Helicobacter pylori in Colorectal Carcinoma Tissue

Introduction

Helicobacter pylori (1) is associated with the development of chronic gastritis, peptic ulcer, and gastric adenocarcinoma (2-4). Previous studies reported positive (5, 6) and negative (7, 8) associations between infection and colorectal neoplasia. Recent reports suggest that H. pylori may be an association factor involved in the development of colorectal cancer in patients infected with certain strains (9, 10). The Swedish group of Grahn et al. (11) was the first who previously described the potential molecular identification of Helicobacter DNA in colorectal cancer biopsies by means of a 16S rDNA PCR amplification assay combined with pyrosequencing analysis. Furthermore, Maggio-Price et al. (12) proved that infection of SMAD3^−/− mice with Helicobacter triggers colon cancer in 50% to 66% of those animals. The aim of this study was to detect and relate the possible role of this microorganism in the etiology of colorectal cancer in colorectal cancer tissue specimens.

Materials and Methods

Patients

The study population consisted of 83 consecutive patients with colorectal adenocarcinoma (55 male, 28 female; mean age, 61.2 ± 10.2 years) who were referred from February 2002 to April 2003 to the Belgrade Institute for Digestive Diseases, University Clinical Center, for endoscopic evaluation and surgical treatment of colorectal cancer. The final clinical diagnosis of the patients was ascertained after endoscopy and confirmed after surgery. Histopathology confirmed adenocarcinoma in 81 and squamocellular carcinoma in 2 patients with colorectal cancer. All patients were tested serologically (IgG H. pylori AT, DPC) for the presence of H. pylori infection prior to the surgical treatment.

Surgery and Endoscopy

During surgical treatment of patients with colorectal cancer, the samples of mucosa (5 × 5 mm) were taken from malignant tissue and from normal tissue as well. After conservation in sterile tubes mixed with 1 mL of RNAlater stabilization solution (Ambion, Austin, TX), they were deep-frozen at −40°C for subsequent PCR analyses (Molecular Gastroenterology Unit, Medical Faculty of Mannheim, Mannheim, Germany and Department of Microbiology, University of Regensburg, Regensburg, Germany).

PCR Analysis

To exclude methodologic bias, the samples were analyzed blindly in two different laboratories employing several different PCR strategies to detect Helicobacter species. After removal of RNAlater, the tissue was immersed in liquid nitrogen and crushed to produce readily digestible pieces. The processed tissue was placed in a solution of proteinase K and digestion buffer, and incubated until most of the cellular protein was degraded. High–molecular weight DNA was prepared from samples according to established protocols (13). The DNA was then used for H. pylori and for K-ras analyses as well. Different PCR assays were done, covering the urease A gene and the 16S gene with three assays each. For urease A, a nested PCR was done as described previously (14). To amplify Helicobacter species–specific 16S rDNA from colorectal cancer samples, two different PCR strategies were done (15). The first PCR was done by using either the highly conserved Escherichia coli 16S rDNA primer pairs C93 and C94 to obtain a 1,465-bp amplicon, covering nearly the whole 16S rDNA gene, or by using the more Helicobacter-specific primer pairs C95 and C96 to generate a 519-bp gene fragment. Nested PCR was then done using the Helicobacter species–specific primer pairs C97 and C98 to obtain a 398-bp gene product. All Helicobacter species–specific amplicons were sequenced subsequently (ABI PRISM 377 DNA Sequencer; Applied Biosystems, Weiterstadt, Germany) and compared with published sequences using the basic BLAST search program. In addition to the conventional PCR protocols described above for 16S rDNA amplification, two nested LightCycler (Roche Diagnostics, Mannheim, Germany) PCR assays were done. The results of both ureA and 16S rDNA PCR are presented in Table 1 (16). Specific amplicons were subsequently purified for sequencing using the High-Pure PCR Product Purification kit (Roche Diagnostics).

K-ras

The K-ras PCR was done as described previously (17). If the K-ras gene is mutated in codon 12, a double band will result, representing the undigested original 99 bp and 78 bp in addition to the 21 bp fragment. The result is visualized on a standard 3% agarose gel (NuSieve; FMC Bioproducts, Rockland, ME).

Biostatistics

All statistical analyses were conducted using the SPSS 11.0.0 statistical software package. The results were considered to be highly significant at P < 0.01.

Results and Discussion

Analyzing the H. pylori IgG seropositivity, we observed 36 positive patients (43.3%). H. pylori PCR was positive in one case (1.2%) of colorectal cancer in the tumor tissue and in five samples (6.0%) of a normal colonic mucosa in the cancer patients. H. pylori seropositivity correlates significantly with H. pylori PCR in normal mucosal samples (Spearman's ρ = 0.289; P < 0.01), which proves that all patients with H. pylori in normal colonic mucosa were, at the same time, seropositive. All patients who were positive to H. pylori ureA gene on PCR were also positive to serologic H. pylori testing. According the χ² test, there was no statistical correlation between H. pylori PCR positivity and colorectal cancer (Table 2 ). One patient was positive both in colorectal cancer and normal mucosal samples as well. We successfully detected the presence of the K-ras gene in 61 (73.4%) malignant and 63 (75.9%) normal mucosal samples in the group of patients with colorectal cancer. There was no statistical difference in PCR K-ras positivity between colorectal cancer malignant and normal mucosa (Table 2), which proves the feasibility of the method. The K-ras PCR showed gene mutations in 19 tumor tissues from patients with colorectal cancer (31.1%) and in 2 cases (3.3%) of normal colonic mucosa from patients with colorectal cancer. There is a significant statistical correlation between K-ras mutation in malignant and normal mucosal samples among patients with colorectal cancer (Table 2).

Serum IgG antibodies were tested against H. pylori (ELISA) and CagA protein (Western blot assay) in patients with colorectal adenocarcinoma, gastric adenocarcinoma, and with other malignancies (cancer controls; ref. 10). Among patients infected with H. pylori, CagA+ seropositivity was associated with increased risk for both gastric and colonic cancer.

The study of Grahn et al. (11) employed PCR amplification and subsequent pyrosequencing analysis to reveal the presence of Helicobacter-specific 16S rDNA variable V3 region sequences in colorectal cancer biopsy specimens. Helicobacter DNA was identified in 26% of colon cancers and in 29% of rectal cancer biopsies.

The animal studies of Maggio-Price et al. (12), using real-time PCR, found that Helicobacter organisms concentrate in the cecum, the preferred site of experimental tumor development. Mucinous adenocarcinomas developed 5 to 30 weeks after infection. Colonic tissue revealed increased transcripts for the oncogene c-myc and the proinflammatory cytokines interleukin-1A, interleukin-6, IFN-γ, and tumor necrosis factor-α, some of which have been implicated in colon carcinoma. These results suggest that bacteria may be important in triggering colorectal cancer.

In our study, we detected the presence of H. pylori genomic material by PCR reaction in colorectal cancer malignant mucosa and normal colonic mucosa as well. We detected only one positive sample from malignant tissue and five positive samples from normal tissue among 83 patients with colorectal cancer. Our established PCR for H. pylori was feasible for colorectal cancer tissue as well. However, H. pylori is not considered to play an important role in the pathogenesis of colorectal cancer. The identification of K-ras mutations in routine PCR analysis correlates with the presence of colorectal cancer.