The effects of genetic and luminal risk factors on cyclooxygenase-2 expression in human colon cancer cells

The process of colon carcinogenesis is dependent upon a variety of genetic mutations and intestinal luminal risk factors. These risk factors cooperate to influence pathophysiological targets directly linked to the carcinogenic process. Cyclooxygenase-2 (COX-2) is one such target, and it is found to be upregulated in 50% of colonic adenomas and 85% of carcinomas. The interaction between genetic mutations and luminal risk factors to affect COX-2 expression was explored in this dissertation work. An activated K-ras oncogene and mutant APC tumor suppressor gene both regulate COX-2 posttranscriptionally. Using colon cancer cells with these genetic alterations, we determined that an activated K-ras stabilizes COX-2 mRNA and works through a PKC-dependent mechanism in a cell line-specific manner. K-ras also affected PTEN expression to potentially suppress cell survival. Wild-type APC downregulated COX-2 protein levels without affecting mRNA. APC interfered with PI-3K-dependent signaling of COX-2, and affected PKC-depending signaling to further modulate COX-2 expression. In addition to genetic alterations, luminal factors obtained from the diet influenced COX-2 expression in these colon cancer cell models. Secondary bile acids like deoxycholic acid (DCA) mediated COX-2 protein stability in addition to its known affects on transcription. DCA's affects on COX-2 protein were partially mediated by APC mutational status and PKC-signaling. Polyamines, another luminal risk factor for colon cancer, regulated COX-2 through the induction of RNA processing by the polyamine-dependent gene, eIF-5A and through a reliance on transport through the cell membrane. The chemopreventive agent, Sulindac, has two metabolites---sulindac sulfone and sulindac sulfide. Previous to this study, little was known about the mechanism of action of the sulfone metabolite. We determined that sulindac sulfone suppressed COX-2 protein levels, thus impinging on this protein's ability to signal downstream genes or to produce prostaglandins. In addition, DFMO, a chemopreventive agent that suppresses polyamine synthesis, significantly induced COX-2 protein levels, predicating the importance of combination chemoprevention to battle COX-2-driven tumorigenesis. The intricate cooperation of genetic mutations and luminal risk factors affecting COX-2 expression and subsequently tumorigenesis, suggest that multiagent interventions may be necessary for successful strategies of colon cancer chemoprevention