Systematic Analysis of Long Non-Coding RNA Genes in Nonalcoholic Fatty Liver Disease

Abstract

The largest solid organ in humans, the liver, performs a variety of functions to sustain life. When damaged, cells in the liver can regenerate themselves to maintain normal liver physiology. However, some damage is beyond repair, which necessitates liver transplantation. Increasing rates of obesity, Western diets (i.e., rich in processed carbohydrates and saturated fats), and cardiometabolic diseases are interlinked to liver diseases, including non-alcoholic fatty liver disease (NAFLD), which is a collective term to describe the excess accumulation of fat in the liver of people who drink little to no alcohol. Alarmingly, the prevalence of NAFLD extends to 25% of the world population, which calls for the urgent need to understand the disease mechanism of NAFLD. Here, we performed secondary analyses of published RNA sequencing (RNA-seq) data of NAFLD patients compared to healthy and obese individuals to identify long non-coding RNAs (lncRNAs) that may underly the disease mechanism of NAFLD. Similar to protein-coding genes, many lncRNAs are dysregulated in NAFLD patients compared to healthy and obese individuals, suggesting that understanding the functions of dysregulated lncRNAs may shed light on the pathology of NAFLD. To demonstrate the functional importance of lncRNAs in the liver, loss-of-function experiments were performed for one NAFLD-related lncRNA, LINC01639, which showed that it is involved in the regulation of genes related to apoptosis, TNF/TGF, cytokine signaling, and growth factors as well as genes upregulated in NAFLD. Since there is no lncRNA database focused on the liver, especially NAFLD, we built a web database, LiverDB, to further facilitate functional and mechanistic studies of hepatic lncRNAs.

Keywords: NAFLD; NASH; RNA-seq; gene expression; liver; lncRNA.

Figure 1

Gene expression profiling of NAFL and NASH patients compared to healthy normal weight and obese individuals. (A) Volcano plots comparing different conditions. The numbers of samples are NAFL (n = 15), NASH (n = 16), healthy normal weight (n = 14), and obese individuals (n = 12). Up-regulated genes are colored in red, and down-regulated genes are in green. (B) Tables of differentially expressed genes. The Others category includes any genes other than protein-coding or lncRNAs based on the biotypes provided by the Ensembl database. (C) Venn diagrams of up- and down-regulated lncRNAs.

Figure 2

Gene expression changes in different stages of NASH compared to obese individuals. (A) Tables of differentially expressed genes. The Others category includes any genes other than protein-coding or lncRNAs based on the biotypes provided by the Ensembl database. (B) Volcano plot of early (n = 138) and moderate NASH patients (n = 68) compared to the control (obese individuals; n = 10). Only 163 up- and 12 down-regulated lncRNAs identified in the previous dataset (GEO accession number, GSE126848) are shown in each volcano plot, as indicated in the figure. Up-regulated genes are colored in red, and down-regulated genes are in green.

Figure 3

Expression profiles of commonly up- or down-regulated lncRNAs. Violin plots show counts per million (CPM) values of each sample group for each dataset, as indicated in the figure. The number of samples for each category is indicated in the figure.

Figure 4

Role of LINC01639 in regulation of gene expression in hepatocytes. (A) Silencing of LINC01639 in Huh-7 cells, Error bars represent mean ± S.E.M. *** p < 0.001. (B) Gene expression profile after silencing of LINC01639 in Huh-7 cells. n = three biological replicates. * p < 0.05; ** p < 0.01. (C) NAFLD cellular model. Huh-7 cells treated with a mixture of fatty acids (FA) consisting of 2 mM oleic acid and 2 mM palmitic acid at the ratio of 1:1. The schematic experimental timeline for NAFLD cellular model generation is presented. (D) Gene expression profile of NAFLD cellular model upon silencing of LINC01639 n = three biological replicates. * p < 0.05; ** p < 0.01. nd = not detected.

Figure 5

LiverDB. (A) The top page of LiverDB. (B) An interactive box plot showing the distribution of gene expression values across samples within each biological condition in the selected study. (C) A volcano plot of selected comparative conditions. (D) A heatmap of differentially expressed genes. (E) A pathway enrichment plot displaying the top results from KEGG pathway analysis as a heatmap. (F) An UpSet plot showing the number of differentially expressed genes that are found across study/contrast pairs.