Crystal Clear

For a guy who never finished high school, Kwong has done OK for himself: He's now chief of the structural biology section at the National Institutes of Health's Vaccine Research Center. "Peter's work not only directly affects our understanding of the AIDS virus, but it's absolutely critical for designing the most optimal HIV vaccines," says Dan Leahy of the Johns Hopkins University School of Medicine. "That's what I find admirable. His science is absolutely rigorous but is also at the cutting edge of insights into human disease and therapy."

Working with Wet Noodles

Kwong made the choice to take on HIV as a graduate student in Wayne Hendrickson's lab at Columbia University in 1987. "I was in Chicago trying to finish up some bungarotoxin work when Wayne called to talk about some different projects. He said, 'Have you heard of this protein called CD4?'" says Kwong. "I said, 'It rings a bell.' Clearly I didn't know much about it." But Kwong was eager to accept the challenge. "Any protein can be as hard as any other. What's interesting is its biological function. Here Wayne was offering me, right off the bat, something that was of profound biological interest. So I said, 'Yeah, let's go for it. Let's solve the structure.'"

The attitude is typical of Kwong. "He has this enormous sense of optimism," says Larry Shapiro of Columbia, who was a graduate student in the Hendrickson lab at the same time as Kwong. "If a problem looks solvable, Peter has no question that he'll solve it. Failure is not an option."

By 1990, Kwong and Hendrickson, along with graduate student Seongeon Ryu, had nailed the structure of the HIV-binding domain of CD4, a result that landed Kwong his first paper in Nature. The structure was revealing in that "you could see the surface that should bind gp120," he says. Kwong wasn't satisfied, though, because in the handshake that admits HIV into the cell, CD4 is just one hand. So Kwong went after gp120.

Trying to grow crystals of this key viral protein was no easy feat, because HIV and its envelope protein gp120 are masters of disguise. The protein is covered with sugars and many of its parts are conformationally disordered. "So the molecule is flopping around like a bowl of spaghetti," says Shapiro. The sugar coating and the flexibility "cloak it from the immune system and also make it extremely hard to work with and a challenge to crystallize."

"The trick was to identify a crystallizable core," adds Leahy, who was a postdoc in the Hendrickson lab at the time. "That's what Peter did." First, the protein had to be expressed in a eukaryotic system. Second, Kwong had to figure out how to strip off its carbohydrates. Finally, working with Joseph Sodroski at the Dana-Farber Cancer Institute and his then-postdoc Richard Wyatt, Kwong labored to identify a more compact, rigid version of gp120, one that had some of its more flexible bits removed.

"Any one of those approaches requires a lot of work and a leap of faith that if it worked, you'd get crystals," says Leahy. "But to string together these three risky and difficult biochemical strategies sort of made it 'difficult cubed.' It takes an enormous amount of psychological stamina to take those kinds of risks. And Peter's got it."

He would need it.

The gp120-CD4 complex

It would take another four years before Kwong would receive even the tiniest hint of success. "It was 1994 or 1995 and I was thinking about doing a postdoc," says Kwong. "I interviewed at several labs and when I came back, I had these little crystals of gp120 with CD4." Those crystals were not the crystals that ultimately yielded a structure, says Kwong. "But having little crystals meant that everything was basically working, and I just had to figure out a couple other tricks to make it good enough to solve the structure."

Those tricks took a couple more years, which Kwong spent as a postdoc in Hendrickson's lab. Although some people suggested that remaining in one lab for so long might be career suicide, for Kwong the choice was obvious. "At some point you have to think: What do I want to do with my life? Is it more important to do something you're interested in doing or to advance your career? I was more interested in solving and seeing that structure than I was in making the right career moves. And that hasn't worked badly for me."

In 1998, Kwong and his collaborators published two papers in Nature and one in Science detailing the structure of the gp120-CD4 complex and reviewing how gp120 interacted with a neutralizing human antibody and with the chemokine coreceptor. Sodroski credits Kwong and Wyatt for their tenacity. "I'm sure you can imagine many postdoctoral fellows saying, 'This project is going to take more than six months? Forget it. I need papers, I need results,'" he says. "Taking on a difficult project like this takes vision, takes courage, and takes an awful lot of intuition in terms of knowing what things to try. That they ultimately succeeded is a testament to their intelligence and persistence."

"Yet Peter makes it look easy," notes Gary Nabel, director of the Vaccine Research Center. "He's so enthusiastic and makes it seem so simple. But if you look, you can count on less than one hand the number of labs who've been able to crystallize HIV envelope."

Kwong's hard work paid off. "To use a diving analogy where you assign points for difficulty and style," Leahy says, "I'd give this a 10 in difficulty, a 10 in execution, and a 10 in importance. Well, maybe it's an 11 in importance."

"Those initial structures gave us a tremendous amount of information about how gp120 is put together," says Sodroski. Among other things, it allowed them to see, in great detail, exactly how gp120 and CD4 interact -a discovery that Sodroski and others say they hope could lead to the design of drugs that disrupt this fateful coupling, potentially leading to new therapeutics.

On to a vaccine

By the late 1990s, however, protease inhibitors had come along, offering infected individuals a way to stave off AIDS. So rather than chase small molecules that could prevent gp120 from grabbing hold of CD4, Kwong started thinking about how the structure could be exploited to develop a vaccine against HIV. In particular, he imagined using the atomic-level details about what gp120 looks like when it binds to CD4 to design mimics that would evoke a productive immune response. When HIV attaches to CD4, and to its coreceptor, Kwong says, "gp120 really does change shape and exposes different surfaces to the immune system." If researchers could generate a protein fragment that retains that exposed conformation, he says, "we could inject that into a naïve individual and teach that person's immune system: Here's what you want to attack, here's what you want to make antibodies against."

At the Vaccine Research Center, where he opened up shop in 2001, Kwong continues to follow that approach. He has engineered stabilized molecules, although so far none has elicited the production of antibodies that protect animals from infection. He did, however, use these mimics to crystallize a broadly neutralizing antibody called b12, which has opened up a new line of attack for vaccine development. Now Kwong and his colleagues are exploring the part of the virus that such neutralizing antibodies target. After all, these are antibodies that actually work. "If we can teach the body to make b12-like antibodies, you'd be immune to HIV," he says. So Kwong is also working on designing structures that look like the bit of gp120 that b12 recognizes. "We're making the mimics now, so in a couple years, we'll see if these things work. I think it's promising."

It's like nothing that's ever been done before. "This is not like Jonas Salk, who took a polio virus, killed it with formaldehyde, shot it into people's arms and said, let the immune system do the work," says Shapiro. "This goes far beyond that. This is understanding the virus at an atomic level, understanding how it evades the immune system, and making pieces that will effectively stoke an immune response. It's enormously ambitious."

"If this really works, it's going to be very big, obviously if it cures AIDS," he says. "But it's like nothing else. This isn't putting a dead virus in somebody's arm. This vaccine will represent a genuine triumph of the human brain over the virus. If it works."