Whole Virus Simulation
Simulation of a one-million-atom virus reveals unexpected twist
Giving new meaning to the phrase computer virus, researchers have created a computer simulation of an entire biological virus comprising approximately one million atoms.
“It wasn’t clear before that one could do a simulation of such a large living system at an atomic level and learn something from it,” says Klaus Schulten, PhD, professor of physics at the University of Illinois at Urbana-Champaign. But when he and graduate students Anton Arkhipov and Peter Freddolino successfully simulated the satellite tobacco mosaic virus (STMV), they revealed some surprising features of the particle in the process. The work was published in the March 2006 issue of Structure, as a collaboration with virologists from University of California, Irvine.
Viruses must do two things: infect cells and transport their genetic material inside a stable container known as a capsid. In the case of the STMV, the capsid consists of 60 identical proteins produced by the virus’s genome. Crystallographers who had imaged the small virus believed all 60 pieces were arranged in complete icosahedral symmetry. The computer simulation, however, showed this to be an incomplete picture of the virus.
Schulten and his colleagues started with the crystallography image of STMV and then allowed the atoms to move according to their physical properties. For just over 10 nanoseconds (broken into 10 million time steps), “we let the laws of physics take over,” says Schulten. The result: Although the capsid remained generally spherical, some of the symmetry was lost. “The virus developed a belt around an equator of the sphere, and that belt engaged in a back and forth motion,” Schulten says.
More important, simulation revealed that, unlike many other viruses, the STMV capsid is unstable without its RNA contents and depends on the RNA to assemble. “It seems that for this virus, the genomic material first aggregates into a sphere, and then recruits the 60 proteins to be a shell around itself,” Schulten says. “This is opposite to what one expected.”
Schulten and his colleagues hope that viral simulations of this type will help researchers understand how viral capsids shift from stable to unstable when they are infecting a cell. It’s possible that one might be able to interfere in an infection at the point when the capsid breaks apart, he suggests. “We want to use information gained from simulations to protect people from viral infections.”
In future projects, Schulten and his colleagues plan to simulate the poliovirus and other viral particles that are 4 to 10 times larger than STMV. Their success with STMV suggests that large scale simulations provide valuable, new information. “Had we done a partial simulation, we wouldn’t have learned as much,” he says.