Currently, the pathogenesis of NAFLD is still unclear and the therapeutic treatments are also limited. In the present study, we identified 45 intersected DEGs between HC-SS group and HC-NASH group, and respectively performed GO and KEGG pathway enrichment analysis to explore the potential effects of these DEGs in NAFLD. The results showed that the GO enrichments were involved in fatty acid metabolism, mesenchyme, extracellular matrix, cell adhesion, and inflammatory and immune response, which also played important roles in tumorigenesis. KEGG analysis showed that the DEGs were primarily enriched in TGF-beta signaling pathway, PI3K-Akt signaling pathway, pathways in cancer, MicroRNAs in cancer, MAPK signaling pathway, and Jak-STAT signaling pathway. Both the results of GO and KEGG analysis all pointed to tumorigenesis. Meanwhile, the overlapping pathways between SS and NASH were PI3K-Akt signaling pathway and pathways in cancer, which suggesting that the two pathways could be an important therapeutic target for NAFLD. Our findings were also consistent with previous reports[9, 11].
WGCNA is a well-established method for studying biological networks and diseases[12]. Due to NAS and steatosis were the two main pathologic indicators in the estimation of NAFLD, we tried to find out the DEGs related to the NAS and steatosis. We totally identify 25 DEGs related to the NAS and steatosis, and PPI network analysis was performed to explore the hub genes in the pathogenesis and progression of NAFLD. Eventually, we determined 10 hub genes (Down-regulated genes: MYC, TGFB3, ADAMTS1, THBS1, RASD1, PCDH20; Up-regulated genes: MAMDC4, CYP7A1, GINS2, and PDLIM3) related to NAS and steatosis.
HCC is the fourth-leading cause of cancer death worldwide, and the morbidity of NAFLD-related HCC is predicted to increase dramatically by 2030, with increases of 82%, 117%, and 122% from 2016 in China, France, and the USA, respectively[3, 13]. Therefore, we explore whether these ten hub genes were associated with the progression in HCC in TCGA database. We found that CYP7A1, GINS2, and PDLIM3 were significantly up-regulated, and MYC, MAMDC4, ADAMTS1, THBS1, and RASD1 were significantly down-regulated in HCC tumor samples compared to normal samples. Surprisingly, we also found that CYP7A1/GINS2/PDLIM3 were correlated with HCC prognosis.
CYP7A1, catalyzing the first and rate-limiting step in the classic bile acid synthesis pathway, has been shown to be involved in the lipid metabolism[14]. Deficiency of CYP7A1 caused by homozygous deletion mutations can inhibit the production of bile acids, leading to the accumulation of cholesterol in liver, reducing LDL receptors and elevating LDL cholesterol[15]. However, CYP7A1 was up-regulated in SS and NASH group compared with HC group in our study. Previous studies shown that CYP7A1 and its associated cholesterol processes were adversely regulated in NAFLD[16], and glucose stimulates CYP7A1 transcription in human hepatocytes[17]. Therefore, up-regulating CYP7A1 in NAFLD may be the consequence rather than cause of disease[18]. In addition, increased CYP7A1 expression and bile acid synthesis ameliorated hepatic inflammation and fibrosis, which proved its effects of anti-tumor[19].
GINS2, a member of GINS family, plays a crucial role in DNA duplication and is highly expressed in various types of cancer[20, 21]. However, very little research can be found about GINS2 in the liver, especially in NAFLD. Previous bioinformatics studies indicated that GINS2 might be the hub genes in the development of NASH to HCC, and predicted poor prognosis in HCC, but there was no further experiment to verify its effects on NAFLD[22, 23].
PDLIM3, highly expressed in skeletal and cardiac muscle, has been suggested to play a pivotal role in myocyte stability, signal transduction, and mechanical signaling, especially in growth and remodeling processes[24]. Interestingly, PDLIM3 was firstly screened out for a new hub gene in the pathogenesis of NAFLD and was associated with the prognosis of HCC. PDLIM3 was highly related to epithelial-mesenchymal transition (EMT) in GSCALite database, which might partially reveal its effects in the pathogenesis in NAFLD and HCC. More future studies are needed to gain more insights about PDLIM3.
In conclusion, we analyzed two public datasets to identify DEGs among HC, SS and NASH. GO and KEGG pathway analysis revealed that the pathogenesis and progression of NAFLD was highly associated with tumorigenesis. Finally, we screened out 10 hub genes related to NAS and steatosis, and three of them were correlated with HCC prognosis. Although further validation is still needed, we provide useful and novel information to explore the potential candidate genes for NAFLD prognosis and therapeutic options.