E. coli has been found as the most common causative agent of UTIs in both community and healthcare settings (21). The extraordinary genetic diversity in E. coli drives the emergence of strains harboring unusual arrangement of virulence genes, including genes from different pathotypes. The significance of such strains to cause infection is not only dependent on their virulence factors, but also on risk factors, such as older age, immunosuppression, frequent use of antimicrobials, and prolonged urinary catheterization in hospitals (21). Here, we characterized 130 E. coli isolates from patients with UTIs during IP or OP therapy in terms of phylogroup, presence of IPEC or UPEC virulence genes, and antimicrobial resistance.
The relatively lower prevalence of E. coli in our IP isolates is attributable to the more pronounced diversity in the causative agents of nosocomial UTIs (8). In line with previous studies (8, 22, 23), there was a predominance of females, among the affected population in both groups. A number of risk factors, including physiological and anatomical changes contribute to the higher UTI prevalence in women (21, 23). The most common phylogroup in our isolates from both IPs and OPs was B2, in agreement with previous studies showing a high prevalence of this group in ExPEC strains (23–25). Isolates of group B1 were not detected in our study in contrast to other studies (3, 22, 24), which may due to the differences in the populations that comprised the study groups. Although it was not our aim, studies have shown that extraintestinal isolates belonging to group B2 are associated with higher virulence and antibiotic resistance compared to other groups such as F or B1. The phylogenetic groups of E. coli have a good correlation with the ESBL production and the virulence potential of isolates (22, 24, 26).
Some studies found that virulence genes are more prevalent among UTI isolates from OPs compared to those obtained from IPs (10). Regarding the virulence markers detected in our isolates, no significant differences could be found between the IP and OP isolates. The cnf was the most common virulence gene in the OPs, while in the IPs, the pap was more prevalent. In several studies, the cnf showed a lower prevalence than the hly and pap (6, 22). The rate of non-virulent isolates was almost similar between the two groups.
Despite the frequent genetic exchange among E. coli strains, there are few studies on the presence of specific virulence traits related to the IPEC pathotypes in UTI-associated strains. In the study of Toval et al., 10.9% of the urine isolates carried IPEC virulence genes, which of them, 89.3% were isolated from IPs(21). In our study, IPEC virulence genes were detected in 49.4% of the OPs and in 51% of the IPs. The eae gene representing EPEC pathotype was not found in the IPs. This adhesin is able to accumulate actin in kidney cells and it may play a certain role in UPEC pathogenesis (6). Additionally, there was a widespread presence of the bfp in our isolates from both groups, suggesting that it may play an important role in the adherence of UPEC to the host. Moreover, we found intermediary strains with characteristics of both IPEC and UPEC. Isolates from the OPs showed relatively more diversity than those from the IPs, regarding the virulence properties of the IPEC. The emergence of novel E. coli variants due to a combination of pathotype-associated traits represents a serious health issue, as was demonstrated by a food-borne outbreak caused by a Shiga toxin-producing EAEC strain (27) or the documentation of an UTI outbreak caused by an EAEC strain(28). The finding that approximately 50% of our isolates from hospitals and the community exhibited virulence properties of IPEC indicates that IPEC virulence markers could be also considered in the risk assessment of UPEC. Regarding the uropathogenic ability of some diarrheagenic genes such as the EAE adhesin or Shiga toxins (6, 29), whether these genes are expressed in vivo and contribute to infection of the human urinary tracts remain to be clarified.
No significant differences could be found between the IP and OP strains regarding virulence factors in the present study and the distribution of virulence factors in both groups was almost similar. It is sometimes assumed that IP isolates may be more virulent than OP isolates, partly because resistance plasmids, selected by antibiotic use, may also code for virulence factors. The similarity in virulence between the IP and OP isolates may indicate that selection of virulent strains takes place as much in the community as in hospitals(30).
Treatment of E. coli infections has been increasingly complicated by the emergence of resistance to the critically important antimicrobials(8). Our isolates showed high resistance rates to some of the first-line empirical therapies (fluoroquinolone and TS) in the community and hospitals. This could be attributed to their inappropriate use in both settings due to their availability and low price (31, 32). Meanwhile, low resistance rates to IMI and FT were observed, indicating these antibiotics as appropriate therapies for UTI in both settings (23, 32, 33). Research suggests that uropathogens isolated from IPs are more resistant to antibiotics than those obtained from OPs (8, 33, 34). A greater proportion of our IP samples showed resistance to antimicrobials than the OP samples and ESBL-producing isolates were more frequently found in the hospital patients, although it was not significant. However, we still found around 36% of ESBL resistance in E. coli isolated from the OPs, which indicate that they are also exposed to high antibiotic selective pressure. In this context, OPs may serve as a source amplifying the spread of resistant E. coli to hospitals. Although carbapenem resistance in our isolates was limited (approximately 10% in each setting), this constitutes a serious threat to health care systems because carbapenems in many cases constitute the last line of therapy and the overuse of these drugs will lead to selection pressure (8, 23). Over the years, we have focused on antibiotic consumption in hospitals as a risk factor for development of antibiotic-resistant bacteria. Special attention should also be paid to consumption of antibiotics in community in order to control the prevalence of antimicrobial resistance.
The emergence of resistance to multiple antibiotics complicates therapy of E. coli -associated UTIs. This difficulty has increased due to the presence of virulence genes in UPEC strains, enhancing the pathogenicity of strains(24). We found associations between virulence and resistance in the isolates from the OPs and the IPs. Carrying virulence genes can confer some benefit on bacteria during the infection processes, which favors the resistant strains (9). The signaling by concentration of antibiotics in an environment plays a predominant role in bacterial responses and evolution in that niche and the most resistant pathogens will be selected. In some cases, there is a positive relationship between the acquisition of resistance and virulence; in other examples, the increased resistance is associated to lower virulence of the microorganisms; and there is also another option, in which there is no significant impact on virulence (35). Future research will reveal whether or not bacteria are becoming more virulent and more resistant in the community or hospitals.
Conclusion
In summary, E. coli isolates from the IPs showed similar virulence and antimicrobial resistance compared to the isolates from OPs. Our results may indicate that consideration should also be given to restrictive antibiotic policy and hygienic standard in the community in order to control the prevalence of antimicrobial resistant and virulent strains. Pathogenic E. coli strains have significant genomic plasticity and many unique genomic traits. In this context, continuous surveillance of virulence and antimicrobial resistance patterns of isolates in different regions is essential. Since resistance and virulence is attributed to local epidemiological factors and uncontrolled use of antimicrobials, further studies involving a large number of cases from different areas are needed.
Statements & Declarations