One of the most spectacular properties of proteins is that they can operate as motors. This is due to the unique property of a protein (as opposed to other polymers and small organic molecules) to spontaneously collapse into a very specific and very rigid 3-dimensional molecular arrangement for the protein. For the most part, proteins are rigid. However, there are flexible regions in the protein that allows the protein to undergo large rigid body motions that, when coupled with chemical reactions, result in a chemically driven motor. We know much about how about some of these motors work. For example, myosin is a motor that contracts our muscle cells.

Myosin exists as little heads on a brushy fiber that spans the length of a muscle cell and this fiber is connected to the both ends of the muscle cell by a springy connector called titin. The myosin heads are stacked against two other pieces of fiber called actin that extends from either end of the cell.

When there is the chemical fuel ATP flooding into the cell, the myosin heads will tug at the two pieces of actin. The cumulative tugging of thousands of myosin motors pulls the two pieces of actin towards the middle of the cell and the muscle cell is compressed.

For a classical physicist, the vastly interesting question is how the myosin generates the force that tugs on the actin fiber. After the myosin latches on to the actin fiber, ATP naturally binds to the myosin in the special ATP binding site. When the phosphate breaks off the ATP, as ATP typically does, something mysterious happens that causes the whole back-end of the myosin to flip thus generating a huge mechanical kick (huge from the point of the myosin). It is this force that, when combined cumulatively with the forces from all the other myosin heads, pulls the two pieces of actin in the muscle cell together and compresses the cell.

The molecular structure of myosin in different states are known, both in the ATP bound state (before the motor stroke), and in the ADP state (after the stroke when ATP has lost one of the phosphates). But we really don’t know what happens in between. From the analysis of the crystal structures, it appears that when the phosphate of the ATP breaks off, a long helix is shifted to cover the space where the phosphate used to be. As a result, the part of the protein at the end of the helix is forced to rotate by some 60º. Here’s a movie, based on static crystal structures, of how the motion might take place (that also includes some real-time Electron Microscopy images of the motion at the end of the movie):

Beautiful optical trap experiments have been able to carefully tease apart individual myosin motors and thus measure the force generated by myosin, which is approximately 1-10 pN per motor, very small from our point of view, but enormous from the point of myosin. But where does the force come from? Understanding the details of this mechanical motion is, for me, one of the great challenges of computational chemistry. Is there some attraction of atoms to fill up the hole? Is there a cocked spring somewhere that releases the loaded helix? Is it some kind of subtle shift in an equilibrium ensemble – which will require a careful free-energy calculation?

I see this challenge as computational because these kind of insights cannot be determined from experiment, the detail is too great. Where is the impasse in simulations? Given that the heart of the chemistry is a classical ATP hydrolysis, this problem will probably not need quantum mechanics – using molecular dynamics ought to be enough. Nevertheless, simulating these huge motions using vanilla molecular dynamics is many orders of magnitude greater than what can be achieved by current computational chemistry setups. Innovate methods of force generation using molecular dynamics protocols will be needed.