Towards an AIDS Vaccine:
TSRI Scientists Describe Unusual Antibody That Targets HIV

By Jason Socrates Bardi

A group of scientists from The Scripps Research Institute (TSRI) and several other institutions has solved the structure of an antibody that effectively neutralizes human immunodeficiency virus (HIV), the virus that causes acquired immunodeficiency syndrome (AIDS).

The antibody binds to sugars on the surface of HIV and effectively neutralizes the virus because of its unique structure, which is described in the latest issue of the journal Science.

"What we found was an unusual configuration of the antibody in which its two Fab domains—the antigen recognition units—are 'interdigitating' with each other," says TSRI Professor Ian Wilson, one of two TSRI professors who led the research. "Nothing like this has ever been seen before."

This new structure is an important step toward the goal of designing an effective vaccine against HIV, and it gives the researchers a new way to design antibodies in general.

"It may enable us to make antibodies that recognize whole new sets of molecules," says TSRI Professor Dennis Burton, the other TSRI investigator who led the research.

The Problem of HIV and Antibodies

HIV causes AIDS by binding to, entering, and ultimately killing T helper cells, which are immune cells that are necessary to fight off infections by common bacteria and other pathogens. As HIV depletes the body of T helper cells, common pathogens can become potentially lethal.

The latest statistics are grim. The World Health Organization estimates that around 40 million people are living with HIV worldwide. During 2001 alone, more than four million men, women, and children succumbed to the disease, and by the end of that year, the disease had made orphans of 14 million children. In the United States, 40,000 people are infected with HIV each year. One of the most compelling medical challenges today is to develop a vaccine that will provide complete prophylactic protection to someone who is later exposed to this virus. An important part of such a vaccine will be a component that elicits or induces effective neutralizing antibodies against HIV in the blood of the vaccinated person.

Also called immunoglobins, antibodies are the basis for many existing vaccines, including those against measles, polio, hepatitis B, and hepatitis A. HIV antibodies are produced by the body's B cells after HIV enters the bloodstream. During such an immune response, the antibodies circulate through the blood. Good antibodies bind to and "neutralize" the virus, making it unable to invade cells. Because neutralizing antibodies attack the virus before it enters cells, they could conceivably be used to prevent HIV infection if they were present prior to virus exposure. A vaccine would seek to elicit these neutralizing antibodies.

This is easier said than done. The body makes lots of antibodies against HIV, but they are almost always unable to neutralize the virus. Much of the viral surface is coated with carbohydrates (sugars), which are hard for the immune system to attack because these sugars are made by human cells and attached by human proteins. In other words, they are "self" and should not be recognized by antibodies.

Interlocking Arms

However, in rare instances some people have produced antibodies that broadly neutralize HIV. One such antibody, called 2G12, was isolated from such an HIV-positive individual about a decade ago by Hermann Katinger, a doctor at the Institute for Applied Microbiology of the University of Agriculture in Vienna, Austria and one of the authors on the paper. This antibody is not like ordinary antibodies.

"The Fab [antigen recognition] arms are interlocked," says Burton. "That is a unique arrangement, and it is good for recognizing a cluster of shapes like sugars on a virus."

The 2G12 antibody forms an unusual "dimer" interface where two antibodies create an unusual multivalent binding interface with multiple binding sites that recognizes an unusual arrangement of 2-3 "oligomannose" sugars on the surface of protein spikes called gp120 that decorate the coat of HIV. This allows the antibody to properly target HIV virions as foreign pathogens. The sugars are human but their arrangement is foreign—and it is this arrangement that the antibodies recognize.

These results are a step in the direction of designing an effective AIDS vaccine because it reveals what these neutralizing antibodies can look like. The next step is to use the structure of the antibody as a template to design an "antigen" that would stimulate the human immune system to make 2G12 or similar broadly neutralizing antibodies against HIV.

The results are also important because the structure of the antibody is something that has never been seen before. "Can we now," asks Wilson, "use this [knowledge] to engineer antibodies with higher affinity against other antigens or clusters of antigens?"

The TSRI study combined experts from several institutions in addition to those at TSRI, including Pauline M. Rudd, and Raymond A. Dwek from the Glycobiology Institute at Oxford University in the United Kingdom. Also involved were researchers in the Department of Biological Science and Structural Biology at Florida State University in Tallahassee.

The research article, "Antibody Domain Exchange is an Immunological Solution to Carbohydrate Cluster Recognition" is authored by Daniel A. Calarese, Christopher N. Scanlan, Michael B. Zwick, Songpon Deechongkit, Yusuke Mimura, Renate Kunert, Ping Zhu, Mark R.Wormald, Robyn L. Stanfield, Kenneth H. Roux, Jeffery W. Kelly, Pauline M. Rudd, Raymond A. Dwek, Hermann Katinger, Dennis R. Burton, and Ian A. Wilson and appears in the June 27, 2003 issue of the journal Science.

The research was supported by The Skaggs Institute for Research, which funds The Skaggs Institute for Chemical Biology at TSRI. Grants from the National Institute of Allergy and Infectious Diseases (NIAID), the National Institute of General Medical Sciences (NIGMS), and the International AIDS Vaccine Initiative (IAVI) also supported the research.


Novel Architecture of Antibody 2G12

A) Overall structure of a "typical" Fab monomer, with the light and heavy chains in grey and purple, respectively. B) Structure of the 2G12 Fab monomer. The heavy chain clearly separates from its usual interaction with the light chain. The monomer does not exist in the crystal, but only in the context of the domain-swapped dimer. C) Structure of the two domain-swapped Fab molecules, as they assemble in the crystal. The light chains are shown in grey, with the heavy chains shown in blue and purple.