Plasma interleukin-6 concentration in Standardbred racehorses determined by means of a novel validated ELISA

Jin-Wen Chen Department of Clinical Studies-New Bolton Center, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA 19348.

Search for other papers by Jin-Wen Chen in
Current site
Google Scholar
PubMed
Close
 PhD
,
Cornelius E. Uboh Department of Clinical Studies-New Bolton Center, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA 19348.
Pennsylvania Equine Toxicology and Research Laboratory, 220 E Rosedale Ave, West Chester, PA 19382.

Search for other papers by Cornelius E. Uboh in
Current site
Google Scholar
PubMed
Close
 PhD
,
Mary A. Robinson Department of Clinical Studies-New Bolton Center, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA 19348.
Pennsylvania Equine Toxicology and Research Laboratory, 220 E Rosedale Ave, West Chester, PA 19382.

Search for other papers by Mary A. Robinson in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
,
Zibin Jiang Department of Clinical Studies-New Bolton Center, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA 19348.

Search for other papers by Zibin Jiang in
Current site
Google Scholar
PubMed
Close
 PhD
, and
Lawrence R. Soma Department of Clinical Studies-New Bolton Center, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA 19348.

Search for other papers by Lawrence R. Soma in
Current site
Google Scholar
PubMed
Close
 VMD

Click on author name to view affiliation information

Abstract

OBJECTIVE To evaluate plasma interleukin 6 (IL-6) concentration in Standardbred racehorses by means of a novel ELISA following validation of the assay for use with equine plasma samples.

SAMPLE Plasma samples obtained from 25 Thoroughbreds for use in assay validation and from 319 Standardbred racehorses at rest 2 to 2.5 hours prior to warm-up and racing.

PROCEDURES A sandwich ELISA was developed with equine anti–IL-6 polyclonal antibody and the biotin-streptavidin chemical interaction to enhance sensitivity. The assay was validated for specificity, sensitivity, precision, and accuracy by use of both recombinant and endogenous proteins.

RESULTS For the assay, cross-reactivity with other human and equine cytokines was very low or absent. Serial dilution of plasma samples resulted in proportional decreases in reactivity, indicating high specificity of the method. Partial replacement of detection antibody with capture antibody or pretreatment of samples with capture antibody caused assay signals to significantly decrease by 55%. The inter- and intra-assay precisions were ≤ 13.6% and ≤ 9.3%, respectively; inter- and intra-assay accuracies were within ranges of ± 14.1% and ± 8.6%, respectively, at concentrations from 78 to 5,000 pg/mL, and the sensitivity was 18 pg/mL. Plasma IL-6 concentration varied widely among the 319 Standardbreds at rest (range, 0 to 193,630 pg/mL; mean, 6,153 pg/mL; median, 376 pg/mL).

CONCLUSIONS AND CLINICAL RELEVANCE This ELISA method proved suitable for quantification of IL-6 concentration in equine plasma samples. Plasma IL-6 concentration was high (> 10,000 pg/mL) in 9.1% of the Standardbred racehorses, which warrants further investigation.

Abstract

OBJECTIVE To evaluate plasma interleukin 6 (IL-6) concentration in Standardbred racehorses by means of a novel ELISA following validation of the assay for use with equine plasma samples.

SAMPLE Plasma samples obtained from 25 Thoroughbreds for use in assay validation and from 319 Standardbred racehorses at rest 2 to 2.5 hours prior to warm-up and racing.

PROCEDURES A sandwich ELISA was developed with equine anti–IL-6 polyclonal antibody and the biotin-streptavidin chemical interaction to enhance sensitivity. The assay was validated for specificity, sensitivity, precision, and accuracy by use of both recombinant and endogenous proteins.

RESULTS For the assay, cross-reactivity with other human and equine cytokines was very low or absent. Serial dilution of plasma samples resulted in proportional decreases in reactivity, indicating high specificity of the method. Partial replacement of detection antibody with capture antibody or pretreatment of samples with capture antibody caused assay signals to significantly decrease by 55%. The inter- and intra-assay precisions were ≤ 13.6% and ≤ 9.3%, respectively; inter- and intra-assay accuracies were within ranges of ± 14.1% and ± 8.6%, respectively, at concentrations from 78 to 5,000 pg/mL, and the sensitivity was 18 pg/mL. Plasma IL-6 concentration varied widely among the 319 Standardbreds at rest (range, 0 to 193,630 pg/mL; mean, 6,153 pg/mL; median, 376 pg/mL).

CONCLUSIONS AND CLINICAL RELEVANCE This ELISA method proved suitable for quantification of IL-6 concentration in equine plasma samples. Plasma IL-6 concentration was high (> 10,000 pg/mL) in 9.1% of the Standardbred racehorses, which warrants further investigation.

The use of an ELISA is a reliable approach to detecting and quantifying proteins of interest, such as pro- and anti-inflammatory factors. But there are few ELISA methods validated for determining plasma concentrations of inflammatory factors in horses, and this limitation delays progress in the investigation of equine physiologic and pathophysiologic immune responses.

Because of their importance in various medical conditions, immune inflammatory factors, such as cytokines, are widely studied in many species. Cytokines are expressed in a variety of cells, especially when cells are under stress in situations such as invasion by foreign substances (eg, viruses and bacteria), allergic reactions, and trauma or injury.1 Interleukin-6 acts as both a pro- and anti-inflammatory cytokine. It is also considered a myokine—a cytokine produced from muscle—and its plasma concentration is increased in response to muscle contraction.2 As a proinflammatory cytokine, IL-6 is secreted by T cells and macrophages to stimulate an immune response during infection and after trauma, especially burns or other tissue damages leading to inflammation. On the other hand, as an anti-inflammatory cytokine, IL-6 is secreted by myocytes to protect muscle tissue from damage during prolonged or intense exercise; plasma concentration of IL-6 is significantly elevated with exercise, and this increase precedes the appearance of other circulating cytokines.2,3 Therefore, IL-6 acts differently when it is secreted from muscle rather than from T cells or macrophages. However, systemic concentrations of IL-6 in racehorses have not been extensively studied, to our knowledge.

The purpose of the study reported here was to develop an indirect ELISA incorporating antigen-specific antibody and biotin-streptavidin chemical interaction to maximize specificity and sensitivity and validate that assay for quantification of IL-6 concentration in equine plasma samples. Interleukin-6 was chosen as the focus of the study because it is a potentially important inflammatory biomarker in racehorses at rest (ie, before a race). The ELISA was then used to measure plasma IL-6 concentration in a large number of Standardbreds at rest.

Materials and Methods

Antibodies and recombinant proteins

The same goat-derived antigen affinity-purified PAb was used as both the CAb and DAb because it contained multiple epitopes for antigen binding. Equine PAb, its biotinylated conjugate, and recombinant protein for IL-6 were commercially available products.a The PAb and its biotin conjugate were derived from the same recombinant equine IL-6 ([Phe 26-Met 208]) that was expressed from an equine gene.4

Source of test samples and quality control plasma samples

Plasma samples were derived from whole blood samples (10 mL each) collected from 25 Thoroughbred horses in the research herd at New Bolton Center and from 319 Standardbred horses 2 to 2.5 hours prior to racing at Harrah's Philadelphia and the Meadows Racetrack and Casino. The horses in the research herd were no longer actively racing. All horses were in good health, and routine foot and dental care, vaccination, and deworming were performed on a scheduled basis. Written consent was obtained from the horse owner for each of the client-owned horses for blood sample collection by veterinarians of the Pennsylvania Harness Racing Commission. Animal treatment and blood sample collection procedures were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania.

The whole blood sample from each horse (Thoroughbreds and Standardbreds) was collected in heparin-containing or NaF + oxalate–containing tubes and centrifuged (1,500 × g for 10 minutes) immediately after collection to harvest plasma for either short-term (< 7 days) storage at −20°C or long-term storage at −70°C. For the purpose of assay validation, the assay was assigned to have 3 quality control plasma samples with high concentrations of IL-6 (> 5,000 pg/mL) as described previously,5,6 and the samples were prepared at 7 concentrations (5,000, 2,500, 1,250, 625, 312, 156, and 78 pg/mL). Three ANP samples were used as negative controls to validate the assay method.

Preparation of reference proteins and selection of calibration buffers

To ensure the accuracy of the immunoassay7 and to validate the range of detection, commercial recombinant equine IL-6 protein was requantified to identify the actual concentration by use of a high-sensitivity protein 250 labeling kitb analyzed on an automated bioanalyzerb according to the manufacturer's protocol. Briefly, both protein ladder and samples were labeled with fluorescent dye in a labeling reaction. Fluorescent dye was reconstituted in dimethyl sulfoxide, and 0.5 μL of reconstituted dye solution was mixed with 5 μL of protein ladder or recombinant equine IL-6 followed by incubation for 30 minutes on ice. The labeling reaction was then terminated by addition of 0.5 μL of ethanolamine and incubation for 10 minutes on ice before performance of microelectrophoresis to separate the ladder and the samples. Finally, the molecular weight and the concentration of the protein fraction of interest were determined.

Selection of the calibration buffers

To acquire the optimal recovery conditions for recombinant equine IL-6, it was serially diluted in various solutionsc,d (1% bovine serum albumin, 10% fetal bovine serum, 1% casein, ANP, polyethylene glycol-treated ANP, PBS solution, and 0.05% Tween 20) to determine the recovery of the spiked protein. The optical density and Pearson correlation coefficient (r2) of the calibration curve were determined. Polyethylene glycol-treated ANP was obtained by treating ANP with polyethylene glycol 6000. Briefly, polyethylene glycol 6000 was added to ANP at a final concentration of 25%; the mixture was vortexed for 1 minute, then placed on ice for 15 minutes. The mixture was centrifuged at 3000 × g for 10 minutes at 4°C, and the supernatant underwent 2-fold dilution in PBS solution; the diluted supernatant was subsequently used as one of the calibrator diluents.

Immunoassay procedure

The immunoassay method was developed to determine the concentration of IL-6 in equine plasma samples by sandwich ELISA with biotin-streptavidin chemical interaction to increase sensitivity. The method was developed with goat anti-equine IL-6 PAb as the CAb, biotinylated goat anti-equine IL-6 PAb as the DAb, and recombinant equine IL-6 as the calibrator protein. Quantification of IL-6 concentration in all equine plasma samples was performed in duplicate, and all procedures were conducted at room temperature (approx 21°C). Briefly, CAb was coated onto a 96-well enzyme immunoassay-radioimmunoassay platee for 2 hours or overnight (approx 15 hours) followed by vigorous washing with 0.05% Tween 20 in PBS solution to remove unbound protein. Nonspecific binding was prevented by incubation of the coated wells with 1% bovine serum albumin for 1 hour followed by vigorous washing. A 100-μL aliquot of each sample or standard was loaded into each well, and 7 selected concentrations of recombinant equine IL-6 (5,000, 2,500, 1,250, 625, 312, 156, and 78 pg/mL) were applied to the plate and incubated for 1 hour to ensure sufficient binding of antigen to the antibody, followed by vigorous washing to remove unbound materials. A DAb was applied to bind the antigen in the wells followed by incubation for 1 hour and then vigorous washing. Streptavidin–horseradish peroxidasea was applied to form the antibody-analyte-antibody–horseradish peroxidase complex, followed by incubation for 20 minutes and then vigorous washing. Finally, the optical density was determined at 450 and 540 nm by application of substrate mixturef in a validated microplate reader.g The concentration was automatically calculated as sOD (derived by subtraction of optical density determined at 540 nm [background optical density] from the optical density determined at 450 nm) on the basis of a 4-parameter logistic fit-curve provided by validated microplate management software.h All standards, quality control plasma samples, and unknown samples were analyzed in duplicate unless stated otherwise.

Optimization of concentrations of the CAb and DAb

Different concentrations of CAb were evaluated along with various concentrations of the DAb, and optimal concentration pairs were determined on the basis of numerous variables including sOD, signal-to-noise ratio, linearity of the calibration curve, and background signal. For a standard curve, the background signal should not be > 0.10 and the correlation coefficient (r2) should not be < 0.98.

Method validation procedures

Validation of the method included a 3-day prestudy sample analysis as previously described.6 A 1-day validation study generated 3 standard curves along with 3 quality control plasma samples. By means of serial 2-fold dilutions, each standard curve incorporated a 7-point calibration, and the 3 quality control plasma samples were prepared at 7 concentrations as described. Three ANP concentrations were analyzed along with the standard curves and quality control plasma samples. The 3-day prestudy validation6 resulted in a total of 9 calibration curves and 126 data points for the quality control plasma samples. In another analysis, 10 replicates of each of the 7 concentrations of the standard samples or quality control plasma samples were analyzed to evaluate intra-assay variability. For the calibrator and quality control plasma sample concentrations, the acceptance criteria were that percentage relative SD and percentage relative error of the back-calculated values must be ≤ 15% and within a range of ± 15%, respectively. Validation of sample analysis consisted of generating a calibration curve and 3 quality control plasma samples with numerous unknown samples for assay during each analysis.

The acceptance criteria were that the quality control plasma sample values must be within a range of either ± 2 SD or ± 10% of the established mean from the prestudy validation, whichever range was greater. If in any analysis the quality control plasma sample values were not within the required range, the assay was marked as failed, and samples were reanalyzed.

Assay specificity

Specificity of the immunoassay was determined by testing for cross-reactivity of the recombinant proteins other than the antigen of interest at 3 concentrations (1, 10, and 100 ng/mL). The recombinant proteinsa,i were derived from equine cDNA sequences (IL-1β, IL-2, IL-4, IL-5, IL-8, IL-10, IL-12, IL-13, IL-15, tumor necrosis factor-α, interferon-γ, and vascular endothelial growth factor A) or from human cDNA sequences (bone morphogenetic protein 1; transforming growth factors β1 β2, and β3; platelet-derived growth factor A; hepatocyte growth factor; and toll-like receptor 1). The percentage cross-reactivity was determined as the ratio of the observed and the actual concentrations used.

To choose the most appropriate sample diluents, equine plasma samples were diluted in various solutions (PBS solution, 0.05% Tween 20, 0.1% Triton X-100, 1% bovine serum albumin, 1% casein, and 10% fetal bovine serum). Each plasma sample was serially diluted (to 128-fold dilution) to assess the change in sOD in response to the concentration of the IL-6 in the diluted samples. The most appropriate diluent with good linearity (≥ 98%) and optimal sOD was used for dilution of the equine plasma samples to acquire the true optical density of each analyte in the sample.

Assay specificity was determined by performing a competitive binding assay. The competitive binding assay used UAb to partially or completely replace DAb or used plasma samples pretreated with UAb; reduced signals were expected to indicate the specificity of the DAb in both instances. Briefly, UAb was mixed with DAb at equal concentration but the total antibody concentration remained constant; in this assay, UAb would compete for binding with epitopes of the antigen, thereby reducing the detection signal. Alternatively, equine plasma samples were pretreated with a concentration of UAb equal to that of the DAb followed by incubation for 1 hour at room temperature, and the reactivity was expected to decrease if the antibody was specific to the antigen.

Epitope analysis was also conducted in a direct antibody competition ELISA.8 Briefly, 96-well microtiter plates were coated with targeted analytes or with nontargeted analytes (eg, IL-2a and IL-4a) at concentrations of 5, 10, 20, 40, 80, 160, or 320 ng/mL for 2 hours at room temperature followed by blocking with 1% bovine serum albumin. The DAb was mixed with UAb at different concentration ratios (1:0, 1:1, 1:2, 1:3, and 0:1), and binding of DAb was revealed by incubation with streptavidin–horseradish peroxidase.

To assess nonspecific background activity, the assay was performed with incubation of equine plasma samples in wells that were coated with PBS solution instead of antibody or coated with nontargeted PAb (eg, anti–IL-2a and anti–IL-4a antibodies) at 3 concentrations (500, 1,000, and 2,000 ng/mL). The assay was also conducted without DAb or with a DAb other than that for IL-6 (ie, anti–IL-2 or anti–IL-4 antibody) at 3 concentrations (200, 400, and 800 ng/mL) to detect any possible nonspecific reaction of the DAb with equine plasma. By use of the selected reagents for this study, the background response did not exceed 5% of the maximal sOD.

Linearity and sensitivity of the assay

The recombinant equine IL-6 was reconstituted in PBS solution containing 1% bovine serum albumin, and the protein was serially diluted (2-fold dilutions) for up to 20 dilutions to determine the detection range of the reference standard. A 7-point calibration curve was generated by serial dilutions of the recombinant equine IL-6 and was used for quantification of recombinant or native equine IL-6 in samples with unknown concentrations of equine IL-6. The mean value was acquired from 10 independent assays, and assay linearity was determined from the calculated correlation coefficient (r2). The standard protein and quality control plasma samples were diluted until no signal was detected and the limit of detection was determined to ascertain the sensitivity of the assay. The limit of detection of the assay was calculated by adding 2 SD to the mean optical density value of 10 zero standard replicates (ie, replicates with no equine IL-6). Low and high ranges of the standard curve were determined.

In vitro stimulation of IL-6

To demonstrate the applicability of the assay to an in vitro situation, 2 whole blood samples (10 mL each) from 24 of the 25 Thoroughbreds from which plasma samples had been obtained previously were treated with lipopolysaccharidec (an endotoxin) or calcium ionophore A23187c to trigger production of cytokines and to assess the effectiveness of the assay. Lipopolysaccharide treatment was performed as previously described.9,10 Briefly, 10 mL of fresh whole blood from each of 12 horses was collected in a 10-mL glass tube coated with heparin and was incubated with lipopolysaccharide (1 μg/mL) in a shaker at 37°C for 24 hours. Another 10-mL sample of whole blood was collected from these horses and used as untreated controls. A sample of fresh whole blood (10 mL) was also collected from another 12 horses. Each sample was treated with calcium ionophore A23187 at a final concentration of 10 μM, and the mixture was incubated at 37°C for 2 hours.11 Another 10-mL sample of whole blood was collected from these horses and used as untreated controls. All treated and untreated blood samples were centrifuged (1,600 × g for 10 minutes) to harvest plasma for quantification of IL-6 concentration.

Extracorporeal shock wave treatment

Approximately 1 month after the experiments to investigate in vitro stimulation of IL-6, 11 of the 25 Thoroughbreds from which plasma and whole blood samples had been obtained previously underwent extracorporeal shock wave treatment on the distal portion of the right forelimb to demonstrate the applicability of the assay to an in vivo situation. The treatment was provided by an electrohydraulic shock wave generator with a focused applicatorj; all horses received the same treatment (total of 2,700 pulses at an energy flux density of 0.55 mJ/mm2 and frequency of 3 Hz12,13). Duration of treatment was approximately 18 minutes to obtain a total amount of energy applied during each treatment of 48.56 J. A blood sample (10 mL) was collected from the horses daily for a week before treatment and at 2 hours after treatment; plasma was isolated for analysis of IL-6 concentration.

Measurement of plasma IL-6 concentration in Standardbreds

Plasma IL-6 concentration was analyzed in a relatively large population of Standardbreds (n = 319) with the validated ELISA. The concentration range was determined in the Standardbred study population.

Data and statistical analyses

A microplate readerg and associated softwareh were used to acquire raw and analyzed data. Optical density values were obtained at 450 nm (OD450 [specific signal]) and 540 nm (OD540 [background signal]). Results were expressed as either mean sOD or mean concentration (pg/mL) of duplicate samples. Inter-or intra-assay percentage relative SDs were assessed by comparing sODs obtained for standard protein or quality control plasma samples on 3 days or sODs from the replicates of standards or quality control plasma samples on 1 plate. Calibration curves were generated by 4-parameter logistic fit-curve analyses of the antigen concentration as the independent variable and the antibody binding as the dependent variable. Statistical analyses were performed by means of a paired Student t test and by calculating the Pearson correlation coefficient (r2). Significance was indicated when the value of P was < 0.05.

Results

A dual-antibody sandwich ELISA was developed and used in the analysis of IL-6 in equine plasma samples. The optimal concentration pairs of the CAb and DAb were determined (CAb, 250 ng/mL; DAb, 150 ng/mL). The linearity of the standard curves was acceptable according to the r2 value (≥ 0.998) and signal intensity.

Matrix effect on recovery of calibrator protein

The result of evaluating various diluents indicated that the recovery rate was at least 88% when bovine serum albumin or polyethylene glycol–treated ANP was used, but decreased with other diluents (Figure 1). The results of the serial dilution of the recombinant protein in various diluents indicated that 1% BSA or polyethylene glycol–treated ANP was associated with the best linearity (with r2 of near unity), whereas other diluents had poor recovery with lower linearity. The ANP had the least linearity (0.63) and the lowest signal response. However, when ANP was treated with 25% polyethylene glycol 6000 to remove abundant proteins, the supernatant had a better linearity and signal, similar to findings for bovine serum albumin.

Figure 1—
Figure 1—

Recovery of recombinant equine IL-6 calibrator in various diluents in a study to develop a sandwich ELISA and validate that assay for quantification of IL-6 concentration in equine plasma samples. Seven diluents (polyethylene glycol-treated ANP [PEG-ANP], 1% bovine serum albumin [BSA], PBS solution [PBSS], 0.05% Tween 20, 10% fetal bovine serum [FBS], 1% casein, and ANP) were evaluated for recovery of the calibrator protein. The optical density (OD) was determined at 450 nm in a validated microplate reader. Those diluents that performed best (PEG-ANP or 1% BSA) on the basis of recovery, linearity, and precision were selected for quantification of the cytokine in plasma samples with unknown IL-6 concentrations.

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Calibration curve

The assay linearity was assessed by generating a calibration curve by use of serial dilutions of recombinant equine IL-6 with 1% BSA. The relationship between the calibrator concentration and sOD had an r2 value > 0.998, which indicated a very high linearity for the detection of the inflammatory factor (Figure 2). The sOD range for the standard curve was from 1.8 to 0.1.

Figure 2—
Figure 2—

Representative standard curve for IL-6 concentration in recombinant equine IL-6 protein solutions. Seven concentrations (0, 78, 156, 312, 625, 1,250, 2,500, and 5,000 pg/mL) of recombinant equine IL-6 protein were used. The optical density [arbitrary units {AU}] was determined at 450 and 540 nm by applying the substrate mixture in a validated microplate reader. The concentration was automatically calculated as sOD (derived by subtraction of optical density determined at 540 nm [background optical density] from the optical density determined at 450 nm) on the basis of a 4-parameter logistic fit-curve provided by validated microplate management software.h Equine IL-6 was quantifiable at concentrations between 78 and 5,000 pg/mL (Pearson correlation coefficient [r2] of this calibration curve was 1.000).

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Matrix effect on recovery of endogenous proteins

For a given horse, quantification of plasma IL-6 concentration varied considerably depending on the dilution buffer used (Figure 3). Hence, sOD was highest in equine plasma samples diluted with 0.05% Tween 20, followed by the sOD in samples diluted with Triton X-100 or PBS solution. Furthermore, 1% BSA and 1% casein were unsuitable for native IL-6 detection because the detection signal and linearity were unacceptable (r2 ≤ 0.49). The most acceptable detection linearity of the analyte and the optimized reactivity signal were achieved with 0.05% Tween 20 in all equine plasma samples (n = 10), suggesting that the detergent was the most suitable of the diluents (r2 ≥ 0.98). However, IL-6 was not detectable in 1 of the 7 samples, even when the sample was diluted with Tween 20 (Figure 4).

Figure 3—
Figure 3—

Effect of sample diluents on detection of IL-6 in equine plasma samples. Plasma samples obtained from 10 Thoroughbreds were diluted in 6 diluents (0.05% Tween 20, 0.1% Triton, PBSS, 1% BSA, 10% FBS, or 1% casein) to assess recovery of endogenous IL-6 and determine the appropriate diluent for accurate IL-6 detection. Tween 20 was chosen as the best diluent for use in the assay. See Figures 1 and 2 for key.

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Figure 4—
Figure 4—

Linearity of native IL-6 concentrations in equine plasma samples following serial dilutions. Seven equine plasma samples (represented by different symbols) obtained from 7 Thoroughbreds were serially diluted in 0.05% Tween 20, and the ELISA was performed to determine change in IL-6–associated signal intensity. See Figure 2 for key.

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Interhorse variability attributable to the matrix effect was observed. Some plasma samples with high concentrations of cytokine generated reasonable detection signals in multiple buffers, whereas others were restricted to dilution with Tween-20 for linear detection, suggesting that greater matrix interference was present in those samples. Data regarding the effect of various diluents on recovery of the analyte indicated that 0.05% Tween-20 was the most suitable buffer for equine plasma samples on the basis of the linearity (r2 ≥ 0.98) and the detection signal obtained (Figure 3). An alternative nonionic detergent, Triton X-100, had a lower detection signal and correlation coefficient (r2 = 0.77), and PBS solution without any detergent also had a lower detection signal and linearity (r2 = 0.71), compared with findings for Tween 20. For plasma sample dilution, 1% bovine serum albumin had both low detection signal and low correlation coefficient (r2 = 0.49).

Serial dilution of equine plasma samples

All equine plasma samples had linear reduction in reactivity when they were serially diluted (Figure 4). In some samples, IL-6 concentration apparently did not change until the 8-fold dilution was reached. These results indicated saturation of the analyte and that appropriate dilution was needed for accurate quantification of analyte-rich samples.

In the analyte-rich samples, sOD was significantly higher when the plasma samples were diluted in Tween 20, compared with findings for undiluted samples, indicating that the detergent can dissociate the matrix complex to expose the epitope of the analyte for antibody binding. These results also indicated that the antigens present in these samples were saturated, according to the extent of the antibody binding, and exceeded the upper limit of the IL-6 calibration curve; therefore, the samples needed to be diluted for accurate quantification. Linearity of signal through serial dilutions was much higher (r2 ≥ 0.998) when the plasma samples with high concentrations of IL-6 were diluted to sODs of approximately 1.5 to 2.0 and the diluted samples were used as the starting point of serial dilution. In this manner, a plasma sample with good linearity could be used as an alternative reference standard.

Enzyme-linked immunosorbent assay specificity

The aim of the study was to develop and validate a new and sensitive ELISA method for detection and quantification of IL-6 in equine plasma samples. The rationale for use of the same PAb as both the CAb and DAb was that the antigen-derived PAb had multiple binding sites sufficient for capture and detection. The readily available PAb could effectively capture IL-6 in equine plasma samples, and its biotinylated conjugate could effectively detect the captured immunogen.

The specificity of a dual-antibody sandwich ELISA for IL-6 detection in equine plasma samples was assessed. Evaluation of 19 recombinant proteins revealed that there was no or extremely low cross-reactivity (< 2%) with anti–IL-6 antibody among the proteins including equine IL-1β, IL-2, IL-4, IL-5, IL-8, IL-10, IL-13, IL-15, tumor necrosis factor-α, interferon-γ, and vascular endothelial growth factor A each at concentrations of 1 to 100 ng/mL (Table 1). Similarly low cross-reactivity with anti–IL-6 antibody was evident for human bone morphogenetic protein 1; transforming growth factors β1 β2, and β3; hepatocyte growth factor; platelet-derived growth factor-A; and toll-like receptor 1 each at concentrations of 1 to 100 ng/mL. These data indicated that the immunoassay was highly specific for quantitation of IL-6 in equine plasma samples and that the immunopeptides of IL-6 used to generate the PAb did not share common epitope sequences with one another.

Table 1—

Assessment of cross-reactivity of the dual-antibody sandwich ELISA developed to determine IL-6 concentration in equine plasma samples.

 Concentration (ng/mL)
Analyte1.0010.00100.00
IL-1β0.390.160.11
IL-20.240.100.07
IL-40.600.140.12
IL-50.170.100.08
IL-8000
IL-100.070.040.01
IL-120.200.150.12
IL-130.360.290.21
IL-150.020.030.02
Tumor necrosis factor-α00.050
Interferon-γ0.150.010.02
Vascular endothelial growth factor A0.040.020.01
Bone morphogenetic protein 10.160.230.19
Transforming growth factor-β10.370.200.16
Transforming growth factor-β21.911.601.32
Transforming growth factor-β30.500.360.31
Platelet-derived growth factor A0.830.660.72
Hepatocyte growth factor1.240.950.64
Toll-like receptor 10.130.020.07

Data are presented as percentage ratios of observed and actual concentrations. Specificity of the immunoassay was determined by testing for cross-reactivity of recombinant proteins other than the antigen of interest at 3 concentrations (1, 10, and 100 ng/mL). The recombinant proteins were derived from equine cDNA sequences (IL-1β, IL-2, IL-4, IL-5, IL-8, IL-10, IL-12, IL-13, IL-15, tumor necrosis factor-α, interferon-γ, and vascular endothelial growth factor A) or from human cDNA sequences (bone morphogenetic protein 1; transforming growth factors β1, β2, and β3; platelet-derived growth factor A; hepatocyte growth factor; and toll-like receptor 1). The percentage cross-reactivity was determined as the ratio of the observed and the actual concentrations used. Evaluation of the 19 recombinant proteins revealed that there was no or extremely low cross-reactivity (< 2%) with anti–IL-6 antibody among the proteins. These data indicated that the immunoassay was highly specific for quantitation of IL-6 in equine plasma samples and that the immunopeptides of IL-6 used to generate the PAb did not share common epitope sequences with one another.

In the assay testing, there was a complete loss of signal in all equine plasma and reference protein samples when CAb was absent or an irrelevant solution or antibody was coated on the microplate wells. For instance, when the microplate wells were coated with PBS solution instead of the IL-6 CAb, no signal was observed in both the reference protein and equine plasma samples, although the rest of the procedure remained the same. The analyte of interest was also not detected when an antibody with another known specificity (eg, anti–IL-2 or anti–IL-4 antibody) was coated at various concentrations in the microplate wells. In addition, when the specific DAb was absent or an irrelevant DAb was applied, no signal was detected. These results indicated there was failure to capture the antigen on the plate when the specific antibody was omitted or when an irrelevant antibody was applied. Therefore, it was evident that the specific antibody was necessary for detection of the antigen of interest. These findings further illustrated the specificity of the antibody used to capture native or recombinant IL-6. Furthermore, not only was there no signal detection when the antigen-specific antibody was absent or when an irrelevant antibody was used, but also no signal was detected when the irrelevant recombinant proteins were applied. All available results confirmed the specificity of the ELISA method.

The competitive binding assay revealed a decrease in sOD when UAb was applied to the microplate wells instead of DAb. However, when samples were pretreated with a concentration of UAb equal to that of the DAb, the detection signal decreased by 55% (P < 0.001), compared with findings from the same samples but without pretreatment (controls; Figure 5). Similarly, when half of the DAb was replaced with UAb, the signal decreased by 62% (P < 0.001), compared with findings for the controls. When the recombinant protein was coated to the microplate wells and the aforementioned mixture of UAb and DAb was applied, the signal decreased by 57%, compared with findings from the same samples but with DAb alone (controls; data not shown), indicating the specificity of the antibody in the detection of the corresponding antigen. Taken together, these results indicated that the antibody was specific for the analyte of interest; moreover, the sOD reduction with competitive binding strongly suggested that the immunoreactivity was specific to the analyte under analysis.

Figure 5—
Figure 5—

Results (mean ± SD) of a competitive binding assay to determine the specificity of the ELISA for IL-6. Plasma samples were obtained from 25 Thoroughbreds. The competitive binding assay used plasma samples pretreated with UAb (Pretreat) or used UAb to partially replace the DAb (Mixture); reduced signals (compared with findings for untreated plasma samples [controls]) were expected to indicate the specificity of the DAb. After pretreatment of equine plasma samples with a quantity of UAb equal to that of the DAb, the signal decreased significantly. In addition, when equal quantities of UAb and DAb were mixed and then applied to bind the antigen of interest, the signal decreased significantly. *For a given paired data set, value is significantly (P < 0.001) decreased, compared with the control value. See Figure 2 for key.

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Linearity, precision, accuracy, and sensitivity of the ELISA

Numerous variables, including linearity, accuracy, precision, and sensitivity of the ELISA, were studied. In the 3-day validation trial, the pooled results from 4 assays indicated a linear calibration curve with r2 of near unity (Figure 2). Expressed as percentage relative SD, interassay precision of calibration standards and quality control plasma samples was 13.6%, whereas intra-assay precision was 9.3% (Table 2). Expressed as percentage relative error, the interassay accuracy of calibration standards and quality control plasma samples was 14.1%, whereas the intra-assay accuracy was 9.7%. The sensitivity of the assay was 18 pg/mL. These results indicated that the day-to-day standard curves generated were stable at picogram-level sensitivity and that the ELISA method was highly reproducible.

Table 2—

Intra- and interassay precision (percentage relative SD) and accuracy (percentage relative error) of the dual-antibody sandwich ELISA developed to determine IL-6 concentration in equine plasma samples.

 Intra-assay dataInterassay data
Predicted concentration (pg/mL)Observed concentration (pg/mL)Percentage relative SD (%)Percentage relative error (%)Observed concentration (pg/mL)Percentage relative SD (%)Percentage relative error (%)
78839.36.48913.614.1
1561426.4–9.017311.710.9
3122897.2–7.42729.1–12.8
6256795.88.66908.410.4
1,2501,1863.7–5.41,3558.98.4
2,5002,6374.65.52,3147.7–7.4
5,0005,4587.19.75,5737.211.5

Recombinant equine IL-6 protein and quality control samples were prepared at 7 concentrations. The data combined the results of recombinant and quality control samples, and each concentration had 10 replicates of each recombinant protein or quality control sample.

Assay applicability

Assay applicability was assessed by use of the validated ELISA to measure changes in native IL-6 concentration following in vitro and in vivo perturbation of IL-6 production (Figure 6). In whole blood samples (n = 12), in vitro lipopolysaccharide treatment resulted in a > 2-fold increase (P < 0.001) in IL-6 concentration, compared with findings in untreated samples. In contrast, IL-6 concentration in whole blood samples (n = 12) did not change significantly following treatment with calcium ionophore. Plasma samples were obtained from 11 Thoroughbreds before and 2 hours after extracorporeal shock wave treatment. This extracorporeal shock wave treatment resulted in a significant (P < 0.01) decrease in plasma IL-6 concentration. However, plasma IL-6 concentration did not change significantly during a 1-week pretreatment period (data not shown). These results indicated that the assay method developed could be used to detect changes in plasma IL-6 concentration.

Figure 6—
Figure 6—

Mean ± SD equine plasma IL-6 concentrations determined by use of the IL-6 ELISA in samples obtained after exposure of horses to extracorporeal shock wave treatment (ESWT) or exposure of equine whole blood samples to lipopolysaccharide (LPS) or calcium ionophore A23187 (CI). Eleven Thoroughbreds each underwent ESWT (2,700 pulses at an energy flux density of 0.55 mJ/mm2 and frequency of 3 Hz) on the distal portion of the right forelimb. A blood sample (10 mL) was collected from the horses before and at 2 hours after treatment; plasma was isolated for analysis of IL-6 concentration. In other experiments, 2 whole blood samples (10 mL each) were collected from 12 Thoroughbreds; each blood sample was treated with LPS or remained untreated (controls). Two whole blood samples (10 mL each) were also collected from another 12 Thoroughbreds; each blood sample was treated with CI or remained untreated (controls). All treated and untreated blood samples were centrifuged to harvest plasma for quantification of IL-6 concentration. Compared with control findings, plasma IL-6 concentration was decreased by ESWT; in vitro, IL-6 concentration was increased after LPS treatment but unaffected by CI treatment. *For a given paired data set, value is significantly (P < 0.01) decreased, compared with the control value. †For a given paired data set, value is significantly (P < 0.001) increased, compared with the control value.

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Plasma IL-6 concentration in Standardbred racehorses

The concentration of IL-6 was measured in plasma samples from 319 Standardbred race horses (Figure 7). The results indicated that the plasma IL-6 concentration varied considerably among horses, from undetectable to 193,630 pg/mL with a mean value of 6,153 pg/mL and a median value of 376 pg/mL. No reactivity was evident in some horses (113/319), and fewer (29/319) had approximately 30-fold increases in plasma IL-6 concentrations, compared with the mean value. Samples with sOD values outside the range of the calibration curve required use of appropriate dilution factors to allow for accurate quantification of plasma IL-6 concentration. It should be stated that while endogenous IL-6 was detected in many of the plasma samples from both the Thoroughbreds from the research herd and the Standardbred racehorses, 120 of 344 (35%) samples did not contain any detectable IL-6.

Figure 7—
Figure 7—

Distribution of plasma IL-6 concentration determined by use of the IL-6 ELISA in samples obtained from 319 Standardbred racehorses at rest prior to warm-up and racing. Among the horses, plasma IL-6 concentration varied greatly; more than a third (113 [35.4%]) of the horses had no detectable plasma IL-6, whereas fewer (29 [9.1%]) had extremely high plasma concentrations of IL-6.

Citation: American Journal of Veterinary Research 77, 1; 10.2460/ajvr.77.1.13

Discussion

In the present study, plasma concentrations of IL-6 in 319 resting Standardbred racehorses were measured by use of a novel ELISA following its validation for use in this species. The use of PAb in solid and liquid phases has previously been shown to enable the successful development of a sandwich ELISA.14,15 It was reported that a sandwich ELISA revealed nonspecific reactivity of tumor necrosis factor-α in nearly all healthy human serum samples16 when PAb in both solid and liquid phases was used. However, the results of the present study indicated that PAb was suitable for specific detection of IL-6 in equine plasma samples, as supported by work from other investigators.17,18 Some investigators have indicated that certain MAbs only recognize monomers but many endogenous cytokines exist as polymers, and this may be attributed to the lower quantification values obtained.19,20 Use of PAbs appears to render a possible solution to this problem and may reveal cytokine concentrations in equine plasma samples with improved accuracy.

The calibrator dilution buffer was a critical factor for effective quantification of biomolecules in the present study. When the recombinant equine IL-6 was spiked with the ANP, the protein was detectable by use of the ELISA, although the recovery was low; however, recovery was higher when the abundant proteins were removed by use of polyethylene glycol. Similar to the results of this study, findings of a previous study17 indicated that low recovery of recombinant tumor necrosis factor protein in human plasma was attributable to the presence of possible binding factors. It was evident that the optical density with the same recombinant human tumor necrosis factor concentration was diminished when more human plasma was used to dilute the recombinant protein. A similar effect was evident with plasma samples obtained from other species, except that rat serum had a greater negative effect on tumor necrosis factor detection. The present study revealed decreased sensitivity and accuracy of the ELISA when plasma was used as the diluent of the calibrators. Equine plasma had a strong negative effect on detection of the recombinant cytokine by the ELISA, and low concentrations of recombinant equine IL-6 were rarely detectable. Although the use of equine plasma or serum as a standard diluent would be comparable when plasma cytokines are assayed, the inhibitory effect of unknown factors such as cytokine receptors suggests that use of equine plasma or serum as standard diluents would be problematic unless such inhibitory factors are eliminated.21

The data obtained in the present study indicated that the ELISA method had high sensitivity for IL-6 detection in equine plasma samples, could detect reduced reactivity when the sample was extensively diluted, and had good linearity when the plasma samples were serially diluted. However, reasonable linearity was not obtained at low dilutions for the plasma samples that contained high concentrations of IL-6. The poor linearity was not attributed to the ELISA method, but to extremely high analyte concentration; high analyte concentration in samples clearly contributed to the lack of linearity because the signal intensity was proportionally decreased when the samples were extensively diluted. Thus, equine plasma samples needed to be adequately diluted for accurate quantification of IL-6 concentration by use of this assay when a high concentration of the antigen was present. The sample diluent was able to break down the antigen complex; following appropriate dilution of samples with unusually high antigen concentration, the analyte was accurately quantified with the ELISA.

The complex composition of plasma samples may interfere with antibody binding to the target protein, thereby reducing detection of the protein of interest. Particularly, equine plasma contains various components that may mask the target protein and make the epitopes inaccessible to the antibody. The presence of other binding agents, such as autoantibody, ligands, or α2-macroglobulin, may also block the detection of the cytokine to some extent, thereby providing a falsely low concentration reading.22,23 Owing to their binding properties, receptors in the plasma samples may also affect the detection of cytokines by immunoassay19,24,25; a detergent can break down the bound complexes and improve detection. In the present study, the sample diluent was an important factor for effective quantification of biomolecules in plasma samples. Various diluents were evaluated, but only Tween 20 was suitable for use in equine plasma samples; other diluents were not as useful because the low signal intensity was not proportional to the dilution factor and had unacceptably poor linearity. Although another detergent, Triton X-100, can also dissociate bioactive tumor necrosis factor-α complexes into monomers and improve its detection as well as modulate the viscosity effect of the matrix for antibody-antigen interaction,8,26 it was not an appropriate diluent for assaying IL-6 in the present study. Furthermore, with use of diluents other than Tween 20, tumor necrosis factor-α was not detected in serum samples from horses with chronic pulmonary disease.18 A previous study8 revealed that nonionic detergents such as Tween 20 can dissociate the protein complex and convert oligomers of human tumor necrosis factor-α, IL-8, and lymphotoxin into monomers. Nevertheless, in the present study, we were able to determine plasma IL-6 concentration near the assay detection threshold, suggesting that the optimized conditions used were sufficient for counteracting any matrix effect. In our opinion, Tween 20 enhanced detection of more cytokine molecules by dissociating protein complexes in the diluted samples and improved the sensitivity of the immunoassay to reflect the actual concentration of the protein of interest in the equine plasma samples. Tween 20 has proven to be an effective plasma diluent for the analysis of IL-6 concentration, and dilution of highly concentrated samples with Tween 20 can increase the detection signal.

Polyclonal antibody can significantly increase the sensitivity of an assay, compared with findings with MAb; for example, there is a 20-fold increase in sensitivity for tumor necrosis factor-α detection when the CAb is polyclonal but only a 6-fold increase when the CAb is monoclonal.27,28 It has been shown that MAb is not helpful in the measurement of pro–IL-1β synthesis.29 Detection of immunoreactive tumor necrosis factor-α is greatly influenced by the type of MAb because tumor necrosis factor-α can be masked up to 75% owing to the presence of receptors for tumor necrosis factor-α.20 Polyclonal antibody is widely used in commercially available ELISA methods. In addition, on the basis of results from another laboratory, equine IL-6 PAb can be used as both the CAb and DAb to detect IL-6 in equine serum samples.14

In the present study, applicability of the ELISA method was also supported by results of the lipopolysaccharide-cytokine induction experiment. Lipopolysaccharide is a broad-spectrum stimulant that can induce cytokine production,30 and expression of proinflammatory cytokines can be stimulated by lipopolysaccharide through toll-like receptor-4 stimulation.31,32 The significant in vitro induction of IL-6 in plasma by lipopolysaccharide observed in the present study has been similarly observed by other investigators.33 Plasma samples from horses after extracorporeal shock wave treatment were used to evaluate the newly validated IL-6 ELISA. The data obtained indicated that a decrease in plasma IL-6 concentration as a result of extracorporeal shock wave treatment could be measured with this assay.

Following completion of the validation process, the assay method was used to measure IL-6 concentrations in plasma samples collected from 319 Standardbred racehorses at 2 racetracks in Pennsylvania. It is interesting to note that many horses had no detectable plasma IL-6, whereas some had a very high concentration (maximum concentration, 194 ng/mL). These results further indicated that the present method can specifically detect IL-6 in equine plasma samples, and the detection signal was abolished when IL-6 was absent from the matrix. Among the 319 horses, 130 (approx 40%) had extremely low or nondetectable concentrations of circulating IL-6, 55 (17%) had low concentrations of 0.101 to 1.0 ng/mL, 108 (34%) had concentrations of 1.001 to 10 ng/mL, and 29 (9.1%) had the highest concentrations of 10.001 to 193,630 ng/mL. The detection range (78 to 5,000 pg/mL) of the assay encompassed IL-6 concentrations in 50% of the samples and identified approximately 35% of samples as negative; only 15% of samples were outliers because of extremely high concentration. However, those high IL-6–content samples can be assayed after dilution; therefore, this method appears to be useful for monitoring concentrations of the cytokine and could provide meaningful diagnostic information in a clinical setting. Previous studies10,14,34 have evaluated circulating IL-6 concentrations in clinical cases. The results of the present study are not comparable to those of a study14 investigating the concentrations found in normal foals, foals with sepsis, or foals that had or had not ingested colostrum. In that study,14 the mean serum IL-6 concentration was 1,995 ng/mL in 15 normal foals (controls), whereas 15 foals with sepsis had significantly lower IL-6 concentrations. Lower concentrations in foals with sepsis are not consistent with the notion that IL-6 is a critical mediator of the acute-phase response; however, severe sepsis leading to death results in decreased production of inflammatory cytokines.10,34 In addition, the difference in IL-6 concentrations between the horses of the present study and those of the previous study14 may be attributed to differences in the assay procedure, given that the method used in the previous study was not validated. Finally, there also may be differences in circulating IL-6 concentrations between adult and juvenile horses.

Interleukin-6 acts as a proinflammatory factor when it is secreted from leukocytes.34 The great variation found in the Standardbreds of the present study should represent their actual physiologic status, and the failure to detect IL-6 in some samples may reflect the absence of inflammation. Interestingly, IL-6 also acts as an anti-inflammatory cytokine when it is secreted from myocytes in response to extensive exercise, thereby protecting muscle from damage.2 Interlukin-6 has extensive anti-inflammatory functions in its role as a myokine and is secreted into the bloodstream in response to muscle contractions.35 Racehorses may have higher expression of IL-6, compared with that of nonracing horses, because IL-6 gene expression is also significantly elevated in leukocytes with exercise.3 Although the horses in the present study were at rest, it is unknown whether exercise training or inflammation contributed to the high plasma IL-6 concentrations in some horses, and the medical background of these horses was unknown except that they were deemed fit to race. Given the possibility of underlying musculoskeletal disorders and inflammatory processes, the high concentration in some of this group of horses is an observation worthy of further investigation.

Results of the present study indicated that the ELISA for determination of plasma IL-6 concentrations in horses was reliable and reproducible and offers promise for future applications. The assay was appropriate for the study of IL-6 concentration in equine plasma samples, and the relationship between plasma IL-6 concentration and many disease processes in horses is unknown.

Acknowledgments

Financial support for this study was provided by the PA Racing Commissions and the PA Harness Horsemen Association at Pocono and Chester Downs and by Meadows Standardbred Owners Association and Horsemen Benevolent and Protection Association at Penn National and Presque Isles Downs.

The authors declare that there were no conflicts of interest.

The authors thank Hilary Goff, Deborah Tsang, and Dr. Dipti Mangal for technical assistance and Dr. Tom Tobin for help with study design and interpretation of the results.

ABBREVIATIONS

ANP

Analyte-negative plasma

CAb

Capture antibody

DAb

Detection antibody

IL

Interleukin

MAb

Monoclonal antibody

PAb

Polyclonal antibody

sOD

Specific optical density

UAb

Unlabeled antibody

Footnotes

a.

R & D Systems, Minneapolis, Minn.

b.

2100 Bioanalyzer, Agilent Technologies, Santa Clara, Calif.

c.

Sigma-Aldrich, St Louis, Mo.

d.

Fisher Scientific, Suwannee, Ga.

e.

Corning Inc, Lowell, Mass.

f.

Neogen Corp, St Joseph, Mich.

g.

iMark Microplate Reader, Bio-Rad Laboratories Inc, Hercules, Calif.

h.

Microplate Manager, version 6.0, Bio-Rad Laboratories Inc, Hercules, Calif.

i.

Kingfisher Biotech, Saint Paul, Minn.

j.

Storz Medical AG, Tägerwilen, Switzerland.

References

  • 1. Dinarello CA. Proinflammatory cytokines. Chest 2000; 118: 503508.

  • 2. Febbraio MA, Pedersen BK. Contraction-induced myokine production and release: is skeletal muscle an endocrine organ? Exerc Sport Sci Rev 2005; 33: 114119.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Donovan DC, Jackson CA, Colahan PT, et al. Exercise-induced alterations in pro-inflammatory cytokines and prostaglandin F2α in horses. Vet Immunol Immunopathol 2007; 118: 263269.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4. Swiderski CE, Sobol G, Lunn DP, et al. Molecular cloning, sequencing, and expression of equine interleukin-6. Vet Immunol Immunopathol 2000; 77: 213220.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5. Martin RM, Patel R, Oken E, et al. Filter paper blood spot enzyme linked immunoassay for adiponectin and application in the evaluation of determinants of child insulin sensitivity. PLoS ONE [serial online]. 2013; 8: e71315. Available at: journals.plos.org/plosone/article?id=10.1371/journal.pone.0071315. Accessed August 6, 2014.

    • Search Google Scholar
    • Export Citation
  • 6. Sizemore N, Dudeck RC, Barksdale CM, et al. Development and validation of two solid-phase enzyme immunoassays (ELISA) for quantitation of human epidermal growth factors (hEGFs). Pharm Res 1996; 13: 10881094.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7. Aziz N, Nishanian P, Mitsuyasu R, et al. Variables that affect assays for plasma cytokines and soluble activation markers. Clin Diagn Lab Immunol 1999; 6: 8995.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8. Petyovka N, Lyach L, Voitenok NN. Homologous ELISA for detection of oligomeric human TNF: properties of the assay. J Immunol Methods 1995; 186: 161170.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Yang RB, Mark MR, Gurney AL, et al. Signaling events induced by lipopolysaccharide-activated toll-like receptor 2. J Immunol 1999; 163: 639643.

    • Search Google Scholar
    • Export Citation
  • 10. Ertel W, Kremer J-P, Kenney J, et al. Downregulation of proinflammatory cytokine release in whole blood from septic patients. Blood 1995; 85: 13411347.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11. Mangal D, Uboh CE, Jiang Z, et al. Interleukin-1β inhibits synthesis of 5-lipooxygenase in lipopolysaccharide-stimulated equine whole blood. Prostaglandins Other Lipid Mediat 2014; 108: 922.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12. Frisbie DD, Kawcak CE, McIlwraith CW. Evaluation of the effect of extracorporeal shock wave treatment on experimentally induced osteoarthritis in middle carpal joints of horses. Am J Vet Res 2009; 70: 449454.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13. Ringer SK, Lischer CJ, Ueltschi G. Assessment of scintigraphic and thermographic changes after focused extracorporeal shock wave therapy on the origin of the suspensory ligament and the fourth metatarsal bone in horses without lameness. Am J Vet Res 2005; 66: 18361842.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14. Burton AB, Wagner B, Erb HN, et al. Serum interleukin-6 (IL-6) and IL-10 concentrations in normal and septic neonatal foals. Vet Immunol Immunopathol 2009; 132: 122128.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15. Abe Y, Miyake M, Sagawa T, et al. Enzyme-linked immunosorbent assay (ELISA) for human tumor necrosis factor (hTNF). Clin Chim Acta 1988; 176: 213217.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16. Abe Y, Miyake M, Miyazaki T, et al. Nonspecific reactions in the sandwich immunoassay for human tumor necrosis factor-α. Clin Chim Acta 1989; 181: 223229.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17. Engelberts I, Möller A, Schoen G, et al. Evaluation of measurement of human TNF in plasma by ELISA. Lymphokine Cytokine Res 1991; 10: 6976.

    • Search Google Scholar
    • Export Citation
  • 18. Lavoie-Lamoureux A, Maghni K, Lavoie J-P. Optimization of a procedure to accurately detect equine TNFα in serum samples. Vet Immunol Immunopathol 2010; 138: 118123.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19. Mire-Sluis AR, Gaines-Das R, Thorpe R. Immunoassays for detecting cytokines: what are they really measuring? J Immunol Methods 1995; 186: 157160.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Kapadia S, Torre-Amione G, Mann DL. Pitfalls in measuring cytokines. Ann Intern Med 1994; 121: 149150.

  • 21. Aukrust P, Liabakk NB, Müller F. Serum levels of tumour necrosis factor-alpha (TNF-α) and soluble TNF receptors in human immunodeficiency virus type I infection: correlation to clinical immunologic and virologic parameters. J Infect Dis 1994; 169: 420424.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Mae N, Liberato DJ, Chizzonite R, et al. Identification of high-affinity anti-IL-1α autoantibodies in normal human serum as an interfering substance in a sensitive enzyme-linked immunosorbent assay for IL-1α. Lymphokine Cytokine Res 1991; 10: 6168.

    • Search Google Scholar
    • Export Citation
  • 23. James K, Milne I, Cunningham A. The effect of α2-macroglobulin in commercial cytokine assays. J Immunol Methods 1994; 168: 3337.

  • 24. Seckinger P, Zhang JH, Hauptmann B, et al. Characterization of a tumor necrosis factor-α inhibitor: evidence of immunological cross-reactivity with the TNF receptor. Proc Natl Acad Sci U S A 1990; 87: 51885192.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25. Whicher J, Ingham E. Cytokine measurements in body fluids. Eur Cytokine Netw 1990; 1: 239243.

  • 26. Morgan CL, Newman DJ, Burrin JM, et al. The matrix effects on kinetic rate constants of Ab-Ag interactions reflect solvent viscosity. J Immunol Methods 1998; 217: 5160.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27. McLaughlin PJ, Elwood NJ, Ramadi LT, et al. Improvement in sensitivity of enzyme-linked immunosorbent assay for tumor necrosis factor. Immunol Cell Biol 1990; 68: 5155.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28. Abe Y, Kimura K, Horiuchi A, et al. Improvement of ELISA sensitivity by allogeneric adsorption of polyclonal antibodies—a technical note for non-experts. Clin Chim Acta 1994; 224: 103105.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29. Dinarello CA. ELISA kits based on monoclonal antibodies do not measure total IL-1β synthesis. J Immunol Methods 1992; 148: 255259.

  • 30. Singh Suri S, Janardhan KS, Parbhakar O, et al. Expression of toll-like receptor 4 and 2 in horse lungs. Vet Res 2006; 37: 541551.

  • 31. Garde AH, Fomsgaard A, Høiby N. An easy microtiter assay for quantitation of cytokine induction by lipopolysaccharide (LPS) and activity of LPS-binding serum components. APMIS 1995; 103: 286292.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32. Hao H-N, Peduzzi-Nelson JD, VandeVord PJ, et al. Lipopolysaccharide-induced inflammatory cytokine production by Schwann's cells dependent upon TLR4 expression. J Neuroimmunol 2009; 212: 2634.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33. Ajuwon KM, Jacobi SK, Kuske JL, et al. Interleukin-6 and interleukin-15 are selectively regulated by lipopolysaccharide and interferon-gamma in primary pig adipocytes. Am J Physiol Regul Integr Comp Physiol 2004; 286: R547R553.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34. Buras JA, Holzmann B, Sitkovsky M. Animal models of sepsis: setting the stage. Nat Rev Drug Discov 2005; 4: 854865.

  • 35. Pedersen BK, Febbraio MA. Muscle as an endocrine organ: focus on muscle-derived interleukin-6. Physiol Rev 2008; 88: 13791406.

All Time Past Year Past 30 Days
Abstract Views 68 0 0
Full Text Views 393 210 31
PDF Downloads 230 107 4
Advertisement