A nonstructural protein of Dengue virus as a means for safe vaccination

Science Spotlight

A nonstructural protein of Dengue virus as a means for safe vaccination

Tomer Hertz, Vaccine and Infectious Disease Division

May 14, 2017

Dengue is a global health problem, the viral infection affects over 300 million people annually.  Dengue virus (DENV) is transmitted by mosquitoes. The virus has four distinct serotypes (DENV1-4) and infection with any one of them can cause a wide range of effects, from being asymptomatic to hemorrhagic fever. Paradoxically, while infection with one serotype results in homologous protection, infection with a second serotype is a major risk for severe disease. This is thought to be caused by antibody enhancement, when antibodies against structural proteins facilitate viral entry into cells with FC receptors (recognizes a protion of the antibody) leading to increased viral load, immune activation, and severe disease. Besides the structural genes, DENV is made up of seven nonstructural genes (NS), one of which, NS1 is 79% conserved and secreted, thus making it a possible anti-viral antibody candidate. In a paper published in the Journal of Immunology, Dr. Tomer Hertz (Fred Hutch, Vaccine and Infectious Disease Division) and colleagues looked at antibodies raised against NS1 in vaccinated mice and infected humans looking for common epitopes that thus could be targeted for protection. Since NS1 is secreted and not present on the viral surface, antibodies against it are thought to not enhance infection by aiding viral entry.

DENV NS1 antibody profiles of vaccinated mice against NS1 peptides from DENV1 (A), DENV2 (B) and DENV3 (C). boxed represent epitope hotspots and the 5 regions of antibody reactivity.

To study NS1 as a possible vaccine candidate, groups of mice were vaccinated with NS1 and one of three adjuvants: MA, SCpG, and alum, or were given a non-lethal DENV2 dose. Next, all animals were challenged with a lethal dose of DENV2. Both MA and SCpG protected animals produced high antibody titers. Alum did not protect and produced low antibody levels, while the low dose DENV2 mice were protected. Sub-lethal DENV2 immunization protected because both strains were of the DENV2 serotype and the first dose conferred homologous protection while producing intermediate levels of NS1 antibodies. In order to further understand the antibodies produced, sera were tested against a peptide library consisting of a microarray of overlapping peptides that covered the entire NS1 from conserved sequences of serotypes 1-3. Using a Fisher exact test with a Bonferroni correction, the researchers compared the NS1/MA response to the NS1/alum group looking for regions of NS1 that produced a robust response. By doing this, five regions on the NS1 were found to produce a robust antibody response over NS1/alum samples (figure). After identifying DENV2 (the vaccine strain) interactions, DENV1-3 NS1 peptides were examined for cross-reactivity. Region 2 seemed to be the most cross reactive, picking up NS1 peptides from all three serotypes tested. Region 1 cross reacted to 1 and 3 while regions 3 and 5 reacted to 2 and region 4 to 3 only. After identification of the 5 regions specific for most antibodies, the authors looked at conservation of the NS1s peptides across serotypes and found the most cross reactive to match the most conserved and that all regions were surface exposed on the full NS1 protein.

To further study NS1 antibodies, samples were taken from DENV2/3 infected children looking at response and epitope binding. As expected antibody levels were higher in convalescent sera than acute infection time points and responses waned at the 12-month mark. By screening the sera against the peptide library results showed that region 2 almost exclusively responded to in the convalescent sera but waned at the 12-month time point. Region 5 was also found in the screen but weaker than region 2 and also waned by 12 months. Overall this study found that region 2 (aa 101-135) was immune-dominant both in vaccinated mice and infected humans and cross reacted between DENV1-3. This region mapped to the wing domain which is thought to be exposed on the dimeric and hexameric forms of NS1. These similarities between the mouse and human sera and it peptide recognition support the idea of eliciting protective epitope from vaccination that would be similar to infection. By mapping regions by the peptide array the group also mapped immunodominant hotspot on NS1, information that could be used for further vaccine design.

Funding for this study was provided by the National Institute of Allergy and Infectious Diseases, NIH.


Hertz T, Beatty PR, MacMillen Z, Killingbeck SS, Wang C, Harris E. 2017. “Antibody Epitopes Identified in Critical Regions of Dengue Virus Nonstructural 1 Protein in Mouse Vaccination and Natural Human Infections.” J Immunol. 198(10):4025-4035. doi: 10.4049/jimmunol.1700029.