Migration and homing of lymphoid cells. Volume 1 [1st ed.] 9780429290763, 9780367259587, 9780367259600, 0367259583, 9781000697483, 1000697487

Citation preview

Migration and Homing of Lymphoid Cells Volume I

Editor

Alan J. Husband, Ph.D. Associate Professor of Immunology Faculty of Medicine University of Newcastle Newcastle, Australia

Boca Raton London New York

CRC Press is an imprint of the Taylor & Francis Group, an informa business

CRCPress Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 Reissued 2019 by CRC Press © 1988 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www. copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

A Library of Congress record exists under LC control number: Publisher's Note The publisher has gone to great lengths to ensure the quality of this reprint but points out that some imperfections in the original copies may be apparent. Disclaimer The publisher has made every effort to trace copyright holders and welcomes correspondence from those they have been unable to contact. ISBN 13: 978-0-367-25958-7 (hbk) ISBN 13: 978-0-367-25960-0 (pbk) ISBN 13: 978-0-429-29076-3 (ebk) Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com

PREFA CE Lymphoid cell migration is crucial to successful immune defense. The continued recir­ culation of small lymphocytes maximizes the opportunity for antigen-presenting cells, ef­ fector-cell precursors, and regulatory cells of appropriate specificity to cooperate in response to antigen encounter, and the subsequent migration of effector cells to target sites ensures an appropriate dissemination of the response. Considerable evidence has accumulated in recent years proving that, while this process may be random within specified compartments, there are pools of lymphocytes, and perhaps even of antigen- presenting cells, defined by nonrandom patterns of migration with respect to tissue specificities and antigen- influenced events. An understanding of the restrictions on cell migration is essential to the development of effective immunization strategies. This book addresses the issues of lymphocyte recirculation leading to inductive interactions in the immune response to antigen, the sites of these interactions, and the subsequent migration and homing of effector cells generated from these responses. In view of the lack of success in establishing effective vaccines against diseases at mucosal sites, particular attention is given to the apparent contrasts between systemic and mucosal lymphoid cell pools, and explanations are sought for mechanisms mediating selectivity of migration and homing. The contributors to this book represent a wide range of expertise from many research centers ensuring coverage of a diversity of interests on cell-traffic research and providing a broad perspective on this key function of the immune system. Regrettably, a contribution by Professor W. L. Ford was prevented by his untimely death, but his additions to our understanding of cell migration remain an enduring bequest to immunology.

Alan J. Husband, Ph.D.

TH E ED ITO R Alan J . H usband, P h.D ., is Associate Professor of Immunology in the Faculty of Med­ icine of the University of Newcastle in Australia. Dr. Husband received his B.Sc.Agr. degree from the University of Sydney in 1972 and subsequently was awarded a Ph.D. degree from the same university for studies in ruminant immunity. He then spent a period of overseas postdoctoral study in the Sir William Dunn School of Pathology at the University of Oxford and returned to Australia in 1977 to the position of Research Scientist with the New South Wales Department of Agriculture. In 1980 he accepted his appointment at the University of Newcastle. Dr. Husband’s research interests in immunology have focused primarily on problems of immune function at mucosal surfaces, particularly the role of cell migration in mucosal effector responses, and he has published extensively in this area. He is a Member of the Australian Society for Immunology, the Australian Society for Microbiology, and the Amer­ ican Association of Immunologists.

C O N TRIBU TO RS Nevin J . A bernethy, B.Sc. Pre-Doctoral Fellow Department of Pathology University of Toronto Toronto, Ontario, Canada Ann Ager, Ph.D . MRC Senior Fellow Department of Immunology University of Manchester Manchester, England Eric B. Bell, Ph.D. Senior Lecturer Department of Immunology University of Manchester Manchester, England Ian G. Colditz, Ph.D. Research Fellow Division of Animal Health Commonwealth Scientific and Industrial Research Organization Armidale, Australia M ark T. Drayson, B.Sc., M .B ., Ch.B. Research Fellow Department of Immunology University of Manchester Manchester, England John B. Hay, Ph.D. Professor Department of Immunology University of Toronto Toronto, Ontario, Canada

Clifford A. Ottaway, M .D ., Ph.D. Associate Professor Department of Medicine and Immunology University of Toronto Toronto, Ontario, Canada

R einhard Pabst, M .D ., Ph.D. Professor Centre of Anatomy Medizinische Hochschule Hannover Hannover, West Germany

Delphine M. V. P arro tt, Ph.D. Gardiner Professor Department of Bacteriology and Immunology University of Glasgow Glasgow, Scotland

Nicholas M. Ponzio, Ph.D. Professor of Pathology Department of Pathology University of Medicine and Dentistry of New Jersey New Jersey Medical School Newark, New Jersey

Roland Scollay, Ph.D. Senior Research Fellow Walter and Eliza Hall Institute Royal Melbourne Hospital Melbourne, Australia

Alan J . H usband, Ph.D . Associate Professor Faculty of Medicine Royal Newcastle Hospital University of Newcastle Newcastle, Australia

G. Jeanette Thorbecke, M .D. Professor of Pathology Department of Pathology New York University School of Medicine New York, New York

Roy L. Kerlin, B.V.Sc. Research Fellow Veterinary Pathology and Public Health University of Queensland Brisbane, Australia

Dennis L. W atson, Ph.D. Principal Research Scientist Division of Animal Health CSIRO Armidale, Australia

TABLE OF CONTENTS VOLU M E I

Chapter 1 Lymphocyte Traffic — Historical Perspectives and Future Directions................................ 1 Delphine M. V. P arro tt Chapter 2 A Tribute to W. L. Ford’s Permanent Contribution to Lymphocyte Recirculation.......... 15 Eric B. Bell Chapter 3 Lymphocyte Migration in the R a t............................................................................................ 19 Ann Ager and M ark T. Dray son Chapter 4 Migration of Cells from the Thymus to the Secondary Lymphoid Organs........................51 Roland Scollay Chapter 5 The Role of the Spleen in Lymphocyte Migration.................................................................63 R einhard Pabst Chapter 6 Antigen-Specific and Nonspecific Patterns of B-Lymphocyte Localization.......... ........ 85 Nicholas M. Ponzio and G. Jeanette Thorbecke Chapter 7 Lymphocyte Migration through Skin and Skin Lesions...................................................... 113 Nevin J . Abernethy and John B. Hay Chapter 8 Migration of Neutrophils and Their Role in Elaboration of Host D efense...................... 135 Ian G. Colditz, Roy L. Kerlin, and Dennis L. W atson Chapter 9 Dynamic Aspects of Lymphoid Cell Migration.................................................................... 167 Clifford A. O ttaway Index

195

TABLE OF CONTENTS V OLU M E II Chapter 10 A Common Mucosal Immune System Revisited...................................................................... 1 Raffaele Scicchitano, Andrezej Stanisz, Peter E rnst, and John Bienenstock Chapter 11 Migration of T Effector Cells: Role of Antigen and Tissue Specificity............................. 35 Alan J . H usband and M argaret L. Dunkley Chapter 12 Migration of Lymphocytes in the Mucosal Immune System................................................53 Julia M. Phillips - Quagliata and Michael E. Lamm Chapter 13 Migration and Differentiation of IgA Precursor Cells in the Gut-Associated Lymphoid T issue.......................................................................................................................... 77 Jeenan Tseng Chapter 14 Origin and Traffic of Gut Mucosal Lymphocytes and Mast C e lls.......................................99 Delphine G uy - G rand and Pierre Vassalli Chapter 15 Lymphocyte Traffic Associated with the Gut: A Review of In Vivo Studies in Sheep............................................................................................................................................. 113 John D. Reynolds Chapter 16 Lymphoid Cell Migration and Homing in the Young Pig: Alternative Immune Mechanisms in Action................................................................................................................ 137 Richard M. Binns and R einhard Pabst Index

175

Volume l

1

Chapter 1 LY M PH O C Y TE TR A FFIC — H ISTO R IC A L PERSPECTIVES AND FU TU R E DIRECTIO NS Delphine M. V. P arrott

TA B LE OF CONTENTS I.

Introduction.........................................................................................................................2

II.

Life Span of Lymphocytes............................................................................................... 2

III.

Lymphocyte Traffic — the Immune Response and M em ory......................................3

IV.

The Route of Entry of Lymphocytes into Lymph Node.............................................. 4

V.

The Recognition of Place: the Definition of Microenvironment and Compartments Recognized by Locomoting C ells................................................................................... 5

VI.

The Relevance of Lymphocyte Locomotion to Lymphocyte Traffic..........................7

VII.

The Effect of Antigen on Lymphocyte Circulation.......................................................8 A. “ The Lymphocyte Trap” ................................................................................... 8 B. Antigen -“ Directed” M igration......................................................................... 9

VIII.

The Future......................................................................................................................... 10

References

12

2

Migration and Homing of Lymphoid Cells I. IN TRO D U C TIO N

It is now almost 30 years since the first publications by James Gowans12 revolutionized our thinking about the physiology of lymphocyte traffic and established it as one of the cornerstones of modem immunology. Until that time, it had been considered that all lym ­ phocytes were short-lived cells which were destroyed within a few hours of entering the blood, and the notion that lymphocytes could survive for many months, even years, and could circulate continuously from blood to lymph was indeed revolutionary. But Gowans would be the first to admit that his observations did not come as “ a bolt from the blue” but were dependent upon others, so it is appropriate that his discoveries should be placed in historical prospective. Prospects past and future is the theme for this short introductory chapter and it is a personal viewpoint which many readers may dismiss in whole or in part, but if I succeed in provoking some reaction then my intention will have been achieved. II. LIFE SPAN O F LYM PHOCYTES A major incentive to research in the 1950s was the prevailing argument over whether lymphocytes were short-lived and rapidly destroyed or long-lived. This argument had its origins in the early 1885 studies of Fleming,3 whose name is usually associated with the discovery of germinal centers, in spleen, lymph nodes, and tonsils, but who also made far­ sighted proposals to explain the observed difference in the numbers of lymphocytes in efferent lymph which was always much richer in cells than lymph arriving at a node in afferent lymphatics. He proposed that either there was continuous cell division within the node (and, remember, he had observed considerable cell division in germinal centers) or that lympho­ cytes could enter the lymph node by crossing the walls of blood vessels so that there was continuous circulation of cells from the blood into the lymph node and back to the blood from the lymph. At that time, he insisted, there was no means of deciding between either proposal. During the first decade of this century, Davis and Carlson4 carried out extensive studies on leukocytes in the blood and the major lymphatics of the dog. They measured the number of lymphocytes entering the blood per hour and the number of lymphocytes in the blood and calculated that lymphocytes must be replaced at least once and possibly 3 to 4 times every 24 hr. Since the level of lymphocytes remained constant, Davis and Carlson concluded that lymphocytes may be destroyed rapidly, developed further, used for repair purposes, or “ they may circulate from lymph to blood and from blood, through the capillary endothelium, into the tissue lymph, and thence back into the lymph, of the larger lymphatic trunks” .4 The idea that lymphocytes might recirculate from blood to lymph was restated by Sjovall in 19365 but dismissed by Yoffey and Drinker in 1939.6 After World War II, technical developments occurred which led to the resolution of at least the main argument over lymphocyte life span. Bollman and his colleagues at the Mayo Clinic7 developed the tech ­ niques of thoracic duct cannulation in the rat and designed a restraining cage in which rats could be maintained with an indwelling cannula for several days. This permitted Mann and Higgins8 to show that drainage from the thoracic duct resulted in the progressive depletion of lymphocytes from the lymph. Another very important technical development followed the use of radioactively labeled compounds for biological research. In 1954, Ottesen9 used radioactive phosphorus to estimate the life span of lymphocytes in human blood and calculated that the majority of lymphocytes had an average life span of 100 to 200 days — at that time a reasonable conclusion from the data, but, nevertheless, an astonishing finding. In 1957, Gowans1 repeated the experiments of Mann and Higgins,8 but extended them to show that, by reinfusing lymphocytes into the blood, the level of circulating lymphocytes could be maintained. It is interesting that he tested an idea which had been put forward by Davis and Carlson and then reiterated by Mann and Higgins that lymphocytes and lymph have a function

Volume /

3

in providing essential nutrients by infusing cell-free lymph or lymph containing killed lym ­ phocytes. Both were ineffective when compared with live lymphocytes. Gowans suggested “ that the continuous entry of living lymphocytes into the blood may be essential for main ­ taining the output of lymphocytes from the thoracic duct” .1 He proceeded to address that suggestion by infusing 32P -labeled lymphocytes into rats which were rapidly recovered from the thoracic duct.” 1 Furthermore, he infused tritium -labeled thymidine which would be incorporated into the DNA of newly divided cells and showed that “ the number of new lymphocytes found each day in the rat amounts to only a small fraction of the normal output of lymphocytes from the thoracic duct” .2 Gowans was a pupil of Lord Howard Florey, an Australian who shared a Nobel prize for the discovery of penicillin and had a life-long interest in the physiology of lymphocytes. Florey headed the Sir William Dunn School of Pathology and inspired many young collab ­ orators to work on lymphocytes.. Another former pupil of Howard Florey in Oxford was Bede Morris who, with a series of colleagues, developed the necessary techniques of lym ­ phatic cannulation in the sheep which have permitted many problems to be addressed which would be impossible to resolve in small laboratory rodents.10 11 In the John Curtin School of Medical Research, Canberra, Australia, Bede Morris, with his Ph.D. student, Joe Hall, carried out experiments1213 which beautifully complemented those of Gowans. Hall and Morris infused 3H-thymidine into a popliteal lymph node and showed by the labeling pattern of lymphocytes in the efferent lymph that under normal conditions without antigenic stim ­ ulation not more than 4% of those lymphocytes were actually produced in the node, and since so few cells entered the node from efferent lymph then at least 85% of cells in efferent lymph must have entered from the blood.12 In other experiments Hall and Morris13 showed that the destruction of lymphocytes within an isolated lymph node by irradiation was followed by prompt restoration at far too rapid a rate to be explained by production of new lymphocytes. At the beginning of the 1960s, as far as lymphocyte “ traffickers” were concerned, lymphocytes were divided into long-lived and short-lived, small lymphocytes and large lymphocytes or lymphoblasts.14 There were also mobilizable and nonmobilizable pools of lymphocytes, that is, those lymphocytes which could be mobilized by thoracic duct drainage and those which could not.15 There were, however, concurrent lines of investigation which were directed towards the source rather than the life span of lymphocytes and these culminated in the major discoveries concerning the functions of the thymus and the bursa of Fabricius and, ultimately, into the division of lymphocytes into T and B. Such discoveries were, of course, directly relevant to lymphocyte traffic studies, but the immunological hot-bed of enthusiasts which indulged in such nonphysiological pursuits as excision of whole organ, whole body irradiation, and the in vitro “ torture” of lymphocytes prepared from lymph nodes provoked an exasperated Bede Morris to exclaim that B and T were most appropriate as the first and last letters of “ bullshit” .78 Nevertheless, the division of lymphocytes in T and B prompted much careful and systematic work on lymphocyte traffic.16 18 At first it was proposed that T cells were long-lived and could recirculate, but B cells could not. We now know that lymphocytes do vary in their life span and in their ability to recirculate from blood to lymph, but these properties cut across the B-T division. Space does not permit a detailed account, and I trust that those who also spent many years on those problems will forgive my scant acknowledgment of their labors. At the present time, life span and lym ­ phocyte recirculation are still very much live issues which are provoking controversy and experiment.79 III. LY M PH O C Y TE TR A FFIC — TH E IM M U N E RESPO N SE A N D M EM O R Y The stimulus to study lymphocyte traffic may have been a need to resolve the question of life span, but within a short space of time it was demonstrated that the immune respon ­

4

Migration and Homing of Lymphoid Cells

siveness of the whole animal is dependent upon lymphocyte circulation. In 1963, McGregor and Go wans19 demonstrated that removal of the recirculating pool by thoracic duct drainage seriously impaired the ability of rats to initiate a primary immune response to sheep red blood cells or to skin grafts and that the deficit could be restored by reinfusion of lymphocytes. Three years later, Gowans and Uhr20 extended the studies to show that memory in the absence of an ongoing immune response is also a property of circulating lymphocytes. Sam Strober in 196821 pointed out that both circulating and noncirculating lymphocytes were important in the initiation of an immune response. Bill Ford devised the ingenious technique of isolating a rat spleen whole and adjusted the perfusate to contain a variable number of lymphocytes.22 He was thus able to demonstrate that the response to sheep red blood cells was directly related to the number of lymphocytes in the perfusate, not the number of lymphocytes in the isolated spleen. In many ways, however, the most telling demonstrations of the importance of lymphocyte traffic in immune responses were those which utilized the technique of the isolated perfused lymph node. Hall and his colleagues in 196723 were the first to demonstrate the function of the cells leaving a lymph node from the efferent lymphatic. They immunized sheep with Salmonella typhi ‘O ’ antigen and demonstrated that removal of efferent lymph and lym ­ phocytes inhibited the production of a systemic immune response. Furthermore, they took efferent lymph cells from an immunized sheep and transferred into an unimmunized recipient twin sheep. The recipient responded with the prompt production of antibody in the absence of added antigen. The authors also made careful and detailed electron microscopy studies on the ultrastructure of the blast cells in efferent lymph and were somewhat “ put off” by that failure to find classical plasma cells in the lymph. With the benefit of hindsight we would nowadays reinterpret all of the early experiments on initiation of immune respon ­ siveness and memory in the light of separate T and B cells, but that in no way diminishes the impact which these early experiments made. A different, though equally significant, manifestation of the importance of lymphocyte migration in immune responses is illustrated by local mucosal immunity. Gowans and Julie Knight24 first noted that large lymphocytes for the thoracic duct migrated into the small intestinal mucosa and transformed into plasma cells. It was several years, though, before these observations were followed up, beginning with Griscelli, Vassalli, and McCluskey in 196925 who showed that although mesenteric lymphoblasts migrated to the gut, lymphoblasts from peripheral lymph nodes did not, and Guy-Grand et al.26 identified two types of migrating lymphoblast, IgA precursors and T blasts, in thoracic duct lymph. But the crucial experiments in respect to lymphocyte traffic and immune responsiveness must be credited to Nat Pierce and Gowans27 in 1975. They compared the number of specific antibody-containing cells in the lamina propria of the intestine of primed rats with or without a thoracic duct fistula after intraduodenal challenge with cholera toxin. Drainage of lymph from the thoracic duct drast­ ically reduced the numbers of specific anticholera toxin antibody-containing cells in the lamina propria, but infusion of lymph that was rich in antitoxin-containing cells resulted in the prompt appearance of such cells in the lamina propria, thus demonstrating unequivocally the function of lymphocyte traffic in local mucosal immunity. IV. TH E R O U TE O F EN TR Y O F LYM PHOCYTES INTO LYM PH N O D E The first experiments of Gowans12 and Hall and Morris11 13 showed that lymphocytes move from blood to lymph, but the route which each lymphocyte took could not be followed in detail without the technique of labeling small thoracic duct lymphocytes with tritiated adenosine and using autoradiography to clarify their whereabouts in the lymphoid tissues. Earlier workers had noticed that lymphocytes were often to be seen in the walls of the postcapillary venules often characterized by plump endothelial walls, but it was Gowans

Volume I

5

and Knight in 196424 who identified these vessels as the route by which lymphocytes enter lymph nodes. They rapidly infused labeled lymphocytes intravenously over 12 min and immediately placed lymph nodes in fixative. At that time, large numbers of lymphocytes could be seen in walls of the postcapillary vessels. Others followed to show that lymphocytes from other sources, namely thymus and spleen, also use this same route to enter the paracortex of the lymph node, an area termed thymus dependent.28 The distinct relationship of HEV (as the characteristic postcapillary, high endothelial venules of lymph nodes and gut-associated tissues are usually termed) with lymphocytes was emphasized by Gowans and Knight24 since not only was it the preferred route of migration of lymphocytes, but under normal conditions it excluded other leukocytes. Vince Marchesi and Gowans29 attributed this difference to the capacity of lymphocytes to cross the HEV by emperipolesis, i.e., they moved intracellularly rather than intercellularly. We now know that the majority of lymphocytes do not take the intracellular route (Schoefl30 and Anderson and Anderson31) and there is less emphasis on the capacity for HEV to exclude other leukocytes, since during inflammation they probably do not. Importance of the unidirection of lymphocyte traffic (i.e., from blood to node but not vice versa) is now also less emphasized because bidirectional traffic is normal in the pig. Gowans and Knight postulated that “ the rapid ‘homing’ of labeled cells into the lymph nodes presumably has its basis in the special affinity of small lymphocytes for the endothelium of the post-capillary’’.24 The search for the molecular basis of that receptor remains an obsession to this day. Gesner and Ginsburg in 196632 first started the search in terms of the carbohydrates of the lymphocyte membrane, but most experiments required that lymphocytes should be preincubated in various enzymes or lectins prior to intravenous injection. Such treatments which are known to alter the migration of lymphocytes through most tissues, including lungs, liver, and spleen, could not be narrowed down to focus only on the interaction of lymphocytes with HEV,3334 although experiments along these lines nevertheless persist. At present, the techniques devised by Woodruff and her colleagues34,35 to examine the adherence of lymphocytes to HEV in vitro are proving to be very profitable in the HEV -receptor search. Chin et al.35 have identified a factor in thoracic duct lymph which can block the binding of lymphocytes to rat HEV in vitro. Gallatin and his colleagues,36 using the Woodruff technique, have developed an antibody which identifies a surface component on lymphocytes, and both groups are working assiduously to demonstrate how these factors and antibodies identified by in vitro methods are, nevertheless, applicable to lymphocyte-traffic studies in vivo. V. TH E R EC O G N ITIO N O F PLACE: THE DEFIN ITIO N O F M IC R O EN V IR O N M EN T AN D C OM PARTM ENTS R EC O G N IZED BY LO C O M O TIN G CELLS The use of autoradiography as a means of following the progress of labeled lymphocytes into lymphoid organs offered a way in which another then-current pressing problem could be approached. Many groups were investigating the differing effectiveness of whole thymus grafts vs. separated preparations of thymocytes, lymph node lymphocytes, or splenic lym ­ phocytes in restoring the impaired immunological responsiveness of neonatally thymectomized mice. The particular deficiency of thymocytes was especially puzzling in this respect. In 1964, the question was posed to Maria de Sousa, a young Portuguese pathologist recently arrived at the Imperial Cancer Research Fund, Mill Hill, London, did thymocytes and spleen cells “ differ in efficiency because they reached different destinations within the lymphoid organs of neonatally thymectomized mice’’.28 The question prompted the careful examination of many sections of lymph nodes and spleens of normal and thymectomized mice and the infusion of radiolabeled suspensions of thymocytes or spleen cells into mice using the same isotope, 3H-adenosine as had been used by Gowans and Knight24 in their experiments in

6

Migration and Homing of Lymphoid Cells

rats. There followed a description of the areas of depletion of the lymphocytes from thymectomized mice in both lymph nodes and spleen and these were termed thymus-dependent areas.28 It was shown that 3H -adenosine-labeled thymus cells localized preferentially in the thymus-dependent areas, but that spleen cells contained an additional population of lym ­ phocytes which migrated to the outer cortex of lymph nodes and the periphery of the splenic follicle. It is ironic that Gowans and Knight24 failed to detect a comparable separate population of lymphocytes in the inoculum of thoracic duct lymphocytes which they injected because of the metabolic differences between T and B lymphocytes in thoracic duct lymph. The B cells took up far less 3H-adenosine than the T cells,18 differences which do not occur in mouse lymphocytes.37 In a subsequent publication from Gowan’s laboratory by Jonathan Howard utilizing rat thoracic duct lymphocytes the length of exposure of autoradiographs was prolonged and lightly labeled B lymphocytes were detected in lymph node nodules.38 Careful examination of autoradiographs and sections of neonatally thymectomized mice by de Sousa had permitted the postulate that the segregation of two separate migrating popu­ lations of cells, thymus-dependent and thymus-independent, occurred and, subsequently, that the ability to segregate is a characteristic of long-lived rather than short-lived cells.16 28 It required, however, the preparation of reasonably pure separated populations of T and B lymphocytes which could be labeled in vitro and injected before the concept that T and B cells have separate migration pathways within lymphoid organs became accepted.38 39 Later discernment of segregation was possible without preparing thymus-less animals, e.g., the binding of sheep erythrocytes to human T cells and the definition of B cells by the presence of surface immunoglobulin and of Fc-receptors and C3 receptors.40 At the present time, monoclonal antibodies identify not only T and B cells, but various subsets of lymphocytes in many species, man, mouse, rat, and sheep. The nature of the phenomenon of segregation of lymphocytes was described clearly in 197116 thus: “ . . . having entered the peripheral lymphoid tissues it would seem reasonable to expect all lymphocytes by virtue of their intrinsic mobility to spread themselves evenly over all territories yet in neonatally thymectonised animals, or in adult thymus-deprived mice, they do not; vast areas normally occupied by the thymus derived cells are left void, with the nodules and medulla in the lymph nodes, and the other layer of the Malpighian body and the red pulp in the spleen fully populated.”

It was realized that understanding segregation would be a major step in identifying the sites of cell-cooperation in vivo, but progress has been slow and has come mainly through perception of interaction of lymphocytes with nonlymphoid cells rather than interactions between T and B lymphocytes. Veldman in 197041 first identified the interdigitating cell in the T areas of the lymph node, and it now appears that the different compartments of the spleen and the lymph node each have their own individual type of reticulum cell and their own mononuclear/macrophage type of cell, the reticulum cell being part of the fixed stroma of the tissue while the mononuclear/macrophage cell is the mobile element, and it seems likely that by interacting with these types of cells that lymphocytes position themselves.42 There is plenty of mor­ phological evidence of such interactions and, of course, a wealth of in vitro experiments demonstrating the importance of lymphocyte/macrophage/dendritic cell interactions in the initiation of immune responses, but hard evidence that such interactions are relevant to lymphocyte traffic studies is more difficult to find. Fossum et al.43 attributed the slow progress of lymphocytes in the nude rat to the increased number of interdigitating cells in these animals. Hendriks44 and his colleagues observed that severing of afferent lymphatics to a lymph node resulted in the disappearance of the nodular aggregates of B lymphocytes and terminal centers and the reduction in height of HEV. Drayton and Ford45 extended these observations to show that despite careful maintenance of blood supply the delivery of

Volume /

7

lymphocytes and blood flow was dramatically reduced in deafferentized lymph nodes. Such studies cannot distinguish between the relative importance of the different components of lymph, whether cellular or not, to these changes, but the evidence points to the importance of both antigen in maintaining the HEV and the nonlymphoid cells in the maintenance of structure. VI. TH E R ELEV A N C E O F LY M PH O C Y TE LO C O M O TIO N TO LY M PH O C Y TE TRAFFIC The studies on the segregation and interaction of lymphocytes with other cells assumed that lymphocytes can and do locomote in a purposeful way1216 24 (see above), though this was not proven at the time. In an early review on lymphocyte traffic Ford and Gowans made a simple statement on the subject of lymphocyte locomotion, “ the motive force for pushing lymphocytes through the labyrinthine reticulum of the white pulp of the spleen and the lymph nodes is not known” .46 Later, after referring to Bill Ford’s experiments on the perfusion of isolated spleens they concluded that the “ intrinsic motility of lymphocytes would seem to be far more important than any vis a tergo in propelling the cells through the organ” ,46 an assumption which was repeated by Parrott and de Sousa16 but not given any serious consideration. Presumably, at that time we had in mind the experiment carried out many years before by Lewis47 and McCutcheon48 who had given good descriptions of the locomotor morphology of lymphocytes though we did not refer to them! Later, Schoefl30 and Anderson and Anderson31 remarked that lymphocytes assume locomotor morphology when squeezing between the endothelial walls of HEV, though the morphology is constrained under these circumstances and the “ hand mirror” -type morphology may not be obvious. Anderson and Anderson31 describe the lymphocyte as making contact with the endothelial cells by means of its pseudopod and microvilli, then adhering by numerous points of contact, and flattening by moving across the endothelial wall. The indifference of most lymphocyte traffickers to the way in which lymphocytes “ get around” inside lymphoid organs and other tissues was and is quite extraordinary. Enthusiasm to study lymphocyte locomotion has come mostly, however, from cell biologists who are primarily interested in the machinery which a cell employs to move itself and in the stimuli which provoke cell movement in vitro, chemokinesis, and chemotaxis.49 When investigations into this topic began in Glasgow, we reasoned that since activated lymphoblasts,49 isolated from lymph or lymph nodes, had been observed to migrate in a nonspecific way in vivo into sites of inflammation and were apparently responding to the same stimulus as monocytes then lymphoblasts would be the most likely population of lymphocytes to respond to che ­ moattractants — and so they did.50 Cultured lymphoblasts and mitogen-stimulated lymph ­ oblasts also responded to chemoattractants such as casein, endotoxin-activated serum, and denatured protein.49*50 The study of normal lymphocyte as distinct from lymphoblast locomotion in vitro using the filter methods commonly used for other leukocytes was unsatisfactory, however, because lymphocytes, unlike neutrophils and monocytes, attach poorly to glass or plastic.49*51 Re ­ cently, however, three-dimensional matrices, including collagen gels which mimic the re ­ ticular framework of a lymph node, have been developed.51 Lymphocytes can move well on collagen gels because they do not require to make adhesions, but can insert blebs into gaps in the gel matrix and use the gel rather like a climbing frame.51 The techniques open up an entirely new prospective for lymphocyte traffic because there is now convincing visual evidence of chemotaxis in normal lymphocytes.52 Using lipopolysaccharide (LPS)-activated serum or the supernatant fluids from the culture of monocytes as a chemotactic source, lymphocytes have been demonstrated by their morphology to orient themselves and locomote directionally.52

8

Migration and Homing of Lymphoid Cells VII. TH E EFFEC T O F A N TIG EN ON LY M PH O C Y TE C IR CU LA TIO N

In complete contrast to the lack of enthusiasm for lymphocyte locomotion studies is the avid pursuit for a specific effect of antigen on lymphocyte traffic. The idea of specificity is embedded in immunology despite ample evidence to the contrary. Many traffic studies have been (and still are) designed with the expectation that some degree of direction and specificity in relation to antigen would be revealed. Most of the expectations were doomed to disappointment!

A. “ The Lymphocyte Trap” Hall and Morris had observed that following administration of antigen there was a rapid increase in the number of lymphocytes leaving via the efferent lymph, but they also observed that the very first response (in the first few hours) was a transient drop in the output of lymphocytes in the efferent lymph and concluded that “ this immediate fall in cell output was a specific response of the lymph node to the presence of antigenic material and could best be explained in terms of a temporary reduction in the rate at which lymphocytes recirculate from blood to lymph” .53 Carefully chosen words! They speculated but did not explore the role of the vascular system in this response. Dresser et al.,54 using a different system in the mouse, studied the effect of injecting sheep red blood cells and adjuvant materials on the migration of 5,Cr-labeled lymph node cells at 24 hr to local lymph nodes draining the site of antigen injection. They observed that there were much larger amounts of radioactivity present in the draining than contralateral nodes, but did not find any change in blood volume to the draining node. There was no evidence of antigenic-specific “ homing” to any draining lymph node. In a thoughtful discussion they proposed that the changes could be due to an “ increase in the efficiency of the mechanical trapping of circulating lympho­ cytes” or to “ secretion of a chemotactic agent which activates stimulated cells to migrate to a draining node” .54 The subject of chemotaxis of lymphocytes did not arouse immediate interest, but the first of those phrases fired the imagination of many groups of workers and there followed a spate of papers about “ lymphocyte trapping” ,55 all based on the assumption that lymphocytes were inhibited from leaving the spleen or lymph nodes rather than inhibited from entering as Hall and Morris53 proposed. Later, Hall56 modified his earlier position, but it was left to Cahill and his colleagues57 to readdress the subject of “ lymphocyte trapping” . They carefully measured the actions of several different antigens in the cell input and output from the popliteal and prefemoral lymph nodes of sheep and came to the following conclu ­ sions: “ There are at least two distinct mechanisms controlling the migration of circulating lymphocytes through an antigen stimulated lymph node. The first controls the increased input of recirculating lymphocytes which occurs only through high endothelial PCV. The second controls the immediate decrease in cell output which . . . does not occur at the PCV . . . they can occur independently and are possibly controlled by different mechanisms.” Cahill et al.57 make the point that the term “ lymphocyte trapping” is inappropriate in the context of these phenomena and should be reserved for the trapping of a small minority of specific antigen-reactive cells. At this time (1976), Peter Herman,58 who had been interested in the microvasculative of lymph nodes, was measuring the changes in blood flow which occurred in the antigenically stimulated rabbit popliteal lymph node. Subsequently, Hay and Hobbs,59 who noted that the increased output in the efferent lymph of sheep occurred at the same time as the regional blood flow was increased, proposed that vascular changes and increased lymphocyte traffic may be related phenomena. Ottaway and Parrott60 set the seal on this topic by devising a technique for measuring blood flow and lymphocyte traffic simultaneously and showed that the migration of small lymphocytes to lymph nodes is directly related to blood flow. Thus, the notion that antigen-induced chemotaxis or increased trapping could account for the large

Volume I

9

increase in lymphocytes in a draining lymph was set aside in favor of increase in blood flow. The brief period of shut down that Cahill et al.57 identified as a separate phenomenon was a concept followed up by other researchers, and the agent causing it was identified as prostaglandin E2, though the mode of action was yet to be elucidated.61

B. Antigen-‘‘Directed’’ Migration A different “ yearning” to reveal antigen-directed migration is also to be found in attempts to unravel the mechanism of cell-mediated immune responses, especially in response to allografts or tumor, and also to understand the assembly of effector cells in the mucosa of the intestine. There were claims and counter claims in respect to antigen-specific accumu ­ lations of lymphocytes and lymphoblasts into the site of rejecting skin allografts, but, on the whole, the results were unsatisfactory. More consistent findings came under category of migration into sites of inflammation, though they provided scant support for antigenspecific migration. Asherson and Allwood62 were concerned with the study of the passive transfer of contact sensitivity and were puzzled by the effectiveness of peritoneal exudates presumed to contain mostly monocytes/macrophages in this respect, because the most obvious response to a contact-sensitizing agent was in the draining lymph nodes. They showed that 51Cr-labeled lymphocytes from draining lymph nodes could move not only into sites of the application of a contact sensitizer, but also into skin sites nonspecifically inflamed with croton oil. A series of experiments summarized in a review by Parrott and Wilkinson49 culminated in the identification of T lymphoblasts in S-phase, but not of small lymphocytes which could move nonspecifically into sites of inflammation whether caused by the contactsensitizing agent which initiated the production of the T lymphoblasts, a noncross-reacting contact sensitizer, or a simple inflammatory agent. A parallel line of thought and experi­ mentation was followed by MacGregor and Logie.63 They were studying the collaborative role of macrophages and lymphocytes, especially those cells present in peritoneal exudates in rats in transferring resistance to Listeria monocytogenes. They also identified T lymph ­ oblasts in S-phase and not small lymphocytes as being effective in transferring resistance and showed that these cells could be encouraged to extravasate into the peritoneal cavity by a nonspecific inflammatory stimuli, e.g., glycogen or thioglycollate, as well as by killed bacteria. Subsequent experiments along the lines set by Asherson and All wood62 and MacGregor and Logie63 did detect significant elements of increased localization of lympho ­ blasts which were accredited to the presence of specific antigen, but major elements of nonspecific accumulation were always present.60 Nevertheless, these experiments do indicate circumstances in which one type of lymphocyte is prompted to leave the bloodstream and was the starting point for one series of experiments on lymphocyte locomotion50 (see above). Another line of thought is to be found in the observations of Griscelli and his colleagues25 who identified two different migration patterns of large lymphoblasts. They found that “ the distribution of labeled cells was found to depend upon the source of the donor cells. Cells from mesenteric lymph nodes or thoracic duct lymph shared a marked preferential accu ­ mulation in lymphoid tissue within or adjacent to the intestine, whereas cells from peripheral nodes accumulated preferentially in peripheral lymph nodes” .25 Small lymphocytes did not display a similar preferential localization nor could Griscelli et al. determine whether the selective accumulation of large dividing cells “ was due to an antigen recognition mechanism or was the result of two different populations of cells with different ‘homing’ mechanisms’’ ,25 But this was the seminal paper which introduced the concept of nonrandom migration which could be influenced by mechanisms other than antigen. Griscelli et al.25 also implanted the thought in the minds of others that antigen in the gut lumen may not be the only determining factor in inducing mesenteric or thoracic duct lymphoblasts to migrate into the intestinal mucosa. This was reinforced by the demonstration that segments of “ antigen -free” gut grafts

10

Migration and Homing of Lymphoid Cells

were also accessable to mesenteric lymphoblasts.56 64 The theme of peripheral vs. mesenteric migration was seized upon and has proved to be very profitable in terms of data, publications, interest, and debate. It did not foster the cause of antigen, but it opened up the prospect that the origin of lymphoid cells could influence their migration.26 64 66 VIII. TH E FUTURE Twenty-five years ago, the findings which resulted from lymphocyte traffic studies in­ cluding the role of lymphocytes in immune responses, the importance of lymphocyte mi­ gration in the propagation of the immune response and in memory, the route of entry of lymphocytes into lymphoid tissue, and their subsequent segregation were all central to immunological thinking. Today, interest in lymphocyte traffic has declined in comparison with, for example, lymphokines and interleukins, the major histocompatibility complex, or the “ T cell receptor” . Lymphocyte traffic studies do not have the glamour attached to more “ molecular” aspects of immunology. Furthermore, there is the all too intrusive question of funding of basic research. Exper­ imentation which requires large numbers of small animals or, even worse, large animals is expensive, and grant-giving bodies require that any research should be or at least appear to be directly applicable to human or animal disease. In terms of lymphocyte-traffic studies, this is not always easy to justify, though the very considerable input into mucosal immunity which this volume contains indicates that many enthusiasts have managed to keep their particular research interest afloat. And it is a safe prediction that this field of interest will continue to flourish because it can be justified in terms of usefulness and because under­ standing the mechanisms which control the assembly of lymphocytes and lymphoblasts in the gut mucosa remain an exciting challenge. Up to the present, the main thrust has been on traffic between rather than within lymphoid tissues. In “ steady state” -type experiments, lymphocyte concentrations have been monitored in the blood stream and lymph both afferent and efferent. Single-time interval-type exper­ iments have simply measured “ roughly” how many cells are present in various tissues after an injection of a bolus of labeled cells — not a very physiological procedure as is so often pointed out. We have learned a great deal about the factors which control the delivery of cells to both lymphoid and nonlymphoid tissues (see Chapters 3, 7, 11, and 12). Future directions of research will surely “ close-in” on individual lymphoid tissues, be they lymph nodes, spleens, Peyer’s patches, or loops of intestine, and examine “ lymphocyte micro­ traffic” which characterizes the social interactions of lymphocytes with other cells, such as dendritic cells. For some time, Bede Morris67 and his colleagues and Brigid Balfour42 and her colleagues have stressed the importance of dendritic-like cells in afferent lymph and the probable effect which those have on the functional architecture of lymph nodes. The effects of deafferenting peripheral lymph nodes on the flow of lymphocytes into a lymph node have already been mentioned. The content of lymphocytes and dendritic cells of afferent lymph varies, however, according to the siting of the lymph node whether peripheral or further down the chain of lymph nodes,67 so that the rate of flux of lymphocytes from the blood into a lymph node may be influenced by the cell contact of afferent lymph. There is indirect evidence that lymphocyte entry from the blood into the mesenteric lymph node is influenced by the large number of lymphocytes entering the node from the wall of the gut.68 But the mesenteric lymph node, like other lymph nodes, is not just a bag of mixed cells,42,69 there is intravascular space, intralymphatic space, as well as the interstitium or extravascular space, and, in fact, we do not know how many lymphocytes entering via the afferent lymphatics will penetrate the interstitium and meet up with lymphocytes from the blood stream. Some could remain segregated within the lymphatic space and pass straight to efferent lymphatics. The con ­

Volume l

11

trolling influence of the mesenteric lymph node itself on the traffic of effector cells to the gut wall is an important new aspect of mucosal immunity. The noncellular content of afferent lymph will also vary according to the siting of each lymph node, and alteration to the chemical composition of lymph could vary its chemotactic properties and impose subtle changes in the speed of locomotion of lymphocytes within nodes in addition to the gross changes brought about by vasoactive substances. There are hints that a mechanism for controlling lymphocyte numbers exists in the recent observations on the effect of natural killer (NK) cells in lymph nodes on allogeneic lym ­ phocytes. These observations show that although allogeneic lymphocytes can transverse the HEV, normally they are promptly killed by NK cells once within the lymph node and their residue scavenged by interdigitating cells.71 Could this be evidence of a homeostatic mech ­ anism which regulates the number of normal but unwanted or damaged lymphocytes? The presence of an unexpected and possible additional controlling mechanism for lym ­ phocyte traffic has been revealed quite recently.72 There are nerve fibers in lymph nodes in close proximity to both the arteriovenous connections and the HEV. Recently it has been shown that T lymphocytes have receptors for vasoactive intestinal peptide (VIP).72 If it is proved that the nerve endings in lymph nodes release VIP locally, then an additional mech ­ anism for regulating T cells but not, presumably, B cells will have been revealed. This is an area ripe for expansion, not only in respect to lymph nodes but also in respect to the gut, for the mucosa of the intestine has many VIPergic nerves. The possibility of being able to see what is going on in vivo inside a lymphoid organ is very exciting. In 1983 Yamaguchi and Schoefl73 showed that it is possible to see the HEV of a mouse’s Peyer’s patch through the thin serosal muscle coat of an exteriorized loop of intestine, and such preparations can be maintained in vivo for several hours. Furthermore, they showed that when fluorescein isothiocyanate-labeled lymphocytes are injected intra­ venously they appeared in the HEV and began to adhere within a few seconds. Ottaway and his colleagues74 have recently exploited this technique further and found that, although lymphocytes adhered readily to the endothelial cells, most adhesions are only temporary. Multiple collisions increased the probability of attachment. They developed a mathematical model of attachment taking into account the geometry of the dome-shaped endothelial cell and inferred from the fit of the model that about 21% of lymphocytes which collide with the cleft regions between cells are secure while hits on dome regions detached. Furthermore, it was observed77 that lymphocytes which have become attached by lodging in the clefts clearly extrude blebs of cytoplasm and climb across the endothelium in a way which was very reminiscent of the time-lapse films of lymphocytes locomoting in vitro. Presumably, it would be a relatively simple matter to exploit the techniques of visualized Peyer’s patch HEV further by, for example, comparing the relative adherence of two different populations of lymphocytes labeled with either fluorescein isothiocyanate or rhodamine isothiocyanate. It would be reassuring if this technique could confirm the selective adherence of intestinally derived lymphocytes to Peyer’s patch tissue in preference to peripherally derived lymphocytes which has been demonstrated in the in vitro slice technique of Woodruff et al.75 At present, we know very little about the disruption of lymphocyte migration which can occur in diseased states, though there is indirect evidence that changes do occur.76 A recent review by Fossum and Ford69 has attempted to bridge the gap between the viewpoint which a histopathologist has of a lymph node in comparison with a cellular immunologist’s. The histopathologist is concerned with understanding a disordered lymph node, particularly as applied to neoplasia and infection, while the objective of a cellular immunologist is to understand how a lymph node works physiologically in defense and recognition. Lympho ­ cyte-traffic studies have concentrated to a very large extent on the normal physiology of a lymph node and could gain from studies of the disordered lymph node. I started thinking about this chapter with a feeling of nostalgia for a “ golden age of

12

Migration and Homing of Lymphoid Cells

immunology” that has passed, but I find myself finishing on a note of optimism because there are still so many important problems which have yet to be tackled.

REFEREN CES 1. Gowans, J. L., The effect of the continuous re-infusion of lymph and lymphocytes on the output of lymphocytes from the thoracic duct of unanaesthetized rats, Br. J. Exp. Pathol., 38, 67, 1957. 2. Gowans, J. L., The recirculation of lymphocytes from blood to lymph in the rat, J. Physiol. (London), 146, 54, 1959. 3. Fleming, W., Studien uber Regeneration der Gewebe. 1. Die Zellvermehrung in den Lymphdrusen und Verwandten Organen, und ihr Einfluss aufDeren Bau, Vol. 66, Max Cohen und Sohn, Bonn, 1885, 4. 4. Davis, B. F. and Carlson, A. J., Contributions to the physiology of lymph. IX. Notes on the leucocytes in the neck lymph, thoracic lymph and blood of normal dogs, Am. J. Physiol., 25, 173, 1909. 5. Sjovall, H., Experimentelle Untersuchungen uber das Blut und die blutbildenden Organe — besonders das lymphatische Gewebe — des Kaninchens bei wiederholten Aderlassen, Haken Ohlssons Buchdruckerei, Lund, 1936, 308. 6. Yoffey, J. M. and Drinker, C. K., The cell content of peripheral lymph and its bearing on the problem of the circulation of the lymphocyte, Anat. Rec., 73, 417, 1939. 7. Bollman, I. L., Cain, J. C., and Grindlay, J. H., Techniques for the collection of lymph from the liver, small intestine or thoracic duct of the rat, J. Lab. Clin. Med., 33, 1349, 1948. 8. Mann, J. D. and Higgins, G. M., Lymphocytes in thoracic duct, intestinal and hepatic lymph, Blood, 5, 177, 1950. 9. Ottesen, J., On the age of human white cells in peripheral blood, Acta Physiol. Scand., 32, 75, 1954. 10. Lascelles, A. K. and Morris, B., Surgical techniques for the collection of lymph from unanaesthetized sheep, Q. J. Exp. Physiol., 46, 199, 1961. 11. Hall, J. G. and Morris, B., The output of cells in lymph from the popliteal node of sheep, Q. J. Exp. Physiol., 47, 360, 1962. 12. Hall, J. G. and Morris, B., The origin of the cells in the efferent lymph from a single lymph node, J. Exp. Med., 121, 901, 1965. 13. Hall, J. G. and Morris, B., Effect of X-irradiation of the popliteal lymph-node on its output of lymphocytes and immunological responsiveness, Lancet, i, 1077, 1964. 14. Everett, N. B. and Tyler, R. W., Lymphopoiesis in the thymus and other tissues, Int. Rev. Cytol., 22, 205, 1967. 15. Caffrey, R. N., Rieke, W. O., and Everett, N. B., Radioautographic studies of small lymphocytes in the thoracic duct of the rat, Acta Haematol., 28, 145, 1962. 16. Parrott, D. M. V. and de Sousa, M., Thymus-dependent and thymus-independent populations: origin, migratory patterns and lifespan, Clin. Exp. Immunol., 8, 663, 1971. 17. Sprent, J., Migration and lifespan of lymphocytes, in B and T Cells in Immune Recognition, Loor, F. and Roelants, G. E., Eds., John Wiley & Sons, New York, 1977, 59. 18. Howard, J., The life span and recirculation of marrow derived small lymphocytes from rat thoracic duct, J. Exp. Med., 135, 185, 1972. 19. McGregor, D. D. and Gowans, J. L., The antibody response of rats depleted of lymphocytes by chronic drainage from the thoracic duct, J. Exp. Med., 117, 303, 1963. 20. Gowans, J. L. and Uhr, J. W., The carriage of immunological memory by small lymphocytes in the rat, J. Exp. Med., 124, 1017, 1966. 21. Strober, S., Initiation of antibody responses by different classes of lymphocytes. V. Fundamental changes in the physiological characteristics of virgin thymus-independent (B) lymphocytes and ’B ’ memory cells, J. Exp. Med., 136, 85, 1972. 22. Ford, W. L., The kinetics of lymphocyte recirculation within the rat spleen, Cell Tissue Kinet., 2, 171, 1969. 23. Hall, J. G., Morris, B., Moreno, C. D., and Bessis, M. C., The ultrastructure and function of the cells in lymph following antigenic stimulation, J. Exp. Med., 125, 91, 1967. 24. Gowans, J. L. and Knight, E. J., The route of recirculation of lymphocytes in the rat, Proc. R. Soc. London Ser. B:, 159, 257, 1964. 25. Griscelli, C., Vassalli, P., and McCluskey, R. T., The distribution of large dividing lymph node cells in syngeneic recipient rats after intravenous injection, J. Exp. Med., 130, 1427, 1969. 26. Guy-Grand, D., Griscelli, C., and Vassalli, P., The gut associated lymphoid system: nature and properties of the large dividing cells, Eur. J. Immunol., 4, 435, 1974.

Volume /

13

27. Pierce, N. F. and Gowans, J. L., Cellular kinetics of the intestinal immune response to cholera toxoid in rats, J. Exp. Med., 142, 1550, 1975. 28. Parrott, D. M. V., de Sousa, M. A. B., and East, J., Thymus-dependent areas in the lymphoid organs of neonatally thymectomized mice, J. Exp. Med., 123, 191, 1966. 29. Marchesi, V. T. and Gowans, J. L., The migration of lymphocytes through the endothelium of venules in lymph nodes: an electron microscopy study, Proc. R. Soc. London Ser. B:, 159, 283, 1964. 30. Schoefl, G. I., The migration of lymphocytes across the vascular endothelium in lymphoid tissue. A re­ examination, J. Exp. Med., 136, 568, 1972. 31. Anderson, A. O. and Anderson, N. D., Lymphocyte emigration from high endothelium venules in rat lymph nodes, Immunology, 31, 731, 1976. 32. Gesner, B. M. and Ginsburg, V., Effect of glycosidases on the fate of transfused lymphocytes, Proc. Natl. Acad. Sci. U.S.A., 52, 750, 1964. 33. de Sousa, M., Cell traffic, in Receptors and Recognition, Vol. 2, Cuatrecasas, P. and Greaves, M. F., Eds., Halsted Press, New York, 1976, 105. 34. Woodruff, J. J., Katz, I. M., Lucas, L. E., and Stamper, H. B., An in vitro model of lymphocyte homing. II. Membrane and cytoplasmic events involved in lymphocyte adherence to specialized high endothelial venules of lymph nodes, J. Immunol., 119, 1603, 1977. 35. Chin, Y. H., Carey, G. D., and Woodruff, J. J., Lymphocyte recognition of lymph node high endo­ thelium. I. Inhibition of in vitro binding by a component of thoracic duct lymph, J. Immunol., 125, 1764, 1980. 36. Gallatin, W. M., Weissman, I. L., and Butcher, E. C., A cell-surface molecule involved in organspecific homing of lymphocytes, Nature (London), 304, 30, 1983. 37. Parrott, D. M. V., Ferguson, A., and de Sousa, M. A. B., Factors affecting the traffic and segregation of lymphoid cell populations, Int. Arch. Allergy Appl. Immunol., 45, 240, 1973. 38. Howard, J. G., Hunt, S. V., and Gowans, J. L., Identification of marrow-derived and thymus-derived small lymphocytes in the lymphoid tissues and thoracic duct lymph of normal rats, J. Exp. Med., 135, 200, 1972. 39. Nieuwenhuis, P. and Ford, W. L., Comparative migration of B and T lymphocytes in the rat spleen and lymph nodes, Cell. Immunol., 23, 254, 1976. 40. Greaves, M. F., Owen, J. J. T., and Raff, M. C., T and B Lymphocytes: Origins, Properties and Roles in Immune Responses, American Elsevier, New York, 1974. 41. Veldman, J. E., Histophysiology and Electron Microscopy of the Immune Response, Ph.D. thesis, Gron­ ingen University, Groningen, The Netherlands, 1970. 42. Hoefsmft, E. C. M., Kamperdijk, E. W. A., and Balfour, B. M., Reticulum cells and macrophages in the immune response, in Mononuclear Phagocytes, Part 2, Van Furth, R., Ed., Lippincott, New York, 1975, 8. 43. Fossum, S., Smith, N. E., and Ford, W. L., The recirculation of T and B lymphocytes in the athymic nude rat, Scand. J. Immunol., 17, 551, 1983. 44. Hendriks, H. R. and Eestermans, Disappearance and reappearance of high endothelial venules and immigrating lymphocytes in lymph nodes deprived of afferent lymphatic vessels: a possible regulatory role of macrophages in lymphocyte migration, Eur. J. Immunol., 13, 663, 1983. 45. Drayton, M. T. and Ford, W. L., Afferent lymph and lymph borne cells: their influence on lymph node function, Immunobiology, 168, 362, 1984. 46. Ford, W. L. and Gowans, J. L., The traffic of lymphocytes, Semin. Hematol., 6, 67, 1969. 47. Lewis, W. H., Locomotion of lymphocytes, Bull. Johns Hopkins Hosp., 49, 29, 1931. 48. McCutcheon, M., Chemotaxis in leukocytes, Physiol. Rev., 26, 319, 1946. 49. Parrott, D. M. V. and Wilkinson, P. C., Lymphocyte locomotion and migration, Prog. Allergy, 28, 193, 1981. 50. Russell, R. J., Wilkinson, P. C., Sless, F., and Parrott, D. M. V., Chemotaxis of lymphoblasts, Nature (London), 256, 646, 1975. 51. Haston, W. S., Shields, J. M., and Wilkinson, P. C., Lymphocyte locomotion and attachment on twodimensional surfaces and in three-dimensional matrices, J. Cell Biol., 92, 747, 1982. 52. Wilkinson, P. C., A visual study of chemotaxis of human lymphocytes using a collagen-gel assay, J. Immunol. Methods, 76, 105, 1985. 53. Hall, J. G. and Morris, B., The immediate effect of antigens on the cell output of a lymph node, Br. J. Exp. Pathol., 46, 450, 1965. 54. Dresser, D. W. A., Taub, R. N., and Krantz, A. R., The effect of localized injection of adjuvant material on the draining lymph node, Immunology, 18, 663, 1970. 55. Zatz, M. M. and Lance, E. M., The distribution of 51Cr-labelled lymphocytes in antigen-stimulated mice. Lymphocyte trapping, J. Exp. Med., 134, 224, 1971. 56. Hall, J. G., Observations on the migration and localization of lymphoid cells, in Progress in Immunology II, Vol. 3, Brent, L. and Holborow, J., Eds., American Elsevier, New York, 1974, 15.

14

Migration and Homing of Lymphoid Cells

57. Cahill, R. N. P., Frost, H., and Trnka, Z., The effect of antigen on the migration of recirculating lymphocytes through single lymph nodes, J. Exp. Med., 143, 870, 1976. 58. Herman, P. G., Lyonnet, D., Fingerhut, R., and Tuttle, R. N., Regional blood flow to the lymph node during the immune response, Lymphology, 9, 101, 1976. 59. Hay, J. B. and Hobbs, B. B., The flow of blood to lymph nodes and its relation to lymphocyte traffic and the immune response, J. Exp. Med., 145, 31, 1977. 60. Ottaway, C. A. and Parrott, D. M. V., Regional blood flow and its relationship to lymphocyte and lymphoblast traffic during a primary immune reaction, J. Exp. Med., 150, 218, 1979. 61. Hopkins, J., McConnell, I., and Pearson, J. D., Lymphocyte traffic through antigen-stimulated lymph nodes. II. Role of prostaglandin E2 as a mediator of cell shutdown, Immunology, 42, 225, 1981. 62. Asherson, G. I. and Allwood, G. G., Inflammatory lymphoid cells: cells in immunized lymph nodes that move to sites of inflammation, Immunology, 22, 493, 1972. 63. McGregor, D. D. and Logie, P. S., The mediator of cellular immunity. VII. Localization of sensitized lymphocytes in inflammatory exudates, J. Exp. Med., 139, 1415, 1974. 64. Parrott, D. M. V. and Ferguson, A., Selective migration of lymphoblasts within the small intestine, Immunology, 26, 571, 1974. 65. Rose, M. L., Parrott, D. M. V., and Bruce, R. G., Migration of lymphoblasts to the small intestine. II. Divergent migration of mesenteric and peripheral immunoblasts to sites of inflammation in the mouse, Cell. Immunol., 27, 36, 1976. 66. Hall, J. G., Hopkins, J., and Orlans, W ., Studies on the lymphocytes of sheep. III. Destination of lymphborne immunoblasts in relation to their tissue of origin, Eur. J. Immunol., 7, 30, 1977. 67. Smith, J. B., McIntosh, G. H., and Morris, B., The traffic of cells through tissues: a study of peripheral lymph in sheep, J. Anat., 107, 87, 1970. 68. Smith, M. E. and Ford, W. L., The recirculating lymphocyte pool of the rat: a systematic description of the migratory behaviour of recirculating lymphocytes, Immunology, 49, 83, 1983. 69. Fossum, S. and Ford, W. L., The organization of cell populations within lymph nodes: their origin, life history and functional relationships, Histopathology, 9, 469, 1985. 70. Trnka, Z. and Morris, B., The physiology and anatomical complexity of the lymphoid system, in The Immunology o f the Sheep, Morris, B. and Niyasaka, M., Eds., Roche, Basel, 1985, 1. 71. Ford, W. L., Rolstad, B., and Fossum, S., The elimination of allogeneic lymphocytes: a useful model of natural killer cell activity in vivo?, Immunol. Today, 5, 227, 1984. 72. Ottaway, C. A., In vitro alterations of receptors for vasoactive intestinal peptide changes the in vivo localization of mouse T cells, J. Exp. Med., 160, 1054, 1984. 73. Yamaguchi, K. and Schoefl, G. L., Blood vessels of the Peyer’s patch in the mouse. In vivo observations, Anat. Rec., 206, 403, 1983. 74. Bjerknes, M., Chang, H., and Ottaway, C. A., Dynamics of lymphocyte-endothelial interactions in vivo, Science, 231, 402, 1986. 75. Chin, Y.-H., Carey, G. D., and Woodruff, J. J., Lymphocyte recognition of lymph node high endo­ thelium. IV. Cell surface structures mediating entry into lymph nodes, J. Immunol., 129, 1911, 1982. 76. de Sousa, M., Lymphocyte traffic in disease in lymphocyte circulation, in Experimental and Clinical Aspects, John Wiley & Sons, New York, 1981, 123. 77. Ottaway, C. A., personal communication. 78. Ford, W., The lymphocyte — its transformation from frustrating enigma to a model of cellular function, in Blood, Pure and Eloquent, Wintroke, M. M., Ed., McGraw-Hill, New York, 1980, 493. 79. Muller, G., Ed., Population dynamics of lymphocyte, in Immunological Reviews, No. 9, Munksgaard, Copenhagen, 1986.

Volume I

IS

Chapter 2 A T R IB U TE TO W . L. F O R D ’S PERM A N EN T CO N TR IB U TIO N TO LY M PH O C Y TE R EC IR CU LA TIO N Eric B. Bell

“ With hindsight it seems that the exploitation of radioisotopes . . . to mark and follow lymphocytes was the key . . . required to establish lymphocyte migration from blood to lymph.” W. L. Ford, 1975

Had it not been for his untimely accidental death in November 1984, this chapter would certainly have been written by Bill Ford. Lymphocyte recirculation was a research interest that occupied much of Bill’s life. Supervised by Sir James Gowans in the Sir William Dunn School of Pathology, Oxford and greatly influenced by the experimental pathologist Lord Florey, Bill Ford was thoroughly imbued with an enthusiasm for lymphocyte migration that led to a number of important contributions — observations that stand as landmarks in the literature — findings that represent milestones pointing the direction for future generations of scientists to follow. By the time Bill Ford joined the Oxford group, James Gowans had rigorously established that lymphocytes were the immune system and that these small, deceptively simple cells were constantly circulating between blood and lymph. Bill’s early papers were concerned with the tempo of lymphocyte migration in the rat. Whereas it was relatively easy to monitor the time it took for radiolabeled lymphocytes to travel through the mesenteric lymph node by sampling thoracic duct lymph, the migration of cells through the spleen required isolation and perfusion of this organ in vitro.13 These studies showed that blood-borne lymphocytes migrated into the white pulp1,2 and emerged from the spleen in the venous return 5 to 6 hr later.3 He was to employ a similar approach to study the traffic of lymphocytes through the isolated perfused mesenteric lymph node4,5 — a tedious surgical procedure requiring endless patience — patience that was tried to the limit on several occasions. Among other aims, Bill was particularly keen to determine whether lymphocytes that have migrated across high endothelial venules (HEV) can return to the blood within the lymph node as others had claimed. He found no evidence of this reverse migration.4 By the end of the 1960s, Burnett’s Clonal Selection Theory of antibody formation was gaining support at the expense of the Instructive Theory. With the recently acquired knowl­ edge of lymphocyte migration, it was possible to experimentally test the Clonal Selection Theory which demanded that lymphocytes were precommitted in specificity and therefore needed to recirculate “ in search of antigen’’. Using an experimental approach that was as elegant in design as the results were definitive, Bill and his colleagues passaged parental lymphocytes from blood to lymph in F, hybrids.6 Lymphocytes which managed to migrate into the thoracic duct were specifically depleted (negatively selected) to alloantigens of the F, hybrid, but not to third-party alloantigens.6 Bill Ford also made practical use of the fact that lymphocytes continuously recirculate. He conceived the idea that by confining a localized field of irradiation to the spleen the entire recirculating pool of lymphocytes might be destroyed without endangering lymphoid precursors.7 A profound lymphopenia was produced by either applying a 32P -impregnated polythene strip to the surface of the spleen or ingeniously by injecting the spleen with radioactive colloids which were phagocytosed in situ * 9 More recently the same principle was applied, but on this occasion lymphocytes carrying an isotope of indium (114mIn) were allowed to migrate into the spleen from the blood; the isotope was again retained by splenic phagocytes and efficiently produced total lymphoid irradiation.10

16

Migration and Homing of Lymphoid Cells

The ability of lymphocytes to selectively migrate across HEV was a phenomenon that received a great deal of Bill’s attention and thought, on both a fundamental and a practical level. The practical aspect led Bill and his colleagues to use lymphocytes as vehicles to carry radioactive isotopes1012 and toxins13 selectively into lymphatic tissues. For example, they successfully labeled human lymphocytes with m In14 which were transfused back into cancer patients.15 This technique is now used as a diagnostic tool to image tumors and as a therapeutic aid in treating lymphocyte-based cancers. More recently, lymphocytes were used to carry the toxic substance ricin into the spleen and lymph nodes of rats13— a promising study which is continuing. The nature of the lymphocyte-HEV interaction was pursued by Bill and co-workers over a number of years. These studies confirmed the effects of trypsin on inhibiting lymphocyte migration into lymph nodes516 and clarified a confused story on the effects of neuramini­ dase.517 A whole range of substances was used to treat lymphocytes in attempts to alter lymphocyte migration.1819 The migration of stimulated and unstimulated lymphocytes was examined;4,5,20 29 detailed migration pathways for T and B cells into spleen and lymph nodes were determined;30 and other investigations showed that having migrated across the HEV, lymphocytes required a period of recovery before further migration was possible.27 While these experiments were in progress it became apparent that during the time taken to collect and label thoracic duct lymphocytes in vitro the migration behavior of the cells altered,26 a finding that enabled him to correct earlier estimates of the tempo of lymphocyte recirculation.3,4,25 Finally, Bill Ford and his student/colleague Paul Andrews, who also died in the same accident, made another important observation about the HEV. They observed that in vivo, HEV selectively incorporated radioactive 35S -sulfate.18,31 Subsequently, they found that these HEV were able to synthesize a sulfated glycolipid which the HEV turned over rapidly.32 This discovery was to prove crucial to the eventual isolation of the HEV in vitro, as will be detailed in Chapter 3 of this book. It remains to be seen whether this glycolipid is directly involved in the lymphocyte-HEV interaction as early results suggested. I have omitted from this brief resume, Bill Ford’s substantial contributions to other areas, e.g ., graft vs. host reactions, transplantation, and lymphocyte life span. Despite the enormous strides made in our understanding of recirculation, Bill realized that the task was just beginning. In 1979, 20 years after Gowans had demonstrated that lymphocytes circulate from blood to lymph, Bill Ford modestly commented33 that “ . . . the migration of lym ­ phocytes now seems much more complicated than could be conceived in 1959” . That Bill Ford should have contributed so much in such a short time was a remarkable achievement. But even more remarkable was his unique ability to form lasting friendships with colleagues around the world.

Eric B. Bell Friend and colleague

REFEREN CES 1. Ford, W. L. and Gowans, J. L., The role of lymphocytes in antibody formation. II. The influence of lymphocyte migration on the initiation of antibody formation in the isolated, perfused spleen, Proc. R. Soc. London Ser. B:, 168, 224, 1967. 2. Ford, W. L., The immunological and migratory properties of the lymphocytes recirculating through the rat spleen, Br. J. Exp. P a t h o l 50, 257, 1969. 3. Ford, W. L., The kinetics of lymphocyte recirculation within the rat spleen, Cell Tissue Kinet., 2, 171, 1969.

Volume I

17

4. Sedgley, M. and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium. I. The entry of lymphocytes into the isolated mesenteric lymph-node of the rat, Cell Tissue Kinet., 9, 231, 1976. 5. Ford, W. L., Sedgley, M., Sparshott, S. M., and Smith, M. E., The migration of lymphocytes across specialized vascular endothelium. II. The contrasting consequences of treating lymphocytes with trypsin or neuraminidase, Cell Tissue Kinet., 9, 351, 1976. 6. Ford, W. L. and Atkins, R. C., Specific selection of lymphocytes by antigen in vivo: unresponsiveness of recirculating lymphocytes after exposure to histocompatibility antigen of F, hybrid rats, Nature (London), 234, 178, 1971. 7. Ford, W. L., The mechanism of lymphopenia produced by chronic irradiation of the spleen, Br. J. Exp. Pathol., 49, 502, 1968. 8. Roser, B. J. and Ford, W. L., Prolonged lymphopenia in the rat. I. The depletion of blood and thoracic duct lymphocyte populations following injection of p-emitting colloids into the spleen or lymph nodes, Aust. J. Exp. Biol. Med. Sci., 50, 165, 1972. 9. Roser, B. J. and Ford, W. L., Prolonged lymphopenia in the rat. II. The immunological consequences of lymphocyte depletion following injection of 185-tungsten trioxide into the spleen or lymph nodes, Aust. J. Exp. Biol. Med. Sci., 50, 185, 1972. 10. Birch, M., Sharma, H. L., Bell, E. B., and Ford, W. L., The carriage and delivery of substances to lymphatic tissues by recirculating lymphocytes. II. Long-term selective irradiation of the spleen and lymph nodes by deposition of indium-114m, Immunology, 58, 359, 1986. 11. Rannie, G. H., Thakur, M., and Ford, W. L., An experimental comparison of radioactive labels with potential application to lymphocyte migration studies in patients, Clin. Exp. Immunol., 29, 509, 1977. 12. Sparshott, S. M., Sharma, H., Kelly, J. D., and Ford, W. L., Factors influencing the fate of indium111 labelled lymphocytes after transfer to syngeneic rats, J. Immunol. Methods,. 41, 303, 1981. 13. Sparshott, S. M., Forrester, J. A., McIntosh, D. P., Wood, C., and Davies, A. J. S., The carriage and delivery of substances to lymphatic tissues by recirculating lymphocytes. I. The concentration of ricin in lymphocyte traffic areas, Immunology, 54, 731, 1985. 14. Wagstaff, J., Gibson, C., Thatcher, N., Ford, W. L., Sharma, H., Benson, W., and Crowther, D., A method for following human lymphocyte traffic using indium-111 oxine labelling, Clin. Exp. Immunol., 43, 435, 1981. 15. Wagstaff, J., Gibson, C., Thatcher, N., Ford, W. L., Sharma, H., and Crowther, D., Human lymphocyte traffic assessed by indium-111 oxine labelling: clinical observations, Clin. Exp. Immunol., 43, 443, 1981. 16. Rannie, G. H., Smith, M. E., and Ford, W. L., Lymphocyte migration into cell-mediated immune lesions is inhibited by exposure to trypsin, Nature (London), 267, 520, 1977. 17. Ford, W. L. and Sedgley, M. E., The migration of lymphocytes across vascular endothelium, Agents Actions, 6, 248, 1976. 18. Ford, W. L., Smith, M. E., and Andrews, P., Possible clues to the mechanism underlying the selective migration of lymphocytes from the blood, in Symp. No. 32 on Cell-Cell Recognition, Curtis, A. S. G., Ed., Cambridge University Press, London, 1978, 359. 19. Smith, M. E. and Ford, W. L., The effect of cytoskeletal inhibitors on the migration of lymphocytes across vascular endothelium, Adv. Exp. Med. Biol., 114, 85, 1979. 20. Ford, W. L. and Simmonds, S., The tempo of lymphocyte recirculation from blood to lymph in the rat, Cell Tissue Kinet., 5, 185, 1972. 21. Smith, M. E., Sparshott, S. M., and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium. III. Concanavalin A delays lymphocytes in normal traffic areas, Exp. Cell. Biol., 45, 9, 1977. 22. Ford, W. L., The recruitment of recirculating lymphocytes in the antigenically stimulated spleen: specific and non-specific consequences of initiating a secondary antibody response, Clin. Exp. Immunol., 12, 243, 1972. 23. Atkins, R. C. and Ford, W. L., Early cellular events in a systemic graft-versus-host reaction. I. The migration of responding and non-responding donor lymphocytes, J. Exp. Med., 141, 664, 1975. 24. Rannie, G. H. and Ford, W. L., Recirculation of lymphocytes: its role in implementing immune responses in the skin, Lymphology, 11, 193, 1978. 25. Smith, M. E., Martin, A. F., and Ford, W. L., The migration of lymphoblasts in the rat. The preferential localization of DNA-synthesizing lymphocytes in particular lymph nodes and other sites, in Essays on the Anatomy and Physiology of Lymphoid Tissues, Vol. 16, Monogr. in Allergy, Tmka, Z. and Cahill, R. N., Eds., S. Karger, Basel, 1979, 203. 26. Smith, M. E. and Ford, W. L., The recirculating lymphocyte pool of the rat. A systematic description of the migratory behaviour of recirculating lymphocytes, Immunology, 49, 83, 1983.

18

Migration and Homing of Lymphoid Cells

27. Smith, M. E. and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium. VI. The migratory behaviour of thoracic duct lymphocytes retransferred from the lymph nodes, spleen, blood or lymph of a primary recipient, Cell. Immunol., 78, 161, 1983. 28. Fossum, S., Smith, M. E., and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium. VII. The migration of T and B lymphocytes from the blood of the athymic nude rat, Scand. J. Immunol., 17, 539, 1983. 29. Fossum, S., Smith, M. E., and Ford, W. L., The recirculation of T and B lymphocytes in the athymic nude rat, Scand. J. Immunol., 17, 551, 1983. 30. Nieuwenhuis, P. N. and Ford, W. L., Comparative migration of B and T lymphocytes in the rat spleen and lymph nodes, Cell. Immunol., 23, 254, 1976. 31. Andrews, P., Ford, W. L., and Stoddart, R. W., Metabolic studies of high walled endothelium of postcapillary venules in rat lymph nodes, in Blood Cells and Vessel Walls: Functional Interactions, CIBA Foundation Symp. No. 71, O ’Connor, M., Ed., CIBA Medical, West Caldwell, N.J., 1980, 211. 32. Andrews, P., Milsom, D. W., and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium. V. Production of a sulphated macromolecule by high endothelial cells in lymph nodes, J. Cell Sci., 57, 211, 1982. 33. Ford, W. L., Lymphocyte migration and immune responses, Prog. Allergy, 19, 1, 1975.

Volume I

19

Chapter 3 LY M PH O C Y TE M IG RA TIO N IN THE RAT Ann Ager and M ark T. Dray son

TA B LE OF CONTENTS I.

Isolation and Culture of HighEndothelial Cells........................................................... 20 A. The High Endothelial Venule............................................................................ 20 B. Maintenance of the High Endothelial VenuleNetw ork..................................20 C. Vascular Endothelial Cell C ulture....................................................................21 D. Isolation of High Endothelial C ells.................................................................. 23 E. Primary Culture of High Endothelial C ells.....................................................23 F. Adhesion of Lymphocytes to Cultured HighEndothelial C ells.................... 26 G. Immunofluorescent Localization of von Willebrand Factor Antigen in Lymph N o d e ....................................................................................................... 26 H. Immunofluorescent Staining of Lymph Node with Antiserum to High Endothelial C e lls.................................................................................................26 I. Discussion.............................................................................................................27

II.

Lymphocyte Migration into Lymph Nodes.................................................................. 29 A. Cell Sources and Radiolabels............................................................................ 29 B. The Role of Antigenic S tim uli.........................................................................32 C. The Role of Afferent Lym ph............................................................................ 38 D. The Site of Selection of Antigen-Specific C ells............................................41

References

46

20

Migration and Homing of Lymphoid Cells I. ISO LA TIO N AND C U LTU R E O F HIGH EN D O TH ELIA L CELLS*

A. The High Endothelial Venule Recirculating lymphocytes enter lymph nodes (LN) from the blood by penetrating the walls of specialized postcapillary venules (PCV). In rodents and humans, endothelial cells lining specialized PCV are plump or cuboidal and these blood vessels are called high endothelial venules (HEV).12 HEV are characterized by numerous lymphocytes in the lumen and in the vessel wall; this feature is used to identify specialized PCV in some species, e.g., sheep, in which the endothelium is not particularly high. The peculiar morphology of HEV has made these blood vessels attractive for light and electron microscope studies and many reports conclude that HEV are readily distinguished from other blood vessels in LN, including nonspecialized PCV which are lined with normal or flat endothelium. HEV are pyroninophilic and show toluidine blue metachromasia due to abundant cytoplasmic RNA.3 5 Ultrastructural studies demonstrate dilated strands of rough, endoplasmic reticulum, dilated Golgi cistemae, and many smooth-walled vesicles6'8 indi­ cating active protein synthetic and secretory apparatus. The perivascular sheath surrounding HEV is thickened in comparison with that around normal PCV. It consists of overlapping pericytes embedded in an extracellular matrix of collagen and electron-dense ground sub­ stance.68 A detailed comparison of the components which make up the extracellular matrices surrounding HEV and normal PCV in LN has not been performed. Biochemical studies also distinguish HEV from other blood vessels of the LN. HEV have substantially higher levels of cytoplasmic nonspecific esterase than other vascular endothe­ lium.3 4 8 The function of this enzyme in HEV is unknown. The metabolism of sodium sulfate by lymphoid tissue in vivo and in vitro has identified a unique biosynthetic pathway leading to the continuous secretion of sulfated glycolipid.910 Autoradiographic studies using 35Ssulfate localize this pathway solely to HEV in LN. A role for this unique sulfated macro­ molecule in the transport of lymphocytes across HEV has been proposed but it remains unproven.11

B. Maintenance of the High Endothelial Venule Network Since the major site of lymphocyte influx into LN is across HEV it is important to understand what maintains the expression of surface determinants for lymphocytes by this network of specialized blood vessels. HEV are present in LN of neonatal and germ-free animals in which lymphocyte recirculation occurs normally.6 71214 The morphology of HEV and the migration of lymphocytes across HEV are, therefore, not dependent on the presence of antigen. A severe reduction in the level of recirculating lymphocytes in rats by chronic thoracic duct drainage does not affect HEV morphology. The levels of cytoplasmic non ­ specific esterase in HEV of these animals remain normal.8 The morphology of HEV is dramatically different in adult rodents in which T lymphocyte function is absent. In congenitally athymic (nude) and neonatally thymectomized rodents, endothelium lining specialized PCV is considerably flatter than in euthymic animals, although these vessels are still distinguishable from normal PCV.1518 However, the capture of trans­ fused syngeneic lymphocytes (T and B cells) by specialized PCV in nude rats is as efficient as in euthymic rats indicating that the flattened endothelial morphology does not seriously impair its ability to interact with lymphocytes.19 Blood flow and lymphocyte influx into the relatively quiescent popliteal LN of rabbit and rat are elevated after administration of a single, large dose of antigen to the footpad20 21 (see Section II.B). The secretion of sulfated glycolipid is elevated over a similar time course and all three parameters reach a peak 4 to 5 days after antigen administration.21 Microangiographic * Section I was contributed by Ann Ager.

Volume I

21

studies by Herman et al.20 concluded that the increased blood flow through the node was achieved by opening of arteriovenous shunts and probably did not involve the formation of new capillaries. However, Anderson et al.5 reported mitotic figures and [3H]-thymidinelabeled nuclei in HEV of LN draining skin allografts in the rat. A careful analysis of serial sections revealed that 3H-thymidine labeling was focal and was restricted to the transition zones from low to high endothelium in PCV leading into the HEV network. No labeling of HEV was found at the transition from high to low endothelium where HEV drain into segmental veins in the medullary region. Proliferation of endothelial cells lining PCV is normally associated with neovascularization. The proliferation of endothelial cells lining HEV following antigen administration does not appear to result in new blood-vessel for­ mation. Anderson et al.5 concluded that it resulted in increased arborization of HEV in the LN cortex by extension of the HEV network from its periphery into existing side branches. There are also dramatic effects on the morphology and function of HEV when a LN is deprived of incoming antigen following ligation of afferent lymphatics. The influx of lym ­ phocytes across HEV and the height of HEV are severely reduced and both effects are correlated with the loss of macrophages from the LN22'24 (see Section II.C). It is clear that the maintenance of a functional HEV network does not simply depend on an intact blood supply. The extent and ability of the HEV network to function at a given time will probably depend on a combination of the level of antigenic stimulation and the influx of nonlymphoid cells via the afferent lymphatics. A greater understanding of the biology of HEV should provide more information about the specific interaction of lymphocytes with these specialized blood vessels which results in the continual migration of lymphocytes into LN.

C. Vascular Endothelial Cell Culture The study of vascular endothelial cell (EC) biology, in general, has increased since the routine culture and identification of these cells were first demonstrated in 1974.25,26 Pure populations of vascular EC can be isolated by digesting the luminal surfaces of major blood vessels using bacterial collagenase. Vascular EC in situ contain von Willebrand factor antigen (vWF:Ag), but underlying smooth muscle cells and fibroblasts do not.27 The localization of vWF:Ag in isolated vascular EC28 provides a means of identification and a measure of contamination of these cultures by unwanted cell types. EC can be isolated from aorta, vena cava, pulmonary artery, and umbilical vein of humans and large domestic animals such as the cow and pig.29 EC proliferate in tissue culture medium supplemented with fetal calf serum for varying periods of time depending on the species and source of the cells, e.g., from 5 to 50 doublings. EC are easily identified by phase-contrast microscopy by their particular morphology and growth characteristics. Proliferation of EC starts 24 to 48 hr after isolation and is confined to the periphery of isolated clumps of 20 to 50 cells (Figure la). EC continue to divide at the edges of these clumps until a confluent monolayer of closely apposed, polygonal cells is formed. Cell division ceases unless EC are detached from the culture vessel and plated at subconfluent density. Proliferation of subcultured EC then continues until a confluent monolayer is formed once more (Figure lb). Cultured vascular EC have provided information about the maintenance of a nonthrombogenic surface and the control of vascular tone by these cells as well as their contribution to diseases of large blood vessels such as atherosclerosis. It is inappropriate to use EC cultured from major blood vessels to study their role in processes involving capillary endothelium such as angiogenesis, inflammation, and vascular permeability. Techniques have, therefore, been developed to culture EC from the microvasculature.30,38 Tissue is disrupted mechanically, the vessel fragments are fractionated on nylon filters of varying pore sizes, and vessels of defined diameters are digested with bacterial collagenase. This gives rise to a heterogeneous population of clumps of two to six endothelial cells and a mixture of pericytes, fibroblasts, and other organ-specific cells. Microvascular

22

Migration and Homing of Lymphoid Cells

FIGURE 1. (a) [3H]-thymidine autoradiograph of porcine aortic endothelial cells 24 hr after isolation showing one heavily labeled nucleus at edge of cell clump. (Hematoxylin-eosin; magnification x 900.) (b) Phase-contrast micrograph of confluent rat aortic endothelial cells showing polygonal, closely apposed cells. (Magnification x 400.)

EC have a similar morphology and growth pattern as large-vessel EC and are, therefore, easily identified in these mixed cultures. EC proliferation is stimulated by specific growth factors, e.g., tumor angiogenesis factor,30 endothelial cell growth factor,31 and the outgrowth from uncontaminated clumps of EC can be selected in a culture dish by mechanical removal of unwanted cells using a drawn out Pasteur pipette.30 More recently, the selective inter­ nalization by microvascular EC of acetylated low-density lipoprotein to which a fluorochrome is attached has allowed a more rapid and easier purification of these cells from mixed cultures by fluorescence-activated cell sorting.32 When EC are isolated from a mixed population of cells rigorous attempts at identification must be carried out. The most commonly employed marker for vascular EC is vWF:Ag which is only found elsewhere in platelets and megak­ aryocytes. Other markers such as angiotensin-converting enzyme,33 plasminogen activator,34 and production of prostaglandin I235 36 have been used, but these markers are not always restricted to microvascular EC in mixed starting cultures. Ideally, a combination of markers should be used. With this information from microvascular EC culture, an attempt was made to isolate

Volume I

23

high endothelial cells (HEC) from LN in order to study lymphocyte interaction with relevant endothelium in vitro.40 Before commencing, a marker for isolated HEC needed to be es ­ tablished. The unique metabolism of sulfate by these cells could provide a means of labeling HEC in lymphoid tissue prior to isolation and of identifying HEC after isolation. Stimulated secretion of sulfated glycolipid21 and HEC proliferation reach a peak 4 days after admin ­ istration of antigen. Popliteal LN undergoing a graft-vs.-host response were therefore used as a source of 35S-labeled and proliferating HEC. LN slices were labeled with 35S-sulfate and, because the half life of sulfated glycolipid in HEC is about 1 hr, the time taken to digest the tissue was kept to a minimum in order to optimize localization of 35S-label in HEC by autoradiography.

D. Isolation of High Endothelial Cells Slices of popliteal LN (1 to 2 mm) were labeled with 50 |xCi/m€ sodium 35S-sulfate for 30 min at 37°C in RPMI 1640 culture medium. The tissue was digested using 0.5% collagenase (Type II, Sigma, Poole, Dorset, U.K.) for 45 min and isolated cells allowed to adhere to plastic coverslips for 60 min. Cells were fixed for 30 min with neutral buffered formalin and processed for autoradiography. Autoradiographs revealed three distinct populations of cells in collagenase digests of 35Slabeled LN: (1) 20- to 30-p.m diameter 35S-labeled, pale-staining cells, (2) 10- to 15-|xm diameter nonlymphoid cells which were unlabeled and lightly stained, and (3) lymphocytes, which were unlabeled and darkly stained. Of the two populations of nonlymphoid cells (1 and 2), 71.7 ± 4.9% (SD, n = 4) were large and labeled and were, therefore, HEC. The majority (>95% ) of the remaining unlabeled cells were medium sized. HEC in fixed and stained preparations were round cells and were found singly, occasionally with one or two lymphocytes attached.

E. Primary Culture of High Endothelial Cells Phase-contrast microscopy of cells isolated from LN by collagenase digestion showed three distinct populations of cells which could be directly identified with those in autora­ diographs. HEC were large, round, single cells occasionally with one to three lymphocytes attached to their surface. Most HEC had a distinct nucleus with at least two nucleoli, and HEC cytoplasm was flat and vesiculated (Figure 2a). The medium-sized nonlymphoid cells identified by autoradiography clearly resembled macrophages by phase-contrast microscopy, having a phase-dark condensed cytoplasm and an indistinct nucleus (Figure 2a). This pop ­ ulation of cells did not proliferate, but persisted in the cultures for up to 10 days. During the first 2 days of culture in RPMI 1640 plus 20% fetal calf serum (FCS), the lymphocytes were washed away and HEC started to move around the culture dish. In so doing, HEC adopted varying morphologies ranging from amoeboid, with at least one pseudopodium, to bipolar, with long cytoplasmic processes. After 2 days, most HEC were no longer vesiculated and the cells started to proliferate. HEC remained motile such that divided cells became singly dispersed, but, occasionally, clumps of five to ten polygonal HEC were seen by day 5 (Figure 2b). At this stage of proliferation and at all later stages, HEC displayed abundant granules dispersed around the nucleus. At confluent densities, HEC underwent a dramatic morphological change to a bipolar one and tended to align in parallel arrays (Figure 2c). To date, 54 primary HEC cultures have been established and maintained to first confluence. At confluence, HEC showed substantial cytoplasmic overlap, but nuclear overlap was not obvious. HEC reached a density of 350 cells per square millimeter having been plated at 2 cells per square millimeter after isolation. Primary HEC cultures occasionally contained clumps of two to three micro vascular EC30 and/or fibroblasts, but neither cell-type proliferated or persisted beyond 5 days after isolation. When plated at subconfluent densities, HEC flattened up to 100 |xm diameter and exhibited

24

Migration and Homing o f Lymphoid Cells

FIGURE 2. Phase-contrast micrographs of high endothelial cells in culture, (a) Primary HEC culture after 2 days showing a flat, vesiculated cell (curved arrow) and amoeboid cells (open arrows), (b) Primary HEC culture after 5 days showing singly dispersed cells, (c) Primary HEC culture after 10 days showing confluent bipolar cells aligned in a parallel array, (d) Adhesion of thoracic duct lymphocytes to HEC. TDL are either attached to the surface of HEC (single arrow) or between HEC and the cultured dish (double arrows). (Magnification x 400.) (From Ager, A., J. Cell Sci., 87, 133, 1987. With permission.)

Volume I 25

26

Migration and Homing of Lymphoid Cells

“ stress fibers” along one axis of the cell. As HEC grew to confluence, stress fibers dis­ appeared and HEC aligned in a parallel array as seen at first confluence. Cultured HEC have been frozen in RPMI 1640 containing 20% FCS and 10% dimethyl sulfoxide at 5 x 1 0 5 cells per milliliter, stored for up to 6 months, and thawed successfully (trypan blue exclusion >95%).

F. Adhesion of Lymphocytes to Cultured High Endothelial Cells In HEV of peripheral LN, lymphocytes adhere to high endothelium and then migrate between the endothelial cells to enter the LN parenchyma. In order to determine whether cultured HEC demonstrated this specific interaction with lymphocytes, the adhesion of lymphocytes to cultured HEC was studied. Thoracic duct lymphocytes (TDL) were collected for 2 to 4 hr at room temperature, washed, and then incubated for 45 min with primary cultures of HEC. Following glutaraldehyde fixation, single TDL were attached to HEC in large numbers (Figures 2d). The ratio of TDL:HEC nuclei in fixed and stained preparations was 0.6 ± 0.2:1 (SEM, n = 12). This accounted for 1.5 to 2% of TDL added. Further examination of living cells by phase-contrast microscopy revealed that lymphocytes were either spherical, reflecting incident light, or significantly flattened, appearing as phasedark cells (Figure 2d). High power microscopy of fixed and stained cells showed that spherical TDL exhibited little intracellular detail and were attached to the exposed surface of HEC, whereas flattened TDL revealed a large nucleus with a low cytoplasm:nucleus ratio and were between HEC and the culture vessel. From a series of experiments, the mean number of TDL underneath HEC was 22.8 ± 3.4% (± SEM, n = 12). Cells other than lymphocytes were rarely observed in these preparations ( < 1 %), but larger cells which could be either blast cells or macrophages have been seen on these occasions.

G. Immunofluorescent Localization of von Willebrand Factor Antigen in Lymph Node The distribution of vWF:Ag in LN endothelium was determined using a rabbit antihuman vWF:Ag antiserum which cross reacts with vWF:Ag in rat vascular endothelium . 37 The luminal surfaces of the hilar artery and vein were positive for vWF:Ag, as were arterioles, venules, and capillaries lined with flat endothelium in the medulla of the LN. The staining pattern of capillary endothelium was distinctly punctate which is the charac­ teristic distribution of vWF:Ag in vascular endothelium. The numerous HEV distributed throughout the paracortex of cervical LN did not stain with vWF:Ag antiserum. Occasionally, some staining was restricted to the lumens of these vessels suggesting the presence of vWF: Ag in blood. Lymphatic endothelium of afferent vessels in the subcapsular sinus and of efferent vessels in the medulla was also negative for vWF:Ag. Primary HEC cultures 5 to 10 days after isolation did not stain for vWF:Ag.

H. Immunofluorescent Staining of Lymph Node with Antiserum to High Endothelial Cells The absence of vWF:Ag in HEV, and thereby a means of identification of isolated high endothelial cells, led to the production of a polyclonal anti-HEC antiserum in order to ascertain if antigenic determinants were shared between cultured HEC and HEV in LN. Rabbits were immunized with 107 cultured HEC in complete Freund’s adjuvant, boosted with 5 x 106 cells in incomplete Freund’s adjuvant or phosphate-buffered saline (PBS) 21 days later, and bled after 10 days. Rabbit anti-rat HEC antiserum was absorbed against rat serum proteins and fetal calf serum proteins. This reduced the titer of the immune sera by ~ 20 - fold, but the pattern of immunofluorescent staining (see below) was not altered. Using this antiserum, cultured HEC were stained uniformly throughout the cytoplasm (Figure 3a). In cryostat sections of LN and Peyer’s patches, anti-HEC antiserum stained HEV uniformly across the vessel wall. Lymphocytes either in the vessel lumen or throughout the LN par-

Volume /

27

FIGURE 3. Immunofluorescent staining of cultured HEC and LN with polyclonal antiHEC antiserum, (a) Subcultured HEC; (b) HEV in LN par­ acortex showing uniform staining throughout vessel wall. (Magnification x 900.) (From Ager, A., J. Cell Sci., 87, 133, 1987. With permission.)

enchyma were not stained (Figure 3b). This antiserum also showed a continuous line of staining associated with capillaries and lymphatic vessels. The continuous, flat endothelium of medullary sinuses and the discontinuous, flat endothelium of the subcapsular sinuses (data not shown) were clearly outlined. The staining pattern of arterioles and venules in the LN revealed that anti-HEC antiserum stained throughout the walls.

I. Discussion Collagenase digestion of lymphoid tissue prelabeled with 35S-sulfate yielded primary cultures in which a majority of the cells (>70% ) were 35S-labeled and were, therefore, HEC. The only other nonlymphoid cells present in significant numbers (>25% ) were identified as macrophages by phase-contrast microscopy. The demonstration that >90% of isolated cells belonged to only two different types of nonlymphoid cell was an unexpected finding. Primary lymph node cultures occasionally contained morphologically distinct microvascular EC and/or fibroblasts ( < 1 % of isolated cells), but neither cell type survived for more than 5 to 6 days. Primary cultures were enriched for HEC as macrophages did not persist beyond

28

Migration and Homing of Lymphoid Cells

7 to 10 days. HEC proliferated to first confluence with an apparent doubling time of 1 to days and underwent dramatic changes in morphology as a consequence of cell motility and cell division. Cells were subcultured and continued to proliferate in long-term culture for up to 2 months. Even though no attempt was made to purify these cultures, at no stage was there more than one morphologically distinct type of cell present. Following antigenic challenge, the proliferation of high endothelium in LN is stimulated five- to tenfold . 5 In order to increase the possibility of isolating HEC that might proliferate in vitro, popliteal LN undergoing a graft-vs.-host response were used initially. When HEC cultures isolated from these “ activated” LN were compared with those isolated from normal cervical LN, similar degrees of purity were obtained. HEC isolated from cervical LN pro ­ liferated in the same way as HEC from popliteal LN. In practice, cervical LN proved to be a more convenient source of HEC cultures and were, therefore, used routinely. Several questions arise from this demonstration of the isolation and culture of HEC. Why does collagenase digestion only release HEC from lymphoid tissue? The yield of 35S-labeled cells from LN slices was only about 10%, therefore, collagenase digestion is very limited. Pericytes (which are normally a serious contaminant of capillary EC cultures) lying under HEC are embedded in a thick extracellular matrix which is presumably not disrupted by collagenase treatment. Even when other cells were isolated, they did not survive for more than 5 to 6 days. It is known that the requirements for capillary EC proliferation in vitro are quite specific, usually involving the addition of growth factors. 30 The conclusion is that the conditions which support rapid growth of HEC in culture do not support capillary EC or fibroblast proliferation. Secondly, why do HEC proliferate in vitro in the apparent absence of specific growth factors? HEC proliferation was stimulated by up to 20% (v/v) fetal calf serum in a dosedependent manner, therefore, either fetal calf serum contains growth factors which are specific for HEC or HEC secrete their own growth factors in tissue culture. Cultured HEC did not retain their peculiar cuboidal morphology in vitro. At subconfluent densities HEC were extremely motile on plastic culture dishes and at confluent densities HEC were bipolar and were aligned in parallel arrays. Since the lack of cuboidal morphology of HEV in nude rats does not impair their ability to transport lymphocytes from the blood it was interesting to find that primary HEC cultures bound lymphocytes at a ratio of lymphocytes:HEC of 0.6:1 which is similar to ratios measured in situ. Anderson et al . 5 measured a ratio of 0.75:1 and Hendriks et al . 23 measured a ratio of 1.6:1 using histological preparations of normal LN. Up to 25% of lymphocytes which adhered to HEC cultures were found to be between the cultured cells and the culture vessel. It is not clear if this observation is related to the migration of lymphocytes across HEV in lymphoid tissue because peripheral blood lymphocytes have been shown to migrate into collagen gels in the absence of specific cell adhesion . 38 39 Lymphocytes do not normally cross the aortic wall so cultured aortic EC were used as a control for the lymphocyte adhesion assay. Preliminary experiments demonstrated that the interaction of lymphocytes with HEC was ~ 50-fold higher than the interaction with aortic EC .40 A rabbit polyclonal antiserum raised against cultured HEC stained HEV in cervical and mesenteric LN and in Peyer’s patches demonstrating that cultured HEC shared antigenic determinants with high endothelium in lymphoid tissue. No attempt was made to study the range of antigenic determinants recognized by this antiserum, therefore, it is not possible to comment on the significance of the staining of vascular and lymphatic vessels which are not lined with high endothelium. HEC line PCV and are, therefore, a type of micro vascular endothelium. The demonstrated lack of vWF:Ag localization in HEV in comparison with its presence in other microvascular EC of rat LN suggests that the precise relationship between these two types of microvascular endothelia is not straightforward with respect to their roles in hemostasis. Further distinctions 2

Volume I

29

between HEC and other vascular EC are established from this study of cultured HEC by their differing patterns of proliferation in vitro and the 50-fold higher affinity for lymphocytes shown by HEC. The main function of HEC in vivo is to transport lymphocytes from the blood to the LN. This first demonstration of the reproducible isolation and culture of specialized postcapillary venule endothelium should allow the expression of surface determinants for lymphocytes by these cells to be studied.

II. LYMPHOCYTE MIGRATION INTO LYMPH NODES* A. Cell Sources and Radiolabels The mammalian immune system is both anatomically dispersed with respect to the major lymphoid organs and heterogeneous in its billions of cellular components. Since Gowans’ discovery of lymphocyte recirculation , 1-2 it has become increasingly apparent that cell move­ ment both between and within lymphoid organs plays a major role in enabling the many components of the immune system to function as an integral unit .4 1 44 An event in one part of the system is dependent upon cellular contributions from other parts and, in turn, the products of that event are disseminated back to the immune system as a whole. This section is concerned largely with the movement of mature lymphocytes between secondary lymphoid organs and is intended to illustrate experimental approaches as much as to discuss the significance of the observations. In the choice of experimental animal one must consider: (1) the availability of inbred strains, (2) the current knowledge of a particular species’ immune system, (3) the availability of tools such as monoclonal antibodies to allow detailed dissection of immunological events, (4) the size and propensity of the animal to surgical manipulation, and, finally, (5) the cost. Mice, rats, and sheep have provided the major subjects of lymphocyte migration studies and, on the whole, have produced concordant results. Their differing attributes in the cat­ egories described above have allowed a variety of experimental approaches not always common to all species. On the one hand, this variety has provided a greater insight into mammlian lymphocyte migration and, on the other hand, an occasional, and perhaps some­ times erroneous, conclusion that there are significant species differences. Movement of a cell is a dynamic process and requires repeated observations in time. Unfortunately, that necessitates at least one violation of the animals’ normal physiology and, furthermore, requires that the cell is in some way marked for later identification. The most popular way of marking lymphocytes has been radioactive labeling and, more lately, use of fluorescent dyes ,42 but chromosome markers, antigenic markers, and even functional markers such as the dependence of secondary antibody responses on memory cells have been exploited .43 This discussion will be confined to the use of radiolabels, but many of the principles and problems are equally applicable to other cell-marker techniques. An ideal method of labeling lymphocytes would produce homogeneous labeling within the population, the label would be permanently retained in the cells and its presence would not damage or change the properties of cells. None of this can be achieved, but by knowledge of the shortcomings of a given radiolabel one can manipulate and/or compensate to achieve sci­ entifically sound results .44,45 Variable labeling within a population of lymphocytes is most common because the pop ­ ulation is itself heterogeneous. 51Cr-sodium chromate labels large lymphocytes approximately twice as much as small lymphocytes,46 although this problem can be overcome by passaging 5 ,Cr-labeled cells from blood to lymph of an intermediate animal before final use, since lymphoblasts recirculate poorly .47 On the other hand, labeling with 3H -thymidine in vitro * Section II was contributed by Mark T. Drayson.

30

Migration and Homing of Lymphoid Cells

labels only large lymphocytes48 and in vivo can be used to label a particular age cohort of lymphocytes.49 In the rat, incubation in vitro with 3H -uridine labels T lymphocytes 10 to 15 times more intensively than B lymphocytes,50 allowing one to predominantly study Tcell migration without purification from B cells51or T/B -cell discrimination in autoradiographs.52 Radiolabels are not permanently retained in cells and detailed knowledge of elution char­ acteristics and reutilization in vivo is essential to experimental design and interpretation of results. 51Cr is a particularly good label in that very little elutes from healthy cells, while cell death in vivo leads to elution and rapid clearance of free label from the tissues and, indeed, the whole animal via the urine .47 This latter characteristic makes 51Cr the label of choice for studies on the rapid destruction of allogeneic lymphoctes in vivo . 53 Indium -oxine conjugates have similarly good elution characteristics but, in contrast to 51Cr, following elution in vivo the label is retained in the tissues for long periods of time by phagocytic cells. This latter characteristic has made oxine conjugates of 114mIn particularly advantageous in the use of lymphocytes to target and deposit radioactivity into lymphoid organs to produce prolonged and selective total lymphoid irradiation . 54 Elution of radiolabel can be compensated for by appropriate manipulation. For instance 30 to 50% of 3H -uridine, 55 and to a lesser extent I4C -uridine , 51 is lost from cells in the first 24 hr after injection, but during this time the remaining label enters an acid insoluble phase and further elution is limited . 51 Conse­ quently, uridine labeling of cells with passage from blood to lymph of an intermediate (16 to 24 hr collection) ensures that virtually all radioactivity detected in a final recipient is associated with accredited recirculating T cells and that very little further elution of label occurs . 51 Even without intermediate blood to lymph passage of cells, very good results can be obtained by double-labeling techniques. In this protocol, a test population of cells is labeled with 14C -uridine and a reference population with 3H-uridine alongside the alter­ nate labeling procedure. By comparing 3H/14C ratios in the first group of recipients with the 14C/3H ratio in a second group receiving the alternately labeled cells one can compensate for the variable elution profiles of the two radiolabels. 57 In addition, because the migratory characteristics of the test and control populations are measured simultaneously in the same animals, small differences of just 3% can be detected57,58 — a principle which has been used equally well for 3H- and 14C -thymidine. 59 Cell toxicity varies between different radiolabels, but for most is a function of concentration of radiolabel in the incubation medium and, thus, of the amount taken up by the cell. 51Crlabeled mouse lymphoid cells do not survive or respond to mitogenic stimulation in vitro after transfer to syngeneic recipients as well as do similar unlabeled cells detected by chromosomal markers . 60 This toxicity may explain in part the higher liver activity found with this label compared to 3H-uridine and, consequently, in our laboratory we routinely use a low labeling concentration for 51Cr of 10 pCi/m€. In searching for suitable radiolabels for human studies, the short half life ( 6 hr) and good imaging characteristics of the 7 emission of 99mTc (sodium pertechnetate) were nullified by severe cell toxicity . 61 In contrast, the low cell toxicity of 75Se-L-selenomethionine was nullified by rapid and uneven elution characteristics.47 11 ‘In-labeled oxine has proved to be a particularly good radiolabel for human studies ,62 producing little cell toxicity and remaining associated with the cells, although, paradoxically, its short half life (2.8 days) makes it less attractive for animal studies. 3Hthymidine and 125I-iodo-deoxyuridine are the most commonly used DNA precursors. 125IUdR is five times more toxic than 3H -thymidine,63 and 1 |xCi/m€ is the upper limit of an acceptable in vitro labeling dose. Nevertheless, the ease of detecting 7 -emissions and their low reutil­ ization characteristics make them attractive .47 For longer term studies one should use 3Hthymidine, but even this is toxic and a dose of 0.75 p,Ci/m€ should not be exceeded .42 The most commonly used sources of cells in mice are spleen and LN cell suspensions; in rats, thoracic duct cells; in sheep, cells in efferent lymph from a variety of sites; and in humans, peripheral blood. Ideally, the cells should be marked in vivo and this can be done

Volume l

31

in sheep by regional perfusion, for instance, of the bone marrow64 or of the small bowel.65 However, such techniques have severe limitations in smaller animals and, in addition, the label cannot be confined to selected cell populations. In consequence, most studies involve the removal of cells, in vitro labeling, and reinjection. The first question to be asked is, are the cells removed from the body in a migratory state, as it is clearly futile to observe migration patterns of cells which would otherwise be sessile. Herein lies the advantage of studying cells removed from the lymph or blood. Nevertheless, spleen and LN cell suspen­ sions contain a large proportion of cells in a migratory state and, as long as account is made for sessile cells contained (germinal center cells, plasma cells, etc.), physiologically sig­ nificant conclusions can be drawn .45 Removing cells from the body and in vitro handling procedures become ever more nec­ essary with attempts to further clarify the subtleties and precise mechanisms of lymphocyte migration. Unfortunately, the ability of lymphocytes to migrate, particularly across spe­ cialized postcapillary venules (SPCV), is dependent upon mechanisms which are very sen­ sitive to in vitro handling. In this laboratory, the traditional way of studying the migration of TDL was to collect lymph overnight (16 hr) on ice; to label in vitro, usually involving ficoll-hypaque separation of red blood cells (RBC); and to incubate for 1 hr with radiolabel at 37°C and multiple washing procedures, before injection of cells into syngeneic recipients. Localization of labeled cells was then assessed in a variety of tissues at sequential time intervals. However, in the early 1980s, the elegant studies of Smith and Ford67 68 demon­ strated that the collection and labeling procedures greatly reduced — though only temporarily — the ability of TDL to migrate across SPCV. They found that if, after this extensive in vitro handling, the cells were passaged from blood to lymph of a syngeneic intermediate, then during that in vivo passage the labeled cells recovered their full migratory potential. As long as lymph collections from the intermediate animal were made at room temperature for only short periods (2 to 4 hr) and in vitro handling was minimized, then the prelabeled “ optimal cells” would migrate entirely well in a final syngeneic recipient. At 30 min after injection, localization of optimal cells in LN had reached an unprecedented 40 to 50% of injected dose/gram of tissue as compared to 7% for “ standard cells” collected and labeled in the traditional manner. Furthermore, in this final recipient, optimal cells migrated through to thoracic duct lymph much faster (a broad 7 to 18 hr peak) and in much greater numbers (40% yield) than did standard cells (a 15 to 18 hr peak and 12% yield). It was demonstrated that these differences in migratory characteristics of optimal cells were not due primarily to removal of dead or nonrecirculating cells and selection of a special subset. Furthermore, holding “ optimal cells” in vitro for 16 hr on ice caused their migratory patterns to return to those of “ standard cells” . It was concluded that the tempo of lymphocyte recirculation from blood to lymph was faster than previously estimated with an average LN transit time of 12 hr. It was particularly gratifying that the thoracic-duct recovery of optimal cells (40% injected dose in first 24 hr after injection) was compatible with the rate at which thoracicduct output fell during chronic duct drainage. Clearly, blood to lymph passage of lymphocytes through an intermediate can be used to ensure ( 1 ) that all cells finally studied are accredited recirculating lymphocytes, (2 ) that elution of contained radiolabels has stabilized, and (3) that the labeled cells recover their full migratory capacity. Furthermore, by varying the timing of collection of labeled cells from the intermediates’ thoracic duct one can exploit the different migration rates of T and B cells in order to preferentially study the migration of one cell type or the other. 51 When studying migration of lymphocyte populations in final recipients it is essential to remember that migration is a dynamic process and, therefore, requires repeated observations in time. Furthermore, exit of cells from the blood may be to any of a large number of different organs which in effect compete with each other for blood lymphocytes. In addition, the length of stay of a lymphocyte in an organ is variable according to the nature of the

32

Migration and Homing of Lymphoid Cells

lymphocyte and the organ in which it finds itself. Nevertheless, if it is recirculating, a cell will return to the blood and exit into another organ often within the time period of observation. Consequently, lymphocyte-migration studies demand meticulous design and large numbers of observations. In particular, sequential observations in time of localization in a large number of different tissues, including the blood. Table 1 presents selected data from Smith and Fords’ extensive studies upon migration of optimal cells . 67 68 Blood levels fall exponentially while an early retention of cells in the lungs is followed by a fairly rapid release of these cells back into the circulation — the “ real” lung localization being only about 3% injected dose per gram of tissue. LN locali­ zation rises rapidly to 50% injected dose per gram by 1 hr while spleen increases to 105% injected dose per gram at this time interval. However, spleen transit time is clearly relatively short since spleen localization has fallen to 56% by 6 hr, while LN localization is still increasing, presumably, mainly due to redistribution of cells from the spleen. It is interesting that blood levels remain stable at about 3% from 1 to 24 hr though this is a net effect of a complicated integration of constant influx and efflux from the blood. At early time intervals when blood levels were high, cervical LN competed rather better for lymphocytes than did mesenteric LN. However, by 6 to 12 hr mesenteric LN localization exceeded cervical LN localization, being 99 and 60% injected dose per gram, respectively, at 24 hr after injection. This apparent discrepancy can be explained by later arrival of lymphocytes in the mesenteric LN via afferent lymphatics from the gut and has been cofirmed by sequential autoradiographic studies. 90 This phenomenon is not exclusive to the mesenteric LN (the celiac LN receives small lymphocytes via lymphatic drainage from the liver), nor is it exclusive to the migration of small lymphocytes. If an animal is injected with 14Cthymidine-labeled lymphoblasts from the cervical LN and 3H -thymidine-labeled lymphoblasts from the mesenteric LN, then it is found that cervical blasts localize preferentially in pe­ ripheral LN (e.g., cervical LN) while mesenteric blasts localize preferentially in the gut wall, Peyer’s patch, and mesenteric LN . 69 However, if the afferent lymphatics to the mes­ enteric LN are interrupted before injection of lymphoblasts, then mesenteric-blast localization in the deafferentized mesenteric LN is reduced by 40% while cervical-blast localization remains unchanged as compared to sham -operated controls .90 Clearly not all localization of mesenteric blasts in mesenteric LN is due to preferential interactions between those lym ­ phocytes and the endothelial cells of the mesenteric specialized postcapillary venules. The migration of lymphoblasts may also be used to illustrate the need for backing up whole organ counting with autoradiography. Smith et al .69 found that thoracic-duct blasts localized mainly in the red pulp of the spleen while axillary LN blasts localized predominantly in the white pulp, clearly very different migration characteristics, yet their respective whole organ localizations were identical.

B. The Role of Antigenic Stimuli In Figure 4 the lymph node is represented as a focus at which the blood and lymphvascular systems converge to create a coalescence of three compartments partially separated by endothelia — a blood vascular, lymph vascular, and extravascular compartment. It is in this latter space that the various elements of the immune system interact, but few cells seen here at any point in time are actually indigenous to the LN. Indeed, a review 70 of cell populations in LN reveals that most LN cells are bone-marrow derived and that, in general terms, nonlymphoid cells such as macrophages are brought to the LN via afferent lymph while the lymphoid cells are delivered via the blood vasculature. Most of these cells either pass through the LN in a matter of hours or are retained to subserve some function on only a temporary basis of days or weeks. An in vitro model of an LN might be a culture vessel with two continuously running portals of entry and at least one exit. Recirculating lymphocytes are brought to the LN in the blood , 1 41,44 interact with the

0.773 ± 0.097 0.377 ± 0.055 0.130 ± 0.030 0.158 ± 0.035 0.181 ± 0.013 6.42 ± 0.82 1.9 ± 0.4

36.6 ± 1.8 51.8b ± 7.4 14.1 ± 4.5 1.83 ± 0.8 1.1 ± 0.5 1.06

2 min 29.6 ± 5.3 29.7 ± 6.3 51.8 ± 16.0 16.9 ± 7.8 10.5 ± 4.1 8.73 ± 3.8 1.8 ± 0.4

10 min 11.6 ± 1.2 10.9 ± 1.6 103.3 ± 14.5 39.7 ± 9.1 31.1 ± 8.9 22.0 ± 5.9 0.6 ± 0.1

30 min 3.3 ± 1.2 3.2 ± 0.8 104.5 ± 14.8 57.4 ± 5.9 51.8 ± 9.1 36.9 ± 5.3 0.6 ± 0.1

60 min 3.3 ± 0.3 3.0 ± 0.3 56.5 ± 7.4 64.9 :t 13.2 68.3 ± 8.4 34.1 ± 5.9 0.6 ± 0.1

6 hr

Time after injection of labeled TDL

2.4 ± 0.6 2.2 ± 0.4 50.8 ± 8.9 72.3 ± 8.7 101.1 ± 20.6 41.2 ± 8.5 0.6 ± 0.1

12 hr 3.0 ± 0.7 2.6 ± 0.5 41.1 ± 2.1 60.2 ± 11.8 98.9 ± 10.8 35.3 ± 7.0 0.8 ± 0.1

24 hr

From Smith, M. E. and Ford, W. L., Immunology, 49, 83, 1983. With permission.

a Five animals at each time point. b Mean and standard deviation of percent injected activity per gram tissue.

Note: Localization of 51Cr-TDL passaged from blood to lymph of an intermediate and collected for 1 hr intervals, spun and injected into 2 to 3 final syngeneic recipients (mean weight 167.8 g, standard deviation 17.3 g).

Liver

Peyer’s patch

Mesenteric LN

Superficial cervical LN

Spleen

Lung

Blood (10 m€)

Tissue*

Weight (g)

Table 1 LYMPHOCYTE MIGRATION IN THE RAT

Volume I 33

34

Migration and Homing of Lymphoid Cells

FIGURE 4. This diagram represents the lymph node as a focus at which the blood and lymphvascular systems coalesce to form the three main compartments of the node — separated from each other by endothelial walls.

endothelia of specialized postcapillary venules (SPCV) , 2 and enter the extravascular space. The quality and magnitude of these interactions are crucial to LN function, but are themselves affected by what arrives in the afferent lymph and processes in the extravascular space. Consider what happens when antigenic material is brought to the LN via the afferent lymphatics. For our model we have the rat popliteal LN stimulated by a footpad injection of sheep red blood cells (SRBC). At sequential time intervals after stimulation, various parameters are measured in the stimulated LN and expressed as a ratio of these parameters in the contralateral unstimulated LN. Figure 5 presents the results of such a series of experiments21 in which SRBC-stimulated rats were injected with 5 ,Cr-labeled thoracic-duct lymphocytes (TDL) 1 hr before kill to measure lymphocyte influx from the blood into the popliteal LN. The same rats were injected with a bolus of 86RbCl 45 sec before kill to measure blood flow by fractional distribution of indicators. The LN were then removed, weighed, and radioactivity measured. It can be seen (Figure 5) that all three parameters (weight, blood flow, and lymphocyte influx) peak at two to three times normal values at 3 to 4 days after footpad injection of antigen. An increase in blood flow tends to precede other changes, but closely parallels changes in weight. Increased lymphocyte influx is a slightly later event and can, at all times, be accounted for by increased delivery of lym ­ phocytes to the LN through increased blood flow. Does one have to inject antigen to induce an increase in lymphocyte influx? Figure 6 shows the changes in blood flow (solid lines) and lymphocyte influx (dotted lines) into popliteal LN after footpad injection of either SRBC (greatest increase in both parameters) or syngeneic RBC (least increase in both parameters). Following injection of syngeneic RBC there is a small but significant increase in lymphocyte influx which interestingly occurs without a significant change in blood flow. Perhaps it is the result of trauma of injection which produces this rise in lymphocyte influx be it all smaller and shorter lived than that induced by antigen. Does one have to have radiosensitive cells — in particular lymphocytes — to evoke

Volume /

35

FIGURE 5. Changes in blood flow (-----­ O----­ ), lymphocyte influx (------­ O-------), and popliteal LN weight (— • —O— *— ) over a period of 8 days after the injection of sheep red blood cells into the right footpad. All values are expressed as ratios of right to left popliteal LN. Geometric mean values are plotted and standard errors from day 4 onward. Measurements were made at 1,6, 12, 18, 24 hr and at 24 hr intervals thereafter. (From Drayson, M. T., Smith, M. E., and Ford, W. L., Immunology, 44, 125, 1981. With permission.)

FIGURE 6. Blood flow and lymphocyte influx to the popliteal LN over a period of 3 days after the foot­ pad injection of erythrocytes. As in Figure 5, geometric means of rightileft ratios and standard errors are plotted. (----­ O-----­ ) = blood flow after sheep erythrocytes; (-----­ O-----­ ) = lymphocyte influx after sheep erythrocytes; (-----­ X-----­ ) = blood flow after rat erythrocytes; (------­ X------­ ) = lymphocyte influx after rat erythrocytes. (From Drayson, M. T., Smith, M. E., and Ford, W. L., Immunology, 44, 125, 1981. With permission.)

antigen-induced changes in lymphocyte influx? In the experiments21 summarized in Figure 7, rats were irradiated with 750 rads 3 days before injection of phosphate-buffered saline (PBS), syngeneic RBC, or SRBC into the right footpad. Twenty-four hours later, we meas­ ured changes in weight, blood flow, and lymphocyte influx by the techniques described above. Irrespective of the substance injected, a threefold increase in lymphocyte influx was

36

Migration and Homing of Lymphoid Cells

FIGURE 7. Recipients examined 24 hr after footpad injection and 4 days after whole body irradiation. Sheep erythrocyte (SRBC), rat erythrocytes (RRBC), or phosphate-buffered saline (PBS) were injected into the right footpads, the left feet were not injected. X = individual values for right:left popliteal LN; — = median value of group. (From Drayson, M. T., Smith, M. E., and Ford, W. L., Immunology, 44, 125, 1981. With permission.)

induced with relatively little change in blood flow. Similar results were found at 48 hr, but by 3 days there was relatively little difference between right stimulated popliteal LN and left internal control LN. It appears that an early increase in lymphocyte traffic can be elicited by the trauma of injection of a substance that need not be antigenic. For a prolonged increase in lymphocyte influx both antigen and radiosensitive cells are necessary. These two latter factors may also contribute to early increases. An increase in blood flow to the whole LN is not essential to all increases in lymphocyte influx, but there may be a hidden change in blood flow to the SPCV network by such mechanisms as arteriovenous shunting . 71 In Figure 8 the upper dotted line shows changes in total LN incorporation of intravenously injected 3H -thymidine after footpad injection of SRBC. A tenfold peak increase is achieved by 3 days after stimulation and, from the radiation experiments described above, it would seem likely that this cell proliferation (primarily lymphoid as assessed by autoradiography) is partly responsible for the increase in lymphocyte traffic. However, one should point out that this proliferation persists for longer ( 6 days) than the increase in lymphocyte influx. Regarding the role of lymphoid proliferation in inducing lymphocyte influx it is worth mentioning that the local growth of syngeneic Roser leukemia cells (phenotypically T cells) in a popliteal LN induces not only an increase in LN weight but also an increase in lymphocyte influx .90 The detailed studies of Anderson et al .58 clearly demonstrated that antigen stimulation induced endothelial proliferation and an overall increase in size of the SPCV network. We have studied high endothelial cell proliferation by autoradiographic examination of popliteal

Volume I

37

FIGURE 8. Changes in popliteal LN incorporation of [3H]-thymidine, production of sulfated glycolipid, and high endothelial cell [3H]-thymidine-labeling index after a right footpad injection of sheep erythrocytes. (****■*•••) = geometric mean and standard error for right:left [3H]-thymidine incorporation. (---­ • ---­ ) = [35S]-sulfate incorporation, right:left ratios, one to two rats at each time point. (A) = mean and standard error of high endothelial cell [3H]-thymidine-labeling index in unstimulated left popliteal LN and ( - ^ - ) = in stimulated right popliteal LN. (From Drayson, M. T., Smith, M. E., and Ford, W. L., Immunology, 44, 125, 1981. With permission.)

LN 4 hr after i.v. administration of 3H -thymidine at sequential time intervals after footpad injection of SRBC. Preliminary results represented in the solid line of Figure 8 show a peak increase in high endothelial cell labeling index of 8 % compared to 0.5 to 0.9% in contralateral control LN. The lower, dotted line (Figure 8 ) represents changes in production of sulfated glycolipid by the SPCV of SRBC -stimulated popliteal LN. Andrews showed autoradiographically that 30 min after footpad injection of 35S-sulfate there was selective uptake of that sulfate by high endothelial cells in the draining LN . 72 It transpired that the sulfate was incorporated into a sulfated glycolipid , 8,9 and there is some evidence to suggest that the molecule may be involved in lymphocyte/endothelial interactions. 72 In this experiment (Figure 8 ), SRBCstimulated and -unstimulated popliteal LN were removed 1 hr after in vivo administration of 35S -sulfate, and radioactivity associated with macromolecules measured . 21 Stimulated LN showed an increased production of this sulfated macromolecule which may have been due to an increase in size of the SPCV network and/or increased metabolism by individual high endothelial cells. In summary, increases in lymphocyte influx may be mediated by (1) increased delivery of lymphocytes to the SPCV by increased LN blood flow and/or arteriovenous shunting, (2) increased size of the SPCV network by endothelial cell proliferation, (3) increased efficiency of extraction of lymphocytes by high endothelial cells, perhaps by changes in metabolic activity and production of sulfated glycolipids. The mechanisms leading up to such phe ­ nomena and their relative contributions are probably complex. Clearly, short-lived increases in lymphocyte influx can be induced by injection of even nonantigenic substances, but

38

Migration and Homing of Lymphoid Cells

prolonged increases in lymphocyte influx probably depend upon antigenic material and lymphoid proliferation.

C. The Role of Afferent Lymph So far, we have considered the effect of antigen arriving via the lymphatics into a relatively quiescent LN. In the next series of experiments, we used cervical LN which in the “ Man ­ chester” rat are comparatively active in terms of lymphoid proliferation, germinal centers, and plasma cells. What happens if we take these cervical LNs and cut off their afferent lymph supply? Presumably after cessation of antigenic stimulus the LN will slip back into a quiescent state, but what of the effect of cessation of afferent lymph flow and its supply of lymph-borne cells? Hendriks and colleagues were able to accomplish a total and prolonged deafferentization of the popliteal LN by removing it from its normal anatomical position in the popliteal fossa to a subcutaneous site on the biceps femoris muscle. 22,23 They found reduced numbers of LN macrophages by 1 week and almost complete disappearance of these cells by 6 weeks after operation. The high endothelial cells had flattened by 3 weeks and few lymphocytes could be seen in transit across the vessel wall. By 8 weeks after operation, lymphoblasts, plasma cells, and germinal centers had disappeared and the gross depletion of both mac­ rophages and lymphocytes had left little more than a reticuloendothelial framework by 12 weeks after operation. These histological studies indicate that many of the nonlymphoid cells in LN are derived from afferent lymph and that in the absence of further antigenic stimulation many of the features of immunological activity rapidly wane and the specialized characteristics of LN tissue disappear. In our own study24 we have dissected away the afferent lymphatics of rat cervical lymph nodes which were then encased in silicone rubber tubes — a procedure which maintains a prolonged and complete deafferentization of these immunologically active LN without dis­ turbing them from their normal anatomical position. At intervals up to 12 weeks later, the weight, blood flow, and lymphocyte influx into these LN were measured simultaneously as described above and the nodes were examined histologically. In addition, incorporation of 125IUdR by deafferentized LN and the ability of their HEV to secrete sulfated glycolipid were measured. In most experiments only the cervical LN on one side were deafferentized, those on the contralateral side being used as internal controls. In Figure 9 changes in weight and blood flow are presented and show that the operative procedure has not altered blood flow at 24 hr later, but that weight has increased by 25%. Histologically, these LN are normal except for a gross distention of the medullary and other lymphatic sinuses with small lymphocytes, perhaps because of a lack of afferent lymph flow to flush them through to the efferent lymphatic. Subsequently, both blood flow and weight declined rapidly, blood flow at 60% of normal at 1 week and 6 % by 6 weeks. At 12 weeks after deafferentization, LN were only a few percent of their original weight and blood flow was undetectable by the 86RbCl technique. Histologically, there was little to suggest that the structure was an LN, having regressed to a relatively acellular mesh work of reticular fibers and blood vessels. Figure 10 displays changes in lymphocyte traffic with a fall to 60% of normal values in the first 24 hr after operation (in the presence of “ normal” blood flow). The continued decline in lymphocyte influx was biphasic with a plateau of 15% normal values between 1 and 3 weeks, but being undetectable by 6 weeks after operation. Histologically, the SPCV appeared normal at 1 day after deafferentization, but by 1 week there was a reduction in height of their endothelial cells and by 3 weeks the SPCV network was clearly involuting with very few recognizable SPCV seen at 6 weeks after operation. LN production of sulfated glycolipid (Figure 10) was decreased after deafferentization and appeared to correlate with reduction in size of the SPCV network — autoradiography of 35S incorporation did not

Volume I

39

FIGURE 9. Blood flow and LN weight after deafferentization expressed as a ratio of the corresponding value for the contralateral LN. Mean ± SE plotted. The number of LN pairs at each time point is stated on the top line. Note the log scale. (From Drayson, M. T. and Ford, W. L., Immunobiology, 168, 362, 1984. With permission.)

suggest a change in uptake by individual high endothelial cells as compared with those of the contralateral control LN. The changes observed in the SPCVs, that is flattening, dimin ­ ished sulfate incorporation, and eventual disappearance of the SPCV network, were preceded by reduced lymphocyte influx and, therefore, cannot be the cause of the reduced influx of lymphocytes. Either the changes in SPCVs could be the result of the decreased entry of lymphocytes or both could be consequences of the lack of a stimulus to lymphocyte entry, such as arrival of antigenic material. Figure 11 is a composite of changes in weight, blood flow, and lymphocyte influx with the addition of LN incorporation of 125IUdR. We cannot offer a dogmatic explanation for the significantly reduced lymphocyte influx at 24 hr after deafferentization. The later decrease of lymphocyte influx appears to be biphasic with no fall between 1 and 3 weeks after deafferentization. It is likely that much of the fall in the first week to 12 to 15% of the lymphocyte influx in control LN may be attributable to deprivation of antigen. Certainly, the rate at which influx drops is closely related to 125IUdR incorporation into lymphoblasts which can be accepted as due to antigenic stimulation. Since the existence of an LN does not require antigenic stimulation, as is testified by the human 12 and sheep fetus 13 and germ free rats , 14 lack of some other element in afferent lymph must be responsible for the later cessation of lymphocyte influx and virtual disappearance of the LN after 12 weeks of deafferentization. The final loss of lymphocyte influx and, indeed, loss of the SPCV network may be due to the denial of lymph borne accessory cells such as macrophages and, in

40

Migration and Homing of Lymphoid Cells

FIGURE 10. Lymphocyte influx and [35S]-sulfated glycolipid secretion after deafferentization expressed as a ratio of the corresponding values for the contra­ lateral LN. Mean ± SE plotted. (From Drayson, M. T. and Ford, W. L., Im ­ munobiology, 168, 362, 1984. With permission.)

particular, dendritic cells. The dramatic decline in LN macrophages by 1 week and almost total loss of both macrophages and interdigitating cells (IDC) from the paracortex between 3 and 6 weeks after operation immediately precedes the final cessation of lymphocyte influx. We found that removal of the silicone rubber tubes at 6 weeks allowed reafferentization of the LN and complete restoration of normal appearance and functional parameters. Hendriks succeeded in partially restoring the histological appearance of SPCVs by injecting antigen and macrophages. 23 In supplementary experiments, we found that UV irradiation of the LN at the time of deafferentization accelerated the reduction in lymphocyte influx without affecting blood flow suggesting that a UV-sensitive cell like the IDC may influence lym ­ phocyte influx . 24 Moreover, we have found that footpad injection of class II positive 3Hthymidine-labeled syngeneic dendritic cells from rat intestinal lymph leads to the localization of these cells in the paracortex of the draining popliteal LN. On autoradiography of thin sections, these cells appear like interdigitating cells and their arrival is associated with marked changes in lymphocyte influx which persist for some weeks .90 In conclusion, the involution of the deafferentized LN is partly due to the lack of antigen, but progression to the complete loss of specialized structure and function is probably due to the lack of other factors including nonlymphoid cells that normally arrive in afferent lymph.

Volume /

41

FIGURE 11. The mean values only from Figures 9 and 10 (except 35S incorpo­ ration) are plotted together with changes in ,25IudR incorporation from 0 to 6 weeks after deafferentization. (From Drayson, M. T. and Ford, W. L., Immunobiology, 168, 362, 1984. With permission.)

D. The Site of Selection of Antigen-Specific Cells Following primary antigenic stimulation, lymphocyte traffic through an LN is substantially increased20’21’72,73 75 and it is through this traffic of lymphocytes that antigen-specific lym ­ phocytes (ASL) are delivered to the LN .44,76 Some of the mechanisms involved in increasing lymphocyte influx have already been discussed, but this section is concerned with the precise point in the lymphocyte migration pathway at which antigen is presented to the recirculating cells. It is generally held that lymphocytes cross SPCV irrespective of their antigenic spec­ ificity; it is only the encounter with antigen in the extravascular space that halts the migration of ASL while the majority of lymphocytes return to the blood via the efferent lymphatic . 2144 More recently however, it has been suggested that entry of lymphocytes into LN is not a random process and that entry of ASL is selectively enhanced by the presentation of antigen on the luminal surface of SPCV. This heterodox hypothesis arose from the finding that in sheep the whole animal became specifically unresponsive to an antigen if deprived of the efferent lymph (removed by chronic drainage) emerging from a popliteal LN undergoing repeated antigenic stimulation . 77 78 Further support for this hypothesis (i.e., antigen selection of ASL within SPCVs) was provided when endothelial cells were shown to present histo ­ compatibility antigens in vitro . 79 80 In the experiments about to be described, it was attempted to test this hypothesis by pin­ pointing the site of ASL selection relative to the blood vessel wall in antigenically stimulated LN. The experimental design centered around measuring the localization of a negatively

42

Migration and Homing of Lymphoid Cells

FIGURE 12. Diagrammatic outline of the experimental design. A 3H-uridine-labeled pop­ ulation of TDL containing ASL was mixed with a 14C-uridine-labeled population of TDL containing no ASL and injected i.v. into a syngeneic recipient. The recipient rat had previously received alloantigenic stimulation of the right popliteal LN, the left popliteal LN was used as an internal control. At 1 or 24 hr after injection of labeled cells the 3H/,4C ratio was compared in the right and left popliteal LN to obtain an index of selective localization of ASL in the stimulated LN. (From Drayson, M. T., Transplantation, 41, 745, 1986. With permission.)

selected population of lymphocytes, i.e., lacking ASL, relative to the localization of a normal population that included ASL (see Figure 12). In primary responses to “ conventional antigen” , such as sheep erythrocytes and flagellin, the proportion of ASL in TDL is 1 in 105 and 1 in 106, respectively . 81,82 Clearly this would not render sufficient difference between a negatively selected population of cells and the reference population for the purpose of our study. We, therefore, used an alloantigen system where the frequency of unprimed ASL is in the region of 5 to 12% in thoracic duct T cells . 57,83,84 The starting point of each experiment was a single population of TDL from AO strain rats. These TDL were divided and one half labeled in vitro with 3H-uridine while the other half was labeled with 14C -uridine. As described previously, uridine labels T cells 10 to 15 times more heavily than B cells, thus allowing us to focus attention upon T -cell migration.

Volume /

43

After labeling, one population of cells was passaged from blood to lymph of a lightly irradiated F 1 hybrid (AO x PVG or AO x DA) while the alternately labeled population was passaged through a lightly irradiated syngeneic intermediate. Passage of parental cells through an Fi hybrid produces an extremely good negative selection since 98% of cells specific for the alloantigen of the second parent (PVG or DA) are removed . 85 Thus, we produced our negatively selected population and a reference population containing 5 to 12% ASL pos ­ sessing alternate 3H and 14C labels. In addition, the blood-to-lymph passage ensured (1) that we were studying accredited recirculating lymphocytes, (2 ) that the radiolabels had entered an acid-insoluble phase and that further elution had stabilized, and (3) that collected cells migrate in a near physiological manner with rapid entry into recipients’ lymphoid tissues. This was achieved by limiting TDL collections to periods of only 4 hr at room temperature and by subjecting cells to only a minimum of in vitro handling before injection into final recipients. These two populations were mixed and injected into final recipients which had previously received (1) a right footpad injection of F, hybrid cells (the same strain as used for negative selection) or (2) a left footpad injection of syngeneic cells. Thus, we could measure the localization of the two populations of cells in an antigen-stimulated popliteal LN and compare their relative localizations to that in the contralateral control popliteal LN. Antigen stimulation was carried out between 12 and 72 hr before injection of labeled cells so that we could study localization during the inductive phase of the immune response — the time at which selection of ASL from the recirculating pool would be most crucial. The timing between injection of cells and measuring their localization in tissue was crucial as to whether one was studying the effect of entry into LNs from the blood (1 hr kill) or a combination of entry and retention (24 hr kill). From Smith and Fords’ findings67 68 described previously, it could be estimated that by 1 hr most cells would have left the blood stream and LN localization would have achieved near maximum levels. On the other hand, no cells which had entered the LN would have had time to leave via efferent lymphatics since maximum transit time is 2 hr and modal transit time 12 hr. Consequently, LN localization at 1 hr would purely reflect lymphocyte entry across SPCV and any enhanced localization of the reference population of cells (containing ASL) as compared to the negatively selected population would reflect enhanced entry of ASL across SPCV. By 24 hr after injection there will have been between 1 and 2 passages of cells through LN. Consequently, three cohorts of cells of roughly equal numbers are considered: cells that (1) entered the LN as a bolus in the first hour and left by 12 to 13 hr, (2) entered the LN at 1 to 12 hr and left by 24 hr, and (3) entered LN at 12 to 24 hr and were still present at 24 hr. Thus, any enhanced localization of the reference population of cells (containing ASL) at 24 hr might be due to selective retention of ASL from the first two of the above cohorts and/or selective entry of ASL from the third cohort of cells. Dray son gives details of these experiments and the methods of calculation . 86 In brief, at 1 or 24 hr after injection an index of selective localization (Isl) was calculated according to the formula

where Ls = localization of syngeneic, passaged TDL and Ln = localization of semiallogeneic, passaged TDL. Operatively, this becomes

44

Migration and Homing of Lymphoid Cells

FIGURE 13. Indices of selective localization in alloantigen-stimulated LN 24 hr after injection of labeled cells. Indices greater than unity indicate selective localization of ASL. Geometric means and standard errors of groups of indices are displayed with the results of a one-tailed students’ t test of significance of positive deviation from unity. Note positive deviations in groups studied 36 hr and 3 days after antigen stimulation and that indices for inguinal LN pairs did not deviate from unity. (From Drayson, M. T., Transplantation, 41, 745, 1986. With permission.)

Any selective localization of ASL in the stimulated right popliteal LN was, thus, signified by an Isl greater than 1.0. Possible differences between the migratory behavior of the two TDL populations through the alloantigen-stimulated LN could be predicted. Prediction of differences in entry depended upon the proportion of cells in the syngeneic, passaged TDL that were ASL (5 to 12%) and the degree of enhanced entry of ASL over that of nonantigenspecific lymphocytes (NSL) (a maximum of fourfold). In the event of enhanced entry of ASL across SPCV, it was calculated that the Isl would be in the range of 1.05 to 1.12. Prediction of differences in retention depended upon the proportion of cells in the syngeneic, passaged TDL which were ASL (5 to 12%) and the number of passages of cells through the LN by 24 hours (1 to 2). The calculated range of Isl was, thus, 1.05 to 1.24. Figure 13 gives the results of experiments in which the 22 final recipients were killed 24 hr after i.v. injection of labeled cells. In experiment 3, (AO x P V G ^ hybrids were used as intermediates for passage of TDL and as donors for stimulation of the right popliteal LN. (AO x DA)Fi hybrids were used in experiment 4. In groups 4B and 3B, studied 3 days after alloantigen stimulation, the mean Isl were 1.5 and 1.16, respectively. Group 4A, studied 36 hr after alloantigen stimulation, had a mean Isl of 1.10. Thus, all these groups had significantly positive deviations from unity which were consistent with the predicted range (1.05 to 24). In group 3A, studied only 16 hr after stimulation, the mean Isl was 1.05, but was not significantly different from unity. All inguinal LN pairs studied produced Isl which were not significantly different from unity (in this strain of rats, inguinal LN are not stimulated by footpad injection of antigen since secondary lymphatic drainage is exclusively to the iliac LN). Thus, at 24 hr after injection of labeled cells these experiments demonstrated a selective localization of ASL in LN at 36 hr and 3 days after alloantigen stimulation. However, they

Volume /

45

FIGURE 14. Indices of selective localization in alloantigen-stimulated LN 1 hour after injection of labeled cells. Indices greater than unity indicate selective localization of ASL. Geometric means and standard errors of groups of indices are displayed. Note the lack of significant deviation from unity in both popliteal and inguinal LN pairs, which was confirmed in a one-tailed students’ t test. (From Drayson, M. T., Transplantation, 41, 745, 1986. With permission.)

did not elucidate whether selective localization was due to selective entry across SPCV or selective retention within the LN extravascular space. Figure 14 depicts the results of 17 rats killed 1 hr after cell transfer and 3 days or 12 hr after antigenic stimulation of the right popliteal LN. In all three groups the indices of selective localization both for popliteal and unstimulated inguinal LN pairs were not significantly different from unity. This demonstrates that there was no selective entry of ASL across SPCV indicating that the selective localization of ASL found at 24 hr (Figure 13) was due to selective retention of ASL within the LN extravascular space after random entry of TDL across HEV. With regard to the absence of enhanced entry of ASL, one should consider what proportion of recirculating lymphocytes delivered to an LN in the arterial blood actually leaves the bloodstream. Two studies in sheep74*87 utilized knowledge of LN blood flow, blood lym ­ phocyte counts, and number of lymphocytes in efferent lymph to calculate that the proportion of lymphocytes entering an LN is about 25%. A more direct method 88 utilizing localization of 51Cr-labeled TDL in an isolated chain of perfused mesenteric LN indicated a figure of about 11%. The latter estimate is probably an underestimate because less than optimal conditions for TDL migration were used (16 hr collections of TDL on ice, unpassaged after labeling) and not all blood entering an LN necessarily passes through a SPCV. Since all blood borne lymphocytes enter LN via SPCV the extraction of blood lymphocytes across the SPCV wall is probably greater than either of the above estimates and may approach 25 to 50%. Thus, the maximum potential for enhanced entry of ASL could not be more than two to four times that of NSL. We have reliably shown that a twofold enhancement does not occur; any less of an enhancement would be of questionable physiological significance, particularly in the light of a threefold increase in total lymphocyte entry to the stimulated LN associated with increased blood flow.

46

Migration and Homing of Lymphoid Cells

We should point out that the present study is confined to the migration of immunologically naive lymphocytes during the inductive phase of a primary immune response. That only two alloantigens were used and that the physiology of a response against conventional or nonalloantigens (particularly where the frequency of ASL is low) may be different. Nevertheless, direct evidence for antigen presentation by the endothelial cells of SPCV in vitro or in vivo is lacking. Indeed, the high endothelial cells of the rat have never, as far as we know, been seen to express class II antigens in vivo. If antigen presentation were to occur in the lumen of SPCV, why should NSL enter the LN? This last question brings to mind a suitable quote to end on, “ there has been little speculation on the functional significance of lymphocyte recirculation apart from the suggestion that it increases the efficiency of regional immune responses by allowing antigen-induced selection of precursors from a pool larger than that accommodated by the regional nodes alone ” . 89

REFERENCES 1. Gowans, J. L., The recirculation of lymphocytes from blood to lymph in the rat, J. Physiol. (London), 146, 54, 1959. 2. Gowans, J. L. and Knight, E. J., The route of recirculation of lymphocytes in the rat, Proc. R. Soc. London Ser. B:, 159, 257, 1964. 3. Smith, C. and Henon, B. K., Histological and histochemical study of high endothelium of post-capillary veins of the lymph node, Anat. Rec., 135, 207, 1959. 4. Ropke, C., Jorgensen, O., and Claesson, M. H., Histochemical studies of high endothelial venules of lymph nodes and Peyer’s patches in the mouse, Z. Zellforsch. Mikrosk. Anat., 131, 287, 1972. 5. Anderson, N. D., Anderson, A. O., and Wyllie, R. G., Microvascular changes in lymph nodes draining skin allografts, Am. J. Pathol., 81, 131, 1975. 6. Claesson, M. H., Jorgensen, O., and Ropke, C., Light and electron microscopic studies of the paracortical post-capillary high endothelial venules, Z. Zellforsch. Mikrosk. Anat., 119, 195, 1971. 7. van Deurs, B. and Ropke, C., The postnatal development of high endothelial venules in lymph nodes of mice, Anat. Rec., 181, 659, 1975. 8. Anderson, N. D., Anderson, A. O., and Wyllie, R. G., Specialised structure and metabolic activities of high endothelial venules in rat lymphatic tissues, Immunology, 31, 455, 1976. 9. Andrews, P., Milsom, D. W., and Ford, W. L., Migration of lymphocytes across specialised vascular endothelium. V. Production of a sulphated macromolecule by high endothelial cells in lymph nodes, J. Cell Sci. , 57, 277, 1982. 10. Andrews, P., Milsom, D. W., and Stoddart, R. W., Glycoconjugates from high endothelial cells. I. Partial characterisation of sulphated glycoconjugate from the high endothelial cells of rat lymph nodes, J. Cell Sci., 59, 231, 1983. 11. Ford, W. L., Allen, T. D., Pitt, M. A., Smith, M. E., and Stoddart, R. W., The migration of lymphocytes across specialised vascular endothelium. VIII. Physical and chemical conditions influencing the surface morphology of lymphocytes and their ability to enter lymph nodes, Am. J. Anat., 170, 377, 1984. 12. Bailey, R. P. and Weiss, L., Light and electron microscopic studies of postcapillary venules in developing human fetal lymph nodes, Am. J. Anat., 143, 43, 1975. 13. Reynolds, J., Gut-associated lymphoid tissue in lambs before and after birth, Monogr. Allergy, 16, 187, 1980. 14. Fossum, S. and Vaaland, J. L., The architecture of rat lymph nodes. I. Combined light and electron microscopy of lymph node cell types, Anat. Histol. Embryol., 167, 229, 1983. 15. Parrott, D. M. V., de Sousa, M. A. B., and East, J., Thymus-dependent areas in the lymphoid organs of neonatally thymectomised mice, J. Exp. Med., 123, 191, 1966. 16. Goldschneider, I. and McGregor, D. D., Migration of lymphocytes and thymocytes in the rat. I. The route of migration from blood to spleen and lymph nodes, J. Exp. Med., 127, 155, 1968. 17. de Sousa, M. A. B., Parrott, D. M. V., and Pantelouris, E. M., The lymphoid tissue in mice with congenital aplasia of the thymus, Clin. Exp. Immunol., 4, 637, 1969. 18. Fossum, S., Smith, M. E., Bell, E. B., and Ford, W. L., The architecture of rat lymph nodes. III. The lymph nodes and lymph borne cells of the congenitally athymic nude rat, Scand. J. Immunol., 12, 421, 1980.

Volume I

47

19. Fossum, S., Smith, M. E., and Ford, W. L., The recirculation of T and B lymphocytes in the athymic, nude rat, Scand. J. Immunol., 17, 551, 1983. 20. Herman, P. G., Yamamoto, I., and Meilins, H. Z., Blood microcirculation in the lymph node during the primary immune response, J. Exp. Med., 136, 697, 1972. 21. Drayson, M. T., Smith, M. E., and Ford, W. L., The sequence of changes of blood flow and lymphocyte influx to stimulated rat lymph nodes, Immunology, 44, 125, 1981. 22. Hendriks, H. R., Eestermans, I. L., and Hoefsmit, E. C. M., Depletion of macrophages and disap­ pearance of postcapillary high endothelial venules in lymph nodes deprived of afferent lymphatic vessels, Cell Tissue Res., 211, 375, 1980. 23. Hendriks, H. R. and Eestermans, I. L., Disappearance and reappearance of high endothelial venules and immigrating lymphocytes in lymph nodes deprived of afferent lymphatic vessels: a possible regulatory role of macrophages in lymphocytes migration, Eur. J. Immunol., 13, 663, 1983. 24. Drayson, M. T. and Ford, W. L., Afferent lymph and lymph borne cells: their influence on lymph node function, Immunobiology, 168, 362, 1984. 25. Jaffe, E. A., Nachman, R. L., Becker, C. G., and Minick, R. C., Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria, J. Clin. Invest., 52, 2745, 1973. 26. Gimbrone, M. A., Cotran, R. S., and Folkman, J., Human vascular endothelial cells in culture, J. Cell B iol, 60, 673, 1974. 27. Bloom, A. L., Giddings, J. C., and Wilks, C. J., Factor VIII on the vascular intima: possible importance in haemostasis and thrombosis, Nature (London), 241, 217, 1973. 28. Hoyer, L. W., de los Santos, R. P., and Hoyer, J. R., Antihemophilic factor antigen. Localisation in endothelial cells by immunofluorescent microscopy, J. Clin. Invest., 52, 2737, 1973. 29. Jaffe, E. A., Ed., Biology of Endothelial Cells, Martinus Nijhoff, The Hague, 1984, chap. 1. 30. Folkman, J. F., Haudenschild, C. C., and Zetter, B. R., Long term culture of capillary endothelial cells, Proc. Natl. Acad. Sci. U.S.A., 76, 5217, 1979. 31. Maciag, T., Cerundolo, J., Ilsley, S., Kelley, P. R., and Forand, R., An endothelial cell growth factor from bovine hypothalamus: identification and partial characterisation, Proc. Natl. Acad. Sci. U.S.A., 76, 5674, 1979. 32. Voyta, J. C., Via, D. P., Butterfield, C. E., and Zetter, B. R., Identification and isolation of endothelial cells based on their increased uptake of acetylated low-density lipoprotein, J. Cell Biol., 99, 2034, 1984. 33. Ryan, U. S., Ryan, J. W., Whitaker, C., and Chiu, A., Localisation of angiotensin converting enzyme (Kininase II). II. Immunocytochemistry and immunofluorescence, Tissue Cell, 8, 125, 1976. 34. Todd, A. S., The histological localisation of fibrinolysin activator, J. Pathol. Bacteriol., 78, 281, 1959. 35. Levin, R. I., Weksler, B. B., Marcus, A. J., and Jaffe, E. A., Prostacyclin production by endothelial cells, in Biology of Endothelial Cells, Jaffe, E. A., Ed., Martinus Nijhoff, The Hague, 1984, chap. 24. 36. Ager, A., Gordon, J. L., Moncada, S., Pearson, J. D., Salmon, J. A., and Trevethick, M. A., Effects of isolation and culture on prostaglandin synthesis by porcine aortic endothelial and smooth muscle cells, J. Cell. Physiol., 110, 9, 1982. 37. Bowman, P. D., Betz, L. A., Ar, D., Wolinksy, J. S., Penney, J. B., Shivers, R. R., and Goldstein, G. W., Primary culture of capillary endothelium from rat brain, In Vitro, 17, 353, 1981. 38. Haston, W. S., Sheilds, J. M., and Wilkinson, P. C., Lymphocytes locomotion and attachment on twodimensional surfaces and in three-dimensional matrices, J. Cell Biol., 92, 747, 1982. 39. Schor, S. L., Allen, T. D., and Winn, B., Lymphocyte migration into three-dimensional collagen matrices: a quantitative study, J. Cell Biol., 96, 1089, 1983. 40. Ager, A., Isolation and culture of high endothelial cells from rat lymph nodes, J. Cell Sci., 87, 133, 1987. 41. Gowans, J. L. and McGregor, D. D., The immunological activities of lymphocytes, Prog. Allergy, 9, 1, 1965. 42. Butcher, E. C. and Ford, W. L., Following cellular traffic: methods of labelling lymphocytes and other cells to trace their migration in vivo, in A Handbook o f Experimental Immunology, 4th ed., Weir, D. M., Ed., Blackwell Scientific, Oxford, 1984. 43. Strober, S. and Dilley, J., Biological characteristics of T and B memory lymphocytes in the rat, J. Exp. Med., 137, 1275, 1973. 44. Ford, W. L., Lymphocyte migration and immune responses, Prog. Allergy, 19, 1, 1975. 45. Ford, W. L. and Smith, M. E., Experimental approaches to lymphocyte traffic; pitfalls of the tracer sample method, Adv. Exp. Med. Biol., 149, 139, 1982. 46. Howard, J. C., Hunt, S. V., and Gowans, J. L., Identification of marrow-defined and thymus-defined small lymphocytes in the lymphoid tissue and thoracic duct lymph of normal rats, J. Exp. Med., 135, 200, 1972. 47. Rannie, G. H. and Donald, K. J., Estimation of the migration of thoracic duct lymphocytes to non­ lymphoid tissues, Cell Tissue Kinet., 10, 523, 1977.

48

Migration and Homing of Lymphoid Cells

48. Labaroff, D. M., Cellular requirements for the rejection of skin allografts in rats, J. Exp. Med., 138, 331, 1973. 49. Everett, N. B. and Tyler, R. W., Lymphopoiesis in the thymus and other tissues: functional implications, Int. Rev. Cytol., 22, 205, 1967. 50. Howard, J. C ., The life-span and recirculation of marrow-derived small lymphocytes from rat thoracic duct, J. Exp. Med., 135, 185, 1972. 51. Ford, W. L. and Simmonds, S. J,, The tempo of lymphocyte recirculation from blood to lymph in the rat, Cell Tissue Kinet., 5, 175, 1972. 52. Nieuwenhuis, P. and Ford, W. L., Comparative migration of B- and T-lymphocytes in the rat spleen and lymph nodes, Cell. Immunol., 23, 254, 1976. 53. Rolstad, B. and Ford, W. L., The rapid elimination of allogeneic lymphocytes: relationship to established mechanisms of immunity and to lymphocyte traffic, Immunol. Rev., 73, 87, 1983. 54. Birch, M., Sharma, H. L., Bell, E. B., and Ford, W. L., The carriage and delivery of substances to lymphatic tissues by recirculating lymphocytes. II. Long term selective irradiation of the spleen and lymph nodes by deposition of indium-114m, Immunology, 58, 359, 1986. 55. Goldschneider, I. and McGregor, D. D., Migration of lymphocytes and thymocytes in the rat. II. Circulation of lymphocytes and thymocytes from blood to lymph, Lab. Invest., 18, 397, 1968. 56. Ford, W. L. and Simmonds, S. J., The tempo of lymphocyte recirculation from blood to lymph in the rat, Cell Tissue Kinet., 5, 175, 1972. 57. Atkins, R. C. and Ford, W. L., Early cellular events in a systemic graft-versus-host reaction. I. The migration of responding and nonresponding donor lymphocytes, J. Exp. Med., 141, 664, 1975. 58. Ford, W. L., The recruitment of recirculating lymphocytes in the antigenically stimulated spleen. Specific and non-specific consequences of initiating a secondary antibody response, Clin. Exp. Immunol, 12, 243, 1972. 59. Thursh, D. R. and Emeson, E. E., Selective DNA synthesis by cells specifically localizing in response to xenogeneic erthrocytes, J. Exp. Med., 138, 659, 1973. 60. DoenhofT, M. J. and Davies, A. J. S., The capacity of mitogen-responsive T cells to ‘home’ to their tissue of origin analysed by means of chromosome markers, J. Exp. Med., 143, 660, 1976. 61. Rannie, G. H., Thakur, M. L., and Ford, W. L., An experimental comparison of radioactive labels with potential application to lymphocyte migration studies in patients, Clin. Exp. Immunol, 29, 509, 1977. 62. Wagstaff, J., Gibson, C., Thatcher, N., Ford, W. L., Sharma, H., Benson, W., and Crowther, D., A method for following human lymphocyte traffic using indium-111 oxine labelling, Clin. Exp. Immunol., 43, 435, 1981. 63. Hofer, K, G. and Hughes, W. L., Radiotoxicity of intranuclear, tritium, 125I Iodine and 13lIodine, Radiat. Res., 47, 94, 1971. 64. Pabst, R., Reynolds, J., and Miyasaka, M., The role of Peyer’s patches and bone marrow in lymphocyte production and migration in lambs, in Immunology of the Sheep, Morris, B. and Miyasaka, M., Eds., Roche, Basel, 1985, 237. 65. Reynolds, J. and Pabst, R., The emigration of lymphocytes from Peyer’s patches in sheep, Eur. J. Immunol., 14, 1, 1984. 66. Ford, W. L., Lymphocyte migration and immune responses, Prog. Allergy, 9, 1, 1975. 67. Smith, M. E. and Ford, W. L., The recirculating lymphocyte pool of the rat: a systematic description of the migrating behaviour of recirculating lymphocytes, Immunology, 49, 83, 1983. 68. Smith, M. E. and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium, Cell. Immunol, 76, 161, 1983. 69. Smith, M. E., Martin, A. F., and Ford, W. L., Migration of lymphoblasts in the rat, Monogr. Allergy., 16, 203, 1980. 70. Fossum, S. and Ford, W. L., The organization of cell populations within lymph nodes: their origin, life history and functional relationships, Histopathology, 9, 469, 1985. 71. Herman, P. G., Utsonomiya, R., and Hessel, S. J., Arteriovenous shunting in the lymph node before and after antigenic stimulus, Immunology, 36, 793, 1979. 72. Andrews, P., Ford, W. L., and Stoddart, R. W ., Metabolic studies of high walled endothelium of postcapillary venules in rat lymph nodes, in Blood Cells and Vessel Walls: Functional Interactions, CIBA Found. Symp. No. 71, O ’Connor, M., Ed., CIBA Medical, West Caldwell, N.J., 1980, 211. 73. Hay, J. B., Johnston, M. G., Vadas, P., Chin, W., Issekutz, T., and Movat, H. Z., Relationships between changes in blood flow and lymphocyte migration induced by antigen, Monogr. Allergy., 16, 112, 1980. 74. Cahill, R. N. P., Frost, H., and Trnka, Z., The effects of antigen on the migration of recirculating lymphocytes through single lymph nodes, J. Exp. Med., 143, 870, 1976. 75. Ottaway, C. A. and Parrott, D. M. V., Regional blood flow and its relationship to lymphocyte and lymphoblast traffic during a primary immune response, J. Exp. Med., 150, 218, 1979.

Volume l

49

76. Hall, J. G. and Morris, B., The effect of X-irradiation of the popliteal lymph node on its output of lymphocytes and immunological unresponsiveness, Lancet, 2, 1077, 1964. 77. McConnell, I., Lachman, P. J., and Hobart, M. J., The restoration of specific immunological virginity, Nature (London), 250, 113, 1974. 78. Hopkins, J., McConnell, I., and Lachmann, P. J., Specific selection of antigen reactive lymphocytes into antigenically stimulated lymph nodes in sheep, J. Exp. Med., 153, 706, 1981. 79. Wagner, C., Burger, D. R., and McColl, E., Antigen presentation and IL-1 production by HLA-DR compatible endothelial cells, Fed. Proc. Fed. Am. Soc. Exp. Biol., 42, 835, 1983. 80. Pober, J. S., Gimbrone, M. A., Cotran, R. S., Reiss, C. S., Burakoff, S. J., Fiers, W., and Ault, K. A., Ia expression by vascular endothelium is inducible by activated T cells and by human "y­interferon, J. Exp. Med., 157, 1339, 1983. 81. Kennedy, J. C., Till, J. E., Siminovitch, L., and McCulloch, E. A., The proliferative capacity of antigen-sensitive precursors of haemolytic plaque-forming cells, J. Immunol., 96, 973, 1966. 82. Armstrong, W. D. and Diener, E., A new method for enumeration of antigen reactive cells responsive to a purified protein antigen, J. Exp. Med., 129, 371, 1968. 83. Wilson, D. B., Blyth, J. L., and Nowell, P. C., Quantitative studies on the mixed lymphocyte interaction in rats. III. Kinetics of the response, J. Exp. Med., 135, 200, 1968. 84. Ford, W. L., Simmonds, S. J., and Atkins, R. C., Early cellular events in a systemic graft-versus-host reaction. II. Autoradiographic estimates of the frequency of donor lymphocytes which respond to each AgB-determined antigenic complex, J. Exp. Med., 141, 681, 1975. 85. Ford, W. L. and Atkins, R. C., Specific unresponsiveness of recirculating lymphocytes after exposure to histocompatability antigen in F, hybrid rats, Nature (London), 24, 178, 1971. 86. Drayson, M. T., The entry of lymphocytes into stimulated lymph nodes: the site of selection of alloantigenspecific cells, Transplantation, 41, 745, 1986. 87. Hay, J. B. and Hobbs, B. B., The flow of blood to lymph nodes and its relation to lymphocyte traffic and the immune response, J. Exp. Med., 145, 31, 1977. 88. Sedgley, M. and Ford, W. L., The migration of lymphocytes across specialized vascular endothelium. I. The entry of lymphocytes into the isolated mesenteric LN of the rat, Cell Tissue Kinet., 231, 1976. 89. Gowans, J. L. and Steer, H. W., The function and pathways of lymphocyte recirculation, in Blood Cells and Vessel Walls: Functional Interactions, CIBA Found. Symp. No. 71, O ’Connor, M ., Ed., CIBA Medical, West Caldwell, N.J., 1980, 113. 90. Drayson, M. T., unpublished observations.

Volume I

51

Chapter 4

MIGRATION OF CELLS FROM THE THYMUS TO THE SECONDARY LYMPHOID ORGANS Roland Scollay

TABLE OF CONTENTS I.

Introduction.......................................................................................................................52

II.

Background to Intrathymic Kinetics..............................................................................52

III.

Quantitative Aspects of Em igration..............................................................................52 A. Stress-Free A nalyses..........................................................................................52 B. Low -Stress A nalyses..........................................................................................53 C. High-Stress Analyses..........................................................................................53 D. Summary of Quantitative A spects.................................................................. 55 E. Emigration in the YoungAnimal....................................................................... 55

IV.

Qualitative Aspects of Em igration............................................................................... 55 A. Stress-Free A nalyses..........................................................................................56 B. High-Stress Analyses..........................................................................................57 C. Summary of Phenotypic Analyses.................................................................. 57

V.

Function in Recent Emigrants........................................................................................ 57

VI.

Conclusions.......................................................................................................................58

Acknowledgments....................................................................................................................... 59 References

59

52

Migration and Homing of Lymphoid Cells I. INTRODUCTION

The first important journey a lymphocyte undertakes is that which takes it from its tissue of origin to the secondary lymphoid organs. The tissue of origin for most, if not all, T cells is the thymus, and this review focuses on the emigration of cells from that site. As we will see, the subject is not without controversy, and both the rate of emigration and the relative maturity of the emigrants have been vigorously argued in the literature. Many approaches to the problem have involved relatively severe experimental manipulation such as surgery and, since the lymphoid system in general is rather stress sensitive, effects due to stress may be important. Indeed, it has been argued 1 that stress effects have led to considerable bias in many experiments and that this may account for the diversity of results obtained in different systems. In the following pages we will briefly review the literature, keeping in mind the question of stress, and see whether any consistent pattern can be found. There are several previous reviews of the subject of thymic emigration, 2 5 but none deals adequately with the questions of stress raised by Rocha , 1 and, indeed, these questions warrant serious consideration. There is, in addition, the problem that not all the studies in the literature are comparable, with rates of migration being expressed in many ways and with many parameters being estimated, guessed at, or extrapolated from other species. Primary data and experi­ mental details are sometimes (often) lacking. We have chosen to use percent of total thy ­ mocytes emigrating per day as the most straightforward measure, as this requires fewer assumptions, for example, about cell turnover in the thymus or the size of the peripheral pool. We have, therefore, attempted to convert all data to this form, but in the absence of raw data our conversions themselves have warranted some assumptions. Where these are particularly uncertain we will point them out. Finally, in the absence of adequate experimental data to the contrary, this author believes that a low migration rate of relatively mature cells is probably closest to the truth. It is with this bias that other data will be interpreted.

II. BACKGROUND TO INTRATHYMIC KINETICS One point which is widely accepted is the high rate of cell production in the thymus . 2' 4 This has been most carefully examined in the mouse , 6*8 where it seems clear that about 25 to 30% of thymocytes are lost and replaced each day. In the mouse, this amounts to approximately five times the blood pool of T cells and about 25% of the total T -cell pool. Limited data obtained in other species support this picture. The controversial point is what happens to this 25 to 30% of lost cells. Clearly, they do not all emigrate and accumulate in the periphery since many peripheral T cells are long lived and the numbers in the adult are quite stable :9 so where do they go? Do they die in the thymus or do they leave first and die elsewhere? Clearly, the maximum emigration would be 25 to 30% of thymocytes per day (i.e., the total production) in a steady-state situation. In the following section we will look at methods used to measure emigration.

III. QUANTITATIVE ASPECTS OF EMIGRATION A. Stress-Free Analyses These are limited to static analyses of normal animals, perhaps assessed at different ages. One method has been to look at sections of thymic tissue and assess the relative numbers of dividing and dying cells. However, a meaningful answer requires detailed knowledge of the lifespan of both mitotic and pyknotic cells and assumes that all dying or dead cells are visible. This may not be the case at the light-microscope level, but mechanisms exist for very rapid clearance of dead cells which are best visualized at the electron-microscope level. 10 Not surprisingly, interpretations of light-microscope analyses run the extremes from

Volume /

53

all or most cells dying in situ*11 to all or most cells leaving1214 and, thus, probably provide little useful information. They do, however, seem to represent the only real “ no stress approach” to quantitation.

B. Low-Stress Analyses This group includes experiments in which the experimental stress is limited to injections (isotopes, etc.) or bleedings. However, even in these cases significant increases in corti­ costeroid levels will follow handling. Perhaps the best approach has been the injection of two radiolabeled nucleoside analogs, 3H-TdR and 125I-UdR, into mice. These precursors are reutilized with differing degrees of efficiency so that local reutilization (i.e., cell death) can be distinguished from loss through emigration. All the studies of this kind concluded that quite high levels of cell death occurred in the thymus,1517 probably >70% of cells lost dying in situ. Assuming a 25% total turnover rate per day, this would mean >18% per day dying in situ and leave a maximum emigration rate of