As the primary model we selected LDL receptor knockout mice
As the primary model we selected LDL-receptor knockout mice because these resemble the human plasma lipoprotein profile more closely. Currently, genetic models with reduced milk cholesterol content, which mimic cholesterol-free formula feeding, are unavailable. Thus, instead of modifying milk cholesterol content we decided to reduce its dietary availability by blocking intestinal cholesterol JP 1302 dihydrochloride in the offspring with ezetimibe. Ezetimibe efficiently interacts with the key intestinal cholesterol absorption transporter Npc1l1 to block cholesterol uptake into enterocytes., Because ezetimibe is excreted readily into mothers\' milk, we supplemented the food of dams for 3 weeks between delivery and weaning with ezetimibe to deliver it via this route to the pups\' intestine and thereby specifically decrease intestinal cholesterol absorption from milk (experimental design summarized in ). Although ezetimibe treatment of dams did not affect milk cholesterol levels (), consistent with our hypothesis, fractional cholesterol absorption was reduced substantially in 3-week-old pups receiving milk from ezetimibe-administered dams (), although Npc1l1 expression was unchanged (). Decreased dietary cholesterol availability resulted in substantially reduced total plasma cholesterol levels () exclusively within (V) LDL lipoproteins (). These physiological changes in the pups resemble the response to cholesterol-free formula feeding in human infants, which also is associated with lower plasma total and LDL-C. In human beings, differences in plasma cholesterol between previously breast- and formula-fed individuals are not detectable anymore in adolescence, however, recur in adulthood in an opposite fashion, with individuals previously fed formula then showing increased LDL-C. Therefore, we also followed up offspring mice until 24 weeks of age (ie, well into adult life). The initial reduction in plasma cholesterol levels observed at 3 weeks was somewhat preserved in the offspring of ezetimibe-treated dams until early adult age (12 weeks) ( and ), whereas at 24 weeks the total plasma cholesterol levels and lipoprotein profiles were comparable ( and ). In addition, at 24 weeks there were no appreciable differences in either hepatic cholesterol content, biliary secretion of cholesterol and bile acids, fecal excretion of neutral sterols and bile acids, or composition of the bile acid pool (). Further characterization of 24-week-old offspring showed a markedly decreased cholesterol absorption in post ezetimibe animals (-27%; < .001) (), conceivably owing to decreased messenger RNA (mRNA) (-51%; < .05) () and protein expression () in the proximal small intestine. At the same time, intestinal mRNA expression of was increased significantly in the postezetimibe group (), likely representing a compensatory mechanism that could explain the comparable plasma cholesterol levels between the groups at 24 weeks of age. Decreased Npc1l1 expression can occur either via direct down-regulation of Npc1l1, or indirectly via suppression of key transcriptional regulators. Reduced expression has been reported in response to Lxr-α activation. However, neither itself nor known target genes, such as the cholesterol transporters , , and , or the hepatic bile acid synthesis enzyme , were changed significantly in our model ( and ). Other negative transcriptional regulators of are Pparα and Pparδ. However, mRNA expression of bona fide intestinal transcriptional targets of Pparα and Pparδ also were unaltered (), suggesting that lower Npc1l1 expression in postezetimibe animals reflects a direct transcriptional repression, likely by epigenetic mechanisms. It recently was suggested that differential hypermethylation of distinct CpG positions in the Npc1l1 promoter might explain the known proximal-to-distal expression gradient of Npc1l1 throughout the intestinal tract with virtually absent Npc1l1 in the colon. Consistent with these findings, we showed using pyrosequencing significantly increased promoter methylation in the large intestine compared with proximal parts (). However, no significant changes in promoter methylation were detected in the proximal intestine between the postezetimibe and the control group, indicating that simple alterations in methylation likely do not explain the observed differences in Npc1l1 expression. Besides DNA methylation, histone modifications represent an important epigenetic regulation system; therefore, we tested the upstream promoter for enrichment of dimethylation and trimethylation of H3K9 and trimethylation of H3K27, histone marks associated with heterochromatization and gene silencing. Interestingly, we observed a marked increase of H3K9me3 in the region encompassing -423 to -607 bp upstream of the transcription start site in 24-week-old postezetimibe offspring compared with controls ( < .05) (). H3K9me2 and H3K27me3 enrichments in the same region also were increased, however, this increase was not statistically significant.