Our work centers on two nascent technologies that not only borrow from the structure of nature’s bioactive molecules but also from nature's approaches to synthesize molecules and assess them for function. The application of these technologies is focused on advances in medicinal chemistry.

DNA-Encoded Chemical Libraries
In the first area, we build synthetic small molecule libraries that are covalently linked encoding DNA scaffolds. Within the past few years, DNA-encoded chemical libraries, DELs, have been adopted at every major pharmaceutical company for hit generation and lead discovery. Several successful cases have been reported yielding multiple clinical candidates. The DEL approach has the ability to generate and assay small molecule libraries of much greater complexity, present demonstrations up to 1010 molecules, than by any other approach.

Rather than screening molecules in discrete assays, DELs can be assessed collectivly for binding to a protein target by a selection assay, all together, in a single tube. The cost, effort, and infrastructure required are reduced dramatically, particularly in comparison to traditional high-throughput screening, HTS. This approach enables medicinal chemistry to capitalize on many of the amazing capabilities of molecular biology, such has high detection sensitivity via DNA amplification, PCR, and massively parallel analysis of DNAs in next-generation DNA sequencing.

Our work in this area involves development of novel libraries and selection strategies for discovery of protein-protein interaction inhibitors that are mediated by short linear peptide motifs, with a particular focus on protein kinases and chromodomains. We are using these newly developed approaches to develop therapeutic lead molecules for application in treating cancer, Alzheimer's disease, and malaria.

DNA-Linked Probes for Protein Assays Conducted via DNA Analysis
In the second area, we have developed an evolution-inspired assay approach to detect sample stimuli using DNA-linked molecules as activity probes. We have developed DNA-encoded probes for detection of several enzymatic activities, as well as synthetic ligand binding, which allow activity detection through DNA sequence analysis. We work to apply this approach in proteomic activity profiling and in screening of small molecules for enzyme inhibition. Developments in genetic analysis technologies, particularly DNA sequencing, have been transformative to biomedical research.

In contrast to genomic information, the barrier to accessing proteomic information, particularly enzyme activity, is dramatically higher. As aberrant enzyme activities are consistently observed in disease, this information is critical for appropriate diagnosis and treatment. As with DNA-encoded libraries, these approaches allow protein assays to harness the power of DNA analysis technologies. Current efforts are focused on developing robust assays for protein targets directly in cell lysates for small molecule screening and in using libraries of protein kinase substrates to understand resistance mechanisms in drug-resistant breast cancer, in collaboration with Dr. Michael Wendt.

We present the use of DNA-encoded libraries of peptides as an approach for discovery and optimization of protein ligands and substrates. We highlight the benefits of a DNA-based approach over traditional peptide library approaches such as cellulose spot and micro arrays.

We have synthesized both combinatorial and scanning positional DNA-encoded libraries of trimethyllysine-containing peptides and peptidomimetics for development of selective inhibitors of the CBX family of chromodomains, ChDs. Additionally, we use collections of DNA-encoded peptide ligands to CBX ChDs as a model system to demonstrate the capabilities of selection assays of DNA- encoded peptides.

Statistical analysis of selections assays indicates low DNA tag bias and adequate robustness for both ligand discovery and determination of quantitative structure activity relationships. Further, we show how the application of photoaffinity labeling with DNA- linked peptides enables both massively parallel determination of affinity constants of peptides to the CBX ChDs by DNA sequencing and also evaluation of binding within live mammalian cells.

In addition to protein binding, we demonstrate selection approaches for selective enrichment of peptide substrates of protein kinases, proteases, and farnesyltransferases. We concentrate specifically on protein tyrosine kinases and investigate the potential of the approach both for new substrate generation as well as the detection of activities in biological samples.