De Novo Design of Buttressed Loops for Diversifying Protein Functions

Abstract

Loops play essential roles in protein functions such as protein-protein interactions, ligand recognition and metal binding. However, designing structured and functional loops remains to be a long-standing challenge for protein engineering. In contrast to alpha helices and beta sheets, loops are substantially more disorganized and structurally less defined. Such intrinsic flexibility makes it difficult to predict or design the loop structures. One feasible strategy for designing loops is to stabilize them by creating extensive interactions between loops and the rest of the protein. In particular, loop buttressing, in which loops are interacting and stabilizing each other, is a trick that has been successfully used by nature. We aim to design and test buttressed loops that adopt rigid conformations and can be further engineered for functions. Inspired by the fold of ankyrin repeat proteins, we developed a computational design protocol that combines canonical secondary structural motifs into structured loops with kinematic closure algorithms to diversify the fold of designed helical repeat proteins. The designed loop conformations were confirmed by both the predictions of AlphaFold and the crystal structures. The scaffold proteins, installed with buttressed loops, were then functionalized by designing new interfaces for mediating protein-protein and protein-peptide interactions. To demonstrate the generalizability of our method, we further applied the loop designing protocol to diversify a de novo TIM barrel. These designs demonstrated that the novel binding grooves or pockets formed between the buttressed loops and the original protein scaffolds provide a unique category of protein interfaces for engineering new functions.