Production of a Recombinant Dengue Virus 2 NS5 Protein and Potential Use as a Vaccine Antigen

Abstract

Dengue fever is caused by any of the four known dengue virus serotypes (DENV1 to DENV4) that affect millions of people worldwide, causing a significant number of deaths. There are vaccines based on chimeric viruses, but they still are not in clinical use. Anti-DENV vaccine strategies based on nonstructural proteins are promising alternatives to those based on whole virus or structural proteins. The DENV nonstructural protein 5 (NS5) is the main target of anti-DENV T cell-based immune responses in humans. In this study, we purified a soluble recombinant form of DENV2 NS5 expressed in Escherichia coli at large amounts and high purity after optimization of expression conditions and purification steps. The purified DENV2 NS5 was recognized by serum from DENV1-, DENV2-, DENV3-, or DENV4-infected patients in an epitope-conformation-dependent manner. In addition, immunization of BALB/c mice with NS5 induced high levels of NS5-specific antibodies and expansion of gamma interferon- and tumor necrosis factor alpha-producing T cells. Moreover, mice immunized with purified NS5 were partially protected from lethal challenges with the DENV2 NGC strain and with a clinical isolate (JHA1). These results indicate that the recombinant NS5 protein preserves immunological determinants of the native protein and is a promising vaccine antigen capable of inducing protective immune responses.

FIG 1

Expression of the recombinant DENV2 NS5 protein produced by E. coli. Coomassie blue-stained polyacrylamide gels of whole bacterial cell extracts induced after cultivation with IPTG for 16 h at 18°C (a) or 4 h at 37°C (b). Western blot analysis of the whole-cell extracts for the first (c) and second (d) induction conditions probed with a mouse anti-NS5 MAb; the black arrows indicate the full NS5 expressed (107 kDa). Lane 1, whole-cell extract of noninduced BLNS5 culture; lane 2, whole-cell extract of induced BLNS5 culture; lane 3, soluble protein fraction of BLNS5 culture after IPTG induction; lane 4, insoluble protein fraction of BLNS5 culture after IPTG induction; lane M, molecular size marker.

FIG 2

Purification of the recombinant NS5. The soluble NS5 was subjected to two sequential purification steps. (a) Following binding to the nickel column, the protein was eluted with increasing imidazole concentrations: 150 mM, 300 mM, 500 mM, and 1 M. The eluted fractions were pooled and dialyzed with Tris-NaCl buffer (50 mM Tris, 500 mM NaCl, 10% glycerol; pH 8.0). (b) The dialyzed material was subjected to gel filtration, resulting in elution of three peaks at 97, 113, and 143.74 min. (c and d) Comassie blue-stained gel (c) and Western blot (d) of the eluted fractions (1 μg of purified protein).

FIG 3

Reactivity of DENV-infected human sera with the recombinant NS5. Sera collected from subjects infected with different DENV serotypes were reacted with purified NS5 (●) and with NS5 subjected to a heat denaturation step (100°C for 20 min) (■) by ELISA. Serum samples were collected from convalescent patients infected with DENV1 (a), DENV2 (b), DENV3 (c), or DENV4 (d). Serum samples were serially diluted (log2), with an initial dilution of 1/100. Each well was adsorbed with 400 ng of recombinant NS5. Statistical analyses were performed by two-way variance analysis tests followed by Bonferroni's posttest. ***, P < 0.01; **, P < 0.1; *, P < 0.5.

FIG 4

Evaluation of the serum antibody responses generated in mice immunized with NS5. BALB/c mice were s.c. immunized with only 10 μg of NS5 (NS5) or NS5 admixed with alum (NS5 + Alum) in three doses at intervals of 2 weeks. (a) Two weeks after each immunization, sera from mice were subjected to ELISA for detection of anti-NS5 serum IgG titers. Values are expressed as mean ± standard deviation. (b) NS5-specific IgG subclass responses after the third vaccine dose. The value at the top represents the IgG1-to-IgG2a subclass ratio. Background values detected in sham-treated mice (PBS) were deduced from values determined in other immunization groups. Data are representative of at least two independent experiments with similar results. Statistical analyses were performed by two-way variance analysis tests followed by Bonferroni's posttest (a) and Student's t test followed by the Mann-Whitney posttest (b). ***, P < 0.01; **, P < 0.1; *, P < 0.5.

FIG 5

Detection of IFN-γ- and TNF-α-secreting cells in spleens of mice immunized with NS5. Two weeks after the last dose, the numbers of IFN-γ- and TNF-α-secreting cells in spleens of immunized mice were determined in ELISPOT assays using pools of NS5-specific MHC-I-restricted (a) or MHC-II-restricted (b) synthetic peptides. Data are representative of at least two independent experiments with similar results. The t test, followed by the Mann-Whitney test, was used to compare means of cytokine-secreting cells. SFU, spot-forming units. *, P < 0.05.

FIG 6

Protective anti-DENV2 immunity induced in mice immunized with NS5. Two weeks after the last immunization, mice were i.c. challenged with 5 × 10⁵ PFU of the DENV2 NGC strain or 1.5 × 10² PFU of the JHA1 clinical isolate. Survival was monitored for 25 days. Each test survival curve was compared to the respective control using the log-rank test, and the P value is represented.