key: cord-0860589-rpoa9gk5
authors: Shi, Yuejun; Shuai, Lei; Wen, Zhiyuan; Wang, Chong; Yan, Yuanyuan; Jiao, Zhe; Guo, Fenglin; Fu, Zhen F.; Chen, Huanchun; Bu, Zhigao; Peng, Guiqing
title: The Preclinical Inhibitor GS441524 in Combination with GC376 Efficaciously Inhibited the Proliferation of SARS-CoV-2 in the Mouse Respiratory Tract
date: 2020-11-13
journal: bioRxiv
DOI: 10.1101/2020.11.12.380931
sha: 5cb1faf7bee82d07de1bc21037166d23fb0541b9
doc_id: 860589
cord_uid: rpoa9gk5

The unprecedented coronavirus disease 2019 (COVID-19) pandemic, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is a serious threat to global public health. Development of effective therapies against SARS-CoV-2 is urgently needed. Here, we evaluated the antiviral activity of a remdesivir parent nucleotide analog, GS441524, which targets the coronavirus RNA-dependent RNA polymerase enzyme, and a feline coronavirus prodrug, GC376, which targets its main protease, using a mouse-adapted SARS-CoV-2 infected mouse model. Our results showed that GS441524 effectively blocked the proliferation of SARS-CoV-2 in the mouse upper and lower respiratory tracts via combined intranasal (i.n.) and intramuscular (i.m.) treatment. However, the ability of high-dose GC376 (i.m. or i.n. and i.m.) was weaker than GS441524. Notably, low-dose combined application of GS441524 with GC376 could effectively protect mice against SARS-CoV-2 infection via i.n. or i.n. and i.m. treatment. Moreover, we found that the pharmacokinetic properties of GS441524 is better than GC376, and combined application of GC376 and GS441524 had a synergistic effect. Our findings support the further evaluation of the combined application of GC376 and GS441524 in future clinical studies. Importance Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes coronavirus disease 2019 (COVID-19), which has seriously threatened global public health and economic development. Currently, effective therapies to treat COVID-19 are urgently needed. In this study, we assessed the efficacy of the preclinical inhibitors GC376 and GS441524 using a mouse-adapted SARS-CoV-2 infected mouse model for the first time. Our results showed that low-dose combined application of GC376 and GS441524 could effectively protect mice from HRB26M infection in the upper and lower respiratory tracts. Hence, the combined application should be developed and considered for future clinic practice.

Among these nsps, M pro and nsp12 (RdRp) of SARS-CoV-2 are functionally and 67 structurally conserved among these viruses and essential for viral replication; they are 68 considered a potential target for the design of antiviral drugs (10-16).

GC376, a dipeptidyl bisulfite adduct salt, exerts strong inhibitory effects on 70 picornaviruses and coronaviruses (11, 14, 15, 17) . The latest research has shown that 71 GC376 can also inhibit the replication of SARS-CoV-2 in Vero E6 cells (12, 13, 16) . 72 Moreover, antiviral treatment with GC376 led to a full recovery in laboratory cats 73 with feline infectious peritonitis (FIP) caused by feline coronavirus (FIPV) (18). 74 Currently, no studies have evaluated the ability of GC376 to inhibit SARS-CoV-2 in 75 animals. In addition, remdesivir (RDV) and its parent nucleoside analog GS-441524 76 inhibit CoVs and other viruses (19) (20) (21) (22) . The latest research also suggests that RDV 77 inhibits SARS-CoV-2 in vivo and in vitro (10, (23) (24) (25) (26) (27) . However, recent research 78 showed that RDV was not intended for lung-specific delivery, and GS-441524 is the 79 predominant metabolite that reaches the lungs, which suggests that GS-441524 is 80 superior to RDV for 29) . 81 Because GC376 and GS441524 target the key proteases M pro and RdRp of 82 SARS-CoV-2 replication, respectively, their combined application may be more 83 effective in the treatment of COVID-19. The combined application of GC376 with 84 RDV produced additive sterilizing additive effects against SARS-CoV-2 in Vero E6 85 cells (13). However, the efficacy of the combined application of GC376 and 86 GS441524 in the inhibition of SARS-CoV-2 replication was not investigated in vivo 87 and in vitro. Our team generated a mouse-adapted SARS-CoV-2 (HRB26M) that 88 5 efficiently replicates in the upper and lower respiratory tract in BALB/c mice (4, 30) . 89 Hence, we assessed the antiviral efficacy of GC376 and GS441524 in an 90 HRB26M-infected BALB/c mouse model. Our results showed that GC376, which 91 effectively inhibited the virus at the cellular level, was less effective in vivo. The 92 ability of GS441524 to block the proliferation of SARS-CoV-2 was significantly 93 higher than GC376 in mice. Notably, the low-dose combined application of GC376 94 and GS441524 could effectively block SARS-CoV-2 infection in the mouse upper and 95 lower respiratory tract.

Structural analysis of GC376 and GS441524 targeting SARS-CoV-2 M pro and 98 nsp12 (RdRp) 99 In this study, we first analyzed the mechanism by which GC376 and GS441524 100 inhibit SARS-CoV-2 replication through structural biology methods. GC376 (Fig. 1a) 101 is a preclinical inhibitor against feline infectious peritonitis virus (FIPV) (31) that has 102 become a broad-spectrum prodrug targeting coronavirus M pro (14, 32, 33) . We 103 determined the high-resolution crystal complex structure of M pro with GC376 at a 104 resolution of 2.35 Å ( Table S1 ). The crystal of M pro -GC376 belongs to the space 105 group P1211, and an asymmetric unit contains two molecules (designated protomer A 106 and protomer B) (Fig. 1b) . The structure of each protomer contains three domains 107 with the substrate-binding site comprised of a His41-Cys145 dyad located in the cleft 108 between domain I and II (Supplementary Fig. 1) . The electron density map clearly 109 showed compound GC376 in the substrate binding pocket of the SARS-CoV-2 M pro , 110 6 and details of the interaction are shown in Fig. 1c and Supplementary Fig. 1 . GC376 111 interacts with His41, Phe140, Leu141, Asn142, Ser144, Cys145, His163, His164, 112 Met165, Glu166, His172, Asp187, Arg188 and Gln189 through many hydrogen bonds 113 and hydrophobic interactions, and binds stably in the groove formed by domain I and 114 domain II. Moreover, we found that GC376 is covalently attached to Cys145 as a 115 hemithioacetal (Fig. 1c) , which prevents the binding and processing of M pro to the 116 substrate (12, 16, 32) . Additionally, GS441524 is the parent nucleoside of RDV (Fig. 117 1d and Supplementary Fig. 2a) , and has shown broad-spectrum inhibition of the 118 replication of various coronaviruses, including SARS- 34, 35) . Since the 119 mechanism by which GS441524 inhibits SARS-CoV-2 replication is similar to that of 120 RDV (28, 34), we showed the cryo-electron microscopy (cryo-EM) ternary structure Fig. 2a, b) . GC376 and GS441524 were also efficacious against HRB26M, with 137 EC 50 values of 0.881±0.109 μM and 5.047±2.116 μM, respectively (Fig. 2d, e) . Our 138 results showed that the ability of GC376 to inhibit viral (HRB26 and HRB26M) 139 replication was better than GS441524 when the agents were applied alone. We (i.n. and i.n.+i.m.) and GC376+GS441524 (i.n. and i.n.+i.m.). Our results showed that 185 GS441524 and GC376+GS441524 markedly inhibited viral replication in nasal 186 turbinates on days 3 and 5 p.i. (Fig. 4b, d and Supplementary Fig. 7b, d) . Only 187 slightly infectious virus was detected on day 5 p.i., although viral RNA was detected 188 in 2 of the 3 mice in the GS441524-treated group (Fig. 4c, d and Supplementary Fig.   189 7c, d). However, i.n. administration of GS441524 did not significantly inhibit viral 190 replication in the lungs compared to that of the mock-treated group on day 5 p.i (Fig. 191 4g, h and Supplementary Fig. 8c, d) . Notably, we observed that i.n. administration 192 of GC376+GS441524 blocked viral replication in the lungs on day 5 p.i., although 193 viral RNA was detected in 1 of the 3 mice ( Fig. 4 g, h and Supplementary Fig. 8 to that of the mock-treated group) ( Fig. 4 and Supplementary Figs. 7 and 8) . approximately 1500-fold and 263-fold compared to that of the mock-treated group, 202 respectively) ( Fig. 4 and Supplementary Figs. 7 and 8) . Although GC376 treatment 203 via the i.n. or combined i.n. and i.m. routes reduced the level of virus replication on 204 day 5 p.i (Supplementary Figs. 7d and 8d) , it did not completely blocked viral 205 replication in nasal turbinates and lungs (Fig. 4) .

Pharmacokinetics study of GC376 and GS441524 alone or in combination 207 To further examine the potential of GC376 and GS441524, we evaluated their 208 pharmacokinetic (PK) properties in SPF SD rats following i.m. administration of 209 GC376 (111 mg/kg), GS441524 (67 mg/kg) and GC376+GS441524 (55.5+33.5 210 mg/kg), which are the same doses used in mice according to weight. The PK results 211 showed that GC376 and GS441524 were rapidly absorbed after i.m. administration, 212 and the peak plasma level was reached 1.30±0.60 h and 2.00±1.10 h after injection, 213 respectively ( Fig. 5 turbinates and lungs, and the specific mechanism needs further study.

Furthermore, we found that the combined application of GC376 and GS441524 230 had a synergistic effect and extended T 1/2 from 3.80±1.17 h to 5.13±2.56 h, T max from 231 2.00±1.10 h to 3.40±1.20 h and the residence time of GS441524 (MRT 0-t from 232 4.50±1.11 h to 6.03±1.37 h) ( Fig. 5c and Table 1 ). However, the injection dose was 233 halved because the clearance rate was not changed, and the MRT 0-t of GC376 234 decreased from 5.80±0.74 h to 2.97±0.41 h. The PK study results showed that GC376 235 reached C max earlier (T max =1.40±0.49) than GS441524 (T max =3.40±1.20 h) to produce 236 a synergistic effect (Fig. 5c) . When these agents were combined, GC376 was the first 237 drug to inhibit SARS-CoV-2 replication. After the plasma concentration of GC376 238 decreased, GS441524 reached its C max (Fig. 5c and Table1) to produce a continuous 239 inhibition of SARS-CoV-2 proliferation and maintain the effective concentration for a 240 longer time. This phenomenon may explain why the combined application of GC376 241 and GS441524 was better than single application alone. In addition, the difference in 242 12 the pharmacokinetics of GC376 and GS441524 between the upper and lower 243 respiratory tract needs to be further evaluated.

In summary, we assessed the efficacy of GC376 and GS441524 to inhibit 

The compounds GC376 (molecular weight: 507.53 g/M) and GS441524 259 (molecular weight: 291.26 g/M) were synthesized at WuXi AppTec with purities 260 higher than 95%. They were dissolved in 5% ethanol, 30% propylene glycol, 45% 261 PEG 400, and 20% water with a concentration of 40 mM/l according to a previously 262 described study (35) . EDTAK2. Then, the whole blood was centrifuged at 7800 g/min for 10 min at 4℃.

Plasma was collected and stored in a freezer at -80°C. Plasma drug concentration was 371 analyzed using LC-MS/MS.

Pharmacokinetic parameters were calculated using WinNonlin software (version 6.4), 373 and a non-atrioventricular model was used for data fitting. The data were analyzed 

The toxicity studies in mice and single-dose PK study in SD rats were performed 380 using protocols that were approved by the Scientific Ethics Committee of Huazhong presented as mean values +/-SD (95% confidence interval). *, P <0.05 was 390 considered statistically significant; **, P < 0.01 was considered highly significant; 391 ***, P < 0.001 and ****, P < 0.0001 were considered extremely significant. All 392 experiments were further confirmed using biological repeats. supporting the findings of this study are available within the paper and its 397 Supplementary Information files, or upon reasonable request from the corresponding 398 author.

Development of an inactivated vaccine candidate 424 for SARS-CoV-2

ChAdOx1 nCoV-19 vaccine prevents SARS-CoV-2 432 pneumonia in rhesus macaques

DNA vaccine protection against 441 SARS-CoV-2 in rhesus macaques

A single dose of an 445 adenovirus-vectored vaccine provides protection against SARS-CoV-2 446 challenge

Safety, tolerability, and immunogenicity 450 of a recombinant adenovirus type-5 vectored COVID-19 vaccine: a 451 dose-escalation, open-label, non-randomised, first-in-human trial

Immunogenicity and safety of a recombinant 457 adenovirus type-5-vectored COVID-19 vaccine in healthy adults aged 18 458 years or older: a randomised, double-blind

Nidovirales: 461 evolving the largest RNA virus genome

Coronaviruses post-SARS: update on replication 463 and pathogenesis

Structural basis for inhibition of the 469 RNA-dependent RNA polymerase from SARS-CoV-2 by remdesivir

Structural Basis for

Inhibiting Porcine Epidemic Diarrhea Virus Replication with the 3C-Like 473

Boceprevir, GC-376, and calpain inhibitors 476

XII inhibit SARS-CoV-2 viral replication by targeting the viral main 477 protease

Both Boceprevir and GC376 480 efficaciously inhibit SARS-CoV-2 by targeting its main protease

Broad-spectrum antivirals against 3C or 3C-like 484 23 proteases of picornaviruses, noroviruses, and coronaviruses

Design, synthesis, and 488 evaluation of inhibitors of Norwalk virus 3C protease

Feline coronavirus drug inhibits the main protease of 493 SARS-CoV-2 and blocks virus replication

Potent inhibition of norovirus 3CL 496 protease by peptidyl α-ketoamides and α-ketoheterocycles

Coronavirus Susceptibility to the Antiviral 505

GS-5734) Is Mediated by the Viral Polymerase and the

Synthesis and antiviral activity of a series of 1'-substituted 509 4-aza-7,9-dideazaadenosine C-nucleosides

Broad-spectrum antiviral GS-5734 514 inhibits both epidemic and zoonotic coronaviruses

Therapeutic efficacy of the small 523 molecule GS-5734 against Ebola virus in rhesus monkeys

lopinavir, emetine, and homoharringtonine inhibit SARS-CoV-2 replication in 527 vitro

Remdesivir and chloroquine effectively inhibit the recently emerged 530 novel coronavirus (2019-nCoV) in vitro

Lianhuaqingwen exerts anti-viral and 534 anti-inflammatory activity against novel coronavirus (SARS-CoV-2)

Comparative 539 therapeutic efficacy of remdesivir and combination lopinavir, ritonavir, and 540 interferon beta against MERS-CoV

Prophylactic and therapeutic remdesivir 543 (GS-5734) treatment in the rhesus macaque model of MERS-CoV infection

Advantages of the Parent Nucleoside

over Remdesivir for Covid-19 Treatment

remdesivir and GS-441524 in two critically ill patients who recovered from 551 COVID-19

Mouse-adapted SARS-CoV-2 554 replicates efficiently in the upper and lower respiratory tract of BALB

Efficacy of a 558 3C-like protease inhibitor in treating various forms of acquired feline 559 infectious peritonitis

Broad-spectrum 562 inhibitors against 3C-like proteases of feline coronaviruses and feline 563 caliciviruses

Protease inhibitors broadly effective 566 against feline, ferret and mink coronaviruses

Remdesivir Inhibits SARS-CoV-2 in Human

Lung Cells and Chimeric SARS-CoV Expressing the SARS-CoV-2 RNA 573 Polymerase in Mice

The nucleoside analog GS-441524 strongly inhibits 576 feline infectious peritonitis (FIP) virus in tissue culture and experimental cat 577 infection studies

Mechanism of Inhibition 579 of Ebola Virus RNA-Dependent RNA Polymerase by Remdesivir. Viruses 11

Remdesivir is a direct-acting antiviral that inhibits RNA-dependent 582 RNA polymerase from severe acute respiratory syndrome coronavirus 2 with 583 high potency

A Dimerization-Dependent Mechanism Drives the Endoribonuclease 586 Function of Porcine Reproductive and Respiratory Syndrome Virus nsp11

Structural Characterization of the Helicase 590 nsp10 Encoded by Porcine Reproductive and Respiratory Syndrome Virus

HKL-3000: 593 the integration of data reduction and structure solution

598 PHENIX: building new software for automated crystallographic structure 599 determination

Coot: model-building tools for molecular graphics

LigPlot+: multiple ligand-protein 603 interaction diagrams for drug discovery

Structural analysis of GC-376 and GS441524 targeting SARS-CoV-2 M pro 619 and RdRp. (a) The dipeptidyl protease inhibitor, GC376. (b) Crystal structure of 620

GC376 interacts covalently with the 621 active cysteine site of SARS-CoV-2 M pro . Electron density at 1.5 σ is shown in gray 622 mesh. Hydrogen bonds are shown as red dashed lines. (d) The chemical structure of 623

PDB 624 ID: 7BV2). (f) Enlarged view of the active site, depicting the interaction between 625 RDV and surrounding amino acids

GC376 and GS441524 potently inhibit SARS-CoV-2 in Vero E6 cells

d-f) Percent inhibition of SARS-CoV-2 632 (HRB26M) by GC376, GS441524 and GC376+GS441524 in Vero E6 cells. The 633 concentrations of GC376 and GS441524 ranged from 0 to 10 µM. The 50% inhibitory 634 concentration (EC 50 ) was calculated using GraphPad Prism 7.0 to assess inhibition 635 ratios at different inhibitor concentrations

Evaluation of i.m. GC376, GS441524 and GC376+GS441524 against 644

One 649 hour after administration of the loading dose of GC376, GS441524, 650 GC376+GS441524 or vehicle solution, the mice were inoculated intranasally with 651 10 3.6 PFU of HRB26M in a volume of 50 μl. On days 3 and 5 p.i., three mice in each 652 32 group were euthanized and their nasal turbinates and lungs were collected. The viral 653 RNA copies and infectious titers in the nasal turbinates

Evaluation of i.n. and i.m. GC376, GS441524 and GC376+GS441524 676 against SARS-CoV-2 infection in mice. Four-to six-week-old female BALB/c mice 677 were administered a loading dose of GC376 (1 µM

One hour after 682 administration of the loading dose of GC376, GS441524, GC376 + GS441524 or 683 vehicle solution, the mice were inoculated intranasally with 10 3.6 PFU of HRB26M in 684 a volume of 50 μl. On days 3 and 5 p.i., three mice in each group were euthanized and 685 34 their nasal turbinates and lungs were collected. The viral RNA copies and infectious 686 titers in the nasal turbinates (a-d) and lungs (e-h) were detected by qPCR and viral 687 titration

Fig.5 Plasma concentration-vs-time profiles following GC376, GS441524 and 691 GC376+GS441524 administration in SD rats. In the single-dose PK study, five SPF 692 SD rats were i.m. injected with GC376 (111 mg/kg), GS441524 (67 mg/kg) and

5 mg/kg) for the determination of serial plasma drug 694 concentrations. Data were analyzed via GraphPad Prism7.0, and the error bars show 695 the SEM of the results from five replicates

 400 We thank research associates at the Center for Protein Research (CPR), 401 Huazhong Agricultural University, for technical support. We also thank Professor