Title Collective methodological emission assay of Thioflavin T for qualitative α-synuclein fibril structures discrimination
Authors Szwachta, Grzegorz ; Šulskis, Darius ; Konopka, Aleksandra ; Jalonicka, Ewelina ; Struniawski, Karol ; Mikalauskaitė, Kamilė ; Sakalauskas, Andrius ; Kozera, Ryszard ; Smirnovas, Vytautas ; Stsiapura, Vitali ; Žiaunys, Mantas ; Hanczyc, Piotr
DOI 10.1016/j.ijbiomac.2025.148994
Full Text Download
Is Part of International journal of biological macromolecules.. Amsterdam : Elsevier B.V.. 2026, vol. 335, part 1, art. no. 148994, p. [1-9].. ISSN 0141-8130. eISSN 1879-0003
Abstract [eng] α-Synuclein aggregation into distinct fibrillar structures is increasingly recognized as a key determinant of Parkinson's disease heterogeneity, yet rapid discrimination of these morpho-types remains challenging. Here, we present a multi-parametric emission assay, supported by machine-learning, to qualitatively distinguish three α-synuclein fibril samples, containing different fibril structures, which were generated under controlled stressors (temperature, glass-bead surface area, quaking). Atomic force microscopy (AFM) revealed that type I fibrils display cross-sectional height variations, type II form long, flat protofilaments and type III comprise thin fibrils emerging from 40 to 60 nm aggregate clumps. Cryogenic electron microscopy (cryo-EM) confirmed the morphological diversity by showing that each sample contained both single-stranded and double-stranded fibril structures. Time-resolved emission of Thioflavin T (ThT) embedded into α-synuclein fibrils was analyzed using multiexponential and lifetime distribution models. The latter approach via reconstruction of decay lifetime distributions using maximum entropy method demonstrated possibility of fibrils discrimination based on fluorescence decay of ThT and allowed to link the fibrils morphology to emission decay kinetics of the dye. Fabry-Perot cavity lasing was used to confirm samples types discrimination. Automated image analysis of lasing frames using K-nearest neighbors model showed 88 % accuracy in samples type recognition, particularly powerful in recognition of quenched-states of ThT being at play in type III fibrils. The results were compared to fibrils, which were generated using artificial cerebrospinal fluid - a medium that yields a single secondary structure of aggregate filaments. These complementary methodological analyses represent a sensitive platform for α-synuclein fibril structures discrimination, with application in early diagnostic assay.
Published Amsterdam : Elsevier B.V
Type Journal article
Language English
Publication date 2026
CC license CC license description