GPCRs dimerization unveiled by millisecond timescale free-energy calculations

Vittorio Limongelli

Faculty of Biomedical Sciences, Institute of Computational Science, University of Lugano USI (Switzerland)

Faculty of Pharmacy, University of Naples “Federico II’ (Italy)

e-mail: vittoriolimongelli@gmail.com

Membrane proteins diffuse in the phospholipid bilayer forming functional dimers and oligomers which can play specific roles during cell cycle and in pathological condition. Unfortunately, the elucidation of membrane protein/protein binding interaction is difficult using standard experimental and computational techniques because of the size and complexity of the systems, and the slow diffusion of proteins in membrane. In the present talk, I introduce an innovative multiscale approach that combines Coarse-Grained molecular dynamics and MetaDynamics (CG-MetaD) [1], allowing to overcome both the size and the timescale limit of the state-of-the-art techniques. Specifically, in CG-MetaD the representation of the system as beads instead of atoms reduces the dimensionality of the system under investigation, while metadynamics enhances the phase space exploration, allowing practical investigation of long timescale and large-scale motions of proteins in membrane. As a result, CG-MetaD super-accelerates the sampling allowing to go over the second timescale, well beyond the timescale accessible by the state-of-the-art simulations.

CG-MetaD has been used to disclose the free energy landscape underlying the dimerization mechanism of the transmembrane helices of the epidermal growth factor receptor. The characterization of the free energy minima allows identifying the active and inactive conformations of the receptor, shedding light on possible activation pathways [1]. Very recently, we have performed millisecond CG-MetaD calculations to describe the dimerization process of the GPCR chemokine receptors CCR5 and CXCR4. The free diffusion of the proteins in membrane and the occurrence of several protein/protein binding events during the simulation, lead to a well-characterized free energy surface and the identification of the receptors’ dimer states [2]. The dimerization interface in all the energy minima is characterized with atomic resolution elucidating the involvement of the lipids and cholesterol molecules. Our findings pave the way to further investigations especially drug design in which the dimer states might be targeted to discover ligands able to interfere with the chemokine GPCRs dimer formation that characterizes severe pathologies like cancer and HIV.

Figure. Coarse grained representation of the transmembrane helices of the epidermal growth factor receptor

and representation of their dimerization free energy surface calculated through CG-MetaD (details in ref 1).

References
1. Lelimousin M, Limongelli V, Sansom MSP. J. Am. Chem. Soc. 138, 10611-10622 (2016)

2. Di Marino D, Limongelli V. Biophys. J. 3, 177a (2016)