A nonstructural protein 1 capture enzyme-linked immunosorbent assay specific for dengue viruses

Abstract

Dengue non-structural protein (NS1) is an important diagnostic marker during the acute phase of infection. Because NS1 is partially conserved across the flaviviruses, a highly specific DENV NS-1 diagnostic test is needed to differentiate dengue infection from Zika virus (ZIKV) infection. In this study, we characterized three newly isolated antibodies against NS1 (A2, D6 and D8) from a dengue-infected patient and a previously published human anti-NS1 antibody (Den3). All four antibodies recognized multimeric forms of NS1 from different serotypes. A2 bound to NS1 from DENV-1, -2, and -3, D6 bound to NS1 from DENV-1, -2, and -4, and D8 and Den3 interacted with NS1 from all four dengue serotypes. Using a competition ELISA, we found that A2 and D6 bound to overlapping epitopes on NS1 whereas D8 recognized an epitope distinct from A2 and D6. In addition, we developed a capture ELISA that specifically detected NS1 from dengue viruses, but not ZIKV, using Den3 as the capture antibody and D8 as the detecting antibody. This assay detected NS1 from all the tested dengue virus strains and dengue-infected patients. In conclusion, we established a dengue-specific capture ELISA using human antibodies against NS1. This assay has the potential to be developed as a point-of-care diagnostic tool.

Fig 1. Human antibodies against NS1 recognize multimeric forms of DENV NS1.

(A) Indirect ELISA demonstrating the specificity of the antibodies. Wells were coated with NS1 from various dengue viruses, blocked, and then purified antibodies were added. The bound antibodies were detected by the addition of HRP-conjugated anti-human IgG followed by TMB substrate. Average OD values ± SEM from 2 experiments, performed in duplicates are shown. (B) Immunoblot analysis. Purified NS1 from various dengue viruses were untreated or incubated at 90°C for 10 minutes, separated on an SDS-PAGE, and then transferred onto a PVDF membrane. The membranes were incubated with the purified antibodies, followed by a HRP-conjugated anti-human IgG and then chemiluminescent substrate. As a loading control, the SDS-PAGE was stained with InstantBlue^TM.

Fig 2. Analysis of association and dissociation of NS1 from various serotypes of DENV to immobilized purified IgG by BLI.

Purified antibodies were immobilized on the biosensors, the antibodies loaded biosensors were incubated in several concentrations (0, 1.56, 3.13, 6.25, 12.5, 25, 50 and 100nM) of recombinant NS1, and then the biosensors loaded with NS1 bound to antibodies were immersed in buffer to allow for the dissociation of the NS1 from the antibodies. The graph shows raw (black line) and fitted (color lines) binding of 50 nM of NS1 from the different serotypes to immobilize purified IgG. The K_D, K_on and K_dis values were calculated using 1:1 model, global fitting analysis using Octet® Data Analysis software, and the mean and standard deviation of two independent experiments were shown.

Fig 3. NS1 capture ELISAs for the detection of NS1 from all dengue serotypes using different combinations of coating and detecting antibodies.

Wells were coated with (A) A2 and D6; (B) D8; or (C) Den3, and then purified recombinant NS1 from various viruses were added. The bound NS1 were detected by the addition of various HRP-conjugated antibodies. (D) Purified NS1 used for the sandwich ELISAs were coated directly on the wells, Den3 or PAb against ZIKV were added, and the bound antibodies were detected after the addition of HRP-conjugated anti-human or HRP-conjugated anti-rabbit. Representative results are shown and the values shown are OD± standard deviation from duplicates.

Fig 4. Competition indirect ELISA.

Wells were coated with DENV-2 NS1, blocked, and incubated with unlabeled A2, D6, or D8. The unlabeled antibodies were washed and then A2-HRP, D6-HRP or D8-HRP were added. The bound HRP-conjugated antibodies were detected after the addition of TMB substrate. Representative results from 2 experiments are shown and the values shown are OD± standard deviation from duplicates.

Fig 5. Detection of NS1 by NS1-capture ELISA that consisted of Den3 as capture antibody and D8-HRP as detecting antibody.

Assay was performed using (A) supernatant from virus infected cells, (B) human plasma spiked with purified NS1 from different dengue serotypes, and (C) human plasma from patients infected with dengue viruses from ≤7 days from fever onset or >7 days from fever onset. (A) The mean OD values ± standard deviation from duplicates are shown. (B) The dotted line represents the limit of detection (LOD) of the assay that was derived from 2-fold the mean OD values of the wells containing no NS1. (C) Individual dots represent samples from different patients and the dots on and above the dotted lines are considered positive for NS1.

Fig 6

(A) Alignment of NS1 amino acid sequences from DENV and ZIKV viruses. Blue boxes are potential epitopes recognized by D6 and red boxes are potential epitopes recognized by A2 and D8. (B) DENV-2 NS1 dimer structure (PDB: 4O6B) prepared using UCSF Chimera [28]. NS1 dimer with one subunit in grey and another subunit in black. Potential epitopes recognized by D6 are colored in blue and the potential epitopes recognized by A2 and D8 are colored in red.