Small Molecule Drug Discovery Research Articles

Abstract PR005: Helping academic investigators develop small molecules for the clinic: The NCI Developmental Therapeutics Program and Stepping-Stones

Abstract Academic researchers working in small molecule drug discovery and development face unique challenges – key among these include limited access to the full range of resources needed, inadequate funding which does not cover costs for routine/iterative product development tasks, and unavailability of expertise for the full range of regulatory critical path steps required for an IND. For academic institutions which lack deep resources, specialized core facilities, or endowments, these challenges are even more dire. In recognition of these realities, the NCI Developmental Therapeutics Program (DTP) expanded its capability through creation of the Stepping-Stones program, which provides resources to academic-based researchers who are working towards advancing their small molecule therapeutic candidates from ‘bench to bedside’. Through the Stepping-Stones program, NCI grant-supported small molecule therapeutics focused projects demonstrating high productivity are eligible to obtain access to NCI/DTP drug development capabilities and resources to fill in ‘gaps’ for specific activities that are not covered through their existing funding, but which are necessary to facilitate advancement along the critical path. Examples of services provided have included molecular profiling for ADME/DMPK studies, screens for secondary pharmacology and in vitro safety, synthesis optimization and chemical synthesis, and formulation development, to name a few. By filling in such discrete ‘gaps’, project teams can ‘de-risk’ a candidate earlier and develop stronger data packages to better position themselves for attracting additional investment for development toward clinical testing, or as the data warrants pursue an earlier re-direction of effort before going too far down the critical path. An overview of Stepping-Stones including details for how to be considered for this program and its interactions with other programs at NCI that support drug discovery and development is described, in addition to an overview of DTP’s discovery and development services which usher R&D programs from discovery through preclinical development and IND-enabling studies to support first-in-human trials. From this presentation, academic based ‘drug hunters’ will obtain knowledge about the Stepping-Stones program as a support mechanism to enable their continued efforts and gain an improved understanding about the ‘ecosystem’ of drug discovery and development programs offered by the NCI for the extramural community of researchers. Citation Format: Sharad Verma, Morgan O' Hayre, Rose Aurigemma. Helping academic investigators develop small molecules for the clinic: The NCI Developmental Therapeutics Program and Stepping-Stones [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Optimizing Therapeutic Efficacy and Tolerability through Cancer Chemistry; 2024 Dec 9-11; Toronto, Ontario, Canada. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(12_Suppl):Abstract nr PR005

Current Status of Computational Approaches for Small Molecule Drug Discovery.

2024 has been an exciting year for computational sciences, with the Nobel Prize in Physics awarded for "artificial neural network" and the Nobel Prize in Chemistry presented for "protein structure prediction and design". Given the rapid advancements in Computer-Aided Drug Design (CADD) and Artificial Intelligence in Drug Discovery (AIDD), a document summarizing their current standing and future directions would be timely and relevant to the readership of Journal of Medicinal Chemistry. This piece of commentary aims to highlight recent developments, key challenges, and potential synergies between these fields, contributing to ongoing discussions in the literature and scientific blogs.

Chemical Space for Peptide-based Antimicrobials.

Multidrug-resistant (MDR) bacteria represent a global public health threat, and antimicrobial peptides (AMPs), derived from naturally occurring linear or cyclic peptides, can provide the solution. However, most AMPs are sensitive to proteases and have poor pharmacokinetics. The EU-funded ERC Advanced Grant SPACE4AMPS aims to identify new AMPs by applying the concepts of chemical space and ligand-based virtual screening, which are well known for small molecule drug discovery, to the world of peptides. We create virtual libraries of peptides and related molecules and use these approaches to select a few tens of compounds for synthesis and detailed. evaluation of antibacterial, toxicity and stability effects. Recent results and prospects of this ongoing project are presented in this review.

Enhancing the Predictive Power of Machine Learning Models through a Chemical Space Complementary DEL Screening Strategy.

DNA-encoded library (DEL) technology is an effective method for small molecule drug discovery, enabling high-throughput screening against target proteins. While DEL screening produces extensive data, it can reveal complex patterns not easily recognized by human analysis. Lead compounds from DEL screens often have higher molecular weights, posing challenges for drug development. This study refines traditional DELs by integrating alternative techniques like photocross-linking screening to enhance chemical diversity. Combining these methods improved predictive performance for small molecule identification models. Using this approach, we predicted active small molecules for BRD4 and p300, achieving hit rates of 26.7 and 35.7%. Notably, the identified compounds exhibit smaller molecular weights and better modification potential compared to traditional DEL molecules. This research demonstrates the synergy between DEL and AI technologies, enhancing drug discovery.

The future of machine learning for small-molecule drug discovery will be driven by data.

Many studies have prophesied that the integration of machine learning techniques into small-molecule therapeutics development will help to deliver a true leap forward in drug discovery. However, increasingly advanced algorithms and novel architectures have not always yielded substantial improvements in results. In this Perspective, we propose that a greater focus on the data for training and benchmarking these models is more likely to drive future improvement, and explore avenues for future research and strategies to address these data challenges.

Discovery of Small Molecule Drugs with Therapeutic Potential for Hyperoxalemia, Hyperoxaluria, and Related Kidney Stones

Discovery of Small Molecule Drugs with Therapeutic Potential for Hyperoxalemia, Hyperoxaluria, and Related Kidney Stones

Combating cancer stem cells: RNA m6A methylation and small-molecule drug discovery

Cancer stem cells (CSCs) are a small population of less differentiated cells with robust self-renewal ability. CSCs have been recognized as the root cause of tumor initiation, progression, relapse, and drug resistance. Recent studies from us and others have highlighted that N6-methyladenosine (m6A), the most prevalent modification in mRNA, plays a crucial role in carcinogenesis and CSC homeostasis. Dysregulation of the m6A modification machinery has been implicated in CSC survival and self-renewal, thereby regulating cancer progression and therapeutic resistance. In this review, we provide an overview of the roles and molecular mechanisms of the RNA m6A modification machinery in CSC survival and self-renewal. Additionally, we summarize the currently known small-molecule inhibitors targeting the dysregulated m6A modification machinery and discuss proof-of-concept studies focusing on the efficacy of these compounds in eliminating CSCs and cancers.

The Development and Application of KinomePro-DL: A Deep Learning Based Online Small Molecule Kinome Selectivity Profiling Prediction Platform.

Characterizing the kinome selectivity profiles of kinase inhibitors is essential in the early stages of novel small-molecule drug discovery. This characterization is critical for interpreting potential adverse events caused by off-target polypharmacology effects and provides unique pharmacological insights for drug repurposing development of existing kinase inhibitor drugs. However, experimental profiling of whole kinome selectivity is still time-consuming and resource-demanding. Here, we report a deep learning classification model using an in-house built data set of inhibitors against 191 well-representative kinases constructed based on a novel strategy by systematically cleaning and integrating six public data sets. This model, a multitask deep neural network, predicts the kinome selectivity profiles of compounds with novel structures. The model demonstrates excellent predictive performance, with auROC, prc-AUC, Accuracy, and Binary_cross_entropy of 0.95, 0.92, 0.90, and 0.37, respectively. It also performs well in a priori testing for inhibitors targeting different categories of proteins from internal compound collections, significantly improving over similar models on data sets from practical application scenarios. Integrated to subsequent machine learning-enhanced virtual screening workflow, novel CDK2 kinase inhibitors with potent kinase inhibitory activity and excellent kinome selectivity profiles are successfully identified. Additionally, we developed a free online web server, KinomePro-DL, to predict the kinome selectivity profiles and kinome-wide polypharmacology effects of small molecules (available on kinomepro-dl.pharmablock.com). Uniquely, our model allows users to quickly fine-tune it with their own training data sets, enhancing both prediction accuracy and robustness.

Lessons learned during the journey of data: from experiment to model for predicting kinase affinity, selectivity, polypharmacology, and resistance.

Recent advances in machine learning (ML) are reshaping drug discovery. Structure-based ML methods use physically-inspired models to predict binding affinities from protein:ligand complexes. These methods promise to enable the integration of data for many related targets, which addresses issues related to data scarcity for single targets and could enable generalizable predictions for a broad range of targets, including mutants. In this work, we report our experiences in building KinoML, a novel framework for ML in target-based small molecule drug discovery with an emphasis on structure-enabled methods. KinoML focuses currently on kinases as the relative structural conservation of this protein superfamily, particularly in the kinase domain, means it is possible to leverage data from the entire superfamily to make structure-informed predictions about binding affinities, selectivities, and drug resistance. Some key lessons learned in building KinoML include: the importance of reproducible data collection and deposition, the harmonization of molecular data and featurization, and the choice of the right data format to ensure reusability and reproducibility of ML models. As a result, KinoML allows users to easily achieve three tasks: accessing and curating molecular data; featurizing this data with representations suitable for ML applications; and running reproducible ML experiments that require access to ligand, protein, and assay information to predict ligand affinity. Despite KinoML focusing on kinases, this framework can be applied to other proteins. The lessons reported here can help guide the development of platforms for structure-enabled ML in other areas of drug discovery.

Open-Source Approach to GPU-Accelerated Substructure Search.

Chemical substructure search is a critical task in medicinal chemistry and small-molecule drug discovery, enabling the retrieval of molecules from databases based on specific chemical features. While systems exist for this purpose, the challenge of efficient and swift searching persists, particularly as data storage migrates to the cloud, introducing new complexities. This study provides a comprehensive analysis of chemical substructure searches, showcasing the benefits of graphics processing unit-accelerated fingerprint screening. The research highlights strategies for optimizing performance, making significant advancements in substructure searching, a pivotal aspect of drug discovery and molecular research. The accessible and scalable nature of the proposed approach makes it a valuable resource for scientists aiming to enhance their substructure search capabilities.

Development of Drug Discovery Platforms Using Artificial Intelligence and Cheminformatics.

Recently, remarkable progress has been achieved in artificial intelligence (AI), including machine learning. Various AI models have been proposed for drug discovery, including the design of small molecules, activity prediction, and three-dimensional (3D) structure prediction of proteins. AI consists of diverse elements, including information retrieval and machine learning, and can be used in a wide range of drug discovery scenarios. In this review, we focused on AI for small-molecule drug discovery with respect to molecular design, activity prediction, and prediction of the binding poses of compounds to target molecules. We also discussed the applications of AI in academic drug discovery.

Direct C4 and C2 C-H Amination of Heteroarenes Using I(III) Reagents via a Cross Azine Coupling.

Aminated nitrogen heterocycles are valuable motifs across numerous chemical industries, perhaps most notably in small molecule drug discovery. While numerous strategies for installing nitrogen atoms onto azaarenes exist, most require prefunctionalization and methods for direct C-H amination are almost entirely limited to position C2. Herein, we report a method for the direct C2 and C4 C-H amination of fused azaarenes via in situ activation with a bispyridine-ligated I(III) reagent, [(Py)2IPh]2OTf, or Py-HVI. Unlike commonly used N-oxide chemistry, the method requires no preoxidation of the azaarene and provides unprecedented direct access to C4 amination products. The resulting N-heterocyclic pyridinium salts can be isolated via simple trituration. The free amine can be liberated under mild Zincke aminolysis, or the amination and cleavage can be telescoped to a one-pot process. The scope of the method is broad; the conditions are mild and operationally simple, and the aminated products are produced in good to excellent yields. Computational studies provide insights into the mechanism of activation, which involves an unusual direct nucleophilic functionalization of an I(III) ligand, as well as a kinetic basis for the observed C2 and C4 amination products.

Label-free, real-time monitoring of membrane binding events at zeptomolar concentrations using frequency-locked optical microresonators

G-protein coupled receptors help regulate cellular function and communication, and are targets of small molecule drug discovery efforts. Conventional techniques to probe these interactions require labels and large amounts of receptor to achieve satisfactory sensitivity. Here, we use frequency-locked optical microtoroids for label-free characterization of membrane interactions in vitro at zeptomolar concentrations for the kappa opioid receptor and its native agonist dynorphin A 1-13, as well as big dynorphin (dynorphin A and dynorphin B) using a supported biomimetic membrane. The measured affinity of the agonist dynorphin A 1-13 to the κ-opioid receptor was also measured and found to be 3.1 nM. Radioligand assays revealed a dissociation constant in agreement with this value (1.1 nM). The limit of detection for the κOR/DynA 1-13 was calculated as 180 zM. The binding of Cholera Toxin B-monosialotetrahexosyl ganglioside was also monitored in real-time and an equilibrium dissociation constant of 1.53 nM was found. Our biosensing platform provides a method for highly sensitive real-time characterization of membrane embedded protein binding kinetics that is rapid and label-free, for drug discovery and toxin screening among other applications.

SMTRI: A deep learning-based web service for predicting small molecules that target miRNA-mRNA interactions

Mature microRNAs are short, single-stranded RNAs that bind to target mRNAs and induce translational repression and gene silencing. Many microRNAs discovered in animals have been implicated in diseases and have recently been pursued as therapeutic targets. However, conventional pharmacological screening for candidate small-molecule drugs can be time-consuming and labor-intensive. Therefore, developing a computational program to assist mature microRNA-targeted drug discovery in silico is desirable. Our previous work (https://doi.org/10.1002/advs.201903451) revealed that the unique functional loops formed during Argonaute-mediated microRNA-mRNA interactions have stable structural characteristics and may serve as potential targets for small molecule drug discovery. Developing drugs specifically targeting disease-related mature microRNAs and their target mRNAs would avoid affecting unrelated ones. Here, we presented SMTRI, a convolutional neural network-based approach for efficiently predicting small molecules that target RNA secondary structural motifs formed by interactions between microRNAs and their target mRNAs. Measured on three additional testing sets, SMTRI outperformed state-of-the-art algorithms by 12.9-30.3% in AUC and 2.0-18.4% in Accuracy. Moreover, four case studies on the published experimentally validated RNA-targeted small molecules also revealed the reliability of SMTRI. Our method is available as a user-friendly web server at http://smtri.net.

Application of Mechanistic Multiparameter Optimization and Large-Scale In Vitro to In Vivo Pharmacokinetics Correlations to Small-Molecule Therapeutic Projects.

Computational chemistry and machine learning are used in drug discovery to predict the target-specific and pharmacokinetic properties of molecules. Multiparameter optimization (MPO) functions are used to summarize multiple properties into a single score, aiding compound prioritization. However, over-reliance on subjective MPO functions risks reinforcing human bias. Mechanistic modeling approaches based on physiological relevance can be adapted to meet different potential key objectives of the project (e.g., minimizing dose, maximizing safety margins, and/or minimizing drug-drug interaction risk) while retaining the same underlying model structure. The current work incorporates recent approaches to predict in vivo pharmacokinetic (PK) properties and validates in vitro to in vivo correlation analysis to support mechanistic PK MPO. Examples of use and impact in small-molecule drug discovery projects are provided. Overall, the mechanistic MPO identifies 83% of the compounds considered as short-listed for clinical experiments in the top second percentile, and 100% in the top 10th percentile, resulting in an area under the receiver operating characteristic curve (AUCROC) > 0.95. In addition, the MPO score successfully recapitulates the chronological progression of the optimization process across different scaffolds. Finally, the MPO scores for compounds characterized in pharmacokinetics experiments are markedly higher compared with the rest of the compounds being synthesized, highlighting the potential of this tool to reduce the reliance on in vivo testing for compound screening.

Do Molecular Fingerprints Identify Diverse Active Drugs in Large-Scale Virtual Screening? (No).

Computational approaches for small-molecule drug discovery now regularly scale to the consideration of libraries containing billions of candidate small molecules. One promising approach to increased the speed of evaluating billion-molecule libraries is to develop succinct representations of each molecule that enable the rapid identification of molecules with similar properties. Molecular fingerprints are thought to provide a mechanism for producing such representations. Here, we explore the utility of commonly used fingerprints in the context of predicting similar molecular activity. We show that fingerprint similarity provides little discriminative power between active and inactive molecules for a target protein based on a known active-while they may sometimes provide some enrichment for active molecules in a drug screen, a screened data set will still be dominated by inactive molecules. We also demonstrate that high-similarity actives appear to share a scaffold with the query active, meaning that they could more easily be identified by structural enumeration. Furthermore, even when limited to only active molecules, fingerprint similarity values do not correlate with compound potency. In sum, these results highlight the need for a new wave of molecular representations that will improve the capacity to detect biologically active molecules based on their similarity to other such molecules.

Machine-Learning-Guided Peptide Drug Discovery: Development of GLP-1 Receptor Agonists with Improved Drug Properties.

Peptide-based drug discovery has surged with the development of peptide hormone-derived analogs for the treatment of diabetes and obesity. Machine learning (ML)-enabled quantitative structure-activity relationship (QSAR) approaches have shown great promise in small molecule drug discovery but have been less successful in peptide drug discovery due to limited data availability. We have developed a peptide drug discovery platform called streaMLine, enabling rigorous design, synthesis, screening, and ML-driven analysis of large peptide libraries. Using streaMLine, this study systematically explored secretin as a peptide backbone to generate potent, selective, and long-acting GLP-1R agonists with improved physicochemical properties. We synthesized and screened a total of 2688 peptides and applied ML-guided QSAR to identify multiple options for designing stable and potent GLP-1R agonists. One candidate, GUB021794, was profiled in vivo (S.C., 10 nmol/kg QD) and showed potent body weight loss in diet-induced obese mice and a half-life compatible with once-weekly dosing.

Portable fiber-based double nanohole optical tweezer for trapping small proteins

We demonstrate the trapping and analysis of individual proteins using a portable optical fiber tweezer setup with a double-nanohole in a gold film coating the fiber’s end and aligned with the fiber core. The instrument was used to trap and analyze cytochrome c, carbonic anhydrase, bovine serum albumin, and PR65 (wild-type and various point mutants). This approach was compared with a free-space optical tweezer setup that requires alignment of the laser focus to the aperture, whereas the fiber-based approach is both portable and alignment-free, which holds promise for applications in antibody discovery, small molecule drug discovery, protein interaction analysis and other applications using the standard well-plate format.

Occurrence of "Natural Selection" in Successful Small Molecule Drug Discovery.

Published compounds from ChEMBL version 32 are used to seek evidence for the occurrence of "natural selection" in drug discovery. Three measures of natural product (NP) character were applied, to compare time- and target-matched compounds reaching the clinic (clinical compounds in phase 1-3 development and approved drugs) with background compounds (reference compounds). Pseudo-NPs (PNPs), containing NP fragments combined in ways inaccessible by nature, are increasing over time, reaching 67% of clinical compounds first disclosed since 2010. PNPs are 54% more likely to be found in post-2008 clinical versus reference compounds. The majority of target classes show increased clinical compound NP character versus their reference compounds. Only 176 NP fragments appear in >1000 clinical compounds published since 2008, yet these make up on average 63% of the clinical compound's core scaffolds. There is untapped potential awaiting exploitation, by applying nature's building blocks─"natural intelligence"─to drug design.

Fluorescence Thermal Shift Assay based High Through-put Screening in Small Molecule Drug Discovery: A Brief Overview

Background: The Early drug discovery process was majorly phenotypic, lacking target specificity, and mechanism of action causing short and/or long-term toxicological impacts on the patients leading to several drug withdrawals from the market. Biochemical procedures and methods used in drug discovery and development make it lengthy in terms of time requiring approx. 15 years with the investment of billions of dollars for a new drug molecule. Technological advancement leading to the identification of protein structures through crystallography paved a new path toward targeted drug discovery. Objective: This paper provides a brief overview of the use of the fluorescence thermal shift assay (FTSA) as a high throughput (HTS) screening technique in small molecule drug discovery. Methods: The article discusses the principles of thermal denaturation techniques and how they are applied in drug discovery. It highlights the advantages of FTSA over other biophysical methods, including its label-free, cost-effective, and high HTS nature. The review also presents a comparative analysis of FTSA with techniques such as isothermal titration calorimetry (ITC), mass spectrometry (MS), nuclear magnetic resonance (NMR), surface plasmon resonance (SPR), X-ray crystallography(X-RD), circular dichroism (CD), and differential scanning calorimetry (DSC). Results: The FTSA technique is described as a powerful tool that addresses the limitations of larger protein requirements, specific assay development, and low HTS associated with other thermal denaturation methods. Unlike biochemical and cellular assays that measure secondary effects, FTSA directly estimates the binding of small molecules to target proteins. Conclusion: The review concludes that FTSA is the only biophysical method that satisfies the three key prerequisites of being label-free, cost-effective, and suitable for high HTS screening in small molecule drug discovery.