The mechanism of RNA capping by SARS-CoV-2

Figures

• 

  Fig. 1. The NiRAN domain NMPylates the N terminus of nsp9. a, Incorporation of α-32P from [α-32P]ATP, GTP, UTP or CTP into nsp8 or nsp9 by WT nsp12, the NiRAN mutant (D218A) or the polymerase mutant (D760A). Reactions were performed in the presence of Mg2+ or Mn2+, and the products were resolved by SDS–PAGE and visualized by Coomassie staining (top) and autoradiography (autorad; bottom). b, Incorporation of α-32P from [α-32P]UTP into nsp9 or the indicated mutants by the NiRAN domain. The reaction products were analysed as described in a. c, Quantification of reaction products shown in b depicting the relative NiRAN-dependent UMPylation activity towards nsp9 or the indicated mutants. Radioactive gel bands were excised and quantified by scintillation counting. d, One-dimensional 31P spectrum of AMP–nsp9 (red) and AMP (blue) recorded in the same buffer as the reference. e, 2D 1H, 31P-HSQC spectra of AMP (top, blue) and AMP–nsp9 (bottom, red). f, The structure of AMP (top) and AMP–nsp9 (bottom) with arrows to indicate the magnetization transfer steps that result in the peaks observed in the 2D NMR spectra. The blue arrows indicate the transfer steps that yield the peaks in the HSQC spectra,…

• 

  Fig. 2. The NiRAN domain RNAylates nsp9. a, Schematic of the nsp9 AMPylation reaction (top) and the proposed nsp9 RNAylation reaction (bottom). b, Time-dependent incorporation of RNA into nsp9 by WT nsp12, the NiRAN mutant (D218A) or the polymerase mutant (D760A). The reaction products were analysed using SDS–PAGE and Coomassie staining. The samples were also treated with RNase A. c, Incorporation of different lengths of 5′-pppRNAs into nsp9 by the NiRAN domain. The reaction products were analysed as described in b. d,e, Incorporation of RNAs (with substitutions in the first and second base) into nsp9 by the NiRAN domain. Reaction products were analysed as described in b (d) and quantified by densitometry (e). For e, data are mean ± s.d. of three replicates. The results in this figure are representative of at least three independent experiments. Source data

• 

  Fig. 3. The NiRAN domain catalyses the transfer of 5′-pRNA from nsp9 to GDP to form the core cap structure GpppA-RNA. a, DeRNAylation of the covalent nsp9–RNALS10 species by WT nsp12, the NiRAN mutant (D218A) or the polymerase mutant (D760A) when incubated with buffer or GDP. The reaction products were analysed as described in Fig. 2b. b, Time-dependent deRNAylation of nsp9–pRNALS10 by WT nsp12, the NiRAN mutant (D218A) or the polymerase mutant (D760A). The reaction products were analysed as described in Fig. 2b. c, DeRNAylation of nsp9–pRNALS10 by nsp12 in the presence of different NTPs, NDPs or PPi. The reaction products were analysed as described in Fig. 2b. d, Incorporation of α-32P from [α-32P]GDP into nsp9–pRNALS10 by WT nsp12, the NiRAN mutant (D218A) or the polymerase mutant (D760A). VCE was used as a control but was incubated with [α-32P]GTP. The reaction products were resolved using urea–PAGE and visualized by toluidine blue O staining (top) and autoradiography (bottom). The results in this figure are representative of at least two independent experiments. nt, nucleotides.

• 

  Fig. 4. nsp14 and nsp16 catalyse the formation of the cap-0 and cap-1 structures. a, TLC analysis of the reaction products from Extended Data Fig. 6a, after extraction from the urea–PAGE gel and treatment with P1 nuclease and CIP. The location of the cold standards (left) was visualized by ultraviolet fluorescence and the reaction products were visualized by 32P autoradiography. b, Pull-down assays depicting the binding of [32P]7MeGpppA-RNALS10 to GST–eIF4E. [32P]7MeGpppA-RNALS10 was produced using SARS-CoV-2 virally encoded proteins or the VCE. Radioactivity in GST pull-down assays was quantified by scintillation counting. Data are mean ± s.d. The results shown in the figure are representative of three independent experiments. nsp14 DA, nsp14(D331A); nsp16 DA, nsp16(D130A). Source data

• 

  Fig. 5. Structural and genetic insights into RNA capping by the kinase domain. a, Front and back views of nsp12, nsp7, nsp8 and nsp9 cryo-EM maps. Green, NiRAN; magenta, RdRp; violet, nsp7; light blue, nsp8; gold, nsp9. b,c, Density maps of the N terminus of nsp9 from this study (b) and from ref. 26 (PDB: 7CYQ) (c). The arrow in c indicates probable density of unmodelled Gly–Ser residues on non-native nsp9. d, Electrostatic surface of the NiRAN active site from PDB 7CYQ bound to nsp9, contoured at ±5 kT e−1. e, Cartoon representation of the nsp9 N terminus interacting with the β9–β8–β10 sheet in NiRAN, and Arg733 in RdRp (PDB: 7CYQ). f, Relative RNAylation activity of the nsp12(R733A) mutant. The reaction products were analysed as described in Fig. 2e. g, Relative GDP-PRNTase activity of the nsp12(R733A) mutant (measured as RNA release from nsp9–pRNALS10). The reaction products were separated by urea–PAGE, stained with SYBR Gold and quantified by densitometry. h, Cartoon representation of the nsp9 C-terminal helix interacting with the NiRAN domain (PDB: 7CYQ). i, Relative RNAylation activity of nsp12 towards indicated nsp9 mutants. ∆C, nsp91–92. The reaction products were analysed as…

• 

  Fig. 6. Proposed model of the SARS-CoV-2 RNA-capping mechanism. a,b, During transcription, the nascent 5′-pppRNA binds to the NiRAN active site in either a cis (a) or trans (b) manner. c, After binding, the N terminus of nsp9 attacks the α-phosphate of the nascent 5′-pppRNA (i), forming the covalent nsp9–pRNA species and releasing PPi (ii). After GDP binding to the NiRAN active site, the β-phosphate of the nsp13-generated GDP attacks the 5′-phosphate on the nsp9–pRNA (iii), releasing capped RNA and regenerating unmodified nsp9 (iv). Subsequent methylation events are carried out by nsp14 and nsp16 to generate the 7MeGpppA2′-O-Me-RNA cap.

• 

  Extended Data Fig. 1. Sequence alignment of the NiRAN domain reveals similarity to the pseudokinase selenoprotein O (SELO). Multiple sequence alignment highlighting conserved kinase-like active site residues in the NiRAN domain among several CoVs, other selected Nidovirales (Arteri-, Mesoni- and Roniviruses) and the human SELO pseudokinase. Top: amino acid sequence surrounding the Lys-Glu ion pair. Bottom: and the amino acid sequence surrounding the active site and the “DFG” motif.

• 

  Extended Data Fig. 2. The canonical eukaryotic mRNA capping mechanism. The 7MeGpppA2′-O-Me cap on eukaryotic RNA is formed co-transcriptionally by four enzymes: 1) an RNA triphosphatase (RTPase), which removes the γ-phosphate from the nascent 5′-triphosphorylated RNA (5′-pppRNA) to yield a 5′-diphosphorylated RNA (5′-ppRNA); 2) a guanylyltransferase (GTase), which transfers the GMP moiety from GTP to the 5′-ppRNA to form the core cap structure GpppN-RNA; 3) a (guanine-N7)-methyltransferase (N7-MTase), which methylates the cap guanine at the N7 position; and 4) a (nucleoside-2′-O)-methyltransferase (2′-O-MTase), which methylates the ribose-2′-OH position on the first nucleotide of the RNA. B denotes any base; GTP, Guanosine triphosphate; GDP, Guanosine diphosphate; PPi, pyrophosphate; SAM, S-Adenosyl methionine.

• 

  Extended Data Fig. 3. The NiRAN domain NMPylates nsp9. a. Incorporation of α-32P from [α-32P]GTP or [α-32P]UTP into nsp7 or nsp9 by WT nsp12, the NiRAN mutant (D218A), or the polymerase mutant (D760A). Reactions were performed in the presence of Mg2+ or Mn2+ and the products were resolved by SDS-PAGE and visualized by Coomassie staining (top) and autoradiography (bottom). b–e. Kinetic analysis depicting the concentration dependence of (b) ATP, (c) CTP, (d) GTP, or (e) UTP on the rate of nsp9 NMPylation by the NiRAN domain. Km (μM), Vmax (pmol NTP/min), kcat (min−1) and kcat/Km (min−1/μM) values are indicated. Plots shown are the mean and SD of triplicate reactions. f–i. Intact mass LC/MS spectra of unmodified nsp9 (cyan) overlayed with NMPylated nsp9 (pink) following incubation with WT nsp12 and (f) ATP, (g) CTP, (h) GTP, or (i) UTP. The observed masses are shown in the insets. The theoretical mass of unmodified nsp9 is 12378.2 Da and the theoretical increase in mass with the addition of each NMP is as follows: AMP, 329 Da; CMP, 305 Da; GMP, 345 Da; UMP, 306 Da. j. Quantification of the stoichiometry of NMPylation of nsp7, nsp8 and nsp9. Reactions were performed in the presence of Mg2+ or Mn2+.…

• 

  Extended Data Fig. 4. Characterization of NiRAN RNAylation activity. a. Intact mass LC/MS spectra (overlayed) of unmodified nsp9 (cyan) or nsp9 after incubation with 5′-pppRNALS10 and WT nsp12 (pink). The theoretical and observed masses are shown in the insets. The Δmass of 3233.17 Da corresponds to monophosphorylated RNALS10 (5′-pRNALS10). b. Incorporation of RNA into nsp9 by nsp12 (0–4 µM) in the presence of Mg2+ or Mn2+. Reaction products were analysed as in Fig. 2b. c. Incorporation of RNA with the indicated 5′ ends into nsp9 by nsp12. Reaction products were analysed as in Fig. 2b. d. Time-dependent incorporation of 5′-pRNALS10 into nsp9 or the nsp9 N1A mutant. Reaction products were analysed as in Fig. 2b. Source data

• 

  Extended Data Fig. 5. Characterization of nsp12 NiRAN GDP-PRNTase activity. a. Schematic representation depicting the mechanism of core cap formation by vesicular stomatitis virus (VSV) polyribonucleotidyltransferase (PRNTase). b. Intact mass LC/MS spectra of nsp9–pRNALS10 after incubation with GDP and WT nsp12 (right) or the NiRAN mutant (left). The theoretical mass of nsp9 is 12378.2 Da and the theoretical mass of nsp9–pRNALS10 is 15611.5 Da. c. Time-dependent deRNAylation of nsp9–pRNALS10 by WT nsp12, the NiRAN mutant (K73A, D221A), or the polymerase mutant (D760A) in the presence of GDP and either Mg2+ or Mn2+. Reaction products were analysed as in Fig. 2b. d. NiRAN-dependent deAMPylation or deRNAylation of nsp9 in the presence of PPi or GDP. Reaction products were analysed as in Fig. 2b. e. Concentration dependence of ATP (blue) or GTP (red) on the rate of phosphate release catalysed by nsp13. Vmax (pmol PO43−/min), Km (μM), kcat (min−1) and kcat/Km (min−1/μM) values are indicated. f. HPLC/MS quantification of GpppA formed during the NiRAN-catalysed deRNAylation of nsp9–pRNALS10. Reaction products were digested with nuclease P1 prior to HPLC analysis. Reactions were performed in triplicate and…

• 

  Extended Data Fig. 6. SARS-CoV-2 nsp14 and nsp16 methyltransferases methylate the nsp12-synthesised core cap structure. a. Incorporation of α-32P from [α-32P]GDP into nsp9–pRNALS10 by WT nsp12, or the NiRAN K73A, D221A mutant (K, DA). Reactions were subsequently incubated with SAM, nsp14 (or the D331A mutant; DA), and nsp10/16 (or the D130A nsp16 mutant; DA). The Vaccinia capping enzyme (VCE) was used as a positive control but incubated with [α-32P]GTP and the Vaccinia 2′-O-MTase. Reaction products were analysed as in Fig. 3d. b. Incorporation of 14C from [methyl-14C]SAM into GpppA-RNALS10 by nsp14 (WT or the D331A mutant; DA), and nsp10/16 (WT or the D130A nsp16 mutant; DA). The VCE and the Vaccinia 2′-O-MTase were used as positive controls. Reaction products were analysed as in Fig. 3d. The 14C signal was detected by phophorimaging.