Epidemiological studies and laboratory animal model assays suggest that a h
igh intake of dietary fat promotes colorectal cancer. Several in vivo and i
n vitro studies support the hypothesis that omega -6 fatty acids promote co
lon tumorigenesis, whereas omega -3 fatty acids lack promoting activity. Fa
t intake in the United States traditionally includes high amounts (30% of t
otal caloric intake) of saturated fat rather than omega -6 fatty acids. The
refore, the present study was designed to compare the modulatory effects of
a high-fat diet containing mixed lipids (HFML), a diet rich in saturated f
atty acids (the average American diet), a diet with fish oil (HFFO) that is
rich in omega -3 fatty acids, and a low-fat corn oil diet (LFCO) on the fo
rmation of chemically Induced colonic aberrant crypt loci (ACF) and tumors,
cyclooxygenase (COX)-2 activity, and apoptosis during experimental colon c
arcinogenesis, At 5 weeks of age, groups of male F344 rats were fed a 5% co
rn oil diet (LFCO). At 7 weeks of age, rats intended for carcinogen treatme
nt received s.c. injections of azoxymethane at a dose level of 15 mg/kg of
body weight once weekly for 2 weeks. Beginning 1 day after the carcinogen t
reatment, groups of rats were then maintained on experimental diets contain
ing 20% HFML or 20% HFFO. Rats were killed at 8, 23, or 38 weeks after azox
ymethane treatment. Colonic ICF and tumors were evaluated histopathological
ly. and apoptosis was evaluated by the terminal deoxynucleotidyl transferas
e-mediated nick end labeling method. Colonic mucosae and tumor samples harv
ested at week 38 were analyzed for COX-2 synthetic activity and expression.
The rats fed the HFML diet showed significantly increased total colonic AC
F (P < 0.001-0.0001) with a multiplicity of <greater than or equal to>4 abe
rrant crypts/focus (P < 0.0001) compared with the effects of the HFFO or LF
CO diets at week 8, 23, and 38. Interestingly, there was a 2- to 3-fold inc
rease (<greater than or equal to>4) in multicrypt foci in rats given the HF
ML diet as compared with such foci in rats fed the HFFO or LFCO diets. By w
eek 23, the HFML, diet had significantly increased the incidence of colonic
tumors (30-60%), and their multiplicity (100-141%) when compared with the
effects of the LFCO or HFFO diets, At week 38, the HFML diet had induced 10
0% colon tumor incidence and a 4-fold multiplicity of adenocarcinomas compa
red with the LFCO and HFFO diets. At weeks 23 and 38, a significantly lower
percentage of apoptotic colonic epithelial cells were observed in the tumo
rs of animals fed the HFML diet as compared with those fed the HFFO diet. T
he HFML diet caused significantly increased levels of COX-2 activity in col
on tumors (P < 0.05-0.01). and these tumors had enhanced levels of COX-2 ex
pression as compared with those in assays with LFCO or HFFO diets. These ob
servations demonstrate for the first time that HFML diets containing high l
evels of saturated fatty acids (such as those in Western diets) promote col
on carcinogenesis. Although the mechanisms involved in colon tumor promotio
n by a HFML diet are not fully known. our results indicate that the modulat
ion of eicosanoid production via the influence on COX activity and the supp
ression of apoptosis may play a key role in HFML diet-induced colon tumorig
enesis.