Physiology and development of Pectinatus cerevisiiphilus and Pectinatus frisingensis, two strict anaerobic beer spoilage bacteria

International Journal of Food Microbiology

ELSEVIER

International Journal

of Food

Microbiolog)

35

( 1997) 2% 39

Physiology and development of Pectinatus cerevisiiphilus and Pectinatus fiisingensis, two strict anaerobic beer spoilage bacteria

Received

3 July

1996: revised

I I October 1996; accepted I7 October 1996

Abstract

The genus Pwtitwtus was isolated recently and the deposited strains were classified as beer spoilage bacteria producing propionate as a major fermentation product. A recent investigation of this genus demonstrated the existence of two species: Pcc,tinrrtLl.sc.e~c,~isiip/li/ll.s. the type strain and Pec,tir7crtus ,f~i.singmsi.s. a new species with a different pattern of growth substrates. Different culture media tested for both species demonstrated a higher specific growth rate for P. ~c,rc,~~i.sii/)/7ilLI.C. However. final biomass production was in every case round 20% higher in P. fri.singol.si,s. A 400% decrease of final biomass production was measured when the species were cultivated on poor culture medium; this decrease was found to be broadly proportional to amounts of acetate excreted in the medium. Both species produced CO, from glucose; however. no significant modifications of biomass and volatile fatty acid production were demonstrated when varying head space composition with regard to CO, levels. Growth experiments on glucose with increasing amounts of ethanol added in the culture. revealed a higher sensitivity of P. crrc,r~i.sii/)/li/Lr.v to ethanol inhibition. Ethanol concentrations over 1.7 M resulted in a complete inhibition of growth for both Pec,tirwtu.s species. Combined effects of culture medium pH. lactate. and glucose concentrations. demonstrated the prevailing role of glucose in the development of the bacteria. However. these three parameters had a different influence on growth characteristics of both Pcctiwtus species. P. c,Crc,c.i.siip/zillr.s grew very poorly with a glucose concentration of 5 mM for pH values below 4. I, This species had optimal pH for growth between 6 and 6.2 and it excreted increasing amounts of acetate nith increase of pH, glucose or lactate in the culture medium. P. frisingo7,si.c showed a wide range of pH allowing a good growth. For glucose concentrations below 20 mM. highest final biomass productions were measured in the culture for pH values round 4.9: this also corresponded to a minimum in acetate excretion. The above results pointed at P. fklsi~qymsi.s as the prevailing species of Prc/i~um.s in beer spoilage. cj 1997 Elsevier Science B.V. K~~~~rror-tl.s: Pcc~ti~zrrru.s: * Corresponding

Beer: Spoilage;

Volatile

fatty acid; Propionic

author.

0168.16O.i.97~$17.00 i 1997 Elsevier Science B.V. All rights reserved PI1 SOlh8-lhOj~9h)OI206-8

acid

"0

20

40

60 80 TIME (h)

100

120

140

Fig. I. Culture medium influence on P. c~err~uiiphih and P. f~%irr~emi.c growth. dctermincd by absorbance measurements at 550 nm. Growth took place on 60 mM glucose in MRS medium for P. ~.c’~eri.tri/~/rilll\ (._ ) and P. f~i.~in~c~~vi.s (m) and in the BCYT medium for P. ~,c,r.c,~.i,ii/~/~ilu~ ( \‘) and P. frrting:cw.~i,~ (+); (averagefipurc for three separate expcrimenta).

1. Introduction Pcctitzutus crrrcisiiphilus has often been described as a beer spoilage bacterium (Haikara et al., 1981 b). This strictly anaerobic mesophilic bacterium was first described by Lee et al. (1978) in bottled beer stored at 30°C. Since then, it has been detected in different beer samples with offflavour (Back et al., 1979: Takahashi. 1983: Haikara, 1985). However, a recent taxonomic study of anaerobic, gram-negative, rod-shaped bacteria from breweries, led to the description of a new species of Pectin&us, PeLtinatus,frisingrr~sis, a 1lew species with different growth substrates pattern and a wider capability to excrete fermentation products (Schleifer et al., 1990). When growing in beer, these two species produce notable amounts of volatile fatty acids, mainly propiand acetate, resulting in unpleasant onate off-flavours. Two main metabolic pathways were described in the literature for the formation of propionate from pyruvate: (i) the acrylate pathway as in Clostridium propimictim (Doelle, 1975) or Buctcroidcs run~inidl (Wallnbfer and Bald-

win, 1967) and (ii) the succinate pathway involving a symmetrical intermediate to propionate (Tholozan et al.. 1988a; Houwen et al., 1991), as previously described in Desu(fihdbus propionicus and Pclohacter propionicus (Stams et al., 1984; Schink et al.. 1987). In this paper dealing with Prctincltu.s, P. c’erc~Giildzilus and P. ,f~i.singmsis growth characteristics will be presented and utilization of different carbon sources will be compared. Furthermore, a metabolic pathway to propionate will be proposed for P. ,fi~i.sit~gtx~i.c based on experiments in dense resting cell suspensions and using [2-‘“Clpyruvate as substrate. The ability of both strains to cause beer spoilage will be discussed from the above results.

2. Materials

and methods

All chemicals, analytical grade and buffers were Merck (Nogent-sur-Marne, purchased from

J.-L. Table 1 Influence of culture Pectinatus

species

P. cerecisiiphilus P. ,frisbrgemi.s

medium Culture MRS” BCYT” MRS BCYT

Tholorun

et al. / International

on P. cerrvisiiphilus medium

(DSM

Specific growth

35 (1997)29-39

Journul of Food Microbiology

20465) and P. fi?singmsis rate (h-l)

0.16 0.11 0.12 0.08

Lag-phase 25 30 20 30

(DSM (h)

20467) growth

Biomass

31

characteristics.

production

(mg TBP,‘mol

Glc)

23.8 4.91 29.82 5.55

__. Growth was performed at 3O”C, under N,:CO, TBP, total bacterial proteins; Glc. glucose. .’ MRS, Man et al.. 1960. ’ BCYT, Samain et al., 1982.

(85:15’%,) head space and with 60 mM glucose

France). The [2-“Clpyruvate was from Isotec (Saint-Quentin-Yvelines, France) with a labelling specificity over 98%. The glucose determination kit was from Sigma Diagnostics (I’Isle d’Abeau Chesnes, France). 2.2. Chenzicul malyses Volatile fatty acids (VFA) and alcohol concentrations were routinely determined by gas chromatography (GC) as previously described (Touzel et al., 1985). Cell growth was followed at 550 nm in butyl rubber-sealed tubes containing 10 ml of culture medium under N, head space, with a 320 RD spectrophotometer (Prolabo, Paris, France). Cell suspensions protein content was determined after Lowry with bovine serum albumin as a standard (Lowry et al., 1951). Organic acids other than VFA were determined after HPLC separation on a Waters Associates (Saint-Quentin-Yvelines, France) OA column (0.65 x 30 cm), with a sample volume of 20 ~1; elution was performed at 0.8 mljmin with 0.005 N H$O, as eluent. Glucose concentrations were determined enzymatically with a glucose oxydase kit (Sigma). 2.3. Organisms

and culture conu’itions

P. ccrevisiiphilus and P. frisingensis were obtained from the German Collection (DSM 20465 and DSM 20467, respectively). Batch anaerobic culture conditions and preparation of media were derived from Miller and Wolin (1974). Cul-

as the carbon

source.

tures were grown in liquid medium under Nz head space. Culture medium were either Lactobucilli MRS medium (Man et al., 1960) from Difco (Detroit, MI), or the basal CBBM medium (Zeikus and Wolfe, 1972) modified by Samain et al. (1982) and called BCYT. Unless otherwise indicated, this last medium was supplemented with pancreatic hydrolysate of casein (0.5 g/l), and with 50 mM of organic buffers for pH regulation, either 2-(N-morpholino)ethanesulfonic acid (MES), 1,4_piperazinediethanesulfonic acid (PIPES) or N-(2-hydroxyethyl)piperazine-N’-(2ethanesulfonic acid) (HEPES). Continuous cultures with pH regulation were made in MRS or CBBM medium without organic buffer addition, as described below. 2.4. Grolvth experiments Growth conditions determinations and amino acid effects on growth kinetics were performed in anaerobic 10 ml culture medium containing tubes under N, head-space (Miller and Wolin, 1974), unless otherwise indicated. Growth substrates were added to the culture at initial concentrations varying from 10 to 40 mM. Influence of head space on growth was measured under Nz, N2:C0, (85: 15%) and H,:CO, (85: 15%) atmosphere. Ethanol concentration effects on growth of both Pectin&us species were studied in BCYT medium under N, atmosphere. Continuous cultures with pH regulation were performed under N, head space in 2 1 stirred fermentors containing 1 1 of BCYT culture medium.

Pcctinatus

species

Culture

medium

Fermentation Propionate

P. (,(,~(,l,isiil)lfilu.t

P. /rl.\r/7gc~rl.t;.\

MRS.’ BCYT,’ BC MRS BCYT BC

products (mM)

70 62

Acctatc 43 I4

11

11

62 52 51

36 II 13

,’ MRS (Man ct al.. 1960): BCYT (Samain et al.. 198’7): BC. BCYT without ” Too littlc growth for reliable determinations

(mM)

Propionate

acetate

I .h 4.3 11 1.7 4. I 4.0

beast extl-act and biotqpticnae.

3. Results

Concentrated cells suspensions (IO”’ cells/ml) were prepared from 1.2 I vials from a 28 mM glucose-grown preculture in BCYT. Late logphase cells were harvested by centrifugation at 30000 x g for 30 min under N,; pellets were rinsed anaerobically (Miller and Wolin, 1974) before resuspension in 4-morpholinetwice propanesulfonic acid (MOPS), pH 6.2. The reaction was performed at 30°C and started by succinate addition; succinate consumption, acetate and propionate productions were followed by HPLC as described above.

Dense cells suspensions (10”’ cells/ml) prepared as described above, were placed in butyl rubber sealed NMR tubes under a N2 atmosphere. The temperature was allowed to equilibrate for 15 min before the reaction was started by the addition of [2-‘?I] 99% enriched ‘3C-pyruvate. The “C-NMR spectra of bacterial suspensions were recorded with a Bruker AM-300 spectrometer operating in the Fourier transform mode at a frequency of 75.45 MHz (Tholozan et al.. 1988a). Peaks in NMR spectra were identified by comparison with spectra obtained with pure reference compounds added to the culture. Quantitative spectra were obtained according to Tholozan et al. (1988b).

P. c,cr.cci.sii~~lliltls and P. ,f~isingrnsis growth was followed for different culture media, the MRS medium used by Lee et al. (1978) for reference strain purification and the BCYT medium (Samain et al., 1982) commonly used for acetogenic bacteria. Each growth experiment was performed three times at 30°C with 200 ml liquid medium in 1 1 vials under N,:CO, (85:15’%) head space. Glucose (60 mM) was used as a carbon source in each experiment. Typical growth kinetics of both species are presented in Fig. 1. In all cases. late log-phase cultures led to a final culture medium pH of 4.1 whatever the culture medium (results not shown). With MRS medium, lag-phases were shorter for both species and specific growth rates of the bacteria were 25% higher in this medium. Moreover. biomass measurements on late logphase cultures demonstrated a growth five times higher for both species in the MRS medium (Tablc 1). Total bacterial protein determinations demonstrated for the two culture media a 20% higher biomass production by P. fiisirqymi.s. When yeast extract and biotrypticase were omitted from BCYT medium, a 30% reduction of P. f~isiqynsis biomass production was measured, while P. c,cr-cl2~i.siipllilu.s biomass was reduced by 300% (results not shown).

J.-L

33

El 0 2 p

0.8

0, W 0.6 Y is 0.4 iis g 0.2 a 100

150

200 TIME (h)

250

300

350

Fig. 2. The influence of ethanol level on growth of Pectincrtus spp. in BCYT medium with glucose 60 mM as carbon source. Symbols: open symbols are used for P. crrrci.siip/du.r with ethanol concentrations of (0) 330, (‘-8) 650 and (fl) 980 mM; closed symbols for P. fri.sin,~en.si.s with ethanol concentrations of (+) 330 mM, ( n ) 980 mM and (X) I .3 M.

Substrate consumption and fermentation products accumulation were determined during these experiments. Propionate was in each case the major fermentation product (Table 2) as previously described (Lee et al.. 1978; Haikara et al., 1981a). Significant variations of acetate accumulations were measured, depending on the growth medium and Prctinutus species (Table 2). A constant propionate to acetate ratio of round 1.6 was routinely obtained in MRS medium for both Pettinatus species. On the other hand, a poorer medium resulted, for both species, in a dramatical decrease of acetate accumulation in the cultures. Acetate excretion level became up to 250% lower than the level reached with MRS medium. The use of a basal mineral medium (BC) instead of the BCYT medium showed unchanged metabolic features for P. frisingensis, though a 30% reduction of biomass production was measured. 3.2. The influence of gus phase composition on gravth P. cerevisiiphilus

and P. fiisingmsis

were grown

on MRS medium and BCYT medium under different head spaces, NL; N,:CO, (85:15’%) and H,:CO, (85:15%). Biomass productions and fermentation balances for a given culture medium were similar for both strains, whatever the head space. A slightly higher biomass production (less than 10%) was noticed on H,:CO,, together with about a 10% higher propionate to acetate ratio (results not shown). 3.3. Ethanol

ejyects on growth parameters

Ethanol alone was not used by the bacteria as a growth substrate. With 60 mM glucose as the carbon source, increasing amounts of ethanol were added, in order to reach final ethanol concentrations of up to 2 M. For both P. crrrcisiiphilus and P. JiYsingensis, growth did not occur at ethanol concentrations over 1.8 M (results not shown). The increase of initial ethanol concentration resulted in a decrease of growth rate and the lag-phase was increased (Fig. 2). Significant differences were measured between the two species, with P. cerrvisiiphilus being the most sensitive.

34

3 IA 0

1

3.5

4

4.5 5 5.5 6 6.5 pH OF CULTURE MEDIUM

7

7.5

Fig. 3. Influence of glucose (Glc) concentration and culture medium pH growth of Pw/iwfu.s species. Biomass was estimated by total bacterial protein (TBP) contents of late log-phase cultures. Symbols: open symbols were used for P. wreri.siiphilus cultures with (C) 6. (0) I2 and (‘~j) 18 mM of glucose: closed symbols for P. fri.siyqw.tir cultures performed with (m) 6. (+) I2 and (0) 18 mM

3.4. Gmbirwd ir$tuenc’e r~f’glumse, lmtute pH on the glow~tll of Pectinmtus species

und

The influence of each factor on growth kinetics of both Pectirzatus species was first determined. Additions of 5-60 mM glucose as a growth substrate led to constant fermentation balances for both Pectinutus species (results not shown). Good growth was observed for both microorganisms on lactate but fermentation balances were shifted to a much higher acetate production in P. ,j~isingm.sis cultures and trace amounts of butyrate were also detected in the cultures (results not shown). In P. ,fkingmsis cultures with 30 mM glucose, an increase of pH values from pH 4 to 6 did not influence growth kinetics. However, an increase of the pH value above 6.0 seems to progressively decrease both the specific growth rate (results not shown) and final biomass production (Fig. 3). The combined effects of glucose concentration, lactate level and pH are shown in Fig. 4. As seen. the final biomass was proportional to the glucose

level. Furthermore, with the exception of a very low pH (pH < 4. l), an increase of lactate concentration in the cultures led to a parallel increase of acetate accumulation (results not shown). Four main differences in biomass synthesis and fermentation products accumulation were shown between the two Pectinatus species: (i) the highest biomass production was measured at pH values ranging from 4.5 to 4.9 in P. ,fkisingensi.s and at a pH of 6.2 -6.3 in P. crrrcisiiplzilus; (ii) P. wrevisiiphilus did not grow at glucose concentrations below 5 mM and at pH values below 4.1, whatever the lactate concentration in the medium as represented for 20 mM lactate concentration in Fig. 4a, whereas combinations including even the lowest values of pH, glucose and lactate concentrations, led to a good growth in P. jksingmsis cultures (Fig. 4b); (iii) accumulations of propionate, the major fermentation product, were progressively increased by the increase of pH, glucose and lactate levels in P. ,fhsingcwsis, whereas this was less pronounced for P. ccrrllisiiplzi1u.s; and (iv)

J.-L.

Tldo:an

rt al. ))/Intonational

Journal of Food Microhiolo~y

r

35 (1997) 29-39

35

b

r e

Fig. 4. Combined effects of pH, glucose and lactate levels on Prctincrtusdevelopment. Lactate concentration was varied from 4 to 40 mM, pH values were varied from 3.7 to 7.2 and glucose concentrations from 5 to 60 mM. Results presented here are values obtained with 20 mm lactate: (a) final absorbance of P. crrerisiipldus cultures: (b) final absorbance of P. f~isin~rrr.ris cultures: (c) acetate production by P. c~e~e~~isiip/~ilzr.s: (d) acetate production by P. ,frisingmCs: (e) propionate production by P. crrc~i.\iipl~ihts; (f) propionate production by P. f~rsinge~zsis. Symbols: open symbols were used for I’. mvwwt~hilus with glucose concentrations of 6 (8’. ). I2 (0) and 18 mM (7) as the carbon source: closed symbols for P. fri.ringensi.r with glucose concentrations of 6 (0). I2 (+) and I8 mM (m).

a particular metabolic feature was observed acetate accumulation in P. fCsingmsi.s, at a value of 4.5 and glucose concentrations below mM, a minimum of acetate production was served.

for pH 20 ob-

3.5. Metabolic

pathwr~y

forpropionmte

production

Preliminary studies in P. cereuisiiphilus suggested that propionate was produced from glucose with succinate as the metabolic intermediate (un-

published results). The involvement of‘ this pathway in P. ,/kisi/y~/lsj.\ wits indicated by the expcrimcnt shown in Fig. 5. Succinatc (IO mM) resulted in 6 mM of propionatc and 2 mM ol acetate within 3 h (Fig. 5). On the other hand, pyuwtc u’as sho~vn to be metabolized by f’. /r.i.si/rgc~si.s. The use of put-e [I-“C]pyruvatc in growing culturcb, led to the accumulation of equal amounts of [Z-’ ‘C] and [3-’ ‘C]propionates. the initial specific position of the “C-label in pyruvatc was randomized in the fermentation products (results not sl10\?;11).

4. Discussion The results obtained here demonstrated similar physiolopical characteristics. but also significant differences between the t\vo Pcwtinu/u.\ species investigated. I’. fi~i.sir7gc~r~si.s showed a higher resistance towards alcohol and it grew over ;i kvidcr range of pH. fa\ouring growth of this spcties in beer. Average lactate level measurements in beers indicated concentrations up to 3 mM

(Charal~ui~bous. 198 I ). For both Pwtimrtm species, lactic acid displayed ;i significant role in growth at ;I low glucose lwel. Icading to higher acetate le\,els. when compared with experiments pc‘rlbrmcd at initial glucose concentrations above 12. mM (Fig. 4). The influence of this compound on Pwti~~r~rr.~ dcvclopmcnt during beer spoilage R;IS thub demonstrated. Classical propionibactcria li‘rment lactate to propionnte and acetate (De Vrics et al.. 1973), according to the overall reaction reported in Table 3. This corresponded in our experiments to propionate levels increasing faster than acetate production with increasing lactate concentrations at fixed pH and glucose concentration values. No interactions between glucose. lactate and pH levels could be Ibund in P. c~crc~1~i.sii~~llilu.v. In P. ,fj.;.si/2~(,/7.~i.~ importance of lactate colihowever. ;I particular ccntration in the medium for strain devclopment. when associated with low pH levels in the culture. was observed. acids determinations during Volatile liltty growth of the two Pvctitwtus species indicated low propionutc le\:el variations compared to those ob-

37

Table 3 Hypothetical

redox

reactions

-+ -+ -+

Glucose+2H, Glucosc+4H,O 3 Lactate Calculations

involved

from Thauer

in

Pectimtusspecies metabolism

on glucose

2 Propionatc+ 2H + + 2H,O 2 Acetatem+2HCO;+4HL+4H’ 2 Propionate fAcetate _ + HCO;

+ H’

AC;,= -358.1 kJ (I) AC;,= -206.7 kJ (2) AC;, = ~ 165.0 kJ (3)

et al., 1977.

served on acetate production. Hypothetical redox reactions involved in glucose conversion to propionate and acetate were presented in Table 3. Calculations from Thauer et al. (1977) indicated that these two reactions generate standard free energy’. It was generally assumed that Y,,,, (grams of cells synthesized per mole of ATP produced) of 10.5 were currently measured in complex media cultures for anaerobic microorganisms (Bauchop and Elsden, 1960). The experimental overall reaction of glucose fermentation to propionate and acetate in P. crwcisiipl~ilus would lead to a theoretical amount of 3.2 and 2.1 mol of ATP synthesized per mole of glucose fermented on the MRS and BCYT culture medium, respectively (Decker et al., 1970; Stouthamer and Bettenhaussen, 1973). For P. J~isingmsis, similar calculations would lead to 2.8 and 2 mol of ATP synthesis on MRS medium and BCYT (or BC), respectively. The corresponding amounts of biomass synthesized on basal medium (9 g dry weight and 10.1 g dry weight/mole of glucose for P. crtwisiipldus and P. jkisingmsis, respectively) would indicate bacteria working at a respective efficiency of 41 and 48%. This is in good agreement with the assumption that anaerobic microorganisms usually show efficiencies below 50% (Decker et al., 1970). Here again, P. jhsingensis displayed better efficiency in growth substrate utilization. Calculated values with MRS medium demonstrated an overestimation of biomass production probably due to medium composition; however, optical density measurements indicated a much higher biomass production for both strains on this last medium. Another explanation could be the existence of additional fermentation products, leading to a more exergonic overall fermentation balance on MRS medium. Acetate was demonstrated to be one of the main ways of ATP synthesis via substrate level

phosphorylation in anaerobic bacteria (Thauer et al., 1977) the good correlation between acetate production level and Pectinutus species growth would indicate acetate kinase as the major pathway for energy synthesis in these bacteria. However, maximal final biomass production at low glucose levels and optimal pH occurred in P. fiisingmris with minimal acetate production. This would suggest either (i) that ATP could also be synthesized during propionate excretion by a propionate kinase activity, one part of the produced propionate being used by the bacteria to regenerate reduced pyridine nucleotides (Gottschalk, 1979) or (ii) that in good growth conditions, acetate levels measured in the medium would result from a combined action of phosphate acetyl transferase’acetate kinase activity and acetate thiokinase activity (Tholozan et al.. 1992). This last point remains to be solved. Up to now, two metabolic pathways to form propionate from pyruvate, have been described in anaerobic bacteria: the acrylate pathway. involved in C. propionicum (Cardon and Barker. 1946) preserved the molecular carbon skeleton all over the pathway to propionate. So, the specific label of metabolic precursors was not randomized during propionate synthesis and [2-‘Clpyruvate used as substrate would have led to [2-‘Clpropionate as the sole labelled position as already demonstrated in C. neopropionicum (Tholozan et al., 1992). On the contrary, the other way to propionate involved two symmetrical molecules, fumarate and succinate, resulting in the equal distribution of the “C-label from the [2“Clpyruvate: in this pathway, the propionate formed through succinate is thus either labelled on the carbon two or on the carbon three (Houwen et al., 1987; Robbins, 1987, 1988). Theoretically. an equal amount of both molecules is formed with this pathway as previously demon-

strated in Drsu(ftibu1hu.s pwpiotkws (Stams et al., 1984) or Pclohuct~r propiotzicus (Schink et al.. 1987), it was also the case in P. ,fj.i.sitzgwsis. This metabolism is thus in good agreement with the previous results on the P. ccwcisiiphilus metabolic pathway, where malonate utilization, a competitive inhibitor of succinate oxidoreductase, inhibited propionate production (Haikara et al., 198la), it was then postulated that the succinate pathway led to propionate in this bacterium.

Acknowledgements The authors thank M. Kubaczka, G. Delattre. A. Ronse and B. Eulalie for helpful technical supply.

References Back. W.. Weiss. N. and Seidel. 11. (1979) Isolieruny und systcmatische Zuordnung bierschadlicher Bakterien. II. Gram-negative anaerobe Stabchen. Brauwissenschaft 32. 233-238. Bauchop, T. and Elsdcn. S.R. (1960) The growth of microorganisms m relation 10 their energy supply. J. Gen. Mtcrobiol. 23. 457 469. Cardon, B.P. and Barker. H.A. (1946) Two nw amino-acid fermenting bacteria. Clostridium propionicum and Diplococcus glycinophilus. J. Bacterial. 52. 629- 634. Charalambous, G. (1981 J In\olatile constituents of beer. In: J.R.A. Pollock (editor), Brewing Science, Vol. 2. Academic Press. London. pp. I68 -255. Decker. K.. Jungermann. K. and Thauer. R.K. (1970) Energ) production in anaerobic organisms. Anpew. Chem. Int. Ed Engl. 9. 138 158. De Vries. W.. van Wyck-Kapteqn, M.C. and Stoulhamer. A.H. (1973) Generation of ATP during qtochromc-linked anaerobic electron transport in propionic acid bacteria. J. Gen. Microbial. 76. 31 41. Doelle. H.W. (I 975) Bacterial Metabolism. 2nd edition. AC+ detnic Press. Nw York. Gottschalk. G. ( 1979) Bacterial Metabolism. Springer Verlag. New York. Haikara. A., En&. T.M. and Lounatmaa, K. (1% la) The genus Pcctinatus. a neu group of anaerobic beer spollqe bacteria. Proc. E. B. C. Congl-. 24. 229 -240. Haikara. A.. Penttila. L.. Enari, T.M. and Lounatmaa. K. (I98 I b) Microbiological. biochemical, and electron microscopic characteriz~ltion of a Pcctinatus strain. Appl. En\,ron. Microbial. 41, 51 I 517. Haikara. A. (1985) Dctcction of Pectinatu? contatninan(s in

beer. ASBC J. 43. 43 45. Houuen. F.P., Dijkcma. C.. Schoenmakers. C.H.H.. Stams. A.J.M. and Zehnder. A.J.B. (1987) “C-NMR study 01’ coculturc. propionatc degradation by a methanoyenic FEMS Microblol. Lett. 41, 269- 274. Houucn. F.P., Dijkema, C.. Statnr, A.J.M. and Zchnder. A.J.B. (1991) Propionate metabolism tn anaerobic bacteria: determination of carboxylation reactions with “C-NMR spectroscopy. Biochim. Biophys. Acta 1056. I26 132. Let, S.Y.. Mabcc, M.S. and Jangaard, N.O. (1978) Pectmatus. a nw genus of the family Bactcroidaceac. Int. J. Syst. Bacterial. 78. 582 594. Lofir).. 0.11.. Rosenbrough. N.J.. I-arr. A.L. and Randal. R.J. (1951 ) Protein measurement with the Folin phenol reagent. J. Blol. Chem. 163. 265-275. Man, J.C.. de Rogosa, M. and Sharpe. M.E. (1960) A medium for the cultivation of Lactobactlli. J. Appl. Bacterial. 23. 130~ 135. Miller. T.L. and Wolin. M.J. (1974) .A scruni bottle tnodliication of the Hungate technique for cultivating obligate anaerobcs. Appl. Environ. Microbial. 27, 985 987. Robbins. J.E. (I 987) ’ 'C nuclear magnetic resonance studies of propionatc catabolism in tnethanogcnic cocultures. Appl. Environ. Microbial. 53. 7260- 2261. Robbins. J.E (1988) A PI-oposed pathhay for catabolism ol propionate in methanogcnic acultures. Appl. Environ. Microblol. 54. I300- I30 I. Samain. E.. ,4lbagnac, G.. Dubourguxr. H.C. and Touzel. J.P. (I 982) Characterization of a new propionic acid bacterium that ferments ethanol and displays a grouth factor-dependant association with a gram negative homoacctogen. FL:MS Microbial. Lett. 15, 69 71. Schink. B.. Krctner, D.R. and Hansen. T.A. C1987) Pathway of propionate formalion from ethanol in Pelobacter propioncus. Arch. Microbial. 147, 121 327. Schleifel-. K.H., Lcutcritz, M., Weiss. N., Ludwig. W.. Kirchhof. G. and Scidel-Riifcr. H. (1990) Taxonotnic study of anaerobic. gram-negative. rod-shaped bacteria from brewcries: emended description of Pectinatus cerevlsiiphiltts and dcscrlplion of Pectinatus friaingensia sp. nav.. Selenomonas Iacticifex hp. nav.. Zymophilua raflinosi\orans gcn. not , sp WV., and Zqmophilus pauci\orans sp. nav. Int. .I. Syat. Bacterial. 10. I9 27. Stouthamer. A.11. and Bcttenhaussen. C. (1973) Utilization ol energy for growth and maintenance in continuous and balch cultures of microorpaniams. Biochim. Biophys. Acta 301. 53 70. Stanis. A.J.M., Krcmer. D.R.. Nicolay. K.. Weenk. G.H. and Hansen, T.A. (1984) Pathway of propionate formarlon in Desulfobulbus propionicus. Arch. Microbial. IN. I h7 173. Takahaahi. N. ( 1983) Presumed Pcctinatus atrains isolated from .japanese beer. Bull. Brew. Sci. 28. I I 14. Thauer. R.K., Jungcrmann. K. end Decker, K. (1977). Energy conwr\ation in chemotrophic anaerobic bacteria. Bactcriol. Rc\. 41. 100 180. Tholozan, J.L.. Samain. E.. Grivct. J.P.. Moletta. R.. Dubour-

guier, H.C. and Albagnac, G. (1988a) Reductive carboxylation of propionate to butyrate in methanogenic ecosystems. Appl. Environ. Microbial. 54.441-445. Tholozan, J.L., Samain. E. and Grivet, J.P. (198813) Isomerization between n-butyrate and iso-butyrate in enrichment cultures. FEMS Microbial. Ecol. 53, 1877191. Tholozan, J.L.. Touzel. J.P.. Samain, E., Grivet, J.P.. Prensier. G. and Albagnac. G. (1992) Clostridium neopropionicum sp. nav., a strict anaerobic bacterium fermenting ethanol to propionate through acrylate pathway. Arch. Microbial.

157, 2499257. Towel. J.P.. Petroff. D. and Albagnac, G. (1985) Isolation and characterization of a new thermophilic Methanosarcina, the strain CHTI 55. Syst. Appl. Microbial. 6, 66-71. Wallnofer, P. and Baldwin. R.L. (1967) Pathway of propionate formation in Bacteroides ruminicola. J. Bacterial. 93, 504505. Zeikus. J.G. and Wolfe. R.S. (1972) Methanobacterium thermoautotrophicum sp. nav., an anaerobic. extreme thcrmophile. J. Bacterial. 109. 7077715.