Cyclooxygenase/lipoxygenase shunting lowers the anti-cancer effect of cyclooxygenase-2 inhibition in colorectal cancer cells

Background: Arachidonic acid metabolite, generated by cyclooxygenase (COX), is implicated in the colorectal cancer (CRC) pathogenesis. Inhibiting COX may therefore have anti-carcinogenic effects. Results from use of non-steroidal anti-inflammatory drugs inhibiting only COX have been conflicting. It has been postulated that this might result from the shunting of arachidonic acid metabolism to the 5-lipoxygenase (5-LOX) pathway. Cancer cell viability is promoted by 5-LOX through several mechanisms that are similar to those of cyclooxygenase-2 (COX-2). Expression of 5-LOX is upregulated in colorectal adenoma and cancer. The aim of this study was to investigate the shunting of arachidonic acid metabolism to the 5-LOX pathway by cyclooxygenase inhibition and to determine if this process antagonizes the anti-cancer effect in colorectal cancer cells. Methods: Three colorectal cancer cell lines (HCA7, HT-29 & LoVo) expressing 5-LOX and different levels of COX-2 expression were used. The effects of aspirin (a non-selective COX inhibitor) and rofecoxib (COX-2 selective) on prostaglandin E-2 (PGE(2)) and leukotriene B-4 (LTB4) secretion were quantified by ELISA. Proliferation and viability were studied by quantifying double-stranded DNA (dsDNA) content and metabolic activity. Apoptosis was determined by annexin V and propidium iodide staining using confocal microscopy, and caspase-3/7 activity by fluorescent substrate assay. Results: COX inhibitors suppressed PGE(2) production but enhanced LTB4 secretion in COX-2 expressing cell lines (P < 0.001). The level of COX-2 expression in colorectal cancer cells did not significantly influence the anti-proliferative and pro-apoptotic effects of COX inhibitors due to the shunting mechanism. Conclusions: This study provides evidence of shunting between COX and 5-LOX pathways in the presence of unilateral inhibition, and may explain the conflicting anti-carcinogenic effects reported with use of COX inhibitors.