APPLIED AND ENVIRONMENTAL MICROBIOLOGY, June 1991, p. 1630-1634 Vol. 57, No. 60099-2240/91/061630-05$02.00/0Copyright © 1991, American Society for Microbiology

Methanogenic Bacteria as Endosymbionts of the CiliateNyctotherus ovalis in the Cockroach Hindgut

HUUB J. GIJZEN,l CEES A. M. BROERS,2 MARTIN BARUGHARE,l AND CLAUDIUS K. STUMM2*Applied Microbiology Unit, Faculty of Science, University ofDar es Salaam, Dar es Salaam, Tanzania,' and

Department of Microbiology, Faculty of Science, University of Nijmegen,NL-6525 ED Nijmegen, The Netherlands2

Received 4 December 1990/Accepted 21 March 1991

Production of methane in the hindgut of the cockroach Periplaneta americana was found to vary, dependingon the feeding regimen. Methane production was positively correlated with the numbers of the ciliateNyctotherus ovalis living in the cockroach hindgut. Defaunation of the cockroaches by means of lowconcentrations of metronidazole (Flagyl) resulted in a quick drop of methane production. Addition of themethanogenic substrates acetate and formate to isolated hindguts stimulated methane production. Inside theciliate cells, autofluorescing bacteria could be demonstrated which were presumed to be methanogens. Electronmicroscopy revealed that the bacteria resembled Methanobrevibacter and that they were closely associated withorganelles which contained infolded membranes and which were presumably hydrogenosomes.

The anaerobic environment in the gastrointestinal tract ofherbivorous animals stimulates the development of specificmicrobial communities, composed of bacteria, fungi, andprotozoa. These intestinal microorganisms are considered tobe beneficial to the host rather than harmful. The microbialecology of the rumen and the termite hindgut has beenstudied extensively, and many genera of bacteria and proto-zoa inhabiting these ecosystems have been described (1, 11).

Besides the mutualistic associations between intestinalmicroorganisms and herbivorous animals, numerous symbi-otic interrelationships also exist between microorganismsthemselves in ruminant and hindgut ecosystems (4, 5, 11).Various interactions have been reported to occur betweenbacteria and protozoa in these ecosystems (11, 13, 18).A novel type of association between ruminal ciliates and

hydrogenotrophic methanogenic bacteria was described byVogels et al. (21) and Stumm et al. (16). Since these metha-nogens, presumably Methanobrevibacter ruminantium,were attached to the outer surface of the ciliate cells, theassociation was referred to as an episymbiosis. It wassuggested that the metabolic interaction between both part-ners in the symbiosis was based on interspecies hydrogentransfer. More recently, Lee et al. (14) reported a similarsymbiosis of methanogenic bacteria with flagellated proto-zoa from termite hindgut, but here the methanogens werelocated inside the protozoan cell. These findings, togetherwith the discovery of methanogenic endosymbionts in sapro-pelic ciliates by van Bruggen et al. (19), suggest that themethanogen-protozoa symbiosis may be a common featureof many anaerobic ecosystems.So far, work on the microbial ecology of the cockroach

hindgut has been fragmentary. Production of considerableamounts of methane by cockroaches is a well-known phe-nomenon, and it indicates the presence of methanogenicbacteria. The site of methane production has been shown tobe the hindgut (5). The results of this study demonstrate thatmethanogenic bacteria are found as endosymbionts in Nyc-totherus ovalis, a ciliate present in high numbers in thehindgut of the cockroach Periplaneta americana. Different

* Corresponding author.

feeding regimens were found to significantly affect the pop-ulation growth rate of the cockroach, the density of hindgutciliates, and the degree of methane production by wholeanimals and isolated hindgut preparations.

MATERIALS AND METHODS

Cockroaches. Cockroaches (P. americana) were collectednear the university campus in Dar es Salaam, Tanzania. Fivecultures of cockroaches were kept in 5-liter glass containerscovered with a small-mesh wire netting (1-mm mesh size).Pieces of polystyrene foam were put into the containers toserve as shelters. Different regimens of feeding and wateraddition were applied to the cultures of P. americana (Table1). The composition of the different diets was chosen toevaluate the effects of different levels of nitrogen and dif-ferent sources of carbohydrates. White tissue paper wasused as a source of cellulose; fish flour was prepared fromthe species Rastineobola argentea, which was dried andsubsequently ground. Water and substrates were added adlibitum into separate petri dishes inside the containers,which were cleaned twice a week. The cultures were startedby adding 25 adult cockroaches to each container. Thecontainers were kept in the dark at room temperature (25 to27°C). Experiments were started after a stabilization periodof 3 months, using adult insects weighing 0.9 to 1.8 g.Methane production. Methane production of intact cock-

roaches was measured by bringing four adults from eachculture into 250-ml serum bottles. The weight of the insectswas recorded in every experiment. The bottles were sealedwith a rubber stopper and a screw cap; subsequently, 0.2 mlof ethane was added as an internal standard. Bottles wereincubated at room temperature. The methane content ofeach bottle was analyzed by gas chromatography (12) every15 min during the first 2 h and every 30 min during thefollowing 4 h. Each incubation was performed in duplicate,and the whole procedure was repeated four times on dif-ferent days to compensate for possible daily fluctuations.To determine methane production in the hindgut of P.

americana, adult insects were dissected under a continuousstream of nitrogen gas. Hindguts from six insects from eachcockroach culture were cut into pieces and pooled in 12 ml of

1630

on March 27, 2021 by guest

http://aem.asm

.org/D

ownloaded from

METHANE PRODUCTION IN THE COCKROACH HINDGUT 1631

TABLE 1. Feed and water regimens for five different culturesof P. americana

Culture Cellulose Fish flour Maize flour Water Addition to water

Ia + + - +II + - - + NH4Cl, 1.5 g/

literIII - + - +IV - - + +V - - + + Metronidazole,

0.1 mg/mla Ratio of cellulose to fish flour = 3:1 (wt/wt).

anaerobic buffer (8), which was modified by the addition ofNa2S 9H20 (2.1 mM) and L-cysteine-HCl (1.8 mM) andadjusted to a pH between 6.8 and 7.0. Preliminary experi-ments had indicated that N. ovalis could survive for at leastseveral hours in the modified buffer (results not shown). Sixsealed serum bottles (18-ml volume), each containing 2 ml ofthe final hindgut suspension, were incubated at room tem-perature, after addition to duplicate samples of 1 ml ofanaerobic buffer containing sodium acetate (50 mM) or

formic acid (217 mM) as methanogenic substrate. To twobottles, 1 ml of anaerobic buffer without an additionalmethanogenic substrate was added. The contents of allbottles were adjusted to pH 6.8 prior to incubation.

After flushing of the bottles with N2 gas and addition of 0.2ml of ethane, methane contents in the bottles were deter-mined in the same way as described for the intact cock-roaches. To compensate for daily fluctuations, the procedurewas repeated four times. All methane values are expressedin volume (milliliters) produced (1 atm [1 atm = 101.29 kPa],25°C).

Metronidazole treatment. In one of the cockroach cultures(culture V), the antiprotozoal drug metronidazole (Flagyl)was added to the drinking water at a concentration of 0.1mg/ml. Both short-term (days 1 to 8) and long-term (after 3months) effects of metronidazole treatment on methaneproduction and on number of N. ovalis cells in hindguts werestudied. The short-term effects were studied in cockroacheswhich were taken from cockroach culture IV and were

subsequently treated in a separate container.Determination of ciliate numbers. Ciliate enumeration was

done on homogeneous samples obtained by suspending thehindguts of four insects into 1 ml of buffer (8) and cuttingthem into small pieces. After thorough rinsing of the hindgutfragments, ciliates were fixed by addition of formaldehyde toa final concentration of 4%. The number of N. ovalis cellswas determined microscopically by using a counting cham-ber (0.2 mm deep). The results obtained by this method were

checked several times by counting all N. ovalis cells in a

20-,ul diluted hindgut suspension. Ciliate counting was re-

peated several times on different days to compensate forpossible daily fluctuations. Average hindgut volume forcockroaches from each culture was estimated by determin-ing the total volume of four to six pooled hindguts.

Microscopy. The presence of methanogenic endosym-bionts was checked by means of Leitz epifluorescencemicroscopy, as described by Doddema and Vogels (6). Thesame technique was used to study free-living methanogens inhindgut contents. For transmission electron microscopy,ciliate cells were first isolated from the hindgut content bymeans of micropipettes. Cells were then fixed at 0°C in 3%glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) andpostfixed on ice in 2% OS04 in the same buffer. The cells

TABLE 2. Methane production by intact adults and by isolatedhindguts of P. americana cultured on different dietsa

Culture ml of CH4/ ml of CH4/hindgut/daybinsect/dayb Controlc Acetate Formate

I 1.81 ± 0.32 1.69 ± 0.43 1.92 ± 0.72 3.6 ± 0.50II 1.03 ± 0.28 0.86 ± 0.52 1.03 ± 0.42 2.8 ± 0.60III 0.93 ± 0.42 0.79 ± 0.48 0.99 ± 0.39 2.1 ± 0.43IV 1.68 ± 0.34 1.25 ± 0.48 1.51 ± 0.45 3.73 ± 0.38V 0.005 NDd ND NDa All values represent means + standard deviations (n = 8).b Calculated from 6-h measurements.c No external source of methanogenic substrate added.d ND, not determined.

were contrasted with 1% uranyl acetate, and thin sectionswere stained with lead citrate before examination with aPhilips EM 200 electron microscope.

Statistical analysis. The statistical significance of differ-ences between the means was assessed by Student's t test.Statistical significance of differences between different cul-tures was assessed by using the single-factor analysis ofvariance method.

RESULTS

Methane production by intact adults and by isolatedhindguts of wild P. americana cultured on different sub-strates is shown in Table 2. The production of methane byintact adults appeared to be significantly higher (P < 0.001)for cockroaches fed on cellulose-fish and maize flour diets(1.68 to 1.81 ml/insect/day) compared with the diet types incultures II and III (0.93 to 1.03 ml/insect/day). The amountsof methane produced by isolated hindguts were somewhatlower than those produced by intact insects. Differences inmethane production between various cockroach cultures,however, were comparable for intact insects and hindguts.Methane production by isolated hindguts could be in-

creased by addition of acetate or formate as external meth-anogenic substrate (Table 2). The addition of acetate hadonly a minor positive effect on methane production, whereasthe addition of formate resulted in more than doubling of themethane production rate.Adult cockroaches used in the experiments weighed be-

tween 0.9 and 1.8 g. The average weight of cockroaches inthe various cultures I to IV, however, did not show anysignificant difference. The least significant difference wascalculated to be 0.25 g. Metronidazole-treated insects (cul-ture V) had a lower average weight, which was significantlydifferent (P < 0.001) from cockroaches in cultures I and IV(Table 3). Therefore, the differences in methane production

TABLE 3. Average weights of adult cockroaches from variouscultures and their specific methane production

Methane production, ml/gCulture Wt/cockroach, g of body wt/day (mean ±(mean±SD;n =8)SEM; n = 8)

I 1.49 ± 0.30 1.21 + 0.10II 1.38 ± 0.24 0.75 ± 0.11III 1.41 ± 0.19 0.65 ± 0.09IV 1.53 ± 0.28 1.09 ± 0.08V 1.20 ± 0.17 0.004 ± 0.0005

VOL. 57, 1991

on March 27, 2021 by guest

http://aem.asm

.org/D

ownloaded from

APPL. ENVIRON. MICROBIOL.

TABLE 4. Number of N. ovalis cells in hindgut of P. americanacultured on different dietSa

Type of diet N. ovalislhindgut Hindgut vol (p.l)' N. ovalis/ml, 104

I 3,970 ± 1,839 65 ± 4 6.11 ± 2.83II 2,806 ± 1,012 52 ± 5 5.40 ± 1.95III 1,262 ± 787 61 ± 4 2.07 ± 1.29IV 3,102 + 1,023 66 ± 3 4.70 ± 1.55V NDC 64 3 ND

a All values represent means ± standard deviations (n = 5).b Average total hindgut volume, including hindgut tissue.ND, not detectable.

cannot be attributed solely to differences in the weight ofindividual groups of cockroaches.The number of N. ovalis cells in the hindguts of P.

americana appeared to be affected by the feed composition(Table 4). The numbers ranged from 4.7 x 104 to 6.1 x 104cells per ml for cockroaches fed on cellulose and maize flourdiets, but only about 2.1 x 104 cells per ml were found incockroaches feeding on a fish flour diet. The least significantdifference was 1,182 cells per hindgut (P < 0.0005).

Inspection of N. ovalis by means of epifluorescence mi-croscopy revealed the presence of numerous methanogenicbacteria within the ciliate cells (probably >500 methanogensper ciliate). Methanogens were present in every N. ovaliscell observed from wild P. americana feeding on all of thevarious diets.

Electron microscopy showed that the bacteria in thecytoplasm of N. ovalis were short, oval rods (0.3 by 0.5 p.m)with tapered ends, resembling the Methanobrevibacter mor-

photype (Fig. la and b). Division stages were frequentlyencountered. Most of the bacteria appeared to be closelyassociated with organelles that had an irregular shape andstrikingly resembled the hydrogenosomes described fromseveral free-living anaerobic protozoa (17, 22). In many ofthe hydrogenosome-like organelles of N. ovalis internalmembranes could be seen (Fig. lc). The methanogenicbacteria and the associated hydrogenosome-like organelleswere not evenly spread throughout the ciliate cytoplasm, butformed clusters or packages (Fig. la).

Addition of the antiprotozoal drug metronidazole to thedrinking water (0.1 mg/ml) had pronounced effects on bothmethane production and the size of the N. ovalis populationin the hindgut. Addition of metronidazole over periods of 3months or longer resulted in a virtually complete inhibitionof methane production by both intact cockroaches andhindgut suspensions (Table 2). Careful microscopic observa-tion of the hindgut content revealed that N. ovalis was

completely eliminated from hindguts of metronidazole-treated animals (Table 4).To study the short-term effects of metronidazole, methane

production and the number of N. ovalis cells were deter-mined for a period of 8 days after the start of metronidazoletreatment (Fig. 2). The results of this experiment revealedthat N. ovalis cells were eliminated from the hindgut within3 days after the metronidazole treatment started. Methaneproduction was reduced to about 6% of the control valuewithin the same period of time.The rate of population increase of the cockroaches also

appeared to be affected by the different diets applied. Thenumber of cockroaches grown on cellulose and maize flourdiets doubled after every 2 to 3 months, whereas on the fishflour diet the doubling time appeared to be more than 5months. When cockroaches were fed on pure cellulose,

without addition of NH4Cl to the drinking water, there wasvirtually no increase in number over a period of 12 months.This was probably due to nitrogen deficiency. The cock-roach culture treated with metronidazole showed doublingtimes of about 9 months.

DISCUSSION

The production of methane by cockroaches is a well-known phenomenon. Cruden and Markovetz (4) reportedvalues of 5.4 to 13.4 ml/day in laboratory-raised Eublaberusposticus as well as substantial amounts by laboratory-raisedP. americana, wild P. americana, and Gromphadorhina sp.In this study, methane emissions from wild P. americanaadults were six to seven times lower than those reported forE. posticus. The differences may be accounted for bydifferences in the diets used and by species differences suchas a major difference in body weight (E. posticus weighsabout 2 to 4.5 g).Cruden and Markovetz (5) reported methane production in

cockroaches, originating from their hindguts, though at thissite no H2 was detectable. It was shown that the majorsource for methanogenesis in the hindgut is formate orH2/CO2 and not acetate. Our results, however, suggest thatmethane production from acetate occurs at least to someextent.Our observation that methanogens in the hindgut of wild

P. americana were found in great numbers as endosym-bionts of the anaerobic ciliate N. ovalis provides newinformation on the origin of methane production by cock-roaches. N. ovalis was only found in the hindgut andmethanogens were not found in association with any otherprotozoal species in the cockroach hindgut or midgut sys-tem. Moreover, only very few free-living methanogens wereobserved microscopically in the cockroach hindgut. Thissuggests that the methanogenic endosymbionts of N. ovalisare the major source of methane production in P. americana.

It is improbable that the considerable decrease of methaneproduction in cockroaches during metronidazole treatmentmight be due to direct inhibition of the methanogenic bacte-ria, as the concentration applied in this study (0.1 mg/ml)was only about 10% of the concentration used in otherstudies (3, 4). Moreover, treatment of anaerobic methano-genic sludge with metronidazole (0.1 mg/ml) did not inhibitmethane production, not even over extended periods of time(results not shown). Consequently, the cessation of methaneproduction from the cockroaches can most probably beattributed to the elimination of N. ovalis cells which serve asa host for the methanogens. The morphological similarity ofthe N. ovalis endosymbiont to Methanobrevibacter suggeststhat the endosymbiont might consume H2/CO2, as is true forboth Methanobrevibacter and the methanogens hithertoisolated from free-living anaerobic protozoa (9, 20). Themethanogenic endosymbionts of many of the latter organ-isms exist in a close spatial association with organelles whichare thought to be hydrogenosomes, providing hydrogen asan energy source for the bacteria (17, 22). Apparently, asimilar situation is present in the gut-inhabiting N. ovalis, inwhich we find the methanogens intimately associated withhydrogenosome-like structures. Thus, also in the N. ovalis-methanogen system an interspecies hydrogen transfer can besupposed to be the physiological basis of the symbiosis. It isnoteworthy in this context that packages of bacteria andorganelles, morphologically similar to the methanogens andhydrogenosome-like organelles described here, are also vis-ible on electron microscopic photographs of N. ovalis thin

1632 GIJZEN ET AL.

on March 27, 2021 by guest

http://aem.asm

.org/D

ownloaded from

METHANE PRODUCTION IN THE COCKROACH HINDGUT 1633

FIG. 1. Transmission electron micrographs of N. ovalis. (a) Low magnification shows the clustering of hydrogenosome-like organelles(arrows) with the associated, electron-dense endosymbionts. V, Food vacuole. Bar, 5 ,um. (b) Endosymbiotic bacteria (B) resemblingMethanobrevibacter; *, division stage. The bacteria are neighbored by hydrogenosome-like organelles (M). Bar, 0.5 Jim. (c) Hydrogenosome-like organelle (M) with infolding of the inner membrane (arrow). Bar, 0.5 ,um.

VOL. 57, 1991

on March 27, 2021 by guest

http://aem.asm

.org/D

ownloaded from

1634 GIJZEN ET AL.

LIp

JIC

L.C*c

E

1.6

1.2

le

14

0 1 2 3 l 5 6 7 8

time (days)FIG. 2. Effect of adding metronidazole (0.1 mg/ml) to drinking

water on number of N. ovalis cells (0) and CH4 production (0).

sections published very recently by Mohr et al. (Fig. 1 and 4in reference 15). However, since the authors focus on

intermediate-like filaments, they do not point out thesestructures. Together with earlier reports on associations ofmethanogens with sapropelic protozoa (2, 19), termite hind-gut flagellates (14), and ruminal ciliates (16, 21), the occur-

rence of methanogens as endosymbionts of a cockroach-inhabiting ciliate suggests that the methanogen-protozoasymbiosis might be a general feature of many anaerobicecosystems.

In several sapropelic protozoa the microbodies were

shown to possess hydrogenase activity, thereby revealingtheir hydrogenosomal nature (7, 22). However, the physio-logical significance of the hydrogenosome-like organelles ofN. ovalis needs further elucidation. The internal membranespresent in these organelles resemble the cristae of mitochon-dria and are similar to those found in the hydrogenosomes ofTrimyema compressum (10) and several anaerobic protozoaexamined by Finlay and Fenchel (7). The ultrastructure ofthe methanogen-associated organelles of N. ovalis can beinterpreted to favor their hypothesis on a mitochondrialorigin of hydrogenosomes (7). More studies are required toelucidate both the nature and function of the hydrogeno-some-like organelles and the ecological significance of thesymbiosis of methanogenic bacteria and the protozoan cell.

ACKNOWLEDGMENTS

We thank G. D. Vogels for critically reading the manuscript andF. Kansiime for his initial work in studying the methanogen-ciliatesymbiosis in cockroaches.

REFERENCES1. Breznak, J. A. 1982. Intestinal microbiota of termites and other

xylophagous insects. Annu. Rev. Microbiol. 36:323-343.2. Broers, C. A. M., G. Brugerolle, C. K. Stumm, and G. D. Vogels.

1990. Psalteriomonas lanterna gen. nov. sp. nov., a free-livingamoeboflagellate isolated from freshwater anaerobic sediments.Eur. J. Protistol. 25:369-380.

3. Cruden, D. L., and A. J. Markovetz. 1979. Carboxymethylcellulose decomposition by intestinal bacteria of cockroaches.

Appl. Environ. Microbiol. 38:369-372.4. Cruden, D. L., and A. J. Markovetz. 1984. Microbial aspects of

the cockroach hindgut. Arch. Microbiol. 138:131-139.5. Cruden, D. L., and A. J. Markovetz. 1987. Microbial ecology of

the cockroach gut. Annu. Rev. Microbiol. 41:617-643.6. Doddema, H. J., and G. D. Vogels. 1978. Improved identification

of methanogenic bacteria by fluorescence microscopy. Appl.Environ. Microbiol. 36:752-754.

7. Finlay, B. J., and T. Fenchel. 1989. Hydrogenosomes in someanaerobic protozoa resemble mitochondria. FEMS Microbiol.Lett. 65:311-314.

8. Gijzen, H. J., K. B. Zwart, P. T. van Gelder, and G. D. Vogels.1986. Continuous cultivation of rumen microorganisms, a sys-tem with possible application to the anaerobic degradation oflignocellulosic waste materials. Appl. Microbiol. Biotechnol.25:155-162.

9. Goosen, N. K., A. M. C. Horemans, S. J. W. Hillebrand, C. K.Stumm, and G. D. Vogels. 1988. Cultivation of the sapropelicciliate Plagiopyla nasuta Stein and isolation of the endosym-biont Methanobacterium formicicum. Arch. Microbiol. 150:165-170.

10. Goosen, N. K., S. Wagener, and C. K. Stumm. 1990. A compar-ison of two strains of the anaerobic ciliate Trimyema compres-sum. Arch. Microbiol. 153:187-192.

11. Hungate, R. E. 1966. The rumen and its microbes. AcademicPress, Inc., New York.

12. Hutten, T. J., M. H. de Jong, B. P. H. Peeters, C. van der Drift,and G. D. Vogels. 1981. Coenzyme M (2-mercaptoethanesulfo-nic acid) derivatives and their effects on methane productionfrom carbon dioxide and methanol by cell-free extracts ofMethanosarcina barkeri. J. Bacteriol. 145:27-34.

13. Kirby, H. 1941. Organisms living on and in protozoa, p. 1009-1113. In G. N. Calkins and F. M. Summers (ed.), Protozoa inbiological research. Columbia University Press, New York.

14. Lee, M. J., P. J. Schreurs, A. C. Messer, and S. H. Zinder. 1987.Association of methanogenic bacteria with flagellated protozoafrom a termite hindgut. Curr. Microbiol. 15:337-341.

15. Mohr, M., A. Ruthmann, K. Eichenlaub-Ritter, S. Kuhn, and P.Traub. 1990. Evidence for intermediate-like filaments in aheterotrichous ciliate. Eur. J. Protistol. 25:255-263.

16. Stumm, C. K., H. J. Gijzen, and G. D. Vogels. 1982. Associationof methanogenic bacteria with ovine rumen ciliates. Br. J. Nutr.47:95-99.

17. Stumm, C. K., and K. B. Zwart. 1986. Symbiosis of protozoawith hydrogen-utilizing methanogens. Microbiol. Sci. 3:10-105.

18. To, L. P., L. Margulis, D. Chase, and W. L. Nutting. 1980. Thesymbiotic microbial community of the Sonoran desert termitePterotermes occidentis. BioSystems 13:109-137.

19. van Bruggen, J. J. A., C. K. Stumm, and G. D. Vogels. 1983.Symbiosis of methanogenic bacteria and sapropelic protozoa.Arch. Microbiol. 136:89-95.

20. van Bruggen, J. J. A., K. B. Zwart, J. G. F. Hermans, E. M.van Hove, C. K. Stumm, and G. D. Vogels. 1986. Isolation andcharacterization of Methanoplanus endosymbiosus sp. nov., anendosymbiont of the marine sapropelic ciliate Metopus contor-tus. Arch. Microbiol. 144:20-23.

21. Vogels, G. D., W. F. Hoppe, and C. K. Stumm. 1980. Associa-tion of methanogenic bacteria with rumen ciliates. Appi. Envi-ron. Microbiol. 40:608-612.

22. Zwart, K. B., N. K. Goosen, M. W. van Schijndel, C. A. M.Broers, C. K. Stumm, and G. D. Vogels. 1988. Cytochemicallocalization of hydrogenase activity in the anaerobic protozoaTrichomonas vaginalis, Plagiopyla nasuta and Trimyema com-pressum. J. Gen. Microbiol. 134:2165-2170.

APPL. ENVIRON. MICROBIOL.

1

c

I

on March 27, 2021 by guest

http://aem.asm

.org/D

ownloaded from