Activation of Peripheral T Follicular Helper Cells During Acute Dengue Virus Infection

We investigated the immune response of peripheral T follicular helper cells during acute dengue virus infection in Thai children. We report significant associations between peripheral T follicular helper cell activation and plasmablast activation, secondary infection, and disease.

Abstract

Background

Follicular helper T cells (T_FH) are specialized CD4 T cells required for B-cell help and antibody production.

Methods

Given the postulated role of immune activation in dengue disease, we measured the expansion and activation of T_FH in the circulation (peripheral T_FH [pT_FH]) collected from Thai children with laboratory-confirmed acute dengue virus (DENV) infection.

Results

We found significant expansion and activation of pT_FH subsets during acute infection with the highest frequencies of activated pT_FH (PD1^hi pT_FH and PD1⁺CD38⁺ pT_FH) detected during the critical phase of illness. Numbers of activated pT_FH were higher in patients with secondary compared with primary infections and in patients with more severe disease. We also found a positive correlation between the frequencies of activated pT_FH and the frequencies of plasmablasts.

Conclusions

To our knowledge, this is the first ex vivo analysis of pT_FH activation during acute DENV infection. Overall, our study supports the model that pT_FH contribute to disease evolution during the critical stage of illness.

A better understanding of dengue pathology is critical to develop therapeutics and improve vaccine efficacy and safety. The hallmark of severe dengue disease, plasma leakage, occurs around the time of defervescence, coincident with clearance of viremia; this period is referred to as the critical phase of illness [1–3]. Epidemiological studies indicate that plasma leakage occurs more frequently in patients undergoing a second dengue virus (DENV) infection with a virus type different from the primary infection [4–6]. A skewed host immune response has been proposed to contribute to the severe forms of the disease. Cytokines secreted by T cells, including interleukin (IL)2 and tumor necrosis factor (TNF), are capable of enhancing inflammation and increasing vascular permeability, but their importance in severe dengue remains uncertain [7–10]. Human leukocyte antigen (HLA) association studies have found significant correlations of specific HLA types with clinical outcome in dengue [11, 12]. Secondary heterotypic DENV infections can reactivate cross-reactive memory T cells that have lower affinity towards the newly infecting DENV type, potentially altering T-cell efficacy [13, 14]. The number of activated CD8 T cells during acute DENV infection were found to be higher in patients with severe disease in some studies [7, 15]. Although previous studies have suggested a role for T cells in dengue pathogenesis, recent studies have pointed to T cells as mediating protection against severe dengue [16, 17].

Previous work by our group and others has largely focused on assessing the phenotype of antigen-specific CD8 T cells during and after acute DENV infection [18–20]. In comparison, the contribution of specific subsets of CD4 T cells in acute DENV infection is less clear. Studies of CD4 T-cell responses to DENV have primarily focused on the characterization of memory DENV-specific CD4 T cells in DENV-immune individuals [21–25]. To our knowledge, ex vivo analyses of CD4 T-cell subsets during acute DENV infection have not been reported.

The discovery of a specialized subset of CD4 T cells, T follicular helper (T_FH) cells, has generated intense interest because they are critical for B-cell development and differentiation and promote the production of pathogen-specific, long-lasting neutralizing antibodies [26–28]. T follicular helper cells are found in secondary lymphoid organs; however, a population of circulating cells, referred to as peripheral T_FH (pT_FH) cells, are more accessible to study and have phenotypic and functional similarities to T_FH cells [29]. A key surface molecule on T_FH is the CXC chemokine type 5 receptor (CXCR5), and CD4⁺CD45RA⁻CXCR5⁺ T cells in the circulation are referred to as pT_FH [30]. Similar to germinal center T_FH, pT_FH cells provide help to B cells in other natural infections [29, 31]. Previous studies have found positive correlations between pT_FH and B cell frequency or plasmablast production [32, 33].

We used flow cytometry to analyze pT_FH at multiple time points during and after acute DENV infection. We found significant activation of pT_FH during the febrile, critical, and early convalescent phases of infection. The pT_FH activation, defined by coexpression of PD-1 and CD38 or high expression of PD-1, was most prominent during the critical phase of illness. We also found a significant positive correlation between the percentage of activated pT_FH and the percentage of plasmablasts. These results highlight the potential importance of pT_FH to dengue illness.

METHODS

Study Subjects and Blood Samples

The clinical study enrolled Thai children 6 months to 14 years of age with acute febrile illnesses, as previously described [1, 3]. Blood samples were collected daily from enrollment until 1 day after defervescence (or a maximum of 5 consecutive days), once in early convalescence (~10 days after study entry), and during late convalescence (6 months–3 years after study entry) [1, 3]. Peripheral blood mononuclear cells (PBMCs) were isolated by centrifugation over Histopaque and cryopreserved. Acute DENV infections were determined by serologic testing of acute and convalescent samples and virus isolation or detection from acute-phase samples; primary and secondary infections were distinguished based on hemagglutination inhibition antibody titer and immunoglobulin (Ig)M/IgG ratio. Time points are reported relative to the day of defervescence (ie, the day at which body temperature dropped and remained below 38°C), which was termed fever day 0. Negative fever days (−1, −2, etc) occurred before defervescence, and positive fever days (+1, +2, etc) occurred after defervescence. Patient diagnosis of dengue fever (DF) or dengue hemorrhagic fever (DHF) was assigned by an expert clinician based on 1997 World Health Organization guidelines [34]. Written informed consent was obtained from each subject and/or his/her parent or guardian. The study protocol was approved by the Institutional Review Boards of the Thai Ministry of Public Health, the Office of the US Army Surgeon General, and the University of Massachusetts Medical School.

Staining and Flow Cytometry

Cryopreserved PBMCs were thawed in Roswell Park Memorial Institute medium/10% fetal bovine serum (FBS) at 37°C. Cells were washed in Hanks’ Balanced Salt Solution/1% FBS and stained with LIVE/DEAD Aqua (Molecular Probes, Invitrogen Corp.) according to the manufacturer’s instructions. Cells were washed and incubated with monoclonal antibodies at 4°C for 30 minutes. The following antibodies were used for T cell staining: CD3 (clone SK7, APC-H7; BD Biosciences), CD4 (clone OKT4, Alexa 700; eBioscience), CD8 (clone RPA-T8, BV711; BD Biosciences), CXCR5 (clone RF8B2, Alexa 488; BD Biosciences), CD45RA (clone HI100, PE, or BV711; BD Biosciences), CD38 (clone HIT2, BV785, or PE; BD Biosciences), PD-1 (clone eBioJ105, APC; eBioscience), CD25 (clone M-A251, BV421; BioLegend), OX40 (Ber-ACT35, PE-Cy7; BioLegend), CD19 (clone HIB19, V500; BD Biosciences), and CD14 (clone MØP9, BV510; BD Biosciences). The following antibodies were used for plasmablast staining: CD19 (clone SJ25C1, PE-Cy7; BD Biosciences), CD38 (clone HIT2, APC; BD Biosciences), CD27 (clone L128, PE; BD Biosciences), CD3 (clone UCHT1, V500; BD Biosciences), and CD14 (clone MΦP9, BV510; BD Biosciences). The following antibodies were used for pT_FH phenotyping: CXCR3 (clone G025H7, BV421; BioLegend), CCR6 (clone 11A9, BV786; BD Biosciences), and inducible T-cell costimulator ([ICOS] clone C398.45A, PE-Cy7; BioLegend). Cells were then washed and fixed with BD Stabilizing Fixative (BD Biosciences). Data were collected on a BD LSR II with BD FACSDiva 8.0.1 software and analyzed using FlowJo 10.2.

Cell Number Calculations

Total lymphocyte cell counts (cells/microliter) were obtained using an automated machine (Sysmex, Kobe, Japan). The number of T cells (cells/microliter) for each subset was calculated using the percentage of cells (based on the lymphocytes/singles subset) determined by flow cytometry.

Statistics

To compare the expression of activation markers on CD4 and CD8 T cell and pT_FH subsets between groups, we used generalized estimating equation (GEE) models. In the models, the autocorrelation between repeated measures for each patient was taken into account by using a first-order autoregressive correlation structure. The GEE models were also used to determine the statistical trends of an increase or decrease in CD4 and CD8 frequencies during acute DENV infection. Last, we determined the correlation between activated pT_FH and plasmablasts taking into account repeated measures of each patient through the rmcorr package [35] in R version 3.3.1 (R Foundation for Statistical Computing, Vienna, Austria; http://www.r-project.org/). With the exception of repeated measures correlation analysis, all statistical analyses were performed using SAS 9.4 software (SAS Institute, Cary, NC). A P < .05 was considered to be statistically significant.

RESULTS

CD4 and CD8 T-Cell Expansion During Acute Dengue Illness

To investigate T-cell activation in vivo, we analyzed 116 PBMC samples obtained from 27 Thai children during and after acute DENV infection using multi-parametric flow cytometry. Nine and 18 patients were diagnosed with primary (1°) and secondary (2°) DENV infections, respectively. Nineteen patients had DF and 8 patients had DHF. A summary of the patient cohort information is found in Supplementary Table S1. The PBMC samples were collected at febrile (fever days −5 to −1), critical (fever days 0 to +1), early convalescence (fever days +3 to +8), and healthy (6 months to 2 years postenrollment) time points. Figure 1A shows our gating strategy to identify CD4 and CD8 T cells. We found an increase in CD8 frequencies coincident with a decrease in CD4 frequencies during acute infection (Figure 1B). Although the frequencies of CD4 T cells decreased, the average numbers of CD4 and CD8 T cells both increased during acute DENV infection (Figure 1C).

Figure 1.

Expansion of CD4 and CD8 T cells during the course of dengue infection. The gating strategy for the flow cytometry analysis of CD4 and CD8 T-cell subsets is shown (A). The frequency of CD8 (filled circles) and CD4 (open circles) T cells is shown during febrile, critical, early convalescence (E. C.) and healthy time points. Generalized estimating equation models were also used to determine the statistical trend for an increase in CD8 and decrease in CD4 frequencies from fever day −5 to E. C. (B). The number of CD8 (filled circles) and CD4 (open circles) T cells is shown during febrile, critical, E. C. (C). Horizontal lines represents the median for all data points, and bars indicate the interquartile range. *, P ≤ .05; ****, P ≤ .0001.

CD4 and CD8 T Cells Are Highly Activated During Acute Dengue Illness

To study the kinetics of T-cell activation, we used antibodies against CD38 and PD-1 because these markers are elevated on CD8 T cells in PBMCs from patients undergoing acute DENV infection [15, 18, 19]. We found significant PD-1 and CD38 coexpression on both CD8 and CD4 T cells during the febrile, critical, and early convalescence phases of infection when compared with samples obtained from the same individuals 6 months to 2 years later (Figure 2). The mean frequencies of activated (PD-1⁺ and CD38⁺) CD8 and CD4 T cells (Figure 2B) were highest during the critical phase of illness (44% and 18%, respectively). We wanted to determine whether there are significant differences in the number of activated CD8 and CD4 T cells in patients with primary versus secondary DENV infections and DF versus DHF. During the critical phase of illness (fever day 0 to +1), the mean (2°:5.44, 1°:4.82, 2°:4.67, and 1°:4.16 log₁₀ cells/mL) and median (2°:5.46, 1°:4.77, 2°:4.80, and 1°:4.21 log₁₀ cells/mL) number of activated (PD-1⁺ and CD38⁺) CD8 and CD4 T cells, respectively, were significantly higher in patients with secondary versus primary DENV infections (Figure 2C). When comparing patients with DF and DHF, we found the mean (DHF:5.53, DF:5.18, DHF:4.73, DF:4.43 log₁₀ cells/mL) and median (DHF:5.56, DF:5.33, DHF:4.89, and DF:4.53 log₁₀ cells/mL) number of activated CD8 T cells to be significantly higher, but this difference did not reach statistical significance in CD4 T cells during the critical phase of illness (Figure 2D).

Figure 2.

Robust activation of CD4 and CD8 T cells during acute dengue infection. Representative flow cytometry plots for CD8 and CD4 T cells showing expression of CD38, PD-1 (A). Percentage of CD8 and CD4 T cells that coexpress PD-1 and CD38 during febrile, critical, early convalescence (E. C.) and healthy time points (B). Number of activated CD8 and CD4 T cells during acute dengue virus infection (C). Donors diagnosed with primary (open triangle) or secondary (filled triangle) infections and dengue fever (DF) (open circle) or dengue hemorrhagic fever (DHF) (filled circle) are shown. Horizontal lines represents the median for all data points, and bars indicate the interquartile range. *, P ≤ .05; **, P ≤ .01; ***, P ≤ .001; ****, P ≤ .0001.

Activation of Peripheral T Follicular Helper Cells During Acute Dengue Illness

We next wanted to determine whether pT_FH were expanded during acute DENV infection. The expression of CXCR5 has been used as a surrogate marker of pT_FH [36]. Shown in Figure 3A are representative flow cytometry plots of CD4⁺CXCR5⁺CD45RA⁻ T cells in a donor with primary DENV infection. The frequency and number of pT_FH (CD4⁺CD45RA⁻CXCR5⁺ T cells) are also shown in Figure 3. The mean frequency of pT_FH was significantly higher during the critical phase of illness when compared with febrile, early convalescent, and healthy time points (Figure 3B). The mean number of pT_FH was also higher during the critical and early convalescent time points when compared with the febrile phase (Figure 3C). Because a number of cell surface molecules including PD-1 have been used recently to characterize activated pT_FH [36], we evaluated the expression of PD-1, CD38, OX40, and CD25 on pT_FH (Figure 4). Based on the intensity of PD-1 expression on pT_FH, we defined 3 subpopulations of PD-1: PD-1^high, PD-1^intermediate, and PD-1^low. Frequencies of PD-1^high pT_FH were significantly elevated at the end of the febrile phase, throughout the critical phase, and at early convalescent time points (Figure 4A, top panel, and B), reaching up to 60% of pT_FH in some donors. We also found elevated frequencies of PD1⁺CD38⁺ pT_FH during acute infection (Figure 4A, middle panel, and B). In contrast, the frequency of OX40⁺CD25⁺ pT_FH cells did not surpass 25% (Figure 4B).

Figure 3.

Expansion of peripheral T follicular helper (pT_FH) cells during acute dengue infection. Representative flow cytometry plots for pT_FH cells (CD4⁺, CD45RA⁻, CXCR5⁺) (A). Percentage (B) and number (C) of pT_FH cells during febrile, critical, early convalescent (E. C.), and healthy time points. Horizontal lines represents the median for all data points, and bars indicate the interquartile range. ***, P ≤ .001; ****, P ≤ .0001.

Figure 4.

Activation of peripheral T follicular helper (pT_FH) cells during acute dengue infection. Representative flow cytometry plots for pT_FH cells expressing high and intermediate (int) PD-1 (top), coexpression of PD1 and CD38 (middle), and coexpression of OX40 and CD25 (bottom) (A). Percentage of PD-1^high, PD-1^int, PD-1⁺CD38⁺, and OX40⁺CD25⁺ pT_FH cells during febrile, critical, early convalescent (E. C.), and healthy time points (B). Horizontal lines represents the median for all data points, and bars indicate the interquartile range. ***, P ≤ .001.

We wanted to further phenotype pT_FH, and we used samples obtained at fever day 0–1 where peak activation occurred. Antibodies to CXCR3 and CCR6 were added to differentiate between T_FH1 (CXCR3⁺CCR6⁻), T_FH2 (CXCR3⁻CCR6⁻), and T_FH17 (CXCR3⁻CCR6⁺) pT_FH subsets. Antibodies to ICOS were included as an additional marker of pT_FH activation (Figure 5) [36]. Representative flow cytometry plots to demonstrate our gating strategy to define T_FH1, T_FH2, and T_FH17 cells are shown in Figure 5A. During the critical period, the majority of pT_FH and activated pT_FH (PD1⁺ICOS⁺) fall within the T_FH1 subset (CXCR3⁺CCR6⁻) (Figure 5B). The PD1⁺ICOS⁺ pT_FH also expressed CD38 (Figure 5C).

Figure 5.

Activation of peripheral T follicular helper (pT_FH) subtypes during acute dengue infection. Representative flow cytometry plots for T_FH1 (CXCR3⁺CCR6⁻), T_FH2 (CXCR3⁻CCR6⁻), T_FH17 (CXCR3⁻CCR6⁺), and CXCR3⁺CCR6⁺ type pT_FH (A). Frequency of T_FH1, T_FH2, T_FH17, and CCR3⁺CCR6⁺ cells during fever day 0–1 (open bars) (B). Frequency of activated T_FH1, T_FH2, T_FH17, and CXCR3⁺CCR6⁺ cells during fever day 0–1 (filled bars) (B). n = 5, error bars indicate the interquartile range for each time point. Representative histogram of CXCR3, CCR6, PD-1, CD38, and inducible T-cell costimulator (ICOS) expression on pT_FH subset from a healthy (black) and an acute sample (red, fever day 1) (C).

We then sought out to determine whether there are significant differences in the number of activated pT_FH in patients with primary versus secondary infection and DF versus DHF (Figure 6). During the critical phase (fever day 0 to +1), the mean (2°:3.46, 1°:2.96, 2°:2.93, 1°:2.59 log₁₀ cells/mL) and median (2°:3.57, 1°:3.18, 2°:3.00, 1°:2.75 log₁₀ cells/mL) number of PD-1^high pT_FH and OX40⁺CD25⁺ pT_FH, respectively, was significantly higher in patients with secondary versus primary DENV infections (Figure 6A and D). During early convalescence, the mean (2°:3.66, 1°:3.39 log₁₀ cells/mL) and median (2°:3.63, 1°:3.51 log₁₀ cells/mL) number of PD-1^int pT_FH was significantly higher in patients with secondary versus primary DENV infections (Figure 6B). The mean (DHF:3.04, DF:2.74 log₁₀ cells/mL) and median (DHF:3.06, DF:2.91 log₁₀ cells/mL) number of OX40⁺CD25⁺ pT_FH was also significantly higher in patients with DHF versus DF during the critical phase (Figure 6D).

Figure 6.

Number of activated peripheral T follicular helper (pT_FH) cells during acute dengue infection. Number of pT_FH cells expressing high PD-1 (A) and intermediate (int) PD-1 (B), PD1⁺ and CD38⁺ (C), and OX40⁺ and CD25⁺ (D) during febrile, critical, and early convalescent (E. C.) time points. Donors diagnosed primary (open triangle) or secondary (filled triangle) and dengue fever (DF) (open circle) or dengue hemorrhagic fever (DHF) (filled circle) infections are shown. Horizontal lines represents the median for all data points, and bars indicate the interquartile range. *, P ≤ .05; **, P ≤ .01.

Correlation of Plasmablast Frequencies With Peripheral T Follicular Helper Cells

T follicular helper cells can induce the differentiation and activation of B cells [37]. As an indication of B-cell activation in vivo, we measured the frequencies of plasmablasts (CD38⁺CD27⁺ B cells) during acute DENV infection; additional PBMCs were available for this analysis from 13 patients (55 samples). We found an expansion of plasmablasts during acute infection (Figure 7A and B). In addition, we found a significant positive correlation between the frequency of activated pT_FH (PD-1^high pT_FH and PD-1⁺CD38⁺ pT_FH) and the frequency of plasmablasts (Figure 7C). The correlation between the frequency of PD-1^high pT_FH and the frequency of plasmablasts had a correlation coefficient of r = 0.41 (P = .0063). The correlation between the frequency of PD-1⁺CD38⁺ pT_FH and the frequency of plasmablasts had a correlation coefficient of r = 0.51 (P = .0005).

Figure 7.

Correlation between activated peripheral T follicular helper (pT_FH) cells and plasmablasts during acute dengue illness. Representative flow cytometry plots for plasmablast cells gated on C19⁺ subset and expressing CD38 and CD27 (A). Percentage of plasmablast cells during febrile, critical, and early convalescent (E. C.) time points (B). Linear correlation between PD-1^high pT_FH cells and plasmablasts and PD-1⁺ and CD38⁺ pT_FH cells and plasmablasts (C). Horizontal lines represents the median for all data points, and bars indicate the interquartile range. ***, P ≤ .001; ****, P ≤ .0001.

DISCUSSION

A deeper knowledge of the timing of T-cell activation is essential to understanding its contribution towards the pathology of dengue. In this sudy, we report the first longitudinal analysis of pT_FH cell activation during acute dengue infection. Our study used PBMCs from children in Thailand undergoing acute DENV infections. Previous studies have primarily focused on CD8 T-cell activation with contrasting findings. Using HLA-DR, CD69, and CD38 to mark activated T cells, very few or many activated CD8 T cells were reported during the febrile phase of dengue illness [15, 18, 38, 39]. We found a greater expansion of CD8 T cells relative to CD4 T cells; however, the absolute numbers of cells per microliter increased over the course of illness for both CD8 and CD4 T cells, with peak numbers detected during the critical phase of illness. Our results support the findings of significant CD8 and CD4 T-cell activation during the febrile phase with peak activation observed during the critical phase. In contrast to CD8 and CD4 T cells, pT_FH exhibited a distinct activation profile with minimal to moderate activation during the febrile phase and greater activation during the critical phase of illness.

We found an increased frequency and number of pT_FH (defined as CXCR5⁺CD45RA⁻CD4⁺ T cells) during acute infection. The frequencies of activated pT_FH varied according to the activation marker(s) used, with high expression of PD-1 or coexpression of PD-1 and CD38 demonstrating the greatest differences in pT_FH activation when compared with healthy time points. The frequencies of PD-1^high and PD-1⁺CD38⁺ pT_FH were highest starting on the day before defervescence and lasting during the critical phase of illness (fever day 0 to +1). Because we had PBMCs obtained at multiple time points during acute DENV infection from each individual analyzed in this study, we were able to identify a peak frequency of activated pT_FH during the critical period. Furthermore, the average number of activated pT_FH were higher in patients with secondary compared with primary DENV infections and in patients with severe versus mild disease. In support of our findings, a recent in vitro study found an increase in pT_FH activation (CXCR5⁺ and PD-1⁺ T cells) when naive CD4 T cells are cocultured with DENV-infected dendritic cells [40]. Altogether, our data suggest that pT_FH activation may be a novel marker for the onset of the critical phase of illness.

There are several mechanisms by which activated pT_FH could contribute to the pathogenesis of severe dengue. Similar to germinal center T_FH cells, pT_FH express CD40L, which is a strong activator of B cells [41, 42]. Studies in human immunodeficiency virus and malaria have demonstrated the ability of pT_FH to activate and help B cells [31, 43]. Peripheral T_FH cell activation of B cells could increase the production of DENV-specific antibodies, which could contribute to plasma leakage through multiple proposed mechanisms, including antibody-mediated enhancement of DENV infection, immune complex formation, and binding to cell-surface NS1 protein [44–47]. Futhermore, our previous work and other published studies have shown that secondary DENV infections induce a potent plasmablast response [48–50]. Our current work extends these observations by demonstrating a correlation between frequencies of pT_FH and plasmablasts. In addition, activated B cells can produce proinflammatory cytokines such as IL6 and TNF-α [51, 52]. Further investigation is required to understand the contribution of pT_FH to B-cell production of proinflammatory cytokines and subsequent cytokine-mediated immunopathology during the critical phase of illness.

In addition to mechanisms of pT_FH-mediated pathology of dengue, there is also great interest in pT_FH as a predictor of neutralizing antibody titer and disease severity. A recent study of individuals with dengue reported that an increased risk for DHF occurred within a specific range of antibody titer [53]. Studies suggest that the breadth and magnitude of pT_FH responses can influence the quality of antibody responses. Human immunodeficiency virus-infected patients with high titers of neutralizing antibodies expressed a larger frequency of memory pT_FH cells (PD-1⁺CXCR3⁻CXCR5⁺CD45RA⁻) compared with patients with low titers of neutralizing antibodies [54]. Alternatively, patients with malaria, with suboptimal antibody responses, had preferential activation of a lower functioning subset of pT_FH cells (PD-1⁺CXCR3⁺CXCR5⁺) [31]. These findings suggest that the activation of different pT_FH subsets may alter the quality of DENV-specific antibodies and possibly influence a subsequent infection and disease outcome.

Our study is subject to several limitations. Because PBMCs were limited, we were unable isolate pT_FH and directly demonstrate their functional capacity ex vivo. One study has identified IL-21 secretion from pT_FH cells after stimulating PBMC from convalescent DENV patients (60 to 120 days postinfection) with DENV peptides [25]. We were unable to detect IL-21 secretion in CD4⁺ T cells by intracellular cytokine staining, but this could be attributed to the transient suppression of T-cell responses observed during acute dengue infections. Considering that our patient cohort comprised children, it is unclear whether these findings would apply to adults. Our sample size was also too small to compare pT_FH responses according to DENV serotype. Nevertheless, our study design had significant statistical power to detect differences in pT_FH activation in patients with primary versus secondary DENV infection and those with DF versus DHF.

CONCLUSIONS

We conclude that pT_FH are highly activated during acute DENV infection. Further work is needed to investigate how pT_FH influence antibody production in patients before and after vaccination and natural infection. This work will provide a better understanding of pT_FH and potentially identify novel immunological indicators of dengue pathology or protection.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Presented in part: 66th American Society for Tropical Medicine and Hygiene, November 2017, Abstract no. 123, Baltimore, MD.

Notes

Authors’ contributions. K. H., A. L. R., and A. M. conceived and designed the experiments and wrote the manuscript text. K. H. performed experiments and prepared figures. K. H. and S. P. conducted statistical analyses. S. G. and S. K. supervised the clinical study, subject enrollment, and collection of clinical data and blood samples. A. S. and L. M. contributed to the analysis of clinical, virologic, and serologic data. All authors contributed to the final manuscript and agree with the results and conclusions.

Acknowledgments.  We thank the donors who generously provided peripheral blood mononuclear cells for use in our studies. We also thank Dr. Marcia Woda for technical expertise and helpful discussion.

Disclaimer.   Material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its presentation and/or publication. The opinions or assertions contained herein are the private views of the author, and are not to be construed as official, or as reflecting true views of the Department of the Army or the Department of Defense. The investigators have adhered to the policies for protection of human subjects as prescribed in AR 70–25. The opinions expressed are those of the authors and do not represent official positions of the National Institutes of Health.

Financial support. This work was funded by National Institutes of Health grant P01AI034533, the Military Infectious Disease Research Program, and utilized core facilities supported by NIH COBRE grant P20 GM10431.

Potential conflicts of interest.  A. L. R. has received compensation as a member of the Scientific Advisory Board on dengue vaccines for Sanofi Pasteur. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.