Background
Campylobacter is considered to be the leading cause of human bacterial gastroenteritis worldwide [
1], accounting for an estimated 500 million infections per year globally [
2]. In severe cases of
C. jejuni infection, individuals may develop post infection complications associated with Guillain Barré Syndrome [
3]. In North China, 36 cases of Guillain Barré Syndrome, resulted from
C. jejuni infection, were reported in 2007 [
4].
Campylobacter species, mainly including
C. jejuni and
C. coli, widely colonize in the intestinal tract of wild and domesticated animals and birds [
5‐
7]. Chicken is one of the most popular animal-based food sources worldwide, which is also the reservoir of
Campylobacter.
Campylobacter-contaminated chicken products are a major cause of human
Campylobacter infection [
8], which highlights its potential public health threat. Several epidemiologic studies on
Campylobacter have been carried out in some parts of China. From 2008 to 2014, Wang et al. reported that the positive rates of
C. jejuni and
C. coli were 18.1 and 19.0% respectively in five provinces of China [
9]. Zhang et al. analyzed the genetic diversity of the
C. jejuni isolates in Eastern China by multilocus sequence typing (MLST) and identified 94 sequence types (STs) belonging to 18 clonal complexes (CCs) [
10]. However, data on the prevalence and genetic diversity of
Campylobacter is still limited in China, especially central China, which is an important transportation junction.
Moreover,
Campylobacter isolates have raised great concerns due to a frequent emergence of resistance to fluoroquinolone, erythromycin, and other drugs [
11,
12], which limits treatment alternatives. Therefore, analysis of antimicrobial resistance of
Campylobacter in the poultry industry will contribute to managing cognate infections and mitigating the emergence of antimicrobial resistant strains. Recent years, the multidrug-resistant
Campylobacter have been frequently isolated, and a high antimicrobial resistance rate of
Campylobacter, especially to fluoroquinolone, has been reported in many areas [
13‐
15]. Multi-drug resistance of
Campylobacter is more severe in China where the resistance to fluoroquinolones was reported to be as high as 98% in some areas [
16,
17]. Although some of the mechanisms accounting for antimicrobial resistance in
Campylobacter have been revealed [
11,
18,
19], some possible factors may also attribute to the raise of antimicrobial resistance, such as the ability of biofilm formation.
Our previous study has shown that the
Campylobacter positive rate was 17.2%, with bacterial count varying from 3.6 to 360 most-probable-number (MPN)/g in the positive samples of chicken meats collected from markets in central China [
20]. Studying the prevalence of
Campylobacter in live chicken and their surroundings will help us further control these pathogens. In this study, we investigated the prevalence, antimicrobial resistance and genetic diversity of
Campylobacter strains isolated from chicken farms and markets in central China, which is one of the most important livestock and poultry circulation centers. We also tested the biofilm-forming ability of the
Campylobacter isolates and analyzed the potential correlation among biofilm formation, genotypes, and antimicrobial resistance.
Discussion
Poultry are recognized as a main reservoir of
Campylobacter. Consumption of poultry is considered to be an important cause of human infection with
Campylobacter, and leads to extensive spread antimicrobial resistance [
28]. In this study,
Campylobacter strains were isolated from 25.2% of the samples collected from chicken farms and markets, including 166
C. jejuni and 40
C. coli. According to several previous reports, the positive detection rate of
Campylobacter in poultry farms varies largely between different regions, ranging from 2 to 100%, and the prevalence of
Campylobacter is lower in Scandinavian countries than in other European countries, North America, and developing countries [
29]. China is the biggest developing country in which a diverse prevalence rate has also been reported in different parts of the country. For example, Huang et al. revealed that
C. jejuni was frequently detected in poultry, with an average isolation rate of up to 18.61% [
30]. Wang et al. showed that the positive rates of
C. jejuni and
C. coli were 18.1 and 19.0% respectively in chicken in five provinces of China [
9]. In Tianjin, the contamination rates of
C. jejuni and other
Campylobacter species were 13.7 and 5.7% respectively [
31]. In this study, our data showed that the positive rate of
C. jejuni was a bit higher than most of the other studies carried out in China. We also found that
Campylobacter existed in the soils and aerosols of chicken farms and markets, suggesting that the pathogens were widely spreading between host and surroundings. This situation makes it harder for us to control
Campylobacter infection. A prevalence and risk assessment of
C. jejuni in chicken in China suggested that key efforts should be made, especially in chick breeding and chicken preparation processes [
32].
In our study, MLST analysis showed a total of 72 different STs belonging to 15 CCs and some unassigned clonal complexes. The major clonal complexes included CC-464, CC-1150, CC-353, and CC-828, which were similar to our previous investigation on chicken meat in the same region [
21]. Most of these CCs (CC-464, CC-1150, CC-353 and CC-828) were also frequently identified in diarrhea patients worldwide [
33,
34]. In North China, the most frequently isolated clonal complexes were CC-21, CC-353, CC-354 and CC-443 [
31,
35], while the dominant clonal complexes of
C. coli were CC-828 and CC-1150 [
36]. In East China, the most common ST type of the
Campylobacter strains isolated from human and food was ST-353, while the dominant ST type from chicken and food was ST-354 [
10]. In Guangdong, a province in southern China, the dominant clonal complex was CC-828 [
37]. It seems that the dominant clonal complexes of
Campylobacter were discrepant in different regions. However, most of the CCs reported in these regions had been isolated in our study, which may be because central China, where all the samples were collected, is one of the most important livestock and poultry circulation centers in our country.
A total of 40 novel STs were identified in this study. Genetic relationship analysis showed that different sources of isolates have a crossed distribution in each clonal group and most of the novel STs only have a minor variation with a close phylogenetic relationship to known CCs. Selection forces, such as differences in temperature, structure and biochemical and immunological habitats, may accelerate the evolution to gain the ability to persist in different enteric environments and survive in different environments during transmission. Clade 1 and 2 contained lots of small genetic branches, which may be due to the adaptive evolution of isolates in these two clades occurred more frequently in our investigated regions.
Another more important selection pressure might be the usage of antibiotics, which could cause heritable genetic mutations and horizontal resistance gene transfer, leading to serious antimicrobial resistance in
Campylobacter [
11,
38]. More seriously, some of the antibiotics to which the
Campylobacter isolates were resistant were used as therapeutic drugs in severe cases of infection [
39]. Although resistance rates varied in different regions, in general high resistance rates, especially to fluoroquinolones, were found in most of the studies in China. For example, in Zhang et al.’s study, the resistance rate of
Campylobacter to ciprofloxacin was 100%, and 94% to tetracycline, 61% to erythromycin, and 50% to ampicillin [
36]. Chen et al. reported that more than 98% of the tested
Campylobacter isolates were resistant to quinolones and tetracycline [
40]. Even as early as in 2002, the prevalence of quinolone resistance of the isolates had been up to 85.9% in Hong Kong [
41]. Low resistance rates of
Campylobacter were only reported in Northwest China [
42]. In our study, a very high resistance rate to β-lactams, tetracyclines and fluoroquinones was observed (Fig.
2), and a high resistance rate to the other drugs, such as erythromycin, was also found in the strong biofilm producers (Table
2). Our previous study showed that all of the fluoroquinolone-resistant strains contained a Thr-86-Ile substitution in GyrA, and that the CmeR-Box variations increased the expression of CmeABC efflux pump which led to the high resistance [
43]. Overexpression of drug efflux pump may not only contribute to fluoroquinolones resistance, but also increase resistance to other drugs [
44‐
46]. Bacteria exposing in efflux inhibitors or mutants in efflux pumps showed decreased biofilm, which suggested that efflux pumps also contributed to their biofilm formation [
47,
48]. Although more resistance mechanisms need to be revealed, efflux pumps seem to play important roles in antimicrobial resistance as well as biofilm formation.
Biofilms are sessile communities of bacterial cells enclosed in a self-produced extracellular polysaccharide matrix, which plays an important role in evading host immune clearance and resisting antimicrobial agents, leading to persistent and chronic infections [
26].
Campylobacter may form a monospecies biofilm, which protects them from environmental stress, including antibiotic treatment [
24]. In our tested strains, 64.6% were identified to be biofilm producers. Comparing with the non-biofilm producers, the biofilm producers possessed a higher resistance rate to ampicillin, neomycin, sulfamethoxazole, amikacin, clindamycin and erythromycin. Although studies on the correlation between biofilm and antimicrobial resistance were limited in
Campylobacter, positive impact of biofilm on reducing the permeation of ampicillin has been reported in other bacteria [
49]. Some regulators, such as LuxS, have also been reported to be linked to biofilm formation and antimicrobial resistance in some bacteria [
50]. We found an exception that the resistance rate to tetracycline was higher in non-biofilm producing isolates than in biofilm producing strains, it may be due to the high distribution of the resistance genes in non-biofilm producing isolates, such as
tet [
51]. It is interesting that all of the soil isolates and aerosol isolates were biofilm producers, which suggested that biofilm might be an important factor to help strain to survive in the surroundings as well as in the host. Our study on the biofilm-forming characteristics of
Campylobacter isolates would help us understand the increasing resistance to antibiotics of
Campylobacter as well as their pathogenicity to host.
In clade 3 and clade 4, 75% of the strains (20/25) were non-biofilm producers and the biofilm-forming abilities in these two clades were significantly lower than other clades (
p < 0.05). The closely related strains may have a common ancestor, and STs developing from one biofilm-forming ST origin may share better biofilm-forming ability. The correlation of the origin and phylogenetic relationship between their
C. jejuni isolates and biofilm-forming abilities has also been reported [
52]. Previous studies also showed that some gene variants were associated with different
C. jejuni multilocus sequence types, such as
fspA [
53] and
capA [
54]. The association between biofilm related genes and multilocus sequence types needs to be further studied. However, isolates within the same clade also exhibited varied abilities to form biofilm in our study. Ben et al. analyzed the genome sequences of strains with different biofilm-forming abilities, and found that three genes were associated with the increased biofilm formation in CC-21 and 43 genes in CC-45, but there was no overlap between these two CCs [
55]. These results suggested a complex genetic correlation between genetic background and biofilm formation.