Screening the Mechanism of Shikonin Against Renal Cell Carcinoma via Network Pharmacology, Molecular Docking, and Cellular Experimental Verification

Summary

This study identified the core targets and pathways of shikonin against RCC via multimethod analysis and in vitro validation, providing a framework for further mechanistic studies.

Abstract

Renal cell carcinoma (RCC) is one of the most common tumors in the urinary system and has the highest mortality rate. Previous investigations have demonstrated that shikonin can treat renal cell carcinoma, but the mechanism remains unclear. Therefore, our study aimed to elucidate the mechanism of shikonin in the treatment of renal cell carcinoma using network pharmacology, molecular docking, and in vitro functional assays, including cell proliferation, migration, and apoptosis, with western blot (WB) validation. Shikonin targets were screened using PharmMapper, SwissTargetPrediction, and other databases, and identified RCC-related targets from Online Mendelian Inheritance in Man (OMIM), GeneCards, and other databases; the potential therapeutic targets were obtained by intersection analysis. A protein-protein interaction (PPI) network was constructed, and Cytoscape was used to screen core targets, while molecular docking was applied to analyze the binding affinity between shikonin and key targets. A total of 374 shikonin targets and 1,087 RCC-related targets were collected, and 98 overlapping target genes were identified. Six core targets (SRC, PIK3CA, PIK3CB, PIK3CD, PTPN11, and PIK3R1) were identified. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses showed that shikonin exerted anti-RCC effects mainly by regulating protein kinase activity, HIF-1, and IL-17 signaling pathways. Molecular docking confirmed that shikonin had a strong binding affinity with these core targets. In vitro studies using human RCC Caki‑1 and 786‑O cells demonstrated that shikonin significantly inhibited cell proliferation and migration in a dose‑dependent manner, promoted cell apoptosis, and upregulated caspase‑3 and caspase‑8 activities. Western blot experiments further verified that shikonin modulated the expression of core target proteins and suppressed the HIF-1 signaling pathway. This study systematically elucidates the pharmacological mechanism of shikonin against renal cell carcinoma, providing a theoretical basis for the development and application of shikonin as a novel anti-RCC agent.

Introduction

Renal cell carcinoma (RCC) is the most common malignant tumor of the urinary system, accounting for approximately 90% of all renal malignant tumors and 2%–3% of adult malignant tumors worldwide. According to the GLOBOCAN 2022 data, there were about 434,800 new cases and 155,953 deaths of renal cell carcinoma globally in 2022, with the incidence and mortality of RCC showing a continuous upward trend. The American Cancer Society estimated 80,980 new cases and 14,510 deaths of renal cell carcinoma in the United States alone in 2025. Renal cell carcinoma has the highest mortality rate among urinary system malignant tumors, and patients with metastatic or drug-resistant RCC still exhibit a poor prognosis. With the widespread application of targeted therapy and immunotherapy, remarkable progress has been made in the clinical treatment of RCC. The first-line standard treatment for advanced or metastatic renal cell carcinoma mainly adopts combination regimens of immune checkpoint inhibitors, or regimens combining immune checkpoint inhibitors with anti-angiogenic tyrosine kinase inhibitors. Despite the continuous advancement of therapeutic approaches, a large number of patients still experience disease progression, drug resistance, or treatment intolerance. Therefore, the development of novel anti-RCC drugs with clear mechanisms, safety, and efficacy remains an urgent clinical need^1,2,3.

Shikonin is the principal bioactive constituent extracted from Radix Arnebiae, and its derivative acetylshikonin has exhibited prominent anti-RCC pharmacological effects. Existing evidence has demonstrated that shikonin exerts anti-RCC activity through multitarget and multipathway regulatory mechanisms. Mechanistically, shikonin can upregulate the expression of the tumor suppressor gene TEK and inhibit the phosphorylation of the AKT/mTOR signaling cascade, thereby suppressing the proliferation, migration, and invasion of RCC cells⁴; In addition, shikonin induces intracellular reactive oxygen species (ROS) accumulation to trigger mitochondrial dysfunction, further initiating multiple programmed cell death patterns including apoptosis, necroptosis, and autophagy in RCC cells⁵. It also modulates the expression of microRNAs such as miR-15b and miR-99b and regulates apoptosis-associated target genes, including FOXO1 and PDCD4, via the MAPK/ERK pathway, with these regulatory effects showing evident cell line specificity⁶. Notably, shikonin retains potent antitumor activity against sunitinib-resistant RCC cell lines, wherein it elicits therapeutic effects by activating necrosome complexes, inhibiting the AKT/mTOR pathway, and inducing G2/M cell cycle arrest⁷. Although the anti-RCC effects of shikonin have been partially validated by existing experimental studies, the core therapeutic targets, key signaling networks, and the underlying molecular regulatory mechanisms of shikonin against RCC remain unclear; moreover, systematic prediction, molecular verification, and experimental confirmation of its hub targets are still lacking, forming an important research gap in current studies.

Network pharmacology and molecular docking serve as complementary technical approaches in modern natural drug research and pharmacological mechanism exploration. From a systematic perspective, network pharmacology can systematically decode the multitarget, multipathway holistic interaction characteristics between bioactive compounds and disease-related targets. In contrast, molecular docking enables molecular-level verification of the binding affinity and stable interaction pattern between small-molecule compounds and core target proteins. The combination of these two methodologies constructs a systematic research closed-loop of holistic target prediction followed by molecular verification. Combining experimental validation with network pharmacology and molecular docking is a mainstream approach for natural drug research, which facilitates the clarification of disease pathogenesis and the elucidation of compound pharmacological mechanisms, and provides a theoretical framework for basic research and translational application of drugs^8,9.

In this study, network pharmacology was applied to predict the candidate targets of shikonin in the treatment of renal cell carcinoma. Molecular docking was used to theoretically verify potential targets, and the key action targets were finally determined through in vitro experiments and further confirmed by Western Blot (WB) experiments. The whole process of this study is shown in Figure 1.

Protocol

All human cell lines used in this study were purchased from authenticated commercial cell banks, and no fresh human tissues or patient samples were collected. Ethical approval was therefore not required for this in vitro cell-based study.

Obtaining and predicting shikonin targets
The keyword "shikonin" was searched in the PubChem database. The SDF structure of shikonin was downloaded, and its simplified molecular-input line-entry system (SMILES) notation was obtained. According to the requirements of the respective target databases, potential shikonin targets were predicted and collected from the PharmMapper, SwissTargetPrediction, HERB, and SEA Search Server databases on March 3, 2026^10,11,12.

Obtaining renal cell carcinoma targets
Using renal cell carcinoma as the keyword, human-derived disease targets were retrieved from the Therapeutic Target Database (TTD), OMIM, and GeneCards, with species limited to Homo sapiens and Relevance Score > 50. A total of 1148 targets were initially obtained, and after summarizing and deduplicating, 1087 unique renal cell carcinoma-related targets were finally acquired^13,14,15.

Identification of intersection targets of shikonin for the treatment of renal cell carcinoma
The predicted gene targets of shikonin and renal cell carcinoma targets were input into Venny 2.1, an online analysis tool. Overlapping targets were identified by intersection analysis and visualized in a Venn diagram.

Establishment of the component-target-disease network
Using shikonin as the core research starting point, we integrated physiological association information between these core targets and RCC, and constructed a molecular interaction network of the "component-target-disease" ternary system using Cytoscape visualization software.

Construction of protein interaction network and core target acquisition of shikonin in the treatment of renal cell carcinoma
Protein interaction database STRING is the core tool to construct protein-protein interaction (PPI) networks, which enables mining and visualization of the interaction relationship between proteins. In this study, the overlapping targets between shikonin and renal cell carcinoma were submitted to the STRING database, with the species set to Homo sapiens for further retrieval. Click the Continue button for subsequent analysis. In the Settings module at the bottom of the analysis results page, the minimum required interaction score was set to a high confidence level of 0.900. At the same time, unrelated proteins and disconnected nodes in the network were hidden to obtain the optimized PPI network mutual mapping. The data were stored in TSV format and imported into Cytoscape software. Corresponding parameters were adjusted so that node size and color shade reflected the corresponding values, and the edge thickness reflected the binding rate score. The PPI network map was established^16,17.

Based on the above PPI network, core targets were accurately screened. The cytohubba plug-in was opened in the Apps module of Cytoscape, and the Calculate button was clicked to start the calculation program. The MCC algorithm was selected from the Top 6 node(s) ranked by option, and the Submit button was clicked to complete the calculation. The top six core targets were screened out for beautification and data download.

GO enrichment analysis and KEGG pathway analysis of shikonin in the treatment of renal cell carcinoma
The Metascape database was used for GO enrichment analysis. In this study, the shikonin-targeted intersections with renal cell carcinoma were imported into the database. First, the selected species was Homo sapiens (human), and then the Custom Analysis mode was selected. Enrichment Analysis was performed for the three dimensions of GO functional enrichment: molecular functions, cellular components, and biological processes. The p-value cut off was set to 0.01, and the enrichment analysis button was clicked to perform analysis using the microcredit platform to construct a GO enrichment analysis visualization of shikonin in the treatment of renal cell carcinoma^18,19.

KOBAS 3.0 database was employed to perform KEGG signaling pathway enrichment analysis on the overlapping targets between shikonin and renal cell carcinoma, and this database can realize pathway annotation and enrichment statistical analysis of genes. The screened intersection targets were imported into the KOBAS 3.0 database; the source of the selected species was H. sapiens (human), the screening type was limited to KEGG signaling pathway, and the p-value threshold for enrichment analysis was set to <0.01. After analysis, the enrichment results of the KEGG pathway were obtained. Based on the enrichment analysis results, the p-value was used as the basis for sorting, and the top 10 KEGG signal pathways were screened out. The visualization processing was completed with the help of a microsignaling platform, and finally, the KEGG enrichment Sankey diagram bubble chart was drawn and saved^20,21.

Molecular docking verification
The three-dimensional (3D) structure of shikonin was retrieved from the PubChem database, and Open Babel software was applied to transform the 3D structure into standard PDB format. The three-dimensional protein structures of the core targets identified by network pharmacology analysis were downloaded in PDB format from the RCSB protein database and AlphaFold database. The crystal structures of the target proteins included PIK3CA (PDB ID: 9B4T), PIK3CD (PDB ID: 8BCY), PIK3R1 (PDB ID: 7CIO), PTPN11 (PDB ID: 9R16), SRC (PDB ID: 9NS1), PIK3CB (AlphaFold-predicted structure, AF-Q8BTI9-F1-model_v6). Water molecules and heteroatoms were removed, polar hydrogens were added, and Gasteiger charges were assigned using AutoDock Tools. The AutoDock Tools software was used to select the docking box of an appropriate size to cover the entire target protein for molecular docking. Finally, visualization of target-active component pairs with optimal binding energy was obtained. PyMol software was used to analyze the binding sites, the type and distance of binding interaction, and to generate a three-dimensional conformational visualization image^22,23.

Cytotoxicity test
Caki-1 cells were inoculated into 96-well culture plates at a density of 5,000 cells per well and treated with shikonin at concentrations of 12.5, 25, 50, and 100 µmol/L for 12 h and 24 h, respectively. For 786-O cells, only24 h treatment was performed with the same concentration gradient. The cytotoxic effect of shikonin on both cell lines was evaluated using the MTT assay. All experiments were performed in biological triplicates with three technical replicates for each condition. Results are expressed as IC₅₀ values. Although the overall inhibitory trend was consistent between the two cell lines, shikonin exhibited stronger cytotoxicity towards Caki-1 cells. Therefore, Caki-1 cells were selected for subsequent functional experiments.

Effect of shikonin on Caki-1 cell migration as determined by scratch-wound assay
Caki-1 cells and 786-O cells in logarithmic phase were seeded into 6-well plates. Parallel scratch lines were created at the bottom of each well using a sterile 200 µL pipette tip. After overnight attachment, PBS was utilized to wash the cultured cells. Cells were treated with different concentrations of shikonin (12.5, 25, 50, and 100 µmol/L) for 12 h. After 12 h, the change in the scratch area at the same site was observed under a microscope.

The formula for calculating the effect of shikonin on the growth activity of Caki-1 cells was as follows: mobility = [(0 h scratch area − 12 h scratch area)/0 h scratch area] × 100%.

Determination of Caspase-3 and Caspase- 8 activities
Caki-1 cells at the logarithmic growth phase were harvested, and the cell concentration was adjusted to 5 × 10⁶ cells/mL, followed by incubation for 12 h. After Caki-1 cells were treated with different concentrations of shikonin (12.5, 25, 50, and 100 µmol/L) and incubated for 24 h, cell detection was performed as described in the Caspase test kit.

Effects of shikonin on the apoptosis rate of Caki-1 cells
Logarithmically growing Caki-1 cells were seeded into 6-well culture plates. Cells were treated with shikonin at concentrations of 12.5, 25, 50, and 100 µmol/L for 24 h in the absence of a control group. Cells were gently resuspended in PBS, collected, and counted. A total of 1 × 10⁵ resuspended cells were centrifuged at 300 × g for 5 min. The supernatant was discarded, and the cell pellet was gently resuspended in 195 µL of Annexin V-FITC binding buffer. Then, 5 µL of Annexin V-FITC was added, followed by 10 µL of propidium iodide staining solution, both with gentle mixing. Cells were incubated at room temperature (20–25 °C) in the dark for 15 min prior to analysis.

Cell signaling pathway analysis
The Caki-1 cells were treated with 12.5, 25, 50, and 100 µmol/L shikonin. After 24 h, the cells were harvested in cold PBS, lysed with cell lysis buffer containing protease inhibitor, and the protein concentration was measured. Protein was isolated on a 10% Sodium dodecyl sulfate (SDS)-polyacrylamide gel. After transfer to PVDF membranes, the membranes were blocked with 5% BSA for 3 h, then incubated overnight at 4 °C with a different primary antibody. Subsequently, the membranes were incubated with the secondary antibody at a dilution of 1:3000. Protein bands were detected with the ECL detection system and analyzed quantitatively using ImageJ software.

Statistical analysis
Experimental results were presented as the mean ± standard deviation (x ± s, n = 3). Before parametric statistical analysis, normality was verified using the Shapiro-Wilk test, and homogeneity of variance was assessed by Levene’s test. One‑way analysis of variance (ANOVA) combined with Tukey's post hoc test was performed using SPSS software. Statistical graphs were constructed with GraphPad Prism software. The independent‑samples t-test was used for pairwise comparisons between groups, and p < 0.05 indicated statistically significant differences.

Results

Obtaining and predicting shikonin targets
Using the four aforementioned databases, potential shikonin targets were predicted, and 426 targets were identified after integration and summarization. The summarized targets were deduplicated, resulting in 374 targets, as shown in Figure 2A.

Obtaining renal cell carcinoma targets
A total of 1,147 targets were obtained after summarizing the data related to the three databases, and...

Discussion

PIK3CA is formally known as phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit alpha isoform. PIK3CA, a central component of the PI3K/AKT/mTOR pathway, plays an essential role in the occurrence, development, and treatment of renal cell carcinoma (RCC). Studies have shown that PIK3CA is a high-frequency mutant gene of this subtype (6/57, 11%), and its mutation is directly related to the abnormal activation of the PI3K/AKT pathway. Moreover, such patients respond well to the therapeutic regimen of an immune c...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
786-O	American Type Culture Collection (ATCC)	RRID: CVCL_1051 (ATCC CRL-1932)	Human clear cell renal cell carcinoma cell line
AlphaFold database	European Molecular Biology Laboratory (EMBL) - EBI	AlphaFold Protein Structure Database	Protein structure prediction database (AlphaFold Protein Structure Database)
AutoDock Tools software	Scripps Research Institute	1.5.7	Molecular docking software for ligand-protein binding prediction
Caki-1	American Type Culture Collection (ATCC)	RRID: CVCL_0234  (ATCC HTB-46)	Human clear cell renal cell carcinoma cell line
Caki-1 cells	Mudanjiang Normal University	HTB-46 (RRID: CVCL_0234)	Human clear cell renal cell carcinoma cell line
Caspase 3 and caspase 8 detection kits	Credit Suisse Biotechnology Co Ltd (QUANZHOU, China)	Caspase-3 Activity Assay Kit, Cat. No. RX302848M, Caspase-8 Activity Assay Kit, Cat. No. RX302849M	Used for detecting caspase activity to assess cell apoptosis
Cytoscape software	Cytoscape Consortium	3.9.1	Used for biological network construction and visualization
DMEM medium	Hyclone	SH30022.01	Cell culture medium for Caki-1 and 786-O cell culture
Fetal bovine serum	Gbico	10099-141	Cell culture supplement for Caki-1 and 786-O cell growth
GeneCards	Weizmann Institute of Science	https://www.genecards.org/	Comprehensive human gene database for target gene annotation
GraphPad Prism	GraphPad, San Diego, CA, USA	version 8.3.0	For statistical graphs construction
HERB database	Chinese Academy of Sciences	http://herb.ac.cn	Traditional Chinese medicine component-target database
ImageJ software	National Institutes of Health (NIH)	Fiji	Image analysis software for cell morphology and Western blot band quantification
inverted microscope	Olympus, IX73, Japan	100× magnification.	Used for observing cellular morphology changes after shikonin treatment
KOBAS 3.0 database	Peking University	http://bioinfo.org/kobas/	Gene ontology and pathway enrichment analysis tool
Metascape database	Memorial Sloan Kettering Cancer Center	https://metascape.org/gp/index.html)	Gene function enrichment and network analysis tool
Microcredit platform	National Center for Bioinformation (China)	http://www.bioinformatics.com.cn/	Online bioinformatics analysis platform for data processing
MTT assay kit	Solarbio Science & Technology Co., Ltd. (Beijing, China)	Cat. No. M8180	Used for detecting cell viability (antiproliferation assay)
OMIM	Johns Hopkins University School of Medicine	https://www.omim.org/	Online Mendelian Inheritance in Man (human genetic disease database)
PharmMapper database	East China University of Science and Technology	https://www.lilab-ecust.cn/pharmmapper/	Small molecule drug target prediction database
PIK3CA, PIK3CB, PIK3CD, and PTPN11	Santa Cruz Biotechnology, Inc. (Dallas, TX, USA)	Santa Cruz Biotechnology, Inc. (Dallas, TX, USA)	Primary antibodies for Western blot analysis of target proteins
PyMol software	Schrödinger, Inc.	2.2.0	Molecular visualization software for protein structure analysis
RCSB protein database	RCSB Protein Data Bank (Rutgers University)	https://www.rcsb.org/	Protein structure database (PDB) for molecular modeling
SEA Search Server database	University of California, San Francisco	https://sea.bkslab.org	Similarity Ensemble Approach for target prediction
Shikonin (purity 98%	leaf organisms originating in China	Y0001439	Small molecule compound used for renal cell carcinoma (RCC) treatment in this study
SPSS	IBM, Armonk, NY, USA	version 20.0	For statistical analysis
STRING	European Molecular Biology Laboratory (EMBL)	https://stringdb.org/	Protein-protein interaction network database
SwissTargetPrediction database	Swiss Institute of Bioinformatics	https://www.swisstargetprediction.ch/	Small molecule target prediction tool for drug discovery
Therapeutic Target Database	Shanghai Institute of Materia Medica	https://ttd.idrbrab.cn	Database of therapeutic targets and drugs
Venny 2.1	Spanish National Center for Biotechnology (CNB-CSIC)	https://bioinfogp.cnb.csic.es/tools/venny/index.html)	Venn diagram tool for data intersection analysis