Sorbitol-Fermenting,β-Glucuronidase–Positive,Shiga Toxin–Negative Escherichia coli O157:H7 in Free-Ranging Red Deer in South-Central Spain

Introduction

E scherichia coli O157:H7 infection has become an important public health problem of the developed world, causing concern not only because of its rise in incidence, but also because of the severity of its complications (Todd and Dundas, 2001). E. coli O157:H7 causes severe gastrointestinal and renal human diseases, ranging from mild diarrhea to hemorrhagic colitis and the hemolytic uremic syndrome (HUS), the leading cause of acute renal failure in children (Hughes et al., 2002). Nevertheless, non-O157 enterohemorrhagic E. coli (EHEC) serotypes have also been implicated in outbreaks of disease worldwide and are currently considered emerging pathogens by the World Health Organization. Unlike other E. coli strains, E. coli O157:H7 typically neither ferments sorbitol nor exhibits β-glucuronidase (GUD) activity, and these differences facilitate the detection of this organism. However, sorbitol-fermenting (SF) and GUD-positive strains have emerged as a cause of human diseases in continental Europe. Indeed, SF E. coli O157 accounts for 17% of sporadic cases of HUS and caused seven outbreaks in Germany between 1988 and 2009; a large outbreak also recently occurred in the United Kingdom (Werber et al., 2011). Some evidence indicates that virulence of SF E. coli O157 is probably greater, because it is more frequently associated with HUS than are non–sorbitol-fermenting (NSF) strains, and HUS patients infected with SF strains require more sessions of hemodialysis and have a higher case fatality rate (Werber et al., 2011).

Although the reservoirs and exposure routes of SF E. coli O157 are still unknown, they are likely to be different from those of NSF E. coli O157:H7 (Werber et al., 2011). Most E. coli O157:H7 infections in humans are foodborne, and the source of infection is an animal reservoir. Healthy cattle and other domestic ruminants are considered natural reservoir hosts of E. coli O157:H7 (Caprioli et al., 2005). However, E. coli O157:H7 strains have also been isolated from wild deer (Dunn et al., 2004; García-Sánchez et al., 2007), and deer have been implicated in the foodborne transmission of E. coli O157:H7 to humans in the United States (Rabatsky-Ehr et al., 2002; Ahn et al., 2009). Further, the occurrence of SF, GUD-positive E. coli O157:H7 strains in the feces of free-ranging deer has been previously reported in the United States (Dunn et al., 2004) and in Spain (García-Sánchez et al., 2007).

In Castilla-La Mancha, a region in south-central Spain, large game hunting and game meat consumption are popular and a major source of financial income. In the present study, our goal was to investigate the prevalence of E. coli O157:H7 in free-ranging red deer sampled from different game estates in the south-central region of Spain to assess their potential as reservoir hosts of SF E. coli O157:H7 strains.

Materials and Methods

Results and Discussion

E. coli O157:H7 was detected and isolated in 4 of the 25 game estates sampled (16%), with prevalence ranging from 6.7% (1/15) in 3 of the estates where positive samples were collected to 10% (1/10) in the fourth one. The overall sample-level prevalence from all 25 estates was 1.5% (4/264). The isolates obtained (four in total) came from four different samples, as only one isolate was obtained from each positive sample. One of them was biochemically typical of E. coli O157:H7, that is, neither fermented sorbitol nor exhibited GUD activity, and carried genes encoding Shiga toxins (Stx) 1 and 2, the intimin subtype γ1, and the EHEC-hemolysin (Table 1). This typical E. coli O157:H7 isolate carried the ter gene cluster and was able to grow on CT-SMAC (Table 1). Nevertheless, the rest of the isolates (three of four) fermented sorbitol, exhibited GUD activity, and carried genes encoding the intimin subtype γ1 and the EHEC-hemolysin, although no Stx-encoding genes were detected (Table 1). All these SF, GUD-positive, Stx-negative E. coli O157:H7 isolates carried the sfp gene cluster, lacked the ter gene cluster, and were unable to grow on CT-SMAC, which are typical features of SF strains isolated from patients, in contrast to NSF strains (Table 1). Therefore, SF, GUD-positive, Stx-negative E. coli O157:H7 strains were detected and isolated in 3 of the 25 game estates sampled (12%), with an overall sample-level prevalence of 1.1% (3/264).

The results of the present study indicate an appreciable prevalence of SF, GUD-positive, Stx-negative E. coli O157:H7 in game estates with free-ranging red deer in this area of Spain. This prevalence was higher than that observed in the only two previous studies that had reported the occurrence of such E. coli O157:H7 variants in the feces of free-ranging deer (Dunn et al., 2004; García-Sánchez et al., 2007). Several authors in Germany have reported that SF E. coli O157 strains can exist as Stx-negative variants during certain phases of their life cycle (Mellmann et al., 2008) and these Stx-negative derivatives of SF E. coli O157 are indeed emerging causes of HUS in Europe (Mellmann et al., 2008). Our findings contribute to earlier investigations identifying deer as a natural source of E. coli O157:H7 but above all indicate that free-ranging red deer may be one of the possible reservoir hosts of Stx-negative derivatives of SF E. coli O157:H7. Sources of human infection with these strains might include vegetables or water contaminated with deer feces but mainly undercooked game meat and meat products obtained from deer. Therefore, the findings of the present study have implications for the zoonotic risk to hunters, people consuming meat and meat products from deer, and others in contact with deer feces. However, further studies will be required to fully elucidate the epidemiology of the Stx-negative derivatives of SF E. coli O157:H7 in wildlife and the degree of zoonotic risk posed by these species.