Potential of ancestral sylvatic dengue-2 viruses to re-emerge

Abstract

Dengue viruses (DENV) are the most important arboviral pathogens in tropical and subtropical regions throughout the world. DENV transmission includes both a sylvatic, enzootic cycle between nonhuman primates and arboreal mosquitoes of the genus Aedes, and an urban, endemic/epidemic cycle between Aedes aegypti, a mosquito with larval development in peridomestic water containers, and human reservoir hosts. All 4 serotypes of endemic DENV evolved independently from ancestral sylvatic viruses and have become both ecologically and evolutionarily distinct; this process may have involved adaptation to (i) peridomestic mosquito vectors and/or (ii) human reservoir hosts. To test the latter hypothesis, we assessed the ability of sylvatic and endemic DENV-2 strains, representing major genotypes from Southeast Asia, West Africa and the Americas, to replicate in two surrogate human model hosts: monocyte-derived, human dendritic cells (moDCs), and mice engrafted with human hepatoma cells. Although the various DENV-2 strains showed significant inter-strain variation in mean replication titers in both models, no overall difference between sylvatic and endemic strains was detected in either model. Our findings suggest that emergence of endemic DENV strains from ancestral sylvatic strains may not have required adaptation to replicate more efficiently in human reservoir hosts, implying that the potential for re-emergence of sylvatic dengue strains into the endemic cycle is high. The shared replication profiles of the American endemic and sylvatic strains suggest that American strains have maintained or regained the ancestral phenotype.

Fig. 1

Phylogenetic analysis of DENV isolates. Phylogenetic tree derived from the envelope protein gene nucleotide sequences of sylvatic and representative endemic DENV-2 strains using Bayesian analysis or maximum likelihood (ML) (PAUP, version 4.10) and drawn using branch lengths obtained using the Rogers–Swofford approximation method. The following ML parameters corresponding to the GTR+G+I model were used: empirical values for nucleotide frequencies (A = 0.32989, C=0.19737, G = 0.25924 and T = 0.21350); and among-site rate variation as: at invariable sites as estimated and γ distribution (discrete approximation) of rates at variable sites. The scale shows a genetic distance of 0.01 or 1% nucleotide sequence divergence. Homologous sequences from dengue sister serotypes 1 and 3 were used as an outgroup to root the DENV-2 tree. Numbers indicate bootstrap values for groups to the right. Asterisks indicate the strains used for this study.

Fig. 2

FACS analysis of blood-derived DCs. (A) Cell surface staining of immature blood-derived moDCs at day 6 post stimulation with IL-4 and GM-CSF cytokine cocktail prior to DENV-1349 infection and infected blood-derived DCs at day 2 post-infection with DENV-1349. Surface expression of CD40, CD80, CD83, CD86 and DC-SIGN was evaluated to determine the maturity of the DCs immediately prior to DENV infection. DC-SIGN is not a marker of moDC phenotype but mediates the infection of human DCs by DENV. Green peaks, isotype control; purple-shaded peaks, non-infected immature DCs; pink peaks, infected mature DCs. Data are representative of three independent experiments. (B) Infected DCs were examined by FACS analysis 48 h post-infection with an α-dengue virus-specific antibody, followed by an FITC-conjugated secondary antibody to determine the percentage of infected cells. Graph represents pooled data generated from 2 individual donors and error bars represent the standard errors of the means.