Cytokine Signature Associated with Disease Severity in Dengue

Abstract

Dengue is the most rapidly spreading viral disease transmitted by the bite of infected Aedes mosquitos. The pathogenesis of dengue is still unclear; although host immune responses and virus serotypes have been proposed to contribute to disease severity. In this study, we examined the circulating dengue virus (DENV) and measured plasma levels of inflammatory mediators. Ninety-eight patients during a dengue outbreak in eastern India in 2016 were included in the study. The presence of DENV was demonstrated by detecting NS1 antigen; IgM capture ELISA and serotypes were discriminated by type-specific RT-PCR and/or sequencing. Plasma samples were assayed for 41-plex cytokine/chemokines using multiplex Luminex assay. Eighty-five (87%) samples were positive by NS1/IgM capture ELISA/RT-PCR. All four serotypes of DENV were detected in this outbreak, with DENV-2 as the predominant type, seen in 55% of cases. Mixed infections were seen in 39% of subjects. Among the host inflammatory biomarkers, GM-CSF, IFN-γ, IL-10, IL-15, IL-8, MCP-1, IL-6, MIP-1β, and TNF-α levels were significantly increased in dengue with and without warning signs, in severe dengue patients in comparison to healthy controls. Four cytokines IFN-γ, GM-CSF, IL-10, and MIP-1β correlated significantly with disease severity and could serve as potential predictor for disease severity. Information on the host biomarkers and the dengue serotype may help guide in optimizing effective intervention strategies.

Keywords: Arbovirus; Cytokine; Dengue; Inflammation; Luminex-bead assay; Severe dengue.

Figure 1

Cytokine levels in healthy controls and dengue positive subjects. Cytokine expression profile among the control vs dengue positive subjects. Of the 41 cytokine examined through Luminex bead array nine cytokines found to be associated with the dengue positivity when compared to the healthy controls. Statistical analyses were performed using Prism^® (version 6, GraphPad Software, CA, USA). Proportions were computed with the corresponding 95% confidence intervals (95% C.I.) according to the Poisson distribution. A ‘P’ value less than 0.05 was considered as statistically significant (* = p < 0.05, ** = p < 0.01 and *** = p < 0.001).

Figure 2

Cytokine levels in control, dengue fever (with and without warning signs) and in severe dengue. Cytokine expression profile among the control, mild dengue (with and without warning signs) and severe dengue subjects. Of the 41 cytokine examined through Luminex bead array, four cytokines were found to be associated with disease severity, when compared to the dengue patients; i.e., dengue fever (with and without warning signs), severe dengue, and the disease free controls. Statistical analyses were performed using Prism^® (version 6, GraphPad Software, CA, USA. A ‘P’ value less than 0.05 was considered as statistically significant (* = p < 0.05, ** = p < 0.01 and *** = p < 0.001). The cytokines IFN-γ, GM-CSF, IL-10 and MIP-1β could be the predictor of disease severity.

Figure 3

sPLS-DA and VIP score of Cytokine expression profile in Dengue fever vs Severe Dengue Patients. The Sparse Partial Least Squares Discriminant Analysis (sPLS-DA) depicting the cluster between the multiplex cytokine expression profile between the dengue fever (with and without warning signs) vs severe dengue patients (Figure 3a). The red circle represents for the multiplex cytokine expression for dengue fever, while the green color depicted for severe dengue patients (3b). The variable importance in the projection (VIP) score for the multiplex cytokine profile summarizes the contributions of the variables depicted in the model (3b). The expression of cytokines IFN-γ, IL-10, G-CSF, GRO, RANTES, and FGF-2 is higher in severe dengue compared to dengue fever patients.

Figure 4

sPLS-DA and VIP score of Cytokine expression profile in DENV-2 vs DENV -1, -3, -4 serotypes in Dengue Patients. The Sparse Partial Least Squares Discriminant Analysis (sPLS-DA) depicting the cluster between the multiplex cytokine expression profile between the DENV-2 vs DENV-1, DENV-3 and DENV-4. The red circle represents for the DENV-1, -3, -4, while the green color depicted for DENV-2 subjects cluster (4a). The variable importance in the projection (VIP) score for the multiplex cytokine profile summarizes the contributions of the variables depicted in the model (4b). The DENV-2 vs DENV-1, -3, -4 group clustered distinctly in terms of the expression profile of the soluble biomarkers.