Naringenin: A potential therapeutic agent for modulating angiogenesis and immune response in hepatocellular carcinoma

Abstract

Naringenin (4,5,7-trihydroxyflavonoid) is a naturally occurring bioflavonoid found in citrus fruits, which plays an important role in metabolic syndrome, neurological disorders, and cardiovascular diseases. However, the pharmacological mechanism and biological function of naringenin on anti-angiogenesis and anti-tumor immunity have not yet been elucidated. Our study firstly demonstrates that naringenin inhibits the growth of hepatocellular carcinoma (HCC) cells both in vivo and in vitro. Naringenin diminishes the ability of HCC cells to induce tube formation and migration of human umbilical vein endothelial cells (HUVECs) and suppresses neovascularization in chicken chorioallantoic membrane (CAM) assays. Meanwhile, in vivo results demonstrate that naringenin can significantly upregulate level of CD8⁺ T cells, subsequently increasing the level of immune-related cytokines in the tumor immune microenvironment. Mechanistically, we found that naringenin facilitate the K48-linked ubiquitination and subsequent protein degradation of vascular endothelial growth factor A (VEGFA) and mesenchymal-epithelial transition factor (c-Met), which reduces the expression of programmed death ligand 1 (PD-L1). Importantly, combination therapy naringenin with PD-L1 antibody or bevacizumab provided better therapeutic effects in liver cancer. Our study reveals that naringenin can effectively inhibit angiogenesis and anti-tumor immunity in liver cancer by degradation of VEGFA and c-Met in a K48-linked ubiquitination manner. This work enlightens the potential effect of naringenin as a promising therapeutic strategy against anti-angiogenesis and anti-tumor immunity in HCC.

Graphical abstract

Highlights

• •Naringenin enhances anti-tumor immunity and inhibits angiogenesis in HCC both in vitro and in vivo.
• •Naringenin inhibits PD-L1 by facilitating the K48-linked ubiquitination and protein degradation of c-Met in HCC.
• •Naringenin downregulates VEGFA expression by enhancing K48-linked ubiquitination modification in HCC.
• •Naringenin may be potential immunotherapeutic and anti-angiogenesis agent against HCC.

1. Introduction

Liver cancer has become one of the leading causes of cancer-related death worldwide, with hepatocellular carcinoma (HCC) accounting for almost 90% of primary liver cancer [1]. At present, HCC treatment includes liver resection, liver transplantation, radiation therapy, and systemic therapy [2]. Due to the unclear early symptoms of liver cancer patients, most diagnoses occur at advanced stages, resulting in poor treatment efficacy and a high mortality rate [3]. Tumor immune escape and angiogenesis are key factors affecting the survival prognosis of HCC patients. Recently, immune checkpoint inhibitors (ICIs) and angiogenesis inhibitors have been developed to prolong survival time [4,5]. However, tumor cells often develop acquired resistance to these therapies, leading to relatively low response rates and the emergence of drug resistance, which greatly hinders their clinical application [6]. Therefore, elucidating the molecular mechanisms of immunotherapy resistance and angiogenesis in HCC as well as searching for new and effective small molecule drugs for the treatment of HCC remains a challenge.

Angiogenesis refers to the process of forming new blood vessels, which is regulated by multiple signaling pathways consisting of various biomolecules in the organism [7]. Among these, vascular endothelial growth factor A (VEGFA) is a potent stimulator of angiogenesis, and its receptors, VEGF receptor 1 (VEGFR1) and VEGFR2, are the key factors mediating angiogenesis and are the main therapeutic targets for anti-angiogenesis to inhibit tumors [8]. Tumor neovascularization significantly impacts tumor growth, infiltration, and metastasis, thus exerting a profound influence on the malignant progression of tumors [9]. An elevated level of VEGFA expression is intricately linked with poor prognosis, including gastric, liver, and esophageal cancers [[10], [11], [12]]. It has been reported that tumor cells characterized by heightened expression levels of VEGFA demonstrate augmented malignancy and exhibit innate resistance to programmed cell death protein-1/programmed death ligand 1 (PD-1/PD-L1) monoclonal antibodies [13]. The inhibition of tumor angiogenesis is an effective clinical therapeutic strategy to improve patient survival prognosis. Although numerous antiangiogenic drugs are currently being developed, only a few antiangiogenic drugs are currently approved for the treatment of HCC, such as sorafenib and regorafenib [5]. However, these drugs have been associated with significant adverse effects in patients, notably cardiovascular toxicity effects including hypertension, myocardial ischemia, myocardial infarction, and bleeding [14]. Furthermore, prolonged administration of these drugs may culminate in the development of acquired resistance to angiogenesis inhibitors [6]. Therefore, it is important to discover new small-molecule anti-angiogenic drugs characterized by low toxicity profiles.

Among all immune checkpoints, the PD-L1/PD-1 signaling pathway stands out as a therapeutic target in numerous malignancies [15]. More than 1,000 clinical trials are presently evaluating antagonistic antibodies that target the inhibitory immune receptor PD-1 and its ligand PD-L1, which have been approved for the treatment of cancer, including solid tumors such as liver cancer, melanoma, non-small cell lung cancer (NSCLC), renal cell carcinoma (RCC), Hodgkin’s lymphoma, bladder cancer, as well as head and neck squamous cell carcinoma (HNSC) [16,17]. Despite the enduring responses and disease remission observed in certain cancer patients treated with PD-1/PD-L1 targeted therapies, their clinical application is constrained by the comparatively lower response rates and the emergence of novel drug resistance mechanisms [18]. Therefore, in order to create combination therapy strategies to overcome the PD-1/PD-L1 blockade, it is imperative to comprehend the regulatory mechanisms of the PD-1/PD-L1 axis.

Naringenin, a naturally occurring flavonoid abundant in fruits and vegetables, such as grapefruit, lemons, and oranges, exhibits a multifaceted pharmacological profile including anti-inflammatory, anti-infective, antioxidant, and anti-tumor properties. Moreover, it serves as a safe and low-toxicity immune system modulator [[19], [20], [21]]. Zhang et al. [22] have elucidated the significant anti-liver cancer properties of naringenin, highlighting its efficacy in inducing apoptosis and modulating critical genes and pathways relevant to liver cancer. Furthermore, Maatouk et al. [23] demonstrated that naringenin promotes proliferation in B cells and T cells and enhances natural killer (NK) cell activity. Multiple investigations have also shown the association between naringenin and PD-L1-related oncogenic signaling pathways, including the transforming growth factor-beta (TGF-β), mitogen-activated protein (MAP) kinase (MAPK), and p53 signaling pathways [[24], [25], [26]]. However, the investigation into the molecular mechanisms underlying naringenin’s therapeutic potential in HCC remains deficient. Therefore, it is crucial to reveal the molecular mechanism through which naringenin suppresses the malignant progression of tumors, particularly in HCC, providing a new clinical approach for prolonging the survival time of cancer patients and improving prognosis.

2. Materials and methods

2.1. Cell lines and cell culture

Chinese Academy of Sciences Cell Bank (Shanghai, China) provided the hepatocellular cancer cell lines (Huh7, Hep1-6). The cells were cultured at 37 °C in a humid incubator with 5% CO₂ in Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) (Invitrogen). All of the experiments were carried out when the cells density reached 80%–90%, in compliance with standard protocol. The cell transfection assay in our study was conducted by using ProFect-3K Transfection Reagent Kit (Cat. No: 82-300015) (Celfolio, MacRoll Biotech Co., Ltd., Guangzhou, China).

2.2. Cell viability assay

Huh7 and Hep1-6 cells were spread over into 96-well cell culture plates. Naringenin was solvated in dimethyl sulfoxide (DMSO). After being incubated for 24 h, the treatment period for the cells was 24 and 48 h, respectively, with the indicated concentrations of naringenin. A control group was also included, and this group was given a medium containing the same concentration of DMSO as the treatment group. Subsequently, cell viability was assessed by the use of Cell Counting Kit-8 (CCK-8) agents, observing the manufacturer’s instructions (Beyotime Biotechnology, Shanghai, China). An independent experiment was conducted in triplicate.

2.3. Colony formation assay

Six-well plates containing 1,000 cells each were used for the cell distribution, and 2 mL of culture medium were added to every well, which was blown several times and mixed well, and incubated at 37 °C for 10 days, then 1 mL of saline or phosphate-buffered saline (PBS) was slowly added along the wall of the wells to wash the cell surfaces, and the cells were fixed for 30 min in 1 mL of methanol, and they were then stained for 30 min with 1% crystal violet. All the experiments were performed in three replicates.

2.4. Soft agar colony formation assay

Following 48 h of drug treatment, the cells were digested with trypsin, and subsequently, the culture medium containing 2 × 10³ cells was amalgamated with 0.66% soft agar. This mixture was then evenly spread out over a layer of 1.32% agar-containing cell culture medium. The experimental procedures were conducted in triplicate to ensure accuracy and reproducibility. After 10 days, the dimensions of the colonies were assessed utilizing a visual micrometer, and the quantification of the colonies was performed.

2.5. Animal models

Six weeks old of C57BL/6 male mice were obtained from the Guangdong Provincial Medical Laboratory Animal Center (Foshan, China) and were housed in standard conditions. All animal experiments were conducted in strict adherence to the animal experiment regulations and ethical standards set forth by the Animal Experiment Center of Guangdong Pharmaceutical University (Guangzhou, China). The Hep1-6 cells (logarithmic growth phase) were injected subcutaneously with 1 × 10⁶ cells into the right back of mice. Each group of six mice received a random assignment to either the treatment or control group. The control group was administered gavage injections of PBS, while the treatment group received naringenin at a rate of 100 mg/kg per day through gavage injections. Tumor size was examined with vernier calipers every two days during the experiment period. Tumor volume (V) was calculated using the formula: V (mm³) = (length) × (width)² × 0.5. The mice received naringenin treatment for 28 days before they were euthanized, at which point the tumors were excised and photographed. All experimental procedures in studies involving animals were in accordance with the ethical standards of the institutions at which the studies were conducted and were approved by Institutional Animal Care and Use Committee of Guangdong Pharmaceutical University (Guangzhou, China) on August 20, 2022 (Approval No.: gdpulac2022185).

2.6. Flow cytometry analysis

Following the collection of mouse lymph, spleen, and tumor tissues, cell suspensions were prepared by grinding the lymph and spleen using a 70-μm cell filter. Subsequently, the spleen cell suspensions were combined with Red Blood Cell Lysis (Solarbio, Beijing, China) for 15 min to isolate immune cells. Tumors were sliced and digested with the appropriate volume of digestion solution for 1 h in a thermostat, followed by isolation of tumor immune cells using the Mouse Tumor Infiltrating Tissue Mononuclear Cell Isolation Kit (Solarbio). An appropriate number of cells was taken and stained with antibodies for 30 min at 4 °C. The monoclonal anti-mouse antibodies CD3-BV421 (BioLegend, San Diego, CA, USA), CD4-PE (BioLegend), and CD8-APC (BioLegend) were selected for utilization in the study. Following a wash with 1× PBS, 300 μL of PBS was used to resuspend the cells. Subsequently, a flow cytometry analysis was conducted using a BD-FACS Celesta flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA), and the resulting data were processed using FlowJo software.

2.7. Chicken chorioallantoic membrane (CAM) assay

The chicken embryos, aged five days, were obtained from Xinxing Dahua Agricultural Poultry and Egg Co., Ltd. (Guangzhou, China). The CAM assay was performed in accordance with the methodology described in our previous research [27]. Briefly, five-day-old chicken embryos were purchased from Xinxing Dahua Agricultural Poultry and Egg Co., Ltd. (Guangzhou, China). We stored the eggs in an incubator at 37 °C with appropriate humidity for at least three days to allow for early embryonic development. On the third day of incubation, we disinfected the surface of the eggs using 75% ethanol to prevent contamination. We then used sterile forceps and a scalpel or scissors and created a window from the air cell end in the egg to expose the CAM. The CAM should be visible and easily accessible for further experimentation. We use a micropipette to directly add a small amount (usually 50−100 µL) of conditioned medium to CAM. First, we ensured the application site is free of debris or other substances to avoid interference. We carefully closed the egg window or covered it to maintain humidity and temperature. And then we returned the eggs to the incubator for a specified duration, typically five days. After the incubation period, we carefully opened the eggs again, observed the CAM, and used photography or imaging software to document the results.

2.8. Immunohistochemistry (IHC)

Tumor tissues obtained from mice were fixed with 4% paraformaldehyde for 48 h. Following fixation, the tissues underwent dehydration using a series of ethanol concentrations and were subsequently embedded in paraffin wax. The embedded tissues were then sectioned into 5-μm thick pieces. The sections underwent deparaffinization for antigen retrieval and were treated with H₂O₂ followed by closure with bovine serum albumin (BSA). Subsequently, the sections were immunostained using anti-PD-L1 antibody (1:200, 28076-1-AP, Proteintech, Wuhan, China) and anti-Ki67 antibody (1:1000, 27309-1-AP, Proteintech) and spent the following night incubated at 4 °C. After the anti-secondary antibody incubation, the sections underwent restaining with hematoxylin. Subsequently, diaminobenzidine (DAB) chromogenic solution (OriGene, Beijing, China) was applied, and the resulting images were captured and analyzed using ImageJ.

2.9. Western blotting analysis

Huh7 and Hep1-6 cells were cultivated in a six-well plate. Following a 48-h exposure to varying naringenin concentrations, cells were harvested and treated with radioimmunoprecipitation assay (RIPA) lysis buffer (Epson, Shanghai, China) supplemented with protein phosphatase inhibitor to extract the protein supernatant. The mixture of the supernatant and 5× loading buffer (Biosharp, Beijing, China) was then heated for 10 min at 100 °C. After the separation of the protein using an sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel, it was subsequently transferred to nitrocellulose membranes, subjected to blocking using 5% skim milk powder, and allowed to incubate for 1 h. The membranes were then incubated with the primary antibody and kept at 4 °C for the duration of the overnight process. Primary antibodies include the following: anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody (1:20000, 28076-1-AP, Proteintech), anti-PD-L1 antibody (1:500, 60004-1-Ig, Proteintech), anti-mesenchymal-epithelial transition factor (c-Met) antibody (1:200, 25869-1-AP, Proteintech), anti-p-MAPK antibody (1:1000, 28733-1-AP, Proteintech), anti-MAPK antibody (1:2000, 11257-1-AP, Proteintech), anti-MAP kinase (MEK) antibody (1:5000, 11049-1-AP, Proteintech), and anti-p-MEK antibody (1:1000, ab96379, Abcam, Cambridge, UK). On the following day, the membranes were then incubated with the secondary antibody for an hour at room temperature and were visible using enhanced chemiluminescence (ECL) (Millipore Corporation, Billerica, MA, USA).

2.10. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

RNAiso Plus (Takara, Dalian, China) was used to extract RNA in accordance with the directions provided by the manufacturer. Reverse-transcribed complementary DNA (cDNA) from total RNA using the Goldenstar™ RT6 cDNA Synthesis Kit (Tsingke Biotechnology Co., Ltd., Beijing, China). StepOnePlus™ Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA) was used to perform qRT-PCR using ChamQ Universal SYBR qPCR Master Mix kit (Vazyme, Nanjing, China). GENEWIZ (Guangzhou, China) performed qPCR utilizing the primers listed below: PD-L1 (forward: AAGAACATTATTCAATTTGTGCATG; reverse: GACTTTCACAGTAATTCGCTTGTAG). GAPDH (forward: GGAGCGAGATCCCTCCAAAAT; reverse: GGCTGTTGTCATACTTCTCATGG). The relative messenger RNA (mRNA) levels of various genes were determined by measuring the 2^−ΔΔC_T values, which represent the levels of gene expression.

2.11. Network pharmacological analysis

Naringenin’s Chemical Abstracts Service (CAS) number (CAS: 480-41-1) was acquired from https://www.chemsrc.com/. The Traditional Chinese Medicine Systems Pharmacology Database (TCMSP; https://old.tcmsp-e.com/tcmsp.php) provided the structural details of naringenin. Targets associated with HCC were sourced from GeneCards (https://www.genecards.org/); potential targets of naringenin were sourced from the SwissTargetPrediction database (http://swisstargetprediction.ch). The Jvenn platform (https://www.bioinformatics.com.cn/static/others/jvenn/) was utilized to create a Venn diagram for visualization based on the intersections of naringenin with possible targets in HCC. Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database (https://cn.string-db.org/) provided protein-protein interaction (PPI) data. Cytoscape software was employed to construct and visualize two PPI interactive networks. These networks encompass the predicted potential targets of naringenin against HCC and a naringenin-target-pathway network. The Database for Annotation, Visualization, and Integrated Discovery (DAVID; https://david.ncifcrf.gov/) was used to analyze biological processes (BP) of potential targets for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses. Bioinformatics refers to an internet-based data analysis and visualization platform (http://www.bioinformatics.com.cn) that is capable of producing bubble and bar charts.

2.12. Molecular docking experiments

The mol2 file of the chemical structure of naringenin and the protein structure Protein Data Bank (PDB) file of related targets were downloaded from the PubChem website and PDB website respectively. The protein structure file was delegated and dehydrated by Pymol software, the mol2 file and the PDB file were imported into AutoDock software for molecular docking, and the outcomes were visualized using Pymol software. Subsequently, bioinformatics analysis and experiments were used to detect the expression of intersecting target genes for HCC.

2.13. Co-immunoprecipitation (Co-IP) assay

Co-IP and Western blot analysis were performed using standard protocols. The following antibodies were used: anti-flag or anti-hemagglutinin (HA). Immunoblotting (IB) assays were performed with the objective of elucidating the interactions between proteins involving the target protein.

2.14. Determination of protein degradation

The designated time points were chosen to administer chlorhexidine (CHX) or proteasome inhibitors (MG132) treatment to the cells, after which total protein was extracted. The quantification of IB assays was then utilized to calculate the protein half-lives.

2.15. Statistical analysis

The statistical analyses were conducted using the SPSS 21.0 statistical software package (IBM, Armonk, NY, USA). Paired t-test or one-way analysis of variance (ANOVA) was performed to analyze the data with normal distribution and followed by Bonferroni’s multiple comparisons test. The data with skewed distributions were tested using Mann-Whitney or Kruskal-Wallis tests. Pairwise comparisons among groups were done with Bonferroni’s correction. Survival curves of HCC patients were plotted according to the Kaplan-Meier method and compared using a log-rank test. P < 0.05 was established to serve as the limit for statistical significance. The data were presented as the mean ± standard deviation (SD) based on a minimum of three independent experiments.

3. Result

3.1. Inhibitory effect of naringenin on HCC cells growth in vitro

First, HCC cells were exposed to different concentrations of naringenin to investigate the impact of naringenin on liver cancer cells. Results from the CCK-8 assay revealed a remarkable suppression of cell survival in both HCC cells, with a more pronounced effect observed at 48 h compared to 24 h (Fig. 1A). Additionally, colony formation assay demonstrated a dose-dependent inhibition of cell proliferation by naringenin (Fig. 1B). Moreover, anchorage-independent growth assay revealed that colonies formed by the naringenin-treated groups were fewer and smaller compared to the control groups (Fig. 1C). These results suggested that naringenin effectively inhibits the proliferative activity of HCC cells.

Fig. 1.

Inhibitory effect of naringenin on hepatocellular carcinoma (HCC) cells growth in vitro. (A) Cell Counting Kit-8 (CCK-8) assay of Huh7 and Hep1-6 cells after treated with different concentrations of naringenin in 24 h (left) and 48 h (right). (B) The ability of cell colony formation was assessed after treatment with different concentrations of naringenin, and the results of the experiments were analyzed. (C) Soft agar assay was performed in Huh7 and Hep1-6 cells after treated with different concentrations of naringenin, and the colony numbers were quantified. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001, paired t-test.

3.2. Naringenin inhibits HCC angiogenesis in vitro

Angiogenesis was one of the hallmarks of cancer and was required for the sustenance of nutrients and oxygen, as well as the ability to evacuate metabolic wastes during tumor growth and metastasis. We further investigate whether naringenin plays an anti-angiogenesis. As shown in Fig. 2A, naringenin significantly reduced the ability of HCC cells to induce tube formation in human umbilical vein endothelial cells (HUVECs) in a dose-dependent pattern (P < 0.01). Migration assays showed that the conditioned medium derived from Huh7 or Hep1-6, which was treated with elevated concentrations of naringenin, significantly reduced the migratory capability of HUVECs (Fig. 2B). Strikingly, CAM experiments demonstrate that conditioned medium derived from Huh7 or Hep1-6, which was treated with naringenin, effectively suppressed neovascularization in a dose-dependent pattern (Fig. 2C). However, the tube formation ability, migratory capability of HUVECs, and neovascularization levels showed no significant change when treated with naringenin alone (Fig. 2). The above results indicate that naringenin inhibits angiogenesis mainly by suppressing the expression of angiogenic factors coming from medium derived from tumor cells.

Fig. 2.

Naringenin inhibits hepatocellular carcinoma (HCC) angiogenesis in vitro. (A) The effect of naringenin on tube formation and human umbilical vein endothelial cells (HUVECs) tube formation with the conditioned medium derived from the indicated HCC cells. (B) The effect of naringenin on HUVEC cell migration with the conditioned medium derived from the indicated HCC cells. (C) Representative images of the blood vessels formed in the chicken chorioallantoic membrane (CAM) assay, after stimulation with the conditioned medium derived from the indicated HCC cells. P < 0.01, paired t-test. ns: not significant.

3.3. Identification of therapeutic targets for naringenin in HCC

To elucidate the potential mechanism of naringenin in anti-tumor and anti-angiogenesis of HCC, initially, we gathered naringenin-related target genes and disease target genes from various databases, including TCMSP, SwissTargetPrediction, GeneCards, and DisGeNET, and obtained a total of 97 naringenin-related target genes and 11,336 disease target genes after removing duplicates. Subsequently, naringenin-disease intersection targets were obtained by establishing a Venn diagram, resulting in the identification of 78 overlapping target genes (Figs. 3A and B). We then conducted GO function enrichment analysis on the 78 targets. The analysis revealed that BP were primarily enriched in regulation of the MAPK cascade, regulation of MAPK activity, regulation of apoptotic process, VEGF signaling pathway, regulation of angiogenesis, and other related processes (Fig. 3C and Table S1). Additionally, we identified five pivotal genes, namely VEGFA, MET, ESR1, PPARG, as well as APP, and utilized AutoDock Vina for molecular docking to confirm their protein targets interact with naringenin. The molecular docking diagrams showed significant binding energies between naringenin and the two targets, namely VEGFA and c-Met, suggesting their potential candidacy as naringenin targets (Figs. 3D and S1A). Based on the findings from the GO enrichment analysis and molecular docking results, Huh7 and Hep1-6 cells exposed to various concentrations of naringenin were subjected to Western blot experiments to evaluate the levels of protein expression of c-Met and VEGFA. Our results demonstrated that naringenin led to decreased expression levels of c-Met and VEGFA in adose-dependent pattern (Figs. 3E and S1B). Meanwhile, our analysis revealed that c-Met and VEGFA displayed high expression levels in tumor tissue, compared with normal tissue (Fig. S1C).

Fig. 3.

Identification of therapeutic targets for naringenin in hepatocellular carcinoma (HCC). (A) Network diagram of naringenin targets. (B) Venn diagram indicating cross genes for naringenin and HCC. (C) The Gene Ontology (GO) enrichment analysis of biological processes (BP). (D) Pattern diagram of molecular docking results of naringenin with vascular endothelial growth factor A (VEGFA), mesenchymal-epithelial transition factor (c-Met). (E) Proteins in HCC cells treated with different concentrations of naringenin were extracted, and the changes in c-Met and VEGFA protein levels were measured. (F) The ability of cell colony formation was assessed in the indicated HCC cells. (G) Soft agar assay was performed in the indicated HCC cells. (H) Representative images of the blood vessels formed in the chicken chorioallantoic membrane (CAM) assay, after stimulation with the conditioned medium derived from the indicated HCC cells. (I) The effect of naringenin on HUVEC tube formation with the conditioned medium derived from the indicated HCC cells. P < 0.01, paired t-test. MAPK: mitogen-activated protein (MAP) kinases; ERK1: extracellular signal-regulated kinases 1; FDR: false discovery rate; GAPDH: glyceraldehyde-3-phosphate dehydrogenase.

Furthermore, we found that overexpression of c-Met-open reading frame (ORF) significantly reversed the anti-tumor effect of naringenin in HCC, as indicated by colony formation assay and anchorage-independent growth assay (Figs. 3F and G). Meanwhile, HUVEC tube formation assay and CAM experiment results indicate that enhanced expression of VEGFA reversed the anti-angiogenesis effect of naringenin in HCC (Figs. 3H and I). Collectively, these results indicate that c-Met and VEGFA are key mediators of therapeutic targets for anti-tumor and anti-angiogenesis of naringenin in HCC.

3.4. Naringenin induced protein degradation of c-Met and VEGFA through Lys-48-linked polyubiquitination

We next investigate how naringenin downregulates c-Met and VEGFA expression in liver cancer. First, we found that naringenin treatment does not affect the mRNA expression of c-Met and VEGFA both in Huh7 and Hep1-6 cells (Fig. 4A). Furthermore, c-Met and VEGFA protein expression levels were detected at different time points by MG132 (inhibitors of protein degradation) or CHX (inhibitors of protein synthesis) treatment. As shown in Fig. S2, naringenin treatment in HCC cell lines had no significant effect on c-Met and VEGFA protein expression after MG132 treatment. However, naringenin treatment in HCC cells significantly induced the degradation of c-Met and VEGFA and decreased protein expression of c-Met and VEGFA after CHX treatment (Fig. 4B), which indicates that naringenin promotes the protein degradation of c-Met and VEGFA in liver cancer cells.

Fig. 4.

Naringenin induced protein degradation of mesenchymal-epithelial transition factor (c-Met) and vascular endothelial growth factor A (VEGFA) through Lys-48-linked polyubiquitination. (A) The messenger RNA (mRNA) levels of c-Met and VEGFA were measured in different concentrations of naringenin treatment. (B) Protein expression of c-Met and VEGFA in indicated hepatocellular carcinoma (HCC) cells at various time points after addition of chlorhexidine (CHX). The expression levels were normalized against loading control glyceraldehyde 3-phosphate dehydrogenase (GAPDH). (C) Polyubiquitination levels of c-Met and VEGFA were measured in the indicated conditions. (D) Polyubiquitination levels of c-Met and VEGFA were measured in HCC cells transfected with ubiquitin (Ub)-K48R. (E) Polyubiquitination levels of c-Met and VEGFA were measured in HCC cells treatment with different concentrations of naringenin. ∗P < 0.05. ns: not significant. HA: hemagglutinin; IP: immunoprecipitation; IB: immunoblotting.

To further explore the effect of naringenin in degrading c-Met and VEGFA protein whether through the ubiquitination and proteasome pathway, HA-K48 or HA-K63 was transfected and performed Co-IP assay. We found that the levels of K48-ubiquitinated c-Met and VEGFA was enhanced in naringenin treatment cells, while the levels of K63-ubiquitinated c-Met and VEGFA showed no significant difference (Fig. 4C). Further, the accumulation of ubiquitinated c-Met and VEGFA induced by naringenin treatment was abolished by transfecting with mutants of ubiquitin (Ub-K48R) (Fig. 4D). Importantly, the K48-ubiquitinating levels of c-Met and VEGFA also increased in a dose-dependent pattern of naringenin treatment in liver cancer (Fig. 4E). These observations suggest that naringenin promotes protein degradation of c-Met and VEGFA through Lys-48-linked polyubiquitination.

3.5. Naringenin inhibits HCC tumor growth in vivo

To provide more insight into how naringenin affects HCC in vivo, we established a subcutaneous tumor model in mice for investigation. As shown in Figs. 5A–C, naringenin treatment exhibited a significant reduction in both size and weight of tumor compared to the control group (P < 0.01). Importantly, we found that the levels of c-Met, Ki67-positive ratio, and the EV density, as indicated by the number of CD31-positive vessels, were significantly decreased in the naringenin-treated group (Fig. 5D). It is noteworthy that naringenin administration did not adversely affect the body weight of the mice, suggesting that naringenin has no significant effect on the quality of life of mice (Fig. 5E). These results suggested that naringenin inhibits HCC growth in vivo.

Fig. 5.

Naringenin inhibits tumor growth in vivo. (A) Hepatocellular carcinoma (HCC) cells were subcutaneously injected into C57BL/6J mice (n = 6/group). Images of the tumors from all mice in each group. (B) Tumor volume of mice from the indicated group were measured (n = 6). (C) Tumor weights of mice from the indicated group were measured (n = 6). (D) Hematoxylin and eosin (H&E) staining, Ki67 and CD31 immunohistochemistry (IHC) staining, and mesenchymal-epithelial transition factor (c-Met) were performed in the control group and naringenin group. Each bar represents the mean ± standard deviation (SD) of three independent experiments. (E) Mouse weight from the indicated group was measured. P < 0.01, paired t-test.

3.6. Naringenin downregulates PD-L1 expression through the c-Met/MAPK signaling pathway

It has been reported that c-Met regulates transcription of PD-L1 through the MAPK signaling pathway, thereby promoting the progression of HCC [27]. Next, we detected mRNA and protein expression of PD-L1 in HCC cells after exposure to different concentrations of naringenin, and the results showed that mRNA and protein expression levels of PD-L1, as well as the phosphorylation levels of MAPK and MEK decreased with increasing naringenin concentration (Figs. 6A and B). Meanwhile, to visually assess the changes in PD-L1 expression, we conducted IHC experiments. As shown in Figs. 6C and D, PD-L1 expression levels of tumor tissues in the naringenin-treated group were significantly reduced, compared with the control group (P < 0.01). Furthermore, we used the c-Met inhibitor to treat the HCC cells and revealed that the protein expression levels of c-Met and PD-L1 as well as the phosphorylation levels of MAPK and MEK, were significantly decreased (Fig. 6E). These results suggested that naringenin inhibits the protein expression level of PD-L1 mainly through suppressing the c-Met/MAPK signaling pathway in liver cancer.

Fig. 6.

Naringenin downregulates programmed death ligand 1 (PD-L1) expression through mesenchymal-epithelial transition factor/mitogen-activated protein (MAP) kinases (c-Met/MAPK) signaling pathway. (A) The messenger RNA (mRNA) expression levels of PD-L1 were assessed following treatment with varying concentrations of naringenin in Huh7 and Hep1-6 cells. (B) The protein levels of PD-L1 and phosphorylation levels of MAPK and MAP kinase (MEK) in Huh7 and Hep1-6 cells were evaluated through Western blot analysis following treatment with varying concentrations of naringenin. (C) Immunohistochemistry (IHC) staining analyses the PD-L1 expression in the indicated group. (D) Flow cytometry analysis was conducted to evaluate PD-L1 expression in the indicated group. (E) Proteins were extracted from hepatocellular carcinoma (HCC) cells treated with a c-Met inhibitor, and the activation status of the MAPK signaling pathway, as well as the alterations in c-Met and PD-L1 protein levels were assessed. ^∗P < 0.05 and ^∗∗P < 0.01. GAPDH: glyceraldehyde 3-phosphate dehydrogenase; FL1-A: fluorescent channel 1-A; PE: phycoerythrin; MFI: mean fluorescence intensity.

3.7. Naringenin enhances anti-tumor immunity in HCC

We further explore the relevance of c-Met and immune cells in HCC. Database analyses showed the most significant negative correlation between c-Met and CD8⁺ T cells (Figs. 7A and B). Subsequently, we explored the impact of naringenin on CD8⁺ T cell activation. Flow cytometry revealed that naringenin-treated mice showed a significantly greater amount of spleen and tumor CD8⁺ T cells, along with a notably greater ability to secrete cytotoxicity-related cytokines including interferon gamma (IFN-γ), tumor necrosis factor-alpha (TNF-α), and granzyme B (GzmB), compared to the control group (Figs. 7C and D, P < 0.001). We further analyzed the correlation between c-Met and stroma as well as immune cells in tumor tissues and found that c-Met was negatively correlated with the degree of immune infiltration in tumors, with higher c-Met expression being associated with lower total immune infiltration in tumors (Fig. 7E). These results suggest that naringenin plays potential effects in enhancing anti-tumor immunity in HCC by decreasing c-Met expression.

Fig. 7.

Naringenin enhances anti-tumor immunity in hepatocellular carcinoma (HCC). (A, B) Correlation analysis: the relationship between mesenchymal-epithelial transition factor (c-Met) and immune cell populations (A) and the relationship between c-Met and activated CD8⁺ T cell, myeloid-derived suppressor cells (MDSC), mast cell, as well as macrophage (B). (C) Flow cytometry analysis was conducted to evaluate the proportion of CD8⁺ T cells in both spleen and tumor tissues following treatment with naringenin, compared to the control group. (D) Serum levels of interferon gamma (IFN-γ), tumor necrosis factor-alpha (TNF-α), and granzyme B (GzmB) secreted by CD8⁺ T cells in the control group and the naringenin-treated group were quantitatively analyzed. (E) The correlation analysis of infiltration between c-Met and stroma as well as immune cells in tumor tissues. P < 0.001. NK: natural killer; TCGA: The Cancer Genome Atlas; LIHC: liver hepatocellular carcinoma.

3.8. Combination therapy with anti-PD-L1 or bevacizumab enhances sensitivity of liver cancer to naringenin

We further used an in vivo mouse model to investigate the therapeutic effect of the combination therapy of naringin and PD-L1 antibody or bevacizumab (a monoclonal antibody that can inhibit VEGF). As shown in Fig. 8, combination therapy with anti-PD-L1 or bevacizumab significantly decreased tumor volume and tumor weight and did not have an effect on body weight of mice. Taken together, these data demonstrate that application of naringenin with PD-L1 antibody or bevacizumab provided better therapeutic effects in liver cancer.

Fig. 8.

Combination therapy with anti-programmed death ligand 1 (PD-L1) or bevacizumab enhances the sensitivity of liver cancer to naringenin. (A) Hep1-6 cells were subcutaneously injected into C57BL/6J mice (n = 5/group). Images of the tumors from all mice in each group. (B) Tumor volumes of mice from the indicated group were measured (n = 5). (C) Tumor weights of mice from the indicated group were measured (n = 5). (D) Mouse weight from the indicated group was measured (n = 5).

3.9. Upregulation of c-Met and VEGFA was associated with HCC patient prognosis

We further evaluated the prognostic relevance of c-Met and VEGFA expression in HCC patients, and Kaplan-Meier curves demonstrated that high levels of c-Met and VEGFA expression were associated with poor prognosis (Fig. 9A). Subsequently, we investigated the mRNA expression levels of c-Met and VEGFA in various cancers. We observed that c-Met expression was statistically significant in 21 types of cancer, among which c-Met was significantly upregulated in 19 types of cancer, including uterine corpus endometrial carcinoma (UCEC), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), lung adenocarcinoma (LUAD), esophageal carcinoma (ESCA), kidney renal papillary cell carcinoma (KIRP), colon adenocarcinoma (COAD), stomach adenocarcinoma (STAD), HNSC, kidney renal clear cell carcinoma (KIRC), lung squamous cell carcinoma (LUSC), liver hepatocellular carcinoma (LIHC), skin cutaneous melanoma (SKCM), bladder urothelial carcinoma (BLCA), thyroid carcinoma (THCA), rectum adenocarcinoma (READ), pancreatic adenocarcinoma (PAAD), pheochromocytoma and paraganglioma (PCPG), kidney chromophobe (KICH), and cholangiocarcinoma (CHOL). Conversely, c-Met expression was significantly decreased in breast invasive carcinoma (BRCA) and prostate adenocarcinoma (PRAD). VEGFA expression was significantly upregulated in 12 types of tumors and was significantly decreased in 6 types of tumors (Fig. 9B). To determine the types of cancers that might benefit from anti-c-Met immunotherapy, conducting pan-cancer analyses to elucidate the immunological role of c-Met is crucial. We extracted c-Met and 60 genes of two classes of immune checkpoint pathways, and our results showed that c-Met was correlated with most immune checkpoints (Fig. 9C). In summary, our study revealed that naringenin has a strong inhibitory effect on the proliferation and growth of HCC cells by enhancing anti-angiogenesis and anti-tumor immunity, suggesting naringenin probably be a potential adjuvant drug in the treatment of HCC (Fig. 10).

Fig. 9.

Upregulation of mesenchymal-epithelial transition factor (c-Met) and vascular endothelial growth factor A (VEGFA) was associated with hepatocellular carcinoma (HCC) patient prognosis. (A) Kaplan-Meier analysis was conducted to assess the overall survival and disease-free survival of patients with high and low expression levels of c-Met and VEGFA, and disease-free survival of patients with high and low expression levels of CD274. (B) c-Met and VEGFA messenger RNA (mRNA) expression levels in different tumors (T) and corresponding normal (N) tissues from The Cancer Genome Atlas (TCGA) and The Genotype-Tissue Expression (GTEx) database by SangerBox. (C) The correlation between c-Met and immune checkpoint expression in pan-cancer. ^∗P < 0.05, ^∗∗∗P < 0.001, and ^∗∗∗∗P < 0.0001. UCEC: uterine corpus endometrial cancer; BRCA: breast cancer; CESC: cervical squamous cell carcinoma; LUAD: lung adenocarcinoma; ESCA: esophageal cancer; KIRP: kidney renal papillary cell carcinoma; COAD: colon adenocarcinoma; PRAD: prostate adenocarcinoma; STAD: stomach adenocarcinoma; HNSC: head and neck squamous cell carcinoma; KIRC: kidney renal clear cell carcinoma; LUSC: Lung squamous cell carcinoma; LIHC: liver hepatocellular carcinoma; SKCM: skin cutaneous melanoma; BLCA: bladder cancer; THCA: thyroid cancer; READ: rectum adenocarcinoma; PAAD: pancreatic adenocarcinoma; PCPG: pheochromocytoma and paraganglioma; KICH: kidney chromophobe carcinoma; CHOL: cholangiocarcinoma; TGCT: testicular germ cell tumors; THYM: thymoma; UVM: uveal melanoma; GBM: glioblastoma multiforme; GBLGG: lower grade glioma and glioblastoma; LGG: low grade glioma; ACC: adrenocortical carcinoma; LAML: acute myeloid leukemia; OV: ovarian serous cystadenocarcinoma; MESO: mesothelioma; KIPAN: pan-kidney cohort; UCS: uterine carcinosarcoma; STES: stomach and esophageal carcinoma; COAD/READ: colon adenocarcinoma/rectum adenocarcinoma; SARC: sarcoma; DLBC: diffuse large B-cell lymphoma; SELP: p-selectin; ICOSLG: inducible T cell costimulator ligand; TNF: tumor necrosis factor; ILB1: interleukin 1 beta; CXCL9: C−X−C motif ligand 9; TNFRSF9: TNF receptor superfamily member 9; IL2RA: interleukin 2 receptor subunit alpha; ITGB2: integrin subunit beta 2; IFNA1: interferon (IFN) alpha 1; CCL5: C−C motif chemokine ligand 5; GZMA: granzyme A; PRF1: perforin 1; ICOS: inducible T-cell co-stimulator; BTN3A2: butyrophilin subfamily 3 member A2; ICAM1: intercellular adhesion molecule 1; TLR4: Toll like receptor 4; ENTPD1: ectonucleoside triphosphate diphosphohydrolase 1; CX3CL1: C−X3−C motif chemokine ligand 1; HMGB1: high mobility group box 1; EDNRB: endothelin receptor type B; HAVCR2: hepatitis A virus cellular receptor 2; C10orf54: chromosome 10 open reading frame 54; TGFB1: transforming growth factor (TGF) beta 1; VTCN1: V-set domain containing T cell activation inhibitor 1; ADORA2A: adenosine A2a receptor; ARG1: arginase 1; TIGIT: T Cell immunoreceptor with Ig and ITIM domains; CTLA4: cytotoxic T-lymphocyte associated protein 4; SLAMF7: SLAM family member 7; IDO1: indoleamine 2,3-dioxygenase 1; PDCD1: programmed cell death 1; BTLA: B and T lymphocyte associated; KIR2DL3: killer cell immunoglobulin like receptor, two Ig domains and long cytoplasmic tail 3; LAG3: lymphocyte activation gene 3.

Fig. 10.

Schematic diagram illustrating that naringenin enhances anti-angiogenesis and anti-tumor immunity via downregulating expression of vascular endothelial growth factor A (VEGFA) and mesenchymal-epithelial transition factor (c-Met), thereby suppressing tumor progression in hepatocellular carcinoma (HCC). Ub: ubiquitin; MAPK: mitogen-activated protein (MAP) kinases; MEK1/2: MAP kinase kinase 1/2; ERK1/2: extracellular signal-regulated kinases 1/2; PD-L1: programmed cell death ligand 1; IFN-γ: interferon-gamma; TNF-α: tumor necrosis factor-alpha; GzmB: granzyme B.

4. Discussion

HCC is characterized by poor response to chemotherapy [28], with patients presenting impaired liver function often exhibiting heightened sensitivity to drug-related adverse reactions [29]. Sorafenib has emerged as a breakthrough in addressing the therapeutic challenges of advanced-stage liver cancer. Nonetheless, its efficacy remains largely unsatisfactory [30]. Beyond sorafenib, numerous clinical investigations into novel molecular targeted therapies have concluded unsuccessfully [31]. However, efforts to explore effective treatment options are still ongoing. In this study, experiments conducted both in vitro and in vivo have demonstrated the anti-tumor effect of naringenin in HCC. The results provide strong evidence that naringenin shows promise as a low-toxicity medication candidate for the treatment of HCC.

Previous research has demonstrated that naringenin can inhibit angiogenesis. Li et al. [32] found that naringenin inhibited angiogenesis in endothelial cells through multiple mechanisms, including downregulation of estrogen-related receptor alpha (ERRα) expression, which inhibited VEGF production. Naringenin also directly inhibited the tyrosine kinase activity of VEGFR-2 (kinase insert domain receptor (KDR)) and VEGF-induced phosphorylation of focal adhesion kinase (FAK), paxillin, and protein kinase B (Akt), thereby interfering with the VEGF signaling pathway. Choi et al. [33] revealed that the inhibition of angiogenesis by naringenin is related to the angiopoietin-2 (Ang2)/tyrosine-protein kinase receptor-2 (Tie2) kinase signaling pathway. However‌, the potential clinical significance and molecular mechanisms of naringenin in anti-angiogenesis in HCC remain largely unexplored. Herein, we identified VEGFA and c-Met as potential targets of naringenin therapy for HCC through network pharmacology, and VEGFA is closely related to neovascularization. Subsequent experimental findings revealed naringenin’s capability to inhibit the tube formation and migration of HUVECs. Moreover, CAM model experiments further confirmed that naringenin can inhibit neovascularization. Notably, we noticed a noteworthy reduction in the expression of the vascular marker CD31 in tumor tissue following treatment with naringenin. Therefore, our study reveals the role of naringenin in inhibiting angiogenesis and suggests it as a low-toxicity anti-angiogenic drug, providing a novel clinical approach for the treatment of HCC.

Previously, Arul and Subramanian [34] revealed that naringenin inhibits proliferation and induces apoptosis in human HCC cells. Similar to previous studies, we observed a dose-dependent antiproliferative activity of naringenin on HCC cells; however, our study differs from previous studies in two ways. First, we used two cell lines, Huh7 and Hep1-6. Second, we used network pharmacology for data analysis, and we discovered the mechanism of relevance of naringenin to cellular immunomodulation. PD-L1, an immune checkpoint molecule, binding to the receptor PD-1, significantly inhibits T-cell activity, and allows tumor cells to escape from immune recognition and clearance [35]. ICIs are currently one of the most promising therapies for advanced liver cancer. While ICIs effectively target cancer cells, they frequently lead to the unintended consequence of attacking normal cells, tissues, and organs, culminating in immunotherapy-related adverse reactions (irAEs), such as hepatotoxicity, skin toxicity, and nephrotoxicity [16]. Therefore, identifying new PD-L1 inhibitors with enhanced efficacy and reduced toxicity and side effects is a crucial avenue in liver cancer treatment. In a previous study, Xu et al. [27] showed that c-Met regulates PD-L1 transcription through the MAPK/nuclear factor-kappaB (NF-κB)p65 pathway in HCC. In this study, we discovered that naringenin can downregulate PD-L1 expression in HCC cells by inhibiting c-Met expression and its downstream MAPK signaling pathway phosphorylation. Interactions between immune cells and tumor cells in the HCC microenvironment play an important role in the development of HCC [36], with cytotoxic CD8⁺ T cells particularly pivotal in anti-infective and anti-tumor responses [37]. Our further demonstration showed that naringenin can influence the tumor microenvironment and enhance the immune-killing capacity of CD8⁺ T cells against HCC cells. This not only offers new insights into the molecular mechanism of naringenin in regulating PD-L1 expression in liver cancer but also presents potential new targeted drugs for clinically managing immune evasion in liver cancer.

Naringenin is a common flavonoid with a variety of biological activities such as antioxidant, anti-inflammatory, anticancer, and immunomodulatory activities [21]. Furthermore, naringenin is characterized by low bioavailability and high intestinal metabolism [25]. It has been reported that naringenin can inhibit the proliferation of a variety of tumor cells without causing significant cytotoxicity to normal cells [21]. In the realm of liver cancer investigation, naringenin is principally concerned with inhibiting oxidative stress [38], modulating the TGF pathway [25], inducing apoptosis, and regulating the MAPK [39] and p53 pathway [34]. In our study, utilizing network pharmacology, we identified two potential targets of naringenin for HCC treatment, namely VEGFA and c-Met. Subsequently, through experimental validation, we demonstrated that naringenin inhibits angiogenesis by targeting VEGFA and suppresses PD-L1 expression by targeting the expression of c-Met protein and the phosphorylation of the MAPK signaling pathway. In HCC, the heightened expression of both c-Met and VEGFA correlates significantly with poor patient prognoses. Moreover, we further reveal the upregulated expression of c-Met and VEGFA in various other cancers. Hence, naringenin, functioning as a dual inhibitor targeting both c-Met and VEGFA, emerges as a promising pharmacological candidate for broad-spectrum utilization in the management of various cancers. This discovery furnishes substantial impetus for forthcoming clinical treatment, potentially offering patients more effective therapeutic options.

However, there are also some limitations in our study. First, the experiments were mainly based on in vitro and in vivo mouse models, and their efficacy and safety need to be verified in clinical trials. Meanwhile, the effects of naringenin on other tumor types need to be further investigated to evaluate its potential as an anti-tumor agent. In addition, the low bioavailability of naringenin may limit its effectiveness in clinical applications, and research should focus on improving the bioavailability of naringenin and developing its derivatives to enhance its anti-tumor effects in the future.

5. Conclusions

Our study reveals that naringenin can effectively inhibit angiogenesis and anti-tumor immunity in liver cancer by degradation of VEGFA and c-Met in a K48-linked ubiquitination manner. This work enlightens the potential effect of naringenin as a promising therapeutic strategy against anti-angiogenesis and anti-tumor immunity in liver cancer.

CRediT authorship contribution statement

Wenmei Wu: Writing – original draft, Data curation, Conceptualization. Xiangyu Qiu: Methodology, Data curation. Xiaofan Ye: Investigation, Data curation. Zhiliang Zhang: Formal analysis, Conceptualization. Siguo Xu: Methodology. Xiuqi Yao: Data curation. Yinyi Du: Methodology. Geyan Wu: Writing – review & editing, Writing – original draft, Conceptualization. Rongxin Zhang: Writing – review & editing, Writing – original draft, Conceptualization. Jinrong Zhu: Writing – review & editing, Writing – original draft, Conceptualization.

Declaration of competing interest

The authors declare that there are no conflicts of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article: