Antibodies produced by the immune system aren’t effective against HIV—and new research potentially explains why. Protein spikes on HIV’s surface are too few and far between for antibodies to adequately latch on to, according to California Institute of Technology researchers reporting their study results in The Proceedings of the National Academy of Sciences and summarized by ScienceDaily.

Antibodies are tiny Y-shaped protein chains that are produced by the B cells and that latch on to protein receptors studding viruses and bacteria. Not only do these antibodies have the ability to neutralize the invader or flag it for elimination by cells of the immune system, but they also help prevent the virus or bacteria from latching on to healthy cells in the body.

Antibodies are Y-shaped for a reason, molecular biologists suggest: It allows each arm of the Y to bind with more than one protein spike on the surface of a disease-causing microorganism. By binding with two arms, rather than one, a single antibody is all the more powerful.

One of the hallmarks of HIV disease—and a major obstacle to vaccine research—is that HIV antibodies don’t work very well in controlling the virus. Past research suggested that HIV’s receptor spikes are coated with a thick sugary substance that antibodies cannot penetrate. The Pasadena, California–based Caltech researchers tested a new theory, that Y-shaped antibodies don’t have the reach necessary to cover receptors that are spread out across HIV’s surface. In turn, the virus escapes being neutralized and is able to go on and infect CD4 cells in the body.

To test this hypothesis, graduate student Joshua Klein, his professor Pamela Bjorkman, PhD, and their colleagues molecularly split two Y-shaped HIV antibodies, called b12 and 4E10, and judged their potency in controlling HIV compared with intact antibodies. Klein’s team found that a two-armed 4E10 antibody was no more effective than a one-armed 4E10 antibody, suggesting that one of the antibody arms wasn’t blocking HIV receptors as it should. They also found that while the two-armed b12 antibody was more effective than a one armed version, it was still far less potent than antibodies to other viruses.

A likely reason for this, the authors suggest, is that HIV typically has only about 15 receptor spikes on its surface, and that these are spread too far apart to be targeted by a single Y-shaped antibody. They compared HIV to a similarly sized flu virus, which typically has 450 receptor spikes within much closer proximity to each other. Klein and his colleagues are encouraging further research to confirm their findings.

Nobel Prize–winning virologist David Baltimore, when asked about the new study, said, “I consider this a very important paper because it changes the focus of the discussion about why HIV antibodies are so poor.”