Discovery of novel non-peptidic and non-covalent small-molecule 3CLpro inhibitors as potential candidate for COVID-19 treatment

Dear Editor, The global coronavirus disease 2019 (COVID-19) pandemic, which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has brought a profound impact on humanity. SARS-CoV-2 enters host cells through the interaction of its spike glycoprotein with angiotensin-converting enzyme 2 (ACE2) of the host and releases its viral RNA genome. The genome is further translated into two polyproteins, PP1a and PP1ab, that contain 16 non-structural proteins, in which the non-structural protein 5 (also called as 3C-Like protease, 3CL) plays a vital role in the life cycle of the virus. As a consequence, multiple peptide-like covalent 3CL inhibitors have been successively reported to show good 3CL inhibitory potencies and antiviral activities. However, the defects of peptidomimetic inhibitors about low membrane permeability and metabolic stability are obvious, in which the approved drug paxlovid (PF-07321332) from Pfizer must be co-administrated with ritonavir as a pharmacokinetic booster. Non-peptidic small molecule inhibitors may have more advantages in metabolic properties and are gradually attracting attention. Our research attempted to discover a novel non-peptidic and noncovalent small molecule 3CL inhibitor so as to give better drug-like properties and good antiviral activities. After screening more than 1,700,000 compounds from the Chinese National Compound Library, we identified 1851 hits targeting 3CL. Taken together with molecular docking and cellbased antiviral assay, we selected two structurally similar compounds, G006 and G004, for further optimization (Fig. 1a). The docking analyses for G006 demonstrated that the quinoline2-one occupies the S1 site to form an important hydrogen bond with Glu166 and Gly143, the 3-chlorophenyl occupies the hydrophobic S2 site to form a π–π interaction with His41. (Fig. 1b and Supplementary Fig. 1). As we have seen, the active site was not fully occupied by G006, which may limit its inhibition potency against 3CL (Fig. 1b). Thus, we have attempted to introduce diversified side chains onto the core piperazine ring in the hope of filling these unoccupied cavities fully. Finally, JZD-07 outstripped its analogs, exhibiting excellent inhibitory activity against the 3CL of both SARS-CoV-2 and SARS-CoV-1, with IC50 of 0.15 μM and 0.11 μM, respectively (Fig. 1c). To explore the selectivity of this 3CL inhibitor, we first examined its inhibitory activities against multiple closely related proteases, which indicated JZD-07 had almost no inhibition for human calpain I, cathepsin (B, D, K, and L), thrombin A, trypsin, caspase-3, DPP-4, and proteasome, while the control 3CL inhibitors GC-376 and PF-07321332 showed certain inhibitory potencies against different cathepsin proteins (Fig. 1d). We also evaluated the inhibitory potencies toward SARS-CoV-2 viral life cycle related targets, and JZD-07 had no inhibition against those enzymes with IC50 > 100 μM (Supplementary Table 1). These data demonstrate that JZD-07 is a highly specific 3CL inhibitor. Based on the inhibitory potency against 3CL, we conducted an antivirus assay of JZD-07 in Vero E6 cells. JZD-07 showed good inhibitory activity against the SARS-CoV-2 original virus with EC50 of 0.82 μM and also showed dose-dependent inhibition on the replication of delta and omicron BA.1. variant with EC50 values of 7.24 μM and 6.03 μM, respectively (Fig. 1e and Supplementary Fig. 2). Besides, JZD-07 exhibited good antivirus activities against HCoV-OC43 and HCoV-229E (Supplementary Fig. 2). Meanwhile, this compound almost has no effect on cell viability with the CC50 in Vero E6 cell over 300 μM (Fig. 1f). These results indicate that JZD-07 has good antiviral activity in vitro and almost has no cytotoxicity. To elucidate the mechanism of action of this non-peptidic small molecule inhibitor, the enzyme kinetics and drug bindingkinetics profiles were studied in detail. Firstly, the inhibition reversibility was investigated by rapid dilution assay. JZD-07 (10 μM) or an irreversible 3CL inhibitor (GC-376, 2 μM) were incubated with SARS-CoV-2 3CL (2 μM) for 20 min, the activities of 3CL incubated with JZD-07 recovered more than 70% after 100-fold dilution. Meanwhile, the IC50 values did not change with preincubation time (Fig. 1g), which indicates that the inhibition against 3CL is reversible. Furthermore, we determined the kinetic parameters using different concentrations of JZD-07. After global fits of the respective enzyme inhibition curves, JZD-07 is shown to be a competitive inhibitor with a Ki value of 0.117 μM (Fig. 1h), comparable to the reported 3CL inhibitor CCF981 and more potent than ML-188 (Supplementary Fig. 3). Recently, the importance of drugbinding kinetics profiles for drug design have been increasingly recognized. We utilized bio-layer interferometry (BLI) assay to measure the binding kinetics and affinity parameters for our compound (Fig. 1i and Supplementary Fig. 4). An overall kon value of JZD-07 is 2.36 × 10 M S, the dissociation behavior is very slow with koff= 0.028 S , and the KD value of JZD-07 is 117 nM. Taken together, the results of enzymatic inhibition and BLI data indicate that JZD-07 is a high affinity, slow-dissociation, and reversible competitive 3CL inhibitor. To elucidate the precise binding modes of the inhibitor with protease, we successfully determined the crystal structure of SARS-CoV-2 3CL in complex with JZD-07 at a resolution of 1.80 Å (Supplementary Table 2). The binding modes showed JZD07 non-covalently bounded into the substrate-binding site and occupied S1’, S1, and S2 subsites (Fig. 1j). The 2-quinolinone group fitted into the S1 subsite by forming hydrogen bonds with the side chains of His163 and Glu166. The carbonyl oxygen atom between the piperazine ring and quinoline ring fixed the conformation of the flexible oxyanion loop by making hydrogen


Dear Editor,
The global coronavirus disease 2019 (COVID-19) pandemic, which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has brought a profound impact on humanity.SARS-CoV-2 enters host cells through the interaction of its spike glycoprotein with angiotensin-converting enzyme 2 (ACE2) of the host and releases its viral RNA genome. 1 The genome is further translated into two polyproteins, PP1a and PP1ab, that contain 16 non-structural proteins, in which the non-structural protein 5 (also called as 3C-Like protease, 3CL pro ) plays a vital role in the life cycle of the virus. 2 As a consequence, multiple peptide-like covalent 3CL pro inhibitors have been successively reported to show good 3CL pro inhibitory potencies and antiviral activities. 3,4However, the defects of peptidomimetic inhibitors about low membrane permeability and metabolic stability are obvious, in which the approved drug paxlovid (PF-07321332) from Pfizer must be co-administrated with ritonavir as a pharmacokinetic booster. 36][7] Our research attempted to discover a novel non-peptidic and noncovalent small molecule 3CL pro inhibitor so as to give better drug-like properties and good antiviral activities.
After screening more than 1,700,000 compounds from the Chinese National Compound Library, 8 we identified 1851 hits targeting 3CL pro .Taken together with molecular docking and cellbased antiviral assay, we selected two structurally similar compounds, G006 and G004, for further optimization (Fig. 1a).The docking analyses for G006 demonstrated that the quinoline-2-one occupies the S1 site to form an important hydrogen bond with Glu166 and Gly143, the 3-chlorophenyl occupies the hydrophobic S2 site to form a π-π interaction with His41.(Fig. 1b and Supplementary Fig. 1).As we have seen, the active site was not fully occupied by G006, which may limit its inhibition potency against 3CL pro (Fig. 1b).Thus, we have attempted to introduce diversified side chains onto the core piperazine ring in the hope of filling these unoccupied cavities fully.Finally, JZD-07 outstripped its analogs, exhibiting excellent inhibitory activity against the 3CL pro of both SARS-CoV-2 and SARS-CoV-1, with IC 50 of 0.15 μM and 0.11 μM, respectively (Fig. 1c).
To explore the selectivity of this 3CL pro inhibitor, we first examined its inhibitory activities against multiple closely related proteases, which indicated JZD-07 had almost no inhibition for human calpain I, cathepsin (B, D, K, and L), thrombin A, trypsin, caspase-3, DPP-4, and proteasome, while the control 3CL pro inhibitors GC-376 and PF-07321332 showed certain inhibitory potencies against different cathepsin proteins (Fig. 1d).We also evaluated the inhibitory potencies toward SARS-CoV-2 viral life cycle related targets, and JZD-07 had no inhibition against those enzymes with IC 50 > 100 μM (Supplementary Table 1).These data demonstrate that JZD-07 is a highly specific 3CL pro inhibitor.
Based on the inhibitory potency against 3CL pro , we conducted an antivirus assay of JZD-07 in Vero E6 cells.JZD-07 showed good inhibitory activity against the SARS-CoV-2 original virus with EC 50 of 0.82 μM and also showed dose-dependent inhibition on the replication of delta and omicron BA.1.variant with EC 50 values of 7.24 μM and 6.03 μM, respectively (Fig. 1e and Supplementary Fig. 2).Besides, JZD-07 exhibited good antivirus activities against HCoV-OC43 and HCoV-229E (Supplementary Fig. 2).Meanwhile, this compound almost has no effect on cell viability with the CC 50 in Vero E6 cell over 300 μM (Fig. 1f).These results indicate that JZD-07 has good antiviral activity in vitro and almost has no cytotoxicity.
To elucidate the mechanism of action of this non-peptidic small molecule inhibitor, the enzyme kinetics and drug bindingkinetics profiles were studied in detail.Firstly, the inhibition reversibility was investigated by rapid dilution assay.JZD-07 (10 μM) or an irreversible 3CL pro inhibitor (GC-376, 2 μM) were incubated with SARS-CoV-2 3CL pro (2 μM) for 20 min, the activities of 3CL pro incubated with JZD-07 recovered more than 70% after 100-fold dilution.Meanwhile, the IC 50 values did not change with preincubation time (Fig. 1g), which indicates that the inhibition against 3CL pro is reversible.Furthermore, we determined the kinetic parameters using different concentrations of JZD-07.After global fits of the respective enzyme inhibition curves, JZD-07 is shown to be a competitive inhibitor with a K i value of 0.117 μM (Fig. 1h), comparable to the reported 3CL pro inhibitor CCF981 and more potent than ML-188 (Supplementary Fig. 3).Recently, the importance of drugbinding kinetics profiles for drug design have been increasingly recognized.We utilized bio-layer interferometry (BLI) assay to measure the binding kinetics and affinity parameters for our compound (Fig. 1i and Supplementary Fig. 4).An overall k on value of JZD-07 is 2.36 × 10 5 M −1 S, −1 the dissociation behavior is very slow with k off = 0.028 S −1 , and the K D value of JZD-07 is 117 nM.Taken together, the results of enzymatic inhibition and BLI data indicate that JZD-07 is a high affinity, slow-dissociation, and reversible competitive 3CL pro inhibitor.
To elucidate the precise binding modes of the inhibitor with protease, we successfully determined the crystal structure of SARS-CoV-2 3CL pro in complex with JZD-07 at a resolution of 1.80 Å (Supplementary Table 2).The binding modes showed JZD-07 non-covalently bounded into the substrate-binding site and occupied S1', S1, and S2 subsites (Fig. 1j).The 2-quinolinone group fitted into the S1 subsite by forming hydrogen bonds with the side chains of His163 and Glu166.The carbonyl oxygen atom between the piperazine ring and quinoline ring fixed the conformation of the flexible oxyanion loop by making hydrogen bonds with the main chain of Gly143 as well as a water molecule which simultaneously contacted with the main chain of Cys145.The O-dichlorobenzene ring is deeply inserted into the S2 subsite by making extensive hydrophobic interactions with multiple residues His41/Met49/Tyr54/Met165/Asp187/Arg188.Notably, it also formed π-π stacking interactions with the side chain of catalytic His41.The morpholinyl group of JZD-07 oriented to the S1 subsite, producing water-mediated hydrogen bonds and hydrophobic interactions with the residue Glu166.These results together provided the molecular details JZD-07 recognized by SARS-CoV-2 3CL pro , which is consistent with its excellent 3CL pro inhibitory activities.
Considering our compound is a novel non-covalent and nonpeptidic 3CL pro inhibitor, we further explored the potential advantages of PK profiles.Oral administration of 20 mg/kg of JZD-07 could give favorable PK parameters with an oral bioavailability of 28.1%.Intraperitoneal administration of 10 mg/kg of JZD-07 also showed an excellent PK profile with an AUC of 2280 h*ng/mL and a C max of 1610 ng/mL (Fig. 1k).Besides, we have launched an acute toxicity experiment, and the results indicate that JZD-07 has a good safety profile (Supplementary Fig. 6).These favorable PK and safety characteristics indicated JZD-07 might be suitable for the treatment of SARS-CoV-2 in vivo infection model.
In the end, we investigated the antiviral potency for JZD-07 in K18-hACE2 mice challenged with the SARS-CoV-2 delta variant.After 2 h of challenge, the mice were injected intraperitoneally with JZD-07 or vehicle (day 0).Then mice were administrated twice at 8 h intervals on day 1.On day 2, the mice were euthanized 4 h after receiving one compound or vehicle treatment (Fig. 1l).Lung tissues were collected for the detection of viral RNA and viral titer, histopathological analysis and immunostaining of lungs were also evaluated.Compared with the vehicle group, the viral copy loads of the lung in the JZD-07 (300 mg/kg bid) treated group were significantly reduced by nearly 1.0 log on day 2.More importantly, the viral titers of the lung of the JZD-07 treated group showed 2.0 log reduction (Fig. 1m).Immunohistochemistry staining targeting viral nucleocapsid protein demonstrated that JZD-07 suppressed viral titer in mice lungs at day 2 (Fig. 1n).All these results indicate JZD-07 has a good therapeutic potential against SARS-CoV-2 delta variant in vivo.
Therefore, we here reported our discovery of a novel nonpeptidic and non-covalent small molecule SARS-CoV-2 3CL pro inhibitor.Based on the hit compounds from high-throughput screening studies and a structure-based drug design strategy, compound JZD-07 stands out and shows an excellent inhibitory potency against 3CL pro .It also exhibits good anti-SARS-CoV-2 activity in Vero E6 cells without cytotoxicity.Mechanistic studies and X-ray crystal structure analysis have shown that JZD-07 is a highly selective, slow-dissociation, non-covalent, and reversible competitive 3CL pro inhibitor.JZD-07 also displays favorable pharmacokinetic profiles in mice with good plasma exposure and bioavailability.More importantly, JZD-07 exhibits a good therapeutic potential against the SARS-CoV-2 delta variant in vivo assay.Overall, these new findings offer a good drug candidate for the clinical treatment of COVID-19.