Veterinary Microbiology 64 (1998) 75±81
A study of relationships among F17 a producing enterotoxigenic and non-enterotoxigenic Escherichia coli strains isolated from diarrheic calves Michel Contrepoisa,*, Yolande Bertina, Pierre Pohlb, Bertrand Picardc, Jean-Pierre Girardeaua a
INRA, Laboratoire de Microbiologie, Centre de Recherches de Clermont-Ferrand-Theix, 63122, Saint-GeneÁs-Champanelle, France b CERVA, Groeselenberg 99, B-1180, Bruxelles, Belgium c Laboratoire de BacteÂriologie±Virologie, Faculte de MeÂdecine de Brest, 29200, Brest, France Received 20 March 1998; accepted 11 September 1998
Abstract We investigated the clonal relationships among 41 enterotoxigenic (ETEC) or nonenterotoxigenic (NETEC) Escherichia coli strains producing the F17 a fimbriae isolated from diarrheic calves in France or Belgium in the early 1980s. Twenty-three of the 26 ETEC strains were highly clonally related, most of them with a O101:K32:H9-serotype. The NETEC strains were also divided in clonal subgroups, most of them with O101:H-serotype. The F17 a positive ETEC strains are no longer isolated from diarrheic calves in these countries. It is postulated that the use of a vaccine including O101, K32 and H9 antigens in addition to K99 (F5) explains the strongly reduced isolation of the O101:K32:H9, K99 (F5) E. coli clone. # 1998 Elsevier Science B.V. All rights reserved. Keywords: Escherichia coli; Cattle-Bacteria; Fimbriae; Clone; Vaccination
1. Introduction The main colonization factors or surface proteins described in pathogenic bovine Escherichia coli are K99 (F5) (Smith and Linggood, 1972; érskov et al., 1975), F41 (Morris et al., 1982), F17-related fimbriae (Bertin et al., 1996a) and the non-fimbrial * Corresponding author. Tel.: +33-4-73624244; fax: +33-473624581; e-mail:
[email protected] 0378-1135/98/$ ± see front matter # 1998 Elsevier Science B.V. All rights reserved. PII S 0 3 7 8 - 1 1 3 5 ( 9 8 ) 0 0 2 5 3 - 3
76
M. Contrepois et al. / Veterinary Microbiology 64 (1998) 75±81
CS31A antigen (Girardeau et al., 1988). Within the F17 family, four variants sharing adhesins with N-acetyl±glucosamine-mediated receptor specificity are known. F17 a was first described as the Fy antigen by Girardeau et al. (1980), then as the Att25 antigen by Pohl et al. (1982) before its molecular characterization by Lintermans et al. (1988, 1991). F17 a fimbriae were identified in NETEC strains and in ETEC strains bearing the K99 (F5) and the F41 fimbrial antigen or the CS31A fibrillae. With these K99-F17-positive strains the production of K99 antigen was highly modulated by growth conditions including oxygen tension and carbohydrate availability (Girardeau et al., 1982). In an epidemiological survey by Contrepois et al. (1985a) 415 E. coli strains isolated from young diarrheic calves in France were studied. Thirty strains produced F17 a (7.2%), 78 were K99-producing ETEC (18.3%) and eight of these ETEC also produced F17 a (10.2% of the ETEC strains). Prevalence of F17 a producing E. coli was similar in Belgium (Pohl and Mainil, 1995). Calf diarrhoea related to K99 ETEC infection is efficiently controlled by K99 vaccination of the dams which produce colostral antibodies inhibiting colonization of the calf intestine by K99 ETEC (Contrepois et al., 1978, 1985b; Moon, 1978; Desmettre et al., 1981). We have also demonstrated that anti-K99 colostral protection was inefficient in calves experimentally infected with bovine ETEC producing both K99 and F17 a colonization factors. Protection was achieved only when colostrum contained antibodies directed against both K99 and F17 a fimbriae (Contrepois and Girardeau, 1985). Specific interest in bovine E. coli producing both K99 and F17 a led us to better characterize the F17 a positive bovine E. coli strains and to establish their clonal diversity. 2. Materials and methods 2.1. Bacterial strains and serology Escherichia coli strains were isolated in the early 1980s from faeces of diarrheic calves, received from regional veterinary laboratories or isolated by the authors in different geographic areas in France (35 strains) or Belgium (6 strains). All strains were isolated from different animals and in one case only from two animals in the same farm. Fimbrial antigens F17 a and K99 were identified using specific antisera and slide agglutination of bacteria grown on Minca agar (Contrepois et al., 1985a) or a Western blotting assay as previously described (Bertin et al., 1996b). The F17 a-positive E. coli were kept on Dorset egg medium at 48C and used for the present study. OKH complete serotypes (Table 1) were established by Drs. F. and I. érskov using standard procedures at the International Center for Escherichia coli and Klebsiella (Copenhagen, Denmark). K means that the strain produces a capsular type A antigen which was not identified; (O101) means that O serogroup was identified as previously described using an immunodot method with a specific O101 rabbit antiserum (Cherifi et al., 1990) and the visualization of an electrophoretic pattern (SDS-PAGE) of LPS similar to the reference E. coli strain B41, O101:K-:H-, K99, F41 (érskov et al., 1975).
M. Contrepois et al. / Veterinary Microbiology 64 (1998) 75±81
77
2.2. Detection of genes encoding K99, STa and F17 Identification of the genes involved in biogenesis of K99 and STa was achieved using the colony hybridization method previously described (Mainil et al., 1986). A multiplex PCR method was used for the genotypic detection of the F17 subtypes (Bertin et al., 1996b). 2.3. Utilization of sugar Growth on minimal agar medium (NaH2PO4, 1.36 g lÿ1; Na2HPO4 2H2O, 10.1 g lÿ1; (NH4)2SO4, 2 g lÿ1; Agar, 15 g lÿ1; MgSO47H2O, 10ÿ2 g lÿ1; MnCl24H2O, 10ÿ3 g lÿ1; FeCl36H2O, 1.35 10ÿ4 g lÿ1; CaCl22H2O, 4 10ÿ4 g lÿ1) supplemented with 5 g of either sorbose, dulcitol, inositol, raffinose, sucrose, rhamnose or adonitol per liter of filtered solution was observed after 48 h of incubation at 378C. A control with glucose identified isolates with growth factor requirements. Bacteria were spotted on the surface of the agar. 2.4. Crude outer membrane protein (OMP) analysis Bacteria grown on Minca agar medium (GuineÂe et al., 1977) for 18 h at 378C were collected in 2 ml of sterile phosphate-buffered saline and centrifuged for 15 min at 4500 g. The pellet was then suspended in 10 ml of Tris buffer at 48C (0.05 M Tris base, 1 mM EDTA [pH 7.8] with HCl). Sonication with a 15% continuous cycle was performed on ice for four periods of 15 s (New Brunswick). After centrifugation for 20 min at 1200 g, the supernatant was centrifuged at 5000 g for 1 h at 48C. The pellet was suspended in 10 mM Tris buffer with 5 mM MgCl2 (pH 8) containing 2% Triton X-100 to solubilize the inner membrane (Schnaitman, 1974). After centrifugation at 50 000 g for 1 h at 48C, the pellet was dissolved in buffer (0.025 M Tris base, 0,192 M glycine, 1% sodium dodecyl sulfate (SDS), pH 8.6). Polyacrylamide gel electrophoresis (PAGE) was performed after the addition of Laemmli buffer (Laemmli, 1970) and heating at 1008C for 5 min. A 10% polyacrylamide gel was used and electrophoresis was run at 20 mA for 5 h. Silver staining of the gel was done as described by Oakley et al. (1980). 2.5. Electrophoretic analysis of esterases E. coli strains were cultured in Fernbach flasks containing 500 ml of L broth (Lennox, 1955) without glucose. The flasks were shaken vigorously for 18 h at 378C in a reciprocating water bath shaker set at about 70 oscillations minÿ1. The preparation of extracts, protein estimation, horizontal slab PAGE (7% w/v acrylamide, Tris glycine buffer, pH 8.6) the estimation of electrophoretic mobility and esterase staining have been described previously (Goullet and Picard, 1985). Each electrophoretic mobility variant was designated as an allozyme. When an esterase was not detected in a strain, it was recorded as a null allozyme. Each distinctive combination of allozymes for the four varieties of esterases (A, B, C and I) was designated an electrophoretic type ET (Goullet and Picard, 1986).
78
M. Contrepois et al. / Veterinary Microbiology 64 (1998) 75±81
3. Results All the F17 E. coli strains identified with a specific antiserum were shown to belong to the F17 a subtype with the genotypic method. The 41 isolates were categorized into two main groups which are 26 ETEC strains and 15 non-ETEC strains and subdivided into subgroups on the basis of OMP patterns A, B, C, D (Fig. 1) and esterase ETs (Table 1). Discrepancies between phenotypic and genotypic detection of the K99 antigen in two strains (not shown) may result either from a loss of expression of K99 genes or loss of genes in culture collection. The ETEC group is composed of a main subgroup I (23 strains) with OMP pattern A1, esterase ET1, O serogroup 101, except one strain (0141) and H9 serogroup. Most of them have capsular antigen K32 but two strains have a K28 serogroup. Strains included in ETEC II subgroup are differentiated from subgroup 1 by their B type OMP pattern. E. coli strain with serotype O8 rel:K32:H19 has a completely different OMP pattern and esterase ET-type. The non-ETEC group (NETEC) is more heterogenous. The two NETEC I and NETEC II subgroups of E. coli strains with O101 and non-motile H-serogroups are distinguished by their OMP patterns A2 and A3, respectively, and their esterase types 3 and 4, respectively. The last four E. coli strains of the NETEC group are completely heterogenous for OKH serogroups, OMP and esterasetypes. The ETEC and NETEC groups and subgroups are not clearly differentiated by the utilization of sugars and the main biotype identified is sorbose±rhamnose-positive, Table 1 Grouping of F17 a positive E. coli isolates Group
No. of isolates
K99
OMP pattern (1)
Esterase ET
Serotype (2) (no. of isolates)
ETEC I
23
A1
1
ETEC II
2
B
1
ETEC III NETEC I
1 7
ÿ
C A2
2 3
NETEC NETEC NETEC NETEC NETEC
4 1 1 1 1
ÿ ÿ ÿ ÿ ÿ
A3 D C C A2
4 5 6 7 8
O101:K32:H9 (12) (O101) (6) O101:K28:H9 (2) O101:K:H9 (1) O141ac:H9 (1) O r o u g h : K : H 9 (1) O101:K28:H9 (1) NT (1) O8rel:K39:H19 O101:K32:Hÿ (2) O101:K:Hÿ (O101) O101:K:Hÿ (4) (O101) O?:K42:H19 O?:K-:H10 O2:K7:H42 (12)
II III IV V VI3
1: OMP pattern identified in Fig. 1. 2: K means K antigen type A present but serogroup not determined; (O101) means a strong reaction with rabbit anti-O101 antibodies and an O101 LPS pattern observed following SDS-PAGE analysis of 608C extracts; NT means not tested.
M. Contrepois et al. / Veterinary Microbiology 64 (1998) 75±81
79
Fig. 1. Representative patterns of the major OMPs of the F17 a-positive E. coli strains (the OMP patterns are numbered as in Table 1). Molecular weight (lane 1), strains 263 (lane 2), AY1 (lane 3), AY5 (lane 4), 753 (lane 5), AY2 (lane 6), 83380 (lane 7).
dulcitol±inositol±raffinose±sucrose-negative and lastly adonitol-positive, the latter characteristic being rather uncommon among E. coli strains (Sojka, 1965). 4. Discussion Serotypes, biotypes and OMP patterns are used to describe E. coli clonal groups (Achtman et al., 1983). The application of multilocus enzyme electrophoresis for identifying and epidemiologically tracing pathogens has established that for many bacterial species, a few geographically widespread clones account for most cases of infectious diseases (Achtman et al., 1983; Selander et al., 1986). E. coli strains producing F17 a fimbriae were a good candidate for a clonal study because E. coli strains producing a fimbrial-type often appear clonally related. In a previous investigation Contrepois et al. (1993) showed that clonal relationships could be demonstrated among E. coli strains producing the CS31A surface antigen. In addition, it is known that bovine ETEC belong to a limited number of O serogroups which are O101, O9, O8 and O20 (Gaastra and de Graaf, 1982). The F41 fimbrial antigen is only expressed by strains with O101 or O9 serogroups (Morris et al., 1980) and CS31A, a K88-F41-related surface antigen (Girardeau et al., 1988) is restricted to bovine ETEC with serogroups O8 or O20 (Girardeau et al., 1988; Contrepois et al., 1993). Although a precise clonal study of bovine ETEC is to our knowledge not available, it can be postulated that plasmid genes mediating biogenesis of the main virulence factors K99 and STa are expressed or ecologically restricted to a few E. coli clones. The results obtained in this study focus on F17 a-producing ETEC which represent 10% of the whole bovine ETEC population (Contrepois et al., 1985a) clearly demonstrate that they are restricted to a main clone and a related subclone of E. coli strains with serotypes O101:K32:H9 or
80
M. Contrepois et al. / Veterinary Microbiology 64 (1998) 75±81
O101:K28:H9 with the exception of a strain with O141ac:H9-serotype. A majority of F17 a-positive NETEC also has a O101 serogroup but they are non-motile and have OMP patterns and esterase ETs which distinguish them from the F17 a-positive ETEC. This study also confirms a specific biotype of F17 a-positive E. coli strains which is in most cases adonitol-positive. Such a result was already underlined by Girardeau et al., 1982 and Contrepois and Gouet (1983) leading Pohl et al. (1984) to propose a selective citrate± adonitol medium for the isolation of F17 a bovine ETEC. There are confirmed reports from Belgium (P. Pohl, unpublished) and unconfirmed reports from France that F17 a-positive bovine ETEC are much less frequently isolated from diarrheic calves than they were in the 1980s. Since the major vaccine against ETEC in both countries contains antigens for O101, K32, H9, K99 and F17 a antigens, vaccination pressure may be responsible for this development. This would be consistent with experimental observations on powerful selection pressures when more than one antigen target is attacked (Contrepois et al., 1978, 1985b; Contrepois and Girardeau, 1985). References Achtman, M., Mercier, A., Kusecek, B., Pohl, P., Heuzenroeder, M., Aronson, W., Sutton, A., Silver, P., 1983. Six widespread bacterial clones among Escherichia coli K1 isolates. Infect. Immun. 39, 315±335. Bertin, Y., Girardeau, J.P., Darfeuille-Michaud, A., Contrepois, M., 1996a. Characterization of 20 K fimbria, a new adhesin of septicemic and diarrhea-associated Escherichia coli strains, that belongs to a family of adhesins with N-acetyl-D-glucosamine recognition, Infect. Immun. 64(1), 332±342. Bertin, Y., Martin, C., Oswald, E., Girardeau, J.P., 1996b. Rapid and specific detection of F17-related pilin and adhesin genes in diarrheic and septicemic Escherichia coli strain by multiplex PCR, J. Clin. Microbiol. 34(12), 2921±2928. Cherifi, A., Contrepois, M., Picard, B., Goullet, Ph., De Rycke, J., Fairbrother, J., Barnouin, J., 1990. Factors and markers of virulence in Escherichia coli from human septicemia. FEMS Microbiol. Lett. 70, 279±284. Contrepois, M., Girardeau, J.P., 1985. Additive protective effects of colostral antipili antibodies in calves experimentally infected with enterotoxigenic Escherichia coli. Infect. Immun. 50(3), 947±949. Contrepois, M., Gouet, Ph., 1983. CineÂtique d'excreÂtion feÂcale de Escherichia coli K99 par des veaux apreÁs infection expeÂrimentale. Ann. Rech. VeÂt. 14(2), 141±146. Contrepois, M., Girardeau, J.P., Dubourguier, H.C., Gouet, Ph., Levieux, D., 1978. Specific protection by colostrum form cows vaccinated with the K99 antigen in new born calves experimentally infected with Escherichia coli Ent K99. Ann. Rech. VeÂt. 9(2), 385±388. Contrepois, M., Martel, J.L., Bordas, C., Hayers, F., Millet, A., Ramisse, J., Sendral, R., 1985a. FreÂquence des pili FY et K99 parmi des souches de Escherichia coli isoleÂes de veaux diarrheÂiques en France, Ann. Rech. VeÂt. 16(1), 25±28. Contrepois, M., Girardeau, J.P., Dubourguier, H.C., Gouet, Ph., 1985b. Vaccination anti-K99 et protection colostrale des veaux infecteÂs expeÂrimentalement avec Escherichia coli K99, Ann. Rech. VeÂt. 16(1), 41±46. Contrepois, M., Bertin, Y., Girardeau, J.P., Picard, B., Goullet, P., 1993. Clonal relationships among bovine pathogenic Escherichia coli-producing surface antigen CS31A. FEMS Microbiol. Lett. 106, 217±222. Desmettre, Ph., Tixier, G., Valette, L., Contrepois, M., Gouet, Ph., Dubourguier, H.C., 1981. Prophylaxie des infections colibacillaires neÂo-natales du veau. Group. Tech. Vet. 224, 69±81. Gaastra, W., de Graaf, F.K., 1982. Host-specific fimbrial adhesins of non invasive enterotoxigenic E. coli strains. Microbial. Reviews 46, 129±161. Girardeau, J.P., Dubourguier, H.C., Contrepois, M., 1980. Attachement des Escherichia coli enteÂropathogeÁnes aÁ la muqueuse intestinale. Group. Tech. Vet. 190, 49±60.
M. Contrepois et al. / Veterinary Microbiology 64 (1998) 75±81
81
Girardeau, J.P., Dubourguier, H.C., Gouet, Ph., 1982. Effect of glucose and amino acids on expression of K99 antigen in E. coli. J. Gen. Microbiol. 128, 2243±2249. Girardeau, J.P., Der Vartanian, M., Ollier, J.L., Contrepois, M., 1988. CS31A, a new K88-related fimbrial antigen on bovine enterotoxigenic and septicemic E. coli strains. Infect. Immun. 56(8), 2180±2188. Goullet, P., Picard, B., 1985. A two-dimensional electrophoretic profile for bacterial esterases. Electrophoresis 6, 132±135. Goullet, P., Picard, B., 1986. Highly pathogenic strains of Escherichia coli revealed by the distinct electrophoretic pattern of carboxylesterase B. J. Gen. Microbiol. 132, 1853±1858. GuineÂe, P.A.M., Veltkamp, J., Jansen, W.H., 1977. Improved Minca medium for the detection of K99 antigen in calf enterotoxigenic strains of E. coli. Infect. Immun. 15, 676±678. Laemmli, U.K., 1970. Cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature (London) 22, 680±685. Lennox, E.S., 1955. Transduction of linked characters of the host by bacteriophage P1. Virology 1, 190±206. Lintermans, P., Pohl, P., Deboek, F., Bertels, A., Schlicker, C., Vandekerkhove, J., Van Damme, J., Van Montagu, M., De Greve, H., 1988. Isolation and nucleotide sequence of the F17 A gene encoding the structural protein of the F17 fimbriae in bovine enterotoxigenic E. coli. Infect. Immun. 56, 1475±1484. Lintermans, J.P., Bertels, A., Schlicker, C., Deboeck, F., Charlier, G., Pohl, P., Norgrem, M., Normark, S., Van Montagu, M., De Greve, H., 1991. Identification, characterization and nucleotide sequence of the F17-G gene which determines receptor-binding of E. coli F17 fimbriae. J. Bacteriol. 173, 3366±3373. Mainil, J.G., Mosley, S.L., Schneider, R.A., Sutch, K., Casey, T.A., Moon, H.W., 1986. Hybridization of bovine Escherichia coli isolates with gene probes for four enterotoxin (STaP, STaH, Stb and LT) and one adhesion factor (K99). Am. J. Vet. Res. 47, 1145±1148. Moon, H.W., 1978. Pili as protective antigens in vaccines for the control of enterotoxigenic Escherichia coli infections. Proc. 2nd Int. Symp. Neonatal Diarrhea, Veterinary Infectious Disease Organization, pp. 393± 410. Morris, J.A., Thorns, C.J., Sojka, W.J., 1980. Evidence for two adhesive antigens on the K99 reference strain Escherichia coli B41. J. Gen. Microbiol. 118, 107±113. Morris, J.A., Thorns, C.J., Scott, A.C., Sojka, W.J., Wells, G.A., 1982. Adhesion in vitro and in vivo associated with an adhesive antigen (F41) produced by a K99 mutant of the reference strain Escherichia coli B41. Infect. Immun. 36, 1146±1153. Oakley, B.R., Kirsch, D.R., Morris, N.E., 1980. A simplified ultrasensitive silver strain for detecting protein in polyacrylamide gels. Anal. Biochem. 105, 361±365. érskov, I., érskov, F., Smith, H.W., Sojka, W.J., 1975. The establishment of K99, a thermolabile transmissible E. coli K antigen, previously called `Kco,' possessed by calf and lamb enteropathogenic strains. Acta Pathol. Microbiol. Scand. Sect. B 83, 31±36. Pohl, P., Lintermans, P., Van Muylen, K., Schotte, M., 1982. Colibacilles enteÂrotoxinogeÁnes du veau posseÂdant un antigeÁne d'attachement diffeÂrent de l'antigeÁne K99. Ann. Med. VeÂt. 126, 569±571. Pohl, P., Lintermans, P., Vandergeheynst, M.C., Van Muylen, K., Schlicker, C., Van Robaeys, G., 1984. Milieu seÂlectif citrate-adonitol pour l'isolement de certaines souches de Escherichia coli K99, Ann. Microbiol. (Inst. Pasteur) 135B, 29±33. Pohl, P., Mainil, J., 1995. F17 positive E. coli. Vet. Record, 623±624. Schnaitman, C.A., 1974. Outer membrane proteins of E. coli. III. Evidence that the major protein of Escherichia coli O111 outer membrane consists of four distinct polypeptide species. J. Bacteriol. 118, 442±453. Selander, R.K., Cougant, D.A., Ochman, H., Muser, J.M., Gilmour, M.N., Whittam, T.S., 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 52, 873±884. Smith, H.W., Linggood, M.A., 1972. Further observations on Escherichia coli enterotoxins with particular regard to those produced by atypical piglet strains and by calf and lamb strains: The transmissible nature of these enterotoxins and of a K antigen possessed by calf and lamb strains. J. Med. Microbiol. 5, 243±250. Sojka, W.J., 1965. Escherichia coli in domestic animals and poultry. Commonwealth Agricultural Bureaux, Farnham Royal, Bucks., UK.