Detection of phylogenetic group B1 Escherichia coli by multiplex PCR: Description of a new amplification pattern

Dear Editor:

Phylogenetic analysis of Escherichia coli isolates has enabled five major groups into this species to be identified, namely A, B1, B2, D and E. Human isolates of groups B2 and D are often involved in extra-intestinal infections, while those of groups A and B1 are frequently associated with commensal microbiota.² Characterisation of phylogenetic groups is of clinical interest. Typically, isolates of groups B2 and D had been reported to frequently have an increased number of virulence factors and decreased rates of antimicrobial resistance than those of group A or B1,^1–4 although more recent studies indicate that some particular clones of phylogenetic group B2 (ST131) or D (ST405) express a multi-resistant phenotype.⁵ Phylogenetic groups were initially recognised by multilocus enzyme electrophoresis,⁶ a rather complex and time-consuming approach. In 2000, Clermont et al.⁷ described a simple approach based on a triplex PCR detecting the genes chuA and yjaA and the DNA fragment TSPE4.C2. This method assigns the phylogenetic group in approximately 80% of the cases. We have used this triplex PCR to define the phylogenetic group of clonally unrelated (as defined by Rep-PCR) multi-resistant clinical isolates of E. coli (manuscript in preparation). In all, 100 E. coli producing extended-spectrum beta-lactamases (ESBL) and 100 isolates lacking ESBL and resistant to at least three of the antimicrobial agents: amoxicillin, nalidixic acid, cotrimoxazole and either or both gentamicin and tobramycin were evaluated. PCR was performed using primers and conditions described previously⁷ using a 2720 Thermal Cycle (Applied Biosystems, Foster City, CA, USA).

Fifty out of the 200 (25%) isolates [38 (76%) ESBL producers and 12 (24%) ESBL non-producers] showed an amplification pattern compatible with group B1, but in 42 (84%) isolates [31 (74%) ESBL producers and 11 (26%) ESBL non-producers] an extra band of approximately 550 bp was also observed (Fig. 1A). This band was not present in the control strains EcoR32 (B1), EcoRS218 (B2), EcoR16 (A) and EcoR39 (D). The experiment was repeated on two different occasions in different days using two different methodologies: in the first one, we used a short program with an annealing temperature (T_m) of 59 °C, and in the second one, we used a longer program with a T_m of 55 °C (both of them described by Clermont et al.⁷). The same results were obtained in the two assays. We are not aware of the description of this finding in previous reports related to phylogenetic group characterisation on E. coli.

Fig. 1.

(A) Results of the triplex PCR for phylogenetic group B1 of E. coli showing (arrow) the extra 550 bp band. (B) PCR assays with all other possible primer combinations for representative isolate belonging to phylogenetic group B1; (a) chuA1/yjaA1; (b) chuA1/tspE4C.1; (c) chuA1/chuA2; (d) chuA1/yjaA2; (e) chuA1/tspE4C.2; (f) yjaA1/tspE4C.1; (g) yjaA1/chuA2; (h) yjaA1/yjaA2; (i) yjaA1/tspE4C.2; (j) tspE4C.1/chuA2; (k) tspE4C.1/yjaA2; (l) tspE4C.1/tspE4C.2. M:100 bp DNA Ladder (Invitrogen, Barcelona, Spain).

In order to characterise the nature of the observed extra band, single PCR assays were performed using primers TspE4C2.1 and TspE4C2.2 and the extra band was not observed. New single PCR assays with all other possible primer combinations were also tested in one representative isolate (Fig. 1B).

The extra ca. 550 bp band was only observed when the primer pair chuA1 and TspE4C2.1 was used. Sequencing of the corresponding amplicon resulted in a 528 bp sequence 100% homologue with a lipase (acetyl-hydrolase) found in the pathogenic strains of E. coli belonging to phylogenetic group B1 [E. coli O111:H− (Genbank access number AP010960.1), E. coli O103:H2 (AP010958.1), E. coli O26:H11 (AP010953.1), E. coli IAI1 (CU928160.2), E. coli 55989 (CU928160.2), E. coli SE11 (AP009240.1) and E. coli E24377A (CP000800.1)]. In conclusion, we have documented that most isolates of E. coli of the phylogenetic group B1, the conventional triplex PCR assay may result in the amplification of a fragment of a gene coding for an acetyl-hydrolase. We suggest that this extra band could be used to characterise a new subgroup of phylogenetic group B1 (B1a), but additional studies should be done in order to evaluate the correlation of the possible subgroups with other typing methodologies such as MLST or PFGE.

Acknowledgements

This study was supported by Ministerio de Sanidad y Consumo, Instituto de Salud Carlos III-FEDER [Spanish Network for the Research in Infectious Diseases (REIPI RD06/0008) and FIS (PI05-0690)]. B. Ruiz del Castillo is supported by a grant from Instituto de Formación e Investigación Marqués de Valdecilla (IFIMAV). We acknowledge Dr. J.R. Johnson (Minneapolis, MN, USA) for helpful discussion.