Disruption of Abcg5 and Abcg8 in mice reveals their crucial role in biliary cholesterol secretion

Abstract

Cholesterol and other sterols exit the body primarily by secretion into bile. In patients with sitosterolemia, mutations in either of two ATP-binding cassette (ABC) half-transporters, ABCG5 or ABCG8, lead to reduced secretion of sterols into bile, implicating these transporters in this process. To elucidate the roles of ABCG5 and ABCG8 in the trafficking of sterols, we disrupted Abcg5 and Abcg8 in mice (G5G8(-/-)). The G5G8(-/-) mice had a 2- to 3-fold increase in the fractional absorption of dietary plant sterols, which was associated with an approximately 30-fold increase in plasma sitosterol. Biliary cholesterol concentrations were extremely low in the G5G8(-/-) mice when compared with wild-type animals (mean = 0.4 vs. 5.5 micromol ml) and increased only modestly with cholesterol feeding. Plasma and liver cholesterol levels were reduced by 50% in the chow-fed G5G8(-/-) mice and increased 2.4- and 18-fold, respectively, after cholesterol feeding. These data indicate that ABCG5 and ABCG8 are required for efficient secretion of cholesterol into bile and that disruption of these genes increases dramatically the responsiveness of plasma and hepatic cholesterol levels to changes in dietary cholesterol content.

Fig 1.

Targeted disruption of mouse Abcg5 and Abcg8 genes. (a) Schematic of the wild-type allele, the targeting construct, and the disrupted allele. The targeting construct was generated as described in Materials and Methods. B, BamHI restriction site. (b) Southern blot analysis of BamHI-digested genomic DNA from offspring of a heterozygote (G5G8^+/−) mating. The probe used in the hybridization was a [α-³²P]dCTP-labeled 275-bp DNA fragment (denoted by an asterisk in a), which was amplified from intron 2 of the Abcg5 gene by PCR using the primers 5′-CTCTGTGACAGGAGCTTGTCCTTC-3′ and 5′-CTAGACGCTCAAGGCAGAGTAATC-3′. (c) Northern blot analysis of hepatic and jejunal RNA from wild-type and G5G8^−/− mice. RNA blotting was performed with pooled samples of total tissue RNA from mice of the indicated genotypes (n = 3 in each group) by using [α-³²P]dCTP-labeled ABCG5 and ABCG8 cDNA probes. (d) Immunoblot analysis of ABCG5 and ABCG8 from liver and jejunum of wild-type and G5G8^−/− mice. A total of 50 μg of pooled membrane protein (n = 3 in each group) was fractionated on SDS/PAGE and immunoblotted with rabbit anti-mouse ABCG5 and ABCG8 antisera developed as described in Materials and Methods. The filters were exposed to Kodak X-Omat Blue films for 5–30 s at room temperature. +/+, wild-type mice; −/−, G5G8^−/− mice; *, nonspecific protein that reacts with the anti-ABCG8 antiserum.

Fig 2.

Plasma and hepatic lipid levels in chow- and cholesterol-fed G5G8^−/− and wild-type mice. Individually housed 20-week-old mice were fed a powdered chow diet (0.02% cholesterol) or the same diet containing 2% cholesterol for 21 days. The mice were killed in the daylight portion of the cycle. Blood and liver were collected. Plasma (a) and hepatic (b) levels of lipids were measured by GC as described in Materials and Methods. +/+, wild-type mice; −/−, G5G8^−/− mice. *, P < 0.01 between wild-type and G5G8^−/− mice (n = 5 in each group).

Fig 3.

Fractional absorption of dietary sterols in wild-type and G5G8^−/− mice. Four-month-old female mice of the indicated genotypes (n = 8 in each group) were housed individually. Each mouse was gavaged with 50 μl of an oil mixture containing deuterated cholesterol, cholestanol, campesterol, sitosterol, and sitostanol. Feces were collected for 3 days, and the sterols were extracted. The fecal sterols were subjected to GC-MS as described in Materials and Methods. Deuterated sitostanol was used as a nonabsorbable fecal marker by which the fractional absorption of the other labeled sterols was calculated. +/+, wild-type mice; −/−, G5G8^−/− mice. *, P < 0.05 between wild-type and G5G8^−/− mice.

Fig 4.

Levels of biliary lipids in G5G8^−/− mice and their littermate controls. (a) Male and female 16-week-old G5G8^−/−, G5G8^+/−, and G5G8^+/+ mice (n = 4–5 in each group) maintained on a chow diet were killed after a 4-h fast. The gallbladder bile was collected, and lipid levels were measured as described in Materials and Methods. (b) Gallbladder bile was collected from the same mice as described for Fig. 2, and the levels of cholesterol, phospholipids, and bile acids were assayed as described in Materials and Methods. +/+, wild-type mice; +/−; heterozygotes; −/−, G5G8^−/− mice. *, P < 0.01 between wild-type and G5G8^−/− mice.

Fig 5.

The bile-acid (a) and neutral sterol (b) content of feces from G5G8^−/− mice and their littermate controls and in vivo cholesterol synthesis (c). Feces were collected from individually housed 17-week-old mice of the indicated genotypes for 3 days while they consumed a chow diet. The daily excretion of total bile acids (a) and neutral sterols (b) was determined as described in Materials and Methods. (c) Four-month-old male and female wild-type (WT) and knockout (KO) mice (n = 6 in each group) maintained on a chow diet were injected i.p. with 40 mCi of [³H]H₂O (1 Ci = 37 GBq). The mice were killed 1 h later (mid-daylight cycle), and the tissues were collected and processed as described in Materials and Methods. +/+, wild-type mice; −/−, G5G8^−/− mice. *, P < 0.05 between wild-type and G5G8^−/− mice.

Fig 6.

Quantitative real-time PCR of liver RNA from wild-type and knockout mice. Total RNA from the livers of mice in each group (n = 5) was isolated and pooled for real-time PCR as described (31, 32). Cyclophilin was used as an internal control for these studies. Each value represents the mRNA level relative to the amount of transcript in the chow-fed wild-type males, which was set arbitrarily to 1. +/+, wild-type mice; −/−, G5G8^−/− mice.