We did not find lpfA1 variants 1, 3–5, or lpfA2 variant 2 in any of the strains studied. The four lpfA-negative STEC strains identified in our study were of human origin (serotypes O8:H16, O117:H7, ONT:H4 and ONT:HNM). Two of them, serotypes O8:H16 and
ONT:H4, were isolated from HUS cases and the only putative virulence factor currently identified in these strains is encoded by the iha gene. The ONT:HNM strain was isolated from a patient suffering from diarrhea and the only virulence factor found in this serotype is encoded by the stx1 toxin gene. Finally, the O117:H7 strain was isolated from an asymptomatic carrier with prolonged shedding and, unusually, it was nonsorbitol fermenting and carried the putative virulence factors iha and astA. In the current study, we could not find a statistical association between the presence of a particular lpfA variant and the severity of the disease. However, we observed that most R428 serotypes maintained the same combination of lpf variants independent of the source of isolation. Therefore, we observed a good association between the lpfA variant and the serotype of the strain, i.e. we identified two strains from serotype O22:H8 that carried lpfA1-2 and lpfA2-1, and two strains from serotype O22:H16 that carried only lpfA2-1. Interestingly, we found that these strains
belonging to the same serogroup and, isolated from cattle, shared the same virulence profiles (Table 1). One more isolate from serotype O22:HNT, Bumetanide which was isolated
from a human diarrheal case, selleck inhibitor carried the lpfA1-2 and lpfA2-1 genes. Similar results occurred in the O174 serogroup, where all the O174:H21 isolates carried the lpfA1-2 and lpfA2-1 gene variants, whereas the other O174 serotypes (O174:H8, O174:H28, O174:NM) carried the lpfA2-1 gene and no common theme with respect to the virulence profile or the source of isolation was observed. The most variable serogroup with respect to the lpfA gene variant content was O8, from which we identified three O8:H16 and four O8:H19 isolates. In the case of the O8:H16 isolates, two were lpfA1-2 and lpfA2-1-positive and carried the same virulence profile, whereas a third isolate was lpfA-negative and iha was the only putative virulence marker. Another difference in these strains, apart from the source of isolation, was the stx genotype; whereas the lpfA-negative strain was stx2 positive, the others were stx1/stx2 positive. In the case of the O8:H19 isolates, two carried the lpfA1-2 and lpfA2-1 genes, and two strains carried only the lpfA2-1gene. Further, all the strains of this serotype carried different virulence gene profiles. Another serotype identified in our study was O178:H19, which included two strains sharing the same origin and carrying the same stx gene and a common virulence profile, but differing in the type of lpfA variant present. One strain was lpfA1-2 positive, whereas the other was both lpfA1-2 and lpfA2-1-positive.