Higher levels of viremia and circulating nonstructural protein 1 (NS1) have been associated with dengue disease severity. In this study, viremia and circulating NS1 levels were determined in 225 serum samples collected from patients in Mexico infected with dengue virus serotypes 1 and 2 (DENV-1 and DENV-2). Patients with dengue hemorrhagic fever (DHF) who were infected with DENV-1 showed higher levels of circulating NS1 than patients with dengue fever (DF) (P = 0.0175). Moreover, NS1 levels were higher in patients with primary infections with DENV-1 than in patient infected with DENV-2 (P < 0.0001) and in patients with primary infections with DENV-2 than in patients with secondary infections with DENV-2 (P = 0.0051). Unexpectedly, viremia levels were higher in patients with DF than in those with DHF infected with either DENV-1 or DENV-2 (P = 0.0019 and P = 0.001, respectively) and in patients with primary infections than those with secondary DENV-2 infections (P < 0.0001). Results indicate that levels of circulating NS1 vary according to the infecting serotype, immunologic status (primary or secondary infection), and dengue disease severity.

Dengue virus (DENV) is the causative agent of dengue fever and the more severe forms of the infection known as dengue haemorrhagic fever and dengue shock syndrome (DHF/DSS). Secondary infections with a serotype different from the primary infection are considered a risk factor for the development of DHF/DSS. One explanation for the increased risk of DHF/DSS development after heterologous secondary infections is the antibody-dependent enhancement (ADE) hypothesis. This hypothesis postulates that pre-existing non-neutralizing antibodies will form immune complexes with the new serotype-infecting virus that in turn will have enhanced capacity to infect macrophages and other Fcγ receptor (FcγR)-bearing cells. Despite the evidence supporting the ADE hypothesis, the molecular mechanisms of ADE are not fully understood. In this work, we present evidence which indicates that intact lipid rafts are required for the ADE infection of U937 cells with DENV. Flow cytometry analysis to measure the percentage of infected cells showed that treatment of differentiated U937 cells with nystatin (30 μg ml−1), filipin (10 μg ml−1) or β-methyl cyclodextrin (30 mM) significantly reduces (P<0.05) the ADE of DENV-4 infection in vitro without any effect on viability or the number of FcγR-bearing cells. Later cholesterol replenishment by supplementing treated cell cultures with bovine fetal serum for 24 h re-established lipid raft integrity and reversed the alteration of the ADE in vitro (P<0.05). Our results suggest that ADE of U937 infection by DENV requires the presence of cholesterol and cholesterol-rich membrane microdomains.