Development of a double-monoclonal antibody sandwich ELISA: Tool for chicken interferon-γ detection ex vivo

Abstract

The aim of the present work was to develop reagents to set up a chicken interferon-γ (ChIFN-γ) assay. Four monoclonal antibodies (mAbs) specific for ChIFN-γ were generated to establish sandwich ELISA based on 2 different mAbs. To improve the detection sensitivity of ChIFN-γ, a double-monoclonal antibody sandwich ELISA was developed using mAb 3E5 as capture antibody and biotinylated mAb 3E3 as a detection reagent. The results revealed that this ELISA has high sensitivity, allowing for the detection of 125 to 500 pg/mL of recombinant ChIFN-γ, and also has an excellent capacity for detecting native ChIFN-γ. This ELISA was then used to detect ChIFN-γ level in chickens immunized with a Newcastle disease virus (NDV) vaccine, the immunized chicken splenocytes were stimulated by NDV F protein as recall antigen. From our results, it appears that the sensitivity range of this sandwich ELISA test is adequate to measure the ex vivo release of ChIFN-γ.

Introduction

Interferon-γ (IFN-γ; also called type II interferon), a cytokine produced predominantly by T-helper type 1 (TH1) cells and Natural Killer cells in response to antigenic or mitogenic stimulation (1,2), plays a critical role in initiating and regulating cell mediated immunity, which is a central player in initiating the TH1 response against intracellular pathogens (3,4).

Chicken provides an important animal model of a number of intracellular infections. Like its mammalian counterpart, chicken IFN-γ (ChIFN-γ) strongly upregulates the expression of class II major histocompatibility complex (MHC) proteins (5–7) so that antimicrobial and antiviral activities of chickens are improved (5,7–10). The ChIFN-γ also enhances immunity against tumors and parasites (11–15). Previous studies showed that the level of IFN-γ following antigenic/mitogenic stimulation allows for accurate evaluation of cell-mediated immunity (CMI) (16,17). Unfortunately, methods of detecting ChIFN-γ are limited. So far, ChIFN-γ is commonly detected based on its ability to inhibit viral replication in vitro or activate the HD11 macrophages. Both of these assays are labor-intensive, time-consuming, and nonspecific methods that exhibit low sensitivity and are difficult to standardize. Although real-time PCR (RT-PCR) or Northern blot can detect very low levels of ChIFN-γ, these methods can be used to analyze ChIFN-γ only at the mRNA, but not the protein level.

Therefore, a qualitative and quantitative assay to accurately and efficiently determine ChIFN-γ levels in biological samples is extremely urgent, especially to study response to infections induced by intracellular bacteria, parasites, and viruses. Until now, there were 2 kinds of assays reported that could successfully evaluate ChIFN-γ in the protein level (18,19). One is a monoclonal antibody (mAb)-based direct binding enzyme-linked immunosorbent assay (ELISA), the other is a quantitative ELISA based on the combination of a rabbit anti-ChIFN-γ serum with a mAb. Both of them could measure ChIFN-γ in a variety of formats and the latter is more sensitive than the former. But these assays are still limited and need to be improved in detecting trace amounts of ChIFN-γ. To address this problem, this study was designed to develop a ChIFN-γ-specific ELISA. We have used recombinant ChIFN-γ, which was generated before (20) to develop mAbs against ChIFN-γ. Using these antibodies we have developed a capture ELISA system for the detection of both recombinant and native ChIFN-γ.

Materials and methods

Chickens

Four-week-old, white Laihang, specific pathogen free (SPF) chickens and 8-week-old BALB/C mice used in this study were provided by the Comparative Medical Center of Yangzhou University (Yangzhou, China), the animals were housed and handled at the Animal Biosafety Facilities and all procedures were approved by the Institutional Animal Experimental Committee.

Vaccines, plasmids, and recombinant proteins

The Newcastle disease mild, living (La Sota strain) vaccine (100992007, Wuhan Chopper Biology, Wuhan, Hubei, China) was used to vaccinate chickens. Chickens were immunized via the oculo-nasal route according to the manufacturer’s instructions. Recombinant plasmid pVAX1-ChIFN-γ was provided by Jiangsu Key Lab of Zoonosis (Jiangsu, China). Newcastle disease virus recall antigen (recombinant protein of NDV F protein) (21), bovine IFN-γ (BovIFN-γ), cervine IFN-γ (CerIFN-γ), chicken IFN-α (ChIFN-α), and chicken interleukin 4 (ChIL-4), all provided by Jiangsu Key Lab of Zoonosis.

Production of recombinant ChIFN-γ proteins and native ChIFN-γ

The production and purification of Escherichia coli-derived recombinants, histidine (His)-tagged ChIFN-γ (His-ChIFN-γ, 330 μg/mL), glutathione S-transferase(GST)-tagged ChIFN-γ (GST-ChIFN-γ, 1500 μg/mL) and the baculovirus-derived recombinant Bac-ChIFN-γ (the supernatant of recombinant virus infected Sf9) were described previously (20,21).

Natural ChIFN-γ was produced by concanavalin A (Con) A-stimulation of spleen cells from SPF chickens. Splenocyte suspensions were prepared as described (22) and adjusted to 10⁷ cells/mL in the growth medium RPMI 1640 (Invitrogen, Thermo Fisher Scientific, Waltham, Massachusetts, USA) containing 10% inactivated fetal bovine serum (Hyclone; Thermo Fisher Scientific), 100 U penicillin/mL, 100 μg streptomycin/mL. Then, 2.5 × 10⁶ cells (250 μL) per well were transferred into flat-bottomed 24-well plates. Equal volumes (250 μL) of medium containing 6, 12, or 24 μg/mL of final concentration Con A were added in triplicate, and cultures were incubated for 4 d. Negative controls received 250 μL RPMI 1640 medium only. After 4 d of incubation at 41°C, 5% CO₂, supernatant was harvested from each well for the measurement of ChIFN-γ production.

Production of monoclonal antibodies

The 8-week-old BALB/c mice were immunized at an interval of 2 wk by subcutaneous injections with 100 μg of recombinant His-ChIFN-γ emulsified in Freund’s adjuvant (Sigma-Aldrich, St. Louis, Missouri, USA), and boosted intravenously with 100 μg of His-ChIFN-γ without adjuvant 3 d prior to fusion. Hybridomas were tested for the presence of antibodies against GST-ChIFN-γ by using an ELISA (see below). Positive cells were cloned 3 times by limiting dilution. The immunoglobulin (Ig) sub-class of mAbs was determined using a mouse mAb isotyping kit (Sigma-Aldrich,) according to the manufacturer’s instructions. BALB/c mice were injected intraperitoneally using positive hybridoma cell line secreting anti-ChIFN-γ mAb. Ascites fluids containing abundant anti-ChIFN-γ mAb were produced by these immunized mice and purified by protein A chromatography (GE Healthcare, Wauwatosa, Wisconsin, USA). Purified mAbs were biotinylated using standard methods (GenScript Biotechnique Company, Nanjing, Jiangsu, China).

The ChIFN-γ ELISA for screening of ChIFN-γ antibodies

The 96-well plates (XiaMen YunPeng Biotechnique Company, Xiamen, Fujian, China) were coated with purified GST-ChIFN-γ diluted at 1.5 μg/mL in 0.05 M/L carbonate-bicarbonate buffer (pH 9.6) for 24 h at 4°C, 50 μL each well. Subsequently, the plates were washed 3 times using phosphate-buffered saline (PBS) containing 0.05% Tween-20 (PBS-T). The plates were then blocked with PBS-T containing 10% (v/v) newborn bovine serum (Lanzhou National Hyclone Bio-engineering Corp., Lanzhou, Gansu, China) for 4 h at 37°C. After washing, 50 μL of hybridoma supernatants were added for 1 h at 37°C and then washed as above. Binding of antibodies was revealed using horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (Sigma-Aldrich), incubated for 30 min at 37°C. The HRP activity was revealed by addition of o-phenylenediamine (OPD) for 15 min at 37°C. The reaction was stopped by addition of 2 M H₂SO₄, 490 nm wavelength was used to detect the color development due to HRP reacting with OPD, OD₄₉₀ value indicates the presence of mAbs against ChIFN-γ.

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and western-blotting

Proteins were separated by SDS-PAGE on a 12% polyacrylamide gel (pH 8.8) with a 5% stacking gel (pH 6.8), in Tris-glycine running buffer (pH 8.7), and transferred onto Polyvinylidene Fluoride (PVDF) membrane. The membranes were blocked with 5% milk protein in PBST, then incubated with anti-ChIFN-γ mAbs in blocking buffer for 1 h. After washing, the membranes were incubated with HRP-conjugated goat anti-mouse IgG (Sigma-Aldrich) in blocking buffer for 0.5 h. Washing 5 times, the membranes were developed using western blotting detection reagent [PBS containing 25% 3, 3′-diaminobenzidine (DAB) and 0.03% H₂O₂]. This DAB substrate solution deposits a brown specific stain in the presence of HRP, then the reaction was quickly stopped by distilled water. After air drying the chromogenic membrane was scanned by the scanner (2580, Epson China Co., Beijing, China).

(COS)-7 cells, Cercopithecus aethiops kidney cells transformed by SV40, were seeded in 24-well-plates and cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; Invitrogen, Thermo Fisher Scientific) containing 10% fetal bovine serum (Hyclone) and incubated overnight to 50% cell confluent at 37°C, 5% CO₂. Recombinant plasmid pVAX1-ChIFN-γ and control plasmid pVAX1 were transduced into COS-7 cells, respectively, using lipofectin reagent (Invitrogen; Thermo Fisher Scientific) following the manufacturer’s instruction. After an additional 48 h incubation at 37°C, 5% CO₂, the supernatant of each well was discarded, plates were washed 3 times using ice cold PBS (pH 7.2), and directly fixed by ice cold methanol for 5 min at room temperature. Then methanol was removed and allow to air dry. After PBS washing, an optimal concentration of mAb was added and incubated for 1 h at room temperature in a shaker, 50 times/min (TS-2, Kylin-Bell Lab Instrument Co., Haimen, Jiangsu, China). After washing, the second antibody fluoresceine isothiocyanate (FITC)-conjugated goat anti-mouse IgG (H + L) (R&D systems, Minneapolis, Minnesota, USA) was added into each well for 30 min incubation at room temperature in dark. After the washing steps, fluorescence of each well was observed and photos were taken by fluorescence microscope (TS 100, Nikon, Japan).

Antigen-capture ELISA

To determine the ability of mAbs to capture ChIFN-γ in various biological test samples, 96-well ELISA plates were precoated with serial dilutions of purified mAbs in 0.01 M PBS at 4°C for 24 h, 100 μL each well. Precoated wells were washed 3 times in PBS-T, then blocked with 2% BSA (in PBS-T, Sangon Bio-engineering Corporation, Shanghai, China) overnight at 4°C. After blocking, 100 μL of each test sample diluted in 1% BSA (in PBS) was added to individual wells and incubated for 2 h at 37°C, followed by a washing step, 100 μL of an optimized dilution of the biotin-labeled mAbs were added and incubation was continued for 1 h at 37°C. The plates were washed 5 times in PBS-T, then incubated with 100 μL of avidin-HRP (diluted in 1: 2500 in PBS-T containing 1% BSA, 20% inactivated newborn bovine serum) for 30 min at 37°C, then exposed to 3,3′,5,5′-tetramethylbenzidine (TMB) peroxidase substrate solution (eBioscience, San Diego, California, USA). The reaction was stopped by addition of 2 M H₂SO₄, OD₄₅₀ of each sample was measured.

Establishment of a double-mAb sandwich ELISA for the detection of ChIFN-γ

In a checker-board analysis, all 4 mAbs were used as capturing or revealing antibodies, and His-ChIFN-γ was used to identify compatible mAb sandwich partners as well as to determine the optimal concentration of coating antibody. The specificity of this sandwich ELISA was verified using Bac-ChIFN-γ and native ChIFN-γ secreted from splenocytes activated by Con A.

The ChIFN-γ stimulated by specific NDV F protein recall antigen

Four-week-old SPF chickens were vaccinated with the Newcastle disease mild vaccine. Three weeks after vaccination, splenocytes from each chicken were isolated for stimulation ex vivo. The NDV F protein was used as a recall antigen to test the response of splenocytes as measured by ChIFN-γ production in the supernatant. Spleen cell suspensions were prepared as described previously. Splenocytes were adjusted to 10⁷ cells/mL in RPMI 1640 and 100 μL cells per well were transferred into flat-bottomed 96-well plates. Equal volumes of medium containing NDV F protein recall antigen (2, 10, 20, 40 μg/mL) were added in triplicate and cultures were incubated for 4 d at 37°C, 5% CO₂. Negative controls received 100 μL RPMI 1640 medium only. After 24 h, 48 h, 72 h, and 96 h of incubation, 100 μL cell supernatant was removed from each well and assayed for ChIFN-γ production.

Results

Generation and characterization of mAbs to ChIFN-γ

From 2 independent fusions, 4 hybridomas (1G10, 2C3, 3E3, 3E5) were cloned and selected for further study based upon their strong ChIFN-γ ELISA reactivity. The IgG isotypes of 1G10, 2C3, 3E3, and 3E5 were IgG1, IgG1, IgG2a, and IgG1, respectively. The specificity of the 4 mAbs to ChIFN-γ was confirmed by ELISA, Western blot, and immunofluorescence analysis. All the results showed that 4 mAbs provided positive identification of recombinant ChIFN-γ without cross-reaction with other control proteins (Table I, Figures 1 and 2).

Table I.

Characterization of monoclonal antibodies (mAbs) specificity to chicken interferon-γ (ChIFN-γ)

	Reactivity of chicken cytokines OD490 nma
mAb	ChIFN-γ	BovIFN-γ	CerIFN-γ	ChIFN-α	ChIL-4
1G10	2.049 ± 0.093	0.071 ± 0.017	0.090 ± 0.004	0.067 ± 0.001	0.090 ± 0.006
2C3	2.097 ± 0.092	0.076 ± 0.002	0.085 ± 0.005	0.080 ± 0.005	0.080 ± 0.007
3E3	2.352 ± 0.062	0.079 ± 0.001	0.097 ± 0.003	0.063 ± 0.039	0.070 ± 0.002
3E5	1.845 ± 0.075	0.085 ± 0.002	0.087 ± 0.003	0.065 ± 0.001	0.072 ± 0.005

^aRecombinant ChIFN-γ, BovIFN-γ, CerIFN-γ, ChIFN-α, and ChIL-4 obtained by E. coli were used as coating antigen (1.5 μg/mL) in the ELISA. The indicated mAbs (1 mg/mL) were incubated for binding to cytokine-coated wells and their binding was detected by reacting with HRP-labeled goat anti-mouse IgG.

Values are means ± SD.

Figure 1.

Western blotting analysis of anti-chicken interferon-γ (ChIFN-γ) monoclonal antibody. Lane 1 and Lane 3 — control proteins from isopropyl-beta-pyranoside-D-1-thio-Galactose (IPTG) induced recombinant bacteria BL21-(pGEX-6P-1) and BL21(DE)3-(pET) were stained with anti-ChIFN-γ mAb 3E3; Lanes 2 and 4: purified GST-ChIFN-γ and His-ChIFN-γ, derived from E. coli, were stained with anti-ChIFN-γ mAb 3E3.

Figure 2.

Immunofluorescent staining analysis of anti-chicken interferon-γ (ChIFN-γ) monoclonal antibody (mAb). Recombinant plasmid pVAX1-ChIFN-γ was transduced into COS-7 cells, after fixing step, cells were stained with anti-ChIFN-γ mAb 3E3.

The establishment of a double-mAb sandwich ELISA for recombinant ChIFN-γ

In order to detect native ChIFN-γ, an antigen-capture ELISA system was set up using purified and biotinylated mAbs as capture and detection antibodies, respectively. Checker-board analysis revealed that even very low levels of recombinant ChIFN-γ could be detected when the biotin-labeled 3E3 was used as the detection antibody in combination with any of the other 4 mAbs used as capturing antibody (Figure 3). A double-mAb sandwich ELISA allowing the detection of 125–500 pg/mL of His-ChIFN-γ with low background was achieved using mAb 3E5 as the capture antibody (85 μg/mL) and biotinylated mAb 3E3 as the detection antibody (0.67 μg/mL) (Figures 4a, 4b, 5b).

Figure 3.

Antibody combinations of different monoclonal antibodies (mAbs). Biotinylated 3E3 as detecting antibody and any of the 4 mAbs as capture reagent to establish the double-mAb sandwich ELISA.

Figure 4.

Establishment of double-monoclonal antibody sandwich ELISA. A — Optimal coating concentration of mAb 3E5. Using mAb 3E5 as capture antibody at the concentration range from 5.4 to 85 μg/mL, and biotinylated mAb 3E3 as detecting antibody, when the coating concentration of 3E5 reached 85 μg/mL, it is the most sensitive combination for the detection of limiting amounts of recombinant chicken interferon-γ (ChIFN-γ). B — Sensitivity of double-monoclonal antibody sandwich ELISA. Detection limit of His-ChIFN-γ (64 ng/mL) was tested in this double-monoclonal antibody sandwich ELISA when 3E5 coated as 85 μg/mL, biotinylated mAb 3E3 used as 0.67 μg/mL.

Figure 5.

Specificity of chicken interferon-γ (ChIFN-γ) sandwich ELISA. A — Serial 2-fold dilutions of His-ChIFN-γ (20 ng/mL) or Bac-ChIFN-γ (the supernatant of recombinant virus infected Sf9) and native ChIFN-γ produced by Con A-activated chicken splenocytes were tested in the sandwich ELISA. B — Serial 2-fold dilutions of His-ChIFN-γ, His-ChIL-4, and His-ChIFN-γ plus His-ChIL-4 were tested in the sandwich ELISA. In this assay, mAb 3E5 was used as coating antigen and the detection reagent was biotinylated 3E3. There is no significant difference between His-ChIFN-γ, His-ChIL-4, plus His-ChIFN-γ groups (P = 0.1884).

Sandwich ELISA for detection of native ChIFN-γ

After establishment of the double-mAb sandwich ELISA shown in Figure 3, the specificity of this ELISA was evaluated by detecting recombinant ChIFN-γ (Bac-ChIFN-γ, His-ChIFN-γ) as well as native ChIFN-γ which is in the supernatant secreted from Con A activated chicken splenocytes. The ELISA could detect not only the recombinant ChIFN-γ but also the native ChIFN-γ (Figure 5a), and is not affected by recombinant ChIL-4 protein (Figure 5b). When the concentration of Con A in the cultured medium was increased from 0 to 12 μg/mL, the level of ChIFN-γ detected also increased. Concentrations of Con A greater than 12 μg/mL did not further increase the level of ChIFN-γ detected (Figure 6). Furthermore, extending the time over which splenocytes were incubated with Con A also led to an increase in the levels of ChIFN-γ detected, although the degree of increase was less pronounced after 48 h. In addition, we found that increasing levels of ChIFN-γ were detected when the number of splenocytes was increased from 1 × 10⁶ cell/mL to 5 × 10⁶ cell/mL (Figure 7). These results indicate that this double-mAb sandwich ELISA is suitable to measure ChIFN-γ in a variety of settings.

Figure 6.

Detection of mitogenic response of chicken splenocytes stimulated with different concentration of Con A at 0, 6, 12, and 24 μg/mL using the sandwich ELISA (P < 0.0001).

Figure 7.

Mitogenic response of different numbers of chicken splenocytes stimulated with Con A (12 μg/mL) using the sandwich ELISA (P < 0.0001).

Sandwich ELISA for detection of native ChIFN-γ stimulated by NDV recall antigen

Splenocytes from 4-week-old NDV immunized chickens were examined 3 wk after live vaccination for their ability to produce ChIFN-γ upon ex vivo stimulation. We found that splenocytes from vaccinated chickens produced ChIFN-γ after NDV-specific antigen recall stimulation with NDV F proteins. The ELISA could detect the highest production of ChIFN-γ after 96 h of exposure to 20–40 μg/mL of NDV F proteins (Figure 8). There was no significant difference in ChIFN-γ production from the antigen-stimulated cultures of unimmunized birds (data not show). These results suggest that this capture ELISA is able to detect native ChIFN-γ released ex vivo.

Figure 8.

Interferon-γ (IFN-γ) responses induced by Newcastle disease virus (NDV) La Sota vaccine. Optimal concentration of recall antigen (at 40 μg/mL) and the time of the harvest for antigen-specific chicken interferon-γ (ChIFN-γ) production of splenocytes from immune chickens. Supernatants of stimulated splenocytes were harvested after 24, 48, 72, or 96 h of activation. The ChIFN-γ production was determined by the ChIFN-γ sandwich ELISA (P < 0.005).

Discussion

Interferon-γ is a significant regulatory cytokine in animals’ and birds’ immune systems, and is therefore an important indicator of immune function. Due to the lack of efficient methods, the study of this important cytokine was hindered. Therefore, it is urgent to establish a convenient assay to detect this important cytokine especially on the protein level.

Previous studies have reported that mAbs with high biological activity could be developed using recombinant antigens derived from E. coli, and could be used to establish a potent sandwich ELISA (24–26). In this paper, 4 specific mAbs against recombinant ChIFN-γ were generated, and the specificity of the 4 mAbs was confirmed by ELISA, Western blot, and immunofluorescence analysis. In order to find the optimal mAb combinations, the 4 mAbs were purified and used in different combinations, with the most sensitive mAb pair (3E5, 3E3) subsequently selected as capture and detecting reagents to develop a capture ELISA with high-sensitivity and low-background.

Our results demonstrated that this capture ELISA based on mAbs could detect both recombinant and native forms of ChIFN-γ. Increasing culture time, mitogen concentration, and numbers of splenocytes activated led to increased detection of ChIFN-γ, indicating that this assay is adequate to measure in vitro release of ChIFN-γ. Furthermore, this capture ELISA was successfully used to detect limited native ChIFN-γ production stimulated ex vivo by NDV recall antigen, suggesting that this ELISA is likely to be useful for measuring native IFN-γ in biological samples and for studying the function of native IFN-γ in vivo.

This sandwich ELISA will be particularly useful in examining the role of ChIFN-γ in the induction of the immune response to several infectious agents and in the development of protective responses in the chicken. Additionally, this assay could be used to detect ChIFN-γ as a surrogate for a robust Th1 response in several chicken models of infectious diseases. We are currently investigating whether the mAbs reported here can be applied to intracellular cytokine staining by flow cytometry or to ELISPOT assays to enumerate IFN-γ-secreting cells. At the very least, these antibodies can serve as useful reagents to develop additional specific IFN-γ immunoassays.