Effects of dietary fish oil on fatty acids and eicosanoids in metastasizing human breast cancer cells.

We examined the relationships between the suppressive effects of dietary fish oil on growth and metastasis of MDA-MB-435 human breast cancer cells in female nude mice and the primary tumor phospholipid fatty acid concentrations, phospholipase A2 activity, and eicosanoid levels. Mice (n = 120) were fed a 23% (wt/wt) corn oil (CO) linoleic acid (LA)-rich diet for seven days before and after 10(6) tumor cells were injected into a mammary fat pad, and then the mice receive one of three isocaloric diets containing 23% total fat but different proportions of CO and menhaden oil (MO) (18% CO-5% MO, 11.5% CO-11.5% MO, 5% CO-18% MO) or a 23% fat diet containing 18% deodorized fish oil supplemented with tocopherol and tert-butylhydroquinone antioxidants (FAO). Primary tumor growth rate was significantly greater in mice fed the 18% CO diet than in the three diets containing higher levels of fish oil (all p < 0.05). The 18% MO diet, but not the 11.5% MO or the 18% FAO diet, suppressed the development of lung metastases compared with the 18% CO diet. Increasing the proportion of MO relative to CO in the diets produced corresponding increases in the primary tumor phospholipid eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) concentrations and reductions in LA and arachidonic acid. There was a significant positive correlation between the LA concentration in these tumors and the extent of lung metastasis (r = 0.504). Tumor phospholipase A2 activity was unaffected by dietary MO intake. Prostaglandin E2 concentration was inversely correlated with phospholipid EPA (r = -0.484) and DHA (r = -0.439), but there was no relationship with lung metastasis. Tumor leukotriene B4 and 5-hydroxyeicosatetraenoic acid levels were not reduced by dietary MO. The 18% FAO- and the 18% MO-fed mice showed similar relationships for the phospholipid fatty acids and prostaglandin E2, despite the lack of effect on metastasis. The strong correlation between phospholipid LA levels and metastasis and the lack of an association with tumor eicosanoids suggest that the 18% MO diet inhibited metastasis because dietary LA was replaced by other fatty acids.