The possibility that soy protein containing isoflavones influences the development of experimental atherosclerosis has been investigated in ovariectomized mice heterozygous for the human CETP transgene and for the LDL-receptor null allele (LDLr(+/-) CETP(+/-)). After ovariectomy at 8 wk of age they were fed a fat/cholesterol-rich diet for 19 wk and divided into three experimental groups: dietary unmodified soy protein containing isoflavones (mg/g of diet), either at low-dose (Iso Low, 0.272, n = 25), or at high-dose (Iso High, 0.535, n = 28); and the atherogenic diet containing an isoflavone-depleted alcohol-washed soy protein as a control group (n = 28). Aortic root lipid-stained lesion area (mean microm2 x 10(3) +/- SD) did not differ among Iso Low (12.3 +/- 9.9), Iso High (7.4 +/- 6.4), and controls (10.7 +/- 12.8). Autoantibody titers against plasma oxidized LDL did not differ among the experimental groups. Using the control mice as the reference value (100%), in vitro mouse peritoneal macrophage uptake of labeled acetylated LDL-cholesterol was lower in the Iso High (68%) than in the Iso Low (85%) group. The in vitro percent removal by exogenous HDL of labeled unesterified cholesterol from macrophages previously enriched with human [4- 14C]-cholesteryl oleate acetylated LDL was enhanced in the Iso High group (50%). In spite of these in vitro potentially antiatherogenic actions, soy protein containing isoflavones did not modify the average size of lipid-stained area in the aortic root.