The Enzymatic Activity of the nsp14 Exoribonuclease Is Critical for Replication of MERS-CoV and SARS-CoV-2

Abstract

Coronaviruses (CoVs) stand out for their large RNA genome and complex RNA-synthesizing machinery comprising 16 nonstructural proteins (nsps). The bifunctional nsp14 contains 3'-to-5' exoribonuclease (ExoN) and guanine-N7-methyltransferase (N7-MTase) domains. While the latter presumably supports mRNA capping, ExoN is thought to mediate proofreading during genome replication. In line with such a role, ExoN knockout mutants of mouse hepatitis virus (MHV) and severe acute respiratory syndrome coronavirus (SARS-CoV) were previously reported to have crippled but viable hypermutation phenotypes. Remarkably, using reverse genetics, a large set of corresponding ExoN knockout mutations has now been found to be lethal for another betacoronavirus, Middle East respiratory syndrome coronavirus (MERS-CoV). For 13 mutants, viral progeny could not be recovered, unless-as happened occasionally-reversion had first occurred. Only a single mutant was viable, likely because its E191D substitution is highly conservative. Remarkably, a SARS-CoV-2 ExoN knockout mutant was found to be unable to replicate, resembling observations previously made for alpha- and gammacoronaviruses, but starkly contrasting with the documented phenotype of ExoN knockout mutants of the closely related SARS-CoV. Subsequently, we established in vitro assays with purified recombinant MERS-CoV nsp14 to monitor its ExoN and N7-MTase activities. All ExoN knockout mutations that proved lethal in reverse genetics were found to severely decrease ExoN activity while not affecting N7-MTase activity. Our study strongly suggests that CoV nsp14 ExoN has an additional function, which apparently is critical for primary viral RNA synthesis and thus differs from the proofreading function that, based on previous MHV and SARS-CoV studies, was proposed to boost longer-term replication fidelity.IMPORTANCE The bifunctional nsp14 subunit of the coronavirus replicase contains 3'-to-5' exoribonuclease (ExoN) and guanine-N7-methyltransferase domains. For the betacoronaviruses MHV and SARS-CoV, ExoN was reported to promote the fidelity of genome replication, presumably by mediating a form of proofreading. For these viruses, ExoN knockout mutants are viable while displaying an increased mutation frequency. Strikingly, we have now established that the equivalent ExoN knockout mutants of two other betacoronaviruses, MERS-CoV and SARS-CoV-2, are nonviable, suggesting an additional and critical ExoN function in their replication. This is remarkable in light of the very limited genetic distance between SARS-CoV and SARS-CoV-2, which is highlighted, for example, by 95% amino acid sequence identity in their nsp14 sequences. For (recombinant) MERS-CoV nsp14, both its enzymatic activities were evaluated using newly developed in vitro assays that can be used to characterize these key replicative enzymes in more detail and explore their potential as target for antiviral drug development.

Keywords: RNA synthesis; guanine-N7-methyltransferase; nidovirus; nonstructural protein; proofreading; replicase; reverse genetics.

FIG 1

Alignment of nsp14 amino acid sequences from selected coronaviruses. Sequences of the ExoN and N7-MTase domains in MERS-CoV (NC-019843), SARS-CoV (NC_004718), SARS-CoV-2 (NC_045512.2), MHV (NP_045298), HCoV-229E (NC_002645), TGEV (AJ271965), and IBV (NP_040829) were used for the analysis. The domains indicated at the top are based on the SARS-CoV-nsp14 secondary structure (PDB 5NFY) (21). Fully conserved residues are shown with white letters on dark gray (above 70% conservation), whereas partially conserved residues are displayed with lighter shades of gray. Catalytic residues and residues involved in formation of zinc fingers are marked with asterisks and circles, respectively. Filled circles indicate zinc fingers targeted by mutagenesis (Fig. 2A), while arrowheads identify the two N7-MTase domain residues mutated to generate the MTase negative control used in biochemical assays. The alignment was generated using Clustal Omega (104) and edited using Jalview version 2.11 (105).

FIG 2

MERS-CoV ExoN knockout mutants are nonviable. (A) Phenotype of MERS-CoV nsp14 ExoN mutants used in this study, scored at 2 dpt. (B) Comparison of plaque phenotype of selected ExoN mutants in HuH7 cells. Plaque assays were performed using supernatants harvested from transfected cell cultures at 3 dpt, which were diluted 10⁻⁴ for wt and mutant E191D and used in undiluted form for the D90A/E92A ExoN knockout double mutant (DM) and the H229C ZF1 mutant. (C) Immunolabeling (2 dpt) of cell cultures consisting of a mixture of (nonsusceptible) BHK-21 cells transfected with in vitro-made full-length MERS-CoV RNA and susceptible (DPP4-expressing) Vero cells used to amplify any infectious progeny released from the transfected BHK-21 cells. (Left) wt virus; (middle) DM mutant; (right) H229C mutant. Cells were labeled for dsRNA (green) and nsp4 (red). Bar, 100 μm.

FIG 3

A SARS-CoV-2 ExoN knockout mutant is nonviable. (A) Plaque phenotypes of wt SARS-CoV (left) and SARS-CoV-2 (right) and their corresponding ExoN motif I knockout double mutants (DM; D90A/E92A) in Vero E6 cells. Plaque assays were performed using supernatants harvested from transfected cell cultures at 2 dpt for SARS-CoV and 3 dpt for SARS-CoV-2. Samples were diluted 10⁻⁶ for SARS-CoV wt, 10⁻⁵ for SARS-CoV-2 wt and SARS-CoV DM, and 10⁻¹ for the SARS-CoV-2 DM mutant. (B) Immunolabeling (2 dpt) as described for Fig. 2C but with Vero E6 cells for amplification of SARS-CoV-2 progeny released from BHK-21 cells transfected with wt (top) or DM (bottom) full-length RNA. Bar, 100 μm.

FIG 4

Impact of ExoN inactivation on intracellular MERS-CoV RNA synthesis. In-gel hybridization analysis of intracellular RNA isolated after 2 or 3 dpt of transfected BHK-21 cells, which were subsequently mixed with HuH7 or Vero cells, as indicated. Purified RNA was separated in an agarose gel and probed with a radiolabeled oligonucleotide probe recognizing the MERS-CoV genome and subgenomic mRNAs.

FIG 5

Characterization of growth kinetics of rMERS-nsp14-E191D and its sensitivity to 5-FU treatment. Vero cells (A) or HuH7 cells (B) were infected at an MOI of 3, supernatant was harvested at the indicated time points, and viral progeny titers were measured by plaque assay from two independent experiments using duplicates (n = 4; values are means ± standard deviations [SD]). (C) Plaque phenotype in HuH7 cells of rMERS-CoV nsp14-E191D and wt control in the absence or presence of the mutagenic agent 5-FU. (D) Dose-response curve of wt and E191D mutant MERS-CoV in the presence of 5-FU concentrations up to 400 μM (MOI, 0.1; n = 4; means ± SD). Statistical significance of the difference with wt virus at each time point (A and B) or concentration (D) was assessed by paired t test. *, P < 0.05; **, P < 0.005.

FIG 6

Expression and purification of recombinant MERS-CoV nsp14. N-terminally His-tagged wt and mutant MERS-CoV nsp14 (∼55 kDa) was expressed in E. coli, affinity purified, and analyzed in a 10% SDS-PAGE gel that was stained with Coomassie blue. The molecular masses of the protein marker (Invitrogen) are given, in kilodaltons.

FIG 7

Optimization of MERS-CoV nsp14 in vitro ExoN assay conditions. The substrate for the assay was a 22-nt synthetic RNA (H4) that was ³²P labeled at its 5′ terminus. Cleavage products were separated by polyacrylamide gel electrophoresis and visualized by autoradiography. (A) Analysis of ExoN activity in the presence of increasing amounts of nsp10, using wt MERS-CoV-nsp14 (left) and the ExoN double-knockout mutant (DM; D90A/E92A; right). The RNA substrate was hydrolyzed for 90 min at 37°C using a fixed concentration of nsp14 (200 nM) and increasing nsp10 concentrations, ranging from 0 to 1,600 nM. (B) Evaluation of the ExoN activity of increasing concentrations (200 to 2,000 nM) of wt or DM nsp14 in the presence of a fixed amount of nsp10 (200 nM).

FIG 8

Time course analysis of the in vitro ExoN activity of MERS-CoV nsp14. The ExoN activity of different recombinant nsp14 proteins (wt, D90A/E92A, E191D, and H229C) was evaluated by incubating 200 nM nsp14 and 800 nM nsp10 for 0, 5, 30, 60, and 90 min at 37°C. As controls, individual proteins (800 nM) were incubated for 90 min. For technical details, see the legend to Fig. 7.

FIG 9

Cross-activation of the in vitro activity of SARS-CoV and MERS-CoV nsp14 by heterologous nsp10. The nsp10 cofactor was exchanged in ExoN assays performed with MERS-CoV and SARS-CoV nsp14, using a 1:4 ratio between nsp14 and nsp10 and a 90-min incubation at 37°C. For technical details, see the legend to Fig. 7.

FIG 10

In vitro ExoN activity of MERS-CoV nsp14 mutants. Residues from the DEDDh catalytic motif and ZF1 motif of the nsp14 ExoN domain and the nsp14 N7-MTase domain were mutated as indicated. Assays were performed using a 1:4 ratio between nsp14 and nsp10 and a 90-min incubation at 37°C. For technical details, see the legend to Fig. 7.

FIG 11

In vitro N7-MTase activity of MERS-CoV nsp14 mutants. The N7-MTase activity of recombinant nsp14 was analyzed in vitro by filter binding assay using synthetic cap analogues as the substrate. (A) Increasing concentrations of MERS-CoV nsp14 were incubated with GpppA and m⁷GpppA in the presence of [³H]SAM for 30 min. (B) The ability of nsp14 to methylate GpppA or m⁷GpppA was determined after reaction times between 0 and 120 min at 30°C. (C) The ability of nsp14 mutants to methylate GpppA was measured four times in duplicate. Values were normalized to the wt control (n = 8; means ± SD).