Introduction
Nonalcoholic fatty liver disease (NAFLD) is a broad-spectrum term used to generalize non-alcoholic fatty liver simple steatosis and non-alcoholic steatohepatitis (NASH). Simple steatosis is the initial stage within the spectrum of NAFLD, which progresses to NASH and increases the risk of developing fibrosis, cirrhosis, and hepatocellular carcinoma [
1,
2]. NAFLD has affected 10 to 48 percent of the general population in different countries of the world [
3‐
5]. As the epidemics of obesity and type 2 diabetes mellitus (T2DM), the incidence and prevalence of NAFLD which is the most common cause of adult liver disease across various countries are growing over time [
3].
It is well-accepted that non-alcoholic hepatic steatosis is primarily caused by insulin resistance (IR) [
1,
6‐
8]. Mitochondrial dysfunction has been shown to be related with the occurrence of hepatic IR of NAFLD patients [
9]. In addition, free fatty acid (FFA)-induced lipid toxicity and the inflammatory response are two main mechanisms that cause hepatic IR in NAFLD [
7]. The mitochondrial dysfunction also contributes to fat accumulation and liver damage by oxidative stress, increases reactive oxygen species generation. NAFLD is also closely related to inflammation and oxidative stress [
10,
11]. It is well recognized that oxidative stress causes cellular dysfunction and is a pathophysiologic cause of NAFLD. Oxidative stress develops when the production of reactive oxygen species surpasses the ability of antioxidants to detoxify them, which causes harm to normal lipid metabolism [
12]. However, there is currently no established single-drug or combined therapy. There are numerous studies on the importance of nano-antioxidants in the prevention and treatment of liver disorders, including hepatic ischemia–reperfusion injury, viral hepatitis, hepatocellular cancer and viral fibrosis [
13‐
16]. Other treatments of targeting reactive oxygen species have been shown to play an protective to injury of rat hepatocytes [
17,
18]. Furthermore, an appealing therapeutic approach for the management of NAFLD involves targeting mitochondria. Antioxidants targeting mitochondrial
•O
2−/H
2O
2, for instance, have the potential to treat NAFLD by counteracting liver inflammation [
19].
The increased fatty acids absorption, increased de novo lipogenesis, and impairment in export and oxidation of fatty acids induced intracellular lipid accumulation and lipid droplets (LDs) formation, which is the histological feature of NAFLD in the liver [
20]. Genetic studies also provided clues to the study of hepatic lipogenesis of NAFLD. Genome-wide association studies (GWAS) have shown that genetic and epigenetic factors acted as regulators in NAFLD. A major genetic factor I148M PNPLA3 variation, a triglyceride hydrolase, increased susceptibility to NAFLD [
21]. Transcriptome and single-cell sequencing studies have also been applied to study the pathogenesis of NAFLD. Numerous signaling pathways and genes have been shown to be involved in the pathogenesis of NAFLD. TLR4-dependent inflammatory factor release pathway, PI3K/Akt signaling pathway and TGF-β/SMAD3-signaling pathway play an important roles in the occurrence of NAFLD [
22‐
24]. Besides,
CYP2E1 has been shown to be involved in LDs formation in NAFLD. Overexpression of
CYP2E1 in vivo or in vitro, the development of hepatic IR was promoted by JNK activation [
25,
26].
CYP3A activity was decreased in hepatoma cell models, high fat diet-induced mice and NAFLD patients [
27]. Meanwhile, variants of
GCKR, TM6SF2 and
MBOAT7 genes have been shown significant contributions to the occurrence of NAFLD [
28]. However, the pathogenesis of NAFLD is not yet completely understood.
In the current study, we performed bioinformatics analysis to identify 134 overlapping genes by analyzing the differential expression gene from the GSE159676 datasets and our RNAseq data and then identified five MCODE (Molecular Complex Detection) modules and 29 hub genes. In one of these five modules, four of six hub genes has been reported to be associated with lipid metabolism [
29‐
32]
. These four hub genes were densely connected with Mitogen-activated protein kinase kinase kinase 4 (
MAP3K4). However, the role of
MAP3K4 in lipid metabolism of NAFLD has not been reported. The present study focused on
MAP3K4 and explored its role and potential mechanism in NAFLD
.
Materials and methods
Clinical data and samples
Nine control samples and seven patients with histologically confirmed NAFLD were included. Control liver samples were collected from normal liver tissues of hepatic hemangioma resection. Histological features were assessed using NAFLD activity score (NAS) [
33]. Two professional pathologists who were blinded to the clinical data performed a histological assessment of liver specimens. The ethics committee of the Second Xiangya Hospital of Central South University granted ethical permission for this experiment. Written informed consent was obtained from all participants. The clinical characteristics between NAFLD patients and controls were displayed in Additional file
1.
Hepatic transcriptome
Hepatic transcriptome analysis was performed by the Next Generation Sequence of RNA extracts from liver specimens as previously described [
34‐
36]. Briefly, total RNA was isolated using the Illustra RNAspin Mini Kit (GE Healthcare, United States). cDNA libraries and sequencing library were generated by TruSeq Stranded Total RNA kit (Illumina Inc.) according to the manufacturer’s instructions. RNA sequencing was performed via Illumina HiSeq2000 high-throughput sequencing system (Illumina Inc.). Adaptor sequences and low-quality reads were removed. Quality control of the raw fastq files was performed using the software Fastq-mcf v1.0.3-r152. Reads were mapped to the annotated Human genome (Human GRCh38/hg38) in Ensembl database. Subsequently, the gene expression profiling, differentially expressed genes, and differentially expressed transcripts were calculated.
Immunohistochemistry
Liver tissues were fixed with paraformaldehyde and embedded in paraffin. The paraffin-embedded tissues were cut into 5 µm slices. After deparaffinization, the slices were incubated 30 min with 10% of goat serum in TBS (50 mM Tris–HCl, pH 7.4, 150 mM NaCl) at room temperature. Subsequently, the tissues were incubated with anti-MAP3K4 antibody (Sigma, United States). After washed with TBS, incubated with the HRP-conjugated secondary antibody (Jackson Immunoresearch, United States), Then, sections were treated with DAB. All stained sections were observed and imaged on Olympus BX41microscope (Olympus, Tokyo, Japan).
RNA interference
The oligos complementary for MAP3K4 RNA and nontarget oligos were synthesized by Gene Pharma Co. Ltd. (Sangon Biotech, China). The target sequence siRNA #1 for human MAP3K4 target sequence is 5'-GAGTCCTGAATCTGATCTAGA-3', and siRNA #2 target sequence is 5'-GTCCAGCAGATCGTTTAAAGT-3'. According to the instructions provided by the manufacturer, HepG2 cells were transfected by the oligos for MAP3K4 interference and control oligo via Lipofectamine 2000 Transfection Reagent (Invitrogen, USA).
BODIPY and immunofluorescence staining
The BODIPY staining was performed as previously described [
35]. After 36 h transfection with siRNA, the cells were treated with oleic acid (OA) + palmitic acid (PA) for 12 h [
34]. The cells were washed with PBS twice and fixed in 4% of paraformaldehyde for 15 min. Then, the cells were permeabilized with PBST (PBS + 0.1% Triton X-100). Subsequently, the cells were blocked with 5% BSA/PBS for 30 min and washed with PBS and incubated in BODIPY 493/503 staining solution (Sigma, United States) for 15 min at 37 °C. Finally, the nuclei were counterstained with DAPI (Invitrogen, United States) for 2 min. All stained sections were observed and imaged on a Zeiss 880 confocal microscope.
Immunoblotting
CGI-58 and Plin-2 antibodies were obtained from Abcam. β-actin, Phospho-JNK (Thr183, Tyr185), JNK, Phospho-S505-cytosolic phospholipase A2 (cPLA2), and cPLA2 antibodies were purchased from Cell Signaling Technology Co. Lysis buffer (2% SDS, 62.5 mM Tris–HCl pH 6.8, and 10% glycerol) was used to lyse the cells. Subsequently, 20 μg protein of each sample was electrophoresed and separated by SDS-PAGE. Then, the proteins were transferred to PVDF membranes (Millipore Corporation, United States). The membrane was incubated in 5% of defatted milk for 1 h at room temperature. The membrane was incubated with the primary antibodies overnight at 4 °C. After incubation, the membrane was washed in PBST three times and subsequently incubated with HRP-conjugated secondary antibodies. Immobilon Western Chemiluminescent HRP substrate (Millipore Corporation, United States) was used to obtain visualization of target protein bands.
Biochemical analysis
According to the manufacturer’s instructions, we detect aspartate aminotransferase (AST) enzyme activity using aspartate aminotransferase activity assay kit (ab105135) and alanine aminotransferase (ALT) enzyme activity using an alanine transaminase activity assay kit (Colorimetric/Fluorometric ab105134).
Data handling and statistical analysis
KEGG pathway enrichment analysis and GO enrichment analysis were performed using Metascape. The database (GSE159676) was obtained from National Center for Biotechnology Information (NCBI). Protein–protein interaction (PPI) enrichment analysis was carried out by the Cytoscape software. When the network contained between 3 and 500 proteins, the MCODE algorithm was further used to identify the densely connected components of the network. Then, MCODE modules were detected from the PPI network using MCODE algorithm [
37]. Hub genes were identified using MCODE algorithm and Betweenness algorithm of plug-in CytoHubba in Cytoscape, version 3.9.1. All data were represented as means ± standard error of the mean (SEM) using Prism 8.3.0 software. The statistical significance of the difference between groups was determined using Student’s t-test in Prism 8.3.0 software, with 0.05 as the cutoff for statistical significance.
Discussions
IN our study, we conducted a comprehensive analysis to identify regulatory factors in NAFLD. 29 hub genes and five MCODE modules were screened using the MCODE algorithm. In the MCODE 5, four of six hub genes (
FABP4,
SERPINE1,
GADD45B and
NAMPT) has been reported to be associated with lipid metabolism [
29‐
32]. Furthermore, all of these hub genes were densely connected with
MAP3K4. The study showed that overexpression of
MAP3K4 was associated with abnormal lipid metabolism in NAFLD. Through interference experiments in vitro, our results indicated that
MAP3K4 might be involved in the biogenesis of LDs by regulating the phosphorylation of JNK and cPLA2 in NAFLD.
FABP4, which located in cytoplasm of adipocyte, enhanced lipolysis by interacting with hormone-sensitive lipase [
32,
50]. Inhibition of SERPINE1 reduced hepatic expression levels of PCSK9, which is well acknowledged as a regulator of lipid metabolism by impairing receptor-mediated low-density lipoprotein cholesterol (LDL-C) clearance [
31,
51]. GADD45B preventsed lipid accumulation by interacting directly with heat shock protein 72 [
30]. In high fat diet-induced mouse, upregulation of NAMPT has been shown to improve hepatic lipid homeostasis by reducing triglyceride levels [
52‐
54]. All the above hub genes were densely connected with
MAP3K4. These results suggested that it was quite reliable to select
MAP3K4 as a candidate gene for further study.
MAP3K4 is known as
MTK1, which is an important member of the MAPK signaling pathway [
55]. MAPK signaling pathway, consisting of a cluster of protein kinase, plays a crucial role in controlling an abroad of biological activities, including cell proliferation, cell motility, cell survival and death, and gene expression [
56] A previous study showed that the JNK cascade was identified as the one governing cPLA2 phosphorylation [
49]. In addition, numerous studies have revealed that MAP3K4 is crucial for JNK activation [
57]
The JNK signal transduction pathway is crucial for the negative regulation of insulin signaling, which is considered the major contributor to the development of hepatic steatosis [
58,
59]. MKK4 and MKK7 are the only two MAP2Ks that activate JNK, and these two upstream MAP2Ks can be activated by MAP3K4 [
57]. In addition, connexin and pannexin genes have been demonstrated to be crucial in liver diseases, including NAFLD [
60]. In non-alcoholic steatohepatitis, connexin32 exerted a protective effect in connexin32 dominant negative transgenic mice [
61]. Connexin32 regulates the expression of JNK and Cdc42. Besides, hepatic inflammation is caused by the activation of pannexin1 in lipoapoptosis [
62]. The pannexin1 activation may be mediated by upstream JNK, which can be activated by MAP3K4. These studies suggested that MAP3K4-JNK pathway might be crucial for hepatic lipid metabolism.
cPLA2α, a calcium-dependent enzyme that cleaves fatty acids which is essential for lipid droplet biogenesis, can be activated and phosphorylated by JNK. cPLA2α contributes to the formation of nascent LDs from the endoplasmic reticulum. Phosphorylation at Ser505 is key for cPLA2α enzyme activity and LDs formation [
49,
63,
64]. Consistent with previous studies, our findings demonstrated that
MAP3K4 knockdown greatly decreased LDs formation by reducing JNK activation and subsequent cPLA2 phosphorylation at Ser-505 [
49,
57]. In addition, due to MAP3K4 control of the insulin-like growth factor 1 receptor, stem cells lacking MAP3K4 kinase activity are less sensitive to insulin stimulation [
65]. This offers fresh insight for further research to investigate the function of MAP3K4 in NAFLD.
There are several limitations about the study. It is necessary to do additional research with larger sample sizes to clarify the relationship between the MAP3K4 expression levels of MAP3K4 and clinical indicators of NAFLD severity. The suspected NAFLD patients should also be considered. Additionally, more experiments are needed to further explore the role of MAP3K4 in NAFLD in vivo and in vitro. Finally, the roles of the other 28 hub genes in NAFLD, and their underlying mechanisms deserved further investigation.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.