SGM Journals
Research Article

Virulence factors and biofilm production among Escherichia coli strains causing bacteraemia of urinary tract origin

Matija Rijavec, Manca Müller-Premru, Breda Zakotnik, Darja Žgur-Bertok

  • 1Department of Biology, Biotechnical Faculty, University of Ljubljana, Ljubljana, Slovenia
  • 2Institute of Microbiology and Immunology, Medical Faculty, University of Ljubljana, Ljubljana, Slovenia
  • 3Department of Infectious Diseases, University Medical Center, Ljubljana, Slovenia
  • Correspondence
    Darja Žgur-Bertok
    darja.zgur.bertok{at}bf.uni-lj.si

    • Journal of Medical Microbiology 2008; 57(11):1329 · https://doi.org/10.1099/jmm.0.2008/002543-0

    View at publisher PubMed

    Abstract

    The aim of the present study was to gain an insight into the role of virulence determinants and biofilm production in bacteraemia of urinary tract origin. For this purpose 105 Escherichia coli isolates from patients with bacteraemia of urinary tract origin, isolated at the Institute of Microbiology and Immunology, University of Ljubljana, Slovenia, were investigated. A total of 88 strains (84 %) were isolated from immunocompromised patients and 17 (16 %) from non-immunocompromised patients. The prevalence of virulence factor (VF)-encoding genes and associations with phylogenetic background, antibiotic resistance, biofilm production and patient status were analysed by PCR and bioassay. Biofilm was produced by 55 (53 %) of the strains. No combination of VFs was highly associated with biofilm production. Of the tested VF-encoding genes, usp, papC and the adhesin-encoding sfa/foc were significantly more prevalent among strains from non-immunocompromised patients. Our results indicate that the uropathogenic specific protein (USP) may be, as judged by predominance and associations of the usp gene, an important VF contributing significantly to bacteraemia of urinary tract origin.

    INTRODUCTION

    Pathogenic Escherichia coli strains are a common cause of extraintestinal infections such as urinary tract infections, neonatal meningitis, as well as bacteraemia (Stamm & Hooton, 1993; Ronald, 2002). Strains that cause extraintestinal infections harbour virulence factors (VFs) that enhance the ability to cause systemic infection (Maslow et al., 1995; Johnson et al., 1998; Johnson & Stell, 2000; Ruiz et al., 2002). While an increase in the occurrence of E. coli bacteraemia and urosepsis has been reported in recent years (McBean & Rajamani, 2001), relatively few studies have investigated the characteristics of E. coli strains causing bacteraemia. Furthermore, it is possible that the virulence genotypes and phylogenetic background of E. coli differs in different geographical regions. While some previous studies have shown that bacteraemia in adults is caused by E. coli strains predominantly from phylogroups B2 followed by D, A and B1 (Johnson et al., 2002), Martínez et al. (2006) demonstrated a predominance of strains of the D phylogroup. The authors postulated that the relative predominance of different phylogroups may be subject to geographical variations. In addition, fully pathogenic group D strains, such as those of serotype O15 : K52 : H1, in certain geographical areas are endemic and exhibit an increasing involvement in bloodstream infections (Martínez et al., 2006). Therefore, the characteristics of isolates from various regions need to be assessed. The aim of the present study was to gain further insight into the role of virulence determinants and biofilm production in bacteraemia. For this purpose the prevalence and associations of VFs, with regard to patient status, phylogenetic background, antibiotic resistance and biofilm formation, among E. coli strains causing urinary bacteraemia from Slovenia were investigated. In addition, several previous investigations have shown that quinolone- and fluoroquinolone-resistant uropathogenic E. coli (UPEC) strains exhibit reduced virulence and invade immunocompromised patients, while susceptible strains are virulent and affect non-immunocompromised patients (Horcajada et al., 2005; Moreno et al., 2005). To gain more insight into the basis of this relationship we analysed the VFs, antibiotic resistances and phylogenetic group of the strains from Slovenia.

    METHODS

    Bacterial infections, isolates and media.

    A total of 105 E. coli isolates from patients with bacteraemia of urinary tract origin, isolated at the Institute of Microbiology and Immunology, Medical Faculty, University of Ljubljana, from patients admitted to various departments of the University Medical Center in Ljubljana, from 2000 and 2001, were studied. The large majority of the infections, 81 (77 %), were community acquired, and the majority of the patients, 88 (84 %), were immunocompromised, and had one or more underlying diseases, such as chronic urinary tract disease (83 patients), diabetes mellitus, malignant, rheumatologic or neurological disease, and/or had undergone immunosuppressive treatment (20 patients), had a urinary catheter or nephrostomy (15 patients), had undergone recent surgery (13 patients) or invasive diagnostic procedures (18 patients). There were 36 males (34 %) and 69 women (66 %) (data not shown). Only one isolate per patient was analysed. The strains were freshly cultivated onto nutrient agar (SIFIN) and Luria–Bertani agar (LB plates).

    Oligonucleotide primers and PCR for phylogenetic group analysis and detection of virulence genes.

    The phylogenetic grouping of the studied isolates was determined by multiplex PCR (Clermont et al., 2000). The primers and PCR conditions used to amplify adhesins, papC (encoding pilus associated with pyelonephritis), sfa/focDE (adhesin-encoding operons: the central region of the sfa/foc operon, encoding S fimbriae and F1C fimbriae), afa (encoding afimbrial adhesin I) (Le Bouguenec et al., 1992), papGIII, fimH (encoding type 1 fimbrial adhesin) (Usein et al., 2001); siderophores fyuA (encoding yersiniabactin) (Johnson & Stell, 2000), iucD (encoding aerobactin), iroN (encoding salmochelin) (Johnson et al., 2000); toxins hlyA (encoding α-haemolysin) (Yamamoto et al., 1995), cnf1 (encoding cytotoxic necrotizing factor 1) (Ambrožič et al., 1998), usp [encoding uropathogenic specific protein (USP)] (Nakano et al., 2001), and capsules K1, K5 (Nowrouzian et al., 2001), were the same as described in the references cited.

    Dot blot hybridization experiments were performed with the DIG DNA labelling and detection kit (Roche) to confirm the PCR assays. Appropriate positive and negative controls were performed.

    Biofilm production.

    The capacity to form biofilms was assayed in microtitre plates essentially as described by O'Toole & Kolter (1998) and Vieira et al. (2004) with slight modification. Briefly, cells were initially grown for 24 h in 7 ml M9 minimal medium at 37 °C with shaking at 120 r.p.m. Subsequently, 100 μl overnight culture was added to 96-well polystyrene microtitre plates and incubated for 24 h without shaking at 37 °C. Unattached bacterial cells were then removed from the culture medium, and the biofilm was stained with 0.2 % (w/v) crystal violet for 15 min (this dye stains the cells but not the polystyrene). The excess crystal violet dye was washed out, and this was followed by washing the samples three times with bidistilled water. To release the dye, 200 μl 96 % ethanol was added to the wells. Subsequently, 125 μl sample was transferred to another well, and the absorbance was measured at 595 nm to estimate the amount of biofilm formed. The experiments were performed in triplicate.

    Antibiotic susceptibility testing.

    Antimicrobial susceptibility testing to the antimicrobial agents ampicillin, ciprofloxacin and trimethoprim/sulfamethoxazole was performed by the disc diffusion method, as well as by MIC determination using microtitre plates (CLSI, 2007).

    Haemagglutination assays and haemolysis.

    Haemagglutination tests to determine mannose-sensitive haemagglutination (MSHA) and mannose-resistant haemagglutination (MRHA) were performed in duplicate by the microtitre method as described by Izumi et al. (2005). Production of α-haemolysin was tested on 5 % sheep blood agar. E. coli strains were inoculated onto blood agar plates, incubated overnight at 37 °C and haemolysis was detected by the presence of a zone of complete lysis of the erythrocytes around the colony.

    Statistical analysis.

    The significance of the results was established using the Fisher's exact test and the level of significance was set at a P value <0.05.

    RESULTS AND DISCUSSION

    E. coli is the leading cause of Gram-negative bloodstream infections; however, data are relatively scarce regarding strain characteristics, particularly from various geographical settings. In the present study, molecular, as well as biological, assays were employed to investigate the phylogenetic background, virulence-associated characteristics and biofilm production of 105 E. coli strains causing bacteraemia of urinary tract origin. The majority, 81 (77 %), of the strains examined here were from community acquired urinary bacteraemia. In addition, 88 (84 %) patients were immunocompromised.

    E. coli strains can be assigned to one of the four main phylogenetic groups: A, B1, D1 and B2 (Herzer et al., 1990). Extraintestinal pathogenic strains belong mainly to phylogenetic groups B2 and to a lesser extent D (Picard et al., 1999). Altogether 54 (51 %) of the studied isolates belonged to group B2, 21 (20 %) to group D, 16 (15 %) to group A and 14 (13 %) to the B1 group.

    As iron is limiting in the bloodstream, iron acquisition systems are important VFs. Of the studied siderophore-encoding genes, iucD was found in as many as 91 % of the tested strains, followed by fyuA found in 74 % strains and iroN in 56 % strains. At least one of the tested siderophores was present in 104 (99 %) of the tested isolates. Our results showed an unexpectedly high prevalence of aerobactin encoding sequences (91 %) among the investigated strains, compared to 79 % reported by Houdouin et al. (2006), and 78 % reported by Moreno et al. (2005), while all other studies reported lower percentages (Johnson et al., 2005; Soto et al., 2007). Such differences might be due to divergences in geographical variation or differences in association with host characteristics. Among the studied adhesins fimH was present in 95 % of the studied isolates, followed by papC in 55 % (papGIII 14 %), sfa/foc 24 % and afa in 3 % of the isolates. Of the toxin-encoding genes usp was present in 34 % of isolates, while hlyA and cnf1 were detected in 26 and 12 %, respectively, of the studied strains. Haemolysis and haemagglutination were employed to confirm production of VFs. A good correlation was found between the presence of hlyA and haemolysis, since 26 % of the tested strains harboured hlyA sequences and 27 % exhibited haemolytic activity. Other studies have revealed that MSHA is associated with the presence of type 1 fimbriae (Salit & Gotschlich, 1977; Duguid et al., 1979). While fimH was detected in 95 % of the investigated strains, only 85 % exhibited MSHA. This discrepancy might be due to phase variation (Hultgren et al., 1985). On the other hand MRHA activity can be mediated by P fimbriae, X, FIC and DR fimbriae (Johnson, 1991). Of the 63 strains exhibiting MRHA, 58 harboured papC, among these 20 also harboured sfa/foc, 4 strains harboured only sfa/foc and 1 only afa. Capsule K1- and K5-encoding genes were present in 26 and 11 % of the studied strains, respectively. Prevalences of the investigated VF-encoding genes were somewhat lower than reported by some other studies of E. coli strains causing bacteraemia, most probably because the majority of the strains were from immunocompromised patients.

    Biofilms are microbial communities of organisms adherent to each other and/or a target surface. Biofilm formation protects bacteria from hydrodynamic flow conditions, for example in the urinary tract, and against phagocytosis and host defence mechanisms, as well as antibiotics (Hanna et al., 2003). More than 50 % of all bacterial infections reported involve biofilm formation (Costerton et al., 1999). A cascade of several precisely, tightly regulated events are required for proper biofilm formation. A majority of the investigated strains in our study (56 %) were in vitro positive for biofilm production. The prevalence of biofilm production was thus higher than reported in another study: 17 % for faecal strains, 43 % for strains isolated from patients with cystitis, 40 % for pyelonephritis and 42 % for bacteraemic E. coli strains (Soto et al., 2007). On the other hand, the same authors reported a high, 63 %, prevalence of biofilm formation among strains from patients with prostatitis. In addition, the latter study showed that haemolysin and type 1 fimbriae expression were significantly associated with biofilm production. Type 1 fimbriae, which promote adhesion to host epithelial cells, have been found to be important in the initial steps of biofilm formation (Prüss et al., 2006). Among the strains investigated in our study biofilm production was not statistically associated with any virulence determinant or combination of virulence determinants (data not shown). Biofilm formation thus was not statistically associated with the multivirulent B2 phylogenetic group but was associated with phylogroups B1 and D (for group B1/D vs non-phylogroup B1/D, P=0.007) (data not shown).

    Prior studies have tried to determine the minimal requirement for bacterial passage into the bloodstream. It was demonstrated that the chromosomal determinants for P fimbriae (Johnson et al., 1988), aerobactin and haemolysin are conserved in antibiotic susceptible UPEC strains that invade non-immunocompromised patients. Two additional studies indicated that papC, fyuA and aer (iucD) (Johnson & Stell, 2000; Bingen-Bidois et al., 2002) represent the minimal prerequisite for bacterial passage and infection of the bloodstream. A recent investigation (Horcajada et al., 2005) indicated that the papG allele II is a critical trait for reaching the kidney, while cnf1 and sfa/focDE are critical for bacteraemia of urinary tract origin. In addition, previous investigations have indicated that a greater complement of VFs is needed for invasion of a non-immunocompromised host than for an immunocompromised one (Maslow et al., 1993; Johnson, 1994; Johnson et al., 1988; Otto et al., 2001). Among the tested VFs in our study only usp, papC and the adhesin-encoding sfa/foc exhibited a statistically significant higher prevalence among strains from non-immunocompromised patients (Table 1⇓). Among the investigated strains those from non-immunocompromised patients exhibited a statistically significant higher prevalence of multivirulence (MV) (MV≥3 and MV≥5) than strains from immunocompromised patients (i.e. more non-immunocompromised patients harboured ≥3 and ≥5 of the tested virulence traits).

    Table 1.

    Host characteristics in relation to phylogenetic group, biofilm production, VFs and antibiotic resistance

    The virulence score (VF score) was calculated for each isolate as the sum of all virulence-associated genes detected (papC, sfa/foc, afa, hlyA, cnf1, usp, fyuA, iroN, iucD, K1, K5), the sum of all the VF scores of the isolates was then calculated, and finally this sum was divided by the number of isolates to give the mean VF score.

    Additionally, it should be noted that strains from phylogenetic group A and B1 were almost exclusively isolated from immunocompromised patients, as was also reported by Moreno et al. (2005), while Bingen-Bidois et al. (2002) found no association between phylogenetic group and patient status. Strains from phylogenetic group B2 were slightly more common among non-immunocompromised patients. Similarly, biofilm-producing strains were also found more frequently in this patient group. However, none of the differences was statistically significant (Table 1⇑).

    Analysis of the correlation between antibiotic resistance and patient status revealed that only resistance to ampicillin was statistically associated with immunocompromised status. Nevertheless, resistant strains, as well as multidrug resistance (MDR) strains, were more often, albeit not significantly, isolated from immunocompromised patients, which is not surprising as such patients are more often treated with antibiotics (Table 1⇑).

    Several previous investigations have shown that ciprofloxacin resistance and MDR are associated with a shift towards less virulent strains and non-B2 phylogenetic groups (Houdouin et al., 2006; Moreno et al., 2005; Horcajada et al., 2005; Johnson et al., 2005). Thus, as expected, susceptible strains from this study exhibited a higher VF score. In addition, our results showed that sensitivity to ciprofloxacin was significantly associated with papC, hlyA, usp and fyuA, but not with cnf1, and not with the aerobactin iron uptake system (Table 2⇓) as determined in a recent study (Piatti et al., 2008). The association between sensitivity to ciprofloxacin and MV (Table 2⇓) was also highly significant. A slightly higher, albeit not statistically significant, capacity to form biofilms was determined among Cipr isolates, 75 % (12), compared to 49 % (46) among Cips isolates. No correlation between biofilm production and resistance/sensitivity to trimethoprim/sulfamethoxazole or ampicillin was observed (data not shown).

    Table 2.

    MDR and ciprofloxacin resistance in correlation with VFs

    The virulence score (VF score) was calculated for each isolate as the sum of all virulence-associated genes detected (papC, sfa/foc, afa, hlyA, cnf1, usp, fyuA, iroN, iucD, K1, K5), the sum of all the VF scores of the isolates was then calculated, and finally this sum was divided by the number of isolates to give the mean VF score. MDR was determined as resistance to antibiotics belonging to at least two different classes of antimicrobials (penicillins, antifolates and quinolones/fluoroquinolones).

    The underlying basis for the association of susceptibility to antibiotics and virulence is still unclear. It is possible that a low presence of certain VFs precedes resistance (Johnson et al., 2005; Moreno et al., 2005) or that low presence of these VFs is followed by acquisition of resistances (Piatti et al., 2008; Vila et al., 2002; Horcajada et al., 2005). The coexistence of both mechanisms is also plausible (Houdouin et al., 2006; Drews et al., 2005). A recent study indicated that a specific chromosomal background, only partially corresponding to the phylogenetic background could precede mutation to antibiotic resistance (Drews et al., 2005).

    The most relevant finding of our study is that usp, USP-encoding sequences, even though detected at a relatively low prevalence among the investigated strains, exhibited a statistically significant association with bacteraemia of urinary tract origin among non-immunocompromised patients. Previous studies have shown USP to be a VF more frequently found in UPEC than in faecal strains: 79.4 % from cystitis, 93.4 % from pyelonephritis and 88.8 % from prostatitis compared to only 24 % from faecal strains from healthy individuals (Yamamoto et al., 2001; Kanamaru et al., 2003). usp sequences were also found to be highly associated with phylogenetic group B2 (Kanamaru et al., 2006). Furthermore, Parret & De Mot (2002) reported that USP itself could be a bacteriocin acting as an endonuclease, since the usp gene showed high homology to genes encoding DNase-type pyocins and colicins. Recently, in a mouse model, it has been shown that usp, as well as malX, a pathogenicity-island marker, pap (encoding P fimbriae) and fyu (encoding the yersiniabactin system), were found to be most closely associated with killer strains (Johnson et al., 2006). The prevalence of the studied isolates harbouring papC, hly, cnf-1 and aer was comparable to that reported by other studies. However, those authors did not test for the prevalence of usp and association with patient status.

    In conclusion, our study indicates, on the basis of the distribution of the usp gene, that USP may be a significant VF with an essential role in contributing to bacteraemia of urinary tract origin in patients with a normal status. USP-encoding sequences were frequently associated with highly virulent strains belonging to phylogroup B2 and with infections in patients with normal, non-immunocompromised status.

    Acknowledgments

    We thank Damijan Nipič and Gavrilo Hadžič for their skilful assistance. This research was supported by grant PO-0508-0487 from the Slovenian Research Agency.

    References

    1. Ambrožič, J., Ostroveršnik, A., Starčič, M., Kuhar, I., Grabnar, M. & Žgur-Bertok, D. (1998). Escherichia coli ColV plasmid pRK100: genetic organization, stability and conjugal transfer. Microbiology 144, 343–352.
    2. Bingen-Bidois, M., Clermont, O., Bonacorsi, S., Terki, M., Brahimi, N., Loukil, C., Barraud, D. & Bingen, E. (2002). Phylogenetic analysis and prevalence of urosepsis strains of Escherichia coli bearing pathogenicity island-like domains. Infect Immun 70, 3216–3226.
    3. Clermont, O., Bonacorsi, S. & Bingen, E. (2000). Rapid and simple determination of the Escherichia coli phylogenetic group. Appl Environ Microbiol 66, 4555–4558.
    4. CLSI (2007). Performance Standards for Antimicrobial Susceptibility Testing, 17th informational supplement, M100–S17:1–177. Wayne, PA: Clinical and Laboratory Standards Institute.
    5. Costerton, J. W., Stewart, P. S. & Greenberg, E. P. (1999). Bacterial biofilms – a common cause of persistent infections. Science 284, 1318–1322.
    6. Drews, S. J., Poutanen, S. M., Mazzulli, T., McGreer, A. J., Sarabia, A., Pong-Porter, S., Rzayev, Y., Willey, B., Green, K. & Low, D. F. (2005). Decreased prevalence of virulence factors among ciprofloxacin-resistant uropathogenic Escherichia coli isolates. J Clin Microbiol 43, 4218–4220.
    7. Duguid, J. P., Clegg, S. & Wilson, M. I. (1979). The fimbrial and non-fimbrial haemagglutinins of Escherichia coli. J Med Microbiol 12, 213–227.
    8. Hanna, A., Berg, M., Stout, V. & Razatos, A. (2003). Role of capsular colanic acid in adhesion of uropathogenic Escherichia coli. Appl Environ Microbiol 69, 4474–4481.
    9. Herzer, P. J., Inouye, S., Inouye, M. & Whittam, T. S. (1990). Phylogenetic distribution of branched RNA-linked multi-copy single-stranded DNA among natural isolates of Escherichia coli. J Bacteriol 172, 6175–6181.
    10. Horcajada, J. P., Soto, S., Gajewski, A., Smithson, A., Jiménez de Anta, M. T., Mensa, J., Vila, J. & Johnson, J. R. (2005). Quinolone-resistant uropathogenic Escherichia coli strains from phylogenetic group B2 have fewer virulence factors than their susceptible counterparts. J Clin Microbiol 43, 2962–2964.
    11. Houdouin, V., Bonacorsi, S., Bidet, P., Bingen-Bidois, M., Barraud, D. & Bingen, E. (2006). Phylogenetic background and carriage of pathogenicity island-domains in relation to antibiotic resistance profiles among Escherichia coli urosepsis isolates. J Antimicrob Chemother 58, 748–751.
    12. Hultgren, S. J., Porter, T. N., Schaeffer, A. J. & Duncan, J. L. (1985). Role of type 1 pili and effects of phase variation on lower urinary tract infections produced by Escherichia coli. Infect Immun 50, 370–377.
    13. Izumi, E., Domingues Pires, P., Bittencourt de Marques, E. & Suzart, S. (2005). Hemagglutinating and hemolytic activities of Enterococcus faecalis strains isolated from different human clinical sources. Res Microbiol 156, 583–587.
    14. Johnson, J. R. (1991). Virulence factors in Escherichia coli urinary tract infection. Clin Microbiol Rev 4, 80–128.
    15. Johnson, J. R. (1994). Pathogenesis of bacteremia during pyelonephritis. Clin Infect Dis 18, 1014–1015.
    16. Johnson, J. R. & Stell, A. L. (2000). Extended virulence genotypes of Escherichia coli strains from patients with urosepsis in relation to phylogeny and host compromise. J Infect Dis 181, 261–272.
    17. Johnson, J. R., Mosely, S. L., Roberts, P. L. & Stamm, W. E. (1988). Aerobactin and other virulence factor genes among strains of Escherichia coli causing urosepsis: association with patient characteristics. Infect Immun 56, 405–412.
    18. Johnson, J. R., Brown, J. J. & Maslow, J. N. (1998). Clonal distribution of the three alleles of the gal(α1–4)gal-specific adhesin gene papG among Escherichia coli strains from patients with bacteremia. J Infect Dis 177, 651–661.
    19. Johnson, J. R., Russo, T. A., Tarr, P. I., Carlino, U., Bilge, S. S., Vary, J. C., Jr & Stell, A. L. (2000). Molecular epidemiological and phylogenetic associations of two novel putative virulence genes, iha and iroNE.coli, among Escherichia coli isolates from patients with urosepsis. Infect Immun 68, 3040–3047.
    20. Johnson, J. R., Kuskowski, M. A., O'Bryan, T. T. & Maslow, J. N. (2002). Epidemiological correlates of virulence genotype and phylogenetic background among Escherichia coli blood isolates from adults with diverse-source bacteremia. J Infect Dis 185, 1439–1447.
    21. Johnson, J. R., Kuskowski, M. A., Gajewski, A., Soto, S., Horcajada, J. P., Jimenez de Anta, M. T. & Vila, J. (2005). Extended virulence genotypes and phyelonephritis or prostatitis. J Infect Dis 191, 46–50.
    22. Johnson, J. R., Clermont, O., Menard, M., Kuskowski, M. A., Picard, B. & Denamur, E. (2006). Experimental mouse lethality of Escherichia coli isolates, in relation to accessory traits, phylogenetic group and ecological source. J Infect Dis 194, 1141–1150.
    23. Kanamaru, S., Kurazono, H., Ishitova, S., Terai, A., Habuchi, T., Nakano, M., Ogawa, O. & Yamamoto, S. (2003). Distribution and genetic association of putative uropathogenic virulence factors iroN, iha, kpsMT, ompT and usp in Escherichia coli isolated from urinary tract infections in Japan. J Urol 170, 2490–2493.
    24. Kanamaru, S., Kurazono, H., Nakano, M., Terai, A., Ogawa, O. & Yamamoto, S. (2006). Subtyping of uropathogenic Escherichia coli according to the pathogenicity island encoding uropathogenic-specific protein: comparison with phylogenetic groups. Int J Urol 13, 754–760.
    25. Le Bouguenec, C., Archambaud, M. & Labigne, A. (1992). Rapid and specific detection of pap, afa, and sfa adhesin-encoding operons in uropathogenic Escherichia coli strains by polymerase chain reaction. J Clin Microbiol 30, 1189–1193.
    26. Martínez, J. A., Soto, S., Fabrega, A., Almela, M., Mensa, J., Soriano, A., Marco, F., Jimenez de Anta, M. T. & Vila, J. (2006). Relationship of phylogenetic background, biofilm production, and time to detection of growth in blood culture vials with clinical variables and prognosis associated with Escherichia coli bacteremia. J Clin Microbiol 44, 1468–1474.
    27. Maslow, J. N., Mulligan, M. E., Adams, K. S., Justis, J. C. & Arbeit, R. D. (1993). Bacterial adhesins and host factors: role in the development and outcome of Escherichia coli bacteremia. Clin Infect Dis 17, 89–97.
    28. Maslow, J. N., Whittam, T. S., Gilks, C. F., Wilson, R. A., Mulligan, M. E., Adams, K. S. & Arbeit, R. D. (1995). Clonal relationships among bloodstream isolates of Escherichia coli. Infect Immun 63, 2409–2417.
    29. McBean, M. & Rajamani, S. (2001). Increasing rates of hospitalization due to septicemia in the US elderly population, 1986–1997. J Infect Dis 183, 596–603.
    30. Moreno, E., Planells, I., Prats, A. M., Moreno, G. & Andreu, A. (2005). Comparative study of Escherichia coli virulence determinants in strains causing urinary tract bacteremia versus strains causing pyelonephritis and other sources of bacteremia. Diagn Microbiol Infect Dis 53, 93–99.
    31. Nakano, M., Yamamoto, S., Terai, A., Ogawa, O., Makino, S., Hayashi, H., Nair, G. B. & Kurazono, H. (2001). Structural and sequence diversity of the pathogenicity island of uropathogenic Escherichia coli which encodes the USP protein. FEMS Microbiol Lett 205, 71–76.
    32. Nowrouzian, F., Adlerberth, I. & Wold, A. E. (2001). P fimbriae, capsule and aerobactin characterize colonic resident Escherichia coli. Epidemiol Infect 126, 11–18.
    33. O'Toole, G. A. & Kolter, R. (1998). Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol Microbiol 28, 449–461.
    34. Otto, G., Magnusson, M., Svensson, M., Braconier, J. & Svanborg, C. (2001). pap genotype and P fimbrial expression in Escherichia coli causing bacteremic and nonbacteremic febrile urinary tract infection. Clin Infect Dis 32, 1523–1531.
    35. Parret, A. H. & De Mot, R. (2002). Escherichia coli's uropathogenic-specific protein: a bacteriocin promoting infectivity? Microbiology 148, 1604–1606.
    36. Piatti, G., Mannini, A., Balistreri, M. & Schito, A. M. (2008). Virulence factors in urinary Escherichia coli strains: phylogenetic background and quinolone and fluoroquinolone resistance. J Clin Microbiol 46, 480–487.
    37. Picard, B., Garcia, J., Gouriou, S., Duriez, P., Brahimi, N., Bingen, E., Elion, J. & Denamur, E. (1999). The link between phylogeny and virulence in Escherichia coli extraintestinal infection. Infect Immun 67, 546–553.
    38. Prüss, B. M., Besemann, C., Denton, A. & Wolfe, A. J. (2006). A complex transcription network controls the early stages of biofilm development by Escherichia coli. J Bacteriol 188, 3731–3739.
    39. Ronald, A. (2002). The etiology of urinary tract infection: tradition and emerging pathogens. Am J Med 113 (Suppl. 1A), 14S–19S.
    40. Ruiz, J., Simon, K., Horcajada, J. P., Velasco, M., Barranco, M., Roig, G., Moreno-Martínez, A., Martínez, J. A., Jiménez de Anta, T. & other authors (2002). Differences in virulence factors among clinical isolates of Escherichia coli causing cystitis and pyelonephritis in women and prostatitis in men. J Clin Microbiol 40, 4445–4449.
    41. Salit, I. E. & Gotschlich, E. C. (1977). Hemagglutination by purified type I Escherichia coli pili. J Exp Med 146, 1169–1180.
    42. Soto, S. M., Smithson, A., Martinez, J. A., Horcajada, J. P., Mensa, J. & Vila, J. (2007). Biofilm formation in uropathogenic Escherichia coli strains: relationship with prostatitis, urovirulence factors and antimicrobial resistance. J Urol 177, 365–368.
    43. Stamm, W. E. & Hooton, T. M. (1993). Management of urinary tract infections in adults. N Engl J Med 329, 1328–1334.
    44. Usein, C.-R., Damain, M., Tatu-Chitoiu, D., Capusa, C., Fagaras, R., Tudorache, D., Nica, M. & Le Bouguénec, C. (2001). Prevalence of virulence genes in Escherichia coli strains isolated from Romanian adult urinary tract infection cases. J Cell Mol Med 5, 303–310.
    45. Vieira, H. L., Freire, P. & Arraiano, C. M. (2004). Effect of Escherichia coli morphogene bolA on biofilms. Appl Environ Microbiol 70, 5682–5684.
    46. Vila, J., Simon, K., Ruiz, J., Horcajada, J. P., Velasco, M., Barranco, M., Moreno, A. & Mensa, J. (2002). Are quinolone-resistant uropathogenic Escherichia coli less virulent? J Infect Dis 186, 1039–1042.
    47. Yamamoto, S., Terai, A., Yuri, K., Kurazono, H., Takeda, Y. & Yoshida, O. (1995). Detection of virulence factors in Escherichia coli by multiplex polymerase chain reaction. FEMS Immunol Med Microbiol 12, 85–90.
    48. Yamamoto, S., Nakano, M., Terai, A., Yuri, K., Nakata, K., Nair, G. B., Kurazono, H. & Ogawa, O. (2001). The presence of the virulence island containing the usp gene in uropathogenic Escherichia coli is associated with urinary tract infection in an experimental mouse model. J Urol 165, 1347–1351.