Dengue fever can be caused by one of four distinct dengue virus (DENV) serotypes that cocirculate in many parts of the world. Point of care serotype-specific nonstructural protein-1 (NS1) capture assays for the rapid serotyping of DENV in human sera would greatly support epidemiological surveillance and potentially also prognosis in individual patients. To ensure both serotype specificity and broad coverage of variants within serotypes, we have applied an innovative approach for the generation and selection of serotype-specific anti-NS1 mAbs. To elicit mAbs against conformational epitopes, NMRI mice were immunized with living HEK 293 transfectants expressing the native target Ags in multiple display on the cell surface. For each serotype, three different NS1 sequence variants were sequentially used for immunization of mice, hybridoma selection, and capture assay development, respectively. Selection of optimal combinations of capturing and detecting mAbs yielded highly sensitive and specific NS1 serotyping ELISAs (st-ELISAs) for the four serotypes. st-ELISA testing of 41 dengue patient sera showed a 100% concordance with the serotype determined by serotype-specific reverse transcriptase real-time quantitative PCR. The respective NS1 variants could be detected for ∼10 d after the onset of illness. Ab-dependent enhancement of DENV infections may be associated with a specific range of pre-existing anti-DENV serological Ab titers. Testing of patient sera with the developed st-ELISAs will not only be useful for epidemiological studies and surveillance, but it may also help to develop and validate assays that can distinguish protective versus enhancing Ab responses for risk assessment for the development of severe dengue disease in individual patients.

Dengue is regarded as the most important arboviral disease of our time, expanding geographically with increasing frequency and magnitude of outbreaks (1, 2). Of the 390 million estimated annual dengue cases, approximately one fourth manifest clinically with dengue fever or severe, life-threatening conditions such as hemorrhage and shock (3, 4). Because timely, adequate management of severe dengue cases can greatly reduce mortality (4), early diagnosis and monitoring of patients is crucial. Approaches to assess the risk of individual patients to develop severe dengue would be most valuable to permit appropriate treatment without delay. Four dengue virus (DENV) serotypes (DENV-1 to -4) are distinguished that differ by ∼25–40% at the amino acid level. Considerable variation is also observed within each of the four serotypes, which can vary from one another by up to 3% at the amino acid level (5, 6). Secondary infection with a heterologous serotype has been established as one of the main risk factors for severe disease progression (712) and has been attributed to enhanced viral replication associated with a specific range of pre-existing anti-DENV Ab titers (13, 14). Therefore, the quantitative analysis of serological anti-DENV responses is thought to contribute to risk assessment for the development of severe dengue in individual patients. Because only those cross-reactive Ab titers recognizing the infecting, heterologous serotype should be considered for such a quantitative analysis, a point of care test for DENV serotyping may support development of serological tests that can differentiate between protective and enhancing Ab responses. Furthermore, a high-throughput serotyping method would be an important tool for epidemiological surveillance and outbreak investigations.

Dengue is classified by the World Health Organization (WHO) as a neglected tropical disease and predominantly affects resource-limited countries (15). Therefore, an ideal serotyping test should be inexpensive, rapid, and simple to perform without the need for sophisticated equipment. Ag detection assays fulfill these criteria and are more suitable as rapid diagnostic test formats than approaches involving viral isolation or nucleic acid detection. Whereas serological assays reflect the history of dengue infections of an individual, Ag capture assays provide information on acute infection. DENV nonstructural protein-1 (NS1) is detectable in blood of infected patients from the first day after the onset of illness up to about day 14 (16) and thus constitutes a useful biomarker for early diagnosis of acute infections. Although several tests based on the general detection of NS1 are commercially available (17), none of them allows for a differentiation of the distinct DENV serotypes.

In a previous study, we have described a novel strategy for the generation of serotype-specific mAbs against DENV NS1, which is based on immunization of mice with living mammalian cells (HEK 293 cell line) expressing the target Ag on the cell surface. Multiple display of the viral Ag on HEK 293 transfectants promotes efficient stimulation of B cells expressing Abs recognizing native epitopes as cell surface receptors. One main advantage of this strategy is that it does not require target Ag purification, which bears the danger of perturbing the native protein structure. By immunizing mice using this approach, we have initially generated panels of serotype-specific mAbs against the NS1 protein of DENV-1 (D1NS1) and DENV-4 (D4NS1) (18). In this study, we aimed at the development of highly specific NS1 capture assays for all four DENV serotypes for the rapid screening of dengue patient sera.

Immunization of female NMRI mice was performed in accordance with the rules and regulations for the protection of animal rights (“Tierschutzverordnung”) of the Swiss Bundesamt für Veterinärwesen. Ethical clearance to analyze anonymized human dengue patient sera was obtained from the Ethics Committee of the Medical Faculty of the Heidelberg University Hospital (“Ethikkommission der Medizinischen Fakultät Heidelberg, Germany”) (reference no. 586/2017).

Immunization strategy.

In a previous study, we have generated panels of serotype-specific mAbs against D1NS1 and D4NS1 (18). This was achieved by applying an efficient mouse immunization strategy using mammalian cells expressing either the D1NS1 or the D4NS1 target Ag on the cell surface, as described earlier (19). In this study, we have used the same strategy to generate panels of serotype-specific mAbs against D2NS1 and D3NS1. Briefly, HEK 293 cells were transfected with the expression plasmid pcDNA3.1 (Invitrogen, San Diego, CA) that was modified to supply inserted genes encoding the NS1 protein of the four DENV serotypes with the secretion signal sequence of bee venom melittin (BVM) as well as a transmembrane (TM) domain (pcDNA3.1_BVM_D1NS1_FLAG_TM_HIS, pcDNA3.1_BVM_D2NS1_FLAG_TM_HIS pcDNA3.1_BVM_D3NS1_FLAG_TM_HIS and pcDNA3.1_BVM_D4NS1_FLAG_TM_HIS) (19), thereby enabling the display of target Ags hooked onto the cell surface. Addition of a FLAG and a hexa-His tag facilitated expression analyses of the target proteins (18). HEK cells (50 ml with a cell density of 106 cells/ml) were transfected with 50 μg of plasmid DNA using 150 μl of Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA). Transfected cells were harvested after 48 h and stored in freezing medium (50% FBS, 40% FreeStyle medium, and 10% dimethyl sulfoxide) at −80°C to preserve the viability of the transfected cells. Before the immunization of NMRI mice, stored aliquots of transfected HEK cells were thawed, washed, and resuspended in 0.9% sodium chloride. Mice were divided into four groups with each receiving i.v. injections of HEK cells expressing one of the four DENV NS1 variants. Each mouse received 106 living transfected HEK cells per dose in two cycles of 4 consecutive days with a break between cycles of 1 wk. Mice showing strong anti-NS1 Ab responses were selected to receive a final boost of 106 transfected HEK cells on 2 consecutive days. Two days after the last injection, selected mice were sacrificed and their spleens were aseptically removed.

Generation and selection of monoclonal hybridoma cell lines.

Mouse spleen cells and PAI myeloma cells were fused to generate Ab-producing hybridoma cell lines as described previously (20). After ∼1 wk, cell culture supernatants were screened for the presence of anti-NS1 Abs against the respective DENV serotypes by indirect ELISA. For that purpose, 96-well Nunc Immuno MaxiSorp plates (Thermo Scientific, Roskilde, Denmark) were coated overnight at 4°C with 50 μl of PBS containing recombinant DENV NS1 (AbD Serotec, Puchheim, Germany) of the respective DENV serotypes at a concentration of 1 μg/ml. NS1 proteins purchased from AbD Serotec are purified recombinant proteins expressed in HEK 293 cells and are therefore presented as native proteins with adequate posttranslational modifications. On the next day, wells were washed with dH2O containing 0.01% Tween 20 and were blocked with 5% nonfat dry milk in PBS for 1 h at 37°C. Subsequently, wells were incubated with hybridoma cell culture supernatant or diluted mouse antisera (as a positive control) for 1 h at 37°C. After washing as described, wells were incubated with HRP-conjugated goat anti-mouse IgG Abs (γ-chain specific; SouthernBiotech, Birmingham, AL) for 1 h at 37°C. Plates were washed and developed with tetramethylbenzidine microwell peroxidase substrate (KPL, Gaithersburg, MD). The reaction was stopped using 0.5 M sulfuric acid and absorbance was measured at 450 nm in a microplate reader (Tecan Sunrise, Grödig, Austria). Hybridoma cell lines with positive cell culture supernatants were selected and repeatedly cloned by limiting dilution.

Cloned hybridoma cell lines were expanded to produce milligram amounts of mAbs. Mouse Ig isotypes and subclasses for all mAbs were determined by indirect ELISA with isotype-specific reagents (SouthernBiotech). All mAbs were of the IgG isotype and different subclasses (IgG1, IgG2a, IgG2b, and IgG3) were identified (Supplemental Table I). mAbs were purified with HiTrap protein A columns according to the manufacturerʼs instructions (GE Healthcare, Uppsala, Sweden) using a low-pressure liquid chromatography system (Model EP-1 Econo pump; Bio-Rad Laboratories). After purification, the collected eluate containing the mAbs was dialyzed in Slide-A-Lyzer dialysis cassettes (Thermo Scientific, Rockford, IL) overnight against PBS and sterile filtered. One aliquot of each mAb at a concentration of 2 mg/ml was biotinylated using EZ-Link Sulfo-NHS-LC-Biotin according to the description of the manufacturer (Thermo Scientific).

To achieve serotype specificity as well as coverage of intraserotype diversity of mAbs, we used for each DENV serotype three antigenically diverse NS1 sequence variants of the same serotype for immunization of mice, hybridoma selection, and capture assay development, respectively. To ensure generation of and selection for mAbs against adequately folded DENV NS1 proteins, HEK cell–expressed target Ags were applied for both immunization of mice and hybridoma selection. For the immunization of mice, NS1 protein variants of recent outbreak strains from South America isolated between 2007 and 2013 (accession nos. KF672760, JX286520, JF808121, and KP188566) were used. Hybridoma selection was performed on recombinant NS1 variants of older isolates from Asia or the Caribbean (AbD Serotec) including D1NS1 of strain Nauru/Western Pacific/1974, D2NS1 of strain Thailand/1984, D3NS1 of strain Sri Lanka/2000, and D4NS1 of strain Dominica/814669/1981. Selected mAbs against the native NS1 protein of the four different serotypes were then further analyzed with whole protein lysates of Huh7 cells infected with the WHO reference strains for DENV-1 (strain Hawaii), DENV-2 (strain New Guinea C), DENV-3 (strain H87), and DENV-4 (strain H241) (21) provided by Progen Biotechnik (Heidelberg, Germany). A maximum likelihood phylogenetic tree illustrating the sequence diversity of the NS1 variants used in this study in context with publicly available sequence variants of the different serotypes was constructed using the Jones–Taylor–Thornton plus G amino acid substitution model (22) contained in MEGA6 (23) (Fig. 1). Zika virus (ZIKV) strains MR766 and H/PF/2013, obtained from the European Virus Archive (Marseille, France) were used to analyze interspecies cross-reactivity of the generated mAbs.

Characteristics of mAbs generated against D1NS1 and D4NS1 were described previously (18). To analyze the serotype specificity of mAbs generated by immunizing with D2NS1 and D3NS1 HEK 293 transfectants, we performed Western blot analyses with whole protein lysates of the transfected HEK cells. To prepare protein lysates, frozen transfected HEK cells were thawed, washed with PBS, and resuspended in RIPA buffer containing 20 mM Tris HCl, 137 mM NaCl, 10% glycerol, 1% Nonidet P-40, 0.25% sodium deoxycholate, and a protease inhibitor mixture (cOmplete Mini; Roche, Mannheim, Germany). RIPA buffer was chosen owing to its nonionic detergents, enabling efficient cell lysis and protein solubilization while avoiding protein degradation and interference with the immunoreactivity of the proteins. After incubation for 30 min on ice, the mix was centrifuged at 10,000 rpm for 10 min and the supernatant was stored at −20°C. For SDS-PAGE, 20 μg of each HEK cell lysate was separated on NuPAGE Novex 4–12% Bis-Tris Zoom gels (Invitrogen) under nonreducing conditions (no heating of samples, no addition of reducing agents). Proteins were subsequently transferred onto nitrocellulose membranes using an iBlot gel transfer device (Invitrogen). Membranes were blocked with 5% nonfat dry milk in PBS and cut into strips. Membrane strips were then incubated with anti-DENV NS1 mAbs, washed with PBS containing 0.05% Tween 20, and thereafter incubated with HRP-conjugated goat anti-mouse IgG Abs (γ-chain specific; SouthernBiotech). Bands were visualized by chemiluminescence using ECL Western blotting substrate (Thermo Scientific).

DENV-1, -3, and -4 WHO reference strains as well as the two ZIKV strains were passaged once on C6/36 cells and stocks were prepared by virus amplification in Vero E6 cells. The DENV-2 virus stock was produced by electroporation of BHK-21 cells with in vitro transcripts and further amplification of the BHK-21 supernatant in Vero E6 cells. Virus stock titers were determined by plaque assay as previously described (18). To analyze the reactivity of mAbs, Huh7 cell lysates were produced. For that purpose, 106 Huh7 cells were infected with a multiplicity of infection of 10 for 36 h (DENV-1, -2, and -4 and ZIKV MR766 and H/PF/2013) and 48 h (DENV-3). Cells were lysed with 100 μl of ice-cold lysis buffer (50 mM Tris HCl [pH 7.4], 150 mM NaCl, 1% Triton X-100, 60 mM β-glycerophosphate, 15 mM 4-nitrophenylphosphate, 1 mM sodium orthovanadate, 1 mM sodium fluoride, and a protease inhibitor mixture).

One hundred forty-three human blood serum samples from 60 laboratory-confirmed dengue patients presenting to the Tropical Medicine Outpatient Unit of the Heidelberg University Hospital (Germany) between 2009 and 2017 were analyzed. The initial diagnosis of patients was made by testing for the presence of the NS1 Ag using the Platelia Dengue NS1 Ag ELISA kit (Bio-Rad Laboratories), by serological analyses of anti-DENV IgG and IgM Abs with Panbio ELISA kits (Alere), or by a commercial PCR assay (Altona Diagnostics). Patients were travelers returning from different DENV endemic countries (23 from Thailand, 14 from Indonesia, 8 from India/Sri Lanka, 9 from Latin America and the Caribbean, and 6 from other countries) with an acute febrile illness consistent with dengue fever. Detailed patient characteristics, including the dengue endemic countries visited before the onset of illness, the days of illness before presenting to the hospital, the age and sex of the patients, the type of test applied for their initial diagnosis, and the severity of disease, are listed in Table I. For several of the patients, serial samples taken at different time points after the onset of illness were available.

The aim of this study was to develop DENV serotype-specific NS1 capture assays. For this purpose, tandems of the generated anti-DENV NS1 mAbs, acting as capturing and detecting reagents for the different serotypes, were used in a sandwich ELISA procedure. For each of these assays, Nunc Immuno MaxiSorp 96-well plates (Thermo Scientific) were coated with 5 μg/ml capturing mAb (50 μl per well) in PBS and incubated overnight at 4°C. On the next day, plates were washed four times with washing buffer (0.05% Tween 20 in dH2O) prior to being incubated with blocking buffer (5% nonfat dry milk in PBS) for 1.5 h at room temperature. After washing as described above, wells were incubated with lysates of DENV-infected Huh7 cells or with serum samples from DENV patients diluted in different sample/reagent buffers (PBS; 0.5% nonfat dry milk in PBS; or 0.5% nonfat dry milk in PBS containing 0.05% Tween 20) for 2 h at room temperature. Following an additional washing step, 50 μl of a corresponding biotinylated detection mAb (5 μg/ml) with the same serotype specificity was added to the wells in the different sample/reagent buffers and incubated for 1.5 h at room temperature. After washing as described above, 50 μl of HRP-labeled streptavidin (Sigma-Aldrich, St. Louis, MO) diluted 1:2000 in sample/reagent buffer was added and incubated for 1 h at room temperature. Plates were washed and tetramethylbenzidine peroxidase substrate solution (KPL) was added. After 10 min the reaction was stopped with 2 M sulfuric acid and absorbance was measured at 450 nm with a microplate reader (Tecan Sunrise). ELISA results were illustrated using GraphPad Prism version 6.0 (GraphPad Software, San Diego, CA).

Serotype-specific reverse transcriptase quantitative PCR (qPCR) assays were performed as described by Alm et al. (24). Briefly, viral RNA was extracted from 140 μl of DENV patient sera using the QIAamp viral RNA extraction kit (Qiagen, Hilden, Germany) in accordance with the manufacturer’s instructions. DENV serotype-specific reverse transcriptase qPCR assays were performed in 25 μl of reaction mixtures containing 5 μl of template RNA, TaqMan fast virus one-step master mix (Applied Biosystems, Foster City, CA), 0.9 μM each primer, and 0.2 μM minor groove binding probe labeled with a FAM reporter dye and a nonfluorescent quencher. Amplification was performed in an Applied Biosystems StepOne Plus qPCR system (Life Technologies, Singapore, Singapore) with the following parameters: reverse transcription at 50°C for 5 min, inactivation at 95°C for 20 s, followed by 45 cycles of fluorescence detection at 95°C for 3 s and annealing at 60°C for 30 s.

Immunization of mice with transfected HEK cells expressing the NS1 protein of the four different DENV serotypes and generation of monoclonal hybridoma cell lines has yielded 36 mAbs against D1NS1 (8 mAbs), D2NS1 (11 mAbs), D3NS1 (13 mAbs), and D4NS1 (4 mAbs). Whereas characteristics of the mAbs produced against D1NS1 and D4NS1 were published previously (18), DENV serotype specificity and cross-reactivity of mAbs generated in this study by immunization with D2NS1 and D3NS1 HEK 293 transfectants was determined by Western blot analysis using whole protein lysates of transfected HEK cells (Supplemental Fig. 1). DENV serotype-specific sandwich ELISAs were developed by testing whole protein lysates from Huh7 cells infected with the DENV-1 to DENV-4 WHO reference strains (Fig. 1) in a standard ELISA procedure. In a first step, all mAbs were tested both as capturing and detecting reagents only for the respective DENV serotype, as shown for D1NS1 serotyping ELISAs (st-ELISAs) in Fig. 2. Based on their capability of reacting with the NS1 protein of the corresponding DENV serotype most efficiently, and based on the prerequisite that at least one of the applied mAbs was entirely serotype specific, mAb combinations NR1.6 and NR1.4 (Fig. 2), NR2.11 and NR2.17, NR3.1 and NR3.4, as well as NR4.3 and NR4.2 were selected as capturing and detecting mAbs, respectively. In a second step, serotype specificity of the assays was analyzed by testing each of the four selected mAb combinations with all four Huh7 cell lysates. All four Ag-capture assays were serotype specific, as indicated by a complete lack of response to the heterologous serotypes. Moreover, none of the four assays cross-reacted with the NS1 protein of the antigenically closely related ZIKV, present in lysates from Huh7 cells infected with two different ZIKV strains (Fig. 3).

FIGURE 1.

Phylogenetic reconstruction of DENV NS1 protein sequences. A maximum likelihood phylogenetic tree including the 12 DENV NS1 protein sequences (352 aa positions) used in this study together with 37 publicly available sequences covering different DENV serotypes (D1–D4) and subtypes was constructed with 1000 bootstrap replicates using the Jones–Taylor–Thornton plus G amino acid substitution model contained in MEGA6. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Bootstrap values >80% are shown along the branches. Sequences of NS1 protein variants used for the immunization of mice (●) and for the screening for anti-NS1 Abs (▪) as well as NS1 sequences of the DENV-1 to -4 WHO reference strains used for the development of the st-ELISAs (▴) are colored according to their serotype. Sequences were retrieved from GenBank, and accession numbers are given in the tree. Scale for genetic distance, 0.05.

FIGURE 1.

Phylogenetic reconstruction of DENV NS1 protein sequences. A maximum likelihood phylogenetic tree including the 12 DENV NS1 protein sequences (352 aa positions) used in this study together with 37 publicly available sequences covering different DENV serotypes (D1–D4) and subtypes was constructed with 1000 bootstrap replicates using the Jones–Taylor–Thornton plus G amino acid substitution model contained in MEGA6. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Bootstrap values >80% are shown along the branches. Sequences of NS1 protein variants used for the immunization of mice (●) and for the screening for anti-NS1 Abs (▪) as well as NS1 sequences of the DENV-1 to -4 WHO reference strains used for the development of the st-ELISAs (▴) are colored according to their serotype. Sequences were retrieved from GenBank, and accession numbers are given in the tree. Scale for genetic distance, 0.05.

Close modal
FIGURE 2.

Development of DENV NS1 st-ELISAs. Sandwich ELISAs were developed by systematically analyzing the capability of different combinations of capturing (shown in panels) and detecting (shown in legends) mAbs to specifically identify NS1 serotype variants in whole protein lysates from Huh7 cells infected with DENV-1 to -4 WHO reference strains (accession nos. KM204119, KM204118, KU050695, and KR011349). This is exemplarily shown for the analysis of DENV-1 whole protein lysate (in micrograms per milliliter) using the eight mAbs (NR1.1 to NR1.8) generated by immunizing mice with D1NS1 transfectants. One mAb (in this graph, NR4.1) generated by immunization of mice with the NS1 protein of a different DENV serotype was included as a negative control. One of the mAb combinations yielding the highest level of NS1 detection for the respective serotypes (for D1NS1, NR1.6 combined with NR1.4) was selected for further analyses.

FIGURE 2.

Development of DENV NS1 st-ELISAs. Sandwich ELISAs were developed by systematically analyzing the capability of different combinations of capturing (shown in panels) and detecting (shown in legends) mAbs to specifically identify NS1 serotype variants in whole protein lysates from Huh7 cells infected with DENV-1 to -4 WHO reference strains (accession nos. KM204119, KM204118, KU050695, and KR011349). This is exemplarily shown for the analysis of DENV-1 whole protein lysate (in micrograms per milliliter) using the eight mAbs (NR1.1 to NR1.8) generated by immunizing mice with D1NS1 transfectants. One mAb (in this graph, NR4.1) generated by immunization of mice with the NS1 protein of a different DENV serotype was included as a negative control. One of the mAb combinations yielding the highest level of NS1 detection for the respective serotypes (for D1NS1, NR1.6 combined with NR1.4) was selected for further analyses.

Close modal
FIGURE 3.

Serotype specificity of the developed NS1 st-ELISAs. Serotype specificity of the selected capturing and detecting mAb combinations that have provided the highest level of NS1 detection for the respective DENV serotype was analyzed by testing whole protein lysates (in micrograms per milliliter) from Huh7 cells infected with DENV-1 to -4 (blue, red, orange, and green) WHO reference strains (accession nos. KM204119, KM204118, KU050695, and KR011349). Additionally, species specificity of the DENV st-ELISAs was confirmed by testing whole protein lysates (in micrograms per milliliter) from Huh7 cells infected with two different ZIKV isolates (black and gray). All developed assays were serotype specific and showed no cross-reactivity with ZIKV NS1.

FIGURE 3.

Serotype specificity of the developed NS1 st-ELISAs. Serotype specificity of the selected capturing and detecting mAb combinations that have provided the highest level of NS1 detection for the respective DENV serotype was analyzed by testing whole protein lysates (in micrograms per milliliter) from Huh7 cells infected with DENV-1 to -4 (blue, red, orange, and green) WHO reference strains (accession nos. KM204119, KM204118, KU050695, and KR011349). Additionally, species specificity of the DENV st-ELISAs was confirmed by testing whole protein lysates (in micrograms per milliliter) from Huh7 cells infected with two different ZIKV isolates (black and gray). All developed assays were serotype specific and showed no cross-reactivity with ZIKV NS1.

Close modal

st-ELISAs for the four DENV serotypes were optimized for the analysis of DENV patient sera by testing different assay buffers used for the incubation of serum samples, detecting mAbs and the streptavidin/peroxidase conjugate. For this purpose, sera of patients infected with known DENV serotype, as determined by reverse transcriptase qPCR, were tested with the four st-ELISAs using different buffers for the individual steps of the overall procedure (Fig. 4). Optimal sensitivity and specificity of the st-ELISAs was achieved by incubating reagents with PBS containing 0.5% milk powder and 0.05% Tween 20. Whereas the milk powder (middle panels) helped to reduce the unspecific background signal, addition of Tween 20 led to an increase of the specific signal (right panels) (Fig. 4).

FIGURE 4.

Optimization of st-ELISAs for the analysis of DENV patient serum samples. The st-ELISAs were optimized by testing different sample/reagent buffers (PBS [left panels], PBS containing 0.5% skim milk powder [middle panels], or PBS containing 0.5% skim milk powder and 0.05% Tween 20 [right panels]). For this purpose, four serum samples (AD) from dengue patients with known infecting serotype were tested with the four st-ELISAs (blue, red, orange, and green) in a 2-fold dilution series starting at a concentration of 1:10. Incubation with PBS containing 0.5% skim milk powder and 0.05% Tween 20 provided an optimal test sensitivity and specificity and was thus used for further analyses. (E) EC indicates serum from a European individual.

FIGURE 4.

Optimization of st-ELISAs for the analysis of DENV patient serum samples. The st-ELISAs were optimized by testing different sample/reagent buffers (PBS [left panels], PBS containing 0.5% skim milk powder [middle panels], or PBS containing 0.5% skim milk powder and 0.05% Tween 20 [right panels]). For this purpose, four serum samples (AD) from dengue patients with known infecting serotype were tested with the four st-ELISAs (blue, red, orange, and green) in a 2-fold dilution series starting at a concentration of 1:10. Incubation with PBS containing 0.5% skim milk powder and 0.05% Tween 20 provided an optimal test sensitivity and specificity and was thus used for further analyses. (E) EC indicates serum from a European individual.

Close modal

Sensitivity and specificity of the developed st-ELISAs in determining the infecting DENV serotype in patient sera were assessed by analyzing dilutions of serum samples from 60 laboratory-confirmed dengue patients and comparing serotyping results with data obtained by DENV serotype-specific reverse transcriptase qPCR assays (Tables I, II). Of the 60 samples, 41 were positive in both st-ELISA and qPCR, with a 100% agreement of serotyping results (17 DENV-1, 16 DENV-2, 5 DENV-3, and 3 DENV-4 patients) (Table I), showing that the accuracy of the st-ELISA is very high. A high ELISA readout was obtained for most of the 1:10 diluted sera, indicating that 20 μl of serum may be sufficient to analyze samples with all four st-ELISAs. Whereas 14 of the 60 sera had negative test results in both of the tests, 2 and 3 were single positives in either the st-ELISA or the qPCR analysis, respectively (Table II). Ninety-three percent of the qPCR-positive samples were thus also positive in the st-ELISA (Table II), revealing a very high sensitivity of the st-ELISAs. Of the 14 samples negative in both st-ELISA and qPCR, 9 were from patients tested at the first day of illness, 2 were collected >2 wk after disease onset, and 2 were diagnosed by the presence of anti-DENV Abs. Of the three st-ELISA–negative and qPCR-positive samples, two were from patients tested at the first day of illness and one was initially diagnosed by qPCR (Table I).

Table I.
Characteristics of dengue patients included in the study as well as qPCR and st-ELISA test results of their sera
Patient IDCountries Visited before Disease OnsetDOIaAge (Years)SexInitial Diagnosis (Positive Test)Severity of DiseaseReverse Transcriptase qPCR Typest-ELISA Type
China, Thailand 19 Unknown NS1 Unknown D1 D1 
Thailand 51 Male NS1 Nonsevere D1 D1 
Malaysia 39 Male NS1 Nonsevere D1 D1 
India 35 Male NS1 Hospitalized D1 D1 
India 12 29 Female NS1 Hospitalized (in India) D1 D1 
Indonesia 47 Female NS1 Nonsevere D1 D1 
Thailand 41 Male NS1 Hospitalized D1 D1 
India 27 Male IgM + IgG Hospitalized D1 D1 
Guadeloupe 39 Female IgM + IgG Nonsevere D1 D1 
10 Borneo, Cambodia, Thailand, Vietnam 25 Female NS1 Nonsevere D1 D1 
11 Philippines 46 Female NS1 Nonsevere D1 D1 
12 Australia, New Zealand, Thailand 33 Male NS1 Nonsevere D1 D1 
13 India 43 Male PCR + NS1 Nonsevere D1 D1 
14 Sri Lanka 54 Male NS1 Nonsevere D1 D1 
15 Caribbean 24 Female NS1 Nonsevere D1 D1 
16 Indonesia 24 Male IgG Nonsevere D1 D1 
17 Indonesia, Singapore 43 Male IgM + seroconversionb IgG Nonsevere D1 D1 
18 Vietnam 25 Female NS1 Nonsevere D1 Neg 
19 Australia, New Zealand, Thailand 22 Female IgM + IgG Hospitalized D2 D2 
20 Indonesia 18 Female Seroconversion IgM Nonsevere D2 D2 
21 Indonesia 47 Male Seroconversion IgM + IgG Nonsevere D2 D2 
22 Indonesia 38 Female NS1 Hospitalized D2 D2 
23 Dominican Republic 48 Female NS1 Hospitalized D2 D2 
24 Thailand 22 Male NS1 Hospitalized D2 D2 
25 Thailand 22 Male NS1 Nonsevere D2 D2 
26 Indonesia 25 Male NS1 Nonsevere D2 D2 
27 Thailand 53 Female NS1 Nonsevere D2 D2 
28 Thailand 51 Male NS1 Hospitalized D2 D2 
29 Sri Lanka 26 Male PCR Hospitalized D2 Neg 
30 Thailand 44 Male NS1 Nonsevere Neg D2 
31 Thailand 40 Female PCR + NS1 Hospitalized D2 D2 
32 Thailand 28 Female NS1 Nonsevere D2 D2 
33 Paraguay 20 Male NS1 Nonsevere D2 D2 
34 Mexico 33 Female NS1 Nonsevere D2 D2 
35 Thailand 28 Female NS1 Nonsevere D2 D2 
36 Indonesia 21 Female IgM + IgG Nonsevere D1, D2 D2 
37 Cuba 61 Unknown NS1 Unknown Neg D3 
38 Indonesia 22 Female NS1 Nonsevere D3 D3 
39 Indonesia 31 Female NS1 Hospitalized D3 D3 
40 Thailand 27 Female NS1 Nonsevere D3 D3 
41 Colombia, Peru 43 Male NS1 Nonsevere D3 D3 
42 Indonesia 28 Male IgM + IgG Nonsevere D3 D3 
43 Venezuela 57 Female IgM Nonsevere D4 D4 
44 South-East Asia 24 Male NS1 Nonsevere D4 Neg 
45 Indonesia 31 Female NS1 Nonsevere D4 D4 
46 Martinique 25 Male IgM + IgG Nonsevere D4 D4 
47 Brazil 43 Male Seroconversion IgM + IgG Nonsevere Neg Neg 
48 India 31 Female NS1 Nonsevere Neg Neg 
49 Thailand 29 Male NS1 Nonsevere Neg Neg 
50 Indonesia 22 Female NS1 Hospitalized Neg Neg 
51 India 25 Male NS1 Hospitalized Neg Neg 
52 Indonesia 29 Male NS1 Nonsevere Neg Neg 
53 Thailand 49 Male NS1 Nonsevere Neg Neg 
54 Thailand 28 Male NS1 Nonsevere Neg Neg 
55 Thailand 11 Female NS1 Nonsevere Neg Neg 
56 Thailand 17 56 Male NS1 Hospitalized Neg Neg 
57 Thailand 28 Male NS1 Nonsevere Neg Neg 
58 Thailand 18 26 Female IgM, NS1 borderline Nonsevere Neg Neg 
59 Thailand 33 Male NS1 Nonsevere Neg Neg 
60 Myanmar 48 Male IgG, IgM borderline Nonsevere Neg Neg 
Patient IDCountries Visited before Disease OnsetDOIaAge (Years)SexInitial Diagnosis (Positive Test)Severity of DiseaseReverse Transcriptase qPCR Typest-ELISA Type
China, Thailand 19 Unknown NS1 Unknown D1 D1 
Thailand 51 Male NS1 Nonsevere D1 D1 
Malaysia 39 Male NS1 Nonsevere D1 D1 
India 35 Male NS1 Hospitalized D1 D1 
India 12 29 Female NS1 Hospitalized (in India) D1 D1 
Indonesia 47 Female NS1 Nonsevere D1 D1 
Thailand 41 Male NS1 Hospitalized D1 D1 
India 27 Male IgM + IgG Hospitalized D1 D1 
Guadeloupe 39 Female IgM + IgG Nonsevere D1 D1 
10 Borneo, Cambodia, Thailand, Vietnam 25 Female NS1 Nonsevere D1 D1 
11 Philippines 46 Female NS1 Nonsevere D1 D1 
12 Australia, New Zealand, Thailand 33 Male NS1 Nonsevere D1 D1 
13 India 43 Male PCR + NS1 Nonsevere D1 D1 
14 Sri Lanka 54 Male NS1 Nonsevere D1 D1 
15 Caribbean 24 Female NS1 Nonsevere D1 D1 
16 Indonesia 24 Male IgG Nonsevere D1 D1 
17 Indonesia, Singapore 43 Male IgM + seroconversionb IgG Nonsevere D1 D1 
18 Vietnam 25 Female NS1 Nonsevere D1 Neg 
19 Australia, New Zealand, Thailand 22 Female IgM + IgG Hospitalized D2 D2 
20 Indonesia 18 Female Seroconversion IgM Nonsevere D2 D2 
21 Indonesia 47 Male Seroconversion IgM + IgG Nonsevere D2 D2 
22 Indonesia 38 Female NS1 Hospitalized D2 D2 
23 Dominican Republic 48 Female NS1 Hospitalized D2 D2 
24 Thailand 22 Male NS1 Hospitalized D2 D2 
25 Thailand 22 Male NS1 Nonsevere D2 D2 
26 Indonesia 25 Male NS1 Nonsevere D2 D2 
27 Thailand 53 Female NS1 Nonsevere D2 D2 
28 Thailand 51 Male NS1 Hospitalized D2 D2 
29 Sri Lanka 26 Male PCR Hospitalized D2 Neg 
30 Thailand 44 Male NS1 Nonsevere Neg D2 
31 Thailand 40 Female PCR + NS1 Hospitalized D2 D2 
32 Thailand 28 Female NS1 Nonsevere D2 D2 
33 Paraguay 20 Male NS1 Nonsevere D2 D2 
34 Mexico 33 Female NS1 Nonsevere D2 D2 
35 Thailand 28 Female NS1 Nonsevere D2 D2 
36 Indonesia 21 Female IgM + IgG Nonsevere D1, D2 D2 
37 Cuba 61 Unknown NS1 Unknown Neg D3 
38 Indonesia 22 Female NS1 Nonsevere D3 D3 
39 Indonesia 31 Female NS1 Hospitalized D3 D3 
40 Thailand 27 Female NS1 Nonsevere D3 D3 
41 Colombia, Peru 43 Male NS1 Nonsevere D3 D3 
42 Indonesia 28 Male IgM + IgG Nonsevere D3 D3 
43 Venezuela 57 Female IgM Nonsevere D4 D4 
44 South-East Asia 24 Male NS1 Nonsevere D4 Neg 
45 Indonesia 31 Female NS1 Nonsevere D4 D4 
46 Martinique 25 Male IgM + IgG Nonsevere D4 D4 
47 Brazil 43 Male Seroconversion IgM + IgG Nonsevere Neg Neg 
48 India 31 Female NS1 Nonsevere Neg Neg 
49 Thailand 29 Male NS1 Nonsevere Neg Neg 
50 Indonesia 22 Female NS1 Hospitalized Neg Neg 
51 India 25 Male NS1 Hospitalized Neg Neg 
52 Indonesia 29 Male NS1 Nonsevere Neg Neg 
53 Thailand 49 Male NS1 Nonsevere Neg Neg 
54 Thailand 28 Male NS1 Nonsevere Neg Neg 
55 Thailand 11 Female NS1 Nonsevere Neg Neg 
56 Thailand 17 56 Male NS1 Hospitalized Neg Neg 
57 Thailand 28 Male NS1 Nonsevere Neg Neg 
58 Thailand 18 26 Female IgM, NS1 borderline Nonsevere Neg Neg 
59 Thailand 33 Male NS1 Nonsevere Neg Neg 
60 Myanmar 48 Male IgG, IgM borderline Nonsevere Neg Neg 
a

Days of illness before presenting to the hospital.

b

Seroconversion from Ab-negative to Ab-positive.

Table II.
Comparison of st-ELISA and reverse transcriptase qPCR test results
TotalE+/P+E/P+E+/PSensitivitya
DENV1 18 17 94% (17/18) 
DENV2 18 16 94% (16/17) 
DENV3 100% (5/5) 
DENV4 75% (3/4) 
All 46 41 93% (41/44) 
TotalE+/P+E/P+E+/PSensitivitya
DENV1 18 17 94% (17/18) 
DENV2 18 16 94% (16/17) 
DENV3 100% (5/5) 
DENV4 75% (3/4) 
All 46 41 93% (41/44) 
a

Sensitivity of serotype detection by st-ELISAs as compared with reverse transcriptase qPCR tests without taking st-ELISA–positive and PCR-negative (E+/P) samples into account.

E, st-ELISA; P, reverse transcriptase qPCR.

The NS1 protein in DENV-2, DENV-3, and DENV-4 patient sera was exclusively detected by the corresponding st-ELISAs. However, of the 17 st-ELISA–positive DENV-1 patient sera, eight were not only detected by the DENV-1–specific assay, but also showed a low level of cross-reactivity with the DENV-2–specific assay. This cross-reactivity was not related to the amount of NS1 present in the sample (as determined by the D1NS1 assay) and may thus occur only with certain sequence variants of D1NS1 (Supplemental Fig. 2).

Taken together, the developed st-ELISAs exhibited a high sensitivity and specificity for the respective DENV serotypes. Both the amounts of serum required as well as the technological demands are considerably lower for the st-ELISAs as compared with RNA extraction followed by reverse transcriptase qPCR.

For 40 of the 43 dengue patients with a positive st-ELISA result, one to four additional serum samples taken at later time points after the onset of illness were available. After identifying the infecting serotype, the subsequently taken samples were tested with the corresponding serotype-specific assay, as exemplarily shown in Fig. 5. st-ELISA results for a 1:10 dilution of the 43 sera collected at the first time point of presentation of the patients as well as for a total number of 83 sera collected at later time points are shown both for the individual serotypes (Fig. 6A) and in a combined graph (Fig. 6B).

FIGURE 5.

Titration of serial dengue patient sera. After analyzing dengue patient sera taken on the day of their first presentation in a 2-fold dilution series starting at a concentration of 1:10 with the four st-ELISAs (blue, red, orange and green), serial samples of the same patients taken at different time points after the onset of illness were tested with the corresponding st-ELISA. Representative results are shown for DENV-1 (A), DENV-2 (B), DENV-3 (C), and DENV-4 (D) patient sera. DOI, day after onset of illness.

FIGURE 5.

Titration of serial dengue patient sera. After analyzing dengue patient sera taken on the day of their first presentation in a 2-fold dilution series starting at a concentration of 1:10 with the four st-ELISAs (blue, red, orange and green), serial samples of the same patients taken at different time points after the onset of illness were tested with the corresponding st-ELISA. Representative results are shown for DENV-1 (A), DENV-2 (B), DENV-3 (C), and DENV-4 (D) patient sera. DOI, day after onset of illness.

Close modal
FIGURE 6.

Persistence of NS1 in serial serum samples from dengue patients. Serum samples taken from patients serially at different time points after the onset of illness were tested with the developed st-ELISAs. Results for 1:10-diluted patient sera are shown separately for the four serotypes (A) as well as in a cumulative graph (B). Results indicate that the infecting serotype of the patients may be detected in samples taken within ∼10 d after the onset of illness.

FIGURE 6.

Persistence of NS1 in serial serum samples from dengue patients. Serum samples taken from patients serially at different time points after the onset of illness were tested with the developed st-ELISAs. Results for 1:10-diluted patient sera are shown separately for the four serotypes (A) as well as in a cumulative graph (B). Results indicate that the infecting serotype of the patients may be detected in samples taken within ∼10 d after the onset of illness.

Close modal

Analysis of the serotype-specific detection of NS1 in serial samples indicates that the developed st-ELISAs are suitable for testing samples taken within ∼10 d after the onset of illness.

Currently, no high-throughput method for the screening of serum samples for the infecting DENV serotype is commercially available, although attempts to develop serotyping ELISAs were already reported several years ago (25). Established PCR-based serotyping methods are expensive, time-consuming, technologically demanding, and require relatively large volumes of blood. Ag capture–based assays, which are simple to perform and provide a high specificity, are proven alternatives to overcome these drawbacks. Due to the significant amino acid sequence homology of DENV serotypes (5, 6), the generation of mAbs exclusively recognizing only one of the four variants is challenging. Recently, the isolation of anti-NS1 Abs by a combination of phage display and a subtractive biopanning strategy to direct Ab selection toward serotype-specific epitopes has been reported (26). A disadvantage of this approach is that Abs lack the in vivo process of affinity maturation that occurs in natural humoral immune responses. However, mAbs with high affinity to the target Ags are a prerequisite for the successful development of sensitive rapid diagnostic tests. In the current study, we have used a strategy of mouse immunization with living transfected HEK cells expressing the native NS1 target proteins on the cell surface (18, 19), thereby presenting the natively folded epitopes efficiently to B cells expressing Abs specific for native NS1 epitopes as B cell receptors. By using this approach, we could isolate panels of serotype-specific mAbs for each of the DENV serotypes. At the same time, intraserotype sequence variation has to be considered when selecting serotype-specific anti-DENV mAbs. By using three antigenically diverse DENV strains of each serotype for immunization, screening for hybridomas, and assay development, we ensured that the developed st-ELISAs are able to detect a wide range of DENV strains within the same serotype. Optimization of st-ELISAs for the analysis of dengue patient sera was achieved not only by testing different combinations of capturing and detecting reagents but also by testing different sample buffers used for serum incubation. An optimal signal-to-noise ratio was achieved with a sample buffer consisting of PBS with 0.5% milk powder and 0.05% Tween 20. Whereas nonspecific binding was eliminated by incubating with milk proteins, we can only speculate that the increased signal observed only in combination with Tween 20 may be due to the surfactant properties of this nonionic detergent, preventing protein aggregation and thereby helping to increase accessibility of the epitopes recognized by the capturing and detecting mAbs. Serotyping of 60 dengue patient sera by the developed st-ELISAs and previously published reverse transcriptase qPCR assays (24) was successful for a total of 41 samples and yielded 100% concordant results for the identification of the infecting serotype. Two and three of the samples were either single positive in the st-ELISA or in the reverse transcriptase qPCR assay; 14 were negative in both assays. A common feature of 13 of the 17 st-ELISA–negative sera was that sera were collected either on the first day of illness (11 patients) or at a very late time point after disease onset (two patients), indicating that levels of NS1 in the blood of these patients were not high enough to return a positive test result. Of the remaining four samples, one was initially reconfirmed by PCR and two by the presence of IgG and/or IgM. Only for one of the patients could no obvious reason for the negative st-ELISA (and reverse transcriptase qPCR) test result be established.

All sera tested in the present study for the establishment and first validation of the st-ELISAs were obtained from returning travelers, who presumably had primary DENV infections. Future field evaluation of the established assays will involve larger sets of dengue patient and nondengue patient sera from dengue endemic areas. None of the mAbs generated showed cross-reactivity with ZIKV NS1.

Individuals with primary dengue only develop long-term, protective immunity against the homologous serotype responsible for the primary infection. The major population of developed Abs is cross-reactive between different serotypes, but nonneutralizing (27, 28). Upon secondary infection with a heterologous DENV serotype, pre-existing Abs may form immune complexes with the heterologous virus that facilitate Fcγ-mediated virus entry and replication in target immune cells, a phenomenon known as Ab-dependent enhancement of infection. This may explain why a second heterologous DENV infection is a well-established principal risk factor for the development of severe dengue disease (712, 29). Modifiers of dengue disease include the interval between primary and secondary DENV infections as well as the specific sequence of two successive DENV infections (8, 30). Recently, it has been shown that the concentration of pre-existing serum Abs most efficiently enhancing DENV infection (peak enhancement titer) lies within a narrow range (14). Quantitative analysis of pre-existing cross-reactive Abs in secondary dengue infections may thus represent a useful biomarker to predict the risk for developing severe dengue disease. For the development and validation of serological assays that can differentiate between protective and enhancing Abs, the st-ELISA–based rapid determination of the serotype of the secondary infection may become a very useful tool.

Moreover, rapid high-throughput screening of patient sera for the infecting serotype by st-ELISAs will be valuable for epidemiological surveillance and outbreak investigations. Considering that dengue epidemics occur mainly in resource-poor settings, the developed st-ELISA represents a suitable technique to be implemented at hospitals in the endemic countries. mAbs selected for the development of highly sensitive st-ELISAs may also be suitable for the future design of lateral flow assay-based rapid diagnostic tests that can be directly applied at field sites.

The online version of this article contains supplemental material.

Abbreviations used in this article:

BVM

bee venom melittin

DENV

dengue virus

NS1

nonstructural protein-1

qPCR

quantitative PCR

st-ELISA

serotyping ELISA

TM

transmembrane

WHO

World Health Organization

ZIKV

Zika virus.

1
Murray
,
N. E.
,
M. B.
Quam
,
A.
Wilder-Smith
.
2013
.
Epidemiology of dengue: past, present and future prospects.
Clin. Epidemiol.
5
:
299
309
.
2
Struchiner
,
C. J.
,
J.
Rocklöv
,
A.
Wilder-Smith
,
E.
Massad
.
2015
.
Increasing dengue incidence in Singapore over the past 40 years: population growth, climate and mobility.
PLoS One
10
:
e0136286
.
3
Bhatt
,
S.
,
P. W.
Gething
,
O. J.
Brady
,
J. P.
Messina
,
A. W.
Farlow
,
C. L.
Moyes
,
J. M.
Drake
,
J. S.
Brownstein
,
A. G.
Hoen
,
O.
Sankoh
, et al
.
2013
.
The global distribution and burden of dengue.
Nature
496
:
504
507
.
4
World Health Organization
.
2017
. Dengue and severe dengue factsheet. Available at: http://www.who.int/mediacentre/factsheets/fs117/en/. Accessed: November 25, 2017.
5
Holmes
,
E. C.
,
S. S.
Twiddy
.
2003
.
The origin, emergence and evolutionary genetics of dengue virus.
Infect. Genet. Evol.
3
:
19
28
.
6
Vasilakis
,
N.
,
S. C.
Weaver
.
2008
.
The history and evolution of human dengue emergence.
Adv. Virus Res.
72
:
1
76
.
7
Halstead
,
S. B.
,
S.
Rojanasuphot
,
N.
Sangkawibha
.
1983
.
Original antigenic sin in dengue.
Am. J. Trop. Med. Hyg.
32
:
154
156
.
8
Sangkawibha
,
N.
,
S.
Rojanasuphot
,
S.
Ahandrik
,
S.
Viriyapongse
,
S.
Jatanasen
,
V.
Salitul
,
B.
Phanthumachinda
,
S. B.
Halstead
.
1984
.
Risk factors in dengue shock syndrome: a prospective epidemiologic study in Rayong, Thailand. I. The 1980 outbreak.
Am. J. Epidemiol.
120
:
653
669
.
9
Kliks
,
S. C.
,
A.
Nisalak
,
W. E.
Brandt
,
L.
Wahl
,
D. S.
Burke
.
1989
.
Antibody-dependent enhancement of dengue virus growth in human monocytes as a risk factor for dengue hemorrhagic fever.
Am. J. Trop. Med. Hyg.
40
:
444
451
.
10
Guzmán
,
M. G.
,
G.
Kourí
,
L.
Valdés
,
J.
Bravo
,
S.
Vázquez
,
S. B.
Halstead
.
2002
.
Enhanced severity of secondary dengue-2 infections: death rates in 1981 and 1997 Cuban outbreaks.
Rev. Panam. Salud Publica
11
:
223
227
.
11
OhAinle
,
M.
,
A.
Balmaseda
,
A. R.
Macalalad
,
Y.
Tellez
,
M. C.
Zody
,
S.
Saborío
,
A.
Nuñez
,
N. J.
Lennon
,
B. W.
Birren
,
A.
Gordon
, et al
.
2011
.
Dynamics of dengue disease severity determined by the interplay between viral genetics and serotype-specific immunity.
Sci. Transl. Med.
3
:
114ra128
.
12
Guzman
,
M. G.
,
M.
Alvarez
,
S. B.
Halstead
.
2013
.
Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: an historical perspective and role of antibody-dependent enhancement of infection.
Arch. Virol.
158
:
1445
1459
.
13
Halstead
,
S. B.
2003
.
Neutralization and antibody-dependent enhancement of dengue viruses.
Adv. Virus Res.
60
:
421
467
.
14
Katzelnick
,
L. C.
,
L.
Gresh
,
M. E.
Halloran
,
J. C.
Mercado
,
G.
Kuan
,
A.
Gordon
,
A.
Balmaseda
,
E.
Harris
.
2017
.
Antibody-dependent enhancement of severe dengue disease in humans.
Science
358
:
929
932
.
15
Horstick
,
O.
,
Y.
Tozan
,
A.
Wilder-Smith
.
2015
.
Reviewing dengue: still a neglected tropical disease?
PLoS Negl. Trop. Dis.
9
:
e0003632
.
16
Hu
,
D.
,
B.
Di
,
X.
Ding
,
Y.
Wang
,
Y.
Chen
,
Y.
Pan
,
K.
Wen
,
M.
Wang
,
X.
Che
.
2011
.
Kinetics of non-structural protein 1, IgM and IgG antibodies in dengue type 1 primary infection.
Virol. J.
8
:
47
.
17
Hunsperger
,
E. A.
,
S.
Yoksan
,
P.
Buchy
,
V. C.
Nguyen
,
S. D.
Sekaran
,
D. A.
Enria
,
S.
Vazquez
,
E.
Cartozian
,
J. L.
Pelegrino
,
H.
Artsob
, et al
.
2014
.
Evaluation of commercially available diagnostic tests for the detection of dengue virus NS1 antigen and anti-dengue virus IgM antibody.
PLoS Negl. Trop. Dis.
8
:
e3171
.
18
Rose
,
N.
,
C. A.
Pinho-Nascimento
,
A.
Ruggieri
,
P.
Favuzza
,
M.
Tamborrini
,
H.
Roth
,
M. T.
Baroni de Moraes
,
H.
Matile
,
T.
Jänisch
,
G.
Pluschke
,
K.
Röltgen
.
2016
.
Generation of monoclonal antibodies against native viral proteins using antigen-expressing mammalian cells for mouse immunization.
BMC Biotechnol.
16
:
83
.
19
Dreyer
,
A. M.
,
J.
Beauchamp
,
H.
Matile
,
G.
Pluschke
.
2010
.
An efficient system to generate monoclonal antibodies against membrane-associated proteins by immunisation with antigen-expressing mammalian cells.
BMC Biotechnol.
10
:
87
.
20
Pluschke
,
G.
,
A.
Joss
,
J.
Marfurt
,
C.
Daubenberger
,
O.
Kashala
,
M.
Zwickl
,
A.
Stief
,
G.
Sansig
,
B.
Schläpfer
,
S.
Linkert
, et al
.
1998
.
Generation of chimeric monoclonal antibodies from mice that carry human immunoglobulin Cγ1 heavy of Cκ light chain gene segments.
J. Immunol. Methods
215
:
27
37
.
21
Añez
,
G.
,
D. A.
Heisey
,
E.
Volkova
,
M.
Rios
.
2016
.
Complete genome sequences of dengue virus type 1 to 4 strains used for the development of CBER/FDA RNA reference reagents and WHO international standard candidates for nucleic acid testing.
Genome Announc.
4
:
e01583-15
.
22
Jones
,
D. T.
,
W. R.
Taylor
,
J. M.
Thornton
.
1992
.
The rapid generation of mutation data matrices from protein sequences.
Comput. Appl. Biosci.
8
:
275
282
.
23
Tamura
,
K.
,
G.
Stecher
,
D.
Peterson
,
A.
Filipski
,
S.
Kumar
.
2013
.
MEGA6: molecular evolutionary genetics analysis version 6.0.
Mol. Biol. Evol.
30
:
2725
2729
.
24
Alm
,
E.
,
G.
Lindegren
,
K. I.
Falk
,
N.
Lagerqvist
.
2015
.
One-step real-time RT-PCR assays for serotyping dengue virus in clinical samples.
BMC Infect. Dis.
15
:
493
.
25
Puttikhunt
,
C.
,
T.
Prommool
,
N.
U-thainual
,
P.
Ong-ajchaowlerd
,
K.
Yoosook
,
C.
Tawilert
,
T.
Duangchinda
,
A.
Jairangsri
,
N.
Tangthawornchaikul
,
P.
Malasit
,
W.
Kasinrerk
.
2011
.
The development of a novel serotyping-NS1-ELISA to identify serotypes of dengue virus.
J. Clin. Virol.
50
:
314
319
.
26
Lebani
,
K.
,
M. L.
Jones
,
D.
Watterson
,
A.
Ranzoni
,
R. J.
Traves
,
P. R.
Young
,
S. M.
Mahler
.
2017
.
Isolation of serotype-specific antibodies against dengue virus non-structural protein 1 using phage display and application in a multiplexed serotyping assay.
PLoS One
12
:
e0180669
.
27
de Alwis
,
R.
,
M.
Beltramello
,
W. B.
Messer
,
S.
Sukupolvi-Petty
,
W. M.
Wahala
,
A.
Kraus
,
N. P.
Olivarez
,
Q.
Pham
,
J. D.
Brien
,
W. Y.
Tsai
, et al
.
2011
.
In-depth analysis of the antibody response of individuals exposed to primary dengue virus infection. [Published erratum appears in 2011 PLoS Negl. Trop. Dis. 5.]
PLoS Negl. Trop. Dis.
5
:
e1188
.
28
Beltramello
,
M.
,
K. L.
Williams
,
C. P.
Simmons
,
A.
Macagno
,
L.
Simonelli
,
N. T.
Quyen
,
S.
Sukupolvi-Petty
,
E.
Navarro-Sanchez
,
P. R.
Young
,
A. M.
de Silva
, et al
.
2010
.
The human immune response to dengue virus is dominated by highly cross-reactive antibodies endowed with neutralizing and enhancing activity.
Cell Host Microbe
8
:
271
283
.
29
Halstead
,
S. B.
2014
.
Dengue antibody-dependent enhancement: knowns and unknowns.
Microbiol. Spectr.
2
.
30
Graham
,
R. R.
,
M.
Juffrie
,
R.
Tan
,
C. G.
Hayes
,
I.
Laksono
,
C.
Ma’roef
,
Erlin
,
Sutaryo
K. R.
,
Porter
,
S. B.
Halstead
.
1999
.
A prospective seroepidemiologic study on dengue in children four to nine years of age in Yogyakarta, Indonesia I. studies in 1995–1996.
Am. J. Trop. Med. Hyg.
61
:
412
419
.

The authors have no financial conflicts of interest.

Supplementary data