TECHNICAL REPORT
Enzyme Linked Immunosorbent Assay for Quantification of Bovine Interleukin-8 to Study Infection and Immunity in the Female Genital Tract James G. Cronin, Rebecca Hodges, Sara Pedersen, Iain M. Sheldon Institute of Life Science, College of Medicine, Swansea University, Swansea, UK
Keywords Bovine, interleukin-8, ELISA, endometrium, inflammation Correspondence James G. Cronin, Institute of Life Science, College of Medicine, Swansea University, Singleton Park, Swansea SA2 8PP, UK. E-mail: [email protected] Submission September 11, 2014; accepted November 4, 2014. Citation Cronin JG, Hodges R, Pedersen S, Sheldon IM. Enzyme linked immunosorbent assay for quantification of bovine interleukin-8 to study infection and immunity in the female genital tract. Am J Reprod Immunol 2014 doi:10.1111/aji.12344
Problem The chemokine IL-8 recruits neutrophils to sites of infection, including the endometrium of the bovine uterus. However, quantification of bovine IL-8 often yields lower concentrations than for other species, which may reflect impaired innate immune responses by bovine cells or inaccurate measurement of IL-8 using the current human IL-8 ELISA method. Method of Study An ELISA was developed and validated for detection of bovine IL-8. Utility of the assay was tested by measuring the response of bovine endometrium and cells to bacteria and pathogen-associated molecular patterns. Results The developed ELISA detected 62.5–2000 pg/mL IL-8, with minimal cross-reactivity to other inflammatory mediators. Concentrations of bovine IL-8 were measured more accurately by the bovine than human IL-8 ELISA. Bovine endometrial IL-8 responses to pathogen-associated molecules were quantitatively similar to other species. Conclusion A bovine-specific IL-8 ELISA was developed, which accurately measured IL-8 secretion from endometrial cells.
Introduction Innate immunity is predicated on pattern recognition receptors such as Toll-like receptors (TLRs) binding to pathogen-associated molecular patterns (PAMPs).1 Binding of PAMPs to TLRs leads to the activation of nuclear factor (NF)-jB and MAPK intracellular signaling pathways, and the secretion of cytokines including Interleukin (IL)-1b, IL-6, IL-10, interferon (IFN)c and tumor necrosis factor (TNF)a, and the production of chemokines such as IL-8 and CCL2.1 Interleukin-8 is a chemokine that primarily recruits neutrophils, which are important for American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
clearance of pathogens and tissue repair following microbial infection.2 In addition, IL-8 has roles in physiology and tissue homeostasis, particularly where there is involvement of neutrophils in tissue remodeling such as in the endometrium and ovary.3 Microbial contamination of the endometrium is ubiquitous in dairy cows after parturition, and pathogenic bacteria cause uterine disease in 40% of animals.4 Uterine disease is characterized by tissue damage, the accumulation of cytokines and chemokines in the endometrium, and an influx of neutrophils into the endometrium to form pus.4–6 Uterine disease is caused by Gram-negative Escherichia 1
CRONIN ET AL.
coli and Gram-positive Trueperella pyogenes, Fusobacterium necrophorum, Bacteroides and Prevotella bacteria, and by bovine herpesvirus 4.7,8 Epithelial and stromal cells are the first to encounter bacteria invading the endometrium, and these cells mount an innate immune response by detecting PAMPs, such as lipopolysaccharide (LPS) from Gram-negative bacteria, and lipopeptides from all classes of bacteria.9 The immune response is characterized by increased expression of IL1B, IL6, and IL8 in endometrial tissues and cells collected from diseased animals.6,10 Similarly, ex vivo organ culture (EVOC) of endometrium secretes IL-1b, IL-6, and IL-8 in response to E. coli and T. pyogenes.11 Furthermore, endometrial epithelial and stromal cells express functional TLR2 and TLR4, and LPS and lipopeptides stimulate the secretion of IL-6 and IL-8.9,12 Bovine herpes virus 4 also stimulates IL-8 secretion from endometrial cells.7 Indeed, IL-8 appears particularly important during uterine disease in cattle because the signs of endometritis can be replicated by infusing recombinant bovine IL-8 into the uterus.13 Previous studies measured the concentrations of bovine IL-8 in biological fluids and cell-culture supernatants using human IL-8 ELISA kits, as the antibody pairs cross-react with bovine IL-8.14,15 Although such methods reveal similar trends for IL-8 production in response to bacteria or PAMPs between cattle and other species, the responses are quantitatively higher in humans than cattle. For example, bovine macrophages required stimulation with up to 100-fold higher concentrations of bacterial PAMPs in order to induce similar concentrations of IL-8 compared with human macrophages treated with the same PAMPs.16 Similarly, human endometrial stromal cells treated with 1 lg/mL LPS for 24 hr produced 2500 pg/mL of IL-817; whereas, bovine endometrial stromal cells treated with the same concentration of LPS secreted only 54 pg/mL, when assayed using the human IL-8 ELISA kit method.9,12 The lower concentrations of IL-8 may reflect impaired innate immune responses by bovine cells compared with human cells or inaccurate measurement of IL-8 using the current human IL-8 ELISA. To address this issue, the objectives of this study were to develop, validate and characterize an ELISA for measuring bovine IL-8, and to demonstrate the utility of the assay using biological fluids and in vitro models of inflammation in the bovine endometrium.
2
Materials and methods ELISA Development Antibodies A sandwich ELISA was developed and validated for bovine IL-8 using antibodies described previously.18 Monoclonal mouse anti-sheep IL-8 was used as a capture antibody (MCA 1660; Abd Serotec, Kidlington, UK), polyclonal rabbit anti-sheep IL-8 (AHP425; Abd Serotec) as a detect antibody, and a polyclonal goat anti-rabbit horseradish peroxidase (HRP) as a tertiary antibody (P0448; DAKO UK Ltd, Ely, UK). The capture antibody was resuspended in 50 lL carbonate/bicarbonate buffer (0.2 M Na2CO3, 0.2 M NaHCO3, pH 9.4; C3041; Sigma, Gillingham, UK), and the detect antibody was resuspended in Dulbecco’s phosphate buffered saline (D-PBS; D8537; Sigma) containing 10% normal mouse serum (M5905; Sigma). The ELISA procedure (described below) was optimized by testing the following ranges of antibody concentrations: 1.43–5 lg/mL capture antibody, 0.146–0.5 lg/mL detect antibody, and 0.025–0.25 lg/mL tertiary antibody. Selection of blocking buffers To select an appropriate reagent for use as a blocking buffer, ELISA plate wells were treated in four duplicate wells with the following blocking reagents in DPBS: 5% bovine serum albumin (BSA; A7906; Sigma), 1% BSA (Sigma), 1% horse serum albumin (HSA; H1138; Sigma), 5% Tween-20 (P9416; Sigma), 5% skimmed milk (#9999; Cell Signaling, Leiden, the Netherlands), or 4% fish skin gelatin (FSG; G7765; Sigma). Developed ELISA procedure Wells of 96-well microtiter ELISA plates (675061; Greiner Bio-one, Stonehouse, UK) were coated with 50 lL per well of 2.5 lg/mL monoclonal mice antisheep IL-8 diluted in 50 lL carbonate/bicarbonate buffer, and the plates incubated for 18 hr at room temperature. The antibody solution was discarded, and the plates were washed three times with 150 lL per well 0.05% (vol./vol.) Tween-20 (P9416; Sigma) in D-PBS (D8537; Sigma) using a platewasher (LT3500; Labtech, Uckfield, East Sussex, UK). Wells were then blocked with 150 lL per well of 4% FSG blocking buffer in D-PBS for 1 hr, at room temperature, to block non-specific signals. A seven-point American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
BOVINE IL-8 ELISA
standard curve was generated by 1 in 2 serial dilution of recombinant bovine IL-8 (4000–62.5 pg/mL; RP0023B; Kingfisher Biotech, Inc. Saint Paul, MN, USA; supplied by 2B Scientific, Upper Heyford, UK) in 4% FSG block, or sample matrix if measuring IL8 in biological fluids, and added to duplicate wells, leaving duplicate analyte-free wells containing 50 lL per well matrix as blank wells. Test samples of mucus, serum, endometrial cell-culture supernatants or EVOC supernatants were loaded in duplicate wells using 50 lL per well. The plates were then incubated for 1.5 hr at room temperature, and then washed as described above. To each well, 50 lL of 0.145 lg/mL (a 1 in 700 dilution) polyclonal rabbit anti-sheep IL-8 in 4% FSG was added and the plates incubated for 2 hr at room temperature and then washed. The plates were then incubated with 50 lL of 0.042 lg/mL HRP polyclonal goat anti-rabbit in 4% FSG for 1 hr, in the dark. The plates were then washed and the reaction visualized by the addition of 50 lL/well colorimetric substrate 3,30 ,5,50 -tetramethylbenzidine (TMB; BD Biosciences Oxford, UK) and incubation in the dark for 15 min. The reaction was stopped by the addition of 50 lL/well 0.5 M sulfuric acid (Sigma), and absorbance was measured at 450 and 550 nm using a plate reader (Fluostar; BMG LABTECH, Ortenberg, Germany). Optical density (OD) values at 550 nm were subtracted from OD values read at 450 nm to correct for optical imperfections in the microtiter plate. The absorbance values for blank wells were subtracted from absorbance values obtained for the standards, and samples and the values for the duplicate wells were averaged. A seven-point standard curve was generated by plotting the average absorbance for each standard concentration on the Y-axis versus the corresponding bovine IL-8 concentration on the X-axis using the MARS Data Analysis software (BMG LABTECH). The sample IL-8 concentrations pg/mL were determined by interpolating from the absorbance value (Y-axis) to bovine IL-8 concentration (X-axis) using four-parameter logistic regression of the standard curve, generated by the MARS Data Analysis software. ELISA Validation Recovery and reproducibility To determine the efficiency of recovery of recombinant bovine IL-8, RPMI 1640 medium (Life Technologies, Paisley, UK) supplemented with 10% FBS was spiked with low, medium, or high (500, 1000, or American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
1500 pg/mL) concentrations of recombinant bovine IL-8 and assayed using the developed bovine IL-8 ELISA. To determine whether the matrix affected OD450 readings, seven-point standard curves were generated by diluting recombinant bovine IL-8 in the following matrices: Dulbecco’s Modified Eagle’s Medium (DMEM; supplemented with 10% FBS; Life Technologies); in vitro Maturation Medium (IVM; supplemented with 10% FBS; Life Technologies); 5% bovine serum albumin (BSA; Sigma); 4% FSG, RPMI 1640 (supplemented with 10% FBS; Life Technologies); and OPTI-MEM (Life Technologies) and assayed using the developed bovine IL-8 ELISA. The intra-assay coefficient of variation (CV) was determined from the average and standard deviation of the absorbance of 12 wells containing 500, 1000, or 2000 pg/mL recombinant bovine IL-8 assayed on the same microtitre plate. The interassay CV was determined from the average absorbance and standard deviation of 12 wells containing 500, 1000, or 2000 pg/mL concentrations of recombinant bovine IL-8 assayed on six independent microtitre plates. Specificity To determine cross-reactivity or interference from cytokines or chemokines commonly induced during the innate immune response, the following recombinant proteins were assayed using the bovine IL-8 ELISA: bovine IL-10 (50 ng/mL to 781.25 pg/mL; RP0379B; Kingfisher Biotech); bovine IL-6 (5 ng/mL to 78 pg/mL; 1861273; Thermo Scientific, Loughborough, UK); bovine IL-1b (2 ng/mL to 31.25 pg/mL; 1861270; Thermo Scientific); bovine IFNc (10 ng/mL to 156.25 pg/mL; 842590 R&D Systems, Abingdon, UK); bovine TNF (8 ng/mL to 125 pg/mL; TNF; 842630; R&D Systems) or bovine chemokine CCL2 (5 ng/mL to 78 pg/mL; RP00227; Kingfisher Biotech). Duplicate seven-point standard curves were generated by 1 in 2 serial dilution of each recombinant protein in 4% FSG and compared with a duplicate seven-point standard curve generated by 1 in 2 serial dilution of recombinant bovine IL-8 (4000 pg/ mL to 62.5 pg/mL) diluted in 4% FSG. Dilution linearity To evaluate linearity of dilution, seven-point standard curves were generated using recombinant bovine IL-8 encompassing the range of 32,000 pg/ mL to 62.5 pg/mL IL-8, with each standard subsequently diluted 1:10 (3200 pg/mL to 6.25 pg/mL), 3
CRONIN ET AL.
1:20 (1600 pg/mL to 3.13 pg/mL), or 1:40 (800 pg/ mL to 1.56 pg/mL) in 4% FSG. The dilutions were evaluated in the developed bovine IL-8 ELISA, and the observed values were plotted against the expected concentrations. To determine the upper limit of quantification, a standard curve was generated by serial dilution of recombinant bovine IL-8 (32,000 pg/mL to 31.25 pg/mL). Correlation Commercially available human IL-8 ELISA kits have previously been validated for measurement of IL-8 in bovine samples.14,15 So, to establish a correlation between the developed bovine IL-8 ELISA and the human IL-8 ELISA DuoSet kit (DY208; R&D Systems, Inc.), recombinant bovine IL-8 protein or supernatants from EVOCs treated for 24 hr with 1 lg/ mL ultrapure LPS (Invivogen, Toulouse, France) were assayed on the developed bovine IL-8 ELISA and the human IL-8 DuoSet ELISA, following the manufacturer’s instructions. Samples were assayed in duplicate, using both IL-8 ELISA systems on the same day. Demonstration of Utility of ELISA Isolation and culture of endometrial cells Endometrial tissue was dissected and processed as described previously to isolate and culture endometrial cells.12,19 Primary endometrial epithelial and stromal cells were cultured in 75 cm2 flasks (Greiner Bio-One, Gloucester, UK) in a humidified atmosphere of air with 5% CO2 at 37°C, with culture medium changed every 48 hr. Epithelial and stromal cell populations were distinguished by cell morphology, and the absence of immune cell contamination confirmed by the absence of CD45, as described previously. 9 Cells were seeded at 2 9 104 per well and treated in 24-well tissue culture plates (TPP, Trasadingen, Switzerland). Ex vivo organ culture Ex vivo organ culture of endometrium was generated as described previously.11 The EVOCs were cultured in six-well plates (TPP) with the epithelial surface uppermost in 3 mL per well of RPMI 1640 (supplemented with 10% FBS; Life Technologies). Plates were incubated in a humidified atmosphere with 5% CO2 in air at 37°C, as described previously.11 4
Isolation and culture of peripheral blood macrophages The isolation and culture of peripheral blood macrophages were performed as previously described,19 with minor modifications. Briefly, jugular venous blood was collected at slaughter into sterile 50-mL tubes (NUNC) containing 5 mL sodium citrate (3.8% w/v in HBSS; Sigma) supplemented with 50 ng/mL gentamicin (Sigma). Tubes were gently inverted 10 times to ensure sufficient mixing before transport to the laboratory at 4°C. Blood was divided into 25-mL volumes and diluted 1:1 with sterile HBSS supplemented with gentamicin (50 ng/mL; Sigma) and amphotericin B (2.5 lg/mL; Sigma). Aliquots of the diluted blood (16 mL) were then layered onto 12mL Ficoll-Paque PREMIUM (GE Healthcare Life Sciences, Buckinghamshire, UK) in a 50-mL tube. Tubes were then centrifuged at 500 g for 40 min at 4°C with the centrifuge brake turned off. Following centrifugation, the peripheral blood mononuclear cell layer was carefully removed and transferred to a sterile 15-ml tube. Cells were washed twice in HBSS supplemented with gentamicin, with centrifugation at 500 g for 15 min, resuspended in RPMI 1640 medium and transferred to 60-mm Petri dishes (Greiner) and incubated at 37°C for 30 min in a humidified atmosphere with 5% CO2 in air at 37°C. After incubation, the medium was removed and cells were washed twice with HBSS supplemented with gentamicin before the addition of RPMI 1640 medium with 10% FBS supplemented with gentamicin (50 ng/mL; Sigma), amphotericin B (Sigma). Cells were cultured for 6–8 days with medium changes every 48 hr to allow the maturation of monocytes. On the day of experimentation, macrophages were removed from the plate using Accutase cell detachment solution (Sigma) and seeded into 24-well plates at 1 9 106 cells. Treatment of Endometrial cells, EVOCs and macrophages To evaluate responses to PAMPs, epithelial cells, stromal cells, macrophages, and EVOCs were treated with LPS (10; 100; and 1000 ng/mL) for 24 hr and supernatants assayed using the developed bovine IL8 ELISA. Supernatants from stromal cells treated with 1 lg/mL ultrapure LPS from E. coli 0111:B4, 1 lg/mL Pam3CSK4 (Pam3CSK4; N-Palmitoyl-S(2,3-bis(palmitoyloxy)-(2RS)-propyl]-(R)-cysteinyl-(S)seryl-(S)-lysyl-(S)-lysyl-(S)-lysyl-(S)-lysine) PAM), 1 lg/mL FSL-1 (S-(2,3-bispalmitoyloxypropyl) CysGly-Asp-Pro-Lys-His-Pro-Lys-Ser-Phe, Pam2CGDPKH American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
BOVINE IL-8 ELISA
PKSF) (all InvivoGen, Toulouse, France) for 24 hr were also assayed using the developed bovine IL-8 ELISA. Finally, supernatants from macrophages treated with heat-killed E. coli (1 9 106, 1 9 107, or 1 9 108 CFU) for 24 hr were assayed using the developed bovine IL-8 ELISA. Bacterial culture and heat inactivation Cultures of E. coli (isolate MS482),8 collected from the uterus of postpartum cows with persistent uterine disease, were grown overnight until they reached log-phase growth in Luria–Bertani medium (Sigma) as described previously.11 The bacteria were diluted to 1 9 108 colony-forming units (CFU)/mL and centrifuged at 1500 g for 10 min at room temperature. The bacteria were then suspended in sterile PBS (Life Technologies Ltd) and heat-inactivated at 75°C for 45 min, and a sample of heat-inactivated bacteria was plated onto Luria–Bertani agar and incubated at 37°C for 48 hr to confirm complete inactivation. The bacteria were then centrifuged at 1500 g for 10 min at room temperature and washed three times in PBS, before suspension at a final concentration equivalent to 1 9 108 CFU/ mL in RPMI 1640 medium with 10% FBS supplemented with gentamicin (Sigma), amphotericin B (Sigma). Isolation of blood and mucus To further demonstrate utility of the IL-8 ELISA, mucus and serum samples were collected from Holstein cows in commercial dairy herds in Cheshire and Staffordshire that calved between May 1, 2011 and January 31, 2012. Each herd was visited weekly by a veterinarian (SP) and procedures conducted under the UK Animal Scientific Procedures Act (1986), with the approval of the UK Government Home Office (License number PPL 40/3478). Farm staff recorded whether calving was normal with no intervention or whether there was dystocia requiring manual intervention. One, three, and five weeks postpartum, 10 ml blood samples (no dystocia, n = 115 and dystocia, n = 63) were collected from the coccygeal vein blood vessels into an evacuated sterile tube without coagulant (Vacutainer; Becton), allowed to clot at 4°C for 6 hr prior to centrifugation at 1000 g for 10 min to harvest serum, which was stored at 20°C. In addition, mucus was collected from the vagina (no dystocia, n = 24 and dystocia, n = 12) using a clean, lubricated, gloved hand inserted through the vulva and into the vagina. The American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
mucus was placed in 30-mL sterile tubes and stored at 20°C. The mucus was processed prior to measurement of inflammatory mediators by ELISA as described previously.20 Briefly, 2 g mucus was placed in 10 mL of cytolyt solution (40% Methanol: 60% distilled water) and mixed with 0.1% Dithiothreitol (Sigma-Aldrich) until the mucus was disrupted, followed by centrifugation at 3000 g for 15 min, and collection of the supernatant. The concentrations of IL-8 were measured in serum and mucus with the developed bovine IL-8 assay.
(a)
(b)
(c)
Fig. 1 IL-8 ELISA development (a) Comparison of blank-corrected OD450 of IL-8 standard curves using the concentrations of tertiary antibody indicated. (b) Comparison of blank-corrected OD450 readings from analyte-free wells blocked with buffers indicated, using four duplicate wells for each buffer. (c) Comparison of blank-corrected OD450 of IL-8 standard curve diluted in the matrices indicated.
5
CRONIN ET AL.
(a) Table I A Comparison of Recombinant Bovine IL-8 (500, 1000, or 1500 pg/mL) Diluted in 4% FSG or RPMI (Supplemented with 10% FBS) in the Bovine IL-8 ELISA IL-8 diluted in 4% FSG (pg/mL)
Average % recovery of IL-8 in RPMI (n = 12)
Range % (n = 12)
500 1000 1500
128 132 144
112–129 120–142 130–169
Table II Cross-Reactivity or Interference by IL-6 for the Bovine IL-8 ELISA was Evaluated by Diluting Recombinant IL-6 in RPMI (Supplemented with 10% FBS), and Values were Interpolated from a Standard Curve Derived from Recombinant Bovine IL-8 Diluted in 4% FSG
(b)
Actual concentrations Observed concentrations of IL-6 (pg/mL) of IL-6 (pg/mL) % Cross-reactivity 5000 2500 1250 625 312.5 156.25 78
189.20 68.39 Below Y Below Y Below Y Below Y Below Y
range range range range range
3.78 2.74 0 0 0 0 0
Statistical Analysis Data are presented as arithmetic mean and standard error of mean (SEM) except where indicated. Data were analyzed by one-way ANOVA using GraphPad Prism 5 (La Jolla, CA, USA) with Dunnett’s post hoc test, and statistical significance described when P < 0.05.
Fig. 2 IL-8 ELISA validation (a) Dilution linearity using bovine IL-8 standards undiluted (diamonds); diluted 1 in 10 (squares); 1 in 20 (triangles); or 1 in 40 (circles) in 4% FSG n = 4 wells (b) Upper limit of quantification determined using a standard curve generated by serial dilution of recombinant bovine IL-8 (32,000 pg/mL–31.25 pg/mL) n = 4 wells.
Results and discussion The present study developed a sandwich ELISA method to detect bovine IL-8 in biological samples, based on previously described antibodies for rumi-
Table III Intra-assay CV for the Bovine IL-8 ELISA was Calculated from Values Derived from Recombinant Bovine IL-8 (Low, Medium, or High: 50, 1000, or 1500 pg/mL, Respectively; n = 12) Diluted in RPMI (Supplemented with 10% FBS) Assayed on the Same Microtitre Plate. Interassay CV was Calculated from Values Derived from Recombinant Bovine IL-8 (Low, Medium, or High: 50, 1000, or 2000 pg/mL, Respectively; n = 6) Diluted in RPMI (Supplemented with 10% FBS) Assayed on six Independent Microtitre Plates Interassay precision Bovine IL-8 (pg/mL) n Mean Standard deviation CV%
6
500 6 502 12 2.4
1000 6 998 32 3.2
Intra-assay precision 2000 6 2005 108 5.4
500 12 640 46 7.2
1000 12 1318 71 5.4
1500 12 2163 176 8.1
American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
BOVINE IL-8 ELISA
nant IL-8.18 Assay development involved iterative testing of a range of concentrations of the capture, detection, and tertiary antibodies; and testing a range of blocking buffers to reduce the background signal in the ELISA. Each antibody was selected on the basis of a relatively high dynamic range standard curve and a reciprocally low non-specific signal in the blank well (Fig. 1a). The concentrations selected were as follows: primary antibody 2.5 lg/mL; secondary antibody 0.145 lg/mL (a 1 in 700 dilution), and tertiary antibody 0.042 lg/mL. To minimize the non-specific signal, commonly used ELISA blocking buffers were assessed. In each assay, four duplicate wells were blocked with 5% BSA, 1% BSA, 1% HSA, 5% Tween-20, 5% skimmed milk or 4% FSG for 1 hr at room temperature. The ELISA was performed and values for non-specific signal generated in blank wells were determined at OD450. Blank wells blocked with 4% FSG exhibited the lowest OD450 reading (Fig. 1b). Thus, 4% FSG was used as the blocking buffer and reagent diluent for subsequent ELISA. The selection of specific concentrations for each antibody and the use of 4% FSG blocking buffer yielded standard curves with an effective range of 62.5–2000 pg/mL IL-8. (a)
Recombinant bovine IL-8 gave relatively higher OD450 readings when diluted in RPMI supplemented with 10% FBS compared to recombinant bovine IL8 standards diluted in 4% FSG (Table I). Therefore, the effect of diluting standards in different matrices on the OD450 readings was assessed and compared to recombinant bovine IL-8 standards diluted in 4% FSG. Recombinant bovine IL-8 standards (4000– 65 pg/mL) were diluted in DMEM, IVM, 5% BSA, 4% FSG, RPMI, or Opti-MEM. Relative blank-corrected OD450 readings varied for each concentration of recombinant bovine IL-8 depending on the matrix (Fig. 1c). To assess cross-reactivity or interference with the developed bovine IL-8 ELISA, a representative spectrum of cytokines and chemokines were assayed to reflect the typical innate immune response.1,9,10,12,21 The assay had little cross-reactivity with biologically meaningful concentrations of bovine IL-1b, IL-10, TNFa, IFNc, or CCL2 (all values below Y range). However, cross-reactivity or interference was exhibited with bovine IL-6, which represented 3.8% of the concentration assayed for a 5000 pg/mL sample of recombinant IL-6 (Table II). However, samples of
Enzyme Linked Immunosorbent Assay for Quantification of Bovine Interleukin-8 to Study Infection and Immunity in the Female Genital Tract James G. Cronin, Rebecca Hodges, Sara Pedersen, Iain M. Sheldon Institute of Life Science, College of Medicine, Swansea University, Swansea, UK
Keywords Bovine, interleukin-8, ELISA, endometrium, inflammation Correspondence James G. Cronin, Institute of Life Science, College of Medicine, Swansea University, Singleton Park, Swansea SA2 8PP, UK. E-mail: [email protected] Submission September 11, 2014; accepted November 4, 2014. Citation Cronin JG, Hodges R, Pedersen S, Sheldon IM. Enzyme linked immunosorbent assay for quantification of bovine interleukin-8 to study infection and immunity in the female genital tract. Am J Reprod Immunol 2014 doi:10.1111/aji.12344
Problem The chemokine IL-8 recruits neutrophils to sites of infection, including the endometrium of the bovine uterus. However, quantification of bovine IL-8 often yields lower concentrations than for other species, which may reflect impaired innate immune responses by bovine cells or inaccurate measurement of IL-8 using the current human IL-8 ELISA method. Method of Study An ELISA was developed and validated for detection of bovine IL-8. Utility of the assay was tested by measuring the response of bovine endometrium and cells to bacteria and pathogen-associated molecular patterns. Results The developed ELISA detected 62.5–2000 pg/mL IL-8, with minimal cross-reactivity to other inflammatory mediators. Concentrations of bovine IL-8 were measured more accurately by the bovine than human IL-8 ELISA. Bovine endometrial IL-8 responses to pathogen-associated molecules were quantitatively similar to other species. Conclusion A bovine-specific IL-8 ELISA was developed, which accurately measured IL-8 secretion from endometrial cells.
Introduction Innate immunity is predicated on pattern recognition receptors such as Toll-like receptors (TLRs) binding to pathogen-associated molecular patterns (PAMPs).1 Binding of PAMPs to TLRs leads to the activation of nuclear factor (NF)-jB and MAPK intracellular signaling pathways, and the secretion of cytokines including Interleukin (IL)-1b, IL-6, IL-10, interferon (IFN)c and tumor necrosis factor (TNF)a, and the production of chemokines such as IL-8 and CCL2.1 Interleukin-8 is a chemokine that primarily recruits neutrophils, which are important for American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
clearance of pathogens and tissue repair following microbial infection.2 In addition, IL-8 has roles in physiology and tissue homeostasis, particularly where there is involvement of neutrophils in tissue remodeling such as in the endometrium and ovary.3 Microbial contamination of the endometrium is ubiquitous in dairy cows after parturition, and pathogenic bacteria cause uterine disease in 40% of animals.4 Uterine disease is characterized by tissue damage, the accumulation of cytokines and chemokines in the endometrium, and an influx of neutrophils into the endometrium to form pus.4–6 Uterine disease is caused by Gram-negative Escherichia 1
CRONIN ET AL.
coli and Gram-positive Trueperella pyogenes, Fusobacterium necrophorum, Bacteroides and Prevotella bacteria, and by bovine herpesvirus 4.7,8 Epithelial and stromal cells are the first to encounter bacteria invading the endometrium, and these cells mount an innate immune response by detecting PAMPs, such as lipopolysaccharide (LPS) from Gram-negative bacteria, and lipopeptides from all classes of bacteria.9 The immune response is characterized by increased expression of IL1B, IL6, and IL8 in endometrial tissues and cells collected from diseased animals.6,10 Similarly, ex vivo organ culture (EVOC) of endometrium secretes IL-1b, IL-6, and IL-8 in response to E. coli and T. pyogenes.11 Furthermore, endometrial epithelial and stromal cells express functional TLR2 and TLR4, and LPS and lipopeptides stimulate the secretion of IL-6 and IL-8.9,12 Bovine herpes virus 4 also stimulates IL-8 secretion from endometrial cells.7 Indeed, IL-8 appears particularly important during uterine disease in cattle because the signs of endometritis can be replicated by infusing recombinant bovine IL-8 into the uterus.13 Previous studies measured the concentrations of bovine IL-8 in biological fluids and cell-culture supernatants using human IL-8 ELISA kits, as the antibody pairs cross-react with bovine IL-8.14,15 Although such methods reveal similar trends for IL-8 production in response to bacteria or PAMPs between cattle and other species, the responses are quantitatively higher in humans than cattle. For example, bovine macrophages required stimulation with up to 100-fold higher concentrations of bacterial PAMPs in order to induce similar concentrations of IL-8 compared with human macrophages treated with the same PAMPs.16 Similarly, human endometrial stromal cells treated with 1 lg/mL LPS for 24 hr produced 2500 pg/mL of IL-817; whereas, bovine endometrial stromal cells treated with the same concentration of LPS secreted only 54 pg/mL, when assayed using the human IL-8 ELISA kit method.9,12 The lower concentrations of IL-8 may reflect impaired innate immune responses by bovine cells compared with human cells or inaccurate measurement of IL-8 using the current human IL-8 ELISA. To address this issue, the objectives of this study were to develop, validate and characterize an ELISA for measuring bovine IL-8, and to demonstrate the utility of the assay using biological fluids and in vitro models of inflammation in the bovine endometrium.
2
Materials and methods ELISA Development Antibodies A sandwich ELISA was developed and validated for bovine IL-8 using antibodies described previously.18 Monoclonal mouse anti-sheep IL-8 was used as a capture antibody (MCA 1660; Abd Serotec, Kidlington, UK), polyclonal rabbit anti-sheep IL-8 (AHP425; Abd Serotec) as a detect antibody, and a polyclonal goat anti-rabbit horseradish peroxidase (HRP) as a tertiary antibody (P0448; DAKO UK Ltd, Ely, UK). The capture antibody was resuspended in 50 lL carbonate/bicarbonate buffer (0.2 M Na2CO3, 0.2 M NaHCO3, pH 9.4; C3041; Sigma, Gillingham, UK), and the detect antibody was resuspended in Dulbecco’s phosphate buffered saline (D-PBS; D8537; Sigma) containing 10% normal mouse serum (M5905; Sigma). The ELISA procedure (described below) was optimized by testing the following ranges of antibody concentrations: 1.43–5 lg/mL capture antibody, 0.146–0.5 lg/mL detect antibody, and 0.025–0.25 lg/mL tertiary antibody. Selection of blocking buffers To select an appropriate reagent for use as a blocking buffer, ELISA plate wells were treated in four duplicate wells with the following blocking reagents in DPBS: 5% bovine serum albumin (BSA; A7906; Sigma), 1% BSA (Sigma), 1% horse serum albumin (HSA; H1138; Sigma), 5% Tween-20 (P9416; Sigma), 5% skimmed milk (#9999; Cell Signaling, Leiden, the Netherlands), or 4% fish skin gelatin (FSG; G7765; Sigma). Developed ELISA procedure Wells of 96-well microtiter ELISA plates (675061; Greiner Bio-one, Stonehouse, UK) were coated with 50 lL per well of 2.5 lg/mL monoclonal mice antisheep IL-8 diluted in 50 lL carbonate/bicarbonate buffer, and the plates incubated for 18 hr at room temperature. The antibody solution was discarded, and the plates were washed three times with 150 lL per well 0.05% (vol./vol.) Tween-20 (P9416; Sigma) in D-PBS (D8537; Sigma) using a platewasher (LT3500; Labtech, Uckfield, East Sussex, UK). Wells were then blocked with 150 lL per well of 4% FSG blocking buffer in D-PBS for 1 hr, at room temperature, to block non-specific signals. A seven-point American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
BOVINE IL-8 ELISA
standard curve was generated by 1 in 2 serial dilution of recombinant bovine IL-8 (4000–62.5 pg/mL; RP0023B; Kingfisher Biotech, Inc. Saint Paul, MN, USA; supplied by 2B Scientific, Upper Heyford, UK) in 4% FSG block, or sample matrix if measuring IL8 in biological fluids, and added to duplicate wells, leaving duplicate analyte-free wells containing 50 lL per well matrix as blank wells. Test samples of mucus, serum, endometrial cell-culture supernatants or EVOC supernatants were loaded in duplicate wells using 50 lL per well. The plates were then incubated for 1.5 hr at room temperature, and then washed as described above. To each well, 50 lL of 0.145 lg/mL (a 1 in 700 dilution) polyclonal rabbit anti-sheep IL-8 in 4% FSG was added and the plates incubated for 2 hr at room temperature and then washed. The plates were then incubated with 50 lL of 0.042 lg/mL HRP polyclonal goat anti-rabbit in 4% FSG for 1 hr, in the dark. The plates were then washed and the reaction visualized by the addition of 50 lL/well colorimetric substrate 3,30 ,5,50 -tetramethylbenzidine (TMB; BD Biosciences Oxford, UK) and incubation in the dark for 15 min. The reaction was stopped by the addition of 50 lL/well 0.5 M sulfuric acid (Sigma), and absorbance was measured at 450 and 550 nm using a plate reader (Fluostar; BMG LABTECH, Ortenberg, Germany). Optical density (OD) values at 550 nm were subtracted from OD values read at 450 nm to correct for optical imperfections in the microtiter plate. The absorbance values for blank wells were subtracted from absorbance values obtained for the standards, and samples and the values for the duplicate wells were averaged. A seven-point standard curve was generated by plotting the average absorbance for each standard concentration on the Y-axis versus the corresponding bovine IL-8 concentration on the X-axis using the MARS Data Analysis software (BMG LABTECH). The sample IL-8 concentrations pg/mL were determined by interpolating from the absorbance value (Y-axis) to bovine IL-8 concentration (X-axis) using four-parameter logistic regression of the standard curve, generated by the MARS Data Analysis software. ELISA Validation Recovery and reproducibility To determine the efficiency of recovery of recombinant bovine IL-8, RPMI 1640 medium (Life Technologies, Paisley, UK) supplemented with 10% FBS was spiked with low, medium, or high (500, 1000, or American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
1500 pg/mL) concentrations of recombinant bovine IL-8 and assayed using the developed bovine IL-8 ELISA. To determine whether the matrix affected OD450 readings, seven-point standard curves were generated by diluting recombinant bovine IL-8 in the following matrices: Dulbecco’s Modified Eagle’s Medium (DMEM; supplemented with 10% FBS; Life Technologies); in vitro Maturation Medium (IVM; supplemented with 10% FBS; Life Technologies); 5% bovine serum albumin (BSA; Sigma); 4% FSG, RPMI 1640 (supplemented with 10% FBS; Life Technologies); and OPTI-MEM (Life Technologies) and assayed using the developed bovine IL-8 ELISA. The intra-assay coefficient of variation (CV) was determined from the average and standard deviation of the absorbance of 12 wells containing 500, 1000, or 2000 pg/mL recombinant bovine IL-8 assayed on the same microtitre plate. The interassay CV was determined from the average absorbance and standard deviation of 12 wells containing 500, 1000, or 2000 pg/mL concentrations of recombinant bovine IL-8 assayed on six independent microtitre plates. Specificity To determine cross-reactivity or interference from cytokines or chemokines commonly induced during the innate immune response, the following recombinant proteins were assayed using the bovine IL-8 ELISA: bovine IL-10 (50 ng/mL to 781.25 pg/mL; RP0379B; Kingfisher Biotech); bovine IL-6 (5 ng/mL to 78 pg/mL; 1861273; Thermo Scientific, Loughborough, UK); bovine IL-1b (2 ng/mL to 31.25 pg/mL; 1861270; Thermo Scientific); bovine IFNc (10 ng/mL to 156.25 pg/mL; 842590 R&D Systems, Abingdon, UK); bovine TNF (8 ng/mL to 125 pg/mL; TNF; 842630; R&D Systems) or bovine chemokine CCL2 (5 ng/mL to 78 pg/mL; RP00227; Kingfisher Biotech). Duplicate seven-point standard curves were generated by 1 in 2 serial dilution of each recombinant protein in 4% FSG and compared with a duplicate seven-point standard curve generated by 1 in 2 serial dilution of recombinant bovine IL-8 (4000 pg/ mL to 62.5 pg/mL) diluted in 4% FSG. Dilution linearity To evaluate linearity of dilution, seven-point standard curves were generated using recombinant bovine IL-8 encompassing the range of 32,000 pg/ mL to 62.5 pg/mL IL-8, with each standard subsequently diluted 1:10 (3200 pg/mL to 6.25 pg/mL), 3
CRONIN ET AL.
1:20 (1600 pg/mL to 3.13 pg/mL), or 1:40 (800 pg/ mL to 1.56 pg/mL) in 4% FSG. The dilutions were evaluated in the developed bovine IL-8 ELISA, and the observed values were plotted against the expected concentrations. To determine the upper limit of quantification, a standard curve was generated by serial dilution of recombinant bovine IL-8 (32,000 pg/mL to 31.25 pg/mL). Correlation Commercially available human IL-8 ELISA kits have previously been validated for measurement of IL-8 in bovine samples.14,15 So, to establish a correlation between the developed bovine IL-8 ELISA and the human IL-8 ELISA DuoSet kit (DY208; R&D Systems, Inc.), recombinant bovine IL-8 protein or supernatants from EVOCs treated for 24 hr with 1 lg/ mL ultrapure LPS (Invivogen, Toulouse, France) were assayed on the developed bovine IL-8 ELISA and the human IL-8 DuoSet ELISA, following the manufacturer’s instructions. Samples were assayed in duplicate, using both IL-8 ELISA systems on the same day. Demonstration of Utility of ELISA Isolation and culture of endometrial cells Endometrial tissue was dissected and processed as described previously to isolate and culture endometrial cells.12,19 Primary endometrial epithelial and stromal cells were cultured in 75 cm2 flasks (Greiner Bio-One, Gloucester, UK) in a humidified atmosphere of air with 5% CO2 at 37°C, with culture medium changed every 48 hr. Epithelial and stromal cell populations were distinguished by cell morphology, and the absence of immune cell contamination confirmed by the absence of CD45, as described previously. 9 Cells were seeded at 2 9 104 per well and treated in 24-well tissue culture plates (TPP, Trasadingen, Switzerland). Ex vivo organ culture Ex vivo organ culture of endometrium was generated as described previously.11 The EVOCs were cultured in six-well plates (TPP) with the epithelial surface uppermost in 3 mL per well of RPMI 1640 (supplemented with 10% FBS; Life Technologies). Plates were incubated in a humidified atmosphere with 5% CO2 in air at 37°C, as described previously.11 4
Isolation and culture of peripheral blood macrophages The isolation and culture of peripheral blood macrophages were performed as previously described,19 with minor modifications. Briefly, jugular venous blood was collected at slaughter into sterile 50-mL tubes (NUNC) containing 5 mL sodium citrate (3.8% w/v in HBSS; Sigma) supplemented with 50 ng/mL gentamicin (Sigma). Tubes were gently inverted 10 times to ensure sufficient mixing before transport to the laboratory at 4°C. Blood was divided into 25-mL volumes and diluted 1:1 with sterile HBSS supplemented with gentamicin (50 ng/mL; Sigma) and amphotericin B (2.5 lg/mL; Sigma). Aliquots of the diluted blood (16 mL) were then layered onto 12mL Ficoll-Paque PREMIUM (GE Healthcare Life Sciences, Buckinghamshire, UK) in a 50-mL tube. Tubes were then centrifuged at 500 g for 40 min at 4°C with the centrifuge brake turned off. Following centrifugation, the peripheral blood mononuclear cell layer was carefully removed and transferred to a sterile 15-ml tube. Cells were washed twice in HBSS supplemented with gentamicin, with centrifugation at 500 g for 15 min, resuspended in RPMI 1640 medium and transferred to 60-mm Petri dishes (Greiner) and incubated at 37°C for 30 min in a humidified atmosphere with 5% CO2 in air at 37°C. After incubation, the medium was removed and cells were washed twice with HBSS supplemented with gentamicin before the addition of RPMI 1640 medium with 10% FBS supplemented with gentamicin (50 ng/mL; Sigma), amphotericin B (Sigma). Cells were cultured for 6–8 days with medium changes every 48 hr to allow the maturation of monocytes. On the day of experimentation, macrophages were removed from the plate using Accutase cell detachment solution (Sigma) and seeded into 24-well plates at 1 9 106 cells. Treatment of Endometrial cells, EVOCs and macrophages To evaluate responses to PAMPs, epithelial cells, stromal cells, macrophages, and EVOCs were treated with LPS (10; 100; and 1000 ng/mL) for 24 hr and supernatants assayed using the developed bovine IL8 ELISA. Supernatants from stromal cells treated with 1 lg/mL ultrapure LPS from E. coli 0111:B4, 1 lg/mL Pam3CSK4 (Pam3CSK4; N-Palmitoyl-S(2,3-bis(palmitoyloxy)-(2RS)-propyl]-(R)-cysteinyl-(S)seryl-(S)-lysyl-(S)-lysyl-(S)-lysyl-(S)-lysine) PAM), 1 lg/mL FSL-1 (S-(2,3-bispalmitoyloxypropyl) CysGly-Asp-Pro-Lys-His-Pro-Lys-Ser-Phe, Pam2CGDPKH American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
BOVINE IL-8 ELISA
PKSF) (all InvivoGen, Toulouse, France) for 24 hr were also assayed using the developed bovine IL-8 ELISA. Finally, supernatants from macrophages treated with heat-killed E. coli (1 9 106, 1 9 107, or 1 9 108 CFU) for 24 hr were assayed using the developed bovine IL-8 ELISA. Bacterial culture and heat inactivation Cultures of E. coli (isolate MS482),8 collected from the uterus of postpartum cows with persistent uterine disease, were grown overnight until they reached log-phase growth in Luria–Bertani medium (Sigma) as described previously.11 The bacteria were diluted to 1 9 108 colony-forming units (CFU)/mL and centrifuged at 1500 g for 10 min at room temperature. The bacteria were then suspended in sterile PBS (Life Technologies Ltd) and heat-inactivated at 75°C for 45 min, and a sample of heat-inactivated bacteria was plated onto Luria–Bertani agar and incubated at 37°C for 48 hr to confirm complete inactivation. The bacteria were then centrifuged at 1500 g for 10 min at room temperature and washed three times in PBS, before suspension at a final concentration equivalent to 1 9 108 CFU/ mL in RPMI 1640 medium with 10% FBS supplemented with gentamicin (Sigma), amphotericin B (Sigma). Isolation of blood and mucus To further demonstrate utility of the IL-8 ELISA, mucus and serum samples were collected from Holstein cows in commercial dairy herds in Cheshire and Staffordshire that calved between May 1, 2011 and January 31, 2012. Each herd was visited weekly by a veterinarian (SP) and procedures conducted under the UK Animal Scientific Procedures Act (1986), with the approval of the UK Government Home Office (License number PPL 40/3478). Farm staff recorded whether calving was normal with no intervention or whether there was dystocia requiring manual intervention. One, three, and five weeks postpartum, 10 ml blood samples (no dystocia, n = 115 and dystocia, n = 63) were collected from the coccygeal vein blood vessels into an evacuated sterile tube without coagulant (Vacutainer; Becton), allowed to clot at 4°C for 6 hr prior to centrifugation at 1000 g for 10 min to harvest serum, which was stored at 20°C. In addition, mucus was collected from the vagina (no dystocia, n = 24 and dystocia, n = 12) using a clean, lubricated, gloved hand inserted through the vulva and into the vagina. The American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
mucus was placed in 30-mL sterile tubes and stored at 20°C. The mucus was processed prior to measurement of inflammatory mediators by ELISA as described previously.20 Briefly, 2 g mucus was placed in 10 mL of cytolyt solution (40% Methanol: 60% distilled water) and mixed with 0.1% Dithiothreitol (Sigma-Aldrich) until the mucus was disrupted, followed by centrifugation at 3000 g for 15 min, and collection of the supernatant. The concentrations of IL-8 were measured in serum and mucus with the developed bovine IL-8 assay.
(a)
(b)
(c)
Fig. 1 IL-8 ELISA development (a) Comparison of blank-corrected OD450 of IL-8 standard curves using the concentrations of tertiary antibody indicated. (b) Comparison of blank-corrected OD450 readings from analyte-free wells blocked with buffers indicated, using four duplicate wells for each buffer. (c) Comparison of blank-corrected OD450 of IL-8 standard curve diluted in the matrices indicated.
5
CRONIN ET AL.
(a) Table I A Comparison of Recombinant Bovine IL-8 (500, 1000, or 1500 pg/mL) Diluted in 4% FSG or RPMI (Supplemented with 10% FBS) in the Bovine IL-8 ELISA IL-8 diluted in 4% FSG (pg/mL)
Average % recovery of IL-8 in RPMI (n = 12)
Range % (n = 12)
500 1000 1500
128 132 144
112–129 120–142 130–169
Table II Cross-Reactivity or Interference by IL-6 for the Bovine IL-8 ELISA was Evaluated by Diluting Recombinant IL-6 in RPMI (Supplemented with 10% FBS), and Values were Interpolated from a Standard Curve Derived from Recombinant Bovine IL-8 Diluted in 4% FSG
(b)
Actual concentrations Observed concentrations of IL-6 (pg/mL) of IL-6 (pg/mL) % Cross-reactivity 5000 2500 1250 625 312.5 156.25 78
189.20 68.39 Below Y Below Y Below Y Below Y Below Y
range range range range range
3.78 2.74 0 0 0 0 0
Statistical Analysis Data are presented as arithmetic mean and standard error of mean (SEM) except where indicated. Data were analyzed by one-way ANOVA using GraphPad Prism 5 (La Jolla, CA, USA) with Dunnett’s post hoc test, and statistical significance described when P < 0.05.
Fig. 2 IL-8 ELISA validation (a) Dilution linearity using bovine IL-8 standards undiluted (diamonds); diluted 1 in 10 (squares); 1 in 20 (triangles); or 1 in 40 (circles) in 4% FSG n = 4 wells (b) Upper limit of quantification determined using a standard curve generated by serial dilution of recombinant bovine IL-8 (32,000 pg/mL–31.25 pg/mL) n = 4 wells.
Results and discussion The present study developed a sandwich ELISA method to detect bovine IL-8 in biological samples, based on previously described antibodies for rumi-
Table III Intra-assay CV for the Bovine IL-8 ELISA was Calculated from Values Derived from Recombinant Bovine IL-8 (Low, Medium, or High: 50, 1000, or 1500 pg/mL, Respectively; n = 12) Diluted in RPMI (Supplemented with 10% FBS) Assayed on the Same Microtitre Plate. Interassay CV was Calculated from Values Derived from Recombinant Bovine IL-8 (Low, Medium, or High: 50, 1000, or 2000 pg/mL, Respectively; n = 6) Diluted in RPMI (Supplemented with 10% FBS) Assayed on six Independent Microtitre Plates Interassay precision Bovine IL-8 (pg/mL) n Mean Standard deviation CV%
6
500 6 502 12 2.4
1000 6 998 32 3.2
Intra-assay precision 2000 6 2005 108 5.4
500 12 640 46 7.2
1000 12 1318 71 5.4
1500 12 2163 176 8.1
American Journal of Reproductive Immunology (2014) ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
BOVINE IL-8 ELISA
nant IL-8.18 Assay development involved iterative testing of a range of concentrations of the capture, detection, and tertiary antibodies; and testing a range of blocking buffers to reduce the background signal in the ELISA. Each antibody was selected on the basis of a relatively high dynamic range standard curve and a reciprocally low non-specific signal in the blank well (Fig. 1a). The concentrations selected were as follows: primary antibody 2.5 lg/mL; secondary antibody 0.145 lg/mL (a 1 in 700 dilution), and tertiary antibody 0.042 lg/mL. To minimize the non-specific signal, commonly used ELISA blocking buffers were assessed. In each assay, four duplicate wells were blocked with 5% BSA, 1% BSA, 1% HSA, 5% Tween-20, 5% skimmed milk or 4% FSG for 1 hr at room temperature. The ELISA was performed and values for non-specific signal generated in blank wells were determined at OD450. Blank wells blocked with 4% FSG exhibited the lowest OD450 reading (Fig. 1b). Thus, 4% FSG was used as the blocking buffer and reagent diluent for subsequent ELISA. The selection of specific concentrations for each antibody and the use of 4% FSG blocking buffer yielded standard curves with an effective range of 62.5–2000 pg/mL IL-8. (a)
Recombinant bovine IL-8 gave relatively higher OD450 readings when diluted in RPMI supplemented with 10% FBS compared to recombinant bovine IL8 standards diluted in 4% FSG (Table I). Therefore, the effect of diluting standards in different matrices on the OD450 readings was assessed and compared to recombinant bovine IL-8 standards diluted in 4% FSG. Recombinant bovine IL-8 standards (4000– 65 pg/mL) were diluted in DMEM, IVM, 5% BSA, 4% FSG, RPMI, or Opti-MEM. Relative blank-corrected OD450 readings varied for each concentration of recombinant bovine IL-8 depending on the matrix (Fig. 1c). To assess cross-reactivity or interference with the developed bovine IL-8 ELISA, a representative spectrum of cytokines and chemokines were assayed to reflect the typical innate immune response.1,9,10,12,21 The assay had little cross-reactivity with biologically meaningful concentrations of bovine IL-1b, IL-10, TNFa, IFNc, or CCL2 (all values below Y range). However, cross-reactivity or interference was exhibited with bovine IL-6, which represented 3.8% of the concentration assayed for a 5000 pg/mL sample of recombinant IL-6 (Table II). However, samples of
