Hepatic or intestinal ABCG5 and ABCG8 are sufficient to block the development of sitosterolemia

Forty years ago, Bhattacharyya and Connor (1) were the first to describe sitosterolemia, a rare autosomal recessive lipid disorder characterized by increased absorption of dietary sterols and accumulation of sitosterol and other plant sterols (phytosterols) in the body. Patients with sitosterolemia also present with reduced biliary cholesterol secretion, decreased cholesterol biosynthesis, xanthomatosis, and premature atherosclerotic cardiovascular disease (2). In 2000, Helen Hobbs and her associates (3) discovered that sitosterolemia was caused by mutations that disrupt the expression or function of the ATP binding cassette transporters ABCG5 or ABCG8. Found on chromosome 2p21 in humans, the genes for ABCG5 and ABCG8 are in a head-to-head orientation and are only separated by 374 base pairs. The expression of ABCG5 and ACBG8 is regulated at the transcriptional level by liver X receptors (LXRs), which can be activated by synthetic LXR ligands or oxysterols derived from cholesterol (4). Localized to the apical membrane of hepatocytes in liver and enterocytes in small intestine, ABCG5 and ABCG8 form an obligate heterodimer (G5G8) that pumps sterols into the bile and intestinal lumen (5). Like humans with sitosterolemia, mice with whole-body deficiency of Abcg5, Abcg8, or both transporters have increased dietary sterol absorption, tissue and blood accumulation of phytosterols, and dramatically reduced biliary cholesterol secretion (6–9). However, mice lacking G5G8 do not display increased atherosclerosis susceptibility even on a LDL receptor-null background (10).

In this issue of the Journal of Lipid Research, Wang and colleagues describe the relative roles of hepatic and intestinal G5G8 in sterol homeostasis. For this work, mice were created with LoxP sites that 3′ flanked exon 2 of Abcg5 and exon 1 of Abcg8. G5G8 expression was disrupted in whole-body, liver, or intestine by crossing the floxed mice with mice carrying the Cre transgene under the control of the CMV-IE enhancer/chicken β-actin/rabbit β-globin, albumin, or villin promoter.

The levels of plasma and hepatic phytosterols in mice with G5G8 deleted in liver (L-G5G8^−/−) or intestine (I-G5G8^−/−) was elevated compared with WT mice but significantly decreased compared with mice with whole-body deletion of G5G8 (G5G8^−/−). Unlike the reductions observed in G5G8^−/− mice, cholesterol concentrations in plasma and liver were similar in WT, L-G5G8^−/−, and I-G5G8^−/− mice. These results showed that G5G8 in either liver or intestine was capable of limiting phytosterol accumulation, which may have allowed hepatic cholesterol biosynthesis to be maintained at a normal level. The findings from the I-G5G8^−/− mice were also consistent with the normalization of serum phytosterol concentrations in a sitosterolemic patient that received a liver with functional G5G8 (11).

Due to the presence of G5G8 in the liver, the level of biliary cholesterol was similar in WT and I-G5G8^−/− mice. However, because phytosterol absorption was increased in the absence of intestinal G5G8, the bile from the I-G5G8^−/− mice had high concentrations of phytosterols. As expected, the concentration of biliary cholesterol in L-G5G8^−/− was significantly decreased compared with that of WT mice. However, L-G5G8^−/− mice had a higher level of biliary cholesterol than G5G8^−/− mice, which the authors ascribed to a potentially higher cholesterol synthesis rate in the L-G5G8^−/− mice.

Even though G5G8^−/− and I-G5G8^−/− mice had similar levels of phytosterols in isolated enterocytes, the cells from the I-G5G8^−/− mice did not have a decreased cholesterol concentration. The difference in cholesterol content between the G5G8^−/− and I-G5G8^−/− enterocytes could have been due to the normal level of biliary cholesterol secretion in the I-G5G8^−/− mice. On a standard rodent diet, G5G8^−/− enterocytes were mainly exposed to dietary phytosterol. In contrast, not only dietary phytosterol but also biliary cholesterol could have been internalized by the I-G5G8^−/− enterocytes resulting in normal cellular cholesterol levels.

In order to gauge the importance of hepatic and intestinal G5G8 in reverse cholesterol transport (RCT), mice were injected with ³H-cholesterol, and the level of ³H-cholesterol in bile and feces was measured. The percentage of ³H-cholesterol in the bile of the mice was qualitatively similar to that observed for biliary cholesterol mass (WT = I-G5G8^−/− > L-G5G8^−/− > G5G8^−/−). Reduced biliary cholesterol secretion undoubtedly contributed to the decrease in fecal ³H-cholesterol excretion for the L-G5G8^−/− and G5G8^−/− versus WT mice. However, the relative decrease in biliary ³H-cholesterol did not closely correspond to the reduction in fecal ³H-cholesterol. I-G5G8^−/− compared with WT mice also had decreased fecal ³H-cholesterol excretion, which could have resulted from increased cholesterol absorption (9). However, fractional cholesterol absorption was similar for I-G5G8^−/− and WT mice. Thus, the levels of ³H-cholesterol the ability of enterocytes to efficiently secrete cholesterol into the intestinal lumen.

The highlighted work of Wang and colleagues has provided important information on the tissue-specific functions of G5G8. Future studies should focus on determining whether cholesterol homeostasis can be maintained when L-G5G8^−/− and I-G5G8^−/− mice are challenged with diets enriched in phytosterols or cholesterol. It will also be important to directly measure whether TICE is modulated in the absence of either intestinal or hepatic G5G8.