ABSTRACT
-
Background/Aims
Excessive lipid accumulation in hepatocytes is a critical cause of metabolic dysfunction-associated steatotic liver disease (MASLD) progression. Ankyrin repeat and SOCS box protein 3 (ASB3) is an E3 ubiquitin ligase that mediates diverse disease processes; however, the direct substrates of ASB3 in lipid metabolism and its role in MASLD remain unexplored.
-
Methods
We generated ASB3 knockout mice fed a high-fat diet to induce MASLD. Oxygen consumption and fatty acid oxidation (FAO) were used to assess lipid metabolism. LC-MS/MS and IP were used to verify the ASB3 target protein. Correlation analysis was conducted on the cohort of MASLD patients vs. the control group.
-
Results
Loss of the ASB3 E3 ubiquitin ligase in hepatocytes strengthens mitochondrial FAO, thereby influencing energy consumption to decrease triglyceride storage and lipid accumulation. Quantitative lysine ubiquitination proteomics revealed that ASB3 directly mediated the ubiquitin levels at two sites (K180 and K639) in carnitine palmitoyl transferase 1A (CPT1A), a rate-limiting enzyme of FAO, to induce CPT1A degradation. Moreover, both constitutive and hepatocyte-specific ASB3 knockout enhance FAO and delay lipid accumulation, liver steatosis, and MASLD progression in a CPT1A-dependent manner. Hepatic ASB3 deficiency also delays fibrosis in MASLD. Analysis of public databases and liver tissue samples from MASLD patients revealed that ASB3 was highly expressed in MASLD patients and was negatively correlated with CPT1A.
-
Conclusions
Our study reveals the key roles of ASB3 in the development of MASLD and suggests a novel therapeutic potential for MASLD.
-
Keywords: Ankyrin repeat and SOCS box protein 3; Metabolic dysfunction-associated steatotic liver disease; Ubiquitin-proteasome system; Carnitine palmitoyl transferase 1A; Fatty acid oxidation
Study Highlights
• ASB3 promotes lipid accumulation in hepatocytes.
• ASB3 directly mediates CPT1A degradation.
• The loss of ASB3 strengthens energy consumption by stabilizing CPT1A.
• The loss of hepatocytic ASB3 delays MASLD progression.
Graphical Abstract
INTRODUCTION
Nonalcoholic fatty liver disease, also referred to as metabolic dysfunction-associated steatotic liver disease (MASLD) based on the new nomenclature, is the most common chronic liver disease, affecting more than 25% of the world’s population [
1,
2], and is defined as the accumulation of fatty acid content greater than 5% of the liver weight. Obesity, insulin resistance (IR), type 2 diabetes mellitus, and metabolic syndrome are high-risk factors for the development of MASLD [
1,
3]. To date, MASLD is a spectrum of diseases ranging from simple steatosis to steatohepatitis, and no approved effective therapeutics are available [
4,
5].
Hepatic lipid accumulation is the initial and critical stage of MASLD [
6]. The cause of hepatic lipid accumulation involves impaired mitochondrial fatty acid oxidation (FAO) [
7,
8]. Hepatic FAO and mitochondrial turnover are compromised in patients with MASLD [
9]. Carnitine palmitoyl transferase 1A (CPT1A) is a key rate-limiting enzyme for FAO and is essential for fatty acids to enter mitochondria [
10]. CPT1A expression is decreased in the livers of obese mice and MASLD patients [
11-
13], whereas its upregulation increases FAO and lipolysis in hepatocytes or adipocytes, antagonizes hepatic steatosis, and improves glucose homeostasis in MASLD [
12,
14,
15]. However, the regulatory effect of decreased CPT1A protein levels on lipid disorder-driven MASLD is not well understood.
Ubiquitination is a crucial posttranslational modification for protein degradation. Ankyrin repeat and SOCS box protein 3 (ASB3) is a member of the E3 ubiquitin ligase family and is composed of a C-terminal SOCS-box structure and an N-terminal ankyrin repeat (ANK) [
16]. ASB3 plays crucial roles in diverse pathological processes of diseases by mediating the polyubiquitination of inflammatory proteins, such as TNF receptor 2 and TNF receptor-associated factor 6 [
17-
19]. However, the direct substrate of ASB3 in metabolism and the role of ASB3 in lipid metabolism remain unexplored, particularly given the lack of reports concerning the role of ASB family members in the development of MASLD.
In this study, using in vitro and in vivo experiments, we found that ASB3 regulates lipid accumulation in hepatocytes via CPT1A degradation. Both constitutive and hepatocyte-specific ASB3 knockout mouse models support the role of ASB3 in promoting MASLD, suggesting that ASB3 is a promising target for improving lipid disorders in MASLD.
MATERIALS AND METHODS
Human studies
Human MASLD specimens and normal liver specimens were obtained from Huashan Hospital. The inclusion of MASLD patient specimens was based on their exhibiting typical clinical manifestations characteristic of MASLD, as confirmed by radiological assessments including computed tomography scans and ultrasound imaging. To ensure validity as normal controls, specimens were procured from liver tissues located at least 2 centimeters away from any pathological lesions or abnormalities, ensuring their status as non-affected tissue. The study was approved by the Research Ethics Committee of Fudan University (no. 2019-198).
Additional Materials and Methods were described in supplementary information.
RESULTS
Effect of ASB3 on lipid accumulation in the liver
Hepatic lipid accumulation is the initial and critical stage of MASLD. We used a HFD-induced fatty liver disease model to investigate whether ASB3 promotes MASLD by affecting hepatic lipid accumulation. We generated ASB3 constitutive knockout mice. The knockout efficiency of ASB3 was confirmed by western blotting (
Supplementary Fig. 1A), and ASB3 was almost undetectable in ASB3-deficient homozygous (HO) mice compared with wild-type (WT) mice. WT mice and
ASB3 HO mice of the same age were then equally divided into two groups and fed an HFD or an NCD at 6 weeks of age for 4 months. Body weight was monitored weekly from 6 to 22 weeks of age during HFD feeding. No significant differences in body weight were observed between the
ASB3 HO mice and the WT mice fed an NCD (
Supplementary Fig. 1B, 1C). The weights of the HFD-fed mice increased gradually, and the weights of HO-HFD mice were significantly lower than those of HFD-fed WT (WT-HFD) mice over the entire period (
Fig. 1A). The mean weight significantly differed between the two groups at 8 weeks (
P<0.05) and 16 weeks (
P<0.05) (
Fig. 1B). Haematoxylin and eosin (H&E) staining revealed more lipid droplets in WT-HFD vs. HO-HFD mice at low magnification, with ballooning degeneration evident in WT-HFD hepatocytes at high magnification (
Fig. 1C). Compared with those from WT-HFD mice, Oil Red O staining of lipid droplets further confirmed the obvious reduction in lipid droplets in the liver of ASB3 HO-HFD mice. Statistical analysis revealed that the percentage of Oil Red O-stained cells in the visual field of the ASB3 HO-HFD mice was significantly lower than that in the WT-HFD mice (
Fig. 1D). In addition, the number of lipid droplets in hepatocytes from WT-HFD mice increased in a dose-dependent manner upon free fatty acid (FFA) treatment, as illustrated by Oil Red O staining, whereas the size and number of lipid droplets were smaller or fewer in
ASB3 HO-HFD mice than in control mice (
Fig. 1E). The quantitative data revealed that the percentage of Oil Red O-positive areas in hepatocytes was significantly lower in
ASB3 HO-HFD mice vs. WT-HFD mice in both the treated and untreated groups (
Fig. 1E). BODIPY staining of lipid droplets in the cells produced similar results (
Fig. 1F). Consistently, ASB3 knockdown in HepG2 cells also resulted in a reduction in FFA-induced lipid droplets (
Supplementary Fig. 1D, 1E). ASB3-overexpressing HepG2 cells (ASB3-OE) exhibited markedly increased lipid droplet accumulation (
Supplementary Fig. 1F, 1G). Taken together, these data revealed that ASB3 knockout affects lipid accumulation and reduces lipogenesis in hepatocytes.
Next, we investigated how ASB3 deficiency mediates the decrease in lipid accumulation in hepatocytes. FAO is an important way to decrease fatty acid accumulation via the promotion of oxygen consumption (OCR). Therefore, we first examined OCR in ASB3-deficient mice
in vivo, and metabolic cage analysis was performed in constitutive ASB3 knockout mice. The results revealed no differences in food intake, carbon dioxide production, respiratory exchange ratio, heat generation, or adipose tissue biology between ASB3 HO-HFD and WT-HFD mice (
Supplementary Fig. 2A–2I). However, OCR was significantly greater in the ASB3 HO-HFD mice housed in the dark (
Fig. 2A). IR is a characteristic of obesity or obesity-related diseases, including MASLD [
20-
23], and liver insensitivity to insulin in individuals with IR leads to impaired glycogen synthesis and activation of the lipid formation pathway [
24]. Glucose and insulin tolerance assays revealed that ASB3 HO mice had significantly improved glucose tolerance and insulin sensitivity under HFD feeding (
Fig. 2B). These data indicate that OCR-induced energy expenditure is increased in ASB3 HO-HFD mice, further revealing that ASB3 plays a critical role in FAO. Consistently, we performed a palmitate (PA) oxidation stress assay to measure FAO by assessing changes in OCR using a Seahorse analyser and observed the effect of ASB3 knockdown on FAO in HepG2 cells. The results revealed that ASB3 was efficiently knocked down at the protein level (
Fig. 2C). In addition, ASB3 knockdown mildly affected basal consumption but significantly increased PA-induced maximal OCR (
Fig. 2C). Similarly, compared with vector control, stable overexpression of ASB3 in cell lines slightly inhibited the basal OCR and markedly reduced the maximal OCR (
Fig. 2D). The RNA sequencing results revealed that energy metabolism-related pathways, such as triglyceride transport, triglyceride metabolism, and cAMP metabolism, were significantly upregulated after ASB3 knockdown (
Fig. 2E, 2F). These data suggest that ASB3 deficiency disrupts FFA-induced lipid accumulation by promoting FAO.
We further studied how ASB3 affects FAO. To explore ASB3 substrates on a broad level, we performed proteomic quantification of lysine ubiquitylation using stable isotope labelling of amino acids in cell culture and affinity enrichment followed by high-resolution LC-MS/MS analysis (
Fig. 3A) [
25]. A total of 5,379 lysine ubiquitylation sites in 2,216 protein groups were identified, among which 5,299 sites in 2,195 proteins were quantified. These modifications were enriched in metabolic pathways, including oxidation-reduction and ATPase activity (
Fig. 3A). α-diGly was used to identify/quantify the ubiquitylation sites using proteomics [
26]. diGLY-modified residues (K180, K639) in the CPT1A protein were detected, and many y- and b-ion series indicated the ubiquitin remnant-containing peptides (
Fig. 3A). No interactions were detected between ASB3 and the other carnitine shuttle components CACT or CPT2 (
Fig. 3A). By ranking the ubiquitin levels from high to low, the ubiquitin levels of two sites (K180 and K639) in the CPT1A protein were found to be significantly downregulated upon ASB3 knockdown (
Fig. 3A).
To further examine whether CPT1A is a potential direct substrate of ASB3, we performed a coimmunoprecipitation assay. The results showed that the Flag-tagged CPT1A protein efficiently pulled down the HA-tagged ASB3 protein, and vice versa, in the 293T and Huh7 cell lines (
Fig. 3B and
Supplementary Fig. 3A). To investigate ASB3-mediated regulation of CPT1A, we performed an ASB3 overexpression assay with cycloheximide (CHX) treatment. ASB3 overexpression induced downregulation of CPT1A protein levels (
Supplementary Fig. 3B). Interference with ASB3 expression decreased CPT1A protein levels (
Supplementary Fig. 3B). In addition, the proteasome inhibitor MG132 increased the accumulation of poly-ubiquitinated CPT1A by blocking its degradation but not disrupting the ASB3-CPT1A interaction (
Fig. 3C, 3D). We then generated hepatocyte-specific ASB3 knockout (
ASB3HKO) mice by crossing
ASB3flox/flox (
ASB3fl/fl) mice with Albumin-Cre transgenic mice. The knockout efficiency of ASB3 was confirmed at the protein and mRNA levels (
Supplementary Fig. 3C). This ASB3-mediated CPT1A ubiquitination effect was further validated
in vivo in hepatocytes from
ASB3fl/fl mice fed an NCD or HFD. CPT1A ubiquitination levels were significantly decreased in hepatocytes from HFD-
ASB3HKO mice (
Fig. 3C). Consistent with these findings, knockdown of ASB3 expression in HepG2 cells decreased CPT1A ubiquitination (
Supplementary Fig. 3D). Using ASB3 truncations, we confirmed that the Ankyrin 5-11 domains of ASB3 were responsible for CPT1A ubiquitination (
Fig. 3D). However, Flag-CPT1A did not interact with HA-ASB3 or its truncations under denatured conditions, whereas CPT1A ubiquitination remained detectable, suggesting that ASB3-mediated ubiquitination is independent of sustained binding (
Supplementary Fig. 3E). Moreover, CPT1A mutations at amino acids 180 and 639 reduced ASB3-mediated ubiquitination, indicating that these are two essential residues for ubiquitination (
Fig. 3E). Importantly, the K180A/639A double mutant exhibited a synergistic reduction in ubiquitination, confirming the critical role of both sites in ASB3-dependent degradation (
Fig. 3E). The subsequent mapping experiments revealed that the Ankyrin 5-11 domains of ASB3 were associated with CPT1A, whereas the C-terminal region segment of CPT1A (amino acids 351–773) was responsible for its association with ASB3 (
Fig. 3F). In summary, these results demonstrated that ASB3 interacts with CPT1A and promotes its ubiquitination, revealing that CPT1A is a substrate of ASB3.
We then performed metabolomics analysis on liver tissue from
ASB3HKO and control
ASB3fl/fl mice. The results revealed 4,768 differentially expressed metabolites between the
ASB3HKO group and the
ASB3fl/fl group (
Supplementary Fig. 4A). Principal component analysis (PCA) revealed pronounced separation between the
ASB3HKO and
ASB3fl/fl groups, demonstrating substantial metabolic profile divergence (
Fig. 4A). Approximately 24.5% of the differentially abundant metabolites in the
ASB3HKO vs. the
ASB3fl/fl comparison were lipid and lipid-like metabolites, including steroids and steroid derivatives, fatty acyls, and glycerophospholipids (
Fig. 4A and
Supplementary Fig. 4B). To verify that ASB3 affects FAO and lipid accumulation through CPT1A in hepatocytes, we examined CPT1A expression and lipid droplets in the livers and hepatocytes of
ASB3HKO mice.
Cpt1a mRNA expression levels in the livers of
ASB3HKO mice were not significantly different from that in the livers of
ASB3fl/fl mice (
Fig. 4B and
Supplementary Fig. 4C). However, CPT1A protein levels were significantly greater in the livers of
ASB3HKO mice than in those of
ASB3fl/fl mice (
Fig. 4C). Thus, these data revealed that the ASB3 silencing-mediated increase in CPT1A is a posttranslational event. Additionally, we further verified that CPT1A activation via L-carnitine treatment did not alter ASB3 expression, confirming that CPT1A does not reciprocally regulate ASB3 (
Supplementary Fig. 4D). Quantitative analysis of BODIPY and Oil Red O staining intensity revealed that L-carnitine significantly reduced lipid accumulation in siNC-transfected control cells; in contrast, this effect was abolished in ASB3 knockout cells (
Supplementary Fig. 4E, 4F). Immunofluorescence staining revealed that ASB3 knockout did not alter the mitochondrial localization of CPT1A (
Supplementary Fig. 4G).
Notably, JC-1 staining under FFA treatment revealed comparable mitochondrial membrane potential between control and ASB3-deficient cells (
Supplementary Fig. 4H). Moreover, RNA-seq analysis revealed no significant changes in the expression of mitochondrial-related genes in ASB3 knockout cells compared with control cells (
Supplementary Fig. 4I). Therefore, our data indicate that the upregulation of FAO resulting from ASB3 deficiency does not occur through alterations in mitochondrial quantity or membrane potential but rather primarily through the regulation of CPT1A in fatty acid transport and utilization.
Hepatocytes were isolated from
ASB3HKO mice and
ASB3fl/fl mice and induced with FFA for 8 hours following treatment with CPT1A with the CPT1A inhibitor etomoxir (Eto, 3 μM) or the control (DMSO). FFA-induced generation of lipid droplets was observed using BODIPY and Oil Red O staining. Compared with those from control mice, hepatocytes from
ASB3HKO mice in the DMSO-treated groups presented significantly decreased sizes and numbers of lipid droplets, whereas the CPT1A inhibitor Eto significantly promoted FFA-mediated lipid droplet formation in hepatocytes from
ASB3fl/fl mice and slightly reversed the decrease in FFA-mediated lipid droplet formation in cells from
ASB3HKO mice (
Fig. 4D). Similar results were observed based on the fluorescence intensity of BODIPY staining for lipid droplets in hepatocytes from
ASB3HKO mice (
Fig. 4D). Moreover, triglyceride levels in cells from the
ASB3HKO mice were significantly lower than those in the cells from the
ASB3fl/fl mice, and Eto treatment significantly increased triglyceride levels in the hepatocytes from the
ASB3fl/fl and
ASB3HKO mice (
Fig. 4E). These results indicate that the decrease in lipid droplet formation and triglyceride levels in the hepatocytes of
ASB3HKO mice is dependent on CPT1A. Moreover, a Seahorse assay revealed that ASB3 knockdown enhanced PA-induced maximal OCR, increased ATP-linked respiration, and promoted spare respiratory capacity (blue line vs. black line in
Fig. 4F), whereas the CPT1A inhibitor Eto treatment partially blocked the ASB3 knockdown-induced increase in FAO (red line vs. blue line in
Fig. 4F). JC-1 staining of Huh7 cells transfected with siASB3 and exposed to FFAs revealed comparable mitochondrial membrane potential to that of the shNC group (
Supplementary Fig. 4H), revealing that the increase in FAO in ASB3-deficient hepatocytes occurred independently of mitochondrial depolarization. Taken together, these data indicate that ASB3 knockdown decreases lipid accumulation and enhances FAO in a CPT1A-dependent manner.
We then explored the role of ASB3 in MASLD by using
ASB3HKO mice. No differences in body weight, adipose tissue, or liver weight were observed between the
ASB3HKO mice and the
ASB3fl/fl mice fed an NCD (
Supplementary Fig. 5A–5E). Liver-directed adeno-associated virus (AAV) viral approaches were used for the
in vivo knockdown of CPT1A in the livers of the mice.
ASB3HKO HO and
ASB3fl/fl mice were divided into four groups: the
ASB3HKO-AAV shNC (HKOshNC),
ASB3HKO-AAV shCPT1A (HKO-shCPT1A),
ASB3fl/fl-AAV shNC (CTL-shNC) and
ASB3fl/fl-AAV shCPT1A (CTL-shCPT1A) groups. Liver-directed AAV virus was injected into
ASB3fl/fl control and
ASB3HKO mice. CPT1A was significantly and efficiently silenced by the AAV shRNA virus vs. the control
in vivo (
Supplementary Fig. 5F). After 12 weeks of HFD feeding, HKO-shNC mice presented a lower body weight compared to CTL-shNC mice (
Fig. 5A). CPT1A knockdown (HKO-shCPT1A) partially rescued this lean phenotype, with no significant difference noted between HKO-shCPT1A and CTL-shCPT1A. This result indicates the essential role of CPT1A in ASB3-mediated weight loss under HFD conditions (
Fig. 5A). Similarly, the liver weight and the proportion of liver weight to body weight in the HKO-shNC group were significantly lower than those in the CTL-shNC group (
Fig. 5B). However, these values were significantly greater in the HKO-shCPT1A vs. HKO-shNC group, and no significant difference was noted between the HKO-shCPT1A and CTL-shCPT1A groups (
Fig. 5B).
The lipid droplets in the liver tissues of the mice were also examined histologically. H&E and Oil Red O staining revealed remarkable lipid droplet accumulation in hepatocytes from CTL-shNC mice compared with HKO-shNC mice, whereas lipid droplet accumulation was increased in the HKO-shCPT1A vs. HKO-shNC group (
Fig. 5C, 5D). Quantification revealed that the number of lipid droplets stained with Oil Red O in the HKO-shNC group was significantly lower than that in the CTL-shNC group but significantly greater in the HKO-shCPT1A vs. HKO-shNC group, with no significant differences noted between the HKOshCPT1A and HKO-shNC groups (
Fig. 5E). We examined FAO activity in hepatocytes from the mice. Conversely, the results revealed that FAO activity in the HKO-shNC group was significantly greater than that in the CTL-shNC group (
Fig. 5F), whereas it was significantly lower in the HKO-shCPT1A group vs. HKO-shNC group, with no significant difference between the HKO-shCPT1A and HKO-shNC groups (
Fig. 5F). These results implied that hepatocyte-specific ASB3 disruption in HFD-fed mice increased the FAO rate by reducing CPT1A degradation.
We then used a methionine-choline deficiency (MCD) diet to induce MASLD to confirm the role of ASB3 in another MASLD mouse model. After 5 weeks of MCD diet intervention, the
ASB3HKO mice had significantly lower body weights, liver weights, and liver-to-body weight ratios than the
ASB3fl/fl mice did (
Fig. 6A). The serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), sensitive indicators of liver injury, were markedly decreased, indicating recovery and protection of liver function (
Fig. 6A). Morphological analysis, H&E staining, Oil Red O staining, and further quantitative analysis of the diameter of vacuoles in H&E-stained sections revealed notable decreases in the diameter and area of lipid droplets within the liver tissue of
ASB3HKO mice (
Fig. 6B). These findings strongly suggest that hepatocyte-specific ASB3 knockout alleviates hepatic lipid accumulation. Additionally, the downregulation of Sirius Red staining and fibrosis-related genes (
Col1a1 and
Acta2) suggested an inhibitory effect of ASB3 knockout on liver fibrosis progression (
Fig. 6C). In terms of the inflammatory response, we observed a marked decrease in inflammatory markers, including
Tnf, Il1b, and
Il6, in the livers of
ASB3HKO mice (
Fig. 6D).
Compared with WT mice, ASB3 HO mice presented significant reductions in body weight and AST and ALT levels (
Supplementary Fig. 6A–6C). Less lipid droplet accumulation and fibrosis were noted in the liver tissue of ASB3 HO mice than in that of WT mice (
Supplementary Fig. 6D–6F). These phenotypes are consistent with the findings in
ASB3HKO mice. No significant differences in lipid transport or lipogenesis genes were observed between WT and ASB3 HO mice (
Supplementary Fig. 6G), suggesting that ASB3 regulates hepatic lipid metabolism specifically through CPT1A-dependent mechanisms.
We also employed a GAN diet (high-fat, high-fructose, and carbon tetrachloride) that faithfully mimics key histopathological features of human MASLD [
27].
ASB3fl/fl mice and
ASB3HKO mice were fed the GAN diet from 12 weeks of age for 3 months. GAN-fed
ASB3HKO mice presented lower body and liver weights (
Fig. 6E). Lipid droplets in hepatocytes were significantly increased in the GAN-fed
ASB3fl/fl mice. At low magnification, more and larger white vacuoles were detected in the
ASB3fl/fl mice than in the GAN-fed
ASB3HKO mice. At high magnification, cells with balloon degeneration were observed in the
ASB3fl/fl mice. The staining of lipid droplets with Oil Red O further confirmed the obvious reduction in the number of lipid droplets in the livers of
ASB3HKO mice compared with those in the livers of
ASB3fl/fl mice (
Fig. 6F). These data together revealed that
ASB3HKO reduces lipid accumulation in liver cells, highlighting the pivotal function of ASB3 in MASLD progression.
Finally, we examined ASB3 expression in the liver tissues of MASLD patients. IHC staining revealed that the ASB3 protein level increased and the CPT1A level decreased significantly in the liver tissue of MASLD patients (
Fig. 7A). This trend was consistently observed in the GAN- and MCD-fed mice (
Fig. 7B). We analysed a bulk RNA-seq dataset from 632 MASLD patients and 32 healthy individuals. ASB3 was slightly upregulated in MASLD patients, who were classified into 4 fibrosis stages on the basis of the NASH-CRN scoring system (
Fig. 7C). Furthermore, a negative correlation between
ASB3 and
CPT1A mRNA expression was observed in the 206 MASLD cohort (R=–0.25,
P<0.001; n=206) (
Fig. 7D). Specifically, we examined the associations of
ASB3 and
CPT1A expression levels with IR markers (
IRF1 and
IRS1), inflammatory factors (
CRP, IL6, and
TNF), and the NAFLD activity score (NAS) in a MASLD cohort (
Fig. 7E).
The results revealed that the transcriptional profiles across disease stages were not significantly correlated (
P>0.05) with
ASB3/CPT1A and the NAS in the cohort (
Supplementary Fig. 7A). However, regarding insulin signalling regulators, ASB3 was positively correlated with
IRF1 (R=0.22,
P=0.001) and IRS1 (R=0.22,
P=0.001), whereas
CPT1A was negatively correlated with these genes (
IRF1: R=–0.14,
P=0.046;
IRS1: R=–0.19,
P=0.004) (
Fig. 7E). Regarding inflammatory mediators, ASB3 was not correlated with CRP (R=–0.032,
P=0.64) (
Supplementary Fig. 7B) but was positively correlated with
IL6 (R=0.20,
P=0.004) and inversely associated with
TNF-α (R=–0.31,
P<0.001) (
Supplementary Fig. 7C, 7D).
CPT1A was strongly correlated with TNF (R=0.33,
P<0.001) and CRP (R=0.29,
P<0.01) but not with
IL6 (R=0.041,
P=0.554) (
Supplementary Fig. 7B–7D). Together, these data reveal a significant correlation of ASB3/CPT1A with inflammatory factors and IR markers. Inflammation and IR are the key clinical features of MASLD, which further suggests the role of ASB3-mediated downregulation of CPT1A in MASLD progression in humans.
Finally, we summarize the new ASB3/CPT1A axis in FAO and its therapeutic potential in preventing MASLD and obesity by targeting the ASB3 ubiquitination pathway (
Fig. 7F).
DISCUSSION
The ubiquitin-proteasome pathway is a major mechanism of intracellular protein degradation and is essential for various physiological processes and diseases [
28,
29]. Lipid metabolism disorders have been recognized in MASLD for decades [
30]. Our study revealed increased ASB3 expression in the liver tissues of MASLD patients. Whole-body and hepatocyte-specific ASB3 knockout in mice protects against HFD-induced MASLD and obesity, as evidenced by reduced body weight and lipid accumulation and improved glucose tolerance, insulin sensitivity, and OCR in a CPT1A-dependent manner. Our results provide a new mechanism underlying HFD-induced MASLD, i.e., ASB3-mediated CPT1A degradation limits energy consumption by restraining FAO, whereas the loss of ASB3 in hepatocytes strengthens mitochondrial FAO, reversing lipid metabolism disorders and preventing MASLD (
Fig. 7F).
ASB3 serves as the substrate recognition component of the E3 ubiquitin ligase complex, interacts with Elongin-B/C to mediate substrate degradation, and plays important roles in human cancers, lipid metabolism, and energy metabolism [
17,
31]. Using quantitative and high-resolution approaches, we identified CPT1A as a new potential direct substrate of ASB3. Similarly, ASB3 induces CPT1A ubiquitination at residues 180 and 639, with mutations in these residues reducing the degree of ubiquitination. Additionally, ASB3 interacts with CPT1A through its ANK region, supporting the concept that a true proteolysis-associated substrate is degraded given its elevated protein ubiquitination level via the interaction of ANK repeats of ASB3 [
32,
33]. CPT1A is the first E3 ubiquitin ligase substrate linked to OCR, revealing the role of ASB3-mediated CPT1A degradation in lipid and energy metabolism.
Excess nutrients and IR increase lipolysis in adipocytes, leading to elevated circulating and hepatic fatty acid levels [
34]. Excessive hepatic TGs, a hallmark of MASLD, are synthesized following the insufficient consumption of FAs by the FAO pathway [
35]. Our study provides multiple lines of evidence that ASB3 regulates hepatic TG homeostasis and obesity in MASLD, particularly through the degradation of CPT1A, which promotes fatty acid utilization via CPT1A-mediated FAO in response to a HFD. In addition, our model also indicated that an increase in mitochondrial FAO reduces diet-induced weight gain in both liver and adipose tissues, which further prevents the systemic progression of obesity-evoked impairment.
The prevalence of MASLD has increased drastically due to the global obesity pandemic; however, no effective therapies have been identified [
30]. Currently, PPARs, glucagon-like peptide-1, and other key molecules are potential targets for the treatment of metabolic diseases, including MASLD [
36]. Although some preclinical studies and clinical trials for agents targeting MASLD, such as the glucagon-like peptide-1 receptor agonist semaglutide, have been conducted, no strategies have been successfully translated into clinical practice to date [
36,
37]. Browning white adipocytes to increase energy expenditure is considered a promising strategy for treating obesity [
38]. Here, we observed that ASB3 knockout model mice presented a lean phenotype and that obesity and MASLD were prevented by increasing energy expenditure via FAO. These results suggest that targeting the ASB3 ubiquitin-proteasome pathway to prevent CPT1A degradation represents a potentially effective therapy for MASLD and obesity. Indeed, enhanced hepatic FAO through CPT1A-mediated gene therapy alters the hepatic lipidomic profile, along with increased autophagy, ketogenesis, and oxidative phosphorylation [
12]. Our study further revealed that the CPT1A inhibitor Eto offsets the influence of ASB3 deficiency and promotes triglyceride storage and lipid accumulation in primary hepatocytes after FFA induction and that AAV-shCPT1A blocks the ASB3 disruption-induced lean phenotype and lipid metabolism disorders
in vivo. We believe that enhancing hepatic FAO through ASB3- and/or CPT1A-mediated gene therapy represents a potential therapeutic option for MASLD and obesity. However, further extensive studies are needed.
FAO is a key bioenergetic pathway that is often dysregulated in diseases, and CPT1A serves as its rate-limiting enzyme. A recent study reported that USP50, a deubiquitinating enzyme, interacts with CPT1A and blocks its proteasomal degradation upon activation of mitochondrial STAT3 [
39]. In addition, CPT1A directly interacts with B-cell lymphoma-2 cells by promoting resistance to apoptosis, affecting the FAO process in lung macrophages [
40]. Our research demonstrated that ASB3 interacts with CPT1A to modulate FAO rates via ubiquitin-mediated CPT1A degradation, revealing novel mechanisms underlying MASLD and indicating that ASB3 is a new regulator of CPT1A. However, further investigations are needed to exclude other CPT1A interactors that may stabilize CPT1A via crosstalk between ubiquitinated proteins and other posttranslational modifications involved in lipid metabolism. Consistently, we observed that CTP1A protein levels were reduced in the Eto treatment group in
ASB3fl/fl mice, but the effect of Eto was not obvious in
ASB3HKO mice. Given that ASB3 is expressed in HFD-fed mice, these data suggest that Etomoxir not only inactivates CPT1A but also likely facilitates ASB3-mediated CPT1A degradation in HFD-fed mice. There is a sex difference in lipid metabolism in HFD-fed mice, with females exhibiting less body weight gain and body fat composition. Therefore, only males are used in this study, which limits the findings.
In conclusion, our study defines the role of ASB3 in regulating β-oxidation associated with metabolic alterations in HFD-induced obesity and MASLD. Our results revealed that ASB3-mediated CPT1A degradation hampers FAO and facilitates lipid accumulation to promote MASLD and that ASB3 disruption stabilizes CPT1A and enhances FAO to prevent MASLD and obesity. Our findings support the feasibility of increasing energy expenditure by targeting ASB3 as a potential therapeutic option for managing MASLD and related disorders.
FOOTNOTES
-
Authors’ contribution
YL, WH, ZW, MG (Mengxiao Ge), LH, HL, and WZ contributed acquisition, analysis, and interpretation of data. WZ, XD, and LW contributed to the acquisition of data and analysis. MG (Ming Guan), DY contributed to the study design. CS, DY, YL, and ZW contributed to the article revision. CS, YL, and DY contributed to the manuscript writing.
-
Acknowledgements
This work was supported by grants from the National Natural Science Foundation of China (81970506, 81770579 to D.Y., 82471775, 82271800 to Y.L., 32270886 to Z.W), the National Key R&D Program of China (2022YFA1106400 and 2020YFA0803201 to Z.W.), Natural Science Foundation of Shanghai Rising-Star Program (23QA1401600 to Y.L.) and general program (23ZR1413000 to Y.L.). The authors thank Prof. Yang Liu for providing small animal nuclear MRIs and Prof. Ronggui Hu for providing His-Ub plasmid.
-
Conflicts of Interest
The authors have no conflicts to disclose.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 1.
Effect of knockdown of ASB3 on the FFA- induced lipid droplets in HepG2 cells. (A) Western blot shows the expression of ASB3 in ASB3 deficiency heterozygous (HE), homozygous (HO) and wild type (WT) mice. (B, C) Comparison of body weight in normal chow diet (NCD)- fed ASB3 HO mice and WT mice. The body weight curve (B) and quantitative data at 22 weeks (C). (D, E) Comparison of FFA induced lipid droplet formation in HepG2 cells treated with ASB3 shRNA (shASB3) or scrambled shRNA controls (shNC) and stained with BODIPY (D) or Oil Red O (E). n=5. (F) Oil Red O staining and quantification of lipid accumulation in ASB3-Normal control (ASB3-NC) and ASB3-overexpressing (ASB3-OE) HepG2 cells treated with 0.5 mM FFA for 24 hours. Scale bar: 100 μm. (G) BODIPY 493/503 fluorescence staining (green) of lipid droplets in ASB3-NC and ASB3-OE hepatocytes under PA treatment. **P<0.01; ns, no significance.
Supplementary Figure 2.
Metabolic cage analysis in ASB3 HO-HFD mice vs. WT-HFD mice. (A, B) Food intake (A) and the quantitative data in the night (dark) and day (light) periods, and total period (B) in the mice; (C, D) Respiratory exchange ratio (RER) (C) and the quantitative data in the night (dark), day (light), and total period (D) in the mice; (E, F) heat generation at night period (E) and quantitative analysis in the night (dark) and day (light) periods, and total period (F) in the mice. (G, H) Carbon dioxide volume (Vco2) at night period (G) and quantitative analysis in the night (dark) and day (light) periods, and total period (H) in ASB3 HO-HFD vs. WT-HFD mice. (I) Relative organ weight (normalized to body weight) of adipose tissues, heart, and liver in ASB3 HO-HFD vs. WT-HFD mice. n=5–6 per group. **P<0.01; ns, no significance.
Supplementary Figure 3.
Comprehensive validation of ASB3-dependent ubiquitination and degradation of CPT1A. (A) Immunoblotting images of ASB3 interacting with CPT1A in Huh7 cell line. (B) CPT1A protein levels in ASB3-overexpression 293T cells (up) or siASB3 293T cells (down) with CHX treatment. (C) Western blot shows ASB3 expression in the liver and lung tissues and the mRNA level of ASB3 in the liver of ASB3HKO and ASB3fl/fl mice in the upper panel. (D) CPT1A ubiquitination in siASB3 treated HepG2 cells with 1 mM FFA treatment. (E) Immunoblotting images of the interaction domains between CPT1A and HA-ASB3 or its truncations under denatured conditions. ****P<0.0001.
Supplementary Figure 4.
Metabolomic profiling and mitochondrial function assay in ASB3L KO and ASB3L KO mice. (A, B) Volcano plots (A) and global component analysis (B) of liver metabolites in the liver tissue of ASB3fl/fl vs. ASB3HKO (n=6 per group). (C, D) Cpt1a mRNA level in ASB3 shRNA Huh7 cells (shASB3) vs. scramble shRNA Huh7 cells (Ctrl). Representation picture and quantification of Oil Red O staining (E) and BODIPY (F) in siNC- or siASB3-transfected hepatocytes treated with L-carnitine (n=3 per group). (G, H) CPT1A (green) and Mito-tracker CMXRos (red) colocalization (G), and JC-1 monomers (green) and aggregates (red) fluorescence ratio (H) in NC or ASB3 knockdown Huh7 cells exposed to 1 mM FFA for 8 hours. n=5–6. (I) genes associated with mitochondrial respiratory chain in NC Huh7 cells or shASB3 Huh7 cells exposed to 1 mM FFA for 8 hours. *P<0.05, ***P<0.001, ns, no significance.
Supplementary Figure 5.
Comparison of normal chow diet (NCD)-fed ASB3HKO and ASB3fl/fl mice. (A–E) Comparison of body weight curve (A), glucose tolerance (B), insulin sensitivity (C), weight of adipose tissues, liver tissues (D) and percentage of adipose tissues and liver tissues over body weight (E) in NCD-fed ASB3HKO vs. ASB3fl/fl mice. (F) The representative western blot results and the quantitative data of CPT1A, ASB3, and GAPDH in liver tissue of the ASB3HKO and ASB3fl/fl mice treated with the indicated virus. n=3. *P<0.05, **P<0.001, ns, no significance.
Supplementary Figure 6.
Comparison of WT and ASB3-/- mice in the progression of MCD-induced MASLD. （A）Relative expression of Asb3 in adipose tissues, lung, and liver of ASB3-/- mice with MCD diets. n=4–6 (B, C). Comparison of body weight (B), liver weight, liver weight/body weight ratio, and serum liver function markers (C) between WT-MCD and ASB3-/- -MCD mice. n=4–10. (D–F) Representative images of liver (D), H&E staining and Oil Red O staining (E) and representative images of Sirius Red staining (F) in WT-MCD and ASB3-/- -MCD mice. (G) Relative mRNA levels of Cd36, Fabp1, Fasn, Acc1, Acox1, and Srebp1c in WT-MCD and ASB3-/- -MCD mice. n=6–7. *P<0.05, **P<0.01, ***P<0.001, ns, no significance.
Supplementary Figure 7.
Expression Levels of ASB3 and CPT1A in relation to different stages and clinical parameters in MASLD. (A) Analysis of ASB3 and CPT1A expression in a cohort of MASLD patients categorized by NAFLD activity score (NAS) using GSE135251 dataset. (B–D) Correlation analysis of ASB3 expression with inflammatory indicators in MASLD patients.
Figure 1.Effect of ASB3 on lipid accumulation in the liver. (A, B) Comparison of HFD-fed ASB3 HO mice (ASB3 HO-HFD) and WT (WTHFD) mice based on body weight (A) and mean body weight at 8 and 16 weeks (B). (C, D) Representative images of H&E staining (C) and quantification of Oil Red O-stained areas (D) in liver tissue from ASB3 HO-HFD mice vs. WT-HFD mice. The top panel is at low magnification, and the bottom panel is at high magnification. The black arrow points to the ballooning degeneration. (E, F) Representative images of FFA-induced percentages of Oil Red O-stained areas (E) and representative images and quantification of neutral lipid droplets stained with BODIPY (F) in hepatocytes from ASB3 HO-HFD and WT-HFD mice. n=3–5 per group. FFA, free fatty acid; HFD, high-fat diet; HO, homozygous; WT, wild-type. *P<0.05, **P<0.01, ***P<0.001.
Figure 2.ASB3 deficiency disrupts lipid accumulation by increasing fatty acid oxidation. (A, B) Dynamic changes in oxygen consumption (OCR) during the night (dark) and day (light) periods and comparisons of OCR during the dark, light, and total periods (A) and the results of the glucose tolerance test and insulin tolerance test (B) in ASB3 HO-HFD and WT-HFD mice. (C, D) Quantitative analysis of basic OCR and maximum respiration OCR and Western blot showing ASB3 expression upon ASB3 shRNA (shASB3) treatment vs. scramble shRNA (shNC) in Huh7 cells (C) and OCR profiles and quantification of maximum respiration in vector control Huh7 cells and ASB3-OE cells exposed to PA (D). (E, F) Volcano plot of differentially expressed genes (DEGs) (E) GO enrichment analysis of DEGs (F) in the RNA-seq data of shNC- and shASB3-transfected Huh7 cells. n=5 per group. HFD, high-fat diet; HO, homozygous; PA, palmitate; WT, wild-type. *P<0.05, ***P<0.001, ****P<0.0001, ns, not significant.
Figure 3.ASB3 regulates CPT1A ubiquitination via direct interactions. (A) Schematic overview of the proteomic quantification approach for identifying lysine ubiquitylation profiles. Summary of the functional landscape and ubiquitination targets identified in our study. Equal amounts of cell lysates (7.5 mg) from shASB3 and shNC cells were used. The diglycine (diGly)-modified lysine and its position are indicated. Here, y1, y2, y3, y4, y5, y6, y7, and y8 are c-terminal fragment y-ion series; b1, b2, b3, and b6 are the N-terminal fragment b-ion series. (B) Immunoblotting images of the interaction of ASB3 with CPT1A in HEK293T cells. (C) ASB3 induces CPT1A ubiquitination in HEK293T cells with or without MG132 (left) and in the hepatocytes of ASB3HKO vs. ASB3fl/fl mice fed an NCD or HFD (right). (D) Identification of the subdomains of ASB3 that induce CPT1A ubiquitination. (E) Effects of CPT1A mutants on ASB3-induced ubiquitination. (F) Identification of subdomains in ASB3 or CPT1A involved in their interactions. HEK293T cells were transfected with the indicated plasmids. ASB3, ankyrin repeat and SOCS box protein 3; CPT1A, carnitine palmitoyl transferase 1A; HFD, high-fat diet; NCD, normal chow diet.
Figure 4.CPT1A-dependent ASB3 regulation of lipid accumulation and FAO. (A) PCA and volcano map analysis of differentially abundant metabolites in the livers of ASB3HKO and ASB3fl/fl mice. (B, C) Comparison of CPT1A RNA (B) and protein (C) levels in the livers of ASB3HKO and ASB3fl/fl mice. (D) Comparison of FFA-induced generation of lipid droplets in hepatocytes from ASB3HKO and ASB3fl/fl mice after treatment with the CPT1A inhibitor etomoxir (Eto, 3 μM) or the DMSO control, as determined using Oil Red O and BODIPY staining. (E, F) Hepatocytes from ASB3HKO and ASB3fl/fl mice were isolated, cultured, and induced with FFA for 8 hours following treatment with Eto or DMSO. (E) Comparison of TG contents in hepatocytes from ASB3HKO and ASB3fl/fl mice. (F) Quantitative analysis of OCR in shASB3-PA vs. shNC-PA cells treated with Eto. n=5 per group. ASB3, ankyrin repeat and SOCS box protein 3; CPT1A, carnitine palmitoyl transferase 1A; FAO, fatty acid oxidation; FFA, free fatty acid; OCR, oxygen consumption; PA, palmitate; PCA, principal component analysis; TG, triglycerides. *PP<0.05, **PP<0.01, ***PP<0.001, ns, not significant.
Figure 5.Hepatocyte-specific ASB3 knockout prevents HFD-induced MASLD in a CPT1A-dependent manner. (A) Comparison of body weight curves and body weights at 20 weeks in the ASB3fl/fl control-AAV shNC (CTL-shNC), ASB3HKO-AAV shNC (HKO-shNC), ASB3fl/fl control-AAV shCPT1A (CTL-shCPT1A), and ASB3HKO-AAV shCPT1A (HKO-shCPT1A) groups (n=5). (B) Comparison of liver images, liver weights, and proportions of liver weight per body weight in the CTL-shNC, HKO-shNC, CTL-shCPT1A, and HKO-shCPT1A groups. (C–F) Comparison of the lipid droplet areas in the liver tissues stained with H&E (C) and oil red O (D) as well as the quantification of oil red O staining (E) and FAO activity (F) in the CTL-shNC, HKO-shNC, CTL-shCPT1A, and HKO-shCPT1A groups. The black arrow points to the lipid vacuole area. n=3–5 per group. ASB3, ankyrin repeat and SOCS box protein 3; CPT1A, carnitine palmitoyl transferase 1A; HFD, high-fat diet; MASLD, metabolic dysfunction-associated steatotic liver disease. *P<0.05, **P<0.01, ns, not significant.
Figure 6.Hepatocyte-specific ASB3 knockout delays MCD-induced and GAN-induced MASLD. (A) Comparison of body weight, liver weight, liver weight/body weight ratio, and serum biomarkers of hepatic ALT and AST function between ASB3fl/fl and ASB3HKO mice fed MCD diets for 5 weeks. (B) Representative hepatic images, H&E and Oil Red O staining, lipid droplet diameters, and lipid droplet areas in the livers of ASB3fl/fl and ASB3HKO mice fed an MCD diet. The black arrow points to the lipid vacuole area. (C, D) Representative images of Sirius Red-stained samples from ASB3fl/fl and ASB3HKO mice and the mRNA levels of fibrosis markers (Col1a1 and Acta2) (C) and inflammation genes (Tnf, Il1b, and Il6) (D) in ASB3fl/fl and ASB3HKO mice fed MCD diets. (E) Body weight, liver weight, and liver/body weight ratio between ASB3fl/fl and ASB3HKO mice fed with GAN diets. (F) Representative images of H&E and Oil Red O staining and quantification of the Oil Red O staining area in ASB3fl/fl and ASB3LHKO mice fed with GAN diets. n=3–6 for in vivo studies. ALT, alanine aminotransferase; AST, aspartate aminotransferase; MCD, methionine-choline deficiency; MASLD, metabolic dysfunction-associated steatotic liver disease. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.
Figure 7.Increased ASB3 expression in the liver tissues of MASLD patients. (A) Comparison of ASB3 and CPT1A IHC staining in MASLD patients and healthy controls and quantification of the data. (B) Comparison of ASB3 and CPT1A IHC staining and quantities in mice fed NCD, MCD and GAN diets. n=5 per group. (C) Analysis of ASB3 expression in a cohort of MASLD patients categorized by fibrosis extent using the SteatoSITE dataset. (D) Correlation analysis between CPT1A and ASB3 expression in MASLD patients from the GSE135251 dataset. (E) Correlation analysis of IRS1, IRF1, CPT1A, and ASB3 expression in MASLD patients from the GSE135251 dataset. (F) A model indicating that ASB3-mediated CPT1A degradation hampers FAO and facilitates lipid accumulation to promote MASLD and that ASB3 disruption stabilizes CPT1A and enhances FAO to prevent MASLD and obesity. ASB3, ankyrin repeat and SOCS box protein 3; CPT1A, carnitine palmitoyl transferase 1A; FAO, fatty acid oxidation; MASLD, metabolic dysfunction-associated steatotic liver disease; MCD, methionine-choline deficiency; NCD, normal chow diet; TG, triglycerides. *P<0.05, **P<0.01, ***P<0.001.
Abbreviations
ankyrin repeat and SOCS box protein 3
aspartate aminotransferase
carnitine palmitoyl transferase 1A
metabolic dysfunction-associated steatotic liver disease
methionine-choline deficiency
principal component analysis
REFERENCES
- 1. Cotter TG, Rinella M. Nonalcoholic Fatty Liver Disease 2020: The State of the Disease. Gastroenterology 2020;158:1851-1864.
- 2. Pouwels S, Sakran N, Graham Y, Leal A, Pintar T, Yang W, et al. Non-alcoholic fatty liver disease (NAFLD): a review of pathophysiology, clinical management and effects of weight loss. BMC Endocr Disord 2022;22:63.
- 3. Younossi ZM. Non-alcoholic fatty liver disease - A global public health perspective. J Hepatol 2019;70:531-544.
- 4. Ferguson D, Finck BN. Emerging therapeutic approaches for the treatment of NAFLD and type 2 diabetes mellitus. Nat Rev Endocrinol 2021;17:484-495.
- 5. Paternostro R, Trauner M. Current treatment of non-alcoholic fatty liver disease. J Intern Med 2022;292:190-204.
- 6. Loomba R, Friedman SL, Shulman GI. Mechanisms and disease consequences of nonalcoholic fatty liver disease. Cell 2021;184:2537-2564.
- 7. Ipsen DH, Lykkesfeldt J, Tveden-Nyborg P. Molecular mechanisms of hepatic lipid accumulation in non-alcoholic fatty liver disease. Cell Mol Life Sci 2018;75:3313-3327.
- 8. Geng Y, Faber KN, de Meijer VE, Blokzijl H, Moshage H. How does hepatic lipid accumulation lead to lipotoxicity in nonalcoholic fatty liver disease? Hepatol Int 2021;15:21-35.
- 9. Moore MP, Cunningham RP, Meers GM, Johnson SA, Wheeler AA, Ganga RR, et al. Compromised hepatic mitochondrial fatty acid oxidation and reduced markers of mitochondrial turnover in human NAFLD. Hepatology 2022;76:1452-1465.
- 10. Schlaepfer IR, Joshi M. CPT1A-mediated Fat Oxidation, Mechanisms, and Therapeutic Potential. Endocrinology 2020;161.
- 11. Nakamuta M, Kohjima M, Morizono S, Kotoh K, Yoshimoto T, Miyagi I, et al. Evaluation of fatty acid metabolism-related gene expression in nonalcoholic fatty liver disease. Int J Mol Med 2005;16:631-635.
- 12. Weber M, Mera P, Casas J, Salvador J, Rodríguez A, Alonso S, et al. Liver CPT1A gene therapy reduces diet-induced hepatic steatosis in mice and highlights potential lipid biomarkers for human NAFLD. Faseb j 2020;34:11816-11837.
- 13. Liu X, Zhang J, Ming Y, Chen X, Zeng M, Mao Y. The aggravation of mitochondrial dysfunction in nonalcoholic fatty liver disease accompanied with type 2 diabetes mellitus. Scand J Gastroenterol 2015;50:1152-1159.
- 14. Tobita H, Sato S, Yazaki T, Mishiro T, Ishimura N, Ishihara S, et al. Alogliptin alleviates hepatic steatosis in a mouse model of nonalcoholic fatty liver disease by promoting CPT1a expression via Thr172 phosphorylation of AMPKα in the liver. Mol Med Rep 2018;17:6840-6846.
- 15. AlNafea HM, Korish AA. Activation of the Peroxisome Proliferator-Activated Receptors (PPAR-α/γ) and the Fatty Acid Metabolizing Enzyme Protein CPT1A by Camel Milk Treatment Counteracts the High-Fat Diet-Induced Nonalcoholic Fatty Liver Disease. PPAR Res 2021;2021:5558731.
- 16. Chung AS, Guan YJ, Yuan ZL, Albina JE, Chin YE. Ankyrin repeat and SOCS box 3 (ASB3) mediates ubiquitination and degradation of tumor necrosis factor receptor II. Mol Cell Biol 2005;25:4716-4726.
- 17. Zhang W, Liu F, Che Z, Wu M, Tang Z, Liu J, et al. ASB3 knockdown promotes mitochondrial apoptosis via activating the interdependent cleavage of Beclin1 and caspase-8 in hepatocellular carcinoma. Sci China Life Sci 2019;62:1692-1702.
- 18. Perez-Garcia J, Espuela-Ortiz A, Lorenzo-Diaz F, Pino-Yanes M. Pharmacogenetics of Pediatric Asthma: Current Perspectives. Pharmgenomics Pers Med 2020;13:89-103.
- 19. Sordillo JE, Kelly RS, Lutz SM, Lasky-Su J, Wu AC. Pharmacogenetics of Bronchodilator Response: Future Directions. Curr Allergy Asthma Rep 2021;21:47.
- 20. Saltiel AR. Insulin signaling in health and disease. J Clin Invest 2021;131.
- 21. London A, Lundsgaard AM, Kiens B, Bojsen-Møller KN. The Role of Hepatic Fat Accumulation in Glucose and Insulin Homeostasis-Dysregulation by the Liver. J Clin Med 2021;10.
- 22. Trouwborst I, Bowser SM, Goossens GH, Blaak EE. Ectopic Fat Accumulation in Distinct Insulin Resistant Phenotypes; Targets for Personalized Nutritional Interventions. Front Nutr 2018;5:77.
- 23. Wu H, Ballantyne CM. Metabolic Inflammation and Insulin Resistance in Obesity. Circ Res 2020;126:1549-1564.
- 24. Irimia JM, Meyer CM, Segvich DM, Surendran S, DePaoli-Roach AA, Morral N, et al. Lack of liver glycogen causes hepatic insulin resistance and steatosis in mice. J Biol Chem 2017;292:10455-10464.
- 25. Hör S, Ziv T, Admon A, Lehner PJ. Stable isotope labeling by amino acids in cell culture and differential plasma membrane proteome quantitation identify new substrates for the MARCH9 transmembrane E3 ligase. Mol Cell Proteomics 2009;8:1959-1971.
- 26. Kim W, Bennett EJ, Huttlin EL, Guo A, Li J, Possemato A, et al. Systematic and quantitative assessment of the ubiquitinmodified proteome. Mol Cell 2011;44:325-340.
- 27. Gallage S, Avila JEB, Ramadori P, Focaccia E, Rahbari M, Ali A, et al. A researcher’s guide to preclinical mouse NASH models. Nat Metab 2022;4:1632-1649.
- 28. Schmidt MF, Gan ZY, Komander D, Dewson G. Ubiquitin signalling in neurodegeneration: mechanisms and therapeutic opportunities. Cell Death Differ 2021;28:570-590.
- 29. Park J, Cho J, Song EJ. Ubiquitin-proteasome system (UPS) as a target for anticancer treatment. Arch Pharm Res 2020;43:1144-1161.
- 30. Pafili K, Roden M. Nonalcoholic fatty liver disease (NAFLD) from pathogenesis to treatment concepts in humans. Mol Metab 2021;50:101122.
- 31. Du WY, Lu ZH, Ye W, Fu X, Zhou Y, Kuang CM, et al. The loss-of-function mutations and down-regulated expression of ASB3 gene promote the growth and metastasis of colorectal cancer cells. Chin J Cancer 2017;36:11.
- 32. Liu B, Jiang S, Li M, Xiong X, Zhu M, Li D, et al. Proteome-wide analysis of USP14 substrates revealed its role in hepatosteatosis via stabilization of FASN. Nat Commun 2018;9:4770.
- 33. Sedgwick SG, Smerdon SJ. The ankyrin repeat: a diversity of interactions on a common structural framework. Trends Biochem Sci 1999;24:311-316.
- 34. Alves-Bezerra M, Cohen DE. Triglyceride Metabolism in the Liver. Compr Physiol 2017;8:1-8.
- 35. Kawano Y, Cohen DE. Mechanisms of hepatic triglyceride accumulation in non-alcoholic fatty liver disease. J Gastroenterol 2013;48:434-441.
- 36. Zhang C, Yang M. Current Options and Future Directions for NAFLD and NASH Treatment. Int J Mol Sci 2021;22.
- 37. Newsome PN, Buchholtz K, Cusi K, Linder M, Okanoue T, Ratziu V, et al. A Placebo-Controlled Trial of Subcutaneous Semaglutide in Nonalcoholic Steatohepatitis. N Engl J Med 2021;384:1113-1124.
- 38. Ma QX, Zhu WY, Lu XC, Jiang D, Xu F, Li JT, et al. BCAABCKA axis regulates WAT browning through acetylation of PRDM16. Nat Metab 2022;4:106-122.
- 39. Li R, Li X, Zhao J, Meng F, Yao C, Bao E, et al. Mitochondrial STAT3 exacerbates LPS-induced sepsis by driving CPT1a-mediated fatty acid oxidation. Theranostics 2022;12:976-998.
- 40. Gu L, Surolia R, Larson-Casey JL, He C, Davis D, Kang J, et al. Targeting Cpt1a-Bcl-2 interaction modulates apoptosis resistance and fibrotic remodeling. Cell Death Differ 2022;29:118-132.
, Chunhua Song4
