THESIS 1 LE“ M $3? 3 ‘3‘ «fine I?" 1'," Ebtate lulu"mg”unullllllllllnl L “a y A METABOLISM OF FERMENTATION INTERMEDIATES IN LAKE SEDIMENTS presented by Derek Reen Lovley has been accepted towards fulfillment of the requirements for Ph . D . degree in Microbiology ////////§<< Major professor) M Date' 7" {7’ £2 MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LIBRARIES “ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. U! . -‘ ‘ nan W E METABOLISM OF FERMENTATION INTERMEDIATES IN LAKE SEDIMENTS BY Derek Reen Lovley A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1982 ABSTRACT METABOLISM OF FERMENTATION INTERMEDIATES IN LAKE SEDIMENTS BY Derek Reen Lovley The metabolism of fermentation intermediates in anaerobic lake sediments was examined. l4C-labelled compounds and inhibitor techniques were used to determine the rates of production and fates of fermentation intermediates in the surface sediments of a eutrophic lake. Acetate and hydrogen were the primary fermentation intermediates with lesser production of propionate, butyrates, valerates, and lactate. The fermentation pattern was similar to those previously described for anaerobic wastes and the rumen but differed markedly from the results previously reported for other lake sediments. Propionate and lactate were metabolized to acetate. The metabolism of propionate, butyrates, and valerates was dependent upon the maintenance of low hydrogen partial pressures by methanogens which were the primary consumers of hydrogen and acetate. The competition between methanogens and sulfate reducers for hydrogen and acetate was studied using kinetic analyses of the uptake of electron donors, inhibitor experiments, and simulation modelling. When sulfate was not limiting sulfate reducers had a higher affinity for hydrogen uptake which permitted them to maintain the hydrogen partial pressure below threshold levels necessary for methanogenesis. However, methanogens outcompeted sulfate reducers for hydrogen in eutrOphic lake sediments since the maximum potential for sulfate reduction was limited by sulfate availability to a value that was much lower than the rate of hydrogen production. In oligotrophic lake sediments sulfate reducers were able to compete effectively with methanogens as evidenced by an increase in methane production from both hydrogen and acetate following the inhibition of sulfate reduction. Sulfate reducers had a higher affinity for acetate at £2 Igi£2_sulfate concentrations. A comparison of the partitioning of terminal carbon and electron flow between sulfate reducers and methanogens in the sediments of lakes of different trophic status as well as simulation modelling of the competition between sulfate reducers and methanogens demonstrated that the relative importance of methane production and sulfate reduction in consuming hydrogen and acetate is primarily controlled by the rate of organic matter decomposition in the sediments. ACKNOWLEDGMENTS I would like to thank Mike Klug for giving me the opportunity to work in his lab. His "can do" attitude, helpful suggestions and constructive criticism were responsible for much of the success of this project. I am grateful to Gary King, Dick Smith and Joe Robinson for their helpful discussions and advice. Jim Tiedje made many important contributions especially in discussions on competitive interactions. Daryl Dwyer provided excellent technical assistance in performing the hydrogen kinetic analysis experiments. The assistance of Ronda Snider, Greg Walker and the many others both at the station and on campus is gratefully acknowledged. Most of all I would like to thank my parents for teaching me a love of learning and my wife, Marie, for her loving support. ii TABLE OF CONTENTS LIST OF TABLESIO0.0.00..0...0....0.00.0.0...OOOOOOOOOOOOOOOOOOO LIST OF FIGURESOOOOIOOOOOOIOOOOOOOOOOOOOOOOOOOIIOOOOOOOIOOOOOOO CHAPTER I. CHAPTER II. CHAPTER III. LITERATURE REVIEW AND EXPERIMENTAL RATIONALE...... THE THEORETICAL METHANE FERMENTATION.............. METHANOGENESIS AND FERMENTATION IN SEDIMENTS...... SULFATE REDUCTION IN FRESHWATER SEDIMENTS......... EXPERIMENTAL RATIONALE............................ LITERATURE CITED.................................. INTERMEDIARY METABOLISM OF ORGANIC MATTER IN THE SEDIMENTS OF A EIJTROPHIC IAKEOCOOOOOOOOOOOOOOO INTRODUCTION...................................... MATERIALS AND METHODS............................. Sediment sampling............................ Measurement of SCFA turnover with C-tracers.................................. Measurement of VFA production with H2, CHC13, and Na2M004 inhibition................ Analytical techniques........................ RESULTS........................................... DISCUSSION........................................ LITERATURE CITED.................................. KINETIC ANALYSIS OF COMPETITION BETWEEN SULFATE REDUCERS AND METHANOGENS FOR HYDROGEN IN SEDDTENTS...so00000000.oooooooooooooooooooooooo INTRODUCTION...................................... MATERIALS AND METHODS............................. Measurements of in situ rates................ Laboratory studies........................... Kinetic analysis............................. RESULTS........................................... DISCUSSION........................................ LITERATURE CITED.................................. iii Page v vii p—a mO‘Ule‘ 13 13 13 13 14 14 15 15 18 20 22 22 23 23 24 26 27 35 43 CHAPTER IV. CHAPTER V. FACTORS CONTROLLING THE RELATIVE IMPORTANCE OF SULFATE REDUCTION AND METHANOGENESIS IN LAKES OF VARYING TROPHIC I‘EVELS...OOOOICCIOOCOOOOICOO0.0 INTRODUCTION...................................... MATERIALS AND METHODS............................. Sediment sampling and incubation............. Radioactive tracer studies................... Analytical techniques........................ Modelling of competition..................... RESULTSOOOOOOOOOOOOI0.0...OOOOOOOOOOOOOOOOOOOOOOOO .EE situ H2 and Acetate Utilization by SRB.... Effect of total sediment metabolism on competition.................................. DISCUSSIONOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO LITERATURE CITEDOOOOOOOOOO...OOOOOOOOOOOOOOOOOOOOO SWARYOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO iv Page 46 46 47 47 49 50 50 54 54 64 69 77 8O LIST OF TABLES Table Page CHAPTER II 1 Percent distribution over time of 14C-labelled compounds produced from [2-140] propionate in surface sediments (0- to 2-cm depth) collected from grav1ty coreSOOOOOOOOOOOOOOOOIOOOOOOOOOOOOOOOO....0 16 2 Pool sizes, turnover rates, and fates of fermentation intermediates in surface sediments (0- to 2-cm depth) collected from gravity cores on 4 September l980........ 17 3 Effect of hydrogen on the metabolism of volatile fatty acids in winter surface sediments collected With an ECkman dredgE................................... 17 4 Production rates of VFAs in winter surface sediments (collected with an Eckman dredge) as determined by inhibitor experimentSoooo0000000000.000000000000000...o.18 5 Relative production of fermentation intermediates in various anaerobic ecosystems............................ 19 CHAPTER III 1 Relative importance of methane production and sulfate reduction in the surface sediments (0-2 cm) of Wintergreen Lake during summer stratification........... 28 2 Hydrogen partial pressure in sediments with and Without “ded enlfateOOOOOOOOOOOOOOOOOOOOIOOOIOO0....0.. 3O 3 Effect of chloroform on the hydrogen uptake potential of methanogens and sulfate reducers........... 34 4 Kinetic parameters for hydrogen uptake in sediments collected from the A site. Progress curves were run concurrently With those Shown in Figure 200000000000 38 Table CHAPTER IV 1 Growth parameters for the simulation model of methanogen and sulfate reducer competition for acetate in sedimentSOOOO0.00...OOOOOOOOOOOOOOOIOOOO 2 Effect of molybdate and sulfate additions on methane production in Lawrence Lake surface sediments.......... 3 Methane production and sulfate reduction in control and molybdate-treated Lawrence Lake sediments over a 20 hour inCUbation periodoooooo00.000000000000000...o 4 Effect of molybdate addition on the conversion of [Z-IACI acetate and 14C02 t0 14CH40000000000000000000o. 5 Effect of inhibitors on hydrogen uptake in freshly- collected and sulfate-depleted sediments............... 6 Methane production, sulfate reduction and related parameters in the sediments of several lakes during sumer stratificationOOOOOOOOOOOOOOOOO0.0....0.0 vi Page 52 57 58 59 61 65 Figure CHAPTER 1 CHAPTER LIST OF FIGURES Page II Disappearance of [2-14C] acetate and production of 14CH4 and 14002 in summer surface sediments (0- to 2-cm depth) collected from gravity cores........ 16 Acetate turnover rates and ratio of 14CH4 ' production to total 14CH4 and 14002 production with depth in summer sediments sampled from gravity cores.......................................... 16 Accumulation of volatile fatty acids and hydrogen after the inhibition of methanogenesis in summer surface sediments collected with an Eckman dredge...... 17 Model of carbon and electron flow in the profundal surface sediments of Wintergreen Lake. The numbers adjacent to compounds represent the molar amounts of each compound produced or metabolized for every 100 mol of the total acetate pool that is metabolized. The numbers adjacent to arrows represent the molar contribution from substrate to product. All acetate and hydrogen produced from initial fermentation and the metabolism of SCFA enter common pools as designated by the boxes. (H2) represents possible reduction of sulfate as well as molecular hydrogen production.................. 20 III Methane production rates and hydrogen partial pressure over time in sulfate-amended and control sediments collected from the A site and incubated at 20 C on a bottle roller. Arrow designates addition of molybdate to sulfate- amended sediments. Values are means of duplicate bottles of each treatment and are representive of the results obtained in several similar experiments. Symbols: H andH , methane production rates in sulfate-amended and control sediments; C - - .. andL‘ ‘ ‘A, hydrogen partial pressure in sulfate-amended and control sediments............... 31 vii Figure CHAPTER 2 CHAPTER 1 Page III (continued) Typical hydrogen uptake progress curves. Symbols: 0, sulfate-amended sediments; X, sulfate—amended sediments treated with chloroform, Q , control sediments. T, represents expected hydrogen partial pressure in sulfate-amended sediments calculated from equation 3 and the appropriate kinetic parameters as described in the text............ 36 IV Methane production, mean and range of standard error on 12 August 0 , and 13 August I , and sulfate concentration with depth on 24 August at the 12.5 m depth site of Lawrence Lake................. 55 Metabolism of [2-14C] acetate in Lawrence Lake surface sediments to 14CH4 and 14C02 at different concentrations of added acetate. Error bars represent standard error of the mean. Standard error for 14002 is contained within the area of the symbols............................................ 62 Examples of methane production, sulfate reduction, and sulfate concentration profiles from simulation model. The simulations depicted in both panels had 100 pM sulfate in the overlying water. The acetate production rates were 3 and 46 umoles per liter per hour for the sediments in panels A and B respectively........................................... 67 The percent methane production of the total of methane production and sulfate reduction in sediments. Curves represent the results from simulation modelling for the 0-2 cm depth interval with sulfate concentrations in the overlying water of 200, 150, 100, and 50 pM. Symbols for percentage methane production of total calculated from rate measurements; LR, Lawrence Lake; BR, Lake Bostalsee; WBR, Wintergreen site B; WAR, Wintergreen Lake site A. Symbols for percentage of methane production from [2-140] acetate: LA, Lawrence Lake; MA, Lake Mendota, WBA, Wintergreen Lake site B; WAA, Wintergreen Lake site A; VA, Lake Vechten.......................... 70 viii CHAPTER I Literature Review and Experimental Rationale The decomposition of sedimented organic carbon in freshwater lakes proceeds primarily through anaerobic metabolism. This is evidenced by benthic respiratory quotients (RQ - C02 released/02 taken up) that exceed unity in most lakes that have been examined (34-36). Potential electron acceptors other than oxygen such as nitrate and sulfate are often depleted from the interstitial water of sediments (30, 47, 48) and methanogenesis becomes the predominate terminal process in the sediments. A compilation of data on five North American lakes has demonstrated that methane is the ultimate fate of 36-58 percent (mean = 47) of the carbon input to eutrOphic lakes (21). The estimate of the total amount of carbon proceeding through methanogenesis in these sediments approaches or exceeds the carbon input estimates since the rate of carbon dioxide production should approximate the rate of methane production if the redox state of the initial substrate is near zero. The Theoretical Methane Fermentation: A theoretical framework for the metabolism of organic carbon in methanogenic environments has been developed based upon results from studies on the rumen, anaerobic waste digestors, enrichments, and mixed and pure cultures. The decomposition of complex organic matter proceeds through three major phases: 1) the hydrolysis and fermentation of initial substrates; 2) the metabolism of fatty acids produced from fermentation and lipid hydrolysis to acetate and hydrogen; and 3) the metabolism of acetate and hydrogen to methane (6, 24, 52). Although fermentative bacteria in pure culture produce a number of fermentation products such as succinate, ethanol and lactate these products are formed in only small amounts in natural systems containing methanogenic bacteria (16, 17, 49, 50). In methanogenic ecosystems acetate, hydrogen and carbon dioxide are the primary fermentation products with lesser amounts of propionate, butyrates, and valerate being formed. Acetate and hydrogen are the ecologically significant fermentation products in methanogenic environments because the maintenance of low hydrogen partial pressures by methanogenic bacteria makes the reoxidation of reduced nicotinamide nucleotides with the formation of hydrogen thermodynamically favorable (49). Thus fermentative bacteria can metabolize pyruvate to acetate rather than oxidizing NADH with the reduction of pyruvate, acetyl-coA, or other fermentation intermediates to form reduced fermentation end products. The ability of methanogenic bacteria to alter the fermentation pattern of fermentative bacteria in this manner has been demonstrated many times (13, 14, 18, 23, 43). Fatty acids of longer chain length than acetate and aromatic compounds are metabolized to acetate and hydrogen by proton-reducing acetogenic bacteria (4, 6, 15, 28, 29). The metabolism of fatty acids with hydrogen production is thermodynamically favorable only at low hydrogen partial pressures and is thus dependent upon hydrogen metabolism by methanogenic bacteria (6, 20, 28, 29). Fatty acids with even chain lengths are metabolized to acetate and hydrogen whereas those with odd chain lengths are metabolized to acetate and propionate (27-29). A bacterium that decomposes fatty acids in this manner, Syntrophomonas wolfei has been isolated in coculture with hydrogen-consuming methanogens or sulfate reducers (29). Prepionate is degraded by a different bacterium which metabolizes propionate to acetate, hydrogen and carbon dioxide (6, 20, 27, 41). A bacterium that metabolizes propionate to acetate, Syntrophobacter wolinii, has also been obtained in coculture with hydrogen consuming bacteria (4). In the third stage of decomposition methanogenic bacteria consume the hydrogen and acetate produced during fermentation and acetogenesis and produce methane. In anaerobic wastes acetate is the precursor of approximately seventy percent of the methane with hydrogen metabolism presumably accounting for most of the balance of methane production (27, 37). Methanogenesis and fermentation in sediments: Despite the acceptance of the decomposition scheme described above as the general pattern for non-gastrointestinal methanogenic systems there has been little data to indicate whether such a pattern is applicable to lake sediments in which methanogenesis dominates carbon flow. Previous studies have focused primarily on the precursors and rates of methanogenesis in sediments. Both hydrogen and acetate are converted to methane (3, 12, 26, 42, 46-48). The reported relative importance of the two precursors varies widely. Acetate turnover rates of 16 (48) and 18 (42) umoles per liter of sediment per hour and 33 umoles per 100 grams of wet sediment (12) have been estimated from the metabolism of [14C] acetate and hydrogen turnover has been estimated from kinetic uptake parameters (42). However the relevance of the reported rates to in_§i£2 rates is uncertain since the methods of sediment collection, storage and incubation in these studies did not mdmic in_§i£g_conditions. The pathways for the metabolism of initial substrates to methanogenic precursors in sediments has only been examined superficially. The effect of addition of various potential fermentation intermediates on methane production has been examined (26, 42, 46). However, such studies are inconclusive since a stimulation of methane production does not mean that the intermediate is important in_§i£u and a lack of stimulation may indicate that the turnover of that compound is near saturation ig_situ rather than that it is an unimportant intermediate. However, it has been convincingly demonstrated that the methanogenic bacteria in lake sediments have a high affinity for hydrogen and are capable of maintaining low hydrogen partial pressures in sediments (42, 46). This evidence suggests that a pattern of carbon flow similar to that preposed for other anaerobic ecosystems is possible in sediments. The only studies on the turnover of fermentation intermediates other than acetate and hydrogen have indicated that carbon flow in anaerobic lake sediments is quite different than that in other anaerobic ecosystems. In a series of reports on the anaerobic metabolism in the sediments of Lake Vecten, Cappenberg and co-workers (7-12) reported that lactate is an important intermediate in carbon flow with a turnover rate 10-fold greater than acetate. As outlined previously, lactate production would be expected to be minimal in environments with low hydrogen partial pressures. Since the acetate turnover rate is only a fraction of the lactate turnover rate in Lake Vechten sediments, most of the carbon proceeding through the lactate pool must not enter the acetate pool. This is contrary to the expected metabolism of lactate to acetate and hydrogen under low hydrogen partial pressure (13). Sulfate reduction in freshwater sediments: Although methanogenesis has been reported to be the predominate terminal process in freshwater sediments in which the overlying water is oxygen-depleted sulfate reduction also occurs (19, 38, 39, 51). Carbon budget calculations indicate that sulfate reduction may be the terminal process for about twenty-five percent of the carbon metabolism in sediments (19, 38). However, these calculations have not been based on simultaneous measurements of methane production and sulfate reduction. Pure cultures of sulfate reducing bacteria can use short chain fatty acids and hydrogen as electron donors (5, 22, 44, 45) and the ability of sulfate reducing bacteria to use these substrates in marine sediments has been demonstrated (2, 33, 40). Sulfate reducers prevent methane production from hydrogen and acetate in marine sediments. Although the mechanisms have not been demonstrated it has been suggested that sulfate reducers outcompete methanogens for these substrates (1, 2, 25, 33, 40, 47). Sulfate reducers in freshwater sediments have the ability to outcompete methanogens for hydrogen and acetate when sulfate is added to the sediments (47). However, most evidence suggests that sulfate reducers lack the ability to compete effectively with methanogens at freshwater sulfate concentrations. Additions of hydrogen stimulates sulfate reduction (19, 39). However, the partial pressures of hydrogen that were added were approximately 100,000 times greater than the in EEEE partial pressure. When sulfate reduction is inhibited in freshwater sediments there is no stimulation of methane production (39) suggesting that sulfate reducers are not metabolizing a significant fraction of hydrogen production in_§i£2. In the same study, inhibition of sulfate reduction partially inhibited acetate oxidation. Since no corresponding increases in methane production were observed the significance of acetate metabolism by sulfate reducer in freshwater sediments is uncertain. Previous studies on sulfate reduction have focused on eutrophic lake sediments which have a high electron acceptor demand and may rapidly become sulfate depleted when samples are removed from the sulfate input from the overlying water. A significant portion of the metabolism in sediments of less productive lakes is also anaerobic (35) and the metabolism in these sediments should also be examined. Experimental rationale: The purpose of the studies reported here was to determine the pathways and controls of carbon and electron flow involved in the production of fermentation products and terminal metabolism in anaerobic freshwater sediments. The production rates of fermentation intermediates were estimated from the turnover rate of 14C-labelled compounds injected into sediments maintained under conditions approximating those in situ and from the accumulation of fermentation products when the metabolism of fermentation products was inhibited. The finding that acetate and hydrogen were the primary fermentation intermediates in eutrOphic sediments led to an examination of the factors that affect the fate of these intermediates in sediments. Since sulfate reducers and methanogens were expected to be the primary consumers of acetate and hydrogen in anaerobic, nitrate-depleted sediment, the factors controlling the competition between sulfate reducers and methanogens for hydrogen and acetate were studied. 10. LITERATURE CITED Abram, J. W., and D. B. Nedwell. 1978. Inhibition of methanogenesis by sulphate reducing bacteria competing for transferred hydrogen. Arch. Microbiol. 117:89-92. Abram, J. W., and D. B. Nedwell. 1978. Hydrogen as a substrate for methanogenesis and sulphate reduction in anaerobic saltmarsh sediments. Arch. Microbiol. 117:93-97. Belyaev, S. 8., Z. I. Finkelstein, and M. V. Ivanov. 1975. Intensity of bacterial methane formation in ooze deposits of certain lakes. Microbiology_44:272-275. Boone, D. R., and M. P. Bryant. 1980. Propionate-degrading bacterium, Syntrophobacter wolinii sp. nov. gen. nov. from methanogenic ecosystems. Appl. Environ. Microbiol..40:626-632. Brandis, A., and R. Thauer. 1981. Growth of Desulfovibrio species on hydrogen and sulfate as sole energy source. J. Gen. Microbiol. 126:249-252. Bryant, M. P. 1979. Microbial methane production—theoretical aspects. J. Anim. Sci. 483193-201. Cappenberg, T. E. 1975. Relationship between sulfate-reducing and methane-producing bacteria. Plant and Soil 43:125-139. Cappenberg, T. E. 1976. Methanogenesis in the bottom deposits of a small stratifying lake. pp. 125-134. In Schlegel, H. G., B. Gottschalk and N. Pfenning (eds.) Microbial production and utilization of gases (H2, CH4, CO). Goltze-Verlag, Gottingen. Cappenberg, T. E. 1979. Kinetics of breakdown processes of organic matter in freshwater sediments. Arch. Hydrobiol. Beih. Ergebn. Limnol. 12:91-94. Cappenberg, T. E., and E. Jongejan. 1978. Microenvironments for sulfate reduction and methane production in freshwater sediments. pp. 129-138. In W. E. Krumbein (ed.) Environmental biogeochemistry and geomicrobiology. Vol. 1. Ann Arbor Science Publications, Ann Arbor. ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Cappenberg, T. E., E. Jongejan, and J. Kaper. 1978. Anaerobic breakdown processes of organic matter in freshwater sediments. pp. 91-99. In M. W. Loutit and J.A.R. Miles (eds.) Microbial Ecology. Springer, Berlin. Cappenberg, T. E., and R. A. Prins. 1974. Interrelations between sulfate-reducing and methan producing bacteria in bottom deposits of a freshwater lake. III. Experiments with ll‘C-labelled substrates. Ant. van Leeuwenhoek J. Microbiol. Serol..4g:457-469. Chen, M., and M. J. Wolin. 1977. Influence of CH4 production by Methanobacterium ruminantium on the fermentation of glucose and lactate by Selenonmonas ruminantium. Appl. Environ. Microbiol. 34:756-759. Chung, K. 1976. Inhibitory effects of H2 on growth of Clostridium cellobioparum. Appl. Environ. Microbiol. 31:342-348. Ferry, J. G., and R. S. Wolfe. 1976. anaerobic degradation of benzoate to methane by a microbial consortium. Arch. Microbiol. 107:33-40. Hungate, R. E. 1975. The rumen microbial ecosystem. Annu. Rev. Ecol. Syst. 6:39-66. Hobson, P. N., S. Bousfield, and R. Summers. 1974. Anaerobic digestion of organic matter. Critical Reviews in Environmental Control 4}131-191. Iannotti, E. L., D. Kafikewitz, M. J. Whilin, and M. P. Bryant. 1973. Glucose fermentation products of Ruminococcus albus grown in continuous culture with Vibro succinogenes: changes caused by interspecies transfer of hydrogen. J. Bacteriol. 114:1231-1240. Ingvorsen, K., J. G. Zeikus, and T. D. Brock. 1981. Dynamics of bacterial sulfate reduction in a eutrOphic lake. Appl. Environ. Microbiol._42:1029-1036. Kaspar, H. F., and K. Whurmann. 1978. Product inhibition in sludge digestion. Microbial Ecol. 4:241-248. Kelley, C. A., and D. P. Chynoweth. 1981. The contribution of temperature and organic input in controlling rates of sediment methanogenesis. Limnol. 0ceanogr._26:89l-897. Laanbroek, H. J., and N. Pfennig. 1981. Oxidation of short-chain fatty acids by sulfate-reducing bacteria in freshwater and marine sediments. Arch. Microbiol. 128:330-335. Latham, J. J., and M. J. WOlin. 1977. Fermentation of cellulose by Ruminococcus flavefaciens in the presence and absence of Methanobacterium ruminantium. Appl. Environ. Microbiol. 24}297-301. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 10 Mah, R. A.; D. M. Ward, L. Baresi, and T. L. Glass. 1977. Biogenesis of methane. Ann. Rev. Microbiol._31:309-341. Martens, C. S., and R. A. Berner. 1977. Interstitial water chemistry of anoxic Long Island Sound sediments. 1. Dissolved gases. Limnol. 0ceanogr._22:lO-25. Mallard, G. E., and J. I. Frea. 1972. Methane production in Lake Erie sediments: temperature and substrate effects. Proc. 15th Conf. Great Lakes Res. 1972:87-93. McCarty, P. L. 1964. The methane fermentation. pp. 314-343. In H. Heukelekian and N. C. Dondero (eds.) Principles and Applications in Aquatic Microbiology. John Wiley & Sons, New York. McInerney, M. J., M. P. Bryant, and N. Pfennig. 1979. Anaerobic bacterium that degrades fatty acids in syntrophic association with methanogens. Arch. Microbiol. 122:129-135. McInerney, M. J., M. P. Bryant, R. B. Hespell, and J. W. Costerton. 1981. Syntrophmonas wolfei gen. nov. sp. nov., an anaerobic, syntrophic fatty acid-oxidizing bacterium. Appl. Environ. Microbiol. 4151029-1039. Molongoski, J. J., and M. J. Klug. 1980. Anaerobic metabolism of particulate organic matter in the sediments of a hypereutrOphic lake. Freshwater Biol._10:507-518. Mountfort, D. 0., and R. A. Asher. 1981. Role of sulfate reduction versus methanogenesis in terminal carbon flow in polluted intertidal sediments of waimea Inlet, Nelson, New Zealand. Appl. Environ. Microbiol. 42:252-258. Mountfort, D. 0., R. A. Asher, E. L. Mays and J. M. Tiedje. 1980. Carbon and electron flow in mud and sandflat intertidal sediments and Delaware Inlet, Nelson, New Zealand. Appl. Environ. Microbiol. 32:686-694. Oremland, R. S., and B. F. Taylor. 1978. Sulfate reduction and methanogenesis in marine sediments. Geochim. Cosmochim. Acta _42:209-214. Rich, P. H. 1979. Differential C02 and 02 benthic community metabolism in a soft-water lake. J. Fish. Res. Board. Can. _3§:1377-1389. Rich, P. H., and A. H. Devol. 1978. Analysis of five North American Lake ecosystems. Verh. Int. Verein. Limnol. 29:1377-1389. Rich, P. H., and R. G. Wetzel. 1978. Detritus in the lake ecosystem. Amer. Natural. 112:57-71. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 11 Smith, P. H., and R. A. Mah. 1966. Kinetics of acetate metabolism during sludge digestion. Appl. Microbiol._14:368-371. Smith, R. L., and M. J. Klug. 1981. Reduction of sulfur compounds in the sediments of a eutrophic lake basin. Appl. Environ. Microbiol. 41}1230-1237. Smith, R. L., and M. J. Klug. 1981. Electron donors utilized by sulfate reducing bacteria in eutrophic lake sediments. Appl. Environ. Microbiol. 42:116-121. Sérensen, J., D. Christensen, and B. B. Jdrgensen. 1981. Volatile fatty acids and hydrogen as substrates for sulfate-reducing bacteria in anaerobic marine sediment. Appl. Environ. Microbiol. 4355-11. Stadtman, T. C., and H. A. Barker. Studies on the methane fermentation. VIII. Tracer experiments on fatty acid oxidation by methane bacteria. J. Bacteriol. 61567-80. Strayer, R. F., and J. M. Tiedje. 1978. Kinetic parameters of the conversion of methane precursors to methane in a hypereutrOphic lake sediment. Appl. Environ. Microbiol. .36:330-340. Weimer, P. J., and J. G. Zeikus. 1977. Fermentation of cellulose and cellobiose by Clostridium thermocellum in the absence and presence of Methanobacterium thermoautotrOphicum. Appl. Environ. Microbiol. 32}289-297. Widdel, F., and N. Pfennig. 1977. A new anaerobic, sporing, acetate-oxidizing, sulfate reducing bacterium, Desulfotomaculum (amend.) acetoxidans. Arch. Microbiol. 112:119-122. Widdel, F., and N. Pfennig. 1981. Studies on dissimilatory sulfate-reducing bacteria that decompose fatty acids. I. Isolation of a new sulfate-reducing bacteria enriched with acetate from saline environments. Description of Desulfobacter postgatei gen. nov. sp. nov. Arch. Microbiol. 1323395'400' Winfrey, M. R., R. R. Nelson, 8. C. Krevickis, and J. G. Zeikus. 1977. Association of hydrogen metabolism with methanogenesis in Lake Mendota sediments. Appl. Environ. Microbiol. 33:312-318. Winfrey, M. R., and J. G. Zeikus. 1977. Effect of sulfate on carbon and electron flow during microbial methanogenesis in freshwater sediments. Appl. Environ. Microbiol._33:275-281. Winfrey, M. R., and J. G. Zeikus. 1979. Anaerobic metabolism of immediate methane precursors in Lake Mendota. Appl. Environ. Microbiol..31:244-253. Wolin, M. J. 1974. Metabolic interaction among intestinal microorganisms. Am. J. Clin. Nutr._2l:1320-1328. 50. 51. 52. 12 Wolin, M. J. 1979. The rumen fermentation: a model for microbial interactions in anaerobic ecosystems. Adv. Microbial. Ecol. 3f49-77. Zaiss, U. 1981. The sediments of the new artificial lake Bostalsee (Saarland, Germany) with particular reference to microbial activity. Arch. Hydrobiol._92:346-358. Zehnder, A.J.B. 1978. Ecology of methane formation. pp. 349-376. In R. Mitchell (ed.) Water pollution microbiology. Vol. 2. John Wiley and Sons, New York. CHAPTER II Intermediary Metabolism of Organic Matter in the Sediments of a Eutrophic Lakel DEREK R. LOVLEY“ AND MICHAEL J. KLUG W. K. Kellogg Biological Station. Michigan State University, Hickory C orners. Michigan 49060. and Department of Microbiology and Public Health, Michigan State University. East Lansing. Michigan 48824 Received 8 September l981/Accepted 3 November 1981 The rates, products, and controls of the metabolism of fermentation intermedi- ates in the sediments of a eutrophic lake were examined. MC-fatty acids were directly injected into sediment subcores for turnover rate measurements. The highest rates of acetate turnover were in surface sediments (0- to 2-cm depth). Methane was the dominant product of acetate metabolism at all depths. Simulta- neous measurements of acetate, propionate, and lactate turnover in surface sediments gave turnover rates of 159, 20. and 3 uM/h, respectively. [2-“C]pro- pionate and [U-“Cllactate were metabolized to ["C]acetate, ”C02, and 1“CH... [”C]formate was completely converted to MC02 in less than 1 min. Inhibition of methanogenesis with chloroform resulted in an immediate accumulation of volatile fatty acids and hydrogen. Hydrogen inhibited the metabolism of C3-C5 volatile fatty acids. The rates of fatty acid production were estimated from the rates of fatty acid accumulation in the presence of chloroform or hydrogen. The mean molar rates of production were acetate, 82%; propionate, 13%; butyrates, 2%; and valerates, 3%. A working model for carbon and electron flow is presented which illustrates that fermentation and methanogenesis are the predominate steps in carbon flow and that there is a close interaction between fermentative bacteria, acetogenic hydrogen-producing bacteria. and methanogens. Although terminal microbial processes in an- aerobic lake sediments have been the subject of many studies (6, 7, 22, 24, 26, 30, 32—34), the intermediate pathways for carbon and electron flow from initial substrates to terminal processes have not been intensively studied. In anaerobic environments such as sludge (11, 15, 17) and the rumen (12, 35) acetate is the dominant fermenta- tion intermediate. However, Cappenberg and Prins (7) report that in the sediments of Lake Vechten, lactate, a fermentation product of mi- nor importance in other anaerobic ecosystems, has a turnover rate that is 10-fold greater than the acetate turnover rate. Propionate, butyrates, valerates, and formate have also been detected in freshwater sediments (22, 23, 31), but their importance as intermediates in in situ carbon metabolism has not been determined. The accu- mulation of C3-C5 volatile fatty acids (VFA) in sediments in the presence of added hydrogen or when methanogenesis is inhibited (J. J. Molon- goski, Ph.D. thesis, Michigan State University, 1978) suggests that these intermediates may be metabolized to acetate with the production of hydrogen as in other non-gastrointestinal sys- tems (2, 4, 5, 15, 16, 20). Formate may be a t Article no. 10149 of the Michigan Agricultural Experiment Station and no. 458 of the Kellogg Biological Station. 13 direct methane precursor in sediments (30) or converted to hydrogen and carbon dioxide (32). The reported difference in the relative impor- tance of lactate and acetate and the lack of data on the role of other short-chain fatty acids (SCFA) in sediment metabolism led to the pres- ent study. Our purpose was to determine the rates of production and fates of the fermentation intermediates in the profundal sediments of a eutrophic lake to identify the central intermedi- ates in carbon metabolism. In view of the impor- tance of terminal processes in controlling carbon flow in other anaerobic ecosystems (4. 5, 16). we also examined the relative importance of sulfate reduction and methanogenesis in controlling the turnover of fermentation intermediates. MATERIALS AND METHODS Sediment sampling. Sediments were collected from within the 6-m depth contour of Wintergreen Lake. a shallow (maximum depth, 6.5 m). eutrophic lake locat- ed in southwestern Michigan. Seasonal changes in input of particulate organic matter (21). metabolite pool size (22). rate of methane production (22. 29). and sulfate reduction (26) have been previously described. Surface sediments were sampled with an Eckman dredge or a gravity corer. Only the unconsolidated surface sediments were collected from the Eckman dredge samples. The water content of these sediments was greater than 90%. The coring apparatus was 14 VOL. 43, 1982 designed to allow unrestricted movement of water through the core tube as it was lowered into the sediment with a winch. The unconsolidated surface sediments remained intact, and the sediment cores were identical to cores taken by hand using SCUBA. The 7-cm-diameter plastic core tube had holes spaced at 2-cm intervals which permitted subcoring over the depth of the core. For direct injection studies subcores of sediment were taken with a 5-ml plastic syringe with the needle end cut off. A subcore of approximately 6 ml of sediment was taken, and the syringe was sealed with a serum bottle stopper (Wheaton Scientific) while sediment was extruded to a final volume of 5 ml. Surface sediments (0- to 2-cm depth) for incubations in tubes were collected from gravity cores with a 5—ml plastic syringe through a l6-gauge needle. The needle was pushed into a rubber stopper as sediment was slowly extruded to seal the syringe for storage. Sub- cores or syringes filled with sediment were stored under water in the dark at in situ temperature. Measurement of SCF A turnover with l‘C-tracers. Turnover experiments were initiated within 3 h of collecting the sediment. A 50-ul sample of the appro- priate MC-labeled substrate that had been prefiushed with oxygen-free nitrogen was directly injected into the sediment subcores with a 250-ul Hamilton syringe. The concentration of ["C]SCFA in the tracer solu- tions was maintained as high as possible to obtain maximal radioactivity without exceeding the expected in situ concentration of the SCFA in the interstitial water. The syringe needle was inserted through the serum stopper. and the tracer was injected as the needle was withdrawn. The subcores were incubated at in situ sediment temperature for appropriate time intervals, and the incubation was stopped by immers- ing the subcores in an ethanol-dry ice bath. The frozen sediments were extruded into anaerobic pressure tubes (Bellco Glass), stoppered. sealed with an alumi- num crimp, and stored at -—10°C until further analysis. Tubes were placed in boiling water for 15 min to stop activity and were promptly analyzed for 1“CH4, "C02, and [”C]SCFA (see below). First-order turnover rate constants, k, of each tracer substrate were determined from triplicate replications of at least three time points. The in situ turnover rates were calculated by multiplying A‘ by the in situ pool size. Acetate turnover rate constants were estimated from the first-order loss of ["Clacetate over time. The slope of a plot of the natural logarithm of [“Clacetate versus time with [”C]acetate expressed as total counts or a percentage of the initial counts is the negative of the first-order rate constant, k. Altema- lively. It was estimated from plots of the natural logarithm of {“CM/[“CM-("CT/“CM)]} versus time, where CM equals the total amount of I‘CH, and 1“C02 produced from ["C]acetate at time points when I“CH, and 1‘CO; production had reached a maximum. and CT equals 1‘CH, and "C02 present at time t. The slope of this line equals k. Propionate turnover rate constants were calculated from the slopes of plots of the natural logarithm of [”Clpropionate versus time. Suflicient [“Cllactate to detect with the radiochroma- tography technique could not be injected into subcores without injecting a solution that had a lactate concen- tration that was significantly higher than the in situ pool. Since ["CJIactate was converted solely to [“Clacetate. “C02, and ”CH4, (see below), the lac- CARBON FLOW 1N LAKE SEDIMENTS 553 tate turnover rate constant (k = fraction of available "C evolved per unit time) was estimated from the linear rate of I“C02 evolution during incubation peri- ods of 10 min or less. before the [”Clacetate that was produced was further metabolized. The amount of label available for 1“CO; production during this time period was therefore equivalent to the ”C disintegra- tions per minute in the carboxyl carbon of the [”C]lac- tate. Sediments (5 ml) were incubated in pressure tubes or 10—ml Vacutainer tubes (Beckton. Dickinson & Co.) for preliminary measurements of acetate turnover, for determining the fate of [U-“Cllactate. and for measur- ing the fraction of total methane produced from hydro- gen and carbon dioxide. Before and during the transfer of sediments. tubes were flushed with a 93% N2-7‘72 C02 gas mixture that had been passed through a heated column of reduced copper. Solutions were added to the tubes through the stopper with a syringe and needle and mixed on‘ a Vortex mixer for 10 to 15 s. The sediments were incubated without any further shaking, and activity was stopped by freezing as above. The fraction of methane that was produced from hydrogen and carbon dioxide was estimated by adding [“Clsodium bicarbonate and comparing the specific activity of the methane produced after 16 h of incubation with the specific activity of the carbon dioxide in the headspace. The latter did not change significantly over the incubation period. The following radiochemicals were used: [2-"C]ac- etate (54 mCi/mmol: New England Nuclear Corp.). [2- I‘C]propionate (55.7 mCi/mol: lntemational Chemical and Nuclear). [“leormate (56 mCi/mmol: interna- tional Chemical and Nuclear), [1-"C]butyrate (14 mCi/mmol; New England Nuclear). ["Clsodium bi- carbonate (0.1 mCi/mmol; New England Nuclear). and [U-“Cllactate (138.6 mCi/mmol: New England Nucle- ar). Measurement of VFA production with H2, CBC]... and Na; M00. inhibition. The method of Kaspar and Whurmann (15) was adapted to measure the produc- tion of VFA that are metabolized with hydrogen production. To determine whether hydrogen inhibited the metabolism of C3-C5 VFA, approximately 80 ml of sediment was transferred to a 100ml serum bottle (Wheaton Scientific). The bottle was stoppered with a butyl rubber stopper (Bellco Glass, Inc.) and shaken by hand. Samples (4 ml) from the serum bottle were placed in pressure tubes containing a 93% N349? C 0: atmosphere. One milliliter of an Nz-flushed solution of the VFA under study was added to the sediment remaining in the serum bottle and. the sediment was shaken. Samples (4 ml) of the amended sediment were added to tubes as above. A 4-ml sample was also taken to measure the initial VFA concentration. Hydrogen (30 ml) was added with a syringe and needle to half of the tubes containing unamended sediment and to half of the tubes containing amended sediment to provide an initial hydrogen partial pressure of approximately 1 atm (100 kPa). All of the tubes were incubated over- night in a horizontal position on a tube roller to enhance hydrogen diffusion into the sediments. A short-term time course was performed to estimate the rates of production of C3-C5 VFA. Hy- drogen was added to the headspace of tubes of sedi- ments incubated on the tube roller as above. Replicate tubes were sacrificed at 2- or 3-h intervals over 554 LOVLEY AND KLUG a 9-h incubation period for VFA determinations. Rates of acetate production were estimated by an adaption of the method of Chynoweth and Mah (8). Chloroforrn and sodium molybdate were added to 80 ml of sediment to give final concentrations of 100 uM and 2 mM. respectively. The chloroform did not inhibit sulfate reduction, and the molybdate did not affect methanogenesis (27; unpublished data). The chloroform and molybdate inhibited acetate uptake by methanogens and sulfate reducers, respectively. Sedi- ments were incubated in the serum bottles or added in 4-ml samples to pressure tubes as above. Subsamples were removed from serum bottles or tubes were sacrificed in triplicate for acetate determinations. The role of sulfate reduction and methanogenesis in controlling the metabolism of fermentation intermedi- ates was determined by treating sediments with either chloroform or sodium molybdate as above. Subsam- ples were removed for measurement of VFA and hydrogen over time. To determine whether sulfate reducers could compete for hydrogen at in situ con- centrations, sulfate reduction was saturated for sulfate by adding sodium sulfate to sediments in serum bottles to a final concentration of 2 mM. Control sediments received no added sulfate. Methane production was measured over a 17.5-h incubation period. Lower rates of methane production in sediments with added sulfate would indicate increased hydrogen or acetate uptake (or both) by sulfate reducers. Analytical techniques. Interstitial water was collect- ed with dialysis samplers (22) or by centrifuging sedi- ment. Samples obtained with both methods gave com- parable results. SCFA were converted to their benzyl esters by a modification of the method of Bethge and Lindstrom (1). Tetraethylammonium hydroxide was added as the counterion, and heptanoate was used as an internal standard. The samples were evaporated to dryness at 70°C under a stream of air that was passed through a trap containing NaOH to remove volatile organic compounds. Benzyl bromide in acetone (typi- cally 1:400, vol/vol) was added to the dried sample and allowed to react for at least 2 h at room temperature. The benzyl esters were separated on a 2-mm (inner diameter) by 2-m glass column packed with 10% Dexsil 300 on 100/120 mesh Supelcoport, (Supelco, Inc.). Helium was used as carrier gas at a flow rate of 30 ml/min. The injector and detector temperatures were 240° and 270°C, respectively. The oven tempera- ture was programmed at 120°C for 4 min, increased to 230°C at a rate of 20°C/min and held at 230°C for 8 min. A 1/8—in. (ca. 3.1-mm; outer diameter) by 2-m stainless steel column of 10% butane diol succinate on Supelco- port (100/120 mesh) was frequently used to confirm the results obtained with the Dexsil column. Nonesterified fatty acids and ethanol were also analyzed on a glass column packed with 10% SP 1220—1% H3PO, on Chromsorb W (Supelco). Operating temperatures. were: injector, 160°C; detector, 180°C; oven, 120°C for VFA and 60°C for ethanol. When necessary VFA were concentrated by making the sample basic with NaOH and evaporating the sample as above. Free acids were acidified with H3130. before analysis. Quantitative analysis of VFA as free acids or as their benzyl derivatives gave comparable results. A .Varian 3700 (Varian Instruments) with dual flame ionization detec- tors was used throughout. A Finnigan model E,-C, gas chromatograph-mass spectrometer was used to further 15 APPL. ENVIRON. MICROBIOL. confirm the structure of some VFA. VFA were sepa- rated on Carbopack C-3‘7c 20 M Carbowax-0.1% H3PO, (Supelco inc.) for this analysis. [“C]SCFA were detected by a modification of the radiochromatography procedure of Zehnder and Brock (36). [”C]SCFA were chromatographed as out- lined above. The fiame ionization detector of the chromatograph was capped with the flow-measuring assembly that was provided with the instrument. The effluent from the detector was passed through a heated stainless steel line to two scintillation vials that were connected in series. The vials contained 3 ml of scintillation-grade ethanolamine (Eastman Kodak Co.) and 4 ml of methanol as a Coytrapping solution. An additional 5 ml of methanol and 7 ml of scintillation cocktail (15 g of 2,5-diphenyloxazole and 1 g of p-bis- (o-methylstyrl)-benzene in 1 liter of scintillation grade toluene) were added to the vials for liquid scintillation counting in a Beckman LS 8500. Trapping efficiencies as determined with ["C]SCFA were consistently greater than 95%. Counting efliciencies were deter- mined with an internal standard of ["C]toluene (4.5 x 10’ dpm/ml: New England Nuclear Corp.). Specific activities of 1‘CH, and 1“CO; were ana- lyzed with a Varian 3700 gas chromatograph with a thermal conductivity detector that was connected in series with a gas proportional counter. The proportion- al counter was constructed in the laboratory and operated on the same principles as commercially avail- able proportional counters. Gases were separated at 45°C on a 3-m column of Porapak N (100/120 mesh, Waters Associates) with a helium flow rate of 20 ml/ min. The injector and detector temperatures were 120 and 160°C. respectively. Hydrogen was determined on the same column. but with nitrogen as the carrier gas. “C02 that was produced from direct injection studies with [“Cllactate was too low to be readily measured with the proportional counter. Samples (1 ml) of headspace gas were injected through a septum and cap (37) into a scintillation vial that contained the C03- trapping solution described above. The vials were shaken to trap all of the "CO; and counted as above. The total 1“C02 in tubes was corrected for dissolved “C-inorganic carbon with an empirical factor that was determined by injecting H"C03‘ into sediment cores (direct injection experiments) or tubes containing sedi- ments (tube incubations). RESULTS Maximum concentrations of SC F A in Winter- green Lake sediments were within the 0 to 4-cm depth interval. Acetate and propionate concen- trations at this depth ranged from 30 to 340 and 10 to 90 pM, respectively. during the 1980 summer stratification. Formate. isobutyrate. bu- tyrate, isovalerate, and valerate were occasion- ally detected at concentrations of 2 uM or less. Lactate was generally undetectable and was never detected at concentrations greater than 4 uM. Ethanol was never detected. In preliminary acetate turnover experiments comparable first-order rate constants were ob- tained by measuring [2-“C]acetate disappear- ance or 14CH, and 1“C02 production (Fig. 1). During summer stratification the rates of acetate VOL. 43, 1982 r 1 l l w o- 1 r— 040'- w 2 u I fie - 4 75 4 u ,oP ' V Q H —\ oso ‘ :1 ,_ .. ' t—l 1U .90 .a - _ < . .L“ ‘50 E u! g E., 020’- o ‘H'Q~ 5 \ 4Q," (:5 3 " , -. 25 5 1 0.I0r- i" in \_.a U z 2 .0 O I Z 1 1 1 1 " on 5 l0 )5 20 ‘00 thE IN MINUTES FIG. 1. Disappearance of [2-“C]acetate and pro- duction of ”CI-I, and 1“C02 in summer surface sedi- ments (0- to 2-cm depth) collected from gravity cores. turnover to CH4 and CO; were highest in the surface sediments (0 to 2 cm) and decreased rapidly below the 2- to 4-cm depth interval (Fig. 2). The high proportion of [2-“C]acetate that was metabolized to 1“CH4 demonstrated that methanogenesis was the predominant terminal process at all depths. ”CH4 accounted for ap- proximately 80% of the total l“CI-I4 and MC02 produced from [2-“C]acetate in surface sedi- ments throughout the stratified period. An exponential first-order loss of [2-“CJpro- pionate was observed in surface sediments. Ra- dioactivity that was lost from the propionate pool after the zero-time samples could be com- pletely accounted for as [”C]acetate, 1“CO; and 4CH4 (Table 1). 1"C02 production from [U- 1‘Cllactate was linear over the 0- to 10-min incubation period as previously shown (27). Sur- DEPTH IN CM 9 0 1 1 ICC 300 500 ID ACETATE tumovrn pM/hr ”CW/“(Hf "co 2 l FIG. 2. Acetate turnover rates and ratio of 1“CH. production to total l‘CI-I.. and l“CO; production with depth in summer sediments sampled from gravity cores. 16 CARBON FLOW IN LAKE SEDIMENTS 555 face sediments that were incubated with [U- l‘C]lactate for 10 min had 32% of the added MC remaining in the lactate pool, 51% as [”CJace- tate, 23% as 1"C02, and less than 1% as ”CH4. No other [”C]SCFA were detected. Turnover rates of SCFA were measured simultaneously on two dates during summer stratification. On 27 June 1980 acetate and propi- onate were the only SCFA that were detected in surface sediments. Acetate had a turnover rate constant of 3.11/h and a pool size of 110 uM yielding a turnover rate of 342 uM/h. The rate constant, pool size, and turnover rate for propio- nate were 1.86/h, 90 uM, and 167 uM/h, respec- tively. On 4 September 1980 SCFA other than ace- tate and propionate were detected, and a more complete examination of carbon flow in surface sediments was conducted (Table 2). Acetate had the highest turnover rate of the SCF A examined. Propionate turnover was less than 20% of ace- tate turnover on a molar basis, and lactate production was minor. The formate turnover rate could not be measured accurately, as [“leormate was completely converted to 1‘CO; in less than 1 min. The isobutyrate and butyrate concentrations were 0.3 and 1 uM. respectively. These concentrations were too low to conduct tracer studies with the available specific activities of the l‘lC-labeled compounds and the radiochromatography technique. lsoval- crate and valerate were not detectable. Thirty- seven percent (standard error, 4) of the total methane production was derived from reduction of carbon dioxide on this date. Methanogenesis but not sulfate reduction was found to be necessary for the metabolism and maintenance of low pool sizes of VFA and hydrogen. The inhibition of methanogenesis with chloroform resulted in an immediate accu- mulation of hydrogen and VFA (Fig. 3). Acetate accumulated most rapidly, with lower accumula- tion rates of propionate, butyrates, and valer- ates. Inhibition of sulfate reduction with 2 mM sodium molybdate did not result in an accumula- tion of VFA or hydrogen even after a 24-h incubation. Increasing the sulfate pool by 2 M had no effect on methane production. The mean and 95% confidence intervals (n = 12) for meth- ane production in sediment that received sulfate TABLE 1. Percent distribution over time of I"C-labeled compounds produced from [2-"C]propionate in surface sediments (0- to 2-cm depth) collected from gravity cores Time (min) [“Clpropionate ["Clacetate Other [“CISCFA "CH, "CO: ”:33” 0 100 0 0 0 0 10 80 9 0 2 6 97 30 48 25 0 10 24 107 556 LOVLEY AND KLUG 17 APPL. ENVIRON. MICROBIOL. TABLE 2. Pool sizes. turnover rates. and fates of fermentation intermediates in surface sediments (0- to 2- cm depth) collected from gravity cores on 4 September 1980 Intermediate Pool size Rate constant Turnover rate Fate _____ _. tttM) (per h) (HM/h) Acetate CO2 CH, Other SCFA Acetate 100 1.59 159 + + 0 Propionate 14 1 .44 20 + + + 0 Lactate 1 2.76 3 + + + 0 Formate 2 N M" NM 0 + 0 0 " ["C]formate was completely converted to MCO; (NM). VFA CONCENIEAHON yM DISSOtle HanOGEN AM i.— ’4— PROPIONME .A “/4— "" , guttOAlES Wtfiyr—H l 1'. l l 1 ' J 4 8 IC l.‘ 6 TIME IN HOURS FIG. 3. Accumulation of volatile fatty acids and hydrogen after the inhibition of methanogenesis in summer surface sediments collected with an Eckman dredge. was 48.1 i 7.1 umol/liter per h compared with a rate of 51.7 t 6.4 umol/liter per h measured for sediments that did not receive sulfate. The effect Of hydrogen partial pressure of approximately 1 atm on the metabolism of vola- :ile fatty acids is illustrated in Table 3. Propio- in less than 1 min. a rate constant was not measurable nate, isobutyrate, isovalerate, and valerate ac- cumulated in sediments which were exposed to increased hydrogen partial pressures. Added hydrogen completely inhibited the metabolism of propionate, isobutyrate, isovalerate, and val- erate that was added to sediment, whereas over 90% of the added VFA were metabolized in controls that received no hydrogen. Butyrate metabolism was only partially inhibited by add- ed hydrogen. However, a compound accumulat- ed that cochromatographed with isobutyrate. Gas chromatography-mass spectrometry con- firmed that the cochromatographing compound was isobutyrate. The sum of butyrate and isobu- tyrate concentrations indicated nearly complete inhibition of the mineralization of butyrate car- bon (Table 3). Incubations of sediment with [1- l“Clbutyrate and added hydrogen resulted in the inhibition of butyrate carbon mineralization and the conversion of approximately 50% of the added [“Clbutyrate to [”C]isobutyrate. In simi- lar experiments hydrogen only partially inhibit- ed the metabolism of added lactate (data not shown). The rate of accumulation of C3-C5 VFA in the TABLE 3. Effect of hydrogen on the metabolism of volatile fatty acids in winter surface sediments collected with an Eckman dredge Concentration (uM I“ " Mean C’t VFA added H3. final , H; + VFA _ _ _ inhibition” Initial Final Propionate 16.7 (2.7)“ 15.4 34.3 (2.5) 114 lsobutyrate 1.6 (0.2) 5.1 6.4 (0.5) 97 Butyrate“ <0.5 26.9 11.6 (0.9) 43 Butyrate' <0.5 34.5 31.2 (1.2) 92 Isovalerate 2.2 (0.3) 2.0 4.2 (.07) 100 Valerate 0.4 (0.3) 20.8 23.4 (0.9) 110 ‘ Sediment additions were: none. only H2, only VFA. and VFA and H3. Over 90% of all added VFA in VFA alone treatment was metabolized during the overnight incubation period. Initial concentrations Of C3-C5 VFA in sediments that were not amended with VFA were less than 0.5 uM. b Mean percent inhibition calculated as (mean final concentration of VFA in sediments with added H3 and VFA - mean final concentration in sediment with only M: added)/(mean initial concentration Of VFA in sediments with VFA added) x 100. ‘ Standard error of mean within parentheses. d Butyrate concentration alone used in calculations. ' Sum of butyrate and isobutyrate concentrations used in calculations. VOL. 43, 1982 18 CARBON FLOW IN LAKE SEDIMENTS 557 TABLE 4. Production rates of VFAs in winter surface sediments (collected with an Eckman dredge) as determined by inhibitor experiments Rate of production (uM/h) and molar % production“ Fatty acid Sediment A Sediment B Sediment C Mean molar % ________ _.__-__ production Rate 92 Rate % Rate % Acetate 6.5 81 7.1 88 4.1 85 85 Propionate 1.2 15 0.7 9 0 3 6 10 Butyrates 0.3 3 0.1 1 0.1 2 2 Valerates ND” 0 0.2 2 0.3 6 3 " Percentage of total number of moles of fatty acids produced. b No detectable increase over the 9-h incubation period. presence of increased hydrogen partial pres- sures and the rate of acetate accumulation in the presence of 100 uM CHCl3 and 2 mM sodium molybdate in sediments that were collected at three separate times in midwinter are shown in Table 4. Although there was considerable vari- ability in the absolute rates of VFA production on different dates, the relative pattern of produc- tion was similar, i.e., acetate > propionate > butyrates and valerates. The ratio of acetate to propionate production was similar to that mea- sured in summer sediments, although the rates of acetate and propionate turnover in winter sediments were less than 10% of summer rates. DISCUSSION The 14C tracer and inhibitor studies both demonstrated that acetate is the dominant car- bon fermentation intermediate in profundal sedi- ments of Winter reen Lake. Rate measurements obtained from 1 C tracer studies are considered to closely represent in situ activities since the sampling method and subsequent manipulations provide a relatively undisturbed sediment sam- ple. The rates of VFA production that were measured in the inhibitor experiments do not represent in situ rates, since sediments were mixed and not incubated at the in situ tempera- ture. The inhibitor approach is also considered to have underestimated the actual acetate pro- duction rates and overestimated the production of C3-C5 SCFA, since increased hydrogen con- centration resulting from the inhibition of meth- anogens or hydrogen addition can be expected to shift carbon fiow toward the production of more reduced fermentation products (8, 10, 25). The inhibitor experiments, however, provided an independent measure of the relationship be- tween propionate and acetate turnover rates and enabled the estimate of the turnover of butyrates and valerates in sediments where low pool sizes precluded the use of tracer studies due to the low specific activities of the available ”C-la- beled compounds. The high rate of acetate turnover relative to the turnover rates of the other fermentation intermediates indicates that most of the acetate was produced in the initial fermentation of sub- strates rather than indirectly through other SCFA. Although some acetate may also be formed by bacteria fermenting hydrogen and carbon dioxide to acetate, the number of these acetogenic bacteria in sediments is low (3), and only a small percentage of the acetate in other anaerobic environments is derived from carbon dioxide (18). The overall pattern of SCFA pro- duction in Wintergreen Lake sediments is simi- lar to that in the rumen and anaerobic sludge (Table 5). Although differences exist in the quali- ty and quantity of the carbon inputs to these systems, these similarities in carbon fiow sug- gest that the steps and controls in initial carbon metabolism are the same in all three systems. Hydrogen partial pressures are low in all three systems (11, 12, 15, 30, 32), and acetate and hydrogen are expected to be the primary fer- mentation products in systems with low partial pressures of hydrogen (12, 35). The noted exception to this general pattern of carbon flow in anaerobic ecosystems is in the sediments of Lake Vechten (Table 5). Although the rate constants for lactate turnover in Winter- green Lake and Lake Vechten are comparable, 2.74 and 2.37/h, respectively, the lactate pool in Lake Vechten sediments is 12.2 ug/g of wet sediment compared with less than 0.35 ug/g of wet sediment in Wintergreen Lake sediments. The lactate pool is higher than the acetate pool in Lake Vechten, which is unique since lactate is generally not detectable in anaerobic sewage sludge (11) and in the rumen. where acetate pool sizes are generally greater than 60 mM. the lactate pool size is much less than 1 mM unless the animal is suddenly shifted to a high-energy diet (9). The inhibition of the turnover of propionate, butyrates, and valerates by increased hydrogen partial pressures in the sediments of Winter- green Lake suggests that these compounds are metabolized with the evolution of molecular hydrogen. Hydrogen inhibits the metabolism of propionate to acetate in sludge (15, 16; P. H. 19 558 LOVLEY AND KLUG APPL. ENVIRON. MICROBIOL. TABLE 5. Relative production of fermentation intermediates in various anaerobic ecosystems Production rates (molar CE)" Ecosystem Acetate Propionate Butyrates Valerates Lactate WGL sediments (summer)” 86 13 ND" ND 2 WGL sediments (winter) 85 10 2 3 ND Anaerobic sludge” 80 14 6 <3 <6 Rumen‘ 72 18 8 2 <1 Lake Vechtenf 9 ND ND ND 91 ° Percentage of total number Of moles of fatty acids produced. ° Contribution to acetate from propionate. butyrate, and lactate turnover subtracted from acetate turnover to estimate acetate coming from initial substrates. WGL. Wintergreen Lake. ‘ ND, Not determined. " Acetate. propionate. and butyrate from basal mesophilic rates of Mackie and Bryant (18). Lactate and valerates from the scheme of Kaspar and Wuhrmann (15). ' From the data of Hungate et al. (13) and Jayasuriya and Hungate (14). I From the data of Cappenberg and Prins (7). Smith, F. M. Bordeaux, and P. S. Shuba, Ab- stracts of papers of the 159th National Meeting of the American Chemical Society, WATR 49, 1970) and the degradation of butyrate to acetate in cocultures (19). The production of [”C]ace— tate as the only [”C]SCFA from [”Cllactate suggests that lactate is also metabolized to ace- tate and hydrogen. Although the in situ fates of C4-C5 VFA could not be determined, it seems likely they are metabolized to acetate either directly or indirectly through the propionate pool as proposed for anaerobic sludge. Whether the metabolism of butyrate to isobutyrate as observed in the hydrogen addition experiments represents a step in in situ metabolism Of butyr- ate has yet to be determined. The immediate accumulation of VFA and hy- drogen after the inhibition of methanogenesis demonstrates the importance of methanogenesis in controlling carbon and electron flow in Win- tergreen Lake sediments. Although approxi- mately 20% of the acetate in surface sediments is completely metabolized to carbon dioxide, pri- marily by sulfate reducers (27), sulfate reducers are not considered to be essential in the metabo- lism of VFA since there was no build-up of VFA when sulfate reduction was inhibited. The accu- mulation of C3-C5 VFA when methanogenesis is inhibited is probably the result of product inhibi- tion by hydrogen and possibly acetate since no known methanogens are capable of directly uti- lizing C3-C5 VFA. Sorensen and co-workers (28) have noted an accumulation of VFA and hydro- gen in marine sediments when sulfate reduction is inhibited which is similar to the accumulation in Wintergreen Lake sediments when methano- genesis is inhibited. They attribute the accumu- lation of propionate and butyrates to the inhibi- tion of the direct metabolism of these compounds by sulfate reducers. However, they did not examine the possibility that the hydrogen or acetate (or both) that accumulated might inhibit the metabolism of propionate and butyr- ate by acetogenic hydrogen-producing bacteria. If the latter were the case the sulfate reducers and methanogens would serve the same function in control of the metabolism of VFA in the two respective habitats. A working model for the metabolism of sedi- mented organic matter in the profundal surface sediments of Wintergreen Lake has been devel- oped (Fig. 4). The hydrogen production rate is based on the assumption that methanogens are the principal hydrogen consumers. and acetate is the only methane precursor other than hydro- gen and carbon dioxide. Although there is a potential for hydrogen uptake by sulfate reduc- ers (27), they are not considered to be significant in in situ hydrogen uptake since additions of sulfate concentrations that are saturating for sulfate reduction do not initially alter the rates of methanogenesis in these sediments. Further- more, the inhibition of sulfate reduction does not stimulate methanogenesis (27). Approximately 40% of the total methane production is derived from hydrogen and carbon dioxide, whereas 80% of the acetate turnover is through methano- genesis. Thus, for each 100 mol of methane produced, [0.6 (fraction of total methane pro- duction from acetate) X 100]/0.8 (fraction of total acetate turnover that goes to methane) = 75 mol of acetate and 0.4 (fraction of methane from hydrogen) X 4 (moles of H3 per CH4) X 100 = 160 mol of hydrogen are metabolized. The re- sultant hydrogen/acetate production ratio is 2.1. which is close to the hydrogen/acetate produc- tion ratio of 2 that is expected from the fermen- tation of an initial substrate with a redox state of 0. The model emphasizes that the controlling steps in carbon flow in profundal sediments of Wintergreen Lake are likely to be the initial fermentation of substrate to acetate. hydrogen. and carbon dioxide and the subsequent metabo- 2O VOL. 43, 1982 CARBON FLOW IN LAKE SEDIMENTS 559 PARTICULATE ORGANIC MATTER l LEACHING AND HYDROLYSIS DISSOLVED ORGANIC MATTER l FERMENTAIION FORMATE l / \ CO2 H2 C02 150 / ll (H2) 13 PROPIONATE / l \ 39012) taco, l3ACETATE 50 C02 c4 AND c5 vrx PROPION ATE co, . l: 2 lACTATE / 4(H2) 2C02 2ACETATE \ ACE TATE 75 l 4/L__j 100 ACETATE 120 C02 FIG. 4. Model of carbon and electron flow in the profundal surface sediments of Wintergreen Lake. The numbers adjacent to compounds represent the molar amounts of each compound produced or metabolized for every 100 mol ofthe total acetate pool that is metabolized. The numbers adjacent to arrows represent the molar contribution from substrate to product. All acetate and hydrogen produced from initial fermentation and the metabolism of SCFA enter common pools as designated by the boxes. (H2) represents possible reduction of sulfate as well as molecular hydrogen production. lism of these compounds by methanogens. The formation of acetate as the major SC FA interme- diate, the flux of more reduced SCFA through the acetate pool. and the maintenance of low pool sizes of SCFA indicate a close coupling between fermentative bacteria. acetogenic hy- drogen-producing bacteria, and methanogens. Based on the pool sizes of SCFA and dissolved inorganic carbon in the sediment. the concentra- tion of dissolved hydrogen must be 30 nM or less to make the oxidation of propionate to acetate thermodynamically favorable (AG 5 0). We con- clude that acetate and hydrogen are the central intermediates in carbon and electron flow in the profundal sediments of Wintergreen Lake and that carbon metabolism is dependent upon the maintenance of low hydrogen concentrations by methanogenic bacteria. Studies on carbon flow in the sediments of less productive lakes are in progress to determine whether this pattern of anaerobic carbon flow can be generalized to other lake sediments where sulfate reduction rather than methanogenesis is the dominant ter- minal process. ACKNOWLEDGMENTS We thank D. Odelson and E. Oliver for gas chromatograph- mass spectrometer analysis and G. King and R. Smith for manuscript review. This research was supported by National Science Founda- tion grants DEB 78-05321 and DEB 81-09994. LITERATURE CITED 1. Bethge, P. 0.. and K. Lindstrom. 1974. Determination of organic acids of low relative molecular mass (C, to C4) in dilute aqueous solution. Analyst (London) 99:137-142. 2. Boone, D. R., and M. P. Bryant. 1980. Propionate-degrad- ing bacterium, Syntrophobacter N‘Ullflil sp. nov. gen. nov.. from methanogenic ecosystems. Appl. Environ. Microbiol. 40:626—632. 3. Braun. M., S. Schoberth. and G. Gottschalk. I979. Enu- meration of bacteria forming acetate from H: and CO; in anaerobic habitats. Arch. Microbiol. 120:201-204. 4. Bryant, M. P. 1976. The microbiology of anaerobic degra- dation and methanogenesis with special reference to sew- age. p. 107-—118. In A. G. Schlegel and J. Barnca (ed.). Microbial energy conversion. Vcrlag Erich Goltze KG. Gottingen. 5. Bryant, M. P. 1979. Microbial methane production—theo- retical aspects. J. Anim. Sci. 48:193-201. 6. Cappenberg, T. E. 1974. Interrelations between sulfate- reducing and methane-producing bacteria in bottom dc- posits ofa freshwater lake. I. Field Observations. Antonie van Leeuwenhoek J. Microbiol. Serol. 40:285-295. 7. Cappenberg, T. E., and R. A. Prins. 1974. Interrelations between sulfate reducing and methane-producing bacteria in bottom deposits of a freshwater lake. lll. Experiments with 1‘Celabelled substrates. Antonie van Leeuwenhoek J. Microbiol. Serol. 40:457—469. 8. Chynoweth, D. P., and R. A. Mah. 1971. Volatile acid formation in sludge digestion. Adv. Chem. Ser. 105:41—54. 9. Counotte, G. H. M., and R. A. Prins. 197811979. Regula- tion ofrumen lactate metabolism and the role of lactic acid 560 ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 21 LOVLEY AND KLUG in nutritional disorders of ruminants. Vet. Sci. Commun. 2:277-303. . Demeyer, D. 1.. and C. J. Van Nevel. 1975. Methanogene- sis. an integrated part of carbohydrate fermentation and its control. p. 366-382. In E. W. McDonald and A. C. T. Warner (ed.). Digestion and metabolism in the ruminant. University of New England. Armidale. Australia. Hobson, P. N., S. Bousfield and R. Summers. 1974. Anaer- obic digestion of organic matter. Crit. Rev. Environ. Control 4:131-191. Hungate. R. E. 1975. The rumen microbial ecosystem. Annu. Rev. Ecol. Syst. 6:39-66. Hungate, R. E., J. Reiehland, and R. Prins. 1971. Parame- ters of rumen fermentation in a continuously fed sheep: evidence ofa microbial rumination pool. Appl. Microbiol. 22:1104-1113. Jayasuriya, G. C. N. and R. E. Hungate. 1959. Lactate conversions in the bovine rumen. Arch. Biochem. Biophys. 82:274-287. Kaspar. H. F., and K. Wuhrmann. 1978. Kinetic parame- ters and relative turnovers of some important catabolic reactions in digesting sludge. Appl. Environ. Microbiol. 36:1-7. Kaspar, H. F., and K. Whurmann. 1978. Product inhibi- tion in sludge digestion. Microb. Ecol. 4:241-248. Lawrence. A. W. 1971. Application of process kinetics to design of anaerobic processes. Adv. Chem. Ser. 105:163- 189. Mackie, R. J., and M. P. Bryant. 1981. Metabolic activity of fatty acid-oxidizing bacteria and the contribution of acetate. propionate. butyrate and CO: to methanogenesis in cattle waste at 40 and 60°C. Appl. Environ. Microbiol. 41:1363-1373. McInerney, J. J., M. P. Bryant. R. B. Hespell, and J. W. Costerton. 1981. Syntroplmnmnas wolfei gen. nov. sp. nov.. an anaerobic. syntrophic fatty acid-oxidizing bacte- rium. Appl. Environ. Microbiol. 41:1029-1039. Melnemey. M. J., M. P. Bryant, and H. Pfennig. 1979. Anaerobic bacterium that degrades fatty acids in syn- trophic association with methanogens. Arch. Microbiol. 122:129-135. Molongoski. J. J., and M. J. Klug. 1980. Quantification and characterization of sedimenting particulate organic matter in a shallow hypereutrophic lake. Freshwater Biol. 10:497-506. Molongoski. J. J., and M. J. Klug. 1980. Anaerobic me- tabolism of particulate organic matter in the sediments of a hypereutrophic lake. Freshwater Biol. 10:507-518. Monokova, S. V. 1975. Volatile fatty acids in bottom 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 35. 36. APPL. ENVIRON. MICROBIOL. sediments of the Rybinsk Reservoir. Hydrobiol. J. 11:45- 48. Robertson, K. C. 1979. Quantitative comparison of the significance of methane in the carbon cycle of two small lakes. Arch. Hydrobiol. Beih. Ergebn. Limnol. 12:123- 135. Rufener, W. H., and M. J. Wolin. 1968. Effect of CCi. on CH. and volatile acid production in continuous cultures of rumen organisms and in a sheep rumen. Appl. Microbiol. 16:1955-1956. Smith. R. L., and M. J. Klug. 1981. Reduction of sulfur compounds in the sediments of a eutrophic lake basin. Appl. Environ. Microbiol. 41:1230—1237. Smith. R. L., and M. J. King. 1981. Electron donors utilized by sulfate-reducing bacteria in eutrophic lake sediments. Appl. Environ. Microbiol. 42:116-121. Sorensen, J., D. Christensen, and B. B. Jorgensen. 1981. Volatile fatty acids and hydrogen as substrates for sulfate- reducing bacteria in anaerobic marine sediment. Appl. Environ. Microbiol. 42:5-11. Strayer, R. F., and J. M. Tledje. 1978. In situ methane production in a small hypereutrophic hard-water lake: loss of methane from sediments by vertical diffusion and ebullition. Limnol. Oceanogr. 23:1201-1206. Strayer, R. F. and T. M. Tiedje. 1978. Kinetic parameters of the conversion of methane precursors to methane in a hypereutrophic lake sediment. Appl. Environ. Microbiol. 36:330—340. Takai. Y. 1970. The mechanism of methane fermentation in flooded paddy soil. Soil Sci. Plant Nutr. 16:238-244. Winfrey, M. R., D. R. Nelson. S. C. Kievlekis. and J. G. Zeikus. 1977. Association of hydrogen metabolism with methanogenesis in Lake Mendota sediments. Appl. Envi- ron. Microbiol. 33:312-318. Winfrey, M. R. and J. G. Zeikus. 1977. Effect of sulfate on carbon and electron flow during microbial methano- genesis in freshwater sediments. Appl. Environ. Micro- biol. 33:275-281. . Winfrey. M. R., and J. G. Zeikus. 1979. Anaerobic metab- olism ofimmediate methane precursors in Lake Mendota. Appl. Environ. Microbiol. 37:244-253. Wolin. M. J. 1979. The rumen fermentation: A model for microbial interactions in anaerobic ecosystems. Adv. Mi- crob. Ecol. 3:49-77. Zehnder, A. J. B.. and T. D. Brock. 1979. Methane forma- tion and methane oxidation by methanogenic bacteria. J. Bacteriol. [37:420-432. . Zehnder. A. J. B.. B. "user, and T. D. Brock. 1979. Measuring radioactive methane with the liquid scintilla- tion counter. Appl. Environ. Microbiol. 37:897-899. CHAPTER III Kinetic Analysis of Competition Between Sulfate Reducers and Methanogens for Hydrogen in Sediments Introduction: It is generally considered that sulfate reducing bacteria (SRB) can inhibit the activity of methanogenic bacteria (MB) when millimolar quantities of sulfate are present. Thermodynamic calculations can be used to predict the exclusion of methane production in sulfate-containing sediments (8,14,29). However, it is invalid to argue that a reaction that is more thermodynamically favorable will exclude another reaction that is also thermodynamically favorable (15). Therefore, MB must be inhibited by toxic metabolites, the lack of methane precursors, or required growth factors in the presence of sulfate. The prevalent conclusion is that SRB inhibit MB by outcompeting them for hydrogen and acetate (1,2,6,14,l8,21,29), but the mechanism(s) for this have not been elucidated. MB are frequently present in sulfate-containing sediments and have the potential to consume methane precursors as evidenced by methane production when sulfate reduction is inhibited or when hydrogen or acetate is added to the sediments (2, 21, 26, 29). Our working hypothesis was that SRB have a higher affinity for hydrogen and acetate than MB, which enables SRB to maintain the pool of these substrates at concentrations too low for MB to effectively utilize when sulfate is not limiting to SRB. The studies reported here concentrated on the competition for hydrogen 22 23 since acetate-utilizing MB are generally absent in natural sediments in which SRB effectively outcompete MB (8,19,26) and the ultimate competition is thus for hydrogen. Materials and Methods: Measurements of in Situ Rates: Sediments were collected during summer stratification from two sites in Wintergreen Lake, a eutrophic lake located in southwestern Michigan. During summer stratification the sediments at the profundal site, site A, lie below an anaerobic, sulfate-depleted hypolimnion (sulfate concentration range 30-16011M) whereas those at the depth of the themocline, site B, have oxygen (range 1-4 mg oxygen per liter) and sulfate (18011M) in the overlying water (16,17). Sulfate reduction rates were measured by the direct injection method of Jorgenson (10) as described in detail by King and Klug (11). Briefly, 10 pl of carrier free 35804" (luCi) was injected into sediment cores incubated at in situ temperatures. The incubation was stopped by quick freezing. The 3582' produced was distilled, trapped, and quantified by liquid scintillation counting. Sulfate reduction rates were calculated by multiplying the rate of conversion of 358042“ to 3582’ by the in situ sulfate pool. Interstitial water was collected with dialysis samplers (l7) and analyzed for sulfate turbidimetrically (28). Methane production was measured on 5 ml subcores taken through ports in cores (7 cm inner diameter) collected using SCUBA. The subcores were extruded into pressure tubes (Bellco Glass) or 20 ml serum bottles (Wheaton Scientific) under an atmosphere of 93 Z nitrogen and 7 Z carbon dioxide. The vessels were stoppered with 24 butyl rubber stOppers (Bellco Glass), sealed with an aluminum crimp, and incubated at in_§i£u_temperatures. The rate of increase in methane concentration in the headspace was measured at intervals over a 20-30 hour incubation period. The tubes were shaken before each methane analysis to equilibrate the dissolved gases with the headspace. Methane was analyzed on a Varian 600D gas chromatograph as described below. Laboratory studies: Sediments for laboratory studies were collected from the A and B site with an Eckman dredge. Depending on the experiment 500, 700, or 800 ml of sediment was transferred under anaerobic conditions to 1 liter reagent bottles (Wheaton Scientific) and sealed with a rubber stopper. A final concentration of either 10 or 20 mM ferrous sulfate (sulfate—amended sediments) or ferrous chloride (control sediments) was added to the sediments. Ferrous salts were used to prevent the accumulation of free sulfide, which is toxic to methanogens at high concentrations (7,29). Ferrous chloride was added to control flasks to eliminate any potential differential effects of excess iron on hydrogen uptake or production. The sediments were incubated at 20 :_ 2°C in the dark without mixing or were placed on a cell production bottle roller (Bellco) and slowly turned. Molydate was added to the sediments as a nitrogen-flushed 0.5 M solution of sodium molybdate to give a final concentration of 5 mM. Molybdate is regarded as an effective and specific inhibitor of sulfate reduction in sediments (20,21,25,26). Carbon dioxide and methane in the headspace of the bottles were analyzed on a Carle basic gas chromatograph equipped with a 25 microthermistor detector. The gases were separated on a 1 m column of Poropak N (Waters Associates) with a helium carrier at a flow rate of 20 ml/min and an oven temperature of 60°C. When greater sensitivity for methane was desired, a Varian 600D gas chromatograph with a flame ionization detector was used. Cases were separated with a helium carrier on a 1 m column of Poropak N at 50°C. Hydrogen was analyzed on a Varian 3700 gas chromatograph with a thermal conductivity detector. The gases were separated on a 3 m column of Poropak N with nitrogen as the carrier at 15 ml per min and an oven temperature of 35°C. The detection limit was 0.04 pascals. One pascal is approximately equivalent to 9.9 x 10"6 atmospheres and a dissolved hydrogen concentration of 8 nM. The bottles were shaken vigorously before sampling to equilibrate the dissolved gases with the headspace. Interstitial water for sulfate analysis was collected by centrifugation and analyzed by high pressure liquid chromatography. Ions were separated at room temperature on a 5 x 0.46 cm Vydac column (Anspec) with a solvent of 1 mM pthallic acid (pH 5.5) at a flow rate of 2 ml/min. Sulfate was detected with a Wescan conductivity detector (Anspec). For the kinetic analysis of hydrogen uptake 4 or 6 m1 samples of sediments were diSpensed into 25 x 142 mm roll tubes (Bellco). The tubes were flushed with oxygen-free nitrogen before and during the transfer. In experiments where chloroform was added to sediments, a 50—75 m1 sample of sediment was first transferred to a 120 ml serum bottle. Chloroform was added directly (final concentration, 0.003% V/V). The sediments were mixed and diSpensed into tubes as above. The tubes were incubated with slow rolling on a tube roller to create 26 a thin film of sediment (27). Hydrogen was added and headspace samples were withdrawn over time and analyzed for hydrogen or methane or both. Two experimental approaches were used to ensure that chloroform did not alter the potential of SRB to take up hydrogen. In the first experiment, sediments were amended with 550 “M (final concentration) of sulfate to saturate SRB for sulfate. The sediments were incubated under saturating hydrogen (50 kPa) on the tube roller, and the rates of sulfate depletion over a 2 h incubation period in sediments treated with chloroform and control sediments were compared. In the second experiment, sediments that had been adapted to 20 mM sulfate were incubated on the tube roller with an initial hydrogen concentration of l kPa. The initial rate of hydrogen uptake was measured in untreated sediments, sediments treated with chloroform and sediments treated with molybdate. If chloroform did not inhibit hydrogen uptake by SRB, then the sum of hydrogen uptake in sediments treated with chloroform and sediments treated with molybdate would equal the hydrogen uptake in untreated sediments. Kinetic Analysis: Hydrogen uptake in sediments has previously been shown to follow MichaeliséMenten kinetics (27). VM°S V= __ (1) K + S where V is the velocity of uptake, V“ is the maximum potential uptake velocity, S is the substrate concentration and K is the substrate concentration at which V = 0.5 V“. Kinetic parameters were estimated 27 from progress curves of hydrogen consumption over time. A linearlized expression of an integrated form of the Michaelis-Menten expression can be derived (23). = . + t K t (2) 1n so/st -1 so—st V“ K where 80 is the initial substrate concentration and St is the substrate concentration at time, t. This method gives kinetic parameters for hydrogen uptake in sediments comparable to those estimated from initial velocity studies (27) and has the added advantage that variability between sediment samples for a particular kinetic analysis can be eliminated since all the substrate concentrations are in effect tested on the same sediment sample. Sediments containing hydrogen-consuming MB and SRB populations can be expected to have a total hydrogen uptake described by a two-term Michaelis-Menten equation. vT =- —— + —— (3) KSRB-i-S KMB'i-S where VT is the total rate of hydrogen uptake, VMSRB and KSRB are the VM and K of the SRB population, and VMMB and KMB are the VM and K for the MB. This two-term equation was used in the analysis of hydrogen uptake in sulfate-containing sediments that had both MB and SRB populations. Kinetic parameters for the two populations were entered into a program which calculated total hydrogen uptake over time. RESULTS: Concurrent methane production and sulfate reduction were observed in the surface sediments (0-2 cm) of both site A and site B (Table 1). 28 .Afifiv wSHM was wcfix Scum oufim < new oumu cowuosku can cowumuucoocoo mummanm o omuwh GOUHuavoum mcmzume was many coauonvmu sawmanm mo Hauou an wovfi>fiv mumu coauoswwu wumeSm a owufiwamhfimwmfi oumu whoa uo M NO uouuo vumwcmum.fl some ”use; you acoawvow mo noufia you moans: m 2 m; H o; S H 2.. an m 2 A; H N8 2 H 3 2 as. Aflauoa mo NV cowuoaeme censuseoum fizzy muwm naoauosvmm enemanm mwumwasm mmcmcuoz cofiumuucmocou mummaam ucoaavmm .cowumoamwumuum uoaaam wcfiuav axe; :mouwuoucaz mo Aao miov mucmaavom wuwwusw msu cw :oHuosvmu wumMHSm was coauosvoua ocmsuma we wocmuuomaa m>wuwaom "a sense 29 Methane production was the dominant process and comprised about the same proportion of the total of methane production and sulfate reduction at both sites. Methane production in sediments from both sites was completely inhibited within 2-5 days at 20°C by the addition of 10 or 20 mM sulfate. Active sulfate reduction in the sulfate-amended sediments was evidenced by the loss of dissolved sulfate and the appearance of black ferrous sulfide over time. There was also an increase in carbon dioxide production in sulfate-amended sediments over that in control sediments. Sulfate-amended sediments in which methane production was inhibited had significantly lower hydrogen partial pressures than FeClz controls and untreated sediments (Table 2). Monitoring over time demonstrated that the inhibition of methane production and the decrease in hydrogen were concurrent (Figure 1). Both control and sulfate-amended sediments had high initial rates of methane production and elevated hydrogen partial pressures; presumably due to disturbances in carbon flow resulting from the initial manipulations with the sediment. The hydrogen partial pressure stabilized in control (FeClZ-amended) sediments at approximately 1 Pa while methane production continued at lower rates. However, in sulfate-amended sediments the methane production rate and hydrogen partial pressure dropped sharply until methane production was no longer detectable. The hydrogen partial pressure continued to slowly decline after methane production had ceased. Addition of SmM (final concentration) sodium molybdate to inhibit sulfate reduction in the sulfate-amended sediments resulted in the 30 comma no Nauom woven suHB mxooa m can» mama as: name m umme um woumnsocH o maoaum>uomno m we uouum wumvcmum.fl.:mmz n manmuooumv uo: was coauoscoua cannume mmumoaecw I “cowuoavoum ocmnume wfinmuowuwv moumouucH + m 25 H 2.0 i usommm 2.3 q~.o.H oo.~ + omauwm msam o~.o.H ~H.H + meowuavnw oz Anamommmv aowuosvoum ucoaummue nouammmum Hmwuumm cement»: «museum: ucoauwom .mummasm woven usonuwa can saws mucoefiwom a“ whammwua Hmwuuma cowouvhm .N manna 31 Figure 1. Methane production rates and hydrogen partial pressures over time in sulfate-amended and control sediments collected from the A site and incubated at 20° C on a bottle roller. Arrow designates addition of molybdate to sulfate-amended sediments. Values are means of duplicate bottles of each treatment and are representative of the results obtained in several similar experiments. SymbolsHandH, methane production rates in sulfate-amended and control sediments; ‘-.and L‘A, hydrogen partial pressure in sulfate—amended and control sediments. 32 .iq/l/ VH3 salowrl 1.: m2: o #2 m2 N? o: 02 00 ON _ _ °°. ‘1'. O o o H N. F O N l ow owl (9d) aanssaad z 33 resumption of methanogenesis at a rate comparable to that in control sediments (Figure 1). This corresponded with an increase in the hydrogen partial pressure which, after an initial accumulation, stabilized at partial pressures similar to those in control sediments. Molybdate had no effect on the hydrogen partial pressure in control sediments (data not shown). Since MB maintained their potential to metabolize hydrogen in sulfate-amended sediments, a suitable inhibitor that would prevent MB from taking up added hydrogen but would not inhibit hydrogen uptake by SRB had to be found before kinetic analysis of hydrogen uptake by SRB could be made. Chloroform (0.003% V/V) inhibited methane production but had no significant effect on the potential of SRB to metabolize hydrogen, as measured by the rate of sulfate reduction or the rate of hydrogen uptake (Table 3). Sulfate-amended sediments had a higher potential for hydrogen uptake than control sediments (Figure 2, Table 4). The addition of chloroform to the control sediments resulted in the accumulation of hydrogen as previously shown (12), but in sulfate-amended sediments a significant potential for hydrogen uptake remained (Figure 2, Table 4). The VM of the population that was inhibited by chloroform in the sulfate-amended sediments can be calculated as the difference between the VM in the sulfate-amended sediments with and without added chloroform. The value obtained, 0.8 mmol H2 per liter of sediment per h, was equivalent to the VM of the control sediments. This indicates that the hydrogen-uptake potential of the MB population was not changed in the sulfate-amended sediment, but that there had been an increase in a hydrogen-consuming potential that was not inhibited by chloroform. 34 .oo~ x .Afiiwaouucoo ca oumu wxmums x oumcnhaoe :ua3 kumouu mucmafivow can Euomouoaco Lugs woummuu mucwEHvom :H oxmuaa :uwouvan mo mums ecu mo aawv I H. mamsvo :ofiuanfizcfi unmouom o .vo«uwa coaumesocfi onu wcauav wouoouoe on fiasco umns coca Hosea mmumu um ceauosvoun ocmnume coon o>mn fiasco muonu oocfim csonw ma coaufinficaw museums new oumafiumo asaucfia.< .ocfi x Hagimueoawoom Houucoo ca mum“ x enemouoano nufiB vmumouu mucwaavom ca oumuv I H. mamsvm :ofiuanfisaa ucmoumm n .ucwaummuu some now mu: “mmmonuamuma ca uouuo wumvcmuw sues com: o ko.ov H.c cm A oxmuab aowouwxm nAo.wv 5.0 cm A coauosemm mummanw nowuoavom ouwmasm pcowuoswoum mcmnuoz wousmmoz mauomouoaeu an coauwawacH ucwouma umuoamuma new .muooavmu mummasm was mcowocmnuma mo Hmfiucouoa manna: cwwouvmp onu co anemouoano mo uoommm .m manna 35 Half-saturation constants, K, for hydrogen uptake were lower in sulfate-amended sediments than in control sediments (Table 4). When the MB in sulfate-amended sediments were inhibited with chloroform, the resultant K was 3-fold lower than the K in control sediments. When the results of kinetic analyses on sediments collected throughout the summer of 1981 from both the A and B site were compiled, the overall mean K values and 95% confidence intervals for hydrogen uptake not inibited by chloroform was 141 1:33 Pa (n=8). This compared with the K for MB in control sediments of 597 :_186 Pa hydrogen (n=8). The theoretical progress curves of hydrogen uptake in sulfate-amended sediments that were calculated from the two-term Michaelis-Menten expression (equation 3) closely corresponded with those observed experimentally (Figure 2). For these calculations VMSRB and KSRB were taken as the mean values from the chloroform-treated, sulfate-amended sediment. It was assumed that VMMB was equal to 0.8 mmoles H2 per liter per hour, as calculated above, and that KMB was equal to the K in control sediments. DISCUSSION: The fact that the inhibition of sulfate reduction in sulfate-amended sediments resulted in an increase in the hydrogen partial pressure and methane production rates to levels found in methanogenic sediments demonstrated that when sulfate concentrations were not limiting SRB inhibited methane production by lowering the hydrogen partial pressure below a threshold level necessary for hydrogen utilization by MB. The inhibition of methane production was not due to the toxic presence of sulfate or sulfide, as previously demonstrated (1,2,6,l4,29), nor to the depletion of some factor other 36 Figure 2. Typical hydrogen uptake progress curves. Symbols: 0, sulfate-amended sediments; X, sulfate-amended sediments treated with chloroform;., control sediments. T represents expected hydrogen partial pressure in sulfate-amended sediments calculated from equation 3 and the appropriate kinetic parameters as described in the text. 37 F e >< e b I 12 x e 10 “.91 I I 0. 0°. ~0 V o o 0' (9d!) EHnSSEHd I Z “l o H 0 TIME (hr) 38 .u: you acoEHvom mo umuHH non comouvm: mo onoEHHHHZA .ucwEummuu some you wo>uno mmouwoua oumoHHaHuu aouw monam> mo uouum wumvsmum can amaze Showouoano Lugs woumoua 3 H m: .55 H «.0 Becwfiiflfigm a: H 5 N5 H N; Beamiimumflam E H as To H ad 3538...“ H8280 Anamommav M can: you N: 28v z> mmumuoEmumm oHuoCHx make uamavaw .N ouawam SH csosm mmocu nuaa haucouusocoo can mums wm>pso wmmuwoum .muHm < ecu Bonn cwuowaaou mucmawvmm GH wxmumn cowouwzn how mumuoamumq oaumcHM ”q canoe 39 than the electron donors necessary for methanogenesis. This conclusion was further supported by the comparable VM for hydrogen uptake by MB in control and sulfate-amended sediments. Thus, the inhibition of methane production by added sulfate differs from the inhibition by oxygen (31) or nitrogen oxides (3) where the added electron acceptor or a product of its metabolism directly inhibits MB. The inhibition of methane production at low hydrogen partial pressures was probably due to the decreased energy yield from methane production. The available free energy for methane production from hydrogen was calculated from the standard free energy of -139.23 kJ (30) and the methane, hydrogen, and carbon dioxide partial pressures to be -16.3 to -16.8 kJ per mole of methane produced in the control sediments shown in Fig. 1. The calculated free energy was approximately -6.7 kJ per mol of methane produced in the sulfate-amended sediments during the initial days of the inhibition of methane production and +4.7 kJ at 6-7 days after the sulfate addition. Though care must be taken in extrapolating from bulk phase pool sizes to those actually experienced by the bacteria, it is clear that the hydrogen partial pressure in the sulfate-amended sediment were sufficiently lowered to significantly reduce the energy available for methane production from hydrogen. The lower hydrogen pool in the sulfate-amended sediments was associated with the lower overall K for hydrogen uptake and, specifically, with the low K for hydrogen uptake by the bacterial population that was not inhibited by chloroform. The K for hydrogen uptake in chloroform-treated, sulfate-amended sediments is considered to represent the K for the SRB population because: (i) there was no 4O detectable hydrogen uptake in the presence of chloroform in sediments not amended with sulfate; (ii) chloroform did not affect hydrogen uptake by SRB; and (iii) molybdate inhibited the hydrogen uptake in sulfate-amended sediments that chloroform did not inhibit. The K for the MB reported here is within the range estimated independently for methanogenic sediments and other methanogenic environments, such as sludge digestors and the rumen (J. A. Robinson and J. M. Tiedje, submitted for publication). Though there was a possibility of hydrogen uptake by bacteria fermenting hydrogen and carbon dioxide to acetate, the importance of these bacteria in methanogenic environments is low relative to methanogens (5,12). The conclusion that MB and SRB were the only two important hydrogen-consuming populations is further supported by the observation that the total hydrogen uptake in the sulfate-amended sediments could be predicted by using the K for the sulfate-depleted control sediment as the K for the population inhibited by chloroform. Under steady state conditions in environments, such as sediments, where there is negligible physical removal or dilution of the microbial pOpulation the substrate pool size can be described by: K S (VM ' y/k) - 1 (A) where y and k are yield and mortality constants and K and VM are expressed on a per cell basis (4,15). Thus, the hydrogen partial pressure should be dependent solely upon the physiological characteristics of the hydrogen-consuming populations. In the sulfate-amended sediments, the lower SRB K for hydrogen uptake (and possibly a higher yield and VM per cell) resulted in a lower hydrogen pool. Some of the inhibition of methane production in sulfate-amended 41 sediments may be attributed to the metabolism of substrates by SRB rather than proton-reducing bacteria and the subsequent lower rates of hydrogen production (6). However, the maintenance of a lower hydrogen partial pressure by SRB that consumed hydrogen was the ultimate cause of the complete inhibition of methane production since the hydrogen partial pressure was independent of the rate of hydrogen production. The maximum potential rate of substrate uptake is equally important as the affinity for substrate in determining the outcome of competition (9). The slow inhibition of methane production in Wintergreen Lake sediments amended with 20 mM sulfate can be explained by the small initial potential for hydrogen uptake of SRB. In freshly collected sediments incubated with saturating hydrogen, the turnover time for 1 mM sulfate (a saturating sulfate concentration) is 204 hours (25). Assuming that all the sulfate reduction was due to hydrogen uptake this yields a maximal VM estimate for the SRB population of 19.6 umol of hydrogen per liter of sediment per h. With the estimate that hydrogen is the precursor for approximately 40% of the methane production in these sediments (12), the rate of hydrogen production can be calculated from the methane production rate (Table 1) as 64 meI per liter of sediment per h or 3-fold higher than the SRB V“ for uptake. Using the VM and K for the MB, the K for the SRB, and the hydrogen partial pressure determined in the present study, it can be calculated from equation 3 that at saturating sulfate concentrations SRB would initially be able to use at most only 10 percent of the total hydrogen consumed by the two populations. Since the in situ sulfate concentration in these sediments is typically at 42 or below the SRB K for sulfate reduction (24), the limitation of SRB by sulfate can be expected to lower the SRB maximum potential for hydrogen uptake (22) and result in an in §i£2_hydrogen uptake by SRB that is much less than 10 percent of the total hydrogen turnover. This result calculated from kinetic parameters agrees well with previous conclusions derived from experimental results (12). MB are able to compete successfully with SRB in Wintergreen Lake sediments despite the lower SRB K for hydrogen uptake because the maximal potential for hydrogen uptake by SRB is limited by sulfate availability. The competition between SRB and MB for acetate is expected to have similar mechanisms as those for hydrogen competition. MB and SRB should coexist in other anaerobic sulfate-containing environments in which the rate of sulfate supply supports a potential for hydrogen and acetate uptake by SRB that is lower than the rate of hydrogen and acetate production. 10. 43 LITERATURE CITED Abram, J. W., and D. B. Nedwell. 1978. Inhibition of methanogenesis by sulphate reducing bacteria competing for transferred hydrogen. Arch. Microbiol. 117:89-92. Abram, J. W., and D. B. Nedwell. 1978. Hydrogen as a substrate for methanogenesis and sulphate reduction in anaerobic saltmarsh sediments. Arch. Microbiol. 117:93-97. Balderston, W. L., and W. J. Payne. 1976. Inhibition of methanogenesis in saltmarsh sediments and whole-cell suspensions of methanogenic bacteria by nitrogen oxides. Appl. Environ. Microbiol. 33f264-269. Billen, G., C. Joiris, J. Wijnant, and G. Gillain. 1980. Concentration and microbiological utilization of small organic molecules in the Scheldt Estuary, the Belgian coastal zone of the North Sea and the English Channel. Estuarine Coastal Mar. Sci. 1}}279-294. Braun, M., S. Schoberth, and G. Gottschalk. 1979. Enumeration of bacteria forming acetate from H2 and C02 in anaerobic habitats. Arch. Microbiol. 120:201-204. Bryant, M. P., L. L. Campbell, C. A. Reddy, and M. R. Crabill. 1977. Growth of Desulfovibrio in lactate or ethanol media low in sulfate in association with Hz-utilizing methanogenic bacteria. Appl. Environ. Microbiol. 3§}1162-1169. Cappenberg, Th. E. 1975. A study of mixed continuous cultures of sulfate-reducing and methane-producing bacteria. Microb. Ecol..2:60-72. Claypool, G. E., and I. R. Kaplan. 1974. The origin and distribution of methane in marine sediments, p. 99-139. .EE.I° R. Kaplan (ed), Natural Cases in Marine Sediments. Plenum Publishing Corp., New York. Healey, F. P. 1980. Slope of the Monod equation as an indicator of advantage in nutrients competition. Microb. Ecol. 5:281-286. Jérgensen, B. B. 1978. A comparison for methods for the quantification of bacterial sulfate reduction in coastal marine sediments. I. Measurements with radiotracer techniques. Geomicrobiol. J. 1:11-27. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 44 King, G. M., and M. J. Klug. 1982. Comparative aspects of sulfur mineralization in sediments of a eutrophic lake basin. Appl. Environ. Microbiol. 43: Lovley, D. R., and M. J. Klug. 1982. Intermediary metabolism of organic matter in the sediments of a eutrOphic lake. Appl. Environ. Microbiol._43:552-560. Mackie, R. I., and M. P. Bryant. 1981. Metabolic activity of fatty acid-oxidizing bacteria and the contribution of acetate, propionate, butyrate and COZ to methanogenesis in cattle waste at 40 and 60°C. Appl. Environ. Microbiol..4l:1363-1373. Martens, C. S., and R. A. Berner. 1977. Interstitial water chemistry of anoxic Long Island Sound sediments. 1. Dissolved gases. Limnol. Oceanogr. 22:10-25. McCarty, P. L. 1972. Energetics of organic matter degradation, p. 91-118. ln_R. Mitchell (ed.), Water pollution microbiology. John Wiley and Sons, Inc., New Ybrk. Molongoski, J. J., and M. J. Klug. 1980. Quantification and characterization of sedimenting particulate organic matter in a shallow hypereutrOphic lake. Freshwater Biol._ig:497-506. Molongoski, J. J., and M J. Klug. 1980. Anaerobic metabolism of particulate organic matter in the sediments of a hypereutrophic lake. Freshwater Biol._lQ:507-518. Mountfort, D. 0., and R. A. Asher. 1981. Role of sulfate reduction versus methanogensis in terminal carbon flow in polluted intertidal sediments of Waimea Inlet, Nelson, New Zealand. Appl. Environ. Microbiol. 42:252-258. Mountfort, D. 0., R. A. Asher, E. L. Mays, and J. M Tiedje. 1980. Carbon and electron flow in mud and sand flat intertidal sediments at Delaware Inlet, Nelson, New Zealand. Appl. Environ. Microbiol..39:686-694. Oremland R. S., and M. P. Silverman. 1979. Microbial sulfate reduction measured by an automated electrical impedance technique. Geomicrobiol. J. 1:355-372. Oremland, R. S., and B. F. Taylor. 1978. Sulfate reduction and methanogenesis in marine sediments. Geochim. Cosmochim. Acta .42:209-214. Ramm, A. E., and D. A. Bella. 1974. Sulfide production in anaerobic microcosms. Limnol. Oceanogr._lg:110-ll8. Segel, I. H. 1975. Enzyme kinetics. John Wiley and Sons, Inc., New York. 24. 25. 26. 27. 28. 29. 30. 31. 45 Smith, R. L., and M. J. Klug. 1981. Reduction of sulfur compounds in the sediments of a eutrOphic lake basin. Appl. Environ. Microbiol. 4i:1230-1237. Smith, R. L., and M. J. Klug. 1981. Electron donors utilized by sulfate reducing bacteria in eutrOphic lake sediments. Appl. Environ. Microbiology 42:116-121. Sprensen, J., D. Christensen, and B. B. J9rgensen. 1981. Volatile fatty acids and hydrogen as substrates for sulfate-reducing bacteria in anaerobic marine sediment. Appl. Environ. Microbiol. 42:5-11. Strayer, R. F., and J. M. Tiedje. 1978. Kinetic parameters of the conversion of methane precursors to methane in a hypereutrophic lake sediment. Appl. Environ. Microbiol. .35:330-340. Tabatabai, M. A. 1974. Determination of sulfate in water samples. Sulfur Inst. J. 19:11-13. Winfrey, M. R., and J. G. Zeikus. 1977. Effect of sulfate on carbon and electron flow during microbial methanogenesis in freshwater sediments. Appl. Environ. Microbiol. 33:275-281. Zehnder, A. J. B. 1978. Ecology of methane formation. p. 349-376. EE.R' Mitchell (ed.) Water pollution microbiology, vol. 2. John Wiley and Sons, Inc., New Ybrk. Zehnder, A. J. B., and K. wuhrmann. 1977. Physiology of a Methanobacterium strain A. Z. Arch. Microbiol. 111:199-205. CHAPTER IV Factors Controlling the Relative Importance of Sulfate Reduction and Methanogenesis in Lakes of Varying Tropic Levels Introduction The factors controlling the relative distribution of methanogenesis and sulfate reduction in freshwater sediments are poorly understood. In nitrate-depleted, anaerobic sediments hydrogen and acetate are the primary fermentation intermediates (12). Thus, the proportion of carbon and electron flow that proceeds through methanogenesis and sulfate reduction can be expected to be primarily dependent on the partitioning of these two resources between the methanogenic bacteria (MB) and the sulfate reducing bacteria (SRB). It has been demonstrated that freshwater SRB can utilize hydrogen and acetate and can outcompete MB for these substrates when sulfate is increased to millimolar concentrations (13, 28). Studies in freshwater sediments not supplemented with sulfate have indicated that SRB are not competitive with MB at in_§i£2 freshwater sulfate concentrations. In eutrOphic lake sediments SRB utilize less than 10% of the hydrogen production and less than 15% of acetate production (12, 13, 24). Inhibition of sulfate reduction does not result in a stimulation of methane production (24) or the accumulation of acetate and hydrogen (12) which further indicates that SRB are consuming an insignificant fraction of the total hydrogen and acetate production. Sulfate reduction rates in eutrophic sediments are less than 15% of 46 47 the methane production rates (13, 31). Others have suggested that SRB are net producers of acetate and hydrogen in freshwater sediments (11, 28). The restriction of previous studies to eutrOphic lake sediments may have underestimated the ability of SRB to compete with MB in freshwater. The sulfate gradient into eutrOphic lake sediments is very steep (18, 23, 38) and the most active zone of sulfate reduction is probably confined to the most superficial sediments (8). Present methods are not adequate for sampling this zone and maintaining in. situ sulfate concentrations. If SRB can effectively compete with MB at freshwater sulfate concentrations then the relative importance of sulfate reduction and methanogenesis as terminal processes in sediment metabolism should be related to the proportion of the zone of actively metabolizing sediment in which sulfate is available to SRB. The depth that sulfate penetrates should in turn be related to the rate of sulfate flux into the sediment and the rate it is consumed within the sediment. Materials and Methods Sediment Sampling and Incubation: Lawrence Lake is a small hardwater lake located in southwestern Michigan. The physical, chemical and biological characteristics of the lake have been described in detail (26). Profundal sediments were sampled from within the 12 meter depth contour. Oxygen was undetectable in the overlying water with the Winkler technique during the sampling period (R. G. Wetzel personal communication). Sulfate concentrations in the interstitial water with depth were determined from dialysis samplers (5) with ports spaced at 1 cm intervals. For 48 measurements of methane production with depth sediments were collected with a gravity corer as previously described (12). Subcores were taken at 2 cm intervals with cut-off 3 ml plastic syringes which were then sealed with a rubber stOpper and incubated underwater at the in situ sediment temperature during transport. The sediments were extruded into 10 ml Vacutainers (Becton-Dickenson) and treated as outlined below. In all subsequent experiments sediments were sampled with an Eckman dredge. Care was taken to collect only the floculent surface sediments and associated water. The sediments were transported on ice to slow sulfate depletion in sealed containers with no gas headspace. Sediments were transferred under a nitrogen atmosphere to anaerobic pressure tubes (Bellco Class) or 10 ml Vacutainers (Becton-Dickenson). Incubations were at 10°C in the dark. Methane production was measured as the rate of accumulation of methane over time. The tubes were vigorously shaken to equilibrate dissolved methane with the headspace before samples were taken. Sulfate reduction rates were estimated from the rate of sulfate depletion in the sediment over a 3.5 hour incubation period. Replicate samples were sacrificed over time and centrifuged to collect the interstial water which was analyzed for sulfate. Solutions of molybdate, acetate and sulfate were flushed with nitrogen and added to the sediments with a syringe and needle. The added volume was less than one percent of the total sediment volume. Molybdate was added to the sediment as the sodium salt to a final concentration of 1 or 2 mM. For hydrogen uptake experiments 100 ml of sediment were placed in 120 ml serum bottle (Wheaton Scientific). Additions of chloroform 49 (0.003% V/V) or molybdate were made prior to transferring 5 ml aliquots into pressure tubes. The sediments were incubated with slow rolling on a tube roller to create a thin film of sediment. The sediments were allowed to equilibrate 1-3 hours. Hydrogen was added to the headspace to give an initial partial pressure of approximately 450 pascals. The rate of hydrogen uptake was measured over a 15 or 30 minute time period. Radioactive tracer studies: Solutions of radioactive compounds were pre-flushed with nitrogen. The final volume of the added solution was five percent or less of the sediment volume. The concentration of the added compounds was less than one percent of the in §i£u_pool size. [2-14C] acetate (54 mCi mmol‘l) and [14C]-sodium bicarbonate (0.1 mCi mmol’l) were purchased from New England Nuclear. The kinetics of acetate uptake were determined using the method of Wright and Hobbie (30). Various final concentrations of acetate were added to sediments along with [2-1401 acetate. The data were analyzed according to the equation of Hobbie and Crawford (7): = (K+ Sn) +5 v v (1) .E f where t is the incubation time, f is the fraction of the added activity converted to product, Sn is the iani 2 substrate concentration, A is the amount of substrate added, V is the maximum uptake velocity and K is the substrate concentration that gives half-maximal velocity of uptake. A plot of t/f versus A yields a straight line with an x-intercept of - K+Sn if uptake follows MichaeliséMenten kinetics. 50 Analytical Techniques: Methane concentrations were measured on a Varian 600 D gas chromatograph with a flame ionization detector. Cases were separated with a helium carrier on a one meter column of Poropack N (Waters Associates) at 50°C. Hydrogen was analyzed on a Varian 3700 gas chromatograph with a thermal conductivity detector. The gases were separated at 35°C on a 3 meter column of Poropak N with nitrogen as the carrier at 15 ml per minute. 14CH1, and 14C02 were separated on the same column but with helium as the carrier gas and an oven temperature of 45°C. The effluent from the detector was passed through a gas proportional counter for quantification of the radioactivity. Total 14C02 in the tubes was corrected for dissolved ll‘C-inorganic carbon with an empirical factor that was determined from the distribution of H14C03“ added to sediment. Sulfate was analyzed by high pressure liquid chromatography. Ions were separated on a 5 x 0.46 cm Vydac column (Anspec) with a solvent of 1 mM pthallic acid (pH 5.5) at a flow rate of 2 ml per minute at room temperature. Sulfate was detected with a Wescan conductivity detector (Anspec). Modelling of Competition: A computer model, called ACETATE, was developed to simulate competition between SRB and MB for acetate in sediments. Uptake of acetate by MB was described by the Michaleis-Menten equation: V x A x X KA‘i'A where v is the velocity of acetate uptake, V is the maximum rate of 51 acetate uptake per unit biomass, X is the biomass, A is the acetate concentration and KA is the acetate concentration that gives an uptake of acetate one half of V. The uptake of acetate by SRB was modelled with a multiplicative model (22) and was computed as V x A x X x SU KA+A KSU+ SU where SU is the sulfate concentration and KSU is the half-saturation constant for sulfate uptake. Growth was computed as: AX = (v x At x Y) - (b x X) where AX is the increase in biomass over the time interval at, Y is the yield of biomass per quantity of acetate taken up and b is the mortality coefficient. The parameters used in the model are listed in Table (1). The KA estimates were from the kinetic analysis of acetate uptake by MB and SRB in Lawrence Lake sediments. The KSU was from the sulfate half-saturation constant for freshwater sediments determined previously (23). The mortality coefficient was estimated from the decline of the maximum potential for methanogenesis in freshwater sediments in which methane production was inhibited with sulfate (unpublished data). Theoretical yields were calculated from thermodynamic considerations using the method of McCarty (15, 16). The maximum rates of acetate uptake per unit biomass were calculated from the above parameters and the expected relationship between substrate pool size and growth parameters at steady state (1, 16). Sulfate flux from the overlying water was assumed to be controlled by diffusion. 52 Table 1. Growth parameters for the simulation model of methanogen and sulfate reducer competition for acetate in sediments.8 Parameter Methanogens Sulfate Reducers KA (HM) 50 3 KSU (11M) - 68 V (pmoles x mg cells"1 x min-1) 0.0109 0.0069 Y (mg cells x mmol acetate’l) 2.136 3.84 b (min'l) 7 x 10"6 7 x 10"6 a The basis for the parameters is outline in the text. 53 where F is the rate of flux and AC is the change in sulfate concentration over the depth interval AZ. D, the diffusion coefficient, was assumed to equal 1.5 x 10"5 cm2 x sec"l which is the middle of the range found by Hessleian (6) for lake sediments in summer at 20°C. Acetate uptake by MB and SRB with depth was modelled by assuming an active depth of acetate production of 3 cm which approximates the depth concluded to be the most active in sediment decomposition (9,12). The rate of acetate production was assumed to be uniform with depth. The sulfate concentration in the overlying water was assumed to remain constant. The sediment was divided into 1 mm layers and individual layers were assumed to be homogeneous. Sulfate flux into each layer from the overlying layer, the sulfate flux out of the layer to the next deeper layer, the amount of acetate produced, the acetate and sulfate uptake and the growth of MB and SRB were calculated at one minute time intervals. The model was generally run for a time equivalent to greater than five years. This time was sufficient to approach as steady state as indicated by insignificant changes in the rate, concentration and biomass parameters over a modelled time period of approximately one year. Therefore the results of the model were not a true steady-state solution but were considered to be a close approximation and sufficient for the purposes of this study in view of the fact that the sediments being modelled would never be expected to be at steady-state. 54 Results In Situ H2 and Acetate Utilization by SRB: Methane production rates were highly variable in subcores of Lawrence Lake profundal sediments taken at the same depth and time (Figure 1). A high day to day variability was also noted especially at the lower depths examined. Sulfate concentrations decreased rapidly with depth indicating active sulfate reduction in the surface sediments (Figure 1). To obtain sediments with the highest sulfate concentrations and to reduce variability within a sediment sample for a given date subsequent studies were performed with well-mixed samples of surface sediments. The addition of molybdate to the sediment completely inhibited sulfate depletion and stimulated methane production (Table 2). The extent of molybdate stimulation of methane production was variable with increasing stimulation on successive dates. Molybdate had no significant effect on methane production in sediments that were preincubated to deplete the sulfate pool (Table 2). The addition of sulfate inhibited methanogenesis suggesting that the potential of SRB to compete with MB was limited by sulfate availability. The increase in the number of moles of methane produced in molybdate-treated sediments was 96% of the number of moles of sulfate reduced in untreated sediments (Table 3). Since the electron equivalents required for the production of methane and the reduction of sulfate are equal (25) this indicates that carbon and electron flow that was blocked from going through sulfate reduction proceeded through methanogenesis. Additions of molybdate increased the production of 14CH4 from both [2-14C] acetate and 14C02 (Table 4). 55 Figure 1. Methane production mean and range of standard error on 12 August., and 13 August-and sulfate concentrations with depth on 24 August at the 12.5 m depth site of Lawrence Lake. 56 .2: zofififizmozoo «ow _ _ _ _ _ _ _ _ I ‘ II I II liltlill I ll ‘ I. l 4 Tiblle III 1 ll ‘ l .l C II I. hzm—zamw ifi 03.. I ‘ II T G I. _ _ _ _ _ _ _ h 09 oo o w v o 5.230.081 20:0300ma eIO “‘3 HldECl 57 .Hooo oumwasw ozu ouoaooo ou xumw ecu :H woumnsocH ucweHwom w .vocHsuouoe uoz o .mowocucmuwo :H 0H mommaam moons any mo :oHumuucoocoo n .HIAmucoEHwow woummuucw :H :oHuoovoua ocmnumzv x AucoEHvow woummuu :H coHuosvoua ocmnuozv m II vmumaaoa 02 N0; :0 + 00.0 m NV £0 03830 I. wouoaoom 02 00; 00.0 + 0~.~ m NV 3.2 03830 02 2.0 8.0 H .3; m 02 3 30a 02 00.0 00.0 H 3.0 0 m2 0N 23 £5 03 me. am; :4 H 00.0 0 2 2.0 020 ca. 0.8 00. I ca 2: 00. 3.0 004 + 33 m 00 0:. 020 02 S; 3. H 0.2 0 002 mm 20 boom fins nqooz 53 35103903 0025030. sum 93 Ah: won—mm some ucmEHvow emumouuco :H moumoHHaom :OHumuucoocou :OHumnoocH mOHumm oumm oumm :oHuozooum mcmnumz mo nonasz oumeSm HmHuHcH :OHuoacoum ocmnuwz .muaoanom oomwusm oxma mucousmg CH :oHuoovoua ocmnuoe co meowquwm mummaom cam mumvnzaoa mo uoommm .N oHan 58 Table 3. Methane production and sulfate reduction in control and molybdate-treated Lawrence Lake sediments over a 20 hour incubation period. Methane Produced Sulfate Reduced (umoles 1-1)a (umoles 1‘1) Control 19.4 (1.8) 58.9 (2.9) Molybdate-treatedb 75.7 (8.2) O Differencec -56.3 58.9 a Mean and standard error in parentheses; n = 5 for methane production; n = 4 for sulfate reduction. b Molybdate final concentration 1 mM. 9 (Control sediments)-(Molybdate-treated sediments). 59 Table 4. Effect of molybdate addition on the conversion of [2-14C] acetate and 14C02 to 14CH4. Percentage Increase with Molybdate Additiona Total Methane 14CH4 from 14CH4 from Production [2-14C1-acetate 14C02 48 (.28)b 74 (.30) NDC 435 (.09) ND 565 a Total methane production determined over a three hour incubation period and 14CH4 production over a 1 hour incubation period for experiment with [2-14C]-acetate. Total methane and 14CH4 production determined over 20 hour period for experiment with 14C02. The sediments for the two experiments were collected on separate dates in October. Percentage increase equals (M004: treated sediments-control sediments) x control"1 x 100. b Coefficient of variation for difference between molybdate-treated and untreated sediments in parentheses. 9 Not determined. 60 From the information that the increase in methane production in molybdate-treated sediments corresponded with the rate of sulfate reduction in untreated sediments it could be calculated from the data in Table 2 that sulfate reduction comprised from 30-81% (mean = 50) of the total of the methane production and sulfate reduction rates. Direct simultaneous measurements of sulfate reduction and methane production in early October (Table 6) indicated that sulfate reduction comprised 72% of the total of both rates. The percentage of [2-14C]-acetate metabolized to 14C02 was 86% and ranged from 40-97% (mean = 74%) throughout the study period. The uptake of added hydrogen was inhibited by chloroform or molybdate alone and the effect of the inhibitors was additive (Table 5). However, even when methane production and sulfate reduction were completely inhibited there was still a significant rate of hydrogen uptake. Hydrogen uptake in the presence of the inhibitors continued even when the sediments were preincubated in the dark to deplete sulfate (less than 2 pM) and presumably other more thermodynamically favorable electron acceptors such as nitrogen oxides and iron (Table 5). Kinetic analysis of [2-140] acetate turnover in freshly collected sediments with a sulfate concentration of 56 uM indicated that the K + Sn for l”'COZ production, 5 uM, was significantly lower than the K + Sn of 33 pH for 14CH4 production (Figure 2). Since prior experiments indicated that the addition of molybdate resulted in a 90% inhibition of the rate of 14C02 production from [2-140]-acetate, the kinetics of acetate metabolism to 14002 primarily represented the kinetics of the SRB. To further ensure that the kinetics for 14C02 production were due to sulfate reduction, sediments were preincubated in the dark for 61 Table 5. Effect of inhibitors on hydrogen uptake in freshly collected and sulfate-depleted sediments. Additions Percent Sediment Type CHCl3 M004 ka Inhibition Fresh Sediment O O .74 :.'14 - Fresh Sediment + 0 .55 :_.02 26 Fresh Sediment O + .55 :_.05 26 1 Fresh Sediment + + .39 i .06 47 SO4-Depleted : Sedimentb 0 0 .84 i .12 - 7*! SO4-Depleted Sedimentb + + .65 + .08 23 a k is the first order rate constant of hydrogen where So-Sf t k = ( So+Sf -l concentrations at the initial time point and at time t. standard deviation of triplicate determinations given. ) x ( 2 ) and So and Sf are the hydrogen Mean and b SO4-depleted sediment was a subsample of the fresh sediment that was stored in the dark for 5 days and had a 804 concentration less than 2.pM. Sediments preincubated in the dark for 4 weeks showed similar results. 62 Figure 2. Metabolism of 2-14C] acetate in Lawrence Lake surface sediments to 1 CH4 and 14C02 at different concentrations of added acetate. Error bars represent standard error of the mean. Standard error for 14C02 is contained within the area of the symbols. 63 .2... m.._.<._.m0< DMDD< ow o ONI owi cm+x _ _ ow ON— oom SHHOH ill 64 approximately two weeks to deplete electron acceptors for acetate oxidation. The sediments were then amended with 500 uM (final concentration) calcium sulfate three days prior to kinetic analysis. The sulfate concentration on the day of the analysis was 350 nM. The K + Sn for 1[‘CH1, and 14002 production were 65 uM and 3 uM respectively. Effect of total sediment metabolism on competition: Comparison of initial rates of sulfate reduction and methane production and the fate of [2-14C] acetate that were measured simultaneously in Lawrence Lake sediments with data from other lake sediments indicated that sulfate reduction was of greater relative importance in the terminal metabolism of Lawrence Lake sediments than in sediments that had higher total rates of methane production and sulfate reduction (Table 6). To examine whether this difference could be explained on the basis of resource competition between SRB and MB and the overall rate of organic matter decomposition in the sediments, the simulation model, ACETATE, was developed and simulations of the activity of SRB and MB within sediments with different rates of carbon metabolism and concentrations of sulfate in the overlying water were run. The model indicated that partitioning of total terminal metabolism between sulfate reduction and methanogenesis was related to the depth in the sediment that sulfate remained above the minimal threshold concentration necessary to support sulfate reduction. As can be seen by comparing the two profiles in Figure 3, sulfate remained above threshold levels at deeper depths in sediments with lower rates of acetate production which permitted SRB to compete with 65 .oz “womoaaew venues zuHa oHnouoouow uoc oumwaom m .<2 moHanHm>m go: some H ANV muoncoaamo Boom some : .Ammv mameN was zouwcas no a .me Eoum some w .AfimV scum wowcmu eoHumuucoocoo was mouse coo: w .muoumuonma mHnu we come eonmaansacs can Amm .mfi .NHV aoum some o .uonouoo manna CH wouooHHoo ucoaHoom cH mu: m.mto mouse HmHuHcH v .mucoEHoow commune c« was ucoEHwom ecu wcH>HHo>o woumz 6H ouowazm o .oumuoom ~0q~INH Boom vooswoun NovS can «20¢d HmuOu mo «:03 no N no venom a .uzo: you uamefinom mo nouHH use moHosn cH commouaxo oumm m oz ooHIo on <2 o.¢m mic scounoo> noz com #0 <2 <2 Nio wmuovcoz Onion oomiomd H<2 0.2 0.02 ofiio mommamumom m ouHm owioq oamiowfi no 0.0 cm NIo ocoouwuouaHz < ouHm 00:00 00~i0m 00 N . 0 00 N0 00033353 nmuw OHHIOm omHIoHH «a c.< w.~ oommuom bouzouamq ucmEHvom nouns nocmnuoz cu maoauosoom mcouuonooum Aaov used : ooNHHonouoz mummasm ocmzuoz HocuoucH 0A2 v 00m mumuoo< N nuaon .:OHumonHumuum uoaaom wcHuzv magma Hmuo>om mo mucoEHnom ecu CH mumuoamumo noumfiou was coHuoseou oumwaom .coHuosooua oomnuoz .c macaw 66 MB over a greater prOportion of the total zone of active decomposition. The total rate of sulfate reduction in sediments with higher rates of acetate production was greater than the rate in sediments with the same sulfate concentration in the overlying water and lower rates of acetate production. However, the relative importance of sulfate reduction to total terminal metabolism was greater in the sediments with lower acetate production. More comprehensive results of the model are presented in Figure 4. As the rate of total terminal metabolism was increased there was a sharp initial increase in the relative importance of methane production followed by a more gradual increase (Figure 4). The exact shape and position of the curve depended on the sulfate concentration in the overlying water; the lower the sulfate concentration the more abrupt the initial increase and the higher the proportion of total terminal metabolism proceeding through methanogenesis. When depth intervals other than 0-2 cm were considered the relationship between methane production and total terminal metabolism was similar to that shown in the figure except that for given values of total terminal metabolism and sulfate in the overlying water the relative importance of methane production increased as the depth of the sediment interval was increased. The relative importance of methane production to total terminal metabolism proceeding through methane production and sulfate reduction in Lawrence and Wintergreen Lakes followed a pattern similar to that predicted with the simulation model (Figure 4). The Lake Bostalsee data did not appear to fit the general trend but this was probably due to the fact that the rate measurements were made on mixed sediments Figure 3 e 67 Examples of methane production, sulfate reduction and sulfate concentration profiles from simulation model. The simulations depicted in both panels had 100 uM sulfate in the overlying water. The acetate production rates were 3 and 46 umoles per liter per hour for the sediments in panels A and B respectively. 68 PIE x .2: ZO_._.ODn_m_m m.._._ r as ezmzamm 55:5 < _ _ _ ow ow o ow ow o .2... ZO_._.