Seminars

May 21 2018

11:00 158 HH

This seminar is sponsored by IBP and MCDB

Paul Bressloff

Self-Organizing Processes in Cell Biology

Summary

Paul Bressloff
Department of Mathematics
University of Utah

Sponsored by the Department of Integrative Biology and Physiology, the Department of Molecular, Cell, and Developmental Biology and the Institute for Quantitative and Computational Biosciences

There are two classical diffusion-based mechanisms of self-organization in cellular and developmental biology. The first involves the formation of a protein concentration gradient, generated by some localized source of protein production. This is thought to play a crucial role in cell differentiation during embryogenesis, for example. The second involves spontaneous Turing pattern formation, where two or more nonlinearly interacting chemical species differing significantly in their rates of diffusion amplify spatially periodic fluctuations in their concentrations, resulting in the formation of a stable periodic pattern. A common way to generate a Turing instability is to have an antagonistic pair of molecular species, a slowly diffusing chemical activator and a quickly diffusing chemical inhibitor. One of the challenges in applying this mechanism to real applications is how to obtain chemical species with significantly different diffusion rates.

In this talk, we present some recent theoretical studies of self-organization in cellular and developmental biology that go beyond the classical diffusion-based mechanisms. We begin by considering an alternative mechanism for the formation of morphogen gradients, based on active vesicular transport along thin, actin-rich filaments known as cytonemes; these provide direct links between source and target cells. We also discuss the important issue of robustness - how organisms generate reliable phenotype from unreliable genotype. In the remainder of the talk, we describe three different mechanisms for spontaneous pattern formation, each of which addresses the general problem of diffusion on multiple scales. First, we consider a model of synaptogenesis in C. elegans, which identifies a homeostatic mechanism for maintaining synaptic density along the ventral cord of C. elegans during larval development. In this case, the inhibitor undergoes fast bidirectional motor transport rather than fast diffusion. Second, we consider a model of cell polarization in fission yeast, which investigates the role of diffusion in the length-dependent switch from monopolar to bipolar growth following cell division (symmetry-unbreaking). We highlight the fact that intracellular pattern formation does not typically involve the classical activator-inhibitor mechanism, being based instead on the mass-conserving redistribution of proteins by cytosolic diffusion and the membrane-dependent cycling of proteins between different conformational states. Finally, we illustrate how a Turing-like mechanism explains the origin of geometric visual hallucinations, where nonlocal synaptic interactions in the visual cortex drive pattern forming instabilities rather than diffusion.

Monday, May 21, 2018
11:00am
Hershey Hall Grand Salon, Room 158



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



 



this is idtest: