Shijun Deng

SUMMARY: Cholesterol overload drives metabolic dysfunction-associated steatohepatitis (MASH). While the liver maintains homeostasis by exporting cholesterol via very low-density lipoprotein (VLDL) secretion, this process is paradoxically suppressed under cholesterol excess by insulin-induced gene 1 (INSIG1), which inhibits sterol regulatory element-binding protein 1 (SREBP1) activity. How VLDL secretion persists during cholesterol overload to prevent lipotoxicity remains unresolved. We identify a cholesterol-responsive interaction between nuclear factor erythroid 2 related factor-1 (NFE2L1) and INSIG1 that sustains cholesterol balance. Hepatic NFE2L1 deficiency elevates INSIG1 levels, suppressing SREBP1 activation and impairing VLDL secretion, leading to cholesterol accumulation and liver injury. Mechanistically, NFE2L1 binds to INSIG1 via its N-terminal homology box 2 (NHB2) domain, with cholesterol enhancing this interaction to drive INSIG1 degradation and SREBP1 activation. In NFE2L1-deficient mice, wild-type NFE2L1 restores SREBP1 activity and VLDL secretion, while the NHB2-deleted mutant (ΔNHB2) fails. Lipidomics reveal that NFE2L1 deficiency reduces serum triglyceride composition, restored exclusively by wild-type NFE2L1. In a murine MASH model, NFE2L1 overexpression activates SREBP1/2, enhances cholesterol secretion, and alleviates liver injury, inflammation and fibrosis, without elevating atherogenic lipoproteins due to compensatory LDL receptor upregulation. Our findings resolve the paradox of cholesterol-driven VLDL secretion and establish the NFE2L1-INSIG1 axis as a therapeutic target for metabolic diseases.

Objective—ATP-binding cassette transporter G1 (ABCG1) mediates cholesterol efflux to lipidated lipoproteins. Conflicting data about cellular localization of ABCG1 and its effect on cholesterol efflux have been reported. Here, we investigated the underlying mechanisms for these different observations. Approach and Results—Confocal microscopy and biotinylation were used to assess cell surface localization of ABCG1. We found that mouse ABCG1 (mABCG1) used in one previous study has a substitution of Leu to Pro at position 550 (mG1-L550P). When the corresponding Leu at position 562 in human ABCG1 (hABCG1) was mutated to Pro (hG1-L562P), the mutant hABCG1, like mG1-L550P, mainly resided intracellularly, whereas wild-type mABCG1 and hABCG1 were localized on the plasma membrane. However, replacement of this Leu with Pro had no significant effect on mABCG1- and hABCG1-mediated cholesterol efflux. Conclusions—Leu at position 550/562 in mABCG1/hABCG1 is critical for their plasma membrane localization but not for ABCG1-mediated cholesterol efflux. Our findings indicate that the substitution of Leu to Pro at position 550 in mABCG1 may contribute to the non–cell surface localization of mABCG1 observed in the previous study.