How does the AIDS virus do it? That shifty pathogen not only evades the human body's immune defenses, but it also turns right around and lethally infects the very T cells deployed by that immune system.

"One of the really big problems in HIV that people have been trying to figure out for the last 20 years," observed structural biologist Peter Kwong, at NIH's National Institute of Allergy and Infectious Diseases (NIAID), is why, when you're infected, the virus can avoid being killed by human antibodies generated against it. Why that's true has not been clear. And our group has finally finished the last piece of our analysis to the puzzle of why it's true."

Kwong, at NIAID's Vaccine Research Center, is first author of a paper in Nature dated Dec. 12, 2002, titled: "HIV-1 evades antibody-mediated neutralization through conformational masking of receptor-binding sites."

"It turns out there's a lot of previous research that we and others have done to analyze this question," Kwong told BioWorld Today. "It started in 1998 when we published the structure of gp120 with CD4 and neutralizing antibodies. The crystal structures of the proteins afforded us access to the precise stereochemistry - the precise atomic coordinates - which allowed us to analyze it in organic chemistry terms. So they're not just these magical entities, but actually things that we can understand. And from that analysis in '98 we figured out numerous protective mechanisms by which HIV evades the immune system.

"But there is one huge mystery that we didn't understand," Kwong continued, "and that's the most vulnerable parts of the virus - the parts that then bind receptors and need to be conserved - need to be exposed. They turned out to be very large, larger than the typical antibody's footprint, which meant that the human immune system should be able to generate antibodies that target these extremely vulnerable regions."

Why No Vaccine To Date? Asked, Answered

"The big question that existed even after we solved the structure was why, in this big, potentially vulnerable region, the body's immune system is not able to generate antibodies to target and kill the virus? The answer, which I think is absolutely fascinating, is that protecting the receptor-binding region of HIV is an energy shield or mask that prevents any ligand - whether antibody or even receptor - from binding. So how then does the virus gain entry? How does it avoid neutralizing antibodies but still uses this same region to bind the receptor for entry?

"And the answer to that," Kwong went on, "is that the virus precisely titrates affinity for a ligand against the receptor binding surface, such that soluble ligands, like antibodies or even soluble CD4, which also doesn't neutralize very well, won't bind tightly enough to kill the virus. But when approaching the cell surface, the virus is able to bind multiple receptors simultaneously.

"The key viral target molecule is the gp120 envelope," Kwong explained. "Because viruses want to hide themselves from the immune system, they don't want to be seen. And what happens with envelope viruses is they actually cloak themselves in part of the lipid bilayer from the host. As it's budding off, it extracts the lipid bilayer. They look exactly like any other cells. The big problem is that that lipid bilayer, although it's a very protective barrier, completely shields the inside of the virus.

"The lipid bilayer not only shields the viral interior but prevents it from entering cells. HIV has to come up with a whole mechanism by which it takes the lipid bilayer from the target cells it wants to infect. In order to do that, it uses its HIV envelope protein, the gp120, which is the major exterior protein, along with the gp41 fusion protein, which is a transmembrane.

"These are relatively small proteins," Kwong pointed out. "The whole molecular weight for gp120 is 500 amino acids; gp41 has developed 300. They're not really small," he added. "If you wanted to bind two membranes together in your body, you'd have a huge assembly of proteins and everything that does this. The virus is subject to a lot of constraints and one of them is the size of the genome. It doesn't have much genomic size; it can't make the proteins very large for infection. So it uses envelope proteins, the 120 and the 41 that make up the viral spike. The viral spike is not only finding the right receptor but finding the right cell. But then it acts to fuse the viral membrane to the target cell membrane, which allows the virus to enter the target cell."

Direct Test Of Killing Live Virus

"The antibody neutralization test we conducted," Kwong recounted, "is a measure of viral killing, so that you have cells that will live and cells infected by virus. We took live virus and tested whether or not they're infected cells. It's a direct test of killing the live virus. When most antibodies recognize the receptor binding sites on gp120," Kwong went on, "an energetically unfavorable shape change occurs, which is correlated with poor neutralizing activity."

A promising immune system HIV killer is the b12 antibody.

"It's an agent that hasn't been understood since it was pulled out in 1994 by another co-author of this study," Kwong said. "It's a receptor-binding-site antibody, but it's somehow able to neutralize very potently almost all strains of HIV. So what we'd like to get in terms of AIDS vaccines is a lot of antibodies that mimic b12. Of all receptor-binding- site antibodies we looked at, b12 is best able to evade this viral energy barrier. And the reason why it's so potent and good at killing is not only that it binds the receptor binding site but in addition it evades the energy barrier.

"So that's the terms of vaccine design. I find it very satisfying because for 20 years we've known that the virus is able to evade being killed by antibodies. And that's the whole basis of why it can create a persistent infection. We now understand the mechanisms," he concluded, "that it uses in a precise molecular way at the atomic level."