Chem.-Biol. Interactions, 79 (1991) 15--30 Elsevier Scientific Publishers Ireland Ltd.

15

METABOLIC ACTIVATION OF CHLORPROMAZINE BY STIMULATED HUMAN POLYMORPHONUCLEAR LEUKOCYTES. INDUCTION OF COVALENT BINDING OF CHLORPROMAZINE TO NUCLEIC ACIDS AND PROTEINS

PETER P. KELDER a, NICOLAAS J. DE MOLa, BERT A. 'T HART b* and LAMBERT H.M. JANSSEN a

aDepartment of Pharmaceutical Chemistry and ~'Department of Pharmacognosy, Faculty of Pharmacy, Utrecht University, PO Box 80082, 3508 TB Utrecht (The Netherlands) (Received June 19th, 1990) (Revision received February 5th, 1991) (Accepted February 14th, 1991)

SUMMARY

Human polymorphonuclear leukocytes (PMNs) have been stimulated with either phorbol 12-myristate 13-acetate (PMA), calcium ionophore A23187 or a combination of both to induce the respiratory burst and myeloperoxidase (MPO) release. Chlorpromazine (CPZ) but not chlorpromazine sulfoxide (CPZSO) inhibited the respiratory burst as measured with lucigenin chemiluminescence. The inhibition was due to interference with processes in the cell leading to the respiratory burst and not to scavenging of produced oxygen radicals that provoke the luminescence. CPZ was metabolized by stimulated PMNs. HPLC analysis revealed formation of CPZSO and an unidentified product. Both products result from decay of chlorpromazine radical cation (CPZ +.), indicating formation of this radical intermediate in CPZSO oxidation by stimulated PMNs. CPZ conversion correlated with H202 production and MPO release. The largest CPZ conversion was observed with phorbol ester plus A23187 stimulation. The conversion was reduced by catalase and sodium azide, an inhibitor of MPO, with 70°7o and 40%, respectively. This indicates only partial involvement of extracellularly released MPO in CPZ metabolism by PMNs. Considerable covalent binding of [3H]CPZ to nucleic acids and proteins of intact stimulated PMNs was observed. This binding was larger upon co-stimulation with phorbol ester and Correspondence to: N.J. De Mol, Dept. Pharmaceutical Chemistry, Utrecht University, PO Box 80082, 3508 TB Utrecht, The Netherlands. Abbreviations: A23187, calcium ionophore A23187; CPZ, CPZSO, CPZ L chlorpromazine, its sulfoxide, its radical cation, respectively; HRP, horseradish peroxidase; MPO, myeloperoxidase; 02=, superoxide anion; PBS, phosphate buffered saline containing 50 ~M CaCl2; PMA, phorbol 12-myristate 13-acetate; PMN(s), polymorphonuclear leukocyte(s). *Present address: Dept. of Chronic and Infectious Diseases, ITRI-TNO, Rijswijk. 0009-2797/91/$03.50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

16

A23187. Azide did not reduce covalent binding. This indicates that covalent binding is not mediated by extracellularly released MPO and that CPZ is probably activated intracellularly. Activation of PMNs and production of H202 is a prerequisite for both CPZ conversion and covalent binding. This study demonstrates that phagocytic cells might contribute to drug metabolism and drug-induced toxicity.

Key words: Chlorpromazine -- Polymorphonuclear leukocyte -- Oxidation -DNA adducts -- Covalent protein binding -- Myeloperoxidase

INTRODUCTION

Polymorphonuclear leukocytes (PMNs) and mononuclear phagocytes constitute the main cellular defence system against microbial infections [1]. These cells are able to phagocytose invading microorganisms which are killed in phagosomes under the influence of reactive oxidizing species, especially hypochlorite [2,3]. The production and release of reactive oxygen species, like superoxide anion (02 9 and H202, and the exocytosis of certain enzymes, e.g. myeloperoxidase (MPO), are required for the microbicidal function. The generation of 02 : and the release of enzymes from granules is evoked by activation of PMN with a variety of chemotaxins and opsonins for which specific receptors are present at the PMN cell surface [1]. Activation of PMN 02 : production can be achieved in vitro with phorbol 12-myristate 13-acetate (PMA) [1,4]. PMA activates a Ca2+-activated, phospholipid dependent enzyme, protein kinase C [4], which in turn activates a membrane bound NADPH dependent oxidase [5]. This NADPH-oxidase, then, reduces O2: to O2 : at a very high turnover rate [3]. The abrupt onset of 02 = consumption and hexose monophosphate shunt activity (to produce NADPH) with the concomitant 02 =generation is known as the 'respiratory burst' [3]. An alternative way to activate NADPH-oxidase is by increased intracellular calcium concentration provoked by calcium ionophores, e.g. A23187. In this respect, A23187 acts synergistically with PMA to induce the respiratory burst [6]. Furthermore, a rise in cytosolic calcium caused by A23187 also leads to exocytosis, and MPO is among the enzymes released [7]. Upon PMN activation a large amount of oxidizing species is generated and MPO is released, which may play a role in metabolic drug activation [8,9]. One of the classes of drugs known to be susceptible for oxidation by peroxidases are phenothiazines [10]. CPZ is an important representative of the widely used phenothiazine neuroleptics. CPZ is metabolized extensively and major biotransformation routes are (a) sulfoxidation, (b) aromatic ring hydroxylation, (c) Nomono- and N-didemethylation and (d) N-oxidation, followed by sulfation or glucuronidation [11]. A relatively stable radical cation (CPZ .+) is assumed to be involved as an intermediate in the oxidation of CPZ to its sulfoxide and ringhydroxylated products, in vivo [12]. In vitro, the production of CPZ .+has been demonstrated during the oxidation of CPZ to chlorpromazine sulfoxide (CPZSO)

17

by horseradish peroxidase (HRP) [10] or hemoglobin [13] in the presence of H202. CPZ .+has been shown to bind covalently to nucleic acids [14,15] and proteins [13,14]. The production of CPZ .+ in a human cellular system may occur when this system contains a high peroxidase level and generates H202 itself. Such a cellular system is the human PMN [1]. In this study we investigated the conversion of CPZ by PMNs stimulated with either PMA or A23187, or a combination of both. The product formation suggests that CPZ .+ is formed as an intermediate. Covalent binding of [ 3H ] CPZ to PMN cell constituents like DNA and proteins were demonstrated. The role of MPO in the CPZ conversion and binding is discussed. MATERIALS AND METHODS

CPZ hydrochloride, Ca-ionophore A23187, PMA, lucigenin, xanthine, HRP (EC 1.11.1.7; activity as defined by Sigma), xanthine oxidase (EC 1.2.3.2), and catalase (EC 1.11.1.6) were purchased from Sigma Chemical Company (St. Louis, MO, U.S.A.). CPZSO was a gift from RhSne-Poulenc (Paris, France). [6,7,8,9-3H]CPZ hydrochloride (spec. act., 23.3 Ci/mmol) was from New England Nuclear (Dreieich, F.R.G.). CPZ-%perchlorate in solid form was prepared according to the method of Levy et al. [16]. The melting point (194°C) and molar absorptivity (12 100 M-1 cm-1) at 525 nm in 5 M H2SO4 were in agreement with those reported by Cheng et al. ]17]. All other chemicals were purchased from E. Merck (Darmstadt, F.R.G.). Solutions were prepared with demineralized water purified through a Millipore purification system (15 M[t water). All incubations were performed in 'phosphate buffered saline, pH 7.4' (PBS), containing 137 mM NaC1, 2.7 mM KC1, 8.1 mM Na2HP04, 1.5 mM KHePQ, and 50 ~M CaC12. Isolation of PMNs PMNs were isolated from the venous blood of healthy volunteers as described by Verburgh et al. [2]. Buffy-coats were freed from erythrocytes by dextran sedimentation, after which PMNs were isolated by Ficoll density-gradient centrifugation. The cells were stored in Hank's balanced salt solution (pH 7.4) at 4°C. Before the experiments the cells were resuspended in PBS, containing 0.1% (w/v) of gelatine to avoid cell aggregation. All experiments were performed with freshly isolated PMNs, of which the viability always exceeded 98% (determined by trypan blue exclusion). Duplicate and triplicate experiments were generally performed with PMNs from different blood donors. Chemiluminescence assay Logarithmic dilutions of either CPZ or CPZSO were included in a reaction mixture of 0.7 ml PBS, containing 100 ~M lucigenin and 16 nM PMA, whether or not with 4 ~M A23187, in flat-bottom vials (2 ml, Sterilin Ltd., Middlesex, U.K.). The vials were placed in a Picolite luminometer (Packard United Technologies,

18

Downers Grave, IL, U.S.A.) to equilibrate at 37°C under gentle stirring. Chemiluminescence production was induced by the addition of 5 × 105 PMNs (pre-equilibrated at 37°C), and luminescence was monitored during 6 s at 2-rain intervals. Where indicated, additional compounds were added to the mixture. Total luminescence during 30 min (area under the curve) was used to quantify overall 02 ~production. Controls consisted of identical mixtures in which CPZ or CPZSO were omitted. This control and an assay in the presence of 5 t~M CPZ were always performed prior to other experiments to check the 02 :-producing capacity of the PMNs. The cells were rejected if the chemiluminescence deviated by more than 30% from the average. Controls for scavenging of 02 : were performed in a cell-free system. 02 : was produced from the conversion of 0.1 mM hypoxanthine to uric acid by 0.4 mI.U./ml xanthine oxidase. Catalase (400 I.U./ml) was included to prevent effects caused by H2Q, which is formed upon dismutation of 02 -

Chlorpromazine conversion by PMNs Reaction mixtures of 0.4 ml PBS contained 5 ~M CPZ with either 16 nM PMA or 4 t~M A23187, or both. The reaction was initiated by addition of 1.0 × 107 PMNs. The incubations were performed during 30 rain at 37°C in Eppendorf cups, which were vortexed (mildly and shortly) every 3 min. After incubation, the cells were precipitated by centrifugation (10 000 rev./min, 10 s) and 360 ~1 supernatant was sampled. The cells were resuspended in a total volume of 200 #1 PBS. Aliquots of 50 #1 of the supernatant and the resuspended cell fraction were diluted each into 150 t~l acetonitrile and vigorously vortexed to extract CPZ, CPZSO and other CPZ metabolites (CPZ and CPZSO are extracted for > 99%). After centrifugation (15 000 rev./min, 5 rain) the extracts were analysed with HPLC. In controls PMA and/or A23187 were omitted. Where indicated other compounds were added to the mixture. The cell supernatants were also assayed for MPO activity. The viability of PMNs, tested with trypan blue exclusion after the various incubations (of 30 min), exceeded always 85%. In a separate experiment CPZ was incubated with 0.1 mM H202 in 2-fold diluted (PBS) supernatant, which had been separated from PMA plus A23187 stimulated PMNs. Non-cellular control mixtures in PBS consisted of 5 ~M CPZ and various of the above-mentioned additives.

Product formation from CPZ.* decay CPZ .+-perchlorate was dissolved in water and immediately added to various buffer solutions whether or not with PMNs, up to a final concentration of 5 tLM. Samples were diluted into acetonitrile, as described above, and analysed by HPLC. CPZ .+-perchlorate decay in water was minimal in the time course of the experiment.

HPLC analysis For HPLC analyses a system was built from a Model 9000-4002 (Kratos) solvent delivery system combined with a SP 8775 (Spectra Physics) autosampler

19

and a Model 757 (Kratos) absorbance detector, set at 248 nm. Samples of 50 gl were chromatographed on a 5 gm Spherisorb Cyano-column (15 cm × 4.6 mm i.d.; Phase Separation Ltd.), thermostatted at 25°C. The mobile phase consisted of a mixture of 80% (w/w) methanol and 20% (w/w) 0.06 M ammonium acetate (pH 6.5). The flow rate was 1.5 ml/min.

Irreversible binding to macromolecules in PMNs The conditions of the incubations were identical to those described under CPZ conversion, except that CPZ was replaced by [6,7,8,9-3H]CPZ (spec. act., 0.70 Ci/mmol). Other chemicals were added as indicated. After incubation the cells were precipitated by centrifugation (10 000 rev./min, 10 s) and the supernatant was separated and assayed for radioactivity and MPO activity. For radioactivity measurements a part of the supernatant was mixed with 10 ml scintillation liquid and counted in a Packard Tricarb liquid scintillation spectrometer model 4430. The cells were either lysed in acetonitrile (3-fold volume with respect to buffer) or lysed with 0.1% Triton X-100, after an additional washing. Cells lysed in acetonitrile were centrifuged. The supernatant and the washed pellet of cell constituents (after dissolution in 0.1 M NaOH) were assayed for radioactivity. The Triton lysate was extracted with phenol to separate nucleic acids from proteins [18]. The nucleic acid fraction was washed consecutively 3 times with equal volumes of chloroform and ether. The nucleic acid content was determined from the absorbance at 258 nm. The radioactivity of a part of this solution was measured as described above. The protein pellet was separated from the phenol layer and washed extensively with chloroform/phenol (1:1), PBS/chloroform (1:1), and PBS/acetonitrile (1:3). The protein pellet was dissolved in 0.1 M NaOH and a part was assayed for radioligand binding as described above. Protein content was determined according to the method of Hartree et al. [19] with bovine serum albumin as a standard. The nucleic acid fraction contained negligible amounts of protein. Using this procedure 13--18 ~g nucleic acid and 210--260 ~g protein were isolated from 107 cells. Assay for MPO MPO activity in the supernatants was determined essentially according to the method of Henson et al. [20]. Samples of 100 ~1 were assayed. RESULTS

Chemiluminescence assay Upon stimulation with PMA or PMA plus A23187 the PMNs start to produce 0 2 : which is detected by means of lucigenin chemiluminescence [21]. The total chemiluminescence induced by PMA plus A23187 is approx. 2.5 times higher than the effects induced by PMA alone (Fig. 1). Stimulation of the PMNs with A23187 alone showed a small and short-during peak of chemiluminescence (Fig. 1). The 02 : production in this case was very small. Non-stimulated cells hardly showed lucigenin chemiluminescence. CPZ has a marked inhibiting effect on the lucigenin luminescence produced by

20

6O

40 x

a. t.)

20

I

I -

•

I

0

~

5

T

•

I

10

•

I

15

•

20

I

25

•

30

time (rain) Fig. 1. Lucigenin chemiluminescence produced by stimulated PMNs. PMNs (5 × 10 r' cells) were incubated with 100 #M lucigenin in 0.7 ml PBS at 37°C. Chemiluminescence, measured as counts/rain (cpm), was induced by stimulation of PMNs with either l0 ng/ml PMA ( * ), 4 ~M A23187 ( • ), or 16 nM PMA plus 4 #M A23187 ( • ) and monitored during 6 s. The curves shown are typical curves from the same PMN batch.

PMNs stimulated with PMA, as well as with PMA plus A23187 (Fig. 2). Although differences exist in the kinetics of the lucigenin chemiluminescence between PMA and PMA plus A23187 stimulated cells (Fig. 1), the degree of reduction by CPZ is similar in both systems (Fig. 2). An inhibition of 50% was reached with CPZ concentrations of 7 t~M and 6/~M in the PMA and the PMA plus A23187 100

100

B 80

80"

60

60"

40

40-

20

20-

C

0

|

|

, , , H ,

i

10

,

,

|H

- ~

0-

#

100

. . . . . . . .

i

10

. . . . . . . .

i

v

,

100

concentration (pM) Fig. 2. Effect of CPZ and CPZSO on the lucigenin chemiluminescence produced by stimulated PMNs. Logarithmic dilutions of CPZ ( • ) or CPZSO ( o ) were included in incubations described in the legend to Fig. 1. PMNs were stimulated with either PMA alone (A) or PMA plus A23187 (B). The luminescence in the presence of CPZ or CPZSO is expressed as the percentage of the luminescence in their absence.

21

stimulated PMNs, respectively. In contrast to CPZ, CPZSO affected lucigenin luminescence much less in both systems (Fig. 2). The A23187 evoked chemiluminescence was diminished to less than 20% by 5 ~M CPZ, whereas CPZSO had no effect (not shown). In a control experiment the possible scavenging of 0 2 : by CPZ was investigated. All incubations contained catalase to prevent effects caused by H20~. In the cell-free system, in which O2-was produced by hypoxanthine and xanthine oxidase, CPZ as well as CPZSO up to a concentration of 50 ~M did not reduce the lucigenin chemiluminescence. HPLC analysis revealed that CPZ was not converted in this system (results not shown). The reduction of chemiluminescence produced in the presence of PMNs was therefore not the result of scavenging of 02 : by CPZ .+, nor was it the result of a direct interaction of CPZ -+with lucigenin. In our further experiments we used a CPZ concentration of 5 ~M which is sufficient to study CPZ metabolism and covalent binding. Reduction of 02 : production is less than 50% under these conditions. The influence on 02 : production by PMNs of other additives used in the experiments, like sodium azide, catalase and HRP, has also been investigated (results not shown). Catalase and HRP hardly had any effect on the 02 "producing capacity of PMNs in the presence of CPZ, whereas sodium azide showed a 2-fold increase of the chemiluminescence. In cell free systems lucigenin chemiluminescence is not affected by azide, or MPO with H202 and chloride [22]. Therefore, the effect of azide is due to increased 0,~ : production. This elfect of azide can be explained by inhibition of released MPO. MPO in combination with H202 and chloride inactivates (in part) NADPH-oxidase [23]. This NADPH-oxidase inactivaton is impaired by MPO inhibition and consequently leads to higher 0 2 : production.

Chlorpromazine conversion by PMNs When 5 gM CPZ was added to a resting PMN suspension (2.5 × 107 cells/ml), about 75% of the amount of CPZ was found in the cell fraction (see e.g. Table III). CPZSO showed a less pronounced affinity for the cells: about 30% of the initial amount was present in the cell fraction (results not shown). This is probably due to the more hydrophilic character of CPZSO. CPZ conversion by PMNs under various experimental conditions is shown in Table I. PMNs did not convert CPZ when not stimulated, or stimulated with A23187 alone. CPZ was converted upon stimulation of PMNs with PMA or PMA plus A23187. CPZSO is the major metabolite (Table I) but also other (unknown) products were observed with HPLC analysis (Fig. 3), of which a metabolite with a retention time of 3.0 rain showed the highest peak. This product is denoted here as compound X. Experiments using the same PMN batch showed that always more CPZ was metabolized and that consistently less CPZSO and more compound X was produced by PMNs which were co-stimulated with PMA and A23187 than by PMNs which were stimulated with PMA alone. CPZSO was not transformed when incubated with stimulated PMNs. The CPZ conversion by stimulated PMNs was reduced by sodium azide and, to an even higher extent by catalase (Table I). H R P has been added to the PMA-stimulated PMNs to

22 TABLE I

CPZ C O N V E R S I O N BY PMNS IN T H E P R E S E N C E OF V A R I O U S A D D I T I V E S Reaction conditions as described in the legend to Fig. 3. C o n c e n t r a t i o n of additives: PMA, 16 nM; A23187, 4 ~M; sodium aside, 1 mM; catalase, 400 I.U./ml; HRP, 0.1 I.U./ml; H20.~, 0.1 raM. Additives

None A23187 PMA + azide + catalase + HRP PMA + A23187 + azide + catalase Resting cells + H202 + azide

Cell supernatant c

CPZ (%)~

100 ± 1 99.5 46 62 80 49 38 67 83 98 82 72

± ± ± ± ± ± ± :~ ± ± ±

1 9 8 5 11 12 7 13 0.5 3 4

CPZSO

(%)a 0.2 0.5 32 14 10 13 24 9 8 1.5 16 15

:e ± :e ~: ± + :e ± ± ± ± +

0.5 0.5 6 2 3 4 7 2 3 0.5 1 3

Compound X

(a.u.)~, 0 0 16 11 13 73 25 6 18 0 1.5 4.3

± ± ± ± ± ± ±

5 4 1 3 8 2 3

± 0.5 ± 0.7

aAmounts of CPZ and CPZSO are expressed as percentage of initial CPZ amount and are the mean ± S.E.M. of three determinations. bAmount of compound X is expressed in arbitrary units (a.u.) and is the mean ± S.E.M. of three determinations. cCpZ was incubated with 0.1 mM H~O,~ in 2-fold diluted supernatant, which had been separated from PMA plus A23187 stimulated PMNs.

substitute for MPO, which is released less by the cells when they are not supplemented with A23187. The presence of HRP in a low amount (0.1 I.U./ml) changed the product formation: less CPZSO and much more of compound X was produced (Table I). Not intentionally stimulated (resting) PMNs to which 0.1 mM H202 had been added, showed no CPZ transformation (Table I). However, in the presence of 1 mM azide, to inhibit catalase, some CPZ was oxidized to CPZSO, and a very small amount of compound X was formed. This reaction was completed within 5 rain. Diluted cell supernatant, from incubations in which PMNs had been stimulated with PMA plus A23187, metabolized CPZ in the presence of 0.1 mM H202, and CPZSO and compound X are formed (Table I). CPZ is also oxidized by hypochlorite which is the product from the reaction of H202 with chloride catalyzed by MPO [3]. In PBS, sodium hypochlorite with a 2-fold molar excess over CPZ, instantaneously oxidized CPZ with + 50% yield of CPZSO, but compound X was not formed. As a control, CPZ was also incubated in cell-free PBS during 30 rain at 37°C. CPZ was stable under these conditions, as it was in the presence of PMA, A23187, sodium azide, catalase, HRP or H202 (up to 0.5 raM).

23

B

A 1 2

E cq

c

O ffl

I

I

0

5

I

10

I

I

15 0 time (min)

I

I

I

5

10

15

Fig. 3. HPLC chromatograms of PMN supernatants. CPZ (5 tLM) was incubated during 30 min in 0.4 ml PBS at 37°C with 107 PMNs which were not stimulated (A) or stimulated with 16 nM PMA plus 4 tLM A23187 (B). Identity of peaks: 1, CPZ; 2, CPZSO; 3, unknown compound, denoted as compound X.

Product formation from CPZ.+decay The product formation after decay of externally added CPZ .+-perchlorate in PBS, whether or not with PMNs, is presented in Table II. Apart from CPZ and CPZSO other minor products were obtained, among which compound X. In the presence of cells CPZ .+decay yielded increased amounts of CPZ and compound X and a diminished amount of CPZSO. Decay of CPZ .+-perchlorate in phosphate buffer pH 6.5 (ionic strength, 0.1 M) yielded approximately equal amounts of CPZ and CPZSO (Table II), in agreement with the report by Cheng et al. [17]. Compound X was also formed in the reaction of CPZ with H202 catalyzed by a low amount (0.1 I.U./ml) of HRP in PBS. When higher (> 10 I.U./ml) HRP amounts were used, CPZ was almost completely oxidized to CPZSO. Therefore the level of CPZ-+ is important for its decay pattern. At higher CPZ .+levels a reaction between two CPZ .+radicals yields equal amounts of CPZ and CPZSO [17], and complete conversion of CPZ results in only CPZSO. At low CPZ.+ concentration a reaction involving two CPZ .*radicals is less favourable. Therefore other decay pathways contribute, possibly leading to the formation of compound X. Compound X was collected after separation by HPLC. The absorption spec-

24 T A B L E II P R O D U C T F O R M A T I O N F R O M CPZ t - P E R C H L O R A T E D E C A Y IN V A R I O U S P H O S P H A T E BUFFERS CPZ t-perchlorate, dissolved in water, w a s directly added to the buffers up to 5 ~M, under mildly

vortexing of the solutions. Buffer

CPZ (%)a

CPZSO (%)a

C o m p o u n d X (a.u.) ~'

PBS c P B S c + P M N s (107 m1-1) Phosphate 0.10 M (pH 6.5)

50 ± 2 76 ± 2 51 ± 1

35 ± 2 8.9 ± 0.5 49 ± 1

5.5 ± 0.5 15 ± 1