Integrated Proteomics and Metabolomics Reveal the Direct Hepatic Protection of Propionate Against Alcoholic Liver Disease via the RGN-PPARα Pathway

Abstract

Background: Propionate, a gut microbiota-derived metabolite, has previously been shown to alleviate chronic alcoholic liver disease (ALD) by preserving intestinal barrier integrity. However, its direct hepatoprotective mechanisms remain unclear. Methods: In this study, employing an acute ALD model to minimize the interference from gut–liver axis effects, we investigated the direct hepatic protection of propionate. Results: Our results demonstrated that propionate administration significantly attenuated hepatic steatosis and oxidative stress. Consistently, in EtOH/OA (oleic acid)-exposed AML-12 hepatocytes, propionate enhanced cell viability and reduced lipid accumulation. Integrated proteomic and metabolomic analyses revealed that propionate altered hepatic proteins and metabolites profiles to stimulate lipolysis, promote fatty acid oxidation, and strengthen antioxidant defenses, consequently restoring lipid homeostasis in ALD mice. Mechanistically, we identified that these beneficial effects may be driven by the upregulation of regucalcin (RGN) following propionate treatments, which, in turn, may activate downstream PPARα signaling via increased levels of p-AMPK, PPARα, ACOX1 and CPT1A. Conclusions: These findings provide novel insight into the liver-centric mechanism through which propionate ameliorates ALD and further support its therapeutic potential in ALD treatment.

1. Introduction

Alcohol-associated liver disease (ALD), resulting from chronic alcohol misuse, ranks among the most prevalent liver disorders worldwide. It is characterized by a progressive pathological spectrum, evolving from initial steatosis to steatohepatitis, fibrosis, cirrhosis, and, ultimately, hepatocellular carcinoma with continued alcohol consumption [1]. Alcohol was estimated to contribute to over 40% of cirrhosis-related deaths and 20% of liver cancer-related deaths globally in 2019 [2], establishing ALD as a leading cause of liver-related mortality. Early intervention and management are crucial for halting ALD progression. However, current clinical options for ALD remain limited [3]. Corticosteroids, such as prednisolone, are the mainstay of treatment, but their utility is hampered by limited long-term efficacy and significant side effects [4,5]. Consequently, beyond the essential recommendation of alcohol cessation, there is a pressing need to identify safe and effective pharmacological strategies for ALD.

In recent years, advances in gut microbiome research have highlighted the therapeutic potential of beneficial microbial metabolites [6,7], particularly short-chain fatty acids (SCFAs). SCFAs, including acetate, propionate, and butyrate, are produced by the gut microbiota through the fermentation of dietary fiber. They enter the systemic circulation via the portal vein and mediate critical interactions between the gut microbiota and host peripheral organs, thereby regulating diverse physiological processes [8]. Among SCFAs, propionate has garnered significant attention due to its good biocompatibility and broad biological functions. For instance, propionate has been shown to protect against Alzheimer’s disease (AD)-like pathological events in AD mice by maintaining mitochondrial homeostasis via the receptors GPR41 and GPR43 [9]. In multiple sclerosis, propionate acts as a promising immunomodulator by promoting the differentiation and function of regulatory T (Treg) cells, thereby correcting the Treg/Th17 imbalance [10]. Furthermore, in cancer research, propionate has been reported to inhibit the epithelial-to-mesenchymal transition (EMT) and reduce the aggressiveness of lung cancer cells [11]. These studies collectively underscore the considerable therapeutic potential of propionate across various disease contexts.

Notably, our previous research established that propionate ameliorated chronic ALD by regulating gut–liver axis homeostasis, with a primary focus on its protective effects on intestinal barrier function, including the enhancement of epithelial integrity, mucus layer composition, and microbial balance [12]. While the gut-centric mechanism is important, it must be emphasized that the liver is the primary site of alcohol metabolism and the main target of its toxicity. Alcohol-induced hepatocellular injury plays a central role in the pathogenesis of ALD [13]. Therefore, it remains to be elucidated whether propionate mediates hepatoprotection through direct, liver-intrinsic mechanisms, independently of its established intestinal effects.

Currently, integrated multi-omics strategies, combining proteomics and metabolomics, have emerged as powerful tools for uncovering the targets and mechanisms of bioactive compounds [14,15]. These approaches provide comprehensive insights into pharmacological actions from both protein expression and metabolic flux perspectives. To address the aforementioned question, the present study established an acute ALD mouse model to minimize the influence of the gut–liver axis and investigated the direct hepatoprotective effects of propionate, focusing on hepatic lipid metabolism and oxidative stress. Integrated proteomic and metabolomic analyses were then employed to elucidate the intrinsic molecular mechanisms governing propionate-mediated hepatoprotection.

2. Materials and Methods

2.1. Reagents and Materials

Sodium propionate (>99.0%, P5436) and sodium oleate (≥99.0%, O7501) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Primary antibodies, including β-actin (#66009-1-Ig), CPT1A (#15184-1-AP), SREBP-1 (#14088-1-AP), ACC1 (#21923-1-AP), ACOX1 (#83731-2-RR), PPARα (#66826-1-Ig) and RGN (#17947-1-AP), were obtained from Proteintech Group (Wuhan, China). GAPDH (#AF7021), AMPK (#AF6423) and p-AMPK (#AF3423) were obtained from Affinity Biosciences (Cincinnati, OH, USA). CYP2E1 (#ab28146) was obtained from Abcam (Cambridge, MA, USA). All secondary antibodies were obtained from Proteintech Group (Wuhan, China). The BCA protein assay kit (P0012) and RIPA lysis buffer (P0013B) were obtained from Beyotime Biotechnology (Shanghai, China). AML-12 cells were obtained from Cell Bank, Type Culture Collection, the Chinese Academy of Sciences (Shanghai, China).

2.2. Animal and Treatments

Male C57BL/6J mice (8 weeks old, 18–20 g) were purchased from Beijing River Laboratory Animal Technology Co., Ltd. (Beijing, China) and acclimated for 1 week under 12 h light/dark cycles (23 ± 2 °C, 50–60% humidity) with ad libitum access to standard chow and water. All procedures were approved by the Institutional Animal Care and Use Committee of Shandong Analysis and Test Center (No. ECAESDATC-2025-009). The acute alcohol-induced liver disease (acute-ALD) model timeline is outlined in Figure 1A. Specifically, mice were first numbered sequentially based on their order of arrival and then randomly assigned into four experimental groups (n = 8 per group) using a computer-generated randomization sequence (Excel software). The groups were as follows: the control group (control), the ethanol intake group (EtOH), the EtOH + 50 mM sodium propionate group (EtOH + L-PA), and the EtOH + 150 mM sodium propionate group (EtOH + H-PA). To ensure allocation concealment, the randomization procedure was performed by an independent researcher who was not involved in subsequent animal experiments. All mice were fed with the control Lieber-DeCarli diet (ReadyDietech, Shenzhen, China) for 4 days to adapt to the liquid diet. Then, the ethanol-fed groups were switched to the ethanol-containing liquid diet, starting at 1% (w/w) and increasing stepwise to 2%, 3% and, finally, 4%; the 4% ethanol diet was maintained for the subsequent 11 days. Sodium propionate was dissolved directly in the liquid diet at 50 mM or 150 mM. Control mice were pair-fed an isocaloric liquid diet in which ethanol was replaced with maltodextrin. On the final day, mice in the ethanol groups received a single oral binge gavage of ethanol (5 g/kg body weight), whereas the control mice were gavaged with an isocaloric maltodextrin solution. Nine hours after the binge gavage, the mice were deeply anesthetized via intraperitoneal administration of 2,2,2-tribromoethanol (dissolved in tert-amyl alcohol) and euthanized by cervical dislocation. The blood was collected and centrifuged to obtain the serum, which was then stored at −80 °C. Liver lobes were excised; portions were snap-frozen in liquid nitrogen and stored at −80 °C for biochemical analyses, and the remainder was fixed in 4% paraformaldehyde for histological examination.

2.3. Biochemical Analysis

Alanine aminotransferase (ALT) aspartate, aminotransferase (AST) activities, low-density lipoprotein (LDL-C), high-density lipoprotein (HDL-C), total cholesterol (TC) and triglyceride (TG) in serum were determined with commercial assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the manufacturer’s instructions. Hepatic TC, TG, CAT, SOD, GSH and MDA contents were quantified in liver homogenate supernatants (10% w/v in PBS) using enzymatic assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

2.4. Histological Analysis

For histologic analysis, paraffin-embedded liver specimens were sectioned at 3 µm. After deparaffinization and rehydration, sections were stained with hematoxylin and eosin (H&E) following standard protocols, dehydrated, cleared, and mounted with neutral balsam. For neutral lipid visualization, 5 µm frozen liver sections were air-dried, fixed in 10% neutral-buffered formalin for 10 min, rinsed with distilled water, and incubated with 0.5% Oil Red O solution for 15 min at room temperature. Following brief differentiation in 60% isopropanol, sections were counterstained with hematoxylin, mounted in aqueous medium, and imaged for lipid droplet quantification.

2.5. Cell Culture and Treatments

AML-12 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum at 37 °C under 5% CO₂. For all experiments, cells were seeded and allowed to attach overnight. The next day, cells were either left untreated or exposed to 200 mmol/L ethanol plus 0.5 mmol/L oleic acid (OA) in the form of sodium oleate for 24 h. Sodium propionate (1 or 2 mmol/L) treatment was performed 3 h before ethanol/OA exposure.

2.6. Cell Viability Analysis

Cell viability was assessed using the Cell Counting Kit-8 (Solarbio, Beijing, China) according to the manufacturer’s instructions. AML-12 cells were seeded at 5 × 10³ cells per well in 96-well plates and exposed to the indicated treatments (sodium propionate, OA or ethanol for 24 h). After treatment, 10 µL of CCK-8 reagent was added to each well, and the plates were incubated at 37 °C for 3 h. The absorbance at 450 nm was recorded on a microplate reader. The data are expressed as the percentages of the untreated control.

2.7. Oil Red O Staining of Cells

The Oil Red O staining of cultured cells was performed using the Oil Red O Stain Kit (Solarbio, Beijing, China) following the manufacturer’s instructions. After treatment, cells were washed twice with PBS and fixed with Oil Red O fixative for 20 min. Following a brief rinse in 60% isopropanol, cells were incubated in freshly prepared Oil Red O working solution for 15 min. Excess dye was removed by differentiation in 60% isopropanol (10 s) and two washes with distilled water. Nuclei were stained with Mayer’s hematoxylin for 1 min and then incubated with Oil Red O buffer for 1 min. After being rinsed again with distilled water, stained cells were kept in distilled water and examined by light microscopy.

2.8. Western Blot

Liver tissue was homogenized in RIPA buffer containing 1% protease and phosphatase inhibitors and centrifuged at 12,000 rpm for 15 min at 4 °C, and the protein concentration was determined by BCA. Equal amounts (10 µg) of proteins were separated by 7–10% SDS-PAGE, transferred to PVDF membranes, and blocked with 5% non-fat milk in TBST. The membranes were incubated overnight at 4 °C with primary antibodies (β-actin, 1:10,000; GAPDH, 1:10,000; CYP2E1, 1:5000; and CPT1A, SREBP-1, ACC1, ACOX1, AMPK, p-AMPK, PPARα and RGN, all at 1:1000) diluted in TBST, followed by incubation with HRP-conjugated secondary antibodies (1:10,000) for 1 h. Bands were visualized using G:BOX Chemi XX9 (Syngen, London, UK) and quantified by Image J 1.8.0 (Bethesda, MD, USA). Values were normalized to β-actin or GAPDH.

2.9. Molecular Docking

Molecular docking of propionate to regucalcin (RGN) was conducted with the AutoDock 4.2.6 program. The RGN protein was prepared in AutoDockTools 1.5.6 by removing water/ligands and adding polar hydrogens; propionate was built in ChemDraw 25.0, energy-minimized, and saved as pdbqt. A 60 × 60 × 60 grid (0.5 Å spacing) centered on the predicted active site was used. Lamarckian GA runs (100) with default parameters were executed, and the lowest-energy pose was selected and rendered in PyMOL 1.8.6.

2.10. Proteomic Analysis

2.10.1. Protein Extraction, Digestion and TMT Labeling

TMT-based quantitative proteomics was performed at Shanghai Applied Protein Technology Co., Ltd. (APT, Shanghai, China). Liver tissue proteins were extracted with SDT buffer (4% SDS, 100 mM Tris-HCl, and1 mM DTT; pH 7.6) and quantified by BCA. Protein digestion by trypsin was performed according to the FASP protocol. The digest peptides of each sample were then desalted on a C18 Cartridge, concentrated by vacuum centrifugation and reconstituted in 40 µL of 0.1% (v/v) formic acid. A 100 μg peptide mixture of each sample was labeled using the TMT reagent according to the manufacturer’s instructions (Thermo Scientific, Waltham, MA, USA).

2.10.2. Strong Cation Exchange (SCX) Fractionation

TMT-labeled peptides were resuspended in buffer A (10 mM KH₂PO₄ and 25% ACN; pH 3.0) and separated on a PolySULFOETHYL 4.6 × 100 mm SCX column (5 µm, 200 Å, PolyLC, Columbia, MD, USA) using an AKTA Purifier (1 mL min⁻¹; 0–100% buffer B containing 500 mM KCl over 60 min). Elution was monitored at 214 nm; 1 min fractions were pooled, desalted in C18 Empore cartridges, and dried in vacuo.

2.10.3. LC-MS/MS Analysis

LC-MS/MS was conducted with a Q Exactive HF-X coupled to an Easy-nLC 1200 (300 nL min⁻¹, Thermo Scientific, Waltham, MA, USA). Peptides were trapped in an Acclaim PepMap100 C18 cartridge (100 µm × 2 cm, Thermo Scientific, Waltham, MA, USA) and separated in an in-house 10 cm × 75 µm, 3 µm C18 column using a 60 min gradient of 5–35% buffer B (0.1% FA/84% ACN). Full MS (300–1800 m/z) was acquired at a 70 k resolution; the ten most intense ions were subjected to HCD (30 NCE, 17.5 k resolution, and 2 m/z isolation) with dynamic exclusion (40 s). The AGC targets were 3 × 10⁶ (MS¹) and 1 × 10⁵ (MS²), a maximum IT of 10 ms, and peptide match preferred.

2.10.4. Protein Identification and Bioinformatic Analysis

The MS raw data for each sample were searched using the MASCOT engine (Matrix Science, London, UK; version 2.2) embedded into the Proteome Discoverer 1.4 software for identification and quantitation analysis. Related parameters and instructions are in

Table 1. MASCOT engine search parameters.

Item	Value
Enzyme	Trypsin
Max. missed cleavages	2
Fixed modifications	Carbamidomethyl (C), iTRAQ 4/8plex (N-term); iTRAQ 4/8plex (K) TMT 6/10/16 plex (N-term); TMT 6/10/16 plex (K)
Variable modifications	Oxidation (M); iTRAQ 4/8 plex (Y); TMT 6/10/16 plex (Y)
Peptide mass tolerance	±20 ppm
Fragment mass tolerance	0.1 Da
Database	See the project report
Database pattern	Decoy
Peptide FDR	≤0.01
Protein quantification	The protein ratios are calculated as the median of only unique peptides of the protein.
Experimental bias	Normalizes all peptide ratios by the median protein ratio. The median protein ratio should be 1 after the normalization.

. Bioinformatic analysis was performed on differentially expressed proteins (DEPs; fold change ≥ 1.2; adjusted p < 0.05). Hierarchical clustering (Cluster 3.0, Java TreeView) and heatmap visualization were used to display expression patterns. Subcellular localization was predicted with CELLO (http://cello.life.nctu.edu.tw/, accessed on 9 December 2024), and conserved domains were annotated via InterProScan 5.36-75.0 (Pfam). GO terms were assigned by Blast2GO 5.2 after local BLAST 2.16.0+ InterProScan, and KEGG pathways were mapped using the KEGG web service (http://geneontology.org/, accessed on 20 December 2024). Functional enrichment (GO/KEGG) was evaluated by Fisher’s exact test with Benjamini–Hochberg correction (FDR < 0.05). PPI networks were constructed by importing STRING or IntAct interactions into Cytoscape3.8.0; node degree was calculated to identify hub proteins.

2.11. Untargeted Metabolomic Analysis

2.11.1. Metabolite Extraction

Untargeted metabolomics was performed at Shanghai Applied Protein Technology Co., Ltd. (APT, Shanghai, China). Liver samples were thawed at 4 °C, mixed with ice-cold methanol/acetonitrile/water (2:2:1, v/v/v), vortexed, sonicated on ice for 30 min, kept at −20 °C for 10 min, and centrifuged (14,000× g, 4 °C, 20 min). The supernatant was dried in vacuo and reconstituted in 100 µL of acetonitrile/water (1:1, v/v) followed by a 15 min, 14,000× g spin at 4 °C; the resulting supernatant was used for LC–MS injection.

2.11.2. LC/MS Analysis

Metabolites were separated in a 2.1 × 100 mm ACQUITY BEH Amide column (1.7 µm) using an Agilent 1290 UHPLC (Agilent Technologies, Santa Clara, CA, USA) coupled to a TripleTOF 6600 (AB Sciex, Framingham, MA, USA). The mobile phases were A (25 mM NH₄Ac + 25 mM NH₄OH in water) and B (ACN); gradient: 95% B (0–0.5 min) → 65% B (4.5 min) → 40% B (5.5 min) → 95% B (6 min), hold for 2 min, 0.5 mL min⁻¹, 25 °C. ESI settings: Gas1/2 60, CUR 30, 600 °C, and ±5.5 kV. Full scan (m/z 60–1000): 0.2 s; IDA-MS² (m/z 25–1000): 0.05 s; CE: 35 ± 15 eV; DP: ±60 V; 10 candidate ions; and 4 Da isotope exclusion.

2.11.3. Metabolite Identification and Analysis

The raw data were converted to mzXML via ProteoWizard and processed in XCMS 3.12.0 (centWave: 10 ppm; peak width: 10–60 s; grouping bw: 5; m/z width: 0.025; minfrac: 0.5). Isotopes/adducts were annotated with CAMERA 1.42.0; features present in <50% of samples in either group were removed. Metabolites were identified by matching accurate mass (<10 ppm) and MS/MS spectra to an in-house standard library. After sum-normalization, multivariate analysis (PCA; OPLS-DA) was performed in ropls (R) with Pareto scaling; model validity was assessed by seven-fold cross-validation and permutation testing (n = 200). Significant changes were defined by VIP > 1 (OPLS-DA) and Student’s t-test p < 0.05; Pearson’s correlation evaluated relationships between variables.

2.12. Statistical Analysis

The data are expressed as means ± standard errors of the means (SEMs). Multiple comparisons were evaluated by one-way ANOVA followed by Fisher’s least significant difference (LSD) test (SPSS 19.0). p-value < 0.05 was considered statistically significant. The figures were generated using GraphPad 8.0 (San Diego, CA, USA).

3. Results

3.1. Propionate Protected Against Hepatic Steatosis in the ALD Mice

To investigate the direct hepatoprotective mechanism of propionate against ALD, an acute ALD model (Figure 1A) was established to minimize the potential involvement of the gut–liver axis. Serum ALT and AST levels were measured to assess liver function. As shown in Figure 1B,C, propionate significantly reduced the elevated levels of ALT and AST, demonstrating the hepatoprotective effect of propionate in acute ALD mice. In addition, histological analysis using H&E and Oil Red O staining (Figure 1D–F) revealed that alcohol intake induced hepatic steatosis, characterized by greater lipid droplet deposition in the livers of the EtOH group compared with the control group. In contrast, propionate treatments markedly alleviated this lipid accumulation. Consistently, hepatic TC and TG levels in ALD mice were significantly decreased after propionate supplementation (Figure 1G,H). The serum lipid parameters, including TC, TG, HDL-C and LDL-C, were also remarkably improved in propionate-treated ALD mice (Figure 1I–L). Furthermore, the expressions of proteins related to lipid metabolism in liver tissues were measured using Western blot analysis. It was observed that propionate treatments led to a significant downregulation of lipid synthesis protein (ACC1 and SREBP) expression, along with an upregulation of lipid oxidation protein (CPT1A) expression in ALD mice (Figure 1M,N). In summary, our findings indicated that propionate exerted a beneficial effect against acute ALD in mice, notably by improving dysregulated hepatic lipid metabolism.

3.2. Propionate Attenuated Lipid Metabolism Disorder In Vitro

Since the effect of propionate on hepatic lipid metabolism in ALD mice, we further investigated its effect on OA- and EtOH-induced lipid metabolism disorders in the hepatocyte AML-12 cell line. First, a series of dose–response experiments was conducted to identify the combination of ethanol and OA concentrations that effectively induced alcoholic liver disease (ALD)-related phenotypes. As depicted in Figure 2A, a concentration of ethanol ≤200 mM reduced cell viability without inducing substantial cell death. Figure 2B and Figure S1 show that 0.5 mM OA triggered evident lipid accumulation, although it exhibited cytotoxicity to AML-12 cells. Therefore, it was determined that 200 mM ethanol and 0.5 mM OA were the best concurrent treatment doses to induce evident cell injury and lipid accumulation in AML-12 cells. Additionally, assessment of the effect of propionate on cell viability revealed that it did not inhibit the growth of AML-12 cells; rather, higher concentrations of propionate enhanced cell proliferation (Figure 2C). On the basis of this, we selected 1 mM and 2 mM propionate for subsequent cell experiments. Consistent with our in vivo findings, propionate improved the viability of cells co-treated with ethanol/OA and suppressed intracellular lipid droplet formation in a dose-dependent manner (Figure 2D,E). Similarly, ethanol/OA co-treatment remarkably elevated the expressions of ACC1 and SREBP-1 while inhibiting the expression of CPT1A. Propionate supplementation, as expected, effectively counteracted this dysregulation in AML-12 (Figure 2F,G). Collectively, these findings further suggested that propionate restored the balance of lipid metabolism in ethanol/OA co-treated hepatocyte cells.

3.3. PA Suppressed Hepatic Oxidative Stress in ALD Mice

Oxidative stress, resulting from the hepatic metabolism of alcohol, also plays a critical role in the pathogenesis of alcoholic liver injury. Hence, to evaluate the effect of propionate on oxidative stress, we measured the levels of key antioxidant agents, including SOD, CAT and GSH, as well as the lipid peroxidation product MDA. As shown in Figure 3A–D, compared with the control group, alcohol intake led to a significant decrease in the hepatic levels of SOD, CAT and GSH, along with an increase in MDA content. However, propionate treatments reversed these alcohol-induced alterations, suggesting that propionate alleviated the hepatic oxidative stress in ALD mice. Moreover, the alcohol-induced increase in CYP2E1 expression was also reduced following propionate administrations, which implied a consequent decline in hepatic ROS production (Figure 3E,F). These results demonstrated that propionate protected liver damage in ALD mice by mitigating hepatic oxidative stress.

3.4. Propionate Altered the Liver Proteomic Profile in ALD Mice

To explore the hepatoprotective mechanism of propionate in ALD, we analyzed the total proteome alteration of liver tissues in mice using TMT-based quantitative proteomics. In this study, a total of 5641 proteins across all liver samples were identified, with 5637 proteins reliably quantified (Figure S2A). The differentially expressed proteins (DEPs) were selected with statistical thresholds of p-value < 0.05 and |fold change (FC)| > 1.2. Relative to the control group, significant alterations were observed in 505 proteins in the EtOH group (283 upregulated and 222 downregulated). This number was reduced to 334 in the propionate-fed group (213 upregulated and 121 downregulated) (Figure 4A,B and Figure S2B). Meanwhile, comparison between the propionate-treated and alcohol-fed groups revealed 112 DEPs, with 47 upregulated DEPs and 75 downregulated DEPs (Figure 4C and Figure S2B). The heatmap of overall DEPs among the three groups is presented in Figure 4D, which shows that the propionate treatments partially restored the alcohol-perturbed proteome. Subsequently, to further clarify the biological significance of these DEPs, Gene Ontology (GO) analysis was performed. As shown in Figure 4E, the DEPs between the EtOH and control groups were mainly involved in lipid metabolism and fatty acid metabolism in biological process analysis, oxidoreductase activity and monooxygenase activity in molecular function, and the endoplasmic reticulum part in cellular component analysis. In the comparison between the PA group and the EtOH group (Figure 4F), the DEPs caused by propionate treatments were significantly enriched in biological processes concerning the carboxylic acid metabolic process and the response to toxic substances and heat generation, which were the main functions involved in lipid metabolism, energy consumption and oxidative stress. Meanwhile, within the category of molecular function, cofactor binding and oxidoreductase activity were the top two enriched functions, which were closely associated with energy metabolism and oxidative stress. In the cell components, these DEPs were mainly related to the extracellular region. As expected, these results underscored that the hepatoprotective effect of propionate is mediated by the regulation of lipid metabolism and oxidative stress in ALD mice.

Further KEGG pathway enrichment analysis of these DEPs was performed. Compared with the control group, DEPs induced by alcohol exposure were enriched in pathways related to chemical carcinogenesis–DNA adducts, xenobiotic metabolism by cytochrome P450 and drug metabolism–cytochrome P450 (Figure S2C). After propionate treatment, the enriched metabolic pathways similarly included xenobiotic metabolism by cytochrome P450 and chemical carcinogenesis–DNA adducts, along with chemical carcinogenesis–reactive oxygen species (Figure 5A). To delineate the directional regulation exerted by alcohol and propionate, separate enrichment analyses for up- and downregulation DEPs were conducted. As shown in Figure S2D, upregulated proteins induced by alcohol exposure were predominantly enriched in arachidonic acid metabolism and linoleic acid metabolism pathways, whereas downregulated proteins were strongly associated with the pentose phosphate pathway and phosphonate/phosphinate metabolism. Notably, in the propionate-treated group, lipid catabolic pathways, including the PPAR signaling pathway and fatty acid degradation, were upregulated, demonstrating a restoration of lipid homeostasis. Concurrently, pathways linked to oxidative stress and xenobiotic metabolism, such as glutathione metabolism and metabolism of xenobiotics by cytochrome P450, were markedly downregulated (Figure 5B). Then, to further validate these pathway-level findings at the individual protein level, we analyzed relevant protein changes and showed them in a heatmap. As depicted in Figure 5C,D, most members of the fatty acid degradation (Lipa, RGN, Bdh1, Decr1, Cpt1a, Hadha and Etfdh) and PPAR signaling pathways (Cyp4a12a, Cyp4a12b, Dbi and Ppara) showed increased abundance following propionate treatment. Within the oxidative stress-related pathway, propionate significantly increased the antioxidant-related proteins’ expression, such as Prdx5, RGN, Glo and Prdx2, while notably decreasing detoxification-related proteins, including Gstm2, Gstm5 and Cbr (Figure 5E). Taken together, the coordinated upregulation of lipid utilization, including lipolysis and fatty acid oxidation, and antioxidant capacity supported a dual mechanism through which propionate alleviated alcoholic liver injury.

3.5. Propionate Altered the Liver Metabolism in ALD Mice

Untargeted metabolomic analysis of liver tissues was performed to identify metabolite variations after propionate treatment in both positive and negative ionization modes. Principal component analysis (PCA) showed tight clustering of quality control samples and clear separation among the three experimental groups, confirming the high reproducibility of the data (Figure 6A,B). Further OPLS-DA revealed distinct metabolic profiles between the EtOH and control groups (Figure S3A,B), as well as between the propionate-treated and EtOH groups (Figure 6C,D), demonstrating that the substantial alteration in the hepatic metabolome was induced by the alcohol and propionate treatments. Metabolites with a VIP score of > 1.0 and p < 0.05 were considered statistically significant contributors to group separation and were selected for differential analysis. Between the EtOH and control groups, 199 differentially expressed metabolites (DEMs) were identified across both ionization modes (Figure 6E,F). In comparison with the EtOH group, propionate treatment significantly altered the abundance of 14 metabolites in positive mode and 55 metabolites in negative mode (Figure 6E,F). In positive mode, these DEMs were mainly categorized as benzenoids and lipid/lipid-like molecules (Figure 6G). In negative mode, the most affected metabolites belonged to lipid/lipid-like molecules and organic acids/derivatives (Figure 6H).

KEGG pathway enrichment analysis was performed to elucidate the key metabolic pathways regulated by propionate in ALD mice. The most significantly enriched pathways included glycerophospholipid metabolism, biosynthesis of unsaturated fatty acids, and the cAMP signaling pathway (Figure 7A). Metabolite Set Enrichment Analysis (MSEA) further indicated that pathways most markedly altered by propionate treatment were predominantly related to lipid metabolism, such as fatty acid metabolism, degradation, and biosynthesis (Figure 7B). Subsequently, key DEMs associated with metabolic pathways were compared across the three groups. As shown in Figure 7C,D, ethyl glucuronide and ethyl sulfate, the direct markers of alcohol metabolism [16], were both reduced in the propionate-treated group, suggesting the accelerated clearance of alcohol and its toxic metabolites in ALD mice. Interestingly, several fatty acids, including pentadecanoic acid, palmitic acid and myristic acid, were significantly increased following propionate treatment in ALD mice, likely due to enhanced lipolysis. Furthermore, correlation and network analysis revealed that the metabolites altered by propionate were predominantly associated with fatty acid and phospholipid metabolism. Within this network, pentadecanoic acid emerged as a central hub exhibiting the strongest correlations with other differential metabolites (Figure 7E,F). Taken together, these metabolomic findings reinforced our hypothesis that propionate ameliorated ALD through regulation of lipid metabolism.

3.6. The Integrated Analysis of Proteomic and Metabolomic

The integrated proteomic and metabolomic analyses were employed to elucidate the pivotal functional nodes of propionate action in ALD mice. As shown in Figure 8A–C, the DEPs and DEMs induced by alcohol exposure co-participated in 94 shared pathways, which were mainly related to biosynthesis of cofactors, metabolism of xenobiotic cytochrome P450 pathways, and arachidonic acid metabolism. Among them, the biosynthesis of cofactors and biosynthesis of unsaturated fatty acids were the most significantly enriched pathways. Compared with the EtOH group, the DEPs and DEMs altered by propionate administration were co-enriched in 36 KEGG pathways (Figure 8D). The top 10 pathways with the highest molecule counts were primarily associated with biosynthesis of cofactors, drug metabolism and bile secretion (Figure 8E). Cholesterol metabolism and arachidonic acid metabolism showed particularly significant enrichment (Figure 8F). Furthermore, the correlation heatmap of propionate-induced DEPs and DEMs revealed that several proteins involved in the lipolysis pathway and PPARα signaling, including Lipa (Q9Z0M5), RGN (Q64374), Cyp4a12a (Q91WL5), Cyp4a12b (A2A974) and Dbi (P31786), were positively correlated with fatty acid metabolites such as palmitic acid, pentadecanoic acid and myristic acid (Figure 8G). These results suggested that propionate may restore hepatic metabolic homeostasis by coordinating protein expression and metabolite accumulation, especially those involved in lipid metabolism.

3.7. Propionate Upregulated RGN-PPARα Signaling in ALD Mice

To identify the key target regulated by propionate, we analyzed the top 15 downregulated proteins after alcohol exposure and the top 15 upregulated proteins following propionate treatment.

Table 2. Top 15 downregulated (EtOH/Pair) and upregulated (PA/EtOH) DEPs.

	Protein	EtOH/Pair	p-Value	Protein	PA/EtOH	p-Value
1	Selenbp2	0.394236566	0.000406755	Mup1	1.717812485	0.00948677
2	Fabp5	0.505568348	0.001085132	Orm1	1.630910175	0.002884733
3	Mup1	0.525579171	0.004906254	Mup17	1.542329918	0.009413115
4	Fdps	0.539585015	6.6498 × 10−5	Mup3	1.436333506	0.001509068
5	Dbi	0.545323409	0.001863242	Hsd17b6	1.327172443	0.000974639
6	Selenbp1	0.547080405	0.000186247	Sugct	1.324584274	0.000567731
7	Pfkfb1	0.572393537	2.36968 × 10−5	Cyp3a11	1.31981035	0.017090046
8	Ldha	0.582398192	0.003369139	Ndufb3	1.307477744	0.006988677
9	Bpifa3	0.582595251	0.005149358	Lipa	1.300033722	0.002755613
10	Fabp1	0.603159633	0.020336272	Fabp2	1.298070357	0.005501241
11	RGN	0.607507231	4.44793 × 10−5	Cyp4a12a	1.28783657	0.024820197
12	Hsd3b5	0.612709438	0.00165556	Bdh1	1.286202476	0.00171725
13	Rnase4	0.613267523	0.010753209	RGN	1.283055217	0.000930183
14	Nme1	0.615888052	0.019934244	Col11a2	1.2763713	0.00895952
15	Mup3	0.62987867c8	0.00111876	Cyp4a12b	1.27631226	0.049248381

shows that alcohol exposure significantly suppressed the expression of Mup1, RGN and Mup3, whereas propionate treatment markedly upregulated their levels. A protein–protein interaction (PPI) network analysis further identified key regulatory proteins such as RGN, cyp4a12a, Mup3, Cyp3a11 and Otc as central nodes within the network, highlighting their function as key mediators through which propionate ameliorates ALD (Figure 9A). Given the established role of RGN in activating PPARα signaling and alleviating oxidative stress [17,18], we hypothesized that propionate might exert its hepatoprotective effects by targeting RGN. To test this hypothesis, we first performed molecular docking to investigate the potential interaction between propionate and RGN. The molecular docking model revealed that propionate could bind within the active pocket of the RGN protein, forming hydrogen bonds with multiple residues, including Asn154, Asn103, Met118, and Arg101 (Figure 9B,C), suggesting a stable and specific interaction between propionate and RGN. Subsequently, we detected RGN expression in liver tissues and found that propionate treatments completely reversed the alcohol-induced decrease in RGN levels. Furthermore, propionate administration significantly increased the expression of key components in the RGN-regulated PPARα pathway, namely, PPARα, ACOX1, and p-AMPK (Figure 9D,E). Collectively, these results demonstrated that propionate may activate the RGN-PPARα signaling pathway, which, in turn, regulates lipid metabolism and attenuates oxidative stress in mice with alcohol-induced liver injury.

4. Discussion

Propionate, a key metabolite derived from the gut microbiota, has demonstrated considerable therapeutic potential across a spectrum of diseases, including hypertensive cardiovascular damage, Alzheimer’s disease, and non-alcoholic steatohepatitis [9,19,20]. Building upon our previous findings that propionate alleviated ALD by protecting intestinal barrier function, this study aimed to further investigate its direct hepatoprotective effects and elucidate the underlying mechanism in ALD. Through the integrated multi-omics study, we demonstrated that propionate exerted direct hepatoprotective effects on ALD potentially through activating the RGN-PPARα signaling pathway, which coordinately rectified lipid metabolic disorders and mitigated oxidative stress.

The pathogenesis of ALD, though multifaceted, centers on a synergistic interplay between hepatic lipid metabolic dysfunction and oxidative stress, which together constitute a core driver of ALD progression [4]. It is reported that lipid homeostasis is profoundly deranged in patients with ALD, with a global increase in lipogenesis and reduced lipid oxidation, leading to lipid accumulation in hepatocytes [21,22]. This is evidenced by the downregulation of key oxidation proteins (CPT1 and CPT2) and the upregulation of lipogenic regulators (SREBP1, ACC1 and Fasn) [23,24]. In turn, the lipid overload induces lipotoxicity, causing endoplasmic reticulum stress and mitochondrial dysfunction [25,26]. Furthermore, steatotic hepatocytes consequently become more susceptible to external toxic signals, resulting in apoptosis and other forms of cell death. In parallel, the excess alcohol metabolism via cytochrome P450 2E1 (CYP2E1) is a major source of reactive oxygen species (ROS) [27], which induces damage to hepatocyte membranes, organelles, and macromolecules. This oxidative damage can be countered through oxidative defense systems, particularly through glutathione (GSH). However, alcohol consumption depletes GSH, leading to profound redox imbalance [4]. Consequently, the elevated oxidative stress results in lipid peroxidation, thereby generating reactive aldehydes like malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE), which further exacerbate hepatocyte damage by promoting inflammation and impairing mitochondrial function.

Our present findings provided direct evidence that propionate intervention disrupted this vicious cycle at both levels. On the one hand, propionate significantly ameliorated hepatic steatosis in ALD mice, as shown by reduced lipid content in the liver and serum, alongside a restored expression of metabolic regulators: upregulated CPT1A and downregulated ACC1 and SREBP1 (Figure 1). This beneficial effect on lipid homeostasis was corroborated in vitro, where propionate suppressed lipid accumulation in AML-12 hepatocytes co-treated with ethanol and oleic acid (Figure 2). On the other hand, propionate robustly attenuated hepatic oxidative stress, normalizing the levels of key antioxidants (SOD, CAT, and GSH) and reducing the lipid peroxidation product MDA. The decreased expression of CYP2E1 also suggested a reduction in the primary enzymatic source of ROS (Figure 3).

Indeed, the proteomic and metabolomic data further elucidated the regulatory role of propionate in lipid metabolism and oxidative stress pathways in ALD mice. Relative to ethanol-exposed mice, propionate administration markedly induced a cohort of proteins that are involved in lipolysis and mitochondrial beta-fatty acid oxidation, including Lipa, Bdh1, Etfdh, Decr, Cyp4a12a, Cyp4a12b, Hadha, Hadhb, Dbi, Cpt1a, and Cpt2 [28,29,30]. This proteomic signature indicated that propionate treatment enhanced hepatic degradation, thereby alleviating lipid overload (Figure 5). Correspondingly, metabolomic profiling revealed that propionate broadly influenced lipid metabolic pathways in ALD mice, particularly those involved in fatty acid biosynthesis, elongation and degradation (Figure 7). Notably, an interesting finding from the metabolomic data was that the hepatic contents of several medium- to long-chain fatty acids (pentadecanoic acid, palmitic acid and myristic acid) were significantly increased upon propionate treatment in ALD mice. While this finding appears contrary to the simplistic expectation that stimulated β-oxidation should deplete its substrates [31], it is consistent with the overarching conclusion that propionate promoted lipid metabolic homeostasis. We proposed that ethanol induced impairment of lipolysis and suppression of fatty acid oxidation, thereby redirecting fatty acids toward esterification and storage, which reduced their detectable free pools in the liver. Propionate treatment reversed this dysregulation by enhancing both lipid mobilization (lipolysis) and oxidative disposal, which was supported by the concomitant upregulation of key proteins related to lipolysis and β-oxidation, such as Lipa, Hadha and Bdh1, in proteomic analysis. The protein Lipa (lysosomal acid lipase), a critical lysosomal enzyme that hydrolyzes cholesteryl ester (CE) and TG, has been reported to alleviate lipid droplet metabolism disorder in non-alcoholic fatty liver disease [32]. Bdh1 (β-hydroxybutyrate dehydrogenase 1), a pivotal enzyme in ketone body metabolism, facilitates the interconversion between β-hydroxybutyrate and acetoacetate. Recent studies have demonstrated that genetic ablation of Bdh1 exacerbated hepatic steatosis [33], whereas its overexpression ameliorated liver injury in fatty liver mice [34]. Similarly, Hadha (hydroxyl CoA dehydrogenase alpha subunit) is a core component of the mitochondrial trifunctional protein complex that regulates β-oxidation flux. Recent evidence indicated that acetylation of Hadha modulated hepatic fatty acid oxidation, positioning it as a key node in fatty liver pathogenesis [35]. Its upregulation alleviated lipid accumulation and oxidative stress, whereas its knockdown aggravated liver damage in non-alcoholic fatty liver disease [36]. Based on these findings, the transient rise in hepatic free fatty acids observed after propionate treatment may, thus, reflect a Lipa-driven surge in lipolytic flux, which momentarily increases the substrate pool for β-oxidation. It is important to note that this study employed an acute ALD model with tissue collection at a relatively early time point post-treatment. Under these conditions, the enhanced β-oxidation capacity signaled by elevated Bdh1 and Hadha may not yet have reached maximal flux, permitting a temporary accumulation of fatty acids prior to their accelerated clearance. This premise was further supported by the observed positive correlation between the levels of these fatty acids and key β-oxidation enzymes in propionate-treated mice (Figure 8). Furthermore, the restoration of certain saturated fatty acids may not be purely adverse in the context of ALD. Prior studies have reported that specific saturated fatty acid diets, such as 12% palmitic acid plus 85% stearic acid, exerted protective effects in ALD models [37]. For instance, pentadecanoic acid has been associated with reduced hepatic steatosis via upregulating peroxisome proliferator-activated receptor α/δ (PPARα/δ) [38,39]. In addition, proteomic profiling indicated a pronounced upregulation of hepatic antioxidant defenses at the protein level, characterized by increased expression of Prdx5, Prdx2 and Glo1 alongside reduced levels of GSTMs (Figure 5). This expression pattern aligned with prior evidence suggesting that propionate treatment promoted a transition toward a reparative phase following oxidative stress [40,41,42]. Altogether, propionate remodeled hepatic lipid metabolism by concurrently stimulating lipolysis, enhancing mitochondrial β-oxidation, and fortifying cellular antioxidant capacity. This coordinated response facilitated a metabolic shift away from lipotoxic lipid storage toward a state of sustained lipid metabolic homeostasis.

Subsequently, we further demonstrated that this regulation of propionate in lipid homeostasis was mechanistically linked to activation of RGN-PPARα signaling axis. To elucidate the molecular mechanism underlying the alleviation of ALD by propionate, we focused on the most significantly altered pathways and proteins identified in our multi-omics analysis. KEGG pathway enrichment analysis highlighted the upregulation of PPARα signaling as a pivotal link between propionate treatment and ALD improvement (Figure 5). Accumulating evidence has indicated that PPARα plays a central role in hepatic lipid homeostasis, and its suppression in ALD contributes to reduced fatty acid oxidation, enhanced de novo lipogenesis, and consequent hepatic steatosis [43]. Pharmacological activation of PPARα has, thus, been shown to exert beneficial effects on ALD, largely through transcriptional regulation of downstream targets involved in lipid metabolism, such as Cyp4a12a/b, Dbi, Acox1, Fasn and CPT1 [44,45,46]. Usually, PPARα activity is modulated by several upstream regulators, including AMPK, Sirt1 and ATGL [47,48,49]. It is worth noting that Regucalcin (RGN), also known as senescence marker protein 30 (SMP30), has also been reported to participate in PPARα activation [17]. RGN is a calcium-binding protein highly expressed in the liver and kidneys, where it helps maintain intracellular calcium homeostasis and modulates multiple signaling cascades. Substantial evidence supports RGN as a key protective molecule in liver pathology. RGN expression is markedly downregulated in various liver disorders, including ALD, non-alcoholic liver disease, and drug-induced liver injury [50]. Restoring RGN expression mitigated liver damage through multiple mechanisms: suppression of ROS accumulation and enhancement of SOD activity, leading to attenuated oxidative stress [18,51]; reducing hepatic triglyceride and total cholesterol accumulation, thereby promoting lipid metabolic homeostasis [52,53]; and inhibition of inflammatory cell infiltration and hepatocyte apoptosis [54]. In our study, RGN protein levels were significantly decreased after ethanol exposure and were robustly restored by propionate treatments (Table 2). Molecular docking and PPI network analyses further supported RGN as a relevant target of propionate in ALD, indicating that propionate may bind to RGN and enhance its protein stability and activity, thereby counteracting the decrease in RGN level in ALD. Therefore, it is reasonable to believe that propionate alleviated ALD, at least in part, by regulating RGN to activate PPARα signaling. Consistent with this, our results showed propionate treatments increased the expressions of hepatic RGN, p-AMPK, PPARα, ACOX1 and CPT1A in ALD mice (Figure 1 and Figure 9). Additionally, given that the circadian clock is a known regulator of hepatic lipid metabolism and PPARα signaling [55,56], it would be interesting to investigate in future studies whether the RGN-PPARα axis identified here is influenced by circadian rhythms, particularly in the context of ALD, where clock disruption has been reported. Taken together, our findings establish RGN-PPARα signaling as a critical axis through which propionate exerts its hepatoprotective effects, while also highlighting the potential of propionate-based dietary strategies as a practical avenue for future ALD management. From a translational standpoint, it is worth briefly noting that while propionate can be derived from gut microbial fermentation of dietary fiber, the concentrations achieved through this route are not comparable to those achieved by oral propionate administration in drinking water [12,57]. Currently, inulin-propionate ester [58] and propionylated high-amylose maize starch [59], which mimic the endogenous release pattern of propionate via gut microbiota fermentation and enable substantial in vivo delivery, represent feasible strategies for efficient propionate supplementation and warrant further investigation in the context of ALD.

5. Conclusions

In summary, our present study demonstrated that propionate administration reshaped hepatic proteomic and metabolomic profiles, concurrently promoting lipolysis, enhancing mitochondrial β-oxidation, and strengthening antioxidant defenses, thereby ameliorating liver damage in an acute ALD model. Mechanistically, propionate treatments specifically upregulated the expression of RGN in liver tissues, leading to activation of the downstream PPARα signaling via increased levels of p-AMPK, PPARα, ACOX1 and CPT1A. These findings not only advanced the mechanistic understanding of propionate in liver protection but also highlighted the potential of propionate as a therapeutic candidate for ALD.