sealPurdue News

October 2000

Small molecules used to block proteins in HIV

WEST LAFAYETTE, Ind. – Taking a cue from the Biblical story of David and Goliath, Purdue University scientists are using small molecules to bring down the molecular giants responsible for viral replication in AIDS.

The new approach blocks the interaction of two identical proteins that make up HIV protease by strategically slinging small molecules between key segments on the proteins as they come together to form an enzyme needed for viral replication. HIV protease is one of three essential enzymes used by the AIDS virus to replicate.

By targeting the site where the proteins come together, the small molecules can serve as a "molecular wedge" to prevent the proteins from interacting, thus thwarting the chain of chemical reactions needed to replicate, says Jean Chmielewski (she-ma-LEW-ski), associate professor of chemistry at Purdue who is a pioneer in the new approach.

"Theoretically, if you can stop the proteins from coming together and performing their biological activity, you'll stop viral replication," she says.

The method is currently being tested at the National Institutes of Health in cells infected with HIV, the virus that causes AIDS. If the studies prove successful, the general approach may someday be used to treat a wide range of diseases and disorders, including some currently untreatable conditions such as autoimmune diseases.

"At this stage, we've been able to synthesize some fairly potent small-molecule inhibitors of HIV protease, which do, in fact, block the ability of this enzyme to come together," Chmielewski says. "That, in turn, blocks its biological activity."

Chmielewski will present details of the new method Sunday (8/20) at the American Chemical Society's national meeting in Washington, D.C.

The new approach is based on the fact that an individual protein usually must work in conjunction with other proteins to carry out vital functions within a cell.

"There are a lot of disease states that are perpetuated by the ability of proteins to communicate or miscommunicate, and if we can block that communication process, we may be able to shut them down entirely," Chmielewski says.

In her HIV model, Chmielewski targeted a site where identical pieces on two proteins come together to form what is called a "dimer." The dimer then becomes an essential enzyme, prompting numerous chemical reactions that allow the virus to replicate.

"This particular enzyme is only active when those two identical pieces join together, so if we can stop the proteins from coming together, we may be able to put the brakes on the replication process," Chmielewski says.

That fact has not escaped drug companies in the last decade. The Food and Drug Administration has approved a number of compounds to target this particular enzyme. But the sites on the virus where those compounds work can mutate fairly easily, making the drugs resistant with time, Chmielewski says.

She says the new approach may help prevent the possibility of drug resistance because it acts on a site that is less likely to mutate.

"We think this might be a way to get in the back door and eliminate or reduce the problem of mutations and the possibility of drug resistance," she says.

The "wedge" molecules were developed by taking pieces of the most active sites on the identical protein segments and putting them into one molecule. After developing a prototype molecule, Chmielewski and her group worked to make smaller and smaller versions of it to obtain an agent small enough to pass through a cell's membrane. The group then developed different versions of the molecule, producing a small library of different compounds.

Chmielewski is now working with collaborators at NIH to test a dozen of those compounds in cells infected with HIV. If the approach proves successful, it may be developed further and carried into human trials within the next five years.

The use of molecular wedges may someday also provide a radical new way to approach other diseases and disorders. Chmielewski notes that due to their large size, proteins are a difficult target, particularly in drug development where scientists must work with very small molecules.

"Because protein surfaces are very large and can interact at various sites, trying to develop a molecule that can both pass through the cell's membrane and block the proteins isn't easy to accomplish," she says. "The molecules we developed are 50 times smaller than the dimeric enzyme they must block, so it's like David taking on Goliath. But I think we've learned a lot over the past ten years about how to do it."

One way to accomplish this feat is to find key portions on the surface of a protein where it joins and interacts with other proteins, she says.

"In the case of HIV protease, there are only two major pieces that actually come together to form the key enzyme, and that's one of the reasons why we chose that as our model system," Chmielewski says.

She is now using this approach to find ways to inhibit various transcription factors, molecules used in the replication of cells. Her studies at Purdue are funded by the National Institutes of Health.

At the ACS meeting in Washington, Chmielewski will outline her approach and serve as chairwoman of a session on this topic. Her panel includes three members from industry and a researcher from Yale who are using small molecules to inhibit protein interactions in various other systems.

Source: Jean Chmielewski, (765) 494-0135,

Writer: Susan Gaidos, (765) 494-2081;

Purdue News Service: (765) 494-2096;


Small molecule dimerization inhibitors of HIV protease

Jean Chmielewski, Reena Zutshi, Michael Shultz, Michael Bowman, Young-Wan Ham, Xuimin Zhao, and George Tora

As more information on macromolecular structure unfolds, it has become evident that inter-protein interactions are a ubiquitous aspect of biological activity. Whereas the nature of these protein-protein interactions is becoming better understood, rational approaches to inhibiting these interactions are still in their infancy. Using dimeric HIV protease as a template, we have developed a strategy to inhibit dimerization based on cross linked, interfacial peptides derived from the interface of this protein. Recent efforts to identify minimal structures that inhibit dimerization, and the design of small molecule inhibitors of HIV protease using a library approach will be described.

* To the Purdue News and Photos Page