Hepatic danger signaling triggers TREM2+ macrophage induction and drives steatohepatitis via MS4A7-dependent inflammasome activation

Abstract

Metabolic dysfunction-associated steatohepatitis (MASH), formerly known as nonalcoholic steatohepatitis (NASH), is an advanced stage of metabolic fatty liver disease. The pathogenic mechanisms of MASH center on hepatocyte injury and the ensuing immune response within the liver microenvironment. Recent work has implicated TREM2⁺ macrophages in various disease conditions, and substantial induction of TREM2⁺ NASH-associated macrophages (NAMs) serves as a hallmark of metabolic liver disease. Despite this, the mechanisms through which NAMs contribute to MASH pathogenesis remain poorly understood. Here, we identify membrane-spanning 4-domains a7 (MS4A7) as a NAM-specific pathogenic factor that exacerbates MASH progression in mice. Hepatic MS4A7 expression was strongly induced in mouse and human MASH and associated with the severity of liver injury. Whole-body and myeloid-specific ablation of Ms4a7 alleviated diet-induced MASH pathologies in male mice. We demonstrate that exposure to lipid droplets (LDs), released upon injury of steatotic hepatocytes, triggered NAM induction and exacerbated MASH-associated liver injury in an MS4A7-dependent manner. Mechanistically, MS4A7 drove NLRP3 inflammasome activation via direct physical interaction and shaped disease-associated cell states within the liver microenvironment. This work reveals the LD-MS4A7-NLRP3 inflammasome axis as a pathogenic driver of MASH progression and provides insights into the role of TREM2⁺ macrophages in disease pathogenesis.

Fig. 1.. MS4A7 is a NAM-enriched factor induced in mouse and human MASH.

(A) UMAP feature plots of Ms4a7 mRNA expression in liver cells from chow and MASH diet-fed mice. (B) RNAscope analysis of Ms4a7 expression in mouse liver. (C) Dual-color RNAscope analysis of Ms4a7 and Trem2 expression in diet-induced MASH mouse liver. (D) qPCR and immunoblotting analyses of hepatic Ms4a7 expression (n=7). (E) Correlation of hepatic Ms4a7 with Trem2 mRNA expression and serum ALT concentrations in a cohort of mice fed a MASH diet for three months. (F) qPCR analysis of hepatic Ms4a7 expression in MASH diet-fed mice (MASH) and two parallel cohorts following a dietary switch to chow for one or two months (MASH-chow). n=8–9 mice per group. (G) UMAP feature plot of MS4A7 mRNA expression in human liver cells. (H to I) qPCR analysis of MS4A7 and TREM2 and (I) RNAscope analysis of MS4A7 expression in non-MASH (n=7) and MASH (n=7) human liver biopsies. (J) Correlation of hepatic MS4A7 with TREM2 and COL1A1 mRNA expression in human MASH liver (n=7). Scale bars, 100 μm (B, C, I). Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for panels D and H (left panel), one-way ANOVA for panel F. ** P<0.01, ***P<0.001. Mann-Whitney test for panel H (right panel), ## P<0.01.

Fig. 2.. Ms4a7 inactivation alleviates diet-induced MASH.

Control WT (n=9) and Ms4a7^−/− (n=8) male mice were fed a MASH diet for 22 weeks, starting at 10–12 weeks of age. (A) Body weight and liver weight. (B) Plasma ALT and AST concentrations. (C) H&E histology and Sirius red staining of liver sections. Quantitation of Sirius red-positive area is shown (n=7–8). (D) Liver hydroxyproline content (n=6). (E) TUNEL staining of liver sections. Quantification of TUNEL-positive dots per field was performed on four mice from each genotype (n=4). (F) Plasma CCL2 and TNF-α concentrations. (G) Immunoblots of total liver lysates from WT and Ms4a7^−/− mice. (H-J) Control WT (n=9) and Ms4a7^−/− (n=10) male mice were fed CDA-HFD diet for 8 weeks, starting at 10–12 weeks of age. (H) Sirius red staining of liver sections. (I) Liver hydroxyproline content (n=6) and quantitation of Sirius red-positive area in H (n=5). (J) Plasma parameters. Scale bars, 200 μm (C, H), 100 μm (E). Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for panels A, B, D-J. *P < 0.05, ** P<0.01, ***P<0.001. Mann-Whitney test for panel C, ## P<0.01.

Fig. 3.. Myeloid-specific deletion of Ms4a7 alleviates diet-induced MASH pathologies.

Ms4a7^f/f and Lyz2-Cre, Ms4a7^f/f male mice were fed MASH diet for 22 weeks, starting at 10–12 weeks of age (n=10). (A) Body weight and liver weight, and plasma triacylglycerol (TAG), free fatty acids (FFA), and cholesterol concentrations. (B) Plasma ALT and AST concentrations. (C) H&E and Sirius red staining of liver sections. Scale bars, 200 μm. (D) Quantitation of Sirius red-positive area in C (n=5) and liver hydroxyproline content (n=6). (E) Plasma CCL2 and TNF-α concentrations. (F) Flow cytometry analysis of liver NAMs (n=6). (G) qPCR analysis of hepatic gene expression. Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for all panels, * P<0.05, ** P<0.01, ***P<0.001.

Fig. 4.. Ms4a7 inactivation restricts pathogenic cell state transition of macrophages and HSCs in the liver.

Control WT (n=9) and Ms4a7^−/− (n=8) male mice were fed a MASH diet for 22 weeks, starting at 10–12 weeks of age (E-G, J, K). Single-nucleus RNAseq analysis was performed on pooled liver nuclei isolated from three pairs of WT and Ms4a7^−/− male mice (A-D, H, I). (A) UMAP representation of 12 major liver cell clusters. (B and C) UMAP illustrating three macrophage subclusters (B) and their percentage within the macrophage cluster (C). (D) Dot plot illustrating percentage of macrophages expressing the indicated genes. (E) qPCR analysis of hepatic gene expression (WT, n=9; KO, n=8). (F) Immunofluorescence staining of liver sections. (G) Flow cytometry analysis of liver NAMs (n=6). (H) UMAP illustrating two HSC subclusters (top) and their percentage within the HSC cluster (bottom). (I) Violin plot illustrating percentage of HSCs expressing Col1a1 and Col1a2. (J) qPCR analysis of hepatic gene expression. (K) Immunofluorescence staining of DECORIN and DESMIN. Scale bars, 100 μm (F, K). Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for panel E, G, J, *P< 0.05, ** P<0.01, ***P<0.001.

Fig. 5.. LD exposure promotes NAM induction and hepatocyte injury in MASH liver.

(A) Ms4a7 (RNAscope) and GPNMB (antibody) immunofluorescence staining of MASH mice liver sections. Scale bar, 20 μm. (B) Flow cytometry analysis of liver cells stained with BODIPY 493/503. Chow-fed mice received a single tail vein injection of PBS or LD and were analyzed 2 hr after treatment. (C) Heatmap of hepatic genes induced by LD treatment and diet-induced MASH. (D and E) Mice fed MASH diet for three months were treated with four daily injections of PBS or LD (n=6). (D) Plasma ALT and AST concentrations. (E) qPCR analysis of hepatic gene expression. (F to G) WT and Ms4a7^−/− male mice were fed MASH diet for 8 weeks, followed by PBS or LD injection for 6 days. (n=6–8). (F) A schematic outline of LD injection. Plasma AST and ALT concentrations in MASH diet-fed WT and Ms4a7^−/− mice treated with six daily LD injections. (G) qPCR analysis of hepatic gene expression. Data represent mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for panels D and E, one-way ANOVA for panels F and G.*P< 0.05, ** P<0.01, ***P<0.001.

Fig. 6.. Association between MS4A7 and the NLRP3 inflammasome pathway.

(A) Volcano plot of differentially expressed liver genes in three pairs of WT and Ms4a7^−/− mice fed MASH diet for 22 weeks, starting at 10–12 weeks of age. Blue and red colors denote significantly downregulated and upregulated genes, respectively. Pathway analysis was performed using the regulated genes. Changed pathways were shown at the top of the plot. (B) Heatmap of hepatic gene expression. (C) Heatmap illustrating opposing effects of Ms4a7 ablation and conditional NLRP3 activation (GSE140742) on hepatic gene expression. (D) qPCR analysis of hepatic gene expression in mice fed chow or MASH for 20 weeks (n=7). (E) correlation of liver Nlrp3, Casp1 and Pycard mRNA abundance with serum ALT; and correlation of hepatic Ms4a7 expression with Nlrp3 mRNA abundance in a cohort of mice fed MASH diet for three months. (F) qPCR analysis of hepatic Nlrp3, Casp1 and Pycard expression in MASH diet-fed mice (MASH) and two parallel cohorts following a dietary switch to chow for one or two months (MASH-chow). n=8–9 mice per group. Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for panel D, one-way ANOVA for panel F. ** P<0.01, ***P<0.001.

Fig. 7.. MS4A7 is required for NLRP3 inflammasome activation through physical interaction.

(A to D) Cultured BMDMs were subjected to NLRP3 inflammasome activation by exposure to LPS (200μg/ml) for 4 hr, followed by 5mM ATP treatment for 1 hr. (A) Immunoblots of total cell lysates. (B) Concentrations of secreted IL-1β in culture media (n=6). (C) ASC immunofluorescence staining. Scale bar, 100 μm. (D) Calcium imaging traces in WT and Ms4a7^−/− BMDMs in response to ATP stimulation. (E-F) Primary liver macrophages were isolated from WT and Ms4a7^−/− mice fed MASH diet for 20 weeks. Cells were subjected to NLRP3 inflammasome activation by exposure to LPS (200μg/ml) for 4 hr, followed by 5mM ATP treatment for 1 hr. (E) Immunoblots of total cell lysates. (F) Concentrations of secreted IL-1β in culture media (n=5). (G) Immunofluorescence staining of MS4A7 and transferrin receptor. Scale bar, 10 μm. Membrane topology of MS4A7 is depicted. (H and I) BMDMs were pre-treated with primaquine (50 mM) for 4 or 24 hr before NLRP3 inflammasome activation assay. (H) Immunoblots of total BMDM lysates. (I) IL-1β concentrations in culture media (n=6). (J) Co-immunoprecipitation of transiently transfected MS4A7 and NLRP3 in 293T cells. (K) Co-immunoprecipitation of endogenous MS4A7 and NLRP3 in cultured BMDMs. Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. Two-tailed Student t test for panel B, F, I, **P<0.01, ***P<0.001.

Fig. 8.. Attenuation of NLRP3 inflammasome pathway underlies the protective effects of Ms4a7 inactivation on diet-induced MASH.

WT and Ms4a7^−/− mice were fed MASH diet for 14 weeks and treated with vehicle or MCC950 (20 mg/kg body weight i.p., every other day) for 8 weeks (n=9–10 for each group). (A) A schematic outline of MCC950 treatment study. (B) Plasma ALT and AST concentrations. (C) Sirius red staining of liver sections. Scale bar, 200 μm. (D) qPCR analysis of hepatic gene expression. (E) Immunoblots of total liver lysates and quantitation. (F) Immunofluorescence staining of liver sections. Scale bar, 100 μm. Data are mean ± s.e.m. In all scatter plots with bars, each data point represents one individual. One-way ANOVA for panel B, D, and E., *P< 0.05, ** P<0.01, ***P<0.001.