Evotec (United Kingdom)

Orthosteric inhibitors of the human heterodimeric DNA mismatch repair complex MutSbeta were identified by high-throughput screening. Following extensive hit confirmation to remove false positives, two series were found to give consistent activity free of likely artefactual effects. Extensive hit profiling confirmed an ATP-competitive mode of action, and X-ray crystallography showed the inhibitors occupying the ATP-binding site of MSH3.

Inhibition of leucine-rich repeat kinase 2 is a genetically supported mechanism for the treatment of Parkinson's disease. We previously disclosed the discovery of an indazole series lead that demonstrated both safety and translational risks. The safety risks were hypothesized to be of unknown origin, so structural diversity in subsequent chemical matter was prioritized. The translational risks were identified due to a low brain Kpu,u in nonhuman primate studies, which raised concern over the use of an established peripheral biomarker as a surrogate for central target engagement. Given these challenges, the team sought to leverage structure- and property-based drug design and expanded efflux transporter profiling to identify structurally distinct leads with enhanced CNS drug-likeness. Herein, we describe the discovery of a "reinvented" indazole series with improved physicochemical properties and efflux transporter profiles while maintaining excellent potency and off-target kinase selectivity, which resulted in advanced lead, compound 23.

BackgroundThe PET (positron emission tomography) radioligand [11C]CHDI-180R was designed to target mutant huntingtin (mHTT) aggregates in Huntington's disease (HD) brains. It has been validated in preclinical HD models and has demonstrated promising kinetic properties in a first-in-human study.AimIn this investigation, focus was to identify mHTT proteoforms that CHDI-180 binds to, which is crucial for understanding its pharmacology and developing second generation tracers.MethodsProbes with photoreactive chloroalkane tag were designed to selectively isolate mHTT target(s). UV-induced covalent photo-cross-linkage is applied to secure the binding between probe and mHTT. Targeted mHTT proteoform(s) cross-linked to ligand is enriched through covalent capture onto beads. After capture, target-ligand complex is selectively released from beads for characterization by using mass spectrometry, mass photometry (MP), gel electrophoresis (AGERA) and negative stain electron microscopy (nsEM). MP and nsEM have been optimized for analysis of various mHTT proteoforms.ResultsWe show optimization of key steps, including coupling of ligands to beads and linker cleavage. Our results indicate that all 6 PAL ligands can be successfully coupled and cleaved. The extent of coupling and cleavage varied based on applied conditions and ligand investigated. Results from a time course of recombinantly derived aggregated mHTT analyzed by nsEM and MP will be presented.ConclusionsOur studies revealed that nsEM and MP are valuable tools to characterize mHTT aggregates. Optimized methods will be applied to characterize mHTT proteoforms to which CHDI-180 binds and to compare CHDI-180 and other HTT PET tracer target(s) in animals and human post-mortem brain samples.

The identification of VHL-binding proteolysis targeting chimeras (PROTACs) that potently degrade the BRM protein (also known as SMARCA2) in SW1573 cell-based experiments is described. These molecules exhibit between 10- and 100-fold degradation selectivity for BRM over the closely related paralog protein BRG1 (SMARCA4). They also selectively impair the proliferation of the H1944 "BRG1-mutant" NSCLC cell line, which lacks functional BRG1 protein and is thus highly dependent on BRM for growth, relative to the wild-type Calu6 line. In vivo experiments performed with a subset of compounds identified PROTACs that potently and selectively degraded BRM in the Calu6 and/or the HCC2302 BRG1 mutant NSCLC xenograft models and also afforded antitumor efficacy in the latter system. Subsequent PK/PD analysis established a need to achieve strong BRM degradation (>95%) in order to trigger meaningful antitumor activity in vivo. Intratumor quantitation of mRNA associated with two genes whose transcription was controlled by BRM (PLAU and KRT80) also supported this conclusion.

BackgroundT-cell immunotherapy with chimeric antigen receptors (CARs) has evolved as part of the standard of care for several hematological malignancies and has transformed the oncology landscape. CAR-T-cell products are traditionally generated from autologous, patient-derived αβT-cells engineered with a CAR for tumor cell targeting. Generating autologous T-cell products for each patient is associated with manufacturing and logistical complexity and high product costs. Furthermore, manufacturing fails in a significant number of cases due to the poor quality and quantity of blood-derived T-cells and restrictions apply in terms of throughput to produce autologous cell products. Consequently, many patients are left without this treatment option, so it is crucial to develop strategies for off-the-shelf T-cell products. Immune cells derived from induced pluripotent stem cells (iPSCs) offer the opportunity to manufacture allogeneic T-cell products with consistently high quality and scalable quantities. Use of iPSCs as a starting material makes it easier to introduce several genetic modifications (e.g. to enhance both cell persistence and tumor infiltration) addressing tumor resistance mechanisms for both liquid and solid tumors.MethodsUsing a validated GMP iPSC line modified with an NY-ESO-1-specific T-cell receptor (TCR) knock-in, we have established a feeder-free differentiation protocol that enables robust production of iPSC-derived αβT-cells (iαβT). Flow cytometry and single cell transcriptome analysis ensured a stringent monitoring of all process stages. To demonstrate functional activity of our iαβT, we performed cytotoxicity and cytokine release assays against tumor cell lines presenting the NY-ESO-1 antigen.ResultsSuccessful knock-in of the NY-ESO-1-TCR-transgene cassette into the iPSC line was confirmed by marker-gene expression. Hematopoietic progenitor cells were induced from knock-in-enriched iPSCs and differentiated into iαβT. During the differentiation process, the T-cell markers CD45, CD5 and CD7 were displayed and cells started to express the TCR. After activation of iαβT, the fraction of NY-ESO-1-TCR-positive cells increased to over 95%. Importantly, iαβT expressed CD8α and CD8β which is crucial for the function of cytotoxic T-cells. Transcriptome analysis validated the efficient differentiation from pluripotent cells towards cells with a T-cell-specific gene expression profile. Co-culture experiments with NY-ESO-1 antigen presenting tumor cell lines confirmed cytotoxic activity of iαβT and their potential to release cytokines.ConclusionsOur scalable iαβT differentiation process enables us to generate CD8+ T-cells that secrete cytokines and show cytotoxic activity, indicating their potential as a promising cell source for TCR-T or CAR-T cancer immunotherapies.

Computational methods in modern drug discovery have become ubiquitous, with methods that cover most of the discovery stages: from hit finding and lead identification to lead optimization. The overall aim of these computational methods is to obtain a more efficient discovery process, by reducing the number of "wet" experiments required to produce therapeutics that have higher probability of succeeding in clinical development and subsequently benefitting end patients by developing highly effective therapeutics having minimal side effects. Virtual Screening is usually applied at the early stage of drug discovery, looking to find chemical matter having desired properties, such as molecular shape, electrostatics, and pharmacophores at desired three-dimensional positions. The aim of this stage is to search in a wide chemical space, including chemistry available from commercial suppliers and virtual databases of predicted reaction products, to identify molecules that would exert a particular biochemical response. This initial stage of the discovery process is very important since the subsequent stages will use the initial chemical motifs that have been found at the hit finding stage, and therefore the most suitable the compound is found, the more likely it is that subsequent stages will be successful and less time and resource consuming. This chapter provides a summary of various Virtual Screening methods, including shape match and molecular docking, and these methods are used in a Virtual Screening workflow that is provided as an example which is described to be run automatically in cloud resources. This automatic in-depth exploration of the chemical space using validated Virtual Screening methods can lead to a more streamlined and efficient discovery process, aiming to deliver chemical matter of high quality and maximizing the required biological effects while minimizing adverse effects. Surely, Virtual Screening pipelines of this nature will continue to play a central role in producing much needed therapeutics for the health challenges of the future.

One area of active research is the use of natural language processing (NLP) to mine biomedical texts for sets of triples (subject-predicate-object) for knowledge graph (KG) construction. While statistical methods to mine co-occurrences of entities within sentences are relatively robust, accurate relationship extraction is more challenging. Herein, we evaluate the Global Network of Biomedical Relationships (GNBR), a dataset that uses distributional semantics to model relationships between biomedical entities. The focus of our paper is an evaluation of a subset of the GNBR data; the relationships between chemicals and genes/proteins. We use Evotec's structured 'Nexus' database of >2.76M chemical-protein interactions as a ground truth to compare with GNBRs relationships and find a micro-averaged precision-recall area under the curve (AUC) of 0.50 and a micro-averaged receiver operating characteristic (ROC) curve AUC of 0.71 across the relationship classes 'inhibits', 'binding', 'agonism' and 'antagonism', when a comparison is made on a sentence-by-sentence basis. We conclude that, even though these micro-average scores are modest, using a high threshold on certain relationship classes like 'inhibits' could yield high fidelity triples that are not reported in structured datasets. We discuss how different methods of processing GNBR data, and the factuality of triples could affect the accuracy of NLP data incorporated into knowledge graphs. We provide a GNBR-Nexus(ChEMBL-subset) merged datafile that contains over 20,000 sentences where a protein/gene-chemical co-occur and includes both the GNBR relationship scores as well as the ChEMBL (manually curated) relationships (e.g., 'agonist', 'inhibitor') -this can be accessed at https://doi.org/10.5281/zenodo.8136752. We envisage this being used to aid curation efforts by the drug discovery community.