218 41 28MB
English Pages 592 [572] Year 2023
Methods in Molecular Biology 2713
Elvira Mass Editor
Tissue-Resident Macrophages Methods and Protocols
METHODS
IN
MOLECULAR BIOLOGY
Series Editor John M. Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, UK
For further volumes: http://www.springer.com/series/7651
For over 35 years, biological scientists have come to rely on the research protocols and methodologies in the critically acclaimed Methods in Molecular Biology series. The series was the first to introduce the step-by-step protocols approach that has become the standard in all biomedical protocol publishing. Each protocol is provided in readily-reproducible step-bystep fashion, opening with an introductory overview, a list of the materials and reagents needed to complete the experiment, and followed by a detailed procedure that is supported with a helpful notes section offering tips and tricks of the trade as well as troubleshooting advice. These hallmark features were introduced by series editor Dr. John Walker and constitute the key ingredient in each and every volume of the Methods in Molecular Biology series. Tested and trusted, comprehensive and reliable, all protocols from the series are indexed in PubMed.
Tissue-Resident Macrophages Methods and Protocols
Edited by
Elvira Mass Life & Medical Sciences Institute (LIMES), University of Bonn, Bonn, Germany
Editor Elvira Mass Life & Medical Sciences Institute (LIMES) University of Bonn Bonn, Germany
ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-3436-3 ISBN 978-1-0716-3437-0 (eBook) https://doi.org/10.1007/978-1-0716-3437-0 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover Illustration Caption: A snapshot from intravital microscopy of the superficial renal cortex in a Cx3cr1gfp/+ Rag2-/- Il2rgY/- mouse depicting tubular capillaries (red), renal tubules (blue), and surfaced rendered Cx3cr1gfp kidney-resident macrophages (green). The tubular capillaries are stained after i.v. injection 2 MDa TRITC-dextran (red), while the renal tubules are stained in blue after endocytosing i.v. injected Alexa Fluor 647-conjugated Ovalbumin (blue). This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.
Preface Macrophages are highly specialized cells that belong to our innate immune system. Beyond their well-recognized immune-related functions, macrophage ontogeny and their role in tissue development and homeostasis have become of great interest. Using fate-mapping models that allow for tracing cells longitudinally, we now know that tissue-resident macrophages originate from fetal progenitors, which colonize the developing organs during embryogenesis. Additionally, after birth, there is a constant influx of monocyte-derived macrophages into every tissue. Thus, every organ harbors macrophages of distinct origins and with different life cycles. The heterogeneity of macrophages is further driven by their microenvironment, where they interact with cells in their subtissular niche (e.g. nerves, blood vessels, epithelial cells, stem cells, etc.), which makes it increasingly difficult to dissect the tissue- and niche-specific core functions of macrophages by standard methods. This is highlighted by recent studies showing that any kind of tissue digestion and subsequent in vitro culturing changes tissue-specific macrophage identity within minutes. In summary, our current knowledge about macrophage development and function forces the scientific field to move beyond the previously described M1/M2 macrophage paradigm to be able to dissect macrophage functions within their specific niches during health and disease. This book highlights the diverse techniques and applications to target, isolate, image, phenotype, and analyze tissue-resident and monocyte-derived macrophages. The collection features an overview of macrophage biology and their core functions as well as an extensive description of how to fate-map the origin of macrophages. Further, sample preparation and analysis from different model organisms, including fruit fly, zebrafish, mice, and humanized mice, are described in detail. In vitro and ex vivo assays to assess macrophage functions, such as efferocytosis and inflammasome activation, are depicted in great detail. As described above, the spatial information of macrophage-tissue cell interaction becomes critical. Thus, several chapters cover imaging techniques from 2D to 3D to intravital imaging analysis. Moreover, bioinformatics tools to quantify and phenotype macrophages are explored, as are methods expanding into metabolomic and epigenetic studies. Together, this detailed edition gives scientists entering the macrophage field information and tools that allow them to dive into the state-of-the-art methodology regarding tissue-resident macrophages. Finally, I would like to express my sincere gratitude to all authors who have taken the time to prepare a chapter and share their knowledge for this book to enable the next generation of macrophage scientists. Bonn, Germany
Elvira Mass
v
Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v xi
1 Macrophage Development and Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nikola Makdissi 2 Fate-Mapping Macrophages: From Ontogeny to Functions . . . . . . . . . . . . . . . . . . Anna Ahlback and Rebecca Gentek 3 Studying Autophagy in Microglia: Overcoming the Obstacles . . . . . . . . . . . . . . . . Ainhoa Plaza-Zabala and Amanda Sierra 4 Hemocyte Nuclei Isolation from Adult Drosophila melanogaster for snRNA-seq. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fabian Hersperger, Melanie Kastl, Katrin Paeschke, and Katrin Kierdorf 5 Isolation of Tissue Macrophages in Adult Zebrafish . . . . . . . . . . . . . . . . . . . . . . . . . Mireia Rovira, Jennifer Pozo, Magali Miserocchi, and Vale´rie Wittamer 6 Genetic and Immunohistochemistry Tools to Visualize Rat Macrophages In Situ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stephen Huang, Dylan Carter-Cusack, Emma Maxwell, Omkar L. Patkar, Katharine M. Irvine, and David A. Hume 7 Phenotyping of Macrophages in Human Immune System Mice. . . . . . . . . . . . . . . Leonie Voss, Carmen Reitinger, and Anja Lux 8 Fate-Mapping of Yolk Sac-Derived Macrophages . . . . . . . . . . . . . . . . . . . . . . . . . . . Iva Splichalova and Elvira Mass 9 Fate-Mapping of Hematopoietic Stem Cell-Derived Macrophages . . . . . . . . . . . . Katharina Mauel and Elvira Mass 10 Isolation and Flow Cytometry Analysis of Macrophages from White Adipose Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ ksel, Dalila Juliana Silva Ribeiro, Seniz Yu Andreas Dolf, and Dagmar Wachten 11 Isolation and Flow Cytometry Analysis of Macrophages from the Dermis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aaron James Forde and Julia Kolter 12 Isolation and Flow Cytometry Analysis of Macrophages from the Kidney . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sarah J. Miller, Alex Yashchenko, and Kurt A. Zimmerman 13 Isolation and Flow Cytometry Analysis of Intestinal Macrophages . . . . . . . . . . . . Maria Francesca Viola and Guy Boeckxstaens 14 Isolation and Characterization of Testis Macrophages Using Flow Cytometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Myriam Sekias, Myriam Baratin, and Marc Bajenoff
1
vii
11 45
71
81
99
117 129 139
149
159
171 183
199
viii
15
16
17
18 19
20 21 22 23
24
25
26
27
28
Contents
Studying Macrophages in the Murine Steatotic Liver Using Flow Cytometry and Confocal Microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhuangzhuang Liu, Pieter A. Louwe, and Charlotte L. Scott Isolation, Ex Vivo Expansion, and Lentiviral Transduction of Alveolar Macrophages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Clara Jana-Lui Busch, Sethuraman Subramanian, Javier Linares, Je´re´my Favret, Ridzky Anis Advent Yuda, and Michael H. Sieweke Translatome Profiling of Tissue-Resident Macrophages Using the RiboTag Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jung-Seok Kim, Zhana Haimon, Sigalit Boura-Halfon, and Steffen Jung Spectral Flow Cytometry Analysis of Resident Tissue Macrophages . . . . . . . . . . . Wan Ting Kong, Mathilde Bied, and Florent Ginhoux Unveiling Macrophage Heterogeneity and Their Spatial Distribution Using Multiplexed Tissue Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . David Alejandro Bejarano and Andreas Schlitzer Three-Dimensional Imaging of Macrophages in Complete Organs. . . . . . . . . . . . Carole Siret and Serge A. van de Pavert Whole-Mount Imaging of Adipose Tissue Macrophages . . . . . . . . . . . . . . . . . . . . . Lydia Sorokin and Luis Henrique Correˆa Functional In Vivo Imaging of Macrophages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anja Wegner, Ralph Palmisano, and Stefan Uderhardt Elucidating Immune Monitoring of Tissue-Resident Macrophages by Intravital Microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Karolin W. Hublitz and Efstathios G. Stamatiades Combined Host-Pathogen Fate Mapping to Investigate Lung Macrophages in Viral Infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sebastian Baasch, Julia Henschel, and Philipp Henneke Measuring the Metabolic State of Tissue-Resident Macrophages via SCENITH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ ello Andrea Vogel, Paulina Garcı´a Gonza´lez, and Rafael J. Argu Analyzing Fcγ-Receptor Interactions on Monocytes with the Proximity Ligation Assay (PLA). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sibel Kara and Falk Nimmerjahn Studying Efferocytosis Dynamics in Tissue-Resident Macrophages Ex Vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Irene Aranda-Pardos, Achmet Imam-Chasan, and Noelia Alonso-Gonzalez Monitoring of Inflammasome Activation of Macrophages and Microglia In Vitro, Part 1: Cell Preparation and Inflammasome Stimulation . . . . . . . . . . . . Marta Lovotti, Matthew S. J. Mangan, Roisı´n M. McManus, Kateryna Shkarina, Matilde B. Vasconcelos, and Eicke Latz
207
231
253
269
281 297 307 323
337
347
363
377
389
407
Contents
29
30
31
32
33 34 35
Monitoring of Inflammasome Activation of Macrophages and Microglia In Vitro, Part 2: Assessing Inflammasome Activation . . . . . . . . . . . Marta Lovotti, Matthew S. J. Mangan, Roisı´n M. McManus, Kateryna Shkarina, Matilde B. Vasconcelos, and Eicke Latz Detection of G-Quadruplex DNA Structures in Macrophages . . . . . . . . . . . . . . . . Melanie Kastl, Fabian Hersperger, Katrin Kierdorf, and Katrin Paeschke Adaptation of Human iPSC-Derived Macrophages Toward an Alveolar Macrophage-Like Phenotype Post-Intra-Pulmonary Transfer into Murine Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Miriam Hetzel, Ingrid Gensch, Mania Ackermann, and Nico Lachmann Tackling Tissue Macrophage Heterogeneity by SplitCre Transgenesis . . . . . . . . . Sigalit Boura-Halfon, Rebecca Haffner-Krausz, Shifra Ben-Dor, Jung-Seok Kim, and Steffen Jung Automated Cell Counting of Macrophages In Situ. . . . . . . . . . . . . . . . . . . . . . . . . . Ouze´na Bouadi and Tuan Leng Tay Morphometric Analyses of Macrophages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jan N. Hansen Combined Analysis of mRNA Expression and Open Chromatin in Microglia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rebekka Scholz, Desire´e Bro¨samle, Xidi Yuan, Jonas J. Neher, and Marc Beyer
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
431
453
463
481
505 519
543
573
Contributors MANIA ACKERMANN • Hannover Medical School, Department of Pediatric Pneumology, Allergology, and Neonatology, Hannover, Germany; Fraunhofer Institute for Toxicology and Experimental Medicine ITEM, Hannover, Germany ANNA AHLBACK • The University of Edinburgh, Institute for Regeneration and Repair, Centre for Reproductive Health & Centre for Inflammation Research, Edinburgh, UK NOELIA ALONSO-GONZALEZ • Institute of Immunology, University of Muenster, Muenster, Germany IRENE ARANDA-PARDOS • Institute of Immunology, University of Muenster, Muenster, Germany RAFAEL J. ARGU¨ELLO • Aix Marseille Univ, CNRS, INSERM, CIML, Centre d’Immunologie de Marseille-Luminy, Marseille, France SEBASTIAN BAASCH • Institute for Imunodeficiency, Center for Chronic Immunodeficiency (CCI), Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany; Institute for Infection Prevention and Control, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany MARC BAJENOFF • Aix-Marseille Universite´, CNRS, INSERM, CIML, Marseille, France MYRIAM BARATIN • Aix-Marseille Universite´, CNRS, INSERM, CIML, Marseille, France DAVID ALEJANDRO BEJARANO • Quantitative Systems Biology, LIMES Institute, University of Bonn, Bonn, Germany SHIFRA BEN-DOR • Department of Life Sciences Core Facilities, Weizmann Institute of Science, Rehovot, Israel MARC BEYER • Immunogenomics & Neurodegeneration, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany; Platform for Single Cell Genomics and Epigenomics at the University of Bonn and the German Center for Neurodegenerative Diseases, Bonn, Germany MATHILDE BIED • Gustave Roussy Cancer Campus, Villejuif, France; Institut National de la Sante´ Et de la Recherche Me´dicale (INSERM) U1015, Equipe Labellise´e—Ligue Nationale contre le Cancer, Villejuif, France GUY BOECKXSTAENS • Center for Neuro-Immune Interaction, Translational Research Center for Gastro-intestinal Disorders, KU Leuven, Leuven, Belgium OUZE´NA BOUADI • Department of Biology, Boston University, Boston, MA, USA SIGALIT BOURA-HALFON • Department of Immunology and Regenerative Biology, Weizmann Institute of Science, Rehovot, Israel DESIRE´E BRO¨SAMLE • Neuroimmunology and Neurodegenerative Disease, German Center for Neurodegenerative Diseases (DZNE), Tuebingen, Germany; Department of Cellular Neurology, Hertie Institute for Clinical Brain Research, University of Tu¨bingen, Tuebingen, Germany CLARA JANA-LUI BUSCH • Center for Regenerative Therapies Dresden (CRTD), Technische Universit€ at Dresden, Dresden, Germany DYLAN CARTER-CUSACK • Mater Research Institute-UQ, Translational Research Institute, Woolloongabba, Brisbane, QLD, Australia
xi
xii
Contributors
LUIS HENRIQUE CORREˆA • Institute of Physiological Chemistry and Pathobiochemistry, Cellsin-Motion Interfaculty Center (CIMIC), University of Mu¨nster, Mu¨nster, Germany ANDREAS DOLF • Flow Cytometry Core Facility, Medical Faculty, University of Bonn, Bonn, Germany JE´RE´MY FAVRET • Center for Regenerative Therapies Dresden (CRTD), Technische Universit€ at Dresden, Dresden, Germany AARON JAMES FORDE • Institute for Immunodeficiency, Center for Chronic Immunodeficiency (CCI), Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany PAULINA GARCI´A GONZA´LEZ • Aix Marseille Univ, CNRS, INSERM, CIML, Centre d’Immunologie de Marseille-Luminy, Marseille, France INGRID GENSCH • Fraunhofer Institute for Toxicology and Experimental Medicine ITEM, Hannover, Germany REBECCA GENTEK • The University of Edinburgh, Institute for Regeneration and Repair, Centre for Reproductive Health & Centre for Inflammation Research, Edinburgh, UK FLORENT GINHOUX • Gustave Roussy Cancer Campus, Villejuif, France; Institut National de la Sante´ Et de la Recherche Me´dicale (INSERM) U1015, Equipe Labellise´e—Ligue Nationale contre le Cancer, Villejuif, France; Universite´ Paris-Saclay, Ile-de-France, France; Singapore Immunology Network (SIgN), A*STAR, Singapore, Singapore; Shanghai Institute of Immunology, Shanghai Jiao Tong University School of Medicine, Shanghai, China; Translational Immunology Institute, SingHealth Duke-NUS Academic Medical Centre, Singapore, Singapore REBECCA HAFFNER-KRAUSZ • Department of Veterinary Resources, Weizmann Institute of Science, Rehovot, Israel ZHANA HAIMON • Department of Immunology and Regenerative Biology, Weizmann Institute of Science, Rehovot, Israel JAN N. HANSEN • Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH - Royal Institute of Technology, Stockholm, Sweden PHILIPP HENNEKE • Institute for Imunodeficiency, Center for Chronic Immunodeficiency (CCI), Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany; Institute for Infection Prevention and Control, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany JULIA HENSCHEL • Institute for Imunodeficiency, Center for Chronic Immunodeficiency (CCI), Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany; Faculty of Biology, University of Freiburg, Freiburg, Germany FABIAN HERSPERGER • Institute of Neuropathology, Faculty of Medicine, University of Freiburg, Freiburg, Germany MIRIAM HETZEL • Hannover Medical School, Department of Pediatric Pneumology, Allergology, and Neonatology, Hannover, Germany STEPHEN HUANG • Mater Research Institute-UQ, Translational Research Institute, Woolloongabba, Brisbane, QLD, Australia KAROLIN W. HUBLITZ • Institute of Microbiology, Infectious Diseases and Immunology (I-MIDI), Charite´-Universit€ atsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany DAVID A. HUME • Mater Research Institute-UQ, Translational Research Institute, Woolloongabba, Brisbane, QLD, Australia
Contributors
xiii
ACHMET IMAM-CHASAN • Institute of Immunology, University of Muenster, Muenster, Germany KATHARINE M. IRVINE • Mater Research Institute-UQ, Translational Research Institute, Woolloongabba, Brisbane, QLD, Australia STEFFEN JUNG • Department of Immunology and Regenerative Biology, Weizmann Institute of Science, Rehovot, Israel SIBEL KARA • Division of Genetics, Department of Biology, University of Erlangen-Nu¨rnberg, Erlangen, Germany MELANIE KASTL • Department of Oncology, Hematology and Rheumatology, University Hospital Bonn, Bonn, Germany; Institute of Clinical Chemistry and Clinical Pharmacology, University Hospital Bonn, Bonn, Germany KATRIN KIERDORF • Institute of Neuropathology, Faculty of Medicine, University of Freiburg, Freiburg, Germany; CIBSS-Centre for Integrative Biological Signalling Studies, University of Freiburg, Freiburg, Germany; Center for Basics in NeuroModulation (NeuroModulBasics), Faculty of Medicine, University of Freiburg, Freiburg, Germany JUNG-SEOK KIM • Department of Immunology and Regenerative Biology, Weizmann Institute of Science, Rehovot, Israel JULIA KOLTER • Institute for Immunodeficiency, Center for Chronic Immunodeficiency (CCI), Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany; Institute for Infection Prevention and Control, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany WAN TING KONG • Gustave Roussy Cancer Campus, Villejuif, France; Institut National de la Sante´ Et de la Recherche Me´dicale (INSERM) U1015, Equipe Labellise´e—Ligue Nationale contre le Cancer, Villejuif, France; Universite´ Paris-Saclay, Ile-de-France, France NICO LACHMANN • Hannover Medical School, Department of Pediatric Pneumology, Allergology, and Neonatology, Hannover, Germany; Fraunhofer Institute for Toxicology and Experimental Medicine ITEM, Hannover, Germany; Hannover Medical School, Cluster of Excellence - Resolving Infection Susceptibility (RESIST, EXC 2155), Hannover, Germany; Hannover Medical School, Biomedical Research in Endstage and Obstructive Lung Disease Hannover (BREATH), German Center for Lung Research (DZL), Hannover, Germany EICKE LATZ • Institute of Innate Immunity, University Hospital Bonn, University of Bonn, Bonn, Germany; German Center for Neurodegenerative Diseases, Bonn, Germany; Department of Infectious Diseases & Immunology, UMass Medical School, Worcester, MA, USA; Centre of Molecular Inflammation Research, Norwegian University of Science and Technology, Trondheim, Norway JAVIER LINARES • Center for Regenerative Therapies Dresden (CRTD), Technische Universit€ at Dresden, Dresden, Germany ZHUANGZHUANG LIU • Laboratory of Myeloid Cell Biology in Tissue Damage and Inflammation, VIB-UGent Center for Inflammation Research, Ghent, Belgium; Department of Biomedical Molecular Biology, Faculty of Science, Ghent University, Ghent, Belgium PIETER A. LOUWE • Laboratory of Myeloid Cell Biology in Tissue Damage and Inflammation, VIB-UGent Center for Inflammation Research, Ghent, Belgium; Department of Biomedical Molecular Biology, Faculty of Science, Ghent University, Ghent, Belgium;
xiv
Contributors
Laboratory of Myeloid Cell Biology in Tissue Homeostasis and Regeneration, VIB-UGent Center for Inflammation Research, Ghent, Belgium MARTA LOVOTTI • Institute of Innate Immunity, University Hospital Bonn, University of Bonn, Bonn, Germany ANJA LUX • Institute of Genetics, Department of Biology, Friedrich-Alexander-Universit€ at Erlangen-Nu¨rnberg, Erlangen, Germany NIKOLA MAKDISSI • Developmental Biology of the Immune System, Life & Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany MATTHEW S. J. MANGAN • Institute of Innate Immunity, University Hospital Bonn, University of Bonn, Bonn, Germany ELVIRA MASS • Developmental Biology of the Immune System, Life & Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany KATHARINA MAUEL • Developmental Biology of the Immune System, Life & Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany EMMA MAXWELL • Mater Research Institute-UQ, Translational Research Institute, Woolloongabba, Brisbane, QLD, Australia RO´ISI´N M. MCMANUS • Institute of Innate Immunity, University Hospital Bonn, University of Bonn, Bonn, Germany; German Center for Neurodegenerative Diseases, Bonn, Germany SARAH J. MILLER • Department of Internal Medicine, Division of Nephrology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA MAGALI MISEROCCHI • Institut de Recherche Interdisciplinaire en Biologie Humaine et Mole´ culaire (IRIBHM), Brussels, Belgium; ULB Neuroscience Institute (UNI), Universite´ Libre de Bruxelles (ULB), Brussels, Belgium JONAS J. NEHER • Neuroimmunology and Neurodegenerative Disease, German Center for Neurodegenerative Diseases (DZNE), Tuebingen, Germany; Department of Cellular Neurology, Hertie Institute for Clinical Brain Research, University of Tu¨bingen, Tuebingen, Germany FALK NIMMERJAHN • Division of Genetics, Department of Biology, University of ErlangenNu¨rnberg, Erlangen, Germany KATRIN PAESCHKE • Department of Oncology, Hematology and Rheumatology, University Hospital Bonn, Bonn, Germany; Institute of Clinical Chemistry and Clinical Pharmacology, University Hospital Bonn, Bonn, Germany RALPH PALMISANO • Optical Imaging Competence Center (OICE), Friedrich-AlexanderUniversit€ at Erlangen-Nu¨rnberg, Erlangen, Germany OMKAR L. PATKAR • Mater Research Institute-UQ, Translational Research Institute, Woolloongabba, Brisbane, QLD, Australia AINHOA PLAZA-ZABALA • Achucarro Basque Center for Neuroscience, Leioa, Spain; Department of Pharmacology, University of the Basque Country (UPV/EHU), Leioa, Spain JENNIFER POZO • Institut de Recherche Interdisciplinaire en Biologie Humaine et Mole´ culaire (IRIBHM), Brussels, Belgium; ULB Neuroscience Institute (UNI), Universite´ Libre de Bruxelles (ULB), Brussels, Belgium CARMEN REITINGER • Institute of Genetics, Department of Biology, Friedrich-AlexanderUniversit€ at Erlangen-Nu¨rnberg, Erlangen, Germany MIREIA ROVIRA • Institut de Recherche Interdisciplinaire en Biologie Humaine et Mole´ culaire (IRIBHM), Brussels, Belgium; ULB Neuroscience Institute (UNI), Universite´ Libre de Bruxelles (ULB), Brussels, Belgium ANDREAS SCHLITZER • Quantitative Systems Biology, LIMES Institute, University of Bonn, Bonn, Germany
Contributors
xv
REBEKKA SCHOLZ • Immunogenomics & Neurodegeneration, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany CHARLOTTE L. SCOTT • Laboratory of Myeloid Cell Biology in Tissue Damage and Inflammation, VIB-UGent Center for Inflammation Research, Ghent, Belgium; Department of Biomedical Molecular Biology, Faculty of Science, Ghent University, Ghent, Belgium; Department of Chemical Sciences, Bernal Institute, University of Limerick, Castletroy, Co. Limerick, Ireland MYRIAM SEKIAS • Aix-Marseille Universite´, CNRS, INSERM, CIML, Marseille, France KATERYNA SHKARINA • Institute of Innate Immunity, University Hospital Bonn, University of Bonn, Bonn, Germany; German Center for Neurodegenerative Diseases, Bonn, Germany AMANDA SIERRA • Achucarro Basque Center for Neuroscience, Leioa, Spain; Department of Neurosciences, University of the Basque Country (UPV/EHU), Leioa, Spain; Department of Biochemistry and Molecular Biology, University of the Basque Country (UPV/EHU), Leioa, Spain; Ikerbasque Foundation, Bilbao, Spain MICHAEL H. SIEWEKE • Center for Regenerative Therapies Dresden (CRTD), Technische Universit€ at Dresden, Dresden, Germany; Aix–Marseille University, CNRS, INSERM, CIML, Marseille, France DALILA JULIANA SILVA RIBEIRO • Institute of Innate Immunity, Department of Biophysical Imaging, Medical Faculty, University of Bonn, Bonn, Germany CAROLE SIRET • Aix-Marseille Univ, CNRS, INSERM, Centre d’Immunologie de MarseilleLuminy (CIML), Marseille, France LYDIA SOROKIN • Institute of Physiological Chemistry and Pathobiochemistry, Cells-in-Motion Interfaculty Center (CIMIC), University of Mu¨nster, Mu¨nster, Germany IVA SPLICHALOVA • Developmental Biology of the Immune System, Life & Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany EFSTATHIOS G. STAMATIADES • Institute of Microbiology, Infectious Diseases and Immunology (I-MIDI), Charite´-Universit€ atsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany SETHURAMAN SUBRAMANIAN • Center for Regenerative Therapies Dresden (CRTD), Technische Universit€ a t Dresden, Dresden, Germany TUAN LENG TAY • Department of Biology, Boston University, Boston, MA, USA; Department of Anatomy & Neurobiology, Boston University Chobanian & Avedisian School of Medicine, Boston, MA, USA STEFAN UDERHARDT • Medizinische Klinik 3 - Rheumatologie und Immunologie, Universit€ atsklinikum Erlangen und Friedrich-Alexander-Universit€ a t ErlangenNu¨rnberg, Erlangen, Germany; Deutsches Zentrum fu¨r Immuntherapie (DZI), FriedrichAlexander-Universit€ at Erlangen-Nu¨rnberg und Universit€ a tsklinikum Erlangen, Erlangen, Germany; Optical Imaging Competence Center (OICE), Friedrich-AlexanderUniversit€ at Erlangen-Nu¨rnberg, Erlangen, Germany SERGE A. VAN DE PAVERT • Aix-Marseille Univ, CNRS, INSERM, Centre d’Immunologie de Marseille-Luminy (CIML), Marseille, France MATILDE B. VASCONCELOS • Institute of Innate Immunity, University Hospital Bonn, University of Bonn, Bonn, Germany MARIA FRANCESCA VIOLA • Developmental Biology of the Immune System, Life and Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany ANDREA VOGEL • Institute for Medical Genetics, Center for Pathobiochemistry and Genetics, Medical University of Vienna, Vienna, Austria
xvi
Contributors
LEONIE VOSS • Institute of Genetics, Department of Biology, Friedrich-AlexanderUniversit€ at Erlangen-Nu¨rnberg, Erlangen, Germany DAGMAR WACHTEN • Institute of Innate Immunity, Department of Biophysical Imaging, Medical Faculty, University of Bonn, Bonn, Germany ANJA WEGNER • Medizinische Klinik 3 - Rheumatologie und Immunologie, Universit€ atsklinikum Erlangen und Friedrich-Alexander-Universit€ a t ErlangenNu¨rnberg, Erlangen, Germany; Deutsches Zentrum fu¨r Immuntherapie (DZI), FriedrichAlexander-Universit€ at Erlangen-Nu¨rnberg und Universit€ a tsklinikum Erlangen, Erlangen, Germany; Optical Imaging Competence Center (OICE), Friedrich-AlexanderUniversit€ at Erlangen-Nu¨rnberg, Erlangen, Germany VALE´RIE WITTAMER • Institut de Recherche Interdisciplinaire en Biologie Humaine et Mole´ culaire (IRIBHM), Brussels, Belgium; ULB Neuroscience Institute (UNI), Universite´ Libre de Bruxelles (ULB), Brussels, Belgium ALEX YASHCHENKO • Department of Internal Medicine, Division of Nephrology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA XIDI YUAN • Neuroimmunology and Neurodegenerative Disease, German Center for Neurodegenerative Diseases (DZNE), Tuebingen, Germany; Department of Cellular Neurology, Hertie Institute for Clinical Brain Research, University of Tu¨bingen, Tuebingen, Germany RIDZKY ANIS ADVENT YUDA • Center for Regenerative Therapies Dresden (CRTD), Technische Universit€ a t Dresden, Dresden, Germany SENIZ YU¨KSEL • Institute of Innate Immunity, Department of Biophysical Imaging, Medical Faculty, University of Bonn, Bonn, Germany KURT A. ZIMMERMAN • Department of Internal Medicine, Division of Nephrology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA
Chapter 1 Macrophage Development and Function Nikola Makdissi Abstract Macrophages were first described over a hundred years ago. Throughout the years, they were shown to be essential players in their tissue-specific environment, performing various functions during homeostatic and disease conditions. Recent reports shed more light on their ontogeny as long-lived, self-maintained cells with embryonic origin in most tissues. They populate the different tissues early during development, where they help to establish and maintain homeostasis. In this chapter, the history of macrophages is discussed. Furthermore, macrophage ontogeny and core functions in the different tissues are described. Key words Tissue-resident macrophages, Hematopoietic waves, Phagocytes, Macrophage ontogeny, Monocytes
1
The History of Macrophages In the late nineteenth century, Elie Metchnikoff first introduced and described macrophages in starfish, which he called phagocytes from the Greek words “phages,” meaning “to eat,” and “cite,” meaning “cell” [1, 2]. This ability to eat and engulf is a common feature of macrophages in vertebrates and macrophage-like cells in many invertebrates [3]. While some macrophage-like cells, such as amoebocytes and coelomocytes, utilize phagocytosis as a digestive mechanism for nutrients, others, such as plasmatocytes in Drosophila melanogaster, are exclusive immune phagocytes. Plasmatocytes are involved in processes such as wound healing and antimicrobial signaling [3]. In vertebrates, macrophages can perform more advanced functions that can be location-specific in mice and humans [4, 5]. In the 1970s, Van Furth and colleagues introduced circulating bone marrow-derived monocytes as macrophage precursors, which dictated the ontogeny model for macrophages for decades [6]. Van Furth’s model still holds true for some tissue-resident macrophages [7–10]. However, genetic fate-mapping techniques and parabiotic models revealed that most resident macrophages, such as microglia
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_1, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
1
2
Nikola Makdissi
in the brain, Kupffer cells in the liver, and alveolar macrophages in the lung, are fetal-derived [11] and self-maintain throughout life [12–16].
2
Macrophage Development In the mouse embryo, hematopoiesis occurs through three distinctive waves. The first wave, primitive hematopoiesis, is initiated at embryonic day 7.5 (E7.5). Primitive erythroid progenitors originate in the yolk sac [17] giving rise to primitive erythroblasts, which will enter the embryo proper to initiate the circulation [18, 19]. In the second wave, called transient hematopoiesis, the erythro-myeloid progenitors (EMPs) arise and expand in the yolk sac between E8.5 and E10.5, and then migrate to the fetal liver. EMPs give rise to pre-macrophage (pMacs), which migrate toward the embryo starting at E9.5. pMacs subsequently colonize the different tissue niches in a Cx3cr1-dependent manner and differentiate into fetal macrophages [20–23]. Furthermore, EMPs have the potential to give rise to monocytes, granulocytes, megakaryocytes, mast cells, erythrocytes, and natural killer cells [17, 24–26]. Those fetal monocytes have been suggested to contribute to the tissue-resident macrophage pool [8, 27, 28]; however, genetic tools distinguishing between monocytes originating during different hematopoietic waves are lacking. The third wave, called definitive hematopoiesis, starts when the hematopoietic stem cells (HSCs) emerge within the hemogenic endothelium in the aorta-gonad-mesonephros region (AGM) at E10.5. Afterward, HSCs migrate to the fetal liver, where they expand and differentiate [29, 30]. After the formation of the bone marrow cavity around E17.5, HSCs migrate to the bone marrow [31]. Shortly after birth, the bone marrow becomes the predominant site for hematopoiesis for life [17]. Even though all tissues have fetal-derived macrophages at birth, HSCs-derived macrophages additionally colonize different tissues at postnatal stages. Ly6Chi monocytes differentiate into macrophages and adopt a tissue-specific signature driven by the niche, such as in the intestine [7], dermis [10], heart [8], and pancreas [8, 9]. Thus, fetal-derived and HSCs-derived macrophages coexist in various organs in the steady-state condition. These populations exhibit different phenotypes and functions, as shown in the peritoneal cavity [32], the lung [33], the adipose tissue [34], and the liver [35]. Under inflammatory conditions, monocytes are recruited into different tissues giving rise to HSCs-derived macrophages. This subset of macrophages is essential in establishing the local inflammatory response or its resolution [36].
Macrophage Development and Function
3
Mammals, birds, and fish share the same macrophage ontogeny. Successive waves of hematopoiesis give rise to macrophages during embryonic development [37]. Human hematopoiesis also follows a similar “wave” model. Definitive erythroid progenitors and bipotential granulocyte/macrophage progenitors arise in the yolk sac and are then found in the liver [38–40]. Human HSCs first appear in the AGM and then in the yolk sac and the liver [39]. Although using genetic models is limited in humans, single cell RNA-Sequencing analysis of human embryos indicated that human tissue-resident macrophages seem to be embryonically derived as well [41].
3
Macrophage Core Functions: From Homeostasis to Pathogenesis Macrophages are resident cells in their tissue niches; they adapt their phenotype from the microenvironment [22, 42] and play a tissue-specific role [43]. Throughout life, macrophages choreograph proper tissue development and function [44]. Furthermore, their ability to initiate and modulate immune responses enables them to play a significant role during pathogenic infection or tissue injury [45, 46]. The following section highlights some of the wellknown functions of tissue-resident macrophages in various tissue niches. Further, examples of the consequences of dysfunctional macrophages emphasize that many tissues rely on proper macrophage core functions for their homeostasis. The central nervous system harbors different subsets of macrophages: microglia, perivascular macrophages, meningeal macrophages, and choroid plexus macrophages [47]. Microglia are the prominent tissue-resident macrophages of the central nervous system. They are fetal-derived and start to populate the brain at E9.5 [48]. During embryogenesis, microglia play a role in neurogenesis by controlling the number of neuronal progenitor cells through phagocytosis [49]. Furthermore, they interact with endothelial cells and neurons, influencing vessel formation [50, 51] and neocortical interneuron migration [52], respectively. Postnatally, microglia produce insulin-like growth factor-1 (Igf-1), which promotes neuronal survival [53]. Microglia also help to modulate the neuronal circuit postnatally through synaptic pruning, a process guided by the complement system [54–56]. Furthermore, they are involved in synaptic modulation and maturation [57]. Later in adulthood, microglia monitor neuronal activity [58] and provide a mechanism for myelin turnover [59]. Dysfunctional microglia are linked to different pathologies such as Parkinson’s disease [60], Alzheimer’s disease [61], schizophrenia [62], and amyotrophic lateral sclerosis [63]. Mutations in the colony-stimulating factor 1 receptor (Csf1r) are linked to various diseases, such as microgliopathy [64] and pediatric-onset
4
Nikola Makdissi
leukoencephalopathy [65]. The latter is characterized by a lack of microglia. Additionally, the somatic mutation BRAFV600E in microglia, leading to their chronic activation, results in neurodegeneration [66]. Kupffer cells are the resident macrophages in the liver. They are fetal-derived macrophages localized in the hepatic sinusoids. In the fetal liver, macrophages are essential for the maturation of erythroblasts [67]. During adulthood, they play a significant role in clearing damaged erythrocytes [68], recycling iron [69], and controlling cholesterol homeostasis [70]. They are crucial to sensing liver injury and pathogens. Upon hepatic injury or diseases, Kupffer cells secrete chemokines, cytokines, and other bioactive molecules, which are essential for the hepatic response [71]. Alveolar macrophages are the main resident macrophages in the lung. They are fetal-derived macrophages localized in the alveoli [27], a location that enables them to surveil their surrounding environment. Alveolar macrophages phagocytose and clear inhaled pathogens and particles without triggering an inflammatory response [72, 73]. Once the pathogen load is over its tolerance limit, alveolar macrophages trigger an inflammatory response via the secretion of chemokines and cytokines [73–75]. Furthermore, dysfunctional alveolar macrophages are linked to pulmonary alveolar proteinosis (PAP), a disease associated with surfactant accumulation within the alveoli [76]. PAP is caused mainly by disruption in granulocyte–macrophage colony-stimulating factor (GM–CSF), an essential factor for developing and maintaining alveolar macrophages [27]. The intestine has the largest macrophage compartment in the body [77]. Anatomically, intestinal macrophages are split into lamina propria macrophages and muscularis macrophages. Despite their fetal origin, some intestinal macrophage populations are gradually replenished by HSCs-derived macrophages [7]. Recently, different reports identified additional subsets of long-lived selfmaintained macrophages [77, 78]. Those subsets are associated with blood vessels and the enteric neurons in the intestine wall, and are important for blood vessel integrity and neuronal survival [79]. However, intestinal macrophages are key players in regulating intestinal immune homeostasis [77, 80] and are essential for the integrity of the gut epithelial barrier [77, 81]. Pathologically, intestinal macrophages play a role in inflammatory bowel diseases linked to the disruption in Il10 signaling [82–85]. Furthermore, they contribute to neurodegeneration in aging and diabetes [85] and the development of diabetic gastroparesis. Adipose tissue macrophages are a heterogeneous population of macrophages with both fetal and HSCs origins [86]. Adipose tissue macrophages have different homeostatic roles in adipose tissue, such as regulating fat storage [86–88] and cholesterol efflux [89]. In obesity, monocytes are recruited to the adipose tissue to
Macrophage Development and Function
5
give rise to macrophages [90] and contribute to tissue inflammation. Adipose tissue macrophages phagocytose the dead adipocytes and tissue debris [91]. Additionally, different reports have linked them to insulin resistance in obesity [92, 93]. In summary, macrophages represent a plastic and heterogeneous cell type. They are great communicators, able to adopt a tissue-specific phenotype based on the microenvironment. Thereby, tissue-resident macrophages contribute to their subtissular niches during development and tissue homeostasis. However, when the core functions of macrophages are disturbed, this may ultimately contribute to or even cause disease onset and progression. References 1. Metschnikoff E (1891) Lecture on phagocytosis and immunity. BMJ 1:213–217 2. Metchnikov E (1883) Untersuchungen ueber die mesodermalen Phagocyten einiger Wirbeltiere. Biologisches Centralblatt 3:560–565 3. Arroyo Portilla C, Tomas J, Gorvel JP, Lelouard H (2021) From species to regional and local specialization of intestinal macrophages. Front Cell Dev Biol 8:1910 4. Kierdorf K, Prinz M, Geissmann F, Gomez Perdiguero E (2015) Development and function of tissue resident macrophages in mice. Semin Immunol 27:369–378 5. Saraiva Camara NO, Braga TT (2022) Macrophages in the human body: a tissue level approach. Elsevier 6. van Furth R, Cohn ZA, Hirsch JG et al (1972) The mononuclear phagocyte system: a new classification of macrophages, monocytes, and their precursor cells. Bull World Health Organ 46:845–852 7. Bain CC, Bravo-Blas A, Scott CL et al (2014) Constant replenishment from circulating monocytes maintains the macrophage pool in the intestine of adult mice. Nat Immunol 15: 929–937. https://doi.org/10.1038/ni.2967 8. Epelman S, Lavine KJ, Beaudin AE et al (2014) Embryonic and adult-derived resident cardiac macrophages are maintained through distinct mechanisms at steady state and during inflammation. Immunity 40:91–104. https://doi. org/10.1016/j.immuni.2013.11.019 9. Calderon B, Carrero JA, Ferris ST et al (2015) The pancreas anatomy conditions the origin and properties of resident macrophages. J Exp Med 212:1497–1512. https://doi.org/10. 1084/jem.20150496 10. Tamoutounour S, Guilliams M, MontananaSanchis F et al (2013) Origins and functional specialization of macrophages and of
conventional and monocyte-derived dendritic cells in mouse skin. Immunity 39:925–938. https://doi.org/10.1016/j.immuni.2013. 10.004 11. Ginhoux F, Guilliams M (2016) Tissueresident macrophage ontogeny and homeostasis. Immunity 44:439–449. https://doi.org/ 10.1016/j.immuni.2016.02.024 12. Merad M, Manz MG, Karsunky H et al (2002) Langerhans cells renew in the skin throughout life under steady-state conditions. Nat Immunol 3:1135–1141. https://doi.org/10.1038/ ni852 13. Jenkins SJ, Ruckerl D, Cook PC et al (2011) Local macrophage proliferation, rather than recruitment from the blood, is a signature of T H2 inflammation. Science (1979) 332: 1284–1288. https://doi.org/10.1126/sci ence.1204351 14. Jakubzick C, Gautier EL, Gibbings SL et al (2013) Minimal differentiation of classical monocytes as they survey steady-state tissues and transport antigen to lymph nodes. Immunity 39:599–610. https://doi.org/10.1016/j. immuni.2013.08.007 15. Hashimoto D, Chow A, Noizat C et al (2013) Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes. Immunity 38:792–804. https://doi.org/10.1016/j. immuni.2013.04.004 16. Ajami B, Bennett JL, Krieger C et al (2007) Local self-renewal can sustain CNS microglia maintenance and function throughout adult life. Nat Neurosci 10:1538–1543. https:// doi.org/10.1038/nn2014 17. Palis J, Robertson S, Kennedy M et al (1999) Development of erythroid and myeloid progenitors in the yolk sac and embryo proper of the mouse. Development 126:5073–5084
6
Nikola Makdissi
18. Ji RP, Phoon CKL, Aristiza´bal O et al (2003) Onset of cardiac function during early mouse embryogenesis coincides with entry of primitive erythroblasts into the embryo proper. Circ Res 92:133–135 19. McGrath KE, Koniski AD, Malik J, Palis J (2003) Circulation is established in a stepwise pattern in the mammalian embryo. Blood 101: 1669–1676. https://doi.org/10.1182/ blood-2002-08-2531 20. Gomez Perdiguero E, Klapproth K, Schulz C et al (2015) Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature 518:547–551. https:// doi.org/10.1038/nature13989 21. McGrath KE, Frame JM, Fegan KH et al (2015) Distinct sources of hematopoietic progenitors emerge before HSCs and provide functional blood cells in the mammalian embryo. Cell Rep 11:1892–1904. https:// doi.org/10.1016/j.celrep.2015.05.036 22. Mass E, Ballesteros I, Farlik M et al (2016) Specification of tissue-resident macrophages during organogenesis. Science (1979) 353: aaf4238. https://doi.org/10.1126/science. aaf4238 23. Stremmel C, Schuchert R, Wagner F et al (2018) Yolk sac macrophage progenitors traffic to the embryo during defined stages of development. Nat Commun 9:75. https://doi.org/ 10.1038/s41467-017-02492-2 24. Bertrand JY, Jalil A, Klaine M et al (2005) Three pathways to mature macrophages in the early mouse yolk sac. Blood 106:3004–3011. https://doi.org/10.1182/blood-200502-0461 25. Mukouyama YS, Chiba N, Mucenski ML et al (1999) Hematopoietic cells in cultures of the murine embryonic aorta-gonad-mesonephros region are induced by c-Myb. Curr Biol 9: 833–836. https://doi.org/10.1016/S09609822(99)80368-6 26. Dege C, Fegan KH, Creamer JP et al (2020) Potently cytotoxic natural killer cells initially emerge from erythro-myeloid progenitors during mammalian development. Dev Cell 53: 229–239.e7. https://doi.org/10.1016/j. devcel.2020.02.016 27. Guilliams M, De Kleer I, Henri S et al (2013) Alveolar macrophages develop from fetal monocytes that differentiate into long-lived cells in the first week of life via GM-CSF. J Exp Med 210:1977–1992. https://doi.org/ 10.1084/jem.20131199 28. Hoeffel G, Chen J, Lavin Y et al (2015) C-Myb + erythro-myeloid progenitor-derived fetal monocytes give rise to adult tissue-resident macrophages. Immunity 42:665–678.
https://doi.org/10.1016/j.immuni.2015. 03.011 29. Medvinsky A, Dzierzak E (1996) Definitive hematopoiesis is autonomously initiated by the AGM region. Cell 86:897–906. https:// doi.org/10.1016/S0092-8674(00)80165-8 30. Boisset JC, van Cappellen W, Andrieu-Soler C et al (2010) In vivo imaging of haematopoietic cells emerging from the mouse aortic endothelium. Nature 464:116–120. https://doi.org/ 10.1038/nature08764 31. Christensen JL, Wright DE, Wagers AJ, Weissman IL (2004) Circulation and chemotaxis of fetal hematopoietic stem cells. PLoS Biol 2: e75. https://doi.org/10.1371/journal.pbio. 0020075 32. dos Cassado AA, D’Impe´rio Lima MR, Bortoluci KR (2015) Revisiting mouse peritoneal macrophages: heterogeneity, development, and function. Front Immunol 6:225. https:// doi.org/10.3389/fimmu.2015.00225 33. Chakarov S, Lim HY, Tan L et al (2019) Two distinct interstitial macrophage populations coexist across tissues in specific subtissular niches. Science 363:eaau0964. https://doi. org/10.1126/science.aau0964 34. Cox N, Geissmann F (2020) Macrophage ontogeny in the control of adipose tissue biology. Curr Opin Immunol 62:1–8 35. Sierro F, Evrard M, Rizzetto S et al (2017) A liver capsular network of monocyte-derived macrophages restricts hepatic dissemination of intraperitoneal bacteria by neutrophil recruitment. Immunity 47:374–388.e6. https://doi. org/10.1016/j.immuni.2017.07.018 36. Ginhoux F, Jung S (2014) Monocytes and macrophages: developmental pathways and tissue homeostasis. Nat Rev Immunol 14:392– 404. https://doi.org/10.1038/nri3671 37. McGrath KE, Frame JM, Palis J (2015) Early hematopoiesis and macrophage development. Semin Immunol 27:379–387 38. Migliaccio G, Migliaccio AR Petti S et al Human embryonic hemopoiesis Kinetics of progenitors and precursors underlying the yolk sac-+ liver transition. J Clin Invest 78: 51–60 39. Ivanovs A, Rybtsov S, Welch L et al (2011) Highly potent human hematopoietic stem cells first emerge in the intraembryonic aortagonad-mesonephros region. J Exp Med 208: 2417–2427. https://doi.org/10.1084/jem. 20111688 40. Aorta-associated CD34+ hematopoietic cells in the early human embryo - PubMed. https:// pubmed.ncbi.nlm.nih.gov/8547678/. Accessed 6 Jan 2023
Macrophage Development and Function 41. Bian Z, Gong Y, Huang T et al (2020) Deciphering human macrophage development at single-cell resolution. Nature 582:571–576. https://doi.org/10.1038/s41586-0202316-7 42. Hoeksema MA, Glass CK (2019) Nature and nurture of tissue-specific macrophage phenotypes. Atherosclerosis 281:159–167 43. Gordon S, Plu¨ddemann A (2017) Tissue macrophages: heterogeneity and functions. BMC Biol 15:53. https://doi.org/10.1186/ s12915-017-0392-4 44. Mass E, Nimmerjahn F, Kierdorf K, Schlitzer A (2023) Tissue-specific macrophages: how they develop and choreograph tissue biology. Nat Rev Immunol 2023:1–17. https://doi.org/ 10.1038/s41577-023-00848-y 45. Gordon S (2007) The macrophage: past, present and future. Eur J Immunol 37:S9–S17. https://doi.org/10.1002/eji.200737638 46. Wynn TA, Chawla A, Pollard JW (2013) Macrophage biology in development, homeostasis and disease. Nature 496:445–455. https://doi. org/10.1038/nature12034 47. Li Q, Barres BA (2018) Microglia and macrophages in brain homeostasis and disease. Nat Rev Immunol 18:225–242. https://doi.org/ 10.1038/nri.2017.125 48. Kierdorf K, Erny D, Goldmann T et al (2013) Microglia emerge from erythromyeloid precursors via Pu.1-and Irf8-dependent pathways. Nat Neurosci 16:273–280. https://doi.org/ 10.1038/nn.3318 ˜ o V, Noc49. Cunningham CL, Martı´nez-Cerden tor SC (2013) Microglia regulate the number of neural precursor cells in the developing cerebral cortex. J Neurosci 33:4216–4233. https://doi.org/10.1523/JNEUROSCI. 3441-12.2013 50. Checchin D, Sennlaub F, Levavasseur E et al (2006) Potential role of microglia in retinal blood vessel formation. Invest Ophthalmol Vis Sci 47:3595–3602. https://doi.org/10. 1167/iovs.05-1522 51. Fantin A, Vieira JM, Gestri G et al (2010) Tissue macrophages act as cellular chaperones for vascular anastomosis downstream of VEGF-mediated endothelial tip cell induction. Blood 116:829–840. https://doi.org/10. 1182/blood-2009-12-257832 52. Squarzoni P, Oller G, Hoeffel G et al (2014) Microglia modulate wiring of the embryonic forebrain. Cell Rep 8:1271–1279. https:// doi.org/10.1016/j.celrep.2014.07.042 53. Ueno M, Fujita Y, Tanaka T et al (2013) Layer v cortical neurons require microglial support for survival during postnatal development.
7
Nat Neurosci 16:543–551. https://doi.org/ 10.1038/nn.3358 54. Stevens B, Allen NJ, Vazquez LE et al (2007) The classical complement cascade mediates CNS synapse elimination. Cell 131:1164– 1178. https://doi.org/10.1016/j.cell.2007. 10.036 55. Schafer DP, Lehrman EK, Kautzman AG et al (2012) Microglia sculpt postnatal neural circuits in an activity and complement-dependent manner. Neuron 74:691–705. https://doi. org/10.1016/j.neuron.2012.03.026 56. Stephan AH, Barres BA, Stevens B (2012) The complement system: an unexpected role in synaptic pruning during development and disease. Annu Rev Neurosci 35:369–389 57. Hoshiko M, Arnoux I, Avignone E et al (2012) Deficiency of the microglial receptor CX3CR1 impairs postnatal functional development of thalamocortical synapses in the barrel cortex. J Neurosci 32:15106–15111. https://doi.org/ 10.1523/JNEUROSCI.1167-12.2012 58. Li Y, Du XF, Liu CS et al (2012) Reciprocal regulation between resting microglial dynamics and neuronal activity in vivo. Dev Cell 23: 1189–1202. https://doi.org/10.1016/j. devcel.2012.10.027 59. Safaiyan S, Kannaiyan N, Snaidero N et al (2016) Age-related myelin degradation burdens the clearance function of microglia during aging. Nat Neurosci 19:995–998. https://doi. org/10.1038/nn.4325 60. Ho MS (2019) Microglia in Parkinson’s disease. In: Advances in experimental medicine and biology. Springer, New York LLC, pp 335–353 61. Hansen DV, Hanson JE, Sheng M (2018) Microglia in Alzheimer’s disease. J Cell Biol 217:459–472 62. Sekar A, Bialas AR, de Rivera H et al (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177– 183. https://doi.org/10.1038/nature16549 63. Clarke BE, Patani R (2020) The microglial component of amyotrophic lateral sclerosis. Brain 143:3526–3539 64. Rademakers R, Baker M, Nicholson AM et al (2012) Mutations in the colony stimulating factor 1 receptor (CSF1R) gene cause hereditary diffuse leukoencephalopathy with spheroids. Nat Genet 44:200–205. https://doi. org/10.1038/ng.1027 65. Oosterhof N, Chang IJ, Karimiani EG et al (2019) Homozygous mutations in CSF1R cause a pediatric-onset leukoencephalopathy and can result in congenital absence of
8
Nikola Makdissi
microglia. Am J Hum Genet 104:936–947. https://doi.org/10.1016/j.ajhg.2019.03.010 66. Mass E, Jacome-Galarza CE, Blank T et al (2017) A somatic mutation in erythro-myeloid progenitors causes neurodegenerative disease. Nature 549:389–393. https://doi.org/10. 1038/nature23672 67. Palis J (2014) Primitive and definitive erythropoiesis in mammals. Front Physiol 5 JAN 68. Willekens FLA, Werre JM, Kruijt JK et al (2005) Liver Kupffer cells rapidly remove red blood cell-derived vesicles from the circulation by scavenger receptors. Blood 105:2141– 2145. https://doi.org/10.1182/blood-200404-1578 69. Guilliams M, Scott CL (2022) Liver macrophages in health and disease. Immunity 55: 1515–1529. https://doi.org/10.1016/j. immuni.2022.08.002 70. Wang Y, van der Tuin S, Tjeerdema N et al (2015) Plasma cholesteryl ester transfer protein is predominantly derived from Kupffer cells. Hepatology 62:1710–1722. https://doi. org/10.1002/hep.27985 71. Krenkel O, Tacke F (2017) Liver macrophages in tissue homeostasis and disease. Nat Rev Immunol 17:306–321. https://doi.org/10. 1038/nri.2017.11 72. Westphalen K, Gusarova GA, Islam MN et al (2014) Sessile alveolar macrophages communicate with alveolar epithelium to modulate immunity. Nature 506:503–506. https://doi. org/10.1038/nature12902 73. MacLean JA, Xia W, Pinto CE et al (1996) Sequestration of inhaled particulate antigens by lung phagocytes: a mechanism for the effective inhibition of pulmonary cell-mediated immunity. Am J Pathol 148:657–666 74. Goritzka M, Makris S, Kausar F et al (2015) Alveolar macrophage-derived type I interferons orchestrate innate immunity to RSV through recruitment of antiviral monocytes. J Exp Med 212:699–714. https://doi.org/10.1084/jem. 20140825 75. Mould KJ, Jackson ND, Henson PM et al (2019) Single cell RNA sequencing identifies unique inflammatory airspace macrophage subsets. JCI Insight 4. https://doi.org/10.1172/ jci.insight.126556 76. Suzuki T, Trapnell BC (2016) Pulmonary alveolar Proteinosis syndrome. Clin Chest Med 37: 431–440 77. Chiaranunt P, Tai SL, Ngai L, Mortha A (2021) Beyond immunity: underappreciated functions of intestinal macrophages. Front Immunol 12:3866
78. de Schepper S, Verheijden S, AguileraLizarraga J et al (2018) Self-maintaining gut macrophages are essential for intestinal homeostasis. Cell 175:400–415.e13. https://doi. org/10.1016/j.cell.2018.07.048 79. Viola MF, Boeckxstaens G (2020) Intestinal resident macrophages: multitaskers of the gut. Neurogastroenterol Motil 32:e13843. https:// doi.org/10.1111/nmo.13843 80. Bain CC, Schridde A (2018) Origin, differentiation, and function of intestinal macrophages. Front Immunol 9. https://doi.org/10.3389/ fimmu.2018.02733 81. Morhardt TL, Hayashi A, Ochi T et al (2019) IL-10 produced by macrophages regulates epithelial integrity in the small intestine. Sci Rep 9: 1223. https://doi.org/10.1038/s41598018-38125-x 82. Ye Z, Hu W, Wu B et al (2021) Predictive prenatal diagnosis for infantile-onset inflammatory bowel disease because of Interleukin-10 signalling defects. J Pediatr Gastroenterol Nutr 72:276–281. https://doi.org/10.1097/ MPG.0000000000002937 83. Hoffmann D, Sens J, Brennig S et al (2021) Genetic correction of IL-10RB deficiency reconstitutes anti-inflammatory regulation in iPSC-derived macrophages. J Pers Med 11: 2 2 1 . h t t p s : // d o i . o r g / 1 0 . 3 3 9 0 / jpm11030221 84. Redhu NS, Bakthavatchalu V, Conaway EA et al (2017) Macrophage dysfunction initiates colitis during weaning of infant mice lacking the interleukin-10 receptor. elife 6. https:// doi.org/10.7554/eLife.27652 85. Delfini M, Stakenborg N, Viola MF, Boeckxstaens G (2022) Macrophages in the gut: masters in multitasking. Immunity 55:1530–1548 86. Cox N, Crozet L, Holtman IR et al (2021) Diet-regulated production of PDGFcc by macrophages controls energy storage. Science (1979) 373:eabe9383. https://doi.org/10. 1126/science.abe9383, 373 87. Pridans C, Raper A, Davis GM et al (2018) Pleiotropic impacts of macrophage and microglial deficiency on development in rats with targeted mutation of the Csf1r locus. J Immunol 201:2683–2699. https://doi.org/10. 4049/jimmunol.1701783 88. Wei S, Lightwood D, Ladyman H et al (2005) Modulation of CSF-1-regulated post-natal development with anti-CSF-1 antibody. Immunobiology 210:109–119. https://doi. org/10.1016/j.imbio.2005.05.005 89. Chawla A, Boisvert WA, Lee CH et al (2001) A PPARγ-LXR-ABCA1 pathway in macrophages is involved in cholesterol efflux and
Macrophage Development and Function atherogenesis. Mol Cell 7:161–171. https:// doi.org/10.1016/S1097-2765(01)00164-2 90. Weisberg SP, McCann D, Desai M et al (2003) Obesity is associated with macrophage accumulation in adipose tissue. J Clin Investig 112:1796–1808. https://doi.org/10.1172/ jci19246 91. Cinti S, Mitchell G, Barbatelli G et al (2005) Adipocyte death defines macrophage localization and function in adipose tissue of obese mice and humans. J Lipid Res 46:2347–2355. https://doi.org/10.1194/jlr.M500294JLR200
9
92. Kanda H, Tateya S, Tamori Y et al (2006) MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Investig 116:1494–1505. https://doi.org/10.1172/ JCI26498 93. Weisberg SP, Hunter D, Huber R et al (2006) CCR2 modulates inflammatory and metabolic effects of high-fat feeding. J Clin Investig 116: 1 1 5 – 1 2 4 . h t t p s : // d o i . o r g / 1 0 . 1 1 7 2 / JCI24335
Chapter 2 Fate-Mapping Macrophages: From Ontogeny to Functions Anna Ahlback and Rebecca Gentek Abstract Macrophages are vital to the physiological function of most tissues, but also contribute to disease through a multitude of pathological roles. They are thus highly plastic and heterogeneous. It is now well recognized that macrophages develop from several distinct progenitors from embryogenesis onwards and extending throughout life. Tissue-resident macrophages largely originate from embryonic sources and in many cases self-maintain independently without monocyte input. However, in certain tissues, monocyte-derived macrophages replace these over time or as a result of tissue injury and inflammation. This additional layer of heterogeneity has introduced many questions regarding the influence of origin on fate and function of macrophages in health and disease. To comprehensively address these questions, appropriate methods of tracing macrophage ontogeny are required. This chapter explores why ontogeny is of vital importance in macrophage biology and how to delineate macrophage populations by origin through genetic fate mapping. First, we summarize the current view of macrophage ontogeny and briefly discuss how origin may influence macrophage function in homeostasis and pathology. We go on to make the case for genetic fate mapping as the gold standard and briefly review different fate-mapping models. We then put forward our recommendations for fate-mapping strategies best suited to answer specific research questions and finally discuss the strengths and limitations of currently available models. Key words Macrophages, Fate mapping, Erythro-myeloid progenitors (EMPs), Hematopoietic stem cells (HSCs), Monocytes
1
Introduction Our understanding of macrophage development has undergone major paradigm shifts over the past two decades. Prior to this, the dominant view was that macrophages originate (exclusively) from the bone marrow (BM) via monocyte intermediates that undergo terminal maturation upon entering tissues [1]. This BM-centric view persisted, although it was known that macrophages populate fetal tissues long before the BM produces any hematopoietic output [2, 3]. These fetal macrophages, therefore, must have other sources. Work from the late 1990s already suggested that the earliest hematopoietic progenitors produced in the extra-embryonic yolk sac (YS) generate microglia, the main resident macrophage
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_2, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
11
12
Anna Ahlback and Rebecca Gentek
population in the brain parenchyma [4]. At the time, studies relied on phenotypic identification of macrophages and their putative progenitors and ex vivo differentiation assays. However, such approaches could not formally demonstrate that fetal progenitors produce macrophages in an unperturbed setting in vivo. Establishing such relationships is made possible through fate mapping, whereby progenitors stably acquire a genetically encoded label that is transmitted to their cellular progeny. It was not until genetic fate mapping was applied to the question of macrophage ontogeny that it was unequivocally shown that microglia indeed originate from YS progenitors [5]. Through the development of additional macrophage fate-mapping models, subsequent landmarking studies demonstrated that YS progenitors generate not only microglia but also macrophages across fetal tissues [6, 7]. These fate-mapping studies also revealed that populations of fetal-derived macrophages persist in adult tissues other than the brain. Over time, these fetal-derived macrophages are supplemented with monocyte-derived ones, and the kinetics and extent of monocyte recruitment are tissue-specific. At steady state, monocyte contribution can range from continuous and predominant, for example, for most intestinal and dermal macrophages, to minimal, for example, for liver Kupffer cells [8–10]. Genetic fate mapping has thus been central to revising our understanding of macrophage ontogeny from the original mononuclear phagocyte system to what we now understand to be a complex, stepwise process, involving discrete progenitors that differ by tissue of origin and time of production [1, 6, 7, 11–14]. Their developmental pattern raises several questions that are key to macrophage biology: Why are some fetal-derived macrophages maintained after birth? To what extent does their origin determine macrophage identity and function? What are the consequences when tissue-resident macrophages are lost and replaced by monocyte-derived macrophages in response to tissue injury or other insults? In this review, we first consider such questions and discuss why one would be interested in delineating macrophage origins. We then make the case for genetic fate mapping as the gold standard to address these questions, give our recommendations for the fatemapping strategies best suited to answer specific research questions, and discuss the strengths and limitations of the currently available models.
2
Developmental Origins and the Microenvironment: Does Ontogeny Matter? In the following, we will briefly summarize our current view of macrophage ontogeny and the relative impact of their origins and environment to highlight applications for genetic fate-mapping studies.
Macrophage Fate Mapping Models
2.1 Current Understanding of Macrophage Ontogeny
13
We now know that macrophages develop from (at least) three types of progenitors that are generated in spatio-temporally distinct but overlapping waves. In the mouse fetus, the first hematopoietic progenitors emerge at embryonic day (E)7.25 in the blood islands of the extraembryonic yolk sac (YS) [15]. These first progenitors are termed primitive since they produce large, nucleated red blood cells known as primitive erythrocytes. Primitive YS progenitors also give rise to megakaryocytes and the first macrophages [4, 5, 7]. A second wave of hematopoietic progenitors with multilineage potential appears at E8.25-E8.5. These erythro-myeloid progenitors (EMPs) continue to produce erythrocytes and megakaryocytes, but also readily produce macrophages as well as granulocytes and mast cells [6–8, 16, 17]. Compared to primitive progenitors, they thus possess an increased lineage potential. EMPs are defined by expression of Kit, CD41, CD93, CD16/32, and absence of Sca1 [18, 19]. They also express the transcription factor Myb, a key regulator of adult hematopoiesis, although, unlike BM HSCs, EMPs do not depend on Myb [6, 7, 11, 20, 21]. This second wave is often referred to as transient- or semi-definitive since EMPs are distinct from primitive progenitors, but cease to exist at fetal stages [22]. EMP output by the YS continues until at least E10.5 [23, 24]. At this time, EMPs are also present in the fetal bloodstream and start to colonize the fetal liver, where they produce lineage-committed precursors. Between E12.5 and E16.5, EMPs give rise to fetal monocytes that go on to seed the developing tissues and differentiate into macrophages [6, 7, 25]. Finally, hematopoietic stem cells (HSCs) first emerge at E10.5 from the intraembryonic hemogenic endothelium [26]. This third wave takes place primarily in the aorta-gonad-mesonephros (AGM) region, but also at additional sites including the vitelline and umbilical arteries, the placenta, and the heart [27–34]. HSCs too migrate to the fetal liver prior to settling in the BM, which becomes the main hematopoietic organ postnatally [35]. They are defined experimentally as cells capable of self-renewing and reconstituting all hematopoietic lineages following transplantation into (lethally) irradiated hosts. Adding further complexity, different pools of HSCs seem to exist [36]. Indeed, the functional definition of HSCs is also met by additional populations of HSCs that emerge slightly earlier in development [14]. Because they appear to be transient in nature, much like EMPs but unlike BM HSCs, these have been termed developmentally restricted HSCs. These progenitors have been identified based on transplantation assays and remain otherwise poorly defined. Consequently, their contribution to macrophage populations is unclear. Since they represent distinct biological entities, we here refer to these progenitors as transient or fetal-restricted HSCs to distinguish them from those found in the BM, which we refer to as adult-type HSCs.
14
Anna Ahlback and Rebecca Gentek
At mid- to late gestation, (both types of) HSCs and EMPs coexist in the fetal liver. This can complicate the interpretation of experimental results concerning macrophage origins. Moreover, both primitive progenitors and EMPs can bypass the fetal liver and directly produce macrophages without a monocyte stage [37, 38], so-called premacrophages (pMacs) [25, 38]. This direct route generates the first macrophages prior to the fetal liver stage, and these macrophages acquire additional specific signatures following recruitment to tissues [38]. Indeed, the very first macrophages can be observed in the YS blood islands by E9.0 and in the brain as early as E9.5 [39]. Because of these shared features and their short succession, as well as different definitions of the respective waves, it remains difficult to pinpoint the precise origin of these very first macrophages even with genetic fate mapping. Therefore, the exact contribution of YS-derived primitive and the first transient-definitive EMPs to macrophages has been debated, especially for microglia. Some researchers consider them entirely primitive [4, 5, 7], whereas according to others, they are primarily EMP-derived with a minor contribution from primitive progenitors [6, 40, 41]. In any case, microglia remain exclusively derived from early YS progenitors throughout life [5]. This complex process results in what has been termed separate “layers” of macrophages that originate from at least three different hematopoietic waves. Throughout life, these developmentally distinct macrophage lineages either (partially) replace each other or coinhabit the same tissue niches. 2.2 Macrophage Specification: Origin, Environment, or Both?
Macrophages are very heterogeneous cells that carry out vital, tissue-specific roles required for normal tissue functioning, development, and protective immunity. These functions can also be co-opted in pathological conditions such as cancer and chronic inflammatory disease. All macrophages share a lineage-defining core gene expression program [42] onto which population- and tissue-specific transcriptional signatures are added that define their identity and enable their functional adaptation [38, 43–45]. Macrophage specification is achieved through successive upregulation of lineage- and population-specific transcription factors in response to signals encountered within the tissue microenvironment, a feature that is captured in the concept of the macrophage “niche” [46]. The discovery that macrophages can originate from different progenitors raised the possibility that they may have discrete functions depending on their origin. A series of studies that ensued provided evidence to the contrary. Alveolar macrophages, for example, originate from fetal monocytes in the perinatal window [44, 47]. Once their network is established, they normally receive minimal BM monocyte input [8]. Following the depletion of the endogenous population, however, alveolar macrophages can be
Macrophage Fate Mapping Models
15
replenished by monocytes. The gene expression profile of the resulting BM-derived alveolar macrophages is strikingly similar to normal, fetal-derived ones [48]. This indicates that both fetal and adult BM monocytes can fully adopt the alveolar macrophage fate. Furthermore, differentiated macrophages from the YS, fetal liverand BM-derived monocytes can all repopulate alveolar macrophages when adoptively transferred into newborn Csf2rb-deficient mice [49]. These mice do not have endogenous alveolar macrophages because they require Gm-Csf (granulocyte macrophage colony stimulating factor), which signals through Csf2rb (colony stimulating factor 2 receptor beta). Since a key function of alveolar macrophages is to clear surfactant from the lung, Csf2rb-deficient mice develop pulmonary alveolar proteinosis (PAP), a condition caused by surfactant build-up [44]. YS macrophages as well as fetal and BM monocytes all occupy the empty alveolar niche in these mice and differentiate into mature alveolar macrophages that stably self-maintain for up to a year. Critically, in this experimental system, surfactant clearance is restored and PAP prevented, irrespective of the origin of alveolar macrophages [49]. This work indicated that the tissue microenvironment has a dominant role in instructing alveolar macrophage fate and functionality, and that different progenitors are all susceptible to this environmental imprinting. Similar conclusions have been drawn for other tissues. When endogenous macrophages are replaced from congenically labeled BM following lethal irradiation of recipient mice, the resulting macrophages transcriptomically largely resemble their normal counterparts [50, 51]. Strikingly, they also share tissue-specific enhancer landscapes with resident macrophages. Microenvironmental cues thus appear to drive the acquisition of selective enhancer repertoires [50, 51]. Collectively, these and similar studies established that distinct types of progenitors, in principle, can differentiate into macrophages that are functionally adapted to their tissue of residence, a process that is facilitated by transcriptional and epigenetic imprinting. This has led to the current consensus that the microenvironment or “nurture” is the major determinant of macrophage identity and function. However, it is important to note that in these experimental approaches, progenitors are introduced into niches that were empty and/or altered by irradiation. This may impact cell–cell interactions and competition with endogenous progenitors. In the case of alveolar macrophages, progenitors were given intranasally with direct access to the luminal alveolar space. They were thus also given access to a niche that may under normal conditions be physically obstructed. Under physiological conditions, niche accessibility and macrophage demands are likely key factors governing recruitment. The nature of the progenitor, on the other hand, may simply be in function of availability at the relevant developmental stage. For example, the first macrophages are recruited to the
16
Anna Ahlback and Rebecca Gentek
primordial brain directly from the YS before the blood-brain barrier (BBB) has been established [52]. When fetal monocytes emerge, there is no longer direct access to the brain parenchyma. The formation of the BBB is thus often considered the reason why microglia remain exclusively of early YS origin. Similarly, lung alveoli form and are widely accessible during a brief perinatal window [53]. Alveolar macrophages are established in this window and subsequently maintained with low BM input in a healthy lung, albeit less limited than in the brain. Alveolar macrophages originate from fetal monocytes, likely because these are the progenitors readily available at the time. Nonetheless, cell-intrinsic differences may still exist. When YS macrophages, fetal and BM monocytes are coinjected into Csf2rbdeficient mice, alveolar macrophages are most efficiently generated from fetal monocytes, their physiological source [49]. Additionally, after Kupffer cell depletion, both engrafting monocytes and remaining resident macrophages act to restore Kupffer cell numbers. However, the remaining resident cells have a competitive advantage as they proliferate earlier than monocyte-derived macrophages [54, 55]. This illustrates that, while distinct progenitors all have the potential to differentiate into tissue-resident macrophages, they may not necessarily do so in unperturbed conditions, such as development, or even following depletion when in competition with remaining endogenous cells. Resolving their true contribution in these conditions requires fate mapping. 2.3 Stage- and Niche-Specific Functions for FetalDerived Macrophages?
In adult organs, macrophage populations with different origins often separate into distinct microanatomical locations. This is the case, for example, in the pancreas, where YS-derived macrophages reside in the interacinar stroma, whereas HSC-derived macrophages localize to the pancreatic islets [56]. Another example is the liver, in which fetal-derived Kupffer cells occupy the sinusoids, whereas macrophages of the capsule are BM-derived [6, 57]. As discussed, lung alveolar macrophages are predominantly fetalderived, but those found in the lung interstitium are of BM origin [58]. Finally, in addition to parenchymal microglia of YS origin, the central nervous system harbors border-associated macrophages that originate from mixed sources [59–62]. These developmentally and spatially distinct macrophage populations may also carry out different functions. Indeed, core organ functions – and the macrophages supporting them – are required throughout life. However, other functions change over the life course and with them, the requirements for resident macrophages. The earliest macrophages are instructed during organogenesis to support organ maturation and vital functions [38], and fetalderived macrophages persisting at later stages may have retained similar roles. On the other hand, macrophages established later, e.g., from the BM, may carry out functions that only become
Macrophage Fate Mapping Models
17
essential to the tissue postnatally. The heart, for example, is first seeded with YS-derived macrophages. A sizeable population of these YS-derived macrophages persists in adult hearts, but they are gradually diluted by incoming BM monocytes [8, 63, 64]. Indeed, most cardiac subpopulations receive monocyte input to some degree, except for self-maintaining Tim4+ Lyve1+ macrophages [65, 66]. The Tim4+ Lyve1+ macrophage population is important for angiogenesis and is primarily located in close contact with cardiac endothelium [58]. During cardiac development, macrophages are necessary for coronary angiogenesis, a role that is carried out specifically by a subset of YS-derived macrophages [67]. Therefore, YS-derived cardiac macrophages with angiogenic functions may be retained from embryogenesis onwards to meet a continued need that mirrors cardiac development. Fetal-derived macrophages are also retained in the mammary gland. They associate preferentially with the vascular endothelium and are necessary for ductal branching during puberty. With the onset of puberty, however, the ductal population become predominated by BM-derived macrophages, which take on an immune surveillance role [68, 69]. Of note, monocytes infiltrating postnatally give rise to ductal macrophages that become phenotypically similar to previously resident fetal-derived macrophages, to the extent that they become indistinguishable from each other by conventional markers [68, 69]. This highlights that macrophage origins cannot be stratified by surface markers but require fate mapping. Finally, most intestinal macrophages are replenished from monocytes throughout life, but a fetal-derived population is specifically retained in the muscularis externa, where it supports intestinal homeostasis through crosstalk with neurons and the vasculature [10, 70]. These examples illustrate how macrophages established from fetal sources may retain functional specificity within certain microanatomical niches. Together with the ever-growing appreciation of their heterogeneity, this highlights the continued need to delineate the origins of macrophage (sub)populations with high resolution and within the tissue context. 2.4 Macrophage Ontogeny: Key to Nonhomeostatic Conditions?
The interplay of developmental origins and microenvironmental imprinting in specifying macrophage identity is even more relevant in nonhomeostatic states, such as infections, injury, and fibrotic or inflammatory disease. Indeed, distinct resident macrophages can show very different responses to such insults. For example, lung interstitial but not alveolar macrophages restrict the growth of Mycobacterium tuberculosis [71]. Alveolar and interstitial macrophages not only showed divergent transcriptomic responses to Mycobacterium tuberculosis challenge but also most of these differences corresponded to epigenetic differences present at baseline and different metabolic activities between the two populations.
18
Anna Ahlback and Rebecca Gentek
Moreover, monocytes are often recruited when the steady state is perturbed to carry out effector functions, and resident macrophage numbers can be diminished. In some instances, and often depending on the nature of the insult, macrophage numbers can be recovered from remaining resident ones. This applies to Kupffer cells following the partial hepatectomy and overdose of N-acetyl-paminophenol [72, 73]. Often, however, monocytes recruited during the insult compensate for the loss of resident macrophages. Examples include the replacement of alveolar macrophages in experimental lung fibrosis, cardiac macrophages post myocardial inflammation and Kupffer cells post pathogen challenge [63, 74, 75]. This may also occur in post-arthritic synovial joints [76]. (Partial) replacement from monocytes can dramatically alter the composition of tissue-resident macrophages with respect to their origins. Persisting monocytes integrate into the local macrophage pool and become phenotypically very similar to or virtually indistinguishable from pre-existing resident macrophages, but their imprinting can be different from macrophages that differentiated at steady state. These monocyte-derived macrophages may thus carry epigenetic “echoes” of the perturbed environment, which may shape their response to subsequent challenges. Indeed, mice in which alveolar macrophages were replaced by monocytes in response to influenza A infection are better protected from subsequent infection with Streptococcus pneumonia [77]. Their alveolar macrophages display substantially altered transcriptional states 2 months after the initial influenza infection and respond to secondary Streptococcus infection with increased IL-6 production. In this case, replacement of pre-existing fetal-derived macrophages with infection-experienced alveolar macrophages is beneficial, a phenomenon now commonly referred to as “trained immunity” [78]. However, it may be detrimental, for example, in chronic lung disease, in which pathologically imprinted, monocytederived macrophages may perpetuate tissue damage over time. To address this, it will be vital to stratify macrophages, not only based on their phenotype and tissue residency but also their origins. Fate mapping is imperative for this. Such studies will be particularly important for chronic conditions that follow cycles of remission and relapse, like rheumatoid arthritis and inflammatory bowel disease. Of note, replacement and/or inappropriate imprinting of resident macrophages may also occur during development in response to environmental challenges, such as maternal immune activation, infections, or exposure to pollutants [79, 80]. Finally, another context in which macrophage ontogeny is important is cancer. Macrophages are abundant components of the tumor microenvironment, where they can have both, pro- and antitumorigenic functions and are also known to shape therapy response. They can originate freshly from the BM or be recruited from surrounding tissues. These pre-existing resident
Macrophage Fate Mapping Models
19
macrophages, again, can have different origins, depending on the affected organ. Intriguingly, fetal-derived macrophages often appear at early stages of tumor development and may promote their local growth and metastasis to remote sites [81–84]. A recent study demonstrated that in hepatocellular carcinoma, a population of macrophages of fetal origins engage in selective crosstalk with endothelial cells that bear resemblance to the fetal liver endothelium [85]. This interaction between fetal-derived macrophages expressing Folr2 (folate receptor beta) and endothelial cells re-expressing Plvap (Plasmalemma Vesicle Associated Protein) generates a protumorigenic, immune-suppressed environment dubbed as an “onco-fetal ecosystem.” Similar interactions may also be found in other tumor types [85, 86]. 2.5 Ontogeny: More than Just Origins
3
The above considerations illustrate that the original “nature versus nurture” debate artificially imposes a dichotomy that does not cover the full biological complexity of tissue-resident macrophages. It is becoming increasingly clear that origins alone do not determine macrophage identity and that nature and nurture do not act independently. Rather, it is the interplay of their sources, the stage they were recruited at and the environmental cues they have been exposed to within their tissue niche and systemically [87]. Defined in this comprehensive way, ontogeny very much shapes macrophage identity and functions. Delineating their ontogeny, therefore, holds the key to understanding macrophage biology in both health and disease.
How to Determine the Origins of Macrophages: The Case for Genetic Fate Mapping Although the interest in macrophage ontogeny has soared over the last decade, experimental tools to study their origins existed long before. These include colony formation or other in vitro differentiation assays, as well as adoptive transfer and parabiosis in vivo. BM chimeras and parabiosis in particular have provided invaluable insight into macrophage biology in health and disease, and they still have their place today. However, they come with caveats that need to be considered and may dissuade the use of these techniques for macrophage ontogeny studies. BM chimeras are widely used in immunology and hematology research. The use of congenically or otherwise-labeled BM allows tracing of grafted BM progenitors and their contribution to resident macrophage populations. Classically, chimeras are established in recipient animals preconditioned by whole-body irradiation, often at lethal doses. This is required to deplete recipient BM and, thus, enable efficient engraftment of transferred progenitors that otherwise would have to compete with the endogenous ones. However, due to the genotoxic nature of irradiation, chimeras
20
Anna Ahlback and Rebecca Gentek
cannot be considered homeostatic. Moreover, some populations, like alveolar macrophages, are radio-sensitive and will be depleted by irradiation, whereas others, like epidermal Langerhans cells and fetal-derived Kupffer cells, are radio-resistant. This must be taken into consideration when assessing the contribution of grafted BM, as it will determine niche availability. To account for these issues, refined approaches have been introduced, such as tissue-protected or “shielded” chimeras, in which only parts of the body are exposed to irradiation, typically the hind legs [54, 88]. This achieves incomplete BM engraftment but can allow for analysis of BM contribution to macrophages in tissues not perturbed by irradiation. Still, BM progenitors tend to be given in excess, at much higher numbers than would normally be available. This can alter cell–cell competition and skew monocyte contribution compared to physiological conditions. Parabiosis is achieved through surgically conjoining two animals to establish a shared blood circulation. This allows free diffusion of blood cells (and soluble factors) between the two parabionts, which typically differ in genetic markers, like the CD45 allele. This model has widely been used to assess the contribution of BM-derived monocytes to tissue-resident macrophages at rest and in disease models [65, 89]. Unlike BM chimeras, parabiosis does not require mice to be preconditioned. Nonetheless, homeostasis is perturbed to some degree in parabiotic mice due to the surgical intervention. Particularly when investigating macrophage populations in the skin, parabiosis may cause issues due to local inflammatory responses where mice are conjoined. Systemic increases in inflammatory mediators and stress levels have also been observed [90], and mice are often treated with antibiotics for prolonged periods. Finally, postoperative mortality can occur [91], and other ethical constraints apply. For these reasons, authorities in many countries no longer allow the use of parabiosis. Unlike chimeras and parabiosis, fate mapping uses genetic approaches to permanently label macrophages according to their origin and is often alternatively referred to as lineage tracing, a less progenitor-centric terminology. Fate mapping depends on the introduction of a genetically encoded label in macrophages or their progenitors via site-specific recombination technology. This is typically achieved through Cre-mediated recombination at loxP recognition sites that will remove a stop codon from a transgenic cassette encoding a fluorescent reporter protein (e.g., in Rosa26lsltdT mice, where lsl stands for “lox-stop-lox”). To fate-map macrophages in an origin-specific manner, expression of Cre recombinase can be targeted to either distinct progenitor types, the hemogenically active endothelium that produces these progenitors, or macrophages themselves. Further specificity can be achieved by making this system inducible through the addition of a modified estrogen receptor to Cre-recombinase. This CreER(T2) complex interacts
Macrophage Fate Mapping Models
21
with heat shock protein (HSP)-90 in the cytoplasm, which prevents its nuclear relocation and, hence, activity. Upon the introduction of the estrogen analog tamoxifen, a ligand for ER, HSP-90 is displaced from the CreER complex, resulting in the release of Cre recombinase to the nucleus. By choosing the time point tamoxifen is introduced, recombination can thus be temporally controlled. The tightest control can be achieved by replacing tamoxifen with its active metabolite, 4-Hydroxy-tamoxifen (4OHT). Compared to tamoxifen, 4OHT acts more rapidly and for a shorter duration [92]. Inducible systems are now the gold standard to fate-map macrophages with fetal origins or determine macrophage turnover during a specified period. Nonetheless, it is worth noting that tamoxifen is not biologically inert and can have side effects that should be controlled for, which we will discuss later. The use of targeted genetic recombination provides several advantages compared to other systems. Firstly, it results in irreversible labeling and, thus, enables tracing macrophages of fetal origin, which was instrumental in the discovery that fetal-derived macrophages persist postnatally. Critically, unlike BM chimeras and parabiosis, genetic fate mapping allows the study of macrophage origins with minimal perturbation of homeostasis. These techniques do not interfere with endogenous cell–cell interactions, progenitor competitions, or niche availability and accessibility. Therefore, the true in vivo contribution of specific progenitors to macrophages can be determined. 3.1 Macrophage Fate-Mapping Models
In the following, we summarize key models currently available to fate-map macrophages of either fetal or BM monocyte origins (see Fig. 1). We highlight their respective strengths and caveats, discuss practical considerations such as the choice of suitable reporters, and make recommendations for models most suitable for different research questions, i.e., their fetal origins and the persistence of fetal-derived macrophages (Table 1) as well as the contribution of adult-type monocytes at either undefined stages (Table 2) or in specified windows (Table 3).
3.1.1 Fate-Mapping Fetal-Derived Macrophages
All models currently used to fate-map macrophages with fetal origins rely on inducible CreER systems (see Table 1). These are induced at specific developmental stages by administration of tamoxifen, typically 4OHT, provided to pregnant dams by intraperitoneal injection. It is important to understand exactly how these models work to make an informed decision which one(s) to use since the cell populations targeted for initial recombination differ between models. These can be endothelial cells, which at the time of induction produce hematopoietic progenitors that go on to generate macrophages (Cdh5-CreER, Tie2MeriCreMer). Alternatively, progenitors are targeted that have (just) undergone the endothelial to hemogenic transition (Runx1CreER) or show varying
22
Anna Ahlback and Rebecca Gentek
Fig. 1 Development of macrophage progenitors and labeling in fate-mapping models. Macrophages develop from at least three types of progenitors, which are generated in successive but overlapping waves during fetal development. Both primitive progenitors and erythro-myeloid progenitors (EMPs) are made in the yolk sac (YS), with primitive progenitors emerging at E7.25, and EMPs being generated from E7.5 until at least E10.5. Adult-type HSCs emerge at E10.5 from the aorta-gonad-mesonephros region (AGM) and fetal-restricted HSCs emerge slightly earlier. Adult-type HSCs ultimately colonize the BM and here generate monocytes postnatally. Using a variety of genetic fate-mapping models, these progenitors can be labeled, and their output can be fate-mapped throughout life. Further explanations in the text and accompanying tables. (Created with Biorender.com)
degrees of commitment to the macrophage lineage, ranging from uncommitted (KitMerCreMer, Cxcr4CreER) to myeloid- or macrophage-biased (Csf1rMeriCreMer, Cx3cr1CreER). Moreover, the genes driving Cre recombinase are also expressed by pMacs or the first differentiated fetal macrophages. This is the case, for example, for Cx3cr1CreER and Csf1rMeriCreMer. Often, fate mapping is used to determine if and to what extent the macrophages of interest originate from YS EMPs. Two models that label EMP-derived macrophages with high efficacy and specificity are Csf1rMeriCreMer (induced at E8.5) and Cdh5-CreER (induced at E7.5) [6, 11, 16, 38]. The Csf1rMeriCreMer model exploits the
E8.5-E9.5, 4OHT
Cxcr4CreER
Fetal HSCs
Fetal macrophages
Does not label YS EMPs
Efficient labeling
E13.5, tam
Some overlap with adult-type HSCs possible Expressed by some macrophages; Check expression in population of interest Expressed by mesodermal progenitors to the YS endothelium; avoid early labeling
Cannot delineate specific fetal origins
Contribution of fetalrestricted progenitors other than EMPs
Fetal-derived macrophages irrespective of specific origins
EMP-derived macrophages
Cx3cr1CreER
More sensitive reporters not tested
Efficient labeling
pMacs
E8.5-E9.5, tam
Some overlap with fetal HSCs EMP-derived macrophages
Use for
Cx3cr1CreER
Efficient labeling of YS hematopoietic output
Notes & limitations
No overlap with fetal Limited labeling efficacy, but EMP-derived and first HSCs more sensitive reporters not macrophages tested
E7.5, 4OHT Endothelial cells producing hematopoietic progenitors (YS)
Cdh5-CreER
Strengths
Csf1rMeriCreMer E8.5, 4OHT EMPs, pMacs, (primitive) macrophages
Induction
Model
Population targeted by Cre(ER)
Table 1 Selected models to fate-map macrophages with fetal origins
(continued)
Werner [94], Simic [132]
Molawi [64], Jokela [98]
Yona [95], Hagemeyer [96], Molawi [64]
Schulz [11], Gomez Perdiguero [6], Mass [38]
Gentek [16]
References
Macrophage Fate Mapping Models 23
Population targeted by Cre(ER)
E7.5, 4OHT Endothelial cells and E8.5, 4OHT YS progenitors
Tie2MeriCreMer
Notes & limitations
Many populations well described
Many populations well described
Also expressed by fetal liver progenitors, HSCs and some macrophages Overlap between waves
Low labeling efficiency (Rosa26lsl-yfp) Overlap between waves
Same line can be used Overlap between waves to label fetal progenitors and HSCs
Strengths
Fetal-derived macrophages irrespective of specific origins
Fetal-derived macrophages irrespective of specific origins
Fetal-derived macrophages irrespective of specific origins
Use for
GomezPerdiguero [6]
Samokhvalov [133], Ginhoux [5], Hoeffel [7]
Sheng [12]
References
In the below models, labeling is induced in endothelial cells producing macrophage progenitors, or either progenitors or macrophages themselves. In some instances, different progenitor and macrophage waves can be targeted using the same mouse line and different time points of label induction YS yolk sac, HSCs hematopoietic stem cells, EMPs erythro-myeloid progenitors, pMacs pre-macrophages, tam tamoxifen, 4OHT 4-hydroxy-tamoxifen
E7.5, 4OHT Progenitors upon E8.5, 4OHT hematopoietic E9.5, 4OHT commitment
E7.5-E8.5, EMPs and other tam hematopoietic (4OHT progenitors not tested)
Induction
Runx1CreER
Kit
MerCreMer
Model
Table 1 (continued)
24 Anna Ahlback and Rebecca Gentek
E9.5, tam Adult-type HSCs No overlap with YS (4OHT not when EMPs tested) produced Little to no overlap with fetal HSCs
Not required
Mds1CreER
Ms4a3Cre
High specificity and efficacy (Rosa26lsltdT )
Gomez-Perdiguero [6]
Monocyte-derived macrophages
Liu et al. [8]
Zhang et al. [135]
Non-EMP-derived Sheng et al. [12] macrophages
Non-EMP-derived Boyer et al. [134], macrophages Schulz e al. [11], Hashimoto et al. [89]
HSC-derived macrophages
Monocyte-derived Prone to male germline macrophages recombination; introduce (gold standard) Cre and reporter separately Low recombination with other lines
More sensitive reporters not tested
Low labeling (Rosa26lsl-yfp) Overlap with fetal HSCs, potentially YS EMPs
References
Non-EMP-derived Gentek et al. [16] macrophages
Use for
These models are either constitutive or induced by the administration of tamoxifen (tam) or 4OHT at fetal stages to label HSCs when they are produced. These approaches should be chosen when the time point of recruitment is not important, for example, to determine the lifetime contribution of monocytes to a macrophage population at steady state 4OHT 4-hydroxy-tamoxifen, EMPs erythro-myeloid progenitors, HSCs hematopoietic stem cells, tam tamoxifen, YS yolk sac
Granulocytemonocyte progenitors (GMPs)
Same line can be used to label fetal progenitors and HSCs
E9.5-E10.5, HSCs tam (4OHT not tested)
KitCreER
Cannot distinguish specific origins
Constitutive Efficient labeling
All HSC-derived cells
Not required
Some overlap with fetal HSCs
Note & limitations
Flt3Cre
Efficient labeling with 4OHT Same line can be used to label fetal progenitors and HSCs
Strengths
Many populations well More sensitive reporters not described tested
E10.5, 4OHT Endothelial cells producing HSCs (AGM)
Induction
Tie2MeriCreMer E10.5, 4OHT Endothelial cells and HSCs
Cdh5
CreER
Model
Population targeted by Cre (ER)
Table 2 Selected models to fate-map macrophages recruited from monocytes at unspecified times
Macrophage Fate Mapping Models 25
26
Anna Ahlback and Rebecca Gentek
Table 3 Selected models to fate-map macrophages recruited from monocytes at specified times
Model CreER
Ccr2
Population targeted by Cre(ER) Monocytes
Strengths high
Ly6C and Ly6Clow monocytes efficiently labeled
Note & limitations Pulse window limited by high monocyte turnover
Cx3cr1CreER Macrophages Efficient labeling Indirect approach, that retain Monocyte decline in labeling limited Cx3cr1 labeling by turnover, expression, indicative of but monocytes dilution with macrophages monocytes stably labeled
Use for
References
Short-term recruitment analysis
Croxford [99], Chen [100]
To complement HSC or monocyte labeling
Yona [95]
Cxcr4CreER
HSCs
Near complete HSC labeling Permanent HSC labeling allows longitudinal assessment of monocyte contribution
Longitudinal Expressed by studies at steady some state or relative macrophages; to insult Check expression in population of interest
Werner [94]
KitCreER
HSCs
Not expressed by Limited labeling Longitudinal (Rosa26lsl-yfp) macrophages studies at steady Permanent HSC state or relative labeling allows to insult, for longitudinal macrophages assessment of that express monocyte Cxcr4 contribution
Sheng Immunity [12]
Ms4a3CreER Granulocyte- High specificity monocyte and efficacy progenitors (Rosa26lsl-tdT) (GMPs)
Pulse window limited by high GMP turnover
Short-term recruitment analysis
Liu [8]
To pulse label BM HSCs or monocytes, labeling is induced by administration of tamoxifen via oral gavage or intraperitoneal injection, typically 2–3 times on consecutive days. These models result in either short-term monocyte labeling allowing analysis of monocyte contribution within a very specific window of time or permanent labeling of monocytes that is, however, initiated at a defined time point important to the study, such as the start of a disease model GMPs granulocyte-monocyte progenitors, HSCs hematopoietic stem cells
expression of Csf1r by EMPs and has been used in landmarking studies demonstrating that EMPs indeed produce the majority of tissue-resident macrophages [6, 11]. The Cdh5-CreER model, on the other hand, targets the initial recombination to endothelial cells, which express Vascular E-Cadherin encoded by Cdh5. In this model, different hematopoietic waves can be labeled, and
Macrophage Fate Mapping Models
27
their macrophage progeny fate-mapped depending on the time of label induction. Resolution between waves is good for the extreme induction time points that correspond to YS and AGM hematopoiesis, respectively [16]. Induction with 4OHT at E7.5 fate-maps YS EMPs, as evidenced by high labeling in microglia, whereas induction at E10.5 labels adult-type HSCs, evidenced by labeling of short-lived leukocytes circulating in the adult blood, which are known to be HSC-descendants [16]. Cdh5 is quickly downregulated once endothelial cells undergo hematopoietic transition. Unlike in the Csf1rMeriCreMer model, macrophages themselves, therefore, cannot recombine and can only inherit the fluorescent label from their progenitors. However, because this approach targets endothelial cells as the highest point of the hierarchy of hematopoiesis, labeling efficacy is amplified, reaching up to 60–80% of YS EMPs [16]. As outlined, subsequent waves of fetal-restricted progenitors are much less well-defined than EMPs. Their contribution to macrophage compartments is thus equally ill-defined. This is because the biology of the underlying hematopoietic programs appears to be very comparable, and (individual) genes useful as Cre drivers specific to different waves have not been identified. Moreover, these waves are initiated shortly after one another, and their production of hematopoietic output overlaps. The resolution of inducible fate mapping for these intermediate time points is thus very poor. One exception from this is the Cxcr4CreER model, which, in conjunction with other fate mappers, can help pinpoint macrophages that originate from fetal-restricted progenitors distinct from YS EMPs. Cxcr4 is expressed by the mesoderm, HSCs, and hemogenic endothelial cells with HSC production capacity [93], but not YS EMPs, pMacs, fetal macrophages, or the YS endothelium at the time it is hemogenically active [94]. Induction with 4OHT at E9.5 results in labeling of fetal-restricted progenitors that precede adult-type HSCs, but does not label YS-derived microglia [94]. Of note, however, Cxcr4 is expressed earlier by mesodermal progenitors that go on to generate YS endothelial cells. Induction at E6.5 labels the mesoderm and, hence, microglia that originate from labeled endothelial cells. Sometimes, deciphering the exact progenitor source is less critical. Rather, one might be interested simply in whether macrophages were established prenatally or originate from such fetalderived macrophages. In other words, the primary interest could be in identifying “all fetal-derived” macrophages irrespective of their specific origins. In these instances, Cx3cr1CreER can be a good approach. Cx3cr1 is expressed by pMacs and terminally differentiated fetal macrophages. Labeling between E8.5 and E9.5 thus captures a range of macrophage states and can result in efficient labeling [95, 96], especially when combined with a sensitive reporter (see below). To further increase labeling, regular
28
Anna Ahlback and Rebecca Gentek
tamoxifen can be used instead of 4OHT, which is normally recommended for better precision of fetal fate mapping. Alternatively, labeling can be induced at later stages of fetal development, such as E13.5 [97, 98], which will primarily label differentiated fetal macrophages. Where such approaches are chosen, it is important to bear in mind that labeled macrophages cannot be assigned to YS (EMP) or “fetal liver” origins, as is sometimes erroneously done. 3.1.2 Fate-Mapping BM Monocyte-Derived Macrophages
When the interest is in monocyte-derived macrophages, it is important to distinguish models that identify these irrespective of when they have been recruited (Table 2) from models in which monocyte contribution can be linked to specific time windows (Table 3). For fate mapping of monocyte-derived macrophages at unspecified times, HSCs or monocytes are either labeled when they are produced during fetal development using inducible systems (Cdh5CreER induced at E10.5, KitCreER induced at E9.5-E10.5, Mds1CreER induced at E9.5) or constitutively labeled based when they pass through a stage marked by expression of specific genes driving Cre recombinase (Flt3Cre, Ms4a3Cre). In both cases, the frequency of labeling is a readout for the lifetime contribution of monocytes to the macrophages of interest, i.e., the extent to which they originated from monocytes until the age at analysis. The Cdh5-CreER model can be used for the purposes of tracing HSC output from their initial appearance in the AGM across the lifespan. Like at E7.5, induction of recombination at E10.5 will label endothelial cells. However, at E10.5, hemogenic activity is primarily located in the AGM, and the YS has largely ceased hemogenic output at this time. This approach thus fatemaps adult-type HSCs as evidenced by labeling of short-lived leukocytes in the adult blood [16]. Using this model, the lifelong contribution of HSCs can thus be quantified, including any potential recruitment before HSCs settle in the BM. However, it is important to note that E10.5 induction in Cdh5-CreER may also label at least some fetal-restricted HSCs. Indeed, discrepancies can be observed between the extent of macrophage labeling in Cdh5CreER pulsed at E10.5 compared to labeling in Ms4a3Cre mice (unpublished data Gentek lab), which reliably fate-map definitive monocytes and will be discussed in the next paragraph. Macrophage labeling is consistently higher in the Cdh5-CreER model, which could be due to labeling of additional progenitors that do not colonize the BM. An alternative explanation is that bona fide adult-type HSCs may seed some tissues prior to settling in the BM and upregulating Ms4a3, or a mixture of both scenarios. This will be interesting to dissect better in the future. To study macrophages originating from monocytes in a comprehensive manner, we recommend the constitutive Ms4a3Cre fatemapping model [8], a tamoxifen-independent model that is rapidly becoming the gold-standard approach. Ms4a3 is highly expressed in
Macrophage Fate Mapping Models
29
granulocyte-monocyte progenitors (GMPs), but not in Ly6Chigh (or Ly6Clow) monocytes. This provides an advantage compared to other models, in which the Cre driver gene is sometimes highly expressed in Ly6Chigh monocytes. At homeostasis, Ms4a3 is also not expressed in any mature tissue-resident macrophage populations, making it the ideal model for delineating monocyte input to these populations, at least in steady-state conditions. Whether Ms4a3 expression is dynamically regulated when homeostasis is perturbed remains to be addressed, however. As for other constitutive Cre drivers, we thus advise to first test this in the biological context of interest. In unperturbed mice, near-complete labeling of circulating monocytes is first observed shortly after birth, making it a useful tool for investigating potential monocyte replacement of fetal-derived macrophages immediately after BM output has started. Moreover, although expression of Ms4a3 has not been formally assessed for fetal macrophages and fetal-restricted progenitors, there is almost no labeling of tissue-resident macrophages at birth [8], indicating that Ms4a3Cre does indeed only label definitive HSC-derived and not EMP-derived monocytes. An important practical consideration for this model is that it is prone to recombination in the male germline [8]. Thus, to produce fate-mapped offspring mice, Ms4a3Cre and a suitable Cre-responsive reporter allele need to be introduced independently through either parent. 3.1.3 Monocyte Input in a Specified Window (PulseChase)
In some instances, one might want to investigate the contribution of recruited monocytes in a specified window in the life of an animal. This is required, for example, to determine the turnover of tissueresident macrophage populations, study the effects of aging on the self-maintenance of resident macrophage populations, or identify monocytes recruited after tissue injury or infection. Such research questions thus necessitate “pulse-chase” approaches in which monocytes are either labeled within a brief window, or labeled permanently but with label induction at a specified time. These approaches involve the induction of labeling, usually in adult animals – the pulse, followed by a period of tamoxifen washout – the chase. For these types of questions, we recommend Cxcr4CreER fate mapping [94]. When tamoxifen is provided to adult animals through either intraperitoneal injection or oral gavage, nearcomplete recombination is induced in BM HSCs. Consequently, HSC-derived lineages, including monocytes, become highly labeled [94]. Importantly, since long-lived HSCs and not shortlived monocytes are targeted, this allows long-term fate mapping of monocyte input to tissue-resident macrophage populations. Indeed, 90% of circulating monocytes remain labeled after 6 months of tamoxifen washout. One caveat of this model is that certain tissue-resident macrophage populations (and other immune cell populations) can express Cxcr4. It is therefore critical to first determine if this applies to the population of interest.
30
Anna Ahlback and Rebecca Gentek
Should the macrophage populations of interest indeed express Cxcr4, then KitCreER could be an alternative. Similar to Cxcr4CreER, providing tamoxifen to adult mice will permanently label HSCs in the BM [12] and, thus, allow long-term fate mapping of their monocyte output. Of note, however, labeling appears to be less efficient than for Cxcr4CreER, at least with the Rosa26lsl-yfp reporter. Alternative reporters have not been tested, to our knowledge. Ccr2CreER and Ms4a3CreER provide yet other options for monocyte fate mapping [8, 99, 100]. Unlike the above models, these approaches directly target recombination to monocytes or GMPs, respectively. Since only existing monocytes or GMPs are labeled, which have a relatively short lifespan, the “pulse” window is therefore short in these models. For example, in Ms4a3CreER mice, substantial labeling of Ly6Chigh monocytes can be observed within 8 days of induction and drastically drops thereafter [8]. Of note, however, labeling is highly efficient, as near complete labeling is observed at peak following just a single shot of tamoxifen. In conclusion, these two models are appropriate for analysis of shortterm recruitment, for example, relative to the start of a disease model. Lastly, a complementary approach that is sometimes exploited is to monitor dilution of pre-existing resident macrophages, rather than directly label monocyte influx. This can be achieved, for example, in Cx3cr1CreER mice [95].
4
Practical Considerations
4.1 Recombination Efficacy, Cre Toxicity, and Zygosity
While they provide means to study macrophage ontogeny with unprecedented resolution, Cre-based fate-mapping models are not without issues. These are important to consider when designing experiments. Key concerns inherent to the Cre/LoxP system are levels of Cre expression and recombination efficacy. These depend on many factors. Levels of Cre expression vary in function of the genes or regulatory elements driving the transgene, and positional effects can also be observed. This is particularly important to consider for knock-in mice targeted to endogenous gene loci as well as Cre drivers for which different versions exist, like Cx3cr1CreER and Ccr2CreER mice [99–102]. The efficacy of any given system thus needs to be assessed for each specific experimental setup. Although maximum recombination efficacy is typically desired for fate-mapping purposes, it cannot always be achieved. Indeed, most systems for fetal fate mapping rely on temporal distinction of progenitor waves that follow each other closely. For these models to work, labeling efficacy thus needs to be balanced with precision. They, therefore, rely on a single dose of tamoxifen (or its derivative 4OHT) and virtually never reach complete labeling.
Macrophage Fate Mapping Models
31
Recombination efficacy can vary substantially between and within litters. To account for this, macrophage or other immune cell populations with known developmental origins need to be analyzed as reference. Commonly used reference populations include brainresident microglia that remain YS-derived throughout life and monocytes in the adult blood, which originate from HSCs in the BM. The frequency of labeling observed in the population of interest can be compared or normalized to these references. In addition to concerns about the precision of labeling, Cre levels may also need to be limited to prevent Cre toxicity. This phenomenon is due to DNA damage at genomic loci other than loxP sites, which leads to cytotoxity, genomic instability, impaired cell cycle progression and can ultimately result in cell death [103]. Such detrimental effects attributed to Cre toxicity have been observed, for example, in microglia following early postnatal Cre induction in Cx3cr1CreER mice [104]. Thus, Cre-negative littermate controls should always be analyzed in parallel to macrophages in fatemapped mice. It is also important to acknowledge the design of the transgenic lines used. In cases where Cre expression is targeted to an endogenous gene locus, zygosity needs to be considered since homozygotes can be rendered deficient for the gene driving Cre expression. This is the case, for example, in Cx3cr1CreER and Cxcr4CreER mice [94, 102]. Therefore, it is not recommended to breed these mice homozygously, which limits the efficacy with which experimental animals can be generated. In some instances, phenotypes can already be observed in heterozygotes. Haploinsufficiency is observed, for example, for Runx1, a key hematopoietic transcription factor. This affects development of the hemogenic epithelium in the YS [105], and is thus relevant to Runx1CreER fate mapping, an approach that labels YS progenitors, including EMPs. 4.2
Reporter Choice
Labeling efficacy is not only dependent on the Cre recombinase but also on the reporter it is combined with. In fact, even for the same Cre(ER) system, recombination efficacy can vary greatly between reporters. Reporter choice thus deserves due consideration. As introduced, Cre-responsive reporter lines typically contain a transgene coding for a fluorescent protein downstream of a stop codon that is flanked by loxP sites. Cre-mediated recombination removes this stop codon and, hence, results in permanent expression of the reporter protein, which is transmitted to any progeny of the cell that originally underwent recombination. These reporter transgenes are normally inserted in a locus that is ubiquitously expressed, often the Gt(ROSA)26Sor (Rosa26) locus. Many different Cre-responsive Rosa26 reporter mice exist. These are widely distributed amongst institutions or (commercially) available through large depositories like The Jackson Laboratory (JAX). They offer a choice between several spectrally distinct reporters,
32
Anna Ahlback and Rebecca Gentek
such as GFP, YFP, CFP, or tdTomato. The decision which reporter to use may often be driven historically or by pragmatic reasons, such as availability of the line within an institution. Early studies delineating the fetal origins of macrophages almost invariably used Rosa26lsl-yfp or Rosa26lsl-rfp mice [6, 7], whereas Rosa26lsl-tdT mice have been more recently introduced to fate-map immune cells including macrophages [16]. This line was generated by the Allen Institute and is also known as “Ai14” or “Ai9” that is very similar in design [106]. Reporter lines differ substantially in their sensitivity and specificity, which need to be considered when designing fate-mapping studies. Rosa26lsl-tdT, for example, is highly sensitive to Cre recombination and thus results in very efficient labeling even with a single dose of 4OHT delivered during fetal development [16]. Its transgene cassette features additional elements designed to enhance the stability of the mRNA transcript (i.e., WPRE, woodchuck hepatitis virus post transcriptional regulatory element). It also exhibits exceptionally bright fluorescence. Therefore, the use of this reporter line has now been adopted by many labs [8, 62, 85, 107]. Indeed, as discussed, it is often desirable to achieve the highest possible recombination efficacy when fate-mapping macrophages. However, to avoid labeling of overlapping waves and achieve precise temporal resolution, fetal fate mappers such as Runx1CreER, Csf1rCreER, and Cdh5-CreER call for a single dose of tamoxifen. In these models, the use of a sensitive reporter such as Rosa26lsl-tdT helps to increase labeling efficacy. The use of a sensitive reporter is also warranted, for example, when the intention is to maximally label adult HSCs or monocytes within a narrow time window to determine their contribution to macrophages relative to an in vivo challenge or insult, such as tumor growth or experimental stroke [94]. However, highly sensitive reporters may suffer from background labeling. For example, recombination may be triggered by very low levels of nuclear Cre. This can occur in inducible models in the absence of tamoxifen, in which CreER may, at a very low rate, spontaneously dissociate from cytoplasmic heat shock proteins and, thus, translocate to the nucleus. Some lines such as Rosa26lsl-tdT may even exhibit some degree of reporter expression without Cre-mediated recombination (e.g., using the Cx3cr1CreER; Rosa26lsl-tdT model, some macrophage populations may be labeled to virtually 100% with tdTomato, even in the absence of tamoxifen). In initial proof of principle experiments, we, therefore, recommend the use of reporter-only (without Cre transgene present) or non-induced (vehicle only) controls in which reporter and CreER transgenes are present, but tamoxifen has not been provided. These are also important to account for potential side effects of tamoxifen and leakiness of the CreER system, which we will consider below.
Macrophage Fate Mapping Models
33
The variety of available reporters also means that more and more specific questions can be asked. For example, one might be interested in delineating subpopulations of macrophages that have passed through a Cx3cr1-expressing state at a specific developmental stage but differ in current expression of Cx3cr1. This distinction can be achieved experimentally by intercrossing Cx3cr1gfp mice, a transcriptional reporter that only labels cells currently expressing Cx3cr1, with Cx3cr1Cre(ER):Rosa26lsl-tdT mice, which label cells with a history of Cx3cr1 expression. Furthermore, additional reporter lines have been developed that carry transgenes for multiple fluorescent proteins. Such multicolor reporters include “Brainbow” [108] and its derivative “Ubow” [109]. Here, recombination results in acquisition of either one of the possible colors in a stochastic manner. In the resulting models, the distribution of fate-mapped macrophages labeled with different colors can be determined by imaging. The formation of monocolored clusters above frequencies that can be expected stochastically is indicative of the expansion of individual macrophage clones. This approach is thus powerful where the interest is not only the source of macrophages but also their mode of expansion during development or following depletion or other insults. It was instrumental in uncovering that the network of epidermal Langerhans cells relies on individual cells with high proliferative potential [109]. Yet other reporters are designed such that cells switch from one to another label upon Cre-mediated recombination. In Rosa26mT/ mG mice, for example, cells express membrane-bound tdTomato prior to recombination and switch to EGFP following Cre activity [110]. These color switch reporters can be useful to monitor, e.g., the expression of genes that are turned on during the differentiation of macrophages from their progenitors, rather than dissecting progenitor origins altogether. Finally, an increasing number of reporters for site-specific recombination use alternative recombinase systems other than Cre/loxP. This allows for the design of combinatorial approaches in which macrophages could be fatemapped according to expression of not one but two genes. We believe this offers possibilities to further refine fate-mapping models in the future and will, therefore, briefly discuss this in the concluding section. 4.3 Limitations of Cre(ER) Systems: Specificity, Precision of Labeling, and Overlap of Hematopoietic Waves
With the notable exception of Ms4a3, individual genes that reliably distinguish distinct hematopoietic programs have not been identified (and may not exist). Current macrophage fate-mappers, therefore, often rely on genes that are shared between endothelial cells, hematopoietic progenitors, and/or even differentiated macrophages. Fate-mapping distinct macrophage lineages thus often depends on temporally defined labeling and, hence, the CreER system and tamoxifen administration.
34
Anna Ahlback and Rebecca Gentek
Indeed, there is a considerable overlap in appearance of hematopoietic progenitors and their macrophage output. The very earliest macrophage progenitors in the YS give rise to macrophages identifiable at E9.0, which then go on to colonize the tissues of the embryo and locally expand. However, EMP output by the YS starts at E8.25 and continues until at least E10.5 [23], meaning macrophages and EMPs are present at the same time and, indeed, share similar gene expression programs. Therefore, labeling of EMPs often leads to unintended labeling of these macrophages as well. This results in an inability to differentiate between true progenitor output and local expansion by fetal macrophages. Furthermore, there is a temporal overlap between the emergence of EMPs and HSCs [23], and the hematopoietic output of these separate progenitors. EMP output continues as late as E16.5, and the first monocyte output by HSCs can be detected as early as E14.5 [6]. For fetal fate-mapping, precise and temporally controlled induction is hence paramount to avoid labeling multiple waves, which appear during development in short succession. Therefore, only one injection is commonly given. Additionally, the use of 4OHT is advised for fetal fate-mapping experiments. Due to the step of conversion of tamoxifen by the liver, the half-life of tamoxifen is much longer than that for its metabolite. Indeed, after tamoxifen treatment, serum levels of 4OHT are sustained for 48–72 h, whereas when 4OHT is given directly, it is cleared within 24 hours [92]. Practically, the need for precision may mean that only relatively low labeling efficacy is achieved in targeted populations. This could lead to difficulties interpreting models since input from other progenitors to the macrophages that are left unlabeled often cannot be formally excluded, at least based on a single model [6, 63]. Even when the above refinements are implemented, we thus recommend the use of complementary models that can report whether other progenitors have contributed to fraction that remains unlabeled in the original model. 4.4 Caveats of Tamoxifen Treatment
Like fetal fate mapping, pulse-chase labeling of BM monocytes for time-resolved analysis of their contribution to macrophage populations requires tamoxifen-mediated induction of CreER systems. Tamoxifen and its metabolite 4OHT can cause considerable side effects both in developing fetuses and adult mice. The compound is an estrogen-analog, and its bioactive metabolites have a high affinity for endogenous estrogen-receptors. It is thus a potent endocrine disruptor, which may result in a litany of confounding effects in female and male mice. In males, even a single dose of 3 mg, which is well in the range of what is used for adult pulse-chase labeling studies, results in a long-term decrease in spermatogenesis [111]. To fate-map macrophages during fetal development, tamoxifen or 4OHT is commonly delivered by intraperitoneal injection of the pregnant female. High doses of tamoxifen rapidly induce abortion
Macrophage Fate Mapping Models
35
if not coadministered with progesterone [112]. In addition, tamoxifen administration at high doses may also increase lethality in pregnant females and disrupt fetal development [112]. Moreover, even at low doses corresponding to those used in fate-mapping experiments, tamoxifen administration during pregnancy often leads to dystocia. To study macrophage origins in adult mice fatemapped at fetal stages, labeled offspring thus often needs to be delivered by Caesarean section and cross-fostered by lactating dams that have not been treated with tamoxifen. This requires breeding of foster colonies, training in Caesarean section and implementation of measures aimed at improving acceptance by foster mothers and minimizing cannibalism, such as smell adaptation. Beyond effects related to the reproductive system, tamoxifen can also directly affect a variety of cell types in a range of other tissues at homeostasis as well as in disease models. Tamoxifen delivered by either intraperitoneal injection or oral gavage is gastrotoxic [113]. Even a single dose of 3 mg induces apoptosis of nearly all gastric parietal cells, leading to metaplasia [113]. In a mouse model of hepatotoxicity, tamoxifen administration increased numbers of tissue-resident macrophages and recruited monocytes in necrotic areas along with a variety of other hepatic parameters [114]. Conversely, mice transplanted with engineered tissue grafts and treated with tamoxifen have a lower degree of macrophage infiltration than controls that have not received tamoxifen [115]. Directly relevant for the purposes of macrophage fate mapping are the effects tamoxifen has on hematopoietic progenitors in the adult. Hematopoietic stem and progenitor cells express estrogen receptors. Even low doses used in fate-mapping experiments can result in reduction of progenitor numbers in the BM [116]. Although fetal hematopoietic progenitors are thought to not express estrogen receptors [117], they may still indirectly be affected by altered estrogen signaling in utero [118]. Indeed, expansion of macrophages has been reported in fetuses treated with tamoxifen at E8.5, at doses commonly used in fetal fatemapping protocols [119]. Finally, an estimated 60% of published studies using tamoxifen inducible in vivo models do not report appropriate tamoxifen-treated controls [111]. This suggests that many tamoxifen-induced effects are still unknown.
5
Concluding Remarks: Future Approaches to Current Limitations Genetic fate mapping has been key to revising our understanding of macrophage ontogeny. Indeed, since the seminal fate-mapping studies of the 2010s, we have come to understand that macrophage identities and functions are shaped by their ontogeny, defined as a combination of their cellular sources and microenvironmental imprinting. For researchers wanting to address questions of
36
Anna Ahlback and Rebecca Gentek
macrophage ontogeny, there is now a suite of useful models to choose from that all have merit for specific questions. Nonetheless, for the reasons discussed here, there is a need for even more refined approaches. Specifically, we need models that better differentiate between fetal progenitor waves. For example, it will be important to develop tools that selectively label fetal-restricted HSCs, but not YS EMPs or adult-type HSCs, in order to better define the biology of these poorly defined progenitors and their contribution to macrophage populations. Another challenge will be to develop constitutive, tamoxifen-independent fate-mapping models for different macrophage lineages, akin to the Ms4a3Cre model for monocytederived macrophages. Combinatorial genetics are a promising candidate strategy toward these aims. Simply put, combinatorial strategies use multiple genes to control recombination, and thus in the case of fate mapping, reporter expression. This can be achieved using multiple different site-specific recombinases [120]. Indeed, although most widely used, Cre recombinase is not the only recombinase available for site-specific recombination. For example, the Dre/rox and Flp/frt systems have long been employed in fate-mapping approaches for developmental and neurobiology, and combinatorial genetic systems utilizing these recombinases have proven feasible and useful to increase specificity, for example, in neuronal lineage tracing studies [121–124]. Yet, combinatorial genetics have been much underutilized by immunologists. By using multiple driver genes and recombinases, it may be possible to circumvent some of the current technical problems with the single-gene based macrophage fate-mapping models, particularly as they relate to the lack of specificity for progenitors and unintended labeling of mature lineages. Additional complexity can be achieved depending on how the reporters for these recombinases are designed. For example, intersectional reporters require two recombination events to express a single reporter. This makes them ideal candidates to increase the specificity of current macrophage fate-mapping models by making label acquisition dependent on two genes [125, 126]. In fact, intersectional reporters that are responsive to up to three separate recombinases have been developed [127]. Alternatively, “exclusive” reporters report only single recombination events, and one recombination event occurring will prevent the other from going ahead. This can generate models in which labeling of undesired cell types can be excluded if they express the recombinase driver gene [128]. Finally, nested reporters allow sequential expression of two distinct reporters, thereby allowing differential labeling, for example, of progenitors and the progeny they give rise to [128– 130]. This could be a way of overcoming the confounding effects of the temporal overlap of distinct progenitor waves. Combinatorial genetics can also be achieved with a single recombinase by splitting the coding transgene in two. In these
Macrophage Fate Mapping Models
37
so-called “split-Cre” models, complementation-competent N-terminal and C-terminal Cre fragments are expressed under control of two distinct promoters. The two Cre fragments on their own lack recombinase activity and only become active if present in the same cell. When combined with a Cre-responsive reporter, only cells expressing both genes (at the same time) will be labeled, and their cellular progeny fate-mapped. This is an alternative approach that circumvents the need for multiple recombinases, but in principle achieves the same as a dual recombinase approach with an intersectional reporter. This strategy has recently been successfully implemented for brain-resident macrophages in an elegant study from the Jung lab [131]. It will be interesting to develop similar models to fate-map, for example, macrophages originating from YS EMPs.
Acknowledgments Work in the Gentek lab was funded by the University of Edinburgh, the Kennedy Trust for Rheumatology Research and Cancer Research, UK. Figures were created with Biorender.com. References 1. van Furth R, Cohn ZA, Hirsch JG et al (1972) The mononuclear phagocyte system: a new classification of macrophages, monocytes, and their precursor cells. Bull World Health Organ 46:845–852 2. Sorokin SP, Hoyt RF, Grant MM (1984) Development of macrophages in the lungs of fetal rabbits, rats, and hamsters. Anat Rec 208:103–121. https://doi.org/10.1002/ AR.1092080112 3. Deimann W, Fahimi HD (1978) Peroxidase cytochemistry and ultrastructure of resident macrophages in fetal rat liver: a developmental study. Dev Biol 66:43–56. https://doi.org/ 10.1016/0012-1606(78)90272-5 4. Alliot F, Godin I, Pessac B (1999) Microglia derive from progenitors, originating from the yolk sac, and which proliferate in the brain. Brain Res Dev Brain Res 117:145–152 5. Ginhoux F, Greter M, Leboeuf M et al (2010) Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science (1979) 330:841–845 6. Gomez Perdiguero E, Klapproth K, Christian S, Busch K et al (2015) Tissueresident macrophages originate from yolksac-derived erythro-myeloid progenitors. Nature 518:547–551
7. Hoeffel G, Chen J, Lavin Y et al (2015) C-Myb(+) erythro-myeloid progenitorderived fetal monocytes give rise to adult tissue-resident macrophages. Immunity 42: 665–678 8. Liu Z, Gu Y, Chakarov S et al (2019) Fate mapping via Ms4a3-expression history traces monocyte-derived cells. Cell 178:1509– 1525.e19 9. Tamoutounour S, Guilliams M, MontananaSanchis F et al (2013) Origins and functional specialization of macrophages and of conventional and monocyte-derived dendritic cells in mouse skin. Immunity 39:925–938. https:// doi.org/10.1016/J.IMMUNI.2013.10.004 10. Bain CC, Bravo-Blas A, Scott CL et al (2014) Constant replenishment from circulating monocytes maintains the macrophage pool in the intestine of adult mice. Nat Immunol 15:929–937. https://doi.org/10.1038/NI. 2967 11. Schulz C, Gomez Perdiguero E, Laurent C, Szabo-Rogers H et al (2012) A lineage of myeloid cells independent of Myb and hematopoietic stem cells. Science (1979) 336:86– 90 12. Sheng J, Ruedl C, Karjalainen K (2015) Most tissue-resident macrophages except microglia are derived from fetal hematopoietic stem cells. Immunity 43:382–393
38
Anna Ahlback and Rebecca Gentek
13. McGrath KE, Frame JM, Palis J (2015) Early hematopoiesis and macrophage development. Semin Immunol 27:379–387 14. Beaudin AE, Boyer SW, Jessica P-C, Hernandez GE et al (2016) A transient developmental hematopoietic stem cell gives rise to innate-like B and T cells. Cell Stem Cell 19: 768–783 15. Palis J, Robertson S, Kennedy M et al (1999) Development of erythroid and myeloid progenitors in the yolk sac and embryo proper of the mouse. Development 126:5073–5084 16. Gentek R, Ghigo C, Hoeffel G, et al (2018) Hemogenic endothelial fate mapping reveals dual developmental origin of mast cells. Immunity 48:1160–1171.e5 17. Iturri L, Freyer L, Biton A, et al (2021) Megakaryocyte production is sustained by direct differentiation from erythromyeloid progenitors in the yolk sac until midgestation. Immunity 54:1433–1446.e5. https://doi.org/10. 1016/J.IMMUNI.2021.04.026 18. Perdiguero EG, Klapproth K, Christian S, Busch K et al (2015) The origin of tissueresident macrophages: when an erythromyeloid progenitor is an erythro-myeloid progenitor. Immunity 43:1023–1024 19. McGrath KE, Frame JM, Fegan KH et al (2015) Distinct sources of hematopoietic progenitors emerge before HSCs and provide functional blood cells in the mammalian embryo. Cell Rep 11:1892–1904 20. Ginhoux F, Guilliams M (2016) Tissueresident macrophage ontogeny and homeostasis. Immunity 44:439–449. https://doi. org/10.1016/j.immuni.2016.02.024 21. McGrath KE, Frame JM, Fegan KH et al (2018) Definitive EMP and pre-HSC emerge in Myb-null murine embryos and retain macrophage potential. Blood 132:2556–2556. https://doi.org/10.1182/BLOOD-201899-118742 22. Lin Y, Yoder MC, Yoshimoto M (2014) Lymphoid progenitor emergence in the murine embryo and yolk sac precedes stem cell detection. Stem Cells Dev 23:1168–1177 23. Frame JM, Fegan KH, Conway SJ et al (2016) Definitive hematopoiesis in the yolk sac emerges from Wnt-responsive hemogenic endothelium independently of circulation and arterial identity. Stem Cells 34:431. https://doi.org/10.1002/STEM.2213 24. Tober J, Yzaguirre AD, Piwarzyk E, Speck NA (2013) Distinct temporal requirements for Runx1 in hematopoietic progenitors and stem cells. Development (Cambridge) 140:
3765–3776. https://doi.org/10.1242/ DEV.094961/-/DC1 25. Stremmel C, Schuchert R, Wagner F et al (2018) Yolk sac macrophage progenitors traffic to the embryo during defined stages of development. Nat Commun 9:75. https:// doi.org/10.1038/s41467-017-02492-2 26. Boisset JC, van Cappellen W, Andrieu-Soler C et al (2010) In vivo imaging of haematopoietic cells emerging from the mouse aortic endothelium. Nature 464:116–120. https:// doi.org/10.1038/nature08764 27. Li Z, Vink CS, Mariani SA, Dzierzak E (2016) Subregional localization and characterization of Ly6aGFP-expressing hematopoietic cells in the mouse embryonic head. Dev Biol. https://doi.org/10.1016/j.ydbio.2016. 05.031 28. Li Z, Lan Y, He W et al (2012) Mouse embryonic head as a site for hematopoietic stem cell development. Cell Stem Cell. https://doi. org/10.1016/j.stem.2012.07.004 29. Shigeta A, Huang V, Zuo J et al (2019) Endocardially derived macrophages are essential for valvular remodeling. Dev Cell. https://doi. org/10.1016/j.devcel.2019.01.021 30. Gordon-Keylock S, Sobiesiak M, Rybtsov S et al (2013) Mouse extraembryonic arterial vessels harbor precursors capable of maturing into definitive HSCs. Blood 122:2338–2345. https://doi.org/10.1182/blood-201212-470971 31. Gekas C, Dieterlen-Lie`vre F, Orkin SH, Mikkola HKA (2005) The placenta is a niche for hematopoietic stem cells. Dev Cell. https:// doi.org/10.1016/j.devcel.2004.12.016 32. Alvarez-Silva M, Belo-Diabangouaya P, Salau¨n J, Dieterlen-Lie`vre F (2003) Mouse placenta is a major hematopoietic organ. Development. https://doi.org/10.1242/ dev.00755 33. Solano ME, Thiele K, Stelzer IA et al (2014) Advancing the detection of maternal haematopoietic microchimeric cells in fetal immune organs in mice by flow cytometry. Chimerism 5:99–102. https://doi.org/10.4161/ 19381956.2014.959827 34. Zeigler BM, Sugiyama D, Chen M et al (2006) The allantois and chorion, when isolated before circulation or chorio-allantoic fusion, have hematopoietic potential. Development 133:4183–4192. https://doi.org/ 10.1242/dev.02596 35. Kieusseian A, de la Grange P, BurlenDefranoux O et al (2012) Immature hematopoietic stem cells undergo maturation in the fetal liver. Development 139:3521–3530
Macrophage Fate Mapping Models 36. Yokomizo T, Ideue T, Morino-Koga S et al (2022) Independent origins of fetal liver haematopoietic stem and progenitor cells. Nature 609:779–784. https://doi.org/10.1038/ S41586-022-05203-0 37. Naito M, Yamamura F, Nishikawa S, Takahashi K (1989) Development, differentiation, and maturation of fetal mouse yolk sac macrophages in cultures. J Leukoc Biol 46:1–10 38. Mass E, Ballesteros I, Farlik M et al (2016) Specification of tissue-resident macrophages during organogenesis. Science (1979) 353: aaf4238 39. Takahashi K, Yamamura F, Naito M (1989) Differentiation, maturation, and proliferation of macrophages in the mouse yolk sac: a lightmicroscopic, enzyme-cytochemical, immunohistochemical, and ultrastructural study. J Leukoc Biol 45:87–96 40. De S, Van Deren D, Peden E et al (2018) Two distinct ontogenies confer heterogeneity to mouse brain microglia. Development 145: dev152306. https://doi.org/10.1242/dev. 152306 41. Kierdorf K, Erny D, Goldmann T et al (2013) Microglia emerge from erythromyeloid precursors via Pu.1- and Irf8-dependent pathways. Nat Neurosci 16:273–280. https:// doi.org/10.1038/nn.3318 42. Gautiar EL, Shay T, Miller J et al (2012) Gene-expression profiles and transcriptional regulatory pathways that underlie the identity and diversity of mouse tissue macrophages. Nat Immunol 13:1118–1128. https://doi. org/10.1038/NI.2419 43. A-Gonzalez N, Guillen JA, Gallardo G et al (2013) The nuclear receptor LXRα controls the functional specialization of splenic macrophages. Nat Immunol 14:831–839. https:// doi.org/10.1038/NI.2622 44. Schneider C, Nobs SP, Kurrer M et al (2014) Induction of the nuclear receptor PPAR-γ by the cytokine GM-CSF is critical for the differentiation of fetal monocytes into alveolar macrophages. Nature Immunol 15(11): 1026–1037. https://doi.org/10.1038/ni. 3005 45. Kohyama M, Ise W, Edelson BT et al (2008) Role for Spi-C in the development of red pulp macrophages and splenic iron homeostasis. Nature 457(7227):318–321. https://doi. org/10.1038/nature07472 46. Guilliams M, Thierry GR, Bonnardel J, Bajenoff M (2020) Establishment and maintenance of the macrophage Niche. Immunity 52:434– 451. https://doi.org/10.1016/J.IMMUNI. 2020.02.015
39
47. Guilliams M, de Kleer I, Henri S et al (2013) Alveolar macrophages develop from fetal monocytes that differentiate into long-lived cells in the first week of life via GM-CSF. J Exp Med 210:1977–1992 48. Gibbings SL, Goyal R, Desch AN et al (2015) Transcriptome analysis highlights the conserved difference between embryonic and postnatal-derived alveolar macrophages. Blood 126:1357–1366 49. van de Laar L, Saelens W, de Prijck S et al (2016) Yolk sac macrophages, fetal liver, and adult monocytes can colonize an empty Niche and develop into functional tissue-resident macrophages. Immunity 44:755–768 50. Gosselin D, Link VM, Romanoski CE et al (2014) Environment drives selection and function of enhancers controlling tissuespecific macrophage identities. Cell 159: 1327–1340. https://doi.org/10.1016/J. CELL.2014.11.023 51. Lavin Y, Winter D, Blecher-Gonen R et al (2014) Tissue-resident macrophage enhancer landscapes are shaped by the local microenvironment. Cell 159:1312–1326 52. Engelhardt B (2003) Development of the blood-brain barrier. Cell Tissue Res 314: 119–129. https://doi.org/10.1007/ S00441-003-0751-Z/FIGURES/3 53. Mund SI, Stampanoni M, Schittny JC (2008) Developmental alveolarization of the mouse lung. Dev Dyn 237:2108–2116. https://doi. org/10.1002/DVDY.21633 54. Scott CL, Zheng F, de Baetselier P et al (2016) Bone marrow-derived monocytes give rise to self-renewing and fully differentiated Kupffer cells. Nature Communications 7(1):1–10. https://doi.org/10.1038/ ncomms10321 55. Bonnardel J, T’Jonck W, Gaublomme D et al (2019) Stellate cells, hepatocytes, and endothelial cells imprint the Kupffer cell identity on monocytes colonizing the liver macrophage niche. Immunity 51:638–654.e9. https://doi.org/10.1016/J.IMMUNI. 2019.08.017 56. Calderon B, Carrero JA, Ferris ST et al (2015) The pancreas anatomy conditions the origin and properties of resident macrophages. J Exp Med 212:1497–1512. https://doi.org/10. 1084/JEM.20150496 57. Sierro F, Evrard M, Rizzetto S et al (2017) A liver capsular network of monocyte-derived macrophages restricts hepatic dissemination of intraperitoneal bacteria by neutrophil recruitment. Immunity 47:374–388.e6
40
Anna Ahlback and Rebecca Gentek
58. Chakarov S, Lim HY, Tan L et al (2019) Two distinct interstitial macrophage populations coexist across tissues in specific subtissular niches. Science 1979:363 59. Goldmann T, Wieghofer P, Jorda˜o MJC et al (2016) Origin, fate and dynamics of macrophages at central nervous system interfaces. Nat Immunol 17:797–805. https://doi.org/ 10.1038/ni.3423 60. van Hove H, Martens L, Scheyltjens I et al (2019) A single-cell atlas of mouse brain macrophages reveals unique transcriptional identities shaped by ontogeny and tissue environment. Nat Neurosci 22:1021–1035. https://doi.org/10.1038/s41593-0190393-4 61. Mrdjen D, Pavlovic A, Hartmann FJ et al (2018) High-dimensional single-cell mapping of central nervous system immune cells reveals distinct myeloid subsets in health, aging, and disease. Immunity 48:380–395.e6. https:// doi.org/10.1016/j.immuni.2018.01.011 62. Utz SG, See P, Mildenberger W et al (2020) Early fate defines microglia and non-parenchymal brain macrophage development. Cell 181:557–573.e18. https://doi. org/10.1016/j.cell.2020.03.021 63. Epelman S, Lavine KJ, Beaudin AE et al (2014) Embryonic and adult-derived resident cardiac macrophages are maintained through distinct mechanisms at steady state and during inflammation. Immunity 40:91–104 64. Molawi K, Wolf Y, Kandalla PK et al (2014) Progressive replacement of embryo-derived cardiac macrophages with age. J Exp Med 211:2151–2158 65. Dick SA, Macklin JA, Nejat S et al (2019) Self-renewing resident cardiac macrophages limit adverse remodeling following myocardial infarction. Nat Immunol 20:29–39 66. Dick SA, Wong A, Hamidzada H et al (2022) Three tissue resident macrophage subsets coexist across organs with conserved origins and life cycles. Sci Immunol 7:7777. https:// doi.org/10.1126/SCIIMMUNOL. ABF7777/SUPPL_FILE/SCIIMMUNOL. ABF7777_MDAR_REPRODUCIBILITY_ CHECKLIST.ZIP 67. Leid J, Carrelha J, Boukarabila H et al (2016) Primitive embryonic macrophages are required for coronary development and maturation. Circ Res 118:1498–1511 68. J€appinen N, Fe´lix I, Lokka E et al (2019) Fetal-derived macrophages dominate in adult mammary glands. Nat Commun:10. https:// doi.org/10.1038/s41467-018-08065-1
69. Dawson CA, Pal B, Vaillant F et al (2020) Tissue-resident ductal macrophages survey the mammary epithelium and facilitate tissue remodelling. Nat Cell Biol 22(5):546–558. https://doi.org/10.1038/s41556-0200505-0 70. de Schepper S, Verheijden S, AguileraLizarraga J et al (2018) Self-maintaining gut macrophages are essential for intestinal homeostasis. Cell 175:400–415.e13 71. Huang L, Nazarova EV, Tan S et al (2018) Growth of Mycobacterium tuberculosis in vivo segregates with host macrophage metabolism and ontogeny. J Exp Med 215: 1135–1152 72. Ait Ahmed Y, Fu Y, Rodrigues RM et al (2021) Kupffer cell restoration after partial hepatectomy is mainly driven by local cell proliferation in IL-6-dependent autocrine and paracrine manners. Cell Mol Immunol 18:2165–2176 73. Zigmond E, Samia-Grinberg S, PasmanikChor M et al (2014) Infiltrating monocytederived macrophages and resident kupffer cells display different ontogeny and functions in acute liver injury. J Immunol 193:344–353. https://doi.org/10.4049/JIMMUNOL. 1400574 74. Misharin AV, Morales-Nebreda L, Reyfman PA et al (2017) Monocyte-derived alveolar macrophages drive lung fibrosis and persist in the lung over the life span. J Exp Med 214:2387–2404 75. Ble´riot C, Dupuis T, Jouvion G et al (2015) Liver-resident macrophage necroptosis orchestrates type 1 microbicidal inflammation and type-2-mediated tissue repair during bacterial infection. Immunity 42:145–158. https://doi.org/10.1016/j.immuni.2014. 12.020 76. Misharin AV, Cuda CM, Saber R et al (2014) Nonclassical Ly6C(-) monocytes drive the development of inflammatory arthritis in mice. Cell Rep 9:591–604. https://doi.org/ 10.1016/J.CELREP.2014.09.032 77. Aegerter H, Kulikauskaite J, Stefania C, Patel H et al (2020) Influenza-induced monocytederived alveolar macrophages confer prolonged antibacterial protection. Nat Immunol 21:145–157 78. Netea MG, Domı´nguez-Andre´s J, Barreiro LB et al (2020) Defining trained immunity and its role in health and disease. Nat Rev Immunol 20:375–388. https://doi.org/10. 1038/S41577-020-0285-6
Macrophage Fate Mapping Models 79. Mass E, Gentek R (2021) Fetal-derived immune cells at the roots of lifelong pathophysiology. Front Cell Dev Biol 9:648313 80. Magalhaes MS, Potter HG, Ahlback A, Gentek R (2022) Chapter Six - Developmental programming of macrophages by early life adversity. In: Cassetta L, Galluzzi L (eds) Mariani SA. Academic Press, International review of cell and molecular biology, pp 213–259 81. Casanova-Acebes M, Dalla E, Leader AM et al (2021) Tissue-resident macrophages provide a pro-tumorigenic niche to early NSCLC cells. Nature. https://doi.org/10.1038/ s41586-021-03651-8 82. Zhu Y, Herndon JM, Sojka DK et al (2017) Tissue-resident macrophages in pancreatic ductal adenocarcinoma originate from embryonic hematopoiesis and promote tumor progression. Immunity 47:323–338.e6. https:// doi.org/10.1016/j.immuni.2017.07.014 83. Loyher P-L, Hamon P, Laviron M et al (2018) Macrophages of distinct origins contribute to tumor development in the lung. J Exp Med 215:2536–2553. https://doi.org/ 10.1084/jem.20180534 84. Etzerodt A, Moulin M, Doktor TK et al (2020) Tissue-resident macrophages in omentum promote metastatic spread of ovarian cancer. J Exp Med 217. https://doi.org/ 10.1084/JEM.20191869 85. Sharma A, Seow JJW, Dutertre C-A et al (2020) Onco-fetal Reprogramming of Endothelial Cells Drives Immunosuppressive Macrophages in Hepatocellular Carcinoma. Cell 183:377–394.e21. https://doi.org/10. 1016/j.cell.2020.08.040 86. Sharma A, Ble´riot C, Currenti J, Ginhoux F (2022) Oncofetal reprogramming in tumour development and progression. Nat Rev Cancer:1–10. https://doi.org/10.1038/ s41568-022-00497-8 87. Park MD, Silvin A, Ginhoux F, Merad M (2022) Macrophages in health and disease. Cell 185:4259–4279. https://doi.org/10. 1016/J.CELL.2022.10.007 88. Baratin M, Simon L, Jorquera A et al (2017) T Cell Zone Resident Macrophages Silently Dispose of Apoptotic Cells in the Lymph Node. Immunity 47:349–362.e5. https:// doi.org/10.1016/j.immuni.2017.07.019 89. Hashimoto D, Chow A, Noizat C et al (2013) Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes. Immunity 38:792–804. https://doi.org/10. 1016/J.IMMUNI.2013.04.004
41
90. Yang C, Liu Z-L, Wang J et al (2021) Parabiosis modeling: protocol, application and perspectives. Zool Res 42:253–261. https://doi. org/10.24272/J.ISSN.2095-8137. 2020.368 91. Spencer CA, Leung TH (2019) Research techniques made simple: parabiosis to elucidate humoral factors in skin biology. J. Invest. Dermatol. 139:1208–1213.e1. https://doi. org/10.1016/J.JID.2019.03.1134 92. Zovein AC, Hofmann JJ, Lynch M et al (2008) Fate tracing reveals the endothelial origin of hematopoietic stem cells. Cell Stem Cell 3:625–636. https://doi.org/10.1016/J. STEM.2008.09.018 93. Dignum T, Varnum-Finney B, Srivatsan SR et al (2021) Multipotent progenitors and hematopoietic stem cells arise independently from hemogenic endothelium in the mouse embryo. Cell Rep 36. https://doi.org/10. 1016/J.CELREP.2021.109675 94. Werner Y, Mass E, Ashok Kumar P et al (2020) Cxcr4 distinguishes HSC-derived monocytes from microglia and reveals monocyte immune responses to experimental stroke. Nat Neurosci 23:351–362 95. Yona S, Kim K-W, Wolf Y et al (2013) Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis. Immunity 38:79–91. https:// doi.org/10.1016/j.immuni.2012.12.001 96. Hagemeyer N, Kierdorf K, Frenzel K, et al (2016) Transcriptome-based profiling of yolk sac-derived macrophages reveals a role for Irf8 in macrophage maturation. EMBO J 35:1730–1744. https://doi.org/10.15252/ embj.201693801 97. Liu F, Dai S, Feng D et al (2020) Distinct fate, dynamics and niches of renal macrophages of bone marrow or embryonic origins. Nat Commun 11:2280. https://doi.org/10. 1038/s41467-020-16158-z 98. Jokela H, Lokka E, Kiviranta M et al (2020) Fetal-derived macrophages persist and sequentially maturate in ovaries after birth in mice. Eur J Immunol 50:1500–1514. https://doi.org/10.1002/eji.202048531 99. Croxford AL, Lanzinger M, Hartmann FJ et al (2015) The cytokine GM-CSF drives the inflammatory signature of CCR2+ monocytes and lcenses autoimmunity. Immunity 43:502–514. https://doi.org/10.1016/j. immuni.2015.08.010 100. Chen H-R, Sun Y-Y, Chen C-W et al (2020) Fate mapping via CCR2-CreER mice reveals monocyte-to-microglia transition in development and neonatal stroke. Sci Adv 6:
42
Anna Ahlback and Rebecca Gentek
eabb2119. https://doi.org/10.1126/sciadv. abb2119 101. Parkhurst CN, Yang G, Ninan I et al (2013) Microglia promote learning-dependent synapse formation through brain-derived neurotrophic factor. Cell 155:1596–1609. https:// doi.org/10.1016/j.cell.2013.11.030 102. Jung S, Aliberti J, Graemmel P et al (2000) Analysis of fractalkine receptor CX(3)CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol 20:4106–4114. https://doi.org/ 10.1128/MCB.20.11.4106-4114.2000 103. Janbandhu VC, Moik D, F€assler R (2014) Cre recombinase induces DNA damage and tetraploidy in the absence of loxP sites. Cell Cycle 13:462–470. https://doi.org/10.4161/cc. 27271 104. Sahasrabuddhe V, Ghosh HS (2022) Cx3Cr1-Cre induction leads to microglial activation and IFN-1 signaling caused by DNA damage in early postnatal brain. Cell Rep 38:110252 105. Lacaud G, Kouskoff V, Trumble A et al (2004) Haploinsufficiency of Runx1 results in the acceleration of mesodermal development and hemangioblast specification upon in vitro differentiation of ES cells. Blood 103:886–889. https://doi.org/10.1182/ BLOOD-2003-06-2149 106. Madisen L, Zwingman TA, Sunkin SM et al (2010) A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat Neurosci 13:133– 140. https://doi.org/10.1038/nn.2467 107. Bain CC, Gibson DA, Steers NJ et al (2020) Rate of replenishment and microenvironment contribute to the sexually dimorphic phenotype and function of peritoneal macrophages. Sci Immunol 5. https://doi.org/10.1126/ sciimmunol.abc4466 108. Livet J, Weissman TA, Kang H et al (2007) Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature 450(7166):56–62. https://doi.org/10.1038/nature06293 109. Ghigo C, Mondor I, Jorquera A et al (2013) Multicolor fate mapping of Langerhans cell homeostasis. J Exp Med 210:1657. https:// doi.org/10.1084/JEM.20130403 110. Muzumdar MD, Tasic B, Miyamichi K et al (2007) A global double-fluorescent Cre reporter mouse. Genesis 45:593–605. https://doi.org/10.1002/DVG.20335 111. Patel SH, O’Hara L, Atanassova N et al (2017) Low-dose tamoxifen treatment in juvenile males has long-term adverse effects
on the reproductive system: implications for inducible transgenics. Sci Rep 7:1–12 112. Ved N, Curran A, Ashcroft FM, Sparrow DB (2019) Tamoxifen administration in pregnant mice can be deleterious to both mother and embryo. Lab Anim 53:630–633 113. Huh WJ, Khurana SS, Geahlen JH et al (2012) Tamoxifen induces rapid, reversible atrophy, and metaplasia in mouse stomach. Gastroenterology 142:21–24.e7 114. Hammad S, Othman A, Meyer C et al (2018) Confounding influence of tamoxifen in mouse models of Cre recombinase-induced gene activity or modulation. Arch Toxicol 92:2549–2561 115. Blum KM, Roby LC, Zbinden JC et al (2021) Sex and Tamoxifen confound murine experimental studies in cardiovascular tissue engineering. Sci Rep 11:8037 116. Sa´nchez-Aguilera A, Arranz L, Martin-Pe´rez D et al (2014) Estrogen signaling selectively induces apoptosis of hematopoietic progenitors and myeloid neoplasms without harming steady-state hematopoiesis. Cell Stem Cell 15: 791–804 117. Igarashi H, Kouro T, Yokota T et al (2001) Age and stage dependency of estrogen receptor expression by lymphocyte precursors. Proc Natl Acad Sci U S A 98:15131. h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / P N A S . 011513098 118. Pepe G, Braga D, Renzi TA et al (2017) Selfrenewal and phenotypic conversion are the main physiological responses of macrophages to the endogenous estrogen surge. Sci Rep 7: 1–14 119. Rojo R, Sauter KA, Lefevre L et al (2018) Maternal tamoxifen treatment expands the macrophage population of early mouse embryos. bioRxiv:296749. https://doi.org/ 10.1101/296749 120. Zhao H, Zhou B (2019) Dual genetic approaches for deciphering cell fate plasticity in vivo: more than double. Curr Opin Cell Biol 61:101–109 121. Fenno LE, Mattis J, Ramakrishnan C et al (2014) Targeting cells with single vectors using multiple-feature Boolean logic. Nat Methods 11:763–772 122. Park JT, Leach SD (2013) TAILOR: transgene activation and inactivation using lox and rox in zebrafish. PLoS One 8:e85218 123. Sajgo S, Ghinia MG, Shi M et al (2014) Dre Cre sequential recombination provides new tools for retinal ganglion cell labeling and manipulation in mice. PLoS One 9:e91435
Macrophage Fate Mapping Models 124. Madisen L, Garner AR, Shimaoka D et al (2015) Transgenic mice for intersectional targeting of neural sensors and effectors with high specificity and performance. Neuron 85:942–958 125. Zhang H, Pu W, Tian X et al (2016) Genetic lineage tracing identifies endocardial origin of liver vasculature. Nat Genet 48:537–543 126. Liu K, Yu W, Tang M et al (2018) A dual genetic tracing system identifies diverse and dynamic origins of cardiac valve mesenchyme. Development 145 127. Plummer NW, Evsyukova IY, Robertson SD et al (2015) Expanding the power of recombinase-based labeling to uncover cellular diversity. Development (Cambridge) 142: 4385–4393. https://doi.org/10.1242/ DEV.129981/-/DC1 128. He L, Li Y, Li Y et al (2017) Enhancing the precision of genetic lineage tracing using dual recombinases. Nat Med 23:1488–1498 129. Liu Q, Liu K, Cui G et al (2019) Lung regeneration by multipotent stem cells residing at the bronchioalveolar-duct junction. Nat Genet 51:728–738 130. Li Y, He L, Huang X et al (2018) Genetic lineage tracing of nonmyocyte population by dual recombinases. Circulation 138:793–805
43
131. Kim J-S, Kolesnikov M, Peled-Hajaj S et al (2021) A binary Cre transgenic approach dissects microglia and CNS border-associated macrophages. Immunity 54:176–190.e7 132. Simic M, Manosalva I, Spinelli L et al (2020) Distinct waves from the hemogenic endothelium give rise to layered lymphoid tissue inducer cell ontogeny. Cell Rep 32:108004. https://doi.org/10.1016/j.celrep.2020. 108004 133. Samokhvalov IM, Samokhvalova NI, Nishikawa S (2007) Cell tracing shows the contribution of the yolk sac to adult haematopoiesis. Nature 446:1056–1061. https://doi.org/10.1038/nature05725 134. Boyer SW, Schroeder AV, Smith-Berdan S, Forsberg EC (2011) All hematopoietic cells develop from hematopoietic stem cells through Flk2/Flt3-positive progenitor cells. Cell Stem Cell 9:64–73. https://doi.org/10. 1016/j.stem.2011.04.021 135. Zhang Y, McGrath KE, Ayoub E et al (2021) Mds1CreERT2, an inducible Cre allele specific to adult-repopulating hematopoietic stem cells. Cell Rep 36:109562. https://doi. org/10.1016/j.celrep.2021.109562
Chapter 3 Studying Autophagy in Microglia: Overcoming the Obstacles Ainhoa Plaza-Zabala and Amanda Sierra Abstract In this chapter, we provide an overview of the main techniques and experimental approaches that can be used to analyze autophagy flux in microglia, the brain-resident macrophages. For this purpose, we first briefly introduce the main peculiarities of microglial biology, describe the basic mechanisms and functions of autophagy, and summarize the evidence accumulated so far on the role of autophagy in the regulation of microglial survival and functions, mainly phagocytosis and inflammation. Then, we highlight conceptual and technical aspects of autophagic recycling and microglial physiology that need to be taken into account for the accurate evaluation of autophagy flux in microglia. Finally, we describe the main assays that can be used to analyze the complete sequence of autophagosome formation and degradation or autophagy flux, mainly in cultured microglia and in vivo. The main approaches include indirect tracking of autophagosomes by autophagic enzymes such as LC3 by western blot and fluorescence-based confocal microscopy, as well as direct analysis of autophagic vesicles by electron microscopy. We also discuss the advantages and disadvantages of using these methods in specific experimental contexts and highlight the need to complement LC3 and/or electron microscopy data with analysis of other autophagic effectors and lysosomal proteins that participate in the initiation and completion of autophagy flux, respectively. In summary, we provide an experimental guide for the analysis of autophagosome turnover in microglia, emphasizing the need to combine as many markers and complementary approaches as possible to fully characterize the status of autophagy flux in microglia. Key words Autophagy flux, Microglia, Autophagosomes, Lysosomes, Western blot, Confocal microscopy, Electron microscopy
1
Microglia and Autophagy: The Generalities Microglia are Central Nervous System (CNS) specialized tissueresident macrophages and, as such, the main immune effectors of the brain parenchyma [1]. In contrast to other major brain cell types, microglia stem from erythromyeloid progenitors of the yolk sac [2] and invade the neuroectoderm-derived developing brain early during embryogenesis [3]. Once in the brain, microglia progenitors respond to brain-specific cues and gradually acquire the
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_3, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
45
46
Ainhoa Plaza-Zabala and Amanda Sierra
CNS-specialized phenotype of mature immune cell [4], contributing to brain-shaping during development and assuring correct brain function during adulthood. Microglia represent around 10% of brain parenchymatic cells, where they efficiently communicate and coordinate with other brain cell types to execute their diverse neuroimmune functions. As the guard cells of the CNS, they constantly scan the brain parenchyma with their fine and motile processes [5, 6]. To this end, microglia express specific sets of receptors that sense and respond to extracellular milieu stimuli [7], including physiological signals and pathological insults. Therefore, through specific interactions with other brain cell types, such as neurons, oligodendrocytes and astrocytes, microglia regulate key brain functions, including neurogenesis, myelination, blood-brain barrier permeability, synaptic turnover, and neuronal connectivity [8], among others. Additionally, as CNS-resident macrophages, microglia are the main phagocytic cells of the brain parenchyma and have the capacity to engulf and degrade apoptotic cells and pathogens and to clear brain-specific extracellular cargo (myelin, neuronal synaptic elements) and toxic forms of proteins (amyloid-beta, alpha-synuclein) [9]. Microglia are also the main regulators of brain inflammation and express and release pro- and anti-inflammatory mediators depending on the context, having the potential to promote brain parenchyma healing or to amplify and perpetuate the inflammatory reaction, feeding forward pathology [10]. All these microglial functions require an intrinsically active and versatile cell that readily scans its immediate environment and adapts in a timely manner to both basal and stress-challenged situations, thus requiring highly efficient and plastic metabolic programs [11, 12]. Autophagy, through the specific recycling of microglial cytoplasmic components, has the potential to regulate metabolic and phenotypic adaptations of microglia to their microenvironment, thus supporting the fundamental neuroimmune actions that deeply affect brain organization and function. Macroautophagy or autophagy is a highly conserved catabolic process that is present in all mammalian cells including microglia. During autophagy, damaged or no longer needed intracellular substrates are enclosed in newly formed autophagic vesicles or autophagosomes, which after subsequent fusion events with acidic endosomes and/or lysosomes, convert into degradative autolysosomes, leading to the degradation of intracellular cargo and efflux of simple biomolecules to the cytosol. Basal levels of autophagy are constitutively active in most cell types and prevent the accumulation of damaged proteins and organelles [13]. In addition, autophagy is up-regulated under stressful situations such as nutrient deprivation, providing the cell with basic biomolecules to fuel on-demand metabolic programs [14]. Thus, autophagy digests a variety of organelles and molecules such as proteins, lipids,
Autophagy Analysis in Microglia
47
Fig. 1 Biochemical markers of autophagosome formation and degradation during the progress of autophagy flux. Autophagy-initiating stimuli usually elicit an inhibition of MTOR that allows the dissociation of ULK1 from the MTOR/ULK1 complex. ULK1, member of the autophagy preinitiation complex, gets activated and starts the biochemical cascade that leads to the sequential action of ATG proteins to form the double membrane autophagosome. First, the phagophore is formed by the action of members of the PtdIns3K complex, which can be tracked experimentally by the presence of ATG16L1 [87] and WIPI2 [88] proteins, among others. During the phagophore extension and closure, LC3 proteins conjugate to the double membrane isolating the autophagosome lumen and serve as indirect biochemical markers of autophagosomes. Subsequently, autophagosomes mature by multiple fusions with acidic endosomes and lysosomes, acquiring lysosomal markers such as LAMP1 and specific cathepsins. During maturation of autophagosomes, LC3-family proteins of the outer membrane are deconjugated by specific enzymes, while LC3-family proteins of the inner membrane are degraded together with the cargo in the autolysosome. Autophagy flux progression analysis can also be complemented with the analysis of autophagic substrates such as the selective autophagy receptor SQSTM1, which binds to LC3 proteins in the inner autophagosome membrane and recruits posttranslationally modified cargo marked for degradation. SQSTM1 is degraded together with cargo in the autolysosome
carbohydrates, and nucleic acids, maintaining the abundance and quality of cellular components and regulating the metabolic status of the cell [15]. These are fundamental functions for the maintenance of cellular physiology. Therefore, the tight regulation of autophagic turnover is applied both to autophagosome formation [16, 17] and degradation [18] (Fig. 1). Autophagosome formation usually starts after metabolic and/or stressful stimuli elicit an inhibition of the autophagy inhibitor MTOR (mechanistic target of rapamycin 1 kinase), the protein kinase that forms a complex with ULK1 (unc-51 like autophagy activating kinase 1) [19, 20]. Inhibition of MTOR allows the dissociation from the complex and activation of ULK1, the gatekeeper that promotes activation of the PtdIns3K (phosphatidylinositol 3-kinase) complex, to de novo generate the double membrane that will enclose the autophagosome [21]. Once the isolation
48
Ainhoa Plaza-Zabala and Amanda Sierra
membrane has been established, the autophagosome grows and completes its formation by the coordinated and sequential action of specific enzymatic effectors known as autophagy-related proteins (from autophagy-related genes, ATGs) [16, 17]. During this process, MAP1LC3/ATG8 (microtubule associated protein 1 light chain 3) (LC3) family protein members conjugate to the whole extent of the inner and outer autophagosomal membranes [22]. In addition, LC3 proteins of the inner autophagosomal membrane bind to specific autophagy receptors, such as SQSTM1 (sequestosome 1), to selectively load intracellular cargo to their lumen while the double membrane of the autophagosome is extending [23]. After the membranes close around the cargo, autophagosomes progressively mature through multiple fusion events with acidic endosomes and/or lysosomes, losing their inner membrane and forming the degradative autolysosome [18]. During this process, autophagosomes considerably decrease their LC3 content because the outer membrane LC3 is directly deconjugated by specific proteases and the inner membrane, which contains LC3, is degraded with lumenal cargo in the autolysosome [22]. Therefore, LC3 is one of the most commonly used indirect biochemical markers of autophagosomes [24, 25]. In agreement with the core functions of autophagy in the maintenance of cellular fitness, autophagy dysregulation contributes to the onset and progression of diverse CNS pathologies [26]. Indeed, the lack of basal autophagy in neurons spontaneously induces neurodegeneration in young mice [27, 28]. However, the impact of autophagy dysregulation on microglial functions and how this may relate to healthy brain aging and pathology has only started to be discovered [29, 30].
2
Autophagy: Regulation of Microglial Cell Biology During the past few years, a considerable amount of knowledge has been gathered on microglial autophagy and the impact of its genetic and/or pharmacological modulation in microglial survival, inflammation, and phagocytosis [29, 30]. In the following paragraphs, we will first briefly summarize what is known so far on the role of autophagy in the regulation of microglial survival, inflammatory response, and phagocytosis. We will then discuss other potential microglial functions that may be affected by autophagy (dys)regulation and propose new lines of research that may help unveil the full potential of autophagic turnover on the regulation of microglial functions. The impact of autophagy on microglial survival is still not clear, although it is well known that genetic deficiency of individual ATG proteins in neural precursor cells induces spontaneous neurodegeneration [27, 28]. Similarly, mice bearing systemic constitutive
Autophagy Analysis in Microglia
49
down-regulation of autophagy exhibit reduced microglial density in the hippocampus [31]. However, the in vivo deletion of single ATG proteins, specifically in microglia, does not affect microglial density in a variety of brain regions and time points analyzed [31– 34], suggesting that microglia rely less on basal autophagic activity than neurons for survival. However, the pharmacological inhibition of basal autophagic turnover in vitro reduced microglial viability in a concentration and time-dependent manner [31], suggesting that the recruitment of compensatory mechanisms and the degree of autophagy inhibition may determine the outcome of autophagy dysregulation in microglial survival. Autophagy modulates inflammatory cytokine expression and release in microglia. Currently, the most accepted idea is that autophagy deficiency in microglia increases the expression of proinflammatory cytokines through dysregulation of NLRP3 inflammasome levels and/or activity [35–38], which leads to increased production of interleukin-1 beta (IL-1β) and related cytokines [39]. The microglial proinflammatory phenotype induced by defective autophagy is most evident in pathological contexts after challenging microglia with inflammatory stimuli such as amyloid-beta [37, 38, 40], alpha-synuclein [36, 41] or bacterial lipopolysaccharides [42–44] both in vitro and in vivo. Nevertheless, it is unclear whether autophagy deficiency alone in the absence of a pathological stressor is enough to shift microglia to an inflammatory phenotype. Indeed, in vivo [32, 33, 36, 37, 41] and in vitro [35, 42, 45–47] models of autophagy deficiency exhibit variable outcomes on basal microglial inflammation. On the other hand, autophagic activity can promote inflammation in certain pathological situations [48– 51] or have no effects on the microglial inflammatory profile [34]. Therefore, autophagic turnover has the potential to affect microglial inflammatory phenotype in a complex manner, which most probably depends on the microglial cell’s intrinsic state but also the extrinsic pathological environment. Autophagy also impacts the function of lysosomal digestion pathways, such as phagocytosis and endocytosis. However, disentangling this interaction is complex due to the molecular overlapping [52, 53] and cross-talk [54, 55] of lysosomal digestion pathways at the intracellular level. This issue has led to the differentiation of two distinct autophagy-related processes in microglia: canonical and noncanonical autophagy, based on the nature of the cargo digested by vesicles containing autophagy effectors. Thus, canonical autophagy refers to classical autophagic turnover whereby cytosolic components are digested through the formation and clearance of an autophagosome, whereas noncanonical autophagy refers to the use of specific autophagic effectors to deliver phagosomal and endosomal cargo into the lysosome. Canonical autophagy in microglia has been linked to the regulation of phagocytosis of apoptotic cells in the neurogenic niche of the
50
Ainhoa Plaza-Zabala and Amanda Sierra
hippocampus [31], wherein newborn cells naturally undergo apoptosis and are efficiently phagocytosed by microglia in physiological conditions [56]. Similarly, autophagy deletion in myeloid cells regulates synaptic material degradation by microglia and modulates social behaviors [57]. On the other hand, noncanonical autophagy in microglia regulates uptake and clearance of amyloid beta and myelin by microglia, affecting disease progression in mouse models of Alzheimer’s disease [52] and multiple sclerosis [53], respectively. Interestingly, autophagic turnover also participates in the clearance of brain extracellular debris. Thus, autophagosomes engulf and degrade neuron-released alpha-synuclein and prevent neurodegeneration in an alpha-synuclein overexpression mouse model of Parkinson’s disease [33]. Overall, these data indicate that canonical autophagy and ATG proteins performing noncanonical functions impact microglial uptake and degradation of different types of brain extracellular debris. Other less explored roles have also been reported for autophagy in microglia. For example, autophagy deficiency in microglia alters their capacity of regulating the density of mature oligodendrocytes, myelination markers, and the lengths of the nodes of Ranvier in a variety of brain regions including the corpus callosum, cingulate cortex, and striatum, increasing seizure susceptibility [32]. Therefore, autophagy and its effector proteins may potentially be involved in diverse and essential functions of microglia. However, there is still a lack of basic understanding of how much microglia rely on autophagic digestion to perform their basal surveying activities, how remodeling of specific cytoplasmic components by autophagy may contribute to microglial phenotypic plasticity, and which autophagy-related genes are essential or dispensable for microglial functions and survival. In addition, it is not known which are the relative contributions of autophagy, phagocytosis, and endocytosis to lysosomal digestion in microglia and which mechanisms regulate their intracellular cross-talk, among others. Thus, it is important to understand the basic peculiarities of microglial biology and autophagy assessing techniques, to optimize the knowledge gathered from these new experiments.
3 Technical Aspects: What You Need to Know for Correctly Analyzing Autophagy in Microglia The experimental monitoring of autophagy flux or the lysosomal turnover of autophagosomes is complicated due to the complex and stepwise biochemical regulation of autophagy and its dynamic nature. Over the years, diverse experimental methods have been developed to capture the complexities of autophagy flux. These experimental methods and their interpretation are explained in
Autophagy Analysis in Microglia
51
detail in a set of updated guidelines [24] written up by researchers in the autophagy field once every approximately 4 years. One of the main messages of the guidelines is that no single experimental method is sufficient to provide an overall picture of autophagic turnover. Thus, the guidelines recommend using complementary methods that address different steps or aspects of autophagic recycling, helping to interpret the regulation of autophagy flux more accurately in different experimental conditions. In the next section, we will summarize key points that need to be considered when experimentally analyzing autophagy in microglia. 1. Autophagy is a dynamic mechanism. Autophagic turnover consists of the formation and degradation of autophagosomes, the specialized double-membrane vesicles that engulf and deliver the cargo to lysosomes. The autophagosomes are de novo and gradually formed each time that an autophagic cycle starts thus increasing their numbers after autophagy induction. When autophagy flux evolves correctly, autophagosomes fuse with endosomal and/or lysosomal vesicles forming the autolysosome, characterized by having a single-membrane and degradative acidic hydrolases, which finally disappears after degradation of cargo. Thus, static measures related to the autophagosome content of the cell at a certain time point can produce misleading results that are far from the actual regulation of autophagic turnover. 2. Autophagy is regulated in a time-dependent manner. The levels of autophagic recycling in the cell respond to time-dependent changes in metabolic and stress-related internal and external cues. Therefore, it is important to analyze the variations of autophagic turnover in response to a specific type of stimulus over a time course. For example, amino acid starvation elicits an acute up-regulation of autophagy flux that increases protein degradation in autolysosomes and amino acid efflux to the cytosol, which leads to subsequent MTOR activation and autophagy flux down-regulation [58], thus producing a biphasic response of autophagy flux to starvation across time. 3. Microglia are professional phagocytes. As professional brain phagocytes, microglia highly rely on their lysosomal degradative compartment, which is the end target of intracellular vesicular transport systems, including autophagy, phagocytosis, and endocytosis. Notably, some autophagic effectors, including LC3, have been shown to play roles in phagocytic [59] and endocytic [52] lysosomal degradative pathways in certain conditions. This may represent a limitation or bias that must be considered when assessing autophagy in microglia exposed to, for example, extracellular phagocytic debris. The reliable dissociation of autophagy from phagocytosis and endocytosis at the
52
Ainhoa Plaza-Zabala and Amanda Sierra
intracellular level in these situations is best performed by transmission electron microscopy (TEM) ultrastructural analysis, although other indirect biochemical markers of canonical autophagy may also serve to complement this data. 4. Microglial physiology may differ in vitro and in vivo. To understand the molecular biology underlying the regulation of autophagosome biogenesis and cellular mechanisms of autophagy execution in microglia, experiments in cultured primary microglia, induced pluripotent stem cell (iPSC)-derived microglia and/or cell lines may be convenient. However, to understand the impact of autophagic recycling in actual microglial functions in the brain, it is more adequate to implement autophagy analysis in more complex ex vivo and in vivo systems. The analysis of autophagosome clearance or autophagy flux in microglia consists of the use of the classical experimental methods accepted in the field for other cell types, which are mainly based on indirect tracking of autophagosome markers such as LC3 by western blot (WB) and/or fluorescent-labeling coupled to confocal microscopy. As a complement, autophagic vesicles (autophagosomes and autolysosomes) can be visualized and analyzed directly by TEM. Analysis including all three methods may be the optimal strategy to thoroughly understand the dynamics of autophagy flux in microglia.
4
Experimental Techniques Used to Analyze Autophagy Flux in Microglia
4.1 Western Blotting of Autophagy-Related Proteins 4.1.1
In Vitro Analysis
The classical method to assess autophagy flux in culture by WB is the LC3 turnover assay (Fig. 2). LC3-family proteins are covalently lipidated to the autophagosomal double membrane while forming and extending, and remain conjugated to the inner membrane until maturation to autolysosomes, so they can serve as an indirect measure of the newly formed autophagosomes under different experimental conditions. However, as autophagy flux involves formation of autophagosomes and their subsequent degradation, LC3 protein quantities in a specific time point will only refer to the pool of autophagosomes present in that specific moment in the cell, and information about the ones formed but already degraded will be lost (Fig. 2a). Indeed, low levels of LC3 after the autophagic stimulus may mean inhibition or induction of autophagy, depending on the velocity of flux [60]. On the contrary, high levels of LC3 may indicate autophagy induction or blockade, the former due to increased autophagosome biogenesis and the latter due to impairments in lysosomal fusion or degradation [60]. To overcome this issue, LC3 turnover assays are designed to compare the amounts of membrane-bound LC3 in cells exposed or not to the autophagic stimulus in the presence and absence of lysosomal inhibitors
Autophagy Analysis in Microglia
53
Fig. 2 LC3 turnover assay by WB can be used to estimate autophagosome formation and degradation in microglia. (a) Microglia in control (EXP-) and experimental stimulus (EXP+) conditions are harvested and processed for LC3 protein western blotting. When the autophagy flux is induced, new autophagosomes that will deliver intracellular cargo to the lysosome start to form. During autophagosome biogenesis, the cytosolic soluble form of LC3 protein (LC3-I) binds to certain lipids of the autophagosomal membrane and converts into conjugated LC3 (LC3-II). LC3-II decorates the whole autophagosomal membrane during its extension, cargo loading and closure, which is the most widely used indirect marker of autophagosomes. Once formed, the autophagosome travels through the cytosol and matures by subsequent fusions with acidic endosomes and lysosomes. During maturation, LC3 of the outer membrane is deconjugated by specific ATG proteases and inner membrane LC3-II is degraded together with cargo, eventually leading to LC3-II devoid degradative
54
Ainhoa Plaza-Zabala and Amanda Sierra
(Fig. 2a). The drugs most commonly used to inhibit the lysosomal turnover of autophagosomes include bafilomycin A1, chloroquine, and ammonium chloride, which inhibit lysosomal acidification and/or fusion with the autophagosome [24, 61–63]. The comparison of lipidated LC3 levels in cells treated and nontreated with lysosomal inhibitors allows to extract information on autophagosome turnover, that is, how many autophagosomes would have been degraded in control and experimental stimulus conditions in a certain period of time (Fig. 2c). The LC3 turnover assay provides information on the overall changes in the levels of autophagy flux. To confirm and complement these results, LC3 protein analysis can be combined with the evaluation of other autophagic effectors (Fig. 1) and/or their posttranslational modifications. For example, to gain insight into the mechanisms of initiation of the autophagic process, phosphorylation of certain residues in core upstream regulators such as ULK1 can be valuable [64]. It is also common to use autophagic substrates such as the selective autophagy receptor SQSTM1 [65] to estimate cargo degradation by autophagy. Nevertheless, SQSTM1 participates in other nonautophagic cellular pathways, such as degradation by the ubiquitin-proteasome system, cellular metabolism, signaling, and apoptosis [66], and may not be a specific autophagy marker in some contexts. In addition, SQSTM1 is significantly regulated at the transcriptional level in certain cell types [65], including microglia [67]. Thus, SQSTM1 protein levels may not always reflect actual autophagic cargo flux, and data derived from its analysis should be interpreted with caution. Finally, the lysosomal response to the autophagic stimulus can be evaluated by estimating lysosomal numbers by structural protein levels such as LAMP1 (lysosomal associated membrane protein 1) and/or ä Fig. 2 (continued) autolysosomes. Thus, LC3-II levels increase during autophagosome formation and decrease during autophagosome degradation. When using LC3 turnover assay, lysosomal inhibitors are added to EXP- and EXP+ conditions to calculate the amount of autophagosomes that would have been degraded during the experimental period of time. (b) Representative blot showing LC3-I (soluble form, 16–18 KDa) and LC3-II (membrane-conjugated form, 14-16 KDa) bands stained with anti-LC3B antibody (NB1002220, Novus Biologicals), and β-actin (housekeeping protein, 42 KDa) bands stained with anti-β-actin antibody (A5316, Sigma-Aldrich) in EXP- and EXP+ microglia in the presence and absence of the lysosomal inhibitor bafilomycin A1 (BAF; 100 nM, 6 h) (Selleckchem, S1413). (c) Left graph shows the raw data (LC3-II/β-actin) obtained after quantification of the intensity of protein bands from LC3 turnover assay blots. Right graph shows classical flux data (BAF+-BAF-) in EXP- and EXP+ conditions obtained from raw data. (d) Autophagosome formation and degradation can be estimated (LC3-II/β-actin) for EXP- (basal) and EXP+ conditions from simple calculations applied to raw data. (e) Left graph shows the estimation of autophagosome formation and degradation (LC3-II/β-actin) for EXP- (basal) and EXP+ conditions applying the calculations described in (d). Right graph shows the output of the calculation of the autophagosome formation ratio (FormR), degradation ratio (DegR), and net autophagosome turnover ratio (NetR) using the formulas specified below. (This figure is adapted from [60])
Autophagy Analysis in Microglia
55
lysosomal function by enzymes like cathepsins. In summary, LC3 turnover assay, combined with other autophagic markers, is one of the most commonly used methods to analyze autophagy flux in cell cultures. In the next section, we will discuss the advantages and disadvantages of using LC3 turnover assay in autophagy research. Advantages
The widespread use of LC3 turnover assay mainly stands on the technical and economic affordability of WB experiments. When turning to the scientific advantages, it is important to highlight that WB of LC3 protein is the only technique that allows the reliable discrimination between cytosolic (soluble) and autophagosomal (membrane-bound) forms of LC3 (Fig. 2b). This distinction is based on evident molecular weight differences between the cytosolic (LC3-I, 16-18KDa) and the lipidated forms of LC3 (LC3-II, 14-16KDa). Thus, WB allows to extract information on LC3-I dynamics, which could theoretically provide information on LC3-I conversion to LC3-II, the actual autophagosome marker. Nevertheless, LC3-I to LC3-II conversion is usually not controlled by a straightforward relationship. Indeed, LC3-I levels could be modified by other factors not related to its conversion to the lipidated LC3-II form, including transcriptional changes [67] and degradation by the ubiquitin-proteasome system [68]. Therefore, it is not recommended to normalize LC3-II levels to LC3-I, but instead to housekeeping loading controls such as β-actin or β-tubulin [24]. In summary, WB allows to distinguish between the cytosolic and membrane-bound forms of LC3, which may be useful, but not critical, to understand the autophagic status of the cell more thoroughly in certain situations.
Limitations
Although the LC3 turnover assay has extensively been used to assess autophagy flux, this method is not without weaknesses. Indeed, protein expression analysis by classical WB presents inherent limitations such as lack of reproducibility and high variability among experiments, producing qualitative or semiquantitative data in best cases. Currently, with the newest WB technologies, it may be possible to obtain real quantitative data using accurate experimental procedures that maintain the linear range of protein band intensities [69]. However, quantitative analysis of WB data is still not a reality in autophagy research literature. The analysis of LC3 turnover by WB estimates differences in the lysosomal degradation of autophagosomes between the basal and the experimental condition in a certain period of time (Fig. 2a). To obtain reliable results, it is important to carefully select the lysosomal inhibitor used, its saturating and non-toxic concentrations at the specific experimental condition, and the time needed to achieve such effects in pilot experiments, since these parameters may vary depending on the cell type. In microglia, the most commonly used lysosomal inhibitor has been bafilomycin A1 so far,
56
Ainhoa Plaza-Zabala and Amanda Sierra
although at dissimilar concentrations and timings [37, 38, 46, 47, 52, 67, 70–72]. Additionally, it is not recommended to incubate mammalian cells for more than 4–6 h with lysosomal inhibitors [24], due to increasing probabilities of unspecific effects at the initiation steps of autophagy and possible toxicity, which is a reason for the invalidation of the experimental data. Given the variability between labs on the interpretation of data coming from this type of assay, individual and collective efforts have been carried out to standardize analysis and optimize extraction of reliable information from LC3 turnover assays [24, 73]. Interestingly, a simple analytic method has recently been described to systematically extract information on the formation and degradation of autophagosomes in microglia from simple calculations applied to raw data obtained by LC3 turnover assay [60] (Fig. 2c–e), which could also be applied for other cell types. This model assumes that the rate of autophagosome formation is equal to autophagosome degradation during steady-state or control conditions, i.e., the net autophagic turnover or autophagosome recycling is constant in basal situations. This allows calculating how autophagosome formation and degradation change after the experimental stimulus, expressed in ratios (Fig. 2e). By comparing these ratios, the net autophagic ratio value can be calculated after the experimental stimulus (Fig. 2e), which gives an idea of how proportional the change on autophagosome formation with respect to degradation is, and thus whether autophagy flux is sustainable across time after application of the experimental stimulus. This is an important issue when drawing scientific conclusions on the effects of physiological and pharmacological stimuli since slight acute instability in autophagic turnover may lead to pathological disequilibrium in the long term, which will impact microglial function and survival. Finally, and as mentioned in previous sections, LC3 protein has other nonautophagy-related functions in vesicular trafficking processes, such as certain forms of phagocytosis, endocytosis, and exocytosis [74]. Therefore, depending on the experimental design, it may not be adequate to use LC3 turnover assay for autophagy flux tracking. One such example could be exposing microglia to extracellular cargo amenable to phagocytosing or endocytosing, such as apoptotic cells, amyloid-beta, alpha-synuclein, synaptosomes, or myelin, among other cargos. In these specific cases, it would be preferable to analyze autophagy flux by direct visual methods such as TEM to discriminate among lysosomal digestion pathways and ensure the specific assessment of autophagic turnover.
Autophagy Analysis in Microglia 4.1.2
In Vivo Analysis
4.2 Fluorescence and Confocal Microscopy-Based Methods 4.2.1
In Vitro Analysis
57
The accurate determination of in vivo autophagy flux in microglia using the LC3 turnover assay is virtually impossible. Indeed, this approach would require the systemic or intracerebral administration of lysosomal inhibitors, which may affect other cell type functions or induce toxicity and, in turn, impact microglial autophagy flux. In addition, specific processing of microglial proteins for WB would entail microglia isolation from brain tissue, probably having to pool a significant number of animals to obtain one sample. On the other hand, microglial isolation by different sorting methods may, on its own, affect microglial autophagic response [75, 76]. Thus, results obtained from LC3 turnover assays in vivo should be interpreted cautiously due to the high probability of yielding-biased results. Fluorescence and confocal microscopy-based methods to track autophagy flux in culture are experimentally similar to WB. Classical autophagosome imaging procedures include immunostaining of endogenous or plasmid-delivered LC3-encoding probes in the presence and absence of lysosomal inhibitors in fixed cells (Fig. 3a), which can be conceptually comparable to LC3 turnover assay by WB. However, imaging methods offer more experimental versatility than WB. Indeed, fluorescence-based microscopy methods allow live imaging of cells across time while tracking their autophagy flux. For live visualization of autophagic turnover, cells can be transfected with autophagy-tracking LC3-based fluorescent plasmid reporters [24, 25] or treated with specific autophagic vesicle binding probes [77]. In the following paragraphs, we will summarize the main fluorescent-reporter plasmids and chemical probes that can be used for image-based analysis of autophagy flux in live or fixed conditions: 1. Tandem RFP-GFP-LC3 plasmid: The best option to track autophagy flux by fluorescence-based imaging methods without using lysosomal inhibitors is by transfecting tandem RFP and GFP tagged LC3 plasmid or similar constructs [78] (Fig. 3b). These plasmids allow for fixed and live imaging of autophagy flux in microglia by taking advantage of the acidsensitive quenching of the GFP protein after the formation of the autolysosome. Thus, tandem RFP-GFP-LC3 transfected microglia autophagosomes appear yellow as a result of the expression of both red and green fluorescence, whereas autolysosomes appear red because green fluorescence is quenched by acidification after lysosomal fusion. The main advantage of this plasmid is that presumable secondary off-target effects of lysosomal inhibitors in the microglial autophagic response are eliminated. However, lysosomal inhibitors can be used as specific internal controls for the correct function of the expression plasmid.
58
Ainhoa Plaza-Zabala and Amanda Sierra
Fig. 3 Autophagy flux can be analyzed by confocal microscopy-based fluorescent tracking of autophagic effectors and substrates. (a) The panel shows immunostaining of endogenous LC3 protein in transgenic EGFP (enhanced green fluorescent protein) expressing primary microglia treated with vehicle and bafilomycin A1 (BAF; 100 nM, 3 h) (Selleckchem, S1413). Confocal microscope (Leica SP8) images were acquired after immunostaining of EGFP (green) (GFP-1020, Aves Labs) and LC3B (red) (NB100-2220, Novus Biologicals) in fixed primary microglia samples. Nuclei were identified by DAPI (4′-6-diamidino-2-phenylindole) (gray) (D9542-10 mg, Sigma-Aldrich). Images are shown in a single microglial z-plane or the whole microglial volume scanning z-stack (7.2 μm). LC3-I shows a diffuse pattern of staining while LC3-II is observed as puncta. Scale bars = 5 μm. (b) The panel shows BV2 microglia transfected with mRFP-GFP-LC3 plasmid (addgene 21074) [78], allowed for 24 h expression and imaged by a Zeiss LSM 880 Airyscan microscope at a single z-plane after fixation and DAPI staining of nuclei (white). LC3-I shows a diffuse pattern of staining while LC3-II is observed as puncta. Autophagosomes express yellow LC3 puncta (combination of RFP and GFP,
Autophagy Analysis in Microglia
59
2. GFP-LC3-RFP-ΔLC3 plasmid: An alternative to control for differential transfection and expression of fluorescently tagged LC3 vectors in individual cells is the second generation tandem flux GFP-LC3-RFP-ΔLC3 reporter that can also be used in live and fixed conditions [79]. This protein construct is cleaved in the cytosol by ATG4 type proteases in functional GFP-LC3 and nonfunctional RFP-ΔLC3. Thus, GFP-LC3 is conjugated to autophagosomal membranes during autophagic activity, whereas RFP-ΔLC3 is unable to conjugate to vesicular membranes and remains in the cytosol, serving as an internal plasmid expression control. Additionally, when autophagosomes mature to autolysosomes, the GFP tagged to membranebound LC3 is quenched, providing information on flux, that is, the formation of autophagosomes and their clearance in lysosomes. When using this reporter, analysis of autophagic activity is performed on a per-cell basis by calculating the GFP:RFP ratio, which decreases when autophagy flux is enhanced in the cell. So far, this reporter assay has not been used to test autophagy flux in microglia. 3. RFP-GFP-SQSTM1 plasmid: Experiments with LC3 tracking reporters can be complemented with autophagic substrate/ receptor monitoring vectors such as RFP-GFP-SQSTM1 [80], which work similarly to LC3 vectors. Thus, SQSTM1 located in autophagosomes leads to yellow fluorescence (combination of GFP and RFP). When the autophagosome fuses with lysosomes, the acidic environment quenches GFP, leading to red (RFP) fluorescence emission. This SQSTM1 reporter can be useful in experimental conditions wherein LC3 proteins significantly bind to single-membrane vesicles, such as endosomes and phagosomes, and, thus, are not appropriate to assess autophagy flux. Indeed, a few reports have shown that SQSTM1 does not significantly bind to single-membrane vesicles in macrophages [81, 82] and, thus, could be used together with other experimental approaches to assess autophagy flux in phagocytes such as microglia. ä Fig. 3 (continued) arrows) while mature autophagosomes or early autolysosomes express red LC3 puncta (RFP only, GFP fluorescence quenched by the acid pH, arrowheads). Changes in RFP and GFP fluorescence of LC3 puncta are usually subtle and not caught by the human eye. GFP/RFP fluorescence ratio can be calculated by using image analysis tools such as FIJI to estimate autophagy flux. Scale bars = 10 μm. (c) The panel shows endogenous LC3 (yellow) and LAMP1 (magenta) puncta immunostaining in the dentate gyrus of transgenic mice expressing EGFP in microglia (cyan). Fixed 50 μm thick hippocampal sections were immunostained for EGFP (cyan) (GFP-1020, Aves Labs), LC3B (yellow) (NB100-2220, Novus Biologicals), and LAMP1 (magenta) (AB528127, DSHB) and imaged in a confocal microscope (Leica Stellaris 5). This type of images can be analyzed using image analysis software such as FIJI. Specifically, EGFP-expressing microglia can be segmented, and LC3 and LAMP1 puncta numbers, size, fluorescence intensity, and colocalization can be analyzed [31]. Scale bars = 15 μm; z = 20 μm
60
Ainhoa Plaza-Zabala and Amanda Sierra
4. Similar to the SQSTM1 plasmid, other expression vectors not relying on LC3 tracking, such as mt-Keima (live) [83, 84] and mt-QC (fixed and live) [85, 86], have been developed to specifically assess mitochondria-selective autophagy or mitophagy. 5. As a complementary approach to autophagy flux reporter plasmid transfection, autophagic-vesicle labeling fluorescent probes can be used to track autophagy flux in microglia. Among the most popular ones is the cytoID commercial kit [77], which fluorescently labels preautophagosomal membranes, autophagosomes, and autolysosomes in live cells without distinction. The fluorescent plasmid reporters and probes can also be analyzed by flow cytometry instead of confocal microscopy [24]. The main advantage of flow cytometry is the high number of cells that can be analyzed per each experimental condition and session. Nevertheless, the visual and spatial information obtained in microscopy-based experiments is lost after flow cytometry, although it may serve as an appropriate complementary experimental method. Similar to WB experiments, LC3 imaging analysis can be supplemented with the concomitant evaluation of phase-specific autophagic effectors (Fig. 1), which allows a more in-depth understanding of stage-dependent aspects of autophagic turnover. When using fixed cells, LC3 immunostaining can be combined with fluorescent staining of phagophore-enriched ATG proteins such as ATG16L1 (autophagy related 16 like 1) [87] and WIPI2 (WD repeat domain, phosphoinositide interacting 2) [88], which allow distinction between extending phagophores and mature autophagosomes. Analysis of the newly forming phagophores may help to discriminate between autophagy induction and blockade in situations of increased numbers of autophagosomes. In addition, staining of selective autophagy substrates such as SQSTM1 [89] in combination with LC3 may be useful to assess autophagy flux progression in certain contexts [24, 25]. Finally, LC3 can also be combined with lysosomal markers such as LAMP1 and/or specific cathepsins to estimate the fusion efficiency between autophagosomes and lysosomes [90] and gain insight into the final stages of autophagy flux. Notably, when imbalance is suspected in any step of autophagy flux, it is important to analyze as many markers as possible for each phase or to complement with different experimental methods because intracellular markers such as LC3 [74], SQSTM1 [66], or LAMP1 [91] play roles and stain cellular compartments beyond autophagic lysosomal digestion. In case of assessing selective removal of specific cellular components such as mitochondria by autophagy (mitophagy), LC3 and other markers described above can be co-immunostained with markers for the
Autophagy Analysis in Microglia
61
assayed cellular component, for example, translocases of outer and inner mitochondrial membranes (TOM/TIM) in mitochondria, among others [92]. Alternatively, when performing live experiments, subcellular components such as mitochondria and lysosomes can be tracked by organelle-specific fluorescent plasmid reporters or chemical probes [24, 25]. Overall, fluorescent staining and confocal microscopy-based methods offer the possibility of independently tracking different stages of autophagic turnover, which may facilitate the understanding of the general regulatory effect of the experimental stimulus over autophagy flux. Advantages
One of the main advantages of fluorescence and imaging-based methods is that spatial information on the relative localization of autophagic and lysosomal compartments within the cell is obtained, which may facilitate the interpretation of autophagy flux dynamics. Similar to WB, although maybe not that straightforward, cytosolic LC3-I and membrane-bound LC3-II can be distinguished after LC3 immunostaining in the cell. Indeed, LC3-I produces a diffuse staining pattern while LC3-II appears as discrete puncta (Fig. 3a, b). An advantage of imaging LC3 is that the experimenter obtains quantitative information on the variation of the size and number of LC3-II puncta (presumable autophagosomes) as well as the size of microglial cytoplasm, which allows to estimate the proportion to which autophagic compartments retract or expand after the experimental stimulus. To reliably estimate autophagy flux in microglia using this methodology, it is advisable to image whole-cell z-stacks [93] of microglia at high resolution with confocal microscopy so that individual LC3-II puncta can be optimally distinguished across the whole microglial volume (Fig. 3a). The quantification of LC3-II puncta alone or in combination with other autophagic markers can be performed by using FIJI custom scripts such as LC3_LAMP_quant (freely available at https://github.com/SoriaFN/Tools/) [94], which enables segmentation of microglia as well as the cytoplasmic LC3 puncta in fluorescently labeled confocal z-stacks. In contrast to WB, the analysis of these images provides quantitative data on the number, size, and subcellular localization of autophagosomes per cell in each experimental condition [31]. Another remarkable advantage of fluorescence-based imaging methods is that they allow for real-time analysis of variations in autophagy flux. In the case of fluorescent reporter plasmids, they permit real-time evaluation of the maturation of autophagic vesicles and their movement through the cytosol after application of the experimental stimulus. Of note, most fluorescent reporter plasmids avoid the necessity of treating cells with lysosomal inhibitors (Fig. 3b) and discard secondary effects of these drugs on autophagic turnover, which may be an advantage when working with moderate to high transfection efficiencies.
62
Ainhoa Plaza-Zabala and Amanda Sierra
Limitations
Although the use of fluorescent-reporter plasmids conveys several advantages, they also present experimental limitations. For example, tracking endogenous LC3 in fixed cells by immunostaining discards experimental artifacts derived from variable plasmid transfection and/or overexpression, especially in hard-to-transfect cells such as microglia. In addition, the use of tandem plasmids in microglia is restricted by the relatively large size of the plasmid and the intrinsic difficulties to achieve high transfection efficiencies in microglial cells. Indeed, plasmid transfection experiments will usually require the use of immortalized microglial cell lines to enhance efficiency. In addition, the most commonly used transient transfection of the plasmid (in relation to the permanent transfection) may produce more variable expression of the tagged LC3 protein, which may account for differences in basal and experimental stimulus-induced expression of the plasmid in each cell, which if not properly handled, may lead to data interpretation bias. However, several reports show effective analysis of autophagy flux in microglial cells using the tandem RFP-GFP-LC3 plasmid [31, 40, 44, 71, 95]. Although fluorescence and confocal microscopy methods may also be used to decipher the specific cargo that is being digested by autophagy in a specific experimental context, this approach provides only a first hint on what may be the actual cargo that autophagy is delivering to the lysosome. Indeed, ultrastructural analysis by TEM would be the experimental method of choice for final confirmation of the type of autophagic cargo digested. Another limitation is that fluorescent probes to label autophagic compartments, such as CytoID, poorly discriminate between autophagic vesicles and other vesicles of the endolysosomal pathway, including phagosomes, endosomes, and lysosomes, which may represent a significant confounding factor in some experimental settings in microglia. Thus, this type of probes should be used with caution and as a complementary approach to other techniques that more accurately measure autophagy.
4.2.2
The analysis of autophagy flux in vivo is mainly performed by immunostaining of autophagic markers in fixed tissue (Fig. 3c). Similar to culture experiments, LC3 immunostaining can be combined with other stage-dependent markers of autophagy flux (Fig. 1). For example, coimmunostaining of LC3 and SQSTM1 can serve as an estimate of cargo loading to autophagosomes and digestion into lysosomes. When combining LC3 with LAMP1 (Fig. 3c), information on fusion efficiency between autophagosomes and lysosomes can be obtained. Summing up the data obtained from these experiments allows the interpretation of changes in autophagy flux occurring at the experimental condition tested. Alternatively, transgenic mice expressing the GFP-LC3 [96] or RFP-GFP-LC3 [97, 98] construct constitutively have been
In Vivo Analysis
Autophagy Analysis in Microglia
63
engineered, which may be adequate to analyze autophagy in microglia. Nevertheless, the limitation of GFP-LC3 mice is that they do not provide information on flux since LC3 staining only informs on the number of autophagosomes present in the cell at a specific moment without taking into account the rates of autophagosome biogenesis and degradation. This obstacle is overcome by RFP-GFP-LC3 mice that possess the advantage of expressing yellow autophagosomes and red autolysosomes (after acid-sensitive quenching of GFP) in different cell types, facilitating LC3 flux estimation from autophagosomes to autolysosomes in vivo. Nevertheless, no reports have been published so far on the use of these mice to analyze autophagy flux in microglia in vivo. 4.3 Transmission Electron MicroscopyBased Methods 4.3.1
In Vitro Analysis
The direct observation and distinction of autophagic vesicles are only possible by electron microscopy-based imaging methods (Fig. 4). TEM exploits the main structural characteristic of autophagosomes, i.e., the presence of a double-membrane surrounding the lumen of the vesicle, to distinguish autophagic compartments from endocytic, phagocytic, and lysosomal vesicles, all of them defined by the presence of a single-membrane, among other characteristics. In addition, TEM allows for the ultrastructural identification of the cargo contained in vesicles, which may also be helpful to distinguish autophagic vesicles of distinct maturational stages. For example, in early nondegradative autophagic vesicles or autophagosomes, intact cytoplasmic portions or specific organelles may be distinguished. On the other hand, in degradative autophagic vesicles or autolysosomes, nondistinguishable electron-dense components may be observed. However, given the continuous nature of vesicle maturation and the presence of structurally similar vesicles in the cell coming from non-autophagic sources (e.g., endocytosis and phagocytosis), it is usually not easy to differentiate between autophagosomes and autolysosomes [99]. The combination of TEM analysis with immune-EM staining of different stagedependent autophagy markers such as LC3 and lysosomal proteins (e.g., LAMP1) complements and facilitates the distinction of vesicles with different degrees of maturation [100–102]. Therefore, the identification of double-membrane containing autophagosomes can reliably be performed by TEM in microglia [31] (Fig. 4). However, the combination of TEM with immunoEM staining may facilitate the distinction between autophagosomes, autolysosomes, and other vesicles coming from nonautophagic sources. Autophagy flux can be analyzed by TEM by adding lysosomal inhibitors to the tested experimental conditions. Thus, TEM is a unique experimental method that allows direct observation and analysis of autophagic structures and their turnover in cells.
64
Ainhoa Plaza-Zabala and Amanda Sierra
Fig. 4 Autophagic vesicles can directly be visualized by TEM. (a) Images show the vesicle types that can be found in primary microglia after analysis of their cytoplasm by TEM. Autophagic-like vesicles are defined as having at least a portion of double membrane and different types of cargo, including granular, membranous, and heterogeneous structures. Lysosomal-like vesicles are characterized by their electron-density and by being enclosed by a single or double membrane. (b) Image panels show a single ultrafine section of primary microglia at low (left) and high (right) magnification. Microglia A (MiA) mainly contains electron-dense lysosomal-like vesicles of variable size. The high magnification image shows that some electron-dense vesicles are enclosed by a single membrane (presumable primary lysosomes) while others are delimited by a double membrane (highlighted in green). Microglia B (MiB) contains autophagic (highlighted in orange) and lysosomal-like vesicles. The high-magnification image shows a portion of the double-membrane containing autophagic-like vesicle that contains heterogeneous cargo inside (highlighted in orange). Scale bars = 2 μm (MiA); 1 μm (MiB) (low magnification) (b); 500 nm (high magnification) (a, b)
Autophagy Analysis in Microglia
65
Advantages
TEM-based methods allow direct visualization of autophagic and other endolysosomal pathway vesicles as well as their cargo and their relative localization in the cells. Additionally, visual ultrastructural analysis can be combined with biochemical marker staining by immunogold, which increases the accuracy of vesicle identification and categorization. Thus, TEM is a unique experimental method wherein autophagic compartments can be directly visualized and analyzed in microglia (Fig. 4).
Limitations
One of the main limitations of TEM is that it is a tedious experimental technique wherein ultrathin sections of fixed cells are imaged and analyzed. Notably, a single image of a cell section showing autophagic and/or other types of endolysosomal vesicles does not provide experimental proof of anything other than the presence of the vesicle in that specific experimental context. Indeed, TEM analysis of autophagic structures is quantifiable and experimentally valuable when rules of correct sampling are followed at the imaging step. Thus, the number of cells imaged and included in the analysis should be enough to ensure adequate statistical power for analysis. In addition, each cell section should be homogeneously scanned by taking representative random images of the whole cytoplasm. For accurate quantification and comparison of data coming from different experimental conditions, it is essential that the number and/or area of autophagic vesicles is normalized to the area of cell cytoplasm analyzed. Even when proceeding like this, only one z-plane of the cell is taken into account for the analysis, but the rest of the volume of the cell is discarded, which can lead to a biased interpretation of results. Thus, although data coming from TEM experiments are highly valuable in terms of direct assessment of autophagic structures, they are usually used as a method that complements the findings obtained by other experimental methods, including fluorescence confocal microscopy and WB methods. Alternatively, 3D reconstruction analysis by specialized electron microscopy techniques such as cryofixation and tomography can be used for direct assessment and quantification of autophagic compartments in a less biased manner [103].
4.3.2
Autophagic vesicles in microglia can also be visualized and analyzed by TEM imaging of brain-derived tissue. However, autophagy analysis in vivo in microglia has several limitations, including the identification of microglia, which is facilitated by immuno-EM analysis. Indeed, microglia contain a small diameter cell body surrounded by fine and relatively long processes. Thus, the percentage of cytoplasm analyzed per microglial cell in each cross-section is low, making it difficult to obtain quantitative data on the autophagic status of microglia. Alternatively, microglia can be FACS-sorted from brain tissue and processed for TEM [104], wherein autophagic vesicles can be identified and quantified similarly to microglial
In Vivo Analysis
66
Ainhoa Plaza-Zabala and Amanda Sierra
cultures (see above). Nevertheless, brain tissue processing for individual cell isolation alters the transcriptional and translational profile of microglia [105], which may also have an impact on their autophagic status, increasing the probability of yielding biased results. Additionally, the number of animals needed to obtain a sufficient quantity of microglia amenable to process for TEM may be high, especially when the focus of research is region-specific microglia instead of whole-brain microglia. In summary, in vivo TEM analysis of autophagy in microglia is possible. However, given its technical complexities, the data obtained may be used as a confirmatory complement for data obtained in vitro and/or in vivo by other experimental methods such as immunofluorescence and WB. 4.4 Concluding Remarks
The evaluation of autophagy flux in microglia is a complex task that requires the use of complementary experimental approaches. The dynamics of autophagosome formation and degradation in microglia are best captured by a combination of marker-based indirect biochemical methods such as LC3 tracking by fluorescence-based confocal microscopy or WB methods with direct assessment of autophagic vesicles by TEM. Combining these approaches with the analysis of complementary autophagy effectors and/or substrates is highly recommended to achieve a more in-depth understanding of mechanisms of initiation and progression of autophagy flux in microglia. It is also crucial to evaluate adaptations and functionality of the lysosomal compartment to ensure the efficient termination of the autophagy flux. Finally, experiments in microglial cultures allow understanding basic mechanisms of autophagy flux execution and regulation in microglia. Nevertheless, in vivo confirmation of these mechanisms is critical for a thorough understanding of the relevance of autophagic mechanisms in microglial physiology and brain function.
References 1. Prinz M, Masuda T, Wheeler MA et al (2021) Microglia and central nervous system–associated macrophages—from origin to disease modulation. Annu Rev Immunol 39:251– 277 2. Ginhoux F, Greter M, Leboeuf M et al (2010) Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science 330:841–845 3. Stremmel C, Schuchert R, Wagner F et al (2018) Yolk sac macrophage progenitors traffic to the embryo during defined stages of development. Nat Commun 9:75–72
4. Stogsdill JA, Kim K, Binan L et al (2022) Pyramidal neuron subtype diversity governs microglia states in the neocortex. Nature 608:750–756 5. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting microglial cells are highly dynamic surveillants of brain parenchyma in vivo. Science 308:1314–1318 6. Davalos D, Grutzendler J, Yang G et al (2005) ATP mediates rapid microglial response to local brain injury in vivo. Nat Neurosci 8:752–758 7. Hickman SE, Kingery ND, Ohsumi TK et al (2013) The microglial sensome revealed by
Autophagy Analysis in Microglia direct RNA sequencing. Nat Neurosci 16: 1896–1905 8. Sierra A, Paolicelli RC, Kettenmann H (2019) ˜ os de Microglı´a: Milestones in a cenCien An tury of microglial research. Trends Neurosci 42:778–792 9. Ma´rquez-Ropero M, Benito E, Plaza-Zabala A et al (2020) Microglial corpse clearance: lessons From macrophages. Front Immunol 11:506 10. Subhramanyam CS, Wang C, Hu Q et al (2019) Microglia-mediated neuroinflammation in neurodegenerative diseases. Semin Cell Dev Biol 94:112–120 11. Bernier LP, York EM, Kamyabi A et al (2020) Microglial metabolic flexibility supports immune surveillance of the brain parenchyma. Nat Commun 11:1559 12. Monsorno K, Buckinx A, Paolicelli RC (2022) Microglial metabolic flexibility: emerging roles for lactate. Trends Endocrinol Metab 33:186–195 13. Pohl C, Dikic I (2019) Cellular quality control by the ubiquitin-proteasome system and autophagy. Science 366:818–822 14. Lahiri V, Hawkins WD, Klionsky DJ (2019) Watch what you (self-) eat: autophagic mechanisms that modulate metabolism. Cell Metab 29:803–826 15. Deretic V, Kroemer G (2022) Autophagy in metabolism and quality control: opposing, complementary or interlinked functions? Autophagy 18:283–292 16. Melia TJ, Lystad AH, Simonsen A (2020) Autophagosome biogenesis: From membrane growth to closure. J Cell Biol 219: e202002085 17. Nakatogawa H (2020) Mechanisms governing autophagosome biogenesis. Nat Rev Mol Cell Biol 21:439–458 18. Yim WWY, Mizushima N (2020) Lysosome biology in autophagy. Cell Discov 6:6 19. Dossou AS, Basu A (2019) The emerging roles of mTORC1 in macromanaging autophagy. Cancers (Basel) 11(10):1422 20. Deleyto-Seldas N, Efeyan A (2021) The mTOR–autophagy axis and the control of metabolism. Front Cell Dev Biol 9:655731 21. Palamiuc L, Ravi A, Emerling BM (2020) Phosphoinositides in autophagy: current roles and future insights. FEBS J 287:222– 238 22. Lee YK, Lee JA (2016) Role of the mammalian ATG8/LC3 family in autophagy: differential and compensatory roles in the
67
spatiotemporal regulation of autophagy. BMB Rep 49:424–430 23. Kirkin V, Rogov VV (2019) A diversity of selective autophagy receptors determines the specificity of the autophagy pathway. Mol Cell 76:268–285 24. Klionsky DJ, Abdel-Aziz AK, Abdelfatah S et al (2021) Guidelines for the use and interpretation of assays for monitoring autophagy (4th edition). Autophagy 17:1–382 25. Mizushima N, Murphy LO (2020) Autophagy assays for biological discovery and therapeutic development. Trends Biochem Sci 45: 1080–1093 26. Griffey CJ, Yamamoto A (2022) Macroautophagy in CNS health and disease. Nat Rev Neurosci 23:411–427 27. Hara T, Nakamura K, Matsui M et al (2006) Suppression of basal autophagy in neural cells causes neurodegenerative disease in mice. Nature 441:885–889 28. Komatsu M, Waguri S, Chiba T et al (2006) Loss of autophagy in the central nervous system causes neurodegeneration in mice. Nature 441:880–884 29. Ju¨lg J, Laura S, Christian B (2021) Canonical and noncanonical autophagy pathways in microglia. Mol Cell Biol 41:389 30. Plaza-Zabala A, Sierra-Torre V, Sierra A (2017) Autophagy and microglia: novel partners in neurodegeneration and aging. Int J Mol Sci 18:598 31. Beccari S, Sierra-Torre V, Valero J et al (2023) Microglial phagocytosis dysfunction in stroke is driven by energy depletion and induction of autophagy. Autophagy 19:1952–1981 32. Alam MM, Zhao XF, Liao Y et al (2021) Deficiency of microglial autophagy increases the density of oligodendrocytes and susceptibility to severe forms of seizures. eNeuro 8: ENEURO.0183-0120.2021 33. Choi I, Zhang Y, Seegobin SP et al (2020) Microglia clear neuron-released α-synuclein via selective autophagy and prevent neurodegeneration. Nat Commun 11:1386 34. Srimat Kandadai K, Kotur MB, Dokalis N et al (2021) ATG5 in microglia does not contribute vitally to autoimmune neuroinflammation in mice. Autophagy 17:3566–3576 35. Yin X, Propson NE, Shuqi D et al (2021) Autophagy deficiency modulates microglial lipid homeostasis and aggravates tau pathology and spreading. Proc Natl Acad Sci U S A 118:e2023418118 36. Cheng J, Liao Y, Dong Y et al (2020) Microglial autophagy defect causes parkinson disease-like symptoms by accelerating
68
Ainhoa Plaza-Zabala and Amanda Sierra
inflammasome activation in mice. Autophagy 16:2193–2205 37. Cho MH, Cho K, Kang HJ et al (2014) Autophagy in microglia degrades extracellular β-amyloid fibrils and regulates the NLRP3 inflammasome. Autophagy 10:1761–1775 38. Houtman J, Freitag K, Gimber N et al (2019) Beclin1-driven autophagy modulates the inflammatory response of microglia via NLRP3. EMBO J 38:e99430 39. Weber A, Wasiliew P, Kracht M (2010) Interleukin-1 (IL-1) pathway. Sci Signal 3: cm2 40. Lu J, Zhang C, Lv J et al (2021) Antiallergic drug desloratadine as a selective antagonist of 5HT2A receptor ameliorates pathology of Alzheimer’s disease model mice by improving microglial dysfunction. Aging Cell 20:e13286 41. Tu HY, Yuan BS, Hou XO et al (2021) α-synuclein suppresses microglial autophagy and promotes neurodegeneration in a mouse model of Parkinson’s disease. Aging Cell 20: e13522 42. Qin Y, Qiu J, Wang P et al (2021) Impaired autophagy in microglia aggravates dopaminergic neurodegeneration by regulating NLRP3 inflammasome activation in experimental models of Parkinson’s disease. Brain Behav Immun 91:324–338 43. Han X, Xu T, Fang Q et al (2021) Quercetin hinders microglial activation to alleviate neurotoxicity via the interplay between NLRP3 inflammasome and mitophagy. Redox Biol 44:102010 44. Ji J, Xue TF, Guo XD et al (2018) Antagonizing peroxisome proliferator-activated receptor γ facilitates M1-to-M2 shift of microglia by enhancing autophagy via the LKB1–AMPK signaling pathway. Aging Cell 17:e12774 45. Jin MM, Wang F, Qi D et al (2018) A critical role of autophagy in regulating microglia polarization in neurodegeneration. Front Aging Neurosci 10:378 46. Heckmann BL, Teubner BJW, BoadaRomero E et al (2020) Noncanonical function of an autophagy protein prevents spontaneous Alzheimer’s disease. Sci Adv 6: eabb9036 47. He Y, She H, Zhang T et al (2017) p38 MAPK inhibits autophagy and promotes microglial inflammatory responses by phosphorylating ULK1. J Cell Biol 217:315–328 48. Friess L, Cheray M, Keane L et al (2021) Atg7 deficiency in microglia drives an altered transcriptomic profile associated with an impaired neuroinflammatory response. Mol Brain 14:87
49. Wang H, Liu Q, Zhang X (2020) C1q/tumor necrosis factor-related protein-1 attenuates microglia autophagy and inflammatory response by regulating the Akt/mTOR pathway. Life Sci 256:117992 50. Yang Z, Zhong L, Zhong S et al (2015) Hypoxia induces microglia autophagy and neural inflammation injury in focal cerebral ischemia model. Exp Mol Pathol 98:219–224 51. Shi H, Wang J, Wang J et al (2018) IL-17A induces autophagy and promotes microglial neuroinflammation through ATG5 and ATG7 in intracerebral hemorrhage. J Neuroimmunol 323:143–151 52. Heckmann BL, Teubner BJW, Tummers B et al (2019) LC3-associated endocytosis facilitates β-Amyloid clearance and mitigates neurodegeneration in murine Alzheimer’s disease. Cell 178:536–551 53. Berglund R, Guerreiro-Cacais AO, Adzemovic MZ et al (2020) Microglial autophagy– associated phagocytosis is essential for recovery from neuroinflammation. Sci Immunol 5: eabb5077 ˜ a-Ramos O, Chiao L, Liu X et al (2022) 54. Pen Autophagosomes fuse to phagosomes and facilitate the degradation of apoptotic cells in Caenorhabditis elegans. elife 11:e72466 55. Li G, Sherchan P, Tang Z et al (2021) Autophagy & phagocytosis in neurological disorders and their possible cross-talk. Curr Neuropharmacol 19:1912–1924 56. Sierra A, Encinas JM, Deudero JJP et al (2010) Microglia shape adult hippocampal neurogenesis through apoptosis-coupled phagocytosis. Cell Stem Cell 7:483–495 57. Kim HJ, Cho MH, Shim WH et al (2017) Deficient autophagy in microglia impairs synaptic pruning and causes social behavioral defects. Mol Psychiatry 22:1576–1584 58. Meijer AJ, Lorin S, Blommaart EF et al (2015) Regulation of autophagy by amino acids and MTOR-dependent signal transduction. Amino Acids 47:2037–2063 59. Martinez J, Malireddi RK, Lu Q et al (2015) Molecular characterization of LC3-associated phagocytosis reveals distinct roles for Rubicon, NOX2 and autophagy proteins. Nat Cell Biol 17:893–906 60. Plaza-Zabala A, Sierra-Torre V, Sierra A (2021) Assessing autophagy in microglia: a two-step model to determine autophagosome formation, degradation, and net turnover. Front Immunol 11:620602 61. Mauvezin C, Nagy P, Juha´sz G et al (2015) Autophagosome-lysosome fusion is
Autophagy Analysis in Microglia independent of V-ATPase-mediated acidification. Nat Commun 6:7007 62. Lazarenko RM, DelBove CE, Strothman CE et al (2017) Ammonium chloride alters neuronal excitability and synaptic vesicle release. Sci Rep 7:5061 63. Halcrow PW, Geiger JD, Chen X (2021) Overcoming chemoresistance: altering pH of cellular compartments by chloroquine and hydroxychloroquine. Front Cell Dev Biol 9: 627639 64. Kim J, Kundu M, Viollet B et al (2011) AMPK and mTOR regulate autophagy through direct phosphorylation of Ulk1. Nat Cell Biol 13:132–141 65. Lamark T, Svenning S, Johansen T (2017) Regulation of selective autophagy: the p62/SQSTM1 paradigm. Essays Biochem 61:609–624 66. Kumar AV, Mills J, Lapierre LR (2022) Selective autophagy receptor p62/SQSTM1, a pivotal player in stress and aging. Front Cell Dev Biol 10:793328 67. Lee JW, Nam H, Kim LE et al (2019) TLR4 (toll-like receptor 4) activation suppresses autophagy through inhibition of FOXO3 and impairs phagocytic capacity of microglia. Autophagy 15:753–770 68. Gao Z, Gammoh N, Wong PM et al (2010) Processing of autophagic protein LC3 by the 20S proteasome. Autophagy 6:126–137 69. Pillai-Kastoori L, Schutz-Geschwender A, Harford JA (2020) A systematic approach to quantitative Western blot analysis. Anal Biochem 593:113608 70. Han B, Jiang W, Cui P et al (2021) Microglial PGC-1α protects against ischemic brain injury by suppressing neuroinflammation. Genome Med 13:47 71. Thangaraj A, Periyasamy P, Guo ML et al (2020) Mitigation of cocaine-mediated mitochondrial damage, defective mitophagy and microglial activation by superoxide dismutase mimetics. Autophagy 16:289–312 72. Yuan B, Shen H, Lin L et al (2017) Autophagy promotes microglia activation through Beclin-1-Atg5 pathway in intracerebral hemorrhage. Mol Neurobiol 54:115–124 73. Rubinsztein DC, Cuervo AM, Ravikumar B et al (2009) In search of an “autophagomometer”. Autophagy 5:585–589 74. Levine B, Kroemer G (2019) Biological functions of autophagy genes: a disease perspective. Cell 176:11–42 75. He Y, Yao X, Taylor N et al (2018) RNA analysis reveals quiescent sequencing
69
microglia isolation methods from postnatal mouse brains and limitations of BV2 cells. J Neuroinflammation 15:153 76. Llufrio EM, Wang L, Naser FJ et al (2018) Sorting cells alters their redox state and cellular metabolome. Redox Biol 16:381–387 77. Guo S, Liang Y, Murphy SF et al (2015) A rapid and high content assay that measures cyto-ID-stained autophagic compartments and estimates autophagy flux with potential clinical applications. Autophagy 11:560–572 78. Kimura S, Noda T, Yoshimori T (2007) Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy 3:452– 460 79. Kaizuka T, Morishita H, Hama Y et al (2016) An autophagic flux probe that releases an internal control. Mol Cell 64:835–849 80. Pankiv S, Clausen TH, Lamark T et al (2007) p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of ubiquitinated protein aggregates by autophagy *. J Biol Chem 282:24131–24145 81. Florey O, Gammoh N, Kim SE et al (2015) V-ATPase and osmotic imbalances activate endolysosomal LC3 lipidation. Autophagy 11:88–99 82. Romao S, Gasser N, Becker AC et al (2013) Autophagy proteins stabilize pathogencontaining phagosomes for prolonged MHC II antigen processing. J Cell Biol 203:757– 766 83. Katayama H, Kogure T, Mizushima N et al (2011) A sensitive and quantitative technique for detecting autophagic events based on lysosomal delivery. Chem Biol 18:1042–1052 84. Sun N, Yun J, Liu J et al (2015) Measuring in vivo mitophagy. Mol Cell 60:685–696 85. Allen GFG, Toth R, James J et al (2013) Loss of iron triggers PINK1/Parkin-independent mitophagy. EMBO Rep 14:1127–1135 86. McWilliams TG, Prescott AR, Allen GFG et al (2016) mito-QC illuminates mitophagy and mitochondrial architecture in vivo. J Cell Biol 214:333–345 87. Tian W, Alsaadi R, Guo Z et al (2020) An antibody for analysis of autophagy induction. Nat Methods 17:232–239 88. Polson HEJ, de Lartigue J, Rigden DJ et al (2010) Mammalian Atg18 (WIPI2) localizes to omegasome-anchored phagophores and positively regulates LC3 lipidation. Autophagy 6:506–522 89. Bjørkøy G, Lamark T, Brech A et al (2005) p62/SQSTM1 forms protein aggregates
70
Ainhoa Plaza-Zabala and Amanda Sierra
degraded by autophagy and has a protective effect on huntingtin-induced cell death. J Cell Biol 171:603–614 90. Zhao YG, Zhang H (2019) Autophagosome maturation: an epic journey from the ER to lysosomes. J Cell Biol 218:757–770 91. Cheng XT, Xie YX, Zhou B et al (2018) Revisiting LAMP1 as a marker for degradative autophagy-lysosomal organelles in the nervous system. Autophagy 14:1472–1474 92. Choubey V, Zeb A, Kaasik A (2022) Molecular mechanisms and regulation of mammalian mitophagy. Cell 11:38 93. du Toit A, Hofmeyr JHS, Gniadek TJ et al (2018) Measuring autophagosome flux. Autophagy 14:1060–1071 94. Arotcarena ML, Soria FN, Cunha A et al (2022) Acidic nanoparticles protect against α-synuclein-induced neurodegeneration through the restoration of lysosomal function. Aging Cell 21:e13584 95. Lu J, Zhou W, Dou F et al (2021) TRPV1 sustains microglial metabolic reprogramming in Alzheimer’s disease. EMBO Rep 22: e52013 96. Mizushima N, Yamamoto A, Matsui M et al (2004) In vivo analysis of autophagy in response to nutrient starvation using transgenic mice expressing a fluorescent autophagosome marker. Mol Biol Cell 15:1101–1111 97. Li L, Wang ZV, Hill JA et al (2014) New autophagy reporter mice reveal dynamics of proximal tubular autophagy. J Am Soc Nephrol 25:305–315
98. Ramachandra Rao S, Fliesler SJ (2022) Monitoring basal autophagy in the retina utilizing CAG-mRFP-EGFP-MAP1LC3B reporter mouse: technical and biological considerations. Autophagy 18:1187–1201 99. Eskelinen EL (2008) To be or not to be? Examples of incorrect identification of autophagic compartments in conventional transmission electron microscopy of mammalian cells. Autophagy 4:257–260 100. Eskelinen EL (2005) Maturation of autophagic vacuoles in mammalian cells. Autophagy 1:1–10 101. Eskelinen EL, Reggiori F, Baba M et al (2011) Seeing is believing: the impact of electron microscopy on autophagy research. Autophagy 7:935–956 102. Nixon RA, Wegiel J, Kumar A et al (2005) Extensive involvement of autophagy in Alzheimer disease: an immuno-electron microscopy study. J Neuropathol Exp Neurol 64: 113–122 103. Jung M, Choi H, Mun JY (2019) The autophagy research in electron microscopy. Appl Microsc 49:11–16 104. Ulland TK, Song WM, Huang SCC et al (2017) TREM2 maintains microglial metabolic fitness in Alzheimer’s disease. Cell 170: 649–663 105. Marsh SE, Walker AJ, Kamath T et al (2022) Dissection of artifactual and confounding glial signatures by single-cell sequencing of mouse and human brain. Nat Neurosci 25: 306–316
Chapter 4 Hemocyte Nuclei Isolation from Adult Drosophila melanogaster for snRNA-seq Fabian Hersperger, Melanie Kastl, Katrin Paeschke, and Katrin Kierdorf Abstract In adult Drosophila, most of the hemocytes are macrophage-like cells (so called plasmatocytes), which serve various functions in organ homeostasis and immune defense. Ontogeny and functions are largely conserved between vertebrate and invertebrate macrophages. Hence, Drosophila offers a powerful genetic toolbox to study macrophage function and genetically modulate these cells. Technological advances in highthroughput sequencing approaches allowed to give an in-depth characterization of vertebrate macrophage populations and their heterogenous composition within different organs as well as changes in disease. Embryonic and larval hemocytes in Drosophila have been recently analyzed in single-cell RNA-sequencing (scRNA-seq) approaches during infection and steady state. These analyses revealed anatomical and functional Drosophila hemocyte subtypes dedicated to specific tasks. Only recently, the Fly Cell Atlas provided a whole transcriptomic single-cell atlas via single-nuclei RNA-sequencing (snRNA-seq) of adult Drosophila including many different tissues and cell types where hemocytes were also included. Yet, a specific protocol to isolate nuclei from adult hemocytes for snRNA-seq and study these cells in different experimental conditions was not available. In this chapter, we give a detailed protocol to purify hemocyte nuclei from adult Drosophila, which can be used in subsequent analyses such as snRNA-seq. Key words Drosophila macrophages, Hemocytes, Plasmatocytes, Nuclei isolation, snRNA-seq
1
Introduction Macrophages are innate immune cells found in nearly all metazoan organisms, including invertebrates such as Drosophila melanogaster. Hemocytes, the Drosophila blood-like immune cells, consist of three different cell types with different innate immune tasks: crystal cells (melanization), lamellocytes (encapsulation), and plasmatocytes (phagocytosis). Interestingly, 95% of all adult hemocytes are plasmatocytes, which are widely considered as the macrophage equivalent in Drosophila. Hence, adult Drosophila plasmatocytes (from here on summarized under the term hemocytes for simplification throughout this chapter) are widely used as a model to study tissue-resident macrophage function in health and disease
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_4, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
71
72
Fabian Hersperger et al.
[1–5]. Adult hemocytes have a stratified developmental origin similar to vertebrate macrophages [6, 7] and share many conserved functions to vertebrate macrophages as recently demonstrated for development of fat tissue [8] or during high fat diet [4]. In contrast to vertebrate macrophages, adult Drosophila hemocytes do not proliferate, and so far no verified hematopoietic hub was defined to reconstitute hemocytes in the adult fly [1, 4]. In the last years, high-throughput sequencing techniques of individual cells or nuclei became more and more popular providing a powerful tool to investigate complex tissues and decipher scientific questions with a much deeper output than old-fashioned genetic approaches such as real-time qPCR. Furthermore, the big advantage of single-cell RNA sequencing (scRNA-seq), as well as single-nucleus RNA sequencing (snRNA-seq), is that whole tissues with a multitude of cell types can be analyzed at once on an individual cell level unravelling heterogeneity within cell types, developmental trajectories, cell-cell interactions, and gene regulatory networks [9]. This powerful toolbox has been extensively used in the field of vertebrate macrophage research and gave new insights into the heterogeneity and diversity of these cells across organs, but also within defined macrophage populations during development, health, and disease [10, 11]. First scRNA-seq experiments were performed on embryonic and larval hemocytes in Drosophila [12–14]. In contrast, adult hemocytes have been more or less unexplored with this technique and, only recently, were included by snRNA-seq in the Fly Cell Atlas [15]. Here, hemocytes were described alongside many other cell types from adult flies. However, a protocol to offer an adult hemocyte-specific isolation protocol for high-throughput sequencing approaches has not been established, yet. Here, we describe a step-by-step protocol based on the “Frankenstein” protocol [16] to isolate and sort nuclei from adult hemocytes for downstream analysis such as snRNA-seq. The protocol provides a valuable tool for other labs to isolate adult hemocytes from different genotypes or disease conditions.
2
Materials Before you begin, expand your fly lines of interest for the experiment. Take into account that you want to analyze same-aged flies and try to avoid using experimental flies with balancers. The protocol is optimized for adult hemocytes isolated from Drosophila imago which are 7 days after eclosion from pupae or older. Book the cell sorter on the day of the experiment and make sure the time slot is sufficient for all nuclei samples you want to analyze. Clean and autoclave enough homogenization pestles for your experiment. Prepare Nuclei Wash and Resuspension Buffer (see also 2.3) freshly on the day of the experiment.
Hemocyte Nuclei Isolation
73
2.1 Transgenic Drosophila melanogaster Lines
1. w1118 line (isogenic control without reporter) (BDSC line #5905).
2.2
1. Homogenization pestles for 1.5 mL microcentrifuge tubes (see also Note 2).
Plastic Ware
2. w1118;HmlΔ-DsRed.nuc (transgenic line with nuclei-bound DsRed reporter in hemocytes) (gift from M. S. Dionne) (see also Note 1).
2. 1.5 mL safe-lock microcentrifuge tubes. 3. 70 μm pre-separation filters. 4. 40 μm cell strainer for pipette tips. 5. 100–1000 μL autoclaved disposable pipette tips. 6. Polystyrene round-bottom FACS tube with lid. 2.3 Buffers and Solutions
1. Nuclei EZ lysis buffer (Sigma, #NUC101). 2. Nuclei Wash and Resuspension Buffer: 1% bovine serum albumin, fraction V, 0.2 U/μL murine RNase Inhibitor (see also Note 3) in 1x phosphate buffered saline (PBS) without Ca2+ and Mg2+. 3. 10 mg/mL 4′,6-diamidino-2-phenylindole (DAPI) in ultrapure water (see also Note 3). 4. DRAQ7 dye. 5. Staining solution: Nuclei Wash and Resuspension Buffer containing DAPI (1:1000) and DRAQ7 (1:100) (see also Note 3).
2.4
Equipment
1. 100–1000 μL single-channel pipette. 2. Microtube centrifuge. 3. Cell sorter.
3
Methods Of note, carry out all steps on ice and avoid warming to ensure high quality of the nuclei samples for further downstream analysis.
3.1 Fly Dissection and Nuclei Isolation
1. Prepare 1.5 mL microcentrifuge tubes for each sample and label them. 2. Fill 500 μL Nuclei EZ lysis buffer into the tubes and place them on ice. 3. Anesthetize the flies with CO2 in their housing tube and transfer them into the Nuclei EZ lysis buffer (see also Note 4). 4. Carefully homogenize the flies using a clean homogenization pestle (see also Note 5).
74
Fabian Hersperger et al.
5. Add another 500 μL of Nuclei EZ lysis buffer with a pipette and gently resuspend the homogenate (see also Note 6). 6. Incubate the sample for 5 min on ice. Gently resuspend the sample with a pipette 1–2 times during the incubation. 7. Filter the homogenate through a 70 μm pre-separation filter into a polystyrene round-bottom FACS tube or directly into a new 1.5 mL microcentrifuge tube. 8. Centrifuge the nuclei in a 1.5 mL microcentrifuge tube for 6 min at 500 g and 4 °C. Remove the supernatant by pouring it out. Usually 50 μL of liquid remain on top of the nuclei pellet. 9. Gently add another 1 mL of Nuclei EZ lysis buffer and resuspend the nuclei. Incubate the sample for 5 min on ice without resuspending during the incubation. 10. Centrifuge the nuclei for 6 min at 500 g and 4 °C. Remove the supernatant as described above. 11. Carefully add 500 μL of Nuclei Wash and Resuspension Buffer. Add the buffer carefully along the wall of the tube without disturbing the pellet to allow a gentle buffer interchange. Incubate the nuclei for 5 min on ice without mixing. 12. After the incubation, add another 500 μL of Nuclei Wash and Resuspension Buffer and gently resuspend the pellet with a pipette. 13. Centrifuge the nuclei for 6 min at 500 g and 4 °C. Remove the supernatant by pouring it out. 14. Gently suspend the nuclei in 1 mL Wash and Resuspension Buffer (see Note 7). 15. Centrifuge the nuclei for 6 min at 500 g and 4 °C. Remove the supernatant as much as possible without disturbing the pellet. Use a pipette for this step. 16. Suspend the nuclei in 200 μL staining solution (Nuclei Wash and Resuspension Buffer supplemented with DAPI and DRAQ7). Incubate the nuclei for 15 min on ice in the dark. 17. Add 1 mL Nuclei Wash and Resuspension Buffer to the tube and mix gently with a pipette. 18. Centrifuge the nuclei for 6 min at 500 g and 4 °C. Remove the supernatant by pouring it out. 19. Suspend the nuclei in 200 μL Nuclei Wash and Resuspension Buffer. 20. Use a 40 μm cell strainer placed on the tip of a pipette to filter the sample. Carefully press the sample through the filter with a 1000 μL pipette. Use this step to transfer the nuclei into a sample tube, appropriate for the used cell sorter. 21. Proceed with nuclei sorting.
Hemocyte Nuclei Isolation
3.2 Hemocyte Nuclei Sorting
75
The nuclei sorting is technically possible on any cell sorter that allows the detection of DAPI, DRAQ7, and a potential cell-type– specific fluorescent nuclear reporter (in our case the hemocyte nuclei-specific DsRed). We sort the nuclei on a BD Aria III using a 70 μm nozzle. To sort the correct events, the following gating strategy is employed (Fig. 1): 1. The nuclei are relatively small. Therefore, the gate in FSC-A and SSC-A is set on the small events with relative low side scatter, as shown in Fig. 1. 2. In the next gate, the DRAQ7 single-positive events contain the nuclei singlets. DRAQ7 is a DNA dye and therefore allows staining of nuclei. Usually, the doublets are visible as well as they show a higher DRAQ7 signal and form a second population. For transcriptomic analysis of single nuclei, such as the 10x genomics approach, it is recommended to exclude the doublets. 3. DRAQ7 vs. DAPI is used to re-gate on nuclei and guarantee that the DRAQ7 signal is really nuclei-specific and not caused by autofluorescent material, which is often a problem in adult Drosophila cell- or nuclei-homogenates. The DRAQ7+DAPI+ events are assumed to be purified nuclei (see Note 8). 4. Next, hemocyte nuclei are defined by gating on DRAQ7+DAPI+DsRed+ nuclei (Fig. 1). This nuclei population of interest is then sorted for further analyses (see Notes 9–12). 5. Sort the hemocyte nuclei into a 1.5 mL microcentrifuge tube containing a few microliters of Nuclei Wash and Resuspension Buffer (containing the RNase inhibitor). Now, the purified hemocyte nuclei can be used for further analyses, such as snRNA-seq (see Note 13).
Fig. 1 Gating strategy for hemocyte nuclei sort. Based on event size and granularity, the first gate is set to a position catching relatively small and non-granular events. Next, the DRAQ7 single-positive events are selected to gate on nuclei singlets. The doublets are nicely visible in the plot with a slightly increased DRAQ7 signal. In the next gate, the nuclei are re-gated to confirm nuclear identity by DAPI and DRAQ7. In the final gate, hemocyte nuclei are now identified by DsRed fluorescence
76
4
Fabian Hersperger et al.
Notes 1. You can also use hemocyte-specific Gal4 driver lines (e.g., srpGal4 [17]) and combine them with an acceptor line using an upstream activator sequence (UAS) to express a nuclear fluorescent marker (Stinger or redStinger) under control of the desired hemocyte-specific Gal4 line. 2. Pestles can be reused between experiments. Used pestles are first rinsed in water, subsequently cleaned with isopropanol, and can be dried and reused (if wanted the pestles can also be autoclaved). 3. All buffers and solutions should be prepared and stored at 4 °C. Prepare all solutions using ultrapure water. Do not store longer than 6 months. 4. The best yield of isolated hemocyte nuclei was achieved by using 100 flies for one sample. Speaking of nuclei numbers, it is possible to isolate approximately 2500–4000 hemocyte nuclei from 100 flies, depending on the cell sorter that is used. We do not recommend to use more than 100 flies as the nuclei isolation can be incomplete. If a higher yield is needed, we recommend preparing several samples and pool them after FACS sorting the nuclei (step 3.2) in one collection tube. Furthermore, single flies can be used with this protocol to analyze the individual hemocyte population of one fly. Please keep in mind that the amount of flies needed largely depends on the subsequent analysis and the scientific research question. 5. The pieces should be equally small and homogenized. The homogenization is incomplete if pieces of the fly such as legs or body walls are still visible. During this process, it might be necessary to take the sample out of the ice. However, we recommend being quick to place the sample back on ice as fast as possible. 6. To avoid that large debris fragments block the pipette tip, we recommend cutting the first few millimeters of the pipette tip. 7. Carefully wash the walls of the tube a few times to take the nuclei into solution that may stick on the wall of the tube. 8. The DRAQ7 gate is the main gate to identify nuclei as the DRAQ7-emission signal is very specific with a distinct peak. DAPI is only used as a confirmation. If only DAPI is used, many more non-nuclear fragments would be sorted due to the larger emission spectrum of DAPI. In Drosophila homogenates, a high quantity of fragments is autofluorescence. DRAQ7 is more specific in this case.
Hemocyte Nuclei Isolation
77
9. If doing an enrichment approach with a cell-type–specific fluorescent reporter, as demonstrated here for hemocytes, make sure to use a non-reporter control (e.g., w1118 or OregonS) to set the gating at the correct position for your reporter-positive nuclei. If you are not using a cell-type–specific reporter, this protocol can be used to sort all DAPI+DRAQ7+ nuclei and perform snRNA-seq from many different cell types without enrichment for one specific cell type. 10. As mentioned above, the here described protocol also allows to isolate all nuclei of the fly. In the subsequent analysis using state-of-the-art methods, such as 10x Genomics, we are able to annotate the sorted cell types according to their gene expression profile with help of the Fly Cell Atlas [15]. However, it is also possible to enrich for another cell type of interest. Therefore, it is necessary to express a nuclear reporter protein in the cells of interest. When choosing your reporter fly lines, take into account that the reporter has to label the nuclei of the cell of interest and should not interfere with the emission of DAPI and DRAQ7, otherwise it is not possible to enrich for the desired cells. 11. For the sorting step, it is essential to have a negative control for the DsRed signal in hemocyte nuclei. The intensity of DsRed can differ between nuclei of the hemocyte population, and therefore the cut-off to the non-labeled nuclei can be challenging. 12. The isolated and sorted nuclei can be further processed and analyzed via different methods, such as a specific staining, to evaluate DNA damage in hemocyte nuclei. 13. The appropriate volume/number of nuclei, which can be sorted in one tube, largely depends on the next analysis steps. If the tube gets too full during the sort, stop the sort in between and divide the sample on several microcentrifuge tubes.
Acknowledgments We would like to acknowledge the Lighthouse Core Facility, Medical faculty, University of Freiburg and its staff, especially J. BodinekWersing and U. Jagadeshwaran, for their assistance with the sorting of nuclei. The Kierdorf lab is supported by project grants of the Fritz Thyssen Foundation and of the DFG. The Kierdorf lab is further supported by the DFG through project grants within SFB/TRR167 (Project ID 259373024), CRC1479 (Project ID 441891347) and within Germany’s Excellence Strategy (grant no. CIBSS—EXC-2189, Project ID 390939984). Research in the Paeschke laboratory is fund by an ERC Stg Grant (638988-
78
Fabian Hersperger et al.
G4DSB) and by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Project-ID 369799452 – TRR237” as well as by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC2151 – 390873048. References 1. Sanchez Bosch P, Makhijani K, Herboso L, Gold KS, Baginsky R, Woodcock KJ, Alexander B, Kukar K, Corcoran S, Jacobs T, Ouyang D, Wong C, Ramond EJV, Rhiner C, Moreno E, Lemaitre B, Geissmann F, Bru¨ckner K (2019) Adult drosophila lack hematopoiesis but rely on a blood cell reservoir at the respiratory epithelia to relay infection signals to surrounding tissues. Dev Cell 51:787–803.e5. https://doi.org/10.1016/j.devcel.2019. 10.017 2. Kierdorf K, Hersperger F, Sharrock J, Vincent CM, Ustaoglu P, Dou J, Gyoergy A, Groß O, Siekhaus DE, Dionne MS (2020) Muscle function and homeostasis require cytokine inhibition of AKT activity in Drosophila. elife 9. https://doi.org/10.7554/eLife.51595 3. Pe´an CB, Schiebler M, Tan SWS, Sharrock JA, Kierdorf K, Brown KP, Maserumule MC, Menezes S, Pila´tova´ M, Bronda K, Guermonprez P, Stramer BM, Andres Floto R, Dionne MS (2017) Regulation of phagocyte triglyceride by a STAT-ATG2 pathway controls mycobacterial infection. Nat Commun 8:14642. https://doi.org/10. 1038/ncomms14642 4. Woodcock KJ, Kierdorf K, Pouchelon CA, Vivancos V, Dionne MS, Geissmann F (2015) Macrophage-derived upd3 cytokine causes impaired glucose homeostasis and reduced lifespan in Drosophila fed a lipid-rich diet. Immunity 42:133–144. https://doi.org/10.1016/j. immuni.2014.12.023 5. Krejcˇova´ G, Danielova´ A, Nedbalova´ P, Kazek M, Strych L, Chawla G, Tennessen JM, Lieskovska´ J, Jindra M, Dolezˇal T, Bajgar A (2019) Drosophila macrophages switch to aerobic glycolysis to mount effective antibacterial defense. elife 8:e50414. https://doi.org/10. 7554/eLife.50414 6. Holz A, Bossinger B, Strasser T, Janning W, Klapper R (2003) The two origins of hemocytes in Drosophila. Dev Camb Engl 130: 4955–4962. https://doi.org/10.1242/dev. 00702 7. Gold KS, Bru¨ckner K (2014) Drosophila as a model for the two myeloid blood cell systems
in vertebrates. Exp Hematol 42:717–727. https://doi.org/10.1016/j.exphem.2014. 06.002 8. Cox N, Crozet L, Holtman IR, Loyher P-L, Lazarov T, White JB, Mass E, Stanley ER, Elemento O, Glass CK, Geissmann F (2021) Diet-regulated production of PDGFcc by macrophages controls energy storage. Science 373. https://doi.org/10.1126/science. abe9383 9. Hedlund E, Deng Q (2018) Single-cell RNA sequencing: technical advancements and biological applications. Mol Asp Med 59:36– 46. https://doi.org/10.1016/j.mam.2017. 07.003 10. Masuda T, Sankowski R, Staszewski O, Bo¨ttcher C, Amann L, Sagar n, Scheiwe C, Nessler S, Kunz P, van Loo G, Coenen VA, Reinacher PC, Michel A, Sure U, Gold R, Gru¨n D, Priller J, Stadelmann C, Prinz M (2019) Spatial and temporal heterogeneity of mouse and human microglia at single-cell resolution. Nature 566:388–392. https://doi.org/ 10.1038/s41586-019-0924-x 11. Guilliams M, Bonnardel J, Haest B, Vanderborght B, Wagner C, Remmerie A, Bujko A, Martens L, Thone´ T, Browaeys R, De Ponti FF, Vanneste B, Zwicker C, Svedberg FR, Vanhalewyn T, Gonc¸alves A, Lippens S, Devriendt B, Cox E, Ferrero G, Wittamer V, Willaert A, Kaptein SJF, Neyts J, Dallmeier K, Geldhof P, Casaert S, Deplancke B, ten Dijke P, Hoorens A, Vanlander A, Berrevoet F, Van Nieuwenhove Y, Saeys Y, Saelens W, Van Vlierberghe H, Devisscher L, Scott CL (2022) Spatial proteogenomics reveals distinct and evolutionarily conserved hepatic macrophage niches. Cell 185:379–396.e38. https:// doi.org/10.1016/j.cell.2021.12.018 12. Cho B, Yoon S-H, Lee D, Koranteng F, Tattikota SG, Cha N, Shin M, Do H, Hu Y, Oh SY, Lee D, Vipin Menon A, Moon SJ, Perrimon N, Nam J-W, Shim J (2020) Single-cell transcriptome maps of myeloid blood cell lineages in Drosophila. Nat Commun 11:4483. https:// doi.org/10.1038/s41467-020-18135-y
Hemocyte Nuclei Isolation 13. Cattenoz PB, Sakr R, Pavlidaki A, Delaporte C, Riba A, Molina N, Hariharan N, Mukherjee T, Giangrande A (2020, e104486) Temporal specificity and heterogeneity of Drosophila immune cells. EMBO J 39. https://doi.org/ 10.15252/embj.2020104486 14. Tattikota SG, Cho B, Liu Y, Hu Y, Barrera V, Steinbaugh MJ, Yoon S-H, Comjean A, Li F, Dervis F, Hung R-J, Nam J-W, Ho Sui S, Shim J, Perrimon N (2020) A single-cell survey of Drosophila blood. elife 9:e54818. https:// doi.org/10.7554/eLife.54818 15. Li H, Janssens J, De Waegeneer M, Kolluru SS, Davie K, Gardeux V, Saelens W, David FPA, Brbic´ M, Spanier K, Leskovec J, McLaughlin CN, Xie Q, Jones RC, Brueckner K, Shim J, Tattikota SG, Schnorrer F, Rust K, Nystul TG, Carvalho-Santos Z, Ribeiro C, Pal S, Mahadevaraju S, Przytycka TM, Allen AM, Goodwin SF, Berry CW, Fuller MT, WhiteCooper H, Matunis EL, DiNardo S, Galenza A, O’Brien LE, Dow JAT, FCA Consortium}, Jasper H, Oliver B, Perrimon N, Deplancke B, Quake SR, Luo L, Aerts S, Agarwal D, Ahmed-Braimah Y, Arbeitman M, Ariss MM, Augsburger J, Ayush K, Baker CC, Banisch T, Birker K, Bodmer R, Bolival B, Brantley SE, Brill JA, Brown NC, Buehner NA, Cai XT, Cardoso-Figueiredo R, Casares F, Chang A, Clandinin TR, Crasta S, Desplan C, Detweiler AM, Dhakan DB, Dona` E, Engert S, Floc’hlay S, George N, Gonza´lez-Segarra AJ, Groves AK, Gumbin S, Guo Y, Harris DE, Heifetz Y, Holtz SL,
79
Horns F, Hudry B, Hung R-J, Jan YN, Jaszczak JS, Jefferis GSXE, Karkanias J, Karr TL, Katheder NS, Kezos J, Kim AA, Kim SK, Kockel L, Konstantinides N, Kornberg TB, Krause HM, Labott AT, Laturney M, Lehmann R, Leinwand S, Li J, Li JSS, Li K, Li K, Li L, Li T, Litovchenko M, Liu H-H, Liu Y, Lu T-C, Manning J, Mase A, MateraVatnick M, Matias NR, McDonoughGoldstein CE, McGeever A, McLachlan AD, Moreno-Roman P, Neff N, Neville M, Ngo S, Nielsen T, O’Brien CE, Osumi-Sutherland D, ¨ zel MN, Papatheodorou I, Petkovic M, O Pilgrim C, Pisco AO, Reisenman C, Sanders EN, Santos GD, Scott K, Sherlekar A, Shiu P, Sims D, Sit RV, Slaidina M, Smith HE, Sterne G, Su Y-H, Sutton D, Tamayo M, Tan M, Tastekin I, Treiber C, Vacek D, Vogler G, Waddell S, Wang W, Wilson RI, Wolfner MF, Wong Y-CE, Xie A, Xu J, Yamamoto S, Yan J, Yao Z, Yoda K, Zhu R, Zinzen RP (2022) Fly cell atlas: a singlenucleus transcriptomic atlas of the adult fruit fly. Science 375:eabk2432. https://doi.org/ 10.1126/science.abk2432 16. Martelotto L (2020) ‘Frankenstein’ protocol for nuclei isolation from fresh and frozen tissue for snRNAseq. https://doi.org/10.17504/ PROTOCOLS.IO.3FKGJKW 17. Bru¨ckner K, Kockel L, Duchek P, Luque CM, Rørth P, Perrimon N (2004) The PDGF/ VEGF receptor controls blood cell survival in Drosophila. Dev Cell 7:73–84. https://doi. org/10.1016/j.devcel.2004.06.007
Chapter 5 Isolation of Tissue Macrophages in Adult Zebrafish Mireia Rovira, Jennifer Pozo, Magali Miserocchi, and Vale´rie Wittamer Abstract Tissue macrophages are essential components of the immune system that also play key roles in vertebrate development and homeostasis, including in zebrafish, which has gained popularity over the years as a translational model for human disease. Commonly, zebrafish macrophages are identified based on expression of fluorescent transgenic reporters, allowing for real-time imaging in living animals. Several of these lines have also proven instrumental to isolate pure populations of macrophages in the developing embryo and larvae using fluorescence-activated cell sorting (FACS). However, the identification of tissue macrophages in adult fish is not as clear, and robust protocols are needed that would take into account changes in reporter specificity as well as the heterogeneity of mononuclear phagocytes as fish reach adulthood. In this chapter, we describe the methodology for analyzing macrophages in various tissues in the adult zebrafish by flow cytometry. Coupled with FACS, these protocols further allow for the prospective isolation of enriched populations of tissue-specific mononuclear phagocytes that can be used in downstream transcriptomic and/or epigenomic analyses. Overall, we aim at providing a guide for the zebrafish community based on our expertise investigating the adult mononuclear phagocyte system. Key words Zebrafish, Macrophages, Microglia, Flow cytometry, mpeg1.1, p2ry12
1
Introduction For years, the teleost fish zebrafish (Danio rerio) has been extensively used as a vertebrate model by developmental biologists due to the transparency and external development of the embryo, and its genetic similarity and amenability, among other features [1]. More recently, zebrafish has also emerged as a powerful tool for the study of hematopoiesis and immunity [2, 3], contributing notably to new insights into the developmental and immune functions of leukocytes [4–8]. Initially conducted on embryos and larvae, immunological investigations in zebrafish have progressively extended to the adult, which is now increasingly used to study the contribution of leukocytes, with a particular interest on macrophages, in a range of physiological processes and diseases, such as cancer, infection, as well as organ regeneration [9–14]. Due to the
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_5, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
81
82
Mireia Rovira et al.
paucity of fish-specific antibodies and the poor antibody crossreactivity between fish and mammals, these studies mostly rely on the use of transgenic animals engineered to express a fluorophore under the control of lineage-specific regulatory elements [15], and allowing to identify zebrafish leukocytes in vivo, on tissue sections or by flow cytometry [3]. While an increasing number of stable lines with fluorescently labeled macrophages have been generated and validated for use in transparent embryos and larvae, one caveat is that their reliability in the adult is not systematic. One striking example is seen with the widely used collection of reporter lines generated using the mpeg1.1 promoter [16]. Indeed, although Tg (mpeg1.1:EGFP) or Tg(mpeg1.1:mCherry) transgenics can reliably serve as pan-macrophage reporters in the embryo and larvae, the situation changes in the adult due to the mpeg1.1 promoter being active in other cell types, including B lymphocytes and non-leukocytic metaphocytes [17–19]. As B cells and metaphocytes both appear at around the juvenile stage [19, 20], this precludes the sole use of mpeg1.1-driven reporters for the isolation of tissue macrophages in the adult. To overcome these limitations and facilitate future studies, we have established optimized protocols to analyze and isolate enriched populations of mononuclear phagocytes from the main zebrafish hematolymphoid organs. One protocol takes advantage of the unique combined expression patterns of the mpeg1.1:GFP [16] and the cd45:DsRed [21] fluorescent reporters, which are both readily available to the zebrafish community. Indeed, in double transgenic animals, GFP is expressed in mononuclear phagocytes, B lymphocytes and nonleukocytic metaphocytes [17], while DsRed labels all leukocytes with the exception of B cells [21]. This combination thus makes it possible to discriminate mononuclear phagocytes from B cells and/or metaphocytes in the brain, liver, skin, and the whole kidney marrow (WKM), the site of adult hematopoiesis in teleosts [22]. Finally, another approach makes use of the Tg(p2ry12:p2ry12-GFP) transgenic line [23] which, in combination with the cd45:DsRed reporter, offers additional means for the specific isolation of microglia, the macrophage population of the central nervous system. We detail the whole procedure step-by-step for each organ, from the dissociation and generation of a single-cell suspension to the creation of the flow cytometry gates allowing for the successful isolation of the populations of interest (Fig. 1).
2
Materials
2.1 Zebrafish Lines (See Notes 1 and 2)
1. Wild-type strain (AB*). 2. Tg(mpeg1.1:EGFP)gl22, here referred to as mpeg1.1:GFP. 3. Tg(ptprc:DsRed)sd3, here referred to as cd45:DsRed.
Zebrafish Tissue Macrophages
83
Fig. 1 Workflow of analysis of macrophages from the Whole Kidney Marrow (WKM), brain, liver and skin. Schematic representation of an adult zebrafish with the anatomical structures of interest labelled. Protocols to obtain single cell suspensions are optimized for every organ. The WKM is the only tissue that does not require enzymatic digestion and is mechanically homogenized in solution
4. TgBAC(p2ry12:p2ry12-GFP)hdb3, here referred to as p2ry12: p2ry12-GFP. 5. Tg(mpeg1.1:GFP; cd45:DsRed). 6. Tg(p2ry12:p2ry12-GFP; cd45:DsRed). For clarity, throughout the text, transgenic animals are referred to without allele designations.
84
Mireia Rovira et al.
2.2 Plasticware and Equipment
1. Scalpels. 2. Microdissection forceps. 3. Microdissection scissors. 4. Dissection scissors. 5. Petri dishes. 6. FACS tubes: 12 × 75 mm polystyrene round bottom tubes (5 mL). 7. Syringes, microfine insulin model (1 mL). 8. 20 G and 26 G needles. 9. 40-μM cell strainers. 10. Pipettes (5, 10, 25 mL). 11. Micropipettes and autoclaved pipette tips (P10, P200, P1000 μL). 12. Sterile unfiltered tips (P10, P200, P1000 μL). 13. Incubator, to be set at 33 °C. 14. Refrigerated swing-bucket centrifuge suitable for FACS tubes (10 min, 290 g at 4 °C). 15. Vortex mixer. 16. Aluminum foil. 17. Cell sorter, e.g. BD FACS ARIA III.
2.3 Media and Reagents
1. Sterile Dulbecco′s Phosphate Buffered Saline (DPBS) 1×. 2. DPBS 0.9×: 500 mL 1×(DPBS) + 55 mL deionized water. Keep on ice. 3. Liberase stock solution (Thermolysin Medium): prepare a stock solution of 5 mg/mL in DEPC water: reconstitute by injecting 1 mL of DEPC water in the 5 mg vial. Place the vial on ice to rehydrate and gently agitate the vial at 2–8 °C until the enzyme is completely dissolved (30 min maximum). Store in single use aliquots at -20 °C. Avoid repeated freezing and thawing. 4. Liberase solution: Prepare a 1/100 dilution of the liberase stock in 0.9× DPBS (calculate a final volume of 1 mL per sample). Keep on ice. 5. FACS Buffer: 2% inactivated fetal calf serum in 0.9× DPBS. Keep on ice. 6. Live dead dye (e.g., SYTOX™ Red).
Zebrafish Tissue Macrophages
3
85
Methods
3.1 Tubes to Prepare Before to Start
1. Label 1.5 mL sterile microcentrifuge tubes and 2 sets of corresponding FACS tubes in serial numbers (see Note 3). 2. Dispense 1 mL of Liberase solution into microcentrifuge tubes, except for WKM samples. Keep the tubes on ice. 3. For WKM samples, dispense 200 μL of ice-cold FACS Buffer into microcentrifuge tubes (see Note 4). 4. In one set of FACS tubes, add 3 mL of FACS Buffer, but leave the tubes for the WKM samples empty. Keep all tubes on ice.
3.2 Dissection of Adult Zebrafish Organs or Tissues
1. Euthanize the fish according to the permissions and ethical rules of local authorities. 2. Dissect the brain: Under a dissection microscope, place the fish on the side, cut the head with a scalpel, remove the eyes using forceps, and transfer the head in a Petri dish with ice-cold 0.9× DPBS. Hold the head with the forceps in one hand, and with the other, perform a small cut with the microdissection scissors in the posterior part of the skull. Be careful not to damage the brain. Break open the skull with the forceps and remove the right and left flank of the skull carefully keeping the head still with a forceps. Collect the brain with one forceps while keeping the head still with the other forceps. Take great care not to dissociate the olfactory bulbs, as they can easily detach from the telencephalon. Place the brain into the corresponding microcentrifuge tube containing 1 mL of ice-cold Liberase solution. 3. Dissect the liver: Make a ventral, midline incision from the anal fin towards the head using microdissection scissors under the microscope. Be careful not to damage the internal organs. Open the lateral muscles and look for the swim bladder, located in the upper cavity. Remove and discard it. If the fish is a female, remove the eggs present in the cavity to visualize the intestine, pancreas, liver, and spleen. Take the posterior intestine located near the anal fin with the forceps and pull towards the middle intestine, removing along all viscera of the body cavity as a full pluck. Carefully transfer into a Petri dish with ice cold 0.9× DPBS. Clean from fat and the gonads using the forceps. Stretch the intestine and locate the liver, which can be identified by its large size with lobed morphology, yellowish/pink color, and extensive vascularization. Place it directly into the corresponding microcentrifuge tube containing 1 mL of ice-cold Liberase solution. 4. Dissect the skin: With the fish placed on the side, make a vertical incision posterior to the operculum. With one forceps, pull the skin along the side of the fish from anterior to posterior while holding the rest of the body with the other forceps. If
86
Mireia Rovira et al.
necessary, carefully pull free any underlying fat that remain attached. Transfer the skin including the scales into the corresponding microcentrifuge tube containing 1 mL of ice-cold Liberase solution. 5. Dissect the WKM: Once all internal organs are removed, the kidney is exposed. Lay the fish on its back and locate the thin, translucent, and pigmented structure that runs along the axis of the vertebral column. If possible, discard the dorsal vein with forceps. Keep the forceps as parallel as possible to the spine to avoid contaminating the sample with bone or muscle debris. Carefully pull out the entire WKM away from the dorsal body wall, from the head (anterior) to the tail (posterior) kidney with the forceps. Place it directly into the corresponding microcentrifuge tube prefilled with 200 μL of ice-cold FACS Buffer (see Note 5). 3.3 Processing the Organs 3.3.1
WKM
1. Predissociate the WKM by gently pipetting up and down with a P200 unfiltered tip. Be careful as the sample can be sticky. 2. Add 300 μL of FACS Buffer and complete dissociation by passing the sample up and down multiple times through a 1 mL syringe with 26 G needle (see Note 6). 3. Filter the suspension through a 40-μm cell strainer directly into an ice-cold and empty FACS tube. Store on ice and protect from light by covering with aluminum foil.
3.3.2
Brain and Liver
1. Incubate in Liberase solution at 33 °C for 45–60 min (brain samples) or 25 min (liver samples) (see Note 7). 2. After 5 min, gently triturate with a P1000 to help physical cell dissociation. Make sure to use unfiltered tips to avoid the samples to stick to the inside of the tip. 3. Every 15 min, gently pipet the sample up and down several times with a P1000 for additional mechanical disruption. 4. Repeat until the tissue is fully dissociated. Do not let the sample digest longer than needed as it can damage cells. 5. At the end of the incubation, gently pass the sample through a 1-mL syringe with a 26 G needle, up and down multiple times. 6. Using the syringe, transfer the cell suspension to the corresponding FACS tube containing 3 mL of ice-cold FACS Buffer (see Note 8).
3.3.3
Skin
1. Using fine scissors, carefully cut the skin sample into small pieces in the microcentrifuge tube containing 1 mL Liberase solution and transfer at 33 °C. 2. After 10 min incubation, triturate the sample and pipet up and down with a P1000. Repeat every 10 min until the skin starts to disaggregate (see Note 9).
Zebrafish Tissue Macrophages
87
3. Once the skin is dissociated, which usually takes around 30 min, the suspension will appear greyish. 4. Keep the tube still until the scales sediment to the bottom. Transfer the cell suspension into a new empty microcentrifuge tube using a P1000 pipette. Take care not to transfer the scales. 5. Homogenize by gently pipetting up and down with a P1000 and a syringe with a 26 G needle. 6. Using the syringe, transfer the solution to the corresponding FACS tube containing 3 mL of ice-cold FACS Buffer (see Note 8). 3.4 Pelleting the Cells
1. Centrifuge the FACS tubes (except the WKM) containing the dissociated tissues (Subheading 3.3) at 290 g at 4 °C for 10 min (see Note 10). 2. Taking great care not to disturb the cell pellet, discard the supernatant using a P1000, leaving about 100 μL above the pellet. 3. To resuspend the cells, add 300 μL of ice-cold FACS Buffer. Disperse the pellet by gently pipetting up and down twice using a P1000.
3.5 Filtering and Staining for Dead Cell Discrimination
1. Using a P1000, pass the suspension through a 40-μm cell strainer into a new and empty FACS tube (second set). 2. Add live/dead dye (in our case 1/1000 Sytox Red dye with a final concentration 5 nM) to each tube. 3. Mix by flicking, and then keep the tubes on ice and in the dark until the scheduled time for analysis (see Note 11).
3.6
Data Collection
1. Collect data on a BD FACSAria III cytometer using the following gating strategy: 2. Start by drawing the first gate (P1) on cells displayed based on side (SSC-A, y axis bi-exponential scale) and forward (FSC-A, x axis linear scale) scattering characteristics. Exclude events with low FSC or high SSC (dead cells, debris and aggregates). This gate needs to be adjusted depending on the tissue (Fig. 2a). 3. To filter cell doublets out, display the P1 events based on the width of the side scatter (SSC-W, y axis linear scale) and the height of the side scatter (SSC-H, x axis bi-exponential scale). Draw the gate for the singlet population (P2) (Fig. 2b). 4. As a second step to ensure doublet exclusion, plot the P2 gate based on forward scatter-width (FSC-W, y axis linear scale) and forward scatter-height (FSC-H, x axis linear scale). Draw the gate for the singlet population (P3) (Fig. 2c). 5. Next, exclude Sytox Red positive dying cells by visualizing P3 on a plot of far red (x axis bi-exponential scale) versus side
88
Mireia Rovira et al.
Fig. 2 General gating strategy to identify single and live cells in adult zebrafish organs using flow cytometry. The WKM is shown as an example. (a) The P1 gate depicts events based on cell size and granularity. (b, c) Exclusion of cell doublets using the combination of pulse width and height side scatter or forward scatter parameters. (d) Selection of live cells (P4, or Sytox red negative fraction). Note that cells are displayed based on SSC-A (y-axis) and Sytox Red fluorescence (x-axis). As an alternative, cells can be plotted based on Sytox red fluorescence and GFP or DsRed (instead of SSC-A), allowing to visualize cell viability directly within the population of interest (not shown). Figures 3, 4, 5, and 6 will show the gating strategy for every organ to follow
scatter-area (SSC-A, y axis bi-exponential scale). Draw P4, which consists of single, live cells (Fig. 2d). 6. Plot the live P4 population based on side (SSC-A, y axis bi-exponential scale) and forward (FSC-A, x axis linear scale) scatter properties. In a tissue-specific manner, draw a new gate (P5) that excludes the lymphoid fraction and is enriched in myeloid cells (see Note 12) (Fig. 3 for WKM, Fig. 4 for brain, Fig. 5 for liver and Fig. 6 for skin).
Zebrafish Tissue Macrophages
89
Fig. 3 Representative dot plots and gating strategy for macrophage isolation in the WKM using Tg(mpeg1.1: GFP;cd45:DsRed). (a) Gating strategy to isolate lymphoid, progenitor and myeloid lineages using lightscattering characteristics. (b) Mature macrophages (red dots), reanalyzed by forward and side scatter, overlap the myeloid and progenitor fractions in the WKM. (c) The P5 gate visualizes the P4 population in a more restrictive manner, with most of the lymphoid population excluded. (d) Macrophages are identified as GFPint; DsRedhigh cells (red circle). Macs, macrophages
7. Once you have discriminated events by size, granularity, and viability, and delineated P5, gate on fluorescently labeled transgenic cells. Plot P5 based on GFP (y axis bi-exponential scale) and DsRed (x axis bi-exponential scale) fluorescence (Fig. 3 for WKM, Fig. 4 for brain, Fig. 5 for liver and Fig. 6 for skin).
Fig. 4 Representative dot plots and gating strategy for isolation of brain microglia using Tg(mpeg1.1:GFP; cd45:DsRed) or TgBAC(p2ry12:p2ry12-GFP;cd45:DsRed). (a) Gating strategy to analyze lymphoid and myeloid cells using light-scattering characteristics. (b) Microglial cells (red dots) reanalyzed by forward and side scatter, localize within the SSCint; FSCint myeloid gate. (c, e) The P5 gate displays the P4 population in Tg (mpeg1.1:GFP;cd45:DsRed) (c) and TgBAC(p2ry12:p2ry12-GFP;cd45:DsRed) (e) in a more restrictive manner. (d, f) The GFPlow; DsRedlow population (red circle) represents microglia in Tg(mpeg1.1:GFP;cd45:DsRed) fish (d), while in the brain of TgBAC(p2ry12:p2ry12-GFP;cd45:DsRed) animals, microglia are the GFP+; DsRed+ double positive population (red circle) (f). MG microglia
Zebrafish Tissue Macrophages
91
Fig. 5 Representative dot plots and gating strategy for isolation of liver macrophages using Tg(mpeg1.1:GFP; cd45:DsRed). (a) Gating strategy to isolate lymphoid and myeloid lineages using light-scattering characteristics. (b) Macrophages (red dots) reanalyzed by forward and side scatter, are contained within the myeloid fraction. (c) After setting up the P4 gate, P5 is a more restrictive gate into the myeloid fraction. (d) Liver macrophages represent a well-defined population of GFP+; DsRed+ double positive cells (red circle). Macs macrophages
8. Make sure to use unlabeled samples (from nontransgenic fish) and single-fluorescent samples (from single GFP and DsRed reporter carriers) to set appropriate detector gains (PMT voltages) and thresholds for fluorescence channels.
92
Mireia Rovira et al.
Fig. 6 Representative dot plots and gating strategy for isolation of skin macrophages using Tg(mpeg1.1:GFP; cd45:DsRed). (a) Gating strategy to isolate lymphoid and myeloid lineages using light-scattering characteristics. Note that this tissue is very different from the other examples given. (b) Mononuclear phagocytes (red dots) reanalyzed by forward and side scatter, localize within the myeloid fraction. (c) The P5 gate visualizes the P4 population in a more restrictive manner, with the lymphoid population excluded. (d) In this setting, skin macrophages are identified as GFP+; DsRed+ double positive (red circle). Macs macrophages
3.7 Expected Outcomes for Each Tissue
1. WKM, using the Tg(mpeg1.1:GFP, cd45:DsRed) line: In this tissue, differentiated mononuclear phagocytes represent a discrete mpeg1.1+ cell population with a specific scatter profile overlapping the myeloid and progenitor fractions (Fig. 3a, b). The majority of the cells shows the characteristics of macrophages, with kidney-shaped nuclei and vacuoles [21]. As the
Zebrafish Tissue Macrophages
93
site of adult hematopoiesis, the WKM also contains large amounts of B lymphocytes that share expression of the mpeg1.1:GFP reporter with macrophages [17], which can thus affect their analysis. Therefore, exclusion of unwanted GFP+ B lymphocytes is firstly achieved by gating on the combined myeloid and scatter fractions (Fig. 3c) and, secondly, using Tg(mpeg1.1:GFP; cd45:DsRed) double transgenic animals, where GFP+; DsRed+ mononuclear phagocytes are discerned from GFP+; DsRed- B cells based on differential DsRed expression. Importantly, WKM mononuclear phagocytes can be further divided into two subpopulations, identified as macrophages (GFPint; DsRedhigh) and putative DC-like cells (GFPhigh; DsRedhigh), respectively (Fig. 3d) (see Note 14). GFP- DsRed+ cells comprise a mix of hematopoietic progenitors, T cells, eosinophils, and neutrophils. The latter is abundant in the WKM [24]. 2. Brain, using the Tg(mpeg1.1:GFP, cd45:DsRed) line: Brain samples contain two main populations of GFP+; DsRed+ cells that can be discriminated based on differential levels of fluorophore expression: microglia, the brain macrophages, which are GFPlow; DsRedlow, and a less abundant GFPhigh; DsRedhigh population (Fig. 4d), which mostly comprises DC-like cells and monocytes (Rovira, Ferrero and Wittamer, unpublished) (see Note 14). In this setting, GFP+; DsRed- cells are B lymphocytes while GFP- DsRed+ cells consist of a mixed population of T lymphocytes, NK cells, and neutrophils, among others. 3. Brain, using the Tg(p2ry12:p2ry12-GFP; cd45:DsRed) line: Microglia can also be easily identified based on expression of the p2ry12:p2ry12-GFP transgene, the purinergic receptor p2ry12 being widely accepted as a canonical microglia marker [25], including in zebrafish [26, 27]. Importantly, expression of the p2ry12 transgene is coupled to CFP expression in the lens (serving as a transgenesis marker), which can make analysis of microglia challenging. Using animals carrying both the p2ry12:p2ry12-GFP and the cd45:DsRed reporters allows to overcome this limitation (Fig. 4f), as in this setting microglia are GFP+; DsRed+ and can be discriminated from unwanted CFP+; DsRed- lens cells (see Note 13). Importantly, due to low p2ry12:p2ry12-GFP transgene expression in microglia (especially in comparison to the mpeg1.1:GFP reporter), fine adjustment of the GFP voltage is required to achieve efficient separation between microglia (GFPlow; DsRed+) and the remaining immune cells (identified as GFP- DsRed+). 4. Liver, using the Tg(mpeg1.1:GFP, cd45:DsRed) line: The liver contains a well-defined population of mpeg1.1 positive cells, which mostly comprises macrophages [28]. These cells are GFP+; DsRed+ and are easily separated from GFP+; DsRed- B
94
Mireia Rovira et al.
lymphocytes (Fig. 5d). The GFP- DsRed+ population contains the remaining leukocytes. 5. Skin, using the Tg(mpeg1.1:GFP, cd45:DsRed) line: The skin contains a heterogenous population of mpeg1.1+ cells, of which B lymphocytes and metaphocytes account for a high proportion [19]. These cells are discerned from skin macrophages based on differential DsRed expression. Accordingly, skin macrophages are GFP+DsRed+, and B cells and metaphocytes are GFP+DsRed- (see Note 14). Drawing a gate on the myeloid fraction (P5) will ensure enrichment for GFP+DsRed+ macrophages (Fig. 6c).
4
Notes 1. Analyses are conducted on fish between 4 and 9 months of age. Animal experiments must conform to national and institutional regulations. 2. Sample preparation for flow cytometry analyses does not require antibody staining since the protocol relies on detection of the endogenous fluorescence in the cells of interest. However, not fluorescent (obtained from nontransgenic fish) and single-fluorescent (obtained from single EGFP and DsRed reporters) control samples are needed to set appropriate laser voltages that identify clear negative and positive populations. 3. The sample preparation time before analysis at the flow cytometer is around 3 h. In practice, this timing will depend on the sample size. If large numbers of samples have to be analyzed, we recommend that two people participate: one in charge of collecting the organs while the second processes the samples. The protocol has no stopping step until the end and the experimenter must proceed quickly. Transgenic fish should be screened in advance and selected the day prior the experiment. If necessary, validation of the reporter lines can be conducted on anesthetized adult fish placed under a fluorescence binocular. Transgenic GFP+ and DsRed+ cells are easily visualized in the skin and gills of Tg(mpeg1.1:GFP) and Tg(cd45: DsRed) animals, respectively. The Tg(p2ry12:p2ry12-GFP) line is identified by a lens-specific expression of cerulean fluorescent protein (CFP) that serves as a transgenesis marker and is easily visible in adult animals. The fish to sample the next morning can be isolated in breeding tanks overnight. 4. Mechanical disruption of soft tissues like the kidney is sufficient to release cells into a single-cell suspension. 5. The order of the dissection of the tissues or organs will depend on your interests. However, some tissues are more sensitive to degradation such as the WKM and the brain. If several organs
Zebrafish Tissue Macrophages
95
are collected from the same animal, we recommend proceeding as followed: cut the head first, place it into a Petri dish with 0.9× DPBS, and keep on ice until processing. Remove the viscera as a full pluck and keep it in cold 0.9× DPBS until processing. Collect the WKM, and then dissect the brain in cold 0.9× DPBS. Quickly peel off the skin. Finally, dissect the liver in cold 0.9× DPBS. 6. To fully dissociate the WKM, always use a P200 and a syringe sequentially. First pipet up and down with the P200 until no clumps are visible. This will avoid clogging the syringe in the subsequent step. The use of 26 GA needles is necessary for complete tissue homogenization. The mixing up and down must be gentle to not harm the cells. The inside diameter of the needle must not be smaller than the diameter of a cell, as this will result in cell lysis. 7. The time required for optimal dissociation using enzymatic digestion will differ between tissues and will depend on the tissue size. Typically, for an adult brain it takes 40–50 min, and 20–30 min for the liver. During incubation, check your samples regularly to prevent overdigestion. 8. Dilution of the homogenized tissue into 3 mL of ice-cold FACS Buffer will ensure enzymatic inactivation of the liberase. 9. To speed up the dissociation process, pass the sample through a syringe with a 20 GA needle. 10. When processing large numbers of samples, we recommend centrifuging maximum 10 tubes at a time as proceeding quickly will prevent detachment of the cell pellets. 11. Always keep the samples on ice to stop cell death and protected from light to prevent loss of fluorescence. As the cells are not fixed, the samples cannot be stored more than a few hours so proceed with the flow cytometry analysis without delay. 12. Using flow cytometry, each of the major hematopoietic lineages can be resolved by light-scatter characteristics. In the zebrafish WKM, the main distinct scatter populations are termed “lymphoid,” “progenitor,” and “myeloid”. Mature mononuclear phagocytes are identified as a distinct forward scatter (FSC)high side scatter (SSC)int population, overlapping the conventional myeloid and precursor fractions. The precursor population is not present in other zebrafish organs, and mononuclear phagocytes largely localize in the myeloid gate. 13. Because eyes are separated from the brain during dissection, contamination of p2ry12:p2ry12-GFP microglia by CFP lens cells in the single cell suspension should be minimal (unlike in the case of cell suspensions produced from embryos and/or larvae).
96
Mireia Rovira et al.
14. Limitations: While tissue macrophages will undoubtedly represent the major population analyzed with this protocol, it is important to keep in mind that other mononuclear phagocyte subsets, such as monocytes and dendritic cells, will likely also be comprised in the isolated population of mpeg1.1+; cd45+ myeloid cells. Although our approach also permits to identify putative dendritic cell-like populations, notably in the brain and in the WKM, whether these cells represent the zebrafish counterparts of mammalian DCs remains to be determined. Likewise, zebrafish monocytes have not been characterized yet. Therefore, more markers and new transgenic lines are needed to discriminate the different subsets of the mononuclear phagocyte compartment in the zebrafish model. Finally, it is also possible that the mpeg1.1 promoter fails to label all macrophages in transgenic fish, although this appears unlikely based on available transcriptomic data. References 1. Link BA, Megason SG (2008) Zebrafish as a model for development. In: Conn PM (ed) Sourcebook of models for biomedical research. Humana Press, Totowa, pp 103–112. https://doi.org/10.1007/978-159745-285-4_13 2. Gore AV, Pillay LM, Venero Galanternik M, Weinstein BM (2018) The zebrafish: a fintastic model for hematopoietic development and disease. Wiley Interdisc Rev 7(3). https://doi. org/10.1002/wdev.312 3. Stachura DL, Traver D (2016) Cellular dissection of zebrafish hematopoiesis. Methods Cell Biol 133:11–53. https://doi.org/10.1016/ bs.mcb.2016.03.022 4. Wattrus SJ, Smith ML, Rodrigues CP, Hagedorn EJ, Kim JW, Budnik B, Zon LI (2022) Quality assurance of hematopoietic stem cells by macrophages determines stem cell clonality. Science 377:7. https://doi.org/10.1126/sci ence.abo4837 5. Cambier CJ, Takaki KK, Larson RP, Hernandez RE, Tobin DM, Urdahl KB, Cosma CL, Ramakrishnan L (2014) Mycobacteria manipulate macrophage recruitment through coordinated use of membrane lipids. Nature 505(7482):218–222. https://doi.org/10. 1038/nature12799 6. Volkman HE, Pozos TC, Zheng J, Davis JM, Rawls JF, Ramakrishnan L (2010) Tuberculous granuloma induction via interaction of a bacterial secreted protein with host epithelium. Science 327(5964):466–469. https://doi.org/ 10.1126/science.1179663
7. Hughes AN, Appel B (2020) Microglia phagocytose myelin sheaths to modify developmental myelination. Nat Neurosci 23(9):1055–1066. https://doi.org/10.1038/s41593-0200654-2 8. Peri F, Nusslein-Volhard C (2008) Live imaging of neuronal degradation by microglia reveals a role for v0-ATPase a1 in phagosomal fusion in vivo. Cell 133(5):916–927. https:// doi.org/10.1016/j.cell.2008.04.037 9. Shwartz A, Goessling W, Yin C (2019) Macrophages in zebrafish models of liver diseases. Front Immunol 10:2840. https://doi.org/ 10.3389/fimmu.2019.02840 10. Marques IJ, Lupi E, Mercader N (2019) Model systems for regeneration: zebrafish. Development 146(18). https://doi.org/10.1242/dev. 167692 11. Simoes FC, Cahill TJ, Kenyon A, Gavriouchkina D, Vieira JM, Sun X, Pezzolla D, Ravaud C, Masmanian E, Weinberger M, Mayes S, Lemieux ME, Barnette DN, Gunadasa-Rohling M, Williams RM, Greaves DR, Trinh LA, Fraser SE, Dallas SL, Choudhury RP, Sauka-Spengler T, Riley PR (2020) Macrophages directly contribute collagen to scar formation during zebrafish heart regeneration and mouse heart repair. Nat Commun 11(1):600. https://doi.org/10. 1038/s41467-019-14263-2 12. Yan C, Brunson DC, Tang Q, Do D, Iftimia NA, Moore JC, Hayes MN, Welker AM, Garcia EG, Dubash TD, Hong X, Drapkin BJ, Myers DT, Phat S, Volorio A, Marvin DL, Ligorio M,
Zebrafish Tissue Macrophages Dershowitz L, McCarthy KM, Karabacak MN, Fletcher JA, Sgroi DC, Iafrate JA, Maheswaran S, Dyson NJ, Haber DA, Rawls JF, Langenau DM (2019) Visualizing engrafted human cancer and therapy responses in immunodeficient zebrafish. Cell 177(7): 1903–1914. e1914. https://doi.org/10. 1016/j.cell.2019.04.004 13. Bevan L, Lim ZW, Venkatesh B, Riley PR, Martin P, Richardson RJ (2020) Specific macrophage populations promote both cardiac scar deposition and subsequent resolution in adult zebrafish. Cardiovasc Res 116(7):1357–1371. https://doi.org/10.1093/cvr/cvz221 14. van Leeuwen LM, van der Kuip M, Youssef SA, de Bruin A, Bitter W, van Furth AM, van der Sar AM (2014) Modeling tuberculous meningitis in zebrafish using Mycobacterium marinum. Dis Model Mech 7(9):1111–1122. https://doi.org/10.1242/dmm.015453 15. Choe CP, Choi S-Y, Kee Y, Kim MJ, Kim S-H, Lee Y, Park H-C, Ro H (2021) Transgenic fluorescent zebrafish lines that have revolutionized biomedical research. Lab Anim Res 37(1). https://doi.org/10.1186/s42826021-00103-2 16. Ellett F, Pase L, Hayman JW, Andrianopoulos A, Lieschke GJ (2010) mpeg1 promoter transgenes direct macrophage-lineage expression in zebrafish. Blood 117(4):e49–e56. https://doi.org/10. 1182/blood-2010-10-314120 17. Ferrero G, Gomez E, Lyer S, Rovira M, Miserocchi M, Langenau DM, Bertrand JY, Wittamer V (2020) The macrophage-expressed gene (mpeg) 1 identifies a subpopulation of B cells in the adult zebrafish. J Leukoc Biol 107(3):431–443. https://doi.org/10.1002/ JLB.1A1119-223R 18. Moyse BR, Richardson RJ (2020) A population of injury-responsive lymphoid cells expresses mpeg1.1 in the adult zebrafish heart. Immunohorizons 4(8):464–474. https://doi.org/10.4049/immunohorizons. 2000063 19. Kuil LE, Oosterhof N, Ferrero G, Mikulasova T, Hason M, Dekker J, Rovira M, van der Linde HC, van Strien PM, de Pater E, Schaaf G, Bindels EM, Wittamer V, van Ham TJ (2020) Zebrafish macrophage developmental arrest underlies depletion of microglia and reveals Csf1r-independent metaphocytes. elife 9. https://doi.org/10.7554/eLife.53403 20. Page DM, Wittamer V, Bertrand JY, Lewis KL, Pratt DN, Delgado N, Schale SE, McGue C, Jacobsen BH, Doty A, Pao Y, Yang H, Chi NC, Magor BG, Traver D (2013) An evolutionarily conserved program of B-cell development and
97
activation in zebrafish. Blood 122(8):e1–e11. https://doi.org/10.1182/blood-201212-471029 21. Wittamer V, Bertrand JY, Gutschow PW, Traver D (2011) Characterization of the mononuclear phagocyte system in zebrafish. Blood 117(26):7126–7135. https://doi.org/10. 1182/blood-2010-11-321448 22. Traver D, Paw BH, Poss KD, Penberthy WT, Lin S, Zon LI (2003) Transplantation and in vivo imaging of multilineage engraftment in zebrafish bloodless mutants. Nat Immunol 4(12):1238–1246. https://doi.org/10.1038/ ni1007 23. Sieger D, Moritz C, Ziegenhals T, Prykhozhij S, Peri F (2012) Long-range Ca2+ waves transmit brain-damage signals to microglia. Dev Cell 22(6):1138–1148. https://doi. org/10.1016/j.devcel.2012.04.012 24. Tang Q, Iyer S, Lobbardi R, Moore JC, Chen H, Lareau C, Hebert C, Shaw ML, Neftel C, Suva ML, Ceol CJ, Bernards A, Aryee M, Pinello L, Drummond IA, Langenau DM (2017) Dissecting hematopoietic and renal cell heterogeneity in adult zebrafish at single-cell resolution using RNA sequencing. J Exp Med 214(10):2875–2887. https://doi. org/10.1084/jem.20170976 25. Paolicelli RC, Sierra A, Stevens B, Tremblay M-E, Aguzzi A, Ajami B, Amit I, Audinat E, Bechmann I, Bennett M, Bennett F, Bessis A, Biber K, Bilbo S, Blurton-Jones M, Boddeke E, Brites D, Broˆne B, Brown GC, Butovsky O, Carson MJ, Castellano B, Colonna M, Cowley SA, Cunningham C, Davalos D, De Jager PL, de Strooper B, Denes A, Eggen BJL, Eyo U, Galea E, Garel S, Ginhoux F, Glass CK, Gokce O, Gomez-Nicola D, Gonza´lez B, Gordon S, Graeber MB, Greenhalgh AD, Gressens P, Greter M, Gutmann DH, Haass C, Heneka MT, Heppner FL, Hong S, Hume DA, Jung S, Kettenmann H, Kipnis J, Koyama R, Lemke G, Lynch M, Majewska A, Malcangio M, Malm T, Mancuso R, Masuda T, Matteoli M, McColl BW, Miron VE, Molofsky AV, Monje M, Mracsko E, Nadjar A, Neher JJ, Neniskyte U, Neumann H, Noda M, Peng B, Peri F, Perry VH, Popovich PG, Pridans C, Priller J, Prinz M, Ragozzino D, Ransohoff RM, Salter MW, Schaefer A, Schafer DP, Schwartz M, Simons M, Smith CJ, Streit WJ, Tay TL, Tsai L-H, Verkhratsky A, von Bernhardi R, Wake H, Wittamer V, Wolf SA, Wu L-J, Wyss-Coray T (2022) Microglia states and nomenclature: a field at its crossroads. Neuron 110(21):3458–3483. https://doi. org/10.1016/j.neuron.2022.10.020
98
Mireia Rovira et al.
26. Ferrero G, Mahony CB, Dupuis E, Yvernogeau L, Di Ruggiero E, Miserocchi M, Caron M, Robin C, Traver D, Bertrand JY, Wittamer V (2018) Embryonic microglia derive from primitive macrophages and are replaced by cmyb-dependent definitive microglia in zebrafish. Cell Rep 24(1):130–141. https://doi.org/10.1016/j.celrep.2018. 05.066 27. Mazzolini J, Le Clerc S, Morisse G, Coulonges C, Kuil LE, van Ham TJ, Zagury JF, Sieger D (2019) Gene expression profiling reveals a conserved microglia signature in larval zebrafish. Glia. https://doi.org/10.1002/glia. 23717
28. Guilliams M, Bonnardel J, Haest B, Vanderborght B, Wagner C, Remmerie A, Bujko A, Martens L, Thone T, Browaeys R, De Ponti FF, Vanneste B, Zwicker C, Svedberg FR, Vanhalewyn T, Goncalves A, Lippens S, Devriendt B, Cox E, Ferrero G, Wittamer V, Willaert A, Kaptein SJF, Neyts J, Dallmeier K, Geldhof P, Casaert S, Deplancke B, Ten Dijke P, Hoorens A, Vanlander A, Berrevoet F, Van Nieuwenhove Y, Saeys Y, Saelens W, Van Vlierberghe H, Devisscher L, Scott CL (2022) Spatial proteogenomics reveals distinct and evolutionarily conserved hepatic macrophage niches. Cell 185(2): 379–396 e338. https://doi.org/10.1016/j. cell.2021.12.018
Chapter 6 Genetic and Immunohistochemistry Tools to Visualize Rat Macrophages In Situ Stephen Huang, Dylan Carter-Cusack, Emma Maxwell, Omkar L. Patkar, Katharine M. Irvine, and David A. Hume Abstract Macrophages contribute to many aspects of development and homeostasis, innate and acquired immunity, immunopathology, and tissue repair. Every tissue contains an abundant resident macrophage population. Inflammatory stimuli promote the recruitment of monocytes from the blood and their adaptation promotes the removal of the stimulus and subsequent restoration of normal tissue architecture. Dysregulation of this response leads to chronic inflammation and tissue injury. In many tissues, their differentiation and survival are dependent on the colony stimulating factor 1 receptor (CSF1R) signalling axis, which is highly conserved across all vertebrates. Complete loss of either CSF1R or its cognate ligands, colony stimulating factor 1 (CSF1), and interleukin 34 (IL-34), results in the loss of many tissue-resident macrophage populations. This provides a useful paradigm to study macrophages. There are many tools used to visualize tissue-resident macrophages and their precursors, monocytes, in mice and humans. Particularly in mice there are genetic tools available to delete, enhance and manipulate monocytes and macrophages and their gene products to gain insight into phenotype and function. The laboratory rat has many advantages as an experimental model for the understanding of human disease, but the analytical resources are currently more limited than in mice. Here, we describe available genetic models, antibodies, and immunohistochemistry (IHC) methods that may be used to visualize tissue-resident macrophages in rats. Key words IBA1, CSF1R, CSF1, IL-34, Reporter rat, CRISPR, MPS
1
Introduction The mononuclear phagocyte system (MPS) is a family of cells including resident-tissue macrophages in every organ of the body, blood monocytes, and committed progenitors in the bone marrow. The development of macrophages from committed progenitors occurs in three successive waves, from the yolk sac, the fetal liver, and the bone marrow. In adult tissues, resident macrophages adapt to their microenvironment to perform tissue-specific functions in homeostasis [1]. Much of our current knowledge of macrophage
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_6, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
99
100
Stephen Huang et al.
ontogeny and function is derived from analysis of inbred mice [2]. Rats are widely recognized as more informative and tractable models for many human diseases, including cardiovascular, neurological, cancer, diabetes, respiratory, and inflammatory disease [3]. Their larger size enables easier surgical procedures and high resolution imaging [4, 5]. However, research on the rat MPS is constrained by more limited availability of resources including genetic models and well-defined antibodies [6]. In vertebrates, differentiation, proliferation, and survival of most macrophages is dependent on CSF1R signalling and its ligands CSF1 and IL-34. Disrupted CSF1R signalling was likely first described in rats [7]. The original toothless rat (Csf1tl/tl) was a spontaneous mutant offspring from an outbred colony of MendelOsborne rats identified in 1974 by Cotton and Gaines by the lack of tooth eruption. The causative mutation located in the Csf1 locus was identified in 2002 [8, 9]. Whereas a mutation in the mouse Csf1 locus (Csf1op/op) has been widely studied, we are not aware of ongoing studies of the Csf1tl/tl rat. The hallmark of both mouse and rat Csf1 mutations is the failure of tooth eruption, which is related to the absence of bone resorbing osteoclasts. In mice, there is progressive recovery of osteoclasts with age whereas in rats, the loss is unremitting [6]. There are other rat mutations that manifest as osteopetrosis or toothless but they are not deficient in macrophages per se (e.g., Incisor absent (ia) rat carrying a Plekhm1 loss of function mutation [10, 11], microphthalmia (mib) rat carrying a Mitf loss of function [12, 13] and the osteopetrotic (op) rat that has an unknown lesion in chromosome 10 (unrelated to the Csf1op/op mouse) [14, 15]). In studies of MPS biology in mice, transgenic reporter lines and antibodies against defined surface markers are used to locate macrophages within tissues and to identify or purify them by flow cytometry, magnetic bead or other separation technologies. Separation based upon surface markers/reporter transgenes has enabled transcriptomic analysis of resident MPS cells from a wide range of mouse tissues [16]. Magnetic bead separation kits are available for the rat based upon an anti-CD11b/c antibody (Miltenyi 130–105634), but it is unclear whether these markers are universally expressed or macrophage-restricted. The resource required for validation of any marker is an animal that is definitively macrophagedeficient and/or a transgenic reporter that is universal and macrophage-restricted. Localization in tissues requires development of specific antibodies and optimization of approaches to detect their cognate antigen in situ and to correlate the location with tissue morphology [17–19]. Several monoclonal antibodies recognizing protein markers expressed by rat macrophages were developed in the 1980s. There has been very little progress since that time and there are
Genetic and IHC Tools to Visualise Rat Macrophages
101
very significant gaps that need to be filled [6, 20, 21]. For example, there is no equivalent of the F4/80 antibody widely used in the mouse, although the mRNA (Adgre1) is highly expressed by rat macrophages and an anti-pig ADGRE1 antibody was recently developed [22, 23]. There is also a lack of antibodies to enable purification and analysis of marrow progenitors from rat. Generation of new antibodies for the rat MPS is an ongoing priority in our group. There has been progress in generating genetic resources for rats as outlined in Table 1 as well as models that use cre-recombinase (Table 2). The characterisation of rats bearing a null mutation in the Csf1r locus (Csf1rko) confirmed the requirement for CSF1 and IL-34 in the generation of macrophages in the embryo and their importance in postnatal development [24, 25]. A transgenic line in which the red fluorescent protein mApple is driven by the Csf1r promoter enabled detection of resident tissue macrophages and the response to administration of recombinant CSF1 [26]. However, as in the mouse, the transgene is also expressed in neutrophils which express Csf1r mRNA but no protein, and in B cells. The mApple transgene is sufficiently bright to enable wholemount imaging of fresh tissues under a confocal microscope (Fig. 1). The mApple reporter can also be detected using antibodies against red fluorescent protein as described below. Extensive analysis of the tissues of the rat Csf1rko confirmed the macrophage-restricted expression of IBA1 (encoded by Aif1). IBA1 is used in many species as a marker for the MPS cells of the brain, microglia and brain-associated macrophages. Allograft inhibitory factor 1 (AIF1) was originally identified in chronically rejecting rat cardiac allografts and localized to macrophages [27]. The rat Csf1rko is associated with almost complete loss of immunoreactive IBA1 in nonimmune tissues throughout development [24] and is also deficient in expression of the Csf1r-mApple reporter. The restricted expression of Aif1 has been used in the recent generation of a knock-in GFP reporter in this locus in rats [28]. We used the Csf1rko rat to test the application and validity of a commercial rabbit monoclonal anti-rat CSF1R antibody (SP211 Abcam; ab183316). While there was some evidence of staining of macrophage-like cells, there was also extensive staining of neuronal structures in the embryonic brain that was unaffected by the Csf1rko (unpublished). In this chapter we will describe the immunohistochemical (IHC) procedures used routinely in our lab to detect antigen in rat tissues. There are many differing protocols for IHC but the typical procedure can be broadly simplified into three steps: (1) preanalytical phase: tissue processing, (2) analytical phase: antigen detection, and (3) postanalytical phase: visual analysis. Although the methods are largely generic, there are details related to tissue processing and antigen retrieval that are idiosyncratic for each antibody.
102
Stephen Huang et al.
Table 1 Available rat strains to study the MPS
Strain name SD-Csf1r
tm
(EGFP)+/-
DA-Csf1rtm (EGFP)+/-
Csf1r-mApple rat
Strain Generation of genetic modification background Homologous recombination in embryonic stem cells with an EGFP-PGK-neomycin cassette targeting Exon 1 of Csf1r delivered by electroporation
Phenotype description
Sprague Loss of CSF1R expression Dawley (SD) [25]
Multi-generational (>7 generations) Dark Agouti (DA) backcross of SD-Csf1r Csf1rtm (EGFP)+/with Dark Agouti
Loss of CSF1R expression [24]
Pronuclear injection of transgene containing the 7.2 kb fms promoter driving mApple expression
SD
mApple expression in macrophages, granulocytes and B cells [31]
Fischer 344 (F344)
Loss of CSF1 expression [9]
F344tl(Toothless) Spontaneous mutation: 10 bp insertion at start of Csf1 gene IL-34-/-
CRISPR-mediated insertion of stop SD/Crl-France Loss of IL-34 expression [32] codon in exon 3.
F344/DuCrlIl-34tm(1,3)
CRISPR-mediated deletion of exon F344/DuCrl 2
Loss of IL-34 expression [Patkar and Huang, Unpublished].
SD-Aif1tm
CRISPR-mediated insertion of GFP SD with P2A linker immediately downstream of Aif1 (IBA1) translation initiation site
Expression of GFP in circulating IBA1+ monocytes and tissue resident macrophages. [28]
Sleeping Beauty transposon Wistar-Kyoto mediated insertion of a transgene (WKY) containing human CD68 promoter driving GFP expression
Expression of GFP in circulating monocytes and all macrophages [33]
(EGFP)Apps
WKY-hCD68GFP
2 2.1
Materials Equipment
1. Tissue processor. 2. Paraffin embedding station. 3. Decloaking chamber. 4. Microtome. 5. Water bath. 6. Hybridization/dry oven. 7. Humidified slide chamber. 8. Microwave. 9. Microscope.
Genetic and IHC Tools to Visualise Rat Macrophages
103
Table 2 Rat strains that require interbreeding to study the rat MPS
Strain name
Generation of genetic modification
Strain background Phenotype
LE-Tg(Cx3cr1-Cre/ ERT2)
Long Evans Expression of tamoxifen Pronuclear injection of BAC (LE) SD inducible Cre recombinase containing Cre-ERT2 F344 driven by endogenous targeted to Cx3Cr1 region Cx3cr1 start codon. (7–9 into Long Evans and Sprague transgene insertions) [34] Dawley embryos. [Unpublished]. Fischer 344 background was produced through backcrossing of the Long Evans line
LE-Tg(Gt(ROSA) 26SorCAGtdTomato)24Jfhy
Pronuclear injection of BAC containing CAG promoterLoxP-Neo-4xSTOP-LoxPtdTomato
LE
Expression of tdTomato where Cre recombinase is expressed. Insertion identified on Chr17: 13,835,107–13,835,780 Nb: Authors also created 2 other lines differentiated by the name 9Jfhy and 14Jfhy. These lines do not contain LoxP-Stop
F344-Tg (CAG-loxP-STOPloxP-ZsGreen)
Microinjection of DNA fragment containing CAG promoter-LoxP-3xSTOPloxP- Zsgreen-SV40polyA into zygotes
F344/ NHsd
Expression of GFP where Cre recombinase is expressed [35]
LE-Rosa26 (CAG-LSLTdTomato)
Microinjection of DNA fragment containing CAG promoter-LoxP-3xSTOPloxP- TdTomato-SV40polyA into zygotes
LE
Expression of RFP where Cre recombinase is expressed [Bryda E. Unpublished]
SD-Tg(CAG-loxPmCherry-loxPEGFP)
Microinjection of DNA fragment containing CAG promoter-LoxP-mCherryloxP- eGFP into zygotes
SD
[Ying Q-L. Unpublished]
2.2 Solutions and Reagents
1. Reverse osmosis (RO) water. 2. Xylene. 3. 100%, 95%, 70% Ethanol in RO water. 4. Phosphate buffered saline (PBS) pH 7.0. 5. Tris-buffered saline (TBS): 20 mM Tris–HCl, 150 mM NaCl in RO water, pH 7.6.
104
Stephen Huang et al.
Fig. 1 Whole mount imaging of the Csf1r-mApple reporter rat highlights the abundance of Csf1r-expressing cells. Localization of mApple positive cells in meninges, lung, liver, white adipose tissue (WAT), testis, and spleen from a wildtype rat. Images were captured using an Olympus FV3000 confocal microscope
6. 4% paraformaldehyde (PFA) in PBS pH 7.4 (see Note 1). 7. Citrate buffer: 10 mM sodium citrate in RO water, pH 6.0. 8. Tris/EDTA buffer: 10 mM Tris, 1 mM EDTA buffer in RO water, pH 9.0. 9. Peroxidase block: 0.3% hydrogen peroxide in TBS. 10. Wash buffer: 0.05% Tween-20 in TBS. 11. DAB peroxidase substrate solution. 12. Hematoxylin (see Note 25). 13. Antibody of choice (see Table 3). 14. Antimouse and/or antirabbit HRP conjugated polymer, depending on choice of primary antibody (see Table 4). 15. Blocking Buffer: antibody dependent (see Table 4).
Genetic and IHC Tools to Visualise Rat Macrophages
105
Table 3 Antibodies used in this protocol Target
Macrophage subsets
Clone (species)
Distributor
IBA1
Nearly all tissue resident macrophage, microglia and circulating monocytes (& Sertoli cells)
Polyclonal (rabbit)
Novachem (019-19741)
CD163
Splenic red pulp macrophages, hematopoietic Monoclonal island macrophages, Kupffer cells, EPR19518 perivascular and meningeal macrophages (rabbit)
Abcam (ab182422)
MHCII
Monocytes and macrophages (& Dendritic cells, B Cells)
Monoclonal OX-6 (mouse)
Abcam (ab23990)
CD206 (MRC1)
Subset of tissue resident macrophages (& endothelial cells)
Polyclonal (rabbit)
Abcam (ab64693)
CD4
Subset of tissue resident macrophages (& T cells)
Recombinant Abcam monoclonal CAL4 (ab237722) (rabbit)
CD11b
Monocytes and macrophages (& neutrophils) Recombinant Monoclonal ERP1344 (rabbit)
Abcam (ab133357)
CD68
Nearly all macrophages, monocytes (& granulocytes)
Monoclonal ED1 (mouse)
Serotec (MCA341GA)
CD31 (Pecam1)
Subset of macrophages (& endothelial cells)
Recombinant polyclonal RM1006 (rabbit)
Abcam (ab281583)
RFP
All CSF1R dependant macrophages (specific to the CSF1R-mApple model)
Polyclonal (rabbit)
Abcam (ab124754)
2.3
Consumables
1. Tissue processing/embedding cassettes. 2. Microtome blades. 3. Superfrost Plus Adhesion slides. 4. Heat-resistant slide rack/container. 5. DPX mounting medium. 6. Coverslips. 7. PAP hydrophobic barrier pen.
3
Methods See Fig. 2 for an overview.
Method 2
CD4
CD11b (see Method 1 Note 30)
1% BSA in TBS
RFP
Method 1
1% BSA in TBS
Background Sniper with 1% goat serum
2h
10% goat serum in 1% BSA in TBS
30 min
30 min
30 min
30 min
30 min
30 min
30 min
15 min
Blocking buffer incubation time
1% BSA in TBS
CD31 Method 2 (see (Pecam1) Note 32)
Method 1 (see Note 31)
1% BSA in TBS
Method 1
CD206 (MRC1)
CD68
1% BSA in TBS
Method 1
MHCII
1% BSA in TBS
Method 2
CD163
1% BSA in TBS
Method 1 (see Note 29)
Blocking buffer
IBA1
Target
Antigen retrieval method
Table 4 Antigen retrieval, blocking and antibody staining methods
1:100
1:5000 (see Note 33)
1:300
1:1000
1:1000
1:1000
1:200
1:500
1:1000
Antibody dilution
1% BSA in TBS
1% BSA in TBS with 0.05% Trion X100
Davinci green
1h
1h
30 min
1h
1h
1% BSA in TBS with 0.05% Trion X100 1% BSA in TBS with 0.05% Trion X100
15 min
1h
1h
2h
Antibody incubation time
1% BSA in TBS
1% BSA in TBS with 0.05% Trion X100
1% BSA in TBS
1% BSA in TBS
Antibody dilution buffer
Anti-rabbit HRP
Anti-rabbit HRP
Anti-mouse HRP
Anti-rabbit HRP
Anti-rabbit HRP
Anti-rabbit HRP
Anti-mouse HRP
Anti-rabbit HRP
Anti-rabbit HRP
Secondary detection reagent
106 Stephen Huang et al.
Genetic and IHC Tools to Visualise Rat Macrophages
107
Fig. 2 Workflow overview of a typical immunohistochemistry experiment 3.1 Preanalytical Phase: Tissue Processing and Section Preparation
1. Immediately transfer tissues/organs to 4% PFA upon dissection (see Note 2). 2. Fix tissues standing in 4% PFA for 24 h at room temperature.
3.1.1 PFA Fixation
3. Exchange the 4% PFA for PBS after the 24 h and store tissues at 4 °C.
3.1.2 Embedding and Sectioning
1. Process and embed tissues in paraffin wax within 48 hours of fixation (see Notes 3–5). 2. Section tissues at 4–12 μm width with a microtome and mount sections onto Superfrost Plus Adhesion slides (see Note 6). 3. Dry slides overnight at 37 °C in hybridization/dry oven.
3.2 Analytical Phase: Antigen Detection
1. Incubate slides on a slide rack in a hybridization/dry oven at 60 °C for 25 min.
3.2.1 Dewax Slides (See Note 7)
2. Immediately immerse slides for 3× 2 min in 100% xylene (see Note 8).
3.2.2 Rehydrate Tissue
1. Immerse slides for 3× 2 min in 100% ethanol (see Note 9). 2. Immerse slides for 2× 2 min 95% ethanol. 3. Immerse slides for 1× 2 min in 70% ethanol.
108
Stephen Huang et al.
4. Wash slides with running RO water for 30 s. 5. Immerse slides in TBS. 3.2.3 Block Endogenous Peroxidase Activity (See Note 10)
1. Cover the tissue completely in peroxidase block solution and incubate at room temperature for 15 min (see Note 11). 2. Remove peroxidase block solution (see Note 12). 3. Wash slides: Rinse with wash buffer (see Note 12), immerse for 3× 30 s in RO water and then immerse in TBS (see Note 13).
3.2.4
Antigen Retrieval
Perform the corresponding antigen retrieval method for an antibody specified in Table 4. Method 1: 1. Immerse slides in citrate buffer (see Note 14). 2. Transfer the container to decloaking chamber and incubate at 95 °C for 20 min. 3. Keep the slides immersed and transfer the container to ice. 4. Allow slides to cool for 10 min. 5. Immerse slides in TBS. Method 2 (see Note 15) 1. Heat Tris/EDTA buffer in the microwave for 3 min with full power. The buffer will boil in the microwave and settle to 95 °C when removed from the microwave. 2. Transfer slides to the heated Tris/ EDTA buffer. 3. Microwave for 10 min at medium power (see Note 16). 4. Microwave for 2 min 30 s at full power (95–100 °C). 5. Microwave for 10 min at low power (see Note 17). 6. Keep the slides immersed in the Tris/EDTA buffer and transfer the container to ice. 7. Allow the slides to cool in the Tris/ EDTA buffer for 10 min. 8. Immerse slides in TBS. Method 3 (see Note 15) 1. Immerse slides in Tris/EDTA buffer. 2. Microwave for 3 min 30 s at medium power (see Note 18). 3. Keep the slides immersed in the Tris/EDTA buffer and transfer the container to ice. 4. Allow the slides to cool in the Tris/EDTA buffer for 60 min.
3.2.5 Block Nonspecific Binding
1. Using a PAP pen, carefully draw a perimeter around the tissue (see Note 19).
Genetic and IHC Tools to Visualise Rat Macrophages
109
2. Cover the tissue completely in the blocking solution specified in Table 3 (see Note 11). 3. Incubate the tissues with blocking solution at room temperature in a humidified slide chamber for the time specified in Table 3. 4. Remove the blocking solution (see Note 12). 3.2.6 Primary Antibody Reaction
1. Dilute the primary antibody and incubate as described in Table 3. All incubations should be performed in a humidified slide chamber at room temperature (see Note 20). 2. Wash slides as described in Subheading 3.2.3 (see Note 21).
3.2.7 Detection and Colour Reaction
1. Incubate tissue with the HRP detection reagent specified in Table 3 for 30 min at room temperature in humified chamber (see Notes 20 and 22). 2. Wash as described in Subheading 3.2.3. 3. Add DAB peroxidase substrate solution and incubate for 2–8 min (see Note 23). 4. Wash as described in Subheading 3.2.3 (see Note 24). 5. Transfer slides to slide rack and immerse in hematoxylin solution for 45 s (see Note 25). 6. Wash slides thoroughly by gently flushing the container with running RO water for 10 min.
3.2.8 Dehydrate and Clear the Tissue (See Note 26)
1. Immerse slides in 70% ethanol for 3 min. 2. Immerse slides for 2× 2 min 95% ethanol. 3. Immerse slides for 3× 3 min in 100% ethanol. 4. Immerse slides for 2× 5 min in xylene (see Note 27). 5. Mount slides using DPX mounting medium (see Note 28). 6. Allow slides to dry in the fume hood for approximately 12 h.
3.3 Postanalytical Phase: Visualization
Slides can be viewed on a brightfield microscope and images captured with an integrated camera system. We routinely use the VS120 (discontinued) and VS200 Olympus slide scanner for high throughput imaging of stained slides. Images captured can be viewed using Olyvia (Olympus software) or Qupath software, and analysis can be performed using either Qupath or FIJI [29, 30] (see Note 34). Figure 3 shows some examples of tissue-resident macrophages stained for IBA1 in Wildtype and Csf1r KO tissues.
110
Stephen Huang et al.
Fig. 3 IBA1 staining in tissues from wildtype and Csf1rko rats. IBA1 positive staining is indicated by the brown pigment and hematoxylin (blue) is used as the counterstain. There is a marked reduction of IBA1 staining in the liver and almost a complete loss in the kidney of Csf1rko rats when compared to wildtype. Specific loss of IBA1 in the interstitial tissues in the lungs and in the marginal zones of the spleen is observed in the Csf1rko rats. Images were captured using the VS120 or VS200 Olympus slide scanner
4 Notes 1. We routinely prepare 4% PFA and store at -20 °C. 2. Handle tissues with care, tissues can easily become damaged during removal and processing. Handling can be reduced by placing tissues directly into tissue processing/embedding cassettes for fixation. Processing time and costs can be reduced by placing multiple tissues of a similar size and consistency into a single cassette. Ensure cassettes can be closed without pressing the tissue. Bones are fixed for 6 days in 4%PFA and then decalcified in EDTA solution (10%w/v EDTA in RO water pH 7.4) prior to fixation and embedding. We use the KOS rapid multifunctional microwave tissue processor to speed up the process of decalcification.
Genetic and IHC Tools to Visualise Rat Macrophages
111
3. If tissues are not in cassettes, transfer to cassette prior to processing. Ensure the tissue is not pressed by the cassette. 4. Tissue processing: tissues are dehydrated in increasing concentrations of ethanol, cleared in xylene and then infiltrated with paraffin wax, is usually performed using an automated tissue processor but can be done manually if required. 5. Embedding is performed using an embedding station. Ensure samples are embedded in an orientation that allows the plane of interest to be sectioned. If multiple tissues are embedded in the same block, ensure they are of similar size and consistency and that there is sufficient space for wax between the tissues to provide support. 6. Cut sections to a thickness that enables visualization of the desired morphology. We routinely cut 10 μM sections to visualize macrophages by IHC. 7. We use a Tissue-Tek Manual Slide Staining Set for immersion of slides during pre- and post stain processing but any containers in which the slides can be fully immersed is sufficient. Perform xylene immersion steps in a fume hood. 8. For each immersion, place slides into a container of xylene for 2 min, remove, very gently shake off the xylene, and transfer to fresh container of xylene. The xylene can be reused for multiple batches of slides; keep the xylene for each of the immersion steps separate and use in the same order each time. 9. For each immersion, place slides into a container of the appropriate concentration of ethanol for 2 min, remove, very gently shake off the ethanol, and transfer to fresh container of ethanol. The ethanol can be reused for multiple batches of slides; keep the ethanol for each of the immersion steps separate and use in the same order each time. 10. Peroxidase blocking reagent is available commercially or can be made as 0.3% hydrogen peroxide in TBS. Except for CD206 the peroxidase blocking step may be performed either before or after antigen retrieval. For CD206, the peroxidase block must be done after incubation with the primary antibody. 11. You can use a transfer pipette to cover the tissue. Importantly, ensure the tissue is completely covered and does not dry out during the incubation. 12. Solution can be quickly removed by holding the slide on the labeled side and ‘flicking’ the solution into a waste container. To wash slides thoroughly, hold slide over a waste container and using a wash squeeze bottle gently rinse with wash buffer: 0.05% tween in TBS.
112
Stephen Huang et al.
13. For IBA1 antibody, replace the 0.05% Tween in TBS with RO water and immerse in RO water rather than TBS. For CD68 wash 3× 3 min in TBS with 0.5% Tween. 14. Make the 10 mM sodium citrate fresh each day. 15. We typically use a microwave for the heating steps in these methods. Time and power will need to be determined empirically for your specific microwave. Our microwave has a constant output of 1100 W with a frequency of 2450 MHz. We typically heat 200 mL of solution in solvent resistant staining trays. Our microwave oscillates between on/off to mediate power output for the set duration of time. We have provided temperatures as a guide for heat retrieval. 16. Medium power at 40% cycles for 8 s on and 12 s off. The buffer temperature after this heating step is approximately 95 °C. 17. Low power at 20% cycles for 5 s on and 15 s off. The buffer temperature after heating step is approximately 85 °C. 18. Medium power at 50% cycles for 10 s on and 10 s off. The buffer does not boil, but heats to approximately 95 °C starting from a room temperature of 24 °C. 19. A PAP pen creates a hydrophobic barrier around the tissue allowing a smaller volume of antibody solution to be used per slide. 20. 50–80 μL is usually sufficient to cover the tissue. Ensure the tissue is completely covered and does not dry out during the incubation. 21. Thoroughly washing the slides is essential for good quality images. 22. We use DAKO Envision HRP detection reagents. 23. We use the DAKO Liquid DAB+ Substrate Chromogen System. Include a positive control to help determine the length of time for the intensity of colour development. For some tissue/ target combinations, colour development may only be visible by microscopy. 24. An optional DAB enhancing step can be performed after this wash if desired. Immerse slides in 0.5% copper sulphate for 10 s and then wash as described in Subheading 3.2.3. 25. We use Gill No. 2 hematoxylin solution. 26. For each immersion, place slides into a container of ethanol or xylene as appropriate, after incubation remove, very gently shake off the ethanol/xylene, and transfer to fresh container. The ethanol and xylene can be reused for multiple batches of slides; keep each of the immersion steps separate and use in the same order each time. Do not use the same ethanol or xylene used in the dewaxing process (see Subheadings 3.2.1 and 3.2.2).
Genetic and IHC Tools to Visualise Rat Macrophages
113
27. Do not leave the slides in xylene for longer than 15 min. 28. If air bubbles are introduced during the mounting process the coverslip can be removed by pacing the slide back into xylene. 29. Antigen retrieval method 3 can also be used as a gentler alternative to method 1. We have also had success using proteinase K antigen retrieval. 30. We have not been able to successfully stain rat embryos or placenta with the CD11b antibody. 31. Use Diva solution in place of sodium citrate. Allow to cool for 20 min. 32. For bones and neonatal rats, use antigen retrieval method 3. 33. For bones and neonatal rats, dilute primary antibody1:1000. 34. Digital image analysis is an ongoing area of development and research. You may find some chapters in this book useful for image analysis.
Acknowledgments IHC methods were adapted from methods provided by Ngari Teakle, who also provided valuable insight for the notes and antibodies used. This work was supported by funding from the National Health and Medical Research council to D.A.H and K. M.I. References 1. Mass E, Ballesteros I, Farlik M, Halbritter F, Gunther P, Crozet L, Jacome-Galarza CE, Handler K, Klughammer J, Kobayashi Y, Gomez-Perdiguero E, Schultze JL, Beyer M, Bock C, Geissmann F (2016) Specification of tissue-resident macrophages during organogenesis. Science 353(6304). https://doi.org/ 10.1126/science.aaf4238 2. Hume DA, Irvine KM, Pridans C (2019) The mononuclear phagocyte system: the relationship between monocytes and macrophages. Trends Immunol 40(2):98–112. https://doi. org/10.1016/j.it.2018.11.007 3. Shimoyama M, Laulederkind SJ, De Pons J, Nigam R, Smith JR, Tutaj M, Petri V, Hayman GT, Wang SJ, Ghiasvand O, Thota J, Dwinell MR (2016) Exploring human disease using the Rat Genome Database. Dis Model Mech 9(10):1089–1095. https://doi.org/10.1242/ dmm.026021 4. Iannaccone PM, Jacob HJ (2009) Rats! Dis Model Mech 2(5–6):206–210. https://doi. org/10.1242/dmm.002733
5. Smalley E (2016) CRISPR mouse model boom, rat model renaissance. Nat Biotechnol 3 4 : 8 9 3 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / nbt0916-893 6. Hume DA, Caruso M, Keshvari S, Patkar OL, Sehgal A, Bush SJ, Summers KM, Pridans C, Irvine KM (2021) The mononuclear phagocyte system of the rat. J Immunol 206(10):2251–2263. https://doi.org/10. 4049/jimmunol.2100136 7. Cotton WR, Gaines JF (1974) Unerupted dentition secondary to congenital osteopetrosis in the Osborne-Mendel rat. Proc Soc Exp Biol Med 146(2):554–561. https://doi.org/10. 3181/00379727-146-38146 8. Dobbins DE, Sood R, Hashiramoto A, Hansen CT, Wilder RL, Remmers EF (2002) Mutation of macrophage colony stimulating factor (Csf1) causes osteopetrosis in the tl rat. Biochem Biophys Res Commun 294(5):1114–1120. https://doi.org/10. 1016/S0006-291X(02)00598-3
114
Stephen Huang et al.
9. Van Wesenbeeck L, Odgren PR, MacKay CA, D’Angelo M, Safadi FF, Popoff SN, Van Hul W, Marks SC Jr (2002) The osteopetrotic mutation toothless (tl) is a loss-of-function frameshift mutation in the rat Csf1 gene: evidence of a crucial role for CSF-1 in osteoclastogenesis and endochondral ossification. Proc Natl Acad Sci U S A 99(22):14303–14308. https://doi.org/10.1073/pnas.202332999 10. Greep RO (1941) An hereditary absence of the incisor teeth. J Hered 32:397–398. https:// doi.org/10.1093/oxfordjournals.jhered. a104973 11. Van Wesenbeeck L, Odgren PR, Coxon FP, Frattini A, Moens P, Perdu B, MacKay CA, Van Hul E, Timmermans JP, Vanhoenacker F, Jacobs R, Peruzzi B, Teti A, Helfrich MH, Rogers MJ, Villa A, Van Hul W (2007) Involvement of PLEKHM1 in osteoclastic vesicular transport and osteopetrosis in incisors absent rats and humans. J Clin Invest 117(4):919–930. https://doi.org/10.1172/ JCI30328 12. Moutier R, Toyama K, Charrier MF (1974) Genetic study of osteopetrosis in the Norway rat. J Hered 65(6):373–375. https://doi.org/ 10.1093/oxfordjournals.jhered.a108554 13. Weilbaecher KN, Hershey CL, Takemoto CM, Horstmann MA, Hemesath TJ, Tashjian AH, Fisher DE (1998) Age-resolving osteopetrosis: a rat model implicating microphthalmia and the related transcription factor TFE3. J Exp Med 187(5):775–785. https://doi.org/10. 1084/jem.187.5.775 14. Moutier R, Ostrowski K, Lamendin H (1989) Microphthalmia - a new recessive mutation in the Norway rat. J Hered 80(1):76–78. https:// doi.org/10.1093/oxfordjournals.jhered. a110798 15. Perdu B, Odgren PR, Van Wesenbeeck L, Jennes K, Mackay CC, Van Hul W (2009) Refined genomic localization of the genetic lesion in the osteopetrosis (op) rat and exclusion of three positional and functional candidate genes, Clcn7, Atp6v0c, and Slc9a3r2. Calcif Tissue Int 84(5):355–360. https://doi. org/10.1007/s00223-009-9229-7 16. Summers KM, Bush SJ, Hume DA (2020) Network analysis of transcriptomic diversity amongst resident tissue macrophages and dendritic cells in the mouse mononuclear phagocyte system. PLoS Biol 18(10):e3000859. https://doi.org/10.1371/journal.pbio. 3000859 17. Ramos-Vara JA, Miller MA (2014) When tissue antigens and antibodies get along: revisiting the technical aspects of immunohistochemistry--the red, brown, and blue technique. Vet
Pathol 51(1):42–87. https://doi.org/10. 1177/0300985813505879 18. Taylor CR (1994) An exaltation of experts: concerted efforts in the standardization of immunohistochemistry. Hum Pathol 25(1):2–11. https://doi.org/10.1016/00468177(94)90164-3 19. O’Hurley G, Sjostedt E, Rahman A, Li B, Kampf C, Ponten F, Gallagher WM, Lindskog C (2014) Garbage in, garbage out: a critical evaluation of strategies used for validation of immunohistochemical biomarkers. Mol Oncol 8(4):783–798. https://doi.org/10.1016/j. molonc.2014.03.008 20. van Goor H, Harms G, Gerrits PO, Kroese FG, Poppema S, Grond J (1988) Immunohistochemical antigen demonstration in plasticembedded lymphoid tissue. J Histochem Cytochem 36(1):115–120. https://doi.org/10. 1177/36.1.3275710 21. van den Berg TK, Dopp EA, Dijkstra CD (2001) Rat macrophages: membrane glycoproteins in differentiation and function. Immunol Rev 184:45–57. https://doi.org/10.1034/j. 1600-065x.2001.1840105.x 22. Pridans C, Irvine KM, Davis GM, Lefevre L, Bush SJ, Hume DA (2020) Transcriptomic analysis of rat macrophages. Front Immunol 11:594594. https://doi.org/10.3389/ fimmu.2020.594594 23. Waddell LA, Lefevre L, Bush SJ, Raper A, Young R, Lisowski ZM, McCulloch MEB, Muriuki C, Sauter KA, Clark EL, Irvine KM, Pridans C, Hope JC, Hume DA (2018) ADGRE1 (EMR1, F4/80) is a rapidlyevolving gene expressed in mammalian monocyte-macrophages. Front Immunol 9: 2246. https://doi.org/10.3389/fimmu. 2018.02246 24. Keshvari S, Caruso M, Teakle N, Batoon L, Sehgal A, Patkar OL, Ferrari-Cestari M, Snell CE, Chen C, Stevenson A, Davis FM, Bush SJ, Pridans C, Summers KM, Pettit AR, Irvine KM, Hume DA (2021) CSF1R-dependent macrophages control postnatal somatic growth and organ maturation. PLoS Genet 17(6): e1009605. https://doi.org/10.1371/journal. pgen.1009605 25. Pridans C, Raper A, Davis GM, Alves J, Sauter KA, Lefevre L, Regan T, Meek S, Sutherland L, Thomson AJ, Clohisey S, Bush SJ, Rojo R, Lisowski ZM, Wallace R, Grabert K, Upton KR, Tsai YT, Brown D, Smith LB, Summers KM, Mabbott NA, Piccardo P, Cheeseman MT, Burdon T, Hume DA (2018) Pleiotropic impacts of macrophage and microglial deficiency on development in rats with targeted mutation of the Csf1r locus. J Immunol
Genetic and IHC Tools to Visualise Rat Macrophages 201(9):2683–2699. https://doi.org/10. 4049/jimmunol.1701783 26. Irvine KM, Caruso M, Cestari MF, Davis GM, Keshvari S, Sehgal A, Pridans C, Hume DA Analysis of the impact of CSF-1 administration in adult rats using a novel Csf1r-mApple reporter gene. J Leukocyte Biol. https://doi. org/10.1002/jlb.ma0519-149r 27. Utans U, Arceci RJ, Yamashita Y, Russell ME (1995) Cloning and characterization of allograft inflammatory factor-1: a novel macrophage factor identified in rat cardiac allografts with chronic rejection. J Clin Invest 95(6):2954–2962. https://doi.org/10.1172/ JCI118003 28. VanRyzin JW, Arambula SE, Ashton SE, Blanchard AC, Burzinski MD, Davis KT, Edwards S, Graham EL, Holley A, Kight KE, Marquardt AE, Perez-Pouchoulen M, Pickett LA, Reinl EL, McCarthy MM (2021) Generation of an Iba1-EGFP transgenic rat for the study of microglia in an outbred rodent strain. eNeuro 8(5):ENEURO.0026. https://doi. org/10.1523/ENEURO.0026-21.2021 29. Bankhead P, Loughrey MB, Fernandez JA, Dombrowski Y, McArt DG, Dunne PD, McQuaid S, Gray RT, Murray LJ, Coleman HG, James JA, Salto-Tellez M, Hamilton PW (2017) QuPath: open source software for digital pathology image analysis. Sci Rep 7(1):16878. https://doi.org/10.1038/ s41598-017-17204-5 30. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A (2012) Fiji: an opensource platform for biological-image analysis.
115
Nat Methods 9(7):676–682. https://doi.org/ 10.1038/nmeth.2019 31. Irvine KM, Caruso M, Cestari MF, Davis GM, Keshvari S, Sehgal A, Pridans C, Hume DA (2020) Analysis of the impact of CSF-1 administration in adult rats using a novel Csf1rmApple reporter gene. J Leukoc Biol 107(2):221–235. https://doi.org/10.1002/ JLB.MA0519-149R 32. Freuchet A, Salama A, Bezie S, Tesson L, Remy S, Humeau R, Regue H, Serazin C, Flippe L, Peterson P, Vimond N, Usal C, Menoret S, Heslan JM, Duteille F, Blanchard F, Giral M, Colonna M, Anegon I, Guillonneau C (2022) IL-34 deficiency impairs FOXP3(+) Treg function in a model of autoimmune colitis and decreases immune tolerance homeostasis. Clin Transl Med 12(8): e988. https://doi.org/10.1002/ctm2.988 33. Turner-Stokes T, Garcia Diaz A, Pinheiro D, Prendecki M, McAdoo SP, Roufosse C, Cook HT, Pusey CD, Woollard KJ (2020) Live imaging of monocyte subsets in immune complexmediated glomerulonephritis reveals distinct phenotypes and effector functions. J Am Soc Nephrol 31(11):2523–2542. https://doi.org/ 10.1681/ASN.2019121326 34. Harvey BK (2022) LE-Tg(OTTC1005) CX3CR1-Cre-ERT2. https://irp.drugabuse. gov/organization/osd/rats/tgr-le-tgottc100 5-cx3cr1-cre-ert2/. Accessed 8 Dec 2022 35. Bryda EC, Men H, Davis DJ, Bock AS, Shaw ML, Chesney KL, Hankins MA (2019) A novel conditional ZsGreen-expressing transgenic reporter rat strain for validating Cre recombinase expression. Sci Rep 9(1):13330. https:// doi.org/10.1038/s41598-019-49783-w
Chapter 7 Phenotyping of Macrophages in Human Immune System Mice Leonie Voss, Carmen Reitinger, and Anja Lux Abstract Human immune system mice, also referred to as humanized mice, are a major research tool for the in vivo study of human immune system function. Upon reconstitution with human hematopoietic stem cells, all major human leukocyte populations develop in immunodeficient mice and can be detected in peripheral blood as well as in lymphatic and nonlymphatic tissue. This includes human macrophages that are intrinsically difficult to study from humans due to their organ-resident nature. In the following chapter, we provide a detailed protocol for generation of human immune system mice. We suggest that these mice are a suitable model to study human macrophage function in vivo. Key words Human hematopoietic stem cell purification, Density gradient centrifugation, Magnetic cell sorting, Immunodeficient mice, HSC transfer, Flow cytometry analysis, Immunofluorescence microscopy
1
Introduction In vivo models are quintessential to obtain a better understanding of complex biological processes, with mouse models as irreplaceable tools for basic and preclinical research. Despite the striking similarities in immune system function between mice and men, present differences limit transfer of findings across species [1, 2]. Hence, in the past, analysis of human immune system function has largely relied on in vitro and ex vivo studies using human cells and tissues. Even at their best, in vitro systems cannot reflect the complexity of numerous organs, cell types and soluble factors cooperating to shape an immune system. This results in a growing need for animal models that can parallel the human immune system. Humanized mice have been developed to overcome these constraints and are now increasingly employed for the in vivo study of human immune cells and tissues [1–3].
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_7, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
117
118
Leonie Voss et al.
Human immune system, or short humanized, mice are generated by transferring human hematopoietic stem cells (HSC) in immunodeficient mice [1–5]. Mouse strains on a nonobese diabetic (NOD) background, like NOD/SCID/γc-/- (NOD.Cg-Prkdcscid IL2rgtm1WjI/SzJ; short NSG), are particularly suitable recipients for stem cell grafts due to their severely compromised immune system which prevents initial rejection and facilitates stable, longterm engraftment of human cells. Immunodeficiency in NSG mice is a consequence of several genetic alterations. Briefly, the SCID mutation of this mouse strain causes a lack of murine B and T cells while a homozygous deletion of the Il2rγ chain impairs the development of NK cells. Additionally, NOD mice show defects in phagocytosis and complement system activity [6–8]. In combination, NSG and related mouse strains show a particularly high humanization efficiency upon transplantation with human HSCs [9]. HSC transfer results in development of a functional human immune system including essential immune cell populations like B cells, T cells, NK cells, monocytes, macrophages, neutrophils and dendritic cells within 8–12 weeks. These human immune cells are detectable in the blood as well as in lymphatic (like spleen and lymph nodes) and nonlymphatic organs (like the liver) [4, 10, 11]. Therefore, this humanized mouse model is suitable for investigating a number of scientific questions regarding the human immune system. This includes the ontogeny of human myeloid cells such as monocytes and macrophages, their homing to nonlymphoid tissues and contribution to immune responses and tissue homeostasis [11–13]. By expression of Fcγ receptors (FcγR), cellular surface receptors for immunoglobulin G (IgG), monocytes and macrophages are also major effector cells of the humoral immune system in terms of killing and clearance of pathogens and tumor cells [14]. Aiming to investigate antibody activity in humanized mice, one however needs to consider that classical NSG mice still express murine FcγR on residual myeloid cells that are able to interact with human IgG [1, 10, 15]. We have therefore previously generated immunodeficient mice comparable to NSG mice with an additional deletion of fcerg1. Consequently, these NSG/FcRγ-/mice lack expression of the FcRγ accessory signaling chain indispensable for expression and function of activating murine FcγRs [16]. Hence, humanized NSG/FcRγ-/- mice are not only suitable for functional analysis of the human immune system but also of human IgG effector functions exclusively mediated by human FcγR expressing immune cells. Regarding protocols for humanization, a wide range of strategies have been established in the past [3, 5]. The protocol detailed below uses HSCs isolated from human umbilical cord blood and transfer into newborn, irradiated mice by intravenous injection [4, 10, 17, 18].
Macrophages in Humanized Mice
2
119
Materials
2.1
Mice
2.2
HSC Purification
Immunodeficient NSG/FcRγ-/- mice are kept under specific pathogen-free (SPF) conditions and in isolated ventilated cages (IVC) according to institutional guidelines and the rules and regulations of animal welfare laws (see Notes 1 and 2). Considering their severely immunocompromised status, mice need to be supplied with acidified drinking water (pH 3.0) to minimize the risk of bacterial infections (see Note 3). Upon irradiation, litters receive antibiotics until weaning at 3–4 weeks. 1. 0.5 M EDTA, pH 8. 2. 50 mL LeucoSep falcons: # 227290, Greiner Bio-One. 3. 15 mL and 50 ml falcons. 4. 96 well V-bottom plate. 5. 1.5 mL microcentrifuge tubes. 6. Neubauer counting chamber. 7. Human separating solution (centrifugation media), density 1.077 g/mL (see Note 4). 8. 1x PBS. 9. Wash buffer: 2 mM EDTA in 1x PBS. 10. MACS buffer: 0.5% FCS, 2 mM EDTA in 1x PBS, sterile filtered, store at 4 °C. 11. MACS LS-column: #130-042-401, Miltenyi Biotec. 12. Cell strainer 40 μm. 13. Cell strainer 100 μm for MACS LS column. 14. QuadroMACS Separator: # 130-090-976, Miltenyi Biotec. 15. MACS® MultiStand: # 130-042-303, Miltenyi Biotec. 16. Human FcR Blocking Reagent. 17. CD34 MicroBeads, human: #130-046-702, Miltenyi Biotec. 18. Trypan blue staining solution. 19. Freezing medium: 10% DMSO in FCS, sterile. 20. FACS buffer: 2% FCS, 0.05% NaN3 in 1x PBS. 21. DAPI solution: 1 μg/mL 4′,6-Diamidino-2-Phenylindole in FACS buffer. 22. Antibody mix detecting human HSCs: CD3 PerCP (clone UCHT1), CD19 PE-Cy7 (clone SJ25C1), CD33-BV510 (clone WM53), CD34 PE (clone 581), CD38 APC (clone HIT2), CD45 APC/Fire750 (clone 2D1), CD56-FITC (clone MEM188) in FACS buffer. Titrate all antibodies before usage. 23. Flow cytometer or spectral analyzer.
120
Leonie Voss et al.
2.3 HSC Transplantation
1. Irradiation device with Cs137 source. 2. Hamilton 100 μL TLC syringe. 3. Hamilton Gauche hollow needle: 32 gauche, 13 mm, needle point style 4, 15°. 4. Sterile 1x PBS. 5. Trypan blue staining solution. 6. Antibiotics-supplemented drinking water (e.g. 0.1 mg/mL Enrofloxacin).
2.4 Verification of Humanization by Flow Cytometry Analysis
1. ddH2O. 2. 10x PBS. 3. 96 well V-bottom plate. 4. Hematocrit capillaries: 75 mm, 75 μL. 5. EDTA tubes (1.5 mL). 6. Antibody mix for analysis of human PBMCs: CD3 V420 (clone SK7), CD14 V450 (clone MSE2), CD16 R668 (clone 3G8), CD19 BYG710 (clone HIB19), CD33 BV510 (clone WM53), human CD45 V547 (clone HI30), murine CD45.1 BV650 (clone 30-F11), CD56 R720 (clone 5.1H11), and CD66b PE-Cy7 (clone G10FS) in FACS buffer. Titrate all antibodies before usage. 7. Fc Block: Unlabeled CD16/32 (clone 2.4G2) at a concentration of 10 μg/mL in FACS buffer (see Note 5). 8. DAPI solution: 1:5000 in FACS buffer.
2.5 Phenotyping of Human Macrophages by Immunofluorescence Microscopy
1. 27G needle: 0.4 × 19 mm. 2. Cryomolds. 3. Cryoembedding compound. 4. Cryostat microtome. 5. Superfrost Plus Adhesion Microscope Slides. 6. Hydrophobic barrier pen. 7. Pasteur pipettes. 8. Goat serum. 9. Staining chamber. 10. Cover slips, 24 × 40 mm. 11. Aqueous Mounting Medium. 12. Antibodies to detect human macrophages: CD45 APC (clone 2D1), CD68 PE (clone Y182A), CD204 PE (clone C79C20), and CD206 PE (clone 15-2). Titrate all antibodies before usage. 13. Immunoflourescence microscope.
Macrophages in Humanized Mice
3
121
Methods Generation of humanized mice requires purification of human HSCs from fresh umbilical cord blood (see Note 6) isolated peripheral blood mononuclear cells (PBMC). All purification steps are carried out under sterile conditions. Subsequently, purified HSCs are stored in liquid nitrogen until transplantation in newborn immunodeficient mice (see Fig. 1a).
3.1
PBMC Isolation
Fresh human umbilical cord blood is collected in 50 mL falcons containing 400 μL 0.5 M EDTA, pH 8. The following volumes are for a 10 mL cord blood sample. Scale accordingly if more is available.
Fig. 1 Generation and characterization of humanized NSG/FcRγ-/- mice. (a) Human PBMCs are purified by density centrifugation of human umbilical cord blood and used for CD34 isolation using Miltenyi CD34 MicroBead Kit and Quadro MACS magnetic cell separator. Purified HSCs are checked for purity and finally injected in newborn, irradiated mice. (b) Representative contour plots showing purity and CD38 expression of HSCs (left) and composition of residual leukocytes (right). HSCs are identified as CD34+CD38low. Within the human leukocytes B cells (CD19+), T cells (CD3+), NK cells (CD56+) and myeloid cells (CD33+) are quantified. (c) Proportion of human CD45+ cells in the blood as well as in spleen and bone marrow of humanized mice (left graph) and quantification of human B cells (CD19+), T cells (CD3+), NK cells (CD56+), monocytes (CD33+) and neutrophils (CD66b+) in the blood within the hCD45+ cell population (right graph) by flow cytometry analysis. Representative contour plots are shown in (d)
122
Leonie Voss et al.
1. Prepare 50 mL LeucoSep tube by adding 15 mL centrifugation media. Centrifuge at 1000 g at room temperature (RT) for 1 min. 2. Dilute blood with 20 mL sterile 1x PBS (1:3) and load on the prepared 50 mL LeucoSep tube. Centrifuge without brake at 1000 g at RT for 10 min. 3. Carefully remove the PBMC cell layer and transfer in a new 50 mL falcon (if applicable: combine PBMCs from different tubes of the same donor). Wash cells by filling up the falcon to 50 mL with wash buffer. Centrifuge at 300 g for 5 min at RT. Discard the supernatant and add 50 mL of PBS. Repeat the centrifugation step. 4. Resuspend PBMCs in 20 mL cold MACS buffer and filter through a 40 μm cell strainer into a new 50 mL falcon. Determine number of viable cells e.g. using a Neubauer counting chamber excluding dead upon 1:1 staining with trypan blue. 3.2 HSC Isolation Using Miltenyi CD34 Microbead Kit
On average, 200,000 HSCs can be isolated from 5 × 107 PBMCs obtained from 10 mL cord blood. 1. Centrifuge PBMCs at 300 g for 10 min at RT. Discard the supernatant carefully. 2. Resuspend PBMCs in 350 μL cold MACS buffer, 100 μL FcR blocking reagent and 50 μL CD34+ MACS beads per 108 PBMC. Incubate cells for 30 min at 4 °C. Mix occasionally to resuspend cell pellet. 3. Add 10 mL of cold MACS buffer. Centrifuge at 300 g for 10 min at 10 °C and discard the supernatant. 4. During centrifugation mount one MACS LS-Column with a 100 μm cell strainer on a Miltenyi Quadro MACS magnetic cell separator and place an empty 15 mL falcon underneath the column. Add 1.5 mL MACS buffer to the filter and drain by gravity flow. Discard the flow-through. 5. Resuspend cells in 700 μL MACS buffer and transfer cells on the cell strainer sitting on the MACS column. Drain by gravity flow. 6. Wash the cell strainer with 3 mL MACS buffer before removal from MACS column. Discard filter. 7. Wash the column four times with 2 mL MACS buffer each. Drain by gravity flow. This flow-through contains the unlabeled PBMCs. Keep on ice until later analysis of purification efficiency. 8. Remove the column from the magnet and place on a new 15 mL falcon. Elute the HSCs by adding 5 mL MACS buffer to the column and pressing with a stamp (see Note 7).
Macrophages in Humanized Mice
123
9. Determine cell number of the viable, eluted HSCs via Neubauer counting chamber excluding dead cells by staining with 1:1 trypan blue. For confirmation of purity and isolation efficiency, PBMC flow-through and isolated HSCs are checked by flow cytometry. Transfer 25,000 HSCs or 200,000 leukocytes to microcentrifuge tubes and keep on ice until continuing with Subheading 3.3. 10. Spin down purified HSCs at 300 g for 5 min at 10 °C and discard the supernatant. 11. Resuspend HSCs in 1 mL freezing media. Initially freeze at 80 °C but keep in liquid nitrogen storage tank for long-term storage until further use. 3.3 Analysis of HSC Purity by Flow Cytometry
1. Centrifuge purified HSCs and leukocytes (from Subheading 3.2, step 9) at 600 g for 5 min at 10 °C. Discard supernatant carefully. 2. Resuspend cells in 150 μL FACS buffer and transfer to 96 well V-bottom plate. 3. Centrifuge at 600 g for 5 min at 10 °C. Discard supernatant. 4. Resuspend cells in 50 μL HSC antibody mix containing fluorescently labeled antibodies against human CD56, CD34, CD3, CD38, CD19, CD45, and CD33. Incubate for 15 min on ice. 5. Wash by adding 100 μL cold FACS buffer and centrifuge at 600 g for 5 min at 10 °C. Discard supernatant. 6. Resuspend cells in 100 μL FACS buffer with DAPI. Centrifuge at 600 g for 5 min at 10 °C. Discard supernatant. 7. Resuspend cells in 100 μL FACS buffer and analyze by flow cytometry quantifying CD34+CD38low HSCs and residual non-HSC populations (see Fig. 1b) (see Note 8).
3.4 Generation of Humanized Mice
1. Newborn, immunodeficient NSG/FcRγ-/- are removed from the breeding cage and irradiated sublethally with a dose of 1.4 Gy using a Cs137 source. Irradiated pups are returned to the cage until HSC transplantation (see Notes 9–12). 2. Immediately before transplantation, quickly thaw frozen HSCs, transfer to 15 mL falcon tube and add 14 mL of PBS. Spin down at 500 g for 5 min at RT, discard supernatant and resuspend in 100 μL PBS per 200,000 HSCs. Determine number of viable cells with a Neubauer counting chamber excluding dead cells by staining with trypan blue and adjust volume by addition of PBS to obtain 1–1.5 × 106/mL.
124
Leonie Voss et al.
3. Mice are injected with HSCs 4–6 h after irradiation. For the duration of the humanization, remove mother from breeding cage. Mix HSCs by pipetting up and down and gather the correct volume of HSCs with the syringe (see Note 13). 4. Injections are performed with a TLC syringe and 32-Gauche hollow needle under manual fixation of the pup. Stretch head/ neck of pup and slowly inject 20,000–30,000 purified HSCs in a total volume of 20 μL intravenously in the vena facialis (see Notes 14–16). 5. Upon irradiation, litters receive antibiotics until weaning at 3–4 weeks. Therefore, supplement drinking water with 0.1 mg/mL Enrofloxacin (see Note 17). 3.5 Verification of Humanization
1. First control of humanization is performed by flow cytometry of peripheral blood between 8 and 12 weeks after reconstitution. 50 μL of peripheral blood is collected upon puncture of the vena facialis or by retro-orbital bleeding in EDTA tubes under anesthesia. 2. Add 450 μL ddH2O and vortex briefly to induce lysis of human and murine erythrocytes. Incubate for 30 s at RT. (see Note 18). 3. Stop lysis by addition of 50 μL of 10x PBS and vortex briefly. Spin down the samples at 600 g for 5 min at RT. Discard the supernatant carefully by pipetting. 4. Resuspend cells in 1 mL 1x PBS for washing. Centrifuge at 600 g for 5 min at RT and carefully discard the supernatant by pipetting. 5. Resuspend cells in 50 μL Fc Block. Transfer cells to 96 well V-bottom plate and incubate on ice for 15 min. Centrifuge at 500 g for 5 min at 10 °C. Discard the supernatant. 6. Resuspend cells in 50 μL antibody mix including fluorescently labeled antibodies against mouse and human CD45 and lineage-markers for human cell populations (CD19, CD3, CD56, CD33, CD66b, CD16, and CD14). Incubate on ice for 15 min. 7. Wash by addition of 100 μL FACS buffer and spin down at 500 g for 5 min at 10 °C. Discard the supernatant. 8. Resuspend cells in 100 μL FACS-DAPI solution. Centrifuge at 500 g for 5 min at 10 °C. Discard supernatant. 9. Resuspend cells in 100 μL FACS buffer and analyze on a flow cytometer quantifying humanization efficiency. Mice containing more than 5% human CD45+ cells are considered successfully humanized (see Fig. 1c, d).
Macrophages in Humanized Mice
125
Fig. 2 Identification of human macrophages in bone marrow and liver of NSG/FcRγ-/- mice. Immunofluorescence analysis of bone marrow (a) or liver (b) sections of humanized mice stained with anti-human CD68 (a), anti-human CD206 (b, left) or CD204 (b, right) to identify human macrophages (shown in red). In all stainings, anti-human CD45 staining (shown in blue) was used to identify human leukocytes. Human macrophages are visible by co-staining (in purple)
3.6 Phenotyping of Human Macrophages by Immunofluorescence Microscopy
Detailed phenotyping of human hematopoietic cells in lymphatic and nonlymphatic tissues is performed upon sacrificing mice by flow cytometry or immunofluorescence microscopy (IF) analysis. Immunofluorescence microscopy is a powerful technique that is widely used by researchers to observe the localization of molecules and cells in tissue sections. The protocol below details analysis of macrophages in cryosections of humanized mouse bone marrow and liver (see Fig. 2). 1. Sacrifice mice and remove organs of interest. Bone marrow can be flushed from the femur using PBS and a 27G needle upon cutting the ends of the bone. 2. Embed organs in cryo-embedding medium in cryomolds and store at -80 °C until further use. 3. Prepare cryosections with a thickness of 5 μm in a cryostat. 4. Fix cryosections in ice-cold acetone for 150 s and dry at room temperature (see Note 19). 5. Use a liquid blocking ImmedgePen to circle around the sample. Let dry for 30 min. 6. Wash slides by carefully rinsing them with PBS using a Pasteur pipette and transfer to staining chamber. 7. Block samples with 100 μL of 5% goat serum diluted in PBS and incubate for 30 min in a staining chamber. 8. Tip the solution from the slides and add mixture of fluorescently labeled staining antibodies. Human macrophages can be identified as hCD45+ and show expression of CD68, CD206, or CD204 [19–21]. Incubate for 1 h in a staining chamber at RT. Exposure to light should be avoided to prevent bleaching of fluorochromes. Wash slides several times with PBS.
126
Leonie Voss et al.
9. For mounting use 24 × 40 mm rectangular cover slips and aqueous mounting medium (2 drops per slide). Let dry for at least 30 min.
4
Notes 1. Maximum hygiene is key in long-term maintenance of severely immunocompromised mouse strains. Mice should be kept and handled separately from immunocompetent mice to avoid cross-contamination with commensal pathogens present in immunocompetent mice. 2. Cages should not be opened outside of the changing station (flow bench). Before opening a cage, disinfect hands and forearms and wear two pairs of gloves. Disinfect gloved hands after opening the cage and before touching the mice. 3. Drinking water (270 mL) is supplied with 500 μL 12.5% hydrochloric acid to obtain pH ~3. 4. Store bottle at RT when opened. 5. Fc Block is used to block residual murine FcγRIIb during staining. 6. Human cord blood should not be older than 24 h. Extended storage may result in decreased purity of isolated HSCs. 7. Before doing so, remove magnet from hood. 8. Ideally, purified HSC samples should include >80% of CD34+CD38low cells (within the living cell population and after doublet discrimination). Residual CD45+ leukocytes in the HSC sample should contain low numbers of T cells (CD3+) and NK cells (CD56+) to prevent graft versus host reactions (GVHD). Residual B cells and monocytes are identified by expression of CD19 and CD33, respectively. 9. To facilitate detection of birth and assignment of independent litters within the same breeding cage, pregnant females can be isolated in individual cages before giving birth and are monitored daily. 10. Before handling newborn mice, always wear a fresh pair of gloves, do not use ethanol to disinfect gloves as the smell will irritate the mothers while accepting back the pups. 11. Pups should be irradiated within 72 h after birth. The optimal time point is 24–48 h as this gives the mothers time to adjust and tolerate temporary removal of the litter for irradiation and transplantation, respectively. Beyond 72 h the skin of the pups is too thick for i.v. injection of HSCs.
Macrophages in Humanized Mice
127
12. Preconditioning of mice, e.g., by irradiation is a prerequisite for efficient engraftment. 13. Do not pipet multiple times with Hamilton pipet as this will cause shearing of cells. 14. The i.v. injection volume must not exceed 30 μL to preserve blood circulation in pups. Transplantation of >50,000 HSCs is counterproductive as this tends to enhance lethal GVHD in humanized mice. Adjust number of cells to be injected accordingly in case the proportion of CD34+/CD38low cells is reduced in the HSC sample. 15. One needle can on average be used for injection of 20–25 pups. Before injection, wash the Hamilton pipette and needle multiple times with sterile PBS. Change or wash the needle between different HSC samples. 16. Successful injection does not cause blister formation under the skin. If i.v. injection is not feasible, intrahepatic injection can be performed alternatively by puncturing the clearly visible liver through the skin. Following injection, remove any blood drops with a paper towel and put the pup back into the cage. 17. Do not add hydrochloric acid to water bottles supplemented with Enrofloxacin. 18. Human erythrocytes are preferentially lysed with ddH2O. Blood has to be at RT before starting the erythrocyte lysis. Do not vortex or incubate too long to prevent lysis of leukocytes. 19. After this step, cryoslides can be stored at -20 °C until further use. For immunofluorescence staining, slides need to reach RT.
Acknowledgments This work was supported by German Research Foundation (DFG) grants SFB1526 and FOR2886 to A.L. References 1. Lux A, Nimmerjahn F (2013) Of mice and men: the need for humanized mouse models to study human IgG activity in vivo. J Clin Immunol 33(1):4–8 2. Shultz LD, Ishikawa F, Greiner DL (2007) Humanized mice in translational biomedical research. Nat Rev Immunol 7(2):118–130 3. Schinnerling K et al (2019) Humanized mouse models of rheumatoid arthritis for studies on immunopathogenesis and preclinical testing of cell-based therapies. Front Immunol 10:203
4. Schwab I, Lux A, Nimmerjahn F (2015) Pathways responsible for human autoantibody and therapeutic intravenous IgG activity in humanized mice. Cell Rep 13(3):610–620 5. Martinov T et al (2021) Building the next generation of humanized hemato-lymphoid system mice. Front Immunol 12:643852 6. Miao M et al (2021) Reevaluation of NOD/SCID mice as NK cell-deficient models. Biomed Res Int 2021:8851986
128
Leonie Voss et al.
7. Shultz LD et al (1995) Multiple defects in innate and adaptive immunologic function in NOD/LtSz-scid mice. J Immunol 154(1): 180–191 8. Zhou Q et al (2014) Humanized NOD-SCID IL2rg–/– mice as a preclinical model for cancer research and its potential use for individualized cancer therapies. Cancer Lett 344(1):13–19 9. Ito M et al (2002) NOD/SCID/gamma(c) (null) mouse: an excellent recipient mouse model for engraftment of human cells. Blood 100(9):3175–3182 10. Lux A et al (2014) A humanized mouse identifies the bone marrow as a niche with low therapeutic IgG activity. Cell Rep 7(1): 236–248 11. Rongvaux A et al (2014) Development and function of human innate immune cells in a humanized mouse model. Nat Biotechnol 32(4):364–372 12. Evren E et al (2021) Distinct developmental pathways from blood monocytes generate human lung macrophage diversity. Immunity 54(2):259–275.e7 13. Evren E et al (2022) CD116+ fetal precursors migrate to the perinatal lung and give rise to human alveolar macrophages. J Exp Med 219(2)
14. Nimmerjahn F et al (2015) FcgR dependent mechanisms of cytotoxic, agonistic, and neutralizing antibody activities. Trends Immunol 36(6):325–336 15. Dekkers G et al (2017) Affinity of human IgG subclasses to mouse Fc gamma receptors. MAbs 9(5):767–773 16. Brandsma AM et al (2016) Clarifying the confusion between cytokine and Fc receptor “common gamma chain”. Immunity 45(2):225–226 17. Danzer H et al (2020) Human Fcγ-receptor IIb modulates pathogen-specific versus selfreactive antibody responses in Lyme arthritis. elife 9 18. Baerenwaldt A et al (2011) Fcγ receptor IIB (FcγRIIB) maintains humoral tolerance in the human immune system in vivo. Proc Natl Acad Sci U S A 108(46):18772–18777 19. Boyette LB et al (2017) Phenotype, function, and differentiation potential of human monocyte subsets. PLoS One 12(4):e0176460 20. Genin M et al (2015) M1 and M2 macrophages derived from THP-1 cells differentially modulate the response of cancer cells to etoposide. BMC Cancer 15:577 21. Larionova I et al (2020) Tumor-associated macrophages in human breast, colorectal, lung, ovarian and prostate cancers. Front Oncol 10:566511
Chapter 8 Fate-Mapping of Yolk Sac-Derived Macrophages Iva Splichalova and Elvira Mass Abstract To better understand the distinct functions of yolk-sac-derived tissue-resident macrophages (TRMs) and bone-marrow-derived macrophages in homeostasis and disease, it is important to trace the ontogeny of these cells. The majority of TRMs originate from erythro-myeloid progenitors (EMPs). EMPs develop into pre-macrophages (pMacs), which can be detected starting at embryonic developmental day (E)9.0, and which give rise to all TRM during early development. pMacs start expressing the gene Cx3cr1, allowing us to genetically target the early yolk-sac wave of pMacs and their progeny. Here, we describe the protocol for the identification of yolk sac-derived TRMs utilizing in utero labelling of the inducible fate mapping Cx3cr1CreERT ; Rosa26LSL-eYFP mouse model. Key words Yolk sac, Embryo, Tissue-resident macrophages, Fate-mapping, EMPs, pMacs, Tamoxifen
1
Introduction The origin of macrophages can be divided into three waves based on the appearance of progenitors with different hematopoietic potential. The first and second waves originate from the yolk sac (YS) and give rise mainly to erythrocytes and macrophages, but also to granulocytes, megakaryocytes, and mast cells [1, 2]. Many fetalderived macrophages originating from the YS persist until adulthood via proliferation and do not rely on replacement from bonemarrow-derived monocytes. The third wave originates from the embryo proper and gives rise to hematopoitic stem cells (HSCs), which later seed the fetal liver and subsequently the bone marrow producing monocyte-derived macrophages, which can contribute to the TRM-pool in some adult tissues [3, 4]. Macrophages play a fundamental role in fetal development. They are involved in organogenesis via phagocytosis of dead cells during footplate remodeling [5], modulation of neuron outgrowth and positioning during the development of the brain [6], tissue vascularization [7], or bone remodeling [8]. Distinguishing YS-derived TRMs from monocyte-derived cells during
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_8, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
129
130
Iva Splichalova and Elvira Mass
embryogenesis, but also in adulthood, can serve us for a better understanding of how these cells are involved in establishment of homeostasis and progression of diseases. An inducible Cre recombination system can be used to genetically target YS-derived TRMs, which is achieved by fusing the Cre protein with estrogen receptor variants [9]. In this case, the Cre is in a cytoplasmic inactive form until activated by the administration of an estrogen-analog hydroxy-tamoxifen (OH-TAM) causing the translocation of the Cre into the nucleus where it can excise loxP sites. Activation of Cre was achieved by tamoxifen in the past, but recent studies demonstrated the indispensability of using OH-TAM, the active metabolite of tamoxifen, in in utero fate-mapping studies due to its faster mode of action, which will result in more accurate timecontrolled labeling. The translocation of Cre to the nucleus is a transient process and is inactive when the OH-TAM is washed out from the system. Here, we show an example of how to fate-map YS-derived macrophages utilizing an inducible mouse model targeting pMacs. To this end, Cx3cr1-CreERT males are mated with Rosa26LSL-eYFP females, which are subjected to a single dose of OH-TAM to obtain embryos in which eYFP expression is induced at E9.0, labeling the first wave of EMP-derived macrophages. Additionally, the preparation of fetal tissues (brain and liver) for flow cytometry analysis is described.
2
Materials
2.1
Mouse Model
Here we use Cx3cr1CreERT (JAX Strain #:020940) and Rosa26LSL(JAX Strain #:006148) mice to generate Cx3cr1CreERT; Rosa26LSL-eYFP embryos where Cre activation is induced at day E9.0 by OH-TAM. Also time points up until E9.5 can be fatemapped with the same protocol to target the first EMP/pMac wave. Please refer to Chapters 2 and 9 in this book for further fate-mapping models of distinct hematopoietic waves.
2.2
Consumables
1. Scissors (standard, blunt-end, and spring scissors angled to side ball tip).
eYFP
2. Forceps (standard and fine). 3. Embryo spoon. 4. 15 mL tubes. 5. 1.5 mL microcentrifuge tubes. 6. 48-well plate. 7. 96-well plate, U-shaped. 8. 10 cm tissue culture dish.
Fate-Mapping of Fetal Macrophages
131
9. 35 mm petri dish. 10. FACS tubes. 11. Cell strainers (70 and 100 μm). 12. 1 mL syringe with 25G needle. 2.3 Buffers and Solutions
1. 4-hydroxytamoxifen stock solution (OH-TAM): 10 mg/mL OH-TAM (Sigma, H7904) in 10% (v/v) molecular grade Ethanol and 90% (v/v) sunflower seed oil (Sigma, 88921). Prepare by decanting the OH-TAM powder into a 15 mL tube. Add 250 μL Ethanol (100%) to the remaining powder in the glass bottle, pipette up and down and transfer the solution into the 15 mL tube. Vortex and sonicate in a sonication bath for 30 min (see Note 1). Add 2.25 mL of sunflower seed oil to obtain a final volume of 2.5 mL OH-TAM solution. Vortex for approximately 5 min at maximum speed and sonicate for 30 min in a sonication bath (see Note 2). Store at 4 °C, protected from light for a maximum of 1 month (see Note 3). 2. Progesterone stock solution: 10 mg/mL Progesterone (Sigma, P3972) in sunflower seed oil (Sigma, 88921). Weight approximately 10 mg of powder into a 15 mL falcon tube. Add sunflower seed oil to obtain a 10 mg/mL stock solution. Vortex until the progesterone is dissolved. Store at 4 °C (see Note 4). 3. 1x Phosphate-buffered saline (PBS). 4. FACS buffer: 0.5% bovine serum albumin (BSA) and 2 mM Ethylenediaminetetraacetic acid (EDTA) in 1x PBS. Filter buffer through a 0.2 μm filter, e.g., using a bottle top vacuum filter. Store at 4 °C (see Note 5). 5. 2x digestion mix: 0.4 mg/mL DNase, 1 mg/mL Collagenase D, 6% fetal calve serum (FCS) in 1x PBS. Calculate the desired volume of the digestion mix (150 μL/sample). Keep on ice until used for tissue digestion. 6. Blocking solution: 1% (v/v) anti-mouse CD16/32 (e.g. clone 93 from Biolegend), 2% (v/v) of rat serum in FACS buffer. Calculate the desired volume of blocking solution (50 μL/ sample). 7. Antibody mix: Calculate antibody mixes based on the number of your samples (see Note 6 + 7). Fill a 1.5 mL tube with the calculated amount of FACS buffer. Add respective antibodies into microcentrifuge tube and mix by pipetting. For the brain: CD45-BUV805, clone 30-F11, 0.25 μg/ mL; CD11b-BUV737, clone M1/70, 1 μg/mL; Cx3cr1-PE-
132
Iva Splichalova and Elvira Mass
CF594, clone SA11F11, 0.5 μg/mL; CD206-BV711; clone C068C2; 1 μg/mL. For the liver: CD45-BUV805, clone 30-F11, 0.25 μg/ mL; CD11b-BUV737, clone M1/70, 1 μg/mL; F4/80, clone BM8, 1 μg/mL; Cx3cr1-PE-CF594, clone SA11F11, 0.5 μg/mL; Tim4-BV786; clone 21H13; 1 μg/mL. 8. Live/dead staining: 5 μg/mL Hoechst33258 in FACS buffer. 2.4
Equipment
1. Ultrasonic unit (see Note 8). 2. Flow cytometer. 3. Centrifuge for FACS tubes and plates. 4. Optional: Epifluorescence microscope (or similar) with GFP detection.
3
Methods
3.1 In Utero Pulse Labeling by OH-TAM
1. On the morning of injection, sonicate the OH-TAM solution in a ultrasonic bath for 10–30 min until it is completely dissolved. Vortex well. 2. Pre-warm progesterone to room temperature. 3. Weigh the pregnant dam and prepare 75 mg/kg OH-TAM and 37.5 mg/kg progesterone (see Note 9) solution from stocksolutions in a 1.5 tube under the fume hood (see Notes 10 and 11). 4. Load the solution into a 1 mL syringe with a 25G needle under the fume hood (see Note 12). 5. In the animal facility, perform intraperitoneal injection by slowly injecting OH-TAM/progesterone solution into the pregnant dam on the day of interest (see Notes 13 and 14). 6. Keep the needle for approx. 10 s in the peritoneum to allow the oil to exit the syringe. 7. After withdrawing the needle, gently massage the abdomen to distribute the solution (see Note 15).
3.2 Preparation of Cell Suspension from Embryonic Tissue for Flow Cytometry Analysis 3.2.1 Embryo and Organ Isolation
The organs of the embryo/fetus are very soft and the protocol of cell suspension preparation can be applied to the embryonic or fetal organs such as the skin, brain, liver, or spleen. Here we show an example of how to prepare cell suspension from E14.5 liver and brain suitable for flow cytometry analysis.
Fate-Mapping of Fetal Macrophages
133
1. Sacrifice the pregnant mouse by cervical dislocation. 2. Perform a V-formed section on the abdomen and take out the uterus containing the embryos into a 10-cm tissue culture dish filled with cold PBS on ice (see Notes 16 and 17). 3. Catch the open side of one uterine horne using forceps and cut the uterus tissue using spring scissors until you reach the uterine junction. Repeat the same procedure for the other side of the uterus (see Note 18). 4. Collect single embryos in their YS into a new 10-cm tissue culture dish filled with cold PBS on ice using an embryo spoon. 5. Use fine foreceps to open the YS, so that embryos remain connected to the placenta via their umbillical vein. 6. Move each embryo into a separate 35 mm tissue culture dish filled with cold PBS on ice and cut off their heads to sacrifice them (see Note 19). 7. Isolate livers and brains using forceps and clean them from surrounding tissue (see Note 20). 8. Transfer isolated tissues into 48 well-plate filled with 150 μL of cold PBS per well. Keep on ice until all tissues have been collected (see Note 21). 3.2.2 Cell Suspension Preparation
1. Prepare sufficient amount of the 2× concentrated digestion mix once you know the number of embryos being analyzed. 2. Fill each well of the 48 well-plate containing tissue with 150 μL of 2× digestion mix. 3. Cut tissues within the well with blunt-end scissors to obtain small pieces (see Note 22). 4. Incubate the tissues in the digestion mix at 37 °C for 30 min. 5. Place the 48 well-plate with samples on ice. 6. Add 500 μL of FACS buffer to each well to stop digestion. 7. Gently pipet the sample with a 1 mL pipet up and down to obtain a single-cell suspension. 8. Filter the samples through a 100 μm filter into the FACS tubes prefilled with 2 mL of FACS buffer. 9. Spin the FACS tubes with samples for 5 min, 400 g, 4 °C. 10. Remove the supernatant.
3.2.3 Staining of Cell Suspension with Antibodies
1. Add 50 μL of blocking solution into each FACS tube and mix by pipetting or careful vortexing. 2. Incubate for 15 min on ice. 3. Transfer 25 μL of the cell suspension into a 96-well plate and add 200 μL FACS buffer (see Note 23).
134
Iva Splichalova and Elvira Mass
Fig. 1 Analysis of Cx3cr1CreERT; Rosa26LSL-eYFP embryos at E14.5, with OH-TAM induction at E9.0. (a) Representative gating strategy for microglia and border-associated macrophages (BAMs) and the eYFP expression in both populations. (b) Representative gating strategy for Kupffer cells (KCs) and their eYFP expression. (c) Labeling efficiency. n = 3
4. Spin down the 96-well plate for 5 min, 400 g, 4 °C. 5. Remove supernatant. 6. Add 20 μL of the antibody mix and resuspend the pellet by pipetting. 7. Incubate on ice for 30 min in the dark (see Note 24). 8. Add 200 μL FACS buffer (see Note 25). 9. Spin down the 96-well plate for 5 min, 400 g, 4 °C. 10. Remove the supernatant. 11. Resuspend cells in 100 μL of FACS buffer. 3.2.4 Flow Cytometry Analysis
1. Filter the sample into the FACS tube through a 70 μm strainer. 2. Add 100 μL Hoechst33258 into the FACS tube containing the sample 1 min before the measurement. 3. Vortex sample and measure using a flow cytometer. An example showing the gating strategy for liver and brain, as well as the labeling efficiency of OH-TAM induction of YFP expression is shown in Fig. 1.
4
Notes 1. Make sure that the water bath does not heat up, otherwise add ice. The OH-TAM suspension must turn into a white homogenous suspension. If not, prolong the time of sonication. 2. The final OH-TAM solution must be a white homogenous suspension. If not, prolong the time of sonication. 3. If you will not use up 2.5 mL OH-TAM within 1 month, you can store the OH-TAM diluted in EtOH and aliquoted at -
Fate-Mapping of Fetal Macrophages
135
20 °C. Make sure that you sonicate the OH-TAM well after thawing, do not do thaw/freeze cycles, and follow labeling efficiency results carefully as the efficiency of OH-TAM may decrease with time. 4. Note, that the solution is clear when stored. 5. Mix BSA well, best is to prepare fresh buffer the day before the experiment. 6. For embryonic tissue the 20 μL of antibody mix is sufficient, you may need to scale this volume up if you are staining in FACS tubes or more cells. 7. The majority of the flow cytometry antibodies contain sodium azide as protection against microbial contamination. Despite this fact, we recommend preparing antibody mixes in sterile conditions. 8. Use floating stands for 15 mL tubes to keep tubes upright. If your water bath is small and heats up quickly, add ice during sonication. The OH-TAM should be warmer than 42 °C as it will lose its efficiency after being exposed to higher temperatures. 9. It is crucial to supplement OH-TAM with progesterone to reduce abortion after tamoxifen administration to the pregnant dam. The OH-TAM treatment often leads to problems during birth, we suggest scarifying the mother and performing a cesarean section on E20, removing the pups, and using foster mothers if the study needs to be performed on the offspring at postnatal ages. 10. Depending on the mouse weight, you will inject about 100–150 μL in total, but a 25G syringe has about 50 μL dead-volume, thus, make sure you prepare enough OH-TAM/progesterone mixture. 11. 75 mg/kg OH-TAM is a high concentration, which will lead to high labelling efficiency, but also a higher abortion rate or increased mortility during labor. To circumvent these problems, you can decrease the OH-TAM concentration by half, especially when performing initial experiments where high labelling efficiency is not required. 12. Despite sonication, OH-TAM stays an emulsion. Thus, to make sure that small pieces are not getting stuck in the syringe during the injection, take up the mixture through the needle. Be very careful when placing the cap back on the needle if you have to transport the syringe to the animal facility. 13. Inject at the middle line of the belly so that you do not accidentlly inject into one of the embryos. Enter the perito-
136
Iva Splichalova and Elvira Mass
neum with the needle at a low angle (approx. 15–20°, instead of the typical 45° angle for interperitoneal injecitons). Make sure by pushing the needly slightly up against the peritoneal wall that you are actually in the peritoneum and not injecting subcutaniously. 14. The day of interest is in this case the day of embryonic development (E). Embryonic development is estimated considering the day of vaginal plug formation as E0.5. Recombination is induced by a single injection of OH-TAM/progesterone solution. Example: The vaginal plug was observed in the morning, and the embryo’s development is considered E0.5. If the induction should take a place at E9.0, the injection will be performed at 8:00 am 9 days after the vaginal plug discovery. If the induction should take place at E9.5, the injection will be performed at 1:00 pm, 9 days after vaginal plug discovery. 15. If you see a big oil spot after syringe withdrawal, make a note for this mouse and its offspring. It is likely that your labelling efficiency will be lower than expected. 16. It is important to work with cold PBS and on ice. If not, embryos will be degraded very fast. 17. If both sides of the uterus have embryos, it is easiest to grab the uterine junction near the cervix by standard foreceps and pull both uterine hornes out of the body. Then cut the uterus on both sides, remove any access of adipose tissue during the procedure. 18. The ball tip helps preventing piercing of the embryos. If you are familiar with dissections you may also use other fine sciccors to carfully cut open the uterus. 19. It is better if the embryo is floating in the dish because dissection of the embryo is easier. 20. Even other organs can be isolated. E12.5 embryos have already visible organs, so it is possible to work without a microscope. In the case, of smaller embryos, we recommend using a microscope or table magnifier. 21. The YFP signal can be detected when using the GFP channel of an Epifluorescence microscope. Like that, all Cre-positive and 1–2 Cre- embryos can be genotyped before running the samples. 22. It is important to use blunt scissors to avoid scratching the bottom of the well. The released plastic particles can kill the cells. 23. You can use the rest of the cell suspension to calculate cell numbers or run an additional antibody panel.
Fate-Mapping of Fetal Macrophages
137
24. You can place the plate into the fridge. If you use aluminum foil and keep the plate on ice, you may need to increase the staining time since the ice/ice-water has a lower temperature than the fridge. 25. If you analyze many tissues and/or many embryos a multichannel pipette should be used.
Acknowledgments We thank Cornelia Cygon for excellent technical support. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy-EXC2151-390873048, GRK2168, SFB1454 and by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 851257). References 1. Mass E, Gentek R (2021) Fetal-derived immune cells at the roots of lifelong pathophysiology. Front Cell Dev Biol 9. https://doi.org/10. 3389/fcell.2021.648313 2. Perdiguero EG, Geissmann F (2016) Development and maintainance of resident macrophages. Nat Immunol 17:2. https://doi.org/ 10.1038/NI.3341 3. Mass E, Nimmerjahn F, Kierdorf K, Schlitzer A (2023) Tissue-specific macrophages: how they develop and choreograph tissue biology. Nat Rev Immunol 2023:1–17. https://doi.org/10. 1038/s41577-023-00848-y 4. Mass E (2018) Delineating the origins, developmental programs and homeostatic functions of tissue-resident macrophages. Int Immunol 30: 493–501. https://doi.org/10.1093/intimm/ dxy044 5. Hopkinson-Woolley J, Hughes D, Gordon S, Martin P (1994) Macrophage recruitment during limb development and wound healing in the embryonic and foetal mouse. J Cell Sci 107: 1159–1167. https://doi.org/10.1242/JCS. 107.5.1159 6. Squarzoni P, Oller G, Hoeffel G, Pont-Lezica L, Rostaing P, Low D, Bessis A, Ginhoux F, Garel S
(2014) Microglia modulate wiring of the embryonic forebrain. Cell Rep 8:1271–1279. https:// doi.org/10.1016/j.celrep.2014.07.042 7. DeFalco T, Bhattacharya I, Williams AV, Sams DM, Capel B (2014) Yolk-sac-derived macrophages regulate fetal testis vascularization and morphogenesis. Proc Natl Acad Sci U S A 111. h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / P N A S . 1 4 00 05 71 1 1/ - / D CSU P P L EME NTAL / PNAS.201400057SI.PDF 8. Jacome-Galarza CE, Percin GI, Muller JT, Mass E, Lazarov T, Eitler J, Rauner M, Yadav VK, Crozet L, Bohm M, Loyher PL, Karsenty G, Waskow C, Geissmann F (2019) Developmental origin, functional maintenance and genetic rescue of osteoclasts. Nature 568:541–545. https://doi.org/10.1038/s41586-019-1105-7 9. Sohal DS, Nghiem M, Crackower MA, Witt SA, Kimball TR, Tymitz KM, Penninger JM, Molkentin JD (2001) Temporally regulated and tissue-specific gene manipulations in the adult and embryonic heart using a tamoxifeninducible Cre protein. Circ Res 89:20–25. https://doi.org/10.1161/HH1301.092687
Chapter 9 Fate-Mapping of Hematopoietic Stem Cell-Derived Macrophages Katharina Mauel and Elvira Mass Abstract Macrophages are cells of the innate immune system, which contribute to the maintenance of tissue homeostasis and form the first line of defense against pathogens. Tissue-resident macrophages that originate from erythro-myeloid-progenitors in the yolk sac colonize the organs early during development and self-maintain in most organs throughout adulthood. Under homeostatic and pathological conditions, circulating monocytes infiltrate the tissue, where they differentiate into macrophages. However, particularly upon inflammation, phenotyping of these distinct macrophage populations using surface markers or antibody stainings is insufficient as their phenotypes converge, at least transiently. A well-established method for the developmental origin of different cell types is the use of in vivo fate-mapping models, where a fluorescent reporter will be expressed under the control of a cell type-specific promoter. Here, we describe the Cxcr4CreERT2; Rosa26LSL-tdTomato mouse fate-mapping model, which labels hematopoietic stem cells and, thus, also monocytes and monocyte-derived macrophages while most tissue-resident macrophages are not targeted. Key words Macrophages, Development, Ontogeny, Fate-mapping, Bone marrow, Hematopoietic stem cells, Monocytes, Spleen, Cxcr4
1
Introduction Macrophages are innate immune cells that monitor pathogens, engulf and digest microbes, and cell debris; thus, they are vital for tissue homeostasis [1]. Most immune cells are derived from hematopoietic stem cells (HSC), as is the case for monocyte-derived macrophages. A decade ago, the prevalent understanding of macrophage biology was revised through the discovery of tissueresident macrophages (TRM) that arise from erythro-myeloid-progenitors (EMP) in the yolk sac, before the onset of definitive hematopoiesis [2]. These HSC-independent macrophages colonize the organs and self-maintain in the adult tissue through local proliferation mainly independent of circulating HSC-derived cells [3, 4].
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_9, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
139
140
Katharina Mauel and Elvira Mass
To better understand the function of TRM in homeostasis and immunity in adult individuals, it is crucial to consider the ontogeny of these unique cells. With the help of fate-mapping experiments and parabiosis studies [5], it was described that some TRMs exclusively originate from yolk sac-derived precursors; that includes especially microglia in the brain and epidermal Langerhans cells in the skin [6]. Other organs are engrafted by a small portion of HSC-derived circulating monocytes in an age-related manner; that includes, e.g., red pulp macrophages in the spleen [7, 8] and peritoneal macrophages [9]. During adulthood, the primary function of TRMs is to surveil the microenvironment and induce inflammation in case of homeostatic disbalances [10], as well as to later induce tissue repair and healing. Under pathological conditions, the niche of TRMs in the respective organ can be severely altered and TRM numbers diminished, giving space for HSC-derived monocytes to infiltrate the empty niche. Due to their different developmental origin, these cells are—at least during the first few days—phenotypically and functionally different and are often characterized by a more inflammatory phenotype [11]. However, the phenotypes of TRMs and monocyte-derived macrophages may converge with time, especially when the latter become longlived as well. Therefore, it is important to address the ontogeny of tissue macrophages in health and disease, which play a crucial role in the maintenance and re-establishment of tissue homeostasis. A standard technique to study the developmental origin of different macrophage populations is the use of fate-mapping models. Upon the modification of two genetic loci, a fluorescent reporter will be expressed by the cell type of interest. First, a genetic locus needs to be identified that is specifically active in either HSCor EMP-derived macrophages. Here we focus on the Cxcr4 gene, which is expressed by long-term HSC [12]. On one allele, the recombinase enzyme Cre is inserted under the control of the Cxcr4 promoter, which induces the specificity of the model toward the HSC-lineage. Into the ubiquitously expressed Rosa26 genetic locus a fluorescent reporter gene is inserted [13]. Common reporters are the green fluorescent protein (GFP) or red fluorescent proteins (tdTomato); the latter is applied in this chapter. The genetic locus is further modified by the insertion of a Stop-codon in front of the fluorescent reporter. This Stop-codon is flanked by a loxP-sequence, which is the genetic sequence that is specifically recognized by the Cre-recombinase enzyme [14]. Taken together, in our example the Cre-recombinase is specifically expressed under to control of the Cxcr4 promoter, which is active in HSC. Therefore, only in cell originating from that lineage, the Stop-codon is excised and the sequence encoding the fluorescent reporter tdTomato is accessible for the polymerase and can be synthesized. Here, we will describe the Cxcr4CreERT2; Rosa26LSL-tdTomato mouse model and the spleen as an exemplary tissue to show how to label HSCs efficiently, providing an inducible fate-mapping model to permanently label all HSC-derived macrophages.
Fate-Mapping of HSC-Derived Macrophages
2
141
Materials
2.1
Mouse Model
2.2
Consumables
Here we use heterozygous Cxcr4CreERT [12] and homozygous Rosa26LSL-tdTomato (Ai14; JAX Strain #: 007914) mice to generate Cxcr4CreERT2; Rosa26LSL-tdTomato mice where Cre activation is induced at 4 weeks of age with tamoxifen. Tamoxifen injections can occur also at a later time point; make sure to include a fourweek wash-out period before the start of your experiment. Please refer to Chapters 2 and 8 in this book for further fate-mapping models of distinct hematopoietic waves. 1. Scissors (standard, blunt-end). 2. Forceps (standard). 3. 15 mL tubes. 4. 50 mL tubes. 5. 1.5 mL and 2 mL microcentrifuge tubes. 6. 96-well plate, U-shaped. 7. 3.5 cm dish (alternatively 6-well plate). 8. FACS tubes. 9. Cell strainer (70 and 100 μm). 10. 1 mL syringe with 26G needle. 11. 10 mL syringe with 26G needle. 12. Metal plunger. 13. Heparin-coated tube. 14. Blood lancet. 15. Aluminum foil.
2.3 Buffers and Solutions
1. Tamoxifen stock solution: 10 mg/mL tamoxifen (Roth, T5648) in 10% (v/v) molecular grade Ethanol and 90% (v/v) corn oil (Sigma, C8267). Prepare by weighing 100 mg of tamoxifen under sterile conditions (see Note 1) in a 15 mL centrifuge tube and add 1 mL of pure Ethanol (100%). Vortex and put the suspension in a 42 °C water bath. Vortex regularly, until the tamoxifen is properly dissolved. Then add 9 mL of corn oil under sterile conditions and mix to obtain a homogeneous solution. Store at 4 °C not more than 4–6 weeks (protected from light). Each mouse will be injected with 100 μL for 5 days (see Note 2). 2. Antibody solution to validate reporter expression in monocytes: Visualization of tdTomato expression in HSC-derived monocytes via CD11b-APC (clone: M1/70, 1 μg/mL) and Ly6C-PE/Cy7 (clone: HK1.4, 1 μg/mL) staining.
142
Katharina Mauel and Elvira Mass
3. Anesthesia: 100 mg/mL ketamin, 20 mg/mL xylazine in sterile 0.9% NaCl (see Note 3) 4. 1x Phosphate-buffered saline (PBS) 5. FACS buffer: 0.5% bovine serum albumin (BSA) and 2 mM Ethylenediaminetetraacetic acid (EDTA) in 1x PBS. Filter buffer through a 0.2 μm filter, e.g. using a bottle top vacuum filter. Store at 4 °C. 6. Digestion mix: 0.2 mg/mL DNase, 0.5 mg/mL Collagenase D, 2.4 mg/mL Dispase and 3% fetal calve serum (FCS, see Note 4) in 1x PBS. Calculate the desired volume of digestion mix (500 μL/sample; see Note 5). Keep on ice until used for tissue digestion. 7. Red-blood-cell (RBC) lysis buffer: 155 mM NH4Cl, 12 mM NaHCO3, 0.1 mM EDTA in dH20. pH 7.1–7.4. 8. Blocking solution: 1% (v/v) anti-mouse CD16/32 (e.g. clone 93 from Biolegend), 2% (v/v) of rat serum in FACS buffer. Calculate the desired volume of blocking solution (50 μL/ sample; see Note 6). 9. Antibody mix: Calculate antibody mixes based on the number of your samples (see Note 7). Fill a 1.5 mL tube with the calculated amount of FACS buffer. Add antibodies (see Table 1 for staining of the spleen) into a microcentrifuge tube and mix by pipetting. 10. Live/dead staining: 5 μg/mL Hoechst33258 in FACS buffer. 2.4
Equipment
1. Water bath. 2. Flow cytometer. 3. Procedure platform with needles (to fix mouse). 4. Centrifuge for FACS tubes and plates. 5. Thermomixer.
Table 1 Antibody panel for spleen tissue First antibody mix
Antigen Ly6G TCRbeta CD19 Nkp46
Conjugate biotin biotin biotin biotin
Concentration [μg/mL] 5 5 5 5
Clone 1A8 H57–597 6D5 29A1.4
Second antibody mix
Streptavidin F4/80 CD45 CD11b CD64
BV785 APC/Cy7 BUV805 BUV661 PerCP/Cy5.5
1 0.5 0.25 1 4
BM8 30-F11 M1/70 X54–5/7.1
Fate-Mapping of HSC-Derived Macrophages
3
143
Methods
3.1 Tamoxifen Administration and Labeling Efficiency Control
1. Let the tamoxifen solution adhere to room temperature. Take up required volume with a 1 mL syringe that has the needle attached (see Note 8). 2. Inject each mouse intraperitoneally (i.p.) with 100 μL of tamoxifen for five consecutive days. To prevent infections, inject into alternating positions every day (left, middle, and right side of the abdomen) (see Note 9). 3. Control successful induction of the reporter expression by collecting a small blood sample from the mouse in a heparincoated tube (for example via tail vein puncture—depending on your animal permit). 4. Add collected blood (10–50 μL) to a microcentrifuge tube, which contains 500 μL of RBC-lysis buffer. 5. Mix and incubate for 5 min on ice. 6. Add 500 μL FACS buffer. 7. Centrifuge at 400 g for 5 min at 4 °C. 8. Discard supernatant and dissolve in 50 μL blocking solution. 9. Incubate 10 min on ice and spin samples at 400 g, 5 min, 4 °C 10. Discard supernatant and resuspend pellet in 25 μL blood staining solution (see Subheading 2.3. Antibody solution to validate reporter expression in monocytes). 11. Incubate 30 min on ice, add 200 μL FACS-buffer and spin samples at 400 g, 5 min, 4 °C. 12. Discard supernatant and resuspend pellet in 100 μL FACSbuffer. 13. Transfer cells through a 70 μm strainer into a FACS tube and add an equal volume of 5 μg/mL Hoechst. 14. Measure tdTomato signal in monocytes (CD11b+Ly6C+) with a flow cytometer. 15. In case of successful expression of the fluorescent reporter, the mice can now be treated (depending on your experimental setup, for example with an infectious agent) or analyzed at the time-point of interest after labeling-induction with tamoxifen.
3.2 Preparation of Cell Suspension from Adult Mouse Spleen for Flow Cytometry Analysis
Depending on the scientific question, different organs might be of interest for flow cytometry analysis. Here, we show an example of how to prepare a cell suspension from adult mouse spleens suitable for flow cytometry analysis. Other organs may require a slight adaptation of the protocol to produce a single-cell suspension.
144
Katharina Mauel and Elvira Mass
3.2.1 Isolation of Mouse Spleen for Flow-Cytometry Analysis
1. On the day of the experiment, inject the mouse i.p. with anesthesia. When the mouse is deeply asleep (see Note 10) fix it on a procedure platform with needles, with its back toward the platform. 2. Open the peritoneal cavity and the diaphragm. Grab the sternum with forceps and open the rib cage using scissors to expose the heart (try not to cut any large blood vessels). 3. From the heart, blood can be collected with a 1 mL syringe and 26 G needle that was flushed with EDTA solution and collected in a heparin tube. It is possible to collect 500 μL or more from an adult mouse. 4. To perfuse the mouse, gently fix the heart with forceps. Then make a small cut on top of the right ventricle (blood will come out). Have a 10 mL syringe prepared with PBS and a 26 G needle and immediately insert the needle into the left ventricle. Gently press the plunger to perfuse the mouse (see Note 11). An indicator of a successful perfusion is the liver, which will become light flesh colored when the blood is flushed out. 5. Collect the spleen by holding the pancreas with forceps and carefully cut off the spleen from the white pancreatic tissue. Place the spleen into a 3.5 cm dish with PBS on ice. 6. Dissect a piece of the tissue (30–40 mg), which is used to prepare a single-cell suspension for flow cytometry.
3.2.2 Cell Suspension Preparation
1. Transfer 30–40 mg of the spleen tissue into a 2 mL microcentrifuge tube and add 500 μL of digestion mix. 2. Cut the spleen within the tube with blunt scissors to obtain small pieces. 3. Incubate the spleen in digestion mix at 37 °C for 30 min while shaking at 1000 rpm. 4. Place the tubes on ice and have 50 mL centrifuge tubes prepared with 100 μm strainers. Wet the strainer with approximately 3 mL FACS buffer, which is collected in the 50 mL centrifuge tubes underneath. 5. Gently pipette the digested spleen suspension up and down and add it on top of the 100 μm strainer. Mesh tissue pieces with a metal plunger. Flush with additional 2–3 mL of FACS buffer and collect all flow-through in the centrifuge tube underneath. 6. Spin the sample tubes for 7 minutes, 400 g, 4 °C. 7. For red blood cell lysis, resuspend the pellet in 1 mL of RBC-lysis buffer and incubate for 5 min on ice. Then add FACS buffer (approximately 7 mL), invert the tube five times and spin (7 min, 400 g, 4 °C). 8. Remove the supernatant
Fate-Mapping of HSC-Derived Macrophages
3.3 Staining of Cell Suspension with Antibodies Conjugated with Fluorochromes
145
1. Add 50 μL of blocking solution into each tube and resuspend the pellet by pipetting. 2. Incubate for 10 min on ice. 3. Transfer the cell suspension into the 96-well plate and top up with 100 μL FACS buffer. 4. Spin down the 96-well plate for 5 min, 400 g, 4 °C. 5. Remove the supernatant by flipping the plate over the sink and tapping it on a tissue. 6. Add 50 μL of the first antibody mix and resuspend each pellet by pipetting. 7. Incubate on ice for 30 min. 8. Add 200 μL FACS buffer. 9. Repeat steps 4–5. 10. Add 50 μL of the second antibody mix and resuspend the pellet by pipetting. 11. Incubate on ice for 30 min in dark (cover the plate with aluminum foil). 12. Add 200 μL FACS buffer. 13. Repeat steps 4–5. 14. Resuspend the cell pellet in 100 μL of FACS buffer.
3.4
Flow Cytometry
1. Filter the samples into FACS tubes through a 70 μm strainer. 2. Add an equal volume (100 μL) of 5 μg/mL Hoechst to the sample. 3. Vortex and measure the sample (see Note 12).
3.5
Data Analysis
1. Export FCS files from your flow cytometer and import them in FlowJo for data analysis. 2. After compensation, discriminate different splenocytes and examine the cell-fate specific labeling with tdTomato using the gating strategy in Fig. 1.
4
Notes 1. Weigh a closed centrifuge tube on the scale, and note the weight. Under the sterile workbench, fill it with tamoxifen powder, close the centrifuge tube, and weigh it again to obtain the exact weight of the tamoxifen powder. 2. If you have larger or smaller cohorts of mice, adjust the amount of tamoxifen solution you need. 3. Always check your animal permit for the concentrations and volumes that are allowed to be injected.
146
Katharina Mauel and Elvira Mass
SSC-A
Lineage- BV785
CD64-PerCP/Cy5.5
(TCRb, CD19, Ly6G, NKp46)
CD45-BUV805
CD11b-BUV661
23.9%
tdTomato
100
% tdTomato
B
SSC-A
FSC-A F4/80-APC/Cy7
FSC-A
CD11b-BUV661
FSC-A
count
SSC-A
FSC-H
Hoechst
A
80
blood monocytes splenic macrophages
60 40 20 0
Fig. 1 Fate-mapping of monocytes and monocyte-derived macrophages using the inducible Cxcr4CreERT2; Rosa26LSL-tdTomato model. (a) Gating strategy for red pulp macrophages (c) Quantification of cell-specific labeling with tdTomato
4. Frozen FCS has to be thawed and heat-inactivated for 30 min at 56 °C. Then aliquots can be stored at -20 °C. 5. To obtain 1 mL of digestion mix, 2.4 mg Dispase is measured with a fine scale, transferred into a 15 mL centrifuge tube and dissolved in 966 μL of PBS. 2 μL of a 100 mg/mL DNase stock solution is added as well as 2 μL of a 0.5 g/mL Collagenase D stock solution. Upon addition of 30 μL FCS, the solution is pre-warmed to 37 °C just before usage. Digestion mix has to be prepared fresh for every experimental day. 6. To obtain a 50 μL blocking solution, 2 μL rat-serum and 0.5 μL Fc-block are dissolved in 47.5 μL FACS-buffer. 7. It is essential to keep in mind, that the indicated dilutions are just an approximation, based on the used lot number. It is important to titrate your antibodies to find the ideal dilution factor for your specific batch. 8. Taking up tamoxifen through the needle ensures that no undissolved particles are taken up and injected. 9. Four weeks after tamoxifen administration, all systemic residues should be eliminated in the mouse. This is important since tamoxifen can have an impact on cellular function.
Fate-Mapping of HSC-Derived Macrophages
147
10. Check for reflexes by pinching the mouse with your fingers or forceps between the toes. 11. Blood will come out of the cut in the upper left ventricle. If the needle is not injected correctly, the lung will inflate and liquid will come out of the mouse nose. 12. In case the machine measures more than 10,000 events/second, dilute the sample with FACS buffer.
Acknowledgements We thank Cornelia Cygon for excellent technical support. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy-EXC2151-390873048, GRK2168, SFB1454 and by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 851257). References 1. Mass E, Nimmerjahn F, Kierdorf K, Schlitzer A (2023) Tissue-specific macrophages: how they develop and choreograph tissue biology. Nat Rev Immunol 2023:1–17. https://doi.org/ 10.1038/s41577-023-00848-y 2. Gomez Perdiguero E, Klapproth K, Schulz C et al (2015) Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature 518:547–551. https:// doi.org/10.1038/nature13989 3. Parwaresch MR, Wacker H-H (1984) Origin and kinetics of resident tissue macrophages: parabiosis studies with radiolabelled leucocytes. Cell Prolif 17:25–39. https://doi.org/ 10.1111/j.1365-2184.1984.tb00565.x 4. Mass E (2018) Delineating the origins, developmental programs and homeostatic functions of tissue-resident macrophages. Int Immunol 30:493–501. https://doi.org/10.1093/ intimm/dxy044 5. Jacome-Galarza CE, Percin GI, Muller JT et al (2019) Developmental origin, functional maintenance and genetic rescue of osteoclasts. Nature 568:541–545. https://doi.org/10. 1038/s41586-019-1105-7 6. Hashimoto D, Chow A, Noizat C et al (2013) Tissue-resident macrophages self-maintain locally throughout adult life with minimal con-
tribution from circulating monocytes. Immunity 38:792–804. https://doi.org/10.1016/j. immuni.2013.04.004 7. Schulz C, Perdiguero EG, Chorro L et al (2012) A lineage of myeloid cells independent of myb and hematopoietic stem cells. Science 335:86–90. https://doi.org/10.1126/SCI ENCE.1219179/SUPPL_FILE/ SCHULZGOMEZ1219179.AVI 8. Sawai CM, Babovic S, Upadhaya S et al (2016) Hematopoietic stem cells are the major source of multilineage hematopoiesis in adult animals. Immunity 45:597. https://doi.org/10.1016/ J.IMMUNI.2016.08.007 9. Bain CC, Hawley CA, Garner H et al (2016) Long-lived self-renewing bone marrowderived macrophages displace embryo-derived cells to inhabit adult serous cavities. Nat Comm u n 7 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / NCOMMS11852 10. Cailhier JF, Partolina M, Vuthoori S et al (2005) Conditional macrophage ablation demonstrates that resident macrophages initiate acute peritoneal inflammation. J Immunol 174:2336–2342. https://doi.org/10.4049/ JIMMUNOL.174.4.2336 11. Guilliams M, Mildner A, Yona S (2018) Developmental and functional heterogeneity of
148
Katharina Mauel and Elvira Mass
monocytes. Immunity 49:595–613. https:// doi.org/10.1016/J.IMMUNI.2018.10.005 12. Werner Y, Mass E, Ashok Kumar P et al (2020) Cxcr4 distinguishes HSC-derived monocytes from microglia and reveals monocyte immune responses to experimental stroke. Nat Neurosci 23:351–362. https://doi.org/10.1038/ s41593-020-0585-y
13. Srinivas S, Watanabe T, Lin CS et al (2001) Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol 1:4. https://doi.org/10. 1186/1471-213X-1-4 14. Van Duyne GD (2015) Cre recombinase. Microbiol Spect 3. https://doi.org/10.1128/ MICROBIOLSPEC.MDNA3-0014-2014
Chapter 10 Isolation and Flow Cytometry Analysis of Macrophages from White Adipose Tissue Dalila Juliana Silva Ribeiro, Seniz Yu¨ksel, Andreas Dolf, and Dagmar Wachten Abstract Macrophages are one of the prominent leukocyte populations in white adipose tissue (WAT) and play an important role during WAT homeostasis and remodeling. Macrophage function in WAT is determined by ontogeny and the local tissue environment. Here, we present a protocol to analyze different macrophage populations from murine WAT using flow cytometry. Key words White adipose tissue, Macrophages, Flow cytometry
1
Introduction The white adipose tissue (WAT) is a heterogeneous tissue that consists not only of adipocytes that store energy in form of lipids, but it also contains a variety of other cell types like endothelial cells, fibroblasts, and different immune cells. WAT dynamically adapts to change in whole body homeostasis to maintain tissue function, e.g., during higher energy intake and obesity development [1]. WAT remodeling is functionally coordinated by the different cell types in the tissue [2]. Macrophages are the most prominent leukocyte populations in WAT and play an important role during WAT homeostasis and remodeling [3]. Under lean conditions, 5–10% of the stromal cells are macrophages [4]. Macrophage function in WAT is determined by ontogeny and the local tissue environment. WAT contains yolk sac-derived, tissue-resident macrophages and bone marrow-derived monocytes/macrophages [3, 5], which control WAT development and expansion or diet-associated inflammation, respectively [6]. Different surface marker expression allows to isolate the different macrophage populations from adult WAT in mice (Fig. 1) [6].
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_10, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
149
150
Dalila Juliana Silva Ribeiro et al.
Stromal vascular fraction
CD45-
CD45+
Lin-: CD335, CD3, CD19, Ly L 6G, SinglecF
F4/80+ CD11b+
Tim4+ CD11c-
alpha
Tim4CD11c+
beta
delta
Lin-: CD335, CD3, CD19, Ly L 6G, SinglecF
F4/80CD11b-
Tim4CD11c-
gamma
Fig. 1 Schematic overview of adipose tissue-resident macrophage subpopulations. Cell surface markers that allow to delineate between the different cell populations are indicated. The macrophage populations can be divided four subpopulations (α, β, δ, γ) [6]. Figure created with Biorender.com
Macrophages subpopulations in WAT have been distinguished by the expression of TIM4 and CD11c [6] and can be further delineated by the expression of MHCII, resulting in two yolk sac-derived tissue-resident macrophages subpopulations (a, b) and two bone marrow-derived monocyte/macrophage subpopulations (g, d). Upon diet-induced obesity, resulting in adipose tissue inflammation, bone marrow-derived monocytes are recruited to the adipose tissue, increasing the number of the g/d populations in WAT [6]. In this chapter, we describe how to isolate and analyze macrophages from WAT by flow cytometry.
Isolation and Analysis of Adipose Tissue Macrophages
2 2.1
151
Materials Consumables
1. 10.0 cm curved scissors. 2. Sterile 50 mL falcon tubes. 3. Sterile 15 mL falcon tubes. 4. Sterile 1.5 mL microcentrifuge tubes. 5. 2 mL microcentrifuge/reaction tubes. 6. 10 mL Serological pipettes. 7. 96-well microplate round bottom (Corning). 8. Cell strainers: 70 μm and 100 μm. 9. Tubes for flow cytometer acquisition (5 mL round tubes).
2.2 Buffers and solutions
1. 1x Dulbecco’s phosphate buffered saline (DPBS), without calcium and magnesium. 2. FACS Buffer: 0.5% (w/v) BSA, 2 mM EDTA in 1x DPBS. 3. Fc Block solution: anti-mouse CD16/32 (Clone 93) 1:100 with 2% rat serum in FACS buffer. 4. Isolation buffer: 0.5% (w/v) BSA in 1x DPBS. 5. Digestion Buffer: 1 mg/mL Collagenase II, 2.5 mM CaCl2 in isolation buffer. 6. Erythrocyte lysis-buffer (e.g., from Biolegend). 7. 70% Ethanol in distilled H2O. 8. Trypan blue solution.
2.3
Antibodies/Dyes
1. Antibody capture compensation beads. 2. Fluorochrome-labeled anti-mouse antibodies against indicated cell surface markers (see Table 1). 3. 7-AAD (7-amino-actinomycin D) Viability Staining Solution.
2.4
Equipment
1. Refrigerated centrifuge. 2. Platform shaker with heating module. 3. Neubauer counting chamber. 4. Micropipettes. 5. Pipette boy. 6. Flow cytometer.
3
Methods Before starting, it is recommended to set the instruments to their required working temperatures. For optimal tissue digestion, the shaker should be set at 37 °C, and the centrifuge should be cooled down to 4 °C for higher cell survival during the experiment.
152
Dalila Juliana Silva Ribeiro et al.
Table 1 Antibodies
3.1 Harvesting of WAT
Antigen
Fluorophore
Concentration
Clone
NKp46 (CD335)
PE
1 μg/mL
29A1.4
CD3
PE
1 μg/mL
17A2
CD19
PE
0.5 μg/mL
1D3/CD19
Ly6G
PE
0.5 μg/mL
1A8
SiglecF
PE
2 μg/mL
S17007L
CD45
APC-Cy7
0.5 μg/mL
30-F11
CD11b
PE-Cy7
0.25 μg/mL
M1/70
F4/80
BV421
0.5 μg/mL
BM8
Tim4
AF647
1.25 μg/mL
F31-5G3
CD11c
BV605
2.5 μg/mL
N418
MHC-II
FITC
1.25 μg/mL
M5/114.15.2
1. Euthanize mice according to a locally approved procedure and disinfect the skin with 70% ethanol. 2. Using clean instruments, make a transverse cut in the abdominal area and open the abdominal cavity. Cut out the gonadal WAT (gWAT) from both sides around the testes (Fig. 2) (see Note 1). 3. Weigh the tissue of interest for each mouse (see Note 2).
3.2 Tissue Digestion and Preparation of Single-Cell Suspension from WAT
1. Transfer the tissues to a 1.5 mL tube and add 1 mL cold WAT isolation buffer. Cut the tissue into small pieces with the 10.0 cm curved scissor (see Note 3). 2. Transfer the whole content from the 1.5 mL tube to a sterile 50 mL falcon tube and add 3 mL of digestion buffer to the sample. We advise performing the tissue digestion from individual mice in separate tubes. We recommend a maximum of 0.6 g of gWAT per 4 mL of digestion buffer. Incubate for 30 min at 190 rpm in a shaker at 37 °C (see Note 4). 3. Stop the enzymatic reaction by adding 12 mL of cold isolation buffer, pipette up and down several times with a pipette boy, place a 100-μm filter to a new empty sterile 50 mL falcon tube, and apply cells to the column (see Note 5). 4. Collect the cell pellet by spinning the sample for 10 min at 500 × g, 4 °C, discard supernatant, resuspend the pellet in 1 mL erythrocyte lysis buffer for 2 min, and stop the reaction by adding 9 mL of isolation buffer.
Isolation and Analysis of Adipose Tissue Macrophages
153
Corpus Testis gWAT A gWAT A
Fig. 2 Representative figure of mice dissection. A transverse cut in the abdominal region is followed by laterally opening the skin, which allows to investigate the adipose tissue. The gonadal WAT (gWAT) is located around the testes. Figure created with Biorender.com
5. Again, collect the cell pellet by centrifuging the sample for 10 min with 500 × g, 4 °C, discard supernatant, and resuspend the pellet in 1 mL FACS buffer. 6. Determine the cell count by diluting 10 μL of the cell suspension 1:2 with Trypan Blue Solution and count the cells using a Neubauer counting chamber. 3.3 Antibody Staining of Cell Surface Markers
Before starting the staining procedure, all antibody solutions, i.e., the full staining mix containing all antibodies, single-staining solutions (see Note 6), and fluorescence minus one (FMO) control (see Note 7) should be prepared in FACS buffer and kept at 4 °C. 1. Resuspend cells in 100 μL FACS buffer in a 1.5 mL reaction tube to a final concentration of 2–3 × 106 cells. Centrifuge the cells for 10 min at 500 × g at 4 °C. 2. Resuspend the cell pellet in 100 μL of Fc Block and incubate for 20 min at 4 °C in a 1.5 mL reaction tube. 3. Plate 80 μL of the cell suspension into a round-bottom 96-well microplate and use the remaining 20 μL from each sample for FMOs and unstained controls. Seed them on the same plate. 4. Centrifuge the plate for 10 min at 500 × g, 4 °C. 5. Discard the supernatant by flipping the plate on a paper towel (see Notes 8 and 9). 6. Resuspend the cell pellet in 20 μL of antibody cocktails, as listed in Table 1 for the different stainings and FMO controls, or in FACS buffer alone for the unstained control.
154
Dalila Juliana Silva Ribeiro et al.
7. Single staining controls should be done with capture beads that are used for compensation and are stained for the same time as regular samples; to this end, add the appropriate quantity of single-staining solutions to match the concentrations indicated in Table 1 into one drop of the capture beads (vortex bead stock first). 8. Incubate samples on ice, protected from light, for 30 min. 9. Washing procedure: add 100 μL of FACS buffer and repeat steps 5 and 6. 10. Resuspend cells in 100 μL of FACS buffer. 11. Filter samples through a 70 μm cell strainer (essential to avoid clogging) and place them in the indicated tubes for flow cytometry acquisition. If starting the staining with 2 × 106 cells, dilute the samples to achieve less than 2000 events per second. Finally, add the cell viability 7-AAD dye at a final dilution of 1: 50 in the full stained samples and FMO controls. Do not add the dye to the unstained control. 12. Place the samples on ice, protected from light, and proceed to the acquisition in the flow cytometer. Determine forward and side scatters with the unstained control samples (Fig. 3). Use unstained controls to determine the background signal for all the detectors. Acquire all single-stained compensation controls and make the appropriate adjustments. Calculate compensation values and apply them for the real samples as well as for FMO controls. 13. For setting up the gating strategy, scatter parameters must be used to exclude cellular debris and doublets (Fig. 3). Selection of the population without a signal for 7-AAD will exclude dead cells (Fig. 3). FMO controls should be used to determine the placement of the subsequential gates. Figure 4 illustrates exemplary how FMOs were used to determine the BV605 signal, i.e., for the selection of CD11c-positive cells (Fig. 4a), and the FITC signal (Fig. 4b), and for the selection of MHCII positive cells (Fig. 4c, d).
4
Notes 1. Inspect the liver and spleen for possible signs of disease, abdominal carcinomas, or organomegaly. Signs of illness should be infrequent events in young wild-type mice. 2. Flow cytometry data can be analyzed in two different ways: frequency of superordinate group, e.g., frequency of CD45+ cells or cells per tissue depot. Hence, it is fundamental to weigh the tissue and calculate the number of cells per mL. The weight
Isolation and Analysis of Adipose Tissue Macrophages
155
Fig. 3 Representative pseudo-color plots of the gating strategy to discriminate the four subpopulations of ATM. The cells of interest were gated from cells of the stromal vascular fraction (SV), followed by doublets and live cell discrimination. CD45 positive signal was used to select cells from hematopoietic origin. Additional immune cell populations were excluded from the analysis using a “dump channel”, in which antibodies from different lineage markers were conjugated with the same fluorochrome. A double-positive signal for F4/80 and CD11b allows the selection of all macrophages, while TIM4, CD11c, and MHCII signals distinguish between the different subpopulations as indicated. For fluorescent channels, parameters were shown using the biexponential scale; this transformation improves the display of negative values, preventing the accumulation of events in the chart edges. The samples for this protocol were acquired in the flow cytometer Attune Nxt (ThermoFisher), following the manufacturers’ instructions on flow rate and sample volume
156
Dalila Juliana Silva Ribeiro et al.
Fig. 4 Example of how FMO controls can be used to determine the gates. (a and c) were stained with all antibodies and followed the same gating strategy as presented in Fig. 2. (b and d) were treated as a and c, except they lack one antibody each, i.e., anti-CD11c conjugated to BV605 (b) and anti-MHC-II conjugated to FITC (d)
of the fat depot depends on the age and diet of the mice. The final cell frequency may depend on the WAT depot, nevertheless all described markers can be applied for all WAT types. 3. Cold isolation buffer will prevent denaturation of the tissue, and smaller pieces of the tissue will result in better digestion efficiency. 4. Attention should be paid to avoid prolonged incubation times with the digestion buffer, since extensive collagenase digestion can degrade the epitopes on the cell surface.
Isolation and Analysis of Adipose Tissue Macrophages
157
5. Adjust the pipet boy to low and filter the solution into a new tube. This step will provide a clean pellet and will avoid contamination of reminiscent lipid particles coming from the walls of the falcon tube. 6. Compensation can be performed using the values generated by individual antibody labeling for each parameter using the cells of interest or capture beads. The latter is recommended when dealing with cell surface markers that show a low expression level and/or when the protein expression is limited to sparse cell populations. As the viability dye belongs to the group of nuclei acid binding-dyes, the single staining control can only be performed with cells for compensation. Unstained cells are used to determine morphological parameters and autofluorescence. 7. FMO controls consist of cells stained with all fluorophores, except the one being measured, and are essential when running a multicolor flow-cytometry experiment. FMOs help to account for the fluorescence spread generated after compensation to identify possible false-positive signals. When setting up a multicolor flow-cytometry panel for the first time, FMOs should be included for all fluorophores. Afterwards, they can be used to separate only rare populations. 8. Staining in plates can make the procedure faster in terms of pipetting time, especially during washing steps. However, this is also more prone to contamination between adjacent samples, which can be detrimental for single-staining controls. We recommend leaving spaces between the wells. It is also possible to perform the staining procedure directly in flow cytometry tubes. 9. As an alternative to prevent non-specific polymer interaction, Brilliant Stain Buffer (InvitrogenTM) can be added prior to the staining. It is used as a complement for multicolor flow-cytometry experiments and is only necessary when more than one polymer dye-conjugated is included.
Acknowledgments We thank Kim Dressler for help with isolating the tissue. Research in the Wachten lab was supported by grants from the DFG – SFB 1454 – project number 432325352, TRR83 – 112927078, TRR333/1 – 450149205, under Germany’s Excellence Strategy – EXC2151 – 390873048 as well as intramural funding from the University of Bonn.
158
Dalila Juliana Silva Ribeiro et al.
References 1. Rosen ED, Spiegelman BM (2014) What we talk about when we talk about fat. Cell 156:20–44 2. Choe SS, Huh JY, Hwang IJ, Kim JI, Kim JB (2016) Adipose tissue remodeling: its role in energy metabolism and metabolic disorders. Front Endocrinol (Lausanne) 7:30 3. Chakarov S, Bleriot C, Ginhoux F (2022) Role of adipose tissue macrophages in obesity-related disorders. J Exp Med 219:e20211948 4. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr (2003) Obesity is associated with macrophage
accumulation in adipose tissue. J Clin Invest 112:1796–1808 5. Cox N, Geissmann F (2020) Macrophage ontogeny in the control of adipose tissue biology. Curr Opin Immunol 62:1–8 6. Cox N, Crozet L, Holtman IR, Loyher PL, Lazarov T, White JB, Mass E, Stanley ER, Elemento O, Glass CK et al (2021) Dietregulated production of PDGFcc by macrophages controls energy storage. Science 373: eabe9383
Chapter 11 Isolation and Flow Cytometry Analysis of Macrophages from the Dermis Aaron James Forde and Julia Kolter Abstract The dermis contains a dense network of tissue macrophages, which contribute to tissue homeostasis, inflammation, and pathogen clearance. Dermal macrophages are partly replenished by circulating monocytes, which fuel the resident population, especially in case of tissue damage or inflammation. The complexity of the tissue, containing blood and lymphoid vessels, hair bulbs, sebaceous glands, and peripheral nerves, allows for the development of distinct macrophages populations. In steady state, discrete subtypes can be distinguished due to their surface marker expression and localization within the dermis. In this chapter, we describe how to extract dermal macrophages from the skin and highlight different gating strategies to identify monocyte and macrophage populations. Key words Macrophages, Monocytes, Dermis, Skin, Mononuclear Phagocytes, Isolation
1
Introduction The skin, the external surface of the body, consists of several layers [1]. The outermost layer, the epidermis, serves as the barrier to the environment. It is mainly composed of keratinocytes and separated from the dermis by a basement membrane, the dermo-epidermal junction. The epidermis contains Langerhans cells (LC), a large population of specialized tissue macrophages which self-maintain in steady state [2]. Interestingly, LCs resemble dendritic cells (DC) in some regards, specifically they are capable of migrating to the lymph nodes to present antigen, and express the DC transcription factor Zbtb46 [3]. Yet, LCs also exhibit features of macrophages with respect to origin and the expression of MafB [3]. In addition, the murine epidermis is home to dendritic epidermal T cells (DETCs), which are yolk sac derived and capable of self-renewal [4]. While the epidermis represents a rather protected environment without its own blood supply, the dermis, the inner layer of the skin, is considerably more heterogeneous. It consists of connective
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_11, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
159
160
Aaron James Forde and Julia Kolter
tissue, mainly collagen produced by fibroblasts, and a multitude of supplying structures such as blood vessels, nerves, sweat glands, and lymphatics. In addition, the dermis contains a variety of immune cells, such as mast cells, dendritic cells, γδ as well as αβ tissue-resident T cells and macrophages. Around and shortly after birth, dermal macrophages, the predominant immune cells in the dermis, rapidly adapt to the extrauterine environment through establishment of a heterogeneous “mature” population. In contrast to LCs in the epidermis, dermal macrophages have been described to be continuously replaced by monocytes in steady state, similar to intestinal macrophages [5, 6]. However, in these earlier studies the complex and diverse nature of dermal macrophages was not yet fully appreciated, as demonstrated by the fact that melanophages (melanin-laden macrophages) were excluded from analysis via flow cytometry due to their strong light-scattering properties [7]. More recently, self-maintaining subsets of macrophages were found to exist locally in the intestine, e.g. associated with blood vessels, the enteric nervous system or Peyer’s patches [8, 9]. Similarly, we identified a subset of embryonically derived dermal macrophages associated with sensory nerves that self-maintain in homeostasis with minimal contribution by monocytes [10, 11]. These macrophages survey the axons of sensory dermal nerves, trim the myelin sheath, and contribute to axon sprouting after mechanical injury. The subset can be identified by high expression of the fractalkine receptor CX3CR1, which is postnatally downregulated in other dermal macrophages. Incoming Ly6Chigh monocytes, which possess an intermediate CX3CR1 expression in comparison, also downregulate the receptor upon differentiation [5, 10]. Accordingly, using the widely available Cx3cr1gfp/+ reporter mice [12, 13], three subsets of adult dermal macrophages can be defined, which reside in the same tissue, yet are diverse with respect to origin, renewal, localization, and function [10]. Within this classification, the CX3CR1low subset represents the largest population of macrophages in the adult dermis (more than 80%). We and others have shown that this subset receives large input by monocytes, yet parts of the population are not or only very slowly exchanged in mouse [7, 10, 14] as well as human skin [15]. In support of this, monocyte-deficient CCR2-/- mice have normal macrophage numbers in the dermis, indicating compensatory proliferation by resident cells [10, 16]. The factors which determine the fate and exchange of cells within this population remain to be uncovered. Previous studies additionally distinguished dermal macrophages based on Ly6C and MHC class II expression [5, 7]. Monocytes were identified as Ly6C+ CCR2+ CD64low cells. In the CD64high fraction, macrophage populations were characterized
Isolation of Dermal Macrophages
161
by the presence or absence of MHC class II expression, which were named P4 and P5, respectively [5, 7]. Within this classification, MHC II- macrophages showed reduced exchange and selfrenewing potential in comparison to MHC II+ macrophages [7, 17]. Mature resident interstitial macrophages also upregulate the mannose receptor CD206 [18], which is not expressed by nerve-associated macrophages [10]. Furthermore, the hyaluronan receptor Lyve1 has been suggested to label macrophages aligning with certain blood vessels [19, 20]. As the predominant immune cell type in the resting dermis, macrophages have essential and non-redundant roles in host defence against opportunistic pathogens such as the bacteria Staphylococcus aureus [21, 22]. Notably, surface expression and, hence, immunophenotype of resident macrophages can change upon injury or infection and the influx of inflammatory cell types may additionally complicate their identification under these conditions. MHC class II expression is a classic example for a surface factor that is strongly regulated upon stimulation and may be induced upon exposure to e.g. IFNγ [23]. Hence, investigators need to be careful with the classification of macrophages based on this marker, as it may vary and only labeling of resident cells (via fate-mapping approaches) would enable a reliable characterization of the previously existing subtypes. In our experience, CD64, however, is stably expressed upon infection or injury and is readily upregulated on monocytes after differentiation [10, 21]. Finally, most studies use ear tissue to extract dermal macrophages. However, it should be noted that the cellular composition between different anatomical skin sites varies and that the frequencies of macrophages, but also other immune cell types such as epidermal γδ T cells, were found to differ e.g. in hairy and glabrous skin [24]. The source of skin should thus be taken into account when comparing results of different studies. Altogether, the careful characterization of dermal macrophages is crucial to thoroughly understand their cellular kinetics and functional properties in steady state and disease. We hence aim to provide investigators with a toolbox to reliably perform these analyses using flow cytometry.
2
Materials
2.1 Required Reagents
1. Digestion buffer: 1x Hanks Balanced Salt Solution (HBSS) without calcium and magnesium, 10% Fetal Calf Serum (FCS), 0.04 mg/mL DNase I, 0.25 U/mL Dispase, 1 mg/ mL Collagenase II. 2. FACS buffer: Dulbecco’s phosphate-buffered saline (DPBS) without calcium and magnesium, 2 mM EDTA, 2% FCS.
162
Aaron James Forde and Julia Kolter
Table 1 Antibody panel for dermal macrophage identification
Antibody/Reagent
Clone
Fluorochrome
Recommended final concentration
CD16/CD32 (Fc block)
93
–
5 μg/mL
CD11b
M1/70
PeCy7
0.07 μg/mL
CD45
30-F11
E450
0.7 μg/mL
CD64
X54–5/7.1
PerCP-Cy5.5
2 μg/mL
CD206
C068C2
PE
1 μg/mL
Ly6C
HK1.4
BV605
0.25 μg/mL
MHC II
M5/114.15.2
FITC
2.5 μg/mL
Ly6G
1A8
APC
0.4 μg/mL
CD3e
145–2 C11
APC
0.4 μg/mL
CD11c
HL3
APC
2 μg/mL
NK 1.1
PK136
Biotin
5 μg/mL
CD19
6D5
Biotin
1.5 μg/mL
SiglecF
ES22-10D8
Biotin
1.5 μg/mL
Fixable viability dye
–
eFluor 780
–
Streptavidin
–
APC
0.4 μg/mL
Antibodies and other reagents required to identify dermal macrophages in whole skin suspensions via flow cytometry as depicted in Fig. 1a
3. Fc block: FACS buffer containing 5 μg/mL anti-CD16/32 antibody (see Table 1). 4. Antibodies & fixable viability dye (optional) (see Table 1). 2.2
Equipment
1. Scissors (see Note 1). 2. Forceps. 3. 1.5 mL micro centrifuge tubes. 4. 50 mL centrifugation tubes. 5. FACS tubes. 6. 70 μM cell strainer. 7. Heat block for 1.5 mL tubes with shaker. 8. Centrifuge for 50 mL tubes and FACS tubes. 9. Flow cytometer.
Isolation of Dermal Macrophages
3 3.1
163
Methods Preparation
1. Prepare a master mix with 1 mL of digestion buffer per sample (see Note 2). Add 1 mL to each 1.5 mL centrifugation tube. 2. Prewarm heating block to 37 °C. 3. Cool centrifuge to 4 °C.
3.2 Sample Processing and Enzymatic Digestion
1. Sacrifice mouse according to ethical guidelines (see Note 3). 2. Cut ears close to the hairline (see Note 4). 3. Using scissors remove any remaining fur from the ears and transfer to a 1.5 mL microcentrifuge tube containing 1 mL digestion buffer. It is recommended to digest 1 ear in 1 mL of digestion buffer. 4. Cut ear into very small pieces using sharp scissors. 5. Vortex tube thoroughly and place into heat block set at 37 °C for 2 h with vigorous shaking (see Note 5). 6. Pass suspension through a 70 μM strainer into 50 mL centrifugation tubes by pipetting and rinse with 10 mL FACS buffer. 7. Centrifuge samples at 400 g for 8 min at 4 °C. 8. Decant supernatant and leave centrifugation tube upside down on a paper towel in order to remove remaining supernatant (see Note 6).
3.3 Staining for Flow Cytometry
1. Prepare a master mix for Fc block (50 μL per sample) (see Note 7). 2. Resuspend pellet in 50 μL Fc block and transfer the sample to a FACS tube by pipetting. 3. Incubate at 4 °C for 5–10 min. 4. In the meantime, prepare the antibody master mix using 50 μL per sample and the required antibodies in 2x concentration (see Note 8). Suggested antibody panels for a standard staining using 8 fluorescence channels can be found in Table 1 (SeeNote 9). 5. Add 50 μL of the 2x antibody mix to each sample, adding up to a total volume of 100 μL (SeeNote 10). 6. Incubate samples at 4 °C for approximately 30 min, protected from light. 7. Add 1 mL of FACS buffer and centrifuge at 400 g for 8 min at 4 °C. 8. Decant supernatant. Proceed to step 11 if using only conjugated dyes.
164
Aaron James Forde and Julia Kolter
9. Optional: If biotin-labeled primary antibodies were used, resuspend pellet in 100 μL FACS buffer containing secondary streptavidin-labeled antibody. Incubate for 15 min at 4 °C, then repeat step 7 + 8. 10. Optional: Resuspend in 500 μL DPBS and add 0.5 μL Fixable Viability dye. Incubate for 15 min at 4 °C protected from light. Add 1 mL of FACS buffer and repeat centrifugation step (seeNote 11). 11. Resuspend the cell pellet in 200 μL FACS buffer, store on ice and cover with aluminum foil to protect from light. Acquire on a flow cytometer within the next hour (see Note 12). 3.4 Flow Cytometry and Gating Strategies
1. To identify dermal macrophages in the single cell suspensions, follow the gating strategy depicted in Fig. 1a. Cells are identified in the forward/side scatter by excluding debris (“cells”, seeNote 13). Next, single cells are gated for by excluding doublets, which represent cells high in the forward and sideward scatter pulse widths (“singlets”). Live cells are identified as events negative for fixable viability dye (“Live”). 2. CD45+ leukocytes are gated for on the basis of CD45 expression. This step will lead to exclusion of other skin cell types such as endothelial cells, fibroblasts, and keratinocytes. 3. Macrophages are CD11b+ cells ((“CD11b+”), seeNote 14). In order to exclude other CD11b+ cell types, one additional channel is required, in which Siglec-F (eosinophils), NK1.1 (NK cells), Ly6G (neutrophils), CD3e (T cells), CD19 (B cells), and CD11c (dendritic cells, epidermal Langerhans cells) are stained (Fig. 1h, seeNote 15). 4. Dermal macrophages can be gated as CD64high cells within the CD45+CD11b+lineage- population (“MΦ”). 5. To identify subpopulations of dermal macrophages, Ly6C and MHC class II can be plotted. Ly6C- macrophages are MHC II low or high and mostly co-express CD206 (Fig. 1a). 6. Adult Cx3cr1gfp/+ reporter mice may be used to identify nerveassociated macrophages (CX3CR1high) and macrophages recently derived from monocytes (CX3CR1int). Ly6C can be used to draw the gate for CX3CR1int cells which co-express Ly6C (Fig. 1d, e). Cells with higher GFP expression are CX3CR1high (i.e., nerve-associated macrophages), cells with lower GFP expression represent CX3CR1low interstitial macrophages (see Note 16). 7. If skin cells are extracted from other sources than the ear, the cellular frequencies may change but the gating strategy remains the same (Fig. 1b, c, f, and g).
Isolation of Dermal Macrophages
Gated on singlets Live
Live/Dead
Gated on singlets Live
Gated on MΦ
CD206
Gated on Live
Gated on CD11b+
Gated on MΦ
CD45
Gated on MΦ
Ly6C
Back: Cx3cr1gfp/+ low
MHC II
G
Tail: Cx3cr1gfp/+
iint high
low
CX3CR1 GFP
CD206
Gated on DCs
Ly6C CD206
CD64
cDC1
Eosinophils
Gated on CD11blow
Gated on CD11b+
iint high
Ly6C
Ly6C
CX3CR1 GFP
Neutrophils
T cells Gated on CD11blow
Gated on CD11b+
MHC II
MHC II
CD64
CD3
Ly6G
CD11c
Ly6C CD11b
SiglecF
DCs CD11c
CD11b
CD11b+
MHC II
MΦ
CD45
LC + cDC2 Gated on CD11b+
Gated on Live
Ly6C
CD64
F
H
Gated on MΦ
Ly6C
Lineage CD45
CX3CR1high CX3CR1int CX3CR1low
CX3CR1 GFP
MHC II
MΦ
E
CX3CR1 GFP
Ly6C
Gated on CD11b+
iint high
CD11blow
Ly6C
Lineage Gated on Live
CD11b
SSC-W
SSC-A
CD45
Live/Dead
FSC-W
Ear: Cx3cr1gfp/+ low
CD206
CD11b+
singlets
FSC-A
CD64
Ly6C
Gated on cells cells
CD45
Ly6C
Tail
MΦ
CD11b
CD45
SSC-A
SSC-W
FSC-W
Gated on MΦ
CD11b+
singlets
FSC-A
D
Live/Dead
Gated on cells cells
C
CD11b
SSC-W
SSC-A
Back
Gated on MΦ
CD11b+
singlets
FSC-W
Gated on CD11b+
Gated on Live
Live
FSC-A
B
Gated on singlets
Gated on cells cells
Lineage
Ear
CD45
A
165
Ly6C
CD45
Fig. 1 Identification of macrophages and other immune cell subsets in the skin. (a) Gating strategy to identify dermal macrophages in single cell suspensions of digested ear skin. Cell debris, doublets, and dead cells are excluded based on their SSC/FSC properties and live/dead staining. Single live cells are then gated by their expression of CD45 and CD11b. Next, lineage+ cells are excluded and dermal macrophages identified by the absence of lineage markers and CD64 expression. (b, c) Flow cytometric analysis of back skin (b) and tail skin (c) using the same gating strategy as in (a). (d) Macrophages gated as in (a) and plotted for CX3CR1-GFP against Ly6C expression to identify CX3CR1low, int and high macrophages. (e) Expression of CX3CR1-GFP, Ly6C, and CD206 by the subsets identified in (d). (f, g) CX3CR1-GFP expression of dermal macrophages in back (f) and tail (g) skin. (h) Backgating on lineage markers to identify other immune cell subsets in the skin. CD11b+ cells include LCs and cDC2, eosinophils and neutrophils. CD11blow cells contain cDC1, T cells as well as NK and B cells (not shown)
4
Notes 1. The quality and sharpness of scissors are essential for efficient digestion of the skin. We recommend using straight stainless steel dissecting scissors (e.g., Hammacher) and regular sharpening.
166
Aaron James Forde and Julia Kolter
2. The digestion mix containing FCS, DNase, and dispase can be prepared in larger batches and stored at -20 °C for at least 1 month. We recommend to add Collagenase II directly to each tube as it is sensitive to multiple freeze-thaw cycles. 3. We strongly recommend to not punch the ears of the mice to obtain biopsies if they shall be used for skin analysis. Ear punches induce strong inflammation in the ears, and severely change the renewal patterns and surface expression of resident macrophages. We recommend to use tail tissue for genotyping (if the respective animal welfare regulations allow for it) or perform biopsies on one ear only and use the contralateral ear for analysis. 4. Other sources of skin, e.g. tail or back skin, may also be used (Fig. 1b, c). However, in the case of back skin the fur must be removed (via shaving or depilatory cream) before harvesting the skin. Additionally, prior to digestion, the underlying fat tissue must be carefully removed using forceps and scalpel. Failure to remove the fat tissue will result in decreased digestion efficiency and increased auto fluorescence upon acquisition. 5. Vigorous shaking is important for the tissue not to settle down in the bottom of the tubes, which in our experience severely reduces the cell extraction efficiency. Vortex immediately before inserting tube into the shaking heating block and shake at maximum power. 6. Do not inverse tubes again to avoid loss of pellet. 7. Fc Block is required to avoid unspecific binding of FcγRexpressing cells to antibodies. Since macrophages express these receptors, binding may lead to exclusion of macrophages due to unspecific positive staining for lineage markers such as CD3 or misleading conclusions about the expression of other surface markers. Hence, blocking of the receptors is an essential step of the protocol. 8. If multiple staining panels are required, the cells can be distributed to several tubes in this step. Typically, we can extract approximately 1 × 106 total cells per ear, including up to 2.5 × 104 macrophages. Thus, depending on the experimental aim, it may not be advisable to split the cells if subsets should be recorded in sufficient numbers. 9. It is recommended to include appropriate controls such as unstained, single stains, and fluorescent minus one (FMO) controls when initially establishing the panels. 10. Alternatively, a washing step can be introduced here to wash off the Fc block from the cells and resuspend the pellet directly in
Isolation of Dermal Macrophages
167
the antibody mix (1x concentrated). In our experience, however, additional centrifugation steps contribute to cell loss and should thus be avoided. 11. The digestion process usually affects cell viability, however, we have not detected cell death in dermal macrophages using the described procedures and incubation times. Despite this, cell viability dyes help to exclude other dead skin cells and debris and are thus recommended, at least for the initial setups or cell sorting. Fixable viability dyes should be stained in protein-free buffers according to manufacturer’s instructions. If the staining panels allow it, other cell viability dyes such as DAPI, 7-AAD or PI can be used and should then be added to the samples approximately 5 min before acquisition. 12. If longer storage (max. up to 48 h) is required, cells can be fixed in 2% paraformaldehyde in PBS for 20 min at 4 °C. Repeat washing steps before resuspending the cells in 200 μL FACS buffer. Cover with aluminum foil and keep the samples at 4 °C until acquisition 13. Macrophages strongly scatter light, especially due to the content of melanin (see Subheading 1). Thus, it is very important to use logarithmic scales for acquisition flow cytometry and to not exclude cells with large size or granularity in forward/ sideward scatters. 14. Macrophages are highly autofluorescent and will always appear brighter in fluorescence compared to other cell types in the tissue, even without the addition of the respective fluorophores. Thus, it is important to use FMOs to draw gates accurately. 15. CD3e+ T cells should always be excluded along with myeloid cell types since, in our experience, CD3e+ cells will otherwise be extracted along with macrophages leading to contamination in downstream analysis e.g. RNA sequencing approaches. As a minimum, CD11c+ and CD3e+ cells should be excluded to ensure a pure macrophage population, as the other cell types usually do not fall into the CD64+ gate. In case of infection or inflammation, the usage of an anti-Ly6G antibody is essential as invading and potentially pre-apoptotic neutrophils may hamper a clear distinction of macrophages. 16. CX3CR1int macrophages co-express Ly6C and/or MHC class II, while CX3CR1low and high macrophages do not express Ly6C and are mostly low for MHC class II.
168
Aaron James Forde and Julia Kolter
Acknowledgments J.K. was funded by the Hans A. Krebs Medical Scientist Programme (Faculty of Medicine, University of Freiburg) and by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – TRR 359 – Project number 491676693. References 1. Kabashima K, Honda T, Ginhoux F, Egawa G (2019) The immunological anatomy of the skin. Nat Rev Immunol 19(1):19–30. https://doi.org/10.1038/s41577-0180084-5 2. Schulz C, Gomez Perdiguero E, Chorro L, Szabo-Rogers H, Cagnard N, Kierdorf K, Prinz M, Wu B, Jacobsen SE, Pollard JW, Frampton J, Liu KJ, Geissmann F (2012) A lineage of myeloid cells independent of Myb and hematopoietic stem cells. Science 336(6077):86–90. https://doi.org/10.1126/ science.1219179 ˜ o CG, Durai V, Albring JC, 3. Wu X, Brisen Haldar M, Bagadia P, Kim K-W, Randolph GJ, Murphy TL, Murphy KM (2016) Mafb lineage tracing to distinguish macrophages from other immune lineages reveals dual identity of Langerhans cells. J Exp Med 213(12): 2553–2565. https://doi.org/10.1084/jem. 20160600 4. Gentek R, Ghigo C, Hoeffel G, Jorquera A, Msallam R, Wienert S, Klauschen F, Ginhoux F, Baje´noff M (2018) Epidermal γδ T cells originate from yolk sac hematopoiesis and clonally self-renew in the adult. J Exp Med 215(12):2994–3005. https://doi.org/10. 1084/jem.20181206 5. Tamoutounour S, Guilliams M, Montanana Sanchis F, Liu H, Terhorst D, Malosse C, Pollet E, Ardouin L, Luche H, Sanchez C, Dalod M, Malissen B, Henri S (2013) Origins and functional specialization of macrophages and of conventional and monocyte-derived dendritic cells in mouse skin. Immunity 39(5):925–938. https://doi.org/10.1016/j. immuni.2013.10.004 6. Sieweke MH, Allen JE (2013) Beyond stem cells: self-renewal of differentiated macrophages. Science 342(6161):1242974. https:// doi.org/10.1126/science.1242974 7. Baranska A, Shawket A, Jouve M, Baratin M, Malosse C, Voluzan O, Vu Manh TP, Fiore F, Bajenoff M, Benaroch P, Dalod M, Malissen M, Henri S, Malissen B (2018) Unveiling skin macrophage dynamics explains both tattoo persistence and strenuous removal.
J Exp Med 215(4):1115–1133. https://doi. org/10.1084/jem.20171608 8. De Schepper S, Verheijden S, AguileraLizarraga J, Viola MF, Boesmans W, Stakenborg N, Voytyuk I, Smidt I, Boeckx B, Dierckx de Casterle I, Baekelandt V, Gonzalez Dominguez E, Mack M, Depoortere I, De Strooper B, Sprangers B, Himmelreich U, Soenen S, Guilliams M, Vanden Berghe P, Jones E, Lambrechts D, Boeckxstaens G (2018) Self-maintaining gut macrophages are essential for intestinal homeostasis. Cell 175(2):400–415.e413. https://doi.org/10. 1016/j.cell.2018.07.048 9. Shaw TN, Houston SA, Wemyss K, Bridgeman HM, Barbera TA, Zangerle-Murray T, Strangward P, Ridley AJL, Wang P, Tamoutounour S, Allen JE, Konkel JE, Grainger JR (2018) Tissue-resident macrophages in the intestine are long lived and defined by Tim-4 and CD4 expression. J Exp Med 215(6):1507–1518. https://doi.org/10. 1084/jem.20180019 10. Kolter J, Feuerstein R, Zeis P, Hagemeyer N, Paterson N, d’Errico P, Baasch S, Amann L, Masuda T, Lo¨sslein A, Gharun K, MeyerLuehmann M, Waskow C, Franzke C-W, Gru¨n D, L€ammermann T, Prinz M, Henneke P (2019) A subset of skin macrophages contributes to the surveillance and regeneration of local nerves. Immunity 50(6):1482–1497. e1487. https://doi.org/10.1016/j.immuni. 2019.05.009 11. Kolter J, Kierdorf K, Henneke P (2020) Origin and differentiation of nerve-associated macrophages. J Immunol 204(2):271–279. https:// doi.org/10.4049/jimmunol.1901077 12. Jung S, Aliberti J, Graemmel P, Sunshine MJ, Kreutzberg GW, Sher A, Littman DR (2000) Analysis of fractalkine receptor CX3CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol 20:4106–4114 13. Yona S, Kim KW, Wolf Y, Mildner A, Varol D, Breker M, Strauss-Ayali D, Viukov S, Guilliams M, Misharin A, Hume DA, Perlman H, Malissen B, Zelzer E, Jung S
Isolation of Dermal Macrophages (2013) Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis. Immunity 38(1): 79–91. https://doi.org/10.1016/j.immuni. 2012.12.001 14. Ginhoux F, Guilliams M (2016) Tissueresident macrophage ontogeny and homeostasis. Immunity 44(3):439–449. https://doi. org/10.1016/j.immuni.2016.02.024 15. McGovern N, Schlitzer A, Gunawan M, Jardine L, Shin A, Poyner E, Green K, Dickinson R, Wang XN, Low D, Best K, Covins S, Milne P, Pagan S, Aljefri K, Windebank M, Miranda-Saavedra D, Larbi A, Wasan PS, Duan K, Poidinger M, Bigley V, Ginhoux F, Collin M, Haniffa M (2014) Human dermal CD14(+) cells are a transient population of monocyte-derived macrophages. Immunity 41(3):465–477. https://doi.org/ 10.1016/j.immuni.2014.08.006 16. Lohrmann F, Forde AJ, Merck P, Henneke P (2021) Control of myeloid cell density in barrier tissues. FEBS J 288(2):405–426. https:// doi.org/10.1111/febs.15436 17. Sheng J, Ruedl C, Karjalainen K (2015) Most tissue-resident macrophages except microglia are derived from fetal hematopoietic stem cells. Immunity 43(2):382–393. https://doi. org/10.1016/j.immuni.2015.07.016 18. Hoeffel G, Debroas G, Roger A, Rossignol R, Gouilly J, Laprie C, Chasson L, Barbon P-V, Balsamo A, Reynders A, Moqrich A, Ugolini S (2021) Sensory neuron-derived TAFA4 promotes macrophage tissue repair functions. Nature 594(7861):94–99. https://doi.org/ 10.1038/s41586-021-03563-7 19. Chakarov S, Lim HY, Tan L, Lim SY, See P, Lum J, Zhang X-M, Foo S, Nakamizo S, Duan K, Kong WT, Gentek R, Balachander A, Carbajo D, Bleriot C, Malleret B, Tam JKC, Baig S, Shabeer M, Toh S-AES, Schlitzer A, Larbi A, Marichal T, Malissen B, Chen J, Poidinger M, Kabashima K, Bajenoff M, Ng LG, Angeli V, Ginhoux F (2019) Two distinct
169
interstitial macrophage populations coexist across tissues in specific subtissular niches. Science 363(6432):eaau0964. https://doi.org/ 10.1126/science.aau0964 20. Lim HY, Lim SY, Tan CK, Thiam CH, Goh CC, Carbajo D, Chew SHS, See P, Chakarov S, Wang XN, Lim LH, Johnson LA, Lum J, Fong CY, Bongso A, Biswas A, Goh C, Evrard M, Yeo KP, Basu R, Wang JK, Tan Y, Jain R, Tikoo S, Choong C, Weninger W, Poidinger M, Stanley RE, Collin M, Tan NS, Ng LG, Jackson DG, Ginhoux F, Angeli V (2018) Hyaluronan receptor LYVE-1-expressing macrophages maintain arterial tone through hyaluronan-mediated regulation of smooth muscle cell collagen. Immunity 49(2): 326–341. e327. https://doi.org/10.1016/j. immuni.2018.06.008 21. Feuerstein R, Forde AJ, Lohrmann F, Kolter J, Ramirez NJ, Zimmermann J, Gomez de Agu¨ero M, Henneke P (2020) Resident macrophages acquire innate immune memory in staphylococcal skin infection. elife 9. https:// doi.org/10.7554/eLife.55602 22. Feuerstein R, Seidl M, Prinz M, Henneke P (2015) MyD88 in macrophages is critical for abscess resolution in staphylococcal skin infection. J Immunol 194(6):2735–2745. https:// doi.org/10.4049/jimmunol.1402566 23. Kiritsy MC, Ankley LM, Trombley J, Huizinga GP, Lord AE, Orning P, Elling R, Fitzgerald KA, Olive AJ (2021) A genetic screen in macrophages identifies new regulators of IFNγ-inducible MHCII that contribute to T cell activation. elife 10:e65110. https://doi.org/ 10.7554/eLife.65110 24. Tong PL, Roediger B, Kolesnikoff N, Biro M, Tay SS, Jain R, Shaw LE, Grimbaldeston MA, Weninger W (2015) The skin immune atlas: three-dimensional analysis of cutaneous leukocyte subsets by multiphoton microscopy. J Invest Dermatol 135(1):84–93. https://doi. org/10.1038/jid.2014.289
Chapter 12 Isolation and Flow Cytometry Analysis of Macrophages from the Kidney Sarah J. Miller, Alex Yashchenko, and Kurt A. Zimmerman Abstract Renal macrophages help maintain homeostasis, participate in tissue injury and repair, and play a vital role in immune surveillance [1–3]. Kidney macrophages can be broken down into two subsets, infiltrating macrophages, which can be further broken down into Ly6Chi and Ly6Clo cells, and kidney resident macrophages. While recent studies have shed light on the differing origins and niches of these cells, a more thorough understanding of kidney macrophage populations and how they may respond to various conditions is needed. This protocol describes how to efficiently isolate murine kidney macrophage populations for flow cytometry analysis. Key words Flow cytometry, Kidney, Kidney resident macrophages, Infiltrating macrophage, Monocyte, Ly6Chi monocyte, Ly6Clo monocyte
1
Introduction The renal macrophage compartment is made up of Ly6Chi monocyte-derived infiltrating macrophages, Ly6Clo monocytederived infiltrating macrophages, and kidney resident macrophages. Both Ly6Chi and Ly6Cloinfiltrating macrophages enter the kidney at low levels in homeostasis, but can dramatically increase their infiltration during inflammation or injury [3, 4]. Ly6Chi macrophages accumulate in renal tissue in response to injury and promote tissue damage by producing inflammatory cytokines in the initial stages of the immune response [5]. Ly6Clo macrophages, which are derived from Ly6Chi macrophages, are pro-fibrotic and promote resolution of injury and wound healing [5–7]. In contrast to these infiltrating subtypes, kidney resident macrophages are tissue resident cells that are first seeded during embryonic development and maintain their population through both in-situ proliferation and recruitment of bone marrow-derived Ly6chimonocyte precursors [1, 4]. These cells are thought to
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_12, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
171
172
Sarah J. Miller et al.
contribute to vascularization of the kidney and kidney development, immune surveillance, and play a role in tissue repair following acute kidney injury [3, 8]. Together, these cells make up approximately 50% of CD45+ immune cells in the murine kidney [1]. A murine kidney is comprised of 16,000 functional units, called nephrons, and much like human kidneys play a vital role in regulating blood pressure, whole-body water content, electrolyte balance, and red blood cell mass [9, 10]. The tissue environment of the kidney can vary dramatically depending on the region being examined and the physiological need of the animal; macrophages occupying the renal medulla are subjected to an increasing osmolarity the closer they get to the renal pelvis and are exposed to bacteria following urinary tract infections [9, 11]. In contrast, macrophages found in the renal cortex do not typically encounter bacteria and enjoy normal osmolarity in their environment. The ability to isolate and analyze renal macrophage populations found in these dynamic environments is vital for understanding the roles they play in homeostasis and disease. This protocol describes the harvesting of murine kidneys under terminal anesthesia followed by cardiac perfusion, tissue processing, and preparation of cells for analysis by flow cytometry. Day 1 is devoted entirely to preparation of the tools and equipment needed. Day 2 begins with euthanasia of the animals and tissue harvest. Avertin (2,2,2-Tribromoethanol) is the anesthetic of choice in this protocol, as isoflurane has been shown to impact immune cell populations by other groups [12]. Once the animal is unconscious, the peritoneal and thoracic cavities are opened up for cardiac perfusion; perfusion is done in this protocol to minimize the chance of immune cells from the blood (i.e., monocytes) contaminating the kidney. Mincing of the kidney tissue is followed by enzymatic digestion, 70 μm filtration, red blood cell lysis, and resuspension in Fc blocking buffer before antibody staining is done. As Day 2 in its entirety is usually a 6-h process from start to finish, the given protocol advises fixing cells for analysis within 48 h. However, if desired, this protocol can easily be adapted to allow analysis of the cells immediately after antibody staining. The final day of the protocol consists of running the stained samples on a flow cytometer, followed by analysis of the data.
2
Materials
2.1 Chemical Reagents
1. Dulbecco’s Phosphate-Buffered Saline (DPBS), 1X, with calcium and magnesium. 2. RPMI without L-glutamine and phenol red. 3. Ammonium-Chloride-Potassium (ACK) Lysing Buffer: 155 mM Ammonium Chloride, 10 mM Potassium Bicarbonate, 0.1 mM EDTA in water.
Analysis of Macrophages from the Kidney
173
4. Trypan blue solution. 5. Type IV DNase I master stock 30,000 units/mL in 0.15 M NaCl (see Note 1). 6. Tribromoethanol master stock: 1 g/mL 2,2,2Tribromoethanol in 2-Methyl-2-butanol (see Note 1). Dissolve overnight at 37 °C. Store at 4 °C protected from light. 7. Avertin working stock (see Note 2): Mix 1 mL Tribromoethanol master stock with 39 mL prewarmed 1X DPBS. Good for 1 week at 4 °C. 8. 1% Heat Shock Treated Bovine Serum Albumin (BSA): 1% w/v in DPBS. Prepare ~5 mL per mouse. Prepare fresh (see Note 2). 9. 2% PFA: 2% v/v in DPBS. Prepare 200 μL per well. Prepare fresh (see Note 2). 10. Digestion Buffer: 1 mg/mL Type I Collagenase (from C. histolyticum, ≥125 CDU/mg solid), 100 units/mL Type IV Deoxyribonuclease I (from bovine pancreas, lyophilized powder, ≥2000 Kunitz units/mg protein) in RPMI without L-glutamine and phenol red. Prepare 1 mL per kidney. Prepare fresh (see Note 2). 11. Fc blocking buffer: Dilute 1:200 anti-mouse CD16/CD32 in 1% BSA. Prepare 1 mL per kidney, 1–2 mL per spleen. Prepare fresh (see Note 2). 12. 70% Ethanol 13. Anesthesia cocktail: e.g., Avertin working stock (0.5 mg/g of body weight). 2.2
Equipment
1. 5 mL sterile syringes. 2. 20 mL sterile syringe. 3. 70 μm cell strainer. 4. Diaper pad. 5. 27-gauge needle. 6. 1 mL syringe. 7. Winged infusion set. 8. 12-well cell culture plate. 9. 100 mm Petri dishes. 10. 2 mL microcentrifuge tube. 11. 1.7 mL microcentrifuge tube. 12. 96-well round bottom plate. 13. 12 × 75mm, 5 mL polystyrene round bottom tube. 14. 12 × 75mm, 5 mL polystyrene round bottom tube with 35 μm cell strainer cap. 15. Dissection board.
174
Sarah J. Miller et al.
Table 1 Antibody panel Antibody
Clone
CD11b APC
M1/70
F4/80 eFluor 450
BM8
Ly6G APC-Cy7
1A8
Live/Dead Zombie Aqua
n/a
CD45 PE
30-F11
CD43 BV605
S7
CD11c PE-Cy7
N418
MHCII Alexa Fluor 700
M5/114.15.2
CD206 FITC
C068C2
CD45 APC
30-F11
CD45 eFluor 450
30-F11
CD45 APC-Cy7
30-F11
CD45 BV605
30-F11
16. Surgical scissors. 17. Student Vannas scissors. 18. Rib-cage holding clamps. 19. Forceps. 20. Razor blades. 21. Cytek Aurora or similar flow cytometry analyzer and FlowJo Software. 2.3 Antibodies and Live/Dead Stain
1. Full antibody panel: CD11b APC (clone M1/70), CD11c PE-Cy7 (clone N418), MHCII Alexa Fluor 700 (clone M5/114.15.2), F4/80 eFluor 450 (clone BM8), Ly6G APC-Cy7 (clone 1A8), CD45 PE (clone 30-F11), CD43 BV605 (clone S7), CD206 FITC (C068C2). 2. Single color controls: CD45 PE, CD45 APC, CD45 PE-Cy7, CD45 Alexa Fluor 700, CD45 eFluor 450, CD45 APC-Cy7, CD45 BV605 (all clone 30-F11). 3. Live/Dead Zombie Aqua dye. See Table 1 for complete list of antibodies needed for protocol.
Analysis of Macrophages from the Kidney
3
175
Method Workflow at a glance: Day 1: Prepare harvest supplies, label tubes, and make stock solutions. 30 min. Day 2: Harvest tissues, isolate, and stain single cells with antibody panel (see Note 3), fix, store at 4 °C overnight. 6+ h. Day 3: Run samples on flow cytometer. 2+ h.
3.1 Isolation and Flow Cytometry Analysis of Kidney Macrophages 3.1.1
Day 1: Prep
1. Gather all tissue harvest supplies (forceps, scissors, rib-cage clamps, diaper pad, 20 mL syringe, winged infusion set, DPBS, dissection board). Dissection board should have diaper pad laid on top, with surgical tools, 20 mL syringe, infusion set, and DPBS placed nearby. 2. Label a 50 mL conical and 2.0 mL microcentrifuge tube for each kidney sample to be collected. Label a 50 mL conical for spleen, which will be used for single color controls (SCC). 3. Label 1.7 mL tubes for SCC, unstained cells, FMO (see Note 4), and set 1 antibody panel. 4. Label 12- and 96-well plates. 5. Prepare master stocks.
3.1.2
Day 2: Harvest
1. Prepare reagents before proceeding. 2. Add 3 mL RPMI to 12-well plate for each sample, then place on ice. Aliquot 1 mL digestion buffer into prelabeled 2 mL microcentrifuge tubes. 3. Anesthetize mice by injection with prepared Avertin working stock (0.5 mg/g of body weight). Measure mouse weight and/or length if applicable. Lay the mouse on its back in a supine position, with legs secured to dissection board by tape or pins for access to the abdomen. Sterilize the abdomen with 70% ethanol. 4. Just above the pelvis, lift the skin and make a small incision using scissors. Gently use scissors to cut up the midline to just beneath the jaw. Use forceps or gloved fingers to separate skin from underlying peritoneal membrane. 5. Carefully cut the peritoneal membrane up the midline to the sternum. Once thoracic cavity is exposed, cut along dorsal side of rib cage to allow the sternum to be lifted toward the jaw and expose the heart. 6. Fill the 20 mL syringe with room temperature DPBS, and attach winged infusion set. Insert needle into left ventricle, then clip the right atrium using scissors and begin slow perfusion. Each animal should be perfused with 10–15 mL DPBS at a rate of ~5 mL/minute.
176
Sarah J. Miller et al.
Fig. 1 Schematic depicting workflow of steps 11–16
7. Following perfusion, remove left kidney from abdominal cavity. Remove kidney capsule before taking kidney weight/length (if applicable), then place in appropriate well of the 12-well plate, on ice. Optional 10-min break point 8. Repeat steps 3–7 for all remaining mice; each mouse will take ~10–15 min to process. Collect the spleen from the final mouse harvested, and add to 12-well plate with kidneys. 9. After all mice have been harvested, use a razorblade (see Note 5) to mince tissue in a petri dish until no large chunks remain and tissue is a paste-like consistency. Carefully add as much of this paste as possible to the appropriately labeled digestion tube using the flat side of the blade. Keep on ice until all kidneys have been minced and are ready for digestion. 10. Place digestion tubes on a rocker at 37 °C for 30 min. 11. Steps 11–16 are depicted in Fig. 1. After 30 min, put digestion samples on ice followed by transferring the sample through a 70 μm cell strainer set inside a prelabeled 50 mL conical. Use the rubber plunger from a 5 mL syringe to “mash” as many cells as possible through the strainer, rinsing with RPMI periodically. A total of ~15 mL RPMI should be used per kidney. Once kidneys are filtered, filter the spleen by mashing the tissue directly on the strainer, again using ~15 mL RPMI to rinse. 12. Spin the cell suspension at 300 g for 5 min at 4 °C. 13. Decant supernatant, then resuspend each cell pellet in 5 mL ACK lysing buffer for 5 min at room temperature. Ensure each pellet is well suspended by vortexing briefly. 14. After 5 min lysis, add 10 mL RPMI, then remove a 10 μL aliquot and add to 10 μL of trypan blue for cell counting. Spin the cell suspension at 300 g for 5 min at 4 °C.
Analysis of Macrophages from the Kidney
177
15. Decant supernatant, then resuspend cells in Fc blocking buffer and place on ice for 30 min. Kidneys should be resuspended in 1 mL, while spleen in 1–2 mL depending on antibody panel used. Optional 30-min break point 16. Count cells, then add ~2 × 106 cells/sample to the labeled 96-well round bottom plate (see Note 6). Add 100 μL of spleen cells to each SCC well, plus one well for an unstained control. Begin making antibodies; final concentrations and volumes needed can be found in Table 2. Table 2 Single color control, FMO Live/Dead, and Set 1 antibody panel, with dilution factors and volume needed for each Single color controls CD45 APC
Final concentration -3
1×10
mg/mL
-3
Volume needed 100 μL
CD45 PE-Cy7
1 × 10
mg/mL
100 μL
CD45 Alexa Fluor 700
1 × 10-3 mg/mL
100 μL
CD45 eFluor 450
1 × 10-3 mg/mL
100 μL
-3
CD45 APC-Cy7
2 × 10
Live/dead zombie aqua
1:500
CD45 PE
1 × 10-3 mg/mL
CD45 FITC
mg/mL
100 μL 100 μL
2.5 × 10
-3
-3
mg/mL
100 μL 100 μL 100 μL
CD45 BV605
1 × 10
Set 1 antibodypanel
Final concentration
Volume needed
CD11b APC CD11c PE-Cy7 MHCII Alexa Fluor 700 F4/80 eFluor 450 Ly6G APC-Cy7 Live/dead zombie aqua CD45 PE CD206 FITC CD43 BV605
1 × 10-3 mg/mL 1 × 10-3 mg/mL 1 × 10-3 mg/mL 1 × 10-3 mg/mL 2 × 10-3 mg/mL 1:500 1 × 10-3 mg/mL 2.5 × 10-3 mg/mL 1 × 10-3 mg/mL
100 μL per sample (i.e. five mice = 500 μL total)
FMO live/dead
Final concentration
Volume needed
CD11b APC CD11b APC CD11c PE-Cy7 F4/80 eFluor 450 Ly6G APC-Cy7 CD45 PE CD206 FITC CD43 BV605
-3
mg/mL
1 × 10 mg/mL 1 × 10-3 mg/mL 1 × 10-3 mg/mL 1 × 10-3 mg/mL 2 × 10-3 mg/mL 1 × 10-3 mg/mL 2.5 × 10-3 mg/mL 1 × 10-3 mg/mL
100 μL
178
Sarah J. Miller et al.
17. Place 100 μL of spleen cells in a 1.7 mL microcentrifuge tube and incubate at 95 °C for 3 min. Add killed cells to Live/Dead SCC well. 18. After appropriate cell numbers have been added to the plate and blocked for 30 min on ice, spin at 300 g for 5 min at 4 °C. 19. Dump supernatant, then wash with 200 μL 1% BSA. Spin plate at 300 g for 5 min at 4 °C. Dump supernatant. 20. Add 100 μL of antibody panel to all sample wells and mix thoroughly by pipetting up and down. For single color controls, always use appropriate color and concentration of CD45 antibody to correspond with macrophage identifying antibodies. Wrap plate well in foil, and incubate 30 min at room temp. 21. After 30 min, spin the plate at 300 g for 5 min at 4 °C. 22. Dump supernatant, then wash with 200 μL 1% BSA. Spin plate at 300 g for 5 min at 4 °C. Dump supernatant (see Note 7). 23. Resuspend cells in 200 μL of 2% PFA and fix for 30 min on ice (see Note 8). 24. Spin plate at 300 g for 5 min at 4 °C. Dump supernatant. 25. Wash cells with 200 μL DPBS. Spin plate at 300 g for 5 min at 4 °C. Dump supernatant. 26. Resuspend in 200 μL DPBS, and store wrapped in foil at 4 °C overnight, or until ready to run on flow cytometer. Run samples within 48 h. 3.1.3 Day 3: Flow Analysis
1. Transfer cells to round bottom polystyrene tubes. 2. Filter samples into 5 mL tube with strainer cap just before running on flow cytometer. 3. Collect 2 × 104 events for unstained cells and each SSC. Collect 3 × 105 events for FMO Live/Dead, and 1 × 106 events for each sample run. 4. Export FCS files and analyze in FlowJo.
3.2 Expected Outcomes
The expected leukocyte yield from a whole kidney is approximately 3 × 106, with >95% viability as determined by flow cytometry. The suggested antibody panel above allows identification of neutrophils, Ly6Clo monocytes, Ly6Chi monocytes, and kidney resident macrophages. Expected amounts of respective cell types can be found in Table 3. Slight deviations will occur based on age and sex of the mice analyzed. Problems such as poor viability and/or low cell yield (see Note 9) as well as a poor staining (see Note 10) can occur.
Analysis of Macrophages from the Kidney
179
Table 3 Approximate cell numbers obtained from an average mouse kidney Cell type
Approx. % of CD45+ cells
Approx. number per kidney
Neutrophils
3
1.25 × 104
Ly6Clo monocytes
3
1.25 × 104
Ly6Chi monocytes
5
1.5 × 104
Kidney resident macrophages
30–40
1.0 × 105
3.3 Quantification and Statistical Analysis
4
Gating strategy for neutrophils, Ly6Clo monocytes, Ly6Chi monocytes, and kidney resident macrophages using provided antibody panel is shown in Fig. 2, with population identifying markers given in Table 4. Quantification of cell abundance can be done in FlowJo by determining the number of each immune cell population as a frequency of live single cells and the frequency of live, CD45+ cells. Total macrophage numbers can also be calculated using the cell counts from the trypan blue staining procedure.
Notes 1. Prepare master stocks on day 1. 2. Prepare on day 2 before beginning protocol. 3. Cells can be run on flow cytometer live immediately after antibody staining and washing if preferred. 4. FMO (fluorophore minus one) stains should be used for each antibody where there is not a clean positive population, and always for the Live/Dead stain. 5. If desired, a tissue homogenizer can also be used at this step. 6. Staining efficiency can decrease when more than 2 × 106 cells are stained per well. 7. If running cells on flow cytometer live, following the 1% BSA wash resuspend cells in 200 μL DPBS and proceed immediately to Day 3, Step 1. 8. Decay of single-color control antibody signal can occur if plate is not kept on ice. 9. Problem 1: Poor viability and/or low cell yield. Overincubation with either digestion buffer or ACK lysis buffer can affect cell viability. Ensure digestion tubes remain on ice until ready to begin the 30-min incubation. When digesting samples or lysing red blood cells, be sure to follow incubation times specified above. Low cell yield can be caused by several factors. Ensure digestion buffer is made fresh on the day of
180
Sarah J. Miller et al.
Fig. 2 Gating strategy using the given antibody panel to identify Ly6Chi monocytes, Ly6Clo monocytes, and kidney resident macrophages
Table 4 Cell surface markers used to identify neutrophils, Ly6Chi monocytes, Ly6Clo monocytes, and kidney resident macrophages Cell type
Identifying markers
Neutrophils
CD45+, Ly6G+
lo
Ly6C monocytes
CD45+, Ly6G-, F4/80low, CD11b+, Ly6Clow, CD43+
Ly6Chi monocytes
CD45+, Ly6G-, F4/80low, CD11b+, Ly6Chi
Kidney resident macrophages
CD45+, Ly6G-, F4/80hi, CD11b+
Analysis of Macrophages from the Kidney
181
harvest, and avoid freezing. When filtering digestion samples, rinse out digestion tube with fresh RPMI and add to 70 μm filter. Make sure the filter is well rinsed. 10. Problem 2: Poor staining. Antibodies can be susceptible to photo-bleaching, and should be kept shielded from light as much as possible during the protocol. Antibodies should be stored in the dark, and preparation of SCCs, FMOs, and set 1 antibody panel should be done as rapidly as possible in low light. For optimum results, be sure the plate is tightly wrapped in foil and protected from light as much as possible once the antibodies have been added. The dilutions given above are optimized for use on the above given cell populations, but should be taken as a starting point. Dilutions should be adjusted if needed to maximize the separation of positive and negative cell populations. References 1. Liu F, Dai S, Feng D, Qin Z, Peng X, Sakamuri S et al (2020) Distinct fate, dynamics and niches of renal macrophages of bone marrow or embryonic origins. Nat Commun 11(1): 2280. https://doi.org/10.1038/s41467020-16158-z 2. Alikhan MA, Ricardo SD (2013) Mononuclear phagocyte system in kidney disease and repair. Nephrology (Carlton) 18(2):81–91. https:// doi.org/10.1111/nep.12014 3. Li Z, Zimmerman KA, Yoder BK (2021) Resident macrophages in cystic kidney disease. Kidney360 2(1):167–175. https://doi.org/10. 34067/kid.0006052020 4. Munro DAD, Hughes J (2017) The origins and functions of tissue-resident macrophages in kidney development. Front Physiol 8:837. https://doi.org/10.3389/fphys.2017.00837 5. Aloria EJG, Song CJ, Li Z, Croyle MJ, Mrug M, Zimmerman KA et al (2021) Ly6chi infiltrating macrophages promote cyst progression in injured conditional Ift88 mice. h t t p s : // d o i . o r g / 1 0 . 3 4 0 6 7 / K I D . 0000882021 6. Peng X, Zhang J, Xiao Z, Dong Y, Du J (2015) CX3CL1-CX3CR1 interaction increases the population of Ly6C(-)CX3CR1(hi) macrophages contributing to unilateral ureteral obstruction-induced fibrosis. J Immunol 195(6):2797–2805. https://doi.org/10. 4049/jimmunol.1403209 7. Li YH, Zhang Y, Pan G, Xiang LX, Luo DC, Shao JZ (2022) Occurrences and functions of
Ly6C(hi) and Ly6C(lo) macrophages in health and disease. Front Immunol 13:901672. https://doi.org/10.3389/fimmu.2022. 901672 8. Lever JM, Hull TD, Boddu R, Pepin ME, Black LM, Adedoyin OO et al (2019) Resident macrophages reprogram toward a developmental state after acute kidney injury. JCI Insight 4(2). https://doi.org/10.1172/jci.insight. 125503 9. Viehmann SF, Bohner AMC, Kurts C, Brahler S (2018) The multifaceted role of the renal mononuclear phagocyte system. Cell Immunol 330:97–104. https://doi.org/10.1016/j. cellimm.2018.04.009 10. Lindstrom NO, McMahon JA, Guo J, Tran T, Guo Q, Rutledge E et al (2018) Conserved and divergent features of human and mouse kidney organogenesis. J Am Soc Nephrol 29(3): 785–805. https://doi.org/10.1681/ASN. 2017080887 11. Farndon SJ, Polanski K, Efremova M, Green K, Del Castillo Velasco-Herrera M, Guzzo C, et al.(2019) Spatiotemporal immune zonation of the human kidney. Science (New York, NY) 365(6460). doi: https://doi.org/10.1126/ science.aat5031 12. Choi JW, Shin BS (2018) Isoflurane decreases interleukin-2 production by increasing c-Cbl and Cbl-b expression in rat peripheral blood mononuclear cells. J Int Med Res 46(7): 2792–2802. https://doi.org/10.1177/ 0300060518770955
Chapter 13 Isolation and Flow Cytometry Analysis of Intestinal Macrophages Maria Francesca Viola and Guy Boeckxstaens Abstract The intestinal macrophage pool represents the largest population of macrophages present within the body. Nevertheless, flow cytometry analysis of intestinal macrophages remains challenging due to historical lack of consensus on surface markers, variations in sample preparation, and a certain capriciousness of the isolation procedure itself. Furthermore, recent studies have uncovered a hitherto unknown heterogeneity of intestinal macrophages, accompanied by a vast increase of subset-identifying surface markers. Here, the isolation procedure for intestinal tissue for flow cytometry analysis is laid out, with particular attention toward the procedures for isolated intestinal layers, and a trouble-shooting section with strategies to avoid common pitfalls and mistakes. Key words Macrophages, Intestinal macrophages, Flow cytometry, Intestine, Colon, Lamina propria, Muscularis externa
1
Introduction
1.1 Intestinal Macrophages: From Monocytes to Macrophages
Intestinal macrophages represent a highly heterogenous, complex, and misunderstood population of phagocytes that reside at the host-lumen interface of the largest mucosal surface of the body. Indeed, when the scientific community began to uncover that most tissue-resident macrophage populations actually derive from embryonically seeded cells, and not via constant replenishment of monocytes, the intestine was heralded as the exception to the rule, as studies demonstrated the complete replacement of embryonically seeded cells by monocytes around the time of weaning [1]. Moreover, it was widely believed that intestinal macrophages underwent a rapid turnover, and were replaced as frequently as every 3–5 weeks [2–4]. As fate-mapping technologies evolved and became more sophisticated, however, it was demonstrated that some macrophages in the gut in adulthood derive from embryonically seeded cells that are able to self-maintain within the tissue
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_13, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
183
184
Maria Francesca Viola and Guy Boeckxstaens
[5, 6]. As such, the intestine harbors an ever-evolving mix of embryonically seeded and bone marrow-derived macrophages that carry out a plethora of functions at the host-lumen interface. While this complexity does pose a hurdle to the analysis of intestinal macrophages, the constant input of circulating monocytes to the mature macrophage pool does offer a “barometer” of the local intestinal milieu, especially during inflammatory conditions. Indeed, the differentiation trajectory from circulating monocyte to mature intestinal macrophage, the so-called waterfall, is well characterized [1, 2]. As such, it is possible to evaluate the influx of monocytes, their differentiation, and the survival of their mature counterparts. This type of assessment is of interest, as the loss of resident macrophages, influx of monocytes, and stasis of differentiation have been reported in inflammation, such as in colitis [3, 7, 8]. A full description of how to perform such an analysis is provided in the analysis section of this chapter. 1.2 Heterogeneity of Intestinal Macrophages
2 2.1
Once fully matured, monocyte-derived macrophages contribute to the pool of mature macrophages, to which also self-maintaining macrophages of embryonic origin contribute. This mature intestinal macrophage pool is characterized by an additional level of complexity due to its heterogeneity. Indeed, mature intestinal macrophages can be found throughout the entire length of the gastrointestinal tract and are located throughout its numerous layers, from beneath the epithelium all the way to the serosa. Macrophages occupying distinct layers differ from their neighbors in terms of both transcriptome and function, and it has been shown that macrophages located in the lamina propria are distinct, both transcriptionally and functionally, from their counterparts in the muscularis externa, and in lymphoid tissues such as Peyer’s patches [6, 9]. To add further complexity, within each tissue, further specialized subsets have been identified and described, with little common consensus and overlap between studies. Finally, not all the identified subsets are characterized by discriminating surface markers, which would aid their detection via flow cytometry or confocal microscopy. We here provide a list, up-to-date at the time of publishing, of subset-specific surface markers that can be used to further discriminate between mature macrophage subpopulations in the intestinal macrophage pool.
Materials Equipment
1. Tools for dissection (forceps and scissors). 2. 10 or 20 mL syringes. 3. 1.5 mL microcentrifuge tubes. 4. 15 mL and 50 mL falcons.
Intestinal Macrophages
185
5. Gavage needles or suitable alternative. 6. Non-sterile petri dishes. 7. Black contrast 93 mm sylgard plates with pins.1 8. Stereoscopic microscope with light source.1 9. Heating chamber with agitation. 10. Stainless steel sieve. 11. Centrifuge with buckets for 15 and 50 mL falcons, FACS tubes, or 96-well plates. 12. 70 μm cell strainers. 13. 5 mL FACS tubes or 96-well plates for staining. 14. Conjugated antibodies of interest. 15. Aluminum foil. 16. Live viability dye (DAPI, 7AAD, PI) or fixable viability dye. 17. FACS tubes with cell strainer or nitrocellulose mesh for cell filtering. Optional: If performing cell purification via Percoll gradient: 18. Syringe fitted with a wide-bore needle. 2.2
2.2.1
Buffers
Common Buffers
Buffers are separated into common buffers, which are always required, and layer-specific buffers. If analyzing the tissue as a whole, without dissection of layers, please prepare buffers in the “Lamina Propria/Whole Tissue” section. If analyzing isolated intestinal layers, both muscularis externa and lamina propria, please prepare buffers of both the “Muscularis Externa” and “Lamina Propria/Whole Tissue” section. Finally, if only one of the two layers is being analyzed, please prepare only the buffers from the relevant section, in addition to the common buffers. It is recommended to check the pH of all buffers prior to beginning the protocol; optimal pH is 7.3–7.4. Whenever possible, prepare buffers fresh, and always store at 4 °C until use. 1. Tissue dissection buffer: 3% FCS and 1.8% HEPES in Hanks′ Balanced Salt solution (with Ca2+ and Mg2+). Prepare 50 mL per sample. It can be stored for 3–4 weeks at 4 °C. 2. FACS buffer: 2% FCS and 2 mM EDTA in PBS. It can be stored for 3–4 weeks at 4 °C.
2.2.2
1
Muscularis Externa
1. Muscularis externa digestion buffer: 3% FCS, 300 mM HEPES, 5 U/mL DNAse I, and 400 U/mL collagenase IV in RPMI 1640. Prepare 9 mL per sample in a 50 mL tube;
If performing dissection of muscularis externa and lamina propria
186
Maria Francesca Viola and Guy Boeckxstaens
pre-warm to 37 °C. Buffer should be prepared as fresh as possible, however, can be stored at 4 °C for up to 2 weeks without enzymes (DNAse I and collagenase IV). Enzymes should be added immediately prior to use. 2.2.3 Lamina Propria/ Whole Tissue (See Note 1)
1. Tissue collection buffer: 1% FCS in HBSS (see Note 2). Prepare at least 20 mL per sample; it can be stored for 3–4 weeks at 4 °C. 2. Epithelial separation buffer: 1% FCS, 100 μg/mL penicillin/ streptomycin, 1 mM EDTA, and 1 mM DTT in HBSS (see Note 2). Prepare 20 mL per sample in a 50 mL tube, pre-warm to 37 °C, and add DTT immediately prior to use. 3. Epithelial wash buffer: 1% FCS, 100 μg/mL penicillin/streptomycin, and 1 mM EDTA in HBSS (see Note 2). Prepare 20 mL per sample in a 50 mL tube; pre-warm to 37 °C. 4. Lamina propria digestion buffer: 5% FCS, 100 μg/mL penicil200 mM HEPES, 50 mM lin/streptomycin, β-mercaptoethanol, 5 U/mL DNAse I, 0.85 mg/mL collagenase V, 1.25 mg/mL collagenase D,2 and 1 mg/mL dispase2 in RPMI 1640. Prepare 20 mL per sample in a 50 mL tube; pre-warm to 37 °C, and only add enzymes immediately prior to use. Optional: If performing Percoll enrichment: 5. Percoll 100% solution: Prepare a solution of Percoll reagent with 10% PBS 10X and 6.55 mM HCl. Prepare 5 mL per sample immediately prior to use. 6. Percoll 44% solution: Prepare 44% solution of Percoll 100% by diluting in RPMI. Prepare 5 mL per sample (plus some excess) immediately prior to use. 7. Percoll 67% solution: Prepare 67% solution of Percoll 100% by diluting in RPMI. Prepare 3 mL per sample (plus some excess) immediately prior to use.
3
Methods
3.1 Sample Preparation
Throughout the sample collection, tissue and buffers should always be kept on ice to avoid tissue degradation. Intestinal tissue can be analyzed as a whole or separated into the muscularis externa and the remaining tissue, containing mucosa and submucosa, indicated here as lamina propria (see Note 3). Digestion of the whole tissue and the isolated lamina propria follow the same protocol, however,
2 Only for digestion of whole tissue. If digesting lamina propria alone (separated from muscularis externa), these enzymes may be omitted.
Intestinal Macrophages
187
differ in the use of enzymes in the tissue digestion. Digestion of the muscularis externa requires a separate protocol, as detailed in the sections that follow. 1. Euthanize the mouse according to protocol. Transcardial perfusion using ice-cold PBS is optional but recommended. Put euthanized mouse on its back; spray abdominal area with 70% ethanol. Lift the skin of the lower abdomen with tweezers, and make a V-shaped incision with scissors; cut away skin to expose abdominal cavity. Gently remove the intestine by cutting the esophagus at the entry of the stomach and separating the tissue from the rear of the abdominal cavity by cutting through mesenteric tissue, finally performing a cut at the end of the colon. 2. Gently unravel the intestine by cutting through the mesenteric tissue. This step needs to be performed without excessive force, as ripping the mesentery will also damage the muscularis externa. Isolate the portion of the intestine of interest, and place it in the dissection buffer (Fig. 1). 3. Fill a syringe fitted to a soft gavage needle with tissue dissection buffer. Then, insert the soft gavage needle into the opening of the tissue, and hold the tip in place using forceps. Gently flush the tissue until clean of luminal content (see Note 4). 4. If analyzing the whole tissue without layer separation: Cut off any remaining mesentery using scissors, and open the tissue longitudinally along the mesenteric border. Optional: Remove Peyer’s patches and analyze separately, if desired. Use tissue dissection buffer to further clean the tissue and place in a falcon containing 20 mL of lamina propria collection buffer. Skip the remaining steps of this section and directly proceed to “Tissue Digestion: Lamina Propria/Whole Tissue” section. If analyzing with layer separation: Proceed to dissection of the muscularis externa by placing the flushed intestinal tissue in a tissue dissection buffer in a sylgard plate, and perform further dissection of 5–7 cm segments using a stereoscopic microscope with a light source. The steps detailed here are further detailed in Fig. 2. 5. Pin the tissue at the extremities so that it is slightly stretched and the mesenteric border is in a visible straight line (i.e., the intestine is not twisted) (Fig. 2a). Using small ball-tip scissors, carefully remove any remaining mesentery (Fig. 2b), and then cut open the tissue along the mesenteric border (Fig. 2c). 6. Once the tissue has been fully opened (Fig. 2d), turn the tissue with the luminal face down, and pin the tissue ends so that it is a flat surface with the serosal side upward (Fig. 2e). Using fine forceps with an angled tip, gently begin scraping the surface of the tissue at one end (Fig. 2f). The muscularis externa will
188
Maria Francesca Viola and Guy Boeckxstaens
Stomach Duodenum Jejunum
IIeum Caecum Colon
Rectum
Fig. 1 Schematic overview of the intestine. Created with BioRender.com. The segments of the small intestine can be identified as follows; the portion between the stomach and the first Peyer’s patch corresponds to the duodenum. The remaining length of the small intestine (until the caecum) can then be divided into two equal pieces, where the segment after the duodenum corresponds to the jejunum and the other, closer to the caecum, corresponds to the ileum
appear as a thin, slightly opaque layer that can be separated from the underlying tissue throughout its length (Fig. 2g). Once the edge of the muscularis externa has been separated from the underlying tissue, continue separating by gentle scraping until the end of the tissue (Fig. 2h, i). Place isolated muscularis externa in a microcentrifuge tube containing 1 mL of muscularis digestion buffer without collagenase, and place on ice. Proceed to “Tissue Digestion: Muscularis Externa.” Optional: Remove Peyer’s patches from remaining lamina propria, and analyze separately, if desired. Taking care to remove portions of tissue where the muscularis externa was not removed, place dissected lamina propria in falcon containing 20 mL tissue collection buffer, and proceed to “Tissue Digestion: Lamina Propria/Whole Tissue.”
Intestinal Macrophages
189
Fig. 2 Step-by-step sequence of how to separate the muscularis externa from the underlying lamina propria. Detailed explanation is provided in the “Sample Preparation” section of this chapter. Tissue was obtained from an adult C57Bl6/J mouse, and the tissue shown in the images corresponds to the ileum 3.2
Tissue Digestion
3.2.1 Muscularis Externa
The following protocol is intended for the digestion of the muscularis externa once it has been dissected from the lamina propria. 1. Mince the tissue using a fine scissor, and then transfer the minced tissue to a 50 mL falcon (it is recommended to cut the pipette tip to facilitate this step). 2. Add the collagenase and the DNAse to the preheated muscularis digestion buffer, and agitate until the enzymes are fully dissolved. Add 8 mL of the muscularis digestion buffer with collagenase to each minced sample. 3. Incubate for 30 min at 37 °C in agitation at 150–300 rpm. Before proceeding to filter, it is recommended to give the samples a vigorous shake. 4. Prepare a second set of 50 mL falcons mounted with 70 μm cell strainers. Filter the samples (it is recommended to use pipettes to favor mechanical dissociation, rather than pouring), and subsequently wash the filters with 10 mL of FACS buffer (see Note 5). 5. Centrifuge the samples at 350 × g for 8–10 min at 4 °C.
190
Maria Francesca Viola and Guy Boeckxstaens
6. Discard the supernatant, and resuspend the pellet in 1 mL of FACS buffer. Optional: Transfer samples to 15 mL falcons to obtain pellets that are more visible. Add a further 9 mL of FACS buffer. 7. Centrifuge the samples at 350 × g for 8–10 min at 4 °C. 8. Discard the supernatant, resuspend the pellet in an appropriate amount of FACS buffer, and proceed to cell counting and extracellular staining. 3.2.2 Lamina Propria/ Whole Tissue
1. Pour the falcons over a metal sieve to collect the tissue. Cut the tissue into pieces of 1–2 cm long with scissors. 2. Transfer the tissue of each mouse to a 50 mL falcon with 25 mL of epithelial separation medium already warmed at 37 °C. Incubate then samples for 8–10 min at 37 °C in agitation at 150–300 rpm to remove epithelial cells. 3. Prepare 50 mL falcons containing 20 mL of preheated lamina propria epithelial wash buffer. 4. Pour the sample over the metallic mesh filter, collect the tissue, and transfer it to the falcons containing preheated lamina propria epithelial wash buffer. Shake first and incubate 10–12 min at 37 °C in agitation at 150–300 rpm. During this step, epithelial cells will separate from the tissue, conferring a cloudy appearance to the buffer (see Note 6). 5. Prepare petri dishes with wash medium, one for every sample. In addition, prepare one microcentrifuge tube containing 1 mL of lamina propria digestion buffer without enzymes per sample. Add enzymes to remaining lamina propria digestion medium, and preheat to 37 °C. 6. Pour the sample over a metallic mesh filter, collect the tissue, and place in petri dish containing wash buffer. Gently agitate the tissue in the buffer to wash and remove any remaining epithelial cells. The tissue should now appear slightly translucent and should not be sticky (see Note 5). 7. Transfer washed tissue into microcentrifuge tube containing lamina propria digestion buffer without enzymes. Mince tissue using scissors. Transfer minced tissue to a 50 mL falcon, and add 19 mL of lamina propria digestion buffer with enzymes. 8. Incubate for 18–20 min at 37 °C in agitation at 150–300 rpm. Before proceeding to filtering, it is recommended to give the samples a vigorous shake. 9. Prepare a set of 50 mL falcons mounted with 70 μm cell strainers. Filter the samples (it is recommended to use pipettes to favor mechanical dissociation, rather than pouring), and subsequently wash the filters with 10 mL of FACS buffer.
Intestinal Macrophages
191
10. Centrifuge the samples at 350 × g for 8–10 min at 4 °C. 11. Discard the supernatant (see Note 7), and resuspend the pellet in 1 mL of FACS buffer. Optional: Transfer samples to 15 mL falcons to obtain pellets that are more visible. Add further 9 mL of FACS buffer. 12. Centrifuge the samples at 350 × g for 8–10 min at 4 °C. 13. Discard the supernatant. If not performing immune cell enrichment using Percoll gradient, resuspend the pellet in appropriate amount of FACS buffer, and proceed to cell counting and extracellular staining. Optional: Immune cell enrichment with Percoll gradient 14. Resuspend cells in 5 mL 44% Percoll solution, and transfer to a 15 mL falcon tube. Using a syringe fitted with a wide-bore needle, slowly layer 3 mL of 67% Percoll solution to the bottom of the 15 mL falcon tube, under the cell suspension. 15. Centrifuge at 820 × g for 20 min at RT without brake. 16. Lamina propria immune cells will collect at the 44%/67% interface. Discard the debris floating at the top the top layer without disturbing the interface. Collect the mononuclear lamina propria cells from the interface, and transfer to a fresh 15 mL falcon tube. 17. Fill up to 10 mL with FACS buffer, and centrifuge at 350 × g for 8–10 min at 4 °C. 18. Discard the supernatant, resuspend the pellet in appropriate amount of FACS buffer, and proceed to cell counting and extracellular staining. 3.3 Antibody Staining
1. Cells can be stained in 5 mL FACS tubes or round-bottom 96-well plates according to preference. Transfer an appropriate amount of cells (see Note 8) to either FACS tubes or a 96-well plate, and centrifuge at 500 g for 3–5 min at 4 °C. Discard supernatant. 2. Incubate with anti-CD16/32 (Fc block) diluted 1/200 in FACS buffer for 12 min at 4 °C. 3. Centrifuge at 500 × g for 3–5 min at 4 °C. Discard supernatant. 4. Optional: Incubate with biotin-conjugated primary antibodies for 20 min at 4 °C protected from light (see Table 1). 5. Centrifuge at 500 × g for 3–5 min at 4 °C. Discard supernatant. 6. Incubate with antibody mix prepared in FACS buffer for 20 min at 4 C protected from light (see Note 9, Table 2). 7. Centrifuge at 500 × g for 3–5 min at 4 °C. Discard supernatant. 8. If using fixable viability dye: Incubate with fixable viability dye diluted in PBS for 15–20 min at 4 °C protected from light (see
192
Maria Francesca Viola and Guy Boeckxstaens
Table 1 List of biotin-conjugated primary antibodies used for staining shown in Fig. 4 Conjugate
Antigen
Dilution
Clone
Biotin
Nkp46
2.5 μg/mL
29A1.4
Biotin
CD19
1.25 μg/mL
6D5
Biotin
TCRß
1.25 μg/mL
H57–597
Table 2 List of extracellular antibodies used for staining shown in Figs. 3 and 4 Conjugate
Antigen
Dilution
Clone
BUV-496
Strep
0.25 μg/mL
N/A
BUV-805
CD45
0.5 μg/mL
30-F11
FITC
CD64
1.25 μg/mL
X54-5/7.1
PE-Cy7
CD11b
0.5 μg/mL
M1/70
APC
Cx3cr1
1 μg/mL
SA011F11
AF-700
MHCII
1.25 μg/mL
M5/114.15.2
APC-Cy7
Ly6C
2.5 μg/mL
HK1.4
Note 10). Live cell viability dyes can be added immediately prior to acquisition (see Note 11). 9. Centrifuge at 500 × g for 3–5 min at 4 °C. Discard supernatant. 10. Wash cells with FACS buffer. Centrifuge at 500 × g for 3–5 min at 4 °C. Discard supernatant. 11. Resuspend in FACS buffer, and proceed to either acquisition at the cytometer. 3.4 Identification of Intestinal Macrophages via Flow Cytometry
A particular challenge when referring to the literature is the lack of consensus regarding the expression of surface markers by intestinal macrophages. Indeed, up until 2012, there was a lack of understanding regarding discriminating markers for intestinal macrophages and dendritic cells, until the identification of CD64 as a reliable marker for mature macrophages [2]. For this reason, it is recommended to verify the gating strategy used by authors prior to interpretation of data present in the literature. In general, mature intestinal macrophages can be unequivocally identified as a Cd11bhi CD64hi population within the immune cell population (see Table 3). However, different considerations should be taken when analyzing the muscularis externa separately from the remaining tissue, and we have therefor split this chapter into two
Intestinal Macrophages
193
Table 3 Essential cell surface markers for the identification of mature intestinal macrophages in both the muscularis externa and the lamina propria Marker
Description
CD45
For identification of immune cells
CD11b
For identification of myeloid cells
CD64
For identification of mature macrophages
sections, one discussing the muscularis externa specifically and one discussing the remaining tissue. It should be noted that the immune cell compartment in the muscularis externa is rather small, and cannot as yet be distinguished from that of the lamina propria when analyzing the tissue as a whole. When analyzing the whole gut, without the separation of the muscularis externa, one is therefore essentially characterizing the macrophages populating the lamina propria, and the contribution from the muscularis externa will be minimal and very difficult to detect. 3.4.1
Muscularis Externa
In the muscularis externa macrophages represent up to 60% of the immune cell population. After doublet exclusion and identification of live cells, the CD45+ population can be easily identified (Fig. 3). Within this CD45+ populations, macrophages can be easily identified as the CD64+CD11b+ population, without the need of further negative lineage gating (Fig. 3). These cells are CX3CR1+ and can be further subsetted using the markers listed in Table 4.
3.4.2
Lamina Propria
In the case of whole tissue analysis, or analysis of the lamina propria, the inclusion of a cocktail of lineage markers (e.g., CD3, CD19, NKp46) is recommended. Therefore, after identification of cells based on FSC/SSC gating, doublet exclusion, and the exclusion of dead cells, the CD45+ population can be easily identified (Fig. 4). It should be noted that if Percoll enrichment was performed, the CD45+ population will represent the majority of the recovered cells. Next, it is recommended to exclude cells expressing lineage markers and gate on cells expressing CD11B (Lin-CD11B +). At this point, if the main interest is in the mature macrophage pool, it is possible to directly identify CD64+CX3CR1+ macrophages. However, it can be interesting to visualize the monocyte to macrophage “waterfall”; to do this, the expression of LY6C and MHCII is shown within the CD11B+. In this plot, monocytes (LY6C+MHCII-) and the monocyte to macrophages intermediates (LY6CintMHCIIint and LY6C-MHCII+) can be identified (Fig. 4). Finally, within the LY6C-MHCII+ population, mature macrophages (CX3CR1+CD64+) can be identified. Further subset discrimination can then be performed using the markers listed in Table 2.
Maria Francesca Viola and Guy Boeckxstaens
Viability
FSC-H
SSC-A
194
FSC-A
FSC-A
CD45
CD11B
FSC-A
FSC-A
CD64
Fig. 3 Representative gating strategy to identify macrophages in the muscularis externa. Sample was obtained from an adult naı¨ve C57Bl6/J mouse Table 4 Additional markers for intestinal macrophages that may be included for identification of specific subsets Marker
Description
CX3CR1
Highly expressed by intestinal macrophages. Not subset specific
F4/80
Highly expressed by intestinal macrophages but also by dendritic cells [10]
MHCII
Expressed by intestinal macrophages but also by antigen-presenting cells. Can be used to define monocyte-to-macrophage differentiation [3]
CD206
Upregulated by mature intestinal macrophages [4]
TIMD4
Upregulated by long-lived, resident intestinal macrophages [5]
CD4
Upregulated by long-lived, resident intestinal macrophages [5]
CD169
Identifies a subset located in proximity to intestinal epithelial crypts which plays a specific role in the recruitment of monocytes during colitis [11]
F11r/ CD321
Identifies a subset of macrophages located in the muscularis externa, in close proximity to the enteric nervous system, involved in the maintenance of enteric neurons [12]
LYVE1
Identifies a subset of macrophages located in close proximity to blood vessels, which are abundant in the serosal layer sheathing the intestine [12, 13]
LY6C
Expressed by monocytes and downregulated by mature macrophages. Can be used to define monocyte-to-macrophage differentiation [3]
FSC-A
Lineage
FSC - A
FSC-A
CD64
LY6C
CD11B
FSC-A
195
CD45
Viability
FSC-H
SSC-A
Intestinal Macrophages
MHCII
CX3CR1
Fig. 4 Representative gating strategy to identify macrophages in the lamina propria. Sample was obtained from an adult naı¨ve C57Bl6/J mouse. Lineage cocktail used here includes TCRß, CD19, and NKp46
4
Notes 1. Buffer volumes indicated are suitable for approximately 10 cm of tissue. If larger portions of the intestine are required for analysis, it is recommended to either increase buffer volume or divide the tissue over multiple falcons containing 20 mL of buffer. 2. The buffers used in this digestion protocol can be prepared with HBSS containing phenol red as a pH indicator. The advantage of a real-time pH indicator is that it allows to monitor buffer pH; however, the use of phenol red may cause increased autofluorescence of the sample. The buffers should thus be pink-red in color and should not change noticeably. If the buffers turn yellow during any step of this protocol, cell viability will be greatly compromised. Ensure all buffers have the correct pH prior to starting. 3. When choosing whether to analyze the intestine as a whole or separated into layers, it is useful to carefully consider whether the muscularis externa macrophage compartment is of interest to the experiment. As this compartment is very small compared to the vast lamina propria macrophage compartment, the relative contribution of the muscularis externa to the data obtained when analyzing the tissue as a whole will be minimal. Furthermore, as yet surface markers that allow for the discrimination of lamina propria and muscularis externa macrophages have as yet not been identified. Therefore, if the muscularis
196
Maria Francesca Viola and Guy Boeckxstaens
externa macrophage compartment is of interest in the frame of the experiment, the layers should be separated during the isolation, and the compartments should be analyzed separately. If the muscularis externa is not of interest, it may be more convenient to analyze the tissue without layer separation, as this greatly reduces sample preparation time, thereby improving cell viability and the quality of the results obtained. 4. The intestine can frequently get clogged during flushing, preventing complete removal of all debris. Cutting the tissue into segments of around 5–7 cm can aid the process, and frequently it also helps to flush from both proximal and distal end of the tissue. 5. Large debris that clog the cell strainer should not be present after digestion. To avoid large debris, ensure that tissue is finely minced prior to digestion and potentially increase sample agitation and digestion time (the latter only if strictly necessary). It should be noted that contaminating mesenteric fat will not be digested efficiently using these enzymes, so it is recommended to carefully remove all mesentery during sample preparation. 6. During incubation with epithelial cell removal buffer, and the subsequent epithelial cell wash buffer, epithelial cells should detach from the tissue, and the buffers should appear cloudy (this is especially visible after the epithelial cell wash step). The remaining tissue should also appear translucent. A clear buffer and opaque tissue that sticks to itself is indicative of insufficient epithelial cell removal, so incubation with these buffers can be extended, and sample agitation may be increased. 7. A pellet that is loose and not adhering to the bottom of the falcon after centrifugation is indicative of low cell viability. To avoid this issue, ensure all buffers are at the correct pH prior to starting, that the tissue has been properly cleaned from fecal material, and that the ratio of tissue to buffer is not exceeded (see Note 1). 8. A suggested appropriate amount of cells would be to use all isolated cells for the muscularis externa. With regard to the lamina propria, if no Percoll enrichment was performed, around 2 × 106 – 6 × 106 cells are required. If Percoll enrichment was performed, it is advisable to use all recovered cells. 9. Due to high levels of autofluorescence in the gut, it is recommended to avoid channels with high levels of autofluorescence if acquiring using a conventional flow cytometer. For example, BV510 and BUV563 have high levels of autofluorescence and should not be used during panel design for conventional flow cytometry.
Intestinal Macrophages
197
10. To shorten protocol length, it is possible to stain using fixable viability dyes to determine cell viability, and extracellular markers, simultaneously. To do this, it is necessary to dilute extracellular antibodies and fixable viability dye in PBS rather than in FACS buffer. 11. When preparing cells for FACS-sorting (rather than flow cytometry), it is always recommended to use live viability dyes to ensure viability of sorted cells. References 1. Bain CC et al (2014) Constant replenishment from circulating monocytes maintains the macrophage pool in the intestine of adult mice. Nat Immunol 15(10):929–937. https://doi.org/ 10.1038/ni.2967 2. Tamoutounour S et al (2012) CD64 distinguishes macrophages from dendritic cells in the gut and reveals the Th1-inducing role of mesenteric lymph node macrophages during colitis. Eur J Immunol 42(12):3150–3166. https://doi.org/10.1002/eji.201242847 3. Bain CC et al (2013) Resident and pro-inflammatory macrophages in the colon represent alternative context-dependent fates of the same Ly6Chi monocyte precursors. Mucosal Immunol 6(3):498–510. https:// doi.org/10.1038/mi.2012.89 4. Schridde A et al (2017) Tissue-specific differentiation of colonic macrophages requires TGFβ receptor-mediated signaling. Mucosal Immunol 10(6):1387–1399. https://doi. org/10.1038/mi.2016.142 5. Shaw TN et al (2018) Tissue-resident macrophages in the intestine are long lived and defined by Tim-4 and CD4 expression. J Exp Med:1–12. https://doi.org/10.1084/ jem.20180019 6. De Schepper S et al (2018) Self-maintaining gut macrophages are essential for intestinal homeostasis. Cell 175(2):400–415.e13. https://doi.org/10.1016/j.cell.2018.07.048 7. Grainger JR et al (2013) Inflammatory monocytes regulate pathologic responses to commensals during acute gastrointestinal
infection. Nat Med. https://doi.org/10. 1038/nm.3189 8. Zigmond E et al (2012) Ly6Chi monocytes in the inflamed colon give rise to proinflammatory effector cells and migratory antigenpresenting cells. Immunity 37(6):1076–1090. https://doi.org/10.1016/j.immuni.2012. 08.026 9. Gabanyi I, Muller PA, Feighery L, Oliveira TY, Costa-Pinto FA, Mucida D (2016) Neuroimmune interactions drive tissue programming in intestinal macrophages. Cell 164(3): 378–391. https://doi.org/10.1016/j.cell. 2015.12.023 10. Geissmann F, Gordon S, Hume DA, Mowat AM, Randolph GJ (2010) Unravelling mononuclear phagocyte heterogeneity. Nat Rev Immunol 10(6):453–460. https://doi.org/ 10.1038/nri2784 11. Asano K et al (2015) Intestinal CD169 + macrophages initiate mucosal inflammation by secreting CCL8 that recruits inflammatory monocytes. Nat Commun 6(44):1–7. https:// doi.org/10.1038/ncomms8802 12. Viola MF et al (2022) Neuro-immune crosstalk in the enteric nervous system from early postnatal development to adulthood. bioRxiv, p. 2022.05.12.491517. https://doi.org/10. 1101/2022.05.12.491517 13. Lim HY et al (2018) Hyaluronan receptor LYVE-1-expressing macrophages maintain arterial tone through Hyaluronan-mediated regulation of smooth muscle cell collagen. Immunity 49(2):326–341.e7. https://doi. org/10.1016/j.immuni.2018.06.008
Chapter 14 Isolation and Characterization of Testis Macrophages Using Flow Cytometry Myriam Sekias, Myriam Baratin, and Marc Bajenoff Abstract Testis-resident macrophages are first responders of the innate immune system against pathogens. They also exert day-to-day functions that are poorly understood. To study testis macrophages, several techniques are used, among which we can find flow cytometry. Flow cytometry is a powerful tool that enables analysis of macrophages at a cellular as well as population level. To analyze testis macrophages using flow cytometry, a specific tissue processing is necessary to extract them. In this protocol, we explain how to extract and analyze the distinct macrophage populations. Key words Testis macrophage, Flow cytometry, Cell isolation, Testis digestion
1
Introduction Macrophages are innate immune cells that are distributed across all tissues [1]. They are professional phagocytic cells that constitute the first line of defense against pathogens [2]. Macrophages also ensure homeostatic functions [3], for instance, mammary gland macrophages are important to maintain extracellular matrix turnover [4], and red pulp macrophages in the spleen are required to recycle iron [5]. Within the tissue, several subtypes of macrophages can be found, all expressing different combination of markers. In the mouse testis, we can discriminate two major macrophage populations, one expressing MHCII and the second one expressing CD206 [6]. In order to characterize those populations, several techniques are used, such as flow cytometry. Over the last decade, flow cytometry has become a powerful tool to explore and characterize protein expression at the cellular level. In this technique, fluorescently labeled antibodies are used to target selected markers found on cells of interest [7]. These antibodies are detected when cells flow through light sources (lasers). Flow cytometry can also discriminate cells based on their size and
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_14, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
199
200
Myriam Sekias et al.
granularity by their light scattering properties. These photonic signals are then transformed into digital data and can be analyzed using several softwares, such as Diva, FlowJo, and Cytobank. When multiplexing fluorescent antibodies into one panel, fluorescence spillover can occur. This then needs to be corrected [8]. This is possible by recording single-stained tubes (cells or beads), and then a compensation matrix can be calculated using the flow cytometer software. If the compensation values calculated in the first place are not optimal, further optimization of the matrix is possible during data analysis. Flow cytometry data can be displayed in a two-dimension dot plot, where a gating strategy is implemented to identify and characterize cells of interest. Macrophages are cells that are hard to extract from tissues; therefore, an enzymatic digestion is necessary to isolate them. In the following protocol, we will describe how to isolate testis macrophages for flow cytometry analysis.
2 2.1
Materials Equipment
1. Tweezers and scissors to dissect mouse tissue. 2. Heating and shaking dry bath. 3. Vortexer. 4. Scale for mg measurements. 5. Flow cytometer. 6. Class II laminar flow hood. 7. Chemical hood (if samples fixation is performed).
2.2
Consumables
1. Nylon mesh, 70 μm. 2. FACS tubes. 3. Plastic container of your choice for organ harvesting (12 or 24 wells plate). 4. 1.5 mL microcentrifuge tubes. 5. Petri dishes.
2.3 Workflow Solutions 2.3.1 Stock Solutions to Be Prepared in Advance
1. 1× PBS: commercially available. 2. 70% ethanol: Dilute 70 mL of absolute ethanol in 30 mL of sterile water. 3. FACS buffer: Under a sterile hood, prepare 1 L of FACS buffer (see Note 1). In sterile 1 L bottle, add 100 mL of sterile 10× PBS. Add 4 mL of sterile 0.5 M EDTA (final concentration 2 mM). Add 20 mL of fetal bovine serum (FBS, 2% final concentration) (see Note 2). Complete with 876 mL of MilliQ water. Filter FACS buffer using a threaded bottle top filter (0.22 μm) (see Note 3).
Testis Macrophage Analysis Using Flow Cytometry
201
4. DNAseI stock: Prepare DNAseI for a stock dilution of 10 mg/ mL in sterile 1× PBS or RPMI (see Note 2). Aliquot stock into 0.5 mL tubes (300 μL per tube) to avoid constant defrosting of the stock. Keep at -20 °C for up to 6 months. 2.3.2 Freshly Prepared Solutions
1. Digestion mix: The digestion mix amount prepared in this section is enough for 20 samples. On the day of the experiment, prepare 3 mg/mL of collagenase D (60 mg for 20 mL of RPMI) and 0.8 mg/mL of dispase II (16 mg for 20 mL RPMI). Add 200 μL of DNaseI stock (final dilution 0.1 mg/ mL). Add 400 μL of FCS (see Note 3). Complete with 19.4 mL of sterile RPMI. Vortex for 5 s. Transfer the necessary volume for the experiment into a new tube (1 mL per testis), and keep on ice. Aliquot the rest into 15 mL tubes and store them at 20 °C (see Note 4). 2. Viability buffer: Prepare this buffer up to 30 min before using it. One hundred μL of viability buffer is needed per sample. Calculate necessary amount of 1× PBS according to sample number, and add 20% to final sample number to compensate for lost volume. Add fixable viability dye of your choice according to the manufacturer’s instruction. Vortex for 5 s. 3. Staining buffer (see Note 5): 100 μL of staining buffer is required per sample. Calculate necessary amount of FACS buffer according to sample number (see Note 6), and add 20% to final sample number to compensate for lost volume. Pipet antibodies mix into FACS buffer (Table 1) (see Note 7). Vortex for 5 s. 4. Fixation buffer (optional): 150 μL of fixation buffer is required per sample. Calculate necessary amount according to samples Table 1 Antibody panel used in this protocol Antibody
Fluorochrome
Clone
Concentration
CD16/32
Purified
93
2.5 μg/mL
CD45
BUV395
30-F11
0.66 μg/mL
F4/80
APC
BM8
0.66 μg/mL
CD64
BV711
X54-5/7.1
0.66 μg/mL
CD11b
PE.Cy7
M1/70
0.22 μg/mL
CD206
PE
C068C2
0.66 μg/mL
MHCII
ef450
M5/114.15.2
0,2 μg/mL
202
Myriam Sekias et al.
number. Cell fixation buffer (e.g., AntigenFix Diapath) is used to fix cells at a 1:1 ratio with FACS buffer. Under the hood, prepare fixation buffer. Vortex for 5 s and keep tube closed under the chemical hood. 5. Counting buffer: 300 μL of flow cytometry counting beads are required per sample. We use 10 μL of vortex CountBrightTM absolute counting beads per sample with 290 μL FACS buffer. Before adding counting beads, the tube is vortexed for at least 1 min (see Notes 8 and 9).
3 3.1
Methods Organ Harvesting
1. Prepare the container of your choice (tubes or plates) with 1× PBS and put it on ice. 2. After euthanizing the mouse, place it horizontally on the bench, with the belly facing up. 3. Put 70% ethanol on the belly (see Note 10). 4. Pull the skin between the hips up with tweezers, and cut the skin perpendicularly to the mouse (you should have the mouse open vertically from the left hip to the right hip). 5. Cut the peritoneum in the same fashion (if it is not already open from step 4). 6. The inguinal fat pad should be revealed on both sides of the mouse. Using fine tweezers, pull the inguinal fat pad toward the mouse head, and the testis should be attached to the bottom end. 7. Harvest the testis by cutting it from the inguinal fat pad. 8. Put the testis in the container filled with 1× PBS on ice. 9. In the lab, weigh testes after placing them on a clean tissue to dry 1× PBS out, and put them back right away in 1× PBS (see Note 11).
3.2
Organ Digestion
1. Put 1 mL of the digestion mix into 1.5 mL microcentrifuge tubes. 2. On a petri dish, cut a very fine piece of the testis scrotal sac on one end. Hold the testis down from the other end with fine tweezers. With another set of tweezers, start applying pressure from the end held with the fine tweezers, and gently push toward the open end of testis squeezing the tissue out of the scrotal sac. 3. Harvest the tissue from step 2 in tubes containing the digestion mix (1 mL per testis) (see Note 12). 4. Set the 1 mL micropipette to 500 μL, and make an angled cut of the edge of a 1 mL tip.
Testis Macrophage Analysis Using Flow Cytometry
203
5. Pipet up and down to homogenize the solution; pieces of testis tubules might get stuck in the tip. Pipet up and down until the solution is homogenized and all tubules are in the mix. 6. Incubate on heating and shaking dry bath at 37 °C for 30 min with constant shaking (800 rpm). Homogenize the solution after 10 min with a 1 mL micropipette. Ten minutes after that, repeat the procedure with a 200 μL micropipette (see Note 13). 3.3
FACS Staining
1. During tissue digestion, put empty FACS tubes on ice, and add 70 μm nylon mesh on top of each tube (one tube per testis). 2. At the end of digestion, pass digested tissue through nylon mesh (70 μm) onto five FACS tubes, and wash the nylon mesh with 1 mL of 1× PBS (see Note 14). 3. Pipette up and down to homogenize the solution, then take 1/10 of the single-cell suspension (200 μL), and place it in a new 5 mL FACS tube (see Note 15). 4. Wash with 1 mL of 1× PBS (see Note 16), spin down for 4 min 400 g at 4 °C, and get rid of the supernatant (see Note 17). 5. Repeat step 4 one more time (see Note 18). 6. Resuspend in 100 μL of viability solution, and incubate for 15 min on ice. 7. Wash with 1 mL of 1× PBS (see Note 16), and spin down for 4 min 400 g at 4 °C. Get rid of the supernatant (see Note 17). 8. Resuspend cell pellet in 100 μL of staining buffer, and incubate for 30 min on ice in the dark (see Notes 19 and 20). 9. Wash with 1 mL of FACS buffer, and spin down for 4 min 400 g, maximum acceleration, and deceleration at 4 °C; get rid of the supernatant (if fixation is required, proceed immediately to step 4: samples fixation). 10. Add 300 μL of counting buffer per sample. 11. Keep samples on ice until sample acquisition. Continue to Subheading 3.5 if acquiring fresh samples. If it is more than an hour wait, it is better to fix them.
3.4 Samples Fixation (Optional)
If acquisition cannot be performed after sample preparation, cells can be fixed following these steps (see Notes 21 and 22): 1. After centrifugation (step 9 in Subheading 3.3FACS staining), add 150 μL of fixation buffer (under a chemical hood). 2. Fix for 20 min on ice (under a chemical hood). 3. Wash with 2 mL of FACS buffer, and spin down for 4 min 650 g at 4 °C; get rid of the supernatant (see Note 23). 4. Add 300 μL of counting buffer per sample, and store at 4 °C until data acquisition.
204
Myriam Sekias et al.
Fig. 1 Testis macrophage gating strategy. (a) Testis events (cells) displayed on forward and side scatter axis reflecting respectively size and granularity of cells. (b) Coincident events can occur during acquisition, which are gated out using forward and side scatter height vs width. (c) Myeloid lineage is positively selected using CD45 and CD11b. (d) Fixable viability dye is implemented to gate out dead cells. (e) Testis macrophages are double positive for F4/80 and CD64 (f). Two main macrophage populations can be distinguished both exclusively expressing either MHCII or CD206. Frequency of these macrophage populations may vary depending on mouse age. FlowJo™ software was used 3.5 Sample Acquisition and Analysis
4
Samples acquisition is performed on the flow cytometer or spectral analyzer of your choice. Here, samples were run on a BD LSRFortessa™ X-20 cell analyzer using DIVA software. Gating strategy and compensation matrix modification were done using FlowJo (Fig. 1).
Notes 1. Prepare all buffers and digestion mix under a Class II laminar flow hood to avoid contaminations. 2. Alternatively, if FCS stock is limited, bovine serum albumin (BSA) can be used at a concentration of 0.1–0.3%. Solubilization of BSA can be facilitated by using ice cold 1× PBS or putting the mix of PBS-BSA at -20 °C for 10 min before homogenization. 3. Alternatively, if your hood does not have a vacuum system, use a pipetboy (pipet filler) onto the threaded bottle top filter. 4. You have the possibility to store digestion mix at -20 °C up to 1 month. But it is better to use it freshly prepared. 5. All buffers should be freshly prepared on the day of the experiment (except for FACS buffer that can be stored at 4 °C).
Testis Macrophage Analysis Using Flow Cytometry
205
6. If you have several brilliant violet (BV) colors in you panel, it is highly recommended to add brilliant stain buffer to the staining buffer before pipetting your antibody mix, to avoid antibodies aggregation. It can be added to a dilution of 1:1 with FACS buffer. 7. Antibodies aggregation will occur over time; this can be amplified with heat and light exposure. To avoid pipetting aggregates, short spin antibody tubes for 10 s in a microcentrifuge. Make sure to pipet with the tip just below surface level (tension). 8. It is very important to vortex counting beads to pipet a homogenous amount of counting beads per experiment. 9. Usually, 50 μL of counting beads is added (according to manufacturer’s recommendations), but we didn’t see a difference in adding less counting beads per sample. 10. Ethanol helps sanitizing the area and avoid fur contamination. 11. Weighing tissues in flow cytometry helps normalizing all samples in term of number of cells per milligram. 12. After placing the testis in the 1.5 mL tube, keep the tube on ice until finishing all samples. 13. It is a very important step that significantly helps in tissue homogenization and macrophage dissociation. 14. If you don’t have a viability dye in your panel, wash with FACS buffer instead of 1× PBS, spin down for 4 min 400 g at 4 °C, get rid of the supernatant, and proceed to step 8 in Subheading 3.3FACS staining. 15. It is important to take only 1/10 of your digested tissue; otherwise cell concentration will be too elevated for an efficient staining and for analysis, depending on the flow cytometer you will be using. 16. Depending on sample number, staining can be performed in 96-well plate (U or V bottom). In this case adapt washing volume to 100 μL of 1× PBS for all washing steps. Also, to enhance reproducibility, use a multichannel micropipette. 17. Depending on if you are working in plates or tubes, the method will be different. For tubes aspirate supernatant using a bench vacuum and a Pasteur pipette with a 200 μL tip on top of it, stop before reaching the pellet. When working with plates, hold the plate firmly, and flick the plate over a sink and then dry the plate on a clean tissue. 18. If you’re using a viability die, you need to wash several times with 1× PBS to get rid of FCS that can capture the dye. 19. Don’t leave cells with staining mix for more than 45 min; otherwise nonspecific antibody binding might occur.
206
Myriam Sekias et al.
20. Keeping samples in the dark is highly recommended to avoid fluorescence quenching. 21. Even if samples are fixed, it is recommended to pass them no more than 24 h postfixation. Be aware that fixation might alter the quality of a fluorescent dye, not all dyes can be fixed, and it might be best suited to fix before staining. 22. The viability dye used in this protocol is suited for cell fixation; be aware that not all viability dyes are compatible with fixation. 23. Cell density is lower upon fixation. Therefore, it is necessary to centrifuge at a higher speed than usual. References 1. Hume DA, Irvine KM, Pridans C (2019) The mononuclear phagocyte system: the relationship between monocytes and macrophages. Trends Immunol 40(2):98–112. https://doi.org/10. 1016/j.it.2018.11.007 2. Sheu KM, Hoffmann A (2022) Functional hallmarks of healthy macrophage responses: their regulatory basis and disease relevance. Annu Rev Immunol 40:295–321. https://doi.org/ 10.1146/annurev-immunol-101320-031555 3. Nobs SP, Kopf M (2021) Tissue-resident macrophages: guardians of organ homeostasis. Trends Immunol 42(6):495–507. https://doi.org/10. 1016/j.it.2021.04.007 4. Wang Y, Chaffee TS, LaRue RS, Huggins DN, Witschen PM, Ibrahim AM, Nelson AC, Machado HL, Schwertfeger KL (2020) Tissueresident macrophages promote extracellular matrix homeostasis in the mammary gland stroma of nulliparous mice. elife 9. https://doi. org/10.7554/eLife.57438
5. Bellomo A, Mondor I, Spinelli L, Lagueyrie M, Stewart BJ, Brouilly N, Malissen B, Clatworthy MR, Bajenoff M (2020) Reticular fibroblasts expressing the transcription factor WT1 define a stromal niche that maintains and replenishes splenic red pulp macrophages. Immunity 53(1): 127–142. e127. https://doi.org/10.1016/j. immuni.2020.06.008 6. Lokka E, Lintukorpi L, Cisneros-Montalvo S, Makela JA, Tyystjarvi S, Ojasalo V, Gerke H, Toppari J, Rantakari P, Salmi M (2020) Generation, localization and functions of macrophages during the development of testis. Nat Commun 11(1):4375. https://doi.org/10.1038/ s41467-020-18206-0 7. McKinnon KM (2018) Flow cytometry: an overview. Curr Protoc Immunol 120:511–5111. https://doi.org/10.1002/cpim.40 8. Szaloki G, Goda K (2015) Compensation in multicolor flow cytometry. Cytometry A 87(11):982–985. https://doi.org/10.1002/ cyto.a.22736
Chapter 15 Studying Macrophages in the Murine Steatotic Liver Using Flow Cytometry and Confocal Microscopy Zhuangzhuang Liu, Pieter A. Louwe, and Charlotte L. Scott Abstract The study of macrophage functions in the context of metabolic dysfunction-associated steatotic liver disease (MASLD) and metabolic dysfunction associated steatohepatitis (MASH) has been hampered by the fact that until recently all macrophages in the liver were thought to be Kupffer cells, the resident macrophages of the liver. With the advent of single-cell technologies, it is now clear that the steatotic liver harbors many distinct populations of macrophages, likely each with their own unique functions as well as subsets of monocytes and dendritic cells which can be difficult to discriminate from one another. Here, we detail the protocols we utilize to (i) induce MASLD/MASH in mice, (ii) isolate cells from the steatotic liver, and (iii) describe reliable gating strategies, which can be used to identify the different subsets of myeloid cells. Finally, we also discuss the issue of increased autofluorescence in the steatotic liver and the techniques we use to minimize this both for flow cytometry and confocal microscopy analyses. Key words Macrophages, Liver, MASLD, MASH, Diet-induced obesity, Autofluorescence, Kupffer cells, Lipid-associated macrophages, Flow cytometry, Confocal microscopy, NAFLD
1
Introduction
1.1 Macrophage Functions in MASLD
Metabolic dysfunction-associated steatotic liver disease (MASLD) previously referred to as nonalcoholic fatty liver disease (NAFLD), represents a spectrum of disease states ranging from the oftenasymptomatic simple steatosis (a buildup of fat in the liver) to the more end stage of the disease now called metabolic dysfunctionassociated steatohepatitis (MASH), which is characterized by excessive inflammation and fibrosis, leading to cirrhosis and even hepatocellular carcinoma (HCC). With the current obesity epidemic, the prevalence of MASLD is increasing at an alarming rate across the world, with ~25% of the world’s population having MASLD [1]. Importantly, not all patients progress from steatosis to MASH, and the factors underlying this progression remain largely unknown. However, multiple “hits” or insults are thought to drive progression to MASH including excess lipid accumulation,
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_15, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
207
208
Zhuangzhuang Liu et al.
insulin resistance, changes in microbiota, and inflammation [2]. Through their response to danger-associated molecular patterns (DAMPs) and other signals generated by the excess lipid burden, hepatic mononuclear phagocytes (MNPs), including monocytes, macrophages, and dendritic cells (DCs), have been proposed to play a crucial role in driving hepatic inflammation in MASLD, contributing to the development of MASH and cirrhosis. Despite this hypothesis, the precise roles played by hepatic macrophages in MASLD/MASH pathogenesis remain unclear. This stems from the fact that until recently, all macrophages in the liver were considered to be Kupffer cells, the resident macrophages of the liver. As such these cells were proposed to play many, often contradictory, roles, being both pro- and anti-inflammatory. The recent advances in single-cell technologies, however, have revealed considerable heterogeneity within the macrophage pool in both the healthy and steatotic liver [3–9]. Thus, efforts are now focused on dissecting the specific roles of the different populations. Here we describe how to isolate, characterize, and localize steatotic liver macrophages using both flow cytometry and confocal microscopy. 1.2 Different Diets to Study MASLD and MASH
One of the main issues with studying MASLD and MASH in mice using diet-induced models is that there is no consensus on the best diet to use to induce MASLD/MASH. There are many distinct diets available and routinely used in labs. These range from simple high-fat diets (HFDs) which contain ~60% fat, to the more complex western diet (WD) which in addition to the fat (~58%), also contains 1–2% cholesterol and are supplemented with sucrose and fructose either in the food itself or in the drinking water. While mice on these diets gain significant amounts of weight over time and hence get MASLD, the progression to MASH is often very slow (Fig. 1). For example, it takes 24–36 weeks on the WD, while mice fed the HFD rarely develop fibrosis [10, 11]. Therefore, diets have been developed which speed up the development of MASH including the methionine and choline deficient (MCD) diet and the choline-deficient L-amino acid defined high-fat (CDA-HF) diet. Mice fed these diets rapidly progress to MASH but do not gain weight (Fig. 1); in fact mice on the MCD diet also rapidly lose weight [10, 12]. While the weight loss associated with the MCD diet limits its usefulness, the CDA-HF diet is rapidly becoming the go-to diet for many labs. Already after 1 week on the CDA-HF diet, mice develop a fatty liver without the presence of fibrosis [13], and as early as 6 to 8 weeks on the CDA-HF diet, mice develop fibrosis [14, 15]. While overall body weight is not increased on this diet, CDA-HF diet-fed mice have a significantly higher liver/body weight ratio which increases with time on the diet [16]. The relatively short feeding time for these diets to induce MASH is thus a huge advantage of these models; however, the lack of obesity remains a concern regarding clinical relevance.
Macrophages in the Steatotic Liver
209
Fig. 1 Different diets induce MASH at distinct rates. C57Bl/6 WT male mice were placed on either a SD, WD, or CDA-HF diet at 5 weeks of age and followed up for 4 or 12 weeks. Comparison of body weight (left), AST/ALT serum values (middle) and PicoSirius red staining (right) from mice fed either the SD or WD for 12 weeks (a) or the SD and CDA-HF diet for 4 weeks (b). Scale bar is 50 μm. Data are representative of at least two individual experiments with six mice per group. **p < 0.01, ****p < 0.0001, Student’s t test
1.3 Isolating Macrophages from the Steatotic Liver
In immunology, it is common place to perform ex vivo enzymatic digestion of tissues to liberate cells for downstream analysis. This often involves removing the organ of interest from the euthanized mice and subsequently cutting it into small pieces and incubating with enzymes until the tissue is digested. Such protocols also exist for studying macrophages in the liver (see Subheading 3.2.1), allowing the easy isolation of macrophages and other CD45+ cells. However, it has also recently been shown that these protocols do not efficiently isolate all CD45+ cells, including some macrophages, nor do they isolate the CD45- cells, such as hepatocytes, stellate cells, or liver sinusoidal endothelial cells [3], as so much of macrophage biology is dependent on their interactions with cells in their local environment, being able to isolate all cells of the liver becomes imperative for macrophage biologists. Thus, protocols have also been developed, which make use of the liver’s dual blood supply to perfuse the liver with enzymes in situ and essentially digest it from the inside out [17, 18]. With such protocols, the number of cells isolated and the proportions of the different cell types much more closely mirror what is observed in the intact tissue by confocal microscopy [3]. As this in vivo digestion is much more time-consuming and is not always possible due to the need for specific equipment, here we describe and compare both protocols.
210
Zhuangzhuang Liu et al.
Fig. 2 An updated view of hepatic macrophage heterogeneity. The homeostatic liver consists of at least three populations of macrophages. The largest population are the resident KCs (ResKCs) which can be identified in mice and humans based on their expression of VSIG4, FOLR2, and TIM4. Additionally, in the mouse CLEC4F and CLEC2 can be used to identify KCs. Alongside the KCs, there is a smaller population of macrophages found in the liver capsule and at the central vein called capsule and central vein macrophages (CapMac) which can be identified in mouse by their lack of KC specific markers and expression of CD207 and CX3CR1, while a similar population of cells can be visualized in the human liver capsule, specific markers of these cells in human are yet to be identified. Finally, the homeostatic liver also contains a small population of macrophages called LAMs that are found specifically around bile ducts. In both mouse and human, these cells express Gpnmb and Spp1 and lack KC markers. In MASLD, these homeostatic macrophage populations are joined by two recruited populations of macrophages. These develop from monocytes recruited to the liver and can exist as either monocyte-derived KCs (moKCs) which express all KC markers but lack TIM4 (initially) or inflammatory LAMs (inf-LAMs) that resemble the LAMs found in the steady state, except these LAMs are now found enriched in zones of steatosis and fibrosis 1.4 Macrophage Heterogeneity in the Healthy and Steatotic Liver
Kupffer cells (KCs) are one of the oldest immune cells known to man, first being described in 1876 by Wilhem von Kupffer, although they were only identified as macrophages two decades later by Tadeusz Browicz [19–22]. Despite this, we still know relatively little about their specific functions as until recently it was not possible to discriminate these cells from other hepatic macrophages, which all express generic macrophage markers such as CD64, F4/80, and MerTK in mice or CD68 in humans. In recent years, thanks to advances in single-cell technologies including flow cytometry and single-cell RNA-sequencing, our understanding of hepatic macrophage heterogeneity has greatly increased, and we now have useful markers to discriminate most of these cell types (Fig. 2 and reviewed in detail in [23]). For example, we now know that murine KCs can be identified based on their expression of CLEC4F, VSIG4, and CLEC2 [3, 4, 6, 24], markers which are not expressed on other hepatic macrophage populations. While under homeostatic conditions, KCs represent a homogenous population of cells, derived during embryogenesis and maintained through self-renewal in the adult, in MASLD and MASH, some KCs die and are subsequently replaced by monocytes which enter the liver and differentiate into monocyte-derived KCs (moKCs). Importantly, expression of TIM4 can be used, at least initially, to
Macrophages in the Steatotic Liver
211
discriminate between resident (Res) and moKCs. This marker is expressed by KCs that have been resident in the liver for prolonged periods, making it a useful marker to discriminate between ResKCs and those that have just developed in the liver. moKCs gain TIM4 expression slowly, with about 50% of the population being TIM4+ after 30 days; thus, by combining KC markers (CLEC4F, VSIG4, and CLEC2) with TIM4 expression, ResKCs and newly generated moKCs can be discriminated. Alongside KCs, other populations of macrophages have also been described in the healthy and steatotic liver (reviewed in detail in [23]). This includes a population of macrophages found at the capsule and around the central veins of the liver, which lack KC markers but express CD207 and CX3CR1 [3, 25] and a population of macrophages termed lipid-associated macrophages (LAMs) that are found in close proximity to the bile duct in the healthy liver. Unfortunately, while these LAMs have a unique gene expression profile, expressing genes including Gpnmb, Spp1, and Trem2 [3], to date we have not identified any useful surface markers that can be used to identify these cells by flow cytometry (GPNMB and TREM2 work by confocal microscopy). Therefore, we have to rely on a negative gating strategy to identify these cells (see Subheading 3.5). While LAMs represent a very small population in the healthy liver, these cells are significantly increased in number in the steatotic liver. Here they are derived from recruited bone marrow monocytes and are found specifically at zones of steatosis, fibrosis, and inflammation, suggesting they may play important roles in disease pathogenesis, although this remains to be studied [3, 4, 26]. With this heterogeneity in mind, here we share our currently gating strategies to identify these distinct cell types in the steatotic liver by flow cytometry. We also share our protocols to identify these cells by confocal microscopy so that their precise location can be examined in different models as often this may shed light on cellular functions. 1.5 Autofluorescence Problems in the Steatotic Liver
The healthy liver already harbors considerable autofluorescence compared with other tissues. However, this is amplified further by the presence of MASLD/MASH. Depending on the diet used to induce MASLD/MASH, this can be more severe in some models than others. For example, the CDA-HF diet induces significantly more autofluorescence than the WD (data not shown). In order to be able to get meaningful results from the flow cytometer and confocal microscope, this autofluorescence has to be taken into consideration. Here, we detail the methods and staining panels we use to prevent issues in interpreting the data due to this autofluorescence signal.
212
2
Zhuangzhuang Liu et al.
Materials
2.1 Diets to Induce MASLD/MASH
1. Western diet: 58% fat and 1% cholesterol (D09061703i, Research diets).
2.1.1
2. Control diet: 11% fat with corn starch (D12328i; Research diets).
Western Diet
3. Sugar water: Mix 23.1 g/L fructose with 18.1 g/L sucrose and top up to 1 L with water. 4. Weighing scales. 2.1.2
CDA-HF Diet
1. CDA-HF diet: 60% fat with 0.1% methionine and no added choline (A06071302i; Research diets). 2. Control diet: 11% fat with corn starch (D12328i; Research diets). 3. Sugar water: Mix 23.1 g/L fructose with 18.1 g/L sucrose and top up to 1 L with water. 4. Weighing scales.
2.2 Isolating Macrophages from the Murine Steatotic Liver 2.2.1
Ex Vivo Digestion
1. RPMI, kept on ice (2 mL per liver to collect and 3–5 mL per liver to digest livers in). 2. Phosphate buffered saline (PBS): 5 mL per liver to perfuse livers with. 3. Collagenase A (0.5 mg/mL). 4. DNase I (10 U/mL). 5. Gentle MACS and Gentle MACS tubes (Miltenyi, one tube per liver). 6. FACS buffer: 2% FCS and 2 mM EDTA in PBS; stored at 4 °C. 7. Shaking water bath at 37 °C. 8. Centrifuge. 9. Osmotic lysis buffer: 155 mM NH4Cl, 12 mM NaHCO3, and 0.1 mM EDTA in dH20. pH 7.1–7.4. 10. 24-well plates. 11. 100 μm cell strainers. 12. 40 μm cell strainers. 13. 50 mL Falcon tubes.
2.2.2
In Vivo Digestion
1. Perfusion pump. 2. 1× water bath at 39 °C and 1 × water bath at 37 °C. 3. Buffer 1: 0.5 mM EGTA, 4.5 mM glucose, 10 mM HEPES, 5.4 mM KCl, 0.85 mM Na2HPO4 × 2 H2O, 136.9 mM NaCl, 0.57 mM Na2H2PO4 × H2O, 4.2 mM NaHCO3, 6 mg/L
Macrophages in the Steatotic Liver
213
phenol red in dH2O. It can be prepared in 1 L and stored for 3 months at 4 °C. Prepare 20 mL (see Note 1) per liver in 50-mL tubes (Tube 1); pre-warmed to 39 °C. 4. Buffer 2: 3.8 mM CaCl2 × 2H2O, 10 mM HEPES, 5.4 mM KCl, 0.85 mM Na2HPO4 × 2 H2O, 136.9 mM NaCl, 0.57 mM Na2H2PO4 × H2O, 4.2 mM NaHCO3, 6 mg/L phenol red, in dH2O. It can be prepared without collagenase A in 1 L and stored for 3 months at 4 °C. 5. Buffer 2 0.2 mg/mL ColA: Prepare 40-mL buffer 2 (see Note 1) per liver in 50-mL tubes (Tube 2); pre-warmed to 39 °C. 6. Buffer 2 0.4 mg/mL ColA: 20–50-mL buffer 2 (see Note 1) supplemented with 0.4 mg/mL collagenase A per liver in 50-m tubes (Tube 3); pre-warmed to 39 °C. 7. 10 U/mL DNase I per liver (to be added to buffer 2 during protocol); keep on ice. 8. 26G needle for perfusion. 9. Perfusion clamps. 10. 10 cm petri dish to mince liver in (one per liver). 11. Centrifuge. 12. Osmotic lysis buffer: 155 mM NH4Cl, 10 mM KHCO3, and 0.1 mM of EDTA in dH20. pH 7.1–7.4. 13. 100 μm cell strainers. 14. 40 μm cell strainers. 15. 50 mL Falcon tubes. 2.3 Flow Cytometry for Macrophages in the Steatotic Liver
1. Antibodies for flow cytometry (see Tables 1 and 2). 2. Compensation beads. 3. Counting beads. 4. Filter top (40 μm) FACS tubes. 5. Conventional or spectral flow cytometer.
2.4 Confocal Microscopy for Macrophages in the Steatotic Liver
1. Antigen Fix (International Medical Products), or similar cell fixation solutions.
2.4.1 Antigen-Fix Perfusion
4. Scalpel.
2. Perfusion pump. 3. PBS. 5. FACS tube with lid pre-filled with antigen fix (~5 mL per liver). 6. 34% sucrose in PBS (5 mL per liver). 7. Rotator for tubes. 8. Embedding compound. 9. Cryomolds.
214
Zhuangzhuang Liu et al.
L+A4:M34aser
Blue
Blue
Y/G
IRC Channel ID
Table 1 Antibody panel for conventional flow cytometer (five laser BD FACS Symphony S6)
FITC BB630P BB660P BB790P PerCP-Cy5.5 PE Pe-CF594
Fluorochrome
Fc Block 2.4G2
Step1: PBS (45min) Ab Clone
CD16/32
Biotin
CD64
2.4G2
X54-5/7.1
Step2: BV/PBS buffer (30min) Dilution Fluorochrome Ab Clone
200
100
488nm 561nm Pe-Cy5
Violet 405nm
Red 633nm
UV 355nm 20mW
Pe-Cy5.5 Pe-Cy7 Pacific Blue BV 510 BV 570 BV 605 BV 650 BV 711 BV750-P BV 786 APC AlexaFluor 700 APC-Cy7
UV-BUV395 BUV450 BUV496 BUV563 BUV615-P BUV661 BUV737 BUV805
2.4.2 Cutting Sections, Staining, and Imaging
BUV450
Live/Dead
N/A
300
Dilution
FITC
MHCII
2.4G2 M5/114.15.2
200 800
PE PE-CF594 PECy5 PECy5 PECy5 PECy5 PECy5
Clec2 SAv CD3e CD19 B220 NK1.1 Ter-119
17D9 N/A 145-2C11 1D3 RA3-6B2 PK136 TER-119
400 400 300 300 300 300 300
PECy7 eF450
Vsig4 CD11c
NLA-14 N418
800 200
BV570
Ly6C
HK1.4
400
Fc Block 2.4G2 CD16/32
BV650
XCR1
ZET
200
BV750
SiglecF
E50-2440
100
AF647 AF700 AF780 BUV395
Tim4 CD31 F4/80 CD11b
54 390 BM8 M1/70
200 200 100 200
BUV563
Ly6G
1A8
200
BUV737 BUV805
CD43 CD45
S7 30-F11
200 400
1. Cryotome. 2. Slide microscope KP frost plus. 3. Humidity chamber. 4. Hydrophobic barrier pen. 5. PBS. 6. Sudan black stock solution; 0.7%w/v in 70% ethanol, filtered through a 22 μm filter. 7. Antibodies for confocal microscopy (see Table 3). 8. Serum from relevant species (see Table 3). 9. Blocking buffer: PBS, 0.5% Saponin, 2% BSA, 1% Donkey or goat serum, and 1% FCS. 10. Antifade mountant. 11. Cover slips. 12. Confocal microscope.
Macrophages in the Steatotic Liver
215
Table 2 Antibody panel for spectral flow cytometer (four laser Cytek Aurora) Laser
Filter setup Start-End Channel ID
Fluorochrome
Ab
Clone
Dilution
Fc Block 2.4G2
2.4G2
200
eF450 Sytox Blue BV480
CD11c
N418
200 8000
BV510 BV570 BV605 BV650
XCR1 Ly6C CD11b Ly6G
ZET HK1.4 M1/70 1A8
200 400 200 400
BV711 BV750 BV785
SiglecF CD31
E50-2440 390
100 300
FITC
MHCII
M5/114.15.2
800
BB790 PE
Clec2
17D9
300
PE-Dazzle594
CD64
X54-5/7.1
100
PECy5 PECy5 PECy5 PECy5 PECy5
CD3e CD19 B220 NK1.1 Ter-119
145-2C11 1D3 RA3-6B2 PK136 TER-119
300 300 300 300 300
PE-Cy7
Vsig4
NLA-14
1000
APC AF647
Tim4
54
200
AF700 AF780
CD45 F4/80
30-F11 BM8
400 100
Total Volume Block
Violet 405nm_100mW
Blue 488nm_50mW
Yellow-Green 561nm_50mW
Red 633nm_80mW
428/15
420-435
V1
443/15
436-451
V2
458/15
451-466
V3
473/15
466-481
V4
508/20
498-518
V5
525/17
516-533
V6
542/17
533-550
V7
581/19
571-590
V8
598/20
588-608
V9
615/20
605-625
V10
664/27
651-678
V11
692/28
678-706
V12
720/29
706-735
V13
750/30
735-765
V14
780/30
765-795
V15
812/34
795-829
V16
508/20
498-518
B1
525/17
516-533
B2
542/17
533-550
B3
581/19
571-590
B4
598/20
588-608
B5
615/20
605-625
B6
661/17
652-669
B7
679/18
669-687
B8
697/19
688-707
B9
717/20
707-727
B10
738/21
728-749
B11
760/23
749-772
B12
783/23
772-795
B13
812/34
795-829
B14
577/20
567-587
Y1
598/20
588-608
Y2
615/20
605-625
Y3
661/17
652-669
Y4
679/18
669-687
Y5
697/19
688-707
Y6
720/29
706-735
Y7
750/30
735-765
Y8
780/30
765-795
Y9
812/34
795-829
Y10
661/17
652-669
R1
679/18
669-687
R2
697/19
688-707
R3
717/20
707-727
R4
738/21
728-749
R5
760/23
749-772
R6
783/23
772-795
R7
812/34
795-829
R8
BB630
PerCP-Cy5.5 PerCP-eFluor710
216
Zhuangzhuang Liu et al.
Table 3 Antibody panel for confocal microscopy
3
Antibody
Clone
Catalogue #
Dilution
Step
Goat anti-mouse CLEC4F
Polyclonal
AF2784
1/100
1
Rabbit anti-mouse CD64
27
MA5-29706
1/200
1
Rat anti-mouse TIM4
RMT4-54
130002
1/100
1
Donkey anti-goat AF488
–
A-11055
1/400
2
Donkey anti-rat Cy3
–
712-166-153
1/400
2
Donkey anti-rabbit AF647
–
A-31573
1/400
2
Methods
3.1 Western Diet vs Choline-Deficient LAmino Acid Defined High-Fat Diet to Induce MASLD/MASH
1. Place mice on the respective diets at 5 weeks of age (see Note 2).
3.2 Isolating Macrophages from the Steatotic Liver
1. Following confirmation of euthanasia, the peritoneal cavity of the mice is opened and the inferior vena cava identified. This vein is then perfused using a 10-mL syringe and 26G needle with ~5 mL PBS to remove blood from the liver. The portal vein should be cut with scissors to allow blood to escape.
3.2.1
Ex Vivo Digestion
2. Weigh mice weekly to follow weight gain. 3. Upon completion of the required time on the diet, typically 12, 24, or 36 weeks for WD and 4, 8, or 12 weeks for CDA-HF diet if studying MASLD/MASH without HCC, euthanized mice following local ethical guidelines by cervical dislocation or CO2 overdose. For perfusion experiments, CO2 overdose is recommended.
2. The liver is then excised from the mice (see Note 3) and put into a 24-well plate containing RPMI (2 mL/well); the plate is kept on ice until all livers are collected. 3. When ready with all livers, add enzymes to pre-warmed RPMI for digestion, and put back in water bath to keep warm. 4. Place livers in GentleMACS tubes and cut roughly with a scissors into pieces. 5. Add 5 mL RPMI containing enzymes (see Note 3) to each tube, and close the lid firmly. 6. Run Liver protocol 1 on GentleMACS dissociator and then place tubes in the water bath at 37 °C shaking for 20–25 min. Shake manually every 5 min. 7. Run Liver protocol 2 on GentleMACS dissociator, and place tubes on ice to stop the digestion.
Macrophages in the Steatotic Liver
217
8. Filter the cell suspension through a 100 μm filter into a 50 mL tube containing 30 mL ice-cold PBS. Wash the filter with an additional 20 mL ice-cold PBS to bring final volume in the tube to 50 mL. 9. Centrifuge at 400 × g for 5 min and discard supernatant carefully. 10. Add 3 mL osmotic lysis buffer to lyse red blood cells, and leave on ice for 3 min. 11. Wash cells with 20 mL PBS and centrifuge at 400× g for 5 min and discard supernatant. 12. Resuspend in 0.5–1 mL FACS buffer and add 50 μL of counting beads (see Note 3). 13. Leave on ice until ready to proceed to staining (see Subheading 3.3: Preparing Cells for Flow Cytometry). 3.2.2
In Vivo Digestion
1. Set up the perfusion pump and water bath as shown in Fig. 3. Run pre-warmed buffer 1 (Tube 1) through the tubing of the perfusion pump for 10 min before beginning. 2. Following confirmation of euthanasia, the peritoneal cavity of the mice is opened, and the inferior vena cava identified. The 26G needle at the end of the tubing of the perfusion pump is then inserted into this vein and clamped in position. The pump is started, and this vein is perfused with buffer 1 (Tube 1) at a flow rate of 6 mL/min until the blood has been removed and
Fig. 3 Perfusion apparatus set up for in vivo digestions. Photograph showing the setup of perfusion pumps and water baths for in vivo digestion protocol
218
Zhuangzhuang Liu et al.
the liver turns a pale color ~2 min (see Note 4). Once the perfusion is seen to be working, the portal vein is cut with scissors to allow efficient perfusion. 3. After the blood has been flushed from the liver, the pump is stopped and the tubing moved to the tube of buffer 2 (Tube 2; also in the water bath) containing 0.2 mg/mL of collagenase A. The pump is then turned on and, the liver is perfused with this buffer (Tube 2) at a flow rate of 6 mL/min until the tube is empty (see Note 5). 4. When the last few milliliters of this buffer are running through the liver, 10 U/mL of DNAse I should be added to Tube 3 containing pre-warmed buffer 2 and 0.4 mg/mL collagenase A. Some of the contents of this tube (~20 mL) are then added to a petri dish. Once the perfusion is finished, the liver is then excised from the mouse and placed in the petri dish. 5. Using two pairs of tweezers, the liver is then torn into small pieces (see Note 6), and the contents of the petri-dish are then added to the remaining buffer 2 in Tube 3 and placed on ice. 6. Once all perfusions have been performed, the tubes containing the shredded livers are placed in a water bath (without shaking) at 37 °C for 20 min. 7. The cells are then filtered through a 100-μm filter into a fresh 50 mL tube and centrifuged at 400 × g for 5 min. Discard the supernatant. 8. Add 3 mL osmotic lysis buffer to lyse red blood cells and leave on ice for 3 min. 9. Wash cells with 20 mL PBS, and centrifuge at 400 × g for 5 min and discard supernatant. 10. Resuspend cells in 50 mL PBS, and filter through a 40 μm filter into a new 50-mL tube. 11. Centrifuge at 50 × g for 1 min. After this spin step, the pellet contains hepatocytes, while the supernatant contains all other cells. Transfer the supernatant to another tube and repeat this step. If hepatocytes are useful for the experiment, resuspend in FACS-buffer and keep on ice. 12. After the second 50 × g spin, transfer the supernatant to a new 50 mL tube, add 50 μL counting beads, and centrifuge at 400 × g for 5 min. Discard supernatant. 13. Resuspend pellet in 1–5 mL FACS buffer depending on pellet size and keep on ice until ready to proceed with staining for flow cytometry.
Macrophages in the Steatotic Liver
3.3 Preparing Cells for Flow Cytometry
219
1. Add 3-7 × 106 cells in FACS buffer to each well of a 96-well plate. 2. Centrifuge at 400 × g for 3 min at 4 °C and discard supernatant. 3. Incubate with 100 μL of staining step 1 mix + anti-CD16/32 “Fc block” for 45 min at 4 °C in PBS (see Note 7). 4. Wash cells by adding 100 μL FACS buffer, centrifuge at 400 × g for 3 min at 4 °C, and discard supernatant. 5. Incubate with 100 μL of staining step 2 mix + anti-CD16/32 “Fc block” for 30 min at 4 °C in a 50:50 mix of PBS and brilliant stain buffer (see Note 7). 6. Wash cells by adding 100 μL FACS buffer, centrifuge at 400 × g for 3 min at 4 °C, and discard supernatant. 7. Resuspend cells in 200 μL FACS buffer; add cell viability dye, if required (see Table 2 and Note 8); filter through a 40 μm filter; and keep on ice until ready to measure. 8. Analyze on conventional or spectral flow cytometer.
3.4 Removal of Autofluorescence Using Spectral Flow Cytometry
1. Run and acquire an unstained sample with autofluorescence signal. 2. In the software SpectroFlo, open an NxN plot, and find a combination that has the highest autofluorescence signal (e.g., V6 vs YG6). 3. Gate the autofluorescent+ and autofluorescent- cells, removing any cells with extreme scatter properties from both gates. 4. Right click on the gate to export events as an FCS file. 5. To designate the autofluorescent signature as a fluorescent tag in SpectroFlo, create a new fluorescent tag for the autofluorescent+ cells, name it, and choose an excitation laser and assign emission wavelength (use an otherwise empty channel). 6. Open the experiment with all compensation controls recorded, add the autofluorescent tag as a fluorophore, choose the additional negative control(s) for spillover calculation in the reference group option, and record unstained cells without autofluorescent (e.g., SD). 7. Import autofluorescent- cell FCS file as a negative control and autofluorescent+ cell FCS file as reference controls in the relevant channels. 8. Unmix the experiment Subheading 3.6).
and
analyze
as
normal
(see
220
Zhuangzhuang Liu et al.
3.5 Phenotypic Identification of Macrophages in the Steatotic Liver
As discussed above, our understanding of hepatic macrophage functions has been stifled by the fact that previously all macrophages in the liver were considered to be Kupffer cells. Now that we know there are many distinct subsets of macrophages in the healthy and steatotic liver, the main aims of our work have been to distinguish between these subsets and other myeloid cells (cDCs, monocytes, neutrophils) by applying rigorous flow cytometric techniques. Here we describe the gating strategies and staining panels we have found most useful for this in the steatotic liver.
3.5.1 Identifying Macrophages in the Steatotic Liver Using Conventional Low Cytometry
Having generated a single-cell suspension, the choice of fluorochromes for the different markers in the flow cytometry panel is absolutely crucial. Alongside the traditional factors to consider in flow cytometry panel design, the autofluorescence of the liver especially after the CDA-HF diet must also be taken into consideration. As is clear from Fig. 4, many channels show substantial signals coming from autofluorescence preventing the identification
Fig. 4 CDA-HF diet induces significant autofluorescence in the macrophage population affecting channel use for conventional flow cytometry. (a) Representative plots showing autofluorescence in indicated channels from different FMO stains in livers of mice fed the CDA-HF diet for 4–10 weeks. Cells shown are gated as macrophages based on CD64, F4/80, and/or CD11b expression. Red boxes indicate channels with substantial autofluorescence issues. (b, c) Representative plots showing FMO (blue) and real antibody staining (red) for indicated markers when the autofluorescence spectra is not corrected (b) compared with when it is (c)
Macrophages in the Steatotic Liver
221
of true real signal versus autofluorescence when those channels are used. Thus, we have carefully designed a panel which overcomes these issues avoiding some of the worst affected channels (see Table 1). Once a suitable panel has been designed, then macrophages can be identified and the different populations studied (Fig. 5). To do this, we first need to identify live CD31-CD45+ single cells. It is important to consider both CD31 and CD45 as previous work has shown that the digestion protocol may lead to the generation of KC-LSEC doublets expressing both CD45 and CD31; thus, with this gating, these can be removed from the analysis [3]. It should be noted that the precise nature of these cells is a controversial issue [27–30]; thus, whether all of these cells should be excluded remains an open question. Following the identification of the live CD31-CD45+ cells, we next exclude cells expressing lineage markers including CD3e, CD19, B220, NK1.1, and Ter-119. At this stage, we can also identify and exclude neutrophils (Ly6G+CD64-) and eosinophils (SiglecF+CD64-) from further analysis (Fig. 5). Then bona fide macrophages can be identified based on their expression of CD64 and F4/80. Within the macrophage pool, KCs can be identified based on VSIG4 expression, and these can be further split into ResKCs and moKCs based on TIM4 expression (Fig. 5). The non-KC macrophage population (VSIG4-) can then be split based on CD11b and CLEC2 expression into two populations, CLEC2+ pre-moKCs; cells en route to become moKCs but that have not yet acquired VSIG4 expression [4, 6] and CD11bhi macrophages. This CD11bhi population consists of both capsule and central vein macrophages and LAMs. However in the steatotic livers, the majority of these cells are LAMs (Fig. 5). As discussed above, we unfortunately do not have a good marker to positively identify LAMs by flow cytometry. CD207 could be added to identify the capsule and central vein macrophages; however, this antibody only works optimally when used to stain cells intracellularly, thus, restricting its usefulness when sorting cells for downstream functional studies. Finally within the CD64-F4/80- non-macrophages, Ly6Chi and Ly6Clo monocytes can be identified within the CD11bhiMHCII- fraction. In the MHCII+ fraction transitioning monocytes (Ly6C+CD64+) and cDCs (Ly6C-CD64-CD11c+MHCII+) can be identified, the latter of which can then be further subdivided into cDC1s and cDC2s based on XCR1 and CD11b expression (Fig. 5). 3.5.2 Identifying Macrophages in the Steatotic Liver Using Spectral Flow Cytometry
With the ability to subtract the autofluorescence spectra from the samples as described above, spectral flow cytometry renders many more channels usable than in conventional flow cytometry (Fig. 4). For example, BV510 which suffers from significant autofluorescence signal, preventing its use in conventional flow cytometry, can now be used in spectral cytometry. Thus here, we are also
222
Zhuangzhuang Liu et al.
Fig. 5 Optimized gating strategy to define different macrophage populations using conventional flow cytometry. Representative gating strategy using optimized panel on a conventional flow cytometer to define distinct myeloid populations in the liver of a C57Bl/6 WT male mouse fed the CDA-HF diet for 4 weeks. Sample was pre-gated on single cells
able to identify the different macrophage populations despite the fact that we are using a four-laser machine instead of a five-laser one. To identify the different macrophage populations, we can take a similar gating strategy to the one discussed above (Fig. 6). First live CD31-CD45+ cells are identified, after which lineage+, Ly6G+ cells, and SiglecF+ cells are gated out. Then, macrophages can be identified using generic macrophage markers CD64 and F4/80. ResKCs and moKCs are identified as VSIG4+TIM4+ and VSIG4+TIM4-, respectively, while in the VSIG4- macrophages, pre-moKCs and LAMs are again identified with CLEC2 and
Macrophages in the Steatotic Liver
223
Fig. 6 Optimized gating strategy to define different macrophage populations using spectral flow cytometry. (a) Representative gating strategy using optimized panel on a spectral flow cytometer to define distinct myeloid populations in the liver of a C57Bl/6 WT male mouse fed the CDA-HF diet for 4 weeks. Sample was pre-gated on single cells. (b) Comparison of absolute numbers of distinct macrophage populations following 4 weeks of CDA-HF diet feeding, using the optimized panels for conventional or spectral flow cytometry
CD11b, respectively. Also, as described above, Ly6Chi and Ly6Clo monocytes can be identified within the CD11bhiMHCII-CD64-F4/80- cells. In the MHCII+CD64-F4/80- cell, fraction transitioning monocytes (Ly6C+CD64+) and cDCs (Ly6C-CD64-CD11c+MHCII+) can be identified, and, finally, the cDCs can be subdivided into cDC1s and cDC2s based on XCR1 and CD11b expression (Fig. 6). Crucially, regardless of whether conventional or spectral flow cytometry is used, similar results are obtained in terms of cell numbers per population, highlighting the validity of both strategies to eliminate the autofluorescence problem and to identify and study hepatic macrophages (Fig. 6).
224
Zhuangzhuang Liu et al.
Fig. 7 Comparison of different techniques for isolating liver macrophages across diets. Absolute numbers of ResKCs (a), moKCs (b), LAMs (c), and pre-moKCs (d) isolated from livers of mice fed the SD, WD, or CDA-HFD for 12 or 4 weeks, respectively, using the ex vivo and in vivo digestion protocols as indicated. ***p < 0.001, ****p < 0.0001 One way ANOVA with Sidak’s multiple comparison test 3.5.3 Comparing Results Obtained with the Different Digestion Protocols across Diets
3.6 Preparing Liver Tissue for Confocal Microscopy
In addition to comparing the analysis methods, we have also compared the results obtained with the different digestion methods. Fitting with our previously published results, the in vivo digestion liberates many more cells than the ex vivo digestion method (Fig. 7). However, despite this difference in cell numbers, importantly all trends observed in the data following in vivo digestion are also observed in the data obtained following the ex vivo digestion. This highlights that both protocols allow changes in the hepatic macrophage pool to be robustly measured (Fig. 7). This is important information, especially given the added level of difficulty associated with and the additional equipment required for the in vivo digestion method. 1. Following confirmation of euthanasia, the peritoneal cavity of the mice is opened and the inferior vena cava identified. The 26G needle at the end of the tubing of the perfusion pump is then inserted into this vein and clamped in position. The pump is started, and this vein is perfused with cold antigen fix for 5 min at 5 mL/min (see Note 8). 2. After perfusion, the liver is carefully excised from the mouse to avoid any damage. The left lobe is identified, and 2–3 slices of 1-mm thickness are taken.
Macrophages in the Steatotic Liver
225
3. The slices (up to three) are then put in a FACS tube containing 5 mL of antigen fix and put on a rotator at 4 °C for 1 h. 4. Antigen fix is then removed and replaced with cold PBS. This is then removed and replaced with more PBS and repeated a third time to wash the liver. 5. The liver pieces are resuspended in PBS and put back on the rotator for 10 min to thoroughly wash off any remaining traces of antigen fix. 6. PBS is discarded and replaced with 34% sucrose solution (5 mL). The samples are put back on the rotator at 4 °C and left for a minimum of 12 h, but best overnight (see Note 9). 7. Embed the liver pieces in embedding compound in cryo-molds (see Note 10), and freeze quickly on ethanol and dry ice. Store at -20 °C if to be used within 1 month or -80 °C for longterm storage. 8. Sections can be cut on a cryotome (typically 20 mm thick), placed on a microscopy slide, and stored at -20 °C for 1 month. 3.7 Staining Steatotic Liver Tissue for Confocal Microscopy
1. Take the microscopy slides with tissue out of the freezer, and allow to come back to room temperature (RT). 2. Circle the pieces of tissue with the hydrophobic pen (see Note 11). 3. Rehydrate the slices in PBS for 5 min. 4. Add blocking buffer to the slices (100–200 μL per slice), and incubate at RT in a humidity chamber in the dark for 30 min. 5. Prepare primary antibodies in the blocking buffer (see Table 3) and centrifuge at 10000 × g for 1 min to pellet aggregates. 6. Remove blocking buffer from slides, and add primary antibody mix (50–100 μL per slice). Incubate for 2 h at RT or overnight at 4 °C in the dark in a humidity chamber (see Note 12). 7. Remove staining mix and wash slides in PBS for 5 min. Meanwhile, prepare secondary antibody mix (see Table 3), and centrifuge at 10000 × g for 1 min to pellet aggregates. 8. Incubate slides with staining mix (50–100 μL) for 1 h at RT. 9. Wash slides in PBS for 5 min. 10. Perform Sudan black quenching protocol [31] by placing slides in 0.7%w/v Sudan black solution for 6 min at RT (see Note 13). 11. Remove the slides from the solution and wash 2–3 times in 50% ethanol for 3 min. 12. Wash slides quickly in dH2O to remove salt.
226
Zhuangzhuang Liu et al.
13. Mount slides with mounting medium. Add 15 μL of the medium to a cover slip, and then invert this so that the mounting medium covers the tissue when you apply it to the slide (see Note 14). Let the mounting medium dry in the dark at RT if imaging the same day or at 4 °C if imaging later (see Note 14). 14. After staining and mounting samples can be imaged on a compatible confocal microscope (Fig. 8).
4
Notes 1. Buffer 1 and Buffer 2 can be made in bulk and stored at 4 °C for up to 3 months. 50 mL of buffer 2 is used per mouse for WD studies, whereas 20 mL buffer 2 is sufficient for mice on the CDA-HF diet. 2. We utilize C57Bl/6 mice for our MASLD studies. For the WD, we use only male mice as females do not gain weight, and hence the progression of the disease is slower. For the CDA-HF diet, both males and females can be used. Males progress slightly faster than females, but both sexes will develop MASH within 4–6 weeks. 3. For a flow cytometry analysis of the liver following the ex vivo digestion protocol, isolating only the left lobe of the liver is sufficient. However, for FACS the whole lobe should be taken to ensure sufficient cell counts. If only using one lobe, then 3 mL digestion media is sufficient. Following the digestion, this can be resuspended in 200–500 μL of FACS buffer, and 20 μL counting beads is sufficient. 4. This perfusion protocol was first described in [18]. If the liver does not go pale, then the needle is no inserted correctly in the vein. If this is the case, try to move the needle within the inferior vena cava. If this does not work, the needle can be inserted into the portal vein (trickier). If the perfusion is not working and cannot be rescued, it is best to stop here as this will result in low cell numbers and huge variation between mice. 5. During this perfusion step, the liver should swell indicating a good digestion. A lack of swelling should be recorded to identify possible causes of outliers in the data. The tube should be empty at the end of this step, but be careful not to allow any air to run through the liver. Be sure to stop the perfusion pump in time. 6. If the perfusion has worked well, there should be no pieces of liver left after this step, and the media in the petri dish should be cloudy with cells. 7. Whether there will be one or two steps for the staining depends on the panel used. If there are biotin-labeled antibodies, these
Macrophages in the Steatotic Liver
227
Fig. 8 Sudan black quenches CDA-HF diet-induced autofluorescence. C57Bl/6 male mice were fed either a standard diet (SD) or CDA-HF diet for 4, 8, or 12 weeks. Livers were perfused with antigen fix, embedded, cut, stained, and imaged on a LSM880. (a) Liver appearance after indicated diet and feeding periods stained using picrosirius red staining. Scale bar (black) is 100 μm. Images shown are representative of n = 4 mice per diet per timepoint. (b, c) Confocal microscopy showing expression of CD64 (magenta), CLEC4F (green), and TIM4 (red) without (B) or with (C) Sudan black quenching. Bottom row shows same samples as top row without staining with each color representing autofluorescence (AutoF) detected in each channel. Scale bar (white) is 100 μm
228
Zhuangzhuang Liu et al.
should be in a first step with the streptavidin included in the second step. Similarly, if you have weaker antibodies, these can be included in a first step to minimize competition for binding. We stain the first step in PBS, because we also add fixable live/ dead dyes to this step if we are using them (see Table 1). These dyes bind to free amines, and, thus, you cannot stain with these reagents in any buffers containing FCS. If using other live/ dead stains, e.g., Sytox blue as a dead cell discriminator (see Table 2), this can be added after the staining protocol is complete. It does not need to be washed off for analysis purposes. 8. The fixation solution contains PFA. All handling of this should thus be done in the fume hood. Separate tubing for the perfusion pump is recommended for antigen fix perfusions compared with enzyme perfusions described above. This protocol was first described in [18]. 9. The 34% sucrose solution is added as it serves as a cryoprotectant; it replaces the water, avoiding ice crystal formation during freezing steps. A stock solution of sucrose can be made, but keep it sterile and ideally freeze in aliquots to prevent bacterial contamination which will damage the tissue integrity. 10. The 34% sucrose solution will form a film around the tissue which makes it difficult to cut; try to remove it by dabbing the tissue slice gently on tissue paper before embedding. 11. Be careful when circling the pieces of tissue with the hydrophobic pen that you do not touch the tissue. Moreover it is best to leave some space between the tissue and the pen and to wait until the pen circle is completely dry before proceeding. Otherwise the pen can interfere with some stains. 12. Staining overnight at 4 °C gives less background than staining for 2 hrs at RT. 13. 0.7%w/v Sudan black solution can be stored at RT in the dark for 2 weeks. If reusing an older batch, stir for 10 min before using to ensure no particles have formed. Additionally Sudan black solution can be reused up to three times. Following quenching tissue color will range from light grey to black depending on the lipid content. Sudan black quenching is compatible with most commonly used fluorochromes used for confocal microscopy including AF488, Cy3, and AF647. It should be noted that Sudan black quenching causes some background signal in the Cy3 and AF647 channel. This is marginal and does not interfere with staining but should be taken into consideration especially if considering staining low abundant proteins. Sudan black quenching weakens the overall fluorescence of fluorochromes. Hence, although possible to use with directly conjugated antibodies, we recommend using unconjugated antibodies with fluorescently-labeled secondary antibodies to boost signals.
Macrophages in the Steatotic Liver
229
14. Mounting media are very viscous, pipette them very slowly, and avoid getting any air bubbles on the cover slip or between the cover slip and the tissue. After mounting the staining is stable for ~1 week, so be sure to image the sections in that timeframe.
Acknowledgments We thank Dr. Johnny Bonnardel who developed the in vivo perfusion protocol and the microscopy protocols on healthy liver tissue that were adapted here for use on the steatotic livers. References 1. Byrne CD, Targher G (2015) NAFLD: a multisystem disease. J Hepatol 62:S47–S64. https://doi.org/10.1016/j.jhep.2014.12.012 2. Buzzetti E, Pinzani M, Tsochatzis EA (2016) The multiple-hit pathogenesis of non-alcoholic fatty liver disease (NAFLD). Metab Clin Exp 65:1038–1048. https://doi.org/10.1016/j. metabol.2015.12.012 3. Guilliams M, Bonnardel J, Haest B et al (2022) Spatial proteogenomics reveals distinct and evolutionarily conserved hepatic macrophage niches. Cell 185:379–396.e38. https://doi. org/10.1016/j.cell.2021.12.018 4. Remmerie A, Martens L, Thone´ T et al (2020) Osteopontin expression identifies a subset of recruited macrophages distinct from Kupffer cells in the fatty liver. Immunity 53:641–657. e14. https://doi.org/10.1016/j.immuni. 2020.08.004 5. Devisscher L, Scott CL, Lefere S et al (2017) Non-alcoholic steatohepatitis induces transient changes within the liver macrophage pool. Cell Immunol 322:74–83. https://doi.org/10. 1016/j.cellimm.2017.10.006 6. Tran S, Baba I, Poupel L et al (2020) Impaired Kupffer cell self-renewal alters the liver response to lipid overload during non-alcoholic steatohepatitis. Immunity 53: 627–640.e5. https://doi.org/10.1016/j. immuni.2020.06.003 7. Xiong X, Kuang H, Ansari S et al (2019) Landscape of intercellular crosstalk in healthy and NASH liver revealed by single-cell Secretome gene analysis. Mol Cell 75:644–660.e5. https://doi.org/10.1016/j.molcel.2019. 07.028 8. Seidman JS, Troutman TD, Sakai M et al (2020) Niche-specific reprogramming of epigenetic landscapes drives myeloid cell diversity in nonalcoholic steatohepatitis. Immunity.
https://doi.org/10.1016/j.immuni.2020. 04.001 9. Krenkel O, Hundertmark J, Abdallah AT et al (2019) Myeloid cells in liver and bone marrow acquire a functionally distinct inflammatory phenotype during obesity-related steatohepatitis. Gut. https://doi.org/10.1136/gutjnl2019-318382 10. Matsumoto M, Hada N, Sakamaki Y et al (2013) An improved mouse model that rapidly develops fibrosis in non-alcoholic steatohepatitis. Int J Exp Pathol 94:93–103. https://doi. org/10.1111/iep.12008 11. Ganz M, Bukong TN, Csak T et al (2015) Progression of non-alcoholic steatosis to steatohepatitis and fibrosis parallels cumulative accumulation of danger signals that promote inflammation and liver tumors in a high fatcholesterol-sugar diet model in mice. J Transl Med 13:193. https://doi.org/10.1186/ s12967-015-0552-7 12. Caballero F, Ferna´ndez A, Matı´as N et al (2010) Specific contribution of methionine and choline in nutritional nonalcoholic steatohepatitis: impact on mitochondrial S-Adenosyl-l-Methionine and Glutathione*. J Biol Chem 285:18528–18536. https://doi.org/ 10.1074/jbc.m109.099333 13. Sugasawa T, Ono S, Yonamine M et al (2021) One week of CDAHFD induces steatohepatitis and mitochondrial dysfunction with oxidative stress in liver. Int J Mol Sci 22:5851. https:// doi.org/10.3390/ijms22115851 14. Hoffmann C, Djerir NEH, Danckaert A et al (2020) Hepatic stellate cell hypertrophy is associated with metabolic liver fibrosis. Sci Rep-UK 10:3850. https://doi.org/10.1038/ s41598-020-60615-0 15. Rokugawa T, Konishi H, Ito M et al (2018) Evaluation of hepatic integrin αvβ3 expression
230
Zhuangzhuang Liu et al.
in non-alcoholic steatohepatitis (NASH) model mouse by 18F-FPP-RGD2 PET. EJNMMI Res 8:40. https://doi.org/10. 1186/s13550-018-0394-4 16. Ikawa-Yoshida A, Matsuo S, Kato A et al (2017) Hepatocellular carcinoma in a mouse model fed a choline-deficient, L-amino aciddefined, high-fat diet. Int J Exp Pathol 98: 221–233. https://doi.org/10.1111/iep. 12240 17. Mederacke I, Dapito DH, Affo` S et al (2015) High-yield and high-purity isolation of hepatic stellate cells from normal and fibrotic mouse livers. Nat Protoc 10:305–315. https://doi. org/10.1038/nprot.2015.017 18. Bonnardel J, T’Jonck W, Gaublomme D et al (2019) Stellate cells, hepatocytes, and endothelial cells imprint the Kupffer cell identity on monocytes colonizing the liver macrophage niche. Immunity 51:638–654.e9. https://doi. org/10.1016/j.immuni.2019.08.017 19. Wake K (2004) Karl Wilhelm Kupffer and his contributions to modern hepatology. Comp Hepatol 3:S2. https://doi.org/10.1186/ 1476-5926-2-s1-s2 20. Kupffer C (1876) Ueber Sternzellen der Leber: Briefliche Mittheilung an Prof. Waldeyer. Arch Mikr Anat 12:353–358. https://doi.org/10. 1007/bf02933897 21. Browicz T (1899) Ueber intravascul€are Zellen in den Blutcapillaren der Leberacini. Arch Mikrosk Anat 55:420–426. https://doi.org/10. 1007/bf02977740 22. Sro´dka A, Gryglewski RW, Szczepariski W (2006) Browicz or Kupffer cells? Pol J Pathol 57:183–185 23. Guilliams M, Scott CL (2022) Liver macrophages in health and disease. Immunity 55: 1515–1529. https://doi.org/10.1016/j. immuni.2022.08.002 24. Scott CL, Zheng F, Baetselier PD et al (2016) Bone marrow-derived monocytes give rise to self-renewing and fully differentiated Kupffer
cells. Nat Commun 7:10321. https://doi. org/10.1038/ncomms10321 25. Sierro F, Evrard M, Rizzetto S et al (2017) A liver capsular network of monocyte-derived macrophages restricts hepatic dissemination of intraperitoneal bacteria by neutrophil recruitment. Immunity 47:374–388.e6. https://doi. org/10.1016/j.immuni.2017.07.018 26. Daemen S, Gainullina A, Kalugotla G et al (2021) Dynamic shifts in the composition of resident and recruited macrophages influence tissue remodeling in NASH. Cell Rep 34: 108626. https://doi.org/10.1016/j.celrep. 2020.108626 27. Simone GD, Andreata F, Bleriot C et al (2021) Identification of a Kupffer cell subset capable of reverting the T cell dysfunction induced by hepatocellular priming. Immunity 54:2089– 2100.e8. https://doi.org/10.1016/j.immuni. 2021.05.005 28. Ble´riot C, Barreby E, Dunsmore G et al (2021) A subset of Kupffer cells regulates metabolism through the expression of CD36. Immunity 54:2101–2116.e6. https://doi.org/10.1016/ j.immuni.2021.08.006 29. Iannacone M, Ble´riot C, Andreata F et al (2022) Response to contamination of isolated mouse Kupffer cells with liver sinusoidal endothelial cells. Immunity 55:1141–1142. https://doi.org/10.1016/j.immuni.2022. 06.012 30. Hume DA, Offermanns S, Bonnavion R (2022) Contamination of isolated mouse Kupffer cells with liver sinusoidal endothelial cells. Immunity 55:1139–1140. https://doi. org/10.1016/j.immuni.2022.06.010 31. Saif M, Kwanten WJ, Carr JA et al (2020) Non-invasive monitoring of chronic liver disease via near-infrared and shortwave-infrared imaging of endogenous lipofuscin. Nat Biomed Eng 4:801–813. https://doi.org/10. 1038/s41551-020-0569-y
Chapter 16 Isolation, Ex Vivo Expansion, and Lentiviral Transduction of Alveolar Macrophages Clara Jana-Lui Busch, Sethuraman Subramanian, Javier Linares, Je´re´my Favret, Ridzky Anis Advent Yuda, and Michael H. Sieweke Abstract Alveolar macrophages (AM) are resident macrophages of the lung and play important roles in the maintenance of tissue homeostasis as well as host defense. Here, we describe how they can be harvested from murine lungs, expanded in vitro, and transduced with lentiviral vectors. Key words Alveolar macrophage, Lungs, Self-renewal, Bronchoalveolar lavage, Primary cell culture, Tissue-resident macrophage, Lentiviral transduction, Conditioned medium, GM-CSF, Genetic manipulation, Genetic engineering
1
Introduction Most functional studies on macrophage biology have been performed with macrophage-like cell lines derived from cancerous material such as RAW264.7 or J774A.1 cells or oncogenetransformed macrophages, all of which can be expanded readily in culture. The use of primary macrophages, however, has been hindered by the limited number of cells that can be extracted from tissues as well as the inability to amplify them in vitro. Here we provide a protocol for isolation of alveolar macrophages (AM) from murine lungs and describe how they can be long-term expanded in vitro using GM-CSF-conditioned medium. We also describe how expanded alveolar macrophages (exAM) can be genetically engineered using lentiviral transduction. This protocol comprises lentivirus production using calcium phosphate precipitation, concentration by ultracentrifugation, titration of lentiviral particles overexpressing fluorescent proteins (e.g., GFP or mCherry), and transduction of lentiviral particles into exAM. Together, the methods outlined here allow the investigation of macrophage functions in primary cells, e.g., the role of Sirt1 in AM self-renewal [1].
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_16, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
231
232
2
Clara Jana-Lui Busch et al.
Materials Prepare all solutions using sterile or sterile-filtered reagents and store at 4 °C until use.
2.1 Conditioned Medium Containing Murine GM-CSF
1. T25, T75, and T175 tissue culture (TC)-treated flasks. 2. 1-L sterile media bottle. 3. J558L cells modified with a murine GM-CSF-expressing construct [2]. 4. IMDM basal medium: IMDM supplemented with 4 mM of L-glutamine. 5. J558L recovery medium: 20% heat-inactivated fetal bovine serum (FBS), 100 U/mL of penicillin-streptomycin in IMDM basal medium. 6. J558L selection medium: 5% FBS, 100 U/mL of penicillinstreptomycin, 1 mg/mL of G418 (Geneticin) in IMDM basal medium. 7. J558L growth medium: 5% FBS, 100 U/mL of penicillinstreptomycin in IMDM basal medium. 8. Mouse GM-CSF ELISA kit (e.g., DuoSet, R&D Systems).
2.2
AM Harvest
1. Dissection tools (forceps, fine scissors). 2. BAL buffer: 2 mM of EDTA (0.5 M of EDTA stock solution, pH 8.0), 0.5% FBS in 1× DPBS, sterile-filter using a 0.2 μm filter, and keep at 4 °C until use (see Note 1). 3. Red blood cell lysis buffer (e.g., hemolysis buffer, Morphisto). 4. 18-G cannula (blunt-end), 1 mL syringe. 5. 70 μm sterile cell strainer.
2.3
AM Culture
1. Complete medium: 2 mM of GlutaMAX™, 1 mM of sodium pyruvate, 100 U/mL of penicillin-streptomycin, 10% FBS in RPMI 1640 without glutamine, sterile-filter using a 0.22 μm filter, and keep at 4 °C until use. 2. 50 mg/mL of gentamicin sulphate solution. 3. Conditioned medium containing GM-CSF (for preparation, see Subheadings 2.1 and 3.1). 4. AM culture medium: roughly 1% conditioned medium containing GM-CSF in complete medium (see Note 11), depending on the production batch (see Subheading 3.1, step 45). Warm-up required amount in water bath to 37 °C before use. 5. Detachment medium: Accutase (e.g., ESGRO Complete™, Merck Millipore) with 1 mM of EGTA. Aliquot and freeze at -20 °C. Warm-up in water bath to 37 °C before use.
Ex Vivo Expansion of Alveolar Macrophages
2.4 AM Cryopreservation
233
1. Freezing medium: 90% FBS, 10% DMSO. 2. Cryovials. 3. Cell freezing container (e.g., Mr. Frosty™ freezing container, Nalgene®).
2.5 Lentiviral Production and Transduction
1. HEK293T cells. 2. T75 or T175 flasks. 3. 12-well and 15 cm tissue culture (TC)-treated dishes. 4. 0.01% w/v Poly-L-lysine solution (10× stock solution): Dilute to 0.001% w/v in sterile water before use. 5. 0.5% Trypsin-EDTA (10× stock solution): Dilute to 0.05% (1×) in 1× DPBS before use. 6. Chloroquine diphosphate: Prepare a 25 mM stock solution by adding 19.39 mL of water to 250 mg of chloroquine. Vortex, sterilize by filtering through a 0.22 μm filter, and store at 4 °C. 7. 250 mM of CaCl2 solution: 3.7 g of calcium chloride dihydrate in 100 mL of deionized water. Sterilize by filtering through a 0.22 μm filter (see Note 2). 8. 2× HEPES-buffered saline (HBS) solution, pH 7.05: 1.6 g of NaCl, 0.07 g of KCl, 0.04 g of Na2HPO4, 0.2 g of sucrose, and 1 g of HEPES in 100 mL of deionized water. Sterilize by filtering through a 0.22 μm filter (see Note 2). 9. 1M HEPES buffer 10. Complete DMEM medium: 10% FBS, 2 mM of GlutaMAX™, 100 U/mL of penicillin-streptomycin, 1 mM of sodium pyruvate in DMEM high glucose. 11. Lentiviral production medium: Complete DMEM medium supplemented with 25 mM of HEPES. 12. Polypropylene ultracentrifuge tubes. 13. Preparative Beckman Coulter ultracentrifuge with SW 32 Ti swinging-bucket rotor (or similar). 14. Fixation buffer (e.g., BD Cytofix™ Fixation Buffer, BD Biosciences). 15. FACS buffer: 1 mM of EDTA, 0.5% FBS in 1x DPBS. 16. 5 mL polystyrene round-bottom FACS tubes with cell strainer cap. 17. Flow cytometer. 18. 1–4 μg/μL of second-generation lentiviral envelope plasmid, encoding for VSV-G glycoprotein, e.g., pMD2.G (Addgene #12259). 19. 1–4 μg/μL of second-generation lentiviral packaging plasmid, encoding for Gag/Pol/Tat/Rev. genes, e.g., psPAX2 (Addgene #12260).
234
Clara Jana-Lui Busch et al.
20. 1–4 μg/μL of second- or third-generation lentiviral transfer plasmid. 21. 12% sodium hypochlorite solution: Dilute 1:10 in water (final concentration 1.2%) before use. 2.6 Common Solutions and Equipment
1. Hemocytometer or Coulter-based automated cell counter system. 2. 0.4% trypan blue solution. 3. Water bath set to 37 °C. 4. Cell culture incubator at 37 °C, 5% CO2. 5. Serological pipettes. 6. 1× DPBS without calcium chloride, without magnesium chloride, pH 7.2. 7. Low-protein-binding filter, e.g., Whatman cellulose acetate (CA) or polyethersulfone (PES) filter unit, FP30/0.45 μm (or bottle-top vacuum filter, pore size 0.45 μm). 8. Whatman Puradisc CA sterile filter, pore size 0.22 μm.
3
Methods
3.1 Preparation of Conditioned Medium Containing Murine GM-CSF
1. Warm up 10 mL of IMDM in a 15 mL conical tube in a water bath set to 37 °C. 2. Transfer a cryovial with 1 × 106 frozen J558L cells from longterm storage to a cell freezing container placed on dry ice to prevent rapid cell thawing. 3. Partially submerge the cryovial for 30 s in the water bath at 37 °C until a small piece of ice crystals remains in the cryovial. 4. Pipet 500 μL of the prewarmed IMDM drop-wise into the cryovial. 5. Transfer the cell suspension into the 15 mL conical tube containing 9.5 mL of prewarmed IMDM. 6. Pellet the cells by centrifugation at 475 × g for 3 min at room temperature (RT), discard supernatant. 7. Resuspend 1 × 106 cells in 5 mL of J558L recovery medium, and seed the cells in a T25 TC-treated flask (corresponding to a cell density of 2 × 105 cells/mL). 8. Incubate for 2 days at 37 °C, 5% CO2. 9. Detach the cells mechanically by rocking the flask vigorously followed by pipetting several times onto the bottom of the flask to wash out remaining loosely attached cells (see Note 3). 10. Transfer cell suspension into a 15 mL conical tube.
Ex Vivo Expansion of Alveolar Macrophages
235
11. Pellet the cells by centrifugation at 475 × g for 3 min at RT, and discard supernatant. 12. Resuspend the cell pellet in 1 mL of J558L recovery medium. 13. Count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system. 14. Resuspend 2 × 106 cells in 5 mL J558L recovery medium, and seed the cells in a T25 TC-treated flask (corresponding to a cell density of 4 × 105 cells/mL). 15. Incubate for 2 days at 37 °C, 5% CO2. 16. Detach and collect the cells in a 15-mL conical tube as in steps 9 and 10. 17. Pellet the cells by centrifugation at 475 × g for 3 min at RT, and discard supernatant. 18. Resuspend the cell pellet in 1 mL of J558L recovery medium. 19. Count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system. 20. Resuspend 4 × 106 cells in 10 mL of J558L recovery medium, and seed the cells in a T75 TC-treated flask (corresponding to a cell density of 4 × 105 cells/mL). 21. Incubate for 2 days at 37 °C, 5% CO2. 22. Detach and collect the cells in a 15 mL conical tube as in steps 9 and 10. 23. Pellet the cells by centrifugation at 475 × g for 3 min at RT, and discard supernatant. 24. Resuspend the cells in 1 mL of J558L selection medium. 25. Count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system. 26. Resuspend 8–10 × 106 cells in 20–25 mL of J558L selection medium, and seed the cells in a T175 TC-treated flask (corresponding to a cell density of 4 × 105 cells/mL). 27. Incubate for 2 days at 37 °C, 5% CO2. 28. Detach and collect the cells in a 15 mL conical tube as in steps 9 and 10. 29. Pellet the cells by centrifugation at 475 × g for 3 min at RT, and discard supernatant. 30. Repeat steps 24–29 twice (see Note 4). 31. Resuspend 10 × 106 cells in 25 mL of J558L growth medium without Geneticin, and seed the cells in a T175 TC-treated flask (corresponding to a cell density of 4 × 105 cells/mL).
236
Clara Jana-Lui Busch et al.
Fig. 1 Preparation of conditioned medium containing murine GM-CSF. (a) Three flasks with modified J558L cells in growth medium, placed tilted in the incubator to avoid contact of medium with the cap or filter. (b) Representative image of a flask with modified J558L cells after 10 days of incubation. (c) Images of cuvettes with control J558L growth medium (1) or J558L growth medium after 10 days, from three independent productions (2,3,4)
32. Change medium every 2 days, and expand the culture for ~1 week up to ten T175 TC-treated flasks, each with a density of 1 × 106 cells/mL or 25 × 106 cells per flask. Passage the cells when they reach a density of 1 × 106 cells/mL or 25 × 106 cells per flask (repeat steps 24–29) (see Note 5). 33. Detach and collect the cells in a 50 mL conical tube as in steps 9 and 10. 34. Pellet the cells by centrifugation at 475 × g for 3 min at RT, and discard supernatant. 35. Resuspend the cell pellet in 20–25 mL of J558L growth medium. 36. Count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system (see Note 6). 37. Resuspend 240 × 106 cells in a sterile 1 L media bottle containing 600 mL of J558L growth medium (corresponding to a cell density of 4 × 105 cells/mL). 38. Seed 200 mL of the cell suspension per T175 TC-treated flask (in total three flasks) (see Note 7). 39. Incubate for 10 days at 37 °C and 5% CO2 without changing the medium (see Fig. 1 and Note 8).
Ex Vivo Expansion of Alveolar Macrophages
237
40. Transfer the medium of the flasks including the floating cells into 50 mL conical tubes. Do not detach remaining cells. 41. Pellet the cells by centrifugation at 475 × g for 5 min at RT, and do not discard the supernatant containing GM-CSF (conditioned medium). 42. Transfer the conditioned medium to a sterile 1 L media bottle, and filter the media through a 0.45 μm filter to remove any contaminating cells and cellular debris (see Note 9). 43. Aliquot the conditioned medium into several 15-mL conical tubes (see Note 10). 44. Store aliquots for up to 1 year at -70 °C to -80 °C. 45. Keep a small aliquot at 4 °C to measure GM-CSF concentration by ELISA following the manufacturer’s instructions (see Note 11). 3.2 Isolation of BAL AM
1. Prewarm the BAL buffer in a water bath at 37 °C (see Note 1). 2. Euthanize mouse by cervical dislocation without damaging the trachea (see Note 12). 3. Expose both lungs and trachea using dissection tools (see Note 12). 4. Using fine scissors, make a small incision in the trachea below the larynx. 5. Insert the 18-G cannula about 5 mm into the trachea using the small incision. 6. Fill up 1 mL syringe with 1 mL of warm BAL buffer, and inject volume into the lungs via the cannula. 7. Recover BAL buffer, and pass through 70 μm cell strainer situated on top of a 15 mL conical tube filled with 3 mL of complete medium. 8. Repeat steps 6 and 7 for 9 additional milliliters and pool aliquots. 9. Pellet cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant. 10. Add 1 mL of hemolysis buffer and incubate 2 min at RT. 11. Stop lysis by adding 9 mL of complete medium. 12. Pellet cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant. 13. Resuspend pellet in 500 μL cold BAL buffer, and determine cell number manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system, under consideration of live and dead cells. 14. Proceed to downstream applications (e.g., long-term culture described in Subheadings 3.3 and 3.4).
238
Clara Jana-Lui Busch et al.
3.3 Establishing Culture
1. Pellet cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant. 2. Resuspend cells at a concentration of 1 × 106 per mL in prewarmed AM culture medium. 3. Seed a non-treated six-well plate with cells at a density of 3–4 × 105 cells per well in 3 mL of prewarmed AM culture medium. 4. Add gentamicin to the AM culture medium at a concentration of 50 μg/mL (1:1000 of stock solution) (see Note 13). 5. Transfer plate to an incubator at 37 °C, 5% CO2. 6. On the next day, replace the medium containing floating cells with 3 mL of fresh prewarmed AM culture medium without gentamicin (see Note 14). 7. Transfer plate to an incubator for 2 days at 37 °C, 5% CO2. 8. To change medium, transfer medium containing floating cells with a serological pipette into a 15 mL conical tube (see Note 15). 9. Add 2 mL of prewarmed AM culture medium to adherent cells in the well to prevent cells from drying-out. 10. Pellet floating cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant. 11. Resuspend pellet in 1 mL of prewarmed AM culture medium and combine with adherent cells. 12. Repeat steps 8–11 every 2 days until reaching 70–80% confluency (see Note 16). 13. To collect cells from a confluent well, transfer supernatant containing floating cells into a 15 mL conical tube on ice. 14. Add 750 μL of detachment medium directly to one well of a six-well plate without washing the well. Incubate 10–30 min at 37 °C until cells start to detach (see Note 17). 15. Pipet detachment medium with a 1000 μL pipette onto plastic bottom to gently wash off cells. 16. Pool detached cells with floating cells in the 15 mL conical tube on ice. 17. Pellet cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant. 18. Resuspend the cell pellet in 1 mL of prewarmed AM culture medium, and assess cell count manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system. Determine live cell count. Typically, in a healthy culture, at least 95% live cells are obtained.
Ex Vivo Expansion of Alveolar Macrophages
239
Table 1 Seeding densities for long-term AM culture Tissue culture dish
Surface area (cm2)
Seeding density
24-well
1.9
0.05–0.1 × 106
12-well
3.5
0.1–0.2 × 106
6-well
9.6
0.3–0.4 × 106
10-cm
56.7
1.6–2.0 × 106
19. Replate cells at a density of 3–4 × 105 live cells per well of a six-well plate in 3 mL of prewarmed AM culture medium (as in step 3 in this section) or 1.1–1.5 × 106 per 10-cm Petri dish in a 10 mL prewarmed AM culture medium (see Note 18). 3.4 Long-Term AM Culture and Cryopreservation of exAM
3.4.1 AM
Cryopreservation of
The isolation and culture of freshly isolated AM have been described in Subheadings 3.2 and 3.3. For long-term maintenance and amplification of exAM, GM-CSF-containing medium must be changed every 2 days as described in Subheading 3.3. Based on the desired scale of the experiment, exAM can be seeded and maintained in different tissue culture well formats (see Table 1). We recommend the six-well or the 10 cm Petri dish, which allows for optimal cell numbers and ease when handling long-term culture. exAM could be kept in culture for at least 10 months without losing their proliferative capacity [3–5]. When exAM are in a proliferative state in culture, they can be detached and replated for experiments and assays or taken through freeze-thaw cycles without any compromise to the growth capacity upon thawing [5]. 1. Prepare and label cryovials to store exAM (see Note 19). 2. When exAM are actively proliferating, detach and count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system (see Subheading 3.3 and Note 20). 3. Pellet cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant. 4. Resuspend 2 × 106 exAM in 500 μL of freezing medium, and transfer into a prelabelled cryovial (see Note 21). 5. Transfer the cryovials into a cell freezing container, and store overnight at -80 °C and then in liquid N2 containers for longterm storage (see Note 22).
240 3.4.2
Clara Jana-Lui Busch et al. Thawing of exAM
1. Warm up 10 mL of AM medium in a 15 mL conical tube in a water bath set at 37 °C. 2. Transfer the cryovial from long-term storage to a cell freezing container placed on dry ice to prevent rapid cell thawing. 3. Partially submerge the cryovial containing 2 × 106 exAM for 30 s in the water bath at 37 °C until a small piece of ice crystals remains in the cryovial. 4. Pipet 500 μL of prewarmed AM medium drop-wise onto the cells in the cryovial. 5. Transfer the entire volume of the cryovial back to the 15 mL conical tube containing 9.5 mL of prewarmed AM medium. 6. Pellet cells by centrifugation at 300 × g for 5 min at 4 °C, and discard supernatant containing freezing medium with DMSO. 7. Resuspend the cells in 1 mL of prewarmed AM medium, and assess cell count and viability manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system (see Note 23). 8. Replate cells at a density of 3–4 × 105 live cells per well of a six-well plate in 3 mL prewarmed AM culture medium (see Table 1 and Note 24). 9. Transfer plate to an incubator overnight at 37 °C, 5% CO2. 10. Discard supernatant containing floating (dying) cells, and add 3 mL of fresh prewarmed AM culture medium (see Note 25). 11. Transfer plate to an incubator for 2 days at 37 °C, 5% CO2 (see Note 26). 12. Passage exAM when confluent (see Subheading 3.3). Maintain an optimal seeding density at each passage (see Table 1).
3.5 LentiviralVector-Mediated Gene Transfer in AM
The lentiviral transduction protocol described here is adapted for exAM manipulation and comprises virus production, concentration by ultracentrifugation, titration of lentiviral particles overexpressing fluorescent proteins (e.g., GFP or mCherry), and transduction of lentiviral particles into exAM. It combines a third-generation lentiviral transfer plasmid with a second-generation packaging system and allows the generation of high-titer lentivirus vectors. Lentivirus production, enrichment, and titration are based on previously published procedures [6, 7]. Handling of lentiviral vectors should be performed in a biosafety level 2 laboratory area and under a biosafety laminar hood when performing any open manipulation of biological materials. Special protective clothing such as lab coats, gloves, face, and eye protection must be worn at all times.
Ex Vivo Expansion of Alveolar Macrophages 3.5.1 Lentivirus Production and Harvesting
241
1. Thaw HEK293T cells following the procedure in Subheading 3.1, steps 1–6 using complete DMEM. 2. Maintain HEK293T cells in T75 or T175 TC-treated flasks in complete DMEM medium by passaging every 2–3 days when roughly 80% confluent (see Notes 27 and 28). 3. Coat a 15 cm TC-treated dish with 10 mL of 0.001% Poly-Llysine solution for 30 min at RT. Wash the dish with 1× DPBS (see Note 29). 4. Harvest the HEK293T cells by adding 5 mL per T175 TC-treated flask of 0.05% trypsin-EDTA, and incubate at 37 °C, 5% CO2, for 3–5 min. Add 10 mL complete DMEM medium to each flask to stop trypsinization (see Note 30). 5. Transfer the cell suspension into 50 mL conical tubes, and pellet the cells by centrifugation at 300 × g for 5 min at RT, and discard supernatant. 6. Resuspend cell pellet in 10 mL of complete DMEM medium, and count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system. 7. Plate 7–8 × 106 HEK293T cells in a 15 cm TC-treated dish (coated with Poly-L-lysine) in a total volume of 20–30 mL complete DMEM medium, and transfer plate to an incubator overnight at 37 °C, 5% CO2 (see Note 31). 8. Add 20–30 μL of 25 mM chloroquine solution to the medium of each dish (resulting in a final concentration of 25 μM), and mix by gently rocking the plate (see Note 32). 9. In a 15 mL conical tube, combine and mix thoroughly 20.8 μg of lentiviral transfer plasmid, 20.8 μg of packaging plasmid psPAX2, and 10.4 μg of envelope plasmid pMD2.G (corresponding to an equimolar ratio) in 3 mL of 250 mM CaCl2 solution (see Notes 33 and 34). 10. Mix the DNA/CaCl2 solution vigorously at 1600 rpm on a vortex mixer, and simultaneously add drop-wise 3 mL of 2× HBS solution (pH 7.05) to each tube (see Note 35). 11. Add 6 mL of the transfection mix to each 15 cm dish dropwise, gently rock the plate without swirling, and transfer plate to an incubator overnight at 37 °C, 5% CO2 (see Note 36). 12. Remove the medium, add 20–30 mL of warm lentiviral production medium to each dish, and transfer the plate to an incubator overnight at 37 °C, 5% CO2 (see Note 37). 13. On the next day (40–44 h after transfection), collect the supernatant containing lentiviral vectors in 50 mL conical tubes, and centrifuge at 500 × g for 5–7 min at 4 °C to remove cells and large cell debris (see Notes 38 and 39).
242
Clara Jana-Lui Busch et al.
14. Filter the supernatant using a 0.45 μm-low-protein-binding filter (CA or PES filter) attached to a syringe or a sterile media bottle to remove cellular debris. 15. Lentivirus-containing supernatant can be used immediately for transduction or aliquoted and stored at -80 °C. Alternatively, the supernatant can be concentrated to increase the titer (see Note 40). 3.5.2 Concentration of Lentiviral Vectors by Ultracentrifugation
1. Transfer aliquots of up to 36 mL of filtered lentiviruscontaining supernatant (from Subheading 3.5.1, step 15) into polypropylene ultracentrifuge tubes, and place the tubes gently inside ultracentrifuge tube holders. 2. Carefully weigh the tubes using a balance located under a laminar hood, and adjust the weight of each tube by adding filtered supernatant or 1× DPBS such that the weight differences between the tubes are less than 0.1 g. 3. Ultracentrifuge them at 110,000 × g for 2.5 h at 4 °C, with maximum acceleration and slow deceleration (see Note 41). 4. Carefully remove the tubes from the ultracentrifuge, and aspirate supernatant by vacuum suction. A tiny virus pellet should be visible as a translucent spot at the bottom of the tube. 5. Immediately add 20–40 μL ice-cold 1× DPBS (without calcium and magnesium) to each tube, and resuspend the pellet by gently pipetting up and down (see Note 42). 6. Pool all concentrated lentivirus from the same vector into a 1.5 mL tube, and using a mini-centrifuge, fast spin at maximum speed for a few seconds to pellet residual membrane fragments and cellular debris. 7. Prepare aliquots in 0.5 mL tubes or cryovials (concentrated virus can be aliquoted in 20–50 μL portions, depending on the total volume of virus-containing liquid). 8. Snap-freeze using dry ice or liquid nitrogen and transfer tubes to -80 °C for long-term storage (see Note 43).
3.5.3 Lentivirus Titration by Flow Cytometry
1. For lentivirus titration, seed 1–3 × 105 HEK293T cells per well in 12-well TC-treated plates containing 1 mL complete DMEM medium 1 day before transduction, and incubate overnight at 37 °C, 5% CO2 (see Note 44). 2. To determine the number of cells at the time of transduction (relevant for calculating the viral titer accurately, see step 16 in this section), perform following steps 3–7. 3. In two wells, remove the medium, and detach the cells by adding 500 μL of 0.05% trypsin-EDTA, and incubate for 3–5 min at 37 °C, 5% CO2.
Ex Vivo Expansion of Alveolar Macrophages
243
4. Add 1 mL of complete DMEM medium to each well to stop trypsinization, and collect the cell suspension into 15 mL conical tubes. 5. Pellet the cells by centrifugation at 300 × g for 3 min at RT, and discard supernatant. 6. Resuspend the cell pellet in 300 μl of complete DMEM medium. 7. Count the cells manually with a hemocytometer (e.g., trypan blue staining) or a Coulter-based automated cell counter system, and calculate the number of cells per well. 8. For transduction, thaw the lentiviral particles at RT, and keep on ice before use. Remove the medium from the wells, and add the appropriate amounts of lentiviral vectors in total volume of 1 mL complete DMEM medium, in duplicates. Additionally, maintain two wells with cells without adding virus as negative control (non-transduced cells) (see Note 45). 9. 12–24 h after transduction, aspirate the lentivirus-containing medium, and add 1 mL of warm complete DMEM medium, and incubate for at least 24 h at 37 °C, 5% CO2 (see Note 46). 10. Remove the medium, detach HEK293T cells by adding 500 μL of 0.05% trypsin-EDTA, and incubate for 3–5 min at 37 °C, 5% CO2. Add 1 mL fresh medium to each well to stop trypsinization. 11. Transfer the cell suspension into 15 mL conical tubes, and pellet the cells by centrifugation at 300 × g for 3 min at RT, and discard supernatant. 12. Fix cells by resuspending in 250 μL of fixation buffer (containing paraformaldehyde), and incubate for 20 min at 4 °C (see Note 47). 13. Add 3 mL of 1× DPBS to fixed cells, and wash the cells by centrifugation at 300 × g for 5 min at 4 °C. 14. Resuspend cells in 300 μL FACS buffer and transfer in FACS tubes. 15. Determine the percentage of fluorescent reporter-expressing cells (FP) by flow cytometry, using non-transduced (negative control) fixed cells for proper gate setting. 16. Calculate lentivirus titer (vector particles numbers, also called transducing units or TU, per mL) according to the following formula (see Note 48): × n ° cells per well × dilution factor Titer (TU/mL) = ½FP ðin%Þ=100 lentiviral volume per well ðmLÞ:
244
Clara Jana-Lui Busch et al.
3.5.4 Lentivirus Transduction of exAM
1. Seed exAM at a density of 3–4 × 105 cells per well of a six-well plate (or 1.1–1.5 × 106 per 10 cm Petri dish) in 3 mL prewarmed AM culture medium. Transfer plate to an incubator at 37 °C, 5% CO2. 2. On the next day, aspirate the medium, and replace with fresh 3 mL of prewarmed AM culture medium. 3. Thaw the lentiviral particles at RT and keep on ice before use. Transduce the exAM by adding lentiviral vector at a multiplicity of infection (MOI) of 5–10 (e.g., An MOI of 5 corresponds to 1.5–2× 106 TU for cells in a six-well plate) (see Note 49). 4. Slightly swirl the plate to evenly distribute the viruses within the medium, and transfer plate to an incubator at 37 °C, 5% CO2. 5. At 12–16 h after transduction, remove the lentiviruscontaining medium, and gently replace with 3 mL of fresh AM culture medium and transfer plate to an incubator at 37 °C, 5% CO2 (see Note 50). 6. Maintain transduced exAM as described in the Subheadings 3.3 and 3.4 until cells are to be assayed (see Note 51). 7. When using fluorescent protein-expressing lentiviral vectors, the expression of fluorescent reporter protein can be detected by epifluorescence and used as an indicator of the transgene expression in transduced cells (see Fig. 2a). In addition, the percentage of transduced exAM might be determined by flow cytometry (see Fig. 2b). Lentiviral vectors are integrated into the host genome, allowing for the stable expression of the transgenes by the transduced cells (see Note 52).
4
Notes 1. BAL buffer must always be prewarmed in a water bath at 37 °C and must be kept at this optimal temperature when performing the lavage as this dramatically increases the recovery of cells [4]. For convenience, we use a digital dry bath or block heater close to the bench where BAL is performed. 2. It is recommended to prepare freshly at the day of transduction. 3. The cells are semi-adherent and will attach to the bottom of the flask at higher cell densities. 4. A total of at least 10 × 106 total cells should be obtained at this step. 5. The cells may be expanded additionally in one to five T175 TC-treated flasks for cryopreservation of selected J558L cells. 6. A total of at least 250 × 106 cells should be obtained at this step.
Ex Vivo Expansion of Alveolar Macrophages
245
Fig. 2 Lentiviral transduction and expression screening in exAM. Concentrated lentiviral particles encoding CBh-mCherry were used to transduce exAM at MOI 5 (non-transduced exAM were used in parallel as control). (a) Representative brightfield (top panels) and epifluorescence (bottom panels) images showing reporter gene (mCherry) expression in non-transduced control exAM (left) and transduced exam (right) on day 3 post lentiviral transduction. Scale bar, 300 μm. Epifluorescence images were taken on an EVOSTMM5000 microscope (Invitrogen). (b) Flow cytometric analysis of reporter gene (mCherry) expression in control (left) and transduced exAM (right) on day 6 post lentiviral transduction
7. Two hundred mL is a large volume for such a vessel, so the flasks have to stand tilted in the incubator to avoid media contact with the caps or filters.
246
Clara Jana-Lui Busch et al.
8. During this period the cells will start to die, and the medium will appear orange to yellow with a high turbidity. We do not measure turbidity (OD600 nm) on a regular basis. For reference, the turbidity of the conditioned medium on day 10 is at least 0.15. 9. When transferring the conditioned medium, leave a residual volume of 500 μL–1 mL on top of the cell pellet to avoid cellular contamination. Do not touch the cell pellet with the pipet tip. 10. We typically aliquot 10 mL per tube, which is sufficient for preparing 1 L of AM medium at 1% conditioned medium containing GM-CSF. 11. We use the Mouse GM-CSF DuoSet from R&D systems to quantify every batch of GM-CSF conditioned medium. The exact amount of conditioned medium needed for preparation of AM medium depends on the concentration of each batch. AM medium typically contains a final concentration of 2 ng/mL of GM-CSF. Higher concentrations (up to 5 ng/ mL) do not affect AM proliferation. In our hands, a typical batch contains around 200 ng/mL GM-CSF and is hence diluted 1:100 (e.g., 1% conditioned medium in complete medium). 12. Do not use CO2 or inhalant anesthetic overdose for euthanasia in order to maintain AM in naı¨ve state. Cervical dislocation (CD) is the preferred method and is a critical step. CD causes disruption of cervical tissues and when performed forcefully may even rupture the trachea. When the trachea is damaged, the 18-G cannula cannot be inserted into the trachea and/or held in place to perform the lavage ten times. When performing the dissection, remove the skin, ribcage, and muscles while avoiding rupture of the jugular vein and other blood vessels to reduce blood streaming and contamination of BAL with red blood cells. 13. Gentamicin is a broad-spectrum antibiotic and effective in inhibiting growth of bacteria possibly contaminating the BAL cells such as lung commensal bacteria or from other external sources when performing a BAL. Gentamicin is omitted after the first medium change. 14. AM will fully adhere to the bottom of the wells within 6–18 h of plating. The supernatant during the first medium change is discarded within this time window, and only the adherent cells are carried to Subheading 3.3, step 7. 15. After the first medium change, the AM can be observed as round-shaped and semi-adherent. Therefore, it is important to collect and pool both adherent and suspended fractions when performing subsequent medium changes, re-plating, or harvesting the cells.
Ex Vivo Expansion of Alveolar Macrophages
247
16. The first doubling of the cell number will occur within 5–10 days, requiring passaging of the cells. If the cell count shows that the cell number has not doubled compared to the number of seeded cells, continue the culture by re-seeding at the recommended cell density following Table 1. For assessing the confluency of the culture, check the cells with an inverted brightfield microscope as follows: In general, there should be roughly the same number of cells in suspension as attached to the plate. In a healthy, proliferating culture, more than 80% of the adherent cells should have a small, round-shaped, shiny morphology. By gently rocking the plate, observe whether the cells in suspension also have a small, round-shaped morphology are live and not clumping or forming structures connected with threads, indicating cell death. 17. The detachment medium containing accutase is preferred for AM for a gentle and efficient detachment while maintaining high viability. Do not pipet the cell suspension extensively; this might affect the viability of the cells. If the cells do not detach within 10 min, collect the already detached cells, and perform another round of incubation with fresh detachment medium. Early passage cells may require incubation time of up to 30 min, while late passage cells require shorter incubation times of 5–10 min. 18. Passaging cells at the right moment is critical as otherwise a reduction in the fraction of adherent cells, and an increase in the fraction of floating cells may be observed that consists of dead or dying exAMs. After the first passage, continue feeding the culture even if a confluency of 70–80% is not reached within 5–10 days, as long as the fraction of floating and adherent cells is roughly the same. If release pressure). (c) Wrong tissue temperature (in our hands a tissue temperature of ~35 °C has proven optimal and can be measured using a laser thermometer => adjust temperature of the heating chamber). (d) Wrong or too much buffer (overuse of PBS dilutes local cytokine/growth factor/electrolyte milieu => use as little exogenous buffer as possible). (e) Laser intensity too high (Fig. 9, Electronic Supplementary Video 2; laser excitation can cause thermal damage => use as little laser output power as possible, especially with higher zoom settings; here a compromise must be made between image quality and tissue physiology).
Functional In Vivo Imaging of Macrophages
333
Fig. 8 Morphology of resident peritoneal macrophages under steady state (on the left-hand side) and after a harmful stimulus (e.g., azide in the buffer, on the right-hand side) showing a morphological change resulting in dystrophic looking macrophages. Scale bar, 20 μm
Fig. 9 Close-up image of one macrophage showing a normal morphology at the beginning of the imaging session (on the left-hand side). After 50 min using a high zoom, the photodamage from the 2P-imaging results in the complete stop of protrusion movements and an amoebic shape (on the right-hand side). Scale bar, 5 μm. See also Electronic Supplementary Video 2
334
Anja Wegner et al.
3. As we are operating in a living tissue, some motion artifacts often cannot be prevented entirely. In anesthetized animals, motion artifacts during imaging can arise from heartbeat or breathing, leading to micrometer-scale displacements of the peritoneum. Again, a compromise must be made between image stability and tissue physiology. Consider the following while trying to stabilize the tissue and optimize imaging quality: (a) Exerting increased pressure onto the peritoneum via the coverslip is not recommended because of the risk of causing tissue damage. (b) Changing the animal’s body position and main axis of motion or anesthetic level may help to diminish breathing-induced motion artifacts. (c) Higher scan rates (at lower resolution) may help to reduce within-frame image distortions. (d) Lateral motion artifacts remaining in and between image stacks can be largely corrected in post-processing using drift-correction algorithms. 4. Another critical aspect in cell imaging is the potential effect of laser excitation on tissue physiology itself. Excessive excitation can cause pathological tissue changes. Photodamage threshold depends on various parameters, including pulse duration, exposure time, and excitation intensity and therefore, needs to be determined experimentally. This can be achieved, for example, by exposing the peritoneal serosa to various illumination conditions and then quantifying any changes in macrophage morphology over the duration of light exposure. Generally, fluorescence imaging should be conducted using as little light exposure as possible. RTM sampling dynamics can be used as a sensitive indicator of intact or disrupted tissue physiology: “If the tissue is unhappy, you can see it immediately on the macrophages”
Acknowledgments SU is by the DFG (448121430, 448121523, 405969122, and 261193037) and through an ERC starting grant (101039438). Imaging was performed at the Optical Imaging Centre Erlangen (OICE) using a DFG-funded microscope system (project number 261193037). We thank OICE staff for technical support.
Functional In Vivo Imaging of Macrophages
335
References 1. Gordon S, Taylor PR (2005) Monocyte and macrophage heterogeneity. Nat Rev Immunol 5(12):953–964 2. Wynn TA, Chawla A, Pollard JW (2013) Macrophage biology in development, homeostasis and disease. Nature 496(7446):445–455 3. Gautier EL, Shay T, Miller J, Greter M, Jakubzick C, Ivanov S et al (2012) Geneexpression profiles and transcriptional regulatory pathways that underlie the identity and diversity of mouse tissue macrophages. Nat Immunol 13(11):1118–1128 4. Amit I, Winter DR, Jung S (2016) The role of the local environment and epigenetics in shaping macrophage identity and their effect on tissue homeostasis. Nat Immunol 17(1):18–25 5. Ginhoux F, Guilliams M (2016) Tissueresident macrophage ontogeny and homeostasis. Immunity 44(3):439–449 6. Guilliams M, Scott CL (2017) Does niche competition determine the origin of tissueresident macrophages?. Nature reviews. Immunology 17(7):451–460 7. van Furth R, Cohn ZA (1968) The origin and kinetics of mononuclear phagocytes. J Exp Med 128(3):415–435
8. Ingersoll MA, Platt AM, Potteaux S, Randolph GJ (2011) Monocyte trafficking in acute and chronic inflammation. Trends Immunol 32(10):470–477 9. Murray PJ, Wynn TA (2011) Protective and pathogenic functions of macrophage subsets. Nat Rev Immunol 11(11):723–737 10. Okabe Y, Medzhitov R (2014) Tissue-specific signals control reversible program of localization and functional polarization of macrophages. Cell 157(4):832–844 11. Freeman SA, Uderhardt S, Saric A, Collins RF, Buckley CM, Mylvaganam S, Boroumand P, Plumb J, Germain RN, Ren D, Grinstein S (2020) Lipid-gated monovalent ion fluxes regulate endocytic traffic and support immune surveillance. Science (New York, N.Y.) 367(6475):301–305 12. Uderhardt S, Martins AJ, Tsang JS, L€ammermann T, Germain RN (2019) Resident macrophages cloak tissue microlesions to prevent neutrophil-driven inflammatory damage. Cell 177(3):541–555.e17
Chapter 23 Elucidating Immune Monitoring of Tissue-Resident Macrophages by Intravital Microscopy Karolin W. Hublitz and Efstathios G. Stamatiades Abstract Intravital microscopy is an invaluable tool to study in real time the dynamic behavior of leukocytes in vivo. We describe herein a simple protocol for time-lapse imaging of tissue-resident macrophages in intact kidney, liver, and spleen in live mice. This method can be used in any commercially available inverted confocal microscope, doesn’t require expensive lasers or optics, exhibits minimal organ perturbation, photo bleaching, or phototoxicity, and, hence, it enables the study of tissue-resident macrophages in situ and in vivo under steady state and inflammation. Key words Kidney, Liver, Spleen, Imaging, Intravital, Monocytes, Macrophages
1
Introduction Recent advances in high-throughput cell profiling techniques have revealed a much greater than anticipated cellular heterogeneity in tissue-resident macrophages. Yet, techniques exclusively relying on dissociation-based cell profiling (e.g. flow cytometry, cytometry by time of flight, single-cell RNA sequencing)—albeit powerful—are likely to produce inaccurate information about the in vivo and in situ functions of tissue-resident macrophages, since these techniques are heavily biased toward preferential analysis of cells that are easily extracted from the tissues. In order to overcome this caveat, techniques for in vivo imaging have been developed. The term intravital microscopy refers to the in vivo imaging of live anesthetized mammals in real time. A lot of excellent reviews have extensively discussed this technique [1– 3]. In vivo imaging allows the study of cells within their native environment, avoiding artifacts observed in vitro. High spatial
Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1007/978-1-0716-3437-0_23. Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_23, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
337
338
Karolin W. Hublitz and Efstathios G. Stamatiades
and/or temporal resolution, dynamic intravital imaging has been utilized to visualize single-cell dynamics in situ and to elucidate the complex biology of tissue-resident macrophage, their cell-cell interactions, and functions [1], revealing patrolling of alveolar macrophages [4]; phagocytosis by Kupffer cells [5]; “cloaking” of damaged tissue by peritoneal macrophages [6]; surveillance of the nerves, brain parenchyma, and epithelial ducts by skin macrophages [7], microglia [8], and mammary gland macrophages, respectively [9]; and scavenging of immune complexes by kidney macrophages [10]. We describe herein a protocol for the real-time imaging of tissue-resident macrophages in the kidney, liver, and spleen of mice. This protocol is simple, doesn’t require expensive optics (e.g. filters, two-photon lasers), can be performed with any commercial inverted confocal microscope (e.g. Leica SP5, Zeiss LSM880), and offers excellent spatiotemporal resolution to study a plethora of biological phenomena (e.g. phagocytosis, cell-cell interactions) for long periods of time (up to 5 h). Due to light scattering, this protocol is not suitable for imaging of deep tissues (deeper than 50 μm). Two-photon microscopy might be more suitable to routinely image deeper tissues. Yet, two-photon laser systems are more complex, expensive, and have lower spatial resolution than “traditional” single-photon microscopy [11].
2 2.1 2.1.1
Materials Reagents General Reagents
1. Glass coverslips and silicon grease to prepare the stage insert. 2. Anesthetic cocktail: 50 mg/kg ketamine, 10 mg/kg xylazine, and 1.7 mg/kg acepromazine in 0.9% NaCl (normal saline, NS). 3. 1 mL syringes and 26G or 27G needles. 4. Isoflurane and oxygen. 5. Ophthalmic ointment. 6. Phosphate buffered saline (PBS) or NS prewarmed at 37 °C. Make small (~5 mL) aliquots and keep sterile. 7. Antiseptic solution: 10% povidone-iodine (e.g. Betadine) prewarmed at 37 °C. 8. Paper strips: 2–4 mm wide, 0.5–1 cm long. The paper strips can be made from Kimwipes, “regular” hands-drying paper towels or similar. 9. Adhesive tape (e.g. Sellotape or similar).
Live Imaging of Tissue-Resident Macrophages 2.1.2 Reagents for In Vivo Injection
339
The injected reagent depends on the specific research needs (e.g. beads, bacteria). In this protocol the following reagents were used: 1. 70 kDa or 2 MDa TRITC-dextran. 2. Anti-F4/80 AF647, Clone BM8. 3. Ovalbumin AF647. 4. C. albicans dTomato, Gratacap et al. [12]. 5. EL4 lymphoma cell line.
2.2
Equipment
1. Inverted confocal microscope with a temperature-controlled microscope enclosure. 2. Heating pad set up at 37 °C. 3. Controllable inhalation anesthesia machine suitable for small animals, for the delivery of oxygen and isoflurane. 4. Hair clipper suitable for small animals. 5. Sterile coarse/blunt scissors and forceps. 6. Small vessel cauterizer. 7. Hot bead sterilizer. 8. Powerful workstation (PC or Mac) for the processing of intravital imaging datasets.
2.3
Software
1. Image analysis software for processing of intravital imaging datasets, e.g. ImageJ and Imaris. 2. Suitable software for processing and annotating intravital microscopy movies, e.g. Adobe After Effects and Adobe Premiere Pro.
3
Methods
3.1 Microscope Setup
1. At least 1 h before imaging, set up the temperature of the microscope enclosure to 37 °C. This will prevent the hypothermia of the anesthetized mouse during imaging (see Note 1). 2. Prepare the microscope stage insert by using silicon grease to secure a glass coverslip on the hole of the stage insert. Transfer the prepared stage insert in the microscope enclosure (see Note 2). 3. Depending on your imaging system, turn on the microscope to “warm-up” the lasers, per manufacturer’s instructions. 4. Inspect the oxygen/isoflurane supply, and make sure that there is enough oxygen/isoflurane for the duration of the imaging period. Depending on your setup, replace oxygen cylinder, and fill up the isoflurane vaporizer. 5. Prewarm to 37 °C any solutions needed, e.g. PBS, NS, and antiseptic solution.
340
Karolin W. Hublitz and Efstathios G. Stamatiades
Fig. 1 Surgical exposure of the kidney for intravital microscopy. (a) Mouse anesthesia. (b) Fur trimming. (c) Skin scrubbing with Betadine and incision (dotted line). (d and e) Body wall incision (dotted line). (f) Exteriorization of the kidney. (g) Stage insert preparation with moist paper strips. The black circle denotes the imaging “window.” (h) Transfer of the animal on the stage insert. (i) Securing the animal on the stage insert using adhesive tape
3.2 Mouse Handling and Preparation
1. Use 7- to 12-week-old male mice (see Notes 3 and 4). 2. Anesthetize the mouse by intraperitoneal injection of anesthetic cocktail. 3. Transfer the mouse on a heating pad set up at 37 °C (Fig. 1a). 4. Provide the animal continuously with 0.5 L/min oxygen and 1–2% isoflurane via a cone (Fig. 1a). 5. Apply ophthalmic ointment to the eyes following anesthesia to prevent corneal drying. 6. Before moving to the next step, confirm deep anesthesia of the mouse by the loss of response to reflex stimulation (toe or tail pinch with firm pressure).
Live Imaging of Tissue-Resident Macrophages
3.3
Surgery
341
1. Trim the fur from the left flank region using a hair clipper suitable for small animals (Fig. 1b; see Notes 5 and 6). 2. Clean the left flank using prewarmed antiseptic solution (Fig. 1c). 3. Lift the skin with a pair of sterile forceps, and using sterile coarse scissors, make a 2–3 cm left flank incision (Fig. 1c and d, see Note 7). 4. Using sterile blunt scissors, separate the underlying dermis from the body wall on either side of the incision (Fig. 1d). 5. Identify the location of the kidney by applying gentle pressure with the thumb and index finger on the abdomen. 6. Using sterile fine scissors, make a ~ 0.5 cm incision in the body wall (Fig. 1e; see Note 8). 7. Applying gentle pressure outside the muscle wall on either side of the kidney using the index finger and thumb to exteriorizse the kidney (Fig. 1f; see Note 9). 8. Apply a couple of drops of sterile, prewarmed (at 37 °C) NS or PBS on the kidney to prevent tissue drying. 9. Prepare the stage insert by applying NS- or PBS-soaked paper strips on the coverslip to form a window for imaging (Fig. 1g; see Note 10). The paper strips will help stabilize the tissue and prevent motion artifacts. 10. Transfer the mouse on the stage insert, in such a way that the kidney will be on the small window made from paper strips (Fig. 1g and h). 11. Further immobilize the animal on the stage insert by two strips of adhesive tape applied gently over the front and back legs of the mouse (Fig. 1i). 12. Keep the kidney/tissue moist by applying sterile prewarmed PBS- or NS-soaked paper around the kidney and surgical area.
3.4 Intravital Microscopy
1. Place the stage insert with the mouse in the microscope, and maintain oxygen/isoflurane inhalation. 2. Use a 10× or 20× objective to locate the kidney and identify a field of interest (see Note 11). 3. Set up the microscope (see Note 12). Scanning at 512 × 512 pixels, with a z-step of 3–4 μm and a pinhole of 1 airy unit, is a good start. Depending on your specific needs, adjust the microscope’s settings accordingly. 4. If needed, define the boundaries for a z-stack. 5. Scan a z-stack, check for any z-drift,and, if necessary, readjust z-stack to correct any z-drift.
342
Karolin W. Hublitz and Efstathios G. Stamatiades
6. Depending on your specific needs, inject i.v. any reagents (e.g. antibodies, high-molecular-weight dextran) before or during image acquisition (see Notes 13 and 14). 7. Apply around 300 μL of prewarmed NS (or PBS) on the kidney every 45 min to keep the tissue moist and prevent tissue drying. 8. In order to prevent the dehydration of the mouse, inject the animal s.c. with around 500 μL NS (or PBS) every hour. 9. Humanely euthanize the mouse according to the institutional protocol after the completion of the imaging period. 3.5 Post-acquisition Processing of Intravital Imaging Datasets
1. Load the raw data in your preferred image analysis (e.g. ImageJ, Imaris) software (see Note 15). 2. Perform any necessary (e.g. smoothening, noise reduction, co-localization, surface rendering, cell tracking, deconvolution) modifications based on your research question (see Note 16). 3. Export movies using the most suitable file format (e.g. *.mov, *.avi) for further processing (see next step). 4. Use your preferred software to process (e.g. increase or decrease frame rate), annotate, and re-export the movies into a format suitable (e.g. *.mp4) for presentation (see Note 17). 5. Representative examples of in vivo imaging of tissue-resident macrophages in the kidney, liver, and spleen are depicted in Figs. 2a–d, 3, and Electronic Supplementary Movies 1–8.
4
Notes 1. Based on our experience, a microscope enclosure (e.g. cage incubator by Okolab) works better to prevent the hypothermia of the anesthetized mouse than a heating pad/blanket. The microscope enclosure should be transparent, to allow direct observation of the living animal (e.g. breathing). 2. In order to minimize any z-drift during imaging, it is recommended to acclimate the stage insert at 37 °C. 3. Use young adult male mice, since in older mice (over 12 weeks old), there is too much visceral fat and extensive perirenal adipose tissue, which hinders surgery and impairs imaging. Additionally, since the kidney and ovary are interconnected with fat tissue, surgical exposure of the kidney in female mice is more challenging, albeit feasible. 4. Depending on the research question, a fluorescent reporter mouse suitable for intravital imaging might be required. Each mouse strain should be validated in house, e.g. for the
Live Imaging of Tissue-Resident Macrophages
343
Fig. 2 Time-lapse intravital microscopy of kidney-resident macrophages in superficial renal cortex of a Cx3cr1gfp/+ Rag2-/- Il2rgY/- mouse in homeostasis and inflammation. (a) The snapshot depicts the tubular capillaries stained in red after i.v. injection 2 MDa TRITC-dextran (red), the renal tubules stained in white after endocytosing i.v. injected Alexa Fluor 647-conjugated Ovalbumin (white), and Cx3cr1gfp kidney-resident macrophages (green). Bar = 50 μm. See also Electronic Supplementary Movie 1. (b) Cx3cr1gfp kidneyresident macrophages (green) around a large blood vessel. The vasculature is stained red with i.v. injection of 70 kDa TRITC-dextran. Arrow shows the direction of blood flow, while arrowheads depict crawling Cx3cr1gfp monocytes inside the vessel. Bar = 20 μm. See also Electronic Supplementary Movie 2. (c) Macrophages (green)-Candida (red) interactions 3 h after i.v. injection C. albicans dTomato. Bar = 20 μm. See also Electronic Supplementary Movie 3. (d) Scavenging of TRITC-dextran (red) by a Cx3cr1gfp kidney-resident macrophage (green). Time stamp is in min:s. Bar = 10 μm. See also Electronic Supplementary Movies 4 and 5
brightness and stability of the fluorescence. For imaging of kidney-resident macrophages, we used the commercially available Cx3cr1gfp/+ knock-in mice [10, 13]. 5. For intravital imaging of the kidney, use only the left kidney, since the surgical exposure of the right kidney is very challenging, due to the presence of the liver. Although the surgical exposure of the kidney is described herein, the spleen can be exteriorized in similar way. For imaging the liver, a midline incision might be necessary, depending on which liver lobe would be used.
344
Karolin W. Hublitz and Efstathios G. Stamatiades
Fig. 3 Snapshots from time-lapse intravital microscopy in a C57BL/6 mouse depicting scavenging of EL4 cells (green) and/or EL4-tumor material (green) by Kupffer cells (red) in the liver (left) and red pulp macrophages (red) in the spleen (right). Macrophages in liver and spleen were labeled with i.v. injection of 10 μg Alexa Flour-647 conjugated F4/80 antibody. Bars = 10 μm. See also Electronic Supplementary Movies 6, 7, and 8
6. It is important to remove all trimmed hair from the surgical area, using a warmed-PBS moist tissue. During surgery leftover trimmed hair could attach on the tissue and impede imaging. 7. We find it very convenient to use a hot bead sterilizer to sterilize the surgical tools just before surgery. 8. There should be minimal bleeding after the incision, since blood vessels on the body wall are clearly visible and, hence, can be avoided. Alternatively, a vessel cauterizer can be used to minimize bleeding. If using a vessel cauterizer be careful not to damage any underlying internal organ (e.g. spleen, liver, kidney). 9. If the incision from the previous step is too big, the kidney will retract back into the abdominal cavity. If too narrow, you will not be able to exteriorize the kidney. Always avoid touching the organ, since even a “gentle” touch could cause tissue injury. Using your finger or a 10-μL pipette tip to gently push-in the abdomen below, the incision could help the exteriorization of the kidney. 10. We recommend preparing the stage insert before surgery, to avoid losing time. 11. Use a suitable microscope objective depending on your specific needs. An objective with long working distance and high numerical aperture (NA) is highly recommended.
Live Imaging of Tissue-Resident Macrophages
345
12. One of the most crucial points of consideration in intravital experiments is laser-induced photodamage. Absorption of laser energy by the tissue leads to toxicity, injury, and hence artifacts. According to our experience, in order to avoid phototoxicity and artifacts, the laser power should be no more than 70–80 μW at the nosepiece—to achieve practicable good fluorescence signals without damaging the specimen. In addition, specific mouse strains that are less prone to phototoxicity [14, 15] could be used for studying macrophages in some tissues (e.g. skin). 13. Although i.v. injections on the anesthetized animal could easily be done into the retro-orbital sinus, cannulation of the tail or jugular vein could make the intravenous delivery of substances easier. 14. The i.v. injection of fluorescent-labeled antibodies is an excellent and rapid way to label cells of interest. For example, the i.v. injection of an anti-F4/80 antibody will label very well and rapidly the Kupffer cells of the liver, red pulp macrophages of the spleen and the kidney macrophages. 15. Although ImageJ is free and powerful, it has its limitations. Depending on your needs, a commercially available software (e.g. Imaris by Bitplane) might be required for your image analysis. 16. The post-acquisition processing of raw intravital imaging datasets requires as much as possible computing power. Large datasets will need a state-of-the-art workstation with powerful processor, fast SSD hard drives, a lot of RAM, and a powerful graphics card. We also recommend two, at least 30 inches, monitors to increase desktop workspace and hence to ease post-acquisition processing. 17. We routinely use Adobe After Effects, Premiere Pro, and Media Encoder to annotate, process, and re-export intravital microscopy movies.
Acknowledgments The work in our laboratory is supported by the German Research Foundation (STA 1657/2-1). The authors would like to thank Prof. Robert Wheeler for C. albicans dTomato and Prof. Morgan Huse for EL4 cell line.
346
Karolin W. Hublitz and Efstathios G. Stamatiades
References 1. Germain RN, Robey EA, Cahalan MD (2012) A decade of imaging cellular motility and interaction dynamics in the immune system. Science 336(6089):1676–1681. https://doi.org/10. 1126/science.1221063 2. Cahalan MD, Parker I (2008) Choreography of cell motility and interaction dynamics imaged by two-photon microscopy in lymphoid organs. Annu Rev Immunol 26(1): 585–626. https://doi.org/10.1146/annurev. immunol.24.021605.090620 3. Coombes JL, Robey EA (2010) Dynamic imaging of host-pathogen interactions in vivo. Nat Rev Immunol 10(5):353–364. https:// doi.org/10.1038/nri2746 4. Neupane AS, Willson M, Chojnacki AK, Vargas ESCF, Morehouse C, Carestia A, Keller AE, Peiseler M, DiGiandomenico A, Kelly MM, Amrein M, Jenne C, Thanabalasuriar A, Kubes P (2020) Patrolling alveolar macrophages conceal bacteria from the immune system to maintain homeostasis. Cell 183(1): 110–125. e111. https://doi.org/10.1016/j. cell.2020.08.020 5. Deppermann C, Kratofil RM, Peiseler M, David BA, Zindel J, Castanheira F, van der Wal F, Carestia A, Jenne CN, Marth JD, Kubes P (2020) Macrophage galactose lectin is critical for Kupffer cells to clear aged platelets. J Exp Med 217(4). https://doi.org/10. 1084/jem.20190723 6. Uderhardt S, Martins AJ, Tsang JS, Lammermann T, Germain RN (2019) Resident macrophages cloak tissue microlesions to prevent neutrophil-driven inflammatory damage. Cell 177(3):541–555. e517. https://doi. org/10.1016/j.cell.2019.02.028 7. Kolter J, Feuerstein R, Zeis P, Hagemeyer N, Paterson N, d’Errico P, Baasch S, Amann L, Masuda T, Losslein A, Gharun K, MeyerLuehmann M, Waskow C, Franzke CW, Grun D, Lammermann T, Prinz M, Henneke P (2019) A subset of skin macrophages contributes to the surveillance and regeneration of local nerves. Immunity 50(6):1482–1497 e1487. https://doi.org/10.1016/j.immuni. 2019.05.009 8. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting microglial cells are highly
dynamic surveillants of brain parenchyma in vivo. Science 308(5726):1314–1318. https://doi.org/10.1126/science.1110647 9. Dawson CA, Pal B, Vaillant F, Gandolfo LC, Liu Z, Bleriot C, Ginhoux F, Smyth GK, Lindeman GJ, Mueller SN, Rios AC, Visvader JE (2020) Tissue-resident ductal macrophages survey the mammary epithelium and facilitate tissue remodelling. Nat Cell Biol 22(5): 546–558. https://doi.org/10.1038/s41556020-0505-0 10. Stamatiades EG, Tremblay ME, Bohm M, Crozet L, Bisht K, Kao D, Coelho C, Fan X, Yewdell WT, Davidson A, Heeger PS, Diebold S, Nimmerjahn F, Geissmann F (2016) Immune monitoring of transendothelial transport by kidney-resident macrophages. Cell 166(4):991–1003. https://doi.org/10.1016/j.cell.2016.06.058 11. So PTC, Dong CY, Masters BR, Berland KM (2000) Two-photon excitation fluorescence microscopy. Annu Rev Biomed Eng 2(1): 399–429. https://doi.org/10.1146/annurev. bioeng.2.1.399 12. Gratacap RL, Rawls JF, Wheeler RT (2013) Mucosal candidiasis elicits NF-kappaB activation, proinflammatory gene expression and localized neutrophilia in zebrafish. Dis Model Mech 6(5):1260–1270. https://doi.org/10. 1242/dmm.012039 13. Jung S, Aliberti J, Graemmel P, Sunshine MJ, Kreutzberg GW, Sher A, Littman DR (2000) Analysis of fractalkine receptor CX3CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol 20(11):4106–4114. https://doi.org/10. 1128/mcb.20.11.4106-4114.2000 14. Li JL, Goh CC, Keeble JL, Qin JS, Roediger B, Jain R, Wang Y, Chew WK, Weninger W, Ng LG (2012) Intravital multiphoton imaging of immune responses in the mouse ear skin. Nat Protoc 7(2):221–234 15. Lammermann T, Afonso PV, Angermann BR, Wang JM, Kastenmuller W, Parent CA, Germain RN (2013) Neutrophil swarms require LTB4 and integrins at sites of cell death in vivo. Nature 498(7454):371–378. https:// doi.org/10.1038/Nature12175
Chapter 24 Combined Host-Pathogen Fate Mapping to Investigate Lung Macrophages in Viral Infection Sebastian Baasch, Julia Henschel, and Philipp Henneke Abstract Macrophage identity, as defined by epigenetic, transcriptional, proteomic, and functional programs, is greatly impacted by cues originating from the microenvironment. As a consequence, immunophenotyping based on surface marker expression is established and reliable in homeostatic conditions, whereas environmental challenges, in particular infections, severely hamper the determination of identity states. This has become more evident with recent discoveries that macrophage-inherent plasticity may go beyond limits of lineage-defining immunophenotypes. Therefore, transgenic fate mapping tools, such as the phage-derived loxP-cre-system, are essential for the analysis of macrophage adaptation in the tissue under extreme environmental conditions, for example, upon encounter with pathogens. In this chapter, we describe an advanced application of the loxP-cre-system during infection. Here, the host encodes a cell type-specific cre-recombinase, while the pathogen harbors a STOP-floxed fluorescent reporter gene. As an instructive example for the versatility of the system, we demonstrate that alveolar macrophages are predominantly targeted after respiratory tract infection with mouse cytomegalovirus (MCMV). Combined host-pathogen fate mapping not only enables to distinguish between infected and non-infected (bystander) macrophages but also spurs exploration of phenotypic adaptation and tracing of cellular localization in the context of MCMV infection. Moreover, we provide a gating strategy for resolving the diversity of pulmonary immune cell populations. Key words Alveolar macrophage, Lung, Mouse cytomegalovirus (MCMV), Host-pathogen interaction, Fate mapping, loxP-cre-system, Host-pathogen fate mapping, Respiratory tract infection, Isolation, Flow cytometry
1
Introduction Macrophages (Mφ), which have long been considered to form a homogenous immune cell population that is constantly replenished by incoming monocytes [1], are now established to exhibit high heterogeneity. They reside in discrete, extremely diverse niches, where they exert specific functions tailored to microanatomical needs [2]. Indeed, it has become evident that macrophage identity,
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_24, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
347
348
Sebastian Baasch et al.
as defined by epigenetic, transcriptional, proteomic, and functional features, is shaped by integrating cues dependent on origin, activation status, the immediate environment, and the time spent in their respective niche [3]. In the lung, two major compartments are separated by a single layer of epithelial cells, namely, alveolar epithelial cells (AEC) type I and II. The first major compartment is the pulmonary interstitium. It contains structures that are common to non-parenchymal tissues, such as connective tissue, lymphatic and blood vessels, nerves, and—specific to the respiratory tract—embedded bronchi and bronchioli. In contrast to the bronchial interstitium, the quantity and diversity of cells and structures decrease in the alveolar interstitium, which surrounds the terminal air sacs of the lung, in order to enable rapid diffusion of oxygen. Accordingly, Mφ that populate the pulmonary interstitium are called “interstitial Mφ.” Based on expression of MHCII and Lyve1 [4] or CD206 [5], they can be divided into two subsets: MHCIIhi Lyve1lo and MHCIIlo Lyve1hi or CD206- and CD206+ interstitial Mφ. Integration of CD11c expression enables a further subclassification of interstitial Mφ 1 (CD11clo CD206+ MHCIIlo), interstitial Mφ 2 (CD11clo CD206+ MHCIIhi), and interstitial Mφ 3 (CD11chi CD206lo MHCIIhi) [6]. Similar to other organs, like the skin [7] or the intestine [8], interstitial Mφ subsets inhabit specific niches. MHCIIlo Lyve1hi and CD206+ interstitial Mφ are associated to blood vessels [4, 5], especially in the bronchial interstitium [5]. CD206- and MHCIIhi Lyve1lo interstitial Mφ populate the alveolar interstitium [5] and have been proposed to reside in the vicinity of peripheral nerves [4]. Association to β3 tubulin expressing nerves has also been observed in CD169+ CD11c- interstitial Mφ in close proximity to the airways [9]. However, according to previous phenotyping, low CD11c expression comprises MHCIIlo Lyve1hi [4]/CD206+ [5] or interstitial Mφ 1/2 [6] subsets, which were located in the bronchial interstitium [5, 6] and in the vicinity of blood vessels [4], as elaborated before. Thus, the specific association of interstitial Mφ with distinct niches requires further investigation and could be in part affected by the challenging preparation and processing of lung sections for microscopy. Functionally, all subsets, but most prominently CD206+ interstitial Mφ [5], secrete IL-10 during steady state and, thus, modulate dendritic cell functions to avert allergic asthma [10, 11]. Interstitial Mφ rely on M-CSF signaling similar to most tissue-resident Mφ [9]. The luminal side of the terminal alveoli, where alveolar macrophages (alveolar Mφ) reside, represents the second compartment of the lung. In contrast to interstitial Mφ, alveolar Mφ are considered to be a homogenous cell population that can be identified by SiglecF+ CD11c+ and CD11blo surface expression. They rely on the transcription factor Pparg, which is induced via GM-CSF and TGF-β [12–14]. GM-CSF is secreted by alveolar epithelial cells
Host-Pathogen Fate Mapping of Lung Macrophages
349
(AEC) type II [15, 16] and to a minor extent by eosinophils [17]. TGF-β is secreted by alveolar Mφ and acts in an autocrine manner [13, 18], yet AEC need to transform inactivated TGF-β into its active form [19–21]. Moreover, the transcription factors Egr2 and Bhlhe40 are important for the maintenance of the alveolar Mφ population [22, 23]. Alveolar Mφ constantly take up surfactant to prevent its accumulation, which would otherwise result in a disease called pulmonary alveolar proteinosis [24]. Additionally, they engulf cell debris to enable gas exchange and prevent respiratory failure after infection with influenza virus [25]. In case of pathogen encounter, alveolar Mφ secrete cytokines and chemokines to initiate immune cell recruitment. Accordingly, alveolar Mφ safeguard the interface of the lower respiratory tract against inhaled particles, bacteria, fungi and viruses. Alveolar Mφ originate from embryonic precursors [26, 27], specifically from fetal monocytes that mature in the postnatal period [12] without significant replacement by circulating bone marrow-derived monocytes [28, 29] at least early in life. Thus, alveolar Mφ are reasonably well characterized in resting conditions. However, their cell-type defining characteristics become blurred under inflammatory conditions, especially due to the immediate influx of monocytes from the periphery into the lung. Monocytederived Mφ differentiate and contribute to the tissue resident alveolar Mφ population by adopting a similar phenotype compared to bona fide alveolar Mφ. As origin is one of the key features to determine macrophage identity, this “contamination” complicates the discrimination and subsequent analysis of monocyte- and embryonic-derived alveolar Mφ. Of note, a diminished SiglecF expression can assist in the identification of monocyte-derived alveolar Mφ [30]. Fate mapping tools, such as the phage-derived loxP-cre-system, have been vital for our understanding of Mφ origin and niche population. The system comprises two components: the recombinase (cre, a), which recognizes short (34 base pairs) loxP (locus of x-over P1) sequences (b), that leads to recombination of loxPflanked (floxed) DNA and subsequent genomic deletion [31]. For fate mapping of cells, mice expressing a cell-specific cre-recombinase transgene are typically crossed to loxP-“STOP”-loxP (LSL)reporter mice, in which the expression of a fluorescent marker (GFP, Tomato, etc.) is prevented by preceding floxed stop signals (e.g., multiple poly-A tails). In cells of progeny mice, where both components are expressed (a and b), the recombination leads to deletion of the stop signal cassettes and enables translation of the fluorescent marker [29]. Hence, the cell of interest is endogenously labeled and can be traced or extracted for further downstream analyses. Thus, cellular fate mapping is used to add spatial information within a given tissue and time of residency and origin to define Mφ identity. Until now the loxP-cre system to study macrophage
350
Sebastian Baasch et al.
biology comes in three flavors: single promotor-driven constitutively expressed cre-recombinase [29, 31], single promotor-driven drug-dependent conditional expressed cre-recombinase [29, 32, 33], and two promotor-driven complementation-dependent (“split”) cre-expression [34, 35] (see also Chapter 18 in this book “Tackling tissue macrophage heterogeneity by spitCre transgenesis”). Exemplary and as outlined before, employing different fractalkine receptor (Cx3cr1)-driven fate mapping approaches, alveolar Mφ have been shown to receive no input from peripheral blood monocytes in steady state [29]. However, a recent Ms4a3-based fate mapping indicates a stronger contribution of bone marrow-derived monocytes to the alveolar Mφ pool, especially in the aging lung [36], which was confirmed in bone marrow chimera experiments that do not require irradiation [30]. The engraftment of monocyte-derived alveolar Mφ into the lung has opposing implication for the host. While infection with murid herpesvirus 4 leads to replacement of alveolar Mφ with monocytes that subsequently confer protection against house dust mite extract-induced asthma [37], monocyte-derived alveolar Mφ express pro-fibrotic genes that highlight a causative role in bleomycin-induced lung fibrosis [38]. Interestingly, after influenza virus infection, monocyte-derived alveolar Mφ protect against secondary bacterial infection [39] but play a detrimental role during secondary viral (influenza virus) infection [30]. Mφ also play a central role in the pathogenesis of CMV as it dedicates gene products specifically to manipulate Mφ [40]. We and others have established that alveolar Mφ are predominant targets in acute MCMV respiratory tract infection [41, 42]. Upon CMV infection, aforementioned alveolar Mφ hallmark functions, such as phagocytosis and cytokine secretion, are impaired. Instead, infected alveolar Mφ become highly mobile and migrate to the lung interstitium and to the draining lymph node [41]. Uninfected bystander alveolar Mφ, on the other hand, show a typical type I interferon signature [41]. Accordingly, it appears that infection directly impacts determinants of Mφ identity. In line with this, infected alveolar Mφ profoundly change their transcriptomic and proteomic profile, which hampers immunophenotyping, for example, via flow cytometry [41]. Employment of a novel application of the loxPcre-system, which enables combined host-pathogen fate mapping, offers great advantages under the highly dynamic conditions of infection. Expression of a host cell type-specific cre-recombinase together with an LSL-reporter fluorescence encoded by the virus allows for the discrimination and subsequent analysis of infected and non-infected Mφ [41]. In this chapter, we show that alveolar Mφ are predominantly targeted after respiratory tract infection with mouse cytomegalovirus (MCMV). To this end, we infect transgenic
Host-Pathogen Fate Mapping of Lung Macrophages
351
LysMcre/+ mice with MCMVLSL-GFP [41] and characterize the GFP+ infected cells via flow cytometry early after infection. Together, we introduce an advancement of Mφ fate mapping in infection with MCMV. Notably, the application is not limited to respiratory tract infections with MCMV and can be adapted to other, especially viral, infections. Thus, investigators may be encouraged to make creative use of the host-pathogen fate mapping. Moreover, we provide a protocol that allows characterization of changes in the cellular and especially myeloid compartment in the lung during steady state and infection.
2
Materials
2.1 Required Reagents
1. LysMcre/+ mice. 2. Viral suspension: 1 × 105 PFU MCMVLSL-GFP in PBS, total volume 40 μL per mouse, store on ice [43]. 3. Digestion buffer: 10% FBS, 1 mg/mL hyaluronidase, 250 μg/ mL collagenase IV, 50 μg/mL Liberase TM, 250 μg/mL DNAse I in PBS. Prepare 4 mL per lung. 4. Isopropanol. 5. Phosphate-buffered saline without Ca2+ and Mg2+ (PBS). 6. Fetal bovine serum (FBS). 7. Anesthesia: 100 mg/kg body weight ketamine, 10 mg/kg body weight xylazine in water for injection (Aqua ad injectabilia). 8. Buffer for flow cytometry (FACS buffer): 2 mM ethylendiaminetetraacetic acid (EDTA), 1% FBS in PBS. 9. Red blood cell/erythrocyte lysis buffer from supplier of your choice. 10. Antibodies and reagents for flow cytometry (see Table 1).
2.2
Equipment
1. Scale. 2. Heating mat. 3. Fine scale. 4. Pipette and tips, 10–100 μL. 5. Pipette and tips, 20–200 μL. 6. Pipette and tips, 100–1000 μL. 7. Scissors. 8. Forceps. 9. Syringe, 1 and 10 mL. 10. Winged infusion set (butterfly needle). 11. 12-well plate.
352
Sebastian Baasch et al.
Table 1 Antibody panel used to characterize pulmonary immune cells with eight fluorescence channels (GFP channel is reserved for viral reporter fluorescence)
Antibody/reagent
Clone
Fluorochrome
Recommended final concentration
CD3e
145-2C11
PE
4 μg/mL
CD11b
M1/70
PeCy7
0.07 μg/mL
CD11c
HL3
APC
2 μg/mL
CD16/CD32 (“Fc block”)
93
–
5 μg/mL
CD64
X54–5/7.1
PerCP-Cy5.5
2 μg/mL
CD103
M290
PE
2 μg/mL
Ly6C
HK1.4
BV605
0.25 μg/mL
Ly6G
1A8
PE
1 μg/mL
SiglecF
E50-2440
BV421
0.4 μg/mL
Fixable viability dye
–
eFluor 780
–
12. Centrifugation tubes (“Falcon tube”), 50 mL. 13. Reaction tube (“Eppendorf tube”), 1.5 mL. 14. Centrifuge. 15. Vortexer. 16. Environmental shaker. 17. Cell strainer, 70 μm. 18. Ice box with ice. 19. Aluminum foil. 20. FACS tubes.
3
Methods
3.1 Respiratory Tract Infection with MCMV
3.1.1
Intranasal Infection
This step is optional or can be adapted to other intranasal infection models if applicable. The steps following Subheading 3.2 will allow the dissection of myeloid immune cells in the lung during steady state or other inflammatory conditions. 1. Weigh mice (see Note 1). 2. Inject anesthesia intraperitoneally (see Note 2). 3. Wait before proceeding (see Note 3). 4. Install viral suspension intranasally (see Note 4).
Host-Pathogen Fate Mapping of Lung Macrophages
353
5. Lay mouse onto a warm heating mat, and monitor breathing until awakening. 6. Put mice back into cage. 3.2 Preparation of Lung Cell Suspension for Flow Cytometry
1. Prepare digestion buffer (4 mL/lung). Transfer 1 mL digestion buffer into a 1.5 mL reaction tube and the residual 3 mL digestion buffer to a 50 mL centrifugation tube.
3.2.1
Preparation
2. Prewarm environmental shaker to 37 °C.
3.2.2
Cell Extraction
1. Sacrifice mice according to ethical guidelines (see Note 5).
3. Cool down centrifuge to 4 °C.
2. Fix the mouse with pins onto a polystyrene pad with the ventral side facing upward. 3. Moisten the mouse carefully with isopropanol (see Note 6), 4. Open the abdominal cavity (see Note 7). 5. Cut diaphragm (see Note 8). 6. Open thorax (see Note 9). 7. Cut left atrium. 8. Cut liver (see Note 10). 9. Perfuse through the right ventricle with cold PBS using a winged infusion set attached to a 10 mL syringe (see Note 11). 10. Remove the right lung close to the mediastinum (see Note 12). 11. Transfer lobes into a 12-well plate well that contains PBS and is placed on ice. Collect lungs from different mice in separate wells, as described in steps 1–11. 12. Transfer lobes in reaction tube with 1 mL digestion buffer. 13. Use scissors to cut lung into small pieces. 14. Decant into 50 mL centrifuge tube with 3 mL digestion buffer (see Note 13). 15. Transfer into environmental shaker 37 °C, 180 rpm for 45 min (see Note 14). 16. In the meanwhile, equip fresh 50 mL centrifugation tubes with 70 μm strainer, and put them on ice. 17. After 45 min digestion, vortex cell suspension vigorously for 10 s (see Note 15). 18. Pass cell suspension through 70 μm strainer, and rinse with 10 mL FACS buffer. 19. Centrifuge at 300 g for 10 min at 4 °C. 20. Decant supernatant, and place centrifugation tube upside down on a paper towel to remove remaining supernatant (see Note 16).
354
Sebastian Baasch et al.
21. Optional: Perform red blood cell lysis according to manufacturer specification (see Note 17). Wash with 3 mL FACS buffer. Centrifuge at 300 g for 6 min at 4 °C. Decant supernatant, and place centrifugation tube upside down on a paper towel to remove remaining supernatant (see Note 16). 22. Resuspend in 1 mL FACS buffer by pipetting. 23. Transfer 50 μL of cell suspension into FACS tube (see Note 18). 3.2.3 Staining for Flow Cytometry
1. Add 1:50 (1 μL) unconjugated anti-CD16/32 directly into cell suspension for “Fc block” (see Table 1) (see Note 19). 2. Incubate at 4 °C for approximately 10 min. 3. In the meantime, prepare the antibody master mix using 50 μL per sample, and double the concentration of antibodies (see Table 1) (see Note 20). 4. Add 50 μL of the 2× antibody master mix to each sample (see Note 21). 5. Incubate samples for approximately 30 min at 4 °C, protected from light. 6. Add 1 mL PBS. 7. Centrifuge at 300 g for 6 min at 4 °C, decant supernatant. 8. Resuspend cell pellet in 500 μL PBS and add 1:1000 viability dye (see Table 1) (see Note 22). 9. Incubate for 15 min at 4 °C, protected from light. 10. Add 1 mL FACS buffer. 11. Centrifuge at 300 g for 6 min at 4 °C. 12. Resuspend in 200 μL FACS buffer, and store on ice and protected from light (aluminum foil) until acquisition.
3.3 Flow Cytometry (Gating Strategies) 3.3.1 Myeloid Cell Populations in the Lung
To identify immune cells of the lung, with a special emphasis on myeloid cell populations, follow the gating strategy as shown in Fig. 1a. 1. Initial gating on forward/sideward scattering “Cells” excludes cell debris of very small size. 2. Doublets (a minimum of two cells attached to each other) represent cells high in the forward and sideward scatter pulse widths and should be excluded to analyze “Singlets.” 3. “Live cells” are negative for the fixable viability dye. 4. To identify neutrophils, conventional dendritic cells (cDC) type I and T-cells, Ly6G (neutrophils), CD103 (cDC1, tissue-resident memory T-cells), and CD3e (T-cells) can be stained with the same fluorochrome in a “pseudo-dump” channel. Next, they can be distinguished based on their differences
Host-Pathogen Fate Mapping of Lung Macrophages
Live cells
Singlets Cells
interstitial Mφ
CD11b
Eosinophils
4a. Gated on Ly6G+CD3+CD103+ Neutrophils cDC1 T-cells
CD11c
5. Gated on non-alveolar Mφ
6. Gated on myeloid CD64
SiglecF
Viability Dye
FSC-W
CD11c
4b. Gated on Ly6G-CD3-CD103SiglecF
FSC-A 7. Gated on CD64-Ly6C-
3. Gated on live cells CD11b
SSC-W
SSC-A
2. Gated on singlets CD11b
1. Gated on cells
Ly6G CD3 CD103
A
355
alveolar Mφ
myeloid inflammatory monocytes Ly6C
CD11c
Ly6C
CD11c
B SiglecF
Gated on alveolar Mφ low Liberase, 30 min low Liberase, 60 min high Liberase, 60 min high Liberase, 30 min
CD11c
CD11c
Fig. 1 Identification of macrophages and other immune cell types in the lung. (a) Gating strategy to identify different immune cell types in single-cell suspensions of the digested lung. Cell debris, doublets, and dead cells are excluded based on their forward-sideward scattering properties and a live/dead staining. A “pseudodump” gate on CD3e-, CD103-, and Ly6G-positive cells and subsequent gating on CD11b and CD11c helps to identify T cells, cDC1, and neutrophils. The remaining cells are gated for alveolar Mφ based on their SiglecF and CD11c expression. Non-alveolar Mφ are gated for CD11b+ myeloid cells and further on Ly6C+ inflammatory monocytes and CD64+ interstitial Mφ. Finally, eosinophils can be identified by their SiglecF expression and the remaining fraction includes Ly6C- monocytes and cDC2 (not shown). (b) CD11c expression on alveolar Mφ after 30 min or 60 min of digestion in buffers with low or high Liberase TM concentration (low Liberase = 50 μg/mL; high Liberase = 250 μg/mL)
in beta 2 integrin expression, i.e., CD11b and CD11c. Neutrophils are positive for CD11b but negative for CD11c, while cDC1 express CD11c, but lack CD11b. T-cells express neither of both (see Note 23). 5. “Alveolar Mφ” express high levels of SiglecF and CD11c (see Note 24). 6. Other myeloid cells can be identified through an initial CD11b gating (“myeloid”) (see Note 25) and their CD64 (“interstitial Mφ”), Ly6C (“inflammatory monocytes”) (see Note 26), or SiglecF (“eosinophils”) (see Note 27) expression.
356
Sebastian Baasch et al.
A
1. Gated on living single cells
GFP
Infected
3b. Gated on Ly6G-CD3-CD103-
2. Gated on uninfected Uninfected
4. Gated on non-alveolar Mφ
SiglecF
CD64
CD11b
SiglecF
Ly6G CD3 CD103
2. Gated on infected
4a. 4 Gat ae ed o n Gated on L y6G 6G 6G+CD3 CD D3+C CD10 D 03+ CD103 Ly6G
6. Gated on CD64-Ly6C-
5. Gated on myeloid interstitial Mφ
alveolar Mφ
Eosinophils CD11b myeloid inflammatory monocytes CD11c
Ly6C
CD11c
Ly6C
CD11c
CD11b
3a. Gated on Ly6G-CD3-CD103Neutrophils cDC1 T-cells
CD11c
B
80
CD11b
100 60
CD11c
CD11c
CD11c
CD64
CD11b
0
SiglecF
20
Ly6G CD3 CD103
40
al in veo te l a rs r t M M itia φ on l M oc φ y N c tes eu D tro C1 ph ils
% of infected cells
infected myeloid cells (LysMcre/+)
Ly6C
Ly6C
Fig. 2 Characterization of MCMV-infected cells in the lung. LysMcre/+ mice were infected with MCMVLSL-GFP and analyzed 12 h post infection. (a) Gating strategy to characterize MCMV-infected cells and uninfected immune cells in single cell suspensions of the digested lung. Infected cells were identified based on their GFP expression after exclusion of cell debris, doublets, and dead cells as described in Fig. 1a. In the interest of illustration, same gates were used as in Fig. 1a with each gate originating from “infected cells” to determine the surface marker expression of infected cells. “Uninfected cells” were gated as in Fig. 1a. (b) Quantification of MCMV-infected cell types (left) and representative flow cytometry of sequential gating strategy (initially gated from “infected cells” (Fig. 2a); right) 3.3.2 Host-Pathogen Fate Mapping Enables Identification of Infected Cells
To identify initially infected cells of the myeloid lineage, LysMcre/+ mice were infected with MCMVLSL-GFP intranasally and analyzed 12 h post infection (see Note 28).
Host-Pathogen Fate Mapping of Lung Macrophages
357
1. Single live cells are gated for as described in Subheading 3.3.1. 2. Infected cells can be identified as GFP-positive, while uninfected cells are GFP-negative (Fig. 2a, see Note 29). 3. Subsequently, infected and uninfected lung immune cells were characterized as described in Subheading 3.3.1 (Fig. 1a).
4
Notes 1. To assess the exact weight of each mouse is important to exactly dose the anesthetics. Application of too little anesthetics leads to insufficient suppression of reflexes and subsequent difficulties to install the viral suspension intranasally. Application of too much anesthetics leads to prolonged recovery with hypothermia and hypoglycemia. 2. Although the anatomical localization of the caecum shows interindividual variation in mice, in our experience, most of the mice have their caecum localized central or on the left side. Thus, we perform intraperitoneal injections in the right caudal quadrant of the abdomen to avoid puncture of the caecum. 3. It is crucial for reproducible installation of virus suspension to wait until a deep stage of anesthesia is reached. A good indicator for sufficient anesthesia is to check spinal reflexes, i.e., the deep pain perception. 4. Hold mice in your palm and slightly stretch head backward using your thumb and index finger. Apply virus suspension slowly with a pipette onto the nostrils waiting until the suspension is breathed in. 5. We do not recommend to sacrifice mice via CO2 inhalation. CO2 exposure leads to hyperventilation and lung emphysema that will compromise, for example, histological analysis. Moreover, the elevated concentration of CO2 will likely affect cellular metabolism, transcriptome, and overall phenotype. Cervical dislocation may lead to rupture of carotid arteries and/or jugular veins that will affect whole body perfusion if needed. In rare cases of additional tracheal damage, blood may run in the lower respiratory tract, which would also compromise the analysis of cellular composition in the lung. Injection with an overdose of anesthetics (e.g., pentobarbital) represents the method of choice with the least experimental interferences but may underlie the narcotics law regulations. 6. The application of liquid disinfectant does not only reduce microbial burden (not sterile!) on the skin and fur of the mouse, but prevents dry hair to attach to the extracted lungs. 7. First cut the skin caudally (“below”) of the xiphoid process of the sternum to visualize the peritoneum. Detach the skin from
358
Sebastian Baasch et al.
the peritoneum and thorax with the blunt sides of the scissor blade. Cut the lose skin to have a clear view of the thorax and peritoneum. Open the peritoneum with scissors. 8. Carefully perforate the diaphragm, so the lung will collapse. Now cut the diaphragm along the costal arch without a high risk to damage the lungs. 9. Open the thorax with scissors that are dedicated to cut through bones, such as rips. Scissors used for soft tissue will become blunt. 10. We recommend to cut into the liver (generous incision) to discharge PBS if the pulmonary circulation is perfused with high pressure and the incision of left atrium is not sufficient. The exact lobe of the liver, which is incised, is not important. In general, it should be avoided to apply too much pressure as it may cause pulmonary edema and potential loss of alveolar Mφ. 11. Perfuse through the right cardiac ventricle to clear the pulmonary circulation of blood until the lung appears pale. In our experience 1–3 mL are sufficient. 12. In mice, the right lung consists of four lobes: cranial, middle, caudal, and the accessory. We use these four lobes for FACS analysis. The removal of the right lung impedes re-inflation of the remaining lung. Thus, we do not recommend to use the collapsed left lung for microscopy. Instead, it can be used for other analyses, e.g., plaque or colony forming unit quantification or molecular analyses. 13. Directly decant the cut lung into the centrifugation tube, without prior vortexing to avoid a high proportion of lung pieces to remain in the reaction tube. If lung pieces stick to the centrifugation tube, pulse centrifuge. 14. The time for digestion varies across different published protocols between 30 and 60 min. We observed slightly higher yield of stromal cells and interstitial Mφ without increased cell death if the lung homogenate was incubated longer. However, the time of digestion needs to be reconsidered when determining the concentration of Liberase TM in the digestion buffer, as it highly impacts on CD11c surface expression especially (Fig. 1b). 15. This step will contribute to obtain a single cell suspension via the separation of loosly attached cell aggregates. 16. Do not invert again to avoid loss of the cell pellet. 17. Despite perfusion there might be minimal residual blood, and high erythrocyte numbers hamper the flow cytometry acquisition. This further needs to be considered, when quantifying, e.g., neutrophil or monocyte numbers.
Host-Pathogen Fate Mapping of Lung Macrophages
359
18. The number of cells increases especially in the acute phase of infection. In our experience 50 μL (out of 1000 μL) will result in a sufficient cell/antibody ratio (determined by staining) under control and infection conditions. This also allows to use several flow cytometry panels including the appropriate staining controls (e.g., single stains, fluorescence minus one (FMO)). 19. Excessive amounts of antibodies are added to the samples for flow cytometry stainings. Thus, low-affinity Fc receptors (CD16 and CD32) that are expressed by Mφ, monocytes, dendritic cells, and B-cells may bind to the Fc region of an antibody and lead to unspecific staining. This can result in false inclusion or exclusion of cells. Thus, “Fc block” is required to avoid unspecific binding of FcγR-expressing cells to antibodies. 20. Double the amount of antibodies is required, because the final volume of the cell suspension including the antibody master mix is 100 μL (2×50 μL). 21. We recommend to prepare unstained, single-stain, and FMO controls for the establishment of a new protocol. 22. We observe significant cell death during the digestion procedure (regardless of time or enzyme composition); therefore, we highly recommend to use a viability dye. As an alternative to fixable viability dyes, DNA-binding stains, such as propidium iodide (PI) or 4,6-diamino-2-phenylindole (DAPI), can be added to the sample 5 min before acquisition. 23. Moreover, the addition of anti-NK1.1 (natural killer cells) and anti-CD19 (B cells) in the same channel as anti-CD3e, antiCD103, and anti-Ly6G can provide a higher resolution of myeloid immune cells. Neutrophils and natural killer cells both express CD11b but can be discriminated due to the high sideward scatter of neutrophils. T and B cells could not be discriminated in this experimental setup. 24. Alveolar Mφ are highly autofluorescent, likely due to their constant uptake of lipid-containing surfactant. Thus, we highly recommend FMO controls. 25. Endothelial cells also express Ly6C but are CD11b-negative. 26. Upon infection, inflammatory monocytes invade the tissue and will rapidly upregulate CD64. 27. Upon activation, alveolar Mφ can upregulate CD11b; thus, we recommend a gating strategy that does not rely on low CD11b expression for the identification of alveolar Mφ during infection. 28. MCMV-infected Mφ downregulate characteristic surface markers within 24 h post infection, which compromises their identification [41]. However, Mφ can still be identified as such 12 h after infection. Once the STOP signal from MCMVLSL-GFP has
360
Sebastian Baasch et al.
been removed by recombination, i.e., infecting LysMcre/+ cells, viral progenies will express GFP regardless of the expression of a cre-recombinase in newly infected cells. This allows to analyze viral spread but requires a second endogenous reporter for initially infected cells, such as ROSA26LSL-Tomato [41]. 29. The GFP reporter of MCMVLSL-GFP is under the control of the human CMV IE promotor, which ensures fast and strong transcription. Thus, a reliable GFP signal can be detected as soon as 6 h after infection based on in vitro experiments with bone marrow-derived Mφ.
Acknowledgments SB was funded by the Hans A. Krebs Medical Scientist Programme (Faculty of Medicine, University of Freiburg). Support was provided by the German Ministry of Education and Research (BMBF; grants 01EO0803, 01GL1746A, and 01EK1602A) and the German Research Council (DFG; grants HE3127/9, HE3127/12, HE3127/16, TRR 167 – Project ID 259373024, and TRR 359 - Project ID 491676693). References 1. van Furth R, Cohn ZA (1968) The origin and kinetics of mononuclear phagocytes. 128:415– 435 2. Okabe Y, Medzhitov R (2016) Tissue biology perspective on macrophages. 17:9–17 3. Ble´riot C, Chakarov S, Ginhoux F (2020) Determinants of resident tissue macrophage identity and function. 52:957–970 4. Chakarov S, Lim HY, Tan L et al (2019) Two distinct interstitial macrophage populations coexist across tissues in specific subtissular niches. 363:eaau0964 5. Schyns J, Bai Q, Ruscitti C et al (2019) Non-classical tissue monocytes and two functionally distinct populations of interstitial macrophages populate the mouse lung. 10:1– 16 6. Gibbings SL, Thomas SM, Atif SM et al (2017) Three unique interstitial macrophages in the murine lung at steady state. 57:66–76 7. Kolter J, Feuerstein R, Zeis P et al (2019) A subset of skin macrophages contributes to the surveillance and regeneration of local nerves. 50:1482–1497. e1487 8. De Schepper S, Verheijden S, AguileraLizarraga J et al (2018) Self-maintaining gut macrophages are essential for intestinal homeostasis. 175:400–415. e413 9. Ural BB, Yeung ST, Damani-Yokota P et al (2020) Identification of a nerve-associated, lung-resident interstitial macrophage subset
with distinct localization and immunoregulatory properties. 5:eaax8756 10. Bedoret D, Wallemacq H, Marichal T et al (2009) Lung interstitial macrophages alter dendritic cell functions to prevent airway allergy in mice. 119:3723–3738 11. Kawano H, Kayama H, Nakama T et al (2016) IL-10-producing lung interstitial macrophages prevent neutrophilic asthma. 28:489–501 12. Guilliams M, De Kleer I, Henri S et al (2013) Alveolar macrophages develop from fetal monocytes that differentiate into long-lived cells in the first week of life via GM-CSF. 210: 1977–1992 13. Yu X, Buttgereit A, Lelios I et al (2017) The cytokine TGF-β promotes the development and homeostasis of alveolar macrophages. 47: 903–912. e904 14. Schneider C, Nobs SP, Kurrer M et al (2014) Induction of the nuclear receptor PPAR-γ by the cytokine GM-CSF is critical for the differentiation of fetal monocytes into alveolar macrophages. 15:1026–1037 15. Huffman JA, Hull WM, Dranoff G et al (1996) Pulmonary epithelial cell expression of GM-CSF corrects the alveolar proteinosis in GM-CSF-deficient mice. 97:649–655 16. Gschwend J, Sherman SP, Ridder F et al (2021) Alveolar macrophages rely on GM-CSF from alveolar epithelial type 2 cells before and after birth. 218:e20210745
Host-Pathogen Fate Mapping of Lung Macrophages 17. Cohen M, Giladi A, Gorki A-D et al (2018) Lung single-cell signaling interaction map reveals basophil role in macrophage imprinting. 175:1031–1044. e1018 18. Branchett WJ, Cook J, Oliver RA et al (2021) Airway macrophage-intrinsic TGF-β1 regulates pulmonary immunity during early-life allergen exposure. 147:1892–1906 19. Koth LL, Alex B, Hawgood S et al (2007) Integrin β6 mediates phospholipid and collectin homeostasis by activation of latent TGF-β1. 37:651–659 20. Munger JS, Huang X, Kawakatsu H et al (1999) A mechanism for regulating pulmonary inflammation and fibrosis: the integrin αvβ6 binds and activates latent TGF β1. 96:319–328 21. Huang X, Wu J, Zhu W et al (1998) Expression of the human integrin β 6 subunit in alveolar type II cells and bronchiolar epithelial cells reverses lung inflammation in β 6 knockout mice. 19:636–642 22. Rauschmeier R, Gustafsson C, Reinhardt A et al (2019) Bhlhe40 and Bhlhe41 transcription factors regulate alveolar macrophage selfrenewal and identity. 38:e101233 23. McCowan J, Fercoq F, Kirkwood PM et al (2021) The transcription factor EGR2 is indispensable for tissue-specific imprinting of alveolar macrophages in health and tissue repair. 6: eabj2132 24. Trapnell BC, Nakata K, Bonella F et al (2019) Pulmonary alveolar proteinosis. 5:1–17 25. Schneider C, Nobs SP, Heer AK et al (2014) Alveolar macrophages are essential for protection from respiratory failure and associated morbidity following influenza virus infection. 10:e1004053 26. Mass E, Ballesteros I, Farlik M et al (2016) Specification of tissue-resident macrophages during organogenesis. 353:aaf4238 27. Gomez Perdiguero E, Klapproth K, Schulz C et al (2015) Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. 518:547–551 28. Hashimoto D, Chow A, Noizat C et al (2013) Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes. 38:792– 804 29. Yona S, Kim KW, Wolf Y et al (2013) Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis. 38:79–91 30. Li F, Piattini F, Pohlmeier L et al (2022) alveolar macrophages Monocyte-derived
361
autonomously determine severe outcome of respiratory viral infection. 7:eabj5761 31. Sauer B, Henderson N (1988) Site-specific DNA recombination in mammalian cells by the Cre recombinase of bacteriophage P1. 85: 5166–5170 32. Indra AK, Warot X, Brocard J et al (1999) Temporally-controlled site-specific mutagenesis in the basal layer of the epidermis: comparison of the recombinase activity of the tamoxifen-inducible Cre-ERT and Cre-ERT2 recombinases. 27:4324–4327 33. Ginhoux F, Greter M, Leboeuf M et al (2010) Fate mapping analysis reveals that adult microglia derive from primitive macrophages. 330: 841–845 34. Kim J-S, Kolesnikov M, Peled-Hajaj S et al (2021) A binary Cre transgenic approach dissects microglia and CNS border-associated macrophages. 54:176–190.e177 35. Hirrlinger J, Scheller A, Hirrlinger PG et al (2009) Split-cre complementation indicates coincident activity of different genes in vivo. 4:e4286 36. Liu Z, Gu Y, Chakarov S et al (2019) Fate mapping via Ms4a3-expression history traces monocyte-derived cells. 178:1509–1525. e1519 37. Machiels B, Dourcy M, Xiao X et al (2017) A gammaherpesvirus provides protection against allergic asthma by inducing the replacement of resident alveolar macrophages with regulatory monocytes. 18:1310–1320 38. Misharin AV, Morales-Nebreda L, Reyfman PA et al (2017) Monocyte-derived alveolar macrophages drive lung fibrosis and persist in the lung over the life span. 214:2387–2404 39. Aegerter H, Kulikauskaite J, Crotta S et al (2020) Influenza-induced monocyte-derived alveolar macrophages confer prolonged antibacterial protection. 21:145–157 40. Baasch S, Ruzsics Z, Henneke P (2020) Cytomegaloviruses and macrophages—friends and foes from early on? 11:793 41. Baasch S, Giansanti P, Kolter J et al (2021) Cytomegalovirus subverts macrophage identity. 184:3774–3793. e3725 42. Farrell HE, Lawler C, Oliveira MT et al (2016) Alveolar macrophages are a prominent but nonessential target for murine cytomegalovirus infecting the lungs. 90:2756–2766 43. Tegtmeyer P-K, Spanier J, Borst K et al (2019) STING induces early IFN-β in the liver and constrains myeloid cell-mediated dissemination of murine cytomegalovirus. 10:1–12
Chapter 25 Measuring the Metabolic State of Tissue-Resident Macrophages via SCENITH Andrea Vogel, Paulina Garcı´a Gonza´lez, and Rafael J. Argu¨ello Abstract Functional reprograming of cells is linked to a process of metabolic rewiring that is adapted for such new functions or microenvironment. Macrophages are present in all tissues and exposed to different microenvironments throughout our body. Profiling energetic metabolism of tissue resident and other heterogeneous populations of macrophages in vitro and ex vivo is technologically very challenging. We have recently developed a method to functionally profile energetic metabolism with single-cell resolution, named SCENITH. This method can be performed rapidly ex vivo and does not require specialized equipment. In this book chapter, we will summarize the tissue processing, the procedure and methods, the analysis and example of results, and a series of frequently asked questions. Key words Immunometabolism, Protein synthesis, Single-cell resolution, Energetic metabolism, Macrophages, Flow cytometry, FACS, Mitochondrial dependence, Glycolytic capacity, Metabolic dependencies, SCENITH
1
Introduction Tissue-resident macrophages are specialized innate immune cells, populating almost every organ of the body. Macrophages are extremely dynamic and can adopt a plethora of functional activation states, orchestrating homeostatic, inflammatory, and reparative responses within their tissue niche. Disturbance of these “gatekeeping” functions has been described to promote loss of tissue functionality and disease development [1–3]. Importantly, acquisition of different functional myeloid phenotypes has been fundamentally linked to the reprogramming of cell intrinsic metabolic processes [4, 5]. Moreover, tissue intrinsic factors such as oxygen levels, nutrient availability, and metabolic crosstalk with adjacent cells can further induce metabolic rewiring in tissue-resident macrophages [6–8]. To produce energy, macrophages are equipped with a series of interconnected catabolic pathways, breaking down
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_25, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
363
364
Andrea Vogel et al.
nutrients and producing energy in the form of ATP. In general, energetic metabolism requires two main ATP-producing pathways, glycolysis, and mitochondrial oxidative phosphorylation (OXPHOS) [5]. In turn, produced ATP is subsequently consumed by anabolic processes such as protein synthesis. Indeed, protein synthesis is one of the most expensive metabolic activities of most cells, and energy production and protein synthesis rates are tightly linked. This concept represents the underlying methodological principle of SCENITH [9], where after blocking a metabolic pathway, the drop in energy is measured by a drop in protein. To measure protein synthesis, puromycin incorporation is a rapid, stable, and convenient readout, due to the capacity to be measured by flow cytometry. Overall, SCENITH allows an easy and rapid profiling of global metabolic dependencies and capacities of tissue in vitro-derived but also resident macrophages. Other current methods to functionally measure cellular metabolism and to profile metabolic dependencies of macrophages require large amounts of purified cells which need to be cultured in special media formulations over several hours [10]. A major limitation of the field is that metabolic states are unstable and sample pre-processing such as buffers for tissue digestion, cell sorting to extract macrophage populations, and following incubations in artificial cell culture environments can rapidly change metabolic activities of macrophages. Further, present techniques are mostly performed as bulk analysis and thereby often fail to resolve cellular heterogeneity. In contrast, SCENITH represents a simple method for complex immune-metabolic profiling of heterogeneous samples with single-cell resolution, allowing the metabolic characterization of rare macrophage populations. By combining treatments with metabolic inhibitors, puromycin incorporation and standard flow cytometry of freshly isolated cell populations, SCENITH enables the functional analysis of metabolic dependencies and capacities of the cells while avoiding metabolic bias introduced by timeconsuming cell sorting, purification, or extended ex vivo culturing periods.
2
Materials
2.1 Preparation of Buffers and Cells
1. FACS buffer: 2% FCS, 2 mM EDTA in 1× PBS. 2. Red blood cell lysis buffer: Dissolve NH4Cl (155 mM), KHCO3 (10 mM), and Na2EDTA (0.1 mM) in 1000 mL ddH2O, and adjust pH to 7.2–7.4 with HCl. Sterile filter the lysis buffer and store at 4 °C. Alternatively, commercially available red blood cell lysis buffers can be used.
Metabolic Profiling of Macrophages by SCENITH
365
3. Enzyme digest buffer: HBSS (without calcium/magnesium), 5% FCS, 10 μM HEPES, 2 mg/mL collagenase A, and 28 U/ mL DNase I, prepare 5 mL digest buffer per sample. 4. Debris removal solution (Miltenyi Biotec, Cat. Number 130-109-398). 2.2 SCENITH Kit (GammaOmics, Cat. Number GO-0002PE50; GO-0002AF488-50; GO-0002AF647-50) (see Note 1)
1. DMSO in PBS. 2. 100 mM 2-deoxyglucose and DMSO in PBS. 3. 1 μM oligomycin in PBS. 4. 2 μg/mL Harringtonine in PBS. 5. 10–20 μM/mL puromycin in PBS (seeNote 2). 6. 100X (pre-titrated) R4743L-E8 conjugated with PE, AF488, or AF647.
2.3 Viability/FcBlocking and Surface Staining
1. Viability staining/Fc-blocking solution: Dilute viability staining concentrate and Fc-blocking concentrate in FACS buffer or PBS according to the manufacturer’s instructions (seeNote 3). 2. 2× Surface staining solution: dilute fluorochrome-conjugated antibodies at 2× concentration in FACS buffer (seeNote 4). The following antibodies (0.2 μg/mL final concentration) have been used to stain for peritoneal macrophages and microglia: Anti-mouse CD45.2 BV650 (clone: 104). Anti-mouse/human B220 AF700 (clone: RA3-6B2). Anti-mouse CD3 AF700 (clone: 500A2). Anti-mouse Ly6G AF700 (clone: 1A8). Anti-mouse CD11b APC (clone: M1/70). Anti-mouse CD11b PE (clone: M1/70). Anti-mouse F4/80 APC (clone: BM8). Anti-mouse P2RY12 PE (clone: S16007D).
2.4 Intracellular Staining
1. Fixation/permeabilization buffer working solution: Mix one part of fix/perm concentrate with three parts fixation/permeabilization diluent (seeNote 5). 2. Permeabilization buffer working solution: Mix one part of 10× permeabilization buffer with nine parts of ddH2O. 3. Intracellular blocking solution: Mix one part of 10× permeabilization buffer with two parts of FCS and seven parts of ddH2O. 4. Anti-puromycin staining solution: 1:100 dilution of antipuromycin antibody (AF488 or AF647 conjugated, clone: R4743L-E8; from the SCENITH kit) in intracellular blocking solution (seeNote 1).
366
Andrea Vogel et al.
2.5 Equipment and Machines
1. 1.5 mL Eppendorf tubes or non-tissue culture treated 96-well plates. 2. Incubator. 3. Flow cytometer.
3
Methods
3.1 Preparation of Single-Cell Suspensions
1. Isolate tissue of interest and dissociate with a suitable mechanical/enzymatic dissociation protocol (seeNotes 6 and 7). (a) For peritoneal macrophage isolation, inject 5 mL of PBS into the peritoneal cavity of a mouse by using a 27G needle. Subsequently, collect peritoneal lavage by using a 23G needle. Pass cell suspension through a 70 μM strainer, and spin down at 500 g for 5 min at room temperature. (b) For microglia, isolate the brain and mince into small tissue pieces. Transfer tissue in 5 mL enzyme digest buffer, and incubate at 37 °C for 30 min under agitation. After successful tissue dissociation, filter cell suspension through a 70 μM strainer. To remove cell and myelin debris, centrifuge cells at 300 g for 10 min at 4 °C, aspirate the supernatant, and resuspend pellet in 3100 μL ice-cold PBS. Transfer the cell suspension to a fresh 15 mL Falcon, and add 900 μL of debris removal solution. Mix well by pipetting up and down ten times. Carefully overlay 4 mL of ice-cold PBS onto the cell solution, creating a gradient. Centrifuge at 3000 g for 10 min at 4 °C with maximum acceleration and maximum break. Remove and discard the top two layers containing myelin and cell debris. Wash cells by adding 14 mL of ice-cold PBS, and spin down at 1000 g for 10 min at 4 °C. 2. Resuspend cells in appropriate medium at a cell density of 5 × 105–2 × 107 cells/mL.
3.2 Inhibitor and Puromycin Treatment
1. For each condition (control, DG, O, DGO, H), transfer 5 × 104–2 × 106 cells in a tube or in a well of a 96-well plate. Experimental duplicates or triplicates should be performed in all conditions. From this point on, try to keep cells at 37 °C and 5%CO2 as long as possible, and avoid long handling periods outside the incubator. 2. Equilibrate cells to 37 °C and 5%CO2 by putting the tubes or plate in the incubator for 20–30 min. 3. Add the corresponding volume of the inhibitor solution to get 1× inhibitor (Co, DG, O) to the respective tubes/wells and
Metabolic Profiling of Macrophages by SCENITH
367
Fig. 1 SCENITH protocol diagram. SCENITH allows the metabolic profiling of cells at a single-cell level by monitoring rapid changes in protein synthesis upon metabolic inhibition, using puromycin incorporation as a translation readout. First, to inhibit ATP production, a cell suspension of macrophages in culture media is treated with 2-deoxy-D-glucose (DG), oligomycin, or other metabolic inhibitors for 10 min to block glycolysis, OXPHOS or other metabolic pathways, respectively. Treatment with DG followed by a treatment with oligomycin (DGO) is also performed to block both glycolysis and OXPHOS simultaneously (complete blockade). Next, cells are treated with puromycin to allow its incorporation into nascent proteins by active ribosomes. Cells are then fixed, permeabilized and stained for intracellular detection of puromycin to be further analyzed by flow cytometry. Cells incubated with the control reagent without inhibitor treatment are used as positive control for basal protein translation levels. The SCENITH protocol is compatible with the use of other metabolic inhibitors for a more detailed metabolic profiling
incubate for 10 min at 37 °C, 5%CO2 in the incubator (i.e., if cells are in a volume of 100 μL add 5 μL of 20× inhibitor solution) (Fig. 1). 4. After 10-min incubation, add the corresponding volume of the 20× DG solution just to the DGO tubes/wells. Mix cell suspension by vortexing or pipetting up and down, and put tubes/plate back at 37 °C, 5%CO2 for additional 5 min (Fig. 1). 5. Without washing add the corresponding volume of the 20× puromycin solution to get a 1× puromycin dilution (i.e., if cells are in a volume of 100 μL, add 5 μL of 20× puromycin solution). Mix cell suspension by gently vortexing or pipetting up and down, and put tubes/plate back at 37 °C, 5%CO2 for 30 min (Fig. 1) (seeNote 8).
368
Andrea Vogel et al.
6. After the incubation immediately put tubes or plate on ice, and add ice-cold FACS buffer, and spin down cell suspension at 400 g for 7 min at 4 °C. Discard the supernatant by aspiring. If the assay is performed in a 96-well plate, repeat the washing step a second time. 3.3 Red Blood Cell Lysis (Optional)
1. If required resuspend cell pellets in red blood cell lysis buffer, mix well, and incubate for 15 min at RT. 2. Fill up tubes or wells with FACS buffer, and centrifuge at 400 g for 7 min at 4 °C. Discard the supernatant by aspiring.
3.4 Fc-Block/ Viability and Surface Staining
1. Resuspend cell pellets in 100 μL of viability staining/Fcblocking solution, and incubate for 15 min at 4 °C in the dark. 2. Subsequently add 100 μL of the prepared 2× surface staining solution, and incubate for 25 min at 4 °C in the dark. 3. Wash cells by filling up tubes or wells with FACS buffer, and spin down cell suspension at 400 g for 7 min at 4 °C. Discard the supernatant by aspiring. If the assay is performed in a 96-well plate, repeat the washing step a second time.
3.5 Intracellular Staining
1. Resuspend cells in 100 μL of fixation/permeabilization buffer working solution, mix well by vortexing or pipetting up and down, and incubate for 20 min at room temperature in the dark. 2. Subsequently, fill up tubes or wells with permeabilization buffer working solution, and spin down cell suspensions at 600 g for 7 min at room temperature. Discard the supernatant by aspiring. If the assay is performed in a 96-well plate, wash cells twice with permeabilization buffer working solution by resuspending the cells in permeabilization buffer working solution followed by centrifugation at 600 g for 7 min at room temperature. After spinning down cells, carefully aspirate and discard the supernatant. 3. Resuspend cells in 50 μL intracellular blocking solution, and incubate for 10 min at room temperature. 4. Without washing add 50 μL of the anti-puromycin staining solution, and incubate for 1 h at 4 °C in the dark (seeNote 9). 5. Fill up tubes or wells with permeabilization buffer working solution, and spin down cell suspensions at 600 g for 7 min at 4 °C. Discard the supernatant by aspiring. If the assay is performed in a 96-well plate, wash cells twice with permeabilization buffer working solution (as indicated in step 2). 6. Resuspend stained cells in 150–200 μL of FACS buffer. 7. Analyze samples by flow cytometry (seeNote 10).
Metabolic Profiling of Macrophages by SCENITH
369
Fig. 2 Representative gating strategy and translation levels of peritoneal macrophages and microglia upon metabolic inhibitor treatment. (a) peritoneal macrophages are gated as live singlets, CD45+B220-CD3-Ly6G-CD11b+F4/80+. (b) Puromycin MFI of peritoneal macrophages treated with control (Co), 2-deoxyglucose (DG), oligomycin (O), 2-deoxy-D-glucose + oligomycin (DGO), or Harringtonine (H). (c) Microglia are gated as live singlets, CD45+B220-CD3-Ly6G-CD11b+F4/80+. (d) Puromycin MFI of microglia treated with the different inhibitors 3.6
Analysis
1. Gate on your macrophage population of interest by your previously established gating strategy. We analyze peritoneal macrophages by gating on microglia by gating on live singlets, CD45+B220-CD3-Ly6G-CD11b+F4/80+, and microglia are defined as live singlets, CD45+B220-CD3-Ly6G-CD11b+P2RY12+ (Fig. 2a). 2. Extract the following geometric MFIs of the channel with the anti-puromycin antibody (Fig. 2b–d). Co = GeoMFI of anti-puromycin fluorophore upon treatment with control. DG = GeoMFI of anti-puromycin fluorophore upon treatment with 2-deoxyglucose.
370
Andrea Vogel et al.
O = GeoMFI of anti-puromycin fluorophore upon treatment with oligomycin. DGO = GeoMFI of anti-puromycin fluorophore upon treatment with 2-deoxyglucose + oligomycin. 3. To quantify the parameters of energy metabolism, the relative impacts of inhibiting a certain bioenergetic pathway are compared to the total inhibition of ATP production, or protein synthesis (Harringtonine), and put into simple algorithms to express relative pathway dependencies and compensation capacities. The percentual glucose dependence is defined by how much protein synthesis activity is dependent on glucose oxidation (glycolysis and glucose-derived acetyl-CoA). It is calculated as the difference in translation levels upon 2-deoxyglucose treatment (DG) and translation levels of control cells (Co), divided by the difference between complete inhibition of ATP production by combinational 2-deoxyglucose and oligomycin treatment (DGO) and control cells (Co). Similarly, the percentual mitochondrial dependence quantifies how much translation is decreased upon inhibition of mitochondrial oxidative phosphorylation and is calculated as the difference in translational activity between cells post oligomycin treatment (OG) and control cells (Co), relative to the difference between complete inhibition of ATP production by combinational 2-deoxyglucose and oligomycin treatment (DGO) and control cells (Co). Additionally, the metabolic parameters glycolytic capacity and fatty acid oxidation and amino acid oxidation (FAO/AOO) can be calculated. Capacities are defined as the maximum compensatory capacity of cells to switch to alternative metabolic cues when another one is inhibited. Glycolytic capacity describes the sustained protein synthesis activity upon inhibition of mitochondrial oxidative phosphorylation, and FAO/AOO capacity quantifies the cellular capacity to exploit fatty acid and amino acid utilization for energy production when glucose oxidation is inhibited. 4. In brief, glucose dependence, mitochondrial dependence, glycolytic capacity, and fatty acid oxidation/amino acid oxidation capacity can be calculated by the equations shown in Fig. 3 (seeNote 10). 5. In the case just one sample per condition can be used, the standard deviations of metabolic dependencies and capacities in cell populations can be calculated. For the respective algorithms please refer to [9]. 6. Two important metabolic parameters that can be informative besides the percentual mitochondrial and glucose dependence (Fig. 3) are the basal level of protein synthesis and the absolute
Metabolic Profiling of Macrophages by SCENITH
371
Fig. 3 SCENITH data interpretation and representation. (a) The bar graphs represent the anti-puromycin MFI of the two populations of cells treated or not with the different metabolic inhibitors (i.e., Co = control; DG = 2deoxy-D-glucose; O = oligomycin). Different normalized metabolic parameters can be extracted (1 to 4); but also the absolute level of translation and the absolute maximum glycolytic capacity and FAO & AAO can be calculated. (b) The glucose dependence (%) and mitochondrial dependence (%) can be represented in an X vs Y graph (left panel). Also the absolute max. Glycolytic capacity (X) vs the basal level of translation (Y) can be represented, where cells with high glycolytic capacity (%) are close to the grey dashed line of slope = 1 (Y = X)
maximum glycolytic capacity. The basal level of translation is calculated as the anti-puromycin MFI of Co minus DGO (or Harringtonine treatment) and the absolute glycolytic capacity as the anti-Puromycin MFI of O minus DGO (or Harringtonine treatment). These parameters can be represented as shown in Fig. 3b. As can be observed in the graph, the cells with low mitochondrial dependence and medium or high glucose dependence are the cells with high glycolytic capacity (i.e., neutrophils). When doing the graph of basal translation versus Abs. maximum glycolytic capacity, the cells that are close to the imaginary line of X = Y are the cells that have high glycolytic capacity.
372
4
Andrea Vogel et al.
Notes 1. We recommend the use of the commercially available SCENITH kit that provides inhibitor combinations, puromycin stocks, and anti-puromycin antibodies. Future enhancements to this kit will include the addition of stabilized reagents, extending its shelf-life and improving ease of use. The kit’s set of metabolic inhibitors will be expanded to provide a broader range for metabolic profiling via SCENITH. The flexibility of this kit allows for the testing of other metabolic inhibitors or the incorporation of metabolites, broadening its utility for diverse experimental setups. 2. For optimal results with the SCENITH kit, prepare small aliquots of puromycin stock following the provided protocols. These aliquots should be stored at -20 °C and protected from light to maintain their stability. It is important to avoid repeated freeze-thaw cycles as these can adversely affect the performance of the reagents. Future iterations of the kit may improve the stability of these reagents, potentially reducing the need for stringent storage conditions. 3. The viability dye must be compatible with fixation and permeabilization buffers and preserve staining patterns after fixation and permeabilization of the cells. Further consider to also include control samples for live/dead cells by pooling freshly isolated cells along with freshly isolated cells that have been placed at 70 °C for 5 min at a 1:1 ratio. Fixable viability dyes stain amine reactive groups and some tissue resident macrophage populations such as Kupffer cells contain lots of membrane bound proteins, which increases the background and might result in suboptimal live/dead gating without proper controls [11]. 4. The fluorescent proteins and fluorophores used for surface staining must be compatible with fixation. Antibody clones used for mass cytometry should be compatible with the SCENITH protocol. Also include appropriate staining controls such as an unstained control for autofluorescence, single stainings and fluorescence minus one (FMO) controls to ensure proper compensation and gating. 5. The intracellular staining protocol is in general compatible with commonly used fixation and permeabilization buffer sets. However, we recommend to use fixation/permeabilization kits which are used to stain nuclear antigens and have a combined fixation/permeabilization step after the surface staining, followed by a subsequent additional permeabilization step.
Metabolic Profiling of Macrophages by SCENITH
373
6. To obtain a viable and metabolically active cell population from the tissue/organ of interest, the respective digestion and dissociation protocol should be tested and optimized using positive and negative controls (puromycin and Harringtonine + puromycin) prior to performing SCENITH. 7. Isolation procedures, incubation of tissues on ice, or stressful digestion protocols could result in decreased viability and metabolic activity of certain macrophage populations. In some cases, the use of tissue pieces before single-cell suspensions for incubation with the inhibitor and puromycin treatment can be recommended. To ensure proper inhibitor/puromycin diffusion inside the tissue, tissue pieces should be partially dissociated by vibratome slicing, chopping, or mincing into smaller pieces of approximately 200 μM. Subsequently, tissue pieces can be put in appropriate medium and inhibitor, and puromycin treatment can be performed as outlined in Subheading 3.2. The time and concentration of puromycin incubation needs to be optimized using control and harringtonine controls and may vary depending on the tissue. After the puromycin incorporation, tissue pieces can be fully dissociated by enzymatic digestion, followed by continuing the remaining protocol steps from Subheading 3.4. “Fc-block/viability and surface staining” onward. If working with frozen samples, first allow them to recover and stabilize after thawing before performing SCENITH. Directly treating thawed cells under cold shock stress with SCENITH inhibitors can lead to inconsistent results with a variable percentage of non-translating cells. 8. Puromycin incubation times depend on the metabolic activity of cells. A 15-min incubation with puromycin generally works fine for metabolically active cells like cell lines in vitro, but incubation times might need to be increased when working with resting cells extracted from tissues. The incubation time with puromycin must be optimized according to the sample (i.e., for blood the incubation time of puromycin is 40 min, more time is optimal for less metabolically active cells). In addition to extending the puromycin incubation time for less active cells, puromycin can be added directly after addition of the inhibitors without any waiting time in between. 9. The intracellular staining with the anti-puromycin antibody can also be combined with staining for additional nuclear/cytoplasmic proteins. 10. For questions regarding to the pattern of puromycin incorporation, please refer to the troubleshooting table (Table 1).
374
Andrea Vogel et al.
Table 1 Potential problems, causes, and suggestions for troubleshooting Problem
Potential cause
Puromycin stock was not correctly Signal for translation (Puro stored MFI) is too low or lower signal in control treatment
An inappropriate concentration of puromycin was used
Incubation time with puromycin was insufficient
Cell culture medium was depleted of nutrients Too many cells during staining
Suggestions Avoid exposure to light and freeze and thawing cycles. Prepare small aliquots of 20X puromycin, and discard after each use The needed puromycin concentration can be celldependent and should be determined experimentally for each sample being tested. Puromycin concentrations that are too high or too low can result in low staining signal Puromycin signal will depend on the metabolic state of the cell. Metabolically active cells will need less incubation times. Do a kinetic experiment of puromycin incubation times ranging from 15-30-45 min Change media 2–8 h before the experiment The amount of anti-puromycin antibody is recommended to stain max five million cells. If working with a higher number of cells, adapt the volume of antibody
If cells show increased translation Signal for translation (Puro Some metabolic inhibitors make upon treatment when MFI) is higher in cells cells switch to other metabolic calculating metabolic treated with inhibitors than sources to compensate. The dependency, the formula would the control activation of this compensatory result in a negative dependency mechanism might result in higher value. Conceptually, a negative levels of translation than the dependency is equal to no control dependency (0) As the inhibitor induces an increase in the translation levels and cells are only relying on the compensatory pathway, then the formula will reveal a percentual metabolic capacity that is over 100%. Values over 100% indicate that cells have the metabolic capacity to sustain even higher levels of translation than control cells (continued)
Metabolic Profiling of Macrophages by SCENITH
375
Table 1 (continued) Problem
Potential cause
Signal of surface markers decrease after treatment with inhibitors
An inappropriate time of treatment Time of treatment with inhibitors depends on the metabolic with inhibitors was performed activity of the cells and has to be determined experimentally In some cases, higher incubation times with inhibitors can lead to antigen internalization. Perform intracellular staining of internalized markers
Signal is present in DGO (all inhibitors)
Oligomycin acts faster than DG
High background from cells
Suggestions
Treat cells 10–15 min with 2-DG before adding oligomycin for 5–10 min. Use this measure as DGO treatment Using Fc block, FCS 2%, or BSA 1–3% during surface and intracellular staining can reduce background
Puromycin signal (MFI) increases after oligomycin treatment
Cells increased translation after mitochondrial inhibition
In some cells, mitochondrial blockade can lead to higher puromycin incorporation. We call this phenomenon MITA “mitochondrial inhibition translation activation” (data unpublished). This seems to link mitochondrial inhibition with activation of the mTOR pathway and allows to reveal a higher mitochondrial activity of cells
Signal doesn’t change with any kind of treatment
Inhibitors don’t work
After thawing, let inhibitors stabilize at 37 °C for 15–30 min before use Inhibitors concentration are celldependent and should be determined experimentally
The concentration and/or time of treatment are not optimized Lots of doublet cells during FACS acquisition
Cells were aggregated
Obtain a single-cell suspension by using 2 mM EDTA in FACS buffer
Viability decreases upon inhibitor treatment
The inhibitor concentrations are not optimized
Inhibitors concentration are celldependent and should be determined experimentally
376
Andrea Vogel et al.
Acknowledgments This work was supported by the French ANR JCJC-Epic-SCENITH ANR-20-CE14-0028; ANR PRC MetaNiche ANR-22PRC; CoPoC Inserm-transfert MAT-PI-17493-A-04 and H2020 Transcan ERANET–“TALETE” (to RJA); and CENTURI postdoctoral fellowship award (to PGG). Rafael Argu¨ello is a Marylou Ingran Scholar of the International Society for the advancement of cytometry (ISAC). References 1. Park MD, Silvin A, Ginhoux F, Merad M (2022) Macrophages in health and disease. Cell 185(23):4259–4279. https://doi.org/ 10.1016/j.cell.2022.10.007 2. Wynn TA, Chawla A, Pollard JW (2013) Macrophage biology in development, homeostasis and disease. Nature 496(7446):445–455. https://doi.org/10.1038/nature12034 3. Epelman S, Lavine KJ, Randolph GJ (2014) Origin and functions of tissue macrophages. Immunity 41(1):21–35. https://doi.org/10. 1016/j.immuni.2014.06.013 4. Russell DG, Huang L, VanderVen BC (2019) Immunometabolism at the interface between macrophages and pathogens. Nat Rev Immunol 19(5):291–304. https://doi.org/10. 1038/s41577-019-0124-9 5. Van den Bossche J, O’Neill LA, Menon D (2017) Macrophage immunometabolism: where are we (going)? Trends Immunol 38(6):395–406. https://doi.org/10.1016/j. it.2017.03.001 6. Wculek SK, Dunphy G, Heras-Murillo I, Mastrangelo A, Sancho D (2022) Metabolism of tissue macrophages in homeostasis and pathology. Cell Mol Immunol 19(3): 384–408. https://doi.org/10.1038/s41423021-00791-9 7. Man K, Kutyavin VI, Chawla A (2017) Tissue development, immunometabolism:
physiology, and pathobiology. Cell Metab 25(1):11–26. https://doi.org/10.1016/j. cmet.2016.08.016 8. Caputa G, Castoldi A, Pearce EJ (2019) Metabolic adaptations of tissue-resident immune cells. Nat Immunol 20(7):793–801. https:// doi.org/10.1038/s41590-019-0407-0 9. Arguello RJ, Combes AJ, Char R, Gigan JP, Baaziz AI, Bousiquot E, Camosseto V, Samad B, Tsui J, Yan P, Boissonneau S, Figarella-Branger D, Gatti E, Tabouret E, Krummel MF, Pierre P (2020) SCENITH: a flow cytometry-based method to functionally profile energy metabolism with single-cell resolution. Cell Metab 32(6):1063–1075. e1067. https://doi.org/10.1016/j.cmet.2020. 11.007 10. Llufrio EM, Wang L, Naser FJ, Patti GJ (2018) Sorting cells alters their redox state and cellular metabolome. Redox Biol 16:381–387. https://doi.org/10.1016/j.redox.2018. 03.004 11. Andreata F, Bleriot C, Di Lucia P, De Simone G, Fumagalli V, Ficht X, Beccaria CG, Kuka M, Ginhoux F, Iannacone M (2021) Isolation of mouse Kupffer cells for phenotypic and functional studies. STAR Protoc 2(4): 100831. https://doi.org/10.1016/j.xpro. 2021.100831
Chapter 26 Analyzing Fcγ-Receptor Interactions on Monocytes with the Proximity Ligation Assay (PLA) Sibel Kara and Falk Nimmerjahn Abstract Proximity ligation assays (PLA) enable the detection and characterization of protein interactions independent of protein abundance or genetic modifications. This technique exploits both antibody and DNA-binding features, providing high selectivity and sensitivity for protein recognition and visualization of single-protein molecules with high spatial accuracy. Here, we describe the general procedure for a direct PLA on splenic monocytes to analyze FcγRIIb homodimerization. However, this method can be applied to other cells and receptors of interest. Key words Proximity ligation assay (PLA), Antibody, Microscopy, Protein interaction, Fc receptor
1
Introduction Myeloid cells, such as monocytes and macrophages, have critical roles during steady state and infection ranging from regulation of metabolic functions [1, 2], orchestrating an immune response to fight invading pathogens [3], clearing cellular debris during tissue development [4], to wound healing and tissue remodeling [5]. Along those lines, monocytes and macrophages are also prominent immune cells in tumor microenvironments where they can act as effector cells in monoclonal antibody-based cancer therapies [6, 7]. Indeed, the clinical efficacy of various cancer therapies has been demonstrated to depend on Fcγ-receptors (FcγRs) expressed broadly on tumor infiltrating myeloid cell subsets [8, 9]. Fc receptors (FcRs) are transmembrane receptors, which have differing affinities for individual immunoglobulin G (IgG) subclasses via binding to their Fc fragment, and are widely expressed on innate and adaptive immune effector cells. In general, FcRs can be distinguished into canonical Type I FcRs, also known as classical FcγRs, which are members of the immunoglobulin superfamily, and into IgG glycosylation-specific Type II FcRs, which belong to
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_26, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
377
378
Sibel Kara and Falk Nimmerjahn
the C-type lectin family and detect anti-inflammatory sialic acidrich IgG glycosylation variants [9]. FcγRs play a crucial role in the immune system by serving as a linker between humoral and cellmediated immune responses. For instance, these receptors are involved in the regulation of effector cell functions ranging from phagocytosis or degranulation to antibody-dependent cellular cytotoxicity (ADCC). Intracellular FcγR signaling can be initiated by several activating FcγRs and is counter-acted by inhibitory FcγRIIb [9, 10]. This unique inhibitory FcγRIIb is expressed along with activating FcγRs on the majority of innate immune cells, including monocytes and macrophages, thereby negatively regulating cell activation [11]. Highlighting the critical function of FcγRIIb, several studies demonstrated that a diminished expression level or functional loss of FcγRIIb is correlated with enhanced susceptibility for autoimmunity [12–14]. These findings emphasize that understanding the biological function of checkpoint proteins such as FcγRIIb is essential to decipher the mechanisms of cellular immune regulation. On the cell membrane level, factors such as membrane composition, membrane environment, protein-protein interactions, and protein localization may play an important role in regulating immune receptor function. Protein-protein interactions are essential for cellular responses and are involved in the majority of cellular processes induced by surface molecules such as immune receptors [15]. Many methods are available to determine the interplay of proteins, including co-immunoprecipitation (Co-IP) [16], tandem affinity purification (TAP) [17], fluorescence resonance energy transfer (FRET) [18], and in situ proximity ligation assay (PLA) [19], but each method has distinct requirements and limitations. PLA is a unique method facilitating the recognition of the interplay of two proteins at highspatial and single-molecule resolution even at low expression levels and without the need for genetic manipulation. PLA is an advanced antibody-based immunoassay that enables highly specific and sensitive detection of endogenous proteins, even if they are expressed weakly or transiently [20–22]. This method makes use of singlestranded oligonucleotides covalently coupled to distinct antibodies identifying the target proteins of interest. If the labeled antibodies come in close proximity on the cell surface, a hybridization of complementary oligonucleotides occurs, followed by a polymerase chain reaction (PCR) with fluorogenic or chromogenic labeled nucleotides for amplification of the signal. As the name indicates, only interactions up to a proximity of 40 nm generate detectable PLA signals visualized as spots that can be recorded by microscopy followed by qualitative or quantitative analysis [20, 21]. Given the versatility of this technique, it allows to identify protein localization by combining the receptor of choice with different markers, such as staining of intracellular compartments [23] or the cell membrane [20], or with other methods to improve specificity, sensitivity, and target range such as flow cytometry [24] or enzyme-linked
Fcγ-R interplay Visualized with PLA
379
immunosorbent assay (ELISA) [25]. Additionally, PLA can be employed to study interactions between proteins within a single cell, among proteins located on the membrane of one cell, and among proteins located on the membrane of different cells [26]. Taken together, analyzing the interplay of proteins in close proximity and identifying their localization provide unique insights for understanding protein function and involvement in cellular responses.
2
Materials
2.1 General Materials and Equipment
1. Plunger of a 3 cc syringe. 2. Cell culture dish. 3. 50 mL conical falcons. 4. 15 mL conical falcons. 5. 1.5 mL microcentrifuge tubes. 6. 70 μm cell strainer. 7. Serological pipettes. 8. Neubauer counting chamber. 9. MACS LS columns. 10. 100 μm cell strainer for MACS LS column. 11. Magnetic separator for column-based systems. 12. Diagnostic 8-well 6 mm numbered microscopy slides. 13. Coverslips thickness 1, 24 × 60 mm. 14. Lid for coverage: cell culture dish lid or 96-well plate lid. 15. Humidity chamber or optional: wet chamber and incubator. 16. Disposable wipers. 17. Tweezers. 18. Ice and Fridge. 19. Centrifuge. 20. Fluorescent and/or bright field microscope (depending on fluorogenic or chromogenic labeling). 21. Material of interest: splenic monocytes isolated from C57BL/6 mice and FcγRIIb-/- mice (see Note 1). 22. Depending on direct or indirect PLA, primary and/or secondary antibodies (see Note 2). 23. Duolink® in situ PLA kits (commercially available) containing the distinct reagents. 24. Monocyte Isolation Kit for negative selection (commercially available) containing the distinct biotinylated antibodies and streptavidin nanobeads. 25. Optional: poly-D-lysine coated glass slides (see Note 3).
380
Sibel Kara and Falk Nimmerjahn
2.2 Buffers and Solutions
1. Trypan blue staining solution. 2. Phosphate-buffered saline (PBS). 3. 200 mM Tris–HCl (pH 7.5). 4. 2 mM Tris–HCl (pH 7.5). 5. Distilled water. 6. 70% Ethanol. 7. MACS buffer: 0.5% bovine serum albumin (BSA), 2 mM EDTA in 1× PBS, sterile filtered. 8. Fixing reagent: 4% paraformaldehyde (PFA) in PBS. 9. Mounting medium: Fluoromount-G with DAPI. 10. PLA analysis program (e.g., ImageJ, Cell Profiler, or BlobFinder).
2.3
For Direct PLA
2.3.1 Duolink® In Situ Probemaker PLUS or MINUS Kit (See Note 4)
Necessitated primary antibodies for detecting the desired proteinprotein interplay (for FcγRIIb: clone Ly17.2) and Duolink® in Situ Probemaker PLUS and MINUS as well as Duolink® In Situ Detection Reagents comprising. 1. Duolink® in situ oligonucleotide PLUS or MINUS: lyophilized activated distinct oligonucleotide for one conjugation of 20 μg antibody. 2. Conjugation buffer: buffer for conjugation reaction. 3. Stop reagent: stops conjugation reaction. 4. Storage solution: buffer for preserving prepared PLA probe. 5. 20× assay reagent: for optimizing antibody diluent. 6. PLA probe diluent: buffer for diluting PLA probe. 7. Blocking solution: blocks unspecific binding sites.
2.3.2 Duolink® In Situ Detection Reagents FarRed (See Note 5)
1. 5× ligation reagent: oligonucleotides that hybridize to PLA probes. 2. 1× ligase (one unit/μL) (see Note 6). 3. 5× amplification FarRed: components needed for rolling circle amplification (RCA) and also oligonucleotide probes labeled with fluorophore that hybridize to the RCA product. 4. 1× polymerase (ten units/μL) (see Note 6). The kits for the species independent assay contain the indicated reagents. Species-specific oligonucleotides and other fluorophores for microscopic analysis can also be ordered depending on the experimental purpose.
Fcγ-R interplay Visualized with PLA
3
381
Methods
3.1 Proximity Ligation Assay (PLA)
The proximity ligation assay can be performed either by direct PLA or by indirect PLA (see Fig. 1). Both techniques are based on an enhanced polymerase chain reaction that is visualized by a dye-based system, e.g., fluorogenic or chromogenic labeling. Basically, the system relies on a pair of oligonucleotides, also named as PLA probes, which contain a unique, PLUS, or MINUS DNA strand bound to a constant region. These specific DNA probes, serving as primers, are covalently linked to the specific antibodies to recognize the target proteins. If the target proteins are within a distance of 40 nm, the PLUS and MINUS oligonucleotides hybridize following addition of a ligase and form a closed DNA circle. This circle serves as a template for the included polymerase leading
Fig. 1 Schematic representation of direct and indirect PLA. (a) The direct PLA is only based on primary antibodies and does not necessitate secondary antibodies. This assay has to be performed if the primary antibodies are derived from the same species or any of the antibodies are raised from species other than mouse, goat, or rabbit. Primary antibodies are directly conjugated to non-species-specific MINUS and PLUS probes with oligomeric sequences bound to them. These labeled probe complexes are used to recognize the target proteins of interest. (b) The indirect PLA takes advantage of two primary antibodies raised from different species that can detect the distinct protein targets. Species-specific secondary antibodies labeled with PLUS or MINUS oligonucleotides are added to the samples to detect bound primary antibodies
382
Sibel Kara and Falk Nimmerjahn
to amplification of the DNA resulting in up to thousand concatemeric copies of the single strand. This amplification process is referred to as rolling circle amplification and the amplified sequence, i.e., the amplicon, remains attached to the probes. Signal localization is achieved via labeling with smaller, usually fluorescently conjugated, oligomers that ultimately become visible as individual spots. These spots, or PLA signals, can be studied by microscopy and analyzed qualitatively and quantitatively through specific programs, e.g., ImageJ [20, 21]. 3.2 Direct PLA of FcγRIIb on Monocytes 3.2.1 Conjugation of Primary Antibodies to PLA Probes PLUS or MINUS
1. The optimal concentration for primary antibodies is 1 μg/μL in PBS (see Note 7). 2. Add 20 μg equals 20 μL of each primary antibody against either target protein to a separate microcentrifuge tube. 3. Add 2 μL of conjugation buffer to the respective antibody, and mix by gently pipetting. 4. Transfer antibody solutions to the vial of lyophilized oligonucleotides PLUS and Probemaker MINUS, respectively, and mix gently by pipetting (see Note 8), and incubate at room temperature overnight (stirring is not required). 5. The next day, add 2 μL of stop reagent to the oligo-antibodyvials to stop the reaction, and incubate at room temperature for 30 min (stirring is not required). 6. Add 24 μL of storage solution and store the PLA probes at 4 °C (see Note 9).
3.2.2 Cell Preparation and Coating on Eight-Well Glass Slides
1. For determining the FcγRIIb, interplay splenic cells are isolated, either of mice expressing FcγRIIb (C57BL/6) or lacking FcγRIIb (FcγRIIb-deficient mice) as negative controls (see Note 1), and transferred into a cell culture dish containing PBS. 2. Remove the plunger from a sterile 3 cc syringe. Use the flat end of the plunger to mince the spleen by crushing the spleen five times in gentle circular motions. 3. Place a 70 μm cell strainer on a 50 mL Falcon, and pass 2 mL PBS through the strainer to prime it, and transfer the splenic cells from the culture dish to the strainer using a serological pipette. 4. Gently pass the cells through the strainer by pressing in a circular motion with the flat end of a fresh plunger, wash the strainer with 10 mL PBS, and then discard the strainer. 5. Top up the Falcon containing the cell suspension with PBS, and pellet the cells by centrifugation at 350 × g for 10 min at 4 °C.
Fcγ-R interplay Visualized with PLA
383
6. Discard the supernatant, resuspend the pellet in 10 mL MACS buffer (see Note 4), and determine the number of viable cells using a Neubauer counting chamber by excluding dead cells upon 1:1 staining with trypan blue. 7. Isolate monocytes with a commercially available Monocyte Isolation Kit for negative selection to obtain unlabeled and therefore untouched cells for analysis (see Note 1). First, adjust the cell concentration to 1 × 108 per mL of MACS buffer, and transfer 1 mL in a fresh 1.5 mL microcentrifuge tube. 8. Add 50 μL of the biotin-antibody mix from the Monocyte Isolation Kit to the cell suspension, mix carefully, and incubate cells for 15 min on ice. 9. Add 25 μL of vortexed (five touches, at max speed) streptavidin nanobeads to cell suspension, and incubate again for 15 min on ice. 10. Place the MACS LS column in a magnetic separator that fits the column and put a fresh 15 mL falcon to collect the isolated, non-biotinylated, and therefore non-magnetically labeled cells. 11. Rinse the column with 3 mL of MACS buffer, add the cell suspension to the column through a 100 μm filter, and collect the fraction containing the unlabeled cells, which are the cells of interest. 12. Wash the cells in the column two times with 3 mL of MACS buffer, and collect the fraction containing the untouched cells in the 15 mL falcon. 13. Centrifuge the cells at 350 × g for 10 min at 4 °C, discard the supernatant, and resuspend cells in 1 mL PBS. 14. Determine number of viable cells for the experiment using a Neubauer counting chamber by excluding dead cells upon 1:1 staining with trypan blue. 15. Transfer 1 × 105 monocytes in 30 μL PBS for coating on diagnostic eight-well microscopy slides, which are pre-coated with poly-D-lysine (see Notes 10–12). 16. Incubate microscopy slides for 2 h in a humidity chamber (37 °C, 5% CO2, 95% RH) to facilitate cell adhesion, while covered with a suitable lid (see Note 13). 3.2.3 Fixing and Blocking of Cells
1. Gently remove excess medium from the samples using a wipe paper sheet (never touch the cells) (see Note 14). 2. Add 30 μL 4% PFA to fix cells, and incubate slides for 15 min at room temperature (see Notes 3 and 15). 3. Gently remove PFA from the wells, wash slide twice for 5 min with PBS, and gently remove excess liquid with a wipe.
384
Sibel Kara and Falk Nimmerjahn
4. Add 30 μL Duolink blocking solution to cells for blocking of unspecific binding sites, place lid to the slide, and incubate for 60 min at 37 °C in a humidity chamber (see Notes 15 and 16). 3.2.4 Addition of Primary Antibodies Coupled to Duolink Probemakers
1. If necessary dilute PLUS or MINUS antibody with 20× Assay Reagent prior preparation of the experimental solution containing both antibody-coupled PLA probes, to avoid high signal-to-noise ratio (see Note 17). 2. Prepare the antibody-coupled PLA probes, 15 μL for each well: 3 μL PLUS antibody + 3 μL MINUS antibody + 9 μL PLA probe diluent (see Note 18). 3. Mix solution by pipetting and incubate for 20 min at room temperature. 4. Take slide from the incubator, and gently remove the blocking solution. 5. Wash cells twice for 5 min with PBS, and gently remove excess liquid. 6. Add 15 μL of the PLUS-MINUS-antibody mixture to the well, place lid to the slide, and incubate overnight at 4 °C (use fridge).
3.2.5 Ligation: Generating the Rolling Circle Template
1. Gently remove PLUS-MINUS-antibody mixture from the samples, wash cells twice for 5 min with PBS, and remove gently medium excess using a wipe. 2. While washing prepare ligation solution, 15 μL for each well: 11.62 μL distilled water +3 μL 5× Ligation reagent +0.38 μL ligase (see Notes 6, 18, and 19). 3. Mix solution by pipetting, add 15 μL of the ligation solution to cells, place lid to the slide, and incubate for 30 min at 37 °C in a humidity chamber.
3.2.6 Amplification with Incorporation of Fluorescent Oligonucleotides
1. Gently remove ligation solution from the samples using a wipe. 2. Wash cells twice for 5 min with PBS, and gently remove excess liquid. 3. While washing prepare amplification solution, 15 μL for each well: 11.81 μL distilled water +3 μL 5× amplification reagent FarRed +0.19 μL polymerase (see Notes 6, 18, and 19). 4. Mix solution by pipetting, add 15 μL of the amplification solution to cells, place lid to the slide, and incubate for 100 min at 37 °C in a humidity chamber. 5. Upon this step, keep cells covered with aluminum foil lid or in the dark, because the amplification solution contains the lightsensitive fluorophore.
Fcγ-R interplay Visualized with PLA
3.3 Mounting of Cells and Preparation for Imaging Preparation
385
1. Batch of coverslips are sterilized in 70% ethanol for at least for 30 min. 2. Before taking PLA slide from the humidity chamber, take out the coverslip with tweezers, and let it dry prior to mounting of the glass slide. 3. Gently remove amplification solution from the wells. 4. Wash cells twice for 10 min with 200 mM Tris–HCl and gently remove medium excess. 5. Wash cells once for 1 min with 2 mM Tris–HCl and gently remove medium excess with a wipe. 6. Let slide dry for 15 min at room temperature, covered with aluminum foil lid. 7. Drop 5–10 μL of Fluoromount-G containing DAPI for cell mounting to the wells. 8. Cover the slide with glass coverslip using tweezers. 9. Place the coverslip angled on top of the edge of the slide, and gently drop it to avoid trapping of air bubbles on the sample. 10. The mounting medium does not solidify! Therefore, the coverslip can be easily replaced. 11. Dry slide at least for 15 min at room temperature, covered with aluminum foil lid.
3.4
Image Analysis
1. The images can be obtained instantly by microscopy, or the slides can be stored in the dark at 4 °C until images are taken (see Note 20). 2. Take images with an appropriate microscope for fluorescent staining, and select suitable filters for acquisition. 3. Use same acquisition parameters between experimental and control samples. 4. When FcγRIIb-FcγRIIb pair is within a proximity of up to 40 nm, red spots will be visible (see Fig. 2). 5. The recorded fluorescent images can be analyzed by ImageJ, Cell Profiler, or BlobFinder.
4
Notes 1. Material of interest could be blood, tissue, adherent cells, or suspension cells. For purifying any cell population of interest for analysis, cells can be isolated by flow cytometry or distinct commercially available magnetic activated cell sorting kits, e.g., MojoSort™.
386
Sibel Kara and Falk Nimmerjahn
Fig. 2 Exemplary PLA images of splenic monocytes (a) lacking FcγRIIb serving as control and (b) expressing FcγRIIb forming PLA-clusters. PLA spots are indicated by red signals and nuclei (blue) are stained with DAPI. Scale bars 5 μm
2. It is critical to evaluate the suitability and selectivity of the antibodiesAntibodies for the recognitionRecognition of the distinct proteins to diminish signal artifacts by unspecific antibodyAntibodies binding resulting as false positive PLAProximity ligation assay (PLA) spots. Ideally a control from a knockout or knock-down sample or from a cell known to be negative for the protein of interest should be included. 3. This method is suitable for fixed cells, cytospin cells, paraffinembedded material, and cells coated on slides. 4. Duolink® In Situ Promemaker PLUS and MINUS reagents and MACS buffer need to be stored at 4 °C. 5. Duolink® In Situ Detection Reagents FarRed reagents need to be stored at -20 °C. 6. Enzymes have to be kept cold (-20 °C) at all times. Therefore, use a freezing block when removing from the freezer. 7. If the antibody concentration is higher than 1 μg/μL, dilute the antibody in PBS. Do not use solutions containing amines as diluents. The antibody buffer has to be free of carriers but may contain up to 0.1% BSA, 5% trehalose, and 0.02% sodium azide. Lower concentrations of antibodies can be used for conjugation as long as the final amount of antibody is 20 μg or does not exceed a volume of 20 mL. 8. Add the primary antibody immediately to lyophilized oligonucleotides after opening the vial. 9. PLA probes are stable up to 1 month at 4 °C. If the antibodies are not stable in storage solution, add necessitated reagents to stabilize them, respectively. 10. No minimum number of cells is required, but avoid excess to have no stacking of cells. The appropriate cell number depends on cell size and glass slide format and the area where cells are transferred on for performing the assay. If too many cells are seeded, it can happen that the resolution and accuracy of the
Fcγ-R interplay Visualized with PLA
387
spatial PLA signal spots cannot be separated and will be visualized as bright blurred signals due to overlay. 11. Diagnostic eight-well microscopy slides are a convenient format to perform PLA, because it enables to visualize cells in one focal plane. Additionally, due to the small area format of the glass slides, the assay is cost-effective, because only 10–30 μL of cell suspension or antibody mixtures per well are required. 12. Add 30 μL of poly-D-lysine (ready-to-use, 1 μg/μL) to eightwell slides and incubate for 1 h at room temperature. Remove solution, wash slide twice with distilled water, and let them dry. 13. To avoid contamination of the samples (e.g., with particles), place a lid (e.g., cell culture dish lid or 96-well-plate lid) on the slides. 14. Gently remove medium excess from the corner of the wells by taking the slide at a slight angle, and aspirate gently medium with a pipette. A low-power aspirator would work best, but pipette aspiration also works to remove the solution. 15. Washing buffers, fixation, and blocking solutions should be at room temperature before use. 16. Drop blocking solution into a fresh microcentrifuge tube, and transfer 30 μL to well. 17. Test your primary antibodies (IgG-class, monoclonal, or polyclonal) in immunohistochemistry (IC) or immunofluorescence (IF) assay to determine the optimal dilution, fixation, and blocking conditions for Proximity ligation assay (PLA). 18. The preparation of a master mix is recommended for multiple samples before applying to the wells. 19. The dilution ratio for the ligase is 1 to 40 and for polymerase 1 to 80. 20. The PLA signal is light sensitive. It is best to view the samples as soon as possible, but samples can be stored up to 4 days at 4 °C.
Acknowledgments This work was supported by grants from the German Research Foundation CRC1181-A07 and TRR305-B03 to F.N. References 1. Epelman S, Lavine KJ, Randolph GJ (2014) Origin and functions of tissue macrophages. Immunity 41(1):21–35 2. Saeki N, Imai Y (2020) Reprogramming of synovial macrophage metabolism by synovial
fibroblasts under inflammatory conditions. Cell Commun Signal 18(1):188 3. Greenhalgh AD, Zarruk JG, Healy LM et al (2018) Peripherally derived macrophages modulate microglial function to reduce
388
Sibel Kara and Falk Nimmerjahn
inflammation after CNS injury. PLoS Biol 16(10):e2005264 4. Herzog C, Pons Garcia L, Keatinge M et al (2019) Rapid clearance of cellular debris by microglia limits secondary neuronal cell death after brain injury in vivo. Development 146(9): dev174698 5. Bosurgi L, Cao YG, Cabeza-Cabrerizo M et al (2017) Macrophage function in tissue repair and remodeling requires IL-4 or IL-13 with apoptotic cells. Science 356(6342):1072–1076 6. Yang L, Zhang Y (2017) Tumor-associated macrophages: from basic research to clinical application. J Hematol Oncol 10(1):58 7. Yang M, McKay D, Pollard JW, Lewis CE (2018) Diverse functions of macrophages in different tumor microenvironments. Cancer Res 78(19):5492–5503 8. Mantovani A, Marchesi F, Malesci A, Laghi L, Allavena P (2017) Tumour-associated macrophages as treatment targets in oncology. Nat Rev Clin Oncol 14(7):399–416 9. Nimmerjahn F, Gordan S, Lux A (2015) FcγR dependent mechanisms of cytotoxic, agonistic, and neutralizing antibody activities. Trends Immunol 36(6):325–336 10. Kara S, Amon L, Lu¨hr JJ, Nimmerjahn F, Dudziak D, Lux A (2020) Impact of plasma membrane domains on IgG fc receptor function. Front Immunol 11:1320 11. Hogarth PM, Pietersz GA (2012) Fc receptortargeted therapies for the treatment of inflammation, cancer and beyond. Nat Rev Drug Discov 11(4):311–331 12. Boross P, Arandhara VL, Martin-Ramirez J et al (2011) The inhibiting fc receptor for IgG, FcγRIIB, is a modifier of autoimmune susceptibility. J Immunol 187(3):1304–1313 13. Floto RA, Clatworthy MR, Heilbronn KR et al (2005) Loss of function of a lupus-associated FcgammaRIIb polymorphism through exclusion from lipid rafts. Nat Med 11(10): 1056–1058 14. Nimmerjahn F, Ravetch JV (2010) Antibodymediated modulation of immune responses. Immunol Rev 236:265–275 15. Braun P, Gingras AC (2012) History of protein-protein interactions: from egg-white
to complex networks. Proteomics 12(10): 1478–1498 16. Lee C (2007) Coimmunoprecipitation assay. Methods Mol Biol 362:401–406 17. Gould KL, Ren L, Feoktistova AS, Jennings JL, Link AJ (2004) Tandem affinity purification and identification of protein complex components. Methods 33(3):239–244 18. Edidin M (2003) Fluorescence resonance energy transfer: techniques for measuring molecular conformation and molecular proximity. Curr Protoc Immunol: Chapter 18 19. Fredriksson S, Gullberg M, Jarvius J et al (2002) Protein detection using proximitydependent DNA ligation assays. Nat Biotechnol 20(5):473–477 20. Gomes I, Sierra S, Devi LA (2016) Detection of receptor heteromerization using in situ proximity ligation assay. Curr Protoc Pharmacol 75:2.16.1–2.16.31 21. Alam MS (2018) Proximity ligation assay (PLA). Curr Protoc Immunol 123(1):e58 22. Stadler C, Rexhepaj E, Singan VR et al (2013) Immunofluorescence and fluorescent-protein tagging show high correlation for protein localization in mammalian cells. Nat Methods 10(4):315–323 23. Sierra S, Luquin N, Rico AJ et al (2015) Detection of cannabinoid receptors CB1 and CB2 within basal ganglia output neurons in macaques: changes following experimental parkinsonism. Brain Struct Funct 220(5):2721–2738 24. Leuchowius KJ, Weibrecht I, Landegren U, Gedda L, So¨derberg O (2009) Flow cytometric in situ proximity ligation analyses of protein interactions and post-translational modification of the epidermal growth factor receptor family. Cytometry A 75(10):833–839 25. Tong QH, Tao T, Xie LQ, Lu HJ (2016) ELISA-PLA: a novel hybrid platform for the rapid, highly sensitive and specific quantification of proteins and post-translational modifications. Biosens Bioelectron 80:385–391 26. Sable R, Jambunathan N, Singh S, Pallerla S, Kousoulas KG, Jois S (2018) Proximity ligation assay to study protein-protein interactions of proteins on two different cells. BioTechniques 65(3):149–157
Chapter 27 Studying Efferocytosis Dynamics in Tissue-Resident Macrophages Ex Vivo Irene Aranda-Pardos, Achmet Imam-Chasan, and Noelia Alonso-Gonzalez Abstract In vitro cocultures of macrophages and apoptotic cells (ACs) provide a practical and useful tool to study efferocytosis. Here, we describe a method for automated quantification and imaging of recognition and engulfment of apoptotic cells by primary macrophages using imaging flow cytometry (IFC). IFC-based analysis allows us to successfully quantify efferocytosis, clearly distinguishing phagocytic from nonphagocytic macrophages and, more importantly, from those in recognition stage, which is not achievable by standard flow cytometrical analysis. To this end, we established a universally employable analysis pipeline to address efferocytosis that can be easily adapted to any macrophage population from samples of different origins. Key words Phagocytosis, Efferocytosis, Macrophages, ImageStream®X, Imaging flow cytometry, Apoptotic cells, Engulfment, Recognition
1
Introduction The phagocytosis of apoptotic cells, also known as efferocytosis, is a mechanism of cell clearance crucial for the resolution of inflammation [1]. In addition, the physiological clearance of circulating cells by tissue-resident macrophages takes place through diverse molecular pathways that correlate with tissue macrophageheterogeneity [2]. Flow cytometry is a well-established technique to measure efferocytosis in different cell types [3, 4] and phagocytosis of bacteria [5]. However, it does not render qualitative data to distinguish engulfment from recognition, which can be achieved with microscopy. Hence, studies of phagocytosis mostly combine microscopy, manual counting, and flow cytometry to obtain quantitative and qualitative data. Imaging flow cytometry combines high-throughput acquisition from flow cytometry with imaging capacities from microscopy and thus provides a useful tool to simultaneously gate and visualize fluorescently labeled cells [6]
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_27, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
389
390
Irene Aranda-Pardos et al.
during efferocytosis, allowing successful differentiation of recognition and engulfment of apoptotic cells. Data analysis is comparable between regular and imaging flow cytometry with the common starting steps consisting of gating desired populations based on their scatter signals and/or fluorescence intensities of stained marker epitopes. However, IFC data additionally contain microscopy images of every acquired channel and, thus, enable (pixelbased) in-depth analysis of cells by a plethora of parameters. These so-called “features” are grouped into different categories, such as size, shape, location, texture, and signal intensity. Furthermore, to utilize the full analytical capacities of IFC, features can be combined by Boolean logic. Although the huge number of combinations provides multiple analysis possibilities, by far not all features are necessary for every experiment. Analysis is driven by the readout that is to be obtained, and there may exist different routes (read features or analysis paths) leading to the same result. In this protocol, we provide a detailed guide for studying efferocytosis ex vivo with primary peritoneal macrophages and apoptotic thymocytes using imaging flow cytometry (Fig. 1). More importantly, the described method can be extended to other primary macrophages or cell types, such as alveolar macrophages and immortalized bone marrow derived macrophages, as we show later (see Note 21 for details).
Fig. 1 The graphical abstract of the described protocol and workflow. (Created with BioRender.com)
Analysis of Efferocytosis by Imaging Flow Cytometry
2
391
Materials
2.1 Buffers and Reagents
1. 70% ethanol, preferably in spray bottle 2. General buffer (GB): 2 mM ethylenediaminetetraacetic acid (EDTA), 0.5% Fetal Calf Serum (FCS) in phosphate-buffered saline (PBS) (kept at 4 °C). 3. 1× PBS. 4. Culture medium (CM): 10% FCS, 1% L-glutamine, 1% penicillin/streptomycin in Roswell Park Memorial Institute (RPMI) 1640. 5. Ammonium-chloride-potassium (ACK) lysis buffer: 150 mM NH4Cl, 10 mM KHCO3, 0.1 mM Na2 EDTA in dH2O, pH 7.2–7.4. 6. 0.1 μM dexamethasone in CM (stock: 10 mM in dimethylsulfoxide [DMSO]). 7. Dye for staining apoptotic cells. In this study, we use CellTrace™ Violet Dye (Invitrogen REF. C34571). 8. Macrophage antibodies: anti-mouse/human CD11b-fluorescein isothiocyanate (FITC) (clone M1/70) or anti-mouse CD11c-Alexa Fluor® 488 (clone N418). 9. 1% paraformaldehyde (PFA) in PBS. 10. Sheath fluid and buffers for flow cytometer operation as noted by the manufacturer or your site’s guidelines.
2.2
Consumables
1. 10 mL syringe 2. 20G needle 3. 15 and 50 mL conical tubes 4. 6-well plates with untreated surface 5. 1.5 mL tubes 6. 100 μm cell strainer 7. 1 mL syringe 8. 10 cm Petri dish 9. Compensation beads, if needed.
2.3
Equipment
1. CO2 chamber for rodent euthanasia. 2. Mouse dissection tools (forceps and scissors). 3. Pipettes. 4. Refrigerated centrifuge suitable for 15- and 50- mL conical tubes. 5. Cell counting system. 6. Humidified cell culture incubator with 5% CO2 at 37 °C.
392
Irene Aranda-Pardos et al.
7. Water bath, set at 37 °C. 8. Inverted microscope. 9. Imaging flow cytometer capable of acquiring high-resolution bright-field and fluorescent images and, if needed, staff for operation. For this study, we use an Amnis®ImageStream®X Mk II, equipped with two detection cameras (12 channels) and acquired samples with 60× magnification and low flow rates (see Note 1). 2.4 Analysis of IFC Data
3
Analysis needs to be performed with a software that can handle raw data containing images. In this study, we use the proprietary IDEAS® software. Although there is a third-party software available to open IFC data, the abovementioned in-depth analysis is currently only possible with IDEAS®.
Methods Perform centrifugations for 5 min at 350 g unless otherwise specified. Culture medium (CM) should be kept at 37 °C prior to use. General buffer (GB) for isolation and detachment is used ice-cold. Cells are cultured in a humidified incubator with 5% CO2 at 37 °C unless otherwise stated.
3.1 Isolation and Culture of Primary Peritoneal Macrophages
The isolation of peritoneal lavage from euthanized mice and culture of peritoneal macrophages for phagocytosis assay: 1. Euthanize mice in a CO2 chamber or with cervical dislocation, as normally performed in your facility. 2. Place the mice in supine position and apply 70% ethanol on the skin of the abdominal area. 3. Open the skin with a small incision, being careful not to damage the peritoneal wall, and separate the skin to properly visualize the peritoneal cavity. 4. Fill a 10 mL syringe with cold GB and inject solution into the peritoneal cavity with a 20 G needle carefully so that organs are not punctured (see Note 2). 5. Gently massage the peritoneum to dislodge any attached cells into the GB solution, and collect back into the syringe (take back as much volume as possible). 6. Transfer solution into a 50 mL conical tube previously filled with 5 mL of culture medium (CM). Place it on ice. 7. Repeat steps 4–6 for two to three times. 8. Centrifuge and discard supernatant.
Analysis of Efferocytosis by Imaging Flow Cytometry
393
9. Resuspend the pellet in 2 mL of CM and count cells (see Note 3). 10. Culture ~1 × 106 macrophages per well of a 6-well plate, with 3 mL of CM. 11. Incubate from 2–3 h to overnight allowing macrophages to attach to the surface. 12. Continue as soon as apoptotic cells are ready (see Subheadings 3.2, 3.3, and 3.4). Remove supernatant and wash away nonattached cells with 1× PBS. 13. Repeat step 12 and add 1.5 mL new CM. 3.2 Isolation of Thymus Tissue
Isolation of thymus from young (6–10 weeks old) mice: 1. Euthanize mouse in a CO2 chamber and place in a supine position, clipping the legs to the working mat. 2. Apply 70% ethanol on the skin of the abdominal area. 3. Make an incision from the bottom to the top and separate the tissue from the skin. 4. Open the peritoneal cavity and make an incision in the diaphragm, until the lungs are exposed. 5. Cut the ribs longitudinally in both the sides to lift them up, exposing the heart, and clip them to the mat if necessary. 6. Visualize the thymus, which is located behind the heart and looks like a white small lobule. 7. Harvest the thymus, cutting the connective tissue around it, and place it on a 1.5 mL tube containing 1× PBS. 8. Place it on ice.
3.3 Preparation of Apoptotic Thymocytes
1. Put thymus on a 100 μm cell strainer placed on a 50 mL tube. 2. Mash the tissue with the plunger of a 1 mL syringe adding PBS through the strainer. Flush the strainer with up to 20 mL of PBS. 3. Centrifuge cell suspension and discard supernatant. 4. Add 0.5 mL ACK lysis buffer for 2 min at room temperature to lyse residual erythrocytes. 5. Add 9 mL GB to stop lysis and centrifuge. Discard supernatant. 6. Plate cells in a 10 cm Petri dish with 15 mL CM (see Note 4). 7. Induce apoptosis by incubating the thymocytes with 0.1 μM dexamethasone in CM for 16 h at 37 °C and 5% CO2. 8. Collect cells and centrifuge. 9. Discard supernatant and proceed with staining.
394
Irene Aranda-Pardos et al.
3.4 Labeling of Apoptotic Cells (ACs)
Volumes showed here are used for one thymus. Scale volumes in case more are needed. 1. Resuspend pellet with 500 μL PBS containing 0.5 μM of CellTrace™ Violet Dye (see Note 5). 2. Incubate at 37 °C with 5% CO2 for 15–20 min. 3. Fill up to 5 mL with GM and centrifuge. 4. Discard supernatant and resuspend cells in 1 mL CM. 5. Count cells and adjust volume to a final concentration of 1 × 107 cells/mL.
3.5 Phagocytosis Assay
1. Coculture apoptotic cells with macrophages in a 5:1 ratio (ACs: macrophage); adding 500 μL of CM with 5 × 106 ACs. In this study, we coculture the cells for 45 min (see Note 6). 2. During coculture, keep the cells at 37 °C with 5% CO2 (see Note 7). 3. Wash the cells three times with PBS, lightly shaking the plate, to remove all the nonphagocytosed apoptotic cells, but being careful not to detach the macrophages.
3.6 Detachment of Macrophages
From now on, work in dark conditions as much as possible, to avoid loss of signal from fluorescently labeled ACs with CellTracker™. 1. Once apoptotic cells have been washed away from the culture, detach macrophages from the plate by carefully pipetting up and down with 1–1.5 mL of cold GB (see Note 8). 2. Transfer cell suspension into a 15-mL Falcon. 3. Repeat step 1 with new GB until no cells remain attached. 4. Centrifuge cells to proceed with staining with labeled antibodies.
3.7 Antibody Labeling of Macrophages for Imaging Flow Cytometry
1. Prepare antibody dilution in GB, following the manufacturer’s instructions. If several antibodies for different markers are used, prepare an Antibody (AB) mix with all of them. 2. Prepare single staining for every fluorophore used, using cells or compensation beads, at the same concentration as AB mix. Include AC stained with CellTracker™ as well. Single-staining samples are critical when performing imaging flow cytometry, so keep them under the same conditions as the sample. 3. Resuspend the samples with 50 μL of AB solution (see Note 9). 4. Incubate cells with ABs for 15–20 min at 4 °C in the dark. Transfer to a 1.5-mL tube, if possible. 5. Add up to 1.5 mL with GB to wash, and centrifuge.
Analysis of Efferocytosis by Imaging Flow Cytometry
395
6. Discard supernatant and resuspend samples with 30–70 μL of PBS with 1% PFA for fixation (see Note 10). 7. Keep at 4 °C in the dark until measurement. 3.8 Sample Acquisition
Follow your site’s guidelines for operation of the instrument. A brief description of features used for acquisition is provided in Table 1. 1. Perform start-up routine as required by your guidelines. 2. Switch off every channel that is not needed for the experiment to save on data size. With the protocol described here, at least the fluorescence channels Ch02 (FITC or Alexa Fluor® 488) and Ch07 (CellTrace™ Violet Dye) are needed. In addition, acquisition of bright-field channels (e. g., Ch01 and Ch09) is highly advised (see Note 11). 3. Load a fully stained phagocytosis sample (take the brightest stained, if known). Switch on lasers needed for excitation of used fluorophores and select the desired magnification. Samples should be run on low flow speed to obtain the highest possible sensitivity of the camera and quality of images. 4. If not present on the analysis area of the software (anymore), open a “scatterplot” with the parameters “Raw Max Pixel_MC_Ch02” and “Raw Max Pixel_MC_Ch07” (see Note 12). Measured Raw Max Pixel intensities should not approach 4095, since this is the saturation value of the charged-coupled device (CCD) camera, and having pixels that bright would be
Table 1 Masks and features needed for the acquisition of samples and a brief description of their purpose Name
Description
Masks
Mask defines the region of interest or area for a certain analysis on each image (more important in the analysis part).
MC and M01
MC is abbreviated for mask combined/common and includes every area of any channel with detected signal combined in a single ROI as largest combined area. M01 accordingly defines a ROI of detected signal in the first channel, in this study’s case, bright-field.
Features
Features are analysis parameters that are always linked to specific masks (and image channels if appropriate).
Raw Max Pixel_MC_XX
Maximum pixel intensity measured for the respective fluorescence channel XX in the largest combined ROI
Area_M01
Measured area in μm2 in mask for Ch01 or bright-field
Aspect Ratio_M01 Measured aspect ratio (short cell axis length/long cell axis length) in mask for Ch01 or bright-field; indicates how spherical or elongated events are on the bright-field image
396
Irene Aranda-Pardos et al.
Fig. 2 Plots with required parameters for the acquisition of phagocytosis samples. (a) Scatterplot of features to exclude oversaturated fluorescence intensity. (b) Scatterplot to gate cells and to exclude SpeedBeads® and debris due to size
detrimental for further analysis. Start adjusting laser outputs to set up and gate accordingly to exclude saturated pixels (R1) if reducing laser output is not enough (Fig. 2a) (see Note 13). Do not change laser settings from now on! 5. Return sample to proceed with the acquisition of single staining for later compensation. For this purpose, the use of the compensation wizard is advised, since it adjusts needed settings in the background. Follow instructions to acquire singlestaining samples one by one (see Note 14). 6. Proceed with the acquisition by loading a phagocytosis sample from a microtube or a plate. In addition, open a scatterplot with Area_M01 and Aspect Ratio_M01 and show events from the above gated (step 4) population. Gate events according to size (Area) to exclude the small SpeedBeads® that run with the machine constantly (Fig. 2b). By doing so, acquired file size is minimized. Acquire at least 20.000 cells, or more if desired, located in the Area gate (R2). Files are saved as raw image files (*.rif). 7. Perform shutdown routine as required by your guidelines. 3.9
Data Analysis
3.9.1 Generation of the Compensation Matrix
Herein described analysis route relies on the generation of custom masks and analysis features, and a brief description is provided in Table 2. 1. Once data are acquired, the first step of analysis is to generate the compensation matrix with the compensation wizard of the analysis software (IDEAS®). Follow instructions until the compensation matrix is calculated. Usually, the software selects
Analysis of Efferocytosis by Imaging Flow Cytometry
397
Table 2 Custom masks and features needed for the analysis of samples and a brief description of their purpose Name
Description
Masks
Analysis relies on the generation of custom masks. As with acquisition, (custom) analysis masks define the region of interest or area for a certain analysis (feature) on each image
System(M02, CD11b, 4)
Defines a broad mask dependent on CD11b staining; used to discriminate between single and aggregating macrophages
Morphology(M02, CD11b)
Defines a restrictive mask dependent on CD11b staining; used for further gating and analysis of distance to ACs
Morphology(M07, ACs)
Defines a restrictive mask dependent on CellTrace™ Violet Dye staining; used for further gating and analysis of distance to Macrophages
Features Area_System(M02, CD11b, 4)
Calculates area of the custom system mask defined by CD11b signal with a weight of 4
Aspect Ratio_System(M02, CD11b, 4)
Measures aspect ratio (short cell axis length/long cell axis length) in custom system mask defined by CD11b signal
Gradient RMS_Morphology(M02, CD11b) _CD11b
Gradient RMS (root mean square) indicates how focused cells are in the custom morphology mask defined by CD11b signal; value is calculated for the CD11b channel
Intensity_Morphology(M07, AC)_AC
Measured fluorescence intensity in the custom morphology mask defined by CellTrace™ Violet Dye signal; value is calculated for the CellTrace™ Violet -Dye channel
Intensity_Morphology(M02, CD11b) _CD11b
Measured fluorescence intensity in the custom morphology mask defined by CD11b signal; value is calculated for the CD11b channel
Modulation_Morphology(M07, AC)_AC
Measures the intensity range normalized between 0 and 1 in the custom morphology mask defined by CellTrace™ Violet Dye signal; value is calculated for the CellTrace™ Violet Dye channel
Spot Count_Morphology(M07, AC)_4
Applies spot counting (unconnected areas of the custom morphology mask defined by CellTrace™ Violet Dye signal) with a connectedness of 4
Area_Morphology(M07, AC)
Calculates area of the custom morphology mask defined by CellTrace™ Violet Dye signal (continued)
398
Irene Aranda-Pardos et al.
Table 2 (continued) Name
Description
Compactness_Morphology(M07, AC)_AC
Calculates the compactness of CellTrace™ Violet Dye staining in the custom morphology mask defined by CellTrace™ Violet Dye signal. Higher value translates into uniform signal without regional intensity differences
Delta Centroid XY_Morphology(M07, AC) _AC_Morphology(M02, CD11b)_CD11b
Measures the distance of calculated centroids of CellTrace™ Violet Dye and CD11b staining in their respective custom morphology masks
proper reference populations for compensation and calculates a usable matrix. However, the obtained values should be checked with evaluating some selected cells with the preview images button (see Note 15). Save calculated matrix as matrix file (*. ctm). 2. With the batch data files option, apply the newly saved compensation matrix to selected rif-files. Once finished, batching will have generated a compensated image file (*.cif) and a data analysis file (*.daf) for every sample that has been selected. Every future analysis and gating is stored in the daf-file with a cross reference to the cif-file, where solely the compensated images are located. 3.9.2 Custom Mask Generation
3.9.3 Generation of Analysis Features
1. Start analysis by opening a daf-file (see Note 16). 2. Generate a custom mask for CD11b staining (System(M02, CD11b, 4)) based on the premade mask M02. A comparison of the premade and custom mask for marking CD11b region is provided in Fig. 3a. Although the custom mask is extending the cells’ borders, it is still much better suited to gate aggregated macrophages. To allow an accurate analysis in the following steps, more restrictive custom masks need to be defined. Generate masks for CD11b fluorescence (Morphology (M02, CD11b)) and AC fluorescence ((Morphology(M07, ACs)) based on the premade masks M02 and M07. Figure 3b shows the comparison between premade and restrictive custom masks (see Note 17). Generate the following features applied on the custom masks from the Subheading 3.9.2 (see Note 18): 1. Area: Apply on masks System (M02, CD11b, 4) and Morphology (M07, AC). 2. Aspect ratio: Apply on mask System (M02, CD11b, 4).
Analysis of Efferocytosis by Imaging Flow Cytometry
399
Fig. 3 Comparison of the area coverage of custom and premade analysis masks. (a) Custom system and premade mask to determine area of interest defined by CD11b staining used for gating single macrophages. (b) Custom (restrictive) Morphology and premade masks to determine area of interests defined by CD11b staining or CellTrace™ Violet Dye staining used for further analysis
3. Gradient root mean square (RMS): Apply on mask Morphology (M02, CD11b)_CD11b. 4. Intensity: Apply on masks Morphology (M02, CD11b) for CD11b channel and Morphology (M07, AC) for AC channel. 5. Spot count: Apply on mask Morphology (M07, AC) with connectedness 4. 6. Compactness: Apply on mask Morphology (M07, AC) for AC channel. 7. Delta centroid XY: Apply simultaneously on masks Morphology (M07, AC) for AC channel and Morphology (M02, CD11b) for CD11b channel. 3.9.4
Gating Strategy
Keep in mind that in imaging flow cytometry gating is done on the level of the images and not on the cells contained in each image. For simplicity, we sometimes will, for example, write gate on macrophages and not gate on images containing macrophages. Start gating events of interest in the following manner (see Note 19): 1. Open a scatterplot with the parameters Area_System(M02, CD11b, 4) and Aspect Ratio_System(M02, CD11b, 4). Gate on single macrophages (Fig. 4a). 2. Open a histogram with the parameter Gradient RMS_Morphology(M02, CD11b)_CD11b and show only the previously
400
Irene Aranda-Pardos et al.
Fig. 4 Successive plots of the analysis route described in this protocol. (a) Scatterplot for gating single macrophages. (b) Histogram for gating focused macrophages. (c) Scatterplot for gating for images with macrophages and AC. (d) Histogram for gating macrophages containing one AC spot. (e) Scatterplot for gating single ACs. (f) Scatterplot for gating ACs in engulfed, attached to and outside macrophages populations and respective example events from the image gallery
gated single macrophages. Gate on focused cells by checking the fluorescence images for CD11b in the gallery above (Fig. 4b). 3. Open a scatterplot with the parameters Intensity_Morphology (M07, AC)_AC and Intensity_Morphology(M02, CD11b)
Analysis of Efferocytosis by Imaging Flow Cytometry
401
_CD11b and show only the previously gated focused cells. Gate on events with appropriate AC and/or macrophage staining (Fig. 4c). 4. Open a histogram with the parameter Spot Count_Morphology(M07, AC)_4 and show only the previously gated events containing AC. Gate on single AC spots (Fig. 4d) (see Note 20). 5. Open a scatterplot with the parameters Area_Morphology (M07, AC) and Compactness_Morphology(M07, AC)_AC and show only the previously gated single AC spots. Gate on single AC spot as defined by the area and compactness (Fig. 4e) (see Note 20). 6. Open a scatterplot with the parameters Delta Centroid XY_Morphology(M07, AC)_AC_Morphology(M02, CD11b) _CD11b and Compactness_Morphology(M07, AC)_AC and show only the previously gated single AC spot. Gate on AC either engulfed by or attached to macrophages or those located outside of macrophages by judging with help of the image gallery (Fig. 4f). 7. Export statistics and images as needed.
4
Notes 1. It is advised to use an ImageStream®X, since this is currently the only model that can acquire high-resolution images with a greater magnification. In special circumstances, a FlowSight® can also be taken into consideration (magnification 20×). Any other flow cytometer is not suitable, because they do not acquire fluorescent micrographs. 2. Avoid puncturing any tissue, especially intestines, as this would result in contamination of peritoneal lavage and might lead to preactivation of peritoneal macrophages by gut-borne bacteria. 3. From an unmanipulated mouse, we can obtain five to ten million cells from the peritoneal cavity, but only ~30% are macrophages. Remaining 50–60% are B cells and 5–10% are T cells. Only macrophages will attach to the culture plate after culturing; thus, the rest of cells can be removed by washing 2 times with 1× PBS later. Calculate accordingly before seeding cells on the 6-well plates (to obtain one million macrophages, seed ~3.3 million total cells). 4. Use one dish per thymus, if preparing more than one. 5. Different CellTracker™ dyes like carboxyfluorescein succinimidyl ester (CFSE) or CellTracker™ Red can be used to label the apoptotic cells as well, depending on the fluorophore selection
402
Irene Aranda-Pardos et al.
for the antibody panel. It is advised to double-check if planned fluorochrome combinations can be acquired by the imaging flow cytometer used. 6. If efferocytosis kinetics is addressed, several cocultures for the different times need to be prepared (e.g., 15 min, 30–45 min, 60 min, and 2 h). Start with the latest of the assay, going backward, to harvest all the samples at the same time in the end. 7. Remove cells from the incubator only to add apoptotic cells to the remaining time points in case kinetics is analyzed. 8. Detachment by pipetting up and down with detachment solution is sufficient for most macrophage populations, but not for more adherent cells like alveolar macrophages. Enzyme-free detachment reagents, such as Accutase®, can be used to harvest these populations. 9. If uncoupled primary antibodies are used, perform secondary antibody staining afterward, following the manufacturer’s instructions. 10. Resuspend samples and single staining in 30 μL if acquiring with the standard sample port for microtubes. If the machine is equipped with an autosampler, which is used, samples should be resuspended with 70 μL of PBS, 1% PFA, transferred to a 96-well round-bottom plate and sealed with a pierceable foil to prevent evaporation. In any case, fixation is crucial for a successful experiment, because samples’ state is retained and remains comparable over time. 11. It is highly recommended to include bright-field channel acquisition, although not needed per se for the described analysis route. It is of great help with setting up an acquisition gate and allows having another parameter available for gating purposes in case of encountered difficulties during analysis, for example, cell aggregation. 12. Raw Max Pixel_MC_XX (XX referring to channel number) scatterplots need to be opened for every fluorescent channel that is analyzed in the experiment, for example, when staining more surface markers on the macrophages. 13. According to the manufacturer, the best practice is to set laser outputs to obtain Raw Max Pixel_MC_XX values of 4000 maximum. Out of experience, values between 500 and 3000 yield good results. Occasionally some events stay oversaturated over a long range of laser power output. In this case, set lasers to obtain the main population between 500 and 3000 and gate accordingly to exclude saturated pixel events. 14. The wizard readjusts the following settings: Every channel is turned on again, bright-field and dark-field (if activated) are turned off, resulting in the acquisition of “pure” fluorescence
Analysis of Efferocytosis by Imaging Flow Cytometry
403
channels for each single staining. Samples can also be acquired manually by changing the before-mentioned settings, if the wizard is not used for any reason. 15. Clicking preview images button applies the calculated matrix on the selected cells. The pop-up window contains fluorescence channels for every cell selected. If, for example, FITCstained cells are previewed and the other channels show dark areas on the cells’ estimated location, the signal was overcompensated and the values in the matrix need to be reduced until obtaining satisfactory results. Vice versa, if areas on the cells’ estimated location are still too bright (compared to background areas), the signal was undercompensated and the values in the matrix need increment. 16. Clicking on a single daf-file will allow the analysis of this sample alone. However, if more samples have been processed, acquired, and need to be analyzed together, files need to be merged with the merge .cif files option. Selected cif-files will be copied in a single file with a corresponding analysis file. Cells in a merged file will still have their original sample/file name assigned, which allows simultaneous analysis, and also the generation of plot overlays with cells from different origins. 17. Premade analysis masks (regions of interest [ROIs]) are usually not accurate enough to distinguish single cells from aggregates and location of certain signals on the images. One of the first steps that need to be performed is to define custom masks. Customs masks are highly specific to the dataset and need to be generated new/adjusted for the data to fit. It is not recommended to use templates containing custom masks on new data without double-checking. 18. Listed features are not exhaustive but contain the minimum number needed for the described analysis path. Dependent on the experiment setup, for example, staining more surface markers, further appropriate features need to be generated. 19. In imaging flow cytometry, analysis is not necessarily performed on single cells. An acquired image can contain different types of cells. This needs to be considered while gating. Thus, sometimes it might be difficult to determine the borders of gates. However, every event acquired is tracked back in the software and has a running count-number assigned. Clicking on a dot in the scatterplot shows this event’s images in the gallery, which is useful to decide gating strategies. This is also valid for histograms, but here “displayed population” in the gallery needs to be set on “selected bin.” Doing so, every event with the same value on the histogram will be shown when clicking into the plot.
404
Irene Aranda-Pardos et al.
Fig. 5 Selected plots from analysis of experiments with different macrophage populations. (a) and (b) show the results of macrophages differentiated from immortalized monocytic progenitors. (c) Comparison of the outcome of the standard and modified analysis step to gate images containing ACs in samples with highly autofluorescent alveolar macrophages. (d) Images of representative alveolar macrophages that do not contain/contain an AC
20. Being defined as one spot does not mean that also only one apoptotic cell is present. Spot counting is highly dependent on the distance between two areas/spots/cells. If two cells are connected, meaning that the staining of each cell is not spatially separated, the previously applied mask for morphology will yield a single area resulting in the following spot count of one. If compactness and/or area is also considered, truly single apoptotic cells can be identified and gated. 21. This analysis route can be employed for experiments with different kinds of macrophages or apoptotic cell types, as shown in Fig. 5a and b (corresponding to plots from Subheading 3.9.4—steps 3 and 6, respectively) for macrophages differentiated from immortalized monocytic progenitors. These cells are not as efficient in efferocytosis compared with peritoneal macrophages, leading to a lower amount of phagocytic cells in general and also allowing detection and analysis of quantitatively more events in “attached” and “outside” gates.
Analysis of Efferocytosis by Imaging Flow Cytometry
405
More importantly, this method can be extended to the analysis of otherwise “problematic” macrophage populations with slight adjustments in the features used. Alveolar macrophages are known to be highly autofluorescent, which affects the sample in a way that positive events cannot be distinguished from negative when above-described gating strategies are applied, since gating according to Subheading 3.9.4—step 3 is not possible. This samples were stained for CD11c instead of CD11b, and a scatterplot with Modulation_Morphology (M07, AC)_AC and Intensity_Morphology(M02, CD11c) _CD11c was used for gating on events with appropriate AC and/or macrophage content (Fig. 5c) to clearly separate alveolar macrophages with and without AC despite their high degree of autofluorescence. Subsequent analysis’ steps to determine the location of AC compared with the macrophage remain unchanged. Figure 5d shows representative cells of the gated subpopulations. References 1. Doran AC, Yurdagul A, Tabas I (2020) Efferocytosis in health and disease. Nat Rev Immunol 20:254–267. https://doi.org/10.1038/ s41577-019-0240-6 2. A-Gonzalez N, Quintana JA, Garcı´a-Silva S, Mazariegos M, de la Aleja AG, Nicola´s-a´vila JA, Walter W, Adrover JM, Crainiciuc G, Kuchroo VK, Rothlin CV, Peinado H, Castrillo A, Ricote M, Hidalgo A (2017) Phagocytosis imprints heterogeneity in tissue-resident macrophages. J Exp Med 214:1281–1296. https:// doi.org/10.1084/jem.20161375 3. Parv K, Westerlund N, Merchant K, Komijani M, Lindsay RS, Christoffersson G (2021) Phagocytosis and efferocytosis by resident macrophages in the mouse pancreas. Front Endocrinol (Lausanne) 12:1–11. https://doi.org/10.3389/fendo.2021.606175
4. Larson SR, Atif SM, Gibbings SL, Thomas SM, Prabagar MG, Danhorn T, Leach SM, Henson PM, Jakubzick CV (2016) Ly6C+ monocyte efferocytosis and cross-presentation of cellassociated antigens. Cell Death Differ 23:997– 1003. https://doi.org/10.1038/cdd.2016.24 5. Rodrı´guez ME, Van Der Pol WL, Van De Winkel JGJ (2001) Flow cytometry-based phagocytosis assay for sensitive detection of opsonic activity of pneumococcal capsular polysaccharide antibodies in human sera. J Immunol Methods 252:33–44. https://doi.org/10.1016/S00221759(01)00329-5 6. Rees P, Summers HD, Filby A, Carpenter AE, Doan M (2022) Imaging flow cytometry. Nat Rev Methods Prim 2. https://doi.org/10. 1038/s43586-022-00167-x
Chapter 28 Monitoring of Inflammasome Activation of Macrophages and Microglia In Vitro, Part 1: Cell Preparation and Inflammasome Stimulation Marta Lovotti, Matthew S. J. Mangan, Ro´isı´n M. McManus, Kateryna Shkarina, Matilde B. Vasconcelos, and Eicke Latz Abstract Inflammasomes are intracellular, multiprotein supercomplexes that mediate a post-translational inflammatory response to both pathogen and endogenous danger signals. They consist of a sensor, the adapter ASC, and the protease caspase 1 and, following their activation, lead to cl1β, as well as lytic cell death. Due to this potent inflammatory capacity, understanding inflammasome biology is important in many pathological conditions. It is increasingly clear that inflammasomes are particularly relevant in macrophages, which express a diverse range of inflammasome sensors. In these two chapters, we detail methods to isolate and differentiate human macrophages, murine bone marrow-derived macrophages, and murine microglia and stimulate the inflammasomes known to be expressed in macrophages, including the AIM2, NLRP3, NLRC4, NLRP1, and non-canonical inflammasomes. Furthermore, we describe the methodology required to measure the various results of inflammasome activation including ASC speck formation, monitoring lytic cell death and cytokine release, as well as caspase-1 activation. Key words Inflammasome, ASC, NLRP3, AIM2, NLRC4, NLRP1, Caspase 1, Gasdermin D
1
Introduction Inflammasomes are intracellular multiprotein complexes that mediate a post-translational inflammatory response to pathogen- or danger-associated molecular patterns (PAMPs and DAMPs, respectively) [1]. They are expressed in both immune and non-immune cell types and have been shown to be particularly important in macrophage-driven inflammation. They consist of an inflammasome-forming sensor, the adapter ASC, and the protease caspase-1. On detection of a stimulus, the inflammasome-forming
Marta Lovotti, Matthew S. J. Mangan, Ro´isı´n M. McManus, Kateryna Shkarina and Matilde B. Vasconcelos contributed equally with all other contributors. Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_28, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
407
408
Marta Lovotti et al.
sensor oligomerizes, recruiting ASC, and enabling it to selfassociate and form ASC fibrils, which can then bind and activate pro-caspase-1. Caspase-1 cleaves and activates the pro-inflammatory cytokines IL-1β and IL-18 and enables their release by triggering lytic cell death through cleavage and activation gasdermin D (GSDMD) [1]. The inflammasome-driven inflammatory reaction is important for the immune response to infection by viral, bacterial, and fungal pathogens [2]. However, inflammasomes can also be aberrantly activated in disease states. Inflammasome sensors are unique amongst pattern recognition receptors as they detect alterations to cellular homeostasis as danger signals as well as directly detecting components of pathogens [3]. This includes the Pyrin inflammasome, which is activated by inhibition of RhoA as well as bile acid analogs, and is particularly true for the NLRP3inflammasome, which is activated by DAMPs that alter homeostasis of the cell, particularly those that trigger a drop in the intracellular potassium concentration [4]. As tissueresident macrophages play a key role in maintaining tissue homeostasis, they are potentially exposed to these DAMPs, and so are a highly relevant cell type to use to understand inflammasome activation during sterile damage or tissue injury. Indeed, the role of tissue-resident macrophages is highly relevant in pathological conditions. One example of this is microglia, where aberrant activation of the NLRP3inflammasome is an important facet of neurodegenerative diseases including Alzheimer’s disease and Parkinson’s disease [5, 6]. Further research understanding the role inflammasomes play in different cells and contexts could contribute knowledge in disease progression as well as providing new avenues for the treatment of these diseases. In this chapter, we detail methods for investigating the inflammasome response in a variety of macrophage models. We have included methods for the isolation of common macrophage models used to assess inflammasome activation, as well as methods for the activation of inflammasome sensors known to be expressed in macrophage populations.
2
Materials
2.1 Common Consumables and Equipment
1. Nunc Delta-surface 6-well polystyrene plates. 2. 50-mL centrifuge tubes. 3. 1.5-mL microcentrifuge tubes. 4. Serological pipettes (5 mL and 25 mL). 5. Scissors. 6. Cell scrapers. 7. Disposable waste bottle. 8. 10-cm tissue culture dishes, tissue culture-treated.
Activating and Monitoring Inflammasome Activation in Macrophages
409
9. Clear 96-well plates, sterile for cell culture. 10. Tissue paper. 11. Parafilm. 12. T75 flasks. 13. Centrifuge. 14. Cell culture incubator. 2.2 Common Buffers, Solutions, and Reagents
1. Milli-Q water. 2. 70% Ethanol. 3. Surface disinfectant. 4. Poly-L-lysine (PLL). 5. 0.5 M EDTA. 6. Bovine serum albumin. 7. Ultrapure lipopolysaccharide from Escherichia coli. 8. Caspase-1 inhibitor‘ VX-765. 9. 1× Dulbecco’s phosphate-buffered saline (DPBS). 10. Complete DMEM: 10% FCS, 1% penicillin/streptomycin. 11. Complete RPMI medium: 10% FCS, 1% penicillin/streptomycin, 1% GlutaMAX. 12. Serum-free cell culture medium: OptiMEM.
2.3 Consumables for the Isolation of Different Myeloid Cell Types
1. Absorbent bench cover.
2.3.1 Human PBMC Isolation and Differentiation
5. MS or LS columns.
2. Blood waste disposal. 3. 500 mL 0.22 μM filter unit. 4. Pre-separation filters. 6. MS or LS magnet. 7. MACS multistand.
2.3.2 Preparation of Murine Bone MarrowDerived Macrophages (BMDMs)
1. Surgical tweezers and scissors. 2. 6-well plates, tissue culture-treated. 3. 25 G needle. 4. 10 mL syringe. 5. 70 μm cell strainer.
2.3.3 Preparation of Murine Microglia
1. Forceps. 2. Curved and straight long-tail tweezers (extra fine). 3. 3 cm Nunc™ Cell Culture/Petri Dishes. 4. Polystyrene foam plate (styrofoam) with 4 syringe tips 27G.
410
Marta Lovotti et al.
2.4 Buffers, Solutions, and Reagents for the Isolation of Different Myeloid Cell Types
1. Buffy coat.
2.4.1 Human PBMC Isolation and Differentiation
5. Detaching buffer: 5 mM EDTA, 1% FCS in DPBS.
2. Ficoll-PAQUE Plus: 5.7 g Ficoll PM400, 9.0 g diatrizoate sodium with edetate calcium in purified water. 3. Red blood cell lysis buffer (optional). 4. MACS buffer: 5 mM EDTA, 1% BSA in DPBS. 6. M-CSF. 7. GM-CSF. 8. CD14+ magnetic beads. 9. Complete RPMI medium: 10% FCS, 1% penicillin/streptomycin, 1% GlutaMAX, 1% sodium pyruvate.
2.4.2 Preparation of Murine Bone MarrowDerived Macrophages (BMDMs)
1. L929 supernatant.
2.4.3 Preparation of Murine Microglia
1. HBSS buffer. 2. Serum-free cell culture medium: DMEM, 1% P/S plus 1× N2 supplement. 3. Trypsin (no EDTA). 4. DNase.
2.5 Consumables for Activation of Inflammasomes
1. Microcentrifuge tubes.
2.5.1 Non-Canonical Inflammasome Activation Using LPS Transfection 2.5.2 Activation of NLRC4 and Non-Canonical Inflammasome Using Salmonella Infection
1. BSL2 designated area and equipment. 2. Individual protective equipment suitable for working with the BSL2 organisms. 3. Round-bottom bacterial culture tubes. 4. Inoculation loops. 5. Filter tips and pipettes. 6. Spectrophotometer. 7. Spectrophotometer cuvettes.
Activating and Monitoring Inflammasome Activation in Macrophages
2.6 Buffers, Solutions, and Reagents for Activation of Inflammasomes
411
1. Lipofectamine 2000. 2. Poly-(dA:dT). 3. m-3M3FBS. 4. Thapsigargin.
2.6.1 AIM2Inflammasome 2.6.2 NLRC4Inflammasome
1. Protective antigen (PA). 2. Lethal factor (Lfn)-Needle proteins (PrgI, MxiH, or BsaK). 3. Lfn-Rod proteins (PrgJ, MxiI, or BsaK).
2.6.3
NLRP1 and CARD8
1. 2 mM Val-boroPro in DMSO. 2. Anthrax protective antigen. 3. Anthrax lethal factor.
2.6.4 NLRP3Inflammasome
1. CRID3.
2.6.5 Pyrin Inflammasome
1. Clostridioides difficile Toxin A (TcdA).
2. Nigericin.
2. Pyrin inhibitor (Colchicine).
2.6.6 Non-Canonical Inflammasome Activation Using LPS Transfection
1. Murine or human Interferon gamma.
2.6.7 Activation of NLRC4 and Non-Canonical Inflammasome Using Salmonella Infection
1. Strain of choice of Salmonella enterica serovar typhimurium, stored as the 50% glycerol stock at -80 °C. For the non-canonical inflammasome activation, we recommend using the mutant strains deficient in the NLRC4 activators, such as ΔorgA/fliC/fljAB.
2. Smooth LPS from Escherichia coli or other Gram-negative bacteria.
2. LB agar supplemented with the appropriate antibiotic (usually streptomycin). 3. LB medium (high salt).
3
Methods
3.1 Protocols for the Isolation of Different Myeloid Cell Types 3.1.1 Human PBMC Isolation and Differentiation
This protocol details a method for the isolation of PBMCs from either leukocyte concentrate (buffy coat or leuko packs) or from whole blood, and the enrichment of CD14+monocytes and their differentiation into macrophages. Leukocyte concentrate is the anti-coagulated fraction of a whole blood sample, obtained by an initial centrifugation step that concentrates red blood cells at the bottom and plasma on top. To obtain macrophages, these
412
Marta Lovotti et al.
monocytes are differentiated in culture for 3 days using M-CSF or GM-CSF. The following protocol is for 60 mL of leukocyte concentrate but can be used directly for whole blood as well. Following differentiation to macrophages, these cells express the AIM2, NLRP3, NLRC4, NLRP1, and Pyrin inflammasomes and enact an inflammasome response to the activators of these inflammasomes. Preparation of the MACS Buffer
1. Dissolve the BSA in 500 mL endotoxin-free PBS (sterile PBS) to a concentration of 0.5%. 2. Add 2 mL of 0.5 M EDTA to the PBS/BSA mix. 3. Filter-sterilize the solution using the filter sterilization module. 4. Store at 4 °C until needed.
Isolation of PBMCs Using Density Gradient
1. Gently resuspend buffy coat by inversion (5–6 times) and wipe off the buffy coat bags with 70% ethanol (see Notes 1–3). 2. Open the buffy coat with a scissors at one of the two hoses and carefully pour 20 mL of the blood into 3 separate 50 mL centrifuge tubes (see Note 4). 3. For whole blood, open the collection tubes and remove blood to 50 mL centrifuge tubes using the serological pipette. 4. Dilute blood 2:3 with sterile PBS and mix gently with a 25 mL pipette (see Note 5). 5. Fill 15 mL of Ficoll into separate 50 mL centrifuge tubes. 6. Carefully layer 35 mL of PBS–blood mixture on top of the Ficoll for the density gradient (see Notes 6 and 7). 7. Centrifuge at room temperature (RT), 700 × g, for 20 min without brake (see Note 8). 8. The interface between Ficoll and supernatant is the PBMC layer (white, Fig. 1). Remove as much plasma as possible and transfer only this white layer into a clean 50 ml centrifuge tube by using a 10 mL pipette. Pre-Spin
Post-Spin
Diluted Blood Ficoll
Platelet Plasma PBMC Ficoll+ Gran RBC
Fig. 1 Layers following separation of diluted blood by Ficoll plaque gradient
Activating and Monitoring Inflammasome Activation in Macrophages
413
9. Adjust the volume with PBS to 50 mL per centrifuge tube and resuspend gently. 10. Centrifuge at 340 × g for 10 min at RT. 11. Discard supernatant. 12. (Optional: red blood cells lysis. Add 2 mL of red blood cell lysis buffer on top of the pellet and resuspend. Incubate for 5 min at RT). 13. Repeat wash with PBS, fill the centrifuge tube to 50 mL, and resuspend gently. 14. Centrifuge at 340 × g for 10 min at RT. 15. Discard supernatant. Monocyte Isolation Using CD14 Beads
1. Resuspend cell pellet in 1 mL MACS buffer +175 μL MACS beads (see Note 9). 2. Incubate for 15 min at 4 °C. 3. Centrifuge at 340 × g for 10 min at 4 °C. 4. During centrifugation, place the LS column on the MACS separator. Put the pre-separation filter on the LS column. Rinse each column with 3 mL cold blocking buffer (see Notes 10 and 11). 5. Resuspend the cells in 1 mL MACS buffer and add the cells to the column. 6. Wash columns 3 times with 3 mL MACS buffer. 7. Remove the column from the separator and add 5 mL MACS buffer. 8. Collect the monocytes into a 50 mL tube by adding 5 mL of MACS buffer to the column and pushing them out using the supplied plunger. 9. Add 40 mL of PBS to the monocytes in the 50 mL tube and collect the cells by centrifugation 360 × g for 5 min. 10. Resuspend pellets in complete RPMI and determine the cell count (see Notes 12 and 13).
Differentiation of Monocytes into M-CSF or GM-CSF Macrophages
1. Using a 6-well Delta-surface plate, seed 2 million/mL monocytes in 5 mL complete RPMI medium, with the addition of 50 U/mL M-CSF or 500 U/mL GM-CSF to induce the differentiation into macrophages for 72 h (see Notes 14 and 15).
Harvesting of M-CSF or GM-CSF Macrophages
1. Transfer supernatants from each well into a 50 mL centrifuge tube, to collect cells that have not adhered. 2. Add 1 mL/well DPBS to wash the cells, then remove, and transfer to the same 50 mL centrifuge tube.
414
Marta Lovotti et al.
3. Add 1 mL of detaching buffer to each well and incubated for 5 min at 37 °C. 4. Scrape cells and transfer them to the corresponding 50 mL centrifuge tube. 5. Spin cells down at 340 × g for 5 min. 6. Discard supernatants and resuspend cells in 10 mL complete RPMI medium (depending on cell density). 7. Determine cell counts. 8. Add 12.5 U/mL M-CSF or 125 U/mL GM-CSF into the diluted cells to seed 80,000 cells/well in a 96-well plate. 3.1.2 Preparation of Murine Bone MarrowDerived Macrophages (BMDMs)
1. Inflammasomes that are commonly detected in BMDMs include the following: NLRP3, AIM2, NLRC4.
Bone Marrow Flushing and Cultivation
1. Kill the mouse and wet the fur with 70% ethanol. 2. Make a small incision with disinfected scissors in the area of the peritoneum. Peel the skin over the hind legs toward the paws. 3. With the scissors, cut around the head of the femur, to separate the hind legs from the hip joints. Also, cut off the paws around the ankle joints. Always try to keep the bones intact. 4. Cut the tendons and muscles. Separate tibia and femur by dislocating the knee joint. Carefully rub away the remaining tendons and muscles with tissues. This results in four cleaned intact bones, two femurs and two tibias. 5. Place the bones into 6-well plate wells, each filled with 2 mL of cold PBS. The following steps have to be performed under sterile conditions. 6. In order to disinfect the bones, shortly transfer them to new 6-well plate wells filled with 1 mL of 70% ethanol and then back to the wells containing PBS. 7. Cut the two ends of the bones. 8. Push a 25 G needle into one of the ends of the bone and flush the cavity with 5 to 10 mL of serum-free medium into a 50 mL centrifuge tube. Continue the flushing (also from the other bone end, if needed) until the bone looks white-transparent. Collect the cells from all the four bones in one tube. 9. To filter the obtained cell solution, place a 70 μm cell strainer on a new 50 mL centrifuge tube. Resuspend the cell solution and transfer it through the strainer into this new centrifuge tube.
Activating and Monitoring Inflammasome Activation in Macrophages
415
10. Centrifuge for 5 min at 340 × g, at RT. 11. Discard the supernatant and resuspend the cell pellet in 16 mL of pre-warmed complete cell culture medium supplemented with 20% L929-conditioned medium. 12. Add 8 mL of the same supplemented medium to each of eight 10 cm tissue culture petri dishes (2 dishes/leg). Then, add 2 mL of the cell suspension on top. 13. Incubate at 37 °C, 5% CO2 for 6 days. Harvesting and Plating the BMDMs
1. After 6 days, remove the cell culture supernatants with a stripette, and wash the remaining cell layer of adherent BMDMs with 10 mL of sterile-cold PBS. 2. Collect the PBS in a 50 mL centrifuge tube, and add 10 mL of cold detaching buffer, i.e., PBS supplemented with 2% FBS and 5 mM EDTA. Scrape the detaching cell layer on the dish with the help of a cell scraper, and tap the sides of the dish. With a stripette, collect the cells in the same 50 mL centrifuge tube. 3. Centrifuge the cells for 5 min, at 340 × g, at RT. 4. Resuspend the cell pellet in 5 mL pre-warmed complete cell culture medium supplemented with 2.5% L929-conditioned medium. 5. Count the cells, and adjust the cell concentration to 1 × 106 cells/mL. 6. For the inflammasome activation experiments, seed BMDMs in flat-bottom, tissue culture-treated 96-well plates at a density of 1 × 105 cells/well (i.e., 100 μL/well). 7. Incubate at 37 °C, 5% CO2 overnight (16–18 h) before further treatment.
3.1.3 Preparation of Murine Microglia Coating of the T75 Flasks
Inflammasomes that are commonly detected in microglia include the following: NLRP3, AIM2, NLRC4 (see Note 16). 1. Add 10 mL PLL (0.01%) to the T75 flasks (coat enough flasks to allow for 2 brains per flask). 2. Incubate for 30 min at 37 °C, 5% CO2 in the incubator. 3. Discard PLL and wash the T75 flasks three times with 10 mL 1× PBS. 4. Add 10 mL complete DMEM to each flask and store at 37 °C until addition of the murine cells.
Brain Dissection
1. Decapitate the mouse pup by gently holding the lower half of the mouse and using scissors, cut off the head with one swift action.
416
Marta Lovotti et al.
2. Store the heads in a 50 mL centrifuge tube on ice while re-locating to the dissection zone. 3. The dissection zone includes the polystyrene foam plate, which has been layered with tissue paper and parafilm that are secured in place with syringe tips. Wipe the parafilm down with ethanol prior to use. 4. Fix the mouse head to the parafilm-styrofoam by piercing the syringe tip through the nose. 5. At the back of the skull is the foramen magnum, which is a small circular opening that connects the brain with the spinal cord. Gently insert the tip of a small scissors horizontally into the foramen magnum and carefully cut to the left of the skull until reaching the eye area. Take care to cut deeply enough that you are cutting through the skull, but not so deeply that you are piercing the underlying brain tissue. Repeat for the right side and the top of the skull, along the midline. 6. Carefully lift up the skin and skull to the left and right to free the brain. 7. Using curved forceps, gently scoop the brain out of the skull. Collect 2 brains per 3 cm dishes that are filled with HBSS. Place the dishes on ice. 8. With the aid of the microscope, remove the meninges from the brain tissue using the extra fine long-tailed tweezers. Take note of the blood vessels and visible folding of the meninges to guide the removal. A brain with meninges looks red-pinkish, whereas after removal of the meninges the brain appears off-white to gray in color. A tip is to try and take hold of the meninges at one end of the brain (say near the olfactory bulb) and remove a strip of the meninges to the back of the brain, as close to the cerebellum as possible. Keep your dissection kit as sharp as possible; meningeal removal is very difficult with blunt instruments. 9. Collect the brain pieces into a 50 mL centrifuge with HBSS and keep on ice. 10. Work as quickly as possible to remove the meninges, and reduce the time that the brain is under the heat of the microscope. Tissue Dissociation and Mixed Glia Culture
1. Let the brain tissue settle to the bottom of the centrifuge tube. Discard the HBSS buffer. 2. Wash cells 3 times with 5–10 mL HBSS. 3. Add 1 mL 1× Trypsin per 12 brains in HBSS buffer and vortex. 4. Incubate for 10 min at 37 °C in the water bath, after 5 min vortex the samples.
Activating and Monitoring Inflammasome Activation in Macrophages
417
5. Stop the Trypsin reaction by adding 9 mL complete DMEM. 6. Add 400 μL DNase to the centrifuge tube. The final concentration is 2 mg in 10 mL when there are 12 murine brains. Adjust accordingly with more or less tissue. 7. Centrifuge at 300 × g for 5 min at 4 °C. 8. Discard the supernatant, taking care not to disturb the cell pellet at the bottom of the tube. Disrupt the cell pellet to separate cells by pipetting up and down using a 10-mL pipette in complete DMEM, and fill 5 mL cell suspension in each T75 flask (2 brains/ flask). This brings the total volume to 15 mL/ flask. 9. Let the cells adhere for 24 h in the incubator at 37 °C, 5% CO2. 10. Discard medium and wash the cells 3× with pre-warmed 1X PBS. 11. Add 13.5 mL complete DMEM and 1.5 mL L929 conditioned medium, a 1:10 dilution of L929 conditioned supernatant. Ensure to note the preparation date and addition of L929 on the flask. 12. The microglia will be ready to shake off after approx. 7 days. Microglia Isolation
1. After approx. 1 week, the maturing microglia can be identified under the microscope as being bright, rounded cells sitting on top of the astrocyte cell layer (Fig. 2). When ready, the microglial cells will start “popping off” this astrocyte layer and the goal is to harvest the microglia before too many have detached into the media. 2. To isolate the microglia, the cells are shaken off by hitting the flasks approx. five times against the palm of your hand. This will not detach the astrocyte layer. 3. It is important to check under the microscope during this process to determine whether there are many microglial cells still attached to the layer or if the astrocyte layer already started to detach from the bottom of the T75 flask. The goal is to leave some microglia, as these cells will proliferate over the next days and can be harvested again. If astrocytes are detaching (which can happen in older cultures), this will not be a pure microglial population. 4. Remove the media containing the detached microglia and centrifuge for 5 min at 300 × g at 4 °C. 5. Quickly add 7 mL of fresh complete DMEM into the T75 flasks so the cells do not dry out. Add 7 mL of the old supernatant to the T75 after centrifugation. Add 750 μL of L929 conditioned medium to the T75 flask. Ensure to note on the flask the date and shake number.
418
Marta Lovotti et al.
Fig. 2 Microglia viewed under the microscope
6. Remove any remaining supernatant from the centrifuge tube and resuspend cells in 1–3 mL of complete DMEM and count the cells. 7. Plate the cells at a pre-determined cell density, depending on the needs of your assay. 8. Leave overnight. 9. Change to serum-free DMEM the next day. 10. Start cell treatments the following day, namely the second day after shaking the microglia in serum-free DMEM. 11. Microglia can be shaken off 3 times and will be ready to harvest again in 2–3 days. 3.2 Activation of Inflammasomes
1. Harvest the cells and seed 100 μL/well in a clear 96-well plate as described above (see Note 17).
3.2.1 AIM2 Inflammasome
2. The following day, to prime the cells and increase the protein levels of pro-IL-1β and IL-18, add 100 μL of complete cell
Activating and Monitoring Inflammasome Activation in Macrophages
419
Fig. 3 Secreted IL-1β in bone marrow-derived macrophages (BMDMs) of wildtype (WT) or AIM2-/- mice, treated with the NLRP3 activator Nigericin and the AIM2 activators poly-(dA:dT) and m-3M3FBS
culture medium with LPS at a final concentration of 200 ng/mL for murine bone marrow-derived macrophages and 50 ng/mL for human macrophages. Make sure to include control wells that do not receive LPS. 3. Incubate the cells for 3 h at 37 °C, 5% CO2. 4. 20 min before the end of the incubation, prepare 10× model dsDNA transfection complexes, consisting of model dsDNA (poly-(dA:dT)) (see Note 18 and Fig. 3) and lipofectamine 2000. Per well, mix in one tube 0.5 μL of lipofectamine 2000 with 5 μL OptiMEM, and in a separate tube 200 ng of poly(dA:dT) (i.e., 0.2 μL of a 1 mg/mL stock) with 5 μL
420
Marta Lovotti et al.
OptiMEM. Incubate for 5 min at RT. Mix the contents of both tubes and incubate for further 10 to 15 min (to allow for the formation of transfection complexes). Since these are prepared as 10× concentrated working solutions, these amounts actually correspond to a final concentration of 2 μg/mL poly-(dA:dT) dsDNA complexed with 5 μL of Lipofectamine 2000. 5. Alternatively: 5 min before the end of the incubation, prepare 10× m-3M3FBS or thapsigargin, i.e., an 850 μM concentrated m-3M3FBS or a 200 μM concentrated thapsigargin working solution (see Note 19). 6. Harvest the supernatant into a new 96-well plate, if interested in analyzing the priming step (e.g., check for TNFα in these supernatants by ELISA/HTRF). 7. Add 90 μL/well of serum-free OptiMEM. 8. Add 10 μL/well of the 10× inflammasome activators directly on top of the relevant wells. 9. Centrifuge the cells for 5 min, at 340 × g, at RT. 10. Incubate the cells for 90 min at 37 °C, 5% CO2. 11. Harvest the supernatant into a new 96-well plate. Use directly or store at -20 °C until ready for further processing. 3.2.2 NLRC4 Inflammasome
1. Harvest the murine or human macrophages and resuspend them at 0.8 × 106 cells/mL in complete RPMI, then plate 100 μL in each a 96-well plate (for ELISA and LDH readouts) or 2500 μL in each well of a 6-well plate (for immunoblot). Sufficient wells of cells should be plated to include an untreated control, as well as wells stimulated with LPS alone. For an example, please see Fig. 4.
Fig. 4 Sample plate layout of treatment conditions to trigger NLRP3inflammasome activation
Activating and Monitoring Inflammasome Activation in Macrophages
421
2. For the assay, the media used depends on the readout desired. For experiments where LDH or immunoblot of the supernatant will be used as a readout, the cells should be stimulated in optiMEM, otherwise use complete RPMI. 3. If IL-1β release is to be used as a readout for inflammasome activation, the macrophages need to be primed with LPS to induce IL-1β expression. In this case, remove the media from the cells and add media containing 100 ng/mL LPS (for BMDM) or 10 ng/mL (for human macrophages). 50 μL per well for a 96-well plate or 1000 μL of media for a 6-well plate is sufficient. 4. Incubate the cells for 3 h before addition of NLRC4 activators. 5. To determine whether the effects observed following NLRC4 stimulation are specific to the inflammasome the macrophages can be pre-incubated with a caspase-1 inhibitor, which will prevent inflammasome-dependent cell death and IL-1β release. 6. To stimulate the NLRC4inflammasome, you can either remove the media from the cells and add the activator at a 1× concentration (100 μL for 96-well plate or 1000 μL for a 6-well plate) or add media on top of the media already present at a 2× concentration (for example add 50 μL of 2× concentrated activator on top of 50 μL of media). 7. Make up the activator mix including the relevant Lfn-needle or Lfn-rod protein with PA at a 1:2 ratio in either optiMEM or complete RPMI (see Notes 20–23). 8. Add the activator/protective antigen mix to the macrophages. 9. Incubate the cells for 1 h at 37 °C. (a) The cells should undergo a very obvious change in morphology as they undergo pyroptosis. 10. For supernatant harvesting from a 96-well plate: Centrifuge the plate at 500 × g for 5 min, then remove the top 50 μL of supernatant. 11. For supernatant and cell harvesting from a 6-well plate: Remove the supernatant to a 1.5 mL microcentrifuge tube and centrifuge at 500 × g for 5 min. Remove the top 950 μL supernatant from the tube to use to precipitate for immunoblotting. The cells can be lysed in 100 μL RIPA buffer. 3.2.3
NLRP1 and CARD8
1. Harvest the cells as described above and seed them into the clear 96-well plates at the concentration of 50,000 cells per well. Incubate overnight at 37 °C, 5% CO2. 2. On the day of experiment: prime the cells with 100 ng/mL LPS and incubate for approximately 4 h to induce IL-1β upregulation. Include the non-stimulated wells as the control.
422
Marta Lovotti et al.
3. Remove the supernatants and wash the cells 1× with the pre-warmed PBS to remove any remaining LPS. 4. Remove the medium and add the assay medium containing 2 μM Val-boroPro. As a control, include the wells containing both Val-boroPro and 20 μM VX-765, which would block the caspase-1 activation downstream of NLRP1/NLRP1b and CARD8 inflammasomes (see Notes 24 and 25). 5. Incubate the plates for 24 h at 37 °C, 5% CO2. 6. For activation with anthrax lethal toxin, remove the media and add a 1:1 mix of anthrax lethal toxin and protective antigen. As a control, include the wells containing both the lethal factor/protective antigen mix and a 20 μM VX-765, which would block the caspase-1 activation downstream of the NLRP1b inflammasome. 7. Harvest the cells and the supernatants separately for further analysis. 3.2.4 NLRP3 Inflammasome
1. Harvest the cells as described above and adjust the final cell concentration to 1 × 106/mL in complete DMEM. 2. Seed 100 μL cells into a clear 96-well plate and incubate overnight at 37 °C, 5% CO2. 3. The following day, remove the supernatant and add 100 μL of serum-free DMEM and incubate overnight at 37 °C, 5% CO2. 4. On the day of experiment, to prime the cells and increase the protein levels of NLRP3 remove the medium, add 100 μL fresh serum-free DMEM with LPS at a final concentration of 100 ng/mL. Make sure to include control wells that do not receive LPS (Fig. 4). 5. Incubate the cells for 3.5 h at 37 °C, 5% CO2 in the incubator. 6. Remove the supernatant and rinse the cells with pre-warmed PBS to remove any remaining LPS. 7. Discard the PBS and add 90 μL serum-free DMEM containing CRID3 at 1 μM to relevant wells. CRID3 is a wellcharacterized and specific NLRP3 inhibitor, an important control for the experiment. An alternative control would be NLRP3-/- cells. 8. After 30 min, add 10 μL of 100 μM nigericin (10×) directly on top of relevant wells, which will be diluted 1/10 to a final concentration of 10 μM. Nigericin triggers K+ efflux, and is a potent NLRP3inflammasome activator (see Note 26). 9. Incubate the cells for 45 min at 37 °C, 5% CO2. 10. Harvest the supernatant into a new 96-well plate and store at -20 °C until ready for further processing.
Activating and Monitoring Inflammasome Activation in Macrophages 3.2.5 Pyrin Inflammasome
423
1. Harvest the murine or human macrophages and resuspend them at 0.8 × 106 cells/mL in complete RPMI, then plate 100 μL in each a 96-well plate (for ELISA and LDH readouts) or 2500 μL in each well of a 6-well plate (for immunoblot). Sufficient wells of cells should be plated to include an untreated control, as well as wells stimulated with LPS alone. For an example, please see Fig. 4. 2. FOR HUMAN MACROPHAGES ONLY Human macrophages need to be incubated with LPS overnight in order to sufficiently increase expression of pyrin to enable a pyrin response. For all wells that will be stimulated to assess the pyrin inflammasome response add LPS at final concentration of 10 ng/mL. 3. For the assay, the media used depends on the readout desired. For experiments where LDH or immunoblot of the supernatant will be used as a readout, the cells should be stimulated in optiMEM, otherwise use complete RPMI. 4. FOR BMDM—If IL-1β release is to be used as a readout for Pyrin inflammasome activation, the BMDM need to be primed in LPS for 3 h to induce IL-1β expression. In this case, remove the media from the cells and add 50 μL of media containing 100 ng/mL LPS for a 96-well plate or 1000 μL of media for a 6-well plate. Incubate the cells for 3 h before addition of TcdA or BAA473. 5. To stimulate the Pyrin inflammasome remove the media from the well and add 100 μL containing 500 ng/mL TcdA for a 96-well plate or 1000 μL for a 6-well plate (see Notes 27 and 28). Additional conditions should be included with a caspase-1 inhibitor (VX765, 40 μM) or with a pyrin inflammasome inhibitor (colchicine, 2 μM). 6. Incubate the cells for 3 h at 37 °C. 7. The cells should become round following stimulation with TcdA. Once this occurs, the cells can be harvested or the media assessed for inflammasome activation. 8. For supernatant harvesting from a 96-well plate: Centrifuge the plate at 500 × g for 5 min, then remove the top 50 μL of supernatant. 9. For supernatant and cell harvesting from a 6-well plate: Remove the supernatant to a 1.5 ml microcentrifuge tube and centrifuge at 500 × g for 5 min. Remove the top 950 μL supernatant from the tube to use to precipitate for immunoblotting. The cells can be lysed in 100 μL RIPA buffer.
424
Marta Lovotti et al.
3.2.6 Non-Canonical Inflammasome Activation Using LPS Transfection
1. Seed the BMDMs or HMDMs in 96-well plates at the concentration of 50,000 cells/well. 2. Prime the cells with 10 ng/mL of human or murine IFN-γ for 16 h (this step is typically performed overnight). 3. Next day, prepare the transfection mixture. For each well (~50,000 cells), prepare separate dilution of LPS (2.5 μg/ 50,000 cells) and Lipofectamine 2000 (1 μL/50,000 cells), each in 25 μL of OptiMEM per well. Incubate for 5 min at RT. 4. Combine LPS and Lipofectamine-containing mixtures and incubate for 20 min at RT. 5. Take out the plates from the incubator and replace the cell culture medium with 150 μL/well of assay medium (OptiMEM or equivalent). 6. Add 50 μL of transfection mixture per well. 7. Centrifuge the plates for 10 min at 500 × g at 37 °C. 8. Place the plates back into the tissue culture incubator and incubate until the desired time points (typically 4–6 h). At this stage, the supernatants can be collected for IL-1β or LDH measurement, and cells can be lysed for Western blot or fixed for immunofluorescence analysis.
3.2.7 Activation of NLRC4 and Non-Canonical Inflammasome Using Salmonella Infection NLRC4 Activation
1. Approximately 2 days before the experiment: using a sterile inoculation loop, streak the bacteria from the frozen glycerol stock onto the LB plate containing the appropriate antibiotic (usually streptomycin). Incubate the plate overnight at 37 °C. The plate with colonies can be then sealed with parafilm and stored for 1 week at 4 °C. 2. The day before an experiment: seed BMDMs or hMDMs in a 96-well plate at the concentration of 50,000 cells/well. Incubate overnight at 37 °C. 3. To prepare the overnight culture of Salmonella, pick up several colonies using a sterile inoculation loop and transfer them into the round-bottom bacterial culture tube containing 2 mL of LB medium. Incubate overnight at 37 °C with shaking at 220 rpm in bacterial incubator (see Note 29). 4. Next morning, measure the optical density of the overnight culture using a spectrophotometer. Dilute the 100 μL of the overnight culture in 900 μL of fresh LB medium, and measure the absorbance at 600 nm (OD600). Use the plane LB to obtain a blank value. Multiply the obtained value by 10 to obtain the number of ODs. 5. To induce the expression of the virulence factors, prepare a subculture by diluting the overnight culture in the 2 mL of fresh LB to obtain OD of 0.05. Incubate for 4 h at 37 °C with shaking at 220 rpm.
Activating and Monitoring Inflammasome Activation in Macrophages
425
6. After setting up the bacterial subculture, prime the cells with 100 ng/mL of LPS for 4 h to induce IL-1β upregulation. 7. After 4 h of incubation: measure the OD of the subculture as described above. Calculate the number of the colony-forming units (CFUs) in the subculture, assuming that 1 OD600 contains approximately 8 × 108 bacteria. 8. To prepare the infection medium, spin down the 1 mL of bacterial culture at 9000 rpm for 1 min and resuspend them in an antibiotic-free macrophage medium. Dilute to obtain the desired MOI (typically 5–20) (see Notes 30 and 31). 9. (Optional) When using inflammasome inhibitors (such as VX-765), add the inhibitors to the respective wells at least 15–30 min before the infection. 10. Remove the priming medium from the cells and wash them 1–2 times with the fresh medium or PBS to remove residual antibiotic. 11. Add the infection medium to the cells. Centrifuge for 5 min at 500 × g to synchronize the infection. 12. Place the plates back to the tissue culture incubator. 13. After 1 h: collect the plates and process for the desired readouts. Non-Canonical Inflammasome Activation
1. This experiment is performed using stationary growth phase Salmonella, which does not normally express NAIP ligands. However, we strongly recommend using the strain deficient for the NAIP ligands, as even their low expression at the stationary phase might be sufficient for NLRC4 activation. 2. Prepare the overnight culture as described above. 3. Measure the OD and calculate the CFUs. 4. Collect 1 mL of the bacteria-containing LB and spin down at 9000 rpm for 1 min. Discard the supernatant and resuspend the bacterial pellet in 1 mL of fresh antibiotic-free medium. 5. Dilute the bacteria in the macrophage medium to obtain a MOI of 50. 6. Wash the cells with a warm medium or PBS to remove the residual antibiotics. 7. Add infection medium to the cells and centrifuge for 5 min at 500 × g to synchronize the infection. Incubate for 1 h at 37 °C. 8. Discard the infection medium and add the fresh medium containing 100 μg/mL of gentamicin to kill the extracellular bacteria. Incubate for 1 h. 9. Change the medium for the fresh medium containing 10 μg/mL gentamicin. 10. Incubate for 16–20 h before collecting the samples.
426
4
Marta Lovotti et al.
Notes 1. Whole blood and blood from leukocyte concentrate can be infectious and contain virus particles from hepatitis or human immunodeficiency virus (HIV). Due caution should be taken when purifying cells from blood, and the personnel purifying the cells should be vaccinated against Hepatitis B. Caution should be taken when handling blood samples and personal protective equipment should be used. 2. Buffy coats can be stored overnight at RT on a shaker prior to the isolation of PBMCs for a maximum of 16 h. 3. Before starting the procedure, plate an absorbent bench cover to line the tissue culture hood. 4. Put the scissors in 70% ethanol to sterilize it. 5. The minimum ratio for mixing the blood and PBS is 1:1, though yield is higher for more diluted blood. 6. Allow the diluted blood to slide down the side of the tube very slowly, to avoid breaking the surface plane and prevent mixing with Ficoll (e.g., hold the tube at an angle, use the pipette at the lowest speed settings). 7. In the event of a blood spillage inside or outside the hood, immediately clean up with Freka-NOL AF. 8. Deceleration disrupts the density gradient by mixing the phases. As an additional safety measure, place the lids onto the centrifuge chambers. 9. CD14+ monocytes are labeled with CD14 microbeads and enriched within the columns placed in the magnetic field of a separator. 10. MACS buffer should be added to the column less than a minute prior to the addition of the sample to prevent column from drying. 11. Whether you use the LS or MS columns depends on the number of PBMCs that are being used. LS columns should be used for higher numbers of cells (3 × 108 - 2 × 109), MS columns for up to 2 × 108 cells. 12. The details are not included in this protocol but the monocytes can also be purified by negative selection if desired. 13. The average buffy coat contains 1 × 109 PBMCs. If purifying from whole blood or a smaller leukocyte concentrate, instead count the number of PBMCs, then adjust the amount of MACS buffer and CD14 microbeads based on the ratio of PBMCs to beads used for the buffy coat. For example, 1 × 108 PBMCs would be resuspended in 100 μL with 20 μL CD14 microbeads.
Activating and Monitoring Inflammasome Activation in Macrophages
427
14. It is essential to use the Delta-surface plates for the macrophage differentiation process as these plates have a low-level surface adhesion which prevents background activation of the cells. 15. Try to differentiate the macrophages in an incubator that is not opened frequently, as this will reduce the yield following differentiation. 16. Murine pups that are just born and up to 2 days old can be used for microglia preparation. 17. AIM2-/- cells can be included as control. 18. The most commonly used activator for AIM2 is the model dsDNA poly-(dA:dT). However, also m-3M3FBS or thapsigargin can be used. These are activators of both the AIM2 and NLRP10 inflammasomes; macrophages only express AIM2. 19. VX-765 (Caspase-1/11 inhibitor) can be included as control. 20. For activation of NLRC4 using lethal factor fusion proteins, the specificity of the different NAIPs for the different bacterial components varies between mouse and human. For NLRC4 activation in human macrophages, needle proteins should be used while rod proteins should be used for murine macrophages. 21. For lethal factor fusion proteins, these can be produced recombinantly rather than purchased, though care should be taken to exclude LPS contamination if it is important. The protocols for protein purification are outside the scope of this protocol but can be found in the following references. 22. Activation of NLRC4 in both human and murine macrophages does not require prior exposure to TLR ligands. These only need to be included if IL-1β release is to be used as a readout. 23. The activity of the batches of both protective antigen and Lfn-needle and Lfn-rod proteins can vary. We recommend that a titration is performed for each new batch of toxin on each cell type for a range of 500 ng/mL to 1 ng/mL to determine the optimal toxin concentration. 24. For activation of NLRP1 using lethal factor and protective antigen. Activation of NLRP1 with lethal factor is specific to certain subtypes of mice that contain the version of the NLRP1b allele that contains the lethal factor cleavage site. This includes the cells from the 129 and Balb/c backgrounds but not B6 background. 25. The activity of the batches of both PA and Lfn can vary. We recommend that a titration is performed for each new batch of toxin on each cell type for a range of 5 μg/mL to 200 ng/mL to determine the optimal toxin concentration.
428
Marta Lovotti et al.
26. An alternative reagent that can be used to trigger NLRP3inflammasome assembly at step #8 is ATP at 10 mM final concentration for 45 min. To simulate neurodegenerative conditions, Aβ at 5 μM or α-Synuclein at 2 μM can be used, but importantly the incubation time changes from 45 min to 24 h. 27. The activity of the batches of TcdA can vary. We recommend that a titration is performed for each new batch of toxin on each cell type for a range of 1 ug/mL to 50 ng/mL to determine the optimal toxin concentration. 28. TcdA, protective antigen, Lfn-needle and Lfn-rod proteins, and lethal factor are proteins and so should be made up and aliquoted into single-use aliquots to avoid freeze/thawing effects. Freeze/thawing proteins will reduce their activity and may result in failure to activate the relevant inflammasome. 29. All experiments with Salmonella should be performed under BSL-2 conditions (unless using heavily attenuated strains, such as AroA, which are classified as BSL-1), and all equipment and medium containing live bacteria should be discarded appropriately. Following infection, live bacteria can be deactivated by PFA fixation; however, other types of samples, such as cell supernatants, may still contain live pathogens and should be handled appropriately. 30. The infectivity of different Salmonella strains and their potency to induce inflammasome activation may vary. We recommend performing the initial testing of the different MOIs to determine the most suitable MOI and incubation time for the strain of choice. 31. Using too high MOIs or prolonged infection times might induce non-specific cell lysis or engage other cell death programs and should be avoided. We recommend including the appropriate controls, such as inflammasome inhibitors or knock-out cells.
Acknowledgments We thank Susanne Schmidt and AG Schultze for the initial protocol for human macrophage differentiation. Thanks to Petr Broz and the members of Broz lab for advice with Salmonella infection protocol. All schematics were created with BioRender. K.S. is a recipient of SNSF Postdoc. Mobility fellowship (P500PB_211096). R.M. is supported by the Alzheimer Forschung Initiative (#20043). E.L. is supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy (EXC2151—390873048) and Helmholtz Association, under the project title “Immunology&Inflammation,” (#ZT-0027).
Activating and Monitoring Inflammasome Activation in Macrophages
429
References 1. Zheng D, Liwinski T, Elinav E (2020) Inflammasome activation and regulation: toward a better understanding of complex mechanisms. Cell Discov 6(1):36 2. Nozaki K, Li L, Miao EA (2022) Innate sensors trigger regulated cell death to combat intracellular infection. Annu Rev Immunol 40(1): 469–498 3. Liston A, Masters SL (2017) Homeostasisaltering molecular processes as mechanisms of inflammasome activation. Nat Rev Immunol 17(3):208–214
4. Mangan MSJ, Olhava EJ, Roush WR, Seidel HM, Glick GD, Latz E (2018) Targeting the NLRP3 inflammasome in inflammatory diseases. Nat Rev Drug Discov 17(8):588–606 5. Hanslik KL, Ulland TK (2020) The role of microglia and the Nlrp3 inflammasome in Alzheimer’s disease. Front Neurol 11:570711 6. McManus RM (2022) The role of immunity in Alzheimer’s disease. Adv Biol 6(5):e2101166
Chapter 29 Monitoring of Inflammasome Activation of Macrophages and Microglia In Vitro, Part 2: Assessing Inflammasome Activation Marta Lovotti, Matthew S. J. Mangan, Ro´isı´n M. McManus, Kateryna Shkarina, Matilde B. Vasconcelos, and Eicke Latz Abstract Inflammasomes are macromolecular complexes that assemble upon the detection of cytoplasmic pathogenassociated or danger-associated signals and induce a necrotic type of cell death termed pyroptosis, facilitating pro-inflammatory cytokine release. Inflammasomes play a critical role in innate immunity and inflammatory response; however, they have also been associated with multiple diseases, including autoinflammatory and neurodegenerative conditions. In the following chapter, we describe methods to detect inflammasome activation and its downstream effects, including detection of ASC oligomerization, detection of activated caspase-1 and cleaved IL-1β, as well as read-outs for inflammasome-mediated cell death. Key words Inflammasome, Pyroptosis, ASC speck, LDH release, Cytokine production, Western blot
1
Introduction Inflammasomes are large intracellular signaling complexes that consist of activated NOD-like (NLR) or AIM2-like (ALR) receptors and the adaptor protein ASC [1]. They serve as platforms for the recruitment and activation of protease caspase-1 and processing of its multiple downstream targets. Among caspase-1 substrates are the pore-forming protein Gasdermin D (GSDMD), which mediates downstream membrane permeabilization leading to the necrotic forms of cell death termed pyroptosis [2], and several pro-inflammatory cytokines, including the interleukin 1 family members IL-1β and IL-18. Both of these cytokines are synthesized
Marta Lovotti, Matthew S. J. Mangan, Ro´isı´n M. McManus, Kateryna Shkarina and Matilde B. Vasconcelos contributed equally with all other contributors. Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_29, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
431
432
Marta Lovotti et al.
as biologically inert cytoplasmic precursors (pro-forms) and are processed by activated caspase-1 into their mature bioactive forms, which are then released through GSDMD pores to trigger a pro-inflammatory response [3]. Later studies introduced an alternative non-canonical inflammasome pathway which does not involve NLRs and ASC, but is triggered by the direct sensing of the intracellular bacterial lipopolysaccharide (LPS) by caspase-11 (or human caspase 4 and caspase 5), also resulting in GSDMD processing and pyroptosis [4]. Inflammasome activation often occurs as part of the initial induction of inflammatory response during infection, trauma, and various other types of sterile tissue damage and is a critical component of both innate immunity and tissue repair [5–7]. In addition to its physiological function in host protection, aberrant inflammasome activity has been linked to a number of pathological conditions in humans, including cancer, chronic autoinflammatory diseases such as gout and rheumatoid arthritis, familiar Mediterranean fever and cryopyrin-associated periodic fever syndrome, and neurodegenerative diseases, as well as acute inflammatory conditions such as sepsis [8–10]. Thus, accurately measuring inflammasome activation is crucial for understanding its contribution toward the studied phenotype or disease model, as well as evaluating the potential of inflammasome activators or inhibitors. Typically, this involves a combination of several techniques, which aim to assess different steps of inflammasome complex assembly and downstream signaling. ASC speck formation is commonly used as a read-out for ASC-dependent inflammasome activation. In primary macrophages, ASC specks can be detected using immunofluorescent staining with the antiASC antibodies [11]. Alternatively, cells expressing fluorescently tagged ASC (such as ASC-mCitrine-expressing primary or immortalized macrophages) can be utilized to directly monitor ASC speck assembly both in vitro and in vivo [12]. In addition to microscopy, ASC oligomerization can be measured using immunoblot, by cross-linking proteins post-lysis. Western blot (and more recently developed semi-quantitative systems, such as capillary electrophoresis) can also be utilized to detect processing and activation of downstream inflammasome components, including caspase 1 (and for non-canonical inflammasomes, caspase 11, 4, and 5), IL-1β and IL-18, and GSDMD. In comparison with a regular Western blot, the simple Western method is more sensitive and is able to detect lower amounts of protein in lower volumes. Its high sensitivity also enables the detection of cleaved caspase-1 and cytokines in the cell supernatants without the need for a separate protein precipitation set. Cytokine release from pyroptotic cells is one of the most commonly employed indirect inflammasome activation read-outs. It is frequently measured in cell supernatants using quantitative
Assessment of Inflammasome Activation
433
techniques such as enzyme-linked immunosorbent assay (ELISA) and homogeneous time-resolved fluorescence (HTRF) assay. Finally, pyroptotic cell death itself can be utilized as a read-out for inflammasome activation. Identification of pyroptotic cells usually relies on the detection of membrane permeabilization using membrane-impermeable nuclear acid stains, such as propidium iodide (PI), CellTox, Cytox, or DRAQ7. These dyes are non-fluorescent in their unbound state, however rapidly gain fluorescence following entry into the permeabilized cells and nucleic acid binding. Additionally, loss of the plasma membrane lipid asymmetry and the phosphatidylserine exposure on the outer membrane leaflet, another common characteristics of the dying cells, can be detected using the binding of the fluorescently labeled Annexin V. Another type of assay to detect pyroptotic cell lysis relies on the detection of the cytosolic enzyme lactate dehydrogenase (LDH), which is released into the extracellular environment following inflammasome activation. The released LDH can be measured using a coupled enzymatic reaction. In this setup, LDH activity can catalyze the conversion of lactate to pyruvate, and during this process, NAD+ is reduced to NADH. NADH is then used by diaphorase to reduce a tetrazolium salt (INT) into formazan, a red product, which is measured at 490 nm. The amount of LDH released in the supernatant is directly proportional to the level of formazan formation. Below, we provide the protocols for measuring different steps of inflammasome activation and pyroptosis using these techniques.
2
Materials
2.1 Common Consumables and Equipment (SeeNotes 1 and 2)
1. Tissue culture equipment (laminar flow cabinet or equivalent). 2. Tissue culture incubator equipped with CO2 supply. 3. Centrifuge. 4. Cell counting equipment. 5. Serological pipettes. 6. General consumables for cell culturing (Petri dishes, tips, cell scrapers, etc.). 7. Disposable or autoclavable waste bottles. 8. Glassware, used for buffer preparation. 9. 15-mL and 50-mL centrifuge tubes. 10. Parafilm. 11. 1.5-mL microcentrifuge tubes. 12. Aluminum foil.
434
Marta Lovotti et al.
13. Cell culture medium (DMEM or RPMI), supplemented with 10% goat serum, antibiotics, and GlutaMax. 14. Opti-MEM reduced serum medium, phenol red free (or equivalent imaging medium). 15. PBS, cell culture grade. 16. Priming agents and inflammasome triggers. 17. Propidium iodide. 18. Heating block. 19. Light microscope. 20. Epifluorescent or confocal microscope equipped with the appropriate set of filters or lasers. 21. Plate reader capable of fluorescence detection (such as Cytation 5, SpectraMax 3, or equivalent). 22. Westen blot imager or other setup for the Western blot detection. 23. Image analysis software (such as FiJi). 24. Statistical analysis software. 2.2 Consumables and Buffers for the ASC Speck Staining
1. Clear-bottom 96-well plates, microscopy suitable. 2. Blocking serum (goat serum, horse serum, or equivalent). 3. Methanol-free 16% paraformaldehyde (PFA), stored protected from light. 4. Fixation buffer: 4% PFA in PBS. 5. Quenching buffer: 50 mM NH4Cl, 0.1 M glycine in PBS, stored at room temperature (RT). 6. Blocking/permeabilization buffer: 10% goat serum, 0.1% Tween 20 (or 0.5% Triton X-100) in PBS. Stored at 4 °C. 7. Wash buffer: 0.1% Tween-20 in PBS, stored at RT. 8. Anti-ASC antibodies. We recommend using mouse monoclonal anti-ASC antibody from Millipore (clone 2EI-7), or purified anti-ASC (TMS-1) antibody from BioLegend (clone HASC-71). For detailed analysis of ASCantibodies suitable for immunofluorescent ASC speck detection, see [1]. 9. Purified mouse IgG control antibodies. 10. Secondary anti-mouse antibody labeled with the respective fluorophores (stored protected from light at all times to prevent fluorophore bleaching). 11. Nuclear counterstains: DAPI, DRAQ5, or equivalent. These dyes should be stored protected from light at 4 °C.
Assessment of Inflammasome Activation
2.3 Reagents and Consumables for Pyroptotic Cell Detection
435
1. Opti-MEM reduced serum medium (phenol red free) or equivalent imaging medium. 2. Membrane-impermeable nucleic acid stains, such as propidium iodide, CellTox Green, Cytox, DRAQ7, or equivalent. 3. Live-cell (membrane-permeable) nuclear counterstain, such as DRAQ5, Hoechst 33342, or equivalent. 4. Purified Annexin V labeled with the appropriate fluorophores.
2.4 Reagents for Pyroptosis Detection Using Propidium Iodide Incorporation
1. 10× staining solution: 50 μM PI in Opti-MEM.
2.5 Reagents and Equipment for Western Blot Analysis and ASC Cross-Linking
1. Protein gel electrophoresis chamber.
2. Lysis buffer: 10% Triton X-100 in PBS can be stored for up to 1 month at RT.
2. Protein electrotransfer chamber. 3. BCA assay kit. 4. Lysis buffer: RIPA 2×: 150 mM NaCl, 50 mM Tris–HCl, 1% nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS in ddH2O. 5. HEPES lysis buffer: 150 mM KCl, 1% nonidet P-40, protease inhibitors in 20 mM HEPES, pH 7.5. 6. Complete, EDTA-free protease inhibitor cocktail (Roche, Cat #11873580001). 7. PhoSTOP Phosphatase inhibitor cocktail (Roche, Cat #4906845001). 8. BS3 (bis(sulfosuccinimidyl)suberate) (Thermo Fisher, Cat # A39266). 9. BS3 (bis(sulfosuccinimidyl)suberate) (Thermo Fisher, Cat # A39266). 10. Methanol (reagent grade). 11. Chloroform. 12. 4× Laemmli sample buffer (LDS). 13. 10× Sample reducing agent (Thermo Fisher, Cat #NP0004). 14. Prestained Protein Ladder, 10 to 180 kDa. 15. Precast NuPAGE 4–12% Bis-Tris protein gel (Thermo Fisher, Cat #NP0336BOX). 16. 20× MOPS or MES NuPAGE running buffer (Thermo Fisher, Cat #NP0001 or #NP0002). 17. Immobilon-FL PVDF transfer membrane (Millipore Cat #IPFL00010). 18. Transfer buffer: 100 mL 10× Tris-glycine buffer (Pierce, Cat #28363), 200 mL methanol and 700 mL ddH2O.
436
Marta Lovotti et al.
Table 1 Antibodies for detection of inflammasome components (human) Antibody
Order #
Dilution
IL-1β
RnD systems #AF-201-SP
1:500
IL-18
mAbProtein #R-I-002
1:500
Cleaved caspase-1
CST #4199T
1:1000
Total caspase-1 (Bally-1)
Adipogen #AG-20B-0048-C100
1:500
NLRP3 (Cryo-2)
Adipogen #AG-20B-0014-C100
1:1000
ASC (AL177)
Adipogen #AG-20B-0006-C100
1:250
Table 2 Antibodies for detection of inflammasome components (mouse) Antibody
Order #
Dilution
IL-1β
RnD systems #AF-401-SP
1:500
Caspase-1 (Casper-1)
Adipogen #AG-20B-0042-C100
1:1000
NLRP3 (Cryo-2)
Adipogen #AG-20B-0014-C100
1:1000
ASC (AL177)
Adipogen #AG-20B-0006-C100
1:250
19. Whatman blotting paper. 20. Sponge pads for immunoblotting. 21. Blocking solution: 3% bovine serum albumin (BSA) in 1× Trisbuffered saline (TBS, pre-made solution). 22. Wash/antibody buffer: 3% BSA, 0.1% Tween-20 in 1× TBS. 23. 1-mL syringe. 24. 21-gauge needle. 25. Primary antibodies (Tables 1 and 2), for cross-linked ASC detection we recommend using anti-ASC antibody, polyclonal AL177 (Adipogen, Cat # AG-25B-0006-C100). 26. Appropriate secondary antibodies (conjugated to fluorophores or HRP, depending on the available detection method). 2.6 Reagents for IL1β Detection Using ELISA
1. Wash buffer: 0.05% Tween-20 in PBS, pH 7.2–7.4. 2. Reagent diluent: 1% BSA in PBS, pH 7.2–7.4, filtered. 3. 1-Step Ultra TMB-ELISA Substrate Solution-1 L (Thermo Fisher, Cat # 34029). 4. Stop Solution: 2 M H2SO4 in ddH2O.
Assessment of Inflammasome Activation
437
5. IL-1β ELISA kit, which contains the capture antibody, detection antibody, streptavidin–HRP (R&D Systems, Cat # DY401). 2.7 Reagents for LDH Assay
1. Pierce LDH Cytotoxicity Assay Kit (Thermo Scientific, cat # 88953), which contains the assay buffer, substrate mice, stop solution, lysis buffer and LDH positive control. If not available, the kits from the other suppliers can be used.
2.8 Homogeneous Time-Resolved Fluorescence (HTRF) Assay
1. 384-well small volume HTRF plate, hi-base, white (Greiner Bio-One GmbH, Cat # 784075). 2. Mouse/Human IL-1β HTRF® kit (Cisbio, Cat #62HIL1BPET or #62MIL1BPEG). 3. Mouse/Human TNF-α HTRF® kit (Cisbio, Cat #62HTNFAPET or #62MTNFAPEG).
2.9 Reagents for the Detection of IL-1β and IL-18 Processing Using WES
1. Simple Western (Protein Simple). 2. 12–230 kDa Wes Separation Module, consisting of 25-capillary cartridge and pre-filled microplates (ProteinSimple). 3. EZ Standard Pack, consisting of ready-to-use Biotinylated Ladder, Fluorescent 5× Master Mix, DTT. 4. Antibodies for human IL-1β (cleaved) and IL-18 neoepitope in Table 1.
3
Methods
3.1 Quantification of ASC Specks Using ASCAntibody Staining and Microscopy
1. Approximately 16–20 h before experiment, seed the macrophages into the tissue culture-treated clear-bottom 96-well plates at the concentration of 50,000 cells/well (see Note 3). 2. Incubate the cells at 37 °C degrees 5% CO2 overnight. 3. On the day of experiment, prime and stimulate the cells using preferred stimuli as described in Chapter 28. 4. At the desired time point, remove the medium from the plates and wash the macrophages 3× very gently with PBS. 5. Add 100 μL of fixation buffer per well, seal the plates and incubate 15 min at RT. To avoid exposure to the PFA fumes, this step should be performed at the ventilated fume hood. PFA fixation will also deactivate Salmonella or other BSL-2 pathogens, enabling further handling of infected plates under BSL-1 conditions. 6. Dispose the fixative into the appropriate organic waste container and wash 3× with PBS.
438
Marta Lovotti et al.
7. Add 100 μL of quenching buffer per well and incubate for 2–3 h. Alternatively, this step can be performed overnight at 4 °C. 8. Remove the quenching buffer and add 100 μL of blocking/ permeabilization buffer. Incubate for 1 h at RT. 9. During incubation, prepare the first staining solution by diluting the primary anti-ASC antibody (1:500) in the blocking buffer (see Note 4). 10. Replace the blocking buffer with the staining solution (50 μL/ well of 96-well plate). 11. Incubate for 1–2 h at RT. Alternatively, the plates can be sealed and incubated overnight at 4 °C. 12. Wash 3× with the wash buffer. 13. Add the appropriate secondary antibody (1:500) pre-diluted in the blocking buffer. 14. Incubate the plates for 45 min at RT, protected from light. 15. Wash 5× 5 min with the wash buffer. 16. To label the nuclei, incubate the samples with 1 μg/mL DAPI in PBS for 15 min, protected from light. 17. Wash 2–3× in PBS. 18. The plates can be imaged immediately or sealed with parafilm, wrapped in foil to prevent fluorophore bleaching, and stored for 1–2 weeks at 4 °C. 19. ASC speck detection can be visualized using epifluorescent or confocal microscopy. Usually, epifluorescence microscopy is sufficient to perform simple speck detection and quantification, while the confocal microscope is typically necessary to perform more elaborate co-localization studies. Use 20× or 40× objective with the filters or lasers for the appropriate for the selected combination of the secondary antibody fluorophore and nuclear dye. During acquisition, use the same parameters (light intensity, exposure, gain, etc.) for each condition. Images should be acquired in an unbiased manner (e.g., by randomly choosing the field of view or using the software-defined positions). After the staining, ASC specks can be visible as bright aggregates of approximately 1–2 μm (larger if using cell types overexpressing fluorescently tagged ASC reporters), typically located next to the cell nuclei. The ASC speck quantification can be performed using image analysis software, such as FiJi (https://imagej.net/software/fiji/) or CellProfiler (https:// cellprofiler.org/). ASC specks and nuclei can be automatically segmented using the appropriate thresholding technique, and the percentage of the ASC-positive cells can be quantified using ratio of ASC specks per nuclei. Statistical analysis software
Assessment of Inflammasome Activation
439
Fig. 1 Example of the data obtained after ASC speck quantification of BMDMs treated with inflammasome activators
(Microsoft Excel and GraphPad Prism) is used to combine and graph the results (Fig. 1) (see Note 5). 3.2 Visualization of the Pyroptotic Cell Death Using Microscopy
1. Seed the cells into the 96-well plates at the appropriate cell density (usually 30,000–40,000 cells/well). For this experiment, we recommend using a lower seeding density than for other assays to aid the clear cell shape discrimination for highresolution live-cell imaging. 2. If applicable: prime the cells with the TLR ligand of choice, as previously described. 3. Before the stimulation: remove the cell culture/priming medium and add 100 μL of the nuclear counterstain solution per well. 4. Incubate for 30–60 min at 37 °C, protected from light. At this stage, all further manipulations need to be performed under dim light conditions to avoid the fluorophore bleaching. 5. Remove the nuclear counterstain solution and wash the cells 2× with the pre-warmed PBS to remove the remaining dye. 6. In each well, add 100 μL of the pre-warmed pyroptosis detection medium containing Opti-MEM supplemented with Annexin V-Pacific Blue (1:300) + CellTox Green (1:10,000). For quantitative analysis, a cell-permeable nuclear counterstain, such as DRAQ5, can be included (see Note 6). 7. Stimulate the cells using the inflammasome activator of choice. 8. For the time-lapse imaging: start image acquisition immediately after stimulation (this step can be delayed for the slowacting inflammasome triggers, such as talabostat or LPS transfection). 9. For endpoint quantification: incubate the plates until the desired timepoint, then acquire the images. The representative
440
Marta Lovotti et al.
Fig. 2 Representative images of the hMDMs left untreated (top) or treated with the NLRC4inflammasome activator PrgI, after 1 h of incubation. CellTox Green staining indicates the membrane permeabilization, and Annexin V is used to visualize the phosphatidylserine exposure on the pyroptotic cells
images of control (non-treated) and pyroptotic cells stained with CellTox green and Annexin V are shown in Fig. 2. 10. For more quantitative analysis, one can also use the ratio between permeabilized (CellTox+) and total (DRAQ5+) cell nuclei. 3.3 Measuring Pyroptosis Using Propidium Iodide (PI) Incorporation
1. Seed the cells in the black 96-well plates at the concentration of 50,000 cells per well. 2. Prime and stimulate using the trigger and the protocol of choice (see sections above for the details). In this experiment, inflammasome activation should be performed in the 90 μL of Opti-MEM or equivalent phenol red-free medium. 3. Add 10 μL of 10× PI staining solution on top to reach the final concentration of 5 μM. 4. To the positive (total lysis) control wells: add 10 μL of the lysis buffer per well (to the final concentration of 1% Triton X-100). Pipette up and down 5–10 times to lyse the cells. 5. For the time course experiments: measure the PI fluorescence using 535 nm excitation and 617 nm emission every 5–15 min (see Note 7). 6. Alternatively, incubate the plates at 37 °C degrees in the dark (for example, wrapped in foil) until the desired timepoint and perform the reading afterward. 7. Following the data acquisition, the percentage of the PI+ cells can be calculated as following: (Sample fluorescence—Background fluorescence) / (Lysed cells fluorescence—background fluorescence), where the background fluorescence typically refers to the fluorescence values obtained from the non-stimulated cells. Example of the data obtained from such measurement is shown in Fig. 3.
Assessment of Inflammasome Activation
441
Fig. 3 Example of the data obtained from the PI incorporation in the cells treated with the NLRP3inflammasome activators
3.4 Immunoblot for Cleaved IL-1β, Cleaved Caspase 1, and Inflammasome Components
1. Pre-cool a centrifuge and place the plate containing the macrophages on ice.
3.4.1 Lysis of Stimulated Cells and Sample Preparation (6 Wells)
5. Spin down at 6000 × g for 15 min at 4 °C to pellet debris and nuclei.
2. Remove culture medium from cells and wash 2× in cold PBS. 3. Add 100 μL lysis buffer to each well and scrape cells. 4. Collect lysates in microcentrifuge tubes.
6. Remove supernatant to a new tube. 7. Quantitate the protein concentration for each sample by BCA assay. 8. Normalize the protein amount so that it is the same in all samples. 9. Mix protein samples with 4× LDS sample buffer and 10× sample reducing agent according to manufacturer’s instructions. 10. Heat samples at 85 °C for 10 min.
3.4.2 Supernatant Harvesting and Precipitation
1. Harvest the supernatant from the different wells into microcentrifuge tubes (the total amount of supernatant should be just over 1 mL). 2. Centrifuge the tubes 500 × g for 5 min. 3. Remove 1 mL of supernatant to a new microcentrifuge tube. At this point, the supernatant can be frozen and stored at -20 °C. 4. For precipitation, take 500 μL of supernatant to a new tube and add 500 μL of methanol and 125 μL of chloroform. 5. Vortex. 6. Centrifuge the tubes at 12,000 × g for 3 min.
442
Marta Lovotti et al.
7. After centrifugation, two separate layers will form with an interface between them. There should be white precipitate at the interface between the two chemicals; this is the protein. 8. Remove the upper layer until just above the interface being careful not to disturb the protein. 9. Add 500 μL methanol. 10. Vortex. 11. Centrifuge the tubes at 12,000 × g for 3 min. 12. The protein pellet should be evident on the bottom of the tube. 13. Remove the methanol by pouring it out of the tube, then blotting on absorbent paper to remove the remaining methanol. 14. Once the pellet is completely dry, add 1× LSB containing sample reduction agent to the pellet (choose an amount that allows you to load the entire resuspended sample in 1 lane of the protein gel. 15. Heat the sample at 95 °C for 5 min to help re-solubilize the protein mix. 16. Run on the NuPAGE gel directly or store at -20 °C. 3.4.3
Gel Electrophoresis
1. Load the protein/LDS/sample reducing agent mix into each well of a NuPAGE 4–12% Bis–Tris protein gel. The volumes the wells can hold vary between gels and should be determined before loading the samples. 2. Load the protein standards into one well on the gel. 3. Separate denatured and reduced proteins on a NuPAGE 4–12% Bis–Tris protein gel in MES or MOPS buffer. MOPS buffer gives better separation across all sizes while MES buffer resolves proteins of lower molecular weight more accurately. 4. Run the gel at 150 V for 1 h or until the lowest standard has reached the bottom. For the protein standards recommended in this protocol, the protein standard is stained in green.
3.4.4 Protein Transfer to PVDF Membrane
1. Activate the PDVF membrane by placing it in methanol until it is coated. 2. Presoak sponges, Whatman paper, and methanol-activated PDVF membrane in transfer buffer until all components are saturated. 3. Assemble the material for transfer—the order for back to front is sponges (2×), Whatman paper, NuPAGE gel, PVDF membrane, Whatman paper, sponges (2×). Ensure that you remove any bubbles between the gel and PVDF membrane.
Assessment of Inflammasome Activation
443
Fig. 4 Immunoblot of the supernatant fraction of hMDM treated as listed and probed for IL-1β or caspase-1. The cleaved forms of IL-1β (p17) and caspase-1 (p20) are highlighted
4. Transfer at 32 V for 1 h (cleaved IL-1β and caspase-1) or 1.5 h (Other inflammasome components). 3.4.5 Developing the Immunoblot (Fig. 4)
1. Transfer the membrane to a 50-mL centrifuge tube filled with 10 mL blocking solution. 2. Incubate for 1 h at RT, or overnight at 4 °C. 3. Incubate membrane with primary antibody diluted in antibody buffer for 1 h at RT, or overnight at 4 °C. 4. Wash membrane with wash buffer 3 times for 10 min. 5. Incubate membrane with secondary antibody (1:20.000) at RT for 2 h. 6. Wash membrane with wash buffer 3 times for 10 min. 7. Develop the blot using a fluorescence scanner such as the LI-COR Odyssey system.
3.5 ASC CrossLinking and Detection Using Western Blot
1. For this protocol, the macrophages should be seeded and stimulated in 6-well plates, ideally 2 × 106 cells/well. 2. Remove the media from the cells into a 1.5-mL microcentrifuge tube and centrifuge at 500 × g for 5 min. 3. Add 0.5 mL lysis buffer to each well of cells and scrape to disrupt the cells. 4. Remove the supernatant from the centrifuged cells and either discard or set aside for other readouts (see Note 8).
444
Marta Lovotti et al.
5. Add the lysis mixture from the relevant well into the matching tube. 6. Pulse the lysate through a 21-gauge needle attached to a 1-mL syringe 20 times to disrupt the DNA. 7. Centrifuge the lysate at 6000 × g for 10 min at 4 °C. 8. Following centrifugation, a pellet should be present at the bottom of the tube. Remove the supernatant to a different tube or discard. 9. Wash the pellet twice with PBS, centrifuging for 6000 × g for 10 mins at 4 °C each time. 10. Resuspend the pellet in 200 μL of PBS. 11. Add BS3 to a final concentration of 2 mM and mix. 12. Incubate the tubes at RT in the dark for 30 min. 13. Centrifuge the tubes at 6000 × g for 10 mins. 14. Remove the supernatant. 15. Resuspend the pellet in 1× LSB containing 10% DTT. 16. Boil the samples for 5 min and 95 °C. 17. Resolve the samples using an SDS–PAGE (as described in Subheading 3.4.3) and transfer to a nitrocellulose or PVDF membrane as detailed in the immunoblot section later in this chapter. 18. For ASC detection, use the ASCantibody diluted to 50 mg/ mL in TBST 1% BSA. 3.6 Detection Inflammasome Activation and IL-1β Release Using WES Capillary Protein Electrophoresis
1. Prepare the EZ standard pack reagents: To the DTT tube, add 40 μL of deionized water (to make a 400 mM solution); to the Fluorescent 5× Master Mix, add 20 μL of the prepared 400 mM DTT solution and 20 μL of 10× sample buffer (provided with the Wes Separation Module); to the biotinylated ladder, add 20 μL of deionized water. Always gently mix by pipetting with the 20- to 200-μL pipette. 2. Combine four parts of sample (supernatant/lysate) with one part of the 5× Fluorescent Master Mix (containing a fluorescently labeled standard, dithiothreitol (DTT), and sample buffer)—e.g., mix 4.8 μL of sample with 1.2 μL of 5× Fluorescent Master Mix. The optimal protein concentration depends on the expression level. If needed, dilute sample with 0.1× sample buffer. 3. Quickly vortex the samples, denature by incubation at 95 °C for 5 min, vortex again, and spin down. Store on ice until further use. 4. Prepare the primary antibody dilution: Dilute the primary antibody (e.g., for cleaved IL-1β) in the provided antibody
Assessment of Inflammasome Activation
445
diluent of the WES module and store the diluted antibodies on ice until the further use. Primary antibodies are usually diluted 1:50 for this method but antibody titrations can be performed. 5. As secondary antibody, the species-specific, ready-to-use detection antibody provided with the WES module is used. 6. Prepare the Luminol-S and Peroxide mix: Combine 200 μL of Luminol-S with 200 μL of peroxide, gently pipette up and down, and store on ice until further use. 7. Following the simple Western instructions, load 5 μL of biotinylated ladder and 3–5 μL of samples; 10 μL of the antibody diluent, the primary and secondary antibodies, the streptavidin–horseradish peroxidase conjugate; 15 μL of the Luminol-S/peroxide chemiluminescent substrate; and 500 μL of the wash buffer into the microplate. The plate is already pre-filled with split running buffer, wash buffer, and 10× sample buffer. 8. Protect the microplate with a lid and centrifuge it for 5 min, 1000 × g at RT. 9. Set the desired assay parameters in the Simple Western software (Compass) as follows: 200 s (sec) loading time of the separation matrix, 15 s loading time of the stacking matrix, 9 s loading time of the sample, 25 min separation time of the sample at 375 V, 90-min incubation time with the primary antibody, and 30-min incubation time with the secondary antibody. 10. Open the WES’s door and insert the 25-capillary cartridge into the cartridge holder (the light will change from orange to blue). Holding the plate firmly on the bench, remove the lid, carefully peel off the evaporation seal, and pop any bubbles (with a pipette tip). Insert the microplate on the plate holder and close the WES’s door. 11. Start the run, which will take approximately 4 h. Once complete, discard the plate and cartridge. 3.7 Detection of IL1β Release Using ELISA
1. To prepare the capture and detection antibodies and IL-1β standard, reconstitute each individual vial according to the manufacturer’s recommendations. 2. Add 100 μL of the diluted capture antibody per well of an ELISA plate. Seal the plate with an adhesive strip and leave overnight at RT (Fig. 5). 3. The following morning, discard the capture antibody and wash the wells 3 times with 150 μL wash buffer. After the last wash, dab the plate against paper towels to ensure all wash buffer is removed. 4. Add 200 μL reagent diluent to each well and leave for 1 h at RT to block the plate and prevent non-specific binding.
446
Marta Lovotti et al.
Fig. 5 Schematic of the IL-1β ELISA setup
5. During this step, prepare serial dilutions of IL-1β standard. Using the lot-specific certificate of analysis for the kit, dilute the standard to a top concentration 1000 pg/mL. Create a standard curve with 7 serial dilutions of known concentration of IL-1β ranging from 1000 pg/mL to 15.6 pg/mL and include a blank sample containing no IL-1β, just media or reagent diluent. 6. Wash the wells 3 times with 150 μL wash buffer. After the last wash, dab the plate against paper towels to ensure all wash buffer is removed. 7. Add 100 μL of the IL-1β standard or sample to each well. Note some samples that were treated with strong stimuli may need to be diluted to ensure being within the range of the standard curve (e.g., a 1:10 dilution). Samples can be diluted in media or reagent diluent. Cover the plate with an adhesive strip and leave for 2 h at RT. 8. Wash the wells 3 times with 150 μL wash buffer and blot the plate against paper towels to ensure all wash buffer is removed. 9. Add 100 μL of the diluted detection antibody per well. Seal the plate and leave for 2 h at RT. 10. Wash the plate with 150 μL wash buffer, and blot against paper towels. 11. Add 100 μL of the diluted streptavidin–HRP per well. Seal the plate and leave for 20 min at RT in the dark. 12. Wash the plate with 150 μL wash buffer, and blot against paper towels. 13. Add 100 μL of substrate solution per well. Incubate for up to 20 min at RT in the dark. 14. Stop the reaction by adding 50 μL of stop solution to each well. To ensure mixing of the reagents, gently tap the plate. 15. Obtain the optical density (OD) by reading the plate on a microplate reader at 450 nm. If possible, use 540 nm or 570 nm as a reference wavelength.
Assessment of Inflammasome Activation
447
Fig. 6 Representative standard curve and sample results of IL-1β release to the supernatant under NLRP3 inflammasome-stimulating conditions
16. To analyze the results, subtract the blank from all wells. Using programs such as Excel or PRISM, create a standard curve with the known OD values and matching concentrations from the IL-1β standard curve (Fig. 6). 17. Using the OD values from the samples, you can extract the unknowns from the standard curve, providing the protein concentration of IL-1β in the samples (Fig. 6). Make sure to scale up any samples that were diluted to fit within the range of the standard curve (see Note 9). 3.8 Measuring Pyroptosis Using Lactate Dehydrogenase Activity Assay (LDH)
1. The general principle of LDH assay is illustrated at Fig. 7. 2. Treat cells under various conditions as described in Chap. 28 to trigger inflammasome activation and pyroptosis. 3. Add 10 μL of the 10× lysis buffer to wells in triplicate for 45 min to trigger cell lysis and release of LDH as a positive control (known as maximum activity control). Add 10 μL of sterile Milli-Q water to wells in triplicate for 45 min (known as spontaneous activity control). Incubate the plate for 45 min at 37 °C, 5% CO2. 4. Collect supernatants for a LDH assay, which can be carried out directly. Alternatively, the supernatants can be stored at -20 °C. 5. Prepare the reaction mix by combining assay buffer and substrate mix in a 1:45 dilution in a 15-mL centrifuge tube. Mix the solution by gently inverting and ensure to protect from light until ready to use.
448
Marta Lovotti et al.
Fig. 7 Schematic demonstrating how LDH release from dying cells can be detected
6. Add 50 μL of the sample supernatant (i.e., target samples, maximum activity control, spontaneous activity control, and 1× LDH control) to a 96-well flat-bottom plate in duplicates. 7. Add 50 μL of the reaction mix to each well, and gently tap the plate to mix the contents. Seal the plate with an adhesive strip. 8. Incubate the plates at 37 °C for 30 min protected from light. 9. Stop the reaction with 50 μL of the stop solution. 10. Measure the absorbance on a plate reader at 490 nm with 600 or 680 nm as reference. 11. To calculate the LDH released (Fig. 7), subtract the spontaneous activity control (water-treated sample) from the target sample (conditions to trigger inflammasome activation), and divide by the total LDH activity [maximum activity control (Triton-treated sample) minus the spontaneous activity control (water treated sample)], and multiply by 100: %Cytotoxicity :
Target sample LDH activity - Spontaneous Activity Control × 100 Maximum Activity Control - Spontaneous Activity Control
The example of the data obtained during this type of assay is provided as in Fig. 8. 3.9 Homogeneous Time-Resolved Fluorescence (HTRF) Assay
Reconstitute the reagents according to the manufacturer’s recommendation. 1. Reconstitute the reagents in the kit according to the manufacturer’s recommendation. Reconstituted reagents can be aliquoted and stored at -80 °C for several months. 2. Stimulate the cells with the inflammasome trigger of interest. 3. Collect supernatants from stimulated cells in 384-well plates. Always keep the supernatants on ice. The samples can also be stored long-term at -20 °C. 4. Dispense the reagents into each 384 well as follows: 2 μL of antibody 1, 2 μL of antibody 2, and 16 μL of the standards or sample. 5. Spin plate at RT, 340 g, for 1 min.
Assessment of Inflammasome Activation
449
Fig. 8 Sample results of LDH release to the supernatant under NLRP3 inflammasome-stimulating conditions
6. Cover the plate with a plate sealer. 7. Incubate the plate in the dark at RT for 3 h. Alternatively, plates can be also incubated at 4 °C overnight. 8. Read the fluorescence emission of the two antibodies on a plate reader. The plate reader must be capable of HTRF measurement.
4
Notes 1. We recommend using most of the reagents and dilutions according to the manufacturers’ recommendations. However, for experiments involving antibodies, the optimal dilution can be determined experimentally using serial titration. Also, the recommended volumes for some reagents (LDH assay mix, BCA reagents) can be scaled down if necessary by using different vessels (384-well plate instead of 96-well) or reducing the general reaction mix volume (e.g., using 100 μL per well instead of recommended 200). 2. Most of the reagents (buffers, staining solutions, kits) should be prepared in advance and brought to room temperature before the beginning of experiment, unless indicated otherwise. 3. When using ASC reporter-expressing macrophages, the cells can be imaged directly after fixation. In this case, a nuclear counterstain can be added to the imaging medium. We typically
450
Marta Lovotti et al.
use PFA fixation for ASC speck analysis. However, in case of co-staining with other antigens, it might be necessary to use other fixative agents (such as methanol or acetone), which might impact antigen localization or epitope recognition. In this case, we recommend performing a sidewise comparison of ASC staining with the same antibody between differentially fixed samples. 4. We recommend including the following controls for this experiment: During stimulation, non-stimulated cells, which should not contain ASC specks, and primed cells, since in some cell types, priming alone can trigger inflammasome activation in a subfraction of the cells; during the staining step: an isotype control: a non-specific antibody matched to the primary antibody’s host species and class (including the light chains), used to discriminate against the non-specific aggregation or the presence of other artifacts; a secondary antibody only control, used to detect the non-specific secondary antibody binding or aggregation; the non-stained cells, used to detect sample autofluorescence. To obtain the best signal-to-noise ratio, it might be necessary to titrate the concentration of the primary antibodies. We have previously found that primary antibody concentrations ranging from 2 to 10 μg/mL usually give the best results [1]. 5. For imaging experiments, the optimal combination of the fluorophores and wavelengths for the nuclear stain, membrane permeabilization detection dye, and Annexin is often determined based on the available laser/filter sets. The cell permeabilization dyes, such as CellTox green, or aggregated ASC, increase their intensity 50- to 100-fold upon DNA binding and have a tendency to bleed into the other channels. To avoid this, we recommend including single fluorophore controls and titrating the dyes on viable and necrotic (permeabilized) cells before performing an experiment. 6. The nuclear labeling by CellTox, PI, or equivalent dyes is non-selective against other types of necrotic cell death (such as late-stage necroptosis, ferroptosis, and others) and thus should be utilized with caution and in combination with other techniques when trying to discriminate between different cell death modalities. This also applies to Annexin V, which labels all the cells with the scrambled membranes, including apoptotic cells. 7. Some inflammasome activators, such as nigericin, can induce pyroptosis rapidly. When performing a time-lapse experiment with such triggers, we recommend setting up the acquisition parameters first, and then adding the ligand directly to the assay medium, so that data acquisition can be started immediately
Assessment of Inflammasome Activation
451
following the stimulation. For the activators with the delayed pyroptosis induction (Val-boropro, LPS transfection, stationary phase Salmonella infection), we recommend starting the acquisition at the later timepoints (usually 4–6 h poststimulation). The most optimal timepoint for each trigger and macrophage subtype can be determined experimentally using population-level assays, such as LDH release or PI incorporation. 8. For ASC cross-linking, harvesting the dying and non-adherent cells by centrifugation and combining this with the cell lysate ensures that all polymerized ASC is collected and analyzed. 9. For ELISA, any samples that fall below the lowest standard, here 15.6 pg/mL, are outside the range of detection of the assay and should be treated as such.
Acknowledgments All schematics were created with BioRender. K.S. is a recipient of SNSF Postdoc. Mobility fellowship (P500PB_211096). R.M. is supported by the Alzheimer Forschung Initiative (#20043). E.L. is supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy (EXC2151—390873048) and Helmholtz Association, under the project title “Immunology&Inflammation,” (#ZT0027). References 1. Latz E, Xiao TS, Stutz A (2013) Activation and regulation of the inflammasomes. Nat Rev Immunol 13:397–411 2. Broz P, Pelegrı´n P, Shao F (2020) The gasdermins, a protein family executing cell death and inflammation. Nat Rev Immunol 20:143–157 3. Chan AH, Schroder K (2019) Inflammasome signaling and regulation of interleukin-1 family cytokines. J Exp Med 217:e20190314 4. Downs KP, Nguyen H, Dorfleutner A et al (2020) An overview of the non-canonical inflammasome. Mol Asp Med 76:100924 5. Kanneganti T (2015) The inflammasome: firing up innate immunity. Immunol Rev 265:1– 5 6. Artlett CM (2013) Inflammasomes in wound healing and fibrosis. J Pathol 229:157–167 7. Bortolotti P, Faure E, Kipnis E (2018) Inflammasomes in tissue damages and immune disorders after trauma. Front Immunol 9:1900
8. Li Y, Huang H, Liu B et al (2021) Inflammasomes as therapeutic targets in human diseases. Signal Transduct Target Ther 6:247 9. Mangan MSJ, Olhava EJ, Roush WR et al (2018) Targeting the NLRP3 inflammasome in inflammatory diseases. Nat Rev Drug Discov 17:588–606 10. McManus RM, Heneka MT (2017) Role of neuroinflammation in neurodegeneration: new insights. Alzheimers Res Ther 9:14 11. Beilharz M, Nardo DD, Latz E et al (2016) NLR proteins, methods and protocols. Methods Mol Biol 1417:145–158 12. Tzeng T-C, Schattgen S, Monks B et al (2016) A fluorescent reporter mouse for inflammasome assembly demonstrates an important role for cell-bound and free ASC specks during in vivo infection. Cell Rep 16:571–582
Chapter 30 Detection of G-Quadruplex DNA Structures in Macrophages Melanie Kastl, Fabian Hersperger, Katrin Kierdorf, and Katrin Paeschke Abstract In addition to the canonical B-DNA conformation, DNA can fold into different secondary structures. Among them are G-quadruplex structures (G4s). G4 structures are very stable and can fold in specific guanine-rich regions in DNA and RNA. Different in silico, in vitro, and in cellulo experiments have shown that G4 structures form so far in all tested organisms. There are over 700,000 predicted G4s in higher eukaryotes, but it is so far assumed that not all will form at the same time. Their formation is dynamically regulated by proteins and is cell type-specific and even changes during the cell cycle or during different exogenous or endogenous stimuli (e.g., infection or developmental stages) can alter the G4 level. G4s have been shown to accumulate in cancer cells where they contribute to gene expression changes and the mutagenic burden of the tumor. Specific targeting of G4 structures to impact the expression of oncogenes is currently discussed as an anti-cancer treatment. In a tumor microenvironment, not only the tumor cells will be targeted by G4 stabilization but also immune cells such as macrophages. Although G4s were detected in multiple organisms and different cell types, only little is known about their role in immune cells. Here, we provide a detailed protocol to detect G4 formation in the nucleus of macrophages of vertebrates and invertebrates by microscopic imaging. Key words DNA, G-quadruplex structures, Macrophages, Immunofluorescence microscopy
1
Introduction DNA does not only exist as a canonical B-form double helix, but can fold into several inter- and intramolecular structures, among those are G-quadruplexes (G4s) [1]. G4s can form within specific guanine-rich regions in DNA and RNA, harboring so-called G4 motifs (consensus G4 motif: GxN1–7Gx-N1–7Gx-N1–7 Gx-N1–7 [2]). Guanines interact via Hoogsteen hydrogen bonding and form G quartets. Stacking of these G quartets, which are further stabilized by a monovalent cation, leads to the formation of a G4 [3]. Different high-throughput sequencing methods revealed over 700,000 potential sites in the genome of higher eukaryotes where G4s could form [4]. The location of G4 motifs is not random. They are overrepresented at telomeres,
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_30, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
453
454
Melanie Kastl et al.
promoters, and transcription factor binding sites [5]. The location is also conserved across different species further highlighting their functional potential [6]. The dynamic folding and unfolding of G4s are regulated by G4-interacting proteins that are either binding (e.g., Zuo1 [7]), stabilizing, or unwinding G4 structures (e.g., helicases like Pif1, WRN, BLM [8–10]). Mutations in these helicases can directly be linked to several diseases [11]. Certain autosomal recessive diseases share mutations in the RecQ helicase that plays an important role in DNA repair. For example, the Werner syndrome that leads to premature aging is caused by mutations in the WRN gene, which encodes for a protein that resembles the RecQ helicases with its central domain [12, 13]. A large number of predicted G4s are located in oncogenes [14]. G4 formation was shown to repress their expression such as transcriptional silencing of c-MYC [15–18]. c-MYC overexpression is linked to several cancer types and high expression levels can be connected to a poor therapeutic prognosis [19]. Based on these findings, G4 formation is currently discussed as a potential new drug target for therapeutic approaches against cancer cells [20]. G4 targeting approaches have already been successful in antiviral therapies targeting their replication [21, 22]. Most viruses exhibit G4-forming sequences [23]. G4 targeting is possible by the application of G4-specific ligands that bind and stabilize G4s (https://g4ldb.com) [24]. A well-known G4-specific ligand is pyridostatin (PDS), which exhibits a high selectivity toward G4 structures in vitro and in vivo [4]. Most bacteria also have putative G4 motifs in their promoter regions [25]. G4 stabilization in Salmonella has been shown to reduce bacterial growth and virulence [26]. However, G4 ligands can cause multiple changes within the targeted cells, as they lead to gene expression changes, replication defects, telomere length changes, and increased genome instability [27–34]. It was observed that G4 stabilization can induce interferon (IFN) β signaling via the formation of micronuclei in cancer cells [35]. This gave rise to the speculation that genome instability events triggered by G4s may lead to a type I IFN response. If G4s form in immune cells and which consequences G4 formation in these cells cause, was not determined, yet. Since modulation of G4s is a promising anticancer therapy [36] and immune cells also can promote or inhibit tumor growth [37], G4 formation in these cells needs to be studied. In particular, tumor-associated macrophages (TAMs) are discussed as connection between inflammation and cancer [38, 39]. Since multiple gene expression changes occur during the activation/ polarization in macrophages [40], it is important to understand if G4s form in these immune cells and which impact these G4s have on the macrophage activation and function. Here, we describe a fast, robust, and feasible method to check for G4 DNA structures in macrophages. We have established a
G-Quadruplex Detection in Macrophages
455
Fig. 1 Staining of G4s in cells using BG4. (a) Schematic overview of the staining. The single-chain antibody BG4 (gray) binds to the G-quadruplex. BG4 contains a FLAG sequence, so the anti-FLAG antibody (dark blue) can bind and introduce a fluorescent Alexa Fluor 488 (green). See for example of G4 stainings in different macrophages (b–d). (b) Labeling of human blood macrophages (c), iBMDMs, and (d) Drosophila S2 cells. Scale bar is 10 μm
robust staining protocol that detects G4 formation using a wellestablished G4-specific antibody (BG4 [41]) in different macrophage cell systems (Fig. 1). This method is applicable for different macrophage types, including macrophages from vertebrate and invertebrates, and allows to screen for changes in G4 levels upon different conditions.
2
Materials
2.1 General Equipment
1. 12-mm autoclaved coverslips in a 24-well plate. 2. Agitator (for incubation of the 24-well plate). 3. Tweezers. 4. Fluorescence microscope.
456
2.2
Melanie Kastl et al.
Reagents
1. 1x PBS. 2. Fix solution: methanol/ acetic acid 3:1 (v/v), prepare fresh. 3. Permeabilization solution: 0.1% Triton X-100 in 1x PBS. 4. Blocking buffer: 2% non-fat dry milk in 1x PBS. 5. Wash solution: 0.1% Tween-20 in 1x PBS. 6. Growth medium for macrophage population of interest. 7. Optional: poly-D-lysin. 8. BG4 single-chain antibody (self-purified antibody, see [41] and Notes 3–5). 9. Rabbit anti-Flag antibody. 10. Goat anti-rabbit Alexa Fluor 488 antibody. 11. Aqueous mounting medium with DAPI.
3 3.1
Methods G4 Staining
1. The macrophages should be seeded on the coverslips the day before the start of the staining protocol (seeNotes 1 and 2). We recommend to start with 7 × 104 cells per cover slide. The cover slides should be placed into a 24-well plate and be covered with growth medium. 2. (Optional) The cells can be treated with a G4 stabilizer like PDS in the morning before the staining. The treatment with a G4 stabilizer will increase the G4 levels and can be used as a positive control and serve as a reference when comparing several treatments, genotypes, or conditions. Alternatively, the macrophage population of interest can also be treated with a G4 destabilizer such as PhPC, which can serve as a negative control [42]. 3. Remove the growth medium, replace it with 250 μL fresh growth medium and 250 μL fix solution (ratio 1:1), and incubate at RT for 5 min. 4. Remove the growth medium/ fix solution mix and add 500 μL fix solution. Remove the fix solution, add 500 μL fix solution, and incubate at RT for 10 min. 5. Wash once with 1x PBS (pipet on–pipet off). 6. Add 500 μL permeabilization solution and incubate at RT for 3 min. 7. Perform 3 washes with 1x PBS at RT for 5 min each and place the plate on an agitator/ nutator (50 rpm) during the wash steps. At this point, the samples can be stored ON at 4 °C. The wells should contain enough PBS to avoid drying out.
G-Quadruplex Detection in Macrophages
457
8. Add 500 μL blocking buffer and incubate for 1 h at RT with agitation (for all steps 50 rpm). 9. Incubate with BG4 (2 μg in 50 μL of blocking buffer) for 2 h at RT with agitation (50 rpm). For details on BG4 purification and validation, seeNotes 3–5. 10. Wash 3 times with wash solution for 10 min each with agitation. 11. Incubate with rabbit anti-Flag antibody (1:800 in blocking buffer) for 1 h with agitation. 12. Wash 3 times with wash solution for 10 min, each with agitation. 13. Incubate with goat anti-rabbit Alexa Fluor 488 (1:1000 in blocking buffer) for 1 h with agitation (Fig. 1a for cartoon on all antibodies used here). The plate should be covered (e.g., aluminum foil) to protect the samples from light. 14. Wash 3 times with wash solution for 10 min each with agitation. 15. Prepare a slide with a small droplet of mounting medium with DAPI, take the coverslips out by using tweezers, and mount the cell-side onto the mounting medium. Let the slide dry overnight (light-protected) at RT and store it afterward at 4 °C. 16. Image the slides (DAPI and Alexa Fluor 488) with a fluorescence microscope and acquire at least 100 nuclei per slide/ condition for quantification (seeNote 6 for sample stainings). 3.2
Quantification
For the quantification of the nuclear G4 levels, we use ImageJ (https://imagej.nih.gov/ij/download.html). Make sure to use the original captured, unmodified file, and open the picture by drag and drop in ImageJ. Convert the stack to two images (DAPI and BG4 channel) by clicking on “Image” ! “Stacks” ! “Stacks to Images” and adjust the threshold in the DAPI picture by selecting the DAPI picture and clicking on “Image” ! “Adjust” ! “Auto Threshold” and select the method that fits best to your settings (e.g., “Default”). Select measurements (“Analyze” ! “Set Measurements”) and direct the measurement to the BG4-stained picture (“Redirect to”), analyze the particles (“Analyze” ! “Analyze Particles”), and set the size from 30 to infinity (unspecific small dots will be excluded by this). If you select “Show Outlines,” you will mark the measured spots and the results are displayed. The “mean” is the integrated density divided by the area (nucleus). The intensity of the BG4 staining in the nucleus is normalized to the area of the nucleus by doing so. Data can be displayed as mean fluorescent intensity or as fold change compared to untreated.
458
Melanie Kastl et al.
3.3 Alternative Methods to Detect G4 Signal
4
Instead of using microscopy, it is also possible to check the G4 levels by FACS. A suitable method was recently published by our lab called BG-flow [43]. This technique is a good alternative to study non-adherent cells. The main difference is that the fixation and staining are performed in tubes and the samples are centrifuged between the different steps.
Notes 1. Cell preparation: different conditions are known to impact the G4 levels in cells, such as stress and damage [44, 45]. Prior to the experiment, make sure that with your experimental conditions the macrophages are viable. Viability and metabolic functions of macrophages can be tested by standard FACS approaches using a second cell sample or other approaches such as an MTT assay or alamarBlue stain [46, 47]. 2. Selection of the macrophage type: take into consideration if your selected macrophage population is adherent or non-adherent. For adherent cells, seed the cells 24 h prior to the start of the staining. For adherent cells, the experimental conditions are optimized so that cells will directly grow on the microscopic cover slides. For non-adherent cells, cover slides need to be pre-treated with 50 μg/mL poly-D-lysin (Sigma A-003-M). Poly-D-lysin is a synthesized extracellular matrix molecule that promotes cell adhesion and allows the attachment of the macrophages to the microscopic slides. 3. BG4 is a single-chain antibody that specifically detects G4 DNA and RNA structures. The antibody can be used for immunofluorescence, FACS, and chromatin immunoprecipitation experiments [43, 48, 49]. Although the antibody is commercially available, we recommend purifying and validating the antibody prior to the experiment (see below) and do not keep the aliquots after purification longer than 6–12 months at 80 °C. 4. BG4 purification: the plasmid which is used to express BG4 in bacteria is commercially available (pSANG10-3F-BG4; Addgene Plasmid No. #55756). We recommend to purify BG4 according to published protocols [50, 51]. Briefly BL21 pLysS competent cells containing the BG4 plasmid are grown in 2xTY medium (1.6% bacto tryptone, 1% bacto yeast extract, 0.5% NaCl) with 50 mg/mL kanamycin and 1% glucose. After expanding the overnight culture to 1 liter (l), BG4 expression was induced with 0.5 mM IPTG (isopropyl-β-d-thiogalactopyranoside) addition and incubation overnight at 25 °C. The cells are lysed in 100 mL TES buffer (50 mM Tris–Cl pH 8.0, 1 mM EDTA, 20% sucrose) and diluted by the addition of
G-Quadruplex Detection in Macrophages
459
120 mL TES buffer prior to centrifugation at 8000 g at 4 °C for 20 min. The supernatant is filtered (0.45 μM) and BG4 is purified using a Ni-NTA Sepharose column (GR Healthcare). BG4 is eluated with 15 mL elution buffer (PBS pH 8 and 250 mM imidazole), concentrated and the buffer is exchanged to inner cell salt buffer (25 mM Hepes (pH 7.6), 110 mM KCl, 10.5 mM NaCl, 1 mM MgCl2) using an Amicon Ultra-15 Centrifugal Filter Unit with 10 kDa cutoff (Millipore). The purified BG4 concentration is measured and stored at -80 °C. 5. BG4 validation: After every BG4 batch purification, it is mandatory to check the quality of the purified protein by Coomassie (a clear and strong band should be visible at 30 kDa). Furthermore, it is recommended to control the specificity of BG4 to G4 structures by biochemical methods (for example, the affinity of BG4 to G4s should be in the 20–200 nM range). For this, we monitor binding of BG4 to G4 structures in comparison with G-rich ssDNA by ELISA as described here [41, 52, 53]. 6. We have performed this staining protocol for three different macrophage cell systems in vitro: (A) an immortalized mouse macrophage cell line (iBMDMs), [54] (B) human macrophages (monocytes from the blood were differentiated into macrophages), and (C) S2 cells, a Drosophilahemocyte cell line. The protocol can be easily applied to similar macrophage in vitro systems such as RAW 264.7 cells, an adherent cell line, or monocyte-derived macrophages isolated and differentiated from blood monocytes (CD14+ positive). If you use non-adherent cells like invertebrate S2 cells, we recommend to pre-treat the coverslips with poly-D-lysin. Additionally, we recommend to seed the cells the evening before or at least 4 h prior to fixation, to avoid changes in G4 levels due to cell stress.
Acknowledgments We thank Stefan Juranek for careful reading of the manuscript. Research in the Paeschke laboratory is funded by the Fritz Thyssen Foundation and Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—“Project-ID 369799452— TRR237” as well as by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy—EXC2151—390873048. The Kierdorf lab is supported by project grants of the Fritz Thyssen Foundation and of the DFG. The Kierdorf lab is further supported by the DFG through project grants within SFB/TRR167 (Project ID 259373024), CRC1479 (Project ID 441891347), and within Germany’s Excellence Strategy (grant no. CIBSS—EXC-2189, Project ID 390939984).
460
Melanie Kastl et al.
References 1. Bochman ML, Paeschke K, Zakian VA (2012) DNA secondary structures: stability and function of G-quadruplex structures. Nat Rev Genet 13(11):770–780. https://doi.org/10. 1038/nrg3296 2. Huppert JL, Balasubramanian S (2005) Prevalence of quadruplexes in the human genome. Nucleic Acids Res 33(9):2908–2916. https:// doi.org/10.1093/nar/gki609. PMID 15914667 3. Gellert M, Lipsett MN, Davies DR (1962) Helix formation by guanylic acid. Proc Natl Acad Sci U S A 48:2013–2018. https://doi. org/10.1073/pnas.48.12.2013 4. Chambers VS, Marsico G, Boutell JM, Antonio MD, Smith GP, Balasubramanian S (2015) High-throughput sequencing of DNA G-quadruplex structures in the human genome. Nat Biotechnol 33(8):877–881. https://doi.org/10.1038/nbt.3295. PMID 26192317 5. Rhodes D, Lipps HJ (2015) G-quadruplexes and their regulatory roles in biology. Nucleic Acids Res 43(18):8627–8637. https://doi. org/10.1093/nar/gkv862. PMID 26350216 6. Capra JA, Paeschke K, Singh M, Zakian VA (2010) G-Quadruplex DNA sequences are evolutionarily conserved and associated with distinct genomic features in Saccharomyces cerevisiae. PLoS Comput Biol 6(7): e1000861. https://doi.org/10.1371/journal. pcbi.1000861. PMID - 20676380 7. De Magis A, Gotz S, Hajikazemi M, FeketeSzucs E, Caterino M, Juranek S, Paeschke K (2020) Zuo1 supports G4 structure formation and directs repair toward nucleotide excision repair. Nat Commun 11(1):3907. https://doi. org/10.1038/s41467-020-17701-8 8. Paeschke K, Bochman ML, Garcia PD, Cejka P, Friedman KL, Kowalczykowski SC, Zakian VA (2013) Pif1 family helicases suppress genome instability at G-quadruplex motifs. Nature 497(7450):458–462. https://doi.org/10. 1038/nature12149. PMID - 23657261 9. Sun H, Karow JK, Hickson ID, Maizels N (1998) The Bloom’s syndrome helicase unwinds G4 DNA. J Biol Chem 273(42): 27587–27592. https://doi.org/10.1074/jbc. 273.42.27587 10. Huber MD, Duquette ML, Shiels JC, Maizels N (2006) A conserved G4 DNA binding domain in RecQ family helicases. J Mol Biol 358(4):1071–1080. https://doi.org/10. 1016/j.jmb.2006.01.077
11. Maizels N (2015) G4-associated human diseases. EMBO Rep 16(8):910–922. https:// doi.org/10.15252/embr.201540607 12. Wu Y, Brosh RM Jr (2010) G-quadruplex nucleic acids and human disease. FEBS J 277(17):3470–3488. https://doi.org/10. 1111/j.1742-4658.2010.07760.x 13. Brosh RM Jr, Bohr VA (2007) Human premature aging, DNA repair and RecQ helicases. Nucleic Acids Res 35(22):7527–7544. https://doi.org/10.1093/nar/gkm1008 14. Eddy J, Maizels N (2009) Selection for the G4 DNA motif at the 5′ end of human genes. Mol Carcinog 48(4):319–325. https://doi.org/10. 1002/mc.20496 15. Siddiqui-Jain A, Grand CL, Bearss DJ, Hurley LH (2002) Direct evidence for a G-quadruplex in a promoter region and its targeting with a small molecule to repress c-MYC transcription. Proc Natl Acad Sci U S A 99(18): 11593–11598. https://doi.org/10.1073/ pnas.182256799 16. Hurley LH, Von Hoff DD, Siddiqui-Jain A, Yang D (2006) Drug targeting of the c-MYC promoter to repress gene expression via a G-quadruplex silencer element. Semin Oncol 33(4):498–512. https://doi.org/10.1053/j. seminoncol.2006.04.012 17. Yang D, Hurley LH (2006) Structure of the biologically relevant G-quadruplex in the c-MYC promoter. Nucleosides Nucleotides Nucleic Acids 25(8):951–968. https://doi. org/10.1080/15257770600809913 18. Ou TM, Lu YJ, Zhang C, Huang ZS, Wang XD, Tan JH, Chen Y, Ma DL, Wong KY, Tang JC, Chan AS, Gu LQ (2007) Stabilization of G-quadruplex DNA and down-regulation of oncogene c-myc by quindoline derivatives. J Med Chem 50(7):1465–1474. https://doi. org/10.1021/jm0610088 19. Mathad RI, Hatzakis E, Dai J, Yang D (2011) c-MYC promoter G-quadruplex formed at the 5′-end of NHE III1 element: insights into biological relevance and parallel-stranded G-quadruplex stability. Nucleic Acids Res 39(20):9023–9033. https://doi.org/10. 1093/nar/gkr612 20. Awadasseid A, Ma X, Wu Y, Zhang W (2021) G-quadruplex stabilization via small-molecules as a potential anti-cancer strategy. Biomed Pharmacother 139:111550. https://doi.org/ 10.1016/j.biopha.2021.111550 21. Perrone R, Butovskaya E, Daelemans D, Palu G, Pannecouque C, Richter SN (2014) Anti-HIV-1 activity of the G-quadruplex
G-Quadruplex Detection in Macrophages ligand BRACO-19. J Antimicrob Chemother 69(12):3248–3258. https://doi.org/10. 1093/jac/dku280 22. Frasson I, Nadai M, Richter SN (2019) Conserved G-Quadruplexes regulate the immediate early promoters of human alphaherpesviruses. Molecules 24(13). h t t p s : // d o i . o r g / 1 0 . 3 3 9 0 / molecules24132375 23. Lavezzo E, Berselli M, Frasson I, Perrone R, Palu G, Brazzale AR, Richter SN, Toppo S (2018) G-quadruplex forming sequences in the genome of all known human viruses: a comprehensive guide. PLoS Comput Biol 14(12):e1006675. https://doi.org/10.1371/ journal.pcbi.1006675 24. Santos T, Salgado GF, Cabrita EJ, Cruz C (2021) G-Quadruplexes and their ligands: biophysical methods to unravel G-Quadruplex/ ligand interactions. Pharmaceuticals (Basel) 1 4 ( 8 ) . h t t p s : // d o i . o r g / 1 0 . 3 3 9 0 / ph14080769 25. Yadav P, Kim N, Kumari M, Verma S, Sharma TK, Yadav V, Kumar A (2021) G-Quadruplex structures in bacteria: biological relevance and potential as an antimicrobial target. J Bacteriol 203(13):e0057720. https://doi.org/10. 1128/JB.00577-20 26. Jain N, Mishra SK, Shankar U, Jaiswal A, Sharma TK, Kodgire P, Kumar A (2020) G-quadruplex stabilization in the ions and maltose transporters gene inhibit Salmonella enterica growth and virulence. Genomics 112(6):4863–4874. https://doi.org/10. 1016/j.ygeno.2020.09.010 27. Kim N (2019) The interplay between G-quadruplex and transcription. Curr Med Chem 26(16):2898–2917. https://doi.org/ 10.2174/0929867325666171229132619 28. David AP, Pipier A, Pascutti F, Binolfi A, Weiner AMJ, Challier E, Heckel S, Calsou P, Gomez D, Calcaterra NB, Armas P (2019) CNBP controls transcription by unfolding DNA G-quadruplex structures. Nucleic Acids Res 47(15):7901–7913. https://doi.org/10. 1093/nar/gkz527 29. Bryan TM (2019) Mechanisms of DNA replication and repair: insights from the study of G-Quadruplexes. Molecules 24(19). https:// doi.org/10.3390/molecules24193439 30. Tan J, Lan L (2020) The DNA secondary structures at telomeres and genome instability. Cell Biosci 10:47. https://doi.org/10.1186/ s13578-020-00409-z 31. Bryan TM (2020) G-Quadruplexes at telomeres: friend or foe? Molecules 25(16).
461
h t t p s : // d o i . o r g / 1 0 . 3 3 9 0 / molecules25163686 32. Paeschke K, Burkovics P (2021) Mgs1 function at G-quadruplex structures during DNA replication. Curr Genet 67(2):225–230. https:// doi.org/10.1007/s00294-020-01128-1. PMID - 33237336 33. Wang Y, Yang J, Wild AT, Wu WH, Shah R, Danussi C, Riggins GJ, Kannan K, Sulman EP, Chan TA, Huse JT (2019) G-quadruplex DNA drives genomic instability and represents a targetable molecular abnormality in ATRXdeficient malignant glioma. Nat Commun 10(1):943. https://doi.org/10.1038/ s41467-019-08905-8 34. De Magis A, Manzo SG, Russo M, Marinello J, Morigi R, Sordet O, Capranico G (2019) DNA damage and genome instability by G-quadruplex ligands are mediated by R loops in human cancer cells. Proc Natl Acad Sci U S A 116(3):816–825. https://doi.org/ 10.1073/pnas.1810409116 35. Miglietta G, Russo M, Duardo RC, Capranico G (2021) G-quadruplex binders as cytostatic modulators of innate immune genes in cancer cells. Nucleic Acids Res 49(12):6673–6686. https://doi.org/10.1093/nar/gkab500 36. Kosiol N, Juranek S, Brossart P, Heine A, Paeschke K (2021) G-quadruplexes: a promising target for cancer therapy. Mol Cancer 20(1):40. https://doi.org/10.1186/ s12943-021-01328-4 37. Zamarron BF, Chen W (2011) Dual roles of immune cells and their factors in cancer development and progression. Int J Biol Sci 7(5): 651–658. https://doi.org/10.7150/ijbs. 7.651 38. Coussens LM, Zitvogel L, Palucka AK (2013) Neutralizing tumor-promoting chronic inflammation: a magic bullet? Science 339(6117): 286–291. https://doi.org/10.1126/science. 1232227 39. Mantovani A, Allavena P, Marchesi F, Garlanda C (2022) Macrophages as tools and targets in cancer therapy. Nat Rev Drug Discov 21(11): 799–820. https://doi.org/10.1038/s41573022-00520-5 40. Palma A, Jarrah AS, Tieri P, Cesareni G, Castiglione F (2018) Gene regulatory network modeling of macrophage differentiation corroborates the continuum hypothesis of polarization states. Front Physiol 9:1659. https://doi. org/10.3389/fphys.2018.01659 41. Biffi G, Tannahill D, McCafferty J, Balasubramanian S (2013) Quantitative visualization of DNA G-quadruplex structures in human cells.
462
Melanie Kastl et al.
Nat Chem 5(3):182–186. https://doi.org/10. 1038/nchem.1548 42. Mitteaux J, Lejault P, Wojciechowski F, Joubert A, Boudon J, Desbois N, Gros CP, Hudson RHE, Boule JB, Granzhan A, Monchaud D (2021) Identifying G-quadruplexDNA-disrupting small molecules. J Am Chem Soc 143(32):12567–12577. https://doi.org/ 10.1021/jacs.1c04426 43. De Magis A, Kastl M, Brossart P, Heine A, Paeschke K (2021) BG-flow, a new flow cytometry tool for G-quadruplex quantification in fixed cells. BMC Biol 19(1):45. https://doi. org/10.1186/s12915-021-00986-6 44. Yin L, Wei J, Lu Z, Huang S, Gao H, Chen J, Guo F, Tu M, Xiao B, Xi C, Zhang K, Li Q, Wu J, Gao W, Jiang K, Yu J, Miao Y (2022) Prevalence of germline sequence variations among patients with pancreatic cancer in China. JAMA Netw Open 5(2):e2148721. https://doi.org/10.1001/jamanetworkopen. 2021.48721 45. Linke R, Limmer M, Juranek SA, Heine A, Paeschke K (2021) The relevance of G-Quadruplexes for DNA repair. Int J Mol Sci 22(22). https://doi.org/10.3390/ ijms222212599 46. Kummrow A, Frankowski M, Bock N, Werner C, Dziekan T, Neukammer J (2013) Quantitative assessment of cell viability based on flow cytometry and microscopy. Cytometry A 83(2):197–204. https://doi.org/10.1002/ cyto.a.22213 47. Riss TL, Moravec RA, Niles AL, Duellman S, Benink HA, Worzella TJ, Minor L (2004) Cell viability assays. In: Markossian S, Grossman A, Brimacombe K et al (eds) Assay guidance manual. Eli Lilly & Company and the National Center for Advancing Translational Sciences, Bethesda 48. Biffi G, Tannahill D, McCafferty J, Balasubramanian S (2013) Quantitative visualization of DNA G-quadruplex structures in human cells. Nat Chem 5(3):182–186. https://doi.org/10. 1038/nchem.1548. PMID - 23422559
49. H€ansel-Hertsch R, Antonio MD, Balasubramanian S (2017) DNA G-quadruplexes in the human genome: detection, functions and therapeutic potential. Nat Rev Mol Cell Biol 18(5): 279–284. https://doi.org/10.1038/nrm. 2017.3. PMID - 28225080 50. Biffi G, Antonio MD, Tannahill D, Balasubramanian S (2013) Visualization and selective chemical targeting of RNA G-quadruplex structures in the cytoplasm of human cells. Nat Chem 6(1):75–80. https://doi.org/10. 1038/nchem.1805. PMID - 24345950 51. Das K, Srivastava M, Raghavan SC (2016) GNG motifs can replace a GGG stretch during G-quadruplex formation in a context dependent manner. PLoS One 11(7):e0158794. https://doi.org/10.1371/journal.pone. 0158794 52. Liu HY, Zhao Q, Zhang TP, Wu Y, Xiong YX, Wang SK, Ge YL, He JH, Lv P, Ou TM, Tan JH, Li D, Gu LQ, Ren J, Zhao Y, Huang ZS (2016) Conformation selective antibody enables genome profiling and leads to discovery of parallel G-quadruplex in human telomeres. Cell Chem Biol 23(10):1261–1270. https://doi.org/10.1016/j.chembiol.2016. 08.013 53. Mao SQ, Ghanbarian AT, Spiegel J, Martinez Cuesta S, Beraldi D, Di Antonio M, Marsico G, Hansel-Hertsch R, Tannahill D, Balasubramanian S (2018) DNA G-quadruplex structures mold the DNA methylome. Nat Struct Mol Biol 25(10):951–957. https://doi.org/10. 1038/s41594-018-0131-8 54. Abdullah Z, Schlee M, Roth S, Mraheil MA, Barchet W, Bottcher J, Hain T, Geiger S, Hayakawa Y, Fritz JH, Civril F, Hopfner KP, Kurts C, Ruland J, Hartmann G, Chakraborty T, Knolle PA (2012) RIG-I detects infection with live Listeria by sensing secreted bacterial nucleic acids. EMBO J 31(21):4153–4164. https://doi.org/10. 1038/emboj.2012.274
Chapter 31 Adaptation of Human iPSC-Derived Macrophages Toward an Alveolar Macrophage-Like Phenotype Post-Intra-Pulmonary Transfer into Murine Models Miriam Hetzel, Ingrid Gensch, Mania Ackermann, and Nico Lachmann Abstract Alveolar macrophages (AMs) represent crucial immune cells in the bronchioalveolar space of the lung. Given the important role in the host defense machinery and lung tissue homeostasis, AMs have been linked to a variety of diseases and thus represent a promising target cell type for novel therapies. The emerging importance of AM underlines the necessity to isolate and/or generate proper cellular models, which facilitate basic biology and translational science. As of yet, most studies focus on the derivation of AM from the murine system. This chapter introduces the use of human-induced pluripotent stem cell (iPSC)derived primitive macrophages, which can be further matured towards an AM-like phenotype upon intrapulmonary transfer into mice. We will give a brief overview on the generation of primitive iPSC-derived macrophages, which is followed by a detailed, step-by-step description of the intra-pulmonary transfer of cells and the follow-up procedures needed to isolate the iPSC-derived, AM-like cells from the lungs posttransfer. The chapter provides an alternative approach to derive human AM-like cells, which can be used to study human AM biology and to investigate novel therapeutic interventions using primitive macrophages from iPSC. Key words Induced pluripotent stem cells, iPSC, Macrophages, Alveolar macrophages, Hematopoietic differentiation, Lung, Intrapulmonary transfer, PMT
1
Introduction Alveolar macrophages (AMs) represent a unique subset of tissueresident macrophages (TRMs) residing in the bronchioalveolar space. Like other TRM subsets, AMs are highly specialized and adapted to their niche, in which they fulfill tissue-specific functions. In fact, AMs are crucial and important cellular mediators of the initial host defense machinery, but they also clear surfactant material, thereby ensuring proper lung function. Thus, it does not come as a surprise that a variety of diseases are associated with the impairment and dysfunction of AMs [1, 2]. AMs are classified as
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_31, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
463
464
Miriam Hetzel et al.
immune scavengers in the lung, which can recognize and phagocytose pathogenic and non-pathogenic intruders that are abundantly inhaled with every breath. Besides their eminent role in clearing these intruders, AMs, under steady-state conditions, are finely balancing the immune cell response throughout the lung, preventing constant inflammatory conditions. Consequently, AMs are highly versatile cells and can be envisioned in a variety of applications. Not only modern immune cell therapy approaches, which aim to adoptively transfer monocyte/ macrophages into the lung to improve pulmonary infections or congenital lung disorders, are in the current focus. Also, a broad range of non-therapeutic applications, which aim to study host– pathogen interactions or to increase our knowledge in the pathophysiology of certain genetic diseases, are on the horizon. To meet this increasing demand for scientific and therapeutic applications, models to directly derive de novo AMs or systems, which have the capacity to generate AMs using in vitro or in vivo techniques are highly desirable. In the murine system, the generation of AMs in vitro seems to be a feasible approach in which fetal liver monocytes can be cultured in the presence of GM-CSF, which leads to the differentiation towards homogenous AMs with therapeutic activity following intra-pulmonary transfer [3, 4]. However, as of yet, similar studies in the human system could not reveal such a straightforward approach. While some efforts are directed to genetically manipulate macrophages by knock-out approaches for, e.g., MafB or c-Maf, the use of primitive monocytes/macrophages with the ability to adopt towards an AM-like phenotype, or the use of systems, which are able to mimic a lung microenvironment, would be highly suited to derive human AMs [5]. Such systems could, for example, rely on the use of TGF-β and GM-CSF, which would induce the expression of EGR2, known to be a crucial factor for the cells to acquire a core AM phenotype [6]. One alternative approach could also be the generation of primitive monocytes/macrophages, which have the plasticity to become an AM-like cell following intra-pulmonary transfer. One of the best-suited examples to derive such cells is pluripotent stem cells (e.g., induced pluripotent stem cells; iPSC). In fact, the derivation of primitive macrophages from iPSC has been proven successful [7] and the use of these cells to derive a plethora of different TRMs such as microglia, Kupffer cells, or AMs has been presented [8–11]. In addition, studies also revealed a MYB-independent origin of iPSC-derived macrophages and the developmental traits of human macrophages could partially be recapitulated using the iPSC system [5, 7]. Thus, human iPSCs are a very elegant tool for the generation of AMs and other TRMs. Along this line, previous studies pointed out that iPSC-derived macrophages can indeed adopt an AM-like phenotype [11] and can be produced in scalable quantities [12, 13]. Especially the latter
Generation of Alveolar Macrophages from Human iPSC
465
is of interest, as these systems would allow for the generation of human macrophages in sufficient quantities for industrial and therapeutic applications. In this chapter, we aim to provide a detailed overview of the derivation of human AMs using human iPSC-derived macrophages, which are able to adopt an AM-like phenotype post-intrapulmonary transfer in immunodeficient mice. The chapter is divided into three parts, presenting the (i) preparation of iPSCderived macrophages, (ii) the step-by-step administration of cells into the lung, and (iii) the detailed re-isolation of cells and proper analysis of cells post intra-pulmonary transfer. Using this information, the reader will gain the know-how to generate human AM-like cells, which can be applied in therapeutic and non-therapeutic settings.
2
Materials
2.1 Instruments and Equipment
1. 37 °C water bath. 2. 37 °C; 5% CO2-humidified incubator. 3. Cell culture cabinet/laminar flow. 4. Benchtop centrifuge. 5. Flow cytometer with 3 or 4 laser configuration (e.g., 405 nm, 488 nm, (561 nm) and 638 nm excitation wavelength). 6. Inverted fluorescence (with 560 excitation wavelength) and phase-contrast microscope. 7. Orbital Shaker (e.g., Celltron, Infors). 8. Refrigerator (4 °C) or ice bath/bucket. 9. Scale. 10. Bend, blunt, anatomical forceps. 11. Cold light lamp. 12. Forceps, anatomical. 13. Heating mat (e.g., ThermoLux). 14. Hemocytometer, Neubauer improved. 15. Intubation stand (Fig. 1). 16. Magnifying glasses. 17. Pliers or nail clipper. 18. Scale and beaker. 19. Scissors. 20. Spatula (width 3 mm) with rounded edges. 21. Styrofoam board. 22. 1 mL syringes. 23. 1.5 mL reaction tubes.
466
Miriam Hetzel et al.
Fig. 1 Intubation stand. Example for an intubation stand. Thread or rubber band can be stretched between the metal knobs on the top. This stand was produced in-house in our technical facility
24. 100- to 1000 μL, 20- to 200 μL, 2- to 20 μL, and 0.5- to 10 μL pipettes with tips. 25. 5 mL, 10 mL, and 25 mL serological pipettes and serological pipette controller. 26. 15 mL and 50 mL conical tube. 27. 20G/22G/24G peripheral intravenous catheter for intratracheal application (see Note 1). 28. 18 G needles for macrophage recovery (see Note 2). 29. 20 G peripheral intravenous catheter for macrophage recovery (see Note 3). 30. 27 G needles. 31. 6-well and 12-well tissue culture dishes. 32. 70 μm cell strainer (e.g., PluriSelect). 33. Adhesive tape (abdominal fixation). 34. Aluminum foil. 35. Glass Pasteur pipettes. 2.2 Reagents, Media, and Buffers
1. 70% ethanol. 2. Complete X-VIVO 15 medium: 2 mM L-glutamine, 1% (V/V) penicillin/streptomycin, 0.2% (V/V) β-mercaptoethanol in X-VIVO 15, or phenol-red free X-VIVO 15 medium (e.g., Lonza # BE02-060Q). Store at 4 °C for up to 1 month. 3. Recombinant human M-CSF and IL-3: 100 μg/mL stock solution, respectively: Reconstitute 100 μg lyophilized cytokine (M-CSF and IL-3; e.g., Peprotech #300-25; 200-03) in 1 mL 0.1% bovine serum albumin (BSA) solution in PBS. Store at -20 °C for up to 6 months.
Generation of Alveolar Macrophages from Human iPSC
467
4. Differentiation medium: 50 ng/mL human M-CSF, 25 ng/ mL human IL-3 in complete X-VIVO medium. Prepare only the volume of medium needed and add cytokines immediately before use. 5. Phosphate-buffered saline (1x PBS, pH 7.4): 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, KH2PO4 in 800 mL deionized water. Adjust pH with HCl. Add deionized water to 1 L and sterile-filter by using 0.22 μm Stericup vacuum filtration system. Store at 4 °C. 6. TrypanBlue. 7. MACS buffer: 0.4% (V/V) 0.5 M Na-EDTA and 2% (V/V) fetal bovine serum in PBS. Store at 4 °C. 8. Flow cytometry antibodies: For phenotyping of human iPSC-derived macrophages: human (h) hCD45 BV421 (clone: HI30; final conc.: 0.5 μg/ mL), hCD14 PE (clone 61D3; final conc.: 1 μg/mL), hCD163 APC (clone: eBioGHI/61; final conc. 1 μg/mL), and hCD11b Pe-Cy7 (clone: ICRF44; final conc.: 2 μg/mL). For characterization of the adaptation of iPSC-derived macrophages to AM-like phenotype: hCD45 PE-Cy7 (clone: HI30; final conc.: 0.25 μg/mL), hCD169 APC (clone: 7–239; final conc.: 1 μg/mL) and hCD11c eFluor450 (clone: 3.9; final conc.: 1 μg/mL), hCD206 PerCP/Cy5.5 (clone: 15–2; final conc.: 1 μg/mL). 9. Human Fc receptor blocking reagent for flow cytometry (e.g., Miltenyi Biotec # 130-059-901). 10. pHrodo™ Red E. coli BioParticles® Conjugate (e.g., Life Technologies # 15140122). 11. Ketamine/midazolam mixture: Mix 1 mL of ketamine (stock solution 100 mg/mL) with 5 mL midazolam (stock solution 1 mg/mL) and 4 mL sterile 0.9% NaCl. Mix well and apply 0.1 mL of the solution per 10 g body weight of the mouse (100 mg/1 mg/kg). 12. Atropine: Mix 1 mL atropine sulfate (stock solution 0.5 mg/ mL) with 4 mL sterile 0.9% NaCl and apply 10 μL of the solution per 10 g body weight of the mouse (0.1 mg/kg). 13. Eye ointment (e.g., Bepanthen eye and nose ointment). 14. Ketamine/xylazine: Mix 2 mL ketamine (stock solution 100 mg/mL) with 1 mL of xylazine (stock solution 20 mg/ mL). Mix well and apply 0.1 mL of the solution per 10 g body weight of the mouse (667 mg/6.67 mg/kg). 15. 0.5 M Sodium-ethylenediaminetetraacetic acid (Na-EDTA, pH 8.0): Weigh 73.1 g EDTA and mix with 400 mL water. Adjust pH to 8.0 with NaOH by adding NaOH pellets
468
Miriam Hetzel et al.
(~10 g). Make up to 500 mL with water. Dilute the stock 1: 1000 with PBS to obtain a 0.5 mM solution. Store at room temperature. 16. 4% Paraformaldehyde (PFA), pH 7.2: 1.33 M PFA in PBS. Dissolve overnight at 50 °C on a stirrer. Adjust pH with 10 N NaOH until solution is clear. Store aliquots at -20 °C. 17. Red cell lysis buffer: 155 mM NH4Cl, 46 mM KHCO3, 0.1% (V/V) 0.5 M Na-EDTA in pure water. Filter by using 0.22 μm Stericup vacuum filtration system. Store at room temperature. 18. Dissociation buffer: 10% (V/V) Collagenase D, 1% (V/V) DNase I in RPMI 1640 medium. Always prepare fresh. 19. Deoxyribonuclease I (DNase I): 10 mg/mL DNase I: Reconstitute 10 mg in 20 mM Tris–HCL pH 7.5 supplemented with 1 mM MgCl2. Store at -80 °C for up to 18 months. 20. Collagenase D: 10 mg/mL in 1x HBSS with Ca2+ and Mg2+. Store at -20 °C for up to 12 months.
3
Methods
3.1 Generation and Characterization of Human iPSC-Derived Macrophages
Carry out all procedures in this chapter under sterile conditions unless otherwise specified. The generation of iPSC-derived macrophages is described in various differentiation protocols [7, 12–15]. A prominent option is the hematopoietic differentiation of iPSC as 3D aggregates allowing the continuous production of macrophages in a classical adherent or scalable suspension culture. However, the harvest and preparation of the iPSC-derived macrophages for the following procedures are consistent and independent of the cultivation method: 1. Perform the cultivation of human iPSC (as described elsewhere [12, 16]) and start the differentiation towards macrophages as described in more detail here [7, 12–15]. A step-by-step approach can also be found here [12]. Once the production of human macrophages has started, proceed to step 2. 2. Prepare required volume of differentiation medium (1.5 mL/ well of a 6-well dish for an adherent culture or 2.5 mL/ well of a 6-well dish for a suspension culture) and pre-warm the medium shortly in a water bath at 37 °C. 3. Use a 1 mL pipette to carefully collect the supernatant containing the macrophages, while tilting the plate by ~45°. Filter the supernatant through a 70 μm cell strainer and collect in a conical tube. 4. Refresh the differentiation medium of the plate slowly to maintain aggregates in culture for continuous macrophage production and incubate for another week at 37 °C, 5% CO2.
Generation of Alveolar Macrophages from Human iPSC
469
5. Spin down the cell solution for 5 min at 300 × g. Aspirate the supernatant and resuspend cell pellet in appropriate volume of PBS (0.5 mL/ well of a 6-well dish) 6. Stain 10 μL of cell-suspension in a 1:10 dilution with TrypanBlue to discriminate dead cells and count living cells using a hemocytometer. 7. Characterize the harvested macrophages in terms of surface marker expression and phagocytosis capacity as a functional readout. To examine the specific surface marker profile of hCD45+/hCD11b+/hCD14+/hCD163+ iPSC-derived macrophages, incubate sample containing 3 × 105 viable cells in 100 μL MACS buffer with 1 μL of Fc receptor blocking reagent for 20 min at 4 °C to prevent unspecific binding. Afterwards, stain with 1 μL of the respective antibody (for final concentrations, see Materials 2.2, 8 flow cytometry antibodies) and incubate at 4 °C for 45 min in the dark. Keep an unstained sample of 3 × 105 viable cells as a control. 8. Stop the antibody reaction by adding 1 mL of PBS to the sample and spin down for 5 min at 300 × g. 9. Take off supernatant, resuspend cell pellet in an appropriate volume of PBS, and perform flow cytometry analysis (Fig. 2a). 10. To examine the phagocytic activity of the iPSC-derived macrophages, seed 2.5 × 105 cells in 500 μL of phenol-red free, complete X-VIVO medium per well of a 12-well dish and incubate with 10 μL pHrodo™ Red E. coli BioParticles® Conjugate for 2 h at 37 °C or at 4 °C as a negative control. As an additional negative control, use an untreated well. 11. Monitor macrophages under fluorescence microscope to confirm appropriate phagocytosis rate (see Note 4). 12. Detach cells by rising with PBS using a 1 mL pipette, collect in conical tube, and spin down for 5 min at 300 × g. 13. Take off supernatant, resuspend cell pellet in an appropriate volume of PBS, and perform analysis with a cytometer, to confirm and quantify efficient phagocytosis (cells phagocytosing the pHrodo™ Red E. coli BioParticles® can be detected in the PE-channel (e.g., 585/42 BP filter) after pre-gating on viable and single cells (Fig. 2b)). 14. The harvested iPSC-derived macrophages should represent at least 90% hCD45 and hCD11b and at least 80% hCD14- and hCD163-expressing cells. Moreover, the frequency of phagocytosing cells should be at least 80% (Fig. 2). Following proper characterization and verification of macrophage functionality, harvested cells can be used for in vivo transfer and adaptation experiments.
470
Miriam Hetzel et al.
Fig. 2 Gating strategy and flow cytometric analysis of human iPSC-derived macrophages. (a) IPSC-derived macrophages have been stained with different flow cytometry antibodies. As a first step, gating on viable cells is performed based on FSC/SSC properties. Subsequently, duplet exclusion is facilitated by single cell gating on SSC-H vs. SSC-A. Respective surface marker expression is presented as histogram overlays compared to unstained controls. (b) Phagocytotic capacity was evaluated using pHrodo™ Red E. coli BioParticles®. Pre-gating on viable and single cells was performed as described before. Histogram represents cells incubated for 2 h with pHrodo™ Red E. coli BioParticles® and non-treated controls
Generation of Alveolar Macrophages from Human iPSC
3.2 Intratracheal Application of iPSCDerived Macrophages
471
Procedures described in this chapter comprise work with living animals. It is thought to serve as exemplary protocol. However, all animal work needs to be approved by local authorities and guidelines may vary in different countries. Only perform these procedures if you have the approval by your local authorities and make sure that staff is well trained before performing these procedures. To enhance the engraftment of the human iPSC macrophages, it is recommended to use an immunodeficient mouse model, which, e.g., lacks endogenous AM and thus provides an open niche for these cells. Furthermore, transgenic expression of human cytokines, especially GM-CSF, promotes the adaptation of the cells. As an example, mouse strains such as C;129S4-Rag2tm1.1FlvCsf1tm1(CSF1)FlvCsf2/Il3tm1.1(CSF2,IL3) Flv Thpotm1.1(TPO)Flv Il2rgtm1.1Flv Tg(SIRPA)1Flv/J (MISTRG; [17]) or C;129S4-Rag2tm1.1FlvCsf2/Il3tm1.1(CSF2,IL3) Flv Il2rgtm1.1Flv/J [18] may be used. 1. Preparation of intubation tube: Remove needle from intravenous catheter. Cut the tip of the needle for example using a nail clipper to obtain a blunt end. Insert the blunt needle back into plastic catheter to obtain an endotracheal tube with guide bar (Fig. 3).
Fig. 3 Preparation of intubation tube. The process for the preparation tube is given step-by-step from left to right. Take out the intravenous catheter from its packaging (first picture). Pull the needle out of the plastic tube. The safety cap will cover the tip of the needle (second picture). Clip the tip of the needle right behind the safety cap using a nail clipper or pliers (third picture) to obtain a blunt end needle (fourth picture). Carefully pull back the needle into the plastic tube to obtain an intubation tube with a guide bar (fifth picture)
472
Miriam Hetzel et al.
2. Weigh the mice to obtain reference value for anesthesia and administer 0.1 mL/10 g bodyweight of the ketamine/midazolam mixture into the peritoneal cavity (see Note 5). Prepare anesthesia and atropine as indicated in the material section. Draw up in syringes, put needle on top of the syringe, release air from needle, and place within reach. 3. Once mice are in deep anesthesia, apply eye ointment to prevent eyes from drying and place mice on heating mat. Apply atropine (10 μL per 10 g bodyweight) s.c. into the nuchal fold to inhibit salivation, which will reduce risk of asphyxia. 4. Place deeply anesthetized mouse on the plastic stand by hanging it with its upper incisors on the thread (see Note 6). Additionally, fix the mouse with tape at the abdomen to reduce load on the upper incisors. 5. Point the cold light lamp toward the lung with contact to the skin. The exact height can be adjusted later. 6. Carefully grab the tongue with the forceps without applying too much pressure. Pull out the tongue to either side of the mouth. Be careful to not hurt the tongue at the lower incisors. 7. Place the spatula at the tongue base and lift the tongue base by carefully pulling the spatula to the front. Release the forceps from the tongue. You should now be able to see the larynx with the trachea and the vocal folds (see Note 7). Adjust the height of the cold light lamp to see the light shine through the trachea for better visualization (see Note 8). 8. Once you have optimized the view on the trachea, carefully insert the catheter into the trachea. The insertion depth is depending on which catheter size is used. Smaller 24-G catheters need to be introduced with almost full length. 22-G or 20-G catheters should not be introduced with their full length as they might damage the carina trachea where the trachea divides into the main bronchi. 9. Remove the needle from the catheter while thoroughly fixing the tube (see Note 9). 10. Mix the cells by pipetting them up and down and apply them in a max. volume of 60 μL on top of the tube (see Note 10, Fig. 4). If the catheter was placed correctly in the trachea, the fluid should be inhaled immediately. If the fluid stays in the tube, it might have been misplaced in the esophagus. In this case, remove the fluid with the pipette from the tube and then carefully pull out the tube. Repeat the intubation procedure as described before.
Generation of Alveolar Macrophages from Human iPSC
473
Fig. 4 Application of macrophages to the lungs. (a) Harvested macrophages are resuspended in PBS and thoroughly mixed by pipetting up and down. The macrophages will be administered in a volume of 60 μL. (b) Carefully pipette the cells into the cannula. If the tube is placed correctly, the cell solution will immediately be inhaled
11. After successful cell application, immediately remove the catheter. Remove the tape from the abdomen and take the mouse off the stand. Place it on a heating mat and monitor it until it regains consciousness (see Note 11). 12. Place the mouse back into cage once it regains consciousness and monitor it in regular intervals. 3.3 Alveolar Macrophage Recovery and Further Analysis
The time point of final analysis depends on the scientific question you want to answer and needs to be in line with what you applied for at your local authorities. However, the macrophages need a sufficient time period to adapt and convert to an alveolar macrophage-like phenotype. Thus, for fully converted cells the analysis should be performed earliest 4 weeks after cell transfer. 1. Prepare 1 mL syringe with 27 G needle with ketamine/xylazine as indicated in the material section. Apply 100 μL of the ketamine/xylazine mixture per 10 g bodyweight i.p. and wait until breathing and heart activity stopped. 2. Pour 70% ethanol onto fur until all parts are completely wet. 3. Pierce needles through distal parts of all four extremities and pin mouse onto aluminum foil-covered Styrofoam board (see Note 12). 4. Open the skin by lifting it with forceps and cutting it with a scissor longitudinal in the middle of the abdomen from genital region up to the mouth. Detach the skin from the subcutaneous tissue by blunt preparation using the closed scissor.
474
Miriam Hetzel et al.
5. Open the peritoneal cavity by lifting the tissue with forceps and cutting it with a scissor. Be careful not to cut into the intestine. Flap the viscera to the left side of the mouse to expose the large abdominal vessels. Cut the vessels to bleed the mouse. You can also perform heart compression to remove as much blood as possible (see Note 13). You will see that the color of the liver will faint the more blood you drain as an indicator of success. 6. Open the thoracic cavity by carefully cutting the diaphragm at the outer edge. Be very careful not to hurt the lungs! Once the diaphragm is cut, the lungs collapse due to the loss of the negative pressure and the diaphragm can be cut open completely. Cut the sternum longitudinal in the medial line up to the throat. While cutting, make sure to lift the thorax with the scissors to have distance to the lung tissue. Be especially careful in the upper part to not hurt the trachea. Cut the ribs lateral in the medial axillary line on both sides to expose the lungs completely. Also at this step, be very careful not to accidentally cut the lung tissue. 7. Remove the skin around the throat (see Note 14). Remove the thyroid gland from the throat by lifting it with the forceps and cutting it with the scissor to expose the trachea. Very carefully, cut the connective tissue surrounding the trachea. 8. Take a 20 G peripheral intravenous catheter and carefully pierce the needle between two ring cartilages in a flat angle into the trachea. Once you pierced through the tissue, pull out the needle and carefully push the tube into the trachea. 9. Draw up 1 mL of PBS into a 1 mL syringe. Connect the syringe to the end of the catheter, fix the catheter with one hand, and push the PBS through the catheter into the lungs using the other hand. It might happen that the PBS runs out through the nose. You can prevent this by pressing your thumb on the nose. Pull out the stamp of the syringe again to recover the fluid from the lungs and obtain the lavage fluid. You will not be able to recover the full 1 mL but rather around 700 μL (see Note 15). 10. Repeat the flushing procedure 2–4 times with 800 μL PBS per flush to maximize the cell yield (see Note 16). Collect all flushes in a 15 mL tube and store on ice. 11. Remove the heart from the thorax and cut off the lungs. Place the lungs in a well of a 6-well dish containing MACS buffer (see Note 17). Store the plate on ice. Dispose of the mouse carcass unless you want to use other organs for analysis. 12. After finishing the dissection of all mice, continue with preparation of lavage fluid and lung tissue.
Generation of Alveolar Macrophages from Human iPSC
475
13. Spin down lavage fluid 10 min at 300 × g (see Note 18). Take off supernatant and resuspend cells in 400 μL MACS buffer (see Note 19). Separate into two 1.5 mL reaction tubes (200 μL each, label as stained and unstained sample) and store on ice until antibody staining. 14. For digestion of the lungs, thoroughly rinse the lungs with PBS and transfer them into a new 6-well. Cut the lung tissue into small pieces using a scissor. Add 2 mL dissociation buffer and incubate on an orbital shaker at 150 rpm for 30 min. 15. Add 1 mL dissociation buffer and incubate for additional 15 min on the shaker. 16. Stop DNase I reaction by adding 3 μL 0.5 mM Na-EDTA to the lung tissue to achieve a final concentration of 0.5 μM EDTA. 17. Transfer mixture to a cell strainer sitting on top of a 50 mL tube. Mince remaining tissue clumps, e.g., by using the stamp of a 2 mL syringe. Rinse cell strainer with MACS buffer. 18. Centrifuge 8 min at 300 × g. 19. Discard supernatant and add 2 mL red cell lysis buffer. Incubate for 3 min at RT. Stop reaction by adding at least 4 mL PBS or MACS buffer. 20. Centrifuge 8 min at 300 × g. 21. Resuspend cells in 400 μL MACS buffer. Separate into two 1.5 mL reaction tubes (200 μL each, label as stained and unstained sample) and store on ice until antibody staining. 22. To isolate the human, iPSC-derived cells for further analysis (e.g., whole transcriptome analysis), perform fluorescenceactivated cell sorting (FACS) on hCD45-positive cells. 23. To characterize the adaptation of iPSC-derived macrophages to an AM-like phenotype, perform flow cytometry analysis of hCD45, hCD169, hCD14, hCD206, and hCD11c expression on the hCD45+, human iPSC-derived cells in the BALF and lung. 24. To prepare cells for antibody labeling first incubate the “stained samples” generated under point 13. (BALF) and 21. (lung homogenate, both already resuspended in 200 μL MACS buffer) with 0.5 μL of Fc receptor blocking reagent for 20 min at 4 °C to prevent unspecific binding. Afterwards, add 1 μL of the respective antibody per sample and incubate at 4 °C for 45 min in the dark. 25. Add 1 mL of PBS to the sample and spin down for 5 min at 300 × g. 26. Take off supernatant, resuspend cell pellet in 100 μL of PBS and perform flow cytometry analysis or FACS (see Note 20 and Fig. 5).
476
Miriam Hetzel et al.
Fig. 5 Flow cytometric analysis of mice transplanted with human iPSC-derived macrophages. Representative gating strategy and flow cytometric analysis of murine broncho-alveolar lavage fluid (BALF) stained with hCD45 antibody (hCD45 Pe-Cy7, clone: HI30) to detect (or isolate) human cells for further analysis. Upper row: Representative gating strategy for unstained sample. Gating on viable cells is performed based on FSC/SSC properties (first picture). Subsequently, duplet exclusion is facilitated by single cell gating on SSC-H vs. SSC-A (second picture). Human cells are identified by expression of hCD45 (expression is shown via autofluorescence channel; third picture). Lower row: Representative staining for hCD45 in a non-transplanted animal and an animal that received 4 × 106 iPSC-derived macrophages 8 weeks prior analysis
4
Notes 1. Catheter size depends on the size of the mouse. Rough estimate: small mice with a weight 30 g—20 G catheter. However, this depends on how you feel comfortable in handling. Some people prefer the small 24 G catheter regardless of the size of the mouse. 2. Exact size is not important. However, they should not be too small. Otherwise, they will bend during preparation of the mouse. 3. It is recommended to use the big catheter here regardless of the size of the mouse as it preoccupies the trachea best and will reduce the risk of spilling lavage fluid.
Generation of Alveolar Macrophages from Human iPSC
477
4. The fluorescence of the bioparticle dye only increases in an acidic environment, thus, only after being phagocytosed, which makes this assay specific and easy to use. 5. Anesthesia given i.p. has a quite long duration. Inhalation anesthesia for example using isoflurane is also possible. However, this requires very skilled personal as the time window for intubation and cell application is quite narrow as isoflurane inhalation must be stopped for the intubation. Thus, time for intubation and cell application is approximately 30–60 s. 6. The thread should not be too thin as this will lead to bruising and tissue damage. A rubber band can be used as well. 7. You can also use magnifying glasses to have a better view. 8. Take your time to adjust the position of the spatula and the lamp. It is crucial to have an ideal view to successfully place the catheter in the trachea. 9. Sometimes the needle is hard to retract because it flattens during clipping of the needle tip. Therefore, it is important to thoroughly hold the tube. Otherwise, the catheter will be pulled out of the trachea together with the needle. 10. Maximum volume that can be tolerated without risking asphyxia. Mice with a body weight over 30 g might tolerate slightly more. However, the volume should be kept low to reduce risk of side effects. 11. The highest risk of side effects and severe events like apnea is within the first 10 min after the procedure. If breathing becomes irregular or arrests, you can try to change the position of the mouse for example by turning it on the back or putting it in a slight head-down position. 12. Do not pin them straight from the top but with a small angle from the side, which will prevent the mouse from moving along the needle. 13. If you want to remove all blood from the lungs for special analysis, you can perform perfusion. However, this needs to be applied for with the authorities as it is performed on a living animal. 14. The musculature around the mouth is very strong and well supplied with blood. You might hit many vessels here that cause massive bleeding and constrain the view. If the blood reaches inside the lung, it might also interfere your results in the lavage. Therefore, it is very important to bleed the mouse as good as possible through the abdominal vessels. If you encounter bleeding in throat or thorax during preparation, make sure to constantly remove the blood with paper towels as good visibility is crucial to not destroy important structures.
478
Miriam Hetzel et al.
15. In case you recover much less than that and you feel that there is a resistance in the lungs preventing you from pulling back the stamp, you might slightly change the position of the tube in the trachea by pulling it a little out or pushing it a little more in. However, be careful not to pull too much as you might pull the tube out of the trachea. If this happens, you can try to push it back through the hole. If you cannot get it back in, you can use a new catheter with needle to get back into the hole. 16. If you want to measure cytokine or protein concentrations in the lavage fluid, you might collect the first flush in a separate tube to not dilute your analyte of interest too much. 17. Most of the cells that converted to AM-like cells should be found in the lavage fluid. However, depending on the efficiency of flushing, there might be many cells sticking to the lung tissue. If you want to analyze these cells, e.g., by flow cytometry, continue with the description of lung processing. 18. If lavage fluid was contaminated with blood during preparation, the cell pellet will appear red. You can perform red blood cell lysis to remove the erythrocytes for further analysis. 19. If a mouse model characterized by the absence of endogenous AM is used, these animals develop pulmonary alveolar proteinosis and the BALF will appear highly turbid and the pellet after centrifugation is quite big due to massive protein accumulation. To reduce the amount of contaminating lipoproteinaceous material, gradient centrifugation can be performed. However, in our hands the improvement was minor and not in relation to the additional time that is necessary to perform it. 20. Also, during the flow cytometry analysis, a lot of small debris/ proteins will be detected if a mouse model lacking AMs is used. As the expected population of donor cells shows in turn a rather low frequency, make sure to acquire enough (e.g., 1 × 106 total) events to properly analyze the population of interest (hCD45+).
Acknowledgments This chapter has received support from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No. 852178), and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy— EXC 2155—project number 390874280. This work was also supported by the Fraunhofer Internal Programs under Grant No. Attract 40-01696.
Generation of Alveolar Macrophages from Human iPSC
479
References 1. Trapnell BC, Nakata K, Bonella F et al (2019) Pulmonary alveolar proteinosis. Nat Rev Dis Primers 5(1):16. https://doi.org/10.1038/ s41572-019-0066-3 2. Belchamber KBR, Donnelly LE (2017) Macrophage dysfunction in respiratory disease. Results Probl Cell Differ 62:299–313. https://doi.org/10.1007/978-3-319-540900_12 3. Subramanian S, Busch CJ, Molawi K et al (2022) Long-term culture-expanded alveolar macrophages restore their full epigenetic identity after transfer in vivo. Nat Immunol 23(3): 458–468. https://doi.org/10.1038/s41590022-01146-w 4. Li F, Okreglicka KM, Piattini F et al (2022) Gene therapy of Csf2ra deficiency in mouse fetal monocyte precursors restores alveolar macrophage development and function. JCI Insight 7(7):e152271. https://doi.org/10. 1172/jci.insight.152271 5. Litvack ML, Wigle TJ, Lee J et al (2016) Alveolar-like stem cell-derived Myb(-) macrophages promote recovery and survival in airway disease. Am J Respir Crit Care Med 193(11): 1219–1229. https://doi.org/10.1164/rccm. 201509-1838OC 6. McCowan J, Fercoq F, Kirkwood PM et al (2021) The transcription factor EGR2 is indispensable for tissue-specific imprinting of alveolar macrophages in health and tissue repair. Sci Immunol 6(65):eabj2132. https://doi.org/ 10.1126/sciimmunol.abj2132 7. Buchrieser J, James W, Moore MD (2017) Human induced pluripotent stem cell-derived macrophages share ontogeny with MYB-independent tissue-resident macrophages. Stem Cell Reports 8(2):334–345. https://doi.org/10.1016/j.stemcr.2016. 12.020 8. Abud EM, Ramirez RN, Martinez ES et al (2017) iPSC-derived human microglia-like cells to study neurological diseases. Neuron 94(2):278–293.e9. https://doi.org/10. 1016/j.neuron.2017.03.042 9. Chen I (2020) Differentiation of human induced pluripotent stem cells (hiPSCs) into osteoclasts. Bio Protoc 10(24):e3854. https://doi.org/10.21769/BioProtoc.3854 10. Tasnim F, Xing J, Huang X et al (2019) Generation of mature Kupffer cells from human induced pluripotent stem cells. Biomaterials
192:377–391. https://doi.org/10.1016/j. biomaterials.2018.11.016 11. Happle C, Lachmann N, Ackermann M et al (2018) Pulmonary transplantation of human induced pluripotent stem cell-derived macrophages ameliorates pulmonary alveolar proteinosis. Am J Respir Crit Care Med 198(3): 350–360. https://doi.org/10.1164/rccm. 201708-1562OC 12. Ackermann M, Rafiei Hashtchin A, Manstein F et al (2022) Continuous human iPSCmacrophage mass production by suspension culture in stirred tank bioreactors. Nat Protoc 17(2):513–539. https://doi.org/10.1038/ s41596-021-00654-7 13. Ackermann M, Kempf H, Hetzel M et al (2018) Bioreactor-based mass production of human iPSC-derived macrophages enables immunotherapies against bacterial airway infections. Nat Commun 9(1):5088. https:// doi.org/10.1038/s41467-018-07570-7 14. Lachmann N, Ackermann M, Frenzel E et al (2015) Large-scale hematopoietic differentiation of human induced pluripotent stem cells provides granulocytes or macrophages for cell replacement therapies. Stem Cell Reports 4(2): 282–296. https://doi.org/10.1016/j.stemcr. 2015.01.005 15. Karlsson KR, Cowley S, Martinez FO et al (2008) Homogeneous monocytes and macrophages from human embryonic stem cells following coculture-free differentiation in M-CSF and IL-3. Exp Hematol 36(9):1167–1175. https://doi.org/10.1016/j.exphem.2008. 04.009 16. Sahabian A, Dahlmann J, Martin U et al (2021) Production and cryopreservation of definitive endoderm from human pluripotent stem cells under defined and scalable culture conditions. Nat Protoc 16(3):1581–1599. https://doi. org/10.1038/s41596-020-00470-5 17. Rongvaux A, Willinger T, Martinek J et al (2014) Development and function of human innate immune cells in a humanized mouse model. Nat Biotechnol 32(4):364–372. https://doi.org/10.1038/nbt.2858 18. Willinger T, Rongvaux A, Takizawa H et al (2011) Human IL-3/GM-CSF knock-in mice support human alveolar macrophage development and human immune responses in the lung. Proc Natl Acad Sci U S A 108(6): 2390–2395. https://doi.org/10.1073/pnas. 1019682108
Chapter 32 Tackling Tissue Macrophage Heterogeneity by SplitCre Transgenesis Sigalit Boura-Halfon, Rebecca Haffner-Krausz, Shifra Ben-Dor, Jung-Seok Kim, and Steffen Jung Abstract Macrophages represent a broad spectrum of distinct, but closely related tissue-resident immune cells. This presents a major challenge for the study of functional aspects of these cells using classical Cre recombinasemediated conditional mutagenesis in mice, since single promoter-driven Cre transgenic models often display limited specificity toward their intended target. The advent of CRISPR/Cas9 technology has now provided a time- and cost-effective method to explore the full potential of binary transgenic, intersectional genetics. Specifically, the use of two promoters driving inactive Cre fragments that, when co-expressed, dimerize and only then gain recombinase activity allows the characterization and manipulation of genetically defined tissue macrophage subpopulations. Here, we will elaborate on the use of this protocol to capitalize on these recent technological advances in mouse genetics and discuss their strengths and pitfalls to improve the study of tissue macrophage subpopulations in physiology and pathophysiology. Key words tissue macrophages, splitCre, CRISPR-Cas9, intersectional genetics, Cre loxP
1 Introduction Macrophages are essential components of the organism’s innate immune defense against microbial pathogens, but also critically contribute to organ development, as well as the maintenance of adult tissue homeostasis [1]. Recent studies have revealed profound developmental and molecular heterogeneity of tissue macrophages [2–5]. Likewise, in addition to generic activities, such as phagocytosis, emerging data have established that tissue macrophages exhibit functions associated with specific physiological organ tasks, such as the ones of heart, lung, and liver [6–8]. Moreover, high-end flow cytometry and single-cell transcriptomics have revealed that individual tissues host multiple macrophage Sigalit Boura-Halfon, Rebecca Haffner-Krausz and Shifra Ben-Dor contributed equally. Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_32, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
481
482
Sigalit Boura-Halfon et al.
subpopulations, often in distinct anatomic locations, such as proximity to vasculature or nerves [9, 10]. For instance, macrophages of the central nervous system (CNS), i.e., brain and spinal cord, can be subdivided into microglia that reside in the CNS parenchyma, and CNS border-associated macrophages (BAM) that comprise perivascular and meningeal cells, as well as choroid plexus macrophages (CPM) [11]. One of the pivotal advances in the study of functional contributions of macrophages in their in vivo context was the ability to target mammalian gene expression in a cell-specific manner using Cre/lox technologies in mice [12–14]. Intense efforts have been made toward establishing “macrophage-specific” Cre driver lines and, collectively, these studies have significantly advanced our insights into these cells [15]. Macrophage-targeted Cre expression has been attempted with promoters driving genes, such as LysM [16], Emr1 (F4/80) [17], Itgam (CD11b) [18], Csf1r [19], and Cx3cr1 [20, 21]. This included originally classical transgenic lines, and more recently, “knock-in” mouse strains that arguably display more accurate expression. However, with rare exceptions, such as the Clec4fcre animals used to target Kupffer cells [6], the existing Cre transgenic mouse strains, although widely used, display suboptimal recombination efficiency and specificity for tissue macrophages. For example, LysMcre mice also target neutrophils and neuro-ectoderm [16, 22] and display limited efficiency in certain tissue macrophages [11, 23]. Csf1rcre mice efficiently target monocytes/macrophages in various tissues, but also dendritic cells (DC), granulocytes, and T lymphocytes [19]. Likewise, also Cx3cr1cre mice target other immune cells including T lymphocytes and DC, as well as non-hematopoietic cells [21, 24–27], although the specificity is somewhat improved with the CreER approach [28]. Given the heterogeneity of tissue macrophages, together with their close degree of relatedness, the expression pattern of a single promoter is clearly insufficient to genetically define distinct macrophage subpopulations. To improve the specificity of Cre-mediated cell targeting, more complex, binary transgenic models have been introduced, including systems that rely on sequential promoter activities [29] or two site-specific recombinases [30, 31]. An alternative approach, successfully adopted in mice [32, 33], relies on a biological “AND gate” strategy. Specifically, in this approach the Cre enzyme is divided into two inactive polypeptide chains, which each are fused to a domain that promotes dimerization, such as the GCN4-coiled coil [34] (see Note 1). Upon their co-expression in a given cell, the 40aa NCre and 284 aa CCre peptides associate with each other to generate an active recombinase (Fig. 1) [32, 35]. Recombination is hence restricted to cells, which display coinciding activities of the two promoters driving the Cre fragments. Notably, like in the case of classical Cre transgenic animals, the overlap of NCre and CCre expression genetically defines the cells.
SplitCre Transgenesis
483
Fig. 1 Design of splitCre proteins and principle of Cre complementation system. (a) Schematic of NCre and CCre proteins. (Modified from Hirrlinger et al. 2009) (b) “AND gate” approach with CCre and NCre expression placed under the control of two different promoters (CX3CR1 promoter (a) and a 2nd promoter (b). Only if both promoters are active, functional complementation takes place and loxP flanked sequences are recombined, thereby activating lox-STOP-lox (LSL) reporter genes
The label of targeted cells does, therefore, not necessarily have to overlap with the expression of the original proteins driven by the two promoters, i.e., the phenotype of the cells at the time of analysis (Fig. 2). Specifically, the cells might have ceased to express one or both of the proteins. Conversely, the necessary overlap of NCre/CCre expression might not be reached due to temporally limited or unsynchronized transcriptional bursts of the two promoters. Irrespectively, recombinase activity of the NCre/CCre dimers in targeted cells can, in combination with different “floxed” alleles, be used for reporter tagging of the targeted cells, their conditional ablation using the DTR system, translatome profiling, and mutagenesis [28, 33, 36, 37] (see also Chapter 17 in this issue of MiMb) (Fig. 1b).
484
Sigalit Boura-Halfon et al.
Fig. 2 Genetic definition of cell populations using binary splitCre transgenesis. Scenarios explaining dissociation of phenotype, i.e., expression of protein 1 and 2 whose promoters also drive the Cre fragments, and genotype of splitCre-targeted cells. Note that targeting of cells by the intersectional approach requires both sufficient temporal overlap of transcription and sufficient threshold. Keep also in mind that half-lives of the proteins expressed from the targeted loci and the NCre and CCre proteins can differ, with the latter likely to be rather ephemeral
The use of binary transgenic splitCre animals in mice is clearly in its infancy, and with every newly established strain, we will learn about the intricacies of the system. Future modifications could for instance include a replacement of the GCN4 domain with intein-
SplitCre Transgenesis
485
based modules that would prevent homodimerizations of the Cre fragments [31]. Rearrangement efficiency of the binary Cre that is inherently lower than the regular Cre recombinase could be boosted [38]. Furthermore, for specific applications, it might be worthwhile to explore inducible splitCre systems [39].
2 2.1
Materials Mice
1. C57BL/6J mice. 2. B6.Cg-Gt(ROSA)26Sor tm9(CAG-tdTomato)Hze/J (Jackson stock no. 007909) [67].
2.2 General Tools and Machines
3. C57BL/6-Lyve1em1(Ncre)JungCx3cr1em1(Ccre)Jung/J stock no. 033319) [34].
(Jackson
4. C57BL/6-Sall1em1(Ncre)JungCx3cr1em1(Ccre)Jung/J stock no. 033318) [34].
(Jackson
1. Nuclease- and pyrogen-free microcentrifuge tubes. 2. Ultra-free centrifugal filters 0.1 μm. 3. Microcentrifuge. 4. Microinjection needles.
2.3
Reagents
1. Homology-directed repair (HDR) templates: When templates are up to 200 bases, use high-quality oligos, e.g., Ultramer® DNA Oligo from IDT with standard desalting purification. For templates longer than 200 bases, high-fidelity single-stranded DNA (ssDNA) fragments, e.g., Megamer® ssGene Fragment from IDT. 2. Guide RNAs: Alt-R® CRISPR-Cas9 crRNA, with desired specific 20 base guide sequence (order in tubes, e.g., IDT). 3. Guide RNAs: Alt-R® CRISPR-Cas9 tracrRNA (e.g., IDT). 4. 100 μg Alt-R® S.p. Cas9 Nuclease V3 (e.g., IDT). 5. 1 M Tris–HCl, pH 7.4. 6. 0.5 M EDTA in RNAse-free, molecular biology grade water. 7. 0.45 μm syringe filter. 8. RNAse-free, molecular biology grade water. 9. Embryo-tested water. 10. Injection buffer: 1 mM Tris pH 7.5, 0.1 mM EDTA: In biosafety hood to maintain sterility, dilute 0.5 M EDTA 1:5 in water to 0.1 M. Add 5 μL 1 M Tris pH 7.5 and 5 μL 0.1 M EDTA to 5 mL H2O. Filter-sterilize through 0.45 μm syringe filter (see Note 2).
486
Sigalit Boura-Halfon et al.
11. Cas9 dilution: dilute Cas9 to 3.1 μM by mixing 1 μL Cas9 into 19 μL injection buffer (e.g., IDT Cas9 supplied as 61 μM, 10 μg/ul) (see Note 3). 12. Injection mix: 31 μL injection buffer, 1.25 μM guide RNA (1 μL from 50 μM stock), 15 ng/μL HDR long ssDNA template (4 μL from 150 ng/μL stock), 0.3 μM Cas9 protein (4 μL from 3.1 μM stock). Flick to mix and spin down 3″ after addition of each reagent. 13. PCR primers 0.025 μmol, desalt purification.
3
Methods
3.1 Design Strategy—Bioinformatic 3.1.1 Strain Considerations
3.1.2 Genomic Considerations
Inbred C57Bl/6 mice are the strain of choice for most targeted mouse models. The use of inbred strains removes the genetic heterogeneity between individual animals; however, it should be noted that there are a number of different C57Bl/6 substrains which differ at diverse loci [40]. Historically, some other mouse strains, such as FVB/N, were more proficient for transgenesis. However, the efficiency of CRISPR/Cas engineering has made it feasible to target transgenes directly into embryos of any desired background strain, including animals that already harbor transgenes, saving time, and superfluous animal use. Specifically, one can perform NCre and CCre transgenesis sequentially. It is critical to keep strain differences in mind while designing CRISPR guides and repair oligos. Most design programs have C57Bl/6 as the standard mouse genome. If you are using a different strain, try to find a database with strain differences, for example “Mouse Phenome Database - The Jackson Laboratory” [41]. If your strain cannot be found, sequence the area surrounding your gene of interest, basing primers on the sequence available, to ensure that your CRISPR reagent sequences are accurate. The design of splitCre systems requires detailed knowledge of the genomic surrounds of your gene of interest: 1. How good of a candidate is your gene and how specific is it for the cell type of interest? Are there genes close in sequence that can make it difficult to find guides/oligo arms? For illustration, we will describe our considerations for the generation of splitCre mice which allowed us to discriminate parenchymal and border-associated CNS macrophages [33]. Cx3cr1 (NM_009987) was chosen to host the CCre fragment since it is broadly active in tissue macrophages, including their precursors [21, 42]. Cx3cr1 is also highly expressed in microglia and likewise, though less so, in BAM (http://mousebrain.org) [61]. Sequence comparison to the mouse genome revealed a
SplitCre Transgenesis
487
genomic repeat at the end of the 3’ UTR, which was, however, irrelevant in our case, as we decided to insert the CCre right after the open reading frame (ORF) of the host protein using a T2A strategy [43]. 2. Does your gene overlap with any other genes (on the same or opposite strands) or with non-coding genes (such as microRNA, which can be intronic)? Cx3cr1 is a two-exon gene, with all of the protein coding sequence in the second exon. There is a potential non-coding overlapping gene in the very 5′ end of the gene, but as the second exon only will be edited, it should not affect the design. 3. Do you wish to replace the gene of interest, making use of its promoter only, or do you wish to preserve the expression of your gene of interest, in addition to the Cre? If the host gene is replaced, will deletion be lethal? If one copy of the gene is replaced with thesplitCrefragment, is the host gene haplosufficient? For splitCre, we recommend not replacing the host gene, but adding the recombinase in order to maintain the native state of the cell as much as possible. 4. Are there splice variants and, if so, do you want to target all of these or only selected ones? Is the beginning or the end of the host gene common to all variants? Variants should be checked using a genome browser. Our preferred is the UCSC genome browser (genome.ucsc.edu) [44]. Actual sequence (mRNA and EST) should be used, and not just gene structure prediction, even from reliable sources (RefSeq [45], GENCODE [46]). In our specific case, Cx3cr1 had no known variants. 5. Should you add the CCre to the N terminus or C terminus of the host gene? Does the protein translation start in the first exon? If so, will you disrupt the promoter by adding the CCre? Will you disrupt enhancer elements otherwise? Predicted enhancer and promoter elements can be seen in the ENCODE cCREs (cis-regulatory elements) track [47]. If there are no promoter elements in the track, the area directly above the transcription start site (TSS) should be taken into account. To simplify the decision, one can check if the gene has been tagged before. If so, were both termini attempted? CCre does not work as a fusion to the host protein and will have to be added as a separate transcriptional (IRES) or translational (T2A /P2A) unit (see Note 4). Positioning of the genes can affect expression (generally with the first expressed better). Therefore, we decided to place the CCre at the C terminus of Cx3cr1, directly before the stop codon using T2A, a self-cleaving 2A peptide (Fig. 3). We do not recommend using IRES for CCre due to its size (approximately 1300 bp), and the lack of consistency in
488
Sigalit Boura-Halfon et al.
Fig. 3 Design strategy of the insertion of CCre fragment at the C terminus of Cx3cr1, directly before the stop codon, using T2A peptide sequence
expression of the second gene. For the splitCre approach, the genomic considerations have to be taken into account for both insertions. The final design may be different for each gene (N term or C term, for example). 3.1.3
Guide Design
Once the preferred site for introduction of the splitCre fragment has been chosen, the guide sequences need to be selected. If both N and C termini of your gene of interest are viable options for CCre insertion, both should be checked, and the better guide should be chosen. The ideal guide should be positioned as close as possible to the insertion site, preferably with the cleavage site on the site of insertion. Insertion within 10 bp of cleavage generally works well [48], though above 10 bp the efficiency drops drastically due to recombination that can occur in between the cleavage site and the insertion. If no suitable guide is found within 10 bp of the insertions, several alternatives are available: (i) Use a guide more than 10 bp distant and in the donor DNA repair template, add silent mutations, which may improve insertion efficiency by reducing potential recombination crossover sites. (ii) If there are no guides closer than 40 bp, or silent mutations cannot be introduced in the repair template, an alternate Cas enzyme can be considered. For example, we have successfully used Cas12a (Cpf1) for AT-rich regions of the genome [49] (see Note 5). (iii) If neither of the previous options are applicable, two guides can be used, to remove a larger flanking region, together with a longer repair oligo to restore the deleted sequences.
SplitCre Transgenesis 3.1.4
Choosing Guides
489
Many software tools are available which predict efficiency and specificity of guide sequences. We advise using several programs and comparing results, optimizing for both on- and off-target activity. Specificity is more critical than efficiency, as we would like the final phenotype to reflect only the targeting of our gene of interest. We generally use CRISPOR (crispor.tefor.net) [50] and Benchling (www.benchling.com) [51, 52] as well as other programs. In the two named sites, scores range from 0 to 100, and guides with above 50 in all algorithms are preferred. In our specific Cx3cr1 case, guides were selected for the least off-target activity. Guide # 3 was chosen (Fig. 4), despite it having a score of 49, since that was close to 50, and importantly, the cut site was much closer to the insertion site (17 bp as opposed to 26 or 112 bp for the other suggested guides). After an initial guide is chosen, more in-depth analysis of potential off-targets is needed. We prefer to work with the off-target list from Benchling and manually inspect each potential off-target by running it, together with its PAM in a Blat search [53] at the UCSC genome browser (genome.ucsc.edu). The hit is then expanded to see what is in the genomic proximity, in order to
Fig. 4 Choosing a gRNA for Ccre insertion site. Examples of possible gRNAs for the insertion of Ccre cassette at the stop codon of Cx3cr1 gene extracted from benchling (a) or Crispor (b). The chosen gRNA is highlighted
490
Sigalit Boura-Halfon et al.
classify the off-target as risky or neutral. Some guides fail in this test (e.g., too many protein coding genes targeted, close relatives of the host protein, other genes in the same pathway, linkage to the on-target site, potential embryonic lethality). In these cases, it may be necessary to select an alternative guide. Note: in mice, if an off-target mutation is detected in a potential founder individual, as long as it is on a different chromosome, it can be crossed away during expansion of the line. 3.1.5
Repair Template
3.2 Gene-Targeted “Knock-In” of splitCre Insertions Using CRISPR/Cas9
The repair templates for the NCre and CCre insertions have to be designed with flanking homology arms. There are several options for introduction of long DNA: single-stranded DNA (ssDNA), double-stranded DNA (dsDNA), or plasmid. Efficiency of integration and the viability of injected embryos have been demonstrated to be significantly higher using ssDNA donors than dsDNA donor [54]; therefore, we have a definite preference for ssDNA. In addition, there is a reduced chance of random integration into the genome. Plasmid DNA might, however, be more readily available and generally is less expensive. We designed long single-stranded repair templates. The size of the homology arms can range from 40 to hundreds of bases, though given the lengths of the inserts in the splitCre approach (NCre ~450 bp, CCre ~1300 bp,) we prefer to use arms of between 100 and 200 bases. Another important consideration for repair template design is whether the intended sequence repair eliminates the ability of the Cas9 RNP to re-cut the repaired allele. This could result in the introduction of insertions or deletions (indels) via nonhomologous end joining (NHEJ) repair in addition to the desired repair. To avoid this, one or more silent changes should be introduced to mutate the PAM sequence. If mutations cannot be introduced into the PAM, silent mutations (2–4) should be added along the guide sequence. If the cleavage site is more than 10 bp away from the insertion site, additional silent mutations can be placed between the cleavage and insertion sites at intervals of less than 10 bp. In the case of Cx3cr1, we added one silent mutation in order to both disrupt the PAM and have no more than 10 homologous bases between the cut site and repair site (Fig. 5) (see Note 6). In this specific case, the position of the guide sequence was after the stop codon. Effects of mutations in the 3’ UTR are generally rather unpredictable. However, caution should be taken if near a splice junction. In general, the less mutations, even silent ones, the better. CRISPR reagents are ordered from commercial sources, according to the design considerations in Subheading 3.1. No additional purification is required above the standard desalt purification. Our preferred source is IDT (https://eu.idtdna.com), though other companies supply comparable reagents (see Note 7).
SplitCre Transgenesis
491
Fig. 5 Schematic presentation of long ssDNA Repair oligo, HDR template, for Ccre insertion at Cx3cr1 stop codon 3.2.1 Annealing of crRNA with gRNA (SeeNote 8)
1. Resuspend crRNA and tracrRNA individually to 100 μM in manufacturer-supplied duplex buffer (e.g., 4 nmols RNA in 40 μL Duplex buffer) (see Note 9). 2. Mix equal volumes of 100 μM crRNA and 100 μM tracrRNA to a concentration of 50 μM gRNA. 3. Anneal in a thermocycler (95 °C for 5 min and then ramp down to 25 °C at 5 °C/min). Alternatively, anneal at 95 °C for 5 min and let cool to room temperature on the bench (about 15 mins). Aliquot immediately into 3 μL aliquots in 1.8 mL microtubes. 4. Resuspend long ss DNA (HDR template) in RNase-free water at a concentration of 150 ng/μL. Aliquot into 8 μL aliquots. 5. Store guide RNAs and repair templates at -80 °C. Reagents can be stored for at least 12 months.
3.2.2 Microinjection of Guides and HDR Templates
Animal procedures, require skills, experience, and specialized equipment. These are normally carried out by core services in accredited animal facilities (see Note 10). These protocols are described extensively in other publications [55–57]. Briefly, the stages involved are (1) superovulation of immature female mice of the required strain and mating with stud males, (2) dissection and collection of 0.5-day post coitum (dpc) embryos, (3) microinjection or electroporation of ribonucleoprotein (RNP) (see Note 11) complexes into the embryo nucleus, (4) embryo transfer to pseudopregnant females for gestation and parturition of mice, and finally (5) screening and breeding of founder mice. This section will describe only protocols specific for splitCre transgenesis, which may not necessarily be performed by the core
492
Sigalit Boura-Halfon et al.
service provider. Some facilities will require the client to prepare the CRISPR reagents and RNP mixes; other facilities include reagent preparation in their service (see Note 12). On the day of microinjection, the Cas9 RNA or RNP mixes should be prepared while the embryos are being collected. 3.2.3 Preparation of RNP Complexes for Microinjection
1. Dilute Cas9 to 3.1 μM in the injection buffer (see Subheading 2.3, steps 10 and 11) (see Note 3). 2. Thaw guide RNAs and repair template on ice. Spin down 3 s to collect microdroplets. 3. Assemble injection mix (see Subheading 2.3, step 12). Transfer the mix to the filter chamber of Ultra-free centrifugal filters 0.1 μm and centrifuge at 16,000 g for 5 min in a microcentrifuge. Remove filter (see Note 13). 4. Incubate in dry bath 37 °C for 10 min. Spin down 3 s to collect microdroplets. 5. Keep on ice prior to loading microinjection needles.
3.2.4 Embryo Microinjection
Two alternative methods of delivery of CRISPR reagents to one-cell mouse embryos are available: microinjection and electroporation. The first reports of direct CRISPR targeting in vivo in mouse embryos employed microinjection, the traditional method of generating mice containing exogenous transgenes [56, 58]. Microinjection is however labor-intensive and requires high skills and specialized equipment. Electroporation as a means to deliver the CRISPR/Cas9 components into mouse zygotes provides a simple, highly efficient alternative that also allows large-scale genome editing [59, 60]. While electroporation is highly efficient for the delivery of short repair templates containing point mutations or short epitope tags, up to 200 bases long, microinjection into the pronucleus of zygotes remains, however, the method of choice for delivery of long repair templates (>200 bases), which do not easily enter the nucleus following electroporation. Since both the CCre and NCre HDR templates are greater than 200 bases, we hence do not recommend electroporation for the generation of splitCre mice. Following microinjection, manipulated embryos are transferred to surrogate host mothers according to standardembryo transfer procedures. We recommend that the embryo transfer is done on the same day as the microinjection. Litters are delivered after 19 days gestation (see Note 14). Pups are considered F0 (see Note 15). At weaning, a tissue sample is taken for genotype analysis (see Note 16).
SplitCre Transgenesis
3.3 Genotyping and Validation
493
Pups must be screened to identify potential founder animals containing the correctly integrated Cre or splitCre transgene insertion. Several sets of primers are required for screening and genotyping of founders. An initial pair can be designed flanking the insertion, which should give a larger band in CCre-positive mice relative to wild type (Fig. 6, set a) (see Note 17). At least one primer should be outside of the homology arms, in order to validate proper genomic location. Because of the size of the Cre insertion together with the homology arms, the amplicon might be too large for efficient PCR amplification, so one set of primers may not be conclusive for CCre, and possibly not for NCre. Therefore, we designed additional sets (two for CCre) with one primer located within the insertion and one outside the homology arms (Fig. 6, sets b and c). These primers can also be used for sequencing (see Note 18). Using the T2A strategy, the splitCre fragments are incorporated into the same transcript as the target gene, and they share a stop codon. Therefore, the entire insert must be fully sequenced (see Note 19), to ensure (1) proper reading frames of the splitCre and the host gene, and (2) that no point mutations were introduced during the repair (troubleshooting see Notes 20, 21 and 22). Once a potential founder mouse has been identified, a second tissue sample should be taken and genotyped to ensure the correct individual has been identified (see Note 23).
Fig. 6 Example of genotyping strategy as used for the establishment of the Sall1ncre:Cx3cr1ccre:tdTomato line
494
Sigalit Boura-Halfon et al.
Primary Breeding
The initially edited founder mouse (F0) could potentially be mosaic and may carry cells of more than one genotype. This is because Cas cleavage and repair can take place after the first cell division, and often more than two versions of the edited chromosome are detected. In addition, not all alleles detected, even in apparent homozygotes, are transmitted in the germline. The founder F0 animals must be backcrossed to wild-type mice, and offspring needs to be genotyped to ensure the correct transmission of the insertion. It may require more than one litter to identify the correctly targeted allele. Once germ line transmission has been ascertained, a screening for off-target cleavage in the F1 pups is recommended. Potentially harmful off-targets should be screened by PCR (with at least 100 bp flanking both sides of the potential edit site) and sequencing to validate that no changes have occurred. If an off-target edit has occurred and is on a different chromosome, it can be bred away. The mouse line can then be maintained as any other classically derived mutant line. It is recommended to establish mouse lines from a few validated founders and maintain them as sublines. For more details on genotyping of CRISPR founder mice, see [61].
3.5 Establishment of splitCre Mouse Lines
Thus far, we described in detail how to establish the Cx3cr1ccre mice. However, since the splitCre system is based on the complementation of two inactive Cre protein fragments, it requires a generation of the NCre polypeptide under the control of a second relevant promoter of choice.
3.5.1 Considerations for Choice of the PromoterDriving NCre Expression
1. Keep in mind that a successful targeting of cells with the splitCre system occurs when expression patterns of the two transgenes show sufficient overlapping including time and level (threshold) (Fig. 2).
3.4
2. What are your target cells? What genes defined them best? 3. What is the pattern of expression of these genes? In the example we have used in this chapter, we wanted to dissect microglia from BAM and therefore asked what genes should be selected to drive NCre expression to best dissect between these two populations. It is recommended to use existing single-cell transcriptome databases that cover your cells of interest. For brain cells, the databases established by the Movahedi and Linnarsson teams (http://www.brainimmuneatlas.org; http:// mousebrain.org) [62, 63], and additional more comprehensive studies [2, 5, 11] showed that expression of Spalt Like Transcription Factor 1 (Sall1) discriminates microglia from other tissue macrophages, including BAM [64] (Fig. 7). Of note, Sall1 expression is, however, not confined to microglia but extends to certain mesenchymal cells [65] and non-immune CNS cells [37]. Conversely, expression of Lyve1,
SplitCre Transgenesis
495
Fig. 7 Choosing the promoter-driving expression of NCre fragment. Screenshot from single-cell database of mouse cortex indicating expression of selected genes relevant for Microglia and CNS border-associated macrophages (http://mousebrain.org). The size of the dot indicates the relative expression of the gene among all clusters. The value is the normalized expression level
encoding the hyaluronan receptor 1, is found on vesselassociated macrophages including BAM in the CNS, but absent from microglia [66]. Lyve1 is, however, also prominently expressed on lymphatic endothelium [67] and only the intersection with Cx3cr1 expression provides the macrophage focus. Once the candidate genes to drive the expression of NCre fragments are selected, the same guidelines and protocols used for the generation of the Cx3cr1ccre animals can be followed to generate the respective transgenic animals, either on wild-type background or to save time directly onto a Cx3cr1ccre background harboring a R26-LSL-reporter allele (herein Subheading 3). 3.5.2 In Vivo Validation— Breeding to a Reporter Strain
Each established mouse line must undergo a thorough validation, as was discussed in Subheadings 3.3 and 3.4. Validated males and females can be interbred to generate homozygous lines. In order to test the new splitCre mouse model, it is recommended to cross it to a reporter strain. We recommend using R26-LSL-tdTtomato animals, also called Ai14 mice (Jackson strain# 007909, B6;Cg-Gt (ROSA)26Sor tm9(CAG-tdTomato)Hze /J [68]), which are a good tool for spatial and temporal tracking of the specific cells in vivo, as they provide a strong marker for the initial screening
496
Sigalit Boura-Halfon et al.
and the allele is efficiently rearranged [69] (see Note 24). The F1 generation results in mice heterozygous for the Ncre, Ccre, and tdTomato genes. Analyze your mice by flow cytometry and imaging. If these animals present only low frequencies of tdTomato+ cells, we recommend to further breed the animals to generate homozygous Ccre and Ncre configurations. In the case of the Sall1ncre:Cx3cr1ccre: tdTomato and Lyve1ncre:Cx3cr1ccre:tdTomato mice, homozygosity significantly boosted the efficiency of the system (see Note 25) and establishing the mouse models as useful tool to distinguish parenchymal microglia from CNS macrophages (Fig. 8) [33]. Beyond imaging, splitCre mouse models can also be used to retrieve the translatomes of the respective cells by combining them with a RiboTag allele [28, 33, 70] (see also Chapter 17 in this issue of MiMb) (see Note 26).
4
Notes 1. Note that the GCN4 domain that promotes dimerization of the NCre and CCre Cre fragments can homodimerize. If the fragments do not show equimolar expression, this could result in the prevention of recombinase activity. Alternative dimerization motifs such as inteins could be explored [31], as well as additional domains supporting dimerization such as splitGFP [71]. 2. The quality of the injection buffer is critical to the survival of the microinjected embryos. A small number of 1-cell embryos should be microinjected with the injection buffer and cultured to blastocyst stage to assure quality of the buffer. Expected blastocyst rates are ~80%. 3. It is possible to prepare diluted Cas9 and the RNP mix in advance and store overnight at 4 °C, or for several days at 80 °C but we recommend preparing immediately prior to microinjection for best results. 4. According to current US regulations, constructs with P2A cannot be shipped outside of the USA. 5. If a different enzyme is chosen, the parameters of the guide design programs have to be changed accordingly. Cas12a for AT-rich regions (e.g., IDT: Alt-R® A.s. Cas12a (Cpf1) Ultra). Requires Cas12 compatible guide (e.g., IDT: Alt-R® A.s. Cas12a crRNA—no tracrRNA required). 6. The insertion of mutation in the PAM site is necessary to prevent subsequent recognition of the repair template as a target of Cas9-sgRNA. 7. Early studies describing CRISPR/Cas9 genome modification in mouse embryos, employed in vitro transcribed sgRNAs and
SplitCre Transgenesis
497
Fig. 8 SplitCre homozygosity improves rearrangement efficiency (a, b) and evidence that Sall1ncre:Cx3cr1ccre: tdTomato and Lyve1ncre:Cx3cr1ccre:tdTomato mice efficiently dissect microglia and BAM (c)
498
Sigalit Boura-Halfon et al.
Cas9 mRNA, usually generated in-house. However, the use of commercially supplied Cas9 ribonucleoprotein (RNP) mixes improves standardization and as a result greatly increases the ease and efficiency of in vivo targeting. 8. We use guide format crRNA:tracrRNA duplex, with no standard modifications. The XT modifications are not required. There is also the option to use single-guide RNA (sgRNA). 9. Since the RNP complexes include RNA, it is crucial to be scrupulous about RNase-free conditions at all stages (gloves, tips, tubes, solutions, laboratory utensils, etc.). 10. All animal procedure must be carried out in accordance with institutional and local regulations. 11. RNPs (commercial) perform better than Cas9 mRNA from earlier protocols. 12. Ordering and preparation of animals, and of CRISPR reagents should be coordinated between the core facility and the client, in advance. 13. The tip of the microinjection needle is typically Import > Pipeline from File” to open the default pipeline “ExamplePercentPositive” on CellProfiler (Fig. 1).
3
Methods
3.1 Creating the Pipeline for Automated Macrophage Quantification
1. Create a new pipeline by editing the modules in the default “ExamplePercentPositive” pipeline (Fig. 2). 2. Right-click on a module and select “Delete [module]” (Fig. 2a) to remove it from the pipeline. Delete the following modules: “RelateObjects,” “FilterObjects,” “MeasureObjectIntensity,” “ClassifyObjects,” and “CalculateMath.” 3. Click on the “+” symbol next to “Adjust modules” (Fig. 2a). Locate a module using the “Find Modules” search bar of the “Add Modules” window and click on “+ Add to Pipeline” (Fig. 2b). Add the following modules: “ColorToGray,” “MaskObjects,” and “SaveImages.” The red “X” next to any module will not affect the pipeline at this stage and can be ignored. 4. Arrange the modules in the order shown (Fig. 2c). Save the new pipeline by clicking on “File > Save Project as.”
Automated Cell Counting of Macrophages In Situ
507
Fig. 1 Accessing the default pipeline. (a) Use the CellProfiler 4.2.4 interface to access the “ExamplePercentPositive” imported pipeline and open (b) the menu of the imported default pipeline. Red boxes highlight the functions to be selected 3.2 Importing Images
1. This pipeline is for quantifying macrophages in 2D images or MIPs of stacks (see Note 1). It starts with the four fixed modules “Images,” “Metadata,” “NamesAndTypes,” and “Groups.” In the first module “Images,” drag .tif or other supported image files into the white space (Fig. 3a). 2. In the second module “Metadata,” select “No” to dissociate the input image metadata from the output image and its measurements (Fig. 3b). 3. In the third module “NamesandTypes,” assign the same name to every image to analyze all of them using the same parameters (see Note 2). In this example, the intensity range was set using “Image bit-depth” (Fig. 3c). 4. In the fourth module “Groups,” select “No” to analyze only one group of images (Fig. 3d; see Note 3). 5. In the fifth module “ColorToGray,” select the correct input image by its name, and set the conversion method to “Split” and the image type to “Channels” (Fig. 3e; see Note 2). The multichannel input image will be split to create one grayscale image per channel. It is possible to add as many channels as required (see Note 4).
3.3 Identification of Macrophage Soma and Nuclear Diameters
1. Check the first five modules to set the pipeline (Fig. 4a). 2. Click on the eye icon of the module “ColorToGray” to make the output of this module visible (Fig. 4a, b).
508
Ouze´na Bouadi and Tuan Leng Tay
Fig. 2 Creating the pipeline for automated macrophage quantification. Instructions to (a) delete and adjust modules, (b) find and add modules, and (c) create the new pipeline. Red boxes highlight the functions to be selected
Automated Cell Counting of Macrophages In Situ
509
Fig. 3 Settings for importing an image. The specific parameters of the (a) “Images” (red box indicates the area to drag-and-drop images), (b) “Metadata,” (c) “NamesAndTypes,” (d) “Groups,” and (e) “ColorToGray” modules are indicated. Settings are specific to the example image “IBA1-DAPI-MIP.tif” shown here
3. Click “Analyze Images” to generate the output of “ColorToGray” (Fig. 4a). 4. In the output window of “ColorToGray,” select the “Zoom” icon and draw a rectangle over the region of interest in the image to visualize representative macrophage somas and nuclei at higher magnification (Fig. 4b). Use the “Measure length” tool to measure their diameters. While the left mouse button is held down, the length (in pixel units) will appear in the bottom right corner of the wizard when the “Measure length” tool is in use. Use the “Home” tool to return the images to their original size (Fig. 4b; see Note 5).
510
Ouze´na Bouadi and Tuan Leng Tay
Fig. 4 Determine the range of diameters (in pixel units) of macrophage soma. (a) Initiation of new pipeline up to the “ColorToGray” module and clicking on “Analyze Images” (red box) to generate a (b) preliminary image to measure the diameter (blue arrow; “Length: 95.1,” red box) of representative macrophage somas in the “ColorToGray” output. This output shows the “Original image” and grayscale images of the two channels (“IBA1” and “DAPI”). Red arrows and annotations in (b) highlight the tools for measuring the soma diameter. Z stack of a mouse brain section immunostained with anti-IBA1 and Alexa Fluor 568-conjugated secondary antibody (red, “Original image”) and counterstained with DAPI (blue, “Original image”) was acquired on a ZEISS AxioImager.Z2 fluorescence microscope with Apotome 3 using a 20X/0.8 M27 objective lens and Zen 3.5 software. Zoom was set at 1 and binning at 1.1. Image stacks were sampled with a voxel size of 0.173 μm × 0.173 μm × 1 μm 3.4 Identification of Macrophages Using DAPI Masks
1. Use the measurements from Subheading 3.3. to set the range of diameters in pixel units for identifying DAPI-labeled nuclei (Fig. 5a) and IBA1-labeled macrophage somas (Fig. 5b; see Note 6). 2. Refine the range of diameters applied based on the generated parameters if the detected macrophages and their outlines do not match the objects in the input image (Fig. 5c). 3. Set the “MaskObjects” module (Fig. 6a) to count IBA1labeled macrophages that overlap with a DAPI-labeled nucleus (Fig. 6b). 4. Set “Number of resulting objects” to “Renumber” to renumber the objects that remain after masking with consecutive numbers. Set “Numbering of resulting objects” to “Retain” to maintain the original numbering of the objects that remain after masking (Fig. 6a).
Automated Cell Counting of Macrophages In Situ
511
Fig. 5 Identification of IBA1-positive macrophages and DAPI-labeled nuclei. Setting parameters in the “IdentifyPrimaryObjects” module for (a) nuclei and (b) macrophage somas. (c) Output of the “IdentifyPrimaryObjects” module for macrophage somas: “IBA1” channel of the input image (grayscale images, top and bottom left) with total detected objects (magenta outline) and saved objects (green outline), 11 IBA1-labeled macrophages saved by the pipeline (indicated in shades of red and blue, top right), and metrics on the saved macrophages (bottom right). Detailed explanation for each function or parameter in this module is provided in the question mark icon highlighted by the red box in (a)
512
Ouze´na Bouadi and Tuan Leng Tay
Fig. 6 Identification of IBA1- and DAPI-labeled macrophages using the mask option. (a) Setting parameters in the “MaskObjects” module. The red box highlights the numbering options “Renumber” and “Retain” for detected macrophages in the final output. (b) Output of the “MaskObjects” module: IBA1-labeled objects saved by the pipeline (colored, left panel); IBA1- and DAPI-labeled macrophages saved by the pipeline (colored with green outline, right panel), IBA1-labeled objects without DAPI (magenta outline, right panel), and DAPI masks (grey, right panel). Here, the letters a to k were added to both panels to highlight the omission of IBA1labeled objects e and k in the “MaskedMacrophages” panel and which will not contribute to the final output of the pipeline
Automated Cell Counting of Macrophages In Situ
3.5 Saving Output Images and Data Tables
513
1. Create output images using the “OverlayOutlines” (Fig. 7a) and “DisplayDataOnImage” (Fig. 7b) modules. 2. Save output images in the desired format (e.g., JPEG) in the “SaveImages” module (Fig. 7c). 3. To export only specific measurements, select “no” for “export all measurement types” (Fig. 8a). 4. Click on “Press the button to select measurements” to select desired measurements for export (Fig. 8a, b). 5. Check all the modules (Fig. 7) and click on “Analyze Images” (Fig. 4a) to run the pipeline. All output JPEG images and one . csv file containing all the quantified macrophages can then be found in the destination folder (Fig. 9).
4
Notes 1. See GitHub (https://github.com/CellProfiler/tutorials/ tree/master/3d_monolayer) for details on analyzing 3D images using CellProfiler. 2. When processing an image, a new name can be assigned in the following modules: “Name to assign these images” (Fig. 3c), “Image name” (Fig. 3e), “Name the primary object to be identified” (Fig. 5a, b), “Name the masked objects” (Fig. 6a), “Name the output image” (Fig. 7a), and “Name the output image that has the measurements displayed” (Fig. 7b). For example, in this chapter, the new grayscale images created in the module “ColorToGray” were renamed “DAPI” and “IBA1” (Fig. 3c). Thereafter, all images must be indicated as such in the drop-down menus: “Select the input image” (Figs. 3e and 5a, b), “Select the objects to be masked” (Fig. 6a), “Select the masking object” (Fig. 6a), “Select the image on which to display outlines” (Fig. 7a), “Select objects to display” (Fig. 7a), “Select the input objects” (Fig. 7b), “Select the image on which to display the measurements” (Fig. 7b), and “Select the image to save” (Fig. 7c). For example, the images “DAPI” and “IBA1” were selected as input images of the module “IdentifyPrimaryObjects” (Fig. 5a, b). The pipeline will fail if naming conventions are inconsistent (e.g., use of incorrect case or symbols) because the downstream modules are executed on previous output images. An error is indicated by a red cross in place of a green tick or an empty box assigned to the module. The error message can be displayed by hovering the cursor over the red cross. 3. In the module “Groups,” a collection of images containing different objects for quantification can be split into groups to be analyzed using the downstream modules of the same
514
Ouze´na Bouadi and Tuan Leng Tay
Fig. 7 Creating and saving an output image. Parameters for (a) “OverlayOutlines,” (b) “DisplayDataOnImage,” and (c) “SaveImages” modules
Automated Cell Counting of Macrophages In Situ
515
Fig. 8 Creating and exporting a datasheet. Setting parameters for (a) “ExportDataSheet” module and (b) “Select measurements” wizard. Red boxes highlight the functions to be selected
pipeline to generate results specific to each object. Groups are formed based on the names assigned to the images in the module “NamesandTypes.” Set “Assign a name” to “Image matching rules” to match images based on the file name, directory, type, and metadata. For instance, this function can be used to match macrophages to counts of puncta from fluorescent in situ hybridization within immunolabeled or fluorescent-tagged macrophage soma. 4. When adding multiple channels for the analysis of macrophage populations that express a combination of markers, the user should ensure that the channels correspond to the acquired channels in the original image. The order of the channels can be different from the order when the image is opened in other software such as Fiji [8] or ImageJ [9]. 5. Setting the range of soma diameters (in pixel units) to accurately detect objects is crucial. An excessive range of diameters will lead to the false-positive detection of too many objects. Measure the soma diameter of several representative
516
Ouze´na Bouadi and Tuan Leng Tay
Fig. 9 Output of pipeline. Final output as (a) JPEG image and (b) .csv file obtained with the “Renumber” option, and (c) JPEG image and (d) .csv file obtained with the “Retain” option. (b, d) Column A indicates the image number. Column B indicates the object number of IBA1+ macrophage soma. Column C indicates the object number of IBA1+ DAPI+ macrophages. The term “nan” for “not a number” is generated when an IBA1+ macrophage does not colocalize with DAPI signal. These numbers are displayed in the JPEG image in (a) and (c). Column D indicates the object number of the parent macrophage that matches the IBA1+ DAPI+ macrophage in column C. (a, c) Here, the parent IBA1+ macrophage #6 is renumbered as #5 in (a, b) and is retained as #6 in (c, d) (white arrowheads), while the parent IBA1+ macrophage #5 is not displayed in (a) or (c)
macrophages of different sizes to acquire an accurate range of diameters. To obtain macrophage soma diameters from different images, select the images one at a time using the “Filter images?” option in the “Images” module (Fig. 10). “Filter images?” allows the selection of one image or a group of images for subsequent visualization or analysis using the pipeline. To work with multiple images, select “Custom” in the drop-down
Automated Cell Counting of Macrophages In Situ
517
Fig. 10 Selection of working images in the imported file list. (a) The custom option in the “Images” module allows the selection of files that will go through the pipeline. (b) Define the criteria for the images to be analyzed using “Select the rule criteria” and clicking on “Apply filters to the file list.” Select “Show files excluded by filters.” The file names in black will be used in the analysis, while the file names in gray will be excluded by the pipeline. Red boxes highlight the crucial functions in this module
menu of “Filter image?” (Fig. 10a). Set the rule criteria as desired under the “Custom” option (Fig. 10b). This applies for the visualization and measurement of multiple representative macrophages across multiple images in the identification of macrophage soma and nuclear diameters. When working with several images, only the last image will be visible in the display window. 6. This automated cell counting pipeline is optimized for images containing separated, non-overlapping objects. In histological sections containing complex cell morphologies or high cell densities, it is recommended to use various combinations of macrophage markers (e.g., CD163, CD169, F4/80, CD68, P2RY12, or IBA1) and nuclear counterstain to create specific “MaskedMacrophages” for quantifying each distinct macrophage population (Figs. 6, 7, 8, 9, and 10).
518
Ouze´na Bouadi and Tuan Leng Tay
Acknowledgements The authors thank S. Guerra, K. He, E. Kraft, and Z. Malone for validating the accuracy and speed of the cell quantification pipeline. O. Bouadi was supported by the Dean’s Fellowship from the Boston University Graduate School of Arts & Sciences. References 1. Ginhoux F, Greter M, Leboeuf M, Nandi S, See P, Gokhan S et al (2010) Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science 330:841–845 2. Gomez Perdiguero E, Klapproth K, Schulz C, Busch K, Azzoni E, Crozet L et al (2015) Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature 518:547–551 3. Hagemeyer N, Kierdorf K, Frenzel K, Xue J, Ringelhan M, Abdullah Z et al (2016) Transcriptome-based profiling of yolk sac-derived macrophages reveals a role for Irf8 in macrophage maturation. EMBO J 35:1730– 1744 4. Hashimoto D, Chow A, Noizat C, Teo P, Beasley MB, Leboeuf M et al (2013) Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes. Immunity 38:792–804
5. Mass E, Ballesteros I, Farlik M, Halbritter F, Gu¨nther P, Crozet L et al (2016) Specification of tissue-resident macrophages during organogenesis. Science 353:aaf4238 6. Stirling DR, Swain-Bowden MJ, Lucas AM, Carpenter AE, Cimini BA, Goodman A (2021) CellProfiler 4: improvements in speed, utility and usability. BMC Bioinformatics 22:433 7. Tay TL, Mai D, Dautzenberg J, Ferna´ndezKlett F, Lin G, Sagar et al (2017) A new fate mapping system reveals context-dependent random or clonal expansion of microglia. Nat Neurosci 20:793–803 8. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T et al (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9:676– 682 9. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH image to ImageJ: 25 years of image analysis. Nat Methods 9:671–675
Chapter 34 Morphometric Analyses of Macrophages Jan N. Hansen Abstract Cell morphology and motility drive the cellular capabilities to interact with the environment. For example, microglia, the longest known tissue-resident macrophages, show a highly branched process tree with which they continuously scan their environment. Computational image analysis allows to quantify morphology and/or motility from images of tissue-resident macrophages. Here, I describe a step-by-step protocol for analyzing the morphology (and motility) of macrophages with our recently described, freely available software MotiQ, which provides a broad band of parameters and thereby serves as a versatile tool for studies of morphology and motility. Key words Morphometric analysis, Morphology, Motility, Image analysis, Macrophages, Microglia, ImageJ
1
Introduction Cell morphology and motility are important indicators for the interaction of cells with their environment. For the longest known tissue-resident macrophages, microglia, it is known that they feature a complex morphology, with a highly branched process tree, and that they continuously scan their environment [1, 2]. They remodel their process tree and motility, reacting to pathological changes in their environment (e.g., [2]). We have developed the software MotiQ to study microglial morphology and motility [3]. MotiQ is not restricted to microglia only but, in principle, can be applied for studying morphology and motility of any cell type and, thus, also for other tissue-resident macrophages. MotiQ has been developed and refined over a long time, with the aim to develop software that (a) is automatized and little user-biased, (b) provides a nearly complete list of parameters applied to date for morphology and motility, and (c) is freely available and broadly accessible. To date, the workflow has been successfully applied by many different users, to data sets from different labs (e.g., [4–12]). Here, I describe a step-by-step guide
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_34, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
519
520
Jan N. Hansen
on how to analyze the morphology and motility of tissue-resident macrophages, from acquiring the images to image analysis with MotiQ. The step-by-step guide can be applied by any laboratory since all software is freely available and no coding skills are required. Similar analysis workflows can also be achieved with alternative solutions, but these either involve tedious process tracing, provide a limited number of parameters, are restricted to 2D images, require coding skills, or include commercial software (for a detailed discussion, see [3]). In contrast, MotiQ largely automatizes image analysis, which allows to analyze big data sets and outputs a broad band of parameters, which makes it applicable for nearly any study question related to morphology, motility, and fluorescence intensity: (a) morphology parameters, which report, e.g., the cellular volume, surface, shape complexity, process tree properties, and polarization of the process tree; (b) motility parameters, which indicate the movement and scanning activity of the cell; and (c) intensity parameters, which allow to correlate morphology and motility with signaling events in live experiments, with cells that are labeled with indicators such as biosensors or calcium dyes.
2
Materials The image analysis is performed on a standard computer, requiring Internet access for installations, and running a Windows, Mac, or Linux operating system (see Note 1). This materials section lists the preparations to be performed before running a MotiQ analysis.
2.1 Installing FIJI and the MotiQ Plugins to Your Computer
1. Go to https://imagej.net/software/fiji/index and download the FIJI version suiting your operating system. 2. Extract the downloaded .zip file: (a) On Windows: Right-click > “Extract All. . .”. (b) On Mac OS: Double-click on the .zip file. (c) On Linux: Right-click > “Extract Here.” 3. The downloaded file does not need to be installed but can be used as is. Place it on your computer as follows. (a) On Windows or Linux: Place the extracted “fiji-” directory into any folder that you have read and write permission to (e.g., your Desktop or Documents folder). (b) On Mac OS: Place the extracted “fiji-macosx” directory into your Documents folder (do not place it in the “Applications” folder to avoid issues triggered by “path randomization” performed by Mac OS).
Morphometric Analyses of Macrophages
521
4. Launch FIJI. (a) On Windows: Enter the “fiji-win” directory, then enter the “Fiji.app” directory, and double-click on the “ImageJ-win.exe” file. (b) On Mac OS: Enter the “fiji-macosx” directory and double-click on the “FIJI.app” icon. (c) On Linux: Enter the “fiji-linux64” directory, then enter the “Fiji.app” directory, and double-click on the “ImageJlinux64” file. 5. In FIJI, go to the menu entry “Help > Update. . .” and wait for the upcoming dialog showing a progress bar with “Checksummer” to finish. An “ImageJ Updater” window will pop up (see Note 2). 6. Click on “Manage update sites” in the “ImageJ Updater” window, which will display a list of update sites. 7. Scroll down the list and select the update site with the Name “MotiQ.” 8. Click “Close” to close the list. 9. Click “Apply changes” in the “ImageJ Updater.” 10. Wait for the installation to finish and for the software to ask you to restart the software. 11. Close FIJI and restart it as described above. 2.2 Preparation of Input Images
Morphometric analysis is performed based on images of fluorescently labeled macrophages, acquired with a confocal/fluorescence microscope. 1. Before image acquisition, the macrophages need to be fluorescently labeled in your sample (see Note 3 and other protocols in this book, e.g., Chapters 15, 20, and 21). 2. Acquire digital images with a confocal microscope (recommended for most images, especially for 3D images) or fluorescence microscope (can be sufficient for 2D images) (see Note 4 and other protocols in this book, e.g., Chapter 15). 3. From the software controlling the microscope, you can try to export each recording into a single .tif file containing only one channel (the channel that labels the entire macrophages) and, for 3D and time-lapse images, containing all slices and frames (see Note 5). If this is possible, continue at Subheading 2.3. If this is not possible, continue here at step 4. 4. Save the images in the standard microscope format provided by your microscope software (see Note 6). 5. Launch FIJI.
522
Jan N. Hansen
6. Load one of the image files of your data set in FIJI via the BioFormatsImporter integration: Drag and drop the file into the FIJI status bar (in the lower left corner of the window of the FIJI software). 7. If a “Bio-Formats Import Options” dialog is shown now: Select “View stack with: Hyperstack,” “Color mode: Composite,” unselect all other check boxes except for the “Autoscale” checkbox, and press “OK.” 8. If a “Bio-Formats Series Options” dialog is shown now: Select the “Series” (= image recordings) that you want to analyze. 9. Select the frontmost image (if multiple images are open) by clicking on its window. 10. If the image contains multiple channels (this is the case if a bar labeled with a “C” is displayed below the image, allowing you to switch between different channels), we need to split it into single-channel images (see Note 5). To this end, click on the menu entry “Image > Color > Split Channels” to burst the image window into separate windows for each channel, close the channel windows that cannot be used for MotiQ analysis, and click on the window that contains the channel to be used for detecting cells with MotiQ. 11. Save the image as a .tif file via the menu entry “File > Save.” 12. Close the saved image. 13. Repeat steps 9–12 for all other images that are still open in FIJI. 14. Repeat steps 6–13 for all other image files to be analyzed in MotiQ. 15. Close FIJI. 2.3 Validate Input Image Files
If you are working with images with multiple slices (3D) or frames (time-lapse) or both, validate that the .tif files contain stacks that are correctly formatted as follows. Otherwise, skip this step and move on to Subheading 2.5. 1. Launch FIJI. 2. Open multiple representative example images through drag and drop into FIJI’s status bar or through the menu entry “File > Open.” 3. Select an opened image window by clicking on it. 4. Click on the menu entry “Image > Hyperstacks > Re-order Hyperstack.” 5. Validate in the upcoming dialog that the number of slices, frames, and channels is correct.
Morphometric Analyses of Macrophages
523
6. If the numbers are incorrect, press “OK,” close all images, and proceed at Subheading 2.4. If the numbers are correct, press “OK,” and proceed here. 7. Close the current image and repeat step 3–6 for all other opened images. 8. Close FIJI. 9. Skip materials Subheading 2.5. 2.4 Re-order Hyperstack in .tif Files (If Applicable)
Subheading
2.4
and
proceed
with
Perform the section under this Subheading only if step 5 in Subheading 2.3 yielded incorrect numbers. For all .tif files in your data set, you need to perform the following: 1. Open the image. 2. Click on the menu entry “Image > Hyperstacks > Re-order Hyperstack.” 3. Use the top 3 selection boxes in the “Re-order Hyperstack” dialog to interchange the definition of channels, slices, and frames so that each stack dimension is correct (i.e., the number of frames that your image should contain is written at “frames,” the number of slices that your image should contain is written at “slices,” the number of channels is “1”) and press “OK.” 4. Click on the menu entry “File > Save. . .”. 5. Close the image.
2.5 Inspect the Images to Find Good Settings
Before running the whole protocol in Heading 3 on your entire data set, it is recommended to inspect your images and test different MotiQ Thresholder and MotiQ 3D Analyzer settings manually so that you know which settings to select when you start the analysis described under Heading 3. You may also try MotiQ Thresholder and MotiQ 3D Analyzer under default settings and validate whether this, by chance, gives good results. However, to find optimal settings, it is recommended to go through the following steps, guiding you to perform several tests on your images to find parameters for MotiQ Thresholder and MotiQ 3D Analyzer. 1. Follow the step-by-step guide to fine-tune parameters to be used in MotiQ Thresholder, provided at the MotiQ wiki (https://github.com/hansenjn/MotiQ/wiki/Find-MotiQThresholder-Settings). The guide in the wiki is always updated for the latest version of MotiQ Thresholder. 2. For a few example images, perform Subheadings 3.1 (if applicable, see below) and 3.2, and with these test images, follow the step-by-step guide to fine-tune settings for MotiQ 3D Analyzer, provided at the MotiQ wiki (https://github. com/hansenjn/MotiQ/wiki/Find-MotiQ-3D-AnalyzerSettings). The guide in the wiki is always updated for the latest version of MotiQ 3D Analyzer.
524
3
Jan N. Hansen
Methods I here describe a protocol for the analysis of a confocal 3D image (stack of multiple slice images) of fixed, fluorescently labeled macrophages. SeeNote 7 for key information on customizing this protocol to a 2D image analysis. SeeNote 8 for key information on customizing this protocol to a time-lapse analysis. The protocol has been developed for the following MotiQ release: https://github. com/hansenjn/MotiQ/tree/v0.2.0. Information on changes in potentially newer versions of MotiQ will be noted on the MotiQ GitHub repository: https://github.com/hansenjn/MotiQ. Subheading 3.1 can be time-consuming. It is only necessary to perform this step if cells in the images are getting very close to each other (see Note 9). If this is not the case, you can skip this step and proceed at Subheading 3.2. Screenshots of the processes explained below are available in the Manual for MotiQ (https://github.com/ hansenjn/MotiQ/tree/master/Manual). In case you are not performing the steps below with an example image while reading the chapter, it can be helpful to look at the screenshots in the Manual to understand the different steps.
3.1 Cropping of Individual Cells
1. Start FIJI. 2. Open the image to be analyzed (it needs to be a single-channel . tif file containing all slices and frames belonging to the image (see Subheading 2.2 and Note 5)) in FIJI by drag and drop of the image into the status bar of FIJI or by using the menu entry “File > Open. . .” 3. Launch MotiQ Cropper via the menu entry “Plugins > MotiQ > MotiQ Cropper.” 4. In the upcoming settings dialog select for “Image:” the option “process the active, open image” and select for “Process stack in:” the option “user-defined order. . .”. 5. Now a dialog pops up that asks you to draw an “ROI” (= region of interest = a selection). Focus on the first cell you aim to select and inspect where the soma/nucleus of the cell is and what protrusions belong to the cell by moving across the different image slices (see Note 10). 6. Draw a polygon circumscribing the cell soma and the protrusions of the cell in a slice of your choice (see Note 11). 7. When the polygon is accurately circumscribing the cell soma and cell protrusions in the displayed slice image, press “OK” in the dialog and confirm the upcoming dialog that suggests to “CONTINUE setting ROIs in next stack image (2)” (see Note 12 for an explanation of the other options) by pressing “OK.”
Morphometric Analyses of Macrophages
525
8. Now, in the currently displayed slice image, MotiQ cropper will automatically set all pixels outside of the selection to zero and move to the closest slice image. It now displays the ROI previously applied. 9. Adjust the ROI so that in this new slice all parts of the cell are still within the ROI. 10. Repeat steps 7–9 until all slice images have been processed. In empty slice images (i.e., slices where the cell to be selected is not visible), do not set any selection and press “OK” to clear all pixels in that slice (see also Notes 11 and 12). 11. MotiQ Cropper automatically saves a .tif file for the selected cell at the file directory where the input image was stored, along with a metadata text file and the selections in a .zip file (see Note 13). 12. Now MotiQ Cropper asks you whether you want to “cut out” (= select) more cells from the same image. If yes, press “Yes” and repeat steps 5–12 for each cell in the image. After selecting the last cell, reply “No” to finish the selection process (for more information on which and how many cells to select, see Note 14). 13. Perform steps 2–12 for all other images to be analyzed. 14. Close FIJI. 3.2 Segmentation of the Images
1. Launch FIJI. 2. Launch MotiQ Thresholder via the menu entry “Plugins > MotiQ > MotiQ Thresholder.” 3. In the upcoming dialog, select “multiple images . . .” for the setting “Process.” 4. If you did not perform Subheading 3.1, unselect the option “Use alternate reference image.” If you performed Subheading 3.1, make sure the option “Use alternate reference image” is enabled and leave the settings nested under this option to default parameters. 5. Select the options for pre-processing and for the threshold algorithm that are best applicable to your data set (see Note 15). If you analyze a 3D image, select under “Stack handling:” the option “apply threshold determined in a maximum-intensity-projection.” If you analyze a 2D image, select for “Stack handling” the option “no stack processing.” Leave all other settings to default parameters. 6. Press “OK.” 7. The “Multi-Task-Manager” dialog pops up. 8. Press “add files.”
526
Jan N. Hansen
9. In the upcoming dialog, navigate to the files that you want to process and select them (i.e., the files produced in Subheading 3.1, ending with “_CUT__X_Y_Z.tif,” or if Subheading 3.1 was skipped, the .tif files prepared in Heading 3 above). 10. Press “Open.” 11. If there are more files to be selected and processed, repeat step 8–10 until all files to be processed are listed in the “MultiTask-Manager.” 12. Press “start processing.” 13. Another Multi-Task-Manager window pops up showing the progress of processing. MotiQ Thresholder will now process one image after the other, which can be time-consuming (see Note 16). When all images are processed, the progress bar in the middle of the dialog shows “analysis done!”. 14. After processing, MotiQ Thresholder automatically saves the output file with a new file ending (“_pBIN.tif”) into the directory in which the input image was stored. 3.3 Validating the Segmented Images
1. Validate the produced “. . .pBIN.tif” files as follows to check whether the segmentation performed by the MotiQ Thresholder was successful. 2. Open some of the produced “. . .pBIN.tif” files in FIJI by drag and drop of the files into the FIJI status bar or by opening them via the menu entry “Files > Open.” 3. Display the “Threshold” dialog in FIJI via clicking on the menu entry “Image > Adjust > Threshold.” 4. Click on an opened image, and then click on the opened “Threshold” window. 5. Make sure that the box “Dark background” is checked, and that the "Display type" (this is the right box where you can select different options) shows “Over/Under.” 6. Click “set” in the “Threshold” dialog, set the “Lower Threshold Level” to 1, and press “OK.” 7. Now you can see what was detected as foreground (original intensity values from the image) and what was detected as background (blue). Check that the segmentation (the “detection of the cells”) is correct (as described in Note 17). 8. Perform steps 4, 6, and 7 for each of the other opened images. 9. If all images checked were correctly segmented (see Note 17), close all images without saving them (if you are asked upon closing), close FIJI, and move on to Subheading 3.4.
Morphometric Analyses of Macrophages
527
10. If some images were not correctly segmented, you need to either exclude them from further analysis or optimize MotiQ Thresholder settings (see Note 15) and repeat Subheadings 3.2 and 3.3 with the optimized settings. 11. Close all images without saving and close FIJI. 3.4 Analysis of the Images
1. Launch FIJI. 2. Launch MotiQ 3D Analyzer via the menu entry “Plugins > MotiQ > MotiQ 3D analyzer” (see Note 7 in case you want to adapt the workflow for 2D images). 3. Enter the settings that you found suitable for your data set (see also Subheading 2.5). (a) If your image is not calibrated (see Note 18), check the box “Re-calibrate image” and enter calibration settings nested under that checkbox. Otherwise, uncheck that box. (b) Enter particle filtering settings according to your type of image and segmentation to exclude particles that are too small to represent a part of the cell/a cell process (see also Note 17). (c) If you exclude Subheading 3.1, enter at “Calculate results for” the option “every particle separately.” (d) If you applied Subheading 3.1, enter at “Calculate results for” the option “all detected particles merged into one object” (see Note 19). (e) Eventually, adjust the values for the “Gauss filter prior to skeletonization” depending on your images/study paradigm (see Note 20). (f) Adjust the “Number format” according to what type of decimal separator your OS/your table calculation software (used in Subheading 3.6) is using. 4. Press “OK.” 5. The “Multi-Task-Manager” dialog pops up. 6. Press “add files,” navigate to the files that you generated in Subheading 3.2 in the upcoming dialog, and select the files ending with “_pBIN.tif,” and press “Open.” 7. If there are more files to be analyzed, repeat step 6 until you listed all files to be processed. 8. Press “start processing.” 9. The current window closes, and another Multi-Task-Manager window pops up showing the progress of analysis. MotiQ 3D Analyzer will now process one image after the other, which can be time-consuming (see Note 16). When all images are
528
Jan N. Hansen
processed, the progress bar in the middle of the dialog shows “analysis done!”. MotiQ 3D Analyzer then has saved the results files with a new file ending in the folder where the files to be analyzed were stored. These files are explored in Subheading 3.5. 3.5 Validating Results
1. Validate the analysis results as follows: Enter the folder saved by MotiQ 3D Analyzer for each analyzed file (the folder name of the output folder is the analyzed file and the added ending “_M3D”) and open the files ending with “3D.tif” in FIJI (“. . .RP. . .-3D.tif,” “. . .H-3D.tif,” and “. . .Skl-3D.tif”). These files show the detected cell, convex hull, and cell skeleton and allow you to verify that the MotiQ 3D Analyzer detected the cells correctly (see Note 21). If you are missing the 3D files (see also the wiki entry https://github.com/hansenjn/ MotiQ/wiki/3D-visualizations-missing), you may use the non-3D files instead for validation (“RP. . .tif,” “. . .H.tif,” “. . .Skl.tif”). Detailed descriptions on how to scrutinize the output files are given with screenshots also in the guidelines for MotiQ 3D Analyzer settings at the MotiQ wiki (https:// github.com/hansenjn/MotiQ/wiki/). 2. If the detected cell and skeleton are grossly incorrect, repeat Subheadings 3.2 to 3.5 or 3.4 to 3.5 with improved settings for the MotiQ Thresholder and/or MotiQ 3D Analyzer (see also Notes 15 and 21). Otherwise, proceed to the next step.
3.6 Convolving Analysis Results
We here show the most simple, manual way for merging the result data from a MotiQ analysis into a separate table for each condition analyzed. These tables can then be used to make additional plots and interpretations (see Subheading 3.7). For researchers experienced in programming, it may be more straightforward to develop a computer code to read the results files and merge them into tables. 1. Open an empty table sheet in software of your choice (e.g., an empty “Excel” document, “Libre Office Calc” document, or “Google Sheets” document). 2. Enter any folder generated by MotiQ 3D Analyzer in the current analysis (folder name ending with “_M3D”). 3. Open the file M3Dl1L.txt in that folder with a text editing tool (e.g., Notepad on Windows, TextEdit on Mac, or gedit on Linux) (see Note 22). 4. Copy all content to the clipboard (e.g., on Windows, press Ctrl + A, and then Ctrl + C; on Mac, press Cmd + A, and then Cmd + C). 5. Paste the content into the spreadsheet at cell A1 (click into cell A1 and press Ctrl + V on Windows or Cmd + V on a Mac). Since
Morphometric Analyses of Macrophages
529
the pasted content is tab-delimited, the pasted content will automatically be spread across multiple rows and columns. Using the content of M3Dl1L, you now created the table header in your table sheet. The table header allows you to later see which column represents what parameter. In the next steps, we will add the analysis results. 6. If you are analyzing images that contain only one time point (non-time-lapse images), delete the second row in the spreadsheet (see Note 22). 7. Close the M3D1L.txt file. 8. For each analyzed file, MotiQ 3D Analyzer has saved a text file outside of the generated “. . .M3D” folder named with the file name of the analyzed file and the ending “. . ._M3Dl1.txt.” Start with the first of these files as follows. 9. Open the corresponding “_M3Dl1.txt” file as the previous text file (hints given in step 3 above). 10. Copy all content (hints for copying are given in step 4 above). 11. Paste all content to the next free row in the spreadsheet (hints for pasting given in step 5 above). 12. Close the text file. 13. Perform steps 9–12 for all other generated “_M3Dl1.txt” files belonging to the same condition. 14. Now you have generated a table that contains all analyzed individual cells as rows, with cell parameters in columns. Save the spreadsheet under the name of the condition for which you merged the data in this table. 15. Repeat steps 1–14 for all other conditions in your data sets. Thereby you will have generated a results table with all results parameters for each condition you are analyzing. 3.7 Data Interpretation
The tables generated in Subheading 3.6 can now be used to produce additional, tables, plots, and analysis results comparing individual parameters of interest between different conditions. Detailed descriptions and examples for each parameter are given in the MotiQ publication [3]. If additional parameters will be added in the future, they will be described on the GitHub webpage: https:// github.com/hansenjn/MotiQ). For example, to study the morphology of a microglial cell, it is recommended to at least investigate the following parameters and compare them between different conditions in your experiment: 1. Ramification Index, which indicates the complexity of the cell shape (1 for a perfect sphere, a higher index refers to a more complicated cell shape).
530
Jan N. Hansen
2. Spanned volume (convex hull), which indicates the volume that a cell may “investigate”/“span” with its processes. 3. Polarity index (binary), which indicates whether the cell distributes cellular protrusions into all directions equally or is stretching toward any direction (0 for the case that the amount of membrane protrusions is balanced when comparing opposite sides of the cell; the higher the value the less similar is the amount of protrusions on opposite sides of the cell, indicating that more protrusions are directed toward a specific side of the cell). 4. Skeleton parameters characterizing the process tree (# branches (all skls), # junctions (all skls), # tips (all skls), tree length (all skls), average branch length (all skls), maximum branch length (all skls)). 3.8 Reduce the File Size of a Finished MotiQ Analysis for Long-Term Storage
The MotiQ analysis provides a stepwise process that allows intermediate validation of the results based on many generated image files. All the files produced by MotiQ can amount to a large file size. A finished MotiQ analysis may be reduced in file size by removing files that can at any time point be regenerated by using the same MotiQ version and the same settings (all applied settings can be looked up in the generated text files later and thereby the analysis can be reproduced). This section provides a guide for cleaning up your MotiQ analysis files and keeping only the minimum required files to reproduce the analysis. However, after running this process, the remaining files do no longer allow any validation of the results unless the image files are regenerated again by running Subheadings 3.2 and 3.4 under the same settings and with the same plugin versions as previously. However, the resulting text files are kept and allow to look up settings and analysis results at any time. Perform the following steps only, if you have completed the analysis of the data set and if you have produced all required figures and plots. Make sure to not accidentally delete the wrong files. 1. Enter the folder system that contains the MotiQ analysis that you want to reduce in size only. 2. Search for all files with the ending “_pBIN.tif” and delete them (these can be reproduced by running Subheading 3.2 again). 3. Search for all folders with the ending “_M3D” and delete them (these can be reproduced by running Subheading 3.4 again). 4. The remaining files can be compressed to further reduce the file size (see Note 23).
Morphometric Analyses of Macrophages
4
531
Notes 1. The performance of the computer will determine the limitations of the analysis. The memory/“RAM” of the computer may, if too small, limit the size of images you can process. The processor speed determines the analysis speed. All MotiQ plugins do not perform multi-core processing, and thus, the analysis speed depends on the speed of a single core of the CPU. However, most analysis steps are automated, allowing you to run analyses outside of working hours, e.g., overnight. 2. The Wiki in the GitHub repository for MotiQ (https://github. com/hansenjn/MotiQ/wiki) shows screenshots for the whole installation process. 3. Fluorescent labeling refers to that the macrophage either (a) expresses a genetically introduced fluorescent protein (e.g., GFP or mCherry) or fusion constructs thereof, or (b) was loaded with a dye prior to recording (e.g., calcium or membrane dyes) that labels the macrophage, or (c) was stained with an immunofluorescent staining labeling the macrophage based on a primary antibody against a protein expressed by the macrophage. The antibody/fluorescent protein/dye needs to label the entire macrophage and all its protrusions (at least the entire macrophage membrane). 4. Confocal microscopy provides the best conditions for later image analysis of morphology and motility with MotiQ since you can achieve a high axial and lateral resolution and avoid of-focus blur in your image, which can hamper to realize a good segmentation with MotiQ Thresholder (see Subheading 3.2 and adjacent Notes). To later analyze the images in MotiQ, a high-resolution objective (e.g., 40× or 63×) should be applied. In the confocal microscope, laser intensities, detectors, pixel size (the distance between two adjacent pixels in the real world; e.g., defined by how fine the scanning is), and more parameters can be fine-tuned to produce a good-quality image. For later MotiQ analysis, the following advice should be considered: (a) If you aim to analyze intensity values with MotiQ, avoid any over- and under-saturation of the acquired images. (b) In theory, it is optimal for the analysis to acquire the best “resolved” images, with the smallest possible pixel size (also called “image resolution” or “image calibration”) since that will give you the most accurate results and allow you to best separate adjacent cells later in the analysis. However, this comes with the compromise that the larger the image file/the more pixels per image, the longer the acquisition time and analysis time become. Thus, it is only beneficial to invest more recording and analysis time if you really plan to look at very small
532
Jan N. Hansen
changes between conditions or have very densely populated tissues/cultures. (c) Recording an image with finer pixel size in a scanning confocal microscope may increase the bleaching of the fluorophores in the sample, which can be especially disadvantageous for low-signal samples, where the bleaching may result in an intensity decrease across slices/ frames in the final image, making image analysis more challenging. (d) For samples providing a low signal-to-noise ratio, you may rather consider line averaging or binning than scanning at a smaller pixel size, since that will give you better quality images. Although this will result in more coarse analysis results compared to a smaller pixel size, it is better to later get coarse but good quality results than to struggle with the analysis approach and end up with bad-quality results. In summary, as usual in microscopy, it is required to find a good compromise between pixel size, signal-to-noise ratio, and acquisition time. For image analysis, it is, however, most important to obtain a good signalto-noise ratio and avoid bleaching since this provides the best ground for a successful image analysis. Note, however, that there is one limit for the pixel size for later MotiQ analysis: Any process of the cell needs to be at least two pixels “thick,” since otherwise, automated detection of continuous cell processes may be difficult for MotiQ. Thus, a recommended pixel size for the study of cell processes will be at the resolution limit ( MotiQ > MotiQ 2D analyzer.” The settings and output files are like those in MotiQ 3D Analyzer. The output folder and files, however, are named with the ending “M2D. . .” instead of “M3D. . .”. MotiQ 2D Analyzer does not output 3D files for inspection as described in Subheading 3.5. Instead, the files with the Endings RP.tif and Skl.tif can be used to inspect the detected cell and the detected skeleton in Subheading 3.5. 8. For time-lapse analysis, all steps can be performed as described except for few exceptions: (1) If you analyze a time-lapse 2D image, select in Subheading 3.1, step 4, for “Process stack in” the option “consecutive order. . .”. (2) Make sure to select “Threshold every time-step separately” in MotiQ Thresholder if you analyze images with fluctuations in the intensity level of the analyzed cell (e.g., in case you are labeling the cells with a dye or sensor to report signaling activity or if you observe strong bleaching of the fluorophore over time). If the intensity signals are stable, unselecting this option is advantageous. If you unselect this option and process a 2D image, select “apply threshold determined in the stack histogram” at “Stack handling:”. (3) If the image is not correctly calibrated (see
534
Jan N. Hansen
also Note 18), make sure to set the correct time interval and time unit and recalibrate the image in MotiQ 2D/3D Analyzer. (4) Select a number for grouping time steps for long-term analysis depending on your experimental question. This number refers to the maximum time that the cell scans its environment under the same conditions or, in other words, refers to how many consecutive time steps are considerable as a functional unit. If you, e.g., perform experiments with live perturbations after a certain time step, you would set the value to the maximum number of consecutive frames under the same condition, e.g., to a value that summarizes all the frames before the perturbation (if the perturbation occurs at time frame 10, set the value to 9, so that you retrieve long-term results for the time steps before and for the time steps after perturbation). For more details, look at the definition of the related parameters such as the “scanning activity” in the MotiQ publication [3]. 9. MotiQ 2D Analyzer and MotiQ 3D Analyzer feature the option to track each particle in the image separately. A particle is referred to a group of pixels that are adjacent to each other. If, in your image, the segmentations of two neighbored cells “touch” (= the cells are so close to each other that at least one pixel of one cell is adjacent to one pixel of the other cell), both cells will be detected as one particle, and thus, you will not receive results on the single-cell level but only for both cells together. Thus, in conditions where cells are “touching,” you need to perform Subheading 3.1 to achieve correct single-cell results. Based on this issue, it may be worth considering seeding the cells rather sparse in in vitro experiments to avoid that neighbored cells “touch,” especially for time-lapse experiments, where Subheading 3.1 is very time-consuming. There is also another reason for performing Subheading 3.1: Since MotiQ 2D and 3D Analyzer can only detect all parts of a cell automatically as one cell if all pixels belonging to the cell are connected with each other, the automated approach skipping Subheading 3.1 may fail, if you provide images with low signalto-noise ratio, where cell processes/protrusions may be interrupted with pixels that will not be detected as foreground in Subheading 3.2. In such a case, it can be advised to perform the cropping since it allows you to assign all particles belonging to one cell even if the cell is “split” into multiple particles. 10. You can move between z slices by moving the bar below the image to the left and the right or by scrolling with the mouse wheel. If you have a 3D time-lapse image, you will have multiple bars below the image, one for slices and one for frames. In that case, the upper bar can be moved with the mouse wheel, and the lower bar by pressing “alt” on Windows/Linux or “option” on Mac while scrolling. To see which processes and
Morphometric Analyses of Macrophages
535
protrusions belong to the same cell, it is recommended to start at a slice position where you see the broadest part of the cell (i.e., the cell soma/ nucleus). Move the “z” bar to such a position. Next, you can verify which cell protrusions/ processes around the cell soma are connected to the cell soma by moving the slice position a little bit forth and back, since connections may be present in other slices than the one that shows the broadest part of the cell. 11. Circumscribing the cell and its protrusions does not need to be very exact unless you do not “cut off” any parts belonging to the cell of interest, and you do not include any parts of a neighbored cell. The cell is circumscribed with a polygon whose points you set by clicking. Move the cursor to the position where you want to set the first point and click. Move the cursor to the next point you want to set and click to fix that point. Set more points to build up the polygon. When you set the last point, perform a double-click, which will set the last point and automatically connect it to the first polygon point to close the ROI. Now, you can still adjust the polygon points by dragging and dropping each point. If you click inside the set polygon and keep the left mouse button pressed, you can move the whole polygon in the whole image while keeping the mouse button pressed. The moving of the polygon is stopped upon release of the mouse button. If you click outside of the polygon, the polygon will be deleted and cannot be restored (thus avoid this unless you want to set a whole new polygon). If you have previously set a selection in MotiQ cropper on another slice of the same image you can restore the selection if needed: Go to the opened FIJI window “ROI Manager” and click on one selection listed there, which will bring that selection into the image and move to the slice image where it was selected; now you can move back to the slice position where you wanted to use that selection again and make use of it (e.g., by adjusting the polygon points and/or just confirm the polygon if it fits at that slice position as well). 12. This intermediate dialog allows more options than just continuing. For example, if you have set all ROIs in all slices where parts of the cell were present, you may select “clear remaining stack images, save, and finish.” Thereby, you do not need to shuffle still through all images where you will anyways not set any ROI. For some images, it may be possible that you can set one ROI that will fit all slices (it must never exclude any parts of the cell of interest and never contain any parts of another cell). In this case, you can just set one ROI, confirm it, and select in the intermediate dialog “apply the current ROI in all remaining stack images.” Lastly, note that this dialog also allows you to reset a slice image that you may
536
Jan N. Hansen
have erroneously provided with the false selection: By activating the checkbox, “Reset specific stack image,” and by selecting the stack image, the selection will be removed from that stack image and the stack image will be re-displayed in whole. There is, however, also an alternative to modify a previously set selection if you realize that it was incorrectly set: Go to the respective slice/frame, set a new ROI that fits correctly, and press “OK,” and now, you will be able to confirm that you want to adjust the ROI for that slice/frame in a special upcoming dialog. 13. The output .zip file contains all selections that were set in MotiQ Cropper (to display them at any later time point you can open the input image in FIJI and drag and drop the .zip file into the status bar of FIJI, which will add them to the “ROI Manager” window in FIJI, from which you can select which of the selections you want to display on the image). The output text file contains metadata for the cropping process. The output .tif file is an image file containing only the selected cell. All files are needed later in Subheading 3.2. Be careful when changing the folder structure in which the original file and the MotiQ cropper output files are placed. It is important that the files generated by MotiQ Cropper and the corresponding input file for each output file are in the same folder since otherwise processing in MotiQ Thresholder (Subheading 3.2) may fail. 14. This note gives advice on how many cells and which cells should be selected from each image with MotiQ Cropper. Since one should always include as many cells as possible in the analysis for maximal statistical power in comparison of different groups, it is optimal to select all cells in the image with MotiQ Cropper and thereby retrieve results for any cell in the image, since this is the most unbiased approach. However, especially in 3D images from tissues not all cells may be with their full cell bodies and protrusions in the image. Thus, tracing incomplete cells may result in artifacts in the results. Here, it may be good to keep consistency, e.g., by defining rules for selecting a cell for further analysis. For example, one can decide to manually select only cells whose cell soma is inside the inner 50% of the slices of the stack, since these cells may be nearly completely in the stack. Furthermore, it can also become extremely time-consuming for large data sets to track all cells from all images. If selection time is limited or if it is not feasible to select all cells in the given time, it may be better to rather include fewer cells per image but include cells from more images than to include all cells from fewer images since the diversity across more images (and thus more regions/experimental replicates) may give more accurate results. In such a
Morphometric Analyses of Macrophages
537
case, one can set up rules such as selecting a fixed number of random cells from each image (e.g., five random cells per image). All these rules should be kept consistent for a whole data set. Note that this note is meant to give practical suggestions, but it is still the responsibility of each researcher to set up justified rules for each of the researcher’s individual study paradigms. 15. The MotiQ wiki provides a checklist to optimize and find better settings for your MotiQ Thresholder processing and also provides example images for how a successful segmentation looks like: https://github.com/hansenjn/MotiQ/wiki. 16. The automated analysis steps can take long for large data sets. The “Multi-Task-Manager” indicates analysis progress and runs one listed image after the other, while not requiring any user interaction. Thus, it can be convenient to just run these automatic analysis steps outside of your working hours or when you are in the laboratory, or during the night, when you do not need your computer. Also, if the analysis takes very long, it may be good, to run the analysis on a higher-performance workstation, if available. The time for processing the images scales with the size of each image and (for MotiQ 2D and 3D Analyzer) with the complexity of the analyzed cells(s) and the number of particles tracked and filtered. If you receive an “Out of Memory” error from FIJI during the processing, it means that the memory (also called “RAM”) that was allocated to your FIJI software is not sufficient. You can set the RAM allocated to FIJI via the menu entry “Edit > Options > Memory & Threads” in FIJI. In the upcoming dialog you can select the memory that can be used by FIJI. However, note that this value shall not be higher than 80% of the RAM your computer has inbuilt (you can find out the memory on a Mac via: apple sign on top left > “about this mac”; on a Windows via: right-click on the windows sign in the lower left of the screen > Settings > System > select “About” on the left bar). If you still receive the error after adjusting the memory/RAM setting, the images that you aim to process are too large to be processed on your computer and you will need to use a computer with more memory/RAM to process them. Lastly, note that you should avoid that your computer hibernates during the processing, since the processing may be largely slowed down or break after waking up your computer. 17. A correct segmentation means that the foreground pixels represent pixels belonging to the cells, whereas the background pixels are pixels that do not refer to any cellular structure (for screenshots and more explanations, see also the MotiQ wiki (see Note 15)). It is acceptable if the foreground pixels also contain some pixels that should be rather background pixels under the
538
Jan N. Hansen
two following circumstances: (1) If those pixels are directly adjacent to the macrophage pixels, they may help to connect membrane protrusions and in that case are even advantageous: As they can thicken cell processes, they aid in detecting also cell processes that have very faint or noisy signals, hardly distinguishable from the background. (2) If the falsely as foreground detected pixels are representing small islands of very few pixels (e.g., 1–10) whereas the cell size is much larger, the segmentation is still acceptable, since you can omit small objects in the particle filtering step inbuilt in MotiQ 2D or 3D Analyzer controlled by the settings “Minimum particle area” (2D) or “Minimum particle volume” (3D). You can set a value at this parameter that is higher than the maximum number of pixels/ voxels of such small particles emerging from falsely as foreground detected pixels. For extended explanations and example images showing how to set the “Minimum particle volume,” see the guidelines for optimizing MotiQ 3D Analyzer settings at the MotiQ wiki (https://github.com/ hansenjn/MotiQ/wiki). 18. You can check whether an image is calibrated in FIJI by opening the image in FIJI and clicking on the menu entry “Image > Properties.” Verify that the “unit of length” is set correctly (e.g., to “micron” or “μm” for most microscopy images) and that the pixel width, pixel height, and voxel depth (referring to the distance between adjacent slices in your stack) are set correctly in the indicated “unit of length.” For time-lapse images, also the frame interval needs to be correctly set. Verify that the indicated values match the information you retrieve from your microscope’s manual/ software. If some of these properties are incorrectly displayed in the image in FIJI, use MotiQ 2D or 3D Analyzer’s recalibrate function to indicate the correct image properties for your images. You only need to scrutinize the calibration of the images that you input into MotiQ Cropper (Subheading 3.1) or, if Subheading 3.1 is skipped, into MotiQ Thresholder (Subheading 3.2), since the output images of MotiQ Cropper and MotiQ Thresholder preserve the calibration information from the input images. 19. The option “Calculate results for every particle separately” in MotiQ 2D or 3D Analyzer allows you to track multiple cells in the same image (see also considerations provided in Note 9). The option “Calculate results for all detected particles merged into one object” is superior for images produced with MotiQ Cropper. Since here the user has already preselected which particles in the image belong to a cell, MotiQ 2D or 3D Analyzer does not need to check that particles are connected to include them and will even consider particles to belong to the same cell if they are not “touching” (providing adjacent
Morphometric Analyses of Macrophages
539
pixels). This is a big advantage, especially for images with a low signal-to-noise ratio, where cell processes/protrusions may be interrupted due to missing signals. The size filter (see Note 17, consideration (2)) is still applied. 20. This value fine-tunes to which degree fine cell processes are considered as the process tree. It is not recommended to set this value to zero since then you may also pick up noisy cell edges as small branches. You may test the effect of adjusting this parameter on your data set by running a few exemplary cells in MotiQ 3D Analyzer with different sigma values for the Gauss filter. The larger the selected sigma, the coarser will be the detected skeleton, whereas the smaller the selected sigma, the finer structures will be included in the skeleton. Extended explanations and examples are provided in the guidelines for finding good MotiQ 3D Analyzer settings at the MotiQ wiki (https://github.com/hansenjn/MotiQ/wiki). 21. Verifying here refers to that you check that the detected skeleton reflects what you see as “processes” in the raw image. For example, if the skeleton is too coarse, you may repeat the analysis with a smaller sigma setting for the “Gauss filter prior to skeletonization,” whereas if the detected skeleton is too noisy, showing branches that are in fact emerging from a “rough” cell edge, you may try to increase the sigma and repeat the analysis. Example images showing how to scrutinize and optimize skeletonization are provided in the guidelines for finding good MotiQ 3D Analyzer settings at the MotiQ wiki (https://github.com/hansenjn/MotiQ/wiki). For fully automated detection of cells (without applying MotiQ cropper (Subheading 3.1)), it is also important to look at the different detected particles in this step and compare to the raw image whether the detected individual particles represent one or multiple cells or whether some cells are split into multiple individual detected particles. You may then eventually adapt the segmentation settings to better connect the pixels of an individual cell or to better split neighbored cells into single-cell particles, or you may consider switching to including Subheading 3.1 in your analysis to ensure that you obtain single-cell results. Another option you may consider for improving the connection of multiple objects belonging to the same cell is that it can help to pre-blur your images before subjecting them to the whole analysis pipeline. To this end, you may open each input image, apply a slide Gaussian blur filter (menu entry in FIJI: “Process > Filters > Gaussian Blur,” e.g., use a sigma of 1–2 px), and save the image, and then, subject the blurred image to the whole MotiQ pipeline (Subheadings 3.1, 3.2, 3.3, 3.4, and 3.5). The blur can reduce noise, improve signal-tonoise ratio, and better connect interrupted processes, and thus
540
Jan N. Hansen
may allow you to perform fully automated detection even for noisy images. 22. For results from non-time-lapse images, the second row in this file is filled with 0.0 values, since only one time point (the time point 0.0) was analyzed. This is normal, and the row with the 0.0 values can be just removed since it is not needed if you have not analyzed a time-lapse image. For time-lapse images, you need to keep this row since this row defines the time point of the measurement values noted in the corresponding column. 23. Compression can be performed using compression tools inbuilt into your operating system or other software that requires installation. For example, the open-source software 7-zip (https://7-zip.org) can be used to largely compress files on a Windows machine: When 7-zip is installed, you can select a folder that you want to compress, perform a right-click, and select “7-zip>Add to archive,” and in the upcoming dialog, you can select compression settings (e.g., select “Archive format: zip” and “Compression level 9 – Ultra” for maximum file size reduction) and start the compression process. When the files are compressed successfully into a zip file, you may delete the uncompressed folder, and at any later time, when you want to use some of the files, you can decompress the .zip file via a right-click on the .zip file and selecting “Extract All. . .”.
Acknowledgements Jan N. Hansen was supported with a postdoctoral fellowship from the Wenner-Gren Foundations, Sweden. References 1. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting microglial cells are highly dynamic surveillants of brain parenchyma in vivo. Science 308(5726):1314–1318. https://doi.org/10.1126/science.1110647 2. Davalos D, Grutzendler J, Yang G, Kim JV, Zuo Y, Jung S, Littman DR, Dustin ML, Gan WB (2005) ATP mediates rapid microglial response to local brain injury in vivo. Nat Neurosci 8(6):752–758. https://doi.org/10. 1038/nn1472 3. Hansen JN, Bruckner M, Pietrowski MJ, Jikeli JF, Plescher M, Beckert H, Schnaars M, Fulle L, Reitmeier K, Langmann T, Forster I, Boche D, Petzold GC, Halle A (2022) MotiQ: an open-source toolbox to quantify the cell motility and morphology of microglia. Mol
Biol Cell 33(11):ar99. https://doi.org/10. 1091/mbc.E21-11-0585 4. Beins EC, Beiert T, Jenniches I, Hansen JN, Leidmaa E, Schrickel JW, Zimmer A (2021) Cannabinoid receptor 1 signalling modulates stress susceptibility and microglial responses to chronic social defeat stress. Transl Psychiatry 11(1):164. https://doi.org/10.1038/ s41398-021-01283-0 5. Fulle L, Offermann N, Hansen JN, Breithausen B, Erazo AB, Schanz O, Radau L, Gondorf F, Knopper K, Alferink J, Abdullah Z, Neumann H, Weighardt H, Henneberger C, Halle A, Forster I (2018) CCL17 exerts a neuroimmune modulatory function and is expressed in hippocampal neurons. Glia 66(10):2246–2261. https://doi.org/10. 1002/glia.23507
Morphometric Analyses of Macrophages 6. Karunakaran I, Alam S, Jayagopi S, Frohberger SJ, Hansen JN, Kuehlwein J, Holbling BV, Schumak B, Hubner MP, Graler MH, Halle A, van Echten-Deckert G (2019) Neural sphingosine 1-phosphate accumulation activates microglia and links impaired autophagy and inflammation. Glia. https://doi.org/10. 1002/glia.23663 7. Klaus C, Hansen JN, Ginolhac A, Gerard D, Gnanapragassam VS, Horstkorte R, Rossdam C, Buettner FFR, Sauter T, Sinkkonen L, Neumann H, Linnartz-Gerlach B (2020) Reduced sialylation triggers homeostatic synapse and neuronal loss in middle-aged mice. Neurobiol Aging. https://doi.org/10. 1016/j.neurobiolaging.2020.01.008 8. Plescher M, Seifert G, Hansen JN, Bedner P, Steinhauser C, Halle A (2018) Plaquedependent morphological and electrophysiological heterogeneity of microglia in an Alzheimer’s disease mouse model. Glia 66(7): 1464–1480. https://doi.org/10.1002/glia. 23318 9. Reusch N, Ravichandran KA, Olabiyi BF, Komorowska-Mu¨ller JA, Hansen JN, Ulas T, Beyer M, Zimmer A, Schmo¨le AC (2021) Cannabinoid receptor 2 is necessary to induce tolllike receptor-mediated microglial activation. Glia. https://doi.org/10.1002/glia.24089
541
10. Schmole AC, Lundt R, Toporowski G, Hansen JN, Beins E, Halle A, Zimmer A (2018) Cannabinoid receptor 2-deficiency Ameliorates disease symptoms in a Mouse Model with Alzheimer’s disease-like pathology. J Alzheimers Dis 64(2):379–392. https://doi.org/ 10.3233/JAD-180230 11. Carvalho K, Faivre E, Pietrowski MJ, Marques X, Gomez-Murcia V, Deleau A, Huin V, Hansen JN, Kozlov S, Danis C, Temido-Ferreira M, Coelho JE, Meriaux C, Eddarkaoui S, Gras SL, Dumoulin M, Cellai L, Neuro CEBBB, Landrieu I, Chern Y, Hamdane M, Buee L, Boutillier AL, Levi S, Halle A, Lopes LV, Blum D (2019) Exacerbation of C1q dysregulation, synaptic loss and memory deficits in tau pathology linked to neuronal adenosine A2A receptor. Brain. https://doi.org/10.1093/brain/awz288 12. Werner Y, Mass E, Ashok Kumar P, Ulas T, Handler K, Horne A, Klee K, Lupp A, Schutz D, Saaber F, Redecker C, Schultze JL, Geissmann F, Stumm R (2020) Cxcr4 distinguishes HSC-derived monocytes from microglia and reveals monocyte immune responses to experimental stroke. Nat Neurosci 23(3): 351–362. https://doi.org/10.1038/s41593020-0585-y
Chapter 35 Combined Analysis of mRNA Expression and Open Chromatin in Microglia Rebekka Scholz, Desire´e Bro¨samle, Xidi Yuan, Jonas J. Neher, and Marc Beyer Abstract The advance of single-cell RNA-sequencing technologies in the past years has enabled unprecedented insights into the complexity and heterogeneity of microglial cell states in the homeostatic and diseased brain. This includes rather complex proteomic, metabolomic, morphological, transcriptomic, and epigenetic adaptations to external stimuli and challenges resulting in a novel concept of core microglia properties and functions. To uncover the regulatory programs facilitating the rapid transcriptomic adaptation in response to changes in the local microenvironment, the accessibility of gene bodies and gene regulatory elements can be assessed. Here, we describe the application of a previously published method for simultaneous high-throughput ATAC and RNA expression with sequencing (SHARE-seq) on microglia nuclei isolated from frozen mouse brain tissue. Key words Microglia, Transcriptomics, Single-nucleus sequencing, Multiomics
1
Epigenetics, Chromatin accessibility, Bioinformatics,
Introduction Microglia represent the most abundant resident macrophage population in the central nervous system (reviewed in [1]). Additionally, they are the most dynamic cell type found in the brain [2]. The study of microglial morphology and cell surface markers by immunostaining, flow cytometry, and noninvasive imaging in vivo has uncovered at least three heterogeneous phenotypes that the cells can adapt and adjust rapidly in response to changes in the microenvironment [3–7]. However, the full complexity of microglia states in the mouse and human brain only became evident with the emergence of high-throughput single-cell (sc) transcriptomic
Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1007/978-1-0716-3437-0_35. Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0_35, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
543
544
Rebekka Scholz et al.
techniques in recent years, leading to the discovery of multiple microglia subclusters associated with brain development [8, 9], homeostasis and aging [10, 11], and different neurological conditions such as amyloid aggregation [12] or multiple sclerosis [13]. To broaden our understanding of the regulation of this diversity of microglial states, it is crucial to consider the underlying epigenetic heterogeneity driving the reported transcriptional changes [14]. The assay for transposase-accessible chromatin with sequencing (ATAC-seq) allows for the mapping of genomic regions that are accessible to transcription factor and RNA polymerase 2 binding as a prerequisite for gene transcription [15]. It is based on the use of adapter-loaded hyperactive Tn5 transposase that introduces a double-strand break in all chromatin sections not shielded by tightly packed nucleosomes and integrates the adapters at the cutting side [15]. While it was originally developed for the analysis of bulk cell populations, it can also be performed on the single-cell level [16]. Although this enables researchers to derive cell-specific chromatin landscapes, a major challenge lies in accurately correlating epigenetic profiles with transcriptomic data derived from the same cell subpopulations but assessed independently [17]. A rapidly growing number of tools for integrating scRNA-seq and scATAC-seq data are available to overcome this obstacle; however, these methods often rely on pre-defined input such as a “gene activity matrix” that can heavily influence the integration performance [18–22]. Alternatively, scRNA-seq and scATAC-seq can also be performed on the same individual cell, thus allowing the unequivocal matching of chromatin accessibility and transcriptomic profiles [23–27]. One of these methods is SHARE-seq (simultaneous high-throughput ATAC and RNA expression with sequencing), developed by Ma and colleagues in 2020 [27]. It is a platebased assay utilizing the split-pool combinatorial barcoding implemented in split-pool ligation-based transcriptome sequencing (SPLiT-seq) and can therefore be established without any specialized equipment compared to droplet-based methods [27– 29]. Furthermore, cells or nuclei can be used for this protocol, which enables the use of fresh-frozen tissue, such as postmortem collected human tissue, as the starting material [27]. Here, we adapted the published SHARE-seq workflow to accommodate our research question, including alternative starting materials and experimental conditions. We describe a method to (a) isolate microglia nuclei from fresh-frozen mouse brain tissue, (b) perform SHARE-seq on the sorted nuclei, and (c) process the resulting sequencing data. Each of these major sections consists of several substeps (see Fig. 1). Briefly, total nuclei are isolated from mouse brain tissue and purified with a sucrose cushion, followed by light fixation and labeling of the myeloid lineage regulating transcription factor
Joint Transcriptomic and Epigenetic Analysis in Microglia
545
Fig. 1 Workflow for the combined analysis of transcriptome and chromatin accessibility of microglia. First, microglia nuclei are isolated from frozen mouse brain tissue, labeled for PU.1, and sorted (a). This is followed by performing SHARE-seq on the nuclei (b), resulting in the generation of separate mRNA and ATAC libraries. After sequencing, the fastq files are subjected to a processing pipeline that generates, among other results, a count matrix and a fragment file for the subsequent combined analysis (c). (The figure was created with Biorender.com)
PU.1 (Subheading 3.2) [30]. PU.1-positive nuclei are subsequently collected for SHARE-seq. Chromatin tagmentation and reverse transcription are performed inside the intact nuclei (Subheadings 3.3 and 3.4). After three rounds of barcoding, the nuclei are lysed, releasing the biotin-tagged first-strand cDNA and tagmented genomic DNA into the supernatant (Subheadings 3.5 and 3.6). For separation of cDNA from genomic DNA, the cDNA is pulled down on streptavidin-coupled magnetic beads (Subheading 3.6). After second-strand synthesis and a first round of PCR amplification, the PCR product is tagmented to add the P5 adapter and amplified to generate the final single-nucleus (sn) RNA library (Subheadings 3.7, 3.9, and 3.10). For the snATAC library, tagmented chromatin fragments are PCR-amplified (Subheadings 3.8 and 3.11). After a quality check, all libraries (with different indices) can be pooled and subjected to sequencing (Subheading 3.12). The generated fastq files are processed using a dedicated preprocessing
546
Rebekka Scholz et al.
pipeline (Subheading 3.13). The output can subsequently be analyzed with standard tools for scRNA-seq and scATAC-seq data. The entire protocol for snRNA and snATAC library generation (Subheading 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, 3.10, and 3.11) takes three consecutive days.
2
Materials If not otherwise specified, use ultrapure nuclease-free water for buffer preparation. Buffers that are needed in multiple steps are listed once at their first occurrence. The RNase inhibitors (RIs) should only be added immediately before use.
2.1 SHARE-seq Preparations
1. Ultrapure, nuclease-free water. 2. TE buffer: 0.1 mM EDTA in 10 mM Tris–HCl, pH 8.0. Prepare in advance and store at room temperature (RT). 3. STE buffer: 50 mM NaCl, 1 mM EDTA in 10 mM Tris–HCl, pH 8.0. Prepare in advance and store at RT. 4. Oligonucleotides: See Electronic Supplementary Tables 1, 2, and 3. 5. Tn5: Unloaded Tn5 transposase can be either bought from Diagenode or other suppliers, or self-produced as described by Picelli and colleagues in 2014 [31] (see Note 1). 6. Tn5 dilution buffer: 0.1 M NaCl, 0.1 mM EDTA, 1 mM DTT, 0.1% NP-40, 50% glycerol in 50 mM Tris–HCl, pH 8.0. Prepare in advance and store at 4 °C for up to 1 month. 7. Glycerol, molecular biology grade. 8. DNA low bind 1.5 mL microcentrifuge tubes. 9. Conical sterile polypropylene centrifuge tubes, 15 mL and 50 mL. 10. DNA low bind 96-well PCR plates. 11. Tightly sealing adhesive PCR plate foils. 12. PCR thermocycler. 13. Thermoshaker for 1.5 mL tubes and 96-well plates. 14. Optional: low-retention pipette tips.
2.2 Microglia Nuclei Isolation and Sorting
1. Petri dish: polystyrene, 90 mm diameter, sterile. 2. Pasteur pipettes: polyethylene, 3 mL, sterile. 3. Dounce tissue grinder set: working volume 2 mL, loose and tight pestle. 4. Scalpel. 5. Cell strainer: 70 μm, sterile.
Joint Transcriptomic and Epigenetic Analysis in Microglia
547
6. Refrigerated swing bucket centrifuge with inlets for 15 mL conical tubes and 1.5 mL microcentrifuge tubes. 7. End-over-end rotator. 8. Fluorescence-activated cell sorting (FACS) instrument. 9. Borosilicate glass FACS tubes. 10. Dulbecco’s phosphate-buffered saline (DPBS), modified, without calcium chloride and magnesium chloride, sterilefiltered. 11. Nuclei PURE Prep Isolation Kit (Sigma-Aldrich, NUC201) (see Note 2). 12. Antibodies: PU.1 rabbit monoclonal antibody (clone 9G7, in this protocol we use the antibody from Cell Signaling Technology), goat anti-rabbit IgG (H+L) cross-adsorbed secondary antibody fluorophore conjugate (polyclonal, working concentration 10 μg/mL), NeuN-PE antibody (clone A60; in this protocol, we use the antibody from Merck Millipore). 13. 4′,6-diamidino-2-phenylindole (DAPI): 1 mg/mL stock solution in ultrapure water. 14. 10% BSA in DPBS. Sterile filter and store at 4 °C for up to 1 day or -20 °C for extended storage times. 15. Nuclei lysis buffer (NLB): 1 mM DTT, 0.1% Triton X-100 in Nuclei PURE lysis buffer (Nuclei PURE Prep Isolation Kit). 16. Nuclei suspension buffer (NSB): 10% BSA, 0.4 U/μL Takara RI (Takara Bio Europe, 2313) in DPBS. 17. Labeling buffer: 1 U/μL Takara RI in NSB. 18. 4% methanol-free formaldehyde (FA) in DPBS. Prepare the 4% solution from the 16% stock immediately before use. 19. Fixation stop solution: 1.17 M glycine, 0.84% BSA in 419 mM Tris–HCl, pH 8.0. Prepare immediately before use. 20. Sucrose cushion solution: 1.8 M sucrose cushion, 0.1 M DTT in sucrose cushion buffer (Nuclei PURE Prep Isolation Kit). 21. Primary antibody solution: 1:50 dilution of NeuN-PE and PU.1 primary antibody in labeling buffer. 22. Secondary antibody solution: 10 μg/mL goat anti-rabbit IgG (H+L) secondary antibody with fluorophore conjugate to a concentration of in labeling buffer. 23. PBSI: 0.04% BSA, 0.025 U/μL SUPERase RI (Thermo Scientific, AM2694), and 0.1 U/μL Enzymatics RI (Enzymatics, Qiagen, Y9240L) in DPBS. Add the RI immediately before use. 24. Neubauer Hemocytometer and light microscope. 25. Trypan blue solution, 0.4%.
548
Rebekka Scholz et al.
2.3 SHARE-seq on Microglial Nuclei— Transposition
1. Benchtop fixed-angle rotor centrifuge for 1.5 mL microcentrifuge tubes. 2. Transposition reaction mix: 38.8 mM Tris-acetate, 77.6 mM potassium acetate, 11.8 mM magnesium acetate, 18.8% N, N-dimethylformamide (DMF), 0.12% NP-40, 0.68 U/μL Enzymatics RI, and 0.4% protease inhibitor cocktail (Sigma-Aldrich, P8340). Prepare immediately before use and do not re-use. 3. Nuclei isolation buffer (NIB): 10 mM NaCl, 3 mM MgCl2, 0.1% NP-40, 0.1 U/μL Enzymatics RI, 0.05 U/μL SUPERase RI in 10 mM Tris–HCl, pH 7.5. The buffer can be stored at 4 ° C for up to 1 month. Only add the RI immediately before use. 4. ATAC-Tn5: Load the Tn5 as described in Subheading 3.1 and store at -20 °C until use.
2.4 SHARE-seq on Microglial Nuclei— Reverse Transcription
1. 0.2 mL PCR tubes.
2.5 SHARE-seq on Microglial Nuclei— Nuclei Barcoding
1. Barcoding mix: 1.667× T4 DNA ligase buffer (supplied with T4 DNA ligase), 0.1667% NP-40, 0.53 U/μL Enzymatics RI, and 0.083 U/μL SUPERase RI. Prepare immediately before use and keep on ice.
2. Reverse transcription mix (RevTr mix): 1.33× reverse transcription buffer (supplied with Maxima H Minus Reverse Transcriptase), 18.75% PEG6000, 625 μM dNTPs, 12.5 μM biotinylated reverse transcription primer, 25 U/μL Maxima H Minus Reverse Transcriptase (Thermo Scientific, EP0751), 0.3 U/μL Enzymatics RI, and 0.3 U/μL SUPERase RI. Prepare immediately before use and keep on ice.
2. Annealed barcoding plates: Prepare the plates as described in Subheading 3.1 and store them at 4 °C or -20 °C for longterm storage. 3. Blocking strand solution 1: 22 μM Round 1 blocking oligonucleotide, 1.8× T4 DNA ligase buffer. Prepare on the same day and store at 4 °C until use. 4. Blocking strand solution 2: 26.4 μM Round 2 blocking oligonucleotide, 1.8× T4 DNA ligase buffer. Prepare on the same day and store at 4 °C until use. 5. Blocking strand solution 3: 23 μM Round 3 blocking oligonucleotide 3, 0.1% Triton X-100. Prepare on the same day and store at 4 °C until use. 6. Ligation mix: 1.25× T4 DNA ligase buffer, 0.13% NP-40, 25 U/μL T4 DNA ligase (NEB, M0202), 0.0625 U/μL SUPERase RI, and 0.4 U/μL Enzymatics RI. Prepare immediately before use and keep on ice. 7. Vortex mixer. 8. Optional: Multi-dispenser pipette and 2.5 mL tips.
Joint Transcriptomic and Epigenetic Analysis in Microglia
2.6 SHARE-seq on Microglial Nuclei— Reverse Crosslinking and Library Splitting
549
1. Magnetic stand for 1.5 mL microcentrifuge tubes and 0.2 mL PCR tubes. 2. Reverse crosslinking mix: 100 mM NaCl, 0.4% SDS, 0.72 U/μ L SUPERase RI in 100 mM Tris–HCl, pH 8.0. Prepare in advance and store at RT. Add the RI immediately before use. 3. Proteinase K solution: 20 mg/mL proteinase K from Tritirachium album, 1 mM calcium acetate, 50% glycerol in 10 mM Tris–HCl, pH 8.0. Prepare in advance and store at -20 °C. 4. Phenylmethylsulfonyl fluoride (PMSF) solution: 100 mM PMSF in anhydrous isopropanol. Aliquot and store at -20 °C (see Note 3). 5. 2× B&W buffer: 2 M NaCl, 1 mM EDTA in 10 mM Tris–HCl, pH 8.0. Prepare in advance and store at RT. 6. 2× B&W buffer + 4RI: 2 M NaCl, 1 mM EDTA, 4 U/μL SUPERase RI in 10 mM Tris–HCl, pH 8.0. Add the RI immediately before use and keep on ice. 7. 1× B&W-T buffer: 1× B&W buffer, 0.05% Tween 20. Prepare in advance and store at RT. 8. 1× B&W-T buffer + RI: 1× B&W buffer, 0.05% Tween 20, varying concentrations of SUPERase RI (1RI: 1 U/μL; 2RI: 2 U/μL). Add the RI immediately before use and keep on ice. 9. Dynabeads MyOne Streptavidin C1 (Thermo Scientific, 65002).
2.7 SHARE-seq on Microglial Nuclei— Template Switch
1. STE buffer + 1RI: 1 U/μL SUPERase RI in 1× STE buffer. Add the RI immediately before use and keep on ice.
2.8 SHARE-seq on Microglial Nuclei— Tagmented DNA Cleanup
1. MinElute PCR Purification Kit (Qiagen, 28004) with buffers PB, EB, and MinElute Spin columns.
2.9 SHARE-seq on Microglial Nuclei— cDNA Amplification
1. cDNA-PCR mix: 1× KAPA HiFi HotStart Ready Mix (Roche, 7958935001), 400 nM RNA-PCR primer, and 400 nM P7 primer. Prepare immediately before use and keep on ice.
2. Template switch mix: 1× reverse transcription buffer, 20% PEG6000, 1 mM dNTPs, 40 U/μL NxGen RI (Lucigen, 30281-2), 2.5 μM template switch oligonucleotide, and 10 U/μL Maxima H Minus Reverse Transcriptase. Prepare immediately before use and keep on ice.
2. Ethanol, absolute, molecular biology grade.
2. SYBR Green stock solution: Dilute SYBR Green I Nucleic Acid Gel Stain (10,000× concentrate in DMSO) to 100× with nuclease-free ultrapure water and store at -20 °C protected from light. Before use, dilute to 10×, and do not re-use this dilution.
550
Rebekka Scholz et al.
3. cDNA qPCR mix: 1× KAPA HiFi HotStart Ready Mix, 400 nM RNA-PCR primer, 400 nM P7 primer, and 1.333× SYBR Green (from 10× stock solution). Prepare immediately before use, and keep on ice protected from light after adding SYBR Green. 4. AMPure XP beads (Beckman Coulter, A63880). 5. 80% ethanol: Dilute 4 parts of molecular biology grade absolute ethanol with 1 part of ultrapure water. Prepare on the same day and store at RT. 6. Qubit Fluorometer (Thermo Scientific). 7. Qubit dsDNA HS Assay Kit (Thermo Scientific, Q32851). 8. 4200 TapeStation System (Agilent). 9. High Sensitivity D5000 Screen Tape & Reagents (Agilent, 5067–5592 and 5067–5593). 10. Quantitative PCR (qPCR) thermocycler and appropriate equipment. 2.10 SHARE-seq on Microglial Nuclei— Tagmentation and RNA Library Preparation
1. 2× Tris-DMF: 10 mM MgCl2, 20% DMF in 20 mM Tris–HCl, pH 7.5. Prepare in advance and store at 4 °C until use. 2. RNA-Tn5: Assemble the Tn5 complex as described in Subheading 3.1 and store at -20 °C until use. 3. Tagmentation PCR mix: 1.72× NEB Next High-Fidelity 2× PCR Master Mix (NEB, M0541), 0.86 μM P7 primer. Prepare immediately before use and keep on ice (see Note 4). 4. High Sensitivity D1000 Screen Tape & Reagents (Agilent, 5067–5584 and 5067–5585).
2.11 SHARE-seq on Microglial Nuclei— ATAC Library Preparation
2.12
Sequencing
2.13
Preprocessing
1. ATAC-PCR mix: 1.72× NEB Next High-Fidelity 2× PCR Master Mix, 0.86 μM P7 Primer. Prepare immediately before use and keep on ice (see Note 4). 2. ATAC-qPCR mix: 1× NEB Next High-Fidelity 2× PCR Master Mix, 0.42 μM P7 primer, 0.42 μM Ad1 primer, 0.8× SYBR Green. Prepare immediately before use, and keep on ice protected from light after adding SYBR Green. Appropriate next-generation sequencing platform, i.e., Illumina NovaSeq 6000 Series or Illumina NextSeq Series including all necessary reagents and kits. 1. Linux system with Docker [32] installed. 2. Data and code provided at https://github.com/masai1116/ SHARE-seq-alignmentV2/ [27] and https://github.com/ grenaud/deML [33] under the GNU general public license v3.0.
Joint Transcriptomic and Epigenetic Analysis in Microglia
551
3. Docker image provided at the Docker Hub repository rebekkascholz/share-seq_repo (https://hub.docker.com/ repository/docker/rebekkascholz/share-seq_repo). 4. “deML-index_share-seq.R” script provided at https://github. com/rebekkascholz/deML-index_share-seq.
3
Methods
3.1 SHARE-seq Preparations
1. Reconstitute all primers to their working stock concentration according to Electronic Supplementary Table 1. 2. To prepare the oligonucleotides that will be loaded onto the Tn5, mix equal amounts of 100 μM Read 1 or Read 2 and Blocked ME Comp. Anneal by using the program in Table 1 (see Note 5). 3. To load the ATAC-Tn5, mix 2.5 μL of the annealed oligonucleotides 1 and 2. Then, add an equimolar amount of Tn5 (see Note 6). Mix by pipetting without introducing bubbles and then incubate for 30 min at 23 °C (see Note 7). 4. After incubation, add 1 volume of glycerol to the loaded Tn5, mix well by pipetting, and store the loaded Tn5 at -20 °C (see Notes 8 and 9). 5. To load the RNA-Tn5, mix 5 μL of annealed oligonucleotides 1 with an equimolar amount of Tn5. Now perform the incubation as described in step 4 and store the solution at -20 °C after adding 50% glycerol. 6. Prepare the barcoding plates by adding 5 μL of 18 μM Round 1 linker to each well of a 96-well DNA low bind plate. Now, add 5 μL of the respective Round 1 barcode (in working stock concentration, 20 μM) to the wells for a final volume of 10 μL/ well. Anneal the oligonucleotides by running the program specified in Table 2 (see Note 10). Prepare Round 2 and Round 3 barcoding plates similarly (5 μL 22 μM Round 2 linker + 5 μL 24 μM Round 2 barcodes; 5 μL 26 μM Round 3 linker + 5 μL 28 μM Round 3 barcodes). Store the plates at -20 °C.
Table 1 Program for Tn5 oligonucleotide annealing Step
Temperature
Time
1
95 °C
2 min
2–76
95 °C
1 min
77
20 °C
Hold
deltaT -1 °C/min
552
Rebekka Scholz et al.
Table 2 Program for barcoding plate oligonucleotide annealing
3.2 Microglia Nuclei Isolation and Sorting
Step
Temperature
Time
1
85 °C
2 min
2–66
85 °C
1 min
67
20 °C
Hold
deltaT -1 °C/min
Please note that prior to working with animals/collecting tissue from animals, approval must be obtained from the local authority. Always work on ice, and pre-cool the homogenizers, buffers, and conical tubes. Add RI to the chilled buffers immediately before use. The protocol was adapted from the Nuclei PURE Prep Isolation Kit (Sigma-Aldrich, NUC-201, Technical Bulletin No. MB-735) to isolate PU.1-positive nuclei from one fresh-frozen mouse brain hemisphere. If more or less tissue is used, scale the reagents accordingly. 1. Transfer the fresh-frozen brain tissue to a 9 cm petri dish. Add 0.5 mL of NLB and mince the frozen tissue into small pieces with a scalpel. 2. Transfer the tissue pieces to a chilled Dounce homogenizer with a 3 mL Pasteur pipette. Wash the petri dish with 1.0 mL of NLB and add this to the homogenizer for a total volume of 1.5 mL. 3. Homogenize slowly with pestle A (loose) (approximately 15 strokes) and make sure to always reach the bottom of the homogenizer. Avoid foaming. Then, switch to pestle B (tight) and homogenize for another 15 times until the suspension appears homogeneous. 4. Incubate the homogenate on ice for 5 min (lysis step) (see Note 11). 5. Add 500 μL of NSB to the homogenate and then filter it through a 70 μm filter (pre-wetted with 500 μL of NSB) into a 15 mL conical tube (see Note 12). 6. Use an additional 3.0 mL of NSB to wash the homogenizer and filter (see Note 13). 7. Centrifuge at 500 g for 5 min at 4 °C to collect the nuclei. Discard the supernatant. 8. Resuspend the pellet to a volume of 1 mL with NSB. 9. For fixation, prepare the 4% FA working solution and fixation stop solution immediately before use (see Note 14).
Joint Transcriptomic and Epigenetic Analysis in Microglia
553
10. Fix the nuclei by adding the FA working solution to the nuclei suspension to a final concentration of 0.5% (i.e., add 143 μL of 4% FA to 1 mL of nuclei suspension). Mix well by inverting and then incubate at RT for 5 min while rotating (see Note 15). 11. Stop the reaction by adding 1× of the freshly made fixation stop solution (i.e., 119.4 μL to 1 mL of nuclei suspension). Incubate for another 5 min on the rotator. 12. Add 2 mL of NSB and then centrifuge at 500 g for 5 min at 4 °C. 13. Discard the supernatant and resuspend the nuclei pellet in 3 mL of NSB. 14. For myelin removal, prepare fresh 1.8 M sucrose cushion solution immediately before use. 15. Add 6 mL of cold 1.8 M sucrose cushion solution to 3 mL of nuclei suspension. Mix gently but well and set on ice. 16. Add 4 mL of ice-cold 1.8 M sucrose cushion solution to the bottom of a pre-cooled 15 mL conical tube. 17. Carefully and slowly layer 9 mL of nuclei suspension on top of it. Be careful not to disturb the bottom layer. 18. Centrifuge the gradient in a swing bucket rotor at 3000 g for 45 min at 4 °C (see Note 16). 19. Remove the tube from the centrifuge carefully. A thin, white nuclei pellet might be visible at the bottom of the tube. First, completely remove the myelin layer on top of the sucrose gradient and then carefully aspirate the supernatant without disturbing the pellet. 20. Resuspend the pellet in 2 mL of NSB and then centrifuge at 500 g for 5 min at 4 °C. 21. Discard the supernatant and then resuspend in 100 μL of the primary antibody solution. Mix well and then incubate at 4 °C in the dark on an end-over-end rotator for 60 min. 22. Add 1 mL NSB. Centrifuge at 500 g for 5 min at 4 °C. 23. Aspirate the supernatant. 24. Add 100 μL of secondary antibody solution. Mix well by pipetting and then incubate for 30 min on a rotation wheel in the dark at 4 °C. 25. After 20 min, add DAPI to a final concentration of 0.5 μg/mL. 26. Add 1 mL NSB to the nuclei suspension, and centrifuge at 500 g for 5 min at 4 °C. 27. Discard the supernatant and resuspend the pellet in 250 μL of NSB. If the suspension is opaque, dilute further. Proceed to fluorescence-activated nuclei sorting (FANS).
554
Rebekka Scholz et al.
Fig. 2 Gating strategy for the isolation of microglia nuclei from whole mouse brain extracts. The PU.1-positive, NeuN-negative population of single, DAPI-positive nuclei is collected by fluorescence-activated nuclei sorting (FANS)
28. Sort for NeuN- PU.1+ nuclei (see Notes 17 and 18). An exemplified gating strategy can be seen in Fig. 2. 3.3 SHARE-seq on Microglial Nuclei— Transposition
The following protocol is intended for two samples with 100,000 nuclei starting material each. Use DNA low bind 1.5 mL microcentrifuge tubes whenever possible. Low-retention pipette tips can be used to increase nuclei recovery. 1. For transposition, prepare the transposition reaction mix fresh before use. 2. Centrifuge the nuclei at 500 g for 5 min at 4 °C. Very carefully, take up the supernatant without disturbing the pellet and leave 5 μL of PBSI for each transposition reaction on top (see Note 19). 3. Add 42.5 μL of the transposition reaction mix to each 5 μL aliquot of nuclei in PBSI. Mix by pipetting and incubate at RT for 10 min. 4. Add 2.5 μL of the appropriate ATAC-Tn5 dilution. Mix and incubate for 30 min at 37 °C shaking with 500 rpm (see Note 19). 5. Combine all reactions from one sample. Centrifuge the samples at 1000 g for 3 min in a swing bucket rotor and then wash once with 500 μL of NIB. Resuspend each sample in 30 μL of NIB.
3.4 SHARE-seq on Microglial Nuclei— Reverse Transcription
1. For reverse transcription, prepare 240 μL of RevTr mix. 2. Add 120 μL of the RevTr mix to each 30 μL of nuclei suspension. Mix well, and then split each sample into 75 μL reactions into 0.2 mL PCR tubes. 3. Perform the reverse transcription in a thermocycler with the steps outlined in Table 3. 4. Combine the reactions from the same sample, add 200 μL of NIB each, and transfer the samples to 1.5 mL microcentrifuge tubes for centrifugation at 1000 g for 3 min.
Joint Transcriptomic and Epigenetic Analysis in Microglia
555
Table 3 Program for reverse transcription Step
Temperature
Time
1
50 °C
10 min
2
8 °C
12 s
3
15 °C
45 s
4
20 °C
45 s
5
30 °C
30 s
6
42 °C
120 s
7
50 °C
180 s
8
50 °C
5 min
9
4 °C
Hold
Comments
Go to step 2: total 3 cycles
5. Remove the supernatant, wash the pellets with 300 μL of NIB each, and then repeat the centrifugation. 6. Resuspend each pellet in 480 μL of NIB and store at 4 °C until the next morning (see Note 20). Optional stopping point (overnight) 3.5 SHARE-seq on Microglial Nuclei— Nuclei Barcoding
1. For cell barcoding, let the pre-prepared barcoding plates come to RT before use. 2. Add 1440 μL of hybridization mix to each 480 μL nuclei suspension in NIB, and mix well to ensure complete resuspension of the nuclei that might have settled overnight. 3. Add 40 μL of the respective nuclei suspension to each well of barcoding plate 1. Use 48 wells per sample. Mix by vortexing lightly and then pulse-spin the plate down (see Note 21). 4. Incubate the plate on a thermoshaker for 30 min at RT with gentle shaking (300 rpm) for hybridization. 5. After 30 min, add 10 μL of round 1 blocking oligonucleotide (22 μM in 2× T4 DNA ligase buffer) to each well and incubate for another 30 min (RT, 300 rpm). 6. After the first barcode is added, all wells from all samples can be pooled together in a 15 mL conical tube (see Note 22). Mix well and then split the nuclei suspension onto the second barcoding plate (approx. 55 μL per well) (see Note 23). Again, incubate for 30 min (RT, 300 rpm). 7. Now, add 10 μL of round 2 blocking oligonucleotide (26.4 μM in 2× T4 DNA ligase buffer) and repeat the incubation.
556
Rebekka Scholz et al.
8. Pool all wells for a second time and split the suspension evenly onto the round 3 barcoding plate (approximately 70 μL/well). Shake at 300 rpm for 30 min at RT. 9. Finally, add 10 μL of round 3 blocking oligonucleotide (23 μM in 0.1% Triton X-100). Repeat the incubation step. 10. Combine all wells and centrifuge at 1000 g for 3 min to pellet the nuclei. 11. Remove the supernatant, wash the pellet with 1 mL of NIB, transfer to a 1.5 mL microcentrifuge tube, and centrifuge again. 12. Repeat the washing step. 13. Resuspend the pellet in 40 μL of NIB (see Note 24). 14. Add 160 μL of freshly prepared ligation mix, mix well, then aliquot to 50 μL reactions, and incubate for 30 min at RT while shaking at 300 rpm. 15. Combine the reactions, add 1 mL of NIB, and centrifuge at 1000 g for 3 min at RT. 16. Remove the supernatant carefully and resuspend the pellet in 51 μL of NIB. 17. Take 1 μL of the nuclei suspension and count the nuclei manually to estimate the recovery. Split the nuclei into reactions of 1000–20,000 nuclei each (see Note 25). 3.6 SHARE-seq on Microglial Nuclei— Reverse Crosslinking and Library Splitting
1. For reverse crosslinking, add 26 μL of 2× reverse crosslinking mix and 2 μL of 20 mg/mL proteinase K to each 25 μL sample. Mix and incubate for 1 h at 55 °C (see Note 26). 2. Add 2.5 μL of 100 mM PMSF to each reaction to inactivate the Proteinase K and incubate at RT for 10 min (see Note 3). 3. In the meanwhile, take 20 μL of the resuspended MyOne Streptavidin C1 Dynabeads (10 μL/reaction) and wash them with 1 mL of 1× B&W-T buffer. Place the tube on a magnetic stand and wait for approx. 1 min until the beads are well attached, and then carefully remove the supernatant. 4. Repeat the washing step with 100 μL of the same buffer. 5. For the last washing step, use 100 μL of 1× B&W-T buffer + 2RI. 6. After washing, resuspend the beads in 100 μL of 2× B&W buffer + 4RI. Add 50 μL to each sample. 7. Rotate the samples on an end-over-end rotator at 10 rpm for 60 min at RT (see Notes 27 and 28). 8. To separate RNA and chromatin fragments, place the samples on a magnetic stand, then transfer the supernatant containing the tagmented DNA to a new microcentrifuge tube, and directly resuspend the beads in 100 μL of 1× B&W-T buffer + 1RI.
Joint Transcriptomic and Epigenetic Analysis in Microglia
3.7 SHARE-seq on Microglial Nuclei— Template Switch
557
1. Place the MyOne C1 Dynabeads with the bound RNA from 3.6 on a magnetic stand to remove the supernatant. Then, wash once with 100 μL of 1× B&W-T buffer + 1RI and once with 100 μL of STE + 1RI. 2. Prepare the template switch mix and add 50 μL to each bead aliquot. Make sure to fully resuspend the beads, and then rotate on an end-over-end rotator at 10 rpm for 30 min at RT. 3. As the beads might have started to clump together, resuspend them by pipetting, then transfer the tubes to a thermomixer, and shake them at 300 rpm for 90 min at 42 °C. Resuspend every 30 min by pipetting. In the meanwhile, proceed to Subheading 3.8 for cleanup of the tagmented DNA fragments. 4. Add 100 μL of nuclease-free water and remove the supernatant (see Note 29). 5. Resuspend the beads in 200 μL of STE for overnight storage at 4 °C (see Note 30).
3.8 SHARE-seq on Microglial Nuclei— Tagmented DNA Cleanup
1. Purify the supernatant containing the tagmented genomic DNA with the QIAGEN MinElute PCR cleanup kit following the MinElute handbook instructions (Qiagen, 28004, HB-2069). First, add 500 μL of buffer PB to each sample for binding. 2. Apply each sample to a MinElute column placed in a 2 mL collection tube. 3. Centrifuge for 1 min at 17,900 g for binding. 4. Discard the flow-through and place the column back in the collection tube. 5. Wash the bound DNA by adding 750 μL of buffer PE. Centrifuge again for 1 min. 6. Discard the flow-through and centrifuge the column again to remove any remaining traces of ethanol from the column bed. 7. Transfer the column to a 1.5 mL microcentrifuge tube. 8. Add 11 μL of buffer EB to the column and leave at RT for 1 min. Then, centrifuge for 1 min. 9. Repeat the elution step with an additional 11 μL of buffer EB. The purified DNA can be stored at 4 °C overnight, or at -20 °C for several days (see Note 31). Optional stopping point
3.9 SHARE-seq on Microglial Nuclei— cDNA Amplification
1. Remove the STE buffer from the Dynabeads (see Subheading 3.7) and resuspend the beads immediately in 55 μL of the freshly prepared cDNA-PCR mix. 2. Carry out five cycles of PCR as outlined in Table 4. After cycles 1 and 2, briefly pause the program and resuspend the beads, as they tend to settle quickly.
558
Rebekka Scholz et al.
Table 4 Program for cDNA-PCR 1 Step
Temperature
Time
1
95 °C
3 min
2
98 °C
30 s
3
65 °C
45 s
4
72 °C
3 min
5
4 °C
Hold
Comments
Go to step 2: total 5 cycles
Table 5 Program for cDNA qPCR Step
Temperature
Time
1
95 °C
3 min
2
98 °C
30 s
3
65 °C
20 s
4
72 °C
3 min
5
4 °C
Hold
Comments
Go to step 2: total 20 cycles
3. In the meanwhile, prepare the qPCR master mix (plan for duplicates for each reaction and at least one negative control reaction). Add 7.5 μL of the master mix to the necessary amount of wells on a qPCR plate. 4. After 5 cycles of PCR, immediately move the reaction tubes to ice, and then, take 2.5 μL of each sample and add it to the qPCR plate. Do this twice for every reaction. Add 2.5 μL of water to the negative control. 5. Place the PCR at 4 °C. Run the qPCR with the conditions outlined in Table 5 (see Note 32). 6. To determine the number of PCR cycles still needed, subtract the baseline (negative control), and then calculate 1/3 of the maximum fluorescence intensity for each well. Identify the cycle where this fluorescence intensity was first reached and use this number as the remaining cycles for the PCR. 7. Run the remaining number of PCR cycles according to Table 6 (see Note 33). 8. Finally, place the PCR tubes on a magnetic stand to separate PCR products in the supernatant from the Dynabeads and transfer the supernatant to new 0.2 mL PCR tubes. Directly
Joint Transcriptomic and Epigenetic Analysis in Microglia
559
Table 6 Program for cDNA-PCR 2 Step
Temperature
Time
1
95 °C
30 s
2
98 °C
30 s
3
65 °C
45 s
4
72 °C
3 min
5
4 °C
Hold
Comments
Go to step 2: total N cycles
resuspend the beads in 100 μL of STE buffer and store them at 4 °C (see Note 34). 9. Purify the cDNA with 0.6× AMPure beads. For this, add 30 μL of thoroughly resuspended, RT-equilibrated beads to each 50 μL reaction, mix well, and incubate at RT for 5 min. 10. Place the tubes on a magnetic stand for 3 min to separate supernatant and beads. Discard the supernatant and wash the beads with 100 μL of freshly prepared 80% ethanol for 30 s. Do not remove the tubes from the magnetic stand, and do not resuspend the beads at this step (see Note 35). 11. Remove the ethanol from the beads, then quickly spin the tubes down, and place them back on the magnetic stand. Now, remove the remaining ethanol with a 10 μL pipette tip. Leave the tubes open to dry the beads—they will turn from shiny to matte (see Note 36). 12. Remove the tubes from the magnet and resuspend the beads in 10.5 μL of buffer EB. Incubate off the magnet for 3 min. 13. Bring the tubes back on the magnetic stand, let the solution clear for 1 min, and then transfer the supernatant containing the eluted, purified cDNA to new tubes (see Note 37). 14. Dilute the cDNA 1:3 with ultrapure nuclease-free water and use 1 μL to measure the concentration by Qubit (see Note 38). Use the remaining 2 μL to check the fragment size distribution on a HS D5000 tape or similar (see Fig. 3a, b; see Note 39). 3.10 SHARE-seq on Microglial Nuclei— Tagmentation and RNA Library Preparation
1. Depending on the cDNA yield from Subheading 3.9, use 20 or 50 ng for subsequent tagmentation (see Note 40). 2. Fill every reaction up to 20 μL with nuclease-free water. 3. Prepare the needed RNA-Tn5 dilutions according to the specific activity of the respective Tn5 batch (see Note 41). 4. Add 25 μL of 2× Tris–DMF to each 20 μL of cDNA and mix well by pipetting. Then, add 5 μL of the appropriate single-
560
Rebekka Scholz et al.
Fig. 3 Fragment size distribution of the amplified SHARE-seq libraries. Agilent 4200 TapeStation profiles of the cDNA prior to tagmentation (a lane 1, b), the final snATAC-seq library (a lane 2, c), and the final snRNA-seq library (d–f). Lane 3 (d) and panel (e) depict the ideal size distribution after tagmentation. Lanes 4 and 5 (d) represent undertagmented (4) or overtagmented (5) libraries, also depicted in (f) (upper and lower panel, respectively). L: DNA ladder
loaded Tn5 dilution. Mix by pipetting without introducing bubbles. 5. Incubate the reactions at 55 °C for 5 min, shaking at 300 rpm. 6. Immediately continue by purifying the tagmented cDNA with the MinElute PCR Purification Kit. To this end, add 250 μL of buffer PB per sample and follow the handbook instructions (Qiagen, 28004, HB-2069; see Subheading 3.8). Elute twice with 10.5 μL of buffer EB. 7. For the final PCR, add 29 μL of the Tagmentation PCR mix to 20 μL of purified, tagmented cDNA. Then, add 1 μL of a library-specific Ad1 primer (25 μM) to each reaction. 8. Mix well and carry out the PCR using the conditions stated in Table 7. 9. Purify the amplified libraries with 0.7× AMPure beads (35–50 μL reaction) and elute to 10.5 μL of buffer EB. Take 1 μL of each final RNA library and dilute it with 2 μL of ultrapure nuclease-free water. Use 1 μL of the dilution to
Joint Transcriptomic and Epigenetic Analysis in Microglia
561
Table 7 Program for cDNA tagmentation PCR Step
Temperature
Time
1
75 °C
5 min
2
98 °C
30 s
3
98 °C
10 s
4
65 °C
30 s
5
72 °C
1 min
6
4 °C
Hold
Comments
Go to step 3: total 7 cycles
quantify the library with a Qubit and 2 μL to check the fragment size distribution on a HS D1000 tape or similar (see Fig. 3d, e; see Note 42). 3.11 SHARE-seq on Microglial Nuclei— ATAC Library Preparation
1. Prepare the ATAC-PCR mix and add 30 μL to each 20 μL of purified tagmented DNA from Subheading 3.8. 2. Perform five cycles of PCR at the conditions outlined in Table 8. 3. In the meantime, prepare the ATAC-qPCR master mix (plan for duplicates for each reaction and at least one negative control reaction). Add 7.5 μL of the master mix to the necessary amount of wells on a qPCR plate. 4. After five cycles of PCR, immediately move the reaction tubes to ice, and then, take 2.5 μL of each sample and add it to the qPCR plate. Perform this twice for every reaction. Add 2.5 μL of water to the negative control. 5. Place the PCR at 4 °C. Run the qPCR with the conditions described in Table 9 (see Note 32). 6. To determine the number of PCR cycles still needed, subtract the baseline (negative control), then calculate 1/3 of the maximum fluorescence intensity for each well. Identify the cycle where this fluorescence intensity was first reached and then use this number as the remaining cycles for the PCR. 7. Run the remaining number of PCR cycles according to Table 10 (see Note 43). 8. Purify the ATAC libraries with 1.2× AMPure beads (54 μL for a 50 μL reaction) as described in Subheading 3.9 (steps 9–13). Elute in 10.5 μL of buffer EB and transfer the supernatant containing the final libraries to new tubes. 9. Dilute the libraries 1:5 with ultrapure nuclease-free water and use 1 μL to measure the dsDNA concentration on a Qubit. Take 2 μL (it might be necessary to dilute this amount) and
562
Rebekka Scholz et al.
Table 8 Program for ATAC-PCR 1 Step
Temperature
Time
1
72 °C
5 min
2
98 °C
30 s
3
98 °C
10 s
4
65 °C
30 s
5
72 °C
1 min
6
4 °C
Hold
Comments
Go to step 3: total 5 cycles
Table 9 Program for ATAC-qPCR Step
Temperature
Time
1
98 °C
30 s
2
98 °C
10 s
3
65 °C
30 s
4
72 °C
1 min
5
4 °C
Hold
Comments
Go to step 2: total 20 cycles
Table 10 Program for ATAC-PCR 2 Step
Temperature
Time
1
98 °C
30 s
2
98 °C
10 s
3
65 °C
30 s
4
72 °C
1 min
5
4 °C
Hold
Comments
Go to step 2: total N cycles
determine the fragment size distribution on a HS D5000 tape. The ideal ATAC library shows a nucleosomal patterning, with peaks at approx. 265, 410, 585, and 775 bp (see Fig. 3a, c; see Note 44).
Joint Transcriptomic and Epigenetic Analysis in Microglia
3.12
Sequencing
563
1. Pool the final snRNA and snATAC libraries for sequencing (see Note 45). 2. Sequence the pool on the Illumina NextSeq or NovaSeq platforms with the following cycles: Read 1: 49 cycles, Index 1: 99 cycles, Index 2: 8 cycles, and Read 2: 49 cycles (see Note 46). Aim for a read depth of at least 25,000 reads/cell.
3.13
Preprocessing
1. All preprocessing steps are performed in a containerized platform-independent environment. Therefore, Docker needs to be installed (https://www.docker.com) [32]. 2. To set up the analysis pipeline, pull the Docker image provided on Docker Hub (https://hub.docker.com/repository/ docker/rebekkascholz/share-seq_repo). 3. Download all necessary scripts and additional files provided here: https://github.com/masai1116/SHARE-seqalignmentV2/ and replicate the directory structure [27]. 4. Since the analysis pipeline does not support demultiplexing by cell barcodes, the samples will be separated with deML beforehand (https://github.com/grenaud/deML) [33]. For this, download a .zip file of the Git repository and unpack it. 5. If not already supplied by the sequencing facility, run bcl2fastq on the bcl output to generate four fastq files (for both reads and both indices). Use the placeholder sample sheet (SampleSheet. csv) provided in the SHARE-seq-alignmentV2 GitHub repositor y (https://github.com/masai1116/SHARE-seqalignmentV2/) (see Note 47). 6. Blc2fastq can be run within the provided Docker container with the following command: docker run -it -v /your/path/:/your/path/ \ --rm rebekkascholz/share-seq_repo:V2 bcl2fastq \ --runfolder-dir /your/path/to/bcl/files/ \ --output-dir /your/path/to/output/fastq/ \ --sample-sheet /your/path/to/SampleSheet.csv \ --create-fastq-for-index-reads --no-lane-splitting
7. For demultiplexing with deML, an index file including all round 1 barcode and i5 index combinations for each biological sample needs to be prepared and saved in the fastq file directory. To create this file, the “deML-index_share-seq.R” script is provided at https://github.com/rebekkascholz/deML-index_ share-seq can be utilized. Save the index file in the deML directory. 8. Start an interactive Docker container session and mount the path to deML and the demultiplexed fastqs: docker run -it -v /your/path/to/output/fastq/:/data/fastq/ \ -v /your/path/to/deML/:/deML/ \ --name deML_demux rebekkascholz/share-seq_repo:V2 /bin/bash
9. First, run make in the deML directory. Then, start demultiplexing with the following command:
564
Rebekka Scholz et al. src/deML -i /deML/index_SHARE.txt \ -f /data/fastq/Undetermined_S1_R1_001.fastq.gz \ -r /data/fastq/Undetermined_S1_R4_001.fastq.gz \ -if1 /data/fastq/Undetermined_S1_R2_001.fastq.gz \ -if2 /data/fastq/Undetermined_S1_R3_001.fastq.gz \ -o /data/fastq/deML-output/demultiplexed. \ -s /data/fastq/deML-output/summary.txt \ -e /data/fastq/deML-output/failed-barcodes.txt
10. For the preprocessing pipeline, create one .yaml file per biological sample with the respective i5 indices for snRNA and snATAC libraries. 11. Rename the fastq files to match the input format for the “SHARE_seqV2_example.sh” script. 12. In the same script, adjust the paths to the running directories and the project names and options. 13. Add the project names and the respective i5 library indices to the configuration file (.yaml). 14. Start the analysis with the following command: docker run --name share-seq_V2 \ -v /your/path/to/raw/data/:/mnt/users/sai/Script/Split-seq_Sai/example_fastq/ \ -v /your/path/to/output/:/mnt/users/sai/Script/Split-seq_Sai/example_output/ \ -v /your/path/to/scripts/files/:/mnt/users/sai/Script/Split-seq_Sai/ \ rebekkascholz/share-seq_repo:V2 \ /mnt/users/sai/Script/Split-seq_Sai/SHARE_seqV2_example.sh &> LOG.txt &
15. The resulting fragment files (snATAC-seq) or count matrices (snRNA-seq) can be used for processing the data with standard tools, such as Seurat [34]. Figure 4 illustrates exemplary readouts from the SHARE-seq data.
4
Notes 1. We experienced strong purity and activity differences between in-house produced Tn5 batches and switched to buying Tn5 from Diagenode with higher activity and small batch effects only. 2. Other nuclei isolation kits can also be utilized to extract nuclei from frozen brain tissue; however, the lysis time might need to be adjusted with different lysis buffers. Moreover, it is crucial to include myelin removal by sucrose gradient centrifugation; therefore, usage of this kit is recommended. 3. PMSF is an irreversible serine protease inhibitor and highly toxic, only handle with the appropriate safety measures and under a fume hood. Refer to the safety data sheet for further information.
Joint Transcriptomic and Epigenetic Analysis in Microglia
565
Fig. 4 Exemplary readouts from the SHARE-seq data. Uniform manifold approximation and projection (UMAP) visualization of transcriptomic (a), chromatin accessibility (b), and joint transcriptomic and epigenetic SHAREseq data (c). Chromatin accessibility tracks aggregated by cell type for the microglia marker gene Siglech [35] are shown in (d). The cluster-specific enrichment of the binding motif for transcription factor SPI1 (PU.1) is depicted in (e). Panel (f) represents the accuracy of transferring cell labels from snRNA-seq to snATAC-seq data. Data analysis was performed with Seurat V4 [34] and Signac [36]. MG microglia, AG astroglia, Olig oligodendrocytes, Neu neurons, OPC oligodendrocyte precursor cells
4. The final concentration in 50 μL reaction volume will be 1× and 0.5 μM, respectively. 5. The annealed oligonucleotides can be stored at -20 °C for at least 1 month. 6. For unloaded Tn5 from Diagenode, add 5 μL of Tn5 stock to the 5 μL oligonucleotide mix. 7. Make sure to not extend the incubation time, as the Tn5 will lose activity after a prolonged time at RT. 8. Diagenode Tn5 now has a concentration of 12.5 μM. 9. In our hands, the loaded Tn5 could be used for several months without losing activity as long as it was stored at -20 °C and not subjected to repeating temperature changes. 10. Use tightly sealing plate covers, i.e., adhesive aluminum foil instead of plastic covers. Due to the high temperatures needed for annealing, covers sometimes loosen at the edges of the
566
Rebekka Scholz et al.
plate. Fill up the wells with reduced volume to 10 μL with ultrapure nuclease-free water in that case. 11. The lysis time depends on the sample and the degree of homogenization. A small fraction of the sample (after myelin removal) can be stained with trypan blue to determine nuclei shape, size and numbers, debris, and intact cells. 12. Pre-wetting the filter helps to reduce nuclei loss. 13. If the filter retains lots of material, the smooth end of the plunger of a 5 mL syringe can be used to recover some additional material. 14. We have found that, for consistent results, it is best to use fresh, unopened FA each time. 15. Counting nuclei at this step, as suggested in the original protocols, is highly unreliable in our hands due to the high myelin content; therefore, we use standardized volumes for our samples. 16. The original protocol states that centrifugation must be done at 30,000 g; however, we have found this to be unnecessary. In our hands, the final nuclei yield was not affected by using a conventional centrifuge. 17. Sort into BSA-coated borosilicate glass FACS tubes or 1.5 mL microcentrifuge tubes filled with 300 μL PBSI to minimize nuclei loss. 18. Nuclei can be counted after sorting to test recovery—for our instrument, we expect a recovery of around 85%. 19. The number of reactions depends on the input nuclei number and the Tn5 activity. We normally split each sample to have an input of 20,000 nuclei/reaction and have also tested various amounts of Tn5 on this nuclei number. Since every batch of Tn5 (whether self-produced or commercial) has a different activity, this needs to be tested individually. 20. This is a stopping point that was necessary due to the length of the protocol; however, one can also directly proceed with barcoding. 21. A multi-dispenser pipette can be used to distribute the nuclei suspensions and blocking strand solutions on the barcoding plates, but is not necessary. The use of multichannel pipettes is discouraged to avoid loss of nuclei or oligonucleotides in reagent reservoirs. 22. Pooling at this step increases the possible barcode combinations for each sample and thereby reduces the likelihood of barcode doublets. To estimate the collision rate, we followed the calculation by Ma and colleagues derived from a birthday
Joint Transcriptomic and Epigenetic Analysis in Microglia
567
paradox solution (http://matt.might.net/articles/countinghash-collisions/) [27]: D -1 n-D þ D D
n
Using 48 wells of the first barcoding plate and 96 wells of plates 2 and 3, the possible barcode combinations (D) are 48 × 96 × 96 = 442,368. Assuming a number of 10,000 nuclei/sub-library (n), the expected number of barcode collisions is 112.2. Therefore, the collision rate is 112.2/ 10,000 = 1.1%. 23. The volume of 55 μL/well accounts for a pipetting loss of 5 μL/well. 24. The buffer volume for resuspending the nuclei depends on the nuclei number. We found this to be adequate for an input of 200,000 nuclei, and the buffer volume might need to be increased for more cells. 25. Because of the low number of nuclei, counting is highly unreliable at this step. Generally, we experienced a recovery of roughly 20,000–40,000 nuclei when starting with 200,000 nuclei in total, so we always split our sample into two reactions at this step without counting every time. 26. This step will also lead to the lysis of the prior intact nuclei and the release of the cDNA and tagmented DNA into the suspension. 27. When handling the beads, avoid centrifugation and drying the beads once the RNA is bound. 28. In this step, the RNA with the biotinylated RT primer will bind to the streptavidin-coated beads, while the barcoded chromatin fragments will stay in the supernatant. 29. Since double-stranded cDNA is present from here on, no RI is needed anymore. 30. This storage step was added for time reasons; however, ideally, one would directly continue with the PCR from the beads (see Subheading 3.9). 31. To continue with the library preparation at a later time point, the eluted DNA can be stored at -20 °C for several days without notable quality reduction. However, repeated freeze– thawing cycles should be avoided. 32. The qPCR is performed to prevent over-amplification of the DNA leading to reduced library complexity. Therefore, the PCR will be stopped while still in the exponential phase of DNA amplification.
568
Rebekka Scholz et al.
33. For one reaction with approx. 15,000 nuclei left after barcoding, expect 7–9 additional cycles. 34. Storing the Dynabeads in buffer at 4 °C allows for a repetition of the PCR from the beads at a later time point (tested for a maximum of 1 week later), in case tagmentation fails. However, not all cDNA fragments will be recovered, leading to lower yield and library complexity. 35. Due to its hygroscopic properties, the ethanol dilution needs to be prepared freshly immediately before use. 36. It is essential to not over-dry the beads at this step since this will affect the DNA recovery significantly. 37. At this point, the cDNA can be stored at 4 °C overnight or 20 °C for some days. 38. Expect an undiluted concentration of 5–15 ng/μL. 39. A single peak with an average size between 1000 and 1500 bp should be expected (see Fig. 3a, b). Generally, more PCR cycles will lead to a shift of the average fragment size toward shorter fragments. Sometimes, especially with yields below 20 ng, primer residues can be observed after purification. If they make up more than 15% of the total library, consider re-purifying the cDNA, since the small fragments will influence the tagmentation performance. 40. We successfully tagmented 10 ng of cDNA; however, the reactions will need to be scaled down accordingly. For reproducibility, it is best to use at least 20 ng/reaction. 41. The Tn5 concentration needed for successful tagmentation of the cDNA depends on the input concentration and average fragment size and on the enzymatic activity; therefore, each new Tn5 batch needs to be tested individually for the optimal outcome. 42. Ideally, the cDNA tagmentation results in one broad peak with a fragment size of 400–700 bp, and a concentration of 4–6 ng/ μL. For examples of sequenceable, over-, and undertagmented libraries, refer to Fig. 3d–f. 43. For one reaction with approx. 15,000 nuclei left after barcoding, expect 5–7 additional cycles. 44. Compared to classical ATAC-seq experiments, there is a right shift of the fragment sizes due to the 99 bp long cell barcode added to each fragment. 45. ATAC and RNA can be sequenced together; however, the ATAC library fragments tend to exceed the clustering ability of the RNA libraries, so we plan ~20% more reads for the RNA than for the ATAC libraries when sequenced together.
Joint Transcriptomic and Epigenetic Analysis in Microglia
569
46. Sequencing can be done on the Illumina NextSeq and NovaSeq platforms; however, for sequencing on the NextSeq, a customized script by Illumina is necessary. 47. Separate fastq files for the index reads are necessary as input for deML. References 1. Prinz M, Priller J (2014) Microglia and brain macrophages in the molecular age: from origin to neuropsychiatric disease. Nat Rev Neurosci 15(5):300–312. https://doi.org/10.1038/ nrn3722 2. Tremblay M-E` (2011) The role of microglia at synapses in the healthy CNS: novel insights from recent imaging studies. Neuron Glia Biol 7(1):67–76. https://doi.org/10.1017/ S1740925X12000038 3. Lawson LJ, Perry VH, Dri P, Gordon S (1990) Heterogeneity in the distribution and morphology of microglia in the normal adult mouse brain. Neuroscience 39(1):151–170. https://doi.org/10.1016/0306-4522(90) 90229-w 4. Schmid CD, Sautkulis LN, Danielson PE, Cooper J, Hasel KW, Hilbush BS, Sutcliffe JG, Carson MJ (2002) Heterogeneous expression of the triggering receptor expressed on myeloid cells-2 on adult murine microglia. J Neurochem 83(6):1309–1320. https://doi. org/10.1046/j.1471-4159.2002.01243.x 5. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting microglial cells are highly dynamic surveillants of brain parenchyma in vivo. Science 308(5726):1314–1318. https://doi.org/10.1126/science.1110647 6. Davalos D, Grutzendler J, Yang G, Kim JV, Zuo Y, Jung S, Littman DR, Dustin ML, Gan W-B (2005) ATP mediates rapid microglial response to local brain injury in vivo. Nat Neurosci 8(6):752–758. https://doi.org/10. 1038/nn1472 7. Edler MK, Mhatre-Winters I, Richardson JR (2021) Microglia in aging and Alzheimer’s disease: a comparative species review. Cell 10(5). https://doi.org/10.3390/cells10051138 8. Matcovitch-Natan O, Winter DR, Giladi A, Vargas Aguilar S, Spinrad A, Sarrazin S, Ben-Yehuda H, David E, Zelada Gonza´lez F, Perrin P, Keren-Shaul H, Gury M, LaraAstaiso D, Thaiss CA, Cohen M, Bahar Halpern K, Baruch K, Deczkowska A, Lorenzo-Vivas E, Itzkovitz S, Elinav E, Sieweke MH, Schwartz M, Amit I (2016) Microglia development follows a stepwise program to regulate brain homeostasis. Science
353(6301):aad8670. https://doi.org/10. 1126/science.aad8670 9. Silvin A, Uderhardt S, Piot C, Da Mesquita S, Yang K, Geirsdottir L, Mulder K, Eyal D, Liu Z, Bridlance C, Thion MS, Zhang XM, Kong WT, Deloger M, Fontes V, Weiner A, Ee R, Dress R, Hang JW, Balachander A, Chakarov S, Malleret B, Dunsmore G, Cexus O, Chen J, Garel S, Dutertre CA, Amit I, Kipnis J, Ginhoux F (2022) Dual ontogeny of disease-associated microglia and disease inflammatory macrophages in aging and neurodegeneration. Immunity 55(8): 1448–1465.e1446. https://doi.org/10. 1016/j.immuni.2022.07.004 10. Olah M, Patrick E, Villani A-C, Xu J, White CC, Ryan KJ, Piehowski P, Kapasi A, Nejad P, Cimpean M, Connor S, Yung CJ, Frangieh MA (2018) A transcriptomic atlas of aged human microglia. Nat Commun 9(1):539. https:// doi.org/10.1038/s41467-018-02926-5 11. Hammond TR, Dufort C, Dissing-Olesen L, Giera S, Young A, Wysoker A, Walker AJ, Gergits F, Segel M, Nemesh J, Marsh SE, Saunders A, Macosko E, Ginhoux F, Chen J, Franklin RJM, Piao X (2019) Single-cell RNA sequencing of microglia throughout the mouse lifespan and in the injured brain reveals complex cell-state changes. Immunity 50(1): 253–271.e256. https://doi.org/10.1016/j. immuni.2018.11.004 12. Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, David E, Baruch K, Lara-Astaiso D, Toth B, Itzkovitz S, Colonna M, Schwartz M, Amit I (2017) A unique microglia type associated with restricting development of Alzheimer’s disease. Cell 169(7):1276–1290.e1217. https://doi.org/10.1016/j.cell.2017.05.018 13. Masuda T, Sankowski R, Staszewski O, Bo¨ttcher C, Amann L, Sagar SC, Nessler S, Kunz P, van Loo G, Coenen VA, Reinacher PC, Michel A, Sure U, Gold R, Gru¨n D, Priller J, Stadelmann C, Prinz M (2019) Spatial and temporal heterogeneity of mouse and human microglia at single-cell resolution. Nature 566(7744):388–392. https://doi. org/10.1038/s41586-019-0924-x
570
Rebekka Scholz et al.
14. Yeh H, Ikezu T (2019) Transcriptional and epigenetic regulation of microglia in health and disease. Trends Mol Med 25(2):96–111. https://doi.org/10.1016/j.molmed.2018. 11.004 15. Buenrostro JD, Giresi PG, Zaba LC, Chang HY, Greenleaf WJ (2013) Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA -binding proteins and nucleosome position. Nat Methods 10(12):1213–1218. https:// doi.org/10.1038/nmeth.2688 16. Lareau CA, Duarte FM, Chew JG, Kartha VK, Burkett ZD, Kohlway AS, Pokholok D, Aryee MJ, Steemers FJ, Lebofsky R, Buenrostro JD (2019) Droplet-based combinatorial indexing for massive-scale single-cell chromatin accessibility. Nat Biotechnol 37(8):916–924. https:// doi.org/10.1038/s41587-019-0147-6 17. Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM, Hao Y, Stoeckius M, Smibert P, Satija R (2019) Comprehensive integration of single-cell data. Cell 177(7): 1888–1902.e1821. https://doi.org/10. 1016/j.cell.2019.05.031 18. Duren Z, Chen X, Zamanighomi M, Zeng W, Satpathy AT, Chang HY, Wang Y, Wong WH (2018) Integrative analysis of single-cell genomics data by coupled nonnegative matrix factorizations. Proc Natl Acad Sci U S A 115(30): 7723–7728. https://doi.org/10.1073/pnas. 1805681115 19. Liu J, Huang Y, Singh R, Vert J-P, Noble WS (2019) Jointly embedding multiple single-cell omics measurements. Algor Bioinform 143. https://doi.org/10.4230/LIPIcs.WABI. 2019.10 20. Welch JD, Kozareva V, Ferreira A, Vanderburg C, Martin C, Macosko EZ (2019) Single-cell multi-omic integration compares and contrasts features of brain cell identity. Cell 177(7):1873–1887.e1817. https://doi.org/10.1016/j.cell.2019.05.006 21. Lin Y, Wu T-Y, Wan S, Yang JYH, Wong WH, Wang YXR (2022) scJoint integrates atlas-scale single-cell RNA-seq and ATAC-seq data with transfer learning. Nat Biotechnol 40(5): 703–710. https://doi.org/10.1038/s41587021-01161-6 22. Zhang Z, Yang C, Zhang X (2022) scDART: integrating unmatched scRNA-seq and scATAC-seq data and learning cross-modality relationship simultaneously. Genome Biol 23(1):139. https://doi.org/10.1186/ s13059-022-02706-x 23. Cao J, Cusanovich DA, Ramani V, Aghamirzaie D, Pliner HA, Hill AJ, Daza RM, McFaline-Figueroa JL, Packer JS,
Christiansen L, Steemers FJ, Adey AC, Trapnell C, Shendure J (2018) Joint profiling of chromatin accessibility and gene expression in thousands of single cells. Science 361(6409):1380–1385. https://doi.org/10. 1126/science.aau0730 24. Liu L, Liu C, Quintero A, Wu L, Yuan Y, Wang M, Cheng M, Leng L, Xu L, Dong G, Li R, Liu Y, Wei X, Xu J, Chen X, Lu H, Chen D, Wang Q, Zhou Q, Lin X, Li G, Liu S, Wang Q, Wang H, Fink JL, Gao Z, Liu X, Hou Y, Zhu S, Yang H, Ye Y, Lin G, Chen F, Herrmann C, Eils R, Shang Z, Xu X (2019) Deconvolution of single-cell multiomics layers reveals regulatory heterogeneity. Nat Commun 10(1):470. https://doi.org/ 10.1038/s41467-018-08205-7 25. Zhu C, Yu M, Huang H, Juric I, Abnousi A, Hu R, Lucero J, Behrens MM, Hu M, Ren B (2019) An ultra high-throughput method for single-cell joint analysis of open chromatin and transcriptome. Nat Struct Mol Biol 26(11): 1063–1070. https://doi.org/10.1038/ s41594-019-0323-x 26. Chen S, Lake BB, Zhang K (2019) Highthroughput sequencing of the transcriptome and chromatin accessibility in the same cell. Nat Biotechnol 37(12):1452–1457. https:// doi.org/10.1038/s41587-019-0290-0 27. Ma S, Zhang B, LaFave LM, Earl AS, Chiang Z, Hu Y, Ding J, Brack A, Kartha VK, Tay T, Law T, Lareau C, Hsu Y-C, Regev A, Buenrostro JD (2020) Chromatin potential identified by shared single-cell profiling of RNA and chromatin. Cell 183(4): 1103–1116.e1120. https://doi.org/10. 1016/j.cell.2020.09.056 28. Zheng GXY, Terry JM, Belgrader P, Ryvkin P, Bent ZW, Wilson R, Ziraldo SB, Wheeler TD, McDermott GP, Zhu J, Gregory MT, Shuga J, Montesclaros L, Underwood JG, Masquelier DA, Nishimura SY, Schnall-Levin M, Wyatt PW, Hindson CM, Bharadwaj R, Wong A, Ness KD, Beppu LW, Deeg HJ, McFarland C, Loeb KR, Valente WJ, Ericson NG, Stevens EA, Radich JP, Mikkelsen TS, Hindson BJ, Bielas JH (2017) Massively parallel digital transcriptional profiling of single cells. Nat Commun 8:14049. https://doi.org/10.1038/ ncomms14049 29. Rosenberg AB, Roco CM, Muscat RA, Kuchina A, Sample P, Yao Z, Graybuck LT, Peeler DJ, Mukherjee S, Chen W, Pun SH, Sellers DL, Tasic B, Seelig G (2018) Singlecell profiling of the developing mouse brain and spinal cord with split-pool barcoding. Science 360(6385):176–182. https://doi.org/ 10.1126/science.aam8999
Joint Transcriptomic and Epigenetic Analysis in Microglia 30. Nerlov C, Graf T (1998) PU.1 induces myeloid lineage commitment in multipotent hematopoietic progenitors. Genes Dev 12(15): 2403–2412. https://doi.org/10.1101/gad. 12.15.2403 31. Picelli S, Bjo¨rklund AK, Reinius B, Sagasser S, Winberg G, Sandberg R (2014) Tn5 transposase and tagmentation procedures for massively scaled sequencing projects. Genome Res 24(12):2033–2040. https://doi.org/10. 1101/gr.177881.114 32. Merkel D (2014) Docker: lightweight Linux containers for consistent development and deployment. Linux J 2014(239):2. 33. Renaud G, Stenzel U, Maricic T, Wiebe V, Kelso J (2015) deML: robust demultiplexing of Illumina sequences using a likelihood-based approach. Bioinformatics 31(5):770–772. https://doi.org/10.1093/bioinformatics/ btu719 34. Hao Y, Hao S, Andersen-Nissen E, Mauck WM, Zheng S, Butler A, Lee MJ, Wilk AJ,
571
Darby C, Zager M, Hoffman P, Stoeckius M, Papalexi E, Mimitou EP, Jain J, Srivastava A, Stuart T, Fleming LM, Yeung B, Rogers AJ, McElrath JM, Blish CA, Gottardo R, Smibert P, Satija R (2021) Integrated analysis of multimodal single-cell data. Cell 184(13): 3573–3587.e3529. https://doi.org/10. 1016/j.cell.2021.04.048 35. Konishi H, Kobayashi M, Kunisawa T, Imai K, Sayo A, Malissen B, Crocker PR, Sato K, Kiyama H (2017) Siglec-H is a microgliaspecific marker that discriminates microglia from CNS-associated macrophages and CNS-infiltrating monocytes. Glia 65(12): 1927–1943. https://doi.org/10.1002/glia. 23204 36. Stuart T, Srivastava A, Madad S, Lareau CA, Satija R (2021) Single-cell chromatin state analysis with Signac. Nat Methods 18(11): 1333–1341. https://doi.org/10.1038/ s41592-021-01282-5
INDEX A Adipose tissue ................................ 4, 136, 150, 153, 270, 307–309, 311–318, 342 Adipose tissue macrophages ...............4, 5, 256, 307–321 AIM2 ................................................. 411, 412, 414, 415, 418, 419, 427 Alveolar macrophages (AMs) .........................2, 4, 14–18, 20, 231–251, 273, 277, 278, 338, 348, 390, 402, 404, 405, 463, 464, 471–475, 478 Antibody .......................................... 54, 82, 93, 100, 101, 104–106, 108, 109, 111–113, 118–120, 123, 125, 131, 134–136, 141–143, 145, 146, 151, 153–157, 162–164, 166, 167, 172, 174–181, 185, 191, 192, 197, 199–201, 205, 213–216, 220, 221, 225, 226, 228, 255, 259–262, 265, 266, 272, 273, 279, 281, 282, 284–285, 288, 289, 293–295, 297, 298, 300–305, 309, 311–318, 320, 321, 326, 331, 332, 342, 344, 345, 351, 352, 354, 359, 365, 369, 372–374, 378–382, 384, 386, 387, 391, 394, 402, 432, 434, 436–439, 443–446, 448–450, 455–458, 467, 469, 470, 475, 476, 510, 531, 547, 553 Apoptotic cells (ACs)............................... 46, 49, 56, 214, 256, 389–391, 393, 394, 397–402, 404, 405 ASC ................................... 407, 408, 431, 432, 434–438, 443, 444, 449–451 ASC speck ................................... 432, 434, 437–439, 450 Autofluorescence.........................76, 157, 166, 195, 196, 211, 219–221, 223, 227, 269, 274, 279, 293, 305, 405, 450, 476 Automatic cell count............................................ 505, 517 Autophagosomes.........................................46–54, 56–60, 62, 63, 66 Autophagy flux .............................................47, 50–63, 66
B Benzyl alcohol and benzyl benzoate (BABB) .... 300–305 Batch analysis................................................................. 505 Bioinformatics ...................................................... 486–490 Bone marrow (BM) ......................................2, 11, 13–18, 20–22, 26, 29–31, 34, 35, 99, 121, 125, 211, 324, 350, 390, 414
Brain................................ 2, 3, 12, 14, 16, 45, 46, 48–52, 57, 65, 66, 82, 83, 85, 86, 88–90, 93, 95, 96, 101, 129–131, 133, 134, 140, 254, 255, 257, 260, 263–266, 298, 299, 303, 338, 366, 415, 416, 482, 494, 499, 506, 510, 543–545, 552, 554, 564 Bronchoalveolar lavage ................................................. 476 Brown adipose tissue (BAT)...................... 256, 307, 313, 315, 317
C Caspase-1 ................................... 407–409, 421–423, 427, 431, 432, 436, 443 Cell isolation........................................................... 66, 254 CellProfiler ...................................................505–507, 513 Chromatin accessibility ............................... 544, 545, 565 Clearing tissue methods................................................ 298 Co-Detection by Indexing (CODEX)............... 281, 282, 284–286, 289–291, 293–295 Colon ........................................................... 187, 291, 292 Colony stimulating factor 1 (CSF1) ...........100–102, 471 Colony stimulating factor 1 receptor (CSF1R)............................................100–102, 105 Conditioned medium ........................232, 234–237, 246, 311, 320, 417 Confocal microscopy ................................. 52, 59, 60, 62, 65, 66, 184, 207–229, 438, 531 Cre/loxP ...................................................................30, 33 CRISPR ...................................... 486, 490–492, 494, 498 CRISPR-Cas9......................................485, 490–492, 496 Cxcr4 ...................................................26, 27, 29, 30, 140
D Density gradient centrifugation ................................... 119 Dermis .................................................... 2, 159–167, 271, 272, 279, 341 Development ...................................... 1–5, 11–14, 16–19, 22, 28, 31–35, 46, 72, 81, 99–101, 112, 113, 118, 129, 136, 149, 171, 172, 208, 269, 363, 377, 481, 505, 544 Diet induced obesity ..................................................... 150
Elvira Mass (ed.), Tissue-Resident Macrophages: Methods and Protocols, Methods in Molecular Biology, vol. 2713, https://doi.org/10.1007/978-1-0716-3437-0, © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
573
TISSUE-RESIDENT MACROPHAGES: METHODS AND PROTOCOLS
574 Index
DNA ....................................... 31, 75, 77, 103, 241, 248, 259–261, 282, 284, 381, 382, 444, 450, 453–459, 485, 487, 490, 491, 498, 499, 545, 546, 548, 549, 551, 553, 555–557, 560, 567, 568 Drosophila macrophages ........................................ 71, 459
E Efferocytosis ........................................389, 390, 402, 404 Electron microscopy .................................................52, 65 Embryo ......................................2, 3, 34, 81, 82, 95, 101, 103, 113, 129–131, 133–137, 298, 300, 303, 304, 486, 490–492, 496, 498 Energetic metabolism ................................................... 364 Engulfment........................................................... 389, 390 Epigenetics ....................................... 15, 17, 18, 544, 565 Erythro-myeloid progenitors (EMPs) ......................2, 13, 14, 22–28, 31, 34, 36, 37, 130, 139, 140
F Fate-mapping ..................................... 1, 11–37, 129–137, 139–147, 161, 183, 256, 349–351 Fc receptor (FcR) ....................................... 359, 377, 467, 469, 475 Flow cytometry ................................. 60, 82, 88, 95, 100, 123, 125, 135, 144, 145, 154, 157, 160–164, 167, 172, 176, 184, 192–193, 199–229, 242–244, 250, 254, 256, 264, 269, 279, 291, 295, 337, 350, 351, 353–359, 364, 367, 378, 385, 389, 467, 469, 470, 478, 481, 496, 543 Flow cytometry analysis ..................................93, 95, 121, 130, 131, 143, 149–157, 159–167, 171–181, 183–197, 200, 226, 269–279, 469, 475, 478, 499 Fluorescence-activated cell sorting (FACS)........... 73, 74, 76, 84–87, 95, 119, 120, 123, 131–135, 141–145, 147, 151, 153, 154, 161–164, 167, 185, 189–192, 197, 200–205, 212, 213, 217–219, 225, 226, 233, 243, 351–354, 358, 364, 365, 367, 368, 375, 458, 475, 547
G Gasdermin D ........................................................ 408, 431 Genetic engineering ...................................................... 231 Genetic manipulation.................................................... 378 Glycolytic capacity................................................ 370, 371 G-quadruplex structures...................................... 453–459 Granulocyte-macrophage colony-stimulating factor (GM-CSF) ...................................... 232, 234–237, 246, 348, 410, 412–414, 464, 471
H Hematopoietic differentiation...................................... 467 Hematopoietic stem cells (HSCs)..................2–4, 13, 14, 22–27, 29–32, 34, 36, 118–124, 126, 127, 129, 139, 140
Hematopoietic waves ..........................2, 14, 26, 130, 141 Hemocytes........................................... 71–73, 75–77, 459 High dimensional.......................................................... 281 Host-pathogen fate mapping .............................. 350, 351 Host-pathogen interaction ........................................... 464 HSC transfer.................................................................. 118 Human hematopoietic stem cell (HSC) purification......................................................... 119
I IBA1..................................101, 102, 105, 106, 109, 110, 112, 505, 510, 511, 513, 516, 517 IDisco .........................................298, 299, 302, 303, 305 Image analysis.............................. 59, 113, 290–292, 312, 339, 342, 345, 434, 438, 520, 524, 531, 532 ImageJ...............................315, 328, 331, 339, 342, 345, 380, 382, 385, 457, 515, 521, 532 ImageStream ........................................................ 392, 400 Imaging.......................................... 33, 57, 59, 60, 63, 65, 100, 101, 109, 214, 226, 281–295, 297–305, 309, 312, 315, 318, 320, 324, 325, 328–334, 338–343, 385, 389, 434, 435, 438, 449, 450, 496, 499, 543 Imaging flow cytometry (IFC)........................... 389, 390, 392, 394, 399, 403 Immunodeficient mice........................118, 121, 465, 471 Immunofluorescence ................................... 66, 125, 297, 320, 424, 458 Immunofluorescence microscopy ................................ 125 Immunofluorescence staining ................... 127, 311–313, 317, 319 Immunoprecipitation (IP) ......................... 254, 255, 257, 264, 265, 458 Induced pluripotent stem cells (iPSCs) ............... 52, 464, 467, 471 Infiltrating macrophage ................................................ 171 Inflammasome ............................. 49, 407–428, 431–434, 436, 438–444, 446, 448, 450 Interleukin-34 (IL-34) ........................................ 100–102 Intersectional genetics ......................................... 484, 495 Intestinal macrophages .....................................4, 17, 160, 183–197 Intestine .............................. 2, 4, 85, 160, 183, 184, 187, 188, 194–196, 282, 286, 292, 330, 348, 400, 474, 506 Intra-pulmonary transfer ..................................... 463–478 Intravital .......................................................329, 337–345 Intravital imaging ....................................... 324, 338, 339, 342, 343, 345 In vivo imaging ...................................323–334, 337, 342 Isolation ......................47, 57, 72, 81–96, 121–123, 131, 144, 149–157, 183–197, 199–206, 231–251, 254, 255, 259, 260, 262, 265, 316, 373, 379, 383, 392–393, 408–411, 426, 547, 552, 554
TISSUE-RESIDENT MACROPHAGES: METHODS K Kidney....................................86, 93, 110, 171–181, 292, 293, 305, 307, 313, 338, 340–345, 506 Kidney-resident macrophages ............................ 171, 176, 179, 180, 343 Kupffer cells (KCs)................................... 2, 4, 12, 15, 16, 18, 20, 105, 134, 208, 210, 211, 220, 273, 276, 277, 338, 344, 345, 372, 464
L Lamina propria ....................................... 4, 184–191, 193, 195, 196, 266 LDH release ................................................ 448, 449, 451 Lentiviral transduction......................................... 231–251 Light sheet microscopy................................................. 297 Lipid-associated macrophages (LAMs).............. 210, 211, 221, 222, 224 Liver .......................................2–4, 12–15, 19, 24, 28, 34, 82, 83, 85, 86, 88, 89, 91, 93, 95, 99, 104, 110, 118, 125, 127, 129–134, 144, 154, 207–229, 260, 270, 271, 274, 276, 277, 282, 286, 292, 293, 298, 305, 338, 342–345, 353, 358, 464, 474, 481, 506 Loxp-cre-system ................................................... 349, 350 Lungs .....................................2, 4, 15–18, 104, 110, 147, 231, 235, 246, 266, 270, 271, 277, 278, 282, 286, 292, 347–360, 393, 463–465, 472–475, 477, 478, 481, 506 Lysosomes .................................................. 46–49, 51, 53, 59–62, 64
AND
PROTOCOLS Index 575
Monocytes ......................................... 1, 93, 99, 118, 129, 140, 149, 160, 171, 183–197, 208, 258, 274, 308, 324, 343, 347, 377, 411, 459, 464 Mononuclear phagocytes..................................82, 92, 93, 95, 96, 208 Mononuclear phagocyte system (MPS).......... 12, 99–101 Morphology.................................85, 100, 111, 247, 248, 254, 313, 324, 333, 334, 397–401, 404, 405, 421, 505, 517, 519, 520, 529, 531, 543 Morphometric analysis.................................................. 521 Motility ........................................................ 519, 520, 531 Mouse cytomegalovirus (MCMV) ...................... 350–353 mpeg1.1 .................................................................... 82, 83, 89–94, 96 mRNA sequencing ........................................................ 487 Multiplexed imaging ..................................................... 282 Muscularis externa ....................... 17, 184–189, 192–196
N NLRC4 ...................................... 410–412, 414, 415, 419, 421, 424–425, 427, 440 NLRP1........................................411, 412, 421, 422, 427 NLRP3................................ 49, 408, 411, 412, 414, 415, 419, 420, 422, 428, 436, 441, 447, 449 Non-alcoholic fatty liver disease (NAFLD)................207, 208, 210–212, 216, 226 Non-alcoholic steatohepatitis (NASH).............. 207–212, 216, 226 Nuclei isolation ..............................................76, 548, 564
O
M
Ontogeny........................... 1, 11–37, 118, 140, 149, 270
Macrophage cellular neighborhoods ........................... 292 Macrophage heterogeneity ........................ 210–211, 253, 281–295, 350, 389, 481–499 Macrophage localization............................................... 308 Macrophage ontogeny ..........................12, 18, 30, 35, 99 Macrophages ................................... 1, 11, 46, 71, 81, 99, 118, 129, 139, 149, 159, 171, 183, 208, 231, 253, 274, 281, 297, 308, 324, 338, 347, 363, 377, 390, 408, 432, 454, 463, 481, 505, 521, 543 Magnetic cell sorting ........................................... 254, 385 Metabolic dependencies ............................. 364, 370, 374 Microglia........................................ 1, 3, 4, 11, 12, 14–16, 27, 31, 45–66, 82, 90, 93, 95, 101, 105, 134, 140, 254–258, 263, 264, 338, 365, 366, 369, 407–428, 431–451, 464, 482, 486, 494–497, 506, 519, 543–569 Microscopy ......................... 57, 112, 225, 324, 337–345, 348, 358, 378, 379, 382, 383, 385, 387, 389, 390, 432, 434, 437, 458, 532, 533, 538 Mitochondrial dependence.................................. 370, 371
P p2ry12 ......................... 82, 83, 90, 93, 95, 365, 369, 517 Peritoneum ................................131, 135, 136, 202, 286, 328, 330, 331, 334, 357, 358, 392, 414 Phagocytes ................................................... 1, 51, 59, 183 Phagocytosis ................................ 1, 3, 48–51, 56, 63, 71, 118, 129, 323, 338, 350, 378, 389, 392, 394–396, 469, 481 Plasmatocytes ..............................................................1, 71 PMT................................................................................. 91 Pre-macrophage (pMacs)............... 2, 14, 22–24, 27, 130 Primary cell culture ....................................................... 231 Protein interaction ........................................................ 378 Protein synthesis ......................................... 364, 367, 370 Proximity ligation assay (PLA) ........................... 378–382, 384–387 Pyroptosis .................................. 421, 431–433, 435, 438, 440, 446, 450, 451
TISSUE-RESIDENT MACROPHAGES: METHODS AND PROTOCOLS
576 Index R
Recognition ..........................................20, 378, 389, 390, 408, 450, 496 Reporter rat ................................................................... 104 Resident tissue macrophages (RTMs) ...... 101, 269, 270, 275–278, 323–327, 334 Respiratory tract infection ................................... 350–353 RiboTag ............................................................... 253–266, 496, 499
S SCENITH ............................................................ 363–373 Self-renewal .......................................................... 159, 210 Single-cell resolution ........................................... 281, 364 Single nuclei RNA-sequencing (snRNA-seq).........71–77, 560, 564, 565 Skin .......................................... 20, 82, 83, 85–89, 92–95, 126, 127, 131, 140, 152, 153, 159–162, 164–167, 175, 187, 202, 246, 270–272, 274, 276, 277, 279, 312, 313, 321, 328–330, 338, 340, 341, 345, 348, 357, 358, 392, 393, 414, 416, 472, 474 Spatial proteomics ......................................................... 281 Spectral flow cytometry ............................. 219, 221, 223, 269, 270 Spleen...........................................85, 104, 110, 118, 121, 131, 140, 142–144, 154, 173, 175–178, 199, 270, 271, 274, 275, 277, 292, 293, 298, 305, 338, 342–345, 382 splitCre ................................................................ 264, 265, 483–499
T Tamoxifen............................................. 21, 24–27, 29, 30, 32, 34, 35, 103, 130, 135, 141, 143, 145, 146, 256, 257
Testis digestion.............................................................. 203 Testis macrophage................................................ 199–206 Time lapse microscopy......................................... 343, 344 Tissue macrophages ............................... 81–96, 140, 253, 256, 257, 331, 389, 481, 482, 486, 494 Tissue-resident macrophage (TRM) ................ 1–3, 5, 12, 16, 18, 20, 26, 29, 35, 45, 71, 102, 105, 109, 129, 139, 140, 149, 150, 253–266, 281, 282, 284, 337–345, 363–373, 389–405, 408, 463, 464, 499, 505, 506, 519, 520 Transcriptomics ............................... 17, 75, 96, 100, 254, 350, 481, 543, 544, 565 Translatome retrieval ........................................... 258, 264 Two-photon imaging.................................................... 331
W Western blot (WB) ..................................... 52–54, 57, 59, 60, 65, 66, 432, 435–436, 443–444 White adipose tissue (WAT) ...................... 104, 149–157, 271, 274, 275, 277, 279, 307, 308, 313, 318–320 Whole-mount .................................... 298, 303, 304, 309, 311–313, 316–320 Whole-mount imaging ........................................ 104, 302
Y Yolk sac ........................................2, 3, 11, 13, 22, 24, 25, 45, 99, 129, 139, 159, 324, 505
Z Zebrafish ........................................................... 81–96, 303