Based on the biological activity of arachidonic acid metabolites, we hypothesized that alterations in the consumption of linoleic acid, the precursor to arachidonic acid, would result in a modification in tumor development when fed during the tumor promotion stage of the mouse skin initiation-promotion model. The effects of seven different levels of dietary linoleic acid (LA), supplied as corn oil in a 15% fat diet, on the incidence and rate of papilloma and carcinoma development were determined. SENCAR mice were placed on one of the experimental diets, containing 1.0, 3.6, 6.0, 7.9, 9.9, 12.5, or 15.0% corn oil, 1 week after initiation with 10 nmol of 7,12-dimethylbenz(a)anthracene and 3 weeks prior to the start of twice weekly promotion with 1 micrograms 12-O-tetradecanoylphorbol-13-acetate (TPA). At 15 weeks of TPA treatment there were significant differences in papilloma number among diet groups, such that an inverse correlation (r = 0.92) was observed between tumor number and level of corn oil; the lowest corn oil diet group had an average of 11.7 tumors/mouse, while the highest corn oil group had 5.4 tumors/mouse. However, there was little difference in tumor incidence among diet groups. A general relationship between diet and carcinoma incidence was also found, such that the highest corn oil diet group had the lowest carcinoma incidence. In an experiment performed with DBA/2 mice, the average number of papillomas/mouse at 17 weeks was 4.5 (1.0% corn oil), 5.6 (7.9%) corn oil), and 2.3 (15.0% corn oil). Papilloma incidence was also affected by diet, with a 79% incidence for the 15.0% corn oil and an incidence of 93% for the 1.0% corn oil group. analyses of the fatty acid composition of epidermal phospholipids in mice fed the experimental diets reflected the dietary LA levels, in that an accumulation of phospholipid LA, accompanied by an overall decrease in arachidonic acid, occurred with increasing dietary corn oil. In spite of the high membrane levels of LA, no measurable amount of epidermal conjugated dienes of LA could be detected. Epidermal prostaglandin E2 levels in acetone-treated mice were similar for all diet groups (approximately 3 pg/micrograms DNA). However, 6 h after topical application with 4 micrograms of TPA, prostaglandin E2 levels were elevated 5- to 10-fold; an inverse correlation (P less than 0.05) was seen with increasing dietary LA, although the concordance with decreased phospholipid arachidonic acid was not strong.(ABSTRACT TRUNCATED AT 400 WORDS)