In this study, we showed that HCV alters sphingolipid metabolism, resulting in a better environment for viral replication. Specifically, HCV increased SM content in the DRM fraction; this step is essential for viral replication since SM is a key component of the membranous replication complex and interacts with RdRp. Employing MS analysis, we identified endogenous SM molecular species (located in the DRM fraction) that increased upon HCV infection, and demonstrated that these endogenous SM molecular species interact directly with RdRp, enhancing HCV replication. Thus, we concluded that HCV modulates sphingolipid metabolism to promote viral replication.
We found that the expression levels of SGMS1/2 and the content of SM and ceramide in HCV-infected humanized chimeric mouse livers was increased (). Our measurement revealed that chronic HCV infection promoted sphingolipid biosynthesis. HCV is known to induce cellular stress 
. A variety of cell stressors increase intracellular ceramide content during the execution phase of apoptosis 
, indicating that ceramide is a proapoptotic lipid mediator. Furthermore, activation of ceramide-metabolizing enzymes such as glucosylceramide synthase and SM synthase can attenuate apoptosis by decreasing the intracellular ceramide content 
. We found that HCV infection correlated with increased mRNA levels of the genes that encode human SM synthases (SGMS1/2
) and glucosylceramide synthase (UGCG
) (data not shown). Thus, the increase in ceramide levels observed in our study was likely to activate enzymes that transfer ceramide to other sphingolipids. On the other hand, Diamond et al. reported on lipidomic profiling performed over the time course of acute HCV infection in cultured Huh-7.5 cells and observed that specific SM molecular species were decreased 72 h after HCV infection 
. Given that their study focused on acute HCV infection, the reason for this discrepancy may be due to the severity of infection, suggesting that the influence of HCV infection on sphingolipid metabolism differs between acute and chronic infections. We also demonstrated that HCV infection correlates with increased abundance of specific SM and ceramide molecular species, with the profiles of individual lipids differing for infection by HCG9 (genotype 1a) and HCR24 (genotype 2a). The precise mechanism and meaning of these differences remain to be elucidated.
Our results indicated that SGMS1 expression had a correlation with HCV replication. This indicates that SM synthesized by SGMS1 contributes to HCV replication. A previous report revealed that in cultured cell lines, SGMS1 localizes in Golgi apparatus while SGMS2 localizes in the plasma membrane 
. Thus, the results of this previous report suggest that SMs synthesized by SGMS1 can be easily incorporated into membranous replication complexes. As for SGMS2, we found that HCV infection significantly increased the expression of SGMS2, although the relationship between SGMS2 and HCV replication was hardly seen in this study. The relationship between SGMS2 and HCV propagation, thus, is an issue that should be elucidated in future studies.
We also demonstrated in this study that reduction of SM molecular species by NA808, a hepatotropic SPT inhibitor with little immunosuppressive activity, inhibits HCV replication in humanized chimeric mice regardless of viral genotype (). Notably, treatment with NA808 (5 mg/kg) restored SM and ceramide levels in the liver to the levels observed in uninfected chimeric mice (). Apparently, a slight reduction in SM had a significant influence on HCV, indicating that SM plays an important role in the HCV life cycle. SM is required for many viral processes in host-pathogen interactions 
. For instance, viral envelopes of human immunodeficiency virus type 1 (HIV-1) and herpes simplex virus (HSV) are enriched with SM, which is necessary for efficient virus infectivity 
. With regard to HCV, in addition to efficient virus infectivity 
, SM is present in the raft domain, which serves as a site of virus replication, together with other sphingolipids and cholesterol 
. Moreover, SM is a component of VLDL whose assembly component and pathway is required for HCV morphogenesis and secretion 
. The above-mentioned observations suggest that SM plays a multifaceted role in the HCV life cycle; therefore, SM is likely to be a good therapeutic target.
HCV is thought to replicate in a specialized compartment characterized as a DRM (designated as the membranous replication complex) 
. SM, cholesterol, and phosphatidylinositol (PI) are thought to be the lipids that make up the membranous replication complex. With regard to PI, several siRNA screening have recently identified type III phosphatidylinositol 4-kinases (PI4K) as crucial host factors for HCV replication 
. In HCV replicon containing cells, PI4P distribution is altered and enriched in the membranous replication complex by PI4KIIIα synthesis. Although the ability of PI to influence membrane bending and regulate intracellular processes (e.g. vesicle fusion, budding, and sorting) has been reported, the role of PI4P in the formation of the membranous replication complex remains to be elucidated. SM and cholesterol organize the solid membrane characterized as the DRM, where HCV replicates 
. In fact, we and other groups demonstrated that reduction of SM and cholesterol suppressed HCV replication 
. We performed the immunofluorescent analysis using lysenin. However, lysenin did not co-localize with NS4B protein. To date, it has been reported that lysenin-binding to SM is increased in the form of SM clusters, and that glycosphingolipids hinder lysein-binding to SM 
Lipid rafts form of HCV replication complex do not have the characters of lysenin-binding to SM.
Further, the role of SM is not only to act as a constituent of the membranous replication complex, but also to bind and activate RdRp 
. In this study, to gain further insight into the HCV membranous replication complex, we attempted to analyze which SM molecular species comprise the membranous replication complex, given that the diversity of molecular species is believed to be responsible for the physiochemical properties of the biomembrane 
(). We found that the composition ratio of SM molecular species observed in this study was quite different between the whole cell and DRM fractions. Further, to identify whether these SM molecular species contribute to HCV replication, we conducted rescue experiments using HCV replicon-containing cells (carrying intact RdRp and active membranous replication complexes) in which each SM molecular species was extrinsically added to replicon cells treated with NA808. However, in this experiment, addition of SM caused cell death. Therefore, we used digitonin-permeabilized semi-intact replicon cells, which enabled us to deliver the extrinsically added SM molecular species directly to the cytosol without catalytic effect and permitted monitoring of intact RdRp and replication complexes. We demonstrated that the specific endogenous SM molecular species (d
0 and d
0) enhance HCV-RNA replication, these species being consistent with the two SM molecular species which mainly constitute the DRM. Collectively, these results suggest that the HCV replication complex characterized as DRM is the specialized compartment that is composed of SM molecular species. These findings will provide new insights into the formation of the HCV replication complex and the involvement of host lipids in the HCV life cycle.