Combination of liquid crystal and deep learning reveals distinct signatures of Parkinson's disease‐related wild‐type α‐synuclein and six pathogenic mutants. Issue 2 (21st December 2021)

• Record Type: Journal Article
• Authors: Yang, Xiuxiu; Zhao, Xiaofang; Zhao, Hansen; Liu, Fengwei; Zhang, Sichun; Zhang, Claire Xi; Yang, Zhongqiang
• Is Part Of: Chemistry, an Asian journal. Volume 17:Issue 2(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-21
• Subjects: Liquid crystal -- α-synuclein -- deep learning -- protein-lipid interaction -- interfaces; Chemistry -- Periodicals; 540.5
• DOI: 10.1002/asia.202101251
• Languages: English
• ISSNs: 1861-4728

Abstract

Abstract: α‐Synuclein is a central player in Parkinson's disease (PD) pathology. Various point mutations in α‐synuclein have been identified to alter the protein‐phospholipid binding behavior and cause PD. Therefore, exploration of α‐synuclein‐phospholipid interaction is important for understanding the PD pathogenesis and helping the early diagnosis of PD. Herein, a phospholipid‐decorated liquid crystal (LC)‐aqueous interface is constructed to investigate the binding between α‐synucleins (wild‐type and six familial mutant A30P, E46K, H50Q, G51D, A53E and A53T) and phospholipid. The application of deep learning analyzes and reveals distinct LC signatures generated by the binding of α‐synuclein and phospholipid. This system allows for the identification of single point mutant α‐synucleins with an average accuracy of 98.3±1.3% in a fast and efficient manner. We propose that this analytical methodology provides a new platform to understand α‐synuclein‐lipid interactions, and can be potentially developed for easy identification of α‐synuclein mutations in common clinic. Abstract : A phospholipid decorated, stimuli‐responsive liquid crystal interface is constructed to investigate the α‐synuclein‐phospholipid interaction. Wild‐type and six familial mutant (A30P, E46K, H50Q, G51D, A53E and A53T) α‐synucleins exhibit distinct signatures at the phospholipid‐decorated liquid crystal‐aqueous interface, which could be further distinguished by deep learning with a 98.3±1.3% accuracy.