The Ebola virus VP35 protein is a suppressor of RNA silencing

Abstract

RNA silencing or interference (RNAi) is a gene regulation mechanism in eukaryotes that controls cell differentiation and developmental processes via expression of microRNAs. RNAi also serves as an innate antiviral defence response in plants, nematodes, and insects. This antiviral response is triggered by virus-specific double-stranded RNA molecules (dsRNAs) that are produced during infection. To overcome antiviral RNAi responses, many plant and insect viruses encode RNA silencing suppressors (RSSs) that enable them to replicate at higher titers. Recently, several human viruses were shown to encode RSSs, suggesting that RNAi also serves as an innate defence response in mammals. Here, we demonstrate that the Ebola virus VP35 protein is a suppressor of RNAi in mammalian cells and that its RSS activity is functionally equivalent to that of the HIV-1 Tat protein. We show that VP35 can replace HIV-1 Tat and thereby support the replication of a Tat-minus HIV-1 variant. The VP35 dsRNA-binding domain is required for this RSS activity. Vaccinia virus E3L protein and influenza A virus NS1 protein are also capable of replacing the HIV-1 Tat RSS function. These findings support the hypothesis that RNAi is part of the innate antiviral response in mammalian cells. Moreover, the results indicate that RSSs play a critical role in mammalian virus replication.

Figure 1. RNAi Suppression by VP35, NS1, and E3L in Mammalian Cells

HEK293T cells or Vero cells were transfected with an expression plasmid for firefly luciferase (Luc), a vector expressing an shRNA against luciferase (shLuc), expression constructs for various RSSs (VP35, NS1, E3L), or a control GFP expression plasmid. Luciferase expression was measured at 2–3 d post transfection. (A and B) Effect of VP35 (100, 300, and 600 ng) on the expression of the silenced luciferase reporter in HEK293T cells (A) and (B) Vero cells. (C and D) Effect of NS1, E3L, and GFP (600 ng) on expression of the silenced luciferase reporter in HEK293T cells (C) and in (D) Vero cells.

Figure 2. VP35 dsRNA-Binding Capacity Is Required for RSS Activity

(A) HEK293T cells were co-transfected with expression vectors for firefly luciferase (Luc), shLuc, the EBOV VP35^wt, or the various mutants (K309A, R312A, and K309/R12A), or a C-terminal deletion mutant R300T (50 ng). The renilla expression plasmid pRL-CMV was co-transfected as internal control. Luciferase expression was measured at 2–3 d post transfection. From the top down, the panels show the relative expression of firefly luciferase, renilla luciferase, and the ratio firefly/renilla. (B) HEK293T cells were co-transfected with expression vectors for firefly luciferase, renilla luciferase, and either the control shLucR, or the active shLuc in the presence and absence (−) of VP35^wt expression plasmid. Luciferase expression was measured at 2–3 d post transfection. Shown is relative luciferase expression corrected for the internal renilla control (firefly/renilla).

Figure 3. The Tat-Minus HIV-1 Complementation System

(A) Schematic of the HIV-rtTA genome. In HIV-rtTA, the Tat-TAR transcriptional axis has been inactivated by mutation of both the Tat protein and the TAR hairpin (as indicated), and replaced by the tetracycline-inducible tetO-rtTA system. For this, the rtTA gene was inserted in place of the Nef gene and eight tetO sites were inserted in the LTR promoter. Upon administration of dox, rtTA can bind to tetO and activate transcription and viral replication. A frame shift mutation was introduced at codon 20 of the Tat gene. (B) Virus production in HEK293T cells transfected with HIV-rtTA-Tat^wt or HIV-rtTA-Tat^fs in the absence (−) or presence (+) of dox. CA-p24 in the culture supernatant was measured at 3 d post transfection.

Figure 4. Tat-Mediated RNAi Suppression Is Essential for HIV-rtTA Virus Production

(A) TZM-bl cells containing a Tat-responsive firefly luciferase reporter gene under control of the HIV-1-LTR were transfected with pBluescript (−), Tat^wt, Tat^F32A, or Tat^Y26A. Luciferase expression was determined 2 d post transfection. (B) Transfection of HEK293T cells with the luciferase reporter expression plasmid (Luc), the vector expressing an shRNA against luciferase (Luc + shLuc), and expression plasmids for Tat^wt , Tat^F32A, or Tat^Y26A. Luciferase expression was determined 2 d post transfection. (C) Virus production 2 d post transfection in HEK293T cells transfected with HIV-rtTA-Tat^fs and expression plasmids for Tat^wt, Tat^F32A, or Tat^Y26A.

Figure 5. Functional Complementation of the Tat RSS Function by VP35, NS1, and E3L

(A and B) HEK293T cells were transfected with HIV-rtTA-Tat^wt, HIV-rtTA-Tat^fs, and expression plasmids for (A) Tat^wt, VP35, E3L, and NS1 (50 ng) or (B) E3L, NS1, and GFP (10, 100, and 500 ng). Virus production was determined 2 d post transfection. (C) To measure transcriptional transactivation capacity of the various proteins, HEK293T cells were transfected with a firefly luciferase reporter under control of the 8tetO promoter (similar to the promoter in the HIV-rtTA constructs), renilla expression vector pRL-CMV, an expression plasmid for rtTA to activate the promoter in the presence of dox, and the indicated amounts of RSSs. Luciferase expression (plotted as firefly/renilla) was determined 2 d post transfection. −, pBluescript (negative control). (D) Transcriptional transactivation capacity of the various RSS proteins was measured using TZM-bl cells that contain a Tat-responsive firefly luciferase reporter gene under control of the HIV-1-LTR. The cells transfected with 0.2 μg of Tat or the indicated RSS expression plasmids. Two to three days after transfection, luciferase expression was measured. −, pBluescript (negative control).

Figure 6. Analyses of HIV-rtTA-Tat^fs Genomic RNA Accumulation Complemented with Tat and VP35

(A) HEK293T cells were transfected with HIV-rtTA-Tat^wt, HIV-rtTA-Tat^fs, and HIV-rtTA-Tat^fs in combination with Tat^wt or increasing amounts of VP35. Two to three days after transfection, virus production was measured via CA-p24 ELISA, and total RNA was isolated. (B) Total RNA (10 μg) was run on a denaturing agarose gel, and after blotting, probed with a Nef-specific radiolabeled probe. Relative genomic RNA accumulation was quantified by phosphorimaging and is indicated below. The RNA accumulation of the HIV-rtTA-Tat^wt was set at 100%.

Figure 7. VP35 dsRNA-Binding Capacity Is Required for trans Complementation of HIV-rtTA-Tat^fs

HEK293T cells were transfected with HIV-rtTA-Tat^wt, HIV-rtTA-Tat^fs, and expression plasmids for Tat^wt, VP35^wt, or the various VP35 mutants (K309A, R312A, and K309/R12A), or a C-terminal deletion mutant R300T (50 ng). Virus production was determined 2 d post transfection.

Figure 8. Antiviral RNAi Activity and IFN Responses Function together in Innate Antiviral Defences

During virus replication, virus-specific dsRNA molecules are generated that are recognized by Dicer and processed into siRNAs. One strand of the siRNA, the guide-strand, is loaded into RISC, which targets viral RNAs for destruction. To escape this antiviral pressure, viruses encode RSS factors that block the RNAi pathway. When virus-specific dsRNA molecules and siRNAs accumulate above a certain threshold, and consequently can no longer be masked by the viral RSS, the cytoplasmatic dsRNA sensors RIG-I/MDA5, PKR, and 2′5′ OAS/RNAseL are activated. This results in the activation of range antiviral responses, including the production of type I IFNs, general translational inhibition, and RNA degradation. This model clarifies why RSSs like NS1, VP35, and E3L have been identified as IFN antagonists.