Keywords

Nonalcoholic steatohepatitis (NASH) ; Diosmetin (Dios); Lipogenesis; Inflammation; Macrophage chemotactic ligand 10 (CXCL10); Signal transducers and activators of transcription 1 (STAT1).
 
 

Introduction

Nonalcoholic fatty liver disease (NAFLD) has become one of the most important causes of liver disease with an estimated global prevalence of 24%,1 encompassing a wide histological spectrum rang from steatosis to nonalcoholic steatohepatitis (NASH), advanced fibrosis, and even hepatocellular carcinoma.2, 3 Given the worldwide epidemic, NASH has caused an enormous clinical and economic burden.4 Some investigators predict that NASH may become the leading factor of liver transplantation in the next about 10 years.5 For NASH prevention, until now, there is no approved therapy aside from addressing lifestyle.6 The occurrence of NASH is characterized by hepatic steatosis and inflammation injury.7, 8 Excessive amounts of lipid accumulation cause toxicity to cells.9 Hepatocytes with lipotoxicity are likely to be more vulnerable to give rise to hepatic inflammatory response.10 The manipulation of lipogenesis and inflammatory response is a promising strategy to prevent NASH. Therefore, it will dramatically be crucial to identify new active compounds for treating NASH.
Diosmetin (3', 5, 7-trihydroxy-4'-methoxy flavone, Dios), a natural flavonoid isolated from citrus fruits,11 has a variety of favorable pharmacological activity in the treatment of diabetes, cancer and oxidative-damage diseases.12-15 The benefits of citrus fruit have been known to improve blood lipid profiles.16 Recent studies demonstrate that Dios inhibits the activation of NLRP3 inflammasome in LPS-induced acute lung injury.17 What’s more, our previous study reveals that Dios exhibits a supplementary effect on promoting fatty acid oxidation in diabetes-mice liver.18 Many compounds in foods of plant origin, particularly dietary flavonoids, exerts the protective effects through regulating lipid metabolism and inflammation against NASH.19-21 According to these results, we reasonably hypothesized that Dios exerted a hepatoprotective effect against NASH.
In the present study, we utilized the established models that HepG2 cells and C57BL/6J mice respectively induced by palmitic acid (PA) and high-fat diet (HFD) to explore the protective effects and the underlying molecular mechanism of Dios against NASH.

Materials and Methods

Chemicals and reagents

Diosmetin (Dios, purity ≥ 98%, Fig. 1A) was purchased from Chengdu Herbpurify CO., LTD (Sichuan, China). The standard diet (SD, HD005) and high-fat diet (HFD, HD001+2% cholesterol, containing 45% fat and 2% cholesterol) were purchased from BiotechHD Co.Ltd. (Beijing, China). The aspartate aminotransferase (AST), alanine aminotransferase (ALT), total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) commercial kits were obtained from Nanjing Jiancheng Bioengineering Institute (Jiangsu, China) and the triglyceride (TG) commercial kit was purchased from Applygen Technologies Inc. (Beijing, China). The Oil Red O and Nile Red staining kits were obtained solarbio (Beijing, China). Dulbecco's modified Eagle medium (DMEM) was purchased from KeyGen Biotech (Jiangsu, China), and fetal bovine serum (FBS) was obtained from Tianhang (Zhejiang, China). Palmitic acid (PA) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Fludarabine (HY-B0069, STAT1 inhibitor) was purchased from MedChemExpress (NJ, USA). siRNA for CXCL10 (human) and the control siRNA (human) were purchased from Tsingke Biological Technology (Beijing, China). The STAT1 overexpression plasmid STAT1-pcDNA3.1 and the control plasmid pcDNA3.1 were obtained from Pulateze (Hunan, China). Lipo6000™ transfection reagent was obtained from Beyotime Biotechnology (Shanghai, China). Antibodies against STAT1, CXCL10, sterol regulatory element-binding proteins-1c (SREBP-1c), carbohydrate response element-binding protein (CHREBP), liver X receptor-alpha (LXRα), liver X receptor-beta (LXRβ), nuclear factor-κB p65 (p65), phosphorylated-p65 (p-p65), tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), β-actin and Lamin B were obtained from Proteintech Group, Inc. (Hubei, China). Antibody against STAT1Y701 was purchased from Affinity Biosciences (Jiangsu, China). Antibody against STAT1S727 was obtained from Beyotime Biotechnology (Shanghai, China).

Animals experiments

All animal experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The animal studies were performed after receiving approval of the Institutional Animal Care and Use Committee (IACUC) in Southwest University, P.R. China (IACUC approval No. yxy202007). All the animals were housed under specific pathogen-free (SPF) conditions at a controlled temperature (22-25 °C) and humidity (50 ± 5%), and alternating 12-hour light/dark cycles. The animals used in this study were housed in individually ventilated cages with natural soft sawdust as bedding up to six per cage and maintained on a normal chow diet with clean water ad libitum.
Eight-week-old male C57BL/6J mice (18-22g) were obtained from Beijing Huafukang Biotechnology Co., Ltd. (SCXK-Jing-2019-0008, Beijing, China). Mice were fed with either SD or HFD for 14 weeks to induce NASH model. Thereafter, the HFD-fed NASH mice were randomly divided into HFD group and HFD + Dios group (n = 6 per group), and administered intragastrically either Dios or not Dios of 60 mg·kg-1·day for four weeks 12. Mice were euthanized by a cervical dislocation under anesthesia, and the livers were removed and either snap-frozen or fixed in buffered 4% formalin for RNA-Seq and followed experiments.
To explore the potential mechanism of STAT1 in the beneficial effect of Dios in NASH, another animal experiment was carried out ( Eight-week-old Male C57BL/6J, SCXK-Jing-2019-0010) that mice were injected with fludarabine (Flu, STAT1 inhibitor, 0.8 mg·kg-1·day) by the abdominal cavity two cycles for 5 days every 2 weeks in the whole experiment, and mice were fed with either SD or HFD for 14 weeks to induce NASH. Thereafter, the HFD-fed mice were randomly divided into different groups (n = 6 per group) including HFD, HFD + Dios, Flu-HFD, and Flu-HFD + Dios, and those mice were either administered or not administered intragastrically Dios (60 mg·kg-1·day) for four weeks. The diet intake of the mouse was monitored per day and the body weight was recorded weekly. After the last administration, all the mice euthanized by a cervical dislocation under anesthesia, and the livers, kidneys, spleens and abdominal fat pads were removed and weighted. Serum was prepared by solidification and centrifugation (4 °C, 12000 × g, 10 min) and stored at -80 °C until the analysis of biochemical parameters. Liver samples were either snap-frozen or fixed in buffered 4% formalin for histological staining, hepatic triglyceride content measurement, quantitative real-time PCR, and western blot assay.

Cell culture and treatment

HepG2 cells were obtained from the Cell Bank at the Chinese Academy of Sciences (Shanghai, China) and cultured in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin in a 5% CO2 humidified incubator at 37 °C. HepG2 cells were treated with different doses (0.10, 0.15, 0.20, 0.25, 0.30, 0.35 and 0.40 mM) of PA for 24 h to induce NASH cell model . To detect the effect of Dios, HepG2 cells were treated with or without Dios under the indicated concentrations combined with PA treatment for 24 h. To study the importance of STAT1 in Dios-mediated hepatoprotective effect, the STAT1 inhibitor fludarabine (10 μM, Fig. S1) was used to pretreat HepG2 cells for 1 h before Dios and PA.

Transfection of overexpression plasmid and siRNA

HepG2 cells (3× 105 cells per well) were seeded in 6-well plates and incubated until the cells reached 70-80% confluence. Cells were then transfected with the STAT1 reporter vector STAT1-pcDNA3.1, the control vector pcDNA3.1 and siCXCL10 (5'-GGUCUUUAGAAAAACUUGATT-3', 3'-UCAAGUUUUUUCUAA-AGACCTT-5') mixed in Lipo6000™ reagent according to the manufacturer’s guidelines. Then, the medium was exchanged with fresh complete medium after 6 h. After 24 h transfection, cells were incubated with 0.2 mM PA in the presence or absence of Dios (80 μM) for 24 hours. After treatment, cells were harvested for further experiments.

RNA Sequencing (RNA-Seq) assay

Total RNA isolated from liver of C57BL/6J mice were used to construct high throughput sequencing libraries using NEBNext® UltraTM RNA Library Prep Ki. High throughput RNA-sequencing was performed using a HiSeq 4000 instrument (Illumina) at Novogene (Beijing, China). Adaptor sequences and low-quality reads were initially filtered from the raw data. Then the remaining ones, called clean reads, were aligned to the reference genome of a mouse, using the HISAT2 v2.0.5 program. Subsequently, unigene expression was calculated as the FPKM (fragments per kilobases of exons for per million mapped reads) with featureCounts v1.5.0-p3. Differential expression genes (DEGs) of two groups were performed using the DESeq2 R package, and when the P-value was less than 0.05 and the log2 ratio was greater than 1 (two-fold change), the unigenes were considered to be differentially expressed. The enrichment analysis of GO (Gene Ontology) and KEGG (Kyoto encyclopedia of genes and genomes) pathway was performed for DEGs using the David which is an online biological information database for annotation, visualization, and integrated discovery 22. Protein-protein interaction (PPI) analysis of differentially expressed genes was based on the STRING database. Hub genes were identified using Cytohubba (a plug-in of Cytoscape software) filtered with the criterion of degrees > 10 criteria (each node had more than 10 interactions).

Biochemical assessment of serum and liver

Serum triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), alanine transaminase (ALT) and aspartate aminotransferase (AST) were quantified using enzymatic assays. The content of TC and TG of liver tissues and HepG2 cells were determined with an enzymatic assay kit and the protein concentrations were determined with a BCA protein assay kit. The results were expressed as mmol per g protein (mmol·g-1). All procedures were carried out according to the manufacturer’s instructions.

Histological assessment

Histologic examination of the liver was performed by hematoxylin eosin staining. Liver samples were fixed briefly in 4% paraformaldehyde and embedded with paraffin. The samples were cut into 5 μm sections, deparaffinized, rehydrated, then stained with hematoxylin eosin or mason trichrome. Images of liver sections were captured with light microscope. The amount of steatosis, activity, and fibrosis were scored using the Fatty Liver Inhibition of Progression (FLIP) Algorithm.23 The SAF score was defined as the total of scores for steatosis (S0-S3), activity including lobular inflammation (0-2) and hepatocellular ballooning (0-2) scores, and the fibrosis (F0-F4) score.

Immunofluorescence analysis

Liver samples embedded with paraffin were dewaxed, blocked and then incubated with phosphorylated STAT1 Tyr701 and phosphorylated STAT1 Ser727 antibodies overnight at 4 °C. Slides were washed three times with PBS and then incubated with secondary antibody labeled fluorescence for 30 min at 37 °C. The nuclei were counterstained with 4′, 6-diamidino-2-phenylindole (DAPI) for 5 min at room temperature. Images were captured using fluorescent microscopy.

Lipid content analysis

Hepatic lipid accumulation was detected by Oil Red O staining. Liver sample cryosections and hepatocytes were prepared for staining with pre-warmed Oil Red O working solution for 20 min, being rinsed with 60% isopropanol 3 times. And then counterstained with hematoxylin for 3 min, gently washed with 60% isopropanol. The images were captured using light microscopy. HepG2 cells were seeded in a 24-well plate and treated with 0.2 mM PA and indicated concentrations of Dios for 24 h. The cells were washed twice with PBS and fixed with 4% formaldehyde for 20 min and then stained with 1 μg·mL-1 Nile red for 30 min at room temperature. Lipid-bound Nile red fluorescence was detected using fluorescence microscopy.

Quantitative real-time polymerase chain reaction(qRT-PCR)

To verify the quantification of gene expression levels, qRT-PCR was performed as previously described 21. Total RNA was extracted from livers and HepG2 cells using Trizol Reagent (Sangon Biotech Co., Ltd., Shanghai, China), and RNA was reverse-transcribed using the Fastingking RT kit (TiangenBiotech CO. LTD., Beijing, China). Quantitative real-time PCR was carried out with SYBR Premix Plus  (TiangenBiotech CO. LTD., Beijing, China) according to the manufacturer’s instructions, and the gene primers listed in Tab. 1. Relative mRNA expression was determined by a comparative method (2-ΔΔCt) using GAPDH as a reference gene.

Western blotting analysis

The proteins were extracted from mouse liver and HepG2 cells, then equal amounts of protein extract were denatured. After separated by electrophoresis in 12% SDS–PAGE, the protein samples were transferred onto Polyvinylidene Fluoride (PVDF) membranes. The membranes were blocked with 5% skimmed milk for 2 h, subsequently incubated with primary antibodies against STAT1 (1:5000), p-STAT1Y701 (1:1000), p-STAT1S727 (1:1000), CXCL10 (1:500), p65 (1:1000), p-p65 (1:1000), IL-6 (1:1000), TNFα (1:1000), LXRα (1:1000), LXRβ (1:1000), CHREBP (1:1000), SREBP-1c (1:1000), β-actin (1:2000) and Lamin B (1:2500) overnight at 4 °C. After three times washing, the membranes were incubated with an HRP-conjugated secondary antibody at room temperature for 2 h. Antibody binding was detected by enhanced chemiluminescence detection kit (Affinity Biosciences, Jiangsu, China) and the digital images were analyzed by Image J. The relative protein levels were normalized to β-actin or Lamin B.

Statistical analysis

Randomization was used to assign samples to the experimental groups and treatment conditions for all in vivo studies. Data collection and acquisition of all in vivo and in vitro experimental paradigms were performed in a blinded manner. All results are expressed as the mean ± SD. Difference among groups were analyzed with one-way analysis of variance (ANOVA) followed by a Bonferroni post hoc analysis using GraphPad Prism 5 software. Post hoc tests were conducted only if F was significant, and there was no variance inhomogeneity. A value of p < 0.05 was considered statistically significant.

Results

Dios decreases lipogenesis and inflammatory response in HFD-induced NASH mice

In order to identify the effect of Dios against through lipogenesis and inflammatory response, a NASH mice model induced by HFD was established and treated by Dios. As depicted in Fig. 1B, Dios sharply reversed lipid accumulation in livers of HFD-induced NASH mice from Oil Red O staining. Furthermore, as shown in HE and Masson staining (Fig. 1C and 1D), the liver tissues exhibited obvious histological changes including steatosis, hepatocyte ballooning, and lobular inflammation, but no fibrosis in the HFD group, and those changes were reduced after Dios treatment. Lipid metabolism plays a critical role in the progress of NASH, the expressions of LXRα, LXRβ, CHREBP, and SREBP-1c protein relevant with lipogenesis were downregulated after Dios treatment (Fig. 1E). As illustrated in Fig. 1F, the amount of p65 phosphorylation and the release of inflammatory factors (TNFα and IL-6) were repressed by Dios.
PA was utilized to establish a NASH model in vitro. The cytotoxicity of PA was measured (Fig. S2A), and there were no apparent changes in HepG2 cells exposed to PA at concentrations of less than 0.2 mM. Fig. S2B and S2C showed that PA treatment for 24 h increased intracellular lipid accumulation and TG content in HepG2 cells in a dose-dependent manner. Thus, the concentration of PA at 0.2 mM was used for achieving maximal lipid accumulation without cytotoxicity in experiments. Dios exhibited no cytotoxicity at the lesser dose than 320 μM in HepG2 cells (Fig. S2D). Dios dose-dependently enhanced the viability of PA-induced HepG2 cells at a concentration of greater than 5 μM (Fig. S2E). Compared with the PA group, both the lipid accumulation and intracellular TG content were reduced significantly by co-treated with Dios and PA in HepG2 cells (Fig. 2A-C). Dios also suppressed expressions of the lipogenic proteins LXRα, LXRβ, CHREBP, and SREBP-1c (Fig. 2D) and decreased that of the inflammatory proteins p-p65 and IL-6 (Fig. 2E) induced by PA in HepG2 cells. Summarizing the above results, we concluded that Dios conferred protection on NASH in vivo and in vitro.

Dios promotes STAT1 and CXCL10 molecular phenotype in livers of NASH mice

Considering the remarkable effects of Dios in vivo and vitro, a liver-transcriptome analysis by RNA sequencing (RNA-Seq) was adopted to evaluate the key genes involved in the blockage of NASH disease by Dios. After normalization and analysis of the sequencing data, the DEGs from three sets of the samples were shown in Fig. 3A and B (A, HFD vs SD; B, HFD + Dios vs HFD). The DEGs of SD, HFD, and HFD + Dios group were compared with Hierarchical cluster analysis (Fig. 3C). From the Venn diagram, the result illustrated 456-shared DEGs were discovered between HFD vs SD and HFD + Dios vs HFD (HFD vs SD: 2236 genes; HFD + Dios vs HFD: 566 genes) (Fig. 3D). To identify the biological features and explore the enrichment-pathways of 456-shared DEGs, the enrichment analysis of GO and KEGG pathway was accomplished by DAVID online tools. The result of GO analysis (Tab. 2) suggested that changes in biological processes (BP) of DEGs were enriched in the immune system process, inflammatory response, chemotaxis, and positive regulation of T cell activation. Changes in cellular components (CC) were mainly enriched in the membrane, plasma membrane, and cytoplasm. Changes in molecular functions (MF) of DEGs were enriched in GTP binding, chemokine activity, and protein kinase binding. Beside, we found that DEGs were enriched in different pathways, such as antigen processing and presentation, type I diabetes mellitus, cytokine-cytokine receptor interaction, cell adhesion molecules (CAMs) and NF-κB signaling pathway (Tab. 2). PPI network consisting of 408 nodes and 3294 edged (Fig. 3E) was also established by STRING database to reveal the interactive relation of 456 DEGs. The top 10 hub genes in the DEGs were confirmed by Cytohubba including TNF, CCL5, CXCL10, STAT1, CXCL9, ITGAX, CD274, IFNG, PSMB8, and CD40 (Fig. 3F). The abbreviations, official names, and functions of these hub genes were shown in Tab. 3. Ultimately, qRT-PCR analysis was applied to examine the expression of six genes from DEGs (Fig. 3G) and the levels of those tested genes were aligned with that of RNA-Seq analysis. From the hub genes combined with the literatures, we selected STAT1 and CXCL10 as key genes for further research. The qRT-PCR result told that Dios dramatically decreased the mRNA levels of STAT1 and CXCL10 induced by HFD (Fig. 3H). Moreover, the levels of p-STAT1Y701, p-STAT1S727, and CXCL10 protein were up- and downregulated in the HFD and HFD + Dios group, respectively (Fig. 3I). Therefore, in the following research, we focused on the functional fraction of STAT1 and CXCL10 and investigate their underlying mechanism in the treatment of NASH by Dios.

Dios prevents CXCL10-mediated lipogenesis and inflammatory response

To further identify the mediation of Dios on CXCL10, we applied siCXCL10 to make CXCL10 silence in HepG2 cells. Fig. 4A showed the transfection efficiency of CXCL10 siRNA at the mRNA level. And siCXCL10-treated HepG2 cells displayed downregulated expression levels of CXCL10 protein (Fig. 4B). As seen in Fig. 4C and 4D, the TG content and lipid accumulation level were significantly decreased after siCXCL10 treatment. Compared with siNull treatment, siCXCL10-treated HepG2 cells displayed downregulated expression levels of LXRα, LXRβ, and p-p65 proteins (Fig. 4E). Thus, lipid accumulation and inflammation caused by PA were efficiently inhibited by siCXCL10, suggesting that CXCL10 was critical to mediate lipogenesis and inflammatory response.

Dios inhibits the nuclear localization of phosphorylated STAT1 to prevent  lipogenesis and inflammatory response

The regulation of Dios on STAT1 was explored in our study. In the immunofluorescence assay, the nuclear localization of p-STAT1Tyr701 (Fig. 5A) and p-STAT1Ser727 (Fig. 5B) in the HFD group was as well decreased in the presence of Dios. Correspondingly, the result obtained from the detection of western blot was consistent with that of immunofluorescence (Fig. 5C). To further recognize the role of STAT1 in the hepatoprotective effects of Dios identify, the specific STAT1 inhibitor (fludarabine) was utilized to inhibit the activation of STAT1. The TG content and lipid accumulation induced by PA were depressed by pretreatment with fludarabine in HepG2 cells (Fig. 6A and 6B), implying a critical role for STAT1 in the amelioration of hepatocyte steatosis. As shown in Fig. 6C, the inhibition of STAT1 apparently decreased the levels of p-STAT1Y701 and p-STAT1S72 proteins in HepG2 cells. Additionally, the inhibition of STAT1 significantly reduced the expression of lipogenic proteins (LXRα, LXRβ, CHREBP, and SREBP-1c) (Fig. 6D) and downregulated inflammatory indicators p-p65 (Fig. 6E) at the protein level. Therefore, we further verified that Dios mitigated the lipogenesis and inflammation were STAT1-dependent.

Dios interferes with STAT1/CXCL10 pathway to ameliorate NASH

As shown in Fig. 7A, the inhibition of STAT1 apparently decreased the levels of CXCL10 protein. Compared with siNull treatment, siCXCL10-treated HepG2 cells displayed no effects on the levels of STAT1, p-STAT1Y701, and p-STAT1S727 protein (Fig. 7B), which indicated that CXCL10 was the downstream gene of STAT1. To further confirm our finding, we transfected HepG2 cells with STAT1 overexpression plasmid. Fig. 7C displayed the transfection efficiency of STAT1 plasmid at the mRNA level. STAT1 overexpression aggravated PA-induced lipid accumulation as observed with O Red O (Fig. 7D) and upregulated the amounts of p-STAT1Y701, p-STAT1S727, and CXCL10 protein (Fig. 7E).
To investigate the contribution of STAT1 in the protective effect of Dios on NASH, a STAT1 inhibitor fludarabine was pretreated to induce low expression of STAT1 in mice. A substantial body-weight reduction was found in NASH mice treated with Dios and fludarabine (Fig. 8A), with no effect on food consumption was found among all groups (Fig. S3B). Fig. 8B showed that Dios and fludarabine markedly alleviated and even reverse pathological changes in liver morphology of NASH mice. Additionally, STAT1-inhibited mice treated with Dios exhibited a decreased index of liver (Fig. 8C) and abdominal fat pad (Fig. S3C and S3D). Furthermore, the results of the biochemical analysis showed that Dios also reduced the levels of serum TC, LDL-C, ALT, and AST, but no difference was seen for serum TG and serum HDL-C (Fig. 8D-I, Fig. S3D-F). The results of the contents of liver TG (Fig. 8E) and TC (Fig. 8F) and Oil Red O staining (Fig. 8A) revealed that there was massive lipid accumulation in the livers of NASH mice, which was fully alleviated by Dios in mice and STAT1-inhibited mice. Meanwhile, the obvious pathological changes of livers was observed including increased hepatic steatosis, hepatocyte ballooning, and lobular inflammation (Fig. 8J). These changes were remarkably ameliorated by Dios and fludarabine in HFD-fed mice. SAF score was performed by histological analysis according to a grading system established for mouse models of NAFLD/NASH, showing that the majority of HFD-fed mice displayed a pathophysiological characteristic of NASH, whereas both Dios and fludarabine dramatically attenuated that characteristic (Fig. S4A-D).
STAT1/CXCL10 was regarded as an crucial pathway relevant with lipogenesis and inflammation in NASH mice. Fig. 9A revealed that the levels of p-STAT1Y701, p-STAT1S727, and CXCL10 protein significantly were eliminated in the HFD + Dios group. Additionally, in comparison with HFD group, the expressions of these proteins were inhibited in the HFD-treated group with STAT1 inhibitor. As depicted in Fig. 9B and 9C, Dios-treated mice displayed decreased protein expression levels of LXRα, LXRβ, CHREBP, SREBP-1c, and p-p65. The protein levels of all of these markers were completely normalized in the fludarabine-treated mice. These indicated that Dios could regulate lipogenesis and inflammatory response by inhibiting STAT1/CXCL10- LXRα/β-p65 pathway.

Discussion

Emerging evidence suggest that histopathological evaluation of biopsy specimens is a gold standard for diagnosing NASH.24 The typical characteristic such as steatosis, lobular inflammation, and hepatocellular ballooning are all observed in NASH model liver but fibrosis is not uncertain.25 Actually, by Oil Red O, HE and Mason trichrome staining in the livers of NASH mice in our study, liver tissues occurred these histological changes without fibrosis. Meanwhile, the SAF scores of liver histology assisted us to recognize that the establishment of NASH mice model was successful.23 As for the impact of Dios on NASH mice, Dios distinctly blocked the pathological changes of liver. In vitro HepG2 cells, Dios eliminated lipid deposition and TG content induced by PA. Our finding was consistent with Zeng’s report.4 In the pathway of lipogenesis, LXRα/β, SREBP-1c and ChREBP are the important transcription factors to regulate hepatic lipogenesis in NAFLD,26, 27 and LXRα/β can activate SREBP-1c and ChREBP to mediate the metabolism of cholesterol and fatty acids.28, 29 Those proteins mediated lipogenesis were inhibited by Dios in NASH mice and HepG2 cells. In the pathway of inflammation, the important marker NF-κB (p65) and the inflammatory factors including TNFα and IL-6 are regarded as the indicator of hepatic inflammation.30, 31 And Dios inhibited the levels of  the inflammatory protein and factors in NASH mice and HepG2 cells. Our study showed that Dios exerted a hepatoprotective effect against NASH, but the key target genes and proteins interfered by Dios in NASH mice were unclear and should be explored. Hence, the RNA-Seq analysis was employed in the further seeking of the novel key targets.
Here a total of 456 mutual DEGs were identified using liver RNA-Seq in our experiment. After the enrichment analysis of those DEGs, most of them were focused on the fields of inflammatory and immune response. Specially, CXCL10 (macrophage chemotactic ligand 10) and STAT1 (signal transducers and activators of transcription 1) were the hub genes among the DEGs, and were likely the major regulated targets for Dios. Hepatocytic lipotoxicity promotes the release of CXCL10, inducing macrophage chemotaxis, increased inflammation, and liver injury.32 STAT1 is known to be a transcription factor that mediates immune and pro-inflammatory responses33 and modulates lipid metabolism, energy expenditure, and mitochondrial biogenesis.34 In the liver of NASH mice and patients, the levels of STAT1 phosphorylation and CXCL10 are increased.33, 35 Similarly, our data showed that the mRNA and protein expressions of STAT1 and CXCL10 were significantly upregulated both in NASH mice and HepG2 cells, however, Dios exhibited the repression on those expressions. In a word, CXCL10 and STAT1 were considered as key targets had a pivotal role for the attenuation of Dios on NASH. However, whether STAT1 and CXCL10 mediated related proteins involved lipogenesis and inflammatory response needed to be discussed.
NASH displays the disorder in lipid metabolism and inflammation,36 while the elevation of CXCL10 and STAT1 expression may contribute to that disorder.37, 38 In other words, if the disorder were mediated by CXCL10 and STAT1, the down-stream proteins involved in the pathways of lipogenesis and inflammation should be interfered. The lipogenic factor LXRα/β and inflammatory protein NF-κB (p65) are diminished in CXCL10-deficient mice.35 Our finding equally manifested that when HepG2 cells were administrated with CXCL10 siRNA, the content of TG and lipid accumulation and the expressions of LXRα/β and p-p65 protein were reduced. Meanwhile, our results indicated that Dios reduced the nuclear localization of p-STAT1Y701 and p-STAT1S727 in livers of NASH mice induced by HFD. In terms of the related reference, saturated free fatty acids induced hepatocyte lipotoxicity in NASH could cause the nuclear localization of p-STAT1Y701,39 and this conclusion supported our results appropriately.  The pretreatment of STAT1-inhibitor fludarabine decreased the lipid accumulation, downregulated the level of STAT1 phosphorylation, and induced the elimination of relative protein levels of lipogenesis and inflammatory response in HepG2 cells. And all inflammatory indicators were inhibited as well after the reduction of STAT1 in our result. Thus, Dios could prevent CXCL10- and STAT1-mediated lipogenesis and inflammatory response. Furthermore, the hepatoprotective mechanism of Dios against NASH through STAT1-CXCL10 was need to be elucidated.
For better identification of the relation between the downstream proteins and STAT1-CXCL10, the STAT1-inhibitor, overexpression vector of STAT1, and CXCL10 siRNA were used here. Combined our study and the numerous evidences, the expression of CXCL10 was positively correlated with the STAT1 phosphorylation.39, 40 However, the levels of STAT1 and p-STAT1 after siCXCL10 treatment in HepG2 cells showed no differences in our result. According to the reference, lipid accumulation and hepatic TG level are increased in the livers of STAT1+/+ mice.38 In our result, the hepatocyte steatosis induced by PA were aggravated due to over-expression of STAT1, but that was overwhelmed by Dios. The inhibition or over-expression of STAT1 led to a decrease or increase of CXCL10 protein expression. Therefore, STAT1 was the upstream of CXCL10. Besides, the STAT1-inhibitor restrained the changes of biochemistry indicators, hepatic pathological parameters and the expression of lipogenesis and inflammation associated proteins in NASH mice, which further revealed that the whole potential hepatoprotective mechanism of Dios on lipogenesis and inflammatory response were most likely through the STAT1/CXCL10 signaling pathway in NASH.
Taken together, the protective mechanism of Dios against NASH was through the modulation of lipogenesis and the alleviation of inflammatory response via STAT1/CXCL10-dependent pathway. The present observations may provide a promising candidate for the treatment of NASH, and Dios can be developed as functional foods or therapeutic medicines for application.

Author contributions

B. Z. designed the research. N. L. performed the experiments, with the assistance of C. Y. and Y. Z. analyzed the data. N. L. wrote the initial manuscript and B. Z. revised the manuscript.

Conflicts of interest

The authors state that they have no conflicts of interest.

Acknowledgments

This work was supported by the Fundamental Research Funds for the Central Universities (XDJK2020B056), and the National Key Research and Development Program of China for Traditional Chinese Medicine Modernization (2017YFC1702605, 2017YFC1702606).

Abbreviations

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CHREBP, carbohydrate response element-binding protein; CXCL10, macrophage chemotactic ligand 10; DEGs, different expression genes; D-GaIN, D-galactosamine; HDL-C, high density lipoprotein cholesterol; HFD, high-fat diet; IL-6, interleukin-6; LDL-C, high density lipoprotein cholesterol; LPS, lipopolysaccharide; LXRα/β, liver X receptor alpha/beta; MCODE, molecular complex detection; NASH, nonalcoholic steatohepatitis; NF-κB, nuclear factor kappa B; PA, palmitic acid; PPI, protein-protein interaction; SREBP-1c, sterol regulatory element-binding proteins-1c; STAT1, signal transducers and activators of transcription 1; TC, total cholesterol; TG, triglyceride; TNF-α, tumor necrosis factor-alpha.

References

(1)     Lee, Y., Doumouras, A. G., Yu, J., Brar, K., Banfield, L., Gmora, S., et al. Complete Resolution of Nonalcoholic Fatty Liver Disease After Bariatric Surgery: A Systematic Review and Meta-analysis. Clin Gastroenterol Hepatol. 2019; 17 (6): 1040-1060.e1011.
(2)     Lau, J. K., Zhang, X., Yu, J. Animal models of non-alcoholic fatty liver disease: current perspectives and recent advances. J Pathol. 2017; 241 (1): 36-44.
(3)     Younossi, Z., Anstee, Q. M., Marietti, M., Hardy, T., Henry, L., Eslam, M., et al. Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol. 2018; 15 (1): 11-20.
(4)     Zeng, X., Yang, J., Hu, O., Huang, J., Ran, L., Chen, M., et al. Dihydromyricetin Ameliorates Nonalcoholic Fatty Liver Disease by Improving Mitochondrial Respiratory Capacity and Redox Homeostasis Through Modulation of SIRT3 Signaling. Antioxid Redox Signal. 2017; 30 (2): 163-183.
(5)     Afzali, A., Berry, K., Ioannou, G. N. Excellent posttransplant survival for patients with nonalcoholic steatohepatitis in the United States. Liver Transpl. 2012; 18 (1): 29-37.
(6)     Han, H. Y., Lee, S. K., Choi, B. K., Lee, D. R., Lee, H. J., Kim, T. W. Preventive Effect of Citrus aurantium Peel Extract on High-Fat Diet-Induced Non-alcoholic Fatty Liver in Mice. Biol Pharm Bull. 2019; 42 (2): 255-260.
(7)     Tomita, K., Freeman, B. L., Bronk, S. F., LeBrasseur, N. K., White, T. A., Hirsova, P., et al. CXCL10-Mediates Macrophage, but not Other Innate Immune Cells-Associated Inflammation in Murine Nonalcoholic Steatohepatitis. Sci Rep. 2016; 6: 28786.
(8)     Schuster, S., Cabrera, D., Arrese, M., Feldstein, A. E. Triggering and resolution of inflammation in NASH. Nat Rev Gastroenterol Hepatol. 2018; 15 (6): 349-364.
(9)     Piccolis, M., Bond, L. M., Kampmann, M., Pulimeno, P., Chitraju, C., Jayson, C. B. K., et al. Probing the Global Cellular Responses to Lipotoxicity Caused by Saturated Fatty Acids. Molecular Cell. 2019; 74 (1): 32-44.e38.
(10) Marra, F., Svegliati-Baroni, G. Lipotoxicity and the gut-liver axis in NASH pathogenesis. J Hepatol. 2018; 68 (2): 280-295.
(11) Liao, W., Ning, Z., Chen, L., Wei, Q., Yuan, E., Yang, J., et al. Intracellular antioxidant detoxifying effects of diosmetin on 2,2-azobis(2-amidinopropane) dihydrochloride (AAPH)-induced oxidative stress through inhibition of reactive oxygen species generation. J Agric Food Chem. 2014; 62 (34): 8648-8654.
(12) Yang, Y., Gong, X. B., Huang, L. G., Wang, Z. X., Wan, R. Z., Zhang, P., et al. Diosmetin exerts anti-oxidative, anti-inflammatory and anti-apoptotic effects to protect against endotoxin-induced acute hepatic failure in mice. Oncotarget. 2017; 8 (19): 30723-30733.
(13) Oak, C., Khalifa, A. O., Isali, I., Bhaskaran, N., Walker, E., Shukla, S. Diosmetin suppresses human prostate cancer cell proliferation through the induction of apoptosis and cell cycle arrest. Int J Oncol. 2018; 53 (2): 835-843.
(14) Michael, H. N., Salib, J. Y., Eskander, E. F. Bioactivity of diosmetin glycosides isolated from the epicarp of date fruits, Phoenix dactylifera, on the biochemical profile of alloxan diabetic male rats. Phytother Res. 2013; 27 (5): 699-704.
(15) Chen, X., Wu, Q., Chen, Y., Zhang, J., Li, H., Yang, Z., et al. Diosmetin induces apoptosis and enhances the chemotherapeutic efficacy of paclitaxel in non-small cell lung cancer cells via Nrf2 inhibition. Br. J. Pharmacol. 2019; 176 (12): 2079-2094.
(16) Roowi, S., Crozier, A. Flavonoids in tropical citrus species. J Agric Food Chem. 2011; 59 (22): 12217-12225.
(17) Liu, Q., Ci, X., Wen, Z., Peng, L. Diosmetin Alleviates Lipopolysaccharide-Induced Acute Lung Injury through Activating the Nrf2 Pathway and Inhibiting the NLRP3 Inflammasome. Biomol Ther (Seoul). 2018; 26 (2): 157-166.
(18) Gong, X. B. Influence of diosmetin on glucose and lipid metabolism and gut microbiota in KK-Ay mice. Southwest university. 2018; Chongqing (China):
(19) Marcolin, E., San-Miguel, B., Vallejo, D., Tieppo, J., Marroni, N., Gonzalez-Gallego, J., et al. Quercetin treatment ameliorates inflammation and fibrosis in mice with nonalcoholic steatohepatitis. J Nutr. 2012; 142 (10): 1821-1828.
(20) Wu, L., Wang, Y., Chi, G., Shen, B., Tian, Y., Li, Z., et al. Morin reduces inflammatory responses and alleviates lipid accumulation in hepatocytes. Journal of Cellular Physiology. 2019; 234 (11): 19785-19798.
(21) Zhou, R., Yi, L., Ye, X., Zeng, X., Liu, K., Qin, Y., et al. Resveratrol Ameliorates Lipid Droplet Accumulation in Liver Through a SIRT1/ ATF6-Dependent Mechanism. Cellular Physiology and Biochemistry. 2018; 51 (5): 2397-2420.
(22) Huang, D. W., Sherman, B. T., Lempicki, R. A. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Research. 2009; 37 (1): 1-13.
(23) Bedossa, P., Consortium, F. P. Utility and appropriateness of the fatty liver inhibition of progression (FLIP) algorithm and steatosis, activity, and fibrosis (SAF) score in the evaluation of biopsies of nonalcoholic fatty liver disease. Hepatology. 2014; 60 (2): 565-575.
(24) Brown, G. T., Kleiner, D. E. Histopathology of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. Metabolism. 2016; 65 (8): 1080-1086.
(25) Takahashi, Y., Fukusato, T. Histopathology of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World J Gastroenterol. 2014; 20 (42): 15539-15548.
(26) Ahmed, M. H., Byrne, C. D. Modulation of sterol regulatory element binding proteins (SREBPs) as potential treatments for non-alcoholic fatty liver disease (NAFLD). Drug Discov Today. 2007; 12 (17-18): 740-747.
(27) Scorletti, E., Byrne, C. D. Omega-3 fatty acids and non-alcoholic fatty liver disease: Evidence of efficacy and mechanism of action. Mol Aspects Med. 2018; 64 135-146.
(28) Higuchi, N., Kato, M., Shundo, Y., Tajiri, H., Tanaka, M., Yamashita, N., et al. Liver X receptor in cooperation with SREBP-1c is a major lipid synthesis regulator in nonalcoholic fatty liver disease. Hepatol Res. 2008; 38 (11): 1122-1129.
(29) Fan, Q., Norgaard, R. C., Bindesboll, C., Lucas, C., Dalen, K. T., Babaie, E., et al. LXRalpha Regulates Hepatic ChREBPalpha Activity and Lipogenesis upon Glucose, but Not Fructose Feeding in Mice. Nutrients. 2017; 9 (7): 678.
(30) Pea, A. d., Leclercq, I., Field, J., George, J., Jones, B., Farrell, G. NF-κB Activation, Rather Than TNF, Mediates Hepatic Inflammation in a Murine Dietary Model of Steatohepatitis. Gastroenterology. 2005; 129 (5): 1663-1674.
(31) Tak, P. P., Firestein, G. S. NF-kappaB: a key role in inflammatory diseases. J Clin Invest. 2001; 107 (1): 7-11.
(32) Ibrahim, S. H., Hirsova, P., Tomita, K., Bronk, S. F., Werneburg, N. W., Harrison, S. A., et al. Mixed lineage kinase 3 mediates release of C-X-C motif ligand 10-bearing chemotactic extracellular vesicles from lipotoxic hepatocytes. Hepatology. 2016; 63 (3): 731-744.
(33) Chmielewski, S., Olejnik, A., Sikorski, K., Pelisek, J., Blaszczyk, K., Aoqui, C., et al. STAT1-dependent signal integration between IFNgamma and TLR4 in vascular cells reflect pro-atherogenic responses in human atherosclerosis. PLoS One. 2014; 9 (12): e113318.
(34) Sisler, J. D., Morgan, M., Raje, V., Grande, R. C., Derecka, M., Meier, J., et al. The Signal Transducer and Activator of Transcription 1 (STAT1) Inhibits Mitochondrial Biogenesis in Liver and Fatty Acid Oxidation in Adipocytes. PLoS One. 2015; 10 (12): e0144444.
(35) Zhang, X., Shen, J., Man, K., Chu, E. S., Yau, T. O., Sung, J. C., et al. CXCL10 plays a key role as an inflammatory mediator and a non-invasive biomarker of non-alcoholic steatohepatitis. J Hepatol. 2014; 61 (6): 1365-1375.
(36) Farrell, G. C., van Rooyen, D., Gan, L., Chitturi, S. NASH is an Inflammatory Disorder: Pathogenic, Prognostic and Therapeutic Implications. Gut Liver. 2012; 6 (2): 149-171.
(37) Neville, L. F., Mathiak, G., Bagasra, O. The immunobiology of interferon-gamma inducible protein 10 kD (IP-10): a novel, pleiotropic member of the C-X-C chemokine superfamily. Cytokine Growth Factor Rev. 1997; 8 (3): 207-219.
(38) Sisler, J., Morgan, M., Raje, V., Grande, R., Derecka, M., Meier, J., et al. The Signal Transducer and Activator of Transcription 1 (STAT1) Inhibits Mitochondrial Biogenesis in Liver and Fatty Acid Oxidation in Adipocytes. PLoS One. 2015; 10 (12): e0144444.
(39) Tomita, K., Kabashima, A., Freeman, B. L., Bronk, S. F., Hirsova, P., Ibrahim, S. H. Mixed Lineage Kinase 3 Mediates the Induction of CXCL10 by a STAT1-Dependent Mechanism During Hepatocyte Lipotoxicity. J Cell Biochem. 2017; 118 (10): 3249-3259.
(40) Burke, S. J., Goff, M. R., Lu, D., Proud, D., Karlstad, M. D., Collier, J. J. Synergistic Expression of the CXCL10 Gene in Response to IL-1β and IFN-γ Involves NF-κB, Phosphorylation of STAT1 at Tyr701, and Acetylation of Histones H3 and H4. The Journal of Immunology. 2013; 191 (1): 323-336.
 
 

Figure Legends

 
Figure 1 Effect of Dios on the indicators of lipogenesis and inflammation in HFD-induced NASH mice. (A) The chemical structure of diosmetin (Dios). Histological evaluations of mice liver tissues in different groups were through Oil Red O staining (B), HE staining (C) and mason trichrome staining (D), respectively. Scale bar 100 μm. (1, steatosis; 2, hepatocyte ballooning; 3, lobular inflammation). (E) The determination of relative proteins expression of lipid metabolism were used Western blot. (F) The expression of IL-6, TNFα, p65 and p-p65 proteins. β-actin was used as an internal control. Data are expressed as mean ± SD (n = 6). *P < 0.05, vs. the SD group; #P < 0.05, vs. the HFD group.
 
Figure 2 Impact of Dios on the indicators of lipogenesis and inflammation induced by PA in HepG2 cells. (A) Intracellular TG content of was detected by using commercial kit after the treatment of 0.2 mM PA in the presence or absence of Dios. (B) Representative images of hepatocellular lipid accumulation which were treated with 0.2 mM PA in the presence or absence of Dios by Oil Red O staining. Scale bar, 20 μm. (C) Representative fluorescence of HepG2 cells was shown after PA and Dios treatment with Nile red staining, and the mean fluorescent intensity was measured as well. Scale bar, 20 μm. (D) The levels of LXRα, LXRβ, CHREBP, SREBP-1c protein expression after the same treatment as above. (E) The levels of IL-6, p65 and p-p65 protein expression after the same treatment as above. Data are presented as mean ± SD (n = 3). *P<0.05, vs. the Control group; #P<0.05, vs. the PA group.
Figure 3 Effect of Dios on molecular phenotype of STAT1 and CXCL10 in mice liver by using RNA-Seq. (A-B) Identification of DEGs in different groups. (A) HFD vs SD group. (B) HFD + Dios vs HFD group. The red points represent upregulated genes. The green points represent downregulated genes. The blue points represent genes with no significant difference. DEGs were screened on P < 0.05 and | log2 (Fold change) | > 1. (C) Heat map analysis was employed to the discrimination of expression profile of DEGs across the samples. Red and blue areas separately represent highly and lowly expressed genes in mice livers among SD, HFD and HFD+Dios groups. (D) Venn diagram of DEGs among different groups. 456 shared DEGs had obtained from this diagram. (E) The PPI network of 456 shared DEGs was analysized through STING database. There were 408 nodes and 3294 edged in the PPI network. (F) The ten hub genes including CXCL10 and STAT1 were confirmed by Cytohubba. (G) The amount of DEGs mRNA expression (FPR2, PSMB8, PTAFR, CCR7, CXCR4, LCK) was measured under the method of quantitative RT-PCR. (H) The mRNA expressions of STAT1 and CXCL10 in mice livers by qRT-PCR assay. (I) The expression of CXCL10, total STAT1, p-STAT1Y701 and p-STAT1S727 proteins were examined through western blot assay. β-actin was used as an internal control. Data are expressed as mean ± SD (n = 3). *P < 0.05, vs. the SD group; #P < 0.05, vs. the HFD group.
 
Figure 4 Dios prevents lipogenesis and inflammatory response, which is CXCL10-dependent. HepG2 cells treated with siRNA either siNull or siCXCL10 for 24h before the stimulation of Dios and PA. The mRNA (A) and protein (B) expression of CXCL10 in HepG2 cells transfected with siCXCL10. (C) Intracellular TG content of HepG2 cells after treatment. (D) Lipid accumulation by Oil Red O staining. Scale bar, 20 μm. (E) The amount of LXRα, LXRβ, p65, and p-p65 expression, and β-actin was used as an internal control. Data are presented as mean ± SD (n = 3). *P<0.05, compared between the marked groups.
 
Figure 5 Dios prevents the nuclear localization of phosphorylated STAT1 in vivo. Representative immunofluorescence of mice liver sections were stained with (A) phospho-STAT1 (Tyr701) and (B) phospho-STAT1 (Ser727) antibodies. The staining was observed following the addition of DAPI by fluorescence microscope. Scale bar, 20μm. (C) The levels of p-STAT1Y701, p-STAT1S727, STAT1, β-actin and LaminB protein expression from the cytosol into the nucleus were analyzed by Western blot. Data are expressed as mean ± SD (n = 3). *P<0.05, vs. the SD group; #P<0.05, vs. the HFD group.
 
Figure 6 Dios prevents lipogenesis and inflammatory response, which is STAT1-dependent. HepG2 cells pretreated with or without fludarabine inhibitor (10 μM) before the stimulation of Dios and PA. (A) Intracellular TG content of HepG2 cells after treatment. (B) Lipid accumulation by Oil Red O staining. The expression of (C) STAT1, p-STAT1Y701, p-STAT1S727, (D) LXRα, LXRβ, CHREBP and SREBP-1c, (D) p65 and p-p65 were examined by Western blot assay and quantified by densitometric analysis. β-actin was used as an internal control. Data are presented as mean ± SD (n = 3). *P<0.05, compared between the marked groups.
 
Figure 7 Dios inhibits CXCL10 by decreasing phosphorylation of STAT1. (A) The level of CXCL10 protein expression in HepG2 cells, which were pretreated with or without fludarabine inhibitor (10 μM) before the stimulation of Dios and PA. (B) The level of STAT1, pSTAT1Y701 and pSTAT1S727 protein expression in HepG2 cells, which were treated with siRNA either siNull or siCXCL10 for 24h before the stimulation of Dios and PA. From C to E, HepG2 cells treated with plasmid either pcDNA3.1 or STAT1-pcDNA3.1 for 24h before the stimulation of Dios and PA. (C) The mRNA expression of STAT1. (D) Lipid accumulation by Oil Red O staining. (E) The levels of STAT1, p-STAT1Y701, p-STAT1S727 and CXCL10 protein, and β-actin was used as an internal control. Data are presented as mean ± SD (n = 3). *P<0.05, compared between the marked groups.
 
Figure 8 Inhibition of STAT1 improves lipid accumulation and inflammation in HFD-induced NASH mice. (A) Weekly body weights of mice and STAT1-inhibited mice were recorded during 18 weeks. (B) Gross examination of liver. (C) Liver index. The contents of liver TG (D) and TC (E) were detected using a commercial kit. The levels of serum TG (F), TC (G), ALT (H) and AST (I) were tested via corresponding kits. (J) Histological analysis of mice livers by Oil Red O staining Scale bars, 100 μm. (K) Histological changes of liver sections was respectively evaluated with hematoxylin eosin staining. Scale bar, 100 μm. Data are presented as mean ± SD (n = 6). *P < 0.05, compared between the marked groups.
 
Figure 9 Interference with STAT1/CXCL10 pathway mediates Dios-conferred hepatoprotective effect. (A) The levels of p-STAT1Y701, p-STAT1S727, STAT1, and CXCL10 protein in the livers of STAT1-inhibited or non-inhibited mice. (B) The levels of LXRα, LXRβ, CHREBP, and SREBP-1C protein in the livers of STAT1-inhibited or non-inhibited mice. (C) The levels of p65 and p-p65 protein in the livers of STAT1-inhibited or non-inhibited mice. β-actin was used as an internal control. Data are presented as mean ± SD (n = 3). *P < 0.05, compared between the marked groups.
 

Table Legends

Table 1 Primers used for real-time RT-PCR analysis.
Table 2 GO and KEGG pathway enrichment analysis of DEGs.
Table 3 Functional roles of 10 hub genes.