Sergiy Levin

Target engagement and drug residence time can be observed in living cells with BRET

The therapeutic action of drugs is predicated on their physical engagement with cellular targets. Here we describe a broadly applicable method using bioluminescence resonance energy transfer (BRET) to reveal the binding characteristics of a drug with selected targets within intact cells. Cell-permeable fluorescent tracers are used in a competitive binding format to quantify drug engagement with the target proteins fused to Nanoluc luciferase. The approach enabled us to profile isozyme-specific engagement and binding kinetics for a panel of histone deacetylase (HDAC) inhibitors. Our analysis was directed particularly to the clinically approved prodrug FK228 (Istodax/Romidepsin) because of its unique and largely unexplained mechanism of sustained intracellular action. Analysis of the binding kinetics by BRET revealed remarkably long intracellular residence times for FK228 at HDAC1, explaining the protracted intracellular behaviour of this prodrug. Our results demonstrate a novel application of BRET for assessing target engagement within the complex milieu of the intracellular environment.

Deciphering the Cellular Targets of Bioactive Compounds Using a Chloroalkane Capture Tag

Phenotypic screening of compound libraries is a significant trend in drug discovery, yet success can be hindered by difficulties in identifying the underlying cellular targets. Current approaches rely on tethering bioactive compounds to a capture tag or surface to allow selective enrichment of interacting proteins for subsequent identification by mass spectrometry. Such methods are often constrained by ineffective capture of low affinity and low abundance targets. In addition, these methods are often not compatible with living cells and therefore cannot be used to verify the pharmacological activity of the tethered compounds. We have developed a novel chloroalkane capture tag that minimally affects compound potency in cultured cells, allowing binding interactions with the targets to occur under conditions relevant to the desired cellular phenotype. Subsequent isolation of the interacting targets is achieved through rapid lysis and capture onto immobilized HaloTag protein. Exchanging the chloroalkane tag for a fluorophore, the putative targets identified by mass spectrometry can be verified for direct binding to the compound through resonance energy transfer. Using the interaction between histone deacetylases (HDACs) and the inhibitor, Vorinostat (SAHA), as a model system, we were able to identify and verify all the known HDAC targets of SAHA as well as two previously undescribed targets, ADO and CPPED1. The discovery of ADO as a target may provide mechanistic insight into a reported connection between SAHA and Huntingtons disease.

Abstract C158: A chemoproteomics strategy for target identification and lead compound optimization using chloroalkane derivatized compounds

The Identification and validation of drug targets is an industry wide challenge. There is a very clear and urgent need for technologies that can identify the interaction partners of small molecules. Chemical proteomics is one technology that has attracted attention as a solution to the drug target identification problem. Here we present a new approach utilizing a chloroalkane (CA) moiety capture handle, which can be chemically attached to small molecules to isolate their respective protein partners. In general derivatization of small molecules with the CA moiety does not impact their cell permeability and has limited impact on potency, allowing for phenotypic assays of the derivatized compound to be performed. The retention of cell permeability also allows for the capture process to be performed from live cells treated with the CA-compound. This process is also compatible with competition assays, which can be used to evaluate and compare other compounds exhibiting a similar mode of action. Here we present a study using Dasatinib-CA, a modified kinase inhibitor and potent anti-cancer drug which binds to a broad range of kinases. First we performed target enrichment experiments by incubating K562 cells with the modified Dasatinib (Dasatinib-CA). Cells were lysed and the Dasatinib-CA, together with the bound targets, was rapidly captured onto magnetic resin coated with HaloTag. Unmodified compound was used to competitively elute putative interacting proteins. Secondly using the same assay format we evaluated the relative target affinities of Dasatinib-CA versus competing molecules. Competition assays were performed by incubating multiple mixtures of Dasatinib analogues and Dasatinib-CA at varying relative concentrations. Eluted proteins were processed using SDS-PAGE and in-gel digestion. For target identification experiments peptides were analyzed using label free mass spectrometry. For the competition assays digested material was labeled with Tandem Mass Tags (TMT) 10plex reagents. Peptides were analyzed using nanoscale LC-MS/MS coupled with a Q Exactive mass spectrometer (Thermo). Protein identification and quantitation was performed with MaxQuant (MaxQuant.org) and data validation and visualization was performed using Perseus (Perseus-framework.org). Using this approach we identified over 30 kinases, including known membrane associated and membrane protein targets. This work presented here highlights a new method for chemical proteomics and demonstrates utility of the platform to enable target identification and to evaluate competitor molecules. Citation Format: Michael Ford, Richard Jones, Ravi Amunugama, Danette Daniels, Rachel Ohana, Sergiy Levin, Thomas Kirkland, Marjeta Urh, Keith Wood. A chemoproteomics strategy for target identification and lead compound optimization using chloroalkane derivatized compounds. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C158.

Abstract B161: Chemoproteomic profiling for target identification using chloroalkane derivatized compounds

Chemoproteomic technologies enable the quantitative and qualitative profiling of small molecule protein interactions. Here we report a new approach utilizing a chloroalkane (CA) capture tag which can be chemically attached to small molecules to enable the isolation of their respective protein targets through selective capture onto an immobilized HaloTag protein. We have combined our chemoproteomics workflow with multiplexed chemical labeling methods to enable the analysis of dose dependent target engagement studies in a single LC-MS/MS experiment. Here we demonstrate this workflow using BIRB796-CA, a modified kinase inhibitor. Target enrichment experiments were performed by incubating HEK293 cells with the BIRB-CA. Cells were then lysed and the BIRB-CA, together with the bound targets, was rapidly captured onto immobilized HaloTag protein. Unmodified compound was used to competitively elute putative interacting proteins. To study dose dependent effects on the protein interaction profile, cells were incubated with multiple mixtures of BIRB and BIRB-CA at varying relative concentrations. The eluted proteins were processed using SDS-PAGE and in-gel digestion. Digested material was labeled with Tandem Mass Tags (TMT) 10plex reagents and analyzed using nanoscale LC-MS/MS coupled with a Q Exactive mass spectrometer (Thermo). Protein identification was performed with MaxQuant (MaxQuant.org) and data validation and visualization was performed using Perseus (Perseus-framework.org). We tested this chemoproteomics workflow using the interaction of MAP kinases (MAPK) with BIRB796, an allosteric kinase inhibitor. Treatment of LPS simulated THP-1 cells with BIRB796 and BIRB-CA revealed minimal impact of the CA modification on the inhibition of TNFα secretion. The interaction of BIRB-CA with known BIRB targets MAPK13, MAPK12, MAPK9 and MAPK14 has been confirmed using label free quantitative proteomic profiling. We have optimized the TMT labeling method for minimal losses and sample processing steps. We are applying the TMT method to BIRB-CA and control pull downs and extending this workflow to include the BIRB/BIRB-CA titration samples. Here we will use post-acquisition data processing of the BIRB/BIRB-CA titration pull down data to assess compound binding characteristics with the inferred extension to compounds with unknown binding specificities. Citation Format: Michael Ford, Richard Jones, Ravi Amunugama, Danette Daniels, Rachel Ohana, Sergiy Levin, Thomas Kirkland, Marjeta Urh, Keith Wood. Chemoproteomic profiling for target identification using chloroalkane derivatized compounds. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr B161.

Abstract 3512: Measuring intracellular target engagement and drug residence time with nanoBRET

We present a biophysical method to directly measure target engagement within intact mammalian cells using bioluminescence energy transfer (BRET). Compound interactions with intracellular targets can be detected with complete specificity by their ability to compete with energy transfer complexes introduced into the cells. These complexes can be detected at physiologically relevant levels by exploiting an extraordinarily bright luciferase (NanoLuc), together with fluorescent tracers optimized for cell-permeability and spectral resolution from the luciferase. We demonstrate applications of the technology for target engagement among key drug target classes, including; kinases, histone deacetylases (HDACs), bromodomains, and the methyltransferase EZH2. Intracellular selectivity and affinity profiles of various reference compounds and approved drugs will be presented. For a panel of HDAC inhibitors, affinity profiles for specific HDAC isozymes strongly correlate with phenotypic potencies (e.g. cell viability). Furthermore, the luminescent output of the energy transfer complex enables a technique to monitor ligand occupancy in real-time. Association and dissociation rates can be derived from the kinetic measurements, providing a means to quantify drug residence time on select targets within intact cell populations. This novel application of intracellular BRET should significantly advance target engagement work flows, and allow for intracellular target affinities to be coupled to phenotypic outcomes. Citation Format: Matthew B. Robers, Melanie Dart, Chad Zimprich, Thomas Kirkland, Sergiy Levin, Thomas Machleidt, Jim Hartnett, Kris Zimmerman, Rachel Ohana, Danette Daniels, Mei Cong, Frank Fan, Keith Wood. Measuring intracellular target engagement and drug residence time with nanoBRET. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3512. doi:10.1158/1538-7445.AM2015-3512

Abstract 2003: Investigating the cellular interactions of BIRB796 analogues using a novel chloroalkane capture tag

Proceedings: AACR 106th Annual Meeting 2015; April 18-22, 2015; Philadelphia, PA Identifying the targets of a bioactive compound is often the rate limiting step toward understanding the molecular mechanism of drug action. Current approaches rely on linking the bioactive compound to a surface or an affinity handle, permitting selective capture of interacting proteins for identification by mass spectrometry. A major consideration with these methods is insuring that the chemical derivatization of the bioactive compound does not disrupt the binding interactions with the cellular targets. We have developed a method based on a novel chloroalkane capture tag that minimally affects compound potency and cell permeability. This allows verification of the pharmacological activity of the modified compound, thus increasing the confidence in the biological relevance of captured proteins. In addition, by allowing the chloroalkane-modified compound to bind the targets within living cells, the cellular architecture during the binding step is preserved and better represents the conditions that the unaltered compound would normally engage these targets. Following binding with the tagged compound in live cells, the cells are lysed and the chloroalkylated compound and its associated targets are rapidly captured onto immobilized HaloTag protein and then released by competitive elution. The identified targets are then validated for direct binding relationship with the bioactive compound by bioluminescence energy transfer. We tested this target capture/target validation work flow using the interaction of MAPK kinases with two allosteric kinase inhibitors (BIRB796 and a BIRB analog exhibiting 100-fold lower potency). RESULTS: Using the two BIRB-chloroalkane derivatives to selectively enrich for targets from HEPG2 cells, we identified and validated multiple relevant MAPK kinases as well as additional off-targets. Interestingly, all the discovered off-targets bind purines. Kinase inhibitors such as BIRB796 which acts by binding to the kinase ATP binding site can interact in a similar manner with other purine binding proteins. Using bioluminescence energy transfer we interrogated the affinity and residence time of the two BIRB compounds to multiple MAPK kinases inside living cells. Our results indicates that the BIRB796 analog exhibits 30-1000 fold reduced affinity to multiple MAPK kinases as well as significant shorter residence time compared to BIRB796. Taken together these results indicate that our workflow can reveal the direct binding relationships between bioactive compounds and their cellular targets and contribute to further understanding of these interactions. Citation Format: Rachel Friedman Ohana, Robin Hurst, Thomas Kirkland, Carolyn Woodroofe, Sergiy Levin, Paul Otto, Tetsuo Uyeda, Michael Ford, Richard Jones, Danette Daniels, Marjeta Urh, Keith Wood. Investigating the cellular interactions of BIRB796 analogues using a novel chloroalkane capture tag. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2003. doi:10.1158/1538-7445.AM2015-2003

Abstract C224: Identification of drug targets via their intracellular interactions.

Phenotypic-based screening of small molecule libraries is a significant trend in drug discovery. However, deciphering the intracellular interactions between proteins and small molecules that mediate phenotypic outcomes remains a major challenge. Current approaches rely on linking the small molecule to a surface or affinity tag (e.g., biotin), allowing selective capture of the corresponding protein target for identification by mass spectrometry. Verifying that the linkage method does not disrupt the pharmacological activity is important for ensuring maintenance of the binding interaction with the intracellular targets. Consequently, although target capture is commonly done from cell lysates, methods compatible with living cells are preferable. We have developed such a method based on a novel chloroalkane tag that minimally affects compound potency and cell permeability. This allows phenotypic outcomes to be recapitulated by the modified compound, increasing confidence in the biological relevance of the captured proteins. The tagged compound is allowed to bind at equilibrium to intracellular targets, then cells are lysed and the chloroalkane bound to its protein targets is quickly captured onto magnetic particles containing immobilized HaloTag. The rapid isolation method minimizes complex collapse, preserves low affinity interactions, and in combination with low background adsorption facilitates target identification by mass spectrometry. Engagement of identified targets by compounds in living cells can be verified using bioluminescence energy transfer (BRET). NanoLuc luciferase fused to the target protein serves as a BRET donor to the compound derivatized with fluorophore adducts. As a model to test this combined target capture/target validation work flow we selected the interaction of histone deacetylase (HDACs) and Vorinostat (SAHA) a broad- HDAC inhibitor. This family of deacetylases exhibits a range of cellular abundance and affinity to SAHA (0.001-1µM), and has members localized in the nucleus or cytoplasm. Results: Treatment of K562 cells with SAHA or a derivative having the chloroalkane tag revealed high cellular potency against HDAC class I/IIb with IC50 of 0.1µM and 0.2µM respectively. These results indicate minimal impact of the chloroalkane modification on the drug potency. For target identification, the SAHA chloroalkane derivative was allowed to bind at equilibrium to intracellular targets followed by rapid capture onto immobilized HaloTag. By this method we were able to identify all the direct interactors of SAHA (i.e., HDAC 1,2,3,6 and 8) regardless of their subcellular localization, abundance, or affinity. The intracellular BRET approach was next used to validate the set of identified HDACs and to rank order their relative binding affinity to SAHA. Our results suggested that these approaches used in combination provide reliable identification of targets in living cells. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C224. Citation Format: Rachel Friedman Ohana, Matt Robers, Thomas A. Kirkland, Carolyn C. Woodroofe, Chad Zimprich, Tetsuo H. Uyeda, Paul Otto, Sergiy Levin, Keith Wood. Identification of drug targets via their intracellular interactions. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C224.