Seminar by Yusuke Himeoka

Marginal states selection with epigenetic modifications for adaptive growth rate regulations

Cells can adapt to a variety of environmental conditions to achieve a higher growth state. This adaptation is achieved by changing the gene expression pattern. Ac- cording to the seminal work by Jacob and Monod[1], such adaptive response is achieved through a signal transduction network that translates environmental condition to the promoter of the genes responsible for adaptation. In spite of the importance of such adaptation mechanism, there remains a question to be addressed: In order to adapt a variety of environmental conditions, cells have to prepare signal transduction networks corresponding to all of them, for which, a huge variety of chemicals and genes would be needed, which may go beyond the capacity of a cell. Also, with such mechanism, cells would not be able to adapt to environmental changes which they have never experienced.
Indeed, recent experiments suggested that cells can adapt even to a novel, unforeseen environment by changing the gene expression pattern. Also, some experiments demonstrated that bacterial cells with artificially gene network without corresponding signal transduction net- works can show adaptive response[2]. In correspondence, a possible theoretical mechanism for the selection of the adaptive state was proposed[3]: When a cell has multiple attractors with different growth-rates, it is shown that that with a higher growth is less perturbed by stochasticity in gene expression dynamics, so that cells tend to be kicked out from low-growth attractors and attracted to those with a higher-growth. However, for this attractor-selection mechanism by noise to work, gene regulation networks (GRN) have to be prepared to have multiple attractors allowing for higher growth rate for a given environmental condition. This will highly limit the applicability of the mechanism. Instead, if the gene expression dynamics have continuous states (line attractors) that show different growth rates, selection of adaptive states can work generally without specific GRN.
Indeed, epigenetic modifications based on several factors, including DNA methylation, histone modification, and their interplay with higher-order chromatin structure are known to play important roles to modify the expression patterns of cells. These modifications can be fixed to various levels, with which gene expressions can be fixed continuously, rather than at a discrete set of values as in attractors.
In this talk, I introduce a simple model which consists of the concentrations of mRNA, Protein, and an epigenetic state variable, and demonstrate that states with higher growth-rates are generally selected by noise, using marginal stability. Without preparing specific gene expression networks with multiple attractors, adaptation to achieve higher growth for given environmental condition is resulted. We will provide general concept of selection from marginal stable states in stochastic processes, and discuss relevance to cellular biology.

[1]  Francois Jacob and Jacques Monod. Genetic regulatory mechanisms in the synthesis of proteins. Journal of molecular biology, 3(3):318–356, 1961. 

[2]  Akiko Kashiwagi, Itaru Urabe, Kunihiko Kaneko, and 
Tetsuya Yomo. Adaptive response of a gene network to environmental changes by fitness-induced attractor selec- tion. PloS one, 1(1):e49, 2006. 

[3]  Chikara Furusawa and Kunihiko Kaneko. A generic 
mechanism for adaptive growth rate regulation. PLoS Comput Biol, 4(1):e3, 2008.