Qushi Huoxue ointment ameliorates metabolic associated steatotic liver disease through autophagy activation and ferroptosis inhibition

Abstract

BACKGROUND

Metabolic associated steatotic liver disease (MASLD) has become a growing global health burden, with its potential to progress to liver fibrosis, cirrhosis, and hepatocellular carcinoma. Qushi Huoxue ointment (QSHXO), a traditional Chinese medicine formula, has demonstrated efficacy in the management of MASLD. However, its underlying mechanisms remain incompletely elucidated.

AIM

To investigate the mechanism by which QSHXO alleviated hepatic lipid deposition and inflammatory injury in MASLD, with a focus on its role in activating hepatocyte autophagy and inhibiting ferroptosis.

METHODS

This study employed a comprehensive research strategy. First, a methionine-choline-deficient diet-triggered MASLD mouse model was established and treated with different doses of QSHXO. The therapeutic effects of QSHXO were comprehensively evaluated using histological analysis, serum biochemical assays, and inflammatory cytokine measurements. Subsequently, bioactive components of QSHXO in serum were identified utilizing liquid chromatography-tandem mass spectrometry. Network pharmacology was then applied to predict potential targets of QSHXO in treating MASLD related to autophagy and ferroptosis. These predicted targets were validated through western blotting, quantitative reverse-transcription polymerase chain reaction, immunohistochemistry, and transmission electron microscopy.

RESULTS

QSHXO significantly ameliorated liver lipid deposition and inflammation in MASLD mice. Specifically, QSHXO promoted autophagic flux, as indicated by upregulation of Beclin1, an increased light chain 3 II/light chain 3 I ratio, and downregulation of P62. Concurrently, QSHXO activated the nuclear factor erythroid 2-related factor 2 pathway, promoting its nuclear translocation and enhancing the expression of downstream targets (SLC7A11 and glutathione peroxidase 4), while reducing hepatic iron deposition; these collectively suggested suppression of ferroptosis. Ultrastructural analysis further confirmed improved mitochondrial morphology and increased autophagic vesicles in QSHXO-treated groups.

CONCLUSION

QSHXO ameliorates MASLD by reducing lipid accumulation, mitigating inflammation, and suppressing hepatocyte damage, which is mediated through the activation of autophagy and inhibition of ferroptosis.

Key Words: Metabolic associated steatotic liver disease; Qushi Huoxue ointment; Autophagy; Ferroptosis; Lipid deposition

Core Tip: Qushi Huoxue ointment can alleviate hepatic lipid accumulation, inflammation, and cell injury in metabolic associated steatotic liver disease mice. Our experimental evidence suggests that these therapeutic effects may be attributed to the concurrent activation of autophagy and inhibition of ferroptosis. The improvement in mitochondrial morphology and the presence of autophagosomes observed under the microscope provide morphological corroboration for this coordinated mechanism.

INTRODUCTION

Metabolic associated steatotic liver disease (MASLD) has become a highly prevalent chronic liver disorder worldwide, with an ever-increasing incidence. The global prevalence of MASLD is approximately 38%[1], whereas in China, the rate has risen rapidly in recent years, reaching about 29%[2]. MAFLD precisely highlights the core pathogenic role of metabolic dysfunction in the occurrence and progression of the disease. Metabolic disorder phenotypes such as obesity, type 2 diabetes mellitus, metabolic syndrome, and cardiovascular diseases are often accompanied by systemic multi-organ dysfunction[3]. Its clinical spectrum ranges from simple steatosis to metabolic associated steatotic hepatitis, fibrosis, cirrhosis, and potentially hepatocellular carcinoma[4]. Although lifestyle interventions, such as dietary management and increased physical activity, remain first-line strategies[5], available pharmacological options for advanced diseases are limited. Current drug therapies often demonstrate restricted efficacy and are accompanied by notable side effects[6]. To date, no specific medication has been approved by the United States Food and Drug Administration for the treatment of MASLD[7].

Traditional Chinese medicine (TCM) has gained increasing recognition for its multi-target approach, minimal adverse effects, and potential in regulating physiological functions and enhancing immunity[8,9]. With its historical basis in treating lipid metabolism disorders and protecting liver function, TCM represents a promising therapeutic alternative for MASLD[10,11]. Among various traditional Chinese medicine formulations, Qushi Huoxue ointment (QSHXO) stands out as particularly deserving of in-depth research, yet the knowledge regarding its multitarget mechanisms remains lacking. QSHXO was developed based on the TCM theory that “phlegm and blood stasis obstruction” constitutes the core pathogenesis of MASLD, often accompanied by spleen deficiency and qi stagnation. Formulated by renowned TCM expert professor Pei-Yu Zhou, this prescription integrates therapeutic principles of eliminating dampness, clearing turbidity, promoting blood circulation, resolving stasis, regulating qi, and strengthening the spleen. Since 2009, professor Xu-Dong Liu team has systematically investigated QSHXO through both clinical and mechanistic studies, demonstrating QSHXO’s efficacy in improving liver function, reducing hepatic inflammation, and decreasing lipid deposition[12-15]. Despite these promising findings, the active components of QSHXO and their precise molecular mechanisms remain incompletely understood.

Autophagy and ferroptosis have attracted increasing research interest due to their potential roles in disease pathogenesis. Autophagy, an essential cellular clearance process, maintains hepatic lipid homeostasis by degrading lipid droplets (LDs) and damaged organelles[16]. It has been shown that dysregulation of autophagy, manifested as impaired autophagic flux, defective lipophagy, and compromised antioxidant responses, is closely linked to MASLD progression[17]. Conversely, ferroptosis is a regulated cell death characterized by iron dependence and lipid peroxidation and is tightly associated with hepatocyte injury and fibrosis in MASLD[18,19]. Notably, a complex crosstalk exists between autophagy and ferroptosis, wherein selective autophagy processes can either facilitate or suppress ferroptotic cell death[20]. In this study, we selected oltipraz as the positive control (Ctrl) drug for in-vivo experiments, primarily based on its potential in regulating autophagy and ferroptosis pathways. As a nuclear factor erythroid 2-related factor 2 (Nrf2) activator, oltipraz has been extensively applied in research on metabolic diseases and steatohepatitis[21,22]. It activates the Nrf2 signaling to alleviate oxidative stress, thereby inhibiting ferroptosis[23]. Meanwhile, Nrf2 can alleviate oxidative stress and inflammation while promoting autophagy, ultimately improving cellular function[24]. Compared to traditional positive Ctrl drugs, oltipraz’s multi-mechanistic profile makes it an ideal candidate for investigating the interplay between autophagy and ferroptosis. However, the role of this interaction in MASLD remains underexplored. In the present study, network pharmacology prediction was integrated with experimental validation to address this gap. Using a methionine-choline-deficient (MCD) diet-triggered mouse model, we revealed that QSHXO ameliorated MASLD by concurrently reactivating autophagy and inhibiting ferroptosis. Our findings may lay a foundation for investigating the future therapeutic applications of QSHXO.

MATERIALS AND METHODS

The overall experimental design is shown in Figure 1.

Figure 1 Experimental design summary diagram. Metabolic associated steatotic liver disease mouse model was first established using the methionine-choline-deficient diet. Following Qushi Huoxue ointment (QSHXO) treatment, serum and liver tissue samples were collected to assess pathological changes, liver function markers, and inflammatory cytokine levels. Subsequently, the bioactive components of QSHXO in serum were identified using liquid chromatography-mass spectrometry/mass spectrometry analysis. Next, network pharmacology was applied to predict the potential target pathways of QSHXO in the treatment of metabolic associated steatotic liver disease, specifically focusing on autophagy and ferroptosis. These predicted targets were further validated through western blot, quantitative reverse-transcription polymerase chain reaction, immunohistochemistry, and transmission electron microscopy. MCD: Methionine-choline-deficient; MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment; TNF-α: Tumor necrosis factor-α; IL-β: Interleukin-β; ELISA: Enzyme-linked immunosorbent assay; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TC: Total cholesterol; TG: Triglyceride; Nrf2: Nuclear factor erythroid 2-related factor 2; Lc3: Light chain 3; GPX4: Glutathione peroxidase 4.

Materials and reagents

The following reagents and antibodies were used in this study: Alanine aminotransferase (ALT) test kit (C009-2-1, Nanjing Jiancheng, Nanjing, Jiangsu Province, China); aspartate aminotransferase (AST) test kit (C010-2-1, Nanjing Jiancheng, Nanjing, Jiangsu Province, China); total cholesterol (TC) test kit (A111-1-1, Nanjing Jiancheng, Nanjing, Jiangsu Province, China); triglyceride (TG) test kit (A110-1-1, Nanjing Jiancheng, Nanjing, Jiangsu Province, China); tumor necrosis factor-α (TNF-α, EK201BHS, LianKe Bio, Hangzhou, Zhejiang Province, China); interleukin (IL)-1β (EK282/4, LianKe Bio, Hangzhou, Zhejiang Province, China); oltipraz (B5958, APExBIO, TX, United States); Beclin1 (AF5128, 1:1000, Affinity, Shanghai, China); light chain 3 (LC3) (AF5402, 1:1000, Affinity, Shanghai, China); P62 (AF5384, 1:1000, Affinity, Shanghai, China); Nrf2 (AF0639, 1:800, Affinity, Shanghai, China); glutathione peroxidase 4 (GPX4) (A11243, 1:1000, Abclonal, Wuhan, Hubei Province, China); SLC7A11 (A2413, 1:1000, Abclonal, Wuhan, Hubei Province, China); and lamin B1 (BF8009, 1:2000, Affinity, Shanghai, China).

Preparation of QSHXO

QSHXO was provided by the TCM Pharmacy of Ruikang Hospital, Affiliated to Guangxi University of Chinese Medicine. The formula consists of 17 medicinal herbs: Bupleurum (Chai Hu) 12 gram, Artemisia capillaris (Yin Chen, 15 gram), Salvia miltiorrhiza (Dan Shen, 15 gram), Lotus leaf (He Ye, 10 gram), Gynostemma (Jiao Gu Lan, 20 gram), Hawthorn (Shan Zha, 20 gram), Cassia seed (Jue Ming Zi, 25 gram), Licorice (Gan Cao, 6 gram), Curcuma (E Zhu, 15 gram), Polygonum cuspidatum (Hu Zhang, 15 gram), Astragalus (Huang Qi, 20 gram), Citrus reticulata (Chen Pi, 10 gram), Poria (Fu Ling, 20 gram), Atractylodes (Bai Zhu, 15 gram), Coix seed (Yi Yi Ren, 30 gram), Alisma (Ze Xie, 15 gram), and Polyporus umbellatus (Zhu Ling, 10 gram). All herbs were mixed and immersed in 10 volumes of distilled water for 30 minutes, followed by a 1-hour decoction. After filtering the extract, the residue was re-decocted with 8 volumes of water for 1-hour. The combined filtrates were concentrated into an ointment at a final concentration of 2.19 g crude drug per gram.

Animal preparation, grouping, and drug administration

A total of 60 specific-pathogen-free male C57BL/6J mice [20 ± 2 gram, 6-8 weeks; Jinwei Biotechnology, Guangzhou, Guangdong Province, China; SCXK (Beijing) 2024-0001] were prepared. We housed animals in a controlled environment (23 ± 1 °C, humidity 55%-60%, 12 hours light/dark cycle). Sterile standard diet and water were provided ad libitum. Our experiments got approval from the Animal Welfare and Ethics Committee of Guangxi University of Chinese Medicine (Approval No. DW20240919-186).

Following 1-week acclimatization, mice were grouped at random (n = 10 per group): (1) Ctrl; (2) MASLD; (3) MASLD + high-dose QSHXO (QSHXO-H, 9 g/kg); (4) MASLD + medium-dose QSHXO (QSHXO-M, 4.5 g/kg); (5) MASLD + low-dose QSHXO (QSHXO-L, 2.25 g/kg); and (6) MASLD + oltipraz (positive Ctrl group, oltipraz, 50 mg/kg). Mice in the Ctrl group received a standard diet throughout the experiment. MASLD was induced in all groups except the Ctrl group by feeding a MCD diet for 10 weeks. The MCD diet (45.5% sucrose, 15.0% corn starch, 10.0% corn oil, 0.0% methionine, 0.0% choline, No. SFD024) was provided by SBEF Biotechnology Company Limited (Beijing, China).

After successful modeling, drug interventions were administered for 6 weeks. Specifically, QSHXO was dissolved in distilled water and administered once daily by oral gavage. Oltipraz was prepared in corn oil containing 5% dimethyl sulfoxide and injected intraperitoneally once daily. The Ctrl and MASLD groups received daily oral gavage of physiological saline as a vehicle treatment. Throughout the intervention period, all MCD-fed mice continued on the MCD diet to maintain the disease state. It should be noted that during the entire 16-week study period, which included a 10-week modeling phase and a subsequent 6-week intervention phase, mice in the Ctrl group were continuously fed a standard chow diet.

Histological analysis

Following the intervention, the mice were anesthetized with 10% chloral hydrate. Initially, blood samples were collected, and the heart was then perfused with frozen phosphate buffered saline to remove the blood. Liver tissues were harvested via median laparotomy and fixed in 4% paraformaldehyde. After paraffin embedding, the tissues were sectioned (5 μm) and stained with hematoxylin and eosin (H&E) for evaluation of general histoarchitecture. For lipid visualization, frozen sections were stained with Oil Red O (ORO) working solution and counterstained with hematoxylin. A light microscope was employed for section observation.

Serum biochemical analysis

Blood was obtained from the retro-orbital plexus, followed by centrifugation (3000 × g, 10 minutes, 4 °C) to isolate serum for subsequent analysis. Serum levels of ALT, AST, TC, and TG were examined using commercial assay kits according to the protocols. Absorbance was measured using a microplate reader. The concentrations were calculated based on standard curves. All samples were analyzed in triplicate.

Enzyme-linked immunosorbent assay

Enzyme-linked immunosorbent assay kits were utilized to quantify TNF-α and IL-1β serum levels as per the protocol. Absorbance was read at 450 nm. Cytokine levels were determined from standard curves. All measurements were conducted in triplicate.

Identification of blood components of QSHXO

QSHXO’s chemical composition in serum was analyzed through ultra-performance liquid chromatography-tandem mass spectrometry. Briefly, serum or drug solution (50 μL) was added with 200 μL of methanol, vortexed (10 minutes), and then centrifuged (13000 rpm, 10 minutes), with the supernatant harvested for analysis. The high-resolution liquid chromatography-mass spectrometry (LC-MS) data collected were processed with compound discoverer 3.3 (Thermo Fisher, MA, United States), followed by a database (mzCloud) search and comparison.

Network pharmacology predictive relevance target analysis

To identify the 17 active components of QSHXO and their corresponding target associations, we first retrieved comprehensive TCM ingredient-target relationships from the BATMAN-TCM 2.0 database (http://bionet.ncpsb.org.cn/batman-tcm/#/home) and matched the major components of TCM with their related targets[25]. Potential MASLD-related targets were identified by screening the comparative toxicogenomics database (https://ctdbase.org/) with a correlation score above 100[26]. Additionally, ferroptosis-related genes (including ferroptosis markers and driver genes) were extracted from the FerrDb database, while the autophagy-related genes were obtained from the autophagy database autophagy database. A Venn diagram was then generated using the ggvenn package in R to visualize the intersections among potential TCM targets in MASLD, ferroptosis, and autophagy functions. Based on these intersections, the TCM-disease-ferroptosis target map and TCM-disease-autophagy target map were constructed. Next, protein-protein interaction (PPI) data for the overlapping targets were retrieved from the STRING database (https://cn.string-db.org/) and the resulting interaction network was constructed and visualized through Cytoscape 3.10.1 software. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were conducted using the clusterProfiler package in R. The top 10 GO-enriched functions are presented in a bar plot, and the top 20 enriched KEGG signaling pathways are shown in a bubble chart.

Western blot analysis

Liver tissues (about 50 mg) were homogenized in RIPA lysis buffer supplemented with protease inhibitors. A bicinchoninic acid assay kit was employed for protein concentration determination. Equal amounts of protein (30 μg) were subjected to 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis for separation, followed by transfer to polyvinylidene fluoride membranes. After blocking in 5% skim milk, membranes underwent incubation (overnight, 4 °C) with primary antibodies against Beclin1, LC3, p62, Nrf2, and GPX4 (dilutions as listed in materials and reagents). Next, goat anti-rabbit secondary antibody was added for incubation (room temperature, 1 hour). Protein bands were visualized using electrochemiluminescence substrate and quantified with ImageJ software. All experiments included Ctrl groups were analyzed with three replicates for each group to ensure the reliability and reproducibility of the results.

Quantitative real-time polymerase chain reaction

Total RNA was isolated from liver tissues utilizing TRIzol reagent (Invitrogen, MA, United States). cDNA was synthesized from 1 μg RNA using a reverse transcription kit. The quantitative reverse-transcription polymerase chain reaction (qRT-PCR) reaction was performed using a reaction system containing SYBR Green PCR Master Mix, specific primers, and cDNA template. The reaction conditions were: 95 °C for 5 minutes, followed by 40 cycles of 95 °C for 15 seconds, 60 °C for 30 seconds, and 72 °C for 30 seconds. The sequence of all primers is shown in Table 1. Relative gene expression level was normalized to β-actin and calculated using the 2^-ΔΔCt method. All experiments were performed in triplicate to ensure experimental reliability.

Table 1 Quantitative real-time polymerase chain reaction primer sequence list.

Gene	Primer	Sequence
Beclin1	Forward	CTGTGGAGTGGAATGAAA
	Reverse	TAGGGAACAAGTCGGTAC
P62	Forward	ATTTCCTGAAGAATGTGGG
	Reverse	CTGCTTGGCTGAGTGTTA
Slc7a11	Forward	TTGGAGCCCTGTCCTATGC
	Reverse	CGAGCAGTTCCACCCAGAC
GPX4	Forward	ATTCTCAGCCAAGGACAT
	Reverse	CAGGATTCGTAAACCACA

GPX4: Glutathione peroxidase 4.

Immunohistochemistry

Paraffin-embedded liver sections (5 μm) were deparaffinized and underwent antigen retrieval. After quenching endogenous peroxidase activity, sections were incubated (overnight, 4 °C) with anti-Nrf2 antibody (1:200), followed by incubation with horseradish peroxidase-conjugated secondary antibody. Diaminobenzidine was applied as the chromogen, and the sections were counterstained with hematoxylin. Images were captured utilizing a slide scanner and analyzed with ImageJ software.

Prussian blue staining

Mouse liver tissue was routinely dehydrated and cleared, then stained (1 hour) with Prussian blue working solution, followed by distilled water washing 3 times. Prussian blue dye was added and stained for 3 minutes. The tissue was then washed with tap water. Next, the tissues were dehydrated using anhydrous ethanol, cleared with xylene, and then sealed with neutral resin. The liver tissue was observed under an optical microscope to assess iron deposition[27].

Transmission electron microscopy

Fresh liver tissue from the experimental mice in each group was sectioned into 1 mm³ pieces, fixed with 2.5% glutaraldehyde solution, and stored (4 °C, overnight). The fixed tissue was then dehydrated, embedded in resin, and sectioned (70-80 nm), followed by staining (37 °C, 10 minutes) with 2% uranyl acetate and 2.6% lead citrate. A Hitachi HT7800 transmission electron microscopy (TEM) was employed for observation.

Statistical analysis

The experiments were conducted with three replicates to ensure the stability and reproducibility of the results. Data were shown as mean ± SD. SPSS 22.0 software was employed for data analysis, and GraphPad Prism 10.0 was utilized for graph generation. For normally distributed data, two independent samples t-test was applied; otherwise, the Mann-Whitney U-test was utilized. Statistical significance was set at P < 0.05, denoted as ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001.

RESULTS

QSHXO ameliorates hepatic steatosis in MASLD mice

The influence of QSHXO in MASLD was investigated in a mouse model, the research results are shown in Figure 2A and B. It was found that mice fed an MCD diet for 10 weeks exhibited significantly reduced body weight and liver weight compared to Ctrl mice. After QSHXO intervention for 8 weeks, mice in the QSHXO-H, QSHXO-M, QSHXO-L, and oltipraz groups all experienced weight gain relative to those in the MASLD group, with the QSHXO high-dose group and the oltipraz group exhibiting the most significant increase (P < 0.001).

Figure 2 Qushi Huoxue ointment attenuates methionine-choline-deficient diet-induced hepatic steatosis. A: Body weight of mice at the beginning and end of the experiment; B: Liver weight of mice at the beginning and end of the experiment; C: Representative hematoxylin and eosin staining images showing macrovesicular steatosis (yellow arrows), and focal inflammatory infiltration (blue arrows); scale bar: 100 μm and 20 μm; D: Representative Oil Red O staining images showing lipid droplets (red area /yellow arrows) and nuclei (blue area); scale bar: 100 μm and 20 μm; E: Quantitative analysis of Oil Red O-positive area (%); F: Serum total cholesterol levels; G: Serum triglyceride levels. Data are presented as means ± SD; ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment; TC: Total cholesterol; TG: Triglyceride.

Histopathological analysis results by H&E staining (Figure 2C) revealed that the MASLD group displayed severe macrovesicular steatosis, hepatocellular ballooning, and focal inflammatory infiltration compared to the Ctrl group. These pathological features were markedly attenuated in mice treated with QSHXO and oltipraz, with mice in the QSHXO-H group showing the most significant pathological improvement. ORO staining was conducted to further quantify hepatic steatosis (Figure 2D and E). As indicated by the results, mice in the MASLD group showed extensive red-stained LDs, whereas QSHXO and oltipraz treatments significantly reduced both the number and staining intensity of LDs, along with a lower percentage of ORO-positive area. In particular, the QSHXO-H group showed the fewest LDs.

Hepatic lipid metabolism was further assessed by measuring TC and TG levels (Figure 2F and G). The results indicated that both QSHXO and oltipraz treatments effectively counteracted the MCD diet-induced elevations in TC and TG levels, indicating a restoration of lipid homeostasis. Among these treatments, the QSHXO-H group demonstrated the most significant improvement (P < 0.01).

QSHXO ameliorates liver injury and inflammation in MASLD mice

Subsequently, liver injury and systemic inflammation were assessed via measuring AST, ALT, TNF-α, and IL-1β serum levels. It was found that the MASLD group exhibited significant elevations in these markers compared with the Ctrl group, indicating severe hepatocellular damage and inflammatory activation (Figure 3A). QSHXO and oltipraz interventions significantly reversed these changes, as evidenced by reduced AST and ALT contents and lowered TNF-α and IL-1β concentrations; these indicated alleviated liver damage and inhibited inflammation. The QSHXO-H group showed the most significantly decreased AST, ALT, TNF-α, and IL-1β levels (P < 0.001). These results supported a clear correlation between hepatic lipid accumulation and liver damage (Figure 3B). Our findings demonstrated that QSHXO not only ameliorated steatosis but also mitigated hepatic inflammation and injury, highlighting its potential as a therapeutic agent for MASLD.

Figure 3 Qushi Huoxue ointment ameliorates liver injury and inflammatory response in metabolic associated steatotic liver disease mice. A: Serum levels of aspartate aminotransferase, alanine aminotransferase, tumor necrosis factor-α, and interleukin-β; B: Schematic diagram illustrating the relationship between hepatic lipid accumulation and liver injury. Data are presented as means ± SD; ^aP < 0.05, ^cP < 0.001. MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; TNF-α: Tumor necrosis factor-α; IL-β: Interleukin-β; TC: Total cholesterol; TG: Triglyceride.

Identification of the components in QSHXO

Using LC-MS/MS, 563 chemical components in QSHXO were identified and 51 components were observed in animal blood after QSHXO treatment. After identifying the intersection of these two sets and excluding the 77 components detected in animal blank serum, the blood components of QSHXO were determined (Table 2), with a total of 19 components classified as blood-entering ingredients. Accumulating studies have also proved that these components have positive effects on MASLD[28-40].

Table 2 Qualitative results of blood ingredients in Qushi Huoxue ointment.

No.	Compound	RT/minute	Formula	Reference ion	m/z	Annotation MW	Calc. MW	Mass error (ppm)	Sort
1	L(-)-carnitine[24]	1.253	C7H15NO3	[M+H]+1	162.11258	161.10519	161.10532	0.00013	Amino acid derivatives
2	Emodin[25]	16.835	C15H10O5	[M+H]+1	271.06018	270.05282	270.0529	0.00007	Anthraquinones
3	Trigonelline[26]	1.402	C7H7NO2	[M+H]+1	138.05501	137.04768	137.04773	0.00005	Quaternary ammonium alkaloid
4	1-heptanesulfonic acid	21.624	C7H16O3S	[M-H]-1	179.07375	180.08201	180.08102	-0.00099	Alkylsulfonic acid
5	18-β-glycyrrhetinic acid[27]	21.286	C30H46O4	[M-H]-1	469.33200	470.33961	470.33928	-0.00033	Triterpenes
6	Cyclo (leucylprolyl)[28]	11.266	C11H18N2O2	[M+H]+1	211.14432	210.13683	210.13704	0.00021	Diketopiperazines
7	DL-homoserine[29]	1.307	C4H9NO3	[M+H]+1	120.06588	119.05824	119.0586	0.00036	Non-proteinogenic amino acids
8	2,3,4,9-tetrahydro-1H-β-carboline-3-carboxylic acid[30]	8.669	C12H12N2O2	[M+H]+1	217.09747	216.08988	216.0902	0.00032	β-carboline alkaloids
9	L-histidine[31]	1.183	C6H9N3O2	[M-H]-1	154.06104	155.06948	155.06831	-0.00116	Proteinogenic amino acids
10	cis-resveratrol[32]	11.887	C14H12O3	[M+H]+1	229.08614	228.07864	228.07886	0.00022	Stilbenes
11	Glycitein[33]	15.457	C16H12O5	[M+H]+1	285.07593	284.06847	284.0686	0.00012	Isoflavones
12	4-oxoproline[34]	2.715	C5H7NO3	[M-H]-1	128.03416	129.04259	129.04144	-0.00115	Amino acid
13	N5-(1,3,5-trimethyl-1H-pyrazol-4-yl)-1H-1,2,4-triazole-3,5-diamine	6.652	C8H13N7	[M+H]+1	208.13333	207.12324	207.12605	0.00281	Heterocyclic compounds
14	2-phenoxypropanoic acid	9.34	C9H10O3	[M-H]-1	165.05479	166.06299	166.06207	-0.00092	Carboxylic acids
15	Caffeic acid[35]	18.577	C9H8O4	[M+H]+1	149.02333	180.04226	148.01604	-32.02622	Phenolic acid
16	(3S,3aR,4S,4aR,7aR,8R,9aR)-3,4a,8-trimethyl-2,5-dioxo-2H,3H,9aH-azuleno[6,5-b]yl(2Z)-2-methylbut-2-enoate]	18.415	C20H26O5	[M+H]+1	369.17236	346.17802	368.16509	21.98706	Esters
17	RKK	20.151	C18H38N8O4	[M+H]+1	431.31555	430.3016	430.30821	0.00661	Amino acids
18	(3R,4S)-4,6,8-Trihydroxy-7-methoxy-3-methyl-3,4-dihydro-1H-isochromen-1-one	10.277	C11H12O6	[M-H]-1	239.05597	240.06339	240.06325	-0.00014	Isocoumarin derivatives
19	Apigetrin[36]	12.243	C21H20O10	[M-H]-1	431.09830	432.10565	432.10569	0.00004	Flavonoids

Network pharmacology predicts the mechanism of QSHXO intervention in MASLD

The potential targets of QSHXO against MASLD were predicted. Specifically, 185 QSHXO components and 1824 corresponding targets were obtained, of which 517 were MASLD-related targets. Additionally, 270 ferroptosis-related genes (including ferroptosis markers and driver genes) and 222 autophagy-related genes were identified. After intersecting potential targets of QSHXO with MASLD, ferroptosis, and autophagy functions, 47 TCM-disease-ferroptosis-related targets and 52 TCM-disease-autophagy-related targets were identified (Figure 4A). Following importing these obtained targets and genes into the STRING database, 83 targets with clear PPIs were retained. The PPI network was constructed and visualized utilizing Cytoscape 3.10.1, where green represented ferroptosis-related genes and yellow represented autophagy-related genes, with a larger node diameter indicating a higher importance of the target in the network (Figure 4B). Moreover, these 83 network targets were subjected to functional enrichment analyses, with the top 10 GO enrichment results shown in a bar chart and the top 20 enriched KEGG signaling pathways in a bubble chart. It was revealed that key signaling pathways related to the targets included autophagy-animal, fluid shear stress, atherosclerosis, lipids, and atherosclerosis (Figure 4C). Based on the literature review and the above PPI network analysis results, “NFE2 L2”, “SLC7A11”, “GPX4”, “MAP1 LC3B”, “SQSTM1”, and “BECN1” were selected as the key targets (Table 3). Furthermore, a relationship network was constructed among these six targets, the top 20 KEGG-enriched signaling pathways, and TCM components. The association between TCM components and key targets was visualized using a Sankey diagram, where NFE2 L2 showed the most significant association with TCM components (Figure 4D).

Figure 4 Network pharmacology predicts the mechaism of Qushi Huoxue ointment intervention in metabolic associated steatotic liver disease. A: Venn diagram of traditional Chinese medicine (TCM)-disease-ferroptosis intersection and TCM-disease-autophagy intersection; B: Protein-protein interaction network diagram; C: Gene Ontology enrichment analysis top 10 bar chart, Kyoto Encyclopedia of Genes and Genomes enrichment analysis top 20 bubble chart; D: TCM-component-key target-pathway network diagram; key target-component Sankey diagram. MASLD: Metabolic associated steatotic liver disease; TCM: Traditional Chinese medicine; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes.

Table 3 Key target network attribute table.

Name	Betweenness	Closeness	Degree
SLC7A11	0.066168	0.480315	6
GPX4	0.002891	0.472868	6
SQSTM1	0.049908	0.508333	12
MAP1 LC3B	0.058196	0.472868	9
BECN1	0.091432	0.512605	12
NFE2 L2	0.698875	0.72619	39

GPX4: Glutathione peroxidase 4.

QSHXO activates autophagy in hepatocytes of MASLD model mice

Next, the protein level of core autophagy markers (Beclin1, LC3, and P62) was assessed by western blot to assess autophagic activity. As shown in Figure 5A and B, hepatic Beclin1 protein level was notably reduced in the MASLD group and the oltipraz group relative to that in the Ctrl group. Intervention with QSHXO counteracted this reduction in a dose-dependent fashion, and QSHXO-H caused the most significant effect (P < 0.001). Furthermore, the LC3II/LC3I ratio, a key indicator of autophagosome (AS) formation, was markedly increased in both QSHXO groups and the oltipraz group, suggesting enhanced autophagic initiation. Meanwhile, substantial accumulation of P62 protein was observed in MASLD liver tissues, indicating impaired autophagic flux and lysosomal degradation[41]. Importantly, QSHXO administration markedly reduced P62 protein levels, supporting the restoration of autophagic-lysosomal clearance function.

Figure 5 Qushi Huoxue ointment activates hepatic autophagy in metabolic associated steatotic liver disease mice. A: Representative western blot images of Beclin1, the light chain 3 II/Light chain 3 I, and P62 protein levels in liver tissues; B: Quantitative analysis of Beclin1, the light chain 3 II/Light chain 3 I ratio, and P62 protein levels; C: Quantitative reverse-transcription polymerase chain reaction analysis of mRNA expressions of Beclin1 and P62 in liver tissues. Data are presented as means ± SD; ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. LC3II/LC3I: Light chain 3 II/Light chain 3 I ratio; MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment.

Whether these changes were reflected at the transcriptional level was then examined using qRT-PCR (Figure 5C). In line with the protein data, Beclin1 mRNA expression was downregulated in the MASLD group but showed dose-dependent upregulation following QSHXO or oltipraz treatments. Conversely, P62 transcript levels were elevated in MASLD mice, and QSHXO intervention demonstrated a dose-dependent regulatory effect, with QSHXO-H treatment causing the most marked performance (P < 0.001). Collectively, these multi-level results demonstrated that QSHXO promoted autophagic activation via enhancing both AS formation and lysosomal degradation, thereby restoring functional autophagic flux in MASLD.

QSHXO can inhibit ferroptosis in hepatocytes of MASLD model mice

According to immunohistochemistry (IHC) analysis results, there were notable differences in Nrf2 expression and localization among groups (Figure 6A and B). The Ctrl group showed moderate cytoplasmic staining, while the MASLD group exhibited only a mild increase. In contrast, both QSHXO and oltipraz treatment groups showed a markedly enhanced Nrf2 expression and prominent nuclear translocation, indicating effective activation of the Nrf2 pathway. Specifically, Nrf2 expression in the QSHXO-H group was most remarkably enhanced.

Figure 6 Qushi Huoxue ointment inhibits ferroptosis in metabolic associated steatotic liver disease mice. A: Representative immunohistochemical images of nuclear factor erythroid 2-related factor 2 (Nrf2) nuclear translocation in liver tissues. Scale bar: 100 μm and 20 μm; B: Quantitative analysis of Nrf2 expression by mean optical density; C: Western blot analysis of Nrf2 (using lamin B1 as internal reference), SLC7A11, and glutathione peroxidase 4 (GPX4) (using β-actin as internal reference) protein levels in liver tissues; D: Quantitative analysis of Nrf2, SLC7A11 and GPX4 protein levels; E: Quantitative reverse-transcription polymerase chain reaction analysis of mRNA expressions of SLC7A11 and GPX4 in liver tissues. Data are presented as means ± SD; ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment; Nrf2: Nuclear factor erythroid 2-related factor 2; GPX4: Glutathione peroxidase 4.

Next, key ferroptosis-related mediators (Nrf2, SLC7A11, and GPX4) were assessed by western blot to further explore the functional consequences of Nrf2 activation (Figure 6C and D). Consistent with the IHC results, Nrf2 protein level was slightly elevated in the MASLD group, reflecting a stress response. After QSHXO and oltipraz interventions, Nrf2 protein level was significantly upregulated. This effect manifested in a dose-dependent manner in QSHXO treatment groups, with the QSHXO-H group demonstrating the most remarkable results (P < 0.001). Notably, Nrf2 activation was accompanied by a marked increase in the protein level of its downstream targets (SLC7A11 and GPX4), suggesting a coordinated antioxidant response[42].

Next, these findings were validated at the transcriptional level using qPCR (Figure 6E). The results indicated that SLC7A11 mRNA expression was noticeably reduced in the MASLD group, but was strongly increased in QSHXO treatment groups in a dose-dependent manner. A similar regulatory pattern was observed for GPX4. This further confirmed that QSHXO enhanced the transcription of key genes involved in ferroptosis suppression by activating the Nrf2 pathway.

In summary, combined validation using IHC, western blot, and qPCR revealed that QSHXO activated the Nrf2 signaling pathway, significantly increasing the transcription level and expression of SLC7A11 and GPX4. These multi-level pieces of evidence collectively suggest that the hepatoprotective effect of QSHXO in the MASLD mouse model is tightly related to its inhibition of ferroptosis.

QSHXO alleviates iron deposition in the liver tissue of MASLD mice

It has been shown that a central characteristic of ferroptosis is dysregulated lipid metabolism, which leads to iron overload and lipid peroxidation, ultimately resulting in cell death[43]. Consequently, controlling intracellular iron homeostasis and reducing iron deposition have emerged as crucial strategies to counteract ferroptosis. Hence, Prussian blue staining was conducted to visualize tissue iron deposition to evaluate whether QSHXO can mitigate aberrant hepatic iron accumulation in MASLD.

In the stained sections, nuclei and background structures appeared red, whereas iron deposits were marked in blue. According to the results, relative to blank mice, liver tissues from MASLD mice exhibited intensified blue staining, indicating pronounced iron accumulation. Treatment with QSHXO effectively attenuated iron deposition, and high-dose QSHXO caused the most significant effect (Figure 7). Interestingly, the oltipraz group showed a stronger Prussian blue signal, suggesting even greater iron retention in the liver. It was hypothesized that this phenomenon may reflect a protective adaptation. Namely, by sequestering redox-active iron into a stable and stored form, cells may reduce the labile iron pool available for driving lipid peroxidation, thereby suppressing ferroptosis.

Figure 7 Qushi Huoxue ointment reduces hepatic iron deposition in metabolic associated steatotic liver disease mice. Representative images of Prussian blue staining in liver tissues. Nuclei are counterstained red, and iron deposits appear as blue granules in the hepatocyte cytoplasm (indicated by black arrows). MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment.

QSHXO improves the mitochondrial ultrastructure in MASLD mice

Both the activation of autophagic flux and the occurrence of ferroptosis induce significant morphological changes at the subcellular level. In the present study, the ultrastructure of hepatocytes was observed using a TEM to obtain direct morphological evidence of the molecular pathway changes. As observed, TEM revealed distinct pathological features across experimental groups (Figure 8). In Ctrl hepatocytes, there was normal cellular architecture, characterized by regular morphology, round nuclei, structurally intact mitochondria with well-defined cristae, and occasional autophagic vesicles, reflecting physiological autophagy. By contrast, MASLD mice exhibited marked ultrastructural alterations, manifested as mitochondrial shrinkage with distorted or absent cristae, significantly increased LD accumulation, and a conspicuous reduction in autolysosomes (ASS); all these indicated impaired autophagic clearance. Notably, treatment with either QSHXO-H or oltipraz markedly ameliorated these abnormalities. Both groups showed enhanced autophagic flux and degradative capacity, evidenced by restored hepatocyte morphology, reduced LD content, and a substantial increase in AS and ASS. Although mild mitochondrial changes persisted, the overall integrity was largely preserved.

Figure 8 Qushi Huoxue ointment ameliorates hepatic ultrastructural damage in metabolic associated steatotic liver disease mice. Representative transmission electron microscopy micrographs of hepatocytes. Compared to the control group, the metabolic associated steatotic liver disease group exhibited mitochondrial shrinkage with disrupted cristae, substantial lipid accumulation, and scarce autolysosome. In contrast, both the Qushi Huoxue ointment-high-dose and oltipraz groups showed preserved mitochondrial architecture, reduced lipid droplets, and increased autophagic vesicles, indicating restored autophagic flux. Scale bars: 2 μm (overview), 500 nm (detail). MASLD: Metabolic associated steatotic liver disease; QSHXO: Qushi Huoxue ointment; M: Mitochondria; LDs: Lipid droplets; ER: Endoplasmic reticulum; AS: Autophagosome; ASS: Autolysosome.

The above ultrastructural observations corroborated our molecular biology findings, collectively revealing the protective mechanisms of QSHXO. In the MASLD group, mitochondrial atrophy and elevated membrane density were observed, indicating ferroptosis at the morphological level. Consistently, elevated Nrf2 levels and downregulated SLC7A11 and GPX4 proteins were detected in this group. Meanwhile, in the QSHXO treatment group, improved mitochondrial morphology and AS structures confirmed that QSHXO successfully activated autophagic flux, thereby eliminating damaged mitochondria and blocking their progression toward ferroptosis. This finding is perfectly aligned with the observed molecular events, such as the increased Beclin1 and LC3II/LC3I ratio and the decreased P62 protein level. Collectively, these findings offer morphological evidence that QSHXO mitigated MASLD-associated hepatic injury via attenuating lipid accumulation, restoring mitochondrial integrity, enhancing functional autophagic flux, and inhibiting ferroptosis.

DISCUSSION

MASLD remains a challenging metabolic disorder with limited therapeutic efficacy from current pharmacological interventions. This research investigated the therapeutic potential and mechanism of the TCM formulation QSHXO in MASLD treatment. We first confirmed that QSHXO significantly improved liver steatosis and liver injury in the MCD diet-induced MASLD model mice. Subsequently, further analysis integrating network pharmacology prediction with systematic experimental validation confirmed that QSHXO exerted its protective effects by dual regulation: Activating the autophagy process and inhibiting ferroptosis. Our findings may provide a new perspective on the action mechanism of this drug.

In this study, QSHXO significantly upregulated Beclin1 expression, increased the LC3II/LC3I ratio, and reduced P62 protein level, indicating that QSHXO successfully activated and restored the impaired autophagic flux in MASLD. As has been evidenced previously, Beclin1 is a key regulator of autophagy initiation[44,45], while the conversion of LC3I to LC3II is a hallmark of AS formation[46,47]. As a selective autophagy substrate, P62 accumulation typically reflects blocked autophagic flux[48,49]. Therefore, our results strongly suggested that restoring functional autophagy to clear damaged organelles and LDs was an important mechanism by which QSHXO alleviated hepatic steatosis. Consistently, a previous study has revealed that impaired autophagy exacerbates hepatic lipid accumulation[50]. Nevertheless, this study is the first to reveal the specific targets of QSHXO in this process.

Furthermore, our research demonstrated that QSHXO counteracted MASLD by inhibiting ferroptosis. Specifically, it was observed that QSHXO treatment promoted Nrf2 nuclear translocation and significantly upregulated its downstream target genes (SLC7A11 and GPX4). Consistently, previous studies have provided evidence that Nrf2 is a central regulator of the cellular antioxidant response[18,51]. SLC7A11 and GPX4 are key proteins executing anti-ferroptosis functions[52,53]. These results demonstrated that QSHXO may enhance the overall cellular antioxidant capacity to resist lipid peroxidation damage. Furthermore, reduced hepatic iron deposition was observed in this study, further confirming the suppression of ferroptosis. This is consistent with the established theory that Nrf2 activation can coordinate a broad cellular defense network[54,55]. This highlighted that QSHXO may be a promising multi-target therapeutic strategy, with efficacy surpassing that of single-target inhibitors.

Interestingly, we noted that the positive Ctrl oltipraz, despite showing similar improvement in Nrf2 activation, exhibited stronger Prussian blue staining signals than the MASLD group. These unexpected findings may reflect a protective “iron sequestration” mechanism where oltipraz promoted iron storage in inert forms, paradoxically demonstrating stronger staining while reducing biologically active iron. This observation challenges the prevailing paradigm in Prussian blue staining interpretation, suggesting that iron deposition patterns, rather than its quantity alone, may hold greater significance for assessing ferroptosis activity.

Our findings revealed that QSHXO concomitantly activated autophagy and suppressed ferroptosis, suggesting a potential cooperative mechanism that contributed to the amelioration of MASLD. Mechanistically, QSHXO promoted the autophagic clearance of damaged organelles while simultaneously enhancing cellular antioxidant defenses via Nrf2-mediated gene expression. This dual approach effectively disrupted the vicious cycle of oxidative stress and lipid peroxidation driving MASLD progression. The preserved mitochondrial ultrastructure and increased ASS formation observed in TEM further provided structural evidence for this functional restoration.

Nevertheless, there are still several limitations that should be acknowledged. First, the MCD diet model, we primarily used rapidly induces steatohepatitis but lacks key metabolic features such as obesity and insulin resistance, which represents a significant contextual limitation. This may affect the direct translatability of our findings to human MASLD. Therefore, validating the key results in diet-induced obesity models, such as those using high-fat or high-fat/high-fructose diets, will be crucial in future studies to more reliably assess QSHXO’s efficacy in a clinically relevant and metabolically complex context. Second, the precise causal relationship between autophagy activation and ferroptosis inhibition requires further investigation using pathway-specific inhibitors or genetic approaches.

Additionally, this study primarily investigated the protective effects of QSHXO in MASLD through mediating the regulation of autophagy and ferroptosis pathways. However, the pathogenesis of MASLD involves a complex signaling network. Although we did not systematically validate the classical pathways such as AMP-activated protein kinase (AMPK), sterol regulatory element-binding protein-1c, phosphatidylinositol 3-kinase/protein kinase B, and nuclear factor kappaB (NF-κB) in our experiments, primarily to concentrate our resources and experimental depth in establishing direct evidence of QSHXO’s effects on autophagy and ferroptosis, we acknowledge the critical roles these pathways play in lipid metabolism and the regulation of inflammation[56-60], and they are likely to have functional intersections with our findings. Based on the existing literature, we make the following focused speculations: First, AMPK, as a key sensor of cellular energy and metabolism, is known to not only directly induce autophagy but also to influence cellular sensitivity to ferroptosis by regulating lipid metabolism and oxidative stress levels[61]. Therefore, the autophagy activation and ferroptosis inhibition induced by QSHXO may share a common upstream regulator the AMPK pathway. Secondly, NF-κB is a core transcription factor of pro-inflammatory responses, while Nrf2 is the main regulator of antioxidant defense, and the two usually act in mutual antagonism[62,63]. Our study found that QSHXO alleviated liver inflammation, which may potentially involve the regulation of the Nrf2-NF-κB balance axis. This could occur through the activation of Nrf2 to enhance cellular antioxidant capacity while simultaneously inhibiting the excessive activation of NF-κB, thereby synergistically alleviating inflammatory stress and providing an upstream explanation for the observed inhibition of ferroptosis. Therefore, future research should thoroughly investigate whether QSHXO improves hepatic lipid metabolism homeostasis by regulating the AMPK/sterol regulatory element-binding protein axis, and whether it alleviates inflammation and oxidative damage by modulating the Nrf2-NF-κB dialogue axis. This would not only place the mechanisms identified in this study within a broader regulatory network but also provide valuable insights into the systems biology underlying the multi-target and integrative regulatory effects of this herbal formulation.

All in all, we hypothesized that QSHXO may initially activate autophagy, which subsequently modulates ferroptosis sensitivity through degradation of key regulators. Future research should employ temporal studies to delineate this sequence of events. Additionally, techniques such as spatial transcriptomics or single-cell RNA sequencing could reveal the heterogeneity of QSHXO’s effects across different liver cell populations, which may help lay a solid foundation to understand the therapeutic mechanism. In summary, our study demonstrated that QSHXO was a multi-target strategy for MASLD with its capacity of concurrently activating autophagy and inhibiting ferroptosis. These findings not only deepen our insights into TCM mechanisms in metabolic diseases but also pave the way for developing innovative multi-pathway interventions for MASLD.

CONCLUSION

Our study demonstrated that QSHXO effectively ameliorated hepatic steatosis and injury in MASLD mice. We systematically validated through a multi-method approach that its protective efficacy is mediated through a coordinated regulation of autophagic flux activation and ferroptosis suppression. Specifically, QSHXO promoted the initiation and degradation stages of autophagy, while concurrently activating the Nrf2-mediated antioxidant pathway to inhibit ferroptosis. These findings not only elucidate the pharmacological basis of QSHXO but also provide a promising multi-target therapeutic strategy for MASLD treatment.