Microbial Life of the Deep Biosphere 9783110300130, 9783110300093

Citation preview

Jens Kallmeyer, Dirk Wagner Microbial Life of the Deep Biosphere Life in Extreme Environments

Life in Extreme Environments

| Edited by Jens Kallmeyer, Dirk Wagner

Volume 1

Microbial Life of the Deep Biosphere |

Editors Dr. Jens Kallmeyer GFZ German Research Centre for Geosciences Section 4.5, Geomicrobiology Telegrafenberg 14473 Potsdam Germany [email protected] Prof. Dr. Dirk Wagner GFZ German Research Centre for Geosciences Section 4.5, Geomicrobiology Telegrafenberg 14473 Potsdam Germany [email protected]

ISSN 2197-9227 ISBN 978-3-11-030009-3 e-ISBN 978-3-11-030013-0 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. © 2014 Walter de Gruyter GmbH, Berlin/Boston Cover image: Thierry Berrod, Mona Lisa Production/Science Photo Library Typesetting: le-tex publishing services GmbH, Leipzig Printing and binding: Hubert & Co. GmbH & Co. KG, Göttingen ♾Printed on acid-free paper Printed in Germany www.degruyter.com

Preface “What are extreme environmental conditions?” Most of the answers that one gets from undergraduate students confirm a rather anthropocentric view: An environment that humans perceive as unpleasant is classified as extreme. De facto, many of Earth’s ecosystems are characterized by “extreme” environmental conditions, because they deviate from those conditions that humans would consider “normal” with regards to temperature, water availability, pressure, salinity, nutrient supply and so on. Despite being considered extreme, these habitats are colonized by a large number of organisms that thrive under the given conditions. The definition of an extreme habitat is based on our anthropocentric view, but as a more general approach, microorganisms can be considered extremophilic when they thrive under physical and chemical conditions that destroy cellular components of most nonextremophilic organisms. In recent times, more and more scientists from various disciplines have become interested in the topic ‘Life in Extreme Environments’. Through multidisciplinary research, completely new concepts were developed of how extremophiles can possibly survive and even thrive in extreme ecosystems. Based on these recent advances, the book series ‘Life in Extreme Environments’ publishes topical volumes in the rapidly growing research field of microbial life in extreme environments. This includes all habitats at the edge of survivability, ranging from equatorial to polar regions, from marine to terrestrial environments and from surface to deep ecosystems. Environmental niches that are, for instance, characterized by extraordinarily hot, cold, acidic, alkaline or dry conditions, or subjected to high salinity, radiation or pressure. The extremophilic microorganisms living in these environments represent numerous and diverse lineages from across all three domains of life: Bacteria, Archaea and Eukarya. Special emphasis is placed on the understanding of the structure and function of microbial communities in extreme environments, their life strategies and adaptation mechanisms as well as their reaction to changing environmental conditions. This book series will be a useful reference for advancing our understanding of the origin of life and for exploring the biotechnology potential of these fascinating microorganisms. The first volume of this series presents a broad overview of our current knowledge of microbial life in deep subsurface environments. Over the last decade, this so-called deep biosphere research has expanded quite dramatically. Since the early days of Morita and Zobell (1955), who set the limit of life at 7.47 meters below the sea floor, the maximum depth to which life reaches into the Earth has been set deeper and deeper and is now exceeding 3 km on land and 1.5 km in marine sediments. Active microbial communities were found in areas that were considered devoid of life, for example, the oceanic crust. Still, we have not seen the true limits of life yet. Despite major technical advances in the last few years, subsurface life exploration is still heavily de-

vi | Preface pending on technical improvements because quite often the abundance and activity of subsurface microbes is orders of magnitude lower than in surface sediments. When compiling this volume, we wanted to cover the diversity of this young but rapidly growing field of research. The first section is devoted to the different major habitats. The chapter of Parke𝑠 et al. provides us with a much-awaited update of their review paper about subseafloor sediment microbiology from 2000. There is hardly any publication about deep subseafloor sediment that does not cite this classic. This chapter allows us to follow the development of the field from a small niche subject into an important research field. Microbial life in subseafloor environments is not restricted to sediments, there are new and exciting findings of life in the oceanic crust, and these are presented by Biddle et al. Of course, research in terrestrial subsurface environments has made a similar leap forwards, and Karsten Pedersen provides us with an update on the state-of-the-art. When searching for life elsewhere in the Universe, one should need to know what to look for. Charles Cockell argues that Earth’s subsurface might be a good analogue for habitats on, or rather, in other planetary bodies. After these more general chapters, the book focuses on several special topics that are currently under much debate. Andreas Teske’s chapter about archaea in deep marine subsurface sediments gives an overview of the current knowledge about this largely uncultivated group of organisms. Although modern molecular techniques have become increasingly popular in recent years, classical cultivation still is very important. Toffin and Alain summarize recent advances in this field with specific remarks about high-pressure cultivation, and other high-tech methods that allow us to grow microbes that would otherwise remain uncultured. Not only microbes, i.e. prokaryotes live in the subsurface. Eukaryotes are also present and might play a much more important role than previously thought. Edgcomb et al. inform us about the current state of knowledge in this area. One of the main drawbacks of molecular techniques is the apparent disconnection between phylogenetic and metabolic information. Only through novel techniques that allow measuring multiple information simultaneously from the same sample can we now actually see which microbe is doing what. Morono et al. present us their recent advances in NanoSIMS research and the challenges that are still lying ahead. This part of the book is closed by a chapter of Karen Lloyd who shows us how just minor differences in sample preparation can have huge impacts on the final results. This should be a note of caution to everybody working in this field and a friendly reminder that there are still many technical challenges ahead of us. The subsurface biosphere is not just of purely scientific interest. As many geotechnological applications are affected by subsurface microbial activity, there is also a growing industrial interest in this field of research. Ollivier et al. introduce us to microbial activity in hydrocarbon reservoirs. Of course, not only hydrocarbons are affected

Preface |

vii

by microbes, Alawi shows us their effects on hydrothermal systems and subsurface storage of carbon dioxide. Even with the most sensitive techniques, metabolic activity might be so low that it cannot be detected and many turnover processes occur over geologic time scales, vastly exceeding the timespan that humans can observe. While LaRowe and Amend focus on thermodynamical controls on subsurface life, DiPrimio introduces basin modeling as a valuable tool to understanding ultraslow abiotic reactions that run over geologic time scales. Røy shows us how to use actual measurements of downcore profiles to quantify metabolic rates. We hope that this volume will provide you with a broad overview of this exciting and rapidly developing field of research and stimulates the debate on this fascinating research field in the near future. March 2014

Dirk Wagner & Jens Kallmeyer

Contents Preface | v Contributing authors | xv R. John Parkes, Henrik Sass, Barry Cragg, Gordon Webster, Erwan Roussel, and Andrew Weightman 1 Studies on prokaryotic populations and processes in subseafloor sediments – an update | 1 1.1 New sites investigated | 1 1.1.1 Southeast Atlantic sector of the Southern Ocean (Leg 177) | 1 1.1.2 Woodlark Basin, near Papua New Guinea, Pacific Ocean (Leg 180) | 4 1.1.3 Leg 185, Site 1149 in the Izu-Bonin Trench, Western Equatorial Pacific | 6 1.1.4 Nankai Trough (Leg 190), subduction zone/accretionary prism, Pacific Ocean | 7 1.1.5 Eastern Equatorial Pacific and Peru Margin Sites 1225–1231 (Leg 201) | 10 1.1.6 Newfoundland Margin (Leg 210) | 12 1.1.7 Carbonate mound (IODP Expedition 307) | 13 1.2 High-pressure cultivation – DeepIsoBUG, gas hydrate sediments | 15 1.3 Subseafloor biosphere simulation experiments | 18 1.4 Conclusions | 20 Jennifer F. Biddle, Sean P. Jungbluth, Mark A. Lever, and Michael S. Rappé 2 Life in the Oceanic Crust | 29 2.1 Introduction | 29 2.2 Sampling tools | 30 2.2.1 Tools for accessing the deep basement biosphere | 32 2.3 Contamination | 36 2.3.1 Contamination induced during drilling | 36 2.3.2 Contamination during fluid sampling | 38 2.4 Direct evidence for life in the deep ocean crust | 38 2.4.1 Textural alterations | 39 2.4.2 Geochemical evidence from fluids | 40 2.4.3 Geochemical evidence from rocks | 41 2.4.4 Genetic surveys | 45 2.5 Future directions | 51

x | Contents Karsten Pedersen 3 Microbial life in terrestrial hard rock environments | 63 3.1 Hard rock aquifers from the perspective of microorganisms | 63 3.2 Windows into the terrestrial hard rock biosphere | 64 3.2.1 Sampling methods for microbes in hard rock aquifers | 64 3.2.2 Yesterday marine – terrestrial today | 65 3.2.3 Basalts and ophiolites | 66 3.2.4 Granites | 68 3.2.5 Hard rocks of varying origin | 70 3.3 Energy from where? | 71 3.3.1 Deep reduced gases | 72 3.4 Activity | 73 3.4.1 Stable isotopes | 73 3.4.2 Geochemical indicators | 74 3.4.3 In vitro activity | 74 3.4.4 In situ activity | 74 3.4.5 Phages may control activity rates | 76 3.5 What’s next in the exploration of microbial life in deep hard rock aquifers? | 76 Laurent Toffin, Karine Alain 4 Technological state of the art and challenges | 83 4.1 Basic concepts and difficulties inherent to the cultivation of subseafloor prokaryotes | 83 4.2 Microbial growth monitoring, method detection limits and innovative cultivation methods | 91 4.3 Challenges and research needs (instrumental, methodological and logistics needs) | 92 Yuki Morono, Motoo Ito, and Fumio Inagaki 5 Detecting slow metabolism in the subseafloor: analysis of single cells using NanoSIMS | 101 5.1 Introduction | 101 5.2 Overview of ion imaging with a NanoSIMS ion microprobe | 102 5.3 Detecting slow metabolism: bulk to single cells | 105 5.3.1 Bulk measurement of subseafloor microbial activity using radiotracers | 105 5.3.2 Observing radioactive substrate incorporation at the cellular level: microautoradiography | 106 5.3.3 Quantitative analysis of stable isotope incorporation using NanoSIMS | 107

Contents | xi

5.4 5.5 5.6

Bridging identification and functional analysis of microbes using elemental labeling | 110 Critical step for successful NanoSIMS analysis: sample preparation | 112 Future directions | 114

Karen G. Lloyd 6 Quantifying microbes in the marine subseafloor: some notes of caution | 121 6.1 Introduction | 121 6.2 Quantification of specific microbial groups in marine sediments | 124 6.3 Assessment of quantitative methods in marine sediments: the Leg 201 Peru Margin example | 128 6.4 Global meta-analysis of FISH, CARD-FISH and qPCR quantifications of bacteria and archaea | 132 6.5 Future outlook | 134 Andreas Teske 7 Archaea in deep marine subsurface sediments | 143 7.1 Introduction | 143 7.2 Archaeal Ribosomal RNA phylogeny | 143 7.3 Marine subsurface Archaea | 144 7.4 Archaeal habitat preferences in the subsurface | 149 7.5 Methanogenic and methane-oxidizing archaea | 152 7.6 Archaeal abundance and ecosystem significance in the subsurface | 154 Bernard Ollivier, Jean Borgomano, and Philippe Oger 8 Petroleum: from formation to microbiology | 161 8.1 Introduction | 161 8.2 Petroleum formation | 161 8.2.1 Petroleum system | 163 8.3 Petroleum microbiology | 166 8.3.1 The sulfate-reducing prokaryotes | 168 8.3.2 The methanoarchaea | 171 8.3.3 The fermentative prokaryotes | 174 8.3.4 Other metabolic lifestyle bacteria | 177 8.4 Conclusion | 179

xii | Contents Virginia Edgcomb, William Orsi, and Jennifer F. Biddle 9 Fungi in the marine subsurface | 187 9.1 Introduction | 187 9.2 The concept of marine fungi | 187 9.3 Fungi in marine near-surface sediments in the deep sea | 189 9.4 Fungi in the deep subsurface | 190 9.4.1 Initial whole community and prokaryote-focused studies of the marine subsurface yielding information on eukaryotes | 190 9.4.2 Eukaryote-focused studies yielding information on fungi in the deep subsurface | 191 9.5 How deep do fungi go in the subsurface? | 197 9.6 Summary | 197 Mashal Alawi 10 Microbes in geo-engineered systems: geomicrobiological aspects of CCS and Geothermal Energy Generation | 203 10.1 Introduction | 203 10.1.1 Carbon Capture and Storage (CCS) | 204 10.1.2 Geothermal energy and aquifer energy storage | 205 10.2 Microbial diversity in geo-engineered reservoirs | 206 10.3 Interactions between microbes and geo-engineered systems | 208 10.3.1 General considerations | 208 10.3.2 Microbial processes in the deep biosphere potentially affected by CCS | 209 10.3.3 Examples from a CCS pilot site, CO2 degasing sites and laboratory experiments | 211 10.3.4 Impact of microbially-driven processes on CO2 trapping mechanisms | 213 10.3.5 Impact of microbially-driven processes on CCS facilities | 214 10.3.6 Impact of microbially-driven processes on geothermal energy plants | 214 10.4 Methods to analyze the interaction between geo-engineered systems and the deep biosphere | 216 10.4.1 Sampling of reservoir fluids and rock cores | 216 10.4.2 Methods to analyze microbes in geo-engineered systems | 216 Charles S. Cockell 11 The subsurface habitability of terrestrial rocky planets: Mars | 225 11.1 Introduction | 225 11.2 The subsurface of Mars – our current knowledge | 226 11.3 Martian subsurface habitability, past and present | 233 11.3.1 Vital elements (C, H, N, O, P, S) | 233

Contents | xiii

11.3.2 11.3.3 11.3.4 11.3.5 11.3.6 11.3.7 11.4 11.5 11.6 11.7 11.8 11.9

Other micronutrients and trace elements | 234 Liquid water through time | 235 Redox couples | 238 Radiation | 239 Other physical and environmental factors | 239 Acidity | 240 Impact craters and deep subsurface habitability | 242 The near-subsurface habitability of present and recent Mars – an empirical example | 243 Uninhabited, but habitable subsurface environments? | 245 Ten testable hypotheses on habitability of the Martian subsurface | 247 Sampling the subsurface of Mars | 250 Conclusion | 251

Rolando di Primio 12 Assessing biosphere-geosphere interactions over geologic time scales: insights from Basin Modeling | 261 12.1 Introduction | 261 12.2 Basin Modeling | 262 12.3 Modeling processes at the deep bio-geo interface | 264 12.3.1 Feeding the deep biosphere (biogenic gas) | 264 12.3.2 Petroleum biodegradation | 267 12.4 Modeling processes at the shallow bio-geo interface | 274 12.5 Conclusions | 275 Doug LaRowe, Jan Amend 13 Energetic constraints on life in marine deep sediments | 279 13.1 Introduction | 279 13.2 Previous work | 280 13.3 Study site overview | 280 13.3.1 Juan de Fuca (JdF) | 281 13.3.2 Peru Margin (PM) | 281 13.3.3 South Pacific Gyre (SPG) | 282 13.4 Overview of catabolic potential | 282 13.5 Comparing deep biospheres | 288 13.6 Electron acceptor utilization | 290 13.7 Energy demand | 292 13.8 Concluding remarks | 293 13.9 Computational methods | 293 13.9.1 Thermodynamic properties of anhydrous ferrihydrite and pyrolusite | 294

xiv | Contents Hans Røy 14 Experimental assessment of community metabolism in the subsurface | 303 14.1 Introduction | 303 14.1.1 The energy source | 303 14.1.2 The carbon budget | 304 14.1.3 Distribution vertical of microbial metabolism the sediment pile | 305 14.2 Quantifiable metabolic processes | 306 14.2.1 Reaction diffusion modeling and mass balances | 307 14.2.2 Measurements of rates of energy metabolism with exotic isotopes | 312 14.3 Summary | 315 Index | 319

Contributing authors Karine Alain Laboratoire de Microbiologie des Environnements Extrêmes Institut Universitaire Européen de la Mer Technopôle Brest-Iroise Plouzané, France e-mail: [email protected] Chapter 4

Charles Cockell Centre for Astrobiology School of Physics & Astronomy Centre for Astrobiology University of Edinburgh Edinburgh, UK e-mail: [email protected] Chapter 11

Mashal Alawi GFZ German Research Centre for Geosciences Section 4.5 Geomicrobiology Potsdam, Germany e-mail: [email protected] Chapter 10

Barry Cragg School of Earth & Ocean Sciences Cardiff University Cardiff, UK e-mail: [email protected] Chapter 1

Jan Amend Department of Earth Sciences Department of Biological Sciences University of Southern California Los Angeles, CA, USA e-mail: [email protected] Chapter 13

Rolando DiPrimio GFZ German Research Centre for Geosciences Section 4.3 Organic Geochemistry Potsdam, Germany e-mail: [email protected] Chapter 12

Jennifer Biddle College of Earth, Ocean and the Environment University of Delaware Lewes, DE, USA e-mail: [email protected] Chapter 2, 9 Jean Borgomano Total CSTJF EP/ EXPLO/ TE/ ISS/ CARB Pau, France e-mail: [email protected] Chapter 8

Virginia Edgcomb Department of Geology and Geophysics Woods Hole Oceanographic Institution Woods Hole, MA, USA e-mail: [email protected] Chapter 9 Fumio Inagaki Geomicrobiology Group Kochi Institute for Core Sample Research JAMSTEC Nankoku, Kochi, Japan and Geobio-Engineering and Technology Group Submarine Resources Research Project JAMSTEC Yokosuka, Japan e-mail: [email protected] Chapter 5

xvi | Contributing authors Motoo Ito Geochemical Research Group Kochi Institute for Core Sample Research JAMSTEC Nankoku, Kochi , Japan and Geobio-Engineering and Technology Group Submarine Resources Research Project JAMSTEC Yokosuka, Japan e-mail: [email protected] Chapter 5 Sean Jungbluth Center for Microbial Oceanography Hawaii Institute of Marine Biology University of Hawaii at Manoa Honululu, HI, USA e-mail: [email protected] Chapter 2 Douglas LaRowe Department of Biological Sciences Department of Earth Sciences University of Southern California Los Angeles, CA, USA e-mail: [email protected] Chapter 13 Mark Lever Department of Bioscience Center for Geomicrobiology Aarhus University Aarhus, Denmark e-mail: [email protected] Chapter 2 Karen G. Lloyd Department Microbiolgy University of Tennessee Knoxville, TN, USA e-mail: [email protected] Chapter 6

Yuki Morono Geomicrobiology Group Kochi Institute for Core Sample Research, Japan JAMSTEC Nankoku, Kochi, Japan and Geobio-Engineering and Technology Group Submarine Resources Research Project JAMSTEC Yokosuka, Japan e-mail: [email protected] Chapter 5 Philippe Oger Laboratoire de Geologie Ecole Normale Supérieur de Lyon Lyon Cedex, France e-mail: [email protected] Chapter 8 Bernard Ollivier Laboratoire de Microbiologie Institute of Oceanography Aix-Marseille Université Marseille Cedex, France e-mail: [email protected] Chapter 8 William Orsi Department of Chemistry and Geochemistry Woods Hole Oceanographic Institution Woods Hole, MA, USA e-mail: [email protected] Chapter 9 R. John Parkes School of Earth & Ocean Sciences Cardiff University Cardiff, UK e-mail: [email protected] Chapter 1

Contributing authors

Karsten Pedersen Department of Civil and Environment Engineering Chalmers University of Technology Göteborg, Sweden and Microbial Analytics Sweden AB Mölnycke, Sweden e-mail: [email protected] Chapter 3 Michael Rappe Center for Microbial Oceanography Hawaii Institute of Marine Biology University of Hawaii at Manoa Honululu, HI, USA e-mail: [email protected] Chapter 2 Erwan Roussel School of Earth & Ocean Sciences Cardiff University Cardiff, UK e-mail: [email protected] Chapter 1 Hans Røy Department of Bioscience Center for Geomicrobiology Aarhus, Denmark e-mail: [email protected] Chapter 14

| xvii

Henrik Sass School of Earth & Ocean Sciences Cardiff University Cardiff, UK e-mail: [email protected] Chapter 1 Andreas P. Teske Department of Marine Sciences University of North Carolina Chapel Hill, NC, USA e-mail: [email protected] Chapter 7 Laurent Toffin Laboratoire de Microbiologie des Environments Extremes, IFREMER Technopole Brest-Iroise Plouzané, France e-mail: [email protected] Chapter 4 Gordon Webster School of Biosciences Cardiff University, Main Building, Park Place, Cardiff, Wales CF10 3AT, UK e-mail: [email protected] Chapter 1 Andrew Weightman School of Biosciences Cardiff University, Main Building, Park Place, Cardiff, Wales CF10 3AT, UK e-mail: [email protected] Chapter 1

R. John Parkes, Henrik Sass, Barry Cragg, Gordon Webster, Erwan Roussel, and Andrew Weightman

1 Studies on prokaryotic populations and processes in subseafloor sediments – an update This chapter provides an update of a year 2000 review of the microbiology of subseafloor sediments [1]. At the time of this review, our Geomicrobiology Group was the main group researching in this area and had been the first to propose the subseafloor biosphere [2]. At this time, the presence of a significant prokaryotic biosphere in subseafloor sediments was still contentious due to perceived low-energy supply coupled with geological time scales, resulting in the view that most microorganisms in subseafloor sediments were either inactive or adapted for extraordinarily low metabolic activity [3]. However, as predicted [2], most cells were subsequently shown to be active [4, 5]. Since the year 2000, a significant number of additional research groups have been investigating the microbiology of subseafloor sediments (> 10, e.g. [5–18]) and they have confirmed our results of the presence of a globally significant subseafloor biosphere. Here, we provide an update of our recent deep biosphere research (7 new sites), including simulation experiments, and place these into a broader context of subseafloor biosphere research. Two aspects which need to be noted at the start of the update are: (1) The general depth trend in intact prokaryotic cells in subseafloor sediments which refers to the update of our original depth plots of acridine orange stained cells [2] which was modified for the 2000 review and (2) The organic acid acetate, which is an important anaerobic metabolic breakdown intermediate of organic matter, as well as a product of H2 /CO2 metabolism, via acetogenesis, and changes in concentration or metabolism of porewater acetate is used as an index for general prokaryotic activity.

1.1 New sites investigated 1.1.1 Southeast Atlantic sector of the Southern Ocean (Leg 177) Ocean Drilling Program (ODP) Leg 177 provided an opportunity to investigate prokaryotic distributions in carbonate-rich, low organic carbon Sites (1088 & 1093) in the Southeast Atlantic sector of the Southern Ocean (󳶳 Fig. 1.1), and to contrast these with porewater acetate concentrations [19]. Calcium carbonate concentrations at the nannofossil ooze Site 1088 (water depth 2082 m and sediment surface temperature ~2.4 °C) was high (88.2 wt%), and ~10 times higher than the deeper water, diatom ooze Site

2 | 1 A subseafloor biosphere update 1093 (water depth 3636 m and sediment surface temperature ~2.6 °C). Prokaryotic celldepth distributions at both sites were lower than the general trend in deep marine sediments (󳶳 Fig. 1.2), but Site 1088 had the lowest cell numbers despite the much shallower water depth. It seems at these sites that the high calcium carbonate content, and hence, low organic carbon, had a greater effect on prokaryotic cell numbers (decreasing) than water column depth or latitude. This was also reflected in porewater acetate concentrations with Site 1088 having consistently low concentrations (0–15 μM), compared to acetate peaks of up to 110 μM at Site 1093, associated with localized diatom rich laminae (󳶳 Fig. 1.2). Geochemical data at both sites also demonstrated low levels

30° S

526

AFRICA Capetown

359

361

Cape Basin

Subtropical Front

40°

360

1089 Ag

1092 1091

Polar Front

Agulhas Plateau

1088

1090 Subantarctic Front

Dec.14,1997

a ulh

sR

idg

e Agulhas Basin

Met eo 704 r Rise

1093

50° Punta Arenas Feb. 6, 1998

1094

Southwest Indian Ridge 703

Bouvet I.

inter sea

w average

Wed d ACC ell Gyre / -Bou ndar y

ice edge

60° 10°W

0°

10°E

20°

Fig. 1.1: Southern Ocean ODP Leg 177 Sites, including Sites 1088 and 1093.

30°

1.1 New sites investigated | 3

(a)

(b)

0

20

Depth (mbsf)

40

60

80

100

120 4.5

5.5

6.5

7.5

8.5

0

9.5

20

Total bacterial populations Log (cells/cm3)

(c)

40

60

80

100

120

100

120

Acetate (µM)

(d)

0

50

Depth (mbsf)

100

150

200

250

300

350 4.5

5.5

6.5

7.5

8.5

Total bacterial population Log (cells/cm3)

9.5

0

20

40

60

80

Acetate (µM)

Fig. 1.2: Depth distributions of total bacterial populations and porewater acetate, Site 1088 (a,b) and 1093 (c,d) Southeast Atlantic sector of the Southern Ocean. Gray shaded area = data below the detection limit of the technique (2.23 × 105 cells/cm3 ).

4 | 1 A subseafloor biosphere update of prokaryotic activity with only limited sulfate removal and low methane concentrations [20]. As carbonate-rich sediments account for ~52% of global seafloor area [21], if prokaryotic cell numbers are consistently lower in carbonate-dominated sediments compared to other sediment types, this would reduce estimates of the total biomass of the subseafloor biosphere.

1.1.2 Woodlark Basin, near Papua New Guinea, Pacific Ocean (Leg 180) Three sites were sampled at water depths from 1150 to 2303 m [22]. Two sites (1109 and 1115) had the global average thermal gradients of ~30 °C/km and were low organic carbon (~0.4%) and low organic matter sedimentation rate sites. Active prokaryotic populations (microscopic cells, culturable prokayotes [anaerobic fermentative heterotrophs, autotrophic and heterotrophic acetogens] and radiotracer activities [sulfate reduction, methanogenesis from acetate and H2 /CO2 , growth – thymidine incorporation into DNA] and geochemistry) were present to all depths sampled at these sites, maximum 801 meters below seafloor (mbsf), and ~15 million years ago (mya). In 2002, these were the deepest subseafloor sediments that the presence of prokaryotes had been detected by a range of complementary methods. Prokaryotic populations and activities were greatest near the sediment surface and decreased with increasing depth, although there were some limited subsurface peaks (󳶳 Fig. 1.3). Consistent with the presence of active prokaryotic populations in deeper layers, there were continuing geochemical changes (porewater sulfate removal and subsequently, methane formation) and corresponding low activity rates (up to 10,000 times lower than near-surface rates). Interestingly, however, depth integration of measurements on the full sediment depth showed the biogeochemical significance of the deeper layers, with 78% of cells, 93% of cell production, and ~90% of prokaryotic activity (methanogenesis and acetate oxidation) occurring in sediments below 20 m. The depth distribution of sulfate reduction activity, in contrast, depended on the rates that occurred, with the higher rates at Site 1109 more rapidly removing sulfate, and thus, restricting most activity to the upper 20 m (65%). Whilst at Site 1115 with lower sulfate reduction rates, sulfate penetrated deeper and sediments below 20 m were responsible for the majority of measured sulfate reduction activity (72%). Cell counts and geochemical data alone measured at the deepest water depth site (1118, 2303 m) also provided strong evidence for significant prokaryotic populations to at least 842 mbsf (󳶳 Fig. 1.4). In addition, there was circumstantial evidence for deep anaerobic oxidation of methane (AOM) providing a new energy source, as fluid flow at depth (~700 mbsf) provided sulfate, and this coincided with removal of methane that had been consistently present from ~240 mbsf. Although at this site there was not an increase in prokaryotic cell numbers due to stimulation of AOM, this has occurred in other deep sediments (e.g. [14, 23, 24]). Also at this site, which had a higher

1.1 New sites investigated |

5

Fig. 1.3: Depth profiles of prokaryotic populations and activities in Woodlark Basin sediments (a) Site1109, (b) Site 1115. (a) Total (O ) and dividing cells (○). The solid lines are Parkes’ general model for cell distributions in marine sediments [1], and dotted lines represent 95%prediction limits. (b) Culturable cells from MPN enrichments; heterotrophic (○) and autotrophic acetogens, (O ) and fermentative heterotrophs (◻). (c) Sulfate reduction (O ) and porewater sulfate (dashed line). (d) Methanogenesis from H2 :CO2 (O ) and in situ methane(dashed line). (e) Acetate metabolism to CO2 (O ) and CH4 (Q ) and porewater acetate (+); (f) thymidine incorporation-rate into DNA (O ). Hollow symbols denote zero values. For Site 1109 //// represents a dolerite layer.

thermal gradient (~63 °C km−1 ), there were peaks in acetate concentrations at depth not present at the lower temperature sites. This could reflect temperature activation of recalcitrant organic matter [25, 26], with acetate accumulation being restricted by acetoclastic sulfate reduction. At the other sites, acetate oxidation was directly measured (󳶳 Fig. 1.3), but low acetate concentrations were consistently present, which demonstrates that acetate was also being produced at depth at these sites. Deep acetate for-

6 | 1 A subseafloor biosphere update

Fig. 1.4: Depth profiles of bacterial populations and activities in sediments at Site 1118 in Woodlark Basin. (a) Total bacterial populations (○) and dividing and divided cells (O ). The solid line shows Parkes’ general model for bacterial distributions in marine sediments [1], and dotted lines represent 95% prediction limits. (b) Porewater sulfate (♦). (c) In situ methane (⬦). (d) Porewater acetate (+).

mation in ~15 mya sediments may seem surprising, but this has been observed in other deep subsurface environments, including Cretaceous age sediments [27] and was consistent with the presence of viable acetogens (󳶳 Fig. 1.3). Low molecular weight hydrocarbons (LMWH) were also detected at sites 1109 and 1115, and their downhole profiles combined with low in situ temperatures suggested that the LMWH components were formed in situ by low-temperature biological processes [28].

1.1.3 Leg 185, Site 1149 in the Izu-Bonin Trench, Western Equatorial Pacific ODP Leg 185 was the first ODP cruise where contamination checks were conducted for microbiology [29, 30]. These tests demonstrated that the inner portion of cores, where the microbiological samples were taken from, were free from any potential sampling contamination. Bacterial populations were present in all samples (deepest at 171.2 mbsf) at this deep-water (5818 m) low-sedimentation-rate site. The highest cell numbers were near the surface (1.4 mbsf; 7.2×106 cells/cm3 ), but then declined rapidly within the upper 10 mbsf. Below this, numbers decreased at a more gradual rate to 7.2 × 105 cells/cm3 at 172 mbsf, a 10-fold reduction. This two-stage bacterial depth distribution has been observed at several other ODP sites (e.g. Amazon Fan [31] and Santa Barbara Basin [32]). Bacterial depth distributions at Site 1149 were well below

7

1.1 New sites investigated |

(a)

(c)

(b)

(d)

(e)

(f)

0 20 40

Depth (mbsf)

60 80 100 120 140 160 180

6

8

10

Log10 cells/cm3

24

26

28

Sulfate (mM)

30 0

4

8

Methane (ppmv)

0

100

200 0

Ammonium (µM)

100

IW Acetate (µM)

200 0

200

400

600

IW Manganese (µM)

Fig. 1.5: Depth profiles at Site 1149, Izu-Bonin Trench, Western Equatorial Pacific. (a) Total prokaryotic cells. Solid sloping line is the regression line of best fit derived from previous ODP legs, dashed lines are the 95% prediction limits [2] (b) Sulfate. (c) Methane. (d) Ammonium. (e) Acetate. (f) Reduced manganese. The shaded area highlights the broad peak in bacterial manganese reduction activity between 26 and 100 mbsf.

those for other subseafloor locations and were predominantly below the lower 95% prediction limits (󳶳 Fig. 1.5). These low bacterial populations probably reflect the low sedimentation rates and low input of bioavailable organic matter that is characteristic for deep-water sites. Consistent with the low cell numbers, there was only limited removal of porewater sulfate, suggesting low bacterial sulfate reduction activity. Most sulfate removal was in the top ~5 mbsf, coinciding with the highest bacterial populations, the presence of small amounts of methane and an increase in porewater manganese and ammonia. In the deeper sediments, however, there was still indirect evidence of continuing low prokaryotic activity, with increases in porewater ammonia, soluble manganese (approx 26 to 100 mbsf), bioavailable acetate and decreasing sulfate. Unexpectedly, manganese reduction, sulfate reduction and a limited amount of methanogenesis seemed to be occurring simultaneously at depth in this low organic matter site, rather than in the expected depth succession. Similar situations were subsequently shown at other deep sediment sites (e.g. [33, 34]).

1.1.4 Nankai Trough (Leg 190), subduction zone/accretionary prism, Pacific Ocean Nankai Trough is a deep trench formed at a subducting plate boundary where there is also active sediment accretion producing a large accretionary prism [35, 36]. Three deep-water sites were analyzed (4751–4844 m), which had relatively low organic carbon concentrations (mean 0.35–0.45% w/w) but steep temperature gradients (base-

8 | 1 A subseafloor biosphere update ment temperatures at Sites 1173 and 1174 were above 100 °C and at Site 1177 were < 70 °C). Depth distribution of prokaryotic cell numbers at Site 1177 and above about 400–500 mbsf at Sites 1173–1174 were similar to other subseafloor sediment sites, but deeper samples at Sites 1173–1174 were very low (< 105 cells cm−3 ). It was, therefore, surprising that amplifiable DNA could not be extracted from Sites 1177 or 1174. However, amplifiable DNA was obtained at three upper depths from Site 1173 (4.15, 98.29 and 193.29 mbsf). Low, but active, prokaryotic populations at these sites was supported by measured rates of methanogenesis and, for the first time, the presence of intact phospholipids (󳶳 Fig. 1.6), which are chemical markers for living prokaryotes [37]. Phylogenetic analysis of the extracted DNA sequences showed a wide variety of uncultured Bacteria and Archaea [35]. Sequences of Bacteria were dominated by an uncultured and deeply branching “deep sediment group” (now called JS1 [38], 53% of sequences). Also present were Planctomycetes (4%), Cyanobacteria and chloroplasts (8%), Betaproteobacteria (11%) and Gammaproteobacteria (14%). The majority of archaeal 16S rRNA gene sequences belonged to uncultured clades of the Crenarchaeota. There was good agreement between sequences obtained independently by cloning and by denaturing gradient gel electrophoresis (DGGE). Nankai Trough sequences were similar to those detected in other marine sediments and anoxic habitats, and so probably represent environmentally important indigenous bacteria. Kinetic analysis of sediment heating experiments to assess hydrocarbon generation in Nankai Trough sediments [36] predicted that organic matter transformation would start at Site 1173 around 300 mbsf and this was in good agreement with in situ thermogenic hydrocarbon formation (e.g. ethane, 󳶳 Fig. 1.6). In addition, below ~400 mbsf there was an increase in rates of methanogenesis, some increases in cell numbers and detection of intact phospholipids. Similar changes occurred at Site 1174, but at depths greater than ~500 mbsf and the increase in cells was more marked. Also corresponding with the predicted increased organic matter reactivity with increasing temperature were increases in porewater acetate and hydrogen (󳶳 Fig. 1.6), which are both important substrates for anaerobic prokaryotes. Overall, however, H2 /CO2 methanogenesis was the dominant methanogenic process in these sediments, whilst acetate and methanol were also important substrates in some samples. Analysis of a functional methanogen (mcrA) gene at Site 1173 showed that both the 4.15 and 193.29 mbsf samples were dominated by Metanobacteriales methanogens, capable of H2 /CO2 methanogenesis, whereas at 98.29 mbsf Methanosarcinales methanogens, which can utilize acetate or methylated compounds, were the dominant sequences. These results show that in deep, sub-surface sediments, thermal activation of buried organic matter can release low molecular weight substrates which can stimulate prokaryotic activity, as suggested from laboratory experiments (e.g. [25, 26], and below). These experiments also showed sulfate production at elevated temperature and this occurred in Nankai Trough subsurface sediments (󳶳 Fig. 1.6 and [39]), and could further stimulate deep prokaryotic activity. In addition, Nankai Trough results clearly demonstrate overlap and interaction between biogenic and thermogenic processes in

1.1 New sites investigated

| 9

Fig. 1.6: Nankai Trough Site 1173 geomicrobiology and biogeochemistry summary. (a) Generation curves from kinetic modeling and experimentally determined rates of potential methanogenesis [36]. (b) Gas concentrations in ppm [71] for methane (diamonds) and ethane (circles), and total cell counts in log10 cm-3 (triangles), light arrows mark depths where intact phospholipids (PL) were detected [37]. (c) Increase in bacterial metabolites with temperature [26].

10 | 1 A subseafloor biosphere update deep, subseafloor sediments (󳶳 Fig. 1.6), which may have important consequences for our understanding of fossil fuel formation [40], and sustain the deep biosphere up to its upper temperature limit (122 °C, [41]).

1.1.5 Eastern Equatorial Pacific and Peru Margin Sites 1225–1231 (Leg 201) Leg 201 was the first dedicated “Deep Biosphere” Drilling Leg (27 January–29 March 2002) [4, 5, 24, 33, 42]. However, active deep bacteria had been detected at several of these sites on previous drilling Legs [43, 44] and repeat sampling would provide unique information about the consistency of deep biosphere populations, as well as more detailed information about these populations. At some sites, the complete sediment column was sampled plus the upper most part of the basaltic basement (1225, 1226, 1231) and prokaryotic cells were present at all sediment depths, although cells were not clearly stimulated at the sediment-basement interface despite evidence for fluid flow through this interface [33]. As previously found for other deep sediments [1], cell populations increased as water column depth decreased, presumably due to higher organic matter quantity and quality at shallow water sites (󳶳 Fig. 1.7). The only exception was the deep-water gas hydrate Site 1230, which had higher cell numbers than other deep-water sites. However, it has been previously shown that gas hydrate containing deep sediments can be particularly biogeochemically active [1]. This water depth trend also strongly suggests that the majority of cells are active, and not dead or dormant cells being buried, as had been previously suggested [3]. This was confirmed at some of these sites by detection of ribosomal RNA in cells (CARDFISH) and by real-time polymerase chain reaction quantification of 16S rRNA genes (qPCR, [4]). Interestingly, the qPCR results indicated that Bacteria rather than Archaea were the dominant prokaryotes within these sediments. 16S rRNA gene libraries and DGGE analysis of Site 1229 Peru Margin sediments, which had a deep brine incursion, and hence, unusually, a deep methane-sulfate interface (~90 mbsf), in addition to the more normal sulfate-methane interface (~30 mbsf), showed marked changes in bacterial diversity and increases in total cells at these interfaces (󳶳 Fig. 1.7). However, changes in archaeal diversity were limited [24]. This further suggests that Bacteria are the major active prokaryotes in these subsurface sediments, with clear activity and diversity changes over geological time scales (e.g. 90 mbsf equals ~0.8 Myr). The dominant Bacteria were Gammaproteobacteria at 6.7 and 86.67 mbsf and Chloroflexi at 30.2 and 42.03 mbsf, with the common subseafloor biosphere phylum JS1 [38] being a minor component. Methanogenic Archaea, however, were detected in both 16S rRNA gene libraries (42.03 mbsf in the methane zone) and by methanogen-specific genes (mcrA, all 4 depths, 6.7, 30.2, 42.03, 86.67 mbsf). This was consistent with measured low rates of active methanogenesis from both H2 /CO2 and acetate.

1.1 New sites investigated |

11

Fig. 1.7: Eastern Equatorial Pacific and Peru Margin Sites, Leg 201. Total cell numbers compared to cell depth profiles at other sites [1]. Cell populations increase as water column depth decreases, except for the deep-water gas hydrate Site 1230. Subsurface increases in cell numbers are highlighted by shaded areas in Sites 1226 and 1229.

Similar stimulation of prokaryotes occurred at an open ocean Site (1226, 󳶳 Fig. 1.7), but in association with repeated lithological depth changes and allied high diatom content. In the three diatom-rich layers between the surface and about 400 mbsf, there was a consistent stimulation of prokaryotic activity (sulfate reduction, growth – thymidine incorporation into DNA) and total cell numbers and/or the proportion of divid-

12 | 1 A subseafloor biosphere update ing and divided cells [24]. It may be that diatomaceous organic matter is considerably less reactive than other sedimentary organic matter and as a consequence can fuel low, but continuing, prokaryotic activity over long periods. The deepest layer (~250 to 320 mbsf) was 7–11 Myr, which markedly extended the known time scale for stimulation of subsurface prokaryotic processes [1]. Furthermore, the diatom layers are controlled by Milankovitch scale cycles via oceanographic variability, intriguingly this links the depth distribution of the deep biosphere prokaryotes in some marine sediments to Earth’s orbital forcing [45]. Furthermore, in the top and bottom diatom-rich layers there was an increasing concentration of dissolved manganese, indicating active prokaryotic manganese reduction in deep sediments. Manganese reduction would normally be expected to be restricted to near-surface layers, but here, due to a combination of high input of minerals and their slow reduction, continuing activity occurred in deeper layers. These sediments also deviate from expected diagenetic sequences in terms of sulfate (brine incursion), iron-reduction, methane formation in sulfate containing layers and oxidized fluids at the sediment-basement interface [33].

1.1.6 Newfoundland Margin (Leg 210) Newfoundland Margin deep sediments are ancient and record the rifting of the North Atlantic Ocean, and thus were an important target to investigate the subseafloor biosphere in old and deep sediments. To enable deep samples to be obtained in the drilling time available, drilling occurred through the top 800 mbsf without coring, then coring was conducted from 800 to 1739 mbsf with excellent recovery (average 85%). The sedimentary succession consisted of background hemipelagic mudrocks with various proportions of interbedded gravity-flow deposits and terminated in diabase sills. Nine samples from Site 1276 were microbiologically analyzed with ages from 46 to 111 My [46]. Prokaryotic cells were present at all depths and distribution was similar to other marine sediments (󳶳 Fig. 1.8). The presence of dividing and live cells indicated that some of these cells were active, and this was supported by the extraction and amplification of archaeal 16S rRNA genes. Resulting 16S rRNA gene libraries showed a low diversity of Archaea with thermophilic Pyrococcus dominating the 958 m depth, and then as soon as methane increases above background concentrations, potential anaerobic methane-oxidizing archaeal (ANME) sequences became dominant. This continued until 1626 mbsf, with temperatures between 60 and 100 °C and high methane concentrations, where Pyrococcus and Thermococcus sequences dominated. This change may reflect the upper temperature limit for ANME prokaryotes [47] and thus other Archaea adapted to higher temperatures, and possibly able to use thermogenic higher hydrocarbons that accumulated below the diabase sill developed. These data provided direct evidence that significant prokaryotic populations are present in subseafloor sediments to greater than kilometer depths and as old

1.1 New sites investigated

Methane (ppm) 3 4 4 1x10 2x10 5x10

0

10

4

Log10 cell number (cells/mL) 5 6 10 10

| 13

7

10

800 Tmin 900

Tmax

Pyrococcus (67%) Rice Cluster VI (33%) 40°C 60°C

1000 ND

Depth (mbsf)

1100

ANME-2 (100%) 50°C 70°C

1200 1300 1400

ANME-3 (100%) 55°C 90°C

1500 Pyrococcus (88%) Thermococcus (12%) 60°C 100°C

1600 1700 0

5

10 15 20 25 HI (mg HC/g TOC)

30

35 0

4

8 12 16 Dividing cells (%)

20

Fig. 1.8: Newfoundland Margin, Leg 210 [46]. Depth profiles of methane (black dots with orange line), prokaryotic cells (red circles), and percentage dividing cells (blue squares). Regression line for prokaryotic cells in other marine deep sediments (solid triangles), prediction limits (...) [1]. Orange arrows show local increases in methane. Hydrogen Index (HI open triangles) measured as mg of hydrocarbon (HC) per g of total organic carbon (TOC). ND, not determined. Dominant archaeal 16S rRNA gene sequences and in situ temperature range are on the right at the depths obtained. The diabase sill is shown as a bold horizontal dashed line.

as 111 My. Considering the 122 °C upper temperature limit for some prokaryotes [41], temperature alone would not limit prokaryotes until much deeper depths.

1.1.7 Carbonate mound (IODP Expedition 307) The Challenger Mound (water depth 781–815 m water depth, 󳶳 Fig. 1.9) is a prominent mound structure (155 m high), which is partially buried with sediment and dead coral rubble on the Southwest Irish continental margin [48, 49]. Two mound sites, Flank (IODP site U1316) and Mound (IODP site U1317), were compared with a nonmound Reference site (IODP site U1318) upslope from the Challenger Mound [48]. This was the first carbonate mound to be drilled (~270 m) and analyzed in detail for microbiology and biogeochemistry (catalyzed reporter deposition-fluorescence in situ hybridization [CARD-FISH], qPCR [16S rRNA and functional genes, dsrA and mcrA], and 16S rRNA gene PCR-DGGE for prokaryotic diversity, and this was compared with the distribution of total and culturable cell counts, radiotracer activity measurements and geochemistry). There was a significant and active prokaryotic community both within and beneath the carbonate mound. As found in the Eastern Equatorial Pacific and Peru Margin Sites, prokaryotic activity at Expedition 307 Sites was quite diverse and activities

14 | 1 A subseafloor biosphere update did not follow the expected depth distributions based on a sequence of reactions providing decreasing energy yield. Although total cell numbers at certain depths were lower than the global average for other subseafloor sediments and prokaryotic activities were relatively low (iron and sulfate reduction, acetate oxidation, methanogenesis) they were significantly enhanced compared with the Reference site. In addition, there was some stimulation of prokaryotic activity in the deepest sediments (Miocene, > 10 Ma), including potential for anaerobic oxidation of methane activity below the mound base. Both Bacteria and Archaea were present, with neither really dominant (overall 50% and 34%, respectively, with considerable variability in proportions between these geographically close sites, 󳶳 Tab. 1.1). These were related to sequences

Fig. 1.9: Location of the Challenger Mound Site IODP Expedition 307.

1.2 High-pressure cultivation – DeepIsoBUG, gas hydrate sediments | 15

commonly found in other subseafloor sediments (Gammaproteobacteria, Chloroflexi, JS1, SAGMEG, MBG-D and MCG). Overall, fewer prokaryotic sequences were detected at depth at all sites despite some activities being elevated in deeper layers. However, the majority of these sequences were mainly related to uncultured groups of prokaryotes from a range of different environments, and therefore, it is unclear what metabolisms are responsible for the measured deep elevated thymidine incorporation or acetate oxidation, particularly at the mound sites, and in the apparent absence of significant iron and sulfate reduction. However, there were some contradictions within the molecular diversity data, for example no Archaea were detected by CARD-FISH, but they were detected by qPCR, PCR-DGGE and indirectly by the presence of archaeal methanogenesis at the mound Sites. In addition, the functional methanogen mcrA gene was not detected at these mounds Sites, yet was detected at the Reference site which had no detectable methane or methanogenesis. Such discrepancies may help to explain some of the differences in prokaryotic diversity at the same deep sediment locations by different research groups, for example, dominance of either Archaea or Bacteria at Leg 201 Sites [4, 5]. Despite these problems, active subseafloor prokaryotic populations were elevated in Mound sites compared to the Reference Site and with an estimate of some 1600 mounds in the Porcupine Basin alone, carbonate mounds may represent a significant prokaryotic subseafloor habitat.

1.2 High-pressure cultivation – DeepIsoBUG, gas hydrate sediments Despite the ubiquitous presence of prokaryotic cells in subseafloor sediments and their large biomass, only a very small proportion of this population can be cultured (e.g. 0.1%, [33]). In addition, there is often a major discrepancy between the prokaryotes detected by molecular genetic approaches and culturing. Also, many phylotypes in clone libraries are unrelated to cultured sequences. Therefore, there is a large prokaryotic diversity in subseafloor sediments which has not been cultured and this severely limits our understanding of this major prokaryotic habitat. A key feature of subsurface environments is elevated pressure, e.g. ~70% of the ocean is at a pressure of 38 MPa or above [51], plus there is up to 10 km (~100 MPa) of sediment in some locations. Thus, the majority of subseafloor prokaryotes live under, and are likely to be adapted to, high pressure, which could be essential for culturing representative subseafloor prokaryotes. We, therefore, developed a new system, Deep-IsoBUG [50], which can maintain sediments under elevated pressure (max 25 MPa) for enrichment, growth and isolation of prokaryotes at pressures up to 100 MPa. When this system is coupled with pressurized subsurface cores obtained using the HYACINTH drilling and core storage

Mound flank site U1316 18.9 90.0 98.8 Mound site U1317 4.9 4.9 coral 20.9 20.9 coral 39.0 106.4 coral 146.6 219.9 Reference site U1318 22.0 221.0

Site/Sample depth (mbsf)

– 25 – 2.8 5.7 3.6 8.7 – 8.8 50 14.3 9.1 80

5.6 9.8 10.8 27.8 – 19.5 – 28.6

– –

– –

4.7 6.6 6.3 8.7 5.6 8.8 50 42.8

4.8 25 –

13.6 –

5.6 4 4.5 2.6 27.7 0.9 – –

23.8 25 –

Proteobacteria Beta Gamma Delta

– – –

Alpha

18.2 –

16.8 14.7 23.4 15.7 16.7 15.9 – –

23.8 – –

– –

4.7 9 5.4 7 5.6 8.8 – –

– 25 100

– 20

2.8 8.2 9 10.4 – 12.4 – –

– – –

Chloro- Actino- Firmiflexi bacteria cutes

– –

2.8 3.2 6.3 3.5 – 4.4 – –

– – –

Nitrospirae

4.5 –

– – – – – – – –

– – –

31.8 –

– – – – 11.1 – – –

9.5 – –

Spiro- JS1 chaetes

– –

14 12.3 3.6 0.9 – 5.4 – –

– – –

OP1

4.5 –

– – – – 5.6 – – –

– – –

OP8

Major phylogenetic groups in 16S rRNA gene libraries or DGGE analysis (%) Bacteria

– –

30.8 18 16.2 4.3 – 9.7 – –

– – –

– –

– – – – 5.6 – – –

9.5 – –

OP11/ NTOD1 B6

Table 1.1: A comparison of bacterial and archaeal phylogenetic groups found in Porcupine Seabight (IODP Expedition 307) sediments.

18.2 –

9.4 8.4 10.8 10.4 22.1 5.4 – 14.3

28.6 – –

22 5

21,588 19,729 26,617 24,173 18 27,678 4 7

21 4 1

Total number of sequences/ Others DGGE bands

16 | 1 A subseafloor biosphere update

– 54.3 67.7 38.5 89.5 20 77.6 – 66.6 100

–

8.6 – – – – – –

– –

MCG

– –

4.3 1.6 32.1 – – – –

–

MBG-B

Crenarchaeota

– –

20 22.6 17.9 8.8 80 14.3 100

75

33.3 –

– – – – – – –

25

SAGMEG MBG-D

– –

2.9 6.5 1.3 1.7 – – –

–

Methanomicrobiales

Archaea Euryarchaeota

– –

5.7 1.6 10.2 – – 8.1 –

–

Halobacteriales

– –

4.2 – – – – – –

–

Others

3 1

20,855 17,780 19,934 19,428 5 16,010 2

4

Data from Webster et al. [48] and Hoshino et al. [70]; calculated from prokaryotic community profiles shown in Hoshino et al. [70]. JS1, OP1, OP8, OP11, OD1 = candidate divisions JS1, OP1, OP8, OP1 and OD1; NT-B6 = novel bacterial group NT-B6. MG1 = Marine Group 1; MCG = Miscellaneous Crenarchaeotal Group; MBG-B = Marine Benthic Group B (or Deep Sea Archaeal Group); SAGMEG = South African Gold Mine Euryarchaeotal Group; MBG-D = Marine Benthic Group D; Others = other lineages, unaffiliated sequences and/or unsequenced DGGE bands.

Mound flank site U1316 18.9 Mound site U1317 4.9 4.9 coral 20.9 20.9 coral 39.0 106.4 coral 219.9 Reference site U1318 22.0 221.0

Thaumarchaeota MG1

1.2 High-pressure cultivation – DeepIsoBUG, gas hydrate sediments | 17

18 | 1 A subseafloor biosphere update system [52], and the PRESS core cutting system [53], DeepIsoBUG enables the recovery and handling of cores at in situ pressures (up to 25 MPa) and subsequent enrichment and isolation of prokaryotes at a range of pressures, without depressurization. The system was first used with subsurface gas hydrate sediments from the Indian Continental Shelf, Cascadia Margin and Gulf of Mexico [50]. Generally, highest cell concentrations occurred in enrichment cultures at close to in situ pressures (14 MPa) with a variety of media, although growth continued up to at least 80 MPa. Predominant bacterial populations in enrichments were Carnobacterium, Clostridium, Marinilactibacillus and Pseudomonas, plus Acetobacterium and Bacteroidetes species in Indian Continental Shelf samples, largely independent of media and pressures. 16S rRNA gene sequences related to all of these Bacteria have been previously detected in deep, subsurface environments, but these were not detected in Indian gas hydrate sediments. Most isolated strains were closely related to those in enrichments (99–100%), although isolates were only piezotolerant, being able to grow at atmospheric pressure. Only the Clostridium and Acetobacterium were obligate anaerobes. No Archaea were enriched. It may be that these sediment samples were not deep enough (total depth 1126–1527 m) to obtain obligate piezophiles.

1.3 Subseafloor biosphere simulation experiments The depth (> 1,626 mbsf ), magnitude and presence of subseafloor prokaryotes in ancient deposits (> 111 My) all make it difficult to understand the energy sources that are available within their “geological dimensions” of time and space. It has been shown, however, that temperature increases accompanying sediment burial can continuously enhance the reactivity of organic matter as reflected in the production of low molecular weight organic acids and, therefore, has the potential for sustaining prokaryotes in deep, subseafloor sediments [25]. This research has subsequently been extended using both long-term (> 500 days) static (thermal gradient system, 0–100 °C) and sequential heating (30–90 °C at different heating rates) sediment slurry experiments to show that prokaryotes produced considerable concentrations of H2 , CH4 and other hydrocarbons, in addition to organic acids, thereby providing energy for themselves and other bacteria. In deep sediments, these compounds had previously been interpreted as being solely products of thermogenic origin, but sterile controls (irradiated) showed that they were predominantly of prokaryotic origin in these heating experiments, and hence, also probably in deep sediments. Even after 300 days at constant temperature (30 or 60 °C) incremental heating to 90 °C stimulated prokaryotic activity, further supporting the temperature-related activation of recalcitrant organic matter. The measured decrease in hydrogen content of organic matter was consistent with its aromatization, and the observed H2 production [26]. Dissolved sulfate, after initially being rapidly reduced by sulfate-reducing bacteria in these slurries, was produced at higher temperatures (~60 °C), which might indicate

1.3 Subseafloor biosphere simulation experiments | 19

(a)

(b) 350

350

2500

300

2500

300

1000

100

250 1500

200 150

1000

100

500

500

50

50

8000

15,000

8000

7000

15,000

7000

6000

3000 5000 2000

10,000 5000 4000 3000 5000

–1 SO2– 4 (µmol L )

4000

Acetate (µmolL–1)

6000 10,000

5000

–1 SO2– 4 (µmol L )

Acetate (µmolL–1)

CO2 (µmol L–1) slurry headspace

150

CH4, H2 (µmol L–1) slurry headspace

1500

200

2000

CO2 (µmol L–1) slurry headspace

CH4, H2 (µmol L–1) slurry headspace

2000 250

2000

1000

1000

0 0

20

40

60

80

Temperature (°C)

100

0 120

0 0

20

40

60

80

100

0 120

Temperature (°C)

Fig. 1.10: Headspace gas and sulfate and acetate changes (a). Basalt amended sediment slurries heated in a Thermal Gradient (83 days) compared to sterile controls (130 days) (b). ◻ CO2 , O CH4 , ♦ H2 , ○ sulfate, V acetate [40].

anaerobic sulfide oxidation by temperature-activated iron oxide minerals [54]. To explore this, and potential additional H2 production from iron minerals, magnetite was added to sediment slurries at 60 °C. This enhanced H2 production, suggesting that iron oxide minerals might be a potential inorganic source of H2 in subseafloor sediments, and in the deep, hot sediments of the Nankai Trough, H2 increased along with increases in sulfate and acetate (󳶳 Fig. 1.6). In the slurry experiments [26], H2 formation at higher temperatures stimulated development of autotrophic bacterial populations and a similar situation may occur in deep marine sediments. To explore this further, a range of minerals was added to sediment slurries incubated at a range of temperatures (0 to ~100 °C, for up to 83 days, [40]). Mineral addition resulted in considerable H2 (maximum 1626 μM) and acetate production (maximum ~10,000 μM), which increased with incubation time and temperature. During incubation, some of the H2 and acetate was probably also consumed to fuel stimulated sulfate removal and CH4 production (󳶳 Fig. 1.10), as both are important substrates for sulfate reduction and methanogenesis. These reactions were conducted by prokaryotes as sterile controls (triple autoclaved), had negligible H2 and CH4 production and sulfate removal. Also, a range of both Archaea and Bacteria similar to those often found in deep sediments were present in the biotic experiments. As a range of iron and non-iron minerals (hematite, labradorite, pyrite, basalt, ilmenite, hornblende,

20 | 1 A subseafloor biosphere update olivine, magnetite, quartz sand) all stimulated H2 formation there must be a general mechanism of H2 formation, in addition to oxidation of reduced ferrous iron by water [55]. Mechanochemistry is a likely process to be involved, whereby reactive surfaces can facilitate the free radical oxidation of water [56]. As the Earth is tectonically active, rocks are constantly subjected to stress, fracturing and faulting, and thus producing reactive surfaces, mechanochemistry should be an important and widespread process. In these slurry experiments, prokaryotes may facilitate mineral H2 formation by creating reactive surfaces by mineral weathering and/or preventing passivation of active surfaces by silica gel barrier formation [57]. Extending the heating experiments to sequential heating up to 155 °C, showed that the previous presence of prokaryotic activity also stimulated reactions at thermogenic temperatures, including formation of free hydrocarbons [40]. These experimental results show the potentially important interplay between the biosphere and geosphere in the functioning of deep sedimentary processes.

1.4 Conclusions The documentation of active prokaryotic cells at the seven additional subseafloor locations studied since the Hydrogeology Journal review [1] further reinforces the ubiquitous presence of prokaryotes in subseafloor sediments. This data plus total cell numbers for near-surface sediments from other cruises, has increased the number of cell counts from ~890 to 1927, however, the cell depth regression is little changed (󳶳 Fig. 1.11). Similar results have also been obtained by other research groups, either using similar microscopic techniques [58] or independent assessment of intact polar membrane lipids and prokaryotic DNA [59]. Although, lower subseafloor biomass estimates of 0.18–3.6% of the global total biomass have been suggested by some researchers when taking into account the low cells concentrations in open-ocean gyre sediments [9]. This lower value is a known underestimate of subseafloor prokaryotic cells, as the sensitive counting technique for low biomass sediments misses some 10– 30% of cells [60] and subsurface increases in cell concentrations at some sites (e.g. 󳶳 Fig. 1.6 & 1.7) have been excluded. Thus, our original estimate that prokaryotes in subseafloor sediments might globally represent 10% of total biomass [2], still seems a reasonable estimate, especially considering elevated prokaryotic activities: in subsurface gas hydrate deposits, at sulfate-methane interfaces, in organic rich layers, with products of deep thermogenic processes diffusing upwards to fuel the base of the biosphere and in oil and gas reservoirs [61, 62]. In addition, it has been suggested that bacterial endospores are as abundant as intact prokaryotic cells in subseafloor sediments, and hence, add to biomass estimates [7]; but only if spores are not stained by dyes such as acridine orange. With increasing molecular genetic analysis of subseafloor sediments, it is starting to become clear that there are some prokaryotic groups which are often dominant in

1.4 Conclusions

Log cell number (cells/cm3)

4

5

6

7

8

9

10

| 21

Log cell number (cells/cm3) 4

11

5

6

7

8

9

10

11

0.001

0 100 200

0.010

300 400 0.100

500 600

800 900 1000

1.000

Depth (m)

Depth (m)

700

10.000

1100 1200

100.000

1300 1400 1000.000 1500

1600 1700

10000.000

Fig. 1.11: Depth distribution of prokaryotic cells in sub-surface sediments from sites in the Pacific and Atlantic Oceans and the Mediterranean Sea. Bold dashed regression line is log cells = 8.08 − 0.69 × log depth (m) r−sq = 0.718, n = 1927 and light dashed lines are 95% lower and upper prediction limits (unpublished results).

subseafloor sediments. For Bacteria these are: Gammaproteobacteria, Chloroflexi and members of the candidate division JS1 [38], which make up 18.9%, 17.3% and 26.1% (sum = 62.3%; present in 62–70% of libraries) of clones, respectively [63]. However, the Alpha-, Beta-, Delta- and Epsilonproteobacteria are also present (in 3–50% of the libraries), but are not so common, averaging only 7.8%, 4.9%, 3.7% and 2.1% of clones, respectively. Of the remaining 21.3% of clones, the Planctomycetes are reasonably abundant (2–26%) at some Peru and Cascadia Margin sites and depths [24, 64], as are the novel groups NT-B2 and NT-B6, both originally found in the Nankai Forearc Basin [65]. For Archaea: the Crenarchaeota dominate with 65% of clones, while only 24.5% of clones belong to the Euryarchaeota and 8.4% to the Thaumarchaeota [63]. The most abundant crenarchaeotal groups are the Miscellaneous Crenarchaeotic Group (MCG) and Marine Benthic Group B (MBG-B; synonymous with the Deep-Sea Archaeal Group, DSAG; [66]) comprising 33% and 26.3% of clones, respectively. The next most abundant groups are the Marine Group 1 (8.4%; Thaumarchaeota), the South African

22 | 1 A subseafloor biosphere update Gold Mine Groups (SAGMEG) 1 and 2 (7.6%; Euryarchaeota) and the thermophilic Euryarchaeota (7.6%); none of the other groupings account for more than about 4.5% overall. These groups have also been found in other sedimentary, aquatic and terrestrial environments, and so are not confined to the deep marine biosphere [67]. The only archaeal phylotypes closely related to cultured species are the euryarchaeotal methanogens, thermophiles and hyperthermophiles. However, these only account for < 8% of the clones and are not a major component of the phylotypes; thus, as with the Bacteria, most of the subseafloor Archaea are from uncultured lineages. Single-cell genomic sequencing of MCG (one cell) and MBG-D (three cells) has shown the presence of extracellular protein-degrading enzymes such as gingipain and clostripain, indicating that these uncultured Archaea may have a previously undiscovered role in protein remineralization in anoxic marine sediments [6]. Importantly, in the sediment heating experiments previously described, prokaryotic populations of both Bacteria and Archaea are representative of many deep sediment types developed, including archaeal thermophiles at high temperatures (65–90 °C), which can dominate deep hot sediments [40]. Despite the above, there has been considerable controversy regarding the relative dominance of the two prokaryotic Domains, Bacteria or Archaea in subseafloor sediments, with some studies showing that Bacteria are dominant [4], whilst others that Archaea dominate [5]. As these two studies included some of the same sediment sites, both studies cannot be correct. For detection of viable cells, Biddle et al. [5] used a standard FISH procedure which has limited sensitivity even in near-surface sediments [68], but the dominance of Archaea was also supported by independent assessment based on 16S rRNA and intact polar lipids. Subsequently, however, it has been suggested that archaeal glycosidic ether lipids may degrade more slowly than bacterial phospholipids in sediments [17] and as a result, archaeal lipids may represent fossil rather than living Archaea. Interestingly, even with improved archaeal DNA quantification, archaeal 16S rRNA genes were only 50% of total prokaryotic DNA [59]. In addition, clear community changes at deep methane-sulfate interfaces for Bacteria in Peru Margin sediments (30 and 90 mbsf, 󳶳 Fig. 1.7), demonstrate a dynamic community responding to changing environmental conditions, in contrast, archaeal diversity was more limited and did not change at interfaces [24]. Stimulation of deep prokaryotic activity at deeper diatom layers/interfaces in an open ocean site demonstrates viable prokaryotes in even deeper and more ancient deposits (7–11 My, deepest ~250 to 320 mbsf, [24]), and their correlation with orbital forcing (Milankovitch Cycles, [45]), via enhanced ocean productivity, elegantly shows how the deep biosphere is an integral part of Earth system processes over geological time scales. This coupled with subseafloor prokaryotes catalyzing a diverse range of metabolic processes, often within the same depth layers [24, 33], 󳶳 Fig. 1.5), demonstrates different controls on prokaryotic activity compared to near-surface sediments where often clear depth sequences of activity occur based on maximum energy yield (e.g. [69]. It may be that severe energy limitation means that no one

References |

23

metabolism/prokaryotic population can dominate and this is probably linked to the recalcitrance of both buried organic and inorganic compounds plus changes in both their type and quantity during deposition over time. However, sediment slurry heating experiments demonstrate that warming of sediments during burial can activate both buried organic and inorganic compounds [40] and thus slowly and continuously supply energy sources for subsurface prokaryotes. For example, in Nankai Trough, below ~300 mbsf there is a broad match between modeled increasing organic matter reactivity and rates of prokaryotic activity [36] and with H2 and acetate formation at depth, plus relative increases in prokaryotic cell numbers (󳶳 Fig. 1.6). A similar situation occurred at Newfoundland Margin sediments, but here CH4 increases were associated with the presence of ANME sequences and presumably utilization of CH4 as a deep energy source (󳶳 Fig. 1.8). However, at higher temperatures (60–100 °C), archaeal thermophiles/hyperthermophiles dominated in the deepest (1626 mbsf) and oldest (111 My) subseafloor sediments that prokaryotic cells have currently been detected [46]. These prokaryotes might be utilizing thermogenic higher hydrocarbons diffusing from below. However, sediment slurry sequential heating experiments suggest an even more close interaction between deep biogenic and thermogenic processes, as biogenic alteration of organic matter greatly stimulates hydrocarbon production at thermogenic temperatures [40]. No doubt that in the future subseafloor prokaryotes will be demonstrated in even deeper sediments.

References [1] [2] [3] [4] [5] [6] [7]

[8] [9]

Parkes RJ, Cragg BA, Wellsbury P. Recent studies on bacterial populations and processes in subseafloor sediments: A review. Hydrogeol J 8, (2000), 11–28. Parkes RJ, Cragg BA, Bale SJ, et al. Deep Bacterial Biosphere in Pacific-Ocean Sediments. Nature 371 (1994), 410–413. D’Hondt S, Rutherford S, Spivack AJ. Metabolic activity of subsurface life in deep-sea sediments. Science 295 (2002), 2067–2070. Schippers A, Neretin LN, Kallmeyer J, et al. Prokaryotic cells of the deep subseafloor biosphere identified as living bacteria. Nature 433 (2005), 861–864. Biddle JF, Lipp JS, Lever MA, et al. Heterotrophic Archaea dominate sedimentary subsurface ecosystems off Peru. Proc Natl Acad Sci USA 103 (2006), 3846–3851. Lloyd KG, Schreiber L, Petersen DG, et al. Predominant archaea in marine sediments degrade detrital proteins. Nature (2013), doi:10.1038/nature12033. Lomstein BA, Langerhuus AT, D’Hondt S, Jørgensen BB, Spivack AJ. Endospore abundance, microbial growth and necromass turnover in deep subseafloor sediment. Nature 484 (2012), 101–104. Kubo K, Lloyd K, Biddle, Amann R, Teske A, Knittel K. Archaea of the Miscellaneous Crenarchaeotal Group are abundant, diverse and widespread in marine sediments. ISME J (2012). Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16213– 16216.

24 | 1 A subseafloor biosphere update [10] D’Hondt S, Spivack AJ, Pockalny R, et al. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106 (2009), 11651–11656. [11] Schippers A, Koweker G, Hoft C, Teichert BMA. Quantification of Microbial Communities in Forearc Sediment Basins off Sumatra. Geomicrobiol J 27 (2010), 170–182. [12] Lever MA, Alperin MJ, Teske A, et al. Acetogenesis in Deep Subseafloor Sediments of The Juan de Fuca Ridge Flank: A Synthesis of Geochemical, Thermodynamic, and Gene-based Evidence. Geomicrobiol J 27 (2010), 183–211. [13] Takano Y, Chikaraishi Y, Ogawa NO, et al. Sedimentary membrane lipids recycled by deep-sea benthic archaea. Nature Geosci 3 (2010), 858–861. [14] Engelen B, Ziegelmueller K, Wolf L, et al. Fluids from the oceanic crust support microbial activities within the deep biosphere. Geomicrobiol J 25 (2008), 56–66. [15] Orcutt BN, Sylvan JB, Knab NJ, Edwards KJ. Microbial Ecology of the Dark Ocean above, at, and below the Seafloor. Microbiol Mol Biol Rev 75 (2011), 361–422. [16] Middelboe M, Glud RN, Filippini M. Viral abundance and activity in the deep subseafloor biosphere. Aquat Microb Ecol 63 (2011), 1–8. [17] Schouten S, Middelburg JJ, Hopmans EC, Sinninghe Damste JS. Fossilization and degradation of intact polar lipids in deep subsurface sediments: A theoretical approach. Geochemica et Cosmochimica Acta 74 (2010), 3806–3814. [18] Lever MA, Rouxel O, Alt JC, et al. Evidence for Microbial Carbon and Sulfur Cycling in Deeply Buried Ridge Flank Basalt. Science 339 (2013), 1305–1308. [19] Wellsbury P, Mather ID, Parkes RJ. Bacterial abundancies and pore-water acetate concentrations in sediments of the Southern Ocean (Sites 1088 and 1093). Proceedings of the Ocean Drilling Program, Scientific Results 177 (2001), 1–12 [on-line]. [20] Gersonde R, Hodell DA, Blum P, et al., Proc ODP, Init Repts, 177 [CD- ROM] Available from: Ocean Drilling Program, Texas A&M University, College Station TX 77845–9547, USA 1999. [21] Brown J, Colling A, Park D, Phillips J, Rothery D, Wright J. Ocean chemistry and deep-sea sediments. Oxford (Pergamon Press). Oxford: Pergamon Press; 1989. [22] Wellsbury P, Mather I, Parkes RJ. Geomicrobiology of deep, low organic carbon sediments in the Woodlark Basin, Pacific Ocean. FEMS Microbiology Ecology 42 (2002), 59–70. [23] Mather ID, Parkes RJ. Bacterial profiles in sediments of the eastern flank of the Juan de Fuca ridge, Sites 1026 and 1027. Proceedings of the Ocean Drilling Program, Scientific Results 168 (2000), 161–165, doi:10.2973/odp.proc.sr.168.023.2000. [24] Parkes RJ, Webster G, Cragg BA, et al. Deep subseafloor prokaryotes stimulated at interfaces over geological time. Nature 436 (2005), 390–394. [25] Wellsbury P, Goodman K, Barth T, Cragg BA, Barnes SP, Parkes RJ. Deep marine biosphere fuelled by increasing organic matter availability during burial and heating. Nature 388 (1997), 573–576. [26] Parkes RJ, Wellsbury P, Mather ID, et al. Temperature activation of organic matter and minerals during burial has the potential to sustain the deep biosphere over geological time scales. Organic Geochemistry 38 (2007), 845–852. [27] Krumholz LR, McKinley JP, Ulrich FA, Suflita JM. Confined subsurface microbial communities in Cretaceous rock. Nature 386 (1997), 64–66. [28] Mather ID, Wellsbury P, Parkes RJ, Maxwell JR. Purge – trap adsorption gas analysis of sediments of the Western Woodlark Basin Sites 1109 and 1115. Proceedings of the Ocean Drilling Program, Scientific Results 180 (2001), 1–14 [on-line]. [29] Cragg B, Wellsbury P, R.J M, Parkes RJ. Bacterial populations in deep-water, low-sedimentation-rate marine sediments and evidence for subsurface bacterial manganese reduction (ODP Site 1149 Izu-Bonin Trench). Proceedings of the Ocean Drilling Program, Scientific Results 185, 2003.

References |

25

[30] Smith DC, Spivack AJ, Fisk MR, Haveman SA, Staudigel H., Party ODPLSS. Tracer-based estimates of drilling-induced microbial contamination of deep sea crust. Geomicrobiol J 17 (2000), 207–219. [31] Cragg BA, Law KM, Cramp A, Parkes RJ. Bacterial profiles in Amazon Fan sediments (Sites 934, 940). Proceedings of the Ocean Drilling Programe, Scientific Results 155 (1997), 565–571. [32] Cragg BA, Parkes RJ, Fry JC, et al. Bacterial profiles in deep sediments of the Santa Barbara Basin Site 893. Proceedings of the Ocean Drilling Programe, Scientific Results 146 (1995), 139– 144. [33] D’Hondt S, Jørgensen BB, Miller DJ, et al. Distributions of microbial activities in deep subseafloor sediments. Science 306 (2004), 2216–2221. [34] Wang GZ, Spivack AJ, Rutherford S, Manor U, D’Hondt S. Quantification of co-occurring reaction rates in deep subseafloor sediments. Geochim Cosmochim Acta 72 (2008), 3479–3488. [35] Newberry CJ, Webster G, Cragg BA, Parkes RJ, Weightman AJ, Fry JC. Diversity of prokaryotes and methanogenesis in deep subsurface sediments from the Nankai Trough, Ocean Drilling Program Leg 190. Environmental Microbiology 6 (2004), 274–287. [36] Horsfield B, Schenk HJ, Zink K, et al. Living microbial ecosystems within the active zone of catagenesis: Implications for feeding the deep biosphere. Earth Planet Sci Lett 246 (2006), 55–69. [37] Zink KG, Wilkes H, Disko U, Elvert M, Horsfield B. Intact phospholipids - microbial “life markers” in marine deep subsurface sediments. Organic Geochemistry 34 (2003), 755–769. [38] Webster G, Parkes RJ, Fry JC, Weightman AJ. Widespread occurrence of a novel division of bacteria identified by 16S rRNA gene sequences originally found in deep marine Sediments. Applied and Environmental Microbiology 70 (2004), 5708–5713. [39] Riedinger N, Brunner B, Formolo MJ, et al. Oxidative sulfur cycling in the deep biosphere of the Nankai Trough, Japan. Geology 38 No. 9, (2010), 851–854. [40] Parkes R, Linnane C, Webster G, et al. Prokaryotes stimulate mineral H2 formation for the deep biosphere and subsequent thermogenic activity. Geology 39 (2011), 219–222. [41] Takai K, Nakamura K, Toki T, et al. Cell proliferation at 122 °C and isotopically heavy CH4 production by a hyperthermophilic methanogen under high-pressure cultivation. Proc Natl Acad Sci USA 105 (2008), 10949–10954. [42] Sorensen KB, Teske A. Stratified communities of active archaea in deep marine subsurface sediments. Applied And Environmental Microbiology 72 (2006), 4596–4603. [43] Parkes RJ, Cragg BA, Fry JC, Herbert RA, Wimpenny JWT. Bacterial biomass and activity in deep sediment layers from the Peru margin. Phil Trans R Soc Lond 331 (1990), 139–153. [44] Cragg BA, Kemp AES. Bacterial Profiles In Deep Sediment Layers From The Eastern Equatorial Pacific Ocean, Site 851. In: Pisias, NG, Mayer, LA, Janecek, TR, Palmer-Julson, A and van Andel, TH (Eds), Proceedings of the Ocean Drilling Programe, Scientific Results 138 (1995) 599–604. [45] Aiello IW, Bekins BA. Milankovitch-scale correlations between deeply buried microbial populations and blogenic ooze lithology. Geology 38 (2010), 79–82. [46] Roussel EG, Cambon-Bonavita MA, Querellou J, et al. Extending the sub-sea-floor biosphere. Science 320 (2008), 1046. [47] Kallmeyer J, Boetius A. Effects of temperature and pressure on sulfate reduction and anaerobic oxidation of methane in hydrothermal sediments of Guaymas Basin. Applied and Environmental Microbiology 70 (2004), 1231–1233. [48] Webster G, Blazejak A, Cragg BA, et al. Subsurface microbiology and biogeochemistry of a deep, cold-water carbonate mound from the Porcupine Seabight (IODP Expedition 307). Environmental Microbiology 11 (2009), 239–257. [49] Ferdelman TG, Kano A, Williams T, Henriet JP, et al. Proceedings of the Integrated Ocean Drilling Program, Expedition Reports 307 (2006), doi:10.2204/iodp.proc.307.102.2006.

26 | 1 A subseafloor biosphere update [50] Parkes RJ, Sellek G, Webster G, et al. Culturable prokaryotic diversity of deep, gas hydrate sediments: first use of a continuous high-pressure, anaerobic, enrichment and isolation system for subseafloor sediments (DeepIsoBUG). Environmental Microbiology 11 (2009), 3140–3153. [51] Abe F, Horikoshi K. The biotechnological potential of piezophiles. Trends Biotechnol 19 (2001), 102–108. [52] Schultheiss PJ, Francis TJG, Holland M, et al. Pressure coring, logging and subsampling with the HYACINTH system. Geological Society London Special Publications 267 (2006), 151–163. [53] Parkes RJ, Martin D, Amann H, et al. Technology for High-pressure Sampling and Analysis of Deep-sea Sediments, Associated Gas Hydrates, and Deep-biosphere Processes. In: T Collett, A Johnson, C Knapp, and R Boswell (Eds): Natural gas hydrates—Energy resource potential and associated geologic hazards: AAPG Memoir 89, p 672–683, 2009. [54] Bottrell SH, Parkes RJ, Cragg BA, Raiswell R. Isotopic evidence for anoxic pyrite oxidation and stimulation of bacterial sulfate reduction in marine sediments. J Geol Soc 157 (2000), 711–714. [55] Stevens TO, McKinley JP. Lithoautotrophic Microbial Ecosystems in Deep Basalt Aquifers. Science 270 (1995), 450–454. [56] Kita I, Matsuo S, Wakita H. H2 Generation by Reaction Between H2 O and Crushed Rock - An Experimental-Study on H2 Degassing From the Active Fault Zone. Journal of Geophysical Research 87 (1982), 789–795. [57] Freund F, Dickinson JT, Cash M. Hydrogen in rocks: An energy source for deep microbial communities. Astrobiology 2 (2002), 83–92. [58] Morono Y, Terada T, Masui N, Inagaki F. Discriminative detection and enumeration of microbial life in marine subsurface sediments. ISME Journal 3 (2009), 503–511. [59] Lipp JS, Morono Y, Inagaki F, Hinrichs KU. Significant contribution of Archaea to extant biomass in marine subsurface sediments. Nature 454 (2008), 991–994. [60] Kallmeyer J, Smith DC, Spivack AJ, D’Hondt S. New cell extraction procedure applied to deep subsurface sediments. Limnol Oceanogr Meth 6 (2008), 236–245. [61] Head IM, Jones DM, Larter SR. Biological activity in the deep subsurface and the origin of heavy oil. Nature 426 (2003), 344–352. [62] Bennett B, Adams JJ, Gray ND, et al. The controls on the composition of biodegraded oils in the deep subsurface - Part 3. The impact of microorganism distribution on petroleum geochemical gradients in biodegraded petroleum reservoirs. Organic Geochemistry 56 (2013), 94–105. [63] Fry JC, Parkes RJ, Cragg BA, Weightman AJ, Webster G. Prokaryotic biodiversity and activity in the deep subseafloor biosphere. FEMS Microbiology Ecology 66 (2008), 181–196. [64] Inagaki F, Nunoura T, Nakagawa S, et al. Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments, on the Pacific Ocean Margin. Proc Natl Acad Sci USA 103 (2006), 2815–2820. [65] Reed DW, Fujita Y, Delwiche ME, et al. Microbial communities from methane hydrate-bearing deep marine sediments in a forearc basin. Applied and Environmental Microbiology 68 (2002), 3759–3770. [66] Inagaki F, Suzuki M, Takai K, et al. Microbial communities associated with geological horizons in coastal subseafloor sediments from the Sea of Okhotsk. Applied and Environmental Microbiology 69 (2003), 7224–7235. [67] Teske AP. Microbial communities of deep marine subsurface sediments: Molecular and cultivation surveys. Geomicrobiol J 23 (2006), 357–368. [68] Pernthaler A, Pernthaler J, Amann R. Fluorescence in situ hybridization and catalyzed reporter deposition for the identification of marine bacteria. Applied And Environmental Microbiology 68 (2002), 3094–3101.

References |

27

[69] Parkes RJ, Cragg BA, Banning N, et al. Biogeochemistry and biodiversity of methane cycling in subsurface marine sediments (Skagerrak, Denmark). Environmental Microbiology 9 (2007), 1146–1161. [70] Hoshino T, Morono Y, Terada T, Imachi H, Ferdelman TG, Inagaki F. Comparative study of subseafloor microbial community structures in deeply buried coral fossils and sediment matrices from the Challenger Mound in the Porcupine Seabight. Frontiers in Microbiology 2 (2011), 231. [71] Moore GF, Taira A, Klaus A, et al. New insights into deformation and fluid flow processes in the Nankai Trough accretionary prism: Results of Ocean Drilling Program Leg 190. Geochemistry, Geophysics, Geosystems 2 (2001), 1058.

Jennifer F. Biddle, Sean P. Jungbluth, Mark A. Lever, and Michael S. Rappé

2 Life in the Oceanic Crust 2.1 Introduction The detection of life in extreme environments and the amazing capacity of microbes to obtain energy from the environment has led to the hypothesis that heat, not water or energy sources, is the final barrier that constrains life on Earth. As such, life is expected to extend far into Earth’s crust, potentially inhabiting an environment as extensive as the world’s oceans [1]. This chapter aims to summarize what is currently known about life in the oceanic crust, which contains all three domains of life and is a vastly underexplored habitat. This is not the first summary of this topic, and additional content can be found in numerous review and primary research papers [2–16], book chapters [17, 18], and books [19–21]. The discovery of hyperthermophilic archaea and recurring geochemical evidence of crustal alteration led to the first hypotheses of a subsurface biosphere in the Earth’s crust [22]. Initial suggestions of this subsurface biosphere came from experiments at seafloor hydrothermal vents, in which elevated DNA levels in hydrothermal plumes were measured that could not be explained by simple seawater entrainment, implied that microbes were likely to live in underlying sediments and rocks. Around this time, advances were also made with experiments showing that microbe could and should grow on the Earth’s crust, furthering the concept of a subsurface crustal biosphere [23–25]. Despite the difficulty of sample retrieval, and low and heterogeneous biomass on crust samples, experimental evidence has now confirmed the existence of crustal life, consisting of a vast microbial community including viruses (reviewed in [5, 6, 13, 14, 26]). The early evidence for a crustal biosphere inevitably led to lines of inquiry intended to address questions concerning its extent, global biomass and transport and distribution. The crustal biosphere extends over a wide temperature range, with evidence for biogenic alteration in the temperature range from 15–80 °C [1]. Yet, most of the subseafloor oceanic crust is at least in theory habitable, based on measured and modeled isotherms, which show a conservative 120 °C isotherm to extend thousands of meters into the crust, which is below the predicted (150 °C; [22]) and experimentally measured temperature maxima of life (122 °C [27]). The volume of this potentially habitable zone in ocean crust equals that of the modern ocean and exceeds that within continental crust by nearly a factor of 2 [1]. Over geologic time, the creation and cooling of new crust has increased the volume that is potentially habitable. Based on compar-

30 | 2 Life in the Oceanic Crust ative cell counts, it has even been proposed that the oceanic crust might be the largest reservoir of microbial life on the planet [1]. The first evidence that life existed in the oceanic crust came from the observation of textural changes in basalts [25], as well as the measurement of DNA that came specifically from basaltic glass [24]. Since then, there has been widely documented textural evidence observed in a variety of locations (e.g. [25, 28–36]); see [37] for comprehensive review. Evidence of microbial habitation of deep crustal basalts has since even been found for eukaryotes, with fossilized fungi documented in 46 Ma old, deep basaltic rocks [38]. Energy is predicted to be available in both shallow and deep basalts, with some limitation in metabolism coming from aged basalts. Here, flow may be severely restricted by mineral precipitates in fractures within, and thick overlying sediments with low permeability that block vertical advection. Support of this notion comes from a compilation of Fe(III)/𝛴Fe ratios measured in basalts ranging in age from several hundred thousand to several hundred million years. Here increasing ratios were measured in the first 10–15 million years after formation, followed by constant ratios afterward [39]. In terms of microbial energy sources, this has been interpreted to mean that the potential for Fe(II) to serve as electron donor and/or energy source to life in subseafloor basalt may be restricted to the initial 10–15 million years after crust formation [39]. More recent data, however, indicate there to still be oxidation potential, and thus possible indigenous, microbial electron donors, 23 million years after formation in certain locations [40]. This chapter will outline some of the main topics in the study of life in the Earth’s crust, from sampling tools to evidence for life, what has not been found but may exist based on known chemical signatures, and what remains to explore and possibly discover in this expansive extreme environment.

2.2 Sampling tools Samples of ocean crust are exceptionally difficult to acquire due to the relatively impenetrable nature of the crustal environment. However, within the past few decades, a suite of tools has been developed that permit the collection of subsurface rocks and fluids from depths ranging from shallow to deep in order to study life in the oceanic crust. Life in the surficial ocean crust has been explored by a variety of mechanisms, including the collection of surface bedrock fragments [e.g. 41, 42] and hydrothermal vent fluid sampling [e.g. 43, 44]; however, these samples may not be reflective of the biosphere that is thought to reside deeper within the ocean crust [45].

2.2 Sampling tools

| 31

Table 2.1: Selected drillship- and seafloor-based basement penetration tools and sampling methods used or theoretically available for study of the ocean crust biosphere.

Wirelinebased drillship operations

Ocean Crust Sample Type

Tool(s)

Used for Subsurface Biosphere Investigation?

Max Penetration Depth (m)

Brief Overview

Example Citation(s)

Rock

DV Joides Resolution Rotary Core Barrel DV Chikyu Rotary Core Barrel

yes

>2000

8200 m max operating depth

[46–49]

yes

∼10,000𝑎

[46, 50]

Proprietary drill pipe w/ outer and inner barrel Circulation Obviation Retrofit Kit (CORK)𝑏

yes

>2800

Accompanied by riser drilling capabilities, permitting deeper drilling penetration depths; 2500 m max operating depth when performing riser drilling operations Type of rotary core barrel; operated by PETROBRAS corp. (Brazil)

yes

NA

Long-term ODP/IODP borehole observatories; multiple design inceptions: CORK, CORK-II, A-CORK, CORK-Lite; accommodate in situ experimentation and basement fluid sampling 6000 m max depth; multiple design inceptions: BMS-1, -2, -3; operated by JOGMEC (Japan)

[13, 52– 57]

Rock

Rock

Fluid

Seafloorbased drilling operations

Rock

Benthic Multicoring Systems (BMS)

yes

20

Rock

Portable Remotely Operated Drill (PROD) Multiple-Barrel Rock Coring System (MCS)

no

125

3000 m max depth; multiple design inceptions: PROD1, -2, -3; operated by Benthic Ltd

[62, 63]

no

1

[62, 64]

Rockdrill (RD2) Rovdrill

no

50

no

80

no

100

Used in conjunction with ROV and HOV platforms; operated by Debra Stakes & MBARI (USA) 4000 m max depth; operated by BGS (UK) 3000 m max depth; multiple design inceptions: Rovdrill1, -2, -3; operated by Nautilus Minerals Inc 3000 m max depth; operated by NIOT (India)

no

80

[67–69]

no

150

yes

?

4000 m max depth; operated by MARUM (Germany) Developed by Gregg Drilling & Testing Inc (USA) Inserted directly using 2000 kg coring apparatus or post-drilling; hollow stainless steel or titanium probes permit venting of fluids

Rock

Rock Rock

Rock

Rock Rock Fluid

𝑎 𝑏

[51]

Autonomous coring system (ACS) MeBo drill rig Gregg’s seafloor drill Short-length Hollow Casing Pipe

Hypothetical max penetration depth based upon length of drill pipe available CORK-Lite designs do not require use of drill-ship operations

[58–61]

[65] [66]

[67]

[67] [59, 60, 70–72]

32 | 2 Life in the Oceanic Crust 2.2.1 Tools for accessing the deep basement biosphere The exploration of life in the oceanic crust has usually involved the use of ocean drilling expeditions conducted by virtue of the Integrated Ocean Drilling Program (IODP) and its previous inceptions (i.e. the Deep Sea Drilling Program (DSDP) and Ocean Drilling Program (ODP)) (󳶳 Table 2.1; 󳶳 Fig. 2.1). In contrast to seafloor-based drilling operations, drillship-based systems are more frequently operated in the deepsea environment and permit deeper drilling penetration depths. Wireline-retrievable coring tools are often used in association with the IODP’s two principal drilling vessels, the Joides Resolution (USA) and Chikyu (Japan). The PETROBRAS corporation (Brazil) has also operated its own wireline-retrievable coring tools that have been used for deep biosphere research [51]. For a complete overview of IODP drilling and coring technology, consult Huey [46]. When combined with the low porosity of most basalts, the widespread use of circulation mud that typically contains seawater and

Fig. 2.1: Subseafloor sampling for rocks and fluids. (A) A chunk of basaltic rock from the ocean crust caught in the extended core barrell (XCB) while drilling at IODP Site U1332 (Photograph by William Crawford, IODP, taken from http://oceanleadership.org/ocean-drilling-tech-exploringseabed-history-with-600000-pounds-of-pipe/). (B) Basalt rocks in core liner after drilling IODP Site 1301 (Photograph by William Crawford, IODP). (C) Basalt rock with internal vein, a potential habitat for crustal microbes (Photograph by William Crawford, IODP). (D) Preparing a CORK for installation in Hole 395A during IODP Expedition 336 (Photograph couresy of IODP, taken from http://oceanleadership.org/scientists-look-to-microbes-to-help-unlock-earths-deep-secrets/). (E) GeoMICROBE sled positioned on a CORK observatory (Photo from http://www.soest.hawaii.edu/ oceanography/mggd/images/6{_}GeoMICROBE{_}sled{_}at{_}CORK{_}observatory.jpg). (F) Large volume fluid sampler deployed on Hole 1026 B (Photo from [55]).

2.2 Sampling tools

| 33

other amendments such as organic polymers and mineral salts prohibits the collection of uncontaminated crustal fluids from within rock cores. Nonetheless, contamination tests performed on rock samples obtained by drilling indicate that, in most cases, the interior of rocks are free of significant microbial contamination (also see next section on “Contamination tests during shipboard drilling operations”). Thus, drilling provides a suitable, and in fact, the only way by which the rock-associated indigenous microbial community within ocean crust can be sampled and studied. Downhole logging is possible using “logging-while-coring” or downhole devices such as the Simple Cabled Instrument for Measuring Parameters in situ (SCIMPI) [73, 74], although existing instrumentation is largely setup for nonbiological investigations (e.g. temperature, resistivity, etc.). However, the Dark Energy Biosphere Investigative Tool (DEBI-T) downhole profiler has been adapted for use in deep subseafloor biosphere investigations [75]. This instrument utilizes deep ultraviolet excitation to induce microbes to fluoresce differently from the organic-rock matrix material. While useful for the study of rock attached-microbes, “logging-while-coring” techniques are also not suited for investigations focused on crustal fluids. Subseafloor IODP drilling borehole observatories known as Circulation Obviation Retrofit Kits (CORKs) currently provide the best means to investigate life inside basalt-hosted, deep basement fluids (󳶳 Fig. 2.1 (D); [55]). CORKs facilitate downhole microbiological experimentation opportunities [13, 76] and, in some iterations, have been equipped with fluid delivery systems that permits the collection of pristine fluids from a re-sealed basement environment [55]. The past two decades have seen the launch of the original CORKs [52] and the evolution of newer versions (CORK-II [77]; ACORK [55, 78]). The original CORK design allowed fluids to interact with the interior of a reactive-iron casing, potentially leading to contamination effects in microbiological samples. Newer CORK-II, A-CORK and CORK-Lite versions incorporate newer features such as sample intake ports positioned away from casing material and fluid delivery lines that are made of microbiologically-friendly polytetrafluoroethylene (PTFE) material and run external to the CORK casing. While there are clear advantages to conducting in situ experiments and obtaining relatively pristine samples from the newer CORK observatories [79], old CORK installations can provide samples useful for comparative microbiological analysis [80]. CORK observatories also facilitate the collection of samples at varying temporal resolution from the same basaltic crust environment [57], which is not possible with in situ coring methods. Seafloor-based drilling and coring operations are a less expensive alternative to drillship operations; various tools have been developed and are maintained by different countries and organizations (󳶳 Table 2.1). While access to cored sample is sometimes limited due to the precious nature of the material, already cored boreholes can serve as access points to basement fluids. Although access to seafloor-based drilling samples is limited and systems are capable of considerably less deep crustal penetration than ship-based drilling, they provide a less expensive alternative that may ultimately lead to their use in complementing drilling vessels for purposes of exploring

34 | 2 Life in the Oceanic Crust the subseafloor biosphere [67]. Many of the seafloor drills listed in 󳶳 Table 2.1 have not been used for subsurface biosphere investigations; however, these tools are theoretically available. While most of the examples noted are relatively sessile seafloor-based systems, the Multiple-Barrel Rock Coring System can be used in conjunction with ROV and HOV platforms [64]. Few microbiological studies have utilized seafloor-based basement penetration methodology (󳶳 Table 2.1). The Benthic Multicoring System (BMS), operated by the Japan Oil, Gas and Metals National Corporation, has been used to generate and case a shallow borehole permitting fluid sampling and in situ microbiological experiments [59, 60, 72]. Another example is the 2000 kg coring apparatus designed by Johnson et al. [70], which does not utilize drilling, but instead relies upon the weight of the instrument to drive metal casing into exposed rocky seamount. Inserted probes contained hollow elements that allowed crustal fluid sampling for microbiology [71]. Sampling of ocean crustal fluids require only that boreholes are producing fluids and that measures are taken to maintain that borehole integrity and ensure pristine sampling. These requirements theoretically permit for a vast array of deep ocean and hydrothermal vent sampling equipment to be used for virtually all hydrothermal fluid sample access points (e.g. probes, CORKs, vents, seeps, etc.) (󳶳 Table 2.2). Sampling tools for fluids come in several types, including syringe/plunger, passive flow and active flow (󳶳 Fig. 2.1 (E), (F); 󳶳 Table 2.2). Due to the low biomass expected in the subsurface ocean crust, the most successful tools used for subsurface fluid sampling to date were capable of collecting large fluid volumes or concentrated the biomass from large volumes of crustal fluids (e.g. via in situ filtration). Passive fluid sampling using a “BioColumn” filter device [53] and downhole Osmo samplers connected in-line with flow through microbial colonization chambers have both been successful in association with CORK borehole observatories, yielding high-integrity samples for subsurface microbiology (󳶳 Table 2.2). Active fluid sampling devices, including the Hydrothermal Fluid and Particle Samples (HFPS) [71, 82], the trio of sampling devices used by Kato and colleagues [60], and the Mobile Pumping System (MPS) combined with both medium- and large-volume bag systems [98, 99] have proven highly successful in sampling subsurface fluids. Active fluid sampling systems such as the GeoMICROBE sled [53] are actively deployed for microbiological analysis; systems such as the GeoMICROBE provide the means to collect crustal fluid samples autonomously in time series fashion (󳶳 Fig. 2.1 (E), 󳶳 Table 2.2). The retrieval of rock samples from the ocean crust provides the opportunity to study microorganisms that are living in direct association with rocks, and thus likely to be involved in the physical and chemical alteration of ocean crust. Drilling conducted by the IODP vessels for the purpose of rock sample collection typically utilizes Rotary Core Barrel (RCB) technology. However, both the Joides Resolution and the Chikyu utilize slight variations of this tool called the extended core barrel (XCB), 󳶳 Fig. 2.1 (A) and the Extended Shoe Coring System (ESCS), respectively. Both the XCB and the ESCS incorporate features that enhance penetration and recovery of basement rock; however,

2.2 Sampling tools

| 35

Table 2.2: Selected ocean incubation and crust sample collection tools used or relevant to deep biosphere investigations. Sample Type

Tool Type

Tool(s)

Brief Description

Example Citation(s)

Miniature Porewater Sampler inside IODP Advanced Piston Corer (APC) Titanium gas-tight sampler

Never used really worked; not theoretically adapted for basalt rock Used for high-temperature fluid sampling in situ sample pressure Small volume fluid samples (1–10 mL) for discrete sampling

[46]

Used for in situ enrichment at hydrothermal vents; never used for deep subsurface biosphere research Multiple day in situ filtration of positively-pressured CORK borehole fluids Connected to CORK wellhead L port via flexible sampling hose Can be deployed in conjunction with CORK and SmartPlug/GeniusPlug observatory system; utilized largely for rock chip in situ incubations Capable of in situ DNA and protein preservation; time-series sampling permits daily resolution for > 1 year

[43, 84– 87]

Used for hydrothermal vent fluid sampling only; can be equipped with either Titanium syringes or gas-tight samplers as fluid collection devices Features titanium intake nossels and multiple temperature probes Used for hydrothermal vent fluid sampling only; capable of in situ filtration and concentration of 10 L of fluids Teflon tube placed inside titanium pipe inserted into ocean crust Never used for deep subsurface biosphere research; feature titanium intake nossels; handles filtered and unfiltered samples Pump composed of Teflon parts; features multiple nossels, including titanium-based Never used for deep subsurface biosphere research; capable of fluid filtration, in situ quantitative PCR and a variety of other functions

[92]

Fluid Syringe/ plunger sampling

SIPPER acrylic sampler

[81, 82] [83]

Passive fluid sampling In situ vent devices

BioColumn in situ filter Titanium bottle with hose Osmo Sampler w/ Flow through Osmo Colonization Systems (FLOCS)

Osmo Sampler w/ Biological OsmoSampling System (BOSS)

[53]

[88] [89–91]

[44, 89]

Active fluid sampling Submersible-Coupled in situ Sensing and Sampling System (SIS)

Hydrothermal Fluid and Particle Sample (HFPS) Hydrothermal Vent BioSampler (HVB)

Teflon tube and impellor Autonomous Microbial Sampler (AMS)

KIPS fluid sampling system

Deep-sea Environmental Sample Processer (D-ESP)

[71, 82] [93]

[72] [94]

[95]

[96, 97]

36 | 2 Life in the Oceanic Crust Sample Type

Tool Type

Tool(s)

Brief Description

Example Citation(s)

Rotary clean seawater sampling system (ROCS)

Six 500 mL polycarbonate, Teflon and silicon cylinders; sample performed using a peristal pump Used to collect up to 10 L of fluids; utilized ROPOS suction pump Polyethylene bag system capable of collecting up to 20 L; utilized ROPOS suction pump Used to collect up to 70 L of fluids; handles filtered and unfiltered samples

[60]

Standard basement core collection tool of Joides Resolution and Chikyu; used often with catchers to increase core retainment Variation of RCB employed on DV Chikyu; used often with catchers to increase core retainment; relatively new tool in IODP arsenal

[46–49], [100]

Osmo Sampler w/ colonization systems and microbial flow cells

Rocks and minerals can be incubated in situ inside CORK borehole observatories

[13, 76, 90, 91]

Autonomous in situ Instrument Colonization System (AISICS) and fluid sample probe

Rocks and minerals can be incubated in situ; developed specifically to study colonization processes; never used for deep biosphere research

[101]

Large-scale fluid (LF) stainless steel sampling system Large volume bag sampling system (BAG) GeoMICROBE/Mobile Pumping System (MPS), Medium-Volume Bag System (MVBS), and Large Volume Bag System (LVBS)

[60] [60]

[57, 98, 99]

Rock Basement coring Rotary Core Barrel (RCB)

Extended Shoe Coring System (ESCS)

In situ deployable experiments

[46, 50]

the ESCS contains upgrades that make it superior to the XCB [46]. Huey [46] provides a complete overview of IODP basement coring technology.

2.3 Contamination Monitoring of sample contamination during drilling and subsequent operations is essential to downstream microbiology and molecular biological studies due to the large potential for cell and nucleic acid contamination of subseafloor samples with cells and nucleic acids from seawater and chemical compounds pumped into boreholes during drilling operations [102].

2.3.1 Contamination induced during drilling Two methods have traditionally been used to monitor contamination in sediment and rock cores during ocean drilling expeditions: fluorescent microbeads (microspheres) and chemical (perfluorocarbon (PFC)) tracers [103–105]. Both methods have their

2.3 Contamination

| 37

strengths and shortcomings. Fluorescent microbeads are suspended in distilled water and placed in a sealed plastic bag that is subsequently mounted to the core-catcher sleeve prior to advancement into the borehole. This bag breaks open during drilling operations, thereby releasing microbeads into drilling fluids. The advantage of this method is that appropriately sized microbeads (0.5–1.0 μm diameter) are likely to be good analogs for the transport of (contaminant) microbial cells in fluids during drilling operations. Detection and quantification is, moreover, methodologically straightforward and can be performed by epifluorescence microscopy – even months after drilling expeditions, on frozen or fixed samples. A pitfall is that homogeneous concentrations of microbeads cannot be achieved by bag breakage and microsphere spillage into drilling fluids. As a result, certain parts of drilling fluids may contain very high concentrations of microbeads, whereas other parts only have low concentrations or may even be devoid of microbeads [105]. Thus, though visual observation of microbeads is evidence for drilling fluid contamination, the absence of microbead observation is not reliable evidence for its absence. A more quantitative and sensitive method is the use of chemical tracers. During past ocean drilling expeditions, the PFC compound perfluoromethylcyclohexane (PMCH; C7 F14 ) has been used. This tracer is injected into the drilling mud stream on board, to a constant concentration of 1 mg l−1 . From there, this mud stream is pumped directly into the borehole to the drill bit. Due to low detection limits (parts per trillion level), PFC tracers offer a highly sensitive method by which drilling fluid intrusion in the nanoliter range per cm3 of sample can be detected [106]. PMCH is also less volatile than initially thought, indicating that loss due to volatilization during drilling operations, sampling, or short-term sample storage on board will not result in significant PFC loss from samples, and thus underestimation of contamination in samples [106, 107]. Potential shortcomings of the PFC method are, however, that samples need to be taken within three hours after core sampling to avoid diffusion of PFC from contaminated core exteriors to clean interiors. Otherwise, false positives might occur, since microbial cells diffuse at a vastly lower rate than PFC. Moreover, recent data from IODP Expedition 337 indicate that PFC can sorb to drilling mud components, sediment and/or rock material. If confirmed, drilling fluid contamination estimated by PFC measurements might have resulted in underestimation of drilling fluid contamination in some cases. This problem of incomplete PFC release during headspace incubation has been reduced through modifications of the PFC incubation protocol [107]. Further tests are currently taking place in the context of the IODP Expedition 347 [108]. Both microspheres and PFC have been used to quantify drilling fluid contamination during drilling operations into ocean basement [104, 106, 109]. Both tracers were quantified on exteriors and in interiors of gabbro cores in the Middle America Trench [109] and indicated high contamination of core exteriors and vastly lower, though still significant, contamination in most core interiors. These findings were partially confirmed by the studies of Smith et al. [104] and Lever et al. [106] on basalt cores, which also demonstrated high contamination of outer rock surfaces. Contrary to the

38 | 2 Life in the Oceanic Crust study on gabbro cores, however, these studies found the PFC content in the interiors of intact rock samples to be very low or even below detection. There was no significant difference, i.e. drilling fluid contamination within basalt cores was not higher, compared to sediment cores obtained by the advanced piston coring technique. A key difference between the study by Morris et al. [109] and the studies by Smith et al. [104] and Lever et al. [106] might be that, in the latter two studies, additional treatments besides washing with water (i.e. washing with methanol or flaming) were used to remove tracer from core exteriors. After 2× washing with artificial seawater, Lever et al. [106]) found PFT on rock exteriors to only have decreased by ∼80%, compared to decreases of ∼99% after additional flame treatment. Without this additional treatment, which was equivalent to flame-sterilizing rock exteriors, the risk of cross-contaminating rock interiors through contact with still highly contaminated rock exteriors during rock crushing was vastly higher. This potential problem was also pointed out by Morris et al. [109] who acknowledge that insufficient removal may have led to transfer of PFT and microspheres from rock exteriors to rock interiors while sampling the latter. Thus, taking together the published data on drilling fluid contamination, it appears that the interiors of intact basalt rocks are only minutely contaminated during the drilling process and thus suitable for microbiological analyses.

2.3.2 Contamination during fluid sampling The sampling of subsurface crustal fluids carries contamination challenges independent of those that can occur during drilling due to the low amount of biomass contained in fluids: fluids emanating from CORK boreholes appear to contain microbial cells that are roughly an order of magnitude less abundant than those in bottom seawater[57]. Thus, from a microbiological perspective, even a small percentage of bottom seawater entrainment during sampling is magnified due to the stark contrast in cell abundance.

2.4 Direct evidence for life in the deep ocean crust When discussing microbial communities associated with hard mineral substrata, such as ocean crust, it is important to distinguish between (a) a free-living/planktonic component that lives within circulating fluids and most likely feeds on dissolved chemicals, (b) an epilithic component that lives attached to rock minerals, e.g. as biofilms, and feeds on dissolved chemicals that flow by in circulating fluids or diffuse out from within rocks and (c) an endolithic component, i.e. microbes that live inside the actual rocks [110]. The planktonic community may consist of a nonindigenous component, that may be entrained through bottom seawater entering the crust or from overlying sediment, in additon to an indigenous component of microbes. The

2.4 Direct evidence for life in the deep ocean crust |

39

latter are either adapted to life within crustal fluids, or are microbes that typically grow attached but temporarily go into planktonic stage, e.g. to disperse. The epilithic community is often overlooked in deep biosphere studies, both due to extremely low recovery of rubble and brecciated rock layers by extant technologies of ocean drilling [77] and the need to sterilize outer surfaces of rocks when drill cores are retrieved [104, 106]. However, biofilms are able to be sensed by downhole tools [75] and are often seen on surficial crust [111], so epiliths are certainly part of the crustal biosphere. Endoliths can be further separated into categories of chasmoendoliths (inhabiting fissures and cracks), cryptoendoliths (living in porous rocks) and euendoliths (actively penetrating into rock) [110].

2.4.1 Textural alterations The earliest, and perhaps most controversial evidence for life in subseafloor basalt comes from observed textural alterations. Textural alterations to basalt rocks that are most visible, even millions of years after their formation, are in most cases produced by euendoliths, which etch tracks and produce fossilized biosignatures, such as borings [25, 30, 112, 113]. While abiotic alterations occur, they can be distinguished from biotic alterations due to differences in banding and appearance. For example, altered basaltic glass from Sites 504B and 896A were described as being abiotically altered from sideromelane to palagonite when zoned alteration bands of regular thickness were seen, compared to the bulbous protusions observed during biotic alteration [114]. Additionally, nucleic acid stains were colocalized with alteration zones at Site 896A, showing the presence of DNA and also the observation of bacteria and a few archaea at alteration fronts [30]. To date, petrographic observations have been utilized to determine the majority of biotic alteration, where structures are examined by eye and are noted as biotic when they are irregular, bulbous, rooted in fractures and colocalized with carbon and DNA signatures [30]. Compiled data show that despite geographic location, the majority of biotic alteration occurs above 300 meters below seafloor (mbsf), with a significant decline below 400 mbsf. While it could be expected that biotic alteration is controlled by fractures within the glass, it is weakly correlated with the fracture density of glass, since fracture density is not the same as permeability, which should control the overall movement of fluids needed for biology [30]. A curious observation of biological alteration is that the degree of alteration does not change significantly across crusts of different ages. This suggests that either the majority of alteration occurs soon after crust formation and is preserved over geologic time, or that bioalteration happens continuously, but is covered up by diagenesis over time [30]. Because of these uncertainties, it is often impossible to date when an alteration has taken place. Also unclear in many petrographic observations is whether or not alteration was caused by prokaryotes or eukaryotes. Fine scale investigations of some filamentous microstructures suggest that fungi may be the cause of microfossils [38, 113, 115].

40 | 2 Life in the Oceanic Crust 2.4.2 Geochemical evidence from fluids Chemical and isotopic compositions of formation fluids emanating at crustal outcrops, obtained from CORK observatories, or mixed with sediment porewater and sampled from sediments directly overlying the basement, provide important insights to the presence and activity of microbes deep within ocean crust. Most research has been on the concentrations of electron acceptors in basaltic fluids of the Eastern flank of Juan de Fuca Ridge. Here, fluids from the seafloor outcrop Baby Bare Springs (BBS) and borehole 1026B are fully depleted of oxygen and nitrate and show decreased concentrations of sulfate, dissolved organic carbon (DOC) [116] and dissolved inorganic carbon (DIC) compared to bottom seawater sampled in the region. These trends are confirmed by analyses on porewater in sediments directly overlying the basaltic basement [117]. The removal of DIC during fluid flow through crust can be explained with chemical precipitation of carbonates in basaltic veins [118]. Purely chemical processes, such as sulfide and iron oxidation within basalt, combined with fluid mixing with O2 -depleted porewater from and microbial O2 respiration within overlying sediments, could account for the depletion of O2 . Yet, reaction path models of fluid flow through basalt at another location (North Pond), render this unlikely as a sole explanation, and suggest that microbial O2 respiration within basalt is likely to account for at least part of the O2 consumption [119]. The full depletion of nitrate during circulation through crust is even less likely to have taken place without the involvement of microbes within crust. Under the in situ conditions, nitrate reduction is likely to be a strictly microbial process. If microbial nitrate reduction were restricted to sediments, then this would require dilution to extinction of fluids circulating through crust with overlying, nitratedepleted sediment porewater – a possible, but unlikely scenario, given that the low permeability of overlying sediments largely limits fluid exchange between crust and sediment to chemical diffusion. Not taking into consideration data from rocks in the same location, the partial depletion in sulfate and DOC could, on the other hand, be attributed to mixing with sulfate-depleted sediment porewater and/or sediment microbial activity. Though most data obtained so far are from the Juan de Fuca Ridge Flank, this does not mean that the same microbial processes are not likely to take place elsewhere within the oceanic crust. Analyses on fluids from the Dorado outcrop on the Cocos Plate, in which both nitrate and sulfate are depleted relative to seawater, confirm the likelihood of microbial nitrate and sulfate reduction within ocean crust [40]. Interestingly, while high-temperature fluids emanating at unsedimented sites on the Juan de Fuca Ridge Crest in the Main Endeavor Field and Axial Volcano (mean: 15 and 17 μM) also have reduced DOC concentrations compared to seawater, low-temperature diffuse fluids from these same locations are enriched in DOC (47 and 48 μM, respectively), and cell numbers [116]. The extent to which the increased DOC concentration

2.4 Direct evidence for life in the deep ocean crust

| 41

and cell counts result from biomass production within the ridge crest, and are fueled by energy derived from rocks, is unknown. Isotopic analyses on formation fluids, obtained from seafloor springs and borehole observatories, provide valuable information on the presence and biochemical pathways of microbial activity deep in oceanic crust. Values of 𝛿13 C-DOC from borehole 1026B and BBS are 5.0 and 13.4 per mil lower than those of bottom seawater in the region [120]. This indicates that the composition of DOC is altered during fluid flow through crust, and acquires a significant chemoautotrophic component, which is in all likelihood produced by microbes that are indigenous to the basalt and use the reductive acetyl CoA pathway for C fixation [49, 120]. Interestingly, formation fluids from Axial Seamount, Easter Island, and West Grotto showed a different trend: 𝛿13 CDOC of fluids leaving the seafloor were higher than those in bottom seawater (Axial: −18.6‰; Easter Island: −19.4‰; West Grotto: −18.4‰) [120]. The reasons are unclear, and may in part include heterotrophic sulfate reduction and methanogenesis, as well as fermentation reactions, which can all increase the 𝛿13 C of residual organic matter [121–123], C fixation by the reverse tricarboxylic acid cycle, which produces 𝛿13 Cvalues in this range [49, 124], or simply mixing with an organic carbon (OC) source that has a higher 𝛿13 C. In addition to carbon isotopic analyses, 𝛿34 S-compositions of sulfate have been measured in porewater extracted from sediments directly overlying the basement across a longitudinal transect of four sites spanning the Eastern flank of the Juan de Fuca Ridge. Since microbial sulfate reducers select against the heavier sulfur isotopes and thus increase the 𝛿34 S of residual sulfate, analyses on 𝛿34 S-sulfate provide clues to microbial sulfate reducing activity. Compared to the 𝛿34 S-sulfate of +21.1‰ in seawater, all sites showed elevated 𝛿34 S-sulfate in porewater from bottom sediment, with boreholes located near the ridge crest only showing minor increases of 0.6– 1.7‰ (1023B: +22.8‰, 1025B: +21.7‰), whereas boreholes further away from the spreading center had increases in 𝛿34 S-sulfate of 9.3–10.1‰ (1026C: 30.4‰, 1028A: 31.2‰) [125]. These elevated 𝛿34 S-sulfate values in bottom sediments, which result from diffusive mixing of basalt fluids and sediment porewater, are evidence of microbial sulfate reduction in basement basalt.

2.4.3 Geochemical evidence from rocks A large body of literature has been published on the chemical and isotopic composition of basalt rocks, with many studies reporting indications of past or present microbial activity. A considerable focus has been on quantifications of iron, sulfur and carbon species, and analyses of their respective stable isotopic compositions. These data allow inferences concerning both the macro- and microscale distribution of microbial activity deep within the oceanic crust. Comparisons between rocks across boreholes

42 | 2 Life in the Oceanic Crust and between sites enable the identification of vertical and horizontal trends in redox conditions and fluid flow across distances of meters to kilometers. In contrast, smallscale comparisons within the same rocks, on the other hand, provide clues to variation in redox conditions on a centimeter to micrometer scale. Dramatic changes in oxidation state have often been observed when going from veins to adjacent rock, or from distinctly colored (often oxidized) bands that surround veins (halos) to parts of rocks that are further away from veins and hence flow of seawater-derived oxidants [49].

2.4.3.1 Iron Iron represents a large elemental fraction of subseafloor basalt. On the Juan de Fuca Ridge Flank, iron alone accounts for on average ∼7.3 weight percent of bulk basalt. Most iron is in the form of Fe2+ , which is a potential microbial electron donor, and may thus be the quantitatively most important driver of chemoautotrophic carbon fixation in most subseafloor basalt environments. Though not direct evidence of microbial activity, bulk analyses of Fe(III)/𝛴Fe provide indications of whether basaltic crust has been oxidized since its formation. The oxidation of Fe(II) to Fe(III) first occurs during cooling and solidification after basalt formation. During this time, seawater reacts with Fe2+ from basalt to form magnetite (Fe3 O4 ) and H2 [126, 127] and oxidizes on average 15 ±5% of total iron [39]. During the next 10–15 million years, Fe oxidation continues, reaching an Fe(III)/𝛴Fe of 45 ±15% as a final stage, after which oxidative alteration is largely absent [39, 40]. During these 10–15 million years, iron may be oxidized by several processes: (1) abiotically/biotically with O2 as electron acceptor, (2) biotically with nitrate as electron acceptor, (3) abiotically by reactions with anoxic seawater, e.g. serpentinization [39], and (4) electron scavenging by microbes using sulfate and/or DIC as electron acceptors [128]. O2 and nitrate are probably important oxidants as long as the basement is directly exposed to bottom seawater (as, e.g. on the Juan de Fuca Ridge)[33] or only covered by a thin, permeable blanket of sediment, which allows vertical advection (e.g. North Pond) [129]. Heat-driven circulation drives bottom seawater containing O2 and nitrate through upper crust and results in abiotic oxidation of Fe2+ by O2 , or biotic oxidation with O2 or nitrate as electron acceptors [39]. As sediment thickness increases over time, vertical advection is blocked. Heatdriven fluid flow through basalt either comes to an end, or – provided there are conduits for fluids to enter and exit – continues, albeit in a horizontal direction, as on the Juan de Fuca Ridge Flank [130, 131]. In the absence of vertical advection, the electron acceptors O2 and nitrate are removed early along the flow path [53, 132]. Yet the Fe(III)/𝛴Fe continues to increase in basalt where O2 or nitrate are absent [39]. Altogether ≥ 50% of iron oxidation occurs after initial crust formation. As with other geochemical measurements, much of the published data on iron minerals in subseafloor basalt comes from the Juan de Fuca Ridge Flank. The anoxic,

2.4 Direct evidence for life in the deep ocean crust

| 43

nitrate-depleted 3.5-million-year-old crust at U1301 contains large amounts of olivine [133]. This iron mineral is known to react with water to produce H2 or small organic molecules such as formate, methanol, or methane via Fischer-Tropsch-Type (FTT) synthesis involving CO2 at temperatures in the range found at U1301 [134–136]. Hence, the significant amounts of olivine at this site may explain the presence of key genes of sulfate reduction and methane-cycling, as well as isotopic values indicative of C fixation [49]. In situ sulfate reducing microbes and methane-cycling Archaea may use H2 to chemolithoautotrophically reduce DIC to OC, and both H2 and/or small organic molecules as energy substrates [49, 137, 138]. In addition to abiotic reactions, biotic iron oxidation in the absence of O2 or nitrate cannot be ruled out. Studies with anoxic media have invoked a role of sulfate reducers and methanogens in iron oxidation via a still poorly understood mechanism termed cathodic depolarization [128, 139] In these incubations, members of both groups showed strong growth after addition of metallic iron to media. For certain strains, this growth response was faster than after H2 addition, suggesting that certain sulfate reducers and methanogens are able to directly harvest electrons released by iron oxidation [128]. In addition to known microbial metabolic reactions, major gaps in our knowledge of microbial Fe cycling cannot be ruled out. For instance, it is possible that microbes are involved in, and perhaps even gain energy by catalyzing Fe oxidation reactions that are currently considered abiotic, such as serpentinization.

2.4.3.2 Sulfur Mineralogical and stable isotopic analyses have been conducted on sulfur pools within subseafloor basalt and provided important insights to oxidation state and microbial sulfur cycling. The relative contributions of acid-volatile sulfide (AVS), chromium-reducible sulfur (CRS), and sulfate-S (SO4 -S) to total sulfur have been quantified and used to infer redox conditions or provide relevant background information to the interpretation of geochemical and microbiological data [49, 140, 141]. Samples with high proportions of SO4 -S (mainly dissolved or as anhydrite (CaSO4 )) that coincide with high Fe(III)/𝛴Fe ratios in rocks and secondary minerals indicate high oxidative alteration of basalt [49]. Hereby, oxidation of reduced sulfur species, e.g. sulfide, can occur with Fe3+ as electron acceptor. In the absence of O2 or nitrate, this reaction occurs abiotically [142, 143] Microbes can then disproportionate the products of this abiotic oxidation, e.g. elemental sulfur, to obtain energy for growth [144]. In other cases, they may simply indicate precipitation of seawater sulfate, which can account for 75% of sulfate found in basalt [140]. Samples dominated by reduced S species (AVS, CRS) can indicate reducing conditions in the absence of microbial activity since basalt formation, or addition of sulfide to basalt by microbial sulfate reduction. In terms of identifying the origin of these sulfur pools, measuring pool size or distribution is typically insufficient to infer

44 | 2 Life in the Oceanic Crust microbial activity. To infer microbial processing it is necessary to determine S stable isotopic composition. Among the first clear evidence of past or present microbial activity came from analyses on sulfide grains from ODP Site 801 on the Pacific plate east of the Mariana trench [148], and confirmed earlier reports of highly 34 S-depleted bulk sulfides from the Iberian Margin (ODP Site 637A) [140]. At ODP Site 801, the 𝛿34 S-composition of pyrite and marcasite grains was determined using an ion microprobe. Values ranged from 0‰, indicative of an unaltered mantle source, to values below −40‰, which point clearly to the addition of sulfide produced by microbial sulfate reduction. These measurements were confirmed by conventional mass spectrometry, as well as analyses on the bulk sulfur species (AVS and CRS). More recently, further studies have confirmed the presence of S species with low isotopic values in subseafloor basalt [49, 141, 146]. Ono et al. [146] found evidence for elevated sulfate reduction in peridotite minerals compared to basalt. This was consistent with the higher H2 production potential of peridotite, which may be a main indigenous electron source for microbial sulfate reducers. Lever et al. [49] detected 𝛿34 S-pyrite with values below −70‰ at IODP Site 1301B on the Juan de Fuca Ridge Flank. These values are in the range of the strongest 34 S-fractionations ever recorded for sulfate reducers [147] and could even indicate a full oxidative and reductive S cycle with multiple cycles of oxidation and reduction having produced these very low values [144]. These analyses on S mineral and isotopic compositions in subseafloor basalt not only provide clues to the microbial processing and alteration of basalt sulfur chemistry, but – given that oceanic basement is a globally important S sink and source [49, 140, 148] – will in the future provide key insights to the role of basalt microbes in the global S cycle in the future.

2.4.3.3 Carbon Subseafloor basalt is a globally significant sink for inorganic carbon from seawater, which is precipitated to form secondary minerals that coat vein surfaces [149, 150]. While neither the precipitation of carbonate, nor its 13 C-isotopic composition in most cases requires biotic processes to be explained [48, 118, 150], measured values are at least consistent with partial biological processing in some cases. For instance, the carbonate values with the lowest 𝛿13 C at IODP Site 1301B (−5.1‰) were measured in a vein that was retrieved in close proximity to rock samples with both the highest OC values of the entire borehole and genes diagnostic of sulfate-reducing, methanogenic and methane-oxidizing microorganisms. Such lower 13 C-carbonate values would be expected if biological remineralization of OC, which has lower 𝛿13 C-values than dissolved inorganic carbon (DIC) from seawater, was contributing a significant amount of the IC precipitated [49]. Data on organic carbon content and stable isotopic composition in subseafloor basalt is so far limited, in part due to the difficulty of analyzing the low quantities of

2.4 Direct evidence for life in the deep ocean crust

| 45

OC present, which are frequently below 0.1 weight % [49, 151–153]. The sources of OC vary with the location, but most likely include FTT synthesis, biological synthesis by indigenous organisms, thermogenic synthesis, entrainment of OC from overlying sediment, and import from seawater [151]. Due to overlaps in 13 C-isotopic fractionations produced by different abiotic, thermogenic and biotic pathways, biological signatures cannot always be identified with certainty. Similarly, due to overlaps in 13 C-values produced by different chemoautotrophic and photoautotrophic pathways, it can be impossible to distinguish OC of allochthonous photoautotrophic origin from OC of autochthonouse chemoautotrophic origin. Furthermore, most studies so far have not included genetic data, which would help narrow the possibilities of interpretation. Values measured across a range of oceanic serpentinites range from −8 to −28‰ suggesting a wide range of sources, with no clear indications of their origin [151–153]. On average, values in gabbros are lower, ranging in 𝛿13 C from −15 to −32‰; here especially the lower values (≤ −30‰) are consistent with chemoautotrophic synthesis via the reductive acetyl CoA pathway, suggesting carbon fixation by basalt microbes via use of indigenous electron donors. The strongest case for chemoautotrophy in subseafloor basalt so far comes from the Juan de Fuca Ridge Flank. Here 𝛿13 C-TOC values below −30‰ were measured in the same or adjacent rock samples to ones in which the combination of metabolic marker genes, 𝛿34 S-pyrite and laboratory-based incubations had demonstrated the presence of active, potentially autotrophic sulfate-reducing and methane-producing microbes [49]. Future studies, which combine C-isotopic analyses, surveys of metabolic and C fixation genes and cultivation experiments will help reveal the extent to which chemolithoautotrophy forms the foundation of basaltic food webs worldwide – and provide insights to samples in which 𝛿13 C-OC analyses alone are inconclusive in resolving the origin of the OC.

2.4.4 Genetic surveys 2.4.4.1 Basalt and gabbro To date, only a limited amount of genetic data has been published on microbial communities inhabiting deeply buried ocean basement. In addition to several studies on chasmoendoliths that inhabit inner surfaces with rocks [49, 154], several studies have looked at communities within solid rocks, which had been powderized and used for the extraction of DNA [48]. The difficulty of performing this work is underscored with the failures of certain efforts to yield interpretable results, possibly due to the lack of clean sampling [47]. The first published study on gabbroic layers was on rock samples from the Atlantis Massif obtained during IODP Expeditions 304 and 305. 16S rRNA gene sequences belonging to Alpha-, Beta-, and Gammaproteobacteria were detected, in addition to a diverse range of functional genes indicative of in situ C-, S-, and N-cycling [48]. Close relatives of many of the 16S rRNA gene sequences detected are also frequently found

46 | 2 Life in the Oceanic Crust in marine water columns, raising the question as to whether these microbes were indeed native to the gabbro environment, or derived from bottom seawater entering at the seafloor and circulating through cracks in the gabbro. Further evidence for the existence of microorganisms indigenous to the deep ocean crust derives from two more recent studies by Lever et al. [49] and Orcutt et al. [154], both on samples from the Juan de Fuca Ridge Flank, sampled during IODP Expedition 301 and IODP Expedition 327, respectively. Lever et al. [49] documented functional genes of methane-cycling Archaea (mcrA; uncultivated Methanosarcinales, ANME-1 Archaea) and sulfate reducers (dsrB; Cluster IV) at several depths in basalt fissures within the basement of IODP Site 1301. Members of these uncultured groups of methane-cycling Archaea and sulfate reducers are widespread in anoxic marine sediments, but typically absent from seawater or human wastewater, making them unlikely to be introduced during drilling operations. Comparisons of functional gene data to stable isotopic compositions of total organic carbon (TOC), carbonate, bulk S pools and individual pyrite granules showed a good match between genetic distributions and indicators of microbial activity. These analyses were complemented by data from multi-year laboratory incubations of basalt pieces. In several of these incubations, methane with a 𝛿13 C composition indicative of microbial methanogenesis was produced, providing strong evidence for an alive, anaerobic in situ microbial community. These data on microbial C- and S-cycling microbial communities have since been confirmed by 16S rRNA gene tag pyrosequencing on rock interiors from the Juan de Fuca Ridge Flank. Here, rock interiors show only insignificant phylogenetic overlaps in bacterial and archaeal community composition with positive controls for drilling fluid contamination and seawater [154]. Moreover, the dominant sequences detected in rocks differ strikingly from those found in overlying sediment. Rocks from the Juan de Fuca Ridge Flank during IODP Expedition 327 were uniformly (to ≥ 90%) dominated by Actinobacteria. By comparison, sediments overlying subseafloor basalt in the same locations were dominated by Chloroflexi, which on average accounted for ∼45–70% of bacterial sequences. While archaeal abundances in sediment were typically about one order of magnitude lower than those of Bacteria, they were three orders of magnitude lower than those of basalt-dwelling Bacteria, and near the detection limit, in basalt. What drives these general differences in microbial community composition is currently unknown. Due to the near-uniformity of temperature between the bottom sediment layers and basalt layers analyzed, temperature is unlikely to have played a role. Bottom sediments as well as circulating fluids also appear to have similar sulfate and dissolved inorganic carbon concentrations. Thus it is plausible that differences in electron donor availability between sediments and basalt are responsible for the divergent communities between sediment and basalt.

2.4 Direct evidence for life in the deep ocean crust

| 47

2.4.4.2 Crustal fluids via borehole observatories Borehole CORK observatories have been used to access microorganisms inhabiting fluids originating from 1.2 and 3.5 million-year-old ridge flank of the Juan de Fuca Ridge and 6.5 million-year-old ridge flank on the Costa Rica Margin (󳶳 Table 2.3). All three locations represent young, hydrologically-active basaltic crustal environments covered by thick (> 100 m) blankets of sediment, which locally restricts fluid circulation into the ocean basement. It is worth the emphasis here that little to no fluid circulation occurs across the sediment-basement interface at these locations despite the adjacency to deep overlying sediments, effectively sealing off a basement microbial community from all environments directly above. Through the use of water sampling and subseafloor borehole colonization experiments, CORK observatories have produced genetic data on the microbial composition of subsurface fluids. Despite the few locations available for this work, there have been overlaps in the microbial communities at the Costa Rica and Juan de Fuca sites (󳶳 Table 2.3). Lineages within the bacterial phyla Firmicutes, Marinimicrobia, Proteobacteria, Planctomycetes, Verrucomicrobia and Bacteroides were documented at all three sites (󳶳 Table 2.3). Despite being sampled nearly a decade apart, CORKs 1026B and 1301A along the Juan de Fuca Ridge Flank showed similar communities (e.g. Desulforudis, Miscellaneous Crenarchaeotic Group). Furthermore, sampling conducted within a single year at boreholes 1025C and 1026B revealed shared lineages (e.g. Desulfobulbus) [80]. Several frequently detected microbial lineages in borehole fluid samples appear to be plausible inhabitants of the deep subsurface based on close phylogenetic relationships with cultivated lineages of chemolithotrophs (e.g. Sulfurimonas, Thiomicrospira). Gene clones phylogenetically related to uncultivated groups are common in borehole fluid samples; these lineages are frequently related to gene clones described from hydrothermally-influenced marine or terrestrial subsurface environments (e.g. Parvarchaeum, Desulforudis, Aminicenantes). Some potential metabolic traits have been described for a handful of these lineages based on genomic analysis [156, 157], though the majority of lineages listed in 󳶳 Table 2.3 are known only by a phylogenetic marker gene sequence. In addition to high quality fluids from subsurface aquifers, CORK observatories also allow the incubation of materials for colonization from subsurface fluids. In situ observatory systems were placed on the Juan de Fuca ridge for four years, after which they were retrieved and analyzed for colonization [13]. Some microbial lineages from colonized rock chips were closely related to others from sampled fluids, suggesting that some members of the pelagic community may be able to settle and grow on an available surface. Other subsurface flow-through experiments, also from the Juan de Fuca, yielded microbes that were able to grow in lab, after enrichment in situ, from Gammaproteobacteria, Alphaproteobacteria and Actinobacteria phlya [76]. CORK observatories have also yielded samples of opportunity, particularly during unintentional in situ enrichment. During replacement of the CORK affixed to Juan de

48 | 2 Life in the Oceanic Crust Table 2.3: Phylogenetic diversity of small subunit ribosomal RNA genes recovered from multiple samples of fluids, environmental isolates, or subseafloor colonization experiments involving ocean crust boreholes.

Phylogenetic affiliationa Archaea Euryarchaeota Archaeoglobales Parvarchaeum Methanococcales𝑓 Methanomicrobia Misc. Euryarchaeotic Group (MEG) Thaumarchaeota Marine Group I Misc. Crenarchaeotic Group (MCG) Bacteria Actinobacteria Microbacteriaceae Aminicenantes𝑔 (OP8) Bacteroidetes CS_B045 Flavobacteriaceae Lutibacter NB1-m NS9 Calescamantes𝑔 (EM19/KB1) Cyanobacteria Synechococcus Firmicutes Caloranaerobacter D8A-2 Desulforudis/Ammonifex Desulfotomaculum Peptococcaceae RF3 Fusibacter Marinimicrobia𝑔 (SAR406) Plactomycetes OM190 Phycisphaeraceae Proteobacteria Alphaproteobacteria Hyphomicrobiaceae Hyphomonadaceae OCS116 Phyllobacteriaceae Rhodobacteraceae IndB1-2 Leisingera SAR11 SB1-18 Sphingomonadaceae Betaproteobacteria Alcaligenaceae Burkholderiaceae Oxalobacteraceae

Costa Rica Rift flank 896Ab

Juan de Fuca Ridge flank 1025Cc 1026Bd 1301Ae

− − − − −

− − − − −

++ + + + −

+ + − + ++

−

−

+ ++

+ +++

− −

− −

+ −

+ ++

−

−

−

++

− + + −

− − − −

− − + −

++ + − ++

−

−

+

++

− − − − − − − −

− − + + − − − +

++ + ++ ++ ++ + + +

+ + ++ − − ++ ++ +

− +

− −

+ −

+ +

− + ++ +

− − − −

+ − + −

+ + − +

+ + − + −

− − + − −

− − + − +

+ + ++ + ++

− − −

− − −

+ ++ −

+ + ++

2.4 Direct evidence for life in the deep ocean crust |

Phylogenetic affiliationa Deltaproteobacteria Desulfobacula Desulfobulbaceae Desulfobulbus Desulfocapsa MSBL7 Desulfohalobiaceae Desulfovibrionaceae Desulfuromonadaceae Nitropinaceae SAR324 Epsilonproteobacteria Sulfurimonas Gammaproteobacteria 9NBGBact_8 Alcanivoracaceae Alteromonadaceae Marinobacter Colwelliaceae Ectothiorhodospiraceae Halomonadaceae JL-ETNP-Y6 JTB35 Methylophaga Moraxellaceae Oceanospirillaceae OM182 Pseudoalteromonadaceae Pseudomonadaceae Xanthomonadaceae Thermomonas Thiomicrospira ZD0405 Thermotogae Kosmotoga Thermosipho Verrucomicrobia Arctic97B-4 Roseibacillus Eukaryota Collodaria 𝑎

Costa Rica Rift flank 896Ab

49

Juan de Fuca Ridge flank 1025Cc 1026Bd 1301Ae

−

−

+

+

− − − − − − − −

+ + − − + − − −

+ + − ++ − + + +

− ++ ++ − + + + +

+

−

+

+++

+ −

− −

+ +

+ ++

− − + − − − + − − − − −

− − − − − − − + − − − −

++ − − − + + + − − − ++ +

++ ++ + ++ + ++ + + ++ ++ + ++

− + −

− − −

− ++ +

++ ++ +

− −

− −

++ ++

− −

− +

− −

++ −

− +

−

−

−

++

Phylogenetic affiliations were determined using SILVA SSU database release 115. However, in some cases manual phylogenetic assignment was required because the SILVA taxonomy was inconsistent. In these instances, lineages were named after the first gene clone derived from the group. Groups commonly found across all samples or that were the most abundant gene clones detected in reliable investigations of microbial community structure, with exception to SAR11, are indicated in boldface. 𝑏 [88] 𝑐 [80] 𝑑 [53, 71, 80, 155] 𝑒 [13, 57, 76] 𝑓 Included due to high abundance in black rust CORK sample [155] 𝑔 Candidate phyla names were adapted from [156]. − Not detected; + Detected once; ++ Multiple samples taken; detected in multiple samples; + + + Multiple samples taken; detected in all samples; clones from Smith et al. [76] not included in this assessment due to the paucity of SSU rRNA genes reported.

50 | 2 Life in the Oceanic Crust Fuca Ridge Flank borehole 1026B in 2004, a black rust was observed on the old CORK system installed in 1996 [155]. This rust yielded 16S rRNA gene clone sequences and cultivates of microbes that differed from surrounding seawater and sediment. Dominant members of the community here were Ammonifex and Methanothermococcus, both thermophilic microbes that were potentially enriched by the warm crustal fluids leaking through the old CORK [155], observed inside and on top of the observatory. The surficial mat was sampled and was found to contain Gammaproteobacteria, Alphaproteobacteria and Bacteroidetes bacteria, along with thermophilic Archaea [88]. The presence of mats at this CORK suggests that reactions between chemical components from reduced hydrothermal fluids with electron acceptors from seawater may foster microbial growth. Yet, caution in interpretation is important. Oxidation of iron from the CORK itself, which resulted in the formation of the observed black rust deposits at 1026B, may have also played a role as microbial energy source and facilitated growth of at least some of the phylotypes observed. Moreover, in the absence of further phylogenetic data from the native rock habitat, it remains uncertain if all, and if not, which phylotypes detected came from basalt vs. surrounding seawater and sediment.

2.4.4.3 Juan de Fuca system: comparing fluids and rocks Interestingly, although microbial communities between different CORK observatories on the Juan de Fuca show striking phylogenetic similarities (󳶳 Table 2.3) [57], the limited published data on rock samples suggest these to harbor significantly different microbial communities [49]. For instance, among the sulfate reducers frequently found in CORK observatories on the Juan de Fuca Ridge Flank have been members of the Firmicutes and Archaeoglobales [53, 57, 90, 155]. However, neither of these groups were detectable, not even with newly-designed, group-specific dsrAB primer pairs, in rock samples from IODP Site 1301 [49]. In the latter, solely the deeply-branching Group IV were detectable [49], which in turn had not been found in dsrAB clone libraries from CORK-derived samples [90, 155]. Similarly, the methanogenic community associated with black rust deposits in the CORK at ODP Site 1026, which consists of a strain of the thermophilic hydrogenotrophic methanogen Methanococcus thermolithotrophicus, differs from that in rocks at IODP Site 1301. The latter is dominated by uncultivated Methanosarcinales closely related to a methanogen with broad substrate requirements that has been enriched from wetland sediment [158], and two phylotypes of the methanotrophic ANME-1 cluster. The comparison of CORKs and rocks from the same type may highlight the different endolith vs. pelagic forms of microbial life in the oceanic crust, or that the surface of true diversity and abundance has only been scratched in this environment.

2.5 Future directions

| 51

2.5 Future directions So far, research on life within the oceanic crust is still at an early exploratory state. While it is known that microbial life, consisting of bacteria, archaea, viruses and fungi and perhaps further into the tree of life, exists deep into the crust, knowledge of these communities is still profoundly limited by access to fresh rock and fluid samples. So far, only a few sites have been sampled and analyzed, and those even have large vertical intervals missing due to poor core recovery of certain lithology types and limitations to depth of penetration. Even within rocks and rock fragments retrieved, the distribution of microbes and microbial biomass remains unclear. It is, for instance, not known what fraction of microbial biomass grows as epiliths on rock surfaces, as chasmoendoliths in rock fissures, as cryptoendoliths in porous rock, or as euendoliths by actively penetrating into rock, let alone how the biomass of these attached-growing microbes compares to that of planktonic forms. Even less is known how and if microbial communities inhabiting different parts of rocks differ phylogenetically and in terms of their ecological niches. In the absence of analyses on the three-dimensional distribution of microbes throughout rocks and circulating fluids – both on microscopic and macroscopic scales – the global biomass of crustal microbes is a matter of speculation. It remains to be shown whether the vast ocean crust, which is distributed over ∼60% of the Earth’s surface and ranges in thickness from 5–10 km, is a globally significant habitat in terms of its microbial biomass, or an ecological desert that is mainly of interest to microbiologists and biochemists with interest in extremophilic forms of life on Earth and beyond. The role of ocean crust-inhabiting microbes in Earth’s biogeochemical cycles is also largely unknown. For instance, past studies have shown ocean crust to be a globally significant sink of carbon. Yet, the extent to which microbes play a role in this carbon sequestration remains to be shown [13]. The same is true for seawater-derived sulfur. Sulfate from seawater is known to accumulate in parts of the ocean crust, and it is evident that microbial sulfate reduction plays a role locally [49]. Yet, chemical precipitation is also likely to be important [140], and the relative extent to which sulfur accumulation is caused by microbes and chemical precipitation processes remains unknown. Currently, it is even uncertain what the main energy sources of life in the ocean crust are. In some locations, the main energy may come from photoautotrophically synthesized organic matter that was originally produced in illuminated surface environments, and enters crust on outcrops due to seawater circulation through crust or by mixing of bottom sediments with formation fluids. Due to the low concentrations of organic carbon in most bottom seawater and the typically highly refractory state of organic compounds at the bottom of sediment columns, one might expect microbial abundance in these places to be very small compared to other habitats in the biosphere. On the other hand, indigenous energy sources, such as electrons from microbial Fe(II) oxidation, H2 produced by serpentinization reactions, or FTT synthesis

52 | 2 Life in the Oceanic Crust of C1-compounds and small-chain organic molecules, might turn out to be quantitatively important [39, 45, 49]. And based on modeled 120 °C isotherms, it has been estimated that the habitable zone in crust ranges from 0.5 km in 1 million years ago (mya) old crust and increases over time to 5 km in 180 mya old crust [1]. If indigenous energy sources are indeed important, and the habitable zone is as large as has been postulated, then the potential for microbial growth and proliferation in the oceanic crust is tremendous. This would mean that in addition to solar irradiance, chemical reactions within rocks are among the major energy sources of life on Earth. Ocean crust would then harbor one of the largest global pools of living biomass, with total biomass exceeding that in all aquatic and soil environments [1]. The implications of a deep, rock-associated biosphere on Earth would be far-reaching with respect to the potential of crustal (sub)surface habitats elsewhere in the universe to support life independent of sunlight. On a more proximal scale, while postulating questions of significance for the coming years, it is important that research remains grounded by already existing data. Based on clearly observable trends in the limited existing data sets, it is already possible to formulate hypotheses that will be testable in the near future. For instance, the discrepancy in microbes between CORK-associated fluids, downhole incubations and rocks from the same locations highlights the potentially diverse nature of life in oceanic crust, and indicates that there are many habitats and forms of crust-associated microbes. Even if these differences are due to sampling biases caused by the fact that different sampling methods pick up different habitat types, it seems likely that phylogenetic differences observed would reflect divergent ecological niches, such as fluidvs. rock-associated life histories. This possibility also indicates that, currently, no one single sampling approach can satisfy all questions about crustal life. Instead, multiple techniques, including drilling, logging and borehole observatories, are needed to study endolithic, epilithic and free- living forms, respectively. Each of these tools provides complementary information that will improve our understanding of life and its role in crustal alteration and global geochemical cycles. To more exhaustively study the distribution of life in ocean crust, it will be important to further expand the use of the already existing, as well as develop innovative new research tools. With the construction of the drilling vessel Chikyu, which was designed to drill to 7 km below the seafloor, the obtenance of rock samples from deeper parts of oceanic crust, and even parts of the Mohorovièiæ discontinuity (moho) at the very boundary of crust and mantle, is within reach. Future CORK designs may enable the closing off of different borehole sections, and thereby enable microbial growth experiments throughout all major segments of the same borehole. Further developments in in situ fluid sampling devices, such as the Quicksilver Probe (Schlumberger Limited, USA), which can be installed within boreholes, will enable the sampling of pristine fluids from rock formations surrounding CORKs. This will eliminate the potential problem of fluid residence times within boreholes prior to sampling. Even short residence times may alter the chemical composition of fluids, making concentration measure-

References | 53

ments on molecules with high microbial turnover rates, e.g. intermediates such as H2 , VFAs and alcohols, especially problematic. Furthermore, the combination of automatic devices with sidewall coring techniques throughout the borehole, will make it possible to conduct and monitor experiments directly on rocks surrounding the boreholes, e.g. by means of isotopic tracer experiments, combined with auto-sampling devices [89, 90] or in situ sample processors [159]. Apart from questions with respect to the distribution and biomass, many unknowns concerning life in the ocean crust have yet to be addressed. These include the relationship of microbes to the age of the crust, the mechanisms by which microbes bore into crust and are involved in crustal transformations, the vertical and horizontal connectivity of ocean crust and the seeding of seafloor and subseafloor basalts [13]. With the increasing efficiency of low biomass molecular methods and the growing field of data on crustal life, the great challenges of these studies may be answered soon, provided that field expeditions continue to collect high-quality samples and continue to improve methods and instruments that enable the study of life in basalt using in-situ observatories.

Acknowledgements We would like to acknowledge all of the scientists, engineers and shipboard personnel that make this work possible. This chapter is dedicated in the memory of James Cowen, a colleague and friend, who was a pioneer in the study of crustal fluids.

References [1] [2] [3] [4] [5] [6] [7] [8]

Heberling C, Lowell RP, Liu L, Fisk MR. Extent of the microbial biosphere in the oceanic crust. Geochemistry Geophysics Geosystems 11 (2010), Q08003. Cowen JP. The microbial biosphere of sediment-buried oceanic basement. Research in Microbiol 155 (2004), 497–506. D’Hondt S, Inagaki F, Ferdelman T, Jørgensen BB, Kato K, Kemp P et al. Exploring subseafloor life with the Integrated Ocean Drilling Program. Scientific Drilling 5 (2007), 26–37. Schrenk MO, Huber JA, Edwards KJ. Microbial provinces in the subseafloor. Annual Review of Marine Science 2 (2010), 279–304. Orcutt BN, Sylvan JB, Knab NJ, Edwards KJ. Microbial ecology of the dark ocean above, at and below the seafloor. Microbiology and Molecular Biology Reviews 72 (2011), 361–422. Edwards KJ, Wheat CG, Sylvan JB. Under the sea: microbial life in volcanic oceanic crust. Nature Reviews Microbiology 9 (2011), 703–712. Biddle JF, Sylvan JB, Brazelton WJ, Tully BJ, Edwards KJ, Moyer CL et al. Prospects for the study of evolution in the deep biosphere. Frontiers in Microbiology 2 (2012), 285. Edwards KJ, Becker K, Colwell F. The deep, dark energy biosphere: intraterrestrial life on Earth. Annual Review of Earth and Planetary Sciences 40 (2012), 551–568.

54 | 2 Life in the Oceanic Crust [9] [10]

[11] [12] [13] [14] [15] [16] [17]

[18]

[19]

[20] [21] [22] [23] [24]

[25] [26]

[27]

[28]

Edwards KJ, Fisher AT, Wheat CG. The deep subsurface biosphere in igneous ocean crust: frontier habitats for microbiological exploration. Frontiers in Microbiology 3 (2012), 8. Wang F-P, Lu S-L, Orcutt BN, Xie W, Chen Y, Xiao X et al. Discovering the roles of subsurface microorganisms: progress and future of deep biosphere investigation. Chinese Science Bulletin 58 (2013), 456–467. Colwell FS, D’Hondt S. Nature and extent of the deep biosphere. Reviews in Mineralogy and Geochemistry 75 (2013), 547–574. Hoehler TM, Jørgensen BB. Microbial life under extreme energy limitation. Nature Reviews Microbiology 11 (2013), 83–94. Orcutt BN, Bach W, Becker K, Fisher AT, Hentscher M, Toner BM et al. Colonization of subsurface microbial observatories deployed in young ocean crust. ISME Journal 5 (2011), 692–703. Lever MA. Functional gene surveys from ocean drilling expeditions a review and perspective. FEMS Microbiology Ecology 84 (2013), 1–23. Orcutt BN, LaRowe DE, Biddle JF, Colwell FS, Glazer BT, Reese BK et al. Microbial activity in the marine deep biosphere: progress and prospects. Frontiers in Microbiology 4 (2013), 189. Schrenk MO, Brazelton WJ, Lang SQ. Serpentinization, carbon, and deep life. Reviews in Mineralogy and Geochemistry 75 (2013), 575–606. Staudigel H, Furnes H. Microbial mediation of oceanic crust alteration. In: Davis EE, Elderfield H (eds). Hydrogeology of the Oceanic Lithosphere. Cambridge University Press: Cambridge. 606–624, 2004. Furnes H, McLoughlin N, Muehlenbachs K, Banerjee NR, Staudigel H, Dilek Y et al. Oceanic pillow lavas and hyaloclastites as habitats for microbial life through time – a review. In: Dilek Y, Furnes H, Muehlenbachs K (eds). Links Between Geological Processes, Microbial Activities & Evolution of Life. Springer: Netherlands. 1–68, 2008. Humphris SE, Zierenberg RA, Mullineau LS, Thomson RE (eds). Seafloor Hydrothermal Systems: Physical, Chemical, Biological and Geological Interactions. American Geophysical Union: Washington, DC, 466pp, 1995. Karl DM (ed). The Microbiology of Deep-Sea Hydrothermal Vents. CRC Press: Boca Raton, FL, 299pp, 1995. Wilcock WSD, DeLong EF, Kelley DS, Baross JA, Cary SC (eds). The Subseafloor Biosphere at Mid-Ocean Ridges. American Geophysical Union: Washington, DC, 399pp, 2004. Deming JW, Baross JA. Deep-sea smokers: windows to a subsurface biosphere. Geochimica et Cosmochimica Acta 57 (1993), 3219–3230. Thorseth IH, Torsvik T, Furnes H, Muehlenbachs K. Microbes play an important role in the alteration of oceanic crust. Chemical Geology 126 (1995), 137–146. Giovannoni SJ, Fisk MR, Mullins TD, Furnes H. Genetic evidence for endolithic microbial life colonizing basaltic glass/seawater interfaces. In: Alt JC, Kinoshita H, Stokking LB, Michael PJ (eds). Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, Vol. 148, 207–214, 1996. Fisk MR, Giovannoni SJ, Thorseth IH. Alteration of oceanic volcanic glass: textural evidence of microbial activity. Science 281 (1998), 978–980. Anderson RE, Brazelton WJ, Baross JA. The deep viriosphere: assessing the viral impact on microbial community dynamics in the deep subsurface. Reviews in Mineralogy and Geochemistry 75 (2013), 649–675. Takai K, Nakamura K, Toki T, Tsunogai U, Miyazaki M, Miyazaki J et al. Cell proliferation at 122 °C and isotopically heavy CH4 production by a hyperthermophilic methanogen under high-pressure cultivation. Proc Natl Acad Sci USA 105 (2008) 10 949–10 954. Furnes H, Staudigel H. Biological mediation in ocean crust alteration: how deep is the deep biosphere? Earth and Planetary Science Letters 166 (1999), 97–103.

References | 55

[29]

[30]

[31] [32] [33]

[34]

[35] [36]

[37]

[38] [39]

[40]

[41] [42]

[43]

[44]

[45] [46]

Fisk MR, Thorseth IH, Urbach E, Giovannoni SJ. Investigation of microorganisms and DNA from subsurface thermal water and rock from the east flank of the Juan de Fuca Ridge. In: Fisher A, Davis EE, Escutia C (eds). Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, Vol. 168, 167–174, 2000. Furnes H, Staudigel H, Thorseth IH, Torsvik T, Muehlenbachs K, Tumyr O. Bioalteration of basaltic glass in the oceanic crust. Geochemistry Geophysics Geosystems 2 (2001), 2000GC000150. Thorseth IH, T Torsvik, V Torsvik, FL Daae, RB Pederson. Diversity of life in ocean floor basalt. Earth and Planetary Science Letters 194 (2001), 31–37. Banerjee NR, Muehlenbachs K. Tuff life: bioalteration in volcaniclastic rocks from the Ontong Java Plateau. Geochemistry Geophysics Geosystems 4 (2003), 2002GC000470. Edwards KJ, McCollom TM, Konishi H, Buseck PR. Seafloor bioalteration of sulfide minerals: results from in situ incubation studies. Geochimica et Cosmochimica Acta 67 (2003), 2843– 2856. Fisk MR, Storrie-Lombardi MC, Douglas S, Popa R, McDonald G, Di Meo-Savoie C. Evidence of biological activity in Hawaiian subsurface basalts. Geochemistry Geophysics Geosystems 4 (2003), 2002GC000387. Einen J, Kruber C, Øvreås L, Thorseth IH, Torsvik T. Microbial colonization and alteration of basaltic glass. Biogeosciences Discussions 3 (2006), 273–307. Walton AW. Microtubules in basalt glass from Hawaii Scientific Driling Project #2 phase 1 core and Hilina slope, Hawaii: evidence of the occurrence and behavior of endolithic microorganisms. Geobiology 6 (2008), 351–364. Staudigel H, Furnes H, McLoughlin N, Banerjee NR, Connell LB, Templeton A. 3.5 billion years of glass bioalteration: volcanic rocks as a basis for microbial life? Earth Science Reviews 89 (2008), 156–176. Schumann G, Manz W, Reitner J, Lustrino M. Ancient fungal life in North Pacific Eocene oceanic crust. Geomicrobiology Journal 21 (2004), 241–246. Bach W, Edwards KJ. Iron and sulfide oxidation within the basaltic ocean crust: implications for chemolithoautotrophic microbial biomass production. Geochimica et Cosmochimica Acta 67 (2003), 3871–3887. Wheat CG, Fisher AT. Massive, low-temperature hydrothermal flow from a basaltic outcrop on 23 Ma seafloor of the Cocos Plate: chemical constraints and implications. Geochemistry Geophysics Geosystems 9 (2008), 2008GC002136. Santelli CM, Orcutt BN, Banning E, Bach W, Moyer CL, Sogin ML et al. Abundance and diversity of microbial life in ocean crust. Nature 453 (2008), 653–656. Mason OU, Di Meo-Savoie CA, Van Nostrand JD, Zhou J, Fisk MR, Giovannoni SJ. Prokaryotic diversity, distribution and insights into their role in biogeochemical cycling in marine basalts. ISME Journal 3 (2009), 231–242. Karl DM, Taylor GT, Novitsky JA, Jannasch HW, Wirsen CO, Pace NR et al. A microbiological study of Guaymas Basin high temperature hydrothermal vents. Deep-Sea Research Part I: Oceanographic Research Papers 35 (1988), 777–791. Robidart J, Callister SJ, Song PF, Nicora CD, Wheat CG, Girguis PR. Characterizing microbial community and geochemical dynamics at hydrothermal vents using osmotically driven continuous fluid samplers. Environmental Science and Technology 47 (2013), 4399–4407. Gold T. The deep, hot biosphere. Proceedings of the National Academy of Sciences of the United States of America 89 (1992), 6045–6049. Huey DP. IODP drilling and coring technology: past and present. Stress Engineering Services Inc.: Houston, TX, 2009; 183pp.

56 | 2 Life in the Oceanic Crust [47]

[48] [49] [50]

[51]

[52]

[53] [54]

[55]

[56] [57] [58] [59]

[60]

[61]

Santelli CM, Banerjee N, Bach W, Edwards KJ. Tapping the subsurface ocean crust biosphere: low biomass and drilling-related contamination calls for improved quality controls. Geomicrobiology Journal 27 (2010), 158–169. Mason OU, Nakagawa T, Rosner M, Van Nostrand JD, Zhou J, Maruyama A et al. First investigation of the microbiology of the deepest layer of ocean crust. PLoS One 5 (2010), e15399. Lever MA, Rouxel O, Alt JC, Shimizu N, Ono S, Coggon RM et al. Evidence for microbial carbon and sulfur cycling in deeply buried ridge flank basalt. Science 339 (2013), 1305–1308. Expedition 331 Scientists. Expedition 331 summary. In: Takai K, Mottl MJ, Nielsen SH and the Expedition 331 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., Tokyo, Vol. 331, 39pp, 2011. von der Weid I, Korenblum E, Jurelevicius D, Rosado AS, Dino R, Sebastian GV, Seldin L. Molecular diversity of bacterial communities from subseafloor rock samples in a deep-water production basin in Brazil. Journal of Microbiology and Biotechnology 18 (2008), 5–14. Davis EE, Becker K, Pettigrew T, Carson B, MacDonald R. CORK: a hydrologic seal and downhole observatory for deep-ocean boreholes. In: Davis EE, Mottl MJ, Fisher AT et al. (eds). Proceedings of the Ocean Drilling Program, Initial Reports. Ocean Drilling Program, College Station , TX, Vol. 139, 43–53, 1992. Cowen JP, Giovannoni SJ, Kenig F, Johnson HP, Butterfield D, Rappé MS et al. Fluids from aging ocean crust that support microbial life. Science 299 (2003), 120–123. Jannasch HW, Davis EE, Kastner M, Morris JD, Pettigrew TL, Plant JN et al. CORK-II: long-term monitoring of fluid chemistry, fluxes and hydrogeology in instrumented boreholes at the Costa Rica subduction zone. In: Morris JD, Villinger HW, Klaus A et al (eds). Proceedings of the Ocean Drilling Program, Initial Reports. Ocean Drilling Program, College Station, TX, Vol. 205, 36pp, 2003. Wheat CG, Jannasch HW, Kastner M, Hulme S, Cowen J, Edwards KJ et al. Fluid sampling from oceanic borehole observatories: Design and methods for CORK activities (1990–2010). In: Fisher AT, Tsuji T, Petronotis K and the Expedition 327 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., Tokyo, 2011; 327: 36pp. Wheat CG, Edwards KJ, Pettigrew T, Jannasch HW, Becker K, Davis EE et al. CORK-Lite: bringing legacy boreholes back to life. Scientific Drilling 14 (2012), 39–43. Jungbluth SP, Grote J, Lin H-T, Cowen JP, Rappé MS. Microbial diversity within basement fluids of the sediment-buried Juan de Fuca Ridge flank. ISME Journal 7 (2013), 161–172. Matsumoto K, Sarata S. Development of deep-sea boring machine system (in Japanese with English abstract). Shigen-to-Sozai 1996; 112: 1015–1020. Higashi Y, Sunamura M, Kitamura K, Nakamura K, Kurusu Y, Ishibashi J et al. Microbial diversity in hydrothermal surface to subsurface environments of Suiyo Seamount, Izu-Bonin Arc, using a catheter-type in situ growth chamber. FEMS Microbiology Ecology 47 (2004), 327– 336. Kato S, Yanagawa K, Sunamura M, Takano Y, Ishibashi J, Kakegawa T et al. Abundance of Zetaproteobacteria within crustal fluids in back-arc hydrothermal fields of the Southern Mariana Trough. Environmental Microbiology 11 (2009), 3210–3222. Ishibashi JI, Marumo K, Maruyama A, Urabe T. Direct access to the sub-vent biosphere by shallow drilling. Oceanography 20 (2007), 24–25.

References | 57

[62]

[63] [64]

[65] [66]

[67]

[68] [69]

[70] [71] [72]

[73] [74]

[75]

[76]

[77]

[78]

Sager W, Dick H, Fryer P, Johnson HP. Report from a workshop “Requirements for robotic underwater drills in U.S. marine geologic research”. November 3–4 2000, College Station, TX, 2003. 82pp. Pallanich J. Prod probes Statoil’s seabed soils. Offshore Engineer 2010; February: 42–44. Stakes DS, Holloway GL, Tucker P, Dawe TC, Burton D, McFarlane JAR, Etchemendy S. Diamond rotary coring from an ROV or submersible for hardrock sample recovery and instrument deployment: the MBARI multiple-barrel rock coring system. Marine Technology Society Journal 31 (1997), 11–20. Wilson M. Drilling at sea. Earthwise 23 (2006), 32–33. Spencer A, Remmes B, Rowson I (eds). A fully integrated solution for the geotechnical drilling and sampling of seafloor massive sulfide deposits. Proceedings Offshore Technology Conference OTC 21439. Houston, TX. 2011: 20pp. Freudenthal T, Wefer G. Drilling cores on the sea floor with the remote-controlled sea-floor drilling rig MeBo. Geoscientific Instrumentation, Methods and Data Systems Discussions 2013; 3: 347–369. Freudenthal T, Wefer G. Scientific drilling with the sea floor drill rig MeBo. Scientific Drilling 5 (2007), 63–66. Krastel S, Wefer G, Hanebuth TJJ, Antobreh AA, Freudenthal T, Preu B, et al. Sediment dynamics and geohazards off Uruguay and the de la Plata River region (northern Argentina and Uruguay). Geo Marine Letters 31 (2011), 271–283. Johnson HP, and the LEXEN Scientific Party. Probing for life in the ocean crust with the LEXEN program. EOS Transactions of the American Geophysical Union 84 (2003), 109–112. Huber JA, Johnson HP, Butterfield DA, Baross JA. Microbial life in ridge flank crustal fluids. Environmental Microbiology 8 (2006), 88–99. Hara K, Kakegawa T, Yamashiro K, Maruyama A, Ishibashi JI, Marumo K, et al. Analysis of the archaeal subseafloor community at Suiyo Seamount on the Izu-Bonin Arc. Advances in Space Research 35 (2005), 1634–1642. Moran K, Farrington S, Massion E, Paull C, Stephen R, Trehu A, Ussler W. SCIMPI: a new seafloor observatory system. OCEANS 2006. Boston, MA. 2006: 6pp. Expedition 341S Scientists and Engineers. Simple cabled instrument for measuring parameters in situ (SCIMPI) and hole 858G CORK replacement. Integrated Ocean Drilling Program Preliminary Report. 2013; 341S: 50pp. Salas EC, Bhartia R, Reid R, Hug W, Nguyen Q, Oswal P, Sullivan K, Edwards KJ. Probing in the dark: preliminary results from the Dark Energy Biosphere Investigative Tool (DEBI-T), IODP 336. AGU Meeting Fall Abstracts 2011: B44B-07. Smith A, Popa R, Fisk M, Nielsen M, Wheat CG, Jannasch HW et al. In situ enrichment of ocean crust microbes on igneous minerals and glasses using an osmotic flow-through device. Geochemistry Geophysics Geosystems 12 (2011), Q06007. Fisher AT, Wheat CG, Becker K, Davis EE, Jannasch H, Schroeder D et al. Scientific and technical design and deployment of long-term subseafloor observatories for hydrogeologic and related experiments, IODP Expedition 301, eastern flank of Juan de Fuca Ridge. In: Fisher AT, Urabe T, Klaus A and the Expedition 301 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., College Station, TX, 2005; 301: 39pp. Becker K, Davis EE. A review of CORK designs and operations during the Ocean Drilling Program. In: Fisher AT, Urabe T, Klaus A and the Expedition 301 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., College Station, TX, 2005; 301: 28pp.

58 | 2 Life in the Oceanic Crust [79]

[80]

[81]

[82]

[83]

[84]

[85]

[86]

[87]

[88]

[89]

[90]

[91]

[92]

Orcutt BN, Barco RA, Joye SB, Edwards KJ. Summary of carbon, nitrogen and iron leaching characteristics and fluorescence properties of materials considered for subseafloor observatory assembly. In: Edwards KJ, Bach W, Klaus A and the Expedition 336 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., Tokyo, 2012; 336: 12pp. Jungbluth SP, Lin H-T, Cowen JP, Rappé MS. Phylogenetic diversity of microorganisms inhabiting basement fluids of the deep subseafloor basaltic crust accessed through ODP boreholes 1025C and 1026B along Juan de Fuca Ridge flank. Frontiers in Microbiology. 2013; In review. Von Damn KL, Edmond JM, Measures CI, Walden B, Weiss RF. Chemistry of submarine hydrothermal solutions at 21°N, East Pacific Rise. Geochimica et Cosmochimica Acta 49 (1985), 2197–2220. Butterfield DA, Roe KK, Lilley MD, Huber JA, Baross JA, Embley RW, et al. Mixing, reaction and microbial activity in the subseafloor revealed by temporal and spatial variation in diffuse flow vents at Axial Volcano. In: Wilcock WSD, DeLong EF, Kelley DS, Baross JA, Cary SC (eds). The Subseafloor Biosphere at Mid-Ocean Ridges. American Geophysical Union: Washington, DC. 2004: 269–289. Di Meo CA, Wakefield JR, Cary SC. A new device for sampling small volumes of water from marine micro-environments. Deep-Sea Research Part I: Oceanographic Research Papers 46 (1999), 1279–1287. Reysenbach AL, Longnecker K, Kirshtein J. Novel bacterial and archaeal lineages from an in situ growth chamber deployed at a Mid-Atlantic Ridge hydrothermal vent. Applied and Environmental Microbiology 66 (2000), 3798–3806. Nercessian O, Reysenbach AL, Prieur D, Jeanthon C. Archaeal diversity associated with in situ samplers deployed on hydrothermal vents on the East Pacific Rise (13°North ). Environmental Microbiology 5 (2003), 492–502. Phillips H, Wells LE, Johnson RV, Elliott S, Deming JW. LAREDO: a new instrument for sampling and in situ incubation of deep-sea hydrothermal vent fluids. Deep-Sea Research Part I: Oceanographic Research Papers 50 (2003), 1375–1387. Alain K, Zbinden M, Le Bris N, Lesongeur F, Quérellou J, Gaill F, et al. Early steps in microbial colonization processes at deep-sea hydrothermal vents. Environmental Microbiology 6 (2004), 227–241. Nigro LM, Harris K, Orcutt BN, Hyde A, Clayton-Luce S, Becker K et al. Microbial communities at the borehole observatory on the Costa Rica Rift flank (Ocean Drilling Program Hole 896A). Frontiers in Microbiology 3 (2012), 232. Jannasch HW, Wheat CG, Plant JN, Kastner M, Stakes DS. Continuous chemical monitoring with osmotically pumped water samplers: OsmoSampler design and applications. Limnology and Oceanography Methods 2 (2004), 102–113. Orcutt B, Wheat CG, Edwards KJ. Subseafloor ocean crust microbial observatories: development of FLOCS (FLow-through Osmo Colonization System) and evaluation of borehole construction materials. Geomicrobiology Journal 27 (2010), 143–157. Kopf A, Saffer DM, Davis EE, Hammerschmidt S, LaBonte A, Meldrum R, et al. The SmartPlug and GeniusPlug: simple retrievable observatory systems for NanTroSEIZE borehole monitoring. In: Kopf A, Araki E, Toczko S and the Expedition 332 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., Tokyo, 332: 20pp, 2011. Massoth GJ, Butterfield DA, Lupton JE, McDuff RE, Lilley MD, Jonasson IR. Submarine venting of phase-separated hydrothermal fluids at Axial Volcano, Juan de Fuca Ridge. Nature 340 (1989), 702–705.

References |

[93]

[94]

[95]

[96]

[97]

[98]

[99]

[100]

[101]

[102]

[103]

[104]

[105]

[106]

[107]

59

Behar A, Matthews J, Venkateswaran K, Bruckner J, Basic G, So E, et al. A deep sea hydrothermal vent bio-sampler for large volume in-situ filtration of hydrothermal vent fluids. NASA Jet Propulsion Laboratories. 6pp, 2005. Taylor CD, Doherty KW, Molyneaux SJ, Morrison AT, Billings JD, Engstrom IB, et al. Autonomous Microbial Sampler (AMS), a device for the uncontaminated collection of multiple microbial samples from submarine hydrothermal vents and other aquatic environments. Deep-Sea Research Part I: Oceanographic Research Papers Vol. 53, 894–916, 2006. Garbe-Schönberg D, Koschinsky A, Ratmeyer V, Jähmlich H, Westernströer U (eds). KIPS – a new multiport valve-based all-Teflon fluid sampling system for ROVs. European Geosciences Union. 2pp, 2006. Scholin C, Doucette G, Jensen S, Roman B, Pargett D, Marin R, et al. Remote detection of marine microbes, small invertebrates, harmful algae and biotoxins using the Environmental Sample Processor (ESP). Oceanography 22 (2009), 158–167. Ussler W, Preston C, Tavormina P, Pargett D, Jensen S, Roman B, et al. Autonomous application of quantitative PCR in the deep sea: in situ surveys of aerobic methanotrophs using the deep-sea environmental sample processor. Environmental Science and Technology 47 (2013), 9339–9346. Cowen JP, Copson DA, Jolly J, Hsieh C-C, Lin H-T, Glazer BT et al. Advanced instrument system for real-time and time-series microbial geochemical sampling of the deep (basaltic) crustal biosphere. Deep-Sea Research Part I: Oceanographic Research Papers 61 (2012), 43–56. Lin H-T, Cowen JP, Olson EJ, Amend JP, Lilley MD. Inorganic chemistry, gas compositions and dissolved organic carbon in fluids from sedimented young basaltic crust on the Juan de Fuca Ridge flanks. Geochimica et Cosmochimica Acta 85 (2012), 213–227. Expedition 336 Scientists. Expedition 336 summary. In: Edwards KJ, Bach W, Klaus A and the Expedition 336 Scientists (eds). Proceedings of the Integrated Ocean Drilling Program. Integrated Ocean Drilling Program Management International, Inc., Tokyo, 336, 30pp, 2012. Callac N, Rommevaux-Jestin C, Rouxel O, Lesongeur F, Liorzou C, Bollinger C et al. Microbial colonization of basaltic glasses in hydrothermal organic-rich sediment at Guaymas Basin. Frontiers in Microbiology 4 (2013), 250. Masui N, Morono Y, Inagaki F. Microbiological assessment of circulation mud fluids during the first operation of riser drilling by the deep-earth research vessel Chikyu. Geomicrobiology Journal 25 (2008), 274–282. Smith DC, Spivack AJ, Fisk MR, Haveman SA, Staudigel H and the Leg 185 Shipboard Scientific Party. Methods for quantifying potential microbial contamination during deep ocean coring. ODP Technical Note 28 (2000), 19. Smith DC, Spivack AJ, Fisk MR, Haveman SA, Staudigel H and the Ocean Drilling Program Leg 185 Shipboard Scientific Party. Tracer-based estimates of drilling-induced microbial contamination of deep sea crust. Geomicrobiology Journal 17 (2000), 207–219. House CH, Cragg BA, Teske A and the Leg 201 Scientific Party. Drilling contamination tests during ODP Leg 201 using chemical and particulate tracers. In: D’Hondt Sl, Jørgensen BB, Miller DJ et al (eds). Proceedings of the Ocean Drilling Program, Initial Reports. Ocean Drilling Program, College Station, TX, Vol. 201, 19pp, 2003. Lever MA, Alperin M, Engelen B, Inagaki F, Nakagawa S, Steinsbu BO et al. Trends in basalt and sediment core contamination during IODP Expedition 301. Geomicrobiology Journal 23 (2006), 517–530. Inagaki F, Hinrichs K-U, Kubo Y and the Expedition 337 Scientists. Deep coalbed biosphere off Shimokita: microbial processes and hydrocarbon system associated with deeply buried coalbed in the ocean. Integrated Ocean Drilling Program Preliminary Report Vol. 337, 62pp, 2012.

60 | 2 Life in the Oceanic Crust [108] Lever MA, personal communication, 2013. [109] Shipboard Scientific Party. Site 1253. In: Morris JD, Villinger HW, Klaus A, et al. Proceedings of the Ocean Drilling Program, Initial Reports. Ocean Drilling Program, College Station, TX, Vol. 205, 2003, 184pp. [110] Golubic S, Friedmann I, Schneider J. The lithobiontic ecological niche, with special reference to microorganisms. Journal of Sedimentary Petrology 51 (1981), 475–478. [111] Bhartia R , Salas EC, Hug WF, Reid RD, Lane AL, Edwards KJ, Nealson KH. Label-free bacterial imaging with deep-UV-laser-induced native fluorescence. Applied and Environmental Microbiology 76 (2010), 7231–7237. [112] McLoughlin N, Wacey D, Kruber C, Kilburn MR, Thorseth IH, Pedersen RB. A combined TEM and NanoSIMS study of endolithic microfossils in altered seafloor basalt. Chemical Geology 289 (2011), 154–162. [113] Ivarsson M, Bengtson S, Belivanova V, Stampanoni M, Marone F, Tehler A. Fossilized fungi in subseafloxor Eocene basalts. Geology 40 (2012), 163–166. [114] Furnes H, Staudigel. Biological mediation in ocean crust alteration: how deep is the deep biosphere? Earth and Planetary Science Letters 166 (1998), 97–103. [115] Ivarsson M, S Bengtso, H Skogby, V Belivanova, F Marone. Fungal colonies in open fractures of subseafloor basalt. Geo Marine Letters 33 (2013), 233–243. [116] Lang SQ, Butterfield DA, Lilley MD, Johnson HP, Hedges JI (2006) Dissolved organic carbon in ridge-axis and ridge-flank hydrothermal systems. Geochimica et Cosmochimica Acta 70 (2006), 3830–3842. [117] Elderfield H, Wheat CG, Mottl MJ, Monnin C, Spiro B. Fluid and geochemical transport through oceanic crust: a transect across the eastern flank of the Juan de Fuca Ridge. Earth and Planetary Science Letters 172 (1999), 151–165. [118] Walker BD, McCarthy MD, Fisher AT, Guilderson TP. Dissolved inorganic carbon isotopic composition of low-temperature axial and ridge-flank hydrothermal fluids of the Juan de Fuca Ridge. Marine Chemistry 108 (2008), 123–136. [119] Orcutt BN, Wheat CG, Rouxel O, Hulme S, Edwards KJ. Oxygen consumption rates in subseafloor basaltic crust derived from a reaction transport model. Nature Communications 4 (2013), 2539. [120] McCarthy MD, Beaupré SR, Walker BD, Voparil I, Guilderson TP, Druffel ERM Chemosynthetic origin of 14 C-depleted dissolved organic matter in a ridge-flank hydrothermal system. Nature Geoscience 4 (2012), 32–36. [121] Penning H, Conrad R. Carbon isotope effects associated with mixed-acid fermentation of saccharides by Clostridium papyrosolvens. Geochimica et Cosmochimica Acta 70 (2006), 2283–2297. [122] Penning H, Claus P, Caper P, Conrad R. Carbon isotope fractionation during acetoclastic methanogenesis by Methanosaeta concilii in culture and a lake sediment. Applied and Environmental Microbiology 72 (2006), 5648–5652. [123] Goevert D, Conrad R. Carbon isotope fractionation by sulfate-reducing bacteria using different pathways for the oxidation of acetate. Environmental Science and Technology 42 (2008), 7813–7817. [124] Zerkle AL, House CH, Brantley SL. Biogeochemical signatures through time as inferred from whole microbial genomes. American Journal of Science 305 (2005), 467–502. [125] Rudnicki MD, Elderfield H, Spiro B. Fractionation of sulfur isotopes during bacterial sulfate reduction in deep ocean sediments at elevated temperatures. Geochimica et Cosmochimica Acta 65 (2001), 777–789. [126] Christie DM, Carmichael ISE, Langmuir CH. Oxidation states of mid-ocean ridge basalt glasses. Earth and Planetary Science Letters 79 (1986), 397–411.

References |

61

[127] Bach W, Erzinger J. Volatile components in basalts and basaltic glasses from the EPR at 9° 30’ N. In: Batiza R, Storms MA, Allan JF (eds). Proceedings of the Ocean Drilling Program, Scientific Results. Ocean Drilling Program, College Station, TX, Vol. 142, 23–29, 1995. [128] Dinh HT, Kuever J, Mußmann M, Hassel AW, Stratmann M, Widdel F. Iron corrosion by novel anaerobic microorganisms. Nature 427 (2004), 829–832. [129] Edwards KJ, Bach W, Klaus A. Mid-Atlantic Ridge flank microbiology: initiation of long-term coupled microbiological, geochemical and hydrological experimentation within the seafloor at North Pond, western flank of the Mid-Atlantic Ridge. IODP Scientific Prospectus, 336. 2010. doi:10.2204/ iodp.sp.336.2010. [130] Fisher AT, Davis EE, Hutnak M, Spiess V, Zühlsdorff L, Cherkaoui A et al. Hydrothermal recharge and discharge across 50 km guided by seamounts on a young ridge flank. Nature 421 (2003), 618–621. [131] Fisher AT, Wheat CG. Seamounts as conduits for massive fluid, heat and solute fluxes on ridge flanks. Oceanography 23 (2010), 74–87. [132] Wheat CG, Jannasch HW, Fisher AT, Becker K, Sharkey J, Hulme S. Subseafloor seawater-basalt-microbe reactions: continuous sampling of borehole fluids in a ridge flank environment. Geochemistry Geophysics Geosystems 11 (2010), 2010GC003057. [133] Fisher AT, Urabe T, Klaus A, Wheat CG, Becker K, Davis E et al. IODP expedition 301 installs three borehole crustal observatories, prepares for three-dimensional, cross-hole experiments in the Northeastern Pacific Ocean. Scientific Drilling 1 (2005), 6–11. [134] Chapelle FH, O’Neill K, Bradley PM, Methé BA, Clufo SA, Knobel LL, Lovley DR. A hydrogenbased subsurface microbial community dominated by methanogens. Nature 415 (2002), 312– 315. [135] McCollom TM, Seewald JS. Abiotic synthesis of organic compounds in deep-sea hydrothermal environments. Chemical Reviews 107 (2007), 382–401. [136] Neubeck A, Duc NT, Bastviken D, Crill P, Holm NG. Formation of H2 and CH4 by weathering of olivine at temperatures between 30 and 70 °C. Geochemical Transactions 12 (2011), 6. [137] Rabus R, Hansen TA, Widdel F. Dissimilatory sulfate- and sulfur-reducing prokaryotes. In: Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E (eds). The Prokaryotes: An Evolving Electronic Resource for the Microbiological Community. Springer: New York, Vol. 2, 659–768, 2006. [138] Whitman WB, Bowen TL, Boone DR. The methanogenic bacteria. In: Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E (eds). The Prokaryotes: An Evolving Electronic Resource for the Microbiological Community. Springer: New York, Vol. 3, 165–207, 2006. [139] Weathers LJ, Parkin GF, Alvarez PJ. Utilization of cathodic hydrogen as electron donor for chloroform cometabolism by a mixed, methanogenic culture. Environmental Science and Technology 31 (1997), 880–885. [140] Alt JC, Shanks WC. Sulfur in serpentinized oceanic peridotites: serpentinization processes and microbial sulfate reduction. Journal of Geophysical Research Solid Earth 103 (1998), 9917–9929. [141] Alt JC, Shanks WC. Microbial sulfate reduction and the sulfur budget for a complete section of altered oceanic basalts, IODP Hole 1256D (eastern Pacific). Earth and Planetary Science Letters 310 (2011), 73–83. [142] Canfield DE. Reactive iron in marine sediments. Geochimica et Cosmochimica Acta 53 (1989), 619–632. [143] Poulton SW, Krom MD, Raiswell R. A revised scheme for the reactivity of iron (oxyhydr)oxide minerals towards dissolved sulfide. Geochimica et Cosmochimica Acta 68 (2004), 3703–3715. [144] Canfield DE, Thamdrup B. The production of 34 S-depleted sulfide during bacterial disproportionation of elemental sulfur. Science 266 (1994), 1973–1975.

62 | 2 Life in the Oceanic Crust [145] Rouxel O, Ono S, Alt J, Rumble D, Ludden J. Sulfur isotope evidence for microbial sulfate reduction in altered oceanic basalts at ODP Site 801. Earth and Planetary Science Letters 268 (2008), 110–123. [146] Ono S, Keller NS, Rouxel O, Alt JC. Sulfur-33 constraints on the origin of secondary pyrite in altered oceanic basement. Geochimica et Cosmochima Acta 87 (2012), 323–340. [147] Sim MS, Bosak T, Ono S. Large sulfur isotope fractionation does not require disproportionation. Science 333 (2011), 74–77. [148] Alt JC, Shanks WC, Crispini L, Gaggero L, Schwarzenbach EM, Früh-Green GL, Bernasconi SM. Uptake of carbon and sulfur during seafloor serpentinization and the effects of subduction metamorphism in Ligurian peridotites. Chemical Geology 322–323 (2012), 268–277. [149] Staudigel H, Hart SR, Schmincke H0U, Smith BM. Cretaceous ocean crust at DSDP sites 417 and 418: carbon uptake from weathering versus loss by magmatic outgassing. Geochimica et Cosmochimica Acta 53 (1989), 3091–3094. [150] Coggon RM, Teagle DAH, Cooper MJ, Vanko DA. Linking basement carbonate vein compositions to porewater geochemistry across the eastern flank of the Juan de Fuca Ridge, ODP Leg 168. Earth and Planetary Science Letters 219 (2004), 111–128. [151] Delacour A, Früh-Green GL, Bernasconi SM, Schaeffer P, Kelley DS. Carbon geochemistry of serpentinites in the Lost City Hydrothermal System (30°N, MAR). Geochimica et Cosmochima Acta 72 (2008), 3681–3702. [152] Früh-Green GL, Connolly JAD, Plas A, Kelley DS, Grobéty B. Serpentinization of oceanic peridotites: implications for geochemical cycles and biological activity. In: Wilcock WSD, DeLong EF, Kelley DS, Baross JA, Cary SC (eds). The Subseafloor Biosphere at Mid-Ocean Ridges. American Geophysical Union: Washington, DC. 119–136, 2004. [153] Alt JC, EM Schwarzenbach, GL Früh-Green, WC Shanks, SM Bernasconi, CJ Garrido et al. The role of serpentinites in cycling of carbon and sulfur: seafloor serpentinization and subduction metamorphism. Lithos 178 (2012), 40–54. [154] Orcutt BN, Lever MA, Baquiran J-P, Edwards KJ, Haddad A, Fisher AT. Microbial community transitions across the deep sediment-basement interface. Frontiers in Microbiology (2013); under revision. [155] Nakagawa S, Inagaki F, Suzuki Y, Steinsbu BO, Lever MA, Takai K et al. Microbial community in black rust exposed to hot ridge flank crustal fluids. Applied and Environmental Microbiology 72 (2006), 6789–6799. [156] Rinke C, Schwientek P, Sczyrba A, Ivanova NN, Anderson IJ, Cheng JF et al. Insights into the phylogeny and coding potential of microbial dark matter. Nature 499 (2013), 431–437. [157] Lloyd KG, Schreiber L, Petersen DG, Kjeldsen KU, Lever MA, Steen AD et al. Predominant archaea in marine sediments degrade detrital proteins. Nature 496 (2013), 215–218. [158] Zhang G, Tian J, Jiang N, Guo X, Wang Y, Dong X. Methanogen community in Zoige wetland of Tibetan plateau and phenotypic characterization of a dominant uncultured methanogen cluster ZC-I. Environmental Microbiology 10 (2008), 1850–1860. [159] Wankel SD, Joye SB, Samarkin VA, Shan SR, Friederich G, Melas-Kyriazi J, Girguis PR. New constraints on methane fluxes and rates of anaerobic methane oxidation in a Gulf of Mexico brine pool via in situ mass spectrometry. Deep-Sea Research Part II: Topical Studies in Oceanography 57, 21–23, 2010.

Karsten Pedersen

3 Microbial life in terrestrial hard rock environments 3.1 Hard rock aquifers from the perspective of microorganisms Rock is commonly classified as igneous, sedimentary or metamorphic. Igneous and metamorphic rocks are hard and have all passed through a period with high pressure and temperatures that were too high for microbial life (> 150 °C). When they cooled down after formation, these rocks were sterile and too compact to allow microbes to enter into the solid rocks. Over geological time, new hard rocks grew old and tectonic processes and glaciations generated fractured networks with room for intruding groundwater. Microbial life followed the groundwater into all rocks that had cooled below the temperature limit for life. The aperture of groundwater-conducting fractures can be anything from very tiny in the micrometer range up to crush zones in faults that may transport large quantities of groundwater (󳶳 Fig. 3.1). Over geological times, fracture surfaces in hard rocks will slowly get new properties because the bare rock surfaces will be altered by mineral dissolution and precipitation processes. Particles such as clay can migrate into the fractures and build clay deposits and eventually, open fractures may be closed by precipitation, e.g. by calcite. Therefore, from the perspective of a microorganism, the environment in a water-conducing fracture that has stood open for a long time can appear very different compared to a new fracture in the host rock. Not only geochemical processes will alter the character of the fracture environment, microbial processes will also influence. Respiration of dissolved O2 in re-charging water will soon make groundwater anaerobic. Oxidation of organic carbon to carbon dioxide may lead to calcite precipitation, iron(III) and manganese(IV) respiration will dissolve the solid oxides to dissolved ions, and sulfate respiration generates sulfide that may precipitate with metal ions. Most dissimilatory microbial processes in systems isolated from an oxygenic atmosphere have that in common that they decrease the redox potential (Eℎ ) of the system which have implications for many Eℎ sensitive geochemical reactions. The groundwater composition and the microbial diversity in various aquifers can be very different also in cases when they are separated only by a couple of cm of rock. It is all about the origin and mixing of the groundwater in each and every aquifer that determines the groundwater properties. At various points in a rock mass, aquifers will connect, and each such connection point will potentially offer gradient conditions that microorganisms can utilize. One example can be if a groundwater that is poor in electron acceptors but rich in electron donors mixes with an acceptor-rich but donor-poor groundwater. Over a very short distance, the conditions for microbial activity will change from bad to good and these conditions will follow downstream the

64 | 3 Microbial life in terrestrial hard rock environments

Fig. 3.1: Examples of fractures in hard rock drill cores. a) tiny fracture with surrounding rock still wet from groundwater, thereby the dark halo. b) Fracture surface with brownish iron oxides. c) Fracture surface with clay material and other precipitates. d) Fracture closed by calcite in the middle with fractures closed by darker material on both sides.

aquifer until the donor or the acceptor have been utilized; then the conditions for activity will again turn to bad until next mixing point. Many microorganisms attach and form biofilms in flowing water systems. It is not far-fetched to assume that attachment just downstream a mixing point will be very favorable for a microbial population with continuous supply of new donors and acceptors, while an unattached life in flowing groundwater will only be favorable when prosperous mixing positions are passed.

3.2 Windows into the terrestrial hard rock biosphere 3.2.1 Sampling methods for microbes in hard rock aquifers Sampling of water and surfaces in hard rock aquifers requires drilling or tunnel excavation. During drilling and tunnel excavation, there is an obvious risk for con-

3.2 Windows into the terrestrial hard rock biosphere | 65

tamination of samples and aquifers and for washout of secondary minerals that may be loosely attached to rock surfaces in aquifers (󳶳 Fig. 3.1). Sampling strategies for groundwater from hard rock aquifer has been reviewed in detail previously [1, 2]. Briefly, the drilling fluids needed may initially contaminate groundwater samples, but the contamination effect from drilling fluids can be mitigated over time [3]. The use of triple tube drilling with front washed drill bits [1, 4] can protect fracture surface materials from washout during drilling. Sampling of groundwater from an aquifer is commonly obtained using packer systems that isolate the aquifers or interest. Addition of tracer substances to the drill fluids and their subsequent analysis can test if contaminating drill water has been washed out by pumping or flushing of the packed of aquifer [1, 2]. It is more difficult to sample undisturbed biofilms on fracture surfaces than sample groundwater because contaminating microorganisms from drilling operations may attach and thereby avoid washout by pumping or flushing. Triple tube drilling can protect fracture surfaces from some of the drill water effects but there will still be risks for washout of fracture materials and contamination from drill water; the range of such washout and contamination effects are of course impossible to evaluate due to the lack of an undisturbed control [5]. An alternative methodology to collect attached microorganisms involves the introduction of solid surfaces in a groundwater flow from a packed off aquifer. The exposure should preferably be for a long time to allow attachment and subsequent growth of attached microorganisms. The time needed will be site-specific to some extent. Weeks, preferably months will be needed in slowly flowing groundwater systems. The solids can be flat glass or rock surfaces, crushed rock, rock or glass beads, just to mention the most obvious materials. Installation of these materials in flow cells can be done on the ground surface on top of pumped boreholes or in tunnels connected to aquifers in boreholes drilled through the tunnel face. A large advantage with tunnels is that it is fairly easy to retain in situ pressure, thereby avoiding the inevitable degassing of dissolved gas that occur when deep groundwater is lifted to the ground surface with borehole pumps.

3.2.2 Yesterday marine – terrestrial today What is terrestrial today may have been subseafloor previously. For example, during glaciations, terrestrial parts of the Fennoscandian Shield close to the coast were immersed under the sea level. The seawater ranged from brackish to marine in salinity. During these periods of immersion, seawater intruded into aquifers of the shield and gave a marine signature down to at most approximately 500 m depth. Sulfate in groundwater, for instance, often originates from such seawater intrusions. Groundwater in the part of the Shield that was above the highest sea level in Scandinavia is often of meteoric origin and is therefore diluted. It is of course straightforward to judge what is terrestrial and subseafloor environments today, but their origin must be kept

66 | 3 Microbial life in terrestrial hard rock environments in mind, especially when fossils and various geochemical signatures are evaluated. Likewise, over very long geological times, hard rock environments have been formed deep in the Earth, often under the seafloor, and uplift has eventually brought these rocks above sea level. Many fracture environments in hard rock have consequently undergone periods of extreme conditions that has shaped them to their present day appearance.

3.2.3 Basalts and ophiolites Extrusive igneous basaltic rocks are formed from rapidly cooling basaltic lava near the surface. Layers of various types of basalt and ashes can create isolated aquifers that eventually discharge as arthesian springs or wells. Samples were collected from alkaline, pH 8.5–10.5, aquifers associated with terrestrial flood basalt in Southeast Washington, USA [6]. The deepest aquifer was situated at 1270 m and both studied aquifers connected to arthesian wells. Microscopic counts were 103 to 105 cells mL−1 and sand traps sampled 105 to 109 cells g−1 . Microbial iron-reduction was demonstrated and the metabolic diversity of the populations was large. Later, six additional boreholes were investigated in the same basaltic structure [7]. It was concluded that autotrophic microorganisms outnumbered heterotrophic microorganisms. Almost two decades later, groundwater from a new drilled well of the Colombia river basalt group was analyzed for microbial diversity with 454 pyrosequencing and quantification performed with quantitative polymerase chain reaction (qPCR) [8]. The results indicated H2 oxidizers, methylotrophs, sulfur reducers, methanotrophas and methanogens. Sequences related to the H2 -oxidizing genus Hydrogenophaga were found in all samples. These diversity data corroborate the previous findings of autotrophs [7] and attests that aquifers in basalt rocks can be populated by a diverse assembly of microorganisms. Ophiolites are ultramafic rocks that have been uplifted from the oceanic crust or the upper mantle and exposed above the sea level. These rocks are often rich in the minerals pyroxene and olivine. In contact with water, ultramafic rocks are altered through serpentinization under the formation of H2 and if CO2 is present under strongly reducing conditions, CH4 and higher hydrocarbons can also be formed. The process of serpentinization and deep life has recently been thoroughly reviewed [9]. In brief, the serpentinization process results in extremely high pH above 10 – it can reach 12.7 [10]. Microbial life in ophiolites under serpentinization consequently experiences very extreme conditions with high pH and low availability of electron acceptors and nutrients [11]. However, there are large amounts of an energetic electron donor available in the form of H2 , which can be used by most microorganisms able to cope with these extreme conditions. The alkaline seeps at the Tablelands Ophiolite in Newfoundland showed presence of H2 utilizing bacteria [12]. Metagenomic analyses indicated potentially autotrophic betaproteobacteria belonging to order Burkholderiales as the

3.2 Windows into the terrestrial hard rock biosphere | 67

most likely H2 oxidizer in the shallow part of the springs. Anaerobic clostridia were indicated to dwell at depth. Subsequent pyrosequencing of the v4v6 16S rDNA revealed Hydrogenophaga sequences to dominate the samples [10]. This genus includes autotrophic H2 utilizers [13, 14]. Hydrogenophaga related 16S rDNA sequences was previously found in cementitious alkaline (pH 12.7) spring waters of Maqarin [15] which is in line with what was found in the Tablelands springs – some strains of Hydrogenophaga appear to be well adapted for a life in high pH environments. The Cedars is a serpentinization site in California, USA that has been investigated for presence of microbial life [16]. The pH reached 12 and H2 constituted up to 51% of the total volume of dissolved gas. Although the cell densities in the spring water was very low in the range from < 101 to 103 cells mL−1 , inserted glass slides in a flow of the alkaline water for between two and three weeks were colonized at 106 to 107 cells cm−2 which is equal to what has been observed with a similar methodology in deep granitic groundwater [17, 18]. Phylogenetic characterization of the microbial populations in The Cedars remains to be done. Both the investigation in the Tablelands and in The Cedars utilized natural springs for collection of samples and they, therefore, represented mixes of alkaline and nonalkaline groundwater which may have influenced the microbial populations. The discussion above (see Section 3.1) about hard rock aquifers from the perspective of microorganisms and favorable mixing points for attachment probably applies well on a serpentinization site. Positions where there is inflow of nonalkaline groundwater to alkaline, H2 -rich groundwater may offer gradients of electron donors and acceptors to alkaliphilic populations. The only way to investigate such positions in detail is to drill holes where aquifers can be packed off and investigated one by one. Life appears to have been present in hard, igneous rocks since the origin of life. Uplifted pillow lava rims in a South African, 3.5 billion-year-old greenstone belt contain mineralized tubes that provide evidence of microbial activity in volcanic rocks during the early history of Earth [19]. Not only do microbes appear to have lived in these rocks, they also actively dissolved and altered the volcanic glass to palagonite and other authigenic minerals [20, 21]. The commonly raised question “how did microbes get down and adapted to the extreme subsurface environment?” may be irrelevant. It is not unlikely that life had been present in aquifers of the crust of the earth ever since its origin. It may be that life came up from the underground when the hostile conditions that ruled the surface during the early days of our planet were mitigated over time. In any circumstance, the evidences of life in 3.5 billion-year-old hard volcanic rock suggest that hard rock aquifers may be one of the oldest habitats for microbial life on Earth.

68 | 3 Microbial life in terrestrial hard rock environments 3.2.4 Granites Aquifers in granitic rock can become completely filled with calcite that precipitate over time (󳶳 Fig. 3.1 (d)). During such a process, it is likely that attached microorganisms, if present, will be closed in and possibly fossilized. The analysis of calcite that filled an aquifer 207 m below ground at the Äspö Underground Rock Laboratory (URL), Sweden, indeed showed the presence of a fossilized biofilm [22]. The 𝛿34 S values for sulfur in sulfide were typical for biogenic sulfide and parts of the calcite had extremely low 𝛿13 C values indicative of heterotrophic microbial activity. Later, another calcite vein from a depth of 450 m at Äspö URL was investigated in greater detail [23]. Further evidences for ancient microbial activity in deep granitic aquifers were obtained. These results provide confidence that findings of present day microbial life and activity in granitic aquifers are much more than a contamination artifact. The exploration of microbial life in granitic aquifers has gained a tremendous advantage from present day plans to dispose High Level Radioactive Wastes (HRLW), mainly spent fuel, in underground repositories at depths of 450 m or more. Underground Rock Laboratories in granitic rock have been built in Canada (Whiteshell URL, closed in 2010), Finland (ONKALO URL), Sweden (Äspö URL) and Switzerland (the Grimsel test site) and another one is under construction in Japan (Mizunami URL). As discussed for sampling methods above, the URL tunnels offer excellent opportunities to drill and sample aquifers. Because these URLs are research facilities, hydrological and geochemical sampling programs are extensive and they generate a wealth of background information about the sites. Further, because the quest to build HRLW repositories that are safe for several hundreds of future generations to come takes very long time, many of the research programs last for decades. This is the case for microbiological investigations in Finland and Sweden where microbiology has been investigated for over two decades by now. The first URL in Sweden was situated in a closed iron mine in Stripa from 1976 until 1990 when it was replaced by the Äspö URL. Working in mines in the absence of a URL certainly offers a good window into the deep biosphere, but it comes with many issues about contamination and questions regarding the influence from mining on the groundwater environment. The construction of tunnels intersects fractures and induces groundwater movements into the tunnels which will drown the mine, or the URL, unless the groundwater is pumped out. Movement of groundwater may disturb microbial processes in the aquifers and it can be difficult to extrapolate results to an undisturbed situation. The problem can be mitigated by selection of aquifers that show little or no influence from the tunneling effects. A very straightforward approach is to select aquifers with groundwater pressures that equal the expected pressure with respect to the natural groundwater table. The greater difference, the greater risk there is for disturbed flow conditions. The actual disturbance will be site specific and must be judged for each case.

3.2 Windows into the terrestrial hard rock biosphere | 69

The Stripa mine had a borehole that started at the deepest mining level (410 m) and continued to 1240 m depth. This borehole was investigated for microbial numbers and activity in three packed-off sections that were placed much deeper than the deepest level of the mine and therefore, the disturbance from the mining activities was assumed to be small or negligible. The deepest sample site (970–1240 m) had 105 cells mL−1 , which was 10 to 100 more than what was observed in the more shallow sites of the borehole [24]. Likewise, the microbes at the deepest site were more metabolically active than those of the shallower sites [25]. Betaproteobacteria related to Ralstonia and Comamonas were detected using 16S rDNA sequencing [18]. These investigations showed no depth relation between numbers and activity of microorganism; the deepest sample was the most densely populated. Because there were only three depths investigated, general conclusions about a depth relation could not be drawn. However, continued investigations on other sites showed similar absence of a depth relation, as will be referred to below. Obviously, each and every aquifer has its own microbial characteristics that follow other principles than just the distance to the ground surface. Prior to construction of URLs, site investigations are performed that include drilling of boreholes. The results can be used as references to data obtained from boreholes in the constructed URL. During the pre-investigations for the Äspö URL, several 1000 m deep boreholes were investigated for numbers and activity of microorganisms. Total numbers were in the range from 105 up to 106 cells mL−1 and 1 to 10% could be cultured using aerobic plate counts [26]. The use of radiotracers and microautoradiography demonstrated that the microbes were active and metabolized short organic acids; attached microorganisms were more active per cell than unattached microorganisms [17]. During construction, microbiology was investigated as the tunnel was driven deeper to its final depth of 460 m. These investigations corroborated the results from surface investigations with respect to numbers of attached and unattached microorganisms [27]. Cloning and sequencing of 16S rDNA showed presence of Acinetobacter, Desulfovibrio, Thiomicrospira and Bacillus to mention the most frequent observations. Homoacetogenic bacteria and methanogens frequently occurred along the tunnel [28] and three new species were isolated and described, Methanobacterium subterraneum [29], Desulfovibrio aespoeensis [30] and Methylomonas scandinavica [31]. At the time of completion of the Äspö URL tunnel, there was a good understanding of microbial abundance, activity and diversity in the aquifers in a range from 70 to 450 m depth. The selection process for a Swedish HLWR included two sites, Laxemar and Forsmark, that both were investigated for microbiology via boreholes [32]. The selected site became Forsmark, north of Stockholm. The construction of the ONKALO URL in Finland started in 2004 and there was, just as for the Äspö URL, thorough pre-investigations of microbial presence and diversity performed at four different sites; the selected site became Olkiluoto [33, 34]. The results generally matched those obtained during the pre-investigations of Äspö URL. During the construction of the ONKALO tunnel in Olkiluoto, investigations continued

70 | 3 Microbial life in terrestrial hard rock environments and groundwater microbiology was investigated over time [35, 36]. Different from Forsmark and Laxemar, the Olkiluoto area showed a pronounced variability of microbial life over depth with a clear peak in culturable microorganisms and diversity between 250 and 350 m depth. This peak coincided with a very sharp sulfate–methane transition zone (SMTZ) with sulfate-rich and methane-poor groundwater between 0 and 250 m depth, methane-rich and sulfate-poor groundwater deeper than 350 m and a mixing situation between 250 and 350 m with sulfate-rich and methane-rich groundwater. A second peak in bacterial numbers was, expectedly, found between 0 to 25 m depth in Olkiluoto. Investigations of a deep borehole in Outokumpu, Eastern Finland, again echoed results from other deep granitic aquifers [37]. Anaerobic sulfate-reducing bacteria and Clostridia were detected at depths down to 1500 m. In short, the investigations of Fennoscandian shield groundwater at four sites in Sweden (Forsmark, Laxemar, Stripa and Äspö) and five sites in Finland (Hästholmen, Kivetty, Olkiluoto, Outokumpu, Romuvaara) showed a large diversity of cultivable microorganisms as deep as the investigations were performed, i.e. approximately 1500 m. A wide variety of aerobic and anaerobic physiological groups of microorganisms were found including oxygen- nitrate- iron- manganese- sulfate-reducing bacteria, acetogens and methanogens. Occasionally, fungi were also observed [38]. Studies of microbiology of granites in other parts of the world show results that reflect what have been found in the Fennoscandian Shield. Investigations of microbial numbers and diversity in aquifers of the Canadian Shield granite was first performed in the Whitshell URL as summarized elsewhere [39]. The total numbers of microorganisms in groundwater were similar to what has been reported for the Fennoscandian Shield. Diversity was analyzed for aerobic, cultivable bacteria and various strains related to Pseudomonas were reported. Drilling and sampling of a granitic rock core at 2 292 m depth in the Henderson mine revealed a large diversity of Proteobacteria, Firmicutes and some novel linages. Archaea was also detected [40]. The authors concluded that their results were consistent with other studies of deep subsurface microbial ecology. Geomicrobiological investigations of groundwater from 1 169 m depth where the Mizunami URL is under construction again showed presence of a wide diversity of microorganisms and cell densities of approximately 5 × 104 cells mL−1 [41]. Groundwater from the Grimsel test site is very diluted because of its high altitude. However, novel lineages of bacteria were found [42]. In summary, investigations of microbial abundance and diversity in deep granitic aquifers around the world point in the same direction. Diversified life is present and active as deep as the investigations stretch, i.e. several thousand meters.

3.2.5 Hard rocks of varying origin Very deep mines in South Africa penetrate several different types of rock, layered over depth with sandstones, sediments and dolomites on top and igneous basalts

3.3 Energy from where? | 71

and metamorphic quartzites at large depths; all with crosscutting, highly permeable dykes. The diversity of microbial populations in groundwater from more than 3000 m depth has been systematically mapped. Novel archaeal communities, distinct from what is found on the surface, were suggested to live in groundwater of the explored goldmines [43]. Cloning of DNA samples from > 3300 m deep sites in alkaline (pH 9.1) 54–60 °C groundwater of the Driefontein mine showed presence of sequences related to Desufotomaculum and Methanobacterium [44]. Further investigations of microbial 16S rDNA diversity in the South African Kalahari Shield showed presence of low bacterial and archaeal diversity [45]. It appeared as if the long-term isolation from the photosphere within Earth’s crust can reduce microbial diversity to single species ecosystems [46]. In opposite to the South African mines, a diversity investigation of saline groundwater from 890 and 1130 m depth below 500 m of permafrost did not reveal archaeal sequences [47]. Instead, the most abundant genera detected related to the sulfate-reducing genus Desulfosporosinus, Halothiobacillus and Pseudomonas.

3.3 Energy from where? Aquifers in hard rock are isolated from the photosphere, but organic material can follow recharging water into the subsurface. The rate and amount of this process varies very much from site to site. Generally, sites close to the sea will be less influenced by intrusion of water with surface-generated organic material compared to sites at high altitudes where the hydraulic gradients for groundwater will be much larger. Salinity of groundwater often increases with depth which may stabilize deep groundwater and reduce mobility to the surface and some fluids at depth can be billions of years old [48]. The influence from the surface will consequently decrease with depth and the flow rates of groundwater in aquifer below the sea level will be very slow, unless there are strong thermal gradients such as those in hydrothermal areas. In deep stagnant aquifers without contact with photosynthesis based surface ecosystems, survival of microorganisms at very slow rates may be possibly analogous to what has been postulated for microorganisms found in subseafloor sediments [49], but can there also be microbial activity in millions of years old groundwater? The content of organic material in hard rocks is, in opposite to many sedimentary systems, very low or none. The active life observed in deep aquifers of hard rock must then rely on sources of energy alternative to those from solar-driven autotrophic processes. This riddle was first discussed by Thomas Gold [50] who suggested that a widespread deep hot biosphere exists at depth in the crust of Earth and that this biosphere is sustained by energy from chemical sources, mainly reduced gases such as H2 and methane in fluids from deep crustal layers. Focus of the scientific community turned towards the possibilities of a chemoautotrophic deep biosphere [51, 52]. The deep biosphere hypothesis by Thomas Gold have since then been tested by many independent inves-

72 | 3 Microbial life in terrestrial hard rock environments tigations that mostly confirm, or at least support, the presence of a deep biosphere in the crust of Earth that is independent of surface photosynthetic ecosystems.

3.3.1 Deep reduced gases The presence of reduced gases H2 and methane with abiotic origins have repeatedly been found in deep granitic groundwater [53, 54]. In some cases, part of the methane could be of biogenic origin with abiogenic H2 as source of electrons for the reduction of carbon dioxide [55]. The modeling of sources of precursors and processes of formation of hydrocarbons are complicated [56] and there is a choice of several different possible processes that will vary from site to site [57]. In some places, radiolysis of water has been suggested as the main source of H2 [58] and in others serpentinization of ultramafic rocks generates large amounts of H2 [9]. Migration of deep methane to the surface through fault and fracture zones in Fennoscandian Shield has been documented [59–61] which shows that there can be a flow of gases from deep crustal layers to the ground in line with Thomas Gold hypothesis. Consequently, there is now little doubt that the electron donors H2 and methane are generally present in deep hard hock.

3.3.1.1 H2 Subsurface lithoautotrophic microbial ecosystems (SLIMEs) were first suggested to exist in crystalline rock aquifers within the Colombia river basalt group [7]. These ecosystems were suggested to be independent of photosynthetic primary production. Instead, a continuous energy source in the form of H2 was inferred from the formation of H2 when crushed basaltic rocks were exposed to anoxic groundwater in the laboratory. The SLIME hypothesis encountered some opposition arguing that the suggested H2 production rate was too slow to support autotrophy as discussed elsewhere [62]. However, an increasing number of reports later suggested the existence of H2 based SLIMEs in hot springs [63, 64], deep mine water [65] and of course in serpentinization environments [12, 66]. In the early days of SLIMEs, interest was focused on autotrophic methanogens. Since then, other groups of H2 lithoautotrophs such as sulfate-reducing bacteria, Hydrogenophaga and acetogens have been found in deep groundwater which increases the diversity of autotrophs in SLIMEs. Basically, the existence of H2 based SLIMEs will depend on the supply of H2 , something that can be difficult to measure, because the presence of H2 utilizers will keep the concentration of H2 at a level at or below K𝑚 , for H2 i.e. less than 1 μM [67]. We need to find ways to analyze the supply rate of H2 over time.

3.4 Activity

| 73

3.3.1.2 Methane Anaerobic oxidation of methane (AOM) with sulfate as electron donor is a well described process for many marine sediment systems [68] but the presence of AOM in terrestrial hard rock environments remains to be proven. Data from investigations of an approximately 100 m wide sulfate–methane transition zone (SMTZ) in Olkiluoto, Finland suggested the presence of an AOM process but conclusive evidence is still pending [69, 70]. While the concentration of H2 in most deep groundwater is low and thereby appears limiting, methane can be found in much larger concentrations [35]. It appears as if it is the availability of an electron acceptor that limits AOM activity. While the fixation of carbon dioxide with H2 is clearly a lithoautotrophic process, AOM utilizes a reduced hydrocarbon molecule which is not a true autotrophic process. However, as long as the methane is of an abiogenic origin, AOM-based ecosystems are just as independent of photosynthetic primary production, as are H2 based SLIMEs.

3.4 Activity Microbial activity can be analyzed by direct and indirect methods. Analysis of stable isotope ratios [71] and the presence of reduced electron acceptors are examples of indirect methods and these methods mainly report past microbial activities rather than present time activities. The use of radiolabeled electron donors and acceptors and gene expression analysis are examples of direct methods that report microbial activity in present time. While indirect methods only require good sampling conditions, direct methods require incubation under in situ conditions to be reliable.

3.4.1 Stable isotopes Stable sulfur (𝛿34 S) and carbon (𝛿13 C) and oxygen (𝛿18 O) isotopic signatures were successfully used to establish that microbial sulfate-reducing activity had occurred in the ancient past in hard rock aquifers of Äspö URL [22, 23]. When 𝛿34 S signatures of other Fennoscandian deep groundwater sites were evaluated, it was concluded that sulfatereduction had occurred in the past [32]. Analysis of 𝛿34 S of sulfate in a present day microbial sulfate-reducing system 420 m underground in ONKALO URL confirmed that SRB were active and sulfate was biologically reduced to sulfide [69]. However, the interpretation of 𝛿34 S must take into account the large variation in the fractionation factor by different SRB which can span from 2 to 42‰ [72]. Therefore, interpretation of values of 𝛿34 S as present day as well as past time activity-indicators must be done with caution unless the SRB diversity is known. Signatures of 𝛿13 C in methane and carbon dioxide can be used to reveal the origin on methane in deep groundwater [55]. In short, stable isotope signatures can be very helpful in the evaluation of past time activity, but less valuable for the analysis of present day activity in hard rock aquifers.

74 | 3 Microbial life in terrestrial hard rock environments 3.4.2 Geochemical indicators Elevated concentrations of reduced electron acceptors such as sulfide (from sulfate) and ferrous iron (from ferric iron), or disappearance of electron donors expected to be present based on geochemical profiles, such as methane, can indicate microbial activity. A very clear example of this was obtained in Olkiluoto, Finland, where a peak in sulfide concentration at 250–350 m depth coincides with a significant increase in the concentration of methane and a loss of sulfate just below the sulfide peak [35]. This particular profile very much resembles the SMTZ found in marine systems where AOM with sulfate is ongoing. Several in situ experiments do indeed support the hypothesis of the existence of an AOM process in Olkiluoto groundwater, but conclusive evidence is lacking, yet [69, 70]. Sequence data showed that up to 60% of the bacterial 16S rDNA pool affiliated with SRB genera in the SMTZ. It can be very difficult to extrapolate depth profiles of microbial activity in hard rock, because fractures are discrete and there can be large variation in geochemistry over short distances. However, when many sample points have been analyzed (at least 25 but preferably more), there is a good chance that indications of profiles can be obtained. Just like in sediments, profiles may suggest what microbial activity and process is dominating over depths, the Olkiluoto case is obvious, in other places, a profile may be less pronounced and it is more difficult to predict where various microbial processes are active [32].

3.4.3 In vitro activity Radiolabeled compounds can be used for the estimation of microbial activity and methods to do so have been in use in microbial ecology for many decades [73]. Populations or single cells can be investigated. Using an array of radiolabeled compounds, the processes of autotrophic methane and acetate formation [28], and transformation of carbon dioxide, formate, acetate, lactate glucose and leucin were investigated in vitro [17, 24–26]. Details and discussion about these experiments have been thoroughly summarized and discussed previously, and are not repeated here [1]. In short, it was shown that autotrophic and heterotrophic microbial populations were viable and active in hard rock aquifers, able to metabolize many different organic and inorganic compounds. The results from these experiments formed the baseline for development and use of in situ experimental systems designed to further study the processes suggested by in vitro experiments to be ongoing in deep hard rock aquifers.

3.4.4 In situ activity Studies of in situ activity in the deep biosphere are difficult to perform. To be truly in situ, pressure must be retained which can be done by “downhole” experiments or ex-

3.4 Activity

| 75

periments in underground tunnels and mines. A methodology comprising three flow cell cabinets (FCCs) mimicking hard rock aquifers was developed and tested in the Äspö hard rock laboratory and mainly used to investigate sulfate-reducing microbial activity [74]. Groundwater can be slowly pumped under in situ pressure and chemistry conditions through flow cells with solid materials for attachment growth of microorganisms in the groundwater. Rates of acetate formation and consumption and the influence of H2 on microbial activity were first investigated [75]. The effects of addition of 3 mM H2 or 2.4 mM acetate were compared with an untreated control. H2 addition reduced the generation time fourfold to two weeks, doubled the sulfide production rate and increased acetate production by approximately 50%. The acetate addition induced acetate consumption. The influence of varying concentrations H2 was studied next with the aim to establish K𝑚 for H2 with sulfate as electron acceptor [67]. A threshold concentration of approximately 1 μM H2 was registered at which sulfate reduction ceased, despite the presence of DOC and acetate, suggesting that H2 was needed for sulfate-reducing activity. Adding excess H2 up to 10 mM only had a small effect on the observed rates, suggesting that the sulfate-reducing population was metabolizing at 𝜇max and it was speculated that phages, i.e. viruses, controlled population density and thereby the overall rates of sulfate-reduction, as discussed in more detail below. The FCCs were moved to the ONKALO tunnel in Olkiluoto and the influence of H2 and methane on sulfate-reduction was investigated [69]. The influence on microbial metabolic activity of 11 mM methane, 11 mM methane plus 10 mM H2 or 2.1 mM O2 plus 7.9 mM N2 (that is, air) was studied. The FCC containing H2 and methane displayed microbial reduction of 0.7 mM sulfate to sulfide, whereas the system containing only methane resulted in 0.2 mM reduced sulfate. The system containing added air became inhibited and displayed no signs of microbial activity. It was concluded likely that a microbial anaerobic methane-oxidizing process coupled to acetate formation and sulfate reduction may be ongoing in aquifers at a depth of 250–350 m in Olkiluoto. Deeper, under the 250–350 m level in the ONKALO tunnel, sulfate is absent in the groundwater and the amount of dissolved methane can be 500 mL L−1 , increasing with depth to 1000 mL L−1 or more. According to 454 pyrosequencing of 16S rDNA, and cultivation attempts, SRB are absent in the deep sulfate-free groundwater [70]. With the indispensable support of sequence data, it was found that sulfate was the only component needed to trigger sulfate reduction in deep sulfate-poor, methane-rich groundwater. A similar conclusion was drawn from experiments in the Äspö URL, where a minute variation in Eℎ caused a community to transition from an iron-oxidizing to a sulfideoxidizing community (Figure 1 in Ref. [76]). The FCC research has demonstrated that variation of only one or a few geochemical parameters of a groundwater can induce very large community transitions with concomitant changes in microbial activity in hard rock aquifers.

76 | 3 Microbial life in terrestrial hard rock environments 3.4.5 Phages may control activity rates The presence of viruses in groundwater that attack microorganisms, phages, must originate from lytic infections of host microorganisms. A lytic infection can only occur in metabolically active microorganisms, which implies that the more phages there are per microorganism, the more active the microbial ecosystem must be. There is now an increasing interest for understanding interactions between phages and the genomes of their host in the subsurface biosphere [77]. The impact of viruses on microbial communities has recently been reviewed [78]. Phages and viruses are inevitable components in all known, naturally evolving ecosystems, and there is now growing evidence that this is also valid for the deep biosphere. Investigation of Äspö URL groundwater for phage abundance returned large numbers of a diverse phage population [79]. The average ratio of virus-like particles (VLP) over the total number of cells (TNC) was 12 indicating active microbial populations. Transmission electron microscopy revealed four distinct bacteriophage groups with at least seven phage families of which some are known to be lytic. The results consequently showed the presence of viruses in deep granitic groundwater up to 450 m depth. If phages in deep groundwater are active and lytic, they will constitute an important group of predators that might control the numbers and activity of microorganisms. Further, their presence suggests that their prey, the microorganisms, are active and growing. The effect from phage activity was studied under in situ-controlled conditions with microorganisms from the ONKALO URL [69]. It was found that microbial activity and the VLP/TNC quotient correlated positively. Consequently, there is growing evidence for the presence of a viral mechanism that controls population density in hard rock aquifers, similar in function to a viral shunt [80]. In addition, phage influence on gene transfer via transformation and transduction will of course be of large importance for genetic exchange in the deep biosphere. The TNC in repeated experiments performed under in situ conditions never exceeded 106 cells mL−1 despite unlimited access to electron donors and acceptors [67, 69, 75]. It was hypothesized that bacteriophages exerted a significant mitigating effect on the numbers of microbial cells. Taken together, it seems safe to conclude that it is bacteriophages that control the number of cells in deep Fennoscandian hard rock aquifers, between 5 × 103 and 1 × 106 cells mL−1 [32, 35, 74], just as they do in marine microbial ecosystems [81].

3.5 What’s next in the exploration of microbial life in deep hard rock aquifers? It may appear contradictory that while most of the terrestrial part of our planet is composed of igneous hard rocks, the knowledge about the deep biosphere in aquifers of these rocks is limited to few places on the continents, in comparison to the wealth

References | 77

of knowledge regarding the deep subseafloor biosphere. Likely, this is mainly due to it being expensive and resource demanding to explore deep, terrestrial hard rock aquifers. The International Continental Drilling Program (ICDP) is designed to deal with large scale drilling and is slowly but steadily adding new sites to our knowledge base. Next hard rock drilling will take place in the mid of Sweden and drill to 2500 m depth during 2014 [82]. Microbial biodiversity of groundwater ecosystems is large [83], and the majority of our diversity data is based on cloning and sequencing of genes in extracted DNA. The growing power of high-throughput sequencing methods will add a new dimension to our understanding of the deep biosphere. There is a current effort to sequence up to 17 subsurface sites around the world with mostly identical primers for Bacteria and Archaea. This effort is denoted the Census of Deep Life project and the first results have already appeared [8, 10, 70, 84]. When all sites are in the database, a first coordinated global evaluation of diversity of the deep biosphere can be performed.

References [1]

Pedersen K. Diversity and activity of microorganisms in deep igneous rock aquifers of the Fennoscandian Shield. In: Subsurface microbiology and biogeochemistry. Fredrickson JK, Fletcher M. New York: Wiley-Liss Inc.; 2001, pp 97–139. [2] Kieft TL, Phelps TJ, Fredrickson JK. Drilling, coring and sampling subsurface environments. In: Manual of environmental Microbiology. Hurst CJ, Crawford RL, Garland JL, Lipson DA, Mills AL, Stetzenbach LD. Washington DC: ASM Press; 20073, pp 799–817. [3] Pedersen K, Hallbeck L, Arlinger J, Erlandson A-C, Jahromi N. Investigation of the potential for microbial contamination of deep granitic aquifers during drilling using 16S rRNA gene sequencing and culturing methods. J Microbiol Methods 30 No. 3 (1997), 179–192. [4] Loren A, Hallbeck L, Pedersen K, Abrahamsson K. Determination and distribution of diesel components in igneous rock surrounding underground diesel Storage Facilities in Sweden. Environ Sci Technol 35 (2001), 374–378. [5] Jägevall S, Rabe L, Pedersen K. Abundance and diversity of biofilms in natural and artificial aquifers of the Äspö Hard Rock Laboratory, Sweden. Microb Ecol 61 (2011), 410–422. [6] Stevens TO, Mckinley JP, Fredrickson JK. Bacteria associated with deep, alkaline, anaerobic groundwaters in Southeast Washington. Microb Ecol 25 (1993), 35–50. [7] Stevens TO, McKinley JP. Lithoautotrophic microbial ecosystem in deep basalt aquifers. Science 270 (1995), 450–453. [8] Lavalleur HJ, Colwell FS. Microbial characterization of basalt formation waters targeted for geological carbon sequestration. FEMS Microbiol Ecol 85 (2013), 62–73. [9] Schrenk MO, Brazelton WG. Serpentinization, carbon and deep life. Rev. Mineral. Geochem 75 (2013), 575–606. [10] Brazelton WJ, Morrill PL, Szponar N, Schrenk MO. Bacterial communities associated with subsurface geochemical processes in continental serpentinite springs. Appl Environ Microbiol 79 (2013), 3906–3916.

78 | 3 Microbial life in terrestrial hard rock environments [11] Szponar N, Brazelton WJ, Schrenk MO, Bower DM, Steele A, Morrill PL. Geochemistry of a continental site of serpentinization, the Tablelands Ophiolite, Gros Morne National Park: A Mars analogue. Icarus 224 (2013), 286–296. [12] Brazelton WJ, Nelson B, Schrenk MO. Metagenomic evidence for H2 oxidation and H2 production by serpentinite-hosted subsurface microbial communities. Front Microbiol 2 (2012), Article 268, 1–16. [13] Willems A, Busse J, Goor M et al. Hydrogenophaga, a new genus of hydrogen-oxidizing bacteria that includes Hydrogenophaga flava comb. nov. (formerly Pseudomonas flava), Hydrogenophaga palleronii (formerly Pseudomonas palleronii), Hydrogenophaga pseudoflava (formerly Pseudomonas pseudoflava and “Pseudomonas carboxydoflava”), and Hydrogenophaga taeniospiralis (formerly Pseudomonas taeniospiralis). Int J Syst Bacteriol 39 No. 3, (1989), 319–333. [14] Yoon K-S, Tsukada N, Sakai Y, Ishii M, Igarashi M, Nishihara H. Isolation and characterization of a new facultatively autotrophic hydrogen-oxidizing Betaproteobacterium, Hydrogenophaga sp. AH-24. FEMS Microbiol Lett 278 (2008), 94–100. [15] Pedersen K, Nilsson E, Arlinger J, Hallbeck L, O’Neill A. Distribution, diversity and activity of microorganisms in the hyper-alkaline spring waters of Maqarin in Jordan. Extremophiles 8 (2004), 151–164. [16] Morrill PL, Kuenen JG, Johnson OJ et al. Geochemistry and geobiology of a present-day serpentinization site in California: The Cedars. Geochim Cosmochim Acta 109 (2013), 222–240. [17] Pedersen K, Ekendahl S. Assimilation of CO2 and introduced organic compounds by bacterial communities in ground water from Southeastern Sweden deep crystalline bedrock. Microb Ecol 23 (1992), 1–14. [18] Ekendahl S, Arlinger J, Ståhl F, Pedersen K. Characterization of attached bacterial populations in deep granitic groundwater from the Stripa research mine with 16S-rRNA gene sequencing technique and scanning electron microscopy. Microbiology 140 (1994), 1575–1583. [19] Furnes H, Banerjee NR, Muehlenbachs K, Staudigel H, de Wit M. Early life records in Archaean pillow lavas. Science 304 (2004), 578–581. [20] Thorseth IH, Furunes H, Heldal M. The importance of microbiological activity in the alteration of natural basaltic glass. Geochim Cosmochim Acta 56 (1992), 845–850. [21] Staudigel H, Furnes H, McLoughlin N, Banerjee NR, Connell LB, Templeton A. 3.5 billion years of glass bioalteration: Volcanic rocks as a basis for microbial life? Earth Sci Rev 89 (2008), 156–176. [22] Pedersen K, Ekendahl S, Tullborg E-L, Furnes H, Thorseth I-G, Tumyr O. Evidence of ancient life at 207 m depth in a granitic aquifer. Geology 25 (1997), 827–830. [23] Heim C, Lausmaa J, Sjövall P et al. Ancient microbial activity recorded in fracture fillings from granitic rocks (Äspö Hard Rock Laboratory, Sweden). Geobiology 10 (2012), 280–297. [24] Pedersen K, Ekendahl S. Incorporation of CO2 and introduced organic compounds by bacterial populations in groundwater from the deep crystalline bedrock of the Stripa mine. J Gen Microbiol 138 (1992), 369–376. [25] Ekendahl S, Pedersen K. Carbon transformations by attached bacterial populations in granitic ground water from deep crystalline bed-rock of the Stripa research mine. Microbiology 140 (1994), 1565–1573. [26] Pedersen K, Ekendahl S. Distribution and activity of bacteria in deep granitic groundwaters of southeastern Sweden. Microb Ecol 20 (1990), 37–52. [27] Pedersen K, Arlinger J, Ekendahl S, Hallbeck L. 16S rRNA gene diversity of attached and unattached groundwater bacteria along the access tunnel to the Äspö Hard Rock Laboratory, Sweden. FEMS Microbiol Ecol 19 (1996), 249–262.

References | 79

[28] Kotelnikova S, Pedersen K. Distribution and activity of methanogens and homoacetogens in deep granitic aquifers at Äspö Hard Rock Laboratory, Sweden. FEMS Microbiol Ecol 26 (1998), 121–134. [29] Kotelnikova S, Macario AJL, Pedersen K. Methanobacterium subterraneum, a new species pf Archaea isolated from deep groundwater at the Äspö Hard Rock Laboratory, Sweden. Int J Syst Bacteriol 48 (1998), 357–367. [30] Motamedi M, Pedersen K. Desulfovibrio aespoeensis sp. nov. a mesophilic sulfate-reducing bacterium from deep groundwater at Äspö hard rock laboratory, Sweden. Int J Syst Bacteriol 48 (1998), 311–315. [31] Kalyuzhnaya MG, Khmelenina VN, Kotelnikova S, Holmquist L, Pedersen K, Trotsenko YA. Methylomonas scandinavica, sp. nov., a new methanotrophic psychrotrophic bacterium isolated from deep igneous rock ground water of Sweden. Syst Appl Microbiol 22 (1999), 565– 572. [32] Hallbeck L, Pedersen K. Culture-dependent comparison of microbial diversity in deep granitic groundwater from two sites considered for a Swedish final repository of spent nuclear fuel. FEMS Microbiol Ecol 81 (2012), 66–77. [33] Haveman SH, Pedersen K, Routsalainen P. Distribution and metabolic diversity of microorganisms in deep igneous rock aquifers of Finland. Geomicrobiol J 16 (1999), 277–294. [34] Haveman SA, Pedersen K. Distribution of culturable anaerobic microorganisms in Fennoscandian shield groundwater. FEMS Microbiol Ecol 39 (2002), 129–137. [35] Pedersen K, Arlinger J, Hallbeck A, Hallbeck L, Eriksson S, Johansson J. Numbers, biomass and cultivable diversity of microbial populations relate to depth and borehole-specific conditions in groundwater from depths of 4 to 450 m in Olkiluoto, Finland. ISME J 2 (2008), 760–775. [36] Nyyssönen M, Bomberg M, Kapanena A, Nousiainena A, Pitkänen P, Itävaara M. Methanogenic and sulfate-reducing microbial communities in deep groundwater of crystalline rock fractures in Olkiluoto, Finland. Geomicrobiol J 29 (2012), 863–878. [37] Itävaara M, Nyyssönen M, Kapanen A, Nousiainen A, Ahonen L, Kukkonen I. Characterization of bacterial diversity to a depth of 1500 min the Outokumpu deep borehole, Fennoscandian Shield. FEMS Microbiol Ecol 77 (2011), 295–309. [38] Ekendahl S, O’Neill AH, Thomsson E, Pedersen K. Characterization of yeasts isolated from deep igneous rock aquifers of the Fennoscandian Shield. Microb Ecol 46 (2003), 416–428. [39] Stroes-Gascoyne S, West JM. Microbial studies in the Canadian nuclear fuel waste management program. FEMS Microbiol Rev 20 (1997), 573–590. [40] Sahl JW, Schmidt R, Swanner E et al. Subsurface microbial diversity in deep-granite-fracture water in Colorado. Appl Environ Microbiol 74 (2008), 143–152. [41] Fukuda A, Hagiwara H, Ishimura T et al. Geomicrobiological properties of ultra-deep granitic groundwater from the Mizunami underground research laboratory (MIU), central Japan. Microb Ecol 60 (2010), 214–225. [42] Konno U, Koukuka M, Komatsu DD et al. Novel microbial populaitoon in deep granite groundwater from Grimsel test site, Switzerland. Microb Ecol 65 (2013), 626–637. [43] Takai K, Moser DP, DeFlaun M, Onstott TC, Fredrickson JF. Archaeal diversity in waters from deep South African gol mines. Appl Environ Microbiol 67 (2001), 5750–5760. [44] Moser DP, Gihring TM, Brockman FJ et al. Desulfotomaculum and Methanobacterium spp. dominate a 4- to 5-kilometer-deep fault. Appl Environ Microbiol 71 (2005), 8773–8783. [45] Gihring TM, Moser DP, Lin L-H et al. The distribution of microbial taxa in the subsurface water of the Kalahari Shield, South Africa. Geomicrobiol J 23 (2006), 415–430. [46] Chivian D, Brodie EL, Alm EJ et al. Environmental genomics reveals a single-species ecosystem deep within Earth. Science 322 (2008), 275–278.

80 | 3 Microbial life in terrestrial hard rock environments [47] Onstott TC, McGown DJ, Bakermans C et al. Microbial communities in subpermafrost saline fracture water at the Lupin Au Mine, Nunavut, Canada. Microb Ecol 58 (2009), 786–807. [48] Holland G, Sherwood Lollar B, Li L, Lacrampe-Couloume G, Slater GF, Ballentine CJ. Deep fracture fluids isolated in the crust since the Precambrian era. Nature 497 (2013), 357–360. [49] D’Hondt S, Rutherford S, Spivack AJ. Metabolic activity of subsurface life in deep-sea sediments. Science 295 (2002), 2067–2070. [50] Gold T. The deep hot biosphere. Proc Natl Acad USA 89 (1992), 6045–6049. [51] Stevens T. Lithoautotrophy in the subsurface. FEMS Microbiol Rev 20 (1997), 327–337. [52] Pedersen K. Exploration of deep intraterrestrial life - current perspectives. FEMS Microbiol Lett 185 (2000), 9–16. [53] Sherwood Lollar B, Voglesonger K, Lin L-H et al. Hydrogeologic controls on episodic H2 release from Precambrian fractured rocks—energy for deep subsurface life on Earth and Mars. Astrobiology 7 (2007), 971–986. [54] Sherwood Lollar B, Frape SK, Weise SM, Fritz P, Macko SA, Welhan JA. Abiogenic methanogenesis in crystalline rocks. Geochim Cosmochim Acta 57 (1993), 5087–5097. [55] Sherwood Lollar B, Frape SK, Fritz P et al. Evidence for bacterially generated hydrocarbon gas in Canadian shield and Fennoscandian shield rocks. Geochim Cosmochim Acta 57 (1993), 5073–5085. [56] Sherwood Lollar B, Lacrampe-Couloume G, Voglesonger K, Onstott TC, Pratt LM, Slater GF. Isotopic signatures of CH4 and higher hydrocarbon gases from Precambrian Shield sites: A model for abiogenic polymerization of hydrocarbons. Geochim Cosmochim Acta 72 (2008), 4778–4795. [57] Apps JA, van de Kamp PC. Energy gases of abiogenic origin in the Earth’s crust. In: The future of Energy gases, U.S. geological Survey Professional Papers. Howell G. Washington: United States Government Printing Office; 19931570, pp 81–132. [58] Lin L-H, Slater GF, Sherwood Lollar B, Lacrampe-Couloumbe G, Onstott TC. The yield and isotopic composition of radiolytic H2 , a potential energy source for the deep subsurface biosphere. Geochim Cosmochim Acta 69 No. 4, (2005), 893–903. [59] Söderberg P, Flodén T. Pockmark development along a deep crustal structure in the northern Stockholm Archipelago, Baltic Sea. Beiträge Meereskunde 62 (1991), 79–102. [60] Söderberg P, Flodén T. Gas seepages, gas eruptions and degassing structures in the seafloor along the Strömma tectonic lineament in the crystalline Stockholm Archipelago, east Sweden. Cont Shelf Res 12 (1992), 1157–1171. [61] Flodén T, Söderberg P. Shallow gas traps and gas migrations models in crystalline bedrock areas offshore Sweden. Baltica 8 (1994), 50–56. [62] Nealson KH, Inagaki F, Takai K. Hydrogen-driven subsurface lithoautotrophic microbial ecosystems (SLIMEs): do they exist and why should we care? Trends Microbiol 13 (2005), 405–410. [63] Chapelle FH, O’Neill K, Bradley PM et al. A hydrogen-based subsurface microbial community dominated bymethanogens. Nature 415 (2002), 312–315. [64] Spear JR, Walker JJ, McCollom TM, Pace NR. Hydrogen and bioenergetics in the Yellowstone geothermal ecosystem. Proc Natl Acad Sci USA 102 (2005), 2555–2560. [65] Lin L-H, Wang P-L, Rumble D et al. Long-term sustainability of a high-energy low-diversity crustal biome. Science 314 (2006), 479–482. [66] Sleep NH, Meibom A, Fridriksson T, Coleman RG, Bird DK. H2 -rich fluids from serpentinization: Geochemical and biotic implications. Proc Natl Acad Sci USA 101 (2004), 12 818–12 823. [67] Pedersen K. Influence of H2 and O2 on sulfate-reducing activity of a subterranean community and the coupled response in redox potential. FEMS Microbiol Ecol 82 (2012), 653–665. [68] Knittel K, Boetius A. Anaerobic oxidation of methane: Progress of an unknown process. Annu Rev Microbiol 63 (2009), 311–334.

References | 81

[69] Pedersen K. Metabolic activity of subterranean microbial communities in deep granitic groundwater supplemented with methane and H2 . ISME J 7 (2013), 839–849. [70] Pedersen K, Bengsson A, Edlund J, Eriksson L. Sulfate-induced diversity transition of a subterranean microbial community in methane-rich, sulfate-poor granitic groundwater. Geomicrobiol J (2014), doi:10.1080/01490451.2013.879508. [71] Des Marais DJ. Stable light isotope biogeochemistry of hydrothermal systems. In: Evolution of hydrothermal ecosystems on Earth (and Mars?). Bock GR, Goode JA. Chichester: John Wiley & Sons Ltd.; 1999, pp 83–98. [72] Detmers J, Brüchert V, Habicht K, Kuever J. Diversity of sulfur isotope fractionations by sulfatereducing prokaryotes. Appl Environ Microbiol 67 No. 2, (2001), 888–894. [73] Grigorova R, Norris JR. Techniques in microbial ecology. In: Methods in microbiology 22. Grigorova R, Norris JR. London: Academic Press; 1990, pp 627. [74] Hallbeck L, Pedersen K. Characterization of microbial processes in deep aquifers of the Fennoscandian Shield. Appl Geochem 23 (2008), 1796–1819. [75] Pedersen K. Subterranean microbial populations metabolize hydrogen and acetate under in situ conditions in granitic groundwater at 450 m depth in the Äspö Hard Rock Laboratory, Sweden. FEMS Microbiol Ecol 81 (2012), 217–229. [76] Anderson CR, James RE, Fru EC, Kennedy CB, Pedersen K. In situ ecological development of a bacteriogenic iron oxide-producing microbial community from a subsurface granitic rock environment. Geobiology 4 (2006), 29–42. [77] Anderson RE, Brazelton WJ, Baross JA. Is the genetic landscape of the deep subsurface biosphere affected by viruses? Frontiers in Microbiology 2 (2011), 1–15. [78] Anderson RE, Brazelton WJ, Baross JA. The deep viriosphere: Assessing the viral impact on microbial community dynamics in the deep subsurface. Rev Mineral Geochem 75 (2013), 649–675. [79] Kyle JE, Eydal HSC, Ferris FG, Pedersen K. Viruses in granitic groundwater from 69 to 450 m depth of the Äspö hard rock laboratory, Sweden. ISME J 2 (2008), 571–574. [80] Suttle CA. Virus in the sea. Nature 437 (2005), 356–361. [81] Suttle CA. marine viruses – major players in the global ecosystem. Nat Rev Microbiol (2007), 801–812. [82] Gee DG, Juhlin C, Pascal C, Robinson P. Collisional Orogeny in the Scandinavian Caledonides (COSC). GFF 132 (2010), 29–44. [83] Griebler C, Luederse T. Microbial diversity in groundwater systems. Freshw Biol 54 (2009), 649–677. [84] Marteinsson VT, Rúnarsson A, Stefánsson A et al. Microbial communities in the subglacial waters of the Vatnajökull ice cap, Iceland. ISME J 7 (2013), 427–437

Laurent Toffin, Karine Alain

4 Technological state of the art and challenges Cultivation of marine subseafloor microorganisms: state-of-the-art solutions and major issues

4.1 Basic concepts and difficulties inherent to the cultivation of subseafloor prokaryotes To grow, every microorganism requires substances for energy generation and cellular biosynthesis. Its environment or growth medium has to provide it with suitable energy sources, nutrients (C, N, P, S and other mineral elements), physical chemical conditions (T°, pH, pressure, etc.) and eventually growth factors [1]. Its growth is largely governed by the Gibbs free energy available and its own maintenance needs. Nowadays, almost the entire (close to 99%) microbial diversity from natural environments is as-yet-uncultured [2]. This low score is obviously not due to any unculturability of microorganisms but rather to the inherent difficulties to grow microorganisms in the laboratory. Indeed, the determination of nutritional requirements of microorganisms and the design of appropriate growth media and conditions involves intensive laboratory studies and prove to be difficult in many instances. In addition, many studies are aimed at isolating microorganisms in clonal culture. Yet it is somewhat paradoxical and sometimes a lost cause to try isolating microorganisms which grow in community and interact with their biotic and abiotic environment in Nature, in pure culture. Despite its difficulty, cultivation remains essential to analyze microbial morphology, physiology and genetics [1, 3]. There are even less microorganisms that have been isolated from the deep marine subsurface sediments, than from other natural environments, including extreme habitats. Most deep subsurface microorganisms detected so far were refractory to cultivation [4, 5], causing subseafloor microorganisms culturability in most studies to far less than 0.1% of all microscopically detected cells [6]. Molecular approaches primarily based on SSU rRNA/DNA gene sequences deposited in public databases have uncovered a vast diversity of unknown microorganisms from deep subseafloor ecosystems and are referenced therein as “uncultured” [7]. The isolation of uncultured microorganisms is one way to understand what the microorganisms corresponding to these sequences could potentially do. Unfortunately, most attempts to cultivate the main representatives of indigenous subseafloor prokaryotes failed. Despite fundamental ecological issues, the enrichment and subsequent isolation of indigenous deep subseafloor microorganisms still remains extremely limited for several technical and biological limitations/considerations.

84 | 4 Technological state of the art and challenges A great challenge for microbiology is the access to uncontaminated and undisturbed deep marine sediment/basalt samples. Samples from the subsurface biosphere are mostly collected from surface operations with riser (Chikyu) or riserless vessels (Joides Resolution), with drillers settled directly on the seafloor (MeBo, [8]) or via coring (󳶳 Fig. 4.1 (a), (b) and (c)). Such sampling requires specific coring or drilling facilities that are implemented in the framework of the international scientific drilling program IODP (Integrated Ocean Drilling Program) [9]. Unfortunately, all processes are sensitive to contamination by nonindigenous microbial populations originating from surrounding seawater or circulation mud fluids [10]. This is the reason why contamination tests are implemented in IODP. This low cultivation success of subseafloor microbes can also be explained by the fact that there have only been a few drilling expeditions fully dedicated to microbiological studies so far. In addition, cultivation approaches are mastered by only a limited number of research groups within the IODP scientific community. The under-representation of deep subsurface microorganisms in culture is also explained by the subseafloor environmental conditions that are still not fully constrained while it is often necessary to approach the conditions of natural habitats as closely as possible to grow their inhabitants. So many microbes remained reluctant to cultivation because we probably failed in recreating viable laboratory conditions. It is established that the marine subseafloor environment is characterized by extreme conditions. High temperature, high pressure, anoxia, recalcitrant organic matter and/or energy-limitation are common in this singular habitat, and are sometimes difficult to simulate in the laboratory. For example, the pressure (hydrostatic and lithostatic) is an important physical parameter in subsurface habitats. During the sampling process, potential damages and/or inactivation of cells due to depressurization or to changes of the physical – chemical conditions might occur. However, with a few exceptions [11], most enrichment cultures and isolation of microorganisms from subsurface sediments were performed at atmospheric pressure with depressurized samples, while microorganisms were subjected to hydrostatic plus lithostatic pressure in situ. So far, strict piezophilic prokaryotes adapted to high pressure, have not been isolated from deep subseafloor sediments. Still, there are obligate piezophilic microorganisms which cannot grow at atmospheric pressure and that have been isolated from other deep environments [12]. A few attempts of cultivating subsurface organisms under in situ high pressure were done using HYACINTH pressure-retaining drilling and core storage system [13] coupled to high pressure incubators [11]; this chain of equipment allows subsurface sediment handling without depressurization. While it is established that activity rates measurements carried out under in situ pressure are more accurate and close to reality than atmospheric pressure incubations [14], the hydrostatic pressure has been considered only in a few cases for enrichment and cultivation of subsurface microorganisms [15–17]. Deep biosphere microbial communities live at very low energy flux [18]. And the bioavailable energy source is one of the limiting factors for prokaryotic life in deep

4.1 Basic concepts and difficulties inherent to the cultivation of subseafloor prokaryotes

a

c

|

85

b

e

d

f

Fig. 4.1: Sampling and subsampling of subseafloor sediment core. (a) 9.5 m-long sediment core in plastic liner in the catwalk of the Joides Resolution during IODP Expedition 329 in South Pacific Gyre. (b) Drill rig MeBo from the Maria S. Merian (from Freudenthal and Wefer [8]). (c) Gravity coring from the Norwegian continental margin during the VICKING cruise on the Pourquoi Pas ? (e) and (f) Microbiological subsampling from sediments cores by using tip-cut syringes from South Pacific Gyre IODP expedition 329 and (d) from West African continental margin during GUINECO cruise on FS METEOR.

86 | 4 Technological state of the art and challenges biosphere. The combination of low energy and nutrient availability often leads to extremely low growth rates in situ. Time scales consecrated to cultivation might be inadequate to scientist’s careers productivity standards. Cultivation of deep biosphere microorganisms is also hampered by the quite low microbial biomass at the most profound depths in sediments [19, 20], by the presumably low activity in situ due to low nutrients and energy fluxes [21], and by the physiological state of cells. Subseafloor sediments harbor as much as endospores as vegetative cells [22]. The biomass turnover is obviously extremely slow (hundreds to thousands of years) and the estimated mean generation time far longer than in other ecosystems [22–24]. Nevertheless, a fraction of microbial cells maintain potential for metabolic activity [22, 25] and at least a fraction of subsurface microbes can be adapted to laboratory growth conditions [26–28]. It is not yet known if a reversion from a dormant to an active state is conceivable for all cells, including those from the sediment parts subjected to the most starved conditions, neither if all subsurface microbes could develop more quickly if provided with adequate nutrients in synthetic culture media [3, 29]. Our knowledge of the subseafloor metabolism is not sufficient either to design suitable growth conditions for microorganisms. The subsurface ecosystem is undoubtedly based on chemosynthesis, with a sedimentary part largely depending on heterotrophy [26, 30, 31] on the organic carbon buried from surface world, and a basaltic part depending at least locally on autotrophy [32] (󳶳 Fig. 4.2). Nevertheless, a range of carbon sources and electron donors could be used. Unusual energy-yielding reactions that require special culture methods or skills might take place in subseafloor sediments. This is notably the case of dehalorespiration [33]. So it is difficult to define the nature and the quantity of nutrients that should be provided in a culture medium. The quantity is particularly critical as every component of a culture medium can have inhibitory or toxic effects on certain strains and given that media with high nutrient concentrations do not reflect the natural habitat of a microorganism [1]. A sudden transfer from energy-limited natural sediment to synthetic medium with high nutrient concentrations can have an inhibitory effect on growth or cause a so-called substrate-accelerated death [3, 5]. Molecular inventories of the microbial diversity based on SSU-rRNA genes are not informative to design culture media. Indeed, many subseafloor lineages do not have cultured representatives [34, 35], and phylogeny does not necessarily translate into physiology within prokaryotic lineages with cultured members. This holds particularly true in the case of subsurface prokaryotes: indeed, there are several examples of isolates from deep sediments that exhibit higher metabolic versatility and are adapted to broader range of physical conditions than their closely-related surface counterparts [5, 15, 26, 36– 38] (󳶳 Table 4.1). So far, an immense majority of generalist heterotrophic bacteria (e.g. [11, 26, 27, 39]) capable of using a wide range of energy sources and terminal electron acceptors, were isolated from subseafloor sediments (󳶳 Table 4.1 and 󳶳 Table 4.2). By contrast, very few specialized sulfate-reducing Bacteria [15, 40] and methanogenic Archaea [28, 36, 38] were isolated (󳶳 Table 4.1 and 󳶳 Table 4.2). Curiously, when wide enrichment collections targeting different physiological groups such as fermenters,

Origin

Peru margin (ODP Leg 201)

Nankai Trough (ODP Leg 190)

Nankai Trough (ODP Exp. 190)

Mariana Forearc (ODP Leg 195)

Japan Sea (ODP site 798B)

Shimokita Peninsula of Japan (Site 9001)

Strain / Order, Class or Phylum

Thermosediminibacter oceani/ Firmicutes

Shewanella profunda/ Gammaproteobacteria

Marinilactibacillus piezotolerans/ Gammaproteobacteria

Marinobacter alkaliphilus/ Gammaproteobacteria

Desulfovibrio profundus/ Deltaproteobacteria

Geofilum rubicundum/ Bacteroidetes

247 / 1180

518 / 900

1 / 2942

4 / 4791

4 / 4791

2 / 252

10–36

15–65

10–50

4–50

4–37

52–76

Subseafloor Temperature range depth (°C) (mbsf) / water depth (m)

ND

0.1–40

ND

0.1–30

0.1–50

ND

Pressure range (MPa)

0.5–6

0.2–10

0–21

0–12

0–6

0–6

Salinity range (%)

Carbon sources

Cellobiose, fructose, galactose, lactose, mannose, rhamnose, trehalose, xylose, yeast extract

Lactate, pyruvate

Yeast extract, tryptone, casein, casamino acids, starch, crude oil, corn oil, olive oil, tween 20, tween 80, 𝑛-pentadecane, 𝑛-tridecane, 𝑛-undecane, 𝑛-decane, 𝑛-nonane, butane, kerosene, ethanol, glycerol, maltose, glucose, lactose, xylose, fructose, acetate, lactate, pyruvate, alanine, glutamate, glutamine, proline

Peptone, tryptone, cellobiose, fructose, galactose, lactose, maltose, mannitol, sorbitol, sucrose, xylose, glucose

Maltose, mannose, arabinose, sorbitol, succinate, glutamate, fumarate, citrate, lactate, pyruvate, yeast extract, peptone, tryptone

Casamino acids, beef extract, tryptone, cellobiose, fructose, galactose, glucose, maltose, mannose, raffinose, sucrose, trehalose, xylose, methanol, inositol, mannitol, sorbitol, lactate, pyruvate

Energy sources

Table 4.1: Physiological features of bacterial and archaeal isolates from subseafloor sediments that have been characterized.

O2 , NO−3 , NO−2

2− SO2− 4 , SO3 , − 3+ S2 O2− 3 ,NO3 , Fe , DMSO, lignosulfonate, reduced carbon source (fermentation)

O2 , NO−3 , fumarate,

O2 , reduced carbon source (fermentation)

O2 , NO−3 , Fe3+ , fumarate, Lcystine, reduced carbon source (fermentation)

S2 O2− 3 , S°, MnO2

Terminal electron acceptor(s)

[74]

[15]

[37]

[73]

[17]

[72]

Reference

4.1 Basic concepts and difficulties inherent to the cultivation of subseafloor prokaryotes |

87

Shimokita Peninsula of Japan (Site 9001)

Shimokita Peninsula of Japan (Site 9001)

Nankai Trough

Nankai Trough

Brevundimonas abyssalis/ Alphapropteobacteria

Sunxiuqinia faeciviva/ Bacteroidetes

Methanoculleus submarinus/ Methanomicrobiales

Methanococcus aeolicus/ Methanococcales

Legend: ND, not determined

Origin

Strain / Order, Class or Phylum

247 / 950

247 / 950

247 /1180

11 / 1180

Subseafloor depth (mbsf) / water depth (m)

15–55

10–55

4–37

2–41

Temperature range (°C)

ND

ND

0.1–25

ND

Pressure range (MPa)

0.3–6

0.6–9

0.5–6

1–4

Salinity range (%)

Carbon sources

H2 , formate

H2 , formate

CO2

Acetate, yeast extract

Yeast extract, tryptone, casein, casamino acids

Xylitol, acetate, 𝛽-hydroxybutyrate, 𝛼-ketoglutarate, lactate, propionate, dextrin, 𝜌-hydroxybutyrate, 𝛼-ketobutyrate, malonate

Energy sources

CO2

CO2

O2 , reduced carbon source (fermentation)

O2

Terminal electron acceptor(s)

[38]

[36]

[76]

[75]

Reference

88 | 4 Technological state of the art and challenges

4.1 Basic concepts and difficulties inherent to the cultivation of subseafloor prokaryotes

| 89

Table 4.2: Summary of previous key studies on prokaryotic cultivable diversity from subseafloor sediments. Origin of subseafloor sediments

Subseafloor depth / water depth

Main results

Reference

Mediterranean sapropels

2.71– 4.38 mbsf/ 1391 m, 0.05– 4.60 mbsf/ 2153 m, 0.05– 3.45 mbsf/ 2330 m

Quantitative analysis of bacterial communities with MPN based on artificial seawater supplemented with (1) amino acids, (2) alcohols, (3) aromatic compounds, (4) long chain fatty acids, (5) a mixture of various carbon compounds or(6) a sediment extract medium. Cultivation of Alpha-, Beta-, Gamma-, Deltaproteobacteria, Actinobacteria, Firmicutes. Half of the isolates are related to Photobacterium and Rhizobium.

[39]

Eastern Equatorial Pacific Ocean and Peru margin (ODP Leg 201)

1–420 mbsf/ 150–5100 m

162 bacterial isolates were obtained on anaerobic media prepared with various carbon sources and with different electron donors. They are related to the Alpha-, Gamma-, Deltaproteobacteria, Firmicutes, Actinobacteria and Bacteroidetes. Most strains belong to the genera Bacillus and Rhizobium. Most isolates are facultatively aerobic.

[26]

Shimokita Peninsula of Japan (Site 9001C)

0.5– 342.5 mbsf/ 1180 m

552 bacterial isolates, large majority of aerobic heterotrophs belonging to the genera Bacillus, Shewanella, Pseudoalteromonas, Halomonas, Pseudomonas, Paracoccus, Rhodococcus, Microbacterium or to the family Flexibacteraceae.

[27]

Indian Continental Shelf, Cascadia margin and Gulf of Mexico

77 mbsf/ 1049 m, 170.5– 171.5 mbsf/ 1315.4 m, 227.08 mbsf/ 1300 m.

Enrichment cultures and isolations performed with deep gas hydrate sediments (sampled without depressurization) in a variety of media under hydrostatic pressure (from 0.1 to 80 MPa). Predominant sequences in enrichments (Carnobacterium, Clostridium, Marinilactibacillus, Pseudomonas, Acetobacterium, Bacteroidetes) are largely independent of media and pressures. Isolation of a majority of facultative anaerobes.

[11]

Shimokita Peninsula of Japan (Site 9001)

0.42, 4.66, 18.34, 48.11 and 106.7 mbsf/ 1180 m

Enrichment cultures performed with a continuous-flow-type bioreactor with polyurethane sponges as microbial habitats and using artificial seawater supplemented with glucose, yeast extract, acetate and propionate: cultivation of Bacteria (notably JS1, Proteobacteria, Bacteroidetes, Planctomycetes and Firmicutes) and archaeal methanogens (Methanobacterium, Methanococcoides, Methanosarcina). Isolation of anaerobes including methanogens.

[28]

Sediments above a basaltic aquifer – Juan de Fuca Ridge IODP Exp. 301

1.3–9.1 mbsf, 240–262 mbsf/ 2656 m

Isolation of sulfate-reducing bacteria: Desulfosporosinus sp., Desulfotomaculum sp., Desulfovibrio sp., Desulfotignum sp. All reduce sulfate using short-chain 𝑛-alcohols , fatty acids or H2 as an electron donor and are also able to ferment ethanol, pyruvate or betaine.

[40]

MPN: Most Probable Number.

90 | 4 Technological state of the art and challenges

Fig. 4.2: Semihypothetical model of microbial metabolisms in subseafloor sediments. In the upper sediments of the seafloor, the carbon cycle is coupled to the reduction of a wide range of electron acceptors that diffuse downward from the water column. These electron acceptors are consumed according to a series of thermodynamically predictable metabolic reactions depending on the free 4+ 3+ 2− energy yielded by their reduction (O2 , NO− 3 , Mn , Fe , SO4 then CO2 ). In deeper sediments, there is no such thing as a universally applicable or absolute model. The energy can come (i) from the organic carbon buried from surface world, (ii) from geofuels formed from geochemical reactions in certain geological settings, (iii) from geofuels formed by thermogenic processes, (iv) from water radiolysis, or (iv) from strong oxidants that diffuse upward from deep aquifers or brines (adapted from DeLong [77] and Ciobanu [78]).

sulfate-reducers and methanogens were performed using subseafloor sediments from various origins, these often led to the isolation of the same few “generalist” Bacteria [11, 26, 27] (󳶳 Table 4.2). This suggests that we will have to renew our cultural methods to grow a fraction of the 99.9% of subseafloor microbial cells that are not yet cultured.

4.2 Microbial growth monitoring

| 91

4.2 Microbial growth monitoring, method detection limits and innovative cultivation methods In many natural environments and notably oligotrophic ones, microbial cells sometimes have a very small size [41], do not reach high density in liquid culture [42] and do not form large colonies on solid media despite long incubation periods [43]. As we face that sort of situation with subseafloor sediments (Ciobanu et al., in press), the methodology of growth observation and measurement is particularly critical. Indeed, the sensitivity of methods varies from one method to another and there are some basic techniques for monitoring growth that must probably be avoided in the framework of subseafloor cultivable diversity. For example, we should probably outlaw optical density measurements with a spectrophotometer and simple observations of liquid culture turbidity by the naked eye as these methods allow only the detection of ∼ 105 cells per milliliter [3]. In the same line of thinking, growth detection on plates or agar shakes should be preferentially performed with a stereomicroscope to not miss microcolonies. Optical microscopy is a more sensitive detection method as it allows the detection of ∼103 –105 cells per milliliter on a slide. Its sensitivity can be largely lowered (< 102 cells per milliliter) when cells are concentrated by filtration before observation. However, it is not always easy to have enough liquid culture for filtration when working with low density slow-growing strains. The microscopic detection of very small (0.3 to 0.8 μm) prokaryotic cells is enhanced when cells are stained (with SYBR® Green or PicoGreen® for example) before observation by fluorescence microscopy. Flow cytometry is a powerful and sensitive tool to enumerate cells on the basis of scattered light or fluorescence (labeled cells), and to measure bacterial growth from only a few hundreds microliters of culture. Analytical methods can also be used for the detection of metabolic end products (HPLC, LC-MS, GC, GC-MS) or the detection of molecules signing activity (ATP-metry [44, 45], etc.). Over the past decade, significant efforts have been made to develop innovative cultivation methodologies to grow recalcitrant microbes from natural environments. Schematically, novel cultivation strategies were based on (i) the refinement of standard cultivation strategies, with, for example, dilutions to extinction under low nutrient concentrations or long-term incubations [2, 46]–[50]; (ii) in situ culture or simulation of in situ conditions, with, for example, the use of natural seawater [43, 51]; (iii) strategies maintaining cell-to-cell communications, like community cultures or use of signal compounds [52]; or (iv) high-throughput cultivation methods or microtechnologies, with, for example, micro-Petri dishes or microencapsulation of cells in gel microdroplets for massively parallel microbial cultivation [53]–[55]. Nowadays, such methods are not applied to deep subseafloor sediments.

92 | 4 Technological state of the art and challenges

4.3 Challenges and research needs (instrumental, methodological and logistics needs): new approaches and future scope for deep subseafloor cultivation One of the main difficulties in working with deep subseafloor samples is linked to the contamination risk with nonindigenous cells. Thus, all steps from sediment sampling/ subsampling to slurry preparation and finally inoculation require contamination checks of the integrity of samples and of their preparation. The risk of contamination is more important with riser-vessels than with riser-less drilling operations. The potential for contamination is also exacerbated by the rotary core barrel (RCB) coring technology. Aboard ship, collection of deep subsurface samples involves specific tools such as drilling or coring systems which increase potential contamination of crude sediments [56, 57]. The development of standardized methods and protocols routinely applied during coring and drilling operations is essential for estimation of potential contamination and quality assurance and quality control of sediment samples. Prevention of contamination during sampling is critical to accurate downstream microbiological analyses. Potential drilling fluid contaminations should be routinely monitored by using quantitative, sensitive and time efficient methods. For instance, Perfluorocarbon tracers (PFT) supplied to the drilling fluid or to drilling mud tank is pumped through a drill pipe into boreholes during drilling operations and subsequently analyzed by an optimized protocol on board [57]. In addition, bacterial-size fluorescent beads can be used during drilling. Microsphere presence/absence within samples is determined by automated microscopic or flow cytometry methods [58]. As these contamination checks have their limits and/or cannot always be used, it is also important to collect drilling fluids/drilling muds and pelagic seawater, during drilling expeditions dedicated to microbiology, for comparison. Microbiological subsampling steps, which include conditioning and storage of noncontaminated sediments prior to cultivation, are also critical. Some choices of conditioning and storage can be made (anoxia or not, pressure or not, etc.) and will increase or limit potential damages to cells and impact on culture efficiency. Sediment handling on board is also challenging. Sterile plastic syringes are generally used to obtain aseptic subcores [59] (󳶳 Fig. 4.1 (d), (e), and (f)). IODP piston core sediments are sufficiently soft to fill the core liner and to be easily subcored. Sediments from deeper horizons may be harder and sterile syringes mini-cores could be hammered into the sediment. When sediments are hard enough (i.e. basalts) to require Rotary Core Barrel (RCB), the core itself is loose within the core liner and exposed to nonsterile seawater during retrieval. Intact whole-round basalt cores could be decontaminated, cracked aseptically, and subsampled to measure drilling fluid contaminations [32] before microbiological analysis. Specialized and standardized core handling systems have to be designed in order to maintain sterile and anaerobic conditions and to remove potentially contaminated outer sediment layers [60].

4.3 Challenges and research needs (instrumental, methodological and logistics needs) |

93

Development of in situ enrichment devices (incubators, colonizers, etc.) deployed and incubated in deep habitats are suitable to incorporate environmental factors which are difficult to reproduce in the laboratory. Such approaches are promising to grow previously uncultivated microorganisms from deep-subseafloor habitats. Future directions to cultivate microorganisms from subseafloor habitats comprise also in situ subseafloor experimentations using newly developed tools and technologies. CORKs (Circulation Obviation Retrofit Kits [61]) can notably be used. Such tools are designed to stop bottom water influx, thus in situ borehole conditions can develop after drilling. CORKs allow in situ growth experiments to investigate microbial colonization and mineral alterations [62]. Incubation chambers were already deployed inside boreholes drilled into the oceanic crust and were actively colonized by microorganisms during deployment [63]. Today, one of the main challenges for microbiologists studying the deep biosphere is to develop strategies and innovative technologies to cultivate as-yet-uncultured microorganisms. Nowadays, in most cases, microbiologists isolate microorganisms after enrichment in liquid media or by directly plating on solidified media. Such approaches are probably not suitable for the slow-growing subseafloor microbes. Microbiology only recognizes pure (axenic) culture for characterization. However in Nature, microorganisms are engaged in complex interaction with their close or distant surrounding microenvironments and their growth depends on other organisms of the community as well. There are many examples of syntrophic relationships between microorganisms that require more than one species for their growth [64, 65], and such associations might be important in subseafloor habitats where energetic fluxes are low and where sharing could be a life strategy. Microorganisms in communities communicate with each other by producing and secreting signaling molecules and cell-to-cell communication appears essential for the growth of single cells within the community in aggregates or biofilms [66]. A large number of approaches have been tested to culture as-yet-uncultured microorganisms from natural habitats by recreating interactions of cells with other relevant species in co-culture [43] but such strategies were never applied to subseafloor sediments. The range of substrates used in the enrichment of deep subseafloor representatives should be expanded by, e.g. including insoluble or a priori refractory compounds or macromolecules, since virtually nothing is known about the microorganisms that are involved in breaking down of complex organic matter. Most standard microbiological techniques are unusable with subseafloor microorganisms. There is a strong need to develop faster and more automatable culture methods. Microscale techniques should be developed and analytical techniques adapted as well. Recently, technological advances have been made to increase culture efficiency. And of the micromanipulation approaches available for position of single cells from a mixture or confinement strategies of small microbial populations [67–69], optical tweezers [70], electrokinetic methods or combined optoelectronic approaches are some of the most useful possibilities. A common method to isolate any individual cell is to mechanically separate the cells by physical boundaries. Clean separation of cells from deep subsurface sediment

94 | 4 Technological state of the art and challenges matrix has the potential to overcome interferences with sediment material and increase the detection limit for total cell counting [20]. Prior to enrichments, viable cell extraction from sediment particles that minimize cell damages might increase culture efficiency. Micromanipulation of individual cell in microwell devices (microdish culture chip) allows the development of microcolonies that can then be recovered for further cultivation by using a micromanipulator. Culture in microcolonies is well adapted to detect slow-growing microorganisms compared to conventional methods. Single-cell isolation from the natural sample combined to Omics is also a promising approach. Omics information could assist in the process of culture media design and metabolism prediction. Then metabolism of uncultured microbes could be inferred from their genomes. Single-cell genomic sequencing is based on Multiple Displacement Amplification (MDA) of DNA from individual cells and does not require any cultivation [71]. Recently, four single cells of uncultured Archaea and Bacteria affiliated to the Miscellaneous Crenarcheotic Group (MCG), to the Marine Benthic Group-D (MBG-D) and to the Chloroflexi (Wasmund et al., 2013), which are abundant in the marine subseafloor, were extracted from sediment and submitted to single-cell genomics. Partial genomes reconstructions showed that the two first lineages (MCG and MBG-D) may have an important role in protein degradation in marine sediments [31]. Some innovative high-throughput culture and isolation techniques that proved to be efficient to grow as-yet-uncultured microorganisms are based on the incorporation of microbial cells within gel microdroplets, followed by an incubation period in a column and a sorting into microplates [53]. The cell-to-cell communications may be maintained by using a flow-through culture in parallel microbioreactors nourished by community culture medium and metabolite products [1]. The innovative aspects of such a development include a (i) culture method coupling and connecting a bioreactor containing a community structure to single cells from this community isolated in gel microdroplets (GMD) [53] which are placed in a multiwell microbioreactor (one GMD per well); (ii) the microbioreactor allowing a complete control of the microenvironment; (iii) the high-throughput handling of several thousand wells containing clonal populations in parallel. Since the early days of microbiology, huge numbers of microbial strains have been isolated from a great variety of natural sources and used for scientific research and in the biotechnological industry. However, large numbers of microbial cultures were lost in the past and are no longer available. This is probably the case for several microbial strains isolated from deep subseafloor sediments. Effective research needs adequate and reliable sources of properly preserved cultures to be supplied promptly on demand for detailed studies on their physiological capabilities and for linking this information to ecological questions [26]. Therefore, subseafloor isolates should be systematically placed in a publicly accessible culture collection. In addition, as already suggested by other colleagues [37], we recommend that an IODP deep subseafloor culture collection be established, for deposits of microorganisms including those requiring specific maintenance such as strict piezophiles.

References |

95

Current cultivation methods are stretched to their limits. They fail to detect subseafloor microbial growth due to low growth rates and/or low cell densities notably. The development and the assessment of very sensitive methodologies for growth detection are urgently required in deep biosphere microbiology. If more efforts were put into cultivation and if more sensitive growth detection methods were routinely used, it is likely that more microorganisms would be amenable to culture.

References [1] [2]

[3] [4] [5] [6] [7] [8] [9] [10]

[11]

[12] [13] [14]

[15]

[16]

Alain K, Querellou J. Cultivating the uncultured: limits, advances and future challenges. Extremophiles 13 (2009), 583–594. Connon SA, Giovannoni SJ. High-throughput methods for culturing microorganisms in verylow-nutrient media yield diverse new marine isolates. Appl Environ Microbiol 68 (2002), 3878– 3885. Zengler K. Central Role of the Cell in Microbial Ecology. Microbiol Mol Biol Rev 73 (2009), 712– 729. Jørgensen BB, Boetius A. Feast and famine - microbial life in the deep-sea bed. Nature Rev Microbiol 5 (2007), 770–781. Sass H, Parkes RJ, eds. Subseafloor sediments: an extreme but globally significant prokaryotic habitat (taxonomy, diversity, ecology). Springer ed; 2011. D’Hondt S, Jørgensen BB, Miller DJ, et al. Distributions of microbial activities in deep subseafloor sediments. Science 306 (2004), 2216–2221. Rappe MS, Giovannoni SJ. The uncultured microbial majority. Ann Rev Microbiol 57 (2003), 369–394. Freudenthal T, Wefer G, Ieee. Shallow Drilling in the Deep Sea: The Sea Floor Drill Rig MEBO. In: Oceans 2009 - Europe, Vols 1 and 2. New York, 180–183, 2009. platforms and vessels. (Accessed May 6, 2013, 2012, at http://www.iodp.org/ships/ platforms.) Masui N, Morono Y, Inagaki F. Microbiological assessment of circulation mud fluids during the first operation of riser drilling by the deep-earth research vessel Chikyu. Geomicrobiol J 25 (2008), 274–282. Parkes RJ, Sellek G, Webster G, et al. Culturable prokaryotic diversity of deep, gas hydrate sediments: first use of a continuous high-pressure, anaerobic, enrichment and isolation system for subseafloor sediments (DeepIsoBUG). Environ Microbiol 11 (2009), 3140–3153. Zeng X, Birrien JL, Fouquet Y, et al. Pyrococcus CH1, an obligate piezophilic hyperthermophile: extending the upper pressure-temperature limits for life. ISME J 3 (2009), 873–876. Schultheiss PJ, Francis TJG, Holland M, et al. Pressure coring, logging and subsampling with the HYACINTH system. Rothwell, RG ed; 2006. Bowles MW, Samarkin VA, Joye SB. Improved measurement of microbial activity in deep-sea sediments at in situ pressure and methane concentration. Limnol Oceanogr Meth 9 (2011), 499–506. Bale SJ, Goodman K, Rochelle PA, et al. Desulfovibrio profundus sp nov, a novel barophilic sulfate-reducing bacterium from deep sediment layers in the Japan Sea. Int J Syst Bacteriol 47 (1997), 515–521. Nilsen RK, Torsvik T. Methanococcus thermolithotrophicus isolated from North Sea oil field reservoir water. Appl Environ Microbiol 62 (1996), 728–731.

96 | 4 Technological state of the art and challenges [17] Toffin L, Bidault A, Pignet P, et al. Shewanella profunda sp nov., isolated from deep marine sediment of the Nankai Trough. Int J Syst Evol Microbiol 54 (2004), 1943–1949. [18] Hoehler TM, Jørgensen BB. Microbial life under extreme energy limitation. Nat Rev Microbiol 11 (2013), 83–94. [19] Parkes RJ, Cragg BA, Bale SJ, et al. Deep Bacterial Biosphere in Pacific-Ocean Sediments. Nature 371 (1994), 410–413. [20] Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16,213– 16 216. [21] D’Hondt S, Rutherford S, Spivack AJ. Metabolic activity of subsurface life in deep-sea sediments. Science 295 (2002), 2067–2070. [22] Lomstein BA, Langerhuus AT, D’Hondt S, Jørgensen BB, Spivack AJ. Endospore abundance, microbial growth and necromass turnover in deep subseafloor sediment. Nature 484 (2012), 101–104. [23] Jørgensen BB, D’Hondt S. A starving majority deep beneath the seafloor. Science 314 (2006), 932–934. [24] Jørgensen BB. Shrinking majority of the deep biosphere. Proc Natl Acad Sci USA 109 (2012), 15 976–15 977. [25] Morono Y, Kallmeyer J, Inagaki F, and the expedition 329 Scientists. Preliminary experiment for cell count using flow cytometry 2011. In D’Hondt S, Inagaki F, Alvarez Zarikian CA, and the expedition 329 Scientists, Proc IODP, 329: Tokyo (Integrated Ocean Drilling Program Management International, Inc.) doi: 10.2204/iodp.proc.329.110.2011 [26] Batzke A, Engelen B, Sass H, Cypionka H. Phylogenetic and physiological diversity of cultured deep-biosphere bacteria from equatorial Pacific Ocean and Peru Margin sediments. Geomicrobiol J 24 (2007), 261–273. [27] Kobayashi T, Koide O, Mori K, et al. Phylogenetic and enzymatic diversity of deep subseafloor aerobic microorganisms in organics- and methane-rich sediments off Shimokita Peninsula. Extremophiles 12 (2008), 519–527. [28] Imachi H, Aoi K, Tasumi E, et al. Cultivation of methanogenic community from subseafloor sediments using a continuous-flow bioreactor. ISME J 5 (2011), 1913–1925. [29] Colwell FS, D’Hondt S. Nature and Extent of the Deep Biosphere. Rev Mineral Geochem 75 (2013), 547–574. [30] Biddle JF, Lipp JS, Lever MA, et al. Heterotrophic Archaea dominate sedimentary subsurface ecosystems off Peru. Proc Natl Acad Sci USA 103 (2006), 3846–3851. [31] Lloyd KG, Schreiber L, Petersen DG, et al. Predominant archaea in marine sediments degrade detrital proteins. Nature 496 (2013), 215–218. [32] Lever MA, Rouxel O, Alt JC, et al. Evidence for Microbial Carbon and Sulfur Cycling in Deeply Buried Ridge Flank Basalt. Science 339 (2013), 1305–1308. [33] Futagami T, Morono Y, Terada T, Kaksonen AH, Inagaki F. Dehalogenation Activities and Distribution of Reductive Dehalogenase Homologous Genes in Marine Subsurface Sediments. Appl Environ Microbiol 75 (2009), 6905–6909. [34] Inagaki F, Nunoura T, Nakagawa S, et al. Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments on the Pacific Ocean Margin. Proc Natl Acad Sci USA 103 (2006), 2815–2820. [35] Teske A, Sørensen KB. Uncultured archaea in deep marine subsurface sediments: have we caught them all? ISME J 2 (2008), 3–18. [36] Mikucki JA, Liu YT, Delwiche M, Colwell FS, Boone DR. Isolation of a methanogen from deep marine sediments that contain methane hydrates and description of Methanoculleus submarinus sp nov. Appl Environ Microbiol 69 (2003), 3311–3316.

References | 97

[37] Takai K, Moyer CL, Miyazaki M, et al. Marinobacter alkaliphilus sp nov., a novel alkaliphilic bacterium isolated from subseafloor alkaline serpentine mud from Ocean Drilling Program Site 1200 at South Chamorro Seamount, Mariana Forearc. Extremophiles 9 (2005), 17–27. [38] Kendall MM, Liu Y, Sieprawska-Lupa M, Stetter KO, Whitman WB, Boone DR. Methanococcus aeolicus sp. nov., a mesophilic, methanogenic archaeon from shallow and deep marine sediments. Int J Syst Evol Microbiol 56 (2006), 1525–1529. [39] Süss J, Engelen B, Cypionka H, Sass H. Quantitative analysis of bacterial communities from Mediterranean sapropels based on cultivation-dependent methods. FEMS Microbiol Ecol 51 (2004), 109–121. [40] Fichtel K, Mathes F, Könneke M, Cypionka H, Engelen B. Isolation of sulfate-reducing bacteria from sediments above the deep-subseafloor aquifer. Front Microbiol 3 (2012), 65. [41] Simu K, Hagstrom A. Oligotrophic bacterioplankton with a novel single-cell life strategy. Appl Environ Microbiol 70 (2004), 2445–2451. [42] Rappe MS, Connon SA, Vergin KL, Giovannoni SJ. Cultivation of the ubiquitous SAR11 marine bacterioplankton clade. Nature 418 (2002), 630–633. [43] Ferrari BC, Binnerup SJ, Gillings M. Microcolony cultivation on a soil substrate membrane system selects for previously uncultured soil bacteria. Appl Environ Microbiol 71 (2005), 8714– 8720. [44] Pridmore RD, Cooper AB, Hewitt JE. ATP as a biomass indicator in 8 North Island Lakes, New-Zealand. Freshw Biol 14 (1984), 73–78. [45] Eydal HSC, Pedersen K. Use of an ATP assay to determine viable microbial biomass in Fennoscandian Shield groundwater from depths of 3–1000 m. J Microbiol Meth 70 (2007), 363–373. [46] Cho JC, Giovannoni SJ. Cultivation and growth characteristics of a diverse group of oligotrophic marine Gammaproteobacteria. Appl Environ Microbiol 70 (2004), 432–440. [47] Button DK, Schut F, Quang P, Martin R, Robertson BR. Variability and isolation of marine-Bacteria by dilution culture - theory, procedures and initial results. Appl Environ Microbiol 59 (1993), 881–891. [48] Stingl U, Tripp HJ, Giovannoni SJ. Improvements of high-throughput culturing yielded novel SAR11 strains and other abundant marine bacteria from the Oregon coast and the Bermuda Atlantic Time Series study site. ISME J 1 (2007), 361–371. [49] Stevenson BS, Eichorst SA, Wertz JT, Schmidt TM, Breznak JA. New strategies for cultivation and detection of previously uncultured microbes. Appl Environ Microbiol 70 (2004), 4748– 4755. [50] Davis KER, Joseph SJ, Janssen PH. Effects of growth medium, inoculum size and incubation time on culturability and isolation of soil bacteria. Appl Environ Microbiol 71 (2005), 826–834. [51] Kaeberlein T, Lewis K, Epstein SS. Isolating “uncultivable” microorganisms in pure culture in a simulated natural environment. Science 296 (2002), 1127–1129. [52] Bruns A, Cypionka H, Overmann J. Cyclic AMP and acyl homoserine lactones increase the cultivation efficiency of heterotrophic bacteria from the central Baltic Sea. Appl Environ Microbiol 68 (2002), 3978–3987. [53] Zengler K, Toledo G, Rappe M, et al. Cultivating the uncultured. Proc Natl Acad Sci USA 99 (2002), 15 681–15 686. [54] Zengler K, Walcher M, Clark G, et al. High-throughput cultivation of microorganisms using microcapsules. In: Leadbetter JR, ed. Environmental Microbiology. San Diego: Elsevier Academic Press Inc; Vol. 124, 2005. [55] Ingham CJ, Sprenkels A, Bomer J, et al. The micro-Petri dish, a million-well growth chip for the culture and high-throughput screening of microorganisms. Proc Natl Acad Sci USA 104 (2007), 18 217–18 222.

98 | 4 Technological state of the art and challenges [56] Smith DC, Spivack AJ, Fisk MR, Haveman SA, Staudigel H, Ocean Drilling Program Leg S. Tracerbased estimates of drilling-induced microbial contamination of deep sea crust. Geomicrobiol J 17 (2000), 207–219. [57] Lever MA, Alperin M, Engelen B, et al. Trends in basalt and sediment core contamination during IODP Expedition 301. Geomicrobiol J 23 (2006), 517–530. [58] Smith DC, Spivack AJ, Fisk MR, Haveman SA, Staudigel H, Party LSS. Methods for quantifying potential contamination during deep ocean coring. ODP Technical Note 28 (2000), 1–19. [59] Parkes RJ, Cragg BA, Bale SJ, Goodman K, Fry JC. A Combined Ecological and Physiological Approach to Studying Sulfate Reduction within Deep Marine Sediment Layers. J Microbiol Meth 23 (1995), 235–249. [60] Wellsbury P, Mather I, Parkes R. Subsampling RCB cores from the western Woodlark Basin (ODP Leg 180) for microbiology. Proceedings of the Ocean Drilling Program, Scientific Results 180 (2001), 1–12. [61] Davis EE, Baecker K, Pettigrew T, Carson B, MacDonald R. CORK: a hydrologic seal and downhole observatory for deep-ocean boreholes. Proc ODP Initial Rep 139 (1992), 43–53. [62] Fisher AT, Wheat CG, Becker K, Davis EE, Jannasch H, Schroeder D, Dixon R, Pettigrew TL, Meldrum R, MacDonald R, Nielsen M, Fisk M, Cowen J, Bach W, Edwards K, Scientific and technical design and deployment of long-trem subseafloor observatories for hydrogeologic and related experiments, IODP expedition 301, eastern flank of Juan de Fuca Ridge. In: Fisher AT, Urabe T, Klaus A, and the Expedition 301 Scientists, Proc IODP 301, doi: 10.2204/iodp.proc.301.103.2005. [63] Orcutt B, Wheat CG, Edwards KJ. Subseafloor Ocean Crust Microbial Observatories: Development of FLOCS (FLow-through Osmo Colonization System) and Evaluation of Borehole Construction Materials. Geomicrobiol J 27 (2010), 143–157. [64] Milucka J, Ferdelman TG, Polerecky L, et al. Zero-valent sulfur is a key intermediate in marine methane oxidation. Nature 491 (2012), 541. [65] Kim YJ, Lee HS, Kim ES, et al. Formate-driven growth coupled with H2 production. Nature 467 (2010), 352–U137. [66] Camilli A, Bassler BL. Bacterial Small-Molecule Signaling Pathways. Science 311 (2006), 1113– 1116. [67] Wessel AK, Hmelo L, Parsek MR, Whiteley M. Going local: technologies for exploring bacterial microenvironments. Nat Rev Microbiol 11 (2013), 337–348. [68] Connell JL, Wessel AK, Parsek MR, Ellington AD, Whiteley M, Shear JB. Probing Prokaryotic Social Behaviors with Bacterial “Lobster Traps”. mBio 1 (2010), 8. [69] Carnes EC, Lopez DM, Donegan NP, Cheung A, Gresham H, Timmins GS, Brinker CJ. Confinement-induced quorum sensing of individual Staphylococcus aureus bacteria. Nat Chem Biol 6 (2010), 41–45. [70] Ashkin A, Dziedzic JM. Optical trapping and manipulation of viruses and Bacteria. Science 235 (1987), 1517–1520. [71] Lasken RS. Single-cell genomic sequencing using Multiple Displacement Amplification. Current Opinion Microbiol 10 (2007), 510–516. [72] Lee YJ, Wagner ID, Brice ME, et al. Thermosediminibacter oceani gen. nov., sp nov and Thermosediminibacter litoriperuensis sp nov., new anaerobic thermophilic bacteria isolated from Peru Margin. Extremophiles 9 (2005), 375–383. [73] Toffin L, Zink K, Kato C, et al. Marinilactibacillus piezotolerans sp nov., a novel marine lactic acid bacterium isolated from deep subseafloor sediment of the Nankai Trough. Int J Syst Evol Microbiol 55 (2005), 345–351. [74] Miyazaki M, Koide O, Kobayashi T, et al. Geofilum rubicundum gen. nov., sp nov., isolated from deep subseafloor sediment. Int J Syst Evol Microbiol 62 (2012), 1075–1080.

References | 99

[75] Tsubouchi T, Shimane Y, Usui K, et al. Brevundimonas abyssalis sp. nov., a dimorphic prosthecate bacterium isolated from deep subseafloor sediment in Japan. Int J Syst Evol Microbiol 63 (2013), 1987–1994. [76] Takai K, Abe M, Miyazaki M, et al. Sunxiuqinia faeciviva sp. nov., a novel facultatively anaerobic, organoheterotrophic bacterium within the Bacteroidetes isolated from deep subseafloor sediment offshore Shimokita, Japan. Int J Syst Evol Microbiol 63 (2013), 1602–1609. [77] DeLong EF. Microbial life breathes deep. Science 306 (2004), 2198–2200. [78] Ciobanu MC. Biosphère de subsurface des marges continentales: diversité, étendue et lien avec le paléoenvironnement. Plouzané: Université de Bretagne Occidentale; 2012. [79] Ciobanu MC, Burgaud G, Dufresne A, Breuker A, Rédou V, Ben Maamar S, Gaboyer F, Vandenabeele-Trambouze O, Lipp JS, Schippers A, Vandenkoornhuyse P, Barbier G, Jebbar M, Godfroy A, Alain K. Microorganisms persist at record depths in the subseafloor of the Canterbury Basin. ISME J (2014, in press). [80] Wasmund K, Schreiber L, Lloyd KG, Petersen DG, Schramm A, Stepanauskas R, Jørgensen BB, Adrian L. Genome sequencing of a single cell of the widely distributed marine subsurface Dehalococcoidia, phylum Chloroflexi. ISME J 8 (2014), 383–397.

Yuki Morono, Motoo Ito, and Fumio Inagaki

5 Detecting slow metabolism in the subseafloor: analysis of single cells using NanoSIMS 5.1 Introduction Numerous microbial ecology studies have demonstrated that the dynamics of microbial activity in any ecosystem depend largely on the availability of energy and nutrient substrates (reviewed by Morita, 1997 [1]). As already reviewed in previous chapters, microbial life has been found in many natural environments, including deep and ancient subseafloor sediments down to 1627 meters below the seafloor (mbsf) [2]. Estimates of naturally occurring microbial populations have suggested that a significant fraction of the total living biomass on earth is present in the subseafloor sedimentary biosphere [3–6]. Chemical profiling of porewater collected from sediment drill cores has suggested that the metabolic activity of subseafloor life is generally extremely low [7– 9] due to severe limitations in the transport of available electron donors and acceptors for cell respiration and growth [10]. Radioactive tracer incubation studies have consistently shown that the theoretical mean generation time of subseafloor microbial cells is extremely low, ranging from a few years to thousands of years [10, 11]. The net activity of the subseafloor sedimentary microbial ecosystem plays an ecologically important role in biogeochemical cycles, such as through degradation of buried organic matter and methane production. Other studies of subseafloor life in sediments of the continental margins have demonstrated that microbial biomass is positively correlated with the concentration of organic matter in the sediment [5, 12]. This indicates that sedimentary microbial ecosystems on most continental margins consist mainly of organic-fueled heterotrophs, with relatively minor autotrophic components (e.g. methanogens) [13]. The composition of carbon isotopes (𝛿13 C) in both extracted intact polar lipids and cellular bodies determined by secondary ion mass spectrometry (SIMS) analysis of core sediments collected off the coast of Peru, revealed constant values similar to the 𝛿13 C of buried organic matter, even in the sulfate–methane transition zones, where 13 C-depleted biogenic methane is expected to mediate microbial carbon metabolism via anaerobic oxidation of methane [14]. Molecular ecological studies based on PCRamplified 16S rRNA (and its corresponding DNA) point to the presence of diverse Archaea and Bacteria in organic-rich subseafloor sediments; most of the associated sequences are phylogenetically distinct from physiologically known isolates [15–17], and the compositions of the communities are stratified with depth and redox-interfaces such as sulfate–methane transition zones [18, 19]. Metagenomic analyses of Peru margin sediment revealed that over 80% of the genetic assemblages were function-

102 | 5 NanoSIMS analysis on subseafloor microbes ally unknown [20], whereas analyses of various continental margin samples identified some key functional genes and activities related to biogeochemical carbon cycles [16, 21–23]. Despite the widespread occurrence of significant microbial populations and communities in the subseafloor sedimentary environment, the physiological nature and nutrient-energy requirements for the long-term maintenance of essential life functions in this environment have seldom been examined [24, 25]. For example, whether the low availability of nutrients and energy substrates constrains the physiological status and growth potential of deeply buried microbial cells, is essentially unknown. More generally, to differentiate whether these subseafloor microbes are alive, growing, simply surviving, dormant, or dead fossils are important. To address these fundamental questions regarding the ecological physiology of subseafloor sedimentary microbial life, a highly sensitive technique known as nanometer-scale secondary ion mass spectrometry (NanoSIMS) was developed in order to facilitate detection of carbon and nitrogen assimilation by subseafloor microbial communities. In this chapter, we review the basis of the NanoSIMS technique, its utility for single-cell analysis of substrate incorporation and for matching identity and function, as well as the technical basics for obtaining high-quality results.

5.2 Overview of ion imaging with a NanoSIMS ion microprobe Secondary ion mass spectrometry (SIMS) is a technique used to analyze the chemical compositions of a solid sample by sputtering the surface of the sample. In SIMS, the surface of the sample is bombarded under high vacuum with a focused primary ion beam (e.g. Cs+ , O+2 or O− ). The collision cascade on the sample surface results in the ejection and ionization of atoms and molecules (secondary ions) from the surface layers of the sample. These secondary ions are accelerated into a mass spectrometer where they are separated according to their energy and mass/charge ratios before being detected. The elemental or isotopic compositions of the sample are determined by the mass/charge ratios of these secondary ions. SIMS is generally considered to be a qualitative analysis because of the large variation of ionization probabilities among different elements and materials. However, quantitative analysis is possible with the use of proper standard materials (e.g. known elemental abundances, known isotopic ratios). SIMS is the most sensitive analysis technique with detection limits ranging from the ppm to ppb level depending on elements. Ion imaging technique with SIMS is a powerful tool to visualize the distributions of isotopes and/or elements in samples, and is becoming common in a variety of science fields including material science [26], cosmochemistry [27] and biology [28]. Castaing & Slodzian [29] initially developed an ion imaging by SIMS in the early 1960s. There are two different modes, ion microscopy and ion microprobe, for acquiring images in SIMS. An ion microscopy mode preserves the lateral position of the sputtered

5.2 Overview of ion imaging with a NanoSIMS ion microprobe |

103

Fig. 5.1: NanoSIMS 50L (AMETEK) installed in Kochi Institute for Core Sample Research, JAMSTEC.

secondary ions from the sample surface to the detector, and then visualizes the distributions of the detected secondary ions. AMETEK CAMECA IMS-1270/1280 equipped with a two-dimensional “stacked CMOS-type active pixel sensor for charged particles” (SCAPS) detector developed by Prof. Yurimoto (Hokkaido University) uses an ion microscopy mode for image acquisition [30]. An ion microprobe mode is operated by rastering a fine focused primary ion beam across the sample surface and by recording the intensities of the secondary ions for each beam position. AMETEK CAMECA NanoSIMS ion microprobe uses an ion microprobe mode for image acquisition (󳶳 Fig. 5.1). Alternatively, ToF (time-of-flight) SIMS can generate ion images of materials, particularly surfaces providing detailed chemical structure information (i.e. elements, functional groups, polymer constituents, molecules). However, due to the different mass analyzer system (time-of-flight type) and short-pulsed low-dosed ion beams that produced by a liquid metal primary ion source (e.g. Ga+ or Bi3+ ), ToF SIMS instrument shows relatively lower lateral resolution than in NanoSIMS, and much longer acquisition times by a pulsed beam [31]. Hereafter, we focus on a NanoSIMS ion microprobe. The CAMECA NanoSIMS 50/50L ion microprobe (󳶳 Fig. 5.1) represents the in-situ microanalysis by SIMS that was designed with the goals of increasing the information that can be obtained from the smallest samples. It offers a fine focusing of a primary ion beam with a spatial resolution of less than 50 nm that is achieved using the Cs+ as a primary ion beam, whereas the O− primary ion beam is focused about 150 nm with ultrahigh sensitivity (< 200 atom detection limit). The NanoSIMS couples a high transmission and high mass resolution mass spectrometer with an array of detectors, enabling simultaneous detection of seven elements or isotopes originating from the same sputtered volume of a sample and maximizing the scientific information that can be obtained from the smallest samples (󳶳 Fig. 5.2). The primary ion beam can be

104 | 5 NanoSIMS analysis on subseafloor microbes

Ion source

Sample

Prima ry Io Beam n

Magnetic Sector Mass Analyzer Secondary Ions

Electron multipliers

Mass 12 Mass 13

Mass 26 Mass 35

12

C-

13

C-

12

C14N-

35

S-

Fig. 5.2: Schematic diagram of NanoSIMS 50L ion microprobe.

scanned across the sample to produce quantitative secondary ion images. Up to seven elemental and/or isotopic images can be acquired simultaneously by seven electron multipliers with sensitivity in the ppm. The capability for images of multiple elements and isotopes within a sample with per-mil precision and accuracy and nm scale spatial resolution is unique to the NanoSIMS. Because isotopic images can be acquired with extremely low primary beam currents (∼ 1 pA), coordinated studies of morphological, structural, chemical and isotopic characteristics of materials at submicrometer scales by NanoSIMS, transmission electron microscopy (TEM), scanning electron microscopy (SEM), and focused ion beam (FIB) systems are becoming routine [32, 33] Owing to these unique capabilities, the NanoSIMS has had a major impact in the field of cosmochemistry, leading to the discoveries of ancient silicate stardust and interstellar organic grains in meteorites and interplanetary dust [27, 34–36]. In both of these examples, the target materials are extremely small (0.2–1 μm), but have isotopic ratios in many elements that significantly differ (20–10,000%) from materials formed in the Solar System. The NanoSIMS is particularly well suited to measuring large isotopic variations at small scales, and accordingly its use in cosmochemistry has been focused on the study of interstellar materials. More recently, the NanoSIMS has been applied to high precision isotopic measurements that are required for many primitive Solar System materials (i.e. refractory inclusions in carbonaceous chondrites) that exhibit moderate isotopic variations (0.1–5%) [37]. In the last decade, SIMS technique has been used in microbiology to match the chemotaxonomic and phylogenetic signature of microbes [38]. Recently, NanoSIMS

5.3 Detecting slow metabolism: bulk to single cells

|

105

ion imaging technique was used in a stable isotope probing study (i.e. 13 C, 15 N labeling) for a single cell to understand microbial metabolic activities, and combination with in-situ hybridization for phylogenetic identification [39–44]. Applications for microbes residing in deep subseafloor biosphere using NanoSIMS ion imaging is discussed in detail in this chapter.

5.3 Detecting slow metabolism: bulk to single cells Because of the extreme depth, no light is available for photosynthetic production along most of the seabed [10]. As a result, can the large number of microbial cells that have been discovered in the subseafloor biosphere be considered “living creatures”? Several lines of evidence have suggested that subseafloor microbes are indeed alive [5, 11, 42]; however, because subseafloor microbial communities consist of species distantly related to cultured organisms from the surface biosphere [16], obtaining culturable representatives is highly challenging. In addition, the life cycle, in situ activity over geological time scales, and the means by which microbes acquire energy in the deep subseafloor have yet to be explored. Microbial life forms found beneath the seafloor primarily rely on organic matter produced by surface phototrophic activity for their nutrients [5]; however, subseafloor sediments contain 10 to 10,000-fold more cells per unit volume than do productive ocean-surface waters [10]. What are the reasons for such a significant difference in microbial colonization of these habitats? Of course, the subseafloor biosphere differs from the water column in a number of ways, but particularly in terms of the spatial arrangement, as the distribution of pore spaces within solids is advantageous for microorganisms because it reduces grazing pressure. On the other hand, considering the microbial cell abundance and relatively low amount of organic matter supplied to the seafloor (on average, 1 g of organic carbon per m2 per year [8, 10]), microbial activity would be expected to be extremely low even at the very surface of the subseafloor environment.

5.3.1 Bulk measurement of subseafloor microbial activity using radiotracers Radioisotopes enable high sensitivity detection of microbial metabolic activity in bulk sediment. Several studies have reported measurable sulfate transformation activity in the 10−19 to 10−16 mol cell−1 day−1 range that is associated with subseafloor microbial cells. This level of activity is equivalent to 10−24 to 10−21 mol cell−1 s−1 [45, 46]. Considering the number of ions in a molar unit of sulfate (6 × 1023 per mole), this level of sulfate reduction activity corresponds to roughly 1 sulfate ion per cell per second, which is far below the energy requirement for even a single revolution of an E. coli flagellum [46]. Even though the measured metabolic activity of subseafloor microbial cells is so low, there is still a potential that the level of activity has been overestimated

106 | 5 NanoSIMS analysis on subseafloor microbes in laboratory incubation experiments. D’Hondt et al. estimated the bulk metabolic activity in subseafloor sediment from geochemical profiles obtained from drilled core samples. The per-cell sulfate reducing activity in subseafloor sediment estimated from −22 the SO2− to 9.0×10−18 mol cell−1 d−1 for 4 ion flux in the cores was in the range 3.8×10 −23 −22 −1 −1 the ocean-margin and 4.1 × 10 to 3.2 × 10 mol cell d for open-ocean sites [8]. Apparently, determining the cell-specific metabolic rate using bulk microbial communities can introduce a high degree of scatter in the data, depending on the approach used and the sample being analyzed. Such variability has been attributed to uncertainty regarding the assumed number, or fraction, of microbial cells that are engaged in the reaction and to differences in the physiological state of microbes under differing experimental conditions [46].

5.3.2 Observing radioactive substrate incorporation at the cellular level: microautoradiography Observing substrate metabolism at the single-cell level has long been a desire of researchers in the field of microbial ecology. Visualization of substrate uptake at the cellular level was first demonstrated by Brock and Brock [47] about 50 years ago. They found it difficult to assess cellular activity by bulk activity measurements with radioisotopes. To solve this problem, they developed an autoradiography technique known as microautoradiography (MAR), which permits visualization of microbial cells that have incorporated a radioisotope-labeled substrate. Autoradiography enables visualization of the localization of a radioactive substance as a result of the formation of silver grains from silver halide upon exposure to radiation. Prior to the time Brock and Brock [47] introduced MAR, it was possible to observe only morphology and substrate incorporation; however, the MAR technique provided a means for directly studying the behavior of microbes using microscopy. One of the initial drawbacks that limited the applicability of MAR for microbial ecology was the limited availability of radioactive substrates incorporating elements with a suitable half-life (such as 14 C, 33/32 P, 3 H, 35 S). However, it is now possible to combine different elements emitting different types of radiation (e.g. alpha or beta) in the same labeling experiment, which has enhanced the utility of MAR [48]. The utility of the MAR technique was enhanced even further when it was recently combined with fluorescent in situ hybridization (FISH) [49–51], thereby enabling quantitative analyses when used with appropriate standard bacteria and radiotracers [52]. Combining the MAR and FISH techniques overcame one of the most significant obstacles to the broad applicability of MAR, namely the inability to identify the microbes incorporating the radioactive substrate, and thereby facilitated application of the technique to diverse habitats [53]. The combination of MAR with catalyzed reporter deposition (CARD)-FISH further extended the capability of examining microbes with low metabolic activity [54]. How-

5.3 Detecting slow metabolism: bulk to single cells

|

107

ever, to the authors’ knowledge, only a single application of this technique to the sedimentary habitat has been reported, which involved freshwater surface sediment, an environment from which target cells can be easily purified [51]. The MAR-FISH approach has yet to be applied to the study of microbes from the subseafloor biosphere, presumably due to characteristics of subseafloor samples that limit its utility. For example, because silver grains formed upon exposure to radiation are usually in the submicron size range but may localize in an area ranging up to several micrometers [48], it is critically important (but difficult) to thoroughly separate the microbial cells from the sediment in order to obtain reliable results. In addition, the spatial resolution of MAR (0.2–2 μm) is not quite sufficient to define a “single cell” within a microbial community, and this factor combined with the necessity of finding appropriate standard microbes for quantitative determination of substrate incorporation potentially limit the application of MAR-FISH in studies of subseafloor microbes.

5.3.3 Quantitative analysis of stable isotope incorporation using NanoSIMS The ability to trace stable isotope incorporation using NanoSIMS has opened up a vast new area of research possibilities in single-cell environmental microbiology [55]. The application of NanoSIMS to the study of microbial ecophysiology was pioneered by McMahon et al. in 2006 [28, 56]. Although they were not the first to demonstrate analysis of microbes using a SIMS-based approach, the introduction of NanoSIMS was a technical breakthrough in terms of the ability to observe stable isotope incorporation at high spatial resolution (50 nm), a level of resolution theoretically unachievable with microscope-based techniques. To date, more than 50 papers reporting the use of NanoSIMS in biology and microbiology have appeared in high-profile journals, indicating that the introduction of NanoSIMS was a groundbreaking development for bioanalysis, and has led to many novel findings in previously unexplored natural systems. A significant advancement made possible by NanoSIMS is the expansion of the range of usable elements to include those previously excluded from use in radioisotope-based methods, such as nitrogen. Lechene et al. exploited this advance to study dinitrogen (N2 )-fixing bacterial symbionts (Teredinibacter urnerae) observed in a region of shipworm (Lyrodus pedicellatus) gills [57]. Despite the essential importance of biological nitrogen fixation, it was previously impossible to quantify the incorporation of nitrogen by individual bacteria or to map the fate of fixed nitrogen in host cells. Shipworms are organisms that primarily feed on wood, and thus must obtain other sources of combined nitrogen for biosynthesis since wood is rich in carbon but not nitrogen. The authors incubated host shipworms in a 15 N-enriched N2 atmosphere. By analyzing individual bacterial symbionts, they found that the 15 N/14 N ratio was increased 39-fold on average compared with the natural ratio. Furthermore, they also found elevated 15 N/14 N ratios in the gill regions immediately adjacent to the

108 | 5 NanoSIMS analysis on subseafloor microbes periphery of individual bacteria, extending even to regions that were free of bacteria, showing that nitrogen fixed by endosymbiotic bacteria within the gills is transported to the proximity of the bacterial cell surface and then used by the shipworm for biosynthesis. Popa et al. [58] and Behrens et al. [59] analyzed the intercellular exchange of carbon and nitrogen in filamentous nitrogen-fixing cyanobacteria, (Anabaena). Anabaena can fix both N2 and carbon dioxide (CO2 ) by differentiating to form heterocysts from vegetative cells. The physically isolated heterocysts are specialized for N2 fixation. NanoSIMS enabled visualization of the transfer of fixed carbon and nitrogen between the vegetative cells and heterocysts, as well as transfer of carbon and nitrogen into epibiotic bacteria that attach to the heterocysts. Using NanoSIMS, Dekas et al. discovered the diazotrophic nature of methane-oxidizing archaea [41]. For some time there was a discrepancy between the calculated rates of oceanic denitrification and N2 fixation, suggesting that other unknown diazotrophic microorganisms may exist that fix substantial amounts of N2 . After metagenomic studies identified nitrogenase genes [60] that also suggested the presence of unidentified diazotrophic organisms, Dekas et al. studied N2 incorporation in aggregates of archaea (ANME-2 group) and bacteria (Desulfosarcina/Desulfococcus [DSS]) that reside in deep-sea sediment and mediate sulfate-dependent anaerobic oxidation of methane (AOM). Sediment samples collected from an active methane seep were anaerobically incubated with methane and 15 N-labeled nitrogen sources. NanoSIMS analyses revealed that the ANME-2 archaea are enriched in 15 N, as are the DSS bacteria residing at the surface of the AOM aggregate, although the ratio is lower. These results showed that ANME-2 archaea can fix nitrogen, which is then shared with DSS bacteria. This finding not only filled gaps in our understanding of the earth’s nitrogen cycle, it resulted in extension of the lower limit of respiratory energy needed to fuel N2 fixation (which costs a number of ATPs) and provided new insights into the relationship between ANME-2 and DSS bacteria. The ability to obtain quantitative information on the ecophysiology of individual microbes in the environment and to compare their metabolic activity/rates is also now possible thanks to NanoSIMS. Musat et al. used NanoSIMS to examine the fixation of 13 C-labeled bicarbonate and the incorporation of 15 N-labeled ammonium in anaerobic phototrophs of oligotrophic meromictic Lake Cadagno [39]. Quantitative measurements of substrate incorporation in individual cells demonstrated that species representing a very small fraction (< 1% of the total cell number) of the microbial community were responsible for 70% and 40% of the total carbon and nitrogen uptake, respectively, and also demonstrated that the uptake rates of substrates for even single cells of the same species differ substantially. Their results showed that microbial populations, even populations composed of the same species, should be regarded as heterogeneous groups of physiologically distinct individual cells. Morono et al. used NanoSIMS to explore the deep subseafloor biosphere, examining microbial activity in 460,000-year-old sediment taken from 219 mbsf and then

5.3 Detecting slow metabolism: bulk to single cells

| 109

amended with various 13 C- and 15 N-labeled substrates. Their work was based upon the idea that even though the deep biosphere is highly energy-starved naturally, the organisms may still be capable of feeding actively if provided a diet rich in organic substrates. Following incubation for about two months, incorporation of almost all of the substrates (except for methane) was observed. As many as 76% of the microbes assimilated the added substrates (such as glucose and amino acids), which demonstrated that a large fraction of subseafloor microbes remain alive. The assimilation rate was in the 10 × 10−18 mol cell−1 d−1 range, which is slow compared with the anaerobic phototrophs discussed above; however, this rate was notably higher (more than 1000 times faster) than the mean rate of organic carbon assimilation [61]. Another finding of this work was that the ratio of nitrogen incorporated from ammonium to the total number of cellular nitrogen consistently exceeded those ratios of carbon (󳶳 Fig. 5.3). The data showed that subseafloor cells preferentially require nitrogen assimilation from ammonia for the better recovery in vitro. This can be contradictory to presence of enough ammonium (∼15 mM) in the sediment porewater. One conceivable explanation is that, although most subseafloor microbes could assimilate nitrogen from ammonium, they might suppress ammonium uptake in situ in order to conserve energy (i.e. although porewater from sediments at 219 mbsf contain 15 mM ammonium, the metabolism required for nitrogen assimilation into amino acids or for carbamoyl phosphate synthesis might consume excess energy required for long-term survival). Another possibility is that subseafloor cells preferentially incorporate high-C/N buried organic matter into biomass rather than ammonium, which may subsequently result (a)

(b) 60

5

29 6

20 19

30 1 4 3

21

27

22

28 26 25 33

Ratio of incorporated C in total C (Atom% per area)

32

31

2

17 18 16 15 14

23 24 11 12

9 10

13

2 µm

8 7

50 40

9 6

30 20 27

10

5 15 16 18 2 14 17 8 20 31 32 3 13 24 22 23 12 4 1921 1 10 26 7 25 11 33 29

30

28

0 0

10

20

30

40

50

Ratio of incorporated N in total N (Atom% per area)

Fig. 5.3: Localization of intracellular 13 C- and 15 N-incorporation in a deep subseafloor microbial community. (a) Example of overlaid ratio images of 13 CGlucose/12 C (red) and 15 Nammonium/14 N (green) visualized using NanoSIMS. (b) Scatter plot of atomic percentages of carbon and nitrogenincorporation in each numbered area shown in (a). Images are reproduced from [42].

60

110 | 5 NanoSIMS analysis on subseafloor microbes in nitrogen-deprived biomass. However, these possibilities need to be carefully examined in future studies by evaluating the carbon and nitrogen composition (i.e. the C/N ratio) and physiological nature of subseafloor cells. What is the ultimate sensitivity of NanoSIMS in terms of detecting substrate incorporation? Lechene et al. discussed the sensitivity of 14 C detection after radiotracer incubation and estimated that NanoSIMS is at least 103 -fold more sensitive than MAR [28]. With respect to stable isotope analyses, the standard deviation can be better than ±1‰. To determine the incorporation of very small amounts of stable isotope, extremely careful analysis of standards (microbial cells of known isotopic composition) and determination of differences in incorporation between cells are required. For example, if we assume that the mass of carbon in a single microbial cell is 20 fg, and that we can detect incorporation of a 13 C-labeled carbon substrate at a quantity equivalent to 1% of the cellular carbon, which represents an increase of about 100% in the natural 13 C/12 C isotopic ratio, this would correspond to the incorporation of 17 amol (amol = 10−18 mol) of carbon into cellular biomass. However, in order to ensure that the data obtained are reliable, repeated analyses of standards are necessary, as well as standardization of sample preparation and analytical procedures. In addition, there are several critical variables in this type of calculation, including the mass of the microbial cells and their elemental composition. Unfortunately, no data regarding the elemental composition of subseafloor cells are available, although it has been assumed that subseafloor cells contain much less carbon than cultured surface microbes (for example, the carbon content of cultured E. coli is roughly 150 fg/cell [Morono et al. unpublished]).

5.4 Bridging identification and functional analysis of microbes using elemental labeling Orphan and coworkers were the first to directly link microbial identification and analysis of metabolic activity by combining FISH and SIMS (FISH-SIMS) [38]. Using the FISH-SIMS approach, the authors found that members of microbial aggregates composed of a specific archaeal group (ANME-2) surrounded by sulfate-reducing bacteria were significantly depleted in 13 C, suggesting that these organisms are involved in anaerobic assimilation of isotopically light methane produced in marine sediments. In addition, the authors also made use of the destructive nature of SIMS-based analyses to observe the isotopic profile of the aggregates in a depth-wise manner. They found that the outer shell of sulfate-reducing bacterial aggregates is less depleted in 13 C than the inner core of ANME-2 cell clusters, indicating that the centrally located ANME-2 cells are the primary methane consumers and that the surrounding bacterial partners receive the methane-derived products. This pioneering study was achieved by identifying FISH-stained cells using microscopy and then reanalyzing them using SIMS. The regions of interest were identified by mapped information before introduc-

5.4 Bridging identification and functional analysis of microbes using elemental labeling | 111

tion of the instrument, sometimes with the aid of a micrometric stage [56]. In SIMS or NanoSIMS, the sample is optically observed by charge-coupled device (CCD)-camera to determine the analysis location. The observable area is limited (i.e. 100 μm×200 μm in square), and it can be difficult to precisely recognize single microbial cells in environmental samples, especially those containing mineral grains, such as subseafloor sediments. To facilitate more precise identification of microbes, in 2008, three laboratories reported development of SIMS-compatible detection techniques that allow simultaneous isotopic measurement and phylogenetic identification of single microbial cells [39, 44, 59]. The techniques use halogens because they produce high ionization yields and have relatively low natural background. The SIMSISH technique, developed by Li et al., replaced the 2󸀠 -deoxycytidine of oligonucleotide probes with 5󸀠 iodo-2󸀠 -deoxycytidine [44]. With this technique, the authors visualized 13 C-methanolincorporating bacterial and archaeal cells in a municipal solid waste reactor. Although SIMSISH requires just replacing probes and is thus quite simple and useful, the number of halogen atoms per probe (4–6 iodine atoms) can limit its application for detecting microorganisms with low ribosomal RNA content [62, 63]. The more sensitive techniques EL (enhanced elemental labeling)-FISH and HISH (halogen in situ hybridization)-SIMS, developed by Behrens et al. [59] and Musat et al. [39], respectively, utilize horseradish peroxidase (HRP)-conjugated oligonucleotide probes and subsequent tyramide signal amplification for depositing a number of fluorine- or bromine-containing fluorophores (e.g. 544Br, BODIPY TMR-X and Oregon Green 488-X, 󳶳 Fig. 5.4). Similar to CARD-FISH, amplification of the signal due to deposition of tyramide increases the signal-to-noise ratio, and the authors used the technique to efficiently identify an association of epibiotic Rhizobium and Anabaena species, as well as species of anaerobic phototrophs. However, because of the relatively large size of the HRP-labeled probes, permeabilization of the cells is required for probe penetration, and this sometimes causes failure in FISH-based identifications [42]. High background fluorine signals have been obtained in analyses of marine sediment samples (Morono et al. unpublished), presumably due to the presence of fluorite (CaF2 ) minerals. In addition, other halide minerals or high-molecular-weight brominated materials found in marine sediments [64] apparently hinder halogenbased identification of microbes found in the subseafloor biosphere. Another possible approach is to use gold nanoparticles for labeling microbial cells. The ionization yield of gold is comparable to that of halogen elements in SIMS. Immunogold staining is a well-known technique for examining histological samples by both transmission [65] and scanning [66, 67] electron microscopy. Recently, Schmidt et al. developed a gold-FISH approach that employs a similar signal amplification process as EL-FISH, but involves deposition of biotinylated tyramides. Streptavidin conjugated with a fluorophore and nanogold particle (1.4 nm) are bound to the biotin for detection using either fluorescence or electron microscopy. Milucka et al. labeled methyl CoM reductase (Mcr) in an aggregate mediating AOM using immunogold

112 | 5 NanoSIMS analysis on subseafloor microbes 12

0

C14N-

1.25

19

2.50 3.75 5.00 (x105 counts per pixel)

0

F-

2.5

5.0 7.5 1.0 (x102 counts per pixel)

Fig. 5.4: Example of HISH (EL-FISH)-SIMS. E. coli cells were hybridized with HRP-labeled oligonucleotide probes and tyramides conjugated with Oregon Green 488-X was deposited on the cells containing HRP-probes. Then the cells were analyzed with NanoSIMS 50L.

staining with anti-Mcr antibodies and identified the site of Mcr localization in ANME-2 cells by NanoSIMS analysis [43]. An interesting application of gold nanoparticles is amplification of signals through autometallography (AMG) [68]. Autometallography is based on the same principle as the physical development of a photograph. In the presence of a few atoms of gold, silver ions are reduced to metallic silver, initiating a deposition reaction that proceeds exponentially until terminated by treatment with acid. This amplification process was later extended to use gold ions as well [69]. Despite the potential of the AMG reaction and its many applications for analysis of histological samples, to date there have been no demonstrations of its use for microbial phylogenetic analyses involving NanoSIMS. The sensitivity of AMG is comparable to that of CARD-FISH, and the ease with which ionized gold can be introduced into microbial cells through cell membrane amplification can expand the range of detectable microbial species. Therefore, although there is a cautionary tale that states that surface-charged minerals can cause nonspecific reactions [70] and that further methodological development is needed prior to application to subseafloor samples, AMG is still an attractive candidate microbial identification tool.

5.5 Critical step for successful NanoSIMS analysis: sample preparation As already reviewed elsewhere [31, 71, 72], NanoSIMS analysis requires careful and proper preparation of samples. The vacuum in the analysis chamber of a NanoSIMS instrument is nominally 10−10 torr, and samples are situated in very close proximity (about 100 μm) to the series of coaxial lenses, which carry a high voltage (16 kV).

5.5 Critical step for successful NanoSIMS analysis: sample preparation

| 113

Therefore, samples that are to be analyzed by NanoSIMS must be solid, flat and vacuum-tolerant. It is also better if the samples are electrically conductive in order to avoid charging of the sample. Usually, the microbial cells are placed on a conductive material such as a silicon wafer, glass slide, or polycarbonate membrane. Glass plates or polycarbonate membranes are coated with a thin layer of gold, platinum, or carbon, for example, so that they will be conductive. It is also possible to coat the microbial cells with a conductive material, as is done in preparation of samples for SEM observation, though it is necessary to increase the presputtering time to obtain steady state of the secondary ions from the cell beneath the coating layer. One of the unique characteristics of the subseafloor, and geologic samples in general, is that the microbial cells exist in a matrix of mineral grains. If we assume a spherical cell has a diameter of 1 μm, the corresponding biovolume of the cell should be roughly 0.5 × 10−12 cm3 . The surface sediment harbors 109 cells per cm3 . As microbial cells comprise only 0.05% of the sediment volume, the quantity of other solids is 1400 × greater than the quantity of microbes (assuming the sediment porosity is 75%). Therefore, greater effort is required to prepare geologic samples for microbiological analyses than samples from aquatic habitats. This problem is obvious for samples in which the microbial cell abundance is low. A NanoSIMS isotope imaging analysis usually encompasses an area of 30 μm × 30 μm. With samples prepared by conventional filtration (suspending the sample in buffer and passing it through a membrane filter), the number of single cells per analysis area will be less than one if the cell abundance in the sediment is less than 107 cells per cm3 . Considering that the CCD view of microbial cells in NanoSIMS is hampered by the presence of a significant number of grains, analyzing low-abundance samples using this technique is a waste of time. Extraction of sediments using density layer(s) is a useful technique for increasing the number of microbial cells for analysis [73, 74]. Sorting samples by cell sorter in order to increase the number of cells per unit of area is another way to increase the efficiency of NanoSIMS. Together with the above approaches, drawing microlines on samples can increase the analytical efficiency of NanoSIMS. As already stated before, the NanoSIMS CCD view enables observation of a 100–200 μm area. Finding the orientation of the sample stub may take a considerable amount of time, however. 󳶳 Figure 5.5 shows a sample with microlines drawn onto it using laser microdissection (LMD). The LMD technique was developed for dissecting target cells from histologic samples and its use has now been expanded to other fields, including microbiology. The advantage of LMD is that in addition to lines, characters that provide information regarding the samples can also be drawn. Using LMD, grids can be drawn on samples to enable matching of CARD-FISH and NanoSIMS ion images, thereby enhancing identification of microbial cells [75]. Unfortunately, preparing samples for SIMS-based isotope quantification may result in unintentional alteration of the ratio of isotopes in microbes, leading to uncertainty as to the validity of results. How to address this issue is sometimes controversial, since each of the steps in the sample preparation protocol is important. Chem-

114 | 5 NanoSIMS analysis on subseafloor microbes

Identified archaeal cell

Fig. 5.5: Filter-labeling with a laser microdissection system for CARD-FISH and NanoSIMS analyses on the identical filter location. The small box-photo shows NanoSIMS image of 13 C-incorporated cells (red) overlaid on CARDFISH image for the detection of archaeal cells. The size of each grid is about 50 μm × 50 μm, marked by a laser microdissection system. Images are reproduced from [42].

ical fixation of cells is commonly used in FISH to deactivate cells and stabilize their structure [76]; however, fixation carries the potential of contaminating cells with carbon from foreign sources [31, 39, 77]. Other steps that are potential sources of carbon and/or nitrogen contamination include hybridization of oligonucleotide probes for FISH and deposition of the fluorogenic tyramides used in CARD-FISH. For example, if 37,000 molecules of an oligonucleotide probe are introduced into a target cell (which is sufficient to produce distinctive fluorescence like that of FISH-stained E. coli cells [63]), the corresponding amounts of contaminant carbon and nitrogen would be around 0.3 fg and 0.1 fg, respectively. Assuming cellular carbon and nitrogen contents of 20 fg and 5 fg, the levels of contaminating carbon and nitrogen associated with the oligonucleotide probe would be 1.5% and 2%, respectively. Considering that the increase in the ratio of 13 C or 15 N following stable isotope incubation usually exceed 100% of the natural ratio of 13 C to 15 N, contamination associated with sample preparation for FISH can be regarded as having a minimal impact on the quantification results. On the other hand, in the case of measurement of natural isotopic abundance, in which more precise quantification is required (usually few tens of permil level [38]), the data should be carefully treated, as the relative impact could be much larger than with stable isotope experiments.

5.6 Future directions Thus far, we have reviewed several highly sensitive techniques for detecting metabolic activity in single microbial cells, and have also discussed related techniques that combine identification of microbes with functional analysis in order to characterize the subseafloor biosphere, with an emphasis on NanoSIMS. However, it is also important to understand what cannot be analyzed with NanoSIMS and how its limitations can be overcome. In principle, SIMS-based approaches can be used to analyze the elemental abundance of a subject material very precisely and quantitatively. Therefore, the incorporation of a stable isotope-labeled substrate can be analyzed in a very sensitive

5.6 Future directions

| 115

manner, as shown already. However, the metabolic activity that results in transformation of the substrate into other compounds that are then excreted from the cell cannot be traced, since no material indicative of the activity is left in the cell. For example, NanoSIMS cannot be used to directly measure sulfate reduction. Fike et al. overcame this problem by capturing transformed sulfide on silver foil or a plate [40]. NanoSIMS analysis of Ag2 S formed on silver foil enabled visualization of the relative distribution of 32 S and 𝛿34 S on a micrometer-scale as well as localization of the 34 S-depleted region that was suggestive of potential metabolic activity of sulfate disproportionators. Milucka et al. combined NanoSIMS and Raman spectroscopy to find intermediate compounds in AOM syntrophy [43]. Based on enrichment culture results and multiple independent experiments, the authors hypothesized that zero-valent sulfur (S0 ) plays an important role in the syntrophic relationship between ANME-2 archaea and DSS bacteria. To confirm this hypothesis, they used confocal micro-Raman spectroscopy with approximately 1 μm resolution, and were able to detect the peak characteristic of elemental sulfur in ANME cells identified by FISH. Also, NanoSIMS analysis indicated that the sulfur content was higher in ANME cells than DSS bacteria. Through an additional array of experiments, including the use of 35 S-based incubation and detection, the authors obtained evidence of intercellular sulfur cycling that involved sulfate reduction by ANME archaea and disulfide disproportionation by DSS bacteria. The examples cited here, together with the history of technical developments in all fields of science, demonstrate that the inspiration to combine techniques is one of the critical forces driving scientific progress. Behrens et al. proposed another interesting combination, NanoSIMS coupled with scanning transmission X-ray microscopy (SXTM) [72], which has a similar spatial resolution as NanoSIMS and has been widely used in analyses of organic materials. The method is based on the chemical speciesspecific absorption of X-rays (near-edge X-ray absorption fine structure [NEXAFS]), and allows for visualization of biomolecule localization and the generation of quantitative information. Although integration of the NanoSIMS and SXTM techniques is still in the conceptual stage and remains technically challenging, if successful, the combined method could be used to synergistically explore hidden aspects of microbial life and microbial interactions. Combining single-cell genomics [78] and NanoSIMS presents interesting possibilities for matching genomic information (genotype) with real phenomena (phenotype). Although both NanoSIMS and single-cell genomic analyses are destructive of the sample material, NanoSIMS can provide meaningful data while stripping only the thin surface atomic layers and thus could potentially be combined with genomic analysis techniques. Genomic analyses of single cells have compensated for the lack of physiological information on lineages without cultivated relatives [79–82]. Genomic information, which can be very broad, dictates the “potential” function of target cells; thus, combining genomic data with substrate incorporation data generated by NanoSIMS would be yet another synergistic combination of technologies.

116 | 5 NanoSIMS analysis on subseafloor microbes Analytical techniques are now available that are sufficiently precise and sensitive to allow direct observation of single cells. Another critical question that must be addressed, however, is how we can extend or connect single-cell activity, identity, and function data to the whole biosphere. How many single cells need to be analyzed to obtain a complete understanding or description of the processes occurring within the whole microbial community? Massive parallel sequencing efforts have revealed that the biosphere is multiple orders of magnitude more complex than previously thought [83]. For the future, it will be required to obtaining an integrated view of the microbe and its surrounding environment at a level of resolution ranging from the single cell to the entire biosphere.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]

[14] [15]

Morita RY, ed. Bacteria in oligotrophic environments: Starvation-survival lifestyle. NY: Chapman & Hall; 1997. Roussel EG, Bonavita M-AC, Querellou J, et al. Extending the Sub-Sea-Floor Biosphere. Science 320 (2008), 1046. Whitman WB, Coleman DC, Wiebe WJ. Prokaryotes: The unseen majority. Proc Natl Acad Sci USA 95 (1998), 6578–6583. Parkes RJ, Cragg BA, Wellsbury P. Recent studies on bacterial populations and processes in subseafloor sediments: A review. Hydrogeol J 8 (2000), 11–28. Lipp JS, Morono Y, Inagaki F, Hinrichs KU. Significant contribution of Archaea to extant biomass in marine subsurface sediments. Nature 454 (2008), 991–994. Kallmeyer J, Pockalny R, Adhikari R, Smith D, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16 213–16 216. D’Hondt S, Jørgensen BB, Miller DJ, et al. Distributions of microbial activities in deep subseafloor sediments. Science 306 (2004), 2216–2221. D’Hondt S, Rutherford S, Spivack AJ. Metabolic activity of subsurface life in deep-sea sediments. Science 295 (2002), 2067–2070. D’Hondt S, Spivack AJ, Pockalny R, et al. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106 (2009), 11 651–11 656. Jørgensen BB, Boetius A. Feast and famine-microbial life in the deep-sea bed. Nat Rev Microbiol 5 (2007), 770–781. Schippers A, Neretin LN, Kallmeyer J, et al. Prokaryotic cells of the deep sub-seafloor biosphere identified as living bacteria. Nature 433 (2005), 861–864. Fischer JP, Ferdelman TG, D’Hondt S, Røy H, Wenzhofer F. Oxygen penetration deep into the sediment of the South Pacific gyre. Biogeosciences 6 (2009), 1467–1478. Colwell FS, Boyd S, Delwiche ME, Reed DW, Phelps TJ, Newby DT. Estimates of Biogenic Methane Production Rates in Deep Marine Sediments at Hydrate Ridge, Cascadia Margin. Appl Environ Microbiol 74 (2008), 3444–3452. Biddle JF, Lipp JS, Lever MA, et al. Heterotrophic Archaea dominate sedimentary subsurface ecosystems off Peru. Proc Natl Acad Sci USA 103 (2006), 3846–3851. Teske A, Sørensen KB. Uncultured archaea in deep marine subsurface sediments: have we caught them all? ISME J 2 (2007), 3–18.

References | 117

[16] Inagaki F, Nunoura T, Nakagawa S, et al. Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments on the Pacific Ocean Margin. Proc Natl Acad Sci USA 103 (2006), 2815–2820. [17] Webster G, John Parkes R, Cragg BA, Newberry CJ, Weightman AJ, Fry JC. Prokaryotic community composition and biogeochemical processes in deep subseafloor sediments from the Peru Margin. FEMS Microbiol Ecol 58 (2006), 65–85. [18] Parkes RJ, Webster G, Cragg BA, et al. Deep sub-seafloor prokaryotes stimulated at interfaces over geological time. Nature 436 (2005), 390–394. [19] Sørensen KB, Teske A. Stratified Communities of Active Archaea in Deep Marine Subsurface Sediments. Appl Environ Microbiol 72 (2006), 4596–4603. [20] Biddle JF, Fitz-Gibbon S, Schuster SC, Brenchley JE, House CH. Metagenomic signatures of the Peru Margin subseafloor biosphere show a genetically distinct environment. Proc Natl Acad Sci USA 105 (2008), 10 583–10 588. [21] Webster G, Blazejak A, Cragg BA, et al. Subsurface microbiology and biogeochemistry of a deep, cold-water carbonate mound from the Porcupine Seabight (IODP Expedition 307). Environ Microbiol 11 (2009), 239–257. [22] Lever MA, Heuer VB, Morono Y, et al. Acetogenesis in Deep Subseafloor Sediments of The Juan de Fuca Ridge Flank: A Synthesis of Geochemical, Thermodynamic and Gene-based Evidence. Geomicrobiol J 27 (2010), 183 –211. [23] Futagami T, Morono Y, Terada T, Kaksonen AH, Inagaki F. Dehalogenation Activities and Distribution of Reductive Dehalogenase Homologous Genes in Marine Subsurface Sediments. Appl Environ Microbiol 75 (2009), 6905–6909. [24] Hoehler TM. Biological energy requirements as quantitative boundary conditions for life in the subsurface. Geobiology 2 (2004), 205–215. [25] Price PB, Sowers T. Temperature dependence of metabolic rates for microbial growth, maintenance and survival. Proc Natl Acad Sci USA 101 (2004), 4631–4636. [26] Christien F, Downing C, Moore KL, Grovenor CRM. Quantitative grain boundary analysis of bulk samples by SIMS. Surf Interf Anal 45 (2013). [27] Messenger S, Keller L, Stadermann F, Walker R, Zinner E. Samples of stars beyond the solar system: silicate grains in interplanetary dust. Science 300 (2003), 105–108. [28] Lechene C, Hillion F, McMahon G, et al. High-resolution quantitative imaging of mammalian and bacterial cells using stable isotope mass spectrometry. J Biol 5 (2006), 20. [29] Castaing R, Slodzian GJ. Optique corpusculaire—premiers essais de microanalyse par emission ionique secondaire. Microscopie 1 (1962), 395–399. [30] Yurimoto H, Nagashima K, Kunihiro T. High precision isotope microimaging of materials. Appl Surf Sci (2003) 203–204. [31] Wagner M. Single-Cell Ecophysiology of Microbes as Revealed by Raman Microspectroscopy or Secondary Ion Mass Spectrometry Imaging. Ann Rev Microbiol 63 (2009), 411–429. [32] Ito M, Messenger S, Keller L, Rahmann ZU, Ross DK, Nakamura-Messenger K. FIB-NanoSIMSTEM Coordinated Study of a Wark-lovering Rim in a Vigarano Type A CAI. Lunar and Planetary Science Conference 2010. [33] Weber P, Graham G, Teslich N, et al. NanoSIMS imaging of Bacillus spores sectioned by focused ion beam. J Microsc 238 (2010), 189–199. [34] Peter H. NanoSIMS perspectives for nuclear astrophysics. New Astronomy Rev (2002), 46. [35] Nguyen A, Zinner E. Discovery of ancient silicate stardust in a meteorite. Science 303 (2004), 1496–1499. [36] Nakamura-Messenger K, Messenger S, Keller L, Clemett S, Zolensky M. Organic globules in the Tagish Lake meteorite: remnants of the protosolar disk. Science 314 (2006), 1439–1442.

118 | 5 NanoSIMS analysis on subseafloor microbes [37] Ito M, Messenger S. Isotopic imaging of refractory inclusions in meteorites with the NanoSIMS 50L. Appl Surf Sci (2008), 255. [38] Orphan VJ, House CH, Hinrichs K-U, McKeegan KD, DeLong EF. Methane-Consuming Archaea Revealed by Directly Coupled Isotopic and Phylogenetic Analysis. Science 293 (2001), 484– 487. [39] Musat N, Halm H, Winterholler Br, et al. A single-cell view on the ecophysiology of anaerobic phototrophic bacteria. Proc Natl Acad Sci USA 105 (2008), 17 861–17 866. [40] Fike DA, Gammon CL, Ziebis W, Orphan VJ. Micron-scale mapping of sulfur cycling across the oxycline of a cyanobacterial mat: a paired nanoSIMS and CARD-FISH approach. ISME J 2 (2008), 749–759. [41] Dekas AE, Poretsky RS, Orphan VJ. Deep-Sea Archaea Fix and Share Nitrogen in Methane-Consuming Microbial Consortia. Science 326 (2009), 422–426. [42] Morono Y, Terada T, Nishizawa M, et al. Carbon and nitrogen assimilation in deep subseafloor microbial cells. Proc Natl Acad Sci USA 108 (2011), 18 295–18 300. [43] Milucka J, Ferdelman T, Polerecky L, et al. Zero-valent sulfur is a key intermediate in marine methane oxidation. Nature (2012), 541–546. [44] Li T, Wu T-D, Mazeas L, et al. Simultaneous analysis of microbial identity and function using NanoSIMS. Environ Microbiol 10 (2008), 580–588. [45] Parkes RJ, Cragg BA, Fry JC, Herbert RA, Wimpenny JWT. Bacterial Biomass and Activity in Deep Sediment Layers from the Peru Margin. Philos Trans R Soc London 331 (1990), 139–152. [46] Hoehler T, Jørgensen B. Microbial life under extreme energy limitation. Nat Rev Microbiol 11 (2013), 83–94. [47] Brock TD, Brock ML. Autoradiography as a Tool in Microbial Ecology. Nature 209 (1966), 734– 736. [48] Nielsen J, Nielsen P. Advances in microscopy: microautoradiography of single cells. Met Enzymol 397 (2005), 237–256. [49] Lee N, Nielsen PH, Andreasen KH, et al. Combination of Fluorescent In Situ Hybridization and Microautoradiography—a New Tool for Structure-Function Analyses in Microbial Ecology. Appl Environ Microbiol 65 (1999), 1289–1297. [50] Ouverney CC, Fuhrman JA. Combined Microautoradiography–16S rRNA Probe Technique for Determination of Radioisotope Uptake by Specific Microbial Cell Types In Situ. Appl Environ Microbiol 65 (1999), 1746–1752. [51] Gray ND, Howarth R, Pickup RW, Jones JG, Head IM. Use of Combined Microautoradiography and Fluorescence In Situ Hybridization To Determine Carbon Metabolism in Mixed Natural Communities of Uncultured Bacteria from the Genus Achromatium. Appl Environ Microbiol (2000), 66. [52] Nielsen JL, Christensen D, Kloppenborg M, Nielsen PH. Quantification of cell-specific substrate uptake by probe-defined bacteria under in situ conditions by microautoradiography and fluorescence in situ hybridization. Environ Microbiol (2003), 5. [53] Wagner M, Nielsen PH, Loy N, Nielsen JL, Daims H. Linking microbial community structure with function: fluorescence in situ hybridization-microautoradiography and isotope arrays. Cur Opin Biotechnol (2006), 17. [54] Alonso C, Pernthaler J. Concentration-Dependent Patterns of Leucine Incorporation by Coastal Picoplankton. Appl Environ Microbiol (2006), 72. [55] Kuypers MMM, Jørgensen BB. The future of single-cell environmental microbiology. Environ Microbiol 9 (2007), 6–7. [56] McMahon G, Glassner BJ, Lechene CP. Quantitative imaging of cells with multi-isotope imaging mass spectrometry (MIMS)—Nanoautography with stable isotope tracers. Appl Surf Sci (2006), 252.

References |

119

[57] Lechene C, Luyten Y, McMahon G, Distel D. Quantitative imaging of nitrogen fixation by individual bacteria within animal cells. Science 317 (2007), 1563–1566. [58] Popa R, Weber PK, Pett-Ridge J, et al. Carbon and nitrogen fixation and metabolite exchange in and between individual cells of Anabaena oscillarioides. ISME J 1 (2007), 354–360. [59] Behrens S, Lösekann T, Pett-Ridge J, et al. Linking Microbial Phylogeny to Metabolic Activity at the Single-Cell Level by Using Enhanced Element Labeling-Catalyzed Reporter Deposition Fluorescence In Situ Hybridization (EL-FISH) and NanoSIMS. Appl Environ Microbiol 74 (2008), 3143–3150. [60] Pernthaler A, Dekas A, Brown C, Goffredi S, Embaye T, Orphan V. Diverse syntrophic partnerships from deep-sea methane vents revealed by direct cell capture and metagenomics. Proc Natl Acad Sci USA 105 (2008), 7052–7057. [61] Jørgensen BB. Deep subseafloor microbial cells on physiological standby. Proc Natl Acad Sci USA 108 (2011), 18 193–18 194. [62] Amann RI, Ludwig W, Schleifer KH. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59 (1995), 143–169. [63] Hoshino T, Yilmaz LS, Noguera DR, Daims H, Wagner M. Quantification of Target Molecules Needed To Detect Microorganisms by Fluorescence In Situ Hybridization (FISH) and Catalyzed Reporter Deposition-FISH. Appl Environ Microbiol 74 (2008), 5068–5077. [64] Gordon WG. The diversity of naturally occurring organobromine compounds. Chem Soc Rev (1999), 28. [65] Verkleij A, Koster A, Müller W, Humbel B. Immuno-Gold Labeling in Transmission Electron Microscopy. NATO ASI Ser A 309 (1999), 339–356. [66] Goldberg M, Fiserova J. Immunogold labelling for scanning electron microscopy. Met Mol Biol 657 (2010), 297–313. [67] Hermann R, Walther P, Müller M. Immunogold labeling in scanning electron microscopy. Histochem Cell Biol 106 (1996), 31–39. [68] Danscher G. Autometallography. A new technique for light and electron microscopic visualization of metals in biological tissues (gold, silver, metal sulfides and metal selenides). Histochemistry 81 (1984), 331–335. [69] Hainfeld J, Powell RD, Stein J, et al. Gold-based autometallography. Microsc Microanal 5 (1999), 486–487. [70] Schmidt H, Eickhorst T, Mussmann M. Gold-FISH: a new approach for the in situ detection of single microbial cells combining fluorescence and scanning electron microscopy. Syst Applied Microbiol 35 (2012), 518–525. [71] Musat N, Foster R, Vagner T, Adam B, Kuypers MM. Detecting metabolic activities in single cells, with emphasis on nanoSIMS. FEMS Microbiol Rev 36 (2012), 486–511. [72] Behrens S, Kappler A, Obst M. Linking environmental processes to the in situ functioning of microorganisms by high-resolution secondary ion mass spectrometry (NanoSIMS) and scanning transmission X-ray microscopy (STXM). Environ Microbiol (2012). [73] Kallmeyer J, Smith DC, Spivack AJ, D’Hondt S. New cell extraction procedure applied to deep subsurface sediments. Limnol Oceanogr Methods 6 (2008), 236–245. [74] Morono Y, Terada T, Kallmeyer J, Inagaki F. An Improved Cell Separation Technique for Marine Subsurface Sediments: Applications for High-throughput Analysis Using Flow Cytometry and Cell Sorting. Environ Microbiol (2013, in press). [75] Imachi H, Aoi K, Tasumi E, et al. Cultivation of methanogenic community from subseafloor sediments using a continuous-flow bioreactor. ISME J 2011. [76] Fox C, Johnson F, Whiting J, Roller P. Formaldehyde fixation. J Histochem Cytochem 33 (1985), 845–853.

120 | 5 NanoSIMS analysis on subseafloor microbes [77] Karlsen F, Kalantari M, Chitemerere M, Johansson B, Hagmar B. Modifications of human and viral deoxyribonucleic acid by formaldehyde fixation. Lab Invest 71 (1994), 604–611. [78] Stepanauskas R. Single cell genomics: an individual look at microbes. Curr Opin Microbiol 15 (2012), 613–620. [79] Lloyd K, Schreiber L, Petersen D, et al. Predominant archaea in marine sediments degrade detrital proteins. Nature (2013). [80] Marcy Y, Ouverney C, Bik EM, et al. Dissecting biological “dark matter” with single-cell genetic analysis of rare and uncultivated TM7 microbes from the human mouth. Proc Natl Acad Sci USA (2007), 0704662104. [81] Podar M, Abulencia CB, Walcher M, et al. Targeted Access to the Genomes of Low-Abundance Organisms in Complex Microbial Communities. Appl Environ Microbiol 73 (2007), 3205–3214. [82] Stepanauskas R, Sieracki ME. Matching phylogeny and metabolism in the uncultured marine bacteria, one cell at a time. Proc Natl Acad Sci USA 104 (2007), 9052–9057. [83] Sogin ML, Morrison HG, Huber JA, et al. Microbial diversity in the deep sea and the underexplored rare biosphere. Proc Natl Acad Sci USA 103 (2006), 12 115–12 120.

Karen G. Lloyd

6 Quantifying microbes in the marine subseafloor: some notes of caution 6.1 Introduction The subseafloor microbial biosphere is one of the largest biomes on Earth, containing about half of all cells found in oceanic environments [1]. These microbes are responsible for deep carbon turnover and other key geochemical cycles [2]. They are also taxonomically diverse on a sub-cm3 spatial scale [3], making it very difficult to study their ecological roles directly in the environment. The best way to know for certain what physiological traits are possible for a given microbe is to grow a strain of it in pure culture in lab. But this method has two major limitations. First, even if a strain expresses a particular phenotype in lab it may not express it in nature, either because environmental conditions are not appropriate for that function, or because the dominant strains in the environment do not carry the trait. Perhaps a larger problem is that, as is often stated, the vast majority of microbes identified through 16S rRNA gene surveys have never been grown in pure culture, so for most environmental microbes, their potential functions remain completely unknown. Culturing efforts thus far in the subseafloor have been unable to cultivate the main taxa identified through DNA studies [4, 5]. This problem is not limited to a few strains, but encompasses even higher order taxa, since many archaeal and bacterial phyla have no cultured representatives [6, 7]. For these reasons, microbes must also be studied “in the wild” by using advanced molecular biological and microscopy techniques, without requiring them to first go through the bottleneck of rapid growth required to reach pure culture in lab. A key component of many of these techniques is the quantification of microbes in situ. If microbes can be accurately quantified, then they can be linked to geochemical reactions measured in the same sediments. In the quest to determine the minimum energy requirements for cells, quantification is essential for determining microbial process rates per cell [8]. Microbes can be quantified either generally, with total cell counts, or by their taxonomy – so that particular groups of microbes can be quantified. Total cell counts are made with a general DNA-binding fluorophore that freely diffuses through cell membranes and likely binds DNA in all cells equally well, so nothing needs to be known about the identity of the cell in order to count it. Cells can be overcounted, since there can be background staining of mineral grains, or undercounted as cells are hidden by mineral grains [9]. One drawback of all cell counting methods is that the detection limit is high relative to many marine subsurface environments [10]. In order to calculate the detection limits of cell counting methods, one must first consider the total number of cells that need to be counted to achieve a statistically significant cell

122 | 6 Quantifying microbes in the marine subseafloor: some notes of caution density. Since the 95% confidence intervals can be approximated as twice the square root of the number of cells counted, Fry et al. [9] recommended counting 400 cells per sample, divided among 3–5 replicate filters, to yield 10% confidence intervals [9]. Counting 400 cells is a reasonable proposition when dealing with sediments containing 108 or 109 cells/mL. However, many deep-sea sediments have far fewer cells, and the time required to achieve this statistical confidence is untenable [10]. The standard in deep subsurface sediments is to achieve a 95% confidence interval of around 30% of the mean [11, 12], which requires counting 35 cells. To determine the number of fields required to count 35 cells at different cell densities, I rearranged the equation for determining microbial cell densities [9]:

𝑇= to

𝑁𝑔 =

𝑁𝑐 𝐴 𝑓 𝐷 𝑁𝑔 𝑎𝑉 𝑁𝑐 𝐴 𝑓 𝐷 𝑇𝑎𝑉

where 𝑇 is the total cell density, 𝑁𝑐 is the number of cells counted, 𝐴 𝑓 is the area of the filter, 𝐷 is the dilution factor for the sample, 𝑁𝑔 is the number of grids (also called fields of view) counted, 𝑎 is the grid or field area and 𝑉 is the volume of sample filtered. I assumed microscope and grid sizes, typical for a Leica epifluorescence microscope of 6400 μm2 for a 10 × 10 grid of 8 μm squares and filter area of 4.9 × 108 μm2 . 󳶳 Figure 6.1 shows that the number of grids that must be counted as a power function of the log10 cell density decrease. Commonly in deep subsurface samples, 200 fields are counted [12], which allows detection down to 105 cells/mL in 󳶳 Fig. 6.1. Out of these 200 fields, on average 165 of them are expected to have no cells, and the rest will have one cell. This means that the operator must look at 200 fields and correctly identify whether each one contains a single cell or no cells. The operator must also resist the urge to “hunt” for cells by stopping more frequently on fields that contain a cell. This would bias the result toward a higher cell density. In addition, false positive error rates would be expected to be roughly the same per field as a sample with higher cell densities, since the sediment particles per field are the same. A false positive error rate of 1 in 10 fields, will have a 6% effect on samples that can be counted in 20 fields (this corresponds to a total cell density of ∼ 2.7 × 106 cell mL−1 ). This same false positive error rate will increase the overall error rate by 57% when 200 fields are counted. It is meaningless to propose an absolute detection limit for cell counts, since it depends on microscope grid sizes, dilution factors, operator experience in avoiding false positives and operator patience. However, it appears that somewhere in the vicinity of 105 cells/mL [10], achieving highly accurate cell counts quickly becomes intractable without robotic autosampling and image analysis. Note that this quantification limit is higher than the 104 cells/mL detection limit determined by Cragg et al. [12]. Researchers working in seawater can overcome cell density limitations by filtering larger volumes of seawater. However, this is not possible in sediments because sedi-

6.1 Introduction

| 123

Fig. 6.1: Relationship between cell density in a sediment sample and the total number of grids that need to be counted to achieve a 95% confidence interval of ∼30% (solid lines) and the number of grids out of that total that will have no cells in them (dotted lines). Calculations are for dilutions of the original sediment of 1 : 20 (red), 1 : 10 (blue), and no dilution (black).

ments and cells are both retained on filters. It is possible to decrease detection limits by first extracting cells from sediments before concentrating them [13]. In this method, carbonates are dissolved with acid, and the cell and sediment slurry is mixed with a detergent to help detach cells from sediment particles, sonicated and/or vortexed and centrifuged through a cushion of Nycodenz to physically separate cells from sediment. Although yields from this method are less than 100%, they result in cell counts within a standard deviation of the mean of conventional total cell counts in marine sediments [13]. A new method that combines the optical removal of nontarget signals with physical removal of sediment grains during processing may make their enumeration more standardized and accurate [14, 15]. In this method, SYBR Green I is used as a DNA stain, because it is red-shifted when bound to sediment particles, as opposed to cells. Photographic subtraction of red-shifted signals leaves only cells to count, and greatly decreases the false positive detection [14]. As cells are being prepared for counting, surrounding sediments can be removed via treatment with strong acids [14] and multilevel density cell separation, resulting in an even cleaner signal [15]. Enumerating cells with general DNA-binding fluorophores, however, provides no information about the identities or functions of the cells. Based on DNA sequence

124 | 6 Quantifying microbes in the marine subseafloor: some notes of caution work, we know that the marine subsurface contains a diverse mixture of Bacteria and Archaea [16, 17]. However, almost all the cells visualized with a general DNA stain in the deep biosphere are small (< 1 μm) and coccoid [14] so it is impossible to use size or morphological characteristics to distinguish different phylogenetic groups.

6.2 Quantification of specific microbial groups in marine sediments The most commonly used methods to quantify particular microbial groups in marine sediments are fluorescent in situ hybridization (FISH), variations of FISH to improve signal-to-noise ratio, quantitative PCR (qPCR), RNA slot blot, deep 16S rRNA gene sequencing, metagenomic/metatranscriptomic sequencing and lipid analysis. In FISH, a DNA probe matching the primary sequence of the target taxon’s 16S rRNA molecules is attached to a fluorophore [18]. This diffuses into aldehyde-fixed cells (although gluteraldehyde protocols exist, most marine sedimentary researchers currently use formaldehyde) and binds to the ribosome. After washing to remove background staining, cells that contain the target 16S rRNA sequence can then be counted on an epifluorescence microscope. This method has the advantage that it targets living cells, since ribosomes degrade rapidly after cell death [19]. However, this relationship between fluorescence activity and cell activity makes FISH very difficult in the deep subsurface biosphere, and cell images require long exposure times of 0.5–5 seconds [20]. The low FISH fluorescence intensity is most likely due to low ribosome contents of the low activity cells in this environment. Methods to boost fluorescence intensity have therefore proved useful for enumerating cells of specific taxa in the deep marine subsurface. Most studies use Catalyzed Reporter Deposition FISH (CARD-FISH) [21]. This method is identical to FISH except that the DNA probe is not bound to a single fluorophore, but instead a horseradish peroxidase (HRP) enzyme which binds a long string of fluorescent tyramides to proteins inside the cells where the probe hybridizes. The key hurdle of CARD-FISH is permeabilizing the cell walls so that the ∼40 kDa enzyme can enter, while avoiding destroying the cell completely with the permeabilization procedures. Another method is called polynucleotide FISH, where multiply fluorescently-labeled RNA probes are used [22, 23], but use of this method in sediments has not been reported in the literature as having been used in the deep subsurface. In any of these methods, it is important that a cell fluoresce with the probe as well as a general DNA stain to avoid false positives. A major limitation to these FISH methods is that they only target 16S rRNA, which provides taxonomic, but not metabolic, physiological, or ecological information. Methods have been developed for binding probes to mRNA [24, 25], or to DNA [26], both of which open the possibility of quantifying cells containing genes of interest for a particular cellular function, not just taxonomy. However, these methods

6.2 Quantification of specific microbial groups in marine sediments | 125

seem to work best on highly active cells, and have such a low efficiency that they are not yet considered to be quantitative, although they promise to be very powerful for linking phylogeny to functions. Another drawback of direct cell counting methods is the high detection limits (󳶳 Fig. 6.1). Although cells at the domain level (i.e. Archaea and Bacteria) are often above the cell counting detection limit, even at deep subsurface sites, specific members of those communities may not be. The same cell separation techniques used for total cell counts [13, 15], may also prove useful for quantifying members of the community with cell abundances less than ∼106 cell/mL. Another solution is to use qPCR, which can detect much lower concentrations of cells in marine sediments than the direct cell counting methods. In qPCR, total DNA is extracted directly from sediments, cleaned and diluted to reduce the effects of coextracted PCR inhibitors [27]. This DNA then undergoes PCR with primers specific for the target of interest (usually 16S rRNA genes, which are often used as a taxonomic proxy) and the number of amplicons present at each PCR cycle are quantified relative to a standard through either SYBR Green or Taqman analyses. SYBR Green is a double-stranded DNA-specific fluorophore, so dimers of primers will also fluoresce if the primer concentration is too high and each qPCR run needs to be subjected to a DNA melt curve to detect the presence of primer-dimers. Taqman oligonucleotide probes fluoresce only when bound to a target sequence of the PCR amplicon. Taqman requires that an additional suitable probe site must be found that matches the target sequences well between the two primers, but it is better at avoiding false positives. In both cases, the resulting amplicons can be sequenced to make sure that only the target sequences were quantified [28, 29]. This is a distinct advantage over cell counting methods, which currently cannot be directly checked to make sure nontarget cells were absent from counts. qPCR can also be used to target only ribosome-containing (and therefore likely viable) cells through reverse transcription qPCR (RT-qPCR). Here, total RNA extracted directly from sediments is enzymatically reverse transcribed to cDNA primed with a single reverse primer, or random hexamers. Then, the cDNA undergoes normal qPCR. This can be performed on either rRNA or mRNA. The advantage of qPCR and RT-qPCR is that any gene or gene transcript with a known sequence can be targeted, so specific metabolic, physiological, or ecological functions can be assayed [30–32]. The challenge is linking the quantities of copies of genes or gene transcripts extracted with a low and highly variable efficiency from the sediment matrix to the number of cells present in situ. In RNA slot blot, otherwise referred to as a Southern blot [33], total extracted RNA is blotted onto a membrane next to RNA standards, hybridized to a radioactively-labeled oligonucleotide probe, and quantified by beta emissions from the membrane. This method is also limited to ribosomal sequences, so is best used for taxonomicbased quantifications, but it has an advantage over RT-qPCR in that it is a direct measurement of RNA quantity without amplification. Linking RNA quantified with RTqPCR or RNA slot blot to cell abundance of a particular group is not straightforward because cellular RNA content can vary with metabolic activity level [19]. So, while mul-

126 | 6 Quantifying microbes in the marine subseafloor: some notes of caution tiplying the RNA fraction of a particular taxonomic group by the total cell count provides a good initial estimate of the absolute cell quantity of that taxonomic group [34], it does not account for the possibility that many of the cells may have low rRNA content, making their fraction of total cells high even though the fraction of total rRNA is low, or vice versa. Deep sequencing of 16S rRNA gene sequences by next generation sequencing methods has also been used to infer the relative quantification of specific microbial groups in marine sediments (e.g. [35, 36]). To produce these large libraries with tens of thousands of individual sequence reads, DNA is first extracted from sediments and amplified by PCR. The argument that next generation sequencing, but not clone libraries (which are almost always < 1000 sequences), is relatively quantitative is that the high sampling depth ensures that the relative quantities of sequences are not skewed by an undersampling bias. Therefore, the ratio of amplicons at the end of PCR amplification can be determined fairly accurately. However, deep sequencing alone does not overcome any biases that happen during amplification. Usually samples go through at least 30 cycles of exponential amplification, and if the primers match one taxonomic group slightly more closely than other groups, the final ratio will be skewed toward that taxon relative to the ratio that existed in the natural samples. For instance, if one taxonomic group has a 1% greater chance than other groups of binding the primers (due to primer mismatches to the other groups, or other kinetic factors), then this group will have a 1.0130 or 35% increase in abundance relative to those other groups after 30 cycles of amplification. This bias can be minimized by choosing primers that are well-matched to an evolutionarily conserved region of the 16S rRNA gene to ensure that all taxa present in the sample have perfectly-matching primers. However, the natural variation in microbial populations means that it is often difficult to do this without introducing degenerate sites. For instance, if most taxonomic groups have the sequence AGGT at a particular region of the 16S rRNA gene, but a few taxonomic groups have AGGA at this region, then the operator can reduce primer bias by introducing a degeneracy at the fourth position, with AGGW, where W stands for the “weakly” bonding Watson-Crick bases A or T. This means that the PCR amplification now has a mixture of two populations of primers, AGGT and AGGA, presumably in equimolar quantities. However, imagine that in these particular sediments, taxa with the AGGT version in their 16S rRNA genes outnumber those with the AGGA version 99 : 1. Since the ratio of degenerate primers is 50 : 50, each cell with AGGA will have an effective concentration of primers relative to cells 100-fold greater than AGGT, and therefore a 100 times greater chance of binding to its primer. Even in a situation where every sequence in the natural population has a perfectly-matching primer pair, the resulting amplicons will therefore be skewed towards the less abundant organism, by an amount dependent upon the number of cycles of exponential amplification used. For this reason, it seems that relative quantification with highly amplified 16S rRNA amplicon libraries should be viewed with caution.

6.2 Quantification of specific microbial groups in marine sediments | 127

Metagenomes and metatranscriptomes offer an opportunity for relative quantifications without primer-dependent amplification [37–39]. Here, DNA (metagenomes) or cDNA reverse-transcribed from RNA (metatranscriptomes) extracted from marine sediments is subjected to high-throughput next generation sequencing where, theoretically, all parts of each cell’s genome can be amplified. Microbial groups can then be quantified relative to other groups by the phylogenetic binning of 16S rRNA gene sequences that appear in the metagenome. Alternately, the relative abundance of a particular microbial group can be assumed based on annotations of the whole metagenomic material. The first method shows promise for the relative quantification of microbial groups in marine sediments, since the uncultured microbial groups prevalent in marine sediments are well-represented in 16S rRNA gene databases. Here, the main concern is making sure that the sequencing depth is great enough to allow for a good representation of 16S genes alongside all the other parts of the genome being sequenced. Using the rest of the genomic information to infer relative abundance of taxonomic groups, however, is hampered by the fact that genomic databases are currently dominated by cultured organisms. The ability to obtain genomic information from single uncultured microbes has only recently become available [40], and the representation among marine sediment organisms is currently very low relative to the 16S rRNA gene databases [41, 42]. Therefore, any taxonomic identification through whole metagenomic or transcriptomic data will be skewed towards cultured bacteria, which comprise 97% of current annotated genomes [43]. As with tag pyrosequencing, these methods do not promise absolute quantifications, only relative quantifications. Different classes of lipids can be both quantitatively and taxonomically informative [44]. This makes them promising quantification methods for particular taxonomic groups. However, the taxonomic resolution provided is not as high as that of individual gene or ribosome sequences. Also, individual microbial cells can differ in their lipid contents [45], which makes it difficult to convert lipid concentrations to cell concentrations, in the same way that it is difficult to convert RNA quantifications to microbial cell numbers. Another obstacle to using lipids as quantitative biomarkers is that, as with DNA, it can be difficult to know whether degraded cell material is included in the quantification. Previously, it was thought that lipids with intact labile polar head groups were representative only of living cells [20, 46], but models and experiments both support the idea that a significant proportion of the intact polar lipids measured in marine sediments represent dead cells or cell detritus [47, 48]. This makes it difficult to know what quantity of DNA or lipids represents living cells vs. extracellular necromass.

128 | 6 Quantifying microbes in the marine subseafloor: some notes of caution

6.3 Assessment of quantitative methods in marine sediments: the Leg 201 Peru Margin example Assessment of the absolute accuracies of these methods is difficult, since spiking exogenous cells into natural sediment cannot perfectly mimic the variety of cells or the variety of cell-biofilm interactions in the natural community. Therefore, there is no commonly accepted best-practice method for quantifying specific microbial taxa in marine sediments. But assessment of quantification methods is warranted because they have produced conflicting results for the Ocean Drilling Program (ODP) Leg 201 Peru Margin cores. Leg 201 was the first ODP cruise dedicated to microbiology, so there was high international interest, and different labs quantified archaea and bacteria on subsamples of the same cores. As quantifications from these samples began to be published, researchers were surprised to find stark differences between ratios of archaea:bacteria measured by different lab groups using different methods on the same cores. These discrepancies have lead to a plethora of new research and interest in both quantitative methods and the deep marine subsurface [16, 20, 38, 47–57]. Explanations have been proposed to explain some of these discrepancies, including primer bias in qPCR measurements [7] and inclusion of necromass in lipid measurements [47, 48]. However, some items, such as the disparate FISH and CARD-FISH measurements, or the ability of qPCR to be absolutely quantitative, deserve further attention. Instead of focusing on the ratio of bacteria and archaea here, as has been done many times in the literature cited above, I will compare the absolute quantities of bacteria and archaea measured with FISH, CARD-FISH and qPCR in these sediments (󳶳 Fig. 6.2). I will first consider the cell count-based methods (FISH and CARD-FISH) and then the qPCR. Site 1229 had the most complete data coverage. Here, most of the bacterial FISH and CARD-FISH cell counts agree within an order of magnitude, suggesting that these may be good methods for quantifying bacteria (󳶳 Fig. 6.2 (b)). However, Schippers et al. [16] show two dips of over an order in magnitude in bacteria at 30 mbsf and 90 mbsf, the shallower of which was not replicated in the Biddle et al. [20] data, even though they measured bacteria at 30 mbsf. Mauclaire et al. [88] did not measure bacteria at 30 mbsf and 90 mbsf. At sites 1227 and 1230, the bacterial cell counts from the different lab groups appear to disagree quite a bit. However, this may be an artifact of low data resolution in Biddle et al. [20]. Mauclaire et al. [88] did not analyze these cores. With the exception of 30 mbsf, bacterial cell counts from the different lab groups tend to agree for 1229, lending support to the idea that these methods are close to quantitative. The archaeal quantifications, on the other hand, show little agreement between Schippers et al. [16] and the other two studies (󳶳 Fig. 6.2 (d)–(f)). Schippers et al. [16] found that archaea were below the CARD-FISH detection limit, negligible compared to bacteria or cell counts. Conversely, both Biddle et al. [20] and Mauclaire et al. 2004 [88]

6.3 Assessment of quantitative methods in marine sediments | 129

Fig. 6.2: qPCR, FISH and CARD-FISH quantifications made by different labs using different methods in Leg 201 Peru Margin sediments from sites (a) and (d) 1227, (b) and (e) 1229, and (c) and (f) 1230, (a)–(c) are bacterial quantifications, and (d)–(f) are archaeal quantifications. Note the different scales of bacteria and archaea. Data from [16, 20, 51, 88] and BDL is below detection limit.

130 | 6 Quantifying microbes in the marine subseafloor: some notes of caution found archaeal cells to be within an order of magnitude of bacterial cells, using FISH and CARD-FISH. Furthermore, the archaeal counts from these two studies are within an order of magnitude of each other for 1229, although few depths were measured by both groups, so the comparison is limited (󳶳 Fig. 6.2 (e)). The good agreement between the two groups’ assessments of archaea means that their methods may be better at quantifying archaea than those of Schippers et al. [16], a subject I will consider in greater depth below. The qPCR values for bacteria and archaea at all sites are more variable than FISH and CARD-FISH counts, oscillating by as much as three orders of magnitude in adjacent sediment depths (󳶳 Fig. 6.2). The dips in bacterial cells in 1229 are not replicated in the qPCR data, however, if the qPCR variability is due to random error, then such details would be swamped by the noise. Theoretically, qPCR quantifications should be offset from cell counts by a factor of the number of 16S rRNA gene copies per genome, but it is unlikely that this high variation results from intersample variation of many orders of magnitude of 16S rRNA gene copies per genome. Another possibility is that qPCR values include DNA that was extracellular, from dead cells. Both of these factors would make qPCR values higher than cell counts. However, for sites 1227 and 1229, bacterial qPCR values oscillate roughly about a mean of 1 : 1 with FISH and CARD-FISH counts, suggesting that genomic 16S rRNA gene copy numbers and extracellular DNA are not responsible for the offset from bacterial cell counts. Site 1230 has consistently higher bacterial qPCR values than bacterial cell counts, but here, copy numbers and extracellular DNA would have to account to a 10–1000 fold increase in qPCR over cell counts. This is untenable for genomic 16S rRNA copy numbers which tend to be fewer than five [43], but extracellular DNA concentrations are less well-constrained, so it is possible that this explains the offset at 1230 (but not at 1227 or 1229, which have no such offset). For archaea, the qPCR values, although variable, agree with the Schippers et al. [16] CARD-FISH conclusions that archaea are below detection limit. Another research group also found that archaea were negligible using qPCR, although they did not report absolute values for either bacteria or archaea [57]. However, the Schippers et al. [16] and [51] archaeal qPCR values are often 2–3 orders of magnitude lower than the archaeal cells counted by Biddle et al. [20] and Mauclaire et al. [88]. Therefore, it is likely that either Schippers et al. [16] and [51] missed archaea, or Biddle et al. [20] and Mauclaire et al. [88] over-counted archaea. One way to determine which of these studies had the more accurate methods is to compare the sums of bacteria and archaea to total prokaryotic cell counts, measured with acridine orange direct counts (AODC), a method for quantifying total cells. The more accurate methods should get closer to replicating AODC counts. An important caveat is that, since FISH methods target only cells with ribosomes, the theoretical presence of large quantities of cells that have died recently, so they have intact cell membranes and genomic DNA, but fully degraded RNA, will cause the AODC counts to be much higher than the sum of bacterial and archaeal FISH or CARD-FISH counts.

6.3 Assessment of quantitative methods in marine sediments | 131

Fig. 6.3: Comparison of sums of bacteria plus archaea measured with qPCR, FISH, or CARD-FISH, to total cell counts with acridine orange direct counts for Leg 201 Peru Margin sites (a) 1227, (b) 1229, and (c) 1230. Gray bars indicate the sulfate depletion depths; these coincide with increases in methane, making them Sulfate Methane Transition Zones (SMTZs). Data from [16, 20, 51, 88].

The Biddle et al. [20] and Mauclaire et al. [88] FISH and CARD-FISH counts are clearly closer to AODC counts than the Schippers et al. [16] CARD-FISH counts (󳶳 Fig. 6.3), with the exception of the Biddle data from 1229. This is due to the fact that archaea were found to be above the detection limit in those studies, indicating that Schippers et al. [16] may have under-counted archaea. It is also apparent that the dips in the Schippers et al. [16] bacterial CARD-FISH data correspond to positive excursions in AODC counts, and the sulfate methane transition zones (SMTZs), where archaea may outnumber bacteria [2]. This is further evidence that the Schippers et al. [16] methods are likely to have missed archaea that were present. Summing the bacterial and archaeal qPCR counts does little to decrease the high variability (󳶳 Fig. 6.3). In general, qPCR sums are much lower than total cell counts.

132 | 6 Quantifying microbes in the marine subseafloor: some notes of caution Other types of quantitative methods have also found archaea to be abundant in the Peru Margin samples. Lipid [20, 46, 48], metagenomic [38], metatranscriptomic [39], and other qPCR [38] analyses have all shown that archaea comprise a nonnegligible proportion of the microbial community. Therefore, it appears that Schippers et al. [16] and [51], and Inagaki et al. [57] severely under-counted archaea. If this is true, then some aspect of their methods should account for this. In these studies, archaea were permeabilized for CARD-FISH using lysozyme, which targets a molecule not known to be present in archaea, peptidoglycan [59]. Other studies have found lysozyme to be inadequate for archaeal cell wall permeabilization when compared to other treatments on the same samples [60, 61]. This suggests that archaea were below the CARD-FISH detection limit in Schippers et al. [16] because their cell walls were never permeabilized adequately to allow the HRP enzyme to enter. For qPCR, all three studies used Taqman technology, with ARCH516 [62] as the Taqman probe. The DNA sequence from ARCH516 has been shown to be a poor match to the archaeal groups commonly found in anoxic marine sediments [7]. Therefore, it is not surprising that these studies under-counted archaea, since the methods used were unlikely to target the archaeal populations present. I will show below that these methods are also associated with under-counted archaea and low yields relative to total cell counts in the larger canon of literature.

6.4 Global meta-analysis of FISH, CARD-FISH and qPCR quantifications of bacteria and archaea As of today, the Leg 201 Peru Margin sediments offer the only opportunity to compare FISH, CARD-FISH, qPCR, lipids, metagenomes and metatranscriptomes measured by many different international research groups on the same samples. However, much can be learned by extending this examination to all currently published quantifications of bacteria and archaea in different marine sediments. We recently compared the two most prevalent methods for quantifying bacteria and archaea in marine sediments – qPCR and FISH/CARD-FISH – in marine sediments and seawater [63]. In seawater samples, we found that there was a very tight 1:1 correlation between the sum of bacteria and archaea quantified with CARD-FISH and the total prokaryotic cells visualized using a general DNA fluorophore. Interestingly, few of these seawater studies used lysozyme as a permeabilization procedure for archaea (only 1 out of 18 studies). Most studies in marine sediments, on the other hand, used lysozyme as a permeabilization procedure for archaea in CARD-FISH, and this environment had low yields of bacteria plus archaea relative to total cell counts as well as low fractions of archaea. This result mirrors the observations above with the Leg 201 Peru Margin cores. We found that studies with CARD-FISH using proteinase K or detergents to permeabilize archaeal membranes were able to replicate the high yields and high archaeal fractions found

6.4 Global meta-analysis of FISH, CARD-FISH and qPCR quantifications ...

|

133

in seawater samples. This is an encouraging result. Marine sediments present a difficult matrix in which to perform visual cell counts, but the results of our meta-analysis suggest that, in high biomass sediments, quantification in sediments can be as accurate as quantifications in seawater. FISH had highly variable yields, possibly due to the difficulty in seeing signals from cells with low metabolic activities [21, 64]. It is important to note that most of the sediments in this meta-analysis were organic-rich, anoxic, marine sediments with low permeability and constant water inundation. We found 11 studies from highly permeable intertidal sands. In each of these, no matter what the quantification method, very few archaea were found, yet yields relative to total cell counts were high. This suggests that these environments are truly dominated by bacteria, and, although archaea may be present, they are in much lower abundance than bacteria. Furthermore, we found that qPCR measurements are unlikely to be absolutely quantitative, since they result in cell estimates that are highly variable relative to total cell counts [63]. This is analogous to the results of the Leg 201 samples as well (󳶳 Fig. 6.2). A high variability of qPCR quantifications is predicted by the high variability of extraction efficiencies even in repeated extractions from identical subsamples [65]. Therefore, the losses during DNA extraction make it difficult to extrapolate qPCR measurements to absolute values in marine sediments. However, qPCR does appear to provide reliable relative quantifications when compared to measurements made with other primers on the same DNA extraction [28, 29, 63]. We found that the coverage of the probes or primers used for any of the quantitative methods must be well-matched to the organisms present in a given sample. As with the Leg 201 samples, ARCH516 has often been used as a Taqman probe or SYBR qPCR primer in deep subsurface samples, but we found it to be systematically associated with low archaeal concentrations, as is expected by its poor match to the types of archaea found in organic-rich marine sediments [7]. Very different archaeal communities can be found in oligotrophic open ocean sites, such as North Pond in the Atlantic [66], and for these communities ARCH516 is a perfect match [63]. Accordingly, ARCH516 results in high archaeal numbers in this environment [53]. FISH and CARD-FISH have the added consideration that not only must the sequence of the probes match the target population, but also it must bind to an intact ribosome. This ribosome may have an intricate secondary structure as well as attached ribosomal proteins blocking probe hybridization. Currently, all published FISH and CARD-FISH studies from marine sediments use probes ARCH915 [67] for archaea and EUB338 [68] or EUB338I-III [69] for bacteria. These probes seem to be good matches to target populations’ sequences and they also appear to reach an accessible portion of the ribosomes. This is supported by the fact that the sum of cell counts made with these two probes nearly perfectly replicates total cell counts in seawater, and in some marine sediment studies [63]. Probes that target other sections of the ribosome, however, may not bind as well. For instance, probes for an uncultured archaeal group worked well on ribosomal sequences from White Oak River sediments that had been inserted into plasmids and expressed

134 | 6 Quantifying microbes in the marine subseafloor: some notes of caution in E. coli in a method called Clone-FISH [29, 70]. However, only one of these probes worked on properly folded ribosomes with the same sequences in a native population [29]. Given the high heterogeneity of good probe binding sites along a ribosomal sequence [71, 72], expansion of FISH methods to specific sub-Domain level populations should be approached with caution. Poor probe binding to folded ribosomes can also plague RT-qPCR measurements if rRNA is not denatured before reverse transcription. Extracted rRNA must first be heat-denatured immediately before the reverse transcription step [73]. Without this denaturation step, the number of rRNA copies may be underestimated.

6.5 Future outlook The number of studies quantifying bacteria and archaea in marine sediments is growing, but much of the sediments underlying Earth’s oceans are yet to be explored (󳶳 Fig. 6.4). In particular, taxon-specific quantifications in oligotrophic deep-sea sediments have only been made in two locations, the Eastern Equatorial Pacific, and North Pond in the Atlantic [16, 53]. Despite being carbon-starved, these ecosystems seem to contain a diverse and functional microbial ecosystem [17, 66]. The importance of such locations is shown by the fact that incorporating total cell counts from those areas resulted in a significant revision of the total size of the microbial biosphere [1]. In general, areas far from land have few quantitative measurements, due to infrequent access to fixed sediment samples obtained with sterile techniques from such locations. Spe-

FISH methods shallow sediments qPCR shallow sediments RNA slot blot shallow sediments

> 1 mbsf > 1 mbsf > 1 mbsf

Fig. 6.4: Approximate locations of quantifications of total archaea and bacteria in marine sediments. Data from: [16, 20, 29, 34, 36, 49–51, 53–55, 57, 76–88, 90–99, 99–107].

6.5 Future outlook

|

135

cific geographic locations with no bacterial and archaeal sedimentary quantifications are Australia, the Pacific Islands, Antarctica, Greenland and the Indian Ocean (except for two studies on the Sumatra Margin [50, 55]). As quantitative methods become more routine and the scientific community comes to a consensus for which methods work best, hopefully this analysis can become standard for every Integrated Ocean Drilling Program expedition. Incorporating such measurements into routine operations will greatly increase the coverage of such measurements and allow for a more nuanced view of the roles and activities of bacteria and archaea in marine sediments. One of the nuances that deserves further study is the possibility that organicrich sediments tend to have an increasing contribution from archaea with depth [63], and organic-poor sediments may have the opposite trend (󳶳 Fig. 6.5). The consensus among studies in organic-rich marine sediments (13 out of 16 cores) is that bacteria dominate surface sediments, but die off more quickly than archaea over time

Fig. 6.5: Increase in the fraction of bacteria with depth in all six cores from oligotrophic environments. NP means North Pond, and data are qPCR-measured bacteria divided by the sum of qPCR-measured bacteria and archaea [53]. EEP means Eastern Equatorial Pacific, and data are CARDFISH measured bacteria divided by total cell counts [16]. Lines are linear regressions with equations on regression fits listed below the figure in matching colors.

136 | 6 Quantifying microbes in the marine subseafloor: some notes of caution and sedimentation, leaving higher archaeal proportions with depth [63]. In the five cores that have been quantified from organic-poor sites, archaea dominate at the surface, but die off more quickly than bacteria, leaving higher bacterial proportions at depth (󳶳 Fig. 6.5). If these trends continue to be upheld as more study sites are included, then this could be very informative of the nature of starvation and stress in this globally-relevant ecosystem. Organic-rich marine subsurface environments are chemically-reducing systems conducive to the presence of strict anaerobes, but electron acceptors are limited, making communities dependent on low energy fermentation reactions [2]. Organic-poor marine subsurface environments are replete with electron acceptors – some sediments even have dissolved oxygen present many meters into the sediment [74, 75]. Here, energy limitation most likely comes from the limitation of electron donors, and the need to devote precious resources to avoiding oxidative damage. Something intrinsic to the nature of bacteria and archaea may better prepare each one for the very different challenges of these two broad environmental types. Methods to accurately and precisely quantify marine subseafloor microbes based on their taxonomic identities and predicted cellular functions are growing. Many innovative method improvements have helped to overcome the difficulties of quantifying cells with low metabolic activity in a challenging sediment matrix. As the database of microbial quantifications increases, patterns are beginning to emerge showing how life functions in this energy-limited environment.

References [1]

[2]

[3]

[4] [5]

[6] [7]

Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16 213– 16 216. D’Hondt S, Jørgensen BB, Miller DJ, Batzke A, Blake R, Cragg BA, Cypionka H, Dickens GR, Ferdelman T, Hinrichs K, Holm NG, Mitterer R, Spivack A, Wang G, Bekins B, Engelen B, Ford K, Gettemy G, Rutherford SD, Sass H, Skilbeck CG, House CH, Aiello IW, Gue G, Inagaki F, Meister P, Naehr T, Niitsuma S, Parkes RJ, Schippers A, Smith DC, Teske A, Wiegel J, Padilla CN, Luz J, Acosta S. Distributions of microbial activities in deep subseafloor sediments. Science 80 (2004), 2216–2221. Parkes RJ, Webster G, Cragg BA, Weightman AJ, Newberry CJ, Ferdelman TG, Kallmeyer J, Jørgensen BB, Aiello IW, Fry JC. Deep subseafloor prokaryotes stimulated at interfaces over geological time. Nature 436 (2005), 390–394. Biddle JF, House CH, Brenchley JE. Microbial stratification in deeply buried marine sediment reflects changes in sulfate/methane profiles. Geobiology 3 (2006), 287–295. Batzke A, Engelen B, Sass H, Cypionka H. Phylogenetic and physiological diversity of cultured deep-biosphere bacteria from Equatorial Pacific Ocean and Peru Margin sediments. Geomicrobiol J 24 (2007), 261–273. Rappé MS, Giovannoni SJ. The uncultured microbial majority. Annu Rev Microbiol 57 (2003), 369–394. Teske A, Sørensen KB. Uncultured archaea in deep marine subsurface sediments: have we caught them all? ISME J 2 (2008), 3–18.

References | 137

[8] [9] [10]

[11] [12]

[13] [14] [15]

[16]

[17]

[18]

[19] [20]

[21]

[22]

[23]

[24] [25]

Hoehler TM, Jørgensen BB. Microbial life under extreme energy limitation. Nat Rev Microbiol 11 (2013), 83–94. Fry JC. Direct Methods and Biomass Estimation, p. 41–86. In: Grigorova, R, Norris, JR (eds.), Methods in Microbiology. Academic Press, 1990. Kallmeyer J. Detection and quantification of microbial cells in subsurface sediments, p. 79– 103. In: Laskin, A, Gadd, G, Sariaslani, S (eds.), Advances in Applied Microbiology, 1st ed. Elsevier Inc. Academic Press, Burlington, 2011. Cragg BA, Kemp AES. Bacterial profiles in deep sediment layers from the Eastern Equatorial Pacific Ocean, site 851. Proc. Ocean Drill. Program, Sci Results 138 (1995), 599–604. Cragg BA, Law KM, Sullivan GMO, Parkes RJ. Bacterial profiles in deep sediments of the Alboran Sea, Western Mediterranean, sites 976–978. Proc. Ocean Drill. Program, Sci Results 161 (1999), 433–438. Kallmeyer J, Smith DC, Spivack AJ, Hondt SD. New cell extraction procedure applied to deep subsurface sediments. Limnol. Oceanogr. Methods 6 (2008), 236–245. Morono Y, Terada T, Masui N, Inagaki F. Discriminative detection and enumeration of microbial life in marine subsurface sediments. ISME J 3 (2009), 503–511. Morono Y, Terada T, Kallmeyer J, Inagaki F. An Improved Cell Separation Technique for Marine Subsurface Sediments: Applications for High-throughput Analysis Using Flow Cytometry and Cell Sorting. Environ Microbiol 15 (2013), 2841–2849. Schippers A, Neretin LN, Kallmeyer J, Ferdelman TG, Cragg BA, Parkes RJ, Jørgensen BB. Prokaryotic cells of the deep subseafloor biosphere identified as living bacteria. Nature 61 (2005), 861–864. Sorensen KB, Lauer A, Teske A. Archaeal phylotypes in a metal-rich and low-activity deep subsurface sediment of the Peru Basin, ODP Leg 201, Site 1231. Geobiology 2 (2004), 151– 161. Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereux R, Stahl DA. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl Environ Microbiol 56 (1990), 1919–1925. Davis BD, Luger SM, Tai PC. Role of ribosome degradation in the death of starved Escherichia coli cells. J Bacteriol 166 (1986), 439–445. Biddle JF, Lipp JS, Lever MA, Lloyd KG, Sørensen KB, Anderson R, Fredricks HF, Elvert M, Kelly TJ, Schrag DP, Sogin ML, Brenchley JE, Teske A, House CH, Hinrichs K-U. Heterotrophic archaea dominate sedimentary subsurface ecosystems off Peru. Proc Natl Acad Sci USA 103 (2006), 3846–3851. Pernthaler A, Pernthaler J, Amann R. Fluorescence In Situ Hybridization and catalyzed reporter deposition for the identification of marine bacteria. Appl Environ Microbiol 68 (2002), 3094–3101. Pernthaler A, Preston CM, Pernthaler J, DeLong EF, Amann R. Comparison of fluorescently labeled oligonucleotide and polynucleotide probes for the detection of pelagic marine bacteria and archaea. Appl Environ Microbiol 68 (2002), 661–667. Delong EF, Taylor LT, Marsh TL, Preston CM. Visualization and enumeration of marine planktonic archaea and bacteria by using polyribonucleotide probes and Fluorescent In Situ Hybridization. Appl Environ Microbiol 65 (1999), 5554–5563. Pernthaler A, Amann R. Simultaneous Fluorescence In Situ Hybridization of mRNA and rRNA in Environmental Bacteria. Society 70 (2004), 5426–5433. Kofoed MVW, Nielsen DÅ, Revsbech NP, Schramm A. Fluorescence in situ hybridization (FISH) detection of nitrite reductase transcripts (nirS mRNA) in Pseudomonas stutzeri biofilms relative to a microscale oxygen gradient. Syst Appl Microbiol 35 (2012), 513–517.

138 | 6 Quantifying microbes in the marine subseafloor: some notes of caution [26]

[27]

[28]

[29]

[30]

[31]

[32]

[33] [34]

[35]

[36]

[37]

[38]

[39] [40] [41]

Hoshino T, Schramm A. Detection of denitrification genes by in situ rolling circle amplification-fluorescence in situ hybridization to link metabolic potential with identity inside bacterial cells. Environ Microbiol 12 (2010), 2508–2517. Lloyd KG, Macgregor BJ, Teske A. Quantitative PCR methods for RNA and DNA in marine sediments: maximizing yield while overcoming inhibition. FEMS Microbiol Ecol 72 (2009), 143– 151. Lloyd KG, Alperin MJ, Teske A. Environmental evidence for net methane production and oxidation in putative ANaerobic MEthanotrophic (ANME) archaea. Environ Microbiol 13 (2011), 2548–2564. Kubo K, Lloyd KG, F Biddle J, Amann R, Teske A, Knittel K. Archaea of the Miscellaneous Crenarchaeotal Group are abundant, diverse and widespread in marine sediments. ISME J 6 (2012), 1949–1965. Nunoura T, Oida H, Miyazaki J, Miyashita A, Imachi H, Takai K. Quantification of mcrA by fluorescent PCR in methanogenic and methanotrophic microbial communities. FEMS Microbiol Ecol 64 (2008), 240–247. Neretin LN, Schippers A, Hamann K, Amann R, Jørgensen BB. Quantification of dissimilatory (bi) sulfite reductase gene expression in Desulfobacterium autotrophicum using real-time RT-PCR. Environ Microbiol 5 (2003), 660–671. Inagaki F, Tsunogai U, Suzuki M, Kosaka A, Machiyama H, Takai K, Nunoura T, Nealson KH, Horikoshi K. Characterization of C1-metabolizing prokaryotic communities in methane seep habitats at the Kuroshima Knoll, Southern Ryukyu Arc, byanalyzing pmoA, mmoA, mxaF, mcrA and 16S rRNA Genes. Appl Environ Microbiol 70 (2004), 7445–7455. Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98 (1975), 503–517. Sahm K, MacGregor BJ, Jørgensen BB, Stahl DA. Sulfate reduction and vertical distribution of sulfate-reducing bacteria quantified by rRNA slot-blot hybridization in a coastal marine sediment. Environ Microbiol 1 (1999), 65–74. Gobet A, Böer SI, Huse SM, van Beusekom JEE, Quince C, Sogin ML, Boetius A, Ramette A. Diversity and dynamics of rare and of resident bacterial populations in coastal sands. ISME J 6 (2012), 542–553. Jørgensen SL, Hannisdal B, Lanzén A, Baumberger T, Flesland K, Fonseca R, Ovreas L, Steen IH, Thorseth IH, Pedersen RB, Schleper C. Correlating microbial community profiles with geochemical data in highly stratified sediments from the Arctic Mid-Ocean Ridge. Proc Natl Acad Sci USA 109, (2012), E2846–E2855. Biddle JF, White JR, Teske AP, House CH. Metagenomics of the subsurface Brazos-Trinity Basin (IODP site 1320): comparison with other sediment and pyrosequenced metagenomes. ISME J 5 (2011), 1038–1047. Biddle JF, Fitz-Gibbon ST, Schuster SC, Brenchley JE, House CH. Metagenomic signatures of the Peru Margin subseafoor biosphere show a genetically distinct environmnet. Proc Natl Acad Sci USA 105 (2008), 10 583–10 588. Orsi WD, Edgcomb VP, Christman GD, Biddle JF. Gene expression in the deep biosphere. Nature 499 (2013), 205–208. Stepanauskas R, Sieracki ME. Matching phylogeny and metabolism in the uncultured marine bacteria, one cell at a time. PNAS 104 (2007), 9052–9057. Lloyd KG, Schreiber L, Petersen DG, Kjeldsen KU, Lever M a, Steen AD, Stepanauskas R, Richter M, Kleindienst S, Lenk S, Schramm A, Jørgensen BB. Predominant archaea in marine sediments degrade detrital proteins. Nature 496 (2013), 215–218.

References | 139

[42]

[43] [44] [45]

[46] [47]

[48]

[49]

[50] [51]

[52]

[53]

[54]

[55]

[56] [57]

Rinke C, Schwientek P, Sczyrba A, Ivanova NN, Anderson IJ, Cheng J-F, Darling A, Malfatti S, Swan BK, Gies E a, Dodsworth J a, Hedlund BP, Tsiamis G, Sievert SM, Liu W-T, Eisen J a, Hallam SJ, Kyrpides NC, Stepanauskas R, Rubin EM, Hugenholtz P, Woyke T. Insights into the phylogeny and coding potential of microbial dark matter. Nature 499 (2013), 431–437. Markowitz VM, Korzeniewski F, Palaniappan K, Szeto E, Werner G, Padki A. The integrated microbial genomes (IMG) system. Nucleic Acids Res 34 (2006), D344–D348. Schouten S, Hopmans EC, Sinninghe Damsté JS. The organic geochemistry of glycerol dialkyl glycerol tetraether lipids: A review. Org Geochem 54 (2013), 19–61. Huguet C, Urakawa H, Martens-Habbena W, Truxal L, Stahl DA, Ingalls AE. Changes in intact membrane lipid content of archaeal cells as an indication of metabolic status. Org Geochem 41 (2010), 930–934. Lipp JS, Morono Y, Inagaki F, Hinrichs K. Significant contribution of Archaea to extant biomass in marine subsurface sediments. Nature 454 (2008), 991–994. Schouten S, Middelburg JJ, Hopmans EC, Sinninghe Damsté JS. Fossilization and degradation of intact polar lipids in deep subsurface sediments: A theoretical approach. Geochim Cosmochim Acta 74 (2010), 3806–3814. Xie S, Lipp JS, Wegener G, Ferdelman TG, Hinrichs K. Turnover of microbial lipids in the deep biosphere and growth of benthic archaeal populations. Proc Natl Acad Sci USA 110 (2013), 6010–6014. Schippers A, Kock D, Höft C, Köweker G, Siegert M. Quantification of microbial communities in subsurface marine sediments of the Black Sea and off Namibia. Front Microbiol 3 (2012), 16. Schippers A, Köweker G, Höft C, Teichert BMA. Quantification of microbial communities in forearc sediment basins off Sumatra. Geomicrobiol J 27 (2010), 170–182. Schippers A, Neretin LN. Quantification of microbial communities in near-surface and deeply buried marine sediments on the Peru continental margin using real-time PCR. Environ Microbiol 8 (2006), 1251–1260. Blazejak A, Schippers A. Real-Time PCR Quantification and Diversity Analysis of the Functional Genes aprA and dsrA of Sulfate-Reducing Prokaryotes in Marine Sediments of the Peru Continental Margin and the Black Sea. Front Microbiol 2 (2011), 253. Breuker A, Schippers A. Data report: total cell counts and qPCR abundance of Archaea and Bacteria in shallow subsurface marine sediments of North Pond: gravity cores collected during site survey cruise prior to IODP Expedition 336. Proc Integr Ocean Drill Progr 336 (2013), 1–7. Webster G, Blazejak A, Cragg B a, Schippers A, Sass H, Rinna J, Tang X, Mathes F, Ferdelman TG, Fry JC, Weightman AJ, Parkes RJ. Subsurface microbiology and biogeochemistry of a deep, cold-water carbonate mound from the Porcupine Seabight (IODP Expedition 307). Environ Microbiol 11 (2009), 239–257. Siegert M, Krüger M, Teichert B, Wiedicke M, Schippers A. Anaerobic Oxidation of Methane at a Marine Methane Seep in a Forearc Sediment Basin off Sumatra, Indian Ocean. Front Microbiol 2 (2011), 249. Jørgensen BB, Boetius A. Feast and famine–microbial life in the deep-sea bed. Nat Rev Microbiol 5 (2007), 770–781. Inagaki F, Nunoura T, Nakagawa S, Teske A, Lever M, Lauer A, Suzuki M, Takai K, Delwiche M, Colwell FS, Nealson KH, Horikoshi K, D’Hondt S, Jørgensen BB. Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments on the Pacific Ocean Margin. Proc Natl Acad Sci USA 103 (2006), 2815–2820.

140 | 6 Quantifying microbes in the marine subseafloor: some notes of caution [58]

[59] [60]

[61] [62]

[63]

[64]

[65]

[66]

[67] [68]

[69]

[70]

[71]

[72]

[73] [74]

Biddle JF, Fitz-Gibbon S, Schuster SC, Brenchley JE, House CH. Metagenomic signatures of the Peru Margin subseafloor biosphere show a genetically distinct environment. Proc Natl Acad Sci USA 105 (2008), 10 583–10 588. Visweswaran GRR, Dijkstra BW, Kok J. Murein and pseudomurein cell wall binding domains of bacteria and archaea–a comparative view. Appl Microbiol Biotechnol 92 (2011), 921–928. Teira E, Reinthaler T, Pernthaler A, Pernthaler J, Herndl GJ. Combining Catalyzed Reporter Deposition-Fluorescence In Situ Hybridization and microautoradiography to detect substrate utilization by bacteria and archaea in the deep ocean. Appl Environ Microbiol 70 (2004), 4411–4414. Molari M, Manini E. Reliability of CARD-FISH procedure for enumeration of Archaea in deepsea surficial sediments. Curr Microbiol 64 (2012), 242–250. Takai K, Horikoshi K. Rapid detection and quantification of members of the archaeal community by quantitative PCR using fluorogenic probes. Appl Environ Microbiol 66 (2000), 5066– 5072. Lloyd KG, May MK, Kevorkian R, Steen AD. Meta-analysis of quantification methods shows that archaea and bacteria have similar abundances in the subseafloor. Appl Environ Microbiol 79 (2013), 7790–7799. Bouvier T, Del Giorgio PA. Factors influencing the detection of bacterial cells using fluorescence in situ hybridization (FISH): A quantitative review of published reports. FEMS Microbiol Ecol 44 (2003), 3–15. Mumy KL, Findlay RH. Convenient determination of DNA extraction efficiency using an external DNA recovery standard and quantitative-competitive PCR. J Microbiol Methods 57 (2004), 259–268. Durbin AM, Teske A. Archaea in organic-lean and organic-rich marine subsurface sediments: an environmental gradient reflected in distinct phylogenetic lineages. Front Microbiol 3 (2012), 168. DeLong EF. Archaea in coastal marine environments. Proc Natl Acad Sci USA 89 (1992), 5685– 5689. Amann RI, Krumholz L, Stahl DA. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic and environmental studies in microbiology. J Bacteriol 172 (1990), 762–770. Daims H, Bruhl A, Amann R, Schleifer K-H, Wagner M. The domain-specific probe EUB338 is insufficient for the detection of all Bacteria: Development and evaluation of a more comprehensive probe set. Syst Appl Microbiol 22 (1999), 434–444. Schramm A, Fuchs BM, Nielsen JL, Tonolla M, Stahl DA. Fluorescence in situ hybridization of 16S rRNA gene clones (Clone-FISH) for probe validation and screening of clone libraries. Environ Microbiol 4 (2002), 713–720. Behrens S, Fuchs BM, Mueller F, Amann R. Is the In Situ Accessibility of the 16S rRNA of Escherichia coli for Cy3-Labeled Oligonucleotide Probes Predicted by a Three-Dimensional Structure Model of the 30S Ribosomal Subunit? Appl Environ Microbiol 69 (2003), 4935– 4941. Behrens S, Ru C, Huber H, Fuchs BM, Amann R. In Situ Accessibility of Small-Subunit rRNA of Members of the Domains Bacteria, Archaea, and Eucarya to Cy3-Labeled Oligonucleotide Probes. Appl Environ Microbiol 69 (2003), 1748–1758. Sambrook J, Russell DW. Molecular Cloning: A Laboratory Manual. CSHL Press, 2001. D’Hondt S, Spivack AJ, Pockalny R, Ferdelman TG, Fischer JP, Kallmeyer J, Abrams LJ, Smith DC, Graham D, Hasiuk F, Schrum H, Stancin AM. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106 (2009), 11 651–11 656.

References | 141

[75] [76]

[77]

[78]

[79]

[80]

[81]

[82]

[83]

[84] [85]

[86]

[87]

[88] [89]

[90] [91]

Røy H, Kallmeyer J, Adhikari RR, Pockalny R, Jørgensen BB, D’Hondt S. Aerobic Microbial Respiration in 86-Million-Year-Old Deep-Sea Red Clay. Science 336 (2012), 922–925. Amaro T, Luna GM, Danovaro R, Billett DSM, Cunha MR. High prokaryotic biodiversity associated with gut contents of the holothurian Molpadia musculus from the Nazaré Canyon (NE Atlantic). Deep Sea Res Part I: Oceanogr Res Pap 63 (2012), 82–90. Breuker A, Stadler S, Schippers A. Microbial community analysis of deeply buried marine sediments of the New Jersey shallow shelf (IODP Expedition 313). FEMS Microbiol Ecol 85 (2013), 578–592. Bühring SI, Elvert M, Witte U. The microbial community structure of different permeable sandy sediments characterized by the investigation of bacterial fatty acids and fluorescence in situ hybridization. Environ Microbiol 7 (2005), 281–293. Burke DJ, Hamerlynck EP, Hahn D. Interactions between the salt marsh grass Spartina patens, arbuscular mycorrhizal fungi and sediment bacteria during the growing season. Soil Biol Biochem 35 (2003), 501–511. Danovaro R, Corinaldesi C, Luna GM, Magagnini M, Manini E, Pusceddu A. Prokaryote diversity and viral production in deep-sea sediments and seamounts. Deep Sea Res Part II: Top Stud Oceanogr 56 (2009), 738–747. Engelen B, Ziegelmüller K, Wolf L, Köpke B, Gittel A, Cypionka H, Treude T, Nakagawa S, Inagaki F, Lever MA, Steinsbu BO. Fluids from the Oceanic Crust Support Microbial Activities within the Deep Biosphere. Geomicrobiol J 25 (2008), 56–66. Ince BK, Usenti I, Eyigor A, Oz NA, Kolukirik M, Ince O. Analysis of Methanogenic Archaeal and Sulfate Reducing Bacterial Populations in Deep Sediments of the Black Sea. Geomicrobiol J 23 (2006), 285–292. Knittel K, Lemke A, Lochte K. Activity, Distribution, and Diversity of Sulfate Reducers and Other Bacteria in Sediments above Gas Hydrate (Cascadia Margin, Oregon). Geomicrobiol J 20 (2003), 269–294. Köchling T, Lara-martín P, González-mazo E, Amils R, Sanz JL. Microbial community composition of anoxic marine sediments in the Bay of Cádiz (Spain). Int Microbiol 14 (2011), 143–154. Leloup J, Loy A, Knab NJ, Borowski C, Wagner M, Jørgensen BB. Diversity and abundance of sulfate-reducing microorganisms in the sulfate and methane zones of a marine sediment, Black Sea. Environ Microbiol 9 (2007), 131–142. Losekann T, Knittel K, Nadalig T, Fuchs B, Niemann H, Boetius A, Amann R. Diversity and Abundance of Aerobic and Anaerobic Methane Oxidizers at the Haakon Mosby Mud Volcano, Barents Sea. Appl Environ Microbiol 73 (2007), 3348–3362. Manini E, Luna GM, Corinaldesi C, Zeppilli D, Bortoluzzi G, Caramanna G, Raffa F, Danovaro R. Prokaryote diversity and virus abundance in shallow hydrothermal vents of the Mediterranean Sea (Panarea Island) and the Pacific Ocean (North Sulawesi-Indonesia). Microb Ecol 55 (2008), 626–639. Mauclaire L, Zepp K, Meister P, McKenzie J. Direct in situ detection of cells in deep-sea sediment cores from the Peru Margin (ODP Leg 201, Site 1229). Geobiology 2 (2004), 217–223. Meyer-Dombard DR, Price RE, Pichler T, Amend JP. Prokaryotic Populations in Arsenic-Rich Shallow-Sea Hydrothermal Sediments of Ambitle Island, Papua New Guinea. Geomicrobiol J 29 (2012), 1–17. Molari M, Manini E. Reliability of CARD-FISH procedure for enumeration of Archaea in deepsea surficial sediments. Curr Microbiol 64 (2012), 242–250. Molari M, Giovannelli D, d’Errico G, Manini E. Factors influencing prokaryotic community structure composition in sub-surface coastal sediments. Estuar Coast Shelf 97 (2012), 141– 148.

142 | 6 Quantifying microbes in the marine subseafloor: some notes of caution [92]

[93]

[94]

[95]

[96]

[97]

[98] [99]

[100]

[101]

[102]

[103] [104]

[105]

[106]

[107]

Nunoura T, Inagaki F, Delwiche ME, Colwell FS, Takai K. Subseafloor microbial communities in methane hydrate-bearing sediment at two distinct locations (ODP Leg204) in the Cascadia Margin. Microbes Environ 23 (2008), 317–325. Nunoura T, Takai K. Comparison of microbial communities associated with phase-separationinduced hydrothermal fluids at the Yonaguni Knoll IV hydrothermal ¢eld, the Southern Okinawa Trough. FEMS Microbiol Ecol 67 (2009), 351–370. Omoregie EO, Mastalerz V, Lange G De, Straub KL, Kappler A, Røy H, Stadnitskaia A, Foucher J, Boetius A. Biogeochemistry and community composition of iron- and sulfur-precipitating microbial mats at the Chefren Mud Volcano (Nile Deep Sea Fan, Eastern Mediterranean). Appl Environ Microbiol 74 (2008), 3198–3215. Omoregie EO, Niemann H, Mastalerz V, Lange GJ De, Stadnitskaia A, Mascle J, Foucher J, Boetius A. Microbial methane oxidation and sulfate reduction at cold seeps of the deep Eastern Mediterranean Sea. Mar Geol 261 (2009), 114–127. Orcutt B, Boetius A, Elvert M, Samarkin V, Joye SB. Molecular biogeochemistry of sulfate reduction, methanogenesis and the anaerobic oxidation of methane at Gulf of Mexico cold seeps. Geochim Cosmochim Acta 17 (2005), 4267–4281. Quan X, Wang Y, Xiong W, He M, Yang Z, Lin C. Description of microbial community structure of sediments from the Daliao River water system and its estuary (NE China) by application of fluorescence in situ hybridization. Environ Earth Sci 61 (2010), 1725–1734. Ravenschlag K, Sahm K, Amann R. Quantitative molecular analysis of the microbial community in marine Arctic sediments (Svalbard). Appl Environ Microbiol 67 (2001), 387–395. Roalkvam I, Jørgensen SL, Chen Y, Stokke R, Dahle H, Hocking WP, Lanzén A, Haflidason H, Steen IH. New insight into stratification of anaerobic methanotrophs in cold seep sediments. FEMS Microbiol Ecol 78 (2011), 233–243. Rosselló-Mora R, Thamdrup B, Schäfer H, Weller R, Amann R. The response of the microbial community of marine sediments to organic carbon input under anaerobic conditions. Syst Appl Microbiol 22 (1999), 237–248. Sievert SM, Ziebis W, Kuever J, Sahm K. Relative abundance of Archaea and Bacteria along a thermal gradient of a shallow-water hydrothermal vent quantified by rRNA slot-blot hybridization. Microbiology 146 No. 6 (2000), 1287–1293. Wilms R, Sass H, Beate K, Cypionka H, Engelen B. Methane and sulfate profiles within the subsurface of a tidal flat are reflected by the distribution of sulfate-reducing bacteria and methanogenic archaea. FEMS Microbiol Ecol 59 (2007), 611–621. Rusch A, Huettel M, Reimers CE, Taghon GL, Fuller CM. Activity and distribution of bacterial populations in Middle Atlantic Bight shelf sands. FEMS Microbiol Ecol 44 (2003), 89–100. Danovaro R, Corinaldesi C, Marco G, Magagnini M, Manini E, Pusceddu A. Deep-Sea Research II Prokaryote diversity and viral production in deep-sea sediments and seamounts. Deep Res Part II 56 (2009), 738–747. Nunoura T, Oida H, Toki T, Ashi J, Takai K, Horikoshi K. Quantification of mcrA by quantitative fluorescent PCR in sediments from methane seep of the NankaiTrough. FEMS Microbiol Ecol 2 (2006), 149–157. Nunoura T, Soffientino B, Blazejak A, Kakuta J, Oida H, Schippers A, Takai K. Subseafloor microbial communities associated with rapid turbidite deposition in the Gulf of Mexico continental slope (IODP Expedition 308). FEMS Microbiol Ecol 69 (2009), 410–424. Oliveira V, Santos AL, Aguiar C, Santos L, Salvador AC, Gomes NCM, Silva H, Rocha SM, Almeida A, Cunha A. Prokaryotes in saltmarsh sediments of Ria de Aveiro: Effects of halophyte vegetation on abundance and diversity. Estuarine Coastal and Shelf Sci 110 (2012), 61–68.

Andreas Teske

7 Archaea in deep marine subsurface sediments 7.1 Introduction Gene sequencing of natural microbial communities in the deep subsurface has provided access to a new biosphere, characterized by evolutionary depth and diversity of novel archaeal lineages. Specific phylum-level lineages and functional groups of archaea occur consistently in deep subsurface sediments, and contrast with differentlystructured communities in the surface biosphere. This chapter provides an overview on currently uncultured archaeal lineages in the marine sedimentary subsurface, their phylogenetic diversity and habitat preferences, and contrasts this barely explored archaeal biosphere with the more limited physiological and phylogenetic spectrum of cultured subsurface archaea, the methanogens.

7.2 Archaeal Ribosomal RNA phylogeny Archaea were established as the third domain of life, next to Bacteria and Eukaryotes, on the basis of ribosomal RNA phylogeny [1]. The commonly adopted “gold standard” for gene-based identification of microorganisms is the small subunit ribosomal ribonucleic acid (ssu rRNA) gene, coding for rRNAs that are integral parts of the ribosome, the multienzyme complex in all living cells that translates genetic information into proteins of multiple physiological functions, after its transcription from DNA-coded genes into messenger RNA. The small subunit ribosomal RNAs serve as scaffolding molecules that maintain the complex three-dimensional structure of the ribosome; they occur in two size classes, 16S rRNA in bacteria and archaea with about 1500 nucleotides length and 18S rRNA in eukaryotes, about 2400 nucletides in length. Very short rRNAs, the 5S rRNA and the longer large subunit rRNA (23S rRNA in bacteria and archaea, 28S rRNA in eukaryotes) are also integral structural components of the ribosome. Due to extremely strong functional and structural constraints that conserve all components of the finely tuned ribosomal machinery, rRNA molecules – and the genes that are coding for them – have evolved sufficiently slowly to reveal, by degree of nucleotide-sequence similarity, the evolutionary paths and phylogenetic relationships of all life forms on Earth [2]. Ribosomal RNA molecules and, to a lesser extent, conserved functional genes, preserve a memory of deep evolutionary divergence that reaches in principle back to the inferred ancestral form of cellular life with inheritable genomes, the last universal common ancestor or progenote [3]. Milestones in the recognition and development of small subunit rRNA as a universal phylogenetic

144 | 7 Archaea in deep marine subsurface sediments marker include the initial discovery of the deep evolutionary divergence between the two prokaryotic domains of life, the Bacteria and the Archaea [4]; the first universal rRNA-based tree of life [5]; the designation of the Bacteria, Archaea and Eukaryotes as the highest-order taxonomic divisions (domains) corresponding to the actual evolutionary divergence of life [1]; and the validation of rRNA phylogenies by complete genome sequencing that confirmed the deep genomic and evolutionary divergences between the bacterial, archaeal and eukaryotic domains [6]. Sequencing surveys of 16S rRNA genes, amplified from extracted microbial community DNA by polymerase chain reaction (PCR), have started to reveal the taxonomic and evolutionary diversity of microbial life in the deep subsurface biosphere, and the extent of microbially colonized deep subsurface environments. With increasingly fine-grained phylogenies corresponding to biochemically and physiologically coherent groups of microorganisms, rRNA phylogenies have become the molecular roadmap of the microbial world; they provide an essential organizing principle for standard reference works in microbiology, such as Bergeys Manual for Determinative Bacteriology (2nd edition) or The Prokaryotes (3rd and 4th edition). With the general availability of PCR and cloning and sequencing of PCR amplicons, 16S rRNA genes became indispensable in mapping microbial diversity in the environment, and demonstrated the extent of previously unrecognized microbial diversity in nature that has so far eluded cultivation efforts [7]. Within the archaeal domain, two subdomain-level clusters were originally recognized: the Crenarchaeota, represented by hyperthermophilic, sulfur-dependent archaea from hot springs and hydrothermal vents and the Euryarcheota, containing the methanogens and extreme halophiles [2]. As the tally of mutually exclusive archaeal lineages increased rapidly [8, 9], it became increasingly difficult to fit them into the crenarchaeotal vs. euryarchaeal divide. The “misfits” that defy categorization include the Korarchaeota [10], the Ancient Archaeal Group lineage [11], the Thaumarcheota [12], the obligate symbiont Nanoarchaeum equitans [13], and the major subsurface lineage Marine Benthic Group B [14]. Current efforts for a genomic census of the widest possible range of bacterial and archaeal lineages [15] recognize the increasingly complex phylogenetic landscape, and propose archaeal “superphyla” consisting of multiple phylum-level lineages united by common ancestry.

7.3 Marine subsurface Archaea The sedimentary marine subsurface is permeated by microbial life; recent global quantifications have shown that the total contribution of the marine sedimentary biosphere amounts to about 0.6% of Earth’s living biomass; cell abundance and density decrease with increasing distance from land, where organic-rich continental margin and coastal sediments provide the subsurface biosphere endmember of high microbial activity and abundance [16]. Yet, even highly oligotrophic open-ocean sedi-

7.3 Marine subsurface Archaea |

145

ments that are oxygenated and depleted of organic carbon nevertheless contain active microbial populations that consume oxygen and presumably remineralize buried organic substrates at very slow rates [17, 18]. The subsurface biosphere is therefore an omnipresent mediator between biological and geological elemental cycles in the subsurface [19]. Microbiological surveys of the deep subseafloor, often performed within the Ocean Drilling Program and its successor programs, have shown that much of this subsurface biosphere is archaeal. The first 16S rRNA gene survey of ODP subsurface sediments, from Hole 798B in the Japan Sea, drilled during ODP Leg 128, detected a new bacterial lineage that could not be subsumed under any other taxonomic category within the 16S rRNA phylogeny of the bacterial world [20]; archaea were not detected. This changed quickly as many more deep sediments recovered by ODP and IODP drilling were examined by 16S rRNA sequencing: ODP 190 site 1173 in the Nankai Trench Accretionary wedge [21] and site 1176 overlying the Nankai subduction zone [22]; ODP leg 201 Eastern Equatorial Pacific site 1225 [23]; Peru Margin and Peru Trench sites 1227 to 1230 during ODP leg 201 [24–28] and Peru Basin site 1231 during ODP leg 201 [29]; ODP leg 204 sites 1244 and 1251 on the Cascadia Margin [25, 30]; ODP leg 210 site 1276 on the Newfoundland Margin [31]. Additionally, significant archaeal community surveys were performed independently of ODP in buried sapropel layers of the eastern Mediterranean [32] and in subsurface sediments around the Pacific Rim: in the New Caledonia Basin and Fairway Basin southwest of New Caledonia [33], in several deep-water locations of the South China Sea [34], in volcanic ash layers interbedded with marine sediments in the Okhotsk Sea [35], and in methane-hydrate rich deep subsurface sediments of the Nankai Trough forearc basin [36]. Within the Integrated Ocean Drilling Program (IODP), 16S rRNA gene analyses of archaeal populations started on the very first expedition (IODP 301) on sediments [37] and borehole crusts [38] from the Juan de Fuca Ridge flanks; sediments from IODP expedition 302 on the Lomonosov Ridge in the Arctic Ocean [39]; sediments and buried coral carbonates from IODP expedition 307 to the Porcupine Seabight in the northeastern Next double page: Fig. 7.1: Archaeal 16S rRNA phylogeny, based on distance analysis of 16S rRNA nucleotide positions 24 to 906 (distance measure HKY85, optimality criterion = minimum evolution, gamma distribution factor of 0.5) performed with PAUP4.0 [98]. The tree topology was checked by 200 bootstrap reruns. References for archaeal lineages: Marine Benthic Group (MBG) A to E [14]; Marine Group I and II [99]; Marine Group III [100]; Deep-Sea Hydrothermal Vent Euryarchaeotal Groups (DHVEG) [11]; Miscellaneous Crenarchaeotal Group (MCG) [35]; South African Goldmine Euryarchaeotal Group (SAGMEG) [49]; Deep-Sea Euryarchaeotal Groups (DSEG) 1–4 and pSIA17 clade [44]; C3 archaea [25]; Marine Hydrothermal Vent Group (MHVG) III and Ancient Archaeal Group (AAG) [11]. The C3 archaeal group is congruent with MCG lineage 15 [51]. DSEG-2 is synonymous with the VAL III clade, originally found in freshwater lakes [101]; DSEG-1 is synonymous with Deep Sea Euryarchaeotal Group and DHVEG-5 matches the Miscellaneous Euryarchaeotic Group, MEG; both [49]. The 16S rRNA sequences SCGC AB539-E09 and SCGC AB540-N05 are from contigs of single cell genomes [52].

100

82

100

97

81

100

100

100 98

99

Peru Margin ODP1227D-37.8(DQ302006) Okhotsk Sea OKHA-5.5 (AB094545) Nankai Trough MA-C1-50 (AY093454) MAC1.50 Peru Basin ODP1231 1.8mBB.1 (AY661826) AY661826 South African Goldmine SAGMA-Q (AB050222) SAGMA.Q South African Goldmine SAGMA-E (AB050209) SAGMA.E

SAGMEG

Peru Margin ODP1229 rRNA(DQ302022) DQ302022 NW Atlantic CRA12-27 cm (AF119123) CRA12.27c SCGC AB539-N05 (ALXL01000072) MBD539N05 Japanese Vent pMC2A203 (AB019737) pMC2A203 Nankai ODP1173 NANK-A153 (AY436524) NANKA153 100 Eel River seep sediment TA1F2 (AF134390) TA1F2 Okhotsk Sea OKHA-1.35 (AB094527) OKHA1.35 MBG-D Guaymas Basin C1R046 (AF419643) C1R046 100 76 Freshwater Reservoir HTA-C7 (AF418931) HTA.C7 99 Fuel-contaminated sediment WCHD3-02 (AF050616) WCHD3.02 Methanomassiliicoccus luminyensis (HQ896499) Massilii 86 Okhotsk Sea OHKA 4.77 (AB094542) OHKA4.77 97 Peru Basin ODP1231 1.8mENZ.1 (AY661821) AY661821 TMEG South African Goldmine SAGMA-T (AB050225) SAGMA.T 100 DQ300569 Aloha clone HF770_036B20 (DQ300569) 99 MG-III Deep-Sea clone DH148-W24 (AF257278) AF257278 South Pacific abyssal sediment (FJ487458) 100 FJ487458 100 Antarctic bathypelagic sediment (EF069336) MG-V EF069336 Tropical estuarine sediment (DQ641797) DQ641797 83 Water Column SBAR16 (M88077) 100 SBAR16 MG-II Water Column Antarctic5 (U11044) ANTARCTIC 100 Yonaguni Knoll IV hydrothermal sediment (AB301993) AB301993 MBG-E NW Atlantic CRA13-11 cm (AF119124) CRA13.11c 91 Japanese Vent pM2CA10 (AB019739) pMC2A10 100 Cascadia Margin ODP1251A-15.24 (AB177273) DHVEG-2 1251A1524 96 Aciduliprofundum boonei (DQ451875) Acidulipr Thermoplasma acidophilum (M38637) Thermoplasmatales 100 Thplasma Picrophilus (X84901) Picrophil oshimaeEquatorial Pacific ODP1225 (AY800211) AY800211 100 South Pacific (FJ487470) DSEG-3 82 FJ487470 Rainbow Vents hydrothermal sediment (AY354110) AY354110 100 Cascadia ODP1251A-15.13 (AB177269) 1251A1513 94 Peru Margin ODP1229 rRNA (DQ302023) DQ302023 100 Peru Margin ODP1227 rRNA (DQ301971) DQ301971 100 South China Sea (EU385849) EU385849 DSEG-2 South Pacific abyssal (FJ487457) FJ487457 100 West Pacific sediment pPCA7.13 (AB049034) pPCA7.13 98 DSEG-1 South Pacific abyssal sediment (FJ487459) FJ487459 99 South China Sea sediment (EU048658) EU048658 Japanese Vent pMC2A232 (AB019745) pMC2A232 100 Peru Basin ODP1231 43mENZ.6 (AY661817) AY661817 100 DSEG-4 South Pacific abyssal sediment (FJ487462) FJ487462 100 South Pacific abyssal sediment (FJ487460) FJ487460 South Pacific abyssal sediment (FJ487461) FJ487461 96 Japanese pISA1 Vent pISA1 (AB019751) South Pacific (FJ487465) FJ487465 100 Pacific abyssal WPA18 (AJ783665) WPA18 Anoxic AS59 (AJ783664) ABS9soilPacific abyssal sed. WPA17 (AJ783664) WPA17 Thermococcus celer Tc.celer DHVEG-5 Palaeococcus ferriphilus Hyperthermophiic euryarchaeota Palaeococ Methanocaldococcus jannaschii

Marine water column Marine Surficial sediment (< 2 mbsf) Marine Subsurface sediment Marine hydrocarbon seep sediment Hydrothermal vent or hot spring Terrestrial subsurface / thermal Terrestrial subsurface / soil Freshwater habitat Cultured species or Candidatus

146 | 7 Archaea in deep marine subsurface sediments

78

91

98

100

0.05 per site 0.05substitutions substitutions/site

84

80

96

100

80

Okhotsk Sea OKHA 1.1 (AB094513) Nankai Trough ODP1173 NANK-A84 (AY436515) NANKA84 Japan Trench methane seep JTB173 (AB015273) MBG-C JTB173 Nankai Trough MA-B1-3 (AY093447) MAB13 Northwest Atlantic CRA 9-27 cm (AF119129) CRA927cm 100 OKHA5.34 Okhotsk Sea OKHA 5.34 (AB094548) Nankai Trough MA-C1-5 (AY093451) MAC15 98 Peru Margin ODP1229 rRNA (DQ302009) DQ302009 Okhotsk Sea OKHA 5.24 (AB094547) MCG OKHA5.24 76 Terrestrial Palaeosol ARC43 (AF005758) ARC43 Freshwater Reservoir HTA-B10 (AF418925) HTAB10 Yellowstone pSL123 hot spring pSL123 (U63345) Peru Margin ODP1229D rRNA (DQ302026) DQ302026 Okhotsk Sea OKHA 1.27 (AB094524) OKHA1.27 100 Peru Margin ODP1227A18.21 (AB177014) 1227A1821 100 SCGC AB539-E09 (ALXK01000069) C3 MCG539E09 Peru Trench ODP1230 rRNA (AB177118) 1230A3309 Desulfurococcus mobilis Dc.mobili 100 Sulfolobus acidocaldarius Hyperthermophilic crenarchaeota Sl.acidoc Pyrobaculum islandicum Pb.island 100 Peru Basin ODP-1231 43mENZ.2 (AY661820) AY661820 East Pacific Rise sediment (EU259351) EU259351 Japanese Vent pISA7 (AB019733) 100 pSIA7 Aegean Sea sediment 20a-6 (AJ299151) ASC20a.6 pISA7 clade Juan de Fuca vent FZ2aA56 (AY166117) VCF2aA56 100 Northwest Atlantic APA 2-17 cm (AF119135) APA2.17cm 93 Equatorial Pacific ODP1225-1H6 (AY800212) AY800212 100 MBG-A Equatorial Pacific ODP1225-1H1 (AY800216) AY800216 Yellowstone hot spring pSL12 (U63343) pSL12 100 Equatorial Pacific ODP1225 (AY800218) AY800218 100 MG-I Equatorial Pacific ODP1225 (AY800219) AY800219 “Cenarchaeum Ca.symbio 98symbiosum” (U51469) Peru Trench ODP1230 rRNA (DQ302030) DQ302030 94 Nankai Trough ODP1173 NANKA8 (AY436511) NANKA8 Japanese Goldmine pHAuA.5 (AB072723) pHAuA.5 100 Equatorial Pacific ODP1225-1H1 (AY800207) AY800207 72 Northwest Atlantic CRA 8-27 cm (AF119128) CRA8.27cm 100 Nankai Trough MA-A1-3 (AY093448) MAA13 100 Peru Margin ODP1227 rRNA (DQ302005) DQ302005 97 Okhotsk Sea OKHA1.18 (AB094523) OKHA1.18 Japanese vent pMC2A36 (AB019720) pMC2A36 100 Northwest Atlantic APA 3-11 cm (AF119137) APA3.11cm 100 Okhotsk Sea OKHA4.94 (AB094544) OKHA4.94 Guaymas Basin C1-R043 (AF419642) C1R043 MBG-B / DSAG Japanese vent pMC2A308 (AB019721) pMC2A308 Okhotsk Sea OKHA1.28 (AB094525) OKHA1.28 Amsterdam Mud Volcano-1A-24 (AY592253) MHVG AY592253 Japan. vent pMC2A15 (AB019718) pMC2A15 100 Cascadia Margin ODP1251A1.15 (AB177260) AB177260 -3 Cascadia Margin ODP1244A-2.3 (AB177229) AB177229 Japanese vent pMCA256 (AB019717) pMCA256 Japan. vent pSSMCA1 AAG (AB019716) 73 100

7.3 Marine subsurface Archaea | 147

148 | 7 Archaea in deep marine subsurface sediments Atlantic Ocean [40, 41], and turbidite sediments from IODP expedition 308 to the Brazos-Trinity and Mars-Ursa Basins on the Gulf of Mexico continental slope [42]. Recently, sequencing surveys of sediments from the IODP Expedition 329 site survey cruise have examined bacterial and archaeal communities of oligotrophic South Pacific sediments [43, 44]. Currently, archaeal 16S rRNA gene sequencing surveys are not only performed for their own sake, but also as a baseline for increasingly sophisticated functional gene analyses and high-throughput metagenome and metatranscriptome surveys. For subsurface and deep ocean microbial habitats, 16S rRNA genes provide the essential taxonomic “grid” for mapping microbial biodiversity in the subsurface, for consistent comparisons and meta-analyses, and for accommodating novel lineages that continue to be discovered in the subsurface [45–48]. The most commonly encountered subsurface phylum-level archaeal lineages are the Miscellaneous Crenarchaeotal Group (MCG), discovered in deep marine sediments [11, 35]; the Marine Benthic Group A to E (MBG-A to MBG-E) first found in surficial sediments of the continental slope and abyssal basins offshore New England [14]; the Deep-Sea Hydrothermal Vent Euryarchaeotal group 1 to 7 (DHVEG-1 to 7), originally detected in hydrothermal vent habitats [11]; and South African Goldmine Euryarchaeotal Group (SAGMEG), first found in terrestrial deep goldmines [49] but also widespread in marine subsurface sediments (󳶳 Fig. 7.1). Continued surveys of deep subsurface sediments, surficial marine sediments and deep-sea hydrothermal vents have uncovered additional, not-yet-cultured archaeal lineages (󳶳 Fig. 7.1). The most frequently found lineages remain the MCG, MBG-B, MBG-D and SAGMEG clades, but other lineages are found in oligotrophic marine sediments (Deep-Sea Euryarchaeotal Group [DSEG] 1 to 4, [44]); some sediment clones appear within the DHVEG clades that were originally found at hydrothermal vents [11]. Thus, the 16S rRNA phylogeny provides the taxonomic skeleton for a complex archaeal subsurface biosphere, and suggests linkages between the sedimentary and the hydrothermal biospheres. The names or acronyms of these archaeal lineages reflect the absence or the problematic ambiguity of ecophysiological clues and hypothesis-generating phylogenetic affiliations; genomic and cultivation approaches will be required to make progress. The Miscellaneous Crenarchaeotal Group (MCG), one of the most frequently detected archaeal subsurface lineages, is currently the best example for multidisciplinary, focused investigations of an uncultured archaeal group. MCG archaea are being investigated by fosmid sequencing [50], molecular quantification in multiple environments [51] and single-cell genome sequencing [52]. Members of this group have been enriched from marine sediments in stable isotope probing experiments with 13 C-labeled acetate [53]. Single-cell genome sequencing of MCG and Marine Benthic Group D cells from subsurface sediments of Aarhus Bay has uncovered genomic pathways for the degradation of detrital protein; the suite of genes includes extracellular peptidases, oligopeptide transporters for transfer into the cell, intracellular peptidases, aminotransferases that deaminate amino acids to 2-keto-acids, and oxidation

7.4 Archaeal habitat preferences in the subsurface | 149

of the 2-keto-acids to acyl-CoA and organic acids [52]. These genome-based inferences were substantiated by enzymatic activity measurements of extracellular peptidases (the types found in the MCG and MBG-D genomes) in the same sediments [52]. Cultivation strategies for MCG are under development; so far, diverse anaerobic conditions using basal media, amino acids, H2 /CO2 gas phase and initial aerobic exposure have enriched specific MCG populations in vitro by an order of magnitude [54]. Since the MCG archaea account for a large proportion of subsurface archaea detected in the sedimentary subsurface, these MCG results substantiate the working hypothesis that the archaeal subsurface biosphere is predominantly heterotrophic. The 𝛿13 C-isotopic composition of prokaryotic cells and (archaeal) intact polar lipids in marine subsurface sediments indicate that methane or DIC are not significant carbon substrates for subsurface cells; most of the subsurface biosphere appears to subsist heterotrophically on buried organic carbon of photosynthetic origin [24, 55, 56].

7.4 Archaeal habitat preferences in the subsurface The 16S rRNA sequences of different archaeal lineages are not randomly distributed in the marine subsurface, but show evidence of biogeographical structure most likely controlled by in situ chemical regime. For example, sediment-hosted methane hydrates [25], volcanic ash layers embedded in marine sediments [35], sulfate–methane transition zones [27], oxygen–nitrate porewater gradients [43, 44], and organic carbon content and redox status of subsurface sediments [48] appear to select in favor of phylogenetically distinct archaeal lineages. Of all subsurface geochemical interfaces, the sulfate–methane transition zone has received the most sustained attention; microbial communities that oxidize methane with sulfate as the electron donor [57] were predicted to be a major component of the subsurface biosphere [58]. Interestingly, 16S rRNA surveys of the sulfate–methane transition zones in subsurface sediments found quite consistent microbial community signatures within and around these sediment horizons: MCG, MBG-B, SAGMEG and MBG-D were repeatedly detected by sequencing reverse-transcribed 16S rRNA [24, 27] and 16S rRNA genes [26, 42, 59], whereas anaerobic, sulfate-dependent methanotrophic archaea (ANME) typical for shallow benthic marine sediments and methane seep sediments remained commonly below detection limit with general archaeal 16S rRNA primers, and required group-specific 16S rRNA and functional gene primers for successful detection in deep subsurface sediments [37]. Obviously, sulfate-dependent methane oxidation is not the only process that sustains subsurface microbial cells in the sulfate–methane transition zone, and heterotrophic metabolisms that are linked to assimilation of buried organic matter predominate [24]. Alternatively, unrecognized methane oxidizers might be hiding in plain sight within uncultured microbial groups. In contrast to the deep subsurface, ANME archaea are easily detectable in shallow sulfate–methane transitions zones of organic-rich, surface sediments [57, 60].

150 | 7 Archaea in deep marine subsurface sediments Seafloor sediments in the open ocean are oxidized and do not harbor the sulfate– methane transition zones that are conspicuous in organic-rich continental margin sediments; in extreme cases, microbial activity is not even sufficient to deplete porewater oxygen throughout the deep sediment column [18, 61]. Recent sequencing surveys are examining drilling sites and marine sedimentary environments representing different trophic regimes: from thoroughly oxidized, extremely organic-lean sediments that accumulate with very slow sedimentation rates in the ultraoligotrophic South Pacific Gyre [17] to reduced, organic-rich, sulfidic or methanogenic sediments on highly productive continental margins, such as the Peru Margin or the Cascadia Margin [62]. This ultraoligotrophic to eutrophic spectrum should influence the composition and activity of subsurface microbial communities, in response to different redox regime, organic carbon content and available substrate spectra. Indeed, a comparison of contrasting sites shows that archaeal lineages change systematically over a spectrum of organic-lean, oxidized marine sediments from abyssal plains to organicrich, reduced sediments on continental margins (󳶳 Fig. 7.2) [48]. The visible impact of organic carbon availability and redox status on microbial community structure in marine subsurface sediments, as reflected in 16S rRNA gene diversity, indicates a basic correspondence between geochemical in situ regime and 16S rRNA gene-based archaeal community structure. More generally, empirically-observed correspondences between subsurface habitat characteristics and microbial community 16S rRNA signature support the working hypothesis that microbial community structure and in situ geochemistry are coupled, even in the metabolically-slow sedimentary subsurface. Decoupling mechanisms, for example accumulation of fossil and/or inactive archaeal cells, remnant populations from past geological and geochemical regimes, or dispersed cells from distant source habitats [63] have to be carefully considered in individual case studies and specific localities. However, such mechanisms are unlikely to overwrite and invalidate global distribution and abundance patterns of active in situ microbial populations that consist of living cells with intact rRNA, and rRNA genes. By the same token, comprehensive meta-analyses of the microbial community composition and abundance in the deep subsurface biosphere, and their correlations to geochemical habitat characteristics and controls [48] should be a high research priority for the immediate future. So far, small subunit rRNA sequencing surveys in all three domains of life are consistent with the notion of a heterotrophic subsurface biosphere that ultimately depends on buried organic matter of planktonic, photosynthetic origin as carbon and energy source [24, 28]. Sequence signatures of lithotrophic and autotrophic subsurface archaeal life comparable to terrestrial subsurface examples [64] have not been identified with certainty. Yet, marine subsurface archaeal lithoautotrophs might occur within the phylum Thaumarchaeota, whose few cultured representatives (Marine Group I archaea) are autotrophic ammonia oxidizers [65]. Marine Group I archaea are usually abundant in the marine water column, but also occur in oxidized marine sediments as long as oxygen or nitrate are present [43, 44]. The working hypothesis of

7.4 Archaeal habitat preferences in the subsurface | 151 23 19

32 31 29

32 32

12

32

32

26

34 32 30

35

27

27

27

32

26

9

27

26 8

23

7

21

27 27

6 15

26 8 23

1 9 1

Leg 201: Site 1231

21 15

2

SPG12

32 29 28

13 5

Leg 201: Site 1225

Leg 201: Site 1226

1. Cren/Thaum: MBG-A affiliated 2. Cren/Thaum: pISA7 3. Cren/Thaum: Fhm5A01* 4. Eury: Fe_A_9* 5. DHVEG: hfmA112* 6. DHVEG: DSEG-1 7. DHVEG: DSEG-2

23 21 19

26

15

23

14 13 12 11 10 8 7 5 3

21 19 17 16 12

27

1

8

26 27 26 26 23 21 19 16 15 13 11 7 6 5 4 3 1

22. Eury: Guaymas Euryarchaeotal Group 23. Eury: SAGMEG affiliated 24. Eury: ANME-1 25. Eury: Methanococci 26. Cren/Thaum: MBG-B/DSAG 27. Cren/Thaum: MCG+C3 28. Cren/Thaum: Fnv51

25 23 19 18 17 7 1

23

3

ZoNéCo MD05-2902 MD05-2896 Exp. 308: -12: New North Core South Core Louisiana Caledonia S. C. Sea S.C. Sea continental Basin slope 8. DHVEG: DSEG-3 9. DHVEG: DSEG-4 10. DHVEG: SM1K20 11. DHVEG: pMC2A384* 12. Thermoplasmata: MG-V 13. Thermoplasmata: HydGC-83* 14. Thermoplasmata: SBAK-shallow-09*

26

24

26 24 23 22 21 16 15 10 8 7 1

25 23 20 17 16 14 13 8 7

Leg 204: Leg 201: Cascadia Peru Margin Margin + Trench

Medit. cold seeps

15. Thermoplasmata: MKCST-A3 16. Other Thermoplasmata: Oligotrophic 17. Eury: Thermococci 18. Eury.: Archaeoglobales 19. Eury: Methanomicrobia 20. Eury: Methanobacteria 21. Eury: MBG-E

29. Cren/Thaum: MHVG-2 30. Cren/Thaum: MHVG-3 31. Cren/Thaum: SCG/1.1b 32. Thermoplasmata: MBG-D 34. Thermoplasmata: Aciduliprofundaceae/DSHVEG 35. Other Thermoplasmata: eutrophic

Fig. 7.2: Relative abundance of uncultured archaeal lineages in 16S rRNA gene clone libraries for all oligotrophic sediment sites for which public relative-abundance data and geochemical information is available [48]. Clades 1–16, colored blue, were either shared with eutrophic endmember sites but comprised less than 1.5% of total clones in any eutrophic site, or were entirely absent from eutrophic endmembers. Selected eutrophic sediment sites include the Peru Margin, Peru Trench and Cascadia Margin; Mediterranean mud volcanoes are included as a surface expression of subsurface processes. Sediment sampling sites are arranged as increasingly oligotrophic from right to left, as suggested by available geochemical parameters [48].

152 | 7 Archaea in deep marine subsurface sediments a heterotrophic subsurface biosphere has to be qualified by the persistent sampling bias towards organic-rich, reducing continental margin sediments where buried organic biomass is omnipresent, and by the general paucity of physiological knowledge on uncultured archaea in the subsurface. Inferring physiological and geochemical key characteristics of deep subsurface archaea requires different approaches, such as advances in cultivation and pure culture isolation, functional gene sequencing and metagenomic surveys.

7.5 Methanogenic and methane-oxidizing archaea So far, the methanogens and (with caveats) the sulfate-dependent methane-oxidizing archaea represent the only group of subsurface archaea where pure cultures or enrichments are available for physiological study; at the same time, functional gene assays allow specific molecular monitoring of these archaea. Therefore, subsurface “archaeologists” have learned more about the subsurface methanogens, their habitat preferences and physiology, than about their uncultured archaeal cousins. The cultured spectrum of subsurface methanogens includes the hydrogenotrophic autotroph Methanobacterium subterraneum from deep terrestrial groundwater [66], the slightly thermophilic (growth optimum at 45 °C), acetate-assimilating hydrogenotroph Methanoculleus submarinus from hydrate-rich deep subsurface sediments (247 m below surface) of the Nankai Trough [67], the hydrogenotrophic and strictly autotrophic species Methanococcus aeolicus from the same sample [68], a new autotrophic strain of the species Methanobacterium palustre and the methanol- and trimethylamine-utilizing species Methanolobus chelungpuianus from deep sandstone samples recovered of a terrestrial earthquake fault in Taiwan [69], the methylotroph Methanolobus profundi from a deep subsurface gas field [70], the methylotrophic and acetoclastic species Methanosarcina horonobensis from subsurface groundwater [71], and numerous methanogen isolates from oil wells, discussed in [69]. A common thread that connects these subsurface methanogens is the metabolic preference for two types of methanogenic substrates; in addition to the classical substrates hydrogen and CO2 for hydrogenotrophic methanogenesis, methylated C1 -substrates are frequently used. These noncompetitive substrates are not consumed by sulfate-reducing bacteria, and are therefore available for methanogens in the marine subsurface. Recently, the range of methanogenic C1 substrates has expanded to include choline (N,N,N-trimethylethanolamine) and N,N-dimethylamineethanolamine; these substrates were used by new strains of the genus Methanococcoides [72], a methylotrophic genus that has been isolated consistently from marine and saline sediments [73]. The phenotype of pure-culture isolates characterized by specific metabolic functions and biogeochemical activities can be linked to environmental genotypes on the basis of protein-coding genes (functional genes) and their gene transcripts and resulting catalytic proteins (enzymes) that are involved in cellular maintenance, biosynthe-

7.5 Methanogenic and methane-oxidizing archaea |

153

sis, substrate transport and metabolism; they also provide a record of how genes are vertically and laterally transferred between organisms over time, and permit insights to the identity and activity of unknown or uncultivated organisms in the environment. Functional genes that are used for functional identification and phylogenetic studies should have the following characteristics: (1) the gene encodes an enzyme used for one specific reaction in a well-characterized group of microorganisms; (2) a large gene sequence database and alignment is available to infer the phylogenetic identity of environmental sequences; (3) sufficient nucleotide sequence conservation so genes can be comprehensively targeted by general or group-specific probes, such as PCR primers; and (4) the evolutionary history is well constrained so organisms which have obtained the gene through lateral transfer can be identified [74]. In practice, functional genes meet all these criteria only to some extent. Highly conserved genes may encode different versions of enzymes that catalyze key reactions within a pathway in different directions. The best archaeal example, the gene encoding the alpha subunit of methyl coenzyme M reductase (mcrA) [75], encodes the enzyme that reduces the coenzyme Mbound methyl group to methane in methanogens and performs the reverse reaction in anaerobic methanotrophs [76, 77]. Such cases demonstrate that, without additional evidence, functional gene proxies or stable C isotopic signatures of reaction educts or products may not indicate the direction of an enzymatic reaction with sufficient certainty [78–80]; microbial rate measurements in natural samples and biochemical tests in enrichments and pure cultures remain the gold standard to verify that a certain process performed by a specific group of microorganisms is actually taking place. Using the functional gene approach, the community composition of methane-cycling archaea has been characterized at several Pacific Rim ocean drilling locations: the Nankai Trough [21], the Peru Trench [25], the Peru Margin [59], the Cascadia Margin [81], the Shimokita Peninsula on the east coast of Japan [82], and the Juan de Fuca Ridge flanks [83]. In addition, mcrA genes have been quantified via real-time PCR in cores from the Cascadia Margin [84], and the Porcupine Seabight in the Atlantic Ocean [40]. At most of these sites, mcrA genes were only reliably detected and analyzed at few sampling depths. Since mcrA and 16S rRNA gene sequences of known methanogens or methanotrophs have turned out to be hard to detect, even at sites with high biogenic methane concentrations and high microbial biomass [40, 42, 59], current methods or commonly used PCR primers need to be improved. Highly degenerate mcrA gene primers with ambiguous nucleotide positions are widely used [85–87], but should be replaced by systematically designed nondegenerate primers that target specific methanogen groups with high specificity and sensitivity [37, 83]. On the other hand, PCR results could indeed indicate that subsurface methanogens may not be as abundant as expected; in this view, small methanogen seed populations have over geologic time produced the vast biogenic methane reservoir found in subseafloor sediments [25]. Interestingly, this possibility is substantiated by metagenomic surveys of subsurface microbial communities at the Peru Margin [88]. Despite the prominent geochemical profiles of sulfate and methane, few functional genes for methanogenesis,

154 | 7 Archaea in deep marine subsurface sediments sulfate-dependent methane oxidation and sulfate reduction were found in the Peru Margin metagenome, and no increase with depth was observed [88]. A third possibility exists that some methane-cycling archaea in subseafloor habitats might be too divergent from their relatives in the surface biosphere to be detected with the same functional gene methods; primer-independent metagenome surveys should in the long run detect such populations. The published mcrA sequences show a considerable diversity of methane-cycling archaea in subseafloor sediments, with genera belonging to the orders Methanosarcinales (Methanosarcina, Methanococcoides, Methanosaeta, ANME-2), Methanocellales (Methanocella), Methanomicrobiales (Methanoculleus), Methanobacteriales (Methanobrevibacter, Methanobacterium), Methanococcales (Methanococcus), ANME-1, and a novel order-level lineage containing the genus Methanomassiliicoccus [reviewed in 74]. The ANME-1 lineage and an uncultured cluster within the Methanosarcinales have been detected in deep subsurface basalt [83]. Members of the genera Methanosarcina, Methanobrevibacter, and Methanobacterium [21, 26, 81, 82] have been found commonly in deep subsurface sediments; nearly identical phylotypes related to Methanosarcina barkeri were detected in the Nankai Trough [21] and the Peru Margin [26]. None of the mcrA groups detected during drilling expeditions are unique to the deep biosphere; in most cases they are characteristic for a wider range of benthic marine environments [reviewed in 74]. Surprisingly, mcrA genes with high sequence similarity to the hydrogenotrophic species Methanobrevibacter arboriphilus appear in sulfate-rich surface sediments of ODP Site 1174 [21]. Nearly identical genes of Methanobrevibacter have been detected repeatedly in mcrA gene subseafloor surveys in the Peru Margin [26] and offshore in the Shimokita Peninsula on the east coast of Japan [82]. The genus Methanobrevibacter is otherwise known from environments that differ strikingly from the energy-depleted deep subsurface biosphere, such as human–animal intestines, activated sludge and decaying wood [89], and it is used as a tracer to identify sewage pollution [90]. These possible contamination problems call attention to improved precautions and strict quality controls in subsurface sampling and sample processing [83, 91].

7.6 Archaeal abundance and ecosystem significance in the subsurface The archaeal domain harbors a wide range of physiological and structural adaptations to survival under severe energy limitation [92]. Physiological investigations of sediment archaea are substantiating this life strategy; for example, carbon assimilation into membrane core lipids appears to be extremely slow and was nearly undetectable even during long-term in situ seafloor incubations [93]. In a comprehensive metatranscriptome study of Peru Margin subseafloor sediments, archaeal gene expression was found to account only for a minor percentage of all recovered gene transcripts [94],

References |

155

with the caveat that transcript annotation always has room for improvement. Attenuated carbon assimilation and gene expression could underlie the very slow biomass turnover in the subseafloor sedimentary biosphere based on biomass quantifications and racemization studies [95], and they are consistent with very low energy availability per microbial cell in the subsurface [96]. If archaea are physiologically predisposed to energy- and substrate-limited subsurface conditions, do they dominate the subsurface biosphere by cell abundance or biomass? Quantifications of Archaea in comparison to Bacteria at different drilling sites using different methodologies have so far produced widely divergent estimates; a recent critical review and meta-analysis of published qPCR and fluorescent in situ hybridization surveys and their respective methodological biases indicates similar abundances for both domains of life in the marine sedimentary subsurface [97]. Since archaea ubiquitously permeate the sedimentary subseafloor biosphere, their combined biogeochemical activities most likely make a crucial contribution to deep diagenesis and sequestration of buried organic matter. Subsurface “archaeology” with genomic and cultivation approaches will continue to provide new insights into the evolutionary depth, genomic potential, physiological repertoire and biogeochemical function of the subseafloor biosphere.

Acknowledgements Andreas Teske was supported by the NSF Science and Technology Center “Dark Energy Biosphere Investigations” (C-DEBI); this is C-DEBI publication No. 199. Subsurface microbiology research in the Teske Lab was supported by NSF, the NASA Astrobiology Institute and ODP/IODP over the last 15 years.

References [1] [2] [3] [4] [5] [6] [7] [8]

Woese CR, Kandler O, Wheelis ML. Towards a natural system of organisms: Proposal for the domains Archaea, Bacteria and Eucarya. Proc Natl Acad Sci USA 87 (1990), 4576–4579. Woese CR. Bacterial evolution. Microbiol Rev 51 (1987), 221–271. Woese CR. The universal ancestor. Proc Natl Acad Sci USA 95 (1998), 6854–6859. Woese CR, Fox GE. The phylogenetic structure of the prokaryotic domains. Proc Natl Acad Sci USA 74 (1977), 5088–5090. Fox GE, Stackebrandt E, Hespell RB, et al. The phylogeny of prokaryotes. Science 209 (1980), 457–463. Bult CJ, two more, et al. Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii. Science 273 (1996), 1058–1073. Pace NR. A molecular view of microbial diversity and the biosphere. Science 276 (1997), 734– 740. Hugenholtz P. Exploring prokaryotic diversity in the genomic era. Genome Biology 3 (2002), reviews0003.1–0003.8

156 | 7 Archaea in deep marine subsurface sediments [9] [10]

[11] [12] [13] [14]

[15] [16]

[17] [18] [19] [20]

[21]

[22]

[23] [24] [25]

[26] [27] [28] [29]

Schleper C, Jurgens G, Jonuscheit M. Genomic studies of uncultivated archaea. Nature Rev Microbiol 3 (2005), 479–488. Barns SM, Delwiche CF, Palmer JD, Pace NR. Perspectives on archaeal diversity, thermophily and monophyly from environmental rRNA sequences. Proc Natl Acad Sci USA 93 (1996), 9188–9193. Takai K, Horikoshi K. Genetic diversity of Archaea in Deep-Sea hydrothermal Vent Environments. Genetics 152 (1999), 1284–1297. Brochier-Armanet C, Boussau B, Gribaldo S, Forterre P. Mesophilic crenarchaeota: proposal for a third archaeal phylum, the Thaumarchaeota. Nature Rev Microbiol 6 (2008), 245–252. Huber H, Hohn MJ, Rachel R, Fuchs T, Wimmer VC, Stetter KO. A new phylum of Archaea represented by a nanosized hyperthermophilic symbiont. Nature 417 (2002), 63–67. Vetriani C, Jannasch HW, MacGregor BJ, Stahl DA, Reysenbach AL. Population structure and phylogenetic characterization of marine benthic archaea in deep-sea sediments. Appl Environ Microbiol 65 (1999), 4375–4384. Rinke C, Schwientek P, Sczyrba A, et al. Insights into the phylogeny and coding potential of microbial dark matter. Nature 499 (2013), 431–437. Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16 213– 16 216. D’Hondt S, Spivack AJ, Pockalny R, et al. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106 (2009), 11 651–11 656. Røy H, Kallmeyer J, Adhikari RR, Pockalny R, Jørgensen BB, D’Hondt S. Aerobic microbial respiration in 86-million-year-old deep-sea red clay. Science 336 (2012), 922–925. Hinrichs KU, Inagaki F. Downsizing the Deep Biosphere. Science 338 (2012), 204–205. Rochelle PA, Cragg BA, Fry JC, Parker RJ, Weightman AJ. Effects of sample handling on estimation of bacterial diversity in marine sediments by 16S rRNA gene sequence analysis. FEMS Microbiol Ecol 15 (1994), 215–225. Newberry CJ, Webster G, Weightman AJ, Fry JC. Diversity of prokaryotes and methanogenesis in deep subsurface sediments from the Nankai Trough, Ocean Drilling Program Leg 190. Environ Microbiol 6 (2004), 274–287. Kormas AK, Smith DC, Edgcomb V, Teske A. Molecular analysis of deep subsurface microbial communities in Nankai Trough sediments (ODP Leg 190, Site 1176). FEMS Microbiol Ecol 45 (2003), 115–125. Teske AP. Microbial communities of deep marine subsurface sediments: molecular and cultivation surveys. Geomicrobiol J 23 (2006), 357–368. Biddle JF, Lipp JS, Lever MA, et al. Heterotrophic Archaea dominate sedimentary subsurface ecosystems off Peru. Proc Natl Acad Sci USA 103 (2006), 3846–3851. Inagaki F, Nunoura T, Nakagawa S, et al. Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments on the Pacific Ocean Margin. Proc Natl Acad Sci USA 103 (2006), 2815–2820. Parkes RJ, Webster G, Cragg BA, et al. Deep subseafloor prokaryotes stimulated at interfaces over geological time. Nature 436 (2005), 390–394. Sørensen KB, Teske A. Stratified communities of active archaea in deep marine subsurface sediments. Appl Environ Microbiol 72 (2006), 4596–4603. Edgcomb VP, Beaudoin D, Gast R, Biddle JF, Teske A. Marine subsurface Eukaryotes: The fungal majority. Environ Microbiol 13 (2011), 172–183. Sørensen, KB, Lauer A, Teske A. Archaeal phylotypes in a metal-rich, low-activity deep subsurface sediment of the Peru Basin, ODP Leg 201, Ste 123. Geobiology 2 (2004), 151–161.

References | 157

[30]

[31] [32] [33]

[34] [35] [36]

[37]

[38] [39] [40]

[41]

[42]

[43] [44] [45] [46] [47] [48]

[49]

Nunoura T, Inagaki F, Delwiche ME, Colwell FS, Takai K. Subseafloor microbial communities in methane-hydrate bearing sediments at two distinct locations (ODP Leg 204) in the Cascadia Margin. Microbes Environments 23 (2008), 317–325. Roussel EG, Cambon-Bonavita MA, Querellou J, Cragg BA, Webster G, Prieur D, Parkes JR. Extending the sub-sea-floor biosphere. Science 320 (2008), 1046. Coolen MJL, Cypionka H, Sass AM, Sass H, Overmann J. Ongoing modification of Mediterranean Pleistocene sapropels mediated by prokaryotes. Science 296 (2002), 2407–2410. Roussel EG, Sauvadet AL, Chaduteau C, Fouquet Y, Charlou JL, Prieur D, Cambon Bonavita MA. Archaeal communities associated with shallow to deep subseafloor sediments of New Caledonia Basin. Environ Microbiol 11 (2009), 2446–2462. Wang P, Li T, Hu A, Wei Y, Guo W, Jiao N, Zhang C. Community structure of archaea from deepsea sediments of the South China Sea. Microb Ecol 60 (2010), 796–806. Inagaki F, et al. Microbial communities associated with geological horizons in coastal subseafloor sediments from the Sea of Okhotsk. Appl Environ Microbiol 69 (2003), 7224–7235. Reed D, Fujita Y, Delwiche ME, Blackwelder DB, Sheridan PP, Uchida T, Colwell FS. Microbial communities from methane hydrate-bearing deep marine sediments in a forearc basin. Appl Environ Microbiol 68 (2002), 3759–3770. Lever MA. 2008. Anaerobic carbon cycling pathways in the subseafloor investigated via functional genes, chemical gradients, stable carbon isotopes and thermodynamic calculations. PhD thesis, The University of North Carolina at Chapel Hill, Dept. of Marine Sciences. Nakagawa S, Inagaki F, Suzuki Y, et al. Microbial Community in Black Rust Exposed to Hot Ridge Flank Crustal Fluids. Appl Environ Microbiol 72 (2006), 6789–6799. Forschner SR, Sheffer R, Rowley DC, Smith DC. Microbial diversity in Cenozoic sediments from the Lomonosov Ridge in the central Arctic Ocean. Environ Microbiol 11 (2009), 630–639. Webster G, Blazejak A, Cragg BA, et al. Subsurface microbiology and biogeochemistry of a deep, cold-water carbonate mound from the Porcupine Seabight (IODP Expedition 307). Environ Microbiol 11 (2009), 239–257. Hoshino T, Morono Y, Terada T, Imachi H, Ferdelman TG, Inagaki F. Comparative study of subseafloor microbial community structures in deeply buried coral fossils and sediment matrices from the Challenger Mound in the Porcupine Seabight. Frontiers in Microbiology 2 (2011), 231. doi:10.3389/fmicb.2011.00231 Nunoura T, Soffientino B, Blazejak A, Kakuta J, Oida H, Schippers A, Takai K. Subseafloor microbial communities associated with rapid turbidite deposition in the Gulf of Mexico continental slope (IODP Expedition 308). FEMS Microbiol Ecol 69 (2009), 410–424. Durbin AM, Teske A. Sediment-Associated Microdiversity within the Marine Group I Crenarchaeota. Environ Microbiol Reports 2 (2010), 693–703. Durbin AM, Teske A. Microbial diversity and stratification of South Pacific abyssal marine sediments. Environ Microbiol 13 (2011), 3219–3234. Fry JC, Parkes RJ, Cragg BA, Weightman AJ, Webster G. Prokaryotic biodiversity and activity in the deep subseafloor biosphere. FEMS Microbiol Ecol 66 (2008), 181–196. Teske A, Sørensen KB. Uncultured Archaea in deep marine subsurface sediments: have we caught them all? ISME J 2 (2008), 3–18. Orcutt BN, Sylvan JB, Knab NJ, Edwards KJ. Microbial ecology of the dark ocean above, at and below the seafloor. Microbiol Mol Biol Rev 75 (2011), 361–422. Durbin AM, Teske A. Archaea in organic-lean and organic-rich marine subsurface sediments: an environmental gradient reflected in distinct phylogenetic lineages. Frontiers in Microbiology 3 (2012), 168, DOI:10.3389/fmicb.2012.00168. Takai K, Moser DP, DeFlaun M, Onstott TC, Fredrickson JK. Archaeal diversity in waters from deep South African gold mines. Appl Environ Microbiol 67 (2001), 5750–5760.

158 | 7 Archaea in deep marine subsurface sediments [50]

[51]

[52] [53]

[54]

[55] [56] [57] [58] [59]

[60]

[61]

[62] [63]

[64] [65] [66]

[67]

[68]

Li PY, Xie BB, Zhang XY, et al. Genetic structure of three fosmid-fragments encoding 16S rRNA genes of the miscellaneous Crenarchaeotic Group (MCG): implications for physiology and evolution of marine sedimentary archaea. Environ Microbiol 14 (2012), 467–479. Kubo K, Lloyd KG, Biddle JF, Amann R, Teske A, Knittel K. Archaea of the Miscellaneous Crenarchaeotal Group (MCG) are abundant, diverse and widespread in marine sediments. ISME J 6 (2012), 1949–1965. Lloyd KG, Schreiber L, Petersen DG, et al. Predominant archaea in marine sediments degrade detrital proteins. Nature 496 (2013), 215–218. Webster G, Rinna J, Roussel EG, Fry JC, Weightman AJ, Parkes RJ. Prokaryotic functional diversity in different biogeochemical depth zones in tidal sediments of the Severn Estuary, UK, revealed by stable-isotope probing. FEMS Microbiol Ecol 72 (2010), 179–197. Gagen EJ, Huber H, Meador T, Hinrichs KU, Thomm M. A novel cultivation-based approach for understanding the Miscellaneous Crenarchaeotic Group (MCG) Archaea from sedimentary ecosystems. Appl Environ Microbiol 79 (2013), 6400–6406. Lipp JS, Morono Y, Inagaki F, Hinrichs KU. Significant contribution of Archaea to extant biomass in marine subsurface sediments. Nature 454 (2008), 991–994. Morono Y, Terada T, Nishizawa M, et al. Carbon and nitrogen assimilation in deep subseafloor microbial cells. Proc Natl Acad Sci USA 108 (2011), 18 295–18 300. Knittel K, Boetius A. Anaerobic oxidation of methane: progress with an unknown process. Annu Rev Microbiol 63 (2009), 311–334. D’Hondt S, Rutherford S, Spivack AJ. Metabolic activity of subsurface life in deep-sea sediments. Science 295 (2002), 2067–2070. Webster G, Parkes RJ, Cragg BA, Newberry CJ, Weightman AJ, Fry JC. Prokaryotic community composition and biogeochemical processes in deep subseafloor sediments from the Peru Margin. FEMS Microbiol Ecol 58 (2006), 65–85. Lloyd KG, Alperin M, Teske A. Environmental evidence for net methane production and oxidation in putative Anaerobic MEthanotrophic (ANME) archaea. Environ Microbiol 13 (2011), 2548–2564. Ziebis W, McManus J, Ferdelman TG, et al. Interstitial fluid chemistry of sediments underlying the North Atlantic gyre and the influence of subsurface fluid flow. Earth Planet Sci Lett 323/324 (2012), 79–91. D’Hondt S, Jørgensen BB, Miller DJ, et al. Distribution of microbial activities in deep subseafloor sediments. Science 306 (2004), 2216–2221. Inagaki F, Takai K, Komatsu T, Kanamatsu T, Fujiioka K, Horikoshi K. Archaeology of archaea: geomicrobiological record of pleistocene thermal events concealed in a deep-sea subseafloor environment. Extremophiles 5 (2001), 385–392. Chapelle FH, O’Neill K, Bradley PM, et al. A hydrogen-based subsurface microbial community dominated by methanogens. Nature 415 (2002), 312–315. Pester M, Schleper C, Wagner M. The Thaumarchaeota: an emerging view of their phylogeny and ecophysiology. Curr Opin Microbiol 14 (2011), 300–306. Kotelnikova A, Macario AJL, Pedersen K. Methanobacterium subterraneum sp. nov., a new alkaliphilic, eurythermic and halotolerant methanogen isolated from deep granitic groundwater. Int J Syst Evol Microbiol 48 (1998), 357–367. Mikucki JA, Liu Y, Delwiche M, Colwelln FS, Boone DR. Isolation of a methanogen from deep marine sediments that contain methane hydrates and description of Methanoculleus submarinus sp. nov. Appl Environ Microbiol 69 (2003), 3311–3316. Kendall MM, Liu Y, Sieprawska-Lupa M, Stetter KO, Whitman WB, Boone DR. Methanococcus aeolicus sp. nov., a mesophilic, methanogenic archaeon from shallow and deep marine sediments. Int J Syst Evol Microbiol 56 (2006), 1525–1529.

References | 159

[69] [70]

[71]

[72]

[73]

[74] [75] [76] [77] [78]

[79] [80]

[81]

[82] [83] [84]

[85]

[86]

Wu SY, Lai MC. Methanogenic archaea isolated from Taiwan’s Chelungpu fault. Appl Environ Microbiol 77 (2008), 830–838. Mochimaru H, Tamaki H, Hanada S, Imachi H, Nakamura K, Sakata S, Kamagata Y. Methanolobus profundi sp. nov., a methylotrophic methanogen isolated from deep subsurface sediments in a natural gas field. Int J Syst Evol Microbiol 59 (2009), 714–718. Shimizu S, Upadhye R, Ishijima Y, Naganuma T. Methanosarcina horonobensis sp. nov., a methanogenic archaeon isolated from a deep subsurface Miocene formation. Int J Syst Evol Microbiol 61 (2011), 2503–2507. Watkins AJ, Roussel EG, Webster G, Parkes RJ, Sass H. Choline and N,Ndimethylethanolamine as direct substrates for methanogens. Appl Environ Microbiol 78 (2012), 8298–8303. Singh N, Kendall MM, Liu Y, Boone DR. Isolation and characterization of methylotrophic methanogens from anoxic marine sediments in Skan Bay, Alaska: description of Methanococcoides alaskense sp. nov., and emended description of Methanosarcina baltica. Int J Syst Evol Microbiol 55 (2005), 2531–2538. Lever MA. Functional gene surveys from ocean drilling expeditions – a review and perspective. FEMS Microbiol Ecol 84 (2013), 1–23. Friedrich MW. Methyl-coenzyme M reductase genes: unique functional markers for methanogenic and anaerobic methane-oxidizing Archaea. Meth Enzymol 26 (2005), 428–442. Hallam SJ, Putnam N, Preston CM, et al. Reverse methanogenesis: testing the hypothesis with environmental genomics. Science 305 (2004), 1457–1462. Shima S, Krüger M, Weinert T, et al. Structure of a methylcoenzyme M reductase from Black Sea mats that oxidize methane anaerobically. Nature 481 (2012), 98–101. Alperin MJ, Hoehler TM. Anaerobic methane oxidation by archaea/sulfate-reducing bacteria aggregates: 1. Thermodynamic and physical constraints. Amer J Science 309 (2009), 869– 957. Alperin MJ, Hoehler TM. Anaerobic methane oxidation by archaea/sulfate-reducing bacteria aggregates: 2. Isotopic constraints. Amer J Science 309 (2009), 958–984. Holler T, Wegener G, Niemann H, et al. Carbon and sulfur back flux during anaerobic microbial oxidation of methane and coupled sulfate reduction. Proc Natl Acad Sci USA 108 (2011), E1484–1490. Yoshioka H, Maruyama A, Nakamura T, et al. Activities and distribution of methanogenic and methane-oxidizing microbes in marine sediments from the Cascadia Margin. Geobiology 8 (2010), 223–233. Imachi H, Aoi K, Tasumi E, et al. Cultivation of methanogenic community from subseafloor sediments using a continuous-flow bioreactor. ISME J 5 (2011), 1913–1925. Lever MA, Rouxel O, Alt JC, et al. 2013. Evidence for microbial carbon and sulfur cycling in deeply buried ridge flank basalt. Science 339 (2013), 1305–1308. Colwell FS, Boyd S, Delwiche ME, Reed DW, Phelps TJ, Newby DT. Estimates of biogenic methane production rates in deep marine sediments at Hydrate Ridge, Cascadia Margin. Appl Environ Microbiol 74 (2008), 3444–3452. Springer E, Sachs MS, Woese CR, Boone DR. Partial gene sequences for the A subunit of methyl-coenzyme M reductase (mcrI) as a phylogenetic tool for the family Methanosarcinaceae. Int J Syst Bacteriol 45 (1995), 554–559. Hales BA, Edwards C, Ritchie DA, Hall D, Pickup RW, Saunders JR. Isolation and identification of methanogen-specific DNA from blanket bog peat by PCR amplification and sequence analysis. Appl Environ Microbiol 62 (1996), 668–675.

160 | 7 Archaea in deep marine subsurface sediments [87]

Luton PE, Wayne JM, Sharp RJ, Riley PW. The mcrA gene as an alternative to 16S rRNA in the phylogenetic analysis of methanogen populations in landfill. Microbiol 148 (2001), 3521– 3530. [88] Biddle JF, Fitz-Gibbon S, Schuster SC, Brenchley JE, House CH. Metagenomic signatures of the Peru Margin subseafloor biosphere show a genetically distinct environment. Proc Natl Acad Sci USA 105 (2008), 10 583–10 588. [89] Whitman WB, Bowen TL, Boone DR. The Methanogenic Bacteria. In: Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E, eds. The Prokaryotes, 3𝑟𝑑 edition. Springer, New York, NY Vol. 3, 165–207, 2006. [90] Ufnar JA, Wang SY, Christiansen JM, Yampara-Iquise H, Carson CA, Ellender RD Detection of the nifH gene of Methanobrevibacter smithii: a potential tool to identify sewage pollution in recreational waters. J Appl Microbiol 101 (2006), 44–52. [91] Lever MA, Alperin M, Inagaki F, et al. Trends in basalt and sediment core contamination during IODP Expedition 301. Geomicrobiol J 23 (2006), 517–530. [92] Valentine D. Adaptation to the energy stress dictate the ecology and evolution of the Archaea. Nature Rev Microbiol 316 (2007), 316–323. [93] Takano Y, Chikaraishi Y, Ogawa NO, et al. Sedimentary membrane lipids recycled by deep-sea benthic archaea. Nature Geosci 3 (2010), 858–861. [94] Orsi WD, Edgcomb VP, Christman GD, Biddle JF. Gene expression in the deep biosphere. Nature 499 (2013), 205–208. [95] Lomstein BA, Langerhuus AT, D’Hondt S, et al. Endospore abundance, microbial growth and necromass turnover in deep subseafloor sediment. Nature 484 (2013), 101–104. [96] Hoehler TM, Jørgensen BB. Microbial life under extreme energy limitation. Nature Microbiol Rev 11 (2013), 83–94. [97] Lloyd, KG, May M, Kevorkian R, Steen AD. Meta-analysis of quantification methods shows archaea and bacteria to be similarly abundant in the subseafloor. Appl Environ Microbiol 79 (2013), 7790–7799. [98] Swofford DL. PAUP*. Phylogenetic analysis using parsimony (and other methods), version 4. 2000. Sinauer Associates, Sunderland, Massachusetts. [99] DeLong EF. Archaea in coastal marine environments. Proc Natl Acad Sci USA 89 (1992), 5685– 5689. [100] López-García P, Moreira D, Lopez-Lopez A, Rodriguez-Valera F. A novel haloarchaeal-related lineage is widely distributed in deep oceanic regions. Environ Microbiol 3 (2001), 72–78. [101] Jurgens G, Glöckner FO, Amann R, Saano A, Montonen L, Likolammi M, Munster U. Identification of novel Archaea in bacterioplankton of a boreal forest lake by phylogenetic analysis and fluorescent in situin situ hybridization. FEMS Microbiol Ecol 34 (2000), 45–56.

Bernard Ollivier, Jean Borgomano, and Philippe Oger

8 Petroleum: from formation to microbiology 8.1 Introduction Economic petroleum accumulations result from the migration in the lithosphere of hydrocarbons formed by the maturation of kerogen. This is a long-term process mainly controlled by temperature and pressure conditions that depend on the geological evolution of the hosting sedimentary basin. Cultural and molecular approaches both reveal that petroleum reservoirs are inhabited by a wide range of mesophilic and thermophilic/hyperthermophilic anaerobic microorganisms able to ferment or oxidize their substrates. Among them sulfatereducing prokaryotes and methanoarchaea are recognized as important contributors to the geomicrobiology of these ecosystems because of their hydrogenotrophic and/or acetoclastic nature. Some of them together with members of the Thermotogales, and other fermentative prokaryotes are considered as indigenous to oil reservoirs. Taking into account the high microbial metabolic diversity existing in oil reservoirs, we may expect the activity of many of these oilfield microorganisms to be of interest in enhancing oil recovery in the next future.

8.2 Petroleum formation Petroleum is a general term for a mixture of hydrocarbons that exist naturally on Earth in gaseous (natural gas), liquid (crude oil) and solid (asphalt) states and are characterized by density lower than that of water [1]. Typical compounds of petroleum contain carbon and hydrogen (C𝑛 H𝑚 ) with small amounts of oxygen, nitrogen and sulfur. Petroleum occurs generally in sedimentary rocks formed under marine or terrestrial environments and buried in the external envelop of the lithosphere (󳶳 Fig. 8.1 (a)). Its complex nature is evidenced by the high numbers (> 1000) of hydrocarbon species that have been identified since petroleum exploitation started in the 19𝑡ℎ century [2]. Under natural conditions, petroleum gas is composed by saturated hydrocarbons (alkanes) with one to three carbon atoms. Petroleum oil is a complex mixture of alkanes with more than four carbon atoms, such as paraffin, cyclo-alkanes and aromatics [3]. Sulfur, nitrogen and oxygen can be present in oil within complex organic molecules such as resins and asphalts. The presence of biomarkers in petroleum (e.g. elements of chlorophyll or cellulose for example) demonstrated the organic origin of oil and gas despite ancient but famous inorganic theories by Porfir’ev [4] and Gold and Soter [5].

162 | 8 Petroleum: from formation to microbiology (a) Photosynthesis CO2 + H2O

CH2O + O2

Sea level

Growth of coral reef

al of teri ma ort nsp ganic a r T or nd nt a ime sed

Limestone (reservoir) oxidated 1 reduced 2 Organic material

Sand (reservoir)

Grey clay and sand

Black mud (source rock)

1 km Lithospheric basement

Hydrostatic pressure (bas)

(b) Sea level

1

1 mm

t se Curren

m

a botto

Porosity Sediment accumulation (Shales)

5 Ma SB Rock

Seal Oil eld

80°C

3

25 M

Maturation of source rock

a SB

Oil migration B

50

M

aS

Oil migration

150°C Subsidence

1 km Heat ow

600

1 : Aerobic microbial respiration 2 : Fermentation and anaerobic microbial respirations 3 : Abiotic (and biotic?) transformation of organic matter

Fig. 8.1: Geological cross-sections illustrating general principals of a petroleum system. (a) Formation and preservation of organic rich sediment (source rock) and reservoir rock. Superficial sedimentary processes allow the concentration and preservation of organic matter in subaqueous conditions. (b) Geological evolution of a sedimentary basin hosting a petroleum system. Under the influence of lithospheric subsidence and heat flow the deposited source rock is preserved from rapid superficial recycling and buried to the right temperature-pressure conditions for oil generation and updip migration in the reservoir rock. The dominant reservoir pore system is of millimeter dimensions allowing strong tension surface forces (rock-fluid interaction) and possible development of biofilms.

8.2 Petroleum formation

| 163

Petroleum formation can be considered as a set of physico-chemico-biological processes within the general organic carbon cycle on Earth, from life, built with elementary chemical components (C,H,O,N), to the production of organic matter and its ultimate decomposition into chemical elements taken by life, thus looping the cycle [6]. This cycle operates in the external Earth envelops, under the influence of solar and earth mantel energy, and can be separated in several interacting subcycles within the atmosphere, the hydrosphere, the biosphere and the lithosphere. Microbial life has an important role in this cycle, especially at surface under aerobic conditions or beneath the surface under anaerobic conditions (e.g. nitrate-, iron-, sulfate reduction and methanogenesis). These subcycles can be very fast (day–year) on the earth surface, in soils or coral reefs, . . . or very long (> 109 years) when they occur in the shallowest part of the lithosphere (0–15 km), in sedimentary basins, and involves the formation and the preservation of petroleum pools over geological times [7]. The interaction between the organic carbon cycle (or the life cycle in general) and geological factors control the formation of petroleum, especially the geological processes which contribute to the burial and preservation of organic matter and petroleum in the lithosphere (󳶳 Fig. 8.1 (b)). Thus, petroleum is not a required step of the organic carbon cycle and it is generally bypassed by the fast looping recycling of organic matter occurring on the earth surface where microbial oxido-reductive activities are important. Indeed the huge volume of liquid petroleum (> 1010 tons, as estimated by Husseini [8]), preserved in the lithosphere, indicates that the destruction processes (bio-physicochemical) are inhibited during very long period of time of the earth evolution. The oldest oil accumulations are found in Proterozoic reservoirs in China (1.4 × 109 years) [9], whereas the oldest oil drop has been dated of 3 × 109 years in Australia [10]. So the formation of hydrocarbon needs life and energy (from sun and earth), but its preservation in the lithosphere requires the creation of subsurface pools, protected from oxidation and biodegradation processes with stable geological conditions. Hereafter, we discuss in more details the petroleum system to provide a framework to discuss the microbial activities in petroleum reservoirs.

8.2.1 Petroleum system The concept of the petroleum system [11] is founded on the integration of all the main geological parameters that contribute to the formation and the preservation of economic petroleum accumulations in the lithosphere. A typical petroleum system can be characterized by two elements, linked by bio-chemico-physical processes over time (󳶳 Fig. 8.1 (b)): (i) the source rock, sedimentary organic-rich material at the origin of the petroleum and (ii) the reservoir, sedimentary rock with porosity that contain the petroleum in a complex geological structure called “trap”.

164 | 8 Petroleum: from formation to microbiology 8.2.1.1 Formation of source rock Source rocks are formed by fine-grained sediments (mud-silt), rich in organic matter, that are deposited in subaqueous environments (seas, lakes, ponds, . . . ) and are preserved from degradation processes (abiotic or biotic) [1]. This organic matter is produced mainly by phytoplankton, prokaryotes (seas or lakes) and land plants. Before the land plants appeared in the Devonian (400 × 106 years ago), phytoplankton was the main organic component of source rocks. The protection from degradation is afforded by (i) anaerobic conditions and (ii) high rate of sedimentation (> mm/year) of inorganic material (mud-silt) to bury the organic matter. Such sedimentary rock can contain on average more than 0.5 % of organic matter. Within the first meters of burial, after depletion of oxygen by aerobes/facultative aerobes and other terminal electron acceptors (e.g. nitrate, Fe(III), sulfate) by facultative aerobes/anaerobes, organic matter is degraded by anaerobes to produce methane. During further burial and slight increases in temperature (+/−50 °C) and pressure (+/−50 bars), microbial degradation, condensation and polymerization partially convert the organic matter (biopolymer) into kerogen [2]. The kerogen, a complex association of dense organic material (fulvic-humic acids, humin) dominated by carbon and oxygen, is the precursor for petroleum. The critical processes in the formation of a source rock are the depositions of organic matter-rich sediments in anoxic subaqueous conditions and the rapid burial of these sediments in subsiding parts of the lithosphere, such as sedimentary basins.

8.2.1.2 Formation of reservoir rock Reservoir rocks are sedimentary rocks formed by physical, chemical and/or biological processes operating at the surface of the earth (sea and land) and resulting in the accumulation of solid grains, mineral, lithic and bioskeletal. They can be separated into two main groups: silico-clastics produced by hydrodynamics (rivers, waves, . . . ) and carbonates related to (marine) biological production of calcium carbonate (coral reef) (󳶳 Fig. 8.1 (a)). The pore space is either preserved from primary deposition features (intergranular) or created by chemical dissolution (vugs, karsts) or tectonic deformation (fracture). In petroleum reservoirs, porosity is generally ranging between 50–5% of the total rock volume. Many physico-chemical processes (diagenesis) can increase or decrease the porosity of the reservoir from the deposition time (at surface) to the depth and time of petroleum emplacement. The porosity (pore space/bulk reservoir rock volume) controls the volume of hydrocarbon trapped in the subsurface at the time of the petroleum emplacement in the trap. For microbiological considerations, it is important to bear in mind that the dominant pore size in petroleum accumulations is in the order of 1 mm in diameter, resulting in significant capillary forces and very high specific surface area that favors the solid-fluid interactions and the formation of biocoatings. In general, porosity decreases exponentially from 50% to 0% (compactioncementation), between the surface and 6,000 m burial, but few percent of porosity

8.2 Petroleum formation

| 165

have been found at 10,000 m depth [12]. Pore types (geometry and topology) probably have a significant effect on microbial activities in the reservoir but this parameter is not fully understood.

8.2.1.3 Trap formation The trap is the geological structure that allows the accumulation of petroleum in the subsurface, by preventing upwards migration (buoyancy) of the hydrocarbon from the deeper source rock to the earth surface and by concentrating the petroleum in discrete volume of rock that can be drilled and exploited. Simple trap generally consists of a tectonic structure (fold) formed by a sealing impermeable rock (shale or salt) overlying the reservoir rock (󳶳 Fig. 8.1 (b)). More complex trap exists but the principle is always identical: a geological object that can retain and concentrate a huge volume of hydrocarbon locally. Trap volume, define by a spill-point and the overall geometry of the structure can reach several billions of cubic meters. Porosity and hydrocarbon saturation (% of pore volume occupied by petroleum) determine the volume of petroleum in place (up to billions of cubic meter) in a single accumulation (nearly 35 × 109 m3 of oil for the Ghawar Field in Saudi Arabia, the largest oilfield in the world) [13].

8.2.1.4 From source rocks to reservoirs: oil generation and migration Three key processes link the two endmembers of the petroleum system, source and reservoir rocks: (i) the maturation of the source rock, (ii) the generation of petroleum and (iii) the migration of petroleum in the trap-reservoir [2]. These processes start just after the initial burial of the rock (hundreds of meters) and can last several millions of years depending on the geological history of the sedimentary basin. The maturation of the source rock refers to the progressive heating of the rock with increasing depth depending on the local rate of subsidence and geothermal gradient (󳶳 Fig. 8.1 (b)). During this process, the kerogen is transformed into petroleum by temperature-related fractioning of the organic compounds into oil and gas. Experimental and subsurface data indicate that the first oil and gas appear just below 50 °C, the gas having some C4 C10 compounds (wet gas), whereas the peak of oil production is around 100 °C. Above this temperature, the amount of gas and the gas-oil-ratio (GOR) increase, and above 150 °C (gas production peak) all the remaining oil and kerogen are cracked into gas with only C1 -C3 compounds (dry gas). This maturation process can continue until all kerogen has been converted into petroleum (+/− 200 °C). Increasing the temperature further (> 200 °C) and the pressure could result in the removal of all organic compounds from the source rock tending to graphite (metagenesis). The generation of petroleum is correlated to the maturation of the source rock (so to its thermal evolution) but it also depends on the type of source rock (kerogen type, permeability, sedimentary heterogeneity, . . . ) and of the rate of heating. To simplify, the oil window is comprised between 60–150 °C and the gas window between 150–200 °C. This

166 | 8 Petroleum: from formation to microbiology process is considered to be controlled only by temperature and pressure in abiotic conditions without contribution of microbial life. However, taking into account that (i) many thermophilic/hyperthermophilic oilfield microorganisms pertaining to the domain Archaea or Bacteria may grow at temperatures ranging between 60 °C and 90 °C corresponding to the oil window, e.g. Archaeglobus fulgidus, we may hypothesize that they could be potential contributors to oil formation (󳶳 Fig. 8.1 (b)). Interestingly, hyperthermophilic anaerobic archaeons have recently been reported to use hydrocarbons as energy sources [14, 15]. The migration of petroleum into the reservoir and the trap is essentially a physical process that brings the hydrocarbon material from the source rock into the reservoir and trap. This process can last several million years because it consists of concentrating billions of m3 of petroleum into a reservoir from a dispersed volume of kerogen (source rock), ten to hundred times bigger, in a sedimentary basin. This migration occurs when the buoyancy forces of the petroleum (gas-oil) exceed the capillary forces in the source rock.

8.2.1.5 Preservation of petroleum accumulation A last important process is the preservation of the petroleum accumulation over geological time. Being some sort of accidents of the global organic carbon cycle, the petroleum systems are not in a state of equilibrium in the lithosphere and tend to be re-introduced in the global material cycles. The plate tectonics or geodynamics, with convergence, divergence, magmatism and the formation of mountain ranges or oceans is a major cause of destruction of petroleum reservoirs, by exhumation, heating or burial. Meteoric water flushing, aerobic biodegration, pressure increase, . . . can also be the causes for alteration of petroleum accumulation. The preservation of petroleum accumulation is probably the exception and not the rule over geological time.

8.3 Petroleum microbiology Despite the drastic physico-chemical conditions existing in oil reservoirs (e.g. high temperatures up to 190 °C in Elgin-Franklin Field-North Sea [16], high salinities up to 400 g/l in Verkhnechona Field-Russia [17]), there is now evidence that microorganisms inhabit such extreme environments considered as closed systems lacking oxygen [18, 19]. This is the case of many mesophilic, thermophilic/hyperthermophilic, and halophilic/hyperhalophilic anaerobic prokaryotes which have been retrieved both by molecular and cultural approaches at many occasions from such ecosystems [18]. However, it is noteworthy that the combination of high temperatures and high salinities is too hostile for life to occur in situ and to our knowledge, the only so far known moderately halophilic thermophilic anaerobes originated from terrestrial ecosystems (e.g. Halothermothrix orenii and Thermohalobacter berrensis) [20, 21]. Surprisingly, the lithostatic pressure, which is also considered as an extreme condition

8.3 Petroleum microbiology

| 167

and should have an impact on the geomicrobiology of deep hot oil reservoirs, has not been considered so far. Thus in contrast to the deep-sea environments, piezophilic microorganisms have not yet been isolated and characterized from subsurface oil reservoirs. Oilfield microbes have a wide range of metabolisms including fermentative and respiratory processes (e.g. sulfate-reduction by various prokaryotes, CO2 reduction by hydrogenotrophic methanoarchaea) with some sulfate-reducers and methanoarchaea having the ability to grow chemolithotrophically [18, 19]. Whether the presence of these microbes results from anthropogenic activities (e.g. seawater flooding of the reservoirs, drilling operations, migration through aquifers) or if they are indigenous to petroleum reservoirs remains questionable [22]. However (i) the optimal physicochemical conditions for growth of several isolated microorganisms fit exactly with that existing in the oil reservoir from where they originated, (ii) similar bacterial and/or archaeal species were recovered from geographically distinct oil reservoirs, and (iii) members of some genera (e.g. Petrotoga and Geotoga) have only been isolated from these subterrestrial ecosystems. These facts are in favor of the indigenous character of some of these microorganisms [22, 23]. In this respect, the challenges for microbiologists today are to demonstrate which microorganisms are indigenous, how long they have stayed in situ (e.g. if they were already present at deposit time of the original sediments in the sedimentary basin), what was their real geomicrobiological significance during oil formation, how they have lived or survived over millions of years and finally what activity and role they perform today? Petroleum reservoirs not only contain hydrocarbons (e.g. aliphatic or aromatic hydrocarbons, etc.) as potential substrates for microorganisms in the oil phase, but also a wide range of organic substrates which accumulate in formation waters and serve as carbon and energy sources for microorganisms. They include fatty acids and especially its main synthesis and degradation intermediate, acetate (up to 20 mM of acetate was found in some petroleum reservoirs), but also formate, propionate and butyrate, aromatic compounds (e.g. benzoate) [24], and more complex organic acids such as naphtenic acids with concentrations in crude oil up to 100 mM [19]. All these substrates may be oxidized under anaerobiosis in the presence of chemical terminal electron acceptors possibly present in oilfield waters (e.g. sulfate) but also by coupling with hydrogenotrophic microbial partners (e.g. hydrogenotrophic sulfatereducing prokaryotes or methanoarchaea) acting as biological terminal electron acceptors. Consequently, microorganisms responsible for the oxidation of some of these compounds (e.g. acetate) have been retrieved from oilfield waters by cultural approaches [18, 19]. However, whereas similar types of metabolism may be expected for hydrocarbon oxidation (e.g. sulfate-reduction or coupling with a prokaryotic hydrogen scavenger), attempts to enrich such microorganisms originating from oil reservoirs was rarely successful [25–27]. Hydrogen possibly produced from abiotic reactions in the deep subsurface may also be oxidized by numerous sulfate-reducing prokaryotes (SRP) and methanoarchaea (MA). Besides oxidative processes, fermentative ones may

168 | 8 Petroleum: from formation to microbiology occur from organic compounds such as amino acids which may accumulate in oil reservoirs during the kerogen maturation process [19]. Because of the availability of sulfate, and CO2 as potential terminal electron acceptors in oil reservoirs, but also that of acetate as electron donor, it is obvious that SRP [28] and MA [29] play a significant geomicrobiological role in these environments. This is particularly true for hydrogen-oxidizing SRP and MA reducing sulfate to sulfide and CO2 to methane, respectively. In addition, acetate may also be oxidized by acetotrophic SRP or fermented by acetoclastic MA. Therefore the use of hydrogen and/or acetate by different populations of microorganisms appear as a fundamental metabolism not only in formation waters but also in the oil phase since hydrocarbon oxidation also depends on the activity of hydrogen oxidizers (SRP or MA) and to a lesser extent on that of acetate users (SRP or MA) [30]. It is noteworthy that the production of acetate may also result from the oxidation of hydrogen coupled to the reduction of CO2 by homoacetogenic bacteria, but their presence in oil reservoirs has been rarely evidenced thus suggesting that homoacetogenesis is, most probably, not an effective metabolic process in situ [18].

8.3.1 The sulfate-reducing prokaryotes Many new SRP species that are known to be pernicious to the oil industry (e.g. responsible for biocorrosion, oil souring, etc.) have been isolated from oilfield ecosystems [18, 19, 28] (󳶳 Table 8.1). They include seven mesophilic (optimum range temperature between 30 and 40 °C) Desulfovibrio species within the Deltaproteobacteria and one mesophilic halophilic Desulfotomaculum species (D. halophilum), phylum Firmicutes, recovered from low-temperature oil reservoirs. Most of these SRPs share the ability to use hydrogen and to incompletely oxidize their organic substrates, e.g. lactate, ethanol, etc., to acetate. This is also the case for thermophilic SRP originating from hot oil reservoirs. They comprise three Desulfotomaculum species (spp.) (Firmicutes) and two Desulfacinum spp. (Deltaproteobacteria), growing optimally at temperatures around 60 °C, together with two Thermodesulfobacterium spp. (Thermodesulfobacteria) with optimal growth occurring at 65 °C (T. thermophilum) and 70 °C (T. commune). Besides all these bacterial species, only one archaeal thermophilic sulfate-reducing species (optimum temperature for growth 76 °C), Archaeoglobus fulgidus strain 7324, was isolated from a North Sea oil reservoir [31] where it was demonstrated to thrive with other thermophilic SRP [32]. In contrast to the type species of this genus (A. fulgidus strain Z) isolated from a shallow hydrothermal vent which was reported to use hydrogen autotrophically in the presence of thiosulfate, but not sulfate as terminal electron acceptor, strain 7324 was shown to use hydrogen only mixotrophically in the presence of isovalerate as electron donor and sulfate as electron acceptor [31]. Both strains completely oxidized their organic substrates. Recently, A. fulgidus was shown to oxidize long-chain fatty acids and alkenes [15, 33]. Complete oxidation of substrates

8.3 Petroleum microbiology

|

169

Table 8.1: Sulfate-reducing prokaryotes isolated from oilfields. Genus

species

Desulfovibrio

gabonensis gracilis longus bastinii capillatus vietnamensis alaskensis rhabdoformis apsheronum toluenicum halophilum kuznetsovii nigrificans thermocisternum infernum subterraneum vibrioformis cetonicum halophilus commune thermophilum norvegicus fulgidus

Desulfobulbus Desulfomicrobium Desulfotignun Desulfotomaculum

Desulfacinum Desulfobacter Desulfobacterium Desulfovermiculus Thermodesulfobacterium Thermodesulforhabdus Archaeoglobus

Hydogen oxidizera

Acetate oxidizera

+ + + + + ND ND ND − + (+) + + + + + − ND + + + − +/−

− − − − − − − − − + − + − − + + + + − − − + +

a

+, utilized; −, not utilized; (+), poorly utilized; (+/−) utilized only in the presence of thiosulfate, but not sulfate; ND, not determined

was demonstrated for several other oilfield SRPs including the mesophilic Desulfobacter vibrioformis and Desulfobacterium cetonicum (Deltaproteobacteria) and the thermophilic, Desulfotomaculum kuznetzovii (Firmicutes), Desulfacinum subterraneum, and Thermodesulforhabdus norvegicus (Deltaproteobacteria) [19]. Desulfovermiculus halophilus was shown not to use acetate, but to completely oxidized butyrate [34]. Desulfotignum toluenicum (strain H3T ), a SRP isolated from an oil reservoir model was able to degrade crude oil, toluene and long-chain fatty acids [35]. The observed metabolic diversity displayed by sulfate-reducing prokaryotes originating from oil reservoirs suggests that they are linked to geomicrobiological cycles and mineralization of the organic matter occurring in petroleum reservoirs. Observations by molecular analysis based on cloning and sequencing PCR-amplified 16S rRNA genes or dissimilatory sulfate reductase (dsrAB genes) also confirm the presence of SRPs. The latter are detected by molecular surveys in about two thirds of the oil reservoirs worldwide [36], at levels up to 5% of the total bacterial diversity (󳶳 Fig. 8.2). Sequences corresponding to the known cultivable SRPs genera from the Firmicutes and the Deltaproteobacteria, e.g. Desulfosporosinus, Desulfocapsa,

170 | 8 Petroleum: from formation to microbiology 100

80

Actinobacteria Bacteroidetes Firmicutes 60

α-proteobacteria β-proteobacteria γ-proteobacteria δ-proteobacteria

40

ε-proteobacteria Other Proteobacteria Deferribacteres

20

Synergistetes Spirochaetes Thermotogae

0

Other Bacteria 1

2

3

4

5

6

7

8

pw pw oil pw oil fw oil fw 70° 85° 30° 50° 47° Malaysia

10 11 12 13 14 15 16 17 18 19

Japan

Oil

Alaska

N. sea

N. sea

Canada

9

oil wh pw pw pw pw pw pw pw pw pw 74° 27° 32° 20° 37° 58° 6.5° China

Canada

Oil sands

Fig. 8.2: Structures of bacterial populations based on sequenced 16S rRNA gene libraries from oil and oil sands reservoirs. 1, 2: Troll oil formation in the North Sea [37, 52]; 3, 4: the Enermark oilfield in Canada [57, 80]; 5–8: the Bokor oilfield in Malaysia [51]; 9–10: the high-temperature Yabase oil reservoir in Japan [41]; 11: a biodegraded, mesothermic petroleum reservoir in the Schrader Bluff Formation of Alaska [43]; 12–15: the Kalamay and Huabei Oil Fields in China [42]; and 16–19: the Athabasca River Oil Sands in Canada [50]. Oil, crude oil; pw, production water; fw, formation water; wh, wellhead. Data for lanes 1, 3–10 redrawn from Meslé et al. [39].

Desulfomicrobium, Desulfobulbus, Desulfuromonas, Desulfovibrio, Desulfobulbus and Desulfotomaculum, are the most frequently retrieved SRP groups. Members of the family Desulfovibrionaceae and Desulfobacteriaceae were retrieved from mesothermic oil reservoirs [36–39], while Desulfotomaculum and Desulfobacter spp. were the most dominant in four Chinese hot petroleum reservoirs [40]. The presence of Desulfotamaculum spp. was correlated with temperature, depth and the concentration of acetate, propionate and sulfate, while Desulfobacter spp. together with Desulfomicrobium and Desulfobulbus spp. showed positive correlation with sulfur and salinity [40].

8.3 Petroleum microbiology

| 171

In addition to the diversity of cultivable SRPs, molecular studies showed the presence of novel putative SRP species from the Deltaproteobacteria belonging to different lineages such as the Desulfobacteraceae [41–43], as well as several sequences from Deltaproteobacteria and Firmicutes which are only remotely related to known SRPs so that their metabolism is difficult to ascertain. Interestingly, sequences corresponding to Archaeoglobulus fulgidus, isolated from a reservoir in the North Sea, were detected in a hot oil reservoir in Japan indicating that this species might be widely represented in subterranean hot oil reservoirs [31, 41].

8.3.2 The methanoarchaea Hydrogenotrophic and acetoclastic methanoarchaea (󳶳 Table 8.2) which are known to compete with SRP are also of geomicrobiological significance in oilfield environments. Mesophilic hydrogenotrophic MA which are known to participate actively in hydrocarbon oxidation in oilfield waters [36] comprise Methanobacterium bryantii, Methanocalculus halotolerans, and Methanoplanus petrolearius growing optimally at 37 °C whereas Methanobacterium ivanovii, Methanothermobacter thermoautotrophicus, and Methanoculleus receptaculi are considered as moderate thermophiles to thermophiles [19, 23, 29, 44]. Methanothermobacter crinale recently isolated from the Shengli oilfield in China showed the highest temperature for growth (85 °C) and should be considered as an hyperthermophilic archaeon [45]. Only one aceticlastic nonhydrogenotrophic MA species has been isolated from oil reservoirs so far, Methanosarcina mazei, which grows under mesothermic conditions [19]. Acetate can Table 8.2: Methanoarchaea isolated from oilfields. Genus

species

Methanobacterium

bryantii ivanovii thermoaggregans thermoautotrophicus crinale halotolerans thermolithotrophicus receptaculi petrolearius euhalobius mazei siciliae evestigatus shengliensis

Methanothermobacter Methanocalculus Methanothermococcus Methanoculleus Methanoplanus Methanohalophilus Methanosarcina Methanohalobium Methermicoccus a

+, utilized; −, not utilized

Hydrogen oxidizera

aceticlastica

methylotrophica

+ + + + + + + + + − − − − −

− − − − − − − − − − + − − −

− − − − − − − − − + + + + +

172 | 8 Petroleum: from formation to microbiology be directly used by this archaeon, but there is evidence that it can also be oxidized in high-temperature reservoirs by syntrophic bacteria using hydrogenotrophic MA as biological electron acceptors to produce methane [26, 46]. Besides hydrogenotrophic and acetoclastic MA, numerous archaeons producing methane only from methylated compounds (e.g. methylamines, methanol) were isolated from saline to hypersaline oil reservoirs [29]. They belong to the genera Methanosarcina (e.g. M. siciliae), Methanohalophilus (e.g. M. euhalobius), Methanohalobium (e.g. M. evestigatum), and Methermicoccus (e.g. M. shengliensis). Although the availability of methanol in formation waters remains largely hypothetical as it is mainly the result of pectin demethylation, it is highly probable that such MA ferment more particularly trimethylamine in situ. In terrestrial ecosystems, many halophilic anaerobic prokaryotes accumulate glycine betaine to face the osmotic stress [47]. A similar accumulation in anaerobic halophiles originating from oil reservoirs may possibly occur. Its release in oilfield waters may lead to its fermentation yielding both acetate and trimethylamine possibly used by MA. MA might as well be able to use other more complex methylated compounds as was recently shown for choline and dimethylethanolamine degradation in deep-sea sediments [48]. In contrast to hydrogen and acetate, there is no competition between SRB and MA for the use of these methylated compounds which may in turn favor methylotrophic methanogenesis in oil reservoirs. Molecular approaches confirm the large predominance of MA in oil reservoirs [29] regardless of the extent of their degradation (󳶳 Fig. 8.3), except in the hottest ones, such as the deepest parts of the Yabase oilfield in Japan, which local temperature is about 98 °C [41]. In these oil reservoirs, MA include hydrogenotrophic members of the order (i) Methanomicrobiales (e.g. Methanoculleus, Methanoregula, and Methanospirillum spp.) [26, 39, 42, 49, 50], (ii) Methanobacteriales (e.g. Methanobacterium and Methanothermobacter spp.) [26, 39, 49, 50], (iii) Methanococcales (e.g. Methanococcus spp.) [37, 41, 51], but also acetoclastic and/or methylotrophic members of the order Methanosarcinales (e.g. Methanosarcina, Methanomethylovorans spp.) [26, 39, 49, 50]. Molecular surveys demonstrate that methanogenesis in the oil compartment and the associated water reservoirs can be driven by different MA exhibiting different methanogenic pathways. In high-temperature oil reservoirs, hydrogenotrophic methanoarchaeal sequences, closely related to that of oilfield isolates, e.g. Methanothermobacterium thermautotrophicus, Methanobacterium subterraneum, Methanothermococcus or Methanococcus spp., are dominating [37, 41, 52]. Hydrogenotrophs closely related to cultivated species also dominate in some low-temperature oil reservoirs [53, 54]. In many reservoirs, molecular surveys also reveal a diversity of yet uncultivated methanoarchaea belonging to four main classes, e.g. Methanosarcinales, Methanomicrobiales, Methanobacteriales, and Methanococcales. Obligate methylotrophic methanogens from the genus Methanolobus dominate in the oil phase of the Bokor oilfield (Malaysia) [51]. Although a Methanolobus sp. was isolated from a gas field environment and a coal seam, none has been isolated from an oil reservoir so far [55, 56]. In contrast, the formation waters from

8.3 Petroleum microbiology

| 173

100

80

60

40

Methanosarcinales Methanomicrobiales Methanobacteriales

20

Methanococcales Hyperthermophiles Other Euryarchaeota

0

Crenarchaeota 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 pw pw oil pw oil fw oil fw oil wh pw pw pw pw pw pw pw pw pw 70° 85°

30°

50°

47°

Malaysia

74°

Japan

Oil

27° 32° 20° 37° 58°

Alaska

N. sea

N. sea

Canada

China

6.5°

Canada

Oil sands

Fig. 8.3: Structures of archaeal populations based on sequenced 16S rRNA gene libraries from oil and oil sands reservoirs. Legends are the same as in 󳶳 Fig. 8.2.

the same field harbor almost exclusively the hydrogenotrophic Methanococcales, Methanococcus maripaludis and M. thermolithoautotrophicus, which suggest that different pathways lead to methane in these oilfields. Similarly, in the mesophilic Enermark oilfield (Canada), the obligate acetoclastic Methanosaeta accounts for 50% of the archaeal sequence reads in the water-associated samples, while the diversity in the oil-associated samples was equally distributed between a genus of obligate methylotrophs (Methanolobus), a genus of hydrogenotrophs (Methanobacterium) and a genus of acetotrophs (Methanosaeta) [57]. Methanosaeta also dominates the archaeal diversity in the production water of a mesophilic oil reservoir in Alaska [43] suggesting the possibility for acetoclastic methanogenesis in situ.

174 | 8 Petroleum: from formation to microbiology 8.3.3 The fermentative prokaryotes Besides SRP and MA, numerous fermentative bacteria have been isolated from oilfield ecosystems [23]. They include mesophilic halotolerant, e.g. Fusibacter paucivorans (Clostridiales), Petrimonas sulfuriphila, (Bacteroidetes), slightly halophilic species, e.g. Spirochaeta smaragdinae (Spirochaetales), Dethiosulfovibrio peptidovorans (Synergistales), Denitrivibrio acetiphilus (Deferribacterales), and moderately halophilic species mainly belonging to the genus Halanaerobium (H. salsuginis, H. congolense, H. acetethylicum, H. kushneri) within the Halanaerobiales. Recently, a bacterium pertaining to a novel genus (Halanaerocella petrolearia) was isolated from a deep subsurface hypersaline oil reservoir [58]. To the exception of Synergistales, these fermentative bacteria are frequently detected in oil reservoirs where they may account for a large proportion of the bacterial diversity [36]. All these species may use sugars producing hydrogen and CO2 , and various organic acids (acetate, formate, lactate, etc. . . ) with the exception of (i) D. peptidovorans which has been reported to use only proteinaceous compounds such as peptone and aminoacids and (ii) Denitrivibrio acetiphilus which only ferments fumarate [23]. Molecular surveys show that fermentative prokaryotes represent a frequent, but variable fraction of the microbial diversity in oil reservoirs. Clostridiales (86% of the fields) and Bacteroidetes (50% of the fields) are the most frequently observed. Other groups are infrequent, but they may represent a significant fraction of the diversity in a specific field. For example, Spirochaetales which are usually observed in the range of a fraction of percent, represent between 3 and 12% in the water-flooded Kalamay and Huabei oil reservoirs [42]. Similarly, the Bokor oilfield in Malaysia is characterized by the unusual abundance of sequences from the Deferribacterales, which represent between 20 and 44% of the bacterial diversity [51]. These sequences differ from those of other Deferribacterales isolated from oil reservoirs, but appear to be most closely related to an isolate from an oil-degrading consortium, and may represent a novel putative important member of the oil-degrading community in this field. Thermophilic fermentative bacteria were frequently enriched and/or isolated from hot petroleum reservoirs. They mainly belong to the order Thermotogales which comprises ten thermophilic to hyperthermophilic genera: Thermotoga, Thermosipho, Fervidobacterium, Geotoga, Petrotoga, Marinitoga, Thermococcoides, Kosmotoga, Oceanotoga and Defluviitoga [23, 59–61] (󳶳 Fig. 8.4). Recently, a mesophilic lineage (Mesotoga) within the Thermotogales has been evidenced by the detection of Thermotogales 16S rRNA gene sequences in many mesothermic environments, including a mesothermic petroleum reservoir [62, 63]. The mesophilic nature of such microorganisms has been established by their first isolation and cultivation in 2011 [64]. Most of these Gram-negative anaerobic bacteria possess an outer sheath-like structure called a “toga” ballooning over the ends of the cell (e.g. Thermotoga and Thermosipho spp.) [65, 66]. These microorganisms ferment carbohydrates with acetate and hydrogen being the major end products, except for Mesotoga spp. which have

8.3 Petroleum microbiology

99 100

| 175

Thermotoga elfii (T) SERB 6459 (X80790) Thermotoga lettingae (T) TMO (AF355615) Thermotoga subterranea (T) SL1 (U22664) Thermotoga hypogea (T) SEBR 7054 (U89768)

82

89

Thermotoga thermarum (T) DSM 5069 (AB039769) 99

Thermotoga maritima (T) MSB8 (M21774) Thermotoga neapolitana (T) DSM 4359 (AB039768)

100

Thermotoga naphthophila (T) RKU10 (AJ872268) Thermotoga petrophila (T) RKU1 (AJ872269) (T) ik275mar (GQ292553)

99

Thermosipho melanesiensis (T) BI429 (Z70248) 100

90

Thermosipho atlanticus (T) DV1140 (AJ577471)

99

Thermosipho geolei (T) DSM 13256 (AJ272022) Thermosipho africanus (T) Ob7 (DQ647057)

74

Thermosipho globiformans (T) MN14 (AB257289)

100

Thermosipho japonicus (T) IHB1 (AB024932) Fervidobacterium ssp.

100 100

Thermococcoides shengliensis (T) 2SM-2 (EU276414) Kosmotoga olearia (T) TBF 19.5.1 (EU980631)

100

100

Kosmotoga arenicorallina (T) S304 1 (AB530678) Kosmotoga arenicorallina (T) S304 2 (AB530678)

100

100

Mesotoga spp.

Marinitoga ssp. Geotoga petraea (T) T5 (HM037999)

100 100

100

Geotoga subterranea (T) CC-1 (L10659) Oceanotoga teriensis (T) OCT74 (EU588727) (T) SulfLac1 (FR850164)

100

Petrotoga mobilis (T) SJ95 (Y15479)

100

Petrotoga olearia (T) SL24 (AJ311703) 100

Petrotoga sibirica (T) SL25 (AJ311702)

92 72

0.02

77

Petrotoga mexicana (T) Met-12 (AY125964) Petrotoga halophila (T) MET-B (AY800102) Petrotoga miotherma (T) DSM10691 (FR733705)

Fig. 8.4: Phylogenetic tree based on 16 rRNA gene sequences representing the position of oilfield microorganisms (bold characters) within the order Thermotogales. Neighbor joining method was used, calculation from 1190 aligned pb, bootstrap from 1000 replicates. Aquifex aeolicus (AJ309733) and Bacillus subtilis (K00637) were used as outgroup (not shown). Bar scale, 0.02 substitution per nucleotide. Modified from Cappelletti et al. [69].

176 | 8 Petroleum: from formation to microbiology not been reported to produce hydrogen (e.g. M. prima and M. infera) or which were demonstrated to use sugars only oxidatively in the presence of elemental sulfur as terminal electron acceptor (e.g. M. infera) [67–69]. Thermotogales that have been recovered from oilfield waters and facilities [23] (󳶳 Fig. 8.4) include thermophilic species with some of them growing at temperatures over 80 °C (Thermotogaacetate, hypogea, T. petrophila, and T. naphtophila) [18, 23], together with Thermosipho geoli, Kosmotoga olearia, and Thermococcoides shengliensis growing optimally between 65 °C and 70 °C. Based on recent results, T. shengliensis should be reclassified as a Kosmotoga sp. [70]. Petrotoga, Geotoga, and Oceanotoga spp. have representatives which originated only from oil reservoirs (󳶳 Fig. 8.4) thus suggesting that they might be indigenous [18, 23, 59]. However, the indigenous character of Thermotogales or other anaerobes more generally to the oilfield ecosystems should be considered with caution as indicated above [22]. Within the domain Bacteria, thermophilic members of the family Thermoanaerobacteraceae which includes the genera Thermoanaerobacter, Thermoanaerobacterium and Caldanaerobacter were also recovered frequently from hot and slightly saline reservoirs [71]. They are known as acetate- and hydrogen-producing bacteria. Thermoanaerobacter thermohydrosulfuricus and T. brockii, first isolated from soil and a thermal spring, respectively, were retrieved from such reservoirs. Beside acetate and hydrogen, Caldanaerobacter subterraneus isolated from an oil reservoir in Southwest France was shown to produce L-alanine from glucose metabolism. The moderately thermophilic Garciella nitratireducens (optimum growth at 55 °C) and Mahella australiensis (optimum growth at 50 °C) isolated from an oilfield separator in the Gulf of Mexico and from an Australian oil reservoir were also reported to ferment sugars [23]. In contrast, the Synergistetes species Thermovirga lienii which has been isolated from a hot oil reservoir in the North Sea, growing optimally at 58 °C and making up to 25% of bacterial diversity in that reservoir, was shown to ferment proteinaceous substrates, amino acids and a limited range of organic acids, but not alkanes, sugars or fatty acids [52, 72]. Stetter et al. [73] first provided evidence of hyperthermophilic fermentative archaea thriving in hot deep reservoirs in the North Sea by enrichment and/or isolation of Thermococcus and Pyrococcus spp. These species generally grow on peptides, polysaccharides, or sugars having the ability to reduce elemental sulfur to sulfide [29]. However, several of these species (e.g. Thermococcus litoralis, T. celer) were first discovered in coastal hydrothermal vents. Thermococcus sibiricus is the only novel species within the genus Thermococcus isolated from a high-temperature oil reservoir (Western Siberia) and fully characterized [74]. It grows optimally at 78 °C on peptides but not on carbohydrates. Its growth was stimulated by alkanes, but the process of oxidation is still unknown [75]. Based on its peculiar physiological features (growth on cellulose, agarose, lamarine, triglycerides and crude oil), T. sibiricus was hypothesized to be indigenous to the nonflooded oil reservoir from where it originated [75]. Thermococcus litoralis is the most frequently hyperthermophilic archaeon encoun-

8.3 Petroleum microbiology

| 177

tered in molecular surveys of hot oil reservoirs [36]; its proportion increases with increasing in situ temperatures [41]. Many of these oilfield fermentative bacteria may use mineral compounds alternatively (e.g. thiosulfate, elemental sulfur, nitrate, ferric iron, etc.) in a nondependent way as terminal electron acceptors. Most Thermotoga spp reduced thiosulfate (e.g. T. elfii, T. subterranea, T. hypogea) elemental sulfur (T. naphtophila) or both (T. petrophila). Depending on species, Petrotoga and Geotoga spp. were shown to reduce thiosulfate and/or elemental sulfur. Kosmotoga olearia and Thermosipho geoli reduce thiosulfate and elemental sulfur, respectively, to sulfide. In contrast to Thermoanaerobacter spp. which reduce thiosulfate to sulfide, Thermoanaerobacterium spp. reduce it to elemental sulfur [23]. While the mesophilic Shewanella putrefaciens (Gammaproteobacteria) was shown to oxidize hydrogen in the presence of thiosulfate and elemental sulfur as electron acceptors, the Bacteroidetes Petrimonas sulfuriphila, isolated from a biodegraded reservoir, only used elemental sulfur when growing on carbohydrates, glycerol and lactate as energy sources [53]. The ability to reduce elemental sulfur has been also demonstrated for oilfield hyperthermophilic archaeons belonging to the order Thermococcales (Thermococcus and Pyrococcus spp.) [29]. There are several other examples of oilfield microorganisms using these mineral sulfur compounds as terminal electron acceptors thus suggesting that they may be involved in the sulfur cycle within oilfield environments and may possibly be active agents of biocorrosion by producing sulfide [19]. These reductive activities have often been explained as a result of H2 detoxication rather than a respiratory process, except in some cases for Thermococcales. Clone libraries, high coverage sequencing of DNA and cultivation demonstrate that fermentative prokaryotes, especially Thermotogales, inhabit hot reservoirs all over the world [26, 37, 41, 52, 76]. However, from the geomicrobiological point of view, their contribution in situ is still unknown and would need further investigation.

8.3.4 Other metabolic lifestyle bacteria Hydrogen may also be oxidized by reservoir isolates. One Thermoanaerobacter strain isolated from petroleum reservoirs was shown to oxidize H2 in the presence of thiosulfate and a similar type of metabolism was also demonstrated for a Thermotoga species [23]. Hydrogen may be also oxidized by dissimilatory Fe(III) reduction either by oilfield bacteria (Thermotoga spp., Deferribacter thermophilus) or archaea (Thermococcus sibiricus) thus highlighting a possible role of these microorganisms in the iron cycle in deep hot reservoirs [77]. Beside Fe(III), Deferribacter thermophilus isolated from oilfield-production waters in the North Sea was shown to reduce Mn (IV), and nitrate [23]. There is evidence that synthrophic associations between Syntrophus spp. and hydrogenotrophic MA exist in oil reservoirs. They have been hypothesized to actively participate in anaerobic hydrocarbon oxidation [53, 78].

178 | 8 Petroleum: from formation to microbiology Nitrate is not originally present in oil reservoirs, but its addition in formation waters to prevent oil souring may have a great impact in the geomicrobiology of these ecosystems. Members of the thermophilic Deferribacteres were retrieved by molecular studies and shown to respond to nitrate addition in a nitrate-treated high-temperature oil production facility [79]. Deferribacteres representatives were assumed to be indigenous to this high-temperature oil reservoir as they were not found in the injection water and play a major role in nitrate-induced souring control at high temperatures because of their ability to reduce nitrate to nitrite or ammonium [79]. Similarly, the monitoring of microbial communities in a mesothermic oilfield, subjected to continuous field-wide injection of nitrate to remove sulfide, indicated unusually high numbers of bacteria able to reduce nitrate such as Sulfurospirillum, Denitrovibrio, Sulfurimonas, Arcobacter, or Thauera [80]. Three anaerobic nitrate-reducing bacterial species have been isolated so far. While Denitrivibrio acetiphilus and Petrobacter succinatimandens use obligatory nitrate to oxidize acetate and/or other organic acids (e.g. formate, fumarate and succinate) [23], Garciella nitratireducens and Petrimonas sulfuriphila ferment sugars and may use nitrate alternatively as electron acceptors [23, 53]. All these species, with the exception of P. succinatimandens reducing nitrate to nitrous oxide, were known to reduce nitrate into ammonium [23]. There is microbiological and molecular evidence that microaerophilic nitrate-reducing bacteria within the Epsilonproteobacteria inhabit low-temperature oil reservoirs where they may represent up to 100% of the bacterial diversity [23, 37, 53, 54, 81]. Such microorganisms were reported to use either H2 and/or H2 S, making them very good candidates for oxidizing this sulfur compound and thus avoiding oil souring and its pernicious effects during the exploitation of crude oils [23, 82]. They can degrade oil by coupling nitrate reduction to the oxidation of reduced, probably organic, sulfur species [81]. The ecological significance of these Epsilonproteobacteria (e.g. Sulfuricurvum, Arcobacter, Sulfurospirillum, Campylobacter spp.), due to their ability to possibly use mineral and/or organic sulfur compounds, has been recently outlined in formation waters with a peculiar emphasis for highly biodegraded petroleum systems [54]. However, the anaerobic process by which these microorganisms may oxidize their substrates in situ is still a matter of debate since oxygen and nitrate are not usually found in nonflooded petroleum reservoirs. Therefore, their dominance within the prokaryotic community in oilfield ecosystems may result from anthropogenic activities during oil exploration and exploitation or natural contaminations (e.g aquifer connections or meteoritic water). This may be also the case of oilfield facultatively aerobic nitrate-reducing bacteria pertaining to the genus Geobacillus (e.g. G. subterraneus, G. uzenensis within the Firmicutes) or Marinobacter (e.g. M. aquaeoli, within the Gammaproteobacteria). They have been reported as hydrocarbonoclastic microbes and may therefore be of ecological relevance in exploited oilfields with the possible entrance of oxygen and/or nitrate addition (see comments above). Another hypothesis regarding the presence of microaerophilic or facultative aerobic microorganisms would have been an unknown anaerobic metabolism that they could perform in situ by using other electron accep-

8.4 Conclusion

|

179

tors different from sulfate and CO2 or by possibly coupling with prokaryotic hydrogen scavengers. Molecular surveys of oil reservoirs, and especially mesothermic reservoirs, often highlight a very diverse bacterial community, including large numbers of Proteobacteria. Although it is sometimes difficult to determine their autigenicity, the sequences obtained are frequently closely related to other known and cultivated oil-reservoir isolates or oil-degrading clones from oil spill or contaminated soils. In fact, bacterial diversity in low-temperature oil reservoirs is mainly characterized by Proteobacteria from the 𝛼, 𝛽, 𝛾, 𝛿, and 𝜀 classes (󳶳 Fig. 8.2) [36, 42, 50, 51, 54, 57] which may account for up to 70% of the total bacterial diversity evidenced by culture-independent methods. 𝛼-proteobacteria and 𝛾-proteobacteria are observed in all but the hot temperature oilfields [36]. 𝛼-proteobacteria, such as Rhodovulum (13.1%), Stappia (11.7%), Rhizobium (14.2%) and Rhodobacter (22.4%) clones were shown to dominate in the mesothermic Enermark oilfield in Canada [57, 80]. Although generally described as aerobic, members of these four genera can grow anaerobically at the expense of nitrate, and have been identified in different subsurface oil-reservoir cultivation-independent studies. Rhodovulum or Stappia clones have been retrieved from oil spills and shown to degrade alkanes in the presence of nitrate. Known oil-degrading, 𝛾-proteobacteria dominate the bacterial diversity in oil and formation waters from the Bokor oil reservoir, amongst which Marinobacterium (34–39%), Pseudomonas (5–25%) and an unidentified group of Alteromonadales (10–32%), represent the large majority of bacterial sequences [51]. While Pseudomonadales are facultative anaerobes, and metabolically versatile, Marinobacter, a related genus to Marinobacterium, is described as strictly aerobic, although it is frequently identified in oil reservoirs [83, 84]. 𝛿-proteobacteria, such as Smithella spp. are found in most reservoirs [36, 37, 85]. Smithella spp. are members of the family Synthrophaceae, which are syntrophic bacteria known to be associated with MA [36]. Actinobacteria, such as Rhodococcus, Propionicella [57], Nocardia spp. [41], which include cellulolytic organisms degrading poorly water-soluble organic compounds are seldom observed. It is unclear why these organisms, many of which require oxygen to degrade cellulose, are present, nor whether the organic matter in oil and coal is chemically close enough to cellulose for these organisms to be able to degrade it. Their detection may point out the limits of the sampling methods that can be applied to these environments. Chloroflexi are also found in about half of the oil reservoirs, especially in low-temperature reservoirs, although their proportion rarely exceeds 10% of the bacterial diversity [43, 50, 57].

8.4 Conclusion In contrast to what is observed in most ecosystems, the view of the microbial community inhabiting oil reservoirs obtained by molecular approaches shows a very good analogy with that obtained by cultural approaches [86]. Both approaches together

180 | 8 Petroleum: from formation to microbiology with a metagenomic study recently performed in an oil reservoir located at a depth of 2.5 kilometres [37] revealed that sulfate-reducing prokaryotes and methanoarchaea in particular, thrive in mesothermic and hot oil reservoirs and that some of them should be considered as indigenous to oilfield ecosystems. Among them, the geomicrobiological significance of hydrogenotrophic methanoarchaea is highly relevant as they have been demonstrated to actively participate to methanogenesis from crude oil in subsurface petroleum reservoirs [26, 27, 36, 78]. Regarding the oxidative process of hydrocarbons, the role to be played by members of the family Syntrophaceae (e.g. Syntrophus sp.) has been highlighted [36, 53, 78], but none of these members has been so far isolated at least in coculture with hydrogen/formate using microorganisms from petroleum reservoirs. It deserves further attention as well as the wide distribution of Proteobacteria in oilfield ecosystems with the presence of some representatives possibly resulting from anthropogenic contaminations. Fermentative bacteria of the phylum Firmicutes, Bacteroidetes, and of the order Thermotogales have been found as common inhabitants of these extreme environments. The involvement of Thermotogales in hydrocarbon oxidation has recently been questioned [26]. Within this order, Geotoga and Petrotoga spp. have only been isolated from production waters of oil reservoirs, thus strongly suggesting the indigenous nature of Thermotogales in these ecosystems. Metabolic activities of these bacteria together with Firmicutes and Bacteroidetes should contribute to favor Microbial Enhanced Oil Recovery (MEOR) by producing gases (CO2 , H2 ), acids (e.g. acetate, propionate), but also solvents (e.g. ethanol, propanol, butanol). Similarly, syntrophic associations between hydrocarbonoxidizing bacteria and hydrogenotrophic methanoarchaea that are known to develop in oilfield waters to produce CO2 and CH4 [25, 27, 78] will be most probably beneficial for MEOR developments in the next future. Despite our knowledge of the involvement of anaerobic microorganisms in oil maturation, two intriguing questions still remain: (i) did indigenous microorganism petroleum reservoirs originate from sedimentary basins where organic matter accumulated to further produced oil and (ii) to what extent may thermophilic/hyperthermophilic microorganisms with possible growth at temperature up to 80–90 °C in oil reservoirs have participated to oil formation from source rock as compared to abiotical processes? Finally, among the physicochemical parameters known of importance in the geomicrobiology of oil reservoirs, the effect of hydrostatic pressure (fluid pressure) which has been neglected so far, should clearly merit further investigations as it may impact the in situ metabolic activities of microorganisms.

Acknowledgements We thank Dr. Pierre Roger and Dr. Wajdi Ben-Hania for revising the manuscript, and Dr. Anne Postec for drawing the phylogenetic tree.

References |

181

References [1] [2] [3] [4] [5] [6] [7] [8] [9]

[10] [11] [12] [13] [14] [15]

[16] [17] [18] [19]

[20]

[21]

[22]

Hunt JM. Petroleum geochemistry and geology. San Francisco: Freeman W, 617pp, 1979. Tissot BP, Welte DH. Petroleum formation and occurrence: a new approach to oil and gas exploration (2nd edn). Berlin: Springer Verlag, 699pp, 1984. Brooks J, Welte D. Advances in petroleum geochemistry. London: Academic Press, 344pp, 1984. Porfir’ev VB. Inorganic origin of petroleum. AAPG Bull 58 (1974), 3–33. Gold T, Soter S. Abiogenic methane and the origin of petroleum. Energy Explor Exploit 1 (1982), 89–104. North FK. Petroleum geology. Boston: Allen & Unwin, 607pp, 1985. Berner RA. The Phanerozoic carbon cycle: CO2 and O2 . Oxford, UK: Oxford University Press, 150pp, 2004. Husseini M. World production of conventional petroleum liquids to 2030: a comparative review. GeoArabia 14 (2009), 215–267. Craig J, Biffi U, Galimberti RF, Ghori KAR, Gorter JD, Hakhoo N, Le Heron DP, Thurow J, Vecoli M. The paleobiolology and geochemistry of Precambrian hydrocarbon source rocks. Mar Petrol Geol 40 (2013), 1–47. Dutkiewicz A, Rasmussen B, Buick R. Oil preserved in fluid inclusions in Archaean sandstones. Nature 395 (1998), 885–888. Perrodon A, Masse P. Subsidence, sedimentation and petroleum systems. Petrol Geol 7 (1984), 5–26. Erhenberg SN, Nadeau PH. Sandstone vs. carbonate petroleum reservoirs: A global perspective on porosity-depth and porosity-permeability relationships. AAPG Bull 89 (2005), 435–445. Mearns E. GHAWAR: an estimate of remaining oil reserves and production decline. The Oil Drum, Europe. http///www.theoildrum.com/node/2494 2007: 58pp. Holmes DE, Risso C, Smith JA, Lovley DR. Anaerobic oxidation of benzene by the hyperthermophilic archaeon Ferroglobus placidus. Appl Environ Microbiol 77 (2011), 5926–5933. Khelifi N, Grossi V, Hamdi M, Dolla A, Tholozan JL, Ollivier B, Hirschler-Réa A. Anaerobic oxidation of fatty acids and alkenes by the hyperthermophilic sulfate-reducing archaeon Archaeoglobus fulgidus. Appl Environ Microbiol 76 (2010), 3057–3060. Fort J. The Elgin/Franklin project: developing the largest high pressure/high temperature fields in the world. Offshore Technol Conf. Houston 2000, OTC paper 12117. Diyashev RN, Ziganshin ES, Ryabchenko VN. Verkhnechonsky Field shows eastern Russia’s potential. Oil Gas J 97 (1999), 53–59. Magot M, Ollivier B, Patel BKC. Microbiology of petroleum reservoirs. Anton Leeuw Int J Gen M 77 (2000), 103–116. Ollivier B, Alazard D. The oil reservoir ecosystem. In: Timmis KN, ed. Handbook of Hydrocarbon Microbiology: Microbial Interactions with Hydrocarbons, Oils, Fats and Related Hydrophobic Substrates and Products. Berlin: Springer Verlag, 2262–2268, 2010. Cayol JL, Ollivier B, Patel BKC, Prensier G, Guezennec J, Garcia JL. Isolation and characterization of Halothermothrix orenii gen. nov., sp. nov., a halophilic, thermophilic, fermentative strictly anaerobic bacterium. Int J Syst Bacteriol 44 (1994), 534–540. Cayol JL, Ducerf S, Patel BKC, Garcia JL, Thomas P, Ollivier B. Thermohalobacter berrensis gen. nov.,sp.nov., a thermophilic, strictly halophilic bacterium from a solar saltern. Int J Syst Evol Microbiol 50 (2000), 559–564. Magot M. Indigenous microbial communities in oilfields. In: Ollivier B, Magot M, eds. Petroleum microbiology. Washington, USA: ASM press, 21–33, 2005.

182 | 8 Petroleum: from formation to microbiology [23] Ollivier B, Cayol JL. The fermentative, iron-reducing and nitrate-reducing microorganisms. In: Ollivier B, Magot M, eds. Petroleum microbiology. Washington, USA: ASM press, 71–88, 2005. [24] Barth T. Organic acids and inorganic ions in waters from petroleum reservoirs, Norwegian continental shelf: a multivariate statistical analysis and comparison with american reservoir formation waters. Appl Geochem 6 (1991), 1–15. [25] Gieg LM, Davidova IA, Duncan KE, Suflita JM. Methanogenesis, sulfate-reduction and crude oil biodegradation in hot Alaskan oilfields. Environ Microbiol 12 (2010), 3074–3086. [26] Mbadinga SM, Li K-P, Zhou L, Wang L-Y, Yang S-Z, Liu J-F, Gu J-D, Mu B-Z. Analysis of alkane-dependent methanogenic community derived from production water of high-temperature petroleum reservoir. Appl Microbiol Biotechnol 96 (2012), 531–542. [27] Wang L-Y, Gao C-X, Mbadinga SM, Zhou L, Liu J-F, Gu J-D, Mu B-Z. Characterization of an alkane-degrading methanogenic enrichment culture from production water of an oil reservoir after 274 days of incubation. Int Biodeterior Biodegr 65 (2011), 444–450. [28] Birkeland NK (2005) Sulfate-reducing Bacteria and Archaea. In: Ollivier B, Magot M, eds, Petroleum microbiology. Washington, USA, ASM Press, 35–54, 2005. [29] Jeanthon C, Nercessian O, Corre E, Grabowski-Lux A. Hyperthermophilic and methanogenic Archaea in oilfields. In: Ollivier B, Magot M, eds. Petroleum microbiology. Washington, USA: ASM press, 55–69, 2005. [30] Aitken CM, Jones DM, Maguire MJ, Gray ND, Sherry A, Bowler BFJ, Ditchfield AK, Larter SR, Head IM. Evidence that crude oil alkane activation proceeds by different mechanisms under sulfate-reducing and methanogenic conditions. Geochim Cosmochim Acta 109 (2013), 162– 174. [31] Beeder J, Nilsen RK, Torsvik T, Lien T. Archaeoglobus fulgidus isolated from hot North Sea oilfield waters. Appl Environ Microbiol 60 (1994), 1227–1231. [32] Nilsen RK, Beeder KJ, Thorstenson T, Torsvik T. Distribution of thermophilic marine sulfate reducers in North Sea oilfield waters and oil reservoirs. Appl Environ Microbiol 62 (1996), 1793– 1798. [33] Fardeau ML, Goulhen F, Bruschi M, Khelifi N, Cayol JL, Igniatidis I, Guyot F, Ollivier B. Archaeoglobus fulgidus and Thermotoga elfii, thermophilic isolates from deep geothermal water of the Paris Basin. Geomicrobiol J 26 (2009), 119–130. [34] Belyakova EV, Rozanova EP, Borzenkov IA, Tourova TP, Pusheva MA, Lysenko AM, Kolganova TV. The new facultatively chemolithoautotrophic, moderately halophilic, sulfate-reducing bacterium Desulfovermiculus halophilus gen. nov., sp. nov., isolated from an oilfield. Microbiology (engl.Transl.) 75 (2006), 161–171. [35] Ommedal H, Torsvik T. Desulfotignum toluenicum sp. nov, a novel toluene-degrading, sulfatereducing bacterium isolated from an oil reservoir model column. Int J Syst Evol Microbiol 57 (2007), 2865–2869. [36] Gray ND, Sherry A, Hubert C, Dolfing J, Head IM. Methanogenic degradation of petroleum hydrocarbons in subsurface environments: remediation, heavy oil formation and energy recovery. Adv Appl Microbiol 72 (2010), 137–161. [37] Kotlar HK, Lewin A, Johansen J, Throne-Holst M, Haverkamp T, Markussen S, Winnberg A, Ringrose P, Aakvik T, Ryeng E, Jakobsen K, Drablos F, Valla S. High coverage sequencing of DNA from microorganisms living in an oil reservoir 2.5 kilometers subsurface. Environ Microbiol Rep 3 (2011), 674–681. [38] Zhang F, She Y-H, Chai L-J, Banat IM, Zhang X-T, Shu F-C, Wang Z-L, Yu L-J, Hou D-J. Microbial diversity in long-term water-flooded oil reservoirs with different in situ temperatures in China. Sci Rep 2 (2012), 760. [39] Meslé M, Dromart G, Oger P. Microbial methanogenesis in subsurface oil and coal. Res Microbiol 164 2013, 959–972.

References | 183

[40] Guan J, Xia L-P, Wang L-Y, Liu J-F, Gu J-D, Mu B-Z. Diversity and distribution of sulfate-reducing bacteria in four petroleum reservoirs detected by using 16S rRNA and dsrAB genes. Intern Biodeter Biodegrad 76 (2013), 58–66. [41] Yamane K, Hattori Y, Ohtagaki H, Fujiwara K. Microbial diversity with dominance of 16S rRNA gene sequences with high GC contents at 74 and 98 degrees C subsurface crude oil deposits in Japan. FEMS Microbiol Ecol 76 (2011), 220–235. [42] Zhao L, Ma T, Gao M, Gao P, Cao M, Zhu X, Li G. Characterization of microbial diversity and community in water flooding oil reservoirs in China. World J Microbiol Biotechnol 28 (2012), 3039–3052. [43] Pham VD, Hnatow LL, Zhang S, Fallon RD, Jackson SC, Tomb J-F, DeLong EF, Keeler SJ. Characterizing microbial diversity in production water from an Alaskan mesothermic petroleum reservoir with two independent molecular methods. Environ Microbiol 11 (2009), 176–187. [44] Cheng L, Qiu T-L, Xia L, Wang W-D, Deng Y, Yin X-B, Zhang H. Isolation and characterization of Methanoculleus receptaculi sp. nov. from Shengli oilfield, China. FEMS Microbiol Lett 285 (2008), 65–71. [45] Cheng L, Dai L, Li X, Zhang H, Lu Y. Isolation and characterization of Methanothermobacter crinale sp. nov., a novel hydrogenotrophic methanogen from the Shengli oilfield. Appl Environ Microbiol 77 (2011), 5212–5219. [46] Mayumi D, Mochimaru H, Yoshioka H, Sakata S, Maeda H, Miyagawa Y, Ikarashi M, Takeuchi M, Kamagata Y. Evidence for syntrophic acetate oxidation coupled to hydrogenotrophic methanogenesis in the high-temperature petroleum reservoir of Yabase oilfield (Japan). Environ Microbiol 13 (2011), 1995–2006. [47] Ollivier B, Caumette P, Garcia JL, Mah RA. Anaerobic bacteria from hypersaline ecosystems. Microbiol Rev 58 (1994), 27–38. [48] Watkins AJ, Roussel EG, Webster G, Parkes RJ, Sass H. Choline and n,n-dimethylethanolamine as direct substrates for methanogens. Appl Environ Microbiol 78 (2012), 8298–8303. [49] Ren H-Y, Zhang X-J, Song Z-Y, Rupert W, Gao G-J, Guo S-X, Zhao L-P. Comparison of microbial community compositions of injection and production well samples in a long-term waterflooded petroleum reservoir. PloS ONE (2011), 6:e23258. doi:10.1371/journal.pone.0023258. [50] Yergeau E, Lawrence J, Sanschagrin S, Waiser M, Korber D, Greera C. Next-generation sequencing of microbial communities in the Athabasca River and its tributaries in relation to oil sands mining activities. Appl Environ Microbiol 78 (2012), 7626–7637. [51] Li D, Midgley DJ, Ross JP, Oytam Y, Abell GCJ, Volk H, Daud WAW, Hendry P. Microbial biodiversity in a Malaysian oilfield and a systematic comparison with oil reservoirs worldwide. Arch Microbiol 194 (2012), 513–523. [52] Dahle H, Garshol F, Madsen M, Birkeland N-K. Microbial community structure analysis of produced water from a high-temperature North Sea oil-field. Anton Leeuw Int J Gen M Microbiol 93 (2008), 37–49. [53] Grabowski A, Tindall BJ, Bardin V, Blanchet D, Jeanthon C. Petrimonas sulfuriphila gen. nov., sp. nov., a mesophilic fermentative bacterium isolated from a biodegraded oil reservoir. Int J Syst Evol Microbiol 55 (2005), 1113–1121. [54] Hubert CR, Oldenburg TBP, Fustic M, Gray ND, Larter SR, Penn K, Rowan AK, Seshadri R, Sherry A, Swainsbury R, Voordouw G, Voordouw JK, Head IM. Massive dominance of Epsilonproteobacteria in formation waters from a Canadian oil sands reservoir containing severely biodegraded oil. Environ Microbiol 14 (2012), 387–404. [55] Doerfert SN, Reichlen M, Iyer P, Wang M, Ferry JG. Methanolobus zinderi sp. nov., a methylotrophic methanogen isolated from a deep subsurface coal seam. Int J Syst Evol Microbiol 59 (2009), 1064–1069.

184 | 8 Petroleum: from formation to microbiology [56] Mochimaru H, Tamaki H, Hanada S, Imachi H, Nakamura K, Sakata S, Kamagata Y. Methanolobus profundi sp. nov., a methylotrophic methanogen isolated from deep subsurface sediments in a natural gas field. Int J Syst Evol Microbiol 59 (2009), 714–718. [57] Kryachko Y, Dong X, Sensen CW, Voordouw G. Compositions of microbial communities associated with oil and water in a mesothermic oilfield. Anton Leeuw Int J Gen M Microbiol 101 (2012), 493–506. [58] Gales G, Chehider N, Joulian C, Battaglia-Brunet F, Cayol J-L, Postec A, Borgomano J, Neria-Gonzales I, Lomans BP, Ollivier B, Alazard D. Characterization of Halanaerocella petrolearia gen. nov., sp. nov., a new anaerobic moderately halophilic fermentative bacterium isolated from a deep subsurface hypersaline oil reservoir. Extremophiles 15 (2011), 565–571. [59] Jayasinghearachchi HS, Lal B. Oceanotoga teriensis gen. nov., sp. nov., a thermophilic bacterium isolated from offshore oil-producing wells. Int J Syst Evol Microbiol 61 (2011), 554–560. [60] Di Pippo JL, Nesbo CL, Dahle H, Doolittle WF, Birkland N-K, Noll KM. Kosmotoga olearia gen. nov., sp. nov., a thermophilic, anaerobic heterotroph isolated from an oil production fluid. Int J Syst Evol Microbiol 59 (2009), 2991–3000. [61] Ben Hania W, Ghodbane R, Postec A, Hamdi M, Ollivier B, Fardeau M-L. Defluviitoga tunisiensis gen. nov., sp. nov., a thermophilic bacterium isolated from a mesothermic and anaerobic whey digester. Int J Syst Evol Microbiol 62 (2012), 1377–1382. [62] Nesbø CL, Dlutek M, Zhaxybayeva O, Doolittle WF. Evidence for existence of “Mesotogas”, members of the order Thermotogales adapted to low-temperature environments. Appl Environ Microbiol 72 (2006), 5061–5068. [63] Nesbø CL, Kumaraswamy R, Dlutek M, Doolittle WF, Foght J (2010) Searching for mesophilic Thermotogales bacteria: “Mesotogas” in the wild. Appl Environ Microbiol 76 (2010), 4896– 4900. [64] Ben Hania W, Ghodbane R, Postec A, Brochier-Armanet C, Hamdi M, Fardeau M-L, Ollivier B. Cultivation of the first mesophilic representative (“mesotoga”) within the order Thermotogales. Syst Appl Microbiol 34 (2011), 581–585. [65] Antoine E, Cilia V, Meunier J, Guezennec J, Lesongeur F, Barbier G. Thermosipho melanesiensis sp. nov., a new thermophilic anaerobic bacterium belonging to the order Thermotogales, isolated from deep-sea hydrothermal vents in the southwestern Pacific Ocean. Int J Syst Bacteriol 47 (1997), 1118–1123. [66] Huber R, Woese CR, Langworthy TA, Fricke H, Stetter KO. Thermosipho africanus gen. nov., represents a new genus of thermophilic Eubacteria within the Thermotogales. Syst Appl Microbiol 12 (1989), 32–37. [67] Ben Hania W, Postec A, Aüllo T, Ranchou-Peyruse A, Erauso G, Brochier-Armanet C, Hamdi M, Ollivier B, Saint-Laurent S, Magot M, Fardeau ML. Mesotoga infera sp. nov., a novel mesophilic member of the order Thermotogales, isolated from an underground gas storage in France. Int J Syst Evol Microbiol 2013, doi: 10.1099/ijs.0.047993–0. [68] Nesbø CL, Bradnan DM, Adebusuyi A, Dlutek M, Petrus AK, Foght J, Doolittle WF, Noll KM. Mesotoga prima gen. nov., sp. nov., the first described mesophilic species of the Thermotogales. Extremophiles 16 (2012), 387–393. [69] Cappelletti M, Postec A, Ollivier B, Zannoni D. Members of the order Thermotogales: From microbiology to hydrogen production. In: Zannoni B, De Philippis R, eds. Advances in Photosynthesis and Respiration, 38. Dordrecht, Springer, 2013 (under press). [70] Nunoura T, Hirai M, Imachi H, Miyazaki M, Makita H, Hirayama H, Furushima Y, Yamamoto H, Takai K. Kosmotoga arenicorallina sp. nov., a thermophilic and obligately anaerobic heterotroph isolated from a shallow hydrothermal system occurring within a coral reef, southern part of the Yaeyama Archipelago, Japan, reclassification of Thermococcoides shengliensis as Kosmotoga shengliensis comb. nov. and emended description of the genus Kosmotoga. Arch Microbiol 192 (2010), 811–819.

References | 185

[71] Grassia GS, McLean M, Glénat P, Bauld J, Sheehy AJ. A systematic survey for thermophilic fermentative bacteria and archaea in high temperature petroleum reservoirs. FEMS Microbiol Ecol 21 (1996), 47–58. [72] Dahle H, Birkeland N. Thermovirga lienii gen. nov., sp. nov., a novel moderately thermophilic, anaerobic, amino-acid-degrading bacterium isolated from a North Sea oil well. Int J Syst Evol Microbiol 56 (2006), 1539–1545. [73] Stetter KO, Huber R, Blöchl E, Kurr M, Eden RD, Fielder M, Cash H, Vance I. hyperthermophilic archaea are thriving in deep North Sea and Alaskan oil reservoirs. Nature 365 (1993), 743–745. [74] Miroshnichenko ML, Hippe H, Stackebrandt E, Kostrikina NA, Chernyh NA, Jeanthon C, Nazina TN, Belyaev SS, Bonch Osmolovskaya EA. Isolation and characterization of Thermococcus sibiricus sp. nov. from a Western Siberia high-temperature oil reservoir. Extremophiles 5 (2001), 85–91. [75] Mardanov AV, Ravin NV, Svelitchnyi VA, Beletsky AV, Miroshnichenko ML, Bonch-Osmolovskaya EA. Metabolic versality and indigenous origin of the archaeon Thermococcus sibiricus, isolated from a Siberian oil reservoir, as revealed by genome analysis. Appl Environ Microbiol 75 (2009), 4580–4588. [76] Mnif S, Bru-Adan V, Godon J-J, Sayadi S, Chamkha M. Characterization of the microbial diversity in production waters of mesothermic and geothermic Tunisian oilfields. J Bas Microbiol 53 (2013), 45–61. [77] Slobodkin AI, Jeanthon C, L’Haridon S, Nazina T, Miroshnichenko M, Bonch-Osmolovskaya E. Dissimilatory reduction of Fe(III) by thermophilic Bacteria and Archaea in deep-subsurface petroleum reservoirs of Western Siberia. Curr Microbiol 39 (1999), 99–102. [78] Jones DM, Head IM, Gray ND, Adams JJ, Rowan AK, Aitken CM, Bennett B, Huang H, Brown A, Bowler BFJ, Oldenburg T, Erdmann M, Larter SR. Crude-oil biodegradation via methanogenesis in subsurface petroleum reservoirs. Nature 451 (2008), 176–181. [79] Gittel A, Kofoed MVW, Sorensen KB, Ingvorsen K, Schramm A. Succession of Deferribacteres and Epsilonproteobacteria through a nitrate-treated high-temperature oil production facility. Syst Appl Microbiol 35 (2012), 165–174. [80] Cornish Shartau SL, Yurkiw M, Lin S, Grigoryan AA, Lambo A, Park H-S, Lomans BP, van der Biezen E, Jetten MSM, Voordouw G. Ammonium concentrations in produced waters from a mesothermic oilfield subjected to nitrate injection decrease through formation of denitrifying biomass and anammox activity. Appl Environ Microbiol 76 (2010), 4977–4987. [81] Watanabe K, Kodama Y, Syutsubo K, Harayama S. Molecular characterization of bacterial populations in petroleum-contaminated groundwater discharged from underground crude oil storage cavities. Appl Environ Microbiol 66 (2000), 4803–4809. [82] Gevertz D, Telang AJ, Voordouw G, Jenneman GE. Isolation and characterization of strains CVO and FWKO B, two novel nitrate-reducing, sulfide-oxidizing bacteria isolated from oilfield brine. Appl Environ Microbiol 66 (2000), 2491–2501. [83] Li D, Hendry P. Microbial diversity in petroleum reservoirs. Microbiol Aust 29 (2008), 25–27. [84] Sette LD, Simioni KCM, Vasconcellos SP, Dussan LJ, Neto EVS, Oliveira VM. Analysis of the composition of bacterial communities in oil reservoirs from a southern offshore Brazilian basin. Anton Leeuw Int J Gen M Microbiol 91 (2007), 253–266. [85] Kobayashi H, Endo K, Sakata S, Mayumi D, Kawaguchi H, Ikarashi M, Miyagawa Y, Maeda H, Sato K. Phylogenetic diversity of microbial communities associated with the crude-oil, large-insoluble-particle and formation-water components of the reservoir fluid from a nonflooded high-temperature petroleum reservoir. J Biosci Bioeng 113 (2012), 204–210. [86] Youssef N, Elshabed MS, McInerney MJ. Microbial processes in oilfields: culprits, problems and opportunities. In: Laskin AI, Sariaslani S, Gadd GM, (Eds): Advances in Applied Microbiology, 66. Burlington, Academic Press; 141–251, 2009.

Virginia Edgcomb, William Orsi, and Jennifer F. Biddle

9 Fungi in the marine subsurface 9.1 Introduction Marine sediments cover more than two thirds of the Earth’s surface and have been estimated to contain a significant fraction of Earth’s prokaryotic biomass [1]. Ribosomal RNA surveys of surficial marine sediments at bathypelagic (1000−4000 m) or abyssopelagic (> 4000 m) depths show considerable eukaryotic diversity (e.g. [2–8]). Through decomposition and mineralization of organic matter, and by grazing in subsurface horizons where bacterial and/or archaeal numbers are high, protists and fungi have the potential to significantly impact carbon cycling in marine sediments. Fungi appear to be some of the most successful eukaryotes in deep-sea extreme near sediment-surface environments (e.g. review by Nagano and Nagahama [9]), and appear to be able to thrive under conditions similar to those observed in the deep sea (e.g. [10]). Many fungi are able to grow under anaerobic conditions, including fermentative yeasts (see discussion in [11]). However, much remains to be learned about the depth range, diversity and activities of eukaryotes in the deep subsurface. Here we review what is known about fungi in deep-sea sedimentary environments and deep subsurface environments.

9.2 The concept of marine fungi Marine fungi are predominantly comprised of an ecologically-defined group of filamentous ascomycetes, their anamorphs and yeasts [12]. Previous studies of marine fungi have primarily focused on tropical mangroves, salt marshes and open ocean regimes where they are known to be involved in the degradation of organic matter (e.g. [13–20]. Fungi are known to play key roles in ecologically important relationships as parasites, pathogens, mutualists and endobionts of marine organisms (e.g. [12, 21]), to represent a significant portion of the biomass in terrestrial systems, and to impact biogeochemical cycles and food webs [3, 22, 23]. Despite the high number of fungal species described to date [24, 25], we are only beginning to gain an understanding of their presence in deep marine environments (e.g. [2, 5–8, 11, 16, 22, 26–31]). The recovery of culturable fungi from deep marine sediments has not necessarily led to a widely-held view of fungi as active participants in subsurface biogeochemical cycling. For example, in a study reporting culturable fungi from sediment cores collected in the Chagos Trench in the Indian Ocean, these cultured sporulating and nonsporulating fungi from sediments up to 0.43 million years old were referred to as

188 | 9 Subsurface Fungi

Fig. 9.1: Heatmap showing the overlap in fungal genera detected in the subseafloor samples (gray) from the Orsi et al. [31] study and those detected in desert dust originating from three different deserts (black). In the heatmap, the columns represent individual genera (listed above), and the rows represent different samples. Note, almost all of the fungi detected in desert dust (left panel) were not detected in the subseafloor (right panel) and vice versa. See 󳶳 Table 9.1 for subseafloor sample code names and metadata. The five numbered genera are those that were detected in the subseafloor and desert dust samples.

“paleobes” [10]. This implies at the very least, that these “paleobes” could serve a very useful purpose in paleoclimate reconstruction and for studies of microbial survival, with the suggestion that they are not actively affecting their environment. More recent data suggest that fungi may not simply be surviving, but may also be actively participating in subsurface carbon and other nutrient cycling [31]. The most current definition of a marine fungus, circa 1979, is that of “those that grow and sporulate exclusively in a marine or estuarine habitat” [17]. The first isolation of a deep-sea fungus was first reported approximately 50 years ago in a study

9.3 Fungi in marine near-surface sediments in the deep sea

| 189

of Atlantic Ocean sediments at 4450 m water depth [32]. With increasing findings of fungi (mostly yeasts) in deep-sea environments [26, 33], including deep-sea sediments (e.g. [27]) hydrothermal vents (e.g. [34]), methane seeps and methane hydrates [28, 35], and deep ocean trenches (e.g. [36, 37]), it is becoming apparent that they might play a significant role in these environments, however, a clear differentiation between marine and terrestrial fungi has been lost. Fungi thought to cause coral diseases are sometimes labeled as terrestrial invaders [38]; however some have been shown to be marine representatives of a typically terrestrial group [39]. There is some evidence that terrestrial/surface-dwelling fungi may be capable of colonizing deep-sea habitats due to their ability to alter their membrane composition to accommodate high hydrostatic pressure [40]. Although no piezophilic fungi have been reported to date, the majority of fungi isolated from deep-sea environments have been shown to be halotolerant [9, 26, 41]. Fungi may be prevalent in deep marine sediments due to the presence of a wide range of carbon and nitrogen sources, including decaying algae and higher plants and animals. While there is increasing evidence for fungi that are truly marine (e.g. [42]), many terrestrial fungi may be delivered to the subsurface attached to detritus or may simply sediment as unattached spores or fungal filaments [43]. Experiments with terrestrial and deep-sea isolates suggest that when transported to the deep sea, terrestrial species are initially stressed but are capable of gradual adaptation for growth under 200 bar pressure at 30 or 5 °C [27]. Fungal spores are known to travel in dust originating from the deserts of Africa, Asia and the Middle East, and this dust has been suggested as a possible source of fungal spores to marine sediments [44]. However, only 10% of the fungal genera detected in a recent RNA-based study by Orsi et al. [31] were among those isolated from atmospheric samples originating from desert dust [44], including Cryptococcus, Rhodotorula, Neurospora, Phoma, and Alternaria (󳶳 Fig. 9.1). This suggests the majority of fungi dispersed from deserts into ocean waters via atmospheric movements are not able to penetrate, and/or survive in marine subsurface sediments.

9.3 Fungi in marine near-surface sediments in the deep sea The occurrence of fungi (filamentous and yeasts) has been studied to a much greater degree in near-surface deep marine environments. Whether Basidiomycetes vs. Ascomycetes dominate deep marine sedimentary environments remains debatable, and may depend on local habitats and carbon sources. Nagahama et al. [45] reported that culturable fungal diversity in surface sediments at 2000 m or greater were dominated by ascomycetous yeasts. In a culture-independent study of ten different deep-sea sediment samples off several Japanese islands, including a sample from the Mariana Trench, the majority of phylotypes were novel sequences not related to previously identified fungal sequences in public databases, and included novel members of Ascomycota [29]. In contrast, the dominance of basidiomycetes was observed by Tak-

190 | 9 Subsurface Fungi ishita et al. [35, 46] in a molecular survey of sediments from deep methane seeps, and by Bass et al. [8] who found yeast forms to dominate fungal diversity in deep ocean environments. These studies, however, have not demonstrated in situ fungal activity in deep-sea environments. Culture-dependent methods recover many fungi from various deep-sea environments, but the majority of those are most similar to terrestrial species [9]. Cultureindependent studies have revealed many novel fungal phylotypes, including lineages recently described as Cryptomycota, so named because it is suspected that they lack the typical fungal chitin-rich cell walls (e.g. [9, 41, 47]). Eurotiomycetes are the most commonly detected fungal taxon within the Ascomycota detected in deep-sea environments, with the majority of species affiliating with members of Aspergillus and Penicillium, two groups known to be globally-distributed (e.g. [9, 27]). Within the phylum Basidiomycota, the most commonly-recovered classes are the Exobasidiomycetes (primarily the genus Malassezia), Microboryomycetes (primarily the yeast orders Sporidiobolales and Erythrobasidiales), and Tremellomycetes (primarily Cryptococcus-related phylotypes) [9]. The activity and dominance of Basidiomycota or Ascomycota in deep marine sediments is likely controlled by a multitude of biogeochemical factors that select for different subseafloor lineages [31].

9.4 Fungi in the deep subsurface 9.4.1 Initial whole community and prokaryote-focused studies of the marine subsurface yielding information on eukaryotes There is evidence of abundant microbes in the deep marine subsurface based on microscopy, and this has been confirmed by nucleic acid and lipid studies [48–52]. The Bacteria and Archaea of the subsurface appear to be mostly heterotrophic [50, 53]. Their source of carbon derives originally from photosynthesis in the overlying water column, and these prokaryotes perform a variety of metabolic functions, including sulfate reduction, methanogenesis and fermentation [54]. The rate at which this necromass is turned over may be in the order of thousands of years in the marine subsurface, with consequent impact on biogeochemical cycling on geological time scales [55]. The first comprehensive survey of microbial life in deeply buried marine sediments, Leg 201 of the Ocean Drilling Program (ODP) (in 2002), collected samples from the equatorial Pacific Ocean and the Peru Margin [50]. Using direct plating and enrichments in different media of sediments collected from the upper sulfate – methane transition zone (SMTZ) at 32.15 meters below seafloor (mbsf) down to 157 mbsf, Biddle et al. [56] found some of the first evidence for eukaryotic life in the deep marine subsurface. In addition to their bacterial isolates, Biddle et al. [56] also recovered fungi affiliating with the ascomycetes, and belonging to the genera Cladosporium, Penicillium, and Acremonium. These were similar to genera previously isolated from deep-sea

9.4 Fungi in the deep subsurface |

191

sediments [27]. A metagenomic analysis of ODP Site 1229 on the Peru Margin by Biddle et al. [57] was applied to further explore the microbial diversity and overall community composition within this environment. In this study, eukaryotic gene relatives were observed throughout the sediment column (1–50 mbsf), and included members of the Apicomplexa, Arthropoda, Ascomycota, Basidiomycota, Chlorophyta, Chordata, Echinodermata, Microsporidia, Nematoda, Platyhelminthes and Streptophyta. This cross-section of eukaryotic diversity clearly supported previous observations of successfully preserved DNA in the marine subsurface [58], since sequences affiliating with most of these taxonomic groups clearly are detrital, and/or originated from the overlying water column. The one taxonomic group that was represented throughout all sediment depths was Ascomycota, and included some of the taxa previously cultivated from Site 1229 sediments [57]. Additional evidence for eukaryotes in the deep subsurface comes from analysis of basement fluids of the sediment-buried Juan de Fuca Ridge Flank [59]. Using a unique sampling system called a Circulation Obviation Retrofit Kit (CORK) observatory at the Integrated Ocean Drilling Program (IODP) borehole 1301A located at 2667 m water depth in the Pacific Ocean, aquifer fluids in 3.5 million year old basalt-hosted basement were sampled. Surprisingly, overall total cell counts were only one order of magnitude less in these waters than in surrounding bottom seawater. DNA-based signatures within clone libraries generated from bottom waters around the borehole included fungi, Alveolata, Excavata, Hacrobia, Picobiliphyta, Metazoa, Rhizaria and Stramenopiles, however only signatures of Radiolaria and Polycystinea were recovered from borehole fluids (see Supplementary Data in [59]). However, the absence of fungi in their clone libraries may have been an artifact of primer or other methodological bias, since Smith et al. [60] observed and cultured microfungi (Rhodotorula sp.) from minerals incubated in the same hole (1301A).

9.4.2 Eukaryote-focused studies yielding information on fungi in the deep subsurface The first investigation of microbial diversity in deep subsurface core samples that targeted eukaryotes exclusively, was an analysis of eukaryotic small subunit ribosomal RNA (18S rRNA) from marine subsurface sediment samples from Peru Margin and Peru Trench [4]. In that analysis, DNA and RNA were extracted from selected subsurface sediment samples collected during ODP Leg 201. The depth distribution of eukaryotes was expected to follow a similar dynamic to what had been previously found for prokaryotes, and so the Edgcomb et al. [4] study analyzed a near-surface sample (3.91 mbsf), and a deeper sample from the SMTZ (35.9 mbsf) from ODP site 1228 (Peru Margin, 262 m water), and a near-surface sample (1.75 mbsf) from ODP 1230 (Peru Trench, 5086 m water). All sediments had a total organic carbon (TOC) (w%) of 3– 10% [61].

192 | 9 Subsurface Fungi Historically, most sequence-based studies of microbial diversity within environmental samples, including the marine subsurface, have utilized PCR amplification of target genes, most commonly small subunit ribosomal RNA (SSU rRNA), from DNA extracted from the environmental sample as a starting material. Because DNA extracts include nucleic acids from active cells, inactive but viable cells, dead cells and extracellular DNA from lysed or degraded cells, DNA pools do not exclusively represent living organisms. To more accurately capture representatives of the active fraction of the microbial community, reverse transcription of rRNA followed by PCR amplification is now commonly used as a proxy for living/metabolically active microbes in SSU rRNA based environmental surveys. The pairing of DNA- and RNA-based analyses in the Edgcomb et al. [4] study allowed a first view into the fraction of diversity detectable with DNA that may represent active eukaryotic microbes that maintain enough ribosomal RNA to be detectable in total RNA pools. The 18S rRNA gene was amplified from both DNA and cDNA preparations for each horizon using a nested PCR approach and eukaryote-specific primers [4]. Of all eukaryotic signatures, Basidiomycota were the dominant sequence types recovered in all cDNA-based clone libraries, suggesting for the first time that fungi dominate the eukaryotic microbial community in the deep marine subsurface [4]. Unexpected sequences in the cDNA pool, such as from animals, green algae and red algae suggest even ribosomal RNA may be preserved for some taxa under certain conditions. Preserved ribosomal RNA from at least some of these groups may have survived in cysts or pollen. The greatest taxonomic diversity was observed in the shallowest sample from 1.75 mbsf (Peru Trench). DNA- and cDNA-based libraries were roughly congruent, however differences between them indicated that working from extracted RNA may be more reliable for identifying active members of the community. Consistent with other studies of deep marine near-surface sediments, some recovered sequences in the Edgcomb et al. [4] study have nearly identical 18S rRNA gene sequences to their closest matches in GenBank from marine surface sediment or pelagic environments, whereas many have clearly different 18S sequences from their closest matches. Sequences affiliating with the Basidiomycetes, included single-celled forms (yeasts) and asexual species (smuts, yeasts, rusts, jelly fungi) that are found in virtually all terrestrial ecosystems, as well as freshwater and marine habitats [12, 17, 62, 63]. Some sequences affiliated with uncultured fungal sequences from Guaymas Basin hydrothermal sediments [3, 6], the Lucky Strike vents at the Mid-Atlantic Ridge [64], Cariaco Basin anoxic water column [7], and nonhydrothermal deep-sea sediments [8]. An rRNA-based approach is especially warranted for subsurface molecular investigations given documentation of DNA ‘paleomes’ in subsurface samples [58]. Extracellular DNA and DNA preserved in structures such as cysts, spores and pollen in sediments, and from deposition of dissolved DNA, DNA bound to detritus, cysts, endospores, intact pollen, cellular exudates and excretion following cell death, protistan grazing and viral lysis is thought to represent one of the largest reservoirs of DNA in the world [65, 66]. Such DNA ‘paleomes’ are a sedimentary record of past

9.4 Fungi in the deep subsurface | 193

Table 9.1: Samples examined in the Orsi et al. study [31] with corresponding metadata. Subseafloor samples Site

North Pond

Hydrate Ridge

Sample code Depth (mbsf) Sampling site

NP 1.6 GeoB 13507-1 120 0 0.2 2.15 34

HR 1.8 IODP 1244a

O2 (μM) Sulfide (μM) TOC (% wt) DIC (mM) NO− 3 (μM)

0 1000 1.5 40 0

Benguela Upwelling System BSP 4.6 GeoB 12805-1 0 3000 3.8 12 0

Eastern Equatorial Pacific EEP 45.3 IODP 1225a

Peru Margin

PM 48.1 IODP 1227a

0 0 0.01 3 0

0 6130 3.6 24 0

microbial communities [67], and have been reported and employed to understand paleo-environments and succession of species as a result of environmental change in core samples of various ages including 100 million-year-old continental drilled black shale [67, 68], and deep marine sediments [58, 69]. To expand upon previous investigations of deep subsurface microbiota, to further test whether fungi are indeed the dominant eukaryotes, and to find additional evidence for activity of subsurface fungi, a study of globally-distributed subsurface samples using an rRNA-based amplicon pyrosequencing approach was conducted [31]. This study used a particularly stringent set of contamination controls to rule out contamination of samples in the laboratory. Amplicon libraries of eukaryotic V4 rRNA were prepared from sediments from the Eastern Equatorial Pacific (45.1 mbsf), Peru Margin (48.1 mbsf), Hydrate Ridge (1.8 mbsf), North Pond (1.6 mbsf), and the Benguela Continental Slope (4.6 mbsf) (󳶳 Table 9.1). Consistent with the findings of Edgcomb et al. [4], analysis of these data revealed the presence of signatures of taxa not expected to be alive in the deeply-buried subsurface (sediments up to 2 million years old), including Diatoms, Viridiplantae and Metazoa. In light of the extensive contamination controls employed in that study (including tests for complete removal of DNA and rRNA contamination from aerosols), and the average pore size of these sediments, which is less than 1 μm, the detection of plant, diatom and probably metazoan rRNA suggests that rRNA from at least some eukaryotes is preserved in deep marine sediments. For example, RNA signatures of the diatom Nitzschia were detected in the cDNA amplicon libraries from the same horizons where fossilized Nitzschia were used to date the Eastern Equatorial Pacific sample (∼2.7 mya) [31]. The preservation of extracellular rRNA is unlikely because of the possibility for degradation by extracellular RNases. However, until preservation of extracellular rRNA in anoxic marine sediments is specifically investigated, rRNA-sequencebased molecular approaches should be integrated with other approaches including

194 | 9 Subsurface Fungi geochemical correlations and microscopy, and analysis of other expressed genes, to obtain a more realistic assessment of the active fraction of subsurface microbial communities. In the Orsi et al. study [31], correlations between broad-ranging taxonomic trends and geochemical factors were investigated in order to distinguish signals that could be coming from potentially active subsurface eukaryotes from those within the rRNA ‘paleome.’ Canonical correspondence analyses (CCA) paired with Multi-Response Permutation Procedure (MRPP) was used to identify groups correlating with TOC, dissolved inorganic carbon (DIC), sulfide, or sediment depth. As a group, sequences from plants showed a relatively high correlation with TOC, potentially reflecting their detrital origin and contribution to TOC and/or preserved pollen. Fungi showed the strongest correlation with TOC across most levels of taxonomic hierarchy, consistent with a potentially saprophytic lifestyle. The relatively strong correlation of fungi with DIC, plants and TOC, may reflect fungal metabolism of organic substrates. Two to three times more fungal-affiliated sequences were recovered from subsurface sediments with higher (1.5–3.7 % wt) vs. lower TOC (0.2–0.01 % wt) (󳶳 Fig. 9.2). This was consistent with the statistically significant influence of TOC on determining the composition of subsurface eukaryotic communities in that study [MRPP𝑝 value = 0.02 (+/−0.01)] [31]. A least squares regression of fungal sequence representation in the subsurface reveals

Fig. 9.2: Least squares regression of the fungal representation in eukaryotic rRNA 454 sequence data from the Orsi et al. [31] study plotted against sediment TOC content. Note the statistically significant (𝑝 < 0.04) correlation between TOC content and fungal abundance.

9.4 Fungi in the deep subsurface | 195

that fungi increase in representation as a function of total organic carbon content (𝑝 < 0.04, 󳶳 Fig. 9.2). Evidence for environmental selection of subsurface fungi across a broad range of subsurface provinces (󳶳 Fig. 9.3) comes from an observed disparity in empirical and estimated fungal richness in anoxic and oxygenated sediments. Empirical and estimated fungal richness in the oxygenated North Pond sediments was at least double that of the anoxic samples [31]. Furthermore, the fungal communities in anoxic, sul-

Fig. 9.3: Unifrac analysis of fungal rRNA data from the Orsi et al. [31] study showing phylogenetic relatedness between subseafloor samples and shallow, near-surface sediment samples taken from Sippewissett Salt Marsh (Cape Cod, MA). The heatmap is a representation of a unifrac dissimilarity matrix. Darker color indicates a similar community while lighter color indicates dissimilarity. Histograms show the abundance of sequences affiliating with major eukaryotic lineages.

196 | 9 Subsurface Fungi

Fig. 9.4: Calcofluor staining of filamentous- (a), spore- (b), and yeast- (c) like cellular forms from Eastern Equatorial Pacific sediments 45 mbsf. Scale bars: 50 μm (a), 10 μm (b), 20 μm (c).

fidic sediments were distinct from the community in an anoxic sediment sample with no detectable sulfide. Taxonomic groups known to contain fermentative representatives, such as, Candida, Rhodotorula, Rhodosporidium, and Trichosporon were found in amplicon libraries from anoxic subsurface samples. Only one oxic sample was used in this comparison. More samples from oxygenated subseafloor samples are needed to confirm or disprove these preliminary results. Microscopic evidence of intact fungal cells and spores in sediments from the Eastern Equatorial Pacific (45 mbsf) was revealed by calcofluor staining (󳶳 Fig. 9.4) [31]. Cellular forms similar to yeasts, filamentous fungi and spores were visualized, consistent with the rRNA detection of fungal taxa with these morphologies at this site,

Fig. 9.5: Regression (R2 ) and Spearman (rho) analyses of fungal diversity (OTU richness) as it relates to sample distance from shore.

9.6 Summary

|

197

supporting the notion of vegetative fungal cells in the marine subsurface that may be contributing to the recycling of organic carbon over geological time scales [31]. If fungi in marine sediments were dispersed from terrestrial locations via runoff, then diversity would be expected to decrease with increased distance from shore due to greater distance from the input material. However, in our data we found an increase in observed and estimated fungal Operational Taxonomic Units (OTUs) with increased distance from shore (󳶳 Fig. 9.5). While the source of fungi in deep sediments remains unknown, physical isolation and environmental selection in the subsurface likely contribute to their phylogenetic uniqueness in different locations.

9.5 How deep do fungi go in the subsurface? Considering the discovery that eukaryotes are part of the subsurface biosphere in borehole fluids [59, 60], it is conceivable that eukaryotes, in particular fungi, continue past the sedimentary-deep biosphere into the basalt below. The source of the borehole fluids in the Juan de Fuca Ridge system are circulated across the basaltic aquifer before entering the boreholes fitted with the CORK installations. It is unknown where the fluids would entrain fungal cells: were they present in seawater and carried or picked up in the sediments or basalts? The concept of fungal life in deep basalts comes from geologic investigations, where evidence of fungal activity is seen. At ODP Site 1224, filamentous fossil structures were seen in carbonate-filled vesicles in an area of lava flow that was 46 Ma [70]. Similarly, filamentous structures are seen in multiple basalts from the Emperor Seamounts, ODP Sites 1203–1206, to a depth of 936 mbsf [71], some of which were shown to have networks similar to those of the Ascomycota [72]. These preserved structures join the evidence from volcanic glasses and ophiolites (reviewed in [72]) to show that the Eukaryotic biosphere stretches into deep basalts. It is well known on the surface environment that fungi interact with basalts, sequestering metals and causing degradation of basalt by production of organic acids [73]. While no known signatures of active fungi have been found from basalts directly, it is extremely likely that fungi are indeed members of the basaltic deep biosphere since they appear in borehole fluids and fossil structures.

9.6 Summary It may come as a surprise that fungi are alive and well in the deep biosphere. The deep marine biosphere has presented many surprises, from organisms surviving on incredibly low energy flux [73] to potentially dominated by Archaea [48, 49]. Subsurface eukaryotes were first a surprise, partially due to their appearance in cultivations, where they were assumed to be a record of paleoclimate and not actually active [10].

198 | 9 Subsurface Fungi However, the growing evidence that fungi are an adapted population for both aerobic and anaerobic sediment, gives credence to the idea that fungi are an active population, despite pore space restrictions and sediment age [31]. The depth and extent of the subsurface fungal biosphere is still being tested. It is likely that the fungal-deep biosphere may extend just as deep as the prokaryotic-deep biosphere in subseafloor sediment and basalt, and may contribute to the marine carbon pump through metabolism of otherwise recalcitrant organic substrates in the marine subsurface.

Acknowledgements Work performed by the authors and included in this overview was made possible by funding from the Center for Dark Energy Biosphere Investigations (C-DEBI) and The Ocean Life Institute (WHOI) to VE (OLI-27071359). We thank Dr Edward Leadbetter (WHOI) for valuable discussions during our studies of subsurface sediments, including those on possible sources of dust-associated aerial contamination.

References [1]

Kallmeyer, J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA (2012), 16 213– 16 216, doi: 10.1073/pnas.1203849109. [2] López-García P, Rodriguez-Valera F, Pedros-Alio C, Moreira D. Unexpected diversity of small eukaryotes in deep-sea Antarctic plankton. Nature 409 (2001), 603–607. [3] Edgcomb VP, Kysela DT, Teske A, Gomez AD, Sogin ML. Benthic eukaryotic diversity in the Guaymas Basin hydrothermal vent environment. Proc Natl Acad Sci USA 99 (2002), 7658– 7662. [4] Edgcomb VP, Beaudoin D, Gast R, Biddle JF, Teske A. Marine subsurface eukaryotes: the fungal majority. Env Microbiol 13 No. 1 (2010), 172–183. [5] López-García P, Philippe H, Gail F, Moreira D. Autochthonous eukaryotic diversity in hydrothermal sediment and experimental microcolonizers at the Mid-Atlantic Ridge. Proc Natl Acad Sci USA 100 (2003), 697–702. [6] López-García P, Vereshchaka A, Moreira D. Eukaryotic diversity associated with carbonates and fluid-seawater interface in Lost City hydrothermal field. Environ Microbiol 9 (2007), 546– 554. [7] Stoeck T, Epstein S. Novel eukaryotic lineages inferred from small-subunit rRNA analyses of oxygen-depleted marine environments. Appl Environ Microbiol 69 (2003), 2657–2663. [8] Bass D, Howe A, Brown N, Barton H, Demidova M, Michelle H et al. Yeast forms dominate fungal diversity in the deep oceans. Proceedings of the Royal Society B- Biological Sciences 274 (2007), 3069–3077. [9] Nagano Y, Nagahama T. Fungal diversity in deep-sea extreme environments. Fungal Ecology 5 (2012), 463–471. [10] Raghukumar C, Raghukumar S, Sheelu G, Gupta SM, Nath BNN, Rao BR. Buried in time: culturable fungi in a deep-sea sediment core from the Chagos Trench, Indian Ocean. Deep-Sea Research I 51 (2004), 1759–1768.

References | 199

[11] Thaler AD, Van Dover CL, Vilgalys R. Ascomycete phylotypes recovered from a Gulf of Mexio methane seep are identical to an uncultured deep-sea fungal clade from the Pacific. Fungal Ecology 5 (2012), 270–273. [12] Kohlmeyer J, Kohlmeyer E. Marine Mycology: The Higher Fungi. Academic Press: New York, USA, 1979. [13] Fell JW, Master IM. Fungi associated with the degradation of mangrove (Rhizophora mangle L.) leaves in south Florida. Estuarine Microbial Ecology. Editors: Stevenson and Colwell. Univ. South Carolina Press, Columbia, the Belle W. Baruch Library in Mar Sci., 1, 1973. [14] Fell JW, Master IM and Newell SY. Laboratory model of the potential role of fungi (Phytophthora spp.) in the decomposition of the red mangrove (Rhizophora mangle) leaf litter. In: Tenore KR, Coull BC (eds). Marine Benthic Dynamics. Univ. of So. Carolina Press, Columbia 11 (1980), 359–392. [15] Hyde KD, Jones EBG, Leao E, Pointing SB, Poonyth AD, Vrjmoed LLP. Role of fungi in marine ecosystems. Biodivers Conserv 7 (1998), 1147–1161. [16] Jobard M, Rasconi S, Sime-Ngando T. Diversity and functions of microscopic fungi: a missing component in pelagic food webs. Aquat Sci 72 (2010), 255–268. [17] Kohlmeyer J. Marine Fungal Pathogens among Ascomycetes and Deuteromycetes. Experientia 35 (1979), 437–439. [18] Newell SY, Fell. JW. Parallel testing of media for measuring frequencies of occurrence for Halophytophthora spp. (Oomycota) from decomposing mangrove leaves. Can J Micro 40 (1994), 250–256. [19] Newell SY, Fell JW. Competition among mangrove oomycotes and between oomycotes and other microbes. Aquatic Microbial Ecology 12 No. 1 (1997), 21–28. [20] Statzell-Tallman A, Belloch, Fell JW. Kwoniella mangroviensis gen. nov., sp. nov. a tremellaceous yeast from mangrove habitats in the Florida Everglades and Bahamas. FEMS Yeast Research 8 (2008), 103–113. [21] Wegley L, Edwards R, Rodriguez-Brito B, Liu H, Rohwer F. Metagenomic analysis of the microbial community associated with the coral Porites asteoides. Environmental Microbiology 9 (2007), 2707–2719. [22] Burgaud G, Arzur D, Durand L, Cambon-Bonavita M-A, Barbier G. Marine culturable yeasts in deep-sea hydrothermal vents: species richness and association with fauna. FEMS Microb Ecol 73 (2010), 121–133. [23] Gadd GM. Geomycology: biogeochemical transformations of rocks, minerals, metals and radionuclides by fungi, bioweathering and bioremediation. Mycological Research 111 (2007), 3–49. [24] Adl SM, Simpson AGB, Farmer MA, Andersen RA, Anderson OR, Barta JR et al. The new higher level classification of eukaryotes with emphasis on the taxonomy of protists. Journal of Eukaryotic Microbiology 52 (2005), 399–451. [25] Richards TA, Bass D. Molecular screening of free-living microbial eukaryotes: diversity and distribution using a meta-analysis. Curr Opin Microbiol 8 (2005), 240–252. [26] Burgaud G, Le Calvez T, Arzur D, Vandenkoornhuyse P, Barbier G. Diversity of culturable marine filamentous fungi from deep-sea hydrothermal vents. Env Microbiology 11 (2009), 1588–1600. [27] Damare S, Raghukumar C, Raghukumar S. Fungi in deep-sea sediments of the Central Indian Basin. Deep Sea Res Part I Oceanographic Research Papers 53 (2006), 14–27. [28] Lai X, Cao L, Tan H, Fang S, Huang Y, Zhou S. Fungal communities from methane hydrate-bearing deep-sea marine sediments in South China Sea. ISME J 1 (2007), 756–762. [29] Nagano Y, Nagahama T, Hatada Y, Nunoura T, Takami H, Miyazaki J, Takai K, Horikoshi K. Fungal diversity in deep-sea sediments – the presence of novel fungal groups. Fungal Ecology 3 (2010), 316–325.

200 | 9 Subsurface Fungi [30] Singh P, Raghukumar C, Meena RM, Verma P, Shouche Y. Fungal diversity in deep-sea sediments revealed by culture-dependent and culture-independent approaches. Fungal Ecology 5 (2012), 543–553. [31] Orsi W, Biddle J, Edgcomb VP. Deep sequencing of subseafloor eukaryotic rRNA reveals active fungi across marine subsurface provinces. PLoS ONE 2013; http://dx.plos.org/10.1371/journal.pone.0056335 [32] Roth FJ, Orpurt PA, Ahearn DJ. Occurrence and distribution of fungi in a subtropical marine environment. Can J Botany 1964; 42:375–383. [33] Connell LB, Barrett A, Templeton A, Staudigel H. Fungal diversity associated with an active deep sea volcano: Vailulu’s Seamount, Samoa. Geomicrobiology Journal 26 (2009), 597–605. [34] Gadanho M, Sampaio JP. Occurrence and diversity of yeasts in the Mid-Atlantic Ridge hydrothermal fields near the Azores Archipelego. Microb Ecol 50 (2005), 408–417. [35] Takishita K, Tsuchiya M, Reimer JD, Maruyama T. Molecular evidence deomonstrating the basidiomycetous fungus Cryptococcus curvatus is the dominant microbial eukaryote in sediment at the Kuroshima Knoll methane seep. Extremophiles 10 (2006), 165–169. [36] Takami H, Inoue A, Fuji F, Horikoshi K. Microbial flora in the deepest sea mud of the Mariana Trench. FEMS Microbiol Lett 152 No. 2 (1997), 279–285. [37] Nagahama T, Abdel-Wahab MA, Nogi Y, Miyazaki M, Uematsu K, Hamamoto M, Horikoshi K. Dipodascus tetrasporeus sp. Nov., an ascosporogenous yeast isolated from deep-sea sediments in the Japan Trench. International J of Systematic and Evolutionary Microbiol 58 (2008), 1040–1046. [38] Ravindran J, Raghukumar C, Raghukumar S. Fungi in Porites lutea: association with healthy and diseased corals. Dis Aquat Organ 47 (2001), 219–228. [39] Geiser DM, Taylor JW, Ritchie KB, Smith GW. Cause of sea fan death in the West Indies. Nature 394 (1998), 137–138. [40] Simonato F, Campanaro S, Lauro FM, Vezzi A, D’Angelo M, Vitulo N et al. Piezophilic adaptation: a genomic point of view. Journal of Biotechnology 126 (2006), 11–25. [41] Le Calvez T, Burgaud G, Mahe S, Barbier G, Vandenkoornhuyse P. Fungal diversity in deep sea hydrothermal ecosystems. App Environ Microbiol 75 (2009), 6415–6421. [42] Burgaud G, Woehlke S, Redou V, Orsi W, Beaudoin D, Barbier G, Biddle JF, Edgcomb VP. Deciphering presence and activity of fungal communities in marine sediments using a model estuarine system. Aquatic Microbial Ecol 70 (2013), 45–62. [43] Lorenz R, Molitoris HP. Cultivation of fungi under simulated deep sea conditions. Mycological Research 101 (1997), 1355–1365. [44] Griffin DW. Atmospheric movement of microorganisms in clouds of desert dust and implications for human health. Clin Microbiol Rev 20 No. 3 (2007), 459–477. [45] Nagahama T, Hamamoto M, Nakase T, Takami H, Horikoshi K. Distribution and identification of red yeasts in deep-sea environments around the northwest Pacific Ocean. Antonie van Leeuwenhoek 80 (2001), 101–110. [46] Takashita K, Yubuki N, Kakizoe N, Inagaki Y, Maruyama T. Diversity of microbial eukaryotes in sediment at a deep-sea methane cold seep: surveys of ribosomal DNA libraries from raw sediment samples and two enrichment cultures. Extremophiles 11 (2007), 563–576. [47] Nagahama T, Takahashi E, Nagano Y, Abdel-Wahab MA, Miyazaki M. Molecular evidence that deep-branching fungi are major fungal components in deep-sea methane cold-seep sediments. Environ Microbiol 13 (2011), 2359–2370. [48] Lipp JS, Morono Y, Inagaki F, Hinrichs KU. Significant contribution of Archaea to extant biomass in marine subsurface sediments. Nature 454 (2008), 991–994.

References | 201

[49] Biddle JF, Lipp JS, Lever MA, Lloyd KG, Sørensen KB, Anderson R et al. Heterotrophic Archaea dominate sedimentary subsurface ecosystems off Peru. Proc Natl Acad Sci USA 103 (2006), 3846–3851. [50] D’Hondt S, Jørgensen, BB, Miller DJ et al. Distributions of microbial activities in deep subseafloor sediments. Science 306 (2004), 2216–2221. [51] Whitman WB, Coleman DC, Wiebe WJ. Prokaryotes: the unseen majority. Proc Natl Acad Sci USA 95 (1998), 6578–6583. [52] Parkes RJ, Cragg BA, Wellsbury P. Recent studies on bacterial populations and processes in subseafloor sediments: A review. Hydrogeol J 8 (2000), 11–28. [53] D’Hondt S, Spivack AJ, Pockalny R et al. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 2009; 106(28):11 651–11 656. [54] D’Hondt S, Rutherford S, Spivack AJ. Metabolic activity of subsurface life in deep-sea sediments. Science 295 No. 5562 (2002), 2067–2070. [55] Lomstein BA, Langerhuus AT, D’Hondt S, Jørgensen BB, Spivack AJ. Endospore abundance, microbial growth and necromass turnover in deep subseafloor sediment. Nature 484 No. 7392 (2012), 101–104. [56] Biddle JF, House CH, Brenchley JE. Microbial stratification in deeply buried marine sediment reflects changes in sulfate/methane profiles. Geobiology 3 No. 4 (2005), 287–295. [57] Biddle JF, Fitz-Gibbon S, Schuster SC, Brenchley JE, House CH. Metagenomic signatures of the Peru Margin subseafloor biosphere show a genetically distinct environment. Proc Natl Acad Sci USA 105 (2008), 10 583–10 588. [58] Coolen MJ, Overmann J. 217,000-year-old DNA sequences of green sulfur bacteria in Mediterranean sapropels and their implications for the reconstruction of the paleoenvironment. Environ Microbiol 9 No. 1 (2007), 238–249. [59] Jungbluth SP, Grote J, Lin H-T, Cowen JP, Rappe MS. Microbial diversity within basement fluids of the sediment-buried Juan de Fuca Ridge Flank. ISME J 1–12 (2012), 1751–7362. [60] Smith A, Popa R, Fisk M, Nielsen M, Wheat CG, Jannasch HW, Fisher AT, Becker K, Sievert SM, Flores G. In situ enrichment of ocean crust microbes on igneous minerals and glasses using an osmotic flowthrough device. Geochem Geophys Geosyst 12 (2011), doi:10.1029/2010GC003424. [61] Meister P, Prokopenko M, Skilbeck CG, Watson M, McKenzie JA. Data report: Compilation of total organic and inorganic carbon data from Peru Margin and eastern equatorial Pacific drill sites (ODP legs 112, 138 and 201). In: Proc. ODP, Sci. Results, Jørgensen BB, D’Hondt S, Miller DJ, eds. (2005), 1–20. [62] Binder M, Hibbett DS. Higher-level phylogenetic relationships of homobasidiomycetes (mushroom-forming fungi) inferred from four rDNA regions. Molecular Phylogenetics and Evolution 22 (2002), 76–90. [63] Hibbett DS, Binder M. Evolution of complex fruiting-body morphologies in homobasidiomycetes. Proceedings of the Royal Society of London Series B-Biological Sciences 269, 1963–1969, 2002. [64] Le Calvez T. Third Annual DOE Joint Genome Institute User Meeting. US Dept. of Energy, Office of Science: Walnut Creek, CA, 2008. [65] Dell’Anno A and Danovaro R. Extracellular DNA play a key role in deep-sea ecosystem functioning. Science 309 (2005), 1497. [66] Corinaldesi C, Barucca M, Luna GM, Dell’Anno A. Preservation, origin and genetic imprint of extracellular DNA in permanently anoxic deep-sea sediments. Mol Ecol 20 (2011), 642–654. [67] Inagaki F, Okada H, Tsapin AI, Nealson KH. Microbial survival: the paleome: a sedimentary genetic record of past microbial communities. Astrobiology 5 No. 2 (2005), 141–153.

202 | 9 Subsurface Fungi [68] Inagaki F, Nunoura T, Nakagawa S, Teske A, Lever M, Lauer A et al. Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments, on the Pacific Ocean Margin. Proc Natl Acad Sci USA 103 (2006), 2815–2820. [69] Boere AC, Rijpstra WI, De Lange GJ, Sinninghe Damste JS, Coolen MJ. Preservation potential of ancient plankton DNA in Pleistocene marine sediments. Geobiology 9 No. 5 (2011), 377–393. [70] Schumann G, Manz W, Reitner J, Lustrino M. Ancient fungal life in North Pacific eocene oceanic crust. Geomicrobiology Journal 21 (2004), 241–246. [71] Ivarrson, M, Lausmaa J, Lindblom S, Broman C, Holm NG. Fossilized microorganisms from the Emperor Seamounts: Implications for the search for a subsurface fossil record on Earth and Mars. Astrobiology 8 (2008), 1139–1157. [72] Ivarsson, M, Bengtson S, Belivanova V, Stampanoni M, Marone F, Tehler A. Fossilized fungi in subseafloor Eocene basalts. Geology 40 (2012), 163–166. [73] Mehta AP, Toma AE, Murr LE. Effect of environmental parameters on the efficiency of biodegradation of basalt rock by fungi. Biotechnology and Bioengineering 21 (1979), 875–885. [74] Hoehler, T and Jørgensen BB. Life under extreme energy limitation. Nature Rev Micro 11 (2013), 83–94.

Mashal Alawi

10 Microbes in geo-engineered systems: geomicrobiological aspects of CCS and Geothermal Energy Generation 10.1 Introduction Reduction of greenhouse gas emissions is one of today’s major challenges. Most anthropogenic CO2 emissions derive from fossil fuels used for generating electricity and it is expected that fossil fuels will provide a major part of the world’s energy portfolio during the twenty-first century [1]. To reduce the emission of greenhouse gases various technical approaches are currently under debate [2, 3]. In this chapter I will focus on the interaction of the microbial deep biosphere in geo-engineered systems used for “Carbon Capture and Storage” (CCS) and geothermal energy generation. Nielson et al. [4] described synergy benefits in combining CCS and geothermal energy generation. The idea behind the coupling of these two techniques is that the net removal of water from the reservoir can be balanced by the injected CO2 . Furthermore, the CO2 plume migration may be controlled by adjusting the different surrounding geothermal sites as pressure sinks and sources. The preliminary design and reservoir engineering for a pilot-scale deployment at the SECARB Cranfield Phase III CO2 Storage Project, in Cranfield, Mississippi, U.S.A was recently presented [5]. However, a competition for appropriate geological sites (i.e. with respect to possible fault reactivation or migration of deep saline brines into shallower freshwater aquifers [6]) and funding is still under debate. It is known that processes such as corrosion, scaling and precipitation of minerals could be induced or enhanced by microorganisms and seriously harm the efficiency and reliability of geo-engineered systems [7–12]. Due to a rapidly increasing amount of data becoming available, the understanding of hydro- and geochemical processes reached a level where many potentially harmful effects can be predicted [13– 18]. However, many aspects of the interactions between geo-engineered systems and deep subsurface microbial ecosystems remain poorly understood [19–21]. Colwell and D’Hondt [22] concluded that geo-engineering can change the subsurface environment by redox stratification, fluid movement, fracturing, seismicity and groundwater fluctuation and thereby most likely transform subsurface deserts into oases for microbial life. Nevertheless, there are also indications that the natural deep subsurface microbial community could be negatively affected by changes in the fluid and reservoir chemistry [23], temperature changes [24], relocation of substances [25, 26] and porosity and permeability changes of the reservoir [27].

204 | 10 Microbes in geo-engineered systems Moreover, the introduced geo-engineered systems provide interesting insights into the deep biosphere.

10.1.1 Carbon Capture and Storage (CCS) In the last decade, CCS was intensively discussed as a transitional step to reduce carbon dioxide emission from coal-fired power plants or industrial plants. Currently 33 CCS facilities are in operation worldwide (21 of which are pilot plants) and another 52 facilities are already planned [28]. The concept of CCS is injecting large amounts of CO2 for long-term storage in suitable geological formations [1, 29, 30]. Large volumes of CO2 can be sequestered in coal beds as well as sedimentary basins, which have a tremendous pore volume, connectivity and appropriate cap rock structures [29–32]. CO2 has already been injected into geological formations for several decades for various purposes, including enhanced oil and gas recovery (EOR/EGR). The first commercial example of enhanced oil recovery was Weyburn-Midale (U.S.A./Canada) where between the years 2000 and 2011 nearly 25 million tons of CO2 were injected [33]. The longest-running CCS-EGR operation in Europe is the Sleipner Gas field off Norway. The worldwide storage potential is dominated by deep saline aquifers, which are porous formations saturated with saline water, thus having no economic or technical use. Another uprising idea is the sequestration of CO2 in basalts (e.g. CarbFix project in Iceland and the Kevin Dome Large Scale Storage Project in U.S.A.) [34, 35]. In particular deep-sea basalts, characterized by seawater-filled pore space and Mg-Ca silicate rocks are adequate [36, 37]. In these reservoirs, CO2 can be fixed by geochemical trapping as stable carbonates. Various reservoir capacities were identified along the eastern ridge flank of the Juan de Fuca plate [38] and other oceanic ridge-flanks [37]. In the first step of the CCS chain, the industrially produced carbon dioxide has to be separated from other effluent gases. The type of technology used for the separation determines the degree of purity of the obtained CO2 and the nature of impurities. Remaining traces of SO2 and/or NO2 , for example, which typically derive from oxyfuel combustion plants, can form H2 SO4 and HNO3 upon contact with water, thus possibly accelerating the alteration and corrosion processes in the reservoir and in the injection facilities (pumps, tubings, borehole cements) [39, 40]. By trapping CO2 in underground formations, its climate-relevant role as a greenhouse gas can be significantly reduced [31]. Four different mechanisms contribute to the immobilization of CO2 in the deep underground: structural trapping, residual trapping, solubility trapping, and mineral trapping [1, 41], each with a characteristic and site-specific relative importance and time scale. In the short-term, structural and residual trapping generally dominate, whereas solubility and mineral trapping might be only of minor significance due to the slowness of diffusion – the transport process controlling the migration of dissolved CO2 after reaching hydrodynamic equilibrium – and of mineral reaction kinetics. However, in the long-term, the situation can reverse.

10.1 Introduction |

205

Numerical models for a Triassic saline aquifer predicted that after a simulation time of 10,000 years, up to about 25% of the injected CO2 could be trapped in newly-formed carbonate minerals, and the vast majority of injected CO2 dissolved in the formation fluid [42, 43].

10.1.2 Geothermal energy and aquifer energy storage The geothermal energy sector is still expanding; nowadays there are 674 geothermal power plants in development worldwide [44]. The report from Matek [44] lists 70 countries that have geothermal projects slated for development. Further development of geothermal plants will also depend on enhanced long-term reliability and cost effectiveness. Here, I will focus on deep geothermal power plants (> 500 m) and sedimentary geologic formations that provide at least suitable conditions for temporary aquifer heat and cold storage (ATES, aquifer thermal energy storage) [45]. Since several geological formations adequate for the installation of ATES are located in a depth of more than 1 km, this aspect of the geotechnical usage of the underground also needs to be discussed. Multiple techniques have been developed to efficiently use geothermal energy in different hydrothermal systems and petrothermal hot dry rock (HDR) formations. Technical approaches like hydraulic fracturing enable the geotechnical usage of deep, sedimentary geothermal reservoirs, which only have a modest productivity initially [46, 47]. The classical setup of a hydrothermal geothermal plant consists of a production and an injection well (a so-called geothermal doublet system) and the hot thermal waters can be used without intensively stimulating the hydraulic reservoir properties. The reinjection through a second well is the most common and recommended method of disposal of the natural fluids. Reinjection of the fluids may also help to maintain the reservoir pressure and prolong the operation time of the plant. If geothermal plants are releasing the natural fluids into surface environments e.g. rivers, high loads of salts and pollutants like arsenic might have strong impacts on the affected ecosystem [48]. Most of the geothermal energy used so far is found in high-enthalpy settings, which are marked by a high geothermal gradient (increase of heat with depth). Apart high temperatures, geothermal reservoirs require high permeability of the rocks to allow high production rates of the hot fluid to the surface [49]. To increase permeability, so-called Enhanced Geothermal Systems (EGS) are created by hydraulic stimulation of the reservoir [49–51]. In areas with average geothermal gradients (ca. 3°C/100 m), like Germany, the fluids with temperatures higher than 100°C, originate from depths greater than 3 km. During hydraulic stimulation, water is pumped into the rock under high pressure, which induces fractures and thereby increase the formation permeability and create a reservoir for the fluid [52].

206 | 10 Microbes in geo-engineered systems

10.2 Microbial diversity in geo-engineered reservoirs The environmental conditions in aquifers or rock formations that are adequate for geotechnical usage are heterogeneous and so are the microbial communities that reside in there. Appropriate geological formations for CCS and geothermal energy are plentiful and diverse and new techniques increase the number of possible on- or offshore sites worldwide, which also increases the diversity of subsurface environments that are affected by anthropogenic activity. Only a few studies analyzed the microbial community in reservoir rocks in the frame of a baseline characterization prior to carbon dioxide injection or geothermal usage. A baseline characterization was performed as part of an EGR project at the planned CO2 storage facility in the Altmark area (Sachsen-Anhalt, Germany) (CLEAN-Project: “CO2 Largescale Enhanced Gas Recovery in the Altmark Natural-Gas Field”) [53, 54]. In this molecular-biological study, fluid samples were analyzed and the identified bacteria were affiliated to hydrogen-oxidizing bacteria, thiosulfate-oxidizing bacteria as well as undescribed species. Since the local mining authority did not permit the injection of carbon dioxide, no data regarding the impact of carbon dioxide in the reservoir could be collected. At the CCS pilot site in Ketzin (Brandenburg, Germany) Europe’s longest-operating on-shore carbon dioxide storage site [17, 31] a microbiological monitoring was performed starting with the drilling of the boreholes [23, 55, 56] until three years after carbon dioxide injection begun. The carbon dioxide is stored in a saline aquifer at depth of 630–710 m. The formation fluid was characterized by approximately 35°C, 6.4 MPa, and a salinity of about 235 g L−1 . Samples were taken from cores (sandstone of the Triassic Stuttgart-Formation) and formation fluids in the well. Most detected microbes were halotolerant or even halophilic. Fingerprinting analysis and Fluorescence in situ Hybridization (FISH) revealed that fermentative halophilic bacteria (Halanaerobium sp., Halobacteroidaceae) and Sulfate Reducing Bacteria (SRB) (Desulfohalobium sp., Desulfotomaculum sp.) were the dominant organisms in the reservoir prior to the carbon dioxide injection [57]. Moreover, archaea that presumably play a crucial role in carbon cycling, could be detected and quantified in the fluids by FISH analysis [23]. Another molecular-biological monitoring was performed on fluids from a geothermically-used aquifer in the Molasse Basin (Southern Germany) [19]. Here the fluid derives from Jurrasic (Malm) carbonate rocks with a very high permeability. SRB of different genera were detected in up to 103°C hot thermal fluids. By analyzing the 16S rRNA and dissimilatory sulfite reductase genes, various SRB and one archaeon were detected. Desulfotomaculum reducens-like bacteria were found in all samples even after passing the heat exchanger, whereby Desulfococcus oleovorans-like bacteria were solely found before the heat exchanger. Contrarily, Candidatus Desulforudis audaxviator-like bacteria and Methanothermobacter thermoautotrophicus were only found in the fluid samples taken after the heat exchanger at a temperature of 61°C. This might

10.2 Microbial diversity in geo-engineered reservoirs

| 207

indicate that the changing conditions during the passage through the plant – e.g. temperature decrease and reduced fluid flow – lead to favorable conditions for growth. Since the fluid flow is high, growth might only be possible in plant components like the heat exchanger, where the flow is partially reduced and biofilms can attach to surfaces. The increased diversity of SRB after heat exchange can be regarded as an evidence for the relevance of microbial activity for the re-injection of fluids and the subsequent cooling of the near-well area over long-term periods [19]. There are indications that Desulfotomaculum and Methanobacterium are ubiquitously distributed, as both genera were also identified as dominant microorganisms in a 4- to 5-kilometerDeep Fault [58]. Stevens et al. [59] characterized the lithoautotrophic microorganisms in deep basalt aquifers from the Columbia River Basalt Group with the result that autotrophic microorganisms outnumbered heterotrophs. Stable carbon isotope measurements implied that autotrophic methanogenesis was coupled to the depletion of dissolved inorganic carbon. A screening of the natural microbial population was also performed in ATES. Lerm et al. [24] investigated an aquifer located in Neubrandenburg (North German Basin, Germany) that is used for seasonal heat storage. The sandstone storage formation is located at a depth of 1228–1268 m. In ATES, distinct shifts in the microbial population can be observed depending on the operational mode, and fluid temperature (46–74°C). Since fluid temperatures in such systems are much lower than in deep geothermal systems, which are usually well above 100°C, a high diversity of bacteria was detected. By microscopically analyzing the fluid samples, and by realtime PCR, higher cell counts as well as higher 16S rRNA and dsrA gene copy numbers were found in fluids from the cold well. Additionally a predominance and higher diversity of SRB was found in the cold well. The last two studies discussed indicate that thermal waters with temperatures below 80°C favor the growth and diversity of SRB, especially, which potentially enhance or induce microbial corrosion [19, 24]. Lerm et al. [24] recently highlighted that the sulfate reducing Candidatus Desulforudis audaxviator, which uses hydrogen originating from radiochemical reactions in deep crustal rocks [60], was found in rather different mine waters [58, 61, 62]. The detection of sequences affiliated with this species in a saline aquifer in the North German Basin [24] as well as in the low mineralized Malm aquifer of the Molasse Basin in southern Germany [19] shows the high adaption potential of these bacteria with respect to the salinity. By environmental genomics it was shown that Candidatus D. audaxviator is even the only player in a so-called single-species ecosystem within a 2.8-kilometer depth South African gold mine [63]. Since temperatures in the deep biosphere are mostly constant, changes of the temperature are initializing relatively fast changes in the microbial abundance and diversity.

208 | 10 Microbes in geo-engineered systems

10.3 Interactions between microbes and geo-engineered systems 10.3.1 General considerations In the deep biosphere, the mobility of substances and thereby substrates, is an important issue, but the microbes can even survive long periods of starvation by extremely reducing their metabolism or spore formation [64, 65]. Over the last years, geomicrobiological analyses of the deep biosphere brought up many questions about the varying physical state of the cells. However, the most fascinating finding might be that there is suddenly no clear consensus on the definition of microbial cell death and it remains unclear how to accurately define the point of no return back to normal metabolism. The time scales of the deep biosphere are too long to adequately answer this question. The mechanisms initializing resuscitation and the involved physiological processes are poorly understood and mostly performed for pathogenic species under laboratory conditions [66] or in other environments like deep permafrost [67]. The tool box applied to the deep biosphere is lacking methods to analyze the physiology and genetics of the ultraslow metabolism of dormant cells or viable but nonculturable cells (VBNC). Valuable outcomes are provided by gene-expression studies of cells while the VBNC state, upon entering into and eventually recovering from this state when provided with nutrients and changing environmental conditions [68]. This scenario is comparable to the dormant cells in many oligotrophic deep biosphere environments like the reservoirs or rock formations used for CO2 storage or geothermal energy. By introducing organic substances (e.g. drill mud), or changing the gas phase (elevated CO2 concentrations) and geochemistry of the aquifer (dislocation of substances, temperature, pH, precipitation), dormant cells might be reactivated and become active again. Additionally, the aspect of spores as a physiological state of the cell should be addressed. The germination of spores under changing conditions might be especially interesting with regards to SRB and therefore microbiologically induced corrosion (MIC) and precipitation. Spores originating from an aquifer used for geothermal energy are resistant to high temperatures (> 130°C) [69] and could therefore be seen as the seeds for a proliferating SRB community in the surface plant installation and subsequently might be involved in MIC of the plant components. To understand the impact of an elevated CO2 concentration on the microbial community in the reservoir, it is of paramount importance to determine the factors that limit growth prior to geotechnical usage. Survival strategies of microorganisms in oligotrophic deep-biosphere ecosystems are manifold [70, 71]. However, minimum factors for microbial lithoautotrophic growth cannot be generally defined, the environmental conditions in the different geo-engineered aquifers are too diverse. Eecke et al. [72] described a hydrogen-limited growth of hyperthermophilic methanogens in some of the analyzed deep-sea hydrothermal vent sites. Furthermore, the authors con-

10.3 Interactions between microbes and geo-engineered systems |

209

clude that H2 limitation may be partly ameliorated by H2 syntrophy with hyperthermophilic heterotrophs. Besides syntrophic relations, the competition of different microbial groups, e.g. methanogens and SRB concurring for H2 , may play an important role during changing conditions [73]. However, these analyses give insights into the complex syntrophic interactions that make predictions concerning the influence of an elevated CO2 concentration on the microbial community difficult.

10.3.2 Microbial processes in the deep biosphere potentially affected by CCS The effect of elevated carbon dioxide levels on deep subsurface microbiology remains, in many aspects, poorly understood. Besides potential changes in geochemistry, microbial community structure and activity due to CO2 -injection, the impact of microbes on geo-engineered systems and vice versa will also be discussed in Section 10.3.3– 10.3.5. Our knowledge of the various interactions between the deep biosphere and CCS is poor and important questions remain unanswered: – Considering that CO2 is a carbon source and electron acceptor for microbial metabolism in the deep biosphere, how can elevated concentrations of carbon dioxide change the microbial population and its activity? – In what quantities and under what conditions are CH4 or H2 S microbially produced inside the reservoir and to what extend might this cause problems? – How strongly can microbes enhance the dissolution of CO2 or the precipitation of new carbonate solid phases? At first, it is important to understand the geochemical changes that take place in the reservoir rock and its fluids prior to the injection of CO2 . The reservoir pressure will increase during storage, the previously stable temperature regime will be perturbed, and the concentration of dissolved compounds and the pH will also be changed. The important aspect of fluid-rock interaction during CO2 sequestration was analyzed in several studies [74–77]. Geochemical changes in the reservoir might already take place before CO2 -injection. Depending on the composition of the applied drill mud and whether or not fracking was used, organic compounds or other substances might mix with the formation fluids. Moreover, CO2 can liberate organic compounds [25, 26] and enhance the dissolution, transformation and precipitation of minerals [78]. Many microbiological processes are directly linked to the geochemistry of the reservoir and thereby also react on geochemical changes that are caused by the storage of CO2 . Also, the availability of liquid water and its salinity are an important aspect heavily discriminating the geochemical response of relatively dry depleted gas reservoirs or water-saturated saline aquifers. 󳶳 Figure 10.1 shows a scheme of a CO2 storage facility with separator unit. The scheme highlights microbial pathways that might be directly affected by elevated CO2 concentrations. The injected CO2 fills the pore space, partially dissolves in the porewa-

210 | 10 Microbes in geo-engineered systems

Fig. 10.1: CO2 storage under microbiological aspects. Modified after Pedersen [79] and Mitchell et al. [80]

ter and can be microbially reduced, mainly by methanogens and acetogenic bacteria and via incomplete fermentation by acetoclastic methanogens. Acetate is produced by acetogenic bacteria and becomes a valuable substrate for FeIII , MnIII, IV and SO2− 4 reducing microorganisms as well as acetoclastic methanogens. It is also discussed that acetate plays a role in the anaerobic oxidation of methane [81]. Elevated concentrations of CO2 might thereby also affect the competition or syntrophic relation of microorganisms, as was shown for anaerobic oxidation of methane and sulfate reduction [82]. Enhanced primary production from lithoautotrophic microbial growth might provide the starting point for a secondary heterotrophic microbial community. Therefore, a more complex reaction of deep biosphere microorganisms on CO2 storage might become plausible. Energy-rich organic polymers, produced by the involved microorganisms, are subsequently degraded by fermentative processes and thereby close the carbon cycle by the emission of CO2 as well as H2 . Additionally, CO2 can be fixed to organic polymers under autotrophic conditions. The involved processes like the reductive citric acid cycle (anaerobic and microaerophilic bacteria) and reductive acetyl CoA pathway (anaerobic bacteria and archaea) are well described [83].

10.3 Interactions between microbes and geo-engineered systems |

211

An important issue is the mobility of certain substances in the deep biosphere that might be affected by CO2 , changes in reservoir pressure or formation of precipitates e.g. carbonates, etc. Sauer et al. [26] constructed an incubation system for high gas partial pressure and incubated organic-rich samples like coals and lignites under simulated CO2 storage conditions. The laboratory experiments showed the different extend of the liberation of various low molecular weight acids (e.g. formate, acetate, oxalate) from different reservoir materials. Also, co-injected gases like SO2 and NO𝑥 have a significant effect on dissolution kinetics as well as total yield of various shortchain organic acids. This high pressure system is not only useful for the incubation of microorganisms, but also if analyzing the chemical degradation or dislocation of substances e.g. low molecular weight acids under high gas saturation and further fluidgas-rock interactions in relation to carbon dioxide sequestration.

10.3.3 Examples from a CCS pilot site, CO2 degasing sites and laboratory experiments At the CCS pilot site in Ketzin (Germany), fingerprinting analyses and FISH data indicated changes in the abundance and community structure of microorganisms in well fluids. Zettlitzer et al. [84] described a temporary reduction of the injectivity of one well that was most likely caused by microbial activity. It was assumed that organic compounds of the drill mud fed the microbial community, which in consequence was enhancing microbial processes like sulfate reduction. The increased microbial activity was boosting mineral precipitation e.g. amorphous iron sulfides and lowering the permeability of the near-well area and thereby decreasing the productivity of the well to a critical value. After a nitrogen lift was conducted to remove iron sulfide and remaining organics (TOC and acetate) the injectivity of the well was re-established and SRB were not detectable anymore [23, 84]. The highest abundance of archaea was found in the fluid samples taken after the CO2 arrival and made up 11% to 16% of the active cells. However, the population stabilized and after 10 months of CO2 -injection no significant changes in total cell counts or population structure were found. The described scenario is an example of the interaction of geotechnical procedures and the surrounding microbial community. Additionally, laboratory long-term experiments under simulated reservoir P-T conditions (5.5 MPa and 40°C) were carried out [85, 86]. Core sections, obtained from the reservoir region at a depth of about 650 m below surface were incubated in high-pressure vessels together with sterile synthetic formation brine. However, these experiments could not show a significant change in the microbial population either. It can be concluded that permanent changes in the population structure of deep subsurface microbial communities might only be detectable after a longer period. A few studies from CCS sites [23, 86] and several from natural CO2 degasing analogues [54, 86–91] showed that microbes in natural environments can cope with highly

212 | 10 Microbes in geo-engineered systems elevated CO2 concentrations. Krüger et al. [88, 92] carefully described effects of elevated CO2 concentrations on the vegetation and microbial populations at a terrestrial CO2 vent at Laacher See (central-west Germany). A comparison of the soil community from a control site and the center of the venting area, showed a change in the soil pH below 10 cm depth. The microbial communities were significantly different at the CO2 -rich sites (CO2 comprising up to 90% and more of soil gas), medium CO2 -rich sites (∼20%), and control locations with background CO2 concentrations. A shift towards anaerobic and acidotolerant to acidophilic species under elevated CO2 concentrations was observed. While this soil habitat is not directly comparable to the deep biosphere, the study can provide insights into how microbial populations shift due to such environmental changes. Potential ecological impacts of CO2 leakage to shallow groundwater and soil/sediments from CCS sites were investigated with regard to the viability and metal reduction capabilities of Shewanella oneidensis MR-1 under CO2 stress [93]. The results clearly show that a sudden increase of the dissolved CO2 can cause stress and cell death. In a comprehensive study on the sulfate reducing bacterium Desulfotomaculum geothermicum and the methanogenic archaeon Methanothermococcus thermolithotrophicus, the differences in their tolerance to CO2 were analyzed [94]. Abiotic dissolution processes were observed but some biomineralization processes of carbonates were also identified for D. geothermicum. The results show how different microbes react to elevated CO2 concentrations. Another interesting outcome was that both strains displayed very different patterns in their conversion of inorganic carbon. M. thermolithotrophicus was mainly producing methane, whereas D. geothermicum induced the formation of biomass. Also, incubation experiments under simulated reservoir conditions (55°C, 5 MPa) with high-temperature oil reservoir samples revealed the strong impact of an increased partial pressure of CO2 on microbial metabolism [95]. After CO2 incubation, the methane production rate was more than doubled compared to the controls with low CO2 conditions. It is assumed that the changed environmental conditions invoke acetoclastic methanogenesis in place of syntrophic acetate oxidation coupled with hydrogenotrophic methanogenesis, which usually dominates in this system [95]. The potential, to which a microbial community can shift and adapt in composition and size, is enormous. It can be assumed that the size of the microbial population’s genetic pool may significantly depend on cell adaptations like resuscitation from dormancy and the ability to form spores. An extremely lowered cell metabolism at times of starvation and unfavorable environmental conditions may be one of the secrets enabling such remarkable population shifts driven by environmental changes. The variation in energy available to microbial populations under CCS conditions was analyzed on two rather different sites [96, 97]. The results indicate that the energy available for Fe(III) reduction increased significantly during CO2 -injection on both sites. Energy available to sulfate reducing microorganisms and methanogens varied little.

10.3 Interactions between microbes and geo-engineered systems

| 213

10.3.4 Impact of microbially-driven processes on CO2 trapping mechanisms One long-term goal of CCS projects is the mineral trapping of CO2 in the reservoir rock. There are three main advantages: a larger volume of CO2 can be stored, the solidified CO2 (in the form of carbonates) has no climate impact and the risk of CO2 leakage from the reservoir can be reduced. The mineralization of injected CO2 into CaCO3 becomes likely if the equilibrium of the reaction Ca2+ + CO2− 3 ↔ CaCO3 is changed by the amount of calcium, increase of the pH and alkalinity and also of an appropriate nucleation substrate (also microbes) [98]. The prospection for storage reservoirs for CO2 , ideally characterized by a high CO2 trapping potential, revealed that the storage in magnesium silicate minerals has many benefits [99]. The naturally-occurring carbonation of magnesium silicate minerals, e.g. olivine and serpentine, enables the fixation of carbon dioxide by trapping it to the geologically-stable mineral magnesite [99, 100]. Mitchell et al. [80] conducted microcosm experiments where urea was used to microbially enhance CO2 trapping processes. The procedure makes use of bacterial − ureolysis, which can cause an increase in pH as well as CO2− 3 , and HCO3 concentrations [101–103]. This technique was first applied in the frame of enhanced gas recovery to reduce the bedrock porosity [104] and for the co-precipitation of radionuclides in groundwater [101, 105]. Mitchell et al. [80, 90] recognized that the injection of hydrolized urea favored the formation of microbial biofilms. These biofilms were resilient to supercritical (sc) CO2 and reduced the permeability of rock cores at high pressure. In further experiments using a pressurized flow reactor, Mitchell et al. [106, 107] could show that permeability in cores can be reduced by 95–99% just by the growth of microbial biofilms. The stability of the biofilm provides shelter for the microbes and seems to be resistant to the mechanical shear stress during the CO2 -injection. During such biomineralization processes, active transport of calcium appears to play a key role in the biochemistry, in addition to pH control [108, 109]. A numerical model was developed to calculate the extent of carbonate precipitation due to the described ureolytic bacterial activity [110]. Schultze-Lam et al. [111] showed that microbes enhance the CaCO3 precipitation via cation adsorption to negatively-charged functional groups on the cell wall by metabolically-driven changes in the solution chemistry and thereby increase mineral saturation and induce nucleation. Sulfate, iron, or nitrate reduction are further microbial processes that were shown to favor the precipitation of CaCO3 [112, 113]. Therefore, engineered biomineralizing biofilms, mineral trapping and solubility-trapping may enhance the performance of CCS facilities (󳶳 Fig. 10.1).

214 | 10 Microbes in geo-engineered systems 10.3.5 Impact of microbially-driven processes on CCS facilities Most of the negative impacts of microbes on geotechnical equipment are known from oil production [69, 114] and geothermal plants [11]. The economically most important problem is MIC of plant components caused by SRB and microbially-induced or enhanced mineral precipitation [8, 10, 11]. An increase in sulfate reduction rate may cause an acidification and enrichment of toxic hydrogen sulfide, which likely precipitates as metal sulfide. Precipitation of amorphous iron sulfides can cause a lowered pore space, thereby decrease the permeability of the reservoir rock and decrease the injectivity of carbon dioxide or geothermal fluid in the borehole-near area or filter area (󳶳 Fig. 10.1). A very particular process of biocorrosion is described for Geobacter sulfurreducens. It was shown that the metabolism of G. sulfurreducens is directly connected to solid electrodes (metal surfaces), exchanging electrons through membranebound redox compounds [115]. An elevated CO2 concentration is provoking biochemical reactions in the reservoir that may negatively affect the reservoir properties, including the long term stability of CO2 sequestration [108, 116].

10.3.6 Impact of microbially-driven processes on geothermal energy plants The geothermal usage of an aquifer can, in some cases, over longer time scales (years to decades), decrease the reservoir temperature by several degrees Celsius. Contrarily, the temperature will increase immediately around the production and injection well and might thereby increase the speed of corrosion caused by physicochemical processes, but also by SRB. Permeability of the reservoir rock can be decreased due to the precipitation of carbonates. This aspect of the geo-engineered system is crucial since the productivity/injectivity of the well determines the productivity of the plant. A significantlylowered permeability of the reservoir rock can be eliminated by acid treatment (or other fracking procedures) and hydraulic flushing, which in the best case leads to a downtime of just several days. There are plants in Europe where additional injection wells had to be drilled to dispose the produced fluid volume. Such additional wells significantly reduce the economic feasibility of a plant. For better process understanding of a geothermal plant, detailed knowledge of the changes in chemistry and microbial community composition of the process fluids is of high importance [19, 117–121]. A detailed geochemical characterization is a prerequisite to understanding the microbial processes and to get an overview of available electron acceptors and donors. An analysis of the gas composition and isotopic signature of gases can indicate which microbial processes may be found in the deep reservoir. A long-term monitoring of these parameters might deliver early indications for process anomalies and changes in the reservoir. Vetter et al. [122] were able to detect

10.3 Interactions between microbes and geo-engineered systems

| 215

changes in the microbial population by a sophisticated biogeochemical monitoring including analyses of the phospholipid fatty acids. Differences in DOC concentration and composition between start-up phase (after drilling) of the wells, the initial and the regular plant operation, were detected. These changes in fluid chemistry and thereby microbial substrate supply might have a much bigger impact on the microbial population than a slight temperature increase or decrease of the reservoir or surrounding sediment. The effects of a changed microbial community or enhanced activity might cause problems in the subsurface or in the above-ground installations of the facility. However, pure geochemical analyses are often not sensitive enough to detect slight changes in organic fractions of the geothermal fluids [24]. The reservoir might host the thermotolerant or thermophilic microorganisms that are involved in corrosion and scaling of plant components, e.g. SRB. The knowledge about their physiology might help to develop counter measures. Increasing cell numbers of fermentative bacteria and SRB in process fluids might be used as potential indicator organisms for early stage process failures due to corrosion or mineral scaling [24]. The impact of MIC on plant components is well documented from other industries [10]. However, the geothermal plants, for example, differ in various aspects from other industrial facilities. The volume of the often saline fluid that passes the geothermal installation is enormous. For example, several deep geothermal plants (2700–6700 m drilled depth) in Bavaria (Southern Germany) have a fluid flow higher than 100 L/s or 8.6 million liters per day (data from [123]). The high loads of salt and the high temperatures of the fluids are enhancing abiotic corrosion processes. Scales in hydro-geothermal systems consist mostly of carbonates, silica, and sulfur minerals [124]. Precipitation is caused by oversaturation or redox reactions in the brine. The intensity of scaling depends on pressure changes, temperature changes, oxygen ingression, and/or corrosion [125]. Nevertheless, MIC is an obvious process under many conditions [10, 24] and SRB found in geothermal plants are well adapted to the environmental conditions [19, 126]. There are different opportunities to prevent or reduce the impact of microbiologically-driven processes like corrosion and enhanced mineral precipitation on the plant structure. After drilling, the subsequent cleaning of the well from residual drill mud by flushing or nitrogen lift might reduce the risk of enhanced microbial activity [23]. The installation of highly corrosion-resistant materials (e.g. alloyed steel or carbon fiber) might also enhance the plant reliability but drastically increase the costs [12]. The temperature regime in many geothermal plants could be regulated up to a level > 90°C where most of the microbes are not active. However, a decrease of the difference in fluid temperature between production well and injection well is directly linked to the produced energy and thereby the efficiency of the geothermal plant.

216 | 10 Microbes in geo-engineered systems

10.4 Methods to analyze the interaction between geo-engineered systems and the deep biosphere The goal of microbiological monitoring is to detect changes in community structure and activity in geo-engineered systems. A baseline study that carefully analyzes the indigenous microbial community and the main microbe-driven geochemical processes in the reservoir, e.g. before carbon dioxide injection or geothermal use, is a prerequisite to understanding the potential changes during the operation [23]. The gained information of microbiological monitoring will also help to identify highly responsive indicator organisms [127]. Moreover, stealthy CO2 leakages or other changes in the reservoir chemistry might be recognized by biological monitoring [128].

10.4.1 Sampling of reservoir fluids and rock cores Rock or fluid samples can be taken during drilling, pumping tests, sampling of the bore sump and deep hole sampling with fluid samplers. Suitable tools for deep borehole sampling are time controlled pressure displacement samplers (PDS) or high-pressure samplers that retrieve the sample under in situ pressure. To retrieve rock samples after drilling, sidewall coring might be an option, albeit an expensive one. An interesting sampling procedure is the so-called U-tube from which – only by reservoir pressure – a small fluid portion can be sampled continuously [129, 130]. Fluid sampling in geothermal plants is comparatively easy, samples can be taken at different sampling points in the facility, e.g. before and after the heat-exchanger. Some geothermal plants have installed bypass systems, not only to monitor fluid chemistry and microbiology, but also to grant the opportunity to change process parameters e.g. temperature, flow velocity and addition of substances, and to see if these changes affect precipitation or MIC. Additionally, residues from filters that have to be installed to trap sediment particles can be sampled to analyze the particle-associated microbial community. A high amount of particulate minerals e.g. silica, iron sulfides or carbonates can clog filter units and thereby shorten the changing intervals and decrease productivity. In some cases, the plant filters or fluids retain evidence for long-term microbial processes in the reservoir e.g. pyrite framboids formed by microbial processes [19, 131].

10.4.2 Methods to analyze microbes in geo-engineered systems Until now, microbiological monitoring concepts in CCS were focusing on potential changes in the community structure during and after the injection [19, 23, 54] and the addition of substrates e.g. urea to enhance biomineralization [80, 107, 132].

References | 217

To gain knowledge of microbially-driven processes in the deep biosphere, classical cultivation-based methods [26, 56, 86, 133, 134] as well as stable isotope- and molecular-biological methods [19, 23, 54, 56, 84, 122] has to be applied in order to obtain a comprehensive overview of the microbial ecosystem and its activity. Only approaches linking geochemical and biological processes can unravel the complex biogeo interactions of the deep biosphere. At the CO2 storage facility in In Salah (Algeria), Jones et al. [133] performed direct cell counting of bacteria on the upper 50 cm of the soil around the boreholes to register changes in the microbial population due to potential leakages of CO2 . Additionally, the microbial activity was monitored by ATP measurements. This baseline study should be useful for a long-term monitoring of the area surrounding the wellhead. Since many microbiological processes are too complex and/or too slow to be analyzed in situ, laboratory experiments using autoclaves (pressure vessels) have to be used to simulate in situ conditions [23, 85, 86, 134]. Unfortunately, the operative and financial effort to perform such experiments is significant, and only a few laboratory studies were conducted under simulated in situ conditions. Kallmeyer et al. [134] introduced a high-pressure thermal gradient block for cultivation in which multiple samples can be simultaneously incubated. Such technical setups are often a prerequisite for analyzing the microbial activity under simulated in situ conditions [26, 135]. An important advantage of laboratory experiments is that complex biological processes can be subdivided into smaller and simpler partial reactions, e.g. by adding intermediates instead of just basic reactants. However, the division of possible bio-geochemical interactions into separate laboratory experiments might not reflect the real in situ conditions of the deep biosphere.

Acknowledgements I would like to thank Marco De Lucia, Bernd Wiese and Simona Regenspurg (GFZ German Research Centre for Geosciences) for their support.

References [1]

[2]

[3]

Working Group III of the Intergovernmental Panel on Climate Change [Metz B, Davidson HCdCO, Loos M, Meyer LA (eds.)], IPCC, 2005: IPCC Special Report on Carbon Dioxide Capture and Storage. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2005. Davidsdottir B, 7.10 – Sustainable Energy Development: The Role of Geothermal Power. In: Comprehensive Renewable Energy, S. Editor-in-Chief: Ali, Editor. 2012, Elsevier: Oxford. p. 273–297. Renewables 2013, Global Status Report. Renewable Energy Policy Network for the 21st Century.

218 | 10 Microbes in geo-engineered systems [4] [5]

[6] [7] [8] [9] [10] [11] [12]

[13]

[14]

[15] [16] [17]

[18] [19] [20]

[21] [22] [23]

[24]

Nielsen CM, Frykman P, Dalhoff F. Synergy Benefits in Combining CCS and Geothermal Energy Production. Energy Procedia 37 (2013), 2622–2628. Freifeld B, et al. Geothermal Energy Production Coupled with CCS: a Field Demonstration at the SECARB Cranfield Site, Cranfield, Mississippi, USA. Energy Procedia 37 (2013), 6595– 6603. Tillner E, et al. Geological CO2 Storage Supports Geothermal Energy Exploitation: 3D Numerical Models Emphasize Feasibility of Synergetic Use. Energy Procedia 37 (2013), 6604–6616. Honegger JL, et al. Detailed study of sulfide scaling at la courneuue nord, a geothermal exploitation of the Paris Basin, France. Geothermics 18 No. 1-2 (1989), 137–144. Beech IB, Sunner J. Biocorrosion: towards understanding interactions between biofilms and metals. Curr Opin Biotechnol 15 No. 3 (2004), 181–186. Coetser SE, Cloete ET. Biofouling and biocorrosion in industrial water systems. Crit Rev Microbiol 31 No. 4 (2005), 213–232. Little BJ, Lee JS, Microbiologically Influenced Corrosion. Wiley, Hoboken, 2007. Valdez B, et al. Corrosion and scaling at Cerro Prieto geothermal field. Anti-Corrosion Methods and Materials 2009;56(1):28–34. Karlsdóttir SN, 7.08 – Corrosion, Scaling and Material Selection in Geothermal Power Production. In: Comprehensive Renewable Energy, S. Editor-in-Chief: Ali, Editor. 2012, Elsevier: Oxford. p. 241–259. Benson SM, Cook P. Chapter 5: Underground geological storage. In: IPCC Special Report on Carbon Dioxide Capture and Storage. Intergovernmental Panel on Climate Change, Interlachen, Switzerland, 5–1 to 5–134, 2005. Ross AR, Pejcic B, Stalker S. Down-hole monitoring of chemical changes associated with CO2 storage – A review of chemical sensor technology. In: Cooperative Research Centre for Greenhouse Gas Technologies, Canberra, Australia. CO2 CRC Publication Number RPT07– 0749, 48, 2007. Stalker L, et al. Feasibility of Monitoring Techniques for Substances Mobilized by CO2 Storage in Geological Formations. Energy Procedia 23 (2012), 439–448. Bozau E, van Berk W. Hydrogeochemical Modeling of Deep Formation Water Applied to Geothermal Energy Production. Procedia Earth and Planetary Science 7 (2013), 97–100. Martens S, et al. Europe’s longest-operating on-shore CO2 storage site at Ketzin, Germany: a progress report after three years of injection. Environmental Earth Sciences 67 No. 2 (2012), 323–334. Oelkers EH. Geochemical aspects of CO2 sequestration. Chemical Geology 217 (3–4 SPEC. ISS.) (2005), 183–186. Alawi M, et al. Diversity of sulfate-reducing bacteria in a plant using deep geothermal energy. Grundwasser 16 No. 2 (2011), 105–112. Basso O, Caumette P, Magot M. Desulfovibrio putealis sp. nov., a novel sulfate-reducing bacterium isolated from a deep subsurface aquifer. Int J Syst Evol Microbiol 55 Pt. 1 (2005), 101– 104. Sand W. Microbial life in geothermal waters. Geothermics 32 (2003), 655–667. Colwell FS, D’Hondt S. Nature and Extent of the Deep Biosphere. Reviews in Mineralogy and Geochemistry 75 No. 1 (2013), 547–574. Morozova D, et al. Monitoring of the microbial community composition in saline aquifers during CO2 storage by fluorescence in situ hybridization. International Journal of Greenhouse Gas Control 4 No. 6 (2010), 981–989. Lerm S, et al. Thermal effects on microbial composition and microbiologically induced corrosion and mineral precipitation affecting operation of a geothermal plant in a deep saline aquifer. Extremophiles 17 No. 2 (2013), 311–327.

References | 219

[25]

[26] [27] [28] [29] [30] [31] [32]

[33] [34]

[35] [36]

[37] [38] [39]

[40] [41] [42]

[43] [44] [45]

Glombitza C, Mangelsdorf K, Horsfield B. A novel procedure to detect low molecular weight compounds released by alkaline ester cleavage from low maturity coals to assess its feedstock potential for deep microbial life. Organic Geochemistry 40 No. 2 (2009), 175–183. Sauer P, Glombitza C, Kallmeyer J. A system for incubations at high gas partial pressure. Frontiers in Microbiology 3 (2012). Luquot L, Gouze P. Experimental determination of porosity and permeability changes induced by injection of CO2 into carbonate rocks. Chemical Geology 265 No. 1–2 (2009), 148–159. http://www.sccs.org.uk/map Accessed October 2013. Benson SM, Cole DR. CO2 sequestration in deep sedimentary formations. Elements 4 No. 5 (2008), 325–331. Bachu S. Screening and ranking of sedimentary basins for sequestration of CO2 in geological media in response to climate change. Environmental Geology 44, No. 3 (2003), 277–289. Schilling F, et al. Status Report on the First European on-shore CO2 Storage Site at Ketzin (Germany). Energy Procedia 1 No. 1 (2009), 2029–2035. Förster A, et al. Reservoir characterization of a CO2 storage aquifer: The Upper Triassic Stuttgart Formation in the Northeast German Basin. Marine and Petroleum Geology 27 No. 10 (2010), 2156–2172. Hitchon B. Best Practices for Validating CO2 Geological Storage: Observations and guidance from the IEAGHG Weyburn-Midale CO2 storage project. Geoscience Publishing, 2012. Gislason SR, et al. Mineral sequestration of carbon dioxide in basalt: A pre-injection overview of the CarbFix project. International Journal of Greenhouse Gas Control 4 No. 3 (2010), 537– 545. Alfredsson HA, et al. The geology and water chemistry of the Hellisheidi, SW-Iceland carbon storage site. International Journal of Greenhouse Gas Control, 2013. 12:399–418. Goldberg D. CO2 Sequestration Beneath the Seafloor. Evaluating the In Situ Properties of Natural Hydrate-Bearing Sediments and Oceanic Basalt Crust. International Journal of the Society of Materials Engineering for Resources 7 No. 1 (1999), 11–16. Goldberg D, Slagle AL. A global assessment of deep-sea basalt sites for carbon sequestration. Energy Procedia 1 No. 1 (2009), 3675–3682. Goldberg DS, Takahashi T, Slagle AL. Carbon dioxide sequestration in deep-sea basalt. Proc Natl Acad Sci USA 105 No. 29 (2008), 9920–9925. Wilke FDH, et al. On the interaction of pure and impure supercritical CO2 with rock forming minerals in saline aquifers: An experimental geochemical approach. Applied Geochemistry 27 No. 8 (2012), 1615–1622. Renard S, et al. Geochemical study of the reactivity of a carbonate rock in a geological storage of CO2 : Implications of co-injected gases. Energy Procedia 4 (2011), 5364–5369. Bachu S, Gunter WD, Perkins EH, Aquifer disposal of CO2 : Hydrodynamic and mineral trapping. Energy Conversion and Management 35 No. 4 (1994), 269–279. Klein E, et al. Evaluation of long-term mineral trapping at the Ketzin pilot site for CO2 storage: An integrative approach using geochemical modeling and reservoir simulation. International Journal of Greenhouse Gas Control 19 (2013), 720–730. Kempka T, et al. Assessment of Long-term CO2 Trapping Mechanisms at the Ketzin Pilot Site (Germany) by Coupled Numerical Modeling. Energy Procedia 37 (2013), 5419–5426. Matek B. Geothermal Power: International Market Overview. GEA, Geothermal Energy Association, 2013. Schmidt T, Mangold D, Müller-Steinhagen H. Central solar heating plants with seasonal storage in Germany. Solar Energy 76 No. 1–3 (2004), 165–174.

220 | 10 Microbes in geo-engineered systems [46]

[47] [48] [49]

[50] [51]

[52]

[53] [54]

[55] [56]

[57]

[58] [59] [60] [61] [62] [63] [64] [65] [66]

Legarth B, Huenges E, Zimmermann G. Hydraulic fracturing in a sedimentary geothermal reservoir: Results and implications. International Journal of Rock Mechanics and Mining Sciences 42 No. 7–8 (2005), 1028–1041. Entingh DJ, Geothermal well stimulation experiments in the United States. Proceedings of the world geothermal congress 2000, Kyushu–Tohoku, Japan; 2000:3689–3694. Bargagli R, et al. Environmental Impact of Trace Element Emissions from Geothermal Power Plants. Archives of Environmental Contamination and Toxicology 33 No. 2 (1997), 172–181. Duchane D, Brown D. Hot Dry Rock (HDR) Geothermal Energy Research and Development at Fenton Hill, New Mexico. Geo-Heat Centre Quarterly Bulletin (Klamath Falls, Oregon: Oregon Institute of Technology) 2002;23(13–19). Barbier E. Geothermal energy technology and current status: an overview. Renewable and Sustainable Energy Reviews 6 No. 1–2 (2002), 3–65. Zimmermann G, et al. Rock specific hydraulic fracturing and matrix acidizing to enhance a geothermal system – Concepts and field results. Tectonophysics 503, No. 1–2 (2011), 146– 154. Grigsby CO, et al. Rock-water interactions in the Fenton Hill, new Mexico, hot dry rock geothermal systems I. fluid mixing and chemical geothermometry. Geothermics 18 No. 5–6 (1989), 629–656. Kühn M, et al. The CLEAN project in the context of CO2 storage and enhanced gas recovery. Environmental Earth Sciences 67 No. 2 (2012), 307–310. Morozova D, et al. The influence of microbial activity on rock fluid interaction: Baseline characterization of deep biosphere for Enhanced Gas Recovery in the Altmark natural gas reservoir. Energy Procedia 4 (2011), 4633–4640. Giese R, et al. Monitoring at the CO2 SINK site: A concept integrating geophysics, geochemistry and microbiology. Energy Procedia 1 No. 1 (2009), 2251–2259. Wandrey M, et al. Microbial community and inorganic fluid analysis during CO2 storage within the frame of CO2 SINK–Long-term experiments under in situ conditions. Energy Procedia 4 (2011), 3651–3657. Wandrey M, et al. Monitoring petrophysical, mineralogical, geochemical and microbiological effects of CO2 exposure — Results of long-term experiments under in situ conditions. Energy Procedia 4 (2011), 3644–3650. Moser DP, et al. Desulfotomaculum and Methanobacterium spp. Dominate a 4- to 5-KilometerDeep Fault. Applied and Environmental Microbiology 71 No. 12 (2005), 8773–8783. Stevens TO, McKinley JP, Lithoautotrophic Microbial Ecosystems in Deep Basalt Aquifers. Science 270 No. 5235 (1995), 450–455. Lin L-H, Long term biosustainability in a high energy, low diversity crustal biotome. Lawrence Berkeley National Laboratory. LBNL Paper LBNL-60468 (2010). Lin L-H, et al. Long-Term Sustainability of a High-Energy, Low-Diversity Crustal Biome. Science 314 No. 5798 (2006), 479–482. Gihring TM, et al. The Distribution of Microbial Taxa in the Subsurface Water of the Kalahari Shield, South Africa. Geomicrobiol J 23 No. 6 (2006), 415–430. Chivian D, et al. Environmental Genomics Reveals a Single-Species Ecosystem Deep Within Earth. Science 322 No. 5899 (2008), 275–278. Hubert C, et al. A Constant Flux of Diverse Thermophilic Bacteria into the Cold Arctic Seabed. Science 325 No. 5947 (2009), 1541–1544. Goldscheider N, Hunkeler D, Rossi P. Review: Microbial biocenoses in pristine aquifers and an assessment of investigative methods. Hydrogeology Journal 14 No. 6 (2006), 926–941. Oliver JD, The viable but nonculturable state in bacteria. J Microbiol, 2005. 43:93–100.

References | 221

[67] [68] [69]

[70] [71] [72] [73] [74] [75] [76]

[77]

[78]

[79] [80] [81] [82] [83] [84] [85]

[86]

[87]

Vorobyova E, et al. The deep cold biosphere: facts and hypothesis. FEMS Microbiology Reviews 20 No. 3–4 (1997), 277–290. Trevors JT, Viable but nonculturable (VBNC) bacteria: Gene expression in planktonic and biofilm cells. Journal of Microbiological Methods 86 No. 2 (2011), 266–273. Rosnes JT, Torsvik T, Lien T. Spore-Forming Thermophilic Sulfate-Reducing Bacteria Isolated from North Sea Oil Field Waters. Applied and Environmental Microbiology 57 No. 8 (1991), 2302–2307. Hoehler TM, Jørgensen BB. Microbial life under extreme energy limitation. Nature Reviews Microbiology 11 No. 2 (2013), 83–94. Røy H, et al. Aerobic Microbial Respiration in 86-Million-Year-Old Deep-Sea Red Clay. Science 336 No. 6083 (2012), 922–925. Ver Eecke HC, et al. Hydrogen-limited growth of hyperthermophilic methanogens at deep-sea hydrothermal vents. Proc Natl Acad Sci USA 109 No. 34 (2012), 13 674–13 679. Robinson J, Tiedje J. Competition between sulfate-reducing and methanogenic bacteria for H2 under resting and growing conditions. Archives of Microbiology 137 No. 1 (1984), 26–32. Gaus I. Role and impact of CO2 –rock interactions during CO2 storage in sedimentary rocks. International Journal of Greenhouse Gas Control 4 No. 1 (2010), 73–89. Assayag N, et al. Water–rock interactions during a CO2 injection field-test: Implications on host rock dissolution and alteration effects. Chemical Geology 265 No. 1–2 (2009), 227–235. Aquilina L, et al. Water-rock interaction processes in the Triassic sandstone and the granitic basement of the Rhine Graben: Geochemical investigation of a geothermal reservoir. Geochimica et Cosmochimica Acta 61 No. 20 (1997), 4281–4295. Fischer S, Liebscher A, Wandrey M. CO2 –brine–rock interaction — First results of long-term exposure experiments at in situ P–T conditions of the Ketzin CO2 reservoir. Chemie der Erde – Geochemistry 70 (2010), 155–164. Rempel KU, et al. An experimental investigation of trace element dissolution in carbon dioxide: Applications to the geological storage of CO2 . Chemical Geology 289 No. 3–4 (2011), 224–234. Pedersen K. Microbial life in deep granitic rock. FEMS Microbiology Reviews 20 No. 3–4 (1997), 399–414. Mitchell AC, et al. Microbially Enhanced Carbon Capture and Storage by Mineral-Trapping and Solubility-Trapping. Environmental Science & Technology 44 No. 13 (2010), 5270–5276. Kuenen JG, Anammox bacteria: from discovery to application. Nat Rev Microbiol 6 No. 4 (2008), 320–326. Boetius A, et al. A marine microbial consortium apparently mediating anaerobic oxidation of methane. Nature 407 No. 6804 (2000), 623–626. Madigan MT, et al. Brock Biology of Microorganisms. Addison-Wesley Longman, Amsterdam, 2011. 13. Zettlitzer M, et al. Re-establishment of the proper injectivity of the CO2 -injection well Ktzi 201 in Ketzin, Germany. International Journal of Greenhouse Gas Control 4 No. 6 (2010), 952–959. Wandrey M, et al. Assessing drilling mud and technical fluid contamination in rock core and brine samples intended for microbiological monitoring at the CO2 storage site in Ketzin using fluorescent dye tracers. International Journal of Greenhouse Gas Control 4 No. 6 (2010), 972– 980. Pellizzari L, et al. The use of tracers to assess drill-mud penetration depth into sandstone cores during deep drilling: method development and application. Environmental Earth Sciences (2013), 1–12. Videmsek U, et al. Abundance and diversity of CO2 -fixing bacteria in grassland soils close to natural carbon dioxide springs. Microb Ecol 58 No. 1 (2009), 1–9.

222 | 10 Microbes in geo-engineered systems [88]

[89] [90] [91] [92] [93]

[94] [95] [96]

[97]

[98] [99]

[100]

[101]

[102] [103] [104] [105] [106] [107] [108]

Krüger M, et al. Effects of elevated CO2 concentrations on the vegetation and microbial populations at a terrestrial CO2 vent at Laacher See, Germany. International Journal of Greenhouse Gas Control 5 No. 4 (2011), 1093–1098. Yakimov MM, et al. Microbial community of a saline mud volcano at San Biagio-Belpasso, Mt. Etna (Italy). Environmental Microbiology 4 No. 5 (2002), 249–256. Oppermann BI, et al. Soil microbial community changes as a result of long-term exposure to a natural CO2 vent. Geochimica et Cosmochimica Acta 74 No. 9 (2010), 2697–2716. Inagaki F, et al. Microbial community in a sediment-hosted CO2 lake of the southern Okinawa Trough hydrothermal system. Proc Natl Acad Sci USA 103 No. 38 (2006), 14 164–14 169. Krüger M, et al. Ecosystem effects of elevated CO2 concentrations on microbial populations at a terrestrial CO2 vent at Laacher See, Germany. Energy Procedia 1 No. 1 (2009), 1933–1939. Wu B, et al. Viability and metal reduction of Shewanella oneidensis MR-1 under CO2 stress: implications for ecological effects of CO2 leakage from geologic CO2 sequestration. Environ Sci Technol 44 No. 23 (2010), 9213–9218. Dupraz S, et al. Impact of CO2 concentration on autotrophic metabolisms and carbon fate in saline aquifers – A case study. Geochimica et Cosmochimica Acta 119 (2013), 61–76. Mayumi D, et al. Carbon dioxide concentration dictates alternative methanogenic pathways in oil reservoirs. Natur Comm 4 (2013). Kirk MF, Variation in energy available to populations of subsurface anaerobes in response to geological carbon storage. Environmental Science and Technology 45, No. 15 (2011), 6676– 6682. Kharaka YK, et al. Changes in the chemistry of shallow groundwater related to the 2008 injection of CO2 at the ZERT field site, Bozeman, Montana. Environmental Earth Sciences 60 No. 2 (2010), 273–284. Stumm W, Morgan JJ, Aquatic Chemistry. John Wiley & Sons: New York, 1996. 3:1040. Gerdemann SJ, et al. Carbon dioxide sequestration by aqueous mineral carbonation of magnesium silicate minerals. 2nd Annual Conference on Carbon sequestration, Alexandria, VA, May 5–8, 2003. Köhler P, et al. Geoengineering impact of open ocean dissolution of olivine on atmospheric CO 2 , surface ocean pH and marine biology. Environmental Research Letters 8 No. 1 (2013), 014009. Mitchell AC, Ferris FG, The coprecipitation of Sr into calcite precipitates induced by bacterial ureolysis in artificial groundwater: Temperature and kinetic dependence. Geochimica et Cosmochimica Acta 69 No. 17 (2005), 4199–4210. Phillips AJ, et al. Engineered applications of ureolytic biomineralization: A review. Biofouling 29 No. 6 (2013), 715–733. Phillips AJ, et al. Potential CO2 leakage reduction through biofilm-induced calcium carbonate precipitation. Environmental Science and Technology 47 No. 1 (2013), 142–149. Ferris FG, et al. Bacteriogenic Mineral Plugging. The Journal of Canadian Petroleum Technology 35 No. 8 (1996), 56–61. Fujita Y, et al. Stimulation of microbial urea hydrolysis in groundwater to enhance calcite precipitation. Environ Sci Technol 42 No. 8 (2008), 3025–3032. Mitchell AC, et al. Resilience of planktonic and biofilm cultures to supercritical CO2 . The Journal of Supercritical Fluids 47 No. 2 (2008), 318–325. Mitchell AC, et al. Biofilm enhanced geologic sequestration of supercritical CO2 . International Journal of Greenhouse Gas Control 3 No. 1 (2009), 90–99. Ménez B, et al. Impact of the deep biosphere on CO2 storage performance. Geotechnologien Sci. Rep. 9 (2007), 150–163.

References |

223

[109] McConnaughey TA, Whelan JF, Calcification generates protons for nutrient and bicarbonate uptake. Earth-Science Reviews 42 No. 1–2 (1997), 95–117. [110] Ebigbo A, et al. Darcy-scale modeling of microbially induced carbonate mineral precipitation in sand columns. Water Resources Research 48 No. 7 (2012), W07519. [111] Schultze-Lam S, et al. Mineralization of bacterial surfaces. Chemical Geology 132 No. 1–4 (1996), 171–181. [112] Van Lith Y, et al. Microbial fossilization in carbonate sediments: a result of the bacterial surface involvement in dolomite precipitation. Sedimentology 50, No. 2 (2003), 237–245. [113] Bell PE, Mills LA, Herman SJ. Biogeochemical Conditions Favoring Magnetite Formation during Anaerobic Iron Reduction. Applied and Environmental Microbiology 53 No. 11 (1987), 2610–2616. [114] Nilsen RK, Torsvik T, Lien T. Desulfotomaculum thermocisternum sp. nov., a Sulfate Reducer Isolated from a Hot North Sea Oil Reservoir. International Journal of Systematic Bacteriology 46 No. 2 (1996), 397–402. [115] Mehanna M, et al. Role of direct microbial electron transfer in corrosion of steels. Electrochemistry Communications 11 No. 3 (2009), 568–571. [116] Bénézeth P, Ménez B, Noiriel C. CO2 geological storage: Integrating geochemical, hydrodynamical, mechanical and biological processes from the pore to the reservoir scale. Chemical Geology 265 No. 1–2 (2009), 1–2. [117] Carvalho MR, Forjaz HV, Almeida C. Chemical composition of deep hydrothermal fluids in the Ribeira Grande geothermal field (São Miguel, Azores). Journal of Volcanology and Geothermal Research 156 No. 1–2 (2006), 116–134. [118] D’Amore F, Giusti D, Abdallah A. Geochemistry of the high-salinity geothermal field of Asal, Republic of Djibouti, Africa. Geothermics 27 No. 2 (1998), 197–210. [119] Gallup DL, Geochemistry of geothermal fluids and well scales, and potential for mineral recovery. Ore Geology Reviews 12 No. 4 (1998), 225–236. [120] Kharaka YK, Cole DR. Geochemistry of Geologic Sequestration of Carbon Dioxide. 2011. p. 133–174. In: Frontiers in Geochemistry: Contribution of Geochemistry to the Study of the Earth (Harmon RS, Parker A, eds). John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9781444329957.ch8. [121] Jun Y-S, Giammar ED, Werth JC. Impacts of Geochemical Reactions on Geologic Carbon Sequestration. Environmental Science & Technology 47 No. 1 (2012), 3–8. [122] Vetter A, et al. Fluid chemistry and impact of different operating modes on microbial community at Neubrandenburg heat storage (Northeast German Basin). Organic Geochemistry 53 (2012), 8–15. [123] Bundesverband Geothermie e.V.; http://www.geothermie.de/fileadmin/useruploads/ aktuelles/projekte/tiefe/deutschland/TG-Projekte_2013_Internet_16-Juli_name.pdf. Accessed October 2013. [124] Mundhenk N, et al. Metal corrosion in geothermal brine environments of the Upper Rhine graben – Laboratory and on-site studies. Geothermics 46 (2013), 14–21. [125] Mundhenk N, et al. Corrosion and scaling as interrelated phenomena in an operating geothermal power plant. Corrosion Science 70 (2013), 17–28. [126] Lerm S, et al. Influence of microbial processes on the operation of a cold store in a shallow aquifer: impact on well injectivity and filter lifetime. Grundwasser 16 No. 2 (2011), 93–104. [127] Pronk M, Goldscheider N, Zopfi J. Microbial communities in karst groundwater and their potential use for biomonitoring. Hydrogeology Journal 17, No. 1 (2009), 37–48. [128] Noble RRP, et al. Biological monitoring for carbon capture and storage – A review and potential future developments. International Journal of Greenhouse Gas Control 10 (2012), 520– 535.

224 | 10 Microbes in geo-engineered systems [129] Zimmer M, Erzinger J, Kujawa C. The gas membrane sensor (GMS): A new method for gas measurements in deep boreholes applied at the CO2 SINK site. International Journal of Greenhouse Gas Control 5 No. 4 (2011), 995–1001. [130] Freifeld BM, et al. The U-tube: A novel system for acquiring borehole fluid samples from a deep geologic CO2 sequestration experiment. Journal of Geophysical Research: Solid Earth 110 No. B10 (2005), B10 203. [131] Schieber J. Sedimentary pyrite: A window into the microbial past. Geology 30 No. 6 (2002), 531–534. [132] Dupraz S, et al. Experimental approach of CO2 biomineralization in deep saline aquifers. Chemical Geology 265 No. 1–2 (2009), 54–62. [133] Jones DG, et al. In: Salah gas CO2 storage JIP: Surface gas and biological monitoring. Energy Procedia 4 (2011), 3566–3573. [134] Kallmeyer J, et al. A high-pressure thermal gradient block for investigating microbial activity in multiple deep-sea samples. Journal of Microbiological Methods 55 No. 1 (2003), 165–172. [135] Deusner C, Meyer V, Ferdelman GT. High-pressure systems for gas-phase free continuous incubation of enriched marine microbial communities performing anaerobic oxidation of methane. Biotechnology and Bioengineering 105 No. 3 (2010), 524–533.

Charles S. Cockell

11 The subsurface habitability of terrestrial rocky planets: Mars 11.1 Introduction One promising area of science is the expansion of our understanding of subsurface environments and their geochemical or even biological potential beyond the Earth. However, because of high mission costs, and the complexity of robotic and human missions, the data on extraterrestrial environments is limited. Of the terrestrial rocky planets that could have harbored, and might still host, subsurface aqueous environments, Mars is a high priority target. Apart from the Earth, of the terrestrial rocky planets Mars has associated with it the largest body of information on its past and present geochemical conditions, both surface and subsurface. In this chapter, I explore some of the existing knowledge about the habitability of the Martian subsurface, making comparisons with the terrestrial subsurface and analogue environments where appropriate, and conclude by suggesting ten testable hypotheses concerning the habitability of the Martian subsurface environment. The surface of Mars today is inhospitable. The surface atmospheric pressure is close to the triple point, such that standing bodies of liquid water cannot persist [1]. The surface of the planet is subjected to an intense ultraviolet (UV) radiation flux compared to the surface of the Earth on account of the lack of a sufficiently thick ozone column to screen UV (200–280 nm) radiation. As a result, the DNA damage-weighted UV irradiance on the surface of the planet is, in the worst-case scenario, about one thousand times higher than on the surface of the Earth [2]. Ultraviolet radiation can be readily screened by thin (tens of microns) layers of dust depending on geometry and size, such that it is rapidly attenuated beneath the surface [3]. However, Mars lacks a magnetic field and so penetrating ionizing radiations are also at a much higher dose in the near-surface environment than on the Earth [4–6]. In the early history of Mars, the atmospheric pressure was probably higher; evidence exists in the form of lake remnants, rivers and other geomorphological features that attest to a time during the early Noachian (4.1–3.7 Ga ago) when there were persistent bodies of liquid water on the surface [7–14]. However, apart from catastrophic outflow channels, much of this water had gone from the surface by the Hesperian (3.7–3.1 Ga ago). These factors have meant that the subsurface of Mars has long been recognized as potentially the most persistently habitable region of Mars over time and one plausible location to search for evidence of past or present life [15].

226 | 11 The subsurface habitability of terrestrial rocky planets: Mars The study of the Martian subsurface has important applications for understanding the habitability of terrestrial rocky planets in general. If subsurface environments are discovered on Mars that are uninhabitable, or contain the prerequisites for life, but did not host life [16], this would encourage a search for the requirements that might be missing or an investigation into why habitable environments on Mars are uninhabited, thus expanding our understanding of the conditions for life within the interior of rocky planets.

11.2 The subsurface of Mars – our current knowledge Despite the obvious importance of the Martian subsurface to astrobiology, our knowledge of its physical and geochemical conditions is scarce. Nevertheless, in recent years, information on the subsurface has been gleaned from radar observations, the study of materials exposed by erosion, canyons or excavated by asteroid and comet impact events and information from ‘analogue’ environments on the Earth, i.e. environments that in one or more parameters are identical or approximate to physicochemical conditions in Martian environments [e.g. 17, 18]. The near-surface (tens of centimeters) of Mars is thought to harbor water ice deposits that vary from 2% wt at the equator to pure ice at the polar regions [19–21] mixed with surface volcanic regolith as determined by the Mars Odyssey Gamma-Ray Spectrometer (GRS) and Neutron Spectrometer (MONS) (󳶳 Fig. 11.1). These instruments detect gamma photons emitted by hydrogen and neutrons emitted in collisions between cosmic rays and elemental nuclei. The Martian cryosphere is estimated to contain an equivalent global layer of water of ∼ 35 m [19].

Fig. 11.1: Hydrogen (presumed to be a proxy for water) distribution in the subsurface of Mars as measured by the Mars Odyssey Gamma-Ray spectrometer (image credit NASA).

11.2 The subsurface of Mars – our current knowledge |

227

Direct observations of the present-day Martian deep subsurface (up to several kilometers depth) were made using the Mars Advanced Radar for Subsurface and Ionospheric Sounding (MARSIS) aboard the Mars Express spacecraft in 2005 [22], which has a theoretical penetration depth of ∼ 5 km. This instrument was a multifrequency, synthetic aperture, orbital sounding radar. The instrument provided information on the subsurface of the north polar layered deposits on Mars down to ∼ 1.8 km, confirming that the deposits are mainly composed of pure water ice. Flyovers of the mid-latitude Chryse Planitia region detected reflections corresponding to a ∼ 250 km circular structure, which is presumed to have been formed by an impact event and shows the deep subsurface structure of basin formation in that region of the planet [23] (󳶳 Fig. 11.2). The MARSIS instrument not only demonstrated that radar can be used to map and gain compositional information in the Martian deep subsurface from orbit, but it also underlined the point that impact events, by causing disruption to geological units or forming basins into which material is subsequently deposited, have had an important influence on the Martian deep subsurface. On account of subduction and atmospheric and hydrological erosion, craters are rare on the Earth and are not considered an important agent in subsurface geological or biological processes. On Mars, however, the lack of deep erosional processes and subduction results in a well-preserved impact record. This is discussed in relation to habitability later in this chapter. The Shallow Radar (SHARAD) instrument on Mars Reconnaissance Orbiter (MRO) has similarly provided direct information on conditions in the subsurface and has a penetration depth of ∼1.5 km. The study of subsurface structures in lobate debris aprons (LDAs), which are broad, lobate features that extend up to 20 km away from steep slopes in equatorial regions of Mars, suggests that there are buried glaciers [24]. The glaciers are covered by a fine layer of the ubiquitous Martian dust and other rocky material. The data from the SHARAD instrument suggests that up to ∼ 28,000 km3 of water ice might be sequestered in LDAs in the Hellas Basin region of Mars alone, equivalent to a global water layer ∼ 20 cm thick. That glaciers exist in equatorial regions of Mars is remarkable. The authors attribute their finding to the large obliquity (axial tilt) changes that Mars experiences. During times when obliquity was ∼ 45 °, models predict that during the southern summer solstice, large amounts of water vapor would be formed. The vapor would be transported northwards and deposited as snow by condensation and precipitation. Eventually, it was covered by debris. Today, these buried glaciers are testament to climatic changes on Mars over time scales of several million years [25]. In general, the water ice on Mars, however, follows its predicted depth of stability under current climatic conditions [20], consistent with the idea that the subsurface ice conditions of Mars follow orbitally-driven climate cycles, with local heterogeneities reflecting differences in topography and material type and preservation of icy deposits from previous epochs.

228 | 11 The subsurface habitability of terrestrial rocky planets: Mars

Fig. 11.2: Radar (Mars Advanced Radar for Subsurface and Ionospheric Sounding (MARSIS) aboard the Mars Express spacecraft) images of the subsurface of Cryse Planitia. Top: Radargrams for MARSIS orbits 1903 (a) and 1892 (b) that both cross Chryse Planitia. Radar-sounding echo intensities are plotted in gray scale as a function of time-delay versus position along the orbit track (north to the right). Bottom: Ground-range projections of the orbit 1903 (c) and 1892 (d) radargrams overlaid on MOLA (Mars Orbiter Laser Altimeter) color-coded shaded relief. Parabolic echoes in the radargrams that project as arcs on the surface are interpreted to be from the near and far rim walls of buried impact basins. The dashed white circles are approximate fits to the arcs in the radargrams. (image credit European Space Agency).

The presence of ice in the subsurface of Mars is confirmed by impact cratering studies. Impact craters penetrate to greater depths than those examined by radar soundings. Layered ejecta blankets around craters – indicative of high volatile (water) content can be used to estimate the depth at which subsurface water ice exists based on known relationships between crater diameters and excavation depth [26, 27]. Quite apart from massive ice deposits at the poles (with an estimated volume of 1.2–1.7 × 106 km3 [28]) and buried ice elsewhere, there is a large literature on glacial

11.2 The subsurface of Mars – our current knowledge |

229

Fig. 11.3: Images of a fresh, 6-meter-wide (20-foot-wide) crater on Mars on Oct 18, 2008, (left) and on Jan 14, 2009 taken by the High-Resolution Imaging Science Experiment camera on NASA’s Mars Reconnaissance Orbiter. Each image is 35 meters (115 feet) across. (image credit NASA/JPL-Caltech/University of Arizona).

and periglacial features on Mars. Evidence for subsurface ice includes inferred features associated with ice, such as polygonal structures, gullies, lobate debris aprons, deformation features in putative permafrost terrain, ice-sublimation related features and parallel sorted stone stripes, among others [29–34]. Pingos, which are produced by liquid water injection into the subsurface, with subsequent freezing (causing upheaval), have been suggested [31, 35]. They are of special interest as their formation mechanism requires bulk liquid water movement. Direct observations of ice have been made on the surface, for example by the Phoenix Lander [36]. Observations from orbit include recently excavated small craters which reveal water ice. The ice is observed to sublimate away after several months [37] (󳶳 Fig. 11.3). Modeling studies suggest that ice can be preserved over these time scales at 45 ° latitude if covered by a 15–50 cm dry layer of regolith and dust. These observations highlight the fact that the Martian subsurface, in the recent geological past, has been subject to changes caused by the much more variable Martian obliquity compared to the Earth, changing ice dynamics on a planetary scale. Today, ice remains an important feature defining the physical and geochemical environment of the Martian subsurface, at least in the top few hundred meters to kilometer depths. In addition to direct information on the subsurface conditions of present-day Mars, orbital data has also given us insights into the conditions and processes in the ancient crust of Mars, particularly during the Noachian. The Nili Fossae region of Mars exhibits an extraordinary range of minerals and is comprised of distinct units (󳶳 Fig. 11.4). Like much of our present data on subsurface conditions, information has been obtained using the Mars Reconnaissance Orbiter (MRO), Compact Reconnaissance Spectrometer for Mars (CRISM) and the HighResolution Science Experiment (HiRISE) instruments. One unit in the Nili Fossae is

230 | 11 The subsurface habitability of terrestrial rocky planets: Mars

Fig. 11.4: Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) image of part of a fracture in the Nili Fossae region near 21.9°N, 78.2°E. The top left shows the Isidis basin image (small square) supposed on a Mars Orbiter Laser Altimeter (red higher elevations, blue lower). The fracture shown, which is 11 km at its narrowest point (top right) is overlain on a Viking digital image (lower left) to show topography. Top right shows results in infrared channels, false colored. Bright green is phyllosilicates, yellow-brown are olivines, purple are pyroxenes. The CRISM data is superposed on High-Resolution Imaging Science Experiment (HiRISE) (lower right) showing that phyllosilicates are in small eroded outcrops of rock and olivines in sand dunes. (image credit NASA/JPL/JHUAPL/Brown University).

a brecciated Fe/Mg-smectite-bearing unit which contains meter to kilometer-sized blocks of altered and unaltered rock and is inferred to represent the ancient crust of Mars, torn up in subsequent impact events [38]. Fe/Mg-smectite clays are the most abundant clays on Mars, followed by chlorites. A second, olivine-rich unit with evi-

11.2 The subsurface of Mars – our current knowledge |

231

Fig. 11.5: False color image of the interior (approximately 460 meters across) of an unnamed crater on the central uplift of the 467-km diameter Huygens crater imaged by the High-Resolution Imaging Science Experiment (HiRISE) on the Mars Reconnaissance Orbiter. The light toned rocks contain carbonates and have been excavated in the central uplift from ∼ 5 km depth. (image credit NASA/JPL-Caltech/University of Arizona).

dence for aqueous alteration may be impact melt or early-stage lava from the Syrtis Major [39]. The presence of prehnite suggests low temperature (200–350 °C) alteration and the wide variety of hydration products including kaolinite, chlorite, mica, opal, zeolites, sulfates, serpentine, smectites and carbonates suggest complex geochemical interactions [38–41]. In particular, asteroid and comet impact craters, which can be regarded as “nature’s drill”, provide insights into the composition of the Martian subsurface and its geochemical characteristics in the ancient past. A study of 31 craters in the Valles Marineris region revealed two very distinctive types of material. One type was a set of deformed, folded and fractured rocks that are a mixture of olivine and high-calcium pyroxene associated with hydrated phases and could be representative of a thick Noachian volcanic deposition up to 18 km in depth. The second set of materials are enriched in low calcium pyroxene and olivine, but also contain serpentine and smectites and attest to possible hydrothermal processes beneath the surface potentially associated with impact events [42]. Central peaks of craters from other regions of Mars also show indications of hydrated phases [41]. These latter massive light-toned rocks are quite homogeneous from crater to crater and have been observed in the southern highlands of Mars [43, 44] suggesting that they may be representative of the Noachian crust on a wide scale. Crater-central uplifts can reveal remarkable data about the processes occurring in the deep subsurface (󳶳 Fig. 11.5). The central uplift of equatorial Leighton Crater has excavated material from 6 km depth in the Martian subsurface. CRISM data shows the presence of carbonates, kaolinite-group elements and Fe/Mg-bearing silicates consistent with serpentine, chlorite, vermiculite and pumpellyite [43]. One model to explain these observations [45] is that carbonate-bearing siliciclastic sediments and volcanics were present at the surface of Mars during the Noachian, similar to units observed 1500 km to the northeast in the Nili Fossae region.

232 | 11 The subsurface habitability of terrestrial rocky planets: Mars Heat from the overlying Hesperian age lavas from the Syrtis Major volcano would have caused water release from the hydrated minerals as well as aqueous CO2 release from the carbonates. Heat would have converted Fe-Mg smectites and olivine to chlorite-smectite and eventually into the serpentine-containing assemblage observed today. Carbonate minerals may have been recrystallized [45]. Impact central uplifts show that the greatest alteration in the Noachian subsurface was occurring at depths greater than ∼ 5 km [41] with material at shallower depths (∼ 2 km) less altered, suggesting an unsaturated subsurface zone [46]. Surface features can reveal important information about the past subsurface geochemistry of Mars in places where the surface is composed of ancient subsurface materials now exposed through erosion. Layered deposits formed in equatorial regions are interpreted to be sandstones formed in shallow fluvial or Aeolian systems [47, 48]. Their formation is contemporaneous with aqueous activity on the surface of Mars before 3.5 Ga ago. Since then, erosion has made these features visible, giving us a direct view of the subsurface of Mars when liquid water was abundant. The presence of deformation bands [49], features that are precursors to fracturing, shows not only that deformation occurred, but that water flow in the past subsurface of Mars might have been influenced, and channeled, by these features. Discoloration along the boundaries of the bands is interpreted to show aqueous alteration of primary minerals [49]. Variations in discoloration along the bands are taken to suggest heterogeneity in past Martian subsurface water flow and spatial differences in subsurface water geochemistry. Whatever the reason for these discontinuities in discoloration, they underscore the important point that the Martian subsurface, like the Earth, was a heterogeneous environment in terms of the locations and extent of water-rock interactions. All of these data can be summarized by observing that clays on Mars, most abundantly iron-magnesium smectites and chlorite, which are found in 78 and 39% of crustal sites examined by Ehlmnann et al. [50], are almost exclusively associated with ancient Noachian terrain. They show that a distinctive set of subsurface processes was occurring at that time in a closed system. These processes could have included magmatic precipitation, hydrothermal activity and low-grade metamorphism. The presence of other minerals including serpentine, silica, prehenite and illite suggests excavation of minerals formed at higher temperatures, for example in subsurface processes of impact crater hydrothermal systems. The lack of chloride and sulfates associated with these deposits (the latter generally found in Hesperian and Amazonian terrains exposed to surface conditions, although there are examples of interbedded clays and sulfates) is further evidence that sedimentary mineral formation did not occur and that the system was closed (i.e. subsurface and isolated from the surface atmospheric conditions) [50–52]. In conclusion to this section, it is perhaps worth highlighting the stark difference in our understanding of the terrestrial and Martian subsurface. In the case of the Earth, our understanding of its subsurface conditions and habitability is largely restricted to

11.3 Martian subsurface habitability, past and present | 233

the modern day, information gained from International Continental Drilling Programs (ICDP), Ocean Drilling Programs (ODP) and other deep-drilling efforts (see other chapters in this volume). On Mars, by contrast, our understanding of the subsurface is essentially restricted to present-day radar observations and geological inferences about subsurface conditions greater than 3.7 Ga ago. This remarkable difference is caused by our lack of direct access to the present-day Martian subsurface, and the lack of plate tectonics and erosional processes on that planet which leave the ancient subsurface of Mars intact and exposed by processes such as asteroid and comet impact events. For this reason, comparative subsurface planetology between Earth and Mars, particularly in assessing habitability, is currently difficult to achieve with reliability and will always be underpinned by large assumptions about planetary processes and their association with life over multibillion year time scales.

11.3 Martian subsurface habitability, past and present Although very little is known about geochemical and physical conditions in the Martian subsurface, past and present, it is possible to constrain some parameters. In this section of the chapter I explore a number of factors that would influence subsurface habitability (including vital elements, liquid water, plausible energy supplies and appropriate physico-chemical conditions) and make observations on our current state of knowledge about Martian subsurface habitability, past and present.

11.3.1 Vital elements (C, H, N, O, P, S) Life requires these six basic elements to construct macromolecules. Carbon atoms are likely to have been, and continue to be, present in the near subsurface of Mars as a consequence of atmospheric exchange (present day 95.32% CO2 ; 800 ppm CO). The detection of carbonates [53] suggests that aqueous interactions with these rocks could generate a source of inorganic carbon through Martian history. Ancient reservoirs of stored carbon dioxide produced during a time when the Martian atmosphere was thicker are an additional plausible surface source that could have made its way into subsurface reservoirs [54]. Organic carbon is more equivocal. However, the infall of carbonaceous chrondites and other organic carbon-bearing material [55] would suggest that organic material should be present as a source of carbon, although on present-day Mars it may be destroyed by reactive oxygen species and ionizing radiation in the near-surface environment [e.g. 4, 5, 55–57]. Reduced magmatic carbon in Martian basalts [57] is also another potential source of carbon. Hydrogen atoms are potentially available from water, which could be split radiolytically [58]. The presence of serpentine in impact crater uplifts [40–42], suggests

234 | 11 The subsurface habitability of terrestrial rocky planets: Mars the presence of sources of hydrogen through serpentinization reactions, at least in the Noachian and possibly today. Nitrogen is present in the modern atmosphere at 2.7%, but it is unknown whether deep subsurface life could fix this nitrogen. Nitrogen fixation could occur by abiotic processes, including impact events, lightening and volcanic activity [59–61]. The concentrations reached and the depths achieved by nitrogen fixed in such processes throughout Martian history is unknown. Boxe et al. [62] use a one-dimensional model to show that fixed nitrogen species, some produced photochemically (for example NO−2 , NO, HNO3 ), could be generated on the surface of Mars and then transported into near-surface environments in thin water films. Similarly, NO and other abioticallyfixed species have been suggested as the earliest nitrogen sources and as biological electron sinks on early Earth [63]. This transient photochemically-produced nitrogen cycle on Mars could provide a source of fixed nitrogen species today, but the depth of its penetration might be low. Without the continuous flow of fixed nitrogen into the deep subsurface of Mars, nitrogen may be, and may have been, one of the major limiting factors for a deep subsurface biosphere. Oxygen atoms could be provided by CO2 , H2 O, sulfates, perchlorates, other reactive oxygen species. Oxygen atoms are bound to many of the biologically-accessible compounds discussed here in association wit other elements (C, H, N, P, S). Phosphate is present in the form of apatite, a common secondary mineral in basalts. Mössbauer, MiniTES and APXS spectra from the Mars Exploration Rovers are interpreted to suggest apatite concentrations (wt%) at between 0.1 and 2.4% [64]. Although the mineral structure of the subsurface is different from surface formations [42], the data nevertheless provide empirical support for apatite being a secondary mineral present in Martian basalts, similarly to Earth, and thereby suggest a potential phosphate source. Sulfur has been widely detected on Mars in the form of sulfate salts including ferric sulfates [65–68]. Jarosite and other sulfur-bearing substances have been detected [69]. The extent of these compounds in the subsurface is not known, but the dominance of the sulfur cycle on Mars [69] suggests that sulfur species would have been distributed from the mantle to the surface throughout Martian history, potentially including sulfur in microbially-accessible gaseous phases such as H2 S and SO2 .

11.3.2 Other micronutrients and trace elements The presence of widespread ultramafic and basaltic rocks on Mars and their alteration products show that micronutrients are available for life. Fe is abundant in the ferrous state in olivines and in the ferric state in a variety of materials from clays to ferric oxides and sulfates [50, 66]. Mg and Ca are present in materials such as clays and pyroxenes, and K and Na in materials such as muscovite, illite and plagioclases [50].

11.3 Martian subsurface habitability, past and present | 235

The inferred presence of minerals such as chromite [64] suggests that many other trace elements and metals are available for life throughout the Martian depth profile. There is likely to be a strong depth dependence in which the form of these elements are found. In the Noachian deep subsurface (and possibly in the deep subsurface today), in confined closed aqueous systems, water may have been heavily enriched in Ca, Mg and Fe in ultramafic ultrabasic environments [70]. In the near-surface environments, both in the present-day and the past, many of these elements would also exist in brines and structurally or trapped oxides and sulfates as well as within basalts [66, 69, 71, 72].

11.3.3 Liquid water through time Liquid water at the surface of Mars today is rendered unstable, partly because much of the surface is at the triple point and partly because the low humidity means that when liquid water is formed, it will rapidly evaporate, even if it does not boil [1]. However, the presence of ancient liquid water in the subsurface of Mars is supported by the observations of clays associated with ancient crustal units exposed by craters and erosion, as described previously. Although the exact mechanisms by which the water interacted with rocks are unknown, the evidence of closed aqueous environments in the ancient deep subsurface of Mars is strong, based on observed aqueous alteration of ancient crustal rocks [50]. The presence of valley networks [7, 8] and a possible northern ocean [10, 11], lakes [12, 13] and evidence of seas [11, 14] on the surface of the planet attests to a time when persistent bodies of liquid water were abundant on the surface of Mars, and by extension, in the subsurface throughout much of the early Noachian (󳶳 Fig. 11.6). After the Noachian, the planet’s declining heat flow is hypothesized to have led to freezing of most of the water in the surface environment and at gradually increasing depth, confining groundwater to a region beneath [11, 73]. One estimate put the cryosphere depth to be ∼ 2.5 km depth at the equator to ∼ 6.5 km depth at the poles [74, 75]. However, downward revisions of the geothermal gradient of Mars suggest that the depth could be up to two to three times greater [76]. Salt solutions would lessen these

Fig. 11.6: Gusev crater (approximately 166 km in diameter) landing site of the Mars Spirit rover in 2003 was likely to be the site of an ancient Martian lake. Image taken by the Mars Odyssey Thermal Emission Imaging System. (image credit NASA/JPL/Arizona State University).

236 | 11 The subsurface habitability of terrestrial rocky planets: Mars

Fig. 11.7: Subsurface water on Mars since the Nochian? Aromatum Chaos, Mars (~43° 8󸀠 W, 1° 7󸀠 S) images by Viking Orbiter (Mars Digital Image Model). The outflow channel feeds into Hydroates Chaos, ultimately extending onto the plains of Chryse Planitia (image credit NASA).

depths by depressing the freezing point. For example, perchlorate at high concentrations would depress the freezing point to 203 K, such that the crysophere would be nonexistent at the equator today. However, subsurface liquid water availability since the Noachian is suggested by the presence of catastrophic outflow channels. Geomorphologically, these features begin from a fracture or region of chaotic terrain and consist of broad depressions tens to thousands of kilometers long with streamlined islands and deposits around craters along their beds [8, 9, 77, 78] (󳶳 Fig. 11.7). It is hypothesized that the mechanisms by which they might be formed include the release of groundwater from the crysophere by impact, earthquake or magmatic intrusion. Some of these channels date back to greater than 3 Ga ago [79]. However, some, such as Athabasca, may have had activity just a few million years ago [80, 81]. If water is the mechanism of their formation, then outflow channels would suggest the presence of liquid water in the subsurface from the early Hesperian through to the geologically very recent past in catastrophic episodes. They might therefore be evidence of the presence of permanent deep groundwater sources throughout Martian history. On present-day Mars, it is reasonable to hypothesize that groundwater could exist deep underground where radiogenic heating and lithostatic pressures would allow liquid water to exist above the freezing point. The presence of salts, which depress the freezing point, would reduce the depth at which these waters were plausible [76]. Impact events would be another mechanism by which the present-day cryosphere could be disrupted to create a link between the subsurface and surface [82]. Evidence has been presented for near-surface present-day liquid water. Gullies, which have characteristic alcoves located on a steep slope with an incised sinuous channel leading down to an apron of deposited material, have been proposed as evidence of present-day liquid water [83–85]. The observation of some of these features high up on impact crater walls and hills is difficult to reconcile with plausible subsurface water sources. It is hypothesized that these features may therefore be formed from two different processes: CO2 -induced fluidization of regolith and subsurface water [86–88].

11.3 Martian subsurface habitability, past and present | 237

Seasonally recurrent dark slope streaks [89] could also represent present-day surface expressions of near-surface salty water with depressed freezing point. Droplets of putative salt solutions on the legs of the Phoenix lander provide evidence for physical and thermodynamic stability of brines on Mars [90] and deliquescence provides one mechanism by which these brines might form [91]. Perennially cold springs in the High Arctic on Earth have been proposed as potential analogue environments to understand the geochemistry and habitability of Martian briny water [92, 93]. Unfortunately, radar searches for putative present-day liquid water have not been successful [94]. A number of factors, including a very dry conductive surface have limited radar measurements in most areas of Mars to less than 100 m, which would explain the lack of detection of subpermafrost groundwater. The low surface roughness and well saturated crustal porosity required to optimize radar penetration depth and signal recovery are met only in a small proportion (< 20%) of the planet. Liquid water could exist today in the form of thin water films on soil grains. Dielectric measurements of soils during the nighttime at the Phoenix landing site suggest the present of liquid water, but the lack of conductivity suggests that the water does not move [95]. Thin films of interfacial liquid water on soil grains of just a few nanometers thickness have been proposed as possible microhabitats [96]. Water is kept in the liquid phase on the surface of grains by Van der Waals forces above a threshold temperature. They are thought to be able to exist as no more than two monolayers at temperatures down to 163 K within the top 20 cm in the Martian subsurface [91, 97, 98]. Whether microorganisms can access thin interfacial layers of water or use it as a solvent, since it is tightly bound to the grains, is not known. However, a serious limitation of this water is that if it remains static (as is suggested for the Phoenix lander site), the microhabitat it creates will geochemically run down, becoming depleted in essential nutrients, making this water a poor environment for the long-term sustenance of life. In the more recent geological history of Mars, bulk liquid water might have become available as a result of warming during higher obliquity. This is possibly the case for ices at the Phoenix landing site [95] (󳶳 Fig. 11.8). During the last 5 Ma, obliquity increases up to 50 ° would generate surface temperatures in excess of 273 K up to 100 days a year. Ulrich et al. [99] investigated features of the Utopia Planitia and suggested that during the last 10 Ma, thaw processes would have generated liquid water, which would have contributed to geomorphological features in the region and made liquid water available to any putative life. High obliquity periods between 7.864 Ma and 7.855 Ma and four periods of duration 100–12,000 years between 9.76 Ma and 9.45 Ma could have represented times when liquid water was stable in surface or near-surface environments.

238 | 11 The subsurface habitability of terrestrial rocky planets: Mars

Fig. 11.8: Color images acquired by NASA’s Phoenix Mars Lander’s Surface Stereo Imager on the 21st and 25th days of the mission, or Sols 20 and 24 (June 15 and 19, 2008) showing ice at ∼ 3 cm depth and its sublimation over four days. (image credit NASA/JPL-Caltech/University of Arizona/Texas A&M University).

11.3.4 Redox couples Since the concentration of CHNOPS and other elements at depth cannot be reliably quantified, so too the concentrations of many redox couples. As for deep subsurface life on Earth, photosynthesis is eliminated, but unlike the Earth, a putative Martian subsurface biosphere could not profit from compounds (such as organics) generated by a productive surface photosynthetic biosphere. Therefore, subsurface Martian life would be limited to chemolithotrophic pathways or anaerobic respirations using organics from meteoritic or endogenous sources on Mars. The potential list of redox couples is exhaustively speculative, but the coupling of ferric and sulfate ions (both detected on Mars) with hydrogen (suggested from the presence of serpentine and other potential hydrogen-evolving products of ultramafic weathering in Noachian terrains) are plausible couples. H2 production from serpentinization reactions has been shown to occur in spinel-containing peridotite, olivine and pyroxene at temperatures of 55 and 100 °C [100], well below the upper temperature limit for life. On the Earth, hydrogen can act as the electron acceptor in the subsurface for redox reactions using electron acceptors known to exist on Mars including sulfate [101–103], ferric iron [103, 104] and carbon dioxide [102, 103, 105]. Nixon et al. [106] discuss the available electron acceptors for iron reduction and conclude that a range of meteoritic organics could be plausible electron donors for iron reduction in the surface and near-surface of Mars. The use of ferrous iron as an electron donor, given the large resources of ferrous-bearing minerals such as olivines, seems plausible. However, Mars probably lacks suitable electron acceptors. Oxygen is at insufficient concentrations for aerobic iron oxidation, anaerobic ferrous iron oxidation linked to nitrate is impossible to assess as there is no detection of sufficient

11.3 Martian subsurface habitability, past and present |

239

nitrate. Although perchlorates could be used as an electron acceptor, to date, iron oxidation with perchlorate has not been shown to conserve energy for growth [106]. Other redox couples could include methanogenesis, using CO2 from the atmosphere or from dissolved carbonates and H2 from serpentinization reactions. Methane itself can be oxidized by microorganisms as a source of energy. Serpentinized ultramafic rocks are known to host thriving microbial communities in the subsurface of the Earth and in surface discharge [18, 107, 108] and could provide analogies to potential water-rock-microbial interactions for the Martian subsurface. Okland et al. [108] discuss the Leka Ophiolite in mid-Norway and suggest anaerobic metabolisms in rock fractures driven by methanogenesis and Fe and Mn oxidation using fixed nitrogen sources (although, see the discussion on fixed N for Mars above) and hydrogen generated in low temperature serpentinization reactions.

11.3.5 Radiation Ultraviolet (UV) radiation is rapidly attenuated in the Martian subsurface, so although the surface flux includes wavelength components down to 200 nm, within a depth of a few tens of microns to millimeters, depending on soil particle size, UV radiation would be extinguished [3, 109]. Ionizing radiations of solar energetic particles (SEP) and galactic cosmic rays (GCR) are more penetrating. The total dose of ionizing radiation experienced on the Martian surface is estimated to be ∼ 0.8 Gy/year, much lower than the fluxes that can be tolerated by radioresistant organisms such as Deinococcus radiodurans, which can withstand doses in excess of 5 kGy without appreciable loss of viability [5]. However, inactivity would result in accumulated damage such that at 2 m depth in the Martian crust, a D. radiodurans population is estimated to suffer a six order of magnitude reduction in viability after 450,000 years. At this depth, a wet heterogeneous regolith on Mars is estimated to receive 0.03 Gy/year and a dry regolith, 0.02 Gy/year, about one order of magnitude lower than monazite-rich sands in Brazil. For a deep subsurface biota just a few meters depth or greater on Mars, particularly one that is active and can repair damage, radiation would not render the subsurface uninhabitable, but it would destroy biomarkers of inactive subsurface life over megayear time scales [6, 110].

11.3.6 Other physical and environmental factors The porosity of the subsurface of Mars may be greater than the Earth at comparable depths and lithologies because of the 3/8 Earth gravity on the planet [76]. As on the Earth, porosity will be controlled by many factors including secondary mineral infilling, sediment deposition, local rock pressure environments, influence of volcanism and other geological processes, but fundamentally there is no reason why the Martian subsurface rock environment should not be accessible to life.

240 | 11 The subsurface habitability of terrestrial rocky planets: Mars In terms of temperature and pressure, we do not have direct measurements of their profiles into the deep subsurface of Mars. As for the Earth, deep habitability is likely to be constrained when the geothermal gradient temperature exceeds the upper temperature limit for life (121 °C; [111]). Geothermal gradients of between ∼ 10 and ∼ 20 °C/km [70] imply temperature habitability depths on the order of ∼ 6–15 km. Lithostatic pressures are unlikely to render environments uninhabitable. As the pressure is proportional to the gravity and Martian gravity is approximately a third of the Earth’s it follows that at depths where temperatures exceed the upper temperature limit for life, pressures will be no higher than experienced in terrestrial deep subsurface settings. pH ranges associated with a variety of subsurface settings are within the range for life. The pH of the Martian near subsurface (centimeter depth) was measured at the Phoenix Lander site where past episodes of liquid water are suggested, and was found to be slightly alkaline, 7.7–7.9, and carbonate-buffered [112]. This pH range is benign for organisms. Although we have no direct measurements of pH in the deep subsurface, past or present, reactions of fluids with mafic and ultramafic rocks control solution chemistry and it would be expected, as on the Earth, that buffered fluids would be anoxic and alkaline or ultrabasic (pH > 10) [18, 108]. Water activity and solutes can constrain the boundaries of active microorganisms. Extremely low water activities can be generated by brines such as chlorides (CaCl2 ) and mixed sulfate brines [71, 72, 113]. The lack of chlorides or sulfates associated with Noachian clays [41] suggests that although deep subsurface ancient brines cannot be excluded, the water was likely not comprised of extremely concentrated brines that would have produced deleterious water activities. On the present-day Mars, seasonally recurrent dark slope streaks [89] could be made of concentrated briny water. Some Martian brines are calculated to have water activities below those required for life and would not be habitable [72].

11.3.7 Acidity Although many environments in the ancient history of Mars may have been neutral to alkaline, the presence of sulfate minerals on the surface of Mars, including jarosite, iron sulfates, gypsum and others suggests the potential for locally acidic conditions. Sulfates are found as Hesperian layered sulfates, polar deposits, interior layered deposits, sediments in craters, within the globally ubiquitous Martian dust (which may contain 5–10% sulfates) and as sulfate veins within rocks [52, 68, 114–117]. Mg- and Fe-bearing sulfates are generally more common than Ca-bearing sulfates [69]. Some of these minerals suggest a period of acidic weathering (pH 2–5) (particularly during the Hesperian) during which exhalations of SO2 from Martian volcanic activity would have produced acidic and oxidizing rain, which subsequently weathered Martian basalts to produce secondary sulfate minerals in low water-rock ratio

11.3 Martian subsurface habitability, past and present | 241

interactions. Thus, acidic liquid water produced during these episodes would have influenced the subsurface environment if it had percolated from the surface into the deep subsurface regions. These sulfate salts are also testament to the fact that the surface and subsurface geochemical cycles of Mars have been strongly influenced by the sulfur cycle, as compared to the Earth, where the carbon cycle dominates [69]. A comprehensive study of a subsurface terrestrial analogue for putative Martian acidic subsurface conditions was undertaken at the Rio Tinto Basin in Spain [118]. The Rio Tinto Basin provides an environment to test the hypothesis that anaerobic metabolisms dependent on iron and sulfur oxidation could operate in deep acidic environments. By sampling groundwater and rocks at subsurface intervals up to a depth of 165 m before and after interactions with sulfides, the Mars Astrobiology Research and Technology Experiment (MARTE) project sought to understand potential biogeochemical interactions. The investigation of the boreholes revealed three zones. A near-surface zone to ∼ 30 m depth that supported fungal populations and is primarily driven by heterotrophy and aerobic respiration with seasonal rainfall. Underlying this section (30–43 m) was a zone of sulfide minerals in which iron and sulfur oxidation occurs under aerobic conditions and from a microbial standpoint is dominated by aerobic iron and sulfur oxidizers. Acidic solutions are responsible for dissolution of silicates and carbonates, the latter potentially providing CO2 to the subsurface microbial community. Beneath this zone is an anaerobic zone that contained organisms inferred to be carrying out anaerobic iron and sulfur oxidation with sulfate-reducers potentially producing H2 S and thence pyrite by reaction with host rocks. The oxygen-dependent processes in the near-surface environment were hypothesized to be partly responsible for driving oxygen depletion in the deeper regions. Both methanogenesis and sulfate reduction were inferred based on gas measurements showing the presence of methane and acidic conditions suggesting hydrogen production in the subsurface. The MARTE group suggested that if acidic Martian fluids reached the subsurface through fractures, oxidants would have been depleted in rock weathering reactions, eventually delivering ferrous and sulfide compounds into the deeper subsurface. In this way, it is hypothesized that the subsurface of Mars should show a zonation with ferric compounds dominating the surface environment and sulfides at depth, a hypothesis which is potentially testable with Martian drilling. Like all analogue environments, the Rio Tinto hosts environmental conditions that in many respects are very different from Mars. The temperatures experienced there are higher than those experienced on Mars in the present and probably in its past. The surface environment and the waters leaching into the formation have oxidation rates much higher than would have occurred on Mars on account of the oxidizing terrestrial atmosphere, and the tectonic conditions are very different to those on Mars where tectonism does not occur. Despite these caveats, the hydrothermal conditions and interactions between oxidative fluids and deep subsurface environments yields

242 | 11 The subsurface habitability of terrestrial rocky planets: Mars insights into the potential fate of Martian fluids and redox couples that might have been supported in acidic regions of Mars. The presence of sulfidic, acidic deep groundwaters, although potentially conducive to life, would not necessarily be a suitable environment for the preservation of biomarkers [56]. Hydrogen peroxide and other reactive oxygen species such as hydroxyl radicals can be formed through the interaction of pyrite and anoxic water. Acidic and oxidizing water, produced during subsurface rock-water interactions would have been deleterious to preserved signatures of life, particularly subsurface filamentous fabrics [119], and may also be detrimental to the preservation of ancient organic records from meteoritic in-fall.

11.4 Impact craters and deep subsurface habitability One significant factor in shaping Martian subsurface geochemical conditions and potential habitats that the subsurface might host are impact craters. Over 42,000 impact craters on Mars with a diameter greater than 5 km have been identified [27]. Largely eroded or subducted on the Earth, these features are much more commonly expressed on the surface of Mars and therefore constitute an important geological influence on subsurface environments. Many of these features were emplaced during the Noachian, when impact cratering rates were higher. Impacts would have fragmented the surface volcanic rock environment, producing a thick, but porous regolith through which water could have flowed. The presence of phyllosilicates [43, 52] and hydrated minerals associated with the uplift regions of Martian complex craters [42, 50] is evidence for this early period of crater-water interaction, with its implications for active subsurface geochemistry. During an impact, kinetic energy is transformed primarily into heat energy, resulting in the formation of hydrothermal systems. These systems, in which liquid water is circulated within impact crater faults, fractures and tubes, have a longevity dependent on the scale of the impact event and the target material [120]. Very large impacts of several hundred kilometers in diameter or greater, which would include Martian basin-forming events, would have lifetimes exceeding a million years. Impact events fracture rocks and increase bulk porosity of target materials with biological consequences [121]. It is known that impact-altered rocks can be up to 30% lower density than unaltered rocks because of these processes [122–125]. One obvious consequence of these changes is to increase the permeability of rocks to fluid flow and thus potentially enhance the availability of nutrients and redox couples in the subsurface. Several craters have been shown to exhibit enhanced fracturing in the subsurface near the point of impact, associated with the high shock pressures and temperatures generated during impact. For example, the Elgygytgyn crater, Russia (diameter 18 km; age 3.5 Ma), has been shown to have over two times the fracture density in the center of

11.5 The near-subsurface habitability of present and recent Mars – an empirical example |

243

the crater compared to several radii distance from the crater [126]. In the Siljan impact structure, Sweden (diameter ∼ 52 km; age ∼ 377 Ma), conductivity has been shown to diminish from the center of the crater outwards, interpreted as reflecting the existence of saline fluids in the heavily fractured center of the subsurface of the crater [123]. The first attempt to directly study the influence of an impact on the deep subsurface biosphere was made in the Chesapeake Bay Impact Structure (diameter ∼ 85 km, age 35.3 Ma) [127–129]. A composite core of 1.76 km depth was acquired with strict contamination control about 9 km from the center of the crater. As with other deep subsurface sites, microbial enumerations displayed a logarithmic decline into the subsurface with a high degree of scatter attributed to the changing lithologies into the crater, in contrast to the relatively uniform nature of marine sediments. Of particular interest was a deep region of the crater exhibiting high microbial abundance (between 106 and 107 organisms/g), which correlated with a region of porous impact suevite at greater than 1500 m depth (󳶳 Fig. 11.9). The enhanced microbial abundance at depth in the Chesapeake crater was attributed to impact-induced fracturing and enhancements in porosity. Mineral evidence at this depth in the form of melt phases show that it was heated to well above the upper temperature limit of life (> 350 °C) during and after impact in a hydrothermal system. Colonization has occurred since the impact. Active biogeochemical cycling is inferred in this region of the crater by the high ferrous iron concentrations (up to 9 μmol/g) and the presence of DNA which yielded a positive Polymerase Chain reaction (PCR) response to primers against the genus Geobacter, suggesting active iron reduction. When applied to considerations of the Martian subsurface, these data, taken alongside data from other craters that exhibit enhanced permeability, fracturing and/or fluid flow in the subsurface of craters, show that impact craters are likely to be places with improved availability of redox couples, nutrients and microbial living space compared to outlying regions. The extensive number of craters on Mars suggests that these are promising targets for subsurface exploration.

11.5 The near-subsurface habitability of present and recent Mars – an empirical example We do not have direct chemical measurements and samples from the deep subsurface of Mars (hundreds to meters to kilometers depth), other than orbital inferences described earlier. Although the Viking landers of the mid-1970s searched for life in the shallow subsurface, NASA’s Phoenix lander sampled the shallow subsurface environment (3 cm depth) with a comprehensive suite of chemical instruments that allowed for the most comprehensive assessment of a subsurface Martian environment for habitability at that time [95]. Phoenix landed at 68.22°N, 234.25°E (aerocentric) and an elevation of −4.1 km against the reference datum in 2008.

244 | 11 The subsurface habitability of terrestrial rocky planets: Mars

Fig. 11.9: Microbial enumerations (log abundance per gram dry weight) through the Chesapeake Bay impact structure. The stratigraphy, TDS concentrations and temperature profile (to 1100 m) is shown (modified from [127]).

11.6 Uninhabited, but habitable subsurface environments?

| 245

At the landing site, water ice was encountered at 3 cm depth (󳶳 Fig. 11.8). Observed as a pure white substance that partially sublimated after four days, it was inferred to be segregated ice, potentially associated with an ice wedge, in a similar form to ice wedges observed in polygonal permafrost terrain on the Earth. Some of the ice at the site had a darker coloration and was composed of ∼ 30% (wt) ice mixed with soil and was inferred to be soil pore ice [130]. Soil grain morphologies did not provide unequivocal evidence of liquid water, but dielectric measurements suggested the presence of liquid water films on soil grains at night. The lack of conductivity in the soil at night suggests that the water does not move. From a habitability point of view, this would suggest that the environment is likely to be geochemically run-down and lacking in nutrient turnover. The presence of perchlorates at ∼ 0.6% wt suggests the possibility of liquid water at or below the triple point on account of freezing point depression [130]. At 20 cm depth the temperature is estimated to be 225 K. Freezing point depression would be required to allow for the presence of liquid water at this temperature. However, given that replicating microbial life has not been convincingly demonstrated below ∼ 15 °C [131], it is not clear that the presence of liquid water at these temperatures would make such an environment habitable. Despite this conclusion, at periods of higher obliquity, up to 50°, which would have occurred during the last 5 Ma, insolation can be as high as 2.5 times the present value, which would potentially create warmer near-surface microenvironments above 273 K for several days each year [132]. Other environmental factors suggest that some additional requirements for habitability are met. The pH of the soil was measured to be buffered at 7.7 to 7.9 [112], even when acid was added. The presence of CHNOPS was assessed similar to the discussion above and found to be favorable. If fixed N was produced in the surface in impact events or atmospheric processes, then the near-surface environment at the Phoenix Landing Site may be more favorable than the deep subsurface. Many questions remain about this site, in particular the source of fixed N (the instruments on Phoenix were not able to deconvolve perchlorate from nitrogen compounds, but the anionic part of the charge balance could be accounted for by measured perchlorate and sulfate levels). However, the Phoenix measurements were significant in providing the first empirical assessment of habitability in a subsurface site, even if only to centimeter depths.

11.6 Uninhabited, but habitable subsurface environments? Despite all the considerations presented in this chapter, the possibility remains that even if the Martian subsurface is habitable, it might be devoid of life [16]. A habitable, but uninhabited Martian subsurface could occur on a completely uninhabited Mars as a result of either a lack of origin of life on Mars or the lack of a successful cross-inoculation of life from the Earth.

246 | 11 The subsurface habitability of terrestrial rocky planets: Mars At the current time, the factors and prerequisites necessary for an origin of life are not sufficiently constrained to be able to quantify the probability that life originated on Mars. If certain environmental conditions had to be met (for example, the presence of deep-sea hydrothermal vents, beaches or other specific environments for chemical complexification during an origin of life), and if Mars lacked those environments, then an origin of life could have been frustrated despite the planet having local environments that are conducive to microbial growth. It is not necessarily the case that habitable conditions for organisms are co-terminus with the presence of conditions required for life to originate, if the physico-chemical conditions required for an origin of life are highly constrained. The possibility of life being transferred from Earth to Mars is amenable to experimental testing. Microorganisms have been shown to survive in impact shock experiments to shock pressures associated with reaching escape velocity from Earth [133– 135] and to survive many years in a desiccated state in space, so long as they are protected from UV irradiation [136]. Temperatures within meteorites can be low enough during atmospheric transit to allow for the survival of organisms [137, 138]. Certainly, rocks have been transferred between the two planets [139, 140]. Nevertheless, despite this impressive evidence, there is no empirical evidence that life has been transferred from Earth to Mars, or that even if it has, it has achieved a successful inoculation of the Martian deep subsurface. An alternative scenario is that Mars is inhabited, but lack of connectivity between deep subsurface habitats means that a newly formed habitable environment remains lifeless. 󳶳 Figure 11.10 shows a hypothetical Amazonian or Hesperian scenario for an impact-generated habitat, formed by the production of subsurface liquid water from melted permafrost [16]. The lack of a substantial surface and near-surface hydrological cycle, an inclement atmosphere, and potentially a permafrost barrier to other hab-

Fig. 11.10: A plausible scenario for the formation of an uninhabited habitat on Mars in an impact crater lake (modified from [16]).

11.7 Ten testable hypotheses on habitability of the Martian subsurface | 247

itable (and inhabited regions) in subpermafrost groundwater would contribute to a failure of cross-inoculation. This would leave an impact-generated subsurface body of water isolated and sterile. These scenarios underline the variety of ways in which habitable, but uninhabited Martian subsurface environments could exist. Importantly, the hypothesis that they exist is experimentally testable. The significance of these environments is that they would provide us with insights into the geochemical processes in the subsurface of ancient and recent Mars, and potentially early Earth, without the confounding effects of biological perturbations [16]. As most of the ancient subsurface of the Earth is now subducted and destroyed, uninhabited Noachian habitats could yield important insights into prebiotic environments on the early Earth that eventually gave rise to and sustained life. We might even learn something about how these prebiotic geochemical conditions produced the chemical conditions for the emergence of life. We would also learn about the extent of geochemical cycles without life and so be better able to quantify the effect of life on planetary crustal processes and geochemical cycles.

11.7 Ten testable hypotheses on habitability of the Martian subsurface The information that exists on the Martian subsurface, from radar and surface mapping and measurements, coupled with our geochemical and biological understanding of subsurface Mars analogue environments on the Earth allows us to suggest plausible hypotheses to be tested concerning Martian subsurface habitability. Here I list just ten tantalizing (and not mutually exclusive) hypotheses based on the foregoing discussions and I provide a paragraph summary of each.

Hypothesis 1: Regions of the subsurface of Mars are or were habitable and inhabited. This hypothesis is the most important deep subsurface hypothesis. Although the ancient, more water-rich clay environments revealed in ancient Noachian terrain have many of the characteristics and requirements to meet habitability criteria, this hypothesis seeks to address whether these environments ever contained life.

Hypothesis 2: Subsurface ice-rich regions are or have been habitable. This hypothesis is a specific subhypothesis of Hypothesis 1 and recognizes that obliquity changes make icy-regions under the surface and in the near-surface environment (for example, the polar regions and locations such as the Phoenix lander site [95]) po-

248 | 11 The subsurface habitability of terrestrial rocky planets: Mars tential locations of habitable environments in the recent geological history of Mars. These regions deserve special attention in the search for extant life.

Hypothesis 3: Regions of the subsurface of Mars are or were habitable, but uninhabited. This hypothesis seeks to find out whether deep subsurface environments could be or have been habitable, but contained no life [16]. It is not contradictory to Hypothesis 1 and 2 in that if Mars is inhabited, uninhabited habitats in the subsurface could still exist in localized regions.

Hypothesis 4: Deep subsurface ices preserve a record of Martian organic and biological processes. Related to Hypotheses 1 and 2 is the possibility that deep ices preserve a record of past or extant organics inventory or life that is frozen [141]. This hypothesis recognizes the high potential for icy regions to once have been habitable when the water was liquid. Regardless of whether they preserve life, icy regions might also preserve a record of past organic meteoritic in-fall, of astrobiological interest in understanding the in-flux of organics into the early Solar System, the past habitability of the Martian subsurface environment and its organic inventory.

Hypothesis 5: Mineral precipitates in some regions of the Martian subsurface host a record of Martian organics or life. Related to Hypothesis 4 is the hypothesis that mineral veins in rocks from the Noachian onwards might preserve signatures of organics or life in the deep subsurface, particularly where liquid water flowed through deformation bands, fractures or pore spaces in ancient ultramafic rocks as they weathered through to clays, producing secondary minerals in which organics and life is trapped and fossilized.

Hypothesis 6: Habitability in the subsurface is heterogeneous from micron to kilometer scales. Geochemical conditions in the subsurface clearly varied from kilometer to subkilometer scales as observed from orbit in ancient Noachian terrains [49]. Conditions also vary on micron scales, as observed, for instance, in the variation in Martian soil grains [95]. This hypothesis addresses the question of whether and how habitability varies spatially in the subsurface of Mars, horizontally and vertically.

11.7 Ten testable hypotheses on habitability of the Martian subsurface |

249

Hypothesis 7: The fractured regions of impact craters provide regions of enhanced habitability on Mars. This hypothesis addresses a specific subquestion raised by Hypothesis 6. In attempting to understand the spatial distribution of habitability better, this hypothesis seeks to understand if deeply fractured impact crater environments might have hosted geochemical conditions more habitable than other locations (regardless of whether they did in fact support life). This hypothesis is driven by the recognition that impact-induced fracturing can enhance fluid flow, weathering and produce deep microbial habitats, potentially associated with hydrothermal environments, and that impact events have been an important geological process in shaping Martian subsurface conditions.

Hypothesis 8: The Martian deep subsurface has sustained the most habitable conditions on the planet since Noachian times. In addition to understanding how habitable conditions vary spatially, this hypothesis investigates the temporal changes in habitability. It is related to Hypothesis 1, but specifically focuses on the temporal sequence of habitability in the deep subsurface. The hypothesis is driven by the recognition that ancient Noachian terrains show the greatest direct evidence for sustained deep subsurface water flow. If deep groundwaters persist on Mars today under the cryosphere, then the Martian deep subsurface may be the only location to have hosted persistently habitable or near-habitable conditions throughout Martian history in contrast to changing and localised habitability or near-habitability on the surface. Clearly the answer to this hypothesis directly influences the potential importance of the deep subsurface as a concern in planetary protection.

Hypothesis 9: The Hesperian and Amazonian were periods with uniquely defined habitability on the surface, but also in the upper subsurface. This hypothesis recognizes that as water available at the surface decreased towards the end of the Noachian and transitioned to acidic aqueous conditions and then eventually into dry Amazonian conditions, this would have generated unique geochemical vertical profiles through the subsurface of Mars. Thus, that at least in the top few hundred meters, the Hesperian and Amazonian might be defined by discrete geochemical periods, not just in the surface environment, but in the subsurface as well, with its implications for habitability. This hypothesis is linked to Hypothesis 6 in that it also considers spatial changes in habitability (for example, changes in depth profiles through time).

250 | 11 The subsurface habitability of terrestrial rocky planets: Mars Hypothesis 10: The Martian subsurface will become more habitable in the future. Obliquity changes in the future will create conditions where near-surface liquid water is likely to be stable and may be stable in the subsurface. This hypothesis addresses the question of where and when the Martian subsurface environment will become more habitable on account of these changes. Finally, this hypothesis also includes the potential for human impact on the Martian deep subsurface. By terraforming Mars [142, 143], the cryosphere might be partially melted, generating subsurface liquid water with its implications for deep subsurface habitability. Although much emphasis has been placed on the introduction of photosynthetic organisms and surface biota on Mars [144], the crucial role that deep biosphere organisms play in global biogeochemical cycles on Earth suggests that the jump-starting of a Martian deep biosphere (speculatively either with an indigenous biota, or a seeded one) by enhancing deep-water circulation might be a first step in the long-term engineering of the Martian environment.

11.8 Sampling the subsurface of Mars How are these hypotheses to be tested? There are at least seven ways in which the subsurface might be investigated to gather data on geochemical conditions and habitability and to address the hypotheses listed in the previous section. This chapter is not a technical review, therefore I briefly summarize these approaches. – Vertical drilling. Deep drilling of Mars has been proposed [145]. In particular, drilling into ice-rich regions of ancient ice in the southern highlands has been proposed as means to address Hypothesis 4 [141, 146]. As with efforts on the Earth, drilling into any subsurface environments on Mars, given the spatial heterogeneity of deposits, would allow for Hypothesis 6 (investigating spatial heterogeneity in habitability) to be addressed on a planetary scale. – Lateral drilling. Another approach to acquiring pristine samples in the subsurface is to drill laterally into existing subsurface excavations, for example the walls of impact craters, canyon systems, outflow channels, polar layered terrains, caves, lava tubes and any excavation, re-entrant or cavern that provides a natural access into the subsurface. – Sampling excavated material. Material excavated by impact can be used to investigate the past subsurface geology, geochemistry and potential habitability of the Martian subsurface. Central uplifts in complex impact craters have already been used to directly characterize the deep subsurface geology on Mars [45, 50] and the morphology of crater ejecta blankets has been used to infer the depth of subsurface volatiles [27]. Given the known relationships between crater diameters and excavation depth, craters provide a way to assess the Martian subsurface at defined depths. Impact ejecta material has been suggested as a way to assay the subsurface of Mars for habitable conditions and biosignatures [147] with impacts

11.9 Conclusion

–

–

–

–

| 251

occurring into pre-existing craters as a means to search the subsurface of potential crater habitats for biosignatures. Sampling eroded material. The presence of regions of the ancient subsurface that are now exposed as result of erosion, for example deformation band clusters [49] provides surface access to investigate ancient subsurface processes. Sampling fluid flow from craters. McLaughlin Crater, a 2.2 km-deep crater located at 21.9 °N has been suggested to show evidence of ancient groundwater discharge [70]. Regions of Mars where topography, faults or other features have allowed water to be discharged from the subsurface might provide a means to sample subsurface deposits. Catastrophic outflow channels [8, 9, 77, 78] are another obvious candidate for the study of post-Noachian groundwater if indeed they have been formed from groundwater discharge. Using caves to access the deep subsurface. The Mars Odyssey Thermal Emission Imaging System (THEMIS) was first used to directly image the entrances of Martian caves [148]. These entrances, conspicuous as dark spots on orbital images, could be formed in a similar way to pit craters, with subsurface material collapsing to leave subsurface void spaces and overhanging rims. It is unknown what the entrances lead into (vertical shafts, large fractures, lava tubes or large cavernous spaces or all of these possibilities). However, these natural subsurface caverns offer access to the subsurface [15], probably to depths of several hundred meters in which vertical and lateral drilling might be used to recover samples. Caves may also contain preserved ice that would be important targets in astrobiological exploration [149]. Gases produced by a subsurface biota. This final method is not a method of rock sample acquisition, but it does offer a way to indirectly assay deep subsurface processes. The notion of detecting subsurface production of gases by a biota has been suggested. Methane is one target gas [150]. The consumption of atmospheric gases that could act as redox half reactions, such as H2 or CO, has also been suggested [151]. The presence of these gases in the Martian atmosphere, produced by photochemical pathways, has been interpreted as evidence that if there is a subsurface biota, it cannot be very active [151]. Although concentrations of biogenic gases would be low, isotopic measurements or the search for trace gases could be used to discriminate abiotic from biota production [150, 151].

11.9 Conclusion In the 1960s and for the next forty years, the surface (and by extension) the subsurface of Mars was widely regarded as a homogenous basaltic environment, with little temporal or spatial variation. However, as the means to map minerals on the surface have improved, so has this view radically changed. The surface is now known to have supported environments that range from acidic, sulfate-rich aqueous conditions to neu-

252 | 11 The subsurface habitability of terrestrial rocky planets: Mars tral and ultramafic (presumably alkaline) environments. As this picture has emerged, radar studies and the investigation of impact crater uplifts and eroded surfaces has revealed the first insights into the structure of the Martian deep subsurface. The picture that emerges is one of a subsurface environment (within the top few hundred meters) that has probably tracked changes observed in surface environments, with neutral and alkaline conditions in the Noachian, giving rise to more acidic groundwaters in the Hesperian and finally transitioning to dry, ice-dominated subsurface environments today with transient obliquity-driven water availability. However, the deep subsurface (∼kilometer depths) may have been more habitable throughout Martian history. Once linked to surface environments throughout the crust, if there is water in these environments today, it may be isolated from the surface and upper subsurface by permafrost, only being transiently connected in catastrophic events. Of all the requisites for habitability that might be most lacking in the deep subsurface, nitrogen is the greatest unknown. A variety of methods can be used in the future to directly investigate the deep subsurface of Mars and in future years these methods will allow scientists to determine whether the subsurface was habitable and whether it was inhabited. The lack of a deep subsurface record of early Earth means that Mars may be the best place for us to study the early prebiotic environment of Earth and understand the conditions that gave rise to, and eventually supported, early life.

References [1]

[2] [3] [4] [5] [6]

[7] [8] [9] [10]

Haberle RM, McKay CP, Schaeffe J, et al. On the possibility of liquid water on present-day Mars. J Geophys Res 2001;106:E10, doi:10.1029/2000JE001360, http://www.agu.org/ journals/ABS/2001/2000JE001360.shtml. Cockell CS, Catling D, Davis WL, et al. The ultraviolet environment of Mars , biological implications past, present and future. Icarus 146 (2000), 343–359. Moores JE, Smith PH, Tanner R, Schuerger AC, Venkateswaran KJ. The shielding effect of small-scale Martian surface geometry on ultraviolet flux. Icarus 192 (2007), 417–433. Kminek G, Bada JL. The effect of ionizing radiation on the preservation of amino acids on Mars. Earth Planet Sci Lett 245 (2006), 1–5. Dartnell L, Desorgher L, Ward J, Coates A. Modelling the surface and subsurface radiation environment: implications for astrobiology. Geophys Res Lett 34 (2007), L02207. Pavlov AA, Vasilyev G, Ostryakov VM, Pavlov AK, Mahaffy P. Degradation of the organic molecules in the shallow subsurface of Mars dust to irradiation by cosmic rays. Geophys Res Lett 39 (2012), L13202. Sharp RP, Malin MC. Channels on Mars. Geol Soc Am Bull 86 (1975), 593–609. Carr MH. Water on Mars. Oxford: Oxford University Press, 1986. Carr MH. Mars – a water-rich planet? Icarus 68 (1986), 187–216. Head JW, Hiesinger H, Ivanov MA, Kreslavsky MA, Pratt S, Thomson BJ. Possible ancient oceans on Mars: Evidence from Mars Orbiter Laser Altimeter data. Science 296 (1999), 2134– 2137.

References | 253

[11] [12] [13] [14] [15] [16] [17] [18]

[19] [20] [21]

[22] [23] [24] [25] [26] [27] [28] [29] [30]

[31] [32]

Clifford SM, Parker TJ. The evolution of the Martian hydrosphere: Implications for the fate of a primordial ocean and the current state of the Northern Plains. Icarus 154 (2001), 40–79. Cabrol NA, Grin EA. The evolution of lacustrine environments on Mars: (Is Mars only hydrologically dormant?) Icarus 149 (2001), 291–328. Cabrol NA, Grin EA. Overview on the Formation of Paleolakes and Ponds in Impact Craters on Mars. Global Planet Change 35 (2002), 199–219. Schwenzer SP, Kring DA. Impact-generated hydrothermal systems: capable of forming phyllosilicates on Noachian Mars. Geology 37 (2009), 1091–1094. Boston PJ, Ivanov MV, McKay CP. On the possibility of chemosynthetic ecosystems in subsurface habitats on Mars. Icarus 95 (1992), 300–308. Cockell CS, Balme M, Bridges JC, Davila A, Schwenzer SP. Uninhabited habitats on Mars. Icarus 217 (2012), 184–193. Abbey W, Salas E, Bhartia R, Beegle LW. The Mojave vadose zone: a subsurface biosphere analogue for Mars. Astrobiology 13 (2005), 637–646. Szponar N, Brazelton WL, Screnk MO, Bower DM, Steele A, Morrill PL. Geochemistry of a continental site of serpentinization, the Tablelands Ophiolite, Gros Morne National Park: A Mars analogue. Icarus 224 (2013), 286–296. Christensen PR. Water at the poles and in permafrost regions of Mars. Elements 2 (2006), 151–155. Bandfield JL. High-resolution subsurface water-ice distributions on Mars. Nature 447 (2007), 64–68. Feldman WC, Pathare S, Maurice TH, et al. Mars Odyssey Neutron data: 2. Search for buried excess water ice deposits at nonpolar latitudes on Mars. J Geophys Res 116 (2011) DOI:10.1029/2011JE003806. Picardi G, Plaut JJ, Bicarri D, et al. Radar soundings of the subsurface of Mars. Science 310 (2005), 1925–1928. Watters TR, Leuschen CJ, Plaut JJ, et al. MARSIS radar sounder evidence of buried basins in the northern lowlands of Mars. Nature 444 (2006), 905–908. Holt JW, Safaeinili A, Plaut JJ, et al. Radar sounding evidence for buried glaciers in the southern mid-latitudes of Mars. Science 322 (2008), 1235–1238. Head JW, Mustard JF, Kreslavsky MA, Milliken RE, Marchant DR. Recent ice ages on Mars. Nature 426 (2003), 797–802. Urbach ER, Stepinski TF. Automatic detection of subkilometer craters in high resolution planetary images. Planet Space Sci 57 (2009), 880–887. Barlow NG. What we know about Mars from its impact craters. Geol Society Amer Bull 122 (2010), 644–657. Zuber MT, Smith DE, Solomon SC, et al. Observations of the north polar regions of Mars from the Mars Orbiter Laser Altimeter (MOLA). Science 282 (1998), 2053–2060. Squyres SW, Carr MH. Geomorphic evidence for the distribution of ground ice on Mars. Science 231 (1986), 249. Mangold N, Maurice S, Feldman WC, Costard F, Forget F. Spatial relationships between patterned ground and ground ice detected by the Neutron Spectrometer on Mars. J Geophys Res 109 (2004) doi: 10.1029/2004JE002235. Mangold N. High latitude patterned ground on Mars: classification, distribution and climatic control. Icarus 174 (2005), 336–359. Seibert NM, Kargel JS. Small-scale Martian polygonal terrain: implications for liquid surface water. Geophys Res Lett 28 (2001) DOI:10.1029/2000GL012093.

254 | 11 The subsurface habitability of terrestrial rocky planets: Mars [33] [34]

[35] [36] [37] [38]

[39] [40] [41] [42]

[43]

[44]

[45] [46] [47]

[48]

[49] [50] [51] [52]

Mangold N. Geomorphic analysis of lobate debris aprons on Mars at MOC scale: Evidence for ice sublimation initiated by fractures. J Geophys Res 108 (2003), doi: 10.1029/2002JE01885. Levy JS, Head JW, Marchant DR. Concentric crater fill in the northern mid-latitudes of Mars: Formation processes and relationships to similar landforms of glacial origin. Icarus 209 (2010), 390–404. Dundas CM, McEwen AS. An assessment of evidence for pingos on Mars using HiRISE. Icarus 205 (2010), 244–258. Smith P, Tamppari LK, Arvidosn RE, et al. H2 O at the Phoenix landing site. Science 325 (2009), 58–61. Bryne S, Dundas CM, Kennedy MR, et al. Distribution of mid-latitude ground ice on Mars from new impact craters. Science 325 (2009), 1674–1676. Ehlmann B, Mustard JF, Swayze GA, et al. Identification of hydrated silicate minerals on Mars using MRO-CRISM: geologic context near Nili Fossae and implications for aqueous alteration. J Geophys Res 144 (2009), E00D08. Wray JJ, B. L. Ehlmann BL. Geology of possible Martian methane source regions. Planet Space Sci 59 (2011), 196–202. Ehlmann B, Mustard JF, Murchie SL. Geologic setting of serpentine deposits on Mars. Geophys Res Lett 37 (2010), LO6201. Ehlmann BL, Mustard JF, Murchie SL, et al. Subsurface water and clay mineral formation during the early history of Mars. Nature 479 (2011), 53–60. Quantin C, Flahut J, Clenet H, Allemand P, Thomas P. Composition and structures of the subsurface in the vicinity of Valles Marineris as revealed by central uplifts of impact craters. Icarus 221 (2012), 436–452. Mustard JF, Ehlmann BL, Murchie SL, et al. Composition, morphology and stratigraphy of Noachian crust around the Isidis Basin. J Geophys Res 114 (2009), doi: 101029/2009/E002249. Poulet F, Mangold N, Platevoet B, et al. Quantitative compositional analysis of Martian mafic regions using the MEx/OMEGA reflectance data. 2: Petrological implications. Icarus 201 (2009), 84–101. Michalski JR, Niles PB. Deep crustal carbonate rocks exposed by meteor impact on Mars Nature Geosciences 3 (2010), 751–755. Rogers AD. Crustal compositions exposed by impact craters in the Tyrrhena Terra region of Mars: considerations for Noachian environments. Earth Planet Sci Lett 301 (2011), 353–364. Grotzinger JP, Arvidson RE, Calvin W, et al. Stratigraphy and sedimentology of a dry to wet eolian depositional system, Burns formation, Meridiani Planum, Mars. Earth Planet Sci Lett 240 (2005), 11–72. McLennan, SM, Grotzinger JP. The sedimentary rock cycle of Mars. In: Bell III JF, ed. The Martian Surface: Composition, Mineralogy, and Physical Properties. Cambridge: Cambridge University Press, 2008; 541–577. Okubo CH, Schultz RA, Chan MA, et al. Deformation band clusters on Mars and implications for subsurface fluid flow. Geol Soc America Bull 121 (2009), 474–482. Ehlmann BL, Berger G, Mangold N, et al. Geochemical consequences of widespread clay mineral formation in Mars’ ancient crust. Space Sci Rev 174 (2013), 329–364. Poulet F, Bibring JP, Mustard JF, et al. Phyllosilicates on Mars and implications for early Martian climate. Nature 438 (2005), 623–627. Bibring J-P, Yangevin Y, Mustard JF, et al. Global mineralogical and aqueous Mars history derived from OMEGA/Mars Express data. Science 312 (2006), 400–404.

References | 255

[53] [54] [55] [56]

[57] [58] [59] [60]

[61] [62] [63] [64]

[65]

[66] [67] [68] [69] [70] [71]

[72] [73] [74]

Ehlmann BL, Mustard JF, Murchie SL, et al. Orbital detection of carbonate-bearing rocks on Mars. Science 322 (2008), 1828–1832. Kurahashi-Nakamura T, Tajika E. Atmospheric collapse and transport of carbon dioxide into the subsurface on early Mars. Geophys Res Lett 33 (2006), doi: 10-1029/2006GL027170 Benner SA, Devine KG, Matveeva LN, Powell DH. The missing organic molecules on Mars. Proc Natl Acad Sci USA 97 (2000), 2425–2430. Davila AF, Fairén AG, Gago-Duport L, et al. Subsurface formation of oxidants on Mars and implications for the preservation of organic biosignatures. Earth Planet Sci Lett 272 (2008), 456–463. Steele A, McCubbin FM, Fries M, et al. A reduced organic carbon component in Martian basalts. Science 337 (2012), 212–215. Lin LH, Hall J, Lippmann-Pipke J, et al. Radiolytic H2 in continental crust: nuclear power for deep subsurface microbial communities. Geochem Geophys Geosyst 6 (2005), Q07003. Segura A, Navarro-González R. Nitrogen fixation on early Mars by volcanic lightning and other sources. Geophys Res Lett 32 (2005), doi:10-1029/2004GL021910. Summers DP, Khare B. Nitrogen fixation on early Mars and other terrestrial planets: experimental demonstration of abiotic fixation reactions to nitrite and nitrate. Astrobiology 7 (2007), 333–341. Manning CV, Zahnle KJ, McKay CP. Impact processing of nitrogen on early Mars. Icarus 199 (2009), 273–285. Boxe CS, Hand KP, Nealson KH, Ying YL, Saiz-Lopez A. An active nitrogen cycle on Mars sufficient to support a subsurface biosphere. Int J Astrobiol 11 (2012), 109–115. Ducluzeau A-L, van Lis R, Duval S, et al. Was nitric oxide the first deep electron sink? Trends Biochem Sci 34 (2008), 9–15. McGlynn IO, Fedo CM, McSween HY. Soil mineralogy at the mars Exploration rover landing sites: an assessment of the competing roles of physical sorting and chemical weathering. J Geophys Res 117 (2012), doi: 10-1029/2011JE003861. Morris RV, Kilingelhofer C, Scroder C, et al. Mössbauer mineralogy of rock, soil, and dust at Meridiani Planum, Mars: Opportunity’s journey across sulfate-rich outcrop, basaltic sand and dust, and hematite lag deposits. J Geophys Res 114 (2006), 27. Bibring J-P, Ardivson RE, Gendrin A, et al. Coupled ferric oxides and sulfates on the Martian surface. Science 317 (2007), 1206–1210. Gendrin A, Mangold N, Bibring J-P, et al. Sulfates in Martrian layered terrains: the OMEGA/Mars Express view. Science 307 (2005), 1587–1591. Langevin Y, Poulet F, Bibring J-P, Gondet B. Sulfates in the north polar region of Mars detected by OMEGA/Mars Express. Science 307 (2005), 1584–1586. Gaillard F, Michalski J, Berger G, McLennan SM, Scaillet B. Geochemical reservoirs and timing of sulfur cycling on Mars. Space Sci Rev 174 (2013), 251–300. Michalski JR, Cuadros J, Niles PB, Parnell J, Rogers AD, Wright SP. Ground water activity on Mars an implications for a deep biosphere. Nature Geoscience 6 (2013), 133–138. Tosca JN, McLennan SM, Clark BC, et al. Geochemical modeling of evaporation processes on Mars: insight from the sedimentary record at Meridiani Planum. Earth Planet Sci Lett 240 (2005), 122–148. Tosca NJ, Knoll AH, McLennan SM. Water activity and the challenge for life on early Mars. Science 320 (2008), 1204–1207. Clifford SM. Polar basal melting on Mars. J Geophys Res 92 (1987), 9135–9152. Fanale FP. Martian volatiles: Their degassing history and geochemical fate. Icarus 28 (1976), 179–202.

256 | 11 The subsurface habitability of terrestrial rocky planets: Mars [75] [76]

[77] [78] [79] [80]

[81]

[82] [83] [84] [85] [86] [87] [88] [89] [90] [91]

[92]

[93] [94] [95] [96]

Clifford SM. A model for the hydrologic and climatic behavior of water on Mars. J Geophys Res 98 (1993), 10973. Clifford SM, Lasue J, Heggy E, Boisson J, McGovern P, Max MD. Depth of the Martian cryosphere: revised estimates and implications for the existence and detection of subpermafrost water. J Geophys Res 115 (2010), DOI: 10-1029/2009JE003462. Tanaka KL. The stratigraphy of Mars. J Geophys Res 91 (1986), E139. Hartmann W, Neukum G. Cratering chronology and the evolution of Mars. Space Sci Rev 96 (2001), 165. Lasue J, Mangold N, Hauber E, et al. Quantitative assessments of the Martian hydrosphere. Space Sci Rev 174 (2013), 155–212. Burr DM, Grier JA, McEwen AS, Keszthelyi LP. Repeated aqueous flooding from the Cerberus Fossae: evidence for very recently extant, deep groundwater on Mars. Icarus 159 (2002), 53– 73. Neukum G, Basilevsky AT, Kneissl T, et al. The geologic evolution of Mars: Episodicity of resurfacing events and ages from cratering analysis of image data and correlation with radiometric ages of Martian meteorites. Earth Planet Sci Lett 294 (2010), 204–222. Schwenzer SP, Abramov O, Allen CC, et al. Puncturing Mars: How impact craters interact with the Martian cryosphere. Earth Planet Sci Lett 335 (2012), 9–17. Malin MC, Edgett KS. Evidence for recent groundwater seepage and run-off on Mars. Science 288 (2000), 2330–2335. Heldmann JL, Mellon MT. Observations on Martian gullies and constraints on potential formation mechanisms. Icarus 168 (2004), 285–304. Goldspiel JM, Squyres SW. Groundwater discharge and gully formation on Martian slopes. Icarus 211 (2011), 238–258. Dienega S, Bryne S, Bridges NT, Dundas CM, McEwan AS. Seasonality of present-day Martian dune-gully activity. Geology 38 (2010), 1047–1050. Reiss D, Erkeling G, Bauch KE, Hiesinger H. Evidence for present-day gully activity on the Russell crater dune field. Geophys Res Lett 37 (2010), DOI: 10-1029/2009GL042192. Schon SC, Head JW. Keys to gully formation processes on Mars: relation to climate cycles and sources of meltwater. Icarus 213 (2011), 428–432. McEwan AS, Ojha L, Dundas CM, et al. Seasonal flows on warm Martian slopes. Science 333 (2011), 740–743. Rennó NO, Bos BJ, Catling DC, et al. Possible physical and thermodynamical evidence for liquid water at the Phoenix landing site. J Geophys Res 114 (2009), doi:10-1029/2009JE003362. Martínez GM, Rennó NO. Water and brines on Mars: Current evidence and implications for MSL. Space Sci Rev 175 (2013), 29–51. 91. Martínez GM, Rennó NO. Water and brines on Mars: Current evidence and implications for MSL. Space Sci Rev 175 (2013), 29–51. Battler MM, Osinski GR, Banerjee NR. Mineralogy of saline perennial cold springs on Axel Heiberg Island, Nunavut, Canada and implications for spring deposits on Mars. Icarus 224 (2013), 364–381. Zorzano MP, Mateo-Martí E, Prieto-Ballesteros O, Osuna S, Renno N. Stability of liquid saline water on present day Mars. Geophys Res Lett 36 (2009), DOI: 10.1029/2009GL040315. Nunes DC, Smrekar SE, Safaeinili A, et al. (2010) Examination of gully sites on Mars with the shallow radar. J Geophys Res 115 (2010), doi:10-1029/2009JE003509. Stoker CR, Zent A, Catling DC, et al. Habitability of the Phoenix landing site. J Geophys Res 115 (2010), E00E20. Möhlmann D. Are nanometric films of liquid supercooled interfacial water bio-relevant? Cryobiology 58 (2009), 256–261.

References | 257

[97] [98] [99] [100] [101] [102]

[103] [104] [105]

[106] [107] [108] [109] [110] [111] [112] [113]

[114] [115] [116] [117] [118]

Möhlmann D. Three types of liquid water in icy surfaces of celestial bodies. Planet Space Sci 59 (2011), 1082–1086. Kereszturi A, Rivera-Valentin EG. Locations of thin water layers on present-day Mars. Icarus 221 (2012), 289–295. Ulrich M, Wagner D, Hauber E, de Vera J-P, Schirrmeister L. Habitable periglacial landscapes in Martian mid-latitudes. Icarus 219 (2012), 345–357. Mayhew LE, Ellison ET, McCollom TM, Trainor TP, Templeton AS. Hydrogen generation from low-temperature water-rock interactions. Nature Geosciences 6 (2013), 478–484. Matias PM, Pereira IAC; Soares CM, Carrondo MA. Sulfate respiration from hydrogen in Desulfovibrio bacteria: a structural biology overview. Prog Biophysics Mol Biol 89 (2005), 292–329. Moser DP, Gihring TM, Brockman FJ, Fredrickson JK, Balkwill DL, Dollhopf ME, Sherwood-Lollar B, Pratt LM, Boice E, Southam G, Wanger G, Baker BJ, Pfiffner SM, Lin L-H, Onstott TC. Desulfotomaculum and Methanobacterium spp. dominate a 4- to 50 km deep fault. Appl Env Microbiol 71 (2005), 8773–8763. Harris SH, Smith RL, Suflita JM. In situ hydrogen consumption kinetics as an indicator of subsurface microbial activity. FEMS Microbiol Ecol 60 (2007), 220–228. Lovley DR. Dissimilatory Fe(III) and Mn(IV) reduction. Microbiol Mol Biol Rev 55 (1995), 259– 287. Kotelnikova S, Pedersen K. Distribution and activity of methanogens and homoacetogens in deep granitic aquifers at Aspö Hard Rock Laboratory, Sweden. FEMS Microbiol Ecol 26 (1998), 121–124. Nixon SL, Cockell CS, Tranter M. Limitations to a microbial iron cycle on Mars. Planet Space Sci 72 (2012), 116–128. Blank JG, Green S, Blake D, et al. An alkaline spring system within the Del Puerto Ophiolite (California, USA): a Mars analog site. Planet Space Sci 57 (2009), 533–540. Okland I, Huang S, Dahle H, Thorseth LH, Pedersen RB. Low temperature alteration of serpentinized ultramafic rock and implications for microbial life. Chem Geol 318 (2012), 75–87. Cockell CS, Raven JA. Zones of photosynthetic potential on Mars and the early Earth. Icarus 2004;169,300–310. Darnell LR, Page K, Jorge-Villar SE, et al. Destruction of Raman biosignatures by ionizing radiation and the implications for life detection on Mars. Anal Bioanal Chem 403 (2012), 131–144. Kashefi K, Lovley DR. Extending the upper temperature limit for life. Science 301 (2003), 934. Hecht MH, Kouvanes SP, Quinn RC, et al. Detection of perchlorate and the soluble chemistry of Martian soil at the Phoenix Mars lander. Science 325 (2009), 64–67. Tosca NJ, McLennan SM, Dyar MD, Sklute EC, Michel FM. Fe oxidation processes at Meridiani Planum and implications for secondary Fe mineralogy on Mars. J Geophys Res 113 (2008), E05005. Clark BC, Baird AK. Is the Martian lithosphere sulfur rich? J Geophys Res 84 (1979), 8395– 8403. Clark BC, Morris RV, McLennan SM, et al. Chemistry and mineralogy of outcrops of at Meridiani Planum. Earth Planet Sci Lett 240 (2005), 73–94. King PL, McLennan SM. Sulfur on Mars. Elements 6 (2010), 107–112. Squyres SW, Arvidson RE, Bell JF, et al. Ancient impact and aqueous processes at Endeavor Crater, Mars. Science 336 (2012), 570–576. Fernández-Remolar DC, Prieto-Ballesteros O, Rodriguez N, et al. Underground habitats in the Rio Tinto Basin: A model for subsurface life habitats on Mars. Astrobiology 8 (2008), 1023– 1047.

258 | 11 The subsurface habitability of terrestrial rocky planets: Mars [119] Hofmann BA. Morphological biosignatures from subsurface environments: recognition on planetary missions. Space Sci Rev 135 (2008), 245–254. [120] Osinski GR, Tornabene LL, Banerjee NR, et al. Impact-generated hydrothermal systems on Earth and Mars. Icarus 224 (2013), 347–363. [121] Cockell CS, Lee P, Broady P, et al. Effects of asteroid and comet impacts on habitats for lithophytic organisms - A synthesis. Met Planet Sci 40 (2005), 1901–1914. [122] Pesonen LJ, Elo S, Lehtinen M, Jokinen T, Puranen R, Kivekäs L. Lake Karikkoselkä impact structure, central Finland: New geophysical and petrographic results. In: Dressler BO, Sharpton VL, eds. Large Meteorite Impacts and Planetary evolution II. Boulder, Colorando. Geol Soc America Special Paper 339 (1999), 131–147. [123] Henkel H. Geophysical aspects of meteorite impact craters in eroded shield environment, with special emphasis on electric resistivity. Tectonophysics 216 (1992), 63–89. [124] Pilkington M, Grieve RAF. The geophysical signatures of terrestrial impact craters. Rev Geophys 30 (1992), 161–181. [125] Plado J, Pesonen LJ, Elo S, Puura V, Suuroja K. Geophysical research on the Kärdla impact structure, Hiiumaa Island, Estonia. Met Planet Sci 31 (1996), 289–298. [126] Gurov YeP, Gurova YeP. Zakonomernosti rasprostraneniya razlomov vokrug meteoritnykh kraterov (na primere kratera El’gygytgyn). Laws of distribution of faults around a meteor crater; example of Elgygytgyn Crater. Doklady Akademii Nauk SSSR 269 (1983), 1150–1153. http://oh1.csa.com/ids70/p_search_form.php?field=au&query= gurov+ye+p&log=literal&SID=d80880d26742730b29ea7378e3776af5 http://oh1. csa.com/ids70/p_search_form.php?field=au&query=gurova+ye+p&log=literal&SID= d80880d26742730b29ea7378e3776af5 [127] Cockell CS, Gronstal AL, Voytek MA, et al. Microbial abundance in the deep subsurface of the Chesapeake Bay impact crater: Relationship to lithology and impact processes. Geol Soc America Special Papers 458 (2009), 941–950. [128] Cockell CS, Voytek MA, Gronstal AL, et al. Impact disruption and recovery of the deep subsurface biosphere. Astrobiology 12 (2012), 231–246. [129] Gronstal AL, Voytek MA, Kirshtein JD, von der Heyde NM, Lowit MD, Cockell CS. Contamination assessment in microbiological sampling of the Eyreville core, Chesapeake Bay impact structure. Geol Soc America Special Papers 458 (2009), 951–964. [130] Cull S, Arvidson RE, Mellon MT, Skemer P, Shaw A, Morris RV. Compositions of subsurface ices at the Mars Phoenix landing site. J Geophys Res 37 (2010), L24203. [131] Clarke A, Morris GJ, Fonseca F, Murray BJ, Acton E, Price HC A low temperature limit for life on Earth. PLoS ONE 8 No. 6, (2013) e66207. [132] Haberle RM, Murphy JR, Scaeffer J. Orbital change experiments with a Mars general circulation model. Icarus 161 (2003), 66–89. [133] Horneck G, Rettberg P, Reitz G, et al. Protection of bacterial spores in space, a contribution to the discussion on panspermia. Origin Life Evol Biosph 31 (2001), 527–547. [134] Burchell MJ, Mann JR, Bunch AW, Brando PFB. Survivability of bacteria in hypervelocity impact. Icarus 154 (2001), 545–547. [135] Burchell MJ, Mann JR, Bunch AW. Survival of bacteria and spores under extreme shock pressures. Mon Not Royal Astron Soc 352 (2004), 1273–1278. [136] Horneck G, Bücker H, Reitz G. Long-term survival of bacterial spores in space. Adv. Space Res 14 (1994), 41–45. [137] Weiss BP, Kirschvink JL, Baudenbacher FJ, et al. A low temperature transfer of ALH84001 from Mars to Earth. Science 290 (2000), 791–795.

References |

259

[138] Fajardo-Cavaros P, Link L, Melosh HJ, Nicholson WL. Bacillus subtilis spores on artificial meteorites survive hypervelocity atmospheric entry: implications for lithopanspermia. Astrobiology 5 (2005), 726–736. [139] Gladman BJ, Burns JA, Duncan. M, Lee P, Levison HF. The exchange of impact ejecta between the terrestrial planets. Science 271 (1996), 1387–1392. [140] Mileikowsky C Cucinotta F, Wilson JW, et al. Natural transfer of viable microbes in space Part I: From Mars to Earth and Earth to Mars. Icarus 145 (2000), 391–427. [141] Smith HD, McKay CP. Drilling in ancient permafrost on Mars for evidence of a second genesis of life. Planet Space Sci 53 (2005), 1302–1308. [142] Fogg MJ. Terraforming: Engineering Planetary Environments. Warrendale, PA: SAE International, 1995. [143] McKay CP, Toon OB, Kasting JF. Making Mars habitable. Nature 352 (1991), 489–496. [144] Friedmann EI, Ocampo-Friedmann R. A primitive cyanobacterium as pioneer microorganism for terraforming Mars. Adv Space Res 15 (1995), 243–246. [145] Zacny K, Bar-Cohen Y, Brennan M, et al. Drilling systems for extraterrestrial subsurface exploration. Astrobiology 8 (2008), 665–706. [146] McKay CP, Stoker CR, Glass BJ. The Icebreaker life mission to Mars: A search for biomolecular evidence of life. Astrobiology 13 (2013), 334–353. [147] Cockell CS, Barlow NG. Impact excavation and the search for subsurface life on Mars. Icarus 155 (2002), 340–349. [148] Cushing GE, Titus TN, Wynne JJ, Christensen PR. THEMIS observes possible cave skylights on Mars. Geophys Res Lett 34 (2007), L17201. [149] Williams KE, McKay CP, Toon, OB, Head JW. Do ice caves exist on Mars? Icarus 209 (2010), 358–368. [150] Summers ME, Lieb J, Chapman E, Yung YL. Atmospheric biomarkers of subsurface life on Mars. Geophys Res Lett 29 (2002), 2171. [151] Weiss BP, Yung YL, Nealson KH. Atmospheric energyfor subsurface life on Mars? Proc Natl Acad Sci USA 97 (2000), 1395–1399.

Rolando di Primio

12 Assessing biosphere-geosphere interactions over geologic time scales: insights from Basin Modeling 12.1 Introduction The term Deep Biosphere was coined by Thomas Gold [1] and refers to subsurface life habitats decoupled from photosynthesis, populated by microbial life. The extent of the deep biosphere is loosely defined to range between 1 m depth (or more depending on terrain) to depths at which the maximum temperature for microbial life is reached. This temperature window in which subsurface life can exist ranges from subzero at permafrost environments to the upper temperature limit of life, which is currently 122 °C [2]. Assuming such rough boundary constraints and in combination with estimates on the number of prokaryotes encountered in different habitats first mass balances indicated that subsurface bacteria and archaea comprise 35–47% of the Earths’ total biomass [3], albeit recent studies indicate the likelihood of these estimates being too high [4]. Independent of the extent of the deep biosphere, microbial life depends on the presence of electron donors and acceptors to gain energy. In the anaerobic environments of subsurface realms, organic substrates are consumed using terminal electron acceptors (e.g. sulfate, nitrate, iron, manganese or carbon dioxide) other than the ubiquitous oxygen in aerobic surface environments. In sedimentary environments, the main source of carbon comes from buried organic matter, which becomes more recalcitrant with increasing depth of burial. Microbial cell counts in subsurface sediments in general show a logarithmic decrease with depth [5], resulting from the decrease in available organic carbon and nutrients in progressively older sediments [6]. However, many situations are known in which microbial life is sustained by organic compounds produced by the thermal conversion of recalcitrant organic carbon at greater depths, beyond the window of microbial life, which have migrated to sites of potential metabolization. Such situations include microbial consortia feeding off thermogenic gas hydrate accumulations [7], petroleum reservoirs [8] or surface expressions of petroleum leakage [9]. Also, an in situ-direct coupling of abiotic substrate generation from buried organic matter and microbial utilization has been postulated to occur in at least one special geologic setting [10], indicating that the deep biosphere and the abiotic geosphere overlap to some extent (see also [5]). The bulk of deep biosphere research to date has focused on understanding the processes, controls and limits of the deep biosphere as we see it today. In view of the fact that surface organic carbon production, deposition, preservation and burial de-

262 | 12 Basin Modeling and the Deep Biosphere fine the occurrence of the main microbial substrate, and that the window of possible subsurface life is controlled by temperature evolution, it becomes evident that a major control on the deep biosphere stems from the geologic evolution of the Earth’s crust. As temperature history and sedimentary burial rates vary drastically during the evolution of a sedimentary basin we can postulate that deep biosphere extent and carbon turnover will also vary significantly both through geologic time and among different basins. In order to investigate such processes, the geologic history of the subsurface needs to be taken into account and linked to the existing understanding of deep biosphere processes. The main focus of this chapter is to show how such links are currently being forged, and what implications can be derived from this type of research.

12.2 Basin Modeling The evolution of a sedimentary basin is the result of a complex combination of physical, chemical and biological processes occurring both sequentially and simultaneously over geologic periods of time. Such processes can be studied using different methods, which are generally subdivided into three main groups: observations, experiments and numerical simulations. In geosciences, observational and experimental methods have dominated, while numerical simulations have only become standard in the last few decades. Numerical simulations of basin evolution, commonly termed “Basin Modeling”, attempt to integrate physical and chemical process understanding in the framework of the temporal evolution of a sedimentary basin. Here it should be clarified that Basin Modeling is a term not only used within the context of the analysis of the sedimentary fill of a basin, but also for the modeling of crustal, lithospheric and mantle process. Hantschel and Kauerauf [11] proposed to term the latter approaches as “Crustal Model” and limit the term Basin Modeling to the study of processes in sediments. The research on Basin Modeling methods was strongly driven by the petroleum industry, and has led to the development of sophisticated software packages which allow to address e.g. sediment compaction, heat and fluid transport, petroleum generation and multiphase fluid flow. With the focus on petroleum, Basin Modeling implies the numerical simulation of physical and chemical processes in sedimentary basins and its application to describe the petroleum system evolution (reservoir, nonreservoir and source rocks). In this case, the term Petroleum System Modeling is also often applied. For an overview of the numerical methods, as well as the physics and chemistry behind Petroleum System Modeling, the interested reader is referred to the textbook by Hantschel and Kauerauf [11]. The usual workflow in Petroleum System Modeling starts off with a geologic model which includes the definition of areal extent, structure and morphology of the geologic sequences based on available data (e.g. seismic data, outcrops, etc.), but also all assumptions of the dominant processes taking place during the geologic evolution of the study area. This model is usually termed the Con-

12.2 Basin Modeling

| 263

Fig. 12.1: Schematic workflow in Petroleum System Modeling.

ceptual Model (󳶳 Fig. 12.1). The Conceptual Model needs to be populated with data describing the properties of the individual units, the thus defined starting model is then discretized for numerical simulation, resulting in a number of cells in the model, each of which is characterized by constant properties. The simulation is performed based on physical and chemical process descriptions (here also biological processes can be described). The simulation results include e.g. the evolution of basin burial (burial history) or the thermal evolution of the basin. Such predictions need then to be compared to the natural system, i.e. measurements of temperature, maturity, etc. and the model is thus calibrated through iteration of the starting conditions or re-definition of the conceptual model. Such Petroleum System Models are not deterministic! Results, even if they reproduce observations in the natural system, need to be assessed regarding the sensitivity of the model predictions to input variables. Many geologic pathways lead to a good calibration, variability of predictions of calibrated models are investigated in the sensitivity analysis (󳶳 Fig. 12.1). The success of petroleum system models in correctly reproducing sedimentary basin evolution, to the extent that model results are today commonly used by the petroleum industry to identify and qualitatively compare potential exploration prospects, indicates that they have reached a level of sophistication, which may be of interest to other fields of research. Here, specifically, the deep biosphere comes to mind. In view of the habitats of microbial communities in the subsurface it becomes evident that plate-tectonics, basin and even petroleum-system evolution are important factors that affect and influence deep subsurface biological systems (see also [8]).

264 | 12 Basin Modeling and the Deep Biosphere Microbial life in the subsurface is constrained to porous systems where nutrients and metabolites are transported by diffusion through the water phase. As sediments are deposited and buried porosity decreases, the pore network tortuosity increases. Temperature and pressure change systematically, depending in part also on the evolution of sediment properties such as porosity and permeability. Also, the evolution of sedimentary organic matter, especially the catagenetic activation of recalcitrant organic matter (e.g. [12]), can be a critical factor in defining the processes occurring and their extent in the deep biosphere. All such changes occurring over geologic time can be addressed and reconstructed using basin models using the appropriate physical, chemical and biologic process descriptions. Thus, the integration of geosphere-scale physical and chemical modeling with biological process descriptions may provide a way forward for deep biosphere research. While this theme is of obvious scientific interest, it is only those aspects, which aim at commercial interests, that have seen an application to date. The very few published examples have focused almost exclusively on biogenic gas formation and the biodegradation of petroleum fluids. These topics are discussed in more detail in the following.

12.3 Modeling processes at the deep bio-geo interface 12.3.1 Feeding the deep biosphere (biogenic gas) Biogenic gas is an important energy resource and the target of industry exploration efforts. Estimated potential biogenic gas reserves may be as high as 20 % of total worldwide conventional gas reserves [13]. The formation of biogenic gas is relatively well understood: methanogenic bacteria metabolize sedimentary organic matter under anoxic, sulfate-free conditions at temperatures below 75 °C [14]. Accordingly, biogenic gas is an important product generated in the deep biosphere, however, a fundamental understanding of biogenic gas formation mechanisms and rates is still largely lacking. For the exploration of this type of resource, rules of thumb are commonly applied: besides the availability of dispersed organic matter in the sediments it appears that the burial rate is a controlling factor. Clayton [14] states that burial rates of 200 to 1000 m/million year appear to be optimal, whereby slower burial results in total oxidation of the organic matter prior to methanogenesis, while higher burial rates lead to the establishment of too high temperatures before methanogenesis can really start. Geologically speaking, river deltas in rapidly subsiding basins are the best sites to find bacterial gas. The potential yield of gas generated by methanogenesis can be estimated by a simple mass balance based upon the mass fraction of carbon converted. Clayton [14] assumes that 10% of the total organic carbon (TOC) can be converted to methane. Newer investigations indicate that simply the presence of organic matter is not enough

12.3 Modeling processes at the deep bio-geo interface |

265

for biogenic gas formation, but that the organic matter needs to be accessible to methanogens, i.e. it is necessary to differentiate between “reactive organic carbon” (ROC) as compared to the TOC [15]. In addition, neither in situ optimum growth conditions for deep biosphere methanogens, nor carbon turnover rates are known (i.e. laboratory derived optimum growth temperatures vary widely for different archaeal strains). Accordingly, since a complete process description is still lacking, the numerical simulation of bacterial gas formation at a basin scale based on complete process understanding has yet to be implemented. Nevertheless, basin models offer much, if not all, of the boundary conditions required for the numerical simulation of bacterial gas formation: temperature, pressure and porosity evolution, the possibility to implement kinetic reaction schemes and full fluid flow modeling. Accordingly the first calculation schemes for biogenic gas generation in time and space are now being presented [16, 17], whereby in both cases bacterial turnover rates and optimum growth conditions are empirical and based on observations from wells (details on the assumptions used are, however, unpublished). An example of the modeled biogenic gas expulsion rate from the BP model [17] is shown in 󳶳 Fig. 12.2, and exemplifies that a clear maximum of generation is assumed to occur at roughly 35 °C at a geologic heating rate. As mentioned earlier, the assumptions behind this maximum in generation/expulsion of biogenic gas remain unclear. Warburton [16] compiled an overview of optimum growth temperatures of methanogenic bacteria found in sedimentary environments reported in the literature, and recognized two main temperature ranges of optimal growth: at around 38 °C and just above 60 °C. Accordingly, the assumption was made that meosphilic and thermophilic

Fig. 12.2: Schematic example of BP’s model for biogenic gas and thermogenic oil and gas generation (modified after Osborne and Barwise [17]).

266 | 12 Basin Modeling and the Deep Biosphere

Fig. 12.3: Comparison of maximum biogenic methane generation rates defined by the authors listed using different analytical and modeling approaches.

methanogens can contribute to bacterial gas formation, with the mesophiles being the dominant species. Based on the reported optimum growth temperatures, Warburton empirically defined a pseudo-kinetic model to reproduce the observations in Basin Modeling. His modeled generation rates of biogenic gas formation, for a sediment containing 1% TOC, reach a maximum of almost 10−10 pmol/d/cm3 (󳶳 Fig. 12.3). A different approach to deciphering interactions between thermally-controlled geologic processes and the deep biosphere was presented by Horsfield and coworkers [10]. Their investigation of sediments from ODP Sites 1173, 1174 and 1177 (Leg 190) using geochemical, microbiological and Basin Modeling methods showed that an overlap of abiotic catagenetic processes and living deep biosphere systems occurs there. They demonstrated that rates of deep biosphere respiration and kinetically controlled abiotic substrate generation, the rates of which were derived from Basin Modeling of the sedimentary evolution at the study sites, coincide. They inferred from these results that biologic utilization of abiotically-generated substrates is taking place. Ex situ metanogenesis experiments at in situ temperatures were performed on selected samples from these wells and showed that the observed methanogenesis rates decreased with increasing depth until the onset of thermal generation, and then increased again. Minimum and maximum methanogenesis rates monitored in these

12.3 Modeling processes at the deep bio-geo interface |

267

experiments were compared to those predicted by Osborne and Barwise [17] in their modeling approach (󳶳 Fig. 12.3), and show that both fall within the same range (10−10 to 10−15 pmol/d/cm3 ). Comparing estimated methanogenesis rates from Horsfield et al. [10] for the two well sites in the Nankai Trough to Osborne and Barwise’s [17] estimated biogenic methane expulsion rates as a function of temperature reveals that in all cases, a main peak of biogenic methane generation and expulsion is observed at temperatures around 30 °C. Following this, peak rates fall to background values for the Osborne and Barwise [17] estimation, whereas in the Nankai Trough dataset, a second increase in biogenic methane generation is evident starting close to the base of the investigated sequences, at temperatures above 60 °C, coinciding with Warburton’s [16] observations. This is the temperature range at which abiotic generation of substrates was modeled to be taking place in the Nankai Trough and confirmed by gas measurements. Interestingly, bacterial cell counts largely mimicked the depth trends of methanogenesis rates in the wells, indicating that here, bacterial activity depends on substrate supply as well, and that this supply can be reconstructed based on the kinetics of abiotic reactions in combination with the reconstruction of the burial history. A different approach for the determination of microbial methane generation rates was used by Sivan et al. [18] and Wallmann et al. [19]. In both cases steady-state transport-reaction models were used to explain measured compound contents in sediments and resulted in similar methane generation rates (󳶳 Fig. 12.3), which also fall within the range of generation rates discussed here derived from different approaches. These examples show that such assessments of process controls can be applied in order to model the generation of biogenic gas, they would however profit from a fundamental, mechanistic understanding of the processes involved. Integration of process understanding with quantitative Basin Modeling is one method that can be applied to enhance our understanding of the deep biosphere. In the examples discussed above, Basin Modeling was used to assess the direct coupling of abiotic substrate generation and microbial utilization. Other situations, where substrates are generated by thermal conversion of sedimentary organic matter and then move to sites where they can be utilized by microbes, are much better known. Biodegraded petroleum accumulations represent such cases.

12.3.2 Petroleum biodegradation The understanding of subsurface petroleum biodegradation has evolved significantly in the last decades. An excellent summary was published by Head et al. [8], and in the following, only the main observations on processes and controls of biodegradation are listed. The biodegradation of petroleum occurs under anaerobic conditions in the subsurface, and is characterized by the preferential metabolization of hydrocarbons over nonhydrocarbons (sulfur-, nitrogen- and oxygen-containing compounds), resulting in

268 | 12 Basin Modeling and the Deep Biosphere a decrease in the economic value of the oil. Petroleum biodegradation occurs at the oil-water contact, where microbial communities can thrive, having direct access to a continuous diffusive flux of hydrocarbons from the oil column as well as to inorganic nutrients from the water leg. There, transport could be either by water flow or also diffusion. The main control on biodegradation is temperature, where at in situ temperatures over 80 °C, biodegraded oils are virtually not encountered. The likelihood of finding a biodegraded fluid increases with decreasing reservoir temperature, however, nonbiodegraded oils are also found at temperatures in part significantly below 80 °C. Wilhelms et al. [20] recognized that such situations were encountered dominantly in inverted basins, i.e. basins which had been subjected to, in part, significant erosion and therefore cooling. In cases where the reservoir formation had been subjected to temperatures above 80 °C and then cooled during uplift, they proposed that the otherwise ubiquitous microbial community had been inactivated, i.e. a form of “pasteurization”, such that fresh oil reaching the reservoirs would remain unaltered. This theory implies that the geologic evolution of a basin has a profound effect on the extent of the deep biosphere, at least with respect to that of petroleum degrading microbial consortia. As Larter et al. [21] stated, the pasteurization theory implies that “the 80 °C limit is the death line for hydrocarbon degraders in basins and perhaps even the base of life itself in deep oligotrophic sediments”. While the mechanisms of anaerobic biodegradation of individual hydrocarbons have been studied microbiologically in the lab in great detail [e.g. 22, 23], evidence that the same mechanisms likely work in subsurface environments was only recently provided by Aitken et al. [24] who identified, in natural biodegraded oils, the same metabolites encountered in the laboratory as a product of biodegradation. The rate at which biodegradation proceeds in deeply-buried petroleum reservoirs is, however, difficult to assess based on laboratory experiments, as these are usually conducted under drastically different conditions than found in the subsurface: where in the deep biosphere biodegradation occurs under anoxic and nutrient limited conditions, in the pore space of lithified sediments and at elevated pressures and temperatures, laboratory experiments usually are performed on bacterial cultures in anoxic settings at elevated temperatures but in direct contact with excess nutrients and usually at ambient pressures. Such conditions are chosen in order to recognize significant rates at human time scales. However, even hexadecane degradation in the lab is a slow process, and it seems likely that subsurface biodegradation rates are orders of magnitude slower. Accordingly, we can expect that the rates monitored in laboratory experiments differ significantly from those in nature. In the absence of hard data regarding subsurface biodegradation rates Yu et al. [25] developed a biodegradation index (BDI) which was flexibly adapted to different basins in order to estimate the degree of biodegradation of individual fields. In this approach, the intensity of biodegradation was a function of the evolution of reservoir temperature during charge, limited by a maximum temperature (termed critical temperature) above which no further degradation takes place and a basin-dependent

12.3 Modeling processes at the deep bio-geo interface |

269

scaling constant. This method did not consider the filling history of the studied reservoirs, and was therefore limited in its application. An extension of the BDI was proposed by Wilhelms et al. [26] and included a tentative description of the filling history by allowing a variable evolution of oil column height during defined time intervals. Nevertheless, this approach also lacked the description of complete filling and spilling dynamics as well as a compositional variability of the charging fluid. Larter et al. [21] estimated subsurface biodegradation rates using a variety of techniques in order to provide the basis for simulating the process in Basin Modeling. They used three approaches to assess natural biodegradation rates: a) a homogeneous and static whole oil column minimum rate assessment, b) a one-dimensional calculation of diffusion controlled hydrocarbon supply to the base of a static oil column with subsequent degradation occurring at the oil-water contact and calibrated to compositional gradients observed in biodegraded oil fields and c) an approach which uses a reservoir charge history (supplying the volume or mass of oil charged in a unit time) and the assumption that oils can be compositionally simplified to a two-component mixture: degradable and nondegradable. Degradation rate of the degradable fraction was varied until a match to observed n-alkane (degradable) to hump (unresolved complex mixture of HCs not resolvable by chromatographic methods, assumed to be representative of the undegradable fraction) ratios was achieved. The results of these calculations indicated that under subsurface conditions zero-order biodegradation rate constants in the biodegradation zone (oil-water contact in cases b and c) are around 10−6 to 10−7 per year, corresponding to field wide minimum rate estimates (also zero-order, case a) of 10−8 kg hydrocarbon/kg oil/year for the entire oil column. These numbers translate to net hydrocarbon degradation rates in the reservoir of around 10−6 mmol oil/L/day, values which are comparable to reported aquifer respiration rates [21]. A few years later Larter et al. [27] expanded this approach to include degradation fluxes (in units of kg/m2 /year) of individual petroleum fractions and compounds. The above-listed studies attempted to link biodegradation processes to the evolution of petroleum systems, especially reservoir charge models. They lacked, however, the possibility to account for dynamic changes in reservoir geometry, large variations in filling rate, petroleum composition, or leakage or spilling out of the reservoir. For the inclusion of such geologically realistic variability two- or three-dimensional geologic reconstructions are required. De Barros-Penteado et al. [28] were the first to attempt to calculate natural in-reservoir biodegradation rates at geologic time scales by combining Petroleum System Modeling and geochemical analyses. They first compared biodegraded and nonbiodegraded fluids from the Potiguar Basin, Brazil, and calculated hydrocarbon losses in the natural environment [29]. Then using a compositional kinetic model of hydrocarbon generation combined with two-dimensional Basin Modeling they reconstructed the filling history of degraded and undegraded accumulations in the study area, allowing thus the calculation of biodegradation rates as a function of petroleum

270 | 12 Basin Modeling and the Deep Biosphere residence time in the reservoir, the temperature history and including hydrodynamics. They expressed biodegradation rates as mass loss in percent of the original mass per million years for individual compound classes. The main observations from this study was that the light oil and gas fraction was the most susceptible to biodegradation, with losses of the C14-fraction reaching 29%/Ma. They concluded that temperature exerted the strongest control on biodegradation rates, with highest rates modeled for temperatures below 40 °C, and that the residence time of the oil in the reservoir played only a minor role. The work of de Barros-Penteado et al. [28] was completed by the development of a compositional and quantitative numerical simulator for the prediction of biodegradation level and corresponding chemical changes [30], with the plan of coupling this simulator to a dynamic basin model at a later stage. The core of this approach focused on developing a mathematical model of the biological processes taking place during biodegradation, whereby stoichiometric equations describing the mass balance between a given hydrocarbon and the generated product were defined for different anaerobic conditions as follows [30]: Denitrifying conditions:

2 1 1 1 1 C𝑥 H𝑦 + ( 𝑥 + 𝑦) NO3 → 𝑥CO2 + ( 𝑥 + 𝑦) N2 + 𝑦H2 O 3 6 3 12 2 Sulfate reducing conditions:

2 2 2 1 1 2 C𝑥 H𝑦 + ( 𝑥 + 𝑦) SO4 → 𝑥CO2 + ( 𝑥 + 𝑦) H2 S + (− 𝑥 + 𝑦) H2 O 5 10 5 10 5 5 Methanogenic conditions:

1 1 1 1 1 C𝑥 H𝑦 + (𝑥 − 𝑦) H2 O → ( 𝑥 − 𝑦) CO2 + ( 𝑥 + 𝑦) CH4 4 2 8 2 8 These global stoichiometric equations were used to describe the complete biodegradation of various model compounds, and to calculate the extent of electron acceptor consumption. The authors also took into account that different hydrocarbon classes (e.g. C6−14 , C14+ 𝑛- and iso-alkanes, NSO compounds, etc.) are degraded in parallel, albeit at different rates. Six bioaccessible model compounds were defined as representative of oil compound classes, and each assigned a value describing its accessibility to biodegradation (arbitrarily ranked from 0 to 1), roughly based on model compound water solubility and diffusivity, as well as ranking of relative biosensitivity (termed intrinsic biodegradation parameter) describing differences in how easily such compounds are degraded as compared to other compound groups. A seventh compound class, the C14+ NSO compounds is also included in this scheme, this fraction is, however, assumed to be resistant to biodegradation. 󳶳 Table 12.1 lists the model compounds used by Haeseler et al. [30] as well as the assumptions regarding bioaccessibility and intrinsic biodegradability. A limiting factor for biodegradation is also the

12.3 Modeling processes at the deep bio-geo interface |

271

Table 12.1: Chemical classes, model compounds, values of bioaccessibility and intrinsic biodegradability used by Haeseler et al. 2010 [30]. Chemical Class C14− SAT C14− ARO C14+ n-alkanes C14+ iso-alkanes C14+ cyclo-alkanes C14+ ARO C14+ NSO

Model Compound

Accessibility

Intrinsic Biodegradability

C9 H20 C9 H12 C18 H38 C18 H38 C18 H34 C18 H24 none

0.5 1.0 0.2 0.2 0.2 0.2 0.0

0.8 0.8 1.0 0.5 0.4 0.3 0.0

availability of terminal electron acceptors in the water phase, such that high, albeit, realistic concentrations per volume water were defined a priori in the model. The values listed in 󳶳 Table 12.1 used by Haeseler et al. [30] represented the initial basic assumptions implemented in the numerical reproduction of biodegradation of petroleum. However, the authors state that such initial values are obviously not well constrained and must be considered as initial estimates which require optimization and validation using natural series of biodegraded oils. They performed this task using a series of biodegraded oils from the Potiguar basin, Brazil [29] and then applied the tuned model to biodegraded fluids from the Williston basin, USA. Haeseler et al. [30] demonstrated that their approach allowed the accurate prediction of biodegradation evolution, permitting the reproduction and prediction of biodegradation progress and ensuing compositional and physical property (e.g. API gravity) evolution of the fluids. They presented convincing evidence that their approach suitably describes the compositional evolution of the liquid phase, however, in their exemplary calculations of evolving products of biodegradation (CO2 , CH4 and H2 S, Table 6 of [30]) they indicate that their model results in the generation of significant amounts of CO2 . High CO2 contents are, on the other hand, not a common feature of biodegraded fluids. Larter and di Primio [31] showed that in the giant Troll gas field, which is characterized by a dry gas overlying a thin biodegraded oil leg, CO2 contents of the gas diminished with increasing level of gas biodegradation, indicating the likelihood of the in-reservoir reduction of carbon dioxide to methane, i.e. the generation of secondary biogenic methane (e.g. [32, 33]). CO2 contents of most biodegraded oilfields generally do not differ from those of neighboring undegraded fields, containing only a few mol %. If the bulk of petroleum degradation results in CO2 formation (e.g. [23]), the lack of enhanced CO2 contents in biodegraded fluids must be due to the reduction of CO2 to methane, which itself then requires a source of hydrogen. Head et al. [8] propose mineral hydrolysis, organic matter maturation (aromatization) or even from the porewater itself, as both thermodynamic and microbiological data indicate [23, 34], as sources of hydrogen.

272 | 12 Basin Modeling and the Deep Biosphere

Fig. 12.4: Exemplary variable definitions as used by Blumenstein et al. [35].

A similar approach to Haeseler et al. [30] but including the dynamic reconstruction of basin evolution, petroleum generation, migration and accumulation was already presented a few years earlier by Blumenstein et al. [35]. Also, in this case petroleum biodegradation is simulated assuming a limited number of compound groups, each with a definition of the proportion degradable and relative degradation rates. The main difference of the Blumenstein et al. [35] approach is that in addition to the compound group degradation properties, a temperature-dependent overall biodegradation rate control was defined. 󳶳 Figure 12.4 shows the assumed decrease in biodegradation rate with increasing geologic temperature as implemented in the model. The biodegradation rate is a zero-order rate constant with respect to the substrate concentration (hydrocarbons), since degradation occurs at the oil-water contact independent of the amount of oil or the column height [21]. Thus the biodegradation rate is described in kg/m2 /Ma and corresponds to the “degradation flux” defined by Larter et al. [21, 27]), but varies as a function of temperature (󳶳 Fig. 12.4). The authors assume a constantly high degradation rate at low temperatures, which decreases to zero at the sterilization temperature of 80 °C [20]. Importantly, the sterilization concept is implemented in such a manner that no biodegradation is considered if the reservoir temperature exceeded 80 °C in the past, even if it has cooled due to erosion in the following geologic history. The compositional subdivision of the petroleum in the Blumenstein et al. [35] approach is based on the engineering compositional characterization used by the PhaseKinetics method developed by di Primio and Horsfield [36]. Such compositional kinetic models allow the prediction of petroleum composition and phase behavior

12.3 Modeling processes at the deep bio-geo interface |

273

Table 12.2: Component definitions, carbon ranges, description of respective biodegradability and relative rates used by Blumenstein et al. [35]. Component

Number of C atoms

Biodegradability

Relative biodegradation rate

Methane Ethane Propane i-Butane n-Butane i-Pentane n-Pentane C6 PK_P10 PK_P20 PK_P30 PK_P40 PK_P50 PK_P60+

1 2 3 4 4 5 5 6 7–14 15–24 25–34 35–44 45–54 55–80

– +++ +++ +++ +++ +++ +++ +++ ++ ++ ++ + + +

– + +++ ++ +++ ++ ++ ++ +++ ++ ++ + + +

using Petroleum System Modeling, accordingly, the compositional format was chosen for the simulation of biodegradation in order to be able to calculate petroleum quality and property changes as a function of increasing degradation, including also volumetric assessment of petroleum phase changes. The compositional subdivision is based on molar contents of the dominant pure hydrocarbon compounds in the gas range (methane through pentane) and compound groups for the C6 , C7−14 , C15−24 , C25−34 , C35−44 , C45−54 and C55−80 ranges (󳶳 Table 12.2). As shown in 󳶳 Table 12.2, Blumenstein and coworkers [35] assign descriptions of biodegradabilty and relative biodegradation rates to each of the compounds or compound groups. These definitions are purely empirical and based on experience, observations and published data. In essence biodegradability, relative degradation rates, as well as the dependence of biodegradation rate with temperature (󳶳 Fig. 12.4) are defined empirically, and can be used to tune the biodegradation model to the observed natural situation. The Blumenstein et al. [35] scheme was linked to a dynamic three-dimensional petroleum system model in a post-processing manner. Results concerning petroleum generation, migration and accumulation were exported for every time step in an ascii format, together with the respective reservoir temperature. Petroleum masses and compositions accumulated in any reservoir were then subjected to the biodegradation calculations based on the premises discussed above. Despite the fact that a feedback back to the petroleum system simulation regarding masses of petroleum, and hence volumes, lost due to biodegradation was not implemented in the study, a good match of evolving petroleum alteration and ensuing changes in PVT properties (Pressure, Volume, Temperature, i.e. physical properties and phase behavior) and petroleum

274 | 12 Basin Modeling and the Deep Biosphere quality for two known biodegraded fields was reached. For this match, the assumed maximum biodegradation rate (󳶳 Fig. 12.4) had to be defined as 15 kg HC/m2 /Ma, a value which corresponds very well with Larter et al. [21] minimum biodegradation rate estimates. The main uncertainties inherent in the combination of Petroleum System Modeling and biodegradation modeling include: a) the correct structural representation of the petroleum reservoir and b) the ability to dynamically model biodegradation in Petroleum System Modeling. Regarding point a): as biodegradation is observed to occur at the oil-water contact, this area of interface between the oil (or gas) leg and the underlying water phase must be represented correctly in the model in order to accurately reproduce biodegradation effects and extent. This is a major problem in Petroleum System Modeling where grid spacing, cell sizes and even seismic resolution limit the accuracy of structural representation for individual petroleum reservoirs. Here, new developments in grid definition, i.e. the implementation of local grid refinements, combinations of models with different gridding, etc. will allow a much better representation of reservoir structure in the future. As to point b): at least one of the software packages available has implemented a biodegradation option into the simulation, taking compositional and volumetric changes, petroleum recharge, mixing and spilling, as well as phase behavior into account in a fully dynamic representation of reservoir processes. While such geologic modeling approaches constrained by empirical assumptions on biodegradation rates and effects probably appear unscientific, fortuitous, and inaccurate for microbiologists, it should be taken into account that they provide a means to extrapolate laboratory findings to the time-temperature field of geologic process, and thus help to investigate how the deep biosphere “ticks” at a basin scale.

12.4 Modeling processes at the shallow bio-geo interface Up to now we have focused on the relatively deep realms of sedimentary basins, and how the deep biosphere interacts with sedimentary organic matter or petroleum products thermogenically generated there from. However, processes and interactions of deep biosphere (or dark energy biosphere, c.f. [37]) with thermogenically produced fluids occurring at shallower levels of burial also have the potential to being relevant at a global scale. Kroeger et al. [38] speculated on the possible contribution of thermogenic methane to the atmosphere over geologic time and possible effects with respect to climate forcing. In their calculations, they demonstrated that vast amounts (tens to hundreds of thousands of Gigatons) of carbon, mainly in the form of methane, could have been released to the atmosphere from sedimentary basins over geologic time. The geosphere does in fact contain a vast repository of carbon which can be transformed into mobile fluids and migrate through the sediments to the surface. As discussed earlier, besides thermogenic processes, the deep biosphere partakes in this

12.5 Conclusions

| 275

transformation via methanogenesis, and such fluxes of carbon represent of course potential feedstocks for the shallower biosphere where e.g. anaerobic oxidation of methane (AOM) can take place [39]. Total fluxes of mobile carbon, here simplified to methane, can be quantified at a basin scale using Petroleum System Modeling by applying kinetic descriptions of hydrocarbon generation, as well as the above-discussed biogenic processes of methane formation or petroleum degradation and secondary biogenic gas formation. While the pure thermogenic mass balance is already possible today, within the constraints of the correct description of sedimentary organic matter contents in the basin, an integrated assessment of all methanogenic and methanotrophic processes is still lacking. Nevertheless, Petroleum System Modeling allows to reconstruct locations of enhanced petroleum flux, i.e. locations of focused fluid flow, and also quantify the rates of flow. Using seismic data, positions of enhanced fluid flow can be visualized in the subsurface, these appear as areas of reduced seismic signal characterized as zones of acoustic turbidity or “white outs” when vague and “gas pipes” or “gas chimneys” when discrete. Ostanin et al. [40] performed such a mapping exercise in the Barents Sea, examining leakage structures leading to gas hydrate zones as well as seafloor pock marks. Their results, put into a geologic temporal framework [40, 42], indicate that episodic petroleum leakage from subsurface gas reservoirs is currently ongoing but was much enhanced following glacial erosion episodes. Especially the basin wide work performed by Rodrigues-Duran et al. [42] shows the possibility of quantifying thermogenic methane fluxes in sedimentary basins. Such work needs to be combined with AOM process understanding in order to be able to characterize the total flux out of a sedimentary basin, as we can expect that only sites of focused methane fluxes can bypass the likely ubiquitous AOM microbial consortia and be released into the hydroor atmosphere.

12.5 Conclusions Petroleum System Modeling can provide deep biosphere researchers with the possibility of investigating processes occurring in the subsurface over geologic time scales. Currently, ongoing efforts focus on deep biosphere processes involving fossil energy resources such as biogenic gas formation or petroleum biodegradation. In these cases, model predictions are still hampered by only very loosely defined reaction rates. Nevertheless, modeling results are now already indicating what reaction rate ranges make sense at geologic heating and burial rates, offering thus some novel input to deep biosphere research. An enticing new direction of investigation involves looking at deep biosphere processes at a basin scale and over geologic time, where first efforts are revealing a highly dynamic system coupling deep biosphere foodstock generation, in situ consumption or migration to locations where microbial metabo-

276 | 12 Basin Modeling and the Deep Biosphere lization can take place. Such coupled efforts could enhance our understanding of the subsurface carbon cycle at a basin or even global scale.

References [1] [2]

[3] [4]

[5] [6] [7]

[8] [9]

[10]

[11] [12]

[13] [14] [15]

[16]

Gold T. The deep, hot biosphere. Proc Natl Acad Sci USA 89 (1992), 6045–6049. Takai K, Nakamura K, Toki T, Tsunogai U, Miyazaki M, Miyazaki J, Hirayama H, Nakagawa S, Nunoura T, and Horikoshi K. Cell proliferation at 122 degrees C and isotopically heavy CH4 production by a hyperthermophilic methanogen under high-pressure cultivation. Proc Natl Acad Sci USA 105 (2008), 10 949–10 954. Whitman WB, Coleman DC, Wiebe WJ. Prokaryotes: the unseen majority. Proc Natl Acad Sci USA 95 (1998), 6578–6583. Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16 213– 16 216. Parkes RJ, Cragg BA, Wellsbury P. Recent studies on bacterial populations and processes in subseafloor sediments: a review. Hydrogeol J 8 No. 1 (2000), 11–28. Westrich JT, Berner RA. The role of sedimentary organic matter in bacterial sulfate reduction: the G model tested. Limnol Oceanogr 29 No. 2 (1984), 236–249. Mangelsdorf K, Haberer RM, Zink K-G, Wilkes H. Molecular indicators for viable microbial communities of the deep biosphere in sediments of the Mackenzie River Delta, Northwest Territories, Canada. Book of Abstracts, Part I, 21th International Meeting on Organic Geochemistry, Kraków, Poland (2003), 275–276. Head IM, Jones DM, Larter SR. Biological activity in the deep subsurface and the origin of heavy oil. Nature 426 (2003), 344–352. Naeth J, di Primio R, Horsfield B, Schaefer B, Shannon PM, Bailey WR, Henriet JP. Hydrocarbon seepage and carbonate mound formation: a basin modelling study from the Porcupine Basin (offshore Ireland). Journal of Petroleum Geology 28 No. 2 (2005), 147–166. Horsfield B, Schenk HJ, Zink K-G, Ondrak R, Dieckmann V, Kallmeyer J, Mangelsdorf K, di Primio R, Wilkes H, Parkes RJ, Fry J, Cragg B. Living microbial ecosystems within the active zone of catagenesis: implications for feeding the deep biosphere. Earth and Planetary Science Letters 246 No. 1–2 (2006), 55–69. Hantschel, T and Kauerauf AI. Fundamentals of Basin and Petroleum Systems Modeling. Springer-Verlag, Berlin, Heidelberg, 2009. Parkes RJ, Wellsbury P, Mather ID, Cobb SJ, Cragg BA, Hornibrook ERC, Horsfield B. Temperature activation of organic matter and minerals during burial has the potential to sustain the deep biosphere over geological timescales. Organic Geochemistry 38 No. 6 (2007), 845–852. Rice DD, Claypool GE. Generation, accumulation and resource potential of biogenic gas. AAPG Bull 65 (1981), 5–25. Clayton C. Source volumetrics of biogenic gas generation. In: Vially R (ed), Bacterial Gas. Editions Technip, Paris, pp. 191–204, 1992. Shuai YH, Zhang SC, Chen JP, Su AG. Source of nutrient substrates for microbes in deep biosphere and characteristics of biogenic gas source rock. Science China Earth Sciences 53 No. 8 (2010), 1163–1168. Warburton I. A “pseudo” kinetic scheme for modelling the formation of biogenic methane. Poster presented at the AAPG Hedberg Conference Petroleum Systems: Modeling the Past, Planning the Future. Nice, France, 1–5 October 2012.

References | 277

[17] Osborne M, Barwise T. Beyond Orgas-BP’s new predictive model for biogenic and thermogenic gas expulsion from source rocks. International Meeting of Organic Chemistry (IMOG) 2011, Oral presentation. [18] Sivan O, Schrag DP, Murray RW. Rates of methanogenesis and methanotrophy in deep-sea sediments. Geobiology 5 (2007), 141–151. [19] Wallmann K, Aloisi G, Haeckel M, Obzhirov A, Pavlova G, Tishchenko P. Kinetics of organic matter degradation, microbial methane generation and gas hydrate formation in anoxic marine sediments. Geochim. Cosmochim. Acta 70 No. 15 (2006), 3905–3927. [20] Wilhelms A, Larter SR, Head IM, Farrimond P, di Primio R, Zwach C. Biodegradation of oil in uplifted basins prevented by deep-burial sterilization. Nature 411 (2001), 1034–1037. [21] Larter S, Wilhelms A, Head I, Koopmans M, Aplin A, di Primio R, Zwach C, Erdmann M, Telnaes N. The controls on the composition of biodegraded oils in the deep subsurface – part 1: biodegradation rates in petroleum reservoirs. Organic Geochemistry 4 (2003), 601–613. [22] Widdel F, Rabus R. Anaerobic biodegradation of saturated and aromatic hydrocarbons. Current Opinion in Biotechnology 12 (2001), 259–276. [23] Zengler K, Richnow HH, Rossello-Mora R, Michaelis W, Widdel F. Methane formation from longchain alkanes by anaerobic microorganisms. Nature 401 (1999), 266–269. [24] Aitken CM, Jones DM, Larter SR. Anaerobic hydrocarbon biodegradation in deep subsurface oil reservoirs. Nature 431 (2004), 291–294. [25] Yu AZ, Cole G, Grubitz G, Peel F. How to predict biodegradation risk and reservoir fluid quality. World Oil April (2002), 1–5. [26] Wilhelms A, Erdmann M, Larter SR. Easy-Fest versus BDIuncertainties in Predrill prediction of biodegradation degree in subsurface petroleum reservoirs. AAPG Bull 88 (2004), Supplement. [27] Larter S, Huang H, Adams J, Bennett B, Jokanola O, Oldenburg T, Jones M, Head I, Riediger C, Fowler M. The controls on the composition of biodegraded oils in the deep subsurface. Part II – geological controls on subsurface biodegradation fluxes and constraints on reservoir-fluid property prediction. AAPG Bull 90 (2006), 921–938. [28] de Barros Penteado H, Behar F, Lorant F, Oliveira DC. Study of biodegradation processes along the Carnaubais trend, Potiguar Basin (Brazil) – part 2. Organic Geochemistry 38 (2007), 1197– 1211. [29] Behar F, Penteado HL, Lorant F, Budzinski H. Study of biodegradation process along the Carnaubais trend, Potiguar Basin (Brazil) – part 1. Organic Geochemistry 37 (2006), 1042–1051. [30] Haeseler F, Behar F, Garnier D, Chenet P-Y. First stoichiometric model of oil biodegradation in natural petroleum systems. Part I – The BioClass 0D approach. Organic geochemistry 41 (2010), 1156–1170. [31] Larter SR, di Primio R. Effects of biodegradation on oil and gas field PVT properties and the origin of oil rimmed gas accumulations. Org Geochem 36 No. 2 (2005), 299–310. [32] Scott AR, Kaiser WR, Ayers Jr WB. Thermogenic and secondary biogenic gases, San Juan Basin, Colorado and New Mexico; implications for coalbed gas producibility. AAPG Bull 78 (1994), 1186–1209. [33] Dessort D, Poirier Y, Sermondadez G, Levache D. Methane generation during biodegradation of crude oil. In: Abstracts of the 21st International Meeting on Organic Geochemistry, Krakow, Poland, 2003. [34] Helgeson HC, Knox AM, Owens DH, Shock EL. Petroleum, oil-field waters and authigenic mineral assemblages – are they in metastable equilibrium in hydrocarbon reservoirs. Geochim Cosmochim Acta 57 (1993), 3295–3339.

278 | 12 Basin Modeling and the Deep Biosphere [35] Blumenstein IO, Krooss BM, di Primio R, Rottke W, Müller E, Westerlage C, Littke R. Biodegradation in numerical basin modeling: a case study from the Gifhorn Trough, N-Germany. International Journal of Earth Sciences 97 No. 5 (2008), 1115–1129. [36] di Primio R, Horsfield B. From petroleum-type organofacies to hydrocarbon phase prediction. AAPG Bull 90 No. 7 (2006), 1031–1058. [37] Edwards KJ, Becker K, Colwell F. The deep, dark energy biosphere: intraterrestrial life on Earth. Annu Rev Earth Planet Sci V 40 (2012), 551–568. [38] Kroeger KF, di Primio R, Horsfield B. Atmospheric methane from organic carbon mobilization in sedimentary basins – the sleeping giant? Earth Science Reviews 107 No. 3–4 (2011), 423–442. [39] Reeburgh WS. Oceanic methane biogeochemistry. Chem Rev 107 (2007), 486–513. [40] Ostanin I, Anka Z, di Primio R, Bernal A. Hydrocarbon leakage above the Snøhvit gas field, Hammerfest Basin SW Barents Sea. First Break 30 No. 11 (2012), 55–60. [41] Ostanin I, Anka Z, di Primio R. Hydrocarbon plumbing systems above the Snøhvit gas field: structural control and implications for thermogenic methane leakage in the Hammerfest Basin, SW Barents Sea. Marine and Petroleum Geology 43 (2013), 127–146. [42] Rodrigues Duran E, di Primio R, Anka Z, Stoddart D, Horsfield B. 3D-Basin modeling of the Hammerfest Basin (southwestern Barents Sea): A quantitative assessment of petroleum generation, migration and leakage. Marine and Petroleum Geology 45 (2013), 281–303.

Doug LaRowe, Jan Amend

13 Energetic constraints on life in marine deep sediments 13.1 Introduction Although it is becoming clear that microorganisms are abundant in marine deep sediments [1–8], it is unclear what percentage of cells are active, how fast they are growing or what controls their diversity and population size [9]. Addressing these issues is a formidable task due to the relative inaccessibility of these environments, the difficulty of cultivating representative microorganisms and the long time scales associated with some of their lifestyles [2, 10–12]. However, quantitative limits on life in the subsurface can be determined by using the physiochemical data that describe their habitats. In particular, the chemical composition can be used to constrain likely metabolic strategies and rates in a given setting. This is accomplished by calculating values of Gibbs energy available from reactions containing different combinations of the electron donors and acceptors that are found in these environments. Not only can Gibbs energies of reaction reveal which catabolic strategies are thermodynamically possible, but they can also help determine which geochemical variables (e.g. temperature, pressure, pH, salinity, composition) are controlling microbial activity. When reduced to an environmentally-appropriate common factor, the energetic potential of all biogeochemical environments can be directly compared to assess how energy limitations affect the amount and type of biomass in them. In the present chapter, geochemical data obtained from sediment cores taken from the Peru Margin, South Pacific Gyre and Juan de Fuca Ridge are used to assess the Gibbs energies of plausible catabolic strategies including, but not limited to, the oxidation of organic matter, methane and hydrogen by a variety of electron acceptors. In conjunction with cell-count data, the results of these calculations illustrate the importance of normalizing energy availability to the limiting substrate and how geochemical data can be used to better understand the distribution of life deep in marine sediments. Perhaps one of the most enticing aspects of using Gibbs energies to characterize the type and level of microbial activity in the deep biosphere is that the geochemical data required to do so are relatively widely available (e.g. Proceedings of the ODP and IODP). These kinds of calculations can be used to predict microbial activity in places where chemical data are available, but where microbiological samples have not been taken. The calculations described below are universal in that they can illuminate which biogeochemical processes could be operating in any environment as long as the prevailing temperature, pressure and composition are appropriately taken into account.

280 | 13 Energetic constraints on life in marine deep sediments

13.2 Previous work The amount of energy available in an environment largely determines the metabolic state of the resident microorganisms [13]. That is, more energy-rich settings can support larger populations, more biomass and faster growth rates than low-energy environments. Because active microorganisms ultimately derive their energy from the catalysis of redox reactions, the Gibbs energies of these types of reactions have the potential to reveal the metabolic state of a microbial ecosystem. This kind of energetic profiling has been carried out successfully for hydrothermal systems in the deep sea [14–20], the shallow sea [21–26] and in terrestrial systems [27–33] and, to a lesser extent, in ocean sediments [34, 35] and basement rock [5, 36–38]. These studies have examined more than 100 organic and inorganic redox reactions under a wide range of environmental conditions. The calculated Gibbs energies of the thermodynamically favored (exergonic) reactions vary between approximately 0 and −120 kJ (mol e− )−1 , indicating which reactions are possible catabolic strategies and how much energy they provide.

13.3 Study site overview The three sites examined in the current study were chosen because they represent three significant types of marine sedimentary settings and are biogeochemically well characterized. Sediments at the Juan de Fuca site (47°45.2󸀠 N, 127°45.8󸀠 W) are influenced by hydrothermal activity resulting from the adjacent ocean spreading center. Although these kinds of sediments do not cover a significant fraction of the ocean floor, the water-rock interactions that typify submarine hydrothermal systems support diverse, dense ecosystems that influence global biogeochemical cycles. Peru Margin sediments (10°58.60󸀠 S, 77°57.46󸀠 W) lie on an active continental shelf where primary productivity is relatively high and therefore, organic matter accumulation is significant. Despite covering only about 7% percent of the ocean floor, nearly 58% of global marine particular organic matter deposition occurs in such zones [39]. With its ultraslow sedimentation rate, great distance from continental land mass, and corresponding low organic carbon sedimentation rate, the South Pacific Gyre site (23°51.04󸀠 S, 165°38.66󸀠 W), represents the majority of the world’s marine sediments. Although most of the geochemical and microbial cell-count data for these sites have been tabulated, several data profiles had to be extracted from figures using Plot Digitizer (http://plotdigitizer.sourceforge.net). For all sites, the density of pore fluids was taken to be 1.035 kg/L, the concentrations of HCO−3 were taken to be equal to the reported dissolved inorganic carbon (DIC) and, in the absence of perfectly overlapping data sets, several chemical concentrations and cell numbers were interpolated such that calculations could be carried out at particular sediment depths. In order to carry out Gibbs energy calculations, the concentrations of all reactant and product species

13.3 Study site overview

| 281

in a given chemical reaction most be known. In the absence of such data, the concentrations of some compounds were estimated. These estimates, the sources of the chemical data and other characteristics of the three study sites are summarized below and in the Appendix.

13.3.1 Juan de Fuca (JdF) Due to its proximity to an active rifting center and a continental margin, the sediments penetrated by drill site U1301 on the JdF ridge are influenced by hydrothermal activity and rapid sedimentation rates (∼75 m/MY given a 3.5 MY old basaltic crust and a 262 m layer of sediments) [40]. The basaltic crust underlying these insulating sediments hosts a dynamic, heat-driven flux of fluids that are a mixture of seawater and hydrothermal solutions that diffuse into the overlying sediments. As a result, the thermal gradient in these sediments (0.228 °C/m) is steeper than that in typical marine sediments, and the concentration profiles of various metabolically-relevant species display complex patterns influenced by both of the sediment column boundaries (see Appendix and 󳶳 Fig. 13.A1). Laboratory studies have shown that microorganisms from these sediments can grow on enriched media via sulfate reduction in the presence of a variety of organic compounds including short-chain alcohols, fatty acids and H2 , and by fermentation of ethanol, pyruvate and betaine [41]; some isolates produced sulfide in the presence of Fe(III) and Mn(IV), but did not show any corresponding growth. + 2+ 2+ Concentration profiles of SO2− 4 , NH4 , Mn , Fe , DIC, dissolved organic carbon (DOC) and CH4 as well as cell numbers, pH, thermal gradient and porosity data for JdF holes U1301C & D were taken from [40]. In the absence of concentration data for HS− and N2 , depth-constant concentrations of 10−5 and 10−4 mol/kg H2 O were used in the relevant energy calculations. The water depth is 2656 m.

13.3.2 Peru Margin (PM) The PM site, under only 78 m of water, underlies a zone of high primary productivity that has resulted in an average sedimentation rate of 24 m/MY [42]. As a result, the sediments here are rich in organic matter (1–12% TOC by weight) [43] and EAs such as O2 and NO−3 disappear in the upper few centimeters of the sediment column. Sulfate, which is highly concentrated in seawater, initially decreases as a function of depth, but similar to JdF sediments, it then increases again owing to diffusion from a deep source. In particular, a Miocene brine supplies sulfate and other ions to the sediment column from below creating complex nutrient profiles (see 󳶳 Fig. 13.A2). Although inner-shelf environments (water depth < 150 m) only cover 5.8% of global ocean settings, a disproportionately large amount of primary production occurs in these waters [39]. The concentration profiles of nitrate, sulfate, DIC, ammonium,

282 | 13 Energetic constraints on life in marine deep sediments sulfide, methane, Mn2+ , and Fe2+ are thought to be influenced by organic matter 2+ 2+ degradation [4]. Concentration profiles of SO2− 4 , Mn , Fe , DIC, acetate, CH4 , pH and bacteria for PM site 1229 were taken from [42]. Porosity data and a constant value of 0.3 millimolal for [HS− ] were taken from [44].

13.3.3 South Pacific Gyre (SPG) The slow sedimentation rate (1.1 m/M.Y.) and water depth (5695 m) at SPG are typical of open-ocean sites that are far from land [45]. Because open-ocean sites comprise half or more of all ocean environments [39], SPG sediments represent, by volume, a substantial portion of Earth’s ecosystems. The low sedimentation rates and correspondingly tiny amounts of organic matter delivered in these sediments results in low biomass concentrations [46]. The near absence of electron donors at this site means that electron acceptors (EAs) such as O2 and NO−3 can penetrate deep into the sediment column here. D’Hondt et al. [46] suggest that oxygen and nitrate are the dominant EAs and that radiolytic hydrogen and organic matter are the most common electron donors (EDs). − 2+ 2+ Concentration profiles of SO2− 4 , NO3 , O2 , H2 , Mn , Fe , DIC, total organic carbon (TOC), pH and cells and porosity for SPG holes U1365A & B were taken from [45]. For sediment depths in which H2 concentration were below detection (< 2.8 nM), a nominal value of 1 nM was assumed to carry out Gibbs energy calculations. In the absence of concentration data for HS− and N2 , depth-constant concentrations of 10−7 and 10−4 mol/kg H2 O were used to calculate the catabolic potential of reactions in which they are product species. In addition, although DOC concentrations were not reported for this site, they were estimated from TOC data according to [DOC] = 10−8∗ (%TOC), which produces DOC concentrations similar to those for hydrogen. Concentration data are largely absent at 43–65 meters below the seafloor (mbsf), so concentration data were interpolated from the values above and below these depths (see Appendix).

13.4 Overview of catabolic potential The 18 redox reactions considered as sources of available catabolic potential are listed in 󳶳 Table 13.1. Because not all of the concentration data that are required to evaluate the energetic potential for all of these reactions have been reported for each site, thermodynamic calculations were only carried out at all three sites for Reactions 3– 5, at two sites for Reactions 6–10 and at one site for the rest of the reactions (see 󳶳 Table 13.1). The electron donors considered include organic matter (OM), H2 , Mn2+ , Fe2+ , and NH+4 and electron acceptors include O2 , NO−3 , Mn(IV), Fe(III), and SO2− 4 . Acetoclastic methanogenesis (Reaction 6) was also considered at two sites. Although many minerals contain oxidized Fe and Mn, only goethite (FeOOH) and pyrolusite

13.4 Overview of catabolic potential |

283

Table 13.1: Catabolic reactions considered in the present study. Reaction 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

−

Location +

CH3 COO + 2O2 → H + 2HCO− 3 + − 5CH3 COO− + 8NO− 3 + 3H → 4N2 + 4H2 O + 10HCO3 − + − 2+ CH3 COO + 4MnO2 + 7H → 2HCO3 + 4Mn + 4H2 O 2+ CH3 COO− + 8FeOOH + 15H+ → 2HCO− 3 + 8Fe + 12H2 O − 2− − − CH3 COO + SO4 → HS + 2HCO3 CH3 COO− + H2 O → CH4 + HCO− 3 − − 4HCOO− + H+ + SO2− 4 → HS + 4HCO3 + 2+ − CH4 + 4MnO2 + 7H → 4Mn + HCO3 + 5H2 O CH4 + 8FeOOH + 15H+ → 8Fe2+ + HCO− 3 + 13H2 O − − CH4 + SO2− → H O + HS + HCO 2 4 3 H2 + 1/2O2 → H2 O + 5H2 + 2NO− 3 + 2H → N2 + 6H2 O + H2 + MnO2 + 2H → Mn2+ + 2H2 O H2 + 2FeOOH + 4H+ → 2Fe2+ + 4H2 O + − 4H2 + SO2− 4 + H → HS + 4H2 O 2+ 2Mn + O2 + 2H2 O → 2MnO2 + 4H+ 4Fe2+ + O2 + 6H2 O → 4FeOOH + 8H+ 2+ 3MnO2 + 4H+ + 2NH+ 4 → N2 + 3Mn + 6H2 O

SPG SPG SPG, PM, JdF SPG, PM, JdF* SPG, PM, JdF PM, JdF PM, JdF PM, JdF PM, JdF PM, JdF SPG SPG SPG SPG SPG SPG SPG JdF

SPG – South Pacific Gyre; PM – Peru Margin; JdF – Juan de Fuca. *FeOOH can represent goethite and anhydrous ferrihydrite. For JdF, DOC is taken to be acetate and for SPG, TOC was used to estimate acetate – see text. With the exceptions of MnO2 and FeOOH, all chemical species are in the aqueous state.

(MnO2 ) were used to represent these electron acceptors. The energetic consequences of this simplification are discussed below. The energetic potential in the marine sediments at JdF, SPG and PM are shown as a function of depth below the seafloor in 󳶳 Fig. 13.1 (a). In this figure, values of the Gibbs energy of reaction, Δ𝐺𝑟 , for reactions listed in 󳶳 Table 13.1 are depicted in units of kJ per mole of electron transferred, kJ(mol e− )−1 . The prevailing temperature, pressure and composition of the pore fluids in these sediments were explicitly taken into account as detailed below in Computational Methods. A prominent feature in this figure is that the reactions separate into two groups for all three sites. The high energy group of reactions, which typically yield values between −70 and −105 kJ(mol e− )−1 , are those in which O2 , NO−3 , and MnO2 are the oxidants, while the low energy group of reactions, yielding between −20 and +5 kJ(mol e− )−1 , are methanogenesis and those in which sulfate and FeOOH are the electron acceptors. In 󳶳 Fig. 13.1 (a), it can also be seen that for some parts of the JdF sediment column, values of Δ𝐺𝑟 are positive for methane oxidation by FeOOH and sulfate, indicating that they are not possible catabolic pathways at these sediment depths. Also, one might expect that because the most exergonic reactions are computed for the oxidation of H2 and OM in SPG sediments, that this site might have the fastest microbial metabolic rates or the largest biomass, and that the

284 | 13 Energetic constraints on life in marine deep sediments (a)

(b)

Juan de Fuca

OM + MnO +

NH

50

2

+ MnO

4

CH + MnO 4

2

Depth , mbsf

50

Depth , mbsf

Juan de Fuca 0

0

2

100

150 OM + SO

200

24

OM + SO

100

CH + MnO 4

OM + FeOOH NH

4

+

4

+ MnO

OM + MnO

OM + FeOOH CH + SO

2

150

200

methanogenesis

250

24

methanogenesis

2

2

CH + FeOOH 4

2-

250

4

CH + FeOOH 4

0

-20

-40

-60

-7

-80

-6

-5

-4

Peru Margin

-1

0

1

0 OM + MnO

CH + MnO

4

4

2

CH + MnO

4

OM + MnO

2

Depth , mbsf

50

100 OM + FeOOH

150

100 CH + SO 4

2-

4

150

2-

-20

OM + SO

4

methanogenesis

0

24

OM + FeOOH

4

CH + SO

2

50

4

OM + SO

2

CH + FeOOH

CH + FeOOH

-40

24

methanogenesis

-60

-80

-100

-7

-6

South Pacific Gyre

-5

-4

-3

-2

-1

0

1

South Pacific Gyre 0

0 H +O 2

H + NO OM + NO

-

2+

2

40

2

2

OM + MnO

50

Mn

2+

20

3

2

+O

2

2

H + SO

70

2

OM + SO

2-

2

2

2

2

2

H + MnO 2

2

2

-

H + NO

50

3

H + FeOOH 2

OM + FeOOH OM + SO

70

4 2-

+O

H +O

40

60

OM + FeOOH

+O

2+

Mn

30

H + FeOOH

60

2+

3

OM + MnO

+O

H + MnO

Fe

2

OM + NO

3

Depth , mbsf

Fe

30

10

2 -

2

20

OM + O

2

OM + O

10

Depth , mbsf

-2

Peru Margin

0

Depth , mbsf

-3

H + SO 2

4

24 2-

4

80

80 0

-20

-40

-60

-80 - -1

ΔG , kJ (mol e ) r

-100

-9

-8

-7

-6

-5

-4 3

log E (J/cm ) r

-3

-2

13.4 Overview of catabolic potential |

285

Opposite page: Fig. 13.1: Catabolic energy potential for the reactions listed in 󳶳 Table 13.1 as a function of depth in three marine sediment columns: a) Gibbs energy in units of kilojoules per mole of electron transferred, kJ (mol e− )−1 and b) the logarithm of energy available per cubic centimeter of sediment, log Er , (J cm-3 ). Each reaction is distinguished by the colors of the lines and their labels, which are abbreviations including just the electron donor and acceptor. Discrete dots rather than a continuous line represent the energy available at Juan de Fuca for methane oxidation by goethite because this reaction is endergonic at some depths. Similarly, methane oxidation by sulfate does not appear in the same plot because it is only exergonic for the bottom few meters in Juan de Fuca sediments.

0

50

Depth , mbsf

SPG 100

150

PM

200

JdF

250 2

3

4

5

6

7

8 3

log (cells / cm )

9

10

Fig. 13.2: Concentration of cells per cubic centimeter in South Pacific Gyre (SPG), Peru Marin (PM) and Juan de Fuca (JdF) sediments as a function of depth.

lower energy yields at the other two sites would result in commensurately lower microbial cell numbers. However, the number of cells in the respective sediment columns, shown in 󳶳 Fig. 13.2, do not reflect this. In fact, the cell counts in JdF and PM sediments are several orders of magnitude higher at all depths than those at SPG, the site of the most exergonic reactions considered. This discrepancy can be readily reconciled by recasting the Gibbs energy calculations presented in 󳶳 Fig. 13.1 (a) to reflect the chemical and physical characteristics at each of these three sites. In 󳶳 Fig. 13.1 (b), the reaction energetics are plotted as energy densities in units of Joules per cm3 of sediment. Energy densities of the 𝑟th reaction in a cm3 of sediment, 𝐸𝑟 , are calculated using:

󵄨󵄨 Δ𝐺 󵄨󵄨 󵄨 󵄨 𝐸𝑟 = 9.65 × 10−4 ⋅ 󵄨󵄨󵄨 𝑟 󵄨󵄨󵄨 [𝑖]𝜙 󵄨󵄨 𝜈𝑖 󵄨󵄨

(1)

where 𝑣𝑖 and [𝑖] stand for the stoichiometric coefficient and molal concentration, respectively, of the 𝑖th limiting electron donor or acceptor and 𝜙 denotes porosity (unitless). The conversion factor 9.65 × 10−4 in Eq. (1) corresponds to the number of kg of H2 O in a cm3 of seawater. Because either the electron donor or acceptor will be a limiting reactant per volume of fluid, the concentration and stoichiometric coefficient of

286 | 13 Energetic constraints on life in marine deep sediments this limiting nutrient was used for values of 𝑣𝑖 and [𝑖] in Eq. (1) to generate the panels in 󳶳 Fig. 13.1 (b). In order to carry out these calculations, the activities of all reactants and products were held constant, simulating a steady state condition. It can be seen in 󳶳 Fig. 13.1 (b) that there is far more energy available at JdF and PM than at SPG. This is due to the fact that although the catabolic reactions considered at SPG are more exergonic per mole of electron transferred, the concentrations of electron acceptors, and, more importantly, electron donors are far lower there than they are at the other two sites. Within the sediment columns at each site, it can be seen that the energy available from different reactions vary by about 6 orders of magnitude per cm3 of sediment. For example, the energy available from Reaction (3), organic matter oxidation by MnO2 , is the most energy yielding reaction at JdF throughout most of the sediment column, yielding 5–0.1 J cm−3 , whereas methane oxidation by FeOOH, where it is exergonic, yields anywhere from 0.04 to < 10−6 J cm−3 . Furthermore, due to concentration gradients in the sediments at all sites, values of 𝐸𝑟 can vary by several orders of magnitude as a function of depth. This is most notable at JdF and PM, but also for the reactions in SPG in which H2 is an electron donor. It should also be pointed out that the most energetic reactions at a particular site vary with depth, suggesting where physiological changes in the microbial communities in these sediment might be observed. This is illustrated for the reactions represented by crossing lines in 󳶳 Fig. 13.1 (b) for PM sediments and, more conspicuously, for JdF. Acetate was used to represent organic matter in all of the Gibbs energy calculations. In the case of PM, acetate concentrations in the sediment were reported in [42], but for JdF, the acetate concentration was taken to be equal to the concentration of DOC. At SPG, acetate concentrations were estimated using the amount of TOC in the sediments (see Section 13.3.2). Acetate is a common organic compound in many natural settings [47], but using a suite of organic compounds would be more reflective of the organic milieu in marine sediments. Using acetate exclusively does have energetic implications, which are shown in 󳶳 Fig. 13.3 (a). For example, values of Δ𝐺𝑟 for sulfate reduction coupled to the oxidation of acetate (Reaction 5 in 󳶳 Table 13.1) and formate (Reaction 7 in 󳶳 Table 13.1) are shown as a function of depth in JdF sediments. Although these compounds are similar, the Gibbs energies of their oxidation per electron nearly span that of the known range for organic compounds [48]. Throughout the sediment column at JdF, formate oxidation by sulfate is nearly twice as exergonic as the analogous reaction with acetate. Furthermore, it can be seen in 󳶳 Fig. 13.3 (a) that the form of the curve representing the energetics of formate oxidation has more pronounced curvature than that for acetate. This is due to the fact that though the same concentration of organic matter is used in both sets of calculations, the stoichiometric coefficients for acetate and formate oxidation reactions are 1 and 4, respectively (see Eq. (3) in Computational Methods). It should be noted that the calculations carried out with organic carbon are an attempt to assess the energetics of readily accessible organic matter. Taking into account the total organic matter (OM) at each site, which includes dissolved and particulate OM, would increase the amount of energy avail-

13.4 Overview of catabolic potential | 287

(a)

-

Acetate: CH COO + SO 3

-

+

24

Formate: 4HCOO + H + SO

-

---> HS + 2HCO 24

(b)

3

-

+

CH COO- + 8FeOOH + 15H --> 2HCO + 8Fe

-

-

---> HS + 4HCO

3

3

2+

3

+ 12H O 2

0

0

JdF 50

Depth , mbsf

Depth , mbsf

50

Acetate

100

Formate

150

150

200

200

250

250 -5

-10

-15

-20

-25 - -1

ΔG , kJ (mol e ) r

-30

Anhydrous ferrihydrite

Goethite

100

JdF 0

-10

-20

-30

-40

-50

- -1

ΔG , kJ (mol e ) r

Fig. 13.3: Comparison of the energetic consequence of representing a) organic matter as acetate versus formate and b) goethite versus anhydrous ferrihydrite in Gibbs energy calculations in Juan de Fuca sediments.

able from OM degradation, but the rates at which this particulate OM is converted into usable OM, is a complex function of numerous environmental variables [49] that is beyond the scope of this chapter. Similarly, the minerals goethite and pyrolusite were used in the Gibbs energy calculations for reactions in which oxidized iron and manganese serve as electron acceptors. Although the exact mineral phases in these sediments are not known, combustion analyses show the presence of iron- and manganese-bearing minerals [40, 42, 45]. As with the choice of acetate to represent OM discussed above, there is an energetic consequence to representing Fe(III) and Mn(IV) as goethite and pyrolusite. This is illustrated in 󳶳 Fig. 13.3 (b), which depicts the Gibbs energy of acetate oxidation by two different phases of Fe(III) minerals, goethite and anhydrous ferrihydrite (FeOOH), in JdF sediments. Despite having the same chemical formula, values of Δ𝐺𝑟 are approximately three times more exergonic for the reaction with anhydrous ferrihydrite than the analogous one with goethite, yet this is only one of many potential mineral phases in which Fe(III) could exist. Furthermore, it should be noted that some mineral phases may not be readily accessible if they are physically separated from the microorganisms that can use them, or by having a crystalline structure that does not readily lend itself to microbial utilization. Clearly, how one represents mineral phases and organic matter can have a dramatic effect on the energetics of microbially metabolized reactions. See Computational Methods for the source of thermodynamic data for anhydrous ferrihydrite and pyrolusite.

288 | 13 Energetic constraints on life in marine deep sediments

13.5 Comparing deep biospheres The energetics of OM degradation with three different electron acceptors are compared in 󳶳 Fig. 13.4. The energetic comparison is made on a per mole e- basis in 󳶳 Fig. 13.4 (a)– (c) and on a per cm3 sediment basis in 󳶳 Fig. 13.4 (d)–(f). It is immediately apparent that the energy yields differ demonstrably at the three sites and that their relative positions change dramatically depending on the normalization procedure. For example, OM oxidation with Mn(IV)(󳶳 Fig. 13.4 (a)) yields similar energies (per mol e− ) at all three sites for the first 70 m of sediment. Below that, the energy yield at JdF decreases from −81 kJ/mol e− to −74 kJ/mol e− at ∼250 mbsf, before sharply increasing to −80 kJ/mol e− at the sediment-water interface [42]. In comparison, the energy yields for this reaction at PM remain fairly constant at approximately −82 kJ/mol e− over the entire depth of the sediment column. The differences in sediment thickness at the three sites permits only an incomplete comparison of energetics with depth. It is striking that the energy yield for OM oxidation with Mn(IV) at JdF is consistently the least exergonic on a per mol e− basis (󳶳 Fig. 13.4 (a)), but the most exergonic on a per cm3 sediment basis (󳶳 Fig. 13.4 (d)). This same shift in relative position among the three sites is seen in 󳶳 Fig. 13.4 (b) and (e) for OM oxidation with Fe(III): on a per mol e− basis, SPG is the most exergonic throughout the sediment columns and JdF is the least exergonic. However, these positions reverse on a per cm3 sediment basis with JdF the most and SPG the least exergonic. For OM oxidation coupled to sulfate reduction (󳶳 Fig. 13.4 (c) and (f)), JdF is the most exergonic regardless of the normalization procedure used, but SPG and PM switch relative positions. The question of which normalization procedure should be used to present the results of energy calculations depends on the subject matter. The amount of energy available per cm3 sediment is arguably a far better indicator of biomass in sediments than when energetic potential is presented per mol e− . This point is best realized by comparing the energetic profiles in 󳶳 Fig. 13.4 to the abundance of microorganism in each sediment column (󳶳 Fig. 13.2). Despite the similar energy yields for the catabolic reactions considered at all three sites on a per mol e− basis (󳶳 Fig. 13.4 (a)–(c)), there are several orders of magnitude less biomass in SPG than the other sites. Similar to the energetics displayed in 󳶳 Fig. 13.1, this discrepancy can be resolved by presenting these energetic calculations in terms of the amount of energy available per cm3 . 󳶳 Figure 13.4 (d)–(f) shows the energy density (𝐸𝑟 ) as a function of depth for the same OM oxidation reactions considered in 󳶳 Fig. 13.4 (a)–(c). For all three reactions, the Opposite page: Fig. 13.4: Direct comparison of the energy available from organic matter oxidation in South Pacific Gyre (SPG), Peru Margin (PM) and Juan de Fuca (JdF) sediments in units of (a)–(c) kilojoules per mole of electron transferred, kJ (mol e− )−1 , and (d)–(f) logarithm of the energy available per cubic centimeter of sediment, log (J cm−3 ), for (a,b) MnO2 , (c,d) FeOOH and (e,f) SO4 2− serving as electron acceptors.

13.5 Comparing deep biospheres | 289

(a)

-

+

-

CH COO + 4MnO + 7H --> 2HCO + 4Mn 3

2

2+

(d)

+ 4H O

3

2

0

-

-

+

CH COO + 4MnO + 7H --> 2HCO + 4Mn 3

2

2+

+ 4H O

3

2

0

Depth , mbsf

SPG 50

50

100

100

150

150

PM

200

SPG

PM

200

JdF

JdF 250

250 -74

-76

-78

-80

-82

-84

-7

-6

-5

(b)

+

CH COO + 8FeOOH + 15H --> 2HCO + 8Fe

2+

3

+ 12H O 2

(e)

-1

0

50

-

+

-

CH COO + 8FeOOH + 15H --> 2HCO + 8Fe 3

2+

3

+ 12H O 2

SPG

50

SPG

100

100

150

150

PM

PM 200

200

JdF

JdF 250

250 0

-5

-10

-15

-8

-20

-7

-6

(c)

CH COO + SO 3

0

24

(f)

3

-

CH COO + SO 3

0

50

50

SPG

100

-2

-1

0

24

-

---> HS + 2HCO

3

SPG

100

PM

150

150

PM

200

-3

r

-

---> HS + 2HCO

-4

log E (J/cm )

r

-

-5

3

- -1

ΔG , kJ (mol e )

Depth , mbsf

1

0

0

Depth , mbsf

-2

r

-

3

-3

log E (J/cm )

r

-

-4

3

- -1

ΔG , kJ (mol e )

200

JdF

JdF 250

250 -2

-4

-6

-8

-10

-12 - -1

ΔG , kJ (mol e ) r

-14

-16

-10

-8

-6

-4

-2 3

log E (J/cm ) r

0

290 | 13 Energetic constraints on life in marine deep sediments amount of energy available per cm3 spans about 7 orders of magnitude between the sites, with SPG at the low end and JdF at the high end of the spectrum. These plots reproduce, at least qualitatively, if not semi-quantitatively, the microbial abundances in the sediment columns (󳶳 Fig. 13.2). It is worth noting that even in the most energyrich JdF sediments, the OM oxidation reaction only yields about 0.1 Joules of energy per cubic centimeter of sediment. Note also that these are only some of the energy yielding reactions at each site. In order to compare the total energy supply in each setting, one would have to know the proportional use of each ED and EA. For example, OM can be degraded by many EAs, providing large differences in energy availability. Furthermore, if mineral phases such as MnO2 that are particularly energetic EAs are not accessible to microorganisms, then other, less energetic reactions such as sulfate reduction could be more typical catabolic strategies in these sediments.

13.6 Electron acceptor utilization The types and relative abundance of electron acceptors encountered in marine sediments varies as a function of geographic location and depth. It is generally assumed that the order in which they are consumed by respiring microorganisms is O2 , NO−3 , Mn(VI), Fe(III), SO2− 4 and CO2 (e.g. [50–52]). This explains the rapid disappearance of O2 , then nitrate and so on as a function of depth in organic rich sediments [47]. The generally-accepted hypothesis for this phenomenon is that the sequence of EA utilization corresponds to the order of Gibbs energy yield of the corresponding organic matter oxidation reactions. That is, a more exergonic reaction (i.e. having a larger value of −Δ𝐺𝑟 ) takes place before less exergonic ones. Another way of saying this is that given a selection of electron acceptors, a microbial community will use the ones that provide the largest amount of energy first, the second most energetic second, and so on. However, because the exergonicity of reactions is a function of environmental parameters such as temperature, pressure and concentrations of reactants and products, the order of what is the most energetic electron acceptor (or donor) can deviate from that listed above. This is clearly shown in 󳶳 Fig. 13.5 (a), which depicts the Gibbs energies of OM transformation by MnO2 , FeOOH and sulfate as well as methanogenesis in JdF sediments. Throughout the sediment column, OM oxidation with Mn(IV) is the most exergonic, consistently yielding approximately −80 kJ / mol e− . The other three reactions are far less exergonic (0 to −20 kJ/ mol e− ), and their relative positions oscillate as a function of depth. For example, OM oxidation with Fe(III) is, at different depths, the most and least exergonic of these three reactions. Energetic comparisons become even more complex by examining the amount of energy per cm3 sediment for the same set of reactions (󳶳 Fig. 13.5 (b)). Again, the reaction with OM + Mn(IV) yields the most energy. A focus on the other three reactions reveals that the energy yield from OM oxidation with Fe(III) is very similar to that from

| 291

13.6 Electron acceptor utilization

(a)

(b) 0

0

50

Depth , mbsf

50

MnO

FeOOH + OM

2

+ OM 100

MnO

100

2

+ OM 150

FeOOH + OM

150

methanogenesis

methanogenesis 200 SO

24

200

+ OM

250

SO

24

+ OM

250 0

-20

-40

-60

-80 - -1

ΔG , kJ (mol e ) r

-100

-3

-2.5

-2

-1.5

-1

-0.5

0

0.5

1

3

log E (J/cm ) r

Fig. 13.5: Catabolic energy potential for organic matter degradation in Juan de Fuca sediments coupled to iron (FeOOH), manganese (MnO2 ) and sulfate reduction and acetoclastic methanogenesis in units of (a) kilojoules per mole of electron transferred, kJ (mol e− )−1 , and (b) Joules per cubic centimeter of sediment, J cm−3 .

methanogenesis throughout the sediment column, but that from OM coupled to sulfate is by far the most exergonic of the three reactions. This is noteworthy because in the absence of accessible MnO2 (which seems to be the case since the Mn profiles there suggest that MnO2 reduction is not especially common throughout the sediment column – see Appendix), iron reducers, sulfate reducers and methanogens could alternate as the dominant microbial communities in these sediments – considerably upsetting the standard EA ordering listed above. In this case, methanogens and Fe(III) and sulfate reducers would be competing as the most active organisms in JdF sediments. By contrast, the Gibbs energies of H2 oxidation seem to follow the standard EA trend. As an example, see the energy profiles in SPG sediments depicted in 󳶳 Fig. 13.1 (a). However, there is another story here as well. Per mole of electron, the energy available from H2 and OM oxidation in SPG follows the canonical order of EAs listed above with reactions with H2 being slightly more exergonic than those with OM for the same EA. Also, per mole of electron, the oxidation of Fe2+ and Mn2+ by O2 , respectively, fall into the energy-rich and energy-poor groups of reactions considered at SPG. Once the energy availability for this site is recast into units of J per cm3 , however, the most energy-rich reactions are, by a wide margin, Fe2+ and Mn2+ oxidation by O2 throughout the sediment column. Simply put, there are far higher concentrations of these two electron donors than H2 or OM in SPG sediments, so, per cm3 , there is much more energy to be had from oxidizing them.

292 | 13 Energetic constraints on life in marine deep sediments

13.7 Energy demand Although the focus of this chapter is energy supply in the deep biosphere, the results presented above also have a bearing on microbial energy demand, or at least how it is sometimes considered. This is because the minimum quantity of energy that microbes must extract from catabolic reactions in order to remain active is often stated in units of kJ per mole. This is alternatively referred to as a minimum Gibbs energy (Δ𝐺min , [53]), the biological energy quantum, (Δ𝐺BQ , [54]), or a threshold Gibbs energy (Δ𝐺thr , [55]), and reported values span a very wide range from −3.8 to −49.6 kJ per reaction turnover [54]. According to these authors, microorganisms can only catalyze reactions that yield at least, e.g. −20 kJ mol−1 [56], an amount of energy related to the mechanism of ATP synthesis. However, the energetic analysis detailed above illustrates a potential pitfall of using this approach. According to this commonly-used bioenergetic minimum theory, many reactions cannot support microbial communities, because their yield is too low on a per mol of substrate basis. However, some of these reactions – as shown above – appear to be very energy-rich on a per cm3 of sediment basis. Based on bioenergetic minimum theory, OM oxidation with Fe(III) and methanogenesis in JdF sediments, for example, should not occur. However, per cm3 of sediment, these reactions yield several orders of magnitude more energy than the knallgas reaction (O2 + H2 ) in SPG, which, according to bioenergetic theory, yields sufficient energy for microorganisms to remain active. It is difficult to reconcile why microorganisms that have access to nearly 1 J cm−3 from one reaction would not use it, while those that only have 10−7 J cm−3 available from another would. The reason that bioenergetic theory leads to this counterintuitive result is that it assumes that there is a fixed stoichiometry between catabolism and ATP synthesis, the ultimate source of Δ𝐺min , Δ𝐺BQ , Δ𝐺thr , or whatever label is used to represent bioenergetic demand. That is, microorganisms ultimately get their energy from catalyzing redox reactions, but ATP synthesis is a dehydration reaction; as such, they cannot be uniquely, stoichiometrically coupled (see [57]). On a more intuitive level, in systems with extremely low levels of H2 , as in SPG sediments, it seems unlikely that microbes could sustain much activity powered by hydrogen oxidation. This is despite its very negative values of Δ𝐺𝑟 (on a per mol basis). Conversely, despite their much more modest values of Δ𝐺𝑟 , catabolic strategies such as FeOOH + OM and methanogenesis in JdF are much more likely to fuel a microbial population due to very high concentrations of OM present in the sediments there. Perhaps the minimum energy a microorganism needs should be normalized like some maintenance energies are, in units of Joules per microorganism per day (e.g [58, 59]) rather than trying to estimate how much ATP needs to be made per mole of substrate. Although this is beyond the scope of the present chapter, it is the focus of a study in progress [60].

13.9 Computational methods |

293

13.8 Concluding remarks In a particular sediment at a particular depth, it is not only the most energy-yielding redox reactions that are catalyzed by microorganisms. Multiple, overlapping respiration and fermentation pathways can coexist [34, 44], microniches where the chemical conditions differ from the bulk sediments could support alternative pathways and very short-lived intermediate chemical species could fuel catabolic reactions. However, at the most fundamental level, the energetics calculations presented above indicate which catabolic reactions are thermodynamically favored to occur in deep marine biosphere environments, and, in particular, at what depths in the sediment columns these reactions are favored. By characterizing which catabolic strategies are possible and which are not, a basic understanding of the biogeochemical dynamics of the deep biosphere can be obtained. It is stressed that these conclusions are only possible because the environmental variables that distinguish each sediment column were taken into account in the calculations. If only the standard state Gibbs energies would have been considered, then all three environments would have looked identical. Furthermore, by taking into account the concentrations of the limiting electron donor or acceptor per cm3 of sediment, the variation of available energy from particular reactions in each of these environments is revealed. By computing this energy density, a more nuanced understanding of the deep biosphere emerges: – The most exergonic reactions per mole of substrate may in fact provide very little energy per cm3 sediment. – Reactions that are not particularly exergonic per mole of substrate may provide the most energy per cm3 sediment. – The relative amounts of biomass can be accounted for (but only if the energetics are quantified per volume or mass, not per mole). – The order of EA usage by microbial communities may deviate from the canonical order of O2 , NO−3 , MnO2 , FeOOH, SO2− 4 , CO2 and disproportionation. Perhaps one of the greatest strengths of the approach taken in this study is that it can be applied to any environment in order to obtain a basic understanding of bioenergetic potential. Calculations of the Gibbs energy of reaction provide a fundamental assessment of what is biologically possible in a given environment and a means to compare how different geochemical variables govern (very) different ecosystems.

13.9 Computational methods There is no easy way to categorize the energetic potential of particular electron donors and acceptors without calculating the Gibbs energy of the reaction (Δ𝐺𝑟 ) at the prevailing temperature, pressure and composition. In this study, values of Δ𝐺𝑟 are calcu-

294 | 13 Energetic constraints on life in marine deep sediments lated using

Δ𝐺𝑟 = −𝑅𝑇 ln

𝐾𝑟 , 𝑄𝑟

(2)

where 𝐾𝑟 and 𝑄𝑟 refer to the equilibrium constant and reaction quotient of the indicated reaction, respectively, 𝑅 represents the gas constant and 𝑇 denotes temperature in Kelvin. Values of 𝐾𝑟 were calculated using the revised-HKF equations of state [61–63], the SUPCRT92 software package [64], and thermodynamic data taken from [65–69]. Values of 𝑄𝑟 were calculated using 𝑣

𝑄𝑟 = ∏ 𝑎𝑖 𝑖 ,

(3)

𝑖

where 𝑎𝑖 stands for the activity of the 𝑖th species and 𝑣𝑖 corresponds to the stoichiometric coefficient of the 𝑖th species in the reaction of interest. Molalities of the 𝑖th species, 𝑚𝑖 , were converted into activities using individual activity coefficients of the 𝑖th species (𝛾𝑖 ), 𝑎𝑖 = 𝑚𝑖 𝛾𝑖 . (4) Values of 𝛾𝑖 were in turn computed as a function of temperature and ionic strength using an extended version of the Debye–Hückel equation [70]. Complete reaction turnover refers to the situation in which the number of moles of reactants and products processed in a chemical reaction is equal to their stoichiometric coefficients. For Reaction (10) in 󳶳 Table 13.1, complete reaction turnover would − − indicate that 1 mole each of CH4 and SO2− 4 are consumed and 1 mole each of HCO3 , HS and H2 O are produced.

13.9.1 Thermodynamic properties of anhydrous ferrihydrite and pyrolusite The standard state thermodynamic properties of anhydrous ferrihydrite at 25 °C were taken from [71], who measured these properties for a slightly hydrated form of this mineral, FeOOH*0.027H2 O, and then estimated these properties for the completely dewatered version by assuming that the associated water molecules have the same thermodynamic properties as liquid H2 O. In the current study, this procedure was expanded to the standard state isobaric heat capacity (𝐶𝑜𝑝 ) data reported by [71] in order to calculate the other standard state thermodynamic properties of anhydrous ferrihydrite as a function of temperature. These values of 𝐶𝑜𝑝 were regressed as a function of temperature (not shown) using the Maier–Kelley equation [72] resulting in the following heat capacity power function coefficients: 𝑎 = 36.387 J K−1 mol−1 , 𝑏 = 0.15358 J K−2 mol−1 , 𝑐 = −424,514 J K mol−1 . The standard state thermodynamic properties of pyrolusite, MnO2 , were taken from [73] and the Maier–Kelley heat capacity coefficients were regressed as a function of temperature using the Maier–Kelley equation from heat capacity data taken

Appendix |

295

from the same source. The resulting values of the 𝐶𝑜𝑝 power function coefficients are 𝑎 = 12.55 J K−1 mol−1 , 𝑏 = 0.009761 J K−2 mol−1 and 𝑐 = −210,500 J K mol−1 . A caveat of using minerals in thermodynamic calculations is that their activities are often taken to be 1, which was also done in this study. Therefore, the resulting Gibbs energies, or energy per cm3 of sediment, do not account for the fact that a cm3 of sediment might contain > 99% or < 1% of this mineral. If the mass of the particular minerals used in this study per cm3 were known in marine sediments, then this quantity could be used to constrain the energy density of these reactions per cm3 sediment. Because they are not known, the limiting substrate for these reactions was always taken to be the aqueous electron donor and never the mineral.

Appendix The concentrations of the species appearing in the catabolic reactions considered in this study (󳶳 Table 13.1), which are required to calculate the Gibbs energies of reaction, are shown as a function of depth for the three study sites in 󳶳 Figs. 13.A1–13.A3. The sources of these data are stated in Section 13.3. Because the concentrations for all species have not continuously been determined as a function of depth, the concentrations of some species have been estimated over the some depth ranges. In most cases, a species concentration missing at a particular depth was assumed to be between the concentrations measured above and below it. However, for several species, larger gaps in the concentration record were filled in as follows. In JdF sediments, concentration of NH+4 from 121–170 mbsf were taken to be a linear interpolation between the concentrations of NH+4 reported above and below this depth interval, the bounding values. In PM sediments, the concentration of SO2− 4 from 41–66 mbsf were taken to be 10 μM because the concentrations were below detection (not stated), the concentration of CH4 between 131–156 mbsf were taken to be a mean of the bounding values and pH was interpolated between values measured at the top and bottom of the sediment column. In SPG sediments, at many depths, H2 concentrations were below detection (2.8 nM), so a nominal values of 1 nM was used and between 43–63 mbsf and 74–76 mbsf and the O2 concentrations were calculated from a 4𝑡ℎ order polynomial whose regression coefficients were determined from O2 concentration at the other sediment depths. The results of all of these estimates and the measured data are shown in 󳶳 Fig. 13.A1–13.A3. Furthermore, the cell counts for SPG sediments were interpolated at many depths. In particular, between 42 and 76 mbsf, a power-law function was applied to the rest of the cell count data to estimate cell numbers in this interval, which is plotted in 󳶳 Fig. 13.2.

296 | 13 Energetic constraints on life in marine deep sediments

0

5

10

15

20

2+

2-

4

30 0

25

5

10

15

20

25

2+

[Fe ] (μmol/kg)

[SO ] (mmol/kg)

DIC (mmol/kg)

30 0

50

100

[Mn ] (μmol/kg) 150

0

50 100 150 200 250 300

0

Depth (mbsf)

50

100

150

200

250

+

[CH ] (log mmol/kg) -6

-5

-4

-3

NH (μmol/kg)

DOC (mmol/kg)

4

-2

0

5

10

15

4

20

400 800 1200 1600 2000

0

Depth (mbsf)

50

100

150

200

250

Fig. 13.A1: Selected geochemical data for Juan de Fuca sediments.

pH 6 6.5 7 7.5 8 8.5 9 9.5

Appendix | 297 [CH ] μmol / kg

[DIC] mmol / kg 5

10

15

20

25

[acetate] mmol / kg

4

0

1000

2000

0

3000 0

2

4

6

8 10 12 14

Depth (mbsf)

50

100

150

2+

[Fe ] μmol / kg 0

3

6

9

12

2+

[Mn ] μmol / kg 15 0

2

4

6

8

10 12

2-

[SO4 ] mmol / kg 0

10

20

30

0

Depth (mbsf)

50

100

150

Fig. 13.A2: Selected geochemical data for Peru Margin sediments.

298 | 13 Energetic constraints on life in marine deep sediments

DIC 1.8 0

2-

[SO ] (mmol/kg)

2.2 2.4 2.6 2.8 25

2

2+

Fe (μmol / kg)

4

26

27

28

29 2

3

4

5

6

7

8

2+

Mn (μmol / kg) 9 2 4

6 8 10 12 14 160

10

Depth (mbsf)

20

30

40

50

60

70

80 -

NO (μmol / kg) 3

O (μmol/kg) 2

30 35 40 45 50 55 60 65 60 80 100120140160180200 0 0

[H ] nmol / kg

pH

2

5 10 15 20 25 30 35 7 7.1 7.2 7.3 7.4 7.5 7.6 7.7

10

Depth (mbsf)

20

30

40

50

60

70

80

Fig. 13.A3: Selected geochemical data for South Pacific Gyre sediments.

TOC, wt% 0.06 0.12 0.18 0.24 0.3

References |

299

Acknowledgements Financial assistance was provided by the Center for Dark Energy Biosphere Investigations (C-DEBI), the NASA Astrobiology Institute – Life Underground (NAI-LU) and the National Science Foundation grant OCE-1207874. This is C-DEBI contribution 169 and NAI-LU contribution 001.

References [1] [2] [3] [4] [5] [6] [7] [8]

[9] [10] [11] [12] [13] [14] [15] [16]

[17]

Whitman WB, Coleman DC, Wiebe WJ. Prokaryotes: The unseen majority. Proc Natl Acad Sci USA 95 (1998), 6578–6583. Schippers A, Neretin LN, Kallmeyer J, et al. Prokaryotic cells of the deep sub-seafloor biosphere identified as living bacteria. Nature 433 (2005), 861–864. Parkes RJ, Cragg BA, Bale SJ, et al. Deep bacterial biosphere in Pacific Ocean sediments. Nature 371 (1994), 410–413. D’Hondt S, Jørgensen BB, Miller DJ, et al. Distributions of microbial activities in deep subseafloor sediments. Science 306 (2004), 2216–2221. Edwards KJ, Bach W, McCollom TM. Geomicrobiology in oceanography: microbe–mineral interactions at and below the seafloor. Trends Microbiol 13 (2005), 449–456. Santelli CM, Orcutt BN, Banning E, et al. Abundance and diversity of microbial life in ocean crust. Nature 453 (2008), 653–657. Cowen JP, Giovannoni SJ, Kenig F, Johnson HP, Butterfield D, Rappé MS. Fluids from aging ocean crust that support microbial life Science 299 (2003), 120–123. Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16 213– 16 216. Jørgensen BB. Shrinking majority of the deep biosphere. Proc Natl Acad Sci USA 109 (2012), 15 976–15 977. Jørgensen BB, D’Hondt S. A starving majority deep beneath the seafloor. Science 314 (2006), 932–934. Jørgensen BB, Boetius A. Feast and famine – microbial life in the deep-sea bed. Nature Rev Microbiol 5 (2007), 770–781. Røy H, Kallmeyer J, Adhikari RR, Pockalny R, Jørgensen BB, D’Hondt S. Aerobic microbial respiration in 86-million-year-old deep-sea red clay. Science 336 (2012), 922–925. Van Briesen JM. Evaluation of methods to predict bacterial yield using thermodynamics. Biodegradation 13 (2002), 171–190. McCollom TM. Geochemical constraints on sources of metabolic energy for chemolithoautotrophy in ultramafic-hosted deep-sea hydrothermal systems. Astrobiology 7 (2007), 933–950. McCollom TM. Geochemical constraints on primary productivity in submarine hydrothermal vent plumes. Deep-Sea Res Part I Oceanogr Res Pap 47 (2000), 85–101. McCollom TM, Shock EL. Geochemical constraints on chemolithoautotrophic metabolism by microorganisms in seafloor hydrothermal systems. Geochim Cosmochim Acta 61 (1997), 4375– 4391. LaRowe DE, Dale AW, Regnier P. A thermodynamic analysis of the anaerobic oxidation of methane in marine sediments. Geobiology 6 (2008), 436–449.

300 | 13 Energetic constraints on life in marine deep sediments [18] Shock EL, McCollom TM, Schulte MD. Geochemical constraints on chemolithoautotrophic reactions in hydrothermal systems. Orig Life Evol Bios 25 (1995), 141–159. [19] Shock EL, Holland ME. Geochemical energy sources that support the subseafloor biosphere. The subseafloor biosphere at mid-ocean ridges. In: Wilcock WSD, DeLong EF, Kelley DS, Baross JA, Cary SC, eds. Geophysical Monograph 144, American Geophysical Union, 153–165. 2004. [20] Amend JP, McCollom TM, Hentscher M, Bach W. Catabolic and anabolic energy for chemolithoautotrophs in deep-sea hydrothermal systems hosted in different rock types. Geochim Cosmochim Acta 75 (2011), 5736–5748. [21] Rogers KL, Amend JP. Energetics of potential heterotrophic metabolisms in the marine hydrothermal system of Vulcano Island, Italy. Geochim Cosmochim Acta 70 (2006), 610–6200. [22] Rogers KL, Amend JP, Gurrieri S. Temporal changes in fluid chemistry and energy profiles in the Vulcano island hydrothermal system. Astrobiology 7 (2007), 905–932. [23] Skoog A, Vlahos P, Rogers KL, Amend JP. Concentrations, distributions, and energy yields of dissolved neutral aldoses in a shallow hydrothermal vent system of Vulcano, Italy. Org Geochem 38 (2007), 1416–1430. [24] Rogers KL, Amend JP. Archaeal diversity and geochemical energy yields in a geothermal well on Vulcano Island, Italy. Geobiology 3 (2005), 319–332. [25] Amend JP, Rogers KL, Shock EL, Gurrieri S, Inguaggiato S. Energetics of chemolithoautotrophy in the hydrothermal system of Vulcano Island, southern Italy. Geobiology 1 (2003), 37–58. [26] Price RE, LaRowe DE, Amend JP. Geochemistry and bioenergetic potential of a shallow-sea hydrothermal vent system off Panarea Island, Aeolian Islands, Italy. Submitted, GCA 2013. [27] Shock EL, Holland M, Meyer-Dombard D, Amend JP, Osburn GR, Fischer TP. Quantifying inorganic sources of geochemical energy in hydrothermal ecosystems, Yellowstone National Park, USA. Geochim Cosmochim Acta 74 (2010), 4005–4043. [28] Spear JR, Walker JJ, McCollom TM, Pace NR. Hydrogen and bioenergetics in the Yellowstone geothermal ecosystem. Proc Natl Acad Sci USA 102 (2005), 2555–2560. [29] Vick TJ, Dodsworth JA, Costa KC, Shock EL, Hedlund BP. Microbiology and geochemistry of Little Hot Creek, a hot spring environment in the Long Valley Caldera. Geobiology 8 (2010), 140–154. [30] Costa KC, Navarro JB, Shock EL, Zhang CL, Soukup D, Hedlund BP. Microbiology and geochemistry of great boiling and mud hot springs in the United States Great Basin. Extremophiles 13 (2009), 447–459. [31] Windman T, Zolotova N, Schwandner F, Shock EL. Formate as an energy source for microbial metabolism in chemosynthetic zones of hydrothermal ecosystems. Astrobiology 7 (2007), 873–890. [32] Inskeep W, Ackerman GG, Taylor WP, Kozubal M, Korf S, Macur RE. On the energetics of chemolithotrophy in nonequilibrium systems: case studies of geothermal springs in Yellowstone National Park. Geobiology 3 (2005), 297–317. [33] Inskeep WP, McDermott TR. Geomicrobiology of acid–sulfate–chloride springs in Yellowstone National Park. In: Inskeep WP, McDermott TR, eds. Geothermal Biology and Geochemistry in Yellowstone National Park: Thermal Biology Institute, Montanta State University, 143–162, 2005. [34] Wang G, Spivack AJ, D’Hondt S. Gibbs energies of reaction and microbial mutualism in anaerobic deep subseafloor sediments of ODP Site 1226. Geochim Cosmochim Acta 74 (2010), 3938– 3947. [35] Schrum HN, Spivack AJ, Kastner M, D’Hondt S. Sulfate-reducing ammonium oxidation: A thermodynamically feasible metabolic pathway in subseafloor sediment. Geology 37 (2009), 939– 942.

References |

301

[36] Bach W, Edwards KJ. Iron and sulfide oxidation within the basaltic ocean crust: Implications for chemolithoautotrophic microbial biomass production. Geochim Cosmochim Acta 67 (2003), 3871–3887. [37] Cowen JP. The microbial biosphere of sediment-buried oceanic basement. Res Microbiol 155 (2004), 497–506. [38] Boettger J, Lin H-T, Cowen JP, Hentscher M, Amend JP. Energy yields from chemolithotrophic metabolisms in igneous basement of the Juan de Fuca ridge flank system. Chem Geol., submitted. [39] Thullner M, Dale AW, Regnier P. Global-scale quantification of mineralization pathways in marine sediments: A reaction-transport modeling approach. Geochem Geophys Geosys 10 (2009), 1–24. [40] Fisher AT, Urabe T, Klaus A, Scientists atE. Site U1301. Proceed Integrat Ocean Drill Prog 301 (2005), doi:10-2204/iodp.proc.301.106.005. [41] Fichtel K, Mathes F, Könneke M, Cypionka H, Engelen B. Isolation of sulfate-reducing bacteria from sedmients above the deep-subseafloor aquifer. Front Microbiol 3 No. 65 (2012). [42] D’Hondt S, Jørgensen BB, Miller DJ, et al. Leg 201 Summary. Proc ODP 2003;201. [43] Jørgensen BB, D’Hondt SL, Miller DJ. Leg 201 Synthesis: Controls on micorbial communities in deeply buried sediments. Proc IODP Scientific Results 201 (2003), 45. [44] Wang G, Spivack AJ, Rutherford S, Manor U, D’Hondt S. Quantification of co-occurring reaction rates in deep subseafloor sediments. Geochim Cosmochim Acta 72 (2008), 3479–3488. [45] D’Hondt S, Inagaki F, Alvarez Zarikian CA, Scientists atE. Site U1365. Proc IODP, 329, 2011. [46] D’Hondt S, Spivack AJ, Pockalny R, et al. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106 (2009), 11 651–11 656. [47] Sarmiento JL, Gruber N. Ocean Biogeochemical Dynamics. Princeton: Princeton University Press; 2006. [48] LaRowe DE, Van Cappellen P. Degradation of natural organic matter: A thermodynamic analysis. Geochim Cosmochim Acta 75 (2011), 2030–2042. [49] Arndt S, Jørgensen BB, LaRowe DE, Middelburg JBM, Pancost RD, Regnier P. Quantifying the degradation of organic matter in marine sediments: A review and synthesis. Earth Sci Rev 123 (2013), 53–86. [50] Claypool GE, Kaplan IR. The origin and distribution of methane in marine sediments. In: Kaplan IR, ed. Natural Gases in Marine Sediments. New York: Plenum Press, 99–139, 1974. [51] Froelich PN, Klinkhammer GP, Bender ML, et al. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim Cosmochim Acta 43 (1979), 1075–1090. [52] Stumm W, Morgan JJ. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters. 3rd ed. New York: John Wiley & Sons; 1996. [53] Schink B. Energetics of synthrophic cooperation in methanogenic degradation. Microbiology and Molecular Biology Reviews 61 (1997), 262–280. [54] Hoehler TM. Biological energy requirements as quantitative boundary conditions for life in the subsurface. Geobiology 2 (2004), 205–215. [55] Curtis GP. Comparison of approaches for simulating reactive solute transport involving organic degradation reactions by multiple terminal electron acceptors. Comp Geosci 29 (2003), 319– 329. [56] Schink B, Thauer RK. Energetics of syntrophic methane formation and the influence of aggregation. In: Lettinga G, Zehnder AJB, Grotenhuis JTC, Hulshoff Pol LW, eds. Granular Anaerobic Sludge; Microbiology and Technology. Proceedings of the GASMAT-workshop. Wageningen: Puduc, 2–17, 1988.

302 | 13 Energetic constraints on life in marine deep sediments [57] LaRowe DE, Dale AW, Amend JP, Van Cappellen P. Thermodynamic limitations on microbially catalyzed reaction rates. Geochim Cosmochim Acta 90 (2012), 96–109. [58] Kirchman DL, Hanson TE. Bioenergetics of photoheterotrophic bacteria in the oceans. Environ Microbio Reports 5 (2013), 188–199. [59] Marschall E, Jogler M, Henssge U, Overmann J. Large-scale distribution and activity patters of an extremely low-light-adapted population of green sulfur bacteria in the Black Sea. Environ Microbio 12 (2010), 1348–1362. [60] LaRowe DE, Amend JP. Catabolic rates, population sizes and doubling/replacement times of microorganisms in the deep subsurface, submitted to American Journal of Science. [61] Helgeson HC, Kirkham DH, Flowers GC. Theoretical prediction of thermodynamic behavior of aqueous electrolytes at high pressures and temperatures: 4. Calculation of activity coefficients, osmotic coefficients, and apparent molal and standard and relative partial molal properties to 600 °C and 5 kb. Amer J Sci 281 (1981), 1249–1516. [62] Tanger JC, Helgeson HC. Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures – Revised equations of state for the standard partial molal properties of ions and electrolytes. Amer J Sci 288 (1988), 19–98. [63] Shock EL, Oelkers E, Johnson J, Sverjensky D, Helgeson HC. Calculation of the thermodynamic properties of aqueous species at high pressures and temperatures – Effective electrostatic radii, dissociation constants and standard partial molal properties to 1000 °C and 5 kbar. J Chem Soc Faraday Trans 88 (1992), 803–826. [64] Johnson JW, Oelkers EH, Helgeson HC. SUPCRT92 – A software package for calculating the standard molal thermodynamic properties of minerals, gases, aqueous species, and reactions from 1 bar to 5000 bar and 0 °C to 1000 °C. Comput Geosci 18 (1992), 899–947. [65] Shock EL, Helgeson HC. Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures – Correlation algorithms for ionic species and equation of state predictions to 5 kb and 1000 °C. Geochim Cosmochim Acta 52 (1988), 2009– 2036. [66] Shock EL, Helgeson HC. Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures – Standard partial molal properties of organic species. Geochim Cosmochim Acta 54 (1990), 915–945. [67] Shock EL, Helgeson HC, Sverjensky D. Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures – Standard partial molal properties of inorganic neutral species. Geochim Cosmochim Acta 53 (1989), 2157–2183. [68] Sverjensky D, Shock EL, Helgeson HC. Prediction of the thermodynamic properties of aqueous metal complexes to 1000 °C and 5 kb. Geochim Cosmochim Acta 61 (1997), 1359–1412. [69] Schulte MD, Shock EL, Wood R. The temperature dependence of the standard-state thermodynamic properties of aqueous nonelectrolytes. Geochim Cosmochim Acta 65 (2001), 3919– 3930. [70] Helgeson HC. Thermodynamics of hydrothermal systems at elevated temperatures and pressures. Amer J Sci 267 (1969), 729–804. [71] Snow CL, Lilova KI, Radha AV, et al. Heat capacity and thermodynamics of a synthetic two-line ferrihydrite, FeOOH*0.027H2 O. J Chem Thermo 58 (2013), 307–314. [72] Maier CG, Kelley KK. An equation for the representation of high-temperature heat content data. J Amer Chem Soc 54 (1932), 3243–3246. [73] Robie RA, Hemingway BS. Low-temperature molar heat capacities and entropies of MnO2 (pyrolusite), Mn3 O4 (hausmanite) and Mn2 O3 (bixbyite) J Chem Thermodyn 17 (1985), 165–181.

Hans Røy

14 Experimental assessment of community metabolism in the subsurface 14.1 Introduction Dissipation of energy from the environment in order to sustain a high degree of order is one of the most basic characteristics of life. Without dissipation of energy there will obviously be no growth, but also no repair of spontaneous denaturation of proteins and nucleic acids, and no sustained membrane potential. Therefore, any living cell must dissipate a certain minimum amount of energy per unit time just to stay alive, a certain basal power requirement [1]. The magnitude of the basal power requirement of the prokaryotes in the deep biosphere is unknown and it is a topic of ongoing research in its own right. But if the basal power requirement is pragmatically set to zero, then knowledge of the actual metabolic rate can be used to estimate the maximum amount of energy that can be assigned to growth of the cells in the microbial community living in the deep biosphere. Such numbers are of fundamental importance to understand the potential for adaptation and evolution in the deep biosphere, and ultimately understand whether the buried community is merely the dying remnants of the organisms that were thriving in the surface environment, or if the microbial community living deeply below ground is an active and thriving part of the biosphere that just lives on a different time scale than Homo sapiens. It is presently not possible to determine the metabolic rate of individual cells in situ in the deep biosphere. Although the cells are scarce compared to the surface, measuring concentration gradients and calculating a flux can be used to assess the combined metabolic activity of the entire community because the flux is a direct result of consumption or production of a compound. Given the current difficulty of assigning physiological and ecological function to the myriad of different organisms in the deep biosphere, the energy flow at the community level provides a sensible starting point to understand microbial dynamics in the subsurface. The division of energy between the groups of organisms responsible for the individual steps in the degradation of organic matter can then be calculated from the thermodynamic equations if the concentrations of the intermediates are known [2], but such calculations are outside the scope here.

14.1.1 The energy source Most microbial communities in the marine deep biosphere are fueled by past photosynthesis. Exceptions exist, especially near plate boundaries and hot spots, where

304 | 14 Experimental assessment of community metabolism in the subsurface hydrothermal vents eject water laden with reducing agents from basalts or ultramafic rocks, and where hydrothermal re-charge penetrate into the basement (9). The extent and total volume of such lithoautotrophic systems is currently under debate. Some authors [3] assume that the volume of this potentially habitable zone in ocean crust might be the largest reservoir of microbial life on Earth. At other sites such ecosystems only thrive on narrow interfaces buried in the seabed and as defined hot spots at the sediment-water interface. Other mechanisms of nonbiological generation of disequilibrium that could fuel microbial life have been suggested, the two main mechanisms are abiotic generation of short-chain organic molecules from sedimentary organic matter due to thermochemical reactions [4, 6] and formation of hydrogen and oxidized radicals from cleavage of water [5] driven by radiation from the naturally occurring thorium and uranium decay-families. The hydrogen yield of radiolysis in sediment is unknown but calculations based on the hydrogen yield in pure water suggests that it is unlikely to contribute substantially to the metabolic activity in all but the most ancient and organic poor sediments [7]. Thus, the most abundant and widespread life in the subsurface is spread out in the vast volumes of the sediment pile [8] and subsides on a slow trickle of organic matter cleaved off old and refractory remains of long dead organisms. These communities subside for millions of years without input of primary produced organic material and without depleting the sediment for carbon. For the carbon budget to balance the combined community metabolism must be exceedingly low.

14.1.2 The carbon budget A first estimate of the maximum rates of microbial carbon oxidation in old marine sediments can be made from a simple mass balance, starting with young sediment at the top of the sediment pile and considering the progressive mineralization of the closed carbon pool as the sediment ages and becomes buried. Marine-dried surface sediment typically contains 0.1–5 % organic matter by weight. Thus, one cm3 of surface sediment contains roughly 200 μg organic carbon or 17 μmol. The mineralization rate in the top cm of a deep-sea sediment is in the order of 1 nmol C cm3 day−1 while the carbon oxidation rate in the surface layer of coastal sediments is near the range 10–100 nmol C cm3 day−1 . This apparently implies that all organic matter would be depleted within a few years after the supply of fresh material stop. But there is surprisingly little depletion of organic matter across the upper meters of marine sediments where the age of the buried material is in the range of thousands of years. This preservation of organic carbon is linked to a rapidly decreasing reactivity of the organic carbon as the sediment ages. The decrease in reactivity of the organic matter in marine sediments with depth can be described by a simple power law function. This function dictates that the reactivity decreases by one order of magnitude each time the age increases by one order of magnitude [10]. The original description by Middelburg was

14.1 Introduction

| 305

based on a compilation of marine organic matter from many different environments and ages from less than one to 35,000 years. Later studies of mineralization along depth in sediment columns have found the same power law relation [11–13], although with a steeper decrease in reactivity as time passes. The adherence to a power law is a mere observation with no mechanistic explanation. The relation has previously been described by mechanistic models that imply successive depletion of separate pools of organic matter with varying degradability in multi-G or reactive continuum models [14, 15]. But no such pools have ever been identified and the mechanistic basis for the loss of degradability during aging of organic matter remains to be resolved. Nevertheless, the power law allows us to explore the range of metabolic rates we should expect to encounter in the deep biosphere.

14.1.3 Distribution vertical of microbial metabolism the sediment pile

󳶳 Figure 14.1 shows two examples of the progression of metabolic activity from the sediment-water interface and into the sediment pile. 󳶳 Figure 14.1 (a) shows data from a eutrophic Danish estuary and cover sediment up to a few thousand years old. 󳶳 Figure 14.1 (b) stems from the extremely oligotrophic North Pacific Gyre and the data cover sediment from 3 to 86 million years old. In both illustrated cases the oxidation rate of carbon decreases according to a power function. 󳶳 Figure 14.1 shows that the rates of microbial respiration just a few meters into the sediment pile is orders of magnitude below that in the surface although the concentration of organic carbon does not change visibly [13]. The total amount of methanogenesis in the estuarine core only corresponds to a few percent of the total organoclastic sulfate reduction that happens in the top one meter although the methane producing zone below is 7 meters thick. The sulfate reduction rate driven by methane oxidation in the sulfate–methane transition zone [16] is elevated by one order of magnitude relative to the zone directly above, because the methane is derived from metabolic activity in a large depth interval below and oxidized in a thin horizon, the sulfate–methane transition zone. But the rate of sulfate reduction coupled to methane oxidation is still orders of magnitude lower than the sulfate reduction rates at the surface. The trend with extremely skewed distribution of metabolic activity is repeated in the data from the oxic site in the North Pacific gyre where 90% of the organic matter mineralization in the 100 m deep sediment column that cover the crust here takes place in the top 6 cm. Due to the power law, most change in metabolic activity happens in the top of the sediment pile too. Tens to hundreds of meters into the sediment we will see long depth intervals with very similar metabolic activity unless there is a discontinuity in the sedimentation record. Most of the potential for change in metabolic rate is associated with loss of reactivity during aging while the organic content of the material deposited in the past has much less of an effect.

306 | 14 Experimental assessment of community metabolism in the subsurface

Fig. 14.1: The distribution of heterotrophic metabolic activity in sediment cores. (a) Rates measured with 35 S radiotracer in an eutrophic estuary that cover sediment ages from present to a few thousand years (Reproduced from [13]). (b) Oxygen consumption rates calculated from the in situ oxygen profile in the sediment at 5,700 m water depth in the oligotrophic North Pacific gyre. (b) Show data from sediment from 3–85 million years old (data from [12]).

The trend of regression line in 󳶳 Fig. 14.1 (a) continues upwards by 2 orders of magnitude to 10,000 nmol cm−3 day−1 aerobic respiration in the mm thin oxic zone. Thus, the 3 meter long estuarine core covers a relative change of 5 orders of magnitude in community metabolism. If the power law extends downward, then drilling to 300 meters depth would add only 4 orders of magnitude more. Thus, most of the transition from the conditions of the surface world and into the deep biosphere happens at relatively shallow depth.

14.2 Quantifiable metabolic processes The most direct quantification of community metabolism would be by direct calorimetric measurement of the liberated metabolic heat. Indeed it is possible to measure minute amounts of heat-generation in sediment samples via microcalorimetry. The most sensitive instruments can reliably quantify down to 100 nW cm−3 heat production (Tein Instruments). This corresponds to the heat generated by a nongrowing microbial community that converts 25 nmol complex organic carbon cm−3 day−1 into CO2 and water while reducing an equivalent amount of sulfate to sulfide. Note that this calculation includes all steps in the degradation of organic matter including hydrolysis, fermentation and ultimate oxidation. Thus, it is in principle possible to measure

14.2 Quantifiable metabolic processes

| 307

the community metabolism in surficial coastal sediments, but in the deep sea and in subsurface sediments the direct measurement of microbial heat generation is futile. This leaves few other options than to quantify bulk metabolic rates from the depletion rate of substrates or from the rate of production of the waste products from the organisms’ energy metabolism. This relies on one fundamental assumption, that the solute transformations are 100% biologically catalyzed. For many redox reactions in near surface sediment this is a safe assumption. However, caution is advised in deep sediments with low rates of metabolism where the assumption cannot readily be tested. We will discuss two primary approaches to quantify community metabolism in the marine subsurface from conversion rates of metabolites. One is mass-balance calculations based on diffusion reaction models of porewater solutes. The second is incubation with radioactive tracers.

14.2.1 Reaction diffusion modeling and mass balances Molecular diffusion in sediment follows simple physical laws. The diffusive flux in the porewater at a given depth is equal to the concentration gradient (𝜕C/𝜕C) multiplied by the diffusion coefficient. The simple Fick’s first law needs two small adjustments to apply to the sediment matrix. It must be corrected for the fact that diffusion is constrained to the pore space and that the porewater only occupies the part of the sediment volume defined by the porosity (𝜑). Second, the diffusion coefficient must be decreased to take into account that the tortuous path that a molecule will travel inside the pore space will be longer than the straight path. Several empirical relations exist between the diffusion coefficients in pure seawater (𝐷𝑚 ) and the diffusion coefficient in sediment (𝐷𝑠 ) at a given porosity [17]. The diffusive flux (𝐽) in sediment combined can be calculated from:

𝐽 = −𝐷𝑠 𝜑

𝜕𝐶 𝜕𝑧

(1)

Molecular diffusion across meter long distances is a slow process. The concentration of a soluble and conservative end product of microbial metabolism will therefore accumulate to a substantial degree before the concentration difference between the loci of production is large enough to drive a diffusive flux out of the active layers that will balance the rate of production. Notable examples of metabolic end products that accumulate in marine sediments are dissolved inorganic carbon (DIC), CH4 , NH+4 and H2 S. None of these compounds can be considered conservative in all geochemical zones, but at least DIC, CH4 and NH+4 can often be assumed to be conservative trough sufficiently deep layers to ascribe changes in the diffusive flux to microbial energy metabolism. On the electron acceptor side the most prominent candidate for quantification of microbial respiration rates is SO2− 4 , although recent research on extremely oligotrophic systems has also utilized NO−3 and O2 [12, 18]. Eq. (1) can be used directly to calculate the diffusive flux across a chosen horizon in the sediment by determin-

308 | 14 Experimental assessment of community metabolism in the subsurface ing the slope (ΔC/Δz) of a tangent to the concentration profile (󳶳 Fig. 14.1 (b)). Further detail can be extracted by applying Eq. (1) layer for layer down through the sediment column. For each layer it is then possible to calculate the local volume specific conversion rate (R) of the solute in question from the difference between the fluxes into and out of the layer. This calculation requires nothing else than a ruler, a pocket calculator and a clear idea of why and where to determine the flux. Alternatively, the depth-specific volumetric conversion rate can be calculated from the first derivative of Eq. (1) (Fick’s second law):

𝑅=

𝜕 𝜕𝐶 (𝐷𝑠 𝜑 ) 𝜕𝑧 𝜕𝑧

(2)

Eq. (2) is typically used in a reverse fashion, where those depth-specific values of R that will produce the best fitting estimation of the measured porewater profile is found by an iterative algorithm [14, 19]. Note that application of Eq. (2) becomes complicated if the profile is not in steady state. The process of calculating conversion rates from porewater profiles is referred to as porewater modeling. Most such modeling, however, only contains one free parameter, namely the production (consumption) rate of the solute in question and there is no need nor intention to link processes within the simple model. Thus it is more accurate to refer to the process a direct calculation by curve fitting.

14.2.1.1 Diffusion and consumption rate of SO2− 4 The SO2− 4 concentration in seawater exceeds all other dissolved electron acceptors by more than an order of magnitude. This allows SO2− 4 to penetrate deep into marine deposits from seawater and from subsurface brines. SO2− 4 is therefore one of the most quantitatively important electron acceptors in the deep biosphere and the volume of sulfate-reducing sediment is only exceeded by that of methanogenic zones. In sediments with a relatively shallow methane–sulfate transition zone there is still more carbon oxidized by sulfate in the upper few meters than by methanogenesis in deposits hundreds of meters thick below, often by orders of magnitude (see above). Nevertheless, the shape of the sulfate profile is mainly shaped by the anaerobic oxidation of methane (AOM) at the interfaces between sulfate-containing compartments and methane-containing compartments. This is not caused by insignificant organoclastic sulfate reduction. The domination of AOM in shaping the profile is because AOM takes place most distant from the source of sulfate, while the majority of organoclastic sulfate reduction takes place so close to the source at the sediment water interface that sulfate can be resupplied by molecular diffusion with no detectable curvature in the sulfate profile. This is a general principle, that profile modeling is simply not sensitive close to the source endpoint. The lack of sensitivity in the surficial layers is aggravated because small local concentration changes must be detected on the background of large and changing concentrations, which is much more challenging than

14.2 Quantifiable metabolic processes

| 309

Fig. 14.2: Example of flux calculations used to quantify the total microbial conversion rate in a geochemical zone. Fluxes illustrated by arrows and given in units of mmol m2 year−1 . The difference in the shape of the methane profiles is because the core from IODP site U1343 was subsampled in the shipboard laboratory while the core from site U1345 was sampled for methane on the catwalk within minutes after core recovery. Reproduced from [20].

measuring deviations from a constant and low concentration. Another blow to calculations of organoclastic sulfate reduction by profile interpretation is small changes in porosity that, according to Eq. (1) cannot be distinguished easily from changes in diffusive flux. Finally Fe(III) mineral that reacts slowly with sulfide and may oxidize a fraction of the sulfide back to sulfate to produce a cryptic sulfur cycling is not revealed by the mass balance of SO2− 4 [11]. There are geochemical scenarios where the curvature 2− of the profile SO4 profiles can be used to calculate respiratory activity, but the most common use of sulfate profiles in this context is to calculate the sulfate flux into AOM zones (󳶳 Fig. 14.2).

14.2.1.2 Diffusion, production and consumption of CH4 Methanogenesis takes over from sulfate reduction as the dominant terminal step where sulfate is depleted. Therefore, methanogenesis is constrained to sediment layers below the zone of sulfate reduction. At atmospheric pressure, the solubility of methane in seawater is ∼1.3 mM, and the amount of methane that can exist in solution increases roughly linearly as pressure increases by 1 bar per 10 meters below sea surface. Due to long diffusive distances in the methanogenic zone the methane partial pressure in the subsurface often builds up to several bars although the rates of metanogenesis is low. At depth methane will stay in solution as long as the hydrostatic pressure exceeds the partial pressure of methane. Once a core with excess

310 | 14 Experimental assessment of community metabolism in the subsurface methane partial pressure is brought to the surface, however, the methane will transfer to the gaseous phase and expand. This gas expansion has two detrimental effects for calculations of methanogenesis based on porewater profiles of methane. First, the loss of gas can cause underestimation of the in situ concentrations that exceeds 1.3 mM (󳶳 Fig. 14.1 (a)). This problem can be partly solved by taking sediment samples for analysis of methane concentration within minutes after core recovery. When such fast sampling is performed there tends to be good agreement between the calculated methane and sulfate fluxes into the methane–sulfate transition zone and the stoichiometry of anaerobic methane oxidation coupled to sulfate reduction (󳶳 Fig. 14.1 (b)). Still, all reported concentrations at or above 1.3 mM methane should be seen as minimum values and only used for interpretation of diffusive fluxes with the greatest caution. The second detrimental effect is that the gas expansion itself can lead to all stages of sediment disturbance from slight elongation of the core to foaming and violent ejection of sediment. As for sulfate, there are core segments where volumetric rates of methane production can be calculated from profile curvature, but these are few and far between. In most cases, the most useful information about metabolic activity that can be extracted from the concentration profile is the summed methanogenesis in the entire methanogenic zone based on the concentration gradient at the outer boundaries.

14.2.1.3 Diffusion and production rate of CO2 and NH+4 The production rate of CO2 is, after the heat production, the second most direct and general assessment of the metabolic activity of heterotrophic communities. Unfortunately, the dissolved inorganic carbon (DIC) participates in many precipitation and dissolution reactions and the low rates of biogenic CO2 production in the subsurface can often not be separated from these with sufficient sensitivity. The mineralization of organic carbon into DIC is, however, associated with a proportional release of organic N as NH+4 . The ratio between C and N production cannot be assessed from the C:N ratio of the particular organic carbon in the sediment because the composition of the bulk material is not necessarily representative for the reactive fractions, and because preferential mineralization of nitrogen cannot be excluded. But the C:N ratio of the mineralized fraction can be read from parameter plots where the concentration of DIC is plotted versus the concentration of NH+4 (󳶳 Fig. 14.3 (a)) If the rate of production of the two solutes is in a fixed ratio, then the diffusive fluxes in the sediment must happen in the same ratio. And as the two solutes diffuse in the same sediment matrix then it follows that the concentrations throughout the profile must also be found in the same stoichiometric ratio, with a slight correction if the molecular diffusion coefficients of the two are dissimilar. 󳶳 Figure 14.3 (a) shows a parameter plot of porewater from the upper 745 meters of the sediment column in the Bering Sea [20]. Low concentration at the origin of the plot stem from the sediment water interface and the initial linear part of the curve up to 30 mM DIC represents the upper 7 meters of the

14.2 Quantifiable metabolic processes

| 311

Fig. 14.3: (a) Parameter plot of DIC and NH+ 4 from IODP Expedition 323 site U1343 trough 745 meters of sediment column in the Bering Sea (data from [20]) showing a stoichiometric ratio between CO2 production and NH+ 4 production of 7.7. (data from [20] and The Integrated Ocean Drilling Program Expedition 323 Scientific Party). (b) Ammonium profiles from site U1343 used for calculation of volume-specific ammonium production rates down along the core. Reproduced from [20].

sediment. The slope of the line indicates the presence of 13.6 mol DIC for each mol NH+4 . A part of this difference can be explained by NH+4 diffusing a factor 1.8 faster than DIC. Correcting the slope by this factor determines the stoichiometric ratio between DIC and NH+4 to 7.7. Note that the correction is based on the diffusion coefficient in pure water since the influence of the sediment matrix on the diffusion coefficients of the two compounds is the same. The 7.7 ratio is surprisingly close to the Redfield ratio (6.6) and constant within a depth interval corresponding to more than 10,000 years of sedimentation history. The deviation from the line in the upper 7 meters is due to carbonate precipitation at depth, whereby DIC no longer increases along with NH+4 . Since the majority of carbon mineralization in the sampled sediment column happens in the upper 7 meters there is little reason to expect radical changes in the ratio of DIC to NH+4 production in the deeper parts of the core. The production rate of DIC can therefore be estimated from the production rate of NH+4 by multiplication by the stoichiometric ratio derived from the parameter plot, in our example, 7.7. The actual depth-specific rates of ammonium production can be found, for example, using the excellent curve-fitting computer program PROFILE [19]. The production rate of DIC can be linked to other terminal processes in the sediment, such as sulfate depletion, but the most robust proxy for microbial metabolism found in the porewater is most often NH+4 .

14.2.1.4 Depletion of organic carbon along depth in the sediment A tempting method to calculate the mineralization rate that is closely related to porewater profile modeling is calculation of the rate of depletion of organic carbon down

312 | 14 Experimental assessment of community metabolism in the subsurface along a sediment core. The time scale in this approach is derived from the sedimentation rate and not from the diffusion coefficient as above. This approach, however, must assume that the input of organic carbon has been constant across geological time. As illustrated in the introduction above, the organic matter in the deep biosphere is characterized by a large pool with a very slow turnover. The amount of carbon in a given layer is therefore much more strongly influenced by the amount of organic matter that was initially deposited at the surface than by the mineralization rate in the deep sediment. This makes quantification of community metabolism based on the depletion rate the carbon content down core very crude at best. Note, however, that once rates of microbial metabolism have been quantified throughout a sediment column by other methods, then it is easy to integrate the rates and recreate the carbon budget through time. This can provide a sanity check on the measured metabolic rates because the integrated rate of carbon metabolism since deposition must agree with a realistic initial content of organic matter.

14.2.2 Measurements of rates of energy metabolism with exotic isotopes The conversion rates of electron acceptors and metabolic end products in the porewater of deeply buried sediments is so slow that quantification by monitoring changes in concentrations over time in laboratory incubation is futile. As an example, the oxygen consumption rate 25 meters below seafloor in 65 million-year-old oxic sediments in the North Pacific gyre estimated from curve fitting is only 1 μM per 1000 years (󳶳 Fig. 14.1 (b)) [12]. One μM is about the standard deviation of oxygen measurements and it would thus take 1000 years of incubation just to detect that oxygen reacts at all. As an alternative, a radioactive tracer of the same chemical composition as the reactive solute under investigation can be added to the porewater. According to the name, the tracer contributes so little to the concentration of the solute under investigation that the concentration increase due to tracer addition can be neglected. After incubation, all radioactivity found in the reaction product must have been produced during the incubation. Not only is the relative change in radioactivity in the product infinitely large, the method also takes advantage of the ability to detect individual radioactive decays. Thus the scintillation counter used to quantify radioactivity is essentially detecting molecules where chemical measurements detect moles, a staggering 23 orders of magnitude principle increase in analytical sensitivity. Radioactive isotopes can be exchanged for enriched stabile isotopes, which can be quantified with good sensitivity with isotope ratio mass spectrometry. The percentage of the tracer that has been converted during incubation is calculated and it is assumed that the same percentage of the background concentration is also converted. For a metabolic process to be measurable with a tracer, a suitable isotope must exist, either radioactive or stabile. Also, the activation energy of reactions that converts the added chemical compound into other compounds must be so low that the

14.2 Quantifiable metabolic processes |

313

isotope is not transferred to other chemical compounds trough equilibrium reactions without net conversion. Finally, it must be possible to separate the product of the reaction from the educts with great efficiency because normally only a minute fraction of the tracer is converted during incubation. The most commonly used tracers used to quantify metabolic rates in sediments are the artificial 𝛽 emitters 3 H, 14 C and 35 S.

14.2.2.1 Sulfate reduction rates measured with 35 SO2− 4 35 SO2− is a near-perfect radiotracer and the measurement of dissimilatory sulfate re4 duction rates with radiotracer [21, 22] serves as illustrative example of radiotracer technique. The high activation energy of 35 SO2− 4 excludes nonbiological isotope exchange between sulfate-bound sulfur and reduced sulfur pools in sediments. Thus, 35 S added as sulfate will only be transferred to the reduced or intermediate inorganic sulfur pools via microbial respiration with sulfate as electron acceptor. Reduced sulfur species can be separated from unreacted sulfate efficiently [22] and the labeled sulfide can then be determined with an extremely high sensitivity via liquid scintillation counting. The decay energy of 35 S is 167 keV, which is sufficient for near 100% detection by liquid scintillation counting, yet low enough to be unproblematic for health and environment at the low activities used in tracer experiments. The decay product 35 Cl is not radioactive and does not interfere with the experiments. The half-life of 35 S is 87 days, long enough for the decay during most incubations to be insignificant, yet short enough that waste can be stored until the activity is below the limits for legal disposal as nonradioactive waste (subject to local regulations). The 35 SO2− 4 tracer is relatively inexpensive and can be purchased carrier-free, dissolved in dilute hydrochloric acid. In contrast to 14C there are no applications of the natural abundance of 35 S that contamination from tracer can interfere with, and the half-life is short enough that a low level contamination with 35 S will eventually disappear by itself. If the sediment is soft enough to sub core, the intact sediment can be incubated with 35 SO2− 4 tracer in cut-off 3–5 mL syringes or glass tubes fitted with syringe plungers [23]. A less desirable incubation method is slurring because the process can profoundly influence the reaction rates, especially during long incubations [24]. It is of paramount importance that the sediment is not contaminated with oxygen during sediment handling, as molecular oxygen is toxic to most sulfate-reducing prokaryotes. This can be accomplished in an anaerobic glove box, but systems with trace amounts of H2 in the atmosphere should be avoided because many sulfate reducers will readily metabolize H2 . The mini-cores in the syringes are injected with μL aliquots of tracer into the middle of each sediment plug and incubated for 1–60 days. The maximum meaningful incubation time is 122 days [25]. The incubation is stopped by suspending the sample in high molarity Zn-acetate followed by freezing to avoid oxidation of Zn35 S. Finally the reduced 35 S is separated from the unreacted 35 SO2− 4 tracer by reductive distillation and the percentage of tracer reduced is determined via liquid scintillation counting.

314 | 14 Experimental assessment of community metabolism in the subsurface One key problem with radiotracer-based measurement of sulfate reduction rates in the deep biosphere is perturbation of the sediment during retrieval and incubation with respect to temperature, pressure and content of dissolved gases. A second limitation is lack of sensitivity. For quantification, an incubation must reduce more than 35 about 0.1 Bq 35 SO2− 4 to H2 S. Incubation of a sulfate-reducing sediment with a community respiration comparable to 1 μM oxygen per 1000 years and a sulfate concentration of 1 mM for 10 days would reduce 1/108 of the added tracer. Thus, to be detectable it is necessary to inject 10 GBq. This is about 5 orders of magnitude more tracer than is used in surface sediments. An uncomfortably large amount of radiotracer for a single incubation and more tracer than the total stock of a typical active laboratory. It is also close to the limits of standard permits for research laboratories and clearly not an amount of radioactivity that allows for high resolution screening of many cores or application as a standard procedure. Note that most research vessels will require a designated isotope van on board to allow work with radiotracers.

14.2.2.2 Conversion rates of methane measured with 14 CH2 COOH, 14 CO2 and 14 CH4 Methanogenesis can be detected like sulfate reduction by adding precursors labeled with 14 C [26–28]. A notable difference to sulfate reduction is that the precursors must be known in advance and only those added as tracers will be detected. In surficial sediments this can be a problem because the acetate concentration is small and the tracer therefore is depleted due to complete turnover of the acetate pool within the incubation. But this is less of a problem in deeply-buried sediments due to low conversion rates. A major challenge for radiotracer-based determinations of methanogenesis is that methane-producing sediments in the subsurface are typically gas charged. Not only do the expanding gases disrupt the sediment structure, but energetics of methane transformations is sensitive to the partial pressure of methane. It is therefore difficult to recreate the in situ conditions for methanogens in the laboratory. Thus, the sediment horizons where good estimations of methanogenesis rates can be estimated by radiotracer technique will generally coincide with the horizons where calculations from porewater profiles also work well. The most prominent niche for radiotracerbased calculations of methanogenesis in the subsurface is at sharp lithological transitions where porewater modeling does not have sufficient spatial resolution [29] and in sediments with porewater advection where the transport coefficient is unconstrained and porewater profile interpretation therefore is not possible. Methane oxidation rates at the interfaces between methane and sulfate can be measured by incubation with 14 CH4 . But in the deep biosphere the technique has few, if any, advantages relative to flux calculations. A special concern with radiotracer-based measurements of methane transformations is that the natural abundance of 14 C is used for radiocarbon dating, and that such measurements are extremely sensitive to contamination with artificially produced 14 C. This problem is aggravated by the long half-life of 14 C (5700 years). Once contami-

14.3 Summary

|

315

nated, a laboratory will never again be suited for handling samples destined for radiocarbon aging.

14.3 Summary The rates of microbial metabolism decrease rapidly with depth below the seafloor and even eutrophic estuaries with high rates of sedimentation resemble the really deep biosphere more than the surface world only a few meters below seafloor. Thus the deep biosphere reaches up close to the sediment water and it might often be more relevant to refer to the buried biosphere rather than the deep biosphere. Rates of microbial metabolism in the buried biosphere are extremely low and often impossible to measure experimentally due to lack of sensitivity, lack of a suitable proxy, and/or due to contamination of samples and loss of dissolved gases from the samples retrieved by scientific drilling. Thus, it is often necessary to select sites for investigations based on the possibility to quantify rates due to specific stratigraphic and geochemical conditions. For example, the deeply oxygenated subtropical gyres where no net anaerobic respiration occurs; sites with strong influence of seawater flow in the upper crust and therefore deep sulfate-reducing layers supplied from both ends with sulfate and therefore no skewing of the sulfate profile by a methane–sulfate transition zone; or sites where the most active layers at the surface have been lost to erosion which allows sulfate to supply the low rates of metabolism in old sediments without total depletion. Phrased differently, the best data come from exotic geochemical settings and in many cases there will be no methods available that can quantify the energy flux trough the microbial community. Due to the generally weak data and often questionable assumptions used in quantification of bulk metabolic rates in the buried biosphere it is imperative to apply several independent proxies for microbial metabolism simultaneously and to compare the calculated rates of community-metabolism with other related parameters. Examples of such key parameters to compare to are the carbon balance in the sediment as mentioned above, microbial growth and microbial substrate incorporation. One novel and powerful estimate of growth is modeling of the turnover of the pool of dead bacterial cells based on the disequilibrium in the distribution of d and l amino acids [30] that is maintained by continuous cell division and death. An equally elegant experimental approach that provides similar data is monitoring of the incorporation of deuterated water into newly synthesized lipids during laboratory incubations [31]. The newly synthesized lipids have the isotopic signature of the porewater irrespectively of the type of energy metabolism of the organisms. Incubation with deuterium-enriched water can therefore be performed without knowledge about what fuels the organisms and without adding anything but water. Analysis of incorporation of labeled substrates via stable isotope probing or on single-cell level via nano-SIMS [32] are promising. But incorporation rates measured after substrate addition must be interpreted as in situ rates with due caution.

316 | 14 Experimental assessment of community metabolism in the subsurface

References [1] [2] [3] [4]

[5]

[6]

[7] [8]

[9] [10] [11]

[12] [13]

[14] [15] [16] [17] [18] [19] [20]

[21]

Hoehler TM, Jørgensen BB. Microbial life under extreme energy limitation. Nature Reviews Microbiology 11 (2013), 83–94. Thauer RK, Jungermann K, Decker K. Energy-conservation in chemotropic anaerobic bacteria. Bacteriological Reviews 41 (1977), 100–180. Heberling C, Lowell R, Liu L, Fisk M. Extent of the microbial biosphere in the oceanic crust. Geochemistry Geophysics Geosystems 11 (2010). Wellsbury P, Goodman K, Barth T, Cragg BA, Barnes SP, Parkes RJ. Deep marine biosphere fueled by increasing organic matter availability during burial and heating. Nature 388 (1997), 573–576. Lin LH, Slater GF, Lollar BS, Lacrampe-Couloume G, Onstott TC. The yield and isotopic composition of radiolytic H-2, a potential energy source for the deep subsurface biosphere. Geochim Cosmochim Acta 69 (2005), 893–903. Horsfield B, Schenk H-J, Zink K, et al. Living microbial ecosystems within the active zone of catagenesis: Implications for feeding the deep biosphere. Earth and Planetary Science Letters 246 (2006), 55–69. D’Hondt S, Spivack AJ, Pockalny R, et al. Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106 (2009), 11 651–11 656. Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D’Hondt S. Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Natl Acad Sci USA 109 (2012), 16 213– 16 216. Middelburg JJ. A simple rate model for organic-matter decomposition in marine-sediments. Geochim Cosmochim Acta 53 (1989), 1577–1581. Middelburg JJ, Meysman FJR. Ocean science – Burial at sea. Science 316 (2007), 1294–1295. Holmkvist L, Ferdelman TG, Jørgensen BB. A cryptic sulfur cycle driven by iron in the methane zone of marine sediment (Aarhus Bay, Denmark). Geochimica Et Cosmochimica Acta 75 (2011), 3581–3599. Røy H, Kallmeyer J, Adhikari RR, Pockalny R, Jørgensen BB, D’Hondt S. Aerobic Microbial Respiration in 86-Million-Year-Old Deep-Sea Red Clay. Science 336 (2012), 922–925. Jørgensen BB, Parkes JR. Role of sulfate reduction and methane production by organic carbon degradation ineutrophic fjord sediments (Limfjorden, Denmark). Limnology and Oceanography 55 (2010), 1338–1352. Berner RA. An idealized model of dissolved sulfate distribution in recent sediments. Geochim Cosmochim Acta 28 (1964), 1497–1503. Boudreau BP, Ruddick BR. On a reactive continuum representation of organic matter diagenesis. Amer J Sci 291 (1991), 507–538. Sansone FJ, Martens CS. Methane oxidation in Cape Lookout Bight, North-Carolina. Limnology and Oceanography 23 (1978), 349–355. Boudreau BP. Diagenetic models and their implementation. Berlin Heidelberg: Springer; 1997. Fischer JP, Ferdelman TG, D’Hondt S, Røy H, Wenzhofer F. Oxygen penetration deep into the sediment of the South Pacific gyre. Biogeosciences 6 (2009), 1467–1478. Berg P, Risgaard-Petersen N, Rysgaard S. Interpretation of measured concentration profiles in sediment porewater. Limnology and Oceanography 43 (1998), 1500–1510. Wehrmann L M, Risgaard-Petersen N, Schrum HN, et al. Coupled organic and inorganic carbon cycling in the deep subseafloor sediment of the northeastern Bering Sea Slope (IODP Exp. 323). Chemical Geology 284 (2011), 251–261. Jørgensen BB. Sulfur cycle of a coastal marine sediment (Limfjorden, Denmark). Limnology and Oceanography 22 (1977), 814–832.

References | 317

[22] Kallmeyer J, Ferdelman TG, Weber A, Fossing H, Jørgensen BB. A cold chromium distillation procedure for radiolabeled sulfide applied to sulfate reduction measurements. Limnology and Oceanography Methods 2 (2004), 171–180. [23] Parkes RJ, Cragg BA, Bale SJ, Goodman K, Fry JC. A combined ecological and physiological approach to studying sulfate reduction within deep marine sediment layers. Journal of Microbiological Methods 23 (1995), 235–249. [24] Jørgensen BB. A Comparison of Methods for the Quantification of Bacterial Sulfate Reduction in Coastal Marine Sediments 1. Measurement with radiotracer techniques. Geomicrobiol J 1 (1978), 11–27. [25] Kallmeyer J, Ferdelman TG, Weber A, Fossing H, Jørgensen BB. A cold chromium distillation procedure for radiolabeled sulfide applied to sulfate reduction measurements. Limnology and Oceanography Methods 2 (2004), 171–180. [26] Kuivila KM, Murray JW, Devol AH, Novelli PC. Methane production, sulfate reduction and competition for substrates in the sediments of Lake Washington. Geochim Cosmochim Acta 53 (1989), 409–416. [27] Parkes JR, Sass H, Webster G, et al. Methods for Studying Methanogens and Methanogenesis in Marine Sediments. In: Timmis KN, ed. Handbook of Hydrocarbon and Lipid Microbiology: Springer Berlin Heidelberg, 3799–3826, 2010. [28] Yoshioka H, Sakata S, Cragg BA, Parkes JR, Fujii T. Microbial methane production rates in gas hydrate-bearing sediments from the eastern Nankai Trough, off central Japan. Geochemical Journal 43 (2009), 315–321. [29] Parkes JR, Webster G, Cragg B, et al. Deep subseafloor prokaryotes stimulated at interfaces over geological time. Nature 436 (2005), 390–394. [30] Lomstein BA, Langerhuus AT, D’Hondt S, Jørgensen BB, Spivac AJ. Endospore abundance, microbial growth and necromass turnover in deep subseafloor sediment. Nature 484 (2012), 101–104. [31] Wegener G, Bausch M, Holler T, Thang NM, Prieto Mollar X, Kellermann MY, Hinrichs KU, Boetius A. Assessing sub-seafloor microbial activity by combined stable isotope probing with deuterated water and 13 C-bicarbonate. Environ Microbiol 14 (2012), 1517–1527. [32] Morono Y, Terada T, Nishizawa M, et al. Carbon and nitrogen assimilation in deep subseafloor microbial cells. Proc Natl Acad Sci USA 108 (2011), 18 295–18 300, doi:10.1073/pnas.1107763108.

Index 16S rRNA 124, 143 16S rRNA gene 207 A Aarhus Bay 148 abiotic substrate generation 266 accretionary prism 7 acetate 210, 211, 286 acetate formation 6 acetate oxidation 14 acetoclastic methanogenesis 210, 212 acetogenesis 1 acetogenic bacteria 210 acetogens 70 acidophilic 212 acidotolerant 212 acridine orange 1 activity 294 alkaliphilic 67 Amazonian 232 ammonium 109 amorphous iron sulfide 211, 214 amplifiable DNA 8 Anabaena 108, 111 anaerobic oxidation of methane (AOM) 4, 73, 210 anaerobiosis 167 Ancient Archaeal Group 144 ANME 12, 115, 149 anthropogenic activities 167 aquifer 63, 206, 207, 214 aquifer thermal energy storage (ATES) 205, 207 Archaea 70, 143 Archaeglobus fulgidus 166 arthesian wells 66 attached and unattached microorganisms 69 autoclaves 217 autometallography 112 autoradiography 106 autotrophy 86 axenic 93 B bacteriophage 76 basalt 66

basalt aquifers 207 baseline study 216 Basin Modeling 262 biocorrosion 214 biodegradation index 268 biodegration 166 biodiversity 77 biofilm 64, 93, 207, 213 Biogenic gas 264 bio-geo interactions 217 biogeochemical cycle 101 biogeographical structure 149 biological energy quantum 292 biological monitoring 216 biomass estimate 4 biomineralization 212, 213, 216 bioreactors 94 biosignatures 39 borehole 206 bulk activity 106 burial rates 262 Burkholderiales 66 bypass systems 216 C C/N ratio 110 CaCO3 precipitation 213 calcite 63 Canadian Shield 70 Candidatus Desulforudis audaxviator 206, 207 Carbon Capture and Storage (CCS) 203, 204, 206, 209–211, 214 – CO2 sequestration see CCS – CO2 Storage see CCS carbon cycle 206, 210 carbon dioxide 209 carbon dioxide sequestration see CCS carbon isotopes 101 carbon oxidation 304 carbonaceous chrondites 233 carbonate 211, 213, 215, 216, 232 carbonate mound 13 carbonate precipitation 213 carbonation 213 CARD-FISH 106, 124

320 | Index catagenesis 264 cation adsorption 213 Cedars 67 cell abundance 113 cell adaptation 212 cell counts 4, 207 cell death 208, 212 cell extraction 123 cell isolation 93 cell numbers 2 cell stress 212 cellular activity 106 chemoautotrophic 71 chemolithotrophic 238 chemosynthesis 86 Chesapeake Bay Impact Structure 243 chlorite 231 Circulation Obviation Retrofit Kits see CORK clays 234 climate impact 213 Clostridia 67 CO2 degasing site 211 CO2 trapping 213 coals 211 colonizer 93 conditioning and storage 92 contamination 65, 92, 154 contamination artifact 68 contamination control 6, 36 contamination test 84 controlled pressure displacement sampler (PDS) 216 CORK 33, 93 corrosion 203, 214, 215 cosmochemistry 102 Crenarchaeota 21, 144 CRISM 231 cultivation 75, 83, 91, 211, 212, 217 culturable 70 culturable cell counts 13 culture collection 94 culture efficiency 92 Culturing 121 Cyanobacteria 108 D deep hole sampling 216 deep hot biosphere 71 deep hot reservoirs 177

deepest mining 69 Deep-IsoBUG 15 Deep-Sea Euryarchaeotal Group 148 Deep-Sea Hydrothermal Vent Euryarchaeotal group 148 Deltaproteobacteria 169 denitrification 108 Desulfohalobium 206 Desulforudis audaxviator see Candidatus D. audaxviator Desulfotomaculum 206, 207 Desulfotomaculum geothermicum 212 Desulfotomaculum reducens 206 detection limit 94 detrital protein 148 Devonian 164 diffusion reaction models 307 diffusive flux 307 direct cell counting 217 dissimilatory sulfate reductase 169 dissimilatory sulfate reduction 313 dissimilatory sulfite reductase 206 dissolution of CO2 213 dissolution of minerals 209 dissolution trapping 204 DOC 75, 215 domains of life 144 dormancy 212 dormant 86, 102 dormant cells 208 Driefontein mine 71 drill mud 208, 209 drilling 215, 216, 250 dsrA gene 207 dykes 71 E ecology 106 ecophysiology 108 EGS 205 electron acceptor 290 electron donors 282 endoliths 51 endospores 86 endosymbiotic bacteria 108 energetic potential 279 energetic profiling 280 energy 71 Energy densities 285

Index | 321

energy flux 84 enhanced gas recovery 204, 213 Enhanced Geothermal Systems (EGS) 205 enhanced mineral precipitation 215 enhanced oil and gas recovery 204 enhanced oil recovery 204 epibiotic bacteria 108 epiliths 51 Epsilonproteobacteria 178 equilibrium constant 294 estimation of microbial activity 74 eukaryotes 30 Euryarcheota 144 evidence of microbial activity 67 excavation 64 extracellular DNA 130 F facultative aerobic microorganisms 178 FCCs 75 Fe(III) reduction 212 Fennoscandian Shield 65 fermentation 306 fermentative bacteria 174, 206, 215 fermentative prokaryotes 174 fingerprinting analysis 206, 211 Firmicutes 169 Fischer Tropsch-Type synthesis 43 FISH-SIMS 110 flagellum 105 flow cells 65 flow cytometry 91 flow reactor 213 flow-through culture 94 fluid pressure 180 fluid sampler 216 fluid sampling 34, 216 fluid-gas-rock interactions 211 fluid-rock interaction 209 Fluorescence in situ Hybridization (FISH) 124, 206, 211 fluorophores 111 formate 211 formation fluid 205, 206, 209 formation of biomass 212 formation permeability 205 fossilized biofilm 68 fossils 66 fracking 209, 214

fracture 63, 164 fracturing 203 Functional genes 153 fungi 39, 70, 187 G Gammaproteobacteria 177 Gamma-Ray Spectrometer 226 gas hydrate 10 gene transcripts 152 generation time 86 genetic exchange 76 genetic pool 212 genotype 115 Geobacter 243 Geobacter sulfurreducens 214 geochemical profiles 74 geochemical trapping 204 geochemistry 74 geo-engineered system 203, 209, 214, 216 geologic trapping of CO2 213 geothermal doublet system 205 Geothermal Energy Generation 203, 205, 214 geothermal fluid 214 geothermal power plant 205, 214 geothermal reservoirs 205 Geotoga 167 Gibbs energies 293 Gibbs energy of reaction 283 gills 107 gold mine 71, 207 greenhouse gas emission 203, 204 greenstone belt 67 groundwater fluctuation 203 groundwater movements 68 Gulf of Mexico 148 Gullies 236 H H2 formation 20 Halanaerobium 206 Halobacteroidaceae 206 halophilic bacteria 206 halotolerant 189 heat exchanger 206, 207 Hellas Basin 227 Hesperian 225, 232, 240 heterotrophs 101 heterotrophy 86

322 | Index High Arctic 237 high gas partial pressure 211 high pressure cultivation 15 high pressure system for cultivation 211 high-pressure sampler 216 HKF equations of state 294 homoacetogenesis 168 Homoacetogenic 69 hot deep reservoirs 176 hot dry rock (HDR) 205 HYACINTH 15 hybridization 105, 106 hydraulic stimulation 205 hydrocarbons 161, 167 hydrogen 167 hydrogen sulfide 214 Hydrogenophaga 66 hydrogenotrophic methanogenesis 152, 212 hydrogen-oxidizing bacteria 206 hydro-geothermal system 215 hydrolysis 306 hydrostatic 84 hydrothermal 232 hydrothermal vent 29, 208 hyperthermophiles 23, 29 hyperthermophilic 208 I ICDP 233 Immunogold 111 impact craters 242 Impact ejecta 250 in situ activity 74 in vitro 74 incubator 93 indicator organisms 216 injection well 214 injectivity 211, 214 inoculation 92 intact polar lipids 22 ion beam 103 ion microprobe 102 ion microscopy 102 ionization yields 111 iron 42 iron oxidation 75 iron reduction 14 iron sulfide 216

isotopic signature of gases 214 isotopic variations 104 J Juan de Fuca (JdF) 281 Juan de Fuca Ridge 145, 153 K karsts 164 kerogen 164 knallgas 292 Korarchaeota 144 L lignites 211 lithoautotrophic systems 304 lithoautotrophs 72, 150 lithosphere 161, 163 lithostatic 84 lithostatic pressure 166 living biomass 101 living creatures 105 LMD 113 Lomonosov Ridge 145 long-term microbial processes 216 long-term monitoring 214, 217 low molecular weight acids 211, 215 Low molecular weight hydrocarbons 6 lytic infection 76 M macromolecules 93 magnesite 213 magnesium silicate minerals 213 Malm aquifer 207 manganese reduction 12 Marine Benthic Group B 144 marine sediment 280 Mars Odyssey 226 Mars Reconnaissance Orbiter 227 MARSIS 227, 228 MARTE 241 maturation 165 MDA 94 metabolic activity 86, 106 metagenome 154 Metagenomic analysis 101 metal reduction 212 metal sulfide 214

Index | 323

meteoric origin 65 methane 72 methane hydrates 149 methane oxidation 305 methane production rate 212 methane sulfate interface 10 methanoarchaea 171, 180 Methanobacterium 207 methanogenesis 7, 8, 14, 207, 212, 241, 305, 309 methanogens 70, 152, 208, 210, 212 Methanothermobacter thermoautotrophicus 206 Methanothermococcus thermolithotrophicus 212 methyl coenzyme M reductase 153 MIC 208, 214–216 microanalysis 103 microautoradiography 69 microbial – genetics 83 – morphology 83 – physiology 83 microbial abundance 70 microbial activity 73, 217 microbial colonization 105 microbial energy demand 292 microbiological monitoring 206, 216 microcalorimetry 306 microcolonies 91 microdroplets 94 micromanipulation 94 microsphere 92 microtechnologies 91 Milankovitch cycles 12 mineral phases 287 mineral precipitation 203, 211 mineral scaling 215 mineral trapping 204, 213 mineral weathering 20 mineralization rate 304, 311 minimum energy requirements 121 mining 68 Miscellaneous Crenarchaeotal Group 148 Molasse Basin 207 Molecular diffusion 307 molecular-biological methods 217

N N2 fixation 108 Nankai Trough 7, 145, 153 Nanoarchaeum equitans 144 nanoparticles 111 NanoSIMS 102, 103, 107, 112–114 New Caledonia Basin 145 Newfoundland Margin 145 Nili Fossae 229 nitrate 178 nitrate reduction 213 nitrogen cycle 108 nitrogen fixation 107, 234 nitrogen lift 211, 215 nitrogenase 108 Noachian 225 noncompetitive substrates 152 normalization 288 North Pacific Gyre 305 nucleation substrate 213 numerical model 205, 213 nutrient availability 86 O obliquity 245 Ocean Drilling Program (ODP) 145, 233 oceanic crust 29 oil production 214 oil reservoir 179, 212 oil-water contact 268 Okhotsk Sea 145 oldest habitats 67 oligonucleotide 114 oligotrophic 91 olivine 213, 232 Olkiluoto 69 Ophiolites 66 optical tweezers 93 optoelectronic approaches 93 organoclastic sulfate reduction 305, 308 osmotic stress 172 oxalate 211 oxyfuel combustion plants 204 P packer systems 65 palagonite 67 pasteurization 268 PCR 101

324 | Index per cell rates 121 perchlorate 234, 236, 245 Perfluorocarbon tracers 92 periglacial 229 permafrost 71, 208, 229 permeability of rock 213, 214 Peru Margin 153, 281 Peru Trench 153 petroleum 161 petroleum accumulation 166 petroleum formation 163 petroleum leakage 261 petroleum system 163 Petroleum System Modeling 262 Petrotoga 167 phages 76 phenotype 115 Phoenix 229, 237, 243 phosphate 234 photosphere 71 Phylogenetic characterization 67 phylogenetic diversity 21 phytoplankton 164 piezophilic prokaryotes 84 Planctomycetes 21 population shifts 212 Porcupine Basin 15, 145 precipitation of carbonates 214 precipitation of minerals 209 pressure adaptation 15 pressure vessels 217 primers 77 prokaryotes 83, 86 prokaryotic activity 4 protein degradation 94 protein-coding genes 152 Pseudomonas 70 pumping test 216 pyrite framboids 216 pyroxenes 234 Q qPCR 124 quality assurance – control 92 R radar 226 radioactive waste 68

radiochemical reactions 207 radioisotopes 105 radiolysis 72 radionuclides 213 radiotracer 69, 312 radiotracer measurements 4, 13 RCB 92 reaction quotient 294 real-time PCR 207 recalcitrant organic matter 18 redox potential 63 redox stratification 203 reductive acetyl CoA pathway 210 reductive citric acid cycle 210 reservoir rock 164, 206, 209, 213, 214 reservoir stimulation 205 residual trapping 204 resuscitation 208 Rhizobium 111 ribosomal RNA 111, 143 ribosome 143 Rio Tinto 241 RNA slot blot 124 Rock 63 rock permeability 214 rock sample 216 RT-qPCR 125 S saline aquifer 205–207, 209 sampling 216 sampling procedure 216 sandstone 207 sapropel 145 scaling 203, 215 seasonal heat storage 207 secondary biogenic methane 271 sedimentation rates 7 sequestration of CO2 204 serpentine 213, 231 serpentinization 43, 51, 66, 238, 239 SHARAD 227 Shewanella oneidensis 212 Shimokita Peninsula 153 sidewall coring 216 silica 215, 216 SIMS 102 simulated in situ conditions 217 single cell genomics 127

Index | 325

single-cell analysis 102 single-cell genome sequencing 148 single-cell microbiology 107 SLIMEs 72 slurry preparation 92 smectites 232 SMTZ 70 solid electrodes 214 solubility-trapping 213 source rock 164 South African Goldmine Euryarchaeotal Group 148 South China Sea 145 South Pacific Gyre (SPG) 148, 150, 282 Southern blot 125 spectrophotometer 91 spores 20, 208, 212 stable carbon isotope measurement 207 stable isotope 73, 110 stagnant aquifers 71 stardust 104 starvation 208, 212 Stripa mine 69 structural trapping 204 Stuttgart-Formation 206 subsampling 92 subseafloor biosphere 105 substrate incorporation 106 substrate-accelerated death 86 suevite 243 sulfate 234 Sulfate Reducing Bacteria (SRB) 206–209, 212, 214, 215 sulfate reduction 7, 14, 211, 214 sulfate-free 75 sulfate–methane transition zones 149 sulfide 75 sulfur cycling 43 sulfur minerals 215 supercritical carbon dioxide 213 SYBR Green I 123 symbionts 107 Synergistetes species 176 syntrophy 93, 209, 210

T Tablelands 67 Taqman 125 tectonic 63 temperature history 262 temperature limit of life 29 terminal electron acceptors 261 terrestrial 76 Thaumarcheota 144 THEMIS 251 thermal fluids 206 Thermodesulfobacteria 168 thermodynamic properties 294 thermophile 174 thermophilic 215 Thermotoga spp. 177 thermotolerant 215 thiosulfate-oxidizing bacteria 206 thymidine incorporation 11 total cell counts 13, 121 transformation 74 transformation of minerals 209 trap formation 165 triple tube drilling 65 tunneling 68 U ultraviolet 225, 239 underground repositories 68 uplift 66 ureolysis 213 Utopia Planitia 237 U-tube 216 V Valles Marineris 231 valley networks 235 vegetative cells 86 viable but nonculturable cells (VBNC) 208 virus 76 W Water activity 240 well fluids 211 wellhead 217

Also of Interest Series: Life in Extreme Environments (edited by Prof. Dr. Dirk Wagner) Microbial Evolution under Extreme Conditions Corien Bakermans, 2015 ISBN 978-3-11-033506-4, e-ISBN 978-3-11-034071-6, Set-ISBN 978-3-11-034072-3 Microbial Life of Cave Systems Annette Summers Engel, 2015 ISBN 978-3-11-033499-9, e-ISBN 978-3-11-033988-8, Set-ISBN 978-3-11-033989-5