TRIM25-Mediated INSIG1 Ubiquitination Promotes MASH Progression Through Reprogramming Lipid Metabolism

Abstract

The global incidence of Metabolic dysfunction-associated steatohepatitis (MASH) is increasing, highlighting the urgent need for new treatment strategies. This study aimed to investigate the involvement of tripartite motif-containing 25 (TRIM25) in MASH progression and explore the therapeutic potential of the TRIM25 inhibitor, C₂₇H₂₆N₂O₄S. Functional studies reveal that TRIM25 promoted lipid accumulation and inflammation by ubiquitinating and degrading insulin-induced gene 1 (INSIG1), thereby enhancing the nuclear translocation of sterol regulatory element-binding protein 2 (SREBP2) and upregulating lipid biosynthesis genes. In vivo experiments using TRIM25 knockout mice demonstrated that TRIM25 deletion ameliorated MASH progression, reduced fibrosis, and decreased inflammatory cell infiltration. It identifies C₂₇H₂₆N₂O₄S as a specific inhibitor of TRIM25. C₂₇H₂₆N₂O₄S effectively decreased INSIG1 ubiquitination and attenuated lipid accumulation in the hepatocytes. To enhance the hepatic delivery of C₂₇H₂₆N₂O₄S, it utilizes exosomes derived from hepatic stellate cells (HSC-EVs). Biodistribution analysis confirmed that the HSC-EVs preferentially accumulated in the liver. In a MASH mouse model, HSC-EV-encapsulated C₂₇H₂₆N₂O₄S (C₂₇H₂₆N₂O₄S@HSC-EV) significantly reduced hepatic lipid accumulation and alleviated MASH severity and fibrosis. This study highlights the critical regulatory role of TRIM25 in MASH and presents C₂₇H₂₆N₂O₄S@HSC-EV as a promising therapeutic approach for MASH treatment.

Keywords: C27H26N2O4S; MASH; TRIM25; exosomes; lipid metabolism; targeted delivery; ubiquitination.

Figure 1

TRIM25 upregulates in hepatocytes during MASH pathogenesis. A) Transcriptomic analysis of liver samples from MASH and normal tissues identified 3864 differentially expressed genes. B) Proteomic analysis identified 714 differentially expressed proteins between MASH and normal liver samples (p < 0.05, log₂FC > 1). C) Venn diagram revealing 340 significantly differentially expressed genes in the transcriptomic and proteomic analyses. D) Functional enrichment analysis of the 340 genes. E) Integrated proteomic and metabolomic analyses identified genes significantly correlated with cholesterol and triglycerides. F and G) TRIM25 expression was measured in human normal liver tissue, MAFLD, and MASH liver tissue using RT‐qPCR (F) and WB analysis G). H,I) TRIM25 expression was assessed using RT‐qPCR (H) and WB (I) analysis in the liver tissues of mice from the HFD‐induced MASH mouse model and control mouse (5 vs 5 mice). J,K) TRIM25 expression was assessed in the liver tissues of mice from the HFHC‐induced MASH mouse model and control mouse (5 vs 5 mice). L) Immunohistochemical analysis confirmed increased TRIM25 expression in human MASH livers. M) Immunohistochemical analysis in HFD or HFHC‐induced MASH mouse. N) TRIM25 mRNA expression in a lipid toxicity model induced in cells with PA/OA. O) TRIM25 protein levels in cell lipid toxicity model. P) Immunofluorescence staining was performed to visualize the expression of TRIM25. Q) Lipidomic analysis of TRIM25‐WT and TRIM25‐KO MIHA cells revealed a significant reduction in total lipid content in TRIM25‐KO cells (3 vs 3). R) Differences in the levels of various lipid classes. S) The log₂FoldChange of the most significantly altered lipid species. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 2

TRIM25 promotes lipid accumulation and inflammation in hepatocytes. A) TRIM25‐overexpressing hepatocyte cell lines generated using adenovirus. B) BODIPY staining revealed increased lipid content in TRIM25‐overexpressing MIHA hepatocytes. C,D) TRIM25 overexpression significantly upregulated mRNA (C) and protein (D) levels of lipid metabolism‐related genes and proinflammatory genes in MIHA. E) Overexpression of TRIM25 significantly increased the levels of IL‐6 and TNF‐α in the cell culture medium, as determined by ELISA assays. This observation suggests that TRIM25 may play a key role in promoting the secretion of these pro‐inflammatory cytokines, potentially implicating its involvement in the regulation of inflammatory responses. F) The assay kits detected that cholesterol and triglyceride levels were significantly elevated in TRIM25‐overexpressing cells. G) WB was performed to validate the knockdown efficiency of TRIM25 in hepatocytes. H) BODIPY fluorescence staining was used to detect intracellular lipid changes following TRIM25 knockdown in MIHA cells. I,J) TRIM25 knockdown significantly downregulated mRNA (I) and protein (J) levels of lipid metabolism‐related and proinflammatory genes in MIHA cells. K) Knockdown of TRIM25 significantly increased the levels of IL‐6 and TNF‐α in the cell culture medium, as determined by ELISA assays. L) Reduced intracellular cholesterol and triglyceride levels in TRIM25 knockdown MIHA cells. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 3

TRIM25 deletion attenuates MASH progression and inflammation in vivo. A) Generation of hepatocyte‐specific Trim25 knockout (HKO) mice. B–D) HFHC‐induced HKO‐HFHC MASH model mice exhibited lower blood glucose levels (B), improved glucose tolerance (C), lower liver weight, and liver weight/body weight ratio (D) compared to controls (Flox‐HFHC) (n = 10 each group). E) Serum ALT and AST levels were significantly reduced in HKO‐HFHC mice. F,G) Serum and liver cholesterol and triglyceride levels were markedly lower in HKO‐HFHC mice. H) BODIPY staining confirmed reduced hepatic lipid accumulation in HKO mice. I) Trim25 knockout downregulated expression of lipid metabolism‐related genes. J) Masson's trichrome, α‐SMA, and Sirius Red staining indicated significantly lower liver fibrosis levels in HKO mice. K) Immunofluorescence analysis revealed reduced macrophage infiltration in HKO mice. L) Flow cytometry demonstrated the inhibition of M1 macrophage polarization and an increased proportion of anti‐inflammatory M2 macrophages in HKO mice. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 4

TRIM25 directly binds to INSIG1. A) Immunoprecipitation and silver staining. B) Immunofluorescence co‐localization revealed that TRIM25 and INSIG1 were predominantly distributed in the cytoplasm. C) Immunoprecipitation of MIHA cells confirmed the interaction between TRIM25 and INSIG1. D) HA‐TRIM25 and Flag‐INSIG1 were transfected into HEK293T cells, followed by co‐immunoprecipitation. E) GST pulldown assays demonstrated that GST‐fused TRIM25 pulled down INSIG1. F) TRIM25 knockdown upregulated INSIG1 protein levels and downregulated lipid metabolism‐related proteins and nuclear localization of SREBP2. G) TRIM25 overexpression decreased INSIG1 levels and upregulated lipid metabolism‐related proteins. H) After nucleocytoplasmic separation, the protein levels of SREBP2 in the cytoplasm and nucleus were measured. It was found that overexpression of TRIM25 promoted the nuclear translocation of SREBP2. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 5

TRIM25 mediates ubiquitin‐dependent degradation of INSIG1. A) HA‐TRIM25 and INSIG1 plasmids were transfected into 293T cells, and the protein levels of INSIG1 were measured at 0, 1, 2, and 3 h after adding CHX. TRIM25 significantly promoted INSIG1 protein degradation. B) CHX was added to TRIM25 knockdown and control MIHA cells, and INSIG1 levels were measured at 0, 1, 2, and 3 h. C) In a cellular lipotoxicity model, INSIG1 degradation was accelerated. D) Treatment with the proteasome inhibitor MG132 increased INSIG1 protein levels, indicating degradation via the ubiquitin‐proteasome pathway. E) TRIM25 overexpression plasmid was transfected into MIHA cells, and the protein level of INSIG1 was detected with or without MG132. It was found that overexpression of TRIM25 did not reduce INSIG1 levels in the presence of MG132. F,G) Ubiquitination assays revealed that TRIM25 knockdown decreased INSIG1 ubiquitination (F), whereas TRIM25 overexpression enhanced it (G). H,I) INSIG1 degradation (H) and ubiquitination (I) levels exhibited a dose‐dependent relationship with TRIM25 expression in 293T cells. J) Schematic diagram shows the distribution of ¹³C‐labeled glucose through the tricarboxylic acid (TCA) cycle and de novo fatty acid synthesis metabolism. Control group, TRIM25 overexpression group, and TRIM25+INSIG1 overexpression MIHA cells were incubated with [U‐¹³C] glucose for 24 hours. K) The distribution of palmitic acid (C16:0) containing different numbers of ¹³C isotope carbons. L) Schematic representing the passage of ¹³C from ¹³C‐labeled acetate through the cholesterol biosynthetic pathway. Different treatment groups of MIHA cells were incubated with ¹³C‐labeled acetate for 48 hours. M) The distribution of palmitic acid (C16:0) containing different numbers of ¹³C isotope carbons. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 6

TRIM25 mediates K11‐ and K48‐linked ubiquitination of INSIG1 at lysines 156 and 158. A) TRIM25 truncated plasmids were constructed based on structural features. B) HA‐TRIM25 truncated, and Flag‐INSIG1 plasmids were transfected into MIHA cells, and immunoprecipitation indicated that TRIM25 binds to INSIG1 through its C‐terminal domain (401‐630aa). C,D) A mutant TRIM25 plasmid (HA‐△TRIM25) lacking amino acids 401–630 was constructed. HA‐△TRIM25 or the control plasmid HA‐TRIM25 were co‐transfected with Flag‐INSIG1 into 293T cells, and immunoprecipitation assays were performed. E) Mutations within the C‐terminal domain abolished TRIM25's ability to ubiquitinate INSIG1. F) Compared to the wild‐type INSIG1 plasmid, co‐transfection of the INSIG1 plasmid with lysine mutations at positions 156 and 158 (K156A/K158A) and TRIM25 into 293T cells revealed that the K156A/K158A mutation rendered the protein expression levels resistant to regulation by TRIM25. G,H) Wild‐type, lysine 156 mutant (K156A), lysine 158 mutant (K158A), or combined lysine 156 and 158 mutant INSIG1 plasmids (K156A/K158A) were co‐transfected with TRIM25 into 293T cells. The results indicated that single or combined lysine mutations at positions 156 and 158 in INSIG1 could render the protein and ubiquitination levels of INSIG1 resistant to TRIM25 regulation. I) Co‐transfection with Flag‐INSIG1 and His‐tagged ubiquitin constructs (K6O, K11O, K27O, K29O, K33O, K48O, and K63O) was performed in MIHA cells, followed by overexpression of HA‐TRIM25 to assess the ubiquitination level of INSIG1. The K48O construct is ubiquitin, with all lysines mutated except for K48. J) 293T cells were transfected with His‐tagged K11 mutant (K11R) or K48 mutant (K48R) ubiquitin‐expression plasmids. Cells were pretreated with MG132 for 1 h, and cell lysates were subjected to IP with a Flag antibody to detect ubiquitinated INSIG1. K) The double mutant (K11R/K48R) ubiquitin abolished TRIM25's regulation of INSIG1 ubiquitination. L) Immunofluorescence demonstrated that TRIM25 overexpression promoted SREBP2 nuclear translocation, but this effect was reversed by INSIG1 overexpression. M) In MIHA cells, following overexpression of TRIM25 or simultaneous overexpression of TRIM25 and INSIG1, the expression of lipid metabolism‐related genes was assessed using RT‐qPCR. N) The mRNA levels of lipid metabolism‐related genes were assessed in MIHA cells following TRIM25 overexpression and SREBP1 knockdown. O) In MIHA cells, mRNA levels of lipid metabolism‐related genes were assessed following INSIG1 knockdown or simultaneous knockdown of INSIG1 and SREBP1. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 7

Screening and Validation of C₂₇H₂₆N₂O₄S as a TRIM25 Inhibitor. A) Virtual screening identified potential TRIM25 inhibitors docked into the TRIM25 binding pocket; candidate compounds were selected based on docking scores. B) Top 15 small‐molecule compounds according to docking score. C) A molecular docking model of TRIM25 protein with the small molecule compound C27H26N2O4S was generated using PyMOL software. D) SPR assay confirmed the binding affinity of C₂₇H₂₆N₂O₄S to TRIM25. E) MIHA cells were treated with varying concentrations of C₂₇H₂₆N₂O₄S for 48 h. Cell viability was plotted against drug concentration, and IC₅₀ was calculated. F,G) C₂₇H₂₆N₂O₄S treatment in MIHA cells demonstrated decreased INSIG1 ubiquitination (F) and increased INSIG1 protein expression (G) under lipid stress conditions. H) Immunofluorescence of SREBP2 exhibited that the inhibitor reversed TRIM25‐induced SREBP2 nuclear translocation. I) BODIPY fluorescence staining demonstrated that the inhibitor reduced lipid accumulation in cell models. J) The inhibitor C₂₇H₂₆N₂O₄S reduced lipid accumulation in the lipotoxicity model of MIHA cells. K) C₂₇H₂₆N₂O₄S downregulated the mRNA expression levels of genes associated with lipid metabolism. L) After tail vein injection of varying concentrations of C₂₇H₂₆N₂O₄S in mice, the survival status and body weight changes were monitored. M) Following 28 days of tail vein injection of a low concentration of C₂₇H₂₆N₂O₄S (100 mg kg⁻¹) in mice, various parameters such as blood routine, liver function, and kidney function were assessed. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 8

Exosome‐mediated delivery of C₂₇H₂₆N₂O₄S enhances liver targeting and therapeutic efficacy. A) Transmission electron microscopy (TEM) images of HSC‐EVs. B) Protein expression levels of CD9, CD63, CD81, ALIX, and calnexin in the HSC‐derived exosomes. C) Nanosight analysis depicts the particle size distribution of exosomes. D) IVIS imaging of the liver accumulation of DiR‐labeled exosomes after intravenous injection. E) Representative IVIS images of the DiR‐labeled exosomes in different organs, including the heart, lung, kidney, and spleen. F) Administration of C₂₇H₂₆N₂O₄S encapsulated in HSC‐EVs (C₂₇H₂₆N₂O₄S@HSC‐EV) improved liver function in MASH mouse models more effectively than free C₂₇H₂₆N₂O₄S (5 vs 5 vs 5). G,H) C₂₇H₂₆N₂O₄S@HSC‐EV treatment significantly reduced blood glucose, and hepatic lipid content (G), with enhanced therapeutic outcomes compared to non‐encapsulated treatment (H) (5 vs 5 vs 5). I,J) Bile composition analysis and histological examination revealed decreased bile lipids (I) and alleviated hepatic fibrosis (J) in C₂₇H₂₆N₂O₄S@HSC‐EV‐treated mice (5 vs 5 vs 5). K,L) Liver tissues were obtained from mice subjected to HC‐EV, C₂₇H₂₆N₂O₄S@HC‐EV, and C₂₇H₂₆N₂O₄S@HSC‐EV tail vein injections. Immunohistochemical staining for INSIG1 was performed to determine INSIG1 protein levels (K), and BODIPY fluorescence staining was used to evaluate lipid droplet content (L). M) RT‐qPCR was used to detect the expression of lipid metabolism‐related genes in the livers of mice from different groups. N) Immunohistochemical analysis indicated a reduction in the number of macrophage infiltrates, a decrease in the proportion of CD86‐positive M1 macrophages, and an increase in the proportion of CD206‐positive M2 macrophages following C₂₇H₂₆N₂O₄S@HSC‐EV treatment (IVIS: In Vivo Imaging System). *p < 0.05, **p < 0.01, and ***p < 0.001.