Cantharidin suppresses hepatocellular carcinoma development by regulating EZH2/H3K27me3-dependent cell cycle progression and antitumour immune response

Prediction and enrichment analysis of potential therapeutic targets and pathways of cantharidin in HCC

To dertermine the pharmacological mechanism of cantharidin in HCC, the targets of cantharidin were obtained from the Herb, Stitch, and SWISS-PROT databases, and the network pharmacology map was constructed based on the interactions between these targets and LIHC (Fig. 1A). Moreover, protein function annotation indicated that these targets of CTD were mainly enzymes, phosphatases, kinases, and oxidoreductases (Fig. 1B). To further explore the therapeutic targets of CTD in HCC, the HepG2 cells were treated with cantharidin and subjected to RNA-seq analysis. Subsequently, integrated analysis was performed based on the RNA-seq results and database information. A Venn diagram was used to confirm the key targets of CTD based on the targets predicted by the databases and the differentially expressed genes (DEGs) identified by RNA-seq in HCC (Fig. 1C). In total, 58 DEGs overlapped with targets from the databases, suggesting that these genes are the key targets of CTD in HCC (Fig. 1D). Furthermore, the interaction network of these target proteins was constructed via the STRING database. As shown in Fig. 1E, MAPK8/9/10/11/14, the PP1 and PP2 phosphatases, NFKB1A, and EGFR were the hub targets of CTD (Fig. 1E). Additionally, the KEGG enrichment analysis results indicated that these genes were significantly enriched in pathways in cancer. Moreover, these genes were also involved in the TNF signalling pathway, the MAPK signalling pathway, inflammatory mediator regulation of TRP channels, the IL-17 signaling pathway, the T cell receptor signaling pathway, and the apoptosis pathway (Fig. 1F). Considering these results collectively we speculated that cantharidin exerts its therapeutic effect by regulating these target genes and their related signalling pathways to inhibit the development of liver cancer.

Fig. 1figure 1

The targets network of cantharidin in HCC. (A) Network diagram of pharmacological regulation of cantharidin targets in HCC. Purple represents the targets of cantharidin from Herb database, dark green shows the targets of cantharidin from Sitch database; green and yellow are the targets from Swiss database. The probability yellow marker gene > 0.03, green marker gene < 0.03. (B) Statistical analysis protein properties of targets from different database. (C) The venn diagram shows the intersection of targets gene sets from three database and differentially expressed genes from cantharidin treated HepG2 cell. (D) Heat map shows the gene expression of predicted cantharidin targets in HepG2 cell. (E) PPI network of cantharidin targets in HCC. Red and green represent two hub networks in cantharidin targets. (F) Pathways with significant enrichment of target genes of cantharidin in HCC

To further explore the potential mechanism of cantharidin in HCC, we analysed our transcriptome data in depth. Based on threshold value of p < 0.05 and |log2FoldChange| > 1, a total of 7008 upregulated genes and 2427 downregulated genes were identified between HepG2 cells with and without cantharidin treatment (Fig. 2A). The expression of these genes was shown in Fig. 2B. Subsequently, GO and KEGG enrichment analyses were performed. These DEGs were significantly enriched in the terms of epithelium migration, epithelial cell migration, and tissue migration (Fig. 2C), suggesting that cantharidin likely inhibits cell metastasis by regulating EMT in liver cancer. Furthermore, the EMT gene set was obtained from the GSEA database and we found that 100 EMT-related genes were differently expressed after CTD treatment (Fig. 2D). Then, a PPI network of EMT-related genes was constructed according to the interaction relationships in the STRING database (Fig. 2E). The results indicated that cantharidin likely regulates these EMT-related genes to inhibit tumour cell growth and metastasis. Next, KEGG enrichment analysis was performed, and the results showed that the top 4 terms were “Cytokine-cytokine receptor interaction”, “ECM-receptor interaction”, “Transcription mis-regulation in cancer” and “Pathways in cancer” (Fig. 2F). Moreover, the top 10 cancer-related pathways were showed in Fig. 2G, including the MAPK, PI3K-Akt, cAMP, Jak-STAT, Wnt, NF-kappa B, AMPK and HIF-1 signalling pathways. Taken together, these finding indicate that CTD might inhibit HCC cell growth by regulating the above multiple pathways.

Fig. 2figure 2

Differentially expressed genes and their function enrichment analysis in cantharidin treated HepG2 cell. (A, B) volcano plot and heat map visualization of the differentially expressed genes after cantharidin treatments in HepG2. Red represents upregulation genes; green represents downregulation genes. (A) Volcano plot of genes. (B) Heat map of differentially expressed genes that logFC > 1, <-1, p < 0.05. (C) GO enrichment of differentially expressed genes. (D) The EMT related genes after cantharidin treatment HepG2 cell was screened using venn diagram. (E) PPI network of EMT related genes. (F) KEGG enrichment of the differentially expressed genes after cantharidin treatment in HepG2 cell. (G) Top 10 significantly enriched signaling pathways in cantharidin treated HepG2 cell

EZH2/H3K27me3 is essential for cell cycle progression in cantharidin treated HCC cells

Gene set enrichment analysis was performed to find promising therapeutic targets specific for CTD. Interestingly, the results showed that the DEGs in the CTD treated group were positively associated with H3K27me3 (Fig. 3A). EZH2 was significantly upregulated in CTD-treated samples compared to non-CTD-treated samples, implying that EZH2 may be the reason for the increase of H3K27me3 level after CTD treatment (Fig. 3B). Moreover, the RT-PCR results confirmed that EZH2 was upregulated in CTD treated HepG2 cells (Fig. 3C). To confirm the CTD-related H3K27me3-modified and repressed genes, we integrated our RNA-seq data from HepG2 cells and ChIP-seq datasets from the Cistrome Data Browser database. In total, 97 genes were occupied and downregulated by H3K27me3 in HepG2 cells (Fig. 3D, E). These data revealed that EZH2/H3K27me3 involved downstream target gene expression might play a critical role in the antitumour therapeutic effects of CTD.

Fig. 3figure 3

Cantharidin treatment is involved in EZH2/H3K27me3 related cell cycle pathway regulation. (A) GO functional annotation shows that cantharidin treatment is positively associated with H3K27me3 pathway based on GSEA analysis. (B) The heatmap shows H3K27me3 related genes expression in cantharidin treated HepG2 cell. Red represents up-regulation genes; blue represents down-regulation genes. (C) The expression of PRC2 complex genes, the key regulatory enzyme of H3K27me3, in cantharidin treated HepG2 cell. The *P < 0.05 was regarded as statistically significant. (D) Venn diagram shows the target genes of cantharidin downregulation genes and H3K27me3 significantly enriched genes from ChIP-seq results in HepG2 cell. (E) The heatmap shows the expression of 97 genes from venn analysis, which may be target genes of H3K27me3 in cantharidin treatment. Red represents up-regulation genes, green represents down-regulation genes. (F) Go enrichment analysis of the predicted target gene set. (G) Pathway analysis of the predicted target gene set. (H) PPI network of these target genes based on the STRING database results

We next uploaded the target genes to Metascape and found that these genes were significantly enriched in the regulation of organelle biogenesis and maintenance, mitochondrial translation, and cell cycle (Fig. 3F). Moreover, these genes were associated with cell cycle related pathways, such as mtorc1 signalling, mitotic spindle, E2F targets, and G2M checkpoint (Fig. 3G). To further confirm the relationships between these genes and EZH2, we constructed a PPI regulation network. The PPI results indicated that 7 proteins, DY310, FBXL19, CENPA, KAT8, CCNA2, NRAS, and CALR, directly interact with EZH2 (Fig. 3H). This information supported the notion that EZH2/H3K27me3 might directly suppress these genes to regulate the cell cycle during CTD therapy for liver cancer.

Prediction of H3K27me3-related potential therapeutic targets of cantharidin in HCC

We further searched disease targets of HCC and retrieved 5875 targets in the disease target database. The Venn diagram showed 22 co-target genes between the cantharidin/H3K27me3-related targets in HepG2 cells and the targets of liver cancer (Fig. 4A). Among them, CCNA2, CENPA, CPE, BRCA2, RFC3, B9D1, RPGR1P1L, NF2, and PSME3 exhibited significantly elevated expression (fold change > 1.5) in the tissues of patients with HCC compared with the normal tissues (Fig. 4B). Moreover, their protein levels were also increased in the tissues of patients with HCC (Fig. 4C). Among of them, CCNA2, CENPA, BRCA2, RFC3, PSME3, ENSA and TERF2 were significantly decreased (log2FC<-1) in CTD treated HCC cells (Fig. 4D). Moreover, the ChIP-seq results showed that H3K27me3 could be enriched in the promoter regions of these dramatically downregulated genes (Fig. 4E). Thus, we hypothesized that the EZH2/H3K27me3 cascade acts as a therapeutic target of CTD-related liver cancer treatment.

Fig. 4figure 4

H3K27me3 related cantharidin treatment target prediction analysis in HCC. (A) Venn diagram shows the genes of liver carcinoma and H3K27me3 related cantharidin treatment targets. (B) The expression of these targets in LIHC. Red represents high expression; blue represents low expression. P < 0.05 is considered significant. (C) Proteins expression of significantly up-regulated genes in LIHC. (D) These genes expression in cantharidin treated HepG2 cell. The * P < 0.05 was regarded as statistically significant. (E) H3K27me3 is dramatically enriched in these genes promoters based on the ChIP-seq results in HepG2

Cantharidin regulated chemokine-related gene expression in liver cancer cells

Further GSEA analysis was performed on the overall expression data of CTD-related genes. The significantly enriched gene sets that positively correlated with CTD were the chemokine biosynthetic and chemokine metabolic modules (Fig. 5A, B). Then, we obtained chemokine- and chemokine receptor-related genes from the GSEA database. The Venn diagram shows that 36 chemokines and 22 chemokine receptors were differentially expressed in cantharidin-treated HepG2 cells (Fig. 5C). Notably, we found that CXCL1/2/3/8 and CCL20/21/24/26 were significantly upregulated, but SEMA3/4/6 were downregulated after CTD treatment (Fig. 5D). Additionally, the levels of CX3CR1, CXCR1, CCRL2, CCR7, CXCR4 and other chemokine receptors with atypical structures were increased and those of five members of the PLXN chemokine receptor family were dramatically decreased in the CTD treatment group (Fig. 5E). Furthermore, the correlations of EZH2 and chemokine were evaluated in HCC. As shown in Fig. 5F, EZH2 expression was significantly associated with that of several chemokines and chemokine receptors, including CXCL6, CCL7, CCL8, CCL22, CXCL10, CCL27, CCL18, CCL16, CXCR4, CCR8, CCR10, CCR1, and CCR7, in HCC.

Fig. 5figure 5

The cantharidin treatment affects chemokine related immune response. (A, B) GO functional annotation shows that cantharidin treatment is positively associated with chemokine biosynthetic and metabolic processes based on GSEA analysis. (C) Venn diagram shows that the differently expressed genes of chemokine and chemokine receptor in the cantharidin treated HepG2 cell. (D, E) The heatmap shows the expression of chemokine (D), and chemokine receptor (E) in the differently expressed genes of cantharidin treated HepG2 cell. (F) The heatmap shows the association of chemokine and chemokine receptor with EZH2 in LIHC, * P < 0.05 was regarded as statistically significant

In addition, the expression of CXC and CCL chemokines involved in the immune response was elevated. We found that the levels of the CTD-related chemokines CXCL1/2/3/8 and CCL20/21/24/26 were significantly correlated with the stromal score and immune score (Fig. 6A, B). Moreover, the relationships between differentially expressed chemokines and immune cell infiltration were evaluated in LIHC. Most chemokines and chemokine receptor were positively associated with immune cells infiltration, including CD4 + cells, NK cells, macrophages, and Treg cells (Fig. 6C, D). Among these genes, the chemokines SEMA4A, SEMA4D, CCL13, and CCL21 and the chemokine receptors C5AR1, PLAUR, CCRL2, FPR2, PTAFR, CCR7 and CXCR4 may be critical genes for immune cell infiltration (Cor > 0.48). Integrated analysis with the transcriptome data indicated that CTD-related DEGs were associated with several immune response signalling pathways, such as MAPK, PI3K-AKT, NF-kappa B, and HIF-1α pathways. Therefore, CTD likely inhibits the progression of HCC by affecting chemokines involved immune cell trafficking and immune signalling responses.

Fig. 6figure 6

Cantharidin related chemokines were involved in tumor immune. (A, B) The correlations between CXC (A) and CCL (B) chemokines and immune score, stromal score, and ESTIMATE score. (C, D) The correlations between the expression of cantharidin related chemokines (C), chemokine receptors (D) and the immune infiltration levels of 25 immune cells, which were analyzed by using the GSCA tools. Purple represents a positive correlation; green represents a positive correlation. * P < 0.05 and # FDR < 0.05 were regarded as statistically significant

Cantharidin inhibits tumour growth and enhances antitumour immunity in vivo

To further confirm the antitumour function of CTD in HCC, we investigated the effect of CTD on tumour growth in vivo. The results showed that tumour volume and weight were significantly decreased in mice treated with a high concentration of CTD and 5-FU (Fig. 7A, B). The tumour inhibition rate of CTD was similar to that of 5-FU (Fig. 7B). However, the CTD group did not exhibit significantly altered total body weight compared with the 5-FU group (Fig. 7C). Moreover, the HE staining results showed that the number of cells was dramatically reduced and that the cell were loosely arranged. The numbers of necrotic and apoptotic cells were increased in the CTD group. As observed the 5-FU group, the number of abnormal vacuoles was increased in the high-concentration CTD group (Fig. 7D). Therefore, CTD could reduce tumour cell growth and promote apoptosis in vivo.

Fig. 7figure 7

The cantharidin improves antitumor immunity in mice. (A) Cantharidin represses the tumor cell growth in vivo.(B) Statistics of the tumor volume, weight, and inhibition rate in mice after therapy with cantharidin and 5-FU. Cantharidin and 5-FU have similar inhibitory effects on tumor growth. Data are presented as the mean ± SD (n = 3), * P < 0.05, ** P < 0.01 vs. the control group. (C) The mice body weight of different model groups. (D) The H&E staining of tumor tissues, including model, 5-FU, CTD high, middle, and low groups. (E-I) Proportion of CD4+ (E), CD8+ (F), Treg (G), and B cell (H) in mice after therapy with cantharidin. To detect the immune cells, blood samples are collected from each treatment group. Cells are stained with anti-CD4, anti‐CD8, anti-CD25, and anti-CD19 antibodies and analyzed by flow cytometry. (I) Statistics of the frequency CD4+, CD8+, Treg, and B cell in mice blood. (J) Elisa assay confirms the expression of immune related genes, IL-2, IL-4, IL-10, TNF-γ, and TNF-α. (K) RT-PCR analyses the expression of PD1 and PD-L1 in the tumor tissues. The * P < 0.05, ** P < 0.01. P value of < 0.05 was regarded as statistically significant. (L) Immunohistochemistry and quantitative analyses the expression of PD1 and PD-L1 in the tumor tissues after therapy with cantharidin

In addition, the CTD-mediated immune response was assessed. Flow cytometry was used to determine the proportions of CD4 + cells, CD8 + cells, Tregs and B cells in the blood of mice (Fig. 7E-H). The results showed that the proportions of CD4+/CD8 + T and B cells were increased after treatment with CTD, while the frequency of Tregs was decreased in the CTD group (Fig. 7I). These results indicated that CTD could inhibit tumour growth by affecting the immune cell distribution to enhance the antitumour immune response in HCC. We then proceeded to further investigate the molecular mechanism of CTD in the immune response. We measured the levels of inflammatory cytokines (TNF-α, IFN-γ, IL-2, IL-4, and IL-10) in the peripheral blood of model mice. As shown in Fig. 7J, IL-2, IL-4, and IFN-γ were elevated, but TNF-α and IL-10 were decreased following treatment with CTD. Next, the expression levels of the immune checkpoint genes PD­1/PD-L1 was were confirmed to investigate the effect of CTD on the immune response. The mRNA and protein expression levels of PD­1/PD-L1 were significantly reduced after CTD treatment (Fig. 7K, L). Therefore, it can be preliminarily concluded that CTD plays a critical role in the regulation of the immune response in exerting its anti-tumorigenic effects.

留言 (0)

沒有登入
gif