Researchers have been trying for nearly 25 years to develop an HIV vaccine, but the development of a vaccine has proven extremely elusive. Yesterday two Caltech scientists presented a report that provides one reason for the difficulties. Many antibodies work by intercepting the virus with a two-pronged receptor before the virus can infect the cell. This is effective for the influenza virus, for instance, because it has 450 "spikes" protruding from it, so the likelihood of two of these spikes bumping into the two prongs of the receptor are fairly high. The Caltech scientists found that using two-pronged receptors was less effective than predicted because an HIV virus might have fewer than 15 spikes--decreasing the likelihood that a receptor will bind to the virus with both prongs. This is important, because when both arms of an antibody bond to a virus, "there can be a hundred to thousandfold increase in the strength of the interaction, which can sometimes translate into an equally dramatic increase in its ability to neutralize a virus."
Another difficulty in designing an effective HIV vaccine is the mutability of the HIV virus. Apparently HIV mutates quickly, so a vaccine would have to produce a wide variety of antibodies to bind with many variations (NY Times, Altman, "Gateses to Finance H.I.V. Vaccine Search").