Sterol O-Acyltransferase 2-Driven Cholesterol Esterification Opposes Liver X Receptor-Stimulated Fecal Neutral Sterol Loss

Abstract

Statin drugs have proven a successful and relatively safe therapy for the treatment of atherosclerotic cardiovascular disease (CVD). However, even with the substantial low-density lipoprotein (LDL) cholesterol lowering achieved with statin treatment, CVD remains the top cause of death in developed countries. Selective inhibitors of the cholesterol esterifying enzyme sterol-O acyltransferase 2 (SOAT2) hold great promise as effective CVD therapeutics. In mouse models, previous work has demonstrated that either antisense oligonucleotide (ASO) or small molecule inhibitors of SOAT2 can effectively reduce CVD progression, and even promote regression of established CVD. Although it is well known that SOAT2-driven cholesterol esterification can alter both the packaging and retention of atherogenic apoB-containing lipoproteins, here we set out to determine whether SOAT2-driven cholesterol esterification can also impact basal and liver X receptor (LXR)-stimulated fecal neutral sterol loss. These studies demonstrate that SOAT2 is a negative regulator of LXR-stimulated fecal neutral sterol loss in mice.

Keywords: Atherosclerosis; Cholesterol; Esterification; Intestine.

Fig. 1

Global deficiency of SOAT2 enhances LXR-stimulated fecal neutral sterol loss. Male wild type (SOAT2^+/+) and SOAT2 total body knockout mice (SOAT2^−/−) were fed a standard chow diet and orally gavaged either with vehicle (VEH) or the LXR agonist T0901317 (T) for 7 days. a Mass fecal neutral sterol loss was determined by gas–liquid chromatography. b Fractional cholesterol absorption was measured using the fecal dual-isotope method. Data represent mean ± S.E from 12 to 14 mice per group and values not sharing a common superscript differ significantly (p ≤ 0.05)

Fig. 2

Intestine or liver specific deletion of SOAT2 is not sufficient to enhance LXR-stimulated fecal neutral sterol loss. Male control mice with two SOAT2 floxed alleles (fl/fl), enterocyte-specific SOAT2 knockout mice (SI⁻/⁻), or hepatocyte-specific SOAT2 knockout mice (L⁻/⁻) were fed a standard chow diet, and were either treated with vehicle (VEH) or the LXR agonist T0901317 (T) for 7 days. a Mass fecal neutral sterol loss was determined by gas–liquid chromatography. b Fractional cholesterol absorption was measured using the fecal dual-isotope method. Data represent mean ± S.E from 9 to 11 mice per group and values not sharing a common superscript differ significantly (p ≤ 0.05)

Fig. 3

The ability of SOAT2 to suppress LXR-stimulated fecal neutral sterol loss relies in part on LXR-stimulated biliary cholesterol secretion. Male wild type (SOAT2^+/+) or global SOAT2 knockout mice (SOAT2^−/−), were studies in the presence of absence of hepatocyte-specific overexpression of NPC1L1 (NPC1L1^Tg) to blunt biliary cholesterol loss. These four genotypes of mice were maintained on a standard chow diet, and were either treated with vehicle (VEH) or the LXR agonist T0901317 (T) for 7 days. a Mass fecal neutral sterol loss was determined by gas–liquid chromatography. b Fractional cholesterol absorption was measured using the fecal dual-isotope method. c Gall bladder bile was collected and analyzed for biliary cholesterol by gas–liquid chromatography. Data represent mean ± S.E from 5 to 14 mice per group and values not sharing a common superscript differ significantly (p ≤ 0.05)

Fig. 4

Expression of cholesterol-regulated gene expression in the liver and small intestine (jejunum). Male wild type (SOAT2^+/+) and SOAT2 total body knockout mice (SOAT2^−/−) were fed a standard chow diet and orally gavaged either with vehicle (VEH) or the LXR agonist T0901317 (T) for 7 days. The relative levels of mRNA for ATP binding cassette transporter A1 (ABCA1), ATP binding cassette transporter G5 (ABCG5), 3-hydroxy-3methylgltaryl-CoA synthase (HMGCS), and LDL receptor (LDLr) were quantified by quantitative real time PCR in the liver (panels a–d) and proximal small intestine (panels e–h). Data represent mean ± S.E from four mice per group and values not sharing a common superscript differ significantly (p ≤ 0.05)