key: cord-0818076-pmm2qy53
authors: Park, Gina J.; Osinski, Adam; Hernandez, Genaro; Eitson, Jennifer L.; Majumdar, Abir; Tonelli, Marco; Henzler-Wildman, Katie; Pawłowski, Krzysztof; Chen, Zhe; Li, Yang; Schoggins, John W.; Tagliabracci, Vincent S.
title: The mechanism of RNA capping by SARS-CoV-2
date: 2022-02-09
journal: bioRxiv
DOI: 10.1101/2022.02.07.479471
sha: 40163f4997662fba62f79249e3d294fe436e89d4
doc_id: 818076
cord_uid: pmm2qy53

The SARS-CoV-2 RNA genome contains a 5’-cap that facilitates translation of viral proteins, protection from exonucleases and evasion of the host immune response1-4. How this cap is made is not completely understood. Here, we reconstitute the SARS-CoV-2 7MeGpppA2’-O-Me-RNA cap using virally encoded non-structural proteins (nsps). We show that the kinase-like NiRAN domain5 of nsp12 transfers RNA to the amino terminus of nsp9, forming a covalent RNA-protein intermediate (a process termed RNAylation). Subsequently, the NiRAN domain transfers RNA to GDP, forming the cap core structure GpppA-RNA. The nsp146 and nsp167 methyltransferases then add methyl groups to form functional cap structures. Structural analyses of the replication-transcription complex bound to nsp9 identified key interactions that mediate the capping reaction. Furthermore, we demonstrate in a reverse genetics system8 that the N-terminus of nsp9 and the kinase-like active site residues in the NiRAN domain are required for successful SARS-CoV-2 replication. Collectively, our results reveal an unconventional mechanism by which SARS-CoV-2 caps its RNA genome, thus exposing a new target in the development of antivirals to treat COVID-19.

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. In CoVs, the nsp13, nsp14, and nsp16 proteins have RTPase 20 , N7-MTase 6 , and 2′-O-MTase 7 activities, respectively.

Thus, it was presupposed that the CoV capping mechanism occurs in a similar fashion to the eukaryotic capping pathway, with the NiRAN domain functioning as the GTase 3,5,21 . However, evidence to support this claim has been lacking.

In this study, we discover that the NiRAN domain transfers monophosphorylated RNA (5′-pRNA) from 5′-pppRNA to the N-terminus of nsp9 as an intermediate step in cap synthesis. The NiRAN domain then transfers 5′-pRNA from RNAylated nsp9 to GDP to form the cap core structure GpppA-RNA. We then reconstitute cap-0 and cap-1 structures using the nsp14 and nsp16 methyltransferases. Furthermore, we present a cryo-EM structure of the SARS-CoV-2 RTC with the native N-terminus of nsp9 bound in the NiRAN active site. Finally, we demonstrate in a reverse genetics system that the N-terminus of nsp9 and the kinase-like active site residues in the NiRAN domain are required for SARS-CoV-2 replication.

The NiRAN domain has been shown to transfer nucleotide monophosphates (NMPs) from nucleotide triphosphates (NTPs) (referred to as NMPylation) to protein substrates, including nsp9 16 and the nsp12 co-factors, nsp7 22 and nsp8 23 . We observed NiRAN-dependent NMPylation of native nsp9, but not native nsp7 or nsp8 (Fig. 1b, Extended Data Fig. 3) . Quantification of 32 P incorporation and intact mass analyses suggests stoichiometric incorporation of NMPs into nsp9 (Extended Data Fig. 3g-j) . Mutation of nsp9 Asn1 to Ala or Asp reduced NMPylation of nsp9 ( Fig. 1c, Extended Data Fig. 4a) , consistent with previous work that suggested NMPylation occurs on the backbone nitrogen of nsp9 Asn1 16 . To provide direct evidence that the amino terminus of nsp9 is NMPylated by the NiRAN domain, we performed nuclear magnetic resonance (NMR) spectroscopy of AMPylated nsp9. The 2D 1 H, 31 P HSQC and 2D HSQC-TOCSY spectra confirm that the AMP is attached to the nitrogen backbone atom of Asn1 via a phosphoramidate linkage ( Fig. 1d-f , Extended Data 4b, c).

Given the ability of the NiRAN domain to transfer NMPs to nsp9 using NTPs as substrates, we wondered whether the NiRAN domain could also utilize 5′-pppRNA in a similar fashion (Fig. 2a) .

We synthesized a 5′-pppRNA 10-mer corresponding to the first 10 bases in the leader sequence (LS10) of the SARS-CoV-2 genome (hereafter referred to as 5′-pppRNA LS10 ) (Extended Data   Table 1 ). We incubated 5′-pppRNA LS10 with nsp9 and nsp12 and analysed the reaction products by SDS-PAGE. Remarkably, we observed an electrophoretic mobility shift in nsp9 that was timedependent, sensitive to RNAse A treatment and required an active NiRAN domain, but not an active RdRp domain (Fig. 2b) . Intact mass analyses of the reaction products confirmed the incorporation of monophosphorylated RNA LS10 (5′-pRNA LS10 ) into nsp9 (Fig. 2c) . The reaction was dependent on Mn 2+ (Extended Data Fig. 5a ) and required a triphosphate at the 5′-end of the RNA (Extended Data Fig. 5b ). Substituting Ala for Asn1 reduced the incorporation of RNA LS10 into nsp9 (Fig. 2d) . We also observed NiRAN-dependent RNAylation of nsp9 using LS RNAs ranging from 2 to 20 nucleotides (Fig. 2e) . Mutation of the first A to any other nucleotide markedly reduced RNAylation (Fig. 2f) . Thus, the NiRAN domain RNAylates the N-terminus of nsp9 in a substrate-selective manner.

The NiRAN domain transfers 5′-pRNA from nsp9 to GDP forming the cap core structure

Negative-sense RNA viruses of the order Mononegavirales, including vesicular stomatitis virus (VSV), have an unconventional capping mechanism in which a polyribonucleotidyltransferase (PRNTase) transfers 5′-pRNA from 5′-pppRNA to GDP via a covalent enzyme-RNA intermediate (Extended Data Fig. 6a ) 24, 25 . Because the NiRAN domain transfers 5′-pRNA to nsp9, we hypothesized that this protein-RNA species may be an intermediate in a similar reaction mechanism to that of the VSV system. To test this hypothesis, we purified the nsp9-pRNA LS10 species by ion exchange and gel filtration chromatography and incubated it with GDP in the presence of nsp12. Treatment with GDP deRNAylated nsp9 in a NiRAN-dependent manner, as judged by the nsp9 electrophoretic mobility on SDS-PAGE (Fig. 3a) and its molecular weight based on intact mass analysis (Fig. 3b) . The reaction was time-dependent, (Fig 3c) , preferred Mg 2+ over Mn 2+ (Extended Data Fig. 6b ) and was specific for GDP--and to some extent GTP--but not the other nucleotides tested (Fig. 3d) . Interestingly, although inorganic pyrophosphate (PPi) was able to deAMPylate nsp9-AMP, it was unable to deRNAylate nsp9-pRNA LS10 (Fig. 3e) . (See

We used Urea-PAGE to analyse the fate of the RNA LS10 during the deRNAylation reaction.

Treatment of nsp9-pRNA LS10 with nsp12 and [a-32 P]GDP resulted in a [ 32 P]-labelled RNA species that migrated similarly to GpppA-RNA LS10 produced by the Vaccinia capping enzyme (Fig. 3f) .

The reaction was dependent on a functional NiRAN domain but not an active RdRp domain. To confirm the presence of a GpppA-RNA cap, we digested the RNA produced from the nsp12 reaction with P1 nuclease and detected GpppA by high performance liquid chromatography/mass spectrometry (HPLC/MS) analysis (Fig. 3g) . Thus, the NiRAN domain is a GDP polyribonucleotidyltransferase (GDP-PRNTase) that mediates the transfer of 5′-pRNA from nsp9 to GDP.

In our attempts to generate GpppA-RNA LS10 in a "one pot" reaction, we found that GDP inhibited the RNAylation reaction (Extended Data Fig. 6c) . However, the formation of GpppA-RNA LS10 could be generated in one pot provided that the RNAylation occurs prior to the addition of GDP (Extended Data Fig. 6c, d) .

The SARS-CoV-2 genome encodes an N7-MTase domain within nsp14 6 and a 2′-O-MTase in nsp16, the latter of which requires nsp10 for activity 7 . Nsp14 and the nsp10/16 complex use Sadenosyl methionine (SAM) as the methyl donor. To test whether NiRAN-synthesized GpppA-RNA LS10 can be methylated, we incubated 32 P-labelled GpppA-RNA LS10 with nsp14 and/or the nsp10/16 complex in the presence of SAM and separated the reaction products by Urea-PAGE ( Fig. 4a) . We extracted RNA from the reaction, treated it with P1 nuclease and CIP, and then analysed the products by thin layer chromatography (TLC) (Fig. 4b) . As expected, the NiRANsynthesized cap migrated similarly to the GpppA standard and the products from the Vaccinia capping enzyme reaction (compare lanes 1 and 4). Likewise, reactions that included SAM and nsp14 migrated similarly to the 7Me GpppA standard and to the products from the Vaccinia capping enzyme reaction following the addition of SAM (compare lanes 2 and 6). Furthermore, treatment of 7Me GpppA-RNA LS10 , but not unmethylated GpppA-RNA LS10 , with nsp10/16 produced the 7Me GpppA2′-O-Me-RNA cap-1 structure (compare lanes 3, 8 and 9). In parallel experiments, we incubated NiRAN-synthesized GpppA-RNA LS10 with nsp14 and/or the nsp10/16 complex in the presence of [ 14 C]-labelled SAM ( 14 C on the donor methyl group) and separated the reaction products by Urea-PAGE. As expected, nsp14 and the nsp10/16 complex incorporated 14 C into GpppA-RNA LS10 to form the cap-0 and cap-1 structures, respectively (Fig 4c) . Thus, the SARS-CoV-2 7Me GpppA2′-O-Me-RNA capping mechanism can be reconstituted in vitro using virally encoded proteins.

Efficient translation of mRNAs is dependent on eIF4E binding to the 7Me GpppA-RNA cap 26 . To test whether the SARS-CoV-2 RNA cap is functional, we incubated [ 32 P]-labelled 7Me GpppA-RNA LS10 with GST-tagged eIF4E. We observed [ 32 P]-labelled RNA in GST pulldowns of [ 32 P] 7Me GpppA-RNA but not the unmethylated derivative (Fig. 4d) . Thus, the 7Me GpppA-RNA cap generated by SARS-CoV-2 encoded proteins is a substrate for eIF4E in vitro, suggesting that the cap is functional.

We determined a cryo-EM structure of the nsp7/8/9/12 complex and observed a nsp9 monomer bound in the NiRAN active site (Fig. 5a, Extended Data Fig. 7-9 , Extended Data Table 2 ). The native N-terminus of nsp9 occupies a similar position to previously reported structures using a non-native N-terminus of nsp9 (Fig. 5b, c) 21 . Our cryo-EM analysis was hindered by the preferred orientation of the complex and sample heterogeneity, yielding final maps with high levels of anisotropy, with distal portions of nsp9 missing, and weak density for the N-lobe of the NiRAN domain (Extended Data Fig. 7, 8) . Therefore, we used our model and the complex structure by Yan et al. 21 (PDBID: 7CYQ) to study the structural basis of NiRAN-mediated RNA capping.

The first four residues of nsp9 extend into the NiRAN active site, forming electrostatic and hydrophobic contacts in and around a groove near the kinase-like active site (Fig. 5d) . Asn1 of nsp9 is positioned inside of the active site, primed for transfer of 5′-pppRNA onto its N-terminus.

Although the terminal NH2 group of nsp9 is the substrate for RNAylation, the local quality of the structures is not high enough to distinguish its exact position. We have modelled the nsp9 acceptor NH2 pointing towards what appears to be the phosphates of the nucleotide analogue UMP-NPP in the active site (Fig 5b) . In the structure by Yan et al. 21 , Asn1 was assigned an opposite conformation and there are unmodeled residues (non-native N-terminus; NH2-Gly-Ser-) visible in the density maps, distorting local structural features (Fig. 5c, arrow) 

Asn2 of nsp9 is in a negatively charged cleft around the NiRAN active site, and contacts Arg733, which extends from the polymerase domain and is partially responsible for positioning nsp9 (Fig.   5e ). Both Leu4 and the C-terminal helix of nsp9 form hydrophobic interactions with a β-sheet (β8-β9-β10) in the N-lobe of the NiRAN domain (Fig. 5e, f) . The N-terminal cap of the nsp9 C-terminal helix also forms electrostatic interactions with a negatively charged pocket on the surface of the NiRAN domain (Fig. 5f) . Nsp12 lacking the RdRp domain (DRdRp; 1-326) neither RNAylates nsp9 nor processes nsp9-pRNA LS10 to form GpppA-RNA (Fig. 5g) . Likewise, deleting the Cterminal helix on nsp9 (ΔC; 1-92) and Ala substitutions of Asn1 and Asn2 abolished RNAylation (Fig. 5h) .

The NiRAN domain resembles SelO, with an RMSD of 5.7 Å over 224 Cα atoms (PDB ID: 6EAC 13 , Extended Data Fig. 10a ). Lys73 (PKA nomenclature; K72) forms a salt bridge with Glu83 (PKA; E91) from the aC (a2) helix and contacts the phosphates (GDP in 7CYQ, or UMP-NPP in our structure; Fig. 5i ). As expected, the "DFG" Asp218 (PKA; D184) binds a divalent cation. Interestingly, the NiRAN domain lacks the catalytic Asp (Extended Data Fig. 1) , (PKA; D166); however, like in SelO, Asp208 is next to the metal binding Asn209 (PKA; N171) and may act as a catalytic base to activate the NH2 group on the N-terminus of nsp9 (Fig. 5i) .

In canonical kinases, the b1-b2 G-loop stabilizes the phosphates of ATP 28 . In contrast, the NiRAN domain contains a b-hairpin insert (b2-b3) where the b1-b2 G-loop should be (Extended Data   Fig. 10b) . This insertion not only makes contacts with the N-terminus of nsp9, but also contains a conserved Lys (K50) that extends into the active site and stabilizes the phosphates of the bound nucleotide. Likewise, Arg116 also contacts the phosphates of the nucleotide. SelO contains a similar set of basic residues pointing into the active site that accommodate the flipped orientation of the nucleotide to facilitate AMPylation (Extended Data Fig. 10b) . Notably, Lys73, Arg116 and Asp218 in SARS-CoV-1 nsp12 are required for viral replication 5 .

To determine the importance of the NiRAN domain and the N-terminus of nsp9 in viral replication, we used a DNA-based reverse genetics system that can rescue infectious SARS-CoV-2 (Wuhan-Hu-1/2019 isolate) expressing a fluorescent reporter 8 (Extended Data Fig. 11a) . We introduced single point mutations in nsp9 (N1A, N1D and N2A) and nsp12 (K73A, D218A and D760A) and quantified the virus in supernatants of producer cells by RT-qPCR to detect the viral N gene. We observed a 400 to 4000-fold reduction in viral load for all the mutants compared to WT (Fig. 5j,   Extended Data Fig. 11a) . To account for the possibility of a proteolytic defect in the mutant viral polyprotein, we tested whether the main viral protease nsp5 (M Pro ) can cleave a nsp8-nsp9 fusion protein containing the Asn1/Asn2 mutations in nsp9. The N1D mutant failed to be cleaved by nsp5, suggesting that the replication defect observed for this mutant is a result of inefficient processing of the viral polyprotein. However, the N1A and N2A mutants were efficiently cleaved by nsp5 (Extended Data Fig. 11b, c) . Collectively, these data provide genetic evidence that the residues involved in capping of the SARS-CoV-2 genome are essential for viral replication.

We propose the following mechanism of RNA capping by CoV: during transcription, the nascent 5′-pppRNA binds to the NiRAN active site, in either a cis (Fig. 6a) or trans (Fig. 6b ) manner and 5′-pRNA is subsequently transferred to the N-terminus of nsp9 forming a phosphoramidate bond (Fig. 6c, panels 1 and 2) . The nsp13 protein produces GDP from GTP, which binds the NiRAN active site and attacks RNAylated nsp9, releasing capped RNA and regenerating unmodified nsp9 ( Fig. 6c, panels 3 and 4) . Subsequently, nsp14 and nsp16 perform sequential N7 and 2′-O methylations, forming a fully functional 7Me GpppA2′-O-Me-RNA cap.

SARS-CoV-2 nsp12 is thought to initiate transcription/replication starting with an NTP, or a short 5′-pppRNA primer 29 . Cryo-EM structures of the RTC suggest that the dsRNA product makes its way out of the RdRp active site in a straight line, supported by the nsp8 helical stalks 21, 30, 31 . In a cis capping model, the helical duplex with nascent 5′-pppRNA would then need to unwind, flex 90°, and extend into the NiRAN active site ~70 Å away (Fig. 6a) . More likely, a separate RTC complex could perform capping in trans (Fig. 6b) . Notably, Perry et al. 32 propose that the nascent RNA strand is separated from the template upon passing through the proof-reading ExoN domain of nsp14 on a neighbouring RTC and threaded towards the NiRAN domain.

SARS-CoV-1 nsp13 has RNA helicase, nucleotide triphosphatase (NTPase), and RNA 5′triphosphatase (RTPase) activities 20 . The RTPase activity implicated nsp13 in the first step of the capping mechanism; however, while nsp13 can act on 5′-pppRNA, this reaction is inhibited in the presence of cellular concentrations of ATP 20 . Thus, we favour the idea that the physiological functions for nsp13 are: 1) to utilize the energy from ATP hydrolysis to unwind double-stranded RNA (helicase), and 2) to hydrolyse GTP to GDP, which can then act as an acceptor for 5′-pRNA in the NiRAN-catalysed capping reaction.

The SARS-CoV-2 capping mechanism is reminiscent of the capping mechanism used by VSV, although there are some differences. The VSV large (L) protein is a multifunctional enzyme that carries out RdRp, PRNTase, and methyltransferase activities to form the cap 24,33,34 . During the reaction, 5′-pRNA is transferred to a conserved His within the PRNTase domain, which adopts a unique α-helical fold that is distinct from that of protein kinases 25 . The presence of two different enzymatic mechanisms of capping, proceeding via covalent protein-RNA intermediates, in Mononegavirales and in Nidovirales is an example of convergent evolution.

Consistent with other reports 16,27 , we observed NiRAN-catalysed NMPylation of nsp9 (Fig. 1b,   Extended Data Fig. 3) . While our results do not necessarily preclude a biologically relevant function for nsp9 NMPylation, it is worth noting that this modification is reversible in the presence of PPi 27 (Fig. 3e) . PPi is produced during the RdRp reaction, making the stability of NMPylated nsp9 difficult to envision in vivo. By contrast, RNAylated nsp9 was not reversible in the presence of PPi. Thus, RNAylation is likely the physiologically relevant modification of nsp9 during viral RNA capping.

Recent work suggested that the NiRAN domain is a GTase that transfers GMP from GTP to 5′-ppRNA, forming a GpppA-RNA cap intermediate 3,21 . In our efforts to reproduce these results, we failed to detect nsp12-dependent GpppA cap formation by TLC (Extended Data Fig. 12a ) or by Urea-PAGE analysis of the RNA (Extended Data Fig. 12b) , in contrast to our control, in which the Vaccinia capping enzyme efficiently generated GpppA-RNA. Because nsp13 and the Vaccinia capping enzyme can hydrolyse GTP to GDP 20 , the cap reported previously 21 appears to be GDP formed from nsp13-and Vaccinia capping enzyme-dependent hydrolysis of GTP.

In summary, we have defined the mechanism by which SARS-CoV-2 caps its genome and have reconstituted this reaction in vitro using non-structural proteins encoded by SARS-CoV-2. Our results uncover new targets for the development of antivirals to treat COVID-19 and highlight the catalytic adaptability of the kinase domain. 

Thin layer chromatograms depicting the reaction products from Fig. 4a following extraction from the Urea PAGE gel and treatment with PI nuclease and CIP. Location of the cold standards (left) was visualized by UV fluorescence and the 32 P by autoradiography. c. Incorporation of 14 C from

DA), and nsp10/16 (or the D130A nsp16 mutant; DA). The VCE and the

MTase were used as positive controls. Reaction products were analysed as in Fig 3f. The 14 C signal was detected by phophorimaging. d. Pull-down assays depicting the binding of

GpppA-RNA LS10 was produced using SARS-CoV-2 virally encoded proteins or the VCE. Radioactivity in GST pull-downs was quantified by scintillation counting. Results represent three independent experiments. Error bars represent the standard deviation (SD)

Structural and genetic insights into RNA capping by the kinase domain. a. Front and back views of nsp12/7/8/9 cryo-EM maps, with respect to the NiRAN domain. The NiRAN domain is in green, the RdRp in magenta

The NiRAN domain is shown in green and nsp9 is in gold. The arrow in (c) indicates additional density that likely corresponds to unmodeled Gly-Ser residues at the non-native Nterminus of nsp9. d. Electrostatic surface view of the NiRAN active site from 7CYQ bound to nsp9 (gold). The N-terminus and the C-terminal helix of nsp9 are shown

PDB ID 7CYQ was used. f. Cartoon representation depicting the interactions between the nsp9 C-terminal helix (gold) and the β8-β9

Interactions between Asn95/96 in nsp9 and D291 in the NiRAN domain are indicated. PDB ID 7CYQ was used. g. Incorporation of 5′-pRNA LS10 into nsp9 and deRNAylation of nsp9-pRNA LS10

or the isolated NiRAN domain (residues 1-326; ∆RdRP). Reaction products were analysed as in Fig. 2b. h. Incorporation of 5′-pRNA LS10 into nsp9 (or the indicated mutants) by nsp12. Reaction products were analysed as in Fig. 2b. i. Cartoon representation of the NiRAN active site. Catalytic residues and GDP are shown as sticks

Relative viral yields from WT or mutant SARS-CoV-2 viruses bearing indicated mutations in nsp9 and nsp12. Data represent averages of two biological replicates. Error bars, SD. Results shown in g and h are representative of at least 2 independent experiments

During transcription, the nascent 5′-pppRNA binds to the NiRAN active site in either a cis (a) or a trans (b) manner. c. Upon binding, the N-terminus of nsp9 attacks the α-phosphate of the nascent 5′-pppRNA (1), forming the covalent nsp9-pRNA species and releasing PPi (2)