Until recently, many scientists were convinced that antibodies capable of preventing HIV infection could not be elicited through vaccination. But the discovery of dozens of potent broadly neutralizing antibodies (bNAbs) and the elucidation of some of their structural targets on HIV’s surface protein—the Envelope glycoprotein trimer that all bNAbs target—have revealed weaknesses that researchers believe can be exploited for both drug and vaccine development.

But the road ahead contains several Olympic-sized hurdles. A number of these challenges center on the immunogens—whole HIV proteins or pieces of protein called epitopes that are added to a vaccine candidate that stimulate the immune system to induce the desired type and amount of immune responses. The idea is that the more focused the immunogen is the more effective the antibody response will be. Yet one worry is that the immunogens will induce only strain-specific antibodies rather than the cross-reacting broadly neutralizing antibodies that are considered essential for a protective sustained response.

To get around this problem, scientists have been using computational tools to identify proteins that can, molecularly speaking, lock in place those epitopes on the surface of HIV that the neutralizing antibody can bind. Because they are essentially freezing the epitope in place these computationally derived protein structures are referred to as scaffolds.

At this week’s AIDS Vaccine 2012 conference, Bill Schief, an associate professor of immunology at The Scripps Research Institute in La Jolla, California, and a member of IAVI’s Neutralizing Antibody Center based at Scripps, discussed work that his lab and others have been doing with protein scaffolds that, after some initial disappointments, seems to have turned the corner. The work being led by Schief’s lab and which involves collaborators at the Vaccine Research Center at the US National Institute of Allergy and Infectious Diseases, involves a different disease, but the findings have implications for AIDS vaccine design.

Schief has been using the respiratory syncytial virus (RSV)—a major cause of respiratory tract infections in infants—as a test case for how to computationally design immunogens that elicit the desired protective antibodies. Schief chose this virus because while there is no vaccine for RSV, the monoclonal antibody (mAB) palivizumab, which targets an epitope on RSV, is prescribed for infants to prevent serious RSV illness in premature infants or those with a respiratory disorder called bronchopulmonary dysplasia.    

Several years ago, Schief’s lab began experimenting with motavizumab, an experimental mAB developed but never approved as an alternative to palivizumab, to try and design epitope scaffolds for RSV. Unfortunately, the scaffolds failed to elicit any neutralizing antibodies in immunized mice.

“So that was depressing,” said Schief. “But we didn’t want to give up. We thought we could make better scaffolds.”

Schief’s lab then switched to a different computational method to design their epitope scaffolds, ending up with six “well-behaved” proteins that seemed stable, had decent secondary structures and bound much more tightly to motavizumab than the epitopes in the previous experiment did. Schief said by freezing the conformation of the epitope, they were able to achieve tight binding.

Next, they immunized rhesus macaques five times over an 18-week period. By the 20th week many of the animals were making very potent neutralizing antibodies, said Schief. “We were very excited because all our work on scaffolds never produced any neutralizing antibodies and everyone would rightfully say maybe scaffolding is never going to work,” he said. “This experiment showed it actually can work.”

Schief said the next step is to try and determine what turned on the mAbs in the animals.