Immunization with germ line-targeting SOSIP trimers elicits broadly neutralizing antibody precursors in infant macaques

Abstract

Adolescents are a growing population of people living with HIV. The period between weaning and sexual debut presents a low-risk window for HIV acquisition, making early childhood an ideal time for implementing an immunization regimen. Because the elicitation of broadly neutralizing antibodies (bnAbs) is critical for an effective HIV vaccine, our goal was to assess the ability of a bnAb B cell lineage-designed HIV envelope SOSIP (protein stabilized by a disulfide bond between gp120-gp41-named "SOS"-and an isoleucine-to-proline point mutation-named "IP"-at residue 559) to induce precursor CD4 binding site (CD4bs)-targeting bnAbs in early life. Infant rhesus macaques received either a BG505 SOSIP, based on the infant BG505 transmitted/founder virus, or the CD4bs germ line-targeting BG505 SOSIP GT1.1 (n = 5 per group). Although both strategies induced durable, high-magnitude plasma autologous virus neutralization responses, only GT1.1-immunized infants (n = 3 of 5) exhibited VRC01-like CD4bs bnAb precursor development. Thus, a multidose immunization regimen with bnAb lineage-designed SOSIPs shows promise for inducing early B cell responses with the potential to mature into protective HIV bnAbs before sexual debut.

Fig. 1.. Development of durable, high-magnitude Env-specific plasma IgG responses after BG505 SOSIP immunization.

(A) Infant RMs received 50 μg of either BG505 SOSIP or germ line-targeting BG505 GT1.1 (n = 5 per group) with the 3M-052-SE adjuvant. (B) Plasma IgG binding responses were measured to the matching immunogens BG505 SOSIP (group 1; blue) or BG505 GT1.1 (group 2; green) for each group by ELISA. A bold line represents the median, and a shaded area represents the range. Black arrows indicate immunization time points. (C) AUC analysis for vaccine-elicited plasma IgG binding responses. Individual data points are empty circles. The group mean weekly AUC value is a filled diamond. The median represents the midpoints of the box, with maximum and minimum values used to create the arms. The median AUC responses were similar between the two groups (P = 0.23).

Fig. 2.. Autologous virus neutralization, recognition of infected cells, and ADCC-mediating vaccine-elicited antibody responses in BG505 SOSIP-immunized infant RMs.

(A to D) HIV-1 neutralization in sera was assayed against the BG505 GT1.1 tier 1 virus [(A) and (B)] or BG505.T332N tier 2 virus [(C) and (D)]. (E) Sera at week 80 was assayed against a panel of mutant viruses to map nAbs targeting GHs on the BG505 Env. Specificity was assigned on the basis of a >three-fold reduction in ID₅₀ to the mutant virus compared with the parental virus, nt, not tested; KI, knock-in mutants (F) Plasma-infected cell IgG binding expressed as the percentage of BG505-infected cells with IgG binding. (G) Plasma ADCC activity against BG505 gp120-coated target cells. The plasma dilution endpoint Ab titers 2 weeks after the fourth, fifth, and sixth immunizations for each animal are shown. In dot plots, horizontal lines represent median responses. For all graphs, BG505 SOSIP-immunized infants are in blue, and BG505 GT1.1-immunized infants are in green. Luc, luciferase.

Fig. 3.. BG505 GT1.1 immunization elicits CD4bs-targeting binding antibodies in infant RMs.

A multiplex bead-based assay was used to assess the specificity of vaccine-elicited plasma IgG binding for the CD4bs epitope by evaluating binding against BG505 GT1.1 and 426c.DM.RS.Core gp120 and their corresponding CD4bs KO antigens BG505 G11.1 N279A.D368R and 426c.DM.RS.KO gp120 in BG505 SOSIP (A)- and BG505 GT1.1 (B)-immunized infant RMs. (C and D) Plasma IgG binding avidity scores as determined by SPR against the 426c.DM.RS.Core gp120 protein (C) and the corresponding CD4bs KO mutant (D). FP, fusion peptide.

Fig. 4.. Visualization of polyclonal antibodies in plasma reveals that CD4bs-targeting antibodies were elicited in the majority of BG505 GT1.1-primed infants.

nsEMPEM of either BG505 GT1.1 (A) or BG505 SOSIP (B) in complex with Fabs derived from plasma IgG at week 80. The neutralization ID₅₀ titer against BG505 GT1.1 (tier 1 virus) and BG505.T332N (tier 2 virus) is shown in the table below each composite figure for each animal immunized with either BG505 SOSIP (blue animal IDs) or BG505 GT1.1 (green animal IDs). Colors indicate previously defined epitopes on the BG505 SOSIP Env and are described in the key. (C and D) Summary of the proportion of epitope targets identified for each animal by nsEMPEM analysis for BG505 GT1.1- and BG505 SOSIP-immunized infant RMs.

Fig. 5.. Three BG505 GT1.1-primed infants developed CD4bs-specific precursor bnAbs.

Serum was screened for the development of VRC01-ike CD4bs precursor bnAb responses using the 426c.TM/293S/GnT1⁻ virus, which engages germ line-reverted forms of VRC01 class bnAbs, and the CD4bs KO virus, 426C.TM.D279K/293S/GnT⁻. Neutralization ID₅₀ titers (A) and fold decreases in ID₅₀ against the 426c parental and mutant viruses (B) are summarized in heatmaps from serum screened at 2 weeks after the fourth, fifth, and sixth immunizations. (C) Kinetics of the development of the CD4bs bnAb precursor signature in plasma of BG505 GT1.1-immunized infant RMs. (D) Three BG505 GT1.1-primed infants exhibited heterologous neutralization in plasma 2 weeks after the fifth and sixth immunizations. Neutralization ID₅₀ titers are summarized in the heatmap.

Fig. 6.. Characterization of mAbs isolated from vaccine antigen-specific B cells from BG505 GT1.1-primed infant RMs.

BG505 GT1.1-specific B cells were single-sorted from infants 8229, 8234, and 8239, and lg VH and VL chains and Vk and VI regions were amplified and sequenced. Paired VH and VK/VI sequences were obtained for 26 antibodies, and the proportion of each family is shown (A). Frequency of VH SHM (B) and length of CDR3 among VH (C) and VL (D) genes of the isolated mAbs. aa, amino acids. (E) Heatmap summarizing IC₅₀ of two mAbs with autologous virus neutralization and CD4bs bnAb precursor neutralization capacity in comparison with HIV-1 bnAbs VRC01 and CH31. (F) Tier 2 autologous virus-neutralizing mAb 220_Rh was assayed against a panel of mutant viruses to map epitope targets of neutralization activity. Specificity was assigned on the basis of a >threefold resistance to neutralization of the mutant virus compared with the parental virus. (G) Negative-Stain EM epitope mapping of isolated rhesus mAbs to BG505 GT1.1 demonstrating binding to the fusion peptide region (orange) or the CD4bs (blue). (H) Heatmap summarizing neutralization IC₅₀ against PVs of the global HIV neutralization panel.

Fig. 7.. Comparison of neutralization responses elicited by BG505 SOSIP immunization in infant versus adult RMs.

(A) Immunization schedule for adult RMs who received either BG505 SOSIP or germ line-targeting BG505 GT1.1 SOSIP trimer (n = 5 per group) with the 3M-052-alum. Neutralization titers were compared between infant RMs at week 54 and adult RMs at week 39 against the BG505 GT1.1 (B) and 365505.T332N (C) PVs. In the dot plots, horizontal lines represent medians. (D) Serum was screened for the development of VRC01-like CD4bs-specific precursor bnAb responses among BG505 GT1.1-immunized infant and adult RMs. Neutralization ID₅₀ titers are summarized in the heatmap for each animal. (E) Fold decrease in ID₅₀ against 426c parental and mutant viruses is summarized in a heatmap. (F) nsEMPEM composite figure of two BG505 GT1.1-primed adult RMs with and without the CD4bs-specific bnAb precursor signature in plasma. Previously defined epitopes on the BG505 SOSIP Env are color coded.