Role of Bacteroides uniformis in Susceptibility of Ta:CF # 1 Mice to Infection by Treponema hyodysenteriae

Zbl. Bakt. 274, 118-125 (1990 ) © Gustav Fischer Verlag, StuttgartlNew York

Role of Bacteroides uniformis in Susceptibility of Ta:CF# 1 Mice to Infection by Treponema hyodysenteriae TOSHIKATSU HAYASHI l , ITARU SUENAGA l , TOMOHIKO KOMEDA l , and TOSHIYUKI YAMAZAKI l 1 1

Animal Health Research Laboratories, Agro Division and Central Research Division, Takeda Chemical Industries, Ltd., 17-85, Ju so-honmachi, 2chome, Yodog awa-ku, Osaka 532, Japan

Received January 4, 1990 . Accepted in revised form April 3, 1990

Summary Among the various strains and colonies of mice tested specific-pathogen-free (SPF) Ta :CF#l mice exhibited the highest suscept ibility to Treponema hyodysenteriae infection. Staphylococcus epidermidis 1047, Streptococcus faecalis 2039 , Lactobacillus murinus 3010, Escherichia coli 1049, and Bacteroides thetaiotaomicron 3008 were administered to the Ta :CF#l mice. B. uniformis IK, which appeared to colonize in the breeding facility under SPF conditions, was isolated from the cecal content of a Ta :CF#l mouse and identified in the mouse flora . Germfree mice were infected with these six bacteria and challenged with T. hyodysenteriae Dj70P3 to elucidate the causal relationship between these bacteria and the sensitivity of mice to T. hyodysenteriae infection . No cecal lesions were observed in gnotobiotic mice eithe r monoassociated with T. hyodysenteriae Dj70P3 or B. uniformis IK or inoculated with T. byodysexteriae Dj70P3 combined with the 5 a.m . strains. Gross cecal lesions appeared onl y in mice inoculated with T. hyodysenteriae Dj70P3 combined with B. uniformis IK. A new SPF mouse colony (CF# l/b Ta) estab lished from gnotobiotic mice by inoculation with B. uniformis IK and the 5 laboratory stock cultures was as susceptible to T. hyodysenteriae as the origin al Ta :CF#l mice. This suggests that B. uniformis IK plays an important role in the high susceptibility of Ta :CF#l mice to T. hyodysenteriae.

Zusammenfassung Von verschiedenen gepriiften Mausestamrnen und -kolonien zeigten Ta:CF#l-SPFMause die hochste Empfanglichkeit fiir eine Treponema hyodysenteriae-Infektion. Den Ta :CF#l-SPF-Mausen wurden Staphylococcus epidermidis 1047 , Streptococcus faecalis 2039, Lactobacillusmurinus 3010, Escherichia coli 1049 und Bacteroides thetaiotaomicron 3008 verabreicht. B. uniformis IK schien in der Tierzuchtanlage unter SPF-Bedingungen die Tiere zu besiedeln ; es wurde aus dem Zokuminhalt einer TA:CF#l-Maus isoliert und inder Mausflora identifiziert. Keimfreie Manse wurden mit diesen sechs Erregern infiziert und im

Treponemal Infection in Ta:CF#l Mice

119

Schutzversuch mit T. hyodysenteriae Dj70P3 gepriift, urn die Kausalbeziehung zwischen diesen Bakterien und der Empfanglichkeit der Mause fur eine T. hyodysenteriae-Infektion zu klaren. Weder in allein mit T. hyodysenteriae Dj70P3 oder B. uniformis IK assoziierten noch in mit T. hyodysenteriae Dj70P3 in Verbindung mit den o.g. 5 Laborstammen geirnpften gnorobiotischen Mausen wurden zokale Lasionen beobachtet, diese traten nur bei mit T. hyodysenteriae Dj70P3 in Verbindung mit B. uniformis IK geimpften Mausen deutlich auf. Eine aus gnotobiotischen Mausen durch lmpfung mit B. uniformis IK und den 5 Laboratoriumsstammen neu gebildete Mausekolonie (CF# l /b Ta) war ebenso empfanglich fiir T. hyodysenteriae wie die urspriinglichen Ta :CF#l-Mause. Dies weist auf eine wichtige Rolle von B. uniformis IK bei der hohen Empfanglichkeit von Ta :CF#l-Mausen gegen T. hyodysenteriae hin,

Introduction

Treponema hyodysenteriae is the etiologic agent of swine dysentery which is characterized by mucohemorrhagic diarrhea (3, 18). Pure cultures of T. hyodysenteriae produce swine dysentery in conventional pigs (1,3,18) but not in germfree pigs (1,9, 10). Lesions typical of swine dysentery were observed in gnotobiotic pigs orally inoculated with T. hyodysenteriae combined with Bacteroides uulgatus, with B. vulgatus and Fusobacterium necrophorum, or with Listeria denitrificans (2, 19). These findings led to the concept that synergy of T. hyodysenteriae and certain other bacteria is necessary for the expression of the pathogenicity of T. hyodysenteraie (10). Experimental infection with T. hyodysenteriae in mice was first reported by Joens and Glock (5) using CFl mice. Bacterial synergism has also been demonstrated in gnotobiotic mice inoculated with T. hyodysenteriae combined with B. vulgatus (6) . Itoh et al. (4) found that the composition of the fecal flora in mice from the same strain from different breeding colonies differed, whereas the composition among different mouse strains in the same breeding colony was similar . We reported previously that the susceptibility of mice to T. hyodysenteriae infection differed among breeding colonies and among mouse strains; Ta:CF#1 mice were the most susceptible to the infection among the mouse strains tested, while K:CF#1 mice belonging to the same strain as Ta :CF#1 mice but produced in a different breeding colony and Slc:ICR mice showed little or no susceptibility to T. hyodysenteriae (16). The se findings prompted us to investigate the relationship between the susceptibility to T. hyodysenteriae and the intestinal flora in mice under the hypothesis of the existence of eliminating organisms or bacteria which act synergistically with T. hyodysenteriae. We recently reported the existence of eliminating organisms for T. hyodysenteriae in the cecal content of Slc:ICR mice (17). The purpose of the present study, in contrast to the previous studies, has been to determine which organisms in the intestinal flora of Ta:CF#1 mice are associated with the high susceptibility of these mice to infection with T. hyodysenteriae. Materials and Methods

Mice. Five-week-old female Ta:CF#1 mice (22) were used. This inbred strain, originally associated with 5 stock cultures (described under microoganisms) is maintained under the specific-pathogen-free (SPF) conditions of the Laboratory Animal Unit, Takeda Chemical Industries, Ltd., Takatsuki, Japan . Germfree mice were obtained by Cesarean section from Ta :CF#1 mice. Gnotobiotic mice were prepared by administering bacteria to the germfree

120

T. Hayashi, I. Suenaga, T. Komeda, and T. Yamazaki

mice in their drinking water. The mice were housed in groups of 5 per cage in a vinyl isolator and were given a sterilized pelleted ration and water ad libitum. Newly established SPF mice, named CF#l/b Ta (CF#l/b), were prepared from the gnotobiotic mice by inoculation with B. uniformis IK and the 5 stock cultures. CF#l/b mice have been produced for more than 8 months since becoming established under controlled SPF conditions. Microorganisms. T. hyodysenteriae Dj70P3, known to be pathogenic in mice (16) was used. The 5 stock cultures, Staphylococcus epidermidis 1047, Streptococcus faecalis 2037, Lactobacillus murinus 3010, Escherichia coli 1049, and B. thetaiotaomicron 3008 were used. These bacteria were originally isolated from an SPF mouse as dominant fecal flora and have been kept in our labo ratory as stock cultures. When an SPF mouse is bacteriologically prepared from a germfree mouse in our Laboratory Animal Unit, the 5 stock cultures are used. B. vulgatus B55-41, an isolate from normal pig feces was kindly supplied by Dr. T. Mitsuoka, The Institute of Physical and Chemical Research, Wako, japan. B. uniformis IK was newly isolated from the cecum of a Ta :CF#l mouse. This organism was distinguishable from B. thetaiotaomicron as it formed smaller colonies on the surface of NBGT blood medium, a selective medium for isolation of Bacteroides (11). B. uniformis IK was subjected to the following reactio ns after growing for 7 days in GAM semisolid without dextrose (Nissui, Tokyo, japan) for identification along with B. uniformis ATCC 8492, B. thetaiotaomicron ATCC 29741 , and B. uulgatus ATCC 8482. Positive results were found in B. uniformis IK for : growth in 20 % bile, esculin hydrolysis, indole production, arabinose, glucose, inulin, maltose, mannose, rhamnose, salicin, sucrose, and xylose. Negative results were found for : catalase production, inositol, mannitol, and trehalose. These results and those for B. uniformis ATCC 8492 were found to be the same. The identification of the strain was also confirmed by gas chromatography (5890A; Hewlett-Packard Co., Palo Alto, CA, USA) and gas chromatography-mass spectrometry (5790A; Hewlett-Packard and jMS DX-300 ; JEOL Ltd., Tokyo, japan) after growth for 7 days in GAM semisolid without dextrose (Nissui) supplemented with 1% glucose. The retention times for the peaks of B. uniformis IK were identical with those obtained for B. uniformis ATCC 8492 (data not shown). Preparation of inocula. T. hyodysenteriae was grown in Trypticase soy broth (TSB; BBL Microbiology Systems, Cockeysville, MD, USA) supplemented with fetal calf serum (10%, vol/vol) (8) in a GasPak jar (BBL) at 3rc for 48 h or in Trpytica se soy agar (TSA; BBL) supplemented with 5% defibrinated horse blood in a GasPak jar at 37°C for 4 days. TSB culture was prepared for the inoculum in germfree and gnotobiotic mice to avoid contami nation. TSA culture was used for the inoculum for SPF mice using the method described previously (15). S. epidermidis 1047, S. faecalis 2037, and E. coli 1049 were grown in Brain heart infusion (Difco Laboratories, Detroit, MI, USA) at 3rc for 24h. B. uniformis IK, B. thetaiotaomicron 3008 , B. vulgatus B55-41, and L. murinus 3010 were grown in GAM semisolid (Nissui) at 37 °C for 48 h. Quantitative cultures. T. hyodysenteriae was isolated from the cecum of mice. The ceca were homogenized in a sterile diluent (4.5 g KHzP0 4 , 6.0 g Na zHP0 4 , 0.5 g L-cysteine HCl · HzO, 0.5 g Tween 80, 1.0 g agar, and 1000 ml distilled water, pH 7.0) (11). Serial dilutions of the homogenates were plated on the surface of TSA supplemented with 5% defibrinated horse blood and 400 ug/rnl of spectinomycin (14) and incubated under an atmosphere of 5% Hz, 25 % COz, and 70% N z at 42 °C for 72 h in an anaerobic glove box (Tabai Espec Co., Osaka, japan). The population and colonization of bacteria, except T. hyodysenteriae, in gnotobiotic mice, were determined by incubating the serially diluted feces; S. epidermidis was isolated on Staphylococcus medium 110 (Eiken, Tokyo, japan), S. faecalis was isolated on EF agar (Nissui), L. murinus was isolated on LBS agar (BBL), E. coli was isolated on DHL agar (Eiken), and B. uniformis, B. thetaiotaomicron, and B. vulgatus were isolated on Bacteroides agar (Nissui). Total aerobic bacteria were counted on TSA supplemented with 5% defibrinated horse blood. Total anaerobic bacteria were counted on GAM agar (Nissui). The isolated bacteria were identified by standard procedures (7, 13). Contaminant monitoring. Contamination in germfree mice was monitored by bacteriolo-

Trepon emal Infection in Ta:CF#l Mice

121

gically examining the feces. The samples were incubated in TGC medium fluid (Nissui) aerobica lly at 37°C, GAM bro th (Nissui) anaerobically at 3rC, and Bacto pota to dextrose bro th (Difco) aerobically at room temperature, each for 7 days. Animal inoculation. The culture broth of the 5 stock cultures, B. uniformis IK, B. thetaiotaomicron 3008, or B. vulgatus B55-41 was introduced into the germfree isolators and was admin istered to the 3-week-old germfree mice ad libitum by a water bottle. After colonization with the bacteria had been confirmed, the 5-week-old gnotobiotic mice were orally inoculated with 0.5 ml of the culture bro th of T. hyodysenteriae Dj70P3 equivalent to approximately 107 CFU, using a blunt-ended metal catheter. Germfree mice were also inoculated with the culture brot h of T. hyodysenteriae Dj70P3 by the same method . Ta:CF#l and CF#l!b mice were inoculated with 0.5 ml of a 50% homogena te of the TSA culture of T. hyodysenteriae Dj70P3 at population levels of approximately 10 7 CFU (15). Necropsy. The mice were killed and sacrificed 10 days after inoculation with T. hyodysenteriae. Gross cecal lesions were examined , and then the cecum was prepared to culture T. hyodysenteriae. Portions of the cecum were fixed in buffered 10% formalin and stained with hematoxylin and eosin for light microscopy. Other portions were fixed in 1% glutaraldehyde in 0.067 M phosphate buffer solut ion (pH 7.0) for scanning electron microscopy accordi ng to the method described by Nakao et al. (12).

Results Isolation of T. hyodysenteriae T . hyodysenteriae was reisolated at levels of approximately 10 7_10 8 CFU/g from the cecum of all the mice challenged with the organism (Tables 1 and 2). The bac terial

Table 1. Gross lesions and colonization by T. hyodysenteriae in the cecum of gnotobiotic and germfree mice" No. of mice positive I No. of mice examined Gro up Organism(s) used for inoculation of germfree mice

Challenge with T.hyodysent eriae

Gross cecal lesion

Colonization with T. hyo dysenteriae

(log CFU/g)

1. 2. 3.

Yes No Yes

19/30 0/6 0/10

30/30 ND C 10/10

(8.7 ± 0.7)b

Yes Yes Yes No

0/10 0/10 0120 0/3

10/10 10/10 20120 ND

(7.S ± 0.7) (S.2 ± 0.3) (S.4 ± 0.6)

4.

5. 6. 7.

B. uniformis IK B. uniformis IK S. epidermidis 1047, S. faecalis 2037, L. murinus 3010 E. coli 1049 , B. theta iotaomicron 3008 B. thetaiotaomicron 300S B. vulgatus B55-41 None (germfree) None (germfree)

a

(6.9 ± 0.9)

Mice were inoculated orall y with 107 CFU of T. hyodysenteriae Dj70P3 and necropsied 10 days after inoculation. b Mean ± SD C Not determind

9 Zbl. Bakt. 274/1

122

T. Hayashi, I. Suenaga, T. Komeda, and T. Yamazaki

Table 2. Gross lesions and colonization with T. hyodysenteriae in the cecum of specificpathogen-free colonies of Ta:CF#l and CF#l /b Ta mice" No. of mice positive 1No. of mice examined Mice

Ta:CF#l

Gross cecal Colonization lesion with T. hyodysenteriae

(log CFU/g)

40/40

40/40

(8A ± OA)b

10/10 10/10

10/10 10/10

(8A ± 0.1) (8.6 ± 0.2)

c

CF#lIb Ta 3 months 8 months a, C

See Table I New SPF mice were prepared from gnotobiotic mice by inoculation with B. uniformis IK and the 5 laboratory stock cultures. Mice were examined 3 and 8 months after the colony had been established.

b

levels, however, were relatively low in gnotobiotic mice challenged with T . hyodysenteriae afte r having been inocula td with the 5 stock cultures .

Isolation of other organisms No bacteria were isolated from the feces of germfree mice. In gnotobiotic mice inoc ulated with the 5 stock cultures, about 105 CFU/g of S. epidermidis, 10 8 CFU/g of S. [aecalis, 10 9 CFU/g of L. murinus, 10 8 CFU/g of E. coli, and 10 9 CFU/g of B. thetaiotaomicron were reisolated from the feces before the T. hyodysenteriae challenge. B. thetaiotaomicron (10 9 CFU/g), B. vulgatus (1011 CFU/g), or B. uniformis (108_ 1010 CFU/g) was reisolated from the feces of the respective monoassociated gnotobiotic mice. B. uniformis was isolated from the feces of the CF#1/b mice 3 months after the colony had been estab lished, at levels of 10 8 to 10 9 CFU/g.

Gross cecal lesions Gross cecal lesions were observed in all the Ta:CF#l and CF#1/b mice challenged with T. hyodysenteriae and in 19 of the 30 gnotobiotic mice inocu lated with B. uniformis IK and challenged with T. hyodysenteriae, but no gross lesions were observed in the cecum of any other experimental group of gnotobiotic or germfree mice (Tables 1 and 2). The macroscopic lesions were large quantities of mucous exudate, edema, catarrhal inflamation, and gas as observed previo usly (5, 16).

Light microscopy Microscopic cecal lesions were observed in Ta :CF#l and CF#lIb mice inoculated with T. hyodysenteriae and gnotobiotic mice inocula ted with T. hyodysenteriae and B. uniformis IK. The cecal lesions were characterized by epithelial erosion, mucosal edema , cryp t dilatation, and increases in the number of goblet cells as demonstra ted previously (5, 16). No cecal lesions were found in the other experimental groups of rmce,

Treponemal Infection in Ta:CF#l Mice

123

Scanning electron microscopy Scanning electron micrographs of the mucosal surface of the Ta:CF#1 and CF#1/b mice revealed an absence lesions in the cecum of noninfected control mice, but dilated crypts, mucus, fibrin, and disrupted and sloughing epithelial cells were observed in the cecum of these mice infected with T. hyodysenteriae as observed previously (16). Mice inoculated with B. uniformis IK alone did not have any lesions, while lesions similar to those seen in the Ta:CF#1 and CF#1/b mice challenged with T. hyodysenteriae were observed in the mice inoculated with B. uniformis IK and challenged with T . hyodysenteriae. The other groups of gnotobiotic and germfree mice had no lesions in the cecum.

Discussion The present study has demonstrated that B. uniformis IK isolated from a Ta:CF#1 mouse and T. hyodysenteriae show a synergistic pathogenic effect in gnotobiotic mice. B. uniformis IK was not one of the 5 stock cultures used as inocula to prepare the Ta:CF#1 mice or other mouse strains from germfree mice in our Laboratory Animal Unit. It seems that B. uniformis IK was derived from a breeding environment source introduced into the laboratory and colonized the cecum of the Ta :CF#1 mice. The synergistic effect of B. uniformis IK and T. hyodysenteriae was confirmed not only in gnotobiotic mice but also in the newly established inbred SPF colony of CF#1/b mice. CF#1/b mice inoculated with B. uniformis IK and the 5 stock cultures were highly susceptible to T . hyodysenteriae 3 months after the colony had been established. When Ta:CF#1 mice were prepared several years ago in the laboratory, their susceptibility to T . hyodysenteriae was low at the beginning of breeding. However, occasional monitoring indicated that the susceptibility gradually increased. After a year the Ta :CF#1 mice acquiredy a stable high susceptibility to T. hyodysenteriae (data not shown ). This sequence indicate s that it takes a long time for B. uniformis to stably colonize the cecum of Ta:CF#1 mice. Whipp et al. (19) reported that when T . hyodysenteriae was present at concentrations greater than 10 6 viable organisms /crrr' of colonic mucosa, signs and lesions characteristic of swine dysentery can be observed in gnotobiotic pigs, and it appears that an important aspect of th synergism is the estabishment of conditions conductive to colonization of the colonic mucosa by T . hyodysenteriae in large number s. How ever, in the present study, T. hyodysenteriae alone colonized the cecum of germfree mice at concentrations of 108 CFU/g, and no cecal lesions were observed. The number of T . hyodysenteriae was the same in gnotobiotic mice inoculated with B. uniformis IK in which lesions were observed . Therefore, the synergistic effect is not simply related to the population of T. hyodysenteriae. Whipp et al. also described that B. vulgatus of either porcine or human origin appears to interact synergistically with T. hyodysenteriae in the pathogenesis of colitis in gnotobiotic pigs; this prerequisite is relatively nonspecific in the sense that it can be fulfilled by anyone of several organisms capable of colonizing the colon (19). Joens et al. (6) have reported that gross cecal lesions occur in gnotobiotic mice inoculated with T. hyodysenteriae and B. vulgatus of porcine origin. However, in the present study no cecal lesions were observed in gnotobiotic mice inoculated with T. hyodysenteriae DJ70P3 combined with B. thetaiotaomicron 3008 or B. vulgatus 55-41, although T. hyodysenteriae colonized the cecum in large numbers. These observations suggest that

124

T. Hayashi, I. Suenaga, T. Komeda, and T. Yamazaki

only a certain restricted strain of B. vulgatus or other species of Bacteroides, e.g, B. uniformis IK, are synergistic with T. hyodysenteriae in mice. In the present study, gross cecal lesions were not observed in all gnotobiotic mice inoculated with B. uniformis IK and T. hyodysenteriae DJ70P3. Therefore, it is suggested that some other synergistic organisms are required for the production of cecal lesions in these mice .

Acknowledgments. We thank Dr. A. lmai and Mr. M. Matsui for helpful advice and for supporting this study and many of our colleagues for their technical assistance. We are also grateful to Dr. A. Shino for assistance with the histopathologic examinations and to Dr. K. Ueno, Institute of Anaerobic Bacteriology, Gifu University, School of Medicine, for technical advice in identifying B. uniformis IK. References

1. Brandenburg, A. c, O. P. Miniats, H. D. Geissinger, and E. Ewert: Swine dysentery : inoculation of gnotobiotic pigs with Treponema hyodysenteriae and Vibrio coli and a Peptostreptococcus. Can. J. Comp oMed. 41 (1977) 294-301 2. Harris, D. L., T.]. L. Alexander, S. C. Whipp, I. M. Robinson, R. D. Glock, and P. j. Matthews: Swine dysentery: studies of gnotobiotic pigs inoculated with Treponema hyodysenteriae, Bacteroides uulgatus, and Fusobacterium necrophorum. J. Am. Vet. Med . Ass. 172 (1978) 468-471 3. Harris, D. L., R. D. Glock, C. R. Christensen, and]. M. Kinyon: Swine dysentery. 1. Inoculation of pigs with Treponema hyodysenteriae (new species) and reproduction of the disease. Vet. Med. Small Anim. Clin. 67 (1972) 61-64 4. Itoh, K., T. Mitsuoka, K. Sudo, and K. Suzuki: Comparison of fecal flora of mice based upon different stra ins and different housing conditions . Z. Versuchstierk . 25 (1983) 135-146 5. joens, L. A. and R. D. Glock: Experimental infection in mice with Treponema hyodysenteriae. Infect. Immun. 25 (1979) 757-760 6. joens, L. A., I. M. Robinson, R. D. Glock, and P.]. Matthews: Production of lesions in gnotobiotic mice by inoculation with Treponema hyodysenteriae. Infect. Immun. 31 (1981) 504-506 7. Krieg, N. R. and]. G. Holt (eds.): Bergey's manua l of systematic bacteriolog y, Vol. 1. Williams and Wilkins Co., Baltimore (1984) 8. Lemcke, R. M.,]. Bew, M. R. Burrows, and R.]. Lysons: The growth of Treponema hyodysenteriae and other porcine intestinal spirochaetes in a liquid medium. Res. Vet. Sci. 26 (1979) 315-319 9. Meyer, R. c., ]. Simon, and C. S. Byerly: The etiology of swine dysentery. 1. Oral inoculation of germ-free swine with Treponema hyodysenteriae and Vibrio coli. Vet. Path. 11 (1974) 515-526 10. Meyer, R. C, J. Simon, and C. S. Byerly: The etiology of swine dysentery. II. Effect of a known microbial flora, weaning and diet on disease production in gnotobiotic and conventional swine. Vet. Path. 11 (1974) 527-534 11. Mitsuoka, T., T. Sega und S. Yamamoto: Eine verbesserte Methodik der qualitativen und quantitativen Analyse der Darmflora von Menschen und Tieren. Zbl. Bakt. Hyg., 1. Abt, Orig . A 195 (1965) 455-469 12. Nakao, M., M. Kondo, and K. Tsuchiya: Light and electron microscop y of the morphological response of Escherichia coli and Serratia marcescens to cefmenoxime (SCE1365), a new broad spectrum cephalosporin. ] . Antibiot . 34 (1981) 1046-1054 13. Sneath, P. H. A., N. S. Mair, M. E. Sharpe, and]. G. Holt (eds.): Bergey's mamual of systematic bacteriology, Vol. 2. Williams and Wilkins Co., Baltimore (1986)

Treponemal Infection in Ta:CF#l Mice

125

14. Songer, J. G., J. M. Kinyon, and D. L. Harris: Selective medium for isolation of Treponema hyodysenteriae . J. Clin. Microbiol. 4 (1976) 57-60 15. Suenaga, I. and T. Yamazaki: Experimental infection with Treponema hyodysenteriae in nude mice. Zbl. Bakt. Hyg., I. Abt. Orig. A 254 (1983) 528-533 16. Suenaga, I. and T. Yamazaki : Experimental Treponema hyodysenteriae infection of mice. Zbl. Bakt. Hyg., I. Abr, Orig. A 257 (1984) 348-356 17. Suenaga, 1. and T. Yamazaki: Eliminating organisms against Treponema hyodysenteriae in the gut of mice. Zbl. Bakt. Hyg., I. Abt. Orig. A 261 (1986) 322-329 18. Taylor, D. J. and T. J. L. Alexander: The production of dysentery in swine by feeding cultures containing a spirochaete . Br. Vet. ]. 127 (1971) 58-61 19. Whipp , S. c., I. M. Robinson, D. L. Harris, R. D. Glock, P. J. Matthews, and T. J. L. Alexander: Pathogenic synergism between Treponema hyodysenteriae and other selected anaerobes in gnotobiotic pig. Infect. Immun. 26 (1979) 1042-1047 Dr. Toshikatsu Hayashi, Animal Health Research Laboratories, Agro Division, Takeda Chemical Industr ies, Ltd., 17-85, Juso-honmachi, 2-chome, Yodogawa-ku, Osaka 532, Japan