Abstract
Certain strains of Shiga toxin-producing Escherichia coli (STEC) which do not have the locus of enterocyte effacement pathogenicity island carry the STEC autoagglutinating adhesin (saa) gene. The distribution of the saa gene in STEC isolates from patients with hemolytic-uremic syndrome (HUS), patients with less severe diarrheal disease, asymptomatic individuals, and healthy cattle was examined. saa-positive strains were detected more frequently (P < 0.001) in STEC strains from bovines (32 of 56 strains) than in those from humans (8 of 91 strains). No significant association (P = 0.135) was found between the saa gene and STEC isolated from patients with HUS (6 of 46 strains) or diarrhea (2 of 29 strains) and from healthy controls (0 of 16 strains).
Shiga toxin-producing Escherichia coli (STEC), also referred to as verocytotoxin-producing E. coli, cause a broad spectrum of disease in humans ranging from mild diarrhea to severe disease, such as hemorrhagic colitis and hemolytic-uremic syndrome (HUS) (2, 8, 20). A wide range of STEC serogroups can cause human disease, although the most commonly isolated serogroups from patients infected with STEC, notably O26, O103, O111, O145, and O157 (16, 21), are those harboring the locus of enterocyte effacement (LEE) (4). One of the genes located on the LEE is eae (E. coli attaching and effacing), which encodes intimin, an outer membrane protein involved in intimate attachment of the bacteria to intestinal host cells (4). However, the presence of the LEE is not essential for pathogenesis, as a number of cases of severe STEC disease, including HUS, have been caused by LEE-negative strains (6, 9, 12). Paton et al. recently described a novel autoagglutinating adhesin, designated Saa (STEC autoagglutinating adhesin), in a LEE-negative O113:H21 STEC strain responsible for an outbreak of HUS (13). Those authors presented evidence that the saa gene was plasmid encoded and associated with the STEC enterohemolysin genes (13, 14). The saa gene was also detected in certain other STEC isolates from humans, including other LEE-negative strains isolated from sporadic HUS cases (14). This raised the possibility that Saa might be a virulence factor for virulent LEE-negative STEC in humans, but there were insufficient numbers of strains available for testing to draw any firm conclusions. In the present study, we examined the distribution of the saa gene in a large collection of STEC isolates from patients with HUS and diarrheal disease and from a healthy control group to test whether there is any association between the presence of saa and severe human STEC disease. We also investigated the distribution of the saa gene in isolates from healthy cattle in the United Kingdom to determine whether this gene may be associated with bovine STEC.
The 91 strains of STEC from the culture collection of the Laboratory of Enteric Pathogens (LEP), London, United Kingdom, comprised 46 strains from patients with HUS (Table 1), 29 isolated from patients with diarrhea who did not develop HUS (Table 2), and 16 from a healthy control group (Table 3) (3, 8, 16, 18, 19). Sixty-one of these strains were differentiated by serotyping, stx typing, and the presence or absence of the eae and STEC enterohemolysin (ehxA) genes (5, 11, 15, 17). The strains were isolated between 1983 and 2000 either at the LEP from fecal samples submitted for detection of STEC or at hospital laboratories and then sent to the LEP for confirmation and further tests. Fecal pat samples collected from healthy cattle from 41 different farms in Scotland between 1996 and 1999 were examined for the presence of STEC strains other than those belonging to serogroup O157 (7). Fifty-six STEC strains were isolated from 423 fecal samples; 30 of these strains were differentiated by serotyping, stx typing, and the presence or absence of the eae and ehxA genes (5, 11, 15, 17) (Table 4). These strains were tested for the presence of saa by PCR with primers 5′-CGTGATGAACAGGCTATTGC and 5′-ATGGACATGCCTGTGGCAAC, which amplify a 119-bp portion of the saa gene (14). To test for differences in the detection of particular genes between the four groups of strains, generalized linear models with binomial errors were performed. In all cases, a P value of <0.05 was taken to indicate significance. Fisher's exact tests were performed to test the significance of the association between the detection of particular genes.
TABLE 1.
Characterization of non-O157 STEC strains isolated from HUS patientsa
| Serotype (no. of strains)b | stx gene type | Presence or absence of:
|
||
|---|---|---|---|---|
| eae | ehxA | saa | ||
| O2:H6 (1) | 2 | − | − | − |
| O5:H- (1) | 1 | + | + | − |
| O9ab:H- (1) | 2 | − | − | − |
| O26:H- (1) | 2 | + | + | − |
| O26:H- (1) | 1 | + | + | − |
| O26:H11 (1) | 2 | + | + | − |
| O26:H11 (3) | 1 | + | + | − |
| O55:H7 (1) | 2 | + | − | − |
| O55:H10 (1) | 2 | − | − | − |
| O91:H21 (1) | 2 | − | + | + |
| O104:H2 (1) | 2 | − | − | − |
| O105ac:H18 (1) | 1+2 | − | + | + |
| O111ac:H- (2) | 1 | + | + | − |
| O111ac:H- (5) | 1+2 | + | + | − |
| O113:H4 (1) | 2 | − | − | − |
| O115:H10 (1) | 1 | − | − | − |
| O121:H19 (1) | 2 | + | + | − |
| O128ab:H2 (1) | 1+2 | − | + | − |
| O128ab:H7 (1) | 2 | + | − | − |
| O128ab:H25 (1) | 2 | + | + | − |
| O134:H25 (1) | 2 | − | + | + |
| O145:H- (1) | 1 | + | + | − |
| O145:H25 (5) | 2 | + | + | − |
| O153:H25 (1) | 2 | − | − | − |
| O163:H19 (2) | 2 | − | + | + |
| O165:H25 (2) | 2 | + | + | − |
| O168:H- (1) | 2 | − | − | − |
| O172:H- (1) | 2 | + | + | − |
| O173:H21 (1) | 2 | − | + | + |
| O?:H4 (1)c | 2 | − | − | − |
| O?:H21 (2) | 2 | − | − | − |
| O?:H40 (1) | 2 | + | − | − |
See references 9, 18, 19, 20, and 21.
Where the serotype is the same, isolates have been differentiated according to stx type and the presence or absence of the eae and ehxA genes.
O?, could not be serogrouped using the current serotyping scheme.
TABLE 2.
Characterization of non-O157 STEC strains isolated from patients with diarrhea but who did not develop HUSa
| Serotype (no. of strains) | stx gene type | Presence or absence of:
|
||
|---|---|---|---|---|
| eae | ehxA | saa | ||
| O4:H10 (1) | 1+2 | − | + | − |
| O5:H- (2) | 1 | + | + | − |
| O26:H11 (5) | 1 | + | + | − |
| O76:H7 (1) | 2 | + | + | − |
| O91:H- (1) | 1+2 | − | − | − |
| O105ac:H18 (1) | 1+2 | − | + | + |
| O118:H12 (3) | 2 | − | − | − |
| O121:H19 (1) | 2 | + | + | − |
| O128ab:H- (2) | 1+2 | − | + | − |
| O128ab:H2 (6) | 1+2 | − | + | − |
| O128ab:H8 (1) | 1+2 | − | − | + |
| O?:H- (1)b | 1 | + | + | − |
| O?:H10 (1) | 2 | − | − | − |
| O?:H19 (1) | 1 | − | − | − |
| Orough:H- (1) | 2 | − | − | − |
| Orough:H45 (1) | 2 | − | − | − |
See references 9, 18, 19, 20, and 26.
O?, could not be serogrouped using the current serotyping scheme.
TABLE 3.
Strains of non-O157 STEC isolated from a healthy control group during a study of infectious intestinal diseasea
| Serotype (no. of strains) | stx gene type | Presence or absence of:
|
||
|---|---|---|---|---|
| eae | ehxA | saa | ||
| O82:H2 (1) | 1+2 | − | + | − |
| O91:H- (1) | 1 | − | + | − |
| O91:H10 (1) | 2 | − | − | − |
| O115:H10 (1) | 1 | − | − | − |
| O118:H1 (1) | 2 | − | − | − |
| O128ab:H2 (3) | 1+2 | − | + | − |
| O146:H21 (2) | 1+2 | − | + | − |
| O162:H6 (1) | 1+2 | − | − | − |
| O162:H8 (1) | 1+2 | − | − | − |
| O?:H- (1)b | 2 | − | + | − |
| O?:H- (1) | 1+2 | − | − | − |
| O?:H18 (1) | 1+2 | − | + | − |
| O?:H21 (1) | 1 | − | + | − |
See reference 3.
O?, could not be serogrouped using the current serotyping scheme.
TABLE 4.
Characterization of non-O157 STEC strains isolated from Scottish cattlea
| Serotype (no. of strains)b | stx gene type | Presence or absence of:
|
||
|---|---|---|---|---|
| eae | ehxA | saa | ||
| O8:H16 (1) | 1+2 | − | + | + |
| O15:H- (2) | 2 | − | − | − |
| O15:H16 (5) | 2 | − | − | − |
| O22:H8 (8) | 1+2 | − | + | + |
| O87:H16 (2) | 2 | − | − | − |
| O91:H49 (1) | 2 | − | + | + |
| O103:H2 (1) | 1 | + | + | − |
| O105ac:H18 (1) | 2 | − | + | + |
| O105ac:H18 (6) | 1+2 | − | + | + |
| O113:H4 (1) | 2 | − | − | − |
| O113:H21 (3) | 2 | − | + | + |
| O136:H- (1) | 2 | − | − | − |
| O159:H7 (1) | 2 | − | + | − |
| O172:H21 (1) | 1+2 | − | − | − |
| E11362/78:H- (1)c | 2 | − | − | − |
| E43478/86:H8 (1)c | 2 | − | + | + |
| E43478/86:H12 (1)c | 2 | − | + | + |
| E8686/77:H16 (2)c | 2 | − | − | − |
| E54071/88:H19 (1)c | 1+2 | − | + | + |
| E7477/77:H25 (1)c | 1 | + | + | − |
| O?:H- (2)d | 2 | − | − | − |
| O?:H2 (1) | 2 | − | − | − |
| O?:H7 (1) | 2 | − | + | + |
| O?:H11 (1) | 2 | − | + | + |
| O?:H16 (3) | 2 | − | − | − |
| O?:H19 (3) | 2 | − | + | + |
| O?:H19 (1) | 1+2 | − | + | + |
| O?:H25 (1) | 2 | − | + | + |
| O?:H40 (1) | 2 | − | + | + |
| O?:H49 (1) | 2 | − | + | + |
See reference 7.
Where the serotype is the same, isolates have been differentiated according to stx type and the presence or absence of the eae and ehxA genes.
Provisional new serotypes (formal “O” group pending).
O?, could not be serogrouped using the current serotyping scheme.
Forty (27%) of the 147 STEC strains in this study had the saa gene (Tables 1-3). None of these 40 saa-positive strains carried the eae gene, and 39 were ehxA positive; the exceptional strain (saa positive and ehxA negative) belonged to serotype O128ab:H8 (Table 2). The saa gene was detected in 6 (13%) of 46 strains from humans with HUS, and 28 (61%) of these 46 strains carried the eae gene, while 12 (26%) had neither (Table 1). Two (7%) of 29 strains from patients with diarrhea or bloody diarrhea who did not develop HUS were saa positive. The eae gene was detected in 10 (34%) strains from this group, and 17 (59%) had neither the eae nor the saa gene (Table 2). None of the isolates from the healthy control group had the saa gene (Table 3). Characterization of the 56 bovine strains showed that 32 (57%) carried the saa gene, 2 (4%) were eae positive, and 22 (39%) were negative for both eae and saa (Table 4). None of the strains harboring the saa gene had the eae gene.
Statistical analysis revealed a significant difference in the frequencies of saa-positive strains between the four groups (χ2 = 45.45; df = 3; P < 0.001). However, if the cattle strains are excluded from the analysis, then this significance was lost (χ2 = 4.00; df = 2; P = 0.135). Therefore, the number of cattle strains containing the saa gene is significantly higher than the number of saa-positive human strains, although there is no difference in the presence of the saa gene between the strains from patients with HUS, patients with diarrhea, and the asymptomatic individuals. Statistical tests also showed a significant difference in the frequencies of isolates containing the eae gene between the four groups (χ2 = 55.89; df = 3; P < 0.001), with the highest frequency of intimin-positive strains being found in the HUS group.
Variables over and above the group categories (HUS, diarrhea, healthy control, and cattle strains), such as toxin type and the presence of the eae and ehxA genes, and the fact that certain serotypes are represented by more strains than are others may influence the results. However, the panel of strains reflects the fact that some serotypes associated with HUS (e.g., O26:H11) or less severe diarrhea (e.g., O128ab:H2) are more commonly isolated than others (2, 9, 16, 20, 21). To date, the saa gene has not been found in STEC strains harboring the eae gene, and the association of saa with bovine strains may be due to the fact that there are so few intimin-positive strains in this group. However, statistical analysis of only the intimin-negative strains showed that there was still a significantly higher frequency of saa-positive strains in STEC strains from cattle than in those from humans (P < 0.001, Fisher's exact test). Thirty-nine of 40 (98%) saa-positive strains also carried the ehxA genes, and 39 of 51 (76%) strains that were negative for the saa and eae genes were also negative for the ehxA gene. Tests showed that for the eae-negative strains, there was a significant association between the presence of the saa and ehxA genes (P < 0.001, Fisher's exact test). These results are consistent with the observation of Paton et al. that the saa gene is plasmid encoded and associated with the plasmid-encoded ehxA gene (13). Only one strain of serotype O128ab:H8 was saa positive but ehxA negative, and further work is being carried out to determine whether the saa gene is plasmid encoded in this strain.
The only saa-positive serotype common to both human and bovine STEC strains in this study was O105ac:H18. Carriage of the saa gene appeared to be specific to certain serotypes within a serogroup, e.g., O113:H21 was saa positive, whereas O113:H4 was saa negative. Of the 22 bovine serotypes that have yet to be designated as a formal “O” type (i.e., provisional types and “O?”), 13 carried the saa gene. This highlights the need to extend the serotyping scheme to include new and emerging STEC serogroups.
Our results showed that there was no significant association between strains of STEC isolated from patients with HUS and the saa gene, although they do support the suggestion that the eae gene is detected more frequently in STEC from this group of patients (1, 10). However, the saa gene was significantly associated with bovine STEC strains, suggesting that Saa may have a role in attachment to the bovine gut. Further work is being carried out, both in vitro and in vivo, to examine the role of Saa in attachment to the bovine gut mucosa.
Acknowledgments
Fecal pat samples were collected as part of the study “Determination of the Prevalence of E. coli O157 in Scottish Beef Cattle” funded by the Scottish Executive Environment and Rural Affairs Department. Bovine STEC were isolated as part of the International Partnership Research Award in Veterinary Epidemiology project entitled “Epidemiology and Evolution of Enterobacteriaceae Infections in Humans and Domestic Animals” funded by the Wellcome Trust.
We thank Doreen Bassett, Jude Evans, Hazel Knight, Judi Lee, Alistair Smith, and Helen Ternent for their excellent technical assistance.
REFERENCES
- 1.Boerlin, P., S. A. McEwan, F. Boerlin-Petzold, J. B. Wilson, R. P. Johnson, and C. L. Gyles. 1999. Associations between virulence factors of Shiga toxin-producing Escherichia coli and disease in humans. J. Clin. Microbiol. 37:497-503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Caprioli, A., A. E. Tozzi, G. Rizzoni, and H. Karch. 1997. Non-O157 Shiga toxin-producing Escherichia coli infections in Europe. Emerg. Infect. Dis. 3:578-579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Evans, J., A. Wilson, G. A. Willshaw, T. Cheasty, D. S. Tompkins, J. G. Wheeler, and H. R. Smith. 2002. Vero cytotoxin-producing Escherichia coli in a study of infectious intestinal disease in England. Clin. Microbiol. Infect. Dis. 8:183-186. [DOI] [PubMed] [Google Scholar]
- 4.Frankel, G., A. Phillips, I. Rosenshine, G. Dougan, J. Kaper, and S. Knutton. 1998. Enteropathogenic and enterohaemorrhagic Escherichia coli: more subversive elements. Mol. Microbiol. 30:911-921. [DOI] [PubMed] [Google Scholar]
- 5.Gross, R. J., and B. Rowe. 1985. Serotyping of Escherichia coli, p. 345-360. In M. Sussman (ed.) The virulence of Escherichia coli. Cambridge University Press, Cambridge, United Kingdom.
- 6.Ito, H., A. Terai, H. Kurazono, Y. Takeda, and M. Nishibuchi. 1990. Cloning and nucleotide sequencing of Vero toxin 2 variant genes from Escherichia coli O91:H21 isolated from a patient with hemolytic uremic syndrome. Microb. Pathog. 8:47-60. [DOI] [PubMed] [Google Scholar]
- 7.Jenkins, C., H. Chart, T. Cheasty, G. A. Willshaw, M. C. Pearce, G. Foster, G. J. Gunn, H. R. Smith, G. Dougan, B. A. Synge, and G. Frankel. 2002. Verocytotoxin-producing Escherichia coli (VTEC) other than serogroup O157 from Scottish cattle. Vet. Rec. 151:58-60. [DOI] [PubMed] [Google Scholar]
- 8.Karmali, M. 1989. Infection by verocytotoxin-producing Escherichia coli. Clin. Microbiol. Rev. 2:15-38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kleanthous, H., H. R. Smith, S. M. Scotland, R. J. Gross, B. Rowe, C. M. Taylor, and D. V. Milford. 1990. Haemolytic uraemic syndromes in the British Isles, 1985-1988: association with verocytotoxin producing Escherichia coli. Part 2. Microbiological aspects. Arch. Dis. Child. 65:722-727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Meng, J., S. Zhao, and M. P. Doyle. 1998. Virulence genes of Shiga toxin-producing Escherichia coli isolated from food, animals and humans. Int. J. Food Microbiol. 45:229-235. [DOI] [PubMed] [Google Scholar]
- 11.Oswald, E., H. Schmidt, S. Morabito, H. Karch, O. Marches, and A. Caprioli. 2000. Typing of intimin genes in human and animal enterohemorrhagic and enteropathogenic Escherichia coli: characterization of a new intimin variant. Infect. Immun. 68:64-71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Paton, A. W., M. C. Woodrow, R. M. Doyle, J. A. Lanser, and J. C. Paton. 1999. Molecular characterization of a Shiga toxin Escherichia coli O113:H21 lacking eae responsible for a cluster of cases of hemolytic-uremic syndrome. J. Clin. Microbiol. 37:3357-3361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Paton, A. W., P. Srimanote, M. C. Woodrow, and J. C. Paton. 2001. Characterization of Saa, a novel auto agglutinating adhesin produced by locus of enterocyte effacement-negative Shiga-toxigenic Escherichia coli strains that are virulent for humans. Infect. Immun. 69:6999-7009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Paton, A. W., and J. C. Paton. 2002. Direct detection and characterization of Shiga toxigenic Escherichia coli by multiplex PCR for stx1, stx2, eae, ehxA, and saa. J. Clin. Microbiol. 40:271-274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Schmidt, H., L. Beutin, and H. Karch. 1995. Molecular analysis of the plasmid-encoded haemolysin of Escherichia coli O157:H7 strain EDL 933. Infect. Immun. 63:1055-1060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Schmidt, H., C. Geitz, P. Tarr, M. Frosch, and H. Karch. 1999. Non-O157:H7 pathogenic Shiga toxin-producing Escherichia coli: phenotypic and genetic profiling of virulence traits and evidence of clonality. J. Infect. Dis. 179:115-123. [DOI] [PubMed] [Google Scholar]
- 17.Thomas, A., H. R. Smith, G. A. Willshaw, and B. Rowe. 1991. Non-radioactively labelled polynucleotide and oligonucleotide DNA probes, for selectively detecting Escherichia coli strains producing Vero cytotoxins VT1, VT2 and VT2 variant. Mol. Cell. Probes 5:129-135. [DOI] [PubMed] [Google Scholar]
- 18.Thomas, A., T. Cheasty, H. Chart, and B. Rowe. 1994. Isolation of Vero cytotoxin-producing Escherichia coli serotypes O9ab:H- and O101:H- carrying VT2 variant gene sequences from a patient with haemolytic uraemic syndrome. Eur. J. Clin. Microbiol. Infect. Dis. 13:1074-1076. [DOI] [PubMed] [Google Scholar]
- 19.Thomas, A., T. Cheasty, J. Frost, H. Chart, H. Smith, and B. Rowe. 1996. Verocytotoxin-producing Escherichia coli, particularly serogroup O157, associated with human infections in England and Wales: 1992-1994. Epidemiol. Infect. 117:1-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Willshaw, G. A., S. M. Scotland, H. R. Smith, and B. Rowe. 1992. Properties of verocytotoxin-producing Escherichia coli of human origin of O-serogroups other than O157. J. Infect. Dis. 166:797-802. [DOI] [PubMed] [Google Scholar]
- 21.Willshaw, G. A., T. Cheasty, H. R. Smith, S. J. O'Brien, and G. K. Adak. 2001. Verocytotoxin-producing Escherichia coli (VTEC) O157 and other VTEC from human infections in England and Wales: 1995 to 1998. J. Med. Microbiol. 50:135-142. [DOI] [PubMed] [Google Scholar]