Roles for endocytic trafficking and phosphatidylinositol 4-kinase III alpha in hepatitis C virus replication

Abstract

Hepatitis C virus (HCV) reorganizes cellular membranes to establish sites of replication. The required host pathways and the mechanism of cellular membrane reorganization are poorly characterized. Therefore, we interrogated a customized small interfering RNA (siRNA) library that targets 140 host membrane-trafficking genes to identify genes required for both HCV subgenomic replication and infectious virus production. We identified 7 host cofactors of viral replication, including Cdc42 and Rock2 (actin polymerization), EEA1 and Rab5A (early endosomes), Rab7L1, and PI3-kinase C2gamma and PI4-kinase IIIalpha (phospholipid metabolism). Studies of drug inhibitors indicate actin polymerization and phospholipid kinase activity are required for HCV replication. We found extensive co-localization of the HCV replicase markers NS5A and double-stranded RNA with Rab5A and partial co-localization with Rab7L1. PI4K-IIIalpha co-localized with NS5A and double-stranded RNA in addition to being present in detergent-resistant membranes containing NS5A. In a comparison of type II and type III PI4-kinases, PI4Ks were not required for HCV entry, and only PI4K-IIIalpha was required for HCV replication. Although PI4K-IIIalpha siRNAs decreased HCV replication and virus production by almost 100%, they had no effect on initial HCV RNA translation, suggesting that PI4K-IIIalpha functions at a posttranslational stage. Electron microscopy identified the presence of membranous webs, which are thought to be the site of HCV replication, in HCV-infected cells. Pretreatment with PI4K-IIIalpha siRNAs greatly reduced the accumulation of these membranous web structures in HCV-infected cells. We propose that PI4K-IIIalpha plays an essential role in membrane alterations leading to the formation of HCV replication complexes.

Fig. 1.

Effect of drug inhibitors on HCV replication. Huh-7.5 cells containing HCV-Con1 replicons were treated with the indicated concentrations of the inhibitors (A) Cytochalasin D or (B) LY294002 for 48 h. HCV RNA levels were measured by real-time RT-PCR and normalized to 18S RNA levels. SEM is shown. *, P < 0.05, as compared with untreated.

Fig. 2.

Localization of HCV replication complexes with Rab5A, Rab7L1, and PI4K-IIIα. Huh-7.5 cells were transfected with GFP-Rab5A, GFP-Rab7L1, or PI4KIIIα-GFP constructs for 24 h, then infected with HCV for 48 h. Cells were fixed and probed with antibodies for (A) dsRNA and (B) NS5A, detecting HCV replication complexes. Insets are zoomed images from boxed areas showing regions of colocalization.

Fig. 3.

Co-fractionation of endogenous PI4K-IIIα with HCV replication complexes. Detergent-resistant membranes were prepared from Huh-7.5 cells with and without HCV replicons by TritonX-100 treatment at 4 °C as described (33). Detergent-resistant (DR) membranes remain at the top of the sucrose gradient (fractions 1–3), whereas detergent-sensitive (DS) membranes sediment in the lower fractions (–9). Protein lysates from pooled fractions were run on SDS/PAGE and probed for PI4K-IIIα, HCV NS5A, calnexin, and caveolin-2 (Cav-2). M, membrane; NM, non-membrane; *, PI4K-IIIα–specific band.

Fig. 4.

PI4K-IIIα is specifically required in HCV replication. Type II (α and β) and type III (α and β) PI4Ks were silenced using a pool of 4 individual siRNAs or, in the case of PI4K-IIIα, a pool and individual siRNAs (nos. 1–3). The effects of these siRNAs on (A) HCV pseudoparticle entry or (B) subgenomic replicon replication were quantified by luciferase activity. si-HCV specifically targets HCV RNA. si-CD81 is a control for HCV entry. Values are relative to irrelevant si-IRR–treated cells. SEM is shown. n = 10. **, P < 0.001 and *, P < 0.05, as compared with IRR.

Fig. 5.

PI4K-IIIα is required for the replication of HCV RNAs but not for their translation. Huh-7.5 cells were treated for 48 h with either irrelevant (IRR, solid line) or individual or pooled PIK4CA siRNAs (dashed lines). Luciferase values were measured over the indicated time course following transfection of (A) sgJFH1-RLuc or (B) sgJFH1-RLuc-GND RNAs. Initial translation of the transfected RNAs occurs between 4 and 24 h as confirmed by cycloheximide (CHX) treatment, which significantly (P < 0.05) reduced luciferase activity in (B). Replication occurs in concert with translation of newly synthesized HCV RNAs between 24 and 48 h. SEM is shown. n = 3. Statistically significant differences (*, P < 0.05) between IRR and PIK4CA siRNA-treated cells occurred only at the 48-h time point in (A).

Fig. 6.

Silencing PI4K-IIIα inhibits membranous web formation. Huh-7.5 cells were treated with (A) irrelevant or (B) PIK4CA siRNAs for 2 days and then infected with 2 infectious HCV particles per cell for 3 days. Cells were fixed and processed for EM. Magnification is 8,260×. (Scale bar, 1 μm.) C and D are higher-magnification images of the cell insets in A and B, respectively. LD, lipid droplet. Magnification is 21,000×. (Scale bar, 200 nm.)