In silico fragment-based discovery of CIB1-directed anti-tumor agents by FRASE-bot.

Chemical probes are an indispensable tool for translating biological discoveries into new therapies, though are increasingly difficult to identify since novel therapeutic targets are often hard-to-drug proteins. We introduce FRASE-based hit-finding robot (FRASE-bot), to expedite drug discovery for unconventional therapeutic targets. FRASE-bot mines available 3D structures of ligand-protein complexes to create a database of FRAgments in Structural Environments (FRASE). The FRASE database can be screened to identify structural environments similar to those in the target protein and seed the target structure with relevant ligand fragments. A neural network model is used to retain fragments with the highest likelihood of being native binders. The seeded fragments then inform ultra-large-scale virtual screening of commercially available compounds. We apply FRASE-bot to identify ligands for Calcium and Integrin Binding protein 1 (CIB1), a promising drug target implicated in triple negative breast cancer. FRASE-based virtual screening identifies a small-molecule CIB1 ligand (with binding confirmed in a TR-FRET assay) showing specific cell-killing activity in CIB1-dependent cancer cells, but not in CIB1-depletion-insensitive cells.

Discovery of Novel Macrocyclic MERTK/AXL Dual Inhibitors.

MERTK and AXL are members of the TAM (TYRO3, AXL, MERTK) family of receptor tyrosine kinases that are aberrantly expressed and have been implicated as therapeutic targets in a wide variety of human tumors. Dual MERTK and AXL inhibition could provide antitumor action mediated by both direct tumor cell killing and modulation of the innate immune response in some tumors such as nonsmall cell lung cancer. We utilized our knowledge of MERTK inhibitors and a structure-based drug design approach to discover a novel class of macrocyclic dual MERTK/AXL inhibitors. The lead compound 43 had low-nanomolar activity against both MERTK and AXL and good selectivity over TYRO3 and FLT3. Its target engagement and selectivity were also confirmed by NanoBRET and cell-based MERTK and AXL phosphorylation assays. Compound 43 had excellent pharmacokinetic properties (large AUC and long half-life) and mediated antitumor activity against lung cancer cell lines, indicating its potential as a therapeutic agent.

Machine Learning-driven Fragment-based Discovery of CIB1-directed Anti-Tumor Agents by FRASE-bot.

Chemical probes are an indispensable tool for translating biological discoveries into new therapies, though are increasingly difficult to identify. Novel therapeutic targets are often hard-to-drug proteins, such as messengers or transcription factors. Computational strategies arise as a promising solution to expedite drug discovery for unconventional therapeutic targets. FRASE-bot exploits big data and machine learning (ML) to distill 3D information relevant to the target protein from thousands of protein-ligand complexes to seed it with ligand fragments. The seeded fragments can then inform either (i) de novo design of 3D ligand structures or (ii) ultra-large-scale virtual screening of commercially available compounds. Here, FRASE-bot was applied to identify ligands for Calcium and Integrin Binding protein 1 (CIB1), a promising but ligand-orphan drug target implicated in triple negative breast cancer. The signaling function of CIB1 relies on protein-protein interactions and its structure does not feature any natural ligand-binding pocket. FRASE-based virtual screening identified the first small-molecule CIB1 ligand (with binding confirmed in a TR-FRET assay) showing specific cell-killing activity in CIB1-dependent cancer cells, but not in CIB1-depleted cells.

Relationship between lysine methyltransferase levels and heterochromatin gene repression in living cells and in silico.

Gene regulation plays essential roles in all multicellular organisms, allowing for different specialized tissue types to be generated from a complex genome. Heterochromatin-driven gene repression, associated with a physical compaction of the genome, is a pathway involving core components that are conserved from yeast to human. Posttranslational modification of chromatin is a critical component of gene regulation. Specifically, tri-methylation of the nucleosome component histone 3 at lysine 9 (H3K9me3) is a key feature of this pathway along with the hallmark heterochromatin protein 1 (HP1). Histone methyltransferases are recruited by HP1 to deposit H3K9me3 marks which nucleate and recruit more HP1 in a process that spreads from the targeting site to signal for gene repression. One of the enzymes recruited is SETDB1, a methyltransferase which putatively catalyzes posttranslational methylation marks on H3K9. To better understand the contribution of SETDB1 in heterochromatin formation, we downregulated SETDB1 through knockdown by a dCas9-KRAB system and examined heterochromatin formation in a chromatin in vivo assay (CiA-Oct4). We studied the contribution of SETDB1 to heterochromatin formation kinetics in a developmentally crucial locus, Oct4. Our data demonstrate that SETDB1 reduction led to a delay in both gene silencing and in H3K9me3 accumulation. Importantly, SETDB1 knockdown to a ∼50% level did not stop heterochromatin formation completely. Particle-based Monte Carlo simulations in 3D space with explicit representation of key molecular processes enabled the elucidation of how SETDB1 downregulation affects the individual molecular processes underlying heterochromatin formation.

PROTAC Linkerology Leads to an Optimized Bivalent Chemical Degrader of Polycomb Repressive Complex 2 (PRC2) Components.

Bivalent chemical degraders, otherwise known as proteolysis-targeting chimeras (PROTACs), have proven to be an efficient strategy for targeting overexpressed or mutated proteins in cancer. PROTACs provide an alternative approach to small-molecule inhibitors, which are restricted by occupancy-driven pharmacology, often resulting in acquired inhibitor resistance via compensatory increases in protein expression. Despite the advantages of bivalent chemical degraders, they often have suboptimal physicochemical properties and optimization for efficient degradation remains highly unpredictable. Herein, we report the development of a potent EED-targeted PRC2 degrader, UNC7700. UNC7700 contains a unique cis-cyclobutane linker and potently degrades PRC2 components EED (DC50 = 111 nM; Dmax = 84%), EZH2WT/EZH2Y641N (DC50 = 275 nM; Dmax = 86%), and to a lesser extent SUZ12 (Dmax = 44%) after 24 h in a diffuse large B-cell lymphoma DB cell line. Characterization of UNC7700 and related compounds for ternary complex formation and cellular permeability to provide a rationale for the observed improvement in degradation efficiency remained challenging. Importantly, UNC7700 dramatically reduces H3K27me3 levels and is anti-proliferative in DB cells (EC50 = 0.79 ± 0.53 µM).

Structural basis of paralog-specific KDM2A/B nucleosome recognition.

The nucleosome acidic patch is a major interaction hub for chromatin, providing a platform for enzymes to dock and orient for nucleosome-targeted activities. To define the molecular basis of acidic patch recognition proteome wide, we performed an amino acid resolution acidic patch interactome screen. We discovered that the histone H3 lysine 36 (H3K36) demethylase KDM2A, but not its closely related paralog, KDM2B, requires the acidic patch for nucleosome binding. Despite fundamental roles in transcriptional repression in health and disease, the molecular mechanisms governing nucleosome substrate specificity of KDM2A/B, or any related JumonjiC (JmjC) domain lysine demethylase, remain unclear. We used a covalent conjugate between H3K36 and a demethylase inhibitor to solve cryogenic electron microscopy structures of KDM2A and KDM2B trapped in action on a nucleosome substrate. Our structures show that KDM2-nucleosome binding is paralog specific and facilitated by dynamic nucleosomal DNA unwrapping and histone charge shielding that mobilize the H3K36 sequence for demethylation.

Enzymatic Macrolactamization of mRNA Display Libraries for Inhibitor Selection.

mRNA display is a powerful, high-throughput technology for discovering novel, peptide ligands for protein targets. A number of methods have been used to expand the chemical diversity of mRNA display libraries beyond the 20 canonical amino acids, including genetic code reprogramming and biorthogonal chemistries. To date, however, there have been few reports using enzymes as biocompatible reagents for diversifying mRNA display libraries. Here, we report the evaluation and implementation of the common industrial enzyme, microbial transglutaminase (mTG), as a versatile biocatalyst for cyclization of mRNA display peptide libraries via lysine-to-glutamine isopeptide bonds. We establish two separate display-based assays to validate the compatibility of mTG with mRNA-linked peptide substrates. These assays indicate that mTG has a high degree of substrate tolerance and low single round bias. To demonstrate the potential benefits of mTG-mediated cyclization in ligand discovery, high diversity mTG-modified libraries were employed in two separate affinity selections: (1) one against the calcium and integrin binding protein, CIB1, and (2) the second against the immune checkpoint protein and emerging therapeutic target, B7-H3. Both selections resulted in the identification of potent, cyclic, low nanomolar binders, and subsequent structure-activity studies demonstrate the importance of the cyclization to the observed activity. Notably, cyclization in the CIB1 binder stabilizes an α-helical conformation, while the B7-H3 inhibitor employs two bridges, one mTG-derived lactam and a second disulfide to achieve its potency. Together, these results demonstrate potential benefits of enzyme-based biocatalysts in mRNA display ligand selections and establish a framework for employing mTG in mRNA display.

Combining pharmacophore models derived from DNA-encoded chemical libraries with structure-based exploration to predict Tankyrase 1 inhibitors.

DNA-encoded chemical libraries (DECLs) interrogate the interactions of a target of interest with vast numbers of molecules. DECLs hence provide abundant information about the chemical ligand space for therapeutic targets, and there is considerable interest in methods for exploiting DECL screening data to predict novel ligands. Here we introduce one such approach and demonstrate its feasibility using the cancer-related poly-(ADP-ribose)transferase tankyrase 1 (TNKS1) as a model target. First, DECL affinity selections resulted in structurally diverse TNKS1 inhibitors with high potency including compound 2 with an IC50 value of 0.8 nM. Additionally, TNKS1 hits from four DECLs were translated into pharmacophore models, which were exploited in combination with docking-based screening to identify TNKS1 ligand candidates in databases of commercially available compounds. This computational strategy afforded TNKS1 inhibitors that are outside the chemical space covered by the DECLs and yielded the drug-like lead compound 12 with an IC50 value of 22 nM. The study further provided insights in the reliability of screening data and the effect of library design on hit compounds. In particular, the study revealed that while in general DECL screening data are in good agreement with off-DNA ligand binding, unpredictable interactions of the DNA-attachment linker with the target protein contribute to the noise in the affinity selection data.

Development of a high-throughput TR-FRET screening assay for a fast-cycling KRAS mutant.

Mutations in the small GTPase protein KRAS are one of the leading drivers of cancers including lung, pancreatic, and colorectal, as well as a group of developmental disorders termed "Rasopathies". Recent breakthroughs in the development of mutant-specific KRAS inhibitors include the FDA approved drug Lumakras (Sotorasib, AMG510) for KRAS G12C-mutated non-small cell lung cancer (NSCLC), and MRTX1133, a promising clinical candidate for the treatment of KRAS G12D-mutated cancers. However, there are currently no FDA approved inhibitors that target KRAS mutations occurring at non-codon 12 positions. Herein, we focused on the KRAS mutant A146T, found in colorectal cancers, that exhibits a "fast-cycling" nucleotide mechanism as a driver for oncogenic activation. We developed a novel high throughput time-resolved fluorescence resonance energy transfer (TR-FRET) assay that leverages the reduced nucleotide affinity of KRAS A146T. As designed, the assay is capable of detecting small molecules that act to allosterically modulate GDP affinity or directly compete with the bound nucleotide. A pilot screen was completed to demonstrate robust statistics and reproducibility followed by a primary screen using a diversity library totaling over 83,000 compounds. Compounds yielding >50% inhibition of TR-FRET signal were selected as hits for testing in dose-response format. The most promising hit, UNC10104889, was further investigated through a structure activity relationship (SAR)-by-catalog approach in an attempt to improve potency and circumvent solubility liabilities. Overall, we present the TR-FRET platform as a robust assay to screen fast-cycling KRAS mutants enabling future discovery efforts for novel chemical probes and drug candidates.

A simulation model of heterochromatin formation at submolecular detail.

Heterochromatin is a physical state of the chromatin fiber that maintains gene repression during cell development. Although evidence exists on molecular mechanisms involved in heterochromatin formation, a detailed structural mechanism of heterochromatin formation needs a better understanding. We made use of a simple Monte Carlo simulation model with explicit representation of key molecular events to observe molecular self-organization leading to heterochromatin formation. Our simulations provide a structural interpretation of several important traits of the heterochromatinization process. In particular, this study provides a depiction of how small amounts of HP1 are able to induce a highly condensed chromatin state through HP1 dimerization and bridging of sequence-remote nucleosomes. It also elucidates structural roots of a yet poorly understood phenomenon of a nondeterministic nature of heterochromatin formation and subsequent gene repression. Experimental chromatin in vivo assay provides an unbiased estimate of time scale of repressive response to a heterochromatin-triggering event.

MERTK activation drives osimertinib resistance in EGFR-mutant non-small cell lung cancer.

Acquired resistance is inevitable in non-small cell lung cancers (NSCLCs) treated with osimertinib (OSI), and the mechanisms are not well defined. The MERTK ligand GAS6 promoted downstream oncogenic signaling in EGFR-mutated (EGFRMT) NSCLC cells treated with OSI, suggesting a role for MERTK activation in OSI resistance. Indeed, treatment with MRX-2843, a first-in-class MERTK kinase inhibitor, resensitized GAS6-treated NSCLC cells to OSI. Both GAS6 and EGF stimulated downstream PI3K/AKT and MAPK/ERK signaling in parental cells, but only GAS6 activated these pathways in OSI-resistant (OSIR) derivative cell lines. Functionally, OSIR cells were more sensitive to MRX-2843 than parental cells, suggesting acquired dependence on MERTK signaling. Furthermore, MERTK and/or its ligands were dramatically upregulated in EGFRMT tumors after treatment with OSI in both xenograft models and patient samples, consistent with induction of autocrine/paracrine MERTK activation. Moreover, treatment with MRX-2843 in combination with OSI, but not OSI alone, provided durable suppression of tumor growth in vivo, even after treatment was stopped. These data identify MERTK as a driver of bypass signaling in treatment-naive and EGFRMT-OSIR NSCLC cells and predict that MRX-2843 and OSI combination therapy will provide clinical benefit in patients with EGFRMT NSCLC.

Use of AD Informer Set compounds to explore validity of novel targets in Alzheimer's disease pathology.

Introduction: A chemogenomic set of small molecules with annotated activities and implicated roles in Alzheimer's disease (AD) called the AD Informer Set was recently developed and made available to the AD research community: https://treatad.org/data-tools/ad-informer-set/. Methods: Small subsets of AD Informer Set compounds were selected for AD-relevant profiling. Nine compounds targeting proteins expressed by six AD-implicated genes prioritized for study by Target Enablement to Accelerate Therapy Development for Alzheimer's Disease (TREAT-AD) teams were selected for G-protein coupled receptor (GPCR), amyloid beta (Aß) and tau, and pharmacokinetic (PK) studies. Four non-overlapping compounds were analyzed in microglial cytotoxicity and phagocytosis assays. Results: The nine compounds targeting CAPN2, EPHX2, MDK, MerTK/FLT3, or SYK proteins were profiled in 46 to 47 primary GPCR binding assays. Human induced pluripotent stem cell (iPSC)-derived neurons were treated with the same nine compounds and secretion of Aß peptides (Aß40 and Aß42) as well as levels of phosphophorylated tau (p-tau, Thr231) and total tau (t-tau) peptides measured at two concentrations and two timepoints. Finally, CD1 mice were dosed intravenously to determine preliminary PK and/or brain-specific penetrance values for these compounds. As a final cell-based study, a non-overlapping subset of four compounds was selected based on single-concentration screening for analysis of both cytotoxicity and phagocytosis in murine and human microglia cells. Discussion: We have demonstrated the utility of the AD Informer Set in the validation of novel AD hypotheses using biochemical, cellular (primary and immortalized), and in vivo studies. The selectivity for their primary targets versus essential GPCRs in the brain was established for our compounds. Statistical changes in tau, p-tau, Aß40, and/or Aß42 and blood-brain barrier penetrance were observed, solidifying the utility of specific compounds for AD. Single-concentration phagocytosis results were validated as predictive of dose-response findings. These studies established workflows, validated assays, and illuminated next steps for protein targets and compounds.

AD Informer Set: Chemical tools to facilitate Alzheimer's disease drug discovery.

Introduction: The portfolio of novel targets to treat Alzheimer's disease (AD) has been enriched by the Accelerating Medicines Partnership Program for Alzheimer's Disease (AMP AD) program. Methods: Publicly available resources, such as literature and databases, enabled a data-driven effort to identify existing small molecule modulators for many protein products expressed by the genes nominated by AMP AD and suitable positive control compounds to be included in the set. Compounds contained within the set were manually selected and annotated with associated published, predicted, and/or experimental data. Results: We built an annotated set of 171 small molecule modulators targeting 98 unique proteins that have been nominated by AMP AD consortium members as novel targets for the treatment of AD. The majority of compounds included in the set are inhibitors. These small molecules vary in their quality and should be considered chemical tools that can be used in efforts to validate therapeutic hypotheses, but which will require further optimization. A physical copy of the AD Informer Set can be requested on the Target Enablement to Accelerate Therapy Development for Alzheimer's Disease (TREAT-AD) website. Discussion: Small molecules that enable target validation are important tools for the translation of novel hypotheses into viable therapeutic strategies for AD.

Multivalent DNA and nucleosome acidic patch interactions specify VRK1 mitotic localization and activity.

A key role of chromatin kinases is to phosphorylate histone tails during mitosis to spatiotemporally regulate cell division. Vaccinia-related kinase 1 (VRK1) is a serine-threonine kinase that phosphorylates histone H3 threonine 3 (H3T3) along with other chromatin-based targets. While structural studies have defined how several classes of histone-modifying enzymes bind to and function on nucleosomes, the mechanism of chromatin engagement by kinases is largely unclear. Here, we paired cryo-electron microscopy with biochemical and cellular assays to demonstrate that VRK1 interacts with both linker DNA and the nucleosome acidic patch to phosphorylate H3T3. Acidic patch binding by VRK1 is mediated by an arginine-rich flexible C-terminal tail. Homozygous missense and nonsense mutations of this acidic patch recognition motif in VRK1 are causative in rare adult-onset distal spinal muscular atrophy. We show that these VRK1 mutations interfere with nucleosome acidic patch binding, leading to mislocalization of VRK1 during mitosis, thus providing a potential new molecular mechanism for pathogenesis.

A chemical probe targeting the PWWP domain alters NSD2 nucleolar localization.

Nuclear receptor-binding SET domain-containing 2 (NSD2) is the primary enzyme responsible for the dimethylation of lysine 36 of histone 3 (H3K36), a mark associated with active gene transcription and intergenic DNA methylation. In addition to a methyltransferase domain, NSD2 harbors two proline-tryptophan-tryptophan-proline (PWWP) domains and five plant homeodomains (PHDs) believed to serve as chromatin reading modules. Here, we report a chemical probe targeting the N-terminal PWWP (PWWP1) domain of NSD2. UNC6934 occupies the canonical H3K36me2-binding pocket of PWWP1, antagonizes PWWP1 interaction with nucleosomal H3K36me2 and selectively engages endogenous NSD2 in cells. UNC6934 induces accumulation of endogenous NSD2 in the nucleolus, phenocopying the localization defects of NSD2 protein isoforms lacking PWWP1 that result from translocations prevalent in multiple myeloma (MM). Mutations of other NSD2 chromatin reader domains also increase NSD2 nucleolar localization and enhance the effect of UNC6934. This chemical probe and the accompanying negative control UNC7145 will be useful tools in defining NSD2 biology.

Reprogramming CBX8-PRC1 function with a positive allosteric modulator.

Canonical targeting of Polycomb repressive complex 1 (PRC1) to repress developmental genes is mediated by cell-type-specific, paralogous chromobox (CBX) proteins (CBX2, 4, 6, 7, and 8). Based on their central role in silencing and their dysregulation associated with human disease including cancer, CBX proteins are attractive targets for small-molecule chemical probe development. Here, we have used a quantitative and target-specific cellular assay to discover a potent positive allosteric modulator (PAM) of CBX8. The PAM activity of UNC7040 antagonizes H3K27me3 binding by CBX8 while increasing interactions with nucleic acids. We show that treatment with UNC7040 leads to efficient and selective eviction of CBX8-containing PRC1 from chromatin, loss of silencing, and reduced proliferation across different cancer cell lines. Our discovery and characterization of UNC7040 not only reveals the most cellularly potent CBX8-specific chemical probe to date, but also corroborates a mechanism of Polycomb regulation by non-specific CBX nucleotide binding activity.

Discovery and Development of Cyclic Peptide Inhibitors of CIB1.

Calcium and integrin binding protein 1 (CIB1) is a small, intracellular protein recently implicated in survival and proliferation of triple-negative breast cancer (TNBC). Considering its interactions with PAK1 and downstream signaling, CIB1 has been suggested as a potential therapeutic target in TNBC. As such, CIB1 has been the focus of inhibitor discovery efforts. To overcome issues of potency and stability in previously reported CIB1 inhibitors, we deploy mRNA display to discover new cyclic peptide inhibitors with improved biophysical properties and cellular activity. We advance UNC10245131, a cyclic peptide with low nanomolar affinity and good selectivity for CIB1 over other EF-hand domain proteins and improved permeability and stability over previously identified linear peptide inhibitor UNC10245092. Unlike UNC10245092, UNC10245131 lacks cytotoxicity and does not affect downstream signaling. Despite this, UNC10245131 is a potent ligand that could aid in clarifying roles of CIB1 in TNBC survival and proliferation and other CIB1-associated biological phenotypes.

UNC5293, a potent, orally available and highly MERTK-selective inhibitor.

Inhibition of MER receptor tyrosine kinase (MERTK) causes direct tumor cell killing and stimulation of the innate immune response. Therefore, MERTK has been identified as a therapeutic target in a wide variety of human tumors. Clinical trials targeting MERTK have recently been initiated, however, none of these drugs are MERTK-specific. Herein, we present the discovery of a highly MERTK-selective inhibitor UNC5293 (24). UNC5293 has subnanomolar activity against MERTK with an excellent Ambit selectivity score (S50 (100 nM) = 0.041). It mediated potent and selective inhibition of MERTK in cell-based assays. Furthermore, it has excellent mouse PK properties (7.8 h half-life and 58% oral bioavailability) and was active in bone marrow leukemia cells in a murine model.

Integrating DNA-encoded chemical libraries with virtual combinatorial library screening: Optimizing a PARP10 inhibitor.

Two critical steps in drug development are 1) the discovery of molecules that have the desired effects on a target, and 2) the optimization of such molecules into lead compounds with the required potency and pharmacokinetic properties for translation. DNA-encoded chemical libraries (DECLs) can nowadays yield hits with unprecedented ease, and lead-optimization is becoming the limiting step. Here we integrate DECL screening with structure-based computational methods to streamline the development of lead compounds. The presented workflow consists of enumerating a virtual combinatorial library (VCL) derived from a DECL screening hit and using computational binding prediction to identify molecules with enhanced properties relative to the original DECL hit. As proof-of-concept demonstration, we applied this approach to identify an inhibitor of PARP10 that is more potent and druglike than the original DECL screening hit.