A coarse-grained model for the simulations of biomolecular interactions in cellular environments.

Xie ZR, Chen J, Wu Y.

J Chem Phys. 2014 ;140(5):054112.

Department of Systems and Computational Biology, Albert Einstein College of Medicine of Yeshiva University, 1300 Morris Park Avenue, Bronx, New York 10461, USA.

 

Abstract

The interactions of bio-molecules constitute the key steps of cellular functions. However, in vivo binding properties differ significantly from their in vitro measurements due to the heterogeneity of cellular environments. Here we introduce a coarse-grained model based on rigid-body representation to study how factors such as cellular crowding and membrane confinement affect molecular binding. The macroscopic parameters such as the equilibrium constant and the kinetic rate constant are calibrated by adjusting the microscopic coefficients used in the numerical simulations. By changing these model parameters that are experimentally approachable, we are able to study the kinetic and thermodynamic properties of molecular binding, as well as the effects caused by specific cellular environments. We investigate the volumetric effects of crowded intracellular space on bio-molecular diffusion and diffusion-limited reactions. Furthermore, the binding constants of membrane proteins are currently difficult to measure. We provide quantitative estimations about how the binding of membrane proteins deviates from soluble proteins under different degrees of membrane confinements. The simulation results provide biological insights to the functions of membrane receptors on cell surfaces. Overall, our studies establish a connection between the details of molecular interactions and the heterogeneity of cellular environments.

Go To PubMed

 

Additional Information

Protein-protein interactions are essential in almost all cellular functions. It is well-known that these interactions are largely affected by their dynamic interplay with the specific environments. However, most of the current experimental methods isolate proteins from their in vivo biological surrounding in order to permit a convenient analysis of protein interactions. This leads to the fact that the discrepancies are usually significant between in vitro and in vivo observations. Using computational simulations, we are trying to link protein interactions back to their cellular environments. Proteins in our model are represented by rigid bodies with physical dimensions. This simplification dramatically improved the computational efficiency. Moreover, the diffusive properties of proteins are distinguished by their cellular locations, e.g. cytoplasm or membrane surfaces. Our theoretical analysis revealed that protein binding in a crowded environment is affected by two opposite factors. If two interacting proteins are initially segregated by other background molecules, crowding reduces the association between these two proteins due to the lagging of their diffusions. On the other hand, if two interacting proteins are initially surrounded by the crowded background or dissociated from a complex, their interaction or re-association is confined within a certain range, which enhances the stability of their binding. We further showed that the dimensional constraints of membrane-bound proteins can significantly impact their binding properties. Future applications of our low-resolution model therefore can shed light on the functional impacts of membrane-involved systems. For instance, studies on the mesoscopic binding properties between extracellular ligands and cell surfaces receptors will help us to test and predict the efficacy or toxicity of new drug candidates in antibody-based cancer therapy.

 

A coarse-grained model for the simulations of