Virus glycoprotein nanodisc platform for vaccine design

Abstract

Transmembrane glycoproteins of enveloped viruses are the targets of neutralizing antibodies and essential vaccine antigens. mRNA-LNP technology allows in situ production of transmembrane glycoproteins upon immunization, but biophysical characterization of transmembrane antigens and in vitro analysis of post-immunization antibody responses typically rely on soluble proteins. Here, we present a methodological platform for assembling transmembrane glycoprotein vaccine candidates into lipid nanodiscs. We demonstrate the utility of the nanodiscs in HIV membrane proximal external region (MPER)-targeting vaccine development by binding assays using surface plasmon resonance (SPR), ex vivo B cell sorting with fluorescence-activated cell sorting (FACS), and by determining the structure of a prototypical HIV MPER-targeting immunogen nanodisc in complex with three broadly neutralizing antibodies (bnAbs), including the MPER bnAb 10E8, to 3.5 Å by cryogenic electron microscopy (cryo-EM), providing a template for structure-based immunogen design for MPER. Overall, the platform offers a tool for accelerating the development of next-generation viral vaccines.

Figure 1.

Nanodisc use in different steps in iterative rational vaccine design. Analytical methods that can directly use nanodiscs as sample material are highlighted.

Figure 2.

HIV Env constructs used for method development and general overview of the nanodisc assembly workflow. a) Naming and gene organization of the glycoprotein constructs used for nanodisc assembly. Introduced mutations and intracellular elements are indicated. ND refers to constructs made for nanodisc platform. b) Schematic overview of HIV Env constructs and MPER targeting antibody 10E8 binding. c) Key steps of the 5-day workflow for glycoprotein nanodisc assembly. Glycoprotein was extracted with detergent from cell surface and bound to affinity matrix. Disc assembly was performed “in-batch” while bound to affinity purification matrix. Assembled discs were then eluted for final SEC polishing purification, followed by quality control steps before subjecting to final analytical methods (SPR, FACS, cryo-EM).

Figure 3.

Different SPR modalities and kinetic analyses of GP nanodiscs. a) Schematic illustrations of three SPR modalities used throughout the study and their strengths and weaknesses. b) Affinity of MPER targeting 10E8 antibody to Env gp151 ND as measured with modalities A and B. Affinities with modality A were measured after storing the sample either 1 week (white), 1 month (grey) or 3 months (black) at 4°C. c) Affinity of 10E8 against Env constructs using modality A. d) SPR data presented in (c), showing increase in affinity originating from reduced off-rate in Env gp151 MPER ND. e) Modality C was used to scout Ebola specific EBOV-296 and EBOV-13C6 IgG epitope accessibility in EBOV GP nanodiscs. Data shows binding of two anti -Ebola antibodies targeting the glycan cap (EBOV-296 and 13C6). Env gp151 ND and anti-HIV MPER antibody PGZL1 were used as a negative controls.

Figure 4.

HIV Env nanodisc FACS probe validation and pilot use in pre-clinical mouse model. a) Nanodiscs were tested for binding to HIV MPER bnAbs 10E8 and DH511 and gp120-specific bnAb BG18 coupled to FACS compensation beads, B6 mouse splenocytes were used as negative control cells, and VRC01 expressing Ramos cells as positive control cells. Empty nanodiscs were assembled with either DOPC lipids (blue) or a mixture of neutral and charged lipids (green). b) Memory (CD19⁺IgM⁻IgD⁻) and naïve (CD19⁺IgM⁺IgD⁺) B cells from mice 6 weeks after immunization with mRNA-LNPs encoding Env gp151 MPER ND nanodisc were analyzed by flow cytometry. Antigen-specific (Env gp151 MPER ND⁺⁺) and epitope-specific (KO⁻) B cells were detected using immunogen-matched WT and 10E8-epitope KO nanodisc probes and WT nanodisc⁺⁺ cells were sorted for BCR sequencing. The complete gating strategy is shown in supplementary Fig. 3. c) Percent antigen-specific memory B cells in the naïve vs. memory compartments of each immunized mouse (top), and proportion of 10E8 epitope-specific (KO⁻) cells within each antigen-specific population. Paired t-test, *p < 0.05, ns: p> 0.05. d) Sorted cells were sequenced and selected antibodies were purified for affinity measurement by SPR using modality B.

Figure 5.

HIV Env nanodisc probe validation for sorting NHP cells. a) Flow cytometry plots showing CD20⁺IgD⁻ memory B cells stained with Env gp151 ND nanodisc on two different fluorophores. Pre-immunization and 2-weeks post-immunization PBMCs from the same RM are shown and the frequency of Nanodisc⁺⁺ cells of IgD⁻ cells is listed. b) Flow cytometry plots showing CD20⁺IgG⁺ memory B cells stained with either soluble Env gp140 protein or Env gp151 ND nanodisc on two different fluorophores. PBMCs from two different RMs 4-weeks post mRNA Env gp151 immunization were split and stained with Env gp140 protein tetramers (left column) or matching nanodisc Env gp151 ND tetramers (right column). The frequency of tetramer⁺⁺ cells of IgG⁺ cells is listed for each plot.

Figure 6.

Structure of Env gp151 MPER ND in complex with bnAbs BG18, VRC01 and 10E8 resolved by cryo-EM. a) Low pass filtered maps of different Fab occupancy states and the location of the bilayer. BG18 Fab is highlighted in green, VRC01 in pink and 10E8 in blue. b) 90° clockwise rotated side views showing the Env tilt angle in relation to lipid bilayer surface. c) Top view and side views of highest resolution maps of both 10E8 occupancy states. Densities are highlighted as follows: gp120 in blue, gp41 in orange, BG18 in green, VRC01 in pink, HC of 10E8 in white, LC of 10E8 in dark grey and MPER peptide in red. d) Highest resolution reconstruction with single 10E8 was used for model building. Insets show interfaces of the three bound bnAbs and key amino acid and glycan interactions with BG18 HC (dark green) and LC (light green) in insets 1–2, VRC01 HC (magenta) and LC (pink) in insets 3–4, and 10E8 HC (white) and LC (dark grey) in insets 5–7. Glycans are highlighted in purple. Insets 2 and 4 exemplify the electron density used to built interacting glycans (transparent blue. e) Epitope-paratope analysis of 10E8 binding interface shown in d. Contacting residues are separated into three components corresponding insets 5–7. f) Binding pocket of 3BC315 (salmon) CDRH3 with key interactions highlighted. g) gp41 protomer distance analysis in the nanodisc complex structure with 10E8 Fab, and soluble Env structure with 3BC315 showing widened interface between protomers A and B due to antibody binding.