Membrane phospholipid remodeling modulates nonalcoholic steatohepatitis progression by regulating mitochondrial homeostasis

Abstract

Background and aims: NASH, characterized by inflammation and fibrosis, is emerging as a leading etiology of HCC. Lipidomics analyses in the liver have shown that the levels of polyunsaturated phosphatidylcholine (PC) are decreased in patients with NASH, but the roles of membrane PC composition in the pathogenesis of NASH have not been investigated. Lysophosphatidylcholine acyltransferase 3 (LPCAT3), a phospholipid (PL) remodeling enzyme that produces polyunsaturated PLs, is a major determinant of membrane PC content in the liver.

Approach and results: The expression of LPCAT3 and the correlation between its expression and NASH severity were analyzed in human patient samples. We examined the effect of Lpcat3 deficiency on NASH progression using Lpcat3 liver-specific knockout (LKO) mice. RNA sequencing, lipidomics, and metabolomics were performed in liver samples. Primary hepatocytes and hepatic cell lines were used for in vitro analyses. We showed that LPCAT3 was dramatically suppressed in human NASH livers, and its expression was inversely correlated with NAFLD activity score and fibrosis stage. Loss of Lpcat3 in mouse liver promotes both spontaneous and diet-induced NASH/HCC. Mechanistically, Lpcat3 deficiency enhances reactive oxygen species production due to impaired mitochondrial homeostasis. Loss of Lpcat3 increases inner mitochondrial membrane PL saturation and elevates stress-induced autophagy, resulting in reduced mitochondrial content and increased fragmentation. Furthermore, overexpression of Lpcat3 in the liver ameliorates inflammation and fibrosis of NASH.

Conclusions: These results demonstrate that membrane PL composition modulates the progression of NASH and that manipulating LPCAT3 expression could be an effective therapeutic for NASH.

Figure 1.. LPCAT3 expression is reduced in NASH patients and liver-specific deletion of Lpcat3 leads to NASH and HCC with age.

(A) Hepatic expression of LPCAT3 in human NAFLD/NASH patients. (B-C) Correlation between hepatic LPCAT3 expression and NAFLD Activity Score (NAS) (B) and fibrosis stage (C) in human NASH patients. 95% confidence band of the best fit line was shown by dashed curve. Data in (A) are presented as means ± SEM. Statistical analysis was performed with one-way ANOVA (A) and linear regression (B-C). **p < 0.01, ***p < 0.001.

Figure 2.. Lpcat3 LKO mice develop spontaneous NASH and HCC with age.

(A) Liver weight (LW) to body weight (BW) ratio of 16-24 months old control Lpcat3^fl/fl (F/F) and Lpcat3^fl/fl Albumin-Cre (LKO) mice on chow diet. (B) Hepatic triglyceride (TG), non-esterified fatty acid (NEFA), free and total cholesterol levels in 16-24 months old F/F and LKO mice on chow diet. (C) Hematoxylin and eosin (H&E) and Sirius Red staining of liver sections from 16-24 months old F/F and LKO mice on chow diet (scale bar, 100 μm). Arrows denote immune cell infiltration and arrow heads show ballooning hepatocytes. (D-E) Quantification of Sirius Red area (D) and NAS (E) for 16-24 months old F/F and LKO mice on chow diet. (F) Liver images, H&E, AFP IHC staining and tumor incidence of 16-24 months old F/F and LKO mice on chow diet (T: tumor. scale bar, 100 μm). Data are presented as means ± SEM. Statistical analysis was performed with student’s t test. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3.. Loss of Lpcat3 in the liver promotes diet-induced NASH and HCC.

(A) H&E, Sirius Red and immunohistochemistry of α-smooth muscle actin (α-SMA) of livers from F/F and LKO mice fed NASH diet for 12 and 30 weeks. Arrows denote immune cell infiltration and arrow heads show ballooning hepatocytes (scale bar, 100 μm). (B-C) Quantification of liver Sirius Red area (B) and NAS (C) of F/F and LKO mice fed NASH diet for 12 and 30 weeks. (D-E) Hepatic mRNA levels of Tnfα and genes involved in fibrosis in F/F and LKO mice fed NASH diet for 12 (D) and 30 weeks (E). (F) Liver images, H&E, AFP staining, tumor incidence and number of tumors per liver of F/F and LKO mice fed NASH diet for 54 weeks (T, tumor. scale bar, 100 μm). Data are presented as means ± SEM. Statistical analysis was performed with student’s t test or multiple t test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 4.. Lpcat3 deficiency in the liver induces ROS production and causes oxidative stress.

(A-B) Enriched terms from Gene Ontology (GO) analysis of significantly upregulated (A) and downregulated genes (B) in 12-week NASH diet-fed LKO livers compared to control F/F. (C) Total ROS levels in primary hepatocytes isolated from F/F and LKO mice fed with NASH diet for 9 weeks. (D) Quantification of mitosox staining in primary hepatocytes isolated from F/F and LKO mice fed with NASH diet for 9 weeks. (E) Immunoblot and quantification of 4-hydroxynonenal (4-HNE) in livers of F/F and LKO mice fed NASH diet for 12 weeks. (F) Immunoblot of MAPK pathways in livers of F/F and LKO mice fed NASH diet for 12 weeks. Data are presented as means ± SEM. Statistical analysis was performed with student’s t test. *p < 0.05, **p < 0.01, ****p < 0.0001.

Figure 5.. Deletion of Lpcat3 in the liver impairs mitochondria homeostasis.

(A) Mitochondrial DNA (mtDNA) content normalized to nuclear DNA (nuDNA) in livers of F/F and LKO mice on chow diet or NASH diet. (B-C) Immunoblots and quantifications of oxidative phosphorylation (OxPhos) complexes in livers (B) or isolated mitochondria (C) from F/F and LKO mice fed NASH diet for 12 weeks. (D) Transmission electron microscopy (TEM) images of liver sections and distribution of hepatic mitochondrial circularity from chow diet or 12-week NASH diet fed mice. Arrows show mitochondria undergoing fission and arrowheads show autophagosome like structures (scale bar, 600 nm). Data in (A-C) are presented as means ± SEM. Data in (D) show median and quantiles. Statistical analysis was performed with multiple t test (A-C) and Kolmogorov-Smirnov test (D). The distribution curve in (D) was fit by Kernel Smooth. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 6.. Loss of Lpcat3 in the liver alters mitochondrial dynamics and promotes stress induced autophagy.

(A-B) Representative immunoblots and quantifications of proteins involved in mitochondrial fusion and fission (A) and autophagy (B) in livers of F/F and LKO mice fed with NASH diet for 12 weeks. (C) Representative images and fluorescence signal readings of autophagic vacuoles staining in primary hepatocytes cultured in serum free medium or treated with cholesterol overnight. Data are presented as mean ± SEM. Statistical analysis was performed with student’s t test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 7.. Lpcat3 deficiency changes mitochondrial membrane composition and oxidative capacity.

(A-B) Fatty acyl composition and total content of phosphatidylcholine (PC) (A) and cardiolipin (B) in inner mitochondrial membrane (IMM) in livers of F/F and LKO mice fed NASH diet for 12 weeks (n=5~6/group). Indicated (sn-1/sn-2) molecular species were confirmed by product ion scanning for aliphatic composition. (C) Representative data of Seahorse mitochondrial stress test in isolated mitochondria from livers of F/F and LKO mice fed NASH diet for 12 weeks. (D) Basal and maximal oxygen consumption rate (OCR) from Seahorse mitochondrial stress test of isolated mitochondria from livers of F/F and LKO mice on chow diet or NASH diet for 12 weeks. Paired data show repetitive experiments. (E) Fatty acid oxidation (FAO) rate of liver homogenate from F/F and LKO mice fed NASH diet for 12 weeks. (F) H₂O₂ production in primary hepatocytes isolated from chow diet fed control and LKO mice with or without palmitate (250 μM) treatment. Data are presented as mean ± SEM. Statistical analysis was performed with multiple t test (A-C), paired t test (D and E), student’s t test (F), and two-way ANOVA (G). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 8.. Overexpression of Lpcat3 in the liver ameliorates diet-induced NASH progression.

(A) H&E, Sirius Red and immunohistochemistry of α-SMA of livers from eGFP and Lpcat3 injected mice fed NASH diet. Arrows denote immune cell infiltration (scale bar, 100 μm). (B-C) Quantifications of Sirius Red area (B) and NAS (C) of eGFP and Lpcat3 injected mice fed NASH diet. (D-E) Enriched GO terms of significantly upregulated (D) and downregulated genes (E) in livers of Lpcat3 injected mice compared to eGFP injected mice based on RNA sequencing data. (F) Relative expression of Lpcat3, inflammation markers and collagen synthesis genes in livers of eGFP and Lpcat3 injected mice fed NASH diet. Data are presented as mean ± SEM. Statistical analysis was performed with student’s t test or multiple t test. *p < 0.05, **p < 0.01.