Modulating the immunodominance hierarchy of immunoglobulin germline-encoded structural motifs targeting the influenza hemagglutinin stem

Abstract

Antibodies targeting epitopes through germline-encoded motifs can be found in different individuals. While these public antibodies are often beneficial, they also pose hurdles for subdominant antibodies to emerge. Here, we use transgenic mice that reproduce the human IGHV1-69^∗01 germline-encoded antibody response to the conserved stem epitope on group 1 hemagglutinin (HA) of influenza A virus to show that this germline-endowed response can be overridden by a subdominant yet cross-group reactive public antibody response. Immunization with a non-cognate group 2 HA stem enriched B cells harboring the IGHD3-9 gene, thereby switching from IGHV1-69- to IGHD3-9-encoded motif-dependent epitope recognition. These IGHD3-9 antibodies bound, neutralized, and conferred cross-group protection in mice against influenza A viruses. A cryoelectron microscopy (cryo-EM) structure of an IGHD3-9 antibody resembled the human broadly neutralizing antibody FI6v3, which uses IGHD3-9. Together, our findings offer insights into vaccine regimens that engage an immunoglobulin repertoire with broader cross-reactivity to influenza A viruses.

Keywords: B cell; CP: Immunology; HA stem; IGHD3-9; IGHV1-69; bnAb; broadly neutralizing antibody; germline-targeting; hemagglutinin; influenza virus; nanoparticle.

Graphical abstract

Figure 1

Characterization of immune responses in IGHV1-69 mice immunized with group 2 HA stem followed by group 1 HA stem (A) Immunization scheme of IGHV1-69 mice. Mice (n = 6) were primed (week 0) with 15 μg of H7ssF followed by two immunizations with H1ssF at weeks 3 and 6. Mice were bled at week 8. (B) Specificity of HA-binding serum antibody measured by ELISA. Mean and SEM are shown for each ELISA curve (n = 6). H1 HA WT, wild-type H1 HA of A/New Caledonia/20/1999; H1 HA Δstem, H1 HA with substitutions in the central stem epitope that knock out antibodies targeting the central stem; H7 HA, wild-type H7 HA of A/Anhui/1/2013. (C) Flow cytometry plot of H1⁺H5⁺ cross-reactive HA-specific B cell population within spleen. Complete gating scheme can be found in Figure S1. (D) Comparison of the CDR H2 amino acid sequences among H1⁺H5⁺ BCRs in IGHV1-69 mice immunized with homologous group 1 (H1ssF × 3, left) and group 2 followed by group 1 stem (H7ssF → H1ssF × 2, right). (E) Comparison of frequency of different reading frames (RFs) of IGHD3-9 gene within H1⁺H5⁺ BCRs in IGHV1-69 mice immunized with homologous group 1 (H1ssF × 3, left) and group 2 followed by group 1 stem (H7ssF → H1ssF × 2, right). (F) CDR H3 core amino acid sequences within H1⁺H5⁺ BCRs in IGHV1-69 mice immunized with group 2 followed by group 1 stem (H7ssF → H1ssF × 2). See also Figures S1 and S2.

Figure 2

HA-binding and virus neutralization profiles of IGHV1-69 + IGHD3-9 mAbs (A) HA-binding profiles of the 9 cross-reactive IGHV1-69 + IGHD3-9 mAbs against group 1 and group 2 HAs by biolayer interferometry (BLI). White indicates no binding (<0.1 nm shift in BLI signal). Pink, salmon, and cayenne indicate low (0.1–0.5 nm shift), medium (0.5–1.0 nm shift), and high binding (1.0–1.5 nm shift), respectively. (B) HA-stem-binding profiles of the 9 cross-reactive IGHV1-69 + IGHD3-9 mAbs against group 1 and group 2 HA stems by ELISA. HA stabilized stem ferritin nanoparticles (ssFs) of H1, H7, and H10 were used as antigens. (C and D) Neutralizing activity of the 9 cross-reactive IGHV1-69 + IGHD3-9 mAbs against group 1 and group 2 viruses. Neutralization activity against H1N1, H5N1, and H2N2 was measured by a reporter-based microneutralization assay (C). Neutralization activity against H7N9 and H10N8 was measured by a pseudotyped lentivirus neutralization assay (D). Group 1 HA-stem-specific (CR6261) and group 2 HA-stem-specific (CR8020) mAbs were used as controls in binding and neutralization assays. See also Table S1.

Figure 3

Cryo-EM structure of IGHV1-69 + IGHD3-9 43_S0008 Fab in complex with HA (A) Overall cryo-EM structure of Fab 43_S0008 in complex with HA. HA trimer is shown in gray and Fab 43_S0008 in green. (B–D) Detailed amino acid interactions of the HA trimer with Fab 43_S0008 light-chain CDR L1 and CDR H3 (B); FR3, CDR H1, and CDR H2 (C); and CDR H3 (D) are depicted in different colors according to the different Fab regions. See also Tables S2 and S3 and Figures S3 and S4.

Figure 4

Structural comparison of 43_S0008 with human IGHV1-69- and IGHD3-9-derived bnAbs (A) HA recognition of IGHV1-69- and IGHD3-9-derived antibodies. HA trimer is colored gray and antibodies are displayed in different colors. The name of the antibody is colored according to the structure and PDB IDs as indicated. (B) Overlay of Fab 43_S0008 with human IGHD3-9-derived FI6v3 and IGHV1-69-derived bnAbs CR9114 and CR6261. (C) Close-up view of key interactions surrounding the W21^b residue in HA2. Comparison of 43_S0008 (green) with human IGHV1-69 bnAbs CR9114 and CR6261 (left) and IGHD3-9 bnAbs FI6v3 and S9-3-37 (right).

Figure 5

Antibody prophylaxis with IGHV1-69 + IGHD3-9 mAb 43_S0008 protects mice from lethal H1N1 and H7N9 influenza virus challenges (A) Experimental design of pre-exposure antibody prophylaxis in mice. BALB/c mice (n = 10) were given 43_S0008, FI6v3, or human IGHV1-69 bnAb at 10 mg kg⁻¹ intraperitoneally 24 h prior to intranasal infection of A/Puerto Rico/8/1934 (H1N1) or A/Anhui/1/2013 (H7N9) virus. The human canonical IGHV1-69 mAbs CR6261 and 315-02-1H01 were used as comparators in the H1N1 and H7N9 challenges, respectively. Control mice received PBS instead of mAbs. (B and C) Changes in body weight (%) and survival curves are shown for H1N1 (B) and H7N9 (C) challenges. Body weight changes are plotted as mean ± SD of each treatment group. Multi-group comparison of Kaplan-Meier survival curves was performed with Mantel-Cox log rank test with Bonferroni correction.

Figure 6

Characterization of IGHD3-9-derived antibodies in humans (A) Frequency of the IGHV1-69 + IGHD3-9 BCR sequences in a total of 22,860,000 sequences collected from three healthy subjects reported in DeWitt et al. (B) Characterization of human IGHD3-9-derived mAbs isolated from the VRC 310 and VRC 315 studies (ClinicalTrials.gov: NCT01086657 and NCT02206464). HA-binding profiles of the IGHD3-9-derived mAbs against various group 1 and group 2 HAs by BLI. White, pink, salmon, and cayenne indicate no binding, low binding, medium binding, and strong binding, respectively. (C) Negative-stain EM two-dimensional (2D) class averages of four representative human IGHD3-9-derived mAbs in complex with H1 HA. (D) Epitope specificity of human IGHD3-9-derived mAbs. HA binding of the IGHD3-9-derived mAbs to WT H1 HA and H1 Δstem HA assessed by BLI. One antibody (310-03-6C_1B10), which was insensitive to Δstem HA, is denoted with square symbols.