With over 500 members in the human genome, protein kinases are one of the largest and most important classes of enzymes in higher organisms. The phospho-transfer reactions that kinases catalyze are vital in the signal transduction pathways that sustain cellular function. The crucial roles of kinases in signal transduction predicate their usefulness as therapeutic and research targets. Many inhibitors of protein kinases bind the ATP portion of the enzyme's active site, even though ATP is a substrate for most kinases. One archetypical example of the problems that arise from this mode of inhibition is the natural product staurosporine. This molecule has been shown to bind numerous kinases with low nano-molar affinity; however, the promiscuity of staurosporine precludes its therapeutic use.

Recently, several groups have addressed the aforementioned specificity problem by designing bisubstrate analogue inhibitors that target both the ATP binding site as well as the peptide binding site of a kinase. With this bisubstrate approach in mind, we have developed a selection strategy to self-assemble two kinase substrate analogues in a non-covalent fashion amenable to in vitro selection against a targeted protein kinase. We have conjugated ATP-analogue-based inhibitors to the leucine zipper domain of Jun, which retains the ability to bind to its native partner, Fos decorated with different peptides to allow for specific interactions with the targeted kinase. Our results demonstrate that our non-covalent bisubstrate inhibitors can abrogate protein kinase function in vitro and provides an entry into the identification of potent and selective kinase inhibitors.