Pathogenic features of urinary Escherichia coli strains causing Asymptomatic Bacteriuria during Pregnancy

Background Asymptomatic bacteriuria is one of the common problems in pregnancy. Pyelonephritis, preterm labor and low birth weight infants have been associated with bacterial infection. Urinary tract infection (UTI) during pregnancy is frequently associated with complications. An observational cross-sectional study including investigated the prevalence of virulence genes, antimicrobial resistance, and its relationship with phylogenetic groups among E. coli strains isolated from pregnant women with asymptomatic bacteriuria who referred to Hafez hospital, Shiraz, Iran. Material and Methods A total of 300 urine samples were screened for Escherichia coli strains. Susceptibility testing was determined by the disk-diffusion method. The phylogenetic groups and 13 virulence genes were identified by PCR. ESBL and AmpC producing isolates were detected using phenotypic methods. PCR was used to identify the bla TEM , bla SHV and bla CTXM genes in ESBL and AmpC-positive isolates. Results Our results revealed that among 300 urine samples, 105 (35%) were positive for E. coli . The data showed that the highest and the lowest resistance rates were observed against nalidixic acid (82.1%), and imipenem (2.8%), respectively. The prevalence of ESBLs and AmpC-β-lactamase, in the E. coli isolates was 41% and 9.5% respectively. bla CTXM was the commonest genotype (93%). Phylogenetic group distribution was as follow: B1 2.8%, A 14.2%, B2 61.9%, and D 4.6%. Our result showed that most of the virulence genes belonged to group B2 and also several virulence genes such as hlyA , cnf-1 , and papGII genes were positively associated with group B2. features

3 Background Pregnant women are typically screened for urinary tract infections (UTIs) in early pregnancy and those with bacteriuria are treated with antibiotics [1]. Its accurate and prompt diagnosis plays an important role in reducing the course of the disease by preventing renal failure following the ascent of the infection in the upper urinary tract [2].
During pregnancy, UTI might be present as asymptomatic bacteriuria or as symptomatic infection [3]. Most infections are caused by Enterobacteriaceae and the most common causative pathogen is E. coli. Uropathogenic Escherichia coli (UPEC) strains are responsible for 80-90% of all cases [4]. E. coli ability to colonize various sites is due in part to genome specific characteristics by acquisition or loss of genes encoding virulence factors and antibiotic resistance genes [5].The interaction between UPEC and epithelial cells is effected by several factors and complex phenomenon, involving various adhesins produced according to the stage of infection while its adherence to epithelial cells plays a critical role for a successful colonization and establishment. The severity of the disease is dependent on the expression of other genes encoding virulence factors [6]. Furthermore, an increasing trend in the spectrum and frequency of antimicrobial-resistant UTIs was observed in the past few years [7]. Not only E. coli resistance to various groups of antibiotics such as β lactams, aminoglycosides and flouroquinolones can be attributed to some genes, but it also intrinsically resistant to some special antibiotics [8]. Phylogenetic analyses have shown that E. coli strains falls into four main phylogenetic groups (A, B1, B2, and D), each of which has a unique panel of genes that characterize its own evolutionary pattern. Various studies have exhibited that groups B2 and D are proportionately higher in pathogenic samples and usually more virulent, whereas groups A and B1 tend to be found at higher rates in commensal samples [9,10].However, there are few studies on virulence factor genes (VFGs), antimicrobial resistance and pattern of 4 phylogenetic groups amongst the E. coli isolated from asymptomatic pregnant women in Iran. The objective of our study was investigated the prevalence of virulence genes, antimicrobial resistance, and its relationship with phylogenetic groups among E. coli strains isolated from asymptomatic pregnant women who referred to Hafez hospital, Shiraz, Iran.

Sample collection and identification
A midstream clean-catch urine sample were obtained from each participant. The urine samples were cultured on MacConkey agar (MA) plates, and plates were incubated incubated in an aerobic atmosphere at 37°C for 18 h. [11]. Bacterial isolates (300) were obtained from pregnant women diagnosed and confirmed by the clinical laboratory.

Antibiotic susceptibility testing (AST)
AST was carried out using the Kirby-Bauer disk diffusion technique as described previously, using single antibiotic disks consisting of sulphamethoxazole-trimethoprim and nitrofurantoin (NI, 300 µg) (Mast, UK). E. coli ATCC 25922 was used as the quality control strain for antibacterial susceptibility testing [12].
Klebsiella pneumonia ATCC 700603 and E. coli ATCC 25922 were used as positive and negative controls respectively. Furthermore, AmpC phenotype was specified by means of compound disk using cefoxitin (FOX), (Mast, UK) with and without boronic acid (Sigma-Aldrich Chemie, , Germany) were used to detect AmpC phenotypes [13].
According to CLSI criteria, all isolates classified as resistant to cefoxitin were suspected to be AmpC producers. An increase in the zone diameter of ≥5 mm in the presence of boronic acid is taken to be a phenotypic confirmation of AmpC production.

Determination of Escherichia coli phylogenetic groups
Isolates were assigned to one of the four main phylogenetic groups of E. coli (A, B1, B2, and D), using the triplex PCR as described by Clermont et al. [10]. The genes chuA, yjaA and TSPE4.C2 were amplified by PCR using the primers listed in Table 1. The E. coli strain ECOR62 and RS218 (B1 and B2 groups respectively) were used as a controls [16].

Statistical analysis
The data was analyzed using SPSS version 18.0 (SPSS Inc., Chicago, IL, USA). The Chisquare test or the Fisher exact test was used to compare categorical variables. A P-value less than 0.05 was considered to be statistically significant.

Results
A total of 300 urine samples were screened and confirmed 105 (35%) as uropathogenic E.
Out of the 105 E. coli isolates examined for β-lactamases, 41% (43) were found to be ESBL producers and 9.5% (10) were AmpC β-lactamase producers. All AmpC producers were also found to be ESBLs positive. Among the ESBL and AmpC-producing isolates, the multiplex PCR assay results indicated that 40(93%) bla CTX-M and 34(79%) bla TEM genes were detected in the E. coli isolates and bla SHV was not found in any of the isolates.

Discussion
In this study, the prevalence of asymptomatic bacteriuria in pregnant women (with recurrent UTIs before pregnancy) was 42%, which was similar to what was observed by a 8 study in Nigeria(47.5%) [17] and a study in Ghana (42.8%) [18]. In contrast, some other studies found lower prevalence than ours, a study in Chitwan, Nepal, conducted by Neupane [20] , and a study in India, conducted by Sujatha R et al. (7.3%) [21]. These difference might be due to factors including the geographical areas being investigated, the social habits of the communities, the socio-economic statuses, and standards of personal hygiene [11]. Quinolones, trimethoprim-sulfamethoxazole, and ß-lactam antibiotics include penicillins and cephalosporins are the most common antibacterial drugs in UTIs' treatment [22]. In this study, E. coli isolates were highly resistant to nalidixic acid, trimethoprim/sulfamethoxazole and ampicillin by more than 70%, and sensitivity values above 80% were found to imipenem, aztreonam and nitrofurantoin. Findings are in line with the earlier studies in Iran [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]. The findings of previous studies as well our current findings indicate that ESBL and AmpC-β-lactamase-producing isolates are typically resistant to others antibiotics such as trimethoprim/sulfamethoxazole and fluoroquinolones [13]. ESBL was detected in 41/105 (43%) of the isolates recovered from patients, and bla CTXM was the commonest genotype (93%). Also, all of the AmpC-producing (9.5%) isolates were ESBLpositive. ESBL-producing E. coli showed the greatest resistance to ampicillin, amoxicillin-clavulanic acid, ceftazidime and trimethoprim-sulphamethoxazole (Table 4). This finding is in agreement with previous studies [24]. This study showed a considerable number of bacteria to harbor the kpsMTII, fimH, papC, iutA, fyuA, traT and sfa/focDE, respectively, which was in contrast to Sáez-López et al., but in line with Forson et al. study on pregnant women in Barcelona [23] and Ghana [18]. In our study, kpsMTII was the only gene found in all 105(100%) isolates. Also, 95 out of 105(90.4%) isolates harbored, fimH genes. papC, papGII and ecp (A and RB) _ the other genes related to the ability to colonize the urinary tract epithelium were detected in 88.5% , 14.2% and 58.1% of the isolates, respectively. Our results are in agreement with those found by other studies [17,[18][19][20][21][22][23]. Also, among the tested VFs in our study, there was only sfaS gene at the lowest rate (2.8%). iutA and fyuA play an important role in iron acquisition systems by up-taking the hydroxamate siderophore aerobactin [25]. The prevalence of the iutA (83.8%) and fyuA (76.1%) genes found in our E. coli isolates correlates with the results published by Forson et al. [18]. Significant associations were observed between SXT resistance and the presence of the siderophores fyuA and iutA. At least one of the tested siderophores was present in 98 (93.3 %) of the tested isolates. Meanwhile, of the toxinencoding genes, hlyA was present in 12.3% of the isolates, while cnf1 was detected in 8.5% of the studied strains. Positive isolates of hlyA, and cnf-1genes were susceptible to ciprofloxacin, which was in line with that of Piatti et al [26]. The distribution of the 105 E.
coli isolates in relation to virulence genes in pregnant women revealed 76.1% (80 isolates) E. coli contained two or more virulence genes. The distribution of virulence genes and the phylogenetic classification are different among countries. For example, in Russia[27] , UTI isolates belonged more often to phylogenetic group A. In Spain and the United States, lower percentages were recorded for the phylogenetic group D [5]; however, in the present study, the most prevalent phylogenetic group was B2 (61.9%). Our result showed that most of the virulence genes belonged to group B2 and also several virulence genes such as hlyA, cnf-1, and papGII genes were positively associated with group B2. All of the 65 isolates in group B2 were positive for kpsMT KII and fimH genes. Strains belonging to group B2 carry more virulence-factor genes compared to strains from other phylogenetic groups, suggesting a relationship between virulence factors and pathogenic potential.
Group D contained the second highest number of E. coli strains. Extraintestinal pathogenic isolates from this group typically have somewhat fewer virulence factor genes and a different mix of VFGs than do group B2 isolates. In our study the lowest prevalence belonged to sfaS (2.8%) found in all 3 isolates of group D. This was in agreement with the report of López-Banda et al. [10]. Group A contains fewer virulence factors genes than group D. E. coli strains belong to this group that expresses kpsMTII, papC, fmH and iutA genes in high percentages. E. coli strains belonging to groups A and B1 do not    Table 4. Percentages of antibiotic resistant UPEC isolates. n (%).