Fig 1.
General overview of subtractive biopanning of a phage library displaying antibody fragments.
Recombinant antigen from all four different DENV serotypes was adsorbed on polystyrene immunotubes. The phage library was exposed to each tube in turn. The first three tubes were used for subtractive purposes to deplete binders against homologous regions amongst the rNS1. Only binders in the last tube against the target antigen were eluted for further characterisation.
Fig 2.
Reactivity of unique isolated phage clones and corresponding reformatted whole mAbs against rNS1 and native NS1 from all four DENV serotypes.
(a) Monoclonal phage displaying antibody fragments from different libraries (either scFv or Fab) were added to immobilised recombinant NS1 (rNS1) and detected with a HRP-conjugated antibody against M13 filamentous phage. (b) Reformatted whole IgG were added to immobilised rNS1 and detected with a HRP-conjugated anti-human IgG antibody. (c) Flavivirus specific antibodies were immobilised and used to capture native NS1 from supernatants of Vero cells infected with DENV serotypes, Zika virus, or non-infected. Captured NS1 was detected using the serotype-specific reformatted mAbs or a flavivirus specific mAb.
Fig 3.
Immunofluorescence assay micrographs showing binding of serotype-specific anti DENV NS1 antibodies.
Vero cells were infected with DENV-1 (a), DENV-2 (b), DENV-3 (c) DENV-4 (d), mock infection (e) or Kunjin virus (f). Viral infection is confirmed using an antibody to flavivirus envelope protein (4G2) stained with Anti-mouse IgG AlexaFluor546 (red), and serotype specific antibody binding (9H2, 4C11, 7G11 and 6A5) was detected with Anti-human IgG AlexaFluor488 (green). All wells also had a nuclear stain (Hoescht) (Blue) added to determine the relative localisation of binding.
Table 1.
Kinetics and affinities of antibody-antigen interactions.
Kinetic data measured using SPR indicating association rate constants (ka) [M-1 s-1], dissociation rate constants (kd) [s-1] ± standard error as well as the equilibrium dissociation constants (KD) kd/ka [M]. X denotes that no measurable interaction was detected by the assay.
Fig 4.
Investigation of serotype-specific mAb pairings with the DENV cross-reactive murine monoclonal Gus2.
(a) Serotype-specific mAbs 9H2, 4C11, 7G11 and 6A5, and pan-reactive mAb Gus2, were tested for binding to an immobilised peptide consisting of DENV-1 NS1 amino acid residues 111–125, an unrelated peptide or no peptide. (b) Gus2 antibody was immobilised onto polystyrene plates and used to capture DENV rNS1. HRP-conjugated, serotype-specific antibodies (9H2, 4C11, 7G11 and 6A5) were then used for detection, showing the ability of these mAbs to sandwich with Gus2 for serotype-specific detection of NS1.
Fig 5.
Investigation of effect of antibody orientation on sandwich ELISA sensitivity.
Immobilised antibodies against DENV NS1 were used to capture dilutions of DENV rNS1. The serotype-specific antibodies were used either as the detecting antibody paired with pan-reactive Gus2 as the capture antibody (a), or as the capture antibodies with detection achieved used the detection antibody from the Panbio® Dengue Early ELISA Kit (b). The results were compared with the complete Panbio® Dengue Early ELISA Kit assay used as per the manufacturer's instructions (c). Background-subtracted data is shown indicating the mean ± SD.
Fig 6.
Proof of principle DENV serotyping assay in human serum using a multiplexed fluorescent microsphere based capture assay.
Microplate wells were coated with Gus2 antibody, which was used to capture dilutions of NS1, separately for each DENV serotype, in human serum. Differentially fluorescent microspheres, conjugated with serotype-specific mAbs, were used to detect the capture of NS1. All four fluorescently labelled serotype-specific microspheres were added in each well. After washing, bound microspheres were detected using excitation and emission filters corresponding to each of the four microspheres (360nm/415nm for 9H2 Anti-DENV-1 (a), 475nm/548nm for 4C11 Anti-DENV-2 (b), 525nm/580nm for 7G11 Anti-DENV-3 (c), 615nm/735nm for 6A5 Anti-DENV-4 (d)).