The Pennsylvania State University the Graduate School the Eberly

The Pennsylvania State University

The Graduate School

The Eberly College of Science

PHYSIOLOGICAL ECOLOGY OF HYDROCARBON SEEP TUBEWORMS

FROM THE GULF OF MEXICO

A Thesis in

Biology

Sharmishtha Dattagupta

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

December 2006

The thesis of Sharmishtha Dattagupta has been reviewed and approved* by the following:

Charles R. Fisher Professor of Biology Thesis Advisor

Michael A. Arthur Professor of Geosciences

Sarah M. Assmann Professor of Biology

Simon Gilroy Associate Professor of Biology

Douglas R. Cavenar Professor of Biology Head of the Department of Biology

*Signatures are on file in the graduate office.

ii ABSTRACT

Cold seeps are located worldwide at active and passive continental margins, and are characterized by the seepage of hydrocarbon-rich fluids from deep reservoirs to the seafloor. Chemosynthetic communities comprising tubeworms, clams and mussels thrive at cold seeps and derive nutrition from reduced chemicals such as methane and sulfide.

This study focuses on thiotrophic vestimentiferan tubeworms from the upper Louisiana slope of the Gulf of Mexico. These tubeworms, dominated by the species Lamellibrachia luymesi and Seepiophila jonesi, are nutritionally dependent on internal sulfide-oxidizing bacterial symbionts. These tubeworms form large, bush-like aggregations that provide habitat for numerous associated fauna, and can persist for centuries. Individual tubeworms grow posterior root-like structures to mine sulfide from the sediment underlying their aggregations. Microbial communities that oxidize hydrocarbons using sulfate as an electron donor produce sulfide in the sediment. Previous to this study, it was hypothesized that tubeworms could release sulfate, a byproduct of sulfide oxidation by their symbionts, through their roots into the underlying sediment. They could thereby enhance microbial sulfate reduction, and ensure themselves a lifetime supply of sulfide.

In this study, I provide the first empirical evidence for this phenomenon. I used a multi-disciplinary approach that included live animal physiology, tissue biochemistry, sediment geochemistry and theoretical modeling, to demonstrate that tubeworms can enhance microbial sulfate reduction through root sulfate release. My experiments with live L. luymesi tubeworms demonstrated that they release 85% of the sulfate produced by sulfide oxidation, and 67% of the protons produced by various metabolic processes across their root. Based on inhibitor experiments, I suggest that they use sulfate-

iii bicarbonate exchangers to mediate sulfate transport across their root epithelium. Further,

I measured the proton-ATPase activity of their plume and root tissues and surmised that these tubeworms might use passive proton channels for root proton transport. In combination, these results suggest that tubeworms could conserve energy by eliminating sulfate and protons using ATP-independent mechanisms across their roots.

In situ geochemical characterization of the tubeworm habitat showed that these tubeworms could exert a significant influence on the chemistry of their environments.

Geochemical data, in combination with a theoretical model, suggest that tubeworms release between 70-90% of the sulfate produced by their symbionts across their roots.

The combination of root sulfide uptake and sulfate release appears to maintain a steady sulfide-to-sulfate ratio in sediments adjoining tubeworm roots. The model suggests that by releasing sulfate, tubeworms can maintain basal metabolic requirements as well as enhanced growth rates under conditions where sediment sulfate concentrations would otherwise become limiting.

iv TABLE OF CONTENTS

LIST OF FIGURES ...... VIII LIST OF TABLES ...... XII PREFACE ...... XIII ACKNOWLEDGEMENTS...... XIV CHAPTER 1: AN INTRODUCTION TO VESTIMENTIFERAN TUBEWORMS ASSOCIATED WITH HYDROCARBON SEEPAGE IN THE GULF OF MEXICO...... 1 COLD SEEP COMMUNITIES ...... 1 GULF OF MEXICO HYDROCARBON SEEP HABITAT CHARACTERISTICS ...... 2 BIOGEOCHEMISTRY OF HYDROCARBON-RICH SEDIMENTS IN THE GULF OF MEXICO ...... 4 VESTIMENTIFERAN TUBEWORM PHYSIOLOGY ...... 6 HYPOTHESES AND QUESTIONS ADDRESSED BY THIS THESIS ...... 10 CHAPTER 2: THE HYDROCARBON SEEP TUBEWORM LAMELLIBRACHIA LUYMESI PRIMARILY ELIMINATES SULFATE AND HYDROGEN IONS ACROSS ITS ROOTS TO CONSERVE ENERGY AND ENSURE SULFIDE SUPPLY...... 14 ABSTRACT ...... 15 INTRODUCTION...... 16 MATERIALS AND METHODS...... 19 Collection and maintenance of tubeworms ...... 19 Measurements of body fluid pH and sulfate concentrations ...... 21 Live tubeworm experiments ...... 22 Measurement of sulfate elimination rates...... 25 Measurement of proton elimination rates...... 26 Data analysis and statistics ...... 27 RESULTS ...... 27 Body fluid parameters...... 27 Laboratory maintenance of tubeworms ...... 27 “Fed” tubeworms: sulfate and proton elimination rates ...... 28 Comparison between “starved” and “fed” tubeworms ...... 29 The effect of inhibitors on sulfate and proton elimination rates...... 31 DISCUSSION...... 32 Sulfate elimination from experimental animals...... 35 The mechanism of sulfate release...... 37 Proton elimination from experimental animals...... 39 The relation between proton and sulfate elimination ...... 41 Conclusion ...... 42 ACKNOWLEDGEMENTS ...... 43 CHAPTER 3: COMPARISON OF PROTON-SPECIFIC ATPASE ACTIVITIES IN PLUME AND ROOT TISSUES OF TWO CO-OCCURING TUBEWORM SPECIES LAMELLIBRACHIA LUYMESI AND SEEPIOPHILA JONESI...... 44 ABSTRACT ...... 45 MATERIALS AND METHODS...... 48 Collection of tissues for analysis...... 48 Preparation of tissue homogenates...... 49 Determination of mitochondrial content: succinate dehydrogenase activity...... 50 Determination of plasma membrane content in fractions ...... 51 Measurement of ATPase activity...... 51 Inhibitor exposure ...... 52 Estimation of proton-specific P- and V-type ATPase activities ...... 53

v RESULTS ...... 53 Mitochondrial activity of fractions...... 53 Na+/K+-ATPase activity of fractions...... 54 Total ATPase activities ...... 55 Inhibitors...... 56 DISCUSSION...... 58 Seep tubeworms have much lower ATPase activities than Riftia pachyptila...... 58 Lamellibrachia luymesi might use proton channels in its roots for proton elimination...... 59 The relative abundance of P- and V-type ATPases varies between seep species ...... 61 Possible caveats of this study and future directions ...... 62 ACKNOWLEDGEMENTS ...... 63 CHAPTER 4: SUBMERSIBLE OPERATED DIALYSIS SAMPLERS FOR COLLECTING PORE WATER FROM DEEP-SEA SEDIMENTS ...... 64 INTRODUCTION...... 66 MATERIALS AND PROCEDURES...... 68 Overall design and construction of peepers...... 68 Preparation of dialysis bags...... 71 De-oxygenation procedure ...... 73 Deployment and recovery of peepers...... 74 Post collection sampling and analyses ...... 75 ASSESSMENT ...... 77 DISCUSSION...... 82 COMMENTS AND RECOMMENDATIONS...... 83 ACKNOWLEDGEMENTS ...... 84 CHAPTER 5: MODIFICATION OF SEDIMENT GEOCHEMISTRY BY THE HYDROCARBON SEEP TUBEWORM LAMELLIBRACHIA LUYMESI: A COMBINED EMPIRICAL AND MODELING APPROACH...... 85 ABSTRACT ...... 86 INTRODUCTION...... 87 MATERIALS AND METHODS...... 90 Pore water sampling scheme...... 90 Chemical analyses...... 92 Estimation of sulfate reduction rates ...... 93 Sulfate diffusion-reaction-supply model ...... 93 Estimations of tubeworm growth and metabolism ...... 101 RESULTS ...... 102 Pore water pH and salinity...... 102 Pore water methane, sulfate and sulfide concentrations...... 105 Pore water sulfur stable isotope ratios...... 109 Sulfate diffusion-reaction-supply model ...... 111 DISCUSSION...... 115 Tubeworm habitat spatial heterogeneity ...... 115 In situ evidence for tubeworm sulfide uptake and sulfate release ...... 116 Tubeworm impact on sediment pH...... 120 Can tubeworms enhance microbial sulfate reduction rates? ...... 120 How does variability in hydrocarbon flux affect tubeworm metabolism and growth?...... 122 ACKNOWLEDGEMENTS ...... 125 Implications and future directions ...... 128 APPENDIX A: ATTEMPTS TO PATCH-CLAMP BATERIOCYTES ISOLATED FROM LAMELLIBRACHIA LUYMESI ...... 130 PURPOSE ...... 130 BACKGROUND ...... 130 METHODOLOGY ...... 133

vi Isolation of bacteriocytes...... 133 Microscopic studies and FISH with isolated bacteriocytes...... 134 Survival of bacteriocytes...... 134 Preliminary patch-clamp studies ...... 135 RESULTS AND DISCUSSION ...... 137 Isolation and identification of bacteriocytes ...... 137 Bacteriocyte survival...... 137 Preliminary patch clamp findings...... 139 CONCLUSIONS ...... 140 ACKNOWLEDGEMENTS ...... 140 APPENDIX B: DEEP-SEA HYDROCARBON SEEP GASTROPOD BATHYNERITA NATICOIDEA RESPONDS TO CUES FROM HABITAT-PROVIDING MUSSEL BATHYMODIOLUS CHILDRESSI ...... 141 ABSTRACT ...... 142 PROBLEM...... 143 MATERIALS AND METHODS...... 144 RESULTS AND DISCUSSION ...... 147 ACKNOWLEDGEMENTS ...... 152 APPENDIX C: TISSUE CARBON, NITROGEN AND SULFUR STABLE ISOTOPE TURNOVER IN TRANSPLANTED BATHYMODIOLUS CHILDRESSI MUSSELS: RELATION TO GROWTH AND PHYSIOLOGICAL CONDITION...... 153 ACKNOWLEDGEMENTS ...... 154 ABSTRACT ...... 155 INTRODUCTION...... 156 MATERIALS AND METHODS...... 158 Study sites ...... 158 Transplants ...... 160 Stable isotope analyses ...... 162 Data analyses ...... 163 RESULTS ...... 166 Growth and condition indices ...... 166 Stable isotope analyses ...... 167 DISCUSSION...... 170 BIBLIOGRAPHY ...... 180

vii LIST OF FIGURES

Figure 1-1: Internal anatomy of Riftia pachyptila showing the location of the major organs and blood vessels. Drawing adapted from Arp et al. (1985)...... 7 Figure 2-1:The extensive tubeworm roots of an aggregation that was excavated and collected using the Bushmaster sampling device (Bergquist et al., 2002; Cordes et al., 2004). This photograph shows roots that are about 1 meter long, extending below the carbonate rock substrate. The roots were observed to break during collection and we believe they might have been even longer in the intact aggregation (CRF, pers. obs.). Photo credits: Dr. E. E. Cordes...... 17 Figure 2-2: Schematic representation of a tubeworm maintenance aquarium. Sulfide dissolved in synthetic seawater (SSW) was introduced into the polyvinylchloride (PVC) hose, which was connected to PVC grating. The sulfide diffused out from the grating through the crushed coral and sediment layers. The aquarium was filled with SSW, which was kept aerated and filtered...... 20 -1 -1 Figure 2-3: Relation between (A) sulfate (µmoles h g wet weight), and (B) protons -1 -1 (µequivalents h g wet weight) eliminated across plumes and roots of individual tubeworms exposed to sulfide for 48 hours prior to the measurement. Data points represented by closed circles in (B) correspond to animals that might have experienced hypoxic conditions and eliminated protons from anaerobic metabolism...... 30 Figure 2-4: Relation between (A) plume, (B) root, and (C) total (plume and root combined) proton and sulfate eliminated from individual tubeworms exposed to sulfide for 48 hours prior to the experiment. Data points represented by closed circles correspond to animals that might have might have experienced hypoxic conditions and eliminated protons from anaerobic metabolism...... 30 Figure 2-5: A summary of the effect of the membrane transport inhibitors, and the solvent (DMSO) on sulfate eliminated across plumes and roots of L. luymesi. An asterisk indicates that the treatment caused significant (P < 0.05) inhibition of sulfate elimination rate...... 31 Figure 2-6: A schematic representation of an aggregation of L. luymesi, depicting a nutrient exchange model for this species. Seawater and sediment pore-water sulfate concentration and pH are from Aharon and Fu (2000). The equation for microbial sulfate reduction shows sulfate reduction coupled with methane oxidation. However, sulfate reduction coupled with higher molecular weight hydrocarbons can also occur (Joye et al., 2004). Bicarbonate uptake across roots is suggested, but has not been directly demonstrated in this study...... 42 Figure 3-1: Mean ± standard deviation of succinate dehydrogenase (SDH) activity in the crude, S1 and S2 fractions of the various tissues analyzed (N=5 for all tissues). L and S correspond to L. luymesi and S. jonesi, respectively...... 54 Figure 3-2: Mean ± standard deviation of Na+/K+-ATPase activity in the crude, S1 and S2 fractions of the various tissues analyzed (N=4 for plume tissues; N=3 for root tissues). L and S correspond to L. luymesi and S. jonesi, respectively...... 55 Figure 3-3: Comparison of percent inhibition of total ATPase activity by various inhibitors in (a) L. luymesi plume and root tissues; (b) L. luymesi and S. jonesi plume tissues; (c) S. jonesi plume and root tissues; and (d) L. luymesi and S. jonesi root

viii tissues. An asterisk implies a significant difference (P<0.05; two-sided t-test). AZ=azide, OU=ouabagenin, NEM= N-ethylmaleimide, VAN=vanadate, BAF=bafilomycin. Percent inhibition is expressed with respect to baseline activity measured in the presence of the solvent used to make inhibitor solutions...... 57 Figure 4-1: (a) Peeper dismantled to show its component parts, (b) Peeper in “open” position, (c) Peeper in “closed” position, (d) Schematic view of peeper handle assembly and (e) Schematic view of inner tube showing actuating handle and details of radial seals...... 70 Figure 4-2: (a) Peepers deployed in situ with plates level with the sediment-water interface, in sediment underlying a Beggiatoa bacterial mat. Handle assembly is in the “open” position. The arrow indicates the peeper used to collect data shown in Figure 4-3. (b) Peeper handle being “closed” by the robotic arm of the Johnson Sea- link submersible...... 71 Figure 4-3: Schematic representation of the set-up used to de-oxygenate peepers before deployment. Peepers were fully assembled before insertion into the chambers, and six peepers could be de-oxygenated at a time. The peeper plates were moved to the very top of the peepers, so that all peeper windows could fit into the acrylic chambers. Before deployment, the peepers were closed and the plates were moved back to the position indicated in Figure 1...... 73 Figure 4-4: Depth profiles of various parameters obtained using the peeper indicated in figure 2...... 78 Figure 4-5: Results from the equilibration experiment performed with dialysis bags inside a cold room maintained at 6-8°C. 21 dialysis bags were filled with sulfate-free ASW and suspended inside a container with two liters of ASW containing 24 mmol L-1 magnesium sulfate. 3 bags each were collected at 7 different time points ranging from 0 to 94 hours, and the sulfate concentrations of the solutions inside the bags were determined. Data points in the graph show the average ± standard deviation of sulfate concentrations inside the bags at various time points. Dashed lines show the average ± standard deviation of sulfate concentration in the solution outside the bag, which did not vary over the course of the experiment. The solid line shows the best- fit curve corresponding to the equation , where Ci is the Ci = 23.6 * #"1! exp(! 0.153 t)%$ sulfate concentration inside the bags at time, t...... 80 Figure 4-6: Results of a model in which sulfate diffusion from the surrounding sediment into a dialysis bag was simulated at various pore water sulfate concentrations...... 82 Figure 5-1: Peepers deployed in situ at four different locations, in replicates of 3. Peepers were deployed either ~30 cm away from tubeworms (“Near”) or > 6 meters away from large aggregations (“far”). Also, they were either deployed in sediments underlying orange Beggiatoa mats (“oily” sediments) or in sediments with apparently low hydrocarbon flux (“plain” sediments)...... 91 Figure 5-2: pH profiles in (a) “plain” and (b) “oily” sediments near and far from tubeworms...... 103 Figure 5-3: Salinity profiles in (a) “plain” and (b) “oily” sediments near and far from tubeworms...... 105 Figure 5-4: Mean methane concentrations (µM) sampled by 3 replicate peepers at 4 locations...... 105

ix Figure 5-5: Pore water sulfate and sulfide concentration profiles at the four locations sampled. Sulfide concentrations in “plain” sediments were in the micromolar range, reaching a maximum of 63 µM (Table 5-2)...... 107 Figure 5-6: Relation between integrated sulfate reduction rates (SRR) estimated from sulfate depth profiles (Table 5-3) and average pore water methane concentrations in corresponding sediments...... 109 Figure 5-7: Pore water sulfate and sulfide sulfur isotope ratios at the four different 34 locations sampled. The mean ± standard deviation of tubeworm blood sulfate δ S is also shown. These tubeworms were collected from the aggregation in the “oily” 34 location. Sulfide δ S values could not be determined for “plain” sediments, due to low sulfide concentrations in “plain” sediment pore water...... 110 Figure 5-8: Sulfate and sulfide concentration profiles generated using model simulations at various maximal sulfate reduction rates (Rm). Tubeworm influence was simulated by assuming sulfide uptake by a one-meter tall tubeworm aggregation. Sulfate release across roots was assumed to be 70% of the total sulfide oxidized by the tubeworm...... 112 Figure 5-9: Sulfate and sulfide sulfur stable isotope profiles generated using model 34 simulations at various maximal sulfate reduction rates (Rm), and using a Δδ S value of either 15‰ or 30‰, as indicated. Tubeworm influence was simulated by assuming sulfide uptake by a one-meter tall tubeworm aggregation. Sulfate release was assumed to be 70% of the total sulfide oxidized by the symbionts...... 114 Figure 5-10: Sulfide-to-sulfate concentration ratios at various depths, plotted along with ratios predicted using the model. The various lines show results from simulations run either in the absence of tubeworm influence, or with tubeworm root sulfate release varying between 0-90 % of the total sulfide oxidized by their symbionts. Data in the top two profiles were collected in this study. The bottom two profiles were plotted using data from cores collected near and far from tubeworm aggregations at the Gulf of Mexico (Aharon and Fu 2003; Arvidson et al. 2004). Previous to this study, there was no data available from deeper than 25 cm...... 118 Figure 5-11: The ratio between sulfide stable isotope values and sulfide concentrations plotted against depth, “near” (< 30 cm away) or “far” (> 150 cm away) from tubeworms. The various lines show ratios predicted using simulations from the model, where sulfate released across tubeworm roots was varied between 0-100%. The best match between the empirical data (collected in 2004) and the model is found at 83% root sulfate release. All data shown in this figure were collected in this study...... 119 Figure 5-12: Maximum individual growth (cm yr-1) predicted by the model for a one- meter long tubeworm, at various integrated sulfate reduction rates, as well as various levels of root sulfate release...... 125 Figure A-1: A mixture of cells isolated from L. luymesi trophosome viewed under bright field (left) and viewed by DAPI staining (right). Putative bacteriocytes containing symbionts are indicated. Some cells appeared to contain myelin structures, indicative of degenerating bacteriocytes in which symbionts have been digested (Dr. Monika Bright, personal communication)...... 136 Figure A-2: FISH using bacteriocyte cell preparations. Panels on the left show binding of FISH probes using fluorescent filters, whereas panels on the right show bright field

x views of the same area of the slide. Scale bars represent 10 µm. (a) Binding of general eubacterial probe to symbiont cells, viewed using an FITC filter; (b) Negative control showing lack of binding of the general archaeal probe to cells...138 Figure A-3: Percentage survival of bacteriocytes exposed to solutions of different compositions, and kept either on ice, or at room temperature...... 138 Figure A-4: Preliminary currents obtained using the “whole-cell” mode from patch clamping isolated L. luymesi bacteriocytes. The pipette solution contained 128 mM K+, 0 mM Na+, and 136.6 mM Cl-, whereas the bath solution contained 460 mM K+, 0 mM Na+, and 580mM Cl-...... 139 Figure B-1: “Initial arm choice”, or percentage of snails that entered either the test or the control arms of the Y-maze first. Bars with asterisks show significant (P<0.05) preference for the test arm...... 151 Figure B-2: The mean ± SE of the proportions of time spent by snails in the test or control arms. Bars with asterisks indicate significant difference (P < 0.05) between times spent in the test and control arms...... 151 Figure C-1: Change in shell length standardized to one year (“ adjusted shell growth”) versus initial shell length of transplant and control mussels from the reciprocal transplant between BP and GC234...... 176 13 15 34 Figure C-2: Mean values of tissue δ C, δ N and δ S of transplant and control mussels from the reciprocal transplant between a-c) BP and GC234, and d-f) BP and BH. Notations for transplants and controls have the name of the origin site first, followed by the name of the host site. For example, BP-BP represents BP controls, and GC- BP represents GC234 to BP transplants. Controls were of two types: manipulated and unmanipulated (see text for details). n = sample size. Error bars represent standard error. Different letters indicate a significant (p < 0.05) difference between data points...... 177 13 15 Figure C-3: Relationships between tissue ∆δ C and ∆δ N values of individual mussels belonging to the transplants between BP, and GC234 or BH...... 178 Figure C-4: Relationships between changes in carbon or nitrogen stable isotope values and shell growth represented by percent change in shell length (%CL) for BP to 13 GC234 transplant mussels and GC234 to BP transplant mussels. a, c) tissue Δδ C 15 vs. %CL and b, d) tissue Δδ N vs. %CL...... 179

xi LIST OF TABLES

Table 2-1: Mean ± standard error of sulfate and proton elimination by tubeworms over a period of 48 hours. Tubeworms were either “fed” sulfide for a period of 48 hours prior to the experiment, or were “starved” for a period of at least 96 hours prior to the experiment. Values with different letter labels indicate a significant difference, using the Bonferroni corrected significance level of 0.0025...... 29 Table 3-1: The total ATPase activity of various tissues examined in this study...... 56 -1 -1 Table 3-2: Estimated P- and V-type ATPase activities (in µmol Pi h g tissue wet weight) of L. luymesi plume (LP), L. luymesi root (LR), S. jonesi plume (SP) and S. jonesi root (SR). Proton-specific ATPase activities are in bold...... 58 Table 4-1: Average ± standard deviation of pH, salinity, sulfate and sulfide concentration, 34 and δ S of sulfate obtained from 12 dialysis bags placed at 10 cm above the sediment-seawater interface...... 82 Table 5-1: Dimensions of sediment “blocks”, as well as other parameters used for the model, such as sulfate and sulfide diffusion coefficients. SA= surface area; D = diffusion coefficient. The SA fraction (fSA) is the fraction of the total root ball hemisphere surface area...... 95 Table 5-2: Chemical and isotope compositions of pore water collected using three replicate peepers deployed in PF (plain-far) and PN (plain-near) locations. nd= non- detectable...... 103 Table 5-3: Chemical and isotope compositions of pore water collected using three replicate peepers deployed in OF (oily-far) and ON (oily-near) locations. nd= non- detectable...... 104 Table 5-4: Integrated sulfate reduction rates estimated from depth profiles at the four different locations sampled. The best-fit equation (corresponding to equation (1)) for each peeper profile is also shown...... 108 Table B-1: A description of the experimental treatments, with abbreviations used to refer to them in the text and figures. All incubations were done at 6°C with 10 gallons of continuously aerated synthetic seawater (SSW) kept in a glass tank. N = number of trials...... 146 Table C-1: Mean (standard deviation) of final shell length, condition index (CI), glycogen content and water content of transplant and control mussels. n = sample size. Values with different letters indicate a significant difference (p < 0.05) between them (statistical tests were performed on the arcsine transformed values)...... 175 Table C-2: Percent change in the tissue stable isotope ratios of transplant mussels in the presence and absence of shell growth, during the study period of about one year. Values were calculated using Equations (6) and (8). †These values could not be calculated as above, as growth of all BP to BH transplant mussels was negligible (< 0.1 %CL). ††This value could not be calculated, as the linear relationship between 13 2 Δδ C and %CL of BH to BP transplant mussels was not significant (p = 0.837; r = 0.012)...... 175

xii PREFACE

Contributions to the research and writing of multi-authored chapters and appendices:

Chapter 2: Lara Miles conducted live tubeworm experiments, and Matthew Barnabei assisted with tubeworm aquarium maintenance.

Chapter 3: Kathryn Luley assisted with protocol development, and Meredith Redding prepared tissue homogenates.

Chapter 4: Guy Telesnicki fabricated sampling devices (and helped with their design),

Kathryn Luley assisted with preparation of sampling devices before deployment,

Benjamin Predmore designed and manufactured de-oxygenation chambers, and Michael

McGinley assisted with the fabrication process.

Chapter 5: Dr. Michael Arthur conceptualized the geochemical model and provided guidance in model development, Guy Telesnicki fabricated sampling devices (and helped with their design), and Kathryn Luley assisted with preparation of sampling devices before deployment.

Appendix A: Laetitia Perfus performed patch-clamp experiments and Dr. Sally Assmann provided guidance.

Appendix B: Jonathan Martin assisted with Y-maze experiments and Dr. Robert Carney conceptualized Y-maze experiments.

Appendix C: Derk Bergquist assisted with data analyses and interpretations, Emily

Szalai performed growth and condition index analyses, and Dr. Stephen Macko determined tissue stable isotope values.

Dr. Charles Fisher provided funding, guidance and ideas for development of all of the projects and made contributions to the text in chapters.

xiii ACKNOWLEDGEMENTS

I would like to thank Chuck Fisher for providing guidance through this dissertation, for giving me the freedom to develop my own ideas and projects, and providing me with the amazing opportunity of diving in submersibles. The members of my Ph.D. committee, Mike Arthur, Sally Assmann, and Simon Gilroy, have all been intimately involved with my thesis projects and provided invaluable guidance. Mike

Arthur has helped me develop a much better understanding of geochemistry, and introduced me to the world of theoretical modeling. I am extremely grateful for his patience and guidance. Sally Assmann allowed me to use her lab facilities and gave me her valuable time to guide me through my patch-clamping attempts. Simon Gilroy helped me develop better biochemical ‘ethics’ and provided many great ideas along the way.

Several aspects of my research would have been impossible without the help and dedication of the pilots of the Johnson Sea-link submersible, and the crew of the research vessel Seaward Johnson II. I’d also like to thank Laetitia Perfus and Burt Thomas for their generous help with experiments and analyses.

My thesis work was made possible through the help of some extremely creative and dedicated undergraduate students. Lara Miles was exceptional with her motherly care for tubeworms exposed to various unfriendly inhibitors. Kate Luley is by far my favorite research cruise partner. I am grateful for our perfectly coordinated five AM sessions of filling countless dialysis bags with deoxygenated seawater on a rocking ship, all without exchanging a word! Meredith Redding took immense care and pride in preparing hundreds of tissue homogenates once she overcame her morbid fear of the -70˚ freezer.

My colleagues in the Fisher lab, past and present, have been entertaining and supportive

xiv throughout the years. Derk Bergquist helped me get started in the lab and introduced me to the world of caving, which I am now hoping to convert into a career. Stephane

Hourdez has provided help, advice (and delicious crepes!) on many occasions. Breea

Govenar has been a close friend and my introduction to the US and its strange grocery stores. Sue Carney has been a great inspiration for sticking through seemingly impossible research projects. Erik Cordes and John Freytag were great to discuss tubeworm science with, and Mark van Horn and Guy Telesnicki made my crazy sampling device ideas a reality. Guy Telesnicki and Kevin Zelnio have been great solace in moments of panic and

Stephanie Lessard-Pilon has been a patient and understanding office-mate during the stressful thesis writing process. Erin McMullin and Gioia Massa, my housemates for two years, taught me to move at a comfortable pace and enjoy my life outside of work.

Speaking of life outside the lab! Michael Fitzgerald has been the best support and friend I could have ever asked for. His patience and intelligence are admirable. My many climbing and biking friends- Jim Bowers, Ylva Lekberg, Denise Woodward, Doug

Fisher, Ieva Perkons, Scott Woods, Kathleen McNally, Brian McCall and Halla

Olafsdottir made my life here in State College a lot of fun. Finally, my sister Shahana

Dattagupta has always been my closest ally, and my parents are the reason I chose to enter the world of science and academics.

Chapter 2 was accepted for publication in the Journal of Experimental Biology and is reproduced here with permission from the Mandy Knowles (Editorial administrator/Editor, JEB). Appendix 2 was previously published in Limnology and

Oceanography 49 (4), pp 1144-1151, and is reproduced here with permission from the publisher (The American Society of Limnology and Oceanography).

xv CHAPTER 1

An introduction to vestimentiferan tubeworms associated with hydrocarbon seepage in the Gulf of Mexico

Cold seep communities

Cold seeps are located worldwide at geologically active and passive margins, and are characterized by the upwelling of reduced fluids to the sediment surface often from deeply buried reservoirs (Sibuet and Olu 1998). The first cold seep system was discovered more than 20 years ago at the Florida Escarpment in the Gulf of Mexico, and the taxa found at this site strongly resembled those found in deep-sea hydrothermal vent communities (Paull et al. 1984). Chemosynthetic organisms such as clams, mussels and tubeworms dominate cold seep macrofaunal communities (Levin 2005; Sibuet and Olu

1998), and these organisms commonly derive nutrition using chemoautotrophic symbionts that can fix carbon using energy derived from reduced chemicals associated with hydrocarbon seepage (reviewed in Fisher 1990; Fisher 1996).

Vestimentiferan tubeworms, numerically dominated by the taxonomic groups of escarpids and lamellibrachids, are found at hydrocarbon seeps worldwide, including continental margins on both sides of the Pacific and Atlantic oceans (Mcmullin et al.

2003 and references therein). These tubeworms are thiotrophs, i.e. they require sulfide for their growth and nutrition (Gardiner and Jones 1993). At most hydrocarbon seeps, sulfide is produced through microbial reduction of seawater sulfate, coupled with oxidation of methane or oil (Boetius and Suess 2004; Joye et al. 2004; Levin 2005; Treude et al.

2003). Thus the biology of vestimentiferan tubeworms is closely linked with biogeochemical processes occurring in their habitats.

1 Gulf of Mexico hydrocarbon seep habitat characteristics

Seepage of hydrocarbons is common at several sites on the continental slope of the Gulf of Mexico. At these sites, oil and gas migrate to the seafloor from reservoirs that are buried deep in the sediment, and are widely distributed across the slope (MacDonald

1998). Many cold seeps are located at active margins where processes associated with plate tectonics drive hydrocarbon migration (Boetius and Suess 2004; Levin 2005; Sibuet and Olu 1998). However, the Gulf of Mexico is a passive margin, and migration of hydrocarbons is driven instead by salt tectonics.

During the late Triassic, rifting of the Pangean supercontinent formed a basin in present day Gulf of Mexico, which subsequently dried up during the Jurassic (Macdonald and Fisher 1996; Pindell 1985). In the Middle Jurassic, movement of landmasses caused the gulf to be almost completely landlocked, which led to massive evaporite formation

(Pindell 1985). Constant incursions of seawater into this basin led to the formation of a thick salt deposit known as the Louann salt formation (Jenyon 1986; Sassen et al. 2001).

In the late Jurassic, the gulf reopened to seawater circulation, and the basin was rapidly loaded with siliclastic sediments brought in primarily by rivers draining the Sierra Madre

Mountains and the Rockies (Kennicutt Ii et al. 1992b; Pindell 1985). Due to the weight of overlying layers, the sediments compacted into layers of sandstone and shale (Mcgookey

1975). The incompressible salt layer began to move and deform in response to the pressure caused by its overburden, forming pillar-like structures called salt diapirs and faults in the overlying rock layers (Jenyon 1986; Macdonald and Fisher 1996). This tectonic action of the salt created traps for hydrocarbons and pathways for migration of oil and gas to the seafloor (Brooks et al. 1985; Macdonald 1998). The bulk of this oil and

2 gas originated from deeply buried (6-10 km) carbonate or shale Mesozoic source rocks

(Kennicutt Ii et al. 1992b; Sassen et al. 2001) that gave rise to shallower (2-3 km) oil and gas accumulations (Sassen et al. 1999).

A majority of chemosynthetic communities are found at the margins of salt structures (Roberts and Aharon 1994), where migration of hydrocarbons is concentrated.

Microbes in the sediment oxidize hydrocarbons using sulfate as a terminal electron acceptor (Aharon and Fu 2000), a process that produces sulfide and carbonate ions

(Macdonald 1998). The carbonate ions are precipitated to form authigenic carbonate rock, large quantities of which provide a substrate for settlement of tubeworms and mussels (Macdonald 1998; Sassen et al. 1994). The carbonate rock also serves to trap hydrocarbons in underlying sediments, enhancing hydrocarbon-driven microbial sulfate reduction (Sassen et al. 1994). The prevalence of chemosynthetic communities on the

Gulf of Mexico slope may also be related to the presence of gas hydrates (Carney 1994), which are ice-like structures in which gases are enclosed in a crystalline water lattice and are formed under conditions of low temperatures and high pressure (Brooks et al. 1985;

1984). Methane gas is a major component of gas hydrates and can be formed from oils either by thermogenic processes deep in the sediment, or by microbial (biogenic) degradation in relatively shallow sediment (Brooks et al. 1984). High concentrations of methane can be released from these hydrates at points of local instability, causing enhanced sulfate reduction in the sediment (Carney 1994).

Mats of sulfide-oxidizing Beggiatoa bacteria growing on the sediment-water interface may also cause higher rates of sulfate reduction in the sediment, by providing a barrier for escape of hydrocarbons into the water column (Sassen et al. 1993). These mats

3 are often considered to be good indicators of areas of high seepage rates (Joye et al. 2004;

Torres et al. 2002; Treude et al. 2003). Beggiatoa mats are usually either pigmented with an orange coloration, or non-pigmented with a yellow-white coloration. Non-pigmented

Beggiatoa are most likely chemoautotrophic, whereas pigmented Beggiatoa are likely heterotrophic, subsisting on by-products of microbial degradation of oil (Nikolaus et al.

2003; Zhang et al. 2005).

Biogeochemistry of hydrocarbon-rich sediments in the Gulf of Mexico

In shallow marine sediments, phytodetritus and organic matter of terrestrial origin provide a rich source of reduced carbon compounds that can be degraded by microbes in the sediment (Berner 1980). Oxygen gets depleted within the first few millimeters below the sediment-seawater interface, below which nitrate and sulfate are the preferred terminal electron acceptors. Since sulfate is abundant in seawater (28-29 mM), it is commonly reduced by anaerobic microbes that breakdown organic matter. At

600-1000 m depths on the continental slope of the Gulf of Mexico, the supply of labile organic matter from terrestrial sources is very limited (Lin and Morse 1991). Instead, a majority of sulfate reduction at these depths is driven by anaerobic oxidation of hydrocarbons such as methane and crude oil (Aharon and Fu 2000; Aharon and Fu 2003;

Joye et al. 2004). The product of sulfate reduction, hydrogen sulfide, sustains the rich communities of tubeworms found at these depths.

At some hydrocarbon seeps such as the Hydrate Ridge on the Cascadia margin, anaerobic oxidation of methane (AOM) fuels extremely high rates of sulfate reduction

(Boetius and Suess 2004; Treude et al. 2003). AOM has garnered a lot of attention in the

4 recent years, as it is an important component of the global carbon cycle (Valentine 2002).

AOM is sometimes mediated by consortia between archaea belonging to the family

Methanosarcinales and sulfate reducing bacteria (SRBs) of the

Desulfosarcina/Desulfococcus group (Boetius et al. 2000; Valentine 2002). In these consortia, the archaea are believed to carry out the process of ‘reverse methanogenesis,’ to oxidize methane, and transfer a reduced intermediate compound to the SRBs that mediate sulfate reduction (Boetius and Suess 2004; Valentine 2002).

However, in the hydrocarbon-rich sediments in the Gulf of Mexico, AOM and sulfate reduction appear to be loosely coupled, indicating that sulfate reduction is fuelled mostly by hydrocarbons other than methane (Joye et al. 2004; Orcutt et al. 2005). Gulf of

Mexico sediments contain crude oil, and abundant hexane extractable aliphatic and aromatic hydrocarbons (Brooks et al. 1984; Kennicutt Ii et al. 1992b). The degradation of hydrocarbons by microorganisms growing under strictly anaerobic conditions was first demonstrated in the late 1980s (Widdel and Rabus 2001). Subsequently, a large number of strains of SRBs have been cultured that can use aliphatic and aromatic hydrocarbons as their sole carbon source. Moreover, microbial sulfate reduction coupled with oxidation of crude oil has also been demonstrated to occur under strictly anaerobic conditions (Rueter et al. 1994). It appears oxidation of crude oil and associated hydrocarbons drives a majority of the microbial sulfate reduction that sustains thiotrophic organisms at Gulf of

Mexico hydrocarbon seeps (Joye et al. 2004).

5 Vestimentiferan tubeworm physiology

Lamellibrachia luymesi (Gardiner and Hourdez 2003) and Seepiophila jonesi

(Gardiner et al. 2001) are the dominant species of vestimentiferan tubeworms found on the upper continental slope of the northern Gulf of Mexico (Bergquist et al. 2002;

Bergquist et al. 2003). Vestimentiferan tubeworm as a group are considered a clade within the Family Siboglinidae, Phylum Annelida, Class Polychaeta (Mcmullin et al.

2003 and references therein). The body of a vestimentiferan tubeworm contains an anteriorly located, well-vascularized gas exchange organ called the plume (Arp et al.

1985, Figure 1-1). Below that is a muscular region called the vestimentum that contains the heart and the brain (Jones 1981). Vestimentiferan tubeworms do not have a mouth or gut; instead they derive their nutrition from intracellular sulfide-oxidizing endosymbiotic bacteria housed in an organ called the trophosome (Cavanaugh et al. 1981; Jones 1981) located in the trunk region of the worm. At the very anterior end of the worm is the segmented opisthosome, which helps to anchor the worm in its tube (Childress and Fisher

1992).

Figure 1-1: Internal anatomy of Riftia pachyptila showing the location of the major organs and blood vessels. Drawing adapted from Arp et al. (1985).

A majority of information regarding the physiology and biochemistry of vestimentiferan tubeworms has been obtained through studies on Riftia pachyptila, the hydrothermal vent tubeworm found at the Galapagos Rift and the East Pacific Rise

(Childress and Fisher 1992). The chemoautotrophic symbionts of vestimentiferan tubeworms utilize the energy from sulfide oxidation to fix carbon using the Calvin cycle, for which they require a regular supply of sulfide, carbon dioxide and oxygen (Childress et al. 1984; Scott et al. 1999). R. pachyptila obtains all three chemicals across its plume

(Arp et al. 1985) and transports them to its symbionts using its vascular blood (Childress et al. 1984)). The trophosome, where the symbionts are located, is highly vascularized and bathed by non-circulating coelomic fluid (Cavanaugh et al. 1981; Gardiner and Jones

1993). This fluid is in equilibrium with the vascular blood and can act as a store for carbon dioxide, oxygen and sulfide (Childress et al. 1984). Both vascular and coelomic

7 fluids contain extracellular hemoglobins that can bind oxygen and sulfide simultaneously, but at separate sites on the protein molecules (Arp et al. 1987; Zal et al. 1996; Zal et al.

1998) The hemoglobins bind oxygen and sulfide with high affinity (Arp and Childress

1983) and provide the symbionts with an environment that has low free-sulfide levels despite the high concentrations of bound sulfide (Fisher et al. 1988). This leads to high levels of carbon fixation by the symbionts (Fisher et al. 1989) and prevents the potential toxic effects of free sulfide, such as inhibition of mitochondrial cytochrome C oxidase

(Hand and Somero 1983). The tubeworm then derives nutrition either via the release of small carbon and nitrogen containing molecules such as succinate and glutamate from the symbionts (Bright et al. 2000; Felbeck and Jarchow 1998) or by digestion of the symbionts (Bright et al. 2000; Hand 1987).

Transport of nutrients into the body of R. pachyptila from its environment has been studied in great detail. R. pachyptila obtains carbon from its environment through

- diffusion of gaseous CO2 and not through uptake of bicarbonate (HCO3 ) ions (Childress

et al. 1993; Goffredi et al. 1997b). CO2 is present at high concentrations in the vent environment at the East Pacific Rise due to the elevated DIC and acidic nature of vent waters (pH 5-6; Goffredi et al. 1999b; Scott et al. 1999). R. pachyptila can concentrate

CO2 in its body fluids to levels far exceeding those present in its environment, by maintaining an alkaline extracellular pH and having high concentrations of carbonic anhydrase in its plume (Goffredi et al. 1999b; Kochevar et al. 1993). Carbonic anhydrase

- catalyses the fast transformation of CO2 into HCO3 , maintaining a strong gradient for diffusion of CO2 across the plume (Goffredi et al. 1997b). Carbonic anhydrase activity is

- also associated with the trophosome (Kochevar et al. 1993), where HCO3 is converted

8 back into CO2, the only form of carbon utilized by the RUBSICO of symbionts to fix carbon via the Calvin cycle (Scott et al. 1999). Hydrogen sulfide, on the other hand, is

- obtained not by simple diffusion of H2S, but by uptake of HS ions, which is also the ionic species bound by the blood (Childress et al. 1984; Goffredi et al. 1997a). The molecular mechanism of HS- transport, or reason why the plume membrane is not

- permeable to H2S has not been elucidated. However, it is proposed that HS uptake is facilitated since it is bound with high affinity by hemoglobins, inside the organism

(Goffredi et al. 1997a; Goffredi et al. 1998).

The symbionts of vestimentiferan tubeworms produce sulfate and protons as products of their metabolism. In R. pachyptila, internal sulfate concentrations (22-25 mmol L-1) do not exceed the range of values found in other marine invertebrates without sulfide-oxidizing symbionts (Goffredi et al. 1999a). Thus R. pachyptila has an efficient, though unknown, mechanism for excretion of sulfate ions. Seawater surrounding the plumes of R. pachyptila has a higher concentration of sulfate than inside the body of the animal. Thus the mechanism for removal of sulfate is likely to be through active membrane transport, requiring energy (Goffredi et al. 1999a). Similarly, the elimination of protons from R. pachyptila is likely to represent a high metabolic cost for the animal

(Girguis et al. 2002). Protons produced by symbionts of R. pachyptila are efficiently removed by unusually high H+- ATPase activity across the plumes of the animals

(Girguis et al. 2002; Goffredi and Childress 2001).

The information obtained from the studies of R. pachyptila can provide valuable insights into the biology of L. luymesi and S. jonesi. However, important distinctions are likely to occur. While R. pachyptila is adapted to the ephemeral vent environment, and

9 maintains a very high growth rate (Lutz et al. 1994), L. luymesi and S. jonesi are found in the stable hydrocarbon seep environment where low levels of sulfide may be available for centuries (Fisher et al. 1997). These seep tubeworms grow very slowly, and age estimates of L. luymesi exceed 170- 200 years (Bergquist et al. 2000; Fisher et al. 1997). Moreover, sulfide concentrations around plumes of L. luymesi were found to be extremely low, whereas concentrations in the sediment below the tubeworms are typically much higher

(Freytag et al. 2001; Julian et al. 1999). Both L. luymesi and S. jonesi grow long and thin posterior extensions (called “roots”) of their bodies and tubes into the sediment. The root tubes of L. luymesi were found to be highly permeable to sulfide (Julian et al. 1999), and

net autotrophic uptake of CO2 across their plumes was demonstrated in the presence of sulfide uptake from the roots (Freytag et al. 2001). Thus it is likely that L. luymesi obtains a significant proportion of its sulfide from across its roots. The role of S. jonesi roots in sulfide uptake has not yet been investigated. Another important distinction between the

vent and the seep environment is the availability of CO2 to tubeworms. In contrast to the

- vent environment where levels of CO2 are high due to acidic vent waters, HCO3 is likely to be the predominant chemical species around plumes of seep tubeworms. Thus seep tubeworms may have adaptations distinct from that of R. pachyptila to take up inorganic carbon from their environment.

Hypotheses and questions addressed by this thesis

The discovery of uptake of sulfide across roots of L. luymesi initiated questions regarding other possible roles of the roots of these tubeworms. Julian et al (1999) suggested that L. luymesi could use its roots to introduce sulfate into the sediment. Since

10 sulfide is produced in the sediment by microbial reduction of sulfate, sulfate concentrations in the sediment could be one of the limiting factors for sustained sulfide production over the long life spans of tubeworms aggregations (Cordes et al. 2005a;

Cordes et al. 2003). Sulfate ions and protons are metabolic byproducts of symbiont- driven oxidation of sulfide in tubeworms (Girguis et al. 2002; Goffredi et al. 1999a). As in the case of R. pachyptila, effective removal of these end products is likely to be an important adaptation for a successful symbiosis. Sulfate concentrations around the plumes of seep tubeworms are likely to be the same as typical oceanic seawater values

(about 29 mmol L-1). However, sulfate is likely to be depleted in the sediment surrounding roots of these tubeworms, due to removal by sulfate-reducing microbial activity (Aharon and Fu 2000; Arvidson et al. 2004). Protons are also consumed in the sediment due to the same reason. Thus, release of sulfate and protons through the roots may occur through passive transport processes that do not impose the high energetic demands of active transport across the plume.

Thus, both L. luymesi and S. jonesi worms may benefit from using their roots for sulfate and proton release. However, the role of S. jonesi roots in sulfide uptake has not been demonstrated to date. L. luymesi and S. jonesi tubeworms almost always co-exist in aggregations at hydrocarbon seep sites on the upper Louisiana slope (Bergquist et al.

2002). While plumes of adult L. luymesi tubeworms can be situated 2 m above the sediment surface where sulfide levels are extremely low (Freytag et al. 2001), S. jonesi tubeworms grow with their plumes close to the sediment surface where sulfide levels may be much higher. Thus it is possible that S. jonesi may use their plumes, in addition to

11 their roots, to obtain sulfide. These differences in growth habits of L. luymesi and S. jonesi may lead to distinct ion transport mechanisms in these two species of tubeworms.

The hypotheses stated above pose several different questions, which were addressed in this thesis: (1) what are the pH values and concentrations of sulfate, sulfide and methane in the sediment surrounding the roots of L. luymesi and S. jonesi tubeworms? How do these compare with concentrations of sulfate and pH in the blood of these animals? Do stable isotope values of sulfate and sulfide in the sediment reflect sulfur recycling? (2) Do L. luymesi and S. jonesi tubeworms release sulfate across their roots? If so, what transport mechanisms do they employ? (3) Do L. luymesi and S. jonesi have high H+-ATPase activities like those demonstrated in R. pachyptila? If so, which exchange surface is this activity associated with: the plumes or the roots? What differences exist between the two species of tubeworms in this regard?

In this thesis, I used a multi-disciplinary approach to address these questions.

Chapter 2 describes the results of whole animal experiments performed with L. luymesi tubeworms that were maintained alive in the laboratory. Chapter 3 describes in vitro experiments to analyze H+-ATPase activities of tissues collected from L. luymesi and S. jonesi. Chapter 4 describes the development of a sampling device that was necessary for characterization of the in situ geochemistry of sediments surrounding tubeworm aggregations. In Chapter 5, the geochemical environment of tubeworms is described and a theoretical model is developed to help explain patterns observed in in situ data.

Appendix A describes an attempt to patch-clamp bacteriocytes isolated from L. luymesi.

Appendices B and C describe studies that were conducted as independent projects unrelated to the main hypothesis addressed by this thesis. Appendix B describes the

12 results of a behavioral study conducted using Bathynerita naticoidea snails collected from hydrocarbon seep sites in the Gulf of Mexico. Appendix C describes the tissue stable isotope turnover of Bathymodiolus childressi mussels transplanted between 3 different hydrocarbon seep sites in the Gulf of Mexico. Both B. naticoidea and B. childressi co-occur with the vestimentiferan tubeworms that were the main focus of this thesis.

13 CHAPTER 2

The hydrocarbon seep tubeworm Lamellibrachia luymesi primarily eliminates sulfate and hydrogen ions across its roots to conserve energy and ensure sulfide supply

Sharmishtha Dattagupta1, Lara L. Miles1, Matthew S. Barnabei1, and Charles R. Fisher1

In press in The Journal of Experimental Biology (date accepted 6/27/06)

1 Department of Biology, The Pennsylvania State University, University Park, PA-16802.

14 Abstract

Lamellibrachia luymesi (Polychaeta: Siboglinidae) is a deep-sea vestimentiferan tubeworm that forms large bush-like aggregations at hydrocarbon seeps in the Gulf of

Mexico. Like all vestimentiferans, L. luymesi obtains its nutrition from sulfide-oxidizing endosymbiotic bacteria, which it houses in an internal organ called the trophosome. This tubeworm has a lifespan of over 170 years and its survival is contingent upon the availability of sulfide during this long period. In sediments underlying L. luymesi aggregations, microbes produce sulfide by coupling sulfate reduction with hydrocarbon oxidation. L. luymesi acquires sulfide from the sediment using a root-like posterior extension of its body that is buried in the sediment. Its symbionts then oxidize the sulfide to produce energy for carbon fixation, and release sulfate and hydrogen ions as byproducts. It is critical for the tubeworm to eliminate these waste ions, and it could do so either across its vascular plume or across its root. In this study, we measured sulfate and proton elimination rates from live L. luymesi and found that they eliminated approximately 85% of the sulfate produced by sulfide oxidation, and approximately 67% of the protons produced by various metabolic processes, across their roots. On the basis of experiments using membrane transport inhibitors, we suggest that L. luymesi has anion exchangers that mediate sulfate elimination coupled with bicarbonate uptake. Roots could be the ideal exchange surface for eliminating sulfate and hydrogen ions for two reasons.

First, these ions might be eliminated across the root epithelium using facilitated diffusion, which is energetically economical. Second, sulfate and hydrogen ions are substrates for bacterial sulfate reduction, and supplying these ions into the sediment might help ensure

L. luymesi a sustained sulfide supply over its entire lifespan.

15 Introduction

Elimination of metabolic waste products is vital for all living organisms. Excess

nitrogen, carbon dioxide and water comprise the major metabolic wastes of most

invertebrates (Brusca and Brusca 2002). However, obligate symbioses including

vestimentiferans have the additional burden of eliminating waste products generated by

their symbionts. Vestimentiferan tubeworms lack a mouth or gut, and derive their

nutrition from intracellular endosymbiotic bacteria housed in an organ called the

trophosome (Cavanaugh et al. 1981; Jones 1981). These bacteria oxidize hydrogen

sulfide and produce sulfate and hydrogen ions as byproducts, and the tubeworm’s

mechanism for eliminating these wastes is among its suite of adaptations crucial for this

symbiosis (Girguis et al. 2002; Goffredi et al. 1998; Goffredi et al. 1999a).

Lamellibrachia luymesi (Gardiner and Hourdez 2003) is a vestimentiferan

tubeworm associated with hydrocarbon seepage at 400-1000 m depths on the upper

Louisiana slope of the Gulf of Mexico. It forms one to two meter tall bush-like

aggregations that can contain over one thousand individuals, and provide habitat for a

wide variety of heterotrophic fauna (Bergquist et al. 2003; Cordes et al. 2005a; Cordes et

al. 2005b). L. luymesi grows extremely slowly, and has a lifespan of at least 170- 250

years (Bergquist et al. 2000; Fisher et al. 1997). Like all vestimentiferans, it requires

sulfide for its survival and growth. Sulfide is produced within sediments at hydrocarbon

seeps by microbial sulfate reduction coupled with oxidation of hydrocarbons such as

methane and oil (Aharon and Fu 2003; Boetius et al. 2000; Joye et al. 2004).

While sulfide levels around the anterior gill-like plumes of adult tubeworms are

generally less than 0.1 µmol l-1 (Freytag et al. 2001), levels in the sediment underlying

16 them are typically greater than 1.5 mmol l-1 (Julian et al. 1999). L. luymesi mines sulfide from the sediment using long, root-like, posterior extensions of its body (Freytag et al.

2001; Julian et al. 1999). Large L. luymesi aggregations can have extensive networks of roots more than a meter deep in the underlying sediment (Figure 1-1). These roots extend below the point at which the tubeworm tubes are attached to a carbonate rock substrate

(Bergquist et al. 2002; Fisher et al. 1997).

Figure 2-1:The extensive tubeworm roots of an aggregation that was excavated and collected using the Bushmaster sampling device (Bergquist et al., 2002; Cordes et al., 2004). This photograph shows roots that are about 1 meter long, extending below the carbonate rock substrate. The roots were observed to break during collection and we believe they might have been even longer in the intact aggregation (CRF, pers. obs.). Photo credits: Dr. E. E. Cordes.

Because sulfate and protons are the two major waste products of chemoautotrophic sulfide oxidation (Girguis et al. 2002; Goffredi et al. 1998; Goffredi et al. 1999a), L. luymesi must possess mechanisms for eliminating sulfate and hydrogen ions. The hydrothermal vent tubeworm Riftia pachyptila is relatively well studied, and has been a model for understanding vestimentiferan physiology. R. pachyptila does not grow a root,

17 but uses its vascular plume to obtain sulfide and exchange other nutrients with its external environment (Arp et al. 1985). Its blood sulfate concentration is 22-25 mmol l-1 (Goffredi et al. 1999a), which is lower than the seawater sulfate concentration of approximately 29 mmol l-1. Thus R. pachyptila is thought to eliminate sulfate ions across its plume surface using membrane transporters that pump sulfate against the concentration gradient, consuming energy in the process (Goffredi et al. 1999a). Likewise, R. pachyptila eliminates protons using a high concentration of proton-specific ATPases in its plume membrane, and this process represents a substantial metabolic cost for the tubeworm

(Girguis et al. 2002; Goffredi and Childress 2001).

Internal pH and sulfate concentrations of L. luymesi have not been previously reported. If they were similar to those of R. pachyptila, L. luymesi would also have to utilize active membrane transport to eliminate sulfate and hydrogen ions across its plume into the surrounding seawater. However, unlike R. pachyptila, L. luymesi could conceivably use its root for waste elimination. Microbial sulfate reduction depletes sulfate and hydrogen ions from the sediment pore-water surrounding L. luymesi roots, creating a favorable gradient for these ions to diffuse out of the roots. L. luymesi could avoid the high energetic demands of sulfate and proton elimination across its plume by passive transport of these ions across its roots.

L. luymesi could derive another important benefit from eliminating sulfate across its roots. In hydrocarbon rich sediments where L. luymesi are found, microbial sulfide production can become limited by sulfate availability (Arvidson et al, 2004; Joye et al,

2004). A number of authors have speculated that L. luymesi might sustain microbial sulfide production by “irrigating” the sediments with sulfate (Cordes et al. 2005a; Cordes

18 et al. 2003; Freytag et al. 2001; Julian et al. 1999). Cordes et al. (2005a) modeled the

sulfide sources and demands of mature tubeworm aggregations and concluded that their

demands could be satisfied over their entire lifespan only if individuals supplied sulfate

across their roots into the sediment.

In this study, we measured the sulfate concentration and pH of L. luymesi body

fluids and confirmed that elimination of these ions across the root surface would be

energetically favorable in their natural habitat. We conducted laboratory experiments

with live L. luymesi to measure sulfate and hydrogen ion elimination rates across their

plume and root surfaces, and used anion transport inhibitors to begin to examine the

molecular mechanism by which sulfate elimination occurs.

Materials and Methods

Collection and maintenance of tubeworms

Small clumps of tubeworms about 50 cm long were collected from the site

GC234, located at a depth of approximately 540 m on the upper Louisiana slope of the

Gulf of Mexico (27°44.7’N, 91°13.3’W). The robotic claw of the Johnson Sea-Link

submersible was used to grasp the carbonate rock to which the tubeworm clumps were

attached and place them in a temperature- insulated box for transport to the surface. After

collection, the tubeworms were maintained on board the ship for up to 15 days in a cooler

with cold (6 °C) circulating seawater. Once a day, about one-third of the seawater in the

cooler was replaced with clean seawater and the tubeworms were “fed” for about one

19 hour by transferring them to a 20-liter bucket containing a 500 µmol l-1 sodium sulfide solution in cold seawater. They were then returned to the maintenance cooler.

Figure 2-2: Schematic representation of a tubeworm maintenance aquarium. Sulfide dissolved in synthetic seawater (SSW) was introduced into the polyvinylchloride (PVC) hose, which was connected to PVC grating. The sulfide diffused out from the grating through the crushed coral and sediment layers. The aquarium was filled with SSW, which was kept aerated and filtered.

The tubeworms were transported in chilled seawater to the laboratory at the

Pennsylvania State University where they were maintained at ambient pressure in specially designed aquaria inside cold rooms (6 °C). The base of each aquarium was fitted with a polyvinylchloride (PVC) grating with holes drilled into it, which was connected to a hose that could be used to introduce sulfide under the sediment in the aquarium (Figure 2-2). The grating was covered with a layer of crushed coral, above which was a ~10 cm layer of sediment collected from the seafloor of the Gulf of Mexico.

The remainder of the aquarium was filled with synthetic seawater (SSW), made using

20 Reef Crystals® (Aquarium Systems Inc.). The SSW in the aquariums was filtered and aerated using a flow-through aquarium filtration system, and about 10% of it was replaced with freshly made SSW once a week. The tubeworms were “fed” twice a week by adding a stock solution containing 6 grams of sodium sulfide dissolved in about 7 liters of SSW through the PVC hose directly into the sediment below the tubeworms, at a rate of approximately 3 liters per hour.

Measurements of body fluid pH and sulfate concentrations

To obtain mixed body fluids (coelomic fluid and vascular blood), tubeworms were dissected within two to three hours after they were collected from the seafloor. The seawater from their tubes was drained, following which they were cut at a point approximately 10 cm above the posterior ends of their tubes. The first few drops of blood were discarded due to the possibility of contamination with seawater, and the remaining blood was collected for analysis. The blood pH was measured using a MI-710 micro- combination pH electrode (Microelectrodes, Inc., Londondery, New Hampshire, USA) connected to a high-impedance Blood Gas Analyzer system (Cameron Instrument

Company, Port Aransas, Texas, USA). The electrode was calibrated using IUPAC standards of pH 6.86 ± 0.01 and 7.41 ± 0.01 (Radiometer Analytical). Hemoglobins and other proteins were then removed from the blood by adding a 1: 1 volume of methanol, following which the solution was centrifuged at 8000 RPM for 10 minutes (Goffredi et al. 1999a). The pellet was discarded and the supernatant was diluted to 25 ml using double-distilled deionized water. The sulfate concentration in the supernatant was determined using turbidimetric analysis, details of which are described below.

Live tubeworm experiments

Sulfide spontaneously oxidizes to form sulfate in the presence of oxygen, so sulfate excretion from tubeworms could not be measured while simultaneously exposing the animals to sulfide. Instead, tubeworms were first exposed to sulfide so they accumulated bound sulfide in their blood, and then transferred to experimental chambers without sulfide to measure sulfate and proton excretion rates. The tubeworms used for our experiments were typically 30- 50 cm long, with wet tissue weights between 2-5 grams.

Assuming that these tubeworms contained 1-4 ml of blood, and the bound sulfide concentrations in their blood were 150 –170 µmol l-1 (Freytag et al. 2001), sulfide oxidation would produce approximately 0.15-0.8 µmoles of sulfate in total. This amount of sulfate would be undetectable in seawater, which has a sulfate concentration of 29 mmol l-1. Thus a sulfate-free artificial seawater (SF_ASW) medium was used when measuring tubeworm sulfate elimination rates. The SF_ASW contained 500 mmol l-1 sodium chloride, 9 mmol l-1 potassium chloride, 9.3 mmol l-1 calcium chloride, 48.5 mmol l-1 magnesium chloride, and 2.5 mmol l-1 sodium bicarbonate. The SF_ASW had a salinity of 36 ‰ and its pH was adjusted to 8.0 using a 0.5 mmol l-1 sodium hydroxide solution.

Sulfate elimination rates of freshly collected tubeworms were determined as a reference for comparison with rates of tubeworms maintained in aquariums. These rates were measured in a cold room (6-8°C) on board the ship using tubeworms that had been collected from the seafloor less than 24 hours previously. Seawater was drained from their tubes, and they were soaked in SF_ASW for 20 minutes to minimize the sulfate

22 carried over by their tubes into the experimental chambers. One or two animals were then inserted into a polycarbonate respirometer, which was filled with SF_ASW and re- circulated continuously using a peristaltic pump. At the end of 48 hours, the solution from the respirometer was collected and its sulfate concentration was determined. The tissue wet-weights of these tubeworms were estimated from the volume of seawater their soft tissues displaced in a 100 ml graduated cylinder, as there was no shipboard balance available.

The remaining experiments were performed with animals maintained in aquaria for a period between one and fifteen months. Before a tubeworm was used for an experiment, it was transferred from the maintenance aquarium to a 2-liter glass graduated cylinder containing SSW, where its plume coloration and reflexes were observed during a period of 15 - 20 hours to ensure that it appeared to be in good physiological condition. A tubeworm was selected for an experiment if it routinely extended its plume outside its tube, its plume was bright red in color, and if it reacted to a sharp knock on the glass cylinder by rapidly withdrawing its plume back into its tube. Before an experiment, tubeworms were “fed” by incubating them with a 500 µmol l-1 solution of sodium sulfide in SSW for 48 hours. SSW was drained from their tubes, and they were soaked in

SF_ASW for 20 minutes. They were then inserted into a “split-chamber” polycarbonate respirometer, which enabled separation of their anterior and posterior halves into distinct, watertight chambers (Freytag et al. 2001; Girguis et al. 2002). These chambers contained

SF_ASW that was continuously re-circulated using a peristaltic pump. A headspace of air supplemented the oxygen dissolved in the SF_ASW in the anterior chamber. Since the concentration of oxygen in air is about 25 times higher than that dissolved in seawater at

23 6˚C (Carpenter 1966), circulation of the SF_ASW through the headspace of air was intended to prevent hypoxic conditions. The volume of the headspace was approximately

50 ml for small worms (up to 3.5 grams wet weight) and approximately 100 ml for larger worms. At the end of 48 hours, the solutions from the plume and root chambers were collected and their sulfate concentrations and pH were determined. The worm was then removed from its tube, and its wet weight was measured.

“Fed” tubeworms were exposed to a sulfate-free medium during the above experiments. As a control for sulfate loss from the animal driven by the large sulfate gradient between their tissues and the SF ASW, we conducted experiments with

“starved” tubeworms. These experiments were performed using the same procedure described above for the “fed” tubeworms, except the tubeworms were incubated in SSW without sulfide for at least 96 hours prior to measuring their sulfate and proton elimination rates.

The effects of inhibitors on tubeworm sulfate and proton elimination rates were measured in a three-part experiment. The inhibitors we used were the general anion exchange inhibitors DIDS (4,4’-diisothiocyanatostilbene-2, 2’-disulfonic acid) and SITS

(4-acetamido-4’ isothiocyanato-2, 2’-stilbenedisulfonic acid), and the potent chloride transport inhibitor NPPB (5-nitro-2- (3-phenylpropylamino) benzoic acid) (Cabantchik and Greger 1992). In these experiments, the procedures used for “feeding” tubeworms and measuring elimination rates were the same as those described above with “fed” tubeworms. First, a tubeworm was “fed” sulfide for 48 hours and its “baseline” sulfate and proton elimination rates were measured. It was then removed from the experimental chamber and “fed” for another 48 hours, following which it was incubated with an

24 inhibitor solution for 3 hours. Its “inhibitor-exposed” sulfate and proton elimination rates were then determined. Finally, it was “fed” for another 48 hours, after which its “post- exposure” sulfate and proton elimination rates were measured. The wet weight of the tubeworm was then determined as above. The various inhibitor solutions used for these experiments were 0.2 mmol l-1 DIDS, 0.4 mmol l-1 SITS, and 0.1 mmol l-1 NPPB. To make an inhibitor solution, either DIDS, SITS, or NPPB was first dissolved in 250 µL of dimethyl sulfoxide (DMSO) and then mixed with 250 ml of SSW so that the final concentration of DMSO in the solution was 0.1% by volume. Control experiments were performed where tubeworms were exposed to a 0.1% solution of DMSO in SSW instead of an inhibitor solution. DIDS, SITS, NPPB and DMSO were all purchased from Sigma

Chemical Company.

Measurement of sulfate elimination rates

Sulfate concentrations of solutions collected from plume and root chambers of experiments described above were measured using barium chloride turbidimetry and all samples were analyzed in duplicate. Specifically, one SulfaVer® reagent packet (Hach

Chemical Company) was added to 25 ml of the sample, and the solution was vortexed.

The reaction was allowed to proceed for exactly 5 minutes, after which the absorbance of the sample was measured at 450 nm using a Beckman DU-64 spectrophotometer

(Beckman Coulter Inc., Fullerton, CA, USA). The sulfate concentration was calculated from a standard curve generated using serial dilutions of a 2500 mg l-1 sulfate standard

(Hach Chemical Company) diluted using SF_ASW. As the sulfate concentration in the

SF_ASW introduced into the plume and root chambers at the beginning of each

25 experiment was empirically determined to be zero, the sulfate measured in the final samples was assumed to be released by the tubeworms. Sulfate elimination rates were calculated from sulfate measurements, incubation times in chambers with SF_ASW, and tubeworm mass.

Measurement of proton elimination rates

pH of solutions collected from experimental chambers were measured to estimate the amount of protons released by tubeworms. The SF_ASW was first equilibrated with air by stirring it continuously for 48 hours at 6°C with a headspace of air above it, to ensure that it reached a steady pH value. A 50 ml sub-sample of the SF_ASW was kept aside in a sealed bottle for subsequent analysis of initial pH and buffering capacity. The SF_ASW was circulated with tubeworms inside split-respirometers for a period of 48 hours, at the end of which pH of the final plume and root chamber samples were measured. The pH of the initial SF_ASW was measured at the same time, and its buffering capacity was determined empirically by titration with a 0.1 mol l-1 hydrochloric acid solution. The titration curve was used to determine the amount of tubeworm proton release into the plume or root chambers necessary to cause the observed differences in pH between the initial and final SF_ASW solutions. Control experiments in which SF_ASW equilibrated as above was circulated in split-respirometers without tubeworms showed that there was no change in pH of SF_ASW in the absence of tubeworms. Proton elimination rates of tubeworms were calculated from proton measurements, incubation times in chambers with SF_ASW, and tubeworm mass.

26 Data analysis and statistics

The program MINITAB (Minitab Inc., State College, PA, USA) was used for all statistical analyses. Student’s t-tests were used to compare sulfate and proton excretion rates between “fed” and “starved” animals, and also for inhibitor studies. When the a priori expectation was that a treatment would reduce the rate of sulfate or proton release, one-sided t-tests were used. For inhibitor experiments where the same individual was used for more than one treatment, paired t-tests for were used for comparisons.

Correlations between parameters were analyzed using the Pearson method.

Results

Body fluid parameters

Mixed blood from freshly collected L. luymesi had sulfate concentration of 23.10

± 1.76 mol L-1 (mean ± SD; n = 15), and pH of 7.12 ± 0.14 (mean ± SD; n = 10).

Laboratory maintenance of tubeworms

Tubeworms were successfully maintained alive in aquaria in the laboratory at 6°C and under atmospheric pressure. Most of the tubeworms extended their plumes outside their tubes when sulfide-rich SSW was being introduced into their aquaria. The trophosomes of tubeworms maintained in aquaria were observed to be light to dark green in coloration, indicating the presence of elemental sulfur reserves (Pflugfelder et al.

2005). Some tubeworms incurred damage to their roots during collection from the seafloor, and a majority of these worms added new root tube material to the posterior ends at which their root tubes had been broken. Laboratory maintained tubeworms

27 eliminated sulfate at a rate of 0.457 ± 0.176 µmoles g-1 wet weight h-1 (mean ± SD, n=

32), which was not significantly lower (p = 0.304, one-sided t-test) than the rate at which freshly collected tubeworms eliminated sulfate (mean ± SD = 0.509 ± 0.144 µmoles g-1 wet weight h-1, n=3).

“Fed” tubeworms: sulfate and proton elimination rates

L. luymesi eliminated both sulfate and hydrogen ions at significantly higher rates across their roots than across their plumes (Table 1-1; sulfate: P= 0.0001, n=32; protons:

P=0.0003, n=15; two-sided paired t-tests). Plume sulfate elimination rates were weakly correlated with root sulfate elimination rates (Figure 2-3A; R= 0.31, P= 0.066).

Conversely, plume proton elimination rates were strongly correlated with root proton elimination rates (Figure 2-3B; R=0.91, P < 0.0001). Proton elimination rates were negatively correlated with sulfate elimination rates, when measured across the plumes

(Figure 2-4A; R= -0.69, P=0.012) and across the root (Figure 2-4B; R= -0.67, P= 0.017).

Total proton elimination rates (plume and root rates combined) were also negatively correlated with total sulfate elimination rates (Figure 2-4C; R= -0.78, P=0.003).

Sulfate Elimination Rate Proton Elimination Rate Experiment Type (µmoles h-1 g-1 wet weight) (µequivalents h-1 g-1 wet weight) Plume Root Plume Root Fed 0.163±0.014a 0.276±0.023a 0.138±0.040a 0.416±0.080c Starved 0.129±0.027a 0.094±0.029b 0.174±0.027a 0.427±0.029c

Comparison between “starved” and “fed” tubeworms

“Starved” tubeworms eliminated sulfate across their plumes at a slightly lower average rate than “fed” tubeworms, but this difference was not statistically significant

(Table 1-1; P = 0.143, one-sided t-test, df=26). On the other hand, “starved” tubeworms eliminated sulfate across their roots at a substantially lower rate than “fed” tubeworms, and this difference was highly significant (Table 1-1; P <0.0001, one-sided t-test, df=26).

“Starved” tubeworms had similar levels of proton elimination as “fed” tubeworms, across their plumes (Table 1-1; P = 0.771, one-sided t-test, df=9) as well as across their roots (P

= 0.530, one-sided t-test, df=9).

Figure 2-3: Relation between (A) sulfate (µmoles h-1 g-1 wet weight), and (B) protons (µequivalents h-1 g-1 wet weight) eliminated across plumes and roots of individual tubeworms exposed to sulfide for 48 hours prior to the measurement. Data points represented by closed circles in (B) correspond to animals that might have experienced hypoxic conditions and eliminated protons from anaerobic metabolism.

Figure 2-4: Relation between (A) plume, (B) root, and (C) total (plume and root combined) proton and sulfate eliminated from individual tubeworms exposed to sulfide for 48 hours prior to the experiment. Data points represented by closed circles correspond to animals that might have might have experienced hypoxic conditions and eliminated protons from anaerobic metabolism.

Figure 2-5: A summary of the effect of the membrane transport inhibitors, and the solvent (DMSO) on sulfate eliminated across plumes and roots of L. luymesi. An asterisk indicates that the treatment caused significant (P < 0.05) inhibition of sulfate elimination rate.

The effect of inhibitors on sulfate and proton elimination rates

Control experiments where tubeworms were exposed to 0.1% DMSO showed that this treatment did not significantly decrease sulfate elimination rates either across the plumes or roots of tubeworms (Figure 2-5A; plume: P=0.447; root: P=0.945; one-sided paired t-tests; n=5). On the other hand, exposure to DIDS or SITS caused a significant decrease in the sulfate elimination rates across roots of L. luymesi, as compared to

“baseline” values (Figure 2-5 B&C; DIDS: P= 0.045, one-sided paired t-test, n=7; SITS:

P=0.040, one-sided paired t-test, n=6). Neither inhibitor significantly decreased sulfate

31 elimination rates across plumes of these animals (DIDS: P= 0.229, one-sided paired t- test, n=7; SITS: P=0.188, one-sided paired t-test, n=6). DIDS appeared to have an irreversible effect on sulfate transport across roots, as sulfate elimination rates of “post- exposure” animals were significantly lower than that of “baseline” values (p=0.015; one- sided paired t-test, n=7). However, “post-exposure” sulfate elimination of SITS-exposed tubeworms was not significantly lower than the “baseline values” (P=0.779, one-sided paired t-test, n=5). Exposure to NPPB did not cause a significant decrease in sulfate elimination across either roots or plumes of tubeworms (Figure 2-5D; plume: P=0.317; root: P=0.327, one-sided paired t-tests; n=4). None of the above inhibitors had a significant effect on proton release across either plumes or roots of tubeworms (p > 0.1 for all inhibitors).

Discussion

Vestimentiferan tubeworms possess a variety of specific physiological adaptations to sustain their obligate symbiosis with sulfide-oxidizing autotrophic bacteria. They successfully support carbon fixation by their symbionts by obtaining sulfide, oxygen and inorganic carbon from their environments and supplying these nutrients to their symbionts via their circulating vascular system (Arp et al. 1985; Cavanaugh et al. 1981;

Childress et al. 1984; Freytag et al. 2001). In addition, they ensure that bacterial sulfide oxidation is not inhibited by a build-up of end products, namely, sulfate and hydrogen ions. The hydrothermal vent tubeworm R. pachyptila eliminates these waste ions across its plume surface (Girguis et al. 2002; Goffredi et al. 1999a). In contrast, we found that the hydrocarbon seep tubeworm L. luymesi primarily used its roots to excrete these waste

32 ions (Table 1-1). The main goal of this study was to identify the mechanism and location of sulfate and proton elimination. Due to experimental constraints, we did not simultaneously expose tubeworms to sulfide while measuring their sulfate and proton elimination rates. Thus the rates we report are likely lower than the rates characteristic of

L. luymesi in their natural habitat.

The plume is considered to be the primary exchange organ for hydrothermal vent vestimentiferans due to its large surface area, extensive vascularization, and short diffusion distances between blood vessels and the external environment (Arp et al. 1985;

Gardiner and Jones 1993). Sulfate and hydrogen ions are produced by symbionts located in the trophosome (Cavanaugh et al., 1981) and in the case of R. pachyptila are carried by the vascular blood to the plume, where they are likely eliminated across the plume epithelium (Goffredi et al., 1999). Unlike R. pachyptila, vascular connections between the trophosome and the body wall have been reported for L. luymesi (Gardiner and Jones

1993; Van Der Land and Nørrevang 1977) that might provide a direct route for transfer of ions between the two tissues in the seep species. Additionally, waste ions might be transferred to the coelomic fluid, which is in equilibrium with the vascular blood for most ions, and is contained within two cavities between the trophosome and the body wall

(Childress et al., 1984; Jones, 1981). The coelomic fluid might mediate transfer of ions from the trophosome to the body wall, where they might be eliminated across the body wall epithelium. In the case of L. luymesi, the root body wall might comprise a significant exchange surface. For example, a mature L. luymesi tubeworm that is about 1 m tall above its point of attachment to the carbonate substrate can have a root that is about the same length (1 m; Cordes et al., 2005a). The root can be approximated as a cylinder with

33 1.5 mm diameter, having a surface area of 47 cm2. The mass of a 1 m tall L. luymesi individual is about 6 grams, based on the mass to length conversion described by Cordes et al. (2003). The plume surface area of L. luymesi has not been measured, but using measurements made on a vestimentiferan of similar morphology, the long-skinny morphotype of Ridgeia piscesae, the plume surface area of a 6 g worm is approximately

53 cm2 (A.C. Anderson, J. F. Flores, S. Hourdez, manuscript submitted). Thus the surface areas of the root and plume are of the same order of magnitude in L. luymesi and both are likely important gas and ion exchange surfaces.

From an energetic standpoint, the root rather than the plume of L. luymesi might be the favorable exchange surface for sulfate and hydrogen ion elimination. Passive facilitated diffusion can mediate membrane transport of an ion in the direction of its electrochemical gradient, whereas energetically expensive active transport is required to transport an ion against its gradient (Byrne and Schultz 1994). In this study, we found that L. luymesi body fluids have an average sulfate concentration of 23 mmol l-1 and an average pH of 7.12. We did not measure intracellular pH and sulfate concentrations of plume or root epithelial cells. However, R. pachyptila intracellular pH was very similar to that of its body fluids (Goffredi et al. 1999a), and we assumed the same for L. luymesi. L. luymesi cells likely have a resting membrane potential of –70mV, which is typical for most animal cells (Lodish et al. 2000). Based on the Nernst equation (Hille 1992), the electrochemical gradient would favor proton efflux from L. luymesi only when external pH values were greater than 8.4. Similarly, sulfate efflux would be favorable if the extracellular sulfate concentration were lower than the intracellular concentration. In their natural habitat, L. luymesi plumes are bathed in seawater that has a sulfate concentration

34 of about 29 mmol l-1 and a pH of about 7.7, whereas their roots are surrounded by sediment pore-water in which sulfate and hydrogen ions are depleted due to microbial sulfate reduction (Aharon and Fu 2000; Arvidson et al. 2004). For example, at sediment depths greater than 20 cm at hydrocarbon-rich sites, pore-water sulfate concentrations can vary between 0 and 18 mmol l-1 and pH can vary between 7.9 and 9.0 (Aharon and Fu

2000). Therefore, L. luymesi could possibly eliminate sulfate and hydrogen ions using passive facilitated diffusion across its root epithelial membrane, while it would require energetically expensive ion pumps to eliminate these ions across its plume membrane.

Sulfate elimination from experimental animals

In order to measure sulfate elimination rates across plume and root surfaces of L. luymesi, we incubated them for 48 hours with a sulfate-free medium inside split-chamber respirometers (Freytag et al., 2001; Girguis et al., 2002). We performed our experiments with either “fed” tubeworms that had previously been exposed to sulfide for 48 hours, or

“starved” tubeworms that had previously been deprived of sulfide for at least 96 hours.

During the experiments, “fed” tubeworms likely eliminated sulfate derived from oxidation of sulfide carried in their blood. Conversely, “starved” tubeworms had minimal bound sulfide in their blood (Freytag et al., 2001) and were unlikely to eliminate sulfate derived from sulfide oxidation. Moreover, since we exposed these tubeworms to a sulfate-free external medium, a portion of the sulfate they eliminated during the experimental period was probably driven by diffusion not mediated by specific membrane transporters. Goffredi et al. (1999) found that the sulfate level in the coelomic fluid of R. pachyptila deprived of sulfide for 48-72 hours was just 5 mmol l-1 lower than

35 that of sulfide-exposed tubeworms. Based on this, we could assume that “fed” and

“starved” tubeworms faced similar gradients when exposed to SF_ASW, and eliminated similar levels of sulfate by unmediated diffusion. This allowed us to estimate the rate of sulfate elimination derived from sulfide oxidation alone, by subtracting sulfate elimination rates of “starved” tubeworms from those of “fed” tubeworms. In doing so, we found that 85% of the total sulfate derived from sulfide oxidation was eliminated across the root surface. There was a substantial and statistically significant difference (0.182

µmole h-1 g-1 wet weight) between the average sulfate elimination rates across roots of

“fed” and “starved” tubeworms (Table 1-1). Conversely, there was a small and statistically insignificant difference (0.033 µmole h-1 g-1 wet weight) between the average sulfate elimination rates across plumes of “fed” and “starved” animals.

In their natural habitat, tubeworms might occasionally experience sulfate-free conditions across their roots (Arvidson et al, 2004). However, their plumes are always exposed to seawater containing 29 mmol l-1. It appears that under our experimental conditions L. luymesi eliminated sulfate across their plumes primarily by unmediated diffusion, while they eliminated most of the sulfate derived from sulfide oxidation across their roots. The extent of plume sulfate elimination would depend on the gill surface area and the sulfate concentration gradient, whereas the extent of root sulfate elimination would depend on the amount of sulfide oxidation the tubeworm underwent during the experimental time period. Plume surface area may differ between individual tubeworms and is not likely to correlate with the rate of sulfide oxidation within the animal. This may explain the relatively weak correlation we found between plume and root sulfate elimination rates of individual tubeworms (Figure 2-3A).

The mechanism of sulfate release

We treated live L. luymesi with inhibitors of membrane anion transport in order to deduce the mechanism of sulfate elimination. The anion exchange inhibitors, DIDS and

SITS significantly inhibited root sulfate elimination (Figure 2-5 B&C), but had no significant effect on plume sulfate elimination. Both DIDS and SITS can bind reversibly, but are also known to have covalent binding capacities leading to irreversible effects

(Cabantchik and Greger 1992). We found that with respect to L. luymesi sulfate transport,

DIDS was the more potent of the two inhibitors and had an irreversible effect, whereas

SITS was less potent and appeared to bind reversibly (Figure 2-5 B& C). Probes of anion transport, such as DIDS and SITS often have broad specificities (Cabantchik and Greger

1992), but sensitivity to these inhibitors indicates the presence of an anion antiport system (Gerencser et al. 1996). Thus L. luymesi roots most likely contain sulfate exchangers through which they mediate the excretion of this ion, whereas their plumes do not appear to have this mechanism. Similar to our findings, Goffredi et al. (1999) did not find evidence of DIDS or SITS-sensitive sulfate exchangers in R. pachyptila plumes.

DIDS and SITS-sensitive sulfate transporters are found in a variety of taxonomic groups including invertebrates (Gerencser et al. 1996; Gerencser et al. 1999; Shimuzu and Bradley 1994), fish (Renfro 1999; Renfro and Pritchard 1983), and mammals

(Markovich 2001; Pritchard and Renfro 1983). DIDS and SITS inhibit the well-studied mammalian band-3 anion exchanger from red blood cell membranes that can mediate transport of chloride, bicarbonate and sulfate (Markovich 2001). In marine organisms, sulfate is commonly exchanged with chloride and bicarbonate that are abundant in

37 seawater (Gerencser et al. 1996; Gerencser et al. 1999; Renfro 1999). Thus, it is plausible that L. luymesi roots have an anion transporter that exchanges sulfate ions for either chloride or bicarbonate ions. To examine whether L. luymesi roots eliminate sulfate using sulfate-chloride antiports, we analyzed sensitivity of root sulfate transport to NPPB, a potent chloride transport inhibitor (Cabantchik and Greger 1992; Culliford et al. 2002;

Gelband et al. 1996). Interestingly, NPPB significantly affects chloride transport across the bacteriocyte membrane in the closely related tubeworm R. pachyptila (De Cian et al.

2003b). We found that NPPB did not have a significant effect on sulfate transport across

L. luymesi roots (Figure 2-5D), indicating that sulfate transport across the L. luymesi root membrane might not occur via a sulfate-chloride exchanger.

Alternately, sulfate elimination across L. luymesi roots might occur via sulfate- bicarbonate exchangers. Sulfate-bicarbonate antiports are found in several organisms including rats (Pritchard and Renfro 1983), teleosts (Renfro 1999) and lobsters

(Gerencser et al. 1999), and are often sensitive to DIDS and SITS. For L. luymesi, uptake of bicarbonate in lieu of sulfate elimination across its roots is reasonable in light of several facts. Bicarbonate levels in sediment pore-water surrounding tubeworms roots are high (Joye et al. 2004; Macdonald 1998). Bicarbonate is produced in the sediments as a byproduct of sulfate reduction coupled with hydrocarbon oxidation, the same process that produces sulfide (Sassen et al. 1994; Valentine 2002). Therefore, bicarbonate uptake by tubeworms from the sediment could enhance sulfide production due to end-product removal. Further, tubeworms could utilize the bicarbonate they take up across their roots for carbon fixation by their symbionts. R. pachyptila takes up inorganic carbon in the form of carbon dioxide by diffusion across its plume surface, facilitated by the high

38 partial pressures of CO2 in acidic vent waters (Childress et al. 1993; Goffredi et al.

1997b). On the other hand, L. luymesi plumes are bathed in seawater with pH of about 7.7

(Aharon and Fu 2000), at which pCO2 levels are negligible. L. luymesi might obtain at least part of its inorganic carbon from the sediment pore-water across its roots. This is consistent with the fact that L. luymesi tissues often have depleted stable carbon isotope values that reflect incorporation of inorganic carbon derived from oxidized methane and crude oil (Kennicutt Ii et al. 1992a; Roberts and Aharon 1994). Finally, carbonate encrustation of tubeworm root tubes could reduce their permeability to sulfate and sulfide. None of the several thousand L. luymesi that have been collected to date had carbonate deposited directly on the root tube surface, although root-balls of the aggregations are often partially embedded in carbonate (Figure 2-1; Cordes et al., 2005a).

It is plausible that L. luymesi limit carbonate precipitation directly on their root tubes by taking up bicarbonate and releasing protons across their roots, thereby decreasing pore- water bicarbonate concentrations and pH (Cordes et al. 2005a).

Proton elimination from experimental animals

We found that “fed” tubeworms eliminated protons across their roots on average three times faster than across their plumes (Table 1-1). The root proton elimination rates of individual tubeworms were strongly correlated with plume elimination rates (Figure 2-

4B), and the relation had a slope of approximately 2.5. This indicates that on an average, for every proton eliminated across the plume, 2.5 to 3 protons were eliminated across the root. Overall, approximately 67% of the total proton elimination occurred across the roots.

39 In our study, we found that “fed” and “starved” tubeworms eliminated protons at very similar rates (Table 1-1). Due to experimental constraints, we were unable to expose the tubeworms to sulfide during the measurement of proton flux. Thus our proton flux values were rather low, and in the same order of magnitude as “pre-sulfide” exposure rates of L. luymesi measured by Girguis et al. (2002). Girguis and co-workers found that proton elimination by R. pachyptila ceased just 1-2 hours after exposure to sulfide was terminated. Therefore, it is likely that in our experiments neither “fed” nor “starved” tubeworms eliminated a significant amount of protons derived from sulfide oxidation.

Instead, the proton flux we measured may have been derived from heterotrophic processes.

Moreover, we included a headspace of air in the anterior chamber to prevent hypoxic conditions during our experiments. However, post-hoc calculations based on heterotrophic oxygen consumption rates of L. luymesi (Freytag et al. 2001) indicated that

3 - 3.5 gram worms might have consumed all the oxygen available to them in the first 30

- 35 hours of the 48 hour experiment, and therefore might have produced protons as a result anaerobic metabolism. In Figures 2-3 and 2-4, we have indicated animals that might have experienced hypoxia using different symbols. We observed no apparent differences with respect to the proton elimination patterns of these animals compared to those that did not experience hypoxic conditions, and the trends we report here do not change if we omit these animals from the analyses.

40 The relation between proton and sulfate elimination

Sulfate transport is dependent on proton gradients in a number of different organisms (Gerencser et al. 1996; Yildiz et al. 1994). Sulfate transport could also be directly coupled with proton transport through proton-sulfate symports (Leustek and

Saito 1999; Renfro and Pritchard 1983). If this type of transporter were used by L. luymesi for sulfate elimination, we would expect to see a positive correlation between sulfate and proton elimination rates of individual tubeworms. In contrast, we observed a significant negative correlation between these rates (Figure 2-4). Moreover, inhibition of sulfate elimination by DIDS and SITS did not affect proton elimination across L. luymesi roots. This evidence combined suggests that proton and sulfate elimination might not be coupled in L. luymesi. The negative correlation between proton and sulfate elimination rates is difficult to explain in terms of known metabolic or membrane processes. Further studies that examine membrane transport in L. luymesi in more detail are needed before a viable explanation can be provided.

Figure 2-6: A schematic representation of an aggregation of L. luymesi, depicting a nutrient exchange model for this species. Seawater and sediment pore-water sulfate concentration and pH are from Aharon and Fu (2000). The equation for microbial sulfate reduction shows sulfate reduction coupled with methane oxidation. However, sulfate reduction coupled with higher molecular weight hydrocarbons can also occur (Joye et al., 2004). Bicarbonate uptake across roots is suggested, but has not been directly demonstrated in this study.

Conclusion

Based on our previous knowledge about L. luymesi and the results of this study, we propose a conceptual model for nutrient uptake and waste elimination processes in L. luymesi (Figure 2-6). In this study, we determined that body fluids of L. luymesi tubeworms have average sulfate concentrations of 23 mmol l-1 and an average pH of 7.12.

This indicates that electrochemical gradients favor elimination of both sulfate and protons by passive facilitated diffusion across the root surface into the surrounding sediment.

Therefore from an energetic perspective, it would be economical for L. luymesi to eliminate sulfate and hydrogen ions across its roots. Consistent with this, we found that

42 under laboratory conditions L. luymesi eliminate approximately 85% of the sulfate produced by sulfide oxidation and approximately 67% of the protons produced by a combination of metabolic processes, across their roots. Our results also suggest that sulfate transport across the root membrane might occur via an antiport that exchanges sulfate for bicarbonate ions. Elimination of sulfate and protons across the roots into the surrounding sediment pore-water might ensure that sulfide production in the sediment around the roots would not be sulfate limited (Cordes et al., 2005a). Moreover, if bicarbonate uptake across the roots does occur, it might supplement L. luymesi inorganic carbon uptake and help to prevent carbonate precipitation on its root tubes.

Acknowledgements

We thank the captain and crew of the Research Vessel (RV) Seward Johnson II as well as the crew and pilots of the Johnson Sea-Link submersibles (Harbor Branch Oceanographic

Institution). We would also like to thank Kathryn Luley for her invaluable assistance at sea, Guy Telesnicki for technical assistance, and Dr. Erik Cordes for helpful discussions.

The Minerals Management Service, Gulf of Mexico Regional OCS office, through contract number 1435-01-96-CT30813, the NOAA National Undersea Research Program at the University of North Carolina, Wilmington, the NOAA Ocean Exploration Program, and the National Science Foundation grant OCE 0117050 supported this work.

43 CHAPTER 3

Comparison of proton-specific ATPase activities in plume and root tissues of two co- occurring tubeworm species Lamellibrachia luymesi and Seepiophila jonesi

Sharmishtha Dattagupta1, Meredith Redding1, Kathryn Luley1, and Charles Fisher1

1 Department of Biology, The Pennsylvania State University, University Park, PA-16802.

44 Abstract

Lamellibrachia luymesi and Seepiophila jonesi are two co-occurring species of vestimentiferan tubeworms found at hydrocarbon seepage sites on the upper Louisiana slope of the Gulf of Mexico. Like all vestimentiferans, they rely on internal sulfide- oxidizing symbionts for nutrition. These symbionts produce hydrogen ions as a byproduct of sulfide oxidation, which the host tubeworm needs to eliminate to prevent acidosis. The hydrothermal vent tubeworm Riftia pachyptila uses a high activity of P- and V-type H+-

ATPases located on its plume epithelium to excrete protons. Unlike R. pachyptila, the seep species grow a posterior root, which they can use in addition to their plumes as a nutrient exchange surface. In this study we measured the ATPase activities of plume and root tissues collected from L. luymesi and S. jonesi, and used a combination of inhibitors to determine the abundances of P- and V-type H+-ATPases. We found that the total H+-

ATPase activity of their plumes were approximately 14 µmol h-1 g-1 wet weight, and that of their roots was between 5-7 µmol h-1 g-1 wet weight. These activities were more than

10 times lower than those measured in R. pachyptila. We suggest that seep tubeworms might use passive channels to eliminate protons across their roots, in addition to ATP- dependent proton pumps located in their plumes and roots. In addition, we found differences between the types of ATPase activities in the plumes of L. luymesi and S. jonesi. While the H+-ATPase activity of L. luymesi plumes is apparently dominated by P- type ATPases, S. jonesi appears to have an unusually high abundance of V-type H+-

ATPases. We suggest that S. jonesi is primarily dependent on its high V-type H+-ATPase activity to drive carbon dioxide uptake across its plume surface. L. luymesi, on the other hand, may rely at least in part on bicarbonate uptake across its root.

45 Introduction

All living organisms maintain ionic homeostasis using membrane transport processes. Transport ATPases, which couple the energy derived from ATP hydrolysis to drive transport of solutes against their electrochemical gradients, are a ubiquitous type of membrane transport protein (Pedersen 1982). They are generally of three different categories: P-, V- and F- type ATPases (Van Winkle 1999). P-type ATPases, named so because they are temporarily phosphorylated during their transport cycle, mediate transport of several different inorganic ions, including sodium, potassium, hydrogen and calcium ions (Møller et al. 1996). The plasma membrane Na+/K+- ATPase is an example of a P-type ATPase. F- and V- type ATPases have multiple domains, and nearly all of them transport protons (Nelson 1992). In multicellular organisms, V-type ATPases are usually found in endomembranes and plasma membranes, whereas F-type ATPases are typically located in inner mitochondrial or thylakoid membranes (Bowman et al. 1988;

Wieczorek et al. 2000).

Compared to other marine invertebrates, the hydrothermal vent tubeworm Riftia pachyptila has unusually high ATPase activity, with a large proportion of its ATPases being P-type and V-type ATPases devoted to proton transport (Goffredi and Childress

2001). R. pachyptila obtains sulfide, carbon dioxide and oxygen from its environment and supplies them to internal sulfide-oxidizing bacterial symbionts, which it relies upon for nutrition (Arp et al. 1985; Childress et al. 1984; Childress et al. 1991). The symbionts produce sulfate and hydrogen ions as end products of sulfide oxidation (Childress et al.

1984; Childress and Fisher 1992; Childress et al. 1991). R. pachyptila uses its highly vascularized plume for sulfide uptake, as well as sulfate and proton elimination (Arp et

46 al. 1985; Goffredi et al. 1998). Since electrochemical gradients are unfavorable for both sulfate and proton elimination across the plume into the surrounding seawater, R. pachyptila expends a substantial amount of energy for eliminating these ions, and uses the high concentration of proton-specific ATPases located on its plume for proton excretion (Girguis et al. 2002; Goffredi and Childress 2001; Goffredi et al. 1998).

Lamellibrachia luymesi and Seepiophila jonesi are close relatives of R. pachyptila, and almost always co-exist in aggregations at hydrocarbon seep sites in the

Gulf of Mexico (Bergquist et al. 2002). Unlike R. pachyptila, L. luymesi and S. jonesi grow root-like posterior extensions of their body, which they could potentially use as a metabolite-exchange surface in addition to their plume. To date, most physiological studies on hydrocarbon seep tubeworms from the Gulf of Mexico have focused on L. luymesi. These studies have established that L. luymesi uses its roots for sulfide uptake, as well as for elimination of sulfate and hydrogen ions (Dattagupta et al. 2006a; Freytag et al. 2001; Julian et al. 1999). Although knowledge about S. jonesi’s physiology is limited, evidence to date suggests that relative to L. luymesi, it might rely more on its plume than its root for metabolite-exchange. While adult L. luymesi plumes are situated over a meter above the sediment surface, where sulfide levels are lower than 0.1 µM, adult S. jonesi plumes are situated close to the sediment surface where sulfide levels can be in the 1-5

µM (Freytag et al. 2001). Since S. jonesi hemoglobins have substantially higher affinity for sulfide than L. luymesi hemoglobins, S. jonesi could potentially acquire a substantial proportion of its sulfide across its plume (Freytag et al. 2003). Moreover, the chitin tubes surrounding S. jonesi roots are significantly less permeable than those of L. luymesi (K.

47 E. Luley, unpublished data), suggesting that S. jonesi might not rely on its root as a significant metabolite-exchange surface.

In this study, we characterize the in vitro ATPase activities of plume and root tissues collected from L. luymesi and S. jonesi. Using inhibitors that target specific types of ATPases, we estimated the proton-specific ATPase activities of these tissues. In a sequential series of experiments, we used ouabagenin to inhibit Na+/ K+-ATPases, N- ethylmaleimide and vanadate to inhibit P-type ATPases, azide to inhibit F-type ATPases, amiloride to inhibit sodium-based ATPases, and bafilomycin to inhibit V-types ATPases

(Bowman et al. 1988; Goffredi and Childress 2001; Lin and Randall 1993). Based on the conjecture that S. jonesi relies more heavily on its plume as a metabolite exchange surface than L. luymesi, we predicted lower proton-specific ATPase activities in L. luymesi plumes than S. jonesi plumes. The internal body fluids of both L. luymesi and S. jonesi have average pH values of 7.1 (Dattagupta et al. 2006a). Their roots are buried in sediment where microbial sulfate reduction depletes protons, and calculations based on electrochemical gradients suggest that these tubeworms could use passive proton transport across their roots instead of ATPase mediated proton transport (Arvidson et al.

2004; Dattagupta et al. 2006a). Based on this, we predicted relatively low proton-specific

ATPase activities in L. luymesi and S. jonesi root tissues.

Materials and Methods

Collection of tissues for analysis

L. luymesi and S. jonesi were collected from the site GC234, located at a depth of approximately 540 m on the upper Louisiana slope of the Gulf of Mexico (27°44.7’N,

48 91°13.3’W)(Macdonald et al. 1990a), using the robotic claw of the Johnson Sea-Link submersible. The tubeworms were transported to the surface in a temperature-insulating box, and maintained alive in cold seawater until they were dissected on board the ship.

Tubeworms were separated by species and dissected within 2-3 hours after collection.

Plume and root body wall tissues were dissected from the tubeworms and kept on ice for up to 30 minutes until they were frozen using liquid nitrogen. They were transported in liquid nitrogen to the laboratory at Penn State University, where they were kept frozen at

-70˚C until further analysis.

Preparation of tissue homogenates

Homogenization and subsequent ATPase assays were performed using modifications of procedures described previously (Goffredi and Childress 2001; Lin and

Randall 1993). Frozen tissues were thawed on ice. Plumes were dissected further to separate plume lamellae tissue from obturaculum, and root body wall was carefully separated from trophosome, coelomic fluid and major blood vessels. The tissues were weighed and homogenized on ice in 30 µl/mg of tissue homogenization buffer to make a crude homogenate (hereafter designated ‘C’). Gill tissue was homogenized using a mechanical tissue homogenizer (Polytron PT 3000, Brinkmann Instruments, Inc.), whereas root body wall was homogenized using a Pyrex glass homogenizer. The homogenization buffer (pH adjusted to 7.3) contained 50 mM imidazole hydrochloride,

20 mM ethylenediamine tetraacetic acid (EDTA), 300 mM sucrose, 0.2 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM dithiothreitol (DTT) and 5 mM ß- mercaptoethanol (bME) (Goffredi and Childress 2001). The homogenates were

49 centrifuged at 2000 g for 7 minutes, to enrich mitochondria in the supernatant and pellet the cell membrane fraction (Lin and Randall 1993). The supernatant (hereafter designated

‘S1’) was stored at -70˚C until further analysis. The pellet was dissolved in 10X volume homogenization buffer containing 6% 3-[(3-Cholamidopropyl) dimethylammonio]-1- propanesulfonate (CHAPS) zwitterionic detergent, in order to solubilize membrane proteins. It was then centrifuged at 2000 g for another 7 minutes, to enrich cell membrane proteins in the supernatant, and pellet cell debris (Lin and Randall 1993). The supernatant

(S2) was divided into 100 µl aliquots and stored at -70˚C until it was used for ATPase assays.

Determination of mitochondrial content: succinate dehydrogenase activity

The activity of the mitochondrial marker enzyme succinate dehydrogenase (SDH) was determined in homogenate fractions C, S1 and S2 as previously described (Munujos et al. 1993). This assay is based on the conversion of 2-(4-iodophenyl)-3-(4-nitrophenyl)-

5-phenyltetrazolium chloride (INT), a colorless compound, to formazan, a blue compound that absorbs at 500 nm. To perform the assay, 10 µl of the homogenate was mixed with 990 µl of the assay buffer (pH adjusted to 8.3) that contained 100 mM Tris hydrochloride, 0.5 mM EDTA, 2 mM potassium cyanide, 2 mM INT, 20 mM succinate

(pH adjusted to 7.4 before it was added to the assay medium), and 12-g/l cremophor EL.

Absorbance at 500 nm was measured every 30 seconds for 6 minutes, and the mitochondrial activity was determined by subtracting absorbance of a control containing all assay components except for succinate.

50 Determination of plasma membrane content in fractions

The activity of the plasma membrane marker enzyme Na+-K+ ATPase was determined in fractions C, S1 and S2. Na+-K+ ATPase activity of fractions is usually estimated by determining ouabain-sensitive ATPase activity (Furriel et al. 2001).

However, Na+-K+ ATPases of vestimentiferan tubeworms appear to be insensitive to oubain, but sensitive to its analogue, ouabagenin (Goffredi and Childress 2001).

Therefore, we estimated Na+-K+ ATPase activity of fractions by determining ouabagenin- sensitive ATPase activity using the methodology described below.

Measurement of ATPase activity

ATPase activity was determined using a spectrophotometric assay using a coupled pyruvate kinase/lactate dehydrogenase enzyme reaction (Goffredi and Childress 2001;

Lin and Randall 1993). The assay medium (pH 7.3) was made fresh on the day the assay was performed, and contained 50 µM fructose diphosphate, 2 mM phosphoenol pyruvate

(PEP), 3 mM adenosine-5’-triphosphate (ATP), 0.2 mM reduced ß-nicotinamide adenine dinucleotide (NADH), 12.7 units/µl lactate dehydrogenase, and 15 units/µl pyruvate kinase, made in a stock solution containing 100 mM sodium chloride, 20 mM potassium chloride, 5 mM magnesium chloride, 30 mM imidazole, and 0.5 mM ethylene glycol bis-

(2-aminoethyl ether)-N,N,N'N'-tetraacetic acid (EGTA). 100 µl of the tissue homogenate was mixed with 900 µl of assay buffer and readings were taken at 340 nm every 30 seconds over a period of 5-10 minutes using a Beckman DU-64 spectrophotometer

(Beckman Coulter Inc., Fullerton, CA, USA), equipped with a temperature controlled cuvette (15˚C). Protein content of the homogenates were determined using the Bradford

51 method, using a 1 mg/ml bovine serum albumin (BSA) standard made using homogenization buffer. ATPase activity was expressed in either protein-specific

-1 -1 inorganic phosphate values (µmol Pi h mg protein) or tissue wet-weight specific values

-1 -1 (µmol Pi h g wet weight).

Inhibitor exposure

Tissue homogenates were incubated with various ATPase-specific inhibitors before the assay was conducted. 50 µl of a stock solution of an inhibitor (or solvent, in the case of controls) was mixed with 50 µl of the tissue homogenate. The mixture was incubated at room temperature for a period of two hours, with the exception of

Bafilomycin, which was incubated for 30 minutes (Goffredi and Childress 2001). The various inhibitor stock solutions used were, 20 mM sodium azide, 40 mM ouabagenin, 20 mM NEM, 20 mM vanadate, and 7 µM bafolimycin. Bafilomycin stock solution was made using 5% DMSO as solvent, and corresponding controls contained the appropriate amount of DMSO. All other inhibitor solutions were made using water. Azide, vanadate and bafolimycin stock solutions were made in advance and the first two were stored at room temperature, whereas bafilomycin was stored at -20 ˚C. NEM stock solution was made fresh on the day of the experiment. The vanadate stock solution was monitored at

400 nm to ensure that decameric vanadium species were negligible or absent in the solution (Aureliano and Gândara 2005).

52 Estimation of proton-specific P- and V-type ATPase activities

We used inhibitor sensitivities to estimate proton-specific ATPase activities as described below. At the concentrations used in our assays, NEM inhibits both P- and V- type ATPases (Forgac 1989; Pedersen and Carafoli 1987). Since vanadate sensitivity was found to be low in most tissues examined, we used NEM-sensitivity to estimate the total

P- and V-type ATPase activity. We used bafilomycin-sensitivity to estimate V-type

ATPase activity, as it is extremely specific to V-type ATPases at concentrations we used in our experiment (Bowman et al. 1988). Since we used EGTA, a Ca2+-chelator in our assays, we assumed that Ca2+-ATPase activity was negligible (Goffredi and Childress

2001; Lin and Randall 1993). We estimated total P-type ATPase activity, which includes

Na+/K+-ATPases, and H+-ATPases, by subtracting bafilomycin-sensitive activity from

NEM-sensitive activity. Finally, we estimated P-type H+-ATPase activity by subtracting ouabagenin-sensitive Na+/K+-ATPase activity from the total P-type ATPase activity.

Results

Mitochondrial activity of fractions

SDH activity, which was used to estimate mitochondrial content, was found to be the highest in the crude homogenates (C) of all tissues examined (Figure 3-1). The fraction S1 was found to contain between 26-77% of the total SDH activity, indicating substantial removal of mitochondria using the first centrifugation step. However, the fraction S2, which was subsequently used to analyze total ATPase activity, still showed some SDH activity. The S2 fractions of L. luymesi and S. jonesi plume tissues contained, respectively, 25% and 22 % of the mitochondrial activity found in their respective crude

53 homogenates. On the other hand, the S2 fractions of L. luymesi and S. jonesi root tissues contained, respectively, 0% and 5% of the total mitochondrial activity. Therefore, mitochondrial removal was more efficient in the root than the plume tissue homogenates.

Figure 3-1: Mean ± standard deviation of succinate dehydrogenase (SDH) activity in the crude, S1 and S2 fractions of the various tissues analyzed (N=5 for all tissues). L and S correspond to L. luymesi and S. jonesi, respectively.

Na+/K+-ATPase activity of fractions

Ouabagenin sensitive activity of fractions, which was used to estimate activity of the plasma membrane marker enzyme Na+/K+-ATPase, indicated that the purification process was successful in enriching plasma membrane in the S2 fraction (Figure 3-2).

Using the mean values of the Na+/K+-ATPase activities, we estimated that the enrichment of plasma membrane in the S2 fraction was 288% and 168%, respectively, for L. luymesi plume and root tissues. Similarly, the enrichment was 159% and 125%, respectively, for

S. jonesi plume and root tissues. Overall the purification procedure appeared to be more efficient for L. luymesi than S. jonesi tissues.

Figure 3-2: Mean ± standard deviation of Na+/K+-ATPase activity in the crude, S1 and S2 fractions of the various tissues analyzed (N=4 for plume tissues; N=3 for root tissues). L and S correspond to L. luymesi and S. jonesi, respectively.

Total ATPase activities

-1 -1 Total ATPase activity in µmol Pi h mg protein was similar in all tissues examined (Table 3-1; P>0.05; two-sided t-test). However, both species had significantly higher protein content, i.e. mg protein per g tissue wet weight (WW), in plume than in root body wall tissues (P<0.0025; two-sided t-tests). As a result, wet-weight specific

-1 -1 plume ATPase activities (µmol Pi h g WW) were significantly higher than root

-1 ATPase activities in both species. S. jonesi plume and root ATPase activities (µmol Pi h g-1 WW) were higher than L. luymesi plume and root ATPase activities, but the difference was not significant (P> 0.05; two-sided t-tests).

Table 3-1: The total ATPase activity of various tissues examined in this study. Plume ATPase activity Root ATPase activity

-1 -1 -1 -1 -1 -1 -1 -1 Species µmol Pi h mg µmol Pi h g µmol Pi h mg µmol Pi h g protein WW protein WW

L. luymesi (N=15) 2.8 ± 1.3 34.9 ± 14.6 2.6 ± 1.4 13.1±6.5 S. jonesi (N=12) 2.0 ± 0.6 41.4 ± 13.4 2.5 ± 0.7 16.3±7.6

Inhibitors

Inhibitor studies were performed with L. luymesi and S. jonesi plume and root tissues, and comparisons were made between each tissue type (Figure 3-3). For L. luymesi, percent inhibition by all inhibitors was higher for plume than for root tissues, and this difference was significant (P<0.05; two-sided t-test) for all inhibitors except

NEM. Similarly, percent inhibition by azide, ouabagenin, and bafilomycin were significantly higher for S. jonesi plume than for root tissues. The higher inhibition by azide in plume tissues of both species was consistent with the higher mitochondrial content found in their S2 fractions (Figure 3-1). L. luymesi and S. jonesi root tissues were very similar with respect to inhibition by all compounds tested, whereas there were some notable differences between their plume tissues. While azide and vanadate inhibition were significantly higher for L. luymesi plume tissue, bafilomycin inhibition was significantly higher for S. jonesi plume tissue. Except for L. luymesi plume tissue, all other tissues appeared to be insensitive to vanadate inhibition.

Figure 3-3: Comparison of percent inhibition of total ATPase activity by various inhibitors in (a) L. luymesi plume and root tissues; (b) L. luymesi and S. jonesi plume tissues; (c) S. jonesi plume and root tissues; and (d) L. luymesi and S. jonesi root tissues. An asterisk implies a significant difference (P<0.05; two-sided t-test). AZ=azide, OU=ouabagenin, NEM= N-ethylmaleimide, VAN=vanadate, BAF=bafilomycin. Percent inhibition is expressed with respect to baseline activity measured in the presence of the solvent used to make inhibitor solutions.

57 -1 -1 Table 3-2: Estimated P- and V-type ATPase activities (in µmol Pi h g tissue wet weight) of L. luymesi plume (LP), L. luymesi root (LR), S. jonesi plume (SP) and S. jonesi root (SR). Proton-specific ATPase activities are in bold.

ATPase activity -1 -1 ATPase type Inhibitor sensitivity (µmol Pi h g WW) LP LR SP SR Total None 35.0 13.1 41.4 16.3 Total P-and V-type ATPase NEM 18.2 5.8 19.0 7.5 V-type H+ ATPase Bafilomycin 3.7 0.6 9.8 0.9 Total P-type ATPase NEM- (Bafilomycin) 14.5 5.2 9.3 6.6 Na+/K+-ATPase Ouabagenin 4.1 0.7 4.3 0.4 P-type H+-ATPase NEM- (Bafilomycin + Ouabagenin) 10.5 4.5 4.9 6.2

Using different inhibitors, we were able to determine the V-type H+-ATPase and the P- type H+-ATPase activities for the root and plume tissues of both species (Table 3-2).

Although the total ATPase activity is very similar for a given tissue between species, the contribution of V-type and P-type is very different in the plumes. Most of the activity in

S. jonesi plume is due to V-type H+-ATPase while most of the activity in L. luymesi plume is due to P-type H+-ATPase.

Discussion

Seep tubeworms have much lower ATPase activities than Riftia pachyptila

Riftia pachyptila, the vestimentiferan tubeworm found at hydrothermal vent sites on the East Pacific Rise, has protein-specific ATPase activities that are approximately four times higher than that of other marine invertebrates that do not harbor symbionts

(Goffredi and Childress 2001). Using an inhibitor study similar to ours, the authors concluded that a majority of R. pachyptila’s ATPase activity was dedicated towards

58 elimination of protons produced by its symbionts during sulfide oxidation. In contrast, we found that the protein-specific ATPase activities of the two hydrocarbon seep tubeworm species, L. luymesi and S. jonesi, were similar to that of non-symbiotic, shallow-living marine invertebrates investigated by Goffredi and Childress (2001).

Like R. pachyptila, L. luymesi and S. jonesi have sulfide-oxidizing symbionts that produce protons as a byproduct of their metabolism. However, R. pachyptila is adapted to an ephemeral vent environment and has a very high growth rate (Lutz et al. 1994), whereas L. luymesi and S. jonesi are adapted to the relatively stable cold-seep environment, and adults of these species have low growth rates (Bergquist et al. 2002;

Bergquist et al. 2000; Fisher et al. 1997). Thus seep tubeworms might have lower metabolic rates than R. pachyptila, consistent with the lower ATPase activities we measured. Alternately, seep tubeworms might rely on passive transport of waste ions such as sulfate and protons across their root surface (Dattagupta et al. 2006a), diminishing their need for high tissue ATPase activities, as discussed below.

Lamellibrachia luymesi might use proton channels in its roots for proton elimination

An important distinction between R. pachyptila and the seep tubeworms we examined is the presence of roots in the latter species. Both L. luymesi and S. jonesi have long posterior root-like structures, but so far, the use of roots for metabolite exchange has been better studied in L. luymesi than in S. jonesi. Based on electrochemical gradients, proton movement across L. luymesi’s plume epithelium would require active transport, whereas it could occur passively across the root epithelium (Dattagupta et al. 2006a). In this study, we found that L. luymesi plume tissue has a total H+-ATPase activity of 14

59 -1 -1 -1 -1 µmol Pi h g wet weight, whereas its root tissue has an activity of only 5 µmol Pi h g wet weight (Table 3-2). Given the fact that L. luymesi eliminates a majority of its protons across its roots (Dattagupta et al. 2006a), it appears that proton channels, rather than H+-

ATPases might mediate proton transport across L. luymesi’s root epithelium.

In laboratory experiments where L. luymesi and R. pachyptila were exposed to similar levels of external sulfide, L. luymesi’s plume proton elimination rates were on average 11.9 µequivalents g-1h-1, compared to R. pachyptila’s elimination rate of 40.5

µequivalents g-1h-1 (Girguis et al. 2002). However, a different study showed that L. luymesi eliminates approximately 67% of its protons across its roots (Dattagupta et al.

2006a). In combination, the results of these studies indicate that the total proton elimination rate of L. luymesi is approximately 36 µequivalents g-1h-1, slightly lower than the mass specific rate of R. pachyptila. However, the mass-specific H+-ATPase activity of L. luymesi is more than ten times lower than that of R. pachyptila. R. pachyptila has a

+ -1 -1 + plume P-type H -ATPase activity of 136-174 µmol Pi h g wet weight and a V-type H -

-1 -1 ATPase activity of 80 µmol Pi h g wet weight (Goffredi and Childress 2001). On the other hand, L. luymesi has a total (plume and root) P-type H+-ATPase activity of 15 µmol

-1 -1 + -1 -1 Pi h g wet weight and a V-type H -ATPase activity of 4 µmol Pi h g wet weight

(Table 3-2). The large discrepancy between these values might be consistent with L. luymesi relying partly on passive proton channels across its root epithelium for proton elimination, in addition to H+-ATPases in its plume and root tissues.

The pH of S. jonesi body fluids is similar to that of L. luymesi (S. Dattagupta, unpublished data), and the two seep species commonly co-exist in aggregations

60 (Bergquist et al. 2002). Since S. jonesi also has roots buried in sediments where pore water pH is depleted, it could potentially eliminate protons using passive transport across

+ -1 its roots. In this study, we found that its plume H -ATPase activity was 14.7 µmol Pi h

-1 + -1 -1 g wet weight, whereas its root H -ATPase activity was 7.1 µmol Pi h g wet weight.

While it appears likely that S. jonesi use their roots for passive proton transport, laboratory experiments that directly measure its proton elimination rates are necessary to demonstrate this.

The relative abundance of P- and V-type ATPases varies between seep species

Based on differences in growth habits of L. luymesi and S. jonesi, we predicted that S. jonesi would rely more on its plume for nutrient exchange than L. luymesi does.

Contrary to our expectations, we found very similar total plume H+-ATPase activities in the two tubeworm species. However, there were some interesting differences in the relative abundances of P- and V-type H+-ATPases in these tissues. While 74% of the H+-

ATPases in L. luymesi plume tissue were P-type ATPases, 66% of the H+-ATPases in S. jonesi plume tissue were V-type ATPases (Table 3-2).

V-type H+-ATPases are important for creating proton gradients that drive secondary membrane transport across the plasma membranes of some invertebrates

(Azuma and Ohta 1998; Wieczorek et al. 2000). In R. pachyptila, V-type ATPases are co- localized with the enzyme carbonic anhydrase on the apical region of the plume epithelium, and are thought to generate acidic extracellular pH conditions that facilitate

+ CO2 uptake (De Cian et al. 2003c). V-type H -ATPases represent on average 24% of the

61 total ATPase activity in S. jonesi plume tissue (Figure 3-3), whereas they represent only

14% of the total ATPase activity in R. pachyptila plumes (Goffredi and Childress 2001).

The higher abundance of V-type H+-ATPases in the S. jonesi plume tissue might indicate

+ that this tubeworm relies heavily on its plume for CO2 uptake. In contrast, V-type H -

ATPases represent only 10% of the total ATPase activity in L. luymesi plume tissue

(Figure 3-3). A previous study suggested that L. luymesi might supplement plume CO2

- uptake with HCO3 uptake across their roots (Dattagupta et al. 2006a). The results of this study are consistent with this possibility.

Possible caveats of this study and future directions

In this study, we used an in vitro approach in which we exposed tissue homogenates to various inhibitors to deduce membrane transport mechanisms. While inhibitors are powerful tools for arriving at mechanistic deductions, lack of inhibitor sensitivity does not indicate the absence of a target protein (Cabantchik and Greger

1992). Moreover, differences in purification efficacy of membrane fractions could lead to artifacts. In our study, we found plasma membrane purification efficiencies were slightly higher for L. luymesi tissues than S. jonesi tissues, which could lead us to underestimate

S. jonesi ATPase activities. Our study provides the first clues to the membrane transport mechanisms used by seep tubeworms for proton elimination. Future studies using purified membrane vesicles, or tissue localization with specific antibodies (De Cian et al.

2003c) could provide more conclusive evidence for the existence of specific membrane proteins such as proton channels in root tissues of seep tubeworms.

62 Acknowledgements

We thank the captain and crew of the Research Vessel (RV) Seward Johnson II as well as the crew and pilots of the Johnson Sea-Link submersibles (Harbor Branch

Oceanographic Institution). The Minerals Management Service, Gulf of Mexico Regional

OCS office, through contract number 1435-01-96-CT30813, the NOAA National

Undersea Research Program at the University of North Carolina, Wilmington, the NOAA

Ocean Exploration Program, and the National Science Foundation grant OCE 0117050 supported this work.

63 CHAPTER 4

Submersible operated dialysis samplers for collecting pore water from deep-sea sediments

Sharmishtha Dattagupta1∗, Guy Telesnicki1, Kathryn Luley1, Benjamin Predmore1†, Michael McGinley1†† and Charles Fisher1

Submitted to Limnology and Oceanography: Methods (date submitted 6/9/2006)

1 Department of Biology, The Pennsylvania State University, University Park, PA-16801 ∗ Corresponding author email: [email protected] † Current address: Department of Zoology, University of Florida, Gainesville, FL-32611 †† Current address: Department of Marine Biology and Biochemistry, University of Delaware Graduate College of Marine Studies, Lewes, DE-19958

64 Abstract

Dialysis samplers, or peepers, have proved to be a viable alternative to cores for collecting pore water from shallow systems such as lakes, streams and wetlands, but have rarely been used to sample deep-sea sediments. Here we describe a newly developed peeper, that was specifically designed to be used by an ROV or manned submersible to obtain samples for poor water profiles down to 60 cm sediment depth from precisely located positions on the sea floor. The peepers we developed can be deployed, opened and closed using a single robotic arm of a submersible. They can be used in some environments where coring is not practical or possible. They can be sealed closed at the end of a deployment, but before removal from the sediment, to maintain sample integrity during recovery of the submersible and until processing of the samples. The same feature allows de-oxygenation of the peeper dialysis cells prior to deployment and ensures that peeper cells remain anoxic until they are opened in situ, after deployment. Each peeper collects samples from 10 cm above the sediment water interface and every 10 cm down to 60 cm depth in the sediment, and provides sufficient volume for multiple chemical analyses, including analysis of stable isotope compositions from every sample.

65 Introduction

Hydrocarbon seepage across the seafloor is widespread at active and passive margins worldwide, and sustains chemosynthetic communities dominated by vesicomyid clams, mytilid mussels and vestimentiferan tubeworms (Sibuet and Olu 1998). These cold seep macro-faunal communities are often fueled by high concentrations of hydrogen sulfide produced in the sediment by microbial associations that couple anaerobic hydrocarbon oxidation with sulfate reduction (Boetius and Suess 2004; Joye et al. 2004;

Levin 2005). Thus collection and analysis of pore water from these sediments is not only useful for geochemical studies, but is also crucial for understanding the biology and the distribution of macro-fauna at these sites (Arvidson et al. 2004; Levin 2005; Treude et al.

2003). A majority of cold seep sites are located at depths greater than 400 m (Sibuet and

Olu 1998), and require submersibles for detailed studies. Sampling sediment pore water in such inaccessible systems is plagued by various challenges. It is complex to manipulate sampling devices using submersibles. Also, it is difficult to maintain integrity of samples between the time they are collected at the seafloor and processed on board a ship.

Deep-sea sediments are commonly collected using cores. Although coring is an effective method for obtaining pore water and for analysis of microbial populations in the sediment, it suffers from some drawbacks. Cores collected using submersibles are usually limited to 30 cm or less in depth, and they do not penetrate sediments that contain carbonate rubble, mussel shells or tubeworm roots, which are all common in hydrocarbon seep systems. Moreover, post-sampling artifacts are a common problem with cores.

Sediments collected using cores need to be centrifuged or squeezed under pressure to

66 obtain pore water samples. Both methods can cause changes in pore water chemistry, by disrupting microbial cells (Bollinger et al. 1992) or by changing the natural sediment structure (Holcombe et al. 2001). Also, decompression or warming of cores during retrieval can cause bacterial cells to disrupt, and change concentrations of some pore water constituents (Aller et al. 1998). The use of microelectrodes with cores reduces post- sampling artifacts, and provides concentration profiles with excellent resolution.

However, they can only be used to detect a few biogeochemically important ions (Krom et al. 1994; Urban et al. 1997), and cannot be used for techniques such as stable isotope analyses.

Dialysis samplers, or “peepers” are an excellent alternative or supplement to cores for collecting pore water samples. Peepers were introduced by Hesslein (1976) and are based on equilibration of fluid contained within a cell with the surrounding medium across a dialysis membrane. A major advantage of this method is the minimal post-collection processing it requires, as the samples are already particle-free (Bollinger et al. 1992;

Hesslein 1976). Peepers have been used extensively in a variety of aquatic systems such as lakes (Bollinger et al. 1992; Carignan et al. 1994), wetlands (Laforce et al. 2000;

Novak et al. 2004), and salt marshes (Howes and Wakeham 1985; Weston et al. 2006).

However, they have rarely been used for sampling deep-sea sediments (Aller et al. 1998;

Hashimoto et al. 1995; Holcombe et al. 2001). A conundrum with using peepers to sample deep sediment pore water is that samples may back-equilibrate with ambient seawater in the time delay that invariably occurs between collection and processing.

Hashimoto et al. (1995) assumed a 10 % loss in hydrogen sulfide from peeper cells in the

67 1.5 hours it took them to recover the sampler. Aller et al. (1998) constructed a scabbard to house the peepers after collection.

In this study, we developed inexpensive peepers specially designed to sample inaccessible systems such as the deep-sea. These peepers can collect pore water samples from 10 cm intervals, up to a sediment depth of 60 cm. They are designed to toggle between two modes; in one mode, the peeper cells are open to the environment for sampling, and in the other mode, the cells are sealed closed to avoid any exchange with the environment. This switching is easy to manipulate using a single robotic arm of a submersible. To avoid artifacts related to introducing oxygen into anoxic sediments

(Carignan 1984; Carignan et al. 1994), we designed chambers in which entire peepers can be deoxygenated, and the cells can be sealed closed before they are deployed in situ. The peeper cells can be opened in situ, and left open for the period of equilibration. Before collecting the peepers from the seafloor, the cells can be sealed closed again, thereby maintaining the integrity of samples during recovery of the submersible and until they are processed on board a ship.

Materials and procedures

Overall design and construction of peepers

Peepers were designed to be easily assembled from and disassembled into their component parts: an inner tube of axially stacked chambers, an outer tube with slots, an actuating handle assembly, a circular plate, and a conical substrate penetrating tip (Figure

4-1). The chambers of the inner tube were designed to house the dialysis bags, and were constructed out of 10 cm long sections of half-inch nominal diameter chlorinated

68 polyvinylchloride (CPVC) hollow pipe, with 1 cm by 4 cm windows milled into them.

Consecutive chambers were connected using solid PVC plugs that served to prevent fluid flow between the chambers, and were secured in place with PVC cement and stainless steel split pins. Seven chambers were stacked axially so that the centers of their windows were positioned exactly 10 cm apart from each other, except the first and second windows that were 20 cm apart. Custom-made butyl rubber radial seals (Iris Rubber

Company, Cicero, Indiana), with ribs at the top and bottom, were cemented on either side of each inner tube window (Figure 4-1e). When fitted over the inner tube, the seals had an outer diameter (OD) of 24.2 mm and the ribs had an OD of 24.8 mm, which exactly matched the inner diameter (ID) of the peeper outer tube. The ribs functioned as o-rings to form a good seal against the inner wall of the outer tube.

The outer tube was a 90 cm long hollow tube with 1 cm by 2.5 cm slots milled into its side that were positioned to align with the inner tube windows. The outer tube was constructed of G-10 Garolite, chosen due to its high tensile strength and stringent ID tolerance (±0.008 cm). The handle assembly was threaded into the top of the outer tube, and a solid PVC cone was fastened to the bottom end using a removable stainless steel split pin. A circular PVC plate with a hole in its center was secured circumferentially on the outer tube by sandwiching it between two shaft collars. The plate was positioned exactly halfway between the first and second slots of the outer tube. During sampling, the plate was positioned at the sediment-seawater interface (Figure 4-2a) to ensure that the first slot was exactly 10 cm above the interface, whereas the remaining slots were at 10 cm depth increment below the interface. The plate also served to prevent channeling of overlying seawater down the length of the peeper.

Figure 4-1: (a) Peeper dismantled to show its component parts, (b) Peeper in “open” position, (c) Peeper in “closed” position, (d) Schematic view of peeper handle assembly and (e) Schematic view of inner tube showing actuating handle and details of radial seals.

The handle assembly (Figure 4-1d) comprised of three parts: (i) a solid PVC shaft attached to the inner tube assembly, (ii) a two-part actuating aluminum handle threaded through a hole in the inner shaft and the slots in the outer tube, and (iii) an outer aluminum handle with slots on the side and two welded coplanar handles. The actuating handle could be moved to shift the position of the inner tube with respect to the outer tube. When the actuating handle was squeezed together with coplanar handle 1, the windows in the inner tube were aligned exactly with the slots on the outer tube. In this position (the “open” position; Figure 4-1b), dialysis bags placed inside the inner tube chambers would be open to the environment. Conversely, when the actuating handle was

70 squeezed together with coplanar handle 2, it aligned the radial seals with the slots in the outer tube. In this position (the “closed” position; Figure 1-1c), the dialysis bags were sealed closed. The handle assembly could be disassembled quickly to facilitate removal of samples from the inner tube chambers for processing.

Figure 4-2: (a) Peepers deployed in situ with plates level with the sediment-water interface, in sediment underlying a Beggiatoa bacterial mat. Handle assembly is in the “open” position. The arrow indicates the peeper used to collect data shown in Figure 4-3. (b) Peeper handle being “closed” by the robotic arm of the Johnson Sea-link submersible.

Preparation of dialysis bags

Cellulose based dialysis membranes can decompose and cause artifacts in samples collected using peepers (Carignan 1984). Therefore, we used non-cellulose based polyvinylidene difluoride (PVDF) membrane (500 kilo-Dalton molecular weight cut off,

16 mm wide tubes, Spectrum Laboratories, Inc.) that is heat-sealable. This membrane material is also resistant to degradation by aromatic and non-aromatic hydrocarbons that might be present in hydrocarbon seep sediments (Joye et al. 2004).

Peeper compartments filled with water less dense than the surrounding pore water can cause shifts in concentration profiles (Grigg 1999). Keeping this in consideration, and

71 because our purpose was to obtain pore water sulfate and sulfide, we filled our peeper cells with sulfate-free artificial seawater (ASW; salinity = 36 ‰, pH = 8.0) containing

500 mmol L-1 sodium chloride, 9 mmol L-1 potassium chloride, 9.3 mmol L-1 calcium chloride, 48.5 mmol L-1 magnesium chloride, and 2.5 mmol L-1 sodium bicarbonate. To prepare the dialysis bags, we cut 6 cm sections of the PVDF membrane tube and soaked it in distilled de-ionized water (DDI water) for one hour. We flushed the insides of the tubes several times with DDI water and heat-sealed one end. We filled the bags with 4 ml of ASW, carefully removed air bubbles, and then sealed the other end to form a cylindrical bag. We stored the bags in cold (4°C) ASW containing 0.05% sodium azide to prevent bacterial growth. Before use, we rinsed the bags with azide-free ASW three times. One bag was placed inside each chamber of the inner tube of a peeper, and covered with a 1.5 cm by 5 cm piece of nylon mesh to protect the dialysis membrane from damage. The mesh was held in place using rubber bands. The peeper was then assembled and placed inside a de-oxygenation chamber.

Figure 4-3: Schematic representation of the set-up used to de-oxygenate peepers before deployment. Peepers were fully assembled before insertion into the chambers, and six peepers could be de-oxygenated at a time. The peeper plates were moved to the very top of the peepers, so that all peeper windows could fit into the acrylic chambers. Before deployment, the peepers were closed and the plates were moved back to the position indicated in Figure 1.

De-oxygenation procedure

Peepers were de-oxygenated inside specially built chambers comprising of 1.5 m long, 4 cm ID acrylic cylinders with PVC end caps (Figure 4-3). A gas inlet was threaded into a hole tapped into the PVC cap to allow introduction of a constant stream of nitrogen gas. Dialysis bags were loaded into peeper chambers and the peepers were fully assembled, with slots in the “open” position. Six de-oxygenation chambers were mounted upright onto a rack and one peeper was loaded tip first into each chamber. The PVC plates on the peepers were moved up above the top windows and served to loosely cap

73 the top of each chamber. The chambers were filled with ASW and nitrogen gas was continuously bubbled through each chamber for a period of approximately six hours.

After two hours of bubbling with nitrogen gas, the oxygen level in the ASW inside each de-oxygenation chamber was determined to be zero using a portable oxygen meter.

Before deployment, the peepers were sealed closed while inside the de-oxygenation chambers and the plates moved back to their position between windows 1 and 2. The peepers were then loaded onto the submersible about 10 minutes before commencing the dive.

Deployment and recovery of peepers

12 peepers were deployed and recovered in the summer of 2004. On a single dive, six peepers were placed inside individual PVC quivers mounted onto the Johnson Sea- link submersible and taken to the sampling location, site GC234 located at a depth of

~540 m on the upper Louisiana slope of the Gulf of Mexico (27°44.7’N, 91°13.3’W).

This site covers an area of several square kilometers, experiences active hydrocarbon seepage and is dominated by large tubeworm aggregations (Bergquist et al. 2003;

Macdonald et al. 1990a). Once at the sampling location, peepers were inserted tip first in a vertical orientation into the sediment until the PVC plates of the peepers were level with the sediment-seawater interface (Figure 4-2a). The robotic claw of the submersible was then used to manipulate the actuating handles of the peepers to “open” them. The peepers were allowed to equilibrate for 5 weeks. At the end of this period, the peepers were “closed” (Figure 4-2b), collected and brought to the surface where they were immediately processed on board the ship.

74 Post collection sampling and analyses

Once on board the ship, peepers were dismantled one at a time and sampled in coordinated manner so each peeper was processed completely within 2 minutes after it was opened. The water samples were used for analysis of pH and salinity, sulfate, sulfide, and methane concentrations, and sulfur stable isotope content. After the peepers were dismantled and the inner tube extracted, the mesh covering from each window was removed. Disposable plastic syringes were used to draw 300 µl samples from all windows for pH and salinity measurements, and were emptied into sealed plastic vials where they were stored at room temperature until analysis. Gas-tight syringes were used to draw 700 µl samples for methane analyses from chambers corresponding to depths 10,

30 and 50 cm. These samples were plunged into 1.5 ml glass vials sealed with rubber stoppers containing 200 µl concentrated phosphoric acid and flushed with nitrogen gas.

They were stored at -20°C until analysis. After syringe samples were withdrawn from the dialysis bags, they were removed from the inner tube chambers. A small slit was made in the bags, and the contents were emptied into plastic screw-top vials containing 1.5 ml of a zinc acetate fixative solution to precipitate sulfide. These samples were mixed well and then centrifuged at high speed for 15 minutes. The supernatant was transferred using a pipette to a different screw-top plastic vial, and its volume was measured. The supernatant was used for sulfate concentration and stable isotope analyses, whereas the zinc sulfide precipitate was used for sulfide concentration and stable isotope analyses.

Both types of samples were stored at 4°C until analysis.

pH and salinity were measured within an hour after sample collection. Salinity was measured using a hand-held refractometer. pH was measured using a MI-710 micro-

75 combination pH electrode (Microelectrodes, Inc., Londondery, New Hampshire, USA) connected to a high-impedance analyzer system (Cameron Instrument Company, Port

Aransas, Texas, USA). Methane concentrations of these samples were determined by headspace analysis using a Shimadzu Gas Chromatograph-14A with a Flame Ionization

Detector and a 2 mL sample loop (Shannon and White 1996). Values were determined using NIST (National Institute of Standards and Technology) traceable methane standards (Sigma-Aldrich Company). The error associated with repeated analyses was <

5%. Water volumes were determined by gravimetric analysis and methane concentrations of the original water samples were calculated from headspace values using Henry's law.

Sulfate concentrations of supernatants were measured using barium chloride turbidimetry using SulfaVer reagents (Hach Chemical Company). The sulfate concentration was calculated from a standard curve generated using serial dilutions of a

2500 mg/L sulfate standard (Hach Chemical Company) diluted using sulfate-free ASW.

Sulfide concentration was determined by Cline’s spectrophotometric method using the sulfide reagent kit (Hach chemical company). The zinc sulfide precipitate was suspended in 1 ml DDI water and 250 µl of the suspension was used for sulfide analysis. Sulfide concentration was calculated using a standard curve generated using serial dilutions of a

7200 mg/L sodium sulfide solution in de-oxygenated DDI water. The exact concentration of each dilution was confirmed using gas chromatography (Childress et al. 1984; Scott et al. 1999).

Barium chloride salt was added in excess to the remaining supernatant to precipitate barium sulfate. To eliminate barium carbonate from the precipitate, it was rinsed once with 1 M hydrochloric acid. It was then dried at 60 °C for 24 hours and used

76 for δ34S analysis of sulfate. The remaining zinc sulfide suspension was also dried and used for δ34S analysis of sulfide. Small quantities of barium sulfate (~0.3 mg) and zinc sulfide (0.2 mg) were weighed into 5x7 tin capsules and mixed with 5-fold quantity of

V2O5 (Hurtgen et al. 2002). The samples were combusted to SO2 at 1000-1050˚C and were analyzed on a VG Optima Series II mass spectrophotometer. Sulfur isotope ratios were expresses in the conventional δ34S notation as per mil (‰) deviations from the S isotope composition of Canon Diablo Troilite. Sulfur isotopic results were typically reproducible within ± 0.2‰.

Assessment

12 peepers were successfully deployed at and collected from a hydrocarbon seep site located at 540 m depth on the seafloor of the Gulf of Mexico, using the manned

Johnson Sea-Link submersible. Sediment depth profiles of pH, salinity, methane, sulfate and sulfide concentrations were obtained from all 12 peepers. Moreover, stable sulfur isotope compositions of both sulfate and sulfide were determined whenever pore water concentrations were high enough to provide sufficient samples for isotope analyses.

Figure 4-4 shows the depth profiles of all parameters measured using a peeper deployed in sediment underlying a Beggiatoa bacterial mat. The profiles are diagnostic of bacterial sulfate reduction occurring in hydrocarbon rich sediments (Aharon and Fu 2000; Aharon and Fu 2003; Arvidson et al. 2004; Boetius et al. 2000; Joye et al. 2004).

Figure 4-4: Depth profiles of various parameters obtained using the peeper indicated in figure 2.

Once the peepers were retrieved on board the ship, they could be rapidly dismantled to obtain the samples. Six peepers were completely processed within 15 minutes after retrieval. The rubber radial seals appeared to maintain sample integrity after collection. To test the effectiveness of the radial seals, two peepers were loaded with dialysis bags filled with DDI water. They were “closed” and attached to the outside of the

Johnson Sea-link submersible and carried to depths up to 650 m during a 4-hour dive.

Since the peepers were exposed to seawater during the entire dive, there was a strong gradient for diffusion of chloride ions into the bags. However, the radial seals prevented

78 diffusion of seawater into the dialysis bags as the salinity of the water contained within the 12 dialysis bags remained at 0 ‰ when sampled at the end at the end of the dive.

Dialysis samplers could produce erroneous data if they are not equilibrated in situ for a sufficient time period (Carignan 1984). Our peepers were equilibrated in situ for 35 days. Equilibration of peeper cells is dependent on a variety of factors including the dimensions of the peeper cell, the dialysis membrane used, the temperature at which equilibration occurs, and the diffusion coefficients of pore water constituents of interest

(Carignan 1984; Laforce et al. 2000; Webster et al. 1998). In order to estimate equilibration times for peeper cells of our specifications, we used a theoretical model in which we simulated diffusion of sulfate (one of the major ions of interest) through the sediment into the peeper cells. Since diffusion coefficients of methane and sulfide in seawater are greater than that of sulfate (Iverson and Jorgensen 1993; Stumm and

Morgan 2000), estimates obtained for sulfate ions are conservative.

The factors affecting the diffusion coefficient of sulfate within the sediment are, temperature, porosity and tortuosity of the sediment (Carignan 1984). Here, the diffusion

coefficient of sulfate was calculated using the equation Ds = Do 1+ n(1! "), where Ds is the sediment diffusion coefficient, Do is the diffusion coefficient in seawater, n = 3 for clay-silt sediments, and φ is the porosity of the sediment (Iverson and Jorgensen 1993).

The porosity value (φ = 0.765) measured at 30 cm depth in the sediments at our study site in the Gulf of Mexico (Cordes et al. 2005a) was used to obtain an estimate of Ds. The

2 -1 2 -1 value of Ds was 0.294 cm day , based on a Do value of 0.501 cm day measured for sulfate in seawater at 5°C (Li and Gregory 1974).

79 30

5 Sulfate concentration (mM)

0 0 20 40 60 80 100 Time (hours)

Figure 4-5: Results from the equilibration experiment performed with dialysis bags inside a cold room maintained at 6-8°C. 21 dialysis bags were filled with sulfate-free ASW and suspended inside a container with two liters of ASW containing 24 mmol L-1 magnesium sulfate. 3 bags each were collected at 7 different time points ranging from 0 to 94 hours, and the sulfate concentrations of the solutions inside the bags were determined. Data points in the graph show the average ± standard deviation of sulfate concentrations inside the bags at various time points. Dashed lines show the average ± standard deviation of sulfate concentration in the solution outside the bag, which did not vary over the course of the experiment. The solid line shows the best-fit curve corresponding to the equation , where Ci is the sulfate concentration inside the bags at time, t. Ci = 23.6 * #"1! exp(! 0.153 t)%$

To determine the diffusion coefficient of sulfate through the dialysis membrane, an equilibration experiment was performed with the dialysis bags inside a cold room maintained at 6 °C (Figure 4-5). In this experiment, the sulfate concentration inside the dialysis bags, Ci, started at zero and approached the outside solution sulfate

-1 concentration, Co (24 mmol L ). The diffusion coefficient was calculated by fitting the data to the equation Ci = Co [1! exp(!at)] (Sten-Knudsen 1978), where the parameter

a = DS! vh . Here D is the diffusion coefficient (cm2 sec-1), S is the surface area

(calculated from the dimensions of the dialysis bags to be 15.7 cm2), α is the distribution coefficient (assumed to be 1), v is the volume of the solution inside the bag (4 cm3), and h is the membrane thickness (0.1 cm). Based on these values, and a best-fit value of a =

80 0.153, the diffusion coefficient of sulfate across the dialysis membrane was calculated to be 0.0936 cm2 day-1.

In the sediment diffusion model, sulfate ions diffusion was assumed to occur from the surrounding sediment into each peeper cell across a sediment area of 4 cm2 (the area of the opening of inner chamber slots). Moreover, sulfate diffusion into dialysis bags was assumed to occur across the surface area of the membrane exposed to the outside, i.e. half of the total bag surface area. Figure 4-6 shows the simulation results of the model at various sediment sulfate concentrations. The simulations indicate that the peeper cells would reach steady state, at least with respect to sulfate concentration, within 15-20 days.

This estimate agrees well with the 20-day equilibration time recommended by Carignan

(1984) for cold (4-6°C) sediments.

In each peeper, one slot was positioned 10 cm above the sediment-seawater interface. The average values of parameters analyzed from 12 such slots are shown in

Table 4-1. The average sulfate concentration was found to be 29.5 mmol L-1 and the average sulfate stable isotope composition was found to be 20‰. Both these values are typical of ambient seawater in the Gulf of Mexico (Aharon and Fu 2000). Thus isotopic equilibration was achieved at least in the top slot.

81 30

20 2mM 10mM 15 15mM 25mM 10 Internal [SO4] (mM)

0 0 5 10 15 20 25 30 35 Time (days)

Figure 4-6: Results of a model in which sulfate diffusion from the surrounding sediment into a dialysis bag was simulated at various pore water sulfate concentrations.

Table 4-1: Average ± standard deviation of pH, salinity, sulfate and sulfide concentration, and δ34S of sulfate obtained from 12 dialysis bags placed at 10 cm above the sediment-seawater interface.

Parameter pH Salinity Sulfate Sulfide δ34S-sulfate

(‰) (mM) (mM) (‰)

Mean ± SD 7.28±0.29 34.3±1.2 29.5±0.8 0.0±0.0 20.2±0.2

Discussion

The pore water chemistry of deep-sea hydrocarbon seep sediments is of interest to geochemists, and also to biologists studying the macro- and micro fauna associated with these environments. Conventional cores that are typically used to sample pore water from these sediments suffer from many drawbacks. The sampling depth rarely exceeds 25-30 cm. Moreover, sediments collected using cores have to be extensively processed,

82 increasing the likelihood of post-sampling artifacts. In shallow sediments, dialysis samplers have been used extensively to supplement cores. However, to date, they have rarely been used to sample inaccessible sediments such as deep-sea sediments.

In this study, a simple yet effective design was developed for a pore water peeper that can be easily deployed and collected using the robotic arm of a manned or remotely operated submersible. The peeper maintains sample integrity for at least 1-2 hours after collection. Also, these peepers can provide concentration and isotope profiles up to 60 cm deep in the sediment. Post-collection sampling is rapid and convenient, thereby minimizing the chances for introducing artifacts. 4 ml of sample can be collected using this method, which provides sufficient sample volume for multiple analyses, including stable isotope analyses.

Comments and recommendations

Although our peepers were designed for use in deep-sea sediments, they could be easily adapted for use in shallow sediments where it is difficult to process samples soon after collection. For example, they can be used in shallow sediments accessible by

SCUBA or from other remote field locations. We used durable materials that could be handled without damage by the robotic arm of a submersible. One out of 12 peepers we deployed broke at the shaft of the inner handle during collection. We had constructed the inner handle shafts using solid PVC, but we recommend replacing that with stainless steel for use with submersibles. The material for the outer tube should be chosen carefully. We found that materials such as PVC that are extruded into tubes have too large a variation in inner diameter to be used for this design. If reduction in costs were a priority, sections of

83 ordinary rubber hose could replace custom radial seals and cheaper non-cellulose based dialysis membranes tied on either end with nylon string could replace the PVDF membrane.

Acknowledgements

We thank the captain and crew of the Research Vessel (RV) Seaward Johnson II as well as the crew and pilots of the Johnson Sea Link submersible (Harbor Branch

Oceanographic Institution). We are indebted to Dr. Edith Widder and Dr. Tamara Frank for devoting time and effort on their research cruise to facilitate collection of the peepers.

Dr. Tamara Frank, Alison Sweeney, Erika Heine, Michael Conda, and Michelle Hoogstra all helped with sampling the peepers. We would also like to thank Dr. Michael Arthur for his input on the peeper design, and Burt Thomas for conducting the methane analyses.

84 CHAPTER 5

Modification of sediment geochemistry by the hydrocarbon seep tubeworm Lamellibrachia luymesi: A combined empirical and modeling approach

Sharmishtha Dattagupta1, Michael Arthur2, Kathryn E. Luley1, Guy Telesnicki1 and

Charles R. Fisher1

1 Department of Biology, Penn State University, University Park, PA-16802.

2 Department of Geosciences, Penn State University, University Park, PA-16802.

Abstract

The sulfide-oxidizing symbiont-containing tubeworm Lamellibrachia luymesi

(Phylum Annelida, Family Siboglinidae) is a dominant member of deep-sea hydrocarbon seep ecosystems on the Gulf of Mexico seafloor. This tubeworm forms large aggregations that can live for centuries and provide habitat for an assortment of associated fauna. A recent diagenetic model has suggested that persistence of these tubeworms for such long time periods is contingent upon their ability to supply sediments with sulfate. To examine this hypothesis, we characterized the tubeworm’s geochemical environment using specially devised submersible operated pore water sampling devices and compared the measured depth profiles with those predicted by a sulfate diffusion- reaction-supply model. We found a large range of sulfide concentrations in the tubeworm habitat, indicating that this species can live under conditions of both high and low sulfide availability. In sediments rich in hydrocarbons, we found compelling evidence that tubeworms enhance microbial sulfide production, likely through a combination of sulfide uptake and sulfate and proton release through their root-like structures into the surrounding sediment. In sediments low in hydrocarbon content, sulfide production is hydrocarbon-limited rather than sulfate-limited, and our model predicts that tubeworm growth would be severely limited by the low sulfide availability.

86 Introduction

Deep-sea ecosystems flourish in the absence of photosynthetic input by deriving energy from reduced compounds such as methane and hydrogen sulfide. At deep-sea hydrocarbon seeps, anaerobic microbes in the sediment couple hydrocarbon oxidation with sulfate reduction to produce sulfide, which sustains rich communities of symbiotic macrofauna such as tubeworms, clams, and mussels (Boetius et al. 2000; Joye et al. 2004;

Sibuet and Olu 1998; Treude et al. 2003; Weber and Jorgensen 2002). At 400-1000 m depths on the seafloor of the Gulf of Mexico, seepage of hydrocarbons supports large aggregations of sulfide-oxidizing tubeworms dominated by the species Lamellibrachia luymesi. This tubeworm has an extraordinary lifespan of 170 to 250 years and provides habitat for a wide variety of macrofauna (Bergquist et al. 2002; Bergquist et al. 2003;

Bergquist et al. 2000; Cordes et al. 2003; Cordes et al. 2005b; Fisher et al. 1997).

Like other vestimentiferan tubeworms, L. luymesi derives nutrition from sulfide oxidizing gamma-proteobacteria that reside in an organ called the trophosome

(Cavanaugh et al. 1981; Childress and Fisher 1992; Mcmullin et al. 2003). L. luymesi acquires sulfide using long posterior extensions of its body (“roots”) that are buried in the sediment (Freytag et al. 2001; Julian et al. 1999), and calculations from an individual- based model indicate that a moderate-sized tubeworm aggregation can require sulfide at an approximate rate of 30 mmol h-1 (Cordes et al. 2003). The survival of L. luymesi aggregations for periods in excess of 200 years is in part contingent upon sufficient sulfide availability (Cordes et al. 2005a).

In hydrocarbon rich sediments, microbial sulfide production may become sulfate limited (Aharon and Fu 2000; Arvidson et al. 2004; Joye et al. 2004). A number of

87 studies have proposed that L. luymesi could release sulfate across its roots deep into the sediment, and thereby enhance its own sulfide supply (Cordes et al. 2005a; Cordes et al.

2003; Freytag et al. 2001; Julian et al. 1999). Sulfate is produced as a byproduct of sulfide oxidation by vestimentiferan symbionts, and must be excreted by the host tubeworm to maintain homeostasis (Dattagupta et al. 2006a; Goffredi et al. 1998;

Goffredi et al. 1999a). Cordes et al. (2005) used a sediment diagenesis model to demonstrate that sources of sulfide such as advection and diffusion of deep sulfide and reduction of seawater sulfate could sustain large L. luymesi aggregations for only about

40 years. They proposed that sulfate release by tubeworm roots was necessary to sustain microbial sulfide production in sufficient quantities to support average sized tubeworm aggregations for hundreds of years. Furthermore, laboratory experiments have shown that

L. luymesi excretes a majority of the sulfate produced by its symbionts across its root surface (Dattagupta et al. 2006a). L. luymesi also eliminates protons, which are another byproduct of sulfide oxidation, primarily across its root. Since protons may be depleted during anaerobic sulfate reduction (Aharon and Fu 2000; Thauer et al. 1977), replenishment of protons through root release may be an additional mechanism for L. luymesi to enhance sulfide production in sediments surrounding its roots.

Mature tubeworm aggregations have extensive roots buried over a meter deep in the sediment, which can exert a significant impact on the geochemistry of the surrounding “rhizosphere” (Cordes et al. 2005a; Cordes et al. 2003; Dattagupta et al.

2006a). To date, empirical evidence of tubeworm influence on sediment geochemistry has been elusive, as focused sampling near tubeworm aggregations requires the use of submersibles. Further, sediments at Gulf of Mexico seeps have typically been sampled

88 using cores that provide pore water profiles only to 20-25 cm depths (Aharon and Fu

2000; Aharon and Fu 2003; Arvidson et al. 2004; Joye et al. 2004). However, L. luymesi roots extend below the carbonate substrate to which the tubeworms are attached, and in mature aggregations, the substrate may be buried by sediment several centimeters thick

(Bergquist et al. 2002; Fisher et al. 1997). Thus sampling to sediment depths up to 25 cm may be insufficient to assess tubeworm root influence on the sediment geochemistry.

Attempts have been made to sample to deeper near tubeworm aggregations using probe samplers (Bergquist et al. 2002; Julian et al. 1999), but this method is inappropriate for obtaining depth profiles with fine resolution. To address whether tubeworms impact the geochemistry of their “rhizosphere”, we developed sampling devices specifically designed to collect pore water at 10 cm increments in sediments surrounding tubeworm aggregations, down to 60 cm below the seawater-sediment interface (Dattagupta et al.

2006b). We analyzed pH, salinity, sulfate, sulfide and methane concentrations, and sulfur stable isotope compositions of sulfate and sulfide in pore water collected from sediments adjacent to tubeworms, as well as from sediments outside the immediate influence of tubeworms. We also developed a geochemical model to simulate the influence of tubeworm sulfide uptake and putative sulfate release on sediment depth profiles of sulfate and sulfide concentrations and isotopic compositions in the sediments.

Using this combination of empirical measurements and theoretical predictions, we present strong evidence for tubeworm influence on the biogeochemistry of their habitat.

89 Materials and Methods

Pore water sampling scheme

In summer 2004, sampling was conducted at the site GC234 located at approximately 540 m depth on the upper Louisiana slope of the Gulf of Mexico

(27˚44.7’N, 91˚13.3’W), using the Johnson sea-link manned submersible. Pore water samples were collected using dialysis samplers (“peepers”) that had one sampling port at

10 cm above the seawater-sediment interface, and six ports at 10 cm depth increments below the interface. Details of the sampling method are described in Dattagupta et al.

(2006b). Peepers were deployed either less than 30 cm away from the base of a tubeworm aggregation (referred to as “near” locations), or in control sediments approximately 6 meters away from any tubeworm aggregation (referred to as “far” locations) (Figure 5-1).

For “near” deployments, two large tubeworm aggregations were selected containing over a thousand individuals and about 1.5 m tall. These aggregations were about 30 meters away from each other. One of the aggregations appeared to be in a zone of active seepage, with orange Beggiatoa mats at its base and new tubeworm recruitment less than a meter away from it. The corresponding “far” controls were also deployed in sediment underlying an orange Beggiatoa mat. Since orange Beggiatoa are commonly associated with conduits of hydrocarbon seepage, and are found in oil saturated sediments (Nikolaus et al. 2003; Sassen et al. 1993; Zhang et al. 2005), we designated these locations “oily”.

The second tubeworm aggregation appeared to be in a zone of much lower seepage rate, without bacterial mats at its base. The corresponding “far” controls were deployed in sediment without visible influence of hydrocarbon seepage (Joye et al. 2004), and these locations were designated “plain”. 12 peepers were deployed in total, with 3 replicates

90 each in 4 different locations: “oily-far” (OF), “oily-near” (ON), “plain-far” (PF) or

“plain-near” (PN). These peepers were incubated in situ for 35 days, which was sufficient to bring them into equilibrium with the sediment pore water (Dattagupta et al. 2006b).

(a) Distance : Far (b) Distance : Near Sediment : “Oily” Sediment : “Oily”

Figure 5-1: Peepers deployed in situ at four different locations, in replicates of 3. Peepers were deployed either ~30 cm away from tubeworms (“Near”) or > 6 meters away from large aggregations (“far”). Also, they were either deployed in sediments underlying orange Beggiatoa mats (“oily” sediments) or in sediments with apparently low hydrocarbon flux (“plain” sediments).

Chemical analyses

Pore water samples were analyzed for pH and salinity, sulfate, sulfide, and methane concentrations, and sulfur stable isotope ratios. Details of the analysis are described in Dattagupta et al. (2006b). Briefly, salinity was measured using a hand-held refractometer and pH using a MI-710 micro-combination pH electrode connected to a high-impedance analyzer system. Methane concentrations were determined by headspace analysis using a Shimadzu Gas Chromatograph-14A equipped with a Flame Ionization

Detector (reproducible within ± 5%). Sulfate concentrations were determined using barium chloride turbidimetry (reproducible within ± 2%) and sulfide concentrations using

Cline’s spectrophotometric method (reproducible within ± 4%). Sulfur stable isotope compositions were analyzed using a VG Optima Series II mass spectrophotometer and sulfur isotope ratios were expressed using the conventional δ34S notation as per mil (‰) deviations from the S isotope composition of Canon Diablo Troilite. Sulfur isotopic results were typically reproducible within ± 0.2‰. The sulfur stable isotope composition of sulfate sulfur in blood was determined for 5 animals collected from the “oily” tubeworm aggregation. The sulfate from the blood was collected as previously described

(Dattagupta et al. 2006a) and precipitated using barium chloride. Blood sulfate sulfur stable isotope composition was determined in the same manner as pore water sulfate samples.

92 Estimation of sulfate reduction rates

Sulfate reduction rates were estimated from sulfate depth profiles using a one- dimensional diffusion-advection-reaction model as described by (Aharon and Fu 2000;

Aharon and Fu 2003). Briefly, sulfate concentration profiles were fitted to an exponential curve (1), where x is the sediment depth, CO is the sulfate concentration at x = 0, and C∞ is the asymptote sulfate concentration. The maximum sulfate reduction rate (SRRmax) was then calculated using equation (2), where DSO4 is the sulfate diffusion coefficient, and ω is the average sedimentation rate at the Gulf of Mexico (6cm/1000 yr) (Aharon and Fu

2000), and the parameter b is derived from equation (1). Since tubeworm aggregations can enhance sedimentation in their vicinity (Macdonald et al. 1989), calculations based on equation (2) represent minimum rates for sediments near tubeworms.

!bx C(x) = (CO ! C" ) e + C" (1)

2 SRRmax = (CO ! C" ) # (DSO4b + $b) (2)

Sulfate diffusion-reaction-supply model

For this model, a one-meter tall tubeworm aggregation containing 1000 individuals was assumed centered within a 200 cm radius cylinder of sediment. For simplicity, the entire aggregation was assumed to be composed of L. luymesi individuals of uniform size. The tubeworm “root ball” was modeled as hemisphere with a 100 cm radius (Cordes et al. 2005a). The surrounding sediment was divided into 7 blocks of 5-10 cm thickness. The centers of blocks 1-7 were, respectively, at 2.5 cm, 10 cm, 20 cm, 30 cm, 40 cm, 50 cm, and 60 cm depths below the sediment-seawater interface. The tubeworm root ball hemisphere was similarly divided into blocks with the same thickness

93 as the sediment blocks. For simplicity, the hemisphere was assumed to comprise of cylinders of successively smaller radii stacked upon each other (Cordes et al. 2005a).

Details of the dimensions of the model blocks are given in Table 1.

94 Table 5-1: Dimensions of sediment “blocks”, as well as other parameters used for the model, such as sulfate and sulfide diffusion coefficients. SA= surface area; D = diffusion coefficient. The SA fraction (fSA) is the fraction of the total root ball hemisphere surface area.

Block Depth Root Root SA Root Sediment Porosity DSO4 DH2S # at cylinder cylinder fraction cylinder volume center radius SA volume 2 3 5 3 5 2 2 (cm) (cm) (cm ) fSA (cm )*10 (cm )*10 (cm /yr) (cm /yr) 1 2.5 100.0 3140 0.060 1.57 4.71 0.810 134.6 254.8 2 10 99.9 6272 0.119 3.13 9.42 0.774 126.5 239.3 3 20 98.9 6209 0.118 3.07 9.49 0.766 124.8 236.1 4 30 96.8 6081 0.116 2.94 9.62 0.765 124.5 235.7 5 40 93.7 5883 0.112 2.76 9.80 0.765 124.5 235.6 6 50 89.3 5608 0.107 2.50 10.06 0.765 124.5 235.6 7 60 83.5 5245 0.100 2.19 10.37 0.765 124.5 235.6

The sediment diffusion coefficients of sulfate and sulfide were calculated at

depths at the center of each sediment block using equation (3), where Do is the diffusion

coefficient corrected for temperature, pressure and salinity, n is set to 2.75, and φ is the

sediment porosity at the particular sediment depth (Cordes et al. 2005a; Iverson and

Jorgensen 1993). Porosity was not measured in this study but calculated using equation

(4) (Cordes et al, 2005), based on empirical data available in Arvidson et al. (2004). In

this equation, φz is the porosity at depth z cm.

D D = o (3) s 1+ n(1! ")

"0.21z !z = 0.076 * e + 0.765 (4)

Ambient seawater sulfate concentration was assumed to be 29-30 mM, and the

sulfide concentration was assumed to be zero (Aharon and Fu 2000; Aharon and Fu

2003). In the model, sulfate and sulfide were allowed to diffuse between each block and

between Block-1 and the overlying seawater. Diffusion between successive sediment

blocks was calculated using Fick’s second law according to equation (5), where Dd is the

95 diffusion coefficient, !C !x is the concentration gradient, and !C !t is the change in concentration per unit time. Hydrocarbon fuelled sulfate reduction was allowed to occur in each sediment block, according to Monod kinetics described by equation (6), where R is the rate of sulfate reduction, Rm is the asymptote maximum value of R, S is the sulfate concentration, and Ks is the saturation constant, which is equal to the sulfate

concentration at which R = Rm 2 (Boudreau and Westrich 1984). Rm was assumed to be

2- -3 -1 8.4 µmol SO4 cm day , and ks was set at 0.444 mM, based on the sulfate reduction kinetics determined for GOM sediments exposed to methane and crude oil at various sulfate concentrations (Luley 2006). For simplicity, we assumed in our model that oil and methane were uniform at all sediment depths and therefore sulfate reduction rates were a function of sulfate concentrations described by the Monod equation. Based on empirical studies with sulfate reducing bacteria of the Desulfovibrio genus grown in culture, sulfate reduction in the model was set to be inhibited at all sulfide concentrations greater than 8 mM, with inhibition percentage varying linearly between 0% at 8 mM and 50% at 15 mM

(Okabe et al. 1992).

The net change in the amount of a sulfur species (either sulfate or sulfide) in a sediment block, Ms, in the time interval ‘dt’ was then defined by the equation (7), where

Mdiff-in is the mass diffusing into a particular sediment block, Mdiff-out is the mass diffusing out of the sediment block, and MSR is the mass reacting to form sulfide. MSR was subtracted from the other terms in the case of sulfate, and added in the case of sulfide. To perform a sensitivity analysis on the influence of Rm on sediment concentration profiles,

the Rm value was multiplied by a “rate fraction” (RF) term ranging from 0.001 to 1.

2 2 !C !t = Dd (! C ! x) (5)

96 R S R = m (6) ks + S

M s (t) = M s (t ! dt) + M diff !in ! M diff !out ± M SR (7)

Sulfide mineralization, specifically pyrite formation, is an important process in sediments where sulfate reduction occurs. This process is influenced by a variety of factors including iron availability, pH, sulfide concentrations and organic matter content of sediments (Morse and Wang 1997). Because these factors were not accounted for in our model, and the rate of sulfide mineralization in sediments at the Gulf of Mexico is not well constrained (Aharon and Fu, 2003), we did not include this process in model construction. Therefore, pore water sulfide concentrations generated by our model are likely overestimates of true values, as sulfide mineralization is likely to decrease free sulfide content.

The influence of the tubeworm aggregation on the sediment geochemistry was further included in the model. Sulfide uptake by tubeworms at various sediment sulfide concentrations was estimated based on empirical sulfide uptake measurements for L. luymesi individuals exposed to various external sulfide concentrations (Freytag et al.

2001). The data were fitted to a logarithmic curve to derive equation (8), where Uind is the uptake rate of an individual tubeworm in µmoles g-1 h-1 and [HS] is the external sulfide

-1 concentration in mmol L . Uind was set at zero at sulfide concentrations less than or equal to 0.1 mM, since equation (8) generates negative values for sulfide uptake at these sulfide concentrations. Individual tubeworm mass was calculated using the empirical relation between tube length and wet weight (Cordes et al. 2003). The total mass of the aggregation (Maggr) was calculated to be 5250 grams, by multiplying individual mass by

97 1000. Total sulfide uptake of the aggregation (Uaggr) was estimated using equation (9), where a metabolic scaling exponent of 0.75 was applied to Maggr. Uptake across each root ball block was calculated by multiplying Uaggr with the fraction of the total root ball area corresponding to that block (fSA; table 5-1).

Uind = 3.67ln[HS] + 8.36 (8)

0.75 Uaggr = (M aggr ) {3.67ln[HS] + 8.36} (9)

Sulfate release was assumed to occur across the tubeworm root surface into the surrounding sediment. Some of the sulfate derived from sulfide oxidation within tubeworms might be released across their plumes (Dattagupta et al. 2006a). Thus total sulfate release from the tubeworm aggregation (Raggr) was computed from the equation

(10), where froot is the fraction of total sulfate that is released by tubeworms across their roots and lies between 0 and 1. Sulfate release into a particular sediment block (Rn) was computed using the equation (11), where wn is a fraction between 0 and 1 and satisfies

7 the criterion that . The likelihood that sulfate would be released into a certain !wn = 1 n=1 sediment block was assumed to be greater when there was a greater sulfate gradient between inside the tubeworm’s body and the sediment block sulfate concentration. The sulfate concentration of L. luymesi body fluids is 23 mM (Dattagupta et al. 2006a).

Therefore, wn was calculated from equation (12), where the factor wn´ was first calculated

2- using equation (13). Here [SO4 ] n is the sulfate concentration in sediment block n. If

2- [SO4 ]n was greater than 23 mM, wn´ was set equal to zero.

Raggr = Uaggr ! froot (10)

Rn = wn ! Raggr (11)

98 wn´ wn = 7 (12) !wn´ n=1

" 23 ! [SO2! ] % w ´ = 4 n * f (13) n #$ 23 &' SA

The isotopic composition (δ34S) of pore water sulfate and sulfide was modeled by taking diffusion, sulfate reduction, tubeworm sulfide uptake and sulfate release into consideration. Seawater sulfate δ34S was set at 20‰, the value measured for GOM bottom water by Aharon and Fu (2000) and in this study. In the absence of sulfate reduction, δ34S values of pore water sulfate are typically invariable with sediment depth

(Bottcher et al. 2006). Therefore we assumed no isotopic discrimination associated with diffusion of sulfate or sulfide through the sediment column. For simplicity, fractionation associated with sulfate reduction (Δδ34S) was assumed to be invariable with depth. To perform a sensitivity analysis, the Δδ34S value was varied between 15-30 ‰ in various runs of the model, based on the fractionation values found for cold-adapted sulfate reducing bacteria (Brüchert et al. 2001). The change in isotope composition, δ34S, of a sulfur species due to either diffusion, or sulfate reduction, was calculated using mixing equations. The change in δ34S in the time interval ‘dt’ based on diffusion was estimated using equation (14), where Mdiff is the mass diffusing in or out of the sediment block, and

34 34 δ Ssource is the δ S composition of the source sediment block from which diffusion occurs. Similarly, the change in δ34S in the time interval ‘dt’ based on sulfate reduction was defined by the equation (15).

34 34 34 M s (t) ! " S(t) = M s (t # dt) ! " S(t # dt) ± M diff ! " Ssource (14)

M (t) 34S(t) M (t dt) 34S(t dt) % 34S(t dt) 34S M ' (15) s ! " = s # ! " # # &{" # # $" } ! SR (

Further, sulfide uptake by L. luymesi was assumed to be accompanied isotopic discrimination of 1.5‰, based on the fact that sulfide uptake and incorporation into organic material by L. satsuma, a tubeworm species closely related to L. luymesi, was found to be approximately 1.5‰ (Miura et al. 2002). The change in sulfide isotopic composition of a sediment block due to tubeworm uptake in the time interval ‘dt’ was estimated using equation (16), where Muptake was the mass of sulfide taken up from a

34 particular sediment block in the time interval, and δ Ssource was the isotope composition of the source sediment block. The sulfur stable isotope composition of tubeworm blood

34 sulfide (δ SHS-TW) was then computed using the equation (17). Since dissimilatory sulfide oxidation by vestimentiferan symbionts is associated with a negligible amount of isotopic fractionation (Vetter and Fry 1998), we assumed that tubeworm blood sulfate had the same sulfur isotopic composition as blood sulfide. The change in sulfide isotopic composition of a sediment block due to tubeworm sulfate release in the time interval ‘dt’ was estimated using equation (18), where Mrel was the mass of sulfate released into a particular sediment block.

34 34 34 M s (t) ! " S(t) = M s (t # dt) ! " S(t # dt) # M uptake %$" Ssource # 1.5'& (16)

7 34 $ M uptake # ! Ssource 34 Block =1 ! SHS"TW = 7 (17) $ M uptake Block =1

100 34 34 34 M s (t) ! " S(t) = M s (t # dt) ! " S(t # dt) + M rel ! " SHS#TW (18)

Estimations of tubeworm growth and metabolism

Using simulations from our model, we calculated the total uptake by an individual tubeworm within an aggregation of 1000 individuals, over a period of 25 years under steady-state conditions. In these simulations, we varied the maximal sulfate reduction rate

-3 -1 -3 -1 (Rm) between 0.0042 µmol cm day to 8.4 µmol cm day , which correspond with integrated sulfate reduction rates of 2.5 mmoles m-2 day-1 to 5000 mmoles m-2 day-1, respectively. Using equation (19) that describes the relation between individual uptake,

Ui, and individual growth, gi (Cordes et al. 2005a; Cordes et al. 2003), we estimated the average growth of an individual tubeworm at various simulated uptake rates generated by our model. Equation (19) assumes a basal metabolic requirement of 1.60 µmol g-1 h-1 based on empirical measurements (Freytag et al. 2001), and a requirement of 4.38 µmol g-1 h-1 to achieve a growth rate of 10 cm yr-1 (Cordes et al. 2003). Although this equation does approximate metabolic scaling with size (Cordes et al. 2005a), it does not account for energy requirements for metabolic processes such as reproduction or tissue repair.

Thus the growth estimates we obtain represent maximum values.

( g + " i % (19) Ui = M i *1.60 + $ 4.40 ! ' - ) # 10& ,

101 Results

Pore water pH and salinity

Pore water pH profiles of “plain-near” and “plain-far” sediments were similar and pH increased slightly with depth (Table 5-1, Figure 5-2). Pore water pH of “oily-far” sediments also increased with depth, whereas pore water pH of “oily-near” sediments decreased sharply at depths between 20-60 cm below the sediment-seawater interface

(Table 5-2). The mean pore water pH of “oily-far” sediments was significantly greater than that of “oily-near” sediments (P <0.0001; two-sided t-test; df=40). Pore water salinity of “plain-near” and “plain-far” sediments were similar to each other and did not vary substantially with depth (Table 5-1, Figure 5-3). However, pore water salinity of

“oily-far” sediments decreased slightly with depth, and values were significantly lower than that of “oily-near” sediments (Table 5-2, Figure 5-3, P=0.006; two-sided t-test; df=40). Freshening of pore water observed in “oily-far” sediments could be due to decomposition of gas hydrates in the underlying sediment (Kennicutt Ii and Brooks

1990).

102 Figure 5-2: pH profiles in (a) “plain” and (b) “oily” sediments near and far from tubeworms. Table 5-2: Chemical and isotope compositions of pore water collected using three replicate peepers deployed in PF (plain-far) and PN (plain-near) locations. nd= non-detectable.

Peeper # Depth pH Salinity [Sulfate] [Sulfide] δ34S- (cm) (‰) (mM) (µM) sulfate -10 6.95 35 29.0 nd 20.1 10 7.44 33 29.6 27.0 20.3 20 7.61 32 23.1 9.8 20.3 PF1 30 7.54 31 20.3 32.3 20.9 40 7.8 35 20.3 14.3 20.5 50 7.64 34 20.6 16.6 20.9 60 7.7 35 17.0 13.1 20.7 -10 7.51 34 30.1 nd 20.1 10 7.47 34 26.7 20.0 20.2 20 7.51 34 21.0 4.7 20.9 PF2 30 7.82 33 20.5 9.9 20.5 40 7.89 32 22.1 nd 20.6 50 7.69 33 18.2 8.6 20.7 60 7.68 35 21.2 6.9 20.9 -10 29.4 nd 10 7.4 34 27.0 7.4 20 7.54 34 19.0 7.9 PF3 30 7.54 33 18.0 7.2 40 7.68 34 21.1 12.6 50 7.69 32 21.2 18.2 20.6 60 7.65 35 20.5 5.0 -10 7.15 35 29.1 6.9 20.7 10 7.51 34 22.2 34.9 20.6 20 7.82 35 21.4 23.9 20.8 PN1 30 7.56 35 21.5 11.5 21.3 40 8.1 35 26.0 12.1 21.9 50 7.49 34 27.2 36.7 22.0 60 7.46 36 29.3 14.2 21.9 -10 7.33 35 30.3 nd 20.3 10 7.63 35 23.5 19.4 20.6 20 7.76 33 27.0 11.7 21.3 PN2 30 7.68 33 23.1 7.5 20.7 40 7.59 35 21.6 5.8 21.1 50 7.46 34 23.2 8.6 21.1 60 7.5 35 24.7 62.7 21.9 -10 7.07 35 28.9 nd 10 7.62 35 24.8 6.1 20 7.61 35 22.8 8.6 PN3 30 7.58 32 22.8 28.7 21.4 40 7.58 34 26.7 7.2 50 7.73 32 24.5 16.6 60 7.45 35 23.0 8.0

103 Table 5-3: Chemical and isotope compositions of pore water collected using three replicate peepers deployed in OF (oily-far) and ON (oily-near) locations. nd= non-detectable.

Peeper # Depth pH Salinity [Sulfate] [Sulfide] δ34S- δ34S- (cm) (‰) (mM) (mM) sulfate sulfide -10 7.11 35 28.5 20.2 10 7.71 34 8.2 10.7 25.5 14.3 20 7.8 34 1.5 9.3 27.8 14.6 OF1 30 7.64 34 1.4 9.7 29.0 15.4 40 7.65 34 1.3 10.9 30.4 16.0 50 8.06 33 1.5 9.3 28.0 16.6 60 7.83 31 1.9 8.8 28.2 16.0 -10 7.56 32 30.0 nd 20.0 10 7.75 35 15.8 3.7 25.8 20 7.86 34 3.4 7.3 27.7 12.7 OF2 30 7.83 33 3.8 9.6 31.7 14.5 40 7.81 34 5.6 6.8 28.2 15.6 50 7.78 33 7.9 11.2 38.0 7.3 60 7.74 34 4.1 29.6 -13.9 -10 7.05 35 28.3 0.0 20.2 10 7.58 34 20.6 5.3 27.3 -13.6 20 7.7 34 12.6 8.6 33.9 -8.5 OF3 30 7.75 35 11.3 5.1 32.2 -7.8 40 7.89 33 6.6 2.7 33.6 5.4 50 7.84 34 5.4 1.2 37.4 12.3 60 7.83 34 9.1 1.6 36.5 3.6 -10 7.93 32 30.7 0.0 20.0 10 7.4 33 8.9 7.1 31.9 10.8 20 7.44 35 2.4 6.3 29.9 16.1 ON1 30 7.5 35 2.4 5.6 27.8 17.6 40 7.33 35 3.0 4.5 28.1 17.4 50 7.23 35 2.6 2.8 32.6 60 7.01 35 1.5 1.8 -10 7.28 35 29.1 0.0 20.2 10 7.47 35 6.0 8.9 40.4 9.1 20 7.48 35 2.5 2.4 34.4 ON2 30 7.3 35 1.1 4.7 14.7 40 7.21 34 2.1 4.1 40.6 14.6 50 7.16 34 2.2 4.8 38.9 60 7.08 34 2.5 2.2 -10 7.13 34 30.7 0.0 20.5 10 7.61 35 11.6 6.6 31.3 5.9 20 7.66 34 2.5 7.8 40.0 17.4 ON3 30 7.44 35 2.9 5.1 41.4 16.8 40 7.23 35 2.2 3.8 52.6 50 7.07 36 3.6 1.8 43.3 60 7.08 36 3.9 1.7 36.9

104

Figure 5-3: Salinity profiles in (a) “plain” and (b) “oily” sediments near and far from tubeworms.

Pore water methane, sulfate and sulfide concentrations

Mean methane concentrations varied by an order of magnitude between “plain” and “oily” sediments (Table 5-2, Figure 5-4). Mean methane levels were 7-150 µM in

“oily-far” sediments and 33-105 µM in “oily-near” sediments. Mean methane levels were

0.1-0.2 µM in “plain-far” sediments, and 0.2-0.7 µM in “plain-near” sediments.

Figure 5-4: Mean methane concentrations (µM) sampled by 3 replicate peepers at 4 locations.

105

In “oily” sediments, sulfate concentrations decreased sharply with depth and were mirrored by increased sulfide concentrations (Table 5-3, Figure 5-5). There was much greater variability in both sulfate and sulfide concentrations between replicate samples taken in “oily-far” sediments than in “oily-near” sediments. In “plain” sediments, sulfate concentrations decreased with depth, but the lowest pore water sulfate concentration measured was 17 mM (Table 5-2, Figure 5-5). Sulfide levels remained low, and varied between 5-63 µM. In “plain-near” sediments, sulfate values increased with depth at depths greater than 20 cm below the sediment-seawater interface. Mean pore water sulfate levels in “plain-near” sediments were significantly greater than those in “plain- far” sediments (P=0.008; two-sided t-test; df=40).

106

Figure 5-5: Pore water sulfate and sulfide concentration profiles at the four locations sampled. Sulfide concentrations in “plain” sediments were in the micromolar range, reaching a maximum of 63 µM (Table

5-2).

107 Calculated sulfate reduction rates were an order of magnitude greater in “oily” sediments than in “plain-far” sediments (Table 5-4). Sulfate reduction rates of “plain- near” sediments could not be calculated using the described method, as the profile did not fit an exponential decrease function (equation (2)). Sulfate reduction rates had a significant linear relation with mean pore water methane concentrations in “oily-near” sediments (Figure 5-6), whereas the corresponding relation for “oily-far” sediments was not significant.

Table 5-4: Integrated sulfate reduction rates estimated from depth profiles at the four different locations sampled. The best-fit equation (corresponding to equation (1)) for each peeper profile is also shown.

Sediment Rate Mean ± Distance Peeper Best- fit equation type (mmol m-2 day-1) SE OF 1 32.49 e-0.158x + 2.33 133.2 Far OF 2 29.26 e-0.123x + 5.42 72.6 73.6±59.1 OF 3 25.28 e-0.060x + 6.54 15.0 Oily ON 1 31.13 e-0.158x + 3.49 127.8 Near ON 2 32.92 e-0.214x + 3.45 247.8 155.6±81.9 ON 3 29.97 e-0.136x + 4.06 91.2 PF 1 16.69 e-0.024x +14.74 1.8 Far PF 2 10.60 e-0.069x +20.51 8.4 7.4±5.2 PF 3 10.08 e-0.085x + 21.56 12 Plain PN 1 No fit - Near PN 2 No fit - - PN 3 No fit -

108

Figure 5-6: Relation between integrated sulfate reduction rates (SRR) estimated from sulfate depth profiles (Table 5-3) and average pore water methane concentrations in corresponding sediments.

Pore water sulfur stable isotope ratios

Seawater sulfate had a δ34S value of 20.2 ± 0.2 ‰, and pore water sulfate δ34S values were enriched with respect to seawater (Table 5-2, Figure 5-7). The δ34S values of pore water sulfate from “plain” sediments remained between 20-22‰ and became slightly heavier with depth. “Plain-near” sulfate δ34S values were on an average significantly heavier than “plain-far” values (P=0.0014; two-sided t-test; df=28). “Oily” sediment pore water sulfate δ34S values were significantly heavier than “plain” sediment values (P<0.001; two-sided t-test; df=60), and were substantially enriched with respect to seawater sulfate. “Oily-near” pore water sulfate δ34S values were significantly heavier than “oily-far” values (P=0.043; two-sided t-test; df =37). Blood sulfate from tubeworms collected from the “oily” tubeworm aggregation had a δ34S value of 10.9 ± 2.6‰ (n=5).

109

Figure 5-7: Pore water sulfate and sulfide sulfur isotope ratios at the four different locations sampled. The mean ± standard deviation of tubeworm blood sulfate δ34S is also shown. These tubeworms were collected from the aggregation in the “oily” location. Sulfide δ34S values could not be determined for “plain” sediments, due to low sulfide concentrations in “plain” sediment pore water.

110 Pore water sulfide δ34S values could only be obtained for “oily” sediments, as sulfide levels in “plain” sediments were too low to obtain sufficient samples for isotope analyses. There was high variability in “oily-far” sulfide δ34S values, especially between

10-30 cm depths below the sediment-seawater interface (Table 5-3, Figure 5-7). At 10 cm depth in “oily-far” sediments, pore water δ34S ranged between -13.6‰ and +14.3‰, and sulfate-sulfide fractionation ranged between 11.25‰ and 40.9‰. In contrast with “oily- far” sediments, the pore water sulfide δ34S values of “oily-near” sediments had low variability. Between 20-40 cm depths, pore water sulfide δ34S values in “oily-near” sediments ranged between 14.6 to 17.5 ‰ with a mean value of 16.4 % (n=11). δ34S values of sulfide samples from 50-60 cm depths in “oily-near” sediments could not be obtained, presumably due to relatively low sulfur content, and possible contamination of zinc sulfide precipitates with zinc carbonate.

Sulfate diffusion-reaction-supply model

The model was implemented to obtain steady-state values of pore water sulfate and sulfide concentrations and δ34S values, in both the presence and absence of tubeworm influence. Pore water concentration profiles of sulfate and sulfide were obtained at sediment sulfate reduction rates varying by three orders of magnitude, by varying the

“rate fraction” (RF) term from 1 to 0.001 (Figure 5-8). Pore water profiles near tubeworm aggregations in Figure 5-8 were obtained by assuming that tubeworms release 70% of the sulfate produced by sulfide oxidation across their roots. Sulfate depletion with depth at corresponding RF values was lower when root sulfate release was assumed to be 90%,

111 but the shape of the profiles remained similar to those shown in Figure 5-8. Sulfate and sulfide concentration profiles generated by the model closely emulated patterns observed in the empirical data (Figure 5-5) with the notable difference that sulfate levels did not drop to zero in the empirical data.

Figure 5-8: Sulfate and sulfide concentration profiles generated using model simulations at various maximal sulfate reduction rates (Rm). Tubeworm influence was simulated by assuming sulfide uptake by a one-meter tall tubeworm aggregation. Sulfate release across roots was assumed to be 70% of the total sulfide oxidized by the tubeworm.

112

The model was also used to generate sulfate and sulfide δ34S depth profiles at various RF values, by assuming that fractionation during sulfate reduction (Δδ34S) was

15‰ - 30‰ (Figure 5-9). The shape of the isotope profiles changed substantially when tubeworm influence was included in the model. Tubeworm sulfide uptake and sulfate release caused δ34S values of pore water sulfide and sulfate to remain relatively constant with depth, with pore water δ34S values being similar to tubeworm blood sulfate values at all depths below 20 cm. When Δδ34S values of either 15‰ or 30‰ were used, the sulfate

δ34S values produced by the model were substantially heavier than the empirical values.

Also, the sulfide δ34S values at 10-30 cm depths obtained from “oily-far” sediments were substantially lighter than those generated by the model. Our model represented a simplified situation in which we assumed that sulfur fractionation due to microbial sulfate reduction remained constant with depth. However, in sediments where sulfate reduction occurs, microbial disproportionation of sulfur intermediates such as elemental sulfur, thiosulfate and sulfite can significantly affect pore water sulfate δ34S values (Habicht and

Canfield 2001). Disproportionation is more likely to occur at very shallow depths in the sediment near the oxic/anoxic interface where sulfide is oxidized to form intermediate sulfur compounds (Aharon and Fu 2003; Chambers and Trudinger 1978; Johnston et al.

2005), and could explain the above discrepancy.

113

Figure 5-9: Sulfate and sulfide sulfur stable isotope profiles generated using model simulations at various 34 maximal sulfate reduction rates (Rm), and using a Δδ S value of either 15‰ or 30‰, as indicated. Tubeworm influence was simulated by assuming sulfide uptake by a one-meter tall tubeworm aggregation. Sulfate release was assumed to be 70% of the total sulfide oxidized by the symbionts.

114

Discussion

Tubeworm habitat spatial heterogeneity

Empirical pore water sulfide and sulfate concentration profiles indicate a high degree of variability in sulfate reduction rates in sediments both near and far from tubeworms (Figure 5-5; Table 5-3). Pore water sulfide concentrations were between 5-63

µM in “plain” sediments, while levels in “oily-far” sediments were between 5-10 mM

(Table 5-2). While we did not measure the total hydrocarbon content of the sediments, we determined the pore water methane concentrations, which are correlated with fluid flow at seepage sites (Joye et al. 2004; Torres et al. 2002). We found that pore water methane levels were an order of magnitude greater in “oily” sediments than in “plain” sediments

(Figure 5-4). Beggiatoa mats are often associated with high fluid flux, and enhanced microbial sulfate reductions rates in sediments underlying them (Arvidson et al. 2004;

Joye et al. 2004; Torres et al. 2002). Our analysis strongly corroborates this association.

We found that the presence of Beggiatoa mats on the sediment surface was a significant predictor for pore water sulfide levels in the underlying sediment (P < 0.0001; one-way

ANOVA: n=83).

Fluid flux and associated sediment microbial sulfate reduction rates at hydrocarbon seeps can vary by orders of magnitude at sites separated by just a few meters (Torres et al. 2002; Treude et al. 2003). Studies at seepage sites in the Gulf of

Mexico have revealed high spatial variability in pore water sulfate depletion (Aharon and

Fu 2003; Arvidson et al. 2004; Joye et al. 2004). In our study, we did not directly measure microbial sulfate reduction rates using tracers, but estimated rates using the

115 diffusion-advection-reaction model described by Aharon and Fu (2000). We found that apparent sulfate reduction rates can vary by an order of magnitude in sediments surrounding tubeworm aggregations that were separated by less than 30 meters (Table 5-

2).

In situ evidence for tubeworm sulfide uptake and sulfate release

Since seawater sulfate concentration is approximately 29 mM, and microbial sulfate reduction produces one mole of sulfide for every mole of sulfate consumed

(Aharon and Fu 2000), the sulfide concentration at a given depth in the pore water should be 30 minus the sulfate concentration. In the absence of processes removing sulfide

2- produced by microbial sulfate reduction, the ratio [H2S]/(29-[SO4 ]) should have a value of 1. However, in hydrocarbon-rich sediments in the Gulf of Mexico, processes such as biological uptake or sulfide mineralization can remove up to 30% of the sulfide produced

2- by microbial sulfate reduction, yielding [H2S]/(29-[SO4 ]) values less than 1 (Aharon and

2- Fu 2000). In this study, we found that in “oily” sediments, the empirical [H2S]/(29-[SO4

]) ratio was remarkably constant near tubeworms and remained between 0.1- 0.2 at all

2- depths ≥ 20 cm. However, far from tubeworms, the [H2S]/(29-[SO4 ]) ratios showed high variability with values ranging between 0.05 - 0.6. To investigate whether tubeworm sulfide uptake and sulfate release could generate these patterns, we used the model to

2- simulate [H2S]/(29-[SO4 ]) ratios and varied the percent sulfate released across the roots between 0-90% in different simulations. We found that in sediments near tubeworms the empirical data matched closely with predictions made by the model when the percentage of sulfate released from the roots was close to 80% (Figure 5-10). Moreover, we

116 2- calculated the [H2S]/(29-[SO4 ]) ratios from values collected using sediment cores in two independent studies (Aharon and Fu 2003; Arvidson et al. 2004). We found that at depths greater than 10 cm, almost all values reported by these studies near tubeworm aggregations fell within the range predicted by 70-90% sulfate release from tubeworm

2- roots (Figure 5-10). Again, in cores collected far from tubeworms, the [H2S]/(29-[SO4 ]) ratios were highly variable and frequently greater than 0.5 (Figure 5-10). Moreover, at 0-

10 cm depths near tubeworms, the data deviated from the model. Roots of mature tubeworm aggregations extend below a carbonate substrate, that may be buried by several centimeters of sediment (Bergquist et al. 2002; Fisher et al. 1997). It is possible that tubeworm sulfide uptake and sulfate release only impacts the geochemistry of pore water deeper than 10 cm below the sediment-seawater interface. Moreover, oxidation of pore water sulfide, which was not included in the model, could play an important role at sediment depths less than 10 cm.

Our analyses also indicate that tubeworms have a significant impact on the sulfur isotope geochemistry of sediments surrounding their roots. The model predicts that both sulfate and sulfide δ34S values near tubeworm aggregations would remain constant at depths below 20 cm, and the tubeworm blood sulfate δ34S value would be similar to the pore water sulfide value (Figure 5-9). Our data from “oily” sediments matched the model predictions well (Figures 5-7 and 5-9). The pore water δ34S value was constant with depth and had very low variability between replicate samples. The tubeworm blood sulfate δ34S was similar to pore water sulfide values, but was lighter than values predicted by the model, suggesting that sulfur isotope discrimination terms used in the model were slight underestimates of true values.

117

Figure 5-10: Sulfide-to-sulfate concentration ratios at various depths, plotted along with ratios predicted using the model. The various lines show results from simulations run either in the absence of tubeworm influence, or with tubeworm root sulfate release varying between 0-90 % of the total sulfide oxidized by their symbionts. Data in the top two profiles were collected in this study. The bottom two profiles were plotted using data from cores collected near and far from tubeworm aggregations at the Gulf of Mexico (Aharon and Fu 2003; Arvidson et al. 2004). Previous to this study, there was no data available from deeper than 25 cm.

118

In “oily-near” sediments, we found a significant linear relation between the ratio of sulfide δ34S values to sulfide concentrations and depth below the sediment-seawater interface (P=0.0003; R2=0.93; n=7). This relation might arise because tubeworms contribute to a constant sulfide δ34S as well as specific sulfide concentrations at various

34 depths in sediments near their roots. We used the model to generate H2S-δ S/[H2S] values at various levels of root sulfate release from tubeworms, ranging from 0% to

100% (Figure 5-12). We found that our empirical data matched the model best when the percent of sulfate released by tubeworms across their roots was 83%, remarkably close to the 85% value measured using laboratory experiments with live L. luymesi individuals

(Dattagupta et al. 2006a).

Figure 5-11: The ratio between sulfide stable isotope values and sulfide concentrations plotted against depth, “near” (< 30 cm away) or “far” (> 150 cm away) from tubeworms. The various lines show ratios predicted using simulations from the model, where sulfate released across tubeworm roots was varied between 0-100%. The best match between the empirical data (collected in 2004) and the model is found at 83% root sulfate release. All data shown in this figure were collected in this study.

119 Tubeworm impact on sediment pH

Experiments with live L. luymesi individuals indicate that they release protons produced by metabolic processes such as autotrophic sulfide oxidation and carbon fixation primarily across their roots (Dattagupta et al. 2006a). In this study, we found that the pH was significantly lower in “oily-near” sediments than in “oily-far” sediments

(Figure 5-2), suggesting that proton release across tubeworm roots does occur in situ. In

“plain” sediments where apparent microbial sulfate reduction rates were low, pH near and far from tubeworms was similar, albeit slightly lower near tubeworms at depths below 30 cm.

Can tubeworms enhance microbial sulfate reduction rates?

Our in situ empirical data, combined with the results of the geochemical model, corroborate laboratory studies that indicate that tubeworms release 70-90% of the sulfate produced during sulfide oxidation by their symbionts across their roots into the surrounding sediment. Microbial sulfate reduction in organic rich sediments can be limited by sulfate availability, and tubeworms might significantly enhance sulfate reduction near their roots by supplying sulfate across their root surface into the surrounding sediment (Cordes et al. 2005a; Dattagupta et al. 2006a).

At hydrocarbon seep sites, pore water methane concentrations are correlated with seepage rates (Torres et al. 2002). In this study, we found high pore water methane concentrations in “oily” sediments, indicating that these locations are hydrocarbon-rich.

Moreover, the relation between sulfate reduction rates and pore water methane concentrations was highly significant in “oily” sediments near tubeworms (Figure 5-6).

120 Far from tubeworms, sulfate reduction rates were poorly correlated with pore water methane concentrations, and were in general lower than in sediments near tubeworms.

For example, at a pore water methane concentration of 170 µM, sulfate reduction rate far from tubeworms was about half of the rate measured in sediments near tubeworms containing only 105 µM methane. These data suggest that microbial sulfate reduction in

“oily” sediments far from tubeworms is sulfate-limited, whereas sulfate supply from tubeworm roots might prevent sulfate-limitation in sediments near tubeworms.

In “plain” sediments, pore water methane concentrations were three orders of magnitude lower than values found in “oily” sediments (Figure 5-4), suggesting that these sediments might be low in hydrocarbon content. Sulfate was not substantially depleted with depth at these locations, and reached a minimum value of 17 mM even at

60 cm depth. Since dissolved sulfate concentrations affect microbial sulfate reduction rates only when they fall below 3 mM or even lower (Boudreau and Westrich 1984;

Luley 2006), it is possible that microbial sulfate reduction in these sediments was hydrocarbon-limited rather than sulfate-limited. In “plain” sediments near tubeworms, sulfate reduction rates appeared to be extremely low, typical of reference sediments at the

Gulf of Mexico that are not impacted by hydrocarbon seepage (Aharon and Fu 2000).

However, depth profiles generated by our model indicate that the combination of sulfide uptake and sulfate release by tubeworms could lead to underestimation of the true sulfate reduction rate using theoretical methods based on sulfate depth profiles alone (Figure 5-

8). By comparing the depth profiles generated by our model with the empirical profiles obtained in “plain-near” sediments, we can estimate that true rates of sulfate reduction in these sediments are more likely in the order of 5 mmoles m-2 day-1, about double the rates

121 measured in reference sediments. At these rates, our model indicates that 70-90% sulfate release from tubeworm roots could enhance sulfide availability by about 4%.

How does variability in hydrocarbon flux affect tubeworm metabolism and growth?

Authigenic carbonates typically serve as substrates for tubeworm settlement

(Bergquist et al. 2002; Fisher et al. 1997), and are associated with areas of prolonged and extensive hydrocarbon seepage (Roberts and Aharon 1994; Sassen et al. 1994).

Tubeworm settlement is believed to occur on exposed carbonate over a period of a few decades and in areas with active surface sulfide expression (Bergquist et al. 2003; Cordes et al. 2003). Fluid flow, that influences migration of hydrocarbons, can vary over short time periods at seepage sites occupied by tubeworm aggregations (Levin 2005; Tyron and

Brown 2004). Therefore, although tubeworm larvae might always settle in areas of high sulfide flux, micro-seepage patterns are likely to change over the 170-250 year long tubeworm lifespan (Bergquist et al. 2000; Fisher et al. 1997).

In this study, we collected pore water samples adjacent to tubeworms surrounded by sediments impacted by both high and low hydrocarbon flux (“oily” and “plain” sediments, respectively). These might represent two extremes of a spectrum of circumstances experienced by tubeworms over their lifespan. By comparing the empirical sulfate and sulfide depth profiles with those generated by our model (Figures 5-5 and 5-

8), we can approximate that sulfate reduction rates in “oily” sediments were at least 1000 times higher than those in “plain” sediments. If mature tubeworms experience such a wide range of sulfide availability during their lifespan, how would that affect their growth and metabolism? This is a complex question that would require extensive temporal and

122 spatial sampling, which we did not achieve during this study. However, we can begin to address this question using simulations from our model.

We used our model to estimate individual growth that can be achieved by a one- meter tall L. luymesi individual with a one-meter long root, in sediments with integrated sulfate reduction rates varying between 2.5 to 5000 mmoles m-2 day-1 (Figure 5-12).

Since we did not consider metabolic processes such as reproduction and tissue repair in the model, these growth estimates represent maximum values. According to the model, at integrated sulfate reduction rates of 2.5 mmoles m-2 day-1, tubeworms fall below their basal sulfide requirements in the absence of root sulfate release, whereas they can achieve basal requirements if they release 70-90% of the sulfate produced by their symbionts.

However, they cannot grow under these conditions even if they release sulfate from their roots. Integrated sulfate reduction rates of 2.5 mmoles m-2 day-1 have been measured in sediments with low hydrocarbon flux at sites containing tubeworm aggregations in the

Gulf of Mexico (Arvidson et al. 2004; Joye et al. 2004), but might not reflect conditions in the immediate vicinity of tubeworm roots.

At integrated sulfate reduction rates of 5 mmoles m-2 day-1 and above, tubeworms can maintain basal sulfide requirements for short periods up to 25 years even in the absence of root sulfate release. However, root sulfate release allows them to achieve much greater growth rates. At an integrated sulfate reduction rate of 50 mmoles m-2 day-1, similar to the average value found at active seep sites in the Gulf of Mexico (Joye et al.

2004), a one-meter tall tubeworm can achieve a maximum average growth of 2.64 cm yr-1 if it releases 70-90% of its sulfate across its roots. This value is remarkably similar to the

123 maximum growth rate of 2.45 cm yr-1 for a one-meter tubeworm calculated using an empirical growth model for L. luymesi tubeworms (Cordes et al. 2005a).

Contrary to expectations, the model also predicts that extremely high sulfate reduction rates do not result in the highest tubeworm growth rates. At rates of 5000 mmoles m-2 day-1, the maximum growth when tubeworms released 90% of the sulfate produced by their symbionts was estimated to be 2.46 cm yr-1, lower than the growth predicted at 50 mmoles m-2 day-1. This might be because an aggregation containing one- meter tall tubeworms cannot release sufficient sulfate to sustain extremely high microbial sulfate reduction rates over extended time periods.

Our model presents a simplified situation in which there is equal resource partitioning between individual tubeworms in an aggregation. However, there is likely to be high variability within individuals of an aggregation, leading to the variability in individual growth rates that has been observed within individuals of the same aggregation

(Bergquist et al. 2002; Fisher et al. 1997). Based on our model, we predict that tubeworm aggregations of the same size might have high variability in growth rates when exposed to different regimes of hydrocarbon flux.

124

Figure 5-12: Maximum individual growth (cm yr-1) predicted by the model for a one-meter long tubeworm, at various integrated sulfate reduction rates, as well as various levels of root sulfate release.

Acknowledgements

We thank the captain and crew of the Research Vessel (RV) Seaward Johnson II as well as the crew and pilots of the Johnson Sea Link submersible (Harbor Branch

Oceanographic Institution). We are indebted to Dr. Edith Widder and Dr. Tamara Frank for devoting time and effort on their research cruise to facilitate collection of the peepers.

We would also like to thank Burt Thomas for conducting the methane analyses. The

Minerals Management Service, Gulf of Mexico Regional OCS office, through contract number 1435-01-96-CT30813, the NOAA National Undersea Research Program at the

University of North Carolina, Wilmington, the NOAA Ocean Exploration Program, and the National Science Foundation grant OCE 0117050 supported this work.

125 CHAPTER 6

Sulfate and proton excretion by hydrocarbon seep tubeworms: a summary

Complex interactions between biogeochemical processes and macro- and micro- fauna are a common theme in hydrocarbon seep ecosystems (Boetius and Suess 2004;

Levin 2005; Treude et al. 2003). At the onset of this study, there were speculations that

Lamellibrachia luymesi, a dominant species of tubeworm found in hydrocarbon seeps in the Gulf of Mexico, was capable of enhancing microbial sulfate reduction in sediments surrounding its aggregations (Freytag et al. 2001; Julian et al. 1999). The idea proposed was that L. luymesi could release sulfate, which is a byproduct of sulfide oxidation by its symbionts, through its roots into sediment. In this study, we used a multi-disciplinary approach to investigate this hypothesis.

The proposition that L. luymesi could supply sulfate across its roots hinged upon the presence of a viable mechanism for sulfate transport across its root epithelium. We conducted experiments with live L. luymesi and determined that they eliminate a majority of the sulfate produced by sulfide oxidation across their roots (Chapter 2). Using inhibitor studies, we deduced that they might have sulfate-bicarbonate antiporters in their root epithelium, which mediate sulfate transport. This study provided the first empirical evidence to substantiate the fact that L. luymesi can transport sulfate across its roots and provided a starting point for further studies into the nature of the transport mechanism.

In addition, the live animal experiments demonstrated that L. luymesi use their roots for proton elimination. Both sulfate and proton elimination across roots can occur in the direction of the respective electrochemical gradients of these ions. This has significant implications for L. luymesi’s physiology. While the closely related

126 vestimentiferan Riftia pachyptila expends a substantial amount of energy for elimination of waste ions (Girguis et al. 2002; Goffredi et al. 1999a), L. luymesi can conserve energy by eliminating these ions using passive membrane transport. R. pachyptila uses a high concentration of proton-ATPases in its plume tissue to mediate proton excretion

(Goffredi and Childress 2001). We measured the proton-ATPase activity of L. luymesi tissues and found that it was approximately ten times lower than that of R. pachyptila

(Chapter 3). This finding is consistent with our conjecture that L. luymesi can use passive transport for proton excretion.

The approaches we used demonstrated that sulfate and proton release across L. luymesi roots occurs under experimental conditions. However, two questions remained.

First, do L. luymesi eliminate sulfate and protons across their roots in their natural habitat? If so, does this enhance microbial sulfate reduction around their roots? To address these questions, we measured several of the in situ geochemical characteristics of the tubeworm habitat. A central objective was to collect pore water samples from sediments surrounding tubeworm roots, which can extend up to a meter deep in the sediment underlying aggregations. However, existing methodology for collecting sediments allowed characterization of pore water only up to 25 cm depths. Therefore, we developed new sampling devices that allowed collection of high-quality pore water samples up to 60 cm deep in the sediments (Chapter 4).

Using these devices, we characterized the pH, sulfate, sulfide, and methane concentrations, as well as sulfur stable isotopes of pore water surrounding tubeworm roots. To better understand this complex data set, we developed a geochemical model to simulate the effects of tubeworm sulfate release on concentration and isotope profiles of

127 pore water sulfate and sulfide. Using this combination of approaches, we found convincing evidence that tubeworms can enhance sulfate reduction rates in sediments by supplying microbes with sulfate across their roots. Our study also showed that tubeworms significantly impact the pH of sediments, presumably through proton release across their roots.

Implications and future directions

Our study indicates that tubeworm aggregations have a significant impact on their habitat geochemistry, and are capable of recycling sulfur using their roots. L. luymesi tubeworms take up sulfide, and release sulfate and protons across their roots. A combination of these processes can greatly enhance sulfate reduction rates in sediments surrounding their aggregations, and ensure them a sufficient sulfide supply over their long lifespan.

Our study focused on L. luymesi, the dominant tubeworm species in aggregations on the upper Louisiana slope (Bergquist et al. 2002). However, L. luymesi almost always co-occurs with another tubeworm, Seepiophila jonesi. In chapter 3, we compared the

ATPase activities of the two tubeworm species and found some interesting differences between the activities in their plumes. Previous studies have also suggested resource partitioning between these two seep species (Freytag et al. 2003), but conclusive evidence for this phenomenon remains elusive. Future studies that focus on S. jonesi physiology are required to obtain a better understanding of possible interactions between the two tubeworm species.

128 Several other questions regarding seep tubeworm physiology remain unanswered.

For example, mechanisms for carbon acquisition by L. luymesi and S. jonesi have never been investigated. Based on findings from our live animal experiments and ATPase activity determinations, we suggest that S. jonesi might take up carbon dioxide by diffusion across its plume, whereas L. luymesi might acquire bicarbonate ions across their roots. Future studies that measure carbonic anhydrase activities in plume tissues of the two species, or directly measure carbon uptake across the plume and root surfaces are needed to investigate this possibility.

Our study also suggests that tubeworm aggregations might have a significant impact on the microbes in sediments surrounding their roots. A recent study has investigated microbial communities in sediments collected near tubeworm aggregations

(Luley 2006). More detailed studies that investigate microbial communities in the direct vicinity of tubeworm roots, and perhaps even microbes growing directly on the chitin tubes surrounding the roots could prove extremely interesting.

129 APPENDIX A

Attempts to patch-clamp live bacteriocytes isolated from Lamellibrachia luymesi

Purpose

This thesis was focused towards identifying the location and mechanism of sulfate and hydrogen ion elimination from seep tubeworms. This appendix describes attempts made to identify mechanisms of membrane transport of ions across L. luymesi cells using an electrophysiological approach. This project was not taken to completion; however, methodological progress made in this study is presented here, and might be beneficial as a starting point for future work in this direction.

Background

Sulfate and hydrogen ions are the major byproducts of sulfide oxidation by L. luymesi symbionts (Dattagupta et al. 2006a). L. luymesi symbionts are located inside cells called bacteriocytes, within an organ called the trophosome (Gardiner and Jones 1993). The trophosome can comprise up to15-30% of a tubeworm’s body weight (Childress et al. 1984) and is composed of several well-vascularized lobules (Felbeck and Jarchow 1998). These lobules contain four cell types, the majority of which are bacteriocytes (De Cian et al. 2003a).

Bacteriocytes contain vacuoles within which one or more symbiotic bacteria are contained

(Felbeck and Jarchow 1998). Thus all symbiont driven metabolism occurs within bacteriocytes, and transfer of nutrients and metabolites across the bacteriocyte cell membrane is likely to be extremely crucial for the sustenance of the tubeworm symbiosis.

130 The host tubeworm provides symbionts with oxygen, hydrogen sulfide, nitrate and inorganic carbon (Childress and Fisher 1992). The symbionts in turn release carbon and nitrogen containing substrates (such as succinate and glutamate (Felbeck and Jarchow 1998), and also sulfate and protons, which are byproducts of sulfide oxidation (Girguis et al. 2002; Goffredi et al.

1999a) . Except for one study (De Cian et al. 2003b), transport of nutrients to and from the symbionts has been studied either at the level of the whole tubeworm, or at the level of isolated symbionts. Using isolated bacteriocytes, deCian et al (2003b) showed that bacteriocytes contain both membrane-attached as well as cytosolic carbonic anhydrase isoforms. They also showed that the bacteriocyte membrane is likely to contain proton-driven sodium-ATPases as well as an anion transporter that exchanges intracellular chloride with external anions. These studies were performed by analyzing internal ionic concentrations after exposing isolated bacteriocytes to various types of membrane transport inhibitors.

Here we attempted to use patch clamping, a powerful electrophysiological tool, to directly analyze membrane transport across L. luymesi bacteriocytes. This technique could not be applied to R. pachyptila bacteriocytes, as their membranes are fragile unless kept under high pressures

(M-C. de Cian, personal communication). However, L. luymesi can be maintained alive in the laboratory under atmospheric pressures (Chapter 2). Thus, bacteriocytes obtained from these tubeworms might be amenable for patch-clamp studies, provided they can be isolated from the tubeworm and maintained live for the duration of the experiment.

The patch-clamp technique was developed in the late 1970s and has since been used extensively to study plant and animal membrane electrophysiology (Hille 1992; Ward 1997). By this method, activities of individual transmembrane proteins as well as the whole ionic current flowing through a cell membrane can be measured (Carpaneto et al. 2003). In this technique,

131 cells are suspended in a solution of known composition (“bath solution”) and a glass microelectrode (“pipette”) with a tip diameter of about 1 micrometer is pressed against the cell membrane (Cahalan and Neher 1992). Gentle suction is applied to the interior of the pipette, which draws a small portion of the cell membrane into the pipette. Under the right conditions, a seal with resistance in the order of 109 Ω (“gigaohm seal”) is formed between the membrane and the pipette glass, as a result of which the interior of the pipette is almost completely electrically isolated from the bath solution.

In this study, we attempted to analyze L. luymesi bacteriocytes using the “whole-cell” mode.

Among the various configurations under which patch-clamp experiments can be performed

(Cahalan and Neher 1992), the “whole-cell” mode is the most common (Ward 1997). In this method, once the gigaohm seal is obtained, applying a strong suction or a voltage pulse of several hundred millivolts (Cahalan and Neher 1992) ruptures the membrane patch at the tip of the pipette. This causes the pipette solution to equilibrate with the cytosol of the cell, and allows the cell membrane potential to be “clamped” to that of the pipette potential (Cahalan and Neher

1992). Under this configuration, currents flowing across the entire cell membrane of an individual cell can be measured (Ward 1997). In such “voltage-clamped” experiments, membrane currents are measured at several different voltage conditions, and the voltage- dependence of the current is determined from the relation between the magnitude of the current

(I) and the voltage (V).

The I-V relation can be used to determine the “reversal” potential of the current, i.e. the potential at which there is a reversal in the sign of the current (Ward 1997). The ions responsible for the measured currents can be determined on the basis of the Nernst potentials of the

132 membrane permeant ions in bath and pipette solutions. The Nernst potential of an ion can be determined using equation

RT ! [X] $ ! $ out Erev = # 2.303 & ' log# & " mF % " [X]in %

Here, Erev is the reversal potential, R is the gas constant, F is the Farraday’s constant, T is the absolute temperature, m is the valence of the ion, and [X] out/[X] in is the ratio of concentrations of the ion X outside and inside the cell. At the reversal potential, the ion in question is under electrochemical equilibrium, and the net flux of this ion across the membrane is zero. Thus, the reversal potential of a current measured using the patch clamp technique should be close to the

Nernst potential of the major ion responsible for the current. A modification of the above equation is used in case the current is due a membrane channel that is permeable to several different ions to varying extents (e.g. (Frachisse et al. 1999). Reversal potentials of voltage- or time- dependent currents are often determined using “tail-current” analyses (Ward 1997).

Methodology

Isolation of bacteriocytes

Bacteriocytes were isolated from 5-10 cm long live L. luymesi using the procedure described

(De Cian et al. 2003b) with a few modifications. Briefly, the worms were dissected and the trophosome was removed. It was rinsed 2-3 times in ice-cold Riftia saline solution (Fisher et al.

1988), and then chopped into several pieces. The cells were dissociated using gentle agitation using a magnetic stirrer for about 20 minutes at 6-8°C. The cell suspension was filtered through a

63 µm sieve to remove large particles. The filtrate was then be centrifuged at 150 g for 5 minutes

(4°C), after which the cell pellet was re-suspended in 5-10 ml of cold Riftia saline solution.

133

Microscopic studies and FISH with isolated bacteriocytes

A mixture of cells was isolated using the procedure outlined above. This mixture was subjected to microscopic studies to positively distinguish bacteriocytes from other cell types.

First, cells were visualized using DNA-specific dyes such as DAPI (4,6-diammino-2- phenylindole), to reveal the cellular nucleus as well as the bacteria within the cells. Second, since

L. luymesi symbionts belong to the phylogenetic group gamma-proteobacteria (Mcmullin et al.

2003), FISH (Fluorescent in-situ hybridization) was performed using a general bacterial probe

(Eub338) to positively identify symbiont containing cells. A general archaeal probe (Ar915) was used as a negative control. The cell suspension was diluted 1:200 using phosphate buffer saline

(PBS). 10 µl of the diluted suspension was added to a glass slide, and the slide was air-dried. The cells on the slide were then desiccated in a series of ethanol solutions (50%, 80% and 95% ethanol made in PBS). Following this, 15 µl of hybridization buffer and 1 µl of the appropriate probe were added to the slide (Orphan et al. 2001). The slides were then allowed to hybridize in a water bath maintained at 48˚C for 90 minutes, following which they were soaked in wash buffer, maintained at 50˚C for 15 minutes. They were then soaked in distilled deionized water

(DDI water) for 2 minutes at room temperature, before they were viewed using a fluorescent microscope equipped with an FITC filter (for the eubacterial probe) or a TRITC filter (for the archaeal probe).

Survival of bacteriocytes

Since patch-clamp studies require live and intact cells, bacteriocyte survival was monitored under conditions conducive for patch clamping. Cell survival was monitored using flow

134 cytometry, where the propidium iodide (PI) dye was used to distinguish between live and dead cells (De Cian et al. 2003b). Live cells exclude PI, but PI can pass through membranes of dead cells and bind to DNA (Nicoletti et al. 1991). Before introducing the bacteriocytes to the flow cytometer, 10 µl of propidium iodide solution (1mg/ml) was added to 1 ml bacteriocyte cell suspension. Cells were maintained under conditions they would be exposed to during patch- clamp studies, for example, at room temperature, and in solutions with low ionic concentrations

(composition mentioned below), while they were monitored for survival.

Preliminary patch-clamp studies

Several attempts were made to patch-clamp bacteriocyte cells, but seal formation was unsuccessful under most conditions. The conditions that were most conducive to seal formation were similar to conditions used to patch-clamp cells isolated from a marine sponge (Carpaneto et al. 2003). Briefly, the bath (outside) solution contained 460 mM KCl, 60 mM MgCl2. 6H2O, and

20 mM HEPES (pH adjusted to 7.2 using KOH). The pipette (inside) solution contained 128 mM

KCl, 4.3 mM MgCl2, 850 µM EGTA and 20 mM HEPES (pH 7.2). The osmomolarity of both solutions was adjusted to 1080 mmol kg-1 using 1M sorbitol (Taylor and Brownlee 2003).

Bacteriocytes were suspended in 500 µl outside solution. The cell suspension was filtered using a

10-micron mesh, after which 10 µl of the cell suspension was viewed under a 40X magnification using an inverted light microscope. Addition of 10 µl of 300 µM LaCl3 caused an increase in pipette seal resistance from 6.2 MΩ to 2 GΩ. Currents were measured in the “whole-cell” mode, using an Axopatch-200A amplifier (Axon Instruments, Foster City, CA), which was connected to a microcomputer via an interface (Digidata 1200 Interface, Axon Instruments). pCLAMP

(Axon Instruments) software was used to acquire and analyze the whole-cell currents.

135 Bacteriocyte

Symbionts

DAPI stained symbiont cells

Myelin body

Figure A-A-1: A mixture of cells isolated from L. luymesi trophosome viewed under bright field (left) and viewed by DAPI staining (right). Putative bacteriocytes containing symbionts are indicated. Some cells appeared to contain myelin structures, indicative of degenerating bacteriocytes in which symbionts have been digested (Dr. Monika Bright, personal communication).

136 Results and Discussion

Isolation and identification of bacteriocytes

Bacteriocytes were successfully isolated using the procedures described. However, a mixture of cells was obtained in all cases, containing bacteriocytes as well as other types of trophosome cells (De Cian et al. 2003b). Cells were visualized using bright field light microscopy, and bacteriocytes were stained using DAPI. The stain appeared to bind to putative symbiont cells within the bacteriocytes (Figure A-1). Moreover, the FISH eubacterial probe bound to the putative symbiont cells, whereas the archaeal probe did not (Figure A-2). Based on these results, it is safe to assume that the cells we isolated were most likely bacteriocytes. These cells were about 7 µm in diameter, which is smaller than the typical bacteriocyte cell size reported previously (De Cian et al. 2003b). It is possible that larger cells did not survive the purification process. Trophosome smears viewed under a light microscope indicated that bacteriocytes as large as 20 µm in diameter were present in the trophosomes of the tubeworms maintained in the aquaria.

Bacteriocyte survival

Bacteriocytes appeared to be robust, and survived for up to 10 hours at room temperature when exposed to either Riftia saline, or patch clamp bath and pipette solutions (Figure A-3).

After 10 hours, cells appeared to die when kept at room temperature, whereas they survived for up to 30 hours when stored on ice. It appears that L. luymesi bacteriocytes can survive for extended periods in vitro and might be suitable for patch clamp studies.

137 (a)

(b)

Figure A-A-2: FISH using bacteriocyte cell preparations. Panels on the left show binding of FISH probes using fluorescent filters, whereas panels on the right show bright field views of the same area of the slide. Scale bars represent 10 µm. (a) Binding of general eubacterial probe to symbiont cells, viewed using an FITC filter; (b) Negative control showing lack of binding of the general archaeal probe to cells.

Figure A-A-3: Percentage survival of bacteriocytes exposed to solutions of different compositions, and kept either on ice, or at room temperature.

138 Preliminary patch clamp findings

The bacteriocytes isolated from L. luymesi were small (~ 7 µm in diameter), and thus difficult to patch-clamp. It appeared that an extracellular matrix surrounding the cells hindered proper seal formation. Attempts to improve seal formation using collagenase treatment of bacteriocytes proved unsuccessful. However, addition of lanthanum ions had a dramatic effect on seal formation, as noted previously on cells isolated from a marine sponge (Carpaneto et al.

2003).

Preliminary current measurements were made in the “whole-cell” configuration (Figure

A-4). Based on the composition of the bath and pipette solutions, the reversal potentials for chloride and potassium ions were -36 mV and +32 mV, respectively. The currents we measured appear to be based on movement on chloride ions, though further studies need to be conducted to affirm this possibility.

1000

800

600 I (mAmp) 400

200 I (mAmp)

0 -100 -60 -20 20 60 100 140 180 Time (sec) -200 V (mV)

Figure A-A-4: Preliminary currents obtained using the “whole-cell” mode from patch clamping isolated L. luymesi bacteriocytes. The pipette solution contained 128 mM K+, 0 mM Na+, and 136.6 mM Cl-, whereas the bath solution contained 460 mM K+, 0 mM Na+, and 580mM Cl-.

139 Conclusions

Preliminary studies show that bacteriocytes are relatively easy to obtain from L. luymesi and that they remain alive in vitro for periods up to 10 hours at room temperature.

Electrophysiology of bacteriocytes could be interesting, as the bacteriocyte membrane is the main location where transfer of nutrients between host tubeworms and symbionts occurs. It appears that bacteriocytes isolated from L. luymesi might be amenable to patch-clamping studies, which might provide crucial information about nutrient transfer across the bacteriocyte membrane.

Acknowledgements

We are indebted to Dr. Laetitia Perfus and Dr. Sally Assmann for help and guidance with this project. We would also like to thank Dr. Chris House for allowing use of his lab facility for performing FISH.

140 APPENDIX B

Deep-sea hydrocarbon seep gastropod Bathynerita naticoidea responds to cues from the habitat providing mussel Bathymodiolus childressi

Sharmishtha Dattagupta*1, Jonathan Martin1, Shu-min Liao2, Robert S. Carney3 and Charles R. Fisher1

Submitted as a Short Communication to Marine Ecology (date submitted 6/6/2006)

1 Department of Biology, The Pennsylvania State University, University, PA 16802, USA 2 Department of Statistics, The Pennsylvania State University, University, PA 16802, USA 3 Department of Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA 70803, USA

141 Abstract

Bathynerita naticoidea (Gastropoda: Neritidae) is a numerically dominant heterotrophic gastropod found at hydrocarbon seep sites on the upper Louisiana slope of the Gulf of Mexico.

Snails of this species are commonly associated with beds of the methanotrophic mussel

Bathymodiolus childressi (Bivalvia: Mytilidae), and their population structure mirrors that of the mussels they are found among. Previous studies have shown that these snails feed on bacteria and decomposing periostracum on the B. childressi shell. In this study, we used a flow-through

Y-maze system to investigate the behavior of B. naticoidea exposed to cues associated with B. childressi mussel beds. We found that the nerite is not attracted to methane, but is strongly attracted to seawater conditioned with B. childressi. The attractant appears to be specific to this type of mussel, and is not a soluble cue produced by conspecific snails.

142 Problem

Chemoreception is widely documented in shallow marine ecosystems (Zimmer and

Butman 2000), where it is used by a variety of organisms for purposes such as food detection

(Salierno et al. 2003), predator avoidance (Rahman et al. 2000), and coordination of larval release and settlement (Krug and Manzi 1999). Though chemical communication is likely to be pervasive in the deep-sea (Herring 2003), it is poorly studied due to difficulties of working in this habitat. Deep-sea cold seep communities contain high densities of organisms (Levin 2005;

Sibuet and Olu 1998), and exist as spatially isolated patches in a larger environment lacking sufficient reduced chemicals to support them (Brooks et al. 1987; Carney 1994). In these ecosystems, chemical cues may be crucial for colonization and persistence by the largely endemic fauna (Carney 1994).

The Gulf of Mexico seafloor has multiple locations at 400- 3000 m depths where hydrocarbon seepage supports extensive populations of chemosynthetic organisms (Brooks et al.

1987; Macdonald 1998; Macdonald et al. 1990a). The mussel B. childressi, which derives most of its nutrition from methanotrophic endosymbionts in its gills (Cary et al. 1988; Childress and

Fisher 1992; Childress et al. 1986; Streams et al. 1997), is a prominent foundation species of these communities between 500 to 2,200m depth. Beds of these mussels can be extensive

(Macdonald et al. 1990b) and provide habitat for several species of heterotrophic fauna

(Bergquist et al. 2005). The gastropod B. naticoidea is often the numerically dominant heterotroph in these mussel bed communities, along a wide geographic range (Bergquist et al.

2005), (Carney 1994). This nerite grazes on the surface of B. childressi shells, ingesting free- living bacteria that are abundant on the shell surface, as well as fragments of mussel periostracum and attached byssal fibers (Zande and Carney 2001). The size structure of B.

143 naticoidea populations is often similar to that of the mussels in the mussel bed where they occur

(Zande and Carney 2001), suggesting a connection between factors influencing the snail and mussel populations. Moreover, fossil records of a Miocene-age cold seep show a neritic species similar to B. naticoidea found with a Bathymodiolus-like methanotrophic mussel (Taviani 1994), further demonstrating a strong association between these taxa. In light of B. naticoidea ’s strong fidelity towards B. childressi, we predicted that the nerite might recognize chemical cues associated with this mussel species, enabling it to locate its preferred habitat. Therefore, we analyzed the behavior of live nerites exposed to potential chemical and biogenic cues using flow- through Y-maze experiments.

Materials and Methods

We collected individuals of B. naticoidea and B. childressi using the Johnson Sea-

Link submersible from Brine Pool NR-1 (27°43’24’’N, 91°16’30”W; ~ 650m depth), a hydrocarbon seep site located on the upper Louisiana slope of the Gulf of Mexico. We transported them alive to the surface in a temperature-insulated box and maintained them on board the ship in buckets of chilled, well-aerated seawater. We changed the seawater in the buckets daily and bubbled it twice daily with methane to “feed” the mussels. We brought the mussels and snails back alive to the laboratory at Pennsylvania State University and maintained them together in an aquarium at 6°C and under atmospheric pressure. For routine aquarium maintenance as well as for experimental treatments, we used synthetic seawater (SSW) made using Reef Crystals (Aquarium Systems Inc.). We bubbled the SSW in the maintenance aquaria with methane four times a day, and filtered and aerated it using a flow-through filtration system.

We replaced approximately 10% of SSW in the aquaria with freshly made SSW once a week.

144 For control experiments with Mytilus edulis, we used mussels purchased from a local seafood market.

We observed snail behavior in a cold room (6°C) using a flow-through Y-maze system.

The mazes consisted of glass Y-shaped tubes (with three 9 cm long arms, and1.6 cm inner diameter), through which a flow of SSW was maintained at a rate of about 15 ml/min using a peristaltic pump. Two arms of the Y-maze received flow from separate 5-gallon reservoirs, one of which contained a test cue, and the other a control cue. We used dye tests to confirm that there was no mixing between solutions in the arms containing the test and control cues, and the flow rate in these arms was equal. We introduced one snail into the third arm (“entry arm”) of the Y- maze. To remove directional bias, we introduced the test cue alternately on the left or right side of the entry arm. We washed the Y-mazes and tubing thoroughly between experiments to remove any residue of the introduced chemical cues and snail mucous. Table 1 describes how we prepared the cues for Y-maze treatments and defines the abbreviations we refer to them by in the remaining text.

We analyzed nerite behavior using two different parameters. First, we determined the

“initial arm choice” from the percentage of snails that entered either the test or the control arm first, and used a binomial test to determine if the probability of choosing the test arm first was significantly greater than half (P < 0.05). Second, we determined the “proportion of time” the snails spent in either the test or control arm. We noted the position of each snail ten times (once every ten minutes), taking the first reading after the snail had traversed three-fourths of the entry arm for the first time. We defined the proportion of time spent as the total number of times, among ten measurements, that a subject was in that arm. If the data were normally distributed

(which was true for all treatments except for MES), we used a one-sided paired t-test to analyze

145 differences between proportions of time spent in the test and control arms. If the data were not normal, we used a Wilcoxon signed rank test. We performed all statistical analyses using the program Minitab ((Minitab Inc., State College, PA, USA).

Table B-0-1: A description of the experimental treatments, with abbreviations used to refer to them in the text and figures. All incubations were done at 6°C with 10 gallons of continuously aerated synthetic seawater (SSW) kept in a glass tank. N = number of trials.

Description of cues Treatment N (Abbreviation) Test Control

SSW with ~300 µM SSW with ~150 µM Methane 23 dissolved methane and ~ 150µM dissolved oxygen (achieved by dissolved oxygen bubbling nitrogen gas) SSW incubated with 5 SSW alone incubated for 5 B. childressi mussels (> 60 mm shell length) for 47 days and bubbled with methane seawater (BCS) 5 days, bubbled with methane once once a day (CS) a day. M. edulis SSW incubated with 10 SSW alone incubated for 5 10 seawater (MES) M. edulis mussels for 5 days days B. naticoidea SSW incubated with 25 SSW alone incubated for 5 19 seawater (BNS) snails for 5 days days Same as BCS, except mussels were maintained for 15 days with water changes on days 5 15d_BCS 18 and 10. The SSW incubated with CS mussels for the last 5 days was used as the test cue BCS passed through a CS passed through a 0.4_BCS 10 0.4µm pore filter 0.4µm pore filter BCS passed through a CS passed through a 0.2_BCS 8 0.2µm pore filter 0.2µm pore filter BCS autoclave treated at Autoclave treated CS, 15 pounds of pressure, 120 °C AC_BCS 17 cooled and filtered through temperature, cooled, and filtered cheesecloth through cheesecloth Same as 15d_BCS, except

S_BCS 11 mussel shells were scrubbed with a CS wire brush before water changes

146 Results and Discussion

B. childressi beds are typically associated with elevated methane levels (Nix et al. 1995;

Smith et al. 2000a). Smith et al. (2000) found methane levels of 42-626 µmol L-1 among the mussel beds at our study site. Therefore, B. naticoidea could potentially use methane as a cue to locate these beds. In this study, we used methane dissolved in seawater as a test cue in Y-maze experiments with nerites. We found that the nerites showed no preference for methane in these experiments (P= 0.661; binomial test; Figure B-1; P= 0.875; one-sided p-aired t-test; Figure 2).

On the other hand, nerites showed a strong preference for seawater acclimated with B. childressi.

62% of the snails entered the arm containing BCS (Bathymodiolus Conditioned Seawater) first

(P=0.039; binomial test; Figure B-1), and the time they spent in this arm was significantly greater than that in the control arm (P< 0.0001; one-sided paired t-test; Figure 2). The attraction the nerites displayed towards B. childressi was apparently species specific, as they showed no preference for seawater acclimated with the shallow-water mussel M. edulis (P= 0.623; binomial test; Figure B-1, and P= 0.541; one-sided paired t-test; Figure B-2).

Individual nerite behavior is strongly influenced by the activities of other conspecifics that routinely crawl on B. childressi shells, depositing pedal mucous, feces and egg cases. We established that they follow mucous trails of conspecifics, as snails chose the same arm of a Y- maze as a previous snail with 100% fidelity (6 trials with mazes not washed between introduction of snails; data not shown). Thus B. naticoidea’ s attraction to BCS could be due to conspecific cues emanating from the mussel shell surface. We examined this possibility using two sets of experiments. First, we tested the behavior of snails exposed to seawater acclimated with only conspecifics, excluding mussels (BNS). We found that they showed no preference for

BNS (P= 0.500; binomial test; Figure B-1, and P= 0.255; one-sided paired t-test; Figure B-2).

147 Second, we tested seawater incubated with B. childressi kept separate from snails for 15 days, along with intermittent water changes (15d_BCS). This treatment should have reduced snail residue on the mussel shells. Initial choice for the 15d_BCS arm was not significant (P=0.119, binomial test; Figure B-1). However 61% of the snails entered the 15d_BCS arm first, which is the same percentage of snails that chose the BCS arm first. The lack of significance in the binomial test for the 15d_BCS treatment could be due to the relatively small sample size (18 trials), as compared to the BCS treatment (47 trials). Moreover, the snails spent significantly greater time in the 15d_BCS arm than in the control arm (P= 0.0014; one-sided paired t-test;

Figure B-2). These results combined indicate that B. naticoidea is attracted to a component in

BCS that originates from B. childressi, and not from conspecific snails.

We further conducted a series of experiments to deduce the nature of the attractant. First we investigated whether the cue was particulate. We filtered BCS through 0.2 or 0.4 µm pore- size filters before using it in Y-maze treatments (0.2_BCS and 0.4_BCS, respectively). These treatments likely eliminated most fragments of periostracum or byssal threads and bacteria originating from mussel shells, which are 0.5 to 20 µm in diameter (Zande and Carney 2001).

The snails had a strong preference for both 0.2_BCS and 0.4_BCS (Figures B-1 & B-2), implying that the cue that attracts them to BCS is most likely not an intact particle. Instead, it is a solute or particle less than 0.2 µm in size. Further we autoclaved the BCS to denature proteins and break down complex carbohydrates (AC_BCS). Only 53% of the snails entered the AC_BCS arm first (P=0.315; binomial test; Figure B-1), but the snails still spent significantly more time in this arm than in the control arm (P=0.046; one-sided paired t-test; Figure B-2). Thus the attractant in BCS is heat sensitive but its effect is not abolished by autoclave treatment. This result is consistent with a complex cue, such as a combination of a heat sensitive peptide or

148 complex carbohydrate, and a heat insensitive component such as an amino acid or sugar. Both simple sugars and amino acids have been implicated as food-locating cues for shallow-water snails (Krug and Manzi 1999; Lombardo et al. 1992). Alternately, the cue could be a relatively small peptide whose activity is only partially suppressed by heat treatment. Low molecular weight peptides are a widespread type of cue used by aquatic organisms (Rittschof 1990;

Rittschof and Bonaventura 1986). Further studies are needed to identify the active component of

BCS.

Finally, since the nerites graze on the organic film on B. childressi shells, we hypothesized that the attractant in BCS originates from this film. We tested this by scrubbing B. childressi shells with a wire brush, likely removing most of the organic film, before using seawater incubated with them as a cue (S_BCS). In accordance with our premise, snails showed no preference for S_BCS (P= 0.274; binomial test; Figure B-1, and P= 0.129; one-sided paired t- test; Figure B-2). Due to the relatively low power of our statistical analysis, we cannot say for certain whether scrubbing the mussel shells abolished snail attraction to mussel cues completely.

However, our preliminary findings indicate the signal B. naticoidea is attracted to in BCS might originate from the organic film on B. childressi shells, and could be a product of bacterial degradation. The organic film on B. childressi shells includes decomposing periostracum, and microbial decomposition of organic material produces peptide cues recognized by other marine invertebrates.

Our study demonstrates that B. naticoidea can perceive and respond to cues associated with B. childressi. The seafloor at our study site located at 650 m depth in the Gulf of Mexico typically experiences currents with average velocities of 10 centimeters per second (Welsh and

Inoue 2000). The flow rates we used in our study (0.125 centimeters per second) were

149 substantially lower than these currents. Thus it is feasible that seafloor currents can transport cues that the nerites can use to locate active B. childressi mussel beds. This could be important if the snails wander off from mussel beds, or if mussels in the bed they occupy die. This mechanism for detecting mussel beds may also be used for settlement by B. naticoidea larvae, but further studies are needed to examine this possibility. Chemicals associated with chemosynthetic environments are commonly thought to be attractants for organisms endemic to them. For example, sulfide was implicated as a possible attractant for vent endemic shrimp

(Renninger et al. 1995). In contrast, we found that a biogenic cue, and not methane, evoked an attraction response from B. naticoidea.

150 100 90 * 80 (%) 70 60 * *

choice 50 40 arm 30 20 Initial 10 0 Methane BCS MES BNS 15_d BCS 0.4_BCS 0.2_BCS AC_BCS S_BCS

% Experimental % Control

Figure B-0-1: “Initial arm choice”, or percentage of snails that entered either the test or the control arms of the Y-maze first. Bars with asterisks show significant (P<0.05) preference for the test arm.

0.6

0.5 * e * m i t

0.4 * * f o

* n 0.3 o i t r

o 0.2 p o r P 0.1

0.0 Methane BCS MES BNS 15_d BCS 0.4_BCS 0.2_BCS AC_BCS S_BCS

Experimental Control

Figure B-0-2: The mean ± SE of the proportions of time spent by snails in the test or control arms. Bars with asterisks indicate significant difference (P < 0.05) between times spent in the test and control arms.

151

Acknowledgements

We thank the captain and crew of the Research Vessel (RV) Seaward Johnson II as well as the crew and pilots of the Johnson Sea Link submersible (Harbor Branch

Oceanographic Institution). We would also like to thank the LSU College of Basic

Sciences Glass Shop for making the Y-mazes. The Minerals Management Service, Gulf of Mexico Regional OCS office, through contract number 1435-01-96-CT30813, the

NOAA National Undersea Research Program at the University of North Carolina,

Wilmington, and the National Science Foundation grant OCE 0117050 supported this work.

152 APPENDIX C

Tissue carbon, nitrogen, and sulfur stable isotope turnover in transplanted Bathymodiolus childressi mussels: Relation to growth and physiological condition

Sharmishtha Dattagupta*1, Derk C. Bergquist1, E. B. Szalai 1, S. A. Macko2 and Charles R. Fisher1

Previously published in Limnology and Oceanography, 49 (4), 2004, 1144-1151.

1 Department of Biology, The Pennsylvania State University, University Park, PA-

16802

2 Department of Environmental Sciences, University of Virginia, Charlottesville, VA-

22903

153 Acknowledgements

We thank the captain and crew of the Research Vessel (RV) Edwin Link as well as the crew and pilots of the Johnson Sea Link submersible (Harbor Branch

Oceanographic Institution). We also thank R. W. Lee for helpful discussions. This manuscript has benefited significantly from suggestions made by S. E. MacAvoy, H.

Sahling, and an anonymous reviewer. The Minerals Management Service, Gulf of

Mexico Regional OCS office, through contract number 1435-01-96-CT30813, the NOAA

National Undersea Research Program at the University of North Carolina, Wilmington, and the National Science Foundation grant OCE0117050 supported this work.

154 Abstract

The growth and physiological condition of the methanotrophic hydrocarbon seep mussel

Bathymodiolus childressi reflects the habitat quality of the sites it occupies on the seafloor of the upper Louisiana slope of the Gulf of Mexico. Here, tissue stable isotope compositions, growth, and physiological health of B. childressi mussels transplanted between different seep sites changed within a year to reflect conditions at the new environment. Tissue stable carbon and nitrogen isotope turnovers, although substantial, were not complete at the end of the one year transplant period; they were strongly correlated with each other, and the extent of turnover of both varied by site and was related to the growth of the mussels. Carbon and nitrogen isotopic turnovers caused by metabolic tissue turnover were about 35% per year. This slow isotopic turnover in B. childressi is presumably due to its relatively slow growth and low metabolic rate. On the other hand, tissue stable sulfur isotope turnover was not correlated with either stable carbon or nitrogen isotope turnover or growth, and was higher in the site with higher levels of sulfide in the environment. This suggested that tissue stable sulfur isotope turnover in these mussels is influenced by sulfide detoxification activities.

155 Introduction

The study of tissue stable isotope turnover is a powerful approach to analyzing dietary and environmental factors that can affect an organism’s metabolic processes such as tissue growth and replacement. In general, tissue stable isotope ratios of an animal reflect its nutritional resources, although small changes can occur without changes in diet as a result of seasonal variation in metabolism (Lorrain et al. 2002). Changes in diet resulting from migration (reviewed in Hobson 1999), availability of prey (Macavoy et al.

2001) or ontogenetic shifts (Herzka et al. 2001) can lead to concomitant changes in tissue stable isotope composition provided the different diets have distinct stable isotope signatures. When an animal switches its diet, its tissue stable isotope composition begins to reflect its new diet at a rate proportional to its growth and metabolic turnover (Fry and

Arnold 1982; Herzka et al. 2001; Macavoy et al. 2001). Consequently, factors such as food availability or habitat conditions, which affect rates of tissue growth and replacement, are also likely to influence rates of tissue stable isotope turnover.

B. childressi is a widely distributed and numerically dominant mussel species at hydrocarbon seep sites on the seafloor of the upper Louisiana slope (ULS) of the Gulf of

Mexico. The gills of this mussel contain methanotrophic endosymbionts (Childress et al.

1986) that provide it with the bulk of its nutrition (Fisher and Childress 1992; Streams et al. 1997). In fact, it is capable of growing in the presence of methane as a sole carbon and energy source (Cary et al. 1988), and based on its tissue carbon isotope value, most of its carbon is methane derived (Brooks et al. 1987). However, B. childressi is also able to filter feed, which might be important for acquiring essential organic nutrients not supplied by its endosymbionts (Page et al. 1990) and perhaps for obtaining bulk nitrogen

156 (Pile and Young 1999). In addition to acquiring nitrogen through filter feeding, B. childressi can utilize other nitrogen resources such as free amino acids, inorganic nitrates or ammonia (Lee and Childress 1995; Lee and Childress 1996; Lee et al. 1992). So far, dietary sulfur sources of B. childressi have not been investigated in detail. However, a study by Brooks et al. (1987) showed that the tissue δ34S values of some B. childressi mussels from the ULS are more similar to the +20‰ δ34S value of seawater sulfate than to the –12 to +2‰ δ34S values of sulfide-utilizing tubeworms and clams living in the same area. This suggested that seawater sulfate might be an important source of sulfur for

B. childressi.

The hydrocarbon seep environment on the ULS is patchy and variable and B. childressi mussels at various sites experience a range of habitat conditions. Three sites inhabited by B. childressi are Brine Pool NR-1 (“BP”), GC234 and Bush Hill (“BH”).

While BH and GC234 are separated by less than a kilometer and are habitat-wise rather similar, BP is about four kilometers away and is environmentally distinct from both BH and GC234. The pool of brine at BP is saturated with methane and lacks sulfide. Levels of sulfide and long chain hydrocarbons are generally higher at GC234 and BH than at the most densely occupied part of the mussel bed at BP, whereas methane levels are generally higher at BP (Nix et al. 1995; Smith et al. 2000b). These environmental parameters likely contribute to the larger maximal size, better health and greater densities of B. childressi mussels at BP than at GC234 and BH (Nix et al. 1995; Smith et al.

2000b). Moreover, the δ13C value of bubbles of methane gas at BP is about –64‰, whereas it is about –49‰ at GC234 and –44‰ to –46‰ at BH (Macdonald et al. 1990b;

Sassen et al. 1999). Since the tissue δ13C signatures of B. childressi mussels reflect their

157 methane source (Brooks et al. 1987), the tissue δ13C values of mussels at BP and GC234

(or BH) are very different (Kennicutt II et al. 1992).

Because tissue stable carbon isotope compositions of B. childressi mussels from BP and GC234 (or BH) are distinct, we predicted that transplanting mussels between the different sites would enable us to measure site-specific, in-situ tissue stable isotope turnover of B. childressi. By measuring growth rates in the same animals, we were able to separate the component of change in tissue stable isotope values due to tissue growth from that due to metabolic tissue turnover. Since mussels at BP are generally in better physiological condition than mussels at GC234 or BH, they are likely to have greater rates of tissue growth and repair. Therefore, we hypothesized that mussels transplanted from GC234 (or BH) to BP would undergo greater tissue stable isotope turnover than reciprocally transplanted mussels. In addition to providing new insights on the biology of

B. childressi, this study adds substantially to our understanding of stable isotope turnover in general and the limitations of tissue stable isotope analyses in hydrothermal vent and cold seep trophic studies (Macavoy et al. 2001; Rieley et al. 1999; Tieszen et al. 1983).

Materials and Methods

Study sites

BP is located at ~650 m depth on the ULS of the Gulf of Mexico (27°43.4’N,

91°16.5’W). This hypersaline (salinity = 121 g/kg) pool is ~22 m long and ~16 m wide and is surrounded by a large (3-7 m wide) continuous bed of B. childressi mussels, the inner rim of which is elevated a few centimeters above the surface of the brine

(MacDonald et al. 1990). For this transplant study, we used mussels from the inner margin of this bed (“inner BP”) where concentrations of methane are high (42 – 626

158 µmol L-1) and sulfide is normally not detectable among the mussels (Smith et al. 2000).

The salinity at siphon level of these mussels is rarely above 39 g/kg (Smith et al. 2000). It is important to note here that higher sulfide levels have been reported at the outer margin of the BP mussel bed (Smith et al. 2000), but we did not sample in that portion of the bed.

The mussel bed at inner BP is densely packed with relatively fast growing and healthy individuals (Smith et al. 2000).

The GC234 site is located at a depth of ~540 m on the ULS (27°44.7’N,

91°13.3’W). This site covers an area of several square kilometers and is dominated by large tubeworm aggregations. Mussel beds, when present, are associated with one or more tubeworm aggregations. Methane levels among mussels in these beds are variable

(non-detectable to 10.7 mmol L–1; Nix et al. 1995) and can sometimes be even higher than that found at the inner BP (Smith et al. 2000). Sulfide concentrations as high as 8 mmol L-1 have been detected among these mussels, and these values are associated with the high methane concentrations mentioned above (Nix et al. 1995). Moreover, oil globules are often released when mussel beds are disturbed at GC234 (CRF, personal observation). Previous studies found that the mussels at this site had lower physiological condition indices than inner BP mussels and did not grow as large as the BP mussels (Nix et al. 1995; Smith et al. 2000).

BH is located at a depth of ~540- 580m on the ULS (27°46.96’N, 91°30.46’W).

Although much smaller in extent, this site is geochemically very similar to GC234, with relatively high sulfide levels (non-detectable to 56 µmol L-1; Nix et al. 1995) and presence of oil in the mussel beds (CRF, personal observation). Previous studies found that the mussels at this site had very low physiological condition indices when compared

159 to both BP and GC234 and did not grow as large as the BP mussels (Nix et al. 1995;

Smith et al. 2000).

Transplants

Two sets of reciprocal transplants of B. childressi mussels were conducted: one between mussel beds at GC234 and inner BP and another between mussel beds at BH and inner

BP. Throughout this paper, the native site of the mussels is referred to as “origin” and the site to which they were transplanted “host”. In September 1994, mussels were collected from the inner edge of the BP mussel community, a bed in GC234 (about 120 mussels each) and a bed in BH (75 mussels) using the Johnson Sea Link I manned submersible

(Harbor Branch Oceanographic Institution). The mussels were transported to the surface in a temperature-insulated box attached to the front of the submersible. They were kept in chilled seawater (8°C) and processed in a cold room (8°C) onboard the ship. The length of each mussel shell was measured to the nearest 0.1 mm using calipers and a commercial color coded, numbered larval fish tag was glued to its umbo for identification (Nix et al.

1995). Within 24 hours after collection, the tagged mussels were transported in temperature-insulated containers filled with chilled seawater to the host site and deployed at the designated mussel bed. The mussels were retrieved 336 days later (August 1995) using collection methods described above. The shell length of each mussel was measured, and marked individuals of a representative size range were sub-sampled from each collection. Tissue from sub-sampled mussels was processed for determination of physiological condition. Three indices of condition were determined: water content (as a percentage of wet weight), condition index (CI; a ratio of ash free dry tissue mass to shell

160 volume) and glycogen content (as a percentage of wet weight) following the procedures of Smith et al. (2000). A subset of the tissue homogenates used for physiological condition analyses were used to determine stable isotope composition.

Control manipulations were not conducted during the same time period as the transplant experiments. Instead, control mussels were collected from inner BP and the mussel bed GC234 in June 1993 using the Johnson Sea Link I submersible. They were measured and tagged as described above for the transplant mussels and were deployed back to their original mussel bed. They were collected 456 days later (September 1994) and their tissues were processed in the same manner as the transplant mussels. The quality of these controls for site to site comparisons of growth and physiological condition is somewhat compromised as small yearly differences in condition index and growth of B. childressi mussels from a single mussel bed have been observed (Smith et al. 2000). However, these yearly differences are of a much smaller magnitude than the differences between sites. To confirm that tissue stable isotope compositions of mussels at our study sites did not change over the course of our study, we also determined the tissue stable isotope compositions of unmanipulated mussels collected from inner BP and

GC234 in August 1995. We compared the stable isotope compositions of these two sets of control mussels (manipulated and unmanipulated) and found no significant difference between their tissue δ13C, δ15N or δ34S values (Figure C-2). Therefore, we used the mean values of tissue stable isotope compositions of the two sets of controls for further analyses. No control manipulations were performed on mussels from BH. Therefore, for

BH controls, we analyzed tissue stable isotope compositions and physiological condition

161 indices of unmanipulated mussels retrieved from BH in August 1995. No growth data were available for the BH controls.

Stable isotope analyses

Gill and mantle tissues of individual mussels were homogenized to a 10x dilution in distilled, deionized water using a Brinkman PT3000 Polytron tissue homogenizer.

Approximately 1 ml samples of these homogenates were incubated in an oven (60°C) until dry and were then ground into a fine powder with a mortar and pestle. The powder was transferred into tin capsules (1 mg of powder for δ13C and δ15N analyses, 6 mg of

34 powder for δ S analyses) and the samples were converted to CO2, N2 and SO2 for isotope analyses using a Carlo Erba Elemental Analyzer (EA) coupled to an OPTIMA stable isotope ratio mass spectrometer (Micromass, Manchester, UK). Carbon and nitrogen isotopic compositions were determined on a single combustion using a dual furnace system composed of an oxidation furnace at 1020oC and a reduction furnace at

650oC. Using the EA, the samples for sulfur were separately pyrolyzed at 1050oC using a single furnace combination oxidation and reduction system. The resulting gases were chemically dried and directly injected into the source of the mass spectrometer using continuous flow. The stable isotopic ratio is reported as follows:

M 3 δ E = [R sample /R standard –1] X 10 (‰) (1) where M is the atomic mass of the heavy isotope of the element E and R is the abundance ratio of the heavy to light isotopes (13C/12C, 15N/14N or 34S/32S) of that element. The standards used for carbon, nitrogen, and sulfur are PeeDee Belemnite limestone (PDB), atmospheric N2 (air) and Canyon Diablo Triolite (CDT), respectively. For sulfur and

162 carbon, the values are corrected for the influence of oxygen isotopes. The reproducibility of the measurement for these elements is typically better than ± 0.2‰ using the continuous flow interface on the OPTIMA. In the laboratory, the samples are commonly measured against tanks of carbon dioxide, nitrogen and sulfur dioxide gases that have been calibrated against NBS 22, atmospheric N2 and NBS 127, respectively.

Data analyses

To ensure normality, all measurements for condition were transformed using an arcsine function (y’ = 2 arcsine (y1/2)) before statistical analysis (Neter et al. 1996).

Significant differences between transplanted and control mussels from either the origin or the host site were tested using ANCOVA with length as a covariate (SAS Version 6.07,

Proc GLM). If the model was significant (p < 0.05), pair-wise comparisons were made using Tukey’s correction for multiple comparisons (SAS Version 6.07, Proc GLM).

Differences in growth parameters between transplant and control populations were tested using a linear regression with the indicator variables in a Ford Walford plot

(Walford 1946). Two methods were used in order to adjust for the different recapture intervals between control and experimental animals prior to applying the Ford Walford plots. In the first method, which assumes equal recapture intervals, no adjustment was made to the length data. This method overestimates the difference between the control and transplant regression lines because it assumes that the group with the shorter recapture interval would have achieved no additional growth if their recapture interval had been lengthened to match the other group. Growth calculated by this method is referred to as “unadjusted growth”. In the second method, the final shell length of all

163 mussels was standardized to 365 days by assuming equal growth per day for each individual. This method may underestimate the difference between the control and transplant regression lines if periods of slow growth coincide with the time period missed by the group with the shorter recapture interval. Growth calculated by this method is referred to as “adjusted growth”. Agreement between the results of the two correction methods increases our confidence in the conclusions.

Significant differences between stable isotope compositions of the two sets of control mussels and the transplant and control mussels were tested using two-tailed

Student’s t-tests at a significance level of 0.05 (Minitab Release 12.21, Minitab Inc.).

Where multiple comparisons were made, the significance level was adjusted using the sequential Bonferroni method (Rice 1989). For each transplant mussel, amount of change in tissue stable isotope value of element E (ΔδME) was calculated as:

ΔδME = |(δME of transplant mussel) – (mean δME of origin control mussels)| (2)

The relationship between ΔδME and growth was tested by a linear regression between ΔδME values and values of percent change in shell length (%CL). %CL values were used as indicators of size-specific growth, and calculated as:

%CL = (Change in shell length /initial shell length) * 100 (3)

Here, values of shell growth were standardized to 365 days, assuming a constant daily growth rate. In order to correct for any potential size-related effects, relationships between tissue δ13 C, δ15 N or δ34 S, and final shell lengths of transplant and control mussels were investigated. No such relationship was found.

To estimate extents of isotopic turnover in populations of transplant mussels, the

M M amount of change in mean δ E (Δδ E mean) of the transplant population was calculated as

164 M M a percent of the total difference (Δδ E total) between the mean δ E of the origin and the host site controls. Here,

M M M Δδ E mean= |(mean δ E of transplant mussels) – (mean δ E of origin control mussels)|(4) and

M M M Δδ E total=|(mean δ E of host control mussels)– (mean δ E of origin control mussels)|(5)

Percent change in mean δME (%ΔδME) of each transplant population was calculated as:

M M M % Δδ E = (Δδ E mean / Δδ E total) * 100 (6)

The above values of %ΔδME estimate isotopic turnover of transplant animals caused by both shell (and tissue) growth and metabolic tissue turnover. To estimate the amount of isotopic turnover due to metabolic tissue turnover alone (i.e., in the absence of growth),

M M the mean value of Δδ E in the absence of shell growth (Δδ EO) was calculated for transplant populations as:

M M Δδ EO = ‘y’- intercept of line relating Δδ E and %CL (7)

To allow direct comparisons between different transplant populations, the ‘y’-intercept was determined from the regression lines made in the %CL range of 0 to 25 %, the %CL range common to all transplant populations for which there was a correlation between

M M M Δδ E and %CL. Percent change in δ E in the absence of shell growth (%Δδ EO) was then calculated as:

M M M %Δδ EO= (Δδ EO / Δδ E total) * 100 (8)

165

Results

Growth and condition indices

BP control mussels had significantly higher condition index (CI) and glycogen content and significantly lower water content than either BH or GC234 control mussels

(Table C-1), which is consistent with previously reported results (Nix et al. 1995; Smith et al. 2000). They also grew significantly more than GC234 control mussels (Figure C-1; p < 0.001 for adjusted and unadjusted growth).

Growth and physiological condition of transplanted mussels tended to resemble those of their host populations. All three physiological condition indices of mussels transplanted from BP to either GC234 or BH were not significantly different from those of their host populations but were significantly different from those of their origin populations (Table C-1). Growth of BP to GC234 transplant mussels was not significantly different from that of their host population (p = 0.117 and 0.349 for adjusted and unadjusted growth respectively) but was significantly different from that of their origin population (p < 0.001 for both adjusted and unadjusted growth; Figure C-1).

Similarly, all three physiological condition indices of mussels transplanted from either

GC234 or BH to BP were significantly different from those of their origin populations, and all except glycogen content of these mussels were not significantly different from those of their host populations (Table C-1). Growth of GC234 to BP transplant mussels was intermediate between and significantly different from that of both their host and origin populations (Figure C-1). Growth of mussels transplanted from BP to BH, and BH to BP was significantly lower than that of BP control mussels (p < 0.001 for both

166 adjusted and unadjusted growth; data not shown), and very little growth was observed for

BP to BH transplant mussels. No growth data were available for BH control mussels, so comparisons could not be made.

Stable isotope analyses

The respective tissue δ13C, δ15N and δ34S values of BH and GC234 control mussels were not significantly different. However, tissue δ13C, δ15N and δ34S values of

BP control mussels were all significantly different from those of BH and GC234 mussels.

The mean tissue δ13C values of BP, BH and GC234 control mussels (–62‰, –39‰, and –

41‰, respectively; Figure C-2a, d) reflected the δ13C values of methane sources at the three sites (Kennicutt II et al. 1992; Brooks et al. 1987), with tissue δ13C values of BP control mussels being significantly more depleted than those of either BH or GC234 control mussels (t = 32.58; df = 14; p < 0.00001 and t = 36.3; df = 36; p < 0.00001, respectively). While mean tissue δ15N values of BH and GC234 control mussels were

2.6‰ and + 2.9‰, respectively, BP control mussels had a mean tissue δ15N value of

-15.5‰, which was significantly more depleted than both (t = 26.26; df =6; p < 0.00001 and t = 52.7; df = 32; p < 0.00001, respectively; Figure C-2b, e). Conversely, BH and

GC234 control mussels had significantly (t =-7.56; df = 29; p < 0.0001 and t = 6.87; df =

35; p < 0.0001, respectively) more depleted mean tissue δ34S values (8.9‰ and +8.4‰, respectively) than that of BP control mussels (+13.1‰; Figure C-2c, f).

All three stable isotope ratios of transplanted mussels resembled those of their host populations more than those of their origin populations (Figure C-2), indicating that

167 tissue stable isotope turnover occurred in all transplants between BH or GC234 and BP.

The mean tissue δ13C and δ15N values of transplanted mussels were intermediate between and significantly different (p < 0.05) from those of both their host and origin populations, except tissue δ15N values of the BP to BH transplant mussels, which were not significantly different from those of their origin population. While the mean tissue δ34S values of GC234 or BH to BP transplant mussels followed a similar pattern, the mean

δ34S values of BP to GC234 or BH transplant mussels were not significantly different from those of their host populations (BP to GC234: t = 2.35; df = 3; p = 0.10; BP to BH: t

= 1.35; df = 3; p = 0.27).

Changes in stable carbon isotope ratios (Δδ13C; Equation 2) of individual mussels transplanted between BP and GC234 were significantly (p < 0.05) correlated with changes in stable nitrogen ratios (Δδ15N) (Figure C-3). However, changes in sulfur stable isotope ratios (Δδ34S) of these mussels were not significantly correlated with either Δδ13C or Δδ15N values (BP to GC234: p = 0.108 and 0.290; GC234 to BP: p = 0.525 and 0.900 for relationships with Δδ13C and Δδ15N respectively). Further, a strong correlation was found between Δδ13C and Δδ15N values of transplant mussels and size specific shell growth (%CL; Figure C-4), and this relationship was significant (p <0.05) except in the case of Δδ13C of BP to GC234 transplant mussels (p= 0.058). Δδ34S and %CL were not significantly correlated for either transplant (BP to GC234: p = 0.346; GC234 to BP: p =

0.601).

Similar results were obtained for mussels transplanted between BP and BH. Δδ13C values of transplant mussels were correlated with Δδ15N values (BP to BH: p = 0.005; BH

168 to BP: p = 0.064; Figure C-3). Again, no correlation was found between Δδ34S values and either Δδ13C or Δδ15N values of these mussels (BP to BH: p = 0.367 and 0.503; BH to

BP: p = 0.481 and 0.407 for relationships with Δδ13C and Δδ15N respectively). Δδ13C and

Δδ34S values of BH to BP transplant mussels were not significantly correlated with %CL values (p = 0.837 and 0.329 for C and S, respectively), but the correlation between Δδ15N of these mussels and %CL was significant (p = 0.045). Growth of BP to BH transplant mussels was almost always negligible, and changes in tissue stable isotope compositions and %CL values were not significantly correlated for these mussels.

The relative amounts of tissue stable isotope turnover of transplant mussels were compared using values of percent changes in mean stable isotope ratios (%Δδ13C,

%Δδ15N, %Δδ34S; Equation 6). Mussel populations transplanted from either GC234 or

BH to BP had higher total %Δδ13C and %Δδ15N, and lower total%Δδ34S, than the respective BP to GC234 or BH transplant populations (Table C-2). Mussel populations transplanted between BP and GC234 had greater tissue stable isotope turnovers than those transplanted between BP and BH.

Tissue stable carbon and nitrogen isotope turnover resulting from metabolic tissue turnover (Table C-2) was calculated using the ‘y’-intercepts of lines relating Δδ13C or

Δδ15N to %CL (Equation 8). The 0 to 25% range of %CL values was common to all transplant populations. Thus, the ‘y’-intercepts of regression lines corresponding to the

GC234 to BP, and BH to BP transplants were calculated using a subset of data points in this %CL range (GC234 to BP: Δδ13C vs. %CL: y=34.1x + 11.3; p = 0.03; r2 = 0.46;

Δδ15N vs. %CL: y=21.3x + 9.5; p = 0.02; r2 = 0.50; BH to BP: Δδ15N vs. %CL: y=9.74 +

169 25.3x; p = 0.045; r2 = 67.4%). These calculations indicate that about 34- 56% of carbon and nitrogen isotopic turnover of these mussels was due to metabolic tissue replacement.

Tissue stable sulfur turnover due to metabolic tissue replacement could not be estimated, as the relation between Δδ34S and %CL was not significant for any of the transplants.

Discussion

Transplanted mussels underwent significant changes in physiological health and isotopic composition within a year (Table C-1; Figure C-2). Growth and condition data for BP, GC234 and BH control mussels were consistent with earlier studies and confirm that the inner edge of BP is more favorable for B. childressi mussels than either GC234 or BH (Nix et al. 1995; Smith et al. 2000). When transplanted to BP, the physiological condition of both GC234 and BH mussels improved significantly (Table C-1) to mirror their host site. GC234 to BP transplant mussels also grew significantly more compared to their origin population (Figure C-1). On the other hand, BP mussels significantly deteriorated in physiological condition (Table C-1) when transplanted to either GC234 or

BH, and BP to GC234 transplant mussels grew significantly less than their origin population (Figure C-1). The change in physiological condition indices upon transplantation to a less favorable habitat is reminiscent of the deterioration in condition of hydrothermal vent mussels when transplanted away from hydrothermal flow (Smith Jr

1985). In addition to improving physiological condition, GC234 to BP and BH to BP transplant mussels experienced greater tissue carbon and nitrogen stable isotope turnover than BP to GC234 and BP to BH transplant mussels, respectively (Table C-2). Thus, the amount of tissue carbon and nitrogen stable isotope turnover in B. childressi mussels is

170 likely to be linked to conditions at the host site through changes in growth and physiological condition.

In this study, a correlation between tissue Δδ13C and Δδ15N of individual transplant mussels was observed (Figure C-3), suggesting similar and perhaps coupled dynamics for carbon and nitrogen turnover in B. childressi tissue. The extent of an organism’s tissue isotopic turnover is a function of both tissue growth and metabolic tissue replacement (Fry and Arnold 1982; MacAvoy et al. 2001; Herzka et al. 2001). The significant correlation between Δδ13C or Δδ15N and shell growth of mussels transplanted between BP and GC234 illustrates the relation between growth and tissue carbon and nitrogen stable isotope turnover in B. childressi. The slopes of the lines (Δδ13C or Δδ15N vs. %CL) corresponding to the BP to GC234 and GC234 to BP transplants were somewhat similar, suggesting that growth induced carbon and nitrogen stable isotope turnovers were similar at the two sites.

In order to deduce the amount of isotopic turnover due to metabolic tissue replacement alone, we estimated stable isotope turnover in mussels with no growth. It is important to note that this estimate is based on shell growth, which is not necessarily a good predictor of tissue growth as mussels may improve their physiological condition and increase tissue mass without any change in shell length. The physiological conditions of BP to GC234 transplant mussels deteriorated significantly on transplantation (Table C-

1), and so they are unlikely to have added tissue mass in the absence of shell growth.

Therefore, the tissue stable isotope turnovers in BP to GC234 transplant mussels with no

M shell growth (%Δδ EO; Equation 8), which were 36% and 34% per year for carbon and nitrogen respectively (Table C-2), are estimates of tissue stable isotope turnover due to

171 metabolic turnover alone, albeit in mussels of compromised condition. On the other hand,

GC234 to BP transplant mussels increased their condition index on transplantation (Table

C-1), and so they may have increased their tissue mass without increasing their shell

M length. Thus, the %Δδ EO values for these mussels (53% to 54%, Table C-2) are likely to be overestimates of tissue stable isotope turnover due to metabolic turnover alone. The

13 15 close agreement between the independently derived estimates of %Δδ CO and %Δδ NO for a particular transplant argues for the validity of these parameters.

The pattern of sulfur stable isotope turnover in B. childressi mussels was distinctly different from that of carbon and nitrogen; Δδ34S of individual transplant mussels was not correlated with either Δδ13C or Δδ15N, and there was no apparent relationship between Δδ15S and growth. The relative degree of sulfur stable isotope turnover in the transplant populations was also opposite of what was observed for carbon and nitrogen (Table C-2). While it was complete at the end of the transplant period in the

BP to GC234 transplant population, it was only 50% complete for the GC234 to BP transplant population. Similarly, sulfur isotopic turnover was about 77% for the BP to BH transplant population and only about 37% for the BH to BP transplant population. These findings suggest that B. childressi mussels may incorporate sulfur through a process other than tissue growth and replacement. As mentioned earlier, mussels at both GC234 and

BH are likely to be exposed to higher levels of sulfide compared to mussels on the inner margin of BP (Nix et al. 1995; Smith et al. 2000). Thus, mussels transplanted from BP to

GC234 (or BH) may encounter high levels of sulfide in their new environment. B. childressi mussels are not known to utilize sulfide as an energy source; on the contrary, they produce sulfur containing compounds such as hypotaurine in order to detoxify

172 environmental sulfide (Pruski et al. 2000). Therefore, BP to GC234 (or BH) transplant mussels may start accumulating sulfide- detoxification compounds (Pruski et al. 2000), which may explain the relatively rapid sulfur isotope turnover observed in these mussels.

The low rates of sulfur isotopic turnover in the reciprocal transplants may also be explained by remobilization of the sulfur from these detoxification compounds to meet metabolic sulfur demands.

Due to the difficulty of collecting direct observational data, stable isotope analyses have been used extensively to obtain nutritional information on deep sea hydrothermal vent and hydrocarbon seep organisms (Conway et al. 1994; Fisher 1995;

Kennicutt Ii et al. 1992a). However, the study of trophic relationships using stable isotopes may be obscured by slow rates of stable isotope turnover (Tieszen et al. 1983).

Reports of tissue stable isotope turnover rates are rare. Studies of red drum larvae

(Herzka and Holt 2000), krill (Frazer et al. 1997), brine shrimp (Fry and Arnold 1982),

Japanese quail (Hobson 1999) and gerbils (Tieszen et al. 1983) have shown high rates of tissue stable isotope turnover associated with high growth and metabolic rates in these organisms. In contrast, studies of broad whitefish (Hesslein et al. 1993) and catfish

(MacAvoy et al. 2001) found relatively slow tissue stable isotope turnover, which was attributed to slow metabolism and growth in these animals. To the best of our knowledge, this paper is the first to report tissue stable isotope turnover in a deep sea organism. We found that the period of a year was not sufficient for the tissues of B. childressi to completely express the stable isotope signature of a new, isotopically distinct, dietary source. This reflects a much slower isotopic turnover rate in tissues of B. childressi mussels than in tissues of some fast growing organisms studied so far. For example,

173 shrimp and squid have shown carbon isotopic half lives ranging between 4-19 days (Fry and Arnold 1982), larval krill have shown replacement of 22-29% of their carbon and 13-

22% of their nitrogen at the end of 8-10 weeks (Frazer et al. 1997) and red drum larvae have shown almost complete carbon and nitrogen isotopic turnover at the end of 15 days

(Herzka and Holt 2000). The slow isotopic turnover rate of B. childressi likely reflects its relatively slow growth and low metabolic rate, characteristics that are typical of a benthic bivalve living in a low temperature (6-8 °C) environment.

174 Table C-1: Mean (standard deviation) of final shell length, condition index (CI), glycogen content and water content of transplant and control mussels. n = sample size. Values with different letters indicate a significant difference (p < 0.05) between them (statistical tests were performed on the arcsine transformed values).

Mussel type n Length (mm) CI (g ml-3) % glycogen % water BP controls 24 63(24) 0.10 (0.03) a 1.7 (0.6) a 84.3 (3.2) a BP to GC234 6 60(19) 0.06 (0.02) b 0.2 (0.2) b 88.8 (1.5) b GC234 to BP 17 62(11) 0.07 (0.02) a 0.3 (0.3) c 83.5 (3.6) a GC234 controls 6 51(20) 0.06 (0.02) b 0.1 (0.2) b 87.5 (2.7) b BP to BH 6 70(28) 0.04 (0.01) b 0.1 (0.2) b 90.9 (1.7) b BH to BP 5 60(10) 0.08 (0.01) a 0.3 (0.1) c 84.6 (1.8) a BH controls 6 58(16) 0.04 (0.01) b 0.04 (0.01) b 92.2 (1.6) b

Table C-2: Percent change in the tissue stable isotope ratios of transplant mussels in the presence and absence of shell growth, during the study period of about one year. Values were calculated using Equations (6) and (8). †These values could not be calculated as above, as growth of all BP to BH transplant mussels was negligible (< 0.1 %CL). ††This value could not be calculated, as the linear relationship between Δδ13C and %CL of BH to BP transplant mussels was not significant (p = 0.837; r2 = 0.012).

Average values At zero shell growth 13 15 34 13 15 Transplant %Δδ C %Δδ N %Δδ S %Δδ CO %Δδ NO BP to GC234 54.2 55.0 156.7 35.9 33.6 GC234 to BP 77.8 71.1 51.3 53.3 52.6 BP to BH 40.5 17.9 76.3 -† -† BH to BP 57.3 63.6 36.8 -†† 56.3

175

Figure C-1: Change in shell length standardized to one year (“ adjusted shell growth”) versus initial shell length of transplant and control mussels from the reciprocal transplant between BP and GC234.

176

Figure C-2: Mean values of tissue δ13C, δ15N and δ34S of transplant and control mussels from the reciprocal transplant between a-c) BP and GC234, and d-f) BP and BH. Notations for transplants and controls have the name of the origin site first, followed by the name of the host site. For example, BP-BP represents BP controls, and GC-BP represents GC234 to BP transplants. Controls were of two types: manipulated and unmanipulated (see text for details). n = sample size. Error bars represent standard error. Different letters indicate a significant (p < 0.05) difference between data points.

177

Figure C-3: Relationships between tissue ∆δ13C and ∆δ15N values of individual mussels belonging to the transplants between BP, and GC234 or BH.

178

Figure C-4: Relationships between changes in carbon or nitrogen stable isotope values and shell growth represented by percent change in shell length (%CL) for BP to GC234 transplant mussels and GC234 to BP transplant mussels. a, c) tissue Δδ13C vs. %CL and b, d) tissue Δδ15N vs. %CL.

179 BIBLIOGRAPHY

Aharon, P., and B. Fu. 2000. Microbial sulfate reduction rates and sulfur and oxygen isotope fractionations at oil and gas seeps in deepwater Gulf of Mexico. Geochimica et Cosmochimica Acta 64: 233-246. ---. 2003. Sulfur and oxygen isotopes of coeval sulfate-sulfide in pore fluids of cold seep sediments with sharp redox gradients. Chemical Geology 195: 201-218. Aller, R. C., P. O. J. Hall, P. D. Rude, and J. Y. Aller. 1998. Biogeochemical heterogeneity and suboxic diagensis in hemipelagic sediments of the Panama Basin. Deep-Sea Research I 45: 133- 165. Arp, A. J., and J. J. Childress. 1983. Sulfide binding by the blood of the hydrothermal vent tube worm Riftia pachyptila. Science 219: 295-297. Arp, A. J., J. J. Childress, and C. R. Fisher. 1985. Blood gas transport in Riftia pachyptila. Biological Society of Washington Bulletin 6: 289-300. Arp, A. J., J. J. Childress, and R. D. Vetter. 1987. The sulfide-binding protein in the blood of the vestimentiferan tube-worm, Riftia pachyptila, is the extracellular haemoglobin. Journal of Experimental Biology 128: 139-158. Arvidson, R. S., J. W. Morse, and S. B. Joye. 2004. The sulfur biogeochemistry of chemosynthetic cold seep communities, gulf of Mexico, USA. Marine Chemistry 87: 97-119. Aureliano, M., and R. M. C. Gândara. 2005. Decavanadate effects in biological systems. Journal of Inorganic Biochemistry 99: 979-985. Azuma, M., and Y. Ohta. 1998. Changes in H+- translocating vacuolar-type ATPase in the anterior silk gland of Bombyx mori during metamorphosis. The Journal of Experimental Biology 201: 479-486. Bergquist, D. C., C. Fleckenstein, J. Knisel, B. Begley, I. R. Macdonald, and C. R. Fisher. 2005. Variations in seep mussel bed communities along physical and chemical environmental gradients. Marine Ecology Progress Series 293: 99-108. Bergquist, D. C., I. A. Urcuyo, and C. R. Fisher. 2002. Establishment and persistence of seep vestimentiferan aggregations on the upper Louisiana slope of the Gulf of Mexico. Marine Ecology Progress Series 241: 89-98. Bergquist, D. C. and others 2003. Community structure of vestimentiferan-generated habitat islands from Gulf of Mexico cold seeps. Journal of Experimental Marine Biology and Ecology 289: 197-222. Bergquist, D. C., F. M. Williams, and C. R. Fisher. 2000. Longevity Record for deep-sea Invertebrate: The Growth Rate of a marine tubeworm is tailored to different environments. Nature 403: 499-500. Berner, R. A. 1980. Early diagenesis: a theoretical approach. Princeton University Press. Boetius, A. and others 2000. A marine microbial consortium apparently mediating anaerobic oxidation of methane. Nature 407: 623-625. Boetius, A., and E. Suess. 2004. Hydrate Ridge: a natural laboratory for the study of microbial life fueled by methane from near-surface gas hydrates. Chemical Geology 205: 291-310. Bollinger, R., H. Brandl, P. Hohener, K. W. Hanselmann, and R. Bachofen. 1992. Sqeeze-water analysis for the determination of microbial metabolites in lake sediments- comparison of methods. Limnology and Oceanography 37: 448-455.

180 Bottcher, M. E., T. G. Ferdelman, B. B. Jorgensen, R. E. Blake, A. V. Surkov, and G. E. Claypool. 2006. Sulfur isotope fractionation by the deep biosphere within sediments of the eastern equatorial Pacific and Peru margin. Proceedings of the Ocean Drilling Program, Scientific Results 201: 1-21. Boudreau, B. P., and J. T. Westrich. 1984. The dependence of bacterial sulfate reduction on sulfate concentration in marine sediments. Geochimica et Cosmochimica Acta 48: 2503-2516. Bowman, E. J., A. Siebers, and K. Alterndorf. 1988. Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proceedings of the National Academy of Sciences of the United States of America 85: 7972-7976. Bright, M., H. Keckeis, and C. R. Fisher. 2000. An autoradiographic examination of carbon fixation, transfer and utilization in the Riftia pachyptila symbiosis. Marine Biology 136: 621-632. Brooks, J. M., H. B. Cox, W. R. Bryant, M. C. Kennicutt Ii, R. G. Mann, and T. J. Mcdonald. 1985. Association of gas hydrates and oil seepage in the Gulf of Mexico. Advances in Organic Geochemistry 10: 221-234. Brooks, J. M., M. C. Kennicutt Ii, R. R. Fay, T. J. Mcdonald, and R. Sassen. 1984. Thermogenic gas hydrates in the Gulf of Mexico. Science 225: 409-411. Brooks, J. M. and others 1987. Deep-sea hydrocarbon seep communities: evidence for energy and nutritional carbon sources. Science 238: 1138-1142. Brüchert, V., C. Knoblauch, and B. B. Jorgensen. 2001. Controls on stable sulfur isotope fractionation during bacterial sulfate reduction in Arctic sediments. Geochimica et Cosmochimica Acta 65: 763-776. Brusca, R. C., and G. J. Brusca. 2002. Invertebrates, Second ed. Sinauer Associates, Inc. Byrne, J. H., and S. G. Schultz. 1994. An introduction to membrane transport and bioelectricity, Second edition ed. Raven Press. Cabantchik, Z. I., and R. Greger. 1992. Chemical probes for anion transporters of mammalian cell membranes. American Journal of Physiology 265: C803-C827. Cahalan, M., and E. Neher. 1992. Patch clamp techniques: an overview. Methods in Enzymology 207: 3-28. Carignan, R. 1984. Interstitial water sampling by dialysis: methodological notes. Limnology and Oceanography 29: 667-670. Carignan, R., S. St-Pierre, and R. Gachter. 1994. Use of diffusion samplers in oligotrophic lake sediments: Effects of free oxygen in sampler material. Limnology and Oceanography 39: 468-474. Carney, R. S. 1994. Consideration of the oasis analogy for chemosynthetic communities at Gulf of Mexico hydrocarbon vents. Geo-Marine Letters 14: 149-159. Carpaneto, A., R. Magrassi, E. Zocchi, C. Cerrano, and C. Usai. 2003. Patch-clamp recordings in isolated sponge cells (Axinella polypoides). Journal of Biochemical and Biophysical Methods 55: 179-189. Carpenter, J. H. 1966. New measurements of oxygen solubility in pure and natural water. Limnology and Oceanography 11: 264-277. Cary, S. C., C. R. Fisher, and H. Helbeck. 1988. Mussel Growth Supported by Methane as Sole Carbon and Energy Source. Science 240: 78-80.

181 Cavanaugh, C. M., S. L. Gardiner, M. L. Jones, H. W. Jannasch, and J. B. Waterbury. 1981. Prokaryotic cells in the hydrothermal vent tubeworm Riftia pachyptila Jones: possible chemoautotrophic symbionts. Science 213: 340-342. Chambers, L. A., and P. A. Trudinger. 1978. Microbiological fractionation of stable sulfur isotopes: A review and critique. Geomicrobiology Journal 1: 249-293. Childress, J. J., A. J. Arp, and C. R. Fisher. 1984. Metabolic and blood characteristics of the hydrothermal vent tube-worm Riftia pachyptila. Marine Biology 83: 109-124. Childress, J. J., and C. R. Fisher. 1992. The Biology of Hydrothermal Vent Animals: Physiology, Biochemistry, and Autotrophic Symbioses, p. 337-441. In M. Barnes, A. D. Ansell and R. N. Gibson [eds.], Oceanogr. Mar. Biol. Annu. Rev. UCL Press. Childress, J. J., C. R. Fisher, J. M. Brooks, M. C. Kennicutt Ii, R. Bidigare, and A. E. Anderson. 1986. A Methanotrophic marine Molluscan (Bivalvia, Mytilidae) Symbiosis: Mussels Fueled by Gas. Science 233: 1306-1308. Childress, J. J., C. R. Fisher, J. A. Favuzzi, R. E. Kochevar, N. K. Sanders, and A. M. Alayse. 1991. Sulfide-driven autotrophic balance in the bacterial symbiont- containing hydrothermal vent tubeworm, Riftia pachyptila Jones. Biological Bulletin 180: 135-153. Childress, J. J. and others 1993. Inorganic carbon uptake in hydrothermal vent tubeworms facilitated by high environmental pCO2. Nature 362: 147-149. Conway, N. M., M. C. Kennicutt Ii, and C. L. Van Dover. 1994. Stable isotopes in the study of marine chemosynthetic- based ecosystems, p. 158-186. In K. Lajtha and R. H. Michener [eds.], Stable isotopes in ecology and environmental science. Blackwell Scientific Publications, Inc. Cordes, E. E., M. A. Arthur, K. Shea, R. S. Arvidson, and C. R. Fisher. 2005a. Modeling the mutualistic interactions between tubeworms and microbial consortia. PLOS Biology 3: e77. Cordes, E. E., D. C. Bergquist, K. Shea, and C. R. Fisher. 2003. Hydrogen sulfide demand of long-lived vestimentiferan tube worm aggregations modifies the chemical environment at deep-sea hydrocarbon seeps. Ecology Letters 6: 212- 219. Cordes, E. E., S. Hourdez, B. L. Predmore, M. L. Redding, and C. R. Fisher. 2005b. Succession of hydrocarbon seep communities associated with the long-lived foundation species Lamellibrachia luymesi. Marine Ecology Progress Series 305: 17-29. Culliford, S. J., J. C. Ellory, H.-J. Lang, H. Englert, H. M. Staines, and R. J. Wilkins. 2002. Specificity of classical and putative Cl- transport inhibitors on membrane transport pathways in human erythrocytes. Cellular Physiology and Biochemistry 13: 181-188. Dattagupta, S., L. L. Miles, M. S. Barnabei, and C. R. Fisher. 2006a. The hydrocarbon seep tubeworm Lamellibrachia luymesi primarily eliminates sulfate and hydrogen ions across its roots to conserve energy and ensure sulfide supply. Journal of Experimental Biology In press. Dattagupta, S., G. Telesnicki, B. Predmore, M. Mcginley, and C. R. Fisher. 2006b. Submersible operated dialysis samplers for collecting pore water from deep-sea sediments. Limnology and Oceanography: Methods Submitted.

182 De Cian, M.-C., A. C. Anderson, X. Bailly, and F. H. Lallier. 2003a. Expression and localization of carbonic anhydrase and ATPases in the symbiotic tubeworm Riftia pachyptila. The Journal of Experimental Biology 206: 399-409. De Cian, M.-C. and others 2003b. Isolated bacteriocyte cell suspensions from the hydrothermal-vent tubeworm Rifita pachyptila, a potent tool for cellular physiology in a chemoautotrophic symbiosis. Marine Biology 142: 141-151. De Cian, M.-C., X. Bailly, J. Morales, J.-M. Strub, A. Van Dorsselaer, and F. H. Lallier. 2003c. Characterization of carbonic anhydrases from Riftia pachyptila, a symbiotic invertebrate from deep-sea hydrothermal vents. Proteins: Structure, Function, and Genetics 51: 327-339. Felbeck, H., and J. Jarchow. 1998. Carbon Release from Purified Chemoautotrophic bacterial Symbionts of the Hydrothermal Vent Tubeworm Riftia pachyptila. Physiological Zoology 71: 294-302. Fisher, C., and J. Childress. 1992. Organic Carbon Transfer from Methanotrophic Symbionts to the Host Hydrocarbon-seep Mussel. Symbiosis 12: 221-235. Fisher, C. R. 1990. Chemoautotrophic and Methanotrophic Symbioses in Marine Invertebrates. Aquatic Sciences 2: 399-436. ---. 1995. Toward an Appreciation of Hydrothermal-Vent Animals: Their Environment, Physiological Ecology, and Tissue Stable Isotope Values. Geophysical Monograph 91: 297-316. ---. 1996. Ecophysiology of Primary Production at Deep-sea Vents and Seeps. Biosystematics and Ecology Series 11: 313-336. Fisher, C. R., J. J. Childress, and E. Minnich. 1989. Autotrophic carbon fixation by the chemoautotrophic symbionts of Riftia pachyptila. Biological Bulletin 177: 372- 385. Fisher, C. R., J. J. Childress, and N. K. Sanders. 1988. The role of vestimentiferan hemoglobin in providing an environment suitable for chemoautotrophic sulfide- oxidizing endosymbionts. Symbiosis 5: 229-246. Fisher, C. R., I. A. Urcuyo, M. A. Simpkins, and E. R. Nix. 1997. Life in the slow lane: growth and longevity of cold-seep vestimentiferans. Marine Ecology 18: 83-94. Forgac, M. 1989. Structure and function of vacuolar class of ATP-driven proton pumps. Physiological Reviews 69: 765-796. Frachisse, J.-M., S. Thomine, J. Colcombet, J. Guern, and H. Barbier-Brygoo. 1999. Sulfate is both a substrate and an activator of the voltage-dependent anion channel of Arabidopsis hypocotyl cells. Plant Physiology 121: 253-261. Frazer, T. K., R. M. Ross, L. B. Quetin, and J. P. Montoya. 1997. Turnover of carbon and nitrogen during growth of larval krill, Euphausia superba : a stable isotope approach. Journal of experimental Biology and Ecology 212: 259-275. Freytag, J. K., P. R. Girguis, D. C. Bergquist, J. P. Andras, J. J. Childress, and C. R. Fisher. 2001. A paradox resolved: Sulfide acquisition by roots of seep tubeworms sustains net chemoautotrophy. Proceedings of the National Academy of Sciences 98: 13408-13413. Freytag, J. K., R. M. Smith, S. R. Breault, S. M. Hourdez, C. S. Jones, and C. R. Fisher. 2003. Sulfide binding properties of hemoglobins isolated from two species of cold seep vestimentiferans. Ph.D. Dissertation. The Pennsylvania State University.

183 Fry, B., and C. Arnold. 1982. Rapid 13C/12C turnover during growth of brown shrimp (Penaeus aztecus). Oecologia (Berlin) 54: 200-204. Furriel, R. P. M., J. C. Mcnamara, and F. A. Leone. 2001. Nitrophenylphosphatase as a tool to characterize gill Na+, K+ ATPase activity in hyperregulating Crustacea. Comparative Biochemistry and Physiology Part A 130: 665-676. Gardiner, S. L., and S. Hourdez. 2003. On the occurrence of the vestimentiferan tube worm Lamellibrachia luymesi van der Land and Nørrevang, 1975 (Annelida: Pogonophora) in hydrocarbon seep communities in the Gulf of Mexico. Proceedings of the Biological Society of Washington 116: 380-394. Gardiner, S. L., and M. L. Jones. 1993. Vestimentifera, p. 371-460. In F. W. Harrison and M. E. Rice [eds.], Microscopic anatomy of invertebrates. Wiley-Liss, Inc. Gardiner, S. L., E. R. Mcmullin, and C. R. Fisher. 2001. Seepiophila jonesi , a new genus and species of vestimentiferan tube worm (Annelida: Pogonophora) from hydrocarbon seep communities in the Gulf of Mexico. Proceedings of the Biological Society of Washington 114: 694-707. Gelband, C. H., P. G. Greco, and J. R. Martens. 1996. Voltage-dependent chloride channels: invertebrates to man. The Journal of Experimental Zoology 275: 277- 282. Gerencser, G. A., G. A. Ahearn, and M. A. Cattey. 1996. Antiport-driven sulfate secretion in an invertebrate epithilium. The Journal of Experimental Zoology 275: 269-276. ---. 1999. Sulfate/Bicarbonate antiport by lobster hepatopancreatic basolateral membrane vesicles. Journal of Experimental Zoology 284: 158-167. Girguis, P. R., J. J. Childress, J. K. Freytag, K. Klose, and R. Stuber. 2002. Effects of metabolite uptake on proton-equivalent elimination by two species of deep-sea vestimentiferan tubeworm, Riftia pachyptila and Lamellibrachia luymesi : proton elimination is a necessary adaptation to sulfide-oxidizing chemoautotrophic symbionts. The Journal of Experimental Biology 205: 3055-3066. Goffredi, S. K., and J. J. Childress. 2001. Activity and inhibitor sensitivity of ATPases in the hydrothermal vent tubeworm Riftia pachyptila: a comparative approach. Marine Biology 138: 259-265. Goffredi, S. K., J. J. Childress, N. T. Desauliniers, and F. H. Lallier. 1997a. Sulfide Acquisition by the vent worm Riftia pachyptila appears to be via uptake of HS-, rather than H2S. The Journal of Experimental Biology 200: 2609-2616. Goffredi, S. K., J. J. Childress, N. T. Desauliniers, R. W. Lee, F. H. Lallier, and D. Hammond. 1997b. Inorganic Carbon Acquisition by the Hydrothermal Vent Tubeworm Riftia pachyptila Depends Upon High External PCO2 and Upon Proton-Equivalent ion Transport by the Worm. The Journal of Experimental Biology 200: 883-896. Goffredi, S. K., J. J. Childress, F. H. Lallier, and N. T. Desaulniers. 1998. How to be a perfect host: CO2 and HS- accumulation and H+ elimination in the hydrothermal vent tube-worm Riftia pachyptila. Cahiers de Biologie Marine 39: 297-300. ---. 1999a. The ionic composition of the hydrothermal vent tubeworm Riftia pachyptila: 2- + - - Evidence for the elimination of SO4 and H and for a Cl / HCO3 shift. Physiological and Biochemical Zoology 72: 296-306.

184 Goffredi, S. K., P. R. Girguis, J. J. Childress, and N. T. Desauliniers. 1999b. Physiological Functioning of Carbonic Anhydrase in the Hydrothermal Vent Tubeworm Riftia Pachyptila. Biological Bulletin 196: 257-264. Grigg, N. J. 1999. Pore-water convection induced by peeper emplacement in saline sediment. Limnology and Oceanography 44: 425-430. Habicht, K. S., and D. E. Canfield. 2001. Isotope fractionation by sulfate-reducing natural populations and isotopic compostion of sulfide in marine sediments. Geology 29: 555-558. Hand, S. C. 1987. Trophosome ultrastructure and the characterization of isolated bacteriocytes from invertebrate-sulfur bacteria symbioses. Biological Bulletin 173: 260-276. Hand, S. C., and G. N. Somero. 1983. Energy metabolism pathways of hydrothermal vent animals: adaptations to a food-rich and sulfide-rich deep-sea environment. Biological Bulletin 165: 167-181. Hashimoto, J., S. Ohta, K. Fujikura, Y. Fujiwara, and S. Sukizaki. 1995. Life habit of Vesicomyid clam, Calyptogena soyoae, and hydrogen sulfide concentration in interstitial waters in Sagami Bay, Japan. Journal of Oceanography 51: 341-350. Herring, P. 2003. The Biology of the deep ocean. Oxford University Press. Herzka, S. Z., and G. J. Holt. 2000. Changes in isotopic composition of red drum (Sciaenops ocellatus) larvae in response to dietary shifts: potential applications to settlement studies. Canadian Journal of Fisheries and Aquatic Sciences 57: 137- 147. Herzka, S. Z., S. A. Holt, and G. J. Holt. 2001. Documenting the settlement history of individual fish larvae using stable isotope ratios: model development and validation. Journal of Experimental Marine Biology and Ecology 265: 49-74. Hesslein, R. H. 1976. An in situ sampler for close interval pore water studies. Limnology and Oceanography 21: 912-914. Hesslein, R. H., K. A. Hallard, and P. Ramlal. 1993. Replacement of sulfur, carbon and nitrogen in tissue of growing broad whitefish (Coregonus nasus) in response to a change in diet traced by delta-34-S, delta-13-C and delta-15-N. Canadian Journal of Fisheries and Aquatic Sciences 50: 2071-2076. Hille, B. 1992. Ionic channels of excitable membranes, Second ed. Sinauer Associates Inc. Hobson, K. A. 1999. Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia 120: 314-326. Holcombe, B. L., R. G. Keil, and A. H. Devol. 2001. Determination of pore-water dissolved organic carbon fluxes from Mexican margin sediments. Limnology and Oceanography 46: 298-308. Howes, B. L., and S. G. Wakeham. 1985. Effects of sampling technique on measurements of porewater constituents in salt marsh sediments. Limnology and Oceanography 30: 221-227. Hurtgen, M. T., M. A. Arthur, N. S. Suits, and A. J. Kaufman. 2002. The sulfur isotopic composition of Neoproterozoic seawater sulfate: implications for a snowball Earth? Earth and Planetary Science Letters 203: 413-429.

185 Iverson, N., and B. B. Jorgensen. 1993. Diffusion coefficients of sulfate and methane in marine sediments: influence of porosity. Geochimica et Cosmochimica Acta 57: 571-578. Jenyon, M. K. 1986. Salt tectonics. Elsevier Applied Science Publishers. Johnston, D. T., J. Farquhar, B. A. Wing, A. J. Kaufman, D. E. Canfield, and K. S. Habicht. 2005. Multiple sulfur isotope fractionations in biological systems: a case study with sulfate reducers and sulfur disproportionators. American Journal of Science 305: 645-660. Jones, M. L. 1981. Riftia pachyptila, new genus, new species, the vestimentiferan tubeworm from the Galapagos Rift geothermal vents. Proceedings of the Biological Society of Washington 93: 1295-1313. Joye, S. B. and others 2004. The anaerobic oxidation of methane and sulfate reduction in sediments from Gulf of Mexico cold seeps. Chemical Geology 205: 219-238. Julian, D., F. Gaill, E. Wood, A. Arp, and C. Fisher. 1999. Roots as a site of hydrogen sulfide uptake in the Hydrocarbon seep Vestimentiferan Lamellibrachia sp. Journal of Experimental Biology 202: 2245-2257. Kennicutt Ii, M. C., and J. M. Brooks. 1990. Recognition of areas effected by petroleum seepage: Northern Gulf of Mexico continental slope. Geo-Marine Letters 10: 221-224. Kennicutt Ii, M. C., R. A. Burke Jr, I. R. Macdonald, J. M. Brooks, G. J. Denoux, and S. A. Macko. 1992a. Stable isotope partitioning in seep and vent organisms: chemical and ecological significance. Chemical Geology (Isotope Geoscience Section) 101: 293-310. Kennicutt Ii, M. C., I. R. Macdonald, P. A. Comet, G. J. Denoux, and J. M. Brooks. 1992b. The origins of petroleum in the northern Gulf of Mexico. Geochimica et Cosmochimica Acta 56: 1259-1280. Kochevar, R. E., N. S. Govind, and J. J. Childress. 1993. Identification and characterization of two carbonic anhydrases from the hydrothermal vent tubeworm Riftia pachyptila Jones. Molecular Marine Biology and Biotechnology 2: 10-19. Krom, M. D., P. Davison, H. Zhang, and W. Davidson. 1994. High-resolution pore-water sampling with a gel sampler. Limnology and Oceanography 39: 1967-1972. Krug, P. J., and A. E. Manzi. 1999. Waterborne and surface-associated carbohydrates as settlement cues for larvae of the specialist marine herbivore Alderia modesta. Biological Bulletin 197: 94-103. Laforce, M. J., C. M. Hansel, and S. Fendorf. 2000. Constructing simple wetland sampling devices. Soil Science Society of America Journal 64: 809-811. Lee, R. W., and J. J. Childress. 1995. Assimilation of inorganic nitrogen by seep mytilid Ia, an undescribed deep-sea mussel containing methanotrophic endosymbionts: fate of assimilated nitrogen and the relation between methane and nitrogen assimilation. Marine Ecology Progress Series 123: 137-148. ---. 1996. Inorganic N assimilation and ammonium pools in a deep-sea mussel containing methanotrophic endosymbionts. Biological Bulletin 190: 373-384. Lee, R. W., E. V. Thuesen, J. J. Childress, and C. R. Fisher. 1992. Ammonium and free amino acid uptake by a deep-sea mussel (Bathymodiolus sp., undescribed) containing methanotrophic bacterial symbionts. Marine Biology 113: 99-106.

186 Leustek, T., and K. Saito. 1999. Sulfate transport and assimilation in plants. Plant Physiology 120: 637-643. Levin, L. A. 2005. Ecology of cold seep sediments: interactions of fauna with flow, chemistry and microbes. Oceanography and Marine Biology: An Annual Review 43: 1-46. Li, Y.-H., and S. Gregory. 1974. Diffusion of ions in seawater and deep-sea sediments. Geochimica et Cosmochimica Acta 38: 703-714. Lin, H., and D. J. Randall. 1993. H+-ATPase activity in crude homogenates of fish gill tissue: inhibitor sensitivity and environmental and hormonal regulation. Journal of Experimental Biology 180: 163-174. Lin, S., and J. W. Morse. 1991. Sulfate reduction and iron sulfide mineral formation in Gulf of Mexico anoxic sediments. The American Journal of Science 291: 55-89. Lodish, H., A. Berk, L. S. Zipursky, P. Matsudaira, D. Baltimore, and J. Darnell. 2000. Molecular Cell Biology, Fourth ed. WH Freeman and Company. Lombardo, F., R. Maramaldo, B. Fratello, and D. Sonetti. 1992. Amino-acids and derivatives as food-finding signals in the fresh-water snail Planorbarius corneus (L). Comparative Biochemistry and Physiology C-Pharmacology, Toxicilogy & Endocrinology 101: 389-398. Lorrain, A., P. Yves-Marie, L. Chauvaud, N. Savoye, A. Donval, and C. Saout. 2002. Differential delta-13-C and delta-15-N signatures among scallop tissues: implications for ecology and physiology. Journal of Experimental Marine Biology and Ecology 275: 47-61. Luley, K. E. 2006. Microbial processes in hydrocarbon seep sediments and their impact on chemosynthetic tubeworm aggregations. Masters in Science. The Pennslvania State University. Lutz, R. A. and others 1994. Rapid growth at deep-sea vents. Nature 371: 663-664. Macavoy, S. E., S. A. Macko, and G. C. Garman. 2001. Isotopic turnover in aquatic predators: quantifying the exploitation of migratory prey. Canadian Journal of Fisheries and Aquatic Sciences 58: 923-932. Macdonald, I. R. 1998. Habitat formation at Gulf of Mexico hydrocarbon seeps. Cah. Biol. Mar. 39: 337-340. Macdonald, I. R., G. S. Boland, J. S. Baker, J. M. Brooks, M. C. Kennicutt Ii, and R. R. Bidigare. 1989. Gulf of Mexico hydrocarbon seep communities. II. Spatial distribution of seep organisms and hydrocarbons at Bush Hill. Marine Biology 101: 235-247. Macdonald, I. R., and C. R. Fisher. 1996. Life without light, p. 87-97. National Geographic. Macdonald, I. R., N. L. Guinasso Jr., J. F. Reilly, J. M. Brooks, W. R. Callender, and S. G. Gabrielle. 1990a. Gulf of Mexico Hydrocarbon Seep Communities: VI. Patterns in Community Structure and Habitat. Geo-Marine Letters 10: 244-252. Macdonald, I. R. and others 1990b. Chemosynthetic mussels at a brine-filled pockmark in the Northern Gulf of Mexico. Science 248: 1096-1099. Markovich, D. 2001. Physiological roles and regulation of mammalian sulfate transporters. Physiological Reviews 81: 1500-1533.

187 Mcgookey, D. P. 1975. Gulf coast Cenozoic sediments and structures: and excellent example of extra-continental sedimentation. Gulf Coast Association of Geological Societies Transactions 25: 104-120. Mcmullin, E. R., S. Hourdez, S. W. Schaeffer, and C. R. Fisher. 2003. Phylogeny and biogeography of deep-sea vestimentiferan tubeworms and their bacterial symbionts. Symbiosis 34: 1-41. Miura, T., M. Nedachi, and J. Hashimoto. 2002. Sulfur sources for chemoautotrophic nutrition of shallow water vestimentiferan tubeworms in Kagoshima Bay. Journal of Marine Biological Association of U.K. 82: 537-540. Morse, J. W., and Q. Wang. 1997. Pyrite formation under conditions approximating those in anoxic sediments: II. Influence of precursor iron minerals and organic matter. Marine Chemistry 57: 187-193. Møller, J. V., B. Juul, and M. Lemaire. 1996. Structural organization, ion tranport, and energy transduction of P-type ATPases. Biochimica et Biophysica Acta 1286: 1- 51. Munujos, P., J. Coll-Canti, F. Gonzalez-Sastre, and F. Javier Gella. 1993. Assay of succinate dehydrogenase activity by a colorimetric-continous method using iodonitrotetrazolium chloride as electron accpetor. Analytical Biochemistry 212: 506-509. Nelson, N. 1992. Evolution of organellar proton-ATPases. Biochimica et Biophysica Acta 1100: 109-124. Neter, J., M. H. Kutner, C. J. Nachtsheim, and W. Wasserman. 1996. Applied linear statistical models, Fourth ed. WCB McGraw-Hill. Nicoletti, I., G. Migliorati, M. C. Pagliacci, F. Grignani, and C. Riccardi. 1991. A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. Journal of Immunological Methods 139: 271-279. Nikolaus, R., J. W. Ammerman, and I. R. Macdonald. 2003. Distinct pigmentation and trophic modes in Beggiatoa from hydrocarbon seeps in the Gulf of Mexico. Aquatic Microbial Ecology 32: 85-93. Nix, E. R., C. R. Fisher, J. Vodenichar, and K. M. Scott. 1995. Physiological ecology of a mussel with methanotrophic endosymbionts at three hydrocarbon seep sites in the Gulf of Mexico. Marine Biology 122: 605- 617. Novak, J. M., K. C. Stone, A. A. Szogi, D. W. Watts, and M. H. Johnson. 2004. Wetlands and aquatic processes. Dissolved phosphorus retention and release from a coastal plain in-stream wetland. Journal of Environmental Quality 33: 394-401. Okabe, S., P. H. Nielsen, and W. G. Characklis. 1992. Factors affecting microbial sulfate reduction by Desulfovibrio desulfuricans in continuous culture: limiting nutrients and sulfide concentration. Biotechnology and Bioengineering 40: 725-734. Orcutt, B., A. Boetius, M. Elvert, V. Samarkin, and S. B. Joye. 2005. Molecular biogeochemistry of sulfate reduction, methanogenesis and the anaerobic oxidation of methane at Gulf of Mexico cold seeps. Geochimica et Cosmochimica Acta 69: 4267-4281. Orphan, V. J. and others 2001. Comparative analysis of methane-oxidizing archaea and sulfate-reducing bacteria in anoxic marine sediments. Applied and Environmental Microbiology 2001: 1922-1934.

188 Page, H. M., C. R. Fisher, and J. J. Childress. 1990. Role of filter-feeding in the nutritional biology of a deep-sea mussel with methanotrophic symbionts. Marine Biology 104: 251-257. Paull, C. K. and others 1984. Biological communities at the Florida Escarpment resemble hydrothermal vent taxa. Science 226: 965-967. Pedersen, P. L. 1982. H+-ATPases in biological systems: an overview of their function. The New York Academy of Sciences. Pedersen, P. L., and E. Carafoli. 1987. Ion motive ATPases. I. Ubiquity, properties and significance to cell function. Trends in Biochemical Sciences 12: 146-150. Pflugfelder, B., C. R. Fisher, and M. Bright. 2005. The color of the trophosome: elemental sulfur distribution in the endosymbionts of Riftia pachyptila (Vestimentifera; Siboglinidae). Marine Biology 146: 895-901. Pindell, J. L. 1985. Alleghenian reconstruction and subsequent evolution of the Gulf of Mexico, Bahamas, and Proto-Caribbean. Tectonics 4: 1-39. Pritchard, J. B., and J. L. Renfro. 1983. Renal sulfate transport at the basolateral membrane is mediated by anion exchange. Proceedings of the National Academy of Sciences 80: 2603-2607. Pruski, A. M., A. Fiala-Medioni, C. R. Fisher, and J. C. Colomines. 2000. Composition of free amino acids and related compounds in invertebrates with symbiotic bacteria at hydrocarbon seeps in the Gulf of Mexico. Marine Biology 136: 411- 420. Rahman, Y. J., R. B. Forward Jr., and D. Rittschof. 2000. Responses of mud snails and periwinkles to environmental odors and disaccharide mimics of fish odor. Journal of Chemical Ecology 26: 679-695. Renfro, J. L. 1999. Recent developments in teleost renal transport. Journal of Experimental Zoology 283: 653-661. Renfro, J. L., and J. B. Pritchard. 1983. Sulfate transport by flounder renal tubule brush border: presence of anion exchange. American Journal of Physiology 244: F488- F496. Renninger, G. H. and others 1995. Sulfide as a chemical stimulus for deep-sea hydrothermal vent shrimp. Biological Bulletin 189: 69-76. Rice, W. R. 1989. Analyzing tables of statistical data. Evolution 43: 223-225. Rieley, G., C. L. Van Dover, D. B. Hedrick, and G. Eglinton. 1999. Trophic ecology of Rimicaris exoculata: a combined lipid and abundance/stable isotope approach. Marine Biology 133: 495-499. Rittschof, D. 1990. Peptide-mediated behaviors in marine organisms. Journal of Chemical Ecology 16: 261-272. Rittschof, D., and J. Bonaventura. 1986. Macromolecular cues in marine systems. Journal of Chemical Ecology 12: 1013-1023. Roberts, H. H., and P. Aharon. 1994. Hydrocarbon-derived carbonate buildups of the northern Gulf of Mexico continental slope: A review of submersible investigations. Geo-Marine Letters 14: 135-148. Rueter, P. and others 1994. Anaerobic oxidation of hydrocarbons in crude oil by new types of sulphate-reducing bacteria. Nature 372: 455-457. Salierno, J. D., S. Rebach, and M. C. Christman. 2003. The effects of interspecific competition and prey odor on foraging behavior in the rock crab, Cancer

189 irroratus (Say). Journal of Experimental Marine Biology and Ecology 287: 249- 260. Sassen, R., S. Joye, S. T. Sweet, D. A. Defreitas, A. V. Milkov, and I. R. Macdonald. 1999. Thermogenic gas hydrates and hydrocarbon gases in complex chemosynthetic communities, Gulf of Mexico continental slope. Organic Geochemistry 30: 485-497. Sassen, R. and others 2001. Massive vein-filling gas hydrate: relation to ongoing gas migration from the deep subsurface in the Gulf of Mexico. Marine and Petroleum Geology 18: 551-560. ---. 1994. Organic geochemistry of sediments from chemosynthetic communities, Gulf of Mexico slope. Geo-Marine Letters 14: 110-119. Sassen, R., H. H. Roberts, P. Aharon, J. Larkin, E. W. Chinn, and R. S. Carney. 1993. Chemosynthetic bacterial mats at cold hydrocarbon seeps, Gulf of Mexico continental slope. Organic Geochemistry 20: 77-89. Scott, K. M., M. Bright, S. A. Macko, and C. R. Fisher. 1999. Carbon dioxide use by chemoautotrophic endosymbionts of hydrothermal vent vestimentiferans: affinities of carbon dioxide, absence of carboxysomes, and delta-13-C values. Marine Biology 135: 25-34. Shannon, R. D., and J. R. White. 1996. The effects of spatial and temporal variations in acetate and sulfate on methane cycling in two Michigan ponds. Limnology and Oceanography 41: 435-443. Shimuzu, M., and T. J. Bradley. 1994. Concentrated excretion of sulfate by the anterior colon of the brine fly, Ephydra hians. Physiological Zoology 67: 54-67. Sibuet, M., and K. Olu. 1998. Biogeography, biodiversity and fluid dependence of deep- sea cold-seep communities at active and passive margins. Deep-Sea Research II 45: 517-567. Smith, E. B., K. M. Scott, E. R. Nix, C. Korte, and C. R. Fisher. 2000a. Growth and condition of seep mussels (Bathymodiolus childressi) at a Gulf of Mexico Brine Pool. Ecology 81: 2392-2403. ---. 2000b. Growth and condition of sep mussels (Bathymodiolus childressi) at a Gulf of Mexico Brine Pool. Ecology 81: 2392-2403. Smith Jr, K. L. 1985. Deep-sea hydrothermal vent mussels: nutritional state and distribution at the Galapagos Rift. Ecology 66: 1067-1080. Sten-Knudsen, O. 1978. Passive transport processes, p. 5-114. In G. Giebisch, D. Tosten, H. Ussing and M. Tosteson [eds.], Membrane transport in biology. Springer- Verlag. Streams, M. E., C. R. Fisher, and A. Fiala-Medioni. 1997. Methanotrophic symbiont location and fate of carbon incorporated from methane in a hydrocarbon seep mussel. Marine Biology 129: 465-476. Stumm, W., and J. J. Morgan. 2000. Aquatic chemistry. John Wiley and Sons. Taviani, M. 1994. The "calcari a Lucina" macrofauna reconsidered: deep-sea faunal oases from Miocene-age cold vents in the Romagna Appennine, Italy. Geo-Marine Letters 14: 185-191. Taylor, A. R., and C. Brownlee. 2003. A novel Cl- inward-rectifying current in the plasma membrane of the calcifying marine phytoplankton Coccolithus pelagicus. Plant Physiology 131: 1391-1400.

190 Thauer, R. K., K. Jungermann, and K. Decker. 1977. Energy conservation in chemotrophic anaerobic bacteria. Bacteriological Reviews 41: 100-180. Tieszen, L. L., T. W. Boutton, K. G. Tesdahl, and N. A. Slade. 1983. Fractionation and turnover of stable carbon isotopes in animal tissues: implications for δ13C analysis of diet. Oecologia (Berlin) 57: 32-37. Torres, M. E. and others 2002. Fluid and chemical fluxes in and out of sediments hosting methane hydrate deposits on Hydrate Ridge, OR I: hydrological provinces. Earth and Planetary Science Letters 201: 525-540. Treude, T., A. Boetius, K. Knittel, K. Wallmann, and B. B. Jorgensen. 2003. Anaerobic oxidation of methane gas above gas hydrates at Hydrate Ridge, NE Pacific Ocean. Marine Ecology Progress Series 264: 1-14. Tyron, M. D., and K. M. Brown. 2004. Fluid and chemical cylcing at Bush Hill: Implications for gas-and hydrate-rich environments. Geochemistry Geophysics Geosystems 5: 1-7. Urban, N. R., C. Dinkel, and B. Wehrli. 1997. Solute transfer across the sediment surface of a eutrophic lake: I. Porewater profiles from dialysis samplers. Aquatic Sciences 59: 1-25. Valentine, D. L. 2002. Biogeochemistry and microbial ecology of methane oxidation in anoxic environments: a review. Antonie van Leeuwenhoek 81: 271-282. Van Der Land, J., and A. Nørrevang. 1977. Structure and relationships of Lamellibrachia (Annelida, Vestimentifera). Kiøbenhavske Selskab af Laerdoms og Videnskabers Elskere 21: 1-102. Van Winkle, L. J. 1999. Biomembrane transport. Academic Press. Vetter, R. D., and B. Fry. 1998. Sulfur contents and sulfur isotopic compositions of thiotrophic symbioses in bivalve molluscs and vestimentiferan worms. Marine Biology 132: 453-460. Walford, L. A. 1946. A new graphic method of describing the growth of animals. Biological Bulletin 90: 141-147. Ward, J. M. 1997. Patch-clamping and other molecular approaches for the study of plasma membrane transporters demystified. Plant Physiology 114: 1151-1159. Weber, A., and B. B. Jorgensen. 2002. Bacterial sulfate reduction in hydrothermal sediments of the Guaymas Basin, Gulf of California, Mexico. Deep-Sea Research I 49: 827-841. Webster, I. T., P. R. Teasdale, and N. J. Grigg. 1998. Theoretical and experimental analysis of peeper equilibration dynamics. Environmental Science and Technology 32: 1727-1733. Welsh, S. E., and M. Inoue. 2000. Loop current rings and the deep circulation in the Gulf of Mexico. Journal of Geophysical Research 185: 16951-16959. Weston, N. B., W. Porubsky, V. A. Samarkin, M. Erickson, S. E. Macavoy, and S. B. Joye. 2006. Porewater stoichiometry of terminal metabolic products, sulfate, and dissolved organic carbon and nitrogen in estuarine intertidal creek-bank sediments. Biogeochemistry 77: 375-408. Widdel, F., and R. Rabus. 2001. Anaerobic biodegradation of saturated and aromatic hydrocarbons. Current Opinion in Biotechnology 12: 259-276.

191 Wieczorek, H., G. Gruber, W. R. Harvey, M. Huss, H. Merzendorfer, and W. Zeiske. 2000. Structure and regulation of insect plasma membrane H+ V-ATPase. The Journal of Experimental Biology 203: 127-135. Yildiz, F. H., J. P. Davies, and A. R. Grossman. 1994. Characterization of sulfate transport in Chlamydomonas reinhardtii during sulfur-limited and sulfur- sufficient growth. Plant Physiology 104: 981-987. Zal, F., F. H. Lallier, J. S. Wall, S. N. Vinogradov, and A. Toulmond. 1996. The multi- hemoglobin system of the hydrothermal vent tube worm Riftia pachyptila. The Journal of Biological Chemistry 15: 8869-8874. Zal, F., E. Leize, F. H. Lallier, A. Toulmond, A. Van Dorsselaer, and J. J. Childress. 1998. S-Sulfohemoglobin and disulfide exchange: The mechanisms of sulfide binding by Riftia pachyptila hemoglobins. Proceedings of the National Academy of Sciences 95: 8997-9002. Zande, J. M., and R. S. Carney. 2001. Population size structure and feeding biology of Bathynerita naticoidea Clarke 1989 (Gastropoda:Neritacea) from Gulf of Mexico hydrocarbon seeps. Gulf of Mexico Science 2: 107-118. Zhang, C. L. and others 2005. Lipid biomarkers and carbon isotope signatures of a microbial (Beggiatoa) mat associated with gas hydrates in the Gulf of Mexico. Applied and Environmental Microbiology 71: 2106-2112. Zimmer, R. K., and C. A. Butman. 2000. Chemical signalling processes in the marine environment. Biological Bulletin 198: 168-187.

192 Curriculum Vitae Sharmishtha Dattagupta E-mail: [email protected] Education 2000- present The Pennsylvania State University Ph.D. Biology 1997- 1999 The Indian Institute of Technology, Mumbai, India M.S. Biotechnology 1994-1999 St. Stephen’s College, Delhi, India B.S. Chemistry (Honors)

Publications • Dattagupta, S., Miles, L. L., Barnabei, M. S., and Fisher, C. R. (2006) The hydrocarbon seep tubeworm Lamellibrachia luymesi primarily eliminates sulfate and hydrogen ions across its roots to conserve energy and ensure sulfide supply. In press. The Journal of Experimental Biology. • Dattagupta, S., Bergquist, D. C., Smith, E. B., Macko, S. E., Fisher, C. R. (2004) Tissue carbon, nitrogen, and sulfur stable isotope turnover in transplanted Bathymodiolus childressi mussels: Relation to growth and physiological condition. Limnology and Oceanography. 49 (4); 1144-1151. • Kumar, M., Dattagupta, S., Kannan, K. K., Hosur, M. V. (1999) Purification, crystallization and preliminary X-ray diffraction study of ribosome inactivating protein: saporin. Biochimica Biophysica Acta. 1429:506-511. • Dattagupta, S., Martin, J., Liao, S-M, Carney, R. S., and Fisher, C. R. (2006) Deep-sea hydrocarbon seep gastropod Bathynerita naticoidea responds to cues from the habitat-providing mussel Bathymodiolus childressi. Accepted by Marine Ecology • Dattagupta, S., Telesnicki, G., Predmore, B. L., McGinley, M. P. and Fisher, C. R. Submersible operated dialysis samplers for collecting pore water from deep-sea sediments. Submitted to Limnology and Oceanography: Methods.

Presentations March 2006. Talk “Do seep tubeworms release sulfate and protons across their roots?” at the 9th Annual Environmental Chemistry Student Symposium, Penn State University. Best Speaker Award. March 2006. Guest Lecture “Sulfate and hydrogen ion elimination across roots of long-lived hydrocarbon seep tubeworms: a mechanism to conserve energy and ensure sulfide supply?” at University of Southern California, Los Angeles, California. September 2005. Talk “Do seep tubeworms release sulfate and protons across their roots?” at the 3rd International Symposium on Deep-Sea Hydrothermal Vent and Seep Biology, San Diego, California. May 2004. Poster. “The possibility of syntrophy between hydrocarbon seep tubeworms and subsurface sulfate-reducing bacteria” at the 2004 Ridge2000- Inter-Ridge Joint Theoretical Institute Short Course and Workshop, Jeju, South Korea. Best Poster Award. March 2004. Poster. “Development of a new sampling device to obtain concentration profiles in deep-sea sediments underlying tubeworm communities” at the 7th Annual Environmental Chemistry Student Symposium, Penn State University. December 2003. Guest lecture. “From hot vents to cold seeps- Research at Dr. Chuck Fisher’s Lab, Penn State University” at National Institute of Oceanography, Dona Paula, Goa, India. February 2003. Invited talk. “Change in tissue stable isotope compositions of transplanted hydrocarbon seep mussels” at Aquatic Sciences Meeting (ASLO), Salt Lake City, Utah. February 2002. Poster. “Change in tissue stable isotope compositions of transplanted hydrocarbon seep mussels” at PSU Graduate Student Exhibition. Third Prize. October 2001. Poster. “Change in tissue stable isotope compositions of transplanted hydrocarbon seep mussels: Influence of growth and environment” at 2nd International Symposium on Deep-Sea Hydrothermal Vent Biology, Brest, France.

Fellowships/ Awards 2006 Best speaker award, 9th Annual Environmental Chemistry Student Symposium 2006 Special Recognition: Women in Science and Engineering (WISE), Penn State University. 2005 CECG (The Center for Environmental Chemistry and Geochemistry) Summer Internship 2004 Inter-RIDGE Outstanding Student Poster Award – First Place 2002 Third place for poster competition, Graduate Student Exhibition, Penn State University 2000- 2001 Braddock Fellowship, The Pennsylvania State University 1999 Institute Silver Medal, Indian Institute of Technology, Mumbai 1996-97,97-98 Summer Research Fellowships, Jawaharlal Nehru Center for Advanced Scientific Research, India 1994-95, 95-96 Delhi University Merit Scholarship

Teaching Experience Guest Lecturer The Pennsylvania State University Bio446: Physiological Ecology Spring 2006 Topic: Physiology of deep-sea hydrothermal vent tubeworms Teaching assistant The Pennsylvania State University Bio406: Symbiosis Spring 2006 (Taught 5 guest lectures) Teaching assistant The Pennsylvania State University Bio230W: Cell and Molecular Biology, Fall 2001, 2002, 2004 Guest Lecturer The Pennsylvania State University Bio406: Symbiosis April 13, 2005 (Topic: Mammalian foregut symbioses)