Prevalence and genetic characterization of Campylobacter from clinical poultry cases in China

ABSTRACT Campylobacter is one of the leading causes of bacterial foodborne diseases globally. Poultry is considered a major reservoir for the transmission of Campylobacter to humans. The occurrence and epidemiology of Campylobacter in healthy poultry have been extensively studied, but the prevalence in diseased poultry has rarely been reported. In this study, a total of 1,563 intestinal samples, 547 gallbladder samples, and 3,729 parenchyma organ samples were collected from diseased chickens, ducks, and geese in Yangzhou Veterinary Hospital between December 2016 and September 2017. The bacterial isolates were analyzed for genotype, resistance, and virulence genes. The overall prevalence of Campylobacter was 17.9%, and a significantly higher prevalence was observed in the diseased chickens (21.8%) than in the diseased ducks (12.1%) and geese (11.9%). Campylobacter was significantly more prevalent in the intestinal tract (39.6%), followed by the gallbladder (11.7%) and parenchyma organ (9.7%). Among 84 whole-genome sequenced Campylobacter isolates, we identified 47 multi-locus sequence typing types, of which ST-51 and ST-825 were the most predominant sequence types (STs) for Campylobacter jejuni and Campylobacter coli, respectively. Phylogenetic analysis showed that Campylobacter isolates from different tissues from the same host tended to cluster together with the same STs, which indicated Campylobacter infection in different tissues in vivo. The antimicrobial resistance genes most frequently identified were gyrA (T86I) (97.6%) and tet (O) (65.5%). Our results revealed the high prevalence of Campylobacter in diseased poultry in China, which will help the farms take the necessary action to develop effective mitigation strategies for reducing Campylobacter infection in poultry. IMPORTANCE Campylobacter is a major cause of campylobacteriosis worldwide, and poultry is the main reservoir for its transmission. Campylobacter was generally considered to be a harmless commensal organism in poultry without pathogenic properties. However, it was proposed that a Campylobacter-like organism may be the cause of vibrionic hepatitis, which poses a significant public health risk. The occurrence and epidemiology of Campylobacter in healthy poultry have been studied systematically, but little is known about the epidemiology of Campylobacter isolates from diseased poultry in China. Therefore, this study determined the prevalence and molecular characterization of Campylobacter from diseased chickens, ducks, and geese in Yangzhou Veterinary Hospital between December 2016 and September 2017, which was critical for improving the diagnosis and prevention of Campylobacter infections.

bloody diarrhea or dysentery syndromes, mostly consisting of cramps, fever, and pain.Occasionally, more serious complications such as arthritis, septicemia, and Guillain-Barré syndrome can also occur with infection with Campylobacter jejuni or Campylobacter coli (5,6).
The most common cause of Campylobacter infections is foodborne, which is primarily caused by poultry, including chickens, ducks, and geese (1,5,7).Chickens dominate the world poultry industry (8).Meanwhile, the duck and goose industries are also important sectors of the poultry industry.China is the world leader in goose production, accounting for 94.1% of global production (9,10).C. jejuni and C. coli are often found naturally in poultry without any pathogenic features, which are primarily colonized in the ceca and colon, while the small intestine is colonized to a lesser extent (6,7,11).Colonization naturally occurs by horizontal transmission from the environment, and the infection rapidly spreads within the flock from one bird to another (12).
However, increasing evidence shows that Campylobacter can invade and destroy the intestinal mucosa and invade other organs, thereby spreading to other internal tissues (13,14).Clinical outcomes of its infection range from asymptomatic infection to life-threatening extraintestinal infections (15).For example, Campylobacter infections are associated with abortion in primates and ruminants by invading the genital tract (3,16).Meanwhile, Hasel et al. observed the heavy colonization of Campylobacter in human kidney abscesses (4).Other studies have also found Campylobacter in abscess infections of the head, brain, and chest walls (4,17,18).
In contrast, extraintestinal infections in poultry were rarely reported.Current studies describe extraintestinal infections in chickens but not in ducks and geese (13,19,20).Studies have identified Campylobacter in the livers of birds with no signs of disease or with vibrionic hepatitis (13).The high colonization of Campylobacter may contribute to the formation of white spots in the livers of commercially farmed birds (13,20).Spotted liver disease in chickens affects mortality and production, ranging from sporadic deaths in chickens to severe decreases in egg production and an increase in mortality of more than 1% per day in egg-laying flocks (13).Infection of extraintestinal organs caused by Campylobacter poses a significant public health risk (20).Therefore, it is important to investigate the distribution and epidemiology of Campylobacter isolated from diseased poultry.
Poultry diseases are known as a critical factor limiting the development of the poultry industry (21).Numerous studies on the occurrence and epidemiology of Campylo bacter in healthy poultry have been carried out (9,(22)(23)(24).However, few studies have focused on the epidemiology of Campylobacter in diseased poultry.This study aimed to determine the molecular epidemiology of Campylobacter isolates from diseased poultry at Yangzhou Veterinary Hospital, China.In addition, the molecular characteristics, genetic relationship, resistome, and virulome features of Campylobacter isolates from the intestinal tract, gallbladder, and parenchymatous organs of diseased poultry were investigated based on whole genome sequencing (WGS) analysis.This study provides preliminary data on Campylobacter from diseased poultry in China.

Sample collection
The source of our samples was diseased poultry from various cities collected and diagnosed at the Yangzhou Veterinary Hospital between December 2016 and September 2017.Poultry sources collected in this study suffered from bacterial, viral, or parasitic infections with symptoms such as enteritis, hepatitis, pericarditis, or other lesions concerning organs.During isolation, one sample was collected from each lesion of a diseased poultry carcass for identification.The collection niches were the tissues where the lesion occurred; for example, it could be parenchymal organs (i.e., heart, liver, spleen, lung, and kidney), gallbladder, and intestines, so the number of samples collected from each individual source varied with the status of the lesion.Thus, the prevalence calculations pertain to samples and not individual animals.The majority of poultry sources came from different farms, with an age range of 5 days to 1 year.A total of 1,563 intestinal samples, 547 gallbladder samples, and 3,729 parenchymal organ samples (i.e., heart, liver, spleen, lung, and kidney) from diseased chickens, ducks, and geese with enteritis, hepatitis, pericarditis, or other symptoms concerning organs were collected in this study.Specifically, 946 intestinal samples, 334 gallbladder samples, and 2,264 parenchyma organ samples were derived from chickens.Then, 148 intestinal samples, 48 gallbladder samples, and 357 parenchyma organ samples were collected from ducks.Additionally, 469 intestinal samples, 155 gallbladder samples, and 1,110 parenchyma organ samples were collected from geese (Table 1).All samples were collected as described previously (20).In brief, all the samples were aseptically placed into sterile Whirl-Pak bags (Nasco, Fort Atkinson, WI, USA), labeled, stored on ice, and immediately transported to the laboratory at Yangzhou University within 24 h.The experiment was strictly conducted according to the Guide for the Care and Use of Laboratory Animals of the Ministry of Health [SYXK(Su) 2017-0045], China, with the permission of the Research Ethics Committee of Yangzhou University.

Isolation and identification of Campylobacter
The isolation and identification of Campylobacter were performed as previously described with some modifications (3).Briefly, all samples were homogenized in phosphate-buffered saline using homogenizers and then enriched in buffered peptone water for 24 h at 42°C in a microaerobic atmosphere of 10% CO 2 , 5% O 2 , and 85% N 2 .After enrichment, approximately 10 µL of the culture was streaked onto Campylo bacter blood-free selective agar-containing charcoal cefoperazone deoxycholate (CCDA) (Oxoid, Basingstoke, United Kingdom) containing six antibiotics (60 µg/mL cefopera zone, 10 µg/mL rifampicin, 20 µg/mL amphotericin B, 6 µg/mL polymyxin B, 10 µg/mL trimethoprim, and 100 µg/mL cycloheximide) and incubated for 48 h at 42°C under microaerobic conditions.Suspected Campylobacter colonies were evaluated based on morphology (gray or brown, wet with metallic luster) and then streaked onto the CCDA plates without antibiotics, which were further incubated at 42°C under microaerobic conditions.All isolates were identified by multiplex PCR using 16S rRNA, mapA, and ceuE primers.All primers used in this study are listed in Table S1.Three to four colonies were picked up and confirmed as C. jejuni or C. coli by multiplex PCR.One colony of C. jejuni or C. coli for each sample was stored at −80°C.

Whole genome sequencing
For strain selection, available isolates were stratified by different hosts (chickens, ducks, and geese), and we randomly selected isolates from different niches to ensure a similar representation of different hosts and different isolation sites.A total of 84 strains were selected, including 46 C. jejuni isolates and 38 C. coli isolates (Table S2).The genomic DNA of the isolates was extracted using the TIANamp Bacteria DNA Kit (TIANGEN, China) according to the manufacturer's instructions.WGS was carried out using the NovaSeq 6000 sequencing platform (Illumina Inc., San Diego, CA, USA) (3).Quality-con trolled paired-end reads were de novo assembled independently using the SPAdes v3.10 assembler (3).The sequence data of all isolates have been deposited in NCBI BioProject PRJNA926478 with the BioSample accession numbers SAMN32874603 to SAMN32874686 (Table S3).

Molecular characteristics analysis
WGS data were used for multi-locus sequence typing (MLST) typing.The sequence types (STs) and clone complexes (CCs) were obtained by submitting the whole genome sequence of isolates to the Campylobacter MLST database (https://pubmlst.org/organisms/campylobacter-jejunicoli).Furthermore, the WGS data were used to iden tify antimicrobial resistance genes and virulence genes.Antimicrobial resistance determinants were screened in each Campylobacter genome using ABRicate, which includes the Resfinder, CARD, ARG-ANNOT, and NCBI ARRGD databases (25).Regarding macrolides, two markers of antimicrobial resistance, including the A2075G mutation in the gene encoding 23S rRNA and the ermB gene, were analyzed (26).gyrA and 23S RNA gene point mutations were determined using Resfinder (25).The bla OXA605 -positive isolates were screened for the single-nucleotide mutation (transversion G→T) in the promoter region of the target gene.NCBI blast tools were used to compare nucleotide sequences (27).Virulence genes were identified using BLAST against the VFDB database (mgc.ac.cn/VFs/) (28).Core genome single-nucleotide polymorphisms of the C. jejuni and C. coli isolates were obtained and given to FastTree2 (reference) for the reconstruction of the Maximum Likelihood tree using ParSNP (29).Furthermore, the tree data were visualized and modified using interactive Tree Of Life.The tree length is ignored for the sake of easier visualization.

Statistical analysis
For the prevalence of Campylobacter spp.from different poultry or niches, the data were analyzed using the chi-square test with the SPSS statistical package (SPSS Inc., Chicago, USA).Statistical significance was set at P ≤ 0.05.

Phylogenetic analysis of Campylobacter isolates
The phylogenetic tree was constructed using maximum likelihood estimation (Fig. 1).Phylogenetic analysis showed high genetic diversity among the C. jejuni and C. coli isolates (Fig. 1).Forty-six C. jejuni isolates were divided into two clades.Interestingly, isolates from diseased chickens were clustered together, indicating a close genetic relationship with the same host.Moreover, we found that the organ isolate HJL0190 and the intestinal isolate HJL0200 from the same diseased chicken individual appeared on the same branch with the same ST type (Table S2).C. jejuni isolates from ducks clustered together with isolates from diseased geese.Notably, 38 C. coli isolates from different hosts or tissues tended to be clustered together.C. coli isolates from diseased chickens were closely clustered with isolates from diseased geese.We identified three pairs of strains that were from different isolation sites of the same individual chicken, present on the same branch, with the same ST type.These strain pairs included HJL0133 and HJL0144, HJL0077 and HJL0103, and HJL0085 and HJL0110 (Table S2), which suggested a multiple organ infection in vivo (Fig. 1).

Resistance gene analysis of Campylobacter
To determine the resistome, we performed in silico analysis to identify genes associated with antimicrobial resistance by means of the ABRicate pipeline using the Resfinder, CARD, ARG-ANNOT, and NCBI ARRGD databases.In total, 17 antimicrobial resistance genes were detected in Campylobacter isolates in this study.The results are shown in Fig. 2. The point mutations in the gyrA gene (T86I), conferring resistance to quinolone, were the most (97.6%, 82/84) prevalent resistance gene and were detected in 44

Virulence gene analysis of Campylobacter
The 84 whole genome-sequenced Campylobacter isolates in this study were tested for the presence of 14 different virulence genes, including genes relating to adhesion (cadF, jlpA, porA), invasion (ciaB, flaC), toxins (cdtA, cdtB, cdtC), and the Type IV secretion system (T4SS, virB11, virB10, virB9, virB8, virB4, and virD4).The presence of each gene in Campylobacter isolates is summarized in Fig. 3. C. jejuni isolates had more virulence genes than C. coli isolates, suggesting that C. jejuni could be more virulent than C. coli.For the genes associated with adhesion, the cadF gene was present in all Campylobacter isolates.Compared to C. coli, a higher proportion of C. jejuni isolates carried jlpA (100.0%) and porA (73.9%) genes for adhesion.The invasion-related genes ciaB and flac were present in all the Campylobacter isolates.Notably, the genes responsible for the production of the cytolethal distending toxin (CDT) (cdtA, cdtB, and cdtC) were found in most C. jejuni isolates.In contrast, they were absent in all C. coli isolates.Among C. jejuni isolated from diseased geese, 6 of 18 isolates tested positive for the cdtA gene.Additionally, only one Campylobacter isolate (HJL0125 isolate) from the intestinal tract of a goose was positive for the T4SS gene cluster.

DISCUSSION
Campylobacter is an important zoonotic pathogen in humans that is prevalent in poultry.
Campylobacter was generally considered to be a commensal organism in poultry without pathogenic properties, but little is known about the epidemiology of Campylobacter isolates from diseased poultry in China.Our result indicated a higher prevalence of Campylobacter in diseased chicken cacasses (21.8%) than in ducks (12.1%) and geese (11.9%), which was consistent with the previous observation that chickens are generally considered the most common source (30).Notably, the prevalence of Campylobacter in the intestinal tract of diseased chickens, ducks, and geese was higher than that in the gallbladder and parenchyma organs, suggesting that Campylobacter were mainly colonized in the poultry intestinal tract.Traditionally, the intestinal tract of food animals, especially poultry, is the main niche for Campylobacter, and the poultry intestinal tract provides a favorable environment for Campylobacter growth (12).However, our results showed that Campylobacter could also be detected in the gallbladder and parenchyma organs at a frequency ranging from 6.1% to 15.3%, implying that extra intestinal colonization by Campylobacter may be a reservoir for the dissemination of Campylobacter in poultry and humans.Laconi et al. also found Campylobacter in the extraintestinal tissues of chickens, including the liver and spleen (14).The consumption of chicken liver contaminated with Campylobacter has been associated with an increased risk of infection and has led to human campylobacteriosis (14).
In the current study, the incidence of intestinal Campylobacter carriage in diseased chickens was found to be 47.3%.This is higher than the prevalence ratio of 25.4% detected in the intestinal tract of healthy chickens from farms or live poultry markets in central China (31).Previous studies from both Poudel et al. and Hue et al. found that C. jejuni was predominant in the intestinal tract of chickens (6,23).However, among the Campylobacter isolated from the intestinal tract of diseased chickens in this study, the most prevalent species was C. coli, accounting for 70.7% (316/447) of total intestinal Campylobacter isolates.This finding indicated that C. coli was increasingly responsible for Campylobacter-associated intestinal tract infections.In addition, this study shows the high prevalence of Campylobacter on the intestinal tract of diseased ducks (29.7%) and geese (27.3%), which agrees with the studies reported by others who found the high prevalence of Campylobacter spp. in ducks and geese from Iran and Poland (24,32).
The C. jejuni population was highly diverse (46 isolates were assigned to 32 STs), and 50.0% of the STs were identified in isolates from chickens.This high genetic diversity observed in C. jejuni isolates from chickens is consistent with previous studies (33).Here, CC574 (15.2%) was the most predominant population among C. jejuni isolates from diseased chickens, in contrast to the result that CC21 was the most common lineage among chickens in China (34).CC443 was the second most common CC among C. jejuni isolates, and the third most common was CC464 (6.5%).Notably, previous studies found that isolates assigned to CC443 and CC464 were observed in both poultry and humans (7,(35)(36)(37), suggesting a potential bidirectional transmission of these strains between them.Our results found that 94.7% of C. coli isolates from diseased poultry belonged to CC828, which was reported as the predominant CC in C. coli from humans, poultry, and the environment (38)(39)(40).Interestingly, the C. coli isolates from chickens clustered with those from geese, and the C. coli isolates from different niches in the same host tended to cluster together with the same STs, suggesting potential transmissions of C. coli between different hosts or different niches in vivo.
We further found that the high percentage (97.6%) of Campylobacter strains with a mutation of the gyrA gene, which confers quinolone resistance, is consistent with previous reports indicating that Campylobacter exhibits a high level of resistance to fluoroquinolone (41).Regrading tetracycline resistance, 65.5% (55/84) of Campylobacter isolates from diseased poultry carried the tet(O) gene.The tet(O) gene has been reported to be the only tetracycline resistance determinant identified in Campylobacter, and it is commonly detected in all tetracycline-resistant Campylobacter isolates.The high prevalence of the tet(O) gene in our isolates suggested a potential high tetracycline resistance in Campylobacter isolates from poultry in China.Indeed, Han et al. confirmed that 94.6% of tetracycline-resistant Campylobacter isolates from broilers at slaughter in China are positive for the carriage of the tet(O) gene (22).The high prevalence of the tet(O) gene in Campylobacter isolates from poultry could be due to the frequent use of tetracycline during feeding.
The prevalence of the bla OXA-605 resistance gene in C. jejuni isolates varied among different hosts, with a higher presence in C. jejuni isolates from chickens (78.9%, 15/19) than from ducks (11.1%, 1/9) and geese (22.2%, 4/18).The bla OXA-61 -like subfamily, which includes bla OXA-605 , is reported to require promoter mutation for resistance in Campylobacter.The G→T transversion has been described to restore the TATA box (from GAAAAT to TAAAAT), making it fully functional, thus increasing oxacillinase production and consequently causing high-level ampicillin resistance (42,43).In general, it was determined that 57.9% (11/19) of C. jejuni isolates from diseased chickens and 5.6% (1/18) of C. jejuni isolates from diseased geese harbored the single-nucleotide mutation.This observation prompted the formulation of a hypothesis suggesting that C. jejuni isolates from diseased chickens may be more resistant to β-lactams than isolates from diseased ducks and diseased geese.Mouftah et al. also found a high prevalence of the bla OXA-605 gene in C. jejuni from broiler carcasses (44).
The distribution of virulence genes in poultry C. jejuni isolates was not statistically different, suggesting that all isolates in this study had the potential to cause poultry illness.Virulence analysis showed that C. jejuni isolates harbored most of the known virulence factors, while C. coli isolates lacked most virulence genes.For example, CDT encoded by genes cdtA, cdtB, and cdtC plays an important role in toxin production and helps in the pathogenesis of Campylobacter (45).Here, we found that cdt gene clusters were abundant in the C. jejuni isolates, while they were absent in the C. coli isolates, indicating that C. jejuni is more virulent than C. coli.
In this study, most samples isolated from diseased poultry carcasses come from different individuals due to sampling principles, which introduces a level of individual variation that may impact the generalizability of the findings.Certainly, the fact that only 84 bacterial isolates out of a total of 1,046 isolates were randomly selected and included in the analysis presents a significant limitation to the study.This represents less than 10% of the total isolates, which could potentially impact the reliability and generalizability of the results.Further exploration is needed in the future.

Conclusion
In conclusion, this study reveals preliminary data on Campylobacter from diseased poultry in China.Here, the overall prevalence of Campylobacter was 17.9%, and a significantly higher prevalence was observed in the diseased chickens (21.8%) than in the diseased ducks (12.1%) and geese (11.4%) (P ＜ 0.05), which was consistent with the fact that chickens were the most common source.C. coli accounted for 53.1% of total Campylobacter isolates.Although Campylobacter was most prevalent in the intestinal tract (39.6%), it was also detected in the gallbladder and parenchyma organs of diseased poultry, with frequencies ranging from 6.1% to 15.3%, implying that extraintestinal colonization by Campylobacter may be a reservoir for Campylobacter.
Forty-seven MLST types were identified among the 84 whole genome sequences of Campylobacter, of which ST-51 and ST-825 were the most common STs for C. jejuni and C. coli, respectively.Phylogenetic analysis showed that Campylobacter isolates from different organs in the same host tended to cluster together with the same STs, suggesting a multiorgan infection of Campylobacter in vivo.Resistome analysis predic ted widespread resistance to fluoroquinolones and tetracycline.Our results revealed the high prevalence of Campylobacter in diseased poultry in China, which will help to develop effective control strategies for reducing Campylobacter infection in poultry.

FIG 3
FIG 3 Distribution of antimicrobial resistance genes.Binary heatmaps show the presence and absence of antimicrobial resistance genes.C. jejuni isolates form the green cluster, and C. coli isolates form the red cluster.Colored cells represent the presence of genes.