Computational design of antagonist peptides based on the structure of secreted frizzled-related protein-1 (SFRP1) aiming to inhibit Wnt signaling pathway

Abstract

Aberrant activation of Wnt/β-catenin signaling pathway, due to the genetic or epigenetic changes, is responsible for tumorigenesis in epithelial cells of different types of cancer such as colorectal cancer. Secreted Frizzled-Related Protein-1 (SFRP1), as one of the antagonist proteins of this pathway, is hyper-methylated in colorectal cancer leading to the formation of Wnt-Fz-LRP and activation of Wnt/β-catenin signaling pathway. We aimed to design antagonist peptides based on SFRP1 structure against wingless-type 2 (Wnt2), a highly expressed ligand in different cancers like colorectal cancer, to inhibit the formation of the initial triple complex of Wnt-Fz-LRP. After homology modeling of SFRP1, molecular docking showed that Wnt2 and SFRP1 interact in the same mode of xWnt8-mFz8 and hWnt3-mFz8 through the thumb and finger binding sites. These binding sites were selected for designing peptides using either substitution or deep learning-based approaches. The efficiency of each designed peptide in interacting with Wnt2 was evaluated by molecular docking. Stability assessment of Wnt2-peptide complexes via molecular dynamic (MD) revealed that the designed peptides could effectively interact with Wnt2 binding sites during the simulation. However, the designed peptides against the thumb site had higher binding affinity and hydrogen bonds compared to the initial sequence. The secondary structure of the designed peptides indicated an alpha-helix structure which is a favorable structure for peptide drugs. Computing the physicochemical properties of peptides predicted a fairly acceptable structure which made them promising candidates in the treatment of cancers like CRC.