Genotypic Characterization of Enterotoxigenic Escherichia coli Strains Causing Traveler's Diarrhea

Enterotoxigenic Escherichia coli (ETEC) is by far the most common pathogen causing traveler's diarrhea (TD) worldwide, being responsible for up to 30 to 60% of all TD cases (1, 2). ETEC adheres to and colonizes the human intestinal mucosa thanks to specific antigenic fimbriae called colonization factors (CFs) (2, 3). Following CF-mediated mucosal adhesion, ETEC elaborates heat-labile toxin (LT) and/or heat-stable toxin (ST), causing secretion of water and electrolytes (4).

Inactivated enterotoxins and some CFs have been used as vaccine candidates in clinical studies (5, 6, 7), but none of them has yielded results in clinical trials sufficiently robust to foster completion of clinical development. Moreover, the great variability of ETEC CFs requires the determination of prevalent CF types in different geographic locations and population types (2, 7). An alternative approach for vaccine development would be to focus on ETEC antigens other than enterotoxins. Other virulence factors (VFs) have been identified from the prototype strain ETEC H10407 (O78:H11/LTI-STh-STp/CFA/I). These “nonclassical” VFs include adhesins (Tia, TibA), a cytoplasmic protein with GTPase activity (LeoA), an autotransporter (EatA), and an enteroaggregative E. coli heat-stable enterotoxin (EAST1) (8). Additionally, the E. coli common pilus (ECP), common to both commensal and pathogenic E. coli, may be present, but its role in ETEC pathogenesis is not yet established (9). To our knowledge, these genes have not been systematically evaluated in epidemiological studies in ETEC strains associated with TD. In this study, we characterized VFs of ETEC causing TD in Spanish travelers abroad.

Fifty-two ETEC isolates causing TD obtained between January 2004 and August 2011 as part of a passive TD surveillance study in Barcelona, Spain, were grown from frozen stock. From these, clinical data were recovered in 30 cases. TD was defined as the occurrence between 12 h after arrival in, and 5 days after departure from, the country visited; of three or more episodes of watery diarrhea within a 24-h period with or without other symptoms; or as the occurrence of unformed stools accompanied by one of the following: abdominal cramps, tenesmus, vomiting, nausea, fever, chills, or prostration. The presence of E. coli and other bacterial enteropathogens was determined by conventional methods, while PCR was used to detect enteroaggregative E. coli (EAEC), enteropathogenic E. coli (EPEC), and ETEC (10). The presence of classical (LT, STh, STp, CFA/I, CS1, CS2, CS3, CS4, CS5, CS6, CS7, CS8, CS12, CS14, CS17, CS18, CS19, CS20, and CS21) and nonclassical (Tia, LeoA, TibA, EatA, EAST1, and ECP) VFs was determined by PCR as previously described (11, 12, 13, 14, 15, 16) (Table 1). Chi-square or Fisher exact tests were used for comparisons between groups as appropriate.

ETEC strains were isolated from patients mostly returning from Africa (15/30, 50%), Asia/Pacific (10/30, 33%), and Central America (4/30, 13%). The median duration of travel was 21 days (range, 7 to 300 days). ST-positive (ETEC-ST) strains (23/52, 44%, including STh and STp) were the most frequent, followed by strains with both LT and ST (ETEC-LT-ST) (14/52, 27%) and strains positive for only LT (ETEC-LT) (15/52, 29%) (Table 2). STh (estA⁺) was the toxin type most frequently identified. This study employs travelers returning with diarrhea. Since most ETEC infections are brief and self-limited, returning travelers will be biased toward more severe cases and/or cases of longer duration. This may actually make the analysis more significant since it may reveal the factors common among the more severe traveler's pathogens, but validation in another study would be required for verification.

The most common CFs were CS21 (58%), CS6 (27%), and CS3 (23%) (Table 2). There were no differences between the geographical area of travel and the CF types identified, with CS21 (Longus) being the most common CF in all areas. The presence of multiple CFs in a given isolate was frequent, these CFs being frequently associated with CS21. The STh⁺ CS21 genotype (with or without other CFs) was the most prevalent among all strains (44%, 23/52). The ECP was detected in 81% of all strains. Meanwhile, the most common nonclassical VFs were EAST1 (65%), autotransporter EatA (48%), and adhesin Tia (21%) (Table 2).

TD owing to infection with ETEC is a considerable problem for travelers in high-risk regions. In several studies, the most prevalent CFs associated with the occurrence of diarrhea worldwide are CFA/I, CS1, CS2, CS3, and CS6; more recently, CS8, CS14, and CS21 have been detected with relatively high frequencies (6). In contrast with some studies performed in developing countries, where CFA/I has been described as the most prevalent CF, accounting for up to 21% of samples (2), in our study, CFA/I was present only in 12% of isolates. Thus, the prevalence of CF antigens may be related to geographical origin.

Among nonclassical VFs, EAST1 (65%, 34/52) and EatA protein (48%, 25/52) were frequently detected in our strains. The presence of EAST1, structurally related to STI peptides, which also leads to increases in cyclic GMP (cGMP) (17), in multiple strains may suggest functional redundancy of toxins with the capacity to provoke elevated levels of cGMP (8). EatA acts as a serine protease, belonging to the SPATE family of proteins (serine protease autotransporters of Enterobacteriaceae). Although a precise function or substrate for EatA is not clear, initial studies in ileal loops suggested that EatA could contribute to accelerated virulence of eatA⁺ strains (18).

Despite our searching for the CFs most commonly described in the literature, 31% of all ETEC strains did not possess any of the tested CFs. In a prior study, Blackburn et al. reported that 58% of ETEC strains were positive for the Escherichia coli common pilus (ECP), a percentage even higher than that of the most prevalent CFs and independent of the presence of CFs, suggesting an important role for ECP in the biology of ETEC, particularly in CF-negative strains and in human infections (9). In our study, the ECP gene (ecpA) was identified in 81% (42/52) of the present isolates. In recent years, a variety of different structures (other than colonization factors) encoded either on plasmids or in the chromosome of ETEC have been identified as putative adhesins. Among these adhesins, Tia is an outer membrane protein that interacts with host cell surface proteoglycans (19) and by itself is sufficient to promote adherence and epithelial cell invasion when cloned into laboratory strains of E. coli.

The success of an ETEC vaccine targeting travelers from industrialized countries to developing countries will depend on a combination of maximally antigenic vaccine preparations and regimens for their delivery which will produce optimal immune responses. While vaccine development to prevent diarrheal illness due to ETEC is feasible, extensive efforts are needed to identify new conserved antigenic targets. Additional studies will be needed to determine the utility of these antigens as well as other autotransporters in ETEC vaccines.

A.M.M. attended a 3-month training course through the CRESIB-UPCH collaborative agreement; training was partially funded by Consejo Nacional de Ciencia, Tecnología e Innovación Tecnológica (CONCYTEC, FONDECYT) in Peru. This work was partially funded by the Agencia Española de Cooperación Internacional para el Desarrollo (AECID), Spain, Programa de Cooperación Interuniversitaria e Investigación Científica con Iberoamérica (D/019499/08, D/024648/09, D/030509/10, and A1/035720/11); Departament d'Universitats, Recerca i Societat de la Informació de la Generalitat de Catalunya, Spain 2009SGR685 (J.R.) and 2009SGR385 (J.G. and E.A.); and National Institutes of Health, Public Health Service award 1K01TW007405 (T.J.O.); J.R. has a fellowship from program I3 of the ISCIII (grant no. CES11/012)

There is no conflict of interest for any of the authors.