We analyzed four DNA vaccines based on DENV-2 NS1: pcENS1, encoding the C-terminal from E protein plus the NS1 region; pcENS1ANC, similar to pcENS1 plus the N-terminal sequence from NS2a (ANC); pcTPANS1, coding the t-PA signal sequence fused to NS1; and pcTPANS1ANC, similar to pcTPANS1 plus the ANC sequence. The NS1 was detected in lysates and culture supernatants from pcTPANS1-, pcENS1- and pcENS1ANC-transfected cells and not in cells with pcTPANS1ANC. Only the pcENS1ANC leads the expression of NS1 in plasma membrane, confirming the importance of ANC sequence for targeting NS1 to cell surface. High levels of antibodies recognizing conformational epitopes of NS1 were induced in mice immunized with pcTPANS1 and pcENS1, while only few pcENS1ANC-inoculated animals presented detectable anti-NS1 IgG. Protection against DENV-2 was verified in pcTPANS1- and pcENS1-immunized mice, although the plasmid pcTPANS1 induced slight higher protective immunity. These plasmids seem to activate distinct patterns of the immune system.

Dengue virus (DENV) is spread through most tropical and subtropical areas of the world and represents a serious public health problem. At present, the control of dengue disease is mainly hampered by the absence of antivirals or a vaccine, which results in an estimated half worldwide population at risk of infection. The immune response against DENV is not yet fully understood and a better knowledge of it is now recognized as one of the main challenge for vaccine development. In previous studies, we reported that a DNA vaccine containing the signal peptide sequence from the human tissue plasminogen activator (t-PA) fused to the DENV2 NS1 gene (pcTPANS1) induced protection against dengue in mice. In the present work, we aimed to elucidate the contribution of cellular and humoral responses elicited by this vaccine candidate for protective immunity. We observed that pcTPANS1 exerts a robust protection against dengue, inducing considerable levels of anti-NS1 antibodies and T cell responses. Passive immunization with anti-NS1 antibodies conferred partial protection in mice infected with low virus load (4 LD50), which was abrogated with the increase of viral dose (40 LD50). The pcTPANS1 also induced activation of CD4+ and CD8+ T cells. We detected production of IFN-γ and a cytotoxic activity by CD8+ T lymphocytes induced by this vaccine, although its contribution in the protection was not so evident when compared to CD4+ cells. Depletion of CD4+ cells in immunized mice completely abolished protection. Furthermore, transfer experiments revealed that animals receiving CD4+ T cells combined with anti-NS1 antiserum, both obtained from vaccinated mice, survived virus infection with survival rates not significantly different from pcTPANS1-immunized animals. Taken together, results showed that the protective immune response induced by the expression of NS1 antigen mediated by the pcTPANS1 requires a cooperation between CD4+ T cells and the humoral immunity.

Two DNA vaccines were constructed encoding the ectodomain (domains I, II and III) of the DENV2 envelope protein (pE1D2) or only its domain III (pE2D2), fused to the human tissue plasminogen activator signal peptide (t-PA). The expression and secretion of recombinant proteins was confirmed in vitro in BHK cells transfected with the two plasmids, detected by immunofluorescence or immunoprecipitation of metabolically labeled gene products, using polyclonal and monoclonal antibodies against DENV2. Besides, results reveal that the ectodomain of the E protein can be efficiently expressed in vivo, in a mammalian system, without the prM protein that is hypothesized to act as a chaperonin during dengue infection. Balb/c mice were immunized with the DNA vaccines and challenged with a lethal dose of DENV2. All pE1D2-vaccinated mice survived challenge, while 45% of animals immunized with the pE2D2 died after infection. Furthermore, only 10% of pE1D2-immunized mice presented some clinical signs of infection after challenge, whereas most of animals inoculated with the pE2D2 showed effects of the disease with high morbidity degrees. Levels of neutralizing antibodies were significantly higher in pE1D2-vaccinated mice than in pE2D2-immunized animals, also suggesting that the pE1D2 vaccine was more protective than the pE2D2.

The lack of an immunocompetent animal model for dengue mimicking the disease in humans is a limitation for advances in this field. Inoculation by intracerebral route of neuroadapted dengue strains in mice is normally lethal and provides a straightforward readout parameter for vaccine testing. However, systemic effects of infection and the immune response elicited in this model remain poorly described. In the present work, BALB/c mice infected by the intracerebral route with neuroadapted DENV2 exhibited several evidences of systemic involvement. DENV-inoculated mice presented virus infective particles in the brain followed by viremia, especially in late stages of infection. Infection induced cellular and humoral responses, with presence of activated T cells in spleen and blood, lymphocyte infiltration and tissue damages in brain and liver, and an increase in serum levels of some pro-inflammatory cytokines. Data highlighted an interplay between the central nervous system commitment and peripheral effects under this experimental condition.

In the last decade, Flaviviruses such as yellow fever (YFV) and Zika (ZIKV) have expanded their transmission areas. These viruses originated in Africa, where they exhibit both sylvatic and interhuman transmission cycles. In Brazil, the risk of YFV urbanization has grown, with the sylvatic transmission approaching the most densely populated metropolis, while concern about ZIKV spillback to a sylvatic cycle has risen. To investigate these health threats, we carried out extensive collections and arbovirus screening of 144 free-living, non-human primates (NHPs) and 5219 mosquitoes before, during, and after ZIKV and YFV outbreaks (2015–2018) in southeast Brazil. ZIKV infection was not detected in any NHP collected at any time. In contrast, current and previous YFV infections were detected in NHPs sampled between 2017 and 2018, but not before the onset of the YFV outbreak. Mosquito pools screened by high-throughput PCR were positive for YFV when captured in the wild and during the YFV outbreak, but were negative for 94 other arboviruses, including ZIKV, regardless of the time of collection. In conclusion, there was no evidence of YFV transmission in coastal southeast Brazil before the current outbreak, nor the spread or establishment of an independent sylvatic cycle of ZIKV or urban Aedes aegypti transmission of YFV in the region. In view of the region’s receptivity and vulnerability to arbovirus transmission, surveillance of NHPs and mosquitoes should be strengthened and continuous.