Classifying Biophysical Subpopulations of Insulin Secretory Granules using Quantitative Whole Cell Structure Analysis

Abstract

Pancreatic {beta}-cells contain insulin secretory granules (ISGs), organelles where proinsulin is converted into insulin. As ISGs mature, they undergo extensive biophysical remodeling, producing a spectrum of subpopulations with heterogeneous molecular and spatial characteristics. However, systematic methods to define ISG subpopulations remain underdeveloped. To address this gap in knowledge, we employed soft x-ray tomography (SXT), which can quantitatively measure the biochemical density of ISGs within whole {beta}-cells. Using unsupervised clustering, we classified subpopulations based on molecular density, size, and spatial positioning. Across different insulin secretory stimuli, we observed shifts towards mature and releasable subtypes, demonstrating that exogenous signals can dynamically remodel ISG subpopulation distributions. We extended this methodology to primary {beta}-cells characterized using volume electron microscopy (vEM). Integrating subpopulations from SXT and vEM uncovered insights inaccessible by a single method in isolation. This strategy establishes a framework for defining therapeutic approaches aimed at enriching physiologically beneficial ISG subpopulations.