Navigation and service

Mesoscale Modeling of Neural Cascades

Neurotransmission is at the heart of brain functioning. A complex network of chemical reactions (molecular signaling cascades) involving hundreds of different molecules, which have to diffuse, meet and interact at the correct time in the correct place, is responsible for the transmission of information through neuronal cells and regulates very complex processes, such as memory, learning, mood, etc. Molecular signaling cascades are initiated by a stimulus (first messenger), acting on a receptor in the post-synaptic neural membrane. This signal is transduced to the cell interior via “transducer” membrane proteins that in turn activate “effector” membrane proteins leading to second messengers’ production.  Second messengers amplify the initial signal and target other effector proteins in the cytosol, resulting in a specific neuronal response to the initial stimulus.  

The emerging picture is that all these events are highly spatially organized. Signaling events initiate within the two-dimensional plane of the membrane and then move through the three-dimensional volume of the cytosol. Moreover, the signaling molecules are non-uniformly distributed. For instance, receptors, G-proteins, and signaling effectors, can be localized in the same or different membrane microdomains, with specialized functions and membrane composition, which constrain their lateral mobility, eventually favoring/disfavoring their interaction. Evidences cumulated over the past two decades suggest that GPCRs can form dimers or oligomers in membrane, which affect their trafficking and/or signaling. In order to understand how membrane composition, diffusion, localization and oligomerization combine together and regulate neurotransmission response, we are developing in collaboration with the ICS-3 institute, a protocol for mesoscale modelling of post-synaptic signalling events (sub-micron resolution) based on the (Generalized) Langevin dynamics and multiparticle collision dynamics simulation schemes. The phenomenological parameters used as input to such mesoscopic simulations are derived in a mean field-type approach by higher resolution (atomistic or quasi-atomistic) simulations. The proposed theoretical/computational research represents an effort to go beyond the single molecule description, towards a more systemic-oriented modeling of information processing at subcellular level.

Lateral diffusion of the CoM of Muscarinic receptor 2 in a neuronal membraneFigure 1 – Lateral diffusion of the center of mass of Muscarinic receptor 2 in a neuronal membrane, calculated through Generalized Langevin Dynamics simulations.