Molecular Subtyping and Antimicrobial Susceptibilities of Campylobacter coli Isolates from Diarrheal Patients and Food-Producing Animals in China

Abstract

The aim of this study was to determine the molecular subtyping and antimicrobial susceptibility characteristics of Campylobacter coli isolates from different sources in China. One hundred thirteen C. coli isolates were subtyped by pulsed-field gel electrophoresis (PFGE), multilocus sequence typing (MLST), and porA and flaA short variable region (SVR) nucleotide sequences. Cluster analysis was performed based on the PFGE and sequence types (ST). Eighty-four PFGE patterns (SmaI) were observed in 113 isolates. Fifty-four STs (28 novel) and three clonal complexes (CC), 86% of which were clustered to CC828, were observed, as well as 52 porA and 37 flaA-SVR sequence alleles. MLST, porA, and flaA-SVR analysis demonstrated that many isolates from diarrheal patients shared identical genotypes with chicken isolates. Minimum inhibitory concentration values of 10 antibiotics were analyzed for 109 isolates isolated in 2011 using the E-test method. The most frequently observed resistance agents were nalidixic acid (100%) and ciprofloxacin (100%), followed by levofloxacin (99%), tetracycline (94%), metronidazole (93%), erythromycin (61%), streptomycin (72%), gentamicin (59%), ampicillin (50%), and chloramphenicol (29%). Multidrug resistance was detected in 108 of 109 C. coli isolates (99%).

Introduction

C ampylobacter is the leading cause of foodborne bacterial disease worldwide. C. jejuni and C. coli are the major species responsible for campylobacteriosis, causing an estimated 95% of human Campylobacter infections (Popovic-Uroic, 1989; Coker et al., 2002; Uzunovic-Kamberovic et al., 2013). Although most cases of human campylobacteriosis are attributed to C. jejuni, C. coli has also been isolated from diarrheal patients, animals, and retail meats (Bi et al., 2010; Wang et al., 2010). One report indicated that 10.6% of adult campylobacteriosis cases in Beijing were attributed to C. coli (Chen et al., 2011) and our recent pilot study with intensive species differentiation analysis indicated that C. coli accounted for 17%–33% of human campylobacteriosis in Shanghai (unpublished).

Bacterial subtyping schemes play an important role in epidemiological investigations to trace sources and routes of pathogen transmission by identifying the outbreak and differentiating the sporadic infection (Schweitzer et al., 2011; O'Mahony et al., 2011). Multilocus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE) have been employed successfully for the longitudinal epidemiology study of Campylobacter spp. from different sources and outbreak investigations (Dingle et al., 2001; Hume et al., 2002; Dingle et al., 2005; Michaud et al., 2005; Thakur et al., 2009).

In recent years, a number of studies have indicated an increase in emerging drug-resistance characteristics among Campylobacter spp. (Kos et al., 2006; Gibreel et al., 2006; Alfredson et al., 2007). Campylobacter can be transferred from food-producing animals to humans, and many reports have demonstrated that the drug-resistant strains involved in human disease are predominantly spread through the food chain (Smith et al., 1999; Piddock et al., 2000; Engberg et al., 2004; Smith et al., 2010; Schweitzer et al., 2011). Determination of the drug resistance status of Campylobacter strains from different sources is crucial for the control and prevention of human infection, particularly for the cases in which therapy is recommended.

In order to determine the genotype and drug resistance profiles of C. coli from China, antibiotic susceptibility tests and molecular subtyping analysis of isolates from different sources were carried out in this study.

Materials and Methods

Isolates collection

Molecular typing was performed on 113 isolates that were isolated from 2 regions (43 isolates from Beijing and 70 isolates from Shanghai) in China from 2010 to 2011. Among these, 48 were from the stool samples of diarrheal patients collected during April to October in each year, 32 were isolated from retail foods (27 from fresh chicken meat and 5 from raw pork), and 33 were from fecal samples obtained from food-producing animals during the same time and in the same region (18 from chickens and 15 from swine). Antimicrobial susceptibility tests were performed on 109 isolates that were isolated in 2011 in Shanghai. Among these, 61 isolates were from the stool samples of diarrheal patients collected from 11 hospitals and 48 isolates were obtained from the fecal samples of food-producing animals in the same region (27 from swine stools collected from 3 abattoirs and 21 from chicken stools collected from 5 poultry markets). In 51 of the 109 isolates, subtyping analysis was also performed. All of the tested isolates were randomly selected from the collection of a pilot study on Campylobacter infection and transmission supported by the Major State Basic Research Development Program in China.

Bacterial and species identification was performed by Gram staining, hippurate hydrolysis tests, and C. coli-specific polymerase chain reaction (PCR), according to previous reports (Klena et al., 2004). The bacteria were cultured on Skirrow's medium (Columbia Agar Base supplemented with 5% sheep blood) at 37°C for 48 h in a microaerobic atmosphere (5% O₂, 10% CO₂, and 85% N₂). The genomic DNA of each isolate was extracted using a DNeasy tissue kit (Qiagen, Valencia, CA) according to the manufacturer's protocol.

PFGE

PFGE was performed according to the standard PulseNet PFGE protocol for C. coli (http://www.cdc.gov/pulsenet/protocols/campy_protocol.pdf) with the SmaI restriction enzyme. PFGE cluster analysis was performed using BioNumerics software (Version 4.1, Applied Maths, Austin, TX), and Dice's coefficient was calculated according to the unweighted-pair group method with arithmetic mean. Isolates were assigned to the same PFGE macrorestriction profile when they clustered with >99% similarity (1.2% optimization and 1.5% position tolerance).

MLST

MLST was performed by sequencing seven housekeeping gene loci (aspA, glnA, gltA, glyA, pgm, tkt, and uncA) according to previously described methods (Dingle et al., 2005). The allele of each gene was obtained from the comparison of the seven alleles in the MLST database (http://pubmlst.org/campylobacter/). For each isolate, the allele combination at the seven loci defined an allelic profile or sequence type (ST). The novel alleles and STs were obtained by submission of the DNA trace files to the database.

FlaA–short variable region (SVR) and PorA sequencing

PCR amplification of flaA-SVR was performed using modified forward and reverse primers flaA-SVR-F (5′-TCCGCAGCAGATGATGCTTCA-3′) and flaA-SVR-R (5′-GCTCTAAC[A,T]CCTGTTTTATC-3′). Nucleotide sequence extension reactions were carried out on both DNA strands using the flaA-SVR primer pairs. The flaA-SVR allele numbers were assigned by comparing the newly generated sequences against existing ones (http://pubmlst.org/perl/bigsdb/bigsdb.pl?db=pubmlst_campylobacter_seqdef&page=sequenceQuery) for fla_nucleotide loci. Alleles of porA were obtained according to the methods described at http://pubmlst.org/campylobacter/info/porA_method.shtml. The novel alleles were assigned by the curator following submission of the DNA trace files to the database.

Phylogenetic analysis based on MLST

The most likely ST founder of the 113 isolates was determined using eBURSTv3 software (Feil et al., 2004). The stringent group definition was based on isolates with six matches to the seven housekeeping gene alleles placed in the same complex and a bootstrapping method (resampling 1000 times) was conducted to provide statistical confidence in the assignment of the ST as the putative founder of the complex. Minimum spanning tree analysis was used to infer the relationships of the genetic population among the isolates from different sources and was constructed using BioNumerics based on the STs and the sequences of porA and flaA-SVR of the isolates.

Antimicrobial susceptibility testing

The minimum inhibitory concentrations (MICs) of 10 antibiotics (erythromycin, streptomycin, gentamicin, ampicillin, chloramphenicol, tetracycline, metronidazole, nalidixic acid, ciprofloxacin, and levofloxacin) against 109 C. coli isolates were determined using an E-test (AB Biodisk) method, as previously described (Zhang et al., 2010). The breakpoint values for erythromycin, ciprofloxacin, and tetracycline used in this study were set as ≥32 μg/mL, 4 μg/mL and 16 μg/mL, respectively, based on Clinical and Laboratory Standards Institute (CLSI, 2006). The breakpoint value for ampicillin, chloramphenicol, gentamicin, and streptomycin used in this study was that the MIC ≥16 μg/mL and the resistance breakpoints for nalidixic acid, levofloxacin, and metronidazole were set as MICs ≥32 μg/mL, 4 μg/mL, and 1 μg/mL, respectively, according to previous publications (Luangtongkum et al., 2006; Miflin et al., 2007; Marinou et al., 2012). C. jejuni ATCC 33560, Staphylococcus aureus ATCC 29213, Escherichia coli ATCC 25922 and the erythromycin-resistant C. coli strain WHO C-9.1 (MIC 32 μg/mL) from the External Quality Assurance System of the WHO Global Foodborne Infections Network were established as erythromycin-resistant controls.

Statistical analysis for antimicrobial resistance rates

The frequencies of antimicrobial resistance patterns from different sources were compared using a χ²-test and Fisher's exact two-tailed test with SPSS19.0 software, where applicable. Values of p<0.05 were considered statistically significant.

Results

PFGE typing

In total, 84 PFGE patterns were detected among the 113 examined isolates. Five closely related groups, sharing >75% similarity in banding patterns, were found for 80 of the tested isolates (Fig. 1). Isolates from different sources were distributed across various clusters. The multidrug-resistant isolates, which are marked with black circles, presented in different clusters (Fig. 1).

FIG. 1.

Dendrogram of pulsed-field gel electrophoresis (PFGE) cluster analysis of 113 Campylobacter coli isolates from different sources. PFGE was performed according to the standard PulseNet PFGE protocol for C. coli with the restriction enzyme SmaI. Cluster analysis was performed using BioNumerics software (Version 4.1, Applied Maths, and Austin, TX). Five clusters were found with ≥75% similarity in banding patterns. Multidrug-resistant isolates (marked with a black circle on the left side of the isolate ID) were distributed in different clusters.

Sequence typing and genetic relation

The background, STs, porA, and flaA-SVR are summarized in Table 1. A total of 54 STs, including 28 novel types, were observed from the 113 isolates. Three clonal complexes (CC)—CC21 (1 isolate), CC1150 (2 isolate), and CC828 (97 isolate)—were generated from 100 isolates; 12 STs from 13 isolates could not be assigned to any of the defined CCs. Twenty-six STs were observed from 48 human isolates. The most frequently observed ST among the human isolates was 1625 (19%, 9/48), followed by 1145 (13%, 6/48). The most frequently observed STs among the chicken fecal or chicken meat isolates were 829 (11%, 5/45) and 860 (11%, 5/45), followed by 1625 (9%, 4/45) and 855 (9%, 4/45). The most frequently observed ST among the swine or pork isolates was 1464 (20%, 4/20). Thirty-four STs were observed in 51 multidrug-resistant isolates. Two CCs—CC1150 (1 isolate) and CC828 (43 isolate) —were generated from 44 isolates; 7 STs from 7 isolates could not be assigned to any of the defined CCs. No specific STs or CCs were related to the drug-resistant profile.

Table 1.

Background and Molecular Subtyping of 113 Campylobacter coli Strains

	Profile
Strains	ST	CC	porA	flaA-SVR	Source
CCF209	825	ST-828	915	17	Fresh chicken meat
CCF210	825	ST-828	915	17	Fresh chicken meat
SH-CCD11C143	825	ST-828	1711	16	Diarrheal patient stool
SH-CCD11C615	825	ST-828	1744	16	Diarrheal patient stool
SH-CCD74	825	ST-828	1456	17	Diarrheal patient stool
SH-CCD11C682	828	ST-828	1428	1326	Diarrheal patient stool
CCSFX20A	829	ST-828	915	947	Swine stool
SH-CCD11C673	829	ST-828	1745	66	Diarrheal patient stool
SH-CCH20	829	ST-828	687	17	Chicken stool
SH-CCH31	829	ST-828	687	17	Chicken stool
SH-CCH32	829	ST-828	687	17	Chicken stool
SH-CCH38	829	ST-828	687	17	Chicken stool
SH-CCH45	829	ST-828	687	17	Chicken stool
SH-CCD11C508	830	ST-828	783	1614	Diarrheal patient stool
CCF193	832	ST-828	1429	224	Fresh chicken meat
CCF200	832	ST-828	1429	224	Fresh chicken meat
CCF220	832	ST-828	1429	224	Fresh chicken meat
SH-CCH17	855	ST-828	561	13	Chicken stool
SH-CCH18	855	ST-828	561	13	Chicken stool
SH-CCH19	855	ST-828	561	13	Chicken stool
SH-CCH36	855	ST-828	561	13	Chicken stool
CCF1341-27	860	ST-828	1432	548	Fresh chicken meat
CCF1341-30	860	ST-828	1432	548	Fresh chicken meat
CCF196	860	ST-828	92	239	Fresh chicken meat
CCF225	860	ST-828	915	17	Fresh chicken meat
SH-CCD11C566	860	ST-828	1717	17	Diarrheal patient stool
SH-CCD11C567	860	ST-828	1717	17	Diarrheal patient stool
SH-CCH28	860	ST-828	1371	769	Chicken stool
SH-CCD11C073	872	ST-828	1774	30	Diarrheal patient stool
SH-CCD43	872	ST-828	1428	494	Diarrheal patient stool
BJ-CCD30	1055	ST-828	1427	860	Diarrheal patient stool
CCF195	1055	ST-828	1427	17	Fresh chicken meat
SH-CCD11C197	1055	ST-828	1713	17	Diarrheal patient stool
SH-CCD6	1055	ST-828	783	820	Diarrheal patient stool
CCSFX22A	1104	ST-828	434	267	Swine stool
SH-CCS11C110	1107	ST-828	293	13	Swine stool
SH-CCS11C305	1107	ST-828	293	13	Swine stool
BJ-CCD09002	1145	ST-828	1428	54	Diarrheal patient stool
SH-CCD11C287	1145	ST-828	1714	494	Diarrheal patient stool
SH-CCD11C365	1145	ST-828	1714	494	Diarrheal patient stool
SH-CCD11C416	1145	ST-828	1714	494	Diarrheal patient stool
SH-CCD11C490	1145	ST-828	1714	494	Diarrheal patient stool
SH-CCD11C518	1145	ST-828	1714	494	Diarrheal patient stool
SH-CCS11C534	1464	ST-828	1716	920	Swine stool
SH-CCS11C535	1464	ST-828	1716	920	Swine stool
SH-CCS11C536	1464	ST-828	1716	920	Swine stool
SH-CCS11C544	1464	ST-828	1716	920	Swine stool
CCSFX23A	1465	ST-828	1719	1613	Swine stool
SH-CCSF11C156	1563	ST-828	1718	1613	Raw pork
SH-CCH41	1586	ST-828	106	67	Chicken stool
SH-CCH42	1586	ST-828	106	820	Chicken stool
CCF263	1593	ST-828	910	30	Fresh chicken meat
SH-CCD11C098	1593	ST-828	639	17	Diarrheal patient stool
BJ-CCD09032	1625	ST-828	687	16	Diarrheal patient stool
BJ-CCD09148	1625	ST-828	1459	16	Diarrheal patient stool
BJ-CCD48	1625	ST-828	1427	16	Diarrheal patient stool
BJ-CCD62	1625	ST-828	1436	16	Diarrheal patient stool
BJ-CCD76	1625	ST-828	1427	1578	Diarrheal patient stool
CCF191	1625	ST-828	687	16	Fresh chicken meat
CCF218	1625	ST-828	1439	66	Fresh chicken meat
CCF250	1625	ST-828	1427	66	Fresh chicken meat
CCF251	1625	ST-828	1427	66	Fresh chicken meat
SH-CCD11C419	1625	ST-828	1715	16	Diarrheal patient stool
SH-CCD11C473	1625	ST-828	687	16	Diarrheal patient stool
SH-CCD11C493	1625	ST-828	687	16	Diarrheal patient stool
SH-CCD62	1625	ST-828	1438	16	Diarrheal patient stool
SH-CCH22	1770	ST-828	1428	770	Chicken stool
SH-CCH23	1770	ST-828	1428	770	Chicken stool
SH-CCH25	2820	ST-828	1427	13	Chicken stool
SH-CCH27	2820	ST-828	1427	13	Chicken stool
SH-CCH26	3985	ST-828	1427	13	Chicken stool
SH-CCS11C545	4938	ST-828	415	569	Swine stool
BJ-CCD09085	5506	ST-828	915	239	Diarrheal patient stool
CCF1341-28	5507	ST-828	1428	16	Fresh chicken meat
CCF1341-29	5507	ST-828	1428	16	Fresh chicken meat
SH-CCD11C271	5507	ST-828	1428	16	Diarrheal patient stool
SH-CCD11C272	5507	ST-828	1428	16	Diarrheal patient stool
CCF208	5510	ST-828	1429	17	Fresh chicken meat
CCF211	5511	ST-828	1427	16	Fresh chicken meat
CCF234	5512	ST-828	1430	776	Fresh chicken meat
CCF264	5515	ST-828	1435	760	Fresh chicken meat
CCF265	5515	ST-828	1435	760	Fresh chicken meat
BJ-CCH2	5516	ST-828	1439	1577	Chicken stool
BJ-CCD09029	5518	ST-828	1457	54	Diarrheal patient stool
BJ-CCD56	5519	ST-828	687	16	Diarrheal patient stool
SH-CCD11C145	6184	ST-828	1371	16	Diarrheal patient stool
SH-CCD11C280	6184	ST-828	1723	30	Diarrheal patient stool
SH-CCS11C531	6315	ST-828	228	343	Swine stool
SH-CCS11C538	6316	ST-828	412	1615	Swine stool
SH-CCS11C543	6316	ST-828	412	1516	Swine stool
SH-CCD11C123	6317	ST-828	561	13	Diarrheal patient stool
SH-CCD11C027	6318	ST-828	910	587	Diarrheal patient stool
SH-CCD11C128	6319	ST-828	1710	17	Diarrheal patient stool
SH-CCS11C542	6320	ST-828	1716	920	Swine stool
SH-CCBF11C157	6321	ST-828	1720	1636	Raw pork
SH-CCSF11C158	6321	ST-828	1720	1636	Raw pork
SH-CCD11C142	6323	ST-828	1722	16	Diarrheal patient stool
SH-CCD11C195	1121	ST-1150	1712	1637	Diarrheal patient stool
SH-CCD13	1121	ST-1150	1440	1580	Diarrheal patient stool
CCSF194	5692	ST-21	92	409	Raw pork
SH-CCD11C504	1450	None	771	569	Diarrheal patient stool
CCSFX9	3225	None	1458	16	Swine stool
CCSF197	3225	None	1458	16	Raw pork
CCF227	3989	None	763	239	Fresh chicken meat
CCF188	5508	None	1455	16	Fresh chicken meat
CCF192	5509	None	1434	1585	Fresh chicken meat
CCF261	5514	None	1427	844	Fresh chicken meat
SH-CCD80	5696	None	177	769	Diarrheal patient stool
SH-CCD11C353	6185	None	915	239	Diarrheal patient stool
SH-CCD11C600	6186	None	687	16	Diarrheal patient stool
SH-CCF11C627	6322	None	1721	261	Fresh chicken meat
SH-CCD11C004	6380	None	915	239	Diarrheal patient stool
SH-CCD11C226	6896	None	1775	66	Diarrheal patient stool

Fifty-two porA alleles were detected from 113 isolates. The four most frequently presented alleles were 1427 and 687, followed by 1428 and 915, which are mostly from chicken or human isolates. Allele 1716 was generated from five swine isolates. Thirty-seven flaA-SVR sequence variants were found and the most common allele, 16, was generated from 23 isolates, 16 of which were from diarrheal patients. The next most frequently observed allele was 17, which is predominantly from diarrheal patients or chicken isolates.

eBURST analysis indicated that the most likely founder ST of the 113 isolates in this study was ST860. ST825, ST829, ST1055, ST1625, and ST828 were the subfounders (Fig. 2). The genetic relatedness based on the STs and the alleles of porA and flaA-SVR from the 113 isolates is presented in Figure 3. The isolates from diarrhea patients showed identical or similar genotypes (with more than seven genes identical) with the isolates from chicken.

FIG. 2.

eBURST analysis for 113 Campylobacter coli isolates. The sequence types (ST) founders of 113 isolates were determined using eBURSTv3 software. The stringent group definition was based on isolates with six matches to the seven housekeeping gene alleles placed in the same complex, and the bootstrapping method (resampling 1000 times) was conducted to provide statistical confidence in the assignment of the ST as the putative founder of the complex. The size of the circle is proportional to the number of isolates with the indicated ST. The most likely founder ST of the 113 isolates in this study is ST860 (blue). ST825 (yellow), ST829 (yellow), ST1055 (yellow), ST1625 (yellow) and ST828 (yellow) are the subfounders.

FIG. 3.

The genetic relatedness of 113 Campylobacter coli isolates from different sources. The minimum spanning tree was constructed using the alleles of porA, flaA-SVR, and sequence types (ST) from 113 Campylobacter coli isolates from different sources (chicken, red; diarrheal patient, green; swine, purple). The size of the circle is proportional to the number of isolates with the indicated ST. The numbers along the lines indicate the difference of the alleles between different isolates (active line: one or two alleles different; dotted line: more than two alleles different). The isolates from diarrheal patients show a more similar genotype to the isolates from chicken.

Results of antimicrobial susceptibility tests

The MIC₅₀ and MIC₉₀ values for each antibiotic are shown in Table 2. The antibiotic-resistance ratio and the statistical analysis for each drug on isolates from different sources are shown in Table 3. Nalidixic acid and ciprofloxacin resistance were found in 100% of C. coli isolates recovered from all sources. Otherwise, the most frequently observed resistance was to levofloxacin (99%, 108/109), tetracycline (94%, 103/109), metronidazole (93%, 101/109), erythromycin (61%, 67/109), streptomycin (72%, 78/109), gentamicin (59%, 64/109), ampicillin (50%, 54/109), and chloramphenicol (29%, 32/109). Multidrug resistance (resistance to 4 or more categories of antibiotics) was observed in 99% (108/109) of isolates.

Table 2.

Fifty Percent Minimum Inhibitory Concentration (MIC) and Ninety Percent Minimum Inhibitory Concentration of 10 Antibiotics for 109 Campylobacter coli Isolates

Antibiotic	MIC range (μg/mL)	MIC₅₀ (μg/mL)	MIC₉₀ (μg/mL)
Erythromycin	0.19−>256	256	>256
Streptomycin	0.38−>256	64	>256
Gentamicin	0.05−>256	>256	>256
Ampicillin	0.125−>256	12	>256
Chloramphenicol	0.5−>256	8	>256
Tetracycline	0.1−>256	>256	>256
Nalidixic acid	48−>256	>256	>256
Ciprofloxacin	16−>32	>32	>32
Levofloxacin	3−>32	>32	>32
Metronidazole	0.02−>256	32	128

Table 3.

Antibiotic-Resistant Ratio and Statistical Analysis for 109 Campylobacter coli Isolates from Different Sources

	Resistance profile, n (%)			Two-tailedp-value
	Diarrheal patient	Chicken	Swine	Human vs. Chicken	Human vs. Swine	Chicken vs. Swine
Strain number	61	21	27
Erythromycin	40 (66%)	11 (52%)	16 (60%)	0.1457	0.8131	0.2878
Streptomycin	37 (61%)	17 (81%)	24 (89%)	0.0907	0.0081	0.683
Gentamicin	41 (67%)	11 (52%)	12 (44%)	0.2236	0.0442	0.5851
Ampicillin	35 (57%)	12 (57%)	7 (26%)	0.9851	0.0064	0.0282
Chloramphenicol	15 (25%)	13 (62%)	4 (15%)	0.0019	0.304	0.0007
Tetracycline	56 (92%)	21 (100%)	26 (96%)	0.3214	0.662	1
Nalidixic acid	61 (100%)	21 (100%)	27 (100%)	1	1	1
Ciprofloxacin	61 (100%)	21 (100%)	27 (100%)	1	1	1
Levofloxacin	61 (100%)	20 (95%)	27 (100%)	1	1	1
Metronidazole	61 (100%)	20 (95%)	20 (74%)	0.2561	<0.0001	0.0642

Examining the isolates from diarrheal patients and the isolates from chickens, the resistance ratios for nine of the antibiotics were found to be similar (p>0.05). Only the resistance ratio of chloramphenicol was noted to be significantly higher in chicken isolates compared to the isolates from diarrheal patients (p=0.0019). However, there are significant differences between the resistance ratios of gentamicin, streptomycin, ampicillin, and metronidazole between the isolates from diarrheal patients and isolates from swine (Table 3). The resistance frequency of swine isolates to streptomycin was higher than that of human isolates (p=0.0081). In contrast, the resistance frequencies of swine isolates to gentamicin, ampicillin, and metronidazole were less than those of the isolates from diarrheal patients. The isolates from diarrheal patients shared similar resistance profiles to the isolates from chickens rather than to those from swine. There was no significant difference between the frequencies of resistance to nalidixic acid, ciprofloxacin, levofloxacin, and tetracycline among the isolates from different sources (p>0.05).

Discussion

C. coli bacteria colonize in the intestines of food-producing animals and can contaminate the associated foods during processing. Consumption of contaminated food has been implicated as the major source of human campylobacteriosis (Mylius et al., 2007; Hermans et al., 2012). Studies from China and other countries have demonstrated that C. coli–associated human campylobacteriosis accounts for approximately 11%–40% of cases (Popovic-Uroic, 1989; Uzunovic-Kamberovic, 2001; Uzunovic, 2003). Molecular subtyping of isolates from the clinical cases and other sources is a potential way to provide information about the source attributes, and source attribution studies can subsequently provide valuable information regarding infection control and disease prevention. The present study found that the majority of the tested isolates (86%) belonged to one lineage (CC828), which infers a common ancestry. This result is consistent with a previous investigation, which indicated that 83.3% of Chinese C. coli isolates from chicken meat belonged to CC828 (Wang et al., 2010).

There have been reports showing that pigs are a common source of C. coli, and they have been previously considered to be the most probable source of human infections (Nielsen et al., 1997; Harvey et al., 1999); however, there are also studies that have demonstrated poultry to be the primary reservoir for both C. jejuni and C. coli bacteria (Siemer et al., 2005; Wingstrand et al., 2006; Zorman et al., 2006). In the present study, both MLST, porA, and flaA-SVR analysis have indicated that the majority of isolates collected from diarrheal patients share identical genotypes with chicken isolates. In addition, the universal C. coli prevalence in chickens also suggests that chickens may be a dominant source of human C. coli infections in China (Wang et al., 2010).

Campylobacter infections are usually self-limiting and do not need antibiotic treatment. However, for severe and prolonged cases, antimicrobial treatment is required. Currently, fluoroquinolones and macrolides are commonly used for the treatment of human campylobacteriosis in China. However, our results indicate that all of the tested isolates are almost 100% resistant to fluoroquinolones and 66% of the human isolates are resistant to erythromycin. The clinical treatment for C. coli-associated campylobacteriosis should therefore be carefully reconsidered. Previous reports have demonstrated that poultry is a major source of fluoroquinolone-resistant Campylobacter-resulting human infections (Iovine et al., 2004a, 2004b; Smith et al., 2010). The judicious use of fluoroquinolone in food-producing animals might be one effective way in which the spread of drug-resistant Campylobacter be reduced in China. It should be noted that the resistance ratios of isolates from diarrheal patients was much higher than has been previously reported (Uzunovic et al., 2013).

The present study indicated the resistant profile of the tested C. coli isolates was quite different from the strains isolated in the United States, which were commonly resistant to tetracycline (66.2%) and less resistant to fluoroquinolone (1.5%) (Thakur et al., 2010). The resistant frequencies were higher than those of isolates from other counties (Lutgen et al., 2009; Dohne et al., 2012; Obeng et al., 2012). High frequencies of multidrug resistance were noted in C. coli isolates from different sources (99%). No specific cluster was found among the 51 multidrug-resistant isolates according to both the PFGE and MLST results. These isolates showed variant patterns, which indicates that they are genetically diverse. The increase in the number of highly resistant C. coli will not only bring complications with regard to disease treatment, but these bacteria will also form a resistant reservoir in the intestines of animals. A resistance integron was recently found in C. coli isolates isolated in China (Qin et al., 2012).

Conclusions

In summary, most C. coli isolates from diarrheal patients shared identical genotypes and antibiotic resistance patterns with isolates from chickens, and high drug resistance was found among isolates from different sources. The data obtained in this study will benefit the established strategy for Campylobacter monitoring and may also improve food safety in China and other parts of the world.

Footnotes

Acknowledgments

This work was supported by the Major State Basic Research Development Program (2013CB127204) and the General Program of National Natural Science Foundation of China (81271789).

Disclosure Statement

No competing financial interests exist.

References

Alfredson

, Korolik

. Antibiotic resistance and resistance mechanisms in Campylobacter jejuni and Campylobacter coli . FEMS Microbiol Lett, 2007; 277:123–132.

, Chen

, Zang

, Shi

. Detection and genotyping of Campylobacter jejuni and Campylobacter coli in food samples by Fla-DGGE. Modern Food Sci Technol, 2010; 26:1148–1152.

Chen

, Sun

, Zeng

, Yu

. Campylobacter enteritis in adult patients with acute diarrhea from 2005 to 2009 in Beijing, China. Chin Med J (Engl), 2011; 124:1508–1512.

[CLSI] Clinical and Laboratory Standards Institute. Methods for Antimicrobial Dilution and Disk Susceptibility Testing of Infrequently Isolated or Fastidious Bacteria. CLSI Document M45-A. Wayne, PA: CLSI, 2006.

Coker

, Isokpehi

, Thomas

, Amisu

, Obi

. Human campylobacteriosis in developing countries. Emerg Infect Dis, 2002; 8:237–244.

Dohne

, Merle

, Altrock

, Waldmann

, Verspohl

, Gruning

, Hamedy

, Kreienbrock

. Antibiotic susceptibility of Salmonella, Campylobacter coli, and Campylobacter jejuni isolated from Northern German fattening pigs. J Food Prot, 2012; 75:1839–1845.

Dingle

, Colles

, Falush

, Maiden

. Sequence typing and comparison of population biology of Campylobacter coli and Campylobacter jejuni . J Clin Microbiol, 2005; 43:340–347.

Dingle

, Colles

, Wareing

, Ure

, Fox

, Bolton

, Bootsma

, Willems

, Urwin

, Maiden

. Multilocus sequence typing system for Campylobacter jejuni . J Clin Microbiol, 2001; 39:14–23.

Engberg

, Neimann

, Nielsen

, Aerestrup

, Fussing

. Quinolone-resistant Campylobacter infections: Risk factors and clinical consequences. Emerg Infect Dis, 2004; 10:1056–1063.

10.

Feil

, Li

, Aanensen

, Hanage

, Spratt

. eBURST: Inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J Bacteriol, 2004; 186:1518–1530.

11.

Gibreel

, Taylor

. Macrolide resistance in Campylobacter jejuni and Campylobacter coli . J Antimicrob Chemother, 2006; 58:243–255.

12.

Harvey

, Young

, Ziprin

, Hume

, Genovese

, Anderson

, Droleskey

, Stanker

, Nisbet

. Prevalence of Campylobacter spp isolated from the intestinal tract of pigs raised in an integrated swine production system. J Am Vet Med Assoc, 1999; 215:1601–1604.

13.

Hermans

, Pasmans

, Messens

, Martel

, Van Immerseel

, Rasschaert

, Heyndrickx

, Van Deun

, Haesebrouck

. Poultry as a host for the zoonotic pathogen Campylobacter jejuni . Vector Borne Zoonotic Dis, 2012; 12:89–98.

14.

Hume

, Droleskey

, Sheffield

, Harvey

. Campylobacter coli pulsed field gel electrophoresis genotypic diversity among sows and piglets in a farrowing barn. Curr Microbiol, 2002; 45:128–132.

15.

Iovine

, Blaser

. Antibiotics in animal feed and spread of resistant Campylobacter from poultry to humans. Emerg Infect Dis, 2004a;10:1158–1159.

16.

Iovine

, Blaser

. Antimicrobial resistance in Campylobacter . Emerg Infect Dis, 2004b;10:1346.

17.

Klena

, Parker

, Knibb

, Ibbitt

, Devane

, Horn

, Miller

, Konkel

. Differentiation of Campylobacter coli, Campylobacter jejuni, Campylobacter lari, and Campylobacter upsaliensis by a multiplex PCR developed from the nucleotide sequence of the lipid A gene lpxA. J Clin Microbiol, 2004; 42:5549–5557.

18.

Kos

, Gibreel

, Keelan

, Taylor

. Species identification of erythromycin-resistant Campylobacter isolates and optimization of a duplex PCR for rapid detection. Res Microbiol, 2006; 157:503–507.

19.

Luangtongkum

, Morishita

, Ison

, Huang

, McDermott

, Zhang

. Effect of conventional and organic production practices on the prevalence and antimicrobial resistance of Campylobacter spp. in poultry. Appl Environ Microbiol, 2006; 72:3600–3607.

20.

Lutgen

, McEvoy

, Sherwood

, Logue

. Antimicrobial resistance profiling and molecular subtyping of Campylobacter spp. from processed turkey. BMC Microbiol, 2009; 9:203.

21.

Marinou

, Bersimis

, Ioannidis

, Nicolaou

, Mitroussia-Ziouva

, Legakis

, Chatzipanagiotou

. Identification and antimicrobial resistance of Campylobacter species isolated from animal sources. Front Microbiol, 2012;doi:10.3389/fmicb.2012.00058.

22.

Michaud

, Menard

, Arbeit

. Role of real-time molecular typing in the surveillance of Campylobacter enteritis and comparison of pulsed-field gel electrophoresis profiles from chicken and human isolates. J Clin Microbiol, 2005; 43:1105–1111.

23.

Miflin

, Templeton

, Blackall

. Antibiotic resistance in Campylobacter jejuni and Campylobacter coli isolated from poultry in the South-East Queensland region. J Antimicrob Chemother, 2007; 59:775–778.

24.

Mylius

, Nauta

, Havelaar

. Cross-contamination during food preparation: A mechanistic model applied to chicken-borne Campylobacter . Risk Anal, 2007; 27:803–813.

25.

Nielsen

, Engberg

, Madsen

. Distribution of serotypes of Campylobacter jejuni and C. coli from Danish patients, poultry, cattle and swine. FEMS Immunol Med Microbiol, 1997; 19:47–56.

26.

Obeng

, Rickard

, Sexton

, Pang

, Peng

, Barton

. Antimicrobial susceptibilities and resistance genes in Campylobacter strains isolated from poultry and pigs in Australia. J Appl Microbiol, 2012; 113:294–307.

27.

O'Mahony

, Buckley

, Bolton

, Whyte

, Fanning

. Molecular epidemiology of Campylobacter isolates from poultry production units in southern Ireland. PLoS ONE, 2011; 6:e28490.

28.

Piddock

, Ricci

, Stanley

, Jones

. Activity of antibiotics used in human medicine for Campylobacter jejuni isolated from farm animals and their environment in Lancashire, UK. J Antimicrob Chemother, 2000; 46:303–306.

29.

Popovic-Uroic

. Campylobacter jejuni and Campylobacter coli diarrhoea in rural and urban populations in Yugoslavia. Epidemiol Infect, 1989; 102:59–67.

30.

Qin

, Wang

, Zhang

, Chen

, Shen

, Deng

, Wu

, Shen

. Identification of a novel genomic island conferring resistance to multiple aminoglycoside antibiotics in Campylobacter coli . Antimicrob Agents Chemother, 2012; 56:5332–5339.

31.

Schweitzer

, Dan

, Kaszanyitzky

, Samu

, Tath

, Varga

, Damjanova

. Molecular epidemiology and antimicrobial susceptibility of Campylobacter jejuni and Campylobacter coli isolates of poultry, swine, and cattle origin collected from slaughterhouses in Hungary. J Food Prot, 2011; 74:905–911.

32.

Siemer

, Nielsen

, On

. Identification and molecular epidemiology of Campylobacter coli isolates from human gastroenteritis, food, and animal sources by amplified fragment length polymorphism analysis and Penner serotyping. Appl Environ Microbiol, 2005; 71:1953–1958.

33.

Smith

, Fratamico

. Fluoroquinolone resistance in Campylobacter . J Food Prot, 2010; 73:1141–1152.

34.

Smith

, Besser

, Hedberg

, Leano

, Bender

, Wicklund

, Johnson

, Moore

, Osterholm

. Quinolone-resistant Campylobacter jejuni infections in Minnesota, 1992–1998. Investigation Team. N Engl J Med, 1999; 340:1525–1532.

35.

Thakur

, White

, McDermott

, Zhao

, Kroft

, Gebreyes

, Abbott

, Cullen

, English

, Carter

, Harbottle

. Genotyping of Campylobacter coli isolated from humans and retail meats using multilocus sequence typing and pulsed-field gel electrophoresis. J Appl Microbiol, 2009; 106:1722–1733.

36.

Thakur

, Gebreyes

. Phenotypic and genotypic heterogeneity of Campylobacter coli within individual pigs at farm and slaughter in the US. Zoonoses Public Health, 2010; 57(Suppl 1):100–106.

37.

Uzunovic

, Smole Mozina

. Campylobacter infections in Zenica-Doboj Canton, Bosnia and Herzegovina—A story. Med Glas (Zenica), 2013; 10:1–11.

38.

Uzunovic-Kamberovic

. Changes in Campylobacter jejuni and Campylobacter coli carriage rates in the Zenica Region [correction of Zenica Canton] of Bosnia and Herzegovina in the pre- and postwar periods. J Clin Microbiol, 2001; 39:2036.

39.

Uzunovic

. Some epidemiologic features of Campylobacter jejuni/coli infections in Bosnia and Herzegovina after the war. Clin Microbiol Infect, 2003; 9:458–460.

40.

Uzunović-Kamberović

, Rijnders

, Stobberingh

, Ibrahimagić

, Kamberović

, Ille

. Molecular characterisation of methicillin-susceptible and methicillin-resistant Staphylococcus aureus in inpatients and outpatients in Bosnia and Herzegovina. Wien Med Wochenschr, 2013; 163:13–20.

41.

Wang

, Zhou

, Meng

. Multilocus sequence typing for molecular typing of Campylobacter coli from poultry meat. Chin J Animal Vet Sci, 2010; 41:234–239.

42.

Wingstrand

, Neimann

, Engberg

, Nielsen

, Gerner-Smidt

, Wegener

, Molbak

. Fresh chicken as main risk factor for campylobacteriosis, Denmark. Emerg Infect Dis, 2006; 12:280–285.

43.

Zhang

, Gu

, He

, Ran

, Xia

, Han

, Li

, Zhou

, Cui

, Zhang

. Molecular typing and antimicrobial susceptibility profiles of Campylobacter jejuni isolates from north China. J Med Microbiol, 2010; 59:1171–1177.

44.

Zorman

, Heyndrickx

, Uzunović-Kamberović

, Smole Mozina

. Genotyping of Campylobacter coli and C. jejuni from retail chicken meat and humans with campylobacteriosis in Slovenia and Bosnia and Herzegovina. Int J Food Microbiol, 2006; 110:24–33.