Uropathogenic Escherichia coli (UPEC) in Jordan: Prevalence of urovirulence genes and antibiotic resistance

1 Introduction

Urinary tract infections (UTIs) frequently occur in humans. The uropathogenic Escherichia coli (UPEC) is significantly associated with etiologies of UTIs, including pyelonephritis and cystitis (Hannan et al., 2012). UPEC has many virulence factors that assist its colonization, invasion, and survival within the host urinary system (Behzadi et al., 2016; Bien et al., 2012; Mao et al., 2012). These factors include adhesins, siderophores, toxins, capsule production, and proteases. In addition, several autotransporter (AT) proteins, which are typically associated with phylogeny (Zude et al., 2014), are correlated with virulence and have been identified in UPEC (Restieri et al., 2007; Zalewska-Piatek et al., 2015). It is believed that AT proteins have structural features that permit their pass through biological membrane systems (Allsopp et al., 2010), in addition to adhesion, invasion, and biofilm formation during pathogenesis (Allsopp et al., 2012, 2010). Among the AT proteins is the serine protease toxin (vat), which is encoded by vat, and that is found in both UPEC and avian pathogenic Escherichia coli (APEC) isolates (Rossiter et al., 2015; Zhuge et al., 2013). FyuA encoding for a heme binding protein called yersiniabactin receptor, and yfcV encoding for one of the chaperone-usher fimbriae subunits that are associated with UPEC (Spurbeck et al., 2012). UPEC chuA gene is also encodes a heme binding protein that mediates heme uptake and transport an plays a role in UTIs (Torres et al., 2001).

The products of shiA, sisA and sisB, are involved in the pathogenesis of UPEC by down-regulating the innate inflammatory response during infection (Lloyd et al., 2009b; Mao et al., 2012). On the other hand, products of eco274 and sivH, only demonstrated a correlation with UTIs, but their roles in pathogenesis remain undefined. (Cusumano et al., 2010; Lloyd et al., 2009a).

While E. coli archetypal UPEC virulence genes (papG, cnfI, hlyA, iroN, and usp) have been extensively studied (Farshad et al., 2012; Jahandeh et al., 2015), several non-archetypal UPEC virulence genes (vat, fyuA, chuA, yfcV, sivH, shiA, sisA, sisB and eco274) need further investigation among clinical isolates.

In the present study, urinary E. coli isolates from Jordan were characterized by molecular techniques to identify the incidence non-archetypal virulence genes of UPEC. In addition, antibiotic susceptibility profile of the isolates was also examined. UTI management has become increasingly difficult because of the emergence of resistance to first line antimicrobials used for empiric therapy (Qin et al., 2013). Furthermore, better understanding of virulence mechanisms and pathogenesis of UTI (UPEC virulence determinants) found in clinical UPEC isolates will help in providing new strategies to treat and prevent UPEC infections.

2 Materials and methods

3 Results

Several E. coli isolates were highly resistant to several antimicrobial agents (Tables 2 and 3). Resistance ranged from 83% for augmentin to 0.4%, for ertapenem. Approximately half (49.8%) of the isolates were ESBL positive (data not shown). Among the isolates, 81.9% were resistant to three or more classes of antimicrobial agents (Table 3).

Using PCR, urovirulence genes were detected at the following rates among the isolates: shiA (92%), sisA (72%), eco274 (44%), sivH (36%), vat (27%), yfcV (25%), sisB (25%), chuA (20%), and fyuA (18%) (Table 4). Among the isolates, 67.4% were identified as UPEC, as they were positive for three or more of the urovirulence genes.

Finally, the profile of antibiotic resistance among the isolates was compared to that reported in other countries (Table 5). Overall, the study’s isolates had higher or similar resistance rates compared to those reported elsewhere (Table 5).

4 Discussion

In the current study, E. coli isolates involved in human UTIs were characterized by PCR to determine prevalence of UPEC virulence genes. In addition, antibiotic susceptibility profile of the isolates was also examined.

UPEC strains are responsible for the majority of uncomplicated urinary tract infections. Many putative virulence genes of UPEC have been previously identified. However, UPEC strains differ in the number and expression levels of virulence genes, leading to variations in ability to grow and persist within the urinary tract (Bien et al., 2012).

Prevalence of non-archetypal urovirulence genes of UPEC strains is poorly defined among UTI patients. One of the aims of the present study was to identify the urovirulence genes of UPEC strains isolated from UTI symptomatic patients. Specifically, the presence of sivH, shiA, sisA, sisB, eco274, chuA, yfcV, vat, and fyuA, was investigated utilizing PCR-based analyses. We considered an isolate to be a UPEC strain, if it possessed three or more of the aforementioned genes.

Statistically significant associations were identified between multiple virulence genes, as some genes were more likely to coexist with other genes. This may suggest the co-carriage of multiple virulence genes on the same plasmid. In general, the prevalence of urovirulence genes presented in the current study is consistent with those from other related studies from the USA and Taiwan (Mao et al., 2012). ShiA and sisA were the most prevalent at 92% and 72%, respectively, while fyuA was the least prevalent at 18%. The other urovirulence genes were prevalent in 20–44% of the isolates. The higher prevalence of sisA in the current study (72%) compared to sisB (25%) is consistent with other studies that reported higher sisA prevalence than sisB among UPEC isolates (Lloyd et al., 2009a; Lloyd et al., 2009b; Mao et al., 2012). This difference may be attributed to carriage of sisA mainly by extra-intestinal pathogenic E. coli strains and of sisB mainly by intestinal E. coli strains (Lloyd et al., 2009b).

Interestingly, SisA and SisB of CFT073 are homologs of ShiA of Shigella flexneri, and play a similar role in suppressing the host defense system during UTIs (Ingersoll and Zychlinsky, 2006; Lloyd et al., 2009b). Therefore, sisA and sisB which are located in pathogenicity islands, can be considered UPEC virulence factor genes. They are found in some but not all UPEC strains, which may explain the reported variability of inflammation among patients of UTIs (Lloyd et al., 2009b). The potential transcriptional regulator gene eco274 which likely encodes a virulence factor involved in pyelonephritis and urosepsis, was prevalent among approximately 44% of the isolates. SivH, which encodes a putative invasive protein, had a prevalence rate of 36%. Although, the specific urovirulence mechanisms of the products of the eco274 and sivH are yet to be identified, their association with UTIs highly suggests their identity as urovirulence genes.

FyuA, chuA, sisB, yfcV, and vat, were prevalent in 18–27% of the isolates. E. coli isolates carrying, fyuA, chuA, yfcV, and vat, were reported to colonize the urinary tract more efficiently than isolates lacking them (Spurbeck et al., 2012). In addition, the latter four genes are highly prevalent among UPEC (Mao et al., 2012; Spurbeck et al., 2012). The fyuA gene was absent from E. coli isolated from asymptomatic bacteriuria in healthy nonpregnant women (Srivastava et al., 2016). Expression of chuA gene was induced by growing UPEC strain CFT073 in urine (Alteri and Mobley, 2007). These findings confirmed the low prevalence FyuA, chuA, sisB, yfcV, and vat genes observed in the current study.

The products of chuA and fyuA, which were identified in a fifth of the isolates, are two receptors that contribute to UPEC pathogenesis during UTIs (Spurbeck et al., 2012). A study demonstrated that chuA and fyuA, contributed to systemic E. coli infections by functioning as heme and yersiniabactin receptors, respectively (Spurbeck et al., 2012).

A previous report has indicated a high correlation between vat (tsh) and UPEC infections but not to fecal E. coli strains (Heimer et al., 2004), and that vat contributed to development of systemic infections (Subashchandrabose et al., 2013). However, vat was not detected in 123 UPEC strains isolated from patients with UTIs (Momtaz et al., 2013). Therefore, vat, which was identified in 27% of this study’s isolates, might be considered as a UPEC virulence factor. More studies are required to understand the contribution of vat to UPEC pathogenesis.

In Jordan and in other countries, the improper prescription of antimicrobials by physicians, purchasing antimicrobials without prescription, and the misuse of antimicrobials by individuals, likely contributes to increasing drug resistance by bacteria (Al-Bakri et al., 2005). In addition, drug resistance among the Enterobacteriaceae is increasing worldwide. A specific concern is the increasing prevalence of ESBL-producing Enterobacteriaceae, which are involved in several difficult-to-control infection outbreaks (Canton et al., 2008). Overall, the study’s isolates had higher or similar resistance rates compared to those reported in Jordan (Abu Shaqra, 2001; Zatorski et al., 2015) and elsewhere (Aypak et al., 2009; Farshad et al., 2012; Sabir et al., 2014), which has the antimicrobial resistance profile of the isolates compared to that from other studies. The observed variations in resistance rates may be attributed to regional differences in strain prevalence and different standards and controls for prescription and use of antimicrobial agents.

According to the European antimicrobial resistance surveillance report which was published in 2011 (EARS-Net, 2012), E. coli resistance rates to aminopenicillins ranged from 39% in Finland, up to 71% in Bulgaria. Resistance to 3rd generation cephalosporins ranged from 4% in Lithuania, Sweden, and Norway, up to 30% in Slovakia, with a mean of 11%. Resistance to fluoroquinolones ranged from 11% in Finland, Sweden, and Norway, up to 42% in Cyprus and Italy. Resistance to aminoglycosides ranged from 3% in Iceland up to 24% in Romania, with a mean of 10%. All European countries had low resistant rates to carbapenems with a mean of <0.1% which is very close to our results for ertapenem at 0.4%. Our results demonstrated much higher resistance levels than those in Europe likely due to misuse and improper prescription of antimicrobials in Jordan. The rates of ESBL-producing isolates in Europe ranged from 68% in Slovakia up to 100% in Hungary, Lithuania, and Romania, with most other countries above 80% (EARS-Net, 2012). In contrast, our study isolates had a lower rate of ESBL phenotype, probably due to more prevalent use of 3rd generation cephalosporins in Europe, than in Jordan.

In conclusion, antimicrobial susceptibilities, and virulence factor profiles of E. coli isolates recovered from UTIs were determined. High rates of multi-drug resistance were reported.

Conflict of interest

The authors have no conflict of interest to declare.