The facts & the project
Interactions between proteins and the famous double helix of DNA are at the heart of cellular function. Proteins are responsible for compacting DNA into chromosomes, for repairing damaged DNA, for copying DNA during cell division, and for driving its transcription into RNA. Crucially, proteins also control which genes will be active at any given moment in any given cell. How proteins recognize their binding sites on DNA, amongst millions or billions of potential sites along the genome, is thus an important question.
A first answer to this question has come from understanding that the DNA base sequence (the arrangement of the four bases: A, T, G and C) changes the structural and mechanical properties of the double helix.
“Molecular simulations have significantly contributed to this discovery by creating the only available comprehensive database of sequence effects on the structure and dynamics of the double helix”, Richard Lavery, director of research at CNRS (Institut de biologie et chimie des protéines in Lyon), said. Carried out in the context of an international consortium of laboratories (the Ascona B-DNA Consortium), their work has led to a reliable prediction of the behavior of a given fragment of DNA.
Simulations have also led to another unexpected answer: “Protein-DNA interfaces turn out to be dynamic and, at least for some proteins, experimentally observed binding sequences are the time-averaged result of multiple bound conformations, rather than of a single conformation”. This finding helps to explain how proteins can quickly locate their ideal binding sites.
“These simulations, which are the new target of the ABC laboratories, require supercomputer resources since they can involve up to 500,00 atoms and must cover microsecond-scale trajectories”, he added.