A PCR-RFLP assay to detect and type cytolethal distending toxin (cdt) genes in Campylobacter hyointestinalis.

Campylobacter hyointestinalis is considered as an emerging zoonotic pathogen. We have recently identified two types of cytolethal distending toxin (cdt) gene in C. hyointestinalis and designated them as Chcdt-I and Chcdt-II. In this study, we developed a PCR-restriction fragment length polymorphism (RFLP) assay that can differentiate Chcdt-I from Chcdt-II. When the PCR-RFLP assay was applied to 17 other Campylobacter strains and 25 non-Campylobacter strains, PCR products were not obtained irrespective of their cdt gene-possession, indicating that the specificity of the PCR-RFLP assay was 100%. In contrast, when the PCR-RFLP assay was applied to 35 C. hyointestinalis strains including 23 analyzed in the previous study and 12 newly isolated from pigs and bovines, all of them showed the presence of cdt genes. Furthermore, a restriction digest by EcoT14-I revealed that 29 strains contained both Chcdt-I and Chcdt-II and 6 strains contained only Chcdt-II, showing 100% sensitivity. Unexpectedly, however, PCR products obtained from 7 C. hyointestinalis strains were not completely digested by EcoT14-I. Nucleotide sequence analysis revealed that the undigested PCR product was homologous to cdtB but not to Chcdt-IB or Chcdt-IIB, indicating the presence of another cdt gene-variant. Then, we further digested the PCR products with DdeI in addition to EcoT14-I, showing that all three cdt genes, including a possible new Chcdt variant, could be clearly differentiated. Thus, the PCR-RFLP assay developed in this study is a valuable tool for evaluating the Chcdt gene-profile of bacteria.

Campylobacter hyointestinalis was first identified from a pig with proliferative enteritis [6, 8] and has recently been isolated from both diseased and healthy animals and raw milk [4, 7, 9, 15]. Furthermore, this species was isolated from human diarrheal stool samples [5, 14, 16]. At present, C. hyointestinalis is considered to be an emerging zoonotic pathogen, and its clinical importance and pathogenic mechanism are under evaluation.

Cytolethal distending toxin (CDT), which is a genotoxin capable of directly damaging DNA in target cells, is considered to be a possible virulence factor of various Gram-negative bacteria including C. hyointestinalis. CDT consists of three subunits, CdtA, CdtB and CdtC, which are encoded by the genes cdtA, cdtB and cdtC, respectively [3]. The CDT holotoxin causes G2/M cell cycle arrest, cytoplasmic distension and eventually cell death via apoptosis [3]. Although the pathogenic mechanism of CDT in vivo is not well understood, CDT produced by Campylobacter jejuni has been reported to cause panmural inflammation with mucosal denudation and necrosis affecting the jejunum, ileum and colon in mice [10].

Recently, we have identified two types of cdt gene clusters in C. hyointestinalis, namely, the cdt-I (Chcdt-I) and cdt-II (Chcdt-II) genes from C. hyointestinalis strains 022 [17] and ATCC35217T [12], respectively. The toxins encoded by these genes (ChCDT-I and ChCDT-II, respectively) both induced cell distention and death in HeLa cells. However, the homologies between these two ChCDTs were only 25.0, 56.0 and 24.8% in their CdtA, CdtB and CdtC subunits, respectively. Since there was a low homology between ChCDT-I and ChCDT-II, particularly regarding their CdtA and CdtC subunits, which are responsible for binding to receptor molecules, it is possible that their target cells might differ. Thus, ChCDT-I and ChCDT-II may have different pathogenic mechanisms in vivo. Therefore, it is important to analyze the distribution of these two cdt gene-variants in C. hyointestinalis to understand the difference in pathogenesis between ChCDT-I and ChCDT-II. In this study, we have developed a PCR-restriction fragment length polymorphism (PCR-RFLP) assay for the detection and differentiation of Chcdt-I and Chcdt-II in C. hyointestinalis. The specificity and sensitivity of the PCR-RFLP assay were evaluated, and the presence and types of cdt genes in 35 C. hyointestinalis strains, including 12 strains newly isolated from pigs and bovines, were successfully determined by the PCR-RFLP assay.

MATERIALS AND METHODS

Bacterial strains and growth media

Thirty-five strains of C. hyointestinalis, which are described in Table 1, were examined in this study. Among them, 23 strains were known to possess Chcdt-I and/or Chcdt-II [12], and the other 12 strains were newly isolated from pigs and bovines. A total of 42 bacterial strains other than C. hyointestinalis including 17 strains of 11 other Campylobacter species and 25 strains of 20 non-Campylobacter species were also included in this study (Table 1).

Table 1.

Bacterial strains used in this study and the distribution of Chcdt-I and Chcdt-II genes

Bacterial species	Strain	Origin	Chcdt-I	Chcdt-II	Reference		Bacterial species	Strain	Origin	cdt
C. hyointestinalis	022	Clinical	+	+	[12]		C. colia)	ATCC33559	Pig	Cccdt
	ATCC 35217T	Pig	-	+	[12]		C. colia)	ATCC43478	Marmoset	Cccdt
	1-1	Pig	+	+	[12]		C. colia)	CO1-017	Clinical	Cccdt
	10-1	Gorilla	+	+	[12]		C. consisus	ATCC33237	Human	-
	2003	Bovine	+	+	[12]		C. consisus	ATCC51562	Human	-
	2030	Bovine	+	+	[12]		C. curvus	ATCC35224	Human	-
	2032	Bovine	+	+	[12]		C. fetusa)	ATCC27374	Sheep	Cfcdt
	2033	Bovine	+	+	[12]		C. helveticus	ATCC51209	Cat	Chelcdt
	2034	Bovine	+	+	[12]		C. hominis	ATCCBAA-381	Human	-
	2035	Bovine	+	+	[12]		C. jejunia)	ATCC33560	Bovine	Cjcdt
	2037	Bovine	+	+	[12]		C. jejunia)	ATCC43432	Human	Cjcdt
	2038	Bovine	+	+	[12]		C. jejunia)	B86	Bovine	Cjcdt
	2039	Bovine	+	+	[12]		C. jejunia)	CO2-037	Clinical	Cjcdt
	3014	Bovine	+	+	[12]		C. laria)	JCM2530	Larus argintatus	Clcdt
	3158	Bovine	+	+	[12]		C. mucosalis	ATCC49352	Pig	-
	3197b)	Bovine	+	+	[12]		C. upsaliensisa)	ATCC43954	Dog	Cucdt
	3477	Bovine	+	+	[12]		C. ureolyticus	ATCC33387	Human	-
	3535b)	Bovine	+	+	[12]		Arcobacter butzuleri	ATCC49616	Clinical	-
	3839b)	Bovine	+	+	[12]		A. cryaerophilus	ATCC43157	Pig	-
	3857b)	Bovine	+	+	[12]		A. skirrowii	ATCC51132	Unknown	-
	87-4b)	Monkey	+	+	[12]		Citrobacter freundii	ATCC43864	Unknown	-
	94-6	Elephant	+	+	[12]		Enterobacter cloacae	NBPC 12935	Unknown	-
	SS	Pork	-	+	[12]		Escherichia colia)	ATCC11775	Human	-
	130206DCC11b)	Bovine	+	+	This study		Escherichia colia)	ATCC25922	Clinical	-
	130206DCC12	Bovine	+	+	This study		Escherichia colia)	GB1371	Clinical	Eccdt-I
	130325D2aC1	Bovine	+	+	This study		E. albertiia)	19982	Clinical	Eacdt
	141007D1C1	Bovine	+	+	This study		Haemophilus ducreyia)	ATCC700724	Human	Hdcdt
	141007D2C1b)	Bovine	+	+	This study		Helicobacter fennelliaea)	ATCC35684	Human	Hfcdt
	S1C-1	Pig	-	+	This study		H. hepaticusa)	ATCC51448	Mouse	Hhcdt
	S2 TDNEFB-1	Pig	+	+	This study		H. pylori	ATCC43504	Human	-
	S2CB-1	Pig	-	+	This study		H. pylori	ATCC43629	Human	-
	S3C-1	Pig	+	+	This study		Klebsiera oxytoca	ATCC8724	Unknown	-
	S4TVB-1	Pig	-	+	This study		K. pneumoniae	ATCC13883	Unknown	-
	S5C-1	Pig	+	+	This study		Providencia alcalifaciensa)	GTC02020	Unknown	Pacdt
	S5TDNPFB-1	Pig	-	+	This study		P. heimbachae	ATCC35613	Penguin	-
							P. rettgeri	ATCC29944	Human	-
							P. rustigianii	ATCC33673	Human	-
							P. stuartii	ATCC29914	Human	-
							Salmonella enterica	ATCCBAA-664	Unknown	-
							Vibrio cholerae	VcN16961	Unknown	-
							Vibrio cholerae	VcBT-AP32541	Unknown	-
							Vibrio cholerae	Vc129	Unknown	-

a) Bacterial species known to possess cdt genes, b) Undigested PCR product was observed from these strains, +: Indicates that the gene was identified by Chcdt-B gene-based PCR-RFLP assay, -: Indicates that the gene was not identified by Chcdt-B gene-based PCR-RFLP assay.

Campylobacter, Helicobacter and Arcobacter spp. were grown on blood agar [blood agar base No. 2 (Oxoid Ltd., Basingstoke, U.K.) supplemented with 5% (v/v) defibrinated horse blood (Nippon Bio Supp. Center, Tokyo, Japan)] under anaerobic conditions (10% CO2, 10% H2 and 80% N2) at 37°C for 2 days or more. Vibrio spp. were cultured using TCBS agar (NISSUI Pharmaceutical Co., Ltd., Tokyo, Japan), and other bacteria were grown in Luria–Bertani broth (Becton, Dickinson and Co., Franklin Lakes, NJ, U.S.A.) at 37°C overnight.

DNA preparation

Template DNAs for PCR were prepared by the boiling method as previously reported [1, 11]. Briefly, bacterial colonies grown on appropriate agar plates were suspended in 500 µl of TE buffer (10 mM Tris-HCl [pH 8.0] and 1 mM ethylenediaminetetraacetic acid). Bacterial suspensions were boiled for 10 min, kept on ice for 10 min and centrifuged at 20,000 ×g for 10 min. Then, the supernatants were collected and used as DNA templates for PCR. As positive controls for the PCR and RFLP assays, pET28a Chcdt-IB and pET28a Chcdt-IIB, which carry the Chcdt-IB and Chcdt-IIB genes of C. hyointestinalis strains 022 and ATCC35217T, respectively, were used [12, 17].

PCR amplification assay

PCR was performed with ChCdt-BF (5′-GCTACTTGGAATATGCAAGG-3′) and ChCdt-BR (5′-TGGTTCTCTATTRAAATCWCC-3′) primer set using an Applied Biosystems Veriti® Thermal Cycler (Thermo Fisher Scientific, Waltham, MA, U.S.A.). Each PCR mixture contained 0.5 and 0.15 µM of ChCdt-BF and ChCdt-RF primers, respectively, 1 µl of DNA template, 0.2 mM of each dNTP, 1 × rTaq DNA polymerase buffer and 0.5 U of rTaq DNA polymerase (Takara Bio Inc., Otsu, Japan) in a total volume of 20 µl. The samples were subjected to an initial denaturation step at 94°C for 3 min followed by 30 cycles of amplification, each cycle consisting of 94°C for 30 sec, 54°C for 30 sec and 72°C for 30 sec. A final extension step at 72°C for 3 min was included. PCR products were analyzed by electrophoresis using a 2% PrimeGel™ Agarose LE gel (Takara Bio Inc.), and bands were visualized with ultraviolet (UV) light after staining with ethidium bromide (1 µg/ml). Images were captured on a ChemiDoc system (Bio-Rad Laboratories, Inc., Hercules, CA, U.S.A.).

RFLP assay

PCR products (2 to 5 µl; 100 to 200 ng) were digested with either 4 U of EcoT14-I or 2 U of EcoT14-I and 2 U of DdeI (New England Biolabs Inc., Ipswich, MA, U.S.A.) under 1 ×H buffer or 1 ×K buffer (Takara Bio Inc.) in a final volume of 10 µl at 37°C overnight. Then, the digested PCR products were analyzed by 3% PrimeGel™ Agarose LE gel electrophoresis as described above.

Detection limit of the PCR-RFLP assay

The C. hyointestinalis strains, ATCC35217T, 022 and 3197, were cultured on blood agar under anaerobic conditions at 37°C for 2 days, and the colonies that grew were suspended in sterile phosphate-buffered saline. The density of each bacterial culture was adjusted to OD600=1.0 and then 10-fold serially diluted in phosphate-buffered saline, and DNA templates were prepared from 100 µl of each dilution by the boiling method as described above. Then, PCR assays were carried out with these DNA templates. Furthermore, to determine the viable bacterial count, 100 µl of each dilution was spread on blood agar plates in triplicate and cultured at 37°C for 2 days under anaerobic conditions, and then, the number of colonies on each plate was counted.

Colony hybridization assay

The distribution of Chcdt-IB and Chcdt-IIB in 12 newly isolated strains of C. hyointestinalis was examined by colony hybridization assay as described previously [11], with minor modifications. In brief, Chcdt-IB and Chcdt-IIB gene-probes were prepared by PCR with each primer set (Table S1) using C. hyointestinalis strains 022 (Chcdt-IB) and ATCC35217T(Chcdt-IIB) as DNA templates, and the obtained PCR products were purified from agarose gel using the Wizard SV® Gel and PCR clean-up system (Promega Corporation, Madison, WI, U.S.A.). Strains were grown on a nitrocellulose membrane (GE Healthcare U.K. Ltd., Buckinghamshire, U.K.) overlaid on blood agar plates under anaerobic conditions at 37°C overnight. Colonies were lysed, and DNA was denatured in situ by the alkaline lysis method followed by UV cross-linking with a UV crosslinker (CX-2000; UVP LLC, Upland, CA, U.S.A.). The processed nitrocellulose membranes were hybridized with each gene probe, and radioactivity was visualized using an FLA-7000 biomolecular imager (GE Healthcare U.K. Ltd.).

Sequencing of cdt genes and PCR products

To determine the entire nucleotide sequence of the cdt gene cluster in 12 newly obtained C. hyointestinalis strains (Table 1), PCR was performed with a combination of forward primers targeting upstream of cdtA or cdtB and reverse primers targeting downstream of cdtC (Table S2). Each PCR mixture contained 0.5 µM of each primer, 1 µl of DNA template, 0.2 mM of each dNTP, 1 ×Ex Taq DNA polymerase buffer and 1.0 U of Ex Taq DNA polymerase in a total volume of 20 µl. The samples were subjected to an initial denaturation step of 3 min at 94°C followed by 30 cycles of 94°C for 30 sec, 50°C for 30 sec and 72°C for 90 sec to 2 min, and then a final extension step at 72°C for 3 to 5 min. If a PCR product of the expected size was obtained, the PCR product was purified using a PCR clean-up system (QIAGEN GmbH, Hilden, Germany) for sequencing. The sequencing reactions were performed by the chain termination method with the BigDye® Terminator v1.1 cycle sequencing kit (Thermo Fisher Scientific). Nucleotide sequences were determined using an ABI PRISM 3130-Avant Genetic analyzer (Thermo Fisher Scientific). The sequence upstream of Chcdt-IB in C. hyointestinalis strains 130206DCC11, 130206DCC12, 130325D2aC1, 141007D1C1, S2TDNEFB-1, S3C-1 and S5C-1, and that of Chcdt-IIB in C. hyointestinalis strain 141007D2C1, were determined with the primers listed in Table S2 by genome walking as described previously [12].

Undigested PCR products (approximately 500 bp in length) were purified from agarose gel by using the Wizard® SV Gel and PCR clean-up system (Promega Corporation). The nucleotide sequences of undigested PCR products were determined as described above, analyzed using MEGA6, and compared with the sequences of Chcdt-IB and Chcdt-IIB.

Nucleotide sequence accession numbers

The cdt gene-sequences determined in this study were deposited to the DNA Data Bank of Japan (DDBJ) with the accession numbers LC175774 to LC175800.

RESULTS

Evaluation of ChcdtB amplification by PCR

By comparing published cdtB sequences of C. hyointestinalis (GenBank accession nos. AB218983 and AB373951) and those of other Campylobacter spp. (AB274783, AB274793, AB274802, AB872889 and AB872911), forward and reverse primers (ChCdt-BF and ChCdt-BR) were designed from the conserved regions of Chcdt-IB and Chcdt-IIB. To evaluate whether these primers can amplify Chcdt-IB and Chcdt-IIB, PCR was carried out using recombinant plasmids, pET28a carrying Chcdt-IB and pET28a carrying Chcdt-IIB, as positive controls. PCR of the positive controls, pET28a carrying Chcdt-IB and pET28a carrying Chcdt-IIB, gave the expected size of the PCR product (507 or 516 bp) in each case (Fig. S1). Sequence analysis further confirmed that these PCR products were amplified from Chcdt-IB and Chcdt-IIB gene-fragments (data not shown).

The sensitivity of the PCR assay was evaluated with the 35 C. hyointestinalis strains listed in Table 1. A PCR product of the expected size was obtained from 23 strains, which were previously reported to possess Chcdt-I and/or Chcdt-II (Table 1). Furthermore, a PCR product of the same size was also obtained from 12 newly isolated C. hyointestinalis strains in which the presence of Chcdt-I and/or Chcdt-II was confirmed by colony hybridization assay and sequence analysis as described below. Then, the specificity of the PCR assay was evaluated with a total of 42 strains including 17 strains of 11 other Campylobacter spp. and 25 strains of 20 non-Campylobacter spp. as listed in Table 1. Irrespective of their cdt gene-possession, no PCR product was obtained from those 42 strains.

Sequence analysis of cdt gene cluster in 12 C. hyointestinalis strains

To calculate the sensitivity of the PCR-RFLP assay developed in this study, we further evaluated the distribution of cdt genes in 12 C. hyointestinalis strains, which were newly isolated from pigs and bovines, by colony hybridization assay. The Chcdt-IB gene-probe was reacted with 8 strains (130206DCC11, 130206DCC12, 130325D2aC1, 141007D1C1, 141007D2C1, S2TDNEFB-1, S3C-1 and S5C-1), while the Chcdt-IIB gene-probe was reacted with all the strains (data not shown). Furthermore, the entire nucleotide sequences of these cdt gene clusters were determined as described in the Materials and Methods section. The sizes of individual cdt genes and the homology of each cdt gene are summarized in Table S3. These data indicated that the 12 C. hyointestinalis strains possessed Chcdt-I and/or Chcdt-II, except for 1 strain, 141007D2C1, in which cdtA was missing.

The detection limit of this assay was evaluated using C. hyointestinalis ATCC35217T(Chcdt-II), 022 (Chcdt-I and Chcdt-II) and 3197 (Chcdt-I, Chcdt-II and Chcdt homologue gene) strains. The detection limits of this PCR assay were determined to be 4.9 ± 1.8 × 102, 4.4 ± 0.47 × 102 and 4.9 ± 3.2 × 102 colony-forming units (CFU)/20 µl of reaction mixture for strains ATCC35217T, 022 and 3197, respectively.

A ChcdtB-based PCR-RFLP assay

EcoT14-I restriction enzyme was selected for the PCR-RFLP assay by comparing the sequences of Chcdt-IB and Chcdt-IIB. When the PCR products obtained from the positive controls were digested with EcoT14-I, the Chcdt-IB DNA fragment yielded two bands of 363 and 144 bp (Fig. 1A, lane 1), and in the case of Chcdt-IIB, the fragment sizes were 261 and 255 bp (Fig. 1A, lane 2). We then evaluated the utility of the PCR-RFLP assay with the PCR products obtained from 35 C. hyointestinalis strains (Table 1). Among the 35 strains, the cdt gene-possession profile of 23 strains was previously analyzed by colony hybridization assay and the same types of cdt were successfully identified by their RFLP patterns, even though both Chcdt-I and Chcdt-II were present (Fig. 1A, lane 4). When the PCR products obtained from 12 other strains were analyzed by the RFLP assay, their RFLP patterns indicated that all C. hyointestinalis strains contained Chcdt-IIB and 7 strains additionally possessed Chcdt-IB (Table 1). These data indicated that the PCR-RFLP assay could identify the cdt gene-profile. However, undigested PCR products were obtained from 7 strains (3197, 3535, 3839, 3857, 87–4, 130206DCC11 and 141007D2C1) among 35 strains when digested with EcoT14-I (Fig. 1A, lane 5).

Fig. 1.

(A) PCR-RFLP with EcoT14-I to evaluate the presence of the ChcdtB gene in C. hyointestinalis. PCR products digested with EcoT14-I were analyzed by electrophoresis on a 3% agarose gel. Undigested PCR product in lane 5 was extracted from the gel, purified, and subjected to a sequencing analysis. Undigested PCR product was identified as a possible ChcdtB variant gene. Lanes: M, 100-bp DNA ladder (Takara Bio Inc.); N, PCR product of C. hyointestinalis 3197 strain; 1, pET28a Chcdt-IB; 2, pET28a Chcdt-IIB; 3, C. hyointestinalis ATCC35217T strain; 4, C. hyointestinalis 022 strain; 5, C. hyointestinalis 3197 strain. (B) PCR-RFLP with EcoT14-I and DdeI to evaluate the presence of the ChcdtB gene in C. hyointestinalis. Lanes: M, 100-bp DNA ladder (Takara Bio Inc.); N, PCR product of C. hyointestinalis 3197 strain; 1, pET28a Chcdt-IB; 2, pET28a Chcdt-IIB; 3, C. hyointestinalis ATCC35217T strain; 4, C. hyointestinalis 022 strain; 5, C. hyointestinalis 3197 strain.

The undigested PCR products obtained from 7 strains were purified and sequenced. This analysis revealed that the undigested PCR products carried sequences of cdt genes with only 45–46% and 64–66% homologies to the corresponding regions of Chcdt-IB and Chcdt-IIB, respectively. These data indicated that the C. hyointestinalis strains might possess new variants of cdt genes.

To avoid the possibility of obtaining ambiguous results from the PCR-RFLP assay, we further digested the PCR products with DdeI in addition to EcoT14-I. When the PCR products obtained from positive controls were digested with these enzymes, the Chcdt-IB gene-fragment and Chcdt-IIB gene-fragment were theoretically digested to yield fragments of 363, 101 and 43 bp, and 261, 113, 70, 41 and 31 bp, respectively. The agarose gel separation result is shown in Fig. 1B. Furthermore, the PCR products of the cdtB gene-homologue were theoretically digested to yield fragments of 302, 72, 51, 27 and 24 bp (Fig. 1B). Then, when we evaluated the utility of this modified PCR-RFLP assay with the PCR products obtained from 35 C. hyointestinalis strains, Chcdt-IB, Chcdt-IIB and cdtB gene-homologue could be clearly distinguished from one another.

DISCUSSION

Although the clinical significance of C. hyointestinalis infection has not yet been clearly established, C. hyointestinalis is implicated as a pathogen of both humans and animals. We have previously reported that two types of the cdt genes, namely Chcdt-I and Chcdt-II, are present in C. hyointestinalis and encode biologically active CDTs [12, 17]. Therefore, CDT is a candidate virulence factor in C. hyointestinalis. To understand the importance of CDT in the pathogenesis of C. hyointestinalis, we attempted to establish a PCR-RFLP assay for the detection and differentiation of Chcdt-I and Chcdt-II in C. hyointestinalis. Since cdtB was previously demonstrated to be more conserved than cdtA and cdtC in C. jejuni, C. coli and C. fetus [2], various cdtB gene-based multiplex PCR and PCR-RFLP assays have been developed to detect and differentiate Campylobacter species [1, 11, 13]. In this study, we have developed a ChcdtB-based PCR-RFLP assay that can detect and differentiate between Chcdt-I and Chcdt-II.

The sensitivity and specificity of the PCR-RFLP assay were shown to be 100%. Furthermore, the PCR-RFLP assay could clearly identify the cdt gene-profile, even though C. hyointestinalis contained both Chcdt-I and Chcdt-II (Fig. 1A, lane 4). Kamei et al. have reported that Chcdt-II may be ubiquitously conserved in C. hyointestinalis [12]. In this study, we evaluated the presence of cdt genes using 12 C. hyointestinalis strains that were newly isolated from pigs and bovines. As expected, Chcdt-IIB was detected in all the C. hyointestinalis strains, while Chcdt-IB was also detected in 7 strains (Table 1). Furthermore, the presence of Chcdt-IIA and Chcdt-IIC in these 12 strains was analyzed by colony hybridization assay, and their nucleotide sequences were determined by sequence analysis (Table S3). Although Chcdt-IIA was not detected in C. hyointestinalis strain 141007D2C1, 11 other strains were demonstrated to possess Chcdt-IIA, Chcdt-IIB and Chcdt-IIC. It appears that Chcdt-IIB and Chcdt-IIC are ubiquitously present in C. hyointestinalis and may be suitable target genes for the detection and identification of this species.

The isolation of C. hyointestinalis is difficult if C. jejuni and C. coli are targeted for isolation because of its susceptibility to cephem antibiotics, which are included in the modified charcoal-cefoperazone-deoxycholate agar and Bolton broth that are normally used for the isolation of C. jejuni and C. coli. Therefore, a PCR-based method is required to detect this bacterium before initiating cultivation. The cdtB gene-based PCR assay developed in this study could detect C. hyointestinalis with 100% sensitivity and specificity (Table 1). The detection limit of the assay was determined to be approximately 102 CFU/20 µl of reaction mixture. These data indicate that the PCR assay developed in this study is useful for the detection of C. hyointestinalis.

However, PCR products obtained from some template DNAs of C. hyointestinalis strains remained undigested by EcoT14-I (Fig. 1A, lane 5). A sequence analysis of these PCR fragments showed that these strains carry the homologous cdtB gene sequence, indicating the detection of a new cdt gene-variant of C. hyointestinalis. Therefore, we further included DdeI digestion in the cdtB gene-based PCR-RFLP assay to distinguish not only Chcdt-IB and Chcdt-IIB but also a possible new cdtB gene-variant. The cdtB gene-based PCR-RFLP assay with EcoT14-I and DdeI could clearly distinguish three different cdtB genes as PCR products including Chcdt-IB and Chcdt-IIB and a possible new cdtB gene-variant in the 35 C. hyointestinalis strains listed in Table 1. Further analysis to characterize the cdt gene-variant is currently ongoing in our laboratory.

In conclusion, the ChcdtB gene-based PCR-RFLP assay developed in this study is useful to detect and differentiate not only Chcdt-I and Chcdt-II but also a possible new cdt gene-variant in C. hyointestinalis. The ubiquitous presence of Chcdt-IIB among the tested C. hyointestinalis strains also confirmed it to be an appropriate target gene for the identification of C. hyointestinalis. Further studies are required for the evaluation of the cdtB gene-based PCR-RFLP assay for the detection and differentiation of cdt genes present in C. hyointestinalis.