Helicobacter pylori babA2 Positivity Predicts Risk of Gastric Cancer in Ardabil, a Very High-Risk Area in Iran

Background: Ardabil, a Northwestern province of Iran, was found to have the highest rate of gastric cancer (GC) in the country (ASRs = 51.8/100,000 for males and 24.9/100,000 for females) and one of the highest gastric cardia cancer rates in the world. The aim of the present study was to assess the associations of the cagA and babA2 status of Helicobacter pylori with GC in the Ardabil population. Materials and Methods: A total of 103 patients with non-atrophic gastritis (56) and GC (47), who underwent endoscopy at the Imam Khomeini Hospital in Ardabil, were assessed. The status of 16S rDNA , cagA and babA2 genes was determined using PCR and histopathological assessment was performed. Results: The following genotypic frequency was observed: cagA + (50.6%), cagA– (49.4%), babA2+ (26.5%), babA2– (73.5%) cagA +/ babA2+ (19.3%), cagA– / babA2+ (7.2%), cagA +/ babA2– (31.3%), cagA– / babA2– (42.2%). Although the frequency of the cagA +, cagA +/ babA2+ and cagA– / babA2+ genotypes in patients with GC (55.6%, 25.9%, and 14.8%, respectively) was higher than in those with NAG (48.2%, 16.1%, and 3.6%, respectively), the difference did not reach significance. In contrast, the presence of the babA2 gene (40.7% vs 19.6%) significantly increased the risk of GC; the age-sex-adjusted odds ratio (95% confidence interval) was 5.068 (1.506-17.058; P= 0.009), by multiple logistic regression. Conclusions: It is proposed that the H. pylori babA2 positivity might be considered as an important determinant of GC risk in Ardabil.


Introduction
Helicobacter pylori is a microaerophilic gram negative spiral bacterium that is able to localize in the stomach mucosa for at least tens of thousands of year. Several epidemiological studies have shown that the H. pylori infection causes a variety of gastrointestinal symptoms such as chronic gastritis and adenocarcinoma (Suerbaum and Michetti, 2002). Although the incidence and mortality of gastric cancer (GC) have declined in recent decades, it is the third-ranking cause of cancer which leads to death and the fifth most common malignant disease in the world (Ferlay et al., 2015). Histologically, GC divided into the intestinal-and the diffuse-type carcinomas. Previous studies have suggested that the interaction between host and environmental factors are jointly responsible for development of GC (Leung et al., 2006).
Stomach cancer is very common in Iran (Alireza et al., 2005a;Alireza et al., 2005b). Ardabil, a volcanic and mountainous land in Northwestern province of Iran that is located in Caspian Sea littoral, includes the highest rate of GC within the country (ASR of 51.8/100,000 in men and 24.9/100,000 in women) (Sadjadi et al., 2003;Babaei et al., 2010). In this area, more than one-third of the GC occurs in the cardia site of the stomach which constitutes

Helicobacter pylori babA2 Positivity Predicts Risk of Gastric Cancer in Ardabil, a Very High-Risk Area in Iran
Esmat Abdi 1 , Saeid Latifi-Navid 1 *, Abbas Yazdanbod 2 , Saber Zahri 1 only 5-10% of the entire stomach, and the incidence for cardia sub site was 26.4 and 8.6 for males and females, respectively Babaei et al., 2009). H. pylori infection has been measured as 89% and 69% in Ardabil province and Iran, respectively (Sadjadi et al., 2003;Malekzadeh et al., 2004;Nouraie et al., 2009). We have previously indicated that all the Iranian H. pylori strains, including those with Ardabil signature, represent the European ancestry, influenced by extensive genetic exchange with neighboring countries. However, they have well preserved ethnic and geographic signatures within the country (Latifi-Navid et al., 2010). Ardabil isolates showed approximately equal contribution from the two European ancestral sources, Ancestral Europe1 and Ancestral Europe2 (AE1 and AE2) that were found in purer form in Central Asia (AE1) and North East Africa (AE2).
CagA is one of the putative virulence factors of H. pylori, which is encoded by the cytotoxin-associated gene A (cagA) that is located in the cag pathogenicity island (cag PAI) and directly translocated into host gastric epithelial cells by a type-IV secretion system (T4SS) after bacterial attachment (Mattar et al., 2007;Ohnishi et al., 2008). It promotes the tumorigenesis of GC by interfering with multiple signaling pathways (Yong et al., 2015).
Approximately, the 50 to 60% of H. pylori strains harbor cagA (Abasiyanik et al., 2002;Basso et al., 2008), whose expression induces the production of IL-8 and mucosal inflammation in cultured gastric cells (Chang et al., 2015;Ferreira et al., 2015). Several studies have been reported that the individuals harboring cagA-positive strains suffer from a high risk for gastric mucosal atrophy and GC, compared to those with strains that lack the cag PAI (Sozzi et al., 2005;Tan et al., 2006). orf17 and particularly cagL, which are other genotypes of cag PAI, significantly increased the risk of peptic ulceration (PU), but not GC. No significant association was found between the cagH and cagG genotypes and the risk of both the PU and the GC (Raei et al., 2015). H. pylori strains may be classified into two types: type I and type II, based on possessing the cag PAI (Bagheri et al., 2013). The East Asian subtype of the cagA gene (A, B, D) strongly linked to severe gastritis and GC, in comparison with western subtype (A, B,C) (Argent et al., 2008).
Another important virulence factor is babA2 (blood group antigen binding adhesion) that encoded by the babA2 gene. BabA is a 78-KDa outer membrane protein (OMP) that binds to lewis b antigens and ABO antigen on the surfaces of gastric epithelial cells (Gerhard et al., 1999). Two babA allelic types identified (babA1 and babA2); however, only the babA2 gene is functionally active. The presence of the 10-bp insertion is the cause of the difference between babA1 and babA2 (Ilver et al., 1998). Expression of the babA2 gene has also been associated with the risk of GC (Lee et al., 2006).
Since the diagnosis of GC in the earlier stages of infection is difficult; therefore, eradication of the H. pylori infection might be the main way of cancer conservation. Therefore, it is necessary to determine which strains of H. pylori lead to GC. H. pylori-specific genotypes could determine the clinical consequences and contribute to understanding of the cause of high tendency for GC (Figueiredo et al., 2002;Peek et al., 2010). The aim of the present study was therefore to determine the H. pylori cagA and babA2 status and their associations with GC in Ardabil population.

Participants of the study
From February 2012 to January 2014, a total number of 103 consecutive adult patients who underwent endoscopy at the Imam Khomeini Hospital in Ardabil, Northwestern Iran, participated in this study. All patients were born and grew up in the rural or urban areas of Ardabil province. Fresh gastric mucosal biopsy specimens were obtained from each patient and classified by endoscopic and histopathological examinations. The study population consisted of patients with none-atrophic gastritis (NAG) and GC. None of the patients had received none steroidal anti-inflammatory drugs or antibiotics within the previous three months, and all the participants signed a written informed consent.

Histopathological assessment
The biopsy samples were fixed in 10% formalin and embedded in paraffin, and were stained by hematoxylineosin, Alcian blue-periodic acid Shiff (pH 2.5) and Giemsa for light microscopy, and histopathological evaluations were also applied according to the updated Sydney classification system (Dixon et al., 1996).

Diagnosis of H. pylori infection
During endoscopy, of each patient four biopsies (two from the antrum and two from the corpus) were taken for histological examination and DNA extraction. The Biopsy specimens were frozen at -80ºC until processing. Detection of H. pylori was performed by using a specific primer set (HP1/HP2) of H. pylori 16S rDNA, selected from some previous published works that produced the PCR fragments of 519-bp (Table 1).

DNA extraction and genotyping of H. pylori
DNA was extracted from the biopsy specimens by using Genomic DNA purification kit (DNGTM-Plus, CinnaGen Co., Iran) according to the manufacturer's recommendations, and stored at -20ºC. After DNA extraction, polymerase chain reaction (PCR) was performed in a volume of 30 µL reaction volume containing 50 ng of genomic DNA, 3 µL of 10X PCR buffer (CinnaGen, Iran), 1 mM MgCl2, 200 µM of each dNTP (CinnaGen Co., Iran), 0.5 µM of each specific primer and 2 U of Taq DNA polymerase (CinnaGen Co., Iran). Amplification was also performed under the following conditions: for cagA: 35 cycles of 40 sec at 94°C, 1 min at 50°C (C-terminus), 55°C (middle region), or 56°C (N-terminus) and 1 min at 72°C, and for babA2: 35 cycles  of 40 sec at 94°C, 1 min at 55°C and 1 min at 72°C. PCR products were visualized by electrophoresis in 1% agarose gel and examined under UV illumination. Primers used for PCR assays of 16S rDNA, cagA and babA2 genes are also listed in Table 1. As controls, amplified fragments of each gene from seven isolates were purified and sequenced with both forward and reverse primers by using BigDye technology on an ABI3700XL DNA sequencer (Applied Biosystems). The BLAST program (http://www.ncbi.nlm. nih.gov) was used to match the nucleotide sequences with the published sequences in GenBank.

Data analysis
Chi-square (χ 2 ) or Fisher's exact tests were used to assess the associations between the cagA and babA2 status and GC, whether independently or in combination. Patients with NAG were considered as controls. Odds ratios (ORs) and the corresponding 95% confidence intervals (CIs) were also calculated. The Forward Stepwise LR (Likelihood Ratio) multiple logistic regression analysis was used for each gene individually, with adjustment for the potential confounders, including a threshold age of 40 years and sex. All the P-values were two-sided and P<0.05 was considered statistically significant. Data were collected and analyzed by using SPSS software (SPSS Science, Chicago, IL) version 19.0.

Discussion
Ardabil, a Northwestern province of Iran, is one of the areas with the highest GC incidence rate in the world. In the present study, we therefore investigated the association of the cagA and babA2 status with the risk of GC in Ardabil population. Through analyzing the data by means of logistic regression, the results showed no significant association between the cagA status of H. pylori and GC risk; however, the importance of CagA EPIYA polymorphisms has not been confirmed in Ardabil population yet; therefore, determining the associations of the CagA EPIYA polymorphisms with GC is important in this high-risk area. Recent studies have shown that the multiple repeats of H. pylori CagA EPIYA-C phosphorylation sites predict risk of gastric ulcer (GU) in Iran (Honarmand-Jahromy et al., 2015). The GU and GC have etiologic factors in common, which are likely to be caused by the H. pylori-induced atrophic gastritis (Hansson et al., 1996). We found that the cagA gene is highly prevalent in Ardabil patients (50.6%), but it could not be considered as an important risk factor for GC. However, in agreement with other studies (Dabiri et al., 2009;Joober et al., 2012), a high frequency of the cagA gene was also observed in GC patients in our study (55.6%). Several studies on Tehran population, of Iran, indicated no significant association between the presence of the cagA gene and GC risk (Jafari et al., 2008;Dabrii et al., 2010;Bagheri et al., 2013). In contrast, study conducted by Cittelly et al. in Colombians showed that the frequency of the cagA gene was significantly higher in GC patients (80%) than in control groups (51.4%) (P<0.01). They proposed that the presence of the cagA gene significantly increased the risk of GC (Cittelly et al., 2002). Also, study performed by Cavalcant et al. from a high-risk area of GC in Brazil demonstrated that the presence of the cagA gene particularly associated with the risk of GC (OR=10.36) (Cavalcante et al., 2012). A recent follow-up study from a high-risk area of GC in Spain showed that infection with cagA+/vacA s1/m1 strains was related to the progression of gastric precancerous lesions (OR = 4.80) compared with those infected with strains that harbored the cagA-/vacAs2/m2 genotype (González et al., 2011). In other study which was performed in a low-risk area of Okinawa of Japan, a strong relationship was found between cagA gene and GC risk in comparison with control groups, in a multiple logistic regression (OR = 6.68) (Matsunari et al., 2012). The prevalence of the cagA gene in Western and Eastern countries was reported to be 60% and 95%, respectively, however, only in Western countries, a significant association with GC was found (Kim et al., 2001;González et al., 2011). A meta-analysis conducted by Zhao et al. also showed that the H. pylori CagA antibody of the serum could not be useful for detecting GC in East Asian countries, with the highest incidence rates of GC (Zhao et al., 2014). This is in agreement with the results of the present study which was conducted in a genotype level and restricted to Ardabil, with the highest GC incidence rate in Iran and different ethnic group characteristics, Azeri with an Altaic language family. Interestingly, it has also been shown that the H. pylori CagA-negative non-virulent isolates may be a potential risk factor of pancreatic cancer (Chen et al., 2015).
Our findings demonstrated that the prevalence of the subjects harboring babA2-strains was significantly higher than the subjects with babA2+ strains. The babA2+ H pylori isolates were present in approximately 26.5% of the study population. The patients infected with the H. pylori strains which carry babA2 gene showed a higher risk of developing GC which was about 5.068 fold, by the multiple logistic regression; therefore, it could be considered an important virulence factor in Ardabil population. We have previously shown no significant difference in the distribution of the babA2 and cagA genes among H. pylori isolates from high (34.2% and 75.6%, respectively) and low (44.4% and 63.9%, respectively) incidence GC areas (P>0.05) (Latifi-Navid et al., 2013). In Iran, the frequency of babA2 genotype was 40.6% (Abadi et al., 2013;Latifi-Navid et al., 2013), which was strongly associated with GC in comparison with the control groups (P=0.0004) (Abadi et al., 2013). Similarly, study conducted by Gerhard et al. showed that frequency of the babA2 gene in GC patients (78%) was higher than those with gastritis (51%), with significant difference (Gerhard et al., 1999). Erzin et al. showed that the frequency of the babA2 gene in GC patients (87.9%) was higher than those with NAG (23.3%) (P=0) (Erzin et al., 2006). Furthermore, in Germany, Turkey, or northern Portugal, babA2 expression was closely associated with the severity of gastrointestinal disease (Gerhard et al., 1999;Azevedo et al., 2008;Erzin et al., 2008), while in Portuguese and Thai populations, the presence of the babA2 gene was not associated with GC risk (Gatti et al., 2006;Chomvarin et al., 2008). These results reflect the fact that the associations of H. pylori virulence genes/ alleles with GC risk have been widely varied according to geographic variation/ or ethnic origin.
We proposed that the H. pylori babA2 positivity might be considered as an important determinant of GC risk in Ardabil. Hence, this study provided us with valuable information which might be helpful to reduce the incidence of GC in high-risk areas, particularly Ardabil. We propose that more investigations with a sufficient number of samples are necessary in order to better depicting the bacterial genotype diversity and the role of environmental and host factors and their associations with GC in this the very high-incidence area.