Discovery and structural characterization of chicoric acid as a SARS-CoV-2 nucleocapsid protein ligand and RNA binding disruptor

doi:10.21203/rs.3.rs-1720953/v1

The nucleocapsid (N) protein plays critical roles in coronavirus genome transcription and packaging, representing a key target for the development of novel antivirals, and for which structural information on ligand binding is scarce. We used a novel fluorescence polarization assay to identify small molecules that disrupt the binding of the N protein to a target RNA derived from the SARS-CoV-2 genome packaging signal. Several phenolic compounds, including L-chicoric acid (CA), were identified as high-affinity N-protein ligands. The binding of CA to the N protein was confirmed by isothermal titration calorimetry, ¹H-STD NMR, and by the crystal structure of CA bound to the N protein C-terminal domain (CTD), further revealing a new modulatory site in the SARS-CoV-2 N protein. Moreover, CA reduced SARS-CoV-2 replication in cell cultures. These data thus open venues for the development of new antivirals targeting the N protein, an essential and yet underexplored coronavirus target.

Nucleocapsid protein

SARS-CoV-2

packaging signal

chicoric acid

phenolic acids

protein-RNA disruptors

HTS

antivirals

The historic COVID-19 pandemic, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has caused millions of deaths worldwide and affected the world’s economy in an unprecedented way^1,2. Despite the enormous efforts of the scientific and medical community to find drugs to fight the disease since its emergence in late 2019, only three small molecule drugs are currently authorized for clinical use. Remdesivir³ and Molnupiravir⁴ are nucleoside analogues with limited efficacy⁵, whereas Nirmatrelvir⁶, a peptide inhibitor of the 3CL protease, developed by Pfizer, is used in combination with Ritonavir and sold under the brand name Paxlovid™. Although these are promising drugs, kidney and liver toxicity and the potential incompatibility of the Paxlovid combination with other drugs may limit their use to the broad population⁷. Moreover, the continued emergence of new, more contagious SARS-CoV-2 variants challenges the efficacy of vaccines currently in use⁸. Therefore, the development of new and specific drugs capable of treating SARS-CoV-2 infections is still an urgent medical need.

To accelerate the discovery of novel anti-SARS-CoV-2 drug candidates, as well as the repurposing of existing drugs, several high-throughput screening (HTS) campaigns have been performed in the last two years. Such assays have explored the biological activities of several SARS-CoV-2 proteins, including the main 3C-like (3C-L) and papain-like (PL-Pro) proteases, the RNA polymerase and the spike (S) and envelope (E) structural proteins^9–14. In addition, numerous cell-based screening assays with SARS-CoV-2 have also been recently reported^12,15−18. Unlike the S and E proteins, however, the structural nucleocapsid (N) protein, the most abundantly expressed SARS-CoV-2 protein in infected cells, has not been fully exploited as a target for drug development against SARS-CoV-2¹⁹.

The N protein is a phosphoprotein that plays fundamental roles in the virus life cycle, including transcription initiation and packaging of the viral genomic RNA²⁰. This protein has a domain architecture comprising two structured RNA-binding modules, the N-terminal (NTD) and C-terminal (CTD) domains, connected by an intrinsically disordered serine and arginine-rich (SR) linker^21–23. The NTD is implicated in the binding and melting of regulatory elements required for transcription initiation^24,25, whereas the CTD drives protein dimerization and is thought to bind the RNA packaging signal (PS) during virus particle assembly^22,26,27. Phosphorylation of the SR linker by host cell kinases, on the other hand, has been shown to modulate the RNA-binding activity of the protein and lead to liquid-liquid phase separation with the viral RNA^26–29. Thus, given its importance in the various stages of the virus replication cycle, the N protein is considered a promising target for the development of SARS-CoV-2 replication inhibitors.

We report here a novel fluorescence polarization-based HTS assay to identify small molecules that can disrupt the interaction of the N protein with an RNA probe derived from the putative SARS-CoV-2 PS sequence. After screening a customized compound library of approximately 3200 bioactive compounds, several phenolic compounds, including L-chicoric acid (CA), showing IC₅₀ values for RNA probe displacement in the low to sub-micromolar ranges, were identified as potential N protein ligands. Complementary orthogonal biophysical methods, including isothermal titration calorimetry (ITC) and ¹H-saturation transfer difference (STD) NMR further revealed that CA binds to the CTD. This was confirmed by the crystal structure of the CTD in complex with CA, determined at 1.7 Å resolution, which represents the first CTD structure bound to a small molecule. In addition, we show that CA inhibited SARS-CoV-2 replication in human lung cells, further suggesting that it may influence N protein-mediated RNA packing in vivo. Taken together, our data provide the first evidence of pharmacologically targeting the SARS-CoV-2 N protein with small molecules, thus paving the way for the rational design of new N protein modulators.

Development of an HTS assay to find small molecules capable of inhibiting the RNA-binding activity of the N protein

In order to establish an assay for the identification of potential inhibitors of the N protein RNA-binding activity, we first searched for an N protein target RNA, and the SARS-CoV-2 PS sequence was a candidate. The putative PS of the SARS-CoV-2/SP02/human/2020/BRA strain (nucleotides 19785-20364) was identified by sequence alignment to the SARS-CoV and MERS-CoV PS sequences^30,31. Within this sequence, we selected the CACUCACUGUCUUUUUUGAUGGUAGAGU stem loop (RNA1) as an N protein target (Figure 1A) because this stem loop is conserved in all SARS-CoV-2 genomes and has an invariable ‘UUUUUU’ motif at the loop.

The RNA-binding activity of the N protein was monitored by a fluorescence polarization (FP) assay using distinct 5’-FITC-labeled RNAs as probes. In addition to RNA1, four other probes were tested (Figure 1A). In the RNA2 probe (CACUCACUGUCAAAAAAGAUGGUAGAGU) the ‘UUUUUU’ motif was replaced by ‘AAAAAA’, whereas in RNA3 (CACUCACUGUCUUUUUU), the stem loop 5’ arm was deleted. RNA4 (CACUCACUGUC) also lacked the ‘UUUUUU’ motif, while RNA5 (AUAUAGCUAC) served as a scramble negative control (Figure 1A). As shown in Figure 1B, optimum binding was observed with the RNA probes 1 and 3 (K_D= 124.2 ± 6.5 and 148.6 ± 7.1 nM), respectively, suggesting that the ‘UUUUUU’ motif, but not the hairpin structure, is critical for the interaction with the N protein. Therefore, RNA1 was chosen for the establishment of the protein-RNA dissociation assay used in our HTS trials.

HTS campaigns reveal polyphenols as privileged scaffolds that disrupt the binding of the N protein to the target RNA

Following the identification of RNA1 as a high-affinity N protein-binding probe, an FP assay was developed. After checking the assay performance under the screening conditions (Figure S1A), we conducted an HTS campaign testing a library of ~3200 approved drugs and bioactive molecules (Figure 1C). Differential values of approximately 170 mP were observed between the positive and negative controls (Figure S1B), resulting in Z’ scores greater than 0.7 for all tested plates (Figure 1D). Thus, the robustness of the HTS assay allowed us to reliably select hit candidates for concentration-response follow-up experiments.

The first criterium for selecting hit candidates from the primary screening was to choose small molecules that reduced the binding of the N protein to RNA1 to less than 30% (Figure 1C), which resulted in a list of 78 compounds. From this preliminary list, 33 compounds were flagged due to autofluorescence, interference with the fluorescence or polarization of the free probe. These criteria led us to select 45 hit candidates, representing an overall hit rate of 1.4%.

To confirm the activity of the hit candidates as inhibitors of the N protein-RNA1 interaction, the selected compounds were subjected to concentration-response experiments and 44 of them had their inhibitory activity confirmed (Table 1).

Table 1. List of selected hit molecules retrieved from the HTS campaign as potential N-protein ligands. Compounds are ranked by their IC₅₀ values for blocking the N protein-RNA1 interaction. Phenyl propanoids with dicaffeoyl motif are highlighted in bold.

Compound	IC₅₀ (µM)	+Error*	-Error*
Chebulinic acid	0.2	0.0	0.0
L-Chicoric acid	0.5	0.1	0.1
Punicalagin	0.7	0.1	0.0
Punicalin	0.7	0.1	0.1
Suramin	0.8	0.1	0.1
Tannic acid	1.4	0.2	0.2
Chlorophyllin B	2.2	0.2	0.3
Corilagin	2.4	0.5	0.4
4,5-Dicaffeoylquinic acid	3.0	0.5	0.3
Methyl Blue	4.2	0.3	0.3
Embelin	5.0	0.9	0.7
Linaclotide	5.5	0.6	0.5
Isochlorogenic acid A	5.5	1.0	0.8
Sennoside A	5.7	0.8	0.7
Lusutrombopag	6.0	0.6	0.5
(R)-(-)-Gossypol acetic acid	6.4	0.4	0.4
Idasanutlin	6.5	0.7	0.6
Sulfamerazine	6.5	0.9	0.8
Eltrombopag	7.1	0.9	0.9
Gossypol acetic acid	7.1	0.5	0.4
IOWH-032	7.7	0.9	0.8
Anacardic Acid	7.7	0.9	0.7
Pranlukast (hemihydrate)	7.8	1.1	1.0
Eltrombopag	8.1	1.0	0.9
Succinobucol	8.1	0.6	0.5
RNPA1000	9.6	1.8	1.4
TMC647055	10.0	1.2	1.0
Montelukast	10.3	1.1	1.0
Micafungin	10.4	0.9	0.8
Zafirlukast	10.4	1.6	1.4
Resazurin	10.5	1.7	1.4
Verteporfin	10.8	3.4	2.6
Chlorophyllin A	10.8	1.2	1.0
Surfactin	11.1	1.5	1.3
Pentagalloylglucose	11.6	1.9	1.7
Hexachlorophene	13.0	1.6	1.3
Ertapenem sodium	13.8	1.4	1.3
MK 0893	14.4	1.1	0,9
Butenafine	15.7	1.3	1.2
Oleic acid	15.7	0.9	0.8
Simeprevir	16.5	0.9	0.9
Sofalcone	18.5	1.6	1.5
Bithionol	23.0	5.6	4.6
Gallic acid	23.3	3.3	2.9

*Reported errors are deviation from mean value for upper (+Error) and lower (-Error) limits of 95 % confidence intervals

Many of the compounds that impaired the binding of the N protein to RNA1 are highly polar. Chebulinic acid (CI), CA, punicalagin (PG), punicalin (PL) and suramin (SU) are submicromolar N protein-RNA1 disruptors (Figure 2). When the N protein was titrated against RNA1 in the presence of CI, CA or PG, we observed that the binding affinity of the N protein to RNA1 was significantly reduced, as revealed by the higher K_D values (Figure S2A), indicating that these compounds inhibit the formation of the N protein-RNA1 complex.

CA binds the CTD and promotes dissociation of the N protein-RNA1 complex

Because caffeic acid derivatives have been recently described as potential inhibitors of SARS-CoV-2 infection³² and CA exhibits antiviral activity against HIV and hepatitis B virus (HBV)^33–35, we decided to further investigate the properties of CA as an N protein ligand. First, the binding affinity of CA for the N protein was determined by ITC, which provided the thermodynamic signature of binding, including K_D, stoichiometry and both the enthalpic and entropic contributions. The results confirmed that CA binds to the N protein with a K_D of 250 ± 7.9 nM at a protein:ligand ratio of 2:1 (Figure 3A), this ratio indicating that there is one CA site per N protein dimer. In addition, the ITC results suggested that the CA binding to the N protein is enthalpy driven.

Given that CA binds to the N protein at nanomolar concentrations, we decided to investigate whether CA could dissociate the N protein-RNA1 complex. In line with the results shown in Figure S2A, we found that CA promoted the dissociation of previously formed N protein-RNA1 complex with a K_D value of 41.1 ± 11.7 µM (Figure 3B). Although this dissociation constant is much higher than the K_D for the CA-N protein interaction, the results are consistent with the fact that RNA1 has a higher binding affinity for the N protein than CA (Figure 1B).

Finally, the interaction of CA with the N protein was examined by the ¹H-STD NMR technique, which allows the identification of the binding epitopes of a ligand when bound to a receptor protein. The results confirmed that CA produces clear STD signals which were mapped to its chemical structure (Figure 3C). Judging by the signal intensity, hydrogen H1 (100% intensity), in CA’s tartaric acid unit, appears to be the most directly involved in N protein binding and thus in closer contact with the protein. In comparison, the H2, H3, H4, H7 and H8 hydrogens, which form the dicaffeoyl units, showed STD signal intensities varying from 70 to 90% (Figure 3C). Importantly, we found that the STD signals were observed with the full-length protein and CTD, but not with the NTD. This suggests the CA-binding site is located on the CTD of N protein.

Together, these results show that CA is a nanomolar N-protein affinity ligand that binds to the CTD and displaces the RNA from the N protein at micromolar concentrations.

The binding mode of CA to the CTD and structural consequences for N protein function

To gain insights into the binding mode of CA to the N protein, we determined the crystal structure of the CTD in complex with this ligand. CTD crystals belonged to the space group P 2₁2₁2₁, with Matthews coefficient³⁶ of 2.20 Å³ Da⁻¹ and solvent content of 44.2%. Datasets for the CTD alone (apo CTD – PDB entry: 7UXX) and in complex with CA (CTD-CA – PDB entry: 7UXZ) were scaled to resolutions of 1.85 and 1.73 Å, respectively. The phases were recovered by molecular replacement, using a previously described SARS-CoV-2 CTD crystal structure (PDB code: 7C22) as the search model³⁷. The models were refined to R_work/R_free values of 17.2 / 20.3% (apo CTD) and 17.4 / 21.5% (CTD-CA). Data collection and refinement statistics are summarized in Table S1.

Six molecules organized as three dimers were found in the asymmetric unit of the CTD. Each protomer in the structures is comprised of five α-helices, two 3₁₀ (η) helices and two antiparallel β-strands (Figure 4A). As reported previously³⁷, the CTD dimer is stabilized by extensive hydrogen bonds connecting the β2 strands and the interaction of residues from the loop between α-helices 1-2 (Arg277, Gly278, Glu280, Gln283 and Asn285). β-strand 1 and α-helix 4 (Gly316, Arg319 and Ile320) also play an essential role in stabilizing the dimer. In addition, the β-hairpins (Leu331 – Leu339), α-helices 3-4 (Ile304, Ala305, Phe314 and Phen315) and α-helix 5 (Phe346 – Leu353) form a hydrophobic core stabilized by Van der Waals interactions across the amino acid side chains.

CA binds to a shallow pocket (volume of ~190 A³) formed between α-helices 1-2 and η-helix 2, close to the C-terminus (Figure 4B). In accordance to the ITC data, only one CA molecule bound to one of the CTD protomers was unambiguously found by inspecting the 2F_o-F_c electron-density map. Nevertheless, this map only partially covers the CA aromatic ring most exposed to solvent (Figure 4B).

A close inspection of the structural complex shows that the binding of CA to the CTD is stabilized mainly by polar contacts of the carboxylate and carbonyl groups from CA’s tartarate and caffeoyl units, involving Arg276/Arg277 and three structural water molecules (Figure 4C). The structure of the complex also shows one CA catechol ring laid inside a hydrophobic canyon formed by η-helix 2, α-helix 2 and the last two C-terminus residues (Phe363 and Pro364).

When the apo and complexed structures were superposed, we observed no significant changes in the overall structures, as judged by the Cα RMSD (0.19 Å deviation). However, we noticed clear conformational changes in the side chains of Phe363, Pro364, Gln289 and Arg276, which contributed to the widening of the CA pocket (Figure 4D and Figure 4E). Arg276 had moved outwards, positioning its guanidine group at ideal distances to form electrostatic (NH1) and water-mediated (NH2) hydrogen-bond interactions with CA’s carboxylate group. Arg277’s main chain amine was already at ideal distances to engage in a hydrogen bond with CA’s other carboxylate group. Arg277 N^ε atom could further form a water-bridged hydrogen bond with Thr271 (Figure 4D), which might probably contribute to stabilizing the polar contact network within the CA-N protein binding site. The catechol ring was also predicted to be important in ligand binding, since Gln289’s polar side chain was turned outwards the pocket, exposing its aliphatic region to the apolar pocket formed by Pro364, Phe363, thus favouring the accommodation of one of CA’s catechol rings, which in turn showed clear electron density (Figure 4E).

The CTD structure in its apo form shows an electrostatic potential distribution that is conserved amongst all N proteins from the Coronaviridae family^21,37, with a major positively charged groove located on one side of the dimer surface (Figure 5A). This region, which is composed by Lys256, Lys257, Lys259 Lys261 and Arg262, is thought to contribute to RNA binding^38–40. We observed that the positively charged groove extended towards the CA-binding pocket through Arg259, Arg276, Arg277 and Arg293 (Figure 5B). Notably, the binding of CA to the CTD not only changes the topology of this region, but also disrupts the continuity of such potential RNA-binding region, with the possibility to affect N protein binding to RNA in a direct (Figure 5C) or indirect way (Figure 5D).

CA displays anti-SARS-CoV-2 activity in in vitro cell assays

After confirming CA as an N protein ligand with potential implication on its RNA binding function, we tested if CA could inhibit SARS-CoV-2 infection in vitro. For this, Calu-3 and Vero CCL81 cells were seeded into 24-well plates at 2x10⁵ and 2.5x10⁵ cells/well, respectively, and infected with SARS-CoV-2 at a multiplicity of infection (MOI) of 1. CA was added to the cell culture medium after infection at 25 and 100 µM final concentrations, whereas DMSO at 0.2% was used as a vehicle control. Assessment of the viral load in the cell culture supernatants collected 48 h post-infection indicated that CA presented antiviral activity at 100 µM only, relative to untreated (vehicle) control. CA treatment caused 10-fold and 100-fold reductions in infectious viral load in Vero CCL81 (p<0.001 - Figure 6A) and Calu-3 (p<0.01 - Figure 6B), respectively. Both cell lines remained viable when treated with CA or DMSO at the tested concentrations (Figure S3). In addition, CA displayed only a slight virucidal activity against SARS-CoV-2 (p<0.05 - Figure 6C), indicating that the observed antiviral effect of CA cannot be attributed to direct action on viral particles. These results thus show that CA inhibits SARS-CoV-2 replication in cell culture at micromolar concentrations, which is consistent with the K_D values for the N protein-RNA1 complex dissociation (Figure 4B).

Despite the great success of newly developed vaccines to prevent contagion and severe forms of infection by SARS-CoV-2, infections are still present, and new waves of contamination by viral variants are recurrent. In this sense, antiviral drugs are needed to improve patient recovery and to prevent disease progression, especially to at-risk patients. The development of such antiviral drugs, in many cases, requires knowledge of the structure and function of potential viral targets and the discovery and characterization of small-molecule modulatory binding sites in these proteins, that afford for structure-based drug design (SBDD) approaches⁴¹.

In this work, we describe a novel fluorescence-based high-throughput screening assay that allows the identification of small molecules that interfere with the RNA-binding activity of the SARS-CoV-2 N protein, the most abundant viral protein expressed in host cells, and which plays fundamental roles in transcriptional regulation and virus assembly^20–29. We further characterized the top hits using a cascade of biophysical assays, and solved, for the first time, the crystal structure of the N protein CTD binding a non-endogenous ligand, chicoric acid, further revealing a new modulatory site in the SARS-CoV-2 N protein.

By screening a customized library of bioactive small molecules (Fig. 1), highly polar compounds stood out, highlighting polyphenols (ellagitannins, CI, PG and PL), a diester of tartaric acid (chicoric acid, CA) and a polysulphonated naphthylurea (suramin, SUR) – Fig. 2. The latter compounds were capable of disrupting the interaction of the full-length SARS-CoV-2 N protein with an RNA probe derived from the SARS-CoV-2 PS sequence (RNA1) in the submicromolar range (Figure S2).

In accordance with the literature, the antiparasitic drug suramin is reported to inhibit SARS-CoV-2 infection in cell culture at 20 µM, by interfering with early steps of virus replication⁴². The natural ellagitannins CI, PG and PL have already been reported to exhibit diverse biological properties⁴³, including antiviral activity against multiple human viruses with in vitro potencies in the low micromolar range ^44–46. Other polyphenols, like catechin gallate and gallocatechin gallate, have also been shown to reduce the binding affinity of an RNA oligonucleotide to the N protein of SARS-CoV, causative of the 2003 coranavirus outbreak, on a biochip assay⁴⁷.

Chicoric acid (CA), a symmetric dicaffeoyl ester of tartaric acid, was highlighted in the present study as a new class of N protein modulator and one of the most potent hit compounds identified in our HTS trials. Importantly, the binding of CA to the N protein could be fully characterized by biophysical methods (Fig. 3) and the crystal structure of CA binding to the SARS-CoV-2 N protein CTD could be determined at 1.7 Å resolution (Fig. 4). Although polyphenols have been shown to interfere with coronavirus N proteins before⁴⁷ and compound PJ34 (SARS-CoV NTD ⁴⁸) and GTP (SARS-CoV-2 CTD) interacted with N proteins, our data show that CA is a new class of N protein ligand that binds to a new modulatory site on the SARS-CoV-2 N-protein CTD. Importantly, the CA-binding site is conserved in SARS-CoV and partially conserved in MERS N proteins (Figure S4). To our knowledge, this is the first description of a non-endogenous SARS-CoV-2 N protein ligand and the first report of this modulatory ligand binding site on a coronavirus N protein.

We present evidence that the N protein CTD is sufficient for interaction with CA, as spotted by the ¹H-STD NMR experiment carried out with the full length, CTD and NTD SARS-CoV-2 N protein (Fig. 3C). The crystallographic data further confirmed that CA binds to the SARS-CoV-2 N protein CTD in a shallow pocket located close to the CTD C-terminus and to the putative positively charged RNA binding groove (Fig. 5). CA binding to N protein is stabilized mainly by a polar contact network involving ionic interactions of CA’ symmetric carboxylates with the N protein’s arginines and water-mediated hydrogen bonds of its caffeoyl carbonyl groups. One of the catechol rings of CA is further involved in ligand binding and induces conformational changes in the N protein, by docking to a hydrophobic pocket formed in the CA-binding site upon ligand binding (Fig. 4).

The two carboxylate and two caffeoyl ester moieties seem important for binding to the N protein and to induce conformational changes that affect its RNA binding function. In addition to the crystallographic observations, an initial structure-activity relationship (SAR) analysis could be performed based on 16 compounds assessed in our screening efforts containing caffeoyl substructures. Those bearing two caffeoyl esters clearly stand out (Figure S5A). In CA, the most potent derivative (IC50 = 0.5 µM), the dicaffeoyl units are bound to tartaric acid, whose carboxylate moieties’ contribution for protein binding have been established (Fig. 4). The other three dicaffeoyl derivatives are isomers of dicaffeoylquinic acid, showing at least 10-fold loss in IC50 values (Figure S5B). Assuming that they bind to the same site as CA, one can perceive that the caffeoyl esters could occupy similar positions, especially for 4,5-dicaffeoylquinic acid and isochlorogenic acid A. However, their single carboxylate moiety at C1 would be further way in comparison to CA, but still able to engage some of the polar contacts observed for CA. This observation can be extended to the other caffeoyl derivatives, such as rosmarinic acid and verbascoside which showed low N protein-RNA1 disruption, and which only contain one or two of the 4 key elements found in CA (Figure S5B).

The functional effects of CA binding to the SARS-CoV-2 N protein could be further explored in the present work. The CA-binding site in N protein is located adjacent to the main positively charged groove thought to interact with the RNA^38,39 (Fig. 5). Notably, the binding of CA to the CTD not only alters the topology but also the charge distribution of this region, which might explain why CA disrupts the N protein-RNA1 complex in solution. This hypothesis is additionally supported by a structural model of the SARS-CoV-2 N protein complexed with an RNA molecule, which we have recently reported⁴⁰. This structural model presupposes two possibilities: one in which the CA-binding site fully overlaps with the RNA interaction site (Fig. 5C), and the second where the CA site is near the RNA site (Fig. 5D). In both scenarios, however, it can be anticipated that the binding of CA to the CTD could preclude RNA interaction, therefore modulating the essential N protein functions related to SARS-CoV-2 genomic RNA binding.

CA showed binding affinities for the N protein of 0.5 µM (Fig. 2); nevertheless, the CA equilibrium constant for the N protein-RNA1 complex dissociation was ~ 40 µM (Fig. 3B). Such dissociation constant is consistent with the higher N protein binding affinity exhibited by RNA1, and with the notion that RNA1 has a much larger interaction surface than CA. These data, aligned with the reported low cell permeability of CA^49,50, also help to explain the relatively high CA concentration required to significantly inhibit SARS-CoV-2 replication in human cells (Fig. 6). Although further optimization of CA as an antiviral agent is needed, our data offer the structural basis for the rational design and development of novel antiviral drugs targeting the SARS-CoV-2 N protein, an essential and yet underexplored target of coronaviruses.

Protein expression and purification

The gene sequence corresponding to the full-length SARS-Cov-2 N protein (GenBank QIG56001.1) was amplified from SARS-CoV-2 RNA and cloned into a pET28a-TEV vector for the expression of a 6xHis-fusion protein, as previously described⁵¹. The NTD (residues Q43-E174) and CTD (residues S250-P364) fragments were amplified from the full-length N construct using primers forward (NTD-F 5’-AACGTGGATCCCAAGGTTTACCCAATAATACTG-3’, CTD-F 5’CTAAGGGATCCGCTGCTGAGGCTTCTAAGAAG3’) and reverse (NTD-R 5’-ACTGCCGCGGCCGCTTTATTCTGCGTAGAAGCCTTTTGG-3’, CTD-R 5’CTTTTTAGCGGCCGCTTATGGGAATGTTTTGTATGCGTC3’), respectively, and inserted into the BamHI/NotI sites of a pET-SUMO vector (Invitrogen), carrying a SUMO sequence at the N-terminus. The expression vectors were used to transform Escherichia coli BL21 (DE3) cells (Novagen, USA). Freshly transformed cells were grown in LB-kanamycin (50 μg/mL) medium to OD600nm 0.8 at 37°C. The temperature of the cultures was lowered to 25°C (full-length N) or to 18°C (NTD and CTD), and protein expression was induced with 0.1 mM (full-length N) or 0.5 mM (NTD and CTD) IPTG for 16 h at the respective temperatures. Cells were harvested by centrifugation (4,000 x g, 10 min) and stored at -80°C. To remove nucleic acids of bacterial origin, the proteins were purified under denaturing conditions using urea and high salt concentration²⁶. Frozen cell pellets were thawed and resuspended in buffer A (50 mM sodium phosphate, pH 7.6, 500 mM NaCl, 10% glycerol, 20 mM imidazole, 6 M urea) and lysed by sonication on ice. Lysed cells were centrifuged at 18,000 x g for 40 min at 4°C to remove cell debris and the supernatants were applied onto a HisTrap FF 5mL column (GE healthcare) pre-equilibrated with buffer A. After washings, proteins were eluted in ten column volumes of buffer B (50 mM sodium phosphate, pH 7.6, 500 mM NaCl, 10% glycerol, 500 mM imidazole, 3 M urea). Fractions containing the protein of interest were pooled and dialyzed against buffer C (50 mM sodium phosphate, pH7.6, 500 mM NaCl, 10% glycerol) overnight at 4°C. Except for the N-full construct, the recombinant proteins (NTD and CTD contructs) were cleaved with the appropriate TEV and SUMO proteases. Cleaved tags were removed by reverse affinity chromatography using buffer A and B without urea. Protein fractions were concentrated and fractionated on a size exclusion Superdex 200 16/600 (full-length N) or Superdex 75 16/600 (NTD and CTD) column, previously equilibrated with buffer C.

For crystallization tests, the CTD was purified using Turbonuclease from Serratia marcescens (Sigma, USA). Briefly, bacterial cells after IPTG induction were suspended in lysis buffer (50 mM Tris HCl pH 8.0; 1 M NaCl, 5% glycerol, 1 mM β-mercaptoethanol) containing 200 units of Turbonuclease and lysed by sonication as described above. Lysed cells were centrifuged, and the supernatant was applied onto a HisTrap FF 5mL column pre-equilibrated with buffer D (50 mM Tris HCl pH 8.0, 500 mM NaCl, 5% glycerol, 1 mM β-mercaptoethanol). Bound proteins were eluted using the same buffer containing 500 mM imidazole. The eluate was dialyzed against buffer D overnight at 4°C. After SUMO cleavage and reverse affinity chromatography, the proteins were fractionated on a Superdex 75 16/600 column pre-equilibrated with buffer E (20 mM Tris HCl, pH 8.0, 100 mM NaCl, 1 mM β-mercaptoethanol).

The quality of all protein preparations was verified by SDS-PAGE and dynamic light scattering (DLS). In addition, UV absorbance at 260/280 nm was used to estimate the amount of nucleic acid in the protein samples. Only protein samples with a monodisperse character and a 260/280 nm ratio of 0.5-0.6 were used in the experiments described below.

Fluorescence polarization assay

Chemically synthesized RNA probes 5’-labelled with fluorescein isothiocyanate (FITC) and purified by HPLC were obtained from Thermo Scientific (USA). Probe sequences were as follows: RNA1 (5’-CACUCACUGUCUUUUUUGAUGGUAGAGU-3’), RNA2 (5’-CACUCACUGUCAAAAAAGAUGGUAGAGU-3’), RNA3 (5’-CACUCACUGUCUUGUUUGAUGGUAGAGU-3’), RNA4 (5’-CACUCACUGUCUUUUUU-3’), RNA5 (5’-CACUCACUGUC-3’) and scramble control RNA6 (5’-AUAUAGCUAC-3’),

Fluorescence polarization (FP) assays were used to determine the binding affinity of the N protein to the RNA probes, solubilized in 50 mM sodium phosphate buffer (pH, 7.6). Purified N protein from 2.5 nM to 15 µM in 50 mM sodium phosphate buffer, 100 mM NaCl (pH, 7.6) was mixed with each RNA at 10 nM final concentration, in 384-well plates. FP data was acquired using a ClarioStar microplate reader (BMG LabTech), with excitation and emission wavelengths set to 485 and 530 nm, respectively. Affinity binding curves were fitted to a Hill1 model using the OriginPro software.

Hight-throughput screening assays

A customized library with 3215 nonredundant compounds from the collections ‘FDA-approved’, ‘anti-COVID’, ‘anti-infection’ and ‘anti-virus’, was purchased from MedChemExpress (NJ, USA). The library, in 384-well plates, was diluted to 1 mM concentration in dimethyl sulfoxide (DMSO) and stored at -20°C. Columns 1, 2, 23 and 24 of all microplates were filled with DMSO for screening controls as described below.

Binding of RNA1 to the N protein was monitored by FP, as described above. Screenings were performed in 384-well, flat bottom, black polypropylene microplates (Greiner #781289), using the binding buffer supplemented with 0.01% triton X-100, in a final volume of 25 µL. The final concentration of RNA, N protein, library compound and DMSO were 10 nM, 500 nM, 20 µM and 2% (v/v) respectively. Initially, the assay plates were filled with the RNA probe solution (19.5 µl) using a MultiDrop dispenser (Thermo Fisher) and the compounds (0.5 µl) were transferred from the library to the assay plates in a Janus-MDT liquid handler platform (PerkinElmer). FP measurements were performed at this stage to detect possible interference from library compounds. The N protein (5 µl) was then transferred to all wells of the assay plates using the MultiDrop dispenser, except for columns 1 and 24 which received buffer, RNA and DMSO only, and were used as negative controls (low control, free probe). On the other hand, columns 2 and 23 received buffer, RNA, protein and DMSO, and were considered as positive controls (high control, bound probe). After adding the protein to the assay mix, the plates were incubated at room temperature for 30 min before FP was measured. mP values of positive (100% binding) and negative (0% binding) controls were used for sample data normalization.

Concentration−response curves and IC₅₀ determination

To confirm the activity and evaluate the potency of selected hit candidates, hit compounds were subjected to dose-response assays. Except for compound concentration, the assay conditions were as described above, where 0.5 µL of the compounds were transferred to the assay plates generating a concentration gradient from 1.8 nM to 50 µM. The concentration-response assays were performed in triplicates and after acquisition of the FP data, the percentage of inhibition for each test compound was calculated as follows: % inhibition = (high control average – read value)/(high control average – low control average) × 100. Normalized data was fit to the log4 parameters equation with variable slope to extract the IC₅₀ values with GraphPad Prism 9 (GraphPad LLC, v. 9.3.1).

Biophysical analysis and hit confirmation

Aggregation assays were performed with the full-length N protein diluted to 10 µM in 50 mM sodium phosphate buffer, 100 mM NaCl (pH, 7.6), with subsequent addition of 20 µM of each test compounds or 1% DMSO used as diluent. The same assay was performed with the protein at 2.5 µM in the presence of 500 µM CA. Samples were evaluated by DLS in a ZetaSizer NanoS (Malvern) equipment, at 10°C, using default parameters set by the equipment.

The affinity of RNA1 for the N protein in the presence of selected hit compounds was inspected by FP. Serial dilutions of the protein:compound mixtures at a 1:2 ratios were prepared in 50 mM sodium phosphate buffer, 100 mM NaCl (pH, 7.6). RNA1 at 10 nM was added to the serial dilutions and FP measurements were performed as described above.

To determine the amount of CA required for N protein-RNA1 dissociation, purified N protein at 2.5 µM in 50 mM sodium phosphate buffer, 100 mM NaCl (pH, 7.6) was incubated 10 nM RNA1 at 4 °C for 60 min prior to the addition of increasing amounts of CA up to 500 µM. The FP data were acquired as described above and the affinity binding curves were fitted to a Hill1 model using the OriginPro software.

The dissociation constant and thermodynamic parameters for the N protein-CA interaction were determined by ITC using a a VP-ITC calorimeter (Malvern). Purified N protein was dialyzed against 50 mM sodium phosphate buffer, 500 mM NaCl (pH, 7.6), overnight at 4 °C, and further diluted to 20 µM. The dialysis buffer was used to prepare CA at 250 µM. Both solutions, from cell and syringe, were prepared in the presence of 0.25% DMSO. CA was titrated against the protein solution (10 µL injections) at 20 °C with 300 s intervals. CA titrations against the buffer and buffer titrations against the protein solution were performed as controls. The isotherm curves, after subtracting the controls, were analyzed using the Microcal Origin software provided with the equipment, and the data were fitted to One Set of Sites model.

Protein crystallization and X-ray data collection

Freshly prepared CTDsamples at 8 mg/mL were subjected to hanging-drop vapor diffusion crystallization trials performed in 24-well VDX plates at 18^oC using Hampton Crystal Screen HT™ solutions. After optimizing the crystallization conditions, CTD crystals were obtained within three days in 100 mM Tris -HCl (pH 8.3), 30% PEG 4000, 0.2 M sodium acetate, with 2 µL drops (1 µL protein /1 µL reservoir solution) and 300 µL reservoir solution. Protein-ligand crystals were grown within three weeks under the same crystallization condition with a reservoir solution containing 4 mM CA. Protein crystals were cryoprotected by rapid soaking in reservoir solution containing 25% glycerol and flash-cooled in liquid nitrogen. X-ray diffraction data were collected under cryogenic conditions (100 K) at 1.327 Å and 0.977 Å wavelength at the Manacá beamline (macromolecular micro and nano crystallography)⁵² of Sirius, the Brazilian synchrotron light source (LNLS, Campinas, Brazil), using a PILATUS 2M detector placed 145 mm from the crystal. The X-ray data were collected using a fine ϕ-slicing strategy, rotated through 360° with a 0.1° oscillation range per frame.

X-ray data processing and structural determination

X-ray diffraction data were automatically processed with XDS⁵³ using the Manacá Automatic Processing Pipeline (ManacáAutoProc) and analyzed and scaled using Pointless, Matthews and Scala from CCP4 package^36,54,55. The phases of the datasets were determined by molecular replacement with Molrep⁵⁶ and Phaser⁵⁷ using the SARS-CoV-2 CTD crystal structure (PDB code: 7C22) as the search model. The atomic structures were refined using REFMAC5⁵⁸, ligands had their geometry restraint information generated for refinement by eLBOW⁵⁹ and then modelled using COOT⁶⁰. All figures were generated using PyMOL.

The volume of the CA binding site was estimated using parKVFinder software⁶¹ using the box adjustment mode around the CA molecule with the following parameters: probe in of 1.4 Å, probe out of 12 Å and removal distance of 0.5 Å.

NMR experiments

All NMR spectra were obtained using an Agilent DD2 500 MHz spectrometer or Varian Inova 600 MHz spectrometer both equipped with a 5 mm triple-resonance probe and a Z pulse-field gradient unit at 298 K. The STD experiments were performed with 400 μM CA and 4 μM N protein samples dissolved in 80 mM sodium phosphate buffer, pH 7.4, prepared with deuterated water. The 1D ¹H-STD spectra were obtained by subtracting the saturated spectra (on- resonance) from the reference spectra (off-resonance), which was automatically performed by phase cycling using the dpfgse satzfer pulse sequence implemented in the VNMRJ software (Agilent). The spectra were acquired using 2048 scans with a selective irradiation frequency of the protein at -0.5 ppm (on-resonance) and 30 ppm (off-resonance). Forty G-shaped pulses of 50 ms separated by 1 ms delays were applied to the samples. The total length of the saturation train was 2.5 s. A T₂ filter was applied to eliminate all protein background. The off-resonance spectra were used as reference spectra and were acquired with 1024 scans keeping all other parameters equal to the 1D ¹H-STD-NMR spectra. For the group epitope mapping analysis, the STD enhancements (A_STD) were determined by the integrals of individual protons of the ligands in the 1D ¹H-STD-NMR spectrum (I_STD) divided by the integral of the same signals at the reference spectrum (I₀) and multiplied by the excess ratio of ligand to protein concentration ([L]/[P]) according to equation 1.

In vitro anti-SARS-CoV-2 activity

Vero CCL81 cells (African green monkey kidney cell line - BCRJ, # 0245) were cultivated in DMEM medium supplemented with 10% fetal bovine serum (FBS), 1% L-glutamine and 1% penicillin/streptomycin. Calu-3 cells (Human lung cell line – ATCC, # HTB-55™) were cultivated in DMEM/F12 (1:1, v/v) medium supplemented with 20% FBS, 1% L-glutamine and 1% penicillin/streptomycin. Both cell lines were grown at 37°C with 5% CO₂.

Antiviral assays were performed with the HIAE-02 SARS-CoV-2/SP02/human/2020/BRA strain (GenBank accession MT126808.1) kindly provided by Prof. Edison Luiz Durigon (USP-SP, Brazil). Virus stocks were propagated in Vero CCL81 cells in 75 cm² flasks. After 30-36 h of growth, culture supernatants were centrifuged to remove cell debris and stored at -80°C. All assays involving infectious virus were performed in the BSL-3 unit of the Emerging Viruses Laboratory (LEVE) at the State University of Campinas, Brazil.

Cell viability in Vero CCL-81 and Calu-3 after CA treatment was measured by the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] (Sigma–Aldrich, USA) method. Briefly, CA diluted in 10% DMEM at 25 and 100 µM final concentration was added to the confluent monolayer of cells grown in 24-well plates. After 48 h growth, the medium was replaced by fresh DMEM containing MTT (200 µg/mL) and cells were incubated for 3 h. DMSO was used to solubilize the formazan crystals and cell viability was measured by OD at 492 nm. The results were expressed according to the equation (T/C) × 100%, where T and C represented the mean optical density of treated and control, respectively. DMSO at 0.2% (v/v) in DMEM medium was used as the vehicle control.

To evaluate the activity of CA on SARS-CoV-2 replication, Calu-3 and Vero CCL-81 cells were seeded into 24-well plates at 2x10⁵ and 2.5x10⁵ cells per well, respectively. The antiviral activity was determined at multiplicity of infection (MOI) of 1. The confluent monolayer of cells was incubated with the virus for 1 h, after which the culture medium was replaced by fresh medium containing CA at 25 and 100 µM final concentration. Culture supernatants were harvested 48 h after virus inoculation and viral load was determined by plaque assay.

Plaque assay

Vero cells were seeded into 24-well plates and incubated for 1h with the supernatants from the antiviral assays, serially diluted to 10^-6. After virus incubation, cells were overlaid with semi-solid medium (1% w/v carboxymethylcellulose in DMEM supplemented with 5% FBS) and incubated for 4 days. After removal of semi-solid medium, cells were fixed with paraformaldehyde 4% and plaques were visualized after 1% methylene blue staining. Viral lysis plaques were counted and the results were expressed as viral plaque forming units (PFU) per mL of sample.

Acknowledgements

This work was supported by the Brazilian Funding Authority for Studies and Projects (FINEP, grant number 01.20.0003.00), the Brazilian Ministry of Sciences, Technology and Innovation (MCTI) and the National Fund for Scientific and Technological Development (FNDCT). This research used facilities of the Brazilian Synchrotron Light Laboratory (LNLS) and the Brazilian Biosciences National Laboratory (LNBio), which are part of the Brazilian Center for Research in Energy and Materials (CNPEM), a private non-profit organization under the supervision of the MCTI. The Manaca beamline (Sirius, LNLS, proposal 20200057) and the LNBio-CNPEM facilities LPP, LEC, ROBOLAB, NMR, LGC and LBE staff is acknowledged for the assistance during the experiments. We also wish to acknowledge all the scientists and staff participating on the CNPEM’s COVID-19 task-force.

Author contributions

Conceived and designed the analysis: GFM, ATC, REM, MLS, KGF, CEB, ACMF, DBBT. Collected the data: EHSB, FAHB, CCCT, ASS, JFS, AN, JCS, JNF, MLS. Contributed data or analysis tools: HVRF, MGC, ACMZ, AFZN, JLPM, MCB, SAR, PSLO. Performed the analysis: GFM, EHSB, FAHB, JFS, AN, HRF, SAR, PSLO, MB, MLS, CEB, ACMF, DBBT.

Protein preparation: FAHB, CCCT, ASS, JCS, MCB, CEB. Biophysical assays: FAHB, SAR, MLS, CEB, ACMF. Protein crystallography, model interpretation and comparison: EHSB, CCCT, ACMZ, AFZN, DBBT, HVRF, PSLO. Virology: JFS, AN, JLPM, REM. HTS assay design and implementation: GFM, FAHB, JNF, MGC, ATC, CEB, ACMF, DBBT. SAR analysis: MB, DBBT. Project management and funding: KGF, DBBT. Designed the paper: ACMF, DBBT. Wrote the paper: GFM, EHSB, FAHB, CEB, ACMF, DBBT with contributions and revisions from all authors.

Data availability

The crystal structures generated during the current study are available in the Protein Data Bank (PDB) under the accession numbers 7UXX (apo CTD) and 7UXZ (CTD-CA), and through the links: https://www.rcsb.org/structure/7UXX and https://www.rcsb.org/structure/7UXZ, respectively.

Nicola, M. et al. The socio-economic implications of the coronavirus pandemic (COVID-19): A review. International Journal of Surgery 78, 185–193 (2020).
Das, S., Wingender, P., Barrett, P., Pugacheva, E. & Magistretti, G. After-Effects of the COVID-19 Pandemic: Prospects for Medium-Term Economic Damage. IMF Working Papers 2021, 1 (2021).
Beigel, J. H. et al. Remdesivir for the Treatment of Covid-19 — Final Report. New England Journal of Medicine 383, 1813–1826 (2020).
Jayk Bernal, A. et al. Molnupiravir for Oral Treatment of Covid-19 in Nonhospitalized Patients. New England Journal of Medicine 386, 509–520 (2022).
Kozlov, M. Merck’s COVID pill loses its lustre: what that means for the pandemic. Nature (2021) doi:10.1038/d41586-021-03667-0.
Owen, D. R. et al. An oral SARS-CoV-2 M pro inhibitor clinical candidate for the treatment of COVID-19. Science (1979) 374, 1586–1593 (2021).
Heskin, J. et al. Caution required with use of ritonavir-boosted PF-07321332 in COVID-19 management. The Lancet 399, 21–22 (2022).
Shah, M. & Woo, H. G. Omicron: A Heavily Mutated SARS-CoV-2 Variant Exhibits Stronger Binding to ACE2 and Potently Escapes Approved COVID-19 Therapeutic Antibodies. Frontiers in Immunology 12, (2022).
Coelho, C., Gallo, G., Campos, C. B., Hardy, L. & Würtele, M. Biochemical screening for SARS-CoV-2 main protease inhibitors. PLOS ONE 15, e0240079 (2020).
Zhu, W. et al. Identification of SARS-CoV-2 3CL Protease Inhibitors by a Quantitative High-Throughput Screening. ACS Pharmacology & Translational Science 3, 1008–1016 (2020).
Lim, C. T. et al. Identifying SARS-CoV-2 antiviral compounds by screening for small molecule inhibitors of Nsp3 papain-like protease. Biochemical Journal 478, 2517–2531 (2021).
Xu, T., Zheng, W. & Huang, R. High‐throughput screening assays for SARS‐CoV‐2 drug development: Current status and future directions. Drug Discovery Today 26, 2439–2444 (2021).
Eydoux, C. et al. A fluorescence-based high throughput-screening assay for the SARS-CoV RNA synthesis complex. Journal of Virological Methods 288, (2021).
Wang, Y. et al. Discovery of SARS-CoV-2-E channel inhibitors as antiviral candidates. Acta Pharmacologica Sinica (2021) doi:10.1038/s41401-021-00732-2.
Smith, E. et al. High-Throughput Screening for Drugs That Inhibit Papain-Like Protease in SARS-CoV-2. SLAS DISCOVERY: Advancing the Science of Drug Discovery 25, 1152–1161 (2020).
Zhang, Z.-R. et al. A cell-based large-scale screening of natural compounds for inhibitors of SARS-CoV-2. Signal Transduction and Targeted Therapy 5, 218 (2020).
Rothan, H. A. & Teoh, T. C. Cell-Based High-Throughput Screening Protocol for Discovering Antiviral Inhibitors Against SARS-COV-2 Main Protease (3CLpro). Molecular Biotechnology 63, 240–248 (2021).
Zhao, T. et al. Warnings regarding the potential coronavirus disease 2019 (COVID-19) transmission risk: Vaccination is not enough. Infection Control & Hospital Epidemiology 1–1 (2021) doi:10.1017/ice.2021.63.
Gorshkov, K. et al. SARS-CoV-2 Nucleocapsid Protein TR-FRET Assay Amenable to High Throughput Screening. ACS Pharmacology and Translational Science 5, (2022).
McBride, R., van Zyl, M. & Fielding, B. C. The coronavirus nucleocapsid is a multifunctional protein. Viruses vol. 6 (2014).
Peng, Y. et al. Structures of the SARS‐CoV‐2 nucleocapsid and their perspectives for drug design. The EMBO Journal 39, (2020).
Dinesh, D. C. et al. Structural basis of RNA recognition by the SARS-CoV-2 nucleocapsid phosphoprotein. PLoS Pathogens 16, (2020).
Kang, S. et al. Crystal structure of SARS-CoV-2 nucleocapsid protein RNA binding domain reveals potential unique drug targeting sites. Acta Pharmaceutica Sinica B 10, (2020).
Grossoehme, N. E. et al. Coronavirus N Protein N-Terminal Domain (NTD) Specifically Binds the Transcriptional Regulatory Sequence (TRS) and Melts TRS-cTRS RNA Duplexes. Journal of Molecular Biology 394, 544–557 (2009).
Keane, S. C., Lius, P., Leibowitzs, J. L. & Giedroc, D. P. Functional Transcriptional Regulatory Sequence (TRS) RNA binding and helix destabilizing determinants of Murine Hepatitis Virus (MHV) Nucleocapsid (N) protein. Journal of Biological Chemistry 287, (2012).
Carlson, C. R. et al. Phosphoregulation of Phase Separation by the SARS-CoV-2 N Protein Suggests a Biophysical Basis for its Dual Functions. Molecular Cell 80, 1092-1103.e4 (2020).
Savastano, A., Ibáñez de Opakua, A., Rankovic, M. & Zweckstetter, M. Nucleocapsid protein of SARS-CoV-2 phase separates into RNA-rich polymerase-containing condensates. Nature Communications 11, (2020).
Perdikari, T. M. et al. SARS‐CoV‐2 nucleocapsid protein phase‐separates with RNA and with human hnRNPs. The EMBO Journal 39, (2020).
Lu, S. et al. The SARS-CoV-2 nucleocapsid phosphoprotein forms mutually exclusive condensates with RNA and the membrane-associated M protein. Nature Communications 12, (2021).
Hsieh, P.-K. et al. Assembly of Severe Acute Respiratory Syndrome Coronavirus RNA Packaging Signal into Virus-Like Particles Is Nucleocapsid Dependent. Journal of Virology 79, 13848–13855 (2005).
Hsin, W.-C. et al. Nucleocapsid protein-dependent assembly of the RNA packaging signal of Middle East respiratory syndrome coronavirus. Journal of Biomedical Science 25, 47 (2018).
Adem, Ş. et al. Caffeic acid derivatives (CAFDs) as inhibitors of SARS-CoV-2: CAFDs-based functional foods as a potential alternative approach to combat COVID-19. Phytomedicine 85, 153310 (2021).
King, P. J. & Robinson, W. E. Resistance to the Anti-Human Immunodeficiency Virus Type 1 Compound l -Chicoric Acid Results from a Single Mutation at Amino Acid 140 of Integrase. Journal of Virology 72, 8420–8424 (1998).
Reinke, R. A. et al. l-Chicoric acid inhibits human immunodeficiency virus type 1 integration in vivo and is a noncompetitive but reversible inhibitor of HIV-1 integrase in vitro. Virology 326, 203–219 (2004).
Zhang, H.-L. et al. Evaluation of Hepatocyteprotective and Anti-hepatitis B Virus Properties of Cichoric Acid from Cichorium intybus Leaves in Cell Culture. Biological and Pharmaceutical Bulletin 37, 1214–1220 (2014).
Matthews, B. W. Solvent content of protein crystals. J Mol Biol 33, 491–497 (1968).
Zhou, R., Zeng, R., von Brunn, A. & Lei, J. Structural characterization of the C-terminal domain of SARS-CoV-2 nucleocapsid protein. Molecular Biomedicine 1, 2 (2020).
Yang, M. et al. Structural Insight Into the SARS-CoV-2 Nucleocapsid Protein C-Terminal Domain Reveals a Novel Recognition Mechanism for Viral Transcriptional Regulatory Sequences. Frontiers in Chemistry 8, (2021).
Chang, C.-K. et al. Multiple Nucleic Acid Binding Sites and Intrinsic Disorder of Severe Acute Respiratory Syndrome Coronavirus Nucleocapsid Protein: Implications for Ribonucleocapsid Protein Packaging. Journal of Virology 83, 2255–2264 (2009).
Ribeiro-Filho, H. V. et al. Structural dynamics of SARS-CoV-2 nucleocapsid protein induced by RNA binding. PLOS Computational Biology 18, e1010121 (2022).
Douangamath, A. et al. Crystallographic and electrophilic fragment screening of the SARS-CoV-2 main protease. Nature Communications 11, (2020).
Salgado-Benvindo, C. et al. Suramin inhibits SARS-CoV-2 infection in cell culture by interfering with early steps of the replication cycle. Antimicrobial Agents and Chemotherapy 64, (2020).
Landete, J. M. Ellagitannins, ellagic acid and their derived metabolites: A review about source, metabolism, functions and health. Food Research International 44, 1150–1160 (2011).
Li, P. et al. Identification of Chebulinic Acid and Chebulagic Acid as Novel Influenza Viral Neuraminidase Inhibitors. Frontiers in Microbiology 11, (2020).
Kesharwani, A. et al. Anti-HSV-2 activity of Terminalia chebula Retz extract and its constituents, chebulagic and chebulinic acids. BMC Complementary and Alternative Medicine 17, (2017).
Lin, L. T. et al. Broad-spectrum antiviral activity of chebulagic acid and punicalagin against viruses that use glycosaminoglycans for entry. BMC Microbiology 13, (2013).
Roh, C. A facile inhibitor screening of SARS coronavirus N protein using nanoparticle-based RNA oligonucleotide. International Journal of Nanomedicine 2173 (2012) doi:10.2147/IJN.S31379.
Lin, S.-Y. et al. Structural Basis for the Identification of the N-Terminal Domain of Coronavirus Nucleocapsid Protein as an Antiviral Target. Journal of Medicinal Chemistry 57, 2247–2257 (2014).
Matthias, A. et al. Permeability studies of alkylamides and caffeic acid conjugates from echinacea using a Caco-2 cell monolayer model. Journal of Clinical Pharmacy and Therapeutics 29, (2004).
Charvat, T. T., Lee, D. J., Robinson, W. E. & Chamberlin, A. R. Design, synthesis, and biological evaluation of chicoric acid analogs as inhibitors of HIV-1 integrase. Bioorganic and Medicinal Chemistry 14, (2006).
Peroni, L. A. et al. Serological Testing for COVID-19, Immunological Surveillance, and Exploration of Protective Antibodies. Frontiers in Immunology 12, (2021).
Nascimento, A. et al. Launch of the Manacá Beamline at Sirius: First Protein Crystallography Structures and New Opportunities for Pharmaceutical Development Using Synchrotrons. Synchrotron Radiation News 34, 3–10 (2021).
Kabsch, W. XDS. Acta Crystallographica Section D Biological Crystallography 66, 125–132 (2010).
Kantardjieff, K. A. & Rupp, B. Matthews coefficient probabilities: Improved estimates for unit cell contents of proteins, DNA, and protein-nucleic acid complex crystals. Protein Science 12, 1865–1871 (2003).
Evans, P. R. An introduction to data reduction: Space-group determination, scaling and intensity statistics. Acta Crystallographica Section D: Biological Crystallography 67, (2011).
Vagin, A. & Teplyakov, A. Molecular replacement with MOLREP. Acta Crystallogr D Biol Crystallogr 66, 22–5 (2010).
McCoy, A. J. et al. Phaser crystallographic software. Journal of Applied Crystallography 40, 658–674 (2007).
Murshudov, G. N., Vagin, A. A. & Dodson, E. J. Refinement of Macromolecular Structures by the Maximum-Likelihood Method. Acta Crystallographica Section D Biological Crystallography 53, 240–255 (1997).
Moriarty, N. W., Grosse-Kunstleve, R. W. & Adams, P. D. electronic Ligand Builder and Optimization Workbench ( eLBOW ): a tool for ligand coordinate and restraint generation. Acta Crystallographica Section D Biological Crystallography 65, 1074–1080 (2009).
Emsley, P. & Cowtan, K. Coot : model-building tools for molecular graphics. Acta Crystallographica Section D Biological Crystallography 60, 2126–2132 (2004).
Guerra, J. V. da S. et al. ParKVFinder: A thread-level parallel approach in biomolecular cavity detection. SoftwareX 12, (2020).

No competing interests reported.

20220516SIMercaldiBezerraBatistaetalchicoricacidasNproteinligand.docx

Discovery and structural characterization of chicoric acid as a SARS-CoV-2 nucleocapsid protein ligand and RNA binding disruptor

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1