Identifying competing endogenous RNA regulatory networks and hub genes in alcoholic liver disease for early diagnosis and potential therapeutic target insights

Introduction

Recent social progress and lifestyle changes are associated with a rising incidence of alcoholic liver disease (ALD) worldwide. ALD is mainly caused by excessive alcohol consumption and the early symptoms of liver dysfunction include hepatomegaly, liver fibrosis, and cirrhosis. If not treated early, these conditions can progress into serious conditions, including liver dysfunction and liver cell carcinoma [1]. According to the latest Global Burden of Disease Project report, released in 2016, 1.2569 million deaths were associated with cirrhosis and chronic liver disease, with alcohol consumption accounting for 27% of these cases. Furthermore, liver cancer is an important health challenge since it is associated with a mortality rate of 245,000 and accounts for 30% of all liver cancer-related deaths. Recent Chinese epidemiological surveys have shown that in some cities in the Zhejiang and Liaoning provinces, the prevalence of ALD ranges from 4.34% to 6.1% and continues to rise annually. Social progress has been associated with a rise in excessive alcohol consumption, which has increased cases of ALD. Currently, the most effective way of preventing liver disease is by abstaining from alcohol. However, only 27% of cases achieve complete recovery of the liver function, with most cases progressing into hepatitis, cirrhosis, or liver cancer [2].

ALD pathogenesis is very complex and, currently, there are no effective drugs and therapeutic targets for its treatment. Thus, there is an urgent need to identify key ALD treatment targets. Therefore, this study downloaded and analyzed publicly available sequencing data from NCBI of liver tissues from ALD patients [3]. Bioinformatics analysis was used to identify the differentially expressed genes that may be associated with the progression from alcoholic fatty liver to alcoholic hepatitis and cirrhosis. Moreover, hub genes were identified, and their mechanisms were evaluated to elucidate any potential therapeutic targets for ALD [4].

The recent increase in the use of chip technology and high-throughput sequencing [5, 6] has markedly increased the amount of information that researchers can obtain, thereby allowing improved examinations of the structure, function, and properties of organisms. For example, high throughput sequencing can measure the expression of 20,000 to 30,000 genes simultaneously [7]. Furthermore, the rapid development of second-generation sequencing technology offers a more precise analytical tool. This technique can be used to more accurately reveal the genetic characteristics of a species and identify specific genetic loci. Subsequently, researchers are able to better understand the natural environment, which has significant implications in medicine. For instance, using second-generation sequencing, researchers at the University of Nijmegen identified a Schinzel–Giedion syndrome-associated pathogenic mutation with diagnostic significance [8]. Schinzel–Giedion syndrome is a highly challenging disease with severe complications, including intellectual disability, malignant tumors, and congenital anomalies. Ultimately, using genetic sequencing [9, 10], Schinzel–Giedion syndrome can be rapidly diagnosed, which significantly shortens the time required for treatment and significantly improves patient prognoses.

In October 2021, Professor Ming He’s team used RNA sequencing to discover that SIRT2 knockout could effectively prevent alcohol-induced production of acute-phase response proteins. Subsequently, in vivo and in vitro experiments have shown that lipocalin-2 (LCN2) can effectively prevent the occurrence of ALD and modulate the antioxidant and immunoregulatory functions of SIRT2. This study identified the SIRT2–C/EBPβ–LCN2 axis as a novel regulatory mechanism underlying ALD [11], alongside its therapeutic potential against ALD. A study by Shirish Barve used 16S rRNA sequencing and whole-genome shotgun metagenomic analysis to verify a reduction in the acetyl-CoA pathway, which was induced by alcohol. Here, they identified a decrease in two key genes that encode terminal enzymes in the conversion of butanol-CoA to butyrate, yet not for (butanol-CoA: acetyl-CoA transferase) and buk (butyrate kinase), in specific taxonomic groups of butyrate-producing bacteria. Future studies should seek to identify any related bacterial species and leverage the results on the functions affected by alcohol to develop novel ALD treatment strategies [12]. Furthermore, a team led by Li-Rui Wang discovered that an alcohol-containing diet significantly downregulated the expression of the AHR gene in intestinal epithelial cells. To investigate the roles and mechanisms of this phenomenon in ALD development, the researchers developed an intestine-specific AHR knockout mouse and surprisingly, found that intestine-specific AHR knockout significantly exacerbated alcohol-induced liver injury, fat accumulation, and liver inflammation. Subsequent studies using 16S rRNA sequencing, metabolomics, and metagenomics, further identified another key therapeutic target for ALD [13].

Advances in RNA-sequencing technology have led to the discovery of a large number of miRNAs, lncRNAs, and circRNAs [14]. Indeed, miRNAs and lncRNAs are the main types of non-coding RNA in animals and plants, whereby miRNAs post-transcriptionally suppress gene expression by pairing with target mRNAs [15]. LncRNAs are a class of non-coding RNAs, which consist of more than 200 nt in length [16], and they are thought to regulate protein-coding genes in several ways. Recently, studies have indicated that lncRNAs can influence miRNA activity through several mechanisms: 1: by acting as miRNA sources; 2: by competitively interacting with miRNAs; 3: by adsorbing to miRNAs, thereby affecting gene expression [17]. Since miRNAs and lncRNAs, which are crucial for human health [18, 19], may also influence the occurrence and progression of liver diseases, they offer significant therapeutic and diagnostic potentials [20, 21].

Here, we used bioinformatics to identify the hub genes that are differentially expressed during ALD progression from mild to severe and to predict their target miRNAs and lncRNAs. Functional enrichment analysis was used to determine the roles of the predicted hub genes, miRNAs, and lncRNAs, as well as to elucidate any potential ALD therapeutic targets.

Results

Discussion

Alcoholic liver disease (ALD) is one of the most common liver diseases worldwide. Recently, the incidence of alcohol-induced liver damage has increased, including among younger age groups, especially young adults, who account for about 70% of the total cases. ALD is mainly caused by excessive alcohol-induced liver cell damage and the disruption of liver function [22]. The growing rate of alcohol abuse has contributed to the rising incidence of ALD. Both chronic and acute alcohol abuse can result in liver disease [23]. Thus, a better understanding of ALD pathogenesis, as well as effective therapeutic and diagnostic targets are urgently needed.

Multiple studies have associated ALD development with various signaling pathways, alcohol-mediated epigenetic modifications, and alterations in miRNA and lncRNA [24]. In this study, we used bioinformatics analyses to identify ALD-associated mRNAs, miRNAs, and lncRNAs, and to construct regulatory networks. We also validated their differential expressions using tissue pathology data.

By analyzing mRNA, miRNA, and lncRNA expressions in normal liver tissue, alcoholic fatty liver, alcoholic hepatitis, and alcoholic cirrhosis samples, we constructed a regulatory network and identified the potential molecular mechanisms underlying ALD development through an mRNA–miRNA–lncRNA network [25]. Analysis of the GSE103580 and GSE164760 datasets revealed that 9738 differentially expressed genes (DEGs) were upregulated and 11112 DEGs were downregulated in association with ALD. To identify the DEGs that change during the stage progressions from alcoholic fatty liver to alcoholic hepatitis (mild) and cirrhosis (severe), we analytically expressed the DEGs to a Venn diagram intersection, which identified 64 upregulated and 39 downregulated DEGs. PPI network analysis of the interaction between these DEGs and the hub genes was performed using Cytoscape. This analysis identified 10 upregulated hub genes (TP53, TCP1, VCL, TUBAIA, BECN1, RAP1A, SFPQ, TSC1, RAB5C, and MYD88) and 10 downregulated hub genes (RAP1A, KLRD1, TP53, NEDD8, SMARCA2, OBSL1, UBE2L3, MRPS15, NDUFA13, and YME1L1). Functional enrichment analysis of these hub genes revealed that they were enriched in extracellular matrix formation, cellular organelle formation, mitochondrial autophagy, and the regulation of the NF-κB signaling pathway. The hsa-miR-92a-3p, hsa-miR-150-5p, hsa-miR-193b-3p, and hsa-miR-520f-3p miRNAs were all identified as potential key factors in ALD progression, thereby highlighting their potential as biomarkers in the early diagnosis of ALD. Additionally, we predicted and analyzed target miRNAs and central regulatory genes to gain insight into the molecular functions of these dysregulated miRNAs in ALD. GO and KEGG enrichment analyses revealed that the mRNA targets of the differentially expressed miRNAs are involved in crucial biological processes, thereby highlighting their importance in major cellular processes. Pathway enrichment analysis revealed that the NF-κB and p53 signaling pathways were involved in ALD pathogenesis. These findings reveal the potential molecular mechanisms underlying ALD.

In summary, this study provides valuable insights into the roles and functions of differentially expressed miRNAs in ALD and reveals their potential involvement in ALD development and progression. Moreover, the identified hub genes are potential ALD biomarkers and form potential therapeutic targets. However, further research is needed to determine the possibility of targeting these hub genes in ALD.

The roles of TP53 (p53) in ALD are the same as in other liver diseases and cancers [26], which include DNA damage response, cell cycle regulation, apoptosis, and antioxidative stress. Metabolic byproducts, such as acetaldehyde and acetate, which are generated during alcohol metabolism can cause DNA damage. By detecting and responding to DNA damage signals, TP53 contributes to the maintenance of genomic stability. When DNA damage occurs, TP53 can prevent the proliferation of damaged cells, promote repair, or induce apoptosis to prevent tumorigenesis [27]. TP53 also plays an important role in cell cycle regulation, whereas alcohol may disrupt cell cycle progression, thereby causing abnormal cell proliferation. TP53 functions during the activation of the cell cycle checkpoint to inhibit cell cycle progression and abnormal cell proliferation. TP53 also regulates apoptosis. In ALD, impaired apoptosis may cause abnormal cell survival and liver damage. Thus, by inducing apoptosis, TP53 can eliminate abnormal cells, thereby preventing their transformation into cancer cells. TP53 can also counter alcohol-induced oxidative stress by modulating the expression of antioxidant enzymes. The byproducts of alcohol metabolism can cause oxidative stress, thereby disrupting cellular redox homeostasis [28]. TP53 can alleviate this stress by inducing the expression of antioxidant enzymes, hence, mitigating oxidative damage. Although changes in TP53 expression might indicate ALD pathogenesis, further research is needed to determine its precise mechanisms in relation to ALD.

Other genes may play important roles in the development of ALD. For example, Ras-proximate-1 (RAP1A), a member of the Ras superfamily, regulates leukocyte adhesion, extracellular matrix attachment, and the release of inflammatory factors [29]. Thus, RAP1A may regulate the interaction between liver cells and inflammatory cells in ALD, as well as modulate the intensity and extent of the inflammatory responses. RAP1A also regulates hepatic stellate cell activation and collagen deposition, thereby influencing the degree of liver fibrosis. It may also participate in the regulation of oxidative stress responses by influencing mitochondrial function, ROS production, and the expression of antioxidant enzymes. T-complex protein 1 (TCP1), also known as T-complex protein-1 ring complex (TRiC) or chaperonin containing TCP1 (CCT), is a molecular complex formed by multiple subunits, which plays a role in protein folding by influencing the unfolding and refolding of proteins. Alcohol-induced liver damage and oxidative stress may cause abnormal protein folding and degradation, thereby affecting liver cell function. Therefore, during ALD, TCP1 may be involved in protein repair and protection from damaged proteins, thereby maintaining normal liver cell function. Through comprehensive functional analyses, we have identified miRNAs and their key target genes, which have potential applications as ALD biomarkers or therapeutic targets.

In addition, although the inclusion of liver cancer data in the validation process resulted in the disappearance of some genes previously identified as hub genes, these genes can still be considered crucial genes in the early regulation of alcoholic liver disease. Despite changes in the expression of some hub genes in the liver cancer data, they can still be regarded as pivotal biomarkers in the early regulation of alcoholic liver disease. This suggests that these genes may exhibit distinct expression patterns and biological functions at different disease stages. Therefore, these genes continue to hold significant research and clinical value in the early diagnosis and treatment of alcoholic liver disease. Further investigation is warranted to elucidate the roles and regulatory mechanisms of these genes in different disease states.

In conclusion, the bioinformatic analyses conducted in this study have provided valuable insights into the roles and functions of differentially expressed miRNAs in the progression of alcoholic liver disease (ALD). The identification of four key miRNAs with regulatory roles at various stages of ALD development holds significant promise for improving the accuracy of early diagnoses of ALD. Furthermore, the identification of hub genes associated with ALD progression presents potential therapeutic targets for the treatment of ALD.

The differential gene expression analysis revealed a comprehensive understanding of the molecular mechanisms underlying ALD, shedding light on the potential roles of TP53, RAP1A, and TCP1 in ALD pathogenesis. These findings have the potential to significantly impact the development of diagnostic and therapeutic strategies for ALD. The functional enrichment analysis of the hub genes has highlighted their involvement in crucial biological processes, cellular components, and molecular functions, providing a deeper understanding of their roles in ALD.

Moreover, the construction of the miRNA–lncRNA network targeting hub genes has unveiled potential regulatory mechanisms underlying ALD, offering new avenues for further research and potential therapeutic interventions. The validation of hub gene expression using tissue pathology data has further strengthened the credibility of the findings, emphasizing the importance of these hub genes in the development of ALD.

Overall, the findings of this study have the potential to advance the field of ALD research by providing a foundation for the development of novel diagnostic biomarkers and therapeutic targets. Further research and validation of the identified miRNAs and hub genes are warranted to fully elucidate their roles in ALD pathogenesis and to determine their potential as targets for therapeutic intervention. Further research is needed to fully establish the therapeutic target value of these key genes in ALD.

Materials and Methods

Supplementary Materials

Supplementary Table 1

Supplementary Table 2