NUDT2 initiates viral RNA degradation by removal of 5′-phosphates

Figures

• 

  Fig. 1. Influence of Nudix depletion on virus growth. a Schematic representation of processes involved in 5′-to-3′ RNA degradation. Eukaryotic and bacterial triphosphorylated messenger (m)RNA are processed by proteins DCP2 and RppH, respectively, to generate monophosphorylated substrates for the 5′-3′ exonuclease XRN1, RNase E or RNase J. We hypothesize that a mammalian Nudix hydrolase could prepare viral RNA for degradation by removing pyrophosphate groups. b Hep3B cells with a CRISPR/Cas9-mediated KO of XRN1 or Hep3B cells transduced with a non-targeting-control vector (NTC) were infected with VSV-GFP at MOI 0.01 for 48 h. GFP expression was continuously quantified by live-cell microscopy. The data points are displayed as the mean of five technical replicates ± SD. One representative experiment of three is shown. c HeLa cells transfected with siRNA pools targeting the indicated Nudix gene for 48 h were infected with VSV-GFP (x-axis) or VSV-AV3-GFP (y-axis). Twenty-four hours after infection, GFP expression was analyzed by fluorometric analysis. The displayed z-scores were calculated from three independent reactions. d HeLa cells were treated with siRNA pools for 48 h targeting the…

• 

  Fig. 2. Identification of NUDT2 as PPP-RNA phosphatase. a RNA was incubated for 3 h with the indicated Nudix hydrolase, and the integrity of the RNA was determined using a Bioanalyzer. One representative experiment of three is shown. b γ-32P-radiolabeled in vitro transcribed 44-mer ssRNA substrate was left untreated or incubated for 90 min with 600 nM NUDT2, NUDT12, NUDT14 or NUDT17 (+) or was left untreated (−). Reaction mixtures were analyzed by TLC and autoradiography. γ-32P: radiolabeled γ-phosphate; RNA: remaining input RNA after incubation. c As (b) but incubation for the time indicated at the top of the panel. 600 nM NUDT2 were used. d As (c) but 600 nM NUDT2 E58A or NUDT2 were used. e As (b) but γ-32P-radiolabeled in vitro transcribed single-stranded (ss-) or double-stranded (ds-) RNA generated by annealing an antisense oligo, or a RIG-I activating hairpin RNA (IVT4) were incubated with 600 nM NUDT2 and 5 U RppH, respectively. f Phylogenetic analysis of all human Nudix hydrolases and bacterial RppH was performed using MAFFT. g Structural alignment of the crystal structures of RppH (brown) with an RNA substrate (red sticks) (PDB: 4S2Y) and NUDT2 (blue) (PDB: 3U53) using PYMOL57.…

• 

  Fig. 3. NUDT2 is active on a broad range of substrates. a EnzChek assay to quantify released phosphate from in vitro transcribed double-stranded triphosphorylated hairpin RNA (IVT4) left untreated (untreated) or incubated with recombinant NUDT2, NUDT E58A, NUDT14, or NUDT14 E121A (600 nM), as indicated. Colorimetric measurements of the samples were performed every 3 min. The line graph shows the mean of three independent experiments with the ribbon indicating ± SD. b Illustration of in vitro transcribed RNA substrates with varying starting nucleotides with or without paired 5′-overhangs. c RNA substrates depicted in (b) were incubated without (dotted line) or with (solid line) NUDT2 at a concentration of 600 nM over a time course of 2 h, and phosphate release was evaluated using the EnzChek assay. Spectrophotometric measurements to quantify phosphate release were performed every 3 min. The line graph shows the mean of three independent experiments with the ribbon indicating ± SD. d The indicated RNA was incubated with NUDT2 (600 nM) for 3 h, and the integrity was analyzed using an Agilent small RNA chip and a Bioanalyzer. e The indicated ribonucleoside triphosphates (2.5 µM) and the…

• 

  Fig. 4. NUDT2 sequentially releases phosphates from PPP-RNA. a A triphosphorylated RNA dinucleotide (pppGpA) was treated with NUDT2 (600 nM) or RppH (5 U) for 2 h at 37 °C. Additionally, the reference pppGpA was incubated in the same reaction buffer (untreated). Products were analyzed by reverse-phase HPLC. The absorbance at 260 nm at the indicated retention time is shown. b As (a) incubation with 600 nM NUDT2 for the indicated time. c As (b) but quantification of the generated RNA after incubation of pppGpA with NUDT2 for the indicated time. Quantification was performed by the integration of the corresponding peak area. d As (b) but incubation with NUDT2 E58A or NUDT14. e As (b), but the triphosphorylated dinucleotide pppApG was used as substrate. f As (e) but quantification of the generated RNA after incubation of pppApA with NUDT2 for the indicated time. Quantification was performed by the integration of the corresponding peak area.

• 

  Fig. 5. NUDT2 is preparing RNA to serve as XRN1 substrate. a In vitro transcribed PPP-renilla RNA was left untreated (mock) or incubated with 600 nM NUDT2, 1 U XRN1, 600 nM NUDT2 together with 1 U XRN1, 5 U RppH, or 5 U RppH together with 1 U XRN1 and 10 U RNase for 4 h. The remaining RNA was quantified by RT-qPCR. The bar plot shows the fold change compared to input RNA as the mean ± SD of three technical replicates. b In vitro synthesized HCV RNA was treated with 5 U RNase, 600 nM NUDT2, 1 U XRN1, or co-treated with 600 nM NUDT2 and 1 U XRN1 and then analyzed on an agarose gel. c In vitro transcribed HCV RNA encoding luciferase was electroporated into HeLa CRISPR KO cells lacking NUDT2 or HeLa cells treated with a non-targeting-control vector (NTC). Viral RNA load was quantified 4 and 6 h post electroporation by the abundance of firefly luciferase RNA. d In vitro transcribed triphosphorylated renilla RNA and capped firefly luciferase RNA were co-electroporated into Hep3B CRISPR/Cas9 KO cells lacking NUDT2, XRN1, or both NUDT2 and XRN1, or Hep3B cells treated with a non-targeting-control vector (NTC). RNA load was quantified 1 and 4 h post electroporation by quantifying the abundance of…

• 

  Fig. 6. Impact of NUDT2 on virus growth. a Bone marrow cells were isolated from Nudt2 knockout (Nudt2−/−), and control mice were tested for the abundance of NUDT2 by mass spectrometry. The mean LFQ intensities ± SD for NUDT2 and GAPDH of six biological replicates are shown. A two-sided students t-test (FDR 0.05) was performed. b Accumulation of infectious virus particles in supernatants of Nudt2+/+ or Nudt2−/− bone marrow infected with either wild-type VSV, VSV-M2 at an MOI of 1 for 16 h. The average TCID50/mL of three biological replicates ± SD are shown. *** p < 0.001 two-way ANOVA with Bonferroni’s post-test. c Accumulation of VSV in supernatants of Nudt2+/+ and Nudt2−/− MEFs infected with VSV at an MOI of 0.001 for 16 h. Average TCID50/mL of three biological replicates ± SD. *** p < 0.001, two-way ANOVA with Bonferroni’s post-test. d Viral RNA load in Nudt2+/+ or Nudt2−/− MEFs infected with VSV. RNA levels were quantified by RT-qPCR analysis using specific primers for the VSV nucleoprotein transcript. Data were normalized to murine ActB RNA. The mean ± SD of three biological repetitions are shown in arbitrary units. e Phase-contrast micrographs of Nudt2+/+ or Nudt2−/− MEFs infected…