Is there a relationship between Helicobacter pylori vacA i1 or i2 alleles and development into peptic ulcer and gastric cancer? A meta-analysis study on an Iranian population

Abstract

Helicobacter pylori has several virulance factor i.e. VacA, CagA, BabA, SabA, AlpA, AlpB and etc. VacA has several polymorphic region in the nucleotide sequence such as s,m,i,d and, c. It has been suggested that each variation in these polymorphic region has been influenced the toxicity of VacA toxin. We performed a comprehensive meta-analysis to determine the main role of VacAi1/i2 in development into peptic ulcer and gastric cancer in an Iranian population.

To the Editor,

Helicobacter pylori is the most highlighted and prominent pathogenic bacterium of the human gastric submucosa; it colonizes the stomachs of half of the world's population [1]. Most of the populations living in developing countries are predominantly infected during childhood and the amount of infection in these areas is nearly 100% [2]. With the introduction of H. pylori as a first-class factor for gastric cancer (GC) by the International Agency for Research on Cancer in 1994, extensive attention was paid by gastroenterologists to this bacterium. Helicobacter pylori has been identified as the aetiological cause of chronic gastritis, peptic ulcer diseases (PUD), gastric adenocarcinoma and mucosa-associated lymphoid tissue lymphoma and some extra-gastrointestinal disorders [[3], [4], [5]]. A significant number of H. pylori-infected individuals remain asymptomatic and PUD is observed in only 15%–20% and GC in 1%–2% of individuals, so the question arises as to why the clinical symptoms are observed in only a small part of the population with the rest remaining as asymptomatic carriers. Based on available evidence, genetic characteristics and H. pylori strain virulence factors, host genome and its epigenetic events, and the environmental conditions can play decisive roles in the incidence of H. pylori-related gastrointestinal outcomes [1,[5], [6], [7]].

The most famous H. pylori virulence factors include vacA and cagA. The cagA (cytotoxin-associated gene A) encodes a toxin of 120–140 kDa, which is phosphorylated at the site of its own EPIYA motif (tyrosine residue) by Src kinase family proteins of the host and, subsequent to phosphorylation, cagA-P induces the ‘hummingbird phenotype’ and incidence of GC through alteration in cytoskeletal rearrangements, cell survival and proliferation, and changes in polarity [8]. According to the previous studies, nearly 100% of H. pylori isolates in the Japanese population produce and express cagA [9]. Iran is the fourth country in terms of GC prevalence in Asia, and nearly 69% of H. pylori isolates in Iranian patients harbour (containing, transporting) cagA [10]. The vacA (vacuolating cytotoxin gene A) is another virulence factor, which releases cytochrome c from the mitochondria by entering into the gastric epithelium, induces apoptosis, destroys tight junctions and intercellular connections, and produces vacuole in the host cells [11]. The vacA consists of five polymorphic regions including signal sequence (s1/s2), middle (m1/m2), intermediate (i1/i2), deletion (d1/d2) and c (c1/c2). Only about 50% of the H. pylori strains encode the vacA [12,13]. Also, the combination of each region determines the vacuolating formation strength. For example, the vacA s1m1 strains express a large amount of toxin with high vacuolating formation strength, whereas s1m2 strains produce a moderate amount of toxin and s2m2 strains are not, or are rarely, toxic [14]. There are controversial results regarding the relationship between i1/i2 segments [15,16]. In this study, for the first time in Iran, we conducted a meta-analysis study to assess the possible role and relationship of vacA i1/i2 in the Iranian population.

Potentially supported documents were collected by searching the PubMed, Scopus, Embase, Elton B. Stephens Company (EBSCO), Google scholar, Scientific Information Database (SID), Islamic Science Citation Database (ISC), and magiran databases. The documents were investigated using the keywords ‘Helicobacter pylori’, ‘Peptic Ulcer Disease’, ‘Gastric Cancer’, ‘VacA protein’ and ‘Iran’. Studies were collected up to April 2020, without language and time constraints, and all the original articles (including case–control, cohort, prospective, retrospective), letters to the editor and congress abstracts that were about determining the frequency of vacA i1/i2 genotypes in the Iranian PUD and GC patients were studied; information including the first author, publication year, location, distribution of gender and age, number of H. pylori strains and distribution of vacA i1/i2 genotypes are listed in Table 1.

Table 1.

Characteristics of included study

Ref.	Publication year	Location	Female/male; age	H. pylori strains	vacA i1/2 genotype
					Total		PU		GC
					i1	i2	i1	i2	i1	i2
Bakhti et al. [17]	2019	Ardabil	NA; 46.52	290	112	120	34	23	32	17
Mottaghi et al. [18]	2014	Tabriz	45/44; NA	89	46	43	6	11	21	3
Bakhti et al. [19]	2015	Ardabil	77/100; 50	177	76	98	34	23	NA	NA
Bakhti et al. [SID]	2014	Ardabil	NA; NA	160	77	90	NA	NA	37	NA
Bakhti et al. [SID]	2014	Ardabil	NA; NA	171	33	33	33	NA	NA	NA
Abdi et al. [20]	2017	Ardabil	44/85; 53.57	103	40	27	NA	NA	24	9
Bakhti et al. [21]	2015	Tehran	NA; 45.34	217	102	106	34	23	29	9
Douraghi et al. [21]	2009	Tehran	91/116; 44.8 ± 16	207	159	94	20	NA	30	NA

SID: Scientific Information Database.

The relationship of the infection with vacA i1/2 strains and development of GC lesions was calculated using the odds ratio index with 95% CI. The random effects model (Dersimonian and Laird method) was used in cases of high heterogeneity including I² index >25% and Cochran's Q-test p < 0.005; the statistical analysis was performed using Comprehensive Meta-Analysis (CMA v. 2.0) software. The publication bias was also measured according to Egger's regression method [4].

The eight studies that met our criteria were entered into the present analysis [12,[17], [18], [19], [20], [21]]. The studies were conducted in the period 2009–2019, five studies were performed in Ardabil, two in Tehran and one in Tabriz. In this study, the data from 1746 patients with an average age of about 48 ± 4 years were evaluated; about 41.9% of the patients were women and the rest were men. In general, 1414 strains of H. pylori were isolated from the patients in the present study. The frequency of vacA i1 in the present study was about 45.61% and the frequency of vacA i2 was about 43.21%. Based on the results of statistical analysis, we did not observe any significant relationship between infection acquisition with H. pylori vacA i1 or i2 and development of PUD or GC in the Iranian patients (Fig. 1).

Fig. 1.

Forrest plots of the vacA i1/i2 association with Helicobacter pylori-related gastrointestinal diseases. (a) vacA i1 in peptic ulcer disease (PUD) and gastric cancer (GC) development; (b) vacA i2 in PUD and GC development; (c) vacA i1 in PUD development; (d) vacA i2 in PUD development; (e) vacA i1 in GC development; (f) vacA i2 in GC development.

In the next step, we evaluated the role of infection with the H. pylori strains cagA + vacA i1/2 with PUD and GC, although no significant relationship was obtained for this. Moreover, because of insufficient information and the ambiguity of some results, we could not determine the role of cagA+/i1 or i2/s1m1 strains or vacA i1 or i2/s1m1 strains in the formation and development of PUD and GC. According to Egger's regression intercept (bias studies) no publication bias was observed in the included studies. It seems that vacA i1 or i2 has no role in the formation and onset of PUD and GC in the Iranian population.

The i locus is located between regions s and m and plays a functional role in the activity and formation of vacuole, such that the strains containing vacA i1 have the ability of vacuolating formation, while vacA i2 strains are unable to produce vacuolation [22]. Our data on the role of constellation in vacA polymorphic regions and the synergistic effects between vacA i1/2 alleles and cagA in the occurrence of PUD and GC are limited (Fig. 2).

Fig. 2.

The schematic of the vacA role in Helicobacter pylori pathogenesis.

In this study, we showed that infection with vacA i1/2 strains had no effect on the development and severity of PUD and GC in the Iranian population. We also demonstrated that the combination of vacA i1/2 and cagA played no roles in the formation of H. pylori-related gastrointestinal issues.

In a study of strains isolated from Iraqi and Iranian patients with gastric ulcer, the vacA i1 genotype was only associated with gastric ulcer in the Iraqi patients [23]. However, studies in East Asia have found a significant relationship between the vacA i1 allele as well as PUD and GC, such that i1 is considered a biomarker for the H. pylori-related gastrointestinal disease. Bagheri et al. revealed the high abundance of vacA i1 allele in Iranian patients [[24], [25], [26]].

In their study on the Swedish population, Karlsson et al. found no significant relationship between vacA i1 and PUD or GC development [27]. González-Rivera et al. found that by inhibiting the nuclear factor of activated T cells, i1 suppresses interleukin-2 production and prevents the formation of chronic inflammation and PUD [28]. However, Memon et al. observed a significant relationship between i1 genotype and duodenal ulcer [29].

In their meta-analysis, Liu et al. stated that infection with the vacA i1 genotype was significantly related to gastric cancer. They stated that the vacA i1 genotype had a significant relationship with GC, particularly in Asian countries (OR 10.89; 95% CI 4.11–20.88) [30]. Sugimoto et al. also showed that there is a significant relationship between vacA i1 genotype and GC development in Asian countries [14]. However, they noted in their study that a great number of s1, m1 and s1m1 strains contained i1. Hence, it is not yet possible to make a definite decision as to whether the vacA i1 genotype causes GC [14]. Numerous studies have also shown that the frequency of the vacA i1 genotype is high in vacA s1m1 strains (based on the results of many studies, these strains increase the risk of PUD and GC), which could alter the results and lead to false negatives [22,23,31,32]. In their meta-analysis, Pourmohammadi et al. showed that the vacA i1 genotype has no role in the GC development. They stated that cagA and vacA s1m1 have a significant relationship with the GC development [33].

Hence, the role of host genetics and environmental factors in the occurrence of PUD and GC, as well as the high frequency of cagA+ and vacA s1m1 strains in individuals with PUD and GC, cannot be definitely decided through the impact of vacA i1/2 alleles in PUD and GC development because of the diversity of the strains and differences in the frequency of vacA i1/2 in strains of different regions of the world. We need to plan and implement more and larger studies.

Conflict of interest

There is no conflict of interest.