Biological cells are very complex and organized systems. Physiological functions are governed by finely tuned dynamical networks of interacting species, going from individual molecules to large stable or meta-stable macromolecular complexes or assemblies. A great number of proteins act as ligand/substrate-dependent activators that contribute to cell signaling pathways (Manning et al., 2002, Olsen et al., 2006) regulating crucial physiological or physiopathological processes. The integration of these processes covers a wide range - from pico-seconds to seconds and beyond - of time scales. 

The problems addressed in this proposal mainly deal with multiscale simulation approaches  that can be used or developed in order to integrate the different scales (in spatio-temporal terms) of complexity in the biological systems, from microscopic or local changes at the level of individual biomolecules, through conformational dynamics of individual proteins and mesoscopic effects involving molecular recognition of interacting objects, to macroscopic or global behavior of cell communication networks (Rual et al., 2005). 

Biological cells are very complex and organized systems. Physiological functions are governed by finely tuned dynamical networks of interacting species, going from individual molecules to large stable or meta-stable macromolecular complexes or assemblies. A great number of proteins act as ligand/substrate-dependent activators that contribute to cell signaling pathways (Manning et al., 2002, Olsen et al., 2006) regulating crucial physiological or physiopathological processes. The integration of these processes covers a wide range - from pico-seconds to seconds and beyond - of time scales. The problems addressed in this proposal mainly deal with multiscale simulation approaches  that can be used or developed in order to integrate the different scales (in spatio-temporal terms) of complexity in the biological systems, from microscopic or local changes at the level of individual biomolecules, through conformational dynamics of individual proteins and mesoscopic effects involving molecular recognition of interacting objects, to macroscopic or global behavior of cell communication networks (Rual et al., 2005). 

Microscopic level – from local perturbation effects to large conformational changes: 

Proteins involved in cellular signaling, such as protein kinases, nuclear receptors or NMDA complex, often display a remarkable conformational plasticity. In solution, the population equilibrium between different conformational states is determined by their energetic landscape. Typically, this equilibrium can be displaced by ligand binding, phosphorylation events or point mutations. Internal modifications such as point mutations or insertion/deletion of amino acids can represent very subtle structural changes and yet have a great impact on the dynamics of the protein or induce large conformational changes at remote sites (Grant et al., 2010). It is becoming more and more clear that signal propagation occurs within individual proteins themselves through allosteric communication routes. Studies have begun to elucidate the mechanistic and thermodynamic impact of cancer mutations on protein kinases at the atomic level (Banavali & Roux, 2005, Dixit & Verkhivker, 2009, Laine et al., 2011). Recent advances in the field have permitted new strategies to emerge that target protein conformational transitions to modulate their function (Laine et al., 2010). Efforts have also been engaged to quantitatively relate equilibrium fluctuations to communication propensity or coupling between distant sites (Bahar et al., 2007, Hilser & Thompson, 2007). However a unifying framework for understanding signal transformation and noise propagation in terms of conformational fluctuations, intrinsic disorder, enthalpy/entropy balance, is still lacking.

Functional networks identification and description:

The second challenge addressed here concerns functional networks construction (Levy et al., 2009, Yosef et al., 2009) and global dynamics. According to the energy landscape theory proteins structural ensemble is modified upon changes in the environment (pH, ionic strength and temperature) (Kumar et al., 2000a, Kumar et al., 2000b). Such changes affect tens or more of proteins binding to the same “hub” that can assume a broad range of different conformations, each of which able to bind to a different partner (Dunker et al., 2005, Haynes et al., 2006) via possibly distinct binding sites (Keskin et al., 2005, Keskin et al., 2008). A largely unsolved question relates to what are the molecular specificity determinants that govern such a broad range of protein-protein associations (Tsai et al., 2009) and how diverse mutations and alterations of molecular signaling pathways resulting in the deregulation of important physiological processes and in physiopathologies (Santos et al., 2010, Schramm et al., 2010). Regarding this matter, a major challenge is to discriminate between the normal physiological processes and those corresponding to pathological functioning and understand how the subsequent changes in structure and dynamics manifest downstream.

Interplay between the challenges:

The main focus of our proposal addresses the problem of how to deal with these different organizational levels in order to link the microscopic characteristics to global behavior of a multi-component biological system. With the recent developments of multiscale representations/approaches, linking microscopic dynamical changes to macroscopic spatio-temporal dynamics is becoming possible. Several methods based on the modelling of the dynamical recognition processes taking place at protein-protein interfaces – with the aid of the already existing molecular dynamics simulations and collective movement analysis (normal modes, elastic modes or consensus modes) - can be extended to reach this goal and establish how these interactions relate to the various signalling pathways.

The methodological approaches as well as the applications in this domain may have a great impact in the medical field, and the interplay can also be foreseen between simulations and experiments.