A molecular survey of Campylobacter jejuni and Campylobacter coli virulence and diversity.

BACKGROUND: The aim of this study was to determine the prevalence of virulence-associated genes and enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) analysis of Campylobacter spp. isolated from children with diarrhea in Iran.
METHODS: A total of 200 stool specimens were obtained from children under 5 years during July 2012 to July 2013. Detection of C. jejuni and C. coli was performed by standard biochemical and molecular methods. The presence of virulence-associated genes and genetic diversity of isolates was examined using PCR and ERIC-PCR analyses.
RESULTS: A total of 12 (6%) Campylobacter spp. were isolated from patients including 10 (4.5%) C. jejuni and 2 (1.5%) C.coli. The flaA, cadF and ciaB genes were present in 100% of isolates, while no plasmid of virB11 gene was present in their genome. The prevalence of invasion-associated marker was 100% among C. coli and was not detected in C. jejuni isolates. The distribution of both pldA and the genes associated with cytolethal distending toxin (CDT) was 58.3% in C. jejuni isolates. Seven distinct ERIC-PCR profiles were distinguished in three clusters using ERIC-PCR analysis. Genotyping analysis showed a relative correlation with geographic location of patients and virulence gene content of isolates.
CONCLUSION: To our knowledge, this is the first molecular survey of Campylobacter spp. in Iran concerning genotyping and virulence gene content of both C. jejuni and C. coli. ERIC-PCR revealed appropriate discriminatory power for clustering C. jejuni isolates with identical virulence gene content. However, more studies are needed to clearly understand the pathogenesis properties of specific genotypes.

INTRODUCTION

Campylobacter spp. especially C. jejuni and C. coli are considered as potential etiological agents that caused many undiagnosed cases of acute diarrhea in children in developing countries including Iran [1-3]. The purpose of our experiments was to determine the rate and molecular survey of C. jejuni and C. coli virulence and also their diversity that leads to the practical applications to elucidate Campylobacter colonization and the control of this organism. Recurrent exposure to these organisms might raise level of specific immunity correlated with age in developing countries. Therefore, children younger than 5 years of age are mainly affected by these organisms [4].

Above organisms are fastidious and need nutrient-rich-based medium and microaerobic atmosphere [5]. This reason may be the main cause that Campylobacter spp. are not applied in routine diagnostic programs of clinical laboratories in most developing countries. The pathogenicity of Campylobacter species is dependent on their ability to bind to the human intestinal cells and CadF protein. This protein, which is encoded by cadF gene, is responsible for Campylobacter binding to extracellular matrix of human intestinal cells [6]. Another gene, flaA, encodes a flagella protein which mediates motility, colonization, and invasion of gastrointestinal tract and it is essential for establishing human infection.

The ciaB (an invasion protein), virB11 (the IV secretory system), and pldA (an outer membrane phospholipase A) genes encode proteins associated with increased bacterial invasion on cultured epithelial cells; however, their exact roles in invasion have remained to be elucidated [7]. Cytolethal distending toxin (CDT) is encoded by three linked genes, including cdtA, cdtB, and cdtC. In epidemiology of infectious diseases, bacterial typing is of great value in source tracking studies. To analyze the genetic relatedness of C. jejuni, several molecular typing methods based on PCR have been developed. enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) has been shown high discriminatory power, good legibility, and ease of use in most epidemiological investigations.

In this study, we aimed to determine the prevalence of virulence-associated genes and ERIC-PCR analysis of C. jejuni and C. coli isolated from children with diarrhea. CDT is one of the most characterized virulence factors in Campylobacter spp. Pathogenesis induces cell cycle arrest in G2 phase and promotes DNA damage together with apoptotic death in human monocytic cells; therefore, its presence is supposed to be associated with the severity of the disease.

MATERIALS AND METHODS

The study was started after obtaining ethical approval from Research and Publication ethics office of Tarbiat Modares University (Tehran, Iran). A total of 200 stool specimens were collected from children with acute diarrhea attended to two major Children’s Hospital Medical Center in Tehran from July 2012 to July 2013. All children under five years of age were included in this study. Children with persistent diarrhea or previous treatment with antibiotics in the last 5 days were excluded from the study. Acute diarrhea was defined as diarrhea which takes about 14 days or less. Demographic data were collected by a co-worker resident in hospital. All samples were transported to the laboratory in a modified Cary-Blair transport medium (5.00 g/L sodium chloride, 1.50 g/L sodium thioglycollate, 1.10 g/L disodium phosphate, and 0.09 g/L calcium chloride, pH 8.4 ± 0.2 at 25ºC( with reduced agar content (1.6 g/L). Identification of C. jejuni and C. coli was performed by the standard culture, Gram staining, and conventional biochemical tests and confirmed by molecular methods [8, 9].

Samples were incubated at 42°C for 48-72 h in microaerophilic conditions onto modified charcoal cefoperazone deoxycholate agar medium (10 g/L meat extract, 10 g/L peptone, 5 g/L sodium chloride, 4 g/L bacteriological charcoal, 3 g/L casein hydrolysate, 1 g/L sodium deoxycholate, 0.25 g/L iron (II) sulfate, 0.25 g/L sodium pyruvate, and 15 g/L agar, pH 7.4 ± 0.2 at 25º) plus campylobacter CCDA selective supplement (cefoperazone 3,200 mg/L). DNA templates were extracted by boiling method [10]. Confirmation of Campylobacter spp. was performed by PCR amplification of cadF gene. A duple-PCR was applied for simultaneous detection of hipO and asp genes, specific to C. jejuni and C. coli, respectively [11]. PCR was performed in a 25-μl reaction mixture, containing 10 ng DNA template, 2.5 μl PCR buffer 10×, 200 μM dNTP, 5 mM MgCl2, 0.1 μM each primer, 1 unit of Taq DNA polymerase, and deionized water. The primer sequences and their designation are shown in Table 1. The C. jejuni ATCC 29428 and C. coli ATCC 43478 were used as reference strains.

Table 1

Primers, PCR conditions, and respective references

Primers	Sequence (5 ' → 3')	Target	PCR condition			Amplicon (bp)	References
			Denaturin	Annealin	Extension
cadFUcadFR	TTGAAGGTAATTTAGATATGCTAATACCTAAAGTTGAAAC	cadF	94οC, 30 s	43οC, 30 s	72οC, 30 s	400	[11]
hipOUhipOR	GAAGAGGTTTGGGTGGTGAGCTAGCTTCGCATAATAACTTG	hipO	94οC, 30 s	53οC, 30 s	72οC, 30 s	735	[11]
aspUaspR	GGTATGATTTCTACAAAGCGAGATAAAAGACTATCGTCGCGTG	asp	94οC, 30 s	53οC, 30 s	72οC, 30 s	500	[11]
flaAUflaAR	TTTCGTATTAACACAAATGGTGCCTGTAGTAATCTTAAAACATTTTG	flaA	94οC, 45 s	46οC, 45 s	72οC, 60 s	1743	this study
cdtjAUcdtjAR	AGGACTTGAACCTACTTTTCAGGTGGAGTAGTTAAAAACC	Cj-cdtA	94οC, 30 s	55οC, 30 s	72οC, 30 s	631	[31]
cdtj BUcdtjBR	ATCTTTTAACCTTGCTTTTGCGCAAGCATTAAAATCGCAGC	Cj-cdtB	94οC, 30 s	56οC, 30 s	72οC, 30 s	714	[31]
cdtjCUcdtjCR	TTTAGCCTTTGCAACTCCTAAAGGGGTAGCAGCTGTTAA	Cj-cdtC	94οC, 30 s	55οC, 30 s	72οC, 30 s	524	[31]
cdtCAUcdtCAR	ATTGCCAAGGCTAAAATCTCGATAAAGTCTCCAAAACTGC	Cc-cdtA	94οC, 30 s	55οC, 30 s	72οC, 30 s	329	[31]
cdtCBUcdtCBR	TTTAATGTATTATTTGCCGCTCATTGCCTATGCGTATG	Cc-cdtB	94οC, 30 s	56οC, 30 s	72οC, 30 s	413	[31]
cdtCCUcdtCCR	TAGGGATATGCACGCAAAAGGCTTAATACAGTTACGATAG	Cc-cdtC	94οC, 30 s	55οC, 30 s	72οC, 30 s	313	[31]
ciaBUciaBR	TGCTAGCCATACTTAGGCGTTTTTTGATAATAGCGGACAATTTGAAA	ciaB	94οC, 30 s	54οC, 30 s	72οC, 30 s	610	this study
pldAUpldAR	AAGCTTATGCGTTTTTTATAAGGCTTTCTCC	PldA	94οC, 30 s	46οC, 30 s	72οC, 30 s	913	[32]
iamAUiamAR	GCGCAAAATATTATCACCCTTCACGACTACTATGCGG	iam	94οC, 30 s	47οC, 30 s	72οC, 30 s	518	[33]
virB11UvirB11R	GAACAGGAAGTGGAAAAACTAGCTCCCGCATTGGGCTATATG	virB11	94οC, 30 s	52οC, 30 s	72οC, 120 s	708	[33]
ERICF ERICR	ATGTAAGCTCCTGGGGATTCAAAGTAAGTGACTGGGTGAGCG	ERIC	94οC, 30 s	52οC, 30 s	72οC, 300 s	variable	[34]

The presence of virulence/invasion-associated genes, including invasion-associated marker (iam), pldA, and ciaB, (responsible for Campylobacter invasion and attachment), virB11 (involved in Campylobacter virulence), and CDT were investigated using specific primers which specifically amplify within the coding region of each gene. The distribution of flaA gene (responsible for Campylobacter attachment) was examined by primers specifically designed according to the flaA locus sequence of C. jejuni strain (GenBank accession no. AF050186.1). Due to the species-allele-specification of CDT sequence, two separate primer pairs were used which were selected according to cdt locus sequence of C. jejuni and C. coli standard strains.

ERIC-PCR assay was performed according to the method introduced by Versalovic and colleagues [12]. The primer sequences, their designation, and amplification conditions are depicted in Table 1. ERIC- PCR amplification reactions were performed in a 25-μl reaction mixture, containing 10 ng genomic DNA, 2.5 μl reaction buffer 10×, 200 μM dNTP, 5 mM MgCl2, 0.2 μM each primer, and 1 unit Taq DNA polymerase. The reaction was placed in a DNA thermal cycler (Mini Bio-Rad, USA). ERIC-PCR patterns were analyzed based on the Dice similarity coefficient using GelClust software [13].

RESULTS

The study population was made up of 200 children with acute diarrhea, including 110 (55%) male and 90 (45%) female with the mean age of 27.4 months (2.3 years). Among 200 stool samples, Campylobacter spp., including 10 (4.5%) C. jejuni and 2 (1.5%) C. coli were isolated from 12 (6%) samples.

The cadF gene was positive in 12 (100%) suspected cultures, and an amplification band of 400 bp was obtained for cadF+ isolates indicative of Campylobacter spp. Duplex-PCR for species identification indicated the presence of 10 (4.5%) C. jejuni and 2 (1.5%) C. coli isolates, respectively (Fig. 1).

Fig. 1

PCR and Duplex- PCRfor species identification of Campylobacter spp. and C. jejuni/ C.coli isolates, respectively; Lane1, cadF gene; lane 2, asp gene; lane 3, hipO gene, and M, 1 kb DNA size marker

All of our Campylobacter isolates harbored flaA and ciaB genes. CDT encoding genes (cdtA, cdtB, and cdtC) involved in CDT production, and pldA was found in 7 (58.3%) of C. jejuni isolates corresponding to amplification bands of 631, 714, 524, and 913 bp, respectively. However, no CDT encoding genes were found among C. coli isolates. The plasmid virB11 gene was also absent among our Campylobacter spp. isolates. The prevalence of iam was 100% among C. coli, while no amplification was obtained for C. jejuni isolates (Table 2).

Table 2

Prevalence of virulence and toxin genes in C. jejuni and C. coli isolates under study

Species (No.)	No. of PCR positive (%)
	cad F	hip O	asp	vir B11	cia B	iam	pld A	fla A	cdt A	cdtB	cdtC
C. jejuni (10)	10 (100)	10 (100)	0 (0)	0 (0)	10 (100)	0 (0)	7 (58.3)	10 (100)	7 (58.3)	7 (58.3)	7 (58.3)
C. coli (2)	2 (100)	0 (0)	2 (100)	0 (0)	2 (100)	2 (100)	0 (0)	2 (100)	0 (0)	0 (0)	0 (0)
Total (12)	12 (100)	10 (83.4)	2 (16.6)	0 (0)	12 (100)	2 (16.6)	7 (58.3)	12 (100)	7 (58.3)	7 (58.3)	7 (58.3)

In total, 10 out of 12 Campylobacter isolates under study produced 7 distinguishable banding profiles by ERIC-PCR genomic fingerprinting, which were corresponded to 10 C. jejuni isolates. Dendrogram of ERIC-PCR was created with UPGMA algorithm, which revealed 3 major clusters with 4, 4, and 2 members. No obvious banding pattern was obtained for C. coli isolates even after multiple attempts (Fig. 2).

Fig. 2

Dendrogram generated from the ERIC-PCR profiles of C. jejuni isolates from humans in relation to their Profile of virulence-associated genes. Strains 1, 2, 3, and 4 share 80% overall similarity, but strains 7, 8, 9, and 10 share 70% overall similarity.   positive   Negative

DISCUSSION

Campylobacteriosis is one of the most common bacterial causes of food-borne illness and leading cause of bacterial diarrheal disease in the world [9]. In the present study, the prevalence of Campylobacter spp. among diarrheic children was about 6%. In a few previous studies performed in Iran, the prevalence of Campylobacter spp. was reported to be 8%, 10.8%, and 8.7% in 2007, 2009, and 2011, respectively [3, 14, 15]. This observation shows the trend of Campylobacter infections to be almost unchanged through years in this country.

The prevalence of Campylobacter species among diarrheic children was reported to be 5.4% in Turkey [14], 7% in India [15] and 11.1% in Lebanon [16]. As reported by WHO [17], the incidence of Campylobacteriosis was 9.3 per 1,000 people in Europe America [17]. All isolates in the present study, either C. Jejuni or C .coli, harbored cadF, flaA, and ciaB genes. These genes are essential virulence factors involved in Campylobacter adhesion and colonization to human intestinal epithelial cells during human infection. The ubiquitous existence of the highly conserved cadF gene in 100% of Campylobacter spp. was previously reported by Konkel and coworkers [18] and was subsequently used by other investigators for successful detection of Campylobacter spp. [19, 20].

The prevalence of virulence-associated genes (ciaB and flaA) was reported to be 80-100% in different studies concerning Campylobacter spp. infections in children with moderate to severe diarrhea [21-23]. Similarly, these genes were also detected within all of our isolates. The ciaB and flaA are both involved in maximal invasion of human intestinal cells. This result can justify broadly existence of these gens in clinical Campylobacter spp.

The virB11 gene was not found in any of Campylobacter isolates under study. This finding is in agreement with the studies by other investigators who did not find virB11 gene among Campylobacter isolates of children from Brazil and Bangladesh [23, 24]. However, a few other studies indicated the prevalence of virB11 to be 10.7-22.7% among clinical isolates [21, 25]. This finding emphasizes the low prevalence of type IV secretion system apparatus in

Campylobacter spp., which can probably be due to plasmid basis of the gene.

In this study, the iam gene that codes for iam was detected in 100% of C. coli isolates, while none of C. jejuni isolates harbored the gene. Similar results have been also reported regarding the high prevalence of iam gene among C. coli as well as its absence or low distribution among C. jejuni isolates of children with diarrhea in Brazil [26]. However, several studies demonstrated no substantial difference in its occurrence among the two species. This result shows that iam frequency is controversial [21, 27, 28].

The distribution of both pldA and the genes associated with CDT production (cdtA, cdtB, and cdtC) was 58.3% in C. jejuni isolates, while none of the genes were detected among C. coli isolates. The CDT toxin induces cell cycle arrest in G2 phase and promotes DNA damage; therefore, its presence is supposed to be associated with the severity of the disease in C. jejuni. However, variations which may occur within cdt gene sequences and may affect their detection through amplification methods should not be ignored during interpretation of negative results in C. coli. Moreover, a direct correlation was observed in this study between detection of pldA gene and the presence of white and red blood cells in stool of patients, which may be due to contribution of pldA gene product in pathological changes in intestinal epithelium. In agreement with our results, Rizal and colleagues [21] reported the presence of cdt and pldA genes among 50% and 55% of C. jejuni isolates, respectively, while none of the genes were detected within their C. coli isolates. Seven distinct ERIC-PCR profiles were distinguished from 10 C. jejuni isolates which were located in three clusters by ERIC-PCR. Cluster analysis showed that all isolates (no. 4) within cluster I were isolated from patients who lived in the south of Tehran. Except one isolate, all others within cluster II (no. 3) have also isolated from the patients in a similar geographical location in Tehran (east and center). Moreover, all of the isolates in cluster I and II revealed identical virulence gene content. Two isolates within cluster III were isolated from west of Tehran, but no clear correlation could be determined between ERIC-PCR profile and virulence genes content of isolates within this cluster. Sahilah and colleagues [29] reported that no specific relationship could be extracted between ERIC-PCR analysis and virulence gene content of their C. jejuni isolates. However, Wardak et al. [30] showed that ERIC-PCR could clearly divide C. jejuni and C. coli into two clusters.

ERIC-PCR was unable to type our C. coli isolates which raises the question that to how extent is the typeability power of ERIC-PCR for C. coli strains. This emphasizes that more studies are needed to clearly understand the ability and role of this typing method in C. coli epidemiological studies. However, it is noteworthy that ERIC-PCR analysis has been proved as a well-documented molecular tool in epidemio-logical studies of C. jejuni strains.

To our knowledge, this is the first molecular survey of C. jejuni and C. Coli genotypes in Iran. Nevertheless, further studies are needed to more clearly understand the correlation between virulence-associated genes and specific genotypes of C. jejuni and C. coli clinical isolates.