Transcriptome-Wide 5-Methylcytosine Functional Profiling of Long Non-Coding RNA in Hepatocellular Carcinoma

Abstract

Background: Growing evidence indicates that methylation status is associated with the pathogenesis of numerous types of cancers. Among these, hepatocellular carcinoma (HCC) is a deadly disease threatening global human health. Although 5-methylcytosine (m5C) has been identified as an important regulatory modification, its distribution in solid tumors, including HCC, remains unclear. The present study aimed to explore the distribution of m5C in HCC.

Materials and methods: Six pairs of human HCC tissues and adjacent non-tumor tissues were collected to analyze the transcriptome-wide m5C methylation of long non-coding RNA (lncRNA). RNA MeRIP-seq was performed to identify m5C peaks on lncRNA and differences in m5C distribution between HCC and adjacent tissues. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathway analyses were explored to predict the possible roles of m5C.

Results: Using m5C peak sequencing, we observed that a sequence motif was necessary for m5C methylation in HCC lncRNA. Unsupervised hierarchical cluster analysis confirmed that lncRNA m5C methylation occurred more frequently in HCC than adjacent non-tumor tissues. RNA sequencing data demonstrated that more genes were up-regulated by methylation in HCC, while methylation down-regulated more genes in adjacent non-tumor tissues. GO and KEGG pathway analyses revealed that genes having a significant correlation with m5C sites in lncRNA were involved in HCC signaling pathways.

Conclusion: Our results revealed the substantially different amounts and distributions of m5C in HCC compared to adjacent non-tumor tissue. We further predicted the cellular functions in HCC that m5C may participate in to provide evidence implicating m5C lncRNA epigenetic regulation in the tumorigenesis and progression in HCC.

Keywords: hepatocellular carcinoma; lncRNA; m5C; tumorigenesis.

Figure 1

Transcriptome-wide m5C methylation and characters of lncRNA in human HCC. (A) Venn diagram of m5C methylation sites identified in lncRNA from human HCC and adjacent non-tumor tissues. (B) Venn diagram of m5C genes in HCC and adjacent tissues. (C) The sequence motif of m5C sites in human HCC and adjacent non-tumor tissues. (D) Percentage of lncRNA harboring different numbers of m5C peaks in the two tissue types, with the majority harboring only one m5C peak.

Figure 2

m5C levels and distribution of chromosome in human HCC and adjacent non-tumor tissue. (A) A cluster tree at the top indicates the relatedness of the two tissues (x-axis) or methylation levels (y-axis). The heatmap uses a color scale to indicate the relative methylation level at each locus. Each row of colored lines (N = 117) represents the methylation level for each CpG locus: red for hypermethylated and yellow for hypomethylated. Each column (n = 12) corresponds to each tissue. (B) Circos plot showing the distribution of m5C methylation sites on each chromosome. Red represents HCC, blue represents adjacent non-tumor tissue.

Figure 3

The differences in lncRNA expression according to m5C methylation. (A) Differentially expressed lncRNAs in HCC and adjacent non-tumor tissues. Genes up-regulated by methylation are red, and genes down-regulated by methylation are blue. (B) Cumulative distribution of lncRNA expression changes between HCC and adjacent non-tumor tissue for m5C up-regulated genes (red) and m5C down-regulated genes (green), whereas blue represents others.

Figure 4

Gene Ontology (GO) term enrichment analysis of m5C genes in HCC lncRNAs. (A-C) The top 10 GO terms for (A) biological processes, (B) molecular functions, and (C) cellular components were significantly enriched for up-methylated m5C genes in HCC. (D–E) The top 10 gene GO terms of (D) biological processes, (E) molecular functions, and (F) cellular components were significantly enriched for the down-methylated m5C genes in HCC.

Figure 5

KEGG pathway analysis of m5C genes in HCC lncRNAs. (A) Bar plot showing the top 10 enrichment scores of significantly enriched pathways for up-methylated m5C genes in HCC. (B) Bar plot showing the top 10 enrichment scores of significantly enriched pathways for down-methylated m5C genes in HCC.