Journal: iScience
Article Title: Diverse intracellular trafficking of insulin analogs by machine learning-based colocalization and diffusion analysis
doi: 10.1016/j.isci.2025.114516
Figure Lengend Snippet: Machine learning framework for live-cell single-particle tracking elucidates variations in intracellular trafficking of insulin analogs (A) A schematic representation of the live-cell assay used to investigate intracellular trafficking patterns. Cells are incubated with labeled insulin (see primary structure in (B) and a fluorescent marker probing an endosomal compartment; early endosomes by Rab5a, late endosomes by Rab7a, endolysosomes, and lysosomes with LysoTracker or transferrin to probe recycling pathways. Temporal 2D molecular movies in two channels are collected to compare insulin behavior with the given endolysosomal or recycling compartment. (B) Primary structure of Atto655-labeled human insulin (Atto655 and insulin are not to scale) highlighting the single point mutation from proline to aspartic acid on position B28, of the fast-acting insulin analog, insulin aspart. (C) Representative spinning disk confocal microscope image of dual-labeled HEK293 cells (Scale bars, 10 μm). Live-cell imaging produces a set of x , y , and t , localizations for each particle, yielding a dataset of single particle trajectories (inset, Scale bars, 0.5 μm). Here, cells are overlaid with the corresponding single particle trajectories of insulin aspart (red, N = 269) and transferrin (blue, N = 2,911). See representative videos of all conditions in , , , , , , , and . (D) Trajectories are fed to DeepSPT, a deep-learning framework, which quantifies diffusional variation between entities, here the two insulin variants (1), and classifies them according to the most similar endolysosomal diffusion (2). LE, late endosomes; LS, lysosomes. (E) Colocalization fingerprinting, a machine learning pipeline for reliable and automatic identification of colocalization based on temporal consistent proximity, and spatial and diffusional features.
Article Snippet: HEK293 cells were obtained from Cell Line Service (CLS, now Cytion, received authenticated at split level 27) and cultured in Dulbecco’s Modified Eagle Medium (DMEM) with 10% Fetal Bovine Serum (FBS) at 37°C with 5% CO 2 .
Techniques: Single-particle Tracking, Incubation, Labeling, Marker, Mutagenesis, Microscopy, Live Cell Imaging, Single Particle, Diffusion-based Assay
cls cell lines service gmbh is a verified supplier 
cls cell lines service gmbh manufactures this product 
