Journal oflmmunological Methods, 44 (1991) 257-263
257
© 1991 Elsevier Science Publishers B.V. All rights reserved 0022-1759/91/$03.50
JIM06132
Short communication
A kinetic assay for eosinophil peroxidase activity in eosinophils and eosinophil conditioned media Steven R. White, Giorgio V.P. Kulp, Stephen M. Spaethe, Eldwin Van Alstyne and Alan R. L e f t Section of Pulmonary and Critical Care Medicine, Department of Medicine, Division of Biological Sciences, University of Chicago, Chicago, IL, U.SJI., and the Pulmonary Research Division, Lilly Research Laboratories, Indianapolis, IN, U.S.A. (Received 10 May 1991, revised received 9 August 1991, accepted 12 August 1991)
The activity of eosinophil peroxidase (EPO) is commonly employed as a measure of eosinophil activation in biologic fluids. Determination of product formation by this enzyme by end-point measurement may be affected profoundly by substrate concentrations, reaction time and degradation of end-product and enzyme. To determine more accurately EPO concentrations in media conditioned by isolated, purified eosinophils, we have developed a kinetic, colorimetric assay to measure EPO concentration as a function of maximum velocity of reaction (Vmax). An automated method for determining Vmax in a 96-well microplate colorimetric assay was utilized over a wide range of substrate concentrations. Concentrations >/3 x 10 -8 g / m l could be determined reliably with this assay. Peroxidase activity was inhibited in a concentration-dependent manner by the addition of 3-amino-l,2,4-triazole (AMT). The EPO concentration in eosinophils determined by this kinetic method was ~ 1.1 x 10 -5 g/106 eosinophils. Eosinophil activation with 10 -6 M f-Met-Leu-Phe (fMLP) caused substantial EPO secretion (9.0 + 1.7% vs. 2.9 + 0.6% total EPO content for control, P = 0.05) and decrease in eosinophil EPO concentration (92.3 + 4.2% of control, P = 0.038). Secretion was enhanced by the addition of 5 / z g / m l cytochalasin B to 10 -6 M fMLP (25.9 + 12.7% total EPO content, P = 0.043 vs. control); similar decreases were noted in eosinophil EPO concentration (71.7 + 16.1% of control, P = 0.043). These data demonstrate that determination of EPO secretion by measurement of Vmax is a reliable, accurate method for precise quantification of this enzyme in media containing purified eosinophils or eosinophil products in the absence of other forms of peroxidase activity. Key words: Eosinophil; Eosinophil peroxidase; Kinetic assay; Colorimetric
assay
Introduction Correspondence to: S.R. White, Section of Pulmonary and Critical Care Medicine, The University of Chicago, 5841 S. Maryland Ave., Box 98, Chicago, IL 60637, U.S.A. Abbreviations: EPO, eosinophii peroxidase; MPO, myeloperoxidase; CYB, cytochalasin B; fMLP, f-Met-Leu-Phe; Tfis, tris(hydroxymethyl)aminomethane-HCl; AMT, 3-amino1,2,4-triazole; Vm~x, maximum velocity of reaction; HBSS, Hanks' balanced salt solution.
Assay for the peroxidase activity of eosinophils has been utilized both as an assay of eosinophil function (Kroegel et al., 1988) and number (Strath et al., 1985) in biologic media. Human eosinophils contain large quantities of eosinophil peroxidase
258 (EPO), which is different in structure from the myeloperoxidase (MPO) of neutrophils (Wever et al., 1982). Strath et al. (1985) demonstrated a simple assay for EPO utilizing o-phenylenediamine as a hydrogen donor, which depends on end-point measurement of the absorbance of the product at 492 nm. This assay accurately estimates eosinophil number; however, the reaction time is prolonged (30 min) and requires quenching by sulfuric acid to stop the reaction and stabilize the end-product. This assay also was not optimized for use in determining the secretion of EPO in media nor to describing the activity of isolated eosinophils. End-point assays that determine the secretion of EPO - and the corresponding level of activation of eosinophils - by end-point measurement of product may not be optimal for assessment of activity in purified suspensions of eosinophils. Enzyme activity may be affected profoundly by a number of factors, including enzyme and substrate concentrations and the time of reaction, such that different concentrations of enzyme in solution may yield similar concentrations of a product. We wished to determine the degranulating activity of eosinophils isolated and purified from peripheral blood. We have developed a colorimetric microplate assay to measure the maximal rate of reaction (Vmax) for concentrations of purified EPO and EPO present in eosinophils and media conditioned by purified eosinophils. This assay avoids the potential problems of an endpoint determination of enzyme concentrations and is optimized for measuring a wide range of EPO concentrations both in solution and in eosinophil cell pellets. This assay may prove useful in the measurement of this enzyme in purified suspensions of eosinophils and eosinophil-conditioned media, and in determining the activity of stimulated eosinophils.
Methods
Isolation of human eosinophils Human eosinophils were isolated from volunteers according to a protocol approved by the
University of Chicago Institutional Review Board. Informed consent was obtained from all human volunteers in this study prior to participation. Whole blood (180 ml) was withdrawn from the antecubital vein of eight human volunteers and placed into containers containing 2 ml of 1:100 heparin. Blood then was mixed 5:1 with 4.5% dextran in 0.9% NaCI and sedimented for 45 min at 22 ° C. The white blood cell layer was collected, divided into 30 ml aliquots, and layered over 10 ml Ficoll-Hypaque (density 1.077 g/ml) in 50 ml tubes. Cells then were centrifuged at 400 x g for 20 min (all centrifugations done at 22 ° C). The pellet containing leukocytes was resuspended in Ca2+-free Hanks' balanced salt solution (HBSS) and twice washed and centrifuged at 400 x g for 10 min. Cells then were suspended in HBSS containing 1 mmol/1 Ca 2÷ and 5% fetal calf serum and diluted to a concentration of 2 x 107/ml. In separate 15 ml tubes discontinuous colloidal polyvinylpyrrolidine-coated silica (Percoll) gradients were prepared (from top to bottom in g / m l [ml]: 1.080 [2.5]; 1.085 [2.5]; 1.090 [3.0]; 1.095 [3.0]; 1.100 [1.5], prepared in HBSS), over which 2 ml of cell suspension was layered. Gradients then were centrifuged at 700 x g for 20 min. The interface containing eosinophils (1.095-1.100 g/ml) was collected, diluted in HBSS containing 1 mmol/1 Ca z÷ and 0.1% gelatin, and then centrifuged at 400 x g for 10 min. Eosinophils then were washed, centrifuged at 400 X g for 10 min and resuspended in HBSS containing 1 mmol/l Ca 2÷ and 0.1% gelatin. Cells were counted and assessed for purity by Wright stain. In these experiments, 180 ml of blood yielded > 5 x 10 6 eosinophils with purity > 92%. Cells were kept on ice until use and activated within 1 h of final purification and collection to prevent deterioration.
Purification of human eosinophil peroxidase Purified EPO used in this assay was a generous gift of Dr. Gerald J. Gleich, Mayo Clinic, Rochester, MN, and was isolated from an eosinophil-rich suspension obtained by leukapheresis of patients with hypereosinophilic syndromes (Agosti et al., 1987). Briefly, eosinophils were disrupted by sonication, and then solubilized, extracted in 0.01 M HC1 and centrifuged at
259 40,000 × g for 20 min. Granular protein extracts then were chomatographed on a Sephadex G-50 column equilibrated with 0.25 M acetate buffer (pH 4.3) with 0.15 M NaCI. Fractions eluting with peroxidase activity were collected and further purified to an A415 nm/a280 nrn ratio > 0.9 by chromatography on carboxymethyl Sepharose as previously described (Carlson et al., 1985). Purity of EPO was assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis demonstrating only two discrete bans (molecular weights ~ 78,000 and 14,000) that correspond to the large and small subunits of EPO (Carlson et al., 1985). No significant contamination was present. EPO was kept frozen at - 7 0 °C until immediately prior to use.
Determination of eosinophil peroxidase concentrations The method of Strath et al. (1985) was modified for this assay. The assay is based on the oxidation of o-phenylenediamine by EPO in the presence of hydrogen peroxide (H202). Hydrogen peroxide (final concentration, 0.0012%0.01%) and o-phenylenediamine (final concentration, 2-16 mM) were dissolved in 100 mM tris(hydroxymethyl)-aminomethane-HCl buffer (Tris), pH 8.0, containing 0.1% Triton X-100, immediately prior to use. Eosinophil suspensions, eosinophil-conditioned medium, and purified EPO (M.W. ~ 70,000) were suspended in HBSS containing 0.1% gelatin and frozen at - 7 0 ° C until immediately prior to use. For the assay, 50 tzl of sample were combined with 75 /~1 of substrate in a polystyrene 96-well microplate and placed into a thermoregulating microplate absorbance spectrophotometer (Thermomax, Molecular Devices, Menlo Park, CA) at 37 ° C. Absorbance at 492 nm was measured every 6 s for 3 min; the maximal velocity of the reaction (Vmax) was calculated by interpolation between successive four points (24 s) utilizing customized software (Softmax v2.01, Molecular Devices) on a Macintosh computer. Purified EPO standards were assayed in triplicate or quadruplicate; cell suspensions or conditioned media samples were assayed in duplicate. Eosinophil suspensions were diluted 1/4 with HBSS containing 1 mmol/l Ca 2÷ immediately prior to assay. Final EPO concentra-
tions then were calculated from standard curves fitted by 4-parameter (iterative log-logit) analysis.
Inhibition of EPO activity To determine the effect of an inhibitor for EPO, 3-amino-l,2,4-triazole (AMT), in this assay, experiments were repeated in the presence of 1-30 mM (final concentration) AMT added to the substrate. Experiments with purified EPO standards then were carried out as above.
Activation of human eosinophils For activation, 5 × 105 eosinophils were incubated in 200/~1 HBSS containing 0.1% gelatin in polystyrene tubes with 10 -8 to 10 -6 M f-MetLeu-Phe (fMLP), fMLP with 5 ~ g / m l cytochalasin B (CYB), or control medium (no agonist) for 30 min at 37°C (one concentration per tube). Eosinophils then were placed on ice and centrifuged at 4 ° C at 400 ×g. Eosinophil pellets diluted to 1 ml with HBSS containing 0.1% gelatin and conditioned medium (undiluted) were collected and stored separately at - 70 ° C until EPO analysis (above). For analysis of EPO concentrations in these experiments, concentrations of substrate (16 mM o-phenylenediamine and 0.01% H202) determined to be optimal in experiments with EPO standards (above) were utilized.
Materials Percoll, tris(hydroxymethyl)aminomethaneHCI (as 99.9% Trisma base), Triton X-100, H202, o-phenylenediamine, 3-amino-l,2,4-triazole, cytochalasin B and gelatin were obtained from Sigma, St. Louis, MO. All reagents were molecular biology grade or higher.
Data analysis Eosinophil peroxidase concentrations are calculated either as g / m l medium for purified EPO standards, or g/10 6 eosinophils for cell activation experiments, and expressed finally as mean _+ standard error of the mean (SEM). Peroxidase concentrations in eosinophil-conditioned medium are also expressed as % total content of the eosinophils (medium/[medium + cell pellet]). For experiments using AMT, inhibition is expressed as mean % baseline Vm~x for the same concentration of EPO; comparisons to baseline were made
260
with the 95% confidence interval (CI). Coml~arisons between control and activated eosinophils were made with the Wilcoxon signed-rank test. Significance was claimed when P < 0.05.
2000 A
.E E
1500
O
1000
Results
x m
E
-
50o -
Determination of EPO concentrations in standard samples Eosinophil peroxidase standards reacted rapidly with substrate. Reactions were complete in < 3 min for all concentrations of E P O >__10 -6 g / m l and all concentrations of substrate utilized, and Vm~x typically was found in the first 30-45 s of reaction (Fig. 1). A discernable Vm~x could be recognized for concentrations of E P O __>3 × 10 -8 g / m l (Fig. 2). Concentrations > 3 x 10 -5 g / m l had a Vmax within the first 12 s of assay and could not be measured accurately without dilution. Sensitivity of the assay was increased by increasing concentrations of o-phenylenediamine and H 2 0 2 (Fig. 2). From these experiments we determined an optimal range of final concentrations of ophenylenediamine to be 4-16 mM and H 2 0 2 t o be 0.0024-0.01%. In subsequent experiments utilizing eosinophils and eosinophil conditioned medium (below), concentration of o-phenylenediamine was set at 16 mM and H 2 0 2 at 0.01%.
I
=.
o
-8
¢, -7
-6
-5
EPO (log g/rnl) Fig. 2. Representative standard curves for the kinetic EPO assay. Standards are shown for two concentrations of o-phenylenediamine and H202 utilized: 16 mM o-phenylenediamine and 0.01% H202 (closed circles), and 3.2 mM o-phenylenediamine and 0.002% H202 (open circles). Utilization of lower concentrations of substrate led to significant limitation of the reaction at higher concentrations of EPO. Each curve represents standards assayed in triplicate or quadruplicate. Curves were fitted to points by 4-parameter (iterative log-logit) analysis.
100%
80%
Inhibition of EPO activity Addition of A M T inhibited the reaction of o-phenylenediamine and H 202 catalyzed by EPO.
c O
60%'
o 1409
mOD/min
744 mOD/min
373 mOD/min
,ooo
x
40%
E omm °mm
500-
20%
~ axlo -~ g/m,
7,,'=.,~-
E3 0
~
1 x l 0 -6 g/ml 0%
~ 60
120
Time
180
(see)
Fig. 1. Representative reaction curves for eo,Anophil peroxidase (EPO) in the kinetic colorimetic assay system. For each concentration > 10 -6 g / m l purified EPO, the end-point optical density (mOD) was similar at the end of 3 min. However, a clear Vm~, could be distinguished for each concentration. Maximum velocity is shown for each concentration.
i
.
1
3
.
.
AMT
.
.
.
.
,
10
30
(mM)
Fig. 3. Inhibition of EPO activity by 3-amino-l,2,4-triazole (AMT). Increasing concentrations of AMT reduced the maximal Vm~ obtained for 10 -5 g / m l EPO in a concentration-dependent manner, n = 4-5 experiments at each point, each done in triplicate.
261 Inhibition was noted for all concentrations of A M T > 3 mM (Fig. 3). Maximum velocity of 10 -5 g / m l E P O with o-phenylenediamine and H z O 2 after addition of 10 mM A M T was 41.1 + 4.1% (95% CI, 29.6 to 52.6%) of control Vm,x (Fig. 3). Addition of 30 mM A M T caused nearcomplete inhibition of the reaction of 10 -5 g / m l E P O with o-phenylenediamine and H 2 0 z (17.8 + 0 . 8 % of control Vm~; 95% CI, 15.7-19.9%) (Fig. 3).
Determination of EPO concentrations in eosinophils and eosinophil-conditioned medium Both eosinophil cell pellets and eosinophilconditioned medium had concentrations of E P O that could be measured with this assay. Cell pellets from unstimulated eosinophils had a baseline E P O concentration of 1.07 + 0.16 x 10 -5 g/106 cells. Subsequent activation with 10 - 6 M fMLP alone decreased pellet E P O concentration to 0.96 + 0.15 × 10 -5 g / 1 0 6 cells (92.3 ± 4.2% of control, P = 0.038, n = 11 experiments). Activation with 10 -6 M fMLP plus 5 / x g / m l CYB decreased cell pellet E P O concentration substantially, from 1.19 ± 0.32 to 0.78 ± 0.31 × 10 -5 g / 1 0 6 cells (71.7 + 16.1% of control, P = 0.043, n = five experiments) (Fig. 4).
110% 100% •=
110%" ~ . . , ~
90%
~ .i
90*/0-
80%
~
80./o-
70%
O
g
ILl
100%"
~
.
~
,
a. 70%" I,U
60% 50%
L
60%" -8 -7 -6 fMLP (log M)
50%
8 -7 fMLP (log M)
40% 35% I:: ¢D C 0 30% t~ 0
0Q fMLP fMLP+ 5 ~ag/rnlcytochalasinB
i
25% 20%
i0 k~
15%
2
C/) 0 " ILl
_2*
10% 5% O%
Control
-8 fMLP (M)
-7
-6
Fig. 5. Eosinophil peroxidase (EPO) concentrations in eosinophil-conditioned medium after stimulation with either f-Met-Leu-Phe (fMLP) or fMLP plus cytochalasin B (CYB). Activation led to concentration-dependent increases in medium EPO concentrations. Concentration is normalized to total content (medium+cell). n = 11 (fMLP) or 5 (fMLP plus CYB) experiments, each done in duplicate. *, P ~/3 × 10 -s g / m l could be determined reliably with this assay (Fig. 2). Addition of A M T inhibited the reaction catalyzed by E P O in a concentration-dependent manner (Fig. 3). Cramer et al. (1984) have demonstrated that A M T inhibited EPO activity in mixed granulocyte suspensions while inhibiting MPO derived from neutrophils only moderately. Our data confirm the ability of A M T to inhibit E P O and demonstrate a concentration-dependent effect of A M T o n VmaxEosinophil peroxidase activity could be measured both in cell suspensions (Fig. 4) and conditioned media (Fig. 5). Activation of eosinophils by either fMLP alone or fMLP plus CYB caused secretion of EPO into media (Fig. 5). These data demonstrate that secretion of small quantities of EPO by eosinophils into media can be determined in this assay and suggest that this assay may be useful in determining the activity of eosinophils. Both E P O and the MPO of neutrophils oxidize o-phenylenediamine (Strath et al., 1985). In the pure suspensions of eosinophils utilized in our experiments, contamination with leukocytes was insignificant ( < 5 % ) ; thus, cross-reaction with MPO in our assay was not a significant problem. Under conditions where significant concentrations of MPO or numbers of neutrophils exist in conjunction with E P O and eosinophils, other assay methods that minimize or abolish the activity of MPO may be required to distinguish activity of E P O from MPO, such as the oxidation of homovanillic acid by E P O but not MPO (Menegazzi et al., 1991). In contrast to the oxidation of ho-
movanillic acid, which requires measurement of oxidation by spectrofluorescence, our assay can be performed in a microplate spectrophotometer. This may allow for rapid measurement of relatively large numbers of samples in a relatively short period of time. In summary, we demonstrate a kinetic assay for determination of E P O concentrations in purified eosinophils and eosinophil-conditioned medium. This assay can determine EPO concentration >t 3 x 10 -8 g / m l , avoids the problems inherent in an end-point assay, and can be performed rapidly in a standard microplate spectrophotometer. This assay may be useful in determining eosinophil activation by precise measurement of both stored, cellular EPO in purified eosinophils and E P O secreted by isolated eosinophils.
Acknowledgements The authors thank Dr. Gerald Gleich, Professor of Medicine, Mayo Graduate School of Medicine, Rochester, MN, for the purified EPO used in these studies. This work was supported by National Heart, Lung and Blood Institute Grants HL-32495 and HL-46368. Dr. White is a recipient of Clinical Investigator Award HL-02484 from the National Heart, Lung and Blood Institute, an Edward Livingston Trudeau Fellowship from the American Lung Association, and is a fellow of the Schweppe Foundation.
References Agosti, J.M., Altman, L.C., Ayars, G.H., Loegering, D.A.,
Gleich, G.J. and Klebanoff,S.J. (1987) The injurious effect of eosinophil peroxidase, hydrogen peroxide and halides on pneumocytesin vitro. J. AllergyClin. Immunol. 79, 496. Carlson, M.G.C., Peterson, C.G.B. and Venge, P. (1985) Human eosinophil peroxidase: purification and characterization. J. Immunol. 134, 1875. Cramer, R., Soranzo, M.R., Dri, P., Menegazzi,R., Pitotti, A., Zabucchi, G. and Patriarca, P. (1984) A simple reliable assay for myeloperoxidase activity in mixed neutrophileosinophil cell suspensions: application to detection of myeloperoxidasedeficiency.J. Immunol. Methods 70, 119.
263 Kroegel, C., Yukawa, T., Dent, G., Chanez, P., Chung, K.F. and Barnes, P.J. (1988) Platelet-activating factor induces eosinophil peroxidase release from purified human eosinophils. Immunology 64, 559. Menegazzi, R., Zabucchi, G., Zuccato, P., Cramer, R., Piccinini, C. and Patriarea, P. (1991) Oxidation of homovanillic acid as a selective assay for eosinophil peroxidase in eosinophil peroxidase-myeloperoxidase mixtures and its
use in the detection of human eosinophil peroxidase deficiency. J. Immunol. Methods 137, 55. Strath, M., Warren, D.J. and Sanderson, C.J. (1985) Detection of eosinophils using an eosinophil peroxidase assay. Its use as an assay for eosinophil differentiation factors. J. Immunol. Methods 83, 209. Wever, R., Hamers, M.N., De Graaf, C.J., Weening, R.S. and Roos, D. (1982) Adv. Exp. Med. Biol. 141,501.
257
© 1991 Elsevier Science Publishers B.V. All rights reserved 0022-1759/91/$03.50
JIM06132
Short communication
A kinetic assay for eosinophil peroxidase activity in eosinophils and eosinophil conditioned media Steven R. White, Giorgio V.P. Kulp, Stephen M. Spaethe, Eldwin Van Alstyne and Alan R. L e f t Section of Pulmonary and Critical Care Medicine, Department of Medicine, Division of Biological Sciences, University of Chicago, Chicago, IL, U.SJI., and the Pulmonary Research Division, Lilly Research Laboratories, Indianapolis, IN, U.S.A. (Received 10 May 1991, revised received 9 August 1991, accepted 12 August 1991)
The activity of eosinophil peroxidase (EPO) is commonly employed as a measure of eosinophil activation in biologic fluids. Determination of product formation by this enzyme by end-point measurement may be affected profoundly by substrate concentrations, reaction time and degradation of end-product and enzyme. To determine more accurately EPO concentrations in media conditioned by isolated, purified eosinophils, we have developed a kinetic, colorimetric assay to measure EPO concentration as a function of maximum velocity of reaction (Vmax). An automated method for determining Vmax in a 96-well microplate colorimetric assay was utilized over a wide range of substrate concentrations. Concentrations >/3 x 10 -8 g / m l could be determined reliably with this assay. Peroxidase activity was inhibited in a concentration-dependent manner by the addition of 3-amino-l,2,4-triazole (AMT). The EPO concentration in eosinophils determined by this kinetic method was ~ 1.1 x 10 -5 g/106 eosinophils. Eosinophil activation with 10 -6 M f-Met-Leu-Phe (fMLP) caused substantial EPO secretion (9.0 + 1.7% vs. 2.9 + 0.6% total EPO content for control, P = 0.05) and decrease in eosinophil EPO concentration (92.3 + 4.2% of control, P = 0.038). Secretion was enhanced by the addition of 5 / z g / m l cytochalasin B to 10 -6 M fMLP (25.9 + 12.7% total EPO content, P = 0.043 vs. control); similar decreases were noted in eosinophil EPO concentration (71.7 + 16.1% of control, P = 0.043). These data demonstrate that determination of EPO secretion by measurement of Vmax is a reliable, accurate method for precise quantification of this enzyme in media containing purified eosinophils or eosinophil products in the absence of other forms of peroxidase activity. Key words: Eosinophil; Eosinophil peroxidase; Kinetic assay; Colorimetric
assay
Introduction Correspondence to: S.R. White, Section of Pulmonary and Critical Care Medicine, The University of Chicago, 5841 S. Maryland Ave., Box 98, Chicago, IL 60637, U.S.A. Abbreviations: EPO, eosinophii peroxidase; MPO, myeloperoxidase; CYB, cytochalasin B; fMLP, f-Met-Leu-Phe; Tfis, tris(hydroxymethyl)aminomethane-HCl; AMT, 3-amino1,2,4-triazole; Vm~x, maximum velocity of reaction; HBSS, Hanks' balanced salt solution.
Assay for the peroxidase activity of eosinophils has been utilized both as an assay of eosinophil function (Kroegel et al., 1988) and number (Strath et al., 1985) in biologic media. Human eosinophils contain large quantities of eosinophil peroxidase
258 (EPO), which is different in structure from the myeloperoxidase (MPO) of neutrophils (Wever et al., 1982). Strath et al. (1985) demonstrated a simple assay for EPO utilizing o-phenylenediamine as a hydrogen donor, which depends on end-point measurement of the absorbance of the product at 492 nm. This assay accurately estimates eosinophil number; however, the reaction time is prolonged (30 min) and requires quenching by sulfuric acid to stop the reaction and stabilize the end-product. This assay also was not optimized for use in determining the secretion of EPO in media nor to describing the activity of isolated eosinophils. End-point assays that determine the secretion of EPO - and the corresponding level of activation of eosinophils - by end-point measurement of product may not be optimal for assessment of activity in purified suspensions of eosinophils. Enzyme activity may be affected profoundly by a number of factors, including enzyme and substrate concentrations and the time of reaction, such that different concentrations of enzyme in solution may yield similar concentrations of a product. We wished to determine the degranulating activity of eosinophils isolated and purified from peripheral blood. We have developed a colorimetric microplate assay to measure the maximal rate of reaction (Vmax) for concentrations of purified EPO and EPO present in eosinophils and media conditioned by purified eosinophils. This assay avoids the potential problems of an endpoint determination of enzyme concentrations and is optimized for measuring a wide range of EPO concentrations both in solution and in eosinophil cell pellets. This assay may prove useful in the measurement of this enzyme in purified suspensions of eosinophils and eosinophil-conditioned media, and in determining the activity of stimulated eosinophils.
Methods
Isolation of human eosinophils Human eosinophils were isolated from volunteers according to a protocol approved by the
University of Chicago Institutional Review Board. Informed consent was obtained from all human volunteers in this study prior to participation. Whole blood (180 ml) was withdrawn from the antecubital vein of eight human volunteers and placed into containers containing 2 ml of 1:100 heparin. Blood then was mixed 5:1 with 4.5% dextran in 0.9% NaCI and sedimented for 45 min at 22 ° C. The white blood cell layer was collected, divided into 30 ml aliquots, and layered over 10 ml Ficoll-Hypaque (density 1.077 g/ml) in 50 ml tubes. Cells then were centrifuged at 400 x g for 20 min (all centrifugations done at 22 ° C). The pellet containing leukocytes was resuspended in Ca2+-free Hanks' balanced salt solution (HBSS) and twice washed and centrifuged at 400 x g for 10 min. Cells then were suspended in HBSS containing 1 mmol/1 Ca 2÷ and 5% fetal calf serum and diluted to a concentration of 2 x 107/ml. In separate 15 ml tubes discontinuous colloidal polyvinylpyrrolidine-coated silica (Percoll) gradients were prepared (from top to bottom in g / m l [ml]: 1.080 [2.5]; 1.085 [2.5]; 1.090 [3.0]; 1.095 [3.0]; 1.100 [1.5], prepared in HBSS), over which 2 ml of cell suspension was layered. Gradients then were centrifuged at 700 x g for 20 min. The interface containing eosinophils (1.095-1.100 g/ml) was collected, diluted in HBSS containing 1 mmol/1 Ca z÷ and 0.1% gelatin, and then centrifuged at 400 x g for 10 min. Eosinophils then were washed, centrifuged at 400 X g for 10 min and resuspended in HBSS containing 1 mmol/l Ca 2÷ and 0.1% gelatin. Cells were counted and assessed for purity by Wright stain. In these experiments, 180 ml of blood yielded > 5 x 10 6 eosinophils with purity > 92%. Cells were kept on ice until use and activated within 1 h of final purification and collection to prevent deterioration.
Purification of human eosinophil peroxidase Purified EPO used in this assay was a generous gift of Dr. Gerald J. Gleich, Mayo Clinic, Rochester, MN, and was isolated from an eosinophil-rich suspension obtained by leukapheresis of patients with hypereosinophilic syndromes (Agosti et al., 1987). Briefly, eosinophils were disrupted by sonication, and then solubilized, extracted in 0.01 M HC1 and centrifuged at
259 40,000 × g for 20 min. Granular protein extracts then were chomatographed on a Sephadex G-50 column equilibrated with 0.25 M acetate buffer (pH 4.3) with 0.15 M NaCI. Fractions eluting with peroxidase activity were collected and further purified to an A415 nm/a280 nrn ratio > 0.9 by chromatography on carboxymethyl Sepharose as previously described (Carlson et al., 1985). Purity of EPO was assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis demonstrating only two discrete bans (molecular weights ~ 78,000 and 14,000) that correspond to the large and small subunits of EPO (Carlson et al., 1985). No significant contamination was present. EPO was kept frozen at - 7 0 °C until immediately prior to use.
Determination of eosinophil peroxidase concentrations The method of Strath et al. (1985) was modified for this assay. The assay is based on the oxidation of o-phenylenediamine by EPO in the presence of hydrogen peroxide (H202). Hydrogen peroxide (final concentration, 0.0012%0.01%) and o-phenylenediamine (final concentration, 2-16 mM) were dissolved in 100 mM tris(hydroxymethyl)-aminomethane-HCl buffer (Tris), pH 8.0, containing 0.1% Triton X-100, immediately prior to use. Eosinophil suspensions, eosinophil-conditioned medium, and purified EPO (M.W. ~ 70,000) were suspended in HBSS containing 0.1% gelatin and frozen at - 7 0 ° C until immediately prior to use. For the assay, 50 tzl of sample were combined with 75 /~1 of substrate in a polystyrene 96-well microplate and placed into a thermoregulating microplate absorbance spectrophotometer (Thermomax, Molecular Devices, Menlo Park, CA) at 37 ° C. Absorbance at 492 nm was measured every 6 s for 3 min; the maximal velocity of the reaction (Vmax) was calculated by interpolation between successive four points (24 s) utilizing customized software (Softmax v2.01, Molecular Devices) on a Macintosh computer. Purified EPO standards were assayed in triplicate or quadruplicate; cell suspensions or conditioned media samples were assayed in duplicate. Eosinophil suspensions were diluted 1/4 with HBSS containing 1 mmol/l Ca 2÷ immediately prior to assay. Final EPO concentra-
tions then were calculated from standard curves fitted by 4-parameter (iterative log-logit) analysis.
Inhibition of EPO activity To determine the effect of an inhibitor for EPO, 3-amino-l,2,4-triazole (AMT), in this assay, experiments were repeated in the presence of 1-30 mM (final concentration) AMT added to the substrate. Experiments with purified EPO standards then were carried out as above.
Activation of human eosinophils For activation, 5 × 105 eosinophils were incubated in 200/~1 HBSS containing 0.1% gelatin in polystyrene tubes with 10 -8 to 10 -6 M f-MetLeu-Phe (fMLP), fMLP with 5 ~ g / m l cytochalasin B (CYB), or control medium (no agonist) for 30 min at 37°C (one concentration per tube). Eosinophils then were placed on ice and centrifuged at 4 ° C at 400 ×g. Eosinophil pellets diluted to 1 ml with HBSS containing 0.1% gelatin and conditioned medium (undiluted) were collected and stored separately at - 70 ° C until EPO analysis (above). For analysis of EPO concentrations in these experiments, concentrations of substrate (16 mM o-phenylenediamine and 0.01% H202) determined to be optimal in experiments with EPO standards (above) were utilized.
Materials Percoll, tris(hydroxymethyl)aminomethaneHCI (as 99.9% Trisma base), Triton X-100, H202, o-phenylenediamine, 3-amino-l,2,4-triazole, cytochalasin B and gelatin were obtained from Sigma, St. Louis, MO. All reagents were molecular biology grade or higher.
Data analysis Eosinophil peroxidase concentrations are calculated either as g / m l medium for purified EPO standards, or g/10 6 eosinophils for cell activation experiments, and expressed finally as mean _+ standard error of the mean (SEM). Peroxidase concentrations in eosinophil-conditioned medium are also expressed as % total content of the eosinophils (medium/[medium + cell pellet]). For experiments using AMT, inhibition is expressed as mean % baseline Vm~x for the same concentration of EPO; comparisons to baseline were made
260
with the 95% confidence interval (CI). Coml~arisons between control and activated eosinophils were made with the Wilcoxon signed-rank test. Significance was claimed when P < 0.05.
2000 A
.E E
1500
O
1000
Results
x m
E
-
50o -
Determination of EPO concentrations in standard samples Eosinophil peroxidase standards reacted rapidly with substrate. Reactions were complete in < 3 min for all concentrations of E P O >__10 -6 g / m l and all concentrations of substrate utilized, and Vm~x typically was found in the first 30-45 s of reaction (Fig. 1). A discernable Vm~x could be recognized for concentrations of E P O __>3 × 10 -8 g / m l (Fig. 2). Concentrations > 3 x 10 -5 g / m l had a Vmax within the first 12 s of assay and could not be measured accurately without dilution. Sensitivity of the assay was increased by increasing concentrations of o-phenylenediamine and H 2 0 2 (Fig. 2). From these experiments we determined an optimal range of final concentrations of ophenylenediamine to be 4-16 mM and H 2 0 2 t o be 0.0024-0.01%. In subsequent experiments utilizing eosinophils and eosinophil conditioned medium (below), concentration of o-phenylenediamine was set at 16 mM and H 2 0 2 at 0.01%.
I
=.
o
-8
¢, -7
-6
-5
EPO (log g/rnl) Fig. 2. Representative standard curves for the kinetic EPO assay. Standards are shown for two concentrations of o-phenylenediamine and H202 utilized: 16 mM o-phenylenediamine and 0.01% H202 (closed circles), and 3.2 mM o-phenylenediamine and 0.002% H202 (open circles). Utilization of lower concentrations of substrate led to significant limitation of the reaction at higher concentrations of EPO. Each curve represents standards assayed in triplicate or quadruplicate. Curves were fitted to points by 4-parameter (iterative log-logit) analysis.
100%
80%
Inhibition of EPO activity Addition of A M T inhibited the reaction of o-phenylenediamine and H 202 catalyzed by EPO.
c O
60%'
o 1409
mOD/min
744 mOD/min
373 mOD/min
,ooo
x
40%
E omm °mm
500-
20%
~ axlo -~ g/m,
7,,'=.,~-
E3 0
~
1 x l 0 -6 g/ml 0%
~ 60
120
Time
180
(see)
Fig. 1. Representative reaction curves for eo,Anophil peroxidase (EPO) in the kinetic colorimetic assay system. For each concentration > 10 -6 g / m l purified EPO, the end-point optical density (mOD) was similar at the end of 3 min. However, a clear Vm~, could be distinguished for each concentration. Maximum velocity is shown for each concentration.
i
.
1
3
.
.
AMT
.
.
.
.
,
10
30
(mM)
Fig. 3. Inhibition of EPO activity by 3-amino-l,2,4-triazole (AMT). Increasing concentrations of AMT reduced the maximal Vm~ obtained for 10 -5 g / m l EPO in a concentration-dependent manner, n = 4-5 experiments at each point, each done in triplicate.
261 Inhibition was noted for all concentrations of A M T > 3 mM (Fig. 3). Maximum velocity of 10 -5 g / m l E P O with o-phenylenediamine and H z O 2 after addition of 10 mM A M T was 41.1 + 4.1% (95% CI, 29.6 to 52.6%) of control Vm,x (Fig. 3). Addition of 30 mM A M T caused nearcomplete inhibition of the reaction of 10 -5 g / m l E P O with o-phenylenediamine and H 2 0 z (17.8 + 0 . 8 % of control Vm~; 95% CI, 15.7-19.9%) (Fig. 3).
Determination of EPO concentrations in eosinophils and eosinophil-conditioned medium Both eosinophil cell pellets and eosinophilconditioned medium had concentrations of E P O that could be measured with this assay. Cell pellets from unstimulated eosinophils had a baseline E P O concentration of 1.07 + 0.16 x 10 -5 g/106 cells. Subsequent activation with 10 - 6 M fMLP alone decreased pellet E P O concentration to 0.96 + 0.15 × 10 -5 g / 1 0 6 cells (92.3 ± 4.2% of control, P = 0.038, n = 11 experiments). Activation with 10 -6 M fMLP plus 5 / x g / m l CYB decreased cell pellet E P O concentration substantially, from 1.19 ± 0.32 to 0.78 ± 0.31 × 10 -5 g / 1 0 6 cells (71.7 + 16.1% of control, P = 0.043, n = five experiments) (Fig. 4).
110% 100% •=
110%" ~ . . , ~
90%
~ .i
90*/0-
80%
~
80./o-
70%
O
g
ILl
100%"
~
.
~
,
a. 70%" I,U
60% 50%
L
60%" -8 -7 -6 fMLP (log M)
50%
8 -7 fMLP (log M)
40% 35% I:: ¢D C 0 30% t~ 0
0Q fMLP fMLP+ 5 ~ag/rnlcytochalasinB
i
25% 20%
i0 k~
15%
2
C/) 0 " ILl
_2*
10% 5% O%
Control
-8 fMLP (M)
-7
-6
Fig. 5. Eosinophil peroxidase (EPO) concentrations in eosinophil-conditioned medium after stimulation with either f-Met-Leu-Phe (fMLP) or fMLP plus cytochalasin B (CYB). Activation led to concentration-dependent increases in medium EPO concentrations. Concentration is normalized to total content (medium+cell). n = 11 (fMLP) or 5 (fMLP plus CYB) experiments, each done in duplicate. *, P ~/3 × 10 -s g / m l could be determined reliably with this assay (Fig. 2). Addition of A M T inhibited the reaction catalyzed by E P O in a concentration-dependent manner (Fig. 3). Cramer et al. (1984) have demonstrated that A M T inhibited EPO activity in mixed granulocyte suspensions while inhibiting MPO derived from neutrophils only moderately. Our data confirm the ability of A M T to inhibit E P O and demonstrate a concentration-dependent effect of A M T o n VmaxEosinophil peroxidase activity could be measured both in cell suspensions (Fig. 4) and conditioned media (Fig. 5). Activation of eosinophils by either fMLP alone or fMLP plus CYB caused secretion of EPO into media (Fig. 5). These data demonstrate that secretion of small quantities of EPO by eosinophils into media can be determined in this assay and suggest that this assay may be useful in determining the activity of eosinophils. Both E P O and the MPO of neutrophils oxidize o-phenylenediamine (Strath et al., 1985). In the pure suspensions of eosinophils utilized in our experiments, contamination with leukocytes was insignificant ( < 5 % ) ; thus, cross-reaction with MPO in our assay was not a significant problem. Under conditions where significant concentrations of MPO or numbers of neutrophils exist in conjunction with E P O and eosinophils, other assay methods that minimize or abolish the activity of MPO may be required to distinguish activity of E P O from MPO, such as the oxidation of homovanillic acid by E P O but not MPO (Menegazzi et al., 1991). In contrast to the oxidation of ho-
movanillic acid, which requires measurement of oxidation by spectrofluorescence, our assay can be performed in a microplate spectrophotometer. This may allow for rapid measurement of relatively large numbers of samples in a relatively short period of time. In summary, we demonstrate a kinetic assay for determination of E P O concentrations in purified eosinophils and eosinophil-conditioned medium. This assay can determine EPO concentration >t 3 x 10 -8 g / m l , avoids the problems inherent in an end-point assay, and can be performed rapidly in a standard microplate spectrophotometer. This assay may be useful in determining eosinophil activation by precise measurement of both stored, cellular EPO in purified eosinophils and E P O secreted by isolated eosinophils.
Acknowledgements The authors thank Dr. Gerald Gleich, Professor of Medicine, Mayo Graduate School of Medicine, Rochester, MN, for the purified EPO used in these studies. This work was supported by National Heart, Lung and Blood Institute Grants HL-32495 and HL-46368. Dr. White is a recipient of Clinical Investigator Award HL-02484 from the National Heart, Lung and Blood Institute, an Edward Livingston Trudeau Fellowship from the American Lung Association, and is a fellow of the Schweppe Foundation.
References Agosti, J.M., Altman, L.C., Ayars, G.H., Loegering, D.A.,
Gleich, G.J. and Klebanoff,S.J. (1987) The injurious effect of eosinophil peroxidase, hydrogen peroxide and halides on pneumocytesin vitro. J. AllergyClin. Immunol. 79, 496. Carlson, M.G.C., Peterson, C.G.B. and Venge, P. (1985) Human eosinophil peroxidase: purification and characterization. J. Immunol. 134, 1875. Cramer, R., Soranzo, M.R., Dri, P., Menegazzi,R., Pitotti, A., Zabucchi, G. and Patriarca, P. (1984) A simple reliable assay for myeloperoxidase activity in mixed neutrophileosinophil cell suspensions: application to detection of myeloperoxidasedeficiency.J. Immunol. Methods 70, 119.
263 Kroegel, C., Yukawa, T., Dent, G., Chanez, P., Chung, K.F. and Barnes, P.J. (1988) Platelet-activating factor induces eosinophil peroxidase release from purified human eosinophils. Immunology 64, 559. Menegazzi, R., Zabucchi, G., Zuccato, P., Cramer, R., Piccinini, C. and Patriarea, P. (1991) Oxidation of homovanillic acid as a selective assay for eosinophil peroxidase in eosinophil peroxidase-myeloperoxidase mixtures and its
use in the detection of human eosinophil peroxidase deficiency. J. Immunol. Methods 137, 55. Strath, M., Warren, D.J. and Sanderson, C.J. (1985) Detection of eosinophils using an eosinophil peroxidase assay. Its use as an assay for eosinophil differentiation factors. J. Immunol. Methods 83, 209. Wever, R., Hamers, M.N., De Graaf, C.J., Weening, R.S. and Roos, D. (1982) Adv. Exp. Med. Biol. 141,501.
