Stochastic Dynamics and Noise Control in Patterning Systems

点击翻译按钮获取中文摘要

DESCRIPTION: The formation of signaling gradients within tissues is a fundamental aspect of animal development. Morphogens specify cells with different identities, causing them to follow distinct developmental programs depending on signal concentration. One of the major challenges for morphogens is to deal with stochastic effects at various levels, such as spatial and temporal fluctuations of the signal in extracellular space or noise in signal transduction. How does a patterning system overcome stochastic fluctuations? How do such stochastic effects at different levels interact? Can noise actually be beneficial? What regulatory strategies create sharp boundaries of gene expression? How can a boundary be placed in the correct spatial location within a reasonable time window while it is sharpening? Through a combination of modeling and experimental studies, this work addresses these fundamentally important questions in the context of formation of segments (rhombomeres) in the zebrafish hindbrain. In the developing vertebrate central nervous system, the vitamin A derivative retinoic acid (RA) patterns multiple segments that underlie the eventual patterns of neurogenesis and defects in its signaling cause disease. The unique aspects of RA signaling in zebrafish make it an intriguing vertebrate model system for studying stochastic effects for developmental patterning. The long-term goal of the proposed research is to understand generic mechanisms of stochastic dynamics in tissue patterning. One of the central hypotheses is that noise in a morphogen system does not necessarily increase uncertainty in patterning and can actually be utilized to improve boundary sharpening. Other mechanisms such as cell sorting and additional morphogens, can be critically important in placing the boundary at the correct location within a short time period. The stochastic dynamics of RA gradient and its downstream responses in vivo will be quantified through Fluorescence Lifetime Imaging Microscopy (FLIM) combined with fluorescent RA responsive transgenics. In particular, noise will be perturbed and measured in vivo, and stochastic interactions will be investigated using novel multiscale modeling frameworks. Three key properties of gene expression boundaries will be scrutinized: sharpness, accuracy, and variability. New relationships among these three properties and the tradeoffs between them, as well as general principles for controlling stochastic fluctuations in developmental patterning will be obtained. These will lead to a better understanding of stochastic effects during embryonic development and the causes of birth defects.