Quantification of methanogenic biomass by enzyme-linked immunosorbent assay and by analysis of specific methanogenic cofactors

Biomass 14 (1987) 195-208

Quantification of Methanogenic Biomass by Enzymelinked Immunosorbent Assay and by Analysis of Specific Methanogenic Cofactors L. G. M. G o r r i s Department of Microbiology,Universityof Nijmegen,ToernooiveldNL-6525 ED Nijmegen,The Netherlands H. A. K e m p a n d D. B. A r c h e r A FRC Institute of Food Research,Norwich Laboratory, Colney Lane, Norwich NR4 7UA, UK (Received 21 October 1987; accepted 29 October 1987)

ABSTRACT The reliability and accuracy with which enzyme-linked immunosorbent assay (EL1SA) and an assay of methanogenic cofactors detect and quantify methanogenic species were investigated. Both assays required standardization with laboratory cultures of methanogenic bacteria and were applied to mixtures of pure cultures and samples from anaerobic digesters. ELISA was shown to be a simple method for detecting and quantifying individual methanogenic species. The range of species which can be assayed is limited by the range of antisera available but, potentially, ELISA can be applied to all methanogens. Although the cofactor assay is not species-specific it can distinguish hydrogenotrophic and acetotrophic methanogens and is quantitative. Key words: Quantification of methanogenic biomass, ELISA, methanogenic cofactor assay, anaerobic digestion, process control.

INTRODUCTION Purification of waste waters by anaerobic degradation of the soluble organic fraction to biogas can be efficiently accomplished by complex 195 Biomass 0144-4565/87/S03.50 -- © 1987 Elsevier Applied Science Publishers Ltd, England. Printed in Great Britain

196

L. G. M. Gorris, H. A. Kemp, D. B. Archer

microbial consortia present in anaerobic digesters. The methanogenic population in such bioreactors consists of both hydrogenotrophic and acetotrophic species. Improved process understanding which can be exploited in control can be obtained by monitoring the methanogenic bacteria in anaerobic digesters. Various methods are available for the quantification of methanogenic biomass and activity.~-5 In this study quantitative analyses of methanogenic biomass in complex and defined mixed cultures were performed using two different methods in order to evaluate their reliability. The first method was a microtitration plate enzyme-linked immunosorbent assay (ELISA) which was developed for the detection and quantification of individual methanogenic bacteria in pure and defined mixed cultures.~ The specificity of the original assay, a single-site ELISA using polyclonal antisera, was later improved by use of monoclonal antibodies and the development of a two-site ELISA in which a combination of polyclonal and monoclonal antisera was employed.6 The high sensitivity and specificity of the refined ELISA render it a useful tool in probing the methanogenic population in complex mixed cultures such as anaerobic sludge. Antisera were available in this study for the quantitative assay of Methanobacterium bryantii strain FR2 and Methanosarcina mazei strain $6, a hydrogenotrophic and an acetotrophic methanogenic species, respectively. The second method is based on analysis of specific methanogenic cofactors, namely the Cl-carrier methanopterin, the redox carrier coenzyme F420 and the CH3-carrier 5-hydroxybenzimidazolylcobamide (vitamin B12-HBI).7 Cofactor assays comparable to the assay used in this study, all employing high-performance liquid chromatography (HPLC), have previously been used to detect and quantify these cofactors in pure cultures of a variety of methanogenic species.5.s.9 A distinct structural difference was found in some of these cofactors between hydrogenotrophic and acetotrophic methanogens. In general, hydrogenotrophs are characterized by the presence of methanopterin and coenzyme F4~_0 with two glutamate residues in the side chain (coenzyme F42,-2) while acetotrophic species contain sarcinapterin, a methanopterin analogue with an additional glutamic acid residue, "~ and coenzyme F42o with four and five glutamate residues (coenzymes F420-4 and F420-5). In both, a coenzyme F420 derivative tentatively identified to be coenzyme F420 with three glutamate residues (coenzymes F420-3) was detected as wellY Based on these differences in cofactor composition, the assay might be used to separately detect and quantify different trophic groups of methanogens in anaerobic digesters.

Quantification of methanogenic biomass

197

Here we describe the results obtained with both methods in quantifying Mb. bryantii FR2 and Ms. mazei $6 present in defined mixtures of pure cultures and in methanogenic sludges to which a known amount of these bacteria was added. In addition, methanogenic sludges of undefined composition were analysed.

MATERIALS AND METHODS

Microorganisms Methanosarcina mazei $6 (DSM 2053) and Methanobacterium bryantii FR2 (DSM 2257) were used as standard preparations in the ELISAs and for the production of polyclonal and monoclonal antibodies. Ms. mazei was grown with methanol (62 mM) as the substrate in a medium described elsewhere. ~ Mb. bryantii was grown on H2/CO 2 (80:20 V/V,at ca 200 kPA of pressure) in a medium containing (per litre): KHzPO4, 0.45 g; K2HPO 4, 0.45 g; NH~CI, 0-45 g; NaC1, 1.35 g; NaHCO3, 2.5 g; MgSOa.7H20, 0.18 g; Na2S.9H20, 0.5 g; sodium acetate, 0-5 g; Lcysteine.HCl, 0.5 g; CaCIz.2H20, 0.12 g; yeast extract, 2.0 g; tryptone soya broth, 0-5 g; sodium resazurin, 1 mg; trace minerals and vitamins solution,~-~ 10 ml each; isobutyric, a-methylbutyric, valeric and isovaleric acid, all at a final concentration of 0.05% (v/v). Escherichia coli strain B was grown on nutrient broth (Difco). Four different types of methanogenic sludge were obtained from laboratory scale fluidized bed reactors about four months after the reactors were started up with bare sand on which bacteria immobilized during maturation. The reactors, operated at 37 °C with a hydraulic retention time of 1.4 h, were treating synthetic waste waters (organic load 2-3 kg VFA-COD m-3 day-~, pH 7.0) containing either acetate, propionate and butyrate (3:1 : 1 w/v) or each of these volatile fatty acids alone as carbon sources in addition to essential salts, minerals and vitamins. A preliminary identification of the methanogens present in the sludge samples was obtained by microscopic observations, using a Leitz epifluorescence microscope. ~3 In all sludges methanogenic bacteria morphologically resembling Methanobacterium spp. and Methanothrix spp. were found to be present as the predominant hydrogenotrophic and acetotrophic methanogens, respectively. Methanothrix appeared to be the most abundant methanogen in all cases. Low amounts of Methanosarcina spp. were observed in all sludges, except in the sludge grown on acetate alone.

198

L. G. M. Gorris, H. A. Kemp, D. B. Archer

Preparation of defined mixtures Samples of pure cultures and sludges were washed in phosphate buffer (PB: 10 mM K2HPOa/KH2PO 4 pH 8"0 containing 0.02% NAN3) and resuspended in this buffer to obtain suspensions with a wet weight content of about 100 mg ml-~. These stock suspensions were then sonicated (MSE Soniprep 150) sufficiently to suspend clumps of cells without causing microscopically detectable physical damage to the cells. This treatment is especially important for the ELISA, for which homogeneous samples without any particulate matter are required. The stock suspensions of the pure cultures of methanogens and E. coli were used to prepare three mixtures containing defined volumes of these three suspensions while a known volume of Mb. bryantii and Ms. mazei stock suspension was added to acetate and propionate grown digester sludge, respectively (Table 1 ). The total concentration of cell protein was determined by biuret ~4 and Folin-Ciocalteu ~5 assay with bovine serum albumin as a standard.

Enzyme-linked immunosorbent assay The preparation of methanogens for immunization and the production of polyclonal and monoclonal antisera have been described previously.6'i6 In this study monoclonal antibodies raised against Ms. mazei TABLE 1 Composition of Defined Mixtures Prepared with Pure Cultures and Methanogenic Sludges"

Stock suspension (code)

Volume ratio of stock suspension per ml mixture (mixture code) FM

Mb. bryantiiFR2 (FS) Ms. mazeiS6(SS) E. coliB (ES) Acetate grown (AS) Propionate grown (PS)

SM

0-6 (36) ~' 0'2 (11) 0'2(17) 0'6(46) 0'2 (47) 0-2 (43) --. . .

EM

AM

PM

0'2 (7) 0"2(10) 0'6 (83) -.

0'34 (69) --(}.46 (31 )

-0-38(26) --0"50 ( 74)

"Butyrate grown sludge (BS) and sludge grown on VFA-mixture (MS) were not used to prepare defined mixtures. bin brackets: % of total protein calculated from the protein content ~5 of the stock suspensions and the volume ratio in the final mixture.

Quantification of rnethanogenic biomass

199

$6 were used in a competitive assay, and two different polyclonal antisera raised against Mb. bryantii FR2 were used in a two-site assay. A detailed description of both assays is given elsewhere.6 Stock suspensions and mixtures were diluted in phosphate buffer (PB) to give a protein concentration of ca 100/zg ml-~. For each sample a series of dilutions ranging from 10-2 t o 10- 6 was then subjected to both ELISAs. Preparations of Ms. mazei and Mb. bryantii containing known amounts of protein ~5were used as standards for quantification.

Assay of specific methanogenic cofactors Methanogenic cofactors were extracted from samples of the various suspensions as described beforeY Aliquots were subjected to analyses with two different binary reversed-phase HPLC systems. The first system (system I) was used to detect specifically derivatives of coenzyme I=420. The HPLC consisted of Waters M6000 and M45 pumps, a 660 programmer and a U6K injector and was equipped with an analytical column (0"46 x 25 cm) packed with 10/~m C~8 LiChrosorb RP-18 (Merck). The detector was an Amicno-Bowman spectrophotofluorimeter with an 8 pl HPLC flow cuvette and with the excitation and emission wavelength at 405 nm and 470 nm, respectively. The flow of the mobile phase, solvent A 27.5 m~x C H 3 C O O H - K O H pH 6"0 and solvent B 20% acetonitrile in 27.5 mM C H 3 C O O H - K O H pH 6"0, was kept constant at 2 ml min- J. A linear gradient from 0% to 100% B in 20 min was started 2 min after injection. The detector signal was integrated with a Hewlett Packard 3390A integrator. A total cofactor spectrum was obtained in the second system (system II) by using a Hewlett Packard 1084B HPLC, equipped with an analytical column (0.46 × 10 cm) packed with 5 p m C~8 LiChrosorb RP-18 and a variable wavelength detector set to 350 nm. Integration of the detector signal was by the 79850B LC terminal. The flow rate of the mobile phase, the same as in System I but with solvents adjusted to pH 4.7, was 1 ml min-~ constantly. A stepwise linear gradient was used after injection: 2 min at 10% B, 10% to 20% B in 4 min, 20% to 60% B in 14 min, 60% to 95% B in 5 min, 15 min at 95% B, 95% to 10% B in 5 min. The cofactor concentrations in the extracts were quantified by using FO (7,8-didemethyl-8-hydroxy-5-deazariboflavin), a synthetic coenzyme F420 analogue, as the internal standard. These cofactor concentrations and the protein contents of the samples ~5 were used to calculate the cofactor content in the original suspension.

L. G. M. Gorris, H. A. Kemp, D. B. Archer

200

RESULTS

Quantification by ELISA The dependence of optical density upon protein concentration in standards and samples is shown in Figs 1 and 2 for the competitive ELISA for Ms. mazei and the two-site ELISA for Mb. bryantii, respectively. Results for protein contents 15 of the samples are given in Table 2. These results were in good agreement with the estimates made by the biuret method. The protein levels of samples constructed by mixing other cell suspensions in known proportions (Table 1) were between 99% and 107% of the theoretical levels, with the exception of sample PM (82%). Detection and quantification of Ms. mazei and Mb. bryantii by ELISA are also given in Table 2. These results are compared with the levels expected from the known protein concentrations and sample compositions (Table 1). The ELISAs were carried out on diluted samples and, in some cases, the amounts of Ms. mazei and Mb. bryantii were below the detection limits of the assays. Although the limits for detection in assays of standards were 3 ng Ms. mazei protein ml-~ and 50 ng Mb. bryantii protein ml-J for routine work limits of 50 ng Ms. mazei protein ml-J and 500 ng Mb. bryantii protein ml-J were adopted. Among the digester samples only MS contained cells with antigenic sites recognized in ELISA using antibody to Ms. mazei. The ELISAs are known to be

1.2

1.0

~" 0.6

d 0.4

0'05

0"5 ~g ml

Fig. 1.

2

5

50

protein

Ms. mazei $6 ELISA standard curve (e), sludge sample PS (D) and defined

sludge mixture PM (o).

Quantification of methanogenic biomass

201

1.6

/

1.4 1.2

E

1.0

o '~ 0-8

//0

d 0.6 0.4

i

0-2 ~-

Sample dilution i

I

105 I

0"025

Fig. 2.

I

10-4 I

i

10-3 I

J,,

0-125 0-25 1.25 2"5 JucJm l 1 protein

I

i

12'5 25

Mb. bryantii F R 2 ELISA standard curve (e) and mixed culture samples E M (tz) and F M (o).

highly specific for Ms. rnazei and Mb. bryantii. L6J6 The sludges probably contained Ms. barkeri and other Methanobacterium spp., as judged by microscopic examination, but specific antibodies are required for their detection and quantification. Quantification by cofactor assay

Representative elution patterns obtained with HPLC analyses I and II of cofactor extract of propionate grown sludge (PS) are shown in Figs 3 and 4, respectively. The amounts of Ms. mazei and Mb. bryantii protein present in the defined mixtures (Table 3) were calculated by comparing the concentrations of selected cofactors measured in these mixtures (data not shown) to cofactor contents measured in the stock suspensions of the methanogens (see legend to Table 3). The detection limit of analysis I, based on coenzyme [=420content, for Ms. mazei and Mb. bryantii was 12 /~g protein and 1-2 #g protein per injected sample, respectively. For analysis II, based on pterin contents, detection limits per injected sample were 0.4 #g Ms. mazei protein and 0.9 ~g Mb. bryantii protein.

L. G. M. Gorris, H. A. Kemp, D. B. Archer

202

TABLE 2 Detection and Quantification of Ms. mazei and Mb. bryantii by ELISA

Suspension code

Totalprotein by Lowry (mg ml- 9

FS FM SS SM ES EM AS AM PS PM BS MS

Ms. mazei

Mb. bryantii

protein by EL1SA (mg ml- i)

protein by EL1SA (mg ml- 9

4.19 7.27 5.85 7.52 16"10 12.50 2"60 2'89 1 2 " 30 6.80 11'30 11.52

" 1.7 (145)' 4.88 (83) 3.53 (101)

321 (77) ~' 3-72(148) ,/ 0.67 (80)

"

,I

1'63 (139) " "

1.44 (171) ,1 1"85 (96)

"

,i

1.50 (69) " 0'25

,/ ,t '~

"Less than 0'05 mg protein ml- ~in the undiluted suspension. ~'Percentage of Mb. bryantii protein detected by ELISA compared to the level of Mb. bryantii protein expected in the sample from the Lowry ~5assay. ' Percentage of Ms. mazei protein detected by ELISA compared to the level of Ms. mazei protein expected in the sample from the Lowry ~s assay. %ess than 0.50 mg protein ml- ~in the undiluted suspension.

~05-~70 nm F~O2

FO

I

L

I

d

I

I

0

5

10

15

20

I

25 time [mm )

Fig. 3. Elution pattern of cofactor containing extract of sludge sample PS obtained with HPLC analysis I. F42o-3, coenzyme F420-3; F420-2, coenzyme F420-2; FO, internal standard FO.

203

Quantification of methanogenic biomass 350 nm

mpt spt

F&~-~ IIII

j

B

I

i

iIo

I

FO hbi

I

I].dmbi

I

I

20

I 30

I

time (min)

Fig. 4. Elution pattern in HPLC analysis II of cofactor extract of sludge sample PS. F342, 7-methylpterin; F420-2, coenzyme F420-2; rapt, methanopterin; spt, sarcinapterin; FO, internal standard FO; hbi, vitamin B~2-HBI; dmbi, vitamin B)2-DMBI.

TABLE 3

Quantification of Ms. mazei and Mb. bryantii in Defined Mixtures by Cofactor Assay with HPLC Analyses I and II Mixture code

Protein content (mg ml- i)

Species

Expected content"

FM SM EM AM PM

Ms. mazei Mb. bryantii Ms. mazei Mb. bryantii Ms. mazei Mb. bryantii Mb. bryantii Ms. mazei

1.17 2'51 3.51 0'84 1.17 0.84 1'93 2.19

Calculated content ~ System I

System H

1.18(101)' 2"52(100) 4.53 (129) 0"94(111) 1.44 (123) 1'02 (122) 1"77 (92) 2.23 (102)

1.52 (130) 2"21 (88) 4.05 (115) 0'76 (91) 1.42 (121) 0"74 (88) 1"77 (92) 2.75 (126)

"Derived from the protein content ~5of the stock suspensions and the volume ratio in the defined mixture. bProtein contents calculated from the concentrations of selected cofactors in the mixtures using cofactor contents measured in stock suspensions as references (in nmol mgprotein): coenzyme F420-3 in Ms. mazei (SS), 0.087; coenzyme F420-2 in Mb. bryantii(FS), 0"85; sarcinapterin in Ms. mazei, 22.7; methanopterin in Mb. bryantii, 15"12. 'Percentage of calculated protein compared to the level of protein expected in the sample from Lowry ~5assay.

204

L. G. M. Gorris, H. A. Kemp, D. B. Archer

The amounts of Methanothrix, Methanosarcina and Methanobacterium spp. observed microscopically in the digester sludges were estimated using pure culture cofactor contents of Mtx. soehngenii, Ms. barkeri MS and Mb. formicicum 8 as references (Table 4). Quantification of the latter methanogen was based on the concentration of methanopteri n in the sludge samples. Since Methanosarcina and Methanothrix both characteristically contain sarcinapterin (spt), vitamin B~_-HBI (hbi) and coenzymes F420-5 and F420-4, none of these compounds can be used to quantify these species individually when they are both present in the same sludge. However, there is a difference in the ratios of the spt and hbi content between these genera. In cultures of Ms. barkeri MS grown on either acetate, methanol or H2/CO 2, and in cultures of Ms. barkeri strain F U S A R O grown on acetate, the ratios spt/hbi are 25.4, 26"5, 22"7 and 21"7, respectively (calculated from data of Van Beelen et al. 5). The ratio for Ms. mazei $6 found in this study, 17.1, is comparable to these values. In contrast, the ratio for Mtx. soehngenii grown on acetate was found to be 272-0. 8 Thus, this ratio may be used to differentiate between Methanosarcina spp. and Methanothrix spp. It is also possible to estimate the proportions of both methanogens separately from the ratio measured in sludges which contain this mixed acetotrophic population. We calculated all spt/hbi ratios that would be found for any combination of Mtx. soehngenii and Ms. barkeri MS in an acetate utilizing mixed population

TABLE 4

Quantification of Methanobacterium, Methanosarcina and Methanothrix spp. in Digester Sludge Samples by Cofactor Assay with HPLC Analysis II Suspension code

AS PS BS MS

Estimated proportion (% of total protein)

Methanobacterium"

Methanothrix ~'

Methanosarcina ~'

1.4 (0.04)' 14-2 (1.75) 13-6 (1.54) 2.5 (0.29)

35.1 (0'88) 31.0 (3.81) 40-0 (4.52) 68.9 (7.93)

np" 1.4 (0.17) 3.3 (0.37) 2.3 (0.26)

"Calculated from the methanopterin (mpt) concentration measured in the sample; reference Mb. formicicum, 121-2 nmol mpt mg ~protein. ~ hDerived from the spt/hbi ratio measured in the sample by comparison to spt/hbi ratios computed for every possible combination of Methanothrix soehngenii and Methanosarcina barkeri (see legend to Fig. 5). 'Amount of methanogen protein (mg ml-t), calculated from the total protein content ~5 and the estimated proportion. 'tNot present as judged by microscopic examination.

Quantification of methanogenic biomass

205

consisting only of these two bacteria and compared them to the spt/hbi ratios measured in the sludges (Fig. 5) to estimate the relative amounts of both species. The sarcinapterin concentrations measured in the sludges were then used to quantify the absolute amounts of both Mtx. soehngenii and Ms. barkeri (Table 4).

DISCUSSION In this study we have investigated the ability of ELISA and assay of methanogenic cofactors to identify and quantify methanogenic bacteria in mixtures of pure cultures and samples from anaerobic digesters. Both assays quantified the methanogenic biomass although there was some variability in the results. ELISA is a species-specific assay, whereas the cofactor analysis is able to assay hydrogenotrophic and acetotrophic methanogenic biomass separately. The presence of Ms. mazei and Mb. bryantii was accurately detected by ELISA in all those samples known to contain the species. It was shown previously that the specificity of the assay is determined by the antibodies used. ~~'~ ELISA has therefore been shown to be a simple method for probing samples of unknown composition for the presence of a particular methanogenic species. The range of species which can be detected is limited only by the range of antisera available. Although the ELISAs described were designed to be highly specific for the target organisms, ELISAs can in principle be designed with specificity to

spt/hbi ratio 45 M 40 30

~

PS f

25

00

i

2'0

i

4~0

L

i

6~0 8~0 Mtx soehngenu~Msbarken MSratio (protein w/w)

Fig. 5. Mtx. soehngenii/Ms, barkeri MS ratio (protein w/w) in a mixed acetotrophic population versus the spt/hbi ratio calculated using spt and hbi contents (nmol cofactor mg ~protein) reported for Mtx. soehngenii(spt, 2.72 and hbi, 0'01) and Methanosarcina barkeri MS (spt, 186"9 and hbi, 7.36) grown on acetate, s Arrows point to the spt/hbi ratios measured in the indicated sludge samples which were used to derive the ratio of Methanothrix and Methanosarcina protein in the acetotrophic population.

206

L. G. M. Gorris, H. A. Kemp, D. B. Archer

strains, species or genera. As more information becomes available on the antigenic mosaic of a wide range of methanogenic species ~7 the use of monoclonal antibodies facilitates the design of ELISAs of differing specificities. All ELISAs require homogeneous samples. In this study samples were obtained by sonication which proved effective at removing sludge from sand support material. ELISA was also used to quantify the amounts of Ms. rnazei and Mb. bryantii in the samples. The values recorded varied from 69% to 171% of those expected from the protein levels and known compositions of the mixtures (111 + 36%; mean_+ standard deviation). The accuracy with which a species is quantified in a natural sample will also be affected by differences in its antigenicity brought about by any effects of growth conditions on the cell surface antigens. Although this aspect was not addressed in the present study it was noticed that the Mb. bryantii standard differed in its antigenicity from a standard grown under different conditions and used previouslyJ' With the cofactor assay a distinction could be made between hydrogenotrophic and acetotrophic methanogenic biomass based on the presence of specific derivatives of methanogenic cofactors. Both analyses employed gave optimal separation of these cofactors. A direct identification of methanogenic bacteria was not possible. By means of microscopic examination and, in the case of acetotrophic species, by comparison of spt/hbi ratios an indirect identification on genus level was obtained. The proportions of Mb. bryantii and Ms. rnazei in the mixtures of known composition were calculated by using cofactor contents measured in the appropriate stock suspension as reference. Cofactor contents measured in the Mb. bryantii FR2 stock suspension (FS) were close to values reported previously for Mb. bryantii Moll. s Both bacteria were cultured in the same medium. The contents of sarcinapterin and coenzyme F420 derivatives in the Ms. mazei $6 stock suspension (SS), however, were 3.4 and 10 times lower, respectively, as values reported before for the same strain grown on methanol in a different medium,s It remains to be investigated whether this indicates that the cofactor content in pure culture is dependent on the composition of the culture medium. Quantification of the amounts of Ms. mazei and Mb. bryantii in the defined mixtures with HPLC analyses I and II yielded values ranging from 92% to 129% (110_+13%) and 88% to 130% (106_+18%) of the expected values, respectively. The proportions of Methanobacterium, Methanosarcina and Methanothrix species in the digester sludges were estimated by the use of pure culture cofactor contents of representative species which were reported

Quantification of methanogenic biomass

207

previously. ~ The calculated proportions were in accordance with the relative abundance of these bacteria as judged by microscopic examination, but the amounts actually present could not be independently verified. The methanogenic cofactors examined in this study function at defined metabolic sites in methanogenesis. ~ Quantification of cofactors in digester sludge may therefore be exploited, for instance, to obtain information on the prevailing metabolic activities of the methanogens or to determine the site of interaction of toxic compounds in methane formation.

ACKNOWLEDGEMENTS This investigation was supported in part by the Foundation for Fundamental Biological Research (BION), which is subsidized by the Netherlands Organization for the Advancement of Pure Research (ZWO). We thank Sara Bramham and Hilary Mellon for excellent technical assistance.

REFERENCES 1. Archer, D. B. (1984). Appl. Environ. Microbiol., 48, 797-801. 2. Delafontaine, M. J., Naveau, H. P. & Nyns, E. J. (1979). Biotechnol. Lett., 1, 71-4. 3. Dolfing, J. & Bloemen, W. G. B. M. (1985). J. Microbiol. Methods, 4, 1-12. 4. Martz, R. E, Sebacher, D. I. & White, D. C. (1983). J. Microbiol. Methods, 1, 53-61. 5. Van Beelen, E, Dijkstra, A. C. & Vogels, G. D. (1983). Eur. J. Appl. Microbiol. Biotechnol., 18, 67-9. 6. Kemp, H. A., Morgan, M. R. A. & Archer, D. B. (1986). Enzyme-linked immunosorbent assay for methanogens using polyclonal and monoclonal antibodies. Proc. Water Treatment Conference, A QUA TECH '86, Amsterdam, The Netherlands, pp. 39-49. 7. Vogels, G. D., Van der Drift, C., Stumm, C. K., Keltjens, J. T. M. & Zwart, K. B. (1984). Antonie v. Leeuwenhoek, 50, 557-67. 8. Gorris, L. G. M. & Van der Drift, C. (1986). Methanogenic cofactors in pure cultures of methanogens in relation to substrate utilization. In Biology of Anaerobic Bacteria, H. C. Dubourguier et al. (eds), Elsevier Science Publishers, Amsterdam, pp. 144-50. 9. Van Beelen, E, Geerts, W. J., Pol, A. & Vogels, G. D. (1983). Analyt. Biochem., 131,285-90. 10. Van Beelen, P., Labro, J. E A., Keltjens, J. T., Geerts, W. J., Vogels, G. D., Laarhoven, W. H., Guijt, W. & Haasnoot, C. A. G. (1984). Eur. J. Biochem., 139, 359-65.

208

L. G. M. Gorris, H. A. Kemp, D. B. Archer

11. Kirsop, B. H., Hilton, M. G., Powell, G. E. & Archer, D. B. (1984). Methanogenesis in the anaerobic treatment of food-processing wastes. In Microbiological Methods for Environmental Biotechnology, J. M. Grainger & J. M. Lynch (eds), Academic Press, London, pp. 139-58. 12. Wolin, E. A., Wolin, M. J. & Wolfe, R. S. (1963). J. Biol. Chem., 238, 2882-6. 13. Doddema, H. J. & Vogels, G. D. (1978). Appl. Environ. Microbiol., 36, 752-4. 14. Herbert, D., Phipps, E J. & Strange, R. E. (1971). Chemical analysis of microbial cells. In Methods in Microbiology, Vol. 5B, J. R. Norris & D. W. Ribbons (eds), Academic Press, London, pp. 209-344. 15. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951). J. Biol. Chem., 193,265-75. 16. Kemp, H. A., Archer, D. B. & Morgan, M. R. A. (1988). The specific analysis of methanogenic bacteria used in the fermentation of food waste. In Advances in Immunoassays for Veterinary and Food Analysis -- I, B. A. Morris, M. N. Clifford & R. Jackman (eds), Elsevier Applied Science Publishers, pp. 143-7. 17. Macario, A. J. L. & Conway de Macario, E. (1985). Trends Biotechnol., 3, 204-8. 18. Whitman W. B. (1985) Methanogenic bacteria. In The Bacteria, Vol. VIII, C. R. Woese & R. S. Wolfe (eds) Academic Press, New York. pp. 3-84.