Atomistic Cell Model Captures Molecular Crowding Effects

    Something like crowd psychology is believed to take hold within the cell, which confines individual molecules so much that they may behave differently than they would, say, within a test tube, a relatively expansive environment. Yet the behaviors of molecular crowds are subtle, unlike the popular delusions, panics, or bouts of irrational exuberance that sometimes occur within human crowds.

    Because molecules do not “go mad in herds,” the way people might, their actions are best assessed individually, as in molecular dynamics simulations. A new simulation, staged by researchers at the RIKEN Institute, has gone so far as to model the cytoplasm in atomistic detail. This simulation has shown that there are major differences between in vitro conditions and the in vivo conditions in the cell.

    Specifically, the simulation has produced evidence that an understanding of in vivo molecular behaviors needs to take better account of interactions, such as protein–protein interactions and electrostatic interactions with ions and metabolites. Such interactions are sometimes neglected if biochemists simply assume that molecular crowds respect the volume exclusion effect—the idea that molecules monopolize a certain volume of the solvent, in this case water, in the solution around them, preventing other molecules from occupying that space.

    The RIKEN researchers presented their work November 1 in the journal eLife, in an article entitled, “Biomolecular Interactions Modulate Macromolecular Structure and Dynamics in Atomistic Model of a Bacterial Cytoplasm.” The article describes a first step toward physically realistic in silico whole-cell models that connect molecular with cellular biology. It also explains how protein–protein interactions may destabilize native protein structures, whereas metabolite interactions may induce more compact states due to electrostatic screening.

    “Protein-protein interactions also resulted in significant variations in reduced macromolecular diffusion under crowded conditions, while metabolites exhibited significant two-dimensional surface diffusion and altered protein-ligand binding that may reduce the effective concentration of metabolites and ligands in vivo,” wrote the article’s authors. “Metabolic enzymes showed weak non-specific association in cellular environments attributed to solvation and entropic effects.”

    For this study, RIKEN scientists modeled the inside of the smallest known bacteria Mycoplasma genitalium, which has a length of approximately 400 nanometers, and dynamically modeled approximately one trillion atoms within the bacterial cell, making this one of the largest molecular dynamic simulations performed to date. The simulation was carried out with GENESIS, a massively parallel molecular dynamics program developed at RIKEN. The calculations, which used 65,536 processing cores of the K supercomputer, took several months to complete, despite the power of this computer.

    “This research has brought us one step closer to the dream of simulating a complete cell at the molecular scale,” asserted Yuji Sugita, one of the leaders of the research team. “The work will also contribute to drug development, as previous studies usually looked at interactions between proteins and a single candidate compound within water. Now, we will be able to also analyze the interactions between the candidate compound and other molecules within the crowded cellular environment.”

    “One limitation of this study is that because of the enormous computing power required, we were only able to conduct short simulations,” Sugita continued. “We believe it is still accurate, but hope to be able to perform this work on even more powerful future computers to reduce the statistical uncertainties and incorporate other interactions into the simulation, such as genomic DNA and cytoskeletal elements.”