IDENTIFICATION OF VIRULENCE FACTORS AMONG ESBL-PRODUCING ESCHERICHIA COLI CLINICAL ISOLATES FROM GAZA STRIP, PALESTINE

This study was done to assess the occurrence of virulence factors and correlation of phylogenetic grouping in ESBLs producing E. coli isolates from Palestine.
Twenty-seven ESBLs producing E. coli isolates were collected between April and June 2013 from three hospitals in Gaza. Detection of genes encoding virulence factors, pathogenicity associated island marker (PAI) and phylogenetic groups were studied by PCR. The correlation of E. coli phylogenetic grouping with the virulence factors in the isolates was studied. The clonal relationships between the isolates were tested by pulsed-field gel electrophoresis (PFGE).
Overall, 44.4% of the 27 E. coli isolates belonged to phylogroup B2, 44.4% to D and 11.2% to A. Among the isolates, fimH, traT and fimA were the most frequent virulence genes and were found over 85% of the isolates. PAIs was found in 8 isolates while the tcpC gene was detected in one isolate. The overall prevalence of PAI was higher in group B2 as compared to groups A and D (p<0.043). A clonal diversity was confirmed among our isolates (27 unrelated PFGE profiles). We report for the first time the prevalence of tcpC as a new virulence marker and PAI in clinical ESBLs-producing E. coli in Palestine.
This study indicates that most of the ESBL- positive isolates showed virulence genes and most strains belonged to phylogenetic groups B2 and D. The study suggested that high number of virulence genes in studied strains may be important factors in the infections development.


INTRODUCTION
Escherichia coli are common inhabitant microorganism of the human intestinal and various animals. However, some E. coli strains can cause urinary tract and blood-stream infections (Pitout, 2010) , (Manges, 2016). Extended-spectrum βlactamases (ESBLs) are enzymes produced by Gram-negative bacteria, confer and increased resistance to β-lactam antibiotics such as ceftazidime, cefotaxime and aztreonam (Paterson and Bonomo, 2005). The overall data on ESBLcontaining E. coli in many countries of Middle East are worrying and this area may be one of the considerable epicenters of the ESBL pandemic globally (Tayh et al., 2016a). The E. coli capacity enhances by many virulence factors (VFs) to cause infections; extraintestinal E. coli strains (ExPEC) carry genes encoding various VFs that can affect processes of host cell such as host defense avoidance mechanisms (capsule or O-specific antigen), combinations of adhesins (S and P fimbriae), toxins (cytonecrotizing and hemolysin factor), and iron acquisition systems (yersiniabactin and aerobactin) (Ananias and Yano, 2008).VFs are important factors in the infections development by different mechanisms such as; help bacteria to attach and invade host cells, lyses host cells via toxins, evade host defenses and sequestering iron from the host which important for bacterial cellular activities. Genes coding for VFs are located with each other on large chromosomal regions recognized as pathogenicity islands (PAIs) (Bingen-Bidois et al., 2002).
Extraintestinal pathogenic E. coli differ with commensal according to phylogenetic attributes and virulence determinants (Johnson and Stell, 2000). Recently, studies indicate that the commensal strains belong to phylogenetic groups B1 and A. However, ExPEC belong particularly to group B2 and, to a lower extent, phylogenetic group D (Zhang et al., 2002) (Duriez et al., 2001). Toll/interleukin-1 (IL-1) receptor (TIR) domain plays an important role in the mammalian to activate innate immunity. Pathogenic microorganisms improved mechanisms to inactivation the TLR dependent host defense and to raise the bacterial virulence for a host (Nagarjuna et al., 2015a). Recently, the TIR homologous protein TcpC has been detected in E. coli strains acts as a virulence factor by inhibiting innate host responses, enhancing persistence of bacteria and increase infection severity. In a previous study, they confirmed a role of tcpC in the urinary tract infection pathogenesis in E. coli but its role in sepsis pathogenesis is unknown (Cirl et al., 2008). In other bacteria, TIR domain contain protein (BtpB) is recognized as a virulence factor that control of local inflammatory responses in the enhancement of chronic brucellosis (Salcedo et al., 2013). SaTlp1 and SaTlp2 are virulence factor of Staphylococcus aureus that interact in mammalian cells with the innate immune signaling (Spear et al., 2012). The virulence factors in ESBL-positive E. coli isolates were studied from different cases in different countries, from healthy school children in India (Zhang et al., 2002), from long-term hospitalized patients in China (Zhao et al., 2015) and diarrheic cattle in France (Wayne, 2014). The objective of this work was to investigate phylogenetic group and major VFs among ESBLs producing E. coli isolates from Palestinian hospitals in Gaza Strip. This study is the first investigation of the prevalence of virulence factors, PAI and tcpC among ESBLs producing E. coli isolates and of the association of virulence factors with the phylogenetic group in ESBLs producing E. coli isolates in Palestine.

Isolation and identification
The samples were obtained from patients visiting Palestinian hospitals in Gaza Strip and were transferred to the laboratory for processing. The isolated bacteria were cultured on plates of blood agar and MacConkey agar. Among the clinical isolates were obtained from urinary tract infection and wound infection, 69 were identified as E. coli by standard biochemical tests. These isolates were confirmed This study was done to assess the occurrence of virulence factors and correlation of phylogenetic grouping in ESBLs producing E. coli isolates from Palestine. Twenty-seven ESBLs producing E. coli isolates were collected between April and June 2013 from three hospitals in Gaza. Detection of genes encoding virulence factors, pathogenicity associated island marker (PAI) and phylogenetic groups were studied by PCR. The correlation of E. coli phylogenetic grouping with the virulence factors in the isolates was studied. The clonal relationships between the isolates were tested by pulsed-field gel electrophoresis (PFGE). Overall, 44.4% of the 27 E. coli isolates belonged to phylogroup B2, 44.4% to D and 11.2% to A. Among the isolates, fimH, traT and fimA were the most frequent virulence genes and were found over 85% of the isolates. PAIs was found in 8 isolates while the tcpC gene was detected in one isolate. The overall prevalence of PAI was higher in group B2 as compared to groups A and D (p<0.043). A clonal diversity was confirmed among our isolates (27 unrelated PFGE profiles). We report for the first time the prevalence of tcpC as a new virulence marker and PAI in clinical ESBLs-producing E. coli in Palestine. This study indicates that most of the ESBL-positive isolates showed virulence genes and most strains belonged to phylogenetic groups B2 and D. The study suggested that high number of virulence genes in studied strains may be important factors in the infections development.
by a species PCR of the uidA gene which encoding the β-glucuronidase (Jouini et al., 2010). E. coli strains were stored in 30% glycerol at -80 °C for further use.

Antibiotic susceptibility testing
Susceptibility testing was performed according to CLSI by the disc diffusion method for the following antimicrobial agents: amoxicillin-clavulanic acid, ampicillin, ceftazidime, cefoxitin, gentamicin, cefotaxime, amikacin, nalidixic acid, tobramycin, ciprofloxacin, kanamycin, imipenem, tetracycline, trimethoprim-sulfamethoxazole, and chloramphenicol (Bio-Rad, France). (Wayne, 2014). The E. coli isolates were examined for ESBL production by double-disk synergy method by using a disk of amoxicillin/ clavulanic acid with two antimicrobial disks (ceftazidime and cefotaxime) ( Kaur et al., 2013), the ESBL producing isolates were selected for more examinations. E. coli ATCC 25922 was used as a control strain.

Isolation of genomic DNA and PCR amplification
All isolates of E. coli confirmed by biochemical and molecular tests were subcultured on tryptone soy broth at 37° C for 18h. The extraction of the DNA was performed by the standard sodium acetate precipitation method. The concentration and the purity of DNA were determined using a NanoDrop™ spectrophotometer at A260/280 nm. PCR amplification was performed in 25 μL reaction volume containing 2 μL DNA template, 1× buffer, 1.5 mM MgCl2, 0.4 mM dNTPs, 0.6 µM forward and reverse primers and 1U Taq DNA polymerase (Biomer). PCR was done with a DNA thermal cycle (applied biosystems thermal cycler). The PCR reaction condition was as follows: initial denaturation at 94 °C for 4 min; 30 cycles of denaturation at 94 °C for 45 sec, annealing for 45 sec at specific temperature (Table 1), extension at 72°C for 45 sec; and a final extension (72 °C, 10 min).

Phylogenetic determination in ESBL-producing E. coli
PCR amplification was used to detect the presence of two genes (chuA, yjaA) and a fragment of DNA known as TspE4C2 to detect phylogenetic grouping (A, B1, B2 and D) of E.coli isolates. Primers sequences are given in Table 1. The interpretation of results was done by dichotomous decision tree (Zhang et al., 2002).

Data analysis
The data of phylogenetic grouping and VFs in ESBLs producing E. coli isolates were analyzed by the Statistical Package for the Social Sciences (SPSS) version 17 software (IBM Corporation, Somers, NY). Data comparison was achieved via analysis of Pearson's Chi-square. The level of statistical significance was set at P< 0.05.

Pulsed-field gel electrophoresis (PFGE) analysis
Genomic relatedness of the 27 ESBLs positive E. coli isolates was determined by PFGE using a restriction enzyme XbaI. The resulting restriction patterns of genomic DNA were analyzed with visual method and by GelCompar II software using the UPGMA algorithm and the Dice similarity coefficients (Slama et al.,  2011).

RESULTS
In this study we screened 27 ESBL-containing strains from urine (n=19) and wound swabs (n=8) from three Palestinian hospitals for the determination of phylogenetic groups, and important VFs associated sequences responsible for extraintestinal pathogenesis.

Antibiotic resistance among the isolates
All 27 E. coli isolates were tested to antimicrobial susceptibility against fifteen antibiotics. The percentage of resistance to antibiotics among ESBL positive isolates is shown in Figure 1. All isolates were resistance to cefotaxime and ampicillin. Resistant to sulfamethoxazole/trimethoprim, tetracycline, nalidixic acid, ciprofloxacin, ceftazidime and kanamycin were 77.8%, 74.1%, 66.7%, 59.3%, 51.9% and 51.9% respectively. Imipenem, amikacin and gentamicin were the most effective antibiotics with the studied isolates, where 85.2%, 81.5% and 74.1% were susceptible to imipenem, amikacin and gentamycin respectively. Sensitivity to amoxicillin-clavulanic acid, cefoxitin, tobramycin and chloramphenicol among ESBL producers were 74.1%, 74.1%, 63% and 63% respectively. Most of the isolates were found to be insensitivity to at least one agent in three or more antimicrobial classes, so they are considered as multidrugresistant (MDR).

Phylogenetic groups and clonal relationship of isolates
The analysis of phylogenetic grouping was classified into four groups (A, B1, B2, and D). Our study observed that ESBLs positive E. coli isolates belonged to the phylogroups B2 (N=12;44.4%), D (N=12;44.4%), and A (N=3; 11%) whereas none of the isolates belonged to phylogroup B1. The PFGE analysis of the E. coli isolates demonstrated a wide genetic diversity.

Correlation of phylogroups with VFs in ESBLs producing E. coli
The correlation of phylogenetic grouping with the virulence factors in different infections has been previously reported. The frequency of the virulence genes (fimA,papG III,cnf1,hlyA,aer,papC,sfa,afa,traT,fimH,eae,and bfp,tcpC and PAI) in the phylogenetic groups (A, B1, B2 and D) was studied. The prevalence of PAI in our isolates was higher in group B2 compared to A and D groups (p<0.043). However, no clear difference in the prevalence of afa, fimH, sfa, traT and tcpC in the E. coli isolates among various phylogenetic groups was seen. TcpC was found in one E. coli isolate and was restricted to phylogenetic group B2 (Figure 3). afa was found to be prevalent more in group D (N=6) than the group B2 (N=2). However, aer was found positive in 6 isolates for group B2 whereas 2 belong to group D (table 3 and figure 3).  They are the products of different genes, which can be detected by the PCR method. In this study, we have studies the antibiotics resistance, phylogenetic grouping and VFs, in 27 ESBL containing E. coli isolated from urine and wound swabs from three Palestinian hospitals. The relationship between the presence of VFs and the phylogenetic groups was also analyzed. The ESBL-containing E. coli isolates show co-resistance to many categories of commonly used antibiotics causing in limitation associated with therapeutic (antibiotics) options. Whereas imipenem was the most effective antibiotics with the studied isolates. Similar reports of gram-negative bacteria were reported from burn units in Palestine (Elmanama et al., 2013; Tayh et al., 2016b). The antibiotics resistance results are very significant for the physicians to choose an appropriate antimicrobial treatment. The higher proportion of resistance and ESBL rate in our study may be a result of the availability of antibiotics without a prescription. Reduction of antibiotic resistance can be by using antibiotics according antimicrobial stewardship guidelines, performing diagnostic testing and antimicrobial susceptibility testing, as well as via developments of new antibiotic (Lee et al., 2013). Extraintestinal E. coli which cause infections have been shown to belong to phylogroups D and B2 (Smith et al., 2007). The results of this study indicated that most isolates belonged to phylogroups B2 (44.4%) and D (44.4%) which is in agreement with previous findings (Slama et al., 2011). Group A was the least frequently isolated phylogenetic group among our isolates, which is in accordance with similar studies (Girardini et al., 2012).
In a recent study in the West Bank, forty-one E. coli isolates were obtained from UTI hospitalized patients in three Palestinian hospitals and the results revealed that 13 isolates belonged to phylogroup B2, 12 isolates to group D, 11 isolates to A and five isolates to B1 (Adwan et al., 2014). In a Turkish study, ESBLpositive E. coli were distributed in groups D (35%), A (35%) and B2 (30%) (Yumuk et al., 2008). In the Canadian study, it was detected that the majority of ESBL-producing isolates were belonged to group D (63%), while 21% of isolates were derived from A and 13% from group B2 (Pitout et al., 2005). PAI is highly prevalent among E. coli isolates causing extraintestinal infections and mostly belonged to phylogenetic groups D and B2 (Östblom et al., 2011).
Our study was proved the prevalence of eleven PAI in E .coli strains isolated from the wound infection and urine with the high prevalence of PAI in group B2. Herzer et al. (Herzer et al., 1990)  . The invasion factor encoded by the aer gene (Iron uptake) was demonstrated in 29.6% of E. coli isolates. The high effectiveness of iron uptake system is mediated by the siderophore aerobactin. The existence of a siderophore may be a significant factor in the infection development (Kuhnert et al., 2000). The frequency of aer gene was detected in 30% of the E. coli isolates causing cystitis in a Spinach study, and this is in agreement with our findings (Ruiz et al.,  2002). However, in a study in France, aer gene was higher than that of our study (80%) (Bingen-Bidois et al., 2002). In another study, that has been done in a Tunisian hospital, it was observed that aer gene was found in 85% of ESBLcontaining E. coli strains (Slama et al., 2011). In our isolates most of aer positive isolates belongs to group B2 and this is in agreement with previous study (Bingen-Bidois et al., 2002).
The TcpC gene was detected in only one isolate. TcpC was first detected by Cirl et al. and they found that tcpC homologous sequences were confirmed in 40% and 21% of E. coli strains from pyelonephritis and cystitis respectively. Their findings proposed that tcpC increases the severity of urinary tract infection in humans and they provided the first proof that E.coli can survive and spread in the host by interfering with TLR signaling (Cirl et al., 2008). The rate of tcpC was lower among our isolates, in comparison with the finding of other studies; In Slovenia, they reported that they detected tcpC in 49 (23%) of the pathogenic E. coli isolates (Erjavec et al., 2010). In a recent study, they investigated the prevalence of the tcpC gene in blood and fecal E. coli strains from India. They found the prevalence of tcpC gene in the phylogenetic groups B2 and D was higher (40.3%) than the B1 and A groups (9.6%) (Nagarjuna et al., 2015a).

CONCLUSION
In conclusion, we investigated phylogenetic grouping and VFs profile in ESBLsproducing E. coli isolates from Palestine. To the best of our knowledge, this is the first report tcpC as new virulence marker and PAI in clinical ESBLs-positive E. coli in Palestine. The results indicate that E. coli from phylogenetic groups B2 and D were predominate in our isolates. Most of the ESBL-positive E. coli strains showed virulence genes. FimH, traT and fimA were highly prevalent among the isolates. This study exhibited that a high