Dynamics of signal transduction and gene expression in single cells

Alexander van Oudenaarden (MIT)

Abstract

In this talk I will address three topics:

Epigenetics Inheritance. The partitioning and subsequent inheritance of cellular factors like as proteins and RNAs is a ubiquitous feature of cell division. However, direct quantitative measures of how such non-genetic inheritance affects subsequent changes in gene expression have been lacking. We tracked families of the yeast Saccharomyces cerevisiae as they switch between two semi-stable epigenetic states. We found that long after two cells have divided they continued to switch in a synchronized manner, whereas individual cells have exponentially distributed switching times. By comparing these results to a Poisson process, we show that the time-evolution of an epigenetic state depends initially on inherited factors, with stochastic processes requiring several generations to decorrelate closely related cells. Finally, a simple stochastic model demonstrates that a single fluctuating regulatory protein can explain the bulk of our results.

Phenotypic Bet-hedging. A classic problem in evolutionary and population biology is to understand how a population optimizes its fitness in fluctuating environments. Rather than maintaining a phenotypically homogenous population, it has been suggested that a 'bet-hedging' strategy might be more beneficial in uncertain environments. Following this strategy, a population consisting of a variety of phenotypes enhances its fitness by ensuring that, at any given time, some of its members are prepared for an unforeseen environmental fluctuation. We experimentally test this hypothesis in vivo using a re-engineered yeast strain that randomly transitions between two phenotypes. Each phenotype is designed to confer a growth advantage over the other phenotype in a certain environment. We show that, in order to optimize population growth, cells have to match their inter-phenotype switching rate to the frequency of environmental changes. Our data suggest that when transition rates are correctly tuned, random bet-hedging constitutes a simple, yet effective, survival strategy to cope with fluctuating environments without the need to actively sense environmental conditions.

Network Identification. Few biological networks have sufficiently well characterized topology and dynamics to build accurate quantitative models. Here we present a general method to probe and model biological networks that lessens the reliance on prior knowledge of the network and provides insights into network topology, dynamics, and function. We use periodic stimuli to measure the frequency dependence of signal transduction in the osmoregulation pathway of Saccharomyces cerevisiae. We apply system identification methods from control engineering to infer a concise dynamic model, which is consistent with the established network topology and correctly describes the intricate network dynamics. Furthermore, the model correctly predicts the behavior of a mutant strain with perturbed feedback regulation. These results suggest that applying control theory principles to biological systems can facilitate an increased functional understanding of complex biochemical networks.

Back to The First q-bio Conference.