10.1245/ASO.2006.05.021
Annals of Surgical Oncology 13:881-886 (2006)
© 2006 Society of Surgical Oncology
Implication of the BRCA2 and Putative ‘‘BRCA3’’ Genes in Dukes’ Stage C, Replication Error–Negative Colon Cancer
Nala S. Sivarajasingham, MD,
Lynn Cawkwell, PhD,
Richard P. Baker, MD,
Sara L. O’Kane, BSc,
Edward F. Smyth, MD,
Jonathan V. T. Tilsed, MD,
Mark B. Watson, PhD,
John Greenman, PhD and
John R. T. Monson, MD
Postgraduate Medical Institute of the University of Hull in Association with the Hull-York Medical School, University of Hull, R&D Building, Castle Hill Hospital, Hull, HU16 5JQ, United Kingdom
Correspondence: Address correspondence and reprint requests to: Lynn Cawk-well, PhD; E-mail: l.cawkwell{at}hull.ac.uk.
 |
ABSTRACT
|
|---|
Background: Although BRCA genes have been implicated in certain tumors, particularly breast tumors, their role in colon tumorigenesis has not been fully explored. We aimed to investigate the association of the BRCA2 and putative ‘‘BRCA3’’ genes in a homogeneous series of right-sided colon cancer specimens.
Methods: Twenty-three Dukes’ stage C, replication error–negative carcinomas were selected from patients with right-sided colon cancer. After histological examination and microdissection, DNA was extracted from normal colon and carcinoma from each patient. Five microsatellite markers spanning the region of BRCA2 and BRCA3 on chromosome 13 (D13S218, D13S219, D13S165, D13S156, and D13S160) and two markers intragenic to BRCA2 and BRCA3 (D13S171 and D13S1308, respectively) were used. Polymerase chain reaction products were analyzed by using a fluorescent allele imbalance assay.
Results: Markers demonstrating the highest allelic imbalance were D13S1308 (53%), D13S171 (33%), and D13S160 (37%).
Conclusions: The intragenic markers D13S1308 (BRCA3) and D13S171 (BRCA2) on chromosome 13 demonstrated a high frequency of allelic imbalance in primary colon carcinoma. This suggests an involvement of BRCA2 and putative BRCA3 in colon tumorigenesis in right-sided, replication error–negative, Dukes’ stage C cancers. Further studies are needed to confirm the precise role of these genes, and any prognostic significance, in colon cancer.
Key Words: BRCA2 • BRCA3 • Colon cancer • Allelic imbalance • Loss of heterozygosity
|
INTRODUCTION
|
|---|
Colorectal cancer arises from the accumulation of genetic events and is usually classified as hereditary or sporadic.1–3 Tomlinson et al.4 reviewed and described three genetic pathways for colorectal cancer formation: the classical, the alternative, and the ulcerative colitis–associated pathways. Briefly, the classical pathway involves chromosomal abnormalities with loss of tumor-suppressor genes and gain of oncogenes; the alternative pathway is characterized by defective mismatch-repair genes resulting in uncorrected nucleotide sequences and replication error (RER) positivity; and ulcerative colitis–associated colorectal cancer arises from areas of flat dysplasia, mostly in the left colon, carrying p53 mutations at an earlier stage of tumorigenesis than that described in sporadic colon cancer.4
There are well-documented reports of loss of heterozygosity at specific sites on chromosomes 5q, 17p, and 18q in colorectal cancer; this suggests the presence of specific tumor-suppressor genes implicated in the progression of this disease.2,3 Vogelstein’s model of the classical pathway hypothesizes the accumulation of genetic events involving, among others, the adenomatous polyposis coli gene on 5q, p53, on 17p and the DCC/SMAD region on 18q.5 Previous studies have also reported an involvement of chromosome 13 in colorectal cancer,6–8 but early studies were performed on mixed series that did not distinguish tumor site, stage, or RER status. Chromosome 13 has been implicated both in progression from adenoma to carcinoma9 and in the metastatic spread to the liver.10 We have previously used a series of right-sided, Dukes’ stage C, RER-negative colon carcinomas to identify regions of allelic imbalance (AI) on chromosome 13 in this homogeneous subset.11
Chromosome 13 is home to a number of genes, including BRCA2 and the putative ‘‘BRCA3’’ gene. BRCA2 gene mutation carriers are found to have an increased risk of developing breast and ovarian cancer in women and breast and prostate cancer in men.12 The Breast Cancer Linkage Consortium showed that in hereditary breast cancer families, 52% were linked to BRCA1 and 32% to BRCA2.13,14 Later studies have shown the linkage attributed to these two genes to be smaller: 10% to 23% of families to BRCA1 and approximately 11% to BRCA2.13,15
BRCA genes are involved in transcriptional regulation of DNA repair, because loss of BRCA1 or BRCA2 is associated with a proliferation of breast and ovarian epithelium.16 It would logically follow that disruption of these genes in other somatic tissues may result in a similar increase in cell proliferation. Indeed, Gorgoulis et al.17 demonstrated AI at marker D13S171 (located within BRCA2) in 70% of non–small-cell lung carcinomas. BRCA2 consists of more than 11,385 base pairs encoding a protein of 3,418 amino acids. To date, more than 450 mutations have been identified in BRCA2, and microdeletions predominate over microinsertions and point mutations.18 The BRCA2 gene promotes homologous recombination,19,20 a major pathway of DNA double strand break repair. It regulates the actions of RAD51, a protein essential for DNA repair, by regulating both its intracellular localization and DNA-binding ability.21,22
In the hunt for further breast cancer genes, Kainu et al.13 performed comparative genomic hybridization studies on tumors from non-BRCA1/2 breast cancer families which they then followed by target linkage analysis and detected somatic deletions in the wild-type gene. This study indicated that loss of 13q was one of the earliest events in hereditary breast cancer, and, on analysis of a well-characterized Swedish breast cancer family, five out of five breast tumors showed a deletion at 13q21–22. A new locus was identified corresponding to a marker at 13q21 (D13S1308); this was estimated to be located at a recombination fraction of.25 from BRCA2, and the putative cancer gene was called BRCAX.13 The involvement of this locus in breast cancer susceptibility was refuted by Thompson et al.,23 who named this locus BRCA3. However, the existence of a putative tumor suppressor gene at D13S1308 (BRCA3) has been proposed more recently in sporadic breast cancers.24 The importance of BRCA2 and the putative BRCA3 gene has not been assessed in sporadic colon cancer.
Because right-sided colon carcinomas behave differently from those that arise in the left side in terms of presentation, prevalence, RER positivity, and prognosis,25–27 the possibility of distinct genetic mechanisms has been hypothesized. Putative markers should therefore be examined in colorectal cancer series with known site, stage, and RER status. Archival specimens of Dukes’ stage C, RER-negative, right-sided colon cancer patients were selected for AI studies to analyze the involvement of the BRCA2 and BRCA3 genes in a homogeneous group of sporadic colon cancers.
|
MATERIALS AND METHODS
|
|---|
Samples
The study was approved by the Hull and East Riding Local Research Ethics Committee. This pilot study used a cohort of 23 RER-negative archival specimens. The series was homogeneous because all samples were right sided and histopathologically proven Dukes’ stage C tumors. All 23 tumors were confirmed as being RER negative by using both a polymerase chain reaction–based microsatellite instability assay and immunohistochemistry with antibodies against hMLH1 and hMSH2.28 For each patient, two specimens were obtained: a sample of normal colon taken from the resection margin and a sample of the primary tumor. The tumor specimens were microdissected and contained at least 50% malignant cells.
DNA was extracted using a standard proteinase K kit (Nucleon; Tepnel, Warrington, UK) according to the manufacturer’s protocol. Seven polymorphic microsatellite markers around the regions of BRCA2 and BRCA3 were selected from human genome databases according to map position, heterozygosity, and amplicon size range (Table 1
; Fig. 1). All microsatellite markers were obtained from Applied Biosystems (Warrington, UK).
Received for publication June 10, 2005.
Accepted for publication November 18, 2005.
TOP
ABSTRACT
INTRODUCTION
.50 was considered to represent AI. 
