Home

Group Members

• Protein folding and unfolded states
• Recognition and repair of DNA damage
• Drug Discovery / Disease
• HIV/AIDS
• Tuberculosis
• Cancer
• Simulation methodologies
• Conformational sampling methods
• Force Field Development

HIV/AIDS 

  
   Since 1981, more than 25 million people have died from acquired immunodeficiency syndrome (AIDS). 12 million children in Africa have been orphaned due to AIDS. More than 20 years after the human immunodeficiency virus (HIV) was discovered to be the causative agent of AIDS, effective and widely available treatments remain elusive. 40 million people are currently living with the disease, and every day 12 thousand are newly infected with HIV. In 2006 alone, three million people died from AIDS; tragically, 380,000 of these were under the age of 15. Former UN Secretary-General Kofi Annan summarized the AIDS pandemic as “the greatest challenge of our generation”.                              
   Although a cure has not been found, substantial progress has been made by pharmaceutical researchers, resulting in drugs that prolong and improve quality of life for at least some AIDS patients. The main avenues include (1) interfering with HIV entry into host cells, (2) blocking reverse transcription, the process through which HIV converts its RNA genome into DNA, and (3) inhibiting the function of HIV protease (HIV PR), a viral protein that catalyzes a crucial chemical step in the maturation of new viral particles. Examples of FDA-approved HIV PR inhibitors are ritonavir, nelfinavir and amprenavir. When successful, these HIV PR inhibitors bind tightly to the active site, keeping HIV PR from binding to and cleaving its substrate viral polypeptides, and thus preventing production of new infectious viral particles. These inhibitors also serve as prime examples of structure-based rational design, in which high-resolution structures (typically from X-ray crystallography) are used to design small molecules with high complementarity to the active site of the target.

  Although current HIV PR inhibitors are often life-saving, it is estimated that up to half of all AIDS patients are infected with HIV strains that are resistant to at least one of the currently approved inhibitors. Over time, patients typically develop HIV strains in which none of the current inhibitors can effectively prevent HIV PR from performing its job. Thus it is of utmost importance to determine how these mutant viruses are able to function in the presence of inhibitors in order to design new and more effective treatments. There is strong indirect evidence that flexibility plays a role in this drug resistance, yet dynamic information is typically not provided by crystal structures, which show only a static view of the protein that is averaged over long times and many molecules in the crystal results. As a prime example, in nearly all crystal structures of HIV PR, the protein adopts one of two structures that differ slightly in the conformation of two flaps in the protein chain that cover and protect the HIV PR active site. In all of the substrate- or drug-bound forms, the flaps adopt a somewhat different conformation than seen for the flaps in free HIV PR. In both forms, the flaps are closed and in contact with each other, preventing entry or exit of either substrates or inhibitors. It has become clear that the flaps must open in order for drugs to enter, but experiments have been unable to show what the critically important but presumably transient open form looks like or how mutations affect the relative energies of the two known forms. This information could enable development of new drugs that bind to mutant HIV PR, or those that target the opening mechanism itself. The latter may prove more difficult for the virus to overcome through mutation than is the current approach of simply filling the HIV PR active site to block binding of its natural substrate.
   In principle, simulations of HIV PR could provide a model for the opening process and explain how drugs gain access to the active site and how this changes with different drugs or HIV strains. Until the present, however, no researchers had reported the ability to simulate HIV PR opening and closing, and a reliable model for the dynamic changes that accompany HIV PR drug binding was not available. 

  Multiple labs have attempted to simulate HIV PR flap opening and closing, as well as drug binding, but none had previously provided results that were in agreement with experimental observations. We have employed a multi-scale approach along with our recently developed force field parameters to achieve unrestrained, all-atom simulations in which large conformational changes of the HIV PR were observed. The simulations represent a breakthrough because, for the first time, they showed the protease undergoing spontaneous conversion between the two forms observed in the experimental structures. Simulations of the inhibitor-bound structure were stable, while removal of the inhibitor resulted in conversion to the form seen in the crystal structure of the unbound protease. Importantly, we observed the protease flaps transiently opening to a much greater degree than observed in any experimental structure, followed by re-closing, providing a model for dynamic changes that permit drug entry. In subsequent research, we placed an inhibitor near our newly determined model of open HIV PR; these simulations resulted in spontaneous binding of the inhibitor to the open active site, inducing closing of the flaps and adoption of a structure of inhibitor-bound HIV PR nearly indistinguishable from that determined by experiment. These results revealed the dynamic mechanism of HIV PR drug binding in atomic detail. Subsequently, an experimental crystal structure of multi-drug resistant HIV PR was published by another lab, revealing incompletely closed flaps that were suggested as a possible source of drug resistance. Based on simulations in solution and crystal environments, we determined instead that the drug-resistant mutant most likely adopts the normal structure in solution. Instead of the reported interpretation that the mutant was unable to close, we showed that the crystal structure was actually the first example of a mechanism of HIV PR allosteric inhibition recently proposed by the McCammon lab, potentially providing a model for a new class of AIDS drugs.

Our recent publications on this topic:

Hornak, V. and Simmerling, C., “Targeting structural flexibility in HIV-1 protease inhibitor binding”, Drug Discovery Today 12:132-138 (2007).  Link

Hornak, V.; Okur, A., Rizzo, R. and Simmerling, C., “HIV-1 protease flaps spontaneously open and reclose in molecular dynamics simulations”, Proc. Nat. Acad. Sci. USA, 103:915-920 (2006). Link

Hornak, V.; Okur, A., Rizzo, R. and Simmerling, C., “HIV-1 Protease Flaps Spontaneously Close to the Correct Structure in Simulations Following Manual Placement of an Inhibitor into the Open State”, J. Am. Chem. Soc., 128: 2812 (2006). Link

Layten, M., Hornak, V. and Simmerling, C., The open structure of a multi drug resistant HIV-1 protease is stabilized by crystal packing contacts, J. Am. Chem. Soc., 128: 13360-13361 (2006) Link