Scripps Research Logo

Development of an Inverse Drug Discovery Platform

Drug candidates are generally discovered using biochemical screens employing an isolated target protein or by utilizing cell-based phenotypic assays. Both non-covalent and covalent hits emerge from such endeavors. Target identification is often very challenging in phenotypic screens, and substantial medicinal chemistry efforts are generally necessary to convert molecules coming from biochemical screens into molecules having the desired membrane permeability, selectivity and potency in a human. As an alternate approach to drug discovery, the Kelly lab together with the Sharpless lab are developing an “Inverse Drug Discovery” strategy in which organic compounds of intermediate complexity harboring weak, but activatable, electrophiles are matched with the protein(s) they react with in cells or cell lysate. An alkyne substructure in each candidate small molecule enables affinity chromatography–mass spectrometry, which produces a list of proteins that each distinct compound reacts with. An important feature of this approach is that it is agnostic with respect to the cellular proteins targeted. Notably, the “Inverse Drug Discovery” approach has features that distinguish it from activity-based protein profiling efforts, which typically employ more-reactive covalent probes that are directed toward a specific class of enzymes or a large number of proteins containing a reactive amino acid side chain like that of Cys.

We are exploring several different activatable electrophiles to exemplify this approach, including arylfluorosulfates (Ar-OSO2F), an underexplored class of sulfur(VI) halides. Arylfluorosulfates have not been utilized much because of difficulty in synthesis and a perceived promiscuous reactivity. The Sharpless Lab has discovered a facile approach to make to arylfluorosulfates from parent phenols using sulfuryl fluoride (SO2F2) gas. Arylfluorosulfates are essentially inert in complex biological media at neutral pH, but can react rapidly and selectively with specific proteins if the binding site provides the correct juxtaposition of protein side chain functional groups and transition state stabilization of the sulfur(VI) fluoride exchange (SuFEx) reaction. Stabilization of the departing fluoride ion is assumed to be critical to facilitating the SuFEx reaction, and may be achieved through interactions with hydrogen bond donors within a protein binding pocket, through the influence of hydrophobic-aqueous interfaces or electric fields found within proteins, or through a combination of these factors. Arylfluorosulfates activated by protein binding undergo chemoselective reactions with proximal Tyr and Lys side chains, and possibly others, to produce a singly-labeled adduct.

In the proteins that we have characterized thus far, the reaction between the recombinant purified protein and arylfluorosulfate appears to occur in a well-defined binding pocket. Typically, these are the same sites used to bind endogenous ligands or cofactors (e.g., ATP, heme, retinoic acid, thyroxine). The reactive sites also tend to contain multiple side chains that are primarily cationic at neutral pH (i.e., those of Arg and Lys). Conjugation with arylfluorosulfates depends on these cationic side chains, which likely lower the pKa of the fluorosulfate-reactive nucleophile (Tyr or Lys side chains) and/or catalyze the SuFEx reaction. In some cases pKa perturbation is not sufficient to achieve reactivity, suggesting a role for the cationic residues in lowering the free energy of activation of the conjugation reaction, perhaps by binding to the departing fluoride ion.

To date, we have synthesized structurally unique arylfluorosulfate probes that target specific members of the intracellular lipid-binding protein family, glutathione S-transferases, members of the AlkBH family that possess RNA demethylase activity, nucleoside diphosphate kinases, heme oxygenase 2, and other proteins. We are also exploring other sulfur(VI)-containing functional groups.

We are also studying additional latent electrophiles not typically studied in a cellular context using our affinity purification-mass spectrometry methodology to further exemplify our “Inverse Drug Discovery” approach. A key goal is to show that very selective covalent antagonists of an important protein target can be made using a medicinal chemistry strategy.