Thermophilic Campylobacter spp., particularly Campylobacter jejuni and Campylobacter coli, have been established as leading causes of food-borne illnesses worldwide (Vetchapitak and Misawa, 2019; European Food Safety Authority, 2021; Sher et al., 2021). The U.S. Centers for Disease Control and Prevention (Centers for Disease Control and Prevention, 2022) estimated that 1.5 million U.S. residents are infected with Campylobacter each year. Most patients have acute, self-limiting gastroenteritis, but some may have severe and long-lasting illnesses, which require antibiotic treatment, particularly in immunocompromised patients (Ma et al., 2014). Additionally, the infection by Campylobacter may be associated with a number of complications such as polyarthralgia, Guillain-Barre syndrome (GBS), Miller Fisher syndrome and even death (Kaakoush et al., 2015).

Campylobacteriosis is transmitted by eating raw or undercooked poultry meat (Centers for Disease Control and Prevention, 2022). Campylobacter contaminates poultry meats prior to or during processing representing a potential health threat to consumers (Suzuki and Yamamoto, 2009). Campylobacter contamination in poultry farms could occur via feed, water, soil, contact animals, biosecurity threats, and vehicles (Ghareeb et al., 2013).

The survival and pathogenicity of Campylobacter species are all influenced by several virulence factors (Casabonne et al., 2016). Bacterial motility, adherence to the intestinal epithelial walls, colonization and cytotoxin production are the main bacterial virulence factors. Several genes related to Campylobacter virulence factors have recently been identified including adhesion and colonization (flaA, flhA, cadF, and racR), invasion-associated markers (ciaB, iam, and virB11), and ganglioside mimicry (wlaN) (Bolton, 2015).

There is a growing antibiotic resistance crisis in clinical medicine since antibiotics were historically used in food animal production either for treatment or for growth promotion, which led to human exposure and infection through a variety of pathways, including meat and poultry products (Price et al., 2007). Moreover, a significant portion of the antibiotics provided are not absorbed by the animals and are excreted in the urine and feces. In Campylobacter infections, antibiotic therapy is commonly required for immunocompromised patients and those with severe campylobacteriosis (Kaakoush et al., 2015). Generally, Campylobacter infections are treated with macrolides (erythromycin, clarithromycin, and azithromycin), although fluoroquinolones (ciprofloxacin) are the most effective drugs to treat diarrhea (Aarestrup et al., 2008). Additional alternative drugs for treatment are tetracycline, doxycycline, and chloramphenicol (Skirrow and Blaser, 2000).

Pastured poultry farms in the USA are considered an important source of animal production that may provide an important opportunity to strengthen rural communities (Conner et al., 2008). Consumer preference of free-range and pasture-raised animal products such as meat, milk, and eggs has grown (Stampa et al., 2020). Because there is a paucity of information on the quality of pasture-raised chickens, many customers feel that these products are of superior quality in contrast to conventionally-farmed chickens, due to their more natural growing conditions (Yeung and Morris, 2001). There is insufficient research on genotyping, presence of virulence determinants, and antibiotic resistance of Campylobacter isolates from pasture-raised poultry farms; therefore, the purpose of this study was to explore genetic relatedness of thermophilic Campylobacter isolated from pasture-raised poultry farms and the following processing operations of broiler carcasses, and to assess antibiotic resistance phenotypes and genotypes as well as to screen the presence of virulence determinants in the retrieved isolates.

The farm description, sample collection and processing, and Campylobacter isolation methods were previously described (Rothrock et al., 2016). Briefly, the samples were collected from feces, pasture soil, cecal content at processing, whole carcass rinsates and final whole carcass products. All samples were collected in the field and were brought back to the laboratories in a cooler packed in ice. The total amount of fecal and soil samples was at least 25 grams per sample. For homogenization, three grams (feces, cecal and soil samples) were diluted 1:3 in 10 mM phosphate buffered saline (PBS) in sterile filtered stomacher bags (Seward Laboratories, Inc., Bohemia, NY, United States). For rinsates, 100 mL of 10 mM PBS were added to each carcass within the storage bag, and the bags were vigorously shaken for 1 minute. The rinsates were collected into the sterile filtered stomacher bags (Seward Laboratories, Inc.). All samples were homogenized for 1 minute with a Stomacher^® 400 Blender (Seward Laboratories, Inc.), and these homogenates were used for the downstream Campylobacter isolation. A volume of 100 μL from the above homogenized suspension was plated onto Campy-Cefex agar (prepared in the laboratory; Stern et al., 1992). The plates were incubated at 42 ± 1°C for 36 to 48 h in a microaerobic condition (85% N₂, 10% CO₂ and 5% O₂) (Hiett et al., 2008; Yeh et al., 2013). Presumptive Campylobacter colonies were selected and enumerated on Brucella agar supplemented with 10% lyzed horse blood for isolation (prepared in the laboratory; Stern et al., 1992). The plates were incubated as described above. Speciation of Campylobacter was carried out using a Q7 BAX Real-Time PCR system according to the manufacturer’s instructions as described previously (Yeh et al., 2022). An end-point multiplex PCR assays were also performed. The 16S rRNA primers specific to Campylobacter in the PCR assays generated amplicons both in C. jejuni and C. coli samples, verifying the isolates as Campylobacter (Linton et al., 1997). The PCR assays with hipO primers amplified a 323-bp product in the C. jejuni samples, but not in the C. coli samples, verifying the isolates as C. jejuni (Caner et al., 2008). The PCR with primers from the ask gene generated about a 550-bp gene fragment that identified the samples of C. coli (Linton et al., 1997). Campylobacter isolates were frozen at −80°C in Luria-Bertani broth with 20% glycerol until downstream analyses were performed.

Within the poultry industry, concerns have been expressed over the microbiological safety of pasture-raised poultry products despite consumer confidence in these types of production. The continuous exposure of the flocks to the pasture environment increases the possibility of contact with other sources of Campylobacter such as wild birds, insects, etc. (Berg, 2001). Due to the growing preference of this type of meat product, the question of whether the welfare benefits for this type of production is aligned with appropriate food safety should be explored. As a result, the current study was carried out to explore the genetic relatedness, virulence, and antimicrobial susceptibility of thermophilic Campylobacter by characterizing 97 isolates from pasture-raised poultry farms and the following processing operations.

Campylobacter sequence-based genotyping techniques yield data that is consistent across host sources, reproducible, and suitable for population genetic study (Dingle et al., 2001). Multi-locus sequence typing (MLST) identifies clonal complexes and links Campylobacter species to specific animal sources (Dingle et al., 2002; Colles et al., 2008). In this study, Campylobacter genotypes identified by MLST were diverse based on the number of samples taken from each flock. These results are in an agreement with the results reported by Colles et al. (2010) who found a great diversity in Campylobacter genotypes isolated from free-range broiler flocks. However, Bull et al. (2006) and Lindmark et al. (2006) reported a lower ST diversity of up to three STs within housed flocks. These discrepant findings highlight the importance of collecting large numbers of samples from a flock in order to identify the full range of variability within a flock. The most common clonal complexes CC607, CC21 and CC353 were predominant among C. jejuni strains in our study. These CCs were reported also as the most common CCs identified from human samples in various geographic regions (Dingle et al., 2001; de Haan et al., 2010; Smid et al., 2013). On the ST level, ST-353 and ST-50 were reported also as the most widely distributed STs among human and broiler C. jejuni isolates (Harvala et al., 2016; Elhadidy et al., 2018). These results highlight the importance of poultry sources for human campylobacteriosis.

The presence of thirteen virulence genes was investigated by PCR to confirm the pathogenic potential of these isolates. Significant differences in the occurrence of virulence genes were observed, C. jejuni isolates had a higher virulence potential than C. coli isolates. These results are in an agreement with those reported by Casabonne et al. (2016) and Wieczorek et al. (2013) from conventionally raised broiler flocks. Our results showed that the flaA gene was detected in all strains, and the flhA gene was found in most of the isolates examined. Similar findings are also reported by Rossler et al. (2020) that flaA and flhA genes were detected in all their isolates collection. Mobility of Campylobacter, involving the coordination of many genes (such as flaA and flhA), is important for passage through the stomach and gut (Gilbreath et al., 2011). The presence of flaA and flhA genes in a high proportion of the isolates examined suggests that motility and virulence mechanism are synchronized during Campylobacter pathogenesis (Wieczorek et al., 2015; Zhang et al., 2016; Frazão et al., 2017; Rossler et al., 2020).

The Campylobacter adhesion to fibronectin F (cadF) gene, encoding an adhesin and fibronectin-binding protein that involves in the process of invasion and influences the microfilament organization in host cells (Zhang et al., 2016), was also detected in most of our isolates (Table 4). Similar observations have been reported that the high frequency of the cadF gene in Campylobacter species was detected from poultry productions and processing operations (Rozynek et al., 2005; Frazão et al., 2017; Rossler et al., 2020). Additionally, Ziprin et al. (1999) demonstrated that Campylobacter cadF-negative strains are not able to colonize in chicken gastrointestinal tract. Therefore, the cadF gene product may play a similar function in human pathogenesis and causing disease.

The Guillain-Barré syndrome associated gene (wlaN) and its gene product have ganglioside-like structures and is responsible for specific lipooligosaccharides (LOS) synthesis (Hermans et al., 2011). This LOS synthesis is thought to be involved in the development of Guillain-Barré and Miller-Fischer syndromes after C. jejuni infection (Gilbert et al., 2000; Linton et al., 2000). The presence of this gene in Campylobacter may increase the risk for suffering post-neurological conditions. Our findings indicate its presence in 36.1% of the total thermophilic Campylobacter isolates, which is in line with many previous studies in conventional poultry management systems (Talukder et al., 2008; Koolman et al., 2015; Wieczorek et al., 2018).

Campylobacter toxins are important virulence marker determinants. One of the toxin groups is the cytolethal distending cytotoxins, which are encoded by the cdt genes and form polycistronic cdt operons. The gene products include CdtA, CdtB, and CdtC cytotoxins, which are toxic to host enterocytes (Carvalho et al., 2013). These cytotoxins play important roles in development of diarrhea by interfering with the proliferation and differentiation of intestinal crypt cells (Scuron et al., 2016). The three subunits are required for the full activity of the toxins (Lapierre et al., 2016). CdtB displays enzymatic Dnase activity resulting in cell-cycle arrest and cell death, while CdtA and CdtC are responsible for the translocation of CdtB across the target cell membrane (Lara-Tejero and Galan, 2001). In this study, these toxin genes were detected in majority of C. jejuni isolates (98.7–97.5%).

Campylobacter survival in the digestive tract is highly dependent on the ciaB gene. This gene can secret a CiaB protein that is responsible for the invasion and colonization of this microorganism in chicken intestines (Hermans et al., 2011). Among Campylobacter isolates, it was found in a frequency of 80.4%. Similarly, in conventionally reared broilers, the ciaB gene was detected in a similar prevalence by Raeisi et al. (2017) and Wieczorek et al. (2018). Because the ciaB gene is important in the early stages of colonization, their removal may causes bacterial failure to survive the stress of passage through the gut followed by colonization failure (Ziprin et al., 2001). Additionally, regulatory protein R (racR) gene and its gene product regulate temperature during growth and colonization of this microorganism in the hosts. The prevalence of racR in our study is 83.5%, which is similar to that reported in conventional broiler flocks by Datta et al. (2003) and Talukder et al. (2008); however, Hanning et al. (2012) demonstrated a lower racR prevalence rate (34%) in a pasture-raised broiler flock study.

The enterochelin binding lipoprotein encoded by siderophore transport (ceuE), which has an important role in virulence and regulation of the siderophore transport system (Hermans et al., 2011), was detected in 72.2% of the Campylobacter isolates.

The use of antibiotics, either overuse or abuse, in food animals contributes to the establishment of antimicrobial resistance (AMR) in commensal and zoonotic enteric bacteria (Varga et al., 2009; van Boeckel et al., 2015). To prevent the spread of AMR Campylobacter through the food chain, it is critical to continuously monitor its antimicrobial resistance and resistance mechanisms. In this study, five (5.2%) Campylobacter isolates were resistant to quinolones and fluroquinolones [nalidixic acid (n = 4) and ciprofloxacin (n = 1)]. The lower resistance of quinolones and fluoroquinolone-resistant isolates in this study may be related to that the U.S. Food and Drug Administration (FDA) banned the use of fluoroquinolones in poultry production in the United States in 2005 (Griggs et al., 2005). However, other studies argued that the FDA’s restriction on fluoroquinolones in chicken production may not be enough to mitigate the resistant Campylobacter in poultry products, because fluoroquinolone-resistant Campylobacter was found in persistent pollutants of poultry products even after discontinuous on-farm fluoroquinolone use (Price et al., 2007). Monitoring the prevalence of resistant strains in chicken flocks, production facilities, consumer poultry products, and human diseases is therefore crucial in order to accurately evaluate the effectiveness of this policy. The low frequency of resistance to quinolones and fluoroquinolones in this study may also related to the fact that antibiotics were not utilized by any farms during the duration of this study. Luangtongkum et al. (2006) reported a significant difference between antimicrobial resistance rates of <2% vs. 46–67% in organic and conventional raised poultry farms, respectively.

The Campylobacter isolates that displayed resistant to quinolones and fluroquinolones [ciprofloxacin (MIC = 8 μg/mL) and nalidixic acid (MIC = 64 μg/mL), respectively] were further examined for the presence of the most common mutation site. A point mutation at position 86 leading to threonine replacement by isoleucine was detected in the QRDR of the gyrA gene from our isolates. A MAMA-PCR was used to determine the presence of this type of mutation (Zirnstein et al., 1999; Payot et al., 2004). In this protocol, a conserved forward primer, CampyMAMAgyrA1, and a reverse mutation detection primer, CampyMAMAgyrA5 were used to generate a 265-bp PCR product, indicating the presence of the Thr-86-Ile (ACA to ATA) mutation in the C. jejuni gyrA gene. This method was used as an alternative to nucleotide sequencing because it is not accessible in ordinary microbiology laboratories. Our results revealed that this mutation was found among all phenotypically resistant isolates. In contrast, this mutation was found to be absent in some quinolone resistance isolates, leading researchers to speculate that it could be linked to alternative resistance mechanisms (Bolton et al., 2013; Elhadidy et al., 2018; Yeh et al., 2022).

Emergence of resistance to erythromycin by Campylobacter isolates has been reported (Deng et al., 2015; Liu et al., 2019; Jehanne et al., 2021). Resistance of Campylobacter to this macrolide is chromosomally mediated, most commonly due to a shift in the target site on the 23S rRNA subunit. These mutations have been identified at locations of 2074 and 2075 (Vacher et al., 2003). The transitory mutation A2075G is the most prevalent among erythromycin-resistant Campylobacter isolates, while the A2074C mutation is less identified among the resistant strains (Vacher et al., 2003). In erythromycin resistant isolates in this study (MIC >64 μg/mL), A2075G was detected in one isolate, while A2074G could not be identified from any isolate. Additionally, ermB was found in three out of nine erythromycin-resistant isolates, while Elhadidy et al. (2020) could not identify ermB gene from any erythromycin-resistant Campylobacter isolates. These findings on the molecular basis of macrolide resistance in Campylobacter revealed the importance of additional resistance mechanisms in Campylobacter encoding erythromycin resistance such as CmeABC is a multi-drug efflux pump system broadly distributed in Campylobacter, representing an important mechanism for antibiotic resistance (Lin et al., 2002).

Interestingly, one isolate in this study was gentamicin resistant (MIC>32 μg/mL). This result is congruent with that of Luangtongkum et al. (2006), who reported that none of the Campylobacter species isolated from conventionally or organically raised broilers were gentamicin resistant. Similarly, Giacomelli et al. (2014) and Elhadidy et al. (2018) could not identify gentamicin resistant isolates among Campylobacter isolates from poultry in Italy and Belgium, respectively. On the other hand, Saenz et al. (2000) found a 25% prevalent rate of gentamicin resistant isolates from broilers in Spain. The common low prevalence of gentamicin resistance may contribute to few usages of this antibiotic during the poultry production (Saenz et al., 2000; Roth et al., 2019).

Our results showed the high prevalence of tetracycline resistance among the isolates (MIC >64 μg/mL) (Table 3). These findings are also reported by other researchers from Kenya, Finland, Iraq, Poland, and USA (Luangtongkum et al., 2006; Nguyen et al., 2016; Pohjola et al., 2016; Wieczorek et al., 2018; Shakir et al., 2021) where C. jejuni and C. coli were isolated from small scale and backyard chicken flocks. In addition, Bailey et al. (2019) found that tetracycline resistances in organic farms were more common than the conventional farms. The reports have demonstrated that the plasmid-encoded tet (O) gene is responsible for tetracycline resistance in Campylobacter (Gibreel et al., 2004; Wozniak-Biel et al., 2018; Elhadidy et al., 2019), and this gene can be horizontally transferred between C. jejuni and C. coli isolates in the intestines of food animals and humans (Kim et al., 2010). Interestingly, the presence of phenotypic tetracycline resistant isolates that did not harbored tet (O) gene may be related to the genetic inactivated of efflux pumps (Jeon et al., 2011). The high rates of resistance reported for tetracycline could be attributed to the overuse during the poultry production (Giacomelli et al., 2014).

In conclusion, MLST analysis showed high genetic diversity among both C. jejuni and C. coli isolates. The identified STs were reported also as the most common STs identified from human samples in various geographic regions. These results highlight the importance of poultry sources for human campylobacteriosis. Additionally, Campylobacter isolated from pasture-raised poultry flocks from this study were generally consistent with Campylobacter previously isolated from conventionally reared broiler flocks in regard to ST prevalence and diversity, antibiotic resistance patters, and virulence. Thus, in terms of public health risk of campylobacteriosis, these results indicate that pasture-raised poultry products appear to be equivalent conventionally reared products, but still represents a potential zoonotic source of Campylobacter that requires further investigation.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.

AA: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. H-YY: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Project administration, Supervision, Writing – original draft, Writing – review & editing. HR: Data curation, Investigation, Methodology, Writing – review & editing. MR: Resources, Writing – review & editing.