Posted by Jef Akst

[Entry posted at 19th November 2009]

 

 

Researchers may have finally answered the question of why many antibodies that target the HIV envelope are still unable to stop the virus from spreading -- a troublesome stumbling block in the protracted search for an effective vaccine.

Human Immunodeficiency Virus Image: Wikimedia commons, NIAID

Slight variations in how those antibodies interact with their target on the HIV envelope cause conformational changes in the target molecule that render the antibodies ineffective, according to a study published this week in Science.

This research "helps us to understand the molecular level details of [how certain antibodies bind] with better clarity than before," said viral immunologist David Montefiore of Duke University, who was not involved in the work. "Understanding the atomic level [detail] that they [describe] here is very helpful for designing immunogens to elicit the right kinds of antibodies we want vaccines to elicit."

In the hunt for an effective HIV vaccine, one promising target site is the CD4 binding site on the HIV-1 envelope. The virus uses this binding site, called a viral spike, to enter the CD4 immune cells it infects. In theory, antibodies that bind to and block this site could protect cells from infection. In practice, however, several antibodies identified from HIV-infected individuals do target the viral spike, but for unknown reasons, most of those antibodies cannot bind and therefore fail to stop the virus's spread. Only one -- the b12 antibody -- appears effective, neutralizing about half of circulating viruses.

In circulating HIV, the viral spike includes three gp120 glycoproteins in a trimer formation. Only b12 can bind gp120 as a trimer, but once each gp120 is isolated from the trimer formation, even the ineffective antibodies can bind to it. Theoretically, vaccine efforts could target b12, but the gp120 trimer is not stable enough to be used as an immunogen in a vaccine. As a result, vaccine efforts targeting b12 mostly use the gp120 monomer, which unfortunately elicits a variety of antibodies that target it, most of which are ineffective.

To determine why only b12 can bind to gp120 in its natural form, Peter Kwong, a structural biologist at the Vaccine Research Center of the National Institutes of Health, and his colleagues performed x-ray crystallography on two ineffective antibodies bound to the gp120 monomer -- F105 and b13. They then compared the atomic details of these antibody-gp120 complexes with the details of the b12-gp120 complex.

As predicted, all three antibodies bound to very similar locations on the gp120 glycoprotein, but very slight differences in binding -- on the order of a few angstroms -- caused dramatically different conformational changes between the envelope protein bound to b12 and the proteins bound to other antibodies. The researchers used computer models to reconstruct what these changed gp120 glycoproteins might look like in trimer form and found that those bound to the ineffective antibodies didn't fit -- the new conformations were incompatible with a functional viral spike. These conformational changes, the authors suggest, may explain why the ineffective antibodies fail to bind altogether, thereby failing to neutralize the virus.

Kwong suggested that eliciting such ineffective antibodies may be HIV's way of activating the host immune system -- which it needs to do in order to spread -- without causing a response that will hurt it. It's "part of the whole decoy system HIV uses to turn on the immune system, [but] not in a manner that it could effectively neutralize the virus," he said.

To create an effective vaccine, then, scientists must figure out a way to elicit antibodies that precisely target the viral spike in a way that does not induce these incompatible conformational changes. The existence of b12, a moderately effective antibody, is nature's proof that "it's possible," Montefiore said, "but we just haven't figured out how to do it yet."

One way to force the immune response to precisely target the viral spike may be to play with the glycans that coat the viral spike of a potential vaccine immunogen, Kwong said. Locations with glycans are ignored by the immune system. By trimming back the glycans surrounding the exact site of ideal antibody binding and adding glycans to other parts that should be avoided, researchers could potentially direct the immune system to generate only those antibodies that would be most effective against HIV.

"We don't have a vaccine yet," Kwong said. "[But] this knowledge provides an explanation of what's going on, [and is] another important step in the process of obtaining the vaccine."