Large library docking for novel SARS-CoV-2 main protease non-covalent and covalent inhibitors

Antiviral therapeutics to treat SARS-CoV-2 are much desired for the on-going pandemic. A well-precedented viral enzyme is the main protease (MPro), which is now targeted by an approved drug and by several investigational drugs. With the inevitable liabilities of these new drugs, and facing viral resistance, there remains a call for new chemical scaffolds against MPro. We virtually docked 1.2 billion non-covalent and a new library of 6.5 million electrophilic molecules against the enzyme structure. From these, 29 non-covalent and 11 covalent inhibitors were identified in 37 series, the most potent having an IC50 of 29 μM and 20 μM, respectively. Several series were optimized, resulting in inhibitors active in the low micromolar range. Subsequent crystallography confirmed the docking predicted binding modes and may template further optimization. Together, these compounds reveal new chemotypes to aid in further discovery of MPro inhibitors for SARS-CoV-2 and other future coronaviruses.


Introduction
recovered by increasing addition of bovine serum albumin (BSA), highlighting the need 123 of detergent or enzyme stabilizing additives (Fig. S1B). We tested three previously 124 reported compounds under our assay conditions. The covalent inhibitor nirmatrelvir had 125 a similar IC50 as the reported potency 6 , whereas two non-covalent inhibitors (PET-UNK-126 29afea89-2 and VLA-UCB-1dbca3b4-15) had measured IC50 values 2 to 5-fold higher 127 compared to reported values 17 , likely due to different substrate and substrate 128 concentrations used in the published assays and those used here (Table S1). These 129 rates provide a reference for comparing the different inhibitors.  (bottom). The substrate contains the P4-P4′ NSP7 extended substrate sequence (blue), the fluorophore (yellow), the fluorescent quencher (purple), and the residues for increasing solubility (green). (B) A list of the viral polypeptide NSP sequences (P4-P4′) that are cleaved by MPro (left). The sequenceLOGO highlighting the substrate specificity of MPro, yielding a P4-P4′ consensus sequence: ATLQ(S/A)XXA (right). (C) The Michaelis-Menten kinetics for the NSP7 substrate with MPro yield parameters indicative of an optimized, efficient substrate. (D) SARS-CoV-2 MPro active site (PDB 6Y2G) 26 (green; sub-pockets S1′, S1, S2, S3, S4), shown here with substrate preferences (pink; P1′, P1, P2, P3, P4) (modeled after PDB 3SNE) 27 , was used to dock 1.2 billion non-covalent molecules and 6.5 million electrophile molecules. Top-ranked molecules were filtered and 395 were synthesized for in vitro testing. Some docking hits were prioritized for compound optimization, crystallography, pan-viral enzymatic activity, and cell-based antiviral activity. For C, experiments were performed in triplicate. for their ability to hydrogen bond with Gly143, His163, or Glu166, and make favorable 148 non-polar contacts with Met49 and Asp187. Ultimately, 220 molecules were prioritized, of 149 which 194 (88%) were successfully synthesized by Enamine. In a primary screen, 150 compounds were tested at a single concentration of 100 µM using the fluorescence-based 151 substrate cleavage assay and 19 showed >30% inhibition of enzyme activity and were 152 prioritized for full concentration-response curves. Overall, 12 molecules were defined as 153 hits, with IC50 values < 300 μM, for an overall hit rate of 6% (12 hits/194 molecules tested); 154 potencies ranged from 97 to 291 µM ( Table 1, Table S1, Fig. S2.1, Fig. S2.2).

156
As mentioned above, DMSO was observed to lower enzyme activity, consequently 157 the actives, initially tested from 10 mM DMSO stocks, were re-tested against MPro from warheads (nitrile, aldehyde) known to react with catalytic cysteines. Several initial docking 164 hits were tested for colloidal aggregation using dynamic light scattering (DLS) and off-165 target counter screens against malate dehydrogenase (MDH) and AmpC β-lactamase 32,33 166 ( Fig. S3). In DLS experiments, some scattering higher than 10 6 is observed indicating 167 potential aggregation. While a few compounds e.g., '3312 showed unspecific inhibition of 168 MDH, off-target activities were reversed by addition of 0.01% Triton X-100. As the MPro 169 enzymatic assay is run with 0.05% Tween-20, an even stronger disruptor of colloidal 170 aggregation than 0.01% Triton-X 100, we deemed the weak aggregation of these 171 compounds not relevant to their activity on Mpro. docking each analog into the MPro structure to assess complementarity.  analogs of each of the four inhibitors were selected for testing in the initial round of 182 optimization (Fig. 2, Table S1). For two initial hits, '0541 and '0273, more potent analogs 183 were identified in two to three rounds of this analog-by-catalog approach ( Table S1). The

184
'0273 analogs Z4924562413 and Z4946671001 had IC50 values of 13 µM and 5 µM, 185 respectively ( Fig. 2A). Analogs of the initial docking hit '0541, such as Z4929615577 and 186 Z4929616137, reached similar potencies of 10 µM and 8 µM, respectively (Fig. 2G). predicted pose, with an RMSD of 1.1 Å (Fig. 2E, Fig. S4). Here, the compound's 202 hydantoin core hydrogen bonds with the backbone amine of Glu166 and Gly143. In 203 addition, the crystal structure of MPro in complex with '6111 confirms the predicted biding 204 pose (RMSD = 1.4 Å) with the isoquinoline placed in the S1 subpocket and the 205 hydrophobic spirocyclic indane group occupying the S2 pocket (Fig. 2F, Fig. S4).     (Table S1). Initial docking 280 hits were evaluated for potential MPro inhibition through colloidal aggregation as 281 previously described for non-covalent docking hits (Fig. S3). Some higher DLS scattering 282 or non-specific inhibition is observed in the AmpC and MDH enzymatic assays. However, 283 adding 0.01% Triton X-100 in the MDH inhibition assay largely recovered enzymatic 284 activity and eliminated any non-specific inhibition thereby suggesting that the measured 285 activities in the detergent-containing MPro enzymatic assays are not caused by 286 aggregation (something also confirmed by subsequent crystallography, see below).

303
We sought to optimize several of the new covalent inhibitors, focusing on the 304 aldehyde '3620 with an IC50 of 55 µM (Table S1). These analogs were identified through 305 multiple strategies, including simply seeking readily available "make-on-demand" including Met49 and Phe181 (Fig. 4B, Fig. 4D, Fig. S8).  Table S3). 358 Meanwhile, no measurable antiviral activity was observed for the covalent aldehyde with an IC50 of 8 µΜ, similar to its SARS-CoV-2 MPro IC50 of 1 µM, however it was a 364 weaker inhibitor for the MERS MPro with an IC50 of 50 µM (Fig. 5B, Table S4). covering 37 different scaffold classes (Fig. 3, Table 1, Table 2). Of these, 15 inhibitors 377 in 3 scaffolds inhibited the enzyme with IC50 values less than 10 µM. The best covalent 378 inhibitor, '7021, was confirmed to act reversibly (Fig. S6), likely reflecting the fast-on/fast-379 off kinetics characteristic of aldehyde covalent inhibitors. We also present an optimized 380 MPro substrate for future inhibitor characterization (Fig. 1). corresponded to the docking predictions (Fig. 3, Fig. 4). Two of the new aldehyde These caveats should not distract from the key observations of this study. Large 407 library docking of both lead-like molecules and covalent electrophiles has revealed 11 408 scaffold families of MPro inhibitors (Fig. 3, Table 1, Table 2), the best of which act in the 409 low µM range (Fig. 2, Fig. 4). Whereas neither hit rates nor affinities rose to levels seen 410 against targets with well-defined binding sites, eight crystal structure of characteristic lead 411 molecules confirmed the docking poses (Fig. 2, Fig. 4), suggesting that, notwithstanding 412 the lower hit rates, when the docking was right it was right for the right reasons. These   The resuspended sample was sonicated for 5 mins or until lysis was complete. Sonicated Compounds were incubated with protease prior to substrate addition at 37 °C for 1 h.

600
For the aldehydes, the top 300,000 ranked molecules were evaluated for torsional 601 strain 31 , and those with a total torsional strain greater than 9.8 (around 3.7 incurred due to strain on atom types on the warhead and this was disregarded, therefore total energy 603 was 6) and single torsional strain greater than 2.5 were excluded (155,386 left In total 21 molecules were prioritized and 16 were successfully synthesized.

619
Make-on-demand synthesis. Non-covalent and covalent compounds purchased 620 from docking screens, as well as analogs, were synthesized by Enamine Ltd. (Table S1).

621
Purities of molecules were at least 90% and most active compounds were at least 95% 622 (based on LC/MS data) (Fig. S10). Regents. An open-source web-based version of the program is available without