Development of Machine Learning Models and the Discovery of a New Antiviral Compound against Yellow Fever Virus

Yellow fever (YF) is an acute viral hemorrhagic disease transmitted by infected mosquitoes. Large epidemics of YF occur when the virus is introduced into heavily populated areas with high mosquito density and low vaccination coverage. The lack of a specific small molecule drug treatment against YF as well as for homologous infections, such as zika and dengue, highlights the importance of these flaviviruses as a public health concern. With the advancement in computer hardware and bioactivity data availability, new tools based on machine learning methods have been introduced into drug discovery, as a means to utilize the growing high throughput screening (HTS) data generated to reduce costs and increase the speed of drug development. The use of predictive machine learning models using previously published data from HTS campaigns or data available in public databases, can enable the selection of compounds with desirable bioactivity and absorption, distribution, metabolism, and excretion profiles. In this study, we have collated cell-based assay data for yellow fever virus from the literature and public databases. The data were used to build predictive models with several machine learning methods that could prioritize compounds for in vitro testing. Five molecules were prioritized and tested in vitro from which we have identified a new pyrazolesulfonamide derivative with EC₅₀ 3.2 μM and CC₅₀ 24 μM, which represents a new scaffold suitable for hit-to-lead optimization that can expand the available drug discovery candidates for YF.

Quantum Machine Learning Algorithms for Drug Discovery Applications

The growing quantity of public and private data sets focused on small molecules screened against biological targets or whole organisms provides a wealth of drug discovery relevant data. This is matched by the availability of machine learning algorithms such as Support Vector Machines (SVM) and Deep Neural Networks (DNN) that are computationally expensive to perform on very large data sets with thousands of molecular descriptors. Quantum computer (QC) algorithms have been proposed to offer an approach to accelerate quantum machine learning over classical computer (CC) algorithms, however with significant limitations. In the case of cheminformatics, which is widely used in drug discovery, one of the challenges to overcome is the need for compression of large numbers of molecular descriptors for use on a QC. Here, we show how to achieve compression with data sets using hundreds of molecules (SARS-CoV-2) to hundreds of thousands of molecules (whole cell screening data sets for plague and M. tuberculosis) with SVM and the data reuploading classifier (a DNN equivalent algorithm) on a QC benchmarked against CC and hybrid approaches. This study illustrates the steps needed in order to be "quantum computer ready" in order to apply quantum computing to drug discovery and to provide the foundation on which to build this field.

Discovery of 5-Nitro-6-thiocyanatopyrimidines as Inhibitors of Cryptococcus neoformans and Cryptococcus gattii

Opportunistic infections from pathogenic fungi present a major challenge to healthcare because of a very limited arsenal of antifungal drugs, an increasing population of immunosuppressed patients, and increased prevalence of resistant clinical strains due to overuse of the few available antifungals. Cryptococcal meningitis is a life-threatening opportunistic fungal infection caused by one of two species in the Cryptococcus genus, Cryptococcus neoformans and Cryptococcus gattii. Eighty percent of cryptococcosis diseases are caused by C. neoformans that is endemic in the environment. The standard of care is limited to old antifungals, and under a high standard of care, mortality remains between 10 and 30%. We have identified a series of 5-nitro-6-thiocyanatopyrimidine antifungal drug candidates using in vitro and computational machine learning approaches. These compounds can inhibit C. neoformans growth at submicromolar levels, are effective against fluconazole-resistant C. neoformans and a clinical strain of C. gattii, and are not antagonistic with currently approved antifungals.

Multiple Computational Approaches for Predicting Drug Interactions with Human Equilibrative Nucleoside Transporter 1

Equilibrativenucleoside transporters (ENTs) participate in the pharmacokinetics and disposition of nucleoside analog drugs. Understanding drug interactions with the ENTs may inform and facilitate the development of new drugs, including chemotherapeutics and antivirals that require access to sanctuary sites such as the male genital tract. This study created three-dimensional pharmacophores for ENT1 and ENT2 substrates and inhibitors using K_t and IC₅₀ data curated from the literature. Substrate pharmacophores for ENT1 and ENT2 are distinct, with partial overlap of hydrogen bond donors, whereas the inhibitor pharmacophores predominantly feature hydrogen bond acceptors. Mizoribine and ribavirin mapped to the ENT1 substrate pharmacophore and proved to be substrates of the ENTs. The presence of the ENT-specific inhibitor 6-S-[(4-nitrophenyl)methyl]-6-thioinosine (NBMPR) decreased mizoribine accumulation in ENT1 and ENT2 cells (ENT1, ∼70% decrease, P = 0.0046; ENT2, ∼50% decrease, P = 0.0012). NBMPR also decreased ribavirin accumulation in ENT1 and ENT2 cells (ENT1: ∼50% decrease, P = 0.0498; ENT2: ∼30% decrease, P = 0.0125). Darunavir mapped to the ENT1 inhibitor pharmacophore and NBMPR did not significantly influence darunavir accumulation in either ENT1 or ENT2 cells (ENT1: P = 0.28; ENT2: P = 0.53), indicating that darunavir's interaction with the ENTs is limited to inhibition. These computational and in vitro models can inform compound selection in the drug discovery and development process, thereby reducing time and expense of identification and optimization of ENT-interacting compounds. SIGNIFICANCE STATEMENT: This study developed computational models of human equilibrative nucleoside transporters (ENTs) to predict drug interactions and validated these models with two compounds in vitro. Identification and prediction of ENT1 and ENT2 substrates allows for the determination of drugs that can penetrate tissues expressing these transporters.

Repurposing the Ebola and Marburg Virus Inhibitors Tilorone, Quinacrine, and Pyronaridine: In Vitro Activity against SARS-CoV-2 and Potential Mechanisms

Severe acute respiratory coronavirus 2 (SARS-CoV-2) is a newly identified virus that has resulted in over 2.5 million deaths globally and over 116 million cases globally in March, 2021. Small-molecule inhibitors that reverse disease severity have proven difficult to discover. One of the key approaches that has been widely applied in an effort to speed up the translation of drugs is drug repurposing. A few drugs have shown in vitro activity against Ebola viruses and demonstrated activity against SARS-CoV-2 in vivo. Most notably, the RNA polymerase targeting remdesivir demonstrated activity in vitro and efficacy in the early stage of the disease in humans. Testing other small-molecule drugs that are active against Ebola viruses (EBOVs) would appear a reasonable strategy to evaluate their potential for SARS-CoV-2. We have previously repurposed pyronaridine, tilorone, and quinacrine (from malaria, influenza, and antiprotozoal uses, respectively) as inhibitors of Ebola and Marburg viruses in vitro in HeLa cells and mouse-adapted EBOV in mice in vivo. We have now tested these three drugs in various cell lines (VeroE6, Vero76, Caco-2, Calu-3, A549-ACE2, HUH-7, and monocytes) infected with SARS-CoV-2 as well as other viruses (including MHV and HCoV 229E). The compilation of these results indicated considerable variability in antiviral activity observed across cell lines. We found that tilorone and pyronaridine inhibited the virus replication in A549-ACE2 cells with IC50 values of 180 nM and IC50 198 nM, respectively. We used microscale thermophoresis to test the binding of these molecules to the spike protein, and tilorone and pyronaridine bind to the spike receptor binding domain protein with K d values of 339 and 647 nM, respectively. Human Cmax for pyronaridine and quinacrine is greater than the IC50 observed in A549-ACE2 cells. We also provide novel insights into the mechanism of these compounds which is likely lysosomotropic.

A Machine Learning Strategy for Drug Discovery Identifies Anti-Schistosomal Small Molecules

Schistosomiasis is a chronic and painful disease of poverty caused by the flatworm parasite Schistosoma. Drug discovery for antischistosomal compounds predominantly employs in vitro whole organism (phenotypic) screens against two developmental stages of Schistosoma mansoni, post-infective larvae (somules) and adults. We generated two rule books and associated scoring systems to normalize 3898 phenotypic data points to enable machine learning. The data were used to generate eight Bayesian machine learning models with the Assay Central software according to parasite’s developmental stage and experimental time point (≤24, 48, 72, and >72 h). The models helped predict 56 active and nonactive compounds from commercial compound libraries for testing. When these were screened against S. mansoni in vitro, the prediction accuracy for active and inactives was 61% and 56% for somules and adults, respectively; also, hit rates were 48% and 34%, respectively, far exceeding the typical 1–2% hit rate for traditional high throughput screens.

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.

Bioactivity Comparison across Multiple Machine Learning Algorithms Using over 5000 Datasets for Drug Discovery

Machine learning methods are attracting considerable attention from the pharmaceutical industry for use in drug discovery and applications beyond. In recent studies, we and others have applied multiple machine learning algorithms and modeling metrics and, in some cases, compared molecular descriptors to build models for individual targets or properties on a relatively small scale. Several research groups have used large numbers of datasets from public databases such as ChEMBL in order to evaluate machine learning methods of interest to them. The largest of these types of studies used on the order of 1400 datasets. We have now extracted well over 5000 datasets from CHEMBL for use with the ECFP6 fingerprint and in comparison of our proprietary software Assay Central with random forest, k-nearest neighbors, support vector classification, naïve Bayesian, AdaBoosted decision trees, and deep neural networks (three layers). Model performance was assessed using an array of fivefold cross-validation metrics including area-under-the-curve, F1 score, Cohen's kappa, and Matthews correlation coefficient. Based on ranked normalized scores for the metrics or datasets, all methods appeared comparable, while the distance from the top indicated that Assay Central and support vector classification were comparable. Unlike prior studies which have placed considerable emphasis on deep neural networks (deep learning), no advantage was seen in this case. If anything, Assay Central may have been at a slight advantage as the activity cutoff for each of the over 5000 datasets representing over 570,000 unique compounds was based on Assay Central performance, although support vector classification seems to be a strong competitor. We also applied Assay Central to perform prospective predictions for the toxicity targets PXR and hERG to further validate these models. This work appears to be the largest scale comparison of these machine learning algorithms to date. Future studies will likely evaluate additional databases, descriptors, and machine learning algorithms and further refine the methods for evaluating and comparing such models.

Repurposing the Ebola and Marburg Virus Inhibitors Tilorone, Quinacrine and Pyronaridine: In vitro Activity Against SARS-CoV-2 and Potential Mechanisms

SARS-CoV-2 is a newly identified virus that has resulted in over 1.3 M deaths globally and over 59 M cases globally to date. Small molecule inhibitors that reverse disease severity have proven difficult to discover. One of the key approaches that has been widely applied in an effort to speed up the translation of drugs is drug repurposing. A few drugs have shown in vitro activity against Ebola virus and demonstrated activity against SARS-CoV-2 in vivo . Most notably the RNA polymerase targeting remdesivir demonstrated activity in vitro and efficacy in the early stage of the disease in humans. Testing other small molecule drugs that are active against Ebola virus would seem a reasonable strategy to evaluate their potential for SARS-CoV-2. We have previously repurposed pyronaridine, tilorone and quinacrine (from malaria, influenza, and antiprotozoal uses, respectively) as inhibitors of Ebola and Marburg virus in vitro in HeLa cells and of mouse adapted Ebola virus in mouse in vivo . We have now tested these three drugs in various cell lines (VeroE6, Vero76, Caco-2, Calu-3, A549-ACE2, HUH-7 and monocytes) infected with SARS-CoV-2 as well as other viruses (including MHV and HCoV 229E). The compilation of these results indicated considerable variability in antiviral activity observed across cell lines. We found that tilorone and pyronaridine inhibited the virus replication in A549-ACE2 cells with IC 50 values of 180 nM and IC 50 198 nM, respectively. We have also tested them in a pseudovirus assay and used microscale thermophoresis to test the binding of these molecules to the spike protein. They bind to spike RBD protein with K d values of 339 nM and 647 nM, respectively. Human C max for pyronaridine and quinacrine is greater than the IC 50 hence justifying in vivo evaluation. We also provide novel insights into their mechanism which is likely lysosomotropic.

Comparing Machine Learning Models for Aromatase (P450 19A1)

Aromatase, or cytochrome P450 19A1, catalyzes the aromatization of androgens to estrogens within the body. Changes in the activity of this enzyme can produce hormonal imbalances that can be detrimental to sexual and skeletal development. Inhibition of this enzyme can occur with drugs and natural products as well as environmental chemicals. Therefore, predicting potential endocrine disruption via exogenous chemicals requires that aromatase inhibition be considered in addition to androgen and estrogen pathway interference. Bayesian machine learning methods can be used for prospective prediction from the molecular structure without the need for experimental data. Herein, the generation and evaluation of multiple machine learning models utilizing different sources of aromatase inhibition data are described. These models are applied to two test sets for external validation with molecules relevant to drug discovery from the public domain. In addition, the performance of multiple machine learning algorithms was evaluated by comparing internal five-fold cross-validation statistics of the training data. These methods to predict aromatase inhibition from molecular structure, when used in concert with estrogen and androgen machine learning models, allow for a more holistic assessment of endocrine-disrupting potential of chemicals with limited empirical data and enable the reduction of the use of hazardous substances.

Computational Approaches to Identify Molecules Binding to Mycobacterium tuberculosis KasA

Tuberculosis is caused by Mycobacterium tuberculosis (Mtb) and is a deadly disease resulting in the deaths of approximately 1.5 million people with 10 million infections reported in 2018. Recently, a key condensation step in the synthesis of mycolic acids was shown to require β-ketoacyl-ACP synthase (KasA). A crystal structure of KasA with the small molecule DG167 was recently described, which provided a starting point for using computational structure-based approaches to identify additional molecules binding to this protein. We now describe structure-based pharmacophores, docking and machine learning studies with Assay Central as a computational tool for the identification of small molecules targeting KasA. We then tested these compounds using nanoscale differential scanning fluorimetry and microscale thermophoresis. Of note, we identified several molecules including the Food and Drug Administration (FDA)-approved drugs sildenafil and flubendazole with K _d values between 30-40 μM. This may provide additional starting points for further optimization.

Comparison of Machine Learning Models for the Androgen Receptor

The androgen receptor (AR) is a target of interest for endocrine disruption research, as altered signaling can affect normal reproductive and neurological development for generations. In an effort to prioritize compounds with alternative methodologies, the U.S. Environmental Protection Agency (EPA) used in vitro data from 11 assays to construct models of AR agonist and antagonist signaling pathways. While these EPA ToxCast AR models require in vitro data to assign a bioactivity score, Bayesian machine learning methods can be used for prospective prediction from molecule structure alone. This approach was applied to multiple types of data corresponding to the EPA's AR signaling pathway with proprietary software, Assay Central. The training performance of all machine learning models, including six other algorithms, was evaluated by internal 5-fold cross-validation statistics. Bayesian machine learning models were also evaluated with external predictions of reference chemicals to compare prediction accuracies to published results from the EPA. The machine learning model group selected for further studies of endocrine disruption consisted of continuous AC50 data from the February 2019 release of ToxCast/Tox21. These efforts demonstrate how machine learning can be used to predict AR-mediated bioactivity and can also be applied to other targets of endocrine disruption.

Machine Learning Models for Estrogen Receptor Bioactivity and Endocrine Disruption Prediction

The U.S. Environmental Protection Agency (EPA) periodically releases in vitro data across a variety of targets, including the estrogen receptor (ER). In 2015, the EPA used these data to construct mathematical models of ER agonist and antagonist pathways to prioritize chemicals for endocrine disruption testing. However, mathematical models require in vitro data prior to predicting estrogenic activity, but machine learning methods are capable of prospective prediction from the molecular structure alone. The current study describes the generation and evaluation of Bayesian machine learning models grouped by the EPA's ER agonist pathway model using multiple data types with proprietary software, Assay Central. External predictions with three test sets of in vitro and in vivo reference chemicals with agonist activity classifications were compared to previous mathematical model publications. Training data sets were subjected to additional machine learning algorithms and compared with rank normalized scores of internal five-fold cross-validation statistics. External predictions were found to be comparable or superior to previous studies published by the EPA. When assessing six additional algorithms for the training data sets, Assay Central performed similarly at a reduced computational cost. This study demonstrates that machine learning can prioritize chemicals for future in vitro and in vivo testing of ER agonism.

Repurposing Pyramax®, quinacrine and tilorone as treatments for Ebola virus disease

We have recently identified three molecules (tilorone, quinacrine and pyronaridine tetraphosphate) which all demonstrated efficacy in the mouse model of infection with mouse-adapted Ebola virus (EBOV) model of disease and had similar in vitro inhibition of an Ebola pseudovirus (VSV-EBOV-GP), suggesting they interfere with viral entry. Using a machine learning model to predict lysosomotropism these compounds were evaluated for their ability to possess a lysosomotropic mechanism in vitro. We now demonstrate in vitro that pyronaridine tetraphosphate is an inhibitor of Lysotracker accumulation in lysosomes (IC50 = 0.56 μM). Further, we evaluated antiviral synergy between pyronaridine and artesunate (Pyramax®), which are used in combination to treat malaria. Artesunate was not found to have lysosomotropic activity in vitro and the combination effect on EBOV inhibition was shown to be additive. Pyramax® may represent a unique example of the repurposing of a combination product for another disease.

Pyronaridine tetraphosphate efficacy against Ebola virus infection in guinea pig

The recent outbreaks of the Ebola virus (EBOV) in Africa have brought global visibility to the shortage of available therapeutic options to treat patients infected with this or closely related viruses. We have recently computationally identified three molecules which have all demonstrated statistically significant efficacy in the mouse model of infection with mouse adapted Ebola virus (ma-EBOV). One of these molecules is the antimalarial pyronaridine tetraphosphate (IC50 range of 0.82-1.30 μM against three strains of EBOV and IC50 range of 1.01-2.72 μM against two strains of Marburg virus (MARV)) which is an approved drug in the European Union and used in combination with artesunate. To date, no small molecule drugs have shown statistically significant efficacy in the guinea pig model of EBOV infection. Pharmacokinetics and range-finding studies in guinea pigs directed us to a single 300 mg/kg or 600 mg/kg oral dose of pyronaridine 1hr after infection. Pyronaridine resulted in statistically significant survival of 40% at 300 mg/kg and protected from a lethal challenge with EBOV. In comparison, oral favipiravir (300 mg/kg dosed once a day) had 43.5% survival. All animals in the vehicle treatment group succumbed to disease by study day 12 (100% mortality). The in vitro metabolism and metabolite identification of pyronaridine and another of our EBOV active molecules, tilorone, suggested significant species differences which may account for the efficacy or lack thereof, respectively in guinea pig. In summary, our studies with pyronaridine demonstrates its utility for repurposing as an antiviral against EBOV and MARV.

Toward the Target: Tilorone, Quinacrine, and Pyronaridine Bind to Ebola Virus Glycoprotein

Pyronaridine, tilorone, and quinacrine were recently identified by a machine learning model and demonstrated in vitro and in vivo activity against Ebola virus (EBOV) and represent viable candidates for drug repurposing. The target for these molecules was previously unknown. These drugs have now been docked into the crystal structure of the ebola glycoprotein and then experimentally validated in vitro using microscale thermophoresis to generate K _d values for tilorone (0.73 μM), pyronaridine (7.34 μM), and quinacrine (7.55 μM). These molecules were shown to bind with a higher affinity than the previously reported toremifene (16 μM). These three structures provide more insight into the structural diversity of ebola glycoprotein inhibitors which can be utilized in the discovery and design of additional inhibitors.

Synergistic drug combinations and machine learning for drug repurposing in chordoma

Chordoma is a devastating rare cancer that affects one in a million people. With a mean-survival of just 6 years and no approved medicines, the primary treatments are surgery and radiation. In order to speed new medicines to chordoma patients, a drug repurposing strategy represents an attractive approach. Drugs that have already advanced through human clinical safety trials have the potential to be approved more quickly than de novo discovered medicines on new targets. We have taken two strategies to enable this: (1) generated and validated machine learning models of chordoma inhibition and screened compounds of interest in vitro. (2) Tested combinations of approved kinase inhibitors already being individually evaluated for chordoma. Several published studies of compounds screened against chordoma cell lines were used to generate Bayesian Machine learning models which were then used to score compounds selected from the NIH NCATS industry-provided assets. Out of these compounds, the mTOR inhibitor AZD2014, was the most potent against chordoma cell lines (IC50 0.35 µM U-CH1 and 0.61 µM U-CH2). Several studies have shown the importance of the mTOR signaling pathway in chordoma and suggest it as a promising avenue for targeted therapy. Additionally, two currently FDA approved drugs, afatinib and palbociclib (EGFR and CDK4/6 inhibitors, respectively) demonstrated synergy in vitro (CI50 = 0.43) while AZD2014 and afatanib also showed synergy (CI50 = 0.41) against a chordoma cell in vitro. These findings may be of interest clinically, and this in vitro- and in silico approach could also be applied to other rare cancers.

Comparing Machine Learning Algorithms for Predicting Drug-Induced Liver Injury (DILI)

Drug-induced liver injury (DILI) is one the most unpredictable adverse reactions to xenobiotics in humans and the leading cause of postmarketing withdrawals of approved drugs. To date, these drugs have been collated by the FDA to form the DILIRank database, which classifies DILI severity and potential. These classifications have been used by various research groups in generating computational predictions for this type of liver injury. Recently, groups from Pfizer and AstraZeneca have collated DILI in vitro data and physicochemical properties for compounds that can be used along with data from the FDA to build machine learning models for DILI. In this study, we have used these data sets, as well as the Biopharmaceutics Drug Disposition Classification System data set, to generate Bayesian machine learning models with our in-house software, Assay Central. The performance of all machine learning models was assessed through both the internal 5-fold cross-validation metrics and prediction accuracy of an external test set of compounds with known hepatotoxicity. The best-performing Bayesian model was based on the DILI-concern category from the DILIRank database with an ROC of 0.814, a sensitivity of 0.741, a specificity of 0.755, and an accuracy of 0.746. A comparison of alternative machine learning algorithms, such as k-nearest neighbors, support vector classification, AdaBoosted decision trees, and deep learning methods, produced similar statistics to those generated with the Bayesian algorithm in Assay Central. This study demonstrates machine learning models grouped in a tool called MegaTox that can be used to predict early-stage clinical compounds, as well as recent FDA-approved drugs, to identify potential DILI.

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 the antimalarial pyronaridine tetraphosphate to protect against Ebola virus infection

Recent outbreaks of the Ebola virus (EBOV) have focused attention on the dire need for antivirals to treat these patients. We identified pyronaridine tetraphosphate as a potential candidate as it is an approved drug in the European Union which is currently used in combination with artesunate as a treatment for malaria (EC₅₀ between 420 nM—1.14 μM against EBOV in HeLa cells). Range-finding studies in mice directed us to a single 75 mg/kg i.p. dose 1 hr after infection which resulted in 100% survival and statistically significantly reduced viremia at study day 3 from a lethal challenge with mouse-adapted EBOV (maEBOV). Further, an EBOV window study suggested we could dose pyronaridine 2 or 24 hrs post-exposure to result in similar efficacy. Analysis of cytokine and chemokine panels suggests that pyronaridine may act as an immunomodulator during an EBOV infection. Our studies with pyronaridine clearly demonstrate potential utility for its repurposing as an antiviral against EBOV and merits further study in larger animal models with the added benefit of already being used as a treatment against malaria.

To date there is no approved drug for Ebola Virus infection. Our approach has been to assess drugs that are already approved for other uses in various countries. Using computational models, we have previously identified three such drugs and demonstrated their activity against the Ebola virus in vitro. We now report on the in vitro absorption, metabolism, distribution, excretion and pharmacokinetic properties of one of these molecules, namely the antimalarial pyronaridine. We justify efficacy testing in the mouse model of ebola infection. We also demonstrate that a single dose of this drug is 100% effective against the virus. This study provides important preclinical evaluation of this already approved drug and justifies its selection for larger animal efficacy studies.

Exploiting machine learning for end-to-end drug discovery and development

A variety of machine learning methods such as naive Bayesian, support vector machines and more recently deep neural networks are demonstrating their utility for drug discovery and development. These leverage the generally bigger datasets created from high-throughput screening data and allow prediction of bioactivities for targets and molecular properties with increased levels of accuracy. We have only just begun to exploit the potential of these techniques but they may already be fundamentally changing the research process for identifying new molecules and/or repurposing old drugs. The integrated application of such machine learning models for end-to-end (E2E) application is broadly relevant and has considerable implications for developing future therapies and their targeting.

Multiple Machine Learning Comparisons of HIV Cell-based and Reverse Transcriptase Data Sets

The human immunodeficiency virus (HIV) causes over a million deaths every year and has a huge economic impact in many countries. The first class of drugs approved were nucleoside reverse transcriptase inhibitors. A newer generation of reverse transcriptase inhibitors have become susceptible to drug resistant strains of HIV, and hence, alternatives are urgently needed. We have recently pioneered the use of Bayesian machine learning to generate models with public data to identify new compounds for testing against different disease targets. The current study has used the NIAID ChemDB HIV, Opportunistic Infection and Tuberculosis Therapeutics Database for machine learning studies. We curated and cleaned data from HIV-1 wild-type cell-based and reverse transcriptase (RT) DNA polymerase inhibition assays. Compounds from this database with ≤1 μM HIV-1 RT DNA polymerase activity inhibition and cell-based HIV-1 inhibition are correlated (Pearson r = 0.44, n = 1137, p < 0.0001). Models were trained using multiple machine learning approaches (Bernoulli Naive Bayes, AdaBoost Decision Tree, Random Forest, support vector classification, k-Nearest Neighbors, and deep neural networks as well as consensus approaches) and then their predictive abilities were compared. Our comparison of different machine learning methods demonstrated that support vector classification, deep learning, and a consensus were generally comparable and not significantly different from each other using 5-fold cross validation and using 24 training and test set combinations. This study demonstrates findings in line with our previous studies for various targets that training and testing with multiple data sets does not demonstrate a significant difference between support vector machine and deep neural networks.

Halogen Substitution Influences Ketamine Metabolism by Cytochrome P450 2B6: In Vitro and Computational Approaches

Ketamine is analgesic at anesthetic and subanesthetic doses, and it has been used recently to treat depression. Biotransformation mediates ketamine effects, influencing both systemic elimination and bioactivation. CYP2B6 is the major catalyst of hepatic ketamine N-demethylation and metabolism at clinically relevant concentrations. Numerous CYP2B6 substrates contain halogens. CYP2B6 readily forms halogen-protein (particularly Cl-π) bonds, which influence substrate selectivity and active site orientation. Ketamine is chlorinated, but little is known about the metabolism of halogenated analogs. This investigation evaluated halogen substitution effects on CYP2B6-catalyzed ketamine analogs N-demethylation in vitro and modeled interactions with CYP2B6 using various computational approaches. Ortho phenyl ring halogen substituent changes caused substantial (18-fold) differences in Km, on the order of Br (bromoketamine, 10 μM) < Cl < F < H (deschloroketamine, 184 μM). In contrast, Vmax varied minimally (83-103 pmol/min/pmol CYP). Thus, apparent substrate binding affinity was the major consequence of halogen substitution and the major determinant of N-demethylation. Docking poses of ketamine and analogs were similar, sharing a π-stack with F297. Libdock scores were deschloroketamine < bromoketamine < ketamine < fluoroketamine. A Bayesian log Km model generated with Assay Central had a ROC of 0.86. The probability of activity at 15 μM for ketamine and analogs was predicted with this model. Deschloroketamine scores corresponded to the experimental Km, but the model was unable to predict activity with fluoroketamine. The binding pocket of CYP2B6 also suggested a hydrophobic component to substrate docking, on the basis of a strong linear correlation ( R2 = 0.92) between lipophilicity ( Alog P) and metabolism (log Km) of ketamine and analogs. This property may be the simplest design criteria to use when considering similar compounds and CYP2B6 affinity.

The EU approved antimalarial pyronaridine shows antitubercular activity and synergy with rifampicin, targeting RNA polymerase

The search for compounds with biological activity for many diseases is turning increasingly to drug repurposing. In this study, we have focused on the European Union-approved antimalarial pyronaridine which was found to have in vitro activity against Mycobacterium tuberculosis (MIC 5 μg/mL). In macromolecular synthesis assays, pyronaridine resulted in a severe decrease in incorporation of ¹⁴C-uracil and ¹⁴C-leucine similar to the effect of rifampicin, a known inhibitor of M. tuberculosis RNA polymerase. Surprisingly, the co-administration of pyronaridine (2.5 μg/ml) and rifampicin resulted in in vitro synergy with an MIC 0.0019-0.0009 μg/mL. This was mirrored in a THP-1 macrophage infection model, with a 16-fold MIC reduction for rifampicin when the two compounds were co-administered versus rifampicin alone. Docking pyronaridine in M. tuberculosis RNA polymerase suggested the potential for it to bind outside of the RNA polymerase rifampicin binding pocket. Pyronaridine was also found to have activity against a M. tuberculosis clinical isolate resistant to rifampicin, and when combined with rifampicin (10% MIC) was able to inhibit M. tuberculosis RNA polymerase in vitro. All these findings, and in particular the synergistic behavior with the antitubercular rifampicin, inhibition of RNA polymerase in combination in vitro and its current use as a treatment for malaria, may suggest that pyronaridine could also be used as an adjunct for treatment against M. tuberculosis infection. Future studies will test potential for in vivo synergy, clinical utility and attempt to develop pyronaridine analogs with improved potency against M. tuberculosis RNA polymerase when combined with rifampicin.

Structure of the ACF7 EF-Hand-GAR Module and Delineation of Microtubule Binding Determinants

Spectraplakins are large molecules that cross-link F-actin and microtubules (MTs). Mutations in spectraplakins yield defective cell polarization, aberrant focal adhesion dynamics, and dystonia. We present the 2.8 Å crystal structure of the hACF7 EF1-EF2-GAR MT-binding module and delineate the GAR residues critical for MT binding. The EF1-EF2 and GAR domains are autonomous domains connected by a flexible linker. The EF1-EF2 domain is an EFβ-scaffold with two bound Ca²⁺ ions that straddle an N-terminal α helix. The GAR domain has a unique α/β sandwich fold that coordinates Zn²⁺. While the EF1-EF2 domain is not sufficient for MT binding, the GAR domain is and likely enhances EF1-EF2-MT engagement. Residues in a conserved basic patch, distal to the GAR domain's Zn²⁺-binding site, mediate MT binding.