How walking slower can make you respond faster

Researchers from the FOM institute AMOLF and the Riken institute in Japan have developed a new algorithm that makes it possible to simulate signal transmission in living cells at the molecular level.

A recent application, reported in a recent issue of the Proceedings of the National Academy of Sciences USA, reveals that diffusion of protein molecules can drastically affect the response of living cells to chemical stimuli.

Living cells continually have to respond and adapt to a changing environment. These changes are detected and processed by protein molecules that chemically and physically interact with one another. These so-called biochemical networks are the information processing devices of life. They allow the cell to perform a large number of computational tasks, analogous to electronic circuits. Yet, while biochemical networks and electronic circuits can perform similar computations, their design principles are markedly different. One key difference is that in a biochemical network, the components, the protein molecules, move by diffusion. In most models of biochemical networks this diffusion was, however, not taken into account, because no numerical techniques existed that allowed the efficient simulation of biochemical networks in time and space.

The researchers developed a numerical technique using a well-known approach in statistical physics, namely that of Green’s functions. These Green’s functions make it possible to analytically compute the propagation of single particles and pairs of particles, allowing for large jumps in time and space in the simulations. Using this trick, particle-based simulations of biochemical networks on biologically relevant length and time scales are now within reach.

The researchers applied their new algorithm to one of the best-studied biochemical networks, namely the Mitogen-Activated-Protein-Kinase (MAPK) system. In this system an enzyme has to activate another protein, the substrate, by chemically modifying it at two sites. The simulations revealed that an enzyme molecule that has just modified a substrate molecule at the first site, can rapidly rebind that substrate molecule to modify the second site, before another enzyme molecule binds it. These enzyme-substrate rebindings may seem innocent, but have major functional consequences: they weaken the sharpness of the response, and they can lead to the loss of bistability – the capacity of a biochemical network to switch between two macroscopic states. Moreover, these enzyme-substrate rebindings can strongly speed up the response; since the probability of a rebinding event increases as the diffusion constant decreases, slower diffusion can, counter-intuitively, lead to a faster response.

Since multi-site protein modification is omnipresent in biochemical networks, the researchers believe that their results are also relevant for understanding signal transmission in other biological systems.

Reference: Spatio-temporal correlations can drastically change the response of a MAPK pathway, K. Takahashi, S. Tanase-Nicola, P. R. Ten Wolde, Proc. Natl. Acad. Sci. USA, Early Online 25 January 2010.

Contact: Pieter Rein ten Wolde, T: +31-20-754 7281

Figure: Snapshot of the simulations. The new algorithm makes it possible to efficiently simulate biochemical networks at the particle level in time and space.