A Dengue Virus Serotype 1 mRNA-LNP Vaccine Elicits Protective Immune Responses

Abstract

Dengue virus (DENV) is the most common vector-borne viral disease, with nearly 400 million worldwide infections each year concentrated in the tropical and subtropical regions of the world. Severe dengue complications are often associated with a secondary heterotypic infection of one of the four circulating serotypes. In this scenario, humoral immune responses targeting cross-reactive, poorly neutralizing epitopes can lead to increased infectivity of susceptible cells via antibody-dependent enhancement (ADE). In this way, antibodies produced in response to infection or vaccination are capable of contributing to enhanced disease in subsequent infections. Currently, there are no available therapeutics to combat DENV disease, and there is an urgent need for a safe and efficacious vaccine. Here, we developed a nucleotide-modified mRNA vaccine encoding the membrane and envelope structural proteins from DENV serotype 1 encapsulated in lipid nanoparticles (prM/E mRNA-LNP). Vaccination of mice elicited robust antiviral immune responses comparable to viral infection, with high levels of neutralizing antibody titers and antiviral CD4⁺ and CD8⁺ T cells. Immunocompromised AG129 mice vaccinated with the prM/E mRNA-LNP vaccine were protected from a lethal DENV challenge. Vaccination with either a wild-type vaccine or a vaccine with mutations in the immunodominant fusion loop epitope elicited equivalent humoral and cell-mediated immune responses. Neutralizing antibodies elicited by the vaccine were sufficient to protect against a lethal challenge. Both vaccine constructs demonstrated serotype-specific immunity with minimal serum cross-reactivity and reduced ADE in comparison to a live DENV1 viral infection.IMPORTANCE With 400 million worldwide infections each year, dengue is the most common vector-borne viral disease. Forty percent of the world's population is at risk, with dengue experiencing consistent geographic spread over the years. With no therapeutics available and vaccines performing suboptimally, the need for an effective dengue vaccine is urgent. Here, we develop and characterize a novel mRNA vaccine encoding the dengue serotype 1 envelope and premembrane structural proteins that is delivered via a lipid nanoparticle. Our DENV1 prM/E mRNA-LNP vaccine induces neutralizing antibody and cellular immune responses in immunocompetent mice and protects an immunocompromised mouse from a lethal DENV challenge. Existing antibodies against dengue can enhance subsequent infections via antibody-dependent enhancement (ADE). Importantly our vaccine induced only serotype-specific immune responses and did not induce ADE.

Keywords: dengue fever; mRNA vaccine; vaccines.

FIG 1

DENV prM/E vaccine design and viral protein expression. (A) Schematic of the DENV genome and engineered mRNA construct. An mRNA encoding the prM and ENV viral proteins was engineered with N-terminal signal peptide sequence, 5′ and 3′ untranslated regions (UTR) flanking the coding sequence, a 3′ poly(A) tail, and a 5′ cap-1 structure. In vitro-synthesized mRNA is encapsulated in a lipid nanoparticle for use in in vitro and in vivo experiments. (B) 293T cells were transfected with the in vitro-transcribed mRNA encoding the wild-type sequence (WT) or a mutant version with amino acid substitutions in the fusion loop epitope (ΔFL). Lysate was analyzed by Western blotting with the domain III-specific 1A1D-2 monoclonal antibody and the fusion loop-specific 4G2 monoclonal antibody. (C) Supernatant from transfected cells was purified and concentrated through ultracentrifugation and analyzed for VLPs by Western blotting with the 1A1D-2 monoclonal antibody or anti-GAPDH. Unpurified cell lysate from WT mRNA-transfected cells is included as a control. Shown are representative blots. (D) Electron microscopy image of VLPs from purified supernatant of transfected 293T cells showing homogenous shape and size of approximately 30 nm.

FIG 2

Optimization of signal peptide and LNP delivery. (A) Constructs were engineered with alternative signal peptides, and in vitro-transcribed mRNA was transfected into differentiated murine muscle myoblast C2C12 cells. Cell lysate was analyzed by Western blotting with 1A1D-2 monoclonal antibody or anti-GAPDH antibodies. (B) Supernatant of transfected C2C12 cells was analyzed by dot blotting with 1A1D-2. (C) In vitro-synthesized WT or ΔFL mRNA was encapsulated in a lipid nanoparticle and administered to C2C12 cells. Lysate was analyzed by Western blotting with 1A1D-2 antibody. Shown are representative blots.

FIG 3

DENV1 prM/E mRNA vaccines induce neutralizing antibody responses. DENV1 prM/E mRNA-LNP vaccines were administered to 10-week-old C57BL/6 mice. (A) Mice were administered 10 μg of mRNA vaccine in a three-shot schedule, and serum was collected at the indicated time points. (B) Serial dilutions of serum from vaccinated mice were analyzed for neutralization activity by an FRNT against DENV1 strain 16007. Neutralization curves at each time point are shown for WT vaccine recipients (left) and ΔFL vaccine recipients (right). The average values ± standard errors of the mean (SEM) of results for five vaccinated mice are shown. (C) Mice were administered a high (10 μg) or low (3 μg) dose of the mRNA vaccines or vaccine encoding GFP. A separate group of mice were infected with wild-type DENV1 by following the same schedule. (D) Antiviral IgG titers were determined by ELISAs, and the endpoint dilution titer was calculated. (E) Serum was analyzed by FRNTs, and the normalized percentage of infection of each group is plotted as the mean ± SEM for each serum dilution. n = 5 mice per group of mice infected with virus or receiving 3-μg vaccine doses. n = 10 mice per group in mice receiving 10-μg doses of the ΔFL and GFP vaccines. n = 15 for mice receiving 10-μg doses of the WT vaccine. (F) EC₅₀ values of the neutralization curves for individual mice are shown. The statistical significance of results for each group in comparison to that for the GFP control was determined via unpaired t test. **, P < 0.01; ***, P < 0.001. Statistical comparisons with P values of >0.05 are not shown in this figure.

FIG 4

DENV1 prM/E mRNA vaccines induce antiviral CD8⁺ and CD4⁺ T cells. DENV1 prM/E mRNA-LNP vaccines were administered to 10-week-old C57BL/6 mice in a three-shot vaccination schedule. Spleens were harvested after the final vaccination dose (day 56) and stimulated with an overlapping peptide array of DENV1 E protein, DENV2 E protein, or DENV1 NS1 protein. Stimulated cells were stained for the intracellular cytokine IFN-γ and analyzed by flow cytometry. Plotted are the IFN-γ⁺ T cells as a percentage of total CD8⁺ T cells (A) or CD4⁺ T cells (B). n = 5 mice in the WT and ΔFL construct-vaccinated groups. (C) Representative flow cytometry plots and gates from a single mouse.

FIG 5

DENV1 prM/E mRNA vaccines protect against a lethal challenge. Ten micrograms of DENV1 prM/E or GFP mRNA-LNP vaccines was administered to AG129 mice in a prime-boost schedule 4 weeks apart. n = 5 mice per group. (A) Serum from vaccinated mice was isolated 2 weeks after the boost and analyzed for neutralization by FRNT of serially diluted serum samples. Plotted are the means ± SEM of results from five mice for each dilution. (B) EC₅₀ values for each mouse are plotted. The vaccinated mice were then challenged with a lethal dose of DENV1 strain Western Pacific. Mice were monitored for weight (C) and survival (D) postchallenge. **, P < 0.01; ***, P < 0.00.

FIG 6

Passive transfer of immune sera protects against a lethal challenge. Serum from naive or WT prM/E mRNA-vaccinated mice was passively transferred into AG129 mice. One day after transfer, mice were challenged with a lethal dose of DENV1 strain Western Pacific. Mice were monitored for weight (A) and survival (B) postchallenge. Survival curves comparing vaccinated and naive serum recipients were analyzed by log rank test. **, P < 0.01.

FIG 7

DENV1 prM/E mRNA vaccination results in reduced ADE levels. Serum from naive mice, WT prM/E mRNA-vaccinated mice, ΔFL prM/E mRNA-vaccinated mice, or mice infected with DENV1 2 weeks after boost were analyzed for enhancement of DENV2 infection. Serial dilutions of serum were incubated with DENV2 and added to Fcγ receptor-positive K562 cells. Fifteen hours later, infected cells were stained for intracellular ENV and quantified by flow cytometry. The percentage of infected cells was normalized to that of infection in the absence of serum. (A) The fold change in percentage of cells infected is shown compared to that of infections in the absence of serum. The average fold enhancement ± SEM for five mice per group is graphed. (B) A representative flow cytometry histogram of the ENV signal for each different treatment at a 1/100 serum dilution is shown. (C) A 1/100 dilution of serum was incubated with DENV2 for 1 h and added to K562 cells at an MOI of 1. At 48 h later, viral titers in the supernatant were quantified via a focus-forming assay. Viral titers were normalized to virus-infected serum within each independent experiment, and the results of four independent experiments are shown. *, P < 0.05; **, P < 0.01. (D) Serial dilutions of serum from vaccinated mice were analyzed for neutralization activity against DENV2 (strain New Guinea C) with an FRNT. The average values ± SEM of results for five vaccinated mice are shown.