Rapid regression of atherosclerosis induced by liver-directed gene transfer of ApoE in ApoE-deficient mice.

Apolipoprotein E (apoE) is a multifunctional protein synthesized by the liver and tissue macrophages. ApoE-deficient mice have severe hyperlipidemia and develop accelerated atherosclerosis on a chow diet. Both liver-derived and macrophage-derived apoEs have been shown to reduce plasma lipoprotein levels and slow the progression of atherosclerosis in apoE-deficient mice, but regression of atherosclerosis has not been demonstrated in this model. We utilized second-generation adenoviruses to achieve hepatic expression of human apoE in chow-fed, apoE-deficient mice with established atherosclerotic lesions of different stages. As expected, hepatic expression of human apoE3 significantly reduced plasma cholesterol levels. Liver-derived apoE also accumulated substantially within preexisting atherosclerotic lesions, indicating that plasma apoE gained access to the arterial intima. Hepatic expression of human apoE3 for 6 weeks resulted in significant quantitative regression of both early fatty streak lesions as well as advanced, complex lesions in both the aortic root and the aortic arch. In addition, hepatic expression of apoE induced substantial morphological changes in lesions, including decreased foam cells and increased smooth muscle cells and extracellular matrix content. In parallel, human apoE4 and apoE2 were also expressed in the liver by using recombinant adenoviruses. ApoE4 reduced cholesterol levels to the same extent as did apoE3 and also prevented progression but did not induce significant regression of preexisting lesions. ApoE2 reduced cholesterol levels to a lesser degree than did apoE3 and apoE4 and lesion progression was reduced, but regression was not induced. In summary, (1) regression of preexisting atherosclerotic lesions in apoE-deficient mice can be rapidly induced by hepatic expression of apoE, despite the absence of macrophage-derived apoE; (2) the morphological changes seen in this model of regression resemble those in other animal models, induced over longer periods of time; (3) liver-derived apoE gained access to and was retained by intimal atherosclerotic lesions; and (4) apoE4 was less effective in inducing regression, despite its effects on plasma lipoproteins that were similar to those of apoE3. The rapid regression of preexisiting atherosclerotic lesions induced by apoE gene transfer in apoE-deficient mice could provide a convenient murine model for investigation of the molecular events associated with atherosclerosis regression.