NCSA NEWS

The quest for a cure

By Barbara Jewett, NCSA

Story posted November 9, 2006

Due to its central role in processing viral polypeptide precursors, HIV-1 protease (HIV-PR) continues to be one of the primary targets of anti-AIDS drug discovery. A greater understanding of the mechanistic events associated with binding of HIV-PR substrates and inhibitors is critical for the design of more potent inhibitors of the enzyme.

Carlos Simmerling and his colleagues at Stony Brook University are viewing the moments at which "starter molecules" for HIV are most vulnerable to new drugs. Using NCSA's SGI Altix supercomputer, called Cobalt, they successfully simulated how HIV protease changes between two forms that already have been determined through experiments. More importantly, however, the group captured the protease in a third, fully open state -- one that had previously been hypothesized but never directly observed.

Coming together
HIV protease is a dimer, formed when two identical chains come together (right and left). The semi-open form is shown, as seen in the crystal structures without an inhibitor. The two flaps (purple and orange, one from each monomer) contact each other and cover the binding pocket, blocking access to the active site. The aspartic acid side chains that the protease uses to carry out its function are at the bottom of the binding pocket (shown in yellow). During simulation without an inhibitor, the flaps are very flexible and the flap tips periodically lose contact with each other.

Flipping flaps
After the flap tips separate, the two monomers rotate with respect to each other and the flaps become separated by about 20 angstrom, opening the active site and allowing substrate or inhibitors to enter. After a short time, the structure closes back to the semi-open form in which it spends most of its time. The binding pocket is again inaccessible to substrate or inhibitors. Occasionally the flap tips swap sides -- the left flap (purple) is in front, while in the normal semi-open form the right flap (orange) is in front. Having the left flap in front is observed in crystal structures of the inhibitor-bound protease, but without an inhibitor this conformation is unstable and rapidly switches back. NMR experiments had suggested that the protease might undergo this kind of transient flap reversal.

In the pocket
In the open form, substrates or inhibitors can enter the binding pocket. In this simulation, we manually placed an inhibitor (green) near the entry to the active site of an open structure from a simulation without an inhibitor. The inhibitor enters the binding pocket and the flaps begin to close. The inhibitor forms specific hydrogen bonds with one of the catalytic aspartic acids and one flap, accelerating the closing process. The other flap closes and helps to pack the bulky side groups of the inhibitor into the pocket.

A new stability
The protease becomes fully closed. For comparison, the crystal structure of the complex is shown as a transparent gray model. The simulation model is nearly identical to the crystal structure even though the closed form was not used for the simulations. Once the protease closes with an inhibitor in the pocket, the flaps become stable and no longer open in the simulations. Note that the left flap (purple) is in front when the inhibitor is bound. When the protease closes without an inhibitor, the right flap is in front. This flap reversal upon inhibitor binding is also observed in the crystal structures.

The potential for overcoming HIV
The change in flap handedness upon ligand binding is clearly reproduced in the team's simulations, but its possible role in HIV-PR function remains unexplained. Understanding the issues that govern HIV-PR flap mobility has profound implications for elucidating the detailed mechanism of this enzyme and in the design of new therapeutic agents, such as allosteric inhibitors intended to interfere with flap opening and thereby with enzymatic function. The team's work was published in the Proceedings of the National Academy of Sciences (103:915-920, 2006) and in the Journal of the American Chemical Society (128: 2812-2813, 2006).

This work was supported by the National Institutes of Health and the Department of Energy.

All images were made using VMD and POV-Ray.

Team members:
Viktor Hornak
Asim Okur
Robert C. Rizzo
Carlos Simmerling