Fabien Cailliez Webpage

Mechanical properties of proteins

For quite a long time it was believed that the knowledge of the structure of proteins would allow an almost complete understanding of their functions. But with the great development of molecular modeling since the 80's and the development of new experimental techniques such as single molecule steering experiments, dynamical and mechanical properties have been shown to have a key role for the function of many biomolecules.

During my Master and my PhD, undergone under the supervision of Dr Richard Lavery at the Laboratoire de Biochimie Théorique of the IBPC (Paris), my research work was focused on the mechanical properties of proteins. My work has been shared between 3 different projects :

  1. The main part of my work was to study E-cadherin molecules. These extracellular proteins are involved in adhesion process between cells and their dysfunction has been observed in many cancers. Their mechanical properties are thought to be of great importance for their function. We used molecular dynamics simulations to study the role of calcium ions in the preservation of their structure. We also compared three observed dimeric forms of terminal segments of E-cadherins to elucidate the role they may play in adhesion phenomena. You can find details of this work in my papers or in my thesis manuscript. You can also have a look at this beautiful website on C-cadherins, which are very similar proteins, that exhibits similar mechanical properties.
  2. A more methodological part of my work was to build a new measure of flexibility at the scale of the residues, but taking into account the structure of the protein. This new index was based on constrained energy minimisations, but can also be obtained from fluctuations in molecular dynamics simulations. One application of this method was to reveal dynamical domains of proteins, by the observation of the response of the protein structure to the applied constraint. This work was done in close collaboration with Isabelle Navizet (now lecturer at Université de Marne-la-Vallée), and was later continued by Sophie Sacquin-Mora.
  3. Finally, I also explored multiscale simulations, developping a mixed CG/MM representation of proteins. This work has been started during a 2-months stay in the laboratory of Prof Steve Harvey at Georgia Tech. The idea was inspired by QM/MM representations that were already available. The aim was to design a representation that could allow the study of very big systems at lower computationnal cost, keeping atomistic details only where necessary, and modeling all the other parts by a gaussian network model (that has been shown to be appropriate to reproduce slow motions in proteins). Unfortunately, I was not able to publish this work during my PhD thesis, but if you are interested, details are given (in french) in my thesis manuscript below.

Overall, these 4 years of research allowed me to experience a great variety of molecular simulation methods useful in biomolecular modeling: molecular dynamics, normal modes, energy minimization, umbrella sampling, MM-PBSA calculations...

Key words

proteins, mechanical properties, molecular dynamics, cellular adhesion, cadherins, constrained energy minimisation, multiscale simulation

Publications

Download