[Retracted] R plasmid in Escherichia coli O103 coding for colonization of the rabbit intestinal tract.
7
0
0
Texte intégral
(2) Vol. 59, No. 6. INFECTION AND IMMUNITY, June 1991, p. 1888-1892. 0019-9567/91/0618880-05$02.00/0 Copyright C 1991, American Society for Microbiology. R Plasmid in Escherichia coli 0103 Coding for Colonization of the Rabbit Intestinal Tract ALAIN REYNAUD,l* MICHEL FEDERIGHI,1 DOMINIQUE LICOIS,2 JEAN-FRANCOIS GUILLOT,3 AND BERNARD JOLY4 Laboratoire d'Analyses Wte'rinaires et Biologiques du Departement du Puy de D6me, RN 89-BP 42, 63000 ClermontFerrand,' Unite de Pathologie du Lapin, INRA de Tours Nouzilly, 37380 Monnaie,2 Universite de Tours, 37000 Tours,3 and Service de Bacteriologie, Faculte de Pharmacie, 63001 Clermont-Ferrand,4 France Received 16 October 1990/Accepted 6 March 1991. T C. A R T E. of the rabbit by the GV strain was related to the presence of this R plasmid.. In 1977, Cantey and Blake (6) were the first to isolate from a rabbit a noninvasive strain of Escherichia coli (RDEC-1) capable of causing severe diarrhea in this animal species without producing heat-labile or heat-stable enterotoxins. Takeuchi et al. showed that the RDEC-1 strain was able to colonize the ileum, cecum, and colon and to adhere to epithelial cells (30). Wolf et al. showed that RDEC-1 possessed a plasmid which mediated the expression of an adhesion factor named AF/Rl, which mediates the interaction of bacteria with normal epithelial cells (35). This strain, belonging to serogroup 015, is therefore akin to the enteropathogenic E. coli isolated from humans (26, 31, 33) or from other species of animals (17, 19). In Belgium, Peeters et al. have also isolated strains which are pathogenic for rabbits after weaning and have the same characteristics as the RDEC-1 strain (20-22). In France, E. coli strains belonging to another serogroup (0103) predominate in epidemic outbreaks of diarrhea in rabbits (3, 5, 23). Recently, Milon et al. (16) found that epidemic and highly pathogenic rhamnose-negative 0103 E. coli attached to rabbit intestinal villi and gave a diffuse adhesion pattern with HeLa cells. We have isolated 60 strains of 0103 E. coli from the cecal contents of diarrheic, weaned rabbits, and we demonstrated that 55 of these strains harbored a plasmid of 117 kb (24). Among them, one strain (GV) possessed such a plasmid which was self-transferable and coded for resistance to several antibiotics. Preliminary observations suggested that this plasmid played a role in colonization of the rabbit intestinal tract and thus contributed to the virulence of this isolate. The aim of the present work was to check if colonization. Bacterial strains. Three independently isolated wild-type strains of E. coli (GV, LY265, and 6100) and one E. coli K-12 strain were used. Strain GV (0103:H2:K-) from a farm on which 20 to 30% of the animals had died of diarrhea after weaning was isolated by us. This strain is resistant to several antibiotics (neomycin, streptomycin, sulphonamides, and oxytetracycline) and does not ferment rhamnose (Rha-). Strain LY265 (0103:H2:K-) was isolated by L. Renault from diarrheic rabbits and is Rha-. Strain 6100 (0?:H2:K-), from a farm without pathological problems, is sensitive to antibiotics and ferments rhamnose (Rha+). It does not possess any plasmid. Strain BM21, a K-12 isolate is nonagglutinable by anti-0103 serum. It does not possess any plasmid and is resistant to nalidixic acid (150 ,ug/ml) by chromosomal mutation (gyrA). Sensitivity to antibiotics. For all the antibiotics tested, an agar diffusion method (Diagnostics Pasteur, Marnes la Coquette, France) was used. Study of plasmids. Extraction of plasmid DNA was performed by the method of Kado and Liu (14), and electrophoretic migration was carried out in 0.7% agarose gels. Plasmid curing was achieved at 42°C in the presence of ethidium bromide by the method described by Bouanchaud et al. (2). Plasmid transfer was carried out by conjugation on agar (34). Toxin assay. A heat-labile enterotoxin assay was performed with a ligated rabbit intestinal loop (32) and the Biolyon kit (Oxoid, Basingstoke, Hampshire, England). A heat-stable enterotoxin assay was performed by inoculation of newborn mice (11).. R *. MATERIALS AND METHODS. Corresponding author. 1888. Downloaded from http://iai.asm.org/ on August 22, 2017 by INRA - old. D E. One rabbit pathogenic Escherichia coli strain, belonging to serogroup 0103, harbors a self-transferable 117-kb plasmid (pREC-1) encoding resistance to several antibiotics. The role of this R plasmid in the colonization of the digestive tract in specific-pathogen-free (E. coli 0103-free) rabbits was studied. Five-weekold rabbits were inoculated with the wild-type strain, with its variant cured of the plasmid, with an E. coli K-12 strain, or with an untypeable E. coli strain from a healthy rabbit. No symptoms and no mortality were observed in animals inoculated with strains without the plasmid pREC-1, but 87.5% of the rabbits infected by the wild strain died, generally with bloody diarrhea, between days 5 and 15 postinfection. The weight gain of animals was strongly reduced. Transfer of the plasmid to the cured strain or to nonvirulent strains led these strains to induce the same pathology but with a lower mortality. Colonization of the gut by the 0103 strain and symptoms of bloody diarrhea are thus related to the presence of the pREC-1 plasmid. The GV strain, which does not produce classical heat-labile enterotoxin or heat-stable enterotoxin and is not invasive, could be considered an enteropathagenic E. coli-like strain. The presence of a conjugative plasmid such as pREC-1 encoding both antibiotic resistance and virulence determinants in 0103 E. coli from rabbits could represent a prominent epidemiological hazard under selective pressure by antibiotic therapy..
(3) R PLASMID IN E. COLI 0103. VOL. 59, 1991. 1. 2. RESULTS Characteristics of the GV strain. Serotype 0103 E. coli GV harbors a single plasmid, pREC-1, with a molecular size of 117 kb (Fig. 1). It produces neither a thermolabile nor a thermostable enterotoxin and is not invasive. Following treatment with ethidium bromide, the strain lost the pREC-1 plasmid (strain GVc) and became sensitive to streptomycin, neomycin, oxytetracycline, and sulfonamides. Transferability of the pREC-1 plasmid. The pREC-1 plasmid was transferred by conjugation to E. coli 6100, BM21, and GVc (Fig. 1). The sensitivity tests carried out on all the 6100/pREC-1, BM21/pREC-1, and GVc/pREC-1 transconjugants showed that the antibiotic resistance was transferred. together.. 7. T C. A R T E. R. 3 4 5 6. 8. 9. 10. D E. FIG. 1. Agarose gel electrophoresis of strain GV (lane 1); strains GVc, BM21, and 6100 (lanes 2, 4, and 6) and their transconjugants (lanes 3, 5, and 7); and reference plasmids of 54, 94.5, and 158 kb (lanes 8 to 10).. Animal experiments. (i) Clinical symptoms and mortality. None of the rabbits inoculated with the plasmid-free strains (6100, BM21, and GVc) developed diarrhea and died, whereas animals inoculated with plasmid-bearing strains had severe diarrhea and died (87.5, 25, 12.5, and 25% deaths and 87.5, 25, 50, and 75% diarrhea attack rates for strains GV, 6100/pREC-1, BM21/pREC-1, and GVc/pREC-1, respectively). Dead animals were observed on day 5 for strain GV and days 9 to 11 for the other strains. All animals that died showed diarrhea generally of the hemorrhagic type. (ii) Weight gain. The mean daily weight gain of the rabbits inoculated with strains 6100, BM21, and GVc was approximately 30 g. Animals inoculated with the strains harboring the pREC-1 plasmid (GV, 6100/pREC-1, BM21/pREC-1, and GVc/pREC-1) showed depressed growth for 10 days. Afterwards, the surviving animals tended to recover a normal daily weight gain, but in those inoculated with the wild-type GV strain, recovery did not take place except in the case of one surviving rabbit, the growth of which was only slightly depressed during the entire experimental period. (iii) Numeration of E. coli excreted by inoculated animals. The bacterial counts of resident E. coli done before inoculation showed that contents of the excreta were about 105 bacteria per g of feces. These E. coli strains, which were present in the guts of rabbits before the experimental inoculations, did not belong to serogroup 0103. Results of the counts performed after inoculation (Fig. 2) showed that the rabbits inoculated with strains GV and BM21/pREC-1 excreted these strains very rapidly at a concentration of 108 to 109 bacterial cells per g of feces. Similar values were. Downloaded from http://iai.asm.org/ on August 22, 2017 by INRA - old. Invasion assay. The invasive capacity of the GV strain was researched by the Sereny test on guinea pig cornea (27). Animal experiments. Fifty-six specific-pathogen-free rabbits, free of serotype 0103 E. coli (8), originating from the Station de Pathologie Aviaire et de Parasitologie, INRA, Tours, France, were used. Two days before inoculation, the animals were put into 14 cages containing four animals each and maintained in a room in which the environment was permanently controlled (9). They received ad libitum a laboratory food free of antibiotics and of anticoccidial agents (UAR 112-91360; UAR, Villemoisson-sur-Orge, France). Bacterial strains were administered to eight 32-day-old rabbits by stomach tubes at a concentration of 107 bacteria per 5 ml per animal. Diarrhea and mortality were recorded daily. The animals were weighed three times a week over a period of 17 days. The samples for bacteriological study were taken immediately before inoculation and then every day until the 14th day after inoculation. Freshly voided feces were collected every day under the cages. The counting of E. coli was carried out by the method of Guillot et al. (12). For each sample, two lactose-positive colonies from the highest dilution were isolated. Complementary tests, including the antibiogram and those for the fermentation of rhamnose (API 20E system; API, la Balme les Grottes Montalieu Vercieu, France) and for agglutination by anti-0103 serum, were carried out with these isolates in order to assess whether they corresponded to the inoculated strains of E. coli. Histological examination. One complementary experiment was carried out by light microscopic (LM) observation. Three rabbits were inoculated with each bacterial strain and sacrificed 6 days postinfection. Segments about 1 cm long were taken from the ileum, fixed, and postfixed by a previously described method (29). Semithin sections (1 ,um) were obtained with a Reichert ultramicrotome, stained with toluidine blue, and examined by LM (Olympus BH2 microscope). Electron microscopic observation of bacterial strains. For negative staining of the strain, the bacterial suspension was resuspended in Penassay broth (Difco Laboratories) (7). Ten microliters of an overnight culture was applied to carboncoated grids for 30 s and was negatively stained with 1% phosphotungstic acid solution (pH, 6.8), and the excess liquid was removed with filter paper. Observation was carried out with a Philips EM400 electron microscope. Adhesion to HEp-2 cells. Adhesion to HEp-2 cells was studied by the method of Cravioto et al. (10). After incubation (3 h at 37°C) of HEp-2 cells on glass coverslips with bacterial cultures in the presence of 0.5% D-mannose (Sigma), cells were washed, fixed in 70% methanol, and stained with Giemsa reagent for LM.. 1889.
(4) 1890 10. REYNAUD ET AL.. INFECT. IMMUN.. COLONY FORMING UNITS /G OF FECES (LOG ) -. 6-. 6 4 2. ;1 =j. 0 0. 2. _ BM21/PREC-1. [711] BM21. 7. (K12). harbouring the R plasmid E. coil without plasmid FIG. 2. Evolution of the number of E. coli excreted by animals inoculated with GV, BM21, and BM21/pREC-1 strains between 1 and 7 days postinoculation. E. coil. obtained with transconjugants 6100/pREC-1 and GVc/ pREC-1. (iv) Histological examination. LM observation revealed a great number of bacteria attached to the epithelial cells of villi and of crypts of the ileum in rabbits inoculated with the GV strain (Fig. 3) or with the transconjugants (data not shown). In all cases, the intestinal villi were shortened. No bacteria in the lining to the mucosa were observed with the E. coli K-12, 6100, or GVc strains. (v) Negative staining of E. coli 0103. After overnight culture in Penassay broth, cells of the GV strain were heavily coated with pili (Fig. 4a). The presence of flagella could also be distinguished. The GV plasmid-cured strain did. T C. A R T E .1. ..11W.'', M. t ..... .: :..:It:... f,:. -.z:-.-.. 'ii:. R. 49,. :. -ie. IL. Ui,. 4p"o. Vig,. VI:. FIG. 3. LM observation of ileal epithelial cells 6 days postinoculation with the GV strain. Numerous E. coli are attached to epithelial cells of villi and of crypts of the ileum. No bacterial cells have invaded the epithelial cells. Magnification, x 1,320.. D E. DISCUSSION The pathogenesis of the 0103 GV strain seems rather similar to that of the RDEC-1 strains (3) or those of other E. coli strains of the 015 serogroup studied by Peeters et al. (21). Nevertheless, the main difference is that diarrhea induced by GV infection is frequently accompanied by severe intestinal hemorrhaging, which has not been reported with the other serogroups. The GV strain harbors a transferable plasmid of about 117 kb, as is commonly observed in E. coli strains of serogroup 0103 isolated from cases of rabbit diarrhea in France (24). Our results show that this plasmid (pREC-1), encoding antibiotic resistance, is involved in the colonization of the intestinal tract of rabbits by E. coli 0103. A variant of the GV strain (GVc), cured of the plasmid, did not induce diarrhea or lethality after inoculation, whereas administration of the parent plasmid-bearing strain was followed by hemorrhagic diarrhea and mortality in experimental animals. The plasmid-cured GV strain demonstrated poor colonization ability (105/g of feces), whereas GV could colonize the intestinal tract with high population sizes (109/g of feces). Moreover, after having acquired the plasmid pREC-1, the nonpathogenic E. coli strains BM21, 6100, and GVc could colonize the gut; their populations 5 days after inoculation were always greater than 108/g of excreta, and symptoms of diarrhea were similar to those obtained in experimental infections with the wild-type GV strain, although they were accompanied by a lower and delayed mortality. The difference in pathogenicity between the wild-type GV strain and the transconjugants which had acquired the pREC-1 plasmid could be due to the presence in the wild-type GV strain of virulence factors encoded by the chromosome in addition to the pREC-1 plasmid. LM examination, experiments with HEp-2 cells, and the Sereny test strongly suggest that GV does not possess any invasive properties. GV does not produce any enterotoxin, but it could produce other toxins, such as Shiga-like toxins, since it has been demonstrated that the RDEC-1 strain synthesizes small amounts of Shigalike toxin type 1 (18). The search for a possible toxin for the GV strain is under way. Other putative virulence factors, in addition to the loss of plasmid pREC-1, could be altered in cured GVc derivatives. Alternatively, the differences in pathogenicity could be due to lower expression of the genes carried by the plasmid pREC-1 in the transconjugants. Histological examination revealed a great number of bacteria attached to the luminal side of enterocytes in rabbits inoculated with the wild-type GV strain, whereas the same strain cured of pREC-1 lost the ability to adhere to intestinal. Downloaded from http://iai.asm.org/ on August 22, 2017 by INRA - old. _ GV. 6. 4 3 6 DAYS POST INOCULATION. 1. not possess any pili (Fig. 4b) but expressed the pili when reconstituted with the plasmid (data not shown). Strains BM21 and 6100 also expressed pili after introduction of plasmid pREC-1. Strain GV demonstrated a slight localized adhesion to HEp-2 cells and was never observed inside the cells. The GV plasmid-cured strain did not attach to HEp-2 cells. (vi) Utilization of rhamnose. In our experience, in agreement with the findings of Camguilhem and Milon (4), all the isolates of rabbit pathogenic 0103 E. coli appear as Rha- in diagnostic tests. However, after loss of plasmid pREC-1, strain GVc exhibited a rhamnose-positive phenotype, which was also observed after curing strain LY265 of a plasmid of similar size. However, the Rha- phenotype was not restored after conjugative transfer of plasmid pREC-1 in the cured GVc and LY265 strains..
(5) VOL. 59, 1991. R PLASMID IN E. COLI 0103. + +. ,i t MP. ... #... ..Z,. ...&. v. A R T E. D E. FIG. 4. Electron microscopic observation of the surfaces GV bacterial strains after phosphotungstic acid staining. (a) Wild-type GV. Flagella (A) and numerous fimbriae (B) can be seen. Magnification, x5O,OOO. (b) GVc cured strain. Only flagella can be observed. Magnification, x35,100.. epithelial cells. The patterns observed were suggestive of the attaching and effacing adherence described by Moon et al. (17). Moreover, we have observed fimbrialike structures on the surface of the GV strain which are not seen on the surface of the GVc cured strain. Since the sequence encoding AF/Rl pili in RDEC-1 has been recently described (36), a specific gene probe could be constructed and used to check whether the same surface proteins promoting attachment of the bacteria to rabbit intestinal brush borders are involved in both the 0103 GV and 015 RDEC-1 strains. After the loss of plasmid pREC-1, strain GVc became able to utilize rhamnose, whereas the parental strain GV was Rha- as has been reported for 0103 pathogenic isolates from rabbits (6). The same effect was observed after curing the 0103 strain LY265 of its 117-kb plasmid. This effect is presumably due to plasmid-encoded proteins either regulating genes involved in the rhamnose pathway or impairing in some other way the bacterial functions required for rhamnose utilization. It should be mentioned in that respect that a permease is required for rhamnose utilization (15). Conjugative plasmids frequently encode membrane proteins (28), which could adversely interact with membrane functions, and Rosas et al. (25) have already demonstrated that envelope modifications due to plasmid loss could be responsible for metabolic differences between wild-type and cured cells. More unexpectedly, the Rha- phenotype was not restored upon reintroduction of plasmid pREC-1 into strain GVc. Moreover, after transfer of the 117-kb plasmid of GV into the LY265c strain, the transconjugants also remained Rha+. This indicates that the persistence of rhamnose utilization after reintroduction of the pREC-1 plasmid was probably not due to a random mutagenic effect of ethidium bromide on the GV strain during the curing process. Whether this effect is due to plasmid sequence modification (e.g., deletion, inver-. R. sion, or insertion of a chromosomal element) during or after conjugational transfer in the cured derivates remains to be investigated. Similarly, reintroduction of pREC-1 either into a cured derivative of GV or LY265 or into E. coli K-12 did not restore completely the colonization ability and the pathogenicity of the wild phenotype. Another factor is likely to be involved in the regulation or expression of bacterial virulence. Moreover, rhamnose fermentation seems to be linked to virulence expression, since the highest colonization and pathogenicity were observed with Rha- E. coli strains (16). The relationship between rhamnose fermentation and virulence remains to be elucidated. Our observations could be compared with those made by Baldini et al. (1) and Moon et al. (17) with enteropathogenic E. coli isolated from humans and from rabbits, in which a plasmid is also involved in colonization (1). Nevertheless, the presence of genes of resistance and virulence on the same extrachromosomal self-transferable structure is very striking although still rare (13). Such a plasmid represents a prominent epidemiological risk, since it can be very easily selected during antibiotic therapy of rabbits. We will now study the spread of this plasmid by means of epidemiological investigations carried out both in France and in other countries. ACKNOWLEDGMENTS We greatly thank J. P. Lafont for the critical review of this manuscript, L. Renault for the gift of strain LY265, and F. and I.. 0rskov for serotyping the rabbit strains. REFERENCES 1. Baldini, M. M., J. B. Kaper, M. M. Levine, D. C. A. Candy, and H. W. Moon. 1983. Plasmid mediated adhesion in enteropathogenic Escherichia coli. J. Pediatr. Gastroenterol. Nutr. 2:534538.. Downloaded from http://iai.asm.org/ on August 22, 2017 by INRA - old. T C. Ad -'. 1891.
(6) 1892. REYNAUD ET AL.. 19. Okerman, L. 1987. Enteric infections caused by non-enterotoxigenic Escherichia coli in animals: occurrence and pathogenicity mechanisms. A review. Vet Microbiol. 14:33-46. 20. Peeters, J. E., G. J. Charlier, and R. Raeymeckers. 1985. Scanning and transmission electron microscopy of attaching effacing Escherichia coli in weanling rabbits. Vet. Pathol. 22: 54-59. 21. Peeters, J. E., R. Geeroms, and V. B. Glorieux. 1984. Experimental Escherichia coli enteropathy in weanling rabbits: clinical manifestations and pathological findings. J. Comp. Pathol. 94: 521-528. 22. Peeters, J. E., P. Pohl, L. Okerman, and L. A. Devriese. 1984. Pathogenic properties of Escherichia coli strains isolated from diarrheic commercial rabbits. J. Clin. Microbiol. 20:34-39. 23. Renault, L., J. Roux, E. Le Bourhis, P. Coudert, D. Licois, and J. F. Guillot. 1983. Description d'un s6rogroupe (0103) d'Escherichia coli enteropathogene chez le lapin au sevrage. Bull. Acad. Vet. France. 56:387-400. 24. Reynaud, A., M. Federighi, B. Joly, P. Hervouet, and R. Camguilhem. 1987. Specificite, antibiosensibilite et virulence des souches d'Escherichia coli isoles dans le caecum des lapins. 1987. Cuni-sciences. 4(1):1-9. 25. Rosas, S. B., A. Calzolari, J. L. La Torre, N. E. Ghittoni, and C. Vasquez. 1983. Involvement of a plasmid in Escherichia coli envelope alterations. J. Bacteriol. 155:402406. 26. Rothbaum, R., A. J. McAdams, R. Gianella, and J. C. Partin. 1982. A clinicopathologic study of enterocyte adherent Escherichia coli, a cause of protracted diarrheoa in infants. Gastroenterology 83:441-454. 27. Sereny, B. Experimental keratoconjunctivitis shigellosa. 1957. Acta Microbiol. Acad. Sci. Hung. 4:367-376. 28. Stanisisch, V. A. 1988. Identification and analysis of plasmids at the genetic level. Methods Microbiol. 21:11-47. 29. Streun, A., P. Coudert, and G. L. Rossi. 1979. Characterization of Eimeria species. II. Sequential morphologic study of the endogenous cycle of Eimeria perforans (Leuckart 1879, Sluiter and Swellengrebel 1912). Z. Parasitenkd. 60:37-53. 30. Takeuchi, A., L. R. Inman, P. D. O'Hanley, J. R. Cantey, and W. B. Lushbaugh. 1978. Scanning and transmission electron microscopical study of Escherichia coli 015 (RDEC-1) enteric infection in rabbits. Infect. Immun. 19:686-694. 31. Taylor, C. J., A. Hart, R. M. Batt, C. McDougal, and L. McLean. 1986. Ultrastructural and biochemical changes in human jejunal mucosa associated with enteropathogenic Escherichia coli (0111) infection. J. Pediatr. Gastroenterol. Nutr. 5:70-73. 32. Taylor, J., P. M. Wilkins, and J. M. Payne. 1961. Relation of rabbit gut reaction to enteropathogenic E. coli. J. Exp. Pathol. 42:43-52. 33. Ulshen, M. H., and J. L. Rollo. 1980. Pathogenesis of Escherichia coli enteritis: evaluation of the gnotobiotic piglet as a model of human infection. Gut 26:270-578. 34. Watanabe, T., and T. Fukasawa. 1961. Episome-mediated transfer of drug resistance in Enterobacteriaceae. I. Transfer of resistance factors by conjugation. J. Bacteriol. 81:669-678. 35. Wolf, M. K., G. P. Andrews, D. L. Fritz, R. W. Sjogren, Jr., and E. C. Boedeker. 1988. Characterization of the plasmid from Escherichia coli RDEC-1 that mediates expression of adhesin AF/Rl and evidence that AF/Rl pili promote but are not essential for enteropathogenic disease. Infect. Immun. 56:18461857. 36. Wolf, M. K., and E. C. Boedeker. 1990. Cloning of the genes for AF/Rl pili from rabbit enteroadherent Escherichia coli RDEC-1 and DNA sequence of the structural subunit. Infect. Immun. 58:1124-1128.. R. A R T E. T C. D E. Downloaded from http://iai.asm.org/ on August 22, 2017 by INRA - old. 2. Bouanchaud, D. H., M. R. Scavizzi, and Y. A. Chabbert. 1969. Elimination by ethidium bromide of antibiotic resistance in enterobacteria and staphylococci. J. Gen. Microbiol. 54:417425. 3. Camguilhem, R., F. Lebas, and C. Labie. 1986. Reproduction experimentale chez le lapin en engraissement d'une diarrh6e provoquee par une souche d'Escherichia coli de s6rogroupe 0103. Ann. Rech. Vet. 17:409424. 4. Camguilhem, R., and A. Milon. 1989. Biotypes and 0 serogroups of Escherichia coli involved in intestinal infections of weaned rabbits: clues to diagnosis of pathogenic strains. J. Clin. Microbiol. 27:743-747. 5. Camguilhem, R., G. Mureau, J. A. Nicolas, J. Brocas, and J. Tournut. 1986. Groupage serologique 0 et antibiosensibilite des souches d'Escherichia coli isolees en France sur des lapins diarrheiques apres le sevrage. Rev. Med. Vet. 137:205-212. 6. Cantey, J. R., and R. K. Blake. 1977. Diarrhea due to Escherichia coli in the rabbit: a novel mechanism. J. Infect. Dis. 135:454-462. 7. Cheney, C. P., S. B. Formal, P. A. Schad, and E. C. Boedeker. 1983. Genetic transfer of a mucosal adherence factor (R1) from an enteropathogenic Escherichia coli strain into a Shigella flexneri strain and the phenotypic suppression of this adherence factor. J. Infect. Dis. 147:711-723. 8. Coudert, P., D. Licois, and J. Besnard. 1988. Establishment of a S.P.F. breeding colony without hysterectomy and hand rearing procedures, p. 137-148. In Proceedings of the 4th Congress of the World Rabbit Science Association, Budapest, vol. 2. World Rabbit Science Association, Budapest. 9. Coudert, P., D. Licois, and A. Streun. 1979. Characterization of Eimeria species. Isolation and study of pathogenicity of a pure strain of Eimeria perforans (Leuckart 1879, Sluiter and Swellengrebel 1912). Z. Parasitenkd. 60:37-53. 10. Cravioto, A., R. J. Gross, S. M. Scotland, and B. Rowe. 1979. An adhesive factor found in strains of Escherichia coli belonging to the traditional infantile enteropathogenic serotypes. Curr. Microbiol. 3:95-99. 11. Giannella, R. A. 1976. Suckling mouse model for detection of heat-stable Escherichia coli enterotoxin: characteristics of the model. Infect. Immun. 14:95-99. 12. Guillot, J. F., E. Chaslus-Dancla, and J. P. Lafont. 1977. Spontaneous implantation of antibiotic-resistant Enterobacteriaceae in the digestive tract of chickens in the absence of selective pressure. Antimicrob. Agents Chemother. 12:697-702. 13. Gyles, C. L., S. Palchaudhuri, and W. K. Maas. 1977. Naturally occurring plasmid carrying genes for enterotoxin production and drug resistance. Science 198:198-199. 14. Kado, C. I., and S. T. Liu. 1981. Rapid procedure for detection and isolation of large and small plasmids. J. Bacteriol. 145:13651373. 15. Lin, E. C. C. 1987. Dissimilatory pathways for sugars, polyols, and carboxylates, p. 244-284. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium, vol. 1. American Society for Microbiology, Washington, D.C. 16. Milon, A., J. Esslinger, and R. Camguilhem. 1990. Adhesion of Escherichia coli strains isolated from diarrheic weaned rabbits to intestinal villi and HeLa cells. Infect. Immun. 58:2690-2695. 17. Moon, H. W., S. C. Whipp, R. A. Argenzio, M. M. Levine, and R. A. Giannelia. 1983. Attaching and effacing activities of rabbit and human enteropathogenic Escherichia coli in pig and rabbit intestines. Infect. Immun. 41:1340-1351. 18. O'Brien, A. D., G. D. Laveck, M. R. Thompson, and S. B. Formal. 1982. Production of Shigella dysenteriae type 1 like cytotoxin by E. coli. J. Infect. Dis. 146:763-769.. INFECT. IMMUN..
(7) RETRACTION R Plasmid in Escherichia coli 0103 Coding for Colonization of the Rabbit Intestinal Tract ALAIN REYNAUD, MICHEL FEDERIGHI, DOMINIQUE LICOIS, JEAN-FRANCOIS GUILLOT, AND BERNARD JOLY. Laboratoire d 'Analyses Vterinaires et Biologiques du Departement du Puy de D6me, RN 89-BP 42, 63000 Clermont-Ferrand, Unite de Pathologie du Lapin, INRA de Tours Nouzily, 37380 Monnaie, Universit& de Tours, 37000 Tours, and Service de Bact6riologie, Facul&t de Pharmacie, 63001 Clermont-Ferrand, France. Volume 59, no. 6, p. 1888-1892: It was reported that an R plasmid harbored by the Escherichia coli 0103 strain GV was responsible for the ability of the strain to colonize the intestinal tract of young rabbits. This result could not be confirmed in later experiments that we performed with various transconjugants that had acquired the R plasmid of strain GV. Under very strictly controlled experimental conditions, none of the transconjugants colonized the rabbit intestinal tract. As a result, Fig. 2, 3, and 4 in our article and the corresponding results contained therein can no longer be considered valid and our conclusion that the R plasmid codes for colonization of the rabbit intestinal tract can no longer be substantiated. We are therefore retracting these parts of the article. Further experiments are being carried out to assess the exact role of the R plasmid.. 4533.
(8)
Documents relatifs