Ebselen derivatives inhibit SARS-CoV-2 replication by inhibition of its essential proteins: PLpro and Mpro proteases, and nsp14 guanine N7-methyltransferase.

Proteases encoded by SARS-CoV-2 constitute a promising target for new therapies against COVID-19. SARS-CoV-2 main protease (Mpro, 3CLpro) and papain-like protease (PLpro) are responsible for viral polyprotein cleavage-a process crucial for viral survival and replication. Recently it was shown that 2-phenylbenzisoselenazol-3(2H)-one (ebselen), an organoselenium anti-inflammatory small-molecule drug, is a potent, covalent inhibitor of both the proteases and its potency was evaluated in enzymatic and antiviral assays. In this study, we screened a collection of 34 ebselen and ebselen diselenide derivatives for SARS-CoV-2 PLpro and Mpro inhibitors. Our studies revealed that ebselen derivatives are potent inhibitors of both the proteases. We identified three PLpro and four Mpro inhibitors superior to ebselen. Independently, ebselen was shown to inhibit the N7-methyltransferase activity of SARS-CoV-2 nsp14 protein involved in viral RNA cap modification. Hence, selected compounds were also evaluated as nsp14 inhibitors. In the second part of our work, we employed 11 ebselen analogues-bis(2-carbamoylaryl)phenyl diselenides-in biological assays to evaluate their anti-SARS-CoV-2 activity in Vero E6 cells. We present their antiviral and cytoprotective activity and also low cytotoxicity. Our work shows that ebselen, its derivatives, and diselenide analogues constitute a promising platform for development of new antivirals targeting the SARS-CoV-2 virus.

Structure-Based Design of Potent Peptidomimetic Inhibitors Covalently Targeting SARS-CoV-2 Papain-like Protease.

The papain-like protease (PLpro) of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) plays a critical role in the proteolytic processing of viral polyproteins and the dysregulation of the host immune response, providing a promising therapeutic target. Here, we report the structure-guide design of novel peptidomimetic inhibitors covalently targeting SARS-CoV-2 PLpro. The resulting inhibitors demonstrate submicromolar potency in the enzymatic assay (IC50 = 0.23 µM) and significant inhibition of SARS-CoV-2 PLpro in the HEK293T cells using a cell-based protease assay (EC50 = 3.61 µM). Moreover, an X-ray crystal structure of SARS-CoV-2 PLpro in complex with compound 2 confirms the covalent binding of the inhibitor to the catalytic residue cysteine 111 (C111) and emphasizes the importance of interactions with tyrosine 268 (Y268). Together, our findings reveal a new scaffold of SARS-CoV-2 PLpro inhibitors and provide an attractive starting point for further optimization.

Integrated data-driven and experimental approaches to accelerate lead optimization targeting SARS-CoV-2 main protease.

Identification of potential therapeutic candidates can be expedited by integrating computational modeling with domain aware machine learning (ML) models followed by experimental validation in an iterative manner. Generative deep learning models can generate thousands of new candidates, however, their physiochemical and biochemical properties are typically not fully optimized. Using our recently developed deep learning models and a scaffold as a starting point, we generated tens of thousands of compounds for SARS-CoV-2 Mpro that preserve the core scaffold. We utilized and implemented several computational tools such as structural alert and toxicity analysis, high throughput virtual screening, ML-based 3D quantitative structure-activity relationships, multi-parameter optimization, and graph neural networks on generated candidates to predict biological activity and binding affinity in advance. As a result of these combined computational endeavors, eight promising candidates were singled out and put through experimental testing using Native Mass Spectrometry and FRET-based functional assays. Two of the tested compounds with quinazoline-2-thiol and acetylpiperidine core moieties showed IC[Formula: see text] values in the low micromolar range: [Formula: see text] [Formula: see text]M and 3.41±0.0015 [Formula: see text]M, respectively. Molecular dynamics simulations further highlight that binding of these compounds results in allosteric modulations within the chain B and the interface domains of the Mpro. Our integrated approach provides a platform for data driven lead optimization with rapid characterization and experimental validation in a closed loop that could be applied to other potential protein targets.

Discovery of Chalcone-Based Hybrid Structures as High Affinity and Site-Specific Inhibitors against SARS-CoV-2: A Comprehensive Structural Analysis Based on Various Host-Based and Viral Targets.

Previous studies indicated that natural-based chalcones have significant inhibitory effects on the coronavirus enzymes 3CLpro and PLpro as well as modulation of some host-based antiviral targets (HBATs). In this study, a comprehensive computational and structural study was performed to investigate the affinity of our compound library consisting of 757 chalcone-based structures (CHA-1 to CHA-757) for inhibiting the 3CLpro and PLpro enzymes and against twelve selected host-based targets. Our results indicated that CHA-12 (VUF 4819) is the most potent and multi-target inhibitor in our chemical library over all viral and host-based targets. Correspondingly, CHA-384 and its congeners containing ureide moieties were found to be potent and selective 3CLpro inhibitors, and benzotriazole moiety in CHA-37 was found to be a main fragment for inhibiting the 3CLpro and PLpro. Surprisingly, our results indicate that the ureide and sulfonamide moieties are integral fragments for the optimum 3CLpro inhibition while occupying the S1 and S3 subsites, which is fully consistent with recent reports on the site-specific 3CLpro inhibitors. Finding the multi-target inhibitor CHA-12, previously reported as an LTD4 antagonist for the treatment of inflammatory pulmonary diseases, prompted us to suggest it as a concomitant agent for relieving respiratory symptoms and suppressing COVID-19 infection.

Host-specific sensing of coronaviruses and picornaviruses by the CARD8 inflammasome.

Hosts have evolved diverse strategies to respond to microbial infections, including the detection of pathogen-encoded proteases by inflammasome-forming sensors such as NLRP1 and CARD8. Here, we find that the 3CL protease (3CLpro) encoded by diverse coronaviruses, including Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), cleaves a rapidly evolving region of human CARD8 and activates a robust inflammasome response. CARD8 is required for cell death and the release of pro-inflammatory cytokines during SARS-CoV-2 infection. We further find that natural variation alters CARD8 sensing of 3CLpro, including 3CLpro-mediated antagonism rather than activation of megabat CARD8. Likewise, we find that a single nucleotide polymorphism (SNP) in humans reduces CARD8's ability to sense coronavirus 3CLpros and, instead, enables sensing of 3C proteases (3Cpro) from select picornaviruses. Our findings demonstrate that CARD8 is a broad sensor of viral protease activities and suggests that CARD8 diversity contributes to inter- and intraspecies variation in inflammasome-mediated viral sensing and immunopathology.

An interaction-based drug discovery screen explains known SARS-CoV-2 inhibitors and predicts new compound scaffolds.

The recent outbreak of the COVID-19 pandemic caused by severe acute respiratory syndrome-Coronavirus-2 (SARS-CoV-2) has shown the necessity for fast and broad drug discovery methods to enable us to react quickly to novel and highly infectious diseases. A well-known SARS-CoV-2 target is the viral main 3-chymotrypsin-like cysteine protease (Mpro), known to control coronavirus replication, which is essential for the viral life cycle. Here, we applied an interaction-based drug repositioning algorithm on all protein-compound complexes available in the protein database (PDB) to identify Mpro inhibitors and potential novel compound scaffolds against SARS-CoV-2. The screen revealed a heterogeneous set of 692 potential Mpro inhibitors containing known ones such as Dasatinib, Amodiaquine, and Flavin mononucleotide, as well as so far untested chemical scaffolds. In a follow-up evaluation, we used publicly available data published almost two years after the screen to validate our results. In total, we are able to validate 17% of the top 100 predictions with publicly available data and can furthermore show that predicted compounds do cover scaffolds that are yet not associated with Mpro. Finally, we detected a potentially important binding pattern consisting of 3 hydrogen bonds with hydrogen donors of an oxyanion hole within the active side of Mpro. Overall, these results give hope that we will be better prepared for future pandemics and that drug development will become more efficient in the upcoming years.

Drug Potency Prediction of SARS-CoV-2 Main Protease Inhibitors Based on a Graph Generative Model.

The prediction of a ligand potency to inhibit SARS-CoV-2 main protease (M-pro) would be a highly helpful addition to a virtual screening process. The most potent compounds might then be the focus of further efforts to experimentally validate their potency and improve them. A computational method to predict drug potency, which is based on three main steps, is defined: (1) defining the drug and protein in only one 3D structure; (2) applying graph autoencoder techniques with the aim of generating a latent vector; and (3) using a classical fitting model to the latent vector to predict the potency of the drug. Experiments in a database of 160 drug-M-pro pairs, from which the pIC50 is known, show the ability of our method to predict their drug potency with high accuracy. Moreover, the time spent to compute the pIC50 of the whole database is only some seconds, using a current personal computer. Thus, it can be concluded that a computational tool that predicts, with high reliability, the pIC50 in a cheap and fast way is achieved. This tool, which can be used to prioritize which virtual screening hits, will be further examined in vitro.

Porcine epidemic diarrhoea virus (PEDV) infection activates AMPK and JNK through TAK1 to induce autophagy and enhance virus replication.

Autophagy plays an important role in defending against invading microbes. However, numerous viruses can subvert autophagy to benefit their replication. Porcine epidemic diarrhoea virus (PEDV) is an aetiological agent that causes severe porcine epidemic diarrhoea. How PEDV infection regulates autophagy and its role in PEDV replication are inadequately understood. Herein, we report that PEDV induced complete autophagy in Vero and IPEC-DQ cells, as evidenced by increased LC3 lipidation, p62 degradation, and the formation of autolysosomes. The lysosomal protease inhibitors chloroquine (CQ) or bafilomycin A and Beclin-1 or ATG5 knockdown blocked autophagic flux and inhibited PEDV replication. PEDV infection activated AMP-activated protein kinase (AMPK) and c-Jun terminal kinase (JNK) by activating TGF-beta-activated kinase 1 (TAK1). Compound C (CC), an AMPK inhibitor, and SP600125, a JNK inhibitor, inhibited PEDV-induced autophagy and virus replication. AMPK activation led to increased ULK1^S777 phosphorylation and activation. Inhibition of ULK1 activity by SBI-0206965 (SBI) and TAK1 activity by 5Z-7-Oxozeaenol (5Z) or by TAK1 siRNA led to the suppression of autophagy and virus replication. Our study provides mechanistic insights into PEDV-induced autophagy and how PEDV infection leads to JNK and AMPK activation.

Eupatin, a Flavonoid, Inhibits Coronavirus 3CL Protease and Replication.

The coronavirus disease 2019 (COVID-19) pandemic has caused more than six million deaths worldwide since 2019. Although vaccines are available, novel variants of coronavirus are expected to appear continuously, and there is a need for a more effective remedy for coronavirus disease. In this report, we isolated eupatin from Inula japonica flowers and showed that it inhibits the coronavirus 3 chymotrypsin-like (3CL) protease as well as viral replication. We showed that eupatin treatment inhibits SARS-CoV-2 3CL-protease, and computational modeling demonstrated that it interacts with key residues of 3CL-protease. Further, the treatment decreased the number of plaques formed by human coronavirus OC43 (HCoV-OC43) infection and decreased viral protein and RNA levels in the media. These results indicate that eupatin inhibits coronavirus replication.

Identification of Darunavir Derivatives for Inhibition of SARS-CoV-2 3CLpro.

The effective antiviral agents that treat severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are urgently needed around the world. The 3C-like protease (3CLpro) of SARS-CoV-2 plays a pivotal role in virus replication; it also has become an important therapeutic target for the infection of SARS-CoV-2. In this work, we have identified Darunavir derivatives that inhibit the 3CLpro through a high-throughput screening method based on a fluorescence resonance energy transfer (FRET) assay in vitro. We found that the compounds 29# and 50# containing polyphenol and caffeine derivatives as the P2 ligand, respectively, exhibited favorable anti-3CLpro potency with EC50 values of 6.3 µM and 3.5 µM and were shown to bind to SARS-CoV-2 3CLpro in vitro. Moreover, we analyzed the binding mode of the DRV in the 3CLpro through molecular docking. Importantly, 29# and 50# exhibited a similar activity against the protease in Omicron variants. The inhibitory effect of compounds 29# and 50# on the SARS-CoV-2 3CLpro warrants that they are worth being the template to design functionally improved inhibitors for the treatment of COVID-19.

Docking-Based Evidence for the Potential of ImmunoDefender: A Novel Formulated Essential Oil Blend Incorporating Synergistic Antiviral Bioactive Compounds as Promising Mpro Inhibitors against SARS-CoV-2.

Essential oils (Eos) have demonstrated antiviral activity, but their toxicity can hinder their use as therapeutic agents. Recently, some essential oil components have been used within safe levels of acceptable daily intake limits without causing toxicity. The "ImmunoDefender," a novel antiviral compound made from a well-known mixture of essential oils, is considered highly effective in treating SARS-CoV-2 infections. The components and doses were chosen based on existing information about their structure and toxicity. Blocking the main protease (Mpro) of SARS-CoV-2 with high affinity and capacity is critical for inhibiting the virus's pathogenesis and transmission. In silico studies were conducted to examine the molecular interactions between the main essential oil components in "ImmunoDefender" and SARS-CoV-2 Mpro. The screening results showed that six key components of ImmunoDefender formed stable complexes with Mpro via its active catalytic site with binding energies ranging from -8.75 to -10.30 kcal/mol, respectively for Cinnamtannin B1, Cinnamtannin B2, Pavetannin C1, Syzyginin B, Procyanidin C1, and Tenuifolin. Furthermore, three essential oil bioactive inhibitors, Cinnamtannin B1, Cinnamtannin B2, and Pavetannin C, had significant ability to bind to the allosteric site of the main protease with binding energies of -11.12, -10.74, and -10.79 kcal/mol; these results suggest that these essential oil bioactive compounds may play a role in preventing the attachment of the translated polyprotein to Mpro, inhibiting the virus's pathogenesis and transmission. These components also had drug-like characteristics similar to approved and effective drugs, suggesting that further pre-clinical and clinical studies are needed to confirm the generated in silico outcomes.

Evaluation of the Inhibition Potency of Nirmatrelvir against Main Protease Mutants of SARS-CoV-2 Variants.

SARS-CoV-2 continues to pose a threat to public health. Main protease (Mpro) is one of the most lucrative drug targets for developing specific antivirals against SARS-CoV-2 infection. By targeting Mpro, peptidomimetic nirmatrelvir is able to inhibit viral replication of SARS-CoV-2 and reduce the risk for progression to severe COVID-19. However, multiple mutations in the gene encoding Mpro of emerging SARS-CoV-2 variants raise a concern of drug resistance. In the present study, we expressed 16 previously reported SARS-CoV-2 Mpro mutants (G15S, T25I, T45I, S46F, S46P, D48N, M49I, L50F, L89F, K90R, P132H, N142S, V186F, R188K, T190I, and A191V). We evaluated the inhibition potency of nirmatrelvir against these Mpro mutants and solved the crystal structures of representative Mpro mutants of SARS-CoV-2 bound to nirmatrelvir. Enzymatic inhibition assays revealed that these Mpro variants remain susceptible to nirmatrelvir as the wildtype. Detailed analysis and structural comparison provided the inhibition mechanism of Mpro mutants by nirmatrelvir. These results informed the ongoing genomic surveillance of drug resistance of emerging SARS-CoV-2 variants to nirmatrelvir and facilitate the development of next-generation anticoronavirus drugs.

Multiscale Simulations of the Covalent Inhibition of the SARS-CoV-2 Main Protease: Four Compounds and Three Reaction Mechanisms.

We report the results of computational modeling of the reactions of the SARS-CoV-2 main protease (MPro) with four potential covalent inhibitors. Two of them, carmofur and nirmatrelvir, have shown experimentally the ability to inhibit MPro. Two other compounds, X77A and X77C, were designed computationally in this work. They were derived from the structure of X77, a non-covalent inhibitor forming a tight surface complex with MPro. We modified the X77 structure by introducing warheads capable of reacting with the catalytic cysteine residue in the MPro active site. The reaction mechanisms of the four molecules with MPro were investigated by quantum mechanics/molecular mechanics (QM/MM) simulations. The results show that all four compounds form covalent adducts with the catalytic cysteine Cys 145 of MPro. From the chemical perspective, the reactions of these four molecules with MPro follow three distinct mechanisms. The reactions are initiated by a nucleophilic attack of the thiolate group of the deprotonated cysteine residue from the catalytic dyad Cys145-His41 of MPro. In the case of carmofur and X77A, the covalent binding of the thiolate to the ligand is accompanied by the formation of the fluoro-uracil leaving group. The reaction with X77C follows the nucleophilic aromatic substitution SNAr mechanism. The reaction of MPro with nirmatrelvir (which has a reactive nitrile group) leads to the formation of a covalent thioimidate adduct with the thiolate of the Cys145 residue in the enzyme active site. Our results contribute to the ongoing search for efficient inhibitors of the SARS-CoV-2 enzymes.

Protein-ligand interactions from a quantum fragmentation perspective: The case of the SARS-CoV-2 main protease interacting with α-ketoamide inhibitors.

We present a hybrid, multi-method, computational scheme for protein/ligand systems well suited to be used on modern and forthcoming massively parallel computing systems. The scheme relies on a multi-scale polarizable molecular modeling, approach to perform molecular dynamics simulations, and on an efficient Density Functional Theory (DFT) linear scaling method to post-process simulation snapshots. We use this scheme to investigate recent α-ketoamide inhibitors targeting the main protease of the SARS-CoV-2 virus. We assessed the reliability and the coherence of the hybrid scheme, in particular, by checking the ability of MM and DFT to reproduce results from high-end ab initio computations regarding such inhibitors. The DFT approach enables an a posteriori fragmentation of the system and an investigation into the strength of interaction among identified fragment pairs. We show the necessity of accounting for a large set of plausible protease/inhibitor conformations to generate reliable interaction data. Finally, we point out ways to further improve α-ketoamide inhibitors to more strongly interact with particular protease domains neighboring the active site.

Aurones: A Promising Scaffold to Inhibit SARS-CoV-2 Replication.

Aurones are a small subgroup of flavonoids in which the basic C6-C3-C6 skeleton is arranged as (Z)-2-benzylidenebenzofuran-3(2H)-one. These compounds are structural isomers of flavones and flavonols, natural products reported as potent inhibitors of SARS-CoV-2 replication. Herein, we report the design, synthesis, and anti-SARS-CoV-2 activity of a series of 25 aurones bearing different oxygenated groups (OH, OCH3, OCH2OCH3, OCH2O, OCF2H, and OCH2C6H4R) at the A- and/or B-rings using cell-based screening assays. We observed that 12 of the 25 compounds exhibit EC50 < 3 µM (8e, 8h, 8j, 8k, 8l, 8m, 8p, 8q, 8r, 8w, 8x, and 8y), of which five presented EC50 < 1 µM (8h, 8m, 8p, 8q, and 8w) without evident cytotoxic effect in Calu-3 cells. The substitution of the A- and/or B-ring with OCH3, OCH2OCH3, and OCF2H groups seems beneficial for the antiviral activity, while the corresponding phenolic derivatives showed a significant decrease in the anti-SARS-CoV-2 activity. The most potent compound of the series, aurone 8q (EC50 = 0.4 µM, SI = 2441.3), is 2 to 3 times more effective than the polyphenolic flavonoids myricetin (2) and baicalein (1), respectively. Investigation of the five more active compounds as inhibitors of SARS-CoV-2 3CLpro based on molecular dynamic calculations suggested that these aurones should detach from the active site of 3CLpro, and, probably, they could bind to another SARS-CoV-2 protein target (either receptor or enzyme).

Targeting SARS-CoV-2 Main Protease: A Successful Story Guided by an In Silico Drug Repurposing Approach.

The SARS-CoV-2 main protease (Mpro) is a crucial enzyme for viral replication and has been considered an attractive drug target for the treatment of COVID-19. In this study, virtual screening techniques and in vitro assays were combined to identify novel Mpro inhibitors starting from around 8000 FDA-approved drugs. The docking analysis highlighted 17 promising best hits, biologically characterized in terms of their Mpro inhibitory activity. Among them, 7 cephalosporins and the oral anticoagulant betrixaban were able to block the enzyme activity in the micromolar range with no cytotoxic effect at the highest concentration tested. After the evaluation of the degree of conservation of Mpro residues involved in the binding with the studied ligands, the ligands' activity on SARS-CoV-2 replication was assessed. The ability of betrixaban to affect SARS-CoV-2 replication associated to its antithrombotic effect could pave the way for its possible use in the treatment of hospitalized COVID-19 patients.

Elucidating Atomistic Insight into the Dynamical Responses of the SARS-CoV-2 Main Protease for the Binding of Remdesivir Analogues: Leveraging Molecular Mechanics To Decode the Inhibition Mechanism.

To combat mischievous coronavirus disease followed by continuous upgrading of therapeutic strategy against the antibody-resistant variants, the molecular mechanistic understanding of protein-drug interactions is a prerequisite in the context of target-specific rational drug development. Herein, we attempt to decipher the structural basis for the inhibition of SARS-CoV-2 main protease (Mpro) through the elemental analysis of potential energy landscape and the associated thermodynamic and kinetic properties of the enzyme-inhibitor complexes using automated molecular docking calculations in conjunction with classical force field-based molecular dynamics (MD) simulations. The crux of the scalable all-atom MD simulations consummated in explicit solvent media is to capture the structural plasticity of the viral enzyme due to the binding of remdesivir analogues and ascertain the subtle interplay of noncovalent interactions in stabilizing specific conformational states of the receptor that controls the biomolecular processes related to the ligand binding and dissociation kinetics. To unravel the critical role of modulation of the ligand scaffold, we place further emphasis on the estimation of binding free energy as well as the energy decomposition analysis by employing the generalized Born and Poisson-Boltzmann models. The estimated binding affinities are found to vary between -25.5 and -61.2 kcal/mol. Furthermore, the augmentation of inhibitory efficacy of the remdesivir analogue crucially stems from the van der Waals interactions with the active site residues of the protease. The polar solvation energy contributes unfavorably to the binding free energy and annihilates the contribution of electrostatic interactions as derived from the molecular mechanical energies.

The research progress of SARS-CoV-2 main protease inhibitors from 2020 to 2022.

The novel coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread worldwide. The main protease (Mpro) of SARS-CoV-2 plays a central role in viral replication and transcription and represents an attractive drug target for fighting COVID-19. Many SARS-CoV-2 Mpro inhibitors have been reported, including covalent and noncovalent inhibitors. The SARS-CoV-2 Mpro inhibitor PF-07321332 (Nirmatrelvir) designed by Pfizer has been put on the market. This paper briefly introduces the structural characteristics of SARS-CoV-2 Mpro and summarizes the research progress of SARS-CoV-2 Mpro inhibitors from the aspects of drug repurposing and drug design. These information will provide a basis for the drug development of treating the infection of SARS-CoV-2 and even other coronaviruses in the future.

In vitro selection of macrocyclic peptide inhibitors containing cyclic γ2,4-amino acids targeting the SARS-CoV-2 main protease.

γ-Amino acids can play important roles in the biological activities of natural products; however, the ribosomal incorporation of γ-amino acids into peptides is challenging. Here we report how a selection campaign employing a non-canonical peptide library containing cyclic γ2,4-amino acids resulted in the discovery of very potent inhibitors of the SARS-CoV-2 main protease (Mpro). Two kinds of cyclic γ2,4-amino acids, cis-3-aminocyclobutane carboxylic acid (γ1) and (1R,3S)-3-aminocyclopentane carboxylic acid (γ2), were ribosomally introduced into a library of thioether-macrocyclic peptides. One resultant potent Mpro inhibitor (half-maximal inhibitory concentration = 50 nM), GM4, comprising 13 residues with γ1 at the fourth position, manifests a 5.2 nM dissociation constant. An Mpro:GM4 complex crystal structure reveals the intact inhibitor spans the substrate binding cleft. The γ1 interacts with the S1' catalytic subsite and contributes to a 12-fold increase in proteolytic stability compared to its alanine-substituted variant. Knowledge of interactions between GM4 and Mpro enabled production of a variant with a 5-fold increase in potency.

tinyIFD: A High-Throughput Binding Pose Refinement Workflow Through Induced-Fit Ligand Docking.

A critical step in structure-based drug discovery is predicting whether and how a candidate molecule binds to a model of a therapeutic target. However, substantial protein side chain movements prevent current screening methods, such as docking, from accurately predicting the ligand conformations and require expensive refinements to produce viable candidates. We present the development of a high-throughput and flexible ligand pose refinement workflow, called "tinyIFD". The main features of the workflow include the use of specialized high-throughput, small-system MD simulation code mdgx.cuda and an actively learning model zoo approach. We show the application of this workflow on a large test set of diverse protein targets, achieving 66% and 76% success rates for finding a crystal-like pose within the top-2 and top-5 poses, respectively. We also applied this workflow to the SARS-CoV-2 main protease (Mpro) inhibitors, where we demonstrate the benefit of the active learning aspect in this workflow.