Objective-We determined the effects of sustained normocholesterolemia on advanced mouse atherosclerosis and whether changes in plaque size and composition can be detected noninvasively by MRI. Methods and Results-Aortic arch segments containing advanced lesions from apolipoprotein E-deficient (apoEϪ/Ϫ) mice (total cholesterol 1281Ϯ97 mg/dL) were transplanted into syngeneic wild-type (WT; 111Ϯ11 mg/dL) or apoEϪ/Ϫ (702Ϯ74 mg/dL) recipient mice on chow diet. Mice underwent serial MRI at 3, 5, 7, and 9 weeks after transplantation. Compared with 3 weeks, correction of dyslipidemia in WT recipient mice resulted in a monotonic decrease (regression) in arterial wall volume, whereas in apoEϪ/Ϫ recipient mice, further plaque progression was noted (PϽ0.05). MRI and histological measurements were closely correlated (Rϭ0.937). The lesional content of macrophages decreased Ͼ90% (PϽ0.001), and smooth muscle cells increased in the WT recipient mice. In vivo T 1 -, T 2 -, and proton density-weighted images of the mouse thoracic aorta differentiated intraplaque lipid and collagen. Conclusions-Plaque changes can be noninvasively monitored by serial in vivo MRI of a mouse regression model. Our ability to image the thoracic aorta and perform in vivo plaque characterization will further enhance atherosclerosis studies. Key Words: atherosclerosis Ⅲ imaging Ⅲ lipoproteins Ⅲ remodeling Ⅲ transplantation P laque regression in mouse models of atherosclerosis has previously been demonstrated primarily by somatic adenoviral gene transfer. 1,2 Such approaches have been limited because of the eventual loss of transgene expression, even with second-generation viral vectors, likely caused by immune responses directed against both the transgene product and adenoviral proteins. 3,4 In addition, regression has typically been inferred from ex vivo histological tissue sampling of numerous animals at multiple time intervals after treatment or perturbation. It would be of obvious advantage to be able to use in vivo MRI to noninvasively monitor, quantify, and characterize plaque in a model where the regressive conditions are constant and sustained indefinitely.Mice have become widely used as models of human atherosclerosis because they offer the advantages of ample progeny, a well-characterized genome similar to that of humans, and the relative ease of genetic manipulations. Apolipoprotein E-deficient (apoEϪ/Ϫ) mice spontaneously develop lesions which resemble those found in humans. 5,6 However, unlike in large animals, in apoEϪ/Ϫ mice neither dietary manipulation nor treatment with statins 7 corrects hyperlipidemia. Tsukamoto et al 8 used adenoviral expression of an apoE transgene in these mice to promote the clearance of the atherogenic lipoprotein particles and correct the dyslipidemia. The expression of the transgene, however, was short-lived, lasting Ϸ3 months. In a modified approach, Desurmont et al 1 used immunocompromised mice in a nude background to prolong the expression of adenoviralexpressed apoE; however, the complete lack of mature cytot...