Virtual Screening of a Chemically Diverse "Superscaffold" Library Enables Ligand Discovery for a Key GPCR Target.

The advent of ultra-large libraries of drug-like compounds has significantly broadened the possibilities in structure-based virtual screening, accelerating the discovery and optimization of high-quality lead chemotypes for diverse clinical targets. Compared to traditional high-throughput screening, which is constrained to libraries of approximately one million compounds, the ultra-large virtual screening approach offers substantial advantages in both cost and time efficiency. By expanding the chemical space with compounds synthesized from easily accessible and reproducible reactions and utilizing a large, diverse set of building blocks, we can enhance both the diversity and quality of the discovered lead chemotypes. In this study, we explore new chemical spaces using reactions of sulfur(VI) fluorides to create a combinatorial library consisting of several hundred million compounds. We screened this virtual library for cannabinoid type II receptor (CB2) antagonists using the high-resolution structure in conjunction with a rationally designed antagonist, AM10257. The top-predicted compounds were then synthesized and tested in vitro for CB2 binding and functional antagonism, achieving an experimentally validated hit rate of 55%. Our findings demonstrate the effectiveness of reliable reactions, such as sulfur fluoride exchange, in diversifying ultra-large chemical spaces and facilitate the discovery of new lead compounds for important biological targets.

Allosteric Modulation of the YAP/TAZ-TEAD Interaction by Palmitoylation and Small-Molecule Inhibitors.

The Hippo signaling pathway is a highly conserved signaling network that plays a central role in regulating cellular growth, proliferation, and organ size. This pathway consists of a kinase cascade that integrates various upstream signals to control the activation or inactivation of YAP/TAZ proteins. Phosphorylated YAP/TAZ is sequestered in the cytoplasm; however, when the Hippo pathway is deactivated, it translocates into the nucleus, where it associates with TEAD transcription factors. This partnership is instrumental in regulating the transcription of progrowth and antiapoptotic genes. Thus, in many cancers, aberrantly hyperactivated YAP/TAZ promotes oncogenesis by contributing to cancer cell proliferation, metastasis, and therapy resistance. Because YAP and TAZ exert their oncogenic effects by binding with TEAD, it is critical to understand this key interaction to develop cancer therapeutics. Previous research has indicated that TEAD undergoes autopalmitoylation at a conserved cysteine, and small molecules that inhibit TEAD palmitoylation disrupt effective YAP/TAZ binding. However, how exactly palmitoylation contributes to YAP/TAZ-TEAD interactions and how the TEAD palmitoylation inhibitors disrupt this interaction remains unknown. Utilizing molecular dynamics simulations, our investigation not only provides detailed atomistic insight into the YAP/TAZ-TEAD dynamics but also unveils that the inhibitor studied influences the binding of YAP and TAZ to TEAD in distinct manners. This discovery has significant implications for the design and deployment of future molecular interventions targeting this interaction.

Mitochondria-Selective Dicationic Small-Molecule Ligand Targeting G-Quadruplex Structures for Human Colorectal Cancer Therapy.

Mitochondria are important drug targets for anticancer and other disease therapies. Certain human mitochondrial DNA sequences capable of forming G-quadruplex structures (G4s) are emerging drug targets of small molecules. Despite some mitochondria-selective ligands being reported for drug delivery against cancers, the ligand design is mostly limited to the triphenylphosphonium scaffold. The ligand designed with lipophilic small-sized scaffolds bearing multipositive charges targeting the unique feature of high mitochondrial membrane potential (MMP) is lacking and most mitochondria-selective ligands are not G4-targeting. Herein, we report a new small-sized dicationic lipophilic ligand to target MMP and mitochondrial DNA G4s to enhance drug delivery for anticancer. The ligand showed marked alteration of mitochondrial gene expression and substantial induction of ROS production, mitochondrial dysfunction, DNA damage, cellular senescence, and apoptosis. The ligand also exhibited high anticancer activity against HCT116 cancer cells (IC50, 3.4 µM) and high antitumor efficacy in the HCT116 tumor xenograft mouse model (∼70% tumor weight reduction).

Discovery of Small Molecule Interleukin 17A Inhibitors with Novel Binding Mode and Stoichiometry: Optimization of DNA-Encoded Chemical Library Hits to In Vivo Active Compounds.

Dysregulation of IL17A drives numerous inflammatory and autoimmune disorders with inhibition of IL17A using antibodies proven as an effective treatment. Oral anti-IL17 therapies are an attractive alternative option, and several preclinical small molecule IL17 inhibitors have previously been described. Herein, we report the discovery of a novel class of small molecule IL17A inhibitors, identified via a DNA-encoded chemical library screen, and their subsequent optimization to provide in vivo efficacious inhibitors. These new protein-protein interaction (PPI) inhibitors bind in a previously undescribed mode in the IL17A protein with two copies binding symmetrically to the central cavities of the IL17A homodimer.

Visualized Tracking and Multidimensional Assessing of Mitochondria-Associated Pyroptosis in Cancer Cells by a Small-Molecule Fluorescent Probe.

Pyroptosis is closely related to the development and treatment of various cancers; thus, comprehensive studies of the correlations between pyroptosis and its inductive or inhibitive factors can provide new ideas for the intervention and diagnosis of tumors. The dysfunction of mitochondria may induce pyroptosis in cancer cells, which can be reflected by the fluctuations of the microenvironmental parameters in mitochondria as well as the changes of mitochondrial DNA level and morphology, etc. To precisely track and assess the mitochondria-associated pyroptosis process, simultaneous visualization of changes in multiphysiological parameters in mitochondria is highly desirable. In this work, we reported a nonreaction-based, multifunctional small-molecule fluorescent probe Mito-DK with the capability of crosstalk-free response to polarity and mtDNA as well as mitochondrial morphology. Accurate assessment of mitochondria-associated pyroptosis induced by palmitic acid/H2O2 was achieved through monitoring changes in mitochondrial multiple parameters with the help of Mito-DK. In particular, the pyroptosis-inducing ability of an antibiotic doxorubicin and the pyroptosis-inhibiting capacity of an anticancer agent puerarin were evaluated by Mito-DK. These results provide new perspectives for visualizing mitochondria-associated pyroptosis and offer new approaches for screening pyroptosis-related anticancer agents.

AI is a viable alternative to high throughput screening: a 318-target study.

High throughput screening (HTS) is routinely used to identify bioactive small molecules. This requires physical compounds, which limits coverage of accessible chemical space. Computational approaches combined with vast on-demand chemical libraries can access far greater chemical space, provided that the predictive accuracy is sufficient to identify useful molecules. Through the largest and most diverse virtual HTS campaign reported to date, comprising 318 individual projects, we demonstrate that our AtomNet® convolutional neural network successfully finds novel hits across every major therapeutic area and protein class. We address historical limitations of computational screening by demonstrating success for target proteins without known binders, high-quality X-ray crystal structures, or manual cherry-picking of compounds. We show that the molecules selected by the AtomNet® model are novel drug-like scaffolds rather than minor modifications to known bioactive compounds. Our empirical results suggest that computational methods can substantially replace HTS as the first step of small-molecule drug discovery.

Cell Painting-based bioactivity prediction boosts high-throughput screening hit-rates and compound diversity.

Identifying active compounds for a target is a time- and resource-intensive task in early drug discovery. Accurate bioactivity prediction using morphological profiles could streamline the process, enabling smaller, more focused compound screens. We investigate the potential of deep learning on unrefined single-concentration activity readouts and Cell Painting data, to predict compound activity across 140 diverse assays. We observe an average ROC-AUC of 0.744 ± 0.108 with 62% of assays achieving ≥0.7, 30% ≥0.8, and 7% ≥0.9. In many cases, the high prediction performance can be achieved using only brightfield images instead of multichannel fluorescence images. A comprehensive analysis shows that Cell Painting-based bioactivity prediction is robust across assay types, technologies, and target classes, with cell-based assays and kinase targets being particularly well-suited for prediction. Experimental validation confirms the enrichment of active compounds. Our findings indicate that models trained on Cell Painting data, combined with a small set of single-concentration data points, can reliably predict the activity of a compound library across diverse targets and assays while maintaining high hit rates and scaffold diversity. This approach has the potential to reduce the size of screening campaigns, saving time and resources, and enabling primary screening with more complex assays.

DeepP450: Predicting Human P450 Activities of Small Molecules by Integrating Pretrained Protein Language Model and Molecular Representation.

Cytochrome P450 enzymes (CYPs) play a crucial role in Phase I drug metabolism in the human body, and CYP activity toward compounds can significantly affect druggability, making early prediction of CYP activity and substrate identification essential for therapeutic development. Here, we established a deep learning model for assessing potential CYP substrates, DeepP450, by fine-tuning protein and molecule pretrained models through feature integration with cross-attention and self-attention layers. This model exhibited high prediction accuracy (0.92) on the test set, with area under the receiver operating characteristic curve (AUROC) values ranging from 0.89 to 0.98 in substrate/nonsubstrate predictions across the nine major human CYPs, surpassing current benchmarks for CYP activity prediction. Notably, DeepP450 uses only one model to predict substrates/nonsubstrates for any of the nine CYPs and exhibits certain generalizability on novel compounds and different categories of human CYPs, which could greatly facilitate early stage drug design by avoiding CYP-reactive compounds.

Consensus Pharmacophore Strategy For Identifying Novel SARS-Cov-2 Mpro Inhibitors from Large Chemical Libraries.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main Protease (Mpro) is an enzyme that cleaves viral polyproteins translated from the viral genome and is critical for viral replication. Mpro is a target for anti-SARS-CoV-2 drug development, and multiple Mpro crystals complexed with competitive inhibitors have been reported. In this study, we aimed to develop an Mpro consensus pharmacophore as a tool to expand the search for inhibitors. We generated a consensus model by aligning and summarizing pharmacophoric points from 152 bioactive conformers of SARS-CoV-2 Mpro inhibitors. Validation against a library of conformers from a subset of ligands showed that our model retrieved poses that reproduced the crystal-binding mode in 77% of the cases. Using models derived from a consensus pharmacophore, we screened >340 million compounds. Pharmacophore-matching and chemoinformatics analyses identified new potential Mpro inhibitors. The candidate compounds were chemically dissimilar to the reference set, and among them, demonstrating the relevance of our model. We evaluated the effect of 16 candidates on Mpro enzymatic activity finding that seven have inhibitory activity. Three compounds (1, 4, and 5) had IC50 values in the midmicromolar range. The Mpro consensus pharmacophore reported herein can be used to identify compounds with improved activity and novel chemical scaffolds against Mpro. The method developed for its generation is provided as an open-access code (https://github.com/AngelRuizMoreno/ConcensusPharmacophore) and can be applied to other pharmacological targets.

High-throughput screening of small-molecules libraries identified antibacterials against clinically relevant multidrug-resistant A. baumannii and K. pneumoniae.

BACKGROUND: The current pipeline for new antibiotics fails to fully address the significant threat posed by drug-resistant Gram-negative bacteria that have been identified by the World Health Organization (WHO) as a global health priority. New antibacterials acting through novel mechanisms of action are urgently needed. We aimed to identify new chemical entities (NCEs) with activity against Klebsiella pneumoniae and Acinetobacter baumannii that could be developed into a new treatment for drug-resistant infections. METHODS: We developed a high-throughput phenotypic screen and selection cascade for generation of hit compounds active against multidrug-resistant (MDR) strains of K. pneumoniae and A. baumannii. We screened compound libraries selected from the proprietary collections of three pharmaceutical companies that had exited antibacterial drug discovery but continued to accumulate new compounds to their collection. Compounds from two out of three libraries were selected using "eNTRy rules" criteria associated with increased likelihood of intracellular accumulation in Escherichia coli. FINDINGS: We identified 72 compounds with confirmed activity against K. pneumoniae and/or drug-resistant A. baumannii. Two new chemical series with activity against XDR A. baumannii were identified meeting our criteria of potency (EC50 ≤50 µM) and absence of cytotoxicity (HepG2 CC50 ≥100 µM and red blood cell lysis HC50 ≥100 µM). The activity of close analogues of the two chemical series was also determined against A. baumannii clinical isolates. INTERPRETATION: This work provides proof of principle for the screening strategy developed to identify NCEs with antibacterial activity against multidrug-resistant critical priority pathogens such as K. pneumoniae and A. baumannii. The screening and hit selection cascade established here provide an excellent foundation for further screening of new compound libraries to identify high quality starting points for new antibacterial lead generation projects. FUNDING: BMBF and GARDP.

Large-Scale Pretraining Improves Sample Efficiency of Active Learning-Based Virtual Screening.

Virtual screening of large compound libraries to identify potential hit candidates is one of the earliest steps in drug discovery. As the size of commercially available compound collections grows exponentially to the scale of billions, active learning and Bayesian optimization have recently been proven as effective methods of narrowing down the search space. An essential component of those methods is a surrogate machine learning model that predicts the desired properties of compounds. An accurate model can achieve high sample efficiency by finding hits with only a fraction of the entire library being virtually screened. In this study, we examined the performance of a pretrained transformer-based language model and graph neural network in a Bayesian optimization active learning framework. The best pretrained model identifies 58.97% of the top-50,000 compounds after screening only 0.6% of an ultralarge library containing 99.5 million compounds, improving 8% over the previous state-of-the-art baseline. Through extensive benchmarks, we show that the superior performance of pretrained models persists in both structure-based and ligand-based drug discovery. Pretrained models can serve as a boost to the accuracy and sample efficiency of active learning-based virtual screening.

High-Throughput Methods for the Discovery of Small Molecule Modulators of Pancreatic Beta-Cell Function and Regeneration.

The progression of type II diabetes (T2D) is characterized by a complex and highly variable loss of beta-cell mass, resulting in impaired insulin secretion. Many T2D drug discovery efforts aimed at discovering molecules that can protect or restore beta-cell mass and function have been developed using limited beta-cell lines and primary rodent/human pancreatic islets. Various high-throughput screening methods have been used in the context of drug discovery, including luciferase-based reporter assays, glucose-stimulated insulin secretion, and high-content screening. In this context, a cornerstone of small molecule discovery has been the use of immortalized rodent beta-cell lines. Although insightful, this usage has led to a more comprehensive understanding of rodent beta-cell proliferation pathways rather than their human counterparts. Advantages gained in enhanced physiological relevance are offered by three-dimensional (3D) primary islets and pseudoislets in contrast to monolayer cultures, but these approaches have been limited to use in low-throughput experiments. Emerging methods, such as high-throughput 3D islet imaging coupled with machine learning, aim to increase the feasibility of integrating 3D microtissue structures into high-throughput screening. This review explores the current methods used in high-throughput screening for small molecule modulators of beta-cell mass and function, a potentially pivotal strategy for diabetes drug discovery.

Recent applications of artificial intelligence in RNA-targeted small molecule drug discovery.

INTRODUCTION: Targeting RNAs with small molecules offers an alternative to the conventional protein-targeted drug discovery and can potentially address unmet and emerging medical needs. The recent rise of interest in the strategy has already resulted in large amounts of data on disease associated RNAs, as well as on small molecules that bind to such RNAs. Artificial intelligence (AI) approaches, including machine learning and deep learning, present an opportunity to speed up the discovery of RNA-targeted small molecules by improving decision-making efficiency and quality. AREAS COVERED: The topics described in this review include the recent applications of AI in the identification of RNA targets, RNA structure determination, screening of chemical compound libraries, and hit-to-lead optimization. The impact and limitations of the recent AI applications are discussed, along with an outlook on the possible applications of next-generation AI tools for the discovery of novel RNA-targeted small molecule drugs. EXPERT OPINION: Key areas for improvement include developing AI tools for understanding RNA dynamics and RNA - small molecule interactions. High-quality and comprehensive data still need to be generated especially on the biological activity of small molecules that target RNAs.

Using Generative Modeling to Endow with Potency Initially Inert Compounds with Good Bioavailability and Low Toxicity.

In the early stages of drug development, large chemical libraries are typically screened to identify compounds of promising potency against the chosen targets. Often, however, the resulting hit compounds tend to have poor drug metabolism and pharmacokinetics (DMPK), with negative developability features that may be difficult to eliminate. Therefore, starting the drug discovery process with a "null library", compounds that have highly desirable DMPK properties but no potency against the chosen targets, could be advantageous. Here, we explore the opportunities offered by machine learning to realize this strategy in the case of the inhibition of α-synuclein aggregation, a process associated with Parkinson's disease. We apply MolDQN, a generative machine learning method, to build an inhibitory activity against α-synuclein aggregation into an initial inactive compound with good DMPK properties. Our results illustrate how generative modeling can be used to endow initially inert compounds with desirable developability properties.

Development of small-molecular-based radiotracers for PET imaging of PD-L1 expression and guiding the PD-L1 therapeutics.

PURPOSE: Programmed cell death protein ligand 1 (PD-L1) is a crucial biomarker for immunotherapy. However, nearly 70% of patients do not respond to PD-L1 immune checkpoint therapy. Accurate monitoring of PD-L1 expression and quantification of target binding during treatment are essential. In this study, a series of small-molecule radiotracers were developed to assess PD-L1 expression and direct immunotherapy. METHODS: Radiotracers of [68Ga]Ga-D-PMED, [68Ga]Ga-D-PEG-PMED, and [68Ga]Ga-D-pep-PMED were designed based on a 2-methyl-3-biphenyl methanol scaffold and successfully synthesized. Cellular experiments and molecular docking assays were performed to determine their specificity for PD-L1. PD-L1 status was investigated via positron emission tomography (PET) imaging in MC38 tumor models. PET imaging of [68Ga]Ga-D-pep-PMED was performed to noninvasively quantify PD-L1 blocking using an anti-mouse PD-L1 antibody (PD-L1 mAb). RESULTS: The radiosyntheses of [68Ga]Ga-D-PMED, [68Ga]Ga-D-PEG-PMED, and [68Ga]Ga-D-pep-PMED were achieved with radiochemical yields of 87 ± 6%, 82 ± 4%, and 79 ± 9%, respectively. In vitro competition assays demonstrated their high affinities (the IC50 values of [68Ga]Ga-D-PMED, [68Ga]Ga-D-PEG-PMED, and [68Ga]Ga-D-pep-PMED were 90.66 ± 1.24, 160.8 ± 1.35, and 51.6 ± 1.32 nM, respectively). At 120 min postinjection (p.i.) of the radiotracers, MC38 tumors displayed optimized tumor-to-muscle ratios for all radioligands. Owing to its hydrophilic modification, [68Ga]Ga-D-pep-PMED had the highest target-to-nontarget (T/NT) ratio of approximately 6.2 ± 1.2. Interestingly, the tumor/liver ratio was hardly affected by different concentrations of the inhibitor BMS202. We then evaluated the impacts of dose and time on accessible PD-L1 levels in the tumor during anti-mouse PD-L1 antibody treatment. The tumor uptake of [68Ga]Ga-D-pep-PMED significantly decreased with increasing PD-L1 mAb dose. Moreover, after 8 days of treatment with a single antibody, the uptake of [68Ga]Ga-D-pep-PMED in the tumor significantly increased but remained lower than that in the saline group. CONCLUSION: PET imaging with [68Ga]Ga-D-pep-PMED, a small-molecule radiotracer, is a promising tool for evaluating PD-L1 expression and quantifying the target blockade of PD-L1 to assist in the development of effective therapeutic regimens.

Multiplexed experimental strategies for fragment library screening against challenging drug targets using SPR biosensors.

Surface plasmon resonance (SPR) biosensor methods are ideally suited for fragment-based lead discovery.  However, generally applicable experimental procedures and detailed protocols are lacking, especially for structurally or physico-chemically challenging targets or when tool compounds are not available. Success depends on accounting for the features of both the target and the chemical library, purposely designing screening experiments for identification and validation of hits with desired specificity and mode-of-action, and availability of orthogonal methods capable of confirming fragment hits. The range of targets and libraries amenable to an SPR biosensor-based approach for identifying hits is considerably expanded by adopting multiplexed strategies, using multiple complementary surfaces or experimental conditions. Here we illustrate principles and multiplexed approaches for using flow-based SPR biosensor systems for screening fragment libraries of different sizes (90 and 1056 compounds) against a selection of challenging targets. It shows strategies for the identification of fragments interacting with 1) large and structurally dynamic targets, represented by acetyl choline binding protein (AChBP), a Cys-loop receptor ligand gated ion channel homologue, 2) targets in multi protein complexes, represented by lysine demethylase 1 and a corepressor (LSD1/CoREST), 3) structurally variable or unstable targets, represented by farnesyl pyrophosphate synthase (FPPS), 4) targets containing intrinsically disordered regions, represented by protein tyrosine phosphatase 1B  (PTP1B), and 5) aggregation-prone proteins, represented by an engineered form of human tau  (tau K18M). Practical considerations and procedures accounting for the characteristics of the proteins and libraries, and that increase robustness, sensitivity, throughput and versatility are highlighted. The study shows that the challenges for addressing these types of targets is not identification of potentially useful fragments per se, but establishing methods for their validation and evolution into leads.

Covalent fragment libraries in drug discovery-Design, synthesis, and screening methods.

As the development of drugs with a covalent mode of action is becoming increasingly popular, well-validated covalent fragment-based drug discovery (FBDD) methods have been comparatively slow to keep up with the demand. In this chapter the principles of covalent fragment reactivity, library design, synthesis, and screening methods are explored in depth, focussing on literature examples with direct applications to practical covalent fragment library design and screening. Further, questions about the future of the field are explored and potential useful advances are proposed.

Discovery of a First-in-Class Small-Molecule Ligand for WDR91 Using DNA-Encoded Chemical Library Selection Followed by Machine Learning.

WD40 repeat-containing protein 91 (WDR91) regulates early-to-late endosome conversion and plays vital roles in endosome fusion, recycling, and transport. WDR91 was recently identified as a potential host factor for viral infection. We employed DNA-encoded chemical library (DEL) selection against the WDR domain of WDR91, followed by machine learning to predict ligands from the synthetically accessible Enamine REAL database. Screening of predicted compounds identified a WDR91 selective compound 1, with a KD of 6 ± 2 µM by surface plasmon resonance. The co-crystal structure confirmed the binding of 1 to the WDR91 side pocket, in proximity to cysteine 487, which led to the discovery of covalent analogues 18 and 19. The covalent adduct formation for 18 and 19 was confirmed by intact mass liquid chromatography-mass spectrometry. The discovery of 1, 18, and 19, accompanying structure-activity relationship, and the co-crystal structures provide valuable insights for designing potent and selective chemical tools against WDR91 to evaluate its therapeutic potential.

Fused Tetrahydroquinolines Are Interfering with Your Assay.

Tricyclic tetrahydroquinolines (THQs) have been repeatedly reported as hits across a diverse range of high-throughput screening (HTS) campaigns. The activities of these compounds, however, are likely due to reactive byproducts that interfere with the assay. As a lesser studied class of pan-assay interference compounds, the mechanism by which fused THQs react with protein targets remains largely unknown. During HTS follow-up, we characterized the behavior and stability of several fused tricyclic THQs. We synthesized key analogues to pinpoint the cyclopentene ring double bond as a source of reactivity of fused THQs. We found that these compounds degrade in solution under standard laboratory conditions in days. Importantly, these observations make it likely that fused THQs, which are ubiquitously found within small molecule screening libraries, are unlikely the intact parent compounds. We urge deprioritization of tricylic THQ hits in HTS follow-up and caution against the investment of resources to follow-up on these problematic compounds.