Study offers new approach to comparing antibody function

Designing vaccine candidates to elicit broadly neutralizing antibodies (bNAbs) has proved a major challenge. 

A minority of HIV-infected people generate such antibodies, but the average time for developing such responses is about 2½ years and only a small fraction of bNAbs protect against a broad range of HIV’s many different subtypes. The long—some say tortuous—development of bNAb responses entails a sort of continuous massaging of the immune system by the infecting virus.

This drives the somatic hypermutation of antibody genes that ultimately boosts the potency and breadth of the resulting bNAbs (see Vaccines to Antibodies: Grow Up, IAVI Report July-Aug. 2010). The antibodies induced thus far by AIDS vaccine candidates are, by comparison, far less mutated, which probably explains why they are far less potent neutralizers of HIV. 

Still, scientists believe it may be possible to reverse engineer immunogens based on the growing roster of isolated bNAbs. A clearer understanding of how antibodies induced by vaccination function differently from those that arise during infection could go some way toward achieving this goal. But the kinds of preclinical studies that could illuminate these differences are inherently difficult to do for HIV. There are at the same time few animal models suitable to the evaluation of human vaccines, in part due to differences in the genetics of the human and simian immune systems. 

A recently published study, however, might have found a partial solution to this conundrum. It describes a model in which vaccine-induced B cell responses in nonhuman primates (NHPs) might be compared to such responses to HIV infection in humans.  The study, led by PhD student Christopher Sundling  and his advisor Gunilla Karlsson Hedestam, a professor in the Department of Microbiology, Tumor and Cell biology at Karolinska Institute in Sweden, compared  vaccine-induced monoclonal antibodies elicited in rhesus macaques that target the HIV Env CD4 binding site (CD4bs) with infection-induced mAbs against CD4bs isolated from individuals with chronic HIV (Sci. Transl. Med. 4 142ra196, 2012). 

The research team, which also included professor Yuxing Li at the Scripps Research Institute (TSRI) and Richard Wyatt, professor of immunology at TSRI and director of viral immunology at IAVI’s Neutralizing Antibody Center, focused its analysis on the CD4bs because it is an exposed and highly conserved site. Many of the bNAbs isolated in HIV-infected individuals target this epitope and the current crop of Env immunogens included in AIDS vaccine candidates appear to elicit anti-CD4bs antibodies that display only a limited potency and breadth of neutralization. 

“We have now established an approach to do these analyses in NHPs, which is critical as the field needs an experimental system to study vaccine-induced antibody responses in a systematic way,” said Karlsson Hedestam. 

Before conducting their comparative analysis of antibodies, the researchers first defined the heavy-chain and light-chain Immunoglobulin (Ig) germline sequences in rhesus macaques and compared these to known human counterparts. 

They then examined V(D)J gene usage—the genes that developing B cells shuffle to generate a diversified portfolio of antibody specificities—in total, Env-specific and CD4bs-specific memory B cells, and found a similar distribution in these populations. Intriguingly, the scientists found an increase in the average length of complementary determining region 3 (CDR3)—the most variable part of the antibody—in the CD4bs-specific B cell population when compared to the total memory or total Env-specific cells.  

The researchers then isolated and characterized eight clonally distinct anti-CD4bs mAbs in the NHPs using site-specific Env sorting probes and compared them to a panel of infection-induced CD4bs-directed mAbs, including VRC01 and b12. Epitope mapping suggested that vaccine-induced mAbs bind close to the axis of the Env trimer, while the infection-induced mAb in humans, VRC01, binds further out on gp120’s outer domain, suggesting possibilities for the design of improved immunogens. 

“If we can restrict access to the Env spike,” said Karlsson Hedestam, “and force selection of antibodies that recognize the functional Env spike more from the side, as the infection-induced bNAbs appear to do, and less from the top, as the vaccine-induced mAbs do--that may result in a response that is more broadly neutralizing.” 

 Although none of the vaccine-induced mAbs neutralized less sensitive tier 2 viruses using the Tzm-bl assay, one of the mAbs—called GE148—caught the eye of researchers. It was found to display weak binding to a variant of the gp120 protein bound by bNAbs but not ordinary anti-HIV antibodies. Karlsson Hedestam said these results suggest that, with further affinity maturation, the GE148 antibody might possibly mature into a better neutralizer.  

Mutating antibodies 
Having established that the B cell sorting probes were specific, many more CD4bs-directed antibodies could be analyzed by sequence. The team  found that the vaccine-induced mAbs were modestly mutated—between 5% and 6%—compared to the closest matching germline sequence among the macaques used in the study.
 
Somatic hypermutation (SHM) is triggered in B cells when their receptors bind antigen and results in the generation of antibodies that have extremely high affinity to their epitopes—a process referred to as affinity maturation. The study doesn’t tell us how HIV vaccine designers might accelerate affinity maturation, but Karlsson Hedestam noted that it does provide tools to better analyze the process. 

“We can see that our vaccine-induced mAbs have the same level of SHM as seen in the general memory B cell population in the same animals and this is known to be similar in the human memory B cell pool,” she said. “Thus, we do not drive SHM poorly; rather it is the chronic infection that drives such extreme levels of SHM in HIV infection, levels that we are not approaching with current vaccine regimens. With the experimental approach described [in this study] I think we are now in a position to identify vaccine approaches that improve upon current approaches.”
 
Nancy Haigwood, director of the Oregon National Primate Research Center at the Oregon Health and Science University, who was not involved in the study, agrees that the approach used in the study offers a novel way to understand and potentially compare different vaccine approaches that are designed to elicit neutralizing antibodies. “The main caveat to relying upon such studies is that they are resource-intensive,” she said.