With the increasing strength of computational resources, computational drug design methodologies have been used as tools in the identification of lead molecules for drug targets. In my thesis, I have utilized structure-based drug design methodology to elucidate information targeting the mammalian transcription factor family of Notch.
As a crystal structure solely exists for the mammalian Notch1 transactivation factor complex (PDB: 2F8X, resolution: 3.25ÌÄ"_), I initially created homology models for the Notch2-Notch4 transcription factor complexes based on the three-dimensional coordinates for each atom of the crystal structure of Notch1. Of interest to our collaborators at the Dana-Farber Cancer Institute, a mutant of the Notch3 complex was tested to address a clinical phenotype seen. Also, weak protein-protein interactions in the Notch1 interface were found to allow for the dissociation of ANK and CSL once MAML1 is removed from the complex.
Stapled SAHM peptide analogs were designed to disrupt the protein-protein interactions of the Notch1-Notch4 transactivation complexes, the loss-of-function or gain-of-function mutations of which are related to numerous diseased states, including cancers. Stapled SAHM mutants disrupting analogous regions of interface homologs resulted in quick molecular dynamics convergence. Stapled SAHM mutants were designed by natural amino acid point mutations to allow for higher binding affinity to non-analogous ANK/CSL interfaces. Novel SAHM1 peptide antagonists were developed through varying the location of the staple and docking to the Notch1 interface; one stapled peptide shows quick MD convergence, analogous to that of SAHM1. The Notch2-Notch4 ANK/CSL interfaces were also targeted at two hot-spot regions by small molecule ligands, which will be subsequently tested computationally by MM/PBSA rescoring and validated by experimental measures by our collaborators.
The quick and cost-efficient computational techniques that help elucidate the mechanisms associated with binding and have been applied include: homology modeling, equilibration and molecular dynamics simulations, and MM/PBSA binding free energy calculations and rescoring.