The acidic domain of the hepatitis C virus NS4A protein is required for viral assembly and envelopment through interactions with the viral E1 glycoprotein

Abstract

Hepatitis C virus (HCV) assembly and envelopment are coordinated by a complex protein interaction network that includes most of the viral structural and nonstructural proteins. While the nonstructural protein 4A (NS4A) is known to be important for viral particle production, the specific function of NS4A in this process is not well understood. We performed mutagenesis of the C-terminal acidic domain of NS4A and found that mutation of several of these amino acids prevented the formation of the viral envelope, and therefore the production of infectious virions, without affecting viral RNA replication. In an overexpression system, we found that NS4A interacted with several viral proteins known to coordinate envelopment, including the viral E1 glycoprotein. One of the NS4A C-terminal mutations, Y45F, disrupted the interaction of NS4A with E1. Specifically, NS4A interacted with the first hydrophobic region of E1, a region previously described as regulating viral particle production. Indeed, we found that an E1 mutation in this region, D72A, also disrupted the interaction of NS4A with E1. Supernatants from HCV NS4A Y45F transfected cells had significantly reduced levels of HCV RNA, however they contained equivalent levels of Core protein. Interestingly, the Core protein secreted from these cells formed high order oligomers with a density matching the infectious virus secreted from wild-type cells. These results suggest that this Y45F mutation in NS4A causes secretion of low-density Core particles lacking genomic HCV RNA. These results corroborate previous findings showing that the E1 D72A mutation also causes secretion of Core complexes lacking genomic HCV RNA, and therefore suggest that the interaction between NS4A and E1 is involved in the incorporation of viral RNA into infectious HCV particles. Our findings define a new role for NS4A in the HCV lifecycle and help elucidate the protein interactions necessary for production of infectious virus.

Fig 1. A Y45F mutation in hepatitis C virus NS4A causes a decrease in infectious viral titer.

(A) Schematic of the NS4A protein. * indicates location of Y45F mutation. Numbers correspond with the amino acid position within NS4A (aa 1–54) or the full-length polyprotein (aa 1690–1711). Strain names are listed as found in the Los Alamos HCV sequence database. (B) Focus forming assay of supernatants harvested from Huh7.5 cells at the indicated days post-electroporation with HCV WT or NS4A Y45F in vitro transcribed RNA. FFU/mL = focus forming units/milliliter. Values are presented as mean ± SD (n = 3). Data are representative of three independent experiments. N.D. = not detected. L.D. = limit of detection. (C) Sequencing of the NS4A region (nt 4710–5251) of cDNA amplified from Huh7.5 cells transfected with HCV WT or NS4A Y45F RNA (JFH1) in (B) at indicated days.

Fig 2. The NS4A Y45F mutation in HCV does not alter viral RNA replication.

(A) Renilla luciferase assay to measure HCV replicon luciferase reporter (JFH1-SGR-luc) activity from Huh7.5 cells at indicated times following electroporation. GND = lethal mutation in HCV NS5B RNA-dependent RNA polymerase. RLU = Renilla luciferase units. Data are presented as mean ± SEM (n = 3). (B) Immunoblot analysis of extracts of Huh7.5 cells at 72 hours post-transfection with indicated HCV RNA. (C) Immunoblot analysis of anti-Flag immunoprecipitated extracts and whole cell lysates (WCL) from Huh7.5 cells transfected with indicated tagged HCV proteins or vector.

Fig 3. The NS4A Y45F mutation inhibits HCV envelopment.

(A, B) Focus forming assay of supernatants for extracellular titer, or cellular lysates for intracellular titer, respectively, from Huh7.5 cells at 48 hours post-electroporation with HCV WT, Y45F, or GND (JFH1) in vitro transcribed RNA. N.D. = not detected. L.D. = limit of detection. (C) Immunoblot analysis for HCV Core protein of cell lysates of Huh7.5 cells at 48 hours post-electroporation with in vitro transcribed HCV RNA subjected to the indicated treatments in a proteinase K protection assay. ΔE1/E2 = complete deletion of E1 and E2 coding regions. For panels A and B, data are presented as mean ± SEM (n = 3). C is representative of 3 independent experiments.

Fig 4. Multiple amino acids in the acidic domain of NS4A are required for HCV envelopment.

(A) Immunoblot analysis of HCV Core protein from lysates of Huh7.5 cells at 48 hours post-electroporation of in vitro transcribed HCV RNA (JFH1) containing the indicated mutations in the NS4A region, after the indicated treatments in a proteinase K protection assay. (B) Quantification of percentage of protease-resistant Core relative to untreated Core from (A). ΔE1/E2 = deletion in the E1 and E2 coding regions. Data are presented as mean ± SEM (n = 3) and were analyzed by one-way ANOVA. *P < 0.05, **P < 0.01. (C) Renilla luciferase assay to measure HCV replicon luciferase reporter (JFH1-SGR-luc) activity from Huh7.5 cells following electroporation of indicated constructs, with individual replicates from three experiments plotted. GND = lethal mutation in HCV NS5B RNA-dependent RNA polymerase. RLU = Renilla luciferase units. (D) Immunoblot analysis of lysates from the 72-hour time point of (C). (E, F) Focus forming assay of supernatants for extracellular titer or cellular lysates for intracellular titer, respectively, from Huh7.5 cells at 48 hours post-electroporation with in vitro transcribed HCV RNA (JFH1) containing the indicated mutations in NS4A. N.D. = not detected. L.D. = limit of detection. For (B, E, F), data are presented as mean ± SEM (n = 3). Panel A is representative of three independent experiments.

Fig 5. NS4A Y45 is required for NS4A interaction with the E1 glycoprotein.

(A) Immunoblot analysis of anti-Flag immunoprecipitated extracts from Huh7.5 cells transfected with indicated HCV proteins or vector. (B) Quantification of the signal of NS4A relative to E1 from the immunoprecipitation from (A). Data are presented as mean ± SEM (n = 3) and were analyzed by unpaired t-test. ***P< 0.001. (C) Immunoblot analysis of anti-HA or anti-IgG immunoprecipitated extracts from Huh7.5 cells at 48 hours post-electroporation with in vitro transcribed HCV HJ3-E1/HA-NS2/YFP [41] RNA which contains an N-terminal HA tag on E1. Panel C is representative of three independent experiments.

Fig 6. NS4A binds the first hydrophobic region of E1.

(A) Schematic of the HCV E1 protein with functional domains indicated and amino acids numbered based on the E1 coding region of JFH1. (B) Immunoblot analysis of anti-Flag immunoprecipitated extracts and whole cell lysate (WCL) from Huh7.5 cells transfected with NS4A-HA WT and Flag-tagged E1, either full-length (WT) or as indicated. Data is representative of 3 independent experiments. (C) Immunoblot analysis of anti-Flag immunoprecipitated extracts and whole cell lysate (WCL) from Huh7.5 cells transfected with Flag-tagged E1 WT or D72A and NS4A-HA or vector. Panel C is representative of 2 independent experiments.

Fig 7. The NS4A Y45F mutation results in production of partially formed virions.

Supernatants harvested from Huh7.5 cells 48 hours post-electroporation of NS4A WT, NS4A Y45F, or ΔE1/E2 in vitro transcribed HCV RNA (JFH1) were analyzed for HCV RNA by RT-qPCR (A) or Core protein by ELISA (B). Data in A are presented as mean ± SEM (n = 3) and analyzed by unpaired t-test. ****P<0.0001. Data in B are representative of 2 independent experiments and are presented as mean ± SD (n = 2). Supernatant harvested from Huh7.5 cells 48 hours post-electroporation with in vitro transcribed HCV WT or NS4A Y45F RNA (JFH1) were concentrated, fractionated over a 10–50% iodixanol gradient, and collected into 10 equal fractions. Fractions were analyzed by focus-forming assay for infectivity, represented as FFU/mL, and RT-qPCR for HCV RNA, shown as a measure of copy number (C) and by immunoblot for Core (D). In the bottom panels of the immunoblots, fractions were treated with 2-mercaptoethanol and DTT and then boiled before immunoblotting for anti-Core. Fractions 1–10 correspond with fractions running from top to bottom of the gradient, with the densities listed at the bottom. Data in C is presented as mean ± SD (n = 3); C-D are representative of 2 independent experiments.