In vitro reconstitution of SARS-coronavirus mRNA cap methylation

Abstract

SARS-coronavirus (SARS-CoV) genome expression depends on the synthesis of a set of mRNAs, which presumably are capped at their 5' end and direct the synthesis of all viral proteins in the infected cell. Sixteen viral non-structural proteins (nsp1 to nsp16) constitute an unusually large replicase complex, which includes two methyltransferases putatively involved in viral mRNA cap formation. The S-adenosyl-L-methionine (AdoMet)-dependent (guanine-N7)-methyltransferase (N7-MTase) activity was recently attributed to nsp14, whereas nsp16 has been predicted to be the AdoMet-dependent (nucleoside-2'O)-methyltransferase. Here, we have reconstituted complete SARS-CoV mRNA cap methylation in vitro. We show that mRNA cap methylation requires a third viral protein, nsp10, which acts as an essential trigger to complete RNA cap-1 formation. The obligate sequence of methylation events is initiated by nsp14, which first methylates capped RNA transcripts to generate cap-0 (7Me)GpppA-RNAs. The latter are then selectively 2'O-methylated by the 2'O-MTase nsp16 in complex with its activator nsp10 to give rise to cap-1 (7Me)GpppA(2'OMe)-RNAs. Furthermore, sensitive in vitro inhibition assays of both activities show that aurintricarboxylic acid, active in SARS-CoV infected cells, targets both MTases with IC(50) values in the micromolar range, providing a validated basis for anti-coronavirus drug design.

Figure 1. Genomic organization of CoV pp1a/pp1ab and location of the nsp14 and nsp16 mutants.

The SARS-CoV genomic RNA is translated in two large polyproteins, pp1a and pp1ab following a -1 ribosomal frame shift. The two polyproteins are then cleaved by viral proteases in order to produce 16 nsps (nsp1 to nsp11 from pp1a and nsp1 to nsp16 from pp1ab). Positions of the point mutants used in this study are indicated. White triangles are used for positions targeting exonuclease motifs of nsp14 and black triangles are used for positions targeting MTase motifs of nsp14 and nsp16 (the putative AdoMet binding site of nsp14 and catalytic tetrad of nsp16).

Figure 2. SARS-CoV proteins nsp14, nsp16 and nsp10 purification, AdoMet-dependent MTase activity on short capped RNA and complex formation of nsp16/nsp10.

Panel A: The SARS-CoV proteins nsp14, nsp16 and nsp10, purified by affinity and size exclusion chromatography (as described in Materials and Methods) were separated by SDS-PAGE (14%) and visualized by Coomassie blue staining. Lane 1 corresponds to the molecular size markers, lanes 2 to 4 to nsp14, nsp16, and nsp10, respectively. Panel B and C: AdoMet-dependent MTase assays performed on short capped RNA substrates. The different purified proteins (nsp10: 1200 nM, nsp14: 50 nM and nsp16: 200 nM) were incubated with GpppAC₅ and ^7MeGpppAC₅ RNA oligonucleotides in presence of [³H]-AdoMet as described in Materials and Methods. The methyl transfer to the capped RNA substrate was determined after 5-, 30-, and 240-min incubation by using a filter-binding assay (see Materials and Methods). Panel D: SARS-CoV His₆-nsp16 protein co-expressed with strep-tag-nsp10 and His₆-nsp16 expressed alone were incubated with Strep-Tactin sepharose. Strep-Tactin-bound protein was eluted with D-desthiobiotin and analysed by SDS-PAGE and Coomassie blue staining. Lane 1 corresponds to the molecular size markers, lane 2 to strep-tag-nsp10 co-expressed with His₆-nsp16 and lane 3 to His₆-nsp16 alone.

Figure 3. AdoMet-dependent MTase assays of nsp14 and nsp16/nsp10 on long, virus-specific, capped RNA substrates.

Panel A: Capped RNAs corresponding to the first 264 nucleotides of the SARS-CoV genome were incubated with SARS-CoV proteins (nsp10: 1.2 µM, nsp14: 50 nM and nsp16: 200 nM). Labeled substrates G*pppAG-264 or ^7MeG*pppAG-264 RNA (the asterisk indicates the labeled phosphate) were incubated alone or in presence of the indicated proteins, digested by nuclease P1 and analyzed by TLC. The origins and the positions of standards GpppA, ^7MeGpppA, ^7MeGpppA_{2' OMe} and GpppA_{2' OMe} (see Materials and Methods) are indicated by black arrows. VV:N7-MTase stands for vaccinia virus N7-MTase and DV:NS5MTase for dengue virus MTase domain of protein NS5, a bi-functional N7- and 2′O-MTase, which were used as positive controls. Panel B: Time course analysis of the N7- and 2′O-methylation by nsp14 and nsp16/nsp10. Labeled G*pppAG-SARS-264 RNA was incubated with a mixture of nsp10 (1.2 µM), nsp14 (50 nM), and nsp16 (200 nM). Methylation of the cap structure was followed during 60 min. The final point (overnight  =  ovn) corresponds to 20 h. As in panel A, TLC analysis of nuclease P1-resistant cap structures is shown. The positions of the origin of migration and of GpppA, ^7MeGpppA, ^7MeGpppA_{2' OMe} and GpppA_{2' OMe} cap analog standards are indicated.

Figure 4. Alanine mutagenesis of nsp14 and nsp16 proteins.

Panel A: Residues of the nsp14 exoribonuclease and MTase catalytic sites were mutated to alanine as indicated in Materials and Methods. Equal amounts (50 nM) of each nsp14 mutant were incubated with GpppAC₅ in the presence of [³H]-AdoMet. Methyl transfer to the RNA substrate was measured after 30 min by using a filter-binding assay (upper panel). The N7-MTase activity of the wt control protein was arbitrarily set to 100%. The bar graph presents the results of 3 independent experiments. The purified His₆-tagged proteins analyzed by SDS-PAGE (14%) are shown in the lower panel. Panel B: Each residue of the putative catalytic tetrad K₄₆-D₁₃₀-K₁₇₀-E₂₀₃ of nsp16 was mutated to alanine. Equal amounts of the different nsp16 mutants (200 nM) were incubated with ^7MeGpppAC₅ in the presence of [³H]-AdoMet and nsp10 (1.2 µM). The methyl transfer to the RNA substrate was measured after 30 min by using a filter-binding assay. The 2′O-MTase activity of the wt protein in the presence of nsp10 was arbitrarily set to 100%. The bar graph represents the mean of 3 independent experiments. The purified His₆-tagged proteins analyzed by SDS-PAGE (14%) are shown in the lower panel.

Figure 5. Inhibition of the nsp14 and nsp16/nsp10 MTase activities.

Nsp14 (50 nM) and nsp16/nsp10 (200 nM/1.2 µM) were incubated with GpppAC₅ (in grey) and ^7MeGpppAC₅ (in black), respectively, in order to measure the methyl transfer to the RNA substrates by filter-binding assay (see Materials and Methods). Panel A: Methyl transfer was measured at a final concentration of 100 µM of each inhibitor candidate. The outcome of the control reaction in absence of inhibitor and at 5% of DMSO was set to 100%. The mean value of three independent experiments is given. 1: control, 2: AdoHcy, 3: sinefungin, 4: SIBA (5′-S-isobutylthio-5′-deoxyadenosine), 5: 3-deaza-adenosine, 6: MTA (5′-deoxy-5′-methylthio-adenosine), 7: 2′,3′,5′-tri-O-acetyl-adenosine, 8: S-5′-adenosyl-L-cysteine, 9: GTP, 10: ^7MeGTP, 11: ribavirin, 12: ribavirin-triphosphate, 13: EICAR-triphosphate, 14: GpppA, 15: ^7MeGpppA, 16: ATA, 17: adamantane-analog (N-({[3-(4-methylphenyl)-1-adamantyl]amino}carbonyl)phenylalanine). Panels B, C and D: Dose-response curves and IC₅₀ values of inhibitors AdoHcy, sinefungin and ATA, respectively. The results of three independent experiments are given. Standard deviations are shown for concentrations that were tested three times. IC₅₀ values were calculated as described in Materials and Methods.