Timing of CD8+ T cell responses in relation to commencement of capillary leakage in children with dengue

Abstract

Immune activation is a feature of dengue hemorrhagic fever (DHF) and CD8+ T cell responses in particular have been suggested as having a role in the vasculopathy that characterizes this disease. By phenotyping CD8+ T cells (CD38+/HLA-DR+, CD38+/Ki-67+, or HLA-DR+/Ki-67+) in serial blood samples from children with dengue, we found no evidence of increased CD8+ T cell activation prior to the commencement of resolution of viremia or hemoconcentration. Investigations with MHC class I tetramers to detect NS3(133-142)-specific CD8+ T cells in two independent cohorts of children suggested the commencement of hemoconcentration and thrombocytopenia in DHF patients generally begins before the appearance of measurable frequencies of NS3(133-142)-specific CD8+ T cells. The temporal mismatch between the appearance of measurable surface activated or NS3(133-142)-specific CD8+ T cells suggests that these cells are sequestered at sites of infection, have phenotypes not detected by our approach, or that other mechanisms independent of CD8+ T cells are responsible for early triggering of capillary leakage in children with DHF.

FIGURE 1

Relationship between kinetics of viremia and the appearance of surface activated CD8⁺ T cells in peripheral blood. Shown in each panel are the mean (±95% CI) of DENV-1 (n = 49; A, C, E) or DENV-3 (n = 30; B, D, F) viremia levels in serial plasma samples from children by FD. The day of defervescence was defined as FD 0. In each panel, the box and whisker plots represent the percentage of CD8⁺ T cells that were double positive for CD38/HLA-DR (A, B), CD38/Ki-67 (C, D), or HLA-DR/Ki-67 (E, F). The number of patients that were evaluated on each day is shown below E and F.

FIGURE 2

Relationship between hemoconcentration and the appearance of surface activated CD3⁺CD8⁺ T cells in peripheral blood. Shown in each panel are the mean (±95% CI) percentage levels of hemoconcentration in serial plasma samples from children with DF (n = 86) or DHF (n = 17) by FD. Hemoconcentration was measured against autologous convalescent samples for 82% of patients. In each panel, the bar graph represents the median (interquartile range) percentage of CD8⁺ T cells that were double positive for HLA-DR/Ki-67 (A), CD38/Ki-67 (B), or CD38/HLA-DR (C), The number of patients that were evaluated on each day for either T cell phenotype or level of hemoconcentration is shown below C.

FIGURE 3

Frequencies of NS3_133–142-specific CD8⁺ T cells during the febrile and afebrile phase. Serial whole blood samples from 103 children with dengue were stained with a pool of tetramers (DENV-1, DENV-2, and DENV-3/4) specific for NS3_133–142-specific CD8⁺ T cells. The scatterplot shows the frequency of tetramer staining NS3_133–142-specific CD8⁺ T cells in individual patient samples according to whether they were detected during the febrile phase (fever day, −6 to −1), afebrile phase (fever day 0–3), or convalescence (fever day 10–15). The frequency of circulating DENV-specific CD8⁺ T cells at early convalescence phase were significantly higher than at febrile phase (p = 0.037, by Mann-Whitney U test). The lower limit of detection (solid horizontal line) was defined as the mean plus 3 SD of the frequency of tetramer-positive staining events in blood samples from patients with no evidence of dengue (62 measurements in 23 nondengue patients).

FIGURE 4

Timing of hemoconcentration in children with DHF. The box and whisker plots represent percentage hemoconcentration (A) and platelet counts (B) in children with DHF (n = 41) by fever day Levels of hemoconcentration were determined by comparison of daily hematocrit values in patients against healthy population age and sex stratified mean hematocrit values (see Table I). Hemoconcentration of >20% was first detected a median of 1 d before defervesence in these children. The number of children with hematocrit or platelet measurements on each fever day is shown beneath the x-axis.

FIGURE 5

Temporal relationship between detection and frequency of HLA-A*1101–restricted NS3_133–142-specific CD8⁺ T cell responses and hemoconcentration, thrombocytopenia and defervescence in children with dengue. Shown in each scatterplot is the percentage of NS3_133–142-specific CD8⁺ T cell detected by tetramer staining in individual patient blood samples against a reference timepoint of day of defervescence (day 0 in A), day when platelet count of <100,000 cells/ml was first detected in each patient (day 0 in B), and day when hemoconcentration of >20% was first detected in each patient (day 0 in C). The values below the x-axis are the number of patients evaluated on each day.