Investigations into the mechanisms by which diabetes accelerates atherosclerosis have been hampered by the lack of suitable animal models. We hypothesized that streptozotocin-treated LDL receptor-deficient mice would be a good model of diabetic atherosclerosis because streptozotocin causes diabetes in the parent C57BL/6 strain and because in these mice diet-induced hypercholesterolemia leads to the formation of advanced atherosclerotic lesions throughout the aorta. Diabetes was induced in 18 mice by intraperitoneal injection of streptozotocin. Low-dose insulin was given subcutaneously to prevent excessive mortality and extreme elevations in triglyceride levels. The control group was subjected to sham injections. Both groups were fed a diet containing .075% cholesterol for six months. Average blood glucose was higher in the diabetic group than in the control group (257 +/- 67 mg/dL versus 111 +/- 7 mg/dL, P < 0.05). Although plasma cholesterol was similar (966 +/- 399 versus 1002 +/- 180 mg/dL) in both groups, VLDL cholesterol was higher whereas LDL cholesterol was lower in the diabetic group. Immunocytochemical analysis demonstrated significantly more advanced glycation end-product (AGE) epitopes in the artery wall of the diabetic group, whereas staining for oxidation-specific epitopes was similar in both groups. Sera of diabetic mice also contained significantly more IgG autoantibodies that bound to several AGE epitopes than did sera from control mice. Despite the presence of hyperglycemia, diabetic dyslipidemia, and enhanced AGE formation in the diabetic mice, both groups had a similar extent of atherosclerosis (diabetic, 17.3 +/- 5.2; control, 16.5 +/- 6.6% of the aortic surface). These data suggest that, at least under conditions of marked hypercholesterolemia; hyperglycemia and enhanced AGE formation do not contribute significantly to atherogenesis in LDL-/- mice.

The role of insulin resistance (IR) in atherogenesis is poorly understood, in part because of a lack of appropriate animal models. We assumed that fructose-fed LDL receptor-deficient (LDLR-/-) mice might be a model of IR and atherosclerosis because (1) fructose feeding induces hyperinsulinemia and IR in rats; (2) a preliminary experiment showed that fructose feeding markedly increases plasma cholesterol levels in LDLR-/- mice; and (3) hypercholesterolemic LDLR-/- mice develop extensive atherosclerosis. To test whether IR could be induced in LDLR-/- mice, 3 groups of male mice were fed a fructose-rich diet (60% of total calories; n=16), a fat-enriched (Western) diet intended to yield the same plasma cholesterol levels (n=18), or regular chow (n=7) for approximately 5.5 months. The average cholesterol levels of both hypercholesterolemic groups were similar (849+/-268 versus 964+/-234 mg/dL) and much higher than in the chow-fed group (249+/-21 mg/dL). Final body weights in the Western diet group were higher (39+/-6.2 g) than in the fructose- (27.8+/-2.7 g) or chow-fed (26.7+/-3.8 g) groups. Contrary to expectation, IR was induced in mice fed the Western diet, but not in fructose-fed mice. The Western diet group had higher average glucose levels (187+/-16 versus 159+/-12 mg/dL) and 4.5-fold higher plasma insulin levels. Surprisingly, the non-insulin-resistant, fructose-fed mice had significantly more atherosclerosis than the insulin-resistant mice fed Western diet (11.8+/-2.9% versus 7.8+/-2. 5% of aortic surface; P<0.01). These results suggest that (1) fructose-enriched diets do not induce IR in LDLR-/- mice; (2) the Western diets commonly used in LDLR-/- mice may not only induce atherosclerosis, but also IR, potentially complicating the interpretation of results; and (3) IR and hyperinsulinemia do not enhance atherosclerosis in LDLR-/- mice, at least under conditions of very high plasma cholesterol levels. The fact that various levels of hypercholesterolemia can be induced in LDLR-/- mice by fat-enriched diets and that such diets induce IR and hyperinsulinemia suggest that LDLR-/- mice may be used as models to elucidate the effect of IR on atherosclerosis, eg, by feeding them Western diets with or without insulin-sensitizing agents.

Apolipoprotein (apo) E-deficient mice develop atherosclerotic lesions that contain epitopes formed during the oxidative modification of lipoproteins, and they demonstrate high titers of circulating autoantibodies against such epitopes, suggesting that this murine strain may provide a model to investigate the atherogenic mechanisms of oxidized lipoproteins (Palinski et al, Arterioscler Thromb. 1994; 14:605-616). To test the hypothesis that lipoprotein oxidation contributes to lesion formation in apoE-deficient mice, we studied the effect of the antioxidant N,N'-diphenyl 1,4-phenylenediamine (DPPD) in mice fed a high-fat diet containing 0.15% cholesterol. Animals were divided into two subgroups matched for sex and plasma cholesterol levels, and DPPD (0.5% wt/wt) was added to the diet of one subgroup. Throughout the 6 months of intervention, DPPD treatment had no significant effect on plasma cholesterol. Plasma levels of DPPD at the end of the experiment were 33.1 mumol/L. As judged by resistance to loss of polyunsaturated fatty acids, lipoproteins (d < 1.019 g/mL) from DPPD-treated animals showed greater resistance to copper-induced oxidation than lipoproteins from control animals. In addition, there was a greater than twofold prolongation of the lag time in the formation of conjugated dienes in the LDL and IDL fractions of DPPD-treated mice. Atherosclerosis was significantly reduced, by 36% in the DPPD-treated mice (14.0 +/- 4.53% of aortic surface area versus 21.9 +/- 11.6%; n = 32; P < .02). These results are consistent with the hypothesis that lipoprotein oxidation contributes to atherogenesis in apoE-deficient mice. However, further studies with other antioxidants are needed to validate this hypothesis.

Maternal hypercholesterolemia during pregnancy is associated with enhanced fatty streak formation in human fetuses and faster progression of atherosclerosis during childhood even under normocholesterolemic conditions. A causal role of maternal hypercholesterolemia in lesion formation during fetal development has previously been established in rabbits. The same experimental model is now used to establish that maternal hypercholesterolemia or ensuing pathogenic events in fetal arteries enhance atherogenesis later in life. Five groups of rabbit mothers were fed chow, cholesterol-enriched chow, or cholesterol-enriched chow plus 1000 IU vitamin E, 3% cholestyramine, or both during pregnancy. Offspring of all groups (n=136) were fed a mildly hypercholesterolemic diet for up to a year and had similar cholesterol levels. Aortic lesion sizes and lipid peroxidation products in plasma and lesions in offspring were determined at birth, 6 months, or 12 months. Lesion progression in offspring of hypercholesterolemic mothers was greater than in all other groups. At each time point, offspring of hypercholesterolemic mothers had 1.5- to 3-fold larger lesions than offspring of normocholesterolemic mothers (P<0.01), with the greatest absolute differences at 12 months. Maternal treatment reduced lesions by 19% to 53%, compared with offspring of untreated hypercholesterolemic mothers (P<0.01), with the greatest effect in the vitamin E groups. At 12 months, lesions in offspring of all vitamin E and cholestyramine-treated mothers were similar to those of normocholesterolemic mothers. Lipid peroxidation end-products in lesions and plasma showed analogous differences between groups as lesions (P<0.01). Thus, pathogenic programming in utero increases the susceptibility to atherogenic risk factors later in life and maternal intervention with cholesterol-lowering drugs or antioxidants reduce postnatal lipid peroxidation and atherosclerosis in their offspring.