Mesoscale Architecture of Beta Cells Upon Glucose and Ex-4 Stimulation

Abstract

To understand biology we must consider how molecular machines and organelles within a cell interact and traffic, similar to the importance of understanding traffic patterns in major cities. We use an imaging across scales approach to understand architectural changes that occur in pancreatic β-cells during insulin secretion. β-cells produce and release insulin in response to glucose and are therefore a target for diabetes therapies. Diabetes is a worldwide problem affecting hundreds of millions of people with increasing patient numbers every year. To acquire cellular architectural information, we are integrating data from cryo-electron tomography, soft x-ray tomography (SXT), and fluorescence microscopy. We imaged INS-1E cells (rat pancreatic β-cell line) under multiple glucose and GPCR agonist concentrations to investigate the cellular organization changes induced during insulin vesicle biogenesis and trafficking. The SXT method allows for investigating the 3D cellular architecture and is an ideal method for investigating cell-to-cell variabilities. Additionally, we are using machine learning methods for image analysis and autosegmentation which increased the number of conditions and samples we can analyze rapidly. Furthermore, we can use the information from cryo-ET and SXT along with with live cell imaging to assemble a data-driven, dynamic model of insulin granules within pancreatic β-cells. The integration of data from different time and spatial scales represents a sophisticated convergence of our understanding of cellular structure and function, which will revolutionize biological discovery, open new dimensions of research, and accelerate advancements in healthcare.