INTRODUCTION
Helicobacter pylori, a microorganism adapted to colonize the gastric mucosa, is considered to be the main etiological agent of enanthematous gastritis (inflammation of the gastric epithelium with simple change mucosal), erosive gastritis (inflammation with loss of integrity of the epithelial lining, not exceeding the muscular layer of the mucosa) and also a risk factor for peptic ulcer and gastric cancer in humans 1,2 . Factors related to the genetic polymorphism of the host, the diversity of bacterial pathogenicity and the environment seem to be related to the broad clinical spectrum related to infection by H. pylori 3 . Several putative genes, such as cagA, cagE, vacA, iceA and babA2, have been identified and are likely to play an important role in the pathogenicity of the bacterium 4–8 .
The cag-PAI is composed of approximately 31 genes, which are responsible for coding type IV secretion system components and inject effector molecules in the host cell. The presence of cag-PAI affects the inflammatory state of the gastric mucosa by polymorphonuclear cell infiltration and increases the production of interleukin-8 (IL-8) 1 . cagA gene (cytotoxin associated gene A) is considered to be the cag-PAI marker. cagA positive strains tend to be more pathogenic, produce more severe lesions of the epithelium and increase the expression of interleukin-1β and IL-8 9,10 . Another member of the cag-PAI, the cagE gene (cytotoxin associated gene E), is also related to an increased production of IL-8 in the gastric epithelial cells 11 .
The vacA gene encodes the vacuolating cytotoxin that damages the gastric epithelial cells. It comprises two variable parts: the s-region, which encodes the signal peptide with the s1 or s2 allele, and the m-region (middle) with the m1 or m2 allele 6,12 . The mosaic combination of the s and m region alleles determines the production of the vacuolating cytotoxin and is associated with the pathogenicity of the bacterium 13 . In general, vacAs1/m1 and s1/m2 strains produce high and moderate levels of vacuolating toxin, respectively, whereas the s2/m2 strains produce little or no toxin 12 . The vacAs1/m1 genotype is considered to be associated with more severe pathologies 14 .
The iceA gene (induced by contact with the epithelium) has two main allelic variants, designated iceA1 and iceA2. The iceA1 allele is up-regulated by the contact of H. pylori with gastric epithelial cells and has been associated with peptic ulcer disease. Meanwhile, the iceA2 allele has been related to asymptomatic gastritis and non-ulcer dyspepsia 7,15 .
The babA gene (blood-group antigen-binding adhesin) encodes a membrane protein, an adhesion called BabA, which binds to the Lewisb blood group antigen on the gastric epithelial cells 8,16 . Although three baballeles have been identified (babA1, babA2 and babB), only the babA2 gene product is necessary for the Lewisb binding activity. Thus, babA2 is responsible for pathogenicity, allowing contact between bacterium and gastric epithelium and facilitating the release of other pathogenicity factors 11 .
We hypothesized that the clinical outcomes of H. pylori infection were influenced by the distribution of the above-mentioned pathogenic factors; therefore, this study aimed at investigating the role of cagE as a pathogenicity biomarker of H. pylori-positive patients, compare it to cagA, vacA, iceA and babA2 genes and correlate these findings with endoscopic diagnoses.
METHODS
Patients and clinical samples
In this study were included 144 patients with dyspeptic symptoms submitted to upper gastrointestinal endoscopy between October 2008 and March 2009 in the Integrated Center for Gastroenterology at the Hospital of the Federal University of Rio Grande, Rio Grande do Sul, Brazil. Patients that had recently (within the last 15 days) received antibiotics or non-steroidal anti-inflammatory drugs (NSAIDs) or had been treated for H. pylori or gastrointestinal bleeding in the last seven days, were excluded. The presence of H. pylori infection in the subjects was determined by histological examination and detection of the ureAgene by polymerase chain reaction (PCR).
Endoscopic diagnosis
The endoscopic diagnosis was established in accordance to the Sydney System classification 17 .
Histological examination
The biopsy samples of the gastric antrum and body destined for histology were fixed in formalin and stained with Hematoxylin-Eosin (H&E) and Giemsa. Histological classification of gastritis was established according to the Sydney System 18 .
Extraction of DNA
After collection, the biopsy samples of the gastric antrum and body were kept in Brain Heart Infusion Broth (Acumedia®, United States of America) with 20% glycerol and stored at -70 °C for further DNA extraction. DNA was extracted of the biopsy samples using DNAzol® Reagent (Invitrogen™, United States of America) and 10µg/µL proteinase K (Promega, United States of America). The samples were separated from the broth and re-suspended in 100µL of proteinase K and 500µL of DNAzol® Reagent. The mixture was incubated at 55°C for 3h and, after this period, 500µL of DNAzol® Reagent was added to it again. After centrifugation at 14,000g for 10min, the supernatant was collected and 500µL cold absolute ethanol was added, followed by centrifugation at 12,000 g for 10min, after which the supernatant was discarded. The DNA pellet was washed two times with 800µL of 75% ethanol, air dried and re-suspended in 50µL of 8mM NaOH. The DNA was stored at -20°C until further usage.
Detection of the ureA gene
The detection of the ureA gene was used to confirm the H. pylori infection in all of the patients 19 . PCRwas performed as described by Rota et al 20 .
Detection of pathogenicity genes by PCR
The presence of the cagA gene was investigated by the amplification of the constant region near the 3′ end of the cagA. The PCR was performed as proposed by Rota et al. 20,21 . and the cagE gene was investigated according to Sozzi et al. 22 . The presence of the vacA and iceA alleles in the biopsy samples was investigated using the primers previously described 23,24 and the PCR was conducted as proposed by Benenson et al 25 . For detection of the babA2 gene, the primers and the PCR conditions applied, were described by Sheu et al. 26 .
RESULTS
From the 144 patients who underwent endoscopy, 57 (39.6%) presented H. pylori under histological examination and PCR, of these 40 were women and 17 were men with an average age of 46.2 years (range, 14-74 years). Based on the endoscopic diagnoses, 45.6% (26/57) of the patients had erosive gastritis, while 54.4% (31/57) had enanthematous gastritis.
The distribution of the cagA, cagE, vacA, iceA and babA2 genes in relation to the endoscopic diagnoses is described in Table 1. A statistically significant association was found between the cagE gene and the diagnosis of erosive gastritis (p=0.029). However, between the cagA, vacA, iceA and babA2 genes and the clinical manifestations, no statistically significant association was observed, although a biological significance was suggested.
Genotype | Enanthematous gastritis(n = 31) | Erosive gastritis(n = 26) | ||
---|---|---|---|---|
n | % | n | % | |
vacAa | ||||
vacAs1/m1 (n = 15) | 6 | 40.0 | 9 | 60.0 |
vacAs1/m2 (n = 10) | 4 | 40.0 | 6 | 60.0 |
vacAs2/m1 (n = 1) | 1 | 100.0 | – | |
vacAs2/m2 (n = 12) | 10 | 83.3 | 2 | 16.7 |
vacA-negative (n = 19) | 10 | 52.6 | 9 | 47.4 |
cagAb | ||||
cagA-positive (n = 26) | 11 | 42.3 | 15 | 57.7 |
cagA-negative (n = 31) | 20 | 64.5 | 11 | 35.5 |
cagEc | ||||
cagE-positive (n = 24) | 9 | 37.5 | 15 | 62.5 |
cagE-negative (n = 33) | 22 | 66.7 | 11 | 33.3 |
iceAd | ||||
iceA1 (n = 11) | 4 | 36.4 | 7 | 63.6 |
iceA2 (n = 30) | 17 | 56.7 | 13 | 43.3 |
iceA-negative (n = 16) | 10 | 62.5 | 6 | 37.5 |
babA2e | ||||
babA2-positive (n = 32) | 19 | 59.4 | 13 | 40.6 |
babA2-negative (n = 25) | 12 | 48.0 | 13 | 52.0 |
ap-value of the Chi-square test = 0.136;
bp-value of the Chi-square test = 0.094;
cp-value of the Chi-square test = 0.029;
The presence of the cagA gene was significantly correlated with the cagE gene (p < 0.001) and only two cagA-positive samples did not present the cagE gene ( Table 2). A statistically significant association was also observed between the presence of cagA, cagE and vacA genes versus babA2 (p<0.05) ( Table 3). We evaluated the distribution of genes in all patients. The combination cagA/cagE was detected in 62.5% (15/24) and 37.5% (9/24) of patients with erosive and enanthematous gastritis, respectively. The biomarkers cagA/cagE/babA2/vacAs1m1/iceA1 combined was present in 15.4% (4/26) of patients with erosive gastritis. In patients with enanthematous gastritis, the combination babA2/vacAs2m2/iceA2 was detected in 22.6% (7/31). Among patients H. pylori-positive, 28.1% (16/57) did not show any of the biomarkers studied here.
Genotype | cagE – positive(n = 24) | cagE – negative(n = 33) |
---|---|---|
cagA – positive (n = 26) | 92.3% (24/26) | 7.7% (2/26) |
cagA – negative (n = 31) | 0% (0/31) | 100% (31/31) |
Genotype | babA2 – positive(n = 32) | babA2 – negative(n = 25) |
---|---|---|
cagAa | ||
cagA – positive (n = 26) | 80.8% (21/26) | 19.2% (5/26) |
cagA – negative (n = 31) | 35.5% (11/31) | 64.5% (20/31) |
cagEb | ||
cagE – positive (n = 24) | 79.2% (19/24) | 20.8% (5/24) |
cagE – negative (n = 33) | 39.4% (13/33) | 60.6% (20/33) |
vacAc | ||
vacAs1/m1 (n = 15) | 73.3% (11/15) | 26.7% (4/15) |
vacAs1/m2 (n = 10) | 90.0% (9/10) | 10.0% (1/10) |
vacAs2/m1 (n = 1) | 100.0% (1/1) | – |
vacAs2/m2 (n = 12) | 83.3% (10/12) | 16.7% (2/12) |
vacA-negative (n = 19) | 5.3% (1/19) | 94.7% (18/19) |
ap-value of the Chi-square test < 0.001;
DISCUSSION
Helicobacter pylori infection has been related to severe gastroduodenal diseases. There is an increasing evidence that the presence of H. pylori genes and their different genotypic combinations are related to development of gastric diseases 11 .
The cagA gene was detected in 57.7% (15/26) of gastric biopsy samples from patients with erosive gastritis. This gene has often been associated with the apoptosis of T helper type 1 (Th1) cells, increased IL-8 production, increased inflammation in the gastric mucosa and a higher risk for developing peptic ulcers or gastric cancers 27 .
On the other hand, the cagE gene was identified in 62.5% (15/24) of gastric biopsy samples from patients with erosive gastritis. This study found a statistically significant association between cagE and erosive gastritis, a more severe mucosal injury. This may be due to the fact that this gene is directly connected with an increase in the production of IL-8 in the gastric cells and with the intensity of epithelial damage 28 .
When evaluating the effect of the combination of genes with the type of gastritis, the presence of the cagA/cagE combination in patients with erosive gastritis was 62.5% (15/24). In patients with enanthematous gastritis, this combination was detected in 37.5% (9/24). The relation between the variables was statistically significant (p=0.047). These results permit to infer that cagE is an important marker of pathogenicity alone or combined with cagA.
Concerning the vacA gene, the combination s1/m1 was mostly detected in gastric biopsy samples derived from patients with an endoscopic diagnosis of erosive gastritis. However, the combination s2/m2 of the vacA gene was frequently observed in samples from patients with enanthematous gastritis. In general, vacAs1/m1 strains have been linked with higher degrees of inflammation and cell infiltration when compared to vacAs2/m2 strains 29,30 . Furthermore, vacAs1/m1 strains produce large amounts of vacuolating toxin and induce a higher vacuolating activity in gastric epithelial cells than vacAs2/m2 strains12,31 .
Regarding the iceA gene, the iceA1 allele was more commonly found in samples from patients with erosive gastritis (63.6%), whereas the iceA2 allele was more commonly identified in specimens from patients with enanthematous gastritis (56.7%), but no statistically significant association was observed. A previous study demonstrated that iceA1 expression was significantly related to the host mucosal response, which led to the hypothesis that the levels of transcription within the host environment may contribute to disease development. In contrast, iceA2 expression may be more influenced by the gene structure, which has a repeated protein structure but is not homologous with known proteins 7 .
A statistically significant association was observed between cagA, cagE, vacA genes and babA2 (p<0.05) (Table 3), but other authors did not find any association between these pathogenicity genes in the samples they investigated 23,32,33 . Our data, however, supports the relationship between the genes cagA, cagE, vacA and babA2 that was described in previous reports 34–37 .
The association of biomarkers cagA/cagE/babA2/vacAs1m1/iceA1 was detected in 15.4% (4/26) of patients with erosive gastritis, this is such an important evidence, considering that these genotypes are more pathogenic. Similar percentages were found by studies conducted in Colombia and in Brazil. In the latter, only the iceA gene was discordant 38,39 . In patients with enanthematous gastritis, the combination of babA2/vacAs2m2/iceA2 was detected in 22.6% (7/31). In a previous study in southern Brazil, the vacAs2m2 and iceA2 alleles were also related with enanthematous gastritis 40 .
According to the results, we concluded that the detection of H. pylori is not in itself sufficient to assess the development of gastric mucosal damage, but the presence of pathogenicity genes is able to give such information. Although the small number of samples can be a limitation in this study, these findings highlight the importance of the detection of biomarkers to evaluate the need of treatment for the microorganism eradication, since in some cases the elimination can lead to the development of other pathologies such as gastric esophageal reflux, asthma and obesity 41 . The cagE gene can be used as a risk biomarker for gastric lesions contributing to a better assessment of the pathogenic potential of H. pyloriand for the infection prognosis of the gastric mucosa.