Genetic characterization of Helicobacter pylori vac A and cag A genes in Thai gastro-duodenal and hepatobiliary patients

Introduction: H. pylori has been detected in patients with hepatobiliary diseases. It is currently unclear whether the H. pylori detected in hepatobiliary patients are genetically similar to those in gastro-duodenal patients. The aim of this study was to determine H. pylori vacA and cagA genotypes in Thai patients with gastro-duodenal and hepatobiliary diseases. Methodology: H. pylori DNA was extracted from samples from gastric biopsies of gastro-duodenal patients (n=100) and from bile samples of hepatobiliary patients (n=80). The vacA and cagA genotypes were performed via polymerase chain reaction (PCR) followed by DNA sequencing. Results: The vacA m1 was found in Thai hepatobiliary patients (90%) at a higher rate compared with gastro-duodenal patients (50%).The combined vacA s1a+c/m1 were mostly found in Thai gastro-duodenal and hepatobiliary patients. The cagA gene was detected in 94% of patients with gastro-duodenal diseases compared with 28.8% in those with hepatobiliary diseases (p<0.05). On the other hand, the Western type cagA was more prominent among hepatobiliary patients (100%) than gastro-duodenal patients (57.4%), and this type was grouped into same cluster with Thai gastro-duodenal patients via phylogenetic analysis. Conclusions: Based on vacA and cagA analysis, we conclude that infection with H. pylori in gastro-duodenal and hepatobiliary patients may be caused by the different H. pylori strains.


Introduction
Helicobacter pylori (H.pylori) is a Gram-negative, non-spore-forming and microaerophilic bacteria.This bacterium is the primary agent in gastro-duodenal diseases, including chronic gastritis and peptic ulcers, which lead to gastric adenocarcinoma [1].H. pylori was recently proposed to be involved in the diseases of the extra-gastroduodenal tract, particularly the hepatobiliary tract [2].Cholangiocarcinoma (CCA) is a primary cancer of the biliary epithelium, and the incidence of this cancer is endemic to Northeast Thailand [3].A report by Boonyanugomol et al. revealed that H. pylori was detected in specimens of hepatobiliary patients (especially in CCA patients) and that this bacteria may be a factor in CCA due to its acceleration of biliary inflammation and proliferation [4].
Vacuolating cytotoxin A (vacA) and cytotoxinassociated gene A (cagA) are the major virulence factors of H. pylori contributing to disease severity [5].VacA is produced in all strains of H. pylori, which induces cytoplasmic vacuoles and increases membrane permeability and apoptotic cell death and leads to gastric epithelial cell damage [5].Variations in H. pylori vacA at signal (s) and middle (m) regions have been studied.Allelic variations in the s-region and mregion have been classified into the s1 (s1a, s1b, and s1c) or s2 and m1 or m2 subtypes, respectively [6].The cagA gene is an important virulence factor of H. pylori and is located at the end of the cag pathogenicity island (cagPAI) [7].Infections with cagA-positive strains are related to gastric ulcers, duodenal ulcers, and gastric cancer.However, the prevalence of cagA-positive strains depends on geographical distribution and the associated human population [8].The 3' of the cagA geneis a variable region encoding the EPIYA motifs (Glu-Pro-Ile-Tyr-Ala).Due to differences in these motifs, the cagA gene can be classified into Western or East Asian types [9].Several studies revealed that infections with East Asian cagA-positive strains are associated with severe inflammation and gastric cancer development compared with Western cagA-positive strains [9][10][11].Therefore, infection with different vacA and cagA genotypes may lead to differing clinical outcomes.
We have previously reported on the prevalence of H. pylori vacA and cagA genotypes in hepatobiliary patients [12].However, data on the genetic association between H. pylori colonized in hepatobiliary and gastro-duodenal tracts remain lacking because those genes have not been completely compared between the two disease groups.The aims of this study were to characterize the genotypes of the vacA and cagA genes of H. pylori detected in Thai gastro-duodenal patients and to compare these genotypes with Thai hepatobiliary patients based on our previously published data [12].The genotype patterns of vacA and cagA in these two groups were analyzed.Elucidating the genetic relationship between H. pylori genotypes detected in gastro-duodenal and hepatobiliary patients may contribute to a better understanding of the pathogenesis of H. pylori in hepatobiliary diseases.

Specimens
Gastric biopsies were performed on dyspeptic patients who underwent gastroendoscopy and examination at the endoscopic unit, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University.Dyspeptic patients were characterized as having nonulcer diseases (NUD), peptic ulcer diseases (PUD), or gastric cancer (GC).In this study, a total of 100 gastric samples were positive for the rapid urease test (RUT), taken to indicate H. pylori infection.
Based on previously published data, bile samples were taken from patients with different hepatobiliary disorders (CCA and cholelithiasis) at the surgical unit of the Srinagarind Hospital, Faculty of Medicine, Khon Kaen University [12].A total of 80 bile samples were found to be positive for the H. pylori ureA gene via polymerase chain reaction (PCR).Patients with history of previous H. pylori treatment were excluded from this study.
Approval for this study was obtained from the human ethics committee of Khon Kaen University (HE581271), and informed consent was obtained from all patients.

DNA extraction
H. pylori genomic DNA was extracted from bile and gastric tissue samples.DNA extractions from bile samples were previously described by Boonyanugomol et al. [4].The DNA samples from biopsies positive for RUT were extracted using a genomic DNA extraction and purification kit (Gentra System, Inc., Minneapolis, USA), according to the manufacturer's instructions.
The PCR was performed in a total volume of 50 µL containing 500 ng of genomic DNA from the sample, 1X PCR buffer (RBC, Bioscience, Taipei, Taiwan), 200 µM of dNTP (Gibco, Gaithersburg, USA), 0.2 µM of each primer (with the exception of 0.5 µM for cagA sequencing), and 1.5 U Taq polymerase (RBC, Bioscience, Taipei, Taiwan).PCR amplification was conducted using a GeneAmp PCR system 9600 (Perkin Elmer, Waltham, USA).The amplification was programmed for 35 cycles.Finally, PCR amplicons were examined via agarose gel electrophoresis and visualized under a UV illuminator after staining with ethidium bromide.

DNA sequence analysis
To confirm the subtypes of the cagA gene, amplicons were purified and sequenced using a DYEnamic ET Dye Terminator Cycle Sequencing Kit from the MegaBACE 1000 DNA Analysis System (GE Healthcare, Life Sciences, Little Chalfont, UK).The cagA types were investigated by translating the amino acid sequence usingthe BMC Search Launcher program (Baylor College of Medicine, Houston, Texas, USA).

Phylogenetic analysis of cagA
The genetic relationship was confirmed by conducting a phylogenetic analysis of a partial sequence of the Western-type cagA detected in patients with gastro-duodenaland hepatobiliary disease and comparing this type with other cagA DNA sequences of gastro-duodenal patients from various geographic locations.To clarify the relationship between cagA detected in gastro-duodenal and hepatobiliary patients in Thailand, a phylogenetic analysis was performed.The non-Thai cagA sequences of gastro-duodenal patients from various regions were also taken from GenBank.The accession numbers used in the tree construction came from the following regions: Japan (AB190943, AB190946, AB190954, AB190956, and AB190949), Australia (AF202973, AF083352, and AF282853), the United States (AB015416, DQ067454, and AB015414), and Europe (AY330639 and AY330644).The partial sequences of cagA were aligned using the CLUSTAL W program.The phylogenetic tree was constructed using a maximum parsimony protocol, and reliability was confirmed via reconstruction with 1,000 replicates (MEGA software version 4.0, Center for Evolutionary Medicine and Informatics, Tempe, USA).

Statistical analysis
Data were subjected to analysis via Fisher's exact test (two-tailed test).A p value of less than 0.05 was regarded as statistically significant.

Results
Patients with gastro-duodenal and hepatobiliary diseases were evaluated for possible relationship to age, gender, and clinical manifestations (Table 2).Males and patients 41-60 years of age were found most frequently in the two patient groups.

Prevalence of H. pylori vacA genotypes in gastroduodenal and hepatobiliary diseases
The genotype of the signal region (s) in vacA genes was type vacA s1 (100%) in both gastro-duodenal and hepatobiliary diseases.A detailed analysis of the vacA s1 type showed that the vacA s1a+c genotype was more predominant in these two groups.However, the vacA s2 and s1b types were absent.The variation in the middle region (m) of the vacA m1 and m2 alleles showed a nearly similar proportion of gastro-duodenal diseases (50% and 45%, respectively).In contrast, the vacA m1 allele (90%) was found more frequently than the m2 allele (10%) in hepatobiliary diseases.When comparing these two groups, the vacA m1 was significantly associated with hepatobiliary diseases, as shown in Table 3 (p<0.05).
Additionally, the gene combination of the vacA signal and the middle regions is shown in Table 4.The distribution of vacA s1a+c/m1 subtype was predominantly found compared to other subtypes in hepatobiliary diseases similar to gastro-duodenal disease.

Prevalence of H. pylori cagA genotypes in gastroduodenal and hepatobiliary diseases
The cagA gene was found in 94% of gastroduodenal patients, but in only 28.8% of hepatobiliary patients.Comparison among the two patient groups revealed that the cagA gene was significantly predominant in the gastro-duodenal patients.
According to PCR and DNA sequencing analysis, Western and East Asian cagA types were found in patients with gastro-duodenal diseases.The Western type (57.4%) was found significantly more frequently than the East Asian type (33%) (p<0.05).
In hepatobiliary patients, all cagA-positive samples were of the Western type (100%), especially in CCA patients.Our results revealed that the Western-type cagA gene was predominant in patients with hepatobiliary diseases.

Phylogenetic analysis of H. pylori cagA Western type detected in gastro-duodenal and hepatobiliary diseases
Due to the predominance of the cagA Western type in hepatobiliary diseases, DNA sequencing of the amplified product was performed.The DNA sequences of the cagA Western type in hepatobiliary diseases were studied for genetic relationships with random samples of the cagA Western type in Thai gastro-duodenal patients (n=20).In this study, the constructed tree showed two general clusters, cluster A (Thai cluster) and cluster B (mixed cluster) (Figure 1).Cluster A was the Thai cluster, which contained cagA sequences in Thai hepatobiliary and Thai gastro-duodenal patients.The interesting findings in cluster A included a genetic relationship between H. pylori detected in the gastro-   duodenal and hepatobiliary tract with a unique genetic presence in the Thai population.However, cluster B was a mixed group that included Asian/European/American/Australian sequences.Findings in cluster B showed that some cagA sequences of Thai patients (gastro-duodenal and hepatobiliary diseases) had genetic similarity with cagA sequences from Japan and the Western geographic region, which diverged from the Thai cluster.

Discussion
Helicobacter pylori is widely known to be the primary agent of gastro-duodenal diseases, including gastric cancer.The gastro-intestinal tract connects with the hepatobiliary tract via the common bile duct at the duodenal papilla; thus, the bacterium in the stomach could pass to the duodenum through the biliary system.Helicobacter species can colonize diverse regions of the digestive system, including the gallbladder, intrahepatic bile duct, and liver [18,19].This knowledge led to investigations into a potential relationship between Helicobacter species and hepatobiliary diseases.The relationship between H. pylori and liver diseases has been assessed in humans, and several reports have demonstrated the prevalence of H. pylori in hepatobiliary diseases [20][21][22].Previous study reported that H. pylori or related similar H. pylori could be detected in liver tissues from patients with cholestatic liver diseases [21].In agreement with our previous report, we demonstrated that H. pylori was found significantly in patients with CCA compared to a cholelithiasis and control groups [4].Additionally, our previous study found that the presence of H. pylori in CCA patients was associated with biliary cell inflammation and proliferation, as determined via inflammatory grading and Ki67-labelling immunohistochemistry techniques, respectively [4].Together, our previous results indicated that H. pylori was strongly associated with CCA through the acceleration of biliary cell inflammation and the disturbance of biliary cell kinetics.We subsequently reported on the genotypes of vacA and cagA genes in hepatobiliary patients [12].However, it remains unclear whether the gastro-duodenal and hepatobiliary patients suffering from H. pylori infection had the same strain.
In the present study, molecular characterization was carried out to determine the genetic characteristics of H. pylori detected in gastro-duodenal diseases and compared with our previous data of hepatobiliary diseases [12].VacA is the major protein toxin secreted by H. pylori, and this toxin is present in all H. pylori strains [5].Multiple effects of VacA on infected cells have been investigated, including the induction of cell vacuolation, inflammatory responses, apoptosis, and cell damage [8].In the present study, we detected the vacA s1 type in both gastro-duodenal and hepatobiliary patients; however, vacA s2 was not found.Within the vacA s1 subtype, the vacA s1a+c subtype appeared to be present at a high proportion in hepatobiliary and gastro-duodenal patients (although with no significant differences in gastro-duodenal patients).Previous studies showed that vacA s1a and s1c, but not s1b and s2 strains, were found in a Southeast Asian population [23].Conversely, vacA s1b and s2 appeared in East Asian and African Arab populations [15,24,25].Our findings are in agreement with a previous report that showed that mixed vacA s1a+c was more frequently found in patients with various gastric diseases in North Thailand [26].We assumed that the prevalence of vacA subtypes varies among different ethnic and geographic populations.The presence of mixed H. pylori vacA s1a and s1c (vacA s1a+c) may result from genetic recombination, leading to mutation inside a strain [27].We suggested that mixed subtypes of vacA (vacA s1a+c) increase the chance of infection with more pylori pathogenic strains [26].
The vacA m-region is involved in cell binding, and vacA m1 and m2 genotypes ultimately affect the specificity to various cellular receptors [28].The glycoprotein receptor protein tyrosine phosphatase (RPTPs) is a receptor for the VacA toxin [29].Our findings showed that H. pylori vacA m1 strains were more frequent in hepatobiliary diseases (p<0.05),whereas vacA m1 and m2 strains were more frequent in gastro-duodenal patients in the same proportion.The difference in vacA m genotypes in hepatobiliary and gastro-duodenal patients may depend on different organs, specific binding to individual cell types, and differing post-translational modifications of RPTPs in each cell type [29].However, gene combination of vacA s-and m-region revealed that vacA s1a+c/m1 was mostly found in both gastro-duodenal and hepatobiliary diseases.The activity of vacA s1/m1 causes more severe inflammation and gastric cell damage than the other subtypes, and an especially low and absent activity has been observed in the s2/m2 subtype [6].We concluded that vacAs1/m1 might be involved with severity of hepatobiliary diseases, especially in CCA.
H. pylori cagA plays an important role in gastric mucosal inflammation.The tyrosine phosphorylation of CagA is widely known to occur at the EPIYA motif [7].
The genetic variation of the EPIYA motif has been investigated, and four distinct segments were determined: EPIYA-A, EPIYA-B, EPIYA-C, and EPIYA-D.The East Asian and Western CagA are characterized by the presence of EPIYA-ABD and EPIYA-ABC, respectively.The co-culture of AGS cells with East Asian and Western CagA strains showed that East Asian CagA strains induced significantly more IL-8 production than Western CagA strains [9].However, H. pylori expressing the CagA Western type were also related to the severity of inflammation and gastric cancer development [9].Most H. pylori samples in gastro-duodenal patients of our study carried the cagA gene (94%), whereas the cagA gene was found in 28.8% of hepatobiliary patients (p< 0.05).The difference in cagA prevalence between the two patient groups may be due to other virulence-associated genes that are associated with the presence of the cagA gene.One suggestion by a previous report stated that bacterial binding to the fucosylated blood group antigen Lewis b (Le b ) is coded by the babA2 gene and is associated with the presence of the cagA gene [30].Thus, differences in the prevalence of cagA between the two patient groups may depend on other factors such as the level of Le b expression in biliary and gastric cells.In addition to the cagA gene, another H. pylori virulence factor may play a role in hepatobiliary diseases and should be further studied, such as our previous report investigating the effects of H. pylori γglutamyltranspeptidase (GGT) in biliary cells [31].
Thailand is located in Southeast Asia, an area that is at the cultural crossroads of East and South Asia and is host to the migration of several geographic populations.The distribution of East Asian CagA strains is predominant in East Asian countries including Japan, China, and Korea.In contrast, Western CagA strains are widely distributed among European, South and Central Asian, North and South American, and African populations [32].In the present study, the Western CagA type was more common than the East Asian CagA type in Thai gastro-duodenal patients.We showed that the Western CagA type was predominant in hepatobiliary diseases and gastro-duodenal diseases.One possible explanation of the Western cagA type predominant in the Thai population might be dependent on geography and ethnic groups.We found that gastroduodenal patients in the Isan ethnic group (who are considered to live in Northeast Thailand) were significantly associated with the Western CagA type (data not shown), which was consistent with a previous epidemiological report that revealed that the Isan ethnic group in Northeast Thailand had a high percentage of CCA [33].Additionally, relationships between host and pathogen genetic variations and other environmental factors play important roles in the variability of infection outcomes [34].Therefore, the host genetics of local ethnic populations in Northeast Thailand may also select for H. pylori Western cagA strain infection.
To clarify the genetic relationships between the H. pylori in the two patient groups, Western cagA sequences in hepatobiliary diseases were analyzed and compared with randomized Western cagA sequences in gastro-duodenal diseases (via phylogenetic analysis).Our constructed phylogenetic tree included two groups.Most of the Western cagA sequences (74%) in Thai gastro-duodenal and Thai hepatobiliary patients were grouped in same cluster, designated the Thai cluster, which diverged from the mixed cluster, which contained Asian/European/American/Australian sequences.However, 26% of the Western cagA sequences of Thai gastro-duodenal patients and Thai hepatobiliary patients were classified as being in the mixed cluster.Several geographic populations live together in Southeast Asia; thus, migration from various regions may be linked to diverse strains of H. pylori, as previously described by Breurecet al. in 2011 [35].These two patient groups were classified into the same group (the Thai population cluster) because they both exhibited the similarity of Western-type cagA sequences.However, we suggest that other H. pylori genes should be used in classifying these two patient groups by phylogenetic analysis.
The high percentage of CCA in Northeast Thailand has been clearly demonstrated to be associated with liver fluke (Opisthorchis viverrini) infection.However, other factors in bile duct carcinogenesis should not be neglected, such as bacterial infection.Our previously published data showed that cagPAI especially cagA of H. pylori can induce inflammation and disturbance of cell kinetic on hepatobiliary cell lines [31,36,37].Therefore, we suggest that, in addition to O. viverrini infection, H. pylori may be a co-factor of the O. viverrini pathogenesis in Thai CCA patients as previously described by our colleagues, showing the significant relationship between O. viverrini and H. pylori [38].However, the association and role of H. pylori in CCA should be further investigated.
Our findings showed that the cagA gene was significantly more predominant in gastro-duodenal patients (94%) than in hepatobiliary patients (28.8%) (p<0.05).Moreover, high prevalence of vacA m1 (90%) and Western-type cagA (100%) were detected in hepatobiliary patients, and low prevalence was detectedin gastro-duodenal patients.Therefore, these findings indicate that H. pylori infection in gastroduodenal and hepatobiliary patients may be caused by the different H. pylori strains.

Conclusions
Our study is the first to show the genetic characterization between H. pylori in diseases of the gastro-duodenal and hepatobiliary tracts.Infection with H. pylori in gastro-duodenal patients and hepatobiliary patients may be caused by different H. pylori strains.However, the pathogenesis of this bacteria in the hepatobiliary tract should be further elucidated.

Table 2 .
Distribution of 180 patients with different clinical characteristics based on age and gender.
ND: no determination due to no information.

Table 3 .
Virulence-associated gene status of H. pylori with differing clinical outcomes in hepatobiliary and gastro-duodenal patients.

Table 4 .
Gene combination of H. pylori vacA s-and m-regions in hepatobiliary and gastro-duodenal patients.