Establishment and Evaluation of a Bead-based Luminex Assay Allowing Simultaneous Quantification of Equine IL-12 and IFN-γ

I. Preliminary Work

Antibody cross-reactivity. Cross-reactivity with equine IL-12. The tested antibodies are listed in Table I. Immunofluorescence, dot blots, and western blots were used for preliminary testing of cross-reactivity of the antibodies.

Cross-reactivity with equine IFN-γ. The cross reaction of mAb to bovine IFN-γ and equine IFN-γ using western blot and flow cytometry was previously reported (1). Therefore, the IFN-γ mAb Clone CC302 (AbD Serotec, Duesseldorf, Germany) and bIFN-γ-I (Mabtech AB, Nacka Strand, Sweden) were tested only in dot blots with equine serum samples (Table I).

Methods used to evaluate cross reactivity. Western blotting. Cellular lysates from IL-12-transfected cells were denaturated (5 min at 95°C) in reducing sample buffer, separated by sodium dode-cylsulfate–polyacrylamide gel electrophoresis (SDS–PAGE; 4% stacking gel, 12% running gel) and transferred to a methanol-activated polyvinylidene difluoride (PVDF) membrane 0.2 μM pore size (Immobilon P; Millipore, Schwalbach, Germany). Membranes were blocked with 5% non-fat milk (Sigma-Aldrich)] in TBS [TBS: dH₂O plus 10 mM Tris-HCl and 150 mM NaCl (Sigma-Aldrich, Munich, Germany)) with 0.1% Tween 20 (TBS-T). Antibodies were then added (1:200 dilution) in TBS-T with 1% non-fat milk and incubated for 2 h at room temperature (RT) on a shaker. After washing, membranes were incubated for 1 h at RT on a shaker with a secondary, alkaline phosphatase (AP)-conjugated antibody (1:5000 dilution). Membranes were washed, incubated in AP buffer (100 mM Tris, 100 mM NaCl, 5 mM MgCl₂, pH 9.5) with 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tetrazolium substrate (NBT/BCIP) (Roche, Mannheim, Germany) on a shaker at RT protected from light. Finally, membranes were washed, dried and evaluated.

Dot blot. Protein samples and the respective controls (10 μl) were spotted on activated PVDF membranes (Immobilon P; Millipore), dried, blocked, and further handled as described for the western blot.

Anti-IL-12 immunofluorescence. Eight hours prior to transfection 3×10⁵ MTH53A (derived from healthy epithelial canine mammary tissue; Small Animal Clinic, University of Veterinary Medicine, Hannover, Germany) and HoMelZh (equine melanoma cell line, provided by Dr. M. Seltenhammer, Veterinary University of Vienna, Austria) cells were seeded in 6-well plates. Cells were grown under standard conditions in complete medium. Transfection was performed as described below. Twenty four hours post transfection, the respective cells were fixed in a solution of 4% of paraformaldehyde/PBS for 20 min at RT, permeabilized and blocked. The primary antibodies (Table I) at a 1:40 dilution and the corresponding secondary antibodies diluted 1:180 were added. Fluorescence microscopy was carried out using a Leica DMI 6000 fluorescence microscope (Leica Microsystems GmbH, Wetzlar Germany).

Antibody biotinylation. For the bead-based Luminex assay, one antibody from each cross-reacting pair (anti-bovine IL-12 clone CC326; AbD Serotec, Duesseldorf, Germany; anti-bovine-IFN-γ mAb clone bIFNγ-I (Mabtech AB, Nacka Strand, Sweden)) was biotinylated to enable its use as a bAb. Biotinylation was performed using an EZ-Link® Sulfo-NHS-LC-Biotinylation Kit (Thermo Fisher Scientific, Perbio Science, Bonn, Germany).

Mammalian expression vectors for recombinant equine IL-12 production. For the generation of IL-12 standards and IL-12 positive controls, two different mammalian cell lines were transfected with different expression vectors encoding equine IL-12. IL-12 was harvested from the supernatants and lysates of the transfected cells. This step is necessary as in contrast to IFN-γ, since currently no purified equine recombinant IL-12 is commercially available.

Equine IL-12 coding expression plasmids: The pUSEr-IRES-IL12 plasmid was provided by Dr. L. Nicolson, University of Glasgow, via Intervet International, Boxmeer, the Netherlands (3, 12).

Two additional IL-12 DNA expression vectors were constructed to accurately analyse the achieved transfection efficiencies via fluorescence microscopy by Green Fluorescent Protein (GFP) detection. The pIRES-hrGFPII expression vector backbone allows the simultaneous but separate expression of hrGFP and the protein of interest (equine IL-12). Additionally, in the pIRES-hrGFPII vectors, the genes of interest are fused to three contiguous copies of the FLAG® epitope allowing detection of the target protein by the FLAG®-Tag (Vitality hrGFP II Mammalian Expression Vectors, catalog no.#240157; Agilent Technologies, Palo Alto, CA, USA).

pIRES-hrGFPII-eIL-12 was generated as previously described (13). pIRES-hrGFPII-Flexi-eIL-12 was generated as follows. Equine IL-12 DNA containing the p35 and p40 IL-12 subunit cDNAs separated by an in-frame [Gly4Ser]3 linker (provided by Dr. L. Nicolson, University of Glasgow, via Intervet International) was amplified using PCR (primer pair NotI_flexi_f5’ CGGCGGCCGCATGGGTCACCAGTGGTTGG3’/IRES_NotI_flexi_r5’CGGCGGCCGCAGGAAGCATTCAGATAGC3’). The generated PCR products were separated on a 1.5% agarose gel, eluted using QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany), and cloned into the bicistronic pIRES-hrGFPII mammalian expression vector (Stratagene, La Jolla, CA, USA) according to the respective manufacturer's instructions. Verification of the constructed plasmid was carried out by specific enzymatic NotI restriction digestion and additionally by conventional sequencing.

Cell transfection. MTH53A and HoMelZh cells were grown as adherent cultures in a humidified atmosphere at 37°C with 5% CO₂ in complete medium [MTH53A cells: medium 199 (Invitrogen, Karlsruhe, Germany), HoMelZh cells: RPMI-1640 (Biochrom AG, Berlin, Germany); supplemented with 5% heat-inactivated fetal calf serum (PAA Laboratories GmbH, Pasching, Austria), 200 U/ml penicillin and 200 ng/ml streptomycin (Biochrom AG)]. Cells were treated and transfected as previously described (13). In brief, 2×10⁵ cells/24 well-plate were seeded. Using TransIT-2020 transfection reagent (Mirus Bio LLC, Madison, Wisconsin, USA) and following the manufacturer's instructions, 12 h after seeding, cells were transfected with three different equine IL-12 mammalian expression vectors (pIRES-hrGFPII-eIL-12, pIRES-hrGFPII-Flexi-eIL12, pUSEr-IRES-IL12). For each transfection approach, a separate incubation of the respective cells with the transfection reagent in the absence of the DNA expression plasmids was carried out as a negative control. Each protocol was performed in triplicate. After transfection, cells were incubated for 24 h in complete medium at 37°C and 5% CO₂. The plasmid DNA uptake efficiency for pIRES-hrGFPII-eIL-12 and pIRES-hrGFPII-Flexi-eIL-12 was verified by fluorescence microscopy due to the simultaneous but independent GFP expression. Transfection supernatants were harvested after 24 h, aliquoted and stored at −20°C until analysis.

PBMC preparation and cell culture. Fresh blood samples (70 ml/sample, one sample per horse) were collected from the jugular vein of six healthy horses into sterile sodium heparin vacutainer tubes (Greiner Bio-One GmbH, Frickenhausen, Germany). Additionally, from each horse a 9.5 ml serum separator vacutainer tube (Greiner Bio-One GmbH, Frickenhausen, Germany) was collected and used for serum separation; aliquots were prepared and stored at −20°C until analysis.

Within 1 h of sample collection, heparinised blood was diluted in phosphate-buffered saline (PBS, pH 7.3), layered on a Ficoll density gradient (d=1.077; Biochrom AG) and diluted to a final concentration of 1×10⁶ cells/ml in complete RPMI-1640 medium (Biochrom AG) supplemented with 5% fetal calf serum (FCS; PAA, Coelbe, Germany), and 1% penicillin-streptomycin (Biochrom AG). One mililiter cell suspension was added to a 24-well tissue culture plate in the presence or absence of the respective stimulating agents. All stimulation tests were performed in triplicate. At the end of the respective culture period, supernatants were collected from each well, aliquoted and stored at −20°C until analysis.

PBMC culture supernatants were used for commercially available ELISAs and the bead-based cytokine Luminex assays (non-multiplex or multiplex assays). Supernatants of non-stimulated PBMCs were used as controls.

PBMC stimulation. PMA/Ionomycin stimulation: Phorbolmyristate acetate (25 ng/ml; PMA, Sigma-Aldrich Chemie GmbH, Munich, Germany) and Ionomycine (500 ng/ml; Sigma-Aldrich Chemie GmbH) were added to the culture media and used to stimulate IFN-γ production by the PBMCs. After 24 h incubation, the supernatants were used as positive controls in the ELISA and Luminex assay.

IFN-γ/Lipopolysaccharide (LPS) stimulation: Equine IL-12 production was stimulated by incubating equine PBMCs for 2 h with 1000 U/ml eIFN-γ (Kingfischer, Biomol, Hamburg, Germany) followed by 10 h incubation with 1 μg/ml LPS (LPS, Escherichia coli, Sigma L2755), as described previously (17). The respective supernatants were used as positive controls in the ELISA and Luminex assays.

Transfection supernatant stimulation: 500 μl of the harvested supernatants from cells transfected with IL-12-encoding vectors were used to stimulate 1×10⁶ PBMCs/ml. After 24 h incubation, supernatants were collected and cytokine concentrations measured.

ELISAs used for comparison with the results of the bead-based analysis. To analyse the IL-12 concentration present in the samples and compare the results with those ones obtained with the generated bead-based assay, two commercially available anti-equine IL-12 ELISAs (IL-12A E90059Eq and IL-12 E90058Eq, Uscn Life science Inc., LOXO GmbH, Dossenheim, Germany) were evaluated.

Supernatants and cellular lysates from IL-12 transfected cells and their respective negative controls were used following the supplier's recommendations.

An anti-equine IFN-γ ELISA (Equine IFN-γ ELISA kit (ALP), 3117-1A-6; MABTECH, Nacka Strand, Sweden) was used to compare the bead-based assay results with the ones obtained with the commercially available ELISA kit.

II. Development and Establishment of the Bead-based Luminex Assay Coupling of mAb to fluorescent beads. For the establishment of the assay, mAbs were independently-coupled to fluorescent beads (Luminex Corp., http://www.luminexcorp.com). In brief, the anti-bovine IL-12 clone CC301 (AbD Serotec) was coupled to bead 22 (Luminex corporation, http://www.luminexcorp.com), anti-bovine IFN-γ clone CC302 (AbD Serotec) to bead 30 and the anti-FLAG® clone M2 (Sigma-Aldrich Chemie GmbH) to bead 42. The coupling reaction was performed using an antibody coupling kit (catalogue no.40-50016; Luminex, Austin, TX, USA) according to the manufacturer's instructions. For antibody coupling, 30-40 μg of the respective mAb and 6.25×10⁶ beads were used. After coupling, the beads were counted to assess the number of beads recovered after each coupling reaction and stored in the dark at 2-8°C. Antibody coupling confirmation was performed for each bead type using a biotinylated anti-mouse antibody as recommended by Luminex Corporation (http://www.luminexcorp.com/prod/groups/public/documents/lmnxcorp/322-antibody-coupling-kit-sell.pdf).

Standard curves for IL-12 and IFN-γ bead-based assay. The standard curves for equine IL-12, equine IFN-γ and FLAG®-Tag were generated using beads coupled to mAbs to anti-IL-12, IFN-γ and FLAG®, respectively. Beads were sonicated, mixed and diluted in blocking buffer (PBS-1% BSA; P3688; Sigma-Aldrich Chemie GmbH) to a final concentration of 6×10⁴ beads/ml each. For each assay, 3×10³ beads/well were used. Cytokine standards were prepared in blocking buffer. The standard curve for anti-IL-12-coupled beads was generated using 1:2 dilutions of the eIL-12 transfected cell lysates (e.g. 5000, 2500, 1250, 625, 312.5, 156.25 and, 78.125 pg/ml, Bradford protein quantification). The equine IFN-γ standard curve was generated using 1:2 dilutions of recombinant equine IFN-γ (e.g. 3000, 1500, 750, 375, 187.5, 93.75, and 46.875 pg/ml), and the standard curve for the competitive immunoassay with 1:2 dilutions of the biotinylated FLAG-BAP® Fusion Protein (3000-0 pg/ml) (Sigma-Aldrich Chemie GmbH).

Bead-based assays. IL-12 concentration assessment using a bead-based competitive immunoassay. This assay was established to assess more accurately the equine IL-12 concentrations of the samples used to generate the standard curves (lysates or supernatants of cells transfected with equine IL-12 encoding vectors). After transfection, protein concentration was primarily estimated using a commercially available Bradford protein quantification assay (catalogue no. 500-0007; Bio-Rad Laboratories, Munich, Germany). The Bradford assay relies on the binding of a dye (Coomassie Blue G250) to the protein. The dye binds all the arginyl and lysyl protein residues of the sample (14). Therefore no differentiation between cells own protein production and the IL-12 expression after transfection can be made.

The difference between the measured protein concentrations in lysates/supernatants of transfected and non-transfected cells was considered as an estimation of the amount of IL-12 in the analysed samples. In the vector used to generate equine IL-12 (pIRES-hrGFPII-Flexi-eIL-12), the IL-12 gene is fused to those of the FLAG® protein, accordingly, the concentration of FLAG® present in the sample is equivalent to the amount of IL-12 produced by the cells.

For the competitive immunoassay, the obtained mean fluorescent intensity (MFI) signal decreases with increasing analyte concentration due to the presence of a competing analyte. The antibody specific to the analyte (FLAG®-Tag) is coupled to the beads' surface and the analyte present in the sample (IL-12 fused to FLAG®-Tag) competes with the labelled analyte (biotinylated Carboxy-terminal FLAG-BAP® Fusion Protein; Sigma-Aldrich Chemie GmbH) for binding to the antibody-coupled beads (anti-FLAG®-coupled beads). Using this method, the anti-FLAG®-coupled beads and a known concentration of biotinylated Carboxy-terminal FLAG-BAP® Fusion Protein enable the FLAG® concentration present in the sample to be determined.

The Carboxy-terminal FLAG-BAP® Fusion Protein (Sigma-Aldrich Chemie GmbH) was biotinylated (EZ-Link® Sulfo-NHS-LC-Biotinylation Kit; Thermo Fisher Scientific, Perbio Science, Bonn, Germany) and the concentration of the biotinylated FLAG-BAP® Fusion Protein (inhibitory concentration 70; IC₇₀) to be applied was determined by titrating the labelled reagent together with the anti-FLAG®-coupled beads (http://www.luminexcorp.com/prod/groups/public/documents/lmnxcorp/competitive-immunoassay-antige.pdf).

A working bead mixture of 12×10⁴ anti-FLAG®-coupled beads/ml was prepared and 25 μl were applied per well. Seven 1:2 dilutions of the eIL-12 transfected cell lysates were made in duplicate; 25 μl of the respective standard or background were applied to the wells and plates were incubated for 30 min on a shaker at RT. After washing, 50 μl of streptavidin–phycoerythrin (Invitrogen, Carlsbad, CA, USA) diluted in blocking buffer were added to the wells and the plate was incubated for 30 min as above. The wells were washed as described earlier and the beads resuspended in 100 μl of blocking buffer. The plate was analysed in a Luminex 200™ instrument (Luminex Corp., http://www.luminexcorp.com). Results were reported as MFI. For the subsequent calculation of equine IL-12 concentration of the lysate samples, the MFI obtained while increasing the lysate concentration, is subtracted from the MFI measured from the inhibitory concentration 70 (IC₇₀) concentration of the biotinylated FLAG-BAP® Fusion Protein. The obtained MFI value is then used to calculate the corresponding protein concentration (equine IL-12) using the FLAG® Tag standard curve and the linear interpolation formula y=xm+n (y: measured mOD, x: unknown protein concentration, m: slope of the line, n: y-intercept).

Standard curve tests. For each standard test, 50 μl of bead solution were added to the well (Polystyrene 96-well plates; Corning, Axygen, Amsterdam, the Netherlands). Seven cytokine dilutions and one background were applied in duplicate. In each well (96-well plate), 50 μl of the corresponding diluted standard concentration or the background samples were applied and plates were incubated for 30 min on a shaker at RT. After washing with PBS, 50 μl of the respective primary bAb diluted in blocking buffer were added to each well and plates were incubated for 30 min as described above. For the anti-equine IL-12 beads, the biotinylated mAb anti-bovine IL-12 clone CC326 (AbD Serotec) was used. The biotinylated anti-bovine-IFN-γ mAb clone bIFN-γ-I (Mabtech AB, Nacka Strand, Sweden) was used with the anti-IFN-γ coupled beads. Thereafter, wells were washed and 50 μl of streptavidin–phycoerythrin (Invitrogen) diluted in blocking buffer were added, respectively. The plate was incubated for 30 min as described above and then washed. After all incubation steps, the beads were resuspended in 100 μl blocking buffer and subsequently analysed as above. The standard curve fitting was performed using the logistic 5p formula y=a+b/(1 + (x/c)^d)^f (y: actual measured MFI, a: estimated response at zero concentration, b: estimated response at infinite concentration, c: mid-range concentration (EC50), d: slope factors-Hill slope-, and f: asymmetry factor; Luminex 200, xPONENT software).

Multiplex IL-12 and IFN-γ bead-based assay. Beads coupled with mAbs against IL-12 and IFN-γ mAb were sonicated, mixed and diluted in blocking buffer to a final density of 6×10⁴ beads/ml each. A standard containing all three cytokines was prepared in blocking buffer, and 1:2 dilutions of the IL-12-transfected cell lysates and recombinant equine IFN-γ prepared. Per well, 50 μl bead mixture and 50 μl of the corresponding standard dilution or background were added and plates were incubated for 30 min on a shaker at RT. After washing with PBS, 50 μl of the primary bAb mixture [biotinylated mAb anti-bovine IL-12 clone CC326 (AbD Serotec) and biotinylated anti-bovine-IFN-γ mAb (Mabtech AB)] were added to each well and plates were incubated for 30 min. Subsequently, wells were washed and 50 μl of streptavidin-phycoerythrin (Invitrogen) diluted in blocking buffer were added. After 30 min incubation and washing, beads were resuspended in 100 μl of blocking buffer and the plate analysed as explained above.

MFIs obtained with the multiplex standards were then compared with those of the non-multiplex assays.

Sensitivity and specificity of the bead-based cytokine detection assays. The sensitivity for each cytokine corresponded to the lowest detectable cytokine concentration in the linear range of the individual cytokine standard curve where the measured MFI was higher than the MFI of the background plus two standard deviations [real measured value: (MFI sample) > (background MFI + 2×SD)]. The analytical specificity of the multiplex assay was determined performing cross-reactivity tests. First only the beads were mixed and the analytes (recombinant eIL-12 and recombinant eIFN-γ) and dAbs applied separately. Secondly, beads and analytes were mixed and finally, beads, analytes and dAbs were applied together in the same well. Afterwards, to test if native cytokines or other components interfere with the assay performance, cell culture supernatants from stimulated PBMCs or supernatants/lysates of cells transfected with eIL-12-encoding vectors, expected to contain native IL-12 and/or IFN-γ, were used as samples in the multiplex assay.

III. Sample Analysis

After establishing the bead-based assay for equine IL-12 and IFN-γ, the assays were tested using different samples. All samples were primarily tested with the corresponding ELISAs and non-multiplex bead-based assays. Subsequently, the multiplex bead-based assay was performed using the same samples to compare these results with the ones obtained with the ELISAs and non-multiplex assays.

Samples used to assess IFN-γ concentrations. Equine serum dilutions (1:10, 1:100, 1:1000), supernatants from stimulated PBMCs [stimulated with: phorbol 12-myristate 13-acetate (PMA)/Ionomycin and equine IL-12-transfection supernatants] and, the respective negative controls.

Samples used to assess IL-12 concentrations. Equine serum dilutions (1:10, 1:100, 1:1000), supernatants from stimulated PBMCs (stimulated with PMA/Ionomycin and IFN-γ/LPS and equine IL-12-transfection supernatants) and the respective controls.

Statistics. Statistical significance was determined using the two-tailed Wilcoxon Mann-Whitney test. Differences were considered statistically significant for p≤0.05.

To compare cytokine results obtained with the ELISA and Luminex assay Pearson's correlation analysis was applied.

Statement of ethical approval. Blood used for this study was collected as part of a routine procedure prior to anaesthesia from horses involved in an anaesthesiology research project. All procedures were approved by the State Office for Consumer and Food Safety in accordance with the German Animal Welfare Law (No. 33.12-42502-04-11/0572).