Computational modeling and molecular mapping of serine protease inhibitor family A5 (SERPINA5) structure, associated with tau expression and Alzheimer's disease.

Abstract

Serine Protease Inhibitor Family AS (SERPINA5), also known as PCI (Protein C Inhibitor) or PAl3 (plasminogen activator inhibitor 3), is a member of the serine proteinase inhibitor family. Our previous studies have shown that increased expression of the SERPINA5 gene is associated with tau expression in the brain and Alzheimer's disease. We hypothesize selective targeting of SERPINA5 has potential as a therapy for Alzheimer's disease. Since the full-length 3D human SERPINA5 structure is still unknown, we performed computational modeling in silico using Schrodinger Suite 2022-2 (Prime software), AlphaFold 2 and OpenFold (trainable PyTorch reproduction of DeepMind's AlphaFold). Molecular mapping was used to identify binding hot spots, which are regions of the protein surface that can bind small molecules or drug fragments and was calculated by ATLAS (FTMap) software. The best structures with different conformational states were selected and molecular mapping was performed for each of them. The druggability hot spots were determined, then ranked according to their ability to bind small molecules and peptides on the SERPINA5 protein surface. These results will be used in molecular docking, molecular dynamics simulations and biological testing research.