ABSTRACT
Fluoroquinolone (FQ)-resistant Campylobacter jejuni is a serious problem worldwide that limits effective treatment of infections. The traditional detection method depends on bacterial isolation and MIC testing, or traditional PCR, which is time-consuming and hard to identify the FQ-resistant C. jejuni in a high abundance wild-type background. This study aimed to develop a rapid and accurate ddPCR assay to detect FQ-resistant C. jejuni mutants based on the crucial resistance mutation C257T (Thr-86-Ile) in gyrA. Our ddPCR gyrA assay showed high specificity and accuracy. Sanger sequencing and the qPCR assay could only recognize gyrA mutant sequences when the ratios of wild-type/mutant were 1:1 or 10:1, respectively. Our ddPCR gyrA assay was able to detect gyrA mutant sequences in the mixtures with up to at least 1000:1 wild-type/mutant ratios, which suggested a significant advantage to distinguish the low mutant signal from the wild-type background. We further monitored the occurrence of gyrA mutations under ciprofloxacin pressure using our ddPCR gyrA assay, and clearly showed that the transition of a dominant C. jejuni subpopulation from wild-type to gyrA C257T mutant, resulting in FQ-resistance. We tested 52 samples from live chickens and retail chicken meat and showed that four samples contained wild-type/mutant mixtures comprising 1.7%, 28.6%, 53.3%, and 87.0% gyrA C257T mutants, respectively. These results demonstrated that the ddPCR gyrA assay was a highly sensitive alternative method to distinguish and quantify FQ-resistant C. jejuni infections that could help guide the appropriate use of FQs in clinical practice.
IMPORTANCE Campylobacter jejuni is considered to be the leading cause of human bacterial gastroenteritis worldwide, and fluoroquinolones (FQs) are the main choices for the treatment of bacterial gastroenteritis in clinical practice. In theory, antimicrobial susceptibility testing should help us to choose the most appropriate drugs for the treatment. However, to test the susceptibility of C. jejuni to FQs, the standardized method is bacteria isolation and MIC measurement, which will take more than 4 days. In addition, a low abundance of FQ-resistant C. jejuni is also hardly distinguished from a high abundance of wild-type background in the mixed infection. Therefore, the development of rapid and accurate detection technology for FQ-resistant C. jejuni is very important. This study provided a ddPCR gyrA assay, which is a highly sensitive alternative method to distinguish and quantify FQ-resistant C. jejuni infections that may help guide the appropriate use of FQs both in veterinary and human clinical practice.
KEYWORDS: FQ-resistant Campylobacter jejuni, ddPCR, detection, gyrA mutant
INTRODUCTION
Campylobacter jejuni is one of the major foodborne pathogens responsible for human bacterial gastroenteritis (1, 2), accounting for approximately 500 million infections per year globally (3). C. jejuni widely colonizes the intestinal tract of wild and domesticated animals and birds, and contaminated meat products are the main source of human infection (4). The detection of this pathogen mainly depends on bacterial isolation and PCR testing (5, 6). In clinical practice, fluoroquinolones (FQs) are one of the main choices for the treatment of bacterial gastroenteritis. To screen microbes for sensitivity to antimicrobials, the standardized method is MIC measurement according to the Clinical and Laboratory Standards Institute (CLSI), which involves bacterial isolation and identification, followed by the MIC test. However, for C. jejuni, this process takes more than 4 days, which hinders the choice of appropriate drugs during this time.
Epidemiological research indicates the widespread dissemination of FQ-resistant C. jejuni in humans and animals. In our previous studies, we identified a Thr-86-Ile mutation in GyrA that conferred FQ resistance in all C. jejuni isolates from chicken farms and retail chicken meat in central China (7, 8). High FQ resistance rates have also been reported in C. jejuni isolates from pigs (9, 10). An increase in the FQ resistance rate among C. jejuni isolates from patients has also been reported (11). For example, 74.3% of C. jejuni clinical isolates recovered from patients with acute gastroenteritis were found to be resistant to ciprofloxacin in Turkey (12). This rate was 48% in Chile (13) and more than 20% in the USA (11). An investigation of Campylobacter in a pediatric department showed that 77.4% of C. jejuni isolates from children were resistant to ciprofloxacin (14). The World Health Organization has reported that the emergence of FQ-resistant Campylobacter species is becoming a public health issue around the world (15). In theory, antimicrobial susceptibility testing should help us to choose the most appropriate drugs for the treatment of hospitalized patients (16). However, bacterial isolation and purification are required before testing, which are time-consuming bottlenecks for rapid responses. In addition, antibiotic-susceptible and -resistant isolates can be simultaneously present in an individual, but the low proportion of resistant strains is not easily detected in samples during the isolation process. Because of the wait time with antimicrobial susceptibility testing, FQs, such as ciprofloxacin, have been used to treat bacterial diarrhea before any resistance data, affecting therapeutic outcomes and promoting drug-resistant mutations among bacteria. Therefore, a rapid and sensitive method to identify FQ-resistant C. jejuni from clinical samples is required to evaluate the respective proportions of susceptible and resistant populations and the potential risk of treatment failure.
FQs target the DNA gyrase and topoisomerase IV of bacteria, inhibiting DNA replication, and transcription, and ultimately leading to cell death (17). Bacterial resistance to FQs is mainly related to chromosomal mutations in the quinolone resistance-determining regions (QRDRs) of DNA gyrase (encoded by the gyrA and gyrB genes), and partly to topoisomerase IV (encoded by parC and parE), as well as plasmid-mediated quinolone resistance (PMQR) genes, such as qnr and aac(6′)-Ib-cr (18, 19). In addition, the overexpression of efflux pumps also plays a role in the resistance to FQs (20, 21). In C. jejuni, the main reported mechanisms of FQ-resistance include mutation of the gyrA gene and alterations to multidrug efflux pumps. Mutations in the promoter of the cmeABC operon were shown to lead to overexpression of the CmeABC efflux pump and partly contributed to low-level resistance to antimicrobial agents in C. jejuni (7). A Thr-86-Ala mutation in GyrA (A256G in gyrA) has also been reported to confer resistance to nalidixic acid but not to FQs (22). However, high-level resistance to FQs (ciprofloxacin MIC >16 g/mL) in C. jejuni has only been reported to be caused by a Thr-86-Ile single mutation in GyrA (C257T in gyrA) (17, 23).
Droplet digital PCR (ddPCR) is a new technology that enables the absolute quantification of nucleic acids with high analytical sensitivity, specificity, and accuracy (24). To increase the sensitivity of resistance detection, a ddPCR gyrA assay, based on the resistant mutations of gyrA, was developed for the detection and quantification of FQ-resistant C. jejuni in this study.
RESULTS
Sanger sequencing of the gyrA QRDR of the wild-type/mutant mixtures.
To assess the sensitivity of Sanger sequencing, the mixtures containing WT (pWT-gyrA) and MT1 (pMT1-gyrA) or MT2 (pMT2-gyrA) were analyzed. As shown in Fig. 1, C/T heterozygous peaks at the 257 site of gyrA in the mixture of WT and MT1, and A/G heterozygous peaks at the 256 site of gyrA in the mixture of WT and MT2 were detected at a 1:1 WT/MT ratio. In contrast, the MT1 and MT2 sequences were only occasionally detected at a 10:1 WT/MT ratio and were not detected at 100:1 and 1000:1 WT/MT ratios. This result suggested that the distinguishability of Sanger sequencing was at a 1:1 ratio of WT/MT.
FIG 1.
Sanger sequencing of the WT/MT mixtures. (A) WT and MT1 mixtures at 1:1, 10:1, 100:1, and 1000:1. (B) WT and MT2 mixtures at 1:1, 10:1, 100:1, and 1000:1. The arrows indicated the mutant peaks.
qPCR gyrA assay of the wild-type/mutant mixtures.
The specificity of the qPCR gyrA assay for detecting the WT (VIC), MT1 (FAM), and MT2 (Cy5) are shown in Table 1 and Fig. 2. The primers and probes are shown in Table 2. In Fig. 2B, a weak Cy5 fluorescence signal was detected when only WT was added to the reaction, which suggested a weak cross-reaction between the MT2 probe and the WT. To avoid the effects of weak fluorescence signals in the presence of mismatched targets, the thresholds of the fluorescence signals (shown in red in Fig. 2) were confirmed by detecting every single target. The qPCR gyrA assay was specific for detecting the WT, MT1, and MT2 sequences. No fluorescence signals were detected from species other than C. jejuni (Table 1).
TABLE 1.
Strains used for specificity analysis of qPCR and ddPCR in this study
| Strainsa or plasmids | qPCR |
ddPCR |
||||
|---|---|---|---|---|---|---|
| WT | MT1 | MT2 | WT | MT1 | MT2 | |
| pWT-gyrA | +b | − | − | + | − | − |
| pMT1-gyrA | − | + | − | − | + | − |
| pMT2-gyrA | − | − | + | − | − | + |
| Campylobacter jejuni NCTC 11168 (WT) | + | − | − | + | − | − |
| Campylobacter jejuni WH31 (WT) | + | − | − | + | − | − |
| Campylobacter jejuni DY01 (MT1) | − | + | − | − | + | − |
| Campylobacter jejuni JSJ01 (MT1) | − | + | − | − | + | − |
| Campylobacter coli JSJ03 | − | − | − | − | − | − |
| Campylobacter coli DY07 | − | − | − | − | − | − |
| Escherichia coli DH5α | − | − | − | − | − | − |
| Escherichia coli CICC 21530 | − | − | − | − | − | − |
| Salmonella pullorum CVCC 519 | − | − | − | − | − | − |
| Salmonella enteritidis CVCC 540 | − | − | − | − | − | − |
| Pasteurella multocida CVCC 1502 | − | − | − | − | − | − |
| Staphylococcus aureus CVCC 545 | − | − | − | − | − | − |
| Clostridium perfringens CVCC 79 | − | − | − | − | − | − |
| Enterococcus faecalis CVCC 1927 | − | − | − | − | − | − |
C. jejuni strains WH31, DY01, JSJ01, and C. coli strains JSJ03, DY06 were isolated in China and were kept in our lab. E. coli DH5α was kept in our lab. Other strains were purchased from the China Institute of Veterinary Drug Control.
+, means test positive; −, means test negative.
FIG 2.
qPCR discrimination plots for the WT/MT mixtures. (A) WT and MT1 mixtures at 1:0, 0:1, 1:1, 10:1, 100:1, and 1000:1. (B) WT and MT2 mixtures at 1:0, 0:1, 1:1, 10:1, 100:1, and 1000:1. Black square indicates the negative control. The threshold is showed in red.
TABLE 2.
Primers and probes used for qPCR and ddPCR assays in this study
| Primer or probe names | Sequences and chemical modifications |
|---|---|
| C-gyrA-F | 5′-CCCGTATAGTGGGTGCTGTT-3′ |
| C-gyrA-R | 5′-TCTTGAGCCATTCTAACCAAAG-3′ |
| C-gyrA-WT-P | 5′-VIC-ATGGAGATACAGCAGTTTAT-MGB-3′ |
| C-gyrA-MT1-P | 5′-FAM-ATGGAGATATAGCAGTTTAT-MGB-3′ |
| C-gyrA-MT2-P | 5′-Cy5-CCACATGGAGATGCAGCAGT-MGB-3′ |
To assess the sensitivity of the qPCR gyrA assay, the mixtures containing WT and MT1 or MT2 were analyzed. As shown in Fig. 2, both WT (VIC) and MT1 (FAM) or MT2 (Cy5) fluorescence signals were detected at 1:1 and 10:1 WT/MT ratios, respectively. In contrast, only the WT (VIC) fluorescence signal was detected at the 100:1 and 1000:1 WT/MT ratios. This result suggested that the distinguishability of the qPCR gyrA assay was at a 10:1 ratio of WT/MT.
Validation of the ddPCR gyrA assay.
The specificity of the ddPCR gyrA assay was evaluated by testing the constructed standard plasmids (Table 2), FQ-susceptible and FQ-resistant C. jejuni strains, and strains other than C. jejuni. As shown in Table 1, the ddPCR gyrA assay could detect and distinguish WT, MT1, and MT2 targets specifically, and the species other than C. jejuni were all negative. When testing the mixture of WT, MT1, and MT2 (1:1:1 ratio), the positive droplets from each plasmid template were in their corresponding fluorescent phase from the three-dimensional diagrams (Fig. 3A).
FIG 3.
Validation of the ddPCR gyrA assay. (A) The three-dimensional diagrams for the ddPCR of the WT, MT1, and MT2 mixtures. Green dots represent wells with the VIC signal (corresponding to the WT); blue dots represent wells with the FAM signal (corresponding to MT1); red dots represent wells with the FAM signal (corresponding to MT2), and gray dots represent wells with no amplification signal. (B to D) The standard curves for the ddPCR are based on serial dilutions of the WT (B), MT1 (C), and MT2 (D).
Next, serial dilutions of WT, MT1, and MT2 were tested. The ddPCR gyrA assay showed good sensitivity and linearity at low copy number concentrations of targets (less than 3 × 105 copies/reaction), and the concentrations provided by the software were close to the actual concentrations added to the reactions (Fig. 3B to D). For the WT, the slope was 0.9884, and the R2 was 0.9961. For MT1, the slope was 0.9951, and the R2 was 0.9967. For MT2, the slope was 0.9697, and the R2 was 0.9921. The minimum limit of detection reached the order of single-digit copies/reaction. However, when the concentrations of plasmids were at 106 orders of magnitude of copies/reaction, the droplets clustered together and concentrations could not be obtained by ddPCR (unpublished data), which suggested that proper dilution was necessary.
ddPCR gyrA assay of the wild-type/mutant mixtures.
To assess the sensitivity of the ddPCR gyrA assay, the mixtures of WT and MT were analyzed. The percentages of MT in the mixtures were calculated using the formula: MT copies/(MT copies + WT copies). As shown in Table 3 and Fig. 4, when MT1 or MT2 was added to the WT at 1:1 to 1000:1 WT/MT ratios, MT1 or MT2 was detected in the mixture, and the percentages of MT were generally consistent with the original ratios of the mixtures. This result suggested that the distinguishability of the ddPCR gyrA assay reached a 1000:1 ratio of WT/MT at least, which means that 0.1% MT1 or MT2 sequences could be detected accurately in WT and MT mixtures.
TABLE 3.
The ddPCR results for the gyrA wild-type and mutant mixtures
| Mutants (mutation) | WT/MT ratio | No. of copies/μL |
Mean % of mutants (SD) |
|||||
|---|---|---|---|---|---|---|---|---|
| WT (VIC) |
MT1 (FAM) or MT2 (Cy5) |
|||||||
| Chip1 | Chip2 | Mean (SD) | Chip1 | Chip2 | Mean (SD) | |||
| MT1 (C257T) | 1:1 | 237.40 | 242.00 | 239.70 (3.25) | 246.20 | 240.30 | 243.25 (4.17) | 50.37 (0.77) |
| 10:1 | 2,062.00 | 2,215.00 | 2,138.50 (108.19) | 201.70 | 184.00 | 192.85 (12.52) | 8.29 (0.88) | |
| 100:1 | 2,750.00 | 2,879.00 | 2,814.50 (91.22) | 20.70 | 21.50 | 21.10 (0.57) | 0.74 (0.004) | |
| 1,000:1 | 2,429.00 | 2,460.00 | 2,444 (21.92) | 5.22 | 7.37 | 6.30 (1.52) | 0.26 (0.06) | |
| MT2 (A256G) | 1:1 | 272.30 | 283.90 | 278.10 (8.2) | 239.10 | 250.40 | 244.75 (7.99) | 46.81 (0.08) |
| 10:1 | 2,362.00 | 2,428.00 | 2,395.00 (46.67) | 245.90 | 249.10 | 247.5 (2.26) | 9.37 (0.09) | |
| 100:1 | 2,784.00 | 2,840.00 | 2,812.00 (39.60) | 20.90 | 20.20 | 20.55 (0.49) | 0.73 (0.03) | |
| 1,000:1 | 2,880.00 | 2,895.00 | 2,887.50 (10.61) | 2.82 | 2.58 | 2.58 (0.17) | 0.093 (0.01) | |
FIG 4.
ddPCR gyrA scatterplots for the WT/MT mixtures: (A) WT and MT1 mixtures at 1:1, 10:1, 100:1, and 1000:1, respectively. (B) WT and MT2 mixtures at 1:1, 10:1, 100:1, and 1000:1, respectively. Gray dots represent wells with no amplification signal; green dots represent wells with the VIC signal (corresponding to the WT); dark blue dots represent wells with the FAM signal (corresponding to MT1); light blue dots represent wells with the FAM plus VIC signal; red dots represent wells with the Cy5 signal (corresponding to MT2), and yellow dots represent wells with the Cy5 plus VIC signal.
Analysis of the gyrA mutations under ciprofloxacin pressure.
To further evaluate our ddPCR gyrA assay, the occurrence of gyrA mutations in FQ-susceptible C. jejuni strain NCTC 11168 (wild-type gyrA) under 10-fold MIC of ciprofloxacin (1 μg/mL) was monitored for 78 h. As shown in Fig. 5A, the optical density at 630 nm (OD630) reduced from 1.6 to 0.3 in the first 6 h after the addition of ciprofloxacin and then began to increase at 42 h posttreatment. As shown in Fig. 5B, the WT maintained a stable level ranging from 5.40 × 104 to 1.07 × 105 copies/μL during the beginning of the monitoring from the 6 h posttreatment, which suggested that the proliferation of wild-type C. jejuni was arrested. In contrast, 1.87 copies/μL of MT1 were detected 48 h after ciprofloxacin treatment, after which MT1 increased rapidly over the following 30 h and became the dominant gyrA type at 78 h (4.83 × 105 copies/μL). During the monitoring, a few copies of MT2 irregularly appeared from 42 to 78 h. However, the abundance was lower than 2.71 × 102 copies/μL, and there was no tendency to proliferate. Finally, the cultures were spread onto plates and 20 colonies were picked at 78 h posttreatment. Of these, 19 possessed the gyrA C257T mutation (MT1) and were ciprofloxacin-resistant, and only one was WT. This result further confirmed the identification results of the ddPCR gyrA assay, that MT1 was the dominant gyrA type in the cultures after 78 h of ciprofloxacin pressure.
FIG 5.
The transition of dominant C. jejuni subpopulations under ciprofloxacin pressure monitored by the ddPCR gyrA assay. (A) The survival curve of C. jejuni under ciprofloxacin pressure. (B) The transition of dominant C. jejuni subpopulations under ciprofloxacin pressure.
Analysis of anal swabs and retail chicken meat samples.
Fifty-two anal swab samples and retail chicken meat samples collected from chicken farms and markets were detected using the ddPCR gyrA assay. As shown in Table 4, one sample showed single VIC-positive droplets, indicative of the WT (FQ-susceptible C. jejuni strain). Twenty-two samples showed single FAM-positive (MT1) droplets, indicative of MT1 (FQ-resistant C. jejuni strain). No single Cy5 signals (MT2) were detected. Four samples showed both VIC and FAM-positive droplets, indicative of both WT and MT1 (mixture of FQ-susceptible and FQ-resistant C. jejuni strains), and the percentages of MT1 in these four mixture samples were 1.7%, 28.6%, 53.3%, and 87.0%, respectively. One sample showed both VIC and Cy5-positive droplets, indicative of both WT and MT2 (mixture of FQ-susceptible C. jejuni strains), and the percentage of MT2 was 66.4%. Further bacterial isolation and FQ susceptibility tests also confirmed our ddPCR gyrA assay results.
TABLE 4.
The ddPCR results for detection of the clinical samples
| ddPCR gyrA assays (no. = 52) | Samples (no.) | Resistant isolates |
Susceptible isolates |
||
|---|---|---|---|---|---|
| No. | MIC (μg/mL) | No. | MIC (μg/mL) | ||
| VIC positive (WT) | 1 | 0 | — | 1 | 0.125 |
| FAM positive (MT1) | 22 | 22 | 4-64a | 0 | — |
| Cy5 positive (MT2) | 0 | 0 | — | 0 | — |
| VIC&FAM positive (WT+MT1) | 4 | 4 | 4-32 | 4 | 0.25-0.5 |
| VIC&Cy5 positive (WT+MT2) | 1 | 0 | — | 2 | 0.125/0.5b |
| Negative | 24 | 0 | — | 0 | — |
The MICs of these 22 isolates were in the ranges of 4 to 64 μg/mL.
The MIC of WT isolate was 0.125 μg/mL, and the MIC of MT2 isolate was 0.5 μg/mL.
DISCUSSION
C. jejuni is one of the most important foodborne pathogens, causing severe human gastroenteritis worldwide (3, 25). Antibiotic therapy is the main strategy for the treatment of C. jejuni infection, and the choice of drugs is particularly important for effective treatment. FQs (i.e., ciprofloxacin) are the most used antibiotics to treat acute bacterial diarrhea in humans and animals (17). Recently, FQ resistance in C. jejuni has increased because of the widespread use of this class of antibiotics (26, 27). In clinical practice, many cases are treated empirically with FQs without epidemiological or drug susceptibility analysis, which has limited the therapeutic effects of FQs. This is likely because such analyses are time-consuming using traditional methods (6). Therefore, in this study, we tried to establish a rapid and sensitive method for the identification of FQ-resistant C. jejuni from clinical samples.
The FQ resistance mechanisms of C. jejuni have been well studied (17). Unlike Escherichia coli and Salmonella, plasmid-mediated quinolone resistance genes have not been found in C. jejuni (18, 28). Although a variety of mutations in the DNA gyrase (encoded by gyrA and gyrB) and topoisomerase IV (encoded by parC and parE) have been identified in C. jejuni isolates by sequencing analysis, only a single point mutation, Thr-86-Ile in GyrA, is currently required for high-level resistance (ciprofloxacin MIC > 16 μg/mL) (29, 30). Asp90-Asn and Ala-70-Thr mutations in GyrA are rare and only confer intermediate-level resistance to FQs (29). Several studies have demonstrated that C. jejuni lacks the parC and parE genes, which strongly suggested that parC and parE cannot be a source of FQ resistance (31). In epidemiological investigations, the Thr-86-Ile mutation was detected in almost all high-level FQ-resistant C. jejuni isolates (7, 30). Based on this relatively specific characteristic of resistant mutations in the genome of C. jejuni, detection of FQ-resistant C. jejuni based on a molecular method was possible. Thus, we developed a ddPCR gyrA assay to detect and quantify FQ-resistant C. jejuni. We designed three probes, one targeting wild-type gyrA, one targeting the resistance mutation C257T in gyrA (Thr-86-Ile), and one targeting the gyrA A256G mutation (Thr-86-Ala), although this mutation was unusual and could not cause FQ resistance (22).
Sanger sequencing is the most used method to identify mutations in genes, and qPCR is another popular method. Therefore, we compared our ddPCR gyrA assay with Sanger sequencing and qPCR methods. With purified single samples, Sanger sequencing, the qPCR gyrA assay, and the ddPCR gyrA assay all showed good levels of detection. However, the ddPCR gyrA assay was able to accurately determine the copy number in each sample with good reproducibility (Fig. 3) even in the WT/MT mixtures (Table 3). Our ddPCR quantitative method did not require the determination of a standard curve, as for qPCR, or bacterial culture to provide viable counts. It is also worth noting that a weak cross fluorescence signal existed between the MT2 probe and WT gyrA target. Weak cross-reactivity is difficult to avoid in the detection of single nucleotide mutations based on qPCR using a TaqMan probe, so proof of threshold was necessary (32). However, cross-detection did not appear in the ddPCR gyrA assay. We inferred that because samples are divided into single droplets in the ddPCR system, any cross-fluorescence signals would be too weak in each droplet to be detected by the ddPCR system. Therefore, compared with the traditional detection method, the ddPCR gyrA assay is a fast and convenient bacterial quantitative method and can avoid nonspecific reactions.
When detecting mixed samples, gyrA mutant sequences were detected in mixtures of up to at least 1000:1 WT/MT ratios (0.1%) using the ddPCR gyrA assay. In contrast, Sanger sequencing and qPCR could only recognize gyrA mutant sequences at ratios of 1:1 or 10:1, respectively. Therefore, Sanger sequencing and qPCR might not be able to detect antibiotic-resistant subpopulations of C. jejuni in the early stage of infection. In principle, the ddPCR reaction mixture is divided into tens of thousands of independent units that are detected individually, so the sensitivity and discernibility of ddPCR are much higher than conventional PCR and qPCR methods. Therefore, our ddPCR gyrA assay offers a significant advantage for the detection of low-abundance FQ-resistant C. jejuni mutants, especially in mixed infections of FQ-resistant and susceptible strains. Recently, ddPCR was also used for detecting resistance mutations in Mycobacterium tuberculosis and Legionella pneumophila, reaching 1% to 0.1% distinguishability (32, 33). A maximum distinguishability of ddPCR of 0.001% could be reached using an improved method to obtain a higher number of individual droplets (24). Therefore, compared with the traditional method, the ddPCR gyrA assay has an obvious advantage in sensitivity and distinguishability.
The advantage of the high sensitivity of the ddPCR gyrA assay was also demonstrated by monitoring the occurrence of gyrA mutations in the wild-type C. jejuni strain under 1 μg/mL ciprofloxacin selective pressure. This concentration of ciprofloxacin has been reported to reflect drug metabolism after oral administration of ciprofloxacin in chickens (34). As shown in Fig. 5, the MT2 sequence only appeared occasionally, and the MT1-type mutant gradually became the dominant subpopulation during culturing in the presence of ciprofloxacin. This result further suggested that the ddPCR gyrA assay was useful for detecting antibiotic-resistant subpopulations of C. jejuni at an early stage of the resistance emergence. Based on the monitoring, we also determined the characteristics of FQ-resistant mutant outbreaks under FQ pressure. This regularity was consistent with the high prevalence rate of FQ-resistant C. jejuni harboring the gyrA C257T mutation in animals and humans (7, 10).
When the ddPCR gyrA assay was applied to clinical samples, it was clear that FQ-resistant C. jejuni harboring the gyrA C257T mutation was the main epidemic strain. Mixed infections with FQ-resistant and susceptible isolates were also frequently detected. It is worth noting that many clones were screened to detect the resistant isolates in the mixed samples, especially in the samples with a low proportion of MT1. This detailed work may not be feasible in the clinic, so resistant isolates may be easily neglected. Furthermore, the gyrA C257T mutation readily occurred after FQ treatment in C. jejuni. Therefore, our ddPCR gyrA assay may have important applications for detecting resistance mutations at an early stage, allowing for timely adjustment of the treatment regimen.
Although the ddPCR gyrA assay has remarkable advantages for the detection of FQ-resistant C. jejuni, it also has some limitations. One limitation is the maximum detection limit, which means that proper dilution may be required for detection in cases of severe infection. Another limitation is that it cannot distinguish dead bacteria from viable bacteria based on DNA detection. In addition, this assay is based on the GyrA Thr-86-Ile mutation. Although this is the main FQ resistance mechanism, which covered most of the FQ-resistant C. jejuni, a small number of resistant strains with other resistance mechanisms may be missed. Therefore, there is a need for continuous monitoring to detect the emergence of new FQ-resistant C. jejuni strains.
In conclusion, based on the reported crucial resistance mutation in gyrA, a ddPCR gyrA assay for detection and quantification of FQ-resistant C. jejuni was developed. Compared with the traditional method, the ddPCR gyrA assay could avoid nonspecific reactions well and had high distinguishability to recognize the FQ-resistant mutants. It was also convenient for bacterial quantification. Therefore, it has multiple advantages in the detection of FQ-resistant C. jejuni and may have important applications in clinical practice to help guide the appropriate use of FQs.
MATERIALS AND METHODS
Bacterial strains and culture conditions.
The strains used for specificity analysis in this study are listed in Table 1. C. jejuni NCTC 11168 was used as an FQ-susceptible control. The FQ-resistant strain C. jejuni DY01, containing a C257T mutation in gyrA, was used as an FQ-resistant control. C. jejuni DY01 was isolated from a chicken farm in Hubei, China (35) and kept in our laboratory. For specificity testing, 14 reference or clinical strains, including four C. jejuni strains and 10 strains of species other than C. jejuni were used. C. jejuni and Campylobacter coli strains were grown on Bolton broth (Oxoid, Basingstoke, UK) or modified charcoal cefoperazone desoxycholate agar (mCCDA) plates containing 1% Campylobacter growth and selective supplements (Oxoid) for 48 h at 42°C in air-tight jars containing AnaeroPack (Mitsubishi, Japan) to generate microaerobic conditions. Escherichia coli and Salmonella were grown in Luria-Bertani medium (LB, Oxoid) at 37°C. Pasteurella multocida and Staphylococcus aureus were grown in tryptic soy broth medium (TSB, BD, Sparks, MD, USA) at 37°C. Clostridium perfringens was cultured anaerobically in a Brain-Heart Infusion medium (BHI, BD) at 37°C. Enterococcus faecalis was cultured in a BHI medium at 37°C.
Preparation of standard plasmids.
Genomic DNA of bacteria was extracted using the MiniBEST Universal Genomic DNA Extraction kit (TaKaRa, Dalian, China) according to the manufacturer’s instructions. An internal part (453 bp) of the QRDR region of gyrA was amplified from C. jejuni NCTC 11168 (wild-type gyrA, defined as WT) or C. jejuni DY01 (gyrA C257T mutant, defined as MT1) using forward primer (gyrA-F: 5′-GCCTGACGCAAGAGATGGTT-3′) and reverse primer (gyrA-R: 5′-TGAGGTGGGATGTTTGTCGC-3′), and then cloned into a pMD18-T vector (TaKaRa). Because no C. jejuni strain contained a naturally occurring A256G mutation in gyrA in our lab at the beginning of this study, the gyrA fragment containing the A256G mutation (defined as MT2) was synthesized and cloned into pMD18-T by Sangon Biotech (Shanghai, China). The three constructed plasmids were confirmed by DNA sequencing and were defined as pWT-gyrA, pMT1-gyrA, and pMT2-gyrA, respectively (Table 1). The concentrations of plasmids were measured using the NanoDrop spectrophotometer (Thermo Fisher Scientific, MA, USA).
Isolation, identification, and FQ susceptibility testing of C. jejuni isolates.
Freshly collected swabs were stored in Cary–Blair modified transport media (Hopebio, Qingdao, China), and was then transported to our laboratory. The samples were resuspended in 0.5 mL PBS and inoculated into 5 mL Bolton broth containing Campylobacter growth and selective supplements, which was incubated for 24 h at 42°C under microaerobic conditions. Following this, 100 μL of the culture was spread onto mCCDA plates containing Campylobacter selective supplements and incubated for 48 h at 42°C under microaerobic conditions (36). Tentative C. jejuni colonies were further purified and identified by PCR targeting hipO as described previously (37). The MIC of ciprofloxacin of C. jejuni isolates was tested according to the CLSI guidelines (38). Strain C. jejuni ATCC 33560 was used for quality control of MIC tests.
gyrA gene amplification and Sanger sequencing.
To assess the sensitivity of the PCR-sequencing approach, standard plasmid pWT-gyrA was mixed with pMT1-gyrA or pMT2-gyrA at 1:1, 10:1, 100:1, and 1000:1 ratios and were then tested by PCR and sequencing. The QRDR of gyrA was amplified by PCR using forward primer (gyrA-F: 5′-GCCTGACGCAAGAGATGGTT-3′′) and reverse primer (gyrA-R: 5′-TGAGGTGGGATGTTTGTCGC-3′′) as previously described (7). The 453-bp PCR products were sent to Sangon Biotech and sequenced using the same primers on a 3730xl DNA Analyzer system (ABI, USA). The sequences obtained were visualized and analyzed with DNAStar (Version 7.1, USA).
Primers and probes designed for qPCR and ddPCR.
The primers and probes were designed based on the wild-type and mutant gyrA sequences of C. jejuni. The primer and probe sequences are listed in Table 2. The primers were used to amplify the core fragment of QRDR in gyrA, and the probes were designed to distinguish wild-type gyrA from two gyrA gene mutants of C. jejuni (MT1: C257T and MT2: A256G).
qPCR validation and study of the wild-type/mutant mixtures.
The qPCR assays were performed using the LightCycler system (Roche, USA). Each of the qPCR mixtures (20 μL) contained 10 μL of Probe qPCR Mix with UNG (TaKaRa), 0.6 μM of each primer, 0.4 μM of each TaqMan probe, 2 μL of the tested DNA sample, and nuclease-free water. The qPCRs were performed with an initial denaturation step at 95°C for 30 s, followed by 40 cycles of denaturation at 95°C for 5 s, and annealing and extension at 60°C for 30 s.
Validation of the qPCR assays for the specific detection of gyrA WT, MT1, and MT2 sequences was performed by testing the pWT-gyrA, pMT1-gyrA, and pMT2-gyrA plasmids, and reference or clinical strains, including four C. jejuni strains and 10 strains of species other than C. jejuni (Table 1). To assess the sensitivity of the qPCR, 1 ng/μL of pWT-gyrA was mixed with 1 ng/μL of pMT1-gyrA or pMT2-gyrA at ratios of 1:1, 10:1, 100:1, and 1000:1. Therefore, the final concentration of the mixed plasmids (pWT-gyrA + pMT-gyrA) was 1 ng/μL. The mixtures were then tested by qPCR with three replicates.
ddPCR validation and study of the wild-type/mutant mixtures.
The ddPCR gyrA assays were performed using the Naica System digital PCR platform (Stilla Technologies, Paris, France). Each of the ddPCR mixtures (25 μL) contained 12.5 μL of 2× PerfeCT qPCR ToughMix UNG (Stilla Technologies), 1.2 μM each primer, 250 nM each TaqMan probe, 100 nM fluorescein, and 2.5 μL of diluted DNA (i.e., 10 pg to 1 μg of DNA per reaction). The mixture was added to the sapphire chip, and the ddPCR assays were performed using the following program: 40°C for 40 min for droplet generation, followed by 95°C for 10 min, 40 cycles of 95°C for 15 s, 60°C for 30 s, and a final pressure release step for approximately 10 min. After PCR amplification, sapphire chips were scanned by the Naica Prism 3 Reader (Stilla Technologies). The resulting images were visualized and analyzed using the Crystal Miner software (Stilla Technologies).
The specificity of the ddPCR gyrA assay was tested using the standard plasmids, and reference or clinical strains (Table 1). To assess the linearity and detection limits of the ddPCR, 10-fold serial dilutions of each standard plasmid were tested with three replicates. To further assess the sensitivity of the ddPCR, pWT-gyrA was mixed with pMT1-gyrA or pMT2-gyrA at ratios of 1:1, 10:1, 100:1, and 1000:1, and these mixtures were tested by the ddPCR gyrA assay with three replicates.
Mutations induced by the addition of ciprofloxacin and ddPCR detection.
To detect the occurrence of resistant mutations of FQ-susceptible C. jejuni under ciprofloxacin pressure, a susceptible C. jejuni strain NCTC 11168 was cultured to OD630 = 1.6 in Bolton broth under microaerobic conditions. The cultures were then centrifuged at 3000 g for 5 min to remove the medium. The pellets were resuspended in 200 mL fresh Bolton broth (OD630 = 1.6) with the addition of 1 μg/mL of ciprofloxacin and then cultured under microaerobic conditions. Every 6 h, OD630 was measured, 10 mL of culture was collected, and the DNA was extracted and 100-fold diluted. The mutations were detected and quantitated using ddPCR gyrA assays. To ensure that low copies of mutations were not missing, if the mutations were not detected, the undiluted DNA would be detected again. Three biological replications were performed. Finally, 10 μL of culture was collected at 78 h after ciprofloxacin pressure and diluted into 90 μL PBS. The diluent was spread on an mCCDA plate without ciprofloxacin and grown for 48 h at 42°C under microaerobic conditions. Following this, 20 colonies were picked randomly, and their gyrA were amplified by PCR and sequenced as described above.
ddPCR gyrA assay on clinical samples.
Fifty-two samples were collected from chicken cloacas and the surface of retail chicken meat using swabs. The chicken cloacas swabs were collected from the 280 to 300 days old Chinese local chickens, which were ready to be sold and then butchered. The retail chicken meats were purchased from the markets. Freshly collected swabs were resuspended in 0.5 mL of PBS, followed by centrifugation at 12000 g for 2 min. The supernatant was removed, and the DNA of sediment was extracted using the DNeasy PowerSoil kit (Qiagen, Dusseldorf, Germany) according to the manufacturer’s instructions. For detection, 2.5 μL of DNA (approximately 10 ng/μL) was tested by a ddPCR gyrA assay, as described above. To further confirm the results of ddPCR gyrA assays, C. jejuni was isolated and purified from the collected samples, and ciprofloxacin susceptibility was tested as described above.
ACKNOWLEDGMENTS
This work was supported by the National Key Research and Development Program of China (2021YFD1800401), the National Natural Science Foundation of China (31873017), the China Agriculture Research System of MOF, and MARA (CARS-41-G13), and the Foreign expert's project of Hubei Province (2020BJH029).
We declare no conflict of interest.
Contributor Information
Qingping Luo, Email: qingping0523@163.com.
Tengfei Zhang, Email: tfzhang23@163.com.
Sandeep Tamber, Health Canada.
REFERENCES
- 1.Coker AO, Isokpehi RD, Thomas BN, Amisu KO, Obi CL. 2002. Human campylobacteriosis in developing countries. Emerg Infect Dis 8:237–244. doi: 10.3201/eid0803.010233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Young KT, Davis LM, Dirita VJ. 2007. Campylobacter jejuni: molecular biology and pathogenesis. Nat Rev Microbiol 5:665–679. doi: 10.1038/nrmicro1718. [DOI] [PubMed] [Google Scholar]
- 3.Epps SV, Harvey RB, Hume ME, Phillips TD, Anderson RC, Nisbet DJ. 2013. Foodborne Campylobacter: infections, metabolism, pathogenesis and reservoirs. Int J Environ Res Public Health 10:6292–6304. doi: 10.3390/ijerph10126292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Vohra P, Chintoan-Uta C, Bremner A, Mauri M, Terra VS, Cuccui J, Wren BW, Vervelde L, Stevens MP, Glycoengineering of Veterinary Vaccines C . 2021. Evaluation of a Campylobacter jejuni N-glycan-ExoA glycoconjugate vaccine to reduce C. jejuni colonisation in chickens. Vaccine 39:7413–7420. doi: 10.1016/j.vaccine.2021.10.085. [DOI] [PubMed] [Google Scholar]
- 5.Vizzini P, Vidic J, Manzano M. 2021. Enrichment free qPCR for rapid identification and quantification of Campylobacter jejuni, C. coli, C. lari, and C. upsaliensis in chicken meat samples by a new couple of primers. Foods 10:2341. doi: 10.3390/foods10102341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gharst G, Oyarzabal OA, Hussain SK. 2013. Review of current methodologies to isolate and identify Campylobacter spp. from foods. J Microbiol Methods 95:84–92. doi: 10.1016/j.mimet.2013.07.014. [DOI] [PubMed] [Google Scholar]
- 7.Zhang T, Cheng Y, Luo Q, Lu Q, Dong J, Zhang R, Wen G, Wang H, Luo L, Wang H, Liu G, Shao H. 2017. Correlation between gyrA and CmeR box polymorphism and fluoroquinolone resistance in Campylobacter jejuni isolates in China. Antimicrob Agents Chemother 61:e00422-17. doi: 10.1128/AAC.00422-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhang T, Luo Q, Chen Y, Li T, Wen G, Zhang R, Luo L, Lu Q, Ai D, Wang H, Shao H. 2016. Molecular epidemiology, virulence determinants and antimicrobial resistance of Campylobacter spreading in retail chicken meat in Central China. Gut Pathog 8:48. doi: 10.1186/s13099-016-0132-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Marotta F, Di Marcantonio L, Janowicz A, Pedonese F, Di Donato G, Ardelean A, Nuvoloni R, Di Giannatale E, Garofolo G. 2020. Genotyping and antibiotic resistance traits in Campylobacter jejuni and coli from pigs and wild boars in Italy. Front Cell Infect Microbiol 10:592512. doi: 10.3389/fcimb.2020.592512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tang M, Zhou Q, Zhang X, Zhou S, Zhang J, Tang X, Lu J, Gao Y. 2020. Antibiotic resistance profiles and molecular mechanisms of Campylobacter from chicken and pig in China. Front Microbiol 11:592496. doi: 10.3389/fmicb.2020.592496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rodrigues JA, Cha W, Mosci RE, Mukherjee S, Newton DW, Lephart P, Salimnia H, Khalife W, Rudrik JT, Manning SD. 2021. Epidemiologic associations vary between tetracycline and fluoroquinolone resistant Campylobacter jejuni infections. Front Public Health 9:672473. doi: 10.3389/fpubh.2021.672473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kayman T, Abay S, Aydin F, Şahin O. 2019. Antibiotic resistance of Campylobacter jejuni isolates recovered from humans with diarrhoea in Turkey. J Med Microbiol 68:136–142. doi: 10.1099/jmm.0.000890. [DOI] [PubMed] [Google Scholar]
- 13.Collado L, Munoz N, Porte L, Ochoa S, Varela C, Munoz I. 2018. Genetic diversity and clonal characteristics of ciprofloxacin-resistant Campylobacter jejuni isolated from Chilean patients with gastroenteritis. Infect Genet Evol 58:290–293. doi: 10.1016/j.meegid.2017.12.026. [DOI] [PubMed] [Google Scholar]
- 14.Schiaffino F, Colston JM, Paredes-Olortegui M, Francois R, Pisanic N, Burga R, Penataro-Yori P, Kosek MN. 2019. Antibiotic resistance of Campylobacter Species in a pediatric cohort study. Antimicrob Agents Chemother 63:e01911-18. doi: 10.1128/AAC.01911-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Khademi F, Sahebkar A. 2020. Prevalence of fluoroquinolone-resistant Campylobacter species in Iran: a systematic review and meta-analysis. Int J Microbiol 2020:8868197. doi: 10.1155/2020/8868197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Goyal D, Watkins LKF, Montgomery MP, Jones SMB, Caidi H, Friedman CR. 2021. Antimicrobial susceptibility testing and successful treatment of hospitalised patients with extensively drug-resistant Campylobacter jejuni infections linked to a pet store puppy outbreak. J Glob Antimicrob Resist 26:84–90. doi: 10.1016/j.jgar.2021.04.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Iovine NM. 2013. Resistance mechanisms in Campylobacter jejuni. Virulence 4:230–240. doi: 10.4161/viru.23753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kotb DN, Mahdy WK, Mahmoud MS, Khairy RMM. 2019. Impact of co-existence of PMQR genes and QRDR mutations on fluoroquinolones resistance in Enterobacteriaceae strains isolated from community and hospital acquired UTIs. BMC Infect Dis 19:979. doi: 10.1186/s12879-019-4606-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chang MX, Zhang JF, Sun YH, Li RS, Lin XL, Yang L, Webber MA, Jiang HX. 2021. Contribution of different mechanisms to ciprofloxacin resistance in Salmonella spp. Front Microbiol 12:663731. doi: 10.3389/fmicb.2021.663731. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rincon G, Radice M, Giovanakis M, Di Conza JA, Gutkind G. 2014. First report of plasmid-mediated fluoroquinolone efflux pump QepA in Escherichia coli clinical isolate ST68, in South America. Diagn Microbiol Infect Dis 79:70–72. doi: 10.1016/j.diagmicrobio.2014.01.007. [DOI] [PubMed] [Google Scholar]
- 21.Baucheron S, Imberechts H, Chaslus-Dancla E, Cloeckaert A. 2002. The AcrB multidrug transporter plays a major role in high-level fluoroquinolone resistance in Salmonella enterica serovar typhimurium phage type DT204. Microb Drug Resist 8:281–289. doi: 10.1089/10766290260469543. [DOI] [PubMed] [Google Scholar]
- 22.Jesse TW, Englen MD, Pittenger-Alley LG, Fedorka-Cray PJ. 2006. Two distinct mutations in gyrA lead to ciprofloxacin and nalidixic acid resistance in Campylobacter coli and Campylobacter jejuni isolated from chickens and beef cattle. J Appl Microbiol 100:682–688. doi: 10.1111/j.1365-2672.2005.02796.x. [DOI] [PubMed] [Google Scholar]
- 23.Lopes BS, Strachan NJC, Ramjee M, Thomson A, MacRae M, Shaw S, Forbes KJ. 2019. Nationwide stepwise emergence and evolution of multidrug-resistant Campylobacter jejuni sequence type 5136, United Kingdom. Emerg Infect Dis 25:1320–1329. doi: 10.3201/eid2507.181572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Heyries KA, Tropini C, Vaninsberghe M, Doolin C, Petriv OI, Singhal A, Leung K, Hughesman CB, Hansen CL. 2011. Megapixel digital PCR. Nat Methods 8:649–651. doi: 10.1038/nmeth.1640. [DOI] [PubMed] [Google Scholar]
- 25.Dasti JI, Tareen AM, Lugert R, Zautner AE, Gross U. 2010. Campylobacter jejuni: a brief overview on pathogenicity-associated factors and disease-mediating mechanisms. Int J Med Microbiol 300:205–211. doi: 10.1016/j.ijmm.2009.07.002. [DOI] [PubMed] [Google Scholar]
- 26.Zhang Q, Lin J, Pereira S. 2003. Fluoroquinolone-resistant Campylobacter in animal reservoirs: dynamics of development, resistance mechanisms and ecological fitness. Anim Health Res Rev 4:63–71. doi: 10.1079/ahr200356. [DOI] [PubMed] [Google Scholar]
- 27.Inglis GD, Taboada EN, Boras VF. 2021. Rates of fluoroquinolone resistance in domestically acquired Campylobacter jejuni are increasing in people living within a model study location in Canada. Can J Microbiol 67:37–52. doi: 10.1139/cjm-2020-0146. [DOI] [PubMed] [Google Scholar]
- 28.Vinue L, Sater MRA, Herriott IC, Huntley MH, Wang M, Jacoby GA, Hooper DC. 2020. Plasmids and genes contributing to high-level quinolone resistance in Escherichia coli. Int J Antimicrob Agents 56:105987. doi: 10.1016/j.ijantimicag.2020.105987. [DOI] [PubMed] [Google Scholar]
- 29.Wang Y, Huang WM, Taylor DE. 1993. Cloning and nucleotide sequence of the Campylobacter jejuni gyrA gene and characterization of quinolone resistance mutations. Antimicrob Agents Chemother 37:457–463. doi: 10.1128/AAC.37.3.457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Liao YS, Chen BH, Teng RH, Wang YW, Chang JH, Liang SY, Tsao CS, Hong YP, Sung HY, Chiou CS. 2022. Antimicrobial resistance in Campylobacter coli and Campylobacter jejuni from human campylobacteriosis in Taiwan, 2016–2019. Antimicrob Agents Chemother 66:e01736-21. doi: 10.1128/AAC.01736-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Payot S, Cloeckaert A, Chaslus-Dancla E. 2002. Selection and characterization of fluoroquinolone-resistant mutants of Campylobacter jejuni using enrofloxacin. Microb Drug Resist 8:335–343. doi: 10.1089/10766290260469606. [DOI] [PubMed] [Google Scholar]
- 32.Hennebique A, Bidart M, Jarraud S, Beraud L, Schwebel C, Maurin M, Boisset S. 2017. Digital PCR for detection and quantification of fluoroquinolone resistance in Legionella pneumophila. Antimicrob Agents Chemother 61:e00628-17. doi: 10.1128/AAC.00628-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Pholwat S, Stroup S, Foongladda S, Houpt E. 2013. Digital PCR to detect and quantify heteroresistance in drug resistant Mycobacterium tuberculosis. PLoS One 8:e57238. doi: 10.1371/journal.pone.0057238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Anadon A, Martinez-Larranaga MR, Iturbe J, Martinez MA, Diaz MJ, Frejo MT, Martinez M. 2001. Pharmacokinetics and residues of ciprofloxacin and its metabolites in broiler chickens. Res Vet Sci 71:101–109. doi: 10.1053/rvsc.2001.0494. [DOI] [PubMed] [Google Scholar]
- 35.Zhang T, Dong J, Cheng Y, Lu Q, Luo Q, Wen G, Liu G, Shao H. 2017. Genotypic diversity, antimicrobial resistance and biofilm-forming abilities of Campylobacter isolated from chicken in Central China. Gut Pathog 9:62. doi: 10.1186/s13099-017-0209-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Zhang A, Song L, Liang H, Gu Y, Zhang C, Liu X, Zhang J, Zhang M. 2016. Molecular subtyping and erythromycin resistance of Campylobacter in China. J Appl Microbiol 121:287–293. doi: 10.1111/jam.13135. [DOI] [PubMed] [Google Scholar]
- 37.Tang JY, Nishibuchi M, Nakaguchi Y, Ghazali FM, Saleha AA, Son R. 2011. Transfer of Campylobacter jejuni from raw to cooked chicken via wood and plastic cutting boards. Lett Appl Microbiol 52:581–588. doi: 10.1111/j.1472-765X.2011.03039.x. [DOI] [PubMed] [Google Scholar]
- 38.Ge B, McDermott PF, White DG, Meng J. 2005. Role of efflux pumps and topoisomerase mutations in fluoroquinolone resistance in Campylobacter jejuni and Campylobacter coli. Antimicrob Agents Chemother 49:3347–3354. doi: 10.1128/AAC.49.8.3347-3354.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]