ANALYTICAL

BIOCHEMISTRY

206,

273-277

(19%)

Coelenterazine Is a Superoxide Anion-Sensitive Chemiluminescent Probe: Its Usefulness in the Assay of Respiratory Burst in Neutrophils Miguel

Lucas’

and Francisca

Solano

Departamento de Bioquimica y Biologia Molecular, Hospital Universitario Facultad de Medicina, Avenida Sa’nchez Pizjua’n 4, 41009 Sevilla, Spain

Received

May

Neutrophils and other phagocytic cells, when stimulated, produce reactive oxygen metabolites which play a key role in the killing of microorganism (1). The oxidants generated are hydroxyl radical, hydrogen perox-

0003.2697/92 Copyright All rights

Macarena,

4, 1992

The oxidation of free coelenterazine by superoxide anion was analyzed and compared to the oxidation by the semisynthetic photoprotein obelin, prepared by incorporation of synthetic coelenterazine into apoobelin. The oxidation of bound coelenterazine was triggered upon binding of calcium to the reconstituted photoprotein. The oxidation of free synthetic coelenterazine, in the absence of the apoprotein, was triggered by superoxide anion. The production of reactive oxygen metabolites by fMet-Leu-Pheand 4b-phorbol lab-myristate 13a-acetate-stimulated neutrophils was studied by means of the luminescence of synthetic coelenterazine. The features of this chemiluminescent probe were compared with those of luminol and are summarized as follows: (a) coelenterazine-dependent chemiluminescence was inhibited by superoxide dismutase; (b) coelenterazine was as sensitive as luminol in detecting the oxidative burst of neutrophils; (c) azide failed to inhibit coelenterazine chemiluminescence; (d) in contrast with luminol, which requires the catalytic removal of hydrogen peroxide, coelenterazine chemiluminescence did not depend on the activity of cell-derived myeloperoxidase. These results indicate the usefulness of coelenterazine as a very sensitive and specific chemiluminescence probe of superoxide anion. o 1992 Academic PWS, IN.

1 To whom

Virgen

correspondence

should

$5.00 0 1992 by Academic Press, of reproduction in any form

he addressed.

ide, hypochlorite, and singlet oxygen and arise from the superoxide anion which is produced by a membranebound oxidase that catalyzes the one-electron reduction of oxygen to superoxide anion at the expense of NADPH (2). Analytical procedures to determine NADPH oxidase activity associated with respiratory burst in phagocytes include reduction of cytochrome c, scopoletin fluorescence, oxygen uptake, and chemiluminescence. Direct or luminol-dependent chemiluminescence (3) and the luminescence of pholasin, the luciferin of the mollusc PhoZas dactylus, have been applied to the estimation of reactive oxygen species (4,5). However, the usefulness of some luminescent indicators is not always clear due to a lack of specificity and interferences in the chemiluminescence reaction. Coelenterazine is the prosthetic group of coelenterate and radiolarian photoproteins, such as obelin and aequorin, and also acts as a conventional luciferin in certain mysids, decapod shrimps, copepods, squid, and fish (6-8). In the course of experiments on the interaction of synthetic coelenterazine and the apoprotein of obelin, sensitivity of free coelenterazine to oxidation by superoxide anion was discovered. Therefore, experiments seeking the possible usefulness of coelenterazine as a luminescence indicator of reactive oxygen metabolites were designed. The experiments described here show that coelenterazine is a sensitive probe for studying the production of superoxide anion by neutrophils.

MATERIAL

AND

METHODS

Cell preparation was as follows: human neutrophils were prepared from freshly venesected blood by dextran sedimentation, Ficoll-Hypaque separation, and hypo273

Inc. reserved.

274

LUCAS

AND

tonic lysis of remaining erythrocytes for 1 min (9). Cells were washed with saline and resuspended in Hepes-buffered Kreb’s medium consisting of 25 mmol/liter Hepes, 120 mmol/liter NaCl, 4.8 mmol/liter KCl, 1.2 mmol/ liter KH,PO,, 1.2 mmoltliter MgS04, 1.3 mmol/liter CaCl,, 15 mmol/liter glucose, 0.1% albumin, pH 7.4. Cells, approximately 1.5 X 10’ per milliliter, were assayed within 2 h. Cell viability, checked with either 0.25% Trypan blue or 50 pmol/liter ethidium bromide, was over 98%. Oxygen radical production was analyzed by measuring either luminol-dependent (10) or coelenterazine-dependent chemiluminescence, as indicated in the legends to the figures. Experiments were performed with a Berthold LB 9500 C luminometer conveniently modified to enable the injection of reagents via a microsyringe through a light-tight septum. All experiments were carried out at 37°C. Samples for neutrophil chemiluminescence determination were prepared by adding aliquots of the neutrophil suspension to Kreb’s-Hepes medium supplemented as indicated in the legends to the figures. Apoobelin was prepared from obelin, extracted from the hydroid Obelia geniculata by Campbell (lo), after stimulation by calcium, removal of coelenteramide, and desalting by gel filtration (11). Regeneration of the photoprotein obelin was carried out by mixing synthetic coelenterazine (see below) and apoobelin in 50 ~1 buffer consisting of 105 mM Tris-HCl, pH 7.4,0.75 mM EDTA, 5 mM mercaptoethanol, and 0.17% gelatin. Luminescence of the reconstituted photoprotein was triggered, in front of the photomultiplier, upon addition of calcium. Oxidation of coelenterazine and luminescence by superoxide anion were studied in a generating system consisting of 100 pM hypoxanthine, 8 mu/ml xanthine oxidase in 50 mM Tris-HCI, pH 7.4, containing 0.1 mM EDTA. Reaction was started by addition of hypoxanthine. Chemicals were as follows: the chemotactic peptide fMet-Leu-Phe (fMLP),’ 4b-phorbol 12b-myristate 13a-acetate (PMA), and luminol(5-amino-2,3dihydro1,4-phthalazinedione) were purchased from Sigma. Coelenterazine, 3,7-dihydro-2-(p-hydroxybenzyl)-6-(phydroxyphenyl)-8-benzylimidazo-[1,2-a]pyrazine-3one, synthesized by the method of Inoue et al. (12), was kindly provided by Dr. A. K. Campbell (Department of Medical Biochemistry, Welsh National School of Medicine, Cardiff). It was stored in the dark; samples were dried under argon and redissolved in methanol immediately before use.

’ Abbreviations l2b-myristate

used: 13a-acetate.

MLP,

fMet-Leu-Phe;

PMA,

4b-phorbol

SOLANO

RESULTS

AND

DISCUSSION

Semisynthetic photoprotein obelin was regenerated by incubation of apoobelin with synthetic coelenterazine. The activity of the regenerated photoprotein approached 30% after 30 min incubation and, after up to 24 h, to 50% of the starting activity in the native obelin sample, when saturating amounts of coelenterazine were used. The chemiluminescence reaction could be triggered by the addition of calcium and, from the parameters of the kinetic of calcium dependency, a cooperative effect was deduced. In fact, we obtained Hill plots with slope values close to 2.4 calculated from the linear regression analysis of log{ VI(M - V)} against log[Ca’+], where Vrefers to the chemiluminescence rate obtained at different free calcium concentrations ([Ca”]) and M to the maximal chemiluminescence rate obtained following the saturation with calcium. Halfmaximal rate was obtained at approximately 2.5 pM Cazc, a value which, in addition to the Hill coefficient (see above), agrees quite well with those obtained in experiments with native obelin (not shown). A striking finding in the course of experiments on obelin regeneration was that coelenterazine was oxidized by the superoxide anion, as determined in a reaction mixture consisting of hypoxanthine and xanthine oxidase. In fact, when coelenterazine was mixed with a medium containing xanthine oxidase, the addition of hypoxanthine led to a luminescent reaction which remained constant for at least 10 min. The addition of superoxide dismutase quenched the chemiluminescence reaction, which suddenly declined back to basal values. Coelenterazine chemiluminescence was prevented when superoxide dismutase was actually present in the reaction mixture (not shown). These results contrast with the observation that the apoprotein apoobelin protected the oxidation of coelenterazine, its own prosthetic group, by superoxide anion. In other words neither native obelin nor the semisynthetic photoprotein obtained by reconstitution from a mixture of apoobelin with synthetic coelenterazine was oxidized by superoxide anion. These results suggested the sensitivity of free coelenterazine to superoxide anion and its usefulness as a chemiluminescence probe. Figure 1 summarizes a comparative study of the oxidation of coelenterazine in its free form, by superoxide anion and by the apoprotein upon binding of calcium to the semisynthetic photoprotein. We further explored this property of coelenterazine in experiments with intact phagocytic cells which are known to produce superoxide among others reactive oxygen metabolites. Stimulation of human neutrophils by fMLP, in the presence of coelenterazine, resulted in the triggering of a chemiluminescent reaction sensitive to inhibition by superoxide dismutase (Fig. 2). The onset of burst activity was detected after approximately 6 s

ANALYSIS

Coelenterazios

OF

COELENTERAZINE

BY

275

SUPEROXIDE

(~1)

FIG. 1. Kinetics of the oxidation of free coelenterazine and synthetic photoprotein obelin: Triangles refer to the rate of oxidation of synthetic obelin upon triggering by calcium ions. Circles refer to the rate of oxidation of free coelenterazine by superoxide anion. The chemiluminescence was analyzed as described under Material and Methods. The concentration of free coelenterazine in superoxide anion experiments ranged from 6 to 1200 nM. The molarity of coelenterazine in the synthetic obelin experiments, close to l-500 nM, was calculated from the coelenterazine concentration in the obelin regeneration experiments, considering a recovery of 50% of the original activity. Luminescence rate values were normalized as percentage of maximal, which approached 2.5 X lo6 counts per 10 s in calcium experiments and 3 x lo5 counts per 10 s in superoxide experiments.

and half-maximal response was reached at nearly 18 s. The sensitivity to superoxide dismutase was observed when the enzyme was added to the incubation mixture either before the stimulus or thereafter, when the respiratory burst was already triggered. Compared to the response to fMLP, that to PMA was quite different with a broadened curve and a delayed response which depended on PMA concentration (not shown). The curves

a A!h t-f---d + 0

OXIDIZATION

1 2 time (min)

I

0

I

I

I

time (min)

6

FIG. 3. Coelenterazine chemiluminescence upon stimulation of neutrophik by fMLP and PMAr 3 X lo4 neutrophils were incubated in 0.5 ml Kreb’s medium, pH 7.4, containing 0.5 pM coelenterazine. Neutrophils were stimulated (see arrowhead) by injecting either 0.1 pM fMLP (trace a) or 10 nM PMA (trace b).

in Fig. 3 are representative of the different response of human neutrophils to fMLP and PMA. For comparative purpose, oxygen radical production by PMA-stimulated neutrophils was monitored by measuring luminol-dependent and coelenterazine-dependent chemiluminescence (Fig. 4). The most striking feature regarding Fig. 4 is that 100 pM azide inhibited luminol-dependent luminescence by nearly 96%, in contrast to a minor effect on coelenterazine-dependent luminescence. However, the profile of luminescent signals in response to PMA was quite different. In fact, in the presence of coelenterazine, the activation of neutrophils by PMA led to an almost immediate response which

3

FIG. 2. Effect of superoxide dismutase on coelenterazine-dependent chemiluminescence in fMLP-stimulated neutrophils: 3 X 10’ neutrophils were incubated in 0.5 ml Kreb’s medium, pH 7.4, containing 1 pM coelenterazine. After 2 min preincubation at 37°C neutrophils were stimulated (see arrows) with 0.2 pM fMLP. Trace b indicates luminescence rate when 20 ag/ml superoxide dismutase was present in the assay mixture. Trace c indicates the quenching of chemiluminescence following the addition of superoxide dismutase to neutrophils after the oxidative burst was triggered as in trace a.

FIG. 4. Effect of azide on luminoland coelenterazine-dependent chemiluminescence in PMA-stimulated neutrophils: 3 X 10’ neutrophils were incubated in 0.5 ml Kreb’s medium supplemented with 10 pM luminol (a,b), 0.5 pM coelenterazine (cd), and 100 FM sodium azide (b,d). PMA, 50 nM, was added as indicated by the arrows.

276

LUCAS

AND

reached maximal intensity after 80 s and declined back to initial values after 4 min, whereas in the presence of luminol, maximal luminescence rate was achieved after 4 min and it took nearly 25 min to return to basal values. These differences may be explained since luminol-dependent chemiluminescence requires a catalyst, namely cell-derived myeloperoxidase, which is inhibited by azide (13,14). The rate of coelenterazine luminescence under stimulation of neutrophils by fMLP was further tested keeping in mind the effect of cytochalasin B and azide. Coelenterazine luminescence was strongly enhanced by cytochalasin B but the shape of the curve was quite similar to that obtained in the absence of cytochalasin B. Azide failed to inhibit significantly coelenterazine-dependent chemiluminescence in fMLP-stimulated neutrophils, but it was inhibited by nearly 25% when the stimulation by fMLP was enhanced by the presence of cytochalasin B (see Fig. 5). The inhibition produced by azide under these conditions is not easy to explain, unless the involvement of a singlet oxygen which may be scavenged by azide is assumed (15). The salient point is the high degree of inhibition produced by azide when the oxidative burst of human neutrophils was assayed by means of luminol-dependent chemiluminescence (Fig. 6), illustrating the well-known catalytic requirement, myeloperoxidase, which couples the removal of hydrogen peroxide to light production by luminol oxidation. Despite its high sensitivity, luminol is not a probe specific to the oxidants produced during the phagocytic burst, although in combination with azide and horseradish peroxidase, it is an probe specific to detecting hydrogen peroxide production by neutrophils (16). When human neutrophils were stimulated by either fMLP or PMA, the course of oxidative burst could be monitored by the coelenterazine-dependent light emission. Coelenterazine was, on an equimolar basis, as sensitive as luminol in detecting reactive oxygen metabo-

SOLANO

FIG. 6. Effect of cytochulasin B and azide on luminol-dependent chemiluminescence in fMLP-stimulated neutrophils: 3 X lo4 neutrophils were incubated in 0.5 ml Kreb’s medium, pH 7.4, containing 10 pM luminol. Test tubes were supplemented as follows: 5 ag/ml cytochalasin B plus 100 pM sodium azide and 4 U/ml horseradish peroxidase (a), 5 ag/ml cytochalasin B (b), 100 pM sodium azide and 4 U/ml horseradish peroxidase (c), and 100 jtM sodium azide (d). Traces a and b show the stimulation by cytochalasin B whereas traces c and d show the inhibition by azide of fMLP-induced chemiluminescence in the presence (a,c) and in the absence (b,d) of exogenously added peroxidase. Arrows indicate the addition of 0.2 pM fMLP. The response to fMLP alone, i.e., with no additional compounds in the test tubes, was almost identical to that shown in trace d.

lites production by human neutrophils. Moreover, coelenterazine-dependent chemiluminescence offers some advantages in experimental designs to follow the activity of NADPH oxidase in intact cells. First, it seems to be rather specific to superoxide anion, as may be deduced from the almost 100% inhibition by superoxide dismutase. Second, the shape of the curves reflects the instantaneous production of superoxide in as much it is not dependent on the release of a catalyst, i.e., myeloperoxidase, as may be deduced from the lack of effect of azide. Therefore the coelenterazine assay does not require extra reagents, i.e., azide and exogenous peroxidase, which could hinder the normal response of neutrophils and the interpretation of the respiratory burst under some experimental conditions. These results indicate the usefulness of synthetic coelenterazine as a chemiluminescent indicator for assaying superoxide production by neutrophils. ACKNOWLEDGMENTS This work was supported by Grant 9210399 from the Fondo de Investigaciones Sanitarias de la Seguridad Social. The author thanks Dr. A. K. Campbell (Department of Medical Biochemistry, University of Wales College of Medicine, CardiE) for providing the obelin and synthetic coelenterazine.

FIG. 5. Effect of cytochnlasin B and azide on coelenterazine-dependent chemiluminescence in fMLP-stimulated neutrophils: 3 X 10’ neutrophils were incubated in 0.5 ml Kreb’s medium, pH 7.4, containing 0.5 PM coelenterazine and supplemented with 5 fig/ml cytochalasin B (a,b) and 100 pM sodium axide (b,d). fMLP, 0.2 pM, was added as indicated by the arrows.

REFERENCES 1. Badwey, J. A., and Karnovsky, 49, 695-726. 2. Babior, Invest.

B. M., Kipnes, 52,741-744.

M. L. (1980)

R. S., and Curnutte,

Annu.

Rev. Biochem.

J. T. (1973)

J. Clin.

ANALYSIS

3. Jenner, D. E., Holt, M. E., and Campbell, min. Cemilumin. 1, 165-171.

OF

COELENTERAZINE

A. K. (1987)

OXIDIZATION

J. Biolu-

10. Campbell,

4. Roberts, P. A., Knight, J., and Campbell, A. K. (1987) Anal. Bio&em. 160, 139-148. 5. Miiller, T., Davies, E. V., and Campbell, A. K. (1989) J. Bidumin. Chemilumin. 3, 105-113. 6. Shimomura, O., Inoue, S., Johnson, F. H., and Haneda, Y. (1980) Comp. Biochem. Physiol. B. 65,435-437. 7. Campbell, A. K., Patel, (1988) Biochem. J. 252, 8. Shimomura, O., Musicki,

A. K., Razavi, Z. S., and McCapra, 143-149. B., and Kishi, Y. (1989) Biochem.

261,913-920. 9. Patel, A. K., Hallet, J. 248,173-180.

M. B., and Campbell,

A. K. (1987)

Biochem.

F. J.

11. Shimomura,

BY

277

SUPEROXIDE

A. K. (1974)

Biochem.

O., and Johnson,

J.

143,411-418.

F. H. (1975)

12. Inoue, S., Sugiura, S., Kakoi, H., Hasizuno, H. (1975) Chem. Lett., 141-144. 13. Klebanoff,

S. J. (1970)

Science

14. DeChatelet, L. R., Long, Thomas, M. J., Henderson, Zmmunol. 129,1589-1593.

Nature

266,236-238.

K., Goto,

T., and Iio,

169,1095-1097.

G. D., Shirley, P. S., Bass, D. A., F. W., and Cohen, M. S. (1982) J.

15. Hasty, N., Merckel, P. B., Radlick, Tetrahedr. Lett., 49-56. 16. Wymann, M. P., von Tscharner, giolini, M. (1987) Anal. B&hem.

P., and Kearns, V., Deranleau,

165, 371-378.

D. R. (1972)

D. A., and Bag-