Background
Dengue viruses (DENVs) are the most important vector-borne human viruses in the world, causing an estimated 50–100 million infections in 100 countries and resulting in a self-limiting febrile illness called dengue fever (DF), which is sometimes associated with haemorrhage [
1]. Approximately 500 000 cases result in the more severe, life-threatening forms, due to plasma leakage, severe haemorrhage, shock and organ failure called either severe dengue disease (SDD) or dengue haemorrhagic fever/dengue shock syndrome (DHF/DSS). Up to 12 500 people (2.5% of all DF cases) die from SDD (DHF/DSS) annually [
1]. Four DENV serotypes (DENV-1 to DENV-4) have been identified, each of which may cause SDD (DHF/DSS).
During DENV infections, high concentrations of the native homo-hexameric form of the nonstructural-1 (NS1) glycoprotein are secreted from infected mammalian cells along with infectious DENV virions [
2]. Purified secreted/extracellular (s/e) DENV-2 NS1 glycoprotein added to normal human sera could be detected by an enzyme-linked immunosorbent assay (ELISA) using a DENV NS1 glycoprotein complex-reactive monoclonal antibody, (MAb) 3D1.4 [
3‐
5], with a sensitivity of 15 ng/ml [
6]; although the DENV-2 NS1 glycoprotein-specific MAb, 1H7.4 [
3], was slightly more sensitive (4 ng/ml) [
6]. However, the ELISA using MAb 1H7.4 failed to detect the s/e NS1 glycoprotein in acute-phase sera from patients with primary DENV-2 infections collected in Thailand [
6]. This assay also only detected the s/e NS1 glycoprotein in less than 50% (DF: 40% and DHF/DSS: 45%) of acute-phase sera from patients with secondary DENV-2 infections, at concentrations from 70 ng/ml up to 15,000 ng/ml [
6]. Since these serum samples had undergone several freeze-thaw cycles, the results demonstrated the lability of the DENV s/e NS1 glycoprotein, probably through disruption of the relevant LD2 epitope (amino acid 25–33), [
4] or loss of the LD2 epitope through cleavage of the NS1 glycoprotein at the conserved (amino acid 100–GKRS–103) dibasic amino-terminal protease cleavage site [
7]. However, a capture ELISA using mouse and rabbit polyclonal antibodies (PAbs) detected the s/e NS1 glycoprotein in a moderate proportion of acute-phase sera from patients with either primary (77% (10/13): 40 to 2,000 ng/ml) or secondary (88% (14/16): 10 to 2,000 ng/ml) DENV-1 infections in French Guiana [
8], identifying the s/e NS1 glycoprotein as a suitable target for DENV diagnostics.
MAbs used in DENV NS1 glycoprotein detection assays should react equally with the NS1 glycoproteins of each DENV serotype. However, MAb 3D1.4 and other MAbs (MAbs 1A12.3, 4H3.4 and 3A5.4) that bound the DENV complex LX1 epitope (amino acid 112–KYSWKTWGKA–121) on the DENV NS1 glycoproteins reacted unequally with synthetic peptides containing the corresponding LX1 epitope of each DENV serotype, as follows: DENV-1 > DENV-3 > DENV-2 = DENV-4 [
4].
Subsequently available commercial DENV s/e NS1 glycoprotein detection ELISAs used MAb 3D1.4 or other likely LX1 epitope-specific MAbs (Pan-E, PanBio/Inverness, Brisbane, Queensland, Australia; Platelia, Bio-Rad Laboratories, Marnes La Coquette, France, or Standard Diagnostics Inc., Kyonggi-do, South Korea). These assays are too expensive ($US 5 to 10/sample) for routine screening of DENV-infected patients in poor, endemic areas, and have low sensitivity for the s/e NS1 glycoprotein of one or more DENV serotypes [
9‐
11]. The Pan-E, Platelia and Standard Diagnostics DENV s/e NS1 glycoprotein detection ELISAs show low to moderate overall sensitivities with acute phase DENV serum samples (Standard Diagnostics: 55%; Platelia: 56.5% [
11]; Pan-E: 56.3%; Platelia: 68.0% [
10]; Pan-E: 64.9%; Platelia: 83.2% [
9]). These assays also showed the highest sensitivity for DENV-1 (Standard Diagnostics: 70%; Platelia: 70% [
11]; Pan-E: 79%; Platelia: 87% [
10]; Pan-E 78.6%; Platelia: 93% [
9]) and the lowest sensitivity for either DENV-2 (Standard Diagnostics: 41%; Platelia: 38% [
11]; Pan-E: 62%; Platelia: 63% [
10]; Pan-E: 76%; Platelia 82% [
9]), or DENV-4 (Standard Diagnostics: 40%; Platelia 73% [
11]; Pan-E: 52%; Platelia: 79% [
10]; Pan-E: 36%; Platelia: 71% [
9]). These results accounted for a large multi-country evaluation giving the lowest sensitivities of Pan-E and Platelia s/e NS1 detection ELISAs with acute-phase sera from Nicaraguan patients (Pan-E: 30%; Platelia: 37%), due to nearly all being infected with DENV-2 (94%), and not with DENV-1 (6%) or DENV-3 (0%) [
10]. Thus, in summary, the strength of these assay reactions varied with DENV serotype, in the following order: DENV-1 > DENV-3 > DENV-2
> DENV-4, similar to the anti-LX1 epitope MAbs [
4].
More recent commercial DENV NS1 glycoprotein detection assays had even lower sensitivities. This was demonstrated in two well-conducted studies of NS1 antigen strip assays, Pan-E (average sensitivity: 45% (range 30–59%), PanBio (average sensitivity: 58.6%: range 48.2–68.4%), and Bio-Rad (average sensitivity: 58.6%: range 48.2–68.4%) and the Standard Diagnostics lateral flow s/e NS1 detection assay (mean sensitivity: 48.5%: range 38.5–58.7%) [
11,
12].
Commercially-available DENV NS1 glycoprotein detection/semi-quantification assays are too expensive for routine use in nearly all DENV endemic regions of the world, and all show significant differences in ability to detect the s/e NS1 glycoprotein of each DENV serotype [
9‐
11,
13‐
16]. Therefore there is an urgent need to develop improved MAbs for use in inexpensive DENV s/e NS1 glycoprotein detection assays.
We previously showed that the s/e NS1 glycoprotein from all four DENV serotypes could be detected in Indonesian patient sera using a simple dot-blot format, particularly when acid treatment and neutralisation steps were used [
17]. However, these assays used an electrochemiluminescence (ECL) substrate coupled with photographic film, which is too expensive for routine use in DENV endemic areas. One advantage was that the glycoproteins were stable for month-long periods when dot-blotted, blocked, and dried onto nitrocellulose membranes before being developed [Falconar, unpublished]. It is also likely that improved immunoperoxidase detection sensitivity could be obtained in Western blot or dot-blot assays by incorporating Ni
2+ ions into the 3,3’diaminobenzidine substrate, as demonstrated in immunohistochemical studies [
18].
In the present study, we aimed to determine: a) whether we could generate mouse MAbs that could more equally detect the s/e NS1 glycoprotein of each serotype without cross-reacting with yellow fever virus (YFV) s/e NS1 glycoprotein or human serum proteins; b) the optimal parameters for quenching endogenous human serum peroxidases in simple dot-blot assays using one of these MAbs and standard 4-chloro-1-naphthol and 3,3’-diaminobenzidine tetrahydrochloride (CND) substrate precipitation; c) the detection sensitivities of such a MAb against the purified s/e NS1 glycoprotein of each DENV NS1 glycoprotein in these dot-blot assays; d) whether such a MAb would show superior detection of the s/e NS1 glycoprotein of each DENV serotype compared with MAb 3D1.4, which bound the LX1 epitope; e) whether the detection sensitivity could be increased by incorporating 1 mM Ni2+ ions into the CND substrate step; f) whether the DENV s/e NS1 glycoproteins in human sera were stable when blotted onto the membranes, blocked, dried and stored at either ambient temperature (28°C) or at 4°C, -20°C or -80°C before being processed; g) whether these assays could be performed within a moderate time period (e.g. 3 hours); and h) the approximate cost of these assays (US$/sample).
Materials and Methods
Ethics statement
The collection of 5 ml blood samples from two clinically healthy human volunteers to obtain the normal human sera used in this study was approved by the University del Norte Ethics Committee. All animal experiments were approved by the Universidad del Norte Ethics Committee following the guidelines for the care and use of experiment animals established by the Colombian government, and following NIH guidelines.
Study site
Barranquilla is the principal seaport of Colombia and lies at the mouth of the Magdalena River on the Caribbean coast. All four DENV serotypes have become endemic in the city, with severe dengue disease (SDD: DHF/DSS) being confirmed since 1997. The isolation of all four DENV serotypes of DENV (DENV-1 (S#14 strain), DENV-2 (S#42 strain), DENV-3 (S#25 strain) and DENV-4 (S#10 strain) in insect (C6/36) cells was performed as described previously [
19,
20].
Purification of MAbs and preparation of immunoaffinity columns
The purification of mouse MAbs of the IgG1, IgG2a and IgG2b subclasses and the preparation of immunoaffinity columns were carried out as described previously [
3,
21,
22]. MAb 3A5.4 was used for these studies due to its ability to react with the s/e NS1 glycoproteins of each DENV serotype, as well as the s/e NS1 glycoprotein of YFV. Briefly, highly concentrated stocks of MAb 3A5.4 were diluted in phosphate buffered saline (PBS) (pH 7.6) and slowly passed through a protein G-Sepharose (P3296, Sigma-Aldrich, USA) column, washed with PBS, eluted using 0.1 M glycine/HCl (pH 2.5), and 0.9-ml fractions were immediately neutralised with 100 μl of 2 M Tris/HCl (pH 7.8). High-MAb-containing fractions were identified using the bicinchoninic acid (BCA) protein assay (BCA-1, Sigma-Aldrich, USA), pooled and repeatedly dialysed against 2 litre volumes of 0.2 M NaHCO
3/0.5 M NaCl (pH 8.9). Cyanogen-bromide-activated Sepharose 4B (0.8 g) (C9142, Sigma-Aldrich, USA) was pre-swollen and then washed with 500 ml of ice-cold HCl (1 mM). MAb 3A5.4 (12–15 mg) in 5 ml of the dialysate was then added, and the capped columns were incubated by inversion overnight at 4°C. The excess sites were then blocked with 1M ethanolamine/PBS/HCl (pH 9.0) for 2 h at ambient temperature. These columns were then subjected to nine cycles of high and low pH oscillations using 0.1 M glycine/HCl containing 1 M NaCl (pH 3.5) and 50 mM Tris/HCl containing 1 M NaCl (pH 9.0), before being washed with 10 mM diethylamine/PBS (pH 11.2) and then PBS containing 0.2% (wt/vol) NaN
3 before being stored at 4°C.
Growth of DENV strains of each DENV serotype and YFV and the immunoaffinity purification of their s/e NS1 glycoproteins
DENV strains of all four DENV serotypes were propagated in both insect (C6/36) cell and mammalian (Vero) cell monolayers as described previously [
5,
19]. For these studies, YFV (17D-204 vaccine strain) was also propagated in both C6/36 and Vero cells. High-titre seed stocks of each virus were prepared in 80 cm
2 cultures of 70% confluent C6/36 cells maintained in Leibovitz L-15 medium (L4386, Sigma-Aldrich, USA) containing 10% (vol/vol) tryptose phosphate broth (T9157, Sigma-Aldrich, USA) with 10% (vol/vol) foetal calf serum (FCS), pyruvate, L-glutamine and antibiotics (referred to hereafter as insect cell growth medium (ICGM)). Cell culture supernatants were collected on days 4 and 8 after infection, clarified by centrifugation and stored at -80°C. These stocks were used to infect 70%-confluent mammalian (Vero) cell monolayers maintained in medium 199 containing Na
2HCO
3, 3.5% FCS, pyruvate, L-glutamine and antibiotics (referred to hereafter as mammalian cell growth medium (MCGM) in 10 large (225 cm
2) flasks. After growth at 37°C in a 5% (vol/vol) CO
2 atmosphere for 4 days, the supernatants were collected and replaced by fresh MCGM and incubated for a further 4 days before being harvested. These supernatants were made up to 20 mM Tris/HCl (pH 7.4), 1 mM PMSF, 5 mM Na
2EDTA, 0.05% (wt/vol) NaN
3, 7% (wt/vol) polyethylene glycol 8000 (89510, Sigma-Aldrich, USA) and 0.4 M NaCl using stock solutions. The DENV particles (virions) were then allowed to aggregate overnight at 4°C before being removed by centrifugation at 8,000
× g for 30 min at 4°C. The clarified supernatants were then slowly (1 ml/min) passed through the MAb 3A5.4 or 3D1.4 immunoaffinity columns. After washing with loading buffer, the bound extracellular/secreted (s/e) NS1 glycoproteins were eluted in their native homo-hexameric form using 20 mM diethylamine in 10 mM Tris/HCl containing 150 mM NaCl, PMSF and EDTA and 0.4 ml fractions were immediately neutralised with 100 μl of 1M Tris/HCl (pH 7.2). Protein concentrations were determined in ELISA plates using 10 μl of sample in 200 μl of BCA reagent (BCA-1, Sigma-Aldrich, USA) with standard concentrations (16 mg/ml to 125 μg/ml serial dilutions) of bovine serum albumin (A7906: Sigma, USA) concentrations prepared in neutralised elution buffer. ELISA plates were incubated at 37°C for 60 min and then absorbance was determined at 570 nm (MRX, Dynax, USA), and protein concentrations were derived from the standard curves. Fractions containing the DENV s/e NS1 glycoproteins were concentrated by centrifugation dialysis at 1,000–2,000
× g (Centricon 10, Amersham, UK) against RPMI-1640 medium (R6504, Sigma, USA) containing a cocktail of protease inhibitors (P1860, Sigma-Aldrich, USA). Protein concentrations were then determined again and fractions were stored at -80°C.
One 25 cm2 flask of DENV-2 infected Vero cells was also used to prepare an infected cell lysate for the Western blot assays by discarding the supernatant and repeatedly washing the cells with RPMI medium before the addition of 2 ml of 32 mM orthophosphoric acid/58 mM Tris base (pH 6.7) (345245/T6066: Sigma-Aldrich, USA) containing 10% sodium dodecyl sulphate (SDS) (L3771, Sigma, USA) (cell-lysis buffer). After repeated passage through a 23-gauge needle to break the DNA, the cell lysate was centrifuged at 200 × g and aliquots of the supernatant were stored at -80°C.
Immunisation of mice and production of mouse MAbs
The immunisation and use of halothane in oxygen to anaesthetise BALB/c mice, and the production and cloning of MAbs were carried out as described previously [
5,
22,
38]. Briefly, a group of three 6-week-old female BALB/c mice (Universidad Nacional, Bogota, Colombia) were immunised by the combined subcutaneous (s.c.; 0.1 ml) and intra-peritoneal (i.p.; 0.4 ml) routes with a mixture of 5 μg of the purified e/s NS1 glycoprotein of each DENV serotype (i.e., 20 μg/mouse) emulsified in complete Freund’s adjuvant (F5881, Sigma, USA). These mice were also each injected with 0.5 μg of mouse interleukin-12 (IL-12) by the i.p. route on days 1, 3, 5 and 7 after the immunisations to increase the production of PAbs of the IgG2a subclass [
39]. Three weeks later, the mice were boosted with the sample antigen emulsified in incomplete Freund’s adjuvant (F5506: Sigma, USA) at the same dose and via the same route, and they also received the same multiple IL-12 injections. Four weeks later, the mice were anaesthetised and blood samples were obtained from the retro-orbital sinus. The highest responding mouse, as determined by the indirect ELISA titres against the s/e NS1 glycoprotein of each DENV serotype, was then immunised with 10 μg each of the DENV-1 (S#14 strain), DENV-2 (S#42 strain), DENV-3 (S#25 strain) and DENV-4 (S#10 strain) s/e NS1 glycoproteins (40 μg in total) in RPMI medium by the combined intravenous (i.v.) route in the tail vein (100 μl) and the i.p. route (300 μl). Three days later, this mouse was anesthetised and humanely killed by cervical dislocation, immersed in 70% ethanol and transferred to a class II laminar flow cabinet. The spleen cells were aseptically removed in RPMI medium containing 10% MAb-cloning tested quality FCS (FCS Premium: Biomeda, USA) and then centrifuged at 200
× g for 10 min at 28°C. SP2/0 Ag14 plasmacytoma cells (ATCC, USA), grown in RPMI medium containing 20% FCS and pyruvate, L-glutamine and antibiotics (PGAB), were harvested. The spleen and plasmacytoma cells were counted and mixed at a 10:1 ratio, co-centrifuged at 250
× g and the supernatant was discarded. One millilitre of 45% (wt/vol) polyethylene glycol 4000 (Merck BDH, UK) containing 5% dimethylsulfoxide (D2650, Sigma, USA)/PBS (pH 8.1) was slowly added to the loosened cell pellet over a period of 1 min with constant agitation before being slowly diluted with 20 ml of Hank’s balanced salt solution (H9394, Sigma, USA). After incubation at 28°C for 30 min, the cells were pelleted by centrifugation at 200
× g, the supernatant was discarded, and the cells re-suspended in 200 ml of RPMI medium containing 20% FBS, 1 × HAT (H0262: Sigma, USA) and PGAB before being added to 96-well culture plates (NUNC, USA) at 200 μl/well. After incubation at 37°C in 5% CO
2/air for 5 days, 100 μl of the medium was changed and thereafter 150 μl was changed every 3 days until the DENV s/e NS1 glycoprotein MAb-positive wells were expanded. The 150 μl volumes collected from each well were tested by ELISA for reactions against the purified DENV s/e NS1 glycoprotein of each DENV serotype, and the contents of the wells containing MAbs that cross-reacted with the s/e NS1 glycoprotein of multiple DENV serotypes were transferred into 48-well plates (Costar, USA). HAT selection medium was subsequently substituted with medium containing HT (H0137, Sigma, USA) after 14 days, from which these hybridomas were never weaned. The culture supernatants were then tested for reactivity with the s/e NS1 glycoprotein from each DENV serotype in dot-blot assays. Western blot assays were also performed to test reactivity against a) the NS1 glycoproteins from each DENV serotype and YFV, b) the 2–mercaptoethanol reduced DENV-2 s/e NS1 glycoprotein, the cell-associated DENV-2 NS1 glycoprotein present in a crude SDS-lysate of DENV-2 infected Vero cells, and c) the DENV-2 s/e NS1 glycoprotein present in a human serum sample after the sera had been treated with 3% H
2O
2/PBS for 30 min to quench endogenous peroxidases. Thus, these assays were used to select wells containing hybridomas that secreted MAbs that specifically reacted with the s/e NS1 glycoprotein of each DENV serotype, and which would not cross-react with proteins in Vero cells or human sera.
Hybridomas that secreted the desired MAbs were cloned twice by limiting dilution in 96-well cell culture plates containing 2–4 × 104/well CD1 (Swiss out-bred) mouse peritoneal macrophage feeder layers. Single wells were identified using an inverted microscopy, the supernatants were screened by ELISA against DENV s/e NS1 glycoproteins and the resultant cloned hybridomas were further sequentially expanded and harvested, pelleted by centrifugation and the cells slowly (-1°C/min) frozen to -80°C after being re-suspended in ice-cold 10% (vol/vol) DMSO/FCS, before being transferred to liquid nitrogen. From this fusion, MAb 2C4.6 and a number of other cloned hybridomas were selected.
The IgG subclass of the MAbs was determined by a kit using goat PAbs specific for each mouse IgG subclass in radial immunoassays (ISO-2, Sigma, USA), according the manufacturer’s instructions.
Immunoassays
The indirect DENV s/e NS1 glycoprotein ELISA was performed as described previously [
5,
21,
23]. Briefly, the immunoaffinity-purified DENV-2 s/e NS1 glycoproteins (see above) or the negative-control antigen, bovine serum albumin, were diluted to 3 μg/ml in carbonate/bicarbonate buffer (pH 9.8), loaded into 96-well ELISA plates (Immulon 2, Dynatech, USA) and adsorbed overnight at 4°C. After washing with PBS containing 0.05% (vol/vol) Tween20 (PBS/T), blocking with 1% (wt/vol) gelatin/PBS for 2 h at 25°C and washing again with PBS/T, 50 μl volumes of the undiluted hybridoma wells were transferred to each ELISA well and allowed to react at 37°C for 60 min. After washing with PBS/T, the bound MAbs were detected with sequential reaction steps using peroxidase-labelled goat anti-mouse IgG heavy and light (H&L) chain reactive (115-035-166, Jackson ImmunoResearch, USA) in PBS/T/G, washing with PBS/T, incubation with
o-phenylenediamine (opd) in citrate/phosphate buffer (pH 5.0) containing H
2O
2 and the absorbance values measured at 490 nm (MRX: Dynex, USA).
The Western blot assays were performed as described previously [
3,
21,
22]. Samples were prepared using 1.0 μg s/e NS1 glycoprotein of each DENV serotype added to 160 μl of stacking buffer containing 0.5% (wt/vol) SDS, 32 μl of 0.25% (wt/vol) bromophenol blue in 50% glycerol/H
2O, with (all DENV serotypes) or without (DENV-2 s/e NS1 glycoprotein) 1% (vol/vol) 2-mercaptoethanol (M3148: Sigma, USA). In addition, 160 μl of a DENV-2 infected Vero cell lysate (see cell culture) and 30 μl of a normal human serum sample (which had been diluted 1 in 4 with PBS) were both treated with 3% (vol/vol) H
2O
2 (H1009, Sigma, USA) for 60 min at 28°C to quench endogenous human serum peroxidases, before the addition of the 5 × stacking buffer and 1.0 μg of DENV-2 s/e NS1 glycoprotein (total volume 160 μl). Both were then treated with 32 μl of 0.05% (wt/vol) bromophenol blue in 50% glycerol/H
20. These samples were then heated at 100°C for 3 min. After cooling and centrifugation, they were then loaded onto 8–9% (wt/vol) preparative SDS/PAGE gels (Miniprotean II: Bio-Rad, UK) and subjected to electrophoresis (15–20 constant mA/gel) and subsequent electro-blotting at 160 mA/gel (Sartoblot II, Sartorius, UK) onto 0.2-μm pore-sized nitrocellulose membranes (BioTraceNT, PAL LifeSciences, USA) before being dried. These membranes were then blocked using 3% (wt/vol) skimmed milk powder (Marvel, Cadbury, UK)/PBS (PBS/T/M) containing 0.02% (wt/vol) NaN
3, washed with PBS/T, dried and then cut into 3-mm wide strips. Preparative blot strips of the reduced (DENV-2 s/e NS1 glycoprotein only) and non-reduced (s/e NS1 glycoproteins of all DENV serotypes and YFV, DENV-2 infected Vero cells and human sera which contained the DENV-2 s/e NS1 glycoprotein) were taped together with waterproof autoclave tape and allowed to react with 0.5 ml of the MAb supernatants of two 500 μl samples (1 ml total) from 48-well cultures diluted with an equal volume of PBS/T containing 2% (wt/vol) skimmed milk powder for 2 h at ambient temperature. After washing with PBS/T, the bound MAbs were detected using sequential reaction steps with peroxidase-labelled goat anti-mouse IgG (H&L) (115-035-166: Jackson ImmunoResearch, USA) in PBS/T/G, washing with PBS/T, PBS and then reacting the membranes with 4-chloro-1-naphthol/3,3’diaminobenzidine (C8890/D8001, Sigma, USA) (CND) substrate in PBS containing 0.006% (vol/vol) H
2O
2.
For the dot-blot assays, 0.5-cm squares were marked on the nitrocellulose membranes using an 8B graphite pencil (Staedtler, Germany). Duplicate, serial, two-fold or three-fold dilutions of the immunoaffinity-purified s/e NS1 glycoproteins prepared in normal human sera collected from anti-DENV IgM and IgG-negative patients [
19] were added at 10 μl/dot and were air dried. Different methods were assessed to quench the endogenous peroxidases in the human sera present on these dot-blots. For this study, solutions of 3% H
2O
2 diluted from a 30% (vol/vol) stock solution (H1009, Sigma Aldrich, USA) in 40% (vol/vol) MeOH/PBS, PBS or H
2O were tested. These dot-blot assays were then processed as described for the Western blot assays to identify the detection sensitivities of MAb 2C4.6 against the s/e NS1 glycoprotein of each DENV serotype when reacted at a concentration of 5–10 μg/ml with serial dilutions of each of them. The sensitivities of MAb 2C4.6 and 3D1.4 against known concentrations of the s/e NS1 glycoprotein of each DENV serotype were then compared. The reactions of other MAbs anti-NS1 MAbs (1G5.4-A1-C3 and 1B6.3) of the IgG2b subclass, which defined a DENV-2 > DENV-4 > DENV-1 > DENV-3 and DENV-2-specific responses against the s/e NS1 glycoprotein of each DENV serotype in Western blot assays respectively [
3,
21,
22], were also tested for their reactions in these dot-blot assays.
In addition, we assessed whether: a) 1 mM Ni2+ (NiCl2) could increase the sensitivities of these assays when added to the 4-chloro-1-naphthol/3,3’ diaminobenzidine (CND) substrate mixture; b) the assays could be successfully processed without loss of detection sensitivities after sample addition and blocking, washing with PBS containing 0.01% NaN3 and storing after drying for 1 month at 28°C (ambient temperature), 4°C, -20°C and -80°C; and c) the total assay time could be reduced to 3 h without significant loss of detection sensitivity when they were performed at ambient temperature.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
The authors both contributed to the design of the experiments, the laboratory work and preparation of the article. Both authors read and approved the final manuscript.