SS148 and WZ16 inhibit the activities of nsp10-nsp16 complexes from all seven human pathogenic coronaviruses

Abstract

Seven coronaviruses have infected humans (HCoVs) to-date. SARS-CoV-2 caused the current COVID-19 pandemic with the well-known high mortality and severe socioeconomic consequences. MERS-CoV and SARS-CoV caused epidemic of MERS and SARS, respectively, with severe respiratory symptoms and significant fatality. However, HCoV-229E, HCoV-NL63, HCoV-HKU1, and HCoV-OC43 cause respiratory illnesses with less severe symptoms in most cases. All coronaviruses use RNA capping to evade the immune systems of humans. Two viral methyltransferases, nsp14 and nsp16, play key roles in RNA capping and are considered valuable targets for development of anti-coronavirus therapeutics. But little is known about the kinetics of nsp10-nsp16 methyltransferase activities of most HCoVs, and reliable assays for screening are not available. Here, we report the expression, purification, and kinetic characterization of nsp10-nsp16 complexes from six HCoVs in parallel with previously characterized SARS-CoV-2. Probing the active sites of all seven by SS148 and WZ16, the two recently reported dual nsp14 / nsp10-nsp16 inhibitors, revealed pan-inhibition. Overall, our study show feasibility of developing broad-spectrum dual nsp14 / nsp10-nsp16-inhibitor therapeutics.

1. Introduction

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused the recent devastating COVID-19 pandemic. Among the coronaviruses (CoVs) classified by the International Committee on Taxonomy of Viruses (ICTV), only seven have infected humans (HCoVs), to-date [[1], [2], [3], [4], [5]]. In addition to SARS-CoV-2, the severe acute respiratory syndrome coronavirus (SARS-CoV) and the Middle East respiratory syndrome coronavirus (MERS-CoV) have caused severe infections in humans and were responsible for the 2003 and 2012 outbreaks with high fatality in China and the Middle East, respectively [[6], [7], [8], [9], [10], [11], [12]]. In contrast, HCoV-NL63, HCoV-229E, HCoV-OC43, and HCoV-HKU1, have been associated with mild respiratory and gastrointestinal symptoms. All seven HCoVs have been transmitted to humans from animals (Supplementary Table S1) [5,12,13]. The HCoVs belong to the Coronaviridae family of viruses with single-stranded positive RNA genome, encoding 16 non-structural proteins (nsps) and four main structural and some accessory proteins [[14], [15], [16]].

Nsp14 and nsp16 are two S-adenosylmethionine (SAM)-dependent methyltransferases that are involved in the viral RNA capping and enabling the virus to evade immune systems of humans. Nsp14 catalyzes the methylation of viral RNA at N7GpppN (forming a cap-0, N7mGpppN), and nsp16 in complex with nsp10 catalyzes the 2′-O-methylation (forming a cap-1, N7mGpppN2′-Om) [17,18]. The methyltransferase activity of SARS-CoV was previously confirmed using N7mGpppACCCCC as substrate and a filter-based radiometric (³H-SAM) assay. [19] Recently, we have reported the full kinetic characterization of methyltransferase activities of SARS-CoV-2 nsp14 and nsp10-nsp16 complex using GpppACCCCC and N7mGpppACCCCC substrates, respectively [20,21], and discovery of bi-substrate [21] and nanomolar inhibitors [22] of nsp14 methyltransferase activity. Through screening of a customized library of 161 potential methyltransferase inhibitors, seven compounds were identified to inhibit SARS-CoV-2 nsp14 methyltransferase activity [21]. Screening these seven compounds against SARS-CoV-2 nsp10-nsp16 complex resulted in identifying two compounds, SS148 and WZ16 (Fig. 1 ), that also inhibited nsp10-nsp16 activity [23]. Using MTase-Glo bioluminescence assay (Promega), it has been shown that the SARS-CoV-2 nsp10-nsp16 methyltransferase activity requires divalent cations, and N7mGpppAUUAAA RNA is a better suited substrate. [24] This RNA substrate matches the naturally occurring ribonucleotides at the 5′-end of SARS-CoV-2 mRNAs [24,25].

Fig. 1.

Chemical structures of SS148 and WZ16. [23].

Given the importance of nsp10-nsp16 complex in protection of viral RNA during replication, this viral protein has been suggested as an important target for development of antiviral therapeutics [15,[25], [26], [27]]. To-our knowledge, no comprehensive kinetic characterization studies has been reported for nsp10-nsp16 complexes for HCoVs, beside SARS-CoV-2. In this study, we report the purification and kinetic characterization of nsp10-nsp16 complexes from all HCoVs in parallel using the two RNA substrates N7mGpppACCCCC and N7mGpppAUUAAA. We also used SS148, WZ16 (Fig. 1), and SAH (the product of the methylation reaction) as tools to probe the differences in patterns of inhibition of SARS-CoV-2 nsp10-nsp16 inhibitors within HCoVs. Our data supports the idea of development of broad-spectrum anti-coronavirus therapeutics.

2. Results

We previously reported the full kinetic characterization of SARS-CoV-2 nsp10-nsp16 methyltransferase (MTase) activity and development of high-throughput screening for discovery of inhibitors for this target towards development of antiviral therapeutics for SARS-CoV-2. [20] Considering the lack of such comprehensive nsp16 kinetic characterization for the other coronaviruses infecting humans, and the importance of discovery of broad-spectrum nsp10-nsp16-inhibitor based anticoronaviral therapeutics, we set to characterize all seven nsp10-nsp16 complexes from these coronaviruses in parallel.

2.1. Purification of nsp10-nsp16 complexes

Nsp10 binding is essential for methyltransferase activity of SARS-CoV-2 nsp10-nsp16 complex [18]. We previously prepared and characterized this complex with a ratio of 8 (nsp10) to 1 (nsp16) for maximum MTase activity [20]. Therefore, all nsp16 proteins were prepared in complex with nsp10 from the same species at the ratio of 8 (nsp10) to 1 (nsp16) to ensure the stability and full methyltransferase activity (Fig. 2 ). The production of active nsp10-nsp16 complexes proved to be challenging and various vector types, cloning tags, expression systems, and purification approaches were tested (Supplementary Table S2). All complexes, except HCoV-HKU1, were prepared by expressing and purifying the nsp10 and nsp16 individually from different expression systems followed by preparing the active complexes at about 8 (nsp10):1 (nsp16) ratio. For HCoV-HKU1 however, nsp10 and nsp16 complex was purified from the co-infection of sf9 cells with the corresponding viruses at the 1:1 ratio. The 8 (nsp10) to 1 (nsp16) molar ratio complex was then prepared by addition of pure HCoV-HKU1 nsp10. For all complexes, the purity and correct molecular weight was confirmed by SDS-PAGE and mass spectrometry (Fig. 2).

Fig. 2.

Production of nsp10-nsp16 complexes from the seven coronaviruses. Complexes were all prepared at 8 (nsp10) to 1 (nsp16) molar ratio from SARS-CoV-2, SARS-CoV, HCoV-OC43, HCoV-229E, HCoV-NL63, and MERS-CoV. However, for HCoV-HKU1, initially nsp10 and nsp16 complex was purified from sf9 cells co-infected by corresponding viruses at ratio of 1 to 1, followed by addition of separately prepared HCoV-HKU1 nsp10 to prepare the 8 (nsp10) to 1 (nsp16) molar ratio complex as described in material and methods. For each complex, 3–6 μg of protein was resolved on the SDS-PAGE and stained with Coomassie Blue. Higher and lower molecular weight bands correspond to nsp16 and nsp10 for each complex, respectively. Higher intensities for nsp10 bands are due to the 8 (nsp10) to 1 (nsp16) ratio of the complexes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2.2. Kinetic characterization of nsp10-nsp16 MTase activity

SARS-CoV-2 nsp10-nsp16 complex is fully active with biotinylated N7mGpppACCCCC as substrate [20]. We employed the same substrate for kinetic characterization of all seven nsp10-nsp16 complexes in parallel. SARS-CoV, HCoV-OC43, HCoV-HKU1 and HCoV-NL63 were all active with this substrate with similar K _m ^app values for SAM (5.7 ± 1.1, 7.1 ± 0.8, 4.8 ± 1.2, and 8.3 ± 0.9 μM, respectively) as that for SARS-CoV-2 nsp10-nsp16 complex (5.0 ± 0.7 μM). However, the K _m ^app for this RNA substrate varied from 0.1 ± 0.01 μM for HCoV-NL63 to 0.7 ± 0.1 μM for SARS-CoV-2 (Table 1 , Supplementary Fig. S1). Nsp10-nsp16 complexes from HCoV-229E and MERS-CoV did not show any activity with this substrate. Despite the variation of k _cat ^app values within the seven nsp10-nsp16 complexes, the catalytic efficiencies (k _cat /K _{m RNA}) of four nsp10-nsp16 complexes from SARS-CoV-2, SARS-CoV, HCoV-OC43 and HCoV-HKU1 are relatively close (56, 65, 78, and 87 μM⁻¹ h⁻¹, respectively), but it was significantly lower (21 μM⁻¹ h⁻¹) for HCoV-NL63 (Table 1).

Table 1.

Kinetic characterization of nsp10-nsp16 complexes from seven coronaviruses using two RNA substrates in parallel.

nsp10-nsp16	RNA (N7mGpppACCCCC)			RNA (N7mGpppAUUAAA)
	Kmapp (μM)		kcatapp (h−1)	Kmapp (μM)		kcatapp (h−1)
	SAM	RNA		SAM	RNA
SARS-CoV-2	5.0 ± 0.7*	0.7 ± 0.1*	39.0 ± 1.8*	2.8 ± 0.4	0.097 ± 0.029	23.5 ± 2.0
SARS-CoV	5.7 ± 1.1	0.3 ± 0.01	19.6 ± 2.1	2.6 ± 0.4	0.092 ± 0.022	20.9 ± 1.4
HCoV-OC43	7.1 ± 0.8	0.2 ± 0.01	15.6 ± 1.4	2.9 ± 0.4	0.075 ± 0.022	18.8 ± 2.2
HCoV-HKU1	4.8 ± 1.2	0.3 ± 0.02	26.2 ± 1.4	2.3 ± 0.03	0.04 ± 0.003	6.9 ± 0.3
HCoV-NL63	8.3 ± 0.9	0.1 ± 0.01	2.1 ± 0.3	0.8 ± 0.03	0.02 ± 0.002	1.1 ± 0.1
HCoV-229E	ND	ND	NA	1.8 ± 0.2	0.06 ± 0.008	19.7 ± 2.6
MERS-CoV	ND	ND	NA	2.7 ± 0.1	0.14 ± 0.01	16.5 ± 2.0

ND: Not determined., NA: Not applicable. *We previously reported the apparent kinetic parameters for SARS-CoV-2 nsp10-nsp16 complex with N7mGpppACCCCC as substrate [20]. We repeated the experiments in parallel in this study for direct comparison.

In search for a suitable RNA substrate for HCoV-229E and MERS-CoV, we tested a more customized SARS-CoV-2 RNA substrate, N7mGpppAUUAAA (PDB ID: 7L6T) [24,25]. Surprisingly, all seven nsp10-nsp16 complexes were active with this substrate with RNA K _m ^app values as low as 20 ± 2 nM for HCoV-NL63. All other complexes showed RNA K _m ^app values below 140 nM. The SAM K _m ^app values were more or less in the same range as for the first tested substrate (0.8 to 2.9 μM) (Table 1, Fig. 3 ). SARS-CoV-2, HCoV-229E and MERS-CoV nsp10-nsp16 complexes showed catalytic efficiency of 242, 328 and 117 μM⁻¹ h⁻¹ for N7mGpppAUUAAA as substrate, respectively. This is a confirmation that this RNA is a better substrate for SARS-CoV-2 nsp10-nsp16 complex (4-fold higher catalytic efficiency) (Table 1), and surprisingly, even better substrate for HCoV-229E nsp10-nsp16 than SARS-CoV-2 or MERS-CoV (Table 1).

Fig. 3.

Kinetic characterization of MTase activities of seven nsp10-nsp16 complexes with N7mGpppAUUAAA substrate. The optimized radiometric MTase assay were used to determine the SAM and RNA K_m^app values for nsp10-nsp16 from (A, B) SARS-CoV-2, (C, D) SARS-CoV, (E, F) HCoV-OC43, (G, H) HCoV-229E, (I, J) MERS-CoV, (K,L) HCoV-HKU1, and (M, N) HCoV-NL63, respectively. All values are presented as the mean ± standard deviation of three independent experiments (n = 3). The values are also presented in Table 1.

2.3. Nsp16 is conserved across coronavirus species

Alignment of the ORF1ab of SARS-CoV-2, SARS-CoV, MERS-CoV, HCoV-OC43, HCoV-HKU1, HCoV-229E and HCoV-NL63 clearly revealed the conservation of enzymes involved in the RNA capping processes (nsp10, nsp13, nsp14, and nsp16) compared to other nonstructural proteins (Fig. 4 ). Alignment of nsp16 amino acid sequences from the seven coronaviruses also indicated significant conservation within this group of coronaviruses (>53% in all cases), with SARS-CoV and SARS-CoV-2 being the closest (>93% sequence identity), and HCoV-229E and HCoV-NL63 relatively more distant from SARS-CoV-2 (57% and 59% identity, respectively) from SARS-CoV-2 (Table 2 , Supplementary Fig. S2). Conservation of nsp10-nsp16 across these coronaviruses indicates that it is possible to develop broad-spectrum nsp10-nsp16 MTase inhibitors and likely anti-coronavirus therapeutics.

Fig. 4.

Alignment of the ORF1ab of the seven HCoVs known to infect humans. ORF1ab of SARS-CoV-2, SARS-CoV, MERS-CoV, HCoV-OC43, HCoV-HKU1, HCoV-229E and HCoV-NL63 were aligned with the coding sequences of nsp proteins using Clustal Omega program. The nsp10, nsp13, nsp14, and nsp16 sequences are corresponding to the encoding residues 4231–4369, 5325–5925, 5926–6452, and 6776–7073 of SARS-CoV-2 ORF1ab, respectively. [26] These four proteins were then manually re-aligned (using Jalview) with the rest of HCoV ORF1ab sequences. Finally, the degree of conservation between these sequences was visualized based on the percentage identity of the residues. Dark blue shows the highest degree of identity, while white represents the lowest percentage. Created gaps are shown in gray. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 2.

Matrix representing percent identity of nsp16 amino acid sequences among seven coronaviruses.

2.4. Broad-spectrum inhibition of nsp10-nsp16 MTases

We have recently reported two compounds, SS148 and WZ16, as dual inhibitors of SARS-CoV-2 nsp10-nsp16 and nsp14 MTase activities [21,23]. To further assess the possibility of developing broad-spectrum inhibitors of coronavirus nsp10-nsp16 MTases, we probed the active site of all seven purified nsp16 proteins by monitoring the inhibition of their activities using these two compounds and the product of the reaction, S-adenosylhomocysteine (SAH) (Fig. 5 , Table 3 , Supplementary Table S3, Supplementary Fig. S3). Interestingly, SAH inhibition was more pronounced with N7mGpppAUUAAA as a substrate for all seven nsp10-nsp16 complexes compared to N7mGpppACCCCC. HCoV-NL63 nsp10-nsp16, showed the highest difference with SAH IC₅₀ values of 80 nM and 2400 nM using these two RNA substrates, respectively (Table 3). The 30-fold lower SAH IC₅₀ may partly reflect the lower SAM K _m (10-fold) of HCoV-NL63 nsp10-nsp16 for N7mGpppAUUAAA substrate. Similarly, SS148 and WZ16 inhibited the MTase activities of all seven coronaviruses, supporting the possibility of development of broad-spectrum nsp10-nsp16 inhibitors.

Fig. 5.

The inhibitory effect of SS148, WZ16 and SAH on nsp10-nsp16 complexes from seven human coronaviruses using the N7-meGpppAUUAAA as substrate. The nsp10-nsp16 complexes from (A, B, C) SARS-CoV-2, (D, E, F) SARS-CoV, (G, H, I) HCoV-OC43, (J, K, L) HCoV-HKU1, (M, N, O) HCoV-NL63, (P, Q, R) HCoV-229E, and (S, T, U) MERS-CoV were tested against SS148, WZ16, and SAH, respectively. IC₅₀ values are also presented in Table 3. All values are presented as the mean ± standard deviation of three independent experiments (n = 3).

Table 3.

Inhibition of methyltransferase activities of nsp10-nsp16 complex from seven human coronaviruses (HCoVs) with different RNA as a substrate.

	IC50 (μM)
	RNA (N7mGpppACCCCC)			RNA (N7mGpppAUUAAA)
	SS148	WZ16	SAH	SS148	WZ16	SAH
SARS-CoV-2	1.7 ± 0.1⁎	4.6 ± 0.2⁎	2.9 ± 0.4⁎	1.0 ± 0.05	2.5 ± 0.2	0.5 ± 0.2
SARS-CoV	2.0 ± 0.3	2.9 ± 0.3	4.0 ± 1.2	1.0 ± 0.1	2.5 ± 0.1	1.5 ± 0.5
HCoV-OC43	2.6 ± 0.3	3.7 ± 0.6	4.4 ± 1.2	1.0 ± 0.2	5.1 ± 0.8	1.4 ± 0.3
HCoV-HKU1	1.3 ± 0.06	4.7 ± 0.4	2.1 ± 0.1	0.4 ± 0.09	4.5 ± 0.09	0.2 ± 0.05
HCoV-NL63	0.4 ± 0.08	4.2 ± 0.5	2.4 ± 0.4	0.2 ± 0.04	5.1 ± 0.3	0.08 ± 0.01
HCoV-229E	NT	NT	NT	1.1 ± 0.2	24 ± 3	0.9 ± 0.2
MERS-CoV	NT	NT	NT	1.5 ± 0.2	9.3 ± 0.6	0.5 ± 0.1

^⁎We previously reported the IC₅₀ values for SARS-CoV-2 nsp10-nsp16 complex using N7mGpppACCCCC as substrate [23]. We repeated the experiments in parallel in this study for direct comparison.

2.5. Homology modeling and molecular docking

To gain more insight into conservation of the ligand binding sites of the nsp16 proteins of these coronaviruses, we compared their structures. To date, crystal structures of the SS148 and WZ16 inhibitors only in complex with SARS-CoV-2 nsp16 are available (PDB IDs: 7R1T, 7R1U). [23] In case of SARS-CoV (2XYR) [19], MERS-CoV (5YNB), and HCoV-OC43 (7NH7) [28], crystal structures of the nsp16 proteins in complex with other ligands such as a pan-methyltransferase inhibitor sinefungin are known. In case of HCoV-229E, HCoV-HKU1, and HCoV-NL63, no structure of the nsp16 protein is available and thus, we used homology modeling. The SS148 and WZ16 inhibitors were placed into the ligand binding pockets by molecular docking.

Superposition of the ligand binding sites of the nsp16 proteins of these seven human coronaviruses revealed that they are highly conserved and the mechanism of their interaction with the SS148 and WZ16 inhibitors is nearly identical (Fig. 6 ). Four main amino acid residues involved in these interactions (Asn43, Asp99, Asp114, and Asp130 in SARS-CoV nsp16) are absolutely conserved among these coronaviruses. The only exception is the residue Asn101, which can form a hydrogen bond with the 2′-O atom of the ribose moiety of the respective ligand. This residue is conserved in SARS-CoV, SARS-CoV-2, MERS-CoV, HCoV-HKU1, and HCoV-NL63, however, it is replaced with hydrophobic residues Tyr101 and Val100 in HCoV-OC43 and HCoV-229E, respectively.

Fig. 6.

Superposition of ligand binding sites of nsp16 proteins of seven human coronaviruses. Detailed view of the nsp16-SS148 (a) and nsp16-WZ16 (b) interactions. Crystal structures of SARS-CoV (cyan), SARS-CoV-2 (green), MERS-CoV (blue), and HCoV-OC43 nsp16 (yellow) and homology models of HCoV-229E (orange), HCoV-HKU1 (red), and HCoV-NL63 nsp16 (violet) were aligned. Ligands and side chains of selected nsp16 amino acid residues are shown in stick representation with carbon atoms colored according to the structure assignment and other atoms colored according to elements: nitrogen, blue; oxygen, red; sulphur, yellow. Selected hydrogen bonds involved in the nsp16-SS148 and nsp16-WZ16 interactions are depicted as dashed black lines. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2.6. Mechanism of action of SS148 and WZ16

Typically, the mechanism of action (MOA) for methyltransferase inhibitors could be determined by IC₅₀ determination at saturation concentrations of substrate and varying SAM concentration (versus SAM), or by keeping SAM concentration at saturation and varying substrate concentrations (versus substrate) as previously described. [23,29] We therefore determined the IC₅₀ values for each compound with each nsp10-nsp16 protein from the 7 HCoVs at 0.5× and 3× of the K_m of each substrate (SAM or RNA) while the second substrate (SAM or RNA) was at close to saturation concentrations (Table 4 , Supplementary Fig. S4). Increase, decrease or no change in IC₅₀ values upon increase in concentration of the substrate could be an indication of competitive, uncompetitive or noncompetitive patterns of inhibition, respectively.

Table 4.

Mechanism of action of SS148 and WZ16. Significant increase and decrease in IC₅₀ values are highlighted by green (competitive pattern of inhibition) and yellow (uncompetitive pattern of inhibition), respectively. No significant change in IC₅₀ values could be an indication of noncompetitive pattern of inhibition. All values are from Supplementary Fig. S4. All experiments were performed in triplicate.

Except for HCoV-NL63 and HCoV-HKU1, nsp10-nsp16 complexes from all HCoVs presented a pattern of uncompetitive for SS148 inhibition with respect to RNA substrate, showing significant decrease in IC₅₀ values at 3× K_m of RNA concentration. This indicates that the presence of RNA substrate is required for binding of SS148. At 3× K_m concentration of SAM, minor increase in IC₅₀ values for SS148 was observed for nsp10-nsp16 complexes from SARS-CoV-2 and SARS-CoV and HCoV-HKU1, which indicates a weak SAM competitive pattern of inhibition (Table 4). However, for HCoV-OC43, HCoV-229E and MERS-CoV, SS148 showed a noncompetitive pattern of inhibition (change in concentration of SAM had no significant effect on SS148 IC₅₀ values). These data indicate that although the nsp16 active site of all HCoVs are highly conserved, yet some minor differences in mechanism of inhibition of SS148 may exist within these proteins.

For WZ16 (Table 4, Supplementary Fig. S4), a pattern of SAM competitive inhibition was observed for nsp10-nsp16 complexes from SARS-CoV, HCoV-HKU1, HCoV-229E, and MERS-CoV, while SARS-CoV-2, HCoV-OC43 and HCoV-NL63 showed noncompetitive pattern (no significant effect of SAM concentration). The effect of RNA, however, was significantly different for various nsp16 complexes.

Overall, our study indicates that all seven nsp10-nsp16 complexes will likely be inhibited by similar small molecules. However, the extent of such inhibition may differ between HCoVs.

3. Discussion

The emergence of HCoVs not only has claimed the lives of millions of people but has also caused unprecedented socioeconomic challenges worldwide. The recent global pandemic was caused by SARS-CoV-2 (COVID-19) which is genetically highly similar to SARS-CoV (SARS in 2003). However, SARS-CoV-2 showed a lot higher transmissibility, which also accelerated the generation of new variants with enhanced virulence, pathogenicity, and transmissibility [30]. Despite the rapid development of different types of vaccines, the fight against COVID-19 proved to be challenging and required administration of several vaccine doses, without yet reaching the full immunity against COVID-19. Generally, vaccines come with logistic challenges in ensuring global access and affordability. Importantly, they are preventative measures and not a cure for those already infected. This necessitates the invention of more accessible alternative therapeutics for targeting the existing and emerging coronaviruses. Most recently, Pfizer discovered Paxlovid™, which received Emergency Use Authorization in December 2021 in the United States [31]. Paxlovid is a Nirmatrelvir/ritonavir combination protease inhibitor that blocks replication of SARS-CoV-2 and reduces the risk for hospitalization and death among patients with mild to moderate COVID-19 [31,32].

Development of additional anti-coronavirus therapeutics by targeting other highly conserved and essential viral nonstructural enzymes is still an unmet need and could lead to broad-spectrum anti-coronavirus treatments. The development of therapeutics that can be effective on a wider range of coronaviruses relies on similarities of the target protein among these viruses. To this end, we further investigated the conservation of nonstructural proteins within larger number of coronaviruses (total of 43 species) and observed the same pattern of conservation as for HCoVs, indicating nsp16 as one of the most conserved nonstructural proteins, along with nsp14 and nsp13 (Supplementary Table S4, and Supplementary Fig. S5 and S6).

Through viral RNA-capping, nsp10-nsp16 complex enables the CoVs to evade the human immune system [33] and its activity is essential for efficient coronavirus RNA synthesis [34]. In vitro, the lack of SARS-CoV nsp10-nsp16 activity resulted in significant reduction in mutant virus titers at late time points in both an immunocompetent respiratory cell line and primary human airway cultures [35]. Similarly, in vivo attenuation was observed in viral replication for SARS-CoV nsp16-mutant infection [35]. Therefore, inhibition of nsp10-nsp16 complex methyltransferase activity could impair the viral mRNA capping, and replication [27,33,36]. The crucial function of nsp10-nsp16 complex in viral replication, combined with high degree of conservation of nsp16 among coronaviruses strongly suggests that this enzyme would be an excellent target for development of anti-coronaviral therapeutics. High druggability of methyltransferase [37,38], further supports feasibility of this approach. Understandably achieving broad spectrum nanomolar inhibition of nsp14/nsp16 from all HCoVs could be challenging. Altogether, our data strongly suggest the feasibility of broad-spectrum nsp10-nsp16-inhibitor development towards next generation anti-coronavirus therapeutics.

4. Materials and methods

4.1. Sequence analysis

The full amino acid sequences of 43 CoVs species including the seven HCoVs here referred to as HCoV-229E (P0C6X1), HCoV-HKU1 (P0C6X3), HCoV-NL63 (P0C6X5), HCoV-OC43 (P0C6X6), MERS-CoV (K9N7C7), SARS-CoV (P0C6X7), and SARS-CoV-2 (P0DTD1) were obtained from the ORF available on the Uniprot database. The nsp16 amino acid sequences of the seven HCoVs were also obtained from the ORF1ab available on Uniprot. The sequences were aligned using Clustal Omega and the sequence similarities and conservation were determined by Jalview software (version 2.11.2.0). Their percent identity matrix was determined using Clustal Omega.

4.2. Nsp10 and nsp16 cloning, expression and purification

Cloning, expression and purification details for all nsp10-nsp16 complexes are described in the Supplementary material and methods and Supplementary Table S2. The SARS-CoV-2 protein cloning, expression and purification was performed as previously described [20].

4.3. Reagents for biochemical assays

Biotinylated N7mGpppACCCCC-Biotin RNA and biotinylated N7mGpppAUUAAA-Biotin RNA substrates were synthesized by BioSYNTHESIS (Lewisville, Texas). 96-well FlashPlates and ³H-SAM were from PerkinElmer (Waltham, MA). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). SAM2 Biotin-Capture Membrane was obtained from Promega (Madison, WI). All reaction buffers contained 0.4 U/μL RNaseOUT ribonuclease inhibitor (Invitrogen, Waltham, MA).

4.4. Kinetic characterization for seven nsp10-nsp16 MTases

Radiometric methyltransferase assays were performed using biotinylated RNA substrates and ³H-SAM in a reaction buffer (50 mM Tris-HCl pH 7.5, 5 mM DTT, 1.5 mM MgCl₂ and 0.01% Triton X-100). [20] Apparent K _m values were determined for N7mGpppACCCCC (RNA1) or N7mGpppAUUAAA (RNA2) by varying RNA concentration (5 nM – 5 μM for RNA1 and 1.25 nM −1.25 μM for RNA2) and keeping SAM at a saturation concentration of 20 μM (5 μM ³H-SAM plus 15 μM cold SAM). Apparent K _m values were determined for SAM in reactions with RNA1 or RNA2 by varying the SAM concentration. The highest concentration of SAM was 25 μM (5 μM ³H-SAM plus 20 μM cold SAM), and RNA concentration was kept at 2 μM. Ten microliter reactions were initiated with SAM and allowed to proceed for 20 min at 23 °C before being quenched with excess GuHCl. 10 μl of reaction mixture each was captured on SAM²® Biotin Capture Membranes (Promega, Madison WI) and incorporated radioactivity was quantitated by liquid scintillation analysis using a Tricarb scintillation counter (PerkinElmer). The kinetic data were analyzed using a Michaelis-Menten model in GraphPad Prism.

4.5. IC₅₀ determination

Two dual inhibitors (SS148, WZ16) of nsp14 and nsp16 MTases and control SAH were tested at various concentrations from 10 nM to 100 μM to determine their half-maximal inhibitory concentration (IC₅₀) values against each nsp10-nsp16 MTase, using biotin-RNA1 or biotin RNA2 as a substrate. The final reaction mixture consisted of 100 nM nsp16, 2 μM ³H-SAM, 0.3 μM RNA1 or RNA2 in 50 mM Tris-HCl, pH 7.5, 1.5 mM MgCl₂, 5 mM DTT, and 0.01% Triton X-100. The reaction time was 10 min at 23 °C. All enzymatic reactions were performed in triplicate, and IC₅₀ values were determined by fitting the data to Four Parameter Logistic equation using GraphPad Prism software.

4.6. Homology modeling and molecular docking

Homology modeling was carried out using previously reported protocols [39,40]. Briefly, structures of SARS-CoV-2 nsp16 in complex with the inhibitors SS148 and WZ16 and structures of SARS-CoV, MERS-CoV, and HCoV-OC43 nsp16 in complex with sinefungin were retrieved from the Protein Data Bank (https://www.rcsb.org) using the entries 7R1T, 7R1U [23], 2XYR [19], 5YNB, and 7NH7 [28], respectively. Homology models of HCoV-229E, HCoV-HKU1, and HCoV-NL63 nsp16 were generated using the Swiss-Model server (https://swissmodel.expasy.org) [41]. Energy minimization of the modeled structures was performed using the Swiss PDB Viewer [42]. Molecular docking of the SS148 and WZ16 inhibitors was carried out using AutoDock v4.2.6 [43]. Structural figures were generated with the PyMOL Molecular Graphics System v2.5 (Schrödinger, LLC).

4.7. Mechanism of action studies

The MOA of SS148 and WZ16 was determined by measuring the IC₅₀ values of both inhibitors for seven nsp10-nsp16 complexes at a fixed concentration of RNA substrate (N7mGpppAUUAAA) at 5× K_m RNA and two concentrations of SAM (0.5× K_m SAM and 3× K_m SAM), and at two concentrations of RNA substrate (0.5× K_m RNA and 3× K_m RNA) and fixed concentration of SAM (5× K_m SAM). Experiments were performed in triplicate and the data were analyzed using GraphPad Prism 9.

Author contributions

F.L designed and performed all in vitro experiments and analyzed data. P.G and AK·Y analyzed and aligned sequences. P.L, P.G, AK·Y designed the expression constructs, T.H and M.Ku expressed and purified proteins. M.Kl and E.B performed Homology modeling and molecular docking. A.SM.L performed MOA experiments. I·C contributed to protein expression and assays, P.L cloned all constructs, A.S and A.H expressed selected proteins in sf9. S·P contributed to manuscript text. M.V conceptualized the idea, designed experiments, analyzed and reviewed data, and supervised all in vitro studies, and led the project. F.L, P.G and M.V compiled data and wrote the manuscript. All authors reviewed and approved the manuscript.

Funding and additional information

MV was funded by the University of Toronto COVID-19 Action Initiative-2020 (#2), COVID-19 Mitacs Accelerate postdoctoral award to AK·Y and S·P, and US NIH grant 1U19AI171110–01.

EB was funded by European Regional Development Fund; OP RDE; Project: “Chemical biology for drugging undruggable targets (ChemBioDrug)”[CZ.02.1.01/0.0/0.0/16019/0000729]. Czech Academy of Sciences [RVO: 61388963] is also acknowledged.

CRediT authorship contribution statement

Fengling Li: Investigation, Methodology, Data curation, Validation, Visualization, Writing – review & editing. Pegah Ghiabi: Investigation, Data curation, Visualization, Writing – original draft, Writing – review & editing. Taraneh Hajian: Investigation, Data curation, Visualization, Writing – review & editing. Martin Klima: Investigation, Methodology, Data curation, Formal analysis, Writing – review & editing. Alice Shi Ming Li: Investigation, Data curation, Visualization, Writing – review & editing. Aliakbar Khalili Yazdi: Investigation, Data curation, Visualization, Writing – review & editing. Irene Chau: Investigation, Data curation. Peter Loppnau: Investigation. Maria Kutera: Investigation. Almagul Seitova: Investigation. Albina Bolotokova: Investigation. Ashley Hutchinson: Investigation. Sumera Perveen: Investigation, Validation. Evzen Boura: Supervision, Formal analysis, Funding acquisition, Writing – review & editing. Masoud Vedadi: Conceptualization, Supervision, Funding acquisition, Data curation, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material Supplementary Figures, Tables, and Matrial and methods

Data availability

All data are provided within the manuscript or in Supporting information.