Protease Structure Reveals Surprises

Current treatments for the hepatitis C virus, which can cause cirrhosis and liver cancer, are toxic with bad side effects. Now, a structure of a hepatitis C virus protease that could become a target for new drugs has been revealed by virologists at Rockefeller University.

Postdoctoral associate Ivo C. Lorenz, research associate Joseph Marcotrigiano, visiting grad student Thomas G. Dentzer, and professor Charles M. Rice report the structure of the catalytic domain of the hepatitis C virus NS2-3 protease (Nature, DOI: 10.1038/nature04975).

The structure is "eye opening," says Ralf Bartenschlager, a virologist at the University of Heidelberg, in Germany. This study "may redirect the interest of drug developers to this enzyme that has been largely neglected," he says.

The hepatitis C virus's RNA genome is translated into a single polyprotein that must then be chopped into the virus's 10 component proteins. Within this polyprotein chain, the NS2-3 protease, which is made of parts of the so-called nonstructural proteins 2 and 3, is responsible for a single cleavage between those two proteins. Although the protease includes parts of both NS2 and NS3, the catalytic domain comes from NS2.

The simplest model of the protease's activity suggests that as soon as the relevant parts of NS2 and NS3 are synthesized, they come together and make the cut. Rice and coworkers show that this simple model can't be right. They find that the catalytic domain of the protease is actually a butterfly-shaped dimer with two identical composite active sites. Of the three key amino acids in each active site, two come from one monomer and one comes from the other.

Although the dimer was a surprise, it actually makes sense, according to Rice. "Why would you encode proteolytic activity that gets used right away for mediating a cleavage?" he asks. If proteolytic activity is concentration dependent, "the dimer formation gives you an added ability to regulate the cleavage."

In addition to the dimer, the structure also reveals that the protease's C-terminal leucine residue is "stuffed into the active site," Rice says, effectively turning off the enzyme after it cleaves the protein.

Although biochemical assays have shown that the protease depends on zinc, there is no zinc associated with the active site. Rice suspects that the zinc may be playing a structural role in the NS3 portion rather than a catalytic role.

Raffaele De Francesco, a hepatitis C expert at Merck's Institute for Research in Molecular Biology, Pomezia, Italy, predicts that "availability of the structure of the elusive NS2-3 protease will trigger intense research aimed at the design or discovery of inhibitors of its proteolytic function." However, the way the enzyme inhibits itself after a single cycle may make it challenging to interfere with the enzyme's activity.

"In order to control this virus, you're going to need a cocktail of specific antivirals," Rice says. "More targets with different resistance profiles are probably going to be the way to go." The structure of protease NS2-3 may give drug hunters yet another target.