Screening Hit Research Articles

VScreenML v2.0: Improved Machine Learning Classification for Reducing False Positives in Structure-Based Virtual Screening.

The enthusiastic adoption of make-on-demand chemical libraries for virtual screening has highlighted the need for methods that deliver improved hit-finding discovery rates. Traditional virtual screening methods are often inaccurate, with most compounds nominated in a virtual screen not engaging the intended target protein to any detectable extent. Emerging machine learning approaches have made significant progress in this regard, including our previously described tool vScreenML. The broad adoption of vScreenML was hindered by its challenging usability and dependencies on certain obsolete or proprietary software packages. Here, we introduce vScreenML 2.0 to address each of these limitations with a streamlined Python implementation. Through careful benchmarks, we show that vScreenML 2.0 outperforms other widely used tools for virtual screening hit discovery.

Discovery and Optimization of First-in-Class Molecular Glue Degraders of the WIZ Transcription Factor for Fetal Hemoglobin Induction to Treat Sickle Cell Disease.

Sickle cell disease (SCD) is a prevalent, life-threatening condition with few treatment options, attributed to a heritable mutation in β-hemoglobin. Therapeutic induction of fetal hemoglobin (HbF) with small molecules has been pursued as a treatment to ameliorate many disease complications but with limited success. Herein, we report the discovery of 10, a novel, potent, and selective molecular glue degrader of the transcription factor WIZ that robustly induces HbF expression as a potential treatment for SCD. 10 was optimized from a phenotypic screening hit utilizing insights from X-ray crystallography and computational modeling to improve potency, selectivity and in vivo exposure. In an hNBSGW mouse xenograft model, 10 demonstrated robust WIZ degradation and HbF induction. These results highlight the potential of WIZ degraders as a promising therapy for sickle cell disease.

Discovery and Preclinical Characterization of Fulacimstat (BAY 1142524), a Potent and Selective Chymase Inhibitor As a New Profibrinolytic Approach for Safe Thrombus Resolution.

Chymase is a serine-protease produced by mast cells. In the past few decades, its role in fibrotic diseases triggered the search for orally available chymase inhibitors. Aiming at reducing adverse cardiac remodeling after myocardial infarction, our research efforts resulted in the discovery of fulacimstat (BAY 1142524). While clinical trials did not demonstrate efficacy in this indication, the recent discovery of a new unexpected biological role of chymase spurred a revival of interest in chymase inhibition: chymase was shown to inactivate plasmin within fibrin-rich clots. Chymase inhibitors are now considered as potential profibrinolytic drugs with low bleeding risk and therefore exceptional safety for the treatment of acute thrombosis settings such as stroke, pulmonary embolism, or venous thrombosis. This article describes the chemical optimization journey from a screening hit to the discovery of fulacimstat (BAY 1142524), a selective chymase inhibitor with a good safety profile, as well as its preclinical in vitro and in vivo characterization.

CCRG-01. MYT1 BLOCKADE AS A NOVEL GLIOMA THERAPEUTICS THROUGH ABROGATIING OF THE G2/M CELL CYCLE CHECKPOINT

Abstract BACKGROUND The cell cycle is tightly regulated by checkpoints, which play a vital role in controlling its progression and timing. Cancer cells exploit the G2/M checkpoint, which serves as a resistance mechanism against genotoxic anticancer treatments, allowing for DNA repair prior to cell division. Manipulating cell cycle timing has emerged as a potential strategy to augment the effectiveness of DNA damage-based therapies. METHODS In this study, we conducted a forward genome-wide CRISPR/Cas9 screening with repeated exposure to the alkylating agent temozolomide (TMZ) to investigate the mechanisms underlying tumor cell survival under genotoxic stress. The screening hits were validated through in patient derived glioma cells and xenografts. RESULTS Our findings revealed that canonical DNA repair pathways, including the Ataxia–telangiectasia mutated (ATM)/Fanconi and mismatch repair, determine cell fate under genotoxic stress. Notably, we identified the critical role of PKMYT1, in ensuring cell survival. Depletion of PKMYT1 led to overwhelming TMZ-induced cytotoxicity in cancer cells. Isobologram analysis demonstrated potent drug synergy between alkylating agents and a Myt1 kinase inhibitor, RP-6306. Mechanistically, inhibiting Myt1 forced G2/M-arrested cells into an unscheduled transition to the mitotic phase without complete resolution of DNA damage. This forced entry into mitosis, along with persistent DNA damage, resulted in severe mitotic abnormalities. Ultimately, these aberrations led to mitotic exit with substantial apoptosis. Preclinical animal studies demonstrated that the combination regimen involving TMZ and RP-6306 prolonged the overall survival of glioma-bearing mice.

Mannose receptor (MRC1) mediates uptake of dextran by bone marrow-derived macrophages.

Macrophages survey their environment using receptor-mediated endocytosis and pinocytosis. Receptor-mediated endocytosis allows internalization of specific ligands, whereas pinocytosis non-selectively internalizes extracellular fluids and solutes. CRISPR/Cas9 whole-genome screens were used to identify genes regulating constitutive and growth factor-stimulated dextran uptake in murine bone-marrow derived macrophages (BMDM). The mannose receptor c-type 1 (MRC1/CD206) was a top hit in the screen. Targeted gene disruptions of Mrc1 reduced dextran uptake but had little effect on fluid-phase uptake of Lucifer yellow. Other screen hits also differentially affected the uptake of dextran and Lucifer yellow, indicating internalization by separate mechanisms. Visualization of dextran and Lucifer yellow uptake by microscopy showed enrichment of dextran in small puncta, which was inhibitable by mannan, a ligand of MRC1. In contrast, Lucifer yellow predominantly was internalized in larger macropinosomes. In addition, IL4-treated BMDMs internalized more dextran than untreated BMDM correlating with increased MRC1 expression. Therefore, dextran is not an effective marker for pinocytosis in BMDMs since it is internalized by receptor-mediated process. Numerous genes that regulate dextran internalization in primary murine macrophages were identified in the whole-genome screens, which can inform understanding of the regulation of MRC1 expression and MRC1-mediated uptake in macrophages.

Discovery of BAY 2413555, First Selective Positive Allosteric Modulator of the M2 Receptor to Restore Cardiac Autonomic Balance.

Autonomic disbalance, i.e., sympathetic overactivation and parasympathetic withdrawal, is a causal driver of disease progression in heart failure. While sympatholytic drugs are established treatments, no drug therapy restoring vagal control of cardiac function is available. We report here the HTS-based discovery of a novel class of 1,8-naphthyridin-4(1H)-one carboxamides acting as positive allosteric modulators (PAMs) of the M2 muscarinic acetylcholine receptor (M2R). M2R is the main postsynaptic myocyte receptor regulating heart rate, electrical conduction, and contractile strength. Extensive optimization of the screening hit in terms of potency, permeation, metabolic stability, and solubility ultimately resulted in the discovery of the first-in-class clinical candidate BAY 2413555 (27). With an overall technical profile compatible with once-daily oral administration in a phase 1 study, no apparent effects on blood pressure, and a mechanism that largely preserves autonomic regulatory capacity, BAY 2413555 could be the tool to finally study the restoration of autonomic balance.

Rational Design of Macrocyclic Noncovalent Inhibitors of SARS-CoV-2 Mpro from a DNA-Encoded Chemical Library Screening Hit That Demonstrate Potent Inhibition against Pan-Coronavirus Homologues and Nirmatrelvir-Resistant Variants.

The recent global COVID-19 pandemic has highlighted treatments for coronavirus infection as an unmet medical need. The main protease (Mpro) has been an important target for the development of SARS-CoV-2 direct-acting antivirals. Nirmatrelvir as a covalent Mpro inhibitor was the first such approved therapy. Although Mpro inhibitors of various chemical classes have been reported, they are generally less active against nirmatrelvir-resistant variants and have limited pan-coronavirus potential, presenting a significant human health risk upon future outbreaks. We here present a novel approach and utilized DNA-encoded chemical library screening to identify the noncovalent Mpro inhibitor 5, which demonstrated a distinct binding mode to nirmatrelvir. A macrocyclization strategy designed to lock the active conformation resulted in lactone 12 with significantly improved antiviral activity. Further optimization led to the potent lactam 26, which demonstrated exceptional potency against nirmatrelvir-resistant variants as well as against a panel of viral main proteases from other coronaviruses.

Structure-Guided Discovery of Selective USP7 Inhibitors with In Vivo Activity.

Inhibition of ubiquitin-specific protease 7, USP7, has been proposed as a mechanism to affect many disease processes, primarily those implicated in oncology. The bound crystal structure of a published high-throughput screening hit with low-micromolar affinity for USP7 identified three regions of the compound for structure-guided optimization. Replacing one side of the compound with different aromatic moieties gave little improvement in affinity, and the central piperidine could not be improved. However, the binding site for the other side of the compound was poorly defined in the crystal structure, which suggested a wide variety of synthetically accessible options for optimization. These were assessed by screening reaction mixtures that introduced different substituents to this other side. Subsequent optimization led to a compound with low-nanomolar affinity for USP7, which showed target engagement in tumors, was tolerated in mice, and showed efficacy in xenograft models.

Discovery and evaluation of novel SHIP-1 inhibitors

Src Homology 2-containing Inositol 5′-Phosphatase-1 (SHIP-1), encoded by INPP5D, has been identified as an Alzheimer’s disease (AD) risk-associated gene through recent genetic and epigenetic studies. SHIP-1 confers AD risk by inhibiting the TREM2 cascade and reducing beneficial microglial cellular processes, including phagocytosis. While several small molecules have been reported to modulate SHIP-1 activity, their limited selectivity and efficacy in advanced models restricted their potential as therapeutic agents or probes for biological studies. Herein, we validated and implemented a high-throughput screening platform to explore new chemotypes that can modulate the phosphatase activity of SHIP-1. We screened 49,260 central nervous system (CNS)-penetrate compounds sourced from commercial vendors using the malachite green-based assay for anti-SHIP-1 activity. Through analysis, prioritization, and validation of the screening hits, we identified three novel types of scaffolds that inhibit the SHIP-1 phosphatase activity with IC50s as low as 46.6 µM. To improve the inhibitory activity of these promising hits, we carried out structure–activity relationship (SAR) studies, resulting in a lead molecule SP3-12 that inhibits SHIP-1 with an IC50 value of 6.1 μM. Kinetic analyses of SP3-12 revealed that its inhibition mechanism is competitive, with a Ki value of 3.2 µM for SHIP-1 and a 7-fold selectivity over SHIP-2. Furthermore, results from testing in a microglial phagocytosis/cell health high content assay indicated that SP3-12 could effectively activate phagocytosis in human microglial clone 3 (HMC3) cells, with an EC50 of 2.0 µM, without cytotoxicity in the dose range. Given its potency, selectivity, and cellular activity, SP3-12 emerges as a promising small molecule inhibitor with potential for investigating the biological functions of SHIP-1.

An Open-Source Implementation of the Scaffold Identification and Naming System (SCINS) and Example Applications.

Organizing and partitioning sets of chemical structures is of considerable practical significance, e.g., in compound library analysis and the postprocessing of screening hit lists. Approaches such as unsupervised clustering are computationally demanding and dataset-dependent; on the other hand, rule-based methods, such as those based on Murcko scaffolds, have linear time complexity but are often too fine-grained, leading to a large number of singletons or sparsely populated classes. An alternative rule-based method that seeks to achieve an optimal balance when grouping compounds into sets is the 'Scaffold Identification and Naming System' (SCINS). To facilitate public use of this previously published method, here, we provide an open-source Python implementation of SCINS, dependent only on RDKit. We show that SCINS can be useful in identifying sparsely and densely populated regions in chemical space in large databases, here exemplified with Enamine REAL Diverse and ChEMBL. We find that Enamine REAL Diverse covers a much smaller SCINS space relative to ChEMBL, whereas the opposite is true when Murcko and generic Murcko scaffolds are considered. Additionally, we show that SCINS can result in chemically intuitive grouping of medium-sized sets of bioactive compounds, which can be useful in compound selection from virtual screening campaigns as well as postprocessing of experimental hit lists. Hence, in this work, we provide both an open-source implementation of SCINS and its characterization with relevant use cases.

From Virtual Screens to Cellular Target Engagement: New Small Molecule Ligands for the Immune Checkpoint LAG-3.

Herein, we performed a virtual screening study to discover new scaffolds for small molecule-based ligands of the immune checkpoint lymphocyte-activation gene 3 (LAG-3). Molecular dynamics (MD) simulations using the LAG-3 structure revealed two putative binding sites for small molecules: the antibody interface and the lipophilic canyon. A 3D pharmacophore screening resulted in the identification of potential ligands for these binding sites and afforded a library of 25 compounds. We then evaluated the screening hits for LAG-3 binding via microscale thermophoresis (MST) and surface plasmon resonance (SPR). Our biophysical screening identified two binders with K D values in the low micromolar range, compounds 3 (antibody interface) and 25 (lipophilic canyon). Furthermore, we investigated the ability of LAG-3 hits to engage LAG-3 on a cellular level using a cellular thermal shift assay (CETSA). In summary, compound 3 shows potential as a lead but is not yet a development candidate.

Aryl phosphate ester-induced pericardial edema in zebrafish embryos is influenced by the ionic composition of exposure media

Pericardial edema – fluid accumulation within the pericardium – is a frequently observed malformation in zebrafish embryo-based chemical toxicity screens. We recently discovered that the severity of triphenyl phosphate (TPHP)-induced pericardial edema was dependent on the ionic strength of exposure media. TPHP is an aryl phosphate ester (APE) widely used as a plasticizer and flame retardant. APEs are characterized by having one or more aryl groups bound to a phosphate center, with TPHP containing only unsubstituted aryl groups. Therefore, the objective of this study was to begin investigating whether, similar to TPHP, pericardial edema induced by other structurally related APEs is dependent on the ionic composition of exposure media. We first mined the peer-reviewed literature to identify other APEs that 1) induced pericardial edema in zebrafish embryos within a minimum of three peer-reviewed publications, and 2) demonstrated a statistically significant induction of pericardial edema in at least 70 % of the studies evaluated. Based on this meta-analysis, we identified four other APEs that caused pericardial edema in zebrafish embryos: isopropylated triphenyl phosphate (IPTPP), cresyl diphenyl phosphate (CDP), tricresyl phosphate (TMPP), and 2-ethylhexyl diphenyl phosphate (EDHPHP). Using TPHP as a positive control and pericardial edema as a readout, we developed concentration-response curves for all four APEs based on static exposure from 24 to 72 h post-fertilization (hpf). We then conducted co-exposures with D-Mannitol (an osmotic diuretic) and exposures within reverse osmosis (RO) water determine whether the ionic composition of exposure media mitigated APE-induced pericardial edema at 72 hpf. Using pericardial edema as an endpoint, the approximate EC50s for TPHP (positive control), IPTPP, CDP, TMPP, and EDHPHP were 6.25, 3.125, 3.125, 25, and 100 µM, respectively, based on exposure from 24 to 72 hpf. Interestingly, similar to our findings with TPHP, co-exposure with D-Mannitol and exposure within ion-deficient water significantly mitigated IPTPP- CDP-, TMPP-, and EDHPHP-induced pericardial edema in zebrafish embryos, suggesting that chemically-induced pericardial edema may be 1) dependent on the ionic composition of exposure media and 2) driven by a disruption in osmoregulation across the embryonic epidermis. Therefore, similar to other assay parameters, our findings underscore the need to standardize the osmolarity of exposure media in order to minimize the potential for false positive/negative hits in zebrafish embryo-based chemical toxicity screens conducted around the world.

Wee1 inhibitor optimization through deep-learning-driven decision making

Deep learning has gained increasing attention in recent years, yielding promising results in hit screening and molecular optimization. Herein, we employed an efficient strategy based on multiple deep learning techniques to optimize Wee1 inhibitors, which involves activity interpretation, scaffold-based molecular generation, and activity prediction. Starting from our in-house Wee1 inhibitor GLX0198 (IC50 = 157.9 nM), we obtained three optimized compounds (IC50 = 13.5 nM, 33.7 nM, and 47.1 nM) out of five picked molecules. Further minor modifications on these compounds led to the identification of potent Wee1 inhibitors with desirable inhibitory effects on multiple cancer cell lines. Notably, the best compound 13 exhibited superior cancer cell inhibition, with IC50 values below 100 nM in all tested cancer cells. These results suggest that deep learning can greatly facilitate decision-making at the stage of molecular optimization.

A Multimodal, In Vivo Approach for Assessing Structurally and Phenotypically Related Neuroactive Molecules.

A recently reported behavioral screen in larval zebrafish for phenocopiers of known anesthetics and associated drugs yielded an isoflavone. Related isoflavones have also been reported as GABAA potentiators. From this, we synthesized a small library of isoflavones and incorporated an in vivo phenotypic approach to perform structure-behavior relationship studies of the screening hit and related analogs via behavioral profiling, patch-clamp experiments, and whole brain imaging. This revealed that analogs effect a range of behavioral responses, including sedation with and without enhancing the acoustic startle response. Interestingly, a subset of compounds effect sedation and enhancement of motor responses to both acoustic and light stimuli. Patch clamp recordings of cells with a human GABAA receptor confirmed that behavior-modulating isoflavones modify the GABA signaling. To better understand these molecules' nuanced effects on behavior, we performed whole brain imaging to reveal that analogs differentially effect neuronal activity. These studies demonstrate a multimodal approach to assessing activities of neuroactives.

Classification and Communication of Critical Findings in Emergency Radiology: A Scoping Review

PurposeTo identify the published standards for the classification and communication of critical actionable findings in emergency radiology and the associated facilitators and barriers to communication and message management or dissemination of such findings. Materials and methodsSearch terms for resources pertaining to critical findings (CFs) in emergency radiology were applied to two databases (PubMed, Embase). Screening of hits using the following pre-established inclusion and exclusion criteria were performed by three analysts with subsequent consensus discussion for discrepancies: (1) the resources include any standards for the classification and communication of imaging findings as critical, or (2) the resource discusses any facilitators to the communication of CFs, or (3) the resource discusses any barriers to the communication of CFs. Resources with explicit focus on a pediatric population or predominant focus on artificial intelligence or natural language processing were omitted. Accompanying gray literature search was used to expand included resources. Data extraction included year, country, resource type, scope or purpose, participants, context, standards to identifying or communicating CFs, facilitators and barriers, method type, recommendations, applicability, and disclosures. ResultsSeventy-six resources were included in the final analysis, including 16 societal or commission guidelines. Among the guidelines, no standardized list of CFs was identified, with typical recommendations suggesting application of a local policy. Communication standards included direct closed-loop communication for high acuity findings, with more flexible communication channels for less acute findings. Applied interventions for CFs management most frequently fell into four categories: electronic (n = 10), hybrid (ie, electronic or administrative) (n = 3), feedback or education (n = 5), and administrative (n = 4). ConclusionThere are published standards, policies and interventions for the management of CFs in emergency radiology. Three-tier stratification (eg, critical, urgent, incidental) based on time sensitivity and severity is most common with most CFs necessitating closed-loop communication. Awareness of systemic facilitators and barriers should inform local policy development. Electronic and administrative communication pathways are useful adjuncts. Further research should offer comparative analyses of different CF interventions with regards to cost-effectiveness, notification time, and user feedback.

Compounds Designs of CK2α Inhibitors Derived from Virtual Screening Hit Compounds by Computational Chemistry with Crystallography.

Protein kinase CK2 type α (CK2α) inhibitors are expected to be a new anticancer drug and a treatment for nephritis. Virtual screening for CK2α inhibitors has been conducted and active compounds with various scaffolds have been obtained. Research on compound optimization is currently in progress for some of them with the aim of improving their activity. This process involves the combination of various computational chemistry methods and crystal analyses. In this review, case studies of structure-based compound designs that have efficiently improved the activity of screening hit compounds, including compounds with a thiadiazole ring and a purine scaffold, are introduced.

LP-03 From neurotoxicity DNT hit screening to hit characterization

LP-03 From neurotoxicity DNT hit screening to hit characterization

SARS-CoV-2 Mpro inhibitor identification using a cellular gain-of-signal assay for high-throughput screening

Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2, SARS2) is responsible for the COVID-19 pandemic and infections that continue to affect the lives of millions of people worldwide, especially those who are older and/or immunocompromised. The SARS2 main protease enzyme, Mpro (also called 3C-like protease, 3CLpro), is a bona fide drug target as evidenced by potent inhibition with nirmatrelvir and ensitrelvir, the active components of the drugs Paxlovid and Xocova, respectively. However, the existence of nirmatrelvir and ensitrelvir-resistant isolates underscores the need to develop next-generation drugs with different resistance profiles and/or distinct mechanisms of action. Here, we report the results of a high-throughput screen of 649,568 compounds using a cellular gain-of-signal assay. In this assay, Mpro inhibits expression of a luciferase reporter, and 8,777 small molecules were considered hits by causing a gain in luciferase activity 3x SD above the sample field activity (6.8% gain-of-signal relative to 100 µM GC376). Single concentration and dose-response gain-of-signal experiments confirmed 3,522/8,762 compounds as candidate inhibitors. In parallel, all initial high-throughput screening hits were tested in a peptide cleavage assay with purified Mpro and only 39/8,762 showed inhibition. Importantly, 19/39 compounds (49%) re-tested positive in both SARS2 assays, including two previously reported Mpro inhibitors, demonstrating the efficacy of the overall screening strategy. This approach led to the rediscovery of known Mpro inhibitors such as calpain inhibitor II, as well as to the discovery of novel compounds that provide chemical information for future drug development efforts.

Exploring novel A2AAR antagonists: Design, synthesis, and evaluation of 2,6,9-trisubstituted purine derivatives as promising antifibrotic agents

A series of 2,6,9-trisubstituted purine derivatives were designed and synthesized with diverse chemical moieties. Through a comprehensive biological evaluation, we identified 4-(6-(methylamino)-2-(phenylethynyl)-9H-purin-9-yl)phenol (6a) as a promising A2AAR antagonist with potent antifibrotic properties. Compound 6a demonstrated significant efficacy in inhibiting CRE promoter activity and in reducing the expression of fibrogenic marker proteins and downstream effectors of A2AAR activation, surpassing the A2AAR antagonist ZM241385 and initial screening hits, 9-benzyl-N-methyl-2-(phenylethynyl)-9H-purin-6-amine (5a) and 9-((benzyloxy)methyl)-N-methyl-2-(phenylethynyl)-9H-purin-6-amine (5j). Further validation revealed that compound 6a effectively inhibited fibrogenic marker proteins induced by A2AAR overexpression or TGF-β1 treatment in hepatic stellate cells, alongside reducing PKA and CREB phosphorylation. These findings suggest that compound 6a exerts its antifibrotic action by modulating the cAMP/PKA/CREB pathway through A2AAR inhibition. Overall, our study provides valuable insights for the development of novel therapeutics that target hepatic fibrosis through A2AAR antagonism.

Machine Learning Guided AQFEP: A Fast and Efficient Absolute Free Energy Perturbation Solution for Virtual Screening.

Structure-based methods in drug discovery have become an integral part of the modern drug discovery process. The power of virtual screening lies in its ability to rapidly and cost-effectively explore enormous chemical spaces to select promising ligands for further experimental investigation. Relative free energy perturbation (RFEP) and similar methods are the gold standard for binding affinity prediction in drug discovery hit-to-lead and lead optimization phases, but have high computational cost and the requirement of a structural analog with a known activity. Without a reference molecule requirement, absolute FEP (AFEP) has, in theory, better accuracy for hit ID, but in practice, the slow throughput is not compatible with VS, where fast docking and unreliable scoring functions are still the standard. Here, we present an integrated workflow to virtually screen large and diverse chemical libraries efficiently, combining active learning with a physics-based scoring function based on a fast absolute free energy perturbation method. We validated the performance of the approach in the ranking of structurally related ligands, virtual screening hit rate enrichment, and active learning chemical space exploration; disclosing the largest reported collection of free energy simulations to date.