Background
Hepatocellular carcinoma (HCC) is a major cause of morbidity and mortality worldwide, with an increasing incidence rate [
1]. Its most important risk factors are hepatitis B virus (HBV) and hepatitis C virus (HCV). Effective treatments against chronic infections with HBV or HCV have significantly reduced the incidence of viral-associated HCC; however, the incidence of HCC associated with metabolic dysfunction-associated steatotic liver disease (MASLD), formerly known as nonalcoholic fatty liver disease (NAFLD) [
2] remains high. Alcoholic steatohepatitis (ASH) is another important risk factor for HCC. Liver fibrosis characterizes disease progression in these chronic liver diseases, and the fibrosis level is a major risk factor for HCC development [
3].
MASLD has been prioritized, and genetic and epigenetic analyses have been performed. Our previous genome-wide association studies and those of others have established a definitive genetic background associated with fibrosis stages [
4‐
6]. Genetic variations affect DNA methylation levels and gene expression, indicating that epigenetic changes are important for MASLD development and progression [
7]. To evaluate the effect of epigenetic status on fibrosis levels, whole hepatic mRNA sequencing was performed, followed by weighted gene co-expression network analysis (WGCNA) [
8]. Two core gene networks associated with MASLD progression were identified, one of which was a scale-free network with four hub genes associated with increased fibrosis and tumorigenesis, while the other was a random network associated with mitochondrial dysfunction. Furthermore, genome-wide hepatic DNA methylation analysis has identified 610 differentially methylated regions (DMRs) associated with fibrosis progression in MASLD [
9]. A method to evaluate DMR networks has been developed, and two DMR networks associated with MASLD progression were detected [
10]. The methylation levels of DMRs in one of the networks (Network 1) decreased with fibrosis progression. Network 1 included genes involved in transcriptional regulation, cytoskeleton organization, and cellular proliferation and is thus potentially associated with tumorigenesis and fibrosis. Meanwhile, Network 2 was potentially associated with metabolic dysfunction. The methylation levels of DMRs in Network 2 increased with fibrosis progression.
Next, the possible occurrence of DMRs associated with fibrosis in MASLD in other liver diseases, such as viral hepatitis, cirrhosis, and HCC, was investigated. Network 2 was observed in viral hepatitis and HCC, with three potential hub genes: fatty acid binding protein 1 (
FABP1), serum/glucocorticoid regulated kinase 2 (
SGK2), and hepatocyte nuclear factor 4 α (
HNF4A) [
11].
FABP1,
SGK2, and
HNF4A methylation levels in cirrhotic livers were higher than those in normal livers, and their methylation levels in HCC samples were comparable to normal levels. Network 1 was not observed in viral hepatitis or HCC; however, it included zinc finger and BTB domain containing 38 (
ZBTB38) [
12] and formin 1 (
FMN1) genes [
13], which may play important roles in the development of fibrosis and cancer. Thus, it is important to investigate the possible relationship between individual DMRs in Network 1 and liver fibrosis and the potential occurrence of HCC in various chronic liver diseases. In this study, the methylation levels of DMRs in Network 1 in the livers of cirrhosis and HCC patients were examined.
Discussion
Liver fibrosis is a major risk factor for HCC development [
1,
2]. Epigenetic studies provide a better understanding of the pathogenesis of liver fibrosis, HCC, and various other diseases. Despite the several studies conducted to this effect [
14‐
16,
21,
22], the changes in methylation levels during the progression from liver fibrosis to HCC remain controversial. The methylation levels of multiple consecutive CpG sites affect gene expression; thus, DMR analysis is more effective for epigenetic research [
23‐
25]. We have previously performed DMR analysis using MASLD livers predisposed to HCC and identified 610 DMRs associated with fibrosis stages [
9]. These DMRs were clustered into two networks (Networks 1 and 2) [
10]. Network 2 contained 430 DMRs and was observed in viral hepatitis and HCC populations with three potential hub genes (
FABP1,
SGK2, and
HNF4A) [
11]. DMR methylation level changes observed in liver fibrosis reverted to normal levels in HCC [
11]; therefore, the DMRs in Network 2 were not considered strong risk factors for HCC regarding liver fibrosis. Network 1 was not observed in cirrhosis and HCC; however, individual methylation changes were considered important for liver fibrosis and HCC. In this study, 12 DMRs were identified and changes in methylation levels were observed in both liver fibrosis and HCC. The methylation levels of these DMRs were confirmed in HCC cell lines. The methylation levels of 24 CpG sites in four genes could distinguish between normal, cirrhotic, and HCC livers; therefore, DMR analysis in liver diseases is important.
We identified 4 genes, namely,
KAZN,
ZBTB38,
FOXK1, and
ZC3H3, the epigenetic changes of which are associated with HCC.
ZBTB38,
FOXK1, and
ZC3H3 directly participated in epigenetic modifications and their transcriptional levels were increased in HCC.
ZBTB38 is a zinc finger transcription factor (ZNF) that is considered a methyl-CpG binding protein [
12,
26]. Its depletion can either promote, reduce, or not affect cell proliferation according to cell type; thus,
ZBTB38 functions as a potential oncogene or tumor suppressor in cancer [
27,
28].
FOXK1 belongs to a family of evolutionarily conserved transcription factors characterized by forkhead DNA-binding domains; it regulates the expression of target genes and contributes to various cellular functions, including the cell cycle, cell growth, proliferation, differentiation, programmed death, metabolism, DNA damage, drug resistance, angiogenesis, and carcinogenesis [
29].
FOXK1 was reported to be upregulated in HCC cells compared with levels in normal liver cells, and its downregulation reduced cell viability [
30], consistent with our findings.
ZC3H3 participates in m6A-methyladenine modification, a post-transcriptional regulatory marker in different RNAs, such as messenger RNAs (mRNAs), transfer RNAs (tRNAs), ribosomal RNAs (rRNAs), circular RNAs (circRNAs), micro RNAs (miRNAs), and long non-coding RNAs (lncRNAs). m6A-methyladenine RNA modification is essential in the initiation and progression of human cancers [
31,
32].
ZBTB38 binds to methylation sites and
ZC3H3 catalyzes m6A methylation of RNA, hence, they are involved in epigenetic changes during tumorigenesis.
FOXK1 plays a role in transcriptional regulation. Therefore, increased transcriptional levels of
ZBTB38,
FOXK1, and
ZC3H3 may affect the epigenetic regulation of several genes, including their own, leading to the development of HCC.
KAZN is a desmosomes component associated with periplakin [
33]. We demonstrated that the methylation levels of
KAZN and its transcriptional levels were decreased in HCC. Indeed, its overexpression stimulates terminal differentiation and reduces cell growth, whereas its knockdown inhibits differentiation and stimulates proliferation [
34]. Notably, the methylation levels of
KAZN were decreased in both cirrhosis and HCC and so were its expression levels as the liver progressed from cirrhosis, dysplastic nodules, and early HCC to advanced HCC. The decreased expression of
KAZN could inhibit differentiation and stimulate the proliferation of liver cells in cirrhosis, leading to the development of HCC.
The methylation levels of CpG sites in the 12 genes could be used to distinguish between normal, cirrhotic, and HCC livers, which was possible with CpG sites in KAZN, ZBTB38, FOXK1, and ZC3H3. The methylation levels of the CpG sites in the 12 or 4 genes could not distinguish between alcoholism- and HCV-derived liver disease. Therefore, these CpG sites could be potentially useful for diagnosing liver cirrhosis and HCC.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.