Simulating Crowded Cytoplasm
Model fits experimental evidence
In biology textbooks, the carefully rendered cross-section of an E. coli cell often resembles a well-organized and spacious apartment, with everything in its place and ample room for movement. But a recent computational recreation of the scene looks more like a Friday night dance floor, with molecules bumped up against one another in every direction. In addition to providing a dramatic, qualitative description of the crowded cytoplasm, this first atomically detailed computational model of E. coli innards is also a tool for quantitative predictions of molecular conduct within the cell. The model is described in the March 2010 issue of PLoS Computational Biology.
“This is an attempt to build a virtual lab, in which we can study various biological and biophysical processes as they might occur inside the cell,” says Adrian Elcock, PhD, coauthor and associate professor of biochemistry at the University of Iowa.
The sea of floating proteins inside every cell is the background against which many cellular reactions take place. Scientists realized years ago that the cytoplasm is generally not an invisible player in those reactions. One of the best-studied examples is macromolecular crowding (also called excluded volume effect). Having large neighbors on every side changes a protein’s effective concentration and influences its movement and ability to react. A biological reaction observed in dilute solution can be much faster or slower than the same reaction inside a crowded cell.
To create the model, Elcock and then graduate student Sean McGuffee, PhD, started by gathering known structural data for 50 of the most common E. coli proteins. They then combined the detailed representations inside a computer model at known cellular concentrations, creating a strikingly dense model of 1008 proteins. The researchers then set that image in motion, running independent Brownian dynamics simulations governed by varying energetic descriptions of intermolecular interactions. The simplest description included only the excluded volume effect: no molecule could take the space of another molecule. The most complex scenario they ran included excluded volume, electrostatic interactions, and favorable short-range hydrophobic interactions. The more complex simulations performed surprisingly well when asked to predict molecular behaviors, such as diffusion and stability, in the E. coli cytoplasm.
The model was able to match experimental observations of how quickly green fluorescent protein diffuses in the E. coli cytoplasm. And it was able to predict the greater stability of the unfolded state of the protein CRABP, cellular retinoic acid binding protein, over the folded state inside E. coli. Although the presence of close neighbors (crowding) generally stabilizes a large folded protein, the specific electrostatic and hydrophobic interactions of unfolded CRABP with other cytoplasmic proteins counteract the crowding effect.
“What this doesn’t mean,” Elcock emphasizes, “is that crowding effects are unimportant. It means that crowding is only part of the story.”
A computational box of 1008 proteins is still a far stretch from the complex E. coli cytoplasm, says Allen Minton, PhD, a pioneer in the study of crowding effects and researcher of physical biochemistry at the National Institutes of Health. “But there are a lot of questions that only this type of computation can answer,” he says. “From a computational point of view, it is a real tour-de-force.”