CAREER: A multi-scale, data-driven model of 3D cell motility

University Of California-San Diego, La Jolla CA

Investigators

Abstract

CAREER: A multi-scale, data-driven model of 3D cell motility This project aims to study how cells migrate in spaces similar to the body by developing new techniques to measure cell movements in three-dimensions (3D). This research will generate understanding of physiologically relevant cell migration behaviors, which are responsible for orchestrating the formation of multicellular organisms. Because cells interact directly and constantly with their surroundings to create movement, migration is context dependent. However, cell motility has historically been studied in two-dimensional (2D) environments. The Principal Investigator (PI) discovered that cells display different migration behaviors in 3D environments that are more physiological. This project will address the fundamental question: How do 3D environmental cues give rise to distinct migration behaviors? The PI will employ a new process for creating 3D environments, a new integrated microscopic imaging technique and advanced computational biology approaches to develop a predictive, quantitative model of how cells move in three dimensions through a synthetic extracellular matrix (ECM). Such a model has the potential to advance our basic understanding of cell motility, which is a critical aspect of eukaryotic development. The project will additionally provide research training opportunities for graduate students and experience-based educational projects on multi-scale systems analysis and career education for high school and undergraduate students. In 3D spaces like bodily tissues, it is generally accepted that the coordination of four local processes (adhesion, cytoskeletal polymerization, contractility, and matrix remodeling) dictate global motility behaviors. Yet, no quantitative framework has been developed to link these localized processes to global cell behaviors. Such a framework would represent a generalized platform for both studying and engineering cell motility, where any signal input (chemical exciter/inhibitor, physical cue, etc.) could be quantified by its effect on these processes to predict the resulting motility behavior output. Historically, motility models have been formulated using a bottom-up approach, where theoretical physical and kinetic equations are fitted to observations. These approaches have advanced the understanding of local processes of motility on 2D planar surfaces. However, mathematical, computational, and experimental limitations have prevented their full extension to physiologically relevant 3D cell motility. The top-down approach used in this project overcomes current limitations to immediately address a fundamental question: How can subcellular molecular processes be connected with the geometry and translocation of the whole cell in a 3D ECM? The goal of this project is to integrate knowledge across three scales of cell motility, extracellular, sub-cellular, and cellular, by taking a data-driven systems approach. In Aim 1, the PI will develop a microfluidic device that will enable highly controlled, systematic variation of ECM matrix properties and high-throughput analysis of their regulatory role in cell migration. Aim 2 will combine into a single platform a time-lapse cell tracking assay and quantitative molecular imaging techniques, which the PI previously developed for measuring the four core motility processes within cells in a 3D ECM. This platform will allow the analysis of single-cell motility at an integrated systems level for the first time. In Aim 3, the PI will apply machine learning and statistical modeling techniques to the single-cell data generated in Aim 2 to develop a predictive framework for emergent 3D cell motility.