Machine Learning Models for Predicting Molecular UV-Vis Spectra with Quantum Mechanical Properties.

Accurate understanding of ultraviolet-visible (UV-vis) spectra is critical for the high-throughput synthesis of compounds for drug discovery. Experimentally determining UV-vis spectra can become expensive when dealing with a large quantity of novel compounds. This provides us an opportunity to drive computational advances in molecular property predictions using quantum mechanics and machine learning methods. In this work, we use both quantum mechanically (QM) predicted and experimentally measured UV-vis spectra as input to devise four different machine learning architectures, UVvis-SchNet, UVvis-DTNN, UVvis-Transformer, and UVvis-MPNN, and assess the performance of each method. We find that the UVvis-MPNN model outperforms the other models when using optimized 3D coordinates and QM predicted spectra as input features. This model has the highest performance for predicting UV-vis spectra with a training RMSE of 0.06 and validation RMSE of 0.08. Most importantly, our model can be used for the challenging task of predicting differences in the UV-vis spectral signatures of regioisomers.

SynRoute: A Retrosynthetic Planning Software.

Computer-assisted synthetic planning has seen major advancements that stem from the availability of large reaction databases and artificial intelligence methodologies. SynRoute is a new retrosynthetic planning software tool that uses a relatively small number of general reaction templates, currently 263, along with a literature-based reaction database to find short, practical synthetic routes for target compounds. For each reaction template, a machine learning classifier is trained using data from the Pistachio reaction database to predict whether new computer-generated reactions based on the template are likely to work experimentally in the laboratory. This reaction generation methodology is used together with a vectorized Dijkstra-like search of top-scoring routes organized by synthetic strategies for easy browsing by a synthetic chemist. SynRoute was able to find routes for an average of 83% of compounds based on selection of random subsets of drug-like compounds from the ChEMBL database. Laboratory evaluation of 12 routes produced by SynRoute, to synthesize compounds not from the previous random subsets, demonstrated the ability to produce feasible overall synthetic strategies for all compounds evaluated.

UV-adVISor: Attention-Based Recurrent Neural Networks to Predict UV-Vis Spectra.

Ultraviolet-visible (UV-Vis) absorption spectra are routinely collected as part of high-performance liquid chromatography (HPLC) analysis systems and can be used to identify chemical reaction products by comparison to the reference spectra. Here, we present UV-adVISor as a new computational tool for predicting the UV-Vis spectra from a molecule's structure alone. UV-Vis prediction was approached as a sequence-to-sequence problem. We utilized Long-Short Term Memory and attention-based neural networks with Extended Connectivity Fingerprint Diameter 6 or molecule SMILES to generate predictive models for the UV spectra. We have produced two spectrum datasets (dataset I, N = 949, and dataset II, N = 2222) using different compound collections and spectrum acquisition methods to train, validate, and test our models. We evaluated the prediction accuracy of the complete spectra by the correspondence of wavelengths of absorbance maxima and with a series of statistical measures (the best test set median model parameters are in parentheses for model II), including RMSE (0.064), R2 (0.71), and dynamic time warping (DTW, 0.194) of the entire spectrum curve. Scrambling molecule structures with the experimental spectra during training resulted in a degraded R2, confirming the utility of the approaches for prediction. UV-adVISor is able to provide fast and accurate predictions for libraries of compounds.

Contraceptive and Infertility Target DataBase: a contraceptive drug development tool for targeting and analysis of human reproductive specific tissues†.

The long and challenging drug development process begins with discovery biology for the selection of an appropriate target for a specific indication. Target is a broad term that can be applied to a range of biological entities such as proteins, genes, and ribonucleic acids (RNAs). Although there are numerous databases available for mining biological entities, publicly available searchable, downloadable databases to aid in target selection for a specific disease or indication (e.g., developing contraceptives and infertility treatments) are limited. We report the development of the Contraceptive and Infertility Target DataBase (https://www.citdbase.org), which provides investigators an interface to mine existing transcriptomic and proteomic resources to identify high-quality contraceptive/infertility targets. The development of similar databases is applicable to the identification of targets for other diseases and conditions.

The Antiviral Drug Tilorone Is a Potent and Selective Inhibitor of Acetylcholinesterase.

Acetylcholinesterase (AChE) is an important drug target in neurological disorders like Alzheimer's disease, Lewy body dementia, and Parkinson's disease dementia as well as for other conditions like myasthenia gravis and anticholinergic poisoning. In this study, we have used a combination of high-throughput screening, machine learning, and docking to identify new inhibitors of this enzyme. Bayesian machine learning models were generated with literature data from ChEMBL for eel and human AChE inhibitors as well as butyrylcholinesterase inhibitors (BuChE) and compared with other machine learning methods. High-throughput screens for the eel AChE inhibitor model identified several molecules including tilorone, an antiviral drug that is well-established outside of the United States, as a newly identified nanomolar AChE inhibitor. We have described how tilorone inhibits both eel and human AChE with IC50's of 14.4 nM and 64.4 nM, respectively, but does not inhibit the closely related BuChE IC50 > 50 µM. We have docked tilorone into the human AChE crystal structure and shown that this selectivity is likely due to the reliance on a specific interaction with a hydrophobic residue in the peripheral anionic site of AChE that is absent in BuChE. We also conducted a pharmacological safety profile (SafetyScreen44) and kinase selectivity screen (SelectScreen) that showed tilorone (1 µM) only inhibited AChE out of 44 toxicology target proteins evaluated and did not appreciably inhibit any of the 485 kinases tested. This study suggests there may be a potential role for repurposing tilorone or its derivatives in conditions that benefit from AChE inhibition.

Machine Learning Models Identify Inhibitors of SARS-CoV-2.

With the rapidly evolving SARS-CoV-2 variants of concern, there is an urgent need for the discovery of further treatments for the coronavirus disease (COVID-19). Drug repurposing is one of the most rapid strategies for addressing this need, and numerous compounds have already been selected for in vitro testing by several groups. These have led to a growing database of molecules with in vitro activity against the virus. Machine learning models can assist drug discovery through prediction of the best compounds based on previously published data. Herein, we have implemented several machine learning methods to develop predictive models from recent SARS-CoV-2 in vitro inhibition data and used them to prioritize additional FDA-approved compounds for in vitro testing selected from our in-house compound library. From the compounds predicted with a Bayesian machine learning model, lumefantrine, an antimalarial was selected for testing and showed limited antiviral activity in cell-based assays while demonstrating binding (Kd 259 nM) to the spike protein using microscale thermophoresis. Several other compounds which we prioritized have since been tested by others and were also found to be active in vitro. This combined machine learning and in vitro testing approach can be expanded to virtually screen available molecules with predicted activity against SARS-CoV-2 reference WIV04 strain and circulating variants of concern. In the process of this work, we have created multiple iterations of machine learning models that can be used as a prioritization tool for SARS-CoV-2 antiviral drug discovery programs. The very latest model for SARS-CoV-2 with over 500 compounds is now freely available at www.assaycentral.org.

Tilorone, a Broad-Spectrum Antiviral for Emerging Viruses.

Tilorone is a 50-year-old synthetic small-molecule compound with antiviral activity that is proposed to induce interferon after oral administration. This drug is used as a broad-spectrum antiviral in several countries of the Russian Federation. We have recently described activity in vitro and in vivo against the Ebola virus. After a broad screening of additional viruses, we now describe in vitro activity against Chikungunya virus (CHIK) and Middle Eastern respiratory syndrome coronavirus (MERS-CoV).

Correction to: Tilorone: a Broad-Spectrum Antiviral Invented in the USA and Commercialized in Russia and beyond.

The mentioned Tables below had data in the bottom half (right hand side) of each table that was formatted incorrectly in the final proof version versus the submitted version.

Tilorone: a Broad-Spectrum Antiviral Invented in the USA and Commercialized in Russia and beyond.

For the last 50 years we have known of a broad-spectrum agent tilorone dihydrochloride (Tilorone). This is a small-molecule orally bioavailable drug that was originally discovered in the USA and is currently used clinically as an antiviral in Russia and the Ukraine. Over the years there have been numerous clinical and non-clinical reports of its broad spectrum of antiviral activity. More recently we have identified additional promising antiviral activities against Middle East Respiratory Syndrome, Chikungunya, Ebola and Marburg which highlights that this old drug may have other uses against new viruses. This may in turn inform the types of drugs that we need for virus outbreaks such as for the new coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Tilorone has been long neglected by the west in many respects but it deserves further reassessment in light of current and future needs for broad-spectrum antivirals.

Repurposing Quinacrine against Ebola Virus Infection In Vivo.

Quinacrine hydrochloride is a small-molecule, orally bioavailable drug that has been used clinically as an antimalarial and for many other applications. A machine learning model trained on Ebola virus (EBOV) screening data identified quinacrine as a potent (nanomolar) in vitro inhibitor. In the current study, quinacrine 25 mg/kg was shown to protect 70% of mice (statistically significant) from a lethal challenge with mouse-adapted EBOV with once-daily intraperitoneal dosing for 8 days.

The Natural Product Eugenol Is an Inhibitor of the Ebola Virus In Vitro.

PURPOSE: Since the 2014 Ebola virus (EBOV) outbreak in West Africa there has been considerable effort towards developing drugs to treat Ebola virus disease and yet to date there is no FDA approved treatment. This is important as at the time of writing this manuscript there is an ongoing outbreak in the Democratic Republic of the Congo which has killed over 1000. METHODS: We have evaluated a small number of natural products, some of which had shown antiviral activity against other pathogens. This is exemplified with eugenol, which is found in high concentrations in multiple essential oils, and has shown antiviral activity against feline calicivirus, tomato yellow leaf curl virus, Influenza A virus, Herpes Simplex virus type 1 and 2, and four airborne phages. RESULTS: Four compounds possessed EC50 values less than or equal to 11 µM. Of these, eugenol, had an EC50 of 1.3 µM against EBOV and is present in several plants including clove, cinnamon, basil and bay. Eugenol is much smaller and structurally unlike any compound that has been previously identified as an inhibitor of EBOV, therefore it may provide new mechanistic insights. CONCLUSION: This compound is readily accessible in bulk quantities, is inexpensive, and has a long history of human consumption, which endorses the idea for further assessment as an antiviral therapeutic. This work also suggests that a more exhaustive assessment of natural product libraries against EBOV and other viruses is warranted to improve our ability to identify compounds that are so distinct from FDA approved drugs.

Efficacy of Tilorone Dihydrochloride against Ebola Virus Infection.

Tilorone dihydrochloride (tilorone) is a small-molecule, orally bioavailable drug that is used clinically as an antiviral outside the United States. A machine-learning model trained on anti-Ebola virus (EBOV) screening data previously identified tilorone as a potent in vitro EBOV inhibitor, making it a candidate for the treatment of Ebola virus disease (EVD). In the present study, a series of in vitro ADMET (absorption, distribution, metabolism, excretion, toxicity) assays demonstrated the drug has excellent solubility, high Caco-2 permeability, was not a P-glycoprotein substrate, and had no inhibitory activity against five human CYP450 enzymes (3A4, 2D6, 2C19, 2C9, and 1A2). Tilorone was shown to have 52% human plasma protein binding with excellent plasma stability and a mouse liver microsome half-life of 48 min. Dose range-finding studies in mice demonstrated a maximum tolerated single dose of 100 mg/kg of body weight. A pharmacokinetics study in mice at 2- and 10-mg/kg dose levels showed that the drug is rapidly absorbed, has dose-dependent increases in maximum concentration of unbound drug in plasma and areas under the concentration-time curve, and has a half-life of approximately 18 h in both males and females, although the exposure was ∼2.5-fold higher in male mice. Tilorone doses of 25 and 50 mg/kg proved efficacious in protecting 90% of mice from a lethal challenge with mouse-adapted with once-daily intraperitoneal (i.p.) dosing for 8 days. A subsequent study showed that 30 mg/kg/day of tilorone given i.p. starting 2 or 24 h postchallenge and continuing through day 7 postinfection was fully protective, indicating promising activity for the treatment of EVD.

Molecular Imaging of Mesothelin-Expressing Ovarian Cancer with a Human and Mouse Cross-Reactive Nanobody.

Mesothelin is an epithelial marker highly expressed at the cell surface of cancer cells from diverse origins, including ovarian and pancreatic adenocarcinomas and mesotheliomas. Previously, we identified and characterized an antimesothelin nanobody (NbG3a) for in vitro diagnostic applications. The main goal of this research was to establish the potential of NbG3a as a molecular imaging agent. Site-specific biotinylated NbG3a (bNbG3a) was bound to streptavidin-conjugated reagents for in vitro and in vivo assays. Initially, we performed microscale thermophoresis to determine the binding affinity between bNbG3a and human ( Kd = 46 ± 8 nM) or mouse ( Kd = 4.8 ± 0.4 nM) mesothelin protein. The human and mouse cross-reactivity was confirmed by in vivo optical imaging using bNbG3a bound to fluorescent streptavidin. We also localized the binding site of nNbG3a on human mesothelin using overlapping peptide scan. NbG3a recognized an epitope within residues 21-65 of the mature membrane bound form of human mesothelin, which is part of the N-terminal region of mesothelin that is important for interactions between mesothelin on peritoneal cells and CA125 on tumor cells. Next, the bNbG3a in vivo half-life after intravenous injection in healthy mice was estimated by ELISA assay to be 5.3 ± 1.3 min. In tumor-bearing animals, fluorescent bNbG3a accumulated in a subcutaneous ovarian xenograft (A1847) and in two syngeneic, orthotopic ovarian tumors (intraovary and intraperitoneal ID8) within an hour of intravenous injection that peaked by 4 h and persisted up to 48 h. MRI analysis of bNbG3a-targeted streptavidin-labeled iron oxides showed that the MRI signal intensity decreased 1 h after injection for a subcutaneous xenograft model of ovarian cancer for bNbG3a-labeled iron oxides compared to unlabeled iron oxides. The signal intensity differences continued up to the final time point at 24 h post injection. Finally, in vivo immunofluorescence 24 or 48 h after bNbG3a intravenous injection showed bNbG3a diffuse distribution of both xenograft and syngeneic ovarian tumors, with local areas of high concentration throughout A1847 human tumor. The data support the use of NbG3a for continued preclinical development and translation to human applications for cancers that overexpress mesothelin.

Bulged DNA substrates for identifying poxvirus resolvase inhibitors.

Resolvase enzymes that cleave DNA four-way (Holliday) junctions are required for poxvirus replication, but clinically useful inhibitors have not been developed. Here, we report an assay for resolvase cleavage activity based on fluorescence polarization (FP) for high-throughput screening and mechanistic studies. Initial analysis showed that cleavage of a fluorescently labeled Holliday junction substrate did not yield an appreciable change in FP, probably because the cleavage product did not have sufficiently increased mobility to yield a strong FP signal. Iterative optimization yielded a substrate with an off-center DNA bulge, which after cleavage released a labeled short stand and yielded a greatly reduced FP signal. Using this assay, 133,000 compounds were screened, identifying 1-hydroxy-1,8-naphthyridin-2(1H)-one compounds as inhibitors. Structure-activity studies revealed functional parallels to Food and Drug Administration (FDA)-approved drugs targeting the related human immunodeficiency virus integrase enzyme. Some 1-hydroxy-1,8-naphthyridin-2(1H)-one compounds showed anti-poxvirus activity.

Evaluation of gyrase B as a drug target in Mycobacterium tuberculosis.

OBJECTIVES: New classes of drugs are needed to treat tuberculosis (TB) in order to combat the emergence of resistance to existing agents and shorten the duration of therapy. Targeting DNA gyrase is a clinically validated therapeutic approach using fluoroquinolone antibiotics to target the gyrase subunit A (GyrA) of the heterotetramer. Increasing resistance to fluoroquinolones has driven interest in targeting the gyrase subunit B (GyrB), which has not been targeted for TB. The biological activities of two potent small-molecule inhibitors of GyrB have been characterized to validate its targeting as a therapeutic strategy for treating TB. MATERIALS AND METHODS: Novobiocin and aminobenzimidazole 1 (AB-1) were tested for their activity against Mycobacterium tuberculosis (Mtb) H37Rv and other mycobacteria. AB-1 and novobiocin were also evaluated for their interaction with rifampicin and isoniazid as well as their potential for cytotoxicity. Finally, AB-1 was tested for in vivo efficacy in a murine model of TB. RESULTS: Novobiocin and AB-1 have both been shown to be active against Mtb with MIC values of 4 and 1 mg/L, respectively. Only AB-1 exhibited time-dependent bactericidal activity against drug-susceptible and drug-resistant mycobacteria, including a fluoroquinolone-resistant strain. AB-1 had potent activity in the low oxygen recovery assay model for non-replicating persistent Mtb. Additionally, AB-1 has no interaction with isoniazid and rifampicin, and has no cross-resistance with fluoroquinolones. In a murine model of TB, AB-1 significantly reduced lung cfu counts in a dose-dependent manner. CONCLUSIONS: Aminobenzimidazole inhibitors of GyrB exhibit many of the characteristics required for their consideration as a potential front-line antimycobacterial therapeutic.

Combining cheminformatics methods and pathway analysis to identify molecules with whole-cell activity against Mycobacterium tuberculosis.

PURPOSE: New strategies for developing inhibitors of Mycobacterium tuberculosis (Mtb) are required in order to identify the next generation of tuberculosis (TB) drugs. Our approach leverages the integration of intensive data mining and curation and computational approaches, including cheminformatics combined with bioinformatics, to suggest biological targets and their small molecule modulators. METHODS: We now describe an approach that uses the TBCyc pathway and genome database, the Collaborative Drug Discovery database of molecules with activity against Mtb and their associated targets, a 3D pharmacophore approach and Bayesian models of TB activity in order to select pathways and metabolites and ultimately prioritize molecules that may be acting as substrate mimics and exhibit activity against TB. RESULTS: In this study we combined the TB cheminformatics and pathways databases that enabled us to computationally search >80,000 vendor available molecules and ultimately test 23 compounds in vitro that resulted in two compounds (N-(2-furylmethyl)-N'-[(5-nitro-3-thienyl)carbonyl]thiourea and N-[(5-nitro-3-thienyl)carbonyl]-N'-(2-thienylmethyl)thiourea) proposed as mimics of D-fructose 1,6 bisphosphate, (MIC of 20 and 40 µg/ml, respectively). CONCLUSION: This is a simple yet novel approach that has the potential to identify inhibitors of bacterial growth as illustrated by compounds identified in this study that have activity against Mtb.

Mega-High-Throughput Screening Platform for the Discovery of Biologically Relevant Sequence-Defined Non-Natural Polymers.

Combinatorial methods enable the synthesis of chemical libraries on scales of millions to billions of compounds, but the ability to efficiently screen and sequence such large libraries has remained a major bottleneck for molecular discovery. We developed a novel technology for screening and sequencing libraries of synthetic molecules of up to a billion compounds in size. This platform utilizes the fiber-optic array scanning technology (FAST) to screen bead-based libraries of synthetic compounds at a rate of 5 million compounds per minute (∼83 000 Hz). This ultra-high-throughput screening platform has been used to screen libraries of synthetic "self-readable" non-natural polymers that can be sequenced at the femtomole scale by chemical fragmentation and high-resolution mass spectrometry. The versatility and throughput of the platform were demonstrated by screening two libraries of non-natural polyamide polymers with sizes of 1.77M and 1B compounds against the protein targets K-Ras, asialoglycoprotein receptor 1 (ASGPR), IL-6, IL-6 receptor (IL-6R), and TNFα. Hits with low nanomolar binding affinities were found against all targets, including competitive inhibitors of K-Ras binding to Raf and functionally active uptake ligands for ASGPR facilitating intracellular delivery of a nonglycan ligand.

Identification of antimicrobial activity among FDA-approved drugs for combating Mycobacterium abscessus and Mycobacterium chelonae.

OBJECTIVES: Rapidly growing mycobacteria have long been neglected in drug discovery efforts and this neglect is reflected in the paucity of therapeutic options available for diseases resulting from these infections. The purpose of this work is to identify new candidate drugs for treating non-tuberculous mycobacteria (NTM) by testing FDA-approved drugs for antimicrobial activity against Mycobacterium abscessus and Mycobacterium chelonae, two emerging NTM drug-resistant pathogens. METHODS: In this study, we screened 1040 FDA-approved drugs against M. abscessus and M. chelonae. RESULTS: Of the drugs screened, 32 compounds exhibited significant antimicrobial activity, with an MIC ≤ 8 mg/L, against M. chelonae, while only 7 compounds showed such activity against M. abscessus. Notably, neostigmine bromide and cinnarizine exhibited highly significant antimicrobial activity against M. chelonae, but had little potency against M. abscessus. Metronidazole and puromycin were the only drugs that acted equipotently against both strains, in decreasing order of effectiveness. CONCLUSIONS: The dearth of identified compounds active against M. abscessus exemplifies its ability to resist drugs as well as the resilience of rapidly growing NTM. Repurposing of approved drugs is a viable alternative to de novo drug discovery and development.

Mannich reaction derivatives of novobiocin with modulated physiochemical properties and their antibacterial activities.

Synthetic derivatives of the natural product antibiotic novobiocin were synthesized in order to improve their physiochemical properties. A Mannich reaction was used to introduce new side chains at a solvent-exposed position of the molecule, and a diverse panel of functional groups was evaluated at this position. Novobiocin and the new derivatives were tested for their binding to gyrase B and their antibacterial activities against Staphylococcus aureus, Mycobacterium tuberculosis, Francisella tularensis and Escherichia coli. While the new derivatives still bound the gyrase B protein potently (0.07-1.8 µM, IC(50)), they had significantly less antibacterial activity. Two compounds were identified with increased antibacterial activity against M. tuberculosis, with a minimum inhibitory concentration of 2.5 µg/ml.

Acylideneoxoindoles: a new class of reversible inhibitors of human transglutaminase 2.

Inhibitors of human transglutaminase 2 (TG2) are anticipated to be useful in the therapy of a variety of diseases including celiac sprue as well as certain CNS disorders and cancers. A class of 3-acylidene-2-oxoindoles was identified as potent reversible inhibitors of human TG2. Structure-activity relationship analysis of a lead compound led to the generation of several potent, competitive inhibitors. Analogs with significant non-competitive character were also identified, suggesting that the compounds bind at one or more allosteric regulatory sites on this multidomain enzyme. The most active compounds had K(i) values below 1.0 µM in two different kinetic assays for human TG2, and may therefore be suitable for investigations into the role of TG2 in physiology and disease in animals.