Sophoricoside from sophora japonica L. is efficacious as monotherapy or in combination with lenvatinib in hepatocellular carcinoma via targeting EGFR.

Introduction
Hepatocellular carcinoma (HCC) ranks the sixth among cancers in terms of its morbidity and the third among factors leading to cancer-associated mortality globally1. Currently, HCC is mainly treated with surgical resection, endovascular intervention, targeted chemotherapy, radiotherapy, and traditional Chinese medicine (TCM) treatment. Drug therapy is an indispensable tool in the treatment of intermediate and advanced HCC, with targeted drugs occupying an important position2. Currently, the approved first-line monotherapy treatment regimens for HCC include sorafenib, lenvatinib and donafenib. Based on the REFLECT results, the overall survival with lenvatinib was not inferior to sorafenib3. Lenvatinib serves as the selective, multi-target tyrosine kinase inhibitor (TKI) with anti-tumour effect by inhibiting neoangiogenesis and lymphangiogenesis, modulating immunity and remodeling the tumor immune microenvironment4. Lenvatinib-based chemotherapy regimens are the current mainstay of treatment, but the presence of severe drug resistance to this agent leads to a poor prognosis for most patients with HCC5. Consequently, it is critical to develop efficient combined treatment for the improvement of survival outcomes for HCC patients. TCM plays a role in HCC management and has significant benefits in improving patients’ clinical symptoms, boosting immunity, improving quality of life and prolonging survival. In recent years, TCM combined with targeted drug therapy has become a promising approach for HCC treatment. For instance, Withaferin A synergistically potentiates sorafenib antitumor activity via ROS-driven endoplasmic reticulum stress and DNA damage in HCC cells6. Flavonoids from Sophora alopecuroides collaboratively enhances sorafenib’s antitumour activity against HCC7. Consequently, it is urgently needed to research and develop novel TCM that can morre effectively treat HCC or improve the efficacy of targeted therapy.
Sophoricoside (Sop), a key compound extracted in herbal Sophora japonica L., is widely used in traditional Chinese medicine and modern pharmaceutical research. It has attracted attention for its diverse pharmacological activities8. Sop exerts anti-inflammatory and immunosuppressive effects. Sop inhibits Bach1/AKT and PI3K/AKT signalling to ameliorate Crohn’s disease-like colitis and acute lung injury. 9,10. Additionally, Sop shows cytotoxicity to breast cancer cells11. Sop attenuates autoimmune impairment through reducing oxidative stress as well as NF-κB signalling activation within hepatocytes12. Despite its known pharmacological properties, possible therapeutic efficacy of Sop and its associated mechanism in HCC have not been extensively investigated.
As the cell surface receptor, epidermal growth factor receptor (EGFR) belongs to a subfamily of receptor tyrosin kinases (RTKs). Homologous ligands of EGFR induce homo-or heterodimerisation of EGFR and cause intracellular tyrosine residues to phosphorylate, and govern multiple biological events including cell growth, migration, apoptosis and metabolic regulation13. EGFR is overexpressed among 68% of HCC patients, which is markedly related to vascular invasion, low differentiation, metastasis and poor survival14. Moreover, over-activation of EGFR signaling is closely linked to the intrinsic or acquired resistance of lenvatinib in HCC patients. To be specific, EGFR-PAK2-ERK5 pathway can be feedback-activated through suppressing fibroblast growth factor receptor (FGFR) through lenvatinib therapy5. HCC cells also become lenvatinib-resistant via EGFR activation and EGFR-STAT3 signaling axis stimulation, which is related to abnormalities in lipid raft stimulation and cholesterol metabolism15. Lenvatinib combined with EGFR inhibition shows significant synergy in HCC in vitro and in vivo5. Therefore, EGFR is the critical therapeutic target for HCC treatment and improving the sensitivity of lenvatinib.
The present work focused on investigating anti-HCC efficacies of Sop both singly and in combination with lenvatinib. Our results showed that Sop significantly inhibited HCC cell survival and proliferation but enhanced antitumor impact of lenvatinib in vitro and in vivo through targeting EGFR and inactivation of AKT and STAT3 pathways. The results suggest that Sop combined with lenvatinib is the candidate treatment against HCC.

Materials and methods

Cells and drugs
This study acquired Huh7 and SK-Hep-1 in Cellcook Biotechnology Company (Guangdong, China), and maintained them within Dulbecco’s modified Eagle’s medium (DMEM, Gibco, USA) that contained 10% fetal bovine serum (FBS, Vivacell) as well as 1% penicillin-streptomycin (Gibco, USA) under 37 °C and 5% CO2 conditions. Sop (HY-N0423) and lenvatinib (HY-10981) were provided by MedChemExpress (MCE, China). The purity of Sop used in our study was 99.36%.

Cell proliferation detection
For examining effects of Sop on HCC cell growth, we seeded cells (2 ~ 4 × 103/well) into the 96-well plates. At specified time points, Cell Counting Kit-8 (CCK-8, Dojindo, China) solution was introduced into every well. Following 1 h of incubation under 37 °C, absorbance was measured with the microplate reader (Tecan) at 450 nm.

Colony formation assay
Cells (4 × 103/well) were inoculated into the 6-well plates, followed by drug treatment every 3 days for 7–10 days. After 4% paraformaldehyde fixation and 1% crystal violet staining, a clone number contained at least 50 cells and statistics were calculated in each group.

Flow cytometry
Following trypsinization, cells underwent centrifugation, harvesting, and rinsing by PBS to conduct flow cytometry. In cell apoptosis assay, cells were subjected to 15 min of Annexin V-FITC and PI staining in line with the Apoptosis Detection Kit (Dojindo) instructions, followed by measurement with the flow cytometer (Beckman). Kaluza Flow Cytometry Analysis Software (http://kaluzasoftware.com/) was employed for data analysis. In cell cycle assay, cells experienced 2 h of 75% ethanol fixation under 4 °C, washing using cold PBS and then addition of working solution. Following incubation at 4 °C and 37 °C for 30 min, flow cytometry (Beckman) was conducted to detect those stained cells in line with Cell Cycle Assay Kit (Dojindo) instructions, and Kaluza software was adopted for data analysis.

Prediction of drug and disease targets from public databases
Sop targets were obtained from Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP), The Encyclopedia of Traditional Chinese Medicine (ETCM), Search Tool for Interacting Chemicals (STITCH), Similarity ensemble approach (SEA), Swiss Target Prediction and Genecards. Additionally, we obtained HCC targets based on the Therapeutic Target Database (TTD), Genecards, Online Mendelian Inheritance in Man (OMIM), Drugbank online, GSE101685 and GSE36376 from Gene Expression Omnibus (GEO). The shared targets of Sop-HCC were identified using Venny online tool.

Network analysis of common targets
This study used STRING for constructing the PPI network of shared targets of Sop-HCC. Cytoscape software 3.10.2 (http://www.cytoscape.org/) was used to calculate PPI networks, identify hub genes and perform visual analysis. Both online approach SRplot and Database for Annotation, Visualisation and Integrated Discovery (DAVID) were utilized for GO as well as KEGG analysis.

Molecular docking and surface plasmon resonance (SPR) assay
The crystallographic structures of the core targets were obtained from the PDB database, while the 2D structure of Sop was retrieved based on PubChem database and later transformed into the 3D format with Chemdraw software 22 (http://www.chemdraw.com.cn). Target protein was then docked to Sop using iGEMDOCK.
Follow the OpenSPRTM Instrument Standard Operating Procedures to install the COOH chip. Start the run at maximum flow rate of 150 µL/min using PBST as assay buffer. After reaching the signal baseline, sample 200 µL of IPA (isopropanol), run for 10 s to evacuate bubbles, rinse the sample loop with buffer after reaching baseline and evacuate with air. After reaching the signal baseline, set the buffer flow rate to 20 µL/min. Activate the chip with EDC/NHS solution (1:1). Incubate 200µL in fixative for 4 min, rinsing the sample loop with buffer and evacuating with air. Add 200µL blocking solution, rinse sample loop with buffer and evacuate with air. Dilute the analytes with Analyte Buffer at the concentrations indicated in the experimental results and sample at 20 µL/min. The analytes are bound to the ligand for 240 s, the ligand dissociates naturally for 360s. TraceDrawer was employed for data analysis.

Cellular thermal shift assay
Following 48 h of treatment with Sop, HCC cell suspensions were aliquoted into PCR tubes. Samples were exposed to discrete temperatures (ranging from 37℃ to 60℃) for 5 min, followed by immediate cooling on ice for 3 min. Cell lysis was then performed through three cycles of freezing in liquid nitrogen and thawing in a 37 °C water bath. Lysates were clarified by centrifugation at 20,000 rpm for 10 min at 4 °C. The supernatant was transferred to new pre-cooled tubes. Total protein concentration was determined using a BCA Protein Assay Kit. Then EGFR levels were analyzed by Western blotting.

Plasmid construction and lentiviral production
To overexpress EGFR, the full-length EGFR was cloned into the lentiviral transfer vector. Lentiviral particles were generated in HEK-293T cells by co-transfecting the expression or control plasmids with the packaging plasmids psPAX2 and pMD2.G. Viral supernatants were harvested 24 h post-transfection and used to transduce HCC cells in the presence of 5 ug/mL polybrene (Solarbio, Beijing, China). Stable cell lines were established by selection with 2 ug/mL puromycin (Solarbio) for one week.

Protein half-life assays
Sop was added to treat HCC cells for a 48-h duration, followed by treatment with cycloheximide (200 ug/mL) for different times to block protein synthesis. At 0, 4, 8 and 12 h, cells were collected and examined through Western blot assay.

Ubiquitination assay
The exponential-phase HCC cells were treated with Sop for 48 h, then treated with 10 µM MG132 (MCE, China) for a 10-h duration. Later, cells were collected for lysis using RIPA buffer containing protease inhibitor cocktail (MCE, China). The protein solution obtained was immunoprecipitated with an anti-EGFR antibody (A11351, Abclonal) and Dynabeads™ Protein-G beads (Invitrogen), and then the ubiquitination level of EGFR was tested with an anti-ubiquitin antibody (3936, Cell Signal Technology) via western blotting.

Western blot
HCC cells treated with drugs for 48 h or xenograft tumor tissue after different treatments were harvested for lysis using RIPA reagent (Beyotime biotechnology, Shanghai, China) that contained phosphatase inhibitors (C0104-A and C0104-B, Lablead) and protease inhibitors (MB2678, Meilunbio). Total protein content was determined using a BCA assay kit (Thermo Fisher Scientific). After heating at 100 °C for 8–10 min, 30 µg of protein samples were subjected to 6–12% SDS-PAGE, followed by transfer onto polyvinylidene difluoride (PVDF) membranes. After 1 h of sealing using 5% BSA, primary antibodies were added for incubation for at least 12 h under 4 °C. Then, membranes were washed with TBST and further probed using secondary antibodies under ambient temperature for a 1-h duration. Membranes were incubated using ECL substrate (Millipore, Billerica, MA, USA), and antibody binding was observed using Chemiluminescent HRP substrates (Millipore, USA) and an Alliance Q9 (UVITEC). Antibodies were as follows: Anti-GAPDH (AB0037, Abways); anti-CDK1 (19532-1-AP), anti-CyclinB1 (28603-1-AP), anti-PCNA (10205-2-AP), anti-EGFR (18986-1-AP, Proteintech); anti-phosphorylated EGFR (AP0301, ABclonal); anti-AKT (9272 S), anti-phosphorylated AKT (9271 S), anti-phosphorylated STAT3 (9145 S), anti-STAT3 (4904 S), anti-cleaved PARP (9541 S), anti-cleaved caspase-3 (9509 S) and anti-PARP (9532 S, CST),.

Animal experiments
In brief, 2 × 106 Huh7 cells were injected subcutaneously into 4-week-old BALB/c male nude mice (SLAC Laboratory Animal, Shanghai) in the right flank. Tumor width and length were detected daily to monitor tumour growth. The formula V (mm3) = 1/2 × length × width2 was used to calculate tumour volume. Eight mice were randomized into two groups in the Sop alone experiment: control (saline, oral administration), Sop (40 mg/kg/day, oral administration). Sixteen mice were randomised into four groups to test whether the combined treatment worked: control (saline, oral administration), lenvatinib (10 mg/kg/day, oral administration), Sop (20 mg/kg/day, oral administration), and lenvatinib + Sop application. After this experiment, each mouse was sacrificed using cervical dislocation, and tumor tissue was excised for measurement of tumor weight and volume. We have performed an analysis of tumor growth inhibition (TGI) and estimated the combination effect. We calculated the TGI for each group (vs. Vehicle) and then calculated the Expected TGI for the combination group assuming additivity using the formula: Expected TGI = TGI_Sop + TGI_Len - (TGI_Sop * TGI_Len)/100. The Observed TGI for the combination was then compared to the Expected TGI. A ratio (Observed/Expected) > 1 indicates synergy. Tumor tissue was immersed within formalin to conduct immunohistochemical (IHC) staining, followed by tumor tissue protein extraction for Western blotting. Each animal experiment gained approval from Ethics Committee of Xiamen University (Reference No.: XMU LAC20230083) and was performed following Guide for the Care and Use of Laboratory Animals.

Immunohistochemical (IHC) staining
Xenograft tumor tissue from the nude mice was subjected to 24 h of immersion within formalin solution, dehydration within ethanol, paraffin embedding and sectioning into 3 μm sections for IHC staining. After 2 h of drying under 60 °C, sections underwent xylene deparaffinization and gradient ethanol rehydration, and antigen retrieval was performed with Tris-EDTA retrieval solution under high pressure and temperature, followed by blocking using 3% hydrogen peroxide as well as 12 h of incubation with anti-EGFR (18986-1-AP, Proteintech), anti-pEGFR (Y1068, Abclonal), Ki-67 (MAB-0672, Maixin Biotechnology, Fuzhou, China) and CDK1 (19532-1-AP, Proteintech) under 4℃. The secondary antibody is subjected to 30 min of incubation using goat anti-rabbit IgG under ambient temperature. Finally, haematoxylin and DAB (Maixin Biotechnology, Fuzhou, China) staining was performed.

Statistical analysis
GraphPad Prism 9.0 (GraphPad Software, La Jolla, CA, USA) was adopted for statistical analysis. Except as otherwise noted, results were represented by mean ± standard deviation (SD). Between-group difference was compared by the Student’s t-test, while among-group difference was compared by the one-way ANOVA followed by Tukey’s post hoc test. P < 0.05 stood for significant difference.

Results

Sop inhibits HCC cell survival, growth and arrests cell cycle
The chemical structure of SOP is shown in Fig. 1A. For investigating if Sop suppressed HCC cells, the commonly used HCC cells (Huh7 and SK-Hep-1) were exposed to 48 h of Sop treatments at varying doses, then CCK-8 assay was conducted to detect half-maximal inhibitory concentrations (IC50). Based on our findings, Sop suppressed Huh7 and SK-Hep-1 cell viability dose-dependently. The IC50 levels of Huh7 and SK-Hep-1 cells were 358.46 ± 21.2 µM and 317.1 ± 24.7 µM, respectively (Fig. 1B). Additionally, Sop showed a significant suppressive effect on other HCC cell lines (Supplemental Fig. 1 A). Next, we conducted CCK-8 (Fig. 1C) and colony formation assays (Fig. 1D) for determining how Sop affected HCC cell proliferation. Our results showed that the proliferation decreased in a gradient with increasing concentrations of Sop. To investigate whether Sop attenuated HCC cell growth via arresting cell cycle or causing apoptosis, flow cytometry was performed. We found that Sop was unable to induce significant apoptosis in these two cell lines (Supplemental Fig. 1B). Whereas, the Huh7 and SK-Hep-1 cell number blocked in G2/M phase apparently elevated following Sop administration (Fig. 1E). For validating this result, we carried out Western blotting assay for detecting proliferation marker PCNA and G2 phase markers cyclin B1 and CDK1 (Fig. 1F). Consistently, the results showed dose-dependent inhibitory effect of Sop on these proteins. Together, these findings suggest that Sop suppressed HCC growth while arresting cell cycle.

Sop suppresses HCC proliferation in vivo
For further verifying anti-HCC efficacy of Sop in vivo, nude mice were subcutaneously injected with Huh7 cells. After 1 week, animals were randomized as control or drug administration group. Our results showed that SOP-treated mice had significantly smaller tumour volumes and weights than controls. (Fig. 2A-C2). In addition, cell proliferation and cell cycle arrest were investigated by detecting Ki-67 and CDK1 expression. As shown by IHC staining, both Ki-67 and CDK1 levels were markedly reduced by Sop treatment (Fig. 2D and E), supporting that Sop administration suppressed cell growth while arresting cell cycle in vivo. The western blot assays testing proliferation and cell cycle markers also showed the similar results (Fig. 2F-G), which conformed to the in vitro results. However, there existed no notable disparity of mouse body weight reduction within both groups (Supplementary Fig. 2 A), signifying that the Sop treatment was well tolerated. Likewise, HE staining for mouse organs from the control and Sop-treated mice manifested no significant histological alterations (Supplementary Fig. 2B), denoting no overt toxicity of Sop. Collectively, these data evinced that Sop repressed HCC growth in vivo without obvious toxicity and adverse effects.

Investigation of the potential mechanism of sop against HCC by network Pharmacology and molecular Docking
For investigating the mechanism underlying anti-HCC effect of Sop, network pharmacology method was conducted. For the prediction of Sop’s targets, TCMSP, STITCH, SEA, ETCM, GeneCards and SWISS databases were analyzed and 176 potential targets of Sop were identified. Moreover, a total of 5103 differential expressed genes of HCC were obtained from OMIM, TTD, GeneCards, Drugbank and GEO databases. Through combination of these results, we screened 82 shared targets (Fig. 3A). Moreover, Cytoscape software and STRING database were used for showing the Sop-HCC target network diagram. This PPI network contained 322 nodes and 3116 edges (Fig. 3B and C). To further define the core target of Sop-HCC, cytoHubba analysis using Maximal Clique Centrality (MCC) algorithm of Cytoscape was conducted. The hub 10 targets were TNF, IL6, EGFR, VEGFA, HRAS, PTGS2, ESR1, SIRT1, SERPINE1 and FGF2 (Fig. 3D). After GO enrichment analysis, we found that these 82 targets could be significantly enriched in regulating proliferation, apoptosis, hypoxia and DNA biosynthetic processes (Fig. 3E). KEGG enrichment analysis also revealed significant enrichment of these genes in pathways associated with metabolism and cancer (Fig. 3F), further supporting Sop’s inhibition against HCC progression. For predicting potential binding between Sop and screened targets, molecular docking method was conducted and the data suggested that the value of binding energy between Sop and EGFR was the highest compared to other predicted molecules (Fig. 3G). Taken together, the bioinformatics analysis suggested that Sop might exert anti-HCC effects via targeting EGFR.

Sop promotes the degradation of EGFR
To further verify whether EGFR is the potential target of Sop, Huh7 and SK-Hep-1 cells were exposed to Sop at varying concentrations. We first detected whether Sop affected the EGFR expression. EGFR and its phosphorylation (pEGFR) levels were analyzed through Western blotting assay, which demonstrated a significant decline of both EGFR and pEGFR expression in HCC cells treated with increasing Sop concentration (Fig. 4A). Tumor tissues in Huh7 xenografts from nude mice were examined for EGFR and pEGFR levels by IHC. The treatment of Sop resulted in an obvious decease of EGFR and pEGFR expression (Fig. 4D). Furthermore, the SPR experiments suggested that the equilibrium dissociation constant (KD) of Sop and EGFR was estimated to be 56.1 µM (Fig. 4B), belonged to the “fast binding/fast dissociation” kinetic characteristics. CETSA experiment as recommended to investigate the direct binding between Sop and EGFR. Interestingly, the results showed no significant thermal stabilization of EGFR upon Sop treatment (Fig. 4C).

Afterwards, the role of Sop treatment in EGFR protein stability was analyzed in the presence of CHX, the inhibitor of protein translation. The results demonstrated that Sop intervention accelerated the degradation of EGFR protein (Fig. 4E) within HCC cells. Moreover, the proteasome inhibitor MG132, rather than the lysosomal inhibitor CQ, could partially abolish the Sop-induced decrease of EGFR expression (Fig. 4F). The ubiquitination modification of EGFR was significantly augmented by Sop (Fig. 4G). Altogether, our findings suggested that Sop attenuated the stability of EGFR through the ubiquitin-proteasomal pathway. Sop promotes ubiquitination-mediated degradation of EGFR. This leads to reduced EGFR protein levels under thermal stress, explaining the absence of observable stabilization in CETSA.
To further confirm EGFR as the functional target mediating Sop’s effects, we overexpressed EGFR in Huh7 and SK-Hep-1 cells. Western blot analysis demonstrated that sophoricoside treatment significantly downregulated EGFR protein expression (Fig. 4H). Notably, compared to control cells without EGFR overexpression, the downregulatory effect of Sop on EGFR was relatively attenuated in EGFR-overexpressing cell lines (Fig. 4I and J), supporting the critical role of EGFR targeting in the observed anti-HCC activity.

Combination of sop and lenvatinib synergistically inhibits HCC cell proliferation in vitro
As EGFR is the main cause of resistance of HCC to lenvatinib, we speculated that Sop might enhanced the suppressive effect of lenvatinib via targeting EGFR. To investigate whether Sop and lenvatinib synergistically inhibits HCC cell proliferation, Sop plus different concentration of lenvatinib were added into HCC cells. IC50 value of lenvatinib decreased gradually as Sop concentration increased (Fig. 5A). Furthermore, Chou-Talalay approach and Compusyn software (https://www.combosyn.com/) were employed for calculating Combination Index (CI), and it was found that CI values were less than 1, indicating that drug combination had strong synergistic inhibition against survival of HCC (Fig. 5B). Similarly, as demonstrated by CCK-8 and colony formation assays, Sop combined with lenvatinib caused higher inhibition effects on HCC cell proliferation (Fig. 5C and D).

Subsequently, this study used flow cytometry to assess the effects of Sop, alone and in combination with Lenvatinib on apoptosis. Sop alone exerted no notable effect influence on HCC cell apoptosis, yet the combination markedly augmented the apoptotic rate as compared to the sole treatment of lenvatinib (Fig. 5E). Consistent with this, the combination up-regulated pro-apoptosis-related protein, cleaved caspase 3 and cleaved PARP (Fig. 5F).
AKT and STAT3 pathways are the downstream pathways of EGFR, and their overactivation is closely associated with lenvatinib resistance. Therefore, we detected their expression in HCC cells treated with Sop and/or lenvatinib. Sop combined with lenvatinib significantly decreased pAKT and pSTAT3 expression, compared to treatment with either Sop or lenvatinib monotherapy (Fig. 5G). The above results suggest that Sop synergistically enhanced the anti-HCC efficacy of lenvatinib by restraining the EGFR-AKT and EGFR-STAT3 signaling.

Sop combined with lenvatinib synergistically attenuates HCC growth in vivo
To further evaluate the anti-HCC activity of Sop combined with lenvatinib on HCC in vivo, we constructed the nude mouse model bearing Huh7 xenograft tumors. Thereafter, animals were randomized as control, Sop monotherapy, lenvatinib monotherapy, and the combination treatment groups. The subcutaneous tumor volumes were determined at 2-day intervals following Sop and lenvatinib administration. As anticipated, tumor weight and volume of single-medication group had evidently diminished compared with control group; while those in combination group were markedly lower than single-drug groups (Fig. 6A and C). Analysis of tumor growth inhibition (TGI) revealed that the combination treatment yielded significantly greater suppression (Observed TGI = 70%) than the effect predicted by simple additivity (Expected TGI = 58%; Observed/Expected ratio = 1.21, > 1), indicating synergistic tumor suppression in vivo. Moreover, no significant between-group difference was found in the body loss, indicating that the combination was well tolerated (Supplemental Fig. 2 C). In consonance with the in vitro findings, western blot detection of the xenografted tumors manifested that Sop combined with lenvatinib markedly downregulated pAKT and pSTAT3 expression (Fig. 6D and E), while enhancing cleaved caspase 3 and cleaved PARP levels (Fig. 6F and G). These data suggest that Sop potentiates lenvatinib’s effect on suppressing HCC in vivo.

Discussion
The diagnosis of many HCC cases is made in the advanced phases, with susceptibility to merely systematic therapies. Natural products enjoy distinct advantages in terms of accessibility, security and affordability, thereby rendering them a favoured choice in eastern lands. Natural products boast diverse medicinal applications, encompassing anti-cancer, antioxidation, anti-inflammation and neuroprotection16–18. Discovering active compounds within natural products that possess minimal toxicity and few adverse effects for treating HCC has recently attracted attention. Sop, a genistein glycoside, manifests a wide spectrum of pharmacological efficacies, encompassing anti-inflammatory traits, anti-cancer potencies, immunosuppressive activities, hormonal and estrogenic actions19,20. To data, whether Sop showed an anti-HCC effect remains unclear. Here, the findings of our study demonstrated that Sop exerted a direct inhibitory effect on HCC. Sop treatment suppressed HCC cell survival and proliferation, and arrested cell cycle in vitro, while giving rise to a notable diminution in tumor growth without obvious toxicity in vivo. Aberrant activation of several cancer-related signaling pathways is related to HCC occurrence and development, such as EGF-EGFR, NF-κB, IL-6/STAT3, Hippo, and PI3K/AKT pathway. At present, drugs targeting some of these pathways have entered clinical research. Our mechanistic investigations using network pharmacology, molecular docking, and SPR experiments revealed that Sop targeted several key proteins and pathways associated with HCC progression. Specifically, Sop was proved to show potent affinity for EGFR and repress its stability, suggesting that EGFR is the possible target of Sop in treating HCC.
Lenvatinib has been approved as the first-line drug to treat HCC. Lenvatinib is the multikinase inhibitor suppressing fibroblast growth factor receptor (FGFR), platelet-derived growth factor receptor α (PDGFRα), vascular endothelial growth factor receptor (VEGFR), and proto-oncogenes KIT and RET. Almost 50% of HCC cases exhibit resistance to lenvatinib. The lenvatinib sensitivity in HCC is influenced by abnormal programmed cell death, cancer stem cell proliferation, metabolic remodeling, epigenetic regulation and transport processes21. Activation and amplification of EGFR and the corresponding downstream pathways (PAK2-ERK5, STAT3-ABCB1 and AKT-mTOR) are related to HCC resistant to lenvatinib. Blockage of EGFR signaling may abolish lenvatinib resistance5,15,22. Therefore, discovering natural products that can function as EGFR inhibitors are anticipated to be a significant means of surmounting lenvatinib resistance. Based on the suppressive function of Sop on EGFR expression, we speculated that Sop might enhance lenvatinib the sensitivity in HCC cells. As expected, Sop combined with lenvatinib synergistically attenuated HCC growth via inhibition of AKT and STAT3 in vitro and in vivo.
AKT pathway is crucial for several cellular physiological events, which shows abnormal activation within cancer, facilitating tumor occurrence and development23. PI3K/AKT signaling axis activation is detected in HCC, which promotes glucose uptake, facilitates glycolysis and increases tumor cell proliferation, and can also induce drug resistance and reduce the radiosensitivity of HCC cells24. Previously, Sop ameliorates Crohn’s disease-like colitis and acute lung injury through inhibiting Bach1/AKT and PI3K/AKT pathways9,10. Through network pharmacology, PI3K/AKT pathway is also identified as a pathway of Sop against HCC. Sop can down-regulate the phosphorylation of AKT, and when combined with lenvatinib, this inhibitory effect is enhanced, implying that the AKT pathway is an important mechanism of Sop action.
Notably, Sop alone does not induce apoptosis, contrasting with the pro-apoptotic effect of its combination with Lenvatinib. Lenvatinib monotherapy can partially inhibit the PI3K-AKT signaling pathway; however, this effect is often counteracted by compensatory activation of survival signals in resistant cells. The sustained activation of AKT signaling inhibits the activation of key apoptotic proteins, such as caspase-3, thereby promoting cell survival. 25 Our findings suggest that Sop significantly enhances the pro-apoptotic effect of lenvatinib in HCC. This synergistic effect provides a novel strategy for overcoming lenvatinib resistance. In light of recent research advances, we propose that Sop may target lenvatinib resistance mechanisms by inhibiting compensatory survival signals, such as the AKT pathway, which are not sufficiently suppressed by lenvatinib alone and may explain why profound apoptosis is not observed with Sop monotherapy.

Conclusion
In summary, our data provide compelling evidence that Sop is efficacious as monotherapy and potentiates the efficacy of lenvatinib, and that this effect is probably regulated by suppressing EGFR-AKT/STAT3 pathway (Fig. 7). The above results suggest that Sop is the promising adjunct therapy for HCC, and warrant further preclinical and clinical investigations to validate its therapeutic potential.

Supplementary Information
Below is the link to the electronic supplementary material.