e Yersinia High-Pathogenicity Island in Escherichia coli and Klebsiella pneumoniae Isolated from Polymicrobial Infections

We examined 12 pairs of strains of Escherichia coli and Klebsiella pneumoniae isolated from mixed infections in human for the presence of the Yersinia high-pathogenicity island (HPI). In one case both isolates carried the HPI, whereas in 11 cases one strain of the pair was HPI-positive. Although there were di$erences in the organization of the Yersinia HPI, all HPI-positive isolates were able to produce yersiniabactin. €e presence of the Yersinia HPI may enhance the capability of strains involved in mixed infections to replicate in irondeprived conditions in the host. K e y w o r d s: HPI, mixed infections, siderophores, yersiniabactin * Corresponding author: R. Koczura, Department of Microbiology, Faculty of Biology, A. Mickiewicz University, ul. Umultowska 89, 61-614 Poznań, Poland; phone: +48 61 8295939; fax: +48 61 8295590; e-mail: [email protected] Koczura R. et al. 1 72 and irp2 RP specific for Y. pestis irp1 and irp2 genes, respectively. Detailed characteristics of the Yersinia HPI was done with a set of PCR reactions with primers complementary to single HPI genes as well as to regions that contained fragments of consecutive genes (Table I). PCR amplifications were done in a 25-μl volume with 2 μl of template DNA isolated with the boiling lysate method (Johnson and Brown, 1996), 2.5 μl of 10 × PCR bu$er (700 mM Tris-HCl, pH 8.6, 166 mM NH 4 (SO 4 ), 25 mM MgCl 2 ), 0.25 μM of each primer, 200 μM of dNTP mix, and 1 U of HiFi Taq polymerase (all reagents provided by Novazym). €e PCR conditions and the sequences of primers were applied according to Karch et al. (1999). PCR products were separated in 1.5% agarose gel, stained with ethidium bromide, visualized under UV and photographed with Bio-Print V.99 system (Vilber-Lourmat, France). All experiments were performed in duplicate. Production of yersiniabactin was detected in a crossfeeding assay (Reissbrodt and Rabsch, 1988) with Yersinia enterocolitica 5030, an indicator strain unable to produce the siderophore but capable of using exogenous yersiniabactin, and Y. enterocolitica 5092, a negative control strain that neither produces nor utilizes yersiniabactin (Haag et al., 1993). It has been previously demonstrated that strains of E. coli and K. pneumoniae can harbour the Yersinia highpathogenicity island (Bach et al., 2000), a genomic island that determines production of yersiniabactin, a siderophore considered to be a virulence factor (Schubert et al., 2002). In the present study, we examined 24 strains of E. coli and K. pneumoniae isolated from 12 polymicrobial infections. €e isolates were screened for yersiniabactin genes. In 11 cases, one strain of a pair (9 E. coli and 2 K. pneumoniae isolates) was positive for the presence of both irp1 and irp2 genes that code for proteins involved in yersiniabactin synthesis. In one case – E. coli 10B and K. pneumoniae 10A – both isolates harboured the HPI genes. Our previous studies have revealed diversity of the HPI among E. coli and K. pneumoniae clinical isolates (Koczura and Kaznowski, 2003a; Koczura and Kaznowski, 2003b). To examine whether there are differences in the HPI structure of E. coli 10B and K. pneumoniae 10A strains, we performed a set of PCR reactions with primers complementary to single HPI genes as well as to regions containing the fragments of consecutive genes. €e results are shown in Table I. €e HPI was located in the vicinity of asnT asparagine-specific tRNA gene. Both isolates failed to give amplification product of irp1/ybtT region. PCR amplification of K. pneumoniae 10A strain was also negative for ybtS/ybtQ region. A di$erence in the structure may suggest that both strains acquired the HPI separately and not by a horizontal gene transfer between them during the infection. To ensure that the HPI-positive isolates can produce functional yersiniabactin, we carried out biological assay with indicator strains. All strains that were positive for irp1 and irp2 genes were able to promote the growth of Y. enterocolitica 5030 indicator strain in iron-deficient conditions, whereas none of them induced the growth of Y. enterocolitica 5092 negative control. €is suggests that the lack of PCR product for regions involved in yersiniabactin biosynthesis, i.e.: ybtS, irp1, and ybtT/fyuA, were probably due to minor alterations of target sequences, as it did not a$ect production of the siderophore. €e ability to produce yersiniabactin is a virulence factor in Yersinia spp. (Carniel, 2001). It has been also shown that it contributes to the virulence of E. coli (Schubert et al., 2002) and K. pneumoniae (Lawlor et al., 2007). Moreover, Hancock et al. (2008) have demonstrated that E. coli strains require the yersiniabactin receptor FyuA for efficient biofilm formation, probably due to superb efficiency of yersiniabactin-mediated iron uptake system in conditions of high cell density and very low iron concentration. Summing up, we showed that E. coli and K. pneumoniae strains involved in mixed infections can produce the same siderophore – yersiniabactin, which may enhance their survival and capability to grow in irondeprived conditions in the infected host and facilitate replication during polymicrobial infections. asnT-int (asnT and int2) 150