Understanding physical constraints on cell biology through multiscale modeling

点击翻译按钮获取中文摘要

Research over the past several decades has made clear that spatial restriction and heterogeneity within cells is an integral components of biochemistry. Understanding and controlling biochemical processes necessarily requires understanding the physical constraints in which they take place. However, the crowded and confined cellular environment makes it difficult to translate measurements made outside of cells into a mechanistic understanding within cells. Similarly, measurements made directly in cells can be difficult to interpret due to the complexity of the cellular interior. Computational and mathematical modeling can meet this challenge by offering a bottom-up approach for connecting experimental observations to underlying physical processes. This program seeks to develop and apply computational models to an archetypal example of physically-constrained biochemistry: protein folding homeostasis in the endoplasmic reticulum (ER), provide valuable insight into how physical constraints within the ER affect its function. The ER is an intracellular organelle that consists of a dynamic network of narrow sheet-like and tubular structures. It is responsible for a broad range of functions including the synthesis, folding and transport of many proteins. Environmental perturbations can lead to a disruption of ER function, called ER stress, which is marked by misbalance between nascent peptides waiting to fold and the chaperones and enzymes needed to fold them. When this occurs, the ER initiates a broad transcriptional program called the unfolded protein response (UPR) to restore homeostasis. Failure of the UPR to mitigate ER stress leads to cell death and is implicated in many diseases such as diabetes, Alzheimer's disease and cancer. The UPR is controlled by three proteins in the ER membrane that detect ER stress through interactions with ER-resident proteins, and transmit stress signals to the nucleus. The objective of this program is to understand how these sensory proteins integrate various physico-chemical features of the ER into a coherent measure of ER stress. To address this objective, we will develop computational models at scales ranging from the atomistic description of proteins to organelle-scale modeling of ER morphology. These models will provide quantitative predictions of the how the ER shape, composition and contents affect the ER stress response, allowing observations made downstream of the ER to be translated into difficult-to-probe mechanisms controlling the stress response within the ER. Furthermore, the tools and methods developed for connecting ER morphology and crowding to stress sensor activity will be widely applicable to processes in other subcellular structures. Lastly, since this research program sits between computational science and cell biology, trainees will be equipped with highly sought-after interdisciplinary skillsets, positioning them at the forefront of computational biology.