HIVe Home
Posted: 13 Jan 2012
Journal Club: Antibodies – what’s the use?

The past several years brought us a wealth of new broadly-neutralizing antibodies (bnAbs), many of which are orders of magnitude more potent and many times more broad than the classical bnAbs, such as b12 or 2G12. How can we use this wealth to prevent or treat HIV infections?

Advances in crystallography and in silico protein folding promise that eventually we will be able to rationally design immunogens that would elicit bnAbs in vaccinated individuals. However, so far this approach hasn’t been proved effective.  Researchers behind the three papers reviewed this week are getting impatient and are moving forward by borrowing a page from the old-school book on immunoprophylaxis – passive immunization.

In passive immunization, antibodies are not elicited in the body in response to an antigen, but are directly injected into a person. David Baltimore’s group at the California Institute of Technology put a new twist on the old approach by combining it with gene therapy (Nature. 2011 Nov 30;481(7379):81-4). They inserted a gene coding for the VRC01 bnAb (or another bnAb) into an adeno-associated virus (AAV) vector and used it to transform muscle cells in mice (Fig. 1). AAV vectors, just like retroviral vectors, integrate into the host genome, which allows stable expression of the transgene for prolonged periods of time. Yet AAV vectors are considered to be less risky than retroviral vectors, because they tend to integrate in a specific place in the host genome, far away from the genes whose disruption may cause cancer. Baltimore’s group observed that Ab concentration in plasma was dose-dependent on the amount of vector administered, and, in all cases, expression was stable for 52 weeks after gene therapy treatment.


Figure 1. Vectored immunoprophylaxis as envisioned by Baltimore et al. AAV vector carrying Ab gene is used to transform fibroblasts. After integration the vector expresses a polypeptide, which is cleaved by the autocatalytic 2A peptide into the heavy (HC) and the light (LC) chains of antibody. The antibody neutralizes the virus after it crosses vaginal epithelium. In the paper, the virus was introduced by intraperitoneal injection. 

The genetically-modified mice used in these experiments lacked their own immune system and were humanized by transplantation of human PBMCs, which made them susceptible to HIV. This allowed the researchers to test whether expression of bnAbs would provide protection against HIV infection. Indeed, the mice injected with AAV containing a bnAb-gene were protected against the challenge with very high doses of HIV, in some cases 100-fold higher than the dose necessary to infect mock-treated animals. However, a caveat to the observed protection is the somewhat unnatural challenge mode in these experiments. The virus was introduced into the peritoneal cavity and, therefore, it was likely to encounter high concentration of bnAbs before it had a chance to establish infection. During sexual transmission, however, the virus is more likely to establish infection in mucosal tissues where concentration of IgGs may be insufficient to provide sterilizing immunity, which may give the virus a chance to evolve resistance to bnAb.

Burton et al. (PNAS July 5, 2011 vol. 108 no. 27 11181-11186) attempted to specifically target the mucosal compartment by applying antibodies topically. They intravaginally injected a high dose of bnAb b12 into macaques and then challenged them 30 minutes later with SHIV-162P, a chimeric SIV virus carrying the HIV envelope protein. The bnAb-treated animals were protected, while the animals injected with placebo became infected. However, similar to the Baltimore study, it is likely that the observed protection was due to the virus being neutralized in the vagina, before it had a chance to enter vaginal mucosa and establish infection there.

When Burton et al. used in the similar manner the b6 antibody, which binds to the virus but does not neutralize it, the protection disappeared (Figure 2). The authors concluded that future attempts to generate Ab-based protection should focus on neutralizing antibodies instead of binding antibodies. Yet, the lack of protection observed with the non-neutralizing antibody could be specific for this particular antibody or due to the fact that it was not present in the right place. The following article, although it deals with a different virus, shows that binding antibodies could have an effect on viral infection and proposes a model to explain it.


Figure 2. Monkeys intravaginally inoculated with either neutralizing antibody b12 (A) or  poorly neutralizing antibody b6 (B) were intravaginally challenged with SHIV. Only b6-treated monkeys became infected as evidenced by the appearance of the virus in blood, while b12-treated monkeys were protected (from Burton et al.).

The study by Vogt et al. (J Virol. 2011 Nov;85(22):11567-80) looked at the effect of passive immunization on infection with West Nile virus in mice. The antibody E28 that they used was poorly neutralizing in vitro. But when it was administered to mice by the intraperitoneal route, it protected the animals from lethal challenge with the virus (Figure 3). Using genetically modified mice as well as mutated versions of the antibody, the researchers showed that the antibody-dependent protection required complement protein C1q, Fc receptor CD16 and the presence of phagocytic cells. These results may explain why the non-neutralizing antibody had no effect in the Burton study — it had to be present in the tissues in order to initiate the Fc-dependent responses.


Figure 3. Antibody E28 does not neutralize West Nile Virus (WNV) in vitro, but protects mice. A) Effect of several antibodies on infectivity of WNV in cell culture. E28 antibody has no effect. B) Passive immunization with E28 antibody lower the viral loads in the brains of mice infected with WNV. C) Passive immunization with E28 antibody improves survival of mice infected with WNV. The 297Q mutation in the heavy chain of the antibody eliminates binding to C1q and Fc receptors – the mutant antibody has no effect on survival (from Vogt et al.).

In conclusion, multiple avenues remain open for exploration of the utility of both neutralizing and non-neutralizing antibodies. While basic research on passive immunization provides answers about the possible mechanisms of protection, gene therapy is moving forward with a translational approach that may show results sooner than some of us are expecting.

 About the author: Yegor Voronin is a Science Officer at the Global HIV Vaccine Enterprise.

Journal Club discusses scientific literature to raise awareness of the recent advances in HIV vaccine research and inspire discourse important to advance HIV vaccine research and development.  Visit the Submission Guidelines if you are interested in submitting a journal club post.