Immunocytochemical and morphometric analysis of hiPSC-derived neurospheres and their conglomerates in Parkinson and Alzheimer disease
Anna A. Kopylova1, Roman V. Shumilin1, Daria D. Volegova1, Elizaveta S. Pereplitsa2, Anna V. Blagova2, Ksenia O. Salina1, Ilya K. Lanikin1, Petr I. Kupriyanov1, Anna V. Zubova2, Marina R. Kapkaeva2, Ivan V. Simkin1, Alla B. Salmina1,2, Stanislav O. Yurchenko1 and Sergei N. Illarioshkin2
1 Laboratory of Cellular Technologies and Tissue Engineering, Center “Soft Matter and Physics of Fluids,” Bauman Moscow State Technical University, Moscow, Russia
2 Laboratory of Neurobiology and Tissue Engineering, Brain Science Institute, Research Center of Neurology, Moscow, Russia
Abstract
Three-dimensional (3D) cultures such as neurospheres derived from human induced pluripotent stem cells (hiPSCs) represent a powerful tool for studying intercellular interactions, mechanisms of neurodegeneration and disease modeling. The aim of this study was to perform a comparative morphometric and immunocytochemical analysis of neurospheres and their conglomerates derived from hiPSCs of patients with Alzheimer disease (AD), Parkinson disease (PD), and healthy donors. hiPSCs were generated from periferal blood leukocytes, reprogrammed and differentiated into neural progenitors, followed by the formation of 3D cultures — neurospheres.
Immunocytochemical staining included proliferation, neuronal differentiation, inflammation markers, cell junctions and mitochondrial dynamics protein. The profiling demonstrated increased mitochondrial markers and decreased proliferation and neuronal differentiation in pathological groups compared to controls.
16 morphometric parameters were evaluated using a custom segmentation algorithm. Statistical analysis revealed group-specific differences: PD and AD-derived neurospheres showed up as more circular, but with more irregular border. AD neurosphere conglomerates were generally smaller, which correlated with decreased proliferation showed by immunocytochemistry study. To further classify neurosphere conglomerates, and to distinguish between the norm and neurodegenaration, we trained a CatBoost model with preliminary selection of 11 more statistically relevant features. Using 80% of the data with 5-fold cross-validation and 20% as a test set, the binary classification of controls versus AD or PD reached Accuracy = 0.833.
These findings suggest that morphometric and immunocytochemical profiling, complemented by machine learning classification, provides a sensitive platform for detecting disease-specific alterations in hiPSC-derived 3D neural cultures and supports their use in neurodegeneration research.
Speaker
Anna Kopylova
Laboratory of Cellular Technologies and Tissue Engineering, Center “Soft Matter and Physics of Fluids,” Bauman Moscow State Technical University, Moscow, Russia
Russia
Discussion
Ask question
