By David N. Leff

Current multidrug treatments against HIV work in one of two ways. Compounds such as AZT attack the reverse transcriptase, which the virion uses to copy its genetic material inside its target cell. Protease inhibitors, the other type of therapeutic in use today, stop the enzymes by which the HIV processes its own proteins.

This leaves the field wide open to a strategy for attacking HIV frontally, on its exposed viral coat, or envelope.

"The envelope protein is not targeted by any of the therapies," observed structural biologist Peter Kim, of Massachusetts Institute of Technology's Whitehead Institute. "I think it's fair to say," he added, "that people would welcome it as a third target for these combination regimens, which have proven successful, because it would be targeting something that occurs before the virus enters the cell."

HIV's envelope protein sits on the surface of the virus. It consists of two glycoproteins with separate functions. The first, gp120, binds the invading virion to receptors on the cell surface, namely CD4 and the recently elucidated chemokine co-receptors. (See BioWorld Today, Nov. 14, 1996, p. 1.) The second, gp41, is responsible for the actual fusion event between the virus and the cell.

"HIV is surrounded by an intact lipid bilayer, as is the cell," Kim noted. "In order for the virus to infect the cell, these two membranes need to fuse. And gp41 carries out this fusion reaction."

Kim, an investigator at the Howard Hughes Medical Institute, is senior author of a paper in the current issue of Cell, dated April 18, 1997, titled "Core structure of gp41 from the HIV envelope glycoprotein."

Getting Crystal Structure Was Breakthrough

"What we did," he told BioWorld Today, "was to study a fragment of gp41 and get its crystal structure. This is the first time that anybody has been able to get a high-resolution -- to 2.0 Angstroms -- of any part of the envelope protein. It's clearly important," he pointed out, "because the envelope is involved in cell recognition and membrane fusion. It's also the major target that the immune system attacks."

Because trying to get the crystalline structure of the entire gp41 protein would be a difficult and daunting challenge, Kim and his colleagues dissected out two key fragments representing about half of it. These they "whittled down by protein dissection to a smaller fragment, but which still contained the essential elements of the function we were trying to investigate -- gp41's fusion process."

What the X-ray crystallography revealed, Kim went on, "was really a big surprise -- the presence of a deep cavity, or pocket, in one part of the gp41 protein. And this structure was filled by a knob-like protrusion from the other fragment. It reminded us of a ball-and-socket arrangement."

That peptide structure also reminded Kim of several other strictly mechanical components, such as spring-loaded harpoons and switch-blade knives.

Alluding to work by other researchers, he said, "These peptides are known to be inhibitors -- good inhibitors -- of HIV infection in the test tube. And they are inhibitors in the micromolar to picomolar concentration range."

The way in which they do their inhibiting "is to bind to the complementary region on the viral gp41. In other words, these two peptides we studied correspond to two different regions of the protein. When you add them one at a time, they bind to the virus and gum it up.

Creating A Drug Is Next Step

"How could this inhibition acquire a clinical or pharmacological form?" he asked rhetorically. "The presumption we are making is that if one could take that peptide up to a small molecule, that would get around the bioavailability problem associated with peptides. Then one might have a therapeutic."

On the downside, Kim went on, "peptides in general don't make good drugs, because they are rapidly degraded by the body and are not very well absorbed. The point is, we know these structures can be inhibited, and thus you can prevent fusion of HIV. So now that we have a high-resolution view of the ball-and-socket, we're hopeful that other people will be able to take this structure and develop, either through rational drug design or combinatorial methods, small molecules that will bind inside the socket. If so, one might be able to prevent the actual fusion event itself."

He continued: "The exciting thing, as far as the structure goes, is that this pocket is incredibly deep, and incredibly hydrophobic. So it looks like the sort of thing that would be a good target for developing a drug with high affinity. And that's really as far as we could go right now."

In 1993, Kim published a paper, also in Cell, "in which we predicted what we called the spring-loaded mechanism for the conformational change in the influenza virus. That prediction," he said, "led to a harpoon-like mechanism --which I think is a good analogy."

Like a sharp, pointed weapon, or a hypodermic syringe and needle, Kim suggested, the region of gp41 projecting beyond the viral membrane "gets inserted into the cell membrane as an intermediate step in fusion, so that it ends up being in both membranes at once, cell and virus."

"I should certainly say," Kim added, "that the HIV gp41 protein structure has a striking similarity to flu, and also to a mouse leukemia virus protein structure.

"It's important to emphasize," Kim concluded, that this structure is just a start in our understanding of the envelope protein. There's really a lot more to do, including getting the structures of gp120 itself and gp120 bound to CD4, and how it interacts with the chemokine co-receptors.

"This is really the first glimpse that we have of the HIV envelope protein. We're genuinely hopeful that others will be able to take the structure and dive into efforts to develop another drug that can be added to the combination regimens for treating HIV." *