753

Sustained expression of COX-2 in the stromal compartment has been shown to play a crucial role in inflammation-associated gastrointestinal tumorigenesis. We recently reported that both zerumbone (ZER), a sesquiterpene occurring in a zingiberaceous plant found in subtropical countries (1), and (±)-13-hydroxy-10-oxo-trans-11-octadecenoic acid (HOA), a linoleate metabolite produced by corn lipoxygenase, exhibited pronounced efficacy for suppressing LPS-induced COX-2 expression in RAW264.7 mouse macrophages. We showed that a topical application of ZER suppressed DMBA/TPA-induced papilloma development and COX-2 protein expression in mouse skin (1), while oral administration attenuated azoxymethane-induced aberrant crypt foci formation and COX-2 induction in rat colons (2). Both ZER and HOA carry an α,β-unsaturated carbonyl moiety that is potentially susceptible to conjugation with biological nucleophiles, implying that each suppresses COX-2 induction through the regulation of redox-sensitive molecules. The chemical and biological natures of this class of compounds have also been shown for isothiocyanates, e.g., sulforaphane. In the present study, we investigated the molecular mechanisms that underlie the suppression of COX-2 by ZER and HOA. Pretreatment of RAW264.7 cells with HOA (20 μM) for 30 minutes markedly suppressed LPS-induced phosphorylation of MAPKKs (MEK1/2 and MKK3/6), MAPKs (p38, ERK1/2, JNK1/2), and Akt, as well as degradation of IκB-α protein and DNA binding of the transcription factors AP-1 and NFκB, along with their transcriptional activities. In contrast, ZER (20 μM) was virtually inactive for attenuating those events, whereas it abolished the phosphorylation of MAPK-activated protein kinase-2 (MK-2), suggesting a direct inhibition of p38 MAPK, while neither HOA or ZER significantly inhibited IκB kinase activity in vitro. In addition, these agents may de-stabilize mRNA of LPS-induced COX-2, iNOS, and IL-6at a post-transcriptional level, since each as well as dexamethasone accelerated the degradation of mRNA from those, as compared with the vehicle-treated control. These results are consistent with our previous results that HOA and ZER disrupted the LPS-induced p38 MAPK signal transduction pathway. Further, our findings are supported by previous reports that have shown a central role of p38 MAPK in stabilizing the mRNA of some proinflammatory genes, including that of COX-2, which contains highly conserved AU-rich elements in the proximal regions of 3’-UTR. Together, our results indicate that 2 electrophilic dietary factors, HOA and ZER, attenuate the expression of COX-2 via distinct and shared action mechanisms. We propose these 2 dietary phytochemicals as reasonable and promising chemopreventive agents for inflammation-associated carcinogenesis. References: (1) Int J Cancer, 110: 481, 2004. (2) Life Sci., 69: 1935, 2001.

[Proc Amer Assoc Cancer Res, Volume 46, 2005]