Science Spotlight

IgG rules HIV mucosal protection

McElrath Laboratory, Vaccine and Infectious Disease Division

Sexual transmission of HIV-1 is through the mucosae of either the genitourinary or gastrointestinal tracts. Because the mucosa is the primary site of infection it’s a good target for HIV prevention by vaccines and immunoprophylaxis. But in order to target prevention in this area a better understanding of mucosal immunology is needed, and specifically, what kinds of antibodies are protective at mucosal portals since most vaccines aim to elicit protective antibodies. In a recent paper published in EBioMedicine, Drs. Astronomo, McElrath (Vaccine and Infection Disease Division), and colleagues evaluated the protective capacity of antibody types in a human ex vivo vaginal infection model and in nonhuman primate (NHP) in vivo rectal transmission models. When asked about the implications of combining ex vivo and in vivo studies in the publication Dr. Astronomo said, “Ex vivo HIV-1 infectivity models utilizing human mucosal tissue explants are a useful tool in HIV-1 prevention research because they bridge between the knowledge gained from discrete in vitro systems and complex in vivo animal models and clinical trials”. The group used a panel of monoclonal antibodies (mAbs) to test multiple antibody isotypes, Fc-receptor mediated functions, non-neutralizing antibody (nnAbs), and broadly-neutralizing antibody (bnAbs) effects. The panel covered both IgG and IgA isotype variants because they are present in the tissues of interest. IgA is typically the main mucosal isotype; however, IgG is the predominant isotype in the lower female genital tract. The IgA antibodies were further divided into subclasses and multimeric forms, allowing a thorough study of antibody isotype effects on HIV infection.

Diagram of human ex vivo explant infection model and an in vivo NHP model to assess the potential of various types of HIV-1 specific Abs to protect against mucosal HIV transmission. Source Dr. Astronomo.

In order to assess the protective potential for Fc-mediated functions, the group first identified the presence of Fc-receptors and effector cells in the mucosa. In order to achieve this, fresh vaginal tissue was dissociated to single cells and analyzed for cellular composition and Fc-receptor (FcR) density. The most abundant leucocytes to express FcRs were macrophages and monocytes (also expresses all FcRs for IgG and IgA) and not natural killer cells, which have generally received more attention for Fc-mediated functions. Once the prevalence of FcRs was confirmed, vaginal explants were used to study the effects of the anti-HIV-1 envelope antibody panel on infection allowing for comparison of bnAbs versus nnAbs and IgG versus IgA. The model allows for HIV infection of human tissue in the presence of tissue resident effector cells and target cells. To model vaccine induced antibodies, HIV was mixed with the individual antibodies then introduced to the explants, as demonstrated in the top panel of the figure. The antibody/controls tested made up 7 groups each containing multiple isoforms of the antibody. The groups were 1) bnAbs against the CD4 binding site 2) nnAbs against gp120 constant region 1 3) nnAbs against gp120 variable loops 1/2 4) nnAbs against gp120 5) nnAbs against gp41 6) isotype controls and 7) antiretroviral controls. When the whole panel was compared, only the bnAbs inhibited infection of the explant, which suggests that in this assay directly blocking viral entry provides more effective protection than non-neutralizing functions. To further evaluate mucosal protection, a NHP model of rectal transmission using SHIV was tested (bottom panel of the figure). NHPs received intravenous (IV) infusions of either of two nnAbs from group 2 above. Even with peak antibody titers in rectal secretions neither antibody was able to protect against rectal HIV transmission, which mirrors the explant results.

Since previous studies have shown some synergistic activities with a combination of antibodies, the vaginal explant model was used to evaluate a nnAb cocktail. Five nnAbs covering three different epitopes were combined and tested against two strains of HIV-1. Similarly to the individual results, the cocktail did not inhibit HIV infection by either strain. To better characterize the protection by bNAbs seen in the explant model and further understand the effects of isoform on mucosal protection, NHPs were used to test the IgG and multiple IgA variants of CH31. After intrarectal instillation of the antibody, the NHPs were challenged with SHIV intrarectally. All animals that received the CH31 IgG isotype were protected while only 50-80% of the CH31 IgA animals were protected. Intravenous infusion of the IgG or monomeric IgA followed by intrarectal challenge yielded similar results. All the IgG-treated animals were protected and only one of six of the IgA-treated animals were. Even though there is natural prevalence for IgA in the gastrointestinal tract, the data suggest there is no advantage for protection by IgA isotype. Of note, results from the NHP study with in vivo intrarectal challenge mirrored those in human vaginal explant ex vivo challenge.  Further work in other mucosae is underway by Dr. Lemos, a researcher from the same group. She is conducting parallel human explant studies to understand the role of neutralizing and non-neutralizing antibodies in the colonic mucosa and the foreskin, which are also sites of HIV transmission. Together, they suggest the potential benefit of neutralizing antibodies in mediating protection at male and female genital sites, as well as the rectal compartment.

Overall, this study aimed to compare IgG versus IgA and bnAb versus nnAb in both an ex vivo and in vivo mucosal model.  Using the models presented here, the field can gain valuable insight into antibody-mediated mucosal protection against HIV-1 more rapidly without the wait for efficacy trial outcomes. By using their models, the group showed that IgG bnAbs were superior at protecting at the mucosal surface. When asked what is next, Dr. Astronomo said, “In HIV Vaccine Trial Network (HVTN) 116 study, men and women will receive intravenous bnAb infusions (VRC01/VRC01LS), followed by ectocervical, vaginal and rectal biopsy collections to assess both mucosal bnAb levels and functionality, the latter through ex vivo HIV-1 challenge of these biopsies”.  

This work was funded by the Bill and Melinda Gates foundation and the National Institute of Allergy and Infectious Diseases (NIAID).