946
Chem. Res. Toxicol. 1998, 11, 946-952
Phototoxicity Associated with Diclofenac: A Photophysical, Photochemical, and Photobiological Study on the Drug and Its Photoproducts Susana Encinas, Francisco Bosca, and Miguel A. Miranda* Departamento de Quı´mica/Instituto de Tecnologı´a Quı´mica UPV-CSIC, Universidad Polite´ cnica de Valencia, Avenida de los Naranjos s/n, 46022 Valencia, Spain Received April 2, 1998
Diclofenac (1) is a photosensitizing nonsteroidal antiinflammatory drug. Its photodecomposition gives rise to chlorocarbazole 2a. This product undergoes photodehalogenation to 3a in a subsequent step. When the photobiological activities of 1, 2a, and 3a are compared by means of the photohemolysis test, it is clearly observed that chlorocarbazole 2a causes cell lysis with a markedly higher efficiency than the parent drug or the secondary photoproduct 3a. Laser flash photolysis studies suggest that photodehalogenation of 2a occurs from its excited triplet state via quenching by ground-state 2a and formation of an excimer. As a consequence, an aryl radical plus an N-centered carbazolyl radical are formed. These radical intermediates appear to be responsible for the observed photobiological effects of diclofenac, via hydrogen abstraction from the target biomolecules, which initiates a type-I photodynamic effect. The efficient peroxidation of model lipids, such as linoleic acid, photosensitized by 2a are in favor of this proposal. Thus, the photosensitizing properties of diclofenac appear to be associated with the photochemical and photobiological activity of its major photoproduct.
Introduction Many drugs are known to induce phototoxic responses after either systemic or topical application.1 The nonsteroidal antiinflammatory drugs (NSAID) deserve special mention due to the incidence of photosensitivity disorders, which is higher than with other types of drugs (1-4). In this context, diclofenac, which is a potent NSAID therapeutically used in inflammatory and painful diseases of rheumatic and nonrheumatic origin, can give rise to photosensitivity disorders (5, 6). The experimental results obtained from photopatch tests studies (7), as well as using the in vivo mouse tail technique, have confirmed the in vivo phototoxic properties of this drug (8). In vitro phototoxicity assays, such as photohemolysis, photobasophil-histamine-release, and Candida albicans growth inhibition (9, 10), have also evidenced the photobiological properties of diclofenac. It is obvious that a detailed knowledge of the photophysical and photochemical properties of a photosensitizing drug is essential to the understanding of its mechanism of action. In this connection, photodegradation of diclofenac (1, 2-(2,6-dichloroanilino)phenylacetic acid) has been the subject of a previous study in which 1-(8-chlorocarbazolyl)acetic acid (2a) and 1-(carbazolyl)acetic acid (3a) were detected as the major photoproducts (11). However, the photodechlorination mechanisms of diclofenac and its photoproduct 2a have not been studied in depth. Another point that deserves attention is the possible contribution of 2a, a chlorocarbazole compound analogue to the phototoxic drug carprofen (12), to the overall 1 Abbreviations: carprofen (CPF), 1,3-cyclohexadiene (CHD), fluorescence quantum yield (φf), nonsteroidal antiinflammatory drug (NSAID), phosphate-buffered saline (PBS), red blood cells (RBC), sodium dodecyl sulfate (SDS), triplet lifetime (τT).
photochemical damage induced by diclofenac. Previously, free radicals directly arising from photodegradation processes of 1 were speculated to be responsible for the in vivo photosensitivity responses (11). The present study deals with the photodegradation of 1, 2a, and 3a and the possible implications in the phototoxic activity. Different photophysical, photochemical, and photobiological data have been obtained to gain further insight into the molecular basis of the observed photosensitizing side effects. These studies have included fluorescence spectroscopy, laser flash photolysis, and photoproduct analysis as well as in vitro phototoxic assays such as photohemolysis and photodynamic lipid peroxidation. The results obtained strongly suggest that the key process in photosensitization by diclofenac is cleavage of the carbon-halogen bond of photoproduct 2a from its excited triplet state, which leads to free radicals. These highly reactive species appear to be capable of producing extensive type-I photodynamic biological damage.
Experimental Procedures Chemicals. Diclofenac (2-(2,6-dichloroanilino)phenylacetic acid) monosodium salt), carprofen (CPF), linoleic acid, and sodium dodecyl sulfate (SDS) were provided by Sigma (St. Louis, MO). 1,3-Cyclohexadiene (CHD) was purchased from Merck (Darmstadt, Germany). Acetonitrile, absolute ethanol, and methanol (HPLC grade) were from SDS (Peypin, France). All other chemicals were of reagent grade. Phosphate-buffered saline (PBS) consisted of 0.05 M phosphate buffer and 0.136 M NaCl solution (pH 7.2). Instrumentation. For HPLC/MS, a Waters Integrity System was used. It consisted of a Waters 996 photodiode array detector, a thermobeam mass detector, and a Waters 2690 separation module. Conventional HPLC analysis was per-
S0893-228x(98)00070-8 CCC: $15.00 © 1998 American Chemical Society Published on Web 07/30/1998
Phototoxicity of Diclofenac formed on a Hitachi apparatus equipped with a Spherisorb column (ODS-2, 10 µm packing), a L-6250 intelligent pump, and a L-400 fixed wavelength ultraviolet detector at a fixed wavelength of 280 nm. Preparative separation of the photoproducts was achieved on a Tracer HPLC instrument with a Lichrosorb column (RP-18, 7 µm packing), using methanol-water-acetic acid (70:29:1) as mobile phase. GC/MS analyses were obtained on a Varian-Saturn II instrument. UV spectra were recorded by means of a Shimadzu UV/vis scanning spectrophotometer (2101 PC) with a slit width of 5 nm. Fluorescence measurements were made on a Hitachi F-2000 fluorescence spectrofluorimeter. The laser flash photolysis system was based on a pulsed Nd: YAG SL404G-10 Spectrum Laser Systems apparatus, using 266 nm as excitation wavelength. The single pulses were ca. 10 ns duration and the energy was ca. 20 mJ/pulse. A Lo255 Oriel xenon lamp was employed as detecting light source. The laser flash photolysis apparatus consisted of the pulsed laser, the Xe lamp, a 77200 Oriel monochromator, an Oriel photomultiplier (PMT) system made up of a 77348 side-on PMT tube, 70680 PMT housing, and a 70705 PMT power supply. The oscilloscope was a TDS-640A Tektronix. The output signal from the oscilloscope was transferred to a personal computer for study. Fluorescence Measurements. Fluorescence emission spectra were taken using dilute ethanol solutions of 1, 2a, 3a, and CPF at room temperature under aerobic and anaerobic (N2) conditions. The samples were excited at ca. 300 nm (absorbance ) 0.1), and emission measurements were performed in the region 315-500 nm. Fluorescence quantum yields (φf) were determined using integrated peak areas by comparison with CPF as standard (φf ) 0.067 in ethanol and N2) (12). Laser Flash Photolysis Measurements. Ethanol solutions of 1, 2a, and 3a (2 × 10-4 M) under nitrogen (bubbling 10 min) were studied by the laser flash photolysis technique at λexcitation ) 266 nm. Parallel experiments were carried out under aerobic (air) conditions to detect quenching of the transient species by oxygen. A kinetic treatment such as Stern-Volmer plot was used to get kq values for 2a and 3a, measuring the triplet lifetimes variations in the presence of oxygen. Compounds 2a and 3a were also used at 3 and 4 × 10-4 M concentrations. Photolysis Studies. Irradiations of 1 in different solvents (10-3 M, 5 mL/tube) were performed with Pyrex-filtered light from an OSRAM-HLQ 125 W medium-pressure Hg lamp located inside an immersion well photoreactor (Applied Photophysics model 3230). The photon flux incident on the tubes (measured by potassium ferrioxalate actinometry) (13) was 5 × 1016 photons s-1. Hence, samples received, on average, 2.3 J cm-2. Photolysis of 1 (1 × 10-3 M)) was studied in methanol and PBS aqueous solutions under aerobic and anaerobic conditions. Parallel experiments were performed for 2a and 3a (compound 3a only in methanolic solutions). In addition, photolyses of deaerated methanolic solutions of 2a (10-3 M) were also studied in the presence of CHD (5 × 10-3 M). Some of the irradiations described above were also performed for CPF as a chlorocarbazole reference. Photoreactions were monitored by reversed-phase HPLC using the conditions mentioned under instrumentation. In addition, the photomixtures of each photoreaction were analyzed by HPLC/MS. Isolation and Identification of the Photoproducts. Isolation of the photoproducts was carried out by reversed-phase HPLC. Compounds 2a and 3a-e (Figure 1) have been previously described and were characterized by their spectral properties (11, 14, 15). Electron impact MS and NMR were used to confirm the structures. The free carboxylic acids (2a and 3a) were partially converted to the corresponding methyl esters in methanol solutions. Alternatively, these compounds were obtained by treatment with methanol/HCl (5%) under reflux for 2 h. The structures were assigned by mass spectrometry. Compound 2a-methyl ester showed diagnostically important peaks at m/z 275 (22) [M+ + 2], 273 (65) [M+], 243 (8) [M+ + 2 - CH3OH], 241 (25) [M+ - CH3OH], 216 (26) [M+ + 2 -
Chem. Res. Toxicol., Vol. 11, No. 8, 1998 947
Figure 1. Chemical structures of diclofenac (1) and its photoproducts 2a-c and 3a-f. COOCH3], 215 (33), 214 (80) [M+ - COOCH3], 213 (100), 178 (25), 177 (20), 151 (20). The MS spectrum of 3a methyl ester showed the following fragmentation pattern: m/z 239 (100) [M+], 207 (18) [M+ - CH3OH], 180 (50) [M+ - COOCH3], 179 (57), 152 (18). As compounds 2b,c and 3f were formed in very small amounts, their structures have been assigned only by their spectral properties obtained from GC/MS analysis (MS spectrum) and from the HPLC/MS system (MS and UV spectrum). Compound 2b: m/z 231 (32) [M+ + 2], 229 (100) [M+], 202 (14) [M+ + 2 - CHO], 200 (38) [M+ - CHO]. Compound 2c: m/z 233 (16) [M+ + 2], 231 (58) [M+], 215 (34) [M+ + 2 - H2O], 213 (100) [M+ + 2 - H2O], 178 (25). Compound 3f: m/z 212 (5) [M+ + 1], 211 (60) [M+], 180 (40) [M+ - OCH3], 179 (100) [M+ - CH3OH], 152 (18). The UV spectra of 2c and 3f were similar to those of the carbazole compounds 2a and 3a,c,d. All of them showed absorption peaks at λ (log ) ca. 240 (4.6), 292 (4.3), 324 (3.7), 336 (3.7) in methanol/water (70/30). The UV spectrum of the ketone 2b was similar to that of the dechlorinated ketone 3b. Thus, 2b showed absorption peaks at λ (log ) 227 (4.6), 258 (4.2), 291 (4.3), 368 (4) in methanol/water (70/30). Photohemolysis Tests. Buffered solutions of 1, 2a, 3a, and CPF (10-4 M) were transferred to plastic cuvettes, and RBC was added so that the resultant suspensions had an optical density of approximately 0.5 at 650 nm (3.3 × 106 cells mL-1). Duplicate samples of the resulting suspensions, as well as samples containing only the drug solutions, were exposed to UVA radiation. For this purpose, a 400 W medium-pressure Hg lamp was used as the light source. The distance between the lamp and the samples was 5 cm. The photon flux incident on the cuvettes (measured as before) was 2 × 1016 photons s-1. Hence, samples received, on average, 12.9 J cm-2 in 1 h. A control experiment under the same conditions was performed in the dark. CPF was used as a reference. After an incubation period of 30 min in the dark, the samples were measured at 650 nm every 10 min of irradiation with an UV/vis spectrophotometer. Further details of the procedure can be found elsewhere (16). Photoperoxidation of Linoleic Acid by the Presence of 2a in Micellar Aqueous Solutions. Solutions of linoleic acid (10-3 M) and sodium dodecyl sulfate (SDS; 10-2 M) in PBS, containing 2a (10-5 M), were irradiated through Pyrex with a 400 W mercury lamp for 5, 8, 15, and 20 min, keeping the temperature at 37 °C by means of a thermostated bath. After irradiation, the samples were incubated for 30 min in the dark. CPF at the same concentration was used as a reference. The reaction was monitored by UV-vis spectrophotometry, following the appearance and subsequent increase of a new absorption maximum at λ ) 233 nm, due to the conjugated dienic hydroperoxides derived from linoleic acid (17). The controls were
948 Chem. Res. Toxicol., Vol. 11, No. 8, 1998
Encinas et al.
Table 1. Irradiations of 1 and 2a in Different Media photoreactiona
mass balanceb
1, MeOH, air 1, MeOH, argon 1, PBS, air 1, PBS, argon 2a, MeOH, air 2a, MeOH, argon 2a, PBS, air 2a, PBS, argon
85 87 53 43 82d 95 33 30
1
2ac
2b
2c
37 0 15 9
4
