Integrated multi-omics demonstrates enhanced antitumor efficacy of donafenib combined with FADS2 inhibition in hepatocellular carcinoma

Highlights

• •Donafenib proves more effective than sorafenib, highlighting its importance in HCC pharmacotherapy.
• •Downregulation of FADS2 at protein and mRNA levels after donafenib treatment by omics analysis.
• •FADS2 blockade inhibits the malignant biological behavior of HCC.
• •Combination of donafenib and FADS2 inhibitor demonstrates synergistic antitumor action in HCC.

Abstract

Pharmacotherapy is crucial for advanced hepatocellular carcinoma (HCC). The multi-kinase inhibitor donafenib offers superior survival benefits over sorafenib. Donafenib has first-line status, but there is limited research for combination therapies with this anticancer agent. This study aimed to delineate donafenib's antitumor effects, including transcriptomics and proteomics to characterize gene expression changes in donafenib-treated HCC cell lines. In vitro and in vivo tumorigenicity studies were conducted to evaluate the combined antitumor effects of donafenib. Proteomic and transcriptomic analyses identified that donafenib downregulated fatty acid desaturase 2 (FADS2) at the protein and mRNA levels. In vitro and in vivo assays revealed an inhibitory effect of FADS2 blockade on HCC cell malignancy. The combination of donafenib and the FADS2 inhibitor sc-26,196 produced synergistic antitumor action, enhancing therapeutic efficacy in HCC cell lines and xenografted tumors in nude mice. These findings highlight the potential of FADS2 as a biomarker for HCC and show a promising combinatorial therapy for its treatment. Thus, we provide a theoretical basis for translating laboratory research into clinical applications.

Introduction

Hepatocellular carcinoma (HCC) is the most prevalent pathological type of primary liver cancer globally, representing approximately 75 %–85 % of cases[1]. Owing to its insidious onset, most patients with HCC are diagnosed at an intermediate or advanced stage, missing the optimal timing for surgical resection. Therefore, pharmacotherapy is crucial for advanced HCC[2].

Donafenib, a multi-kinase inhibitor, is the first targeted therapy to demonstrate superior survival benefits and safety compared with sorafenib in a large-scale Phase III clinical trial with a single-agent head-to-head comparison. It has been adopted as a first-line treatment for patients with advanced HCC[3]. However, the extent to which donafenib prolongs patient survival is limited, restricting its clinical application[4]. Given that cancer pathogenesis involves complex pathways and multiple targets, combining drugs targeting different molecules may offer additional benefits for tumor treatment.

The "fatty acid desaturase 2″ (FADS2) gene, also known as the delta-6 fatty acid desaturase gene, primarily regulates polyunsaturated fatty acid synthesis via the introduction of a double bond into their hydrocarbon chain, serving as a crucial rate-limiting enzyme[5]. FADS2 is highly expressed in various cancerous tissues, including breast cancer, lung cancer, esophageal cancer, colorectal cancer, and ovarian cancer, suggesting that it may be a potential therapeutic target for cancer[[6], [7], [8], [9], [10]].

Donafenib lacks basic research, with no studies reporting a pharmacological link to FADS2. Multidrug combinations are a common mode of oncology treatment in clinical practice, where drug-drug interactions result in altered drug effects, and combinations have the potential to result in increased efficacy and decreased toxicity. Therefore, this study investigated whether FADS2 blockade enhances the anti-tumor effect of donafenib in HCC. The findings herein may provide a preclinical strategy for the treatment of patients with HCC using novel drug combinations.

Materials and methods

Cell culture

HuH-7 and HCCLM3 cells were cultured in DMEM (GP22080081336, Wuhan Xavier Biotechnology Co. Ltd) supplemented with 10 % fetal bovine serum (FBS, Tianjin Anuo Ruikang Biotechnology Co. Ltd), 100 U/mL penicillin and 100 U/mL streptomycin (Shanghai Ebixin Biotechnology Co., China), and routinely incubated at 37 °C in 5 % CO₂.

Construction of xenograft tumor model in nude mice

We constructed a subcutaneous tumor formation model using 4-week-old BALB/c male nude mice (weighing 14–16 g) purchased from Beijing Huafukang Company (Laboratory Animal Production License No SCXK, Beijing; 2019–0008; Laboratory Animal Quality Certificate No 110,322,231,102,939,481). The mice were acclimatized in a barrier environment for 1 week before the experiment. A suitable HuH-7 cell suspension containing 5 × 10⁶ in 0.1 mL was prepared. Each nude mouse was subcutaneously inoculated with this cell suspension in the right axilla. Tumor growth was observed every 2 d until the transplanted tumor volume reached 100 mm³. The nude mice were randomly grouped for drug intervention using the random number method. Tumor diameter was measured using Vernier calipers. After this period, the mice were euthanized using the cervical dislocation method, and the subcutaneously grafted tumors were removed. Thereafter, the transplant tumor volume was calculated using the formula V (mm³) = ab²/2, where ‘V’ represents the tumor volume, 'a' represents the longest diameter of the tumor, and 'b' represents the shortest diameter of the tumor. At the end of the experiment, the Tumor Weight Inhibition (TWI) was calculated using the formula TWI% = (1 − T: C) × 100 %, where 'T' represents the average subcutaneous tumor volume of treated mice in the administration group and 'C' represents the average subcutaneous tumor volume of control mice in the control group. This study evaluated the effects of drug combinations using the Kim Jung Mean Q method. The formula Q = (EA + B)/[EA + (1 − EA) × EB] was used, where EA and EB represent the rate of tumor inhibition when the two single drugs are applied, and EA + B represents the rate of tumor inhibition after combining the two drugs. A Q value >1.15 indicates a synergistic effect, a value between 1.15 and 0.85 indicates an additive effect and a value <0.85 indicates an antagonistic effect.

Proteome data analysis

Eight groups of samples were used in this study, including four cases each in the control and experimental groups. Three hundred microliters of 8 M urea were added to the sample, and a protease inhibitor was added at 10 % of the solution. Following centrifugation at 14,100 × g for 20 min, the supernatant was collected. The protein concentration was measured using the Bradford method, and the remaining sample was frozen at −80 °C. Protein (50 µg) was extracted from each sample for reduction treatment. Raw data were further processed using the bioinformatics analysis platform, BMKCloud (www.biocloud.net).

Transcriptome data analysis and quality control

The initial data were processed using the BMK Cloud. Raw data in FASTQ format were initially processed using an internal Perl script. All subsequent analyses were based on clean, high-quality data. Clean data were aligned to the reference genome sequence. Based on the reference genome, only the sequences with a perfect match or one mismatch were subjected to further analysis and annotation. Alignment with the reference genome was performed using Hisat2 software. The StringTie assembly method, based on reference annotation of the transcriptome (RABT), was employed to identify known transcripts and predict new transcripts from the alignment results of Hisat2[11].

Gene function annotation and differential expression analysis

Gene function annotation and differential expression analyses were performed. Differential expression analysis between the two groups was performed using DESeq2. DESeq2 uses a model based on a negative binomial distribution to identify differential gene expression data[12]. The Benjamini and Hochberg method was used to control the FDR (False Discovery Rate) and obtain adjusted p-values. Genes with adjusted p-values <0.01 and FC ≥2, as determined by DESeq2 analysis, were classified as differentially expressed.

GO and KEGG enrichment analysis

DEPs (different proteins) and DEGs (different genes) were subjected to GO (Gene Ontology) enrichment analysis using the ClusterProfiler package. This package is based on the Wallenius non-central hypergeometric distribution, which accounts for gene-length bias. KEGG (Kyoto Encyclopedia of Genes and Genomes), a database resource, was used to gain insights into the high-level functions of biological systems using molecular-level information. The enrichment of DEGs in KEGG pathways was analyzed using the KOBAS database and ClusterProfiler software[13].

Clinical relevance analysis of FADS2 in HCC

GEPIA (Gene Expression Profiling Interactive Analysis, http://gepia.cancer-pku.cn/) is an interactive gene expression analysis platform for analyzing cancer and normal samples, integrating resources from public databases, such as The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression[14]. The KM Plotter (kmplot, https://kmplot.com/analysis/) is a prognostic analysis website for evaluating the relationship between specific gene expression and the survival rate of patients with cancer[15]. The kmplot analysis was used to evaluate the relationship between FADS2 expression and the overall survival rate of patients with HCC. HCC data from TCGA were used for FADS2 GSEA (Gene Set Enrichment Analysis) and clinical correlation analyses. The "ggplot2″ and "limma" R packages were used to visualize the data. The Human Protein Atlas (HPA) database (https://www.proteinatlas.org/) provides a variety of tissues and cells of human protein distribution information, checking each protein in various human normal tissues, tumor tissue, cell lines, and the distribution and expression of blood cells. In this study, we investigated the immunohistochemical expression of FADS2 in HCC patients using the HPA database.

Western blot analysis

The antibodies used in this experiment were as follows: Anti-FADS2 antibody (PS01915), Sodium Potassium ATPase antibody (T55159F), B-Raf antibody (T55319), Raf-1 antibody (T55953), PCNA antibody (TA0239), MCM2 antibody (T55200), β-actin antibody (P30002), Goat Anti-Mouse IgG HRP (M21001) and Goat Anti-Rabbit IgG-HRP (M21002). The cells were harvested and lysed in ice-cold RIPA lysis buffer (1 × Tris-buffered saline, 1 % Nonidet P-40, 0.5 % sodium deoxycholate, and 0.1 % sodium dodecyl sulfate [SDS]). Proteins were transferred to a polyvinylidene difluoride membrane with a pore size of 0.22 µm using a constant current of 300 mA. Samples were blocked with TBST (Tris–HCl, NaCl, and Tween-20) buffer containing 5 % skim milk (w/v) for 2 h to prevent non-specific binding. The protein bands were visualized using an enhanced chemiluminescence assay (Clarity Western ECL Substrate; Bio-Rad, Hercules, CA, USA).

Cell proliferation assay

HuH-7 and HCCLM3 cells in the logarithmic growth phase were seeded in 96-well plates, and 4 × 10³ cells were seeded in 100 µL complete medium in each well. The cells were then cultured overnight at 37 °C in an incubator with 5 % CO₂, and the next operation was performed after cell attachment. After 48 h of drug treatment, 10 µL of CCK-8 reagent was added to each well, and the absorbance value was detected using a multifunctional microplate reader (wavelength set at 450 nm) after 2 h of incubation.

Combination index (CI) analysis

The combination drugs were administered at a fixed ratio to ensure that the concentration of each drug remained constant within each group. For the HuH-7 cell line, concentrations of donafenib ranged from 1.6 to 24 µmol/L, whereas sc-26,196 concentrations were set at 4 to 60 µmol/L. For HCCLM3 cells, donafenib concentrations ranged from 2 to 30 µmol/L, whereas sc-26,196 concentrations ranged from 4 to 60 µmol/L. Based on the dose-effect relationship between single drug use and combination treatment, the synergistic or antagonistic effects of donafenib and sc-26,196 on tumor growth in HuH-7 and HCCLM3 cells were further evaluated using CompuSyn software and Chou-Talalay method[16]. The combination index (CI) was used to assess the effect of multiple drugs when used together; a CI value <1 indicates synergy, with lower values indicating greater synergy; CI >1 indicates antagonism, with higher values indicating greater antagonism, while a CI = 1 indicates additivity.

Cell apoptosis experiment

The experimental group was divided into donafenib alone, sc-26,196 alone and combination group. Each group was set up in three wells. For HuH-7 cell line: donafenib 7.5 μmol/L, sc-26,196 16.5 μmol/L, donafenib 7.5 μmol/L+sc-26,196 16.5 μmol/L. HCCLM3 cell line: donafenib 9.5 μmol/L, sc-26,196 20.5 μmol/L, donafenib 9.5 μmol/L+sc-26,196 20.5 μmol/L. Cells were treated according to the AnnexinV-Fitc/PI kit instructions and detected using a CytoFlex flow cytometer. Data were analyzed using FlowJo software.

Cell cycle experiment

The experimental groups included donafenib alone, sc-26,196 alone, and sc-26,196 combined with donafenib. Each group had three replicates, with concentrations identical to those used in apoptosis experiments. The treated cells were processed according to the instructions provided by the Cell Cycle Kit and subsequently analyzed using a CytoFlex flow cytometer. Data were analyzed using FlowJo software.

Cell scratch experiment

Culture plates were marked with a horizontal line on the back using a marker pen before cell seeding. HuH-7 and HCCLM3 cells in the logarithmic growth phase were seeded into 6-well plates. When the cells had filled the bottom of the dish and adhered well, the cell scratch was made with a 200ul gun tip perpendicular to the well plate, ensuring that the width of each scratch was as uniform as possible. After aspirating the medium and gently washing the cells three times with sterilized PBS, serum-free medium containing drugs was added and photographed using a Dmi8 inverted microscope. The control group and the experimental group were set up the same as the cell apoptosis and cycle experiment.

Statistical analyses

SPSS 26.0 software was used to analyze the experimental data, and the results of measurement data were expressed as mean ± standard deviation. T-test and analysis of variance were used for comparisons between the two groups, and p < 0.05 was considered statistically significant.

Results

Tumor inhibition of donafenib in a nude mouse model

To better understand the biological functions of donafenib in vivo, we established a BALB/c nude mouse xenograft model. Fig. 1A illustrates subcutaneous tumor formation in nude mice. Following dosing, the tumor volume was measured every 2 d Mice were euthanized after 16 d, and tumor tissues were retained for comparison (Fig. 1B). Quantified tumor volumes in Fig. 1C reveal a significant difference between the experimental and control groups on day 16 (donafenib group/control group = 0.41, p<0.001), indicating slowed tumor growth following donafenib intervention. Fig. 1D shows that the tumor mass of the experimental group was 65.31 % of the control group (p<0.001), indicating a significant reduction in tumor mass. Fig. 1E shows that there was no statistically significant difference in body weight between the two groups on day 16.

Fig. 1.

Antitumor effect of donafenib on HCC in vivo. A. Nude mouse subcutaneous tumor formation model. B. Tumor tissue images of mice in the control and donafenib groups. C. Quantitative graphs depicting tumor volume in the control group and donafenib groups. D. Quantitative graphs illustrating tumor mass in the control and donafenib groups. E. Quantitative graphs illustrating mice mass in the control and donafenib groups.

Proteomic analysis of HCC cells before and after donafenib treatment

Upregulated proteins were identified with a fold change (FC) ≥2 and p ≤ 0.01, while downregulated proteins had FC ≤1/2 and p ≤ 0.01. Under these conditions, 103 DEPs were identified: 42 upregulated and 61 downregulated proteins (Fig. 2A). Fig. 2B shows principal component analysis (PCA) results, revealing significant differences between the two groups with minimal variation within each group. Fig. 2C presents volcano and heat maps of DEPs, indicating consistent expression within the same sample type and marked differences between various sample types, suggesting high representativeness and biological reproducibility. GO functional annotation and enrichment analysis showed enrichment in biological processes (Fig. 2D) such as anatomical structural development, cell adhesion, and lipid metabolic processes. Regarding cellular structures (Fig. 2E), DEPs were primarily involved in multicellular organism development, extracellular regions, and collagen trimers. For molecular functions (Fig. 2F), these proteins were significantly enriched in metal ions, carbohydrate derivatives, and heme binding. KEGG enrichment analysis demonstrated signaling pathways involving DEPs, primarily the PI3K-Akt signaling pathway, PPAR signaling pathway, and proteoglycans in cancer (Fig. 2G). Notably, the FADS2 had the highest absolute value of log2FC among the downregulated DEPs (Table 1).

Fig. 2.

Proteomics analysis of HCC tissues before and after donafenib treatment. A. Number of differentially expressed proteins. B. Principal Component Analysis (PCA) performed on all samples. C. Volcano plot and expression heat map of differentially expressed proteins. d-F. Functional annotation and enrichment analysis using Gene Ontology (GO) conducted for the identified differentially expressed proteins. G. KEGG pathway enrichment analysis conducted on differentially expressed proteins.

Table 1.

Top 10 upregulated and downregulated proteins with the largest fold change between the control and donafenib groups.

#ID protein	Gene symbol	Log2FC	P value	Regulated
ENSG00000134824	FADS2	−5.100477422	0.000265344	down
ENSG00000110651	CD81	−3.061854826	0.000429058	down
ENSG00000108106	UBE2S	−3.020053804	0.008012755	down
ENSG00000159461	AMFR	−2.543267701	4.06E-05	down
ENSG00000171557	FGG	−2.449830024	3.79E-05	down
ENSG00000142192	APP	−2.325856241	0.009246851	down
ENSG00000134962	KLB	−2.171662303	0.001272968	down
ENSG00000196411	EPHB4	−2.162588791	0.002064807	down
ENSG00000139329	LUM	−2.143300435	0.003518893	down
ENSG00000100292	HMOX1	−2.089861635	0.000782455	down
ENSG00000165806	CASP7	3.271745582	0.000461476	up
ENSG00000111011	RSRC2	2.338214017	0.006268646	up
ENSG00000119125	GDA	2.290756083	0.00256821	up
ENSG00000120896	SORBS3	2.263416532	0.000269974	up
ENSG00000149269	PAK1	2.124767392	0.002750141	up
ENSG00000135521	LTV1	2.057776126	0.001596714	up
ENSG00000149503	INCENP	1.983364746	0.005660194	up
ENSG00000083937	CHMP2B	1.953527838	0.002349193	up
ENSG00000184863	RBM33	1.937826332	0.002439264	up
ENSG00000163840	DTX3L	1.832859605	0.008625644	up

FADS2 protein expression level after donafenib treatment

On the basis of the fold order results, FADS2 was the most significant downregulated gene. Western blotting experiments confirmed these results, showing decreased FADS2 expression in HuH-7 and HCCLM3 cells after donafenib treatment (Figs. 3A and 3D). This was consistent with the proteomic results, with downregulation of FADS2 being dose-dependent. In addition, the protein expression levels of B-Raf (Figs. 3B and 3E) and Raf-1 (Figs. 3C and 3F), which are common targets of donafenib, were also assessed in the HuH-7 and HCCLM3 cell lines. The expression of both proteins was differentially reduced compared to the control group.

Fig. 3.

Verification of FADS2 protein level changes after donafenib intervention in HuH-7 and HCCLM3 cell lines. A. Protein expression analysis of FADS2 in HuH-7. B. Protein expression analysis of B-Raf in HuH-7. C. Protein expression analysis of Raf-1 in HuH-7. D. Protein expression analysis of FADS2 in HCCLM3. E. Protein expression analysis of B-Raf in HCCLM3. F. Protein expression analysis of Raf-1 in HCCLM3.

Transcriptome analysis of HCC cells before and after donafenib treatment

In the DEGs screening, 3606 DEGs were identified using the criteria of FC≥ 2 and FDR< 0.01, with 1266 upregulated and 2340 downregulated genes (Fig. 4A). The PCA results (Fig. 4B) showed significant differences between the sample groups. The volcano plot and expression heat map of DEGs (Fig. 4C) showed consistent expression patterns within duplicate sample sets and substantial differences between different sample sets, highlighting the representability and biological reproducibility of DEGs. GO functional annotation and enrichment analysis revealed DEGs enriched in biological processes such as cell differentiation regulation, transcriptional RNA II promoter regulation and DNA replication (Fig. 4D). For cellular components, DEGs were involved in the host cell nucleus, extracellular matrix and chromosomes (Fig. 4E). Molecular function analysis revealed enrichment in protein dimerization, calcium ion binding, and growth factor activity (Fig. 4F). KEGG pathway analysis (Figs. 4G-I) revealed enrichment in pathways such as cytokine-cytokine receptor interaction, PI3K-Akt signaling, and cell cycle pathways. Focusing on FADS2 mRNA expression, we observed reduced FADS2 expression in donafenib-treated HCC cells (FDR=4.92E-69, log2FC=−2.33), consistent with the proteomic analysis, suggesting that donafenib potentially enhances tumor cell inhibition by suppressing FADS2 expression.

Fig. 4.

Transcriptome analysis of HCC cells before and after donafenib treatment. A. Number of differentially expressed genes. B. PCA analysis of all samples. C. Volcano plot and expression heatmap of differentially expressed genes (DEGs). d-F. GO functional annotation and enrichment analysis of DEGs. G. KEGG pathway type classification. H. KEGG bubble diagram of DEGs. I. KEGG enrichment network of DEGs.

Functional prediction and clinical relevance analysis of FADS2

Given the concordance between proteomic and transcriptomic results, we studied FADS2 expression and prognosis in HCC tissues, along with its functional mechanisms. Gene Expression Profiling Interactive Analysis (GEPIA) data showed FADS2 expression in pan-cancer (Fig. 5A) and HCC tissues (Fig. 5B). Using the KM Plotter, we explored the correlation between FADS2 expression and HCC patient prognosis, revealing a potential association of high FADS2 expression with poor prognosis in HCC patients (Fig. 5C). Fig. 5D displays proteins that may interact with FADS2. GSEA indicated that FADS2 is mainly involved in cellular responses to copper ions, inorganic compound detoxification, positive regulation of the G1/S transition of the mitotic cell cycle, regulation of mitochondrial endopeptidase activity, and metal ion stress response (Fig. 5E). KEGG enrichment suggested a potential association of FADS2 with ribosome pathways (Fig. 5F). Fig. 5G presents a heatmap of the clinical correlation analysis between FADS2 and age, sex, grade, stage, and TNM staging. Immunohistochemical analysis of HPA showed strong staining intensity and was detected in >75 % of the cell density. This protein is mainly localized in the cytoplasm and membrane, suggesting that it may play an important role in the biological process of HCC (Fig. 5H).·

Fig. 5.

Functional prediction and clinical relevance analysis of FADS2. A.FADS2 expression in pan-cancer in GEPIA. B.FADS2 expression in HCC and non-tumor tissues in GEPIA. C. Survival analysis of FADS2 in patients with HCC. D. Protein-protein interaction of FADS2. E. GO enrichment analysis of FADS2. F. KEGG enrichment analysis of FADS2. G. Heatmap depicting the clinical relevance analysis of FADS2. H. Immunohistochemical image of FADS2 in HCC obtained by Human Protein Atlas.

Effect of FADS2 blockade on malignant biological behavior of HCC cells

To confirm the effect of FADS2 on HCC cells, we treated two types of HuH-7 and HCCLM3 cell lines with the FADS2 inhibitor sc-26,196. The cck8 and apoptosis experiments showed that sc-26,196 inhibited HCC cell proliferation (Figs. 6A and 6B) and induced apoptosis (Figs. 6C and 6D). The cell cycle results exhibited a slight increase in the proportion of the G1/S phase in the sc-26,196-treated HCC cell lines, but no statistical difference was found (Figs. 6E and 6F). The cell scratch assays (Figs. 6G and 6H) demonstrated that sc-26,196 suppressed HCC cell invasion and migration. These results suggest that FADS2 blockade may have an inhibitory effect on the malignant behavior of HCC cells.

Fig. 6.

Effects of FADS2 blockade on malignant behavior of HCC cells. A. Effects of sc-26,196 on HuH-7 cell proliferation detected via the CCK8 assay. B. Effects of sc-26,196 on HCCLM3 cell proliferation detected via the CCK8 assay. C. Effects of sc-26,196 on HuH-7 cell apoptosis detected via flow cytometry. D. Effects of sc-26,196 on HCCLM3 cell apoptosis detected via flow cytometry. E. Effects of sc-26,196 on HuH-7 cell cycle detected via flow cytometry. F. Effects of sc-26,196 on HCCLM3 cell cycle detected via flow cytometry. G. Effects of sc-26,196 on HuH-7 cell motility detected using the scratch experiment. H. Effects of sc-26,196 on HCCLM3 cell motility detected using the scratch experiment.

Donafenib combined with FADS2 blockade exhibits synergistic anti-tumor effects in HCC cells

Based on the above findings, we proposed further investigating the synergistic anti-tumor effects of combining donafenib and FADS2 blockade in HCC cells. Cells were treated with doses of donafenib, sc-261,961, or their combination for 48 h. Both donafenib (Figs. 7A and 7B) and sc-26,196 (Figs. 7C and 7D) treatments reduced HuH-7 and HCCLM3 cell viability in a dose-dependent manner. Treatment with donafenib and sc-29,196 in a fixed ratio showed significant anti-tumor activity at lower concentrations (inhibition rate > 50 %) (Table 2). According to the Chou-Talalay method, a CI value <1 indicates a synergistic effect of the combined drugs. CI values were <1 in most concentration ranges when the dual drug combination was used in HuH-7 and HCCLM3 cell lines, suggesting better synergy with the combination treatment (Figs. 7E and 7F). Western blot results showed downregulation of FADS2 (Figs. 7G and 7H), as well as proliferation-related proteins PCNA (Figs. 7I and 7J) and MCM2 (Figs. 7K and 7L) by the combination treatment compared to single-agent treatment. Subsequent comparisons of single-drug and combination-group effects on HCC cell apoptosis, cell cycle, and metastasis revealed that the combined treatment had the most significant effect on inducing apoptosis (Figs. 7M and 7N), increasing the G1/S phase (Figs. 7O and 7P), and inhibiting cell motility (Figs. 7Q and 7R). These results suggest that FADS2 inhibition could enhance the anti-tumor effect of donafenib.

Fig. 7.

Synergistic anti-tumor effect of combining donafenib with sc-26,196 on HCC cells. A. Inhibition effect of donafenib on HuH-7 cell proliferation. B. Inhibition effect of donafenib on HCCLM3 cell proliferation. C. Inhibition effect of sc-26,196 on HuH-7 cell proliferation. D. Inhibition effect of sc-26,196 on HCCLM3 cell proliferation. E. Fa-CI plot of combination drugs in HuH-7 cells. F. Fa-CI plot of combination drugs in HCCLM3 cells. G. FADS2 protein expression of the single agent group and the combination group in HuH-7. H. FADS2 protein expression of the single agent group and the combination group in HCCLM3. I. PCNA protein expression of the single agent group and the combination group in HuH-7. J. PCNA protein expression of the single agent group and the combination group in HCCLM3. K. MCM2 protein expression of the single agent group and the combination group in HuH-7. L. MCM2 protein expression of the single agent group and the combination group in HCCLM3. M. Effects of the single agent group and the combination group on HuH-7 cell apoptosis. N. Effects of the single agent group and the combination group on HCCLM3 cell apoptosis. O. Effects of the single agent group and the combination group on HuH-7 cell cycle. P. Effects of the single agent group and the combination group on HCCLM3 cell cycle. Q. Effects of the single agent group and the combination group on HuH-7 cell motility. R. Effects of the single agent group and the combination group on HCCLM3 cell motility.

Table 2.

Dose effect in the combination group.

Group	HuH-7			HCCLM3
	donafenib (μmol/L)	sc-26,196 (μmol/L)	Proliferation inhibition rate	donafenib (μmol/L)	sc-26,196 (μmol/L)	Proliferation inhibition rate
1	1.6	4	(26.60±8.00)%	2	4	(29.59±5.59)%
2	4.8	12	(43.48±6.01)%	6	12	(52.11±4.25)%
3	8	20	(65.65±3.77)%	10	20	(81.42±7.39)%
4	11.2	28	(87.02±3.04)%	14	28	(79.57±3.12)%
5	14.4	36	(90.14±2.81)%	18	36	(87.17±2.48)%
6	17.6	44	(92.63±2.41)%	22	44	(92.23±0.73)%
7	20.8	52	(96.56±1.58)%	26	52	(96.71±0.40)%
8	24	60	(97.33±0.85)%	30	60	(97.89±0.68)%

Donafenib/sc-26,196 combination exhibits synergistic therapeutic effects in nude mouse model

The in vitro experiments demonstrated synergistic anti-tumor effects of donafenib combined with sc-26,196 in HCC cells. We established a nude mouse xenograft tumor model to confirm these effects in vivo. Fig. 8A presents images of nude mice with tumors following single drug and combination treatments. Tumor volume measurements every 2 days post-treatment showed that the combination-treated group had smaller tumor volumes than the single-drug group at all time points (Fig. 8B). The Kim Jung Mean Q value was 1.21, indicating that the combination most effectively inhibited tumor growth. After 2 weeks of treatment, tumor mass was significantly reduced in the combination group compared to monotherapy (Fig. 8C). The body weight of mice in the combination group did not change significantly compared to the control or single-drug groups (Fig. 8D).

Fig. 8.

Donafenib/sc-26,196 combination exhibited a synergistic therapeutic effect in nude mouse model. A. Images of nude mice with tumors in the single drug and combination treatment groups. B. Quantification of tumor volume in each group. C. Quantification of tumor tissue mass in each group. D. Quantification of mice mass in each group.

Discussion

The occurrence and development of HCC regulated by multiple systems and genes, necessitates simultaneous intervention in multiple signaling pathways or targets for comprehensive inhibition of tumor growth and metastasis, enhancing treatment effectiveness and sustainability[17,18],. Omics approaches have become a powerful strategy for systematically and accurately identifying molecules that regulate tumorigenesis and serve as potential targets for drug therapy[19,20],. This study employed proteomics as the primary high-throughput research method, complemented by transcriptomics for validation, and in vivo and in vitro experiments to characterize key functional molecules regulating HCC development and donafenib therapy.

Donafenib is a new generation of oral small-molecule kinase inhibitors used as first-line targeted therapy for HCC. Results from a phase III clinical trial (ZGDH3) comparing donafenib head-to-head with sorafenib in patients with advanced HCC showed that median overall survival (mOS) was extended by 6.1 months in the donafenib arm compared to the sorafenib arm, and that donafenib was associated with a significantly lower incidence of grade ≥3 adverse events than the sorafenib group[3]. Preclinical studies have confirmed that this drug can inhibit the activity of various receptor tyrosine kinases, such as vascular endothelial growth factor receptor (VEGFR) and platelet-derived growth factor receptor (PDGFR), and directly inhibit various Raf kinases and the downstream Raf/MEK/ERK signaling pathway, thereby exerting multiple inhibitory and multi-target blocking effects in anti-tumor therapy. It can inhibit the proliferation of tumor cells and the formation of tumor blood vessels, thus exerting multiple inhibition and multi-target blocking anti-tumor effects[21]. There is currently little research on donafenib and it is crucial to explore the complex molecular regulatory network of donafenib to elucidate the mechanism of action and explore more potential targets.

Recently, combination therapies have made significant progress in the treatment of HCC. For example, the combination of PD-1 inhibitor pembrolizumab with a multikinase inhibitor lenvatinib has shown greater anti-tumor activity than monotherapy[22]. Similarly, the combination of nivolumab with ipilimumab has shown the potential to improve response rates and overall survival in clinical trials[23]. In addition, combinations of targeted agents, such as atezolizumab in combination with bevacizumab, have shown superior efficacy to sorafenib alone in the IMbrave150 trial[24]. In chemotherapy and targeted therapies, strategies combining transarterial chemoembolization (TACE) with systemic therapy are being investigated to improve overall efficacy[25]. Individualized medicine approaches are helping to develop more precise treatment plans through genomic analysis. Comprehensive and personalized treatment strategies are expected to improve the prognosis and quality of life of HCC patients, and future clinical trials will further validate and optimize these treatment options.

Certain drugs have been shown to enhance therapeutic efficacy by blocking one or more targets in the signaling pathway, increasing the bioavailability of another drug, or stabilizing another drug in the system. For example, cyclosporine (Cyclosporine capsules) can increase the bioavailability of paclitaxel by maintaining P-glycoprotein expression, whereas the combination of flavonoid genistein with 5-fluorouracil applied to HT-29 colon cancer cells leads to a decrease in the survival signaling pathway Glut-1 and an increase in pro-apoptotic gene expression (P53 and P21)[26,27],. In HCC, sirtuin 1 and 2 inhibitors enhance the inhibitory effects of sorafenib on tumor cells[28,29],. Currently, anti-cancer combination therapy involves combining two or more innovative/improved drugs or a combination of several new drugs with a standard of care. In oncology, combination therapy aims to improve efficacy, guided by mechanistic research[30,31],. This study investigated the mechanism of action of donafenib in HCC, addressing a research gap domestically and internationally. We found that combining donafenib with a FADS2 inhibitor led to growth inhibition in HCC cell lines and tissues. This suggested potential dosage reduction to mitigate drug resistance and adverse reactions.

FADS2 has been proven to play a regulatory role in several types of tumors. In colorectal cancer, FADS2 promotes cell proliferation by increasing prostaglandin E2 metabolism[32]. Upregulated FADS2 activity is associated with melanoma and lung tumor growth, and inhibition of FADS2 can substantially reduce tumor growth[5]. FADS2 knockdown in lung cancer cells reduces the mRNA levels of iron death-associated factors, increasing the sensitivity of cells to iron death[33]. High FADS2 expression accelerates ovarian cancer cell lipid metabolism and tumor invasion. Pharmacological inhibition and knockdown of FADS2 delay ovarian cancer tumor growth, and tumor stem cell formation and decrease platinum resistance[10]. In addition, inhibition of FADS2 has the potential to suppress tumor growth, and molecular inhibitors may suppress tumor progression in combination with chemotherapeutic agents. FADS2 inhibitor sc-26,196 increases the radiosensitivity of glioblastoma cells[34], reduces colorectal cancer tumor size[35], and enhances carfukzinub efficacy in liposarcoma treatment[36].

Using transcriptomic and proteomic analysis of donafenib-treated HCC cell lines, we found that donafenib reduced FADS2 expression. FADS2 inhibition can suppress the proliferation and metastatic ability of HCC cells while increasing apoptosis by biological experiments, suggesting that targeting FADS2 may negatively regulate the malignant biological behavior of HCC. Notably, we found that donafenib and the FADS2 inhibitor sc-26,196 had synergistic anti-tumor effects, which were confirmed by both in vitro and in vivo experiments.

Our research had several limitations. Using both proteomics and transcriptomics, FADS2 expression was found to be upregulated in HCC cells. However, the cause or effect of tumorigenesis remains unclear regarding the increase in FADS2 levels, as its role in tumors is not fully understood. Further studies are required to confirm the specific mechanism of targeting FADS2 for tumor inhibition before applying the inhibition strategy clinically. In addition, toxicological experiments on sc-26,196 in vivo were not performed. As sc-26,196 is a FADS2 inhibitor used only for research purposes, the safety of its clinical application remains unclear. Further validation of FADS2 expression in patient samples and correlation analysis with clinical indicators will also provide reference points. In addition, histological techniques combined with cytological and molecular experiments can be used to screen for pathway changes after HCC intervention with donafenib combined with FADS2 inhibition.

Donafenib, as the first-line target therapy for HCC, has rarely been reported in domestic and international studies, and our research belongs to the source innovation. The object of research and any information uncovered will be of certain value to scientific research and clinical practice. This study suggests that combining donafenib and sc-26,196 may have a stronger therapeutic effect than a single agent. This indicated that combining multi-kinase inhibitors with FADS2 inhibition may be a promising therapeutic strategy for HCC.

Conclusion

This study elucidated the synergistic antitumor effect of the targeted drug donafenib in combination with a FADS2 inhibitor in HCC using multi-omics analysis and in vitro and in vivo biological experiments for closed-loop validation. We provide a theoretical foundation for understanding targeted drug mechanisms, individualized dosing, developing novel therapeutic strategies, and translating basic oncological research findings into clinical practice.

Data availability

Statistics supporting the findings of this study are all available from PubMed (https://pubmed.ncbi.nlm.nih.gov/).

Consent for publication

Written informed consent for publication was obtained from all participants.

Ethics approval and informed consent

Animal experiments were conducted in strict accordance with the principles approved by the Medical Ethics Committee of Peking University Cancer Hospital. All procedures conformed to the Declaration of Helsinki.

CRediT authorship contribution statement

Hui Li: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Yafeng Dai: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation. Di Wu: Writing – original draft, Investigation, Formal analysis, Data curation. Song Gao: Validation, Supervision, Resources, Conceptualization. Jianhai Guo: Software, Resources, Formal analysis, Data curation. Pengjun Zhang: Validation, Supervision, Software, Resources, Data curation, Conceptualization. Hui Chen: Funding acquisition, Supervision, Validation, Visualization. Fuxin Kou: Writing – original draft, Supervision, Project administration. Shaoxing Liu: Validation, Supervision, Formal analysis, Data curation. Aiwei Feng: Writing – original draft, Validation, Supervision. Baojiang Liu: Writing – original draft, Validation. Dongdong Hou: Writing – original draft. Xu Zhu: Writing – review & editing, Visualization, Validation, Resources, Funding acquisition, Data curation, Conceptualization.

Declaration of competing interest

The authors declare no conflicts of interest.