Human CD4-Binding Site Antibody Elicited by Polyvalent DNA Prime-Protein Boost Vaccine Neutralizes Cross-Clade Tier-2-HIV Strains

Abstract

The vaccine elicitation of HIV-neutralizing antibodies with tier-2-neutralization breadth has been a challenge. Here, we report the isolation and characteristics of a CD4-binding site specific monoclonal antibody, HmAb64, from a human volunteer immunized with a polyvalent gp120 DNA prime-protein boost vaccine. HmAb64 derived from heavy chain variable germline gene IGHV1-18, light chain germline gene IGKV1-39, and had a 3^rd heavy chain complementarity determining region (CDR H3) of 15 amino acids. On a cross-clade panel of 208 HIV-1 pseudo-virus strains, HmAb64 neutralized 21 (10%), including tier-2 neutralization resistant strains from clades B, BC, C, and G. The cryo-EM structure of the antigen-binding fragment of HmAb64 bound to a conformation between prefusion closed and occluded open forms of envelope trimer, using both heavy and light CDR3s to recognize the CD4-binding loop, a critical component of the CD4-binding site. A gp120 subunit-based vaccine can thus elicit an antibody capable of tier 2-HIV neutralization.

Keywords: CD4-binding site; DNA immunization; HIV-1; Tier-2 neutralization; envelope glycoprotein; gp120; monoclonal antibody; vaccine.

Figure 1 |. DNA prime and protein boost elicited serum HIV-1 Env specific neutralizing antibody responses targeting the CD4-binding site.

(a) Immunization scheme. Serum samples were collected two weeks after each immunization and PBMCs were collected two weeks after the last immunization. (b) Development of gp120-specific IgG titers following key immunization steps. (c) Neutralization with the volunteer serum at peak level against a panel of pseudotyped viruses expressing primary Env antigens from different subtypes of HIV-1 as indicated. (d) Competition of volunteer sera against CD4bs-specific mAb b12-captured infection to target cells by HIV-1 JR-FL Env-expressing pseudovirus as previously reported . The ability of serum competition is presented as the serum dilutions that can achieve 50% inhibition. Unrelated human serum was used as the placebo control.

Figure 2 |. Vaccine elicited antibody HmAb64 targets the CD4-binding site and neutralizes tier-2 HIV-1 strains.

(a) HmAb64 binding to gp120 Env proteins. HmAb64 showed strong binding to wild type gp120 antigens from multiple HIV-1 isolates but not to their D368R mutants, indicating CD4-binding site specificity. (b) Neutralization analysis. HmAb64 neutralized 21 viruses (10%) from diverse clades of the 208-strain NIAID panel. (c) Neutralization titers and tier designation of VRC panel viruses neutralized by HmAb64.

Figure 3 |. Binding of HmAb64 to autologous or heterologous gp120 antigens.

(a) Binding to either autologous gp120 antigens included in the PDPHV vaccine (left) or heterologous gp120 antigens (right). (b) Surface plasma resonance binding kinetics of HmAb64.

Figure 4 |. Binding of HmAb64 to HIV-1 on cell surface.

(a) Examples of FACS analyses and gating showing the specific binding of HmAb64 to HIV-1 SF162 infected human CD4+ T cells but not to uninfected cells. CD4bs mAb VRC01 was used as the positive control and normal human IgG was used as the negative control. (b) Percentage of HmAb64 binding to SF162 infected CD4+ T cells using cells from three individual human donors.

Figure 5 |. Surface plasmon resonance assay of HmAb64 binding to HIV-1 Env antigens.

(a) Titration of sCD4 (D1D2 domain) binding to various HIV-1 Env antigens (Bal gp120, A244 gp120 and BG505-SOSIP). (b) Binding kinetics of HmAb64 to HIV-1 Env antigens. (c) Binding kinetics of HmAb64 to HIV-1 Env antigens in the presence of increasing amount of sCD4(D1D2). (d) Binding kinetics of VRC01 to HIV-1 Env antigens in the presence of increasing amount of sCD4(D1D2).

Figure 6 |. HmAb64 blocked binding of gp120 Env proteins to human CD4+ T cells in a FACS-based assay.

(a) HmAb64 blocked binding of gp120 to CD4+ T cells. CD4-binding site antibody Leu3A was used as a positive control. (b) HmAb64 blocking is dose dependent. (c) HmAb64 blocked binding of diverse gp120 antigens to CD4+ T cells. Env protein from SIVmac251 was used as a negative control.

Figure 7 |. Cryo-EM structure of HmAb64 in complex tier 2 CNE40 SOSIP Env.

(a) Cryo-EM density of HmAb64 in complex with CNE40 Env in two perpendicular views. The gp120, gp41, N-linked glycans as well as heavy and light chains of bound HmAb64 were colored light gray, dark gray, green, blue and slate blue, respectively. (b) Refined cryo-EM structure of HmAb64 in complex with CNE40 Env. The CD4-binding loop in gp120 was colored orange. (c) Epitope of HmAb64 in CNE40 Env. It is of note that the epitope (blue) located only to one side of the CD4-binding loop (orange). (d) HmAb64 recognized the open-occluded conformation similar to those recognized by antibodies b12-and ab1303. (e) HmAb64 and CD4 recognized the same side of the CD4-binding loop. (f) Key HmAb64 interactions with the CD4-binding loop of gp120. (g) Mapping of HmAb64 epitope on CNE40 gp120. Epitope residues were highlighted in gray shade with those forming hydrogen bonds colored red and those forming salt bridges underlined. Amino acids that were disordered in the cryo-EM structure were shown in light gray font. Secondary structures of the antibody-binding regions were shown for reference. (h) Mapping of hmAb64 paratope. Paratope residues were highlighted in gray shade with those forming hydrogen bonds colored red and those forming salt bridges underlined. Only two SHM residues were involved in direct contact with gp120. The sequences were numbered according to the Kabat nomenclature.