RNA-seq data analysis of stimulated hepatocellular carcinoma cells treated with epigallocatechin gallate and fisetin reveals target genes and action mechanisms

Abstract

Hepatocellular carcinoma (HCC) is an essentially incurable inflammation-related cancer. We have previously shown by network analysis of proteomic data that the flavonoids epigallocatechin gallate (EGCG) and fisetin (FIS) efficiently downregulated pro-tumor cytokines released by HCC through inhibition of Akt/mTOR/RPS6 phospho-signaling. However, their mode of action at the global transcriptome level remains unclear. Herein, we endeavor to compare gene expression alterations mediated by these compounds through a comprehensive transcriptome analysis based on RNA-seq in HEP3B, a responsive HCC cell line, upon perturbation with a mixture of prototypical stimuli mimicking conditions of tumor microenvironment or under constitutive state. Analysis of RNA-seq data revealed extended changes on HEP3B transcriptome imposed by test nutraceuticals. Under stimulated conditions, EGCG and FIS significantly modified, compared to the corresponding control, the expression of 922 and 973 genes, respectively, the large majority of which (695 genes), was affected by both compounds. Hierarchical clustering based on the expression data of shared genes demonstrated an almost identical profile in nutraceutical-treated stimulated cells which was virtually opposite in cells exposed to stimuli alone. Downstream enrichment analyses of the co-modified genes uncovered significant associations with cancer-related transcription factors as well as terms of Gene Ontology/Reactome Pathways and highlighted ECM dynamics as a nodal modulation point by nutraceuticals along with angiogenesis, inflammation, cell motility and growth. RNA-seq data for selected genes were independently confirmed by RT-qPCR. Overall, the present systems approach provides novel evidence stepping up the mechanistic understanding of test nutraceuticals, thus rationalizing their clinical exploitation in new preventive/therapeutic modalities against HCC.

Keywords: ADAM, a disintegrin and metalloproteinase with thrombospondin motifs; ADAMTS9, ADAM metallopeptidase with thrombospondin type 1 motif 9; CLIC3, Chloride Intracellular Channel 3; CTGF, Connective Tissue Growth Factor; DEGs, differentially expressed genes; DMSO, dimethyl sulfoxide; ECM, extracellular matrix; EGCG, epigallocatechin gallate; EMT, epithelial to mesenchymal transition; Epigallocatechin gallate; FIS, fisetin; Fisetin; GO, Gene Ontology; Gene Ontology; HCC, hepatocellular carcinoma; HSPA2, Heat Shock Protein Family A (Hsp70) Member 2; HSPB1, Heat Shock Protein Family B (Small) Member 1; Hepatocellular carcinoma; MEM, minimum essential medium; MMP11, Matrix Metallopeptidase 11; MMP9, Matrix Metallopeptidase 9; MMPs, matrix metalloproteinases; PDGFRB, Platelet Derived Growth Factor Receptor Beta; RNA-sequencing; RT-qPCR, reverse transcription-quantitative real time PCR; Reactome Pathways; SD, standard deviation; SEM, standard error of mean; SERPINE1, Serpin Family E Member 1; STIM, stimulated; TF, transcription factor; Transcription factors.

Conflict of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

© 2020 The Authors.

Figures

Graphical abstract

Fig. 1

Chemical structure of EGCG (A) and FIS (B).

Fig. 2

Volcano plots illustrating DEGs in EGCG-stimuli versus DMSO-stimuli (A) and FIS-stimuli versus DMSO-stimuli (B). Red and green dots represent up- and down-regulated genes, respectively; grey dots correspond to non-statistically significant altered genes. The horizontal dashed line indicates a statistical threshold corresponding to an adjusted p-value of 2 Fold Change, y-axis: p-value in negative log10 scale. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Venn diagram depicting common DEGs (adjusted p-value 

Fig. 4

Heatmap visualizing the hierarchical clustering of gene expression for the 695 common genes. Each row represents one of the common genes. Columns represent expression averages of replicates for each investigated group. Expression values are normalized counts from DESeq2, transformed by the variance-stabilizing transformation and mean-centered. Red color indicates relative over-expression, while green color indicates relative under-expression. DMSO-STIM: DMSO-treated stimulated cells, EGCG-STIM: EGCG-treated stimulated cells, FIS-STIM: FIS-treated stimulated cells. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

GO enrichment analysis associating the co-modified DEGs with statistically significant biological processes (corrected p-value 

Fig. 6

Reactome Pathway enrichment analysis associating the co-modified DEGs with statistically significant biological pathways (corrected p-value 

Fig. 7

RT-real time qPCR validation of RNA-seq data for selected genes. RT-qPCR data represent mean ± SEM (n = 3–4, *p-value 2 scale) derived from (A) EGCG-STIM and (B) FIS-STIM versus DMSO-STIM.