Design of tetratricopeptide repeat proteins with novel binding specifications

The design of proteins with new functionalities and desired binding specificities has a wide array of practical applications and also reveals insight into the basis for molecular recognition. Here, we present two related approaches to protein design based on the tetratricopeptide repeat (TPR) protein scaffold, as well as several practical applications of these redesigned proteins. First, we present the rational redesign of a TPR scaffold along with its corresponding peptide ligand. By introducing a minimal number of mutations, we created orthogonal protein---peptide recognition pairs. We show that the binding properties of these recognition pairs can be understood, quantitatively, using straightforward chemical considerations. In a second approach, we designed a strategy to create TPRs with more diverse binding specificities. Here, we created a large naive TPR protein library designed to encode a high degree of diversity so that TPRs could potentially be isolated that bind any protein or peptide sequence of interest. In order to screen this library, we adapted the split-green fluorescent protein (GFP) reassembly assay for screening for protein-protein interactions. Using these two approaches, we have developed a series of TPR proteins that bind several different peptide targets with low micromolar dissociation constants.;The recognition pairs we have developed are also practically useful. We demonstrate the facile replacement of these proteins for antibodies in both detection and purification applications. The new protein---peptide pair has a dissociation constant that is weaker than typical antibody---antigen interactions, yet the recognition pair is highly specific and we have shown that this affinity is sufficient for both Western blotting and affinity purification. Moreover, we demonstrate that this more moderate affinity is actually advantageous for purification applications, because extremely harsh conditions are not required to dissociate the TPR---peptide interaction. The results we present are important, not only because they represent a successful application of protein design but also because they help define the properties that should be sought in other scaffolds that are being developed as antibody replacements. We also extended this work to explore applications for TPR---peptide recognition pairs in developing tools to study and perturb biological processes. We have explored the applications for TPR---peptide recognition pairs in targeting proteins for degradation and for the development of tools for tracking protein localization in live cells.