Clonality of Enterococcus Faecalis Isolates in Stool and Urine Samples of Patients with CommunityAcquired Urinary Tract Infection, and its Link with Virulence Gene Determinants and Antimicrobial Resistance Phenotypes

Background Community-acquired urinary tract infection (CA-UTI) could be caused through endogenous or exogenous routes. To show this relationship, we investigated molecular fingerprints and genotypes of paired Enterococcus faecalis isolates from the urine of symptomatic patients and their fecal samples. Out of the studied patients, 63 pairs of E. faecalis isolates were obtained simultaneously from their urine and feces samples. All the strains were sensitive to vancomycin, linezolid, nitrofurantoin, and daptomycin (MIC value: ≤4 µg/ml), whileresistance to tetracycline (Urine: 88.9%; stool: 76,2%) and minocycline (Urine: 87.3%, stool: 71.4%) wasdetected in most of them. The most common detected virulence genes were included efbA , ace , and gelE . RAPD-PCR and PFGE analyses showed same patterns of molecular fingerprints between paired of the isolates in 26.9% and 15.8% of the patients, respectively. 1 min, 72 °C for 1 min, and a final extension at 72 °C for 10 min. The PCR products were visualized using a UV transilluminator after electrophoresis in a 1% agarose gel and staining with the red safe solution (Bioneer, South Korea). To confirm correct amplification of the target genes, direct sequencing of one amplified product for each gene was carried out using ABI 3730X capillary sequencer (Pishgam, Macrogen, Seoul, Korea).


Abstract Background
Community-acquired urinary tract infection (CA-UTI) could be caused through endogenous or exogenous routes. To show this relationship, we investigated molecular fingerprints and genotypes of paired Enterococcus faecalis isolates from the urine of symptomatic patients and their fecal samples.

Results
Out of the studied patients, 63 pairs of E. faecalis isolates were obtained simultaneously from their urine and feces samples. All the strains were sensitive to vancomycin, linezolid, nitrofurantoin, and daptomycin (MIC value: ≤4 µg/ml), whileresistance to tetracycline (Urine: 88.9%; stool: 76,2%) and Background Urinary tract infections (UTIs) are the most common bacterial infections both in the community and hospital settings at all age groups. Although uropathogenic Escherichia coli is the most common cause of community-acquired urinary tract infections in humans [1], Enterococcus species, especially Enterococcus faecalis (E. faecalis), are considered as the second most important cause of UTI among uropathogenic bacteria [2,3]. E. faecaliscan also cause surgical wound infection, bacteremia, endocarditis, neonatal sepsis, and meningitis [4]. E. faecalis is predominantly inhabitant of the human gastrointestinal tract, where they form part of the normal intestinal flora in approximate amounts of 10 8 colonies per gram of feces [5]. This rate of colonization could predispose our urinary tract to recurrent infections via the perinea urethral route. This type of infection, which is known as community-acquired urinary tract infection (CA-UTI), is generally attributed to women. This infection may be host specific, due to the existence of receptors for bacterial adhesins, or mediated by potent virulence factors that are necessary for their pathogenesis in the urinary tract [6].
Management of CA-UTI involves administration of antibiotics based on susceptibility patterns of responsible bacteria in each region. Prompt elimination of the infection is needed to avoid severe complications in the infected patients [6].
Some virulence factors have been proposed for E. faecalis to describe its involvement in UTI; however, pathogenesis of this bacterium and its link with symptoms and complications of the infection is unclear yet. These virulence determinants, such as aggregation substance (asa1), gelatinase (gelE), cytolysin (cylA), enterococcal surface protein (esp), collagen-binding-protein (ace) and PavA-like fibronectin-binding protein (efbA), could facilitate initial colonization, biofilm formation, destruction of the host tissue, and evasion from host immune response. While in the hospitals, factors, such as the use of indwelling medical devices, can facilitate colonization of the urinary tract [6]; however, few data exist about mechanisms that are employed by this bacterium for its colonization in non-hospital settings. Diversity in colonization rate among different strains of this bacterium in different tissues and their pathogenicity could explain degree of complications that are occurring in the infected patients. While Enterococcal surface protein (Esp), adhesion to collagen of E. faecalis (ACE), aggregation substance (AS), PavA-like fibronectin-binding protein (EfbA), cytolysin (CYL), and gelatinase (GelE) are proposed as main virulence factors of E. faecalis, no virulence genotype has been suggested for discrimination of the pathogenic from non-pathogenic strains [7][8][9].
Comparison of phenetic, genomic and virulence characteristics of the strains causing UTI with those unable to cause this infection could provide more data about this link. This study was aimed to investigate diversity of virulence determinants, antibiotic resistance profiles, and genomic relationship of E. faecalis strains in urine samples of symptomatic patients with community acquired UTI compared with those isolated from their stool samples.

Results
Patients and Clinical isolates of E. faecalis A total of 126 E. faecalis isolates were obtained from 63 patients with CA-UTI. Of these, 63 were derived from urine and 63 were from fecal specimens, simultaneously.).The isolates showed positive results for esculin hydrolysis, 6.5%NaCl, non-fermentation of arabinose, and catalase tests and their identity were confirmed by species specific PCR assay. The mean age for the studied patients was 43 years, which ranged between 6 and 87 years old. vancomycin, ampicillin, penicillin, nitrofurantoin and linezolid in the urine and feces isolates. All the studied strains were susceptible to daptomycin (MIC value: ≤4 µg/ml). Except for minocycline, no significant difference was detected between the resistance rates in the strains collected from the urine and stool samples. Comparison of pairs of the isolates from urine and feces specimens showed same resistance patterns among 39 patients (61.9%); however, 17 (26.9%) and 8 (12.6%) pairs of them showed difference in resistance phenotype to one and greater classes of antimicrobials, respectively (Table 3). Multi-drug resistance (MDR) phenotype was detected in two pairs of the isolates. This phenotype was more common in urine samples of the patients with CA-UTIs originating from unrelated strains to the intestinal tract (11.1%, 5/63). All the MDR strains showed tetracyclines/gentamicin (120 µg)/fluoroquinolones resistance pattern.  a. TET, tetracycline; MIN, minocycline; GM120, gentamicin 120 µg; CP, ciprofloxacin; LEV, levofloxacin; GAT, gatifloxacin. Resistance phenotypes were determined for all antibiotics, except daptomycin, by disk diffusion (Kirby-Bauer) method according to CLSI 2014 guidelines (MastGroupLtd,United Kingdom). Antibiotic concentration for each disk was as follows: Penicillin G (10 units), ampicillin (10 µg), vancomycin (30 µg), tetracycline (30 µg), minocycline (30 µg), ciprofloxacin (5 µg), levofloxacin (5 µg), gatifloxacin (5 µg), nitrofurantoin (300 µg), high level gentamicin-resistant enterococci (HLGRE, 120 µg) and linezolid (30 µg). b. Patients with similar resistance patterns in both fecal and urine samples.
Association of antibiotic resistance patterns and virulence determinants among E. feacalis isolates (esp + /efbA + /asa1 + /ace + /cyl + /gelE + ) and 6 pairs (35.2%) showed partial genotypes ( Genotypic patterns of these strains are shown in Table 5. There was a significant correlation between the complete genotype and the determined endogenous infection based on the RAPD patterns (pvalue = 0.02). pairs with different RAPD types) were selected for PFGE analysis. Considering a cut off value of 87%, eleven strains (18.9%) showed common pulsotypes (CT) and thirty-two strains (55%) were singletons (ST).Amongst the CT, 10 pairs of the strains from the urine and fecal samples showed similar pulsotypes (Fig. 1). The characterized pulsotypes in patients with endogenous infections and their link with antibiotic resistance patterns and virulence determinants are shown in Fig. 1 and Table 6.
Comparison of E. faecalis isolates in urine and stool samples of patients with exogenous UTI based on antibiotic resistance patterns and virulence determinants are shown in Table 7.   [14].
Most of these bacteria are members of the fecal microbiota and can cause the infection through endogenous route [6,15]. E. faecalis isolates have been recognized as the second uropathogen in some countries [2,3].
In our study, we found a high prevalence of resistance to tetracycline and minocycline among E.  [16,17], but higher than the results obtained by other researchers from India, and Brazil. [14,18]. Arbitrary usage of antibiotics for the treatment of infections or agriculture could explain a higher rate of resistance to this antibiotic compared with other antimicrobials. In our study, the observed rates of resistance to tetracycline and minocycline in the fecal isolates were higher than those reported in the studies conducted by other researchers in healthy peoples [19][20][21]. This higher frequency of resistance among the fecal isolates could be caused by possible transmission of Enterococci from animal reservoirs through the food chain. The frequency of resistance to gentamicin(120 µg), ciprofloxacin, levofloxacin, and gatifloxacin in urine specimens was 28.6%, 20.6%, 14.3% and 12.7%, which was relatively similar to those detected in fecal specimens (15.8%, 12.6%, and 6.3%, respectively). This rate is consistent with the report published by Sallem et al.  [8]. Similarly, it seems that Ace protein (Adhesion to collagen of E. faecalis) bind to extracellular matrix proteins of the urinary tract, and plays an important role in early stage colonization and pathogenesis of UTI [28], while gelatinase (gelE), is a secreted protease, that is involved in dissemination of bacterium by degradation of polymerized fibrin [29]. The frequency of ace and gelE genes in fecal specimens was higher than those reported from healthy volunteers in Tunisia [21]. In our study, the frequency of esp, asa and cyl genes were 77.8%, 79.4% and 54% in urine and 74.6%, 65% and 46% in fecal specimens, respectively. Our result was in accordance with previous reports published by other studies in HA-UTIs [30][31][32][33] and in opposing with some other reports [34,35]. In the case of esp, its frequency among our isolates was higher than those reported in Tunisia among E. faecalis isolates from healthy volunteers (25.4%) [21].
There is little information about multiple virulence determinants among E. faecalis isolates associated Enterococci from patients with HA-UTI and found that most of the strains carried two virulence determinants in the urinary tract [37]. In the current study, 39.6% of the strains in urine specimens contained all the virulence determinants, while 28.5% and 25.3% carried five and four virulence determinants that were different from the aforementioned results. This discrepancy could be due to the difference in sample types and geographic locations; however, providing more accurate conclusions is not possible, since there is little information about community-acquired UTI through Enterococci and its association with related virulence determinants. Khalid investigated the occurrence of five virulence-associate genes in E. faecalis isolates associated with CA-UTIs and found that 28% of the strains contained all the virulence determinants, while 36% and 32% harbored four and five genetic markers of virulence [24]. In our previous study investigated concomitant distribution of virulence genes among E. faecalis isolates and found that 28.5% of strains contain all virulence determinants, 28.5% and 30%, five and four virulence determinants [22]. These results were in agreement with the current study results. In the current study, 36   cycles, and a final extension step at 72 o C for 10 min. Gel electrophoresis was used to interpret the results as described by [12]. Similarity of all banding profiles was analyzed by the GelCompar II software. E. faecalis ATCC 29212 was used as the control strain in this assay.

Pulsed field gel electrophoresis (PFGE)
Genomic DNA was prepared in agarose plugs as described by Turabelidze et al. with some modifications [13]. In brief, after cell lysis by lysozyme and then incubation with proteinase K, DNA was digested with Sma I. The PFGE procedure was carried out using a contour-clamped homogeneous electric field apparatus (CHEF DRII, Bio-Rad Laboratories, USA). Digested genomic DNA of Salmonella enterica serotype Braenderup (H9812) was used as size marker. The PFGE patterns were determined using the Dice coefficient in GelCompar II version 2.0 (Applied Maths, Belgium). Accordingly, isolates that differed by ≤ 3 bands were assigned to the same pulse type (PT), while isolates that differed by ≥ 4 bands were assigned to different types [12]. Pulsed field gel electrophoresis.

Ethics approval and consent to participate
This study was approved by the ethics committee of Shahid Beheshti University of Medical Sciences.
Since all urine and stool samples were routinely submitted for isolation and identifcation of E. fecalis strains, and all samples were pseudonymized, no informed consent was obtained.
Zohreh Ghalavand analyzed data and drafted manuscripts. Kiandokht Ghanati, Masoud Alebouyeh, and Marjan Rashidan provided research material and critical revision of this article. All authors read and approved the fnal manuscript.
Funding: This work was supported from Shahid Beheshti University of Medical Sciences, Tehran, Iran.
The funders had no role in designing the study, collecting data, and interpreting or deciding to submit the work for publication.
Availability of data and materials: Data supporting the findings of this study of infectious diseases and the Center for Tropical Medical Research are available. However, there are restrictions on the availability of this data and are therefore not available to the public. However, the data are available at the request of the authors with a reasonable request and with permission from the infectious diseases and tropical medicine research center.

Consent for publication
Not applicable.