Emergence of bla TEM, bla CTX-M, bla SHV and bla OXA genes in multidrug-resistant Enterobacteriaceae and Acinetobacter baumannii in Saudi Arabia.

Multidrug-resistant (MDR) patterns due to extended-spectrum β-lactamase (ESBL) production in pathogenic bacteria are now becoming prevalent in hospitals worldwide, posing a public health challenge. The aim of the present study was to determine the antibiotic susceptibility patterns and distribution of the bla TEM, bla CTX-M, bla SHV and bla OXA ESBL resistance genes in MDREnterobacteriaceae and Acinetobacter baumannii (A. baumannii). A cross-sectional study was conducted between September 2017 and August 2018 in the King Abdullah Hospital (Bisha, Saudi Arabia). Bacterial isolates were collected from the clinical samples of patients; these were identified and screened for ESBL production and their antibiotic susceptibility was examined using standard microbiology methods. Multiplex-PCR runs were performed to identify genes encoding ESBL producers. DNA sequencing analysis was used to identify the specific gene variants. Of the 274 isolates, 173 (63.1%) exhibited MDR patterns to different antibiotics. A. baumannii revealed the highest resistance rates for cefuroxime (100%), gentamicin (88%) and amikacin (86%). Klebsiella pneumoniae (K. pneumoniae) isolates had the highest resistance rates for cefuroxime (98%), aztreonam and trimethoprim/sulfamethoxazole (87% for each). Escherichia coli (E. coli) exhibited high resistance rates for trimethoprim/sulfamethoxazole (92%) and cefuroxime (87%). Of the 173 MDR isolates, 78 (45.1%) exhibited ESBL production. Of these, 88.9% (72/78) carried ESBL genes. The most prevalent gene-encoding isolates were bla TEM (84.7%), followed by bla CTX-M (33.3%), bla SHV (2.7%) and bla OXA-1 (1.4%). A single bla TEM gene was predominantly produced by K. pneumoniae (60.7%), A. baumannii (78.9%) and Proteus mirabilis (80%), whereas bla CTX-M was harbored by E. coli (33.3%). The co-existence of two different genes in a single bacterium was revealed in 22.2% of isolates, commonly between bla TEM and bla CTX-M (19.4%). Sequencing analysis revealed that bla CTX-M-15 and bla TEM-1 were predominant variants of the bla CTX-M and bla TEM genes, respectively. The present study revealed a diversity of ESBL genes in Gram-negative bacterial isolates, with bla TEM being the most prevalent type. The emergence of various ESBL genes with several co-existing genotypes is alarming, rendering extensive surveillance studies necessary to understand the transmission and epidemiology of such resistant gene-carrying isolates. Copyright: © Ibrahim et al.

Introduction

In recent years, the production of extended-spectrum β-lactamases (ESBL) has become the main mechanism of resistance to β-lactam and other antibiotics in Enterobacteriaceae and Acinetobacter baumannii (A. baumannii) (1). ESBL enzymes confer resistance to penicillins, cephalosporins, monobactams and other antibiotic classes (2). Multidrug-resistant (MDR) patterns due to ESBL production in pathogenic bacteria are now becoming prevalent in hospitals worldwide, posing a public health challenge, including treatment failure, prolonged hospital stay and increased mortality rates (3-5).

Recently, >300 different ESBL types have been described in Gram-negative bacteria (6,7). The blaTEM and blaSHV types have been recognized as the most prevalent ESBL genes conferring antibiotic resistance in pathogenic bacteria worldwide (8-10). Previous studies have revealed that the number of clinical isolates harboring the blaCTX-M gene type has also increased in the last few years (2,11). The blaCTX-M family includes >130 β-lactamase variants classified into five distinct groups: blaCTX-M-1, blaCTX-M-2, blaCTX-M-8, blaCTX-M-9 and blaCTX-M-25 (12).

The genetic diversity of ESBL-producing Enterobacteriaceae and A. baumannii has progressively increased, posing challenges to hospital authorities due to their ability to confer antibiotic susceptibility and limit therapeutic options (13,14). The characterization of resistance genes encoding ESBL-producing microorganisms is a powerful tool for developing evidence-based guidelines for combating antibiotic resistance in the clinical setting (11).

In Saudi Arabia, Gram-negative bacteria-harboring ESBL resistance genes have been studied, with most studies emerging from the central and eastern regions (5,15-19). Despite the spread of antibiotic resistance among bacterial pathogens in the southern region, data on the distribution of ESBL resistance genes and their resistance profile among Enterobacteriaceae and A. baumannii remain limited (20,21). A previous study performed at the Aseer Central Hospital, a regional hospital in the southern region, identified MDR patterns among 98.1% of Acinetobacter species recovered from patients at intensive care units (22). In addition, another study conducted in the same hospital identified a high distribution of class D carbapenemase-encoding genes in A. baumannii, mainly ISAba1/OXA-23 and ISAba1/OXA-24 carbapenemases, which is alarming and presents an emerging public health threat (23). The emergence of MDR A. baumannii bacteremia in the southern region of Saudi Arabia has been well documented as an important health problem (22,24,25). A recent study reported the high frequency of MDR Gram-negative bacteria and a rate of ESBL production of 27% in patients at the King Abdullah Hospital, a referral hospital in Bisha, in the southern region of the country (20).

Due to the lack of information on the genotyping of ESBL producers and their MDR patterns in southern Saudi Arabia, the present study aimed to determine the antibiotic susceptibility patterns and distribution of ESBL genes in Enterobacteriaceae and A. baumannii isolates collected from clinical specimens of patients. The findings of the present study facilitated the implementation of infection control measures and provided epidemiological data to prevent spreading of MDR bacteria. The data also provided guidelines for the use of antibiotics in the clinical settings and improved the management of patients suffering from infections caused by Enterobacteriaceae and A. baumannii.

Materials and methods

Study design and setting

A cross-sectional study was conducted between September 2017 and August 2018 at the King Abdullah Hospital (Bisha, Saudi Arabia). A total of 274 Enterobacteriaceae and A. baumannii were recovered from patients as a part of the treatment and diagnosis procedure for infections. The clinical samples were collected from the patients as a part of routine investigations of infectious agents in the hospital microbiology laboratory. Therefore, consent letters were not obtained from the patients as per the study nature. Various clinical samples of urine (n=96), sputum (n=84), wound swab (n=51), blood (n=27), high vaginal swab (n=5), tracheal aspirate (n=5), umbilical discharge (n=3), cerebrospinal fluid (n=2) and eye swab (n=1) collected from 274 patients suffering from different bacterial infections were included in the study. Discharge from umbilical stump was collected from neonatal and/or infant patients in the wards using sterile swab soaked with normal saline and submitted to the laboratory. High vaginal swab was taken from female patients by clinicians and sent for microbiological examinations. Clinical specimens with incomplete personal information of the patients were excluded from the study. The average age of the patients was 46.0±25.5 years, including 158 females and 116 males. In total, 32 patients were aged 2-17 years, 97 were aged 18-40 years, 57 were aged 41-65 years and 88 were aged >65years. Patients of <2 years old were excluded from the study. The hospital is a referral hospital in the southern region (365 beds) with different specialties serving the Bisha province and the surrounding areas. The Research and Ethics committee at the College of Medicine, University of Bisha (Bisha, Saudi Arabia) reviewed and approved the present study protocol (approval no. UBCOM/1438-05/04).

Isolation and identification of bacteria

Enterobacteriaceae and A. baumannii were collected from the microbiology laboratory of King Abdullah Hospital during the routine processing of the clinical specimens of patients. Preliminary isolation and identification of bacteria were based on conventional microbiological methods. Briefly, Isolation of bacterial pathogens from specimens of urine, stool, sputum and other body fluids were performed by inoculating one loopful of each sample onto MacConkey agar plates (Oxoid, Ltd.) using a sterile disposable plastic loop (10-µl loop). Specimens of wounds, eye, umbilical and vaginal swabs were inoculated directly onto MacConkey agar plates by streaking them onto a small area of the plate. A disposable sterile plastic loop (1-µl loop) was used for cross-streaking to spread the inoculum over the surface of each plate to obtain single colonies of the suspected bacterial pathogen. Specimen of blood (5 ml) were extracted under aseptic conditions, transferred immediately into sterile bottles containing brain heart infusion broth (Oxoid, Ltd.), incubated at 37˚C with 5% CO2 and examine daily for turbidity for ≤7 days. If turbidity was observed, a 10-µl loopful of the blood sample was subcultured onto MacConkey agar plates for isolation of the suspected Gram-negative pathogen after aerobic incubation at 37˚C for 24 h. The isolate was tentatively identified based on the colony morphology, gram staining and oxidase test and the API 20 E Gram-Negative Microbial Identification Kit (cat. no. 20160; bioMerieux SA). Then, full identification of bacterial isolates was confirmed using the Phoenix system identification assay (Becton, Dickinson and Company). One single bacterial isolate from the clinical sample of each patient was included in the present study.

Screening of ESBL production

Phenotypic ESBL production among isolates was examined using a double-disc synergy test (DDST) as previously described (26) and the decreased susceptibility to cefuroxime, ceftazidime and cefotaxime was examined according to the Clinical and Laboratory Standard Institute (CLSI) recommendations (27). Bacterial isolates yielded positive results with DDST, when subjected to a multiplex-PCR amplification assay to detect blaTEM, blaSHV, blaCTX-M and blaOXA-1 resistance genes.

Antibiotic susceptibility testing of ESBL-producing bacteria

Antibiotic susceptibility testing of the Enterobacteriaceae and A. baumannii was performed using the Kirby-Bauer disk diffusion method on Mueller-Hinton agar medium (Oxoid) as per the CLSI guidelines (27). The following antibiotics with known concentrations recommended by the CLSI were examined: Amikacin (30 µg), amoxicillin/clavulanate (20/10 µg), aztreonam (30 µg), cefepime (30 µg), cefotaxime (30 µg), cefoxitin (30 µg), ceftazidime (30 µg), cefuroxime (30 µg), ciprofloxacin (5 µg), colistin (10 µg), gentamicin (10 µg), imipenem (10 µg), meropenem (10 µg), nitrofurantoin (50 µg), piperacillin (100 µg), piperacillin/tazobactam (100/10 µg), tobramycin (10 µg) and trimethoprim/sulfamethoxazole (23.75 µg/1.25 µg; Oxoid). E. coli American Type Culture Collection (ATCC) 25922 served as a control strain for antibiotic susceptibility examination. The final antibiotic susceptibility results of bacterial pathogens were interpreted using the 2017 CLSI breakpoints to categorize the isolates as susceptible or resistant. All intermediate results were considered resistant strains. ‘A susceptible category indicates that the isolates of the patient respond to the usually achievable concentrations of that antibiotic when the dosage is recommended to treat that type of infection and bacterial species. Conversely, the resistant category indicates that the isolates of the patient are not inhibited by the usually achievable concentrations of that antibiotic with the dosages usually used with that drug’ (27). Isolates were defined as MDR when they were resistant to at least three antibiotics from different classes.

Multiplex-PCR for the detection of ESBL genes

Multiplex-PCR runs were performed to identify the blaTEM, blaSHV, blaCTX-M and blaOXA-1 genes. A pair of six specific oligonucleotide primers (Eurofins Scientific) were used in the PCR reaction, as previously described (28). DNA was extracted from ESBL-producing isolates using the boiling method as previously described (29). The amplification of ESBL genes was then carried out in a total reaction volume of 50 µl. Each reaction mixture contained 25 µl HotStarTaq Plus Master Mix (cat. no. 203643; Qiagen GmbH), 4 µl DNA template, a variable volume of a specific primer group (Table I) and 9 µl nuclease-free water. The Eppendorf Master cycler Gradient instrument (Eppendorf) was used for the amplification of target genes with the following optimal cycling conditions: Initial heat activation at 95˚C for 5 min, 35 cycles of denaturation at 94˚C for 45 sec, annealing at 53˚C for 45 sec, extension at 72˚C for 1 min and a final extension at 72˚C for 10 min. The amplification products were visualized under ultraviolet illumination at a wavelength of 312 nm, after running at 85 volts for 60 min on 2% agarose containing ethidium bromide (1 µg/ml). A 100-bp DNA ladder (cat. no. 239045; Qiagen GmbH) was used as a standard molecular weight to determine the size of PCR products. DNA from reference blaCTX-M, blaTEM, blaSHV and blaOXA-like-positive strains was used as a positive control.

Table I

The frequency of extended-spectrum β-lactamase resistance genes among Gram-negative bacteria (n=72) as detected by multiplex PCR using different primers.

		Oligonucleotide primer (28)
Resistance gene	N (%)	Sequence	Length	Band size (base pairs)
bla TEM	61 (84.7)	F: 5'-CATTTCCGTGTCGCCCTTATTC-3'	22	800
		R: 5'-CGTTCATCCATAGTTGCCTGAC-3'	22
bla CTX-M	24 (33.3)
blaCTX-M	20 (27.2)	F: 5'-TTAGGAARTGTGCCGCTGYA-3'	20	688
group 1		R: 5'-CGATATCGTTGGTGGTRCCAT-3'	21
blaCTX-M	3 (4.2)	F: 5'-CGTTAACGGCACGATGAC-3'	18	404
group 2		R: 5'-CGATATCGTTGGTGGTRCCAT-3'	21
blaCTX-M	3 (4.2)	F: 5'-TCAAGCCTGCCGATCTGGT-3'	19	561
group 9		R: 5'-TGATTCTCGCCGCTGAAG-3'	18
blaCTX-M	0.0	F: 5'-AACRCRCAGACGCTCTAC-3'	18	326
group 8/25		R: 5'-TCGAGCCGGAASGTGTYAT-3'	19
bla SHV	2 (2.7)	F: 5'-AGCCGCTTGAGCAAATTAAAC-3'	21	713
		R: 5'-ATCCCGCAGATAAATCACCAC-3'	21
bla OXA-1	1 (1.4)	F: 5'-GGCACCAGATTCAACTTTCAAG-3'	22	564
		R: 5'-GACCCCAAGTTTCCTGTAAGTG-3'	22

F, forward; R, reverse.

DNA sequencing

Random PCR products of blaTEM-, blaSHV-, blaCTX-M- and blaOXA-1-positive samples were selected for DNA sequencing to identify the specific gene subtypes. A total of ~30 µl PCR products were sealed in sterile Eppendorf tubes and sent to Macrogen, Inc. for sequencing. PCR products were sequenced on an ABI PRISM® 3730XL Analyzer (96 capillary types) using the same primer sets (Table I). The results were obtained from the website of the company. Similarities in the nucleotide sequences were compared on the GenBank database of the National Center for Biotechnology Information website using the Basic Local Alignment Search Tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi). The sequences obtained for blaTEM, blaCTX-M and blaSHV genes have been deposited in the GenBank database (http://getentry.ddbj.nig.ac.jp/) under accession numbers LC636038 to LC636063.

Statistical analysis

Statistical analysis was performed using SPSS 22.0 (IBM Corp.). Simple descriptive statistics were used to calculate antibiotic-resistant patterns, and the prevalence and distribution of ESBL resistance genes. Outcome data were stated as proportions in frequencies and means ± standard deviation.

Results

Bacterial isolates

A total of 274 Enterobacteriaceae and A. baumannii isolates obtained from the clinical samples of patients were used in the present study. The isolates were recovered from samples of urine, sputum, wound swab, blood, high vaginal swab, tracheal aspirate, umbilical discharge, eye swab and cerebral spinal fluid. This indicated that Enterobacteriaceae and A. baumannii could cause infections throughout body systems and sites, including genito-urinary, respiratory, bloodstream, central nervous and soft tissue infections, which was consistent with a recent study (30).

Resistance patterns of Enterobacteriaceae and A. baumannii

Out of the 274 isolates, 63.1% (n=173) exhibited MDR patterns to different classes of antibiotics. Fig. 1 illustrated the MDR patterns among different isolates. The MDR pattern was determined in 100% of A. baumannii, 63.5% of Klebsiella pneumoniae (K. pneumoniae), 54.8% of Proteus mirabilis (P. mirabilis) and 51.8% of E. coli isolates.

Figure 1

Frequency of multidrug-resistant Enterobacteriaceae and Acinetobacter baumannii producing extended-spectrum β-lactamase. MDR, multidrug-resistant.

Table II summarized the antibiotic susceptibility of MDR isolates. A. baumannii exhibited the highest resistance rates to the tested antibiotics (86% for amikacin, 88% for gentamicin and 100% for cefuroxime). K. pneumoniae exhibited the highest resistance rates for cefuroxime (98%), aztreonam and trimethoprim/sulfamethoxazole (87% each), and cefotaxime (83%). E. coli exhibited high resistance rates for trimethoprim/sulfamethoxazole (92%), cefuroxime (87%) and ceftazidime (71%). P. mirabilis exhibited high resistance rates for trimethoprim/sulfamethoxazole (100%), amoxicillin/clavulanate, cefotaxime, cefuroxime (88% each), cefepime, ciprofloxacin (82% each) and ofloxacin (77%).

Table II

Percentage of antimicrobial resistance among multidrug-resistant Acinetobacter baumannii and Enterobacteriaceae family members.

	A. baumannii	K. pneumoniae	E. coli	Proteus mirabilis	Enterobacter cloacae	Citrobacter freundii	Providencia rettgeri	Proteus vulgaris	Morganella morganii	K. oxytoca
Agent	(n=49)	(n=54)	(n=38)	(n=17)	(n=6)	(n=1)	(n=2)	(n=2)	(n=2)	(n=2)
Amikacin	86(42)	41(22)	16(6)	35(6)	33(2)	0.0 (0)	50(1)	50(1)	50(1)	50(1)
Amoxicillin/clavulanate	92(45)	69(37)	37(14)	88(15)	50(3)	100(1)	50(1)	50(1)	50(1)	50(1)
Aztreonam	96(47)	87(47)	68(26)	65(11)	83(5)	100(1)	50(1)	50(1)	100(2)	100(2)
Cefepime	94(46)	82(44)	66(25)	82(14)	50(3)	0.0 (0)	50(1)	0.0 (0)	50(1)	0.0 (0)
Cefotaxime	96(47)	83(45)	66(25)	88(15)	33(2)	100(1)	50(1)	50(1)	50(1)	0.0 (0)
Ceftazidime	94(46)	80(43)	71(27)	59(10)	50(3)	0.0 (0)	0.0 (0)	50(1)	50(1)	0.0 (0)
Cefuroxime	100(49)	98(53)	87(33)	88(15)	83(5)	100(1)	100(2)	100(2)	100(2)	50(1)
Ciprofloxacin	96(47)	61(33)	58(22)	82(14)	33(2)	100(1)	100(2)	0.0 (0)	50(1)	100(2)
Colistin	6(3)	4.0(2)	3.0(1)	6.0(1)	0.0 (0)	0.0 (0)	0.0 (0)	0.0 (0)	100(2)	0.0 (0)
Foxitin	92(45)	46(25)	18(7)	59(10)	50(3)	100(1)	0.0 (0)	0.0 (0)	100(2)	50(1)
Gentamicin	88(43)	61(33)	37(14)	53(9)	17(1)	0.0 (0)	50(1)	50(1)	100(2)	0.0 (0)
Imipenem	94(46)	35(19)	13(5)	29(5)	17(1)	0.0 (0)	50(1)	50(1)	100(2)	0.0 (0)
Meropenem	92(45)	33(18)	13(5)	24(4)	0.0 (0)	0.0 (0)	0.0 (0)	0.0 (0)	100(2)	50(1)
Nitrofurantoin	96(47)	69(37)	37(14)	71(12)	67(4)	0.0 (0)	100(2)	0.0 (0)	50(1)	50(1)
Ofloxacin	94(46)	63(34)	50(19)	77(13)	33(2)	100(1)	100(2)	0.0 (0)	50(1)	100(2)
Piperacillin	94(46)	61(33)	47(18)	65(11)	50(3)	0.0 (0)	50(1)	50(1)	100(2)	50(1)
Piperacillin/tazobactam	94(46)	54(29)	40(15)	35(6)	50(3)	0.0 (0)	50(1)	50(1)	100(2)	50(1)
Tobramicin	90(44)	59(32)	37(14)	53(9)	33(2)	0.0 (0)	50(1)	50(1)	50(1)	50(1)
Trimethoprim/sulfamethoxazole	96(47)	87(47)	92(35)	100(17)	100(6)	100(1)	100(2)	50(1)	50(1)	100(2)

Distribution of ESBL genes among MDR isolates

Of the 173 MDR Enterobacteriaceae and A. baumannii isolates, 78 (45.1%) exhibited ESBL production. Of the 78 isolates, 72 (92.3%) carried ESBL genes as determined using multiplex-PCR. Fig. 2 presents an example of multiplex-PCR results revealed during the present study. Of the 72 isolates, 53 were Enterobacteriaceae (K. pneumoniae, 28; E. coli, 18; P. mirabilis, 5 and Enterobacter cloacae, 2) and 19 were A. baumannii (Fig. 3). Collectively, the most prevalent ESBL resistance genes in the isolates were blaTEM (84.7%) followed by blaCTX-M (33.3%), blaSHV (2.7%) and blaOXA-1 (1.4%). The most frequent blaCTX-M group was the blaCTX-M-group 1 (27.2%), followed by the blaCTX-M-group 2 and 9 (4.2%), whereas none of the isolates carried the blaCTX-M-group 8/25 (Table I).

Figure 2

Agarose gel electrophoresis of multiplex-PCR amplification products. (A) Acinetobacter baumannii (lanes 1 and 3) and Klebsiella pneumoniae (lane 5) showed positive expression of the blaTEM gene (800 bp). Acinetobacter baumannii (lanes 2, 4 and 6) and Escherichia coli (lane 7) showed negative findings to ESBL gene expression. (B) Klebsiella pneumoniae (lane 1), Escherichia coli (lanes 2, 3 and 6) and Enterobacter cloacae (lane 4), positive blaCTX-M gene (688 bp). Acinetobacter baumannii (lane 5) and Proteus mirabilis (lane 7) showed negative findings to ESBL gene expression. (C) Lane 2, co-existence of blaTEM and blaSHV genes in Klebsiella pneumoniae isolate. Acinetobacter baumannii (lanes 3 and 6) showed positive blaTEM gene expression. Klebsiella pneumoniae (lane 4) showed positive blaSHV gene expression (713 bp). Acinetobacter baumannii (lanes 1, 5 and 7) showed negative findings to ESBL gene expression. Lane M, DNA marker (1,500-100 bp);. ESBL, extended-spectrum β-lactamase.

Figure 3

Prevalence and characterization of resistant genes encoding extended-spectrum β-lactamase-producing Enterobacteriaceae and Acinetobacter baumannii. ESBL, extended-spectrum β-lactamase.

Frequency of ESBL genes in Enterobacteriaceae and A. baumannii

Fig. 3 illustrates the distribution of ESBL genes among Gram-negative bacterial isolates. A single blaTEM gene was predominantly produced by K. pneumoniae (60.7%), A. baumannii (78.9%) and P. mirabilis (80%), while blaCTX-M was commonly produced by E. coli (33.3%).

Out of the total PCR-positive isolates (n=72), the co-existence of two different genes in a single isolate was revealed in 16 (22.2%) strains. This combination was mainly observed between blaTEM and blaCTX-M (19.4%; 14/72) and between blaTEM and blaSHV (2.8%, 2/72) genes. Among the isolates carrying a combination of blaTEM and blaCTX-M genes (n=14), the majority were K. pneumoniae (50%, n=7) and E. coli (28.6%, n=4; Fig. 3).

Distribution of ESBL genes according to isolate sources

The distribution of various ESBL resistance genes among the clinical samples of patients is revealed in Table III. Bacterial isolates encoding various ESBL resistance genes were commonly recovered from sputum (n=30), followed by urine (n=18), wound (n=13) and blood (n=8) specimens. A single blaTEM gene was frequently detected among isolates from sputum (73.3%; 22/30), urine (61.1%; 11/18) and wound (53.8%; 7/13) specimens. The highest frequency of a single blaCTX-M gene was detected among wound isolates (23.1%; 3/13) compared with sputum (10%), blood (12.5%; 1/8) and urine (11.1%; 2/18) isolates. The co-existence of blaTEM and blaCTX-M genes was observed in 37.5% (3/8) of blood isolates, 23.1% (3/13) of wound isolates and 22.2% (4/18) of urine isolates.

Table III

Distribution of extended-spectrum β-lactamase resistance genes in Gram-negative isolates recovered from clinical samples of patients.

	Single gene			Combined genes
Source of Gram-negative bacteria	blaTEM (n=45)	blaCTX-M (n=10)	blaOXA-1 (n=1)	blaTEM and blaCTX-M (n=14)	blaTEM and blaSHV (n=2)	Total
Sputum (n=30)	22	3	1	3	1	30
Urine (n=18)	11	2	0	4	0	18
Wound (n=13)	7	3	0	3	0	13
Blood (n=8)	4	1	0	3	0	8
Tracheal aspirate (n=2)	0	1	0	1	0	2
Eye swab (n=1)	1	0	0	0	0	1

Sequencing analysis of resistance genes encoding ESBL producers

The sequencing analysis of the blaCTX-M, blaTEM and blaSHV types is revealed in Table IV. Out of the 20 blaCTX-M family members, the majority (75%; n=15) carried the blaCTX-M-15 subtype. blaCTX-M-15 was identified among K. pneumoniae (n=7) and E. coli (n=5) isolates. Out of the 18 blaTEM family members, TEM-1 was the most prevalent variant (83.3%, n=15). This subtype was common among A. baumannii (n=7) and K. pneumoniae (n=5) isolates.

Table IV

Sequencing analysis results of blaCTX-M, blaTEM and blaSHV genes produced by Enterobacteriaceae members and Acinetobacter baumannii.

ESBL gene	Total (n=40)	K. pneumoniae (n=13)	E. coli (n=12)	P. mirabilis (n=2)	E. cloacae (n=3)	A. baumannii (n=10)
blaCTX-M subtype (n=20)
CTX-M-15	15	7	5	1	1	1
CTX-M-71	1	0	1	0	0	0
CTX-M-101	1	0	0	0	0	1
CTX-M-127	1	0	1	0	0	0
CTX-M-181	1	0	0	0	1	0
CTX-M-182	1	0	1	0	0	0
blaTEM subtype (n=18)
TEM-1	15	5	2	0	1	7
TEM-115	1	0	1	0	0	0
TEM-159	1	0	0	1	0	0
TEM-169	1	0	1	0	0	0
blaSHV subtype (n=2)
SHV-28	1	0	0	0	0	1
SHV-226	1	1	0	0	0	0

Discussion

The emergence of MDR patterns due to ESBL-producing Enterobacteriaceae and A. baumannii is becoming a global concern (13,31). The present study determined antibiotic susceptibility patterns and characterized ESBL genes among clinical isolates of Enterobacteriaceae and A. baumannii. Collectively, 63.1% of isolates exhibited MDR to different antibiotics. In addition, MDR patterns were identified in 63.5% of K. pneumoniae and 51.8% of E. coli isolates. In Riyadh, the capital of Saudi Arabia, MDR patterns were identified in 67% of uropathogenic E. coli isolates at a tertiary healthcare center (19). These values were higher than those reported in Libya, where the MDR phenotype was detected in 33.2% of E. coli and 42% of K. pneumoniae isolates from patients with urinary tract infections (32). The results were also consistent with the observed high prevalence of MDR patterns (100%)among clinical isolates of Enterobacteriaceae carrying ESBL resistance genes collected in Ethiopia (33).

In the present study, the overall resistance rates of MDR Enterobacteriaceae and A. baumannii were very high for most examined antibiotics, except for colistin. Previous studies have reported that co-resistance to several antibiotic classes of penicillins, cephalosporins, aminoglycosides, fluoroquinolones, trimethoprim/sulfamethoxazole and carbapenems was common among ESBL-producing Gram-negative bacteria (7,19,33,34). These findings indicated that the emergence of ESBL-producing microorganisms could cause susceptibility to various antibiotics.

A multiplex-PCR assay has been proposed to rapidly detect several resistance genes encoding ESBL-producing Gram-negative bacteria (28,35). However, complete gene sequencing for the blaTEM, blaSHV, blaCTX-M and blaOXA-1 types is essential to differentiate narrow-spectrum β-lactamases from ESBL (35). In the present study, among 81 phenotypically identified ESBL isolates, 88.9% carried one or more of the following resistant genes: blaTEM, blaCTX-M, blaSHV and blaOXA. Previous studies have reported that bacteria carrying ESBL genes confer resistance to extended-spectrum cephalosporins, β-lactam agents and other antibiotic classes (7,19,20). This phenomenon may pose serious public health risks, as it would result in substantial limitations in therapeutic options. Thus, appropriate control measures, including establishing screening strategies for identifying ESBL-producing bacteria, are required to prevent such strains.

blaTEM was the most prevalent gene detected in ESBL-producing Gram-negative bacteria in the present study. This was inconsistent with a study from the Eastern region of Saudi Arabia, where blaCTX-M (97.4%) was more frequent than blaSHV (23.1%) and blaTEM (0.0%) in Enterobacteriaceae (16). Similarly, the predominance of the blaCTX-M type in ESBL-producing Gram-negative bacteria in the Eastern region have been documented by other studies (4,36). Worldwide studies have reported different ESBL resistance genes produced by Gram-negative bacteria. For instance, blaCTX-M was the most prevalent type in the Asian Pacific region, followed by blaSHV and blaTEM (1). In Nigeria, the most frequent gene types among isolates from patients with surgical site infections were blaSHV, blaCTX-M and blaOXA (37). In Burkina Faso, the most prevalent ESBL resistance genes were blaCTX-M (40.1%), blaTEM (26.2%) and blaSHV (5.9%) in Enterobacteriaceae (38). These results, coupled with the present findings, revealed that the prevalence of ESBL gene types can vary between locations and geographical regions.

MDR ESBL-producing K. pneumoniae and A. baumannii have become common causes of healthcare-related infections (11,31). In the present study, the prevalence of blaTEM in K. pneumoniae was revealed to be 60.7%. Increasing rates of the blaTEM gene have been reported among clinical isolates of K. pneumoniae in Al-Qassim (70.9%) (8) and Riyadh (54.05%), in the Central region of Saudi Arabia (17). On the other hand, the frequency of blaTEM was high among A. baumannii isolates, which was consistent with a previous study from the Makkah city in the western region of the country (31). These high rates, which indicated the dissemination of such ESBL-producing isolates, is alarming for multiple hospitals. The high prevalence of the blaTEM gene detected in K. pneumoniae and A. baumannii isolates may increase the incidence rate of infection caused by these ESBL producers across different regions. This renders extensive surveillance studies in local and national hospitals in Saudi Arabia necessary to understand the transmission and epidemiology of resistance genes encoding ESBL-producing bacteria. However, using molecular methods in local hospitals to detect resistance genes may help develop effective new antimicrobial treatments against ESBL producers and improve the infection control system. In the present study, E. coli commonly produced the blaCTX-M gene (33.3%). Similarly, the predominance of E. coli carrying blaCTX-M genes has been reported in the western region of Saudi Arabia (39). Previous studies revealed that blaCTX-M is the most prevalent gene among uropathogenic E. coli isolates from patients with hospital and community-acquired infections (19,29). Furthermore, ESBL-producing E. coli collected from fecal colonization was revealed to produce the blaCTX-M gene as it has been reported in a previous study (40). Several factors and mechanisms contribute to the spread of bacterial clones carrying the blaCTX-M gene in Saudi Arabia, including plasmid dissemination and the clonal spread of bacterial strains, the frequent use of cephalosporins and the large population of migrant workers (13). According to Yasir et al (39), the high diversity in the E. coli clones may have arisen due to the fact that ~50% of the population of Saudi Arabia are expatriates from developing countries, including Pakistan, India, Bangladesh, the Philippines and African countries where self-medication in patients is evident.

The present study revealed that blaCTX-M-15 was the most frequent subtype of the blaCTX-M type. This was consistent with data from Saudi Arabia (6,16,17) and several other parts of the world (1,2). These findings indicated that blaCTX-M-15 is a public health concern, since it is the most widespread gene worldwide. The emergence of the blaCTX-M-15 variant has been revealed to be attributed to the horizontal gene transfer of genetic elements and the clonal expansion of microorganisms (19,41). Furthermore, the widespread and unnecessary use of ceftriaxone and cefotaxime have contributed to the emergence and spread of blaCTX-M resistance genes (2).

Multiple ESBL resistance genes in a single bacterium render that strain more difficult to treat with several antibiotic drugs (16). In the present study, the co-existence of two different ESBL genes in the same strain was detected in 22.2% of isolates. However, the most common combination of ESBL resistance genes was between blaTEM and blaCTX-M, which was consistent with studies from Pakistan (42) and Algeria (12). However, the combined production of blaTEM and blaCTX-M genes was more frequently detected in K. pneumoniae (50%) and E. coli (28.6%) isolates. These figures were lower than those reported in Nepal, where two or more ESBL genes were present in 100% of Klebsiella spp. and 56.2% of E. coli clinical isolates from a teaching hospital (7). The value reported in the present study was considerably higher than the 3.4% reported in uropathogenic E. coli from the Eastern region of Saudi Arabia (29).

The present findings revealed that sputum was the most frequent source of various ESBL resistance genes in Gram-negative bacteria. This may be due to the several sputum samples collected from patients at intensive care unit (ICU) wards. A previous study indicated that the characterization of antibiotic susceptibility of bacterial pathogens from the sputum of patients in the ICU with ventilator-associated pneumonia can help control this type of infection (43). In addition, it is known that most patients admitted to the ICU are immunocompromised and/or undergoing invasive procedures, which would lead to prolonged antibiotic therapy (25). The extended stay, selective pressure and frequent use of antibiotic treatment of patients in the ICU contribute to the increase in ESBL producers (43).

The present study has several limitations that need to be addressed in future studies. Firstly, the study was laboratory-based; therefore, clinical data of the patients were not obtained to analyze the risk factors for ESBL infection and understand the epidemiological spread of ESBL genes. Secondly, the study was conducted in a single center in southern Saudi Arabia. The results can therefore not be representative of all parts of the southern region. Multicenter studies are required to confirm these findings. Thirdly, the AmpC β-lactamase class and other types of ESBL enzymes, (such as blaVEB, blaPER, blaGES and blaBEL) which confer significant antibiotic resistance among Gram-negative bacteria, were not examined.

In conclusion, the ESBL resistance genes were a significant cause of MDR patterns and conferred susceptibility to various antibiotic agents in Enterobacteriaceae and A. baumannii. The present study reported high levels of various resistance genes in ESBL-producing isolates, with blaTEM being the most prevalent type. In addition, the co-existence of two different ESBL genes has been frequently detected in a single bacterial pathogen (12,42). The blaCTX-M-15 gene is the predominant variant among isolates carrying the blaCTX-M type. The emergence of various ESBL-resistant and coexisting genes in Enterobacteriaceae and A. baumannii is alarming and may significantly limit the efficacy of therapeutic options in hospital settings. However, extensive surveillance studies at both the local and national levels are urgently required to obtain an understanding of the transmission and epidemiology of resistance genes in ESBL-producing bacteria. Using molecular methods at local hospitals to detect resistance genes in ESBL-producing bacteria is recommended to improve the infection control system and help set effective antibiotic therapy plans.