Effect of Some Psychotropic Drugs on Luminol

Home , Chlorprothixene, Thioxanthene

Effect of Some Psychotropic Drugs on Luminol Ð Dependent Ð ț Chemiluminescence Induced by O2 , OH, HOCl Vera Hadjimitova*, Trayko Traykov, Milka Mileva and Stefan Ribarov Department of Medical Physics and Biophysics, Sofia University School of Medicine, Sofia 1431, Bulgaria, Fax: +359-2-517-176. E-mail: [email protected] * Author for correspondence and reprint requests Z. Naturforsch. 57c, 1066Ð1071 (2002); received June 6/July 15, 2002 Chemiluminescence, Free Radicals, Psychotropic Drugs We studied antioxidant activity of six neuroleptics (chlorpromazine, levomepromazine, promethazine, trifluoperazine and thioridazine) and two antidepressants (imipramine and amitriptyline) in the range of concentration of 10Ð7Ð10Ð4 m. We applied luminol-dependent chemiluminescence to test the ability of these drugs to scavenge the biologically relevant oxygen-derived species: hydroxyl radical, superoxide radical, hypochlorous acid in vitro. We found that the phenothiazines were powerful scavengers of hydroxyl and superoxide radicals. Chlorprothixene, amitriptyline and imipramine had no scavenge activity to the superoxide radical. All drugs showed a moderate scavenger effect on hypochloric anion.

α Introduction might inactivate 1-antiproteinase at the spot of Reactive oxygen radicals have been implicated inflammation and it contributes to protiolitic dam- in pathophysiology of many neurologic disorders ages though HOCl has a secondary role on bacte- and brain dysfunction (Kontos and George, 1985; ricide effect of phagocytes. Siesjo et al., 1989; Chan, 1994). The evidence The phenothiazines Ð their most widely known shows that reactive oxygen radicals are involved representative being chlorpromazine (CPZ) Ð and in brain injuries such as cerebral ischemia and re- their metabolites are distributed in many body tis- perfusion (Chan, 2001). It has been demonstrated sues reaching highest concentration in brain, lung, that approximately 2% to 5% of the electron flow liver and spleen (AHFS Drug Information 89). in isolated brain mitochondria produces superox- The common therapeutic plasma concentration of Ð these drugs is usually in the range of 0.05Ð5 µm in ide anion radicals (O2 ) and hydrogen peroxide the brain the concentration is about two times big- (H2O2) (Boveris and Chance, 1973). These con- stantly produced reactive oxygen species (ROS) ger (Krushkov and Lambev 1993). The main ther- are reduced by enzyme and nonenzyme antioxi- apeutic effect of CPZ is realized mostly at treat- dant protective systems of organism. Far more re- ment of psychic disorders. It is established active hydroxyl radicals (țOH) are generated at however that CPZ inhibits lipid peroxidation in - disturbance of the balance upon increase of the vivo (Roy et al., 1984) and in vitro (Slater, 1968). Ð CPZ and trifluoperazine (TFPZ) protect the myo- concentration of O2 and H2O2. The invasion of phagocytes at the spot of the damaged or inflamed cardial phospholipids from LP connected with tissue results in local increase of ROS concentra- ischemic tissue damages (Janero and Burghardt, tion. Together with the increased production of 1989). As far as initiation and development of LP Ð is connected with availability of ROS, the manifes- O2 and H2O2, HOCl is also produced. HOCl tation of antioxidant properties could be ascribed to the ability of drugs to scavenge ROS. Abbreviations:O2, superoxide radical; KO2, potassium Our aim was to establish the capability of a group superoxide; țOH, hydroxyl radical; ROS, reactive oxygen species; NaOCl, sodium hypochlorite; CL, chemilu- widely used psychotropic drugs to interact with minescence; CPZ, chlorpromazine; LVPZ, levomepro- ROS in vitro by means of luminol-dependent mazine; TRDZ, thioridazine; PMZ, promethazine; chemiluminescence (CL). The present paper covers TFPZ, trifluoperazine; CPX, chlorprothixene; AMI, am- the investigations on the ability of neuroleptics: itriptyline; IMI, imipramine; PBS, phosphate buffer solution; CL-SI, chemiluminescence scavenging index; LP, chlorpromazine (CPZ), promethazine (PMZ), le- lipid peroxidation. vomepromazine (LVPZ),thioridazine (TRDZ), tri-

0939Ð5075/2002/1100Ð1066 $ 06.00 ” 2002 Verlag der Zeitschrift für Naturforschung, Tübingen · www.znaturforsch.com · D V. Hadjimitova et al. · Psychotropic Drugs and Chemiluminescence in Model Systems 1067 fluoperazine (TFPZ), chlorprothixene (CPX) and control sample, drug was omitted. Each sample antidepressants: imipramine (IMI), amitriptyline was incubated for 10 min. at 37 ∞C. Then, 20 µlof Ð ț (AMI) to scavenge O2 , OH and HOCl. xanthine oxidase (100 IU/l in PBS) were added (one unit xanthine oxidase will convert 1.0 µmol of Materials and Methods xanthine to uric acid per min at pH 7.5 at 25 ∞C). Subsequently, the CL signal was measured for Chemicals and preparations 5 min. The ratio of CL in the presence and in the The following psychotropic drugs were studied: absence of the drug was termed CL scavenging phenothiazines (chlorpromazine, levomepromazi- index (CL-SI). ne, thioridazine, promethazine and trifluoperazine), three-cycled antidepressants-imipramine Luminol-dependent CL in a system of potassium and amitriptyline, and thioxanthene neuroleptic- superoxide (KO2) Ð produced superoxide (assay II) chlorprothixene. Chlorpromazine, thioridazine, The assay was carried out using 1 ml samples of promethazine hydrochlorides and trifluoperazine PBS, pH 7.4, containing 0.1 mm luminol and the dihydrochloride were obtained from Sigma Chem- drug (in control sample, drug was omitted). The ical Co. (St. Louis, MO, USA). Levomepromazine CL was measured immediately after addition of (2-methoxy-10-[3-dimethylamino-2-methylpropyl]- 20 µlKO solution. In this case the release of the phenothiazine) hydrochloride was kindly donated 2 superoxide is a fast process. Therefore, CL was by Pharmachim Co. (Sofia, Bulgaria). Chlorpro- measured using the “flash assay” option of the thixene, amitriptyline and imipramine were ob- MultiUse program, every 50 milliseconds. The tained from Sigma Chemical Co. (St. Louis, MO, ratio of CL in the presence and in the absence of USA). Sodium hypochlorite and most of the other the drug was termed CL scavenging index (CL- reagents were obtained from Sigma Chemical Co. SI). (St. Louis, MO, USA) and were of finest grade. The chemiluminescence reagent was prepared Luminol-dependent CL in a system of iron-depen- by dissolving luminol in a small amount of 0.01 m dent hydroxyl radical formation (assay III) NaOH. Then the solution was diluted to luminol concentration of 1 mm with a 50 mm phosphate One ml samples of PBS, pH 7.4, containing: 3+ buffer solution (PBS) and the pH was adjusted to 0.1 mm luminol, 0.1 mm Fe , 0.1 mm EDTA, 7.4 with 0.01 m HCl. 0.1 mm ascorbate, 1 mm H2O2 and either of tested Drugs were dissolved in PBS, pH 7.4. Their final drug at concentrations between 1 and 100 µm,ora concentrations in the samples investigated are buffer for the controls. The CL was measured shown in the Figure legends. using the “flash assay” option of the MultiUse program, every 50 milliseconds. The ratio of CL in Methods the presence and in the absence of the drug was termed CL scavenging index (CL-SI). Luminol-dependent CL was used for registration of ROS (Allen, 1975; Allen and Loose, 1976; Itoh et Luminol-dependent CL in a system of NaOCl- al., 1989). For this purpose an LKB 1251 lumino- generated hypochlorite (assay IV) meter (Bioorbit, Turku, Finland) set at 37 ∞C was used. It was connected with an AT-type computer The sample contained the following substances via serial interface and MultiUse program ver. 1.08 in 1 ml PBS: 0.1 mm luminol 0.06 mm NaOCl and (Bioorbit, Turku, Finland) was used for collecting the tested drug at concentrations between 1 and µ of the data. Four types of CL assays were used. 100 m, or a buffer for the controls. The chemiluminescence was registered after addition of NaOCl using the “flash assay” option of the Luminol-dependent CL in a system of xanthine- MultiUse program, every 50 milliseconds. The ra- xanthine oxidase-generated superoxide (assay I) tio of CL in the presence and in the absence of One ml PBS, pH 7.4, containing 0.1 mm luminol, the drug was termed CL scavenging index (CL- 1mm xanthine and the drug at concentrations as SI). indicated in the figure legends were used. In the 1068 V. Hadjimitova et al. · Psychotropic Drugs and Chemiluminescence in Model Systems

Results and Discussion gram every 50 milliseconds. The results obtained are summarized in Fig. 2a. We did not find any The effects of drugs on the luminol-dependent significant changes of CL-SI below drug concen- Ð µ CL in a system with enzymatically generated O2 trations of 10 m. At higher concentrations, CPZ, (assay I) are shown in Fig. 1. We assume that the PMZ and TFPZ slightly reduced CL-SI. At the CL ratio in the presence and in the absence of concentration of 0.1 mm, the phenothiazines de- Ð the tested drugs should show the O2 scavenging creased CL-SI 2.5 to 6.5-fold. properties of the drugs. Therefore, this ratio was Figurs 1b. and 2b. show data concerning the effect termed CL-SI (chemiluminescence scavenging in- of the CPX, IMI and AMI on the luminol-depen- dex). dent CL in a xanthine-xanthine oxidase and KO2 µ Ð At concentrations above 1 m, phenothiazines system of O2 generation. The results obtained de- decreased moderately CL-SI in a xanthine-xan- monstrate that the CPX, AMI and IMI in the sys- thine oxidase system. The effect was most distinct tems do not affect substantially the CL-response. for CPZ, while being vague for TFPZ. These re- The results obtained show that at concentrations sults possibly indicate that phenothiazines were lower than 10 µm, the investigated drugs do not ex- Ð able to scavenge O2 . hibit a detectable scavenging effect. Drug-induced decrease of the luminol-depen- The formation of țOH is achieved in most of the Ð dent CL in a system of xanthine-xanthine oxidase- O2 generating systems, in particular by activated generated superoxide radicals may be due to the phagocytes. The main determining factor of the ac- Ð well-known fact that phenothiazines are inhibitors tual toxicity of O2 and H2O2 for the cells is con- of a number of enzymes (Lullmann-Rauch and nected with the availability of metal ions, catalys- Scheid, 1975; Abdalla and Bechara, 1994; Moto- ing the formation of țOH radicals (Halliwell and hashi, 1995). There are no data showing that xan- Gutteridge, 1986). thine oxidase is among them. Nevertheless, the We investigated the ability of the selected drugs Ð ț ability of drugs to scavenge O2 was tested also in to interact with OH in a system containing the a system with potassium superoxide Ð generated Fe2+-EDTA complex (Fig. 3). In this system țOH Ð Ð O2 (assay II). In this case, the release of O2 is are formed in the water phase in which the rest of an extremely fast process, and CL was registered the participants of the system also are Ð drugs and by the “flash assay” option of the MultiUse pro- luminol. Strongest țOH scavenging properties are

Fig. 1. Effect of psychotropic drugs on the luminol- dependent CL in a system of xanthine-xanthine Ð oxidase- generated O2 (assay I). The reaction mixture contained 1 ml phosphate buffer solution (PBS), pH 7.4, 0.1 mm luminol, 1 mm xanthine and drug at concentrations as indicated. In the control sample drug was omitted. After addition of 20 µl xanthine oxidase (100 IU/l in PBS) the chemiluminescence was measured for 5 min (one unit xanthine oxidase will convert 1.0 µmol of xanthine to uric acid per min at pH 7.5 at 25 ∞C). The ratio of CL in the presence and in the absence of the drugs was termed CL-SI. V. Hadjimitova et al. · Psychotropic Drugs and Chemiluminescence in Model Systems 1069

Fig. 2. Inhibition by psychotropic drugs of the luminol-dependent CL in a system with KO2-generated superoxide radicals by phenothiazines (assay II). The sample cuvette contained 1 ml phosphate buffer solution (pH 7.4), 0.1 mm luminol and drug. In control sample the drug was omitted. The chemiluminescence was registered after addition µ of 20 lKO2 (1 mm in dimethyl sulfoxide). For CL-SI see Fig. 1.

Fig. 3. Effect of drugs investigated on the luminol- dipendent CL in a system of iron ion-dependent hydroxyl radical formation (assay III). One ml samples of phosphate buffer solution, pH 7.4, containing: 0.1 mm luminol, 0.1 mm Fe3+, 0.1 mm EDTA, 0.1 mm ascorbate, 1 mm H2O2 and either of tested drug at concentrations between 1 and 100 µm, or a buffer for the controls. The CL signal was measured for 1 min. For CL-SI see Fig. 1.

shown by TRDZ and LVPZ at concentration concentration of 0.1 mm show CL-SI 85% and 0.1 mm decreasing CL-SI about 12 and 8 times, 77%, respectively. This suggests a slight ability for respectively. PMZ reveals the least scavenger ef- interaction with țOH. Upon decrease of substance fect from the rest of the investigated phenothi- concentration, CL-SI decreased and at 1 µm for all azines. PMZ and AMI decreased the amount of substances it is in the interval 76Ð92%. Only țOH in the system about 2 times. CPX and IMI at LVPZ is an exemption Ð for it CL-SI hardly 1070 V. Hadjimitova et al. · Psychotropic Drugs and Chemiluminescence in Model Systems

Fig. 4. Effect of drugs investigated on the luminol- dipendent CL in a system of NaOCl-generated hypochlorite (assay IV). The sample contained the following substances in 1 ml phosphate buffer solution: 0.1 mm luminol 0.06 mm NaOCl and the tested drug at concentrations between 1 and 100 µm, or a buffer for the controls. The chemiluminescence was registered after addition of NaOCl using the “flash assay” option of the MultiUse program, every 50 milliseconds. For CL-SI see Fig. 1.

reached 41%. The results obtained indicate proba- et al., 1990; Suzuki et al., 1991; 1992). At concentra- bly that all investigated drugs scavenge țOH to a tion 0.1 mm all neuroleptics investigated by us Ð higher extent than O2 . CPZ (Bindoli et al., 1988; decrease CL-SI index by 2 times approximately. Breugnot et al., 1990), PMZ (Poli et al., 1989) and Concerning the antidepressants investigated, we TFPZ (Janero and Burghardt, 1989) are pointed did not find a significant interaction with OClÐ out as good antioxidants. A possible explanation (Fig. 4b). There is a little evidence available for the of these properties is the ability of the drugs to capability for a direct interaction with hypochlorite Ð ț scavenge O2 and OH radicals. of the drugs that we have investigated. α It is known that CPZ protects 1-antiproteinase We emphasize that our results are obtained from HOCl attack, which may occur upon phago- in vitro only. The value of these tests is that they cyte activation (Jeding et al., 1995). The interaction enable one to investigate the possibility of direct of the drugs with hypochlorite ions (OClÐ) mea- antioxidant effects of compounds in vivo. The fact sured by luminol-dependent CL is presented on that such effects could be possible (as shown by Fig. 4. The amount of OClÐ added to the model sys- in vitro testing) does not mean that they actually tem is considered with the amount eliminated by occur in vivo. Future studies that are underway phagocytes during phagocytosis Ð 60 µm (Gresham will examine these questions. V. Hadjimitova et al. · Psychotropic Drugs and Chemiluminescence in Model Systems 1071

Abdalla D. S. P. and Bechara E. J. H. (1994), The effect Janero D. R. and Burghardt B. (1989), Prevention of oxi- of chlorpromazine and Li2CO3 on the superoxide dis- dative injury to cardiac phospholipid by membrane- mutase and glutathione peroxidase activities of rat active “stabilizing agents”. Res. Commun. Chem. brain, liver and erythrocytes. Biochem. Mol. Biol. Int. Pathol. Pharmacol. 63, 163Ð173. 34, 1085Ð1090. Jeding I., Evans P. J., Akanmu D., Dexter D., Spencer AHFS Drug Information 89 (Eds. Mc Evoy GK) (1989), J. D., Aruoma O. I., Jenner P. and Halliwell B. (1995), American Society of Hospital Pharmacists, Inc., Characterization of the potential antioxidant and pro- Bethesta, MD, pp. 1134Ð1137. oxidant actions of some neuroleptic drugs. Biochem. Allen R. C. (1975), The role of pH in the chemilumines- Pharmacol. 49, 359Ð365. cent response of the myeloperoxidase-halide-HOOH Kontos H. A. and George E. (1985), Brown memorial antimicrobial system. Biochem. Biophys. Res. Com- lecture. Oxygen radicals in cerebral vascular injury. mun. 63, 684Ð691. Circ. Res. 57, 508Ð16. Allen R. C. and Loose L. D. (1976), Phagocytic activa- Krushkov I. M. and Lambev I. T. (1993), A Reference tion of a luminol-dependent chemiluminescence in Book in Pharmacotherapeutics, Medicina i Fizkult- rabbit alveolar and peritoneal macrophages. Biochem. ura, Sofia. Biophys. Res. Commun. 69, 245Ð252. Lullmann-Rauch R. and Scheid D. (1975), Intraalveolar Bindoli A., Rigobello M. P., Favel A. and Galzigna L. foam cells associated with lipidosis-like alterations in (1988), Antioxidant action and photosensitizing ef- lung and liver of rats treated with tricyclic psycho- fects of three different chlorpromazines. J. Neu- tropic drugs. Virchows Arch. B. Cell Pathol. 19, rochem. 50, 138Ð41. 255Ð268. Boveris A. and Chance B. (1973), The mitochondrial Motohashi N. (1995), Phenothiazines and calmodulin. generation of hydrogen peroxide. General properties Anticancer Res. 11, 1125Ð1164. and effect of hyperbaric oxygen. Biochem. J. 134, Poli G., Cheeseman K. H., Biasi F., Chiarpotto E., Dian- 707Ð716. zani M. U., Esterbauer H. and Slater T. F. (1989), Pro- Breugnot C., Maziere C., Salmon S., Auclair M., Santus R., methazine inhibits the formation of aldehydic pro- Morliere P., Lenaers A. and Maziere J. C. (1990), Phe- ducts of lipid peroxidation but not covalent binding nothiazines inhibit copper and endothelial cell-induced resulting from the exposure of rat liver fractions to peroxidation of low density lipoprotein. A comparative CCl4. Biochem. J. 264, 527Ð32. study with probucol, butylated hydroxytoluene and vi- Roy D., Pathak D. N. and Singh R. (1984), Effects of tamin E. Biochem. Pharmacol. 40, 1975Ð1980. chlorpromazine on the activities of antioxidant en- Chan P. H. (1994), Oxygen radicals in focal cerebral isch- zymes and lipid peroxidation in the various regions of emia. J. Brain Pathol. 4,59Ð65. aging rat brain. J. Neurochem. 42, 628Ð633. Chan P. H. (2001), Reactive oxygen radicals in signaling Siesjo B. K., Agardh C. D. and Bengtsson F. (1989), Free and damage in the ischemic brain. J. Cereb. Blood radicals and brain damage. Cerebrovasc. Brain Metab. Flow Metab. 21,2Ð14. Rev. 1, 165Ð211. Gresham H. D., Zheleznyak A., Mormol J. S. and Brown Slater T. F. (1968), The inhibitory effects in vitro of phe- E. J. (1990), Studies on the molecular mechanisms of nothiazines and other drugs on lipid-peroxidation sys- human neutrophil Fc receptor-mediated phagocytosis. tems in rat liver microsomes, and their relationship to Evidence that a distinct pathway for activation of the the liver necrosis produced by carbon tetrachloride. respiratory burst results in reactive oxygen metabo- Biochem. J. 106, 155Ð160. lite-dependent amplification of ingestion. J. Biol. Suzuki M., Asako H., Kubes P., Jennings S., Grisham Chem. 265, 7819Ð7826. M. B. and Granger D. N. (1991), Neutrophil-derived Halliwell B. and Gutteridge J. M. (1986), Oxygen free oxidants promote leukocyte adherence in postcapil- radicals and iron in relation to biology and medicine: lary venules. Microvasc. Res. 42, 125Ð138. some problems and concepts. Arch. Biochem. Bio- Suzuki M., Miura S., Suematsu M., Fukumura D., Kurose phys. 246, 501Ð14. I., Suzuki H., Kai A., Kudoh Y., Ohashi M. and Itoh K., Majima T., Edo K., Mizugaki M. and Ishida N. Tsuchiya M. (1992), Helicobacter pylori-associated (1989), Suppresive effect of 1-methyladenosine on the ammonia production enhances neutrophil-dependent generation of chemiluminescence by mouse perito- gastric mucosal cell injury. Am. J. Physiol. 263, 719Ð neal macrophages stimulated with opsonized zymo- 725. san. Tohoku J. Exp. Med. 157, 205Ð214.