Highly sensitive chemiluminescence method for determining myeloperoxidase in human polymorphonuclear leukocytes

ANALYTICAL

BIOCHEMISTRY

199,

191-196

(1991)

Highly Sensitive Chemiluminescence Method for Determining Myeloperoxidase in Human Polymorphonuclear Leukocytes Kiyoshi Uehara,*,t Kazutoshi Hori,*$

Minoru Nakano,*,l and Satoshi Koga*$

*Photon Medical Research Center, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka 431-31, Japan; tDepartment of Anesthesiology and Reanimatology, Gunma University School of Medicine, Maebashi, Gunma 371, Japan; $4th Department of Internal Medicine, Hyogo College of Medicine, Nishinomiya, Hyogo 663, Japan; and §Taiho Pharmaceutical Company Limited, Kandanishiki-cho, Chiyoda-ku, Tokyo 101, Japan

Received

June

7,1991

Myeloperoxidase activity was assayed by a chemiluminescence method, using a cypridina luciferin analog as a chemiluminescence probe, after extraction from peripheral human polymorphonuclear leukocytes. The chemiluminescence method was based on the detection of ‘0, generated by myeloperoxidase-catalyzed HOBr formation followed by the interaction of HOBr with H,O, at pH 4.5. With this method, myeloperoxidase in less than 100 polymorphonuclear leukocytes could be detected and myeloperoxidase in lo6 polymorphonuclear leukocytes would be calculated to be 14.4 pmol. Eosinophil extract, which contains eosinophil peroxidase, catalyzed ‘0, generation to a great extent, compared with the polymorphonuclear leukocyte extract at pH 4.5. Myeloperoxidase activity in extract of neutrophi1 fraction could be greatly influenced by eosinophil COntaIlllIIatiOIl.

0 1991

Academic

Press,

Inc.

It has been reported that polymorphonuclear leukocytes (PMNs)’ infiltrated in tissue (l-3) and constituted in peripheral blood (4) can be detected by measuring myeloperoxidase (MPO), a plentiful constituent of neutrophils, using conventional peroxidase assays, such as o-dianisidine and guaiacol methods. However, such dye-oxidizing methods could only permit the extraction of MPO from 1 X lo4 neutrophils and have some diffi-

r To whom correspondence ’ Abbreviations used: PMN, PMN extract; EE, eosinophil eosinophil peroxidase; SOD, methyl-6(p-methoxyphenyl) one.

should be addressed. polymorphonuclear leukocyte; PMNE, extract; MPO, myeloperoxidase; EPO, superoxide dismutase; MCLA, 2-3,7-dihydroimidazo(1,2-alpyrazin-3-

0003-2697191 $3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

culties in measuring initial rates of peroxidase activities. Thus, it is important to solve the above problems by establishing a sensitive and accurate method for detecting MPO activity. We have already reported that a cypridina luciferin analog, 2-methyl-6-(p-methoxyphenyl)-3,7-dihydroimidazo[l,2-alpyrazin-3-one hydrochloride (MCLA), reacts specifically with O,, as well as ‘02, to produce light in the visible region and emphasized that O;-induced luminescence can be quenched by superoxide dismutase (SOD), thereby leaving ‘O,-induced luminescence, even if both 0; and ‘0, are generated (5). It has been known that MPO catalyzes to produce HOBr in the presence of H,O, and Br- at acidic pH, and the HOBr produced reacts with H,O,, yielding ‘0, (6). Thus, we attempted to determine MPO activity in PMNs extract to generate ‘0,) which could be detected by a sensitive chemiluminescence method with MCLA.

MATERIALS

AND

METHODS

(1) Solubilization of MPO from neutrophils in peripheral blood. PMNs were purified from fresh heparinized

blood obtained from normal volunteers. The blood was subjected to Ficoll-Hypaque density gradient centrifugation and PMNs collected were treated by a hypotonic solution to remove contaminating erythrocytes. Resultant PMNs contained more than 95% neutrophils and O--5% eosinophils. Washed PMNs were suspended in 50 mM K-phosphate buffer at pH 6.0 containing 0.02% hexadecyltrimethylammonium bromide and then sonicated over ice in 30 consecutive 0.5-s bursts at 0.5-s intervals and at a power setting of 30 W (Branson Sonifier 191

192 25t 020-

B

2 El5-

18-

* ‘0

,15E El2* u - 9. 6. 3. 01

2 I INCUBATION

NaN3 OR NaCN

TIME,

2 min

FIG.

1. MCLA-dependent chemiluminescence in PMNE (6000 cells) systems (A). Reaction mixture contained 10 pM MCLA, 5 mM KBr, 0.5 H,Oz, 0.5 FM SOD, 20 pM desferrioxamine, PMNE (6000 cells), and 0.1 M acetate buffer at pH 4.5 in a total volume of 2 ml. The additional substances were 50 mM histidine, 1 mM NaN,, or 1 mM NaCN. The results are expressed as the single measurement of chemiluminescence in one representative experiment. Effect of the number of cells on maximal chemiluminescence intensity (Max. CI) (B) and integratedchemiluminescence intensity (ICI) during a 6-min period (C) of MCLA-dependent chemiluminescence. The reaction mixtures were the same as those given for (A), save that PMNE at a variety of concentrations was used. Reaction were carried out in H,O (line 1) or D,O (92% enriched D,O) (line 2). In these figures, 2 X 106cpm of integrated chemiluminescence intensity correspond to 1 arbitrary unit. Each value represents the mean of duplicate determinations in one representative experiment. mM

250). The sonicated sample was centrifuged at 40,OOOg for 20 min and the supernatant obtained, called PMN extract (PMNE), was used for the MCLA-dependent chemiluminescence experiments. (2) Solubilization of eosinophil peroxidase from eosinophils in peripheral blood. Eosinophils were purified from fresh heparinized blood obtained from a normal volunteer, according to the method described by GHrtner (7). The purity was more than 95%. Washed eosinophils suspended in the detergent-phosphate buffer solution were sonicated and centrifuged by the same procedure as that described for the preparation of PMNE. The clear supernatant obtained, called eosinophi1 extract (EE), was used for MCLA-dependent chemiluminescence experiments. (3) Enzymes. Human leukocytes were lysed with cethyltrimethylammonium bromide and MPO in the extracts was then purified as described by Morita et al. (8). The final product exhibited an A4JAzw = 0.86 (R, value) and had a specific activity of 150 U/mg of protein. (4) Reagents. MCLA was purchased from Tokyo Kasei Co. and was dissolved to about 60 pg/ml in doubly distilled water. This solution was stored in a l.O-ml aliquot at -40°C until needed. MCLA concentrations were based upon ed3,,= 9600 M-l cm-‘. Deuterium oxide (D,O, 99.9%) was obtained from MSD Isotopes Division of Merck Frosst Canada Inc. Bovine erythrocyte SOD was

purchased from Sigma. All other chemicals were reagent grade. (5) Assays. MPO activity was assayed at pH 7.0, 25°C in terms of A,,, during the oxidation of Guaiacol (Sigma Chemical Co.) in the presence of H,O, (9). Units were calculated according to the equation proposed by Klebanoff et al. (9). Protein was measured by the method described by Lowry et al. (10). Chemiluminescence was measured with a Luminescence Reader (Aloka BLR-301) as described previously (11,12). The standard reaction mixture for the luminescence measurements contained purified MPO (O-50 PM) or PMNE (O-6 X lo3 cells), 5 mM KBr, 10 I.~M MCLA, 20 PM desferrioxamine, 0.5 ELM SOD, and 0.5 mM H,O, in 2 ml of continuously stirred 0.1 M acetate buffer at pH 4.5. In some cases, NaN,, NaCN, or histidine was also present. In all cases, the reactions were initiated by addition of 50 ~1 of MPO (or PMNE) through a microsyringe (Hamilton Co.) and maintained at 25°C in a Luminescence Reader. The integrated chemiluminescence intensity was computed by a integrator equipped in a Luminescence Reader. RESULTS

(1) MCLA-Dependent System Addition of PMNE MCLA, 0.5 mM H,O,,

Chemiluminescence

from PMNE

to a mixture containing 10 5 mM KBr, 0.5 PM SOD, 20

PM j.&M

CHEMILUMINESCENCE

0’

I 2’

I

’

3

‘

4 KBr,

’ 6’

5

I a’

7

METHOD

I IO’

9

OF

I

mM

FIG. 2. Effect of three components on MCLA-dependent luminescence. Complete reaction mixture contained 10 pM MCLA, 5 mM KBr, 0.5 mM H,O,, 0.5 pM SOD, 20 FM desferrioxamine, PMNE (1.68 X lo4 cells), and 0.1 M acetate buffer at pH 4.5 in a total volume of 2 ml. Concentration of each components was varied: (A) MCLA, (B) H,O,, (C) KBr. Each value represents the mean of duplicate determinations in one representative experiment.

desferrioxamine, and 0.1 M acetate buffer at pH 4.5 caused a marked luminescence, which was significantly inhibited by 50 mM histidine, a scavenger of ‘0, (13), and was completely inhibited by 1 mM NaN,, a scavenger of ‘0, (14) and an inhibitor of MPO (15), or 1 I3IM NaCN, a well-known peroxidase inhibitor. These results are shown in Fig. 1. Klebanoff and Waltersdorph (16) have reported that desferrioxamine at high concentrations is an electron donor as well as a inhibitor of MPO. However, desferrioxamine at 20 PM did not exhibit inhibitory effect on the MPO activity measured by the guaiacol method. As expected, desferrioxamine at 20 PM did not quench the MPO-catalyzed luminescence, but significantly suppressed the background (a nonspecific MCLA-dependent luminescence), probably by chelating iron contaminant in the reaction mixtures. Thus desferrioxamine at low concentration was added to all reaction mixtures. In contrast to histidine at 50 IIIM, this amino acid at concentrations between 0.5 and 1 mM

MYELOPEROXIDASE

DETECTION

193

did not quench the MPO-catalyzed luminescence, indicating that histidine could be acting as a ‘0, scavenger instead of acting as an electron donor for higher oxidation states of MPO. From these results, it could be tentatively concluded that ‘0, is the product of an MPOcatalyzed reaction which elicits MCLA luminescence. If MCLA-dependent luminescence can be derived only from ‘02, it is not necessary to add SOD to quench O;-derived chemiluminescence in the system. To scavenge 0,) which is possibly generated via MPO complex III by the MPO + H,O, reaction, SOD was added to all reaction mixtures. Matheson et al. have reported that SOD to excess quenches ‘0,) probably by histidine residues in SOD protein (17). It has been proved that 0.5 pM SOD does not quench a monomole emission of ‘Ap (lo,) at 1.27 pm (5). In our experiments, the addition of SOD did not affect the MPO-catalyzed luminescence, but slightly suppressed nonspecific MCLA-dependent luminescence. Under the above experimental conditions, except that PMNEtat a variety of concentrations was used, the maximal chemiluminescence intensity and integrated chemiluminescence intensity during a 6-min period, when tested as a function of PMN number, increased linearly with increasing PMN number up to 6000 and 4000 cells, respectively (Figs. 1B and 1C). As shown in these figures, the replacement of H,O in the reaction mixture with D,O (92% enriched D,O) was enhanced by a factor of 1.3 in the maximal chemiluminescence intensity and by a factor of 2.5 in the integrated chemiluminescence intensity. For experiments in 92% enriched D,O, the apparent pH, measured with a glass electrode, was adjusted to 4.8 to give a pD of 4.5 (18). From Fig. lB, the following relation between maximal chemiluminescence intensity (y) and cell number (x) could be obtained: y = 37x.

PI

When EE was tested with respect to ‘0, generation under the same experimental conditions as those described in the legend to Fig. lB, the following relation between maximal chemiluminescence intensity (y) and cell number (x) could be obtained: y = 227x.

PI

(2) Effects of Three Components at a Variety of Concentrations on the MCLA-Dependent Luminescence With fixed concentrations of H,O,, KBr, and PMNE, maximal chemiluminescence intensity increased significantly, reached a maximum at 5 PM, and gradually decreased thereafter, with increasing MCLA (Fig. 2A).

194

UEHARA

ET

AL.

25t

‘0 20; 815L %o2 = 5-

0

IO

20 MPO,

30 PM

40

50

20 MPO,

30 PM

40

50

FIG. 3. Effect of various concentrations of purified MPO on maximal chemiluminescence intensity (A) and on integrated chemiluminescence intensity during a 6-min period (B). The reaction mixtures were the same as those in the legend to Fig. 1A save that PMNE was replaced by purified MPO. The reaction was carried out in Hz0 (line 1) or D,O (92% enriched D,O) (line 2). Each value represents the mean of duplicate determinations in one representative experiment.

The effects of various concentrations of H,O, and of KBr on MCLA-dependent luminescence were also investigated. Figure 2B demonstrates the dependence of maximal chemiluminescence intensity upon [H202] at fixed concentrations of PMNE, MCLA, and KBr, while Fig. 2C gives chemiluminescence data for a function of [KBr] at fixed concentrations of PMNE, MCLA and H,O,. On the basis of these findings and the data in Figs. 2A-2C, adequate concentrations of three factors depending upon maximal chemiluminescence intensity would be 10 PM MCLA, 0.5 mM H,O,, and 5 IIIM KBr, in the presence of PMNE from 100 to 6000 PMNs. (3) Calibration To elucidate the relationship between MPO activity and MCLA-dependent luminescence, MPO at a variety of concentrations was incubated with fixed concentra-

TABLE

1

Test of Added MPO on PMNE by the Chemiluminescence Method

Recovery

PMNE Exp.

No. 1

2

(PM)

Theoretical

Measured”

Recovery (o/o)

0 13.27 26.22

23.04 35.99

9.77 22.03 33.52

(100) 95.6 93.1

0 8.06 16.13

17.63 25.70

9.57 17.89 24.31

(100) 101.5 94.6

Added MPO” (PM)

a Each value represents the one representative experiment.

mean

of duplicate

determinations

tions of H,O,, MCLA, KBr, SOD, desferrioxamine, and acetate buffer in H,O (the standard mixture) and in D,O, and maximal chemiluminescence intensity or integrated chemiluminescence intensity during the 6-min period measured was plotted against MPO concentration, to obtain a linear function of MPO concentration with respect to each of the experiments (Figs. 3A and 3B). D,O enhanced the maximal chemiluminescence intensity by a factor of 1.3 and integrated the chemiluminescence intensityby a factor of 2.5, values which are in good agreement with those obtained with a NaOCl + H,O, system (19). The most simple procedure is considered to measure the maximal chemiluminescence intensity. This (Fig. 3A, line 1) allows the calibration of MCLA-dependent luminescence in the PMNE system in terms of MPO content, using the equation maximal light intensity of PMNE system = 5.13 X lo3 [MPO], where [MPO] represents pure MPO concentration in picomoles. From the calculated value of [MPO], the amount of MPO in 1 X lo6 PMNs could be calculated from the equation picomoles of MPO in 1 X lo6 PMNs = [MPO] X 2 X 103/known PMN number. Five healthy humans, aged 25-35 years, donated PMNs, which was assayed in terms of pmol/106 cells. The values obtained were 14.4 +- 2.8 (mean f. SD) pmol/106 cells. It has been reported that with an immunological method, PMNs contain 20 pmol of MP0/106 cells (20). (4) Recovery

in

The PMNs suspended in 50 mM K-phosphate buffer containing 0.02% hexadecyltrimethylammonium bromide at pH 6.0 were divided into three groups, and pure MPO was loaded on each of two groups, followed by the MPO extraction procedure (see Materials and Meth-

CHEMILUMINESCENCE

METHOD

OF MYELOPEROXIDASE

ods). One group, in which MPO was not loaded, was also treated by the MPO extraction procedure. These samples were assayed by the MCLA chemiluminescence method. This method gave good recovery with PMNs as shown in Table 1.

DISCUSSION

MCLA has already been used as the basis for the measurement of ‘0, generated in a purified MPO-H,O,-halide system (21) and heme-compound 13-hydroperoxy linoleic acid systems (19) in the presence of added SOD. Even though both 0; and ‘02, generated in a system, react with MCLA to emit the same light, O;-derived luminescence could easily be quenched by a catalytic amount of SOD without any effect on ‘O,-derivedchemiluminescence (5,19). The strong inhibition of MCLA-dependent luminescence by 50 mM histidine (a ‘0, scavenger) or 1 mM NaN, (a ‘0, scavenger and an inhibitor of MPO) and the effect of D,O on MCLA-dependent luminescence indicate that ‘0, elicits MCLA-dependent luminescence. The effectiveness of D,O in the MCLA-dependent luminescence, however, was very low, compared with that of monomol emission of ‘0, (‘A\,) (22), but was essentially the same as that in the MCLA-dependent luminescence of the NaOCl + H,O, system (a wellknown ‘O,-generating system). Such a low effectiveness of D,O may be attributable to the high reactivity of ‘0, with MCLA. The second-order rate constant for the MCLA + ‘0, reaction has been reported to be 2.9 X 10’ M-lS-l(s).

Singlet oxygen in PMNE (MPO)-H,O,-HBr systems could be generated by the following reactions at pH 4.5: MPO

195

DETECTION

MCLA with HOC1 is very slow, i.e., lz = 1 X lo3 M-‘s-’ (5). Thus, HOBr generated by Eq. [3] would not interfere in the ‘O,-derived luminescence. PMNs used for the present experiments contained from 0 to 5% eosinophil. It has been reported that eosinophils, when activated, generate ‘0, even at neutral pH, which could be detected by near-infrared spectrometry (24). Furthermore, eosinophils contain eosinophil peroxidase (EPO) at a concentration of 194.8 pmol/lO” cells (calculated using the molecular weight of EPO of 77,000) (25), corresponding to about 13 times that of MPO in neutrophils. The present work indicates that, in terms of MCLA-dependent luminescence (maximal chemiluminescence intensity/cell number), EE generates ‘0, to a great extent, compared with PMNE, when tested under the same experimental conditions. Judging from Eqs. [l] and [2] and 5% eosinophils in PMN preparation, about 30% of ‘O,-derived luminescence (MCLAdependent luminescence) would originate from EPO. Cramer et al. (26) have reported that with two conventional methods (guaiacol and o-dianisidine methods), 3.3% eosinophil is enough to raise the peroxidase activity of the normal PMN preparation by about 40% with respect to that in pure neutrophils.

REFERENCES 1. Bradley, G. (1982)

P. P., Priebat, D. A,, Christensen, J. Inuest. Dermatol. 78, 206-209.

2. Krawisz, J. E., Sharon, ology 87, 1344-1350.

P., and Stenson,

3. Grisham, M. B., Hernandez, J. Physiol. 251, 567-574. 4. Raush, P. G., and Moore,

W. F. (1984)

L. A., and Granger, T. G. (1975)

Blood

Gastroenter-

D. N. (1986) 46,

Am.

913-919.

5. Mashiko, S., Suzuki, N., Koga, S., Nakano, M., Goto, T., Ashino, T., Mizumoto, I., and Inaba, H. (1991) J. Biolumin. Chemilumin. 6,69-72.

H,O, + HBr

+

HOBr + H,O

[31

6. Kanofsky, J. R., Wright, J., Miles-Richardson, A. I. (1984) J. Clin. Znuest. 74, 1489-1495.

HOBr + H,O,

+

‘0, + HBr + H,O

[41

7. Glrtner,

Judging from the possible mechanism of the 2-methyl6-phenyl-3,7-dihydroimidazo[l,2-a]pyrazin-3-one + 0; reaction (23), it seems likely that ‘0, adds to MCLA, yielding a dioxetane analog that decarboxylates and protonates to an excited carbonyl compound that emits light at 465 nm during its return to ground state. This is not the only route through which MCLA-dependent luminescence can be elicited. Thus univalent oxidation and deprotonation would yield a MCLA radical which could then add 0,, giving the dioxetane analog. HOBr, as well as HOCl, would be a strong oxidant to abstract electron from MCLA. Even though HOC1 at 10 mM oxidizes 10 pM MCLA at pH 4.5, this oxidant does not elicit MCLA-dependent luminescence. The reaction of

R. D., and Rothstein,

I. (1980)

Immunology

40,

G. E., and Tauber,

133-136.

8. Morita, Y., Iwamoto, H., Aibara, S., Kobayashi, gawa, E. (1986) J. Biochem. 99,761-770. 9. Klebanoff, Methods Academic 10. Lowry, (1951) 11. Nakano, Biochem.

S. J., Waltersdorph, in Enzymology (Packer, Press, New York.

Sugioka,

K., Ushijima,

and Hase-

A. M., and Rosen, H. (1984) in L., Ed.), Vol. 105, pp. 399-403,

0. H., Rosebrough, N. J., Farr, J. Biol. Chem. 193.265-275. M.,

T.,

A. L., and Randall,

Y., and Goto,

T. (1986)

R. J. Anal.

159,363-369.

12. Nakano, M., Kimura, H., Hara, M., Kuroiwa, M., Kato, M., Totsune, K., and Yoshikawa, T. (1990) Anal. Btichem. 187,277-280. 13. Bellus, D. (1979) Adu. Photochem. 14. Hasty, N., Merkel, P. B., Radlick, Tetrahedron Lett. 1, 49-52. 15. Winterbourn, C. C., Garcia, them. J. 228,583-592.

11, 105-265. P., and Kearns,

R. C., and Segal,

D. R. (1972)

A. W. (1985)

Bio-

196

UEHARA

16. Klebanoff, S. J., and Waltersdorph, A. M. (1988) Arch. Biochem. Biophys. 264,600-606. 17. Matheson, I. B. C., Etheridge, R. D., Kratowich, N. R., and Lee, J. (1975) Photochem. Photobiol. 21, 165-171. 18. Salomaa, P., Schaleger, L. L., and Long, F. A. (1964) J. Am. Chem. Sot. 86, 1-7. 19. Koga, S., Nakano, M., and Uehara, K. (1991) Arch. Biochem. Biophys., 289,223-229. 20. Cberg, G., Lindmark, G., Moberg, L., and Venge, P. (1983) Br. J. Haematol. 55, 701-708. 21. Nakano, M. (1990) in Methods in Enzymology (Packer, L., and

ET

AL.

Glazer, A. N., Eds.), Vol. 186, pp. 585-591, Academic Press, New York/London. 22. Kanofsky, J. R. (1983) J. Bid. Chem. 258,5991-5993. 23. Kobayashi, T., Saga, K., Shimizu, S., and Goto, T. (1981) Agric. Biol. Chem. 45,1403-1408. 24. Kanofsky, J. R., Hoogland, J. Biol. Chem. 263,9692-9696.

H., Wever,

R., and Weiss,

25. Carlson, M. G. C., Peterson, C. G. B., and Venge, Zmmunol. 134, 1875-1879. 26. Cramer, P. D. R., Soranzo, M. R., Comin, A., Miotti, triarca, P. (1982) Blood 60, 323-327.

S. J. (1988) P. (1985)

J.

V., and Pa-