Detection of Arcobacter species in different intestinal compartments of broiler chicken during slaughter and processing

Abstract Arcobacter spp. are commonly present on meat products. However, the source of contamination on chicken meat is under dispute. Since different studies reported contradictory results on the occurrence of Arcobacter spp. inside the intestinal tract of chicken, our study examined four intestinal compartments at four significant production steps during broiler slaughter and processing in the slaughterhouse. Altogether, 157 intestinal tracts from 19 flocks were examined qualitatively and semiquantitatively applying a selective enrichment. Further verification was performed by mPCR and rpoB sequencing. Arcobacter spp. were only detected sporadically in intestinal contents after bleeding (2/32) and in none after scalding (0/32). After defeathering, Arcobacter spp. were detected in 62% (18/29) of the intestinal contents with 28% (8/29) of the duodenal, 21% (6/29) of the jejunal, 3% (1/29) of the cecal, and 55% (16/29) of the colonic samples tested positive with loads up to 24,000 MPN/g in the colonic content. Further 88% (7/8) of colonic tissue samples were tested positive. After evisceration, the prevalences (58/64) and loads of Arcobacter spp. display comparable levels in the intestinal contents like after defeathering. In conclusion, our data point out that Arcobacter spp. are most likely detected in the colonic intestinal compartment of the chicken after defeathering and evisceration. Therefore, not only cross‐contamination originating from the environment inside the slaughterhouse may cause carcass contamination with Arcobacter spp. on broiler chicken carcasses. The detection of Arcobacter spp. in duodenal and jejunal contents as well as in the colonic tissue indicates that there possibly exists an Arcobacter reservoir inside the chicken.

However, most authors regard Arcobacter spp. in animals as commensals (Ramees et al., 2017). Over the last years, attempts were made to assess the impact of Arcobacter on humans. In 2002, the International Commission on Microbiological Specifications for Foods (ICMSF) classified the species A. butzleri and A. cryaerophilus as a serious hazard for human health (ICMSF, 2002). Several sporadic cases of gastroenteritis, bacteremia, endocarditis, and peritonitis associated with Arcobacter have been reported in humans (Ho et al., 2006). Furthermore, a large study in Belgium determined Arcobacter as the fourth most common pathogen group in fecal samples of enteritis patients (Van den Abeele, Vogelaers, Van Hende, & Houf, 2014).
Arcobacter spp. were isolated from various sources like feces, sewage, water, seafood, milk, vegetables, and meat products (Collado & Figueras, 2011;Ramees et al., 2017;Wesley & Miller, 2010). The prevalence of Arcobacter spp. in meat products is high, especially in products of poultry origin, followed by pork and beef (Ho et al., 2006;Kabeya et al., 2004). The contamination of poultry meat products most probably occurs during processing in the poultry slaughterhouse (Gude, Hillman, Helps, Allen, & Corry, 2005;Hsu & Lee, 2015). While Arcobacter spp. were detected in the intestinal content of chicken in several studies (Ho, Lipman, & Gaastra, 2008;Van Driessche & Houf, 2007), others detected them only on chicken carcasses and in the environment inside the slaughterhouse (Atabay & Corry, 1997;Gude et al., 2005;Houf, De Zutter, Van Hoof, & Vandamme, 2002). Since there is no standardized protocol for the detection of Arcobacter spp., various methods have been applied in corresponding studies, which makes it difficult to compare results of different studies.
However, since Arcobacter spp. are commonly present on poultry products (Houf et al., 2002) and may pose a hazard to human health, it is necessary to clarify the routes of transmission of Arcobacter spp.
in the chicken processing chain. The purpose of this study was to examine four sections of the intestinal tract (duodenum, jejunum, cecum, colon) of broiler chicken at four significant production steps (bleeding, scalding, defeathering, evisceration) along the slaughter line on a qualitative and semiquantitative level.

| Sample collection and further processing
All samples were obtained from one slaughterhouse on 13 non-consecutive days over 20 months. Whole carcasses from a total of 19 flocks were collected at different stages along the process chain. For each flock, samples were collected at different processing points.
Samples of at least 7 different flocks were collected at 4 different days for each processing point. In total, 32 carcasses after bleeding, 32 after scalding, and 29 after defeathering were investigated. For all flocks included in the study the already separated intestines after evisceration (n = 64) were also collected. Additionally, 4 carcasses of two different flocks were collected at each of the four sampling points (n = 32) to compare the Arcobacter spp. load in colonic tissue versus content. Furthermore, environmental samples were taken in the slaughterhouse, including scalding water (n = 6) and plucking fingers (n = 6), which were dismantled from the defeathering machine to be examined. All samples were stored at 4°C until further processing within 24 hr. After removing the intestinal tracts from the carcasses, the intestinal content of duodenum (pars descendens duodeni and pars ascendens duodeni), jejunum (without the distal part of the ileum), both caeca, and colon were aseptically collected.
Concerning the samples where colonic tissue and colonic content were analyzed in parallel, the colonic section was opened longitudinally after collecting the content and rinsed with aseptic water. The tissue with the intestinal mucosa was incised and homogenized in Arcobacter enrichment broth.
To process the scalding water, 50 ml was centrifuged for 10 min at 5000 × g. The supernatant was discarded, and 45 ml of Arcobacter enrichment broth was added to the remaining sediment of 5 ml and thoroughly mixed for 2 min. The plucking fingers were processed by adding them to a tube containing 10 ml Arcobacter enrichment broth, mixing them thoroughly for 2 min. The remaining liquid was incubated as described above.
For semiquantitative detection, serial 10-fold dilutions of the initial dilutions were prepared in Arcobacter enrichment broth. After incubation, 10 µl of each dilution was transferred to Arcobacter selective agar plates composed of Arcobacter enrichment broth (without lysed horse blood) and 1.2% Agar Bacteriological No. 1 (Oxoid) and further incubated.
For the qualitative detection of Arcobacter spp., the homogenates were incubated as described above before plating 50 µl on Arcobacter selective agar plates. Suspicious colonies (diameter of 1 mm, beige to transparent) were picked and subcultured on Mueller-Hinton agar plates (Oxoid) supplemented with 5% defibrinated sheep blood (MHB) and incubated. To confirm the presence of Arcobacter spp., DNA was isolated from any dilution that showed bacterial growth on the semiquantitative plates and of suspicious colonies grown on MHB plates from the qualitative samples using the Chelex method as described previously (Karadas et al., 2013) and subsequently identified by mPCR according to Houf, Tutenel, De Zutter, Van Hoof, and Vandamme (2000). In brief, the total volume of 25 µl PCR reaction mixture included 2 µl DNA template, 2.5 µl of 10× PCR buffer (Qiagen), 2.8 mM MgCl 2 , 0.2 mM of each dNTP (Thermo Fisher Scientific), 0.75 U of Taq DNA polymerase (Qiagen), 1 µM of each primer ARCO, BUTZ, CRY1, CRY2, and 0.5 µM of the primer SKIR (Primers listed in Table 1). After an initial denaturation step (94°C for 2 min), the PCR involved 32 cycles of denaturation (94°C for 45 s), primer annealing (61°C for 45 s), chain extension (72°C for 30 s), and a final elongation step (72°C for 5 min). After gel electrophoresis in 3% agarose gel, PCR products were visualized with GR Green (Excellgen) under UV light.
The semiquantitative load of Arcobacter spp. was determined according to the MPN-method based on ISO/TS10272-3:2010/ Cor.1:2011 for detection and quantification of Campylobacter spp., by adjusting media, incubation time and temperature, and by reducing the sample weight to 1 g. The confirmation of the presence of Arcobacter spp. was verified by the mPCR above mentioned (Houf et al., 2000).
For all qualitatively detected isolates identified as Arcobacter spp. by mPCR, species verification was performed by rpoB sequencing according to (Korczak et al., 2006). In brief, the total volume of 50 µl PCR reaction mixture included 4 µl DNA template, 5 µl 10× PCR buffer, 2.5 mM MgCl 2 , 0.2 mM of each dNTP, 1 U of Taq DNA polymerase and 0.4 µM of the primers CamRpoB-L and RpoB-R (Primers listed in Table 1). After an initial denaturation step (95°C for 3 min), PCR involved 35 cycles of denaturation (94°C for 30 s), primer annealing (54°C for 30 s), chain extension (72°C for 30 s), and a final elongation step (72°C for 7 min). After gel electrophoresis in 3% agarose gel, PCR products were visualized with GR Green under UV light.
The amplified DNA was purified by the GeneJET PCR Purification Kit (Thermo Fisher Scientific) and sequenced with CamRpoB-L and RpoB-R primers by Eurofins Genomics (Ebersberg). The sequences were analyzed using BioNumerics version 7.1 (Applied Maths) and standard Nucleotide BLAST (NCBI) and compared with the NCBI nucleotide collection database to verify the Arcobacter species.

| RE SULTS
Arcobacter spp. were detected in each of the 19 examined flocks at least at one of the four sampling sites. However, after the first two production steps, the prevalences of Arcobacter spp. in the four sections of the intestinal tract were low. After bleeding, Arcobacter spp. were only detected in the intestinal content of 6% (2/32) of the colonic samples whereas no Arcobacter spp. were detected in any intestinal content after scalding ( Table 2). The Arcobacter load in both colonic content samples after bleeding was relatively low (2.3 and 230 MPN/g). In contrast, after defeathering and evisceration, the prevalences and Arcobacter loads increased. After defeathering, Arcobacter spp. were detected in the intestinal content of 28% (8/29) of the duodenal, 21% (6/29) of the jejunal, 3% (1/29) of the cecal, and 55% (16/29) of the colonic samples ( Table 2). The highest Arcobacter load was determined in the colonic content (up to 24,000 MPN/g),   tected at all four investigated processing steps. Therefore, the cecum is not likely to be the reservoir of Arcobacter spp. inside the chicken intestines, which has also been suggested by Ho et al. (2008). In the colonic section, the highest prevalences and bacterial loads were determined, which both declined toward the more orad situated duodenal and jejunal intestinal sections. After defeathering, higher prevalences and bacterial loads of Arcobacter spp. were determined in the colonic tissue compared with the colonic content. However, after the evisceration process, the prevalences and median bacterial loads did not differ between the colonic tissue and content. These data suggest that

| D ISCUSS I ON
Arcobacter spp. are possibly localized in the tissue/mucus layer of the intestinal tract, as has already been reported for Campylobacter (Awad, Hess, & Hess, 2018). However, we cannot completely rule out that the Arcobacter spp. isolated from colonic tissue samples partially derives from residues of the colonic content. Further samples need to be examined to affirm colonic tissue/mucosa as an Arcobacter reservoir in chicken. Assuming Arcobacter spp. reside inside the chicken-but were not detected in our study in contents of the intestinal tracts and colonic tissues after bleeding and scalding-it can be speculated that certain processes (not yet determined) after scalding allowed detection of Arcobacter spp. at a later stage in the slaughter line (i.e., after defeathering). As no other major processing steps are between both sampling points and the time between both sampling stages is relatively short, the physical forces during the defeathering process should be considered in future investigations.
Furthermore, one has to consider possible cross-contamination during the defeathering process. Several authors assumed that Arcobacter spp. can establish and proliferate in the environment inside the poultry slaughterhouse and several sources of Arcobacter contamination have been claimed, for example process water, slaughterhouse environment, and the defeathering machine itself (Ferreira, Fraqueza, Queiroz, Domingues, & Oleastro, 2013;Gude et al., 2005;Houf, De Zutter, Verbeke, Van Hoof, & Vandamme, 2003;Kjeldgaard, Jorgensen, & Ingmer, 2009). The few Arcobacter-positive samples detected in the colonic content after bleeding and scalding with only low Arcobacter loads might rather indicate cross-contamination of the colonic content originating during the transport to the slaughterhouse or the environment within (Corry & Atabay, 2001;Ho et al., 2008;Van Driessche & Houf, 2007). Arcobacter spp. are described as being able to survive the scalding temperatures, and therefore, the scalding water is suspected to contribute to the contamination within the slaughterhouse (Ho et al., 2008;Van Driessche & Houf, 2008). In our study, Arcobacter spp. were also detected in the scalding water samples. However, the Arcobacter loads determined in the scalding water (up to 230 MPN/ml) could not explain the higher Arcobacter loads in the intestinal contents determined after defeathering (up to 24,000 MPN/g). Therefore, an additional source with a consistent flow of high loads of Arcobacter spp. during the defeathering process, such as compartments of the defeathering machine (Plucker) or process water, h to be assumed.
Several characteristics of Arcobacter spp. might facilitate their ability to establish in the environment inside the slaughterhouse.
Detection of Arcobacter in several water bodies has been reported (Hsu & Lee, 2015). Also, Arcobacter spp. can form biofilms under conditions most likely to be found in poultry slaughterhouses, and chicken meat juice supports the growth of A. butzleri at cold temperatures (Ferreira et al., 2013;Kjeldgaard et al., 2009). In combination with the ability to withstand certain disinfection substances that are generally used in slaughterhouses (Rasmussen, Kjeldgaard, Christensen, & Ingmer, 2013), Arcobacter spp. possibly reside and multiply in the environment inside the slaughterhouse. Contamination of plucking fingers by scalding water or contact with cross-contaminated carcasses has been reported (Allen, Tinker, Hinton, & Wathes, 2003;Houf et al., 2003

| CON CLUS ION
In this study, we were not able to define the source of Arcobacter contamination in the poultry slaughterhouse. However, our data contributed to the understanding of Arcobacter transmission inside the poultry slaughterhouse and pointed out that cross-contamination processes seem to be of multifactorial origin and the defeathering procedure might be of importance for Arcobacter cross-contamination in the production line. Since Arcobacter spp. are commonly present on poultry products, the transmission routes of Arcobacter spp. into the poultry slaughterhouse need to be further investigated.

ACK N OWLED G M ENTS
This work was in parts supported by the German Federal Ministries of Education and Research (BMBF) by grant 01Kl1712 (Arco-Path).
Open access funding enabled and organized by Projekt DEAL.
[Correction added on 28 September 2020, after first online publication: Projekt DEAL funding statement has been added.]

CO N FLI C T O F I NTE R E S T S
None declared. Thomas Alter: Conceptualization (lead); Funding acquisition (lead);

AUTH O R CO NTR I B UTI O N
Writing-original draft (supporting); Writing-review & editing (equal).

E TH I C S S TATEM ENT
None required.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analyzed during this study are included in this published article.