Saffron and Calculus bovis combination exerts anti-hepatic fibrotic effect in liver fibrosis rats via the mitogen-activated protein kinases pathway

Abstract

BACKGROUND

Liver fibrosis is a global health issue that lacks effective treatments. Tibetan medicine, with a long history, has accumulated rich experience in the treatment of chronic liver diseases. The saffron (Saf) and Calculus bovis (Cal b) combination is among the most commonly used medicines in clinical practice in Tibetan medicine for hepatic disease. Its characteristic therapies and drug compatibility provide unique ideas for the treatment of liver fibrosis and have research value and application potential.

AIM

To investigate the efficacy of the Saf-Cal b therapy in treating liver fibrosis and explored its underlying mechanism.

METHODS

We initially established a carbon tetrachloride-induced rat liver fibrosis model to assess Saf-Cal b’s anti-fibrotic effects. Subsequently, we conducted network pharmacology analysis to identify the potential therapeutic targets and pathways of Saf-Cal b in liver fibrosis intervention. Finally, we performed in vivo validation of key regulatory targets.

RESULTS

Saf-Cal b combination therapy exerted superior effects in ameliorating liver fibrosis in model rats compared with Saf or Cal b monotherapy. Through network pharmacology prediction, key targets of the combination were identified. Mechanistic validation revealed that Saf-Cal b inhibited the p38 mitogen-activated protein kinases pathway, which in turn suppressed the transforming growth factor-β/small mother against decapentaplegic pathway. This sequential inhibition led to reduced activation of hepatic stellate cells, a central event in liver fibrosis progression.

CONCLUSION

These findings demonstrate that Saf-Cal b combination therapy is more effective than either monotherapy in alleviating liver fibrosis, with its therapeutic effect mediated through the p38 mitogen-activated protein kinases/transforming growth factor-β/small mother against decapentaplegic signaling axis, providing a potential therapeutic strategy for liver fibrosis.

Key Words: Liver fibrosis; Tibetan medicine; Saffron; Calculus bovis; Mitogen-activated protein kinases pathway

Core Tip: Liver fibrosis, a health issue of global concern, lacks effective treatments. Using Tibetan medicine experience and network pharmacology, this study explored the potential mechanism of the saffron-Calculus bovis combination against liver fibrosis in rats. After CCl₄ induction of rats, intragastric intervention was performed, followed by assessments of liver pathology, liver function, inflammation, and hepatic stellate cell activation. The results showed that the combination exerted more significant anti-fibrotic effects compared with single treatment, and it may act by regulating the mitogen-activated protein kinase pathway.

INTRODUCTION

Liver disease causes 2 million deaths worldwide annually, with mortality primarily attributed to end-stage liver fibrosis[1]. Liver fibrosis is a healing reaction that is induced in the liver in response to various harmful stimuli. It is characterized by the excessive accumulation of extracellular matrix (ECM) components and the formation of scar tissue, which ultimately leads to a progressive distortion of tissue architecture and functional impairment of the liver. The development of liver fibrosis is a dynamic process that entails ongoing liver injury, inflammatory responses, and an imbalance in the processes of ECM deposition and degradation[2]. Liver fibrosis is an intermediate stage and common result of chronic liver diseases. If not treated in time, it may progress to cirrhosis, liver failure, or even hepatocellular carcinoma, thereby posing a severe threat to life and health. While significant advancements have been made in liver fibrosis research in recent years, specific and effective drugs for clinical treatment are lacking[3]. Consequently, the development of effective anti-fibrotic therapeutic agents is a critical research priority and remains the forefront of research.

Tibetan medicine, with a millennia-old historical legacy, has evolved into a comprehensive medical system characterized by a systematic, well-established, and distinctive theoretical foundation. As a result of the special geographical environment and dietary habits of the Tibetan people, as well as long-term clinical practices, Tibetan medicine has developed unique insights into liver diseases[4]. In current clinical practice of Tibetan medicine, there are 22 compound formulations used for the treatment of chronic liver diseases. Among these, the formulations containing Saffron (Saf) and Cal b include Shisanwei Shugan Capsules, Ershiwuwei Songshi Pills, and Bawei Honghua Qingganre Powder, etc. Saf is the most prevalent medicinal component and incorporated in 18 formulations (81.8%). Calculus bovis (Cal b) is present in 13 preparations (59.1%)[5]. The Saf-Cal b combination is the predominant therapeutic pairing for chronic hepatic disorders in Tibetan medicinal interventions[6].

Crocus sativus Linnaeus, commonly referred to as Saf or Tibetan Saf, is a perennial herbaceous species in the Iridaceae family, genus Crocus. This commercially valuable spice crop exhibits a broad geographical distribution and is predominantly cultivated in temperate regions spanning Europe, the Mediterranean basin, and Central Asia. In the 2020 edition of the Chinese Pharmacopoeia, Saf is pharmacologically characterized by its sweet flavor profile and neutral thermal property, with specific tropism toward the heart and liver meridians[7]. This medicinal agent exhibits multiple therapeutic properties, including blood-activating and stasis-resolving effects, blood-cooling and detoxification actions, and mood-regulating and sedative functions. Clinically, Saf is indicated for various pathological conditions such as menstrual disorders, postpartum complications, heat-toxin-induced dermatological manifestations, depressive disorders, and psychosomatic symptoms including anxiety and agitation[8].

Cal b is the dried gallstone from the cow Bos taurus domesticus Gmelin. The 2020 edition of the Chinese Pharmacopoeia reports that Cal b has distinct bitter medicinal properties and a cooling nature, with specific meridian tropism targeting the cardiac and hepatic systems[7]. This traditional medicinal substance exhibits multi-modal therapeutic actions, including cardiac heat-clearing and phlegm-resolving properties, orifice-opening and hepatic-cooling effects, and wind-extinguishing and detoxification capabilities. Clinical applications primarily include neurological disorders (e.g., febrile delirium and cerebrovascular events with phlegm obstruction), convulsive disorders (including epilepsy and seizure-related conditions), psychiatric manifestations (such as manic episodes), and inflammatory conditions (including pharyngolaryngeal inflammation, oral ulcerations, and suppurative dermatological lesions)[9].

Hepatic stellate cell (HSC) activation is a crucial driving force underlying fibrogenic progression in chronic liver diseases[10,11]. Following HSC activation, α-smooth muscle actin (α-SMA) expression is significantly upregulated, establishing α-SMA as a reliable molecular marker for HSC activation. Some mechanistic studies have demonstrated that the p38 mitogen-activated protein kinase (MAPK) signaling pathway plays a crucial regulatory role in modulating α-SMA protein expression in activated HSCs, while simultaneously influencing cell cycle progression, thereby contributing to the pathogenesis and progression of hepatic fibrosis[12]. Qiu et al[9] found that activation of p38/MAPK promotes liver cell proliferation and chronic liver inflammation[13]. p38 MAPKs not only participate in cell proliferation and differentiation, but also in cell apoptosis. p38 MAPK influences the development of liver fibrosis through multiple pathways, including regulating the inflammatory response of the liver, inducing cell apoptosis, and regulating cytokine release[14]. To explore the potential mechanism of the Saf-Cal b in intervening hepatic fibrosis, we employed network pharmacology. Based on the “disease-gene-drug-target” interaction network, network pharmacology analyzes drug mechanisms at the systemic level. When predicting targets and pathways, it integrates multi-omics data to rapidly screen key targets, enrich related pathways, reduce experimental blindness, reveal the synergistic effects of drugs on multiple targets, and improve research efficiency[11]. This study explored the potential mechanism of the anti-hepatic fibrotic effect of the Saf-Cal b combination. The ultimate goal is to enhance the scientific understanding of Tibetan medicinal interventions for hepatic fibrosis and provide new perspectives for ethnopharmacological research.

MATERIALS AND METHODS

Drugs and reagents

The following drugs and reagents were used in this study: Saf (Jiuyuanci Traditional Chinese Medicine Co., Ltd., Bozhou City, China), Cal b (Tongrentang, Beijing, China), Biejia Ruangan tablet (BJRG; Furui Medical Technology Co., Ltd., Inner Mongolia, China), carbon tetrachloride (CCl₄, Macklin, Shanghai, China), and olive oil (Macklin, Shanghai, China).

The solutions were prepared as follows. CCl₄ solution: A mixture of CCl₄ and olive oil was prepared at a 1:1 ratio. Saf solution: Five grams of Saf was added to deionized water; the sample was soaked, decocted twice in distilled water, and concentrated to 500 mL to obtain Saf solution (0.01 g/mL). The solution was stored in a refrigerator at 4 °C until use. Cal b solution: Cal b powder (0.35 g) was added to 100 mL of distilled water, sonicated, and mixed for 20 minutes to obtain the final solution (0.0035 g/mL). The solution was stored in a refrigerator at 4 °C until use. Saf-Cal b solution: Cal b powder (0.35 g) was added to 100 mL of 0.01 g/mL Saf solution. The mixture was ultrasonicated for 20 minutes. The solution was stored in a refrigerator at 4 °C until use. Referring to the drug ratio in classic Tibetan medicines for treating liver fibrosis (e.g., Ershiwuwei Songshi Wan), we set the ratio of Saf to Cal b at 1:1[7]. BJRG solution: BJRG (0.6 g) was added to 10 mL of 0.9% saline solution. The mixture was sonicated for 20 minutes and homogenized, yielding a final solution of 0.06 g/mL. The solution was stored in a refrigerator at 4 °C until use.

Experimental animal procedures

Sprague Dawley rats (male, 200 ± 20 g, n = 145) were purchased from SPF (Beijing, China) Biotechnology Co., Ltd. (Beijing, China), certification number: SCXK [jing] 2019-0010. The rats were housed in the Animal Experiment Center of Liangxiang Campus of Beijing University of Chinese Medicine, under specific pathogen-free conditions. The certificate number was BUCM-2023041904-2025. The animal protocol was designed to minimize pain or discomfort. Prior to experimentation, rats were acclimatized to laboratory conditions for one week, with controlled environment parameters: Temperature 23 ± 2 °C, 12 hours light/12 hours dark cycle, humidity 50% ± 10%, and ad libitum access to standard chow and tap water.

The rats were acclimatized for one week. On the second week, the rats were randomly allocated into the following groups: Control rats, CCl₄ rats, BJRG rats, Saf rats, Cal b rats, and Saf-Cal b rats (10 rats/group). CCl₄ (2 mL/kg, CCl₄:olive oil = 1:1) was administered by gavage twice a week for 4 or 8 weeks to induce liver fibrosis in all groups except the control group[15]. Intragastric gavage was performed using straight gavage needles appropriate for the rat size (200 ± 20 g body weight: 16-gauge, 2-inch length, 2.0 mm ball diameter) with conscious animals to avoid unnecessary stress. Saf (0.1 g/kg), Cal b (0.035 g/kg), Saf 0.1 g/kg + Cal b 0.035 g/kg, and BJRG (0.6 g/kg) were administered by gavage after 1 week. BJRG was used as a positive control. Because a dose of Saf and Cal b has not been established, the dose was calculated following the recommended dose in the 2020 edition of the Chinese Pharmacopoeia[7]. Control and CCl₄ rats received an identical volume of 0.9% saline (Figure 1A). At the end of weeks 4 and 8, rats were euthanized by intravenous injection of pentobarbital sodium (150 mg/kg) for tissue collection.

Figure 1 The saffron and Calculus bovis treatment inhibited liver fibrosis in CCl₄-induced liver fibrosis rats at 4 weeks and 8 weeks. A: An overview of the experimental protocol is shown. Six-week-old rats received CCl₄ gavage for 4 weeks and 8 weeks, with concurrent drug administration initiated on the second week of gavage; B: Hematoxylin and eosin staining (scale bar: 100 μm); C: Masson’s staining (scale bar: 100 μm) of liver tissue from the indicated groups; D: Transmission electron microscopy was performed to examine the ultrastructure of liver tissue (scale bar: 500 nm). Note: In this study, “4w” and “8w” are defined as the 4^th and 8^th week, respectively, after the initial CCl₄ treatment. BJRG: Biejia Ruangan tablet; Saf: Saffron; Cal b: Calculus bovis; HE: Hematoxylin and eosin; TEM: Transmission electron microscopy.

Histopathology and transmission electron microscopy

To evaluate liver histopathological changes (steatosis, lobular inflammation, hepatocellular ballooning), hematoxylin-eosin staining was used. Collagen accumulation, a key feature of liver fibrosis, was detected via Masson’s trichrome staining. Procedures followed previous methods with minor modifications[16]. Briefly, 4 μm paraffin-embedded liver sections were deparaffinized in xylene and rehydrated through graded ethanol. For hematoxylin-eosin staining, sections were incubated in hematoxylin (Sigma-Aldrich, St. Louis, MO, United States) for 5 minutes, differentiated in 1% hydrochloric acid-ethanol for 30 seconds, and counterstained with eosin (Sigma-Aldrich, St. Louis, MO, United States) for 3 minutes. Masson staining used Weigert’s iron hematoxylin (Sigma-Aldrich, St. Louis, MO, United States) for 10 minutes, Masson’s trichrome solution (Sigma-Aldrich, St. Louis, MO, United States) for 30 minutes, and 1% acetic acid differentiation for 1 minutes. Stained sections were dehydrated, cleared, mounted with neutral balsam, and imaged via Olympus BX53 (Tokyo, Japan) microscope.

Transmission electron microscopy (TEM) observed hepatocyte ultrastructures (mitochondrial morphology, endoplasmic reticulum integrity). 1 mm³ liver tissues were fixed in 4% glutaraldehyde (LABLEAD, Beijing, China) in 0.1 mol/L phosphate buffered saline (PBS) (pH = 7.4) at 4 °C for 24 hours. After 3 PBS washes, tissues were post-fixed with 1% osmium tetroxide (SPI, West Chester, PA, United States) at room temperature for 2 hours, acetone-dehydrated, embedded in epoxy resin 812 (SPI, West Chester, PA, United States), and polymerized at 60 °C for 48 hours. 60 nm sections (Leica EM UC7, Wetzlar, Germany) stained with 2% uranyl acetate (SPI, West Chester, PA, United States) and lead citrate (SPI, West Chester, PA, United States) were examined under Hitachi Corp (Tokyo, Japan) TEM.

Assessment of liver injury parameters

At the time of sampling, the wet weight of the liver and spleen, as well as the body weight of the rats, were measured. The liver and spleen indices were calculated using the formula: Organ index = organ wet weight (g)/body weight (g) × 100%. Serum was separated from whole blood by centrifugation. The serum levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and total bilirubin (TBIL) were determined using a fully automated biochemical analyzer. The hyaluronic acid (HA) content was measured with Mindray reagents using a fully automated chemiluminescent immunoassay analyzer. Liver hydroxyproline (HYP) levels were determined using the HYP assay kit (Nanjing Institute of Bioengineering, China), following the manufacturer’s instructions.

The levels of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), matrix metalloproteinase 2 (MMP-2) and tissue inhibitors of metalloproteinases-1 (TIMP-1) in liver were measured with ELISA kits (Thermo Fisher, MA, United States; RayBiotech, GA, United States) following the manufacturers’ protocols. Glutathione peroxidase (GSH-PX) was evaluated using the GSH-PX assay kit (Nanjing Institute of Bioengineering, China). Malondialdehyde (MDA) levels were determined with the Lipid Peroxidation MDA Assay Kit (Beyotime, China). Liver superoxide dismutase (SOD) levels were determined with the SOD Assay Kit (Beyotime, China).

Network pharmacology: Collection of potential targets of Saf and Cal b in anti-liver fibrosis

The chemical components of Saf and Cal b were retrieved from Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) (http://tcmspw.com/tcmsp.php) and Bioinformatics Annotation Database for Molecular Mechanism of Traditional Chinese Medicine (BATMAN-TCM) (http://bionet.ncpsb.org/batman-tcm). From TCMSP, active components with oral bioavailability ≥ 30% and drug-likeness ≥ 0.18 (per absorption, distribution, metabolism, and excretion properties) and their targets were selected; from BATMAN-TCM, targets meeting a score cutoff ≥ 20 and adjusted P value < 0.05 were included, with literature supplementing missing components and targets.

Target proteins were standardized via UniProt (https://www.uniprot.org/). Duplicates were removed to establish a Saf-Cal b component-target database and imported into Cytoscape 3.8.2 to construct a drug-component-target network. Liver fibrosis-related genes were retrieved from GeneCards (https://www.genecards.org/) and OMIM (http://www.omim.org/) using “liver fibrosis”, “hepatic fibrosis”, and ”cirrhosis”. After UniProt standardization (Homo sapiens, excluding non-human targets), overlapping targets between Saf-Cal b and liver fibrosis were identified via Venn diagram analysis (https://bioinfogp.cnb.csic.es/tools/venny/index.html).

Analysis of targets of Saf and Cal b in liver fibrosis

A protein-protein interactions (PPI) network diagram was generated for the overlapping targets identified above using the STRING database (https://string-db.org/cgi/). Cluster analysis was performed using the Cytoscape 3.8.2 software plug-in molecular complex detection (MCODE); the best protein module and hub gene was selected by the score of the node[17].

Gene function annotation and Kyoto Encyclopedia of Genes and Genomes pathway analyses

The Database for Annotation, Visualization and Integrated Discovery (http://david.abcc.ncifcrf.gov/) was used for visualization and analysis of Saf-Cal b target proteins. The target genes were pasted into the gene list section of Database for Annotation, Visualization and Integrated Discovery and labeled as “official gene symbol”. The species was set as “Homo sapiens” with a threshold of P < 0.05. The results obtained from Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were analyzed.

Immunofluorescence

The preparation of paraffin sections from liver tissue was performed as for histopathology. For antigen retrieval, the sections were boiled in 6 mmol/L citrate buffer (pH = 6.0) for 15 minutes and then cooled at room temperature for 20 minutes. To reduce non-specific binding, the sections were blocked with PBS containing 5% bovine serum albumin for 1 hour at room temperature. Immunofluorescence analysis was performed following previously described methods[16]. Antibodies against p38 MAPK (1:200, Cell Signaling Technology, MA, United States) and α-SMA (1:200, ProteinTech, IL, United States) were used as primary antibodies. α-SMA was used to label HSCs. For secondary antibody staining, immunoglobulin G-horseradish peroxidase (1:250; ZSGB-BIO, Beijing, China) was used, and the nucleus was stained with DAPI. The cells were observed and analyzed using a fluorescent inverted microscope.

Western blot

Protein extracts were prepared from rat liver tissue using radio-immunoprecipitation assay buffer containing 1% phenylmethylsulfonyl fluoride and 1% phosphatase and protease inhibitors. Western blot analysis was performed using standard methods. The primary antibodies included p-p38 (1:1000), p38 (1:1000), phosphorylated c-Jun N-terminal kinases (JNK) (1:1000), JNK (1:1000), phosphorylated extracellular signal-regulated kinase (ERK) (1:2000), ERK (1:1000), and type I collagen (Col-I, 1:1000) (all from Cell Signaling Technology, MA, United States). Glyceraldehyde-3-phosphate dehydrogenase (1:5000; LABLEAD, China) was used as a loading control.

Real time polymerase chain reaction

Total RNA was extracted from liver tissue using the FastPure whole cell/tissue isolation kit (Accurate Biotech, Changsha, China). RNA was reverse transcribed into complementary DNA (cDNA) using a cDNA synthesis kit. cDNA was subjected to real-time polymerase chain reaction analysis on a Bio Rad CFX-96 system using a reverse transcription kit (Accurate Biotech, Changsha, China). The rat-specific polymerase chain reaction primers were purchased from Hunan Aikeli Bioengineering Company, China (Table 1).

Table 1 Polymerase chain reaction primers.

Target gene	Forward sequence (5′-3′)	Reverse sequence (5′-3′)
P38 MAPK	GGATATTTGGTCCGTGGGCT	TCATGGCTTGGCATCCTGTTA
ERK	CATGAAGGCCCGAAACTACCT	AGCCACTGGTTCATCTGTCG
JNK	CTCTCCAGCACCCGTACATC	TTGACAGACGGCGAAGAGAC
GAPDH	GGCAAGTTCAACGGCACAG	GCCAGTAGACTCCACGACAT

MAPK: Mitogen-activated protein kinase; JNK: C-Jun N-terminal kinases; ERK: Extracellular signal-regulated kinase; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase.

Statistical analysis

All experimental data are presented as mean ± SD. Statistical analyses were performed with GraphPad Prism software (version 9.5; GraphPad Software Inc., San Diego, CA, United States). Intergroup comparisons were conducted through one-way analysis of variance followed by Fisher’s least significant difference post hoc test. P < 0.05 indicated statistical significance. All in vivo experiments in this study included six biological replicates and three technical replicates per sample to ensure the reliability of the results. The statistical methods of this study were reviewed by Fang Liu from Beijing University of Chinese Medicine, who confirmed the appropriateness of the analytical approaches.

RESULTS

Saf-Cal b ameliorated liver fibrosis in CCl₄-induced liver fibrosis model rats

To evaluate the therapeutic effect of Saf-Cal b on liver fibrosis, we established liver fibrosis rat models. The rats were acclimatized for one week, followed by CCl₄ gavage for 4 weeks or 8 weeks. After the first week of gavage, drug treatment was administered for 3 weeks and 7 weeks (Figure 1A). Upon treatment completion, liver and spleen tissues were collected for histological analysis.

Hematoxylin and eosin staining revealed preserved hepatic architecture in controls, with intact lobules, radial hepatic cords/sinusoids, and no inflammation (Figure 1B). The 4-week CCl₄ rats developed perivenular inflammation, hepatocyte steatosis, and spotty necrosis, which progressed to confluent necrosis, lobular disorganization, and exacerbated inflammation by 8 weeks. In the group treated with Saf-Cal b, the hepatic tissue structure was arranged neatly, and the hepatocyte structure tended to be normal. Notably, the degree of normalization of hepatic tissue structure after Saf-Cal b treatment was superior to that of single drug treatment, specifically manifested by more complete hepatic lobule structure, more physiological arrangement of hepatic cords and sinusoids, and more obvious improvement in hepatocyte morphology and cytoplasmic homogeneity. Histopathological assessment using Masson’s trichrome staining confirmed extensive collagen deposition in CCl₄ rats, while treatment with Saf-Cal b resulted in progressive reduction of fibrotic matrix from 4 weeks to 8 weeks (Figure 1C).

Ultrastructural analysis via TEM revealed marked mitochondrial injury of hepatocytes in CCl₄ rats, including progressive swelling, cristae disruption, and quantitative reduction, with maximal severity at week 8 (Figure 1D). The combination therapy demonstrated superior cytoprotective effects compared with monotherapy, markedly attenuating mitochondrial damage and maintaining structural integrity.

The effects of Saf-Cal b on hepatic fibrosis: Hepatoprotection, anti-inflammation, and antioxidant action

At both 4 weeks and 8 weeks, the liver index and spleen index of rats in the CCl₄ group were significantly higher than those in the control group (P < 0.001; Figure 2A). Compared with results in the model group, the liver index and spleen index of animals treated with BJRG, the positive control, were significantly reduced at both 4 and 8 weeks. Similarly, compared with the model group, both Cal b and Saf monotherapy also reduced the liver index and spleen index at 8 weeks. Notably, at 4 weeks, the spleen index in the Saf-Cal b combination treatment group was significantly lower than that in the Cal b monotherapy group (P < 0.05). At 8 weeks, both the spleen index and liver index in the Saf-Cal b combination treatment group were significantly lower than those in the Cal b monotherapy group (P < 0.01).

Figure 2 Effects of the saffron and Calculus bovis treatment on liver function, inflammatory responses, and oxidative stress. A and B: Serum aspartate aminotransferase, alanine aminotransferase, and total bilirubin activities; C: Serum tumor necrosis factor-α and interleukin-1β levels; D: Glutathione peroxidase, superoxide dismutase and malondialdehyde concentration in the liver in the indicated groups. Data are shown as mean ± SD (n = 6). BJRG: Biejia Ruangan tablet; Saf: Saffron; Cal b: Calculus bovis; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TBIL: Total bilirubin; TNF: Tumor necrosis factor; IL: Interleukin; GSH-PX: Glutathione peroxidase; SOD: Superoxide dismutase; MDA: Malondialdehyde. ^aP < 0.001 vs control group, ^bP < 0.05 vs CCl₄ group, ^cP < 0.01 vs CCl₄ group, ^dP < 0.001 vs CCl₄ group, ^eP < 0.05 vs saffron and Calculus bovis group, ^fP < 0.01 vs saffron and Calculus bovis group, and ^gP < 0.001 vs saffron and Calculus bovis group.

After 4 and 8 weeks, the serum markers of liver injury (AST, ALT and TBIL) were significantly elevated in rats in the CCl₄ group compared with the control group (P < 0.001; Figure 2B). At 4 weeks, the BJRG-positive control exhibited significant hepatoprotective effects, specifically manifested by a significant reduction in AST and TBIL levels (P < 0.05). The levels of liver enzymes in the Cal b treatment group were significantly reduced (P < 0.05). Notably, compared with results in the Saf group, the TBIL level in the Saf-Cal b combination treatment group was significantly lower (P < 0.05). By 8 weeks, the positive control drug BJRG showed significant hepatoprotective effects, as indicated by reduced levels of AST, ALT, and TBIL (P < 0.05). Similarly, both Cal b and Saf monotherapy also reduced the levels of liver injury markers at 8 weeks. At 8 weeks, the Saf-Cal b combination treatment group had lower levels of liver enzymes (AST, TBIL) compared with the Saf group and Cal b group (P < 0.05); additionally, the ALT level in the combination treatment group was significantly lower than that in the Cal b group (P < 0.05).

Cytokine profiling by ELISA quantification revealed significant upregulation of pro-inflammatory mediators (TNF-α, IL-1β) in liver tissues of CCl₄ model rats compared with controls at both 4 weeks and 8 weeks (Figure 2C; P < 0.001). The levels were significantly reduced in the BJRG-positive control group at both 4 weeks and 8 weeks. At 4 weeks, monotherapy with Saf and Cal b reduced the levels of TNF-α and IL-1β (P < 0.05). At 8 weeks, Saf and Cal b monotherapy led to reduced TNF-α levels, while IL-1β levels were reduced by monotherapy with Saf only (P < 0.05). Notably, significantly reduced inflammatory cytokine levels were observed in the Saf-Cal b group compared with the Saf and Cal b groups at both 4 and 8 weeks (P < 0.05).

Oxidative stress evaluation revealed significant impairment of hepatic antioxidant defense mechanisms in CCl₄ rats compared with controls, characterized by reduced GSH-PX and SOD activities, along with elevated MDA levels (P < 0.001; Figure 2D). The BJRG-positive control showed significantly increased GSH-PX activity and SOD at 4 weeks and 8 weeks, and significant improvement in MDA only at 4 weeks (P < 0.05). At 4 weeks, compared with the model group, all treatment groups showed significant improvements in GSH-PX and SOD levels (P < 0.05), while only the Saf group and the Saf-Cal b group showed significantly improved levels of MDA (P < 0.05). The Saf-Cal b monotherapy group showed significant improvement in all parameters compared with model rats at 8 weeks (P < 0.05). At 8 weeks, Saf-Cal b had similar effects on GSH-PX as monotherapy treatments, while it was more effective in increasing SOD and decreasing MDA compared with monotherapy treatments (P < 0.05).

The effects of Saf-Cal b on hepatic fibrosis: Balancing fibrolysis and fibrogenesis

At both 4 weeks and 8 weeks, the level of HYP, a marker of collagen deposition, was significantly elevated in the CCl₄ rat group (P < 0.001; Figure 3A). The HYP level in the BJRG-positive control group was significantly reduced at both 4 and 8 weeks. At 4 weeks, the HYP levels in all groups except for the Cal b group were significantly decreased (P < 0.05). Notably, at 4 weeks and 8 weeks, the HYP level in the Saf-Cal b combination treatment group was significantly lower than that in the Cal b monotherapy group and the Saf monotherapy group (P < 0.05).

Figure 3 Effects of various drugs on hepatic fibrogenesis and degradation in rats. A: Hydroxyproline content in rat liver tissue and hyaluronic acid level in serum at 4 weeks and 8 weeks; B: Tissue inhibitors of metalloproteinases-1 and matrix metalloproteinase-2 content in rat liver tissues at 4 weeks and 8 weeks. Data are shown as mean ± SD (n = 6). BJRG: Biejia Ruangan tablet; Saf: Saffron; Cal b: Calculus bovis; HYP: Hydroxyproline; HA: Hyaluronic acid; TIMP: Tissue inhibitors of metalloproteinases; MMP: Matrix metalloproteinase. ^aP < 0.001 vs control group, ^bP < 0.05 vs CCl₄ group, ^cP < 0.01 vs CCl₄ group, ^dP < 0.001 vs CCl₄ group, ^eP < 0.05 vs saffron and Calculus bovis group, and ^fP < 0.01 vs saffron and Calculus bovis group.

Serological analysis revealed that the level of HA, which is positively correlated with HSC activation[18], was significantly increased in the serum of rats in the CCl₄ model group (P < 0.001; Figure 3A). The HA level in the BJRG-positive control group was significantly reduced at 8 weeks, and the level was also significantly decreased after therapeutic intervention (P < 0.01). At 4 weeks, the HA content was significantly reduced in both the Saf monotherapy group and the Saf-Cal b combination treatment group compared with the CCl₄ model group (P < 0.05). By 8 weeks, the HA levels in all drug-treated groups were significantly decreased (P < 0.01). Notably, the HA level in the Saf-Cal b combination treatment group was lower than that in the Saf monotherapy group and the Cal b monotherapy group (P < 0.05).

We next examined ECM production and degradation. TIMP-1 and MMP-2 precisely regulate ECM metabolism through a dynamic equilibrium in hepatic fibrosis[19,20]. TIMP-1 was elevated and MMP-2 was reduced in the liver tissue of CCl₄ rats at both 4 and 8 weeks (P < 0.01; Figure 3B). At 8 weeks, the BJRG-positive control group showed a significant decrease in TIMP-1 levels and a significant increase in MMP-2 levels, suggesting improved deposition and balance of ECM in liver tissues (P < 0.05). The Cal b group, Saf group, and Saf-Cal b group all showed significantly downregulated TIMP-1 levels and significantly upregulated MMP-2 levels at 4 and 8 weeks (P < 0.05). Notably, at 4 weeks, the effect of the Saf-Cal b treatment on improving TIMP-1 levels was significantly superior to that of Saf and Cal b, and the effect on improving MMP-2 levels was significantly superior to that of Cal b (P < 0.05). Similarly, at 8 weeks, the efficacy of the Saf-Cal b combination on improving TIMP-1 and MMP-2 levels was also significantly superior to that of Saf and Cal b (P < 0.05). Together, these experimental results confirm that Saf-Cal b treatment has a significant ameliorative effect on hepatic fibrosis in rats.

Network pharmacology analysis: Analyzing the targets of Saf and Cal b in anti-liver fibrosis

Our results indicate that Saf-Cal b treatment has a significant ameliorative effect on hepatic fibrosis. To investigate the mechanism of the anti-fibrotic action of the Saf-Cal b combination, we used network pharmacology methods to predict the potential key targets and signaling pathways. We first used network pharmacology prediction to identify potential targets of Saf and Cal b in anti-liver fibrosis. A total of 3026 potential targets of Saf and Cal b were collected using the TCMSP database and BAT-MAN database (Table 2). We also collected 755 liver fibrosis-related targets using the Genecard database and OMIM database. Venn diagram analysis revealed 342 intersecting targets between the targets of Saf and Cal b and liver fibrosis-related targets (Figure 4A). On the basis of these targets, we constructed the “plant active ingredient-target” network diagram of Saf-Cal b (Figure 4B and C). We constructed a PPI network of the interactions of the intersecting targets (Figure 4D).

Figure 4 Mechanism of saffron and Calculus bovis against hepatic fibrosis based on network pharmacology. A: Venn diagram of overlapping targets of saffron and Calculus bovis (Saf-Cal b) in the treatment of hepatic fibrosis; B: Network diagram of active ingredients and targets of Saf-Cal b in the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform database; C: Network diagram of active ingredients and targets of Saf-Cal b in the Bioinformatics Annotation Database for Molecular Mechanism of Traditional Chinese Medicine database; D: Protein-protein interaction network; E: Core clustering network; F: Bar chart of Gene Ontology functional enrichment analysis (biological processes, cellular components, molecular function) of targets of Saf-Cal b in the treatment of hepatic fibrosis; G: Bubble chart of Kyoto Encyclopedia of Genes and Genomes pathway analysis of targets of Saf-Cal b in treating hepatic fibrosis. BP: Biological processes; CC: Cellular components; MF: Molecular functions; PI3K-Akt: Phosphatidylinositol 3-kinase/protein kinase B.

Table 2 The components of saffron and Calculus bovis selected from the database.

Medicine	Component	Component ID	Database
Saffron	Quercetin	MOL000098	TCMSP
	Kaempferol	MOL000422
	Isorhamnetin	MOL000354
	Crocetin	MOL001406
	Crocetin	MOL001406	BATMAN-TCM
	Kaempferol	MOL000422
	Picrocrocin	MOL001408
Calculus bovis	Bilirubin	MOL008838	TCMSP
	Methyl desoxycholate	MOL008839
	ZINC01280365	MOL008846
	CLR	MOL000953
	Cherianoine	MOL008843	BATMAN-TCM
	Bilicerdin	MOL008834
	Cholic Acid	MOL009807
	CLR	MOL000953

TCMSP: Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform; BATMAN-TCM: Bioinformatics Annotation Database for Molecular Mechanism of Traditional Chinese Medicine; CLR: Cholesterin.

The 342 targets were analyzed for protein interaction relationships (confidence > 0.9), and the protein interaction relationships were expressed in Cytoscape 3.8.2. Cluster analysis was performed using the MCODE plugin, and the highest score clustering network was obtained using K-core values (Figure 4E). MCODE analysis results showed that MAPK, IL-6R, mothers against decapentaplegic homolog 3 (SMAD3), EGF, CEBPB, IFNG, SMAD2, ERBB2, NOTCH1, CD4, IGF1, JUN, JAK2, MAPK1, SMAD4, AKT1, CCND1, HIF1A, GRB2, PDGFRB, FOS, PDGFB, SRC, PTPN11, EGFR, PIK3CA, MYC, and PDGFRA genes were the hub genes for clustering. Integrated network pharmacology analysis of PPI networks and MCODE module clustering identified the MAPK signaling pathway (highest K-core value) as the central potential mechanism underlying the anti-fibrotic effects of the Saf-Cal b combination (Figure 4E). The MAPK cascade critically modulates transforming growth factor-β (TGF-β)/SMAD3 signaling transduction, thereby driving HSC activation and ECM deposition, which are central to the pathogenesis of liver fibrosis[12,21]. Thus, we focused on the MAPK signaling pathway for in-depth mechanistic investigation.

Gene Ontology enrichment analysis of the 342 target proteins was conducted using the Majorbio Cloud Platform with a significance threshold of P < 0.001. The results identified 114 biological processes, 15 cellular components, and 21 molecular functions with significant enrichment. The top 10 ranked terms per category revealed that biological processes predominantly involved transcriptional regulation (RNA polymerase II promoter), signal transduction, and modulation of cell proliferation/apoptosis; cellular components was enriched in cytoplasmic, plasma membrane, nuclear, and extracellular compartments; while molecular function highlighted critical functional domains, including protein kinase activity (serine/threonine/tyrosine), ATP zinc ion binding, and DNA/enzyme interactions (Figure 4F).

KEGG pathway enrichment analysis identified 44 significantly enriched signaling pathways (P < 0.001). The top 10 entries (ranked by descending gene count) included pathways in cancer, endocrine resistance, proteoglycans in cancer, lipid metabolism and atherosclerosis, advanced glycation end products-receptor for advanced glycation end products signaling in diabetic complications, human cytomegalovirus infection, Kaposi sarcoma-associated herpesvirus infection, hepatitis B infection, phosphatidylinositol 3-kinase/protein kinase B signaling pathway, and chemical carcinogenesis-receptor activation (Figure 4G).

The Saf-Cal B combination suppresses the MAPK signaling pathway and inhibits HSC activation in CCl₄ rats

Next, we evaluated the effects of Saf, Cal b, and Saf-Cal b on the MAPK pathway. Quantitative real time polymerase chain reaction analysis showed that the mRNA levels of p38 MAPK, JNK, and ERK in the liver tissues of CCl₄ rats were all increased at 4 weeks and 8 weeks (P < 0.001; Figure 5A and B). In the BJRG-positive control group, the mRNA levels of these factors were significantly decreased at both 4 weeks and 8 weeks (P < 0.05). The Cal b group, Saf group, and Saf-Cal b group also showed significantly downregulated p38 MAPK and JNK mRNA levels at 4 weeks and 8 weeks (P < 0.05). While the mRNA level of ERK in the Saf-Cal b group was significantly downregulated at 4 weeks (P < 0.05), monotherapy treatment had no significant effect. At 8 weeks, Saf and Saf-Cal b treatments significantly downregulated ERK mRNA (P < 0.05). Notably, compared with Saf or Cal b monotherapy, the Saf-Cal b regimen had a superior effect on downregulating p38 and JNK mRNA levels in CCl₄ rats (P < 0.05).

Figure 5 Potential mechanism of saffron and Calculus bovis against hepatic fibrosis. A: Effects of each drug on mRNA levels of p38 mitogen-activated protein kinase, c-Jun N-terminal kinases, and extracellular signal-regulated kinase at 4 weeks; B: Effects of each drug on mRNA levels of p38 mitogen-activated protein kinase, c-Jun N-terminal kinases, and extracellular signal-regulated kinase at 8 weeks; C: Immunofluorescence results of p38 and α-smooth muscle actin in rat liver tissues at 4 weeks and 8 weeks; D: Protein expression of type I collagen, phosphorylated-p38, phosphorylated c-Jun N-terminal kinases, phosphorylated extracellular signal-regulated kinase, transforming growth factor-β, and mothers against decapentaplegic homolog 3 in each group at 4 weeks; E: Protein expression of type I collagen, phosphorylated-p38, phosphorylated c-Jun N-terminal kinases, phosphorylated extracellular signal-regulated kinase, transforming growth factor-β, and mothers against decapentaplegic homolog 3 in each group at 8 weeks. α-SMA: Α-smooth muscle actin; COI-I: Type I collagen; TGF-β: Transforming growth factor-β; SMAD3: Mothers against decapentaplegic homolog 3; BJRG: Biejia Ruangan tablet; Saf: Saffron; Cal b: Calculus bovis; JNK: C-Jun N-terminal kinases; ERK: Extracellular signal-regulated kinase; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase. ^aP < 0.001 vs control group, ^bP < 0.05 vs CCl₄ group, ^cP < 0.01 vs CCl₄ group, ^dP < 0.001 vs CCl₄ group, ^eP < 0.05 vs saffron and Calculus bovis group, and ^fP < 0.01 vs saffron and Calculus bovis group.

Activation of HSCs is a central event in hepatic fibrosis, with α-SMA serving as a classical marker of HSC activation[22]. Following activation, HSCs exhibit massive secretion of ECM components. Immunofluorescence analysis demonstrated markedly enhanced co-expression and co-localization of p38 MAPK and α-SMA in HSCs of CCl₄ rats, with a time-dependent increase in fluorescence intensity correlating with fibrotic progression. Suppressed expression of these markers was observed in all treatment groups, with the Saf-Cal b combination exhibiting more pronounced inhibition of p38 MAPK/α-SMA co-localization compared with monotherapy regimens (Figure 5C).

Col-I has been established as a core pathological biomarker in hepatic fibrosis[23]. Western blot analysis revealed that while the protein expression of Col-I was significantly elevated in CCl₄-induced fibrotic rats at both 4 weeks and 8 weeks (P < 0.01), Col-I levels were significantly reduced in all treatment groups (P < 0.05; Figure 5D and E). The Col-I protein level in the Saf-Cal b group was significantly lower than that in the Cal b group at 4 weeks and 8 weeks (P < 0.05).

Additionally, the phosphorylated protein levels of p38, JNK, and ERK in CCl₄ model rats were significantly upregulated (P < 0.05). We observed that all drug treatments significantly inhibited the phosphorylated protein levels of p38 MAPK and JNK at 4 weeks and 8 weeks (P < 0.05). However, only Saf-Cal b treatment significantly inhibited the phosphorylated protein level of ERK at 4 weeks and 8 weeks (P < 0.05). Notably, p38 MAPK phosphorylation was significantly lower at 4 weeks and 8 weeks in the Saf-Cal b group compared with the Saf and Cal b groups (P < 0.05). JNK phosphorylation was also significantly lower in the Saf-Cal b group at 8 weeks compared with the Saf and Cal b groups (P < 0.05).

p38 MAPK may directly regulate HSC activation, thereby influencing hepatic fibrogenesis[9]. Structural biology studies revealed significant conformational homology between the ATP-binding domain of the TGF-β type I receptor and the kinase domain of p38 MAPK[24]. Inhibition of p38 MAPK signaling leads to downregulation of the TGF-β/SMAD3 pathway[25]. Our experimental data showed that the protein expression levels of TGF-β and SMAD3 were significantly elevated in CCl₄-induced fibrotic rats at both 4 weeks and 8 weeks (P < 0.01; Figure 5E). The protein expression levels of TGF-β in all drug-treated groups were significantly downregulated at 4 weeks and 8 weeks compared with levels in the CCl₄ model group (P < 0.05). While SMAD3 protein was induced in CCl₄ model rats, Saf-Cal b treatment exerted a significant inhibitory effect on SMAD3 levels at 4 weeks (P < 0.05). Notably, at 8 weeks, TGF-β protein expression was significantly lower in the Saf-Cal b group compared with the Cal b group, and SMAD3 protein expression was significantly lower than that in the Saf and Cal b groups (P < 0.05).

DISCUSSION

Liver fibrosis is a common pathological stage in most chronic liver diseases, representing a significant global health issue, with no targeted therapeutic options currently available[26]. Recent studies have increasingly demonstrated the potential of Tibetan medicine in the treatment of liver fibrosis[16]. In Tibetan medicine theory, the Saf-Cal b combination is the predominant therapeutic pairing for chronic hepatic diseases. However, the pharmacological mechanisms of this drug combination remain unclear.

Saf-Cal b is a commonly used drug pair in Tibetan medicine for treating chronic liver diseases. Because of the scarcity and high cost of natural bezoar, substitutes such as artificially synthesized derivatives and bioengineered bezoar preparations have been developed. Artificially synthesized derivatives show similar efficacy and replicate the main components of natural bezoar, while bioengineered preparation balances yield and quality. Synthetic preparations, with high cost-effectiveness and stable efficacy, are highly recognized in Tibetan clinical practice; thus, artificial bezoar was used in this experiment.

The CCl₄-induced toxic liver fibrosis rat model simulates the pathological state of human liver fibrosis, resulting in mitochondrial damage of hepatocytes, decreased ATP levels, and increased reactive oxygen species, thereby triggering oxidative stress[27]. This model stimulates the production of inflammatory factors in the liver, further igniting the inflammatory response and promoting the progression of liver fibrosis[28]. This study systematically evaluated the temporal therapeutic effects of the combination of Saf and Cal b in the CCl₄-induced liver fibrosis rat model at two time points (4 weeks and 8 weeks).

In the Guidelines for the Integrated Traditional Chinese and Western Medicine Treatment of Liver Diseases, BJRG is recommended for the treatment of liver cirrhosis induced by hepatitis B[29]. Therefore, we used BJRG as a positive control drug in this study. Our experimental results revealed that both BJRG and Saf-Cal b exert anti-hepatic fibrosis effects, and there are both similarities and differences in their anti-hepatic fibrosis mechanisms. BJRG and Saf-Cal b exerted highly consistent effects at the TGF-β/SMAD3 pathway level; this is consistent with the conclusion from Yang et al’s study[27]. BJRG and Saf-Cal b also exhibit differences in their mechanisms of action against liver fibrosis. At 4 weeks, BJRG had no effects on rat AST, HA, TIMP-1, and MMP-2 levels or ERK phosphorylation; at 8 weeks, it still failed to significantly improve rat MDA levels. In contrast, Saf-Cal b exerted a regulatory effect on all the above factors[30].

Histopathological assessment revealed that, compared with monotherapy, the Saf-Cal b combination resulted in significant improvements in liver structure, characterized by alleviated fatty degeneration and necrosis of hepatocytes and reduced ECM accumulation. The change in HYP and HA levels, markers of liver fibrosis severity, confirmed the anti-liver fibrosis effect of Saf-Cal b, and the effects were better than those of monotherapy treatments. Furthermore, Saf-Cal b effectively restored ECM homeostasis, as evidenced by a rebalancing of TIMP-1 and MMP-2. Lipid peroxidation reaction in chronic liver injury is an important inducer of liver fibrosis[31]. Notably, Saf-Cal b demonstrated a stronger hepatoprotective effect and alleviated hepatic oxidative damage, with significant reductions in AST, ALT, and TBIL levels, enhancing GSH-PX and SOD activities while reducing MDA levels, indicating that Saf-Cal b possesses superior anti-fibrotic potential. From the change of inflammatory factors and oxidative stress indicators, the Saf-Cal b combination demonstrated a more significant therapeutic effect at the 8-week time point than at the 4-week time point, which indicated better effects over a longer treatment time. These results provide insights into why the Saf-Cal b combination was the most common combination treatment for chronic liver disease in Tibetan medicine.

To clarify the molecular mechanisms by which the Saf-Cal b regulates liver fibrosis, we conducted network pharmacology analyses. The analysis of both the predicted targets and signaling pathways served as a foundation for the subsequent validation experiments. From the pathway results, we observed a low correlation between certain top-ranked pathways and hepatic fibrosis. Consequently, we decided to prioritize the target data, supplemented by insights from published mechanisms. Since a detailed examination of these unrelated pathways was beyond the primary scope of this study, we deem it prudent to reserve them for future research. Using PPI and MCODE analyses, MAPK signaling was identified as a potential mechanism underlying the anti-fibrotic effects of the Saf-Cal b combination. KEGG pathway enrichment analysis revealed that the anti-fibrotic mechanisms of Saf-Cal b involve multiple pathways. Notably, substantial evidence indicates close associations between MAPK signaling and the identified pathways[32,33]. Our results showed that as liver fibrosis progressed, the expression of p38 MAPK and α-SMA in the liver tissues was significantly upregulated, accompanied by the activation of HSCs. Following Saf-Cal b intervention, the expression of p38 MAPK was significantly downregulated, effectively inhibiting HSC activation[34]. Multiple studies have indicated that p38 MAPK can influence HSC activation by regulating the downstream TGF-β signaling pathway, thereby promoting the progression of liver fibrosis[35,36]. Our results showed that Saf-Cal b significantly inhibits the mRNA and protein expression levels of p38 MAPK, JNK, and ERK and their phosphorylation. These changes in kinase activity may further downregulate the TGF-β/SMAD3 pathway, ultimately directly influencing the development of liver fibrosis.

Our study also generated some notable results. At the 4-week timepoint, Cal b treatment exhibited relative superiority in improving serum liver function parameters, whereas Saf monotherapy exerted more pronounced effects on restoring hepatic fibrotic metabolic homeostasis. The Saf-Cal b combination therapy exhibited enhanced efficacy over monotherapy, manifesting as more robust antioxidant, anti-inflammatory, and fibrometabolic regulatory activities. A differential inhibition of p38 phosphorylation was observed with the combination, despite the absence of significant changes in the mRNA expressions of p38, JNK, and ERK. A more pronounced advantage emerged at 8 weeks, where the combination therapy exerted superior modulation of p38, JNK, and ERK gene expression and more potent suppression of p38 and JNK phosphorylation, along with improved negative regulation of the TGF-β/SMAD3 pathway. Together, these findings demonstrated that the combination treatment exerted a more profound effects compared with monotherapy. These results further provide robust evidence for the application of Saf-Cal b in liver fibrosis. This study has several limitations. While the pharmacological effects and mechanisms of Saf-Cal b were validated through network pharmacology combined with in vivo experiments, systematic assessment of the overall efficacy via in vitro models is lacking. These issues require further investigation.

CONCLUSION

Our experimental data demonstrate that Saf-Cal b exerts significant anti-fibrotic effects against hepatic fibrosis. The therapeutic efficacy of Saf-Cal b treatment surpassed that of Saf or Cal b monotherapy. Further experimental results suggested that the Saf-Cal b combination simultaneously inhibited the p38 MAPK and TGF-β/SMAD pathways, thereby intervening in the progression of hepatic fibrosis. Finally, we present a concise overview of the core research framework of this study in Figure 6.

Figure 6 Graphical abstract. Saf: Saffron; Cal b: Calculus bovis; PPI: Protein-protein interactions; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; ERK: Extracellular signal-regulated kinase; MAPK: Mitogen-activated protein kinase; JNK: C-Jun N-terminal kinases; TGF-β: Transforming growth factor-β; SMAD3: Mothers against decapentaplegic homolog 3.