Shape-shifting HIV has many tricks for evading the immune system, which makes developing a vaccine particularly challenging. But researchers are beginning to gain ground: independent teams have solved the structure of two key proteins, providing crucial information about the AIDS-causing virus.

Although the virus stimulates the immune system to produce many antibodies, it has a highly variable protective membrane that allows it to evade the attack and survive in the body long term. This causes a major problem for developing an HIV vaccine.

And this is why candidate vaccines designed to stimulate the production of antibodies against HIV have failed. Today, the vaccines furthest along in clinical trials all work by inducing a cellular immune response instead, which is useful but only effective once the virus has already infected cells.

A broadly neutralizing antibody response is the holy grail of HIV vaccine development. James Bradac , NIAID

However, some patients who have lived with the virus for a long time produce 'broadly neutralizing' antibodies, also called superantibodies, which seem to be effective against many strains of HIV.

"Coming up with a vaccine that induces a broadly neutralizing antibody response is the holy grail of HIV vaccine development," says James Bradac, a virologist and chief of preclinical research and development in the division of AIDS at the National Institute of Allergy and Infectious Diseases in Bethesda, Maryland.

Broad effect

Dennis Burton and Ian Wilson, immunologists at The Scripps Research Institute in La Jolla, California, have looked at 4E10, the most broadly acting HIV antibody known so far. They have worked out the structure that it has when it is bound to gp41, the protein (or antigen) that it recognizes on the virus's surface, and they have published their results in Immunity1.

The pair hope to use the information to design a vaccine that will stimulate the production of antibodies like 4E10. "We can make a mimic of the antigen that will elicit the same type of antibodies we initially studied," says Wilson. He and Burton call the approach retrovaccinology.

Protein unbound

The structure of the second protein is published in this week's Nature2 by researchers led by the structural biologist Stephen Harrison of Harvard Medical School, Boston. They reveal the crystal structure of the virus surface protein gp120 from the simian immunodeficiency virus, which is closely related to HIV.

The researchers studied the structure of the protein as it is before it binds to a helper T cell, a type of immune cell that HIV infects. The bound structure was solved several years ago, so the new information helps to show how the molecule changes shape when it recognizes and binds to the cell.

"Each one of these studies gives us a better understanding of how HIV infects a cell and hence how we might stop that happening," says Bradac.