Abstract
Objectives: To detect resistance genes to fluoroquinolones and β-lactams in Salmonella strains from a Saudi hospital.
Methods: From October 2015 to December 2016, a total of 149 Salmonella strains were collected from stool specimens from patients admitted to King Fahad Hospital of the University, AlKhobar, Saudi Arabia using CHROMagar Salmonella. The organism identification and antimicrobial susceptibility testing were performed using Vitek 2 system. Strain serogrouping was performed using Wellcolex color Salmonella kit. Fluoroquinolone resistance genes, extended-spectrum β-lactamases (ESBLs), and AmpC β-lactamase were determined using polymerase chain reaction (PCR). Enterobacterial repetitive intergenic consensus sequence-based PCR (ERIC-PCR) was used to determine clonal relatedness.
Results: The resistance rates to cefotaxime were 1.3% and ciprofloxacin 19.5%. Plasmid mediated quinolone resistance (PMQR) genes, qnrB and qnrS, were detected in 8 strains, qnrB (n=5) and qnrS (n=3), respectively. No ESBLs, AmpC, or mutations in the topoisomerases were detected. Salmonella isolates formed 7 clusters with similarity.
Conclusions: This study reveals the emergence of fluoroquinolone resistant Salmonella in the region imposing public health concerns.
Acute gastroenteritis caused by Salmonella species is a worldwide health issue for humans and animals.1-3 Although most cases of salmonellosis are uncomplicated it can be very serious and life-threatening in infants, elderly, and immuocompromised patients and may require antibiotic treatment.1,4 Fluoroquinolones and cephalosporins are the drugs of choice for invasive salmonellosis treatment.1,5
The prevalence of Salmonella strains with resistance to fluoroquinolones and beta lactams have increased globally with important impact on hospitalization, therapeutic failure, and mortality.2,3 Antimicrobial resistance to fluoroquinolones and cephalosporins has developed under selective antibiotic pressure in Salmonella species especially in food producing animals and it became a major concern.3,5 The main mechanism of fluoroquinolone resistance is consequence of mutations in the quinolone resistance determining regions (QRDR) of DNA gyrase (gyrA and gyrB) and topoisomerase IV (parC and parE). The mutations confer high-level resistance.6-9 The overexpression of efflux pumps is another fluoroquinolone resistance mechanism.6-9 Plasmid mediated quinolone resistance (PMQR) genes, qnrA, qnrB, qnrC, qnrD, qnrS, aac (6’)-Ib-cr, and qepA are associated with low level resistance to fluoroquinolones with minimum inhibitory concentrations (MIC) around 0.12-4 µg/ml.3,7,9,10 Plasmid mediated quinolone resistance genes are causing treatment failure and are an infection control concern since they are encoded on plasmids, mobile DNA elements, and can be easily acquired via horizontal transfer by different Enterobacteriaceae species.3,8,10 The major β-lactam resistance mechanism in Salmonella is the production of β-lactamases such as extended spectrum β-lactamases (ESBLs) and plasmid mediated AmpC β-lactamases (pAmpC).11-13 According to literature, these β-lactamases can be co-harbored on plasmids along with PMQR genes.1,4 There are few reports of Salmonella infections and antibiotic resistance in Saudi Arabia.14,15 In addition, there are no reports delineating the molecular mechanisms of fluoroquinolones and/or β-lactam resistance in Salmonella strains from Saudi hospitals. The aim of this study was to determine the prevalence of resistance to fluoroquinolones and β-lactams in Salmonella isolates from patients admitted to a Saudi hospital. In addition, this study also aimed to characterize the contribution of mutations in QRDRs (gyrA, gyrB, parC, and parE) and/or PMQR genes, qnrA, qnrB, qnrC, qnrD, qnrS, aac (6’)-Ib-cr, and qepA to fluoroquinolone resistance as well as the involvement of ESBLs and/or pAmpC β-lactamases in third generation cephalosporin resistance.
Methods
Salmonella identification and antimicrobial susceptibility testing
This study was conducted at King Fahad Hospital of the University (KFHU), Alkhobar, Saudi Arabia from October 2015 to December 2016. Non-duplicate specimens were collected from patients admitted to the hospital during the study period. The ethical committee reviewed and approved the study at Imam Abdulrahman Bin Faisal University (IRB 2017-01- 203). Screening for Salmonella was performed on stool specimens using Chromagar Salmonella (Chromagar, Paris, France) as instructed by the manufacturer. Pink colonies on Chromagar were confirmed to be Salmonella using Vitek 2 system (BioMe´rieux, Marcy l’Etoile, France). In addition, Vitek 2 system was used to identify Salmonella from specimens other than stool such as blood, wounds, and urine. Wellcolex color Salmonella kit (Remel Europe, London, UK) was used for Salmonella serotyping as instructed by the manufacturer. Vitek 2 AST-N291 card was used to detect antimicrobial susceptibility testing for trimethoprim-sulfamethoxazole, ampicillin, cefoxitin, imipenem, meropenem, cefotaxime, and cefepime. Susceptibility testing for ciprofloxacin was performed using E test strips (Epsilometer assay; BioMe´rieux, Marcy l’Etoile, France). Escherichia coli (E. coli) ATCC 25922 strain and Pseudomonas aeruginosa ATCC 27853 strain were used as quality control strains. Salmonella strains not susceptible to ciprofloxacin MIC ≥0.12 µg/ml and/or resistant to cefotaxim MIC ≥4 µg/ml were included in this study. Minimum inhibitory concentrations results were interpreted using the criteria of Clinical and Laboratory Standards Institute (CLSI).
Molecular analysis
Polymerase chain reaction (PCR) method was used to detect genes responsible for resistance to fluoroquinolones and β-lactams using specific primers and conditions previously described (Table 1).11,12,16 Positive controls were used in each PCR. For fluoroquinolone resistant Salmonella strains, PMQR genes were tested using primers amplifying qnrA, qnrB, qnrC, qnrD, qnrS, aac (6’)-Ib-cr, and qepA. In addition, gyrA, gyrB, parC, and parE were examined for QRDR mutations. Salmonella isolates resistant to third generation cephalosporins were tested for ESBL production using TEM, SHV, and CTX-M primers. Screening for pAmpC genes in strains resistant to third generation cephalosporins and/or cefoxitin was carried out using Philisa AmpC ID kit (Streck Company, Omaha, NE, USA) as recommended by the manufacturers. The kit detects the following pAmpC genes DHA, CMY, EBC, FOX, ACC, and MOX. ABI 3730xl DNA analyzer (Applied Biosystems, Foster city, CA, USA) was used for amplicon sequencing.
Primers used in this study.
Enterobacterial repetitive intergenic consensus sequence-based PCR (ERIC-PCR) was used to determine the clonal relatedness among Salmonella strains as previously described.17 Gel compar II software version 6 (Applied Maths, Sint-Martens-Latem, Belgium) was used to analyze the DNA fingerprint patterns.
The variables investigated in the current study were non-continuous variables. Thus, the statistical methods used in the manuscript were only descriptive that included assessment of prevalence and percentage of Salmonella resistance to antimicrobial agents tested in this study.
Results
From October 2015 to December 2016, a total of 149 Salmonella species were isolated from patients admitted to KFHU. They were isolated from 76 females and 73 males with ages from 1 to 90 years and average of 25.3 years old. Children younger than 5 years old were most affected with 44 cases (29.5%). The most common source of these isolates was stool (n=140) strains followed by blood (n=6), urine (n=2), and wounds (n=1). The most common Salmonella serogroups detected were B (n=42), D (n=36), and C (n=21) isolates. Serogroups E (n=10), A (n=4), and G (n=3) were also detected. There were 33 untypable Salmonella strains.
All strains were susceptible to imipenem, meropenem, and cefepime. A total of 35 strains were resistant to ampicillin while 6 strains were resistant to cefoxitin which makes the rate of resistance to ampicillin (23.5%; 35/149) and cefoxitin (4%; 4/149). For the third generation cephalosporins, 2 strains were resistant to cefotaxime with a resistance rate of 1.3% (2/149). The resistance rate to trimethoprim-sulfamethoxazole was 17.4% (26/149). Based on Etest results, 29 out of 149 strains were resistant to ciprofloxacin with a resistance rate of 19.5% (29/149).
Using PCR, qnrB and qnrS were detected in 5 and 3 isolates, respectively (Table 2). No qnrA, qnrC, qnrD, aac (6’)-Ib-cr, and qepA genes were detected. In addition, no mutations in the QRDR region of gyrA, gyrB, parC, and parE were identified in these isolates. No ESBLs or pAmpC genes were detected in these isolates.
Characteristics of Salmonella strains harboring PMQR genes.
Using ERIC-PCR, Salmonella strains grouped into 7 clusters with clonal relatedness scores ranging from 89% to 98% (data not shown). Environmental samples collected from different hospital wards did not grow any Salmonella strains and no outbreaks were reported during the study duration.
Discussion
Hospitalization, clinical therapeutic failure, and mortality due to fluoroquinolone resistant Salmonella have increased worldwide.1,6 It is interesting to know that most prevalent cases were identified in children younger than 5 years. Our data correlates with data published from China, Thailand, and USA.1,2,5,18 It is not clear why Salmonella is associated most with children younger than 5 years. Further epidemiological and immunological studies are needed to explain this association.
The prevalence of fluoroquinolone resistance in this study is 19.5% which is significant increase in the Eastern province of Saudi Arabia compared to 3% fluoroquinolone resistance rate in studies conducted in the same region from 2008-2011.14 Multiple factors can contribute to this increase of resistance including over the counter use of antibiotics and misuse of the antibiotics empirically when patients do not necessarily need antibiotic treatment. The fluoroquinolone resistance rate in this study is comparable to rates in other countries such as Palestine (15%), Philippines (14.9%), and Finland (21.3%) while it is higher than that in USA (2.4%), Hong Kong (7.1%), Sri Lanka (8%), and Ghana (6.6%).1,4,5,19-21 In addition, it is lower than fluoroquinolone resistance rates in other countries such as Taiwan (48.1%), Thailand (46.2%), Romania (60%), and Korea (36.5%).3,5,18 Plasmid mediated quinolone resistance genes, qnrB and qnrS, were detected in 8 strains while fluoroquinolone resistance mechanisms could not be identified in 21 strains. It is possible that resistance in these 21 strains may be due to the over-expression of efflux pumps. In addition, efflux pumps may contribute partly to resistance is the 8 strains encoding PMQR genes. However, we did not test for efflux pump overexpression in this study.
Plasmid mediated quinolone resistance genes confer low level fluoroquinolone resistance.6-8,22 However, their detection is crucial since they facilitate the selection of QRDR mutations which results in higher fluoroquinolone resistance.7,10,23 In addition, they are encoded on plasmids which can transfer between organisms creating a major infection control and public health concern.10,23 Plasmid mediated quinolone resistance genes can also be encoded with other antimicrobial resistance determinants such as ESBL limiting the therapeutic options.1,4
The resistance to the third generation cephalosporins was 1.3% which is comparable to that in USA (4.1%) while it is lower than that in China (11%).1,2 Cefepime was fully susceptible in our study compared to 10% resistance in studies recently published from China.2 No ESBLs or pAmpC were detected in these isolates suggesting that resistance to cefotaxime was due to other mechanisms not examined such as permeability changes or other β-lactamases.
Salmonella strains were distributed in 7 clusters showing different susceptibility profiles. The strains were isolated from patients admitted to the hospital from different geographical locations in Saudi Arabia. In addition, no outbreaks were reported during the study period. Taken together, these data suggest that these cases are sporadic and not related to any outbreak.
The performance of this study in a single center is a limitation because having multiple centers involved will present more comprehensive data regarding characterization of Salmonella strains and their susceptibility profiles phenotypically and genotypically. The overexpression of efflux pumps and changes of permeability were not tested which is another limitation factor.
In conclusion, this study showed the prevalence of fluoroquinolone resistance in Saudi Arabia. It also detected for the first time Salmonella isolates harboring plasmid mediated quinolone resistant determinants, qnrB and qnrS. Additionally, this article revealed the significance of conducting nationwide surveillance and epidemiological studies to determine the prevalence and antibiotic resistance mechanisms of an important organism such as Salmonella.
Acknowledgment
The authors would like to thank Imam Abdulrahman Bin Faisal University for supporting this study. This study was supported by Deanship for Scientific Research (DSR) at Imam Abdulrahman Bin Faisal University (IAU), Project number 2014050). The authors would like also to thank Streck Company in Omaha, NE, USA for providing Philisa ampC ID kits free of charge.
Footnotes
Disclosure. Authors have no conflict of interests, and the work was not supported or funded by any drug company.
- Received March 20, 2018.
- Accepted June 13, 2018.
- Copyright: © Saudi Medical Journal
This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.