Distinct Characteristics of Escherichia coli Isolated from Patients with Urinary Tract Infections in a Medical Center at a Ten-Year Interval

Escherichia coli causing urinary tract infections (UTIs) are one of the most common outpatient bacterial infections. This study aimed to compare the characteristics of E. coli isolated from UTI patients in a single medical center in 2009–2010 (n = 504) and 2020 (n = 340). The antimicrobial susceptibility of E. coli was determined by the disk diffusion method. PCRs were conducted to detect phylogenetic groups, ST131, K1 capsule antigen, and 15 virulence factors. Phylogenetic group B2 dominated in our 2009–2010 and 2020 isolates. Moreover, no phylogenetic group E strains were isolated in 2020. E. coli isolates in 2020 were more susceptible to amoxicillin, ampicillin/sulbactam, cefuroxime, cefmetazole, ceftazidime, cefoxitin, tetracycline, and sulfamethoxazole/trimethoprim, compared to the isolates in 2009–2010. Extensively drug-resistant (XDR)-E. coli in 2009–2010 were detected in groups B1 (5 isolates), B2 (12 isolates), F (8 isolates), and unknown (1 isolate). In 2020, XDR-E. coli were only detected in groups A (2 isolates), B2 (5 isolates), D (1 isolate), and F (4 isolates). The prevalence of virulence factor genes aer and fimH were higher in E. coli in 2009–2010 compared to those in 2020. In contrast, afa and sat showed higher frequencies in E. coli isolates in 2020 compared to E. coli in 2009–2010.


Introduction
Urinary tract infection (UTI) is one of the most common outpatient bacterial infections worldwide with a lifetime incidence of 50-60% in adult women and contribute to a substantial financial burden on society [1]. Lower UTI refers to the inflammation and

Antimicrobial Susceptibility of E. coli Isolated from UTI Patients Is Associated with Patient Age
We previously reported that E. coli strains isolated from patient age group >80 were more resistant to amoxicillin in combined with clavulanate, cefazolin, cefixime, cefmetazole, ceftriaxone, cefuroxime, ciprofloxacin, levofloxacin, than the average resistant rate [5]. In contrast, E. coli strains isolated from patient age group ≤3 were more susceptible to amoxicillin in combined with clavulanate, cefixime, cefuroxime, ciprofloxacin, levofloxacin, and piperacillin/tazobactam, than the average resistant rate [5]. The distribution of antimicrobial resistance of E. coli isolated from different host age groups was shown in Table  S1. Overall, E. coli isolated from the elderly were more resistant to cefazolin, cefuroxime, ceftriaxone, ciprofloxacin, and levofloxacin, in 2009-2010 and 2020 (p < 0.05) (Table S1). We next determined the antimicrobial susceptibility of E. coli isolated from two extreme age groups of UTI patients in 2009-2010 and 2020 (Table 4).

Discussion
The longitudinal survey to investigate the distinct characteristics of E. coli isolated from a single medical center at a 10-year interval is rare. In this study, we compared the bacterial characteristics, including phylogenetic groups, antimicrobial susceptibility, and virulence factors, of E. coli strains isolated from patients with UTI in 2009-2010 and 2020. We applied modified phylogenetic analysis to classify E. coli strains into eight groups in this study. Although phylogenetic group B2 was predominant in our 2009-2010 (62.1%) and 2020 (68.8%) E. coli isolates, we found 38 (4.6%) and 51 (6.0%) isolates in our 844 isolates belonged to phylogenetic groups C and E, respectively. Moreover, the results showed a decrease of groups C and E at a 10-year interval. Iranpour et al. reported the predominant phylogenetic group was B2 (39.3%), followed by unknown (27.1%), E (9.3%), C and clade I (each 6.4%), B1 (5%), F and D (each 2.9%), and A (0.7%) in E. coli causing UTI in Iran [29]. These results suggest the geographical difference of characteristics of E. coli isolated from UTI patients in different countries. Our previous study showed that phylogenetic group B2 dominated in UTI isolates (541/907, 59.6%), followed by group D (188/907, 20.7%), group A (95/907, 10.5%), and group B1 (83/907, 10.5), according to old phylogenetic analysis method [30]. Compared to our previous report [30], we found a significant decrease of group D (20.7% vs. 8.5%) and group A (10.5% and 3.1%) isolates in this study. Gordon et al. reported that 21% of E. coli strains were either incorrectly assigned to a phylogenetic group by the old triplex method or could not be assigned to one of the phylogenetic groups A, B1, B2, D, or E using the MLST data [31]. Therefore, this inconsistency may result from the incorrect assignment of E. coli isolates to a phylogenetic group, and thus the application of modified phylogenetic analysis to precisely classify E. coli strains is necessary.
Banerjee et al. reported ST131 was a dominant, antimicrobial-resistant clonal group associated with healthcare settings, elderly hosts, and persistent or recurrent symptoms [32]. Our findings show a stable existence of ST131 in a medical center in Taiwan. However, the prevalence of ST131 in patients with recurrent UTI was 17.2% (unpublished data), higher than the prevalence of this clone in UTI patients' first episode in this study. We found 29 (87.9%) and 5 (83.3%) of phylogenetic group C isolates in 2009-2010 and 2020, respectively, were MDR-E. coli (Table 3). In contrast, only 140 (44.7%) and 118 (50.44%) B2 isolates in 2009-2010 and 2020, respectively, were MDR-E. coli (Table 3).
Recently, Chakraborty et al. collected 33 bla NDM-5 -producing E. coli isolates which were highly resistant to β-lactams, including novel β-lactam/β-lactamase inhibitor combinations (ceftazidime/avibactam, imipenem/relebactam, and meropenem/vaborbactam). They found these isolates were assigned to different sequence types (STs) and indicated a predominance of isolates exhibiting ST167 in Switzerland and Germany (n = 10) (phylogenetic group C), followed by ST405 (group E), ST1284 (group C), and ST361 (group C) [33]. These results suggest that the new assigned phylogenetic groups, C and E, may be responsible for the spread of MDR-E. coli worldwide. Therefore, the characteristics of E. coli phylogenetic groups C and E are worth investigating and comparing with other groups.
We found E. coli isolated in 2020 were, in general, more susceptible to many antimicrobials. These results suggest the decrease of antimicrobial resistance in 2020 may result from the implementation of the national antimicrobial stewardship program from 2013 in Taiwan. In addition, we showed that E. coli strains isolated from age group >80 were more resistant to most commonly used antimicrobial agents than strains isolated from age group ≤3 (Table 4). These results are consistent with our previous findings [5]. In addition, Pulcini et al. also revealed that the elderly people in nursing homes had a risk around 40% higher than their community-dwelling peers of having antibiotic-resistant Enterobacteriaceae cultured from their urine samples [34]. Moreover, E. coli showed resistance to amoxicillin/clavulanate, nalidixic acid, ofloxacin, ciprofloxacin, ceftriaxone, and ESBLs were all more prevalent in nursing home samples than in community samples [34]. This observation may be caused by the high frequency and long-term use of antimicrobials in the elderly compared to the younger generation. As a result, antimicrobial stewardship and infection prevention and control programs should be tightly implemented in the elderly.
Consistent with previous reports [35,36], we found the high frequency of type 1 fimbrial adhesin gene fimH (94.0%) in our 844 isolates (Table 5). These results suggest a critical role of FimH in establishing successful colonization of E. coli in the urinary tract. Our results showed a change in the prevalence of virulence factors in E. coli isolates in 2020 compared to 2009-2010 ( Table 5). The significance of this evolutionary tract in E. coli uropathogenesis remains to be studied. In addition, the effect of different combinations of virulence factors in urinary tract colonization of E. coli is worth investigating. Here, our results revealed a decrease of virulence factors in E. coli isolated from the elderly compared to the younger generation (Table 6), which were consistent with our previous report, which showed the prevalence of virulence factors and antibiotic resistance of E. coli causing UTI were associated with patient age [5]. E. coli isolated from the elderly were more resistant to antimicrobials and had fewer virulence factors [5]. The role(s) of specific virulence factors in E. coli causing UTI in different patient age groups remains to be studied experimentally. Moreover, direct evidence demonstrating that the elderly are more vulnerable to low virulent E. coli caused by the decline of host immune protection or increase in host risk factors is still lacking.
In this study, we compared the characteristics of E. coli isolated from UTI patients in a single medical center at a 10-year interval to determine the regional evolutionary change of E. coli. We found a decrease in antimicrobial resistance and a difference in phylogenetic group composition and virulence factor distribution of E. coli at a 10-year interval. However, the driving force behind these phenotypic and genotypic changes among E. coli causing UTI at a 10-year interval remains to be investigated. In addition, whether these changes are associated with the virulence of E. coli is unclear. Moreover, our results showed that the characteristics of E. coli isolated from UTI patients were strongly associated with patient age. E. coli isolated from the elderly were more resistant to antimicrobials and had fewer virulence factors. Accordingly, the distinct characteristics of E. coli isolated from different age groups revealed in this study may be beneficial for clinical physicians to precisely control E. coli-caused UTI in different age groups of patients in the future.

Sampling and Isolation of E. coli
E. coli isolates were recovered from patients with UTIs at National Cheng Kung University hospital during 2009-2010 and 2020. This study was approved by the NCKUH Research Ethics Committee (B-ER-110-144). E. coli isolates were identified in the clinical laboratory by colony morphology, Gram stain, biochemical tests, and the Vitek system (bioMérieux, Marcy l'Etoile, France) according to the manufacturer's recommendations. A total of 504 and 340 non-duplicate E. coli isolates were collected in 2009-2010 and 2020, respectively. E. coli isolates were stored at −80 • C in lysogeny broth (LB) containing 20% glycerol (v/v) until tested.

Phylogenetic Grouping and Escherichia coli Sequence Type 131 (ST131) Detection
Based on PCR amplification patterns of specific genetic markers (arpA, chuA, trpA, yjaA, and TSPE4.C2), E. coli strains were divided into eight phylogenetic groups: A, B1, B2, C, D, E, F, and clade I, according to the previous study [27]. The primers used for phylogenetic typing are listed in Table S3.
Primers ST131_for (5'-GACTGCATTTCGTCGCCATA-3') and ST131_rev (5'-CCGGCG GCATCATAATGAAA-3') in combined purified genomic DNA as a template were used to perform PCR to detect E. coli ST131, according to the previous study [38]. Amplification mixtures were performed with the following cycling conditions: an initial denaturation at 94 • C for 3 min, 30 cycles of 94 • C for 30 s, 60 • C for 30 s, and 72 • C for 30 s, and one final cycle of 72 • C for 5 min.

Statistical Analysis
Pearson's Chi-square tests or Student t-tests were used for comparing categorical variables. All statistical analyses were performed using IBM SPSS statistics version 24.0 (IBM Corporation, Armonk, NY, USA). A p-value < 0.05 was taken as significant.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/pathogens10091156/s1, Table S1: The distribution of antimicrobial-resistant E. coli in different age groups in 2009-2010 and 2020, Table S2: The distribution of virulence factors in E. coli in different age groups in 2009-2010 and 2020, Table S3: Oligonucleotide primers used in this study. Institutional Review Board Statement: Ethical review and approval were waived for this study, due to the use of these clinically obtained isolates were remnants from patient samples, and the data were analyzed anonymously.