Dengue NS1 Antibodies Are Associated With Clearance of Viral Nonstructural Protein-1

Abstract

Background

Dengue vascular permeability syndrome is the primary cause of death in severe dengue infections. The protective versus potentially pathogenic role of dengue nonstructural protein-1 (NS1) antibodies are not well understood. The main goal of this analysis was to characterize the relationship between free NS1 concentration and NS1 antibody titers in primary and secondary dengue infection to better understand the presence and duration of NS1 antibody complexes in clinical dengue infections.

Methods

Hospitalized participants with acute dengue infection were recruited from Northern Colombia between 2018 and 2020. Symptom assessment, including dengue signs and symptoms, chart review, and blood collection, was performed. Primary versus secondary dengue was assessed serologically. NS1 titers and anti-NS1 antibodies were measured daily.

Results

Patients with secondary infection had higher antibody titers than those in primary infection, and there was a negative correlation between anti-NS1 antibody titer and NS1 protein. We demonstrate that in a subset of secondary infection, there were indeed NS1 antigen-antibody complexes on the admission day during the febrile phase that were not detectable by the recovery phase. Furthermore, dengue infection status was associated with higher circulating sialidases.

Discussion

The negative correlation between antibody and protein suggests that antibodies may play a role in clearing this viral protein.

Dengue viruses (DENVs) are among the leading causes of pediatric morbidity and mortality globally [1]. Severe dengue most commonly occurs in the setting of a second infection with a different serotype. Increased pathology during secondary infection results from poorly neutralizing antibodies, enhanced virus replication, and increased proinflammatory immune responses, all of which lead to increased vascular permeability [2–4]. Dengue vascular permeability syndrome leads to plasma leakage and shock, the primary cause of death in severe dengue infections [5]. There are several possible mechanisms leading to vascular permeability syndrome. During dengue infection, endothelial cells become activated, which can lead to plasma leakage in surrounding tissues and trigger the activation of the hemostatic system [6]. There is also growing evidence that dengue virus nonstructural protein-1 (NS1), produced in high concentrations during intracellular virus replication, contributes to vascular permeability [7–14].

NS1 serves as part of the replication complex in the form of a glycosylated dimer on intracellular membranes and the infected cell surface but it is also found as a secreted hexamer circulating in the blood during acute illness [8]. Secreted NS1 is associated with endothelial hyperpermeability and vascular dysfunction, which contribute to vascular permeability syndrome [7–12, 15, 16]. The mechanism has been described in cultured endothelial cells, demonstrating that NS1 protein increases expression of sialidases such as neuraminidase 2 (Neu2) as well as heparinases [9]. These enzymes then degrade endothelial glycocalyx components such as sialic acid and heparan sulfate, resulting in endothelial hyperpermeability. Thus, dengue NS1 may be an effective vaccine target [16] or target for monoclonal antibody therapies [17–19].

The protective versus potentially pathogenic role of dengue NS1 antibodies, however, is not well understood. There is evidence that NS1 antibodies may play an important protective effect, preventing endothelial permeability in vivo and in vitro [16], while other studies suggest that NS1 antibody and/or NS1 antigen-antibody complexes may be pathogenic. NS1 antigen-antibody complexes enhance complement activation, which is associated with severe dengue [7] and destruction of platelets. NS1 antibodies may cross-react with endothelial cells, thereby inducing apoptosis and endothelial damage [20], as well as opsonize and activate human platelets, inducing thrombocytopenia [21].

It is known that individuals with secondary dengue clear viremia and NS1 antigen earlier and faster than those with primary infection [14], and that there are increased levels of NS1 antigen-antibody complexes in secondary infection [22]. The kinetics of serum free NS1 versus NS1 protein-NS1 antibody complex in primary versus secondary dengue infection have not yet been defined. The main goal of our analysis was to characterize the relationship between free NS1 concentration and NS1 antibody titers in primary and secondary dengue infection to better understand the presence and duration of NS1 antibody complexes in clinical dengue infections. The secondary goals of this study were to investigate the relationship between NS1 protein and increases expression of sialidases in vivo and the relationship between NS1 antibody and thrombocytopenia. These analyses may give insights into the possible pathogenic role of anti-NS1 antibody and free NS1.

METHODS

Inclusion Criteria

Men, women, and children 8 years of age and older with a clinical diagnosis of dengue, including fever >38°C at enrollment and a laboratory-confirmed NS1 antigen test, were recruited from Northern Colombia from 2018 to 2020.

Ethics Statement

All participants provided written informed consent and children received written assent in addition to the consent of their parent or guardian. The study samples were collected under approval from the Universidad El Bosque Institutional Ethics Review Board (UB 387–2015) under the protocol “Surveillance of sentinel infectious events prevalent in Colombia” and the Clínica Fundación with a nonhuman subjects’ determination from the George Washington University Institutional Review Board.

Outcome Measures

Dengue signs and symptoms were reviewed, and blood collection was performed daily until 5 days after defervescence or until the patient was discharged, whichever came first. Clinical signs and symptoms collected daily included fever, headache, retroorbital pain, arthralgias, myalgias, rash, abdominal pain, vomiting, diarrhea, somnolence, hypotension, hepatomegaly, mucosal hemorrhage, hypothermia, hemoconcentration, platelets < 100 000/µL, shock, organ failure, hemorrhage with hemodynamic compromise and ascites, and pleural effusion or pericardial effusions documented by ultrasound. Individuals were stratified as dengue with alarm signs, without alarm signs, and severe dengue as per the World Health Organization diagnostic criteria [23].

Clinical Phase Definitions

The febrile phase refers to the period during which the patient was febrile (excluding the defervescence day). The critical phase includes the last day of fever (defervescence day, day 0) and the first day without fever (day 1 postdefervescence) and is named the “critical phase” as dengue patients are most likely to experience shock and risk of death during this time period. Finally, the recovery phase in this study is defined as starting on day 2 postdefervescence until the patient was discharged.

Sample Collection

After informed consent, 3 mL of blood was collected and serum was used to confirm dengue NS1 positivity on the Inbios NS1 Detect rapid test according to the manufacturer's instructions. The remaining serum was deidentified, labeled with a unique study identifier, and frozen to −80°C prior to shipping to George Washington University on dry ice for further analysis.

Samples were analyzed from various timepoints with the following definitions (Supplementary Table 2):

1. Admission day sample was collected on the first day upon hospital admission with at least 24 hours of fever.

2. Defervescence day sample was collected on the last day of fever.

3. Recovery phase sample was collected 2 days postdefervescence.

4. Last day collected sample was collected on the last day that the patient was hospitalized.

Case Classification

Primary infections were defined as NS1 antigen-confirmed dengue infections during the febrile phase that were seronegative for dengue immunoglobulin G (IgG) as measured by 2 different enzyme-linked immunosorbent assay (ELISA) formats and no neutralizing antibody detection. Secondary infections were defined as NS1 antigen-confirmed dengue infections during the febrile phase that were seropositive for dengue IgG by at least 1 of 2 different ELISA formats and had a detectable neutralizing titer for at least 1 dengue serotype. If the neutralization and IgG ELISA data were discrepant, then infections were considered secondary if the IgG ELISA was positive as IgM antibodies may provide some early neutralization in primary infection.

Data Management

All participants were assigned a unique identification number. All data was void of personal identifiers and was stored in the REDCap database at George Washington University.

RT-PCR for DENV Serotyping

The infecting DENV serotype was identified by extracting RNA from admission samples using QIamp RNA extraction kit (Qiagen). Reverse transcription polymerase chain reaction (RT-PCR) for DENV serotypes 1, 2, 3, and 4 was performed using Dengue Serotyping Real Time PCR detection kit (Viasure) according to manufacturer's instructions. Thermocycler conditions were performed as follows: 45°C for 15 minutes, 95°C for 2 minutes, and 45 cycles of 95°C for 10 second and 60°C for 50 second. Data was collected using Viia 7 Real-Time PCR (Applied Biosystems) and CFX Real Time PCR (BioRad). Samples were considered positive if the cycle threshold (Ct) value was < 40 and the internal control amplified.

Enzyme-Linked Immunosorbent Assays

For all the following ELISA assays, plasmas were heat-inactivated at 56°C for between 30 and 60 minutes.

Dengue IgG Detection

Admission samples were tested using DENV Detect IgG ELISA kit (InBios) to measure total DENV IgG antibodies. Samples and controls were diluted 1:100 and run against dengue and control antigens and analyzed according to the manufacturer's instructions.

Indirect Laboratory-Based DENV IgG and IgM Antibody Detection

ELISAs were performed as previously described [24]. Briefly, ELISA plates coated with recombinant DENV-1, 2, 3, and 4 E proteins (Aalto Bio Reagents) were blocked with B3T buffer, and incubated with 4-fold serial dilutions of sera starting at 1:400 or 2-fold serial dilutions of sera starting at 1:100, for detection of IgG and IgM, respectively, with peroxidase-conjugated goat anti-human IgG or IgM antibody (Jackson ImmunoResearch). Positivity in this assay was defined by analyzing 4 flavivirus-naive plasmas against the DENV-2 E protein. Mean optical density (OD) and standard deviations (SD) were calculated for both IgG and IgM, and used to define positivity as 3 SD above the mean after background subtraction.

NS1 Antigen Detection

NS1 antigen was measured longitudinally in all individuals with samples collected for at least 2 days postdefervescence using the Dengue Virus NS1 ELISA kit (EUROIMMUN) according with manufacturer's instructions. Plasma samples were diluted 1:2, and samples above the linear standard curve range were further serially diluted 2-fold up to 1:64.

Quantitative NS1 Antibody Detection

The ELISA procedures were described previously [25]. Briefly, 96-well microplates were coated with 1 µg/mL of NS1 (The Native Antigen Company) for each serotype or phosphate-buffered saline (PBS) as a control. Human plasma samples were diluted to 1:200, 1:2000, and 1:20 000. Purified IgG (Athens Research and Technology, Inc) was used for the standard curve, and goat anti-human IgG-horseradish peroxidase (Southern Biotech) for the secondary. Plates were read at OD 450 nm on a PerkinElmer EnSpire. Background from the PBS plate was subtracted. IgG concentrations (ng/mL) were interpolated from the standard curve using a sigmoidal nonlinear regression in GraphPad Prism 10. OD values outside the linear range (0.25 < O.D < 2.75) on the standard curve were further diluted to achieve OD values within the linear range.

Sialidase Neuraminidase 2 Concentration

The Human Sialidase 2 ELISA kit (Reddot Biotech) was used according with manufacturer's instructions. Admission or day 2 samples (if no more admission day plasma was available) were tested for Neu2 concentration with nondengue-infected controls. These Colombian controls were sex matched, and age matched within ± 15 years, except for 4 controls where the closest available age match was up to 27 years from the index case age. Plasma samples were run neat (undiluted), and, if needed, further diluted 1:6 and 1:12.

Acid Dissociation of NS1 antigen-antibody Complexes

Longitudinal samples from 14 individuals were analyzed for the presence of NS1 antigen-antibody complexes by disassociating the complexes and retesting the sample for NS1 antigen. The acid dissociation was performed as previously described [22]. Briefly, 50 μL of dissociation buffer (1.5 M glycine/hydrochloric acid, pH 2.8) was added to 50 μL of serum followed by an incubation at 37°C for 1 hour. The reaction was stopped using 50 μL of neutralizing buffer (1.5 M Tris/hydrochloric acid, pH 9.7). The samples were analyzed at a final dilution of 1:3 and 1:6 using a Dengue Virus NS1 ELISA kit (EUROIMMUN).

Dengue Neutralization Assays

Admission day and last collection day samples were tested for neutralization of 4 DENV viruses using a method described previously [26]. Briefly, DENV-1 (West-Pac 74), DENV-2 (S16803), DENV-3 (CH53489), and DENV-4 (H241) were propagated in Vero cells (American Type Culture Collection [ATCC]) and titrated on Raji-DC-SIGN cells (National Institutes of Health [NIH] AIDS Repository) to produce 20% infection of cells. Plasma was 4-fold serially diluted starting from 1:30 to 1:7680 and incubated with each serotype for 30 minutes in a humidified 37°C CO₂ incubator. Virus-only and cell-only conditions were positive and baseline controls. After incubation, DC-SIGN Raji cells were added at 1.2 × 10⁵ cells per well. Plates were washed the next day and then stained with anti-DR antibodies (BD Clone L243) conjugated to phycoerythrin, peridinin chlorophyll, or allophycocyanin (BD), for 30 minutes on ice. After washing, cells were fixed, permeabilized, and stained with a pan-dengue monoclonal antibody 2H2 purified from 2H2 hybridoma cell supernatant (ATCC) conjugated to fluorescein isothiocyanate (FITC; in house) for 30 minutes at room temperature and read on a FACS CANTO II (BD). Cells were gated by fluorochrome and 2H2-FITC. A nonlinear curve fit was used to determine the serum titer yielding 50% neutralization of infection (NT₅₀) using GraphPad Prism 8.

Statistical Analyses

We used a mixed model regression approach to estimate the correlation of log(NS1) and average log(DENV-1–4) over time [27]. This method accounts for the within-subject correlation of the repeated measures. A normal approximation was used to calculate the confidence interval using Fisher z transformation.

RESULTS

Participants’ Clinical Characterization

Hospitalized dengue participants were enrolled during the febrile phase. On a daily basis, symptoms were assessed, charts were reviewed, and blood was collected until 5 days postdefervescence. Severe cases were transferred to higher acuity regional hospitals with intensive care capabilities, and thus, no more samples were collected. The cohort was evenly distributed for sex and mainly pediatric, with a median age of 14 years (Table 1). The majority of the participants (86%) experienced dengue with alarm signs, and the mean days of fever that were documented were approximately 5 days.

Infecting Serotypes and Clinical Disease Status

Plasma from admission day was analyzed for DENV RNA and serotyped. For those whose samples were positive for RNA and could be serotyped (n = 32), the dominant infection was with DENV-1 (n = 21), while 4 were DENV-2 and 7 DENV-3 (Supplementary Table 1). Primary versus secondary infection status was assigned as described in “Methods” section. Fourteen participants were categorized as primary with the rest as secondary (n = 22) (Supplementary Table 1).

There were no differences in median age, documented fever days, or in alarm signs between participants experiencing primary or secondary infection (Table 1). Overall, there were no differences in clinical symptoms between primary and secondary infection over the 3 infection phases (Supplementary Table 2). All experienced fever with a majority of participants exhibiting classical symptoms such as headache, retroorbital pain, myalgias, and arthralgias, which all started to wane by the recovery phase. There were significantly more participants who experienced headaches at defervescence day in primary infection compared to secondary infection (Supplementary Table 2) (P = .0003). Few participants in this cohort displayed signs or symptoms of vascular permeability and no participants with ascites, pleural effusion, or pericardial effusions, shock, or severe end organ damage. One participant had hemoconcentration and 3 participants had hypotension.

NS1 Protein and Antibody Levels

The kinetics of the NS1 protein levels were determined by measuring NS1 protein concentrations in serum from patients whose sampling included at least 2 days postdefervescence (n = 24; 9 primary and 15 secondary) (Figure 1 and Supplementary Figure 1). The NS1 concentrations peaked around the day of defervescence and declined over the next 5 days in both primary and secondary infection. By 4 days postdefervescence, circulating NS1 levels had fallen to near undetectable levels in almost all participants. There were no statistical differences in NS1 levels between primary and secondary infections, although a trend toward lower levels was noted in secondary infection (mean relative units, 20 in secondary vs 117 in primary infection).

The NS1 antibody titers for all 4 dengue serotypes for each patient were determined longitudinally (Supplementary Table 3) and the average titer was calculated. We utilized a mixed model regression approach to estimate the correlation of log (NS1) and average log (anti-DENV-1–4 NS1) over time. The results demonstrated a moderate negative correlation between NS1 concentration and anti-NS1 antibodies (1–4) in both primary (r = −0.498, P < .0001) and secondary (r = −0.567, P < .0001) infection, with a lower correlation in primary infection (Figure 2A) compared to secondary infection (Figure 2B). In secondary infection, the mean anti-NS1 antibody titer was greater 2 days prior to defervescence compared to primary infection but this trend was not statistically significant (3.758 vs 2.751, P = .1028).

Role for antigen-antibody Complexes

To determine if the negative correlation between NS1 concentration and anti-NS1 antibodies was due to enhanced clearance of antigen-antibody immune complexes, we performed a dissociation assay on the admission day sample from 15 people with secondary dengue infection to measure NS1 levels before and after immune complex dissociation (Figure 3). Almost all plasma had high circulating NS1 levels, and half (8 of 15) had increased titers (>100 relative units [RU]/mL) after dissociation, indicating the presence of antigen-antibody complexes (Figure 3A). Thus, antibody-NS1 complexes were commonly detectable at this time. Four participants had such high NS1 titers that no dissociation could be detected because they were out of range, while 3 did not have detectable complex dissociation. By the recovery phase, NS1 levels were undetectable in 8 participants, and dissociation had no effect (Figure 3B), suggesting true NS1 clearance. Two participants still had detectable complexes as defined by an increase >100 RU/mL in NS1 concentration after dissociation. These data suggest that NS1 antigen-antibody complexes formed during the febrile phase led to clearance of NS1 by the postdefervescence phase, but that there was heterogeneity in this response. We selected 3 participants to test dissociation of complexes longitudinally (Supplementary Figure 2). All 3 had complexes that could be disrupted at the day of admission; however, in every subsequent plasma sample there were no complexes detected. For 2 of the 3 individuals, NS1 levels became undetectable after admission day, further supporting the hypothesis that complex formation could lead to fast protein clearance.

Because there is also a potentially pathogenic role for NS1 antibodies in mediating thrombocytopenia [21], we tested for a relationship between NS1 antibody titer and thrombocytopenia. There was no significant difference in antibody titer between patients with thrombocytopenia (platelets < 100 000/µL) on day 1 of admission and those without, suggesting that there was no pathogenic role for NS1 antibodies in mediating platelet counts (Figure 4).

Association Between Sialidase and Dengue Infection

Dengue NS1 protein has been shown to increase expression of sialidases in vitro [11]. Sialidase expression leads to consumption of the glycocalyx resulting in increased vascular permeability. There were significantly increased levels of circulating sialidase Neu2 in the plasma of dengue cases on admission day compared to uninfected control plasma (Figure 5A). We also observed that Neu2 did significantly correlate with NS1 protein (Figure 5B) but was not significantly negatively correlated with anti-NS1 antibody titer (Figure 5C).

DISCUSSION

As expected, we found that patients with secondary dengue infection had higher anti-NS1 antibody titers than those in primary infection and that there was a negative correlation between anti-NS1 antibody titer and NS1 protein, suggesting that these antibodies may play a role in clearing this viral protein. Even within the secondary infection group, patients with high antibody levels were more likely to have undetectable NS1 protein by the recovery phase. These observations are consistent with previous studies [14].

The elucidation of the role of NS1 antibodies in viral clearance and disease pathogenesis is of great importance for dengue vaccine design. There are currently 3 live attenuated dengue vaccines in clinical development, whose efficacy may depend on the types of antibodies induced: the Dengvaxia dengue vaccine [28], the Takeda dengue vaccine (TAK-003) built on a dengue serotype 2 backbone, and the Instituto Butantan, US NIH, and Merck (MSD) dengue vaccine (Butantan-DV) that carries nonstructural proteins to all 4 serotypes [28]. The effects of vaccination on vascular permeability remain to be determined. Further research to define the role of NS1 antibodies and NS1 antigen-antibody complexes from vaccination and natural infection in clinical vascular permeability is needed.

Here we describe the kinetics of NS1 antibody, free and in complex, over time during the dengue disease course. A majority of patients had detectable NS1 antigen-antibody complexes at admission day (during the febrile phase), and by the recovery phase 80% had low to undetectable levels of NS1 protein with no evidence of dissociation. We hypothesize that anti-NS1 antibodies play a dominant role in NS1 clearance during this time, although we cannot preclude a pathogenic role for these antibodies. Because these antibodies have been described as potentially enhancing complement and opsonizing platelets, both of which could lead to thrombocytopenia, we tested whether patients who developed thrombocytopenia had higher titers of anti-NS1 antibodies, but found no association. We did not, however, perform a quantitative measure of platelets. The mechanisms of thrombocytopenia in dengue are likely multifactorial and, given the presence of thrombocytopenia detected in early primary dengue infection prior to antibody production, other mechanisms should be considered. Future clinical studies should consider the specific effects of NS1 antibodies and complexes on complement and platelets.

Lastly, we investigated the pathogenic role of NS1 during dengue infection by measuring circulating sialidases, which, if induced by NS1, could degrade the endothelial glycocalyx and increase vascular permeability. Indeed, we found that during dengue infection, circulating sialidase levels were higher compared to uninfected controls. These data provide insight into expected levels of sialidases in dengue fever that may help inform calculations of effect size for future clinical trials to treat dengue vascular permeability.

Our recruitment was limited due to the coronavirus disease 2019 (COVID-19) pandemic and Colombian quarantine restrictions that coincided with the beginning of this study, and therefore we were not powered in our sample size to compare between disease severity groups. In addition, there was no clinical sample prior to dengue infection to assist in determination of primary versus secondary infection. Participants had sample collection for a variable number of days before and after defervescence depending on how rapidly they recovered, limiting the number of samples from the same time point in dengue disease course available for comparison. Finally, this article describes associations found in natural infection, and clinical trials are needed to define a causal relationship between NS1 antibody titer and NS1 antigen clearance.

In conclusion, we found a negative correlation between anti-NS1 antibodies and protein, suggesting a potential protective effect of NS1 antibody in viral clearance. We also defined the increased levels of circulating sialidases during dengue fever. Further research to define the protective versus pathogenic role of NS1 antibody from natural primary infection and after dengue vaccine administration on clinical vascular permeability, complement, and platelets is needed.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Acknowledgments. We thank Allied Research Society for their assistance in this study. P. S. and K. P. are military service members or federal/contracted employees of the United States Government. This work was prepared as part of their official duties. Title 17 U.S.C. 105 provides that “copyright protection under this title is not available for any work of the United States Government.” Title 17 U.S.C. 101 defines a US Government work as work prepared by a military service member or employee of the US Government as part of that person's official duties.

Author contributions. R. M. L., A. C., P. S., G. D., C. B., R. A., A. P., I. Z., S. H., G. S., and K. P. conceived and designed the analysis. M. P. P., E. M., P. S., C. B., L. E., K. S., C. H. G., and W. R. collected data and contributed data. M. P. P., P. S., J. G., S. N., L. H., and C. H. G. performed the analysis. M. P. P., R. M. L., and A. Y. C. wrote the paper.

Disclaimer. Contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Advancing Translational Sciences or the National Institutes of Health. The views expressed in this article reflect the results of research conducted by the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, nor the United States Government.

Financial support. This work was supported by the National Center for Advancing Translational Sciences, National Institutes of Health (grant numbers UL1TR001876 and KL2TR001877 to A. C.).

References

Author notes