Cyclooxygenase (COX) enzymes catalyze the rate-limiting step in the conversion of arachidonic acid to prostaglandins. While COX-1 is constitutively expressed in most tissues and is involved in tissue homeostasis, COX-2 is highly inducible, is frequently overexpressed in cancers, and plays roles in inflammation, immune function, angiogenesis, and tumor invasiveness. Both general COX inhibitors, as well as COX-2-specific inhibitors, have been shown to inhibit carcinogenesis in a number of animal models. However, COX inhibitors have been shown to have COX-independent effects. So we have used both COX-2 transgenic and COX-2 knockout mice to determine the role of COX-2 in carcinogenesis using the mouse skin model. FVB mice overexpressing COX-2 in the skin via the keratin 14 (K14) promoter were initially shown to be resistant to skin tumorigenesis induced by 7,12-dimethylbenz[a]anthracene (DMBA) initiation and 12-O-tetradecanoylphorbol-13-acetate (TPA) promotion. However, using DMBA only or using the nonphorbol ester tumor promoter, anthralin, with DMBA, K14.COX2 transgenic mice developed many more tumors (3.7 and 5.5 fold, respectively) than wild type FVB mice. Thus, overexpression of COX-2 appears to promote skin tumorigenesis and resistance to tumorigenesis is limited specifically to TPA tumor promotion, which correlated to a reduced TPA induction of proliferation, ornithine decarboxylase, interleukin-1α, and tumor necrosis factor-α expression in transgenic versus wild type mice. Similarly, when SKH-1 hairless mice overexpressing COX-2 in the skin via the keratin 5 (K5) promoter were subjected to ultraviolet (UV) light-induced tumorigenesis, transgenic mice developed 2.5 fold more tumors/mouse than wild type SKH-1 mice. To complement these studies with transgenic overexpressing mice, COX-1 and COX-2 knockout mice were backcrossed onto the SKH-1 background for UV carcinogenesis studies (others have already shown that COX-1 and COX-2 knockout mice are resistant to DMBA/TPA skin tumorigenesis; Cancer Res., 62:3395, 2002). However, since homozygous COX-2 -/- mice do not survive beyond ~6 weeks of age on the SKH-1 background, heterozygous COX-1 +/- and COX-2 +/- mice were used. While COX-1 +/- mice developed UV-induced tumors similarly to wild type SKH-1 mice, COX-2 +/- mice developed 3 fold fewer tumors/mouse with a 10-week increase in latency compared to wild type mice. Thus, loss of only one allele of COX-2 was enough to decrease tumorigenesis in this system. Overall, mouse skin COX-2 expression levels were directly correlated to susceptibility to tumorigenesis. Supported by NIH grants CA100140, ES07784.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA