Lack of association among TNF-α gene expression, -308 polymorphism (G > A) and virulence markers of Helicobacter pylori

Background Tumor necrosis factor plays a critical role in the pathogenesis of gastric diseases such as gastric cancer, and an abnormal inflammatory response has frequently been observed in dyspeptic patients. Helicobacter pylori infection can induce a gastric mucosal inflammatory response that may be influenced by -308 (G > A) polymorphisms and gene expression of the TNF-α gene. Methods One hundred and thirty-four gastric biopsy samples were collected from patients of both genders (61♂ and 73♀, mean age 40.3 ± 24.2 years) with gastric symptoms. The -308 (G > A) polymorphism of TNF-α was characterized using polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP). The expression level was measured using real-time PCR, and relative quantification (RQ) was calculated using the comparative CT method (2-ΔΔCT). Results The analysis revealed an increase in TNF-α gene expression in patients with gastritis; on the other hand, no statistical differences were observed in patients with gastric cancer. In addition, no association was found among -308 polymorphism genotypes, virulence markers, or TNF-α gene expression. Conclusions Helicobacter pylori induces a large increase in TNF-α expression in patients with gastritis, regardless of tissue inflammation, but after the tissue becomes neoplastic, the presence of bacteria did not influence expression. These results suggest that the TNF-α pathway may play an important role in the progression from gastritis to gastric cancer


Background
Helicobacter pylori (H. pylori) is a gram-negative bacterium that colonizes more than half of the world population, causing chronic infection. It comprises a major risk factor for gastric cancer (GC), which is the fourth most commonly diagnosed cancer and the second most deadly one, with over 750,000 new cases annually worldwide [1,2].
Several factors are correlated with the severity of clinical outcomes. These factors include the virulence markers of H. pylori, such as the genes cagA, vacA and dupA, and host genetic factors. However, bacterial factors or host genetic factors alone are not sufficient to explain why so few individuals harboring H. pylori develop gastric disease [3,4].
Evidence suggests that the main pathophysiological event in the H. pylori infection of gastric mucosa is the induction of a gastric mucosal inflammatory response. This reaction is induced by the contact of H. pylori with gastric cells, and is followed by neutrophil and mononuclear cell activation and the transcription and synthesis of pro-inflammatory and anti-inflammatory cytokines [4,5].
Tumor necrosis factor alpha (TNF-α) is a very important multifunctional pro-inflammatory cytokine. It provides a rapid form of host defense against infection by gramnegative bacteria. It is produced by activated monocytes and macrophages, which play a key role in inflammatory response [6]. However, this cytokine is also associated with gastric acid inhibition and paracrine effects, both of which may increase the risk of developing gastric cancers [7,8].
Several polymorphisms in the promoter region of the TNF-α gene have been associated with a variety of inflammatory diseases and cancer [9,10]. TNF-α-308 G > A (rs1800620), one of the most frequently studied polymorphisms on this gene, seems to have an influence on transcriptional activities and may contribute to the development of gastric disease [11]. Furthermore, previous studies showed that H. pylori virulence markers, such as cagA, vacA, and dupA genes, play important roles in gastric mucosal inflammation and injury through activated inflammatory cell infiltration [12][13][14].
CagA-cytotoxin is a virulence factor that has been well established as being associated with the development of peptic ulceration or gastric adenocarcinoma. This protein is translocated into the epithelial cytosol, thus causing changes that favor the persistence of the bacterium, but which also lead to changes in gastric cells without the direct adherence of bacteria in the stomach [15].
Another important virulence marker is vacuolating cytotoxin A (vacA), which encodes the secreted toxin VacA. The VacA toxin exhibits multiple cellular activities, and all H. pylori strains possess a vacA gene. This gene displays allelic diversity in two main regions: the signal (s) region (s1 and s2) and the middle (m) region (m1 and m2). Different combinations are associated with vacuolating capabilities. In addition, vacA s1/m1 strains are correlated with greater vacuolating activity and gastric disease, while vacA s2/m2 strains appear to be nonvacuolating [16][17][18].
Thus, the absence of an association with H. pylori positivity is understandable, as is the strong association between cagA and s1/m1 of vacA factors, and as well as their association with gastritis and gastric cancer. The secretion of the vacuolating toxin vacA has been found to induce vacuole formation in the gastric epithelial cellular membrane, as well as the modulation of apoptosis rather than the secretion of inflammatory cytokines. These mechanisms, in turn, influence gastric carcinogenesis [19].
The strong association between the inflammatory process and gastric carcinogenesis is very important in the evaluation of genetic polymorphisms in cytokine genes. The polymorphism in the TNF gene promoter region may influence not only cytokine gene expression, but also susceptibility to disease and pathogenesis [20]. Furthermore, the relationship among genes, host cytokine polymorphisms, their expression, and the severity of clinical outcomes in the ethnically diverse Brazilian population needs to be further characterized.
We believe that the relationship between the host genetic factors and the genetic factors of H. pylori can determine the severity of gastric disease in infected individuals. Thus, the aims of the current study were to determine the expression levels of the TNF-α cytokine gene in tissues samples from gastric cancer patients, chronic gastritis patients and uninfected controls, to evaluate the TNF-α -308 G > A polymorphism, and to determine the influence of H. pylori, cagA, vacA and dupA virulence markers on gastric carcinogenesis.

Methods
This study was approved by the Research Ethics Committee of Sacred Heart University (Universidade do Sagrado Coração) in Bauru, São Paulo, Brazil (under registration n. 068/12). Written informed consent was obtained from every patient involved in the study. All tissue samples were obtained by means of endoscopy or gastric surgery, procedures which were performed in the Department of Gastroenterology at Marília Medical School, the São Paulo Hospital, or the João de Barros Barreto University Hospital.

Study population
One hundred and thirty-four gastric biopsy samples were collected from patients of both genders (61♂/73♀, mean age 40.3 ± 24.2 years) with gastric symptoms. Histological parameters were analyzed and revealed 70 patients with chronic gastritis (G), 40 with normal gastric tissue (C), and 24 with gastric cancer (GC) tissue. The histological parameters were classified according to the Sydney system [21] and Lauren's classification [22]. Patients exposed to chemotherapy or radiotherapy before surgery or those who had received antimicrobial therapy, anti-cholinergic and anti-inflammatory agents, or proton pump inhibitors up to 30 days before endoscopy were excluded.
DNA extraction and H. Pylori status DNA from gastric biopsies was extracted using the QIAamp® tissue kit (Qiagen, Germany) according to the manufacturer's recommendations. The presence of H. pylori and virulence markers from cagA, vacA and dupA genes were all detected using polymerase chain reaction, as in previous studies [23][24][25][26][27].

Genotyping of TNF-α
The -308 (G > A) polymorphism of TNF-α was characterized using both polymerase chain reaction (PCR) and the restriction fragment length polymorphism (RFLP) technique according to the Wilson's protocol [28].The 107 base pair fragment was amplified using the primers 5′ AGGCAATAGGTTTTGAGGGCCAT 3 ′ and 5′ TCCTCCCTGCTCCGATTCCG 3′ , and the amplification condition that was utilized included 35 cycles: 45 s at 94°C, 45 s at 61°C, and 45 s at 72°C. The amplicons were digested with NcoI (Fermentas, USA) overnight at 37°C, visualized in 3 % agarose gel, stained with ethidium bromide, and analyzed using an Alpha Imager 2200 (Alpha Innotech Corporation™). The fragments visualized were 107 bp in length in the case of allele A and 87 and 20 bp in length in the case of allele G.

RNA Extraction and cDNA Synthesis
After collection, the biopsy fragments were immediately placed into an RNAlater solution (Applied Biosystems™, USA) and stored at -20°C until use. Total RNA was isolated using an RNeasy Mini Kit (Qiagen, Germany), and RNA concentration and quality were measured using the NanoDrop 2000 Spectrophotometer (NanoDrop, USA). The concentration was adjusted to 500 ng, and cDNA synthesis was conducted using High-Capacity cDNA Reverse Transcription Kits (Applied Biosystems™, USA) following the manufacturer's instructions.

Quantitative analysis of TNF-α gene expression
In order to determine TNF-α gene expression (Hs01113624_g1), the normal gastric tissue sample, which was negative for H. pylori infection, was designated as a calibrator. All reactions were repeated, and GUSB (Hs00187320_m1), UBC (Hs00221499_m1), and TBP (Hs00187332_m1) were used as reference genes.
The expression level was measured using the TaqMan® System (Life Technologies, USA) within the Applied Biosystems 7500 Fast Real-Time PCR system (Applied Biosystems™, USA). Relative quantification (RQ) was calculated using the comparative CT method (2 -ΔΔ ct).

Statistical analysis
The χ 2 test was used to compare the TNF-α allele and genotype distributions. Odds rates (ORs) were calculated using a dominant model (i.e., combining heterozygotes and homozygotes for the minor allele). To analyze gene expression, the two-tailed Student's t-test was used. Statistical analyses were performed using the GraphPad Prism 5 statistical package. Statistical significance was accepted at p < 0.05. Table 1 summarizes the characteristics of the 134 study subjects, 71 (53 %) of whom tested positive for H. pylori infection and 63 (47 %) of whom tested negative for H. pylori infection. H. pylori presence was associated with the development of G and GC (p < 0.001), and an association between cagA and s1/m1 alleles of the vacA gene in samples from G and GC was observed (p < 0.001).

H. Pylori infection and virulence markers
For the other analyses, the group with gastric cancer tissue that also tested negative for H. pylori was excluded due to the low sample number.
With the exception of the gastritis group (p = 0.076), genotype distributions were not within the Hardy-Weinberg equilibrium in the gastric cancer group (p = 0.012) or in the control group (p = 0.0065). Statistical analysis did not detect an association between the polymorphisms and gastric disease or between the polymorphisms and the presence of H. pylori.

TNF-α expression in patients with gastritis and gastric cancer
It is important to note that, in the expression analysis, samples from all of the test groups were compared to the normal gastric tissue samples, which were negative for H. pylori infection (control group RQ = 1.072).
The expression analysis revealed increased expression in patients with gastritis (RQ: 2.912) in relation to the control group (p = 0.0001). In the second analysis, the gastritis group was divided according to the presence of H. pylori: TNF-α expression was significantly higher in patients with gastritis who had tested positive for H. pylori (p = 0.0001; RQ = 3.774) ( Table 3). However, no statistical difference was observed in patients with gastritis and without H. pylori (p = 0.3697; RQ: 1.297) ( Table 3).
Conversely, there was no statistical difference observed when the control group (RQ = 1.072) was compared to patients with gastric cancer. Expression in neoplastic tissue (RQ = 1.163; p = 0.6332) did not present any alterations (Table 3).

Correlation among TNF-α gene expression, -308 (G > a) polymorphism, and H. Pylori virulence markers
No association was observed between the genotypes or alleles of the -308 (G > A) polymorphism and TNF-α gene expression in any of thesamples. First, gene expression of the samples withG/G genotypes was compared to that ofG > A + A/A genotypes in patients with gastritis (p = 0.8269) and with gastric cancer (p = ns).The small number of A/A genotypes,particularly in the gastric cancer samples, may have influenced theanalysis. In subsequent analyses, gene expression was compared tothe presence of cagA gene (p = 0.4677), to the presence of thedupA gene (p = 0.0518), and to genotypes s1/m1 and s2/m2 of the vacA gene (p = 0.3578). The presence of more pathogenic of H. pyloristrains did not induce changes in TNF-α expression. In addition, cagA, vacA and dupA genes were not found to be associated with differential gene expression of TNF-α cytokine.

Discussion
H. pylori infection in gastric mucosa causes an inflammatory response in the host characterized by infiltration of immune cells such as macrophages, neutrophils, and lymphocytes. This reaction is most likely induced by the contact of the bacterium with the epithelium; however, it has been estimated that only 20 % is adhered to the gastric mucosa and therefore able to activate the inflammatory response [29].
Thus, it is evident that virulence factors play a more important role in the progression of gastric carcinogenesis than the mere presence of the bacteria does.
Li et al. [6] evaluated several studies on the TNF-α-308 G > A polymorphism in different ethnic populations. In Caucasian populations, they demonstrated a weak association between the TNF-α-308 A variant and gastric cancer risk. Meanwhile, no associations were found between this A variant and gastric carcinogenesis in East Asian samples or other ethnic populations, including the Brazilian population. Two of the most recent Brazilian articles cited in this metaanalysis evaluated patients with chronic gastritis and gastric cancer and did not find any association between the TNF-α-308 A polymorphism and gastric carcinogenesis [4,36].
The association between inflammation and disease progression is well described in a gastric cancer model, which begins with chronic gastritis that persists and may induce changes such as metaplasia, dysplasia and, finally, cancer [37]. We also observed that, in patients with chronic gastritis, TNF-α mRNA was overexpressed in relation to the control patients, and the association was greater among patients who had gastritis and were also H. pylori positive. On the other hand, there was no increased expression in patients with gastric cancer. These results indicate the involvement of this cytokine in initial carcinogenesis. A recent study with a mouse model showed that TNF-α is also produced in excess in early colonic lesions in cases of colorectal cancer, and the data supported the closest association between TNF-α and early tumor progression [38].
Bartchewsky et al. [39] showed that patients with chronic gastritis infected by cagA-positive strains of H. pylori had the highest cytokine mRNA and protein levels. However, we did not find an association between higher cytokine mRNA level and patients infected by cagA-positive strains. Santos et al. [4] found similar results regarding the presence of the TNF-α-308 G > A polymorphism and infection by cagA-positive strains. This polymorphism apparently does not affect the presence of H. pylori or the progression of gastric disease.
More recently, Caliskan et al. [40] performed a study to determine the relationship between among H. pylori genotypes and their effects on cytokine release levels. These authors studied five genotypes: cagA and vacA  s1m2 (genotype 1); cagA and vacA s1m1 (genotype 2); cagA, vacA s1m2, and babA2 (genotype 3); vacA s2m2 (genotype 4); and cagA and vacA s2m2 (genotype 5). These researchers found that all of these genotypes increased IL-1β, IL-6, IL-8, IL 10, and TNF-α levels in THP-1 cells. However, genotype 5 was found to cause higher amounts of IL-1β, IL-6, TNF-α, and IL-10, whereas genotype 1 induced the highest levels of IL-8. Thus, they suggested that the host cytokine response varies depending on the individual specific immune response to different H. pylori strains.
In another recent study, Michalkiewicz et al. [41] analyzed gene expression of innate immunity and cytokines in the gastric mucosa from H. pylori-infected children and uninfected children and found high mRNA expression of IL-6, IL-10, IFN-γ, and TNF-α. However, H. pylori cagApositive strains were associated only with higher IL-6 and IL-10 mRNA expression in relation to the cagA-negative strains. Meanwhile, TNF-α presented increased protein levels and was correlated with bacterial density.
Our results are consistent with those of several authors, and we offer two hypotheses. It is possible that bacterial density, not the H. pylori genotypes, stimulates mRNA synthesis of TNF-α. It is also possible that an increase in protein levels occurs without a significant increase in gene expression. The important finding is that, regardless of the pathway, TNF-α induces an intense inflammatory response that inhibits gastric acid production and suggests a worse prognosis.
In addition, we found that, contrary to results reported in the literature, TNF-α-308 G > A polymorphisms did not influence TNF-α mRNA expression in cases of gastric disease. In patients with postoperative sepsis, Baghel et al. [20] revealed that the AA homozygous genotype presents a greater capacity for producing TNFα cytokines than other genotypes. Higuchi et al. [42] reported that the A variant of the TNF-α-308 polymorphism presented higher transcriptional activity and increased TNF-α cytokine production. Hosono et al. [43] found increased tumor necrosis factor receptor 1 expression in human colorectal adenomas, though TNF-α expression did not differ significantly.
These contradictory results may be explained, at least in part, by the fact that the Brazilian population is composed of a genetic miscegenation of various ethnic groups. This variation may influence allele and/or genotype frequencies and their correlations with gene expression. Furthermore, most studies evaluate protein levels of TNF-α, which are not necessarily associated with the mRNA levels of TNF-α. A recent study published by de Oliveira et al. [44] found results similar to ours. These authors reported that the polymorphic alleles of TNF-α (-308) did not have any effect on TNF-α or IL-10 expression levels. It is important to note that this study was performed on a Brazilian population and reinforces the hypothesis mentioned above.

Conclusion
Ours results show that the presence of Helicobacter pylori induces a large increase in TNF-α expression in patients with gastritis, regardless of tissue inflammation, but after the tissue becomes neoplastic, the presence of bacteria did not influence expression. These results suggest that the TNF-α pathway may play an important role in the progression from gastritis to gastric cancer.

Ethics committee approval
This study was approved by the Research Ethics Committee of Sacred Heart University (Universidade do Sagrado Coração) in Bauru, SP, Brazil (under registration n. 068/12).

Consent
Written informed consent was obtained from all patients or their legal guardians for publication of this case report.