Design Of Covalent Inhibitors Research Articles

Total syntheses of cyclohelminthol I-IV reveal a new cysteine-selective covalent reactive group.

Biocompatible covalent reactive groups (CRGs) play pivotal roles in several areas of chemical biology and the life sciences, including targeted covalent inhibitor design and preparation of advanced biologic drugs, such as antibody-drug conjugates. In this study, we present the discovery that the small, chlorinated polyketide natural product cyclohelminthiol II (CHM-II) acts as a new type of cysteine/thiol-targeting CRG incorporating both reversible and irreversible reactivity. We devise the first syntheses of four simple cyclohelminthols, (±)-cyclohelminthol I-IV, with selective chlorinations (at C2 and C5) and a Ni-catalyzed reductive cross coupling between an enone, a vinyl bromide and triethylsilyl chloride as the key steps. Unbiased biological profiling (cell painting) was used to discover a putative covalent mechanism for CHM-II in cells with subsequent validation experiments demonstrating mechanistic similarity to dimethyl fumarate (DMF) - a known (covalent) drug used in the treatment of multiple sclerosis. Focused biochemical experiments revealed divergent thiol-reactivity inherent to the CHM-II scaffold and through further chemical derivatization of CHM-II we applied activity-based protein profiling (ABPP)-workflows to show exclusive cysteine-labelling in cell lysate. Overall, this study provides both efficient synthetic access to the CHM-II chemotype - and neighboring chemical space - and proof-of-concept for several potential applications of this new privileged CRG-class within covalent chemical biology.

The study of halogen effect on the reactivity of the serine-targeting covalent warheads.

Halogens favorably contributes to the drug potency and metabolic stability via electrostatic interactions. Herein, the halogen effects on the reactivity of the halogenated 2,2,2-trifluoroacetophenones as serine-targeting covalent warheads were investigated. Our results showed that introducing halogen atoms, especially Cl or Br, into the phenyl scaffold would influence the electron density around the ring, which led to different time-dependent inhibition response to the target serine hydrolase (hCES1A). Co-crystallography analysis not only verified that halogenated molecules preferred to form covalent adducts, but also provided the conformational information for the design of covalent inhibitors targeting to hCES1A protein for the treatment of drug-induced acute enteritis.

Targeting PfCLK3 with Covalent Inhibitors: A Novel Strategy for Malaria Treatment.

Malaria still causes over 600,000 deaths annually, with rising resistance to frontline drugs by Plasmodium falciparum increasing this number each year. New medicines with novel mechanisms of action are, therefore, urgently needed. In this work, we solved the cocrystal structure of the essential malarial kinase PfCLK3 with the reversible inhibitor TCMDC-135051 (1), enabling the design of covalent inhibitors targeting a unique cysteine residue (Cys368) poorly conserved in the human kinome. Chloroacetamide 4 shows nanomolar potency and covalent inhibition in both recombinant protein and P. falciparum assays. Efficacy in parasites persisted after a 6 h washout, indicating an extended duration of action. Additionally, 4 showed improved kinase selectivity and a high selectivity index against HepG2 cells, with a low propensity for resistance (log MIR > 8.1). To our knowledge, compound 4 is the first covalent inhibitor of a malarial kinase, offering promising potential as a lead for a single-dose malaria cure.

CovalentInDB 2.0: an updated comprehensive database for structure-based and ligand-based covalent inhibitor design and screening.

The rational design of targeted covalent inhibitors (TCIs) has emerged as a powerful strategy in drug discovery, known for its ability to achieve strong binding affinity and prolonged target engagement. However, the development of covalent drugs is often challenged by the need to optimize both covalent warhead and non-covalent interactions, alongside the limitations of existing compound libraries. To address these challenges, we present CovalentInDB 2.0, an updated online database designed to support covalent drug discovery. This updated version includes 8303 inhibitors and 368 targets, supplemented by 3445 newly added cocrystal structures, providing detailed analyses of non-covalent interactions. Furthermore, we have employed an AI-based model to profile the ligandability of 144 864 cysteines across the human proteome. CovalentInDB 2.0 also features the largest covalent virtual screening library with 2 030 192 commercially available compounds and a natural product library with 105 901 molecules, crucial for covalent drug screening and discovery. To enhance the utility of these compounds, we performed structural similarity analysis and drug-likeness predictions. Additionally, a new user data upload feature enables efficient data contribution and continuous updates. CovalentInDB 2.0 is freely accessible at http://cadd.zju.edu.cn/cidb/.

Reactivities of acrylamide warheads toward cysteine targets: a QM/ML approach to covalent inhibitor design.

Covalent inhibition offers many advantages over non-covalent inhibition, but covalent warhead reactivity must be carefully balanced to maintain potency while avoiding unwanted side effects. While warhead reactivities are commonly measured with assays, a computational model to predict warhead reactivities could be useful for several aspects of the covalent inhibitor design process. Studies have shown correlations between covalent warhead reactivities and quantum mechanic (QM) properties that describe important aspects of the covalent reaction mechanism. However, the models from these studies are often linear regression equations and can have limitations associated with their usage. Applications of machine learning (ML) models to predict covalent warhead reactivities with QM descriptors are not extensively seen in the literature. This study uses QM descriptors, calculated at different levels of theory, to train ML models to predict reactivities of covalent acrylamide warheads. The QM/ML models are compared with linear regression models built upon the same QM descriptors and with ML models trained on structure-based features like Morgan fingerprints and RDKit descriptors. Experiments show that the QM/ML models outperform the linear regression models and the structure-based ML models, and literature test sets demonstrate the power of the QM/ML models to predict reactivities of unseen acrylamide warhead scaffolds. Ultimately, these QM/ML models are effective, computationally feasible tools that can expedite the design of new covalent inhibitors.

Discovery of a Tunable Heterocyclic Electrophile 4-Chloro-pyrazolopyridine That Defines a Unique Subset of Ligandable Cysteines.

Electrophilic small molecules with novel reactivity are powerful tools that enable activity-based protein profiling and covalent inhibitor discovery. Here, we report a reactive heterocyclic scaffold, 4-chloro-pyrazolopyridine (CPzP) for selective modification of proteins via a nucleophilic aromatic substitution (SNAr) mechanism. Chemoproteomic profiling reveals that CPzPs engage cysteines within functionally diverse protein sites including ribosomal protein S5 (RPS5), inosine monophosphate dehydrogenase 2 (IMPDH2), and heat shock protein 60 (HSP60). Through the optimization of appended recognition elements, we demonstrate the utility of CPzP for covalent inhibition of prolyl endopeptidase (PREP) by targeting a noncatalytic active-site cysteine. This study suggests that the proteome reactivity of CPzPs can be modulated by both electronic and steric features of the ring system, providing a new tunable electrophile for applications in chemoproteomics and covalent inhibitor design.

Development of Highly Potent and Selective Covalent FGFR4 Inhibitors Based on SNAr Electrophiles.

Fibroblast growth factor receptor 4 (FGFR4) is thought to be a driver in several cancer types, most notably in hepatocellular carcinoma. One way to achieve high potency and isoform selectivity for FGFR4 is covalently targeting a rare cysteine (C552) in the hinge region of its kinase domain that is not present in other FGFR family members (FGFR1-3). Typically, this cysteine is addressed via classical acrylamide electrophiles. We demonstrate that noncanonical covalent "warheads" based on nucleophilic aromatic substitution (SNAr) chemistry can be employed in a rational manner to generate highly potent and (isoform-)selective FGFR4 inhibitors with a low intrinsic reactivity. Key compounds showed low to subnanomolar potency, efficient covalent inactivation kinetics, and excellent selectivity against the other FGFRs, the kinases with an equivalent cysteine, and a representative subset of the kinome. Moreover, these compounds achieved nanomolar potencies in cellular assays and demonstrated good microsomal stability, highlighting the potential of SNAr-based approaches in covalent inhibitor design.

Machine Learning Models to Interrogate Proteome-Wide Covalent Ligandabilities Directed at Cysteines.

Machine learning (ML) identification of covalently ligandable sites may accelerate targeted covalent inhibitor design and help expand the druggable proteome space. Here, we report the rigorous development and validation of the tree-based models and convolutional neural networks (CNNs) trained on a newly curated database (LigCys3D) of over 1000 liganded cysteines in nearly 800 proteins represented by over 10,000 three-dimensional structures in the protein data bank. The unseen tests yielded 94 and 93% area under the receiver operating characteristic curves for the tree models and CNNs, respectively. Based on the AlphaFold2 predicted structures, the ML models recapitulated the newly liganded cysteines in the PDB with over 90% recall values. To assist the community of covalent drug discoveries, we report the predicted ligandable cysteines in 392 human kinases and their locations in the sequence-aligned kinase structure, including the PH and SH2 domains. Furthermore, we disseminate a searchable online database LigCys3D (https://ligcys.computchem.org/) and a web prediction server DeepCys (https://deepcys.computchem.org/), both of which will be continuously updated and improved by including newly published experimental data. The present work represents an important step toward the ML-led integration of big genome data and structure models to annotate the human proteome space for the next-generation covalent drug discoveries.

Constructing Innovative Covalent and Noncovalent Compound Libraries: Insights from 3D Protein-Ligand Interactions.

Noncovalent interactions between small-molecule drugs and protein targets assume a pivotal role in drug design. Moreover, the design of covalent inhibitors, forming covalent bonds with amino acid residues, requires rational reactivity for their covalent warheads, presenting a key challenge as well. Understanding the intricacies of these interactions provides a more comprehensive understanding of molecular binding mechanisms, thereby guiding the rational design of potent inhibitors. In this study, we adopted the fragment-based drug design approach, introducing a novel methodology to extract noncovalent and covalent fragments according to distinct three-dimensional (3D) interaction modes from noncovalent and covalent compound libraries. Additionally, we systematically replaced existing ligands with rational fragment substitutions, based on the spatial orientation of fragments in 3D space. Furthermore, we adopted a molecular generation approach to create innovative covalent inhibitors. This process resulted in the recombination of a noncovalent compound library and several covalent compound libraries, constructed by two commonly encountered covalent amino acids: cysteine and serine. We utilized noncovalent ligands in KLIFS and covalent ligands in CovBinderInPDB as examples to recombine noncovalent and covalent libraries. These recombined compound libraries cover a substantial portion of the chemical space present in the original compound libraries and exhibit superior performance in terms of molecular scaffold diversity compared to the original compound libraries and other 11 commercial libraries. We also recombined BTK-focused libraries, and 23 compounds within our libraries have been validated by former researchers to possess potential biological activity. The establishment of these compound libraries provides valuable resources for virtual screening of covalent and noncovalent drugs targeting similar molecular targets.

Exploring Protein Bioconjugation: A Redox-Based Strategy for Tryptophan Targeting.

Amino acid bioconjugation technology has emerged as a pivotal tool for linking small-molecule fragments with proteins, antibodies, and even cells. The study in Nature by Chang and Toste introduces a redox-based strategy for tryptophan bioconjugation, employing N-sulfonyloxaziridines as oxidative cyclization reagents, demonstrating high efficiency comparable to traditional click reactions. Meanwhile, this tool provides feasible methods for investigating the mechanisms underlying functional tryptophan-related biochemical processes, paving the way for protein function exploration, activity-based proteomics for functional amino acid identification and characterization, and even the design of covalent inhibitors.

Danshensu inhibits SARS-CoV-2 by targeting its main protease as a specific covalent inhibitor and discovery of bifunctional compounds eliciting antiviral and anti-inflammatory activity

The coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has posed a serious threat to human. Since there are still no effective treatment options against the new emerging variants of SARS-CoV-2, it is necessary to devote a continuous endeavor for more targeted drugs and the preparation for the next pandemic. Salvia miltiorrhiza and its active ingredients possess wide antiviral activities, including against SARS-CoV-2. Danshensu, as one of the most important active ingredients in Salvia miltiorrhiza, has been reported to inhibit the entry of SARS-CoV-2 into ACE2 (angiotensin-converting enzyme 2)-overexpressed HEK-293T cells and Vero-E6 cells. However, there is a paucity of information regarding its detailed target and mechanism against SARS-CoV-2. Here, we present Danshensu as a covalent inhibitor of 3-chymotrypsin-like protease (3CLpro) against SARS-CoV-2 by the time-dependent inhibition assay (TDI) and mass spectrometry analysis. Further molecular docking, site-directed mutagenesis, circular dichroism (CD) and fluorescence spectra revealed that Danshensu covalently binds to C145 of SARS-CoV-2 3CLpro, meanwhile forming the hydrogen bonds with S144, H163 and E166 in the S1 site. Structure-based optimization of Danshensu led to the discovery of the promising compounds with good inhibitory activity and microsomal stability in vitro. Due to Danshensu inhibiting lung inflammation in the mouse model, we found that Danshensu derivatives also showed better anti-inflammatory activity than Danshensu in lipopolysaccharide (LPS)-stimulated RAW264.7 macrophage cells. Thus, our study provides not only the clue of the efficacy of Salvia miltiorrhiza against SARS-CoV-2, but also a detailed mechanistic insight into the covalent mode of action of Danshensu for design of covalent inhibitors against SARS-CoV-2 3CLpro, highlighting its potential as a bifunctional molecule with antivirus and anti-inflammation.

Development of covalent inhibitors: Principle, design, and application in cancer

AbstractCovalent inhibitors have been a rapidly growing field in drug discovery due to their therapeutic potential and unique advantages in cancer therapy. As opposed to noncovalent inhibitory drugs, covalent inhibitors reversibly or irreversibly modify proximal nucleophilic amino acid residues on proteins, aiming to selectively recognize and bind to protein targets and addressing some of the challenges faced by noncovalent drugs. Most successful targeted covalent inhibitors depend primarily on binding‐site cysteine residues, but this has limitations for certain protein targets that lack targetable cysteine residues. Recently, the rational design of covalent inhibitors or covalent probes targeting other nucleophilic residues, such as lysine, tyrosine, serine, has turned out to be another promising strategy for cancer therapy. Thus, the development of novel strategies to extend the scope of covalent binding and improve the binding properties is required. This review gives a summary of the development of covalent inhibitors targeting noncysteine from different aspects, including target identification, structure–activity relationships, drug discovery strategies, and binding properties, in the hope of providing a scientific reference for future covalent drug discovery as a means of expanding research in cancer therapy.

Discovery of a first-in-class Aurora A covalent inhibitor for the treatment of triple negative breast cancer

Aurora kinases, which belong to the serine/threonine protein family, play critical roles in the regulation of the cell cycle and mitotic spindle assembly. They are frequently highly expressed in various types of tumors, and the use of selective Aurora kinase inhibitors has become a potential treatment option for cancer therapy. Despite the development of some reversible Aurora kinase inhibitors, none has been approved for clinical use yet. In this study, we report the discovery of the first-in-class irreversible Aurora A covalent inhibitors that target a cysteine residue at the substrate binding site. These inhibitors were characterized in enzymatic and cellular assays, and 11c exhibited selective inhibition to normal and cancer cells, as well as to Aurora A and B kinases. The covalent binding of 11c to Aurora A was confirmed by SPR, MS, and enzyme kinetic analysis, and Cys290-mediated covalent inhibition was supported through a bottom-up analysis of inhibitor-modified targets. Moreover, Western blotting assays were conducted on cells and tissues, and cellular thermal shift assays (CETSA) were further performed on cells to demonstrate the selectivity to Aurora A kinase. 11c displayed comparable therapeutic efficacy in an MDA-MB-231 xenograft mouse model relative to the positive control ENMD-2076, while requiring only half the dose of ENMD-2076. These results confirmed that 11c may be a promising drug candidate for the treatment of triple negative breast cancer (TNBC). Our work may provide a new perspective on the design of covalent inhibitors of Aurora kinase.

Docking and Selectivity Studies of Covalently Bound Janus Kinase 3 Inhibitors.

The Janus kinases (JAKs) are a family of non-receptor cytosolic protein kinases critical for immune signaling. Many covalently bound ligands of JAK3 inhibitors have been reported. To help design selective JAK inhibitors, in this paper, we used five model proteins to study the subtype selectivity of and the mutational effects on inhibitor binding. We also compared the Covalent Dock programs from the Schrodinger software suite and the MOE software suite to determine which method to use for the drug design of covalent inhibitors. Our results showed that the docking affinity from 4Z16 (JAK3 wild-type model), 4E4N (JAK1), 4D1S (JAK2), and 7UYT (TYK2) from the Schrödinger software suite agreed well with the experimentally derived binding free energies with small predicted mean errors. However, the data from the mutant 5TTV model using the Schrödinger software suite yielded relatively large mean errors, whereas the MOE Covalent Dock program gave small mean errors in both the wild-type and mutant models for our model proteins. The docking data revealed that Leu905 of JAK3 and the hydrophobic residue at the same position in different subtypes (Leu959 of JAK1, Leu932 of JAK2, and Val981 of TYK2) is important for ligand binding to the JAK proteins. Arg911 and Asp912 of JAK3, Asp939 of JAK2, and Asp988 of TYK2 can be used for selective binding over JAK1, which contains Lys965 and Glu966 at the respective positions. Asp1021, Asp1039, and Asp1042 can be utilized for JAK1-selective ligand design, whereas Arg901 and Val981 may help guide TYK2-selective molecule design.

Mechanistic investigation of SARS-CoV-2 main protease to accelerate design of covalent inhibitors

Targeted covalent inhibition represents one possible strategy to block the function of SARS-CoV-2 Main Protease (MPRO), an enzyme that plays a critical role in the replication of the novel SARS-CoV-2. Toward the design of covalent inhibitors, we built a covalent inhibitor dataset using deep learning models followed by high throughput virtual screening of these candidates against MPRO. Two top-ranking inhibitors were selected for mechanistic investigations—one with an activated ester warhead that has a piperazine core and the other with an acrylamide warhead. Specifically, we performed a detailed analysis of the free energetics of covalent inhibition by hybrid quantum mechanics/molecular mechanics simulations. Cleavage of a fragment of the non-structured protein (NSP) from the SARS-CoV-2 genome was also simulated for reference. Simulations show that both candidates form more stable enzyme-inhibitor (E-I) complexes than the chosen NSP. It was found that both the NSP fragment and the activated ester inhibitor react with CYS145 of MPRO in a concerted manner, whereas the acrylamide inhibitor follows a stepwise mechanism. Most importantly, the reversible reaction and the subsequent hydrolysis reaction from E-I complexes are less probable when compared to the reactions with an NSP fragment, showing promise for these candidates to be the base for efficient MPRO inhibitors.

Rational Tuning of the Reactivity of Three-Membered Heterocycle Ring Openings via SN 2 Reactions.

The development of small-molecule covalent inhibitors and probes continuously pushes the rapidly evolving field of chemical biology forward. A key element in these molecular tool compounds is the "electrophilic trap" that allows a covalent linkage with the target enzyme. The reactivity of this entity needs to be well balanced to effectively trap the desired enzyme, while not being attacked by off-target nucleophiles. Here we investigate the intrinsic reactivity of substrates containing a class of widely used electrophilic traps, the three-membered heterocycles with a nitrogen (aziridine), phosphorus (phosphirane), oxygen (epoxide) or sulfur atom (thiirane) as heteroatom. Using quantum chemical approaches, we studied the conformational flexibility and nucleophilic ring opening of a series of model substrates, in which these electrophilic traps are mounted on a cyclohexene scaffold (C6 H10 Y with Y=NH, PH, O, S). It was revealed that the activation energy of the ring opening does not necessarily follow the trend that is expected from C-Y leaving-group bond strength, but steeply decreases from Y=NH, to PH, to O, to S. We illustrate that the HOMONu -LUMOSubstrate interaction is an all-important factor for the observed reactivity. In addition, we show that the activation energy of aziridines and phosphiranes can be tuned far below that of the corresponding epoxides and thiiranes by the addition of proper electron-withdrawing ring substituents. Our results provide mechanistic insights to rationally tune the reactivity of this class of popular electrophilic traps and can guide the experimental design of covalent inhibitors and probes for enzymatic activity.

Mechanistic Modeling of Monoglyceride Lipase Covalent Modification Elucidates the Role of Leaving Group Expulsion and Discriminates Inhibitors with High and Low Potency.

Inhibition of monoglyceride lipase (MGL), also known as monoacylglycerol lipase (MAGL), has emerged as a promising approach for treating neurological diseases. To gain useful insights in the design of agents with balanced potency and reactivity, we investigated the mechanism of MGL carbamoylation by the reference triazole urea SAR629 (IC50 = 0.2 nM) and two recently described inhibitors featuring a pyrazole (IC50 = 1800 nM) or a 4-cyanopyrazole (IC50 = 8 nM) leaving group (LG), using a hybrid quantum mechanics/molecular mechanics (QM/MM) approach. Opposite to what was found for substrate 2-arachidonoyl-sn-glycerol (2-AG), covalent modification of MGL by azole ureas is controlled by LG expulsion. Simulations indicated that changes in the electronic structure of the LG greatly affect reaction energetics with triazole and 4-cyanopyrazole inhibitors following a more accessible carbamoylation path compared to the unsubstituted pyrazole derivative. The computational protocol provided reaction barriers able to discriminate between MGL inhibitors with different potencies. These results highlight how QM/MM simulations can contribute to elucidating structure–activity relationships and provide insights for the design of covalent inhibitors.

Fast and Effective Prediction of the Absolute Binding Free Energies of Covalent Inhibitors of SARS-CoV-2 Main Protease and 20S Proteasome.

The COVID-19 pandemic has been a public health emergency with continuously evolving deadly variants around the globe. Among many preventive and therapeutic strategies, the design of covalent inhibitors targeting the main protease (Mpro) of SARS-CoV-2 that causes COVID-19 has been one of the hotly pursued areas. Currently, about 30% of marketed drugs that target enzymes are covalent inhibitors. Such inhibitors have been shown in recent years to have many advantages that counteract past reservation of their potential off-target activities, which can be minimized by modulation of the electrophilic warhead and simultaneous optimization of nearby noncovalent interactions. This process can be greatly accelerated by exploration of binding affinities using computational models, which are not well-established yet due to the requirement of capturing the chemical nature of covalent bond formation. Here, we present a robust computational method for effective prediction of absolute binding free energies (ABFEs) of covalent inhibitors. This is done by integrating the protein dipoles Langevin dipoles method (in the PDLD/S-LRA/β version) with quantum mechanical calculations of the energetics of the reaction of the warhead and its amino acid target, in water. This approach evaluates the combined effects of the covalent and noncovalent contributions. The applicability of the method is illustrated by predicting the ABFEs of covalent inhibitors of SARS-CoV-2 Mpro and the 20S proteasome. Our results are found to be reliable in predicting ABFEs for cases where the warheads are significantly different. This computational protocol might be a powerful tool for designing effective covalent inhibitors especially for SARS-CoV-2 Mpro and for targeted protein degradation.

Rational design, optimization, and biological evaluation of novel α-Phosphonopropionic acids as covalent inhibitors of Rab geranylgeranyl transferase

Rab geranylgeranyltransferase (GGTase-II, RGGT) catalyses the post-translational modification of eukaryotic Rab GTPases, proteins implicated in several pathologies, including cancer, diabetes, neurodegenerative, and infectious diseases. Thus, RGGT inhibitors are believed to be a potential platform for the development of drugs and tools for studying processes related to the abnormal activity of Rab GTPases. Here, a series of new α-phosphonocarboxylates have been prepared in the first attempt of rational design of covalent inhibitors of RGGT derived from non-covalent inhibitors. These compounds were equipped with electrophilic groups capable of binding cysteines, which are present in the catalytic cavity of RGGT. A few of these analogues have shown micromolar activity against RGGT, which correlated with their ability to inhibit the proliferation of the HeLa cancer cell line. The proposed mechanism of this inhibitory activity was rationalised by molecular docking and mass spectrometric measurements, supported by stability and reactivity studies.

Consensus Prediction of Human Gut Microbiota-Mediated Metabolism Susceptibility for Small Molecules by Machine Learning, Structural Alerts, and Dietary Compounds-Based Average Similarity Methods.

The human gut microbiota (HGM) colonizing human gastrointestinal tract (HGT) confers a repertoire of dynamic and unique metabolic capacities that are not possessed by the host and therefore is tentatively perceived as an alternative metabolic ″organ″ besides the liver in the host. Nevertheless, the significant contribution of HGM to the overall human metabolism is often overlooked in the modern drug discovery pipeline. Hence, a systematic evaluation of HGM-mediated drug metabolism is gradually important, and its computational prediction becomes increasingly necessary. In this work, a new data set containing both the HGM-mediated metabolism susceptible (HGMMS) and insusceptible (HGMMI) compounds (329 vs 320) was manually curated. Based on this data set, the first machine learning (ML) model, a new structural alerts (SA) model, and the K-nearest neighboring dietary compounds-based average similarity (AS) model were proposed to directly predict the HGM-mediated metabolism susceptibility for small molecules, and exhibit promising performance on three independent test sets. Finally, consensus prediction (ML/SA/AS) for DrugBank molecules revealed an intriguing phenomenon that a typical Michael acceptor ″α,β-unsaturated carbonyl group″ is a very common warhead for the design of covalent inhibitors and inclined to be metabolized by HGM in anaerobic HGT to generate the reduced metabolite without the reactive warhead, which could be a new concern to medicinal chemists. To the best of our knowledge, we gleaned the first HGMMS/HGMMI data set, developed the first HGMMS/HGMMI classification model, implemented a relatively comprehensive program based on ML/SA/AS approaches, and found a new phenomenon on the HGM-mediated deactivation of an extensively used warhead for covalent inhibitors.