Science Spotlight

Creating antibodies with antibodies

From the McGuire, Taylor, and Stamatatos labs, Vaccine and Infectious Disease Division

Vaccines designed to prevent infections such as HIV and influenza have successfully elicited pathogen-specific antibodies, yet these antibodies fail to protect from infection. Encouragingly, a subset of HIV-1-infected individuals produce broadly neutralizing antibodies (bNab), and vaccine development is focused on reproducing these antibodies in HIV-negative people. However, the production of bNabs is difficult to elicit via vaccines because their germline progenitors do not bind HIV-derived vaccine immunogens, preventing bNab maturation. Traditionally, vaccines are designed to elicit polyclonal antibody responses to an epitope on an ent ire pathogen or large subunit. In a pair of recent papers in the Journal of Experimental Medicine, Vaccine and Infectious Disease Division members Tara Bancroft (formerly of the Taylor and Stamatatos labs) and Pia Dosenovic of the Nussenzweig lab at Rockefeller University, along with the Stamatatos and McGuire labs, along with their VIDD collaborators Roland Strong and Marie Pancera, explored the use of anti-idiotype antibodies in HIV vaccine design.

To test the feasibility of anti-iditoypic antibodies in vaccine design, Bancroft and colleagues selected a bNab called b12 that has been isolated from infected patients. Although b12 is known to successfully neutralize a breadth of HIV viruses, it has yet to be induced via vaccine stimulation, as its germline precursor form does not bind HIV immunogens. To bypass this limitation, the authors created anti-idiotype antibodies against the inferred germline form of b12 (iglb12) by immunizing mice with iglb12 itself. They then isolated the resultant anti-iglb12 antibodies and screened for those that exclusively bound iglb12. To test if these antibodies could bind B cells, the cell type that produces antibodies, with b12-like B cell receptors (BCRs)—the potential b12 germline precursors—in human HIV-negative blood, they incubated the anti-idiotypes with human blood. Their experiment successfully identified B cells specific to iglb12.

Although their technique succeeded, the authors were surprised to find that when they single-cell sorted and sequenced the iglb12-specific B cells, their BCRs only moderately matched the amino acid sequence of iglb12. This relatively low concordance caused the authors to hypothesize that iglb12 may be auto-reactive. To test this, they incubated iglb12 with human liver cells and indeed observed self-recognition. They further confirmed the autoreactivity of iglb12 by generating a transgenic mouse model where all B cells express iglb12-like antibodies. Investigations of these mice found that most B cells were deleted in the bone marrow before maturing, with the remaining B cells become anergic and unresponsive to antigen stimulation. Despite the unresponsiveness, the B cells could be activated by anti-iglb12 antibodies. These experiments suggested that the failure of iglb12 to respond to vaccine antigens, as these antibodies are selected out of the B cell repertoire, but also that anti-idiotype antibodies can be used as immunogens to induce these and other rare germline-precursor B cells to activate and proliferate.

In collaboration, the Nussenzweig, Stamatatos, and McGuire labs performed similar breakthrough work by creating an anti-idiotype antibody binding a different class of bNab called VRC01. VRC01 neutralizes a range of HIV viruses by binding and inhibiting the HIV envelope CD4 binding site that is necessary for the virus to enter a CD4+ T cell. Unlike b12, immunogens have been designed that stimulate germline VRC01, however, previous work from the Stamatatos lab and others, demonstrated that competing non-VRC01 B cells are also activated by HIV-derived antigens, potentially dampening the desired immune response. To increase specificity for VRC01, the authors created an anti-idiotype antibody, iv8, that targets naïve VRC01 precursors. Using a transgenic mouse model, they found that both iv8 and a specially modified HIV-1 derived immunogen could stimulate VRC01 precursor B cells, but that iv8 avoided stimulating undesirable off-target B cells as well. In this study, iv8 also identified B cells with VRC01-like features in human blood, providing another example of the novel uses for anti-idiotype antibodies in HIV vaccine design.

Top: a germline targeting Env can engage bANb precursor B cells (purple), but will also generate off-target responses. Bottom: anti-idiotype monoclonal antibodies will selectively expand bNAb precursor B cells prior to immunization with a germline targeting Env.It will like also elicit off target responses against the antibody (gray), which  would not be Env-reactive and therefore not boosted by Env.
Top: a germline targeting Env can engage bANb precursor B cells (purple), but will also generate off-target responses. Bottom: ai-Mabs will selectively expand bNAb precursor B cells prior to immunization with a germline targeting Env.It will like also elicit off target responses against the antibody (gray), which would not be Env-reactive and therefore not boosted by Env. Figure provided by Andy McGuire.

These two studies demonstrate, for the first time, proof-of-concept that anti-idiotype antibodies can bind and activate bNab HIV-specific germline precursors B cells in vivo. This represents an exciting strategy for vaccine design because it eliminates several of the challenges that have currently halted the HIV vaccine field, including bypassing the need to design immunogens to target germline precursor antibodies and eliminating the non-specific B cells responses that can overwhelm HIV bNab development. Furthermore, unlike vaccine immunogens, anti-idiotype antibodies are very stable, are easily mass produced, and could be administered in concert with multiple anti-idiotypes of different specificities. Although promising, anti-idiotypes would only be the first step in the vaccine design process, as after initial identification and stimulation, the germline-precursor antibodies would need to be further guided into mature bNabs. However, their use as tools to screen rare B cells of interest, as well as act as immunogens themselves, presents an important development in the HIV vaccine field.

 “Previous efforts to develop germline targeting immunogens rely heavily on structure-based design, which require high-resolution crystal structures that reveal the molecular details of how particular antibodies bind to their antigen," McGuire explained. "We are excited because these studies demonstrate that we have potentially found a work around for structure-based design that allows us to target any B cell precursor, not only for HIV-1 but for other pathogens as well."

Dosenovic P, Pettersson AK, Wall A, Thientosapol ES, Feng J, Weidle C, Bhullar K, Kara EE, Hartweger H, Pai JA, Gray MD, Parks KR, Taylor JJ, Pancera M, Stamatatos L, Nussenzweig MC, McGuire AT. 2019. Anti-idiotypic antibodies elicit anti-HIV-1-specific B cell responses. J Exp Med.  pii: jem.20190446. doi: 10.1084/jem.20190446. [Epub ahead of print]

This work was supported by the Bill and Melinda Gates Foundation Collaboration for AIDS Vaccine Discovery, the National Institutes of Health, the Howard Hughes Medical Institute, and the Vaccine and Infectious Disease initiative grant.

UW/Fred Hutch Cancer Consortium members Andrew McGuire and Justin Taylor contributed to this work.

--

Bancroft T, DeBuysscher BL, Weidle C, Schwartz A, Wall A, Gray MD, Feng J, Steach HR, Fitzpatrick KS, Gewe MM, Skog PD, Doyle-Cooper C, Ota T, Strong RK, Nemazee D, Pancera M, Stamatatos L, McGuire AT, Taylor JJ. 2019. Detection and activation of HIV broadly neutralizing antibody precursor B cells using anti-idiotypes. J Exp Med. pii: jem.20190164. doi: 10.1084/jem.20190164. [Epub ahead of print].

This work was supported by the National Institute of Allergy and Infectious Diseases.

UW/Fred Hutch Cancer Consortium members Andrew McGuire and Justin Taylor contributed to this work.