The Biogenesis of Dengue Virus Replication Organelles Requires the ATPase Activity of Valosin-Containing Protein

Abstract

The dengue virus (DENV) causes the most prevalent arthropod-borne viral disease worldwide. While its incidence is increasing in many countries, there is no approved antiviral therapy currently available. In infected cells, the DENV induces extensive morphological alterations of the endoplasmic reticulum (ER) to generate viral replication organelles (vRO), which include convoluted membranes (CM) and vesicle packets (VP) hosting viral RNA replication. The viral non-structural protein NS4B localizes to vROs and is absolutely required for viral replication through poorly defined mechanisms, which might involve cellular protein partners. Previous interactomic studies identified the ATPase valosin-containing protein (VCP) as a DENV NS4B-interacting host factor in infected cells. Using both pharmacological and dominant-negative inhibition approaches, we show, in this study, that VCP ATPase activity is required for efficient DENV replication. VCP associates with NS4B when expressed in the absence of other viral proteins while in infected cells, both proteins colocalize within large DENV-induced cytoplasmic structures previously demonstrated to be CMs. Consistently, VCP inhibition dramatically reduces the abundance of DENV CMs in infected cells. Most importantly, using a recently reported replication-independent plasmid-based vRO induction system, we show that de novo VP biogenesis is dependent on VCP ATPase activity. Overall, our data demonstrate that VCP ATPase activity is required for vRO morphogenesis and/or stability. Considering that VCP was shown to be required for the replication of other flaviviruses, our results argue that VCP is a pan-flaviviral host dependency factor. Given that new generation VCP-targeting drugs are currently evaluated in clinical trials for cancer treatment, VCP may constitute an attractive broad-spectrum antiviral target in drug repurposing approaches.

Keywords: NS4B; dengue virus; endoplasmic reticulum; valosin-containing protein; viral replication organelles.

Figure 1

VCP ATPase inhibition impairs DENV replication. (A) Huh7.5 cells were infected with DENV2 16681s at a MOI of 0.05. The day after, the cells were treated with DMSO, 50 nM NMS-873, 0.5 µM CB-5083, or 10 µM NITD008 as positive control. After 24 h (2 days post-infection), cell supernatants were collected, and plaque assays were performed. (B) Huh7.5 cells were infected with the DENV2 16681s Renilla luciferase (Rluc)-expressing reporter virus (DVs-R2A) at a MOI of 0.001. One day post-infection, the infected cells were treated with various concentration of CB-5083 or NMS-873. After 24 h (2 days post-infection), Rluc assays were performed. Cell viability was measured in uninfected cells by CellTiter-Glo luminescent assays. The plotted data are relative to the DMSO treatment control. The dot line is a marker of a 50% decrease in viability or replication. NTR: non-translated region; 5′ CS: 5′ cyclization sequence; 2A: Tosea asigna virus 2A cleavage site at the C-terminus of Rluc to ensure proper processing after polyprotein synthesis. (C,D) Huh7.5 cells were transduced with lentiviruses expressing wild-type VCP or the depicted dominant negative VCP E305Q/E578Q mutant (MOI = 1). Four days post-transduction, the cells were collected and analyzed by Western blotting using the indicated antibodies (C) or subjected to MTT assays (D). (E,F) The cells were transduced as in (C). Two days post-transduction, the cells were infected with DENV2 16681s at a MOI of 0.05. Two days post-infection (four days post-transduction), extracellular infectious titers and relative intracellular DENV RNA levels were determined using plaque assays (E) and RT-qPCR (F), respectively. (G) Cells were transduced as in (C). Two days post-transduction, cells were infected with DVs-R2A at a MOI of 0.001. Two days post-infection (4 days post-transduction), viral replication was measured using Rluc assays. All results are representative of two or three independent experiments. **: p-value ≤ 0.01; ***: p-value ≤ 0.001.

Figure 2

VCP associates with the DENV NS4B in infected cells. (A) Huh7.5 cells were infected with DENV2 16681s (MOI = 1) or left uninfected. Two days later, cells were fixed, immunolabeled with anti-VCP and anti-NS4B antibodies, and imaged by confocal microscopy. The bottom right panels show single channel and merged images of the magnified area indicated with the dashed square (~2.4-fold magnification). (B) Huh7.5 were prepared exactly as in (A) and labeled with anti-NS4B and anti-dsRNA antibodies. White arrows indicate colocalization foci.

Figure 3

VCP associates with DENV NS4B when expressed alone as a precursor or a mature protein. (A) Huh7.5-T7 cells were transfected with a plasmid expressing DENV2 16681s NS4A-2K-NS4B-HA. Sixteen hours post-transfection, the cells were fixed, immunolabeled with anti-VCP and anti-NS4B antibodies, and imaged by confocal microscopy. Scale bar: 20 µm. (B) Huh7.5-T7 cells were transfected with the plasmids expressing the indicated viral proteins. Sixteen hours post-transfection, cells extracts were prepared and subjected to co-immunoprecipitation directed against HA. Resulting eluates and cell extracts were analyzed by Western blotting using the indicated antibodies. The DENV NS4A-2K-NS4B typically exhibits a lower molecular weight than expected because of a faster in-gel migration. The black dots indicate non-specific signals generated by the anti-NS4B antibodies. (C) Huh7.5-T7 cells were transfected with plasmids expressing the indicated DENV proteins. Sixteen hours post-transfection, the cells were fixed, immunolabeled with anti-VCP and anti-HA antibodies, and imaged by confocal microscopy. White arrows indicate colocalization foci. Scale bar: 20 µm.

Figure 4

VCP ATPase activity is required for the stability of the DENV convoluted membranes. (A) Huh7.5 cells were infected with DENV2 16681s at a MOI of 1. Two days post-infection, cells were treated with DMSO, 20 µM CB-5083, or 20 µM NMS-873. After a 4 h treatment, the cells were prepared for imaging by transmission electron microscopy. CM: convoluted membranes; VP: vesicle packets. (B,C) More than 33 cells from each condition were analyzed for CM abundance (B) and size (C). ***: p-value ≤ 0.001; NS: not significant. (D) Huh7.5 cells were infected with DENV2 16681s at a MOI of 1. Two days post-transfection, cells were treated with DMSO, 20 µM CB-5083 for four hours, or 20 µM NMS-873 for one hour. Cell extracts were prepared and analyzed by Western blotting with the indicated antibodies. The relative abundance of NS3 and NS4B (shown in red) was quantified after normalization to actin levels using the ImageLab software (Bio-Rad).

Figure 5

The DENV vesicle packet biogenesis requires VCP ATPase activity. (A) Schematic representation of the plasmid-induced DENV replication organelle system (pIRO-D) with a side-by-side comparison between the transcribed RNA with the DENV RNA genome. NTR: non-translated region; prom.: promoter; 5′ CS: 5′ cyclization sequence; IRES: internal ribosme entry site. (B–D) Huh7-Lunet-T7 cells were transfected with pIRO-D plasmid and, 4 h post-transfection, treated with DMSO, 2.5 µM CB-5083, or 1 µM NMS-873 for 12 h. (B) Fixed cells were labeled with anti-NS3 antibodies and imaged by widefield microscopy. (C) The transfection efficiency upon treatment with the different compounds was determined based on the % of NS3-positive cells. Mean and SD from two experiments are shown. (D) Total lysates of transfected and treated cells were analyzed by Western blotting using the indicated antibodies. The relative abundance of NS3 and NS4B (shown in red) was quantified after normalization to actin levels using the Fiji software. (E) Electron microscopy analysis of samples treated as in (B). (F) Quantification of the number of cells exhibiting the VPs. At least 15 cells from 2 different experiments have been analyzed for each condition. (G) The diameter of VPs was determined.