Publication Summary

A multi-targeted probe-based strategy to identify signaling vulnerabilities in cancers

Suman Rao,1,2,3 Deepak Gurbani,4 Guangyan Du,2,3 Robert A. Everley,1 Christopher M. Browne,2,3,5 Apirat Chaikuad,6,7 Li Tan,2,3 Martin Schroder,6,7 Sudershan Gondi,4 Scott B. Ficarro,2,3,5 Taebo Sim,8,9 Nam Doo Kim,10 Matthew J. Berberich,1 Stefan Knapp,6,7,11 Jarrod A. Marto,2,3,5 Kenneth D. Westover,4 Peter K. Sorger,1,* and Nathanael S. Gray2,3,12,*

1Laboratory of Systems Pharmacology, Harvard Medical School, Boston, MA 02115, USA; 2Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA 02115, USA; 3Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02115, USA; 4Departments of Biochemistry and Radiation Oncology, The University of Texas Southwestern Medical Center at Dallas, Dallas, TX 75390, USA; 5Blais Proteomics Center, Dana-Farber Cancer Institute, Boston, MA 02115, USA; 6Institute of Pharmaceutical Chemistry, Goethe-University Frankfurt, 60438 Frankfurt am Main, Germany; 7Buchmann Institute for Life Sciences and Structural Genomics Consortium Goethe-University Frankfurt, 60438 Frankfurt am Main, Germany; 8Chemical Kinomics Research Center, Korea Institute of Science and Technology (KIST), Seoul 02792, Republic of Korea; 9KU-KIST Graduate School of Converging Science and Technology, Korea University, Seoul 02841, Republic of Korea; 10NDBio Therapeutics Inc., Incheon 21984, Republic of Korea; 11German Cancer Network, Frankfurt Site, 60438 Frankfurt am Main, Germany; 12Lead Contact

*Correspondence: peter_sorger [at] (P.K.S.), nathanael_gray [at] (N.S.G.)

Figure 5. Ribbon-Structure Representing Cysteine Positions across the 23 Kinases Covalently Targeted by SM1-71. Through our seven different orthogonal approaches, we confirmed covalent inhibition of 23 kinases across five different spatial positions. The majority of kinases were labeled in the DFG–1 (D1) position followed by the P-loop region (P1-P3). We also identified one kinase labeled in the activation loop region (O1).

Cell Chemical Biology (2019) 26, 818–829. doi:10.1016/j.chembiol.2019.02.021


Deregulation of kinase proteins alters cellular signaling and can result in diverse diseases including cancer. The majority of recently approved kinase inhibitor cancer drugs work by reversibly binding kinases. While covalent kinase inhibitors that irreversibly bind and inhibit these proteins are potentially more effective as therapeutic agents, only a fraction of the kinome has been covalently targeted. Rao et al. describe using a multi-target chemical probe to scan kinases for potential sites of covalent bonding to identify 23 new targets for drug discovery.


Key Findings

  • Our study used SM1-71, a covalent kinase inhibitor with an ortho-positioned acrylamide warhead that allows it to adopt different conformations and interact with cysteines across multiple spatial positions.
  • Using SM1-71, we identified 23 covalent kinase targets across multiple positions, nine of which have not been covalently targeted by an inhibitor.
  • Our study establishes a practical workflow for using multi-target inhibitors and complementary chemoproteomic and cellular assays to scan the kinome for covalently modifiable targets.


    Covalent kinase inhibitors, which typically target cysteine residues, represent an important class of clinically relevant compounds. Approximately 215 kinases are known to have potentially targetable cysteines distributed across 18 spatially distinct locations proximal to the ATP-binding pocket. However, only 40 kinases have been covalently targeted, with certain cysteine sites being the primary focus. To address this disparity, we have developed a strategy that combines the use of a multi-targeted acrylamide-modified inhibitor, SM1-71, with a suite of complementary chemoproteomic and cellular approaches to identify additional targetable cysteines. Using this single multi-targeted compound, we successfully identified 23 kinases that are amenable to covalent inhibition including MKNK2, MAP2K1/2/3/4/6/7, GAK, AAK1, BMP2K, MAP3K7, MAPKAPK5, GSK3A/B, MAPK1/3, SRC, YES1, FGFR1, ZAK (MLTK),MAP3K1, LIMK1, and RSK2. The identification of nine of these kinases previously not targeted by a covalent inhibitor increases the number of targetable kinases and highlights opportunities for covalent kinase inhibitor development.


    Funding Sources

    This work was supported by grants Welch I-1829 (to K.D.W), CPRIT RP140233 (to K.D.W.), P50-GM107618, U54HL127365, U54-CA225088, and KU-KIST Graduate School of Converging Science and Technology Program. S.R. was supported by a Jonathan M. Goldstein and Kaia Miller Goldstein Systems Pharmacology Fellowship.


    Related references

    1. Tan, L., Gurbani, D., Weisberg, E.L., Jones, D.S., Rao, S., Singer, W.D., Bernard, F.M., Mowafy, S., Jenney, A., Du, G., Nonami, A., Griffin, J.D., Lauffenburger, D.A., Westover, K.D., Sorger, P.K., and Gray, N.S. (2017) Studies of TAK1-centered polypharmacology with novel covalent TAK1 inhibitors. Bioorg Med Chem. 25(4):1320-1328. doi:10.1016/j.bmc.2016.11.034 PMID:28038940

    2. Tan, L., Gurbani, D., Weisberg, E.L., Hunter, J.C., Li, L., Jones, D.S., Ficarro, S.B., Mowafy, S., Tam, C.P., Rao, S., Du, G., Griffin, J.D., Sorger, P.K., Marto, J.A., Westover, K.D., and Gray, N.S. (2017) Structure-guided development of covalent TAK1 inhibitors. Bioorg Med Chem. 25(3):838-846. doi:10.1016/j.bmc.2016.11.035 PMID:28011204 PMCID:PMC5484537