HCV Broadly Neutralizing Antibodies Use a CDRH3 Disulfide Motif to Recognize an E2 Glycoprotein Site that Can Be Targeted for Vaccine Design

Abstract

Hepatitis C virus (HCV) vaccine efforts are hampered by the extensive genetic diversity of HCV envelope glycoproteins E1 and E2. Structures of broadly neutralizing antibodies (bNAbs) (e.g., HEPC3, HEPC74) isolated from individuals who spontaneously cleared HCV infection facilitate immunogen design to elicit antibodies against multiple HCV variants. However, challenges in expressing HCV glycoproteins previously limited bNAb-HCV structures to complexes with truncated E2 cores. Here we describe crystal structures of full-length E2 ectodomain complexes with HEPC3 and HEPC74, revealing lock-and-key antibody-antigen interactions, E2 regions (including a target of immunogen design) that were truncated or disordered in E2 cores, and an antibody CDRH3 disulfide motif that exhibits common interactions with a conserved epitope despite different bNAb-E2 binding orientations. The structures display unusual features relevant to common genetic signatures of HCV bNAbs and demonstrate extraordinary plasticity in antibody-antigen interactions. In addition, E2 variants that bind HEPC3/HEPC74-like germline precursors may represent candidate vaccine immunogens.

Keywords: E2; HCV; broadly neutralizing antibodies; disulfide bond; epitope; glycoprotein; structure.

Figure 1.. The shared CDRH3 motif in E2 front layer-specific bNAbs adopts different orientations.

(A) Sequence alignment of the recombined antibody heavy chain (top) and light chain (bottom) variable gene sequences of HEPC3, HEPC74, and AR3C. CDR loops were defined based on IMGT nomenclature and colored blue (CDRH1), orange (CDRH2), and red (CDRH3). Dots indicate identical amino acids; dashes indicate gaps. (B) V_H, D, and J_H gene usage for HEPC3, HEPC74, and AR3C. (C) Sequence alignment of CDRH3s from mature HEPC3 and HEPC3 germline precursor genes determined by IMGT/V-QUEST. Dots indicated identical amino acids and dashes indicate regions encoded by other gene segments or Nnucleotide additions. Two cysteines encoded by the D gene segment are highlighted in bold. (D) Side (top) and top (bottom) views of Fab structures of HEPC3, HEPC74, and AR3C. The crystal structures were superimposed on their V_H and V_L domains. Protein backbones are shown as ribbons and CDR loops are blue (CDRH1), orange (CDRH2), and red (CDRH3). Disulfide bonds are shown as yellow sticks. See also Figure S2, Table S1.

Figure 2.. HEPC3 and HEPC74 recognize E2ecto similarly.

The crystal structure of HEPC3E2core_1a53-HEPC46 (A), HEPC3-E2ecto_1a53-HEPC46 (B), HEPC3-E2ecto_1b09 (D), and HEPC74E2ecto_1b09 (E) complexes are displayed as cartoon representations. E2 – grey, HEPC3-HC – purple, HEPC3-LC – pink, HEPC74-HC – dark green, HEPC74-LC – light green, HEPC46-HC – dark blue, HEPC46-LC – light blue. Disulfide bonds are shown as yellow sticks and E2 glycans are shown as sticks with light blue, red, and dark blue colors for carbon, oxygen, and nitrogen atoms, respectively. Dashed lines indicated disordered regions. (C) E2 regions included in E2core and E2ecto structures. E2 regions are colored by structural components: HVR1 (orange), AS412 (dark green), front layer (yellow), VR2 (red), β-sandwich (violet), CD81bl (blue), VR3 (light grey), post-VR3 (dark grey), back layer (green), and stem (black). Dashed line indicates the G-S-S-G linker, deleted glycosylation sites in E2 core construct are highlighted in red. See also Figure S1, S5, Table S2–S5.

Figure 3.. HEPC3 and HEPC74 recognize HCV E2 using a disulfide-linked CDRH3 motif.

(A) Surface representations of the HEPC3-E2core_1a53-HEPC46, HEPC74-E2ecto_1b09, and AR3CE2core_1a154 (PDB 4MWF) (Kong et al., 2013) structures. The HEPC46 Fab is not shown for clarity. E2 glycans are shown as sticks with light blue, red, blue colors for carbon, oxygen, and nitrogen atoms, respectively. E2cores – grey, HEPC3-HC – purple, HEPC3-LC – pink, HEPC74HC – dark green, HEPC74-LC – light green, AR3C-HC – cyan, AR3C-LC – light cyan. (B) Comparison of the heavy chain CDR (CDRH) loop positions in E2 core complex structures with HEPC3, HEPC74, and AR3C. E2core structures are shown as cartoon representation (top) and as surface representation (bottom). CDRH loops are blue (CDRH1), orange (CDRH2), and red (CDRH3) tubes. Disulfide bonds and the α1-helix are shown as yellow sticks and dashed ovals, respectively, in the top images. E2 surfaces in the bottom images are colored by structural components: AS412 (dark green), front layer (yellow), VR2 (red), β-sandwich (violet), CD81bl (blue), VR3 (light grey), post-VR3 (dark grey), back layer (light green). See also Figure S3.

Figure 4.. The intra-CDRH3 disulfide bond in E2 front layer-specific bNAbs is required for maximal binding and neutralization.

(A) Binding to E2ecto_1a53 (top) and neutralizing activity against 1a53 strain HCVpp (bottom) of wild-type bNAbs (black) and corresponding CDRH3 2Cys→2Ala (green) or CDRH3 2Cys→2Ser (orange) double mutants. Means ± s.d. of duplicates are shown. One experiment representative of two independent experiments is shown. (B) Binding to E2ecto_1a53 (top) and neutralizing activity against 1a53 strain HCVpp (bottom) of inferred germline precursors HEPC3rua, HEPC74rua, and AR3Crua (black) and corresponding CDRH3 2Cys→2Ala (green) and CDRH3 2Cys→2Ser (orange) double mutants. One experiment representative of two independent experiments is shown. Means ± s.d. of duplicates are shown. (C) Superposition of HEPC3-E2ecto_1b09 (light green), HEPC3-E2core_1a53 (pink), and AR3CE2core_1a154 (light cyan) structures. Structures were superimposed on the E2cores and are shown as cartoon representations with CDRH3 loops in green (HEPC74), purple (HEPC3), and cyan (AR3C). Disulfide bonds are shown as yellow sticks.

Figure 5.. Soluble E2 ectodomain structures reveal conformations of disordered or truncated regions of E2core.

(A) Structures of HEPC3-E2core_1a53-HEPC46_, HEPC3-E2ecto_1b09, and HEPC3-E2core_1a53-HEPC46 complexes displayed as cartoon representations (HVR1 (orange), AS412 (dark green), front layer (yellow), VR2 (red), ß-sandwich (violet), CD81bl (blue), VR3 (light grey), post-VR3 (dark grey), and back layer (green)) with regions of E2 that were truncated or disordered in E2core structures indicated. Disulfide bonds are yellow sticks; E2core glycans are sticks with light blue, red, blue colors for carbon, oxygen, and nitrogen atoms, respectively. Dashed lines indicate disordered regions. (B) Observed disulfide bonds in E2core and E2ecto structures. Disulfide bonds visible only in E2ecto structures are highlighted in bold. (C) Ordered glycans in E2core and E2ecto structures. Glycans that contact HEPC3 are highlighted in bold. (D) Close proximity of Cys581-Cys585 and Cys569-Cys597 disulfide bonds (E2ecto_1b09 structure). In the E2core_1a154 structure (Kong et al., 2013), the disulfide pattern is: Cys561–Cys581 and Cys585–Cys597. (E) Structure of the VR2 loop (E2ecto_1b09 structure) (red). The Cys459-Cys486 disulfide bond that is disordered in the E2 core structures is shown. In the E2core_2a structure (Khan et al., 2014), Cys486 disulfide bonds with Cys620, likely due to deletion of the portion of E2 that includes Cys452, the Cys620 partner. (F) CDRH1 interactions with Asn448 glycan with interacting residues shown as sticks. Potential H-bonds are shown as black dashed lines, and residues at the interface are indicated. (G) HEPC3 light chain interactions with Asn430 glycan with interacting residues shown as sticks. Potential H-bonds are shown as black dashed lines, and residues at the interface are indicated. See also Figure S4.

Figure 6.. The AS412 region can adopt multiple conformations.

(A) Structures of the N-terminal portion of E2ecto including the AS412 region for the HEPC3-E2ecto1b09 and HEPC3E2ecto1a53-HEPC46 complexes. (B) Structures of AS412 peptides in complex with the indicated bNAbs. Only backbone atoms are shown. N- and C- termini are labeled. (C) Superposition of AS412 structures from (A) and (B). Structures were aligned on the C-termini.

Figure 7.. Germline precursors of E2 front layer-specific bNAbs neutralize diverse HCV strains.

(A) Sequence alignment of antibody heavy chain variable domains of the inferred germline precursors HEPC3rua, HEPC74rua, and AR3Crua. CDRs were defined based on IMGT nomenclature and colored blue (CDRH1), orange (CDRH2), and red (CDRH3). Dots indicate identical amino acids and dashes indicate gaps. (B) Heat map showing the binding of HEPC3, HEPC74, AR3C and their germline precursors to a panel of HCV genotype 1 E2ecto proteins. The EC₅₀ value for each E2ecto-mAb combination is shown, with dark red, orange, yellow, or white shading indicating high, intermediate, low, or no detectable binding, respectively. EC₅₀ values greater than 10,000 ng/mL are indicated by the > symbol. One experiment representative of two independent experiments is shown. (C) Heat map showing the neutralization capacity of HEPC3, HEPC74, AR3C and their germline precursors measured using a panel of genotype 1 HCVpp. IC₅₀ values for each virus-mAb combination are shown, with IC₅₀ values greater than 10,000 ng/mL indicated by the > symbol. See also Figure S6, S7.