Computer-Assisted Discovery of Retinoid X Receptor Modulating Natural Products and Isofunctional Mimetics

Rational design and virtual screening identify mimetics of the RXR agonist valerenic acid

The ligand-sensing transcription factor retinoid X receptor (RXR) is the universal heterodimer partner of nuclear receptors and involved in multiple physiological processes. Its pharmacological modulation holds therapeutic potential in cancer and neurodegeneration but many available RXR ligands lack specificity. The sesquiterpenoid valerenic acid has been identified as RXR agonist with unprecedented subtype and homodimer preference. Here, we identified simplified mimetics of the complex natural product by rational design and virtual screening that exhibited similar activity profiles on RXR and informed about structural elements contributing to the favorable activity.

Structural Fusion of Natural and Synthetic Ligand Features Boosts RXR Agonist Potency

The retinoid X receptors (RXRs) are ligand-activated transcription factors involved in, for example, differentiation and apoptosis regulation. Currently used reference RXR agonists suffer from insufficient specificity and poor physicochemical properties, and improved tools are needed to capture the unexplored therapeutic potential of RXR. Endogenous vitamin A-derived RXR ligands and the natural product RXR agonist valerenic acid comprise acrylic acid residues with varying substitution patterns to engage the critical ionic contact with the binding site arginine. To mimic and exploit this natural ligand motif, we probed its structural fusion with synthetic RXR modulator scaffolds, which had profound effects on agonist activity and remarkably boosted potency of an oxaprozin-derived RXR agonist chemotype. Bioisosteric replacement of the acrylic acid to overcome its pan-assay interference compounds (PAINS) character enabled the development of a highly optimized RXR agonist chemical probe.

MARS Plus: An Improved Molecular Design Tool for Complex Compounds Involving Ionic, Stereo, and Cis-Trans Isomeric Structures

MARS (Molecular Assembling and Representation Suite) (Hsu et al. J. Chem. Inf. Model. 2019, 59, 3703-3713) is a toolbox for the molecular design of organic molecules. MARS uses integer arrays to represent the elements and connectivity between elements of a molecule. It provides a collection of operations to manipulate the elemental composition and connectivity of a molecule (or a pair of molecules), enabling the creation of novel chemical compounds. In this work, the original MARS is extended to handle complex molecular structures, including geometric (cis-trans) isomers, stereo isomers, cyclic compounds, and ionic species. The extended version of MARS, referred to as MARS+, has a more comprehensive coverage of the chemical space and therefore can explore molecules with a greater chemical and physical diversity. Compared to other molecular design tools, MARS+ is designed to perform all possible manipulations on a given molecule or a pair of molecules. Molecular structure manipulation can be conducted in either a controlled or a random fashion. Furthermore, every structure manipulation has a counterpart so that the operation can be reversed. Nearly any possible chemical structure can be generated with MARS+ via a combination of molecular operations. The capabilities of MARS+ are examined by the design of new ionic liquids (ILs). The results show that MARS+ is a useful tool for computer-aided molecular design (CAMD) and molecular structure enumeration.

Structural Modification of the Natural Product Valerenic Acid Tunes RXR Homodimer Agonism

Retinoid X receptors (RXR) are ligand-sensing transcription factors with a unique role in nuclear receptor signaling as universal heterodimer partners. RXR modulation holds potential in cancer, neurodegeneration and metabolic diseases but adverse effects of RXR activation and lack of selective modulators prevent further exploration as therapeutic target. The natural product valerenic acid has been discovered as RXR agonist with unprecedented preference for RXR subtype and homodimer activation. To capture structural determinants of this activity profile and identify potential for optimization, we have studied effects of structural modification of the natural product on RXR modulation and identified an analogue with enhanced RXR homodimer agonism.

De Novo Design of κ-Opioid Receptor Antagonists Using a Generative Deep-Learning Framework

Likely effective pharmacological interventions for the treatment of opioid addiction include attempts to attenuate brain reward deficits during periods of abstinence. Pharmacological blockade of the κ-opioid receptor (KOR) has been shown to abolish brain reward deficits in rodents during withdrawal, as well as to reduce the escalation of opioid use in rats with extended access to opioids. Although KOR antagonists represent promising candidates for the treatment of opioid addiction, very few potent selective KOR antagonists are known to date and most of them exhibit significant safety concerns. Here, we used a generative deep-learning framework for the de novo design of chemotypes with putative KOR antagonistic activity. Molecules generated by models trained with this framework were prioritized for chemical synthesis based on their predicted optimal interactions with the receptor. Our models and proposed training protocol were experimentally validated by binding and functional assays.

De Novo Design of κ-Opioid Receptor Antagonists Using a Generative Deep Learning Framework

Likely effective pharmacological interventions for the treatment of opioid addiction include attempts to attenuate brain reward deficits during periods of abstinence. Pharmacological blockade of the κ-opioid receptor (KOR) has been shown to abolish brain reward deficits in rodents during withdrawal, as well as to reduce the escalation of opioid use in rats with extended access to opioids. Although KOR antagonists represent promising candidates for the treatment of opioid addiction, very few potent selective KOR antagonists are known to date and most of them exhibit significant safety concerns. Here, we used a generative deep learning framework for the de novo design of chemotypes with putative KOR antagonistic activity. Molecules generated by models trained with this framework were prioritized for chemical synthesis based on their predicted optimal interactions with the receptor. Our models and proposed training protocol were experimentally validated by binding and functional assays.

Rational Design of a New RXR Agonist Scaffold Enabling Single-Subtype Preference for RXRα, RXRβ, and RXRγ

The three retinoid X receptor subtypes (RXRα, RXRβ, RXRγ) exhibit critical regulatory roles in cell proliferation and differentiation, metabolism, and inflammation. Due to their importance in nuclear receptor signaling, RXRs are widely distributed and pan-RXR agonists cause adverse effects, but the three highly conserved RXR ligand binding sites render the development of subtype-selective ligands a major challenge. We have fused elements of known RXR ligands to obtain a new RXR agonist chemotype on which minor structural modifications enabled the development of tools with single-subtype preference for RXRα, RXRβ, and RXRγ. Molecular modeling indicated different binding conformations and interaction patterns with the RXR LBDs as factors of preferential binding. In a phenotypic adipocyte differentiation experiment, only the RXRα preferential tool enhanced the adipogenic effects of pioglitazone, suggesting this subtype as particularly relevant in adipogenesis and highlighting the set of subtype-preferential RXR agonist tools as suitable for functional cellular studies.

Identification of novel off targets of baricitinib and tofacitinib by machine learning with a focus on thrombosis and viral infection

As there are no clear on-target mechanisms that explain the increased risk for thrombosis and viral infection or reactivation associated with JAK inhibitors, the observed elevated risk may be a result of an off-target effect. Computational approaches combined with in vitro studies can be used to predict and validate the potential for an approved drug to interact with additional (often unwanted) targets and identify potential safety-related concerns. Potential off-targets of the JAK inhibitors baricitinib and tofacitinib were identified using two established machine learning approaches based on ligand similarity. The identified targets related to thrombosis or viral infection/reactivation were subsequently validated using in vitro assays. Inhibitory activity was identified for four drug-target pairs (PDE10A [baricitinib], TRPM6 [tofacitinib], PKN2 [baricitinib, tofacitinib]). Previously unknown off-target interactions of the two JAK inhibitors were identified. As the proposed pharmacological effects of these interactions include attenuation of pulmonary vascular remodeling, modulation of HCV response, and hypomagnesemia, the newly identified off-target interactions cannot explain an increased risk of thrombosis or viral infection/reactivation. While further evidence is required to explain both the elevated thrombosis and viral infection/reactivation risk, our results add to the evidence that these JAK inhibitors are promiscuous binders and highlight the potential for repurposing.

Natural product drug discovery in the artificial intelligence era

Natural products (NPs) are primarily recognized as privileged structures to interact with protein drug targets. Their unique characteristics and structural diversity continue to marvel scientists for developing NP-inspired medicines, even though the pharmaceutical industry has largely given up. High-performance computer hardware, extensive storage, accessible software and affordable online education have democratized the use of artificial intelligence (AI) in many sectors and research areas. The last decades have introduced natural language processing and machine learning algorithms, two subfields of AI, to tackle NP drug discovery challenges and open up opportunities. In this article, we review and discuss the rational applications of AI approaches developed to assist in discovering bioactive NPs and capturing the molecular "patterns" of these privileged structures for combinatorial design or target selectivity.

Validation strategies for target prediction methods

Computational methods for target prediction, based on molecular similarity and network-based approaches, machine learning, docking and others, have evolved as valuable and powerful tools to aid the challenging task of mode of action identification for bioactive small molecules such as drugs and drug-like compounds. Critical to discerning the scope and limitations of a target prediction method is understanding how its performance was evaluated and reported. Ideally, large-scale prospective experiments are conducted to validate the performance of a model; however, this expensive and time-consuming endeavor is often not feasible. Therefore, to estimate the predictive power of a method, statistical validation based on retrospective knowledge is commonly used. There are multiple statistical validation techniques that vary in rigor. In this review we discuss the validation strategies employed, highlighting the usefulness and constraints of the validation schemes and metrics that are employed to measure and describe performance. We address the limitations of measuring only generalized performance, given that the underlying bioactivity and structural data are biased towards certain small-molecule scaffolds and target families, and suggest additional aspects of performance to consider in order to produce more detailed and realistic estimates of predictive power. Finally, we describe the validation strategies that were employed by some of the most thoroughly validated and accessible target prediction methods.

Targeting Nuclear Receptors in Neurodegeneration and Neuroinflammation

Nuclear receptors, also known as ligand-activated transcription factors, regulate gene expression upon ligand signals and present as attractive therapeutic targets especially in chronic diseases. Despite the therapeutic relevance of some nuclear receptors in various pathologies, their potential in neurodegeneration and neuroinflammation is insufficiently established. This perspective gathers preclinical and clinical data for a potential role of individual nuclear receptors as future targets in Alzheimer's disease, Parkinson's disease, and multiple sclerosis, and concomitantly evaluates the level of medicinal chemistry targeting these proteins. Considerable evidence suggests the high promise of ligand-activated transcription factors to counteract neurodegenerative diseases with a particularly high potential of several orphan nuclear receptors. However, potent tools are lacking for orphan receptors, and limited central nervous system exposure or insufficient selectivity also compromises the suitability of well-studied nuclear receptor ligands for functional studies. Medicinal chemistry efforts are needed to develop dedicated high-quality tool compounds for the therapeutic validation of nuclear receptors in neurodegenerative pathologies.

Oxaprozin Analogues as Selective RXR Agonists with Superior Properties and Pharmacokinetics

The retinoid X receptors (RXR) are ligand-activated transcription factors involved in multiple regulatory networks as universal heterodimer partners for nuclear receptors. Despite their high therapeutic potential in many pathologies, targeting of RXR has only been exploited in cancer treatment as the currently available RXR agonists suffer from exceptional lipophilicity, poor pharmacokinetics (PK), and adverse effects. Aiming to overcome the limitations and to provide improved RXR ligands, we developed a new potent RXR ligand chemotype based on the nonsteroidal anti-inflammatory drug oxaprozin. Systematic structure-activity relationship analysis enabled structural optimization toward low nanomolar potency similar to the well-established rexinoids. Cocrystal structures of the most active derivatives demonstrated orthosteric binding, and in vivo profiling revealed superior PK properties compared to current RXR agonists. The optimized compounds were highly selective for RXR activation and induced RXR-regulated gene expression in native cellular and in vivo settings suggesting them as excellent chemical tools to further explore the therapeutic potential of RXR.

Molecular Scaffold Hopping via Holistic Molecular Representation

Molecular descriptors encode a variety of molecular representations for computer-assisted drug discovery. Here, we focus on the Weighted Holistic Atom Localization and Entity Shape (WHALES) descriptors, which were originally designed for scaffold hopping from natural products to synthetic molecules. WHALES descriptors capture molecular shape and partial charges simultaneously. We introduce the key aspects of the WHALES concept and provide a step-by-step guide on how to use these descriptors for virtual compound screening and scaffold hopping. The results presented can be reproduced by using the code freely available from URL: github.com/ETHmodlab/scaffold_hopping_whales .

Cheminformatics in Natural Product-based Drug Discovery

This review seeks to provide a timely survey of the scope and limitations of cheminformatics methods in natural product-based drug discovery. Following an overview of data resources of chemical, biological and structural information on natural products, we discuss, among other aspects, in silico methods for (i) data curation and natural products dereplication, (ii) analysis, visualization, navigation and comparison of the chemical space, (iii) quantification of natural product-likeness, (iv) prediction of the bioactivities (virtual screening, target prediction), ADME and safety profiles (toxicity) of natural products, (v) natural products-inspired de novo design and (vi) prediction of natural products prone to cause interference with biological assays. Among the many methods discussed are rule-based, similarity-based, shape-based, pharmacophore-based and network-based approaches, docking and machine learning methods.

Comprehensive Set of Tertiary Complex Structures and Palmitic Acid Binding Provide Molecular Insights into Ligand Design for RXR Isoforms

The retinoid X receptor (RXR) is a ligand-sensing transcription factor acting mainly as a universal heterodimer partner for other nuclear receptors. Despite presenting as a potential therapeutic target for cancer and neurodegeneration, adverse effects typically observed for RXR agonists, likely due to the lack of isoform selectivity, limit chemotherapeutic application of currently available RXR ligands. The three human RXR isoforms exhibit different expression patterns; however, they share high sequence similarity, presenting a major obstacle toward the development of subtype-selective ligands. Here, we report the discovery of the saturated fatty acid, palmitic acid, as an RXR ligand and disclose a uniform set of crystal structures of all three RXR isoforms in an active conformation induced by palmitic acid. A structural comparison revealed subtle differences among the RXR subtypes. We also observed an ability of palmitic acid as well as myristic acid and stearic acid to induce recruitment of steroid receptor co-activator 1 to the RXR ligand-binding domain with low micromolar potencies. With the high, millimolar endogenous concentrations of these highly abundant lipids, our results suggest their potential involvement in RXR signaling.

Control of Synthetic Feasibility of Compounds Generated with CReM

Synthetic feasibility of compounds generated with de novo approaches is one of the main issues, which may limit their applicability. Many of the de novo generation approaches do not address this issue. Here, we studied the recently implemented chemically reasonable mutations approach (CReM) and the ways how one could indirectly control synthetic complexity of generated compounds and how this affected the target scores for Guacamol benchmark tasks. We found a clear trade-off between synthetic complexity and target scores and demonstrated that CReM-based solutions were competitive to reference approaches, which were explicitly biased by synthetic feasibility of generated compounds.

A Turing Test for Molecular Generators

Machine learning approaches promise to accelerate and improve success rates in medicinal chemistry programs by more effectively leveraging available data to guide a molecular design. A key step of an automated computational design algorithm is molecule generation, where the machine is required to design high-quality, drug-like molecules within the appropriate chemical space. Many algorithms have been proposed for molecular generation; however, a challenge is how to assess the validity of the resulting molecules. Here, we report three Turing-inspired tests designed to evaluate the performance of molecular generators. Profound differences were observed between the performance of molecule generators in these tests, highlighting the importance of selection of the appropriate design algorithms for specific circumstances. One molecule generator, based on match molecular pairs, performed excellently against all tests and thus provides a valuable component for machine-driven medicinal chemistry design workflows.

NURA: A curated dataset of nuclear receptor modulators

Nuclear receptors (NRs) are key regulators of human health and constitute a relevant target for medicinal chemistry applications as well as for toxicological risk assessment. Several open databases dedicated to small molecules that modulate NRs exist; however, depending on their final aim (i.e., adverse effect assessment or drug design), these databases contain a different amount and type of annotated molecules, along with a different distribution of experimental bioactivity values. Stemming from these considerations, in this work we aim to provide a unified dataset, NURA (NUclear Receptor Activity) dataset, collecting curated information on small molecules that modulate NRs, to be intended for both pharmacological and toxicological applications. NURA contains bioactivity annotations for 15,247 molecules and 11 selected NRs, and it was obtained by integrating and curating data from toxicological and pharmacological databases (i.e., Tox21, ChEMBL, NR-DBIND and BindingDB). Our results show that NURA dataset is a useful tool to bridge the gap between toxicology- and medicinal-chemistry-related databases, as it is enriched in terms of number of molecules, structural diversity and covered atomic scaffolds compared to the single sources. To the best of our knowledge, NURA dataset is the most exhaustive collection of small molecules annotated for their modulation of the chosen nuclear receptors. NURA dataset is intended to support decision-making in pharmacology and toxicology, as well as to contribute to data-driven applications, such as machine learning. The dataset and the data curation pipeline can be downloaded free of charge on Zenodo at the following DOI: https://doi.org/10.5281/zenodo.3991561.

De Novo Drug Design of Targeted Chemical Libraries Based on Artificial Intelligence and Pair-Based Multiobjective Optimization

Artificial intelligence and multiobjective optimization represent promising solutions to bridge chemical and biological landscapes by addressing the automated de novo design of compounds as a result of a humanlike creative process. In the present study, we conceived a novel pair-based multiobjective approach implemented in an adapted SMILES generative algorithm based on recurrent neural networks for the automated de novo design of new molecules whose overall features are optimized by finding the best trade-offs among relevant physicochemical properties (MW, logP, HBA, HBD) and additional similarity-based constraints biasing specific biological targets. In this respect, we carried out the de novo design of chemical libraries targeting neuraminidase, acetylcholinesterase, and the main protease of severe acute respiratory syndrome coronavirus 2. Several quality metrics were employed to assess drug-likeness, chemical feasibility, diversity content, and validity. Molecular docking was finally carried out to better evaluate the scoring and posing of the de novo generated molecules with respect to X-ray cognate ligands of the corresponding molecular counterparts. Our results indicate that artificial intelligence and multiobjective optimization allow us to capture the latent links joining chemical and biological aspects, thus providing easy-to-use options for customizable design strategies, which are especially effective for both lead generation and lead optimization. The algorithm is freely downloadable at https://github.com/alberdom88/moo-denovo and all of the data are available as Supporting Information.

BRADSHAW: a system for automated molecular design

This paper introduces BRADSHAW (Biological Response Analysis and Design System using an Heterogenous, Automated Workflow), a system for automated molecular design which integrates methods for chemical structure generation, experimental design, active learning and cheminformatics tools. The simple user interface is designed to facilitate access to large scale automated design whilst minimising software development required to introduce new algorithms, a critical requirement in what is a very fast moving field. The system embodies a philosophy of automation, best practice, experimental design and the use of both traditional cheminformatics and modern machine learning algorithms.

CReM: chemically reasonable mutations framework for structure generation

Structure generators are widely used in de novo design studies and their performance substantially influences an outcome. Approaches based on the deep learning models and conventional atom-based approaches may result in invalid structures and fail to address their synthetic feasibility issues. On the other hand, conventional reaction-based approaches result in synthetically feasible compounds but novelty and diversity of generated compounds may be limited. Fragment-based approaches can provide both better novelty and diversity of generated compounds but the issue of synthetic complexity of generated structure was not explicitly addressed before. Here we developed a new framework of fragment-based structure generation that, by design, results in the chemically valid structures and provides flexible control over diversity, novelty, synthetic complexity and chemotypes of generated compounds. The framework was implemented as an open-source Python module and can be used to create custom workflows for the exploration of chemical space.

A Novel Biphenyl-based Chemotype of Retinoid X Receptor Ligands Enables Subtype and Heterodimer Preferences

The nuclear retinoid X receptors (RXRs) are key ligand sensing transcription factors that serve as universal nuclear receptor heterodimer partners and are thus involved in numerous physiological processes. Therapeutic targeting of RXRs holds high potential but available RXR activators suffer from limited safety. Selectivity for RXR subtypes or for certain RXR heterodimers are promising strategies for safer RXR modulation. Here, we report systematic structure-activity relationship studies on biphenyl carboxylates as new RXR ligand chemotype. We discovered specific structural modifications that enhance potency on RXRs, govern subtype preference, and vary modulation of different RXR heterodimers. Fusion of these structural motifs enabled specific tuning of subtype preferential profiles with markedly improved potency. Our results provide further evidence that RXR subtype selective ligands can be designed and present a novel chemotype of RXR modulators that can be tuned for subtype and heterodimer preferences.

Alternative retinoid X receptor (RXR) ligands

Retinoid X receptors (RXRs) control a wide variety of functions by virtue of their dimerization with other nuclear hormone receptors (NRs), contributing thereby to activities of different signaling pathways. We review known RXR ligands as transcriptional modulators of specific RXR-dimers and the associated biological processes. We also discuss the physiological relevance of such ligands, which remains frequently a matter of debate and which at present is best met by member(s) of a novel family of retinoids, postulated as Vitamin A5. Through comparison with other natural, but also with synthetic ligands, we discuss high diversity in the modes of ligand binding to RXRs resulting in agonistic or antagonistic profiles and selectivity towards specific subtypes of permissive heterodimers. Despite such diversity, direct ligand binding to the ligand binding pocket resulting in agonistic activity was preferentially preserved in the course of animal evolution pointing to its functional relevance, and potential for existence of other, species-specific endogenous RXR ligands sharing the same mode of function.

Automated De Novo Drug Design: Are We Nearly There Yet?

Medicinal chemistry and, in particular, drug design have often been perceived as more of an art than a science. The many unknowns of human disease and the sheer complexity of chemical space render decision making in medicinal chemistry exceptionally demanding. Computational models can assist the medicinal chemist in this endeavour. Provided here is an overview of recent examples of automated de novo molecular design, a discussion of the concepts and computational approaches involved, and the daring prediction of some of the possibilities and limitations of drug design using machine intelligence.

A review of the molecular design and biological activities of RXR agonists

An attractive approach to combat disease is to target theregulation of cell function. At the heart of this task are nuclear receptors (NRs); which control functions such as gene transcription. Arguably, the key player in this regulatory machinery is the retinoid X receptor (RXR). This NR associates with a third of the NRs found in humans. Scientists have hypothesized that controlling the activity of RXR is an attractive approach to control cellular functions that modulate diseases such as cancer, diabetes, Alzheimer's disease and Parkinson's disease. In this review, we will describe the key features of the RXR, present a historic perspective of the first RXR agonists, and discuss various templates that have been reported to activate RXR with a focus on their molecular structure, biological activity, and limitations. Finally, we will present an outlook of the field and future directions and considerations to synthesize or modulate RXR agonists to make these compounds a clinical reality.

Machine Learning Consensus To Predict the Binding to the Androgen Receptor within the CoMPARA Project

The nuclear androgen receptor (AR) is one of the most relevant biological targets of Endocrine Disrupting Chemicals (EDCs), which produce adverse effects by interfering with hormonal regulation and endocrine system functioning. This paper describes novel in silico models to identify organic AR modulators in the context of the Collaborative Modeling Project of Androgen Receptor Activity (CoMPARA), coordinated by the National Center of Computational Toxicology (U.S. Environmental Protection Agency). The collaborative project involved 35 international research groups to prioritize the experimental tests of approximatively 40k compounds, based on the predictions provided by each participant. In this paper, we describe our machine learning approach to predict the binding to AR, which is based on a consensus of a multivariate Bernoulli Naive Bayes, a Random Forest, and N-Nearest Neighbor classification models. The approach was developed in compliance with the Organization of Economic Cooperation and Development (OECD) principles, trained on 1687 ToxCast molecules classified according to 11 in vitro assays, and further validated on a set of 3,882 external compounds. The models provided robust and reliable predictions and were used to gather novel data-driven insights on the structural features related to AR binding, agonism, and antagonism.

NP-Scout: Machine Learning Approach for the Quantification and Visualization of the Natural Product-Likeness of Small Molecules

Natural products (NPs) remain the most prolific resource for the development of small-molecule drugs. Here we report a new machine learning approach that allows the identification of natural products with high accuracy. The method also generates similarity maps, which highlight atoms that contribute significantly to the classification of small molecules as a natural product or synthetic molecule. The method can hence be utilized to (i) identify natural products in large molecular libraries, (ii) quantify the natural product-likeness of small molecules, and (iii) visualize atoms in small molecules that are characteristic of natural products or synthetic molecules. The models are based on random forest classifiers trained on data sets consisting of more than 265,000 to 322,000 natural products and synthetic molecules. Two-dimensional molecular descriptors, MACCS keys and Morgan2 fingerprints were explored. On an independent test set the models reached areas under the receiver operating characteristic curve (AUC) of 0.997 and Matthews correlation coefficients (MCCs) of 0.954 and higher. The method was further tested on data from the Dictionary of Natural Products, ChEMBL and other resources. The best-performing models are accessible as a free web service at http://npscout.zbh.uni-hamburg.de/npscout.

Discovery of Novel Molecular Frameworks of Farnesoid X Receptor Modulators by Ensemble Machine Learning

The bile acid activated transcription factor farnesoid X receptor (FXR) has revealed therapeutic potential as a molecular drug target for the treatment of hepatic and metabolic disorders. Despite strong efforts in FXR ligand development, the structural diversity among the known FXR modulators is limited. Only four molecular frameworks account for more than 50 % of the FXR modulators annotated in ChEMBL. Here, we leverage machine learning methods to expand the chemical space of FXR-targeting small molecules by employing an ensemble of three complementary machine learning approaches. A counter-propagation artificial neural network, a k-nearest neighbor learner, and a three-dimensional pharmacophore descriptor were combined to retrieve novel FXR ligands from a collection of more than 3 million compounds. The ensemble machine learning model identified six new FXR modulators among ten top-ranked candidates. These active hits comprise both FXR activators and antagonists with micromolar potencies. With four novel FXR ligand scaffolds, these computationally identified bioactive compounds appreciably expand the chemical space of known FXR modulators and may serve as starting points for hit-to-lead expansion.

Scaffold-Hopping from Synthetic Drugs by Holistic Molecular Representation

The discovery of novel ligand chemotypes allows to explore uncharted regions in chemical space, thereby potentially improving synthetic accessibility, potency, and the drug-likeness of molecules. Here, we demonstrate the scaffold-hopping ability of the new Weighted Holistic Atom Localization and Entity Shape (WHALES) molecular descriptors compared to seven state-of-the-art molecular representations on 30,000 compounds and 182 biological targets. In a prospective application, we apply WHALES to the discovery of novel retinoid X receptor (RXR) modulators. WHALES descriptors identified four agonists with innovative molecular scaffolds, populating uncharted regions of the chemical space. One of the agonists, possessing a rare non-acidic chemotype, revealed high selectivity on 12 nuclear receptors and comparable efficacy as bexarotene on induction of ATP-binding cassette transporter A1, angiopoietin like protein 4 and apolipoprotein E. The outcome of this research supports WHALES as an innovative tool to explore novel regions of the chemical space and to detect novel bioactive chemotypes by straightforward similarity searching.