Helicobacter pylori is one of the most common pathogens affecting humans, reported to infect approximately 35% to 70% of the world’s population. Many individuals infected with H. pylori will develop asymptomatic gastritis, but 10% develop peptic ulcer (gastric or duodenal) or gastric cancer. The clinical outcome of the infection depends on a combination of bacterial, host, and environmental factors1.
Different virulence genes, such as the cytotoxin-associated (cagA) gene and the vacuolating cytotoxin (vacA) gene, have been described in H. pylori infections. Studies conducted in Brazil have shown an association between the s1m1 genotype and cagA positivity and the development of gastrointestinal diseases such as peptic ulcers and gastric cancer2−4. According to data from the National Cancer Institute (INCA), the frequency of gastric diseases such as peptic ulcers and gastric adenocarcinoma is high in the northern region of Brazil, particularly in the State of Pará5. Most cases of gastric cancer have been reported in the northeastern municipalities of the state. However, only a few studies have investigated the occurrence of H. pylori in the State of Pará, and these studies have been restricted to the capital, Belém. In Belém, there is a high prevalence of H. pylori infection among adult patients with gastric disorders, which ranges from 64% to 74% in patients with gastritis4,6 and is 82% in patients with gastric ulcer7.
The objective of the present study was to evaluate differences in presence of virulence markers (cagA and vacA genes) between H. pylori strains isolated from patients with chronic gastritis in 2 cities within the State of Pará.
Patients
Gastric biopsies were collected from 2 groups. There were 168 patients from Belém with chronic gastritis who were seen at the Ofir Loyola Hospital in Belém and 151 patients from Bragança with chronic gastritis who were seen at the Santo Antonio Maria Zaccaria Hospital in Bragança (Figure 1).
During endoscopy, 4 gastric biopsy fragments were collected. Two antral biopsies were analyzed by histological methods, and 2 antral specimens were also analyzed by molecular methods.
Histological evaluation
The biopsies were fixed in 10% buffered formalin solution, processed in alcohols, and embedded in paraffin, and 4 µm thick sequential sections were cut and stained with hematoxylin and eosin. The histopathological findings of chronic inflammation and polymorphonuclear activity were scored on a scale from 0 to 3 using the criteria described in the updated Sydney classification system1,8, with 0 indicating no inflammation, 1 light inflammation, 2 moderate inflammation, and 3 severe inflammation.
Deoxyribonucleic acid extraction
Total DNA was extracted from frozen gastric biopsy specimens by the addition of 10µL proteinase K and 300µL lysis buffer (200mM Tris-HCl, 25mM EDTA, 300mM NaCl, 1.2% sodium dodecyl sulfate). The lysate was extracted with an equal volume of phenol-chloroform, precipitated with isopropanol, and washed with 70% ethanol.
PCR amplification and detection of amplified DNA products
Polymerase chain reaction (PCR) amplification for the detection of H. pylori DNA in gastric mucosa was performed as described previously9. The previously described F1-F and B1-R primers10 were used for detection of cagA, and vacA was amplified using the oligonucleotide primers described by Atherton et al.11, the vacA signal region (vacA s1 or s2: primers SS1-F and SS2-F/VA1-R, respectively) and middle region (vacA m1 or m2: primers VA3-F/VA3-R and VA4-F/VA4-R, respectively).
The PCR products were visualized by electrophoresis on 2% agarose gel stained with ethidium bromide and examined under UV illumination.
Statistical analysis
Data were analyzed using Bioestat 5.0 software (available in http://www.mamiraua.org.br/downloads/programas). The log-likelihood ratio G-test with Yates’ correction for continuity12 was used to compare frequencies and to evaluate the association between bacterial genotypes and histological findings.
The G test was used to compare frequencies, and to evaluate the association between bacterial genotypes and histological findings. A level of significance of 5% was adopted.
The epidemiological data of the 2 groups showed what the mean age of patients from Belém, the state capital, was 45 years (range: 18 to 91 years), and the mean age of patients from Bragança was 36 years (range: 18 to 64 years).
Infection with H. pylori was more frequent among patients from Bragança. Bacterial DNA was observed in 142 (94%) of the 151 patients in Bragança, while H. pylori DNA of was isolated in 130 (77%) of the 168 patients in Belém. Five of the 130 patients from Belém and 12 of the 142 patients from Bragança were co-infected with at least two different H. pylori isolates, because the DNA associated with both the m1 and the m2 alleles was detected. Thus, the number of isolates analyzed for the prevalence of cagA and allelic variants of vacA was reduced to 125 in Belém and to 130 in Bragança.
The results of the amplification of the different alleles of the 2 major H. pylori virulence factors, cagA and vacA, are shown in Figure 2. All possible combinations of the vacA alleles were identified. The most prevalent vacA s-region genotype was s1, whose frequency ranged from 79% (99/125) in Belém to 95% (124/130) in Bragança. For the vacA m-region, genotype m1 was the most prevalent among Belém strains (80%, 100/125) and among Bragança strains (96%, 125/130). The most frequent combination of vacAalleles found in patients from both Belém and Bragança was s1m1, with a significant difference between the 2 cities (G = 20.63, p ≤ 0.01).
FIGURE 2 Prevalence of the cytotoxin-associated (cagA) gene and of the different allele combinations of vacuolating cytotoxin (vacA) gene in patients with chronic gastritis.
The cagA gene was detected in 95/125 (76%) of Belém patients and in 87% (113/130) of Bragança patients, with no significant differences between the 2 cities.
Analysis of the association between the degree of inflammation and neutrophil activity and the 2 major H. pylori virulence factors, cagA and vacA, indicated a higher degree of inflammation and neutrophil activity in patients infected with s1m1 cagA-positive (s1m1 cagA+) strains when compared to non-virulent strains (s1m1 cagA−, s1m2 cagA−, s2m1 cagA−, s2m2 cagA−) (Table 1).
TABLE 1 – Association between the degree of inflammation and neutrophil activity and virulence factors (vacA and cagA) in Helicobacter pylori.
| Genotype | Degree of inflammation | Neutrophil activity | ||||||
|---|---|---|---|---|---|---|---|---|
| 1(%) | 2–3 (%) | G test | P | 1 (%) | 2–3 (%) | G test | P | |
| Belém | ||||||||
| s1m1-cagA+ | 13 (14.0) | 82 (86.0) | 19.8632 | <0.01 | 20 (21.0) | 75 (79.0) | 10,4287 | <0.01 |
| NV | 12 (70.0) | 5 (30.0) | 11 (64.0) | 6 (36.0) | ||||
| Bragança | ||||||||
| s1m1-cagA+ | 2 (2.0) | 111 (98.0) | 37.9049 | <0.01 | 4 (3.0) | 109 (97.0) | 37,7517 | <0.01 |
| NV | 11 (65.0) | 6 (35.0) | 12 (70.0) | 5 (30.0) | ||||
NV: non-virulent strain (s1m1 cagA”, s1m2 cagA”, s2m1 cagA”, s2m2 cagA”).
The H. pylori genome is genetically diverse and different virulence genes contribute to the risk and severity of disease outcome. Several studies have demonstrated geographical differences in the prevalence of vacAalleles and cagA status among H. pylori isolates2,3,13.
In the present study, a high prevalence of genotype s1m1 cagA+ was observed among patients from the 2 cities studied. The prevalence of s1m1 cagA+ strains was higher in Bragança than in Belém. However, the incidence of gastric cancer is higher in Belém. This finding might be explained by the fact that reporting of cases is performed by the cancer referral hospital in Belém, and does not consider the origin of the patient as demonstrated by the records of the Brazilian Ministry of Health, which also suggests the underreporting of diagnosed gastric cancer cases in Bragança.
Regardless of potential underreporting of cases in Bragança, mortality due to gastric cancer has been increasing since 2005, from 2/100,000 inhabitants to 6/100,000 inhabitants in 2007 and 15/100,000 inhabitants by 2010. In contrast, these rates remained constant in Belém, with a decline from 187/100,00 inhabitants in 2007 to 166/100,000 inhabitants in 20085.
Studies conducted in Brazil have shown that vacA s1m1 and cagA+ genotypes increase the risk of gastric cancer and peptic ulcers2−4. In the present study, comparison of bacterial genotypes and histopathological findings showed that patients from the 2 cities who carried s1m1 cagA+ strains had a higher degree of inflammation and neutrophil activity in the gastric mucosa.
Helicobacter pylori is a well-established risk factor, but not a sufficient cause for the development of stomach cancer14. In this respect, numerous epidemiological studies have indicated diet to be an important exogenous risk factor14,15. There is marked diversity in dietary habits in the Amazon region as a whole due to differences in environmental conditions. Whereas the diet of the rural population consists of fruits, game, and fish, complemented by cassava flour, the source of protein for the population living in the state capital is often dried fish and beef jerky, along with a high intake of canned products14,15. Other lifestyle-related factors such as smoking, alcohol consumption, and stress, which have been associated with gastric carcinogenesis, are also more frequent in urban areas15.
Therefore, in addition to H. pylori infection, other factors probably contribute to the elevated incidence of gastric cancer in the cities of the State of Pará.
