Advances in Cellular Target Engagement and Target Deconvolution.

Abstract

Proof of target engagement, the interaction between a lead molecule and its protein target, is a critical determinant of success in drug discovery. Structural and biophysical analysis of compound binding to purified target protein is a core element of many drug discovery projects, for example, x-ray crystallography to provide a molecular view of drug–target interaction and surface plasmon resonance to measure the kinetics of drug binding. Biochemical assays to determine target engagement often come in the form of enzyme inhibition assays, and these readouts are frequently applied in hit identification campaigns. While target engagement in biochemical settings provides critical insights around binding modes and avenues to rapidly identify hit series, these findings are not guaranteed to translate into desired phenotypic effects. Differences in cellular and buffered environments, including the state of the target of interest as well as concentrations of cofactors, may raise challenges when trying to correlate phenotypic effects directly with biochemical readouts. Thus, the ability to monitor target engagement in a live-cell setting provides the critical link between biochemical binding and phenotype. Cellular target engagement tells whether a biochemical binding event occurs in a cell and, if so, whether properties such as potency and kinetics are retained. Furthermore, cellular target engagement, when matched with phenotypic responses, offers direct evidence on whether a particular method of drugging a target is appropriate. Direct measurement of target engagement in cells and tissues is still in its relative infancy. However, methods are emerging that allow for measuring target engagement in a high-throughput and miniaturized manner amenable for hit ID as well as understanding target engagement in the most relevant disease models. Quantitative methods of reading out target engagement events have also improved the accuracy of cellular target engagement readouts. In this special issue of SLAS Discovery, we explore multiple aspects of cellular target engagement, from direct measurement of compound–target interaction using BRET technologies, through measurement of compound-induced changes in protein stability and localization, to in vivo drug distribution and target engagement studies. We also highlight technologies that enable a more holistic view of target engagement by mapping multiple compound–protein interactions in a single study. When applied in combination, these methods can build a strong platform of evidence linking compound to target to biological response, thereby impacting project attrition and medicine discovery success. Several articles in this special issue utilize bioluminescence resonance energy transfer (BRET) technology to measure interactions between compound and target protein, or between two different proteins. The application of these methods to cellular target engagement assays has been particularly influenced by the introduction of Nanoluciferase as a small, extremely bright, genetically encodable protein that can be used as a “NanoBRET” energy donor in combination with a long-wavelength fluorophore energy acceptor such as a HaloTag.1Machleidt T. Woodroofe C. Schwinn M. et al.NanoBRET–A Novel BRET Platform for the Analysis of Protein-Protein Interactions.ACS Chem. Biol. 2015; 10: 1797-1804Google Scholar Jin and colleagues2Jin H.Y. Tudor Y. Choi K. et al.High-Throughput Implementation of the NanoBRET® Target Engagement Intracellular Kinase Assay To Reveal Differential Compound Engagement by SIK2/3 Isoforms.SLAS Discov. 2020; 25: 215-222Google Scholar describe the development and application of 384-well NanoBRET assays to measure compound–target engagement for SIK kinases. By expressing SIK-Nanoluc fusion proteins for their primary target, SIK2, and for a countertarget, SIK3, and establishing BRET assays with a pan-kinase fluorescent tracer molecule, the authors were able to establish robust competition assays and run a small-molecule screen that identified SIK2:SIK3 selective molecules. Ong et al.3Ong L.L. Vasta J.D. Monereau L. et al.A High Throughput BRET Cellular Target Engagement Assay Links Biochemical-to-Cellular Activity for Bruton’s Tyrosine Kinase.SLAS Discov. 2020; 25: 176-185Google Scholar describe a miniaturized, NanoBRET ligand displacement assay for Bruton’s tyrosine kinase (BTK). With this target, high levels of target occupancy are believed to be critical for efficacy; therefore, assays that measure target engagement and residence time in cells are critical in the discovery of next-generation inhibitors. Using a Nanoluciferase-tagged BTK construct and a fluorophore-labeled probe compound, the authors were able to develop a robust competition assay that can be applied to measure both steady-state IC50 values and drug–target residence time by running the assay in kinetic mode. Phillipou and co-workers4Phillipou A.N. Lay C.S. Carver C.E. et al.Cellular Target Engagement Approaches to Monitor Epigenetic Reader Domain Interactions.SLAS Discov. 2020; 25: 163-175Google Scholar have applied NanoBRET technology to develop a range of assays for BET bromodomain proteins, which are epigenetic targets of interest in cancer and inflammatory disorders, and they describe how these assays can be used in a drug discovery cascade to drive potency and selectivity. Thus, assays for both compound–target and target–histone interactions are described, and the authors extend their work to develop a NanoBRET assay in which a Nanoluciferase fragment (HiBit) is tagged to the endogenous BRD4 protein using CRISPR/Cas9 technology. Nanoluciferase technology can also be applied to measure a change in membrane protein localization that occurs following compound–target binding: Soave and colleagues5Soave M. Kellam B. Woolard J. NanoBiT Complementation to Monitor Agonist-Induced Adenosine A1 Receptor Internalization.SLAS Discov. 2020; 25: 186-194Google Scholar used the recently developed NanoBiT technology, where Nanoluciferase is split into a small, high-affinity HiBiT tag and a larger NanoLuc subunit (LgBiT), to measure ligand-dependent internalization of a G-protein-coupled receptor. By tagging the adenosine A1 receptor with HiBiT and adding purified LgBiT protein to the media, complementation occurs at the cell surface and agonist-dependent receptor internalization is measured as a loss of luminescence using a bioluminescent imaging microscope. In their native cellular environment, proteins typically reside in a complex with other partners. While this provides a strong rationale for the use of cell assays in a drug discovery cascade, it can also present new opportunities for drug discovery: Siddiqui et al.6Siddiqui F.A. Parkkola H. Manoharan G.B. et al.Medium-Throughput Detection of Hsp90/Cdc37 Protein–Protein Interaction Inhibitors Using a Split Renilla Luciferase-Based Assay.SLAS Discov. 2020; 25: 195-206Google Scholar describe the development of a cell lysate-based split Renilla luciferase assay to identify inhibitors that act at the protein–protein interface of Hsp90 and Cdc37. This assay is particularly applicable to identifying allosteric, C-terminal Hsp90/Cdc37 inhibitors (ATP competitive inhibitors are not detected in the assay), and the authors propose that this approach can mitigate the known toxicity associated with inhibitors against the Hsp90 ATP-binding site. The cellular thermal shift assay (CETSA) method, which utilizes the established thermodynamic principle of ligand-induced stabilization, is now established as a cellular target engagement method. In this special issue, Seashore-Ludlow and colleagues7Seashore-Ludlow B. Axelsson H. Lundbäck T. Perspective on CETSA Literature—Towards More Quantitative Data Interpretation.SLAS Discov. 2020; 25: 118-126Google Scholar present a comprehensive analysis of the approximately 270 peer-reviewed CETSA publications to date, with a focus on experimental design and how the choice of protocol used can influence experimental outcome and interpretation of the data. The authors also provide a perspective on future opportunities in the CETSA field, including application to more translatable cell and tissue samples where input material is scarce and new methods are required. In their analysis, Seashore-Ludlow et al. report that 85% of CETSA publications use Western blot as the primary endpoint. In this format, lack of throughput and sample requirements can be a challenge. In this special issue, Herledan and colleagues8Herledan A. Lejeune-Dodge A. Leroux F. et al.Drug Target Engagement Using Coupled Cellular Thermal Shift Assay—Acoustic Reverse-Phase Protein Array.SLAS Discov. 2020; 25: 207-214Google Scholar describe a new method—CETSA-acoustic reverse-phase protein array—that offers the potential to address these challenges. The authors have developed a method whereby nanoacoustic transfer of the soluble protein fraction is used to position an array of samples on a nitrocellulose membrane. Protein can then be detected by fluorescent imaging using a target-specific primary and fluorescent secondary antibody. Henderson et al.9Henderson M.J. Holbert M.A. Simeonov A. et al.High-Throughput Cellular Thermal Shift Assays in Research and Drug Discovery.SLAS Discov. 2020; 25: 137-147Google Scholar provide an analysis of high-throughput (HT) CETSA methods, which utilize conventional plate-based assay readouts to detect soluble, folded protein, and therefore allow application in a higher-throughput setting, for example, analysis of hits from a high-throughput screen where it has been possible to rank hits according to the EC50 value of protein stabilization. Several formats of HT CETSA have been described, including antibody-based (measuring endogenous protein) and reporter-based, where a tag is added to the target protein. The authors provide a perspective on how best to apply these methods in a drug discovery setting. The methods described above can be applied to measure the binding of a compound to its targeted protein in cells. However, in other circumstances the experimenter may wish to carry out a hypothesis-free experiment to map compound binding across a wider range of proteins, for example, deconvolution of a compound identified in a phenotypic screen, or analysis of the off-target binding of a compound with a safety liability. In this special issue, several methods are described that can be used to deconvolute the pharmacological activity of an active compound; thus, Seashore-Ludlow et al. reference the mass spectrometry (MS) format of CETSA (CETSA MS; also referred to as thermal proteome profiling [TPP]) that can applied to map compound–protein binding without the requirement to label either moiety. This topic has recently been reviewed elsewhere in some detail.10Dai L. Prabhu N. Yu L.Y. et al.Horizontal Cell Biology: Monitoring Global Changes of Protein Interaction States with the Proteome-Wide Cellular Thermal Shift Assay (CETSA).Annu. Rev. Biochem. 2019; 88: 383-408Google Scholar Freeth and Soden11Freeth J. Soden J. New Advances in Cell Microarray Technology to Expand Applications in Target Deconvolution and Off-Target Screening.SLAS Discov. 2020; 25: 223-230Google Scholar describe an array-based target deconvolution method for membrane proteins, which are traditionally challenging to address using proteomic methods. Cells are transfected using an array of cDNAs encoding a membrane protein library, and a “fingerprint” of binding is established for the therapeutic molecule that enables an unbiased and comprehensive analysis of membrane targets. In this article, the authors describe two novel builds on the technology: First, they have adapted the method to express a library of secreted proteins that are tethered to the cell membrane by fusion to an inert transmembrane and intracellular domain. Second, they describe a novel approach to uncover receptor–receptor interactions using whole-cell “baits” expressing a receptor of interest. A critical element of target deconvolution is the generation of chemical probes that can be used to label subsets of proteins in situ. Song and Zheng12Song J. Zheng G. Bioorthogonal Reporters for Detecting and Profiling Protein Acetylation and Acylation.SLAS Discov. 2020; 25: 148-162Google Scholar describe the development of bioorthogonal reporters for detecting and profiling protein acetylation and acylation. Lysine acetylation/acylation is an important posttranslational modification, catalyzed by a class of enzymes called lysine acetyltransferases. The authors describe a method whereby acetyl or acyl groups can be derivatized with reactive handles and then modified with CoA for incorporation as posttranslational modifications. These reactive handles (e.g., alkyne azide) are amenable to subsequent functionalization with reporter groups such as fluorophores for imaging or biotin for enrichment and mass spectrometry analysis, all of which allow one to elucidate lysine acetyltransferase biology, including important topics such as pathway effects, mechanisms, and substrate selectivities. In the final article of this special issue, Maynard and Hart13Maynard J. Hart P. The Opportunities and Use of Imaging to Measure Target Engagement.SLAS Discov. 2020; 25: 127-136Google Scholar provide a perspective on the use of in vivo and ex vivo imaging to measure target engagement. Whereas the majority of the articles in this issue study isolated cells, often using a model system, the in vivo situation is much more complex with the target protein residing in a multicellular organ, with the associated barriers to compound access and understanding of pharmacology. The authors describe the methods that can be used to measure compound biodistribution, either in the live phase using techniques such as positron emission tomography (PET) or near-infrared imaging, or ex vivo using mass spectrometry imaging. These methods can also be used to measure downstream changes in cells and tissues in response to drug treatment, for example, glucose turnover using 18F-FDG PET or metabolic changes using mass spectrometry imaging. Declaration of Conflicting Interests The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Martin Main is employed by Medicines Discovery Catapult and his research and authorship of this article was completed within the scope of his employment with Medicines Discovery Catapult.