SR4 was previously synthesized from a validated protocol at the Chemical GMP Synthesis Facility, Beckman Research Institute of the City of Hope (CA, USA) (14). All other chemicals and reagents were purchased from Sigma-Aldrich (Merck KGaA), unless otherwise specified.

All animal experiments were conducted in accordance with a protocol approved by the City of Hope National Medical Center's Institutional Animal Care and Use Committee (Duarte, USA; approval no. 12004). Male 9-week-old db/db mice (BKS.Cg-Dock7^m +/+ Lepr^db/J) weighing 40-45 g were obtained from Jackson Laboratory. A total of 32 animals were divided into two groups: Vehicle control group and SR4-treated group; eight animals per group were used for the efficacy studies and another eight animals per group were used for the metabolic/indirect calorimetry studies. Mice were kept in specific pathogen-free conditions at 22˚C with a 12/12 h light/dark cycle. The animals had unrestricted access to water and standard chow diet (LabDiet^® 5K52; Scott Distributing; ScottPharma Solutions). After 1 week of acclimatization, mice received either the vehicle control (4% DMSO in corn oil) or SR4 dissolved in vehicle (10 mg/kg of BW) via oral gavage (14). Dosing was performed three times a week (Mon, Wed and Fri) in the early morning across a 5-week period. Food consumption was monitored throughout the present study. The animals were monitored daily for health and behavior. Mice were weighed weekly and body composition (lean mass and fat mass) was measured by quantitative magnetic resonance imaging (EchoMRI™ 3-in-1; v2.1; Echo Medical Systems). All animals survived until the end of the present study and did not meet any of the humane endpoints for early euthanasia, which included failure to eat or drink for 24 h, inability to make normal posture and behavior, signs of severe distress and >20% weight loss. At study completion, mice were euthanized via CO₂ inhalation (30-70% total chamber volume/min) followed by cervical dislocation and observation of cessation of breathing to confirm mortality. Blood was then collected through cardiac puncture and the plasma was separated for subsequent analysis. Tissue samples were collected and either snap-frozen and stored at -80˚C or preserved in 10% neutral buffer formalin for further biochemical and immunohistochemical analyses, respectively.

To assess glucose tolerance, mice were fasted for 16 h overnight. A 2 g/kg BW D-glucose solution was then administered intraperitoneally. Tail vein blood (~5-10 µl) was collected serially at 0, 30, 60, 90 and 120 min post-challenge through the tail-nick method (19) The blood collection procedure was approved by the Institutional Animal Care and Use Committee of City of Hope in accordance with the NIH ‘Guidelines for Survival Blood Collection in Mice and Rats’ (https://oacu.oir.nih.gov/). Blood glucose concentration was measured with an Accu-Chek Compact Plus glucometer (Roche Diagnostics Ltd.). The total area under the curve (AUC) was calculated using the trapezoidal method.

Indirect calorimetry was performed in a separate experiment using the PhenoMaster (version 5.9.3; cat. no. 2016-5420; TSE Systems) at the Comprehensive Metabolic Phenotyping Core at Beckman Research Institute (CA, USA). Mice were treated with SR4 or vehicle for 5 weeks (n=8/group). The animals were allowed to acclimate to the cages for 2 days before two cycles of 24 h measurements. Both oxygen consumption (V0₂) and carbon dioxide output (VC0₂) were normalized to lean mass as determined by EchoMRI^™. The respiratory exchange ratio (RER) was calculated as VC0₂/V0₂. With this, the energy expenditure (EE), standardized for BW, was calculated using the following formula: EE=(3.185 + 1.232 x RER) x V0₂.

Plasma was extracted from blood samples following centrifugation at 2,000 x g at 4˚C for 10 min. Plasma insulin was measured using a commercial kit according to the manufacturer's protocol (Crystal Chem Ultra Sensitive Mouse Insulin Elisa kit; cat. no. 90080; Crystal Chem, Inc.). Concentrations of plasma alanine transaminase (ALT), aspartate aminotransferase (AST), total triglycerides (TG) and cholesterol were quantified using commercially available kits (ALT Activity Assay Kit, cat. no. MET-512; AST Assay Kit, cat. no. MET-5127; Triglyceride Quantification kit, cat. no. STA-39; Total Cholesterol Assay Kit, cat. no. STA-384) from Cell BioLabs, Inc., according to the manufacturer's instructions. Levels of glycated hemoglobin (HbA1c) in the blood were measured using the mouse HbA1c assay kit (cat. no. 80310; Crystal Chem, Inc) according to the manufacturer's instructions. Liver TG were measured as previously described (17). Hepatic nucleotides (ATP and AMP) were measured using UV-high-performance liquid chromatography. Briefly, 100 µl of liver homogenates was removed and mixed with 100 µl of deionized water. Furthermore, 18.5 µl of 1 M potassium hydroxide, 30 µl of 150 mM monopotassium phosphate (KH₂PO₄)/150 mM potassium chloride (KCl) solution (Ph 6.0) and 0.5% acetonitrile were added. The final mixture was centrifuged at 17,000 x g for 5 min at room temperature. The supernatant was then collected and transferred to an auto-injector vial and 10 µl was injected into a Thermo ODS Hypersil^™ C18 column (3 µm; 150x4.6 mm; Thermo Fisher Scientific, Inc.). Isocratic separation at 25˚C was achieved using a mobile phase consisting of 150 mM KH₂PO₄, 150 mM KCl (pH 6.0) and 0.5% acetonitrile. The flow rate was set at 0.8 ml/min and a total run time of 20 min. Nucleotides were detected by their absorbance at 260 nm using a Shimadzu SPD-40 UV-Vis detector (Shimadzu Scientific Instruments, Inc.) and compared with the elution position of standards (17).

At the end of the present study, the liver from each mouse was harvested and weighed. Liver samples were either flash-frozen or directly fixed in 10% formalin for downstream assays. Formalin-fixed liver samples were sectioned (5 µm) and stained with H&E for 10-15 min at room temperature. Separately, frozen liver samples from each group were embedded in OCT compound (Sakura Finetek USA, Inc.), sliced and stained with Oil Red O (Sigma-Aldrich; Merck KGaA) at room temperature for 15 min. All stained slides were viewed and images were captured under a bright field microscope (cat. no. AX70; Olympus Corporation) equipped with a digital camera.

Multiple enzymes associated with oxidative stress and antioxidant defense, including glutathione (GSH), glutathione S-transferase (GST) and glutathione peroxidase (GPx), were measured in liver crude homogenates using established methods as previously described (20). The level of malondialdehyde (MDA), a marker of lipid peroxidation, was quantified using the thiobarbituric acid reactive substances assay as described previously (21).

Fresh mitochondria were isolated from the livers of mice (n=4-5/group) as described previously (12). Briefly, mice were euthanized by CO₂ inhalation and cervical dislocation as aforementioned, before livers were immediately removed, minced and placed in ~10-fold volume of ice-cold mitochondria isolation medium [250 mM sucrose, 10 mM Tris/hydrogen chloride (HCl), 1 mM EGTA and 1% fatty acid/endotoxin-free BSA; pH 7.4]. Tissues were homogenized with 25-30 strokes in a Potter-Elvehjem tissue grinder (Sigma-Aldrich; Merck KGaA). After centrifugation at 800 x g for 10 min at 4˚C, the fats and lipids were carefully aspirated. The remaining supernatant was then filtered through a sterile 0.40 µm nylon mesh membrane and centrifuged at 8,000 x g for 10 min at 4˚C. The supernatant and any white debris were removed. The mitochondrial pellet was resuspended in ice-cold mitochondrial assay solution (MAS buffer; 70 mm sucrose, 220 mM mannitol, 10 mM KH₂PO4, 5 mm magnesium chloride, 2 mM HEPES, 1 mM EGTA and 0.2% fatty acid-free BSA; pH 7.2) and this centrifugation was repeated. The final pellet was resuspended in a minimal volume of MAS buffer. Mitochondrial protein concentration was measured using the DC protein assay kit according to the manufacturer's protocol (Bio-Rad Laboratories, Inc.) with BSA as a standard. Isolated liver mitochondria were seeded at 1 µg/well in a Seahorse 96-well plate. Basal respiration (mitochondrial respiration state 2), ADP-stimulated respiration (mitochondrial respiration state 3), non-ADP-stimulated respiration (mitochondrial respiration state 4; proton leak), FCCP-stimulated respiration (mitochondrial respiration state 3u; maximal respiration) and spare respiratory capacity (SRC) were quantified using a Seahorse XF96 flux analyzer (Seahorse Biosciences, Inc.) as described previously (22).

Liver tissue samples were homogenized in a lysis buffer containing 50 mM Tris/HCl (pH 7.4), 150 mM sodium chloride, 1 mM EDTA, 5 mM sodium pyrophosphate, 1 mM sodium orthovanadate, 50 mM sodium fluoride, 1% NP-40, 1 mM PMSF and a protease inhibitor cocktail tablet (Roche Diagnostics Ltd.). Homogenates were then centrifuged for 15 min at 9,600 x g at 4˚C and supernatants were collected. Protein concentration was determined using a DC protein assay kit according to the manufacturer's instructions. For western blotting, 10 µg of protein per sample were resolved by SDS-PAGE on 4-15% Criterion TGX^™ gels (Bio-Rad Laboratories, Inc.) and then transferred to nitrocellulose membranes. Membranes were blocked with 5% skimmed milk in Tris-buffered saline containing 0.05% Tween 20 for 1 h at room temperature before overnight incubation at 4˚C with a 1:1,000 dilution of primary antibodies against total and phosphorylated AMPK and total and phosphorylated ACC (cat. nos. 2532, 2537, 8578 and 3661; Cell Signaling Technology, Inc.). After a series of washes, the membranes were stained with a 1:3,000 dilution of peroxidase-labeled secondary antibodies (goat anti-rabbit IgG HRP conjugate; cat. no. 1706515; Bio-Rad Laboratories, Inc.) at room temperature for 2 h, and an ECL system (Western Lightning^® Chemiluminescence Reagent; cat. no. NEL104001EA; Revvity, Inc.) was used to visualize the immunoreactive proteins. Equal loading of proteins was confirmed by stripping and reprobing the membranes with β-actin antibodies (1:3,000 dilution; cat. no. 3700; Cell Signaling Technology, Inc.).

Total RNA was isolated from liver samples using the RNeasy kit (Qiagen, Inc.) according to the manufacturer's instructions. Complementary DNA (cDNA) was synthesized from the isolated RNA using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, Inc.). qPCR was performed on an ABI-7500 PCR system (Thermo Fisher Scientific, Inc.) using Power SYBR^® Green master mix (Thermo Fisher Scientific, Inc.) for gene amplification and detection. The list of primer pairs used in the present study and their sequences are listed in Table I. The following thermocycling conditions were used for the PCR: Initial denaturation at 50˚C for 2 min; cDNA was denatured at 95˚C for 10 min; and 40 cycles 95˚C for 15 sec and 60˚C for 60 sec. The relative mRNA levels of all the target genes were quantified using the comparative 2^-ΔΔCq method with β-actin as an internal control (23).

Total RNA was extracted from liver tissue samples (n=3 control and n=2 SR4 treatment) using the RNeasy Mini Kit (Qiagen, Inc.), per the manufacturer's protocol. RNA libraries were prepared using the TruSeq Stranded mRNA Library Prep Kit (Illumina, Inc.) and sequenced on the HiSeq^® 2500 platform (Illumina, Inc), generating single-end 51 bp reads. Sequence alignment to the mouse reference genome (GRCm38/mm10) was performed using TopHat2(version 2.1.1; http://ccb.jhu.edu/software/tophat/index.shtml). Read quantification and normalization utilized the ‘Empirical Analysis of Digital Gene Expression in R’ (‘edgeR’; version 4.1) package (24), based on raw count data. Differentially expressed genes (DEGs) analysis was conducted using the ‘DESeq2’ package (25), applying the Benjamini-Hochberg method to control the false discovery rate (FDR). Genes with an FDR<0.05 and an absolute fold change (FC) ≥1.5 were considered DEGs. A volcano plot was generated using VolcaNoseR (version v1.0.3; https://huygens.science.uva.nl/VolcaNoseR/), with Manhattan distance used to identify the top 10 DEGs (26).

Functional enrichment analysis of DEGs was performed using the Database for Annotation, Visualization and Integrated Discovery platform, version 6.8(27), incorporating Gene Ontology (GO; https://geneontology.org/) biological processes and Kyoto Encyclopedia of Genes and Genomes (KEGG; https://www.kegg.jp) pathway annotations. Significant enrichment was defined by adjusted P<0.05 and gene counts ≥3. Furthermore, visualization was conducted using SRplot (https://www.bioinformatics.com.cn/en). To explore protein-protein interactions (PPIs), a PPI network was constructed using the STRING database (version 12.0; https://string-db.org/) (28), with a minimum confidence score of 0.7. Evidence channels for the interactions included in the network were derived from experimental data, curated databases, co-expression, gene fusion, co-occurrence, neighborhood and text mining. The resulting network was visualized and analyzed in Cytoscape (version 3.10.3; https://cytoscape.org/) (29). Significant network modules were identified using the ‘Molecular Complex Detection’ (‘MCODE’) Cytoscape plugin (30), with the following parameters: K-core=2, maximum depth=100, node score cut-off=0.2 and degree cut-off=2. Module functional annotation was further analyzed using STRING. Finally, hub genes within the PPI network were identified using the Maximal Clique Centrality (MCC) algorithm in the ‘cytoHubba’ plugin (31).

Statistical analyses were performed using GraphPad Prism (version 10.3.1; GraphPad; Dotmatics). Data are presented as mean ± SEM. BW changes were compared between groups using repeated measure two-way ANOVA coupled with post hoc Tukey's test. Comparison between two groups was analyzed by Welch's t-test to adjust for unequal sizes and variances between samples. P<0.05 was considered to indicate a statistically significant difference.

To investigate the effects of SR4 in diabetic mice, either vehicle control (4% DMSO in corn oil) or SR4 was administered through oral gavage to male db/db mice three times a week for 5 weeks. SR4-treated mice showed significant weight loss after only 3 weeks of treatment and at the end of the present study these animals exhibited ~13% overall weight loss compared with vehicle control (50.4±0.4 vs. 57.7±0.5 g, respectively; Fig. 1A and B). In addition, SR4-treated animals displayed a greater reduction in fat mass without any marked change in lean mass (Fig. 1C). Notably, the decrease in BW and improvement in body mass composition were not associated with less food consumption as both SR4 treatment and control demonstrated similar daily food intake (Fig. 1D).

To determine if the reductions in weight and improved body composition by SR4 was associated with energy expenditure, indirect calorimetry experiments were performed over a 3-day period (with one day acclimation). Both oxygen consumption and total daily expenditure, but not RER, were significantly increased in animals treated with SR4 compared with the control (Fig. 2A-C). Notably, the apparent increase in oxygen consumption and energy expenditure were only detected in the dark phase and not in the light phase (Fig. 2D and E). No significant change was observed in the daily RER in both dark and light cycles between SR4 and control animals (Fig. 2F).

Db/db mice are characterized by a non-functional leptin receptor, leading to obesity and insulin resistance and ultimately causing hyperglycemia. This hyperglycemia develops progressively, with initial insulin resistance followed by a decrease in insulin secretion, resulting in sustained high blood glucose levels (32). As obesity is one of the key risk factors for insulin resistance in db/db mice, the effects of SR4 on glucose and insulin sensitivity were investigated. SR4 treatment significantly reduced fasting blood glucose after 5 weeks of treatment, with mean plasma glucose of 280.2±0.2 mg/dl in the control group vs. 245±0.4 mg/dl in the SR4-treated group (Fig. 3A). Similarly, HbA1c levels were also significantly decreased by SR4 at the end of treatment (Fig. 3B). Additionally, SR4 reduced the plasma insulin levels significantly (Fig. 3C). To further investigate the effects of SR4 on hyperglycemia, intraperitoneal glucose tolerance tests were performed after a 16 h fast. Glucose tolerance was quantified as the AUC integrated from 0-120 min. Control db/db mice exhibited impaired glucose tolerance and SR4 treatment significantly reduced the peak glucose levels after the glucose load (Fig. 3D). Consequently, glucose AUC was significantly suppressed in SR4-treated mice compared with control (Fig. 3E). Together, these results indicated that SR4 could improve glycemic control and insulin sensitivity in db/db mice.

Plasma TG levels of SR4-treated mice were lower by 28% compared with that of the vehicle control group (168.1±12.8 vs. 235.8±9.0 mg/dl; Fig. 4A). SR4 also significantly decreased total cholesterol by 29% compared with control (139.6±18.2 vs. 196.5±14.3 mg/dl; Fig. 4B). As T2D is associated with the development of fatty liver in the db/db mice (33), the liver pathology was next examined upon SR4 treatment. Compared with the control group, SR4 treated mice had 30% lower liver wet weight (3.6±0.2 vs. 2.5±0.2 g; Fig. 4C). Biochemical analyses also revealed that SR4 significantly decreased hepatic TG contents by 32% compared with control animals (78.6±4.1 vs. 116.0±2.8 mg/g protein; Fig. 4D).

Liver tissues were further analyzed to observe changes in lipid accumulation. H&E staining of liver sections revealed that db/db control animals exhibited numerous large vacuoles filled with excess lipids, which is characteristic of the hepatic steatosis that develops in these mice (Fig. 4E, upper panel). In addition, Oil Red O staining demonstrated an accumulation of numerous larger fat droplets in these mice (Fig. 4E, lower panel). In the SR4-treated group, hepatocellular vacuolation was minimally observed and a notable improvement in lipid accumulation with a marked reduction in the number and size of lipid droplets was detected in the liver of these animals. To determine the molecular mechanisms involved in the inhibitory effect of SR4 on hepatic steatosis, the expression of numerous known hepatic genes was measured genes using qPCR. SR4 significantly suppressed the expression of a number of genes involved in fatty acid and cholesterol synthesis, including acetyl-CoA carboxylase (ACC; Acaca), ATP citrate lyase (Acly), CCAAT/enhancer-binding protein α (Cebpa), fatty acid synthase (Fasn), PPARγ (Pparg), sterol regulatory element binding protein-1c (Srebf1) and stearoyl-coenzyme A desaturase 1 (Scd1; Fig. 4F). Excessive lipid accumulation in the liver is often associated with liver damage, including elevated levels of liver enzymes ALT and AST. SR4 treatment significantly decreased the plasma levels of these enzymes by 40 and 37%, respectively, compared with the control group (Fig. 4G and H). These results indicated that SR4 improved the overall pathology of the liver and attenuated the obesity-induced hepatic steatosis and liver injury in db/db mice.

Mitochondrial uncouplers such as SR4 are known to activate AMPK by disrupting mitochondrial function and reducing ATP production, which can lead to an increase in the AMP:ATP ratio, a key signal for AMPK activation (12). Chronic treatment with SR4 in db/db mice led to a significant increase in the hepatic AMP:ATP ratio (Fig. 5A) as well as an increase in phosphorylation of AMPK and its downstream target ACC (Fig. 5B).

An additional known effect of mitochondria uncouplers is to increase oxygen consumption and mitochondria respiration (9,12,15). Therefore, the mechanisms by which SR4 treatment can affect liver mitochondria function and bioenergetics using the Seahorse flux analyzer were examined (Fig. 5C). Mitochondria freshly isolated from the livers of mice treated with SR4 showed an increase in basal respiration (state 2) of 19.0% compared with control. Similarly, ADP-linked respiration (state 3) increased to ~80% while maximal respiration (state 3u) further increased to 120%, demonstrating increased levels of oxygen consumption linked to ATP production (Fig. 5C). This increase in oxygen consumption rate (OCR) further coincides with a 40% increase in proton leak across the inner mitochondrial membrane. Furthermore, SR4-treatment significantly increased the SRC, the capability to increase energy production when faced with higher demands and often used as a parameter in assessing mitochondria function, by ~195% (Fig. 5D). A previous study showed that impaired mitochondrial function in db/db mice can lead to oxidative stress (5). Next, the effects of SR4 on liver oxidative stress and GSH-linked antioxidant system were examined. Hepatic TBAR levels of mice treated with SR4 were lower by 53% compared with those of the control group (8.6±0.45 nmol MDA/mg protein vs. 17.9±1.45 nmol MDA/mg protein; Fig. 5E). Concomitantly, the activity levels of antioxidant enzymes GSH, GST and GPx were significantly higher in the liver of SR4-treated mice compared with the control (Fig. 5F-H). Together, these data suggest that SR4 uncoupling activity could activate hepatic AMPK, improve mitochondria function and alleviate oxidative stress.

To further investigate the effects of SR4 treatment on liver function and metabolic signaling pathways, untargeted whole-transcriptome sequencing of liver tissues isolated from db/db mice treated with or without SR4 was performed. This analysis identified 642 DEGs, with 217 being upregulated and 425 downregulated by SR4 treatment. Fig. 6A presents a volcano plot highlighting these DEGs based on significance criteria (adjusted P<0.05 and absolute FC≥1.5). The top 10 most strongly regulated genes included the elongation of very long chain fatty acids protein 6 (Elovl6), ephrin type-B receptor 2, cell death-inducing DFFA-like effector A, malic enzyme 1, arrestin domain-containing 3, fatty acid synthase (Fasn), ACC a (Acaca), cytochrome P450 3A41A, cytochrome P450 4A12B and metallothionein 1. The functional relevance of these DEGs to metabolic liver diseases, including diabetes, obesity and hepatic steatosis, is summarized in Table II (34-53). Overall, the genes most highly modulated by SR4 were primarily associated with fatty acid synthesis, lipid accumulation and liver steatosis, inflammatory responses and fibrotic processes, which suggested that SR4 may exert protective effects through the coordinated regulation of hepatic metabolic and stress-response pathways.

To evaluate the biological significance of the DEGs, GO and KEGG pathway enrichment analyses were performed. GO biological process analysis indicated that the 217 upregulated DEGs were significantly enriched in pathways related to ‘Xenobiotic metabolic process’, ‘Steroid metabolic process’, ‘Amino acid metabolic process’, ‘Urea cycle’ and ‘Glutathione metabolic process’ (Fig. 6B). By contrast, the 425 downregulated DEGs were primarily associated with ‘Lipid metabolic process’, ‘Fatty acid biosynthetic process’, ‘Lipid storage’, ‘Glycolytic process’ and the ‘Pentose phosphate shunt’. KEGG pathway analysis revealed that upregulated DEGs were enriched in ‘Metabolic pathways’, ‘Alanine, aspartate and glutamate metabolism’, ‘Chemical carcinogenesis-DNA adducts’, ‘Steroid hormone biosynthesis’, ‘Biosynthesis of amino acids’, ‘Retinol metabolism’ and ‘Metabolism of xenobiotics by cytochrome P450’ (Fig. 6C). Downregulated DEGs were significantly associated with the ‘PPAR signaling pathway’, ‘Carbon metabolism’, ‘Fatty acid metabolism’, ‘AMPK signaling pathway’, ‘Glycolysis/Gluconeogenesis’ and the ‘Pentose phosphate pathway’. These results indicated that the transcriptional changes induced by SR4 treatment involve key pathways regulating nutrient metabolism, energy homeostasis and oxidative stress, processes that are involved in the pathogenesis of T2D and progression to MASLD.

To assess the functional relationships among the identified DEGs, a PPI network was constructed using the STRING database with a high-confidence interaction score threshold of 0.7. The resulting network, derived from the 642 DEGs, comprised 328 nodes (representing 328 of the DEGs) and 764 edges, forming its largest connected component (Fig. 7A). This number of observed edges was significantly greater than expected for a random network of similar size (P≤1x10^-16), indicating that the encoded proteins are more interconnected than would occur by chance and are likely involved in biologically related processes. Among the 328 genes in the network, 118 were upregulated and 210 were downregulated. Network clustering analysis using the ‘MCODE’ plugin in Cytoscape identified ≥20 subnetwork modules. The top five modules were functionally enriched in distinct metabolic processes. Cluster 1 was associated with steroid hormone biosynthesis and retinol metabolism, cluster 2 with the pentose phosphate pathway (PPP) and glycolysis, cluster 3 with fatty acid synthesis and pyruvate metabolism, cluster 4 with lipid droplet metabolism and cluster 5 with cellular amino acid catabolic pathways (Fig. 7B-F).

Due to the central role of AMPK in regulating nutrient and energy metabolism, along with its increased protein phosphorylation and pathway enrichment in the present study, a focused PPI subnetwork was constructed to explore functionally connected gene modules. This subnetwork encompassed DEGs implicated in AMPK signaling, amino acid metabolism, lipid metabolism and glucose metabolism, including key components of the PPP and pyruvate metabolism, as identified through GO and KEGG enrichment analyses. The resulting network comprised 65 nodes and 191 edges, underscoring a high degree of interconnectivity among genes involved in core metabolic processes (Fig. 7G). Notably, multiple interactions within the subnetwork converged on key AMPK regulatory nodes, including Acaca and Acacb (encoding ACC 1 and 2, respectively), Pparg (PPARγ) and mammalian target of rapamycin (mTor), highlighting the intricate integration of lipid biosynthesis with energy-sensing pathways. To pinpoint central regulatory elements within the subnetwork, hub gene analysis was performed using the MCC algorithm within the ‘cytoHubba’ plugin. This analysis identified a number of prominent hub genes, with Fasn and Acly exhibiting the highest MCC scores (Fig. 7H). Additionally, a putative AMPK/PPARγ-associated regulatory network was identified, linking fatty acid synthesis, TG production, lipid droplet formation and storage (Fig. 7I). Collectively, these findings suggested that genes involved in lipid metabolism serve key roles in mediating the SR4-induced metabolic reprogramming in the liver.