Antibiotic resistance, phylogenetic typing, and virulence genes profile analysis of uropathogenic Escherichia coli isolated from patients in southern Iraq

Of the most common infectious diseases that occur mainly by uropathogenic Escherichia coli (UPEC) is urinary tract infections (UTIs). The purpose of this study was to investigate virulence factors, antibiotic resistance, and phylogenetic groups among UPEC strains isolated from patients with UTI in southern Iraq. A total of 100 UPEC isolates were collected from urine samples of UTI patients from various hospitals in southern Iraq, and confirmed by morphological and biochemical tests. Antimicrobial susceptibility testing on isolates was performed by disk diffusion method. Multiplex PCR techniques were used to evaluate the phylogenetic groups based on Clermont method and to detect the presence of six virulence factor genes. The majority of isolates belonged to the phylogenetic groups B2 (46%) and C (13%). The most prevalent virulence factors were fimH (96%), followed by aer (47%), papC (36%), cnf1 (17%), hly (15%), and afa (8%). Phenotypic testing showed that the isolates were most resistant to piperacillin, ticarcillin, amoxicillin/clavulanic acid (92%, 91%, and 88%, respectively) and most sensitive to amikacin and imipenem, respectively. The maximum antibiotic resistance and virulence factors were observed in the phylogenetic group B2. The results showed that the UPEC isolates had all six virulence factors with high frequency and the highest drug resistance. Besides, the results showed a direct relationship between virulence factors, gene diversity, phylogenetic background, and antimicrobial resistance in the UPEC isolates.


Introduction
Urinary tract infections (UTIs), after respiratory infections, are one of the most common infections among hospitalized patients and referrals to laboratories (Demirci et al. 2019). Escherichia coli accounts for over 80-90% of communityacquired UTIs and 30-50% of hospital-acquired UTIs and is one of the major contributors to hospitalization with severe complications and high healthcare costs. The infection is more prevalent in women, and half of women experience this condition at least once in their lifetime; recurrence of infection is common. There are two pathogenic groups of E.coli, intestinal pathogenic E. coli (InPEC) and extraintestinal pathogenic E. coli (ExPEC) (Kaper et al., 2004;Xia et al. 2015), which include meningitis-associated E. coli, sepsis-associated E.coli, and uropathogenic E. coli (Ørskov & Ørskov, 1985). The ExPEC strains, compared with commensal E. coli strains, have bigger genomes and express more virulence factors (Rasko et al., 2008).
Vital virulence factors in the UPEC developing resistance to the host defense system and contributing to adhesion, invasion, and damage to the host cell include adhesins, toxins, siderophores, polysaccharide-based protective coatings, inosines, and serum resistance-related proteins (Johnson et al. 2005). The genes expressing virulence factors are located on bacterial chromosomes, plasmids, and even bacteriophages and can be transferred horizontally or vertically between bacteria (Piatti et al., 2008). The UPEC binds to the epithelial cell lining of the urinary tract with the help of Communicated by Agnieszka Szalewska-Palasz. adhesins such as type 1 fimbriae, A-fimbrial (afa), fimbrial adhesins P (pap), and S-fimbrial adhesins (sfa) (Dobrindt et al., 2001;Mulvey et al. 1998). Cytotoxic necrosis factor 1 (cnf1) is produced by about 40% of UPECs, which, along with many other factors, contributes to bacterial spread and viability in the urinary tract. Aerobactin, an iron-chelating agent, confers bacteria the ability of colonizing in iron-deficient environments such as urinary tract (Wiles et al. 2008). The activity of the cytolytic factor of α-hemolysin encoded by hlyA gene contributes to bacterial invasion into the epithelial barrier (Trifillis et al., 1994).
Phylogenetic studies are of particular importance in evaluating the genetic evolution of E. coli and can be investigated by polymerase chain reaction (PCR), multi-locus enzyme electrophoresis, or ribotyping. According to phylogenetic studies, E. coli isolates are divided into seven phylogenetic groups B2, B1, A, D, F, E, C, and clade 1 (Clermont et al. 2013), which are different in environmental niches, life history characteristics, and tendency to develop disease (Gordon et al. 2008). Most of the extraintestinal E. coli strains that are producing UTIs belong to the B2 phylogenetic group, and a few falls into group D, with the majority of commensal strains being in other groups. Phylogenetic studies of E.coli, using multiplex PCR technique, is based on the presence of three genes, ChuA and YjaA, arpA, and anonymous DNA fragment TspE4.C2 (Clermont et al. 2019(Clermont et al. , 2013. The emergence of antibiotic resistance in pathogenic bacteria is one of the global treatment problems. Currently, reports show that the rate of resistance in UPEC bacteria is increasing. This is especially important in countries with misuse and overuse of antibiotics. Determination of antibiotic resistance patterns in common pathogenic bacteria is important to guide experimental and specific therapies against specific pathogens, including UPEC strains (Al-Naqshbandi et al. 2019).
The purpose of this study was to investigate the relationship between virulence factors (involved in adhesion and toxin production) and the antibiotic resistance profile with phylogenetic groups in clinical UPEC strains isolated from outpatient in some hospitals in southern Iraq.

Isolation of Escherichia coli isolates
In this study, 385 urine samples were collected from patients with UTI symptoms from May 2017 to January 2018 from Qalat Saleh, Al-Sadr, Al-Zahrawi, and Children Hospital in Maysan Governorate in southern Iraq. In total, 100 clinical isolates (62 females and 38 males) of UPEC were collected. The streaking method was used to discrete pure isolates and then the isolates were identified by API 20E Kit (bioMerieux,France), and stored at − 20 °C.

Phylotyping and detection of virulence genes
The DNA was extracted by boiling and stored at − 20 °C until testing. Multiplex PCR based on Clermont method was performed to phylotype the phylogenetic groups of E. coli isolates (Clermont et al. 2013). Also, multiplex PCR technique was used to analyze the prevalence of virulence genes of fimH, afa, hlyA, papC, cnf1, and aer. Table 1 shows the sequences of primers used for each gene and the reaction conditions of Multiplex PCR (Bio-Red, Germany) for each gene cluster.

Statistical analysis
To compare the occurrence of phenotypic markers in UPEC, chi-square and two-tailed Fisher's exact tests were used. These tests also were used to describe the association of probable virulence factors with other factors. P < 0.05 was considered statistically significant.

Results
In total, 100 isolates, of the 385 urine specimens from UTI patients, were identified as UPEC and of which 62% and 38% were related to females and males, respectively. The age range of patients in this study was between 4 months and 78 years. The highest prevalence of the disease in all isolates was observed in the age range of 0-5 years (29 samples), followed by 18 patients in the range age of 30-39 years in women and 17 patients in the range age of 20-29 years in men ( Fig. 1). There was a statistically significant relationship between the frequency of bacteria and gender (P = 0.006).

Prevalence of virulence genes
The results showed that 96% of the isolates had at least one variant of the genes encoding virulence factors; Fig. 3 shows an example of PCR results. The highest and lowest prevalence were related to the fimH gene and the afa gene, respectively. The results also showed a high prevalence of virulence genes among the isolates. The distribution of virulence genes in this study was fimH (96%), aer (47%), papC (36%), cnf1 (16%), hlyA (15%), and afa (8%) respectively.

Prevalence of phylogenetic groups
In the present study, the prevalence of phylogenetic groups was as follows: B2, 46% of isolates; C, 13% of isolates; E and A, 11% of isolates; D, 7% of isolates; clade I, 3% of isolates; and unknown, 12%. The highest and zero prevalence phylogenetic group was related to B2, and B1 and F, respectively (Fig. 3).

Relationship between the phylogenetic group and patient profile
In female patients, the frequency of phylogenetic groups was as follows: B2 (63%), C (61.5%), clade I (66.6%), and D (57.14%). Phylogenetic groups E and A showed almost similar distribution in women and men with 45.45% and 54.5%, respectively. Unknown group with 66.6%, which was the most isolated sample, belonged to women. The highest and lowest frequencies of phylogenetic groups were in the range of 0-4 (30%) and 10-19 years (4%), respectively. Phylogenetic groups B2 (63%), C (61.5%), U (66.6%), D (57.14%), and clad I (66.6) were more common in women than men. In contrast, phylogenetic groups E (54.5%) and A (54.5%) were more common in men. The highest and the lowest frequencies of phylogenetic groups were in the range of 0-4 years and 50-78 years, respectively. The results obtained by chi-square test show a significant relationship between phylogenetic groups and age (P < 0.05) (Fig. 4). No significant difference between the distribution of phylogenetic groups in isolates of male and female patients was shown by statistical analysis.

Virulence gene and phylogenetic groups patterns
All the studied isolates exhibited 17 virulence gene patterns, referred to as Ec: four isolates belonged to the Ec1 pattern that did not have any virulence genes; the most frequent single gene was fimH identified in 33% of isolates (Ec2); the most two frequent genes were fimH and aer identified in 13% of isolates (Ec3); the most three frequent genes were fimH, aer, and papC identified in 16% of the isolates (Ec4); the most four frequent genes were fimH, aer, papC, and cnf1 identified in 5% of isolates (Ec5); and finally, the most five frequent genes were fimH, aer, papC, cnf1, and hlyA genes identified in 4% of isolates (Ec6). A total of six genes together was not detected in any of the isolates. Also, according to the results, among phylogenetic groups, the majority of the patterns belonged to B2 group with 15 patterns (88.23%), followed by group C and A with 5 patterns (29.41%), Group D and E with 3 patterns (17.64%), and clade I with 2 patterns (11.76%). Isolates with unknown phylogenetic group were seen in 3 patterns (17.64%) ( Table 2).

Distribution of antibiotic resistance among phylogenetic groups
In the present study, the rate of multidrug resistance among the isolates ranged from 3 to 8 classes of antibiotics of which, 94% of isolates showed MDR resistance to at least three classes of antibiotics. The frequency of samples among the classes of antibiotics showed that most isolates belong to five classes (31.91%), four classes (25.53%), and six classes (22.34%), respectively. The highest prevalence of MDR strains was observed in phylogenetic group B2. The MDR distribution among phylogenetic groups ranks between groups B2 (47.8%), C (13.8%), A (11.7%), E (10.6%), U (6.3%), D (7.4%), and clade I (2.1%), respectively.
The highest rates of resistance among phylogenetic groups belonged to B2, followed by groups D, C, E, A, unknown, and clade I, respectively. Of the 16 antibiotics studied, phylogenetic group D showed the highest antibiotic resistance rate to the seven antibiotics of Piperacillin, Ticarcillin, Amoxicillin clavulanic acid, Doxycycline, Fig. 4 Relationship between phylogenetic groups and patient age  Total  96  47  36  16  15  8  100 Norfloxacin, Gentamicin, and Imipenem. Phylogenetic clade I showed the highest antibiotic resistance rate to four antibiotics of Piperacillin, Ticarcillin, Trimethoprim-sulfamethoxazole, and Ceftazidime. Phylogenetic group C showed the highest antibiotic resistance rate to three antibiotics of Aztreonam, Ceftriaxone, and Amikacin. Phylogenetic group A showed the highest antibiotic resistance rate to two antibiotics of Nitrofurantoin and Ciprofloxacin. The phylogenetic groups of B2, E, and U showed the highest antibiotic resistance rate to Cefoxitin, Piperacillin, and Cefixime, respectively. The distribution of antibiotic resistance among phylogenetic groups is shown in Table 3. Significant differences were seen between phylogenetic groups and antibiotic resistance (P < 0.05).
P value obtained with chi-square test. The statistically significant values are shown in bold.

Relationship between virulence genes, phylogenetic groups, and antibiotic resistance
Of the 219 virulence genes observed among the isolates studied, the highest frequency was seen in phylogenetic group B2 and D with 55.25% and 10.5%, respectively, and the lowest gene prevalence was observed in group E and clade I with 6.39% and 2.73%, respectively. All examined virulence genes were observed in phylogenetic group B2. Five virulence genes were observed in the phylogenetic groups of D, C, and U. Four virulence genes were observed in the phylogenetic group A, and three virulence genes were observed in the phylogenetic groups of E and clade I. Table 4 shows the prevalence rates of virulence gene between phylogenetic groups. Table 5 shows the relationship between virulence genes and antibiotic resistance. According to the results, there was a significant relationship between papC gene and Amoxicillin-Clavulonic acid, Norfloxacin, Ciprofloxacin, and Impenem, between fimH gene and resistance to Piperacillin, Ticarcillin, and Nitrofurantoin, between afa gene and resistance to Ceftazidime, Gentamicin, and Amikacin, between aer and resistance to Ciprofloxacin and Norfloxacin, and cnf1 gene and Piperacillin. There was not any significant relationship between hlyA gene and antibiotic resistance.
P value obtained with chi-square test. The statistically significant values are shown in bold.

Discussion
Urinary tract infections, one of the most common human infections, have become a serious risk to public health due to the unexpected increase in antibiotic resistance. The use of molecular phylogenetic classification and the determination of resistance and sensitivity patterns of E. coli in patients can prevent the spread of many resistant infections, and help the economics and health of different communities with appropriate antibiotic treatment (Abd ALameer 2015).
Our outcomes showed the presence of UTIs in all age groups and an approximately prevalence of twofold in females compared to males (62 females versus 38 males). The anatomical differences in the urinary system between men and women causes an increase in the frequency of disease in women; E.coli is part of the normal flora of the gastrointestinal system, and urethra is shorter and wider in women increasing the chances of bacterial entry and colonization, resulting in ascending UTIs (Dadi et al. 2020). The highest prevalence of the disease was in the age range of 0-5 years, which is not in line with the results obtained by Shah et al. (2019). In a study conducted by Tabasi et al. (2015), the most prevalent disease was in the age range of 31-40 years in women and 51-60 years in men, which is consistent with the present results in terms of prevalence of disease in women but not in terms of age. Previous studies have shown that the prevalence of UTIs is highly correlated with socioeconomic status, educational level, and sexual activity (Tabasi et al. 2015;Emiru et al. 2013;Al-Gasha'a et al. 2020).
The highest and the lowest resistance rate among E. coli isolates was related to Piperacillin and to Imipenem respectively, which is consistent with the studies performed by others (Ahmed et al. 2019;Salehzadeh and Zamani 2018). López-Banda et al. (2014) showed that resistance to Imipenem and Amikacin antibiotics was lower than other antibiotics. Katongole et al. (2019) also showed a low rate of resistance to Imipenem and Amikacin. Given these similarities to the lowest antibiotic resistance to Imipenem and Amikacin, it can be suggested that the pattern of antibiotic administration in different parts of the world is the same and that these two drugs may be proposed as the last line of clinical treatment of UTIs when there is no other choice, avoiding antibiotic resistance against these two drugs.
Physicians usually treat the UTIs empirically, so awareness of epidemiological data in the area is essential to prevent unnecessary use of antibiotics in the treatment of infections and reduce unpleasant outcomes (Flores-Mireles et al.  . The outcomes indicated that the majority of isolates were resistant to most antimicrobials tested, and 94% of the isolates were resistant to more than three classes of antibiotics that were classified as the MDR group (Hadifar et al., 2016). The results showed that the prevalence rate of MDR isolates in southern Iraq was higher than in other countries. Compared to others, while North America and Europe have the lowest rates, Asian and African countries appear to experience higher levels of MDR-UPEC, which could be due to inappropriate and overuse of antibiotics in previous years (Ventola 2015). Phylogenetic identification and analysis is a way to study the diversity and characteristics of E. coli in order to resolve alternative treatment options and establish control programs (Müştak et al., 2015). The phylogenetic studies revealed that the ExPEC isolates are generally positioned in the phylogenetic group B2 and at a lower rate in group D (Ejrnaes, 2011). In the present study, the evaluation of phylogenetic groups for UPEC isolates showed that most isolates belonged to phylogenetic group B2 and clade I showed the lowest frequency. The results and patterns obtained from other studies suggest that most UPEC isolates fall into group B2 followed by groups C, which is in agreement with the present study (Tewawong et al. 2020;Boroumand et al. 2021), and group B1 and F were not observed among our isolates, which is in line with other studies (Boroumand et al., 2021). One of the reasons for the differences in the present study results with the above studies can be attributed to differences in the geographical area studied and differences in the number of isolates. Studies have shown that the highest microbial resistance rates are in phylogenetic group B2, which is in line with the present results (Noie Oskouie et al., 2019).
UPEC isolates have different types of pathogenic genes that, depending on the presence or absence of some of these factors, lead to the appearance or absence of clinical manifestations. Bacteria can acquire these genes through horizontal transfer and provide flexibility in the size of the bacterial genome (4.5 to 5.5 Mb.), so isolates with larger genomes will be more pathogenic (Hozzari et al., 2020). Identifying virulence factors, not only familiarizes us with the pathogenic process of bacteria, but also plays an important role in the development of vaccines and drugs (Dadi et al. 2020;Langermann et al. 2000).
The present results also showed that 96% of the isolates had at least one variant of the genes encoding virulence factors, and the distribution of virulence genes were as follows: fimH, aer, papC, cnf1, hlyA, and afa, respectively. FimH, due to its high percentage among UPEC isolates and its role in bacterial binding to urinary tract cells and colonization, is considered as a potential candidate to develop vaccine preventing urinary tract infections as the antibodies produced by the FimC-FimH vaccine did not alter the E. coli niche in the gut microflora (Langermann et al., 2000(Langermann et al., , 1997. Bacterial access to iron plays an important role in urinary tract infections (Wiles et al. 2008), and many isolates in this study encode the aer gene, which is necessary for iron uptake.
The highest and lowest prevalence of these genes was observed in phylogenetic group B2 and clade I, respectively. The results also exhibited a high prevalence of genes encoding virulence factors among UPEC isolates, which is in agreement with other studies.
Our results are in consistent with the results obtained by Lee et al. (2016) and Tarchouna et al. (2013), which reported highest and lowest prevalence of fimH gene, respectively, but differed from the results obtained by Zaki et al. (2015) who examined 91 hospital isolates in Egypt. It has been reported that the genes, fimH, papC, afa, aer, hlyA, and cnf1 highly distributed among all isolates (Yilmaz and Aslantas 2020; Zaki and Elewa 2015), which is also in line with the present results.
In this study, the isolates were divided into 17 patterns (Ec1-Ec17) based on the distribution of virulence genes; the pattern Ec2 containing only the fimH gene was the most prevalent pattern with 33 samples. The lowest prevalence of virulence genes was observed in Ec14, Ec17, Ec8, and Ec15 patterns with only one isolate. Also, none of the six genes was detected in the four isolates belonging to the Ec1 pattern. Out of the 17 different patterns, the phylogenetic group B2 was present in 15 patterns, phylogenetic groups A and C in 5 patterns, and phylogenetic group D in 3 patterns. These results are consistent with the findings by Jalali et al. (2015). Ali Abdi and Rashki Ghalehnoo (2015) detected 31 different patterns and reported that phylogenetic group B2 was present in most patterns (15 patterns). However, phylogenetic groups D and A each was present in 9 patterns, roughly in agreement with the results of the present study in terms of the prevalence of virulence genes in phylogenetic groups. Tarchouna et al. (2013) positioned the UPEC isolates in 23 patterns that were slightly different from the present study in the distribution of virulence genes, and reported the highest prevalence belonged to fimH gene. Some researchers mentioned that there is an inverse relationship between the number of pathogenic factors and the number of antibiotics to which the strain is resistant (Da Silva and Mendonça 2012;Brennan et al. 2018), and some reported opposite results, which showed a positive correlation between virulence and resistance (Ballesteros-Monrreal et al. 2020).
Our results also showed that there is positive and negative relationship between pathogenic factors and antibiotic resistance that can conclude there are not always a positive correlation between virulence and resistance (Ballesteros-Monrreal et al. 2020), and it depends on the type of E.coli, resistance mechanism, antibiotics studied (Cepas and Soto 2020), and gradual evolutionary mechanisms responsible for antibiotic resistance that, independently from any change in gene encoding virulence factors, could be arisen any time just in consequence of the continuous exposure to antibiotic agents (Yazdanpour et al. 2020).

Conclusion
The results showed that the highest prevalence rates in the studied isolates were related to phylogenetic group B2. Therefore, it can be stated that the isolates have high pathogenicity having virulence genes highly, as the results of different studies indicate that the two phylogenetic groups, B2 and D, carry more virulence factors than other phylogenetic groups. The high resistance of the isolates to antibiotics indicates that it is necessary to detect resistant isolates rapidly and timely in order to select appropriate treatment options and to prevent the spread of resistance. Also, it is recommended that urinary tract infections should be treated according to the regional pattern of sensitivity and resistance in order to prevent the spread of drug-resistant isolates. The results of the present study showed that the urine of patients with urinary tract infection could be a potential source for the spread of E. coli isolates with different virulence and resistance factors. Therefore, according to the high prevalence of urinary tract infection, the distribution of resistance and virulence factors, and the therapeutic failure, consequently, pyelonephritis, cystitis, and prostatitis, it is necessary to detect these isolates promptly and accurately.