Is There a Role for Cyclin D1 in Ulcerative Colitis-Related Neoplasia?

Ulcerative colitis (UC), a form of inflammatory bowel disease (IBD), is associated with a significantly increased risk of colorectal cancer development. The specific cause of this is unknown. Cyclin D1 (CCND1) plays a significant role in cell growth regulation by enabling the cell-cycle progression through the G1/S checkpoint. A common single nucleotide polymorphism (SNP) exists within exon 4 of CCND1. This SNP (A870G) correlates with altered alternative mRNA splicing, with exon 5 excluded from the polymorphic transcript. Exon 5 is critical in regulating the stability of the CCND1 protein, and its exclusion enhances CCND1 protein stability. The mutant A870G SNP is associated with the development of a variety of tumors including nasopharangeal and colorectal cancers. The goal of this study was to define associations between the CCND1 A870G SNP and UC-related cancer. Our preliminary data from U95A Affymetrixã cDNA microarrays experiments showed 3 times greater CCND1 expression in dysplastic UC colonic mucosa vs. non-dysplastic UC, and 6.5 times greater expression in UC-cancer vs. non-dysplastic UC. These data emphasize a significant differential expression of CCND1 in dysplastic/cancerous colonic mucosa as compared to benign tissue in UC. We hypothesized that the presence of the A870G polymorphism promotes neoplastic progression in colonic UC through abnormal cell-cycle stimulation secondary to enhanced CCND1 stability. We performed a case-case/case-control study to correlate the presence of the A870G SNP with UC-associated dysplasia and colorectal cancer. The study population consisted of 232 non-Jewish Caucasian individuals with non-dysplastic UC (n=79), UC-dysplasia (n=15), UC-cancer (n=9), and sporadic colorectal cancer (n=61), as well as 68 unrelated controls. The A870G polymorphism was genotyped from germ-line DNA using SNP-specific polymerase chain reactions (PCR) and Taqman allelic discrimination. CCND1 A870G allele and genotype frequencies were compared using chi-squared analyses. Comparisons were performed between disease and control populations, as well as between non-dysplastic UC and UC-dysplasia/UC-cancer populations. These comparisons did not reveal a significant association of A870G allele and/or genotype frequencies with disease status, thus disproving our hypothesis. However, our data do reveal some interesting trends, with the frequency of the homozygote mutants being higher in the UC-dysplasia (5/15 [33%]) and lower in the UC-cancer (1/9 [11%]) populations when compared to both the non-dysplastic UC (16/79 [20%]) and control (16/68 [24%]) populations. However, we must conclude that this current study lacks statistical power due to low numbers in both the UCdysplasia and UC-cancer groups. In summary, this current study does not reveal a significant correlation between the CCND1 A870G polymorphism and UC-associated cancer. However, these results show a trend towards differing CCND1 genotype distributions in UC-associated cancer. This trend, in addition to our microarray results, implies a continued requirement for investigation of the role of Cyclin D1 in UCrelated cancer