Genomic Characterization of Uropathogenic Escherichia coli Isolates from Tertiary Hospitals in Riyadh, Saudi Arabia

Uropathogenic Escherichia coli (UPEC) is the most common cause of urinary tract infections (UTIs) in hospitalised and non-hospitalised patients. Genomic analysis was used to gain further insight into the molecular characteristics of UPEC isolates from Saudi Arabia. A total of 165 isolates were collected from patients with UTIs between May 2019 and September 2020 from two tertiary hospitals in Riyadh, Saudi Arabia. Identification and antimicrobial susceptibility testing (AST) were performed using the VITEK system. Extended-spectrum β-lactamase (ESBL)-producing isolates (n = 48) were selected for whole genome sequencing (WGS) analysis. In silico analysis revealed that the most common sequence types detected were ST131 (39.6%), ST1193 (12.5%), ST73 (10.4%), and ST10 (8.3%). Our finding showed that blaCTX-M-15 gene was detected in the majority of ESBL isolates (79.2%), followed by blaCTX-M-27 (12.5%) and blaCTX-M-8 (2.1%). ST131 carried blaCTX-M-15 or blaCTX-M-27, and all ST73 and ST1193 carried blaCTX-M-15. The relatively high proportion of ST1193 in this study was notable as a newly emerged lineage in the region, which warrants further monitoring.


Introduction
Urinary tract infections (UTIs) are the most common infection worldwide, with considerable economic consequences, morbidity, and mortality. UTIs cause inflammation of the urethra (urethritis), bladder (cystitis), and kidney (pyelonephritis) [1]. The infection affects individuals regardless of age and gender; however, its incidence is highest in females owing to the physiological and structural characteristics of the urethra [2].
Uropathogenic Escherichia coli (UPEC) is the most common cause of UTIs in hospitalised and non-hospitalised patients [3]. UPEC strains harbour numerous virulence factors that contribute to their ability to cause disease. UPEC strains can be distinguished from other E. coli pathotypes by specific virulence-associated determinants, including diverse adhesins, toxins, siderophores, capsule variants, and other miscellaneous traits [4]. Molecular typing methods are used to differentiate and characterise UPEC strains from each other by advanced molecular investigation methods. Such methods include multilocus sequence typing (MLST) and WGS [5,6]. These methods have been used to type UPEC strains isolated from clinical specimens and can be used in combination to elucidate the global epidemiology of UPEC strains [7].
Given their importance, it is imperative to understand the major lineages of UPEC and their role in the global dissemination of high-risk pathogens. The identification and characterisation of different UPEC lineages are essential for improving the understanding, diagnosis, and treatment of UTIs. One of the major lineages of UPEC that is of particular concern is ST131, which is responsible for a large percentage of UTIs and bloodstream infections [8]. This clone has been studied extensively to gain insight into its pathogenesis, antimicrobial resistance (AMR), and epidemiology [9]. ST131 is typically associated with high AMR and is considered one of the most successful and vital lineages. It is characterised by its ability to acquire and spread plasmids and can acquire resistance to multiple antimicrobials, including extended-spectrum cephalosporins and carbapenems. Recently, ST1193 has been identified as an emerging lineage of UPEC responsible for causing UTIs and bloodstream infections [10]. This clone is following in the footsteps of ST131, which has been reported previously as the second most frequent clone among ESBL and fluoroquinolone-resistant E. coli isolates [11][12][13][14] In Saudi Arabia, several studies have reported an increased frequency of AMR and the prevalence of UPEC, which requires urgent attention to understand the phenotypic and genotypic traits [15][16][17]. Studies examining the molecular mechanisms of UPEC are lacking in Saudi Arabia. Here, we utilised WGS analysis to perform genomic characterization of UPEC isolates recovered from two hospitals in Riyadh, Saudi Arabia, namely, King Abdul-Aziz Medical City (KAMC) and King Abdullah bin Abdul-Aziz University Hospital (KAAUH).

Antimicrobial Susceptibility Testing
The antimicrobial susceptibility of the examined strains to different antimicrobial agents was routinely tested in the clinical microbiology laboratories at KAAUH and KAMC. Among the 165 isolates, 56 (33.9%) were ESBL-producing E. coli isolates, of which 22% (n = 22/100) were collected from KAAUH and 52.3% (n = 34/65) were collected from KAMC. The percentage of resistance to the tested antibiotics is shown in Table 1. Ciprofloxacin resistance was detected in 35.8% (n = 59/165) of isolates, 35 of which were ESBL-producing. All isolates examined in this study were susceptible to tigecycline and carbapenem (imipenem and meropenem).

Discussion
In this study, we investigated the prevalence, clonal relatedness, and antibiotic susceptibility of UPEC isolates from patients with UTIs at two tertiary care hospitals in Riyadh, Saudi Arabia. We used both phenotypic and genotypic methods to identify ESBLproducing E. coli isolates and found that 33.9% (56/165) of the UPEC isolates tested positive for ESBL. The ESBL-producing UPEC isolates were highly resistant to other commonly used antibiotics, such as ciprofloxacin and trimethoprim.

Discussion
In this study, we investigated the prevalence, clonal relatedness, and antibiotic susceptibility of UPEC isolates from patients with UTIs at two tertiary care hospitals in Riyadh, Saudi Arabia. We used both phenotypic and genotypic methods to identify ESBL-producing E. coli isolates and found that 33.9% (56/165) of the UPEC isolates tested positive for ESBL. The ESBL-producing UPEC isolates were highly resistant to other commonly used antibiotics, such as ciprofloxacin and trimethoprim.
Our findings showed the prevalence of ESBL-producing UPEC (33.9%) in both KAAUH and KAMC, which was nearly consistent with other reports from Saudi Arabia that found 35% in 2015 [15] and 33% in 2018 [16]. However, the prevalence of ESBLs in KAMC increased from 35% in the isolates collected for investigations in 2012-2013 to 51.5% in the isolates gathered for this study in 2019-2020. [15]. This poses major healthcare and economic burdens that lead to serious and complicated UTIs with limited treatment options. Our study aimed to enhance the AMR WGS-based surveillance network to monitor the spread of UPEC within Riyadh city and among different hospitals.
This study characterised 48 ESBL-producing UPEC isolates using the WGS approach. In silico investigation was performed using the generated sequences in this study to determine the clonal structure of the examined isolates by means of phylogrouping and serotyping, followed by MLST and SNP-derived phylogenetic analysis, to determine the clonal relatedness. Our findings showed that the majority of the ESBL-producing UPEC isolates in our study (n = 30/48, 62.5%) belonged to phylogenetic group B2. This is consistent with earlier UPEC findings in Saudi Arabia, which demonstrated the dominance of this group over other phylogenetic groups observed in this study, including A, B1, D, F, and G [15,17].
The results of this study revealed the presence of multidrug-resistant UPEC strains with high virulence potential, including ST10, ST69, ST73, ST131, ST405, and ST1193. ST131 was the most predominant sequence type in our study, which is a global pandemic clone responsible for community and hospital-acquired UTI and bloodstream infections [18][19][20]. This successful clone was first identified in 2008 as capable of producing ESBLs and is currently considered the most common multidrug-resistant, ESBL-producing UPEC strain worldwide [21][22][23][24].
The geographical distribution of ST131 has yet to be completely understood; however, it has been found in humans, animals, and food sources in Europe, North America, Canada, Japan, Korea, Asia, the Middle East, and Africa [7,23,25]. Subsequent research confirmed the worldwide prevalence of ST131 and its wide range of virulence and resistance genes present in transferable plasmids [26,27]. Additionally, ST131 is highly prevalent amongst fluoroquinolone-resistant E. coli [9].
It is clear that ST131 is highly diverse and can vary significantly in terms of AMR and virulence. This clone has been identified as the primary cause of UTIs in adults and children worldwide, partly due to its ability to develop resistance to a broad range of antibiotics readily. Furthermore, ST131 is known to be involved in outbreaks of healthcare-associated UTIs that have high antibiotic resistance, making them challenging to treat [28][29][30].
The molecular phylogeny of E. coli ST131 was studied, and three major subclades were revealed: A, B, and C. The ST131 subclone delineation is primarily based on fimH alleles, serotypes, and the carriage of AMR genes [7,31]. Our finding divided ST131 into two clades, clade A (fimH41 and fimH141) and clade C (fimH30). ST131 fimH30 isolates are typically serotype O25:H4, while fimH41 and fimH141 belong to serotype O16:H5.
The carriage of AMR genes also contributes to the distinction between the clades. The O25:H4-ST131 fimH30 subclone is a major global concern and serves as a background for simultaneous resistance to multiple drugs with different mechanisms of action and resistance. The O25:H4-ST131 fimH30 clade C2 (also known as C2-H30Rx) is rapidly expanding, associated with the production of CTX-M-15 and fluoroquinolone resistance [21,32,33]. This subclone was the most detected ST131 in our study, associated with a high resistance rate, as illustrated in Tables 3 and 4. Moreover, we have observed that other subclones of ST131, including C1-M27, are associated with the production of CTX-M-27. The C1-M27 was first detected during the late 2000s among ESBL-producing E. coli in Japan [24,34,35]. Along with C1-M27, bla CTX-M-27 gene was detected in other ST131 subclones, including fimH41 and fimH141. Each of these subgroups has its own unique genetic markers and phenotypic traits.
Studies have found that carbapenem-resistant strains of E. coli ST131 have emerged in recent years because of the acquisition of plasmids carrying carbapenemase encoding genes, such as bla OXA-48, bla NDM, and bla KPC-2 [30,36,37]. Therefore, it is crucial to be aware of the potential risks associated with this clone and take measures to prevent its spread.
Currently, ST1193 is an emerging multidrug-resistant clone rapidly spreading worldwide, mimicking the success of the highly successful ST131 clone [10]. It was first reported in 2012 among fluoroquinolone-resistant E. coli isolates recovered from humans and dogs in Australia between 2007 and 2008 [10]. ST1193 evolved from clonal complex (CC) 14 in the early 1990s through the transition of type 1 pili from fimH27 to fimH64 and the acquisition of three QRDR mutations in gyrA (p.S83L and p.D87N), parC (p.S80I), and parE (p.L416F) [10,13]. To the best of our knowledge, this is the first report of ST1193 fimH64 in Saudi Arabia carrying QRDR mutations and β-lactamase genes (bla CTX-M-15 and bla EC-5 ). This clone has been reported in various countries and is associated with serious infections, including urinary tract and bloodstream. To prevent the further spread of this clone, it is important to use effective antibiotic stewardship and infection control measures and improve the surveillance and identification of high-risk clones.
In conclusion, we described the genomic characterisation of UPEC isolates recovered from two hospitals in Riyadh, Saudi Arabia. WGS has been proven to be a powerful epidemiological tool for investigating UPEC [6]. Members of the ST131 lineage constitute a key UPEC clone, which, in our study, was unusually associated with high virulence in addition to broad antibiotic resistance. ST1193 is a recently evolving lineage that can carry bla CTX-M-15 and warrants close monitoring. Further studies are required to limit the spread of major UPEC lineages that can display high virulence potential and a broad spectrum of drug resistance, including the recent dissemination of carbapenemase genes such as bla NDM and bla OXA [38][39][40].

Bacterial Isolates and Phenotypic Testing
Non-duplicate E. coli UPEC isolates (n = 165) were recovered from patients with UTIs from two tertiary care hospitals in Riyadh, Saudi Arabia, KAMC (n = 65) and KAAUH (n = 100), from May 2019 to September 2020. Demographic data of the study participants were obtained from each hospital's electronic medical record system. UPEC isolates were collected from the clinical microbiology laboratories of each hospital after routine diagnostics. Patient demographics, including gender, age, and hospitalisation details, were retrieved from the medical record system. Bacterial identification and antimicrobial susceptibility testing were performed using the VITEK II instrument (BioMerieux, Marcy-l'Etoile, France).

Whole Genome Sequencing
Approximately one-third of the UPEC isolates (n = 48/165, 29.1%) were selected for whole genome sequencing to represent most ESBL-producing isolates collected in this study. Genome sequencing was performed using the MiSeq Illumina platform with a 2 × 300 bp paired-end reads protocol. Prior to sequencing, DNA was extracted using the QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany) and a Qubit Fluorometric Quantitation Thermo Fisher (Invitrogen, Waltham, MA, USA) was used to measure DNA quantity and integrity. According to the manufacturer's instructions, DNA library preparation was performed using the Nextera XT DNA Library Prep Kit (Illumina, Cambridge, UK).