key: cord-266710-3wdy16tw
authors: Fintelman-Rodrigues, Natalia; Sacramento, Carolina Q.; Ribeiro Lima, Carlyle; Souza da Silva, Franklin; Ferreira, André C.; Mattos, Mayara; de Freitas, Caroline S.; Cardoso Soares, Vinicius; da Silva Gomes Dias, Suelen; Temerozo, Jairo R.; Miranda, Milene D.; Matos, Aline R.; Bozza, Fernando A.; Carels, Nicolas; Alves, Carlos Roberto; Siqueira, Marilda M.; Bozza, Patrícia T.; Souza, Thiago Moreno L.
title: Atazanavir, Alone or in Combination with Ritonavir, Inhibits SARS-CoV-2 Replication and Proinflammatory Cytokine Production
date: 2020-09-21
journal: Antimicrob Agents Chemother
DOI: 10.1128/aac.00825-20
sha: 
doc_id: 266710
cord_uid: 3wdy16tw

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is already responsible for far more deaths than previous pathogenic coronaviruses (CoVs) from 2002 and 2012. The identification of clinically approved drugs to be repurposed to combat 2019 CoV disease (COVID-19) would allow the rapid implementation of potentially life-saving procedures. The major protease (Mpro) of SARS-CoV-2 is considered a promising target, based on previous results from related CoVs with lopinavir (LPV), an HIV protease inhibitor. However, limited evidence exists for other clinically approved antiretroviral protease inhibitors. Extensive use of atazanavir (ATV) as antiretroviral and previous evidence suggesting its bioavailability within the respiratory tract prompted us to study this molecule against SARS-CoV-2. Our results show that ATV docks in the active site of SARS-CoV-2 Mpro with greater strength than LPV, blocking Mpro activity. We confirmed that ATV inhibits SARS-CoV-2 replication, alone or in combination with ritonavir (RTV) in Vero cells and a human pulmonary epithelial cell line. ATV/RTV also impaired virus-induced enhancement of interleukin 6 (IL-6) and tumor necrosis factor alpha (TNF-α) levels. Together, our data strongly suggest that ATV and ATV/RTV should be considered among the candidate repurposed drugs undergoing clinical trials in the fight against COVID-19.

( Fig. S1 and S2 ). ATV bound more spontaneously because of its hydrogen bonds with Mpro, whereas LPV depends on hydrophobic interactions (Fig. S2) .

Molecular dynamics analysis revealed that the root mean square deviation (RMSD) for the SARS-CoV-2 Mpro backbone presented different conformations in complex with ATV or LPV (see Fig. S3 ). LPV was initially at a 3.8-Å distance from the catalytic residue Cys145 (see Fig. S4A and S5A). After conformational changes, LPV was 7.17 Å distant from the active site ( Fig. 1A and C) , likely limiting its antiviral activity. Another critical residue, His41, was satisfactorily at a distance of 2.89 Å from bound LPV ( Fig. 1A and C). ATV interacted with neither His41 nor Cys145 at initial analysis ( Fig. S4B and S5B) . Nevertheless, ATV's position remained stable within the active site independently of conformational changes ( Fig. 1B and D) . The steric occupation of the cleft in the enzymatic active site by ATV, which blocks the residues of the catalytic amino acids, can be explained by its stronger interactions with Mpro than LPV (Tables S1 to S3). ATV inhibits SARS-CoV-2 Mpro enzymatic activity. Next, we evaluated whether ATV could inhibit SARS-CoV-2 Mpro activity by partially purifying the enzyme in cellular fractions obtained from SARS-CoV-2-infected cells and performing zymographic profiles. To ensure that the proteinase profiles were not dependent on cellular enzymes, similar fractions of mock-infected cells were also prepared. The results from cysteine proteinase zymographic profiles in gelatinolytic gels reveled a cellular-related band of approximately 70 kDa under both conditions (Fig. 2 , Nil lanes). This activity was blocked by the drug E-64, an epoxide that acts as an irreversible inhibitor of cysteine proteases (Fig. 2 , E-64 lanes). In the infected cells, a region of activity was observed between 31 and 38 kDa that was not present in the mock fraction (Fig. 2 ). This zone of molecular weight is consistent with the expected size of SARS-CoV-2 Mpro. The enzyme activity was inhibited by exposure of the gels to 10 M ATV, without affecting cellular cysteine proteinase (Fig. 2 , ATV lanes). As a control, the activity of SARS-CoV-2 Mpro in fractions from infected cells was evaluated by treatment with RTV, which inhibited activity in the molecular range of 31 to 38 kDa without a change in the 70-kDa region (Fig. 2 , RTV lanes). These data are consistent with predictions from the molecular modeling and dynamics that ATV targets SARS-CoV-2 Mpro.

SARS-CoV-2 is susceptible to ATV and ATV/RTV in different cell types. We extended our investigation to evaluate the susceptibility of SARS-CoV-2 to ATV in different cellular systems. Vero cells are a well-known model to produce high virus titers. ATV alone or in combination with RTV decreased infectious virus production and RNA levels in this cell line ( Fig. 3A and B, respectively). ATV/RTV was more potent than ATV, with 50% effective concentration (EC 50 ) values of 0.5 Ϯ 0.08 M and 2.0 Ϯ 0.12 M, respectively (Fig. 3B) . Positive controls CQ, LPV/RTV, and remdesivir displayed potencies of 1.0 Ϯ 0.07 M, 5.3 Ϯ 0.5 M, and 0.5 Ϯ 0.08 M, respectively (Fig. 3B ). Our positive controls are consistent with results in the literature (21) , validating our analysis. The ATV/RTV, ATV, CQ, LPV/RTV, and remdesivir values for 50% cytotoxic concentration (CC 50 ) were 280 Ϯ 3 M, 312 Ϯ 8 M, 259 Ϯ 5 M, 91 Ϯ 3 M, and 512 Ϯ 30 M, respectively. Our results indicate that the selectivity index (SI; which represents the ratio between the CC 50 and EC 50 values) for ATV/RTV, ATV, CQ, LPV/RTV, and remdesivir were 560, 156, 259, 18, and 1,020, respectively, which shows that ATV/RTV and ATV have therapeutic potential above that of CQ and LPV/RTV, compounds that advanced toward clinical trials early after the pandemic outbreak. Since the results regarding the pharmacologic activity of ATV and ATV/RTV against SARS-CoV-2 replication in Vero cells were promising, we next investigated whether the proposed drug therapies could inhibit virus replication in a human epithelial pulmonary cell line (A549). ATV alone showed a nearly 10-fold increase in potency for inhibiting SARS-CoV-2 replication in A549 cells (Fig. 3C ) compared to that in Vero cells (Fig. 3B ). ATV/RTV and CQ were similarly potent in inhibiting virus replication in both cell types ( Fig. 3B and C) . Drugs repurposed in this study, ATV and ATV/RTV, were more potent than positive controls in inhibiting SARS-CoV-2 replication in A549 cells. Potencies for ATV/RTV, ATV, CQ, LPV/RTV, and remdesivir were 0.60 Ϯ 0.05 M, 0.22 Ϯ 0.02 M, 0.89 Ϯ 0.02 M, 0.9 Ϯ 0.5 M, and 0.6 Ϯ 0.02 M, respectively. In vitro results confirmed the rationale that SARS-CoV-2 would be susceptible to ATV, including in cells derived from the respiratory tract.

ATV and ATV/RTV prevent cell death and proinflammatory cytokine production in SARS-CoV-2-infected monocytes. Severe COVID-19 has been associated with levels of lactate dehydrogenase (LDH), interleukin 6 (IL-6), and leukopenia (22) . Viral infection in the respiratory tract often triggers the migration of blood monocytes to FIG 3 The antiviral activity of ATV and ATV/RTV against SARS-CoV-2. Vero (A and B) or A549 (C) cells were infected with SARS-CoV-2 at an MOI of 0.01 and exposed to the indicated concentrations of atazanavir (ATV), atazanavir/ritonavir (ATV/RTV; 3:1), chloroquine (CQ), remdesivir (RDV), or lopinavir/ritonavir (LPV/RTV; 4:1). After 2 days, viral replication in the culture supernatant was measured by TCID 50 /ml (A) or RT-PCR (B and C). The data represent means Ϯ standard errors of the means (SEMs) from three independent experiments.

Antimicrobial Agents and Chemotherapy orchestrate the transition from innate to adaptive immune responses (23) . For these reasons, ATV and ATV/RTV were tested at suboptimal (1 M) and optimal (10 M) doses, with respect to their in vitro pharmacological parameters against SARS-CoV-2. ATV/RTV, CQ, and remdesivir were similarly efficient at reducing the amount viral genome equivalent in the human monocytes (Fig. 4A ). Virus infection increased cellular mortality by 75%, which was prevented by ATV, ATV/RTV, and remdesivir (Fig. 4B ). LPV/RTV was inefficient at reducing viral RNA levels and cell death ( Fig. 4A and B) . Moreover, we observed that infections by SARS-CoV-2 triggered the expected increase in the IL-6 levels in the culture supernatant, which ranged from 20-to 60-fold depending on the cell donor (Fig. 4C) . The virus-induced enhancement of IL-6 levels was significantly prevented by treatment with ATV, ATV/RTV, and CQ (Fig. 4C ). Another biomarker of uncontrolled proinflammatory cytokine response, tumor necrosis factor alpha (TNF-␣), was upregulated 40-fold during virus infection (Fig. 4D) . ATV, ATV/RTV, and remdesivir (10 M) significantly prevented the induction of TNF-␣ release (Fig. 4D ). Altogether, our results confirm that ATV and ATV/RTV should not be ignored as an additional therapeutic option against COVID-19. 

In these 2 decades of the 21st century, the human vulnerability to emerging viral diseases has been notable (24) . The emergence of infectious disease highlights the undeniable fact that existing countermeasures are inefficient at preventing virus spillover and disease outbreaks. Preclinical data on the susceptibility of an emerging virus to clinically approved drugs can allow for the rapid mobilization of resources toward clinical trials (8) . This approach proved feasible for combating the Zika, yellow fever, and chikungunya outbreaks experienced in Brazil over the past 5 years, when our group demonstrated that sofosbuvir, a blockbuster drug against hepatitis C, could represent a compassionate countermeasure against these diseases (25) (26) (27) (28) (29) .

Currently, the rate of SARS-CoV-2 dissemination has become one of the most rapidly evolving pandemics known in modern times, with the number of cases and deaths doubling every week, and the peak of the pandemic has yet to arrive in some territories (6) . The existence of several ongoing clinical trials against COVID-19 reinforces the suggestion that drug repurposing represents the fastest approach to identify therapies to emerging infectious disease (8) .

Among therapies initially included in the Solidarity trial, the most interesting results come from remdesivir, whereas CQ and LPV/RTV showed limited clinical benefit (9) . LPV/RTV reduced mortality in critically ill patients by 5% (11) . On the other hand, this therapy showed no clinical benefit in a large clinical trial (30) . Although the combination therapy with protease (LPV/RTV) and RNA polymerase (ribavirin) inhibitors and immunomodulators (IFN-␤) reduced the viral loads in COVID-19 patients (31), these drugs seem to be impractical for early treatment because of IFN's price-safety profile. The history of antiretroviral research teaches us that combinations are necessary. Positive laboratory and clinical results with RNA polymerase inhibitors, such as remdesivir, ribavirin, and favipiravir (21, 31, 32) , against SARS-CoV-2 suggest they could be more effective if combined with active protease inhibitors. We highlight ATV and ATV/RTV because our assay readout to quantify infectious virus particles reveals (i) a good profile of antiviral activity, (ii) higher potencies in respiratory cells, and (iii) the ability to reduce levels of proinflammation mediator in monocytes.

We believe that early repurposing of LPV/RTV for COVID-19 was based on previous evidence during the SARS-CoV outbreak in 2002 and again for MERS-CoV (33) . Information on the susceptibility of SARS-CoV-2 to other antiviral protease inhibitors approved since 2003, such as ATV, has been scarce. Since this year, ATV has become a wider prescribed drug among HIV-infected individuals than LPV, including for critically ill patients (17) . ATV shows a safer profile than LPV in both short-and long-term therapeutic regimens (16, 34) . ATV has a documented bioavailability to reach the respiratory tract (19, 35) , which led to its proposed use against pulmonary fibrosis (20) . Under our experimental conditions, ATV was superior to LPV/RTV, which may motivate further clinical trials.

The potency of LPV/RTV against SARS-CoV-2 was lower than those of ATV and ATV/RTV. Nevertheless, remdesivir was more potent than ATV or ATV/RTV. The improved potency of ATV, in comparison to that of LPV, may be at least in part due to its multiple hydrogen bond-driven interactions within the Mpro active site. Other investigators have also recognized a wider range of interactions of ATV and Mpro compared to those with LPV (18, 36) , although none provided functional evidence through phenotypic assays as presented here. Neither ATV nor LPV displayed any interactions with the catalytic dyad of Cys145 and His41 at the start of the molecular dynamics simulations. However, important interactions were observed at the end, such as LPV-His41 and ATV-Glu166. Glu166 is one of the residues that promotes the opening of Mpro for its substrate to interact with the active site (37, 38) .

Highly pathogenic respiratory viruses, such as influenza A virus, have been associated with a cytokine storm, described as an uncontrolled proinflammatory cytokine response (39, 40) . Cytokine storms also seem to be highly relevant for pathogenic human CoVs (41) . Contemporary investigations on SARS-CoV-2 strongly suggest the involvement of cytokine storms in disease severity (22) . COVID-19 mortality is associated with enhanced IL-6 levels and consistent with cell death, as measured by LDH release (22) . We showed that ATV and ATV/RTV decreased IL-6 release in SARS-CoV-2infected human primary monocytes. Moreover, we also included in our analysis TNF-␣, another hallmark of inflammation during respiratory virus infections (22, 58) . Our results reveled that cellular mortality and cytokine storm-associated mediators were reduced after treatment with the repurposed antiretroviral drugs used in this study.

As the SARS-CoV-2 pandemic goes on and the Solidarity trials fail to demonstrate the benefit of LPV/RTV, preclinical data or clinically approved protease inhibitors, such as ATV-ATV/RTV, need to be catalogued. The higher potency of ATV-ATV/RTV over that of LPV/RTV is the contribution of our study to highlight a new option among clinically approved drugs that should be considered in ongoing clinical trials for an effective treatment for COVID-19.

Reagents. The antivirals ATV, ATV/RTV, and CQ were received as donations from Instituto de Tecnologia de Fármacos (Farmanguinhos, Fiocruz). ATV/RTV was prepared daily in the proportion of 3:1 as pharmaceutical pills composed of 300 mg ATV and 100 mg RTV. Remdesivir and LPV/RTV (4:1 ratio) were purchased from Selleckchem (Houston, TX). Enzyme-linked immunosorbent assay (ELISA) kits were purchased from R&D Bioscience. All small molecule inhibitors were dissolved in 100% dimethyl sulfoxide (DMSO) and subsequently diluted at least 10 4 -fold in culture or reaction medium before each assay. The final DMSO concentrations showed no cytotoxicity. The materials for cell culture were purchased from Thermo Scientific Life Sciences (Grand Island, NY), unless otherwise mentioned. Cells and virus. African green monkey kidney (Vero, subtype E6) and A549 (human lung epithelial cells) cells were cultured in high-glucose Dulbecco's modified Eagle medium (DMEM) with 10% fetal bovine serum (FBS; HyClone, Logan, UT) and 100 U/ml penicillin and 100 g/ml streptomycin (pen/strep; Thermo Fisher) at 37°C in a humidified atmosphere with 5% CO 2 .

Human primary monocytes were obtained after 3 h of plastic adherence of peripheral blood mononuclear cells (PBMCs). PBMCs were isolated from healthy donors by density gradient centrifugation (Ficoll-Paque; GE Healthcare). PBMCs (2.0 ϫ 10 6 cells) were plated on 48-well plates (Nalge Nunc) in RPMI 1640 without serum for 2 to 4 h. Nonadherent cells were removed, and the remaining monocytes were maintained in DMEM with 5% human serum (HS; Millipore) and penicillin/streptomycin. The purity of human monocytes was Ͼ95% as determined by flow cytometric analysis (FACScan; Becton, Dickinson) using anti-CD3 (BD Biosciences) and anti-CD16 (Southern Biotech) monoclonal antibodies.

SARS-CoV-2 was prepared in Vero E6 cells from an isolate contained on a nasopharyngeal swab obtained from a confirmed case in Rio de Janeiro, Brazil. Viral experiments were performed after a single passage in a cell culture in a 150-cm 2 flask with DMEM plus 2% FBS. Observations for cytopathic effects were performed daily, which peaked 4 to 5 days after infection. All procedures related to virus culture were handled in a biosafety level 3 (BSL3) multiuser facility according to WHO guidelines. Virus titers were determined as the tissue culture infectious dose at 50% (TCID 50 /ml). Virus stocks were kept in Ϫ80°C ultralow freezers. The virus strain was sequenced to confirm the virus identity.

Cytotoxicity assay. Monolayers of 1.5 ϫ 10 4 Vero cells in 96-well plates were treated for 3 days with various concentrations (semilog dilutions from 600 to 10 M) of ATV, ATV/RTV, or CQ. Then, 5 mg/ml 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) in DMEM was added to the cells in the presence of 0.01% of N-methyl dibenzopyrazine methyl sulfate (PMS). After incubating for 4 h at 37°C, the plates were measured in a spectrophotometer at 492 nm and 620 nm. The 50% cytotoxic concentration (CC 50 ) was calculated by a nonlinear regression analysis of the dose-response curves.

Yield reduction assay. Cells were infected at a multiplicity of infection (MOI) of 0.01. Vero or A549 cells were infected at densities of 5 ϫ 10 5 cells/well. Human primary monocytes were infected at a density of 2 ϫ 10 5 to 8 ϫ 10 5 cells/well, depending on the endogenous characteristic of the cell donor. Infections were performed in 48-well plates for 2 h at 37°C. The cells were washed, and various concentrations of compounds were added to DMEM with 2% FBS. After 48 h, virus in the supernatants was quantified by real-time reverse transcription-PCR (RT-PCR) and/or by TCID 50 /ml. A variable slope nonlinear regression analysis of the dose-response curves was performed to calculate the concentration at which each drug inhibited virus production by 50% (EC 50 ).

Virus titration. Monolayers of Vero cells (2 ϫ 10 4 cell/well) in 96-well plates were infected with a log-based dilution of supernatants containing SARS-CoV-2 for 1 h at 37°C. Cells were washed, fresh medium was added with 2% FBS, and 3 to 5 days postinfection, the cytopathic effect was scored in at Fintelman-Rodrigues et al.

Antimicrobial Agents and Chemotherapy least 10 replicates per dilution by independent readers. The reader was blind with respect to source of the supernatant. A Reed and Muench scoring method was employed to determine TCID 50 /ml (42) . Molecular detection of virus RNA levels. The total RNA from the culture supernatants was extracted using a QIAamp Viral RNA kit (Qiagen), according to the manufacturer's instructions. Quantitative RT-PCR was performed using a QuantiTect Probe RT-PCR kit (Qiagen) in an ABI PRISM 7500 sequence detection system (Applied Biosystems). Amplifications were carried out in 25-l reaction mixtures containing 2ϫ reaction mix buffer, 50 M each primer, 10 M probe, and 5 l of RNA template. Primers, probes, and cycling conditions recommended by the Centers for Disease Control and Prevention (CDC) protocol were used to detect SARS-CoV-2 (43) . The standard curve method was employed for virus quantification. For reference to the cell amounts used, the housekeeping gene RNase P was amplified. The threshold cycle (C T ) values for this target were compared to those obtained for different cell amounts, 10 7 to 10 2 , for calibration.

Measurements of inflammatory mediators and cell death marker. The levels of TNF-␣, IL-6, and LDH were quantified in the monocyte supernatants from infected and uninfected cells. ELISAs for TNF-␣ and IL-6 required 100 l of supernatants to be exposed to capture antibody in 96-well plates. After a 2-h incubation period at room temperature (RT), the detection antibody was added. Plates were incubated for another 2 h at RT. Streptavidin-horseradish peroxidase (HRP) and its substrate were added and incubated for 20 min, and the optical density was determined using a microplate reader set to 450 nm.

Extracellular lactate dehydrogenase (LDH) was quantified using a Doles kit according to the manufacturer's instructions. Supernatant was centrifuged at 5,000 rpm for 1 min to remove cellular debris. A total of 25 l of supernatant was placed into 96-well plates and incubated with 5 l of ferric alum and 100 l of LDH substrate for 3 min at 37°C. NAD (oxidized form) was added before the addition of a stabilizing solution. After a 10-min incubation, plates were measured in a spectrophotometer at 492 nm.

Molecular docking. ATV (PubChem CID 148192) and LPV (PubChem CID 92727) were used as inhibitors of the SARS-CoV-2 Mpro. ATV and LPV were prepared using a generalized amber force field (GAFF), and their charges were obtained using the AM1-BCC loading scheme (44, 45) .

Molecular docking experiments were performed with DOCK 6.9 (46) to identify the binding site of Mpro. SARS-CoV-2 Mpro structure was obtained from Protein Data Bank (RCSB PDB; http://www .rcsb.org) under the accession code 6LU7 (47) . The active site region was identified by using a complexed peptide {N-[(5-methylisoxazol-3-yl)carbonyl]alanyl-L-valyl-nϳ1ϳ-[(1r,2z)-4-(benzyloxy)-4-oxo-1-{[(3r)-2oxopyrrolidin-3-yl]methyl}but-2-enyl]-L-leucinamide} as a guide. The creation of the DOCK 6.9 input files for docking was performed using Chimera 1.14 (48) .

The docking of ligands was performed in a box of 10-Å edges with its mass center matching that of the complexed peptide. Each scan produced 20 conformations for each ligand, with the best score being used for molecular dynamics simulations.

Molecular dynamics. Since the tertiary structure (3D) of the SARS-CoV-2 Mpro is a homodimer, we focused the molecular dynamics on only one chain, chain A. Molecular dynamics calculations were performed using NAMD 2.9 (49) and Charmm27* force field (50) at pH 7, i.e., with deprotonated Glu and Asp, protonated Arg and Lys, and neutral His with a protonated N atom. This all-atom force field has been able to fold properly many soluble proteins (51) (52) (53) . The soluble proteins were centered in a cubic box of transferable intermolecular potential with 3 points (TIP3P) water molecules (54); the box extended 1.2 nm outside the protein on its four lateral sides, and the appropriate numbers of Na ϩ and Cl Ϫ ions were added to ensure system neutralization. The electrostatic interactions were calculated using the particle mesh Ewald method and a cutoff of 1.2 nm (55). The same cutoff of 1.2 nm was used for the Van der Waals interactions. The nonbonded pair lists were updated every 10 fs. The analysis was based on molecular dynamics (MD) simulation of 100 ns at 310 K.

Protein extraction. Protein extracts containing SARS-CoV-2 Mpro activity were obtained from Vero cell monolayers on 25-cm 2 flasks that were infected for 1 h at an MOI of 0.1 at 37°C and 5% CO 2 . After 1 or 2 days of infection, the supernatant was harvested, and monolayers were washed 3 times with in sterile cold PBS (pH 7.2). Next, cells were suspended in 1 ml of lysis buffer (100 mM Tris-HCl [pH 8.0], 150 mM NaCl, 10% glycerol, and 0.6% Triton X-100) and kept at 4°C. The soluble protein fraction was isolated as the supernatant after centrifugation (100,000 ϫ g, 30 min, 4°C) and stored at Ϫ20°C until further use. The protein concentrations of the samples were determined using the BCA protein assay kit.

Zymographic assays. Proteinases were assayed after electrophoresis on 10% SDS-PAGE gels with 0.1% copolymerized gelatin (56) . Briefly, each of the gel slots was loaded with 12 g of soluble proteins dissolved in Laemmli's buffer, and following electrophoresis at a constant voltage of 200 V at 4°C, they were soaked for 1 h at 25°C in washing buffer (0.1 mM sodium acetate buffer [pH 5.5] containing 2.5% TX-100). Proteinase activity was detected by incubating (16 h at 37°C) the gels in reaction buffer (0.1 mM sodium acetate buffer pH 5.5 containing 1.0 mM DTT), in the presence or absence of the same concentrations of 10 M E-64, ATV, RTV, or the ATV/RTV combination. Hydrolysis of gelatin was visualized by staining the gels with 0.2% amido black (57) .

Statistical analysis. The assays were performed blinded by one professional, codified, and then read by another professional. All experiments were carried out at least three independent times, including a minimum of two technical replicates in each assay. The dose-response curves used to calculate EC 50 and CC 50 values were generated by variable slope plot from Prism GraphPad software 8.0. The equations to fit the best curve were generated based on R 2 values of Ն0.9. Student's t test was used to assess statistical significance at P values of Ͻ0.05. The statistical analyses specific to each software program used in the bioinformatics analysis are described above.

Data availability. The complete genome of the virus strain used in this study is publicly available at https://nextstrain.org/ncov (Brazil/RJ-314/2020 or GISAID EPI ISL no. 414045).

The molecular biology of coronaviruses

Origin and evolution of pathogenic coronaviruses

Identifying SARS-CoV-2 related coronaviruses in Malayan pangolins

Middle east respiratory syndrome coronavirus (MERS-CoV

An interactive web-based dashboard to track COVID-19 in real time

2020. Opinion. Will our economy die from coronavirus?

Coronavirus puts drug repurposing on the fast track

WHO R&D Blueprint: informal consultation on prioritization of candidate therapeutic agents for use in novel coronavirus 2019 infection

Effect of high vs low doses of chloroquine diphosphate as adjunctive therapy for patients hospitalized with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection: a randomized clinical trial

A trial of lopinavir-ritonavir in adults hospitalized with severe COVID-19

Approved Antiviral drugs over the past 50 years

Small molecules targeting severe acute respiratory syndrome human coronavirus

Coronaviruses: an overview of their replication and pathogenesis

Pharmacokinetics and pharmacodynamics of cytochrome P450 inhibitors for HIV treatment

Effects of switching from lopinavir/ritonavir to atazanavir/ritonavir on muscle glucose uptake and visceral fat in HIV-infected patients

Treatment guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents: an update

Old drugs for newly emerging viral disease

Preclinical pharmacokinetics and tissue distribution of long-acting nanoformulated antiretroviral therapy

Protective effect of atazanavir sulphate against pulmonary fibrosis in vivo and in vitro

Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

The host immune response in respiratory virus infection: balancing virus clearance and immunopathology

Zika virus and neurological diseaseapproaches to the unknown

Yellow fever virus is susceptible to sofosbuvir both in vitro and in vivo

Beyond members of the Flaviviridae family, sofosbuvir also inhibits chikungunya virus replication

Sofosbuvir protects Zika virus-infected mice from mortality, preventing short-and long-term sequelae

The clinically approved antiviral drug sofosbuvir inhibits Zika virus replication

Efficacy of sofosbuvir as treatment for yellow fever: protocol for a randomised controlled trial in Brazil (SOFFA study)

No clinical benefit from use of lopinavir-ritonavir in hospitalised COVID-19 patients studied in RECOVERY

Triple combination of interferon beta-1b, lopinavir-ritonavir, and ribavirin in the treatment of patients admitted to hospital with COVID-19: an open-label, randomised, phase 2 trial

Experimental treatment with favipiravir for COVID-19: an open-label control study

Comparative therapeutic efficacy of remdesivir and combination lopinavir, ritonavir, and interferon beta against MERS-CoV. 1

HIV protease inhibitors: a review of molecular selectivity and toxicity

UPLC-MS/MS quantification of nanoformulated ritonavir, indinavir, atazanavir, and efavirenz in mouse serum and tissues

Predicting commercially available antiviral drugs that may act on the novel coronavirus (SARS-CoV-2) through a drug-target interaction deep learning model

The crystal structures of severe acute respiratory syndrome virus main protease and its complex with an inhibitor

Identification of potential binders of the main protease 3CL pro of the COVID-19 via structure-based ligand design and molecular modeling

Cytokine and chemokine profiles in lung tissues from fatal cases of 2009 pandemic influenza A (H1N1): role of the host immune response in pathogenesis

Role of macrophage cytokines in influenza A virus infections

Pathogenic human coronavirus infections: causes and consequences of cytokine storm and immunopathology

A simple method of estimating fifty percent endpoints

CDC. 2020. Coronavirus disease 2019 (COVID-19). CDC

Fast, efficient generation of high-quality atomic charges. AM1-BCC model: II. Parameterization and validation

Development and testing of a general amber force field

DOCK 6: impact of new features and current docking performance

Structure of M pro from SARS-CoV-2 and discovery of its inhibitors

UCSF Chimera-a visualization system for exploratory research and analysis

Scalable molecular dynamics with NAMD

Development and current status of the CHARMM force field for nucleic acids

Folding atomistic proteins in explicit solvent using simulated tempering

Communication: mul-SARS-CoV-2 Is Susceptible to Atazanavir Antimicrobial Agents and Chemotherapy tiple atomistic force fields in a single enhanced sampling simulation

How fast-folding proteins fold

Comparison of simple potential functions for simulating liquid water

Particle mesh Ewald: an N·log(N) method for Ewald sums in large systems

Electrophoretic analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulfate and copolymerized substrates

Heterogeneity of cysteine proteinases in Leishmania braziliensis and Leishmania major

Induction of TNF-alpha in human macrophages by avian and human influenza viruses