Extraintestinal pathogenic Escherichia coli in Saudi Arabia: A review of antimicrobial resistance and molecular epidemiology

Extra-intestinal pathogenic Escherichia coli (ExPEC) is commonly associated with causing urinary tract and bloodstream infections. Over the past two decades, the antimicrobial resistance of ExPEC has increasingly been reported [1]. Given that Saudi Arabia annually hosts mass religious events, such as Hajj, this review investigated several aspects of antimicrobial resistance of ExPEC in this country including the current prevalence of resistance and molecular epidemiology of ExPEC isolates. Generally, the overall prevalence of antibiotic resistance of ExPEC in Saudi Arabia is on increase. The current emergence of colistin resistance in ExPEC represents a major challenge to public health. Local molecular epidemiological studies have shown the dominance of E. coli sequence type 131 (E. coli ST131) over other major ExPEC STs. This is an important observation given that this clone has been associated with high multidrug resistance and extended-spectrum β-lactamases carriage. To reduce the burden of this resistance in the future, it would be crucial to avoid uncontrolled use of antibiotics in either clinical settings or animal food industry.

In the recent years, the level of antibiotic resistance of E. coli, particularly ExPEC, to firstline agents such as ciprofloxacin and cephalosporins, has increased markedly [1]. Furthermore, this resistance has further increased globally to include many powerful antibiotic families such as polymyxins [3].
The Kingdom of Saudi Arabia is the annual host of mass religious events, known as Umrah and Hajj. These events are associated with high risk of infectious disease transmission [4] which makes the kingdom a potential center for the spread of antimicrobial resistant bacterial strains around the world.
This review aimed to summarize the current knowledge of antimicrobial resistance of ExPEC in Saudi Arabia, focusing on the overall local prevalence of antimicrobial resistance and molecular epidemiology of ExPEC isolates. The PubMed database was used to search the literature from 1987 to 2018, with the following phrases: Escherichia coli, extra-intestinal pathogenic E. coli, ESBL-producing E. coli, molecular epidemiology of extra-intestinal pathogenic E. coli, Saudi Arabia, antibiotics and antimicrobial resistance. Reference lists of relevant research articles were searched for more information.

Current prevalence of antimicrobial resistance of ExPEC in Saudi Arabia
In this section, the current local prevalence of antimicrobial resistance of ExPEC isolates, to major antibiotic classes is demonstrated.

Penicillins/penicillin-beta-lactamase inhibitor combinations
Early local reports showed that E. coli resistance to penicillins, such as amoxicillin and ampicillin, was reported in more than 50 % of isolates [5,6]. Nonetheless, penicillin-beta-lactamase inhibitor combinations showed better activity against E. coli compared to penicillins [7] (Table 1). Recently, the spectrum of ExPEC resistance to penicillins has increased significantly. For instance, it has been demonstrated that all UPEC isolates were resistant to ampicillin [8], and that 85 and 100 % of UPEC isolates were resistant to piperacillin and amoxicillin, respectively [9]. Although ExPEC resistance to penicillin-betalactamase inhibitor combination agents, particularly piperacillin/tazobactam, has recently been found to be low [10,11], Al-Agamy and colleagues have shown that resistance to amoxicillin/clavulanic acid and piperacillin/tazobactam was found in 96 and 50 % of the tested UPEC isolates, respectively [9].
The current high local resistance levels to penicillins and penicillin-beta-lactamase inhibitor combinations among ExPEC isolates have strongly been associated with ESBL carriage. It has been found that ESBL-producing ExPEC isolates were completely resistant to ampicillin, pipracillin and amoxicillin/clavulanic acid [9]. Resistance to piperacillin/tazobactam was also high among ESBL-producing ExPEC isolates, with 77.4 % of these isolates showing resistance to this agent [9] (Table 1).

Cephalosporins
Previous data demonstrated that about 70% of all E. coli isolates were susceptible to firstgeneration cephalosporins including cefazolin [6]. However, it was found that cefalotin was associated with low activity against E. coli [7,12]. E. coli resistance to second-generation cephalosporins such as cefoxitin and cefuroxime, was found to be significantly variable between agents, with a relatively low resistance to cefoxitin (range: 5 -7 %) [6] when compared to much higher resistance to cefuroxime (range: 13 -56 %) [6,13]. Furthermore, the range of E. coli resistance to third-generation cephalosporins was between 6 and 49 % of all E. coli isolates [6,9,10,12,13], while it was between 2 and 54% of all E. coli isolates for fourth-generation cephalosporins [9,10,12].
Recently, the level of ExPEC resistance to cephalosporins has increased markedly. For example, Al-Mijalli has reported that about 99 % of the all tested E. coli isolates were resistant to cefalotin, ceftriaxone, cefuroxime and ceftazidime, while resistance to cefoxitin and cefepime was reported in 14 and 96.7 % of isolates, respectively [14]. Another report has demonstrated that ESBL-producing UPEC isolates were associated with complete resistance to cefotaxime, ceftazidime and cefepime, and it also showed that 70.3% of these isolates were cefoxitin-resistant [9] (Table 1).

Monobactams
Previous local reports showed that the range of E. coli resistance to monobactams such as aztreonam, was between 36 and 63% [6,10,13]. However, the level of ExPEC resistance to aztreonam has recently increased. For instance, it has been found that 74.4% of ESBL-producing UPEC isolates were aztreonam-resistant [14] ( Table 1). Another study has demonstrated that all tested ESBL-producing UPEC isolates were aztreonam-resistant, while aztreonam resistance remained very low among non-ESBL-producing ExPEC isolates and it was only detected in 1.65% of the tested isolates [9].
Currently, although the majority of recently published reports have concluded that ExPEC isolates remained totally susceptible to imipenem and meropenem [9,16], AlYousef and others have demonstrated that 15 and 22.5 % of UPEC isolates were resistant to imipenem and ertapenem, respectively [8] (Table 1). This finding is important given that carbapenems are used for treating serious infections caused by multidrug resistant (MDR) ExPEC isolates, and such resistance makes the management of patients more complicated.

Resistance to aminoglycosides
Until early 2000s, the local prevalence of E. coli resistance to aminoglycosides was low; with less than 10% of E. coli isolates showing resistance to aminoglycosides [5,6,8,10,17]. However, a substantial increase in the level of resistance of ExPEC to some aminoglycosides agents, particularly amikacin and tobramycin, has been reported. For example, it has been found that 78% of ExPEC isolates were resistant to amikacin, 47.8% were non-susceptible to gentamicin while 56.5 % were resistant to tobramycin [18] (Table 1).

Resistance to fluoroquinolones
Early local studies reported high E. coli resistance to fluoroquinolones, such as ciprofloxacin, norfloxacin and levofloxacin. For example, resistance to ciprofloxacin ranged between 25.8 and 51 % [6,11], while resistance to norfloxacin was observed in 10 % of E. coli isolates [6]. More recent studies have shown higher levels of fluoroquinolone resistance in ExPEC isolates than was reported previously. For instance, Al Mously and others have detected ciprofloxacin resistance in about 70% of all tested ExPEC isolates [15], while Alyamani and co-authors have demonstrated that about 80% of ExPEC isolates were ciprofloxacinresistant [19]. Norfloxacin resistance of ExPEC has also recently increased, and this resistance has been reported in as many as 44% of ExPEC isolates [10]. ExPEC resistance to levofloxacin has also been high, ranging between 28 and 63 % [8,11,14] (Table 1).

Resistance to colistin
Globally, E. coli resistance to colistin has recently been reported [3]. Until 2016, E. coli resistance to colistin was not reported locally. Although the majority of past local studies did not include colistin in the antibiotic panels used to determine the antibiotic susceptibility profiles of E. coli, a previous study showed that all tested ExPEC isolates were completely susceptible to colistin [9]. However, Sonnevend and others described the first case of colistin resistance among ExPEC isolates in 2016, and they found a colistin resistant ExPEC blood isolate harboring the gene conferring resistance to colistin, namely mcr-1 [20]. This finding is of great importance given that colistin is currently used to treat many MDR ExPEC isolates that show resistance to powerful antibiotic families e.g. carbapenems, and such resistance might lead to the emergence and dissemination of pandrug resistant (PDR) ExPEC isolates in the future.

Resistance to other antibiotics
Many previous reports showed that E. coli resistance to nitrofurantoin has been very low, with more than 90% of isolates showing susceptibility to this agent [6,15]. More recently, the prevalence of nitrofurantoin resistance among UPEC isolates has increased, and this resistance was observed in as high as 15 % of UPEC isolates [14] (Table 1).
Before 2010, E. coli resistance to tetracycline ranged between 50 and 54.5 % of total E. coli isolates [6]. More recently, the spectrum of ExPEC resistance to tetracycline has increasingly been reported, and this resistance has been found in as high as 65.2 % of ExPEC isolates [18]. Tigecycline is highly active against E. coli and a previous report found a complete susceptibility of E. coli isolates to this agent [13]. However, higher tigecycline resistance has recently been detected locally, with 13% of isolates showing resistance to tigecycline [18] ( Table 1).
According to many previous reports, the local prevalence of E. coli resistance to sulfamethoxazole-trimethoprim ranged between 40 and 66 % of E. coli isolates [13,17] (Table 1). With regard to ExPEC resistance to fosfomycin, Al-Agamy and others have found full susceptibility to fosfomycin among a collection of UPEC isolates [9]. Interestingly, by contrast, AlYousef and colleagues have recently demonstrated that 37.5% of ExPEC isolates were resistant to fosfomycin [8] ( Table 1).
Molecular characterization of ESBL genotypes among ExPEC isolates has shown that bla CTX-M genes, encoding for CTX-M enzymes, were highly prevalent when compared to the traditional β-lactamases encoding genes such as bla TEM, bla SHV and bla OXA , and that bla CTX-M-15 was the most common CTX-M variant [9,16] (Table 2).
Interestingly, the spread of CTX-M-15 enzyme has been strongly associated with high levels of resistance to penicillins, cephalosporins, aminoglycosides, nitrofurans and fluoroquinolones among ExPEC isolates [9,16]. These findings highlight the major role of ESBL carriage in driving the high resistance levels among ExPEC isolates.

Molecular epidemiology of MDR ExPEC in Saudi Arabia
In Saudi Arabia, multilocus sequence typing (MLST)-based studies have generally shown considerable clonal diversity within ExPEC populations [19,[22][23][24]. However, E. coli ST131 was the most common ST type among the tested ExPEC isolates [19,22,24]. This section focuses on the most common ExPEC STs circulating locally.

E. coli ST131 complex
Recent studies have demonstrated that E. coli ST131 was the predominant ST type among all ExPEC STs in Saudi Arabia (Table 3) [19,22,24,25]. The local prevalence of ST131 isolates ranged between 17.3 and 61.7% of all ExPEC isolates. Alqasim and co-authors have found that ST131 accounted for 61.7% of all tested isolates, that H30 was the most frequent ST131 sub-clone, and that H30 isolates were higher than other ST131 sub-clones in terms of ESBL carriage and fluoroquinolone resistance [25]. Alghoribi and others have shown that this clone was highly assoacited with ESBL production and antimicrobial resistance compared to other ExPEC clones [22]. Abd El Ghany et al have characterized 10 carbapenem resistant ExPEC isolates recovered from bacteremia, and identified the presence of one E. coli ST131 isolate harboring the New Delhi metallo-β-lactamase gene, bla NDM-1 [23].

E. coli ST73 complex
In Saudi Arabia, Alghoribi et al have demonstrated that E. coli ST73 was the second most dominant ExPEC ST, accounting for 11.4 % of the total isolates identified in the study [22] ( Table 3). However, Alyamani et al have demonstrated that E. coli ST73 was less commonly identified among the tested urine isolates, with only 1 of 58 (1.7 %) isolates belonging to this clone [19].

E. coli ST69 complex
Locally, a previous study found that E. coli ST69 isolates accounted for 7.4% of the total STs identified in the study [22] (Table 3). However, another report did not identify E. coli ST69 in the UTI strain collection [19]. Abd El Ghany et al found one E. coli ST69 isolate of ten ExPEC bacteremia isolates harboring the carbapenem resistance determinant, bla NDM-1 [23].

E. coli ST95 complex
In Saudi Arabia, Alghoribi et al have found that E. coli ST95 accounted for 5.4 % of the total isolates identified in the study [22]. However, two recent studies did not identify this ST in their strain collections [19,23].

E. coli ST10 complex
Genotyping studies in Saudi Arabia have identified members of ST10 complex in their strain collections. For example, Alyamani et al have identified 7 of 58 (12 %) E. coli urine isolates belonging to E. coli ST10 complex [19]. Another study has found that ST10 isolates accounted for 6.4% of the total tested ExPEC isolates [22] (Table 3).

Other clonal ExPEC STs
Local studies have identified a number of other STs that are variable in terms of antimicrobial resistance levels and ESBL carriage. For instance, a study has shown that E. coli ST38 accounted for 7.4% of the total isolates, and this ST was highly associated MDR phenotype and high CTX-M carriage [22]. Alyamani et al have also found that 6.9 % of all tested E. coli urine isolates belonged to E. coli ST38 [19].
The MDR E. coli clone ST405 was identified among a collection of UPEC isolates, accounting for about 3 % of the total isolates [22]. Abd El Ghany et al has recently identified 8 out of 10 carbapenem resistant E. coli bacteremia isolates belonging to two ST complexes: ST448 complex and ST23 complex [23] (Table 3). ST448 isolates were associated with carrying the carbapenem resistance gene, bla OXA-181 , while the ST23 isolates harbored the carbapenem resistance determinant, bla NDM-5 .

CONCLUDING REMARKS
This review has shown that the local prevalence of antimicrobial resistance of ExPEC to many front-line antibiotics has increased drastically in the recent years. More importantly, ExPEC isolates have currently developed resistance to the last-line antibiotics such as carbapenems and polymyxins.
The current high antimicrobial resistance of ExPEC can be attributed to many factors. Firstly, the dominance of MDR ExPEC clones, particularly ST131, has recently been identified locally. This is alarming given that this clone is highly associated with resistance to many firstline agents such as fluoroquinolones and third generation cephalosporins, making the treatment of patients more complicated. In the future, it would be important to launch rapid diagnostic tests to identify these clones in order to reduce the burden of resistance.
Unoptimized use of antibiotics has also resulted in high antimicrobial resistance locally Therefore, monitoring the use of antibiotics in clinical settings or animal food industry is essential to combat resistance issue.

Conflict of interest
No conflict of interest is associated with this work.

Contribution of authors
I declare that this work was done by the author named in this article and all liabilities pertaining to claims relating to the content of this article will be borne by the author.

Open Access
This is an Open Access article that uses a funding model which does not charge readers or their institutions for access and distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/ 4.0) and the Budapest Open Access Initiative (http://www.budapestopenaccessinitiative.org/rea d), which permit unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.