Association of Salmonella Serotypes with Quinolone Resistance in Broilers

Abstract Fluoroquinolone is widely used for the treatment of bacterial diseases, and the emergence of quinolone resistance has become a serious concern in recent years, owing to an increase and inappropriate use of antimicrobials. Here, we attempted to understand the differences in the emergence frequency of quinolone-resistant bacterial variants in three Salmonella serotypes S. Infantis, S. Schwarzengrund, and S. Manhattan—which are mainly found in broiler industries in Japan. Emergence frequency tests for quinolone-resistant variants using enrofloxacin-containing agar plates and sequence analysis in the quinolone resistance-determining region (QRDR) of gyrA in DNA gyrase were performed. The results showed no significant difference in the emergence frequency among the three serotypes, and most of the resistant variants had mutations in the QRDR region. These findings suggest that differences in the serotypes tested are not associated with the emergence frequency of quinolone-resistant variants.


Introduction
Fluoroquinolone is a widely used antimicrobial agent for the treatment of bacterial infections in humans. It is used as a first-line treatment for severe gastroenteritis, including Salmonella infection, in adults 1) . Emergence of fluoroquinolone-resistant Salmonella increases the risk of treatment failure in patients. At present, three fluoroquinolonesnorfloxacin (NFLX), ofloxacin (OFLX), and enrofloxacin (ERFX)-have been approved for the treatment of bacterial diseases in broiler chicken in Japan. To date, although fluoroquinolone resistance has been reported in limited serotypes (S. Typhimurium in cattle and S. Chorelaesuis in pigs) of Salmonella in Japan 2, 3) , fluoroquinolone resistance has not been observed in Salmonella isolates from poultry.
The development of fluoroquinolone resistance in Enterobacteriaceae, including Salmonella, can be associated with multiple substitutions of amino acids in the target enzymes (DNA gyrase and topoisomerase IV), decreased permeability of drugs, and/or activation of efflux mechanisms 4) . A point mutation in the quinolone resistance-determining region (QRDR) of gyrA, which encodes the GyrA subunit of DNA gyrase, has been recognized to be responsible for quinolone resistance 5) . Moreover, additional mutations in the QRDR doi: 10.14252/foodsafetyfscj.2018012 of gyrA and parC are required for resistance development to fluoroquinolones 4) . The emergence of quinolone-resistant Salmonella with a QRDR mutation has been associated with efflux pump activation in experimental investigations 6) . Additionally, activation of the AcrAB-TolC efflux pump has been reported to be associated with the emergence of quinolone resistance in S. Typhimurium 7) and S. Choleraesuis 8) . Therefore, evaluation of emergence frequencies of quinolone-resistant strains and efflux pump activities of the strains may contribute to the estimation of fluoroquinoloneresistance development potential in the strains.
Salmonella causes various foodborne illnesses and is transferred from animals to humans via animal products 1) . Chicken meat is a common source of foodborne salmonellosis in Japan 9) . Salmonella enterica subsp. enterica serovar Infantis is known to be the major serotype present in broiler meat in Japan 10) . In the last decade, not only S. Infantis but also S. Schwarzengrund and S. Manhattan have been isolated from broiler samples 11) . Of the three serotypes prevalent in the Japanese broiler industry, fluoroquinolone-resistant S. Schwarzengrund has been reported in Thailand and Taiwan as well 12,13) . As a change in Salmonella serotypes prevalent in broiler was observed in Japan, we aimed to determine the possibility of quinolone resistance emergence among three Salmonella serotypes. Therefore, in this study, we investigated the frequency of emergence of quinolone-resistant mutants and their efflux pump activities using S. Infantis, S. Schwarzengrund, and S. Manhattan isolates from broiler chickens.

Materials and Methods
A total of 48 nalidixic acid (NA)-susceptible Salmonella isolates, including 14 strains of S. Infantis, 16 strains of S. Schwarzengrund, and 18 strains of S. Manhattan, from broiler chickens and retail chicken meats collected between 2010 and 2013, were used. The minimum inhibitory concentrations (MICs) of NA and ERFX were determined using broth microdilution methods with commercially available plates (Eiken Chemical Co., Ltd. Tokyo, Japan).
The emergence frequency of quinolone-resistant variants was determined as the ratio of the average number of colonies on agar plates with and without fluoroquinolone. Each strain suspension was adjusted to 10 10 CFUs/mL and inoculated onto Mueller Hinton (MH) agar plates containing different concentrations of ERFX (4 × MIC (4MIC) and 2 × MIC (2MIC)).
Mutation in the QRDR domain of gyrA was analyzed by direct DNA sequencing using two selected strains that appeared in each of the fluoroquinolone-containing agar plates. Briefly, to extract bacterial DNA, the bacterial suspension was boiled in distilled water and centrifuged at 10,000 g for 5 min. The supernatant was stored at -20°C as template DNA. Next, gyrA was amplified using TaKaRa ExTaq (TaKaRa Bio Inc., Kusatsu, Japan) with previously reported primer sets (STGYRA1 and STGYRA2) 14) . The PCR conditions were as follows: 94°C for 3 min, 35 cycles at 94°C for 30 sec, 55°C and 72°C for 30 sec, and 72°C for 10 min. The amplified PCR product was purified using Wizard ® SV Gel and PCR Clean-UP System (Promega, Fitchburg, WI, USA) according to the manufacturer's protocol. DNA sequencing was performed on an ABI Prism 3130 Genetic Analyzer using a BigDye Terminator ver. 3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Waltham, MA, USA).
Quantitative PCR analysis of AcrB expression was performed as described by Usui et al 15) . Briefly, bacterial RNA was extracted using ISOGEN (Nippon Genetics Co. Ltd., Tokyo, Japan). To eliminate genomic DNA, the extracted RNA was treated with recombinant DNase I (TaKaRa Bio Inc.). cDNA was synthesized from these RNA samples using a PrimeScript RT reagent kit (TaKaRa Bio Inc.). Real-time PCR was performed using the Step One Plus TM Real-Time PCR system (Thermo Fisher Scientific), gene-specific primers 7,16) (acrB: acrB-rt1 and acrB-rt2; 16S rRNA: Salm 16S-F and Salm 16S-R1), and the THUNDERBIRD ® SYBR qPCR Mix (Toyobo Co., Ltd, Osaka, Japan). The PCR conditions were as follows: 95°C for 1 min, followed by 40 cycles at 95 °C for 10 s, 60°C for 15 s, and 72°C for 30 s. The expression of acrB was normalized with respect to that of 16S rRNA. The ΔΔCt method was used to calculate fold induction of transcription of a target gene by comparison with a value relative to wild-type strain growth in MH broth.
All statistical analyses to determine differences were performed using one-way analysis of variance followed by Tukey's multiple comparison test. P-values < 0.05 indicated significance.

Results and Discussion
The resistant variants of all three serotypes did not emerge on MH agar containing 4MIC of ERFX. Following culture on MH agar containing 2MIC of ERFX, the average emergence frequencies of the resistant variants were 5.1×10 −9 , 5.0×10 −9 , and 9.1×10 −9 for S. Infantis (n = 14), S. Schwarzengrund (n = 16), and S. Manhattan (n = 18), respectively (Fig. 1a). No significant difference in the emergence frequency of quinolone-resistant variants was observed. To determine mutations in the QRDR region of DNA gyrase, two variants per parental strain were selected and subjected to direct DNA sequencing. In the selected variants, the Ser-83 or Asp-87 mutations in gyrA were found in 85 of the 96 variants ( Table 1). The Ser-83/Asp-87 mutation rates were 42.8/57.1, 50.0/46.9, and 41.7/30.6% in S. Infantis, S. Schwarzengrund, and S. Manhattan, respectively, indicating that the Ser-83/ Asp-87 mutation is observed in most of the resistant strains of the three serotypes. A previous study in Japan showed a higher prevalence of NA resistance in S. Schwarzengrund (21.4%) than in S. Infantis (8.0%) and S. Manhattan (11.8%), although fluoroquinolone resistance was not observed in these strains 11) .
Next, to evaluate the efflux pump activities, we examined the expression levels of acrB in the three serotypes. The average acrB expression levels in S. Infantis (n = 14), S. Schwarzengrund (n = 16), and S. Manhattan (n = 18) were 0.39-, 0.40-, and 0.27-fold of the expression in S. Infantis ATCC 51741, respectively (Fig. 1b). Thus, no significant differences were observed in the average expression of acrB in the three serotypes. Our previous study showed higher arcB expression in quinolone-resistant Salmonella strains than in quinolone-susceptible strains 8) . In this study, the susceptible strains were selected to evaluate the potential emergence of quinolone resistance in each serotype of Salmonella. Additionally, following antimicrobial drug treatment, the Salmonella strains showed increased activation of the AcrAB-TolC efflux pump 15,17) . Further, as Salmonella strains isolated from poultry samples were used, it is unknown whether the bacteria were previously exposed to antimicrobials.
The fluoroquinolone-resistant Salmonella serotype has been reported worldwide 2,3,12,13,18) . However, the fluoroquinolone-resistant strain of S. Infantis found in Serbia exhibited high clonality 18) . Moreover, in Japan, fluoroquinoloneresistant strains of Salmonella in food-producing animals  rarely emerged 2,3) . Although the frequency of emergence of quinolone resistance among the three Salmonella serotypes was not different in this study, continuous surveillance for antimicrobial susceptibility in Salmonella from foodproducing animals is essential to prevent the spread of the resistant bacteria imposed by novel risk factors.