Structure-guided design of a prefusion GPC trimer induces neutralizing responses against LASV

Abstract

Lassa virus (LASV) belongs to the Arenaviridae family and causes severe hemorrhagic fever in humans. Although many vaccine candidates for Lassa fever exist, no vaccines have been approved for clinical use currently. The precursor glycoprotein complex (GPC), which is expressed as a trimer on the viral surface, is the main target for vaccine development. However, it has been a significant challenge to elicit effective neutralizing antibodies against LASV. In this study, we designed and produced a prefusion GPC trimer antigen of LASV, named GPCv2. Based on the structural information of GPC, we made modifications by replacing the amino acid at position 328 with proline and appending the trimerization domain. This resulted in a highly expressed prefusion trimeric form of GPCv2 that retained important conformational epitopes and stimulated higher levels of neutralizing antibodies. Moreover, vaccination with GPCv2 protected mice from LASV pseudovirus challenge. Additionally, immune repertoire sequencing showed that the induced immune clones in the trimeric group were more convergent and has its own unique V-J pairing bias compared with monomeric group. These findings demonstrate the potential of GPCv2 as a promising candidate antigen for an effective vaccine against LASV.

Fig. 1. Construction and expression of the GPCv1 and GPCv2.

a Linear schematic of the GPCv1 and GPCv2 constructs. A G4S linker was inserted between GP1 and GP2; T4fib was fused to the C terminal. b A molecular representation of GPCv2. c A representative elution chromatograph of PBS (gray dotted line) and GPCv1 (blue line) as well as GPCv2 (red line). d SDS-PAGE analysis of GPC v1 and GPCv2 (left: −DTT; right: +DTT). e A comparison of the N-glycosylation profile of GPCv1 and GPCv2. The percentage of each group is shown for each glycosite in the pie chart, and the proportion of each glycoform is shown in the bar chart. f Characteristics of N-glycosylation sites. The diagram illustrates the structural schematic of the modified GPCv2. The orange parts represent the glycosylation sites on the 7UOT structure, the yellow ones show the sites simulated from the 5vk2 structure, and the green ones indicate the glycosylation sites on GPCv2.

Fig. 2. Antigenicity of GPCv1 and GPCv2.

a Confirmation of the antigenicity of GPCv1 and GPCv2 using ELISA. b, c Binding profiles of purified GPCv1 and GPCv2 for different mAbs from four groups of human Lassa-neutralizing antibodies: GP1-A, GPC-A, GPC-B, and GPC-C. d Summary of binding affinity. e Location of putative epitopes on a structural model of LASV GPC (surface model). Epitope color code: GPC-B, blue; GPC-C, purple; and GP1-A, green.

Fig. 3. Evaluation of immunization performed using different GPC variants.

a A schematic overview of the mouse immunization study. Mice (n = 6 per group) were either mock-immunized with PBS (gray column) or vaccinated with GPCv1 (blue column) or GPCv2 (red column) intramuscularly. The time points for vaccination and bleeding are indicated with appropriate symbols. b–d The antibody response was analyzed with ELISA using the indicated antigen. e Neutralizing antibodies were analyzed in an HIV-based pseudovirus assay. Lines represent the average titers of all animals in each vaccine group. f Neutralizing antibodies were analyzed in LASV minigenome. g Inhibition of binding of the immune sera by the indicated mAbs (37.7H/37.2D/12.1 F/25.1 C/36.1 F). h Splenocytes were extracted at day 35 post-immunization and stimulated with the GL-9 peptides from LASV GPC for IFN-γ enzyme-linked immunospot (ELISpot) assays. i Percentage of IFN-γ/TNF-α and CD107a-producing CD4 + T cells at day 14 after the third immunization. *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001.

Fig. 4. Protective effects of LASV vaccines.

a Workflow of GPCv1 and GPCv2 protective vaccination experiments. BALB/c mice (n = 4/group) were intramuscularly inoculated with vaccines at 0, 2, and 4 weeks, followed by challenge with luciferase-expressing pseudoviruses via the intraperitoneal route at 5 weeks, and they were then imaged at 1 dpi. b Protection against pseudovirus challenge in mice vaccinated with GPCv1 and GPCv2. The first line shows the pseudovirus infection of 10-week-old BALB/c mice as a control. The bioluminescence in the GPCv1 and GPCv2 vaccination groups was imaged at 1 dpi, 1 day apart. c Total flux events in each group were compared. p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 5. Characterization of BCR repertoire diversity in immunized mice.

a Representative cell clustering for GPCv1 and GPCv2. b For BCR clonotype analyses, the top 10 contrasts and their proportion between immunized groups and PBS group are shown. c A Venn diagram showing the variety of clones. Comparing the number of shared and unique elements among three samples: GPCv1, GPCv2, and PBS, with the corresponding numbers labeled in different regions of the diagram. d Frequencies of IGHV-J gene pair usage.