Prevalence of plasmid-mediated resistance genes among multidrug-resistant uropathogens in Egypt.

BACKGROUND: The emergence of multidrug-resistant (MDR) uropathogens has become a public health threat and current knowledge of the genotypic basis of bacterial resistance is essential for selecting appropriate treatment options.
OBJECTIVES: To determine the prevalence of antimicrobial resistance among MDR uropathogens and to elucidate the molecular bases of plasmid-mediated resistance.
METHODS: Bacterial isolates were recovered from urine specimens of 150 out-patients with signs and symptoms of urinary tract infections (UTIs) at El-Demerdash Hospital, Cairo, Egypt. Standard methods were used for identification, antimicrobial susceptibility testing was performed according to CLSI guidelines.
RESULTS: Among the recovered isolates, 22.7% and 77.3% were Gram-positive, and negative, respectively. Of which; 43.3% were MDR with 60% harboring plasmids. Extended spectrum β-lactamase (ESBL) genes bla CTX-M, bla SHV, and bla TEM were detected on plasmids of 89.7%, 41%, and 84.6% of the tested isolates, respectively. The aminoglycoside resistance gene aac6'-Ib/aac-6'-Ib-cr was found on plasmids of 92.3% of the tested isolates followed by qnrS (92.3%), qnrB (46.2%), and qnrA (7.7%). The most prevalent quinolone efflux pump gene was oqxB (38.5%), followed by oqxA (20.5%), then qepA (10.3%).
CONCLUSION: High levels of resistance to nitrofurans, β-lactam/β-lactamase inhibitor, cephalosporins, aminoglycosides, and fluoroquinolones were detected, and their use as empirical treatment for UTIs has become questionable.

Introduction

Urinary tract infection (UTI) is one of the most common infections worldwide which may be caused by Gram-negative or Gram-positive bacteria, as well as by some fungi. The most common causative organism is Escherichia E. coli1. Antimicrobial resistance (AMR) has become a major threat to public health in many countries. There has been a steady increase in AMR to the agents commonly used in treatment of UTIs2,3. In most UTI cases, empirical therapy is initiated before the results of urine culture and sensitivity are available, thus it is necessary to have AMR surveillance4. The emergence and spread of multidrug-resistant (MDR) organisms, which show resistance to three or more classes of antimicrobials5, is increasing over time; and UTI cases requiring intravenous therapy due to the lack of effective oral treatment has become a challenge for physicians, complicating a previously simple-to-treat infection3.

Extended Spectrum Beta Lactamases (ESBLs) have emerged as a chief mechanism of resistance among uropathogens6. These ESBLs are enzymes that trigger the resistance against β-lactam antibiotics by hydrolysis of the β-lactam ring7. Regrettably, ESBL-producing organisms usually carry resistance determinants to other antimicrobial agents as well,such as aminoglycosides and fluoroquinolones, leaving a limited range of treatment options6. The aim of this study was to reveal the prevalence of antimicrobial resistance and the molecular bases of plasmid-mediated resistance among bacterial uropathogens in one of the major clinical settings in Cairo, Egypt.

Materials and methods

Specimen collection

Starting October 2015 to May 2016, a total of 150 bacterial isolates were recovered from urine specimens of patients suffering signs and symptoms of UTIs at the outpatient clinics of El-Demerdash Hospital, Cairo, Egypt. All specimens were mid-stream urine and patients were instructed on how to collect specimens to avoid contamination. Patients included in the study were adults (ages ranging from 25 – 45 years), symptomatic, with pyuria (Pus cells ≥20/HPF), and the bacterial count in urine was >105 cfu/ml. The study was approved by Faculty of Pharmacy, Ain Shams University Ethics Committee Nr. 212 and an informed consent was obtained from patients after explaining the study purpose.

Identification of the recovered bacterial isolates

Isolates were categorized based on their Gram reactions, followed by identification using standard methods. Identification to the species level was done for MDR isolates by using API® 20E identification kit and API® Staph identification kit (BioMérieux, France) for Gram-negative and Gram-positive isolates respectively.

Antimicrobial susceptibility testing

The Kirby-Bauer disk diffusion test was performed according to Clinical and Laboratory Standards Institute (CLSI) guidelines8,9 using commercially available antimicrobial disks (Oxoid, UK). The reference strains E. coli ATCC® 25922 and Staphylococcus S. aureus ATCC® 25923 were used for quality control. All MDR isolates were selected for further studying.

Determination of minimum inhibitory concentration of multidrug-resistant isolates

Minimum inhibitory concentration (MIC) values of MDR isolates were determined by broth microdilution method according to CLSI guidelines9,10 using ceftriaxone, cefepime, meropenem, gentamicin, and ciprofloxacin. The reference strains E. coli ATCC® 25922 and S. aureus ATCC® 25923 were used for quality control.

Extraction of DNA plasmids from multidrug-resistant isolates

The extraction of DNA plasmids from MDR isolates was done using Zyppy™ Plasmid Miniprep Kit (Zymo Research, USA) according to the manufacturer's instructions. The extracted DNA plasmids were analyzed via agarose gel electrophoresis11 and visualized by UV transilluminator.

Amplification of plasmid-encoded resistance genes

Amplification of antibiotic resistance genes was carried out by polymerase chain reaction (PCR) using the appropriate primers (Table 1); and the DNA plasmids of the MDR isolates as templates. Primers were manufactured by LGC Biosearch Technologies, USA. The amplified products were analyzed via agarose gel electrophoresis, and the expected DNA product size was determined by comparing to a 100 bp DNA ladder (New England Biolabs, UK).

Table 1

Primers sequences, expected product sizes, and annealing temperatures (Ta) of the tested genes

Gene	Primer	Primer sequence (5′ → 3′)	Expected product size (bp)	Ta (°C)	References
blaCTX-M	Pf Pr	CGCTTTGCGATGTGCAG ACCGCGATATCGTTGGT	550	52	Bonnet et al.12
blaSHV	Pf Pr	GGTTATGCGTTATATTCGCC TTAGCGTTGCCAGTGCTC	867	52	Rasheed et al.13
blaTEM	Pf Pr	ATGAGTATTCAACATTTCCG CTGACAGTTACCAATGCTTA	867	50
aac(6′)-Ib /aac(6′)Ib-cr	Pf Pr	TTGCGATGCTCTATGAGTGG CGTTTGGATCTTGGTGACCT	358	46	Hamed et al.14
qnrA	Pf Pr	GCCCGCTTCTACAATCAAGT GGCAGCACTATTACTCCCAAG	347	60
qnrB	Pf Pr	TATGGCTCTGGCACTCGTT GCATCTTTCAGCATCGCAC	192	60
qnrS	Pf Pr	TCGGCACCACAACTTTTCAC TCAC'ACGCACGGAACTCTAT	255	60	Hamed et al.15
qepA	Pf Pr	TCTACGGGCTCAAGCAGTTG ACAGCGAACCGATGACGAAG	312	55
oqxA	Pf Pr	CTCTCCTTTCTGCTCGTCGG AATAGGGGCGGTCACTTTGG	489	67
oqxB	Pf Pr	TAGTGCTGGTGGTGCTGGTA GGGTAGGGAGGTCTTTCTTCG	480	68

Notes: blaCTX-M, blaSHV, and blaTEM genes code for ESBLs; aac(6′)-Ib gene codes for aminoglycoside 6′-N-acetyltransferase type Ib; aac(6′)-Ib-cr gene codes for aminoglycoside 6′-N-acetyltransferase type Ib ciprofloxacin-resistant variant; qnrA, qnrB, and qnrS genes are PMQR determinants coding for quinolone resistance; qep A, oqxA, and oqxB genes code for plasmid-mediated quinolone efflux pump resistance.

Abbreviations: Ta, calculated annealing temperature; ESBLs, extended-spectrum beta-lactamases; PMQR, plasmidmediated quinolone resistance.

The antibiotic resistance genes amplified in this study included ESBL genes (CTX-M, SHV, and TEM); the aac(6′)-Ib gene conferring resistance to aminoglycosides, and its bifunctional variant aac(6′)Ib-cr conferring resistance to both aminoglycosides and ciprofloxacin; low level resistance plasmid-mediated quinolone resistance (PMQR) genes (qnrA, qnrB, qnrS), and quinolone efflux pump genes (qepA, oqxA, oqxB).

Sequencing of selected PCR products

Some selected PCR products of amplified genes were sent for sequencing at GATC, Germany using ABI 3730 xl DNA Sequencer. The alignment and assembly of the obtained forward and reverse sequence files into the final consensus was done using BioEdit v7.2.5 software16.

Transformation

Plasmids extracted from the MDR isolates were used to transform competent E. coli DH5α prepared according to the modified Hanahan method17 to test the phenotypic resistance of the transformants. Transformants were cultured on LB/ampicillin, LB/gentamicin, and LB/ciprofloxacin agar plates at concentrations of 100 µg/ml, 25 µg/ml, and 50 µg/ml, respectively.

Statistical analysis

Categorical variables were analyzed using the Chi-square test to determine statistical significance. Statistical analysis including descriptive statistics, frequency tables, and cross-tabulations was performed using Statistical Package for the Social Sciences software IBM® SPSS® version 2018. A value of P<0.05 was considered statistically significant, and significance was two-sided.

Results

Of the 150 recovered isolates; 79 (52.7%) were from female patients, and 71 (47.3%) were from male patients. Gram-negative bacilli (GNB) represented 77.3% of the total isolates (n=116); 107 of which (92.2%) were identified as members of Enterobacteriaceae, 7 (6%) were Pseudomonas spp., and 2 (1.7%) were Acinetobacter spp. On the other hand, 34 isolates (22.7%) were Gram-positive cocci (GPC); 23 of which (67.6%) were coagulase-negative Staphylococcus spp., 6 (17.6%) were Staphylococcus aureus, and 5 (14.7%) were Enterococcus spp.

The antimicrobial susceptibility patterns of the recovered isolates are shown in Table 2. Out of 150 isolates; 65 (43.3%) were MDR; 51 of which (78.5%) were Gram-negative including E. coli (24/51; 47.1%), Klebsiella (K.) pneumoniae (15/51; 29.4%), K. terrigena (4/51; 7.8%), Proteus mirabilis (5/51; 9.8%), Acinetobacter (A.) baumannii (2/51; 3.9%), and Pseudomonas (P.) aeruginosa (1/51; 2%). Meanwhile; 14 (21.5%) of the 65 MDR isolates were Gram-positive including S. aureus (4/14; 28.6%), S. haemolyticus (3/14; 21.4%), S. xylosus (3/14; 21.4%), S. epidermidis (2/14; 14.3%), and S. lentus (2; 14.3%). Results of the antimicrobial suscepibility of the recovered isolates are presented in table S1. The prevalence of antimicrobial resistance of the tested MDR Gram-negative and Gram-positive isolates to different anti-microbial agents is shown in Figures 1 and 2, respectively.

Table 2

Antimicrobial susceptibility patterns of the recovered 150 isolates

Antibiotics	Resistant isolates, n (%)
	Enterobacteriaceae (n=107)	Pseudomonas spp. (n=7)	Acinetobacter spp. (n=2)	Staphylococcus aureus (n=6)	Coagulase -negative Staphylococcus spp. (n=23)	Enterococcus spp. (n=5)
	R	R	R	R	R	R
Co-amoxiclav	53 (49.5)	nd	nd	2 (33.3)	9 (39.1)	nd
Ampicillin/ sulbactam	55 (51.4)	nd	2 (100)	3 (50.0)	9 (39.1)	nd
Cefoxitin	24 (22.4)	nd	nd	3 (50.0)	10 (43.5)	nd
Ceftriaxone	42 (39.3)	nd	1 (50.0)	1 (16.7)	8 (34.8)	nd
Ceftazidime	44 (41.1)	3 (42.9)	2 (100)	3 (50.0)	10 (43.5)	nd
Cefepime	46 (43.0)	1 (14.3)	1 (50.0)	3 (50.0)	9 (39.1)	nd
Imipenem	17 (15.9)	0 (0.0)	2 (100)	1 (16.7)	2 (8.7)	nd
Gentamicin	26 (24.3)	1 (14.3)	2 (100)	1 (16.7)	8 (34.8)	nd
Doxycycline	12 (11.2)	nd	0 (0.0)	1 (16.7)	1 (4.3)	0 (0.0)
Ciprofloxacin	43 (40.2)	2 (28.6)	2 (100)	1 (16.7)	11 (47.8)	1 (20.0)
Levofloxacin	35 (32.7)	2 (28.6)	1 (50.0)	2 (33.3)	8 (34.8)	1 (20.0)
Co-trimoxazole	40 (37.4)	nd	2 (100)	2 (33.3)	6 (26.1)	nd
Vancomycin	nd	nd	nd	1 (16.7)	1 (4.3)	1 (20.0)
Azithromycin	nd	nd	nd	3 (50.0)	8 (34.8)	nd
Erythromycin	nd	nd	nd	3 (50.0)	8 (34.8)	2 (40.0)
Clindamycin	nd	nd	nd	3 (50.0)	9 (39.1)	nd
Nitrofurantoin	40 (37.4)	nd	nd	6 (100)	11 (47.8)	0 (0.0)

Abbreviations: R, resistant; nd, not determined (due to lack of interpretation data in CLSI guidelines).

Figure 1

Prevalence of antimicrobial resistance of the 51 tested MDR GNB isolates to Different antimicrobial agents. Prevalence was expressed as percent of resistant isolates relative to total tested isolates for each antimicrobial agent

Figure 2

Prevalence of antimicrobial resistance of the 14 tested MDR GPC isolates to different anti-microbial agents. Prevalence was expressed as percent of resistant isolates relative to total tested isolates for each antimicrobial agent

Minimum inhibitory concentrations of MDR isolates

The obtained MIC results of MDR GNB and GPC isolates are shown in supplementary tables S2 and S3, respectively.

Extraction of DNA Plasmids from MDR isolates

DNA plasmids were extracted from 39 (60%) of the 65 MDR isolates. The extracted plasmids were analyzed via agarose gel electrophoresis, and the band sizes were compared to a 1 kb DNA ladder (New England Biolabs, UK).

Amplification of some plasmid-encoded resistance genes

Results of PCR amplification of the ESBL genes (CTX-M, SHV, and TEM); the aac(6′)-Ib gene, plasmid-mediated quinolone resistance (PMQR) genes (qnrA, qnrB, qnrS), and quinolone efflux pump genes (qepA, oqxA, oqxB) are depicted in figures S1, S2 and S3. The prevalence of amplified antibiotic resistance genes among MDR bacterial uropathogens is shown in Figure 3.

Figure 3

Prevalence of some selected antibiotic resistance genes among MDR bacterial uropathogens. Prevalence was expressed as percent of isolates carrying the tested genes relative to total tested isolates (n=39)

Genotypes of MDR isolates

Among the 65 MDR isolates; 22 different genotypes were observed based on the PCR detection of antimicrobial resistance genes on the extracted DNA plasmids of the MDR isolates, as shown in Table 3.

Table 3

Percentage of plasmid-mediated antimicrobial resistance genotypes of MDR isolates and statistically significant associations

MDR isolates harboring plasmids	Genotypes	No. of isolates (%*)	Significant associations	Pearson chi-square
Positive (n=39)	blaCTX-M/blaTEM/aac(6′)-Ib/qnrS	5 (7.7%)	aco-amoxiclav/blaSHV	0.026
	blaCTX-M/blaTEM/aac(6′)-Ib/qnrS/oqxB	4 (6.2%)	acefoxitin/blaSHV	0.002
	blaCTX-M/blaSHV/blaTEM/aac(6′)-	4 (6.2%)	aimipenem/blaSHV	0.012
	Ib/qnrB/qnrS
	blaCTX-M/blaSHV/aac(6′)-Ib/qnrB/qnrS	3 (4.6%)	ameropenem/blaSHV	0.01
	blaCTX-M/blaSHV/blaTEM/aac(6′)-Ib/qnrS	2 (3.1%)
	blaCTX-M/blaTEM/aac(6′)-	2 (3.1%)	bblaCTX-M/aac(6′)-Ib	0.001
	Ib/qnrS/oqxA/oqxB
	blaCTX-M/aac(6′)-Ib/qnrB/qnrS	2 (3.1%)	bblaSHV/qnrB	0.018
	blaCTX-M/blaTEM/aac(6′)-Ib/qnrB/qnrS	2 (3.1%)	bblaSHV/qnrS	0.031
	blaCTX-M/blaSHV/blaTEM/aac(6′)-Ib/qnrB	2 (3.1%)	bblaSHV/oqxB	0.035
	blaSHV/blaTEM	1 (1.5%)	bblaTEM/qnrB	0.047
	blaCTX-M/blaTEM/aac(6′)-	1 (1.5%)	bqnrB/oqxB	0.01
	Ib/qnrB/qnrS/oqxA/oqxB
	blaCTX-M/blaTEM/aac(6′)-	1 (1.5%)
	Ib/qnrA/qnrS/oqxB
	blaCTX-M/blaSHV/blaTEM/aac(6′)-	1 (1.5%)
	Ib/qnrS/oqxA/oqxB/qepA
	blaTEM/qnrB/qnrS	1 (1.5%)
	blaCTX-M/blaSHV/blaTEM/aac(6′)-	1 (1.5%)
	Ib/qnrB/qnrS/oqxA/oqxB/qepA
	blaSHV/blaTEM/aac(6′)-Ib/qnrB/qnrS	1 (1.5%)
	blaCTX-M/qnrS/oqxA/oqxB	1 (1.5%)
	blaCTX-M/blaTEM/aac(6′)-	1 (1.5%)
	Ib/qnrA/qnrS/oqxB/qepA
	blaCTX-M/blaSHV/blaTEM/aac(6′)-	1 (1.5%)
	Ib/qnrS/oqxA/oqxB
	blaCTX-M/blaTEM/aac(6′)-Ib/qnrA/qnrS	1 (1.5%)
	blaCTX-M/blaTEM/aac(6′)-	1 (1.5%)
	Ib/qnrB/qnrS/oqxA/oqxB/qepA
	blaTEM/aac(6′)-Ib/qnrS/oqxB	1 (1.5%)
Negative (n=26)	---	26 (40%)

Notes: genotypes, plasmid-mediated antimicrobial resistance.

Percentages were calculated with reference to the number of MDR isolates (n=65).

Significant association between antibiotic resistance and PCR detection of the respective gene on plasmids.

Significant co-existence of resistance genes on plasmids of the same isolate. Abbreviations: MDR, multidrug-resistant; PCR, polymerase chain reaction.

The results of transformation along with PCR amplification for the tested isolates are shown in Table S4. In case of plasmids harboring ESBL coding genes (n=39), successful transformation and gene expression was achieved with plasmids extracted from 28 isolates (71.8%) harboring such plasmids; demonstrated by the ability of transformants to grow on LB/ampicillin agar plates. Out of 36 MDR isolates that carried the aminoglycoside resistance gene aac(6′)-Ib/aac(6′)-Ib-cr, the plasmids extracted from 15 isolates (41.7%) were successfully transformed and resistance genes were expressed; which was demonstrated by the ability of transformants to grow on LB/gentamicin agar plates. Isolates that carried any of the plasmid-mediated quinolone resistance (PMQR) genes (aac(6′)-Ib-cr, qnrA, qnrB, qnrS, qepA, oqxA, and oqxB) were found to be 38 out of the 39 tested MDR isolates (97.4%); plasmids extracted from 10 of which (26.3%) were successfully transformed and resistance genes were expressed; demonstrated by the ability of transformants to grow on LB/ciprofloxacin agar plates.

Statistical analysis has shown that there is significant association between antimicrobial resistance and PCR detection of the respective genes on DNA plasmids. There is also significant co-existence of PCR-detected antibiotic resistance genes on DNA plasmids of the same isolate (P value <0.05). The statistical association and respective P values are shown in Table 3.

Discussion

As reported, UTIs are becoming more difficult to treat due to the emergence and prevalence of a wide range of antibiotic resistance mechanisms3. Accordingly, in this study we assessed both the phenotypic and genotypic bases of antimicrobial resistance of some MDR uropathogens against the most common antimicrobial agents used in treatment of UTIs.

The antimicrobial susceptibility of the GPC collected in this study (n=34) showed that the lowest resistance was observed to doxycycline, vancomycin, and imipenem (5.9%, 8.8%, 10.3%). On the other hand, the highest resistance was observed to nitrofurantoin (50.0%); cefoxitin, ceftazidime (44.8% each); ampicillin/sulbactam and cefepime (41.4% each). The antimicrobial susceptibility of the GNB (n=116) showed that the lowest resistance was observed to doxycycline (11.0%) and imipenem (16.4%). The highest resistance was observed to ampicillin/sulbactam (52.3%); co-amoxiclav (49.5%); ceftazidime (42.2%); cefepime (41.4%); and ciprofloxacin (40.5%). These results limit the use of nitrofurans, cephalosporins, β-lactam/β-lactamase inhibitors, and fluoroquinolones as empirical treatment of UTIs, while tetracyclines and carbapenems still retain their efficacy in treating UTIs based on in vitro data. Some other studies deduced that imipenem showed highest efficacy and may be the drug of choice for empirical therapy of UTIs based on the in vitro data19–21.

In this study, PCR amplification was used to detect some plasmid-mediated antimicrobial resistance genes associated with the 39 MDR isolates harboring plasmids. The ESBLs genes were detected in all 39 (100%) isolates with CTX-M gene showing highest prevalence (89.7%), followed by TEM (84.6%) then SHV (41%). Our findings agreed with the study conducted by Goudarzi et al. in 2015 which also revealed a dominant presence of CTX-M (74.9%), followed by TEM (70%), then SHV (59.9%)22. Of the 39 MDR isolates harboring plasmids, 36 (92.3%) carried the aac6′-Ib/aac-6′-Ib-cr gene conferring resistance to aminoglycosides and ciprofloxacin, 11 of which (30.6%) carried the three ESBLs genes together, and 23 (63.9%) carried two of the ESBLs genes. A study conducted by Paterson et al. revealed that ESBL-producing K. pneumoniae isolates that were ciprofloxacin-resistant possessed multiple β-lactamases23, to which our results are in accordance. Our results also showed that there is a significant association between the presence of CTX-M gene and aac6′-Ib/aac-6′-Ib-cr gene in the same isolate (P=0.001). Accordingly, isolates that produce ESBLs also carry resistance genes to aminoglycosides and fluoroquinolones, thus reducing treatment options. This cross-resistance is more prominent in urinary isolates3. Therefore, carbapenems, which are less prone to hydrolysis by such enzymes, have become the preferred therapy for infections with ESBL-producing pathogens24–27.

The most prevalent qnr gene in our study was qnrS gene (36/39; 92.3%), followed by qnrB (18/39; 46.2%), and qnrA (3/39; 7.7%). This prevalence pattern is in accordance to that obtained in a study conducted by several recent studies28–30. The quinolone efflux pump resistance genes were detected in 15 (38.5%) out of the 39 MDR isolates and their prevalence was highest for oqxB (15/39; 38.5%), followed by oqxA (8/39; 20.5%), and qepA (4/39; 10.3%). It should be noted that the acquisition of PMQR genes alone results in low levels of resistance to fluoroquinolones, and does not cause MICs to exceed the breakpoints of these agents31, but rather facilitates the selection of mutants of higher-level resistance32. This was evident in our study by the lack of statistically significant association between the presence of PMQR genes and resistance to fluoroquinolones, which means the presence of PMQR genes alone, did not confer resistance to fluoroquinolones.

Conclusion

High levels of resistance to antimicrobials commonly used for treatment of UTIs was detected among MDR uropathogens. The current efficacy of nitrofurans, β-lactam/β-lactamase inhibitor, cephalosporins, aminoglycosides, and fluoroquinolones has become questionable. Carbapenems, tetracyclines, and vancomycin have yet to retain their efficacy in treatment of UTIsbased on in vitro data. No significant correlation was observed between the presence of PMQR genes and fluoroquinolone resistance, indicating that PMQR genes alone do not grant phenotypic resistance to fluoroquinolones, however the resistance may have been due to co-existence of ESBL and/or aac6′-Ib/aac-6′-Ib-cr genes in the same isolate or even on the same plasmids.