Hepatitis C Virus nonstructural protein 4B induces lipogenesis via Merlin-Hippo pathway

Hepatitis C virus (HCV) infection is often associated with hepatic steatosis, even hepatocellcular carcinoma. Nonstructural protein 4B (NS4B)a highly hydrophobic non-structural protein, induces lipogenesis, but the underlying mechanism remains incompletely understood. In the present study, NS4B expression in Huh7.5 cells could reduce moesin-ezrin-radixin–like protein (Merlin, NF2 gene coding protein) expression, Yap phosphorylation, and increase AKT phosphorylation and sterol regulatory element-binding proteins (SREBPs) expression, which regulate lipogenic gene expressions including fatty acid synthase (FAS). In Huh7.5 cells, Merlin silencing reduced Yap phosphorylation, AKT phosphorylation, and especially SREBPs levels lead to reduction of lipid drop synthesis. In addition, HCV NS4B infected Huh7.5 cells exhibited lipid droplets, but lipid droplets significantly reduced after Merlin over-expressed. In conclusion, these results indicated that NS4B may play an important role in HCV-associated liver pathogenesis via Merlin-Hippo signal pathway. our findings showed that NS4B inhibited Merlin and Yap phosphorylation, and also up-regulated AKT phosphorylation to lead to lipogenesis. The results may provide a novel mechanism of hepatic steatosis associated with HCV infection.


Introduction
Chronic liver disease caused by virus infection distribute widely in the word, and always associated with high morbidity and mortality, one of the leading causes is HCV infection (1).The host liver can be damaged progressively once infected by HCV and the infection is persistent (2,3). Host factors synthesis can be promoted obviously once infected by HCV for virus propagation, such as cellular lipid generation and metabolism (4,5). It has been clearly shown that lipid generation and metabolism can be altered by HCV in infected hepatocytes (4,6,7), and it is the main reason why hepatic steatosis always found in HCV infected patients. Li,et al. (8) reported that HCV exploited intrinsic response and hijacked lipid metabolism through the interaction between HCV 3-untranslated region (3-UTR) and DDX3X, and in turn activated IKKα,which enhanced core-associated lipid droplet formation for viral assembly. The mechanistic basis of these functional effects still needs far more investigations to be clear.. Although hepatic steatosis always can be found in HCV infected patients (9,10), and we also can be sure that lipid generation and metabolism is strongly associated with HCV infection, the molecular mechanisms that related with lipogenises during HCV infection are poorly understood.
Generaly speaking, HCV genome encodes a polyprotein of more than 3,010 amino acids that is cleaved at endoplasmic reticulum (ER) by host and viral proteases, including structural and nonstructural proteins. NS4B, one of nonstructural proteins, 27 kDa, a hydrophobic protein locating through the ER membrane (11,12) with N-terminal tail, functions remain confused. Elazar, et al. (13) found that membrane anchoring could be mediated by N-terminal 26 residues of NS4B, which were important for HCV replication in vitro experiment. Furthermore, NS4B also be related with NIH-3T3 cells transformation with Ras (14) or without Ras (15).
HCV infection can significantly affects lipid-related metabolism in vivo including lipids and circulating lipoprotein, which has been indicated by clinical research evidence (16). The liver is the main center of lipid production via uptaking lipoproteins and deliver lipids to other organs (17). HCV infected patients often suffer from type Ⅱdiabetes cause the insulin resistance, one of pathological features in HCV infected patients. Furthermore, hepatic fibrosis, steatosis, hepatocellular carcinoma and resistance to anti-viral treatment(18) also found to be related with insulin resistance. Recent evidences showed that hepatic steatosis can be induced by HCV core protein. They found that the microsomal triglyceride transfer protein activity and very low density lipoprotein secretion (19) were inhibited by HCV core protein, and then expression and transcriptional activity of peroxisome proliferatorsactivated receptor (PPAR)α (20) was impaired,resulting in the SREBP1 and PPARα(21) activation.
SREBPs are endoplasmic reticulum binding transcription factors that up-regulate the activation of enzymes that facilitate the cholesterol and fatty acids synthesis, and lipoproteins cellular uptake. Park and coworkers demonstrated that HCV NS4B modulated SREBP1 via the AKT pathway (22). Wu, et al. (23) demonstrates that HCV NS4B protein activates PI3K/AKT pathway via up-regulating Snail in HCC and inhibiting Hippo signal pathway.
In this present study, our findings showed that NS4B inhibited Merlin and Yap phosphorylation, and also up-regulated AKT phosphorylation to lead to lipogenesis. The results may provide a novel mechanism of hepatic steatosis associated with HCV infection.
pCDNA3.1-NS4B was kindly send by Yi-Zheng (Huadu District People's Hospital, China). Full-length Merlin sequence was inserted into pFLAG-CMV2 vectors to construct pFLAG-CMV2-Merlin. PCR and DNA sequencing detection methods were used to testify the integrity and correctness of recombinant plasmids. The plasmids were transfected into cells by lipofectamine-2000 (Invitrogen, Karlsruhe, USA) according to manufacturers' instructions when the cells confluence approximately 80%.

Western blotting analysis
Cells were lysed in cell lysis buffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 10% glycerol, and protease inhibitor mixture) for 20 min on ice. ECL kit (Abcam) was purchased to detecte proteins. The equivalent amount of protein was electrotransferred to the nitrocellulose membrane with 10% sds-page treatment. Phosphate-buffered saline (PBS) including 5% nonfat dry milk was used to block the membrane for 1 h and then hatch 2 h at room temperature with one of

Oil red O test
When cultured cell density reached 5-6*10 5 cells/ml, medium drained and cells were washed twice with 0.01 M PBS, The cells were fixed with phosphate buffer containing 10% formaldehyde for 10 mins, rinsed with PBS for 1min and 60% isopropanol for 15 seconds. The cells were stained with filtered Oil red O working solution with light avoid for 1min at 37℃, rinsed with 60% isopropanol for 5 15 seconds and PBS 3 times for 3 min. Cells stained with hematoxylin counterstain for 1 min and rinsed with PBS 3 times for 3 min. Tablet was sealed and observed under microscope.

Statistical analysis
Strip grayscale analysis (Quantity one) was processed to analyse the protein western bot results. All experimental data acquired the mean of three independent experiments (n=3) and expressed as the mean ± standard error.. Tukey's test was used to determine difference between different groups. All data were analyzed with SPSS(22.0) software. P<0.05 was considered statistically difference.

1.
NS4B regulates Hippo pathway via NF2  Fig.1 A), which indicated that NS4B could medicate Merlin expression form gene level. Next, western blot assays verified the gene tests. As shown in Fig.1  h later when different amount of NS4B plasmid transfected( Fig.1 E,F,). These results showed that HCV NS4B do mediate Hippo signal pathway.
Furthermore, we also detected some protein expression or phosphorylation of AKT signal pathway.

Merlin mediated AKT signal pathway
We have found HCV NS4B can regulate Hippo signal pathway and AKT signal pathway, we hypothesized that these two pathways connected here. To test this hypothesis, We transfected different amount of Merlin expression plasmid into Huh7.5 cells and then examined the protein expression and phosphorylation levels of AKT and NF-κB. Increased expression of Merlin significantly reduced p-AKT and NF-κB, however total AKT expression has no alternation( Fig.2 A-B ). We also These results suggested that HCV NS4B mediated AKT signal pathway via inhibiting Merlin-Hippo signal pathway.

NS4B and Merlin induces lipogenesis and SREBP-1 expression
In order to test whether Merlin has a regulatory effect on lipid metabolism in Huh7.5 cells, we verified by Oil red O tests with high expression of Merlin and silence of NF2 gene in Huh7.5 cells, and high expression of NS4B as control. In Huh7.5 cells, lipid droplets significantly increase after pCDNA3.1-NS4B transfection (Fig.3 B,F), whereas the lipogenesis decreased after Merlin over-expression (Fig.3 C,F). Lipid droplets reduced in cells expressing NS4B after transfection with Merlin palsmid (Fig.3 A,F).

G-J) and
Si-NF2 (Fig.3 M,N), and decreased by Merlin (Fig.3 K,L), which consistent with our Oil red O results (Fig.3 A-F). Altogether, these results implied that NS4B works in the context of HCV to trigger lipogenesis via Merlin reduced.

Discussion
Hippo pathway plays a pivotal role in regulating tissue growth size through mediating cell growth, proliferation and apoptosis (24,25). Recently, researchers reported that the effect of Hippo pathway can be found in different cell types including hepatic cells (26). They found that the change of expression and phosphorylation levels of Yap, one of the key substrates of Hippo pathway, would regulate the growth of hepatic tissue directly (27,28), and this is partially why Hippo pathway is one of the important pathway in the mediating mechanism of hepatic cell. Merlin was reported to be a tumor suppressor (29) and Merlin significantly connected with key factors of Hippo pathway (30,31). In this paper we detected Merlin protein expression level in Huh7.5 cells transfected with HCV NS4B plasmid.
Our results showed that Merlin was down-regulated( Fig. 1A-F). We speculate that the Hippo pathway probably participates in the mechanism of cell biology affected by NS4B. To identify this hypothesis, we also detected the phosphorylation level of Yap and found that the phosphorylation level of Yap exhibited a positive correlation with NS4B protein level (Fig. 1A-F There is no doubt about HCV is strongly related with lipid accumulation in hepatic tissue. Hepatic steatosis can be found in 40%-86% HCV infected patients. Some studies had reported the correlation between HCV and lipid metabolism in hepatic cells (10,17,21,33,34). NS4B was reported to accelerate the lipid synthesis through AKT pathway (35). In this paper, we checked the effect of NS4B on AKT pathway and the results (Fig. 1G-L) were consistent with the previous reports (35). However, there are no more reports about the mechanism about the effect of NS4B on AKT at present. Interestingly, there are some reports about the correlation of Merlin with AKT pathway (36,37), therefore, we conjectured that Merlin is involved in the effect of NS4B on AKT pathway. To test this hypothesis, we measured the key proteins expression and activity of AKT pathway in Huh7.5 cell line after Merlin being over-expressed or silenced separately. The results showed that the total AKT expression could not be influenced by Merlin (Fig. 2). But the activity of AKT has negative correlation with Merlin 8 expression (Fig. 2). These results indicated that Merlin might be one of the upstream regulators of AKT pathway, and NS4B probably mediates AKT pathway via Merlin.The roles of AKT pathway in the synthesis of lipid droplets and Merlin in the synthesis of fatty acid had been confirmed (38). We believed that Merlin was related with lipid droplets synthesis as well. We detected the lipid droplets in Huh7.5 cell line after Merlin being over-expressed or silenced separately. We find that the lipid droplet volume is obviously up-regulated when NS4B over-expressed in Huh7.5 cells (Fig. 3B,F).
Interestingly, we also found that the lipid droplet volume is obviously up-regulated when Merlin is silenced (Fig. 3D,F), but the lipid droplet volume is obviously down regulated when Merlin is overexpressed (Fig. 3C,F). At the same time, to confirm this correlation, we measured the lipid droplets related protein, SREBP-1. Results showed that the SREBP-1 protein can be up-regulated in Huh7.5 cells either NS4B within( Fig. 3G-J) or Merlin gene is silenced (Fig. 3M,N). Conversely, SREBP-1 was down-regulated in Huh7.5 cells when Merlin was over-expressed (Fig. 3K,L).
Although the relationship between NS4B and lipid droplets still needs more research and we also realized that it is still too early to draw a final conclusion. We believe that HCV NS4B is one of the related proteins with lipid droplets in Huh7.5 cells and in view of the reported results(39), we speculate the Hippo pathway might be a bridge between NS4B and AKT pathway.

Sponsored by the innovation and promotion project of the First Affiliated Hospital of Guangzhou
University of Chinese Medicine: 2015QN08.

Conflict of interest
The authors declare no conflicts of interest

Availability of data and materials
The datasets used and/or analyzed during the present study are available from the corresponding author upon reasonable request.