APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1981, p. 775-7820099-2240/81/030775-08$02.00/0

Vol. 41, No. 3

Fate of Immediate Methane Precursors in Low-Sulfate, Hot-Spring Algal-Bacterial Mats

KENNETH A. SANDBECK AND DAVID M. WARD*

Department ofMicrobiology, Montana State University, Bozeman, Montana 59717

The fates of acetate and carbon dioxide were examined in several experimentsdesigned to indicate their relative contributions to methane production at varioustemperatures in two low-sulfate, hot-spring algal-bacterial mats. [2-'4C]acetatewas predominantly incorporated into cell material, although some "CH4 and14C02 was produced. Acetate incorporation was reduced by dark incubation inshort-term experiments and severely depressed by a 2-day preincubation indarkness. Autoradiograms showed that acetate was incorporated by long filamentsresembling phototrophic microorganisms of the mat communities. [3H]acetatewas not converted to C3H4 in samples from Octopus Spring collected at theoptimum temperature for methanogenesis. NaH14CO3 was readily converted to14CH4 at temperatures at which methanogenesis was active in both mats. Com-parisons of the specific activities of methane and carbon dioxide suggested that ofthe methane produced, 80 + 6% in Octopus Spring and 71 + 21% in WiegertChannel were derived from carbon dioxide. Addition of acetate to 1 mM did notreduce the relative importance of carbon dioxide as a methane precursor insamples from Octopus Spring. Experiments with pure cultures of Methanobac-terium thermoautotrophicum suggested that the measured ratio of specific activ-ities might underestimate the true contribution of carbon dioxide in methanogen-esis.

Hot-spring algal-bacterial mats represent anatural environment where thermophilic anaer-obic decomposition of organic matter may bestudied. Algal mats of alkaline, siliceous hotsprings have also been suggested as modernanalogs of the environments in which siliceousPrecambrian stromatolites once formed (7, 28).Thus, studies of their formation and decompo-sition may aid in interpretation of early lifeforms preserved as microfossils. Previous studiesindicated the complete decomposition of algal-bacterial organic matter (7), but detailed studiesof anaerobic decomposition are lacking (4, 37).A complete anaerobic food chain was suggestedby the observation of methane production (29)and sulfate reduction (30) as terminal anaerobicprocesses in algal-bacterial mats of low- andhigh-sulfate hot springs, respectively. This studywas directed at determining the importance ofimmediate methane precursors in low-sulfate,hot-spring algal-bacterial mats.

Acetate and carbon dioxide appear to be themajor immediate methane precursors in manyanaerobic environments (17, 36). The relativeimportance of these methane precursors can becompared because acetate conversion to meth-ane occurs by reduction of the intact methylgroup and not via CO2 as an intermediate (17,25, 26, 32). In different environments, the rela-

tive amounts of methane formed from thesesubstrates varies (14, 17, 19, 34). Acetate hasbeen shown to account for 73 to 90% of themethane produced in anaerobic waste digestors,60% of the methane formed in rice paddy soil,and 70% of the methane formed in lake sedi-ments. Carbon dioxide has been shown to ac-count for 14 to 30% of the methane formed inanaerobic waste digestors, 20 to 30% of the meth-ane formed in rice paddies, and 36 to 98% of themethane formed in lake sediments. In contrastto the usual importance of acetate in methano-genesis, only 5.3% of the methane formed in therumen was derived from acetate (21), presum-ably because of its uptake by the animal.

In this paper we report that CO2 is the majorsource of methane in hot-spring algal-bacterialmats. Uptake by other inhabitants of this micro-bial community, possibly photoheterotrophs,may preclude acetate as a major methane pre-cursor.

MATERIALS AND METHODS

Study areas. The major research area used in thisstudy was Octopus Spring, an alkaline hot spring (pH8.0) located about 0.15 km SSE of Great FountainGeyser in the White Creek drainage in YellowstoneNational Park. At this spring, experiments were un-dertaken only in the southernmost effluent channel.

775

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

776 SANDRECK AND WARD

Another research area used in this study was in ameadow, also in the Lower Geyser Basin of Yellow-stone National Park, adjacent to Firehole Lake Drive.Springs in this meadow were collectively referred toas Serendipity Springs because of their chance discov-ery in early 1968 by M. L. and T. D. Brock (8). Thestudy area in the Serendipity Springs group was anartificial effluent channel constructed by Fraleigh andWeigert (8) by diverting the effluent of a spring sothat it constantly flowed down a plywood channel (1.2m wide by 24 m long). The pH of the piped in waterranged from 6.0 to 7.0 (8).Some experiments were performed only in the 50 to

55°C region at Octopus Spring and 60°C at the Wie-gert channel where methanogenesis was greatest (K.A. Sandbeck and D. M. Ward, in preparation).

Sampling. Whole cores were removed from thealgal-bacterial mat with a no. 4 brass cork borer (50.3mm2) and transferred directly to 1-dram (ca. 1.8-g)glass vials (Kimble; 14.5 by 45 mm) which were sealedanaerobically (13; except that no copper-reducing col-umn was used in the field) under a stream of 100%helium (Linde). Recessed butyl rubber stoppers (A. H.Thomas; size 00) were used to effect a seal and keepthe vials anaerobic during later manipulations. Stop-pers were secured by lengthwise taping with maskingtape, except during "light" incubation, when tape wasfixed only at the vial neck. During "dark" incubation,vials were additionally wrapped in several layers ofaluminum foil. Anaerobically tubed samples wereeither incubated in the effluent channel or quicklyplaced in insulated coolers that contained water whichwas SoC warmer than the in situ temperature. Duringtransit to the laboratory (ca. 2 h), samples cooledslightly, but this procedure ensured that they wouldbe kept within 5°C of their indigenous temperatures.In the laboratory, samples were transferred to darkincubators that matched the in situ temperature. Alladditions were made from anoxic sterile stock solu-tions, using a 1-ml disposable syringe (Plastipak sy-ringe; Becton, Dickinson & Co.), at the time of samplecollection unless otherwise stated.

Fate of [2-14C]acetate. Replicate vials (two tofour) received either 0.2 ml of a 1-ytCi/ml stock solutionor 0.1 ml of a 10-,uCi/ml stock solution of [2-'4C]acetate(sodium salt, 44 mCi/mmol; New England NuclearCorp.). Unless otherwise stated, biological activity wasterminated after 2 h by the addition of 0.5 ml ofFormalin accompanied by extremely vigorous shakingto ensure that the Formalin permeated the gelatinoussample. Gas headspace subsamples were removed foranalysis of '4CO2, C02, '4CH4, and CH4 as describedbelow. The pH (VWR pH Master pH meter plus glasscombination electrode) and headspace and liquid vol-umes (determined by displacement with water) weremeasured after removal of samples so that resultscould be corrected to a per core basis (see below).Incorporated radioactivity was determined by filteringa 0.1-ml volume of a homogenized sample (Teflontissue homogenizer) diluted with 0.9 ml of distilledwater through a 0.45-,um membrane filter (MilliporeCorp.). After the filtrate had been collected, the filterwas rinsed with 0.5 to 1.0 ml of distilled water, dried,and exposed to concentrated HCI fumes overnight toremove 14C-labeled carbonates. Radioactivity in the

filters and in 0.1 ml of the filtrate was determined in10 ml of Aquasol (New England Nuclear Corp.) on amodel LS 100-C liquid scintillation counter (BeckmanInstruments, Inc.). Correction for differences in count-ing efficiency were made by the automatic externalstandard method. Data were corrected to a per corebasis for comparison with radioactivity in other frac-tions. These methods resulted in efficient recovery ofthe added "4C. Results are presented as a percentageof the recovered radioactivity to eliminate the varia-bility inherent in small-volume additions.Fate of [3Hlacetate. Replicate vials received 0.1

ml of a 20-,uCi/ml stock solution of [3H]acetate (so-dium salt, 2 Ci/mmol; New England Nuclear Corp.).Vials were poisoned with Formalin after 3 h of incu-bation in the effluent channel or 4 days of incubationin a laboratory incubator. Gaseous subsamples wereanalyzed for C3H4 as described in the analytical meth-ods section. Incorporated and unincorporated radioac-tivities were determined as described above.

Conversion of NaH'4CO3 to 14CH4. Replicatevials received 0.1 ml of a 20-,uCi/ml stock solution (pH8.0) of NaH'4CO3 (44.5 mCi/mmol; New England Nu-clear Corp.). CH4, '4CH4, C02, and 14CO2 were followedover time as described in the analytical methods sec-tion. Methane (nanomoles) produced from CO2 wasobtained by dividing '4CH4 disintegrations per minuteper core by the specific activity of CO2 (disintegrationsper minute per nanomole).The relative importance of CO2 as a methane pre-

cursor was determined by calculating the CH4 specificactivity/CO2 specific activity ratio (14). In some ex-periments, hydrogen (0.1 ml of 100% H2 [Linde]) orsodium acetate (0.1 ml to achieve a final concentrationof 1 mM in an estimated 2.5-ml sample) was added asdescribed in the Results section to enhance differentmethane formation reactions (34). In some cases, vialswere flushed with 02-free helium (Linde) before deter-mination of the specific activity ratio (see Results).

Isotopic selection of 12C over 14C in NaH14CO3 re-duction to '4CH4 or the error in measurement of theCH4 specific activity/CO2 specific activity ratio waschecked by similar additions of NaH'4C03 to purecultures of H2-utilizing thermophilic methanogenicbacteria resembling Methanobacterium thermauto-trophicum obtained from 50, 55, 60, and 65°C regionsof the Octopus Spring algal-bacterial mat (Sandbeckand Ward, in preparation). Isolates were grown in amineral salts medium with H2 as the only energysource. Incubation was at the temperature at whicheach isolate was obtained. Determination of the CH4specific activity/CO2 specific activity ratio was doneafter 24 h, before significant depletion of CO2.

Analytical methods. Gas samples were removedfrom the headspace of vials by using a helium flushedGlasspak syringe (Becton-Dickinson) attached to aMininert valve (Supelco) to eliminate loss of sampleowing to pressure differences. Gas subsamples (0.2 ml)were analyzed for CH4, '4CH4, C02, and 14CO2 by gaschromatography-gas proportion analysis, using amethod similar to that of Nelson and Zeikus (20).Details of the method were described by Ward andOlson (30), except that gas concentrations and radio-activity were quantified by using a Spectra-Physicsmodel 4100 computing integrator and a Spectra-Phys-

APPL. ENVIRON. MICROBIOL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

METHANE PRECURSORS IN ALGAL-BACTERIAL MATS 777

ics Minigrator, respectively. C3H4 was detected in thesame way but with the gas proportional counter fur-nace turned off to prevent combustion of methane. Toensure detection by this method, C3H4 was preparedby incubating 2,uCi of [3H]acetate with the contentsof an anaerobic dairy cow manure digester until no

further increases in C3H4 were detected. To ascertainthe limit of detection of C3H4, the response to C3H4was determined. It was calculated that no more than0.29% of the [3H]acetate added to algal-bacterial matsamples could have been converted to C3H4 or it wouldhave been detected. The total amounts of CH4 and14CH4 per core were calculated by comparison of sub-volume to the gas headspace volume. Total amountsof C02 and 14CO2 per core were determined by correc-

tion for the difference between subsample and head-space volumes and also for gas solubility and dissocia-tion equilibria by the method of Stainton (27). Sinceactivity is not proportional to core length (29), com-

parisons were made on a per core basis. The signifi-cance of differences between mean values was deter-mined by a two-sample t test.Autoradiography and microscopy. Autoradi-

ograms of material incubated with [2-'4C]acetate were

prepared by the method of Brock and Brock (3). Athin film of homogenate was smeared onto a glass slide(precleaned glass slides; VWR Scientific) and allowedto air dry. The slides were put through a series of fivedistilled-water rinses (1 min each) to remove any

unincorporated radioactivity. Slides were then dippedfor 5 s in photographic emulsion (Kodak NTB2) undera Kodak no. 2 safelight. Slides were exposed for 24 to26 days in total darkness and then developed in totaldarkness with Kodak D-19 developer and fixed withKodak fixer. Slides were examined with a Leitz Ortho-lux II microscope equipped for interference-contrastoptics. Photomicrographs were taken with a NikonMicroflex model EFM semiautomatic photomicro-graphic attachment at 50OX, using Kodak Panatomic-X film. Exposure time was 0.5 to 1 s. Negatives were

then enlarged on silver bromide print paper (KodaBromide F-4) to achieve better contrast.

Autofluorescence was observed with a Leitz Ortho-lux II microscope equipped with ultraviolet light ver-

tical illumination from an HBO 200-W mercury lamp.The ultraviolet light passed through a Leitz B cubeexcitation-emission filter combination.

RESULTS

Microscopic observations. Surface samplesfrom algal-bacterial mat cores were examined byfluorescence microscopy. The red autofluores-cence of chlorophyll a in cyanobacteria sug-gested the predominance of organisms resem-

bling Synechococcus lividus in both hot springsand the presence of microorganisms resemblingMastigocladus laminosus and Phormidium sp.(4, 5) as minor components of the Wiegert Chan-nel mat. Samples which were incubated anaer-

obically and in which substantial methanogene-sis had occurred also showed the presence ofrod-shaped bacteria which exhibited a blue-green-colored autofluorescence similar to the au-

tofluorescence of M. thermoautotrophicum (36;Sandbeck and Ward, in preparation) isolatedfrom the same mats. No blue-green-fluorescingsarcina- or pseudosarcina-shaped cells were ob-served.Fate of acetate. In an initial experiment

(data not shown), a 2-h incubation of samplesfrom various temperatures in the OctopusSpring mat with 0.2 1LCi of [2-'4C]acetate did notresult in production of '4CH4. Significantamounts of '4C were recovered in the cell frac-tion, and small amounts of '4C02 were detected.When the amount of added [2-14C]acetate wasincreased to 1 PCi, 14CH4 was detected after a 2-h incubation with samples collected at temper-atures at which methanogenesis was active inboth Octopus Spring and Wiegert Channel mats(Table 1). '4CO2 was produced at all tempera-tures in both mats. The predominant fate ofacetate appeared to be its uptake by cells, as 58to 86% of the radioactivity was recovered in thecell fraction. Dark incubation of samples col-lected at 550C in both mats significantly reducedacetate incorporation (P = 0.037 for OctopusSpring; P = 0.007 for Wiegert Channel), sug-gesting that acetate incorporation might be lightdependent. An autoradiogram prepared from [2-14C]acetate-labeled material obtained from Oc-topus Spring (550C) is shown in Fig. 1. The labelwas taken up by very long, filamentous micro-organisms resembling phototrophic microorga-nisms of the mats. Similar results were noted for[2-'4C]acetate-labeled material obtained at othertemperatures in the Octopus Spring and WiegertChannel mats.Dark incubation appeared to stimulate 14CH4

production in both mats (Table 1), but the data

TABLE 1. Fate of [2-'4Cjacetate (1 tCi) incubatedanaerobically for 2 h in hot-spring algal-bacterial

mat cores

% of recovered "C in:Condition

CO2 CH4 Cells Filtrate

Octopus Spring450C 1.59 NDa 74.08 24.34500C 2.53 ND 58.86 38.61550C, light 1.32 0.13 44.71 53.85550C, dark 0.58 0.91 25.79 72.74600C 2.07 0.04 66.21 31.69650C 5.20 ND 66.99 24.81

Wiegert Channel450C 0.98 ND 64.01 35.01500C 0.82 ND 71.45 27.73550C, light 1.01 0.32 86.10 12.57550C, dark 4.15 11.06b 52.81 31.99600C 1.09 0.07 85.24 13.60

aNot detected.b Standard deviation, 11.89.

VOL. 41, 1981

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

778 SANDBECK AND WARD

FIG. 1. Autoradiogram of an Octopus Spring algal-bacterial mat sample collected and incubated anaero-bically for 2 h at 55°C with [2-14C]acetate (1 ,uCi). Magnification, Xl,037.

were insufficient to indicate that the apparentincrease was significant (P = 0.48 for OctopusSpring, P = 0.33 for Wiegert Channel).The possible interaction between acetate-usingmethanogenic bacteria and acetate-incorporat-ing phototrophic microorganisms was furthertested by preincubating samples in darkness for2 days before the addition of [2-'4C]acetate. Asmethanogenesis was ongoing (data not shown),these conditions should have reduced the com-petitiveness of phototrophs, and enhanced thecompetitiveness of methanogens, for acetate.These incubation conditions should also haveeliminated the possibility that oxygen, whichmight have been produced during short-termincubation in the light, inhibited methanogenicbacteria. Small amounts of '4CO2 and 14CH4 wereproduced with time after [2-14C]acetate additionin all samples. After 71 h of incubation with [2-'4C]acetate, a period presumably sufficient forsignificant metabolism of acetate, the distribu-tion of '4C in various fractions was compared(Table 2). The dark preincubation period waseffective in reducing the incorporation of acetate(3 to 8% recovered in the cell fraction), furtherindicating the light dependence of acetate up-take. In all cases, except a single replicate at55°C in Wiegert Channel, 278% of the 14C re-

TABLE 2. Fate of [2-l4Cjacetate (1 ,uCi) added tohot-spring algal-bacterial mat cores 48 h aftercollection and anaerobic incubation in the dark

% of recovered 14C in:0Condition

C02 CH4 Cells Filtrate

Octopus Spring500C 0.39 0.12 4.39 95.11550C 0.29 5.44 3.70 90.27600C 0.36 0.20 6.64 92.82

Wiegert Channel500C 0.07 0.13 2.67 97.16550C 1.24 38.94b 8.38 53.54600C 0.18 0.11 1.93 97.79a Samples assayed 71 h after label addition.b Standard deviation, 31.33.

mained in the filtrate (presumably as unmetab-olized acetate). Although increased amounts of4CH4 were produced in samples from 55°C inthe Wiegert Channel mat (the data do not indi-cate confidence in the mean value, however;Table 2), surprisingly little 14CH4 was producedat temperatures at which methanogenesis wasactive in both mats. Considering the lengthyincubation period and the possibility that '4CH4could have been produced indirectly via reduc-

APPL. ENVIRON. MICROBIOL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

METHANE PRECURSORS IN ALGAL-BACTERIAL MATS 779

tion of '4CO2 (which was also produced from [2-"C]acetate), the data do not indicate that dark-ness caused a shift in acetate metabolism fromincorporation to conversion to methane.The direct reduction of [3H]acetate to C3H4 in

an anaerobic dairy cow manure digester (datanot shown) indicated a method that could beused to study the reduction of the methyl groupdirectly. [3H]acetate was added to samples col-lected at 500C in Octopus Spring (the optimumtemperature for methane production in this matat the time of the experiment). C3H4 was notdetected upon either short (3-h) or long (4-day)incubation (Table 3). Nearly all of the 3H wasrecovered in the filtrate. Little 3H was detectedin the cell fraction, presumably because thehigher-specific-activity [3H]acetate had less ef-fect than [2-'4C]acetate on the acetate concen-tration in samples.Carbon dioxide as a methane precursor.

When NaH'4C03 was added to samples fromboth mats, 14CH4 production ensued (Fig. 2).The CH4 specific activity/CO2 specific activityratio indicated the fraction of methane derivedfrom C02 reduction. However, the specific activ-ity ratio was found to increase from about 0.5 to0.8 in the 6 to 8 h after sample collection andNaH04CO3 addition. This increase also occurredif NaH14CO3 was added substantially after sam-

TABLE 3. Fate of[3Hlacetate incubatedanaerobically in Octopus Spring algal-bacterial

mat cores collected at 50°C

Incubation % of recovered 'H in:

tine CH4 Cells Filtrate

3 h NDa 1.23 98.754 days ND 2.22 97.75

a Not detected.

WIEGERT CHANNEL 600(

OCTOPUS SPRING 55°C

ples were collected (46 h). Thus, the change didnot appear to reflect changes in the relativeimportance of C02 as a methane precursor butappeared to be an artifact of the dispersal ofradiolabel to sites where methane productionoccurred within the gelatinous mat sample. Thefollowing procedure was adopted in subsequentanalyses. NaH'4CO3 was incubated with samplesfor a period sufficient to avoid initial changes inthe specific activity ratio (24 h). The headspaceswere then purged with helium. The specific ac-tivity of methane produced during a 3-h reincu-bation period was then compared with the spe-cific activity of carbon dioxide evolved duringthe same period.When data from different experiments were

pooled, mean specific activity ratios of 0.80 +0.06 (twice standard error) for the OctopusSpring mat (55°C) and 0.71 + 0.21 (twice stan-dard error) for the Wiegert Channel mat (6000)indicated that most of the methane produced inthese mats was formed by reduction of C02.Similar results indicating the predominance ofC02 as a methane precursor were noted at alltemperatures in Octopus Spring (Table 4).Additional experiments were performed on

samples collected at 550C in Octopus Spring toevaluate the potential contributions of acetateand C02 in methanogenesis. When H2 was addedto enhance the reduction of C02 to methane, therelative importance ofC02 was slightly increased(Table 5) (P = 0.013). This suggested that otherreactions might be ongoing to account for a smallamount of the methane being produced. Theaddition of acetate (to 1 mM) did not reduce therelative importance of C02, as would be expectedif acetate were being converted to methane inthese samples.To test the error in measuring the specific

activity ratio or isotopic preference of 12C02 over14CO2, reduction of NaH4CO3 to "CH4 wastested in pure cultures of M. thermoautotroph-icum grown under conditions in which methaneshould be produced entirely by C02 reduction.Specific activity ratios of 0.84, 0.81, 0.83, and0.83 (mean 0.83 ± 0.01, twice standard error)

TABLE 4. Relative importance of CO2 as a methaneprecursor at various temperatures of the Octopus

Spring algal-bacterial matTemp (00) CH4 sp act/C02 8p act'

45 0.8150 0.8755 0.8860 0.90

a (CH4 specific activity . CO2 specific activity) x100 = % CH4 from CO2. Readings taken 3 h afterheadspace flushed with 100% helium (5 min).

cr0

0

cDi00

LL

I

(-

-lJ0

0 1 1

10 20HOURS AFTER SAMPLING

FIG. 2. Production of methane from CO2 in hot-spring algal-bacterial mat cores.

VOL. 41, 1981

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

780 SANDBECK AND WARD

TABLE 5. Influence of hydrogen or acetate on therelative importance of CO2 as a methane precursor

in Octopus Spring algal-bacterial mat samplescollected and incubated anaerobically at 55°C

Condition CH4 sp act/CO2 spCondition ~~~~act"

Control.0.73Hydrogen" 0.86Acetatec.0.86

' (CH4 specific activity . CO2 specific activity) x100 = % CH4 from CO2. Headspace flushed at 24 hwith 100% helium (5 min); analysis at 27 h.

b H2 (0.1 ml of 100% H2) added at the time ofsampling and again immediately after flushing withhelium.

'"Sterile anoxic sodium acetate (0.1 ml) to achievea final concentration of 1 mM (assuming a 2.5-mlsample) added at the time of sampling and immedi-ately after flushing with helium.

were found for isolates grown at 50, 55, 60, and65°C, respectively.

DISCUSSIONMethanogenic bacteria have been shown to

use H2, acetate, formate, methanol, carbon mon-oxide, and various methylamines as energysources (1). However, in natural environments,the predominant methane precursors arethought to be H2/CO2 and acetate (17, 36). SinceCO2 reduction and acetate conversion to meth-ane occur by different mechanisms, NaH'4C03,[2-'4C]acetate, and [3H]acetate were used to se-lectively study the contribution of either meth-ane production reaction during methane produc-tion in low-sulfate, hot-spring algal-bacterialmats.When [2-'4C]acetate was added to samples,

'C was rapidly and predominantly incorporatedinto cellular material. This phenomenon wasobserved at all temperatures in both the OctopusSpring and Wiegert Channel mats. The incor-poration of [2-'4C]acetate into cellular carbonwas decreased by dark incubation (and by ad-dition of label after 2 days of incubation in thedark [Table 2]), which suggests that photo-trophic microorganisms may be involved in ace-tate incorporation. Autoradiograms of [2-14C]-acetate-labeled algal-bacterial mat materialfrom Octopus Spring and Wiegert Channel wereprepared so that incorporated 14C could be re-lated to the type of cell involved in acetateincorporation. Representative results from theOctopus Spring mat (55°C) (Fig. 1) clearly in-dicated that very long filamentous bacteria in-corporated most of the acetate into cellular ma-terial. It is reasonable to guess that acetate wasincorporated by the filamentous phototrophicbacterium Chloroflexus aurantiacus because:

(i) this organism is common to both of the algal-bacterial mats sampled, (ii) light/dark experi-ments indicated the involvement of photo-trophic microorganisms, and (iii) the ability ofC. aurantiacus to grow photoheterotrophicallyusing acetate as a carbon source has been shownpreviously (16, 23, 24). Pierson (B. K. Pierson,Ph.D. thesis, Oregon State University, Eugene,1973) and Bauld and Brock (2) were led to thesame conclusion when they also noticed incor-poration of radiolabeled acetate into Chloro-flexus-like filaments in hot spring algal mats.A minor percentage of [2-'4C]acetate was con-

verted to 14CH4 in short-term labeling experi-ments (Table 1). This might have been relatedto competition for acetate, or inhibition by 02produced, by phototrophic microorganisms.When the problems associated with phototrophswere eliminated by preincubation in the dark,[2-'4C]acetate was not significantly converted to"4CH4 at several temperatures in both mats dur-ing a lengthy incubation in a period of activemethanogenesis. Although increased "4CH4 pro-duction occurred at 55°C in the Wiegert Channelmat, very little '4CH4 was produced at temper-atures at which methanogenesis is very active(e.g., 50 and 60'C) in the same mat. It is difficultto resolve whether "4CH4 was produced directlyfrom [2- 4C]acetate or indirectly by reduction of14CO2- '4CO2 produced from [2-'4C]acetate (Ta-bles 1 and 2) could be explained by anaerobicacetate oxidation by sulfate-reducing bacteria(33). Sulfate reduction has been previously re-ported in the Octopus Spring algal-bacterial mat(6), and rapid reduction of 35SO42- to H2"55 alsooccurred in the Octopus Spring and WiegertChannel mats (unpublished data of this labora-tory). The possibility that '4CH4 detected in [2-'4C]acetate labeling experiments may have comefrom reduction of 14C02 rather than by directreduction of the methyl carbon of acetate seemsespecially likely in the experiment in which [2-"4C]acetate was added 48 h after sampling andsamples were not assayed until 71 h after labeladdition (Table 2). Direct conversion of [;'H]-acetate to C3H4 could not be demonstrated insamples from Octopus Spring (50°C) (Table 3).The high relative importance of CO2 as a

methane precursor (80 + 6% for Octopus Spring[55°C] [Table 4] and 71 + 21% for WiegertChannel [60°C]) was consistent with the resultsindicating the lack of importance of acetate as amethane precursor. Additions of hydrogen andacetate have been shown to markedly affect therelative importance of methanogenic reactionsin Lake Mendota sediment (34) and in an anaer-obic dairy cow manure digester (unpublisheddata of this laboratory). The possible contribu-tion of methane precursors other than CO2 was

APPL. ENVIRON. MICROBIOL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

METHANE PRECURSORS IN ALGAL-BACTERIAL MATS 781

suggested by the slight increase in the relativeimportance of CO2 reduction to methane uponH2 addition (Table 5). However, the failure ofacetate to decrease the importance of CO2 re-duction to methane is additional evidence thatacetate is not an important methane precursorin Octopus Spring. This level of added acetate(1 mM) was near K, values for acetate whichhave been reported for isolated acetate-usingmethanogenic bacteria (35, 40).

Despite the high relative importance of C02in methanogenesis, specific activity ratios werealways less than 1.0 (which indicates that 100%of the methane formed is derived from CO2reduction). Other known methane precursorssuch as methylamines (11, 22, 31, 40) or methylsulfides (38) might contribute to the formationof methane. Zinder et al. (39) showed the for-mation of methyl sulfides in the degradation ofOctopus Spring algal-bacterial mats. Methane-producing bacteria are also known to carry outone of the greatest biological carbon isotopediscriminations known (17). In an experiment todetermine the specific activity ratio in pure cul-tures of M. thermoautotrophicum, the CH4 spe-cific activity/CO2 specific activity ratio was 0.83+ 0.01 (twice standard error), suggesting an ap-parent isotopic discrimination between 12C and14C of about 17%. A preference of 12C over 13C of2.5 to 3.4% has been reported during C02 reduc-tion to CH4 by M. thermoautotrophicum (9, 10).As the mass difference between 12C and 14C isgreater than the mass difference between 12Cand 13C, the discrimination between 12C and 14Cshould be greater than that between 12C and 13C(15). Thus, some of the specific activity dilutionbetween 4OCH4 and 14CO2 could be explained bya preference for the lighter nonradioactive 12C.The difference between the apparent and ex-pected specific activity ratios in pure culturescould also be due to error in the four determi-nations needed to calculate the specific activityratio. The apparent discrimination (and/orerror) measured in pure cultures is sufficient tocorrect specific activity ratios determined in fieldwork to near 1.0, suggesting that C02 reductionmay account for nearly all of the methane pro-duced in the anaerobic degradation of hot-springalgal-bacterial mats.High temperature is not a barrier to microbial

conversion of acetate to methane, as Zinder andMah (40) reported the isolation of a thermo-philic strain of Methanosarcina sp. capable ofgrowth on acetate at 550C. Blue-green autoflu-orescence of methanogenic bacteria has beensuggested as presumptive evidence for their rec-ognition in natural materials (18). Only rod-shaped, blue-green autofluorescing cells wereobserved in the decomposing mat samples. In

the Octopus Spring mat, the photoheterotrophicC. aurantiacus is active within the entire 1- to3-mm photic zone because of its ability to growat low light intensity beneath the cyanophyte S.lividus, which requires higher light intensity (2,4, 7). Maximum rates of methanogenesis werealso reported to be in the 2- to 4-mm depthintervals in Octopus Spring (29). The proximityof active photoheterotrophic microorganismsand anaerobic processes may lead to acetateconsumption by photoheterotrophic bacteriarather than by methanogenic bacteria. It hasalso been suggested that acetate may not be animportant energy source for sulfate-reducingbacteria in a high-sulfate, hot-spring algal-bac-terial mat (30). Here, despite partial oxidation of[2-'4C]acetate to 14CO2, acetate accumulatesupon dark incubation even during active sulfatereduction. A similar situation exists in the rumenand gastrointestinal fermentations, where meth-ane is produced mainly from C02 because thehost animal absorbs acetate for its energy me-tabolism (12, 17).

ACKNOWLEDGMENTSWe thank Mike Winfrey, Eric Beck, and the U.S. National

Park Service for assistance in obtaining results.The material is based upon work supported by the National

Science Foundation (project DEB-7824070) and the MontanaDepartment of Natural Resources and Conservation (project402-782).

LITERATURE CITED

1. Balch, W. E., G. E. Fox, L J. Magrum, C. R. Woese,and R. S. Wolfe. 1979. Methanogens: reevaluation ofa unique biological group. Microbiol. Rev. 43:260-296.

2. Bauld, J., and T. D. Brock. 1973. Ecological studies ofChloroflexis, a gliding photosynthetic bacterium. Arch.Mikrobiol. 92:267-284.

3. Brock, M. L., and T. D. Brock. 1968. The application ofmicroautoradiographic techniques to ecological studies.Int. Assoc. Theor. Appl. Limnol. 15:1-29.

4. Brock, T. D. 1978. Thermophilic microorganisms and lifeat high temperatures. Springer Verlag, New York.

5. Castenholz, R. W. 1969. Thermophilic blue-green algaeand the thermal environmnent. Bacteriol. Rev. 33:476-504.

6. Doemel, W. N., and T. D. Brock. 1976. Vertical distri-bution of sulfur species in benthic algal mats. Limnol.Oceanogr. 21:237-244.

7. Doemel, W. N., and T. D. Brock. 1977. Structure,growth, and decomposition of laminated algal-bacterialmats in alkaline hot springs. Appl. Environ. Microbiol.34:433-452.

8. Fraleigh, P. C., and R. G. Wiegert. 1975. A modelexplaining successional change in star.ding crop of ther-mal blue-green algae. Ecology 56:656-664.

9. Fuchs, G. D., R. Thauer, H. Ziegler, and W. Stichler.1979. Carbon isotope fractionation by Methanobacte-rium thermoautotrophicum. Arch. Microbiol. 120:135-139.

10. Games, L. M., J. M. Hayes, and R. P. Gunsalus. 1978.Methane-producing bacteria: natural fractionations ofthe stable carbon isotopes. Geochim. Cosmochim. Acta42:1295-1299.

VOL. 41, 1981

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.

782 SANDBECK AND WARD

11. Hippe, H., D. Caspari, K. Fiebig, and G. Gottschalk.1979. Utilization of trimethylamine and other N-methylcompounds for growth and methane formation by Meth-anosarcina barkeri. Proc. Natl. Acad. Sci. U.S.A. 76:494-498.

12. Hungate, R. E. 1967. Hydrogen as an intermediate in therumen fermentation. Arch. Mikrobiol. 59:158-164.

13. Hungate, R. E. 1969. A roll rube method for cultivationof strict anaerobes, p. 117-132 In J. R. Norris and D.W. Ribbons (ed.), Methods in microbiology, vol. 3B.Academic Press, Inc., New York.

14. Jeris, J. S., and P. L. McCarty. 1965. The biochemistryof methane fermentation using 14C tracers. J. WaterPollut. Control Fed. 37:178-192.

15. Jorgensen, B. B. 1978. A comparison of methods for thequantification of bacterial sulfate reduction in coastalmarine sediments. I. Measurement with radiotracertechniques. Geomicrobiol. J. 1:11-27.

16. Madigan, M. T., S. R. Petersen, and T. D. Brock. 1974.Nutritional studies on Chloroflexus, a filamentous pho-tosynthetic, gliding bacterium. Arch. Microbiol. 100:97-103.

17. Mah, R. A., D. M. Ward, L. Baresi, and T. L. Glass.1977. Biogenesis of methane. Annu. Rev. Microbiol. 31:309-341.

18. Mink, R. W., and P. R. Dugan. 1977. Tentative identi-fication of methanogenic bacteria by fluorescence mi-croscopy. Appl. Environ. Microbiol. 33:713-717.

19. Mountfort, D. O., and R. A. Asher. 1978. Changes inproportions of acetate and carbon dioxide used as meth-ane precursors during the anaerobic digestion of bovinewaste. Appl. Environ. Microbiol. 35:648-654.

20. Nelson, D. R., and J. G. Zeikus. 1974. Rapid method forthe radioisotopic analysis of gaseous end products ofanerobic metabolism. Appl. Microbiol. 28:258-261.

21. Oppermann, R. A., W. 0. Nelson, and R. E. Brown.1961. In vivo studies of methanogenesis in the bovinerumen: dissimilation of acetate. J. Gen. Microbiol. 25:103-111.

22. Patterson, J. A., and R. B. Hespell. 1979. Trimethyl-amine and methylamine as growth substrates for rumenbacteria and Methanosarcina barkeri. Curr. Microbiol.3:79-83.

23. Pierson, B. K., and R. W. Castenholz. 1974. A photo-trophic gliding filamentous bacterium of hot springs,Chloroflexus aurantiacus, gen. and sp. nov. Arch. Mi-crobiol. 100:5-24.

24. Sirevag, R., and R. Castenholz. 1979. Aspects of carbonmetabolism in Chloroflexus. Arch. Microbiol. 120:151-153.

25. Smith, M. R., and R. A. Mah. 1978. Growth and meth-anogenesis by Methoanosarcina strain 227 on acetate

and methanol. Appl. Environ. Microbiol. 36:870-879.26. Smith, M. R., and R. A. Mah. 1980. Acetate as a sole

carbon and energy source for growth of Methanosar-cina strain 227. Appl. Environ. Microbiol. 39:993-999.

27. Stainton, M. P. 1973. A syringe gas stripping procedurefor gas chromatographic determination of inorganic andorganic carbon in freshwater and carbonates in sedi-ments. J. Fish. Res. Board Can. 30:1411-1445.

28. Walter, M. R. 1972. A hot spring analog for the deposi-tional environment of Precambrian iron formations ofthe Lake Superior region. Econ. Geol. 67:965-972.

29. Ward, D. M. 1978. Thermophilic methanogenesis in ahot-spring algal-bacterial mat (71 to 30°C). Appl. En-viron. Microbiol. 35:1019-1026.

30. Ward, D. M., and G. J. Olson. 1980. Terminal processesin the anaerobic degradation of an algal-bacterial matin a high-sulfate hot spring. Appl. Environ. Microbiol.40:67-74.

31. Weimer, P. J., and J. G. Zeikus. 1978. One carbonmetabolism in methanogenic bacteria. Cellular charac-terization and growth of Methanosarcina barkeri.Arch. Microbiol. 119:49-57.

32. Weimer, P. J., and J. G. Zeikus. 1978. Acetate metab-olism in Methanosarcina barkeri. Arch. Microbiol.119:175-182.

33. Widdel, F., and N. Pfennig. 1977. A new anaerobic,sporing bacterium, Desulfotomaculum (emend.) acetox-idans. Arch. Microbiol. 112:119-122.

34. Winfrey, M. R., and J. G. Zeikus. 1979. Anaerobicmetabolism of immediate methane precursors in LakeMendota. Appl. Environ. Microbiol. 37:244-253.

35. Zehnder, A. J. B., B. A. Huser, T. D. Brock, and K.Wuhrmann. 1980. Characterization of an acetate-de-carboxylating, non-hydrogen-oxidizing methane bacte-rium. Arch. Microbiol. 124:1-11.

36. Zeikus, J. G. 1977. The biology of methanogenic bacteria.Bacteriol. Rev. 41:514-541.

37. Zeikus, J. G. 1979. Thermophilic bacteria: ecology, phys-iology and technology. Enzyme Microbiol. Technol. 1:243-252.

38. Zinder, S. H., and T. D. Brock. 1978. Production ofmethane and carbon dioxide from methane thiol anddimethyl sulphide by anaerobic lake sediments. Nature(London) 273:226-228.

39. Zinder, S. H., W. N. Doemel, and T. D. Brock. 1971.Production of volatile sulfur compounds during thedecomposition of algal mats. Appl. Environ. Microbiol.34:859-860.

40. Zinder, S. H., and R. A. Mah. 1979. Isolation and char-acterization of a thermophilic strain ofMethanosarcinaunable to use H2-CO2 for methanogenesis. Appl. Envi-ron. Microbiol. 38:996-1008.

APPL. ENVIRON. MICROBIOL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 19

Nov

embe

r 20

21 b

y 16

5.22

5.19

8.19

2.