Photochemistry and Photobiology, 2013, 89: 1479–1489
Photooxidation Mechanism of Levomepromazine in Different Solvents ~ ero-Santiago1,2, Carmelo García*1, Virginie Lhiaubet-Vallet2, Jerome Trzcionka2, Luis E. Pin 1 and Miguel A. Miranda2 Rolando Oyola1, Karen Torres1, Jaysika Leguillu 1 2
Department of Chemistry, University of Puerto Rico at Humacao, Humacao, PR cnica de Vale ncia, Valencia, Spain Instituto de Tecnología Química UPV-CSIC, Universitat Polite
Received 25 April 2013, accepted 17 July 2013, DOI: 10.1111/php.12147
ABSTRACT Unwanted photoinduced responses are well-known adverse effects of most promazine drugs, including levomepromazine (LPZ, Levoprome® or Nozinan®). This drug is indicated in psychiatry primarily for the treatment of schizophrenia and other schizoaffective disorders. Levomepromazine’s particular sedative properties make it especially fit for use in psychiatric intensive care. Nevertheless, it is photolabile under UV-A and UV-B light in aerobic conditions resulting in the formation of its sulfoxide. The LPZ photochemistry in acetonitrile (MeCN) is completely different from that in methanol (MeOH) and phosphate buffer solutions (PBS, pH = 7.4). The major photoproduct in PBS and MeOH under aerobic conditions is levomepromazine sulfoxide (LPZSO), although the amount is considerably higher in the aqueous environment. The corresponding main photoproduct in MeCN could not be characterized. The destruction quantum yields of LPZ in PBS, MeOH and MeCN are 0.13, 0.02 and 109 M1 s1) and PIND (k = [1.8 0.1] 9 1010 M1 s1). However, the triplet lifetime was found to be solvent dependent. The shortest lived LPZ triplet is found in PBS, with s = 5.1 ls (Table 1). It should be noted, however, that this value was calculated using the Stern–Volmer analysis, the lifetime of 3LPZ* is calculated at infinite dilution, and that the maleate counterion can deactivate the triplet (2). Therefore, as previously indicated, it corresponds to the intrinsic 3LPZ* lifetime in the absence of any self-quenching effect. This is relevant for in vivo situations,
where low concentrations of LPZ are present. A much lower value of 0.75 ls has been reported for 2-methoxyphenothiazine in diglyme (40), although in this case, self-quenching processes were not completely eliminated either. An excited triplet energy of ~60 kcal mol1 was obtained for 3LPZ* from the maximum emission wavelength in the phosphorescence spectrum, close to the value reported by Saucin and Vorst [ET = 58 kcal mol1 in ethanol glass (49)]. The TD-DFT predicted values for both the energy and the maximum wavelength of the triplet state are also impressively close (Table 1). The triplet-state molar absorption coefficients (eT) of 3 LPZ*were determined using PIND as the triplet acceptor in deaerated MeOH or MeCN. As PIND is not soluble in PBS 7.4, it was assumed that the corresponding eT value is the same as that obtained for MeOH. From the respective triplet absorbances maxima, values of the order of (1.5 0.02) 9 104 M1 cm1 were calculated using Eq. (3). These values are within the range of accuracy of those reported for TR (1.8 9 104 M1 cm1 in MeCN and 1.5 9 104 M1 cm1 in MeOH) (35). As previously mentioned, the bimolecular quenching constant of PIND and 3 LPZ* is close to the diffusion limit rate constant (k 1010 M1 s1), resulting in a high probability for the energy transfer process (>95%). The intersystem crossing quantum yields were determined using TX and the lowest practical laser pulse energies to minimize nonlinear effects (e.g. photoionization). From the linear dependence of the TX triplet–triplet absorption, used as an actinometer, and of LPZ as a function of the laser energy at the excitation wavelength of 355 nm (data not shown), the /T values were calculated using Eq. (4) and are given in Table 1. These values are near 0.70, close to the corresponding value of TR (/T = 0.74). This result is very important in establishing that the major deactivation pathway of LPZ is the intersystem crossing. Furthermore, the 3LPZ* properties are not substantially affected by the solvent, as occuring in the case of 3 TR* (35). Photochemistry of LPZ-MS in PBS 7.4
Figure 3. Time domain fluorescence of LPZ in MeCN under air (──) and the corresponding lamp response (– – –).
Figure 4. Transient absorption spectra of a 0.28 mM LPZ solution in N2-saturated MeCN after a 355 nm laser pulse (Energy < 3 mJ): (●) 9.9 ls, (s) 29.7 ls, (▼) 59.4 ls and (D) 247 ls. Inset: decay curves for the transients at 480 nm.
The steady-state photolysis of LPZ in PBS 7.4 is important toward understanding the behavior of this drug in vivo. Irradiation of a 0.035 mM aqueous solution of LPZ-MS at 313 nm produces an absorption increase in the regions of 230–300 and 320–350 (Fig. 5). This concentration is about eight times smaller than the one used for the LFP experiments. This is very important because the lower the ground state concentration, the larger the lifetime of the 3LPZ*. Three isosbestic points can be identified at 221, 302 and 315 nm, in agreement with the behavior reported by Karpinska during the first 40 min of irradiation (9). Subsequent to this stage, these authors observed the disappearance of the isosbestic points, most likely because they induced the photodegradation of the LPZSO. In our case, the resulting absorption spectra of the photolyzed LPZ-MS match the absorption spectrum of the sulfoxide (LPZSO-MS, Fig. 1). The photolysis was measured under identical experimental conditions, but using fluorescence spectroscopy as probe (Fig. 6). The result shows a photoproduct with a huge emission with a maximum at 375 nm, which coincides with the emission maximum of the sulfoxide (Fig. 1). It is not surprising that the fluorescence emission of LPZSO is greater than that of LPZ because all sulfoxide derivatives normally have bigger φf-values. For instance, the same behavior is observed for CMZ (7) and CMZSO (25). The
Photochemistry and Photobiology, 2013, 89
Figure 5. Absorption spectra for the photolysis of 0.035 mM LPZ-MS in PBS (pH = 7.4) under aerobic conditions at 313 nm (A313 = 0.15). A time interval of 60 s was used to record each spectrum with t(a) = 0 s. Inset: concentration∕time plot for the photolysis of LPZ-MS in PBS 7.4 under anaerobic conditions irradiated with 313 nm (A313 = 0.15, I0 = 6.9 9 109 E s1). (●) Concentration of LPZ-MS and (▼) Concentration of LPZSO-MS during irradiation. The dashed line represents the linear regression for 0–90 s.
photodegradation of LPZ-MS follows a first order mechanism in PBS 7.4 (Fig. 5). The quantum yields for the photodegradation of LPZ-MS and photoformation of LPZSO-MS (measured at the beginning of the photoreaction) are 0.13 and 0.14, respectively. This means that 18% of 3LPZ* is converted to the corresponding sulfoxide. The presence of molecular oxygen in the reaction matrix obviously plays a key role in the LPZ photooxidation mechanism. During the steady-state photolysis of LPZ-MS in PBS, no formation of the cation radical (kmax = 566 nm) was observed under aerobic conditions (Fig. 5), contrary to the experiment under almost anaerobic conditions (Fig. 7). Under aerobic conditions, the concentration of O2 is enough to generate the LPZ+ by the electron transfer to 3LPZ* and to react further to produce LPZSO. The bimolecular rate constant of the cation radical of cyamemazine (CMZ+) with molecular oxygen is 2 9 107 M1 s1 (11). Assuming that LPZ+ has a similar rate constant, its reaction with O2 would be too fast and no LPZ+ can be observed under aerobic conditions. However, at low oxygen concentrations, no molecular oxygen is available for the consecutive reaction after the production of LPZ+. Therefore, the lifetimes of both 3LPZ* and LPZ+ increase and there should be enough time to see its absorption, considering that the lifetime of LPZ+ is over a few milliseconds under anaerobic conditions (50). The absorption at 566 nm observed for LPZ+ (Fig. 7) is similar to the reported maximum wavelength for the cation radicals of 2-methoxypromazine and LPZ (562 and 565 nm) (51,52); CPZ (kmax = 530 nm) (53), PZ (kmax = 518 nm) (48) and trifluoropromazine (TFPZ, kmax = 500 nm) (48). This observation is relevant for the formation mechanism of LPZSO and is also an evidence for the monophotonic formation of the cation radical. The same experiment under anaerobic conditions shows that LPZ-MS is photostable and no formation of the cation radical is observed (data not shown). Based on these results, it could be postulated that the cation radical is formed by an electron transfer between 3LPZ* with ground-state oxygen (3O2, Scheme 1). To confirm that LPZSO is formed by the reaction of
1485
Figure 6. Emission spectra for the photolysis of a 0.072 mM LPZ-MS solution in PBS7.4 under air conditions (A313 = 0.30). A time interval of 60 s was used to record each spectrum with t(a) = 0 s. Excitation wavelength = 313 nm; slits = 2.5 nm; integration time = 0.10 s.
Figure 7. Absorption spectra for the photolysis of 0.035 mM LPZ-MS in PBS (pH = 7.4) under almost anaerobic conditions at 313 nm (A313 = 0.15). A time interval of 60 s was used to record each spectrum with t(a) = 0 s. Inset: spectra in the 450–650 nm region.
the LPZ cation radical with molecular oxygen, and not from a reaction of singlet oxygen with the LPZ ground state, the very same photoreaction previously described was performed in the presence of NaN3, which is an excellent quencher of singlet oxygen (54). It is expected that the rate of formation of LPZSO measured at 334 nm will not change upon addition of NaN3 if singlet oxygen does not play a role in its formation. Indeed, the results are consistent with this expectation (Fig. 8). Conversely, the reaction between the cation radical of the phenothiazine derivatives and superoxide is not favored, as reported by Hovey (12). Therefore, we expect a similar behavior for the reaction between LPZ+ and superoxide. Photochemistry of LPZ in MeOH The LPZ stationary photolysis in MeOH under aerobic conditions induces an absorption decrease in the region of 200–224, 257–265 and 302–321 nm (Fig. 9). In these regions, absorption minima are observed at 210, 261 and 310 nm. However, the
1486
~ero-Santiago et al. Luis E. Pin
Figure 8. Kinetics for the photolysis of 0.035 mM LPZ-MS in PBS 7.4 under air conditions measured at 334 nm in the presence of (●) 0 mM
NaN3 and (s) 0.46 mM NaN3.
absorption increases in the wavelength regions of 225–257, 266–301 and 322–346 nm. Correspondingly, absorption maxima are obtained at 244, 277 and 335 nm. These absorption peaks are similar to the LPZSO maxima in MeOH (k = 248, 277, 295 and 333 nm). During short periods of photolysis, the observed characteristics match the absorption properties of LPZ and LPZSO. These results differ from those reported by Vargas et al., as theirs show a decay in absorbance at 254 nm, when a hypsochromic shift should have been be observed (6). Furthermore, unlike LPZ-MS photochemistry in PBS 7.4, no isosbestic points can be clearly identified in MeOH. This contrasts sharply with the five isosbestic points observed in the plot of extinction coefficient vs k for LPZ and LPZSO in this solvent. This could imply that the photochemistry in this environment is complex and other competing photoproducts are masking the properties of LPZSO. Besides, LPZSO is more photoreactive than LPZ in this solvent. To elucidate the complexity of the photochemistry of LPZ in MeOH, the same experiment was repeated as described above, but by measuring the fluorescence spectra as probe (Fig. 10). The results clearly show that the emission at 448 nm decreases, which is consistent with the photodegradation of LPZ. However, there is an increase in the fluorescence emission in the region of 325–395 nm and the maximum shifts during the course of photolysis. In the first 10 min of photolysis, the maximum is found at 373 nm, which coincides with the emission maximum of LPZSO (kmax = 375 nm). This suggests that, at short irradiation times, LPZSO formation is favored. After 10 min of photolysis, the maximum hypsochromic shifts to 355 nm and an isosbestic point appears at 412 nm. This confirms that the photochemical behavior of LPZ in this solvent is complex and several photoproducts are formed depending on photolysis time, in contrast to the behavior observed in PBS 7.4. The HPLC of the photolyzed solution shows that the major photoproduct is LPZSO and some of the minor photoproducts are N-oxide LPZ and N,S-dioxide LPZ. The corresponding absorption and emission spectra of LPZ and N-oxide LPZ are exactly the same. This behavior was also observed for the absorption and emission spectra of LPZSO and N,S-dioxide LPZ. Also, it was impossible to identify the minor photoproduct with an emission maximum near 355 nm. Although most of the major photoproducts of the bulk photolysis of LPZ in MeOH were isolated and characterized, their fluorescence
Figure 9. Absorption spectra for the photolysis of a 0.035 mM MeOH solution of LPZ under aerobic conditions at 313 nm (A313 = 0.15). A time interval of 300 s was used to record each spectrum with t(a) = 0 s.
Figure 10. Emission spectra for the photolysis of a 0.035 mM LPZ solution in MeOH under air conditions.
spectra show that they are all contaminated with the photoproduct with a maximum emission at 355 nm. It is impossible to detect this impurity using other spectroscopic techniques such as NMR and IR. The results show that LPZ is more photostable in MeOH than in PBS (pH = 7.4), but its photochemistry is more complex. This fact was confirmed by photolyzing LPZ with the same light intensity for the same time in both solvents. Under our experimental conditions, LPZ photodegraded completely after 10–15 min in the aqueous environment, whereas only 19% disappeared in MeOH. All other parameters being equal, a photodegradation quantum yield of only 0.02 can be extrapolated for LPZ in MeOH taking its value in PBS 7.4 as reference [Eq. (1)]. This value differs by a factor of 10 from the one reported by Vargas et al., which is 0.18 in the UVB region (6). Vargas et al. further reports that 80% of LPZ is converted into LPZSO, while this study found a conversion yield of only 55%. The major photoproduct of this photoreaction in MeOH is still LPZSO, but the corresponding N-oxide and N-oxide sulfoxide are detectable byproducts. These other photoproducts are not observed in PBS 7.4 because the electron pair of the nitrogen of the alkyl-amino chain
Photochemistry and Photobiology, 2013, 89 is not available, which is obviously a requirement for their formation. Photochemistry of LPZ in MeCN The LPZ stationary photolysis in MeCN under aerobic conditions has not been previously studied. This solvent has a dielectric constant of e = 37.5, similar to the value of MeOH (e = 30). Besides, the corresponding concentrations of dissolved molecular oxygen at 25°C are also similar; 10.2 mM in MeOH and 9.1 mM in MeCN. Nevertheless, the photochemistries of LPZ under the same conditions in both solvents are different (Figs. 8 and 10). In MeCN, an absorption decrease is observed at 211, 258 and 314 nm, while it increases at 228, 279 and 364 nm (Fig. 11). As a consequence, three isosbestic points form at 270, 300 and 336 nm. Furthermore, contrary to its photolysis in MeOH and PBS 7.4, the main photoproduct in MeCN is not LPZSO. In fact, the major photoproducts of the LPZ photodecomposition in MeCN could not be detected, apparently due to the retention of these compounds by the nonpolar HPLC column. The chromatograms only show the presence of LPZ and traces of LPZSO. Moreover, the corresponding N-oxide and other related derivatives were not detected either. Taking into consideration all the results of this study, it can be established that the presence of molecular oxygen does not alter the lifetime of the excited singlet state (1LPZ*). The fluorescence quantum yield is very low and, therefore, the quantum yield of formation of the excited triplet state is relatively high in all solvents tested (φT > 0.60). Furthermore, it was observed that the triplet lifetime depends more on the solvent and the molecular oxygen
*
S
hν
N R1
concentration than on any other parameter. However, the molar absorption coefficient is solvent independent. In general, LPZ is photostable under anaerobic conditions, but it photodegrades very quickly under aerobic conditions. The photochemistry of LPZ in MeCN, on the other hand, is completely different from that in MeOH and PBS (pH = 7.4) under aerobic conditions. The major photoproduct found for the photolysis of LPZ-MS in phosphate buffer and MeOH under
Figure 11. Absorption spectra for the photolysis of a 0.0345 mM solution of LPZ-FB in MeCN under aerobic conditions at k > 300 (A313 = 0.19). A time interval of 300 s was used to record each spectrum with t(a) = 0 s.
1
S N R1
OCH3
*
OCH3
3
S N R1
OCH3 O2
OH, O2
OH O S N R1
OCH3
O2, H2O
O O S N R1
O2 O2 OCH3
OH
H2O2 O S N R1
1487
OCH3 Scheme 1. Mechanism proposed for the formation of LPZSO.
S N R1
OCH3
1488
~ero-Santiago et al. Luis E. Pin
aerobic conditions is LPZSO, although the amount is considerably higher in the aqueous environment. The characterization of the corresponding main photoproduct in MeCN could not be achieved with the available separation methods. Nevertheless, only traces of LPZSO and no other oxidation products were detected for this solvent.
CONCLUSIONS To elucidate its photooxidation mechanism, the photophysics and photochemistry of LPZ and LPZ-MS were studied in different solvents, under aerobic and anaerobic conditions. The results of this work are summarized in the reaction mechanism given in Scheme 1. This scheme shows that LPZSO does not form as a result of a reaction of singlet oxygen with ground state LPZ as previously proposed by Hovey (12). Nonetheless, consistent with this author’s findings, our results show that the LPZ+ is actually formed by an electron transfer process between 3LPZ* and molecular oxygen upon irradiation with 313 nm. This proves that the cation radical of the promazine derivatives can form in three different ways: (a) in a biphotonic process with k > 300 nm, as proposed by Chignell et al. (55) and Kochevar et al. (47) for CPZ; (b) in a monophotonic process with k < 300 nm; and (c) in monophotonic process at wavelengths higher than 300 nm, provided there is an acceptor that serves as a catalyst for the electron transfer process. However, no LPZSO is formed by a singlet oxygen reaction in PBS. Moreover, the oxidation photoproducts in MeOH (N-oxide and N-oxide sulfoxide) may form by a reaction between singlet oxygen and the ground state of LPZ, but they are only minor products. The cation radical, in turn, reacts with ground-state triplet oxygen, producing a peroxyl free radical, as previously proposed by Iwaoka and Kondo for the photooxidation of CPZ in water with 254 nm (56). This last radical is then responsible for the formation of LPZSO. Acknowledgements—This work has been supported in part by NIH– MBRS grant SO6GM08216 to UPR–Humacao and the UPRH-FOPI Program (CGR). Financial support by the Spanish Government (Ramon y Cajal contract to VLV) is gratefully acknowledged. We especially thank the High Performance Computing Facility at UPR Río Piedras for assistance with the theoretical calculations and Elizabeth Hodges for the revision of the manuscript.
REFERENCES 1. Nogrady, T. (1988) Medicinal Chemistry: A biochemical approach, 2nd edn. Oxford University Press, New York. 2. Oyola, R. (2001) Design, construction and use of a nanosecond laser transient absorption spectrokinetic system for the study of the photochemical and photophysical properties of relevant biological molecules: dTpA, dApT and tricyclic antidepressive drugs. Ph. D. Thesis, University of Puerto Rico, Río Piedras. 3. Dahl, S. G. and H. Hall (1981) Binding affinity of levomepromazine and two of its major metabolites to central dopamine and a-adrenergic receptors in the rat. Psychopharmacology 74, 101–104. 4. Sivaraman, P., R. Rattehalli and M. Jayaram (2012) Levomepromazine for schizophrenia. Schizophrenia Bull. 38, 219–220. 5. Dietz, I., A. Schmitz, I. Lampey and C. Schulz (2013) Evidence for the use of levomepromazine for symptom control in the palliative care setting: A systematic review. BMC Palliat. Care 12, 1–11. 6. Vargas, F., K. Carbonell and M. Camacho (2003) Photochemistry and in vitro phototoxicity studies of levomepromazine (methotrimeprazine), a phototoxic neuroleptic drug. Pharmazie 5, 315–319.
7. Morliere, P., J. Haigle, K. Aissani, P. Filipe, J. N. Silva and R. Santus (2004) An insight into the mechanisms of the phototoxic response induced by cyamemazine in cultured fibroblasts and keratinocyte. Photochem. Photobiol. 79(2), 163–171. 8. Rasheed, A., M. Javed, S. Nazir and O. Khawaja (1994) Interaction of chlorpromazine with tricyclic anti-depressants in schizophrenic patients. J. Pakistan Med. Asoc. 44, 233–234. 9. Karpinska, J., A. Soko, A. Bernatowicz, A. Szul^ecka and U. Kotowska (2012) Studies on photodegradation of levomepromazine and olanzapine under simulated environmental conditions. Photochem. Photobiol. Sci. 11, 1575–1584. 10. Dahl, S. G. and H. Refsum (1976) Effects of levomepromazine, chlorpromazine and their sulfoxides on isolated rat atria. Eur. J. Pharm. 37, 241–248. 11. Morliere, P., F. Bosca, M. A. Miranda, J. V. Castell and R. Santus (2004) Primary photochemical processes of the phototoxic neuroleptic cyamemazine: a study by laser flash photolysis and steady-state irradiation. Photochem. Photobiol. 80, 535–541. 12. Hovey, M. C. (1982) Micellar photochemistry. Photooxidations with intramicellar-generated singlet oxygen. J. Am. Chem. Soc. 104, 4196–4202. 13. Thankamoniamma, M., M. Narayanapillai, A. M. Braun and E. Oliveros (2012) Self sensitized photooxidation of n-methyl phenothiazine: acidity control of the competition between electron and energy transfer mechanisms. Photochem. Photobiol. Sci. 11, 1744– 1755. 14. Xiang, B. and L. Kevan (1994) Photooxidation of phenothiazine derivatives in silicas of different pore sizes. Langmuir 10, 2688–2693. 15. Nath, S. and A. V. Sapre (2001) Photoinduced electron transfer from chloropromazine and promethazine to chloroalkanes accompanied by cleavage of C-Cl bond. Chem. Phys. Lett. 344, 138–146. 16. Cheng, H. Y., P. H. Sackett and R. L. McCreery (1978) Kinetics of chlorpromazine cation radical decomposition in aqueous buffers. J. Am. Chem. Soc. 100, 962–967. 17. Rosenthal, I. and T. Bercovici (1977) Dye-sensitized photooxidation of chlorpromazine. J. Heterocyclic Chem. 14, 355–357. 18. Galzigna, L., M. P. Schiappelli and M. Scarpa (1997) A peroxidase-catalyzed sulfoxidation of promethazine. Free Radic. Res. 27, 501–504. 19. Baciocchi, E., T. Del Giacco, O. Lanzalunga, A. Lapi and D. Raponi (2007) The singlet oxygen oxidation of chlorpromazine and some phenothiazine derivatives. Products and reaction mechanisms. J. Org. Chem. 72, 5912–5915. 20. Baciocchi, E., T. Del Giacco and A. Lapi (2004) Oxygenation of benzyldimethylamine by singlet oxygen. Products and mechanism. Org. Lett. 6, 4791–4794. 21. Baciocchi, E., T. Del Giacco and A. Lapi (2006) Quenching of singlet oxygen by tertiary aliphatic amines. Structural effects on rates and products. Helv. Chim. Acta 89, 2273–2280. 22. Hall, R. D., G. R. Buettner, A. Motten and C. F. Chignell (1987) Near-infrared detection of singlet molecular oxygen produced by photosensitization with promazine and chlorpromazine. Photochem. Photobiol. 46, 295–300. 23. Petrushenko K. B., A. I. Vokin, V. K. Turchanino an and Y. L. Frolov (1983) Energy degradation processes of the electron excitation of the phenothiazine molecule. Bull. Acad. Sci. USSR Div. Chem. Sci.(Engl. Transl.) 32, 2151–2152. 24. Demas, J. N. and G. Crosby (1971) The measurement of photoluminescence quantum yields: a review. J. Phys. Chem. 75, 991–1024. 25. Bosca, F., P. Moliere, M. A. Miranda, J. Castell and R. Santus (2006) Primary steps of the photochemical reactions of 2-cyano-10(3-[dimethylamino, n-oxide]-2-methylpropyl)-5-oxide-phenothiazine, the photoproduct of cyamemazine, a phototoxic neuroleptic: comparison with the sulfoxide. Photochem. Photobiol. Sci. 5, 336–342. 26. Loennechen, T. and S. G. Dahl (1990) High-performance liquid chromatography of levomepromazine (methotrimprazine) and its main metabolites. J. Chromatography A 503, 205–215. 27. Pi~nero, L. E. (2006) Synthesis, characterization, photophysics, and photochemistry of 2-chlorinated phenothiazine derivatives in several solvents. Master Thesis, UPR-Río Piedras, Puerto Rico. 28. García, C., L. Pi~nero, R. Oyola and R. Arce (2009) Photodegradation of 2-chloro substituted phenothiazines in alcohols. Photochem. Photobiol. 85, 160–170. 29. Tosa, M., C. Paizs, C. Majdik, L. Poppe, P. Kolonits, J. A. Silberg, L. Novak and F. Irimie (2001) Selective oxidation methods for
Photochemistry and Photobiology, 2013, 89
30.
31. 32.
33.
34. 35.
36.
37. 38. 39. 40.
41.
42. 43.
44.
preparation of N-alkylphenothiazine sulfoxide and sulfones. Heterocycl. Commun. 7(3), 277–282. Tosa M., C. Paizs, C. Majdik, L. Novak, P. Kolonits, F. Irimie and L. Poppe (2002) Synthesis of optically active 3-substituted-10-alkyl10H-phenothiazine-5-oxide by enantioselective biotransformations. Tetrahedron: Asymmetry 13, 211–221. Jouyban, A. and B. H. Yousefi (2003) A quantitative structure property relationship study of electrophoretic mobility of analytes in capillary zone electrophoresis. Comp. Biol. Chem. 27, 297–303. Amovillia C., V. Barone, R. Cammic, E. Cancesd, M. Cossib, B. Mennuccia, C. S. Pomellie and J. Tomasi (1998) Recent advances in the description of solvent effects with the polarizable continuum model. Adv. Quantum Chem. 32, 227–261. Charaf-Eddin A., A. Planchat, B. Mennucci, C. Adamo and D. Jacquemin (2013) Choosing a functional for computing absorption and fluorescence band shapes with TD-DFT. J. Chem. Theory Comput. 9, 2709–2760. Kuhn, H. J., S. E. Braslavsky and R. Schmidt (1989) Chemical actinometry. Pure Appl. Chem. 61(2), 187–210. García, C., R. Oyola, L. E. Pi~nero, R. Arce, J. Silva and V. Sanchez (2005) Substitution and solvent effect on the photophysical properties of several series of 10-alkylated phenothiazine derivatives. J. Phys. Chem. A 109, 3360–3371. Lhiaubet-Vallet, V., N. Belmadoui, M. J. Climent and M. A. Miranda (2007) The long-lived triplet excited state of an elongated ketoprofen derivative and its interactions with amino acids and nucleosides. J. Phys. Chem. B 111, 8277–8282. Bensasson, R. and E. J. Land (1971) Triplet-triplet extinction coefficient via energy transfer. Trans. Faraday Soc. 67, 1904–1915. Cosa, G. (2004) Laser techniques in the study of drug photochemistry. Photochem. Photobiol. 80, 159–174. Merkel, P. and J. P. Dinnocenzo (2008) Thermodynamic energies of donor and acceptor triplet states. J. Photochem. Photobiol. A 193, 110–121. Bensasson, R., C. R. Goldschmidt, E. J. Land and T. G. Truscott (1978) Laser intensity and the comparative method for determination of the triplet quantum yield. Photochem. Photobiol. 28, 277–281. Allonas, X., C. Ley, C. Bibaut, P. Jacques and J. Fouassier (2000) Investigation of the triplet quantum yield of thioxanthone by timeresolved thermal lens spectroscopy: solvent and population lens effects. Chem. Phys. Lett. 322, 483–490. Yip, R. W., A. G. Szabo and P. K. Tolg (1973) Triplet state of ketones in solutions. Quenching rate studies of thioxanthenone triplets by flash absorption. J. Am. Chem. Soc. 95(13), 4471–4472. Aaron, J., M. Maafi, C. Kersebet, C. Parkanyi, A. M. Shafik and N. Motohashi (1996) A solvatochromic study of new benzo [a] phenothiazines for the determination of dipole moments and specific solute-solvent interactions in the first excited singlet state. J. Photochem. Photobiol. A 101, 127–136. Karpinska, J., A. Sokol and M. Skoczylas (2008) An application of UV-derivative spectrophotometry and bivariate calibration algorithm
45.
46. 47.
48.
49. 50. 51.
52. 53.
54. 55.
56. 57. 58.
1489
for study of photostability of levomepromazine hydrochloride. Spectrochimica Acta Part A 71, 1562–1564. Barra, M., G. S. Calabrese, M. T. Allen, R. W. Redmond, R. Sinta, A. A. Lamola, R. D. Small and J. Scaiano (1991) Photophysical and photochemical studies of phenothiazine and some derivatives: explanatory studies of novel photosensitizers for photoresist technology. Chem. Mater. 3, 610–616. Pi~nero, L., X. Calderon, J. Rodriguez, I. Nieves, R. Arce, C. García and R. Oyola (2008) Spectroscopic and electrochemical properties of 2-aminophenothiazine. J. Photochem. Photobiol. A 195, 85–91. Garcia, C., G. A. Smith, M. Grant, I. E. Kochevar and R. W. Redmond (1995) Mechanism and solvent dependence for the photoionization of promazine and chlorpromazine. J. Am. Chem. Soc. 117, 10871–10878. Elisei, F., L. Latterini, G. G. Aloisi, U. Mazzucato, G. Viola, G. Miolo, D. Vedaldi and F. Dall’Acqua (2002) Excited state properties and in vitro phototoxicity studies of three phenothiazine derivatives. Photochem. Photobiol. 75(1), 11–21. Saucin, M. and A. Van der Vorst (1980) Photodynamic potentialities of some phenothiazine derivatives. Radiat. Environ. Biophys. 17, 159–168. Madej, E. and P. Wardman (2006) Pulse radiolysis and cyclic voltammetry studies of redox properties of phenothiazine radicals. Rad. Phys. Chem. 75, 990–1000. Kelder, P. P., N. J. DeMol and L. H. Janssen (1991) Mechanistic aspects of the oxidation of phenothiazine derivatives by methemoglobin in the presence of hydrogen peroxide. Biochem. Pharmacol. 42, 1551–1559. Levy, L. (1972) Stability of some phenothiazine free radicals. J. Med. Chem. 15, 898–905. Vazquez, A., J. Tudela, R. Varon and C. F. García (1992) Determination of the molar absorptivities of phenothiazine cation radicals generated by oxidation with hydrogen peroxide/peroxidase. Anal. Biochem. 202, 245–248. Bancirova, M. (2011) Sodium azide as a specific quencher of singlet oxygen during chemiluminescent detection by luminol and cypridina luciferin analogues. Luminescence 26, 685–688. Motten, A., G. R. Buettner and C. F. Chignell (1985) Spectroscopic studies of cutaneous photosensitizing agents - VIII. A spin-trapping study of light induced free radical from chlorpromazine and promazine. Photochem. Photobiol. 42(1), 9–15. Iwaoka, T. and M. Kondo (1974) Mechanistic studies on the photooxidation of chlorpromazine in water and ethanol. Bull. Chem. Soc. Jap. 47, 980–986. Maroi, Y., A. M. Braun and M. Graetzel (1979) Light-initiated electron transfer in functional surfactant assemblies. 1. Micelles with transition metal counterions. J. Am. Chem. Soc. 101, 567–572. Alkaitis, S., G. Beck and M. Gratzel (1975) Laser photoionization of phenothiazine in alcoholic and aqueous micellar solution. Electron transfer from triplet states to metal ion acceptors. J. Am. Chem. Soc. 97, 5723–5729.
Photooxidation Mechanism of Levomepromazine in Different Solvents ~ ero-Santiago1,2, Carmelo García*1, Virginie Lhiaubet-Vallet2, Jerome Trzcionka2, Luis E. Pin 1 and Miguel A. Miranda2 Rolando Oyola1, Karen Torres1, Jaysika Leguillu 1 2
Department of Chemistry, University of Puerto Rico at Humacao, Humacao, PR cnica de Vale ncia, Valencia, Spain Instituto de Tecnología Química UPV-CSIC, Universitat Polite
Received 25 April 2013, accepted 17 July 2013, DOI: 10.1111/php.12147
ABSTRACT Unwanted photoinduced responses are well-known adverse effects of most promazine drugs, including levomepromazine (LPZ, Levoprome® or Nozinan®). This drug is indicated in psychiatry primarily for the treatment of schizophrenia and other schizoaffective disorders. Levomepromazine’s particular sedative properties make it especially fit for use in psychiatric intensive care. Nevertheless, it is photolabile under UV-A and UV-B light in aerobic conditions resulting in the formation of its sulfoxide. The LPZ photochemistry in acetonitrile (MeCN) is completely different from that in methanol (MeOH) and phosphate buffer solutions (PBS, pH = 7.4). The major photoproduct in PBS and MeOH under aerobic conditions is levomepromazine sulfoxide (LPZSO), although the amount is considerably higher in the aqueous environment. The corresponding main photoproduct in MeCN could not be characterized. The destruction quantum yields of LPZ in PBS, MeOH and MeCN are 0.13, 0.02 and 109 M1 s1) and PIND (k = [1.8 0.1] 9 1010 M1 s1). However, the triplet lifetime was found to be solvent dependent. The shortest lived LPZ triplet is found in PBS, with s = 5.1 ls (Table 1). It should be noted, however, that this value was calculated using the Stern–Volmer analysis, the lifetime of 3LPZ* is calculated at infinite dilution, and that the maleate counterion can deactivate the triplet (2). Therefore, as previously indicated, it corresponds to the intrinsic 3LPZ* lifetime in the absence of any self-quenching effect. This is relevant for in vivo situations,
where low concentrations of LPZ are present. A much lower value of 0.75 ls has been reported for 2-methoxyphenothiazine in diglyme (40), although in this case, self-quenching processes were not completely eliminated either. An excited triplet energy of ~60 kcal mol1 was obtained for 3LPZ* from the maximum emission wavelength in the phosphorescence spectrum, close to the value reported by Saucin and Vorst [ET = 58 kcal mol1 in ethanol glass (49)]. The TD-DFT predicted values for both the energy and the maximum wavelength of the triplet state are also impressively close (Table 1). The triplet-state molar absorption coefficients (eT) of 3 LPZ*were determined using PIND as the triplet acceptor in deaerated MeOH or MeCN. As PIND is not soluble in PBS 7.4, it was assumed that the corresponding eT value is the same as that obtained for MeOH. From the respective triplet absorbances maxima, values of the order of (1.5 0.02) 9 104 M1 cm1 were calculated using Eq. (3). These values are within the range of accuracy of those reported for TR (1.8 9 104 M1 cm1 in MeCN and 1.5 9 104 M1 cm1 in MeOH) (35). As previously mentioned, the bimolecular quenching constant of PIND and 3 LPZ* is close to the diffusion limit rate constant (k 1010 M1 s1), resulting in a high probability for the energy transfer process (>95%). The intersystem crossing quantum yields were determined using TX and the lowest practical laser pulse energies to minimize nonlinear effects (e.g. photoionization). From the linear dependence of the TX triplet–triplet absorption, used as an actinometer, and of LPZ as a function of the laser energy at the excitation wavelength of 355 nm (data not shown), the /T values were calculated using Eq. (4) and are given in Table 1. These values are near 0.70, close to the corresponding value of TR (/T = 0.74). This result is very important in establishing that the major deactivation pathway of LPZ is the intersystem crossing. Furthermore, the 3LPZ* properties are not substantially affected by the solvent, as occuring in the case of 3 TR* (35). Photochemistry of LPZ-MS in PBS 7.4
Figure 3. Time domain fluorescence of LPZ in MeCN under air (──) and the corresponding lamp response (– – –).
Figure 4. Transient absorption spectra of a 0.28 mM LPZ solution in N2-saturated MeCN after a 355 nm laser pulse (Energy < 3 mJ): (●) 9.9 ls, (s) 29.7 ls, (▼) 59.4 ls and (D) 247 ls. Inset: decay curves for the transients at 480 nm.
The steady-state photolysis of LPZ in PBS 7.4 is important toward understanding the behavior of this drug in vivo. Irradiation of a 0.035 mM aqueous solution of LPZ-MS at 313 nm produces an absorption increase in the regions of 230–300 and 320–350 (Fig. 5). This concentration is about eight times smaller than the one used for the LFP experiments. This is very important because the lower the ground state concentration, the larger the lifetime of the 3LPZ*. Three isosbestic points can be identified at 221, 302 and 315 nm, in agreement with the behavior reported by Karpinska during the first 40 min of irradiation (9). Subsequent to this stage, these authors observed the disappearance of the isosbestic points, most likely because they induced the photodegradation of the LPZSO. In our case, the resulting absorption spectra of the photolyzed LPZ-MS match the absorption spectrum of the sulfoxide (LPZSO-MS, Fig. 1). The photolysis was measured under identical experimental conditions, but using fluorescence spectroscopy as probe (Fig. 6). The result shows a photoproduct with a huge emission with a maximum at 375 nm, which coincides with the emission maximum of the sulfoxide (Fig. 1). It is not surprising that the fluorescence emission of LPZSO is greater than that of LPZ because all sulfoxide derivatives normally have bigger φf-values. For instance, the same behavior is observed for CMZ (7) and CMZSO (25). The
Photochemistry and Photobiology, 2013, 89
Figure 5. Absorption spectra for the photolysis of 0.035 mM LPZ-MS in PBS (pH = 7.4) under aerobic conditions at 313 nm (A313 = 0.15). A time interval of 60 s was used to record each spectrum with t(a) = 0 s. Inset: concentration∕time plot for the photolysis of LPZ-MS in PBS 7.4 under anaerobic conditions irradiated with 313 nm (A313 = 0.15, I0 = 6.9 9 109 E s1). (●) Concentration of LPZ-MS and (▼) Concentration of LPZSO-MS during irradiation. The dashed line represents the linear regression for 0–90 s.
photodegradation of LPZ-MS follows a first order mechanism in PBS 7.4 (Fig. 5). The quantum yields for the photodegradation of LPZ-MS and photoformation of LPZSO-MS (measured at the beginning of the photoreaction) are 0.13 and 0.14, respectively. This means that 18% of 3LPZ* is converted to the corresponding sulfoxide. The presence of molecular oxygen in the reaction matrix obviously plays a key role in the LPZ photooxidation mechanism. During the steady-state photolysis of LPZ-MS in PBS, no formation of the cation radical (kmax = 566 nm) was observed under aerobic conditions (Fig. 5), contrary to the experiment under almost anaerobic conditions (Fig. 7). Under aerobic conditions, the concentration of O2 is enough to generate the LPZ+ by the electron transfer to 3LPZ* and to react further to produce LPZSO. The bimolecular rate constant of the cation radical of cyamemazine (CMZ+) with molecular oxygen is 2 9 107 M1 s1 (11). Assuming that LPZ+ has a similar rate constant, its reaction with O2 would be too fast and no LPZ+ can be observed under aerobic conditions. However, at low oxygen concentrations, no molecular oxygen is available for the consecutive reaction after the production of LPZ+. Therefore, the lifetimes of both 3LPZ* and LPZ+ increase and there should be enough time to see its absorption, considering that the lifetime of LPZ+ is over a few milliseconds under anaerobic conditions (50). The absorption at 566 nm observed for LPZ+ (Fig. 7) is similar to the reported maximum wavelength for the cation radicals of 2-methoxypromazine and LPZ (562 and 565 nm) (51,52); CPZ (kmax = 530 nm) (53), PZ (kmax = 518 nm) (48) and trifluoropromazine (TFPZ, kmax = 500 nm) (48). This observation is relevant for the formation mechanism of LPZSO and is also an evidence for the monophotonic formation of the cation radical. The same experiment under anaerobic conditions shows that LPZ-MS is photostable and no formation of the cation radical is observed (data not shown). Based on these results, it could be postulated that the cation radical is formed by an electron transfer between 3LPZ* with ground-state oxygen (3O2, Scheme 1). To confirm that LPZSO is formed by the reaction of
1485
Figure 6. Emission spectra for the photolysis of a 0.072 mM LPZ-MS solution in PBS7.4 under air conditions (A313 = 0.30). A time interval of 60 s was used to record each spectrum with t(a) = 0 s. Excitation wavelength = 313 nm; slits = 2.5 nm; integration time = 0.10 s.
Figure 7. Absorption spectra for the photolysis of 0.035 mM LPZ-MS in PBS (pH = 7.4) under almost anaerobic conditions at 313 nm (A313 = 0.15). A time interval of 60 s was used to record each spectrum with t(a) = 0 s. Inset: spectra in the 450–650 nm region.
the LPZ cation radical with molecular oxygen, and not from a reaction of singlet oxygen with the LPZ ground state, the very same photoreaction previously described was performed in the presence of NaN3, which is an excellent quencher of singlet oxygen (54). It is expected that the rate of formation of LPZSO measured at 334 nm will not change upon addition of NaN3 if singlet oxygen does not play a role in its formation. Indeed, the results are consistent with this expectation (Fig. 8). Conversely, the reaction between the cation radical of the phenothiazine derivatives and superoxide is not favored, as reported by Hovey (12). Therefore, we expect a similar behavior for the reaction between LPZ+ and superoxide. Photochemistry of LPZ in MeOH The LPZ stationary photolysis in MeOH under aerobic conditions induces an absorption decrease in the region of 200–224, 257–265 and 302–321 nm (Fig. 9). In these regions, absorption minima are observed at 210, 261 and 310 nm. However, the
1486
~ero-Santiago et al. Luis E. Pin
Figure 8. Kinetics for the photolysis of 0.035 mM LPZ-MS in PBS 7.4 under air conditions measured at 334 nm in the presence of (●) 0 mM
NaN3 and (s) 0.46 mM NaN3.
absorption increases in the wavelength regions of 225–257, 266–301 and 322–346 nm. Correspondingly, absorption maxima are obtained at 244, 277 and 335 nm. These absorption peaks are similar to the LPZSO maxima in MeOH (k = 248, 277, 295 and 333 nm). During short periods of photolysis, the observed characteristics match the absorption properties of LPZ and LPZSO. These results differ from those reported by Vargas et al., as theirs show a decay in absorbance at 254 nm, when a hypsochromic shift should have been be observed (6). Furthermore, unlike LPZ-MS photochemistry in PBS 7.4, no isosbestic points can be clearly identified in MeOH. This contrasts sharply with the five isosbestic points observed in the plot of extinction coefficient vs k for LPZ and LPZSO in this solvent. This could imply that the photochemistry in this environment is complex and other competing photoproducts are masking the properties of LPZSO. Besides, LPZSO is more photoreactive than LPZ in this solvent. To elucidate the complexity of the photochemistry of LPZ in MeOH, the same experiment was repeated as described above, but by measuring the fluorescence spectra as probe (Fig. 10). The results clearly show that the emission at 448 nm decreases, which is consistent with the photodegradation of LPZ. However, there is an increase in the fluorescence emission in the region of 325–395 nm and the maximum shifts during the course of photolysis. In the first 10 min of photolysis, the maximum is found at 373 nm, which coincides with the emission maximum of LPZSO (kmax = 375 nm). This suggests that, at short irradiation times, LPZSO formation is favored. After 10 min of photolysis, the maximum hypsochromic shifts to 355 nm and an isosbestic point appears at 412 nm. This confirms that the photochemical behavior of LPZ in this solvent is complex and several photoproducts are formed depending on photolysis time, in contrast to the behavior observed in PBS 7.4. The HPLC of the photolyzed solution shows that the major photoproduct is LPZSO and some of the minor photoproducts are N-oxide LPZ and N,S-dioxide LPZ. The corresponding absorption and emission spectra of LPZ and N-oxide LPZ are exactly the same. This behavior was also observed for the absorption and emission spectra of LPZSO and N,S-dioxide LPZ. Also, it was impossible to identify the minor photoproduct with an emission maximum near 355 nm. Although most of the major photoproducts of the bulk photolysis of LPZ in MeOH were isolated and characterized, their fluorescence
Figure 9. Absorption spectra for the photolysis of a 0.035 mM MeOH solution of LPZ under aerobic conditions at 313 nm (A313 = 0.15). A time interval of 300 s was used to record each spectrum with t(a) = 0 s.
Figure 10. Emission spectra for the photolysis of a 0.035 mM LPZ solution in MeOH under air conditions.
spectra show that they are all contaminated with the photoproduct with a maximum emission at 355 nm. It is impossible to detect this impurity using other spectroscopic techniques such as NMR and IR. The results show that LPZ is more photostable in MeOH than in PBS (pH = 7.4), but its photochemistry is more complex. This fact was confirmed by photolyzing LPZ with the same light intensity for the same time in both solvents. Under our experimental conditions, LPZ photodegraded completely after 10–15 min in the aqueous environment, whereas only 19% disappeared in MeOH. All other parameters being equal, a photodegradation quantum yield of only 0.02 can be extrapolated for LPZ in MeOH taking its value in PBS 7.4 as reference [Eq. (1)]. This value differs by a factor of 10 from the one reported by Vargas et al., which is 0.18 in the UVB region (6). Vargas et al. further reports that 80% of LPZ is converted into LPZSO, while this study found a conversion yield of only 55%. The major photoproduct of this photoreaction in MeOH is still LPZSO, but the corresponding N-oxide and N-oxide sulfoxide are detectable byproducts. These other photoproducts are not observed in PBS 7.4 because the electron pair of the nitrogen of the alkyl-amino chain
Photochemistry and Photobiology, 2013, 89 is not available, which is obviously a requirement for their formation. Photochemistry of LPZ in MeCN The LPZ stationary photolysis in MeCN under aerobic conditions has not been previously studied. This solvent has a dielectric constant of e = 37.5, similar to the value of MeOH (e = 30). Besides, the corresponding concentrations of dissolved molecular oxygen at 25°C are also similar; 10.2 mM in MeOH and 9.1 mM in MeCN. Nevertheless, the photochemistries of LPZ under the same conditions in both solvents are different (Figs. 8 and 10). In MeCN, an absorption decrease is observed at 211, 258 and 314 nm, while it increases at 228, 279 and 364 nm (Fig. 11). As a consequence, three isosbestic points form at 270, 300 and 336 nm. Furthermore, contrary to its photolysis in MeOH and PBS 7.4, the main photoproduct in MeCN is not LPZSO. In fact, the major photoproducts of the LPZ photodecomposition in MeCN could not be detected, apparently due to the retention of these compounds by the nonpolar HPLC column. The chromatograms only show the presence of LPZ and traces of LPZSO. Moreover, the corresponding N-oxide and other related derivatives were not detected either. Taking into consideration all the results of this study, it can be established that the presence of molecular oxygen does not alter the lifetime of the excited singlet state (1LPZ*). The fluorescence quantum yield is very low and, therefore, the quantum yield of formation of the excited triplet state is relatively high in all solvents tested (φT > 0.60). Furthermore, it was observed that the triplet lifetime depends more on the solvent and the molecular oxygen
*
S
hν
N R1
concentration than on any other parameter. However, the molar absorption coefficient is solvent independent. In general, LPZ is photostable under anaerobic conditions, but it photodegrades very quickly under aerobic conditions. The photochemistry of LPZ in MeCN, on the other hand, is completely different from that in MeOH and PBS (pH = 7.4) under aerobic conditions. The major photoproduct found for the photolysis of LPZ-MS in phosphate buffer and MeOH under
Figure 11. Absorption spectra for the photolysis of a 0.0345 mM solution of LPZ-FB in MeCN under aerobic conditions at k > 300 (A313 = 0.19). A time interval of 300 s was used to record each spectrum with t(a) = 0 s.
1
S N R1
OCH3
*
OCH3
3
S N R1
OCH3 O2
OH, O2
OH O S N R1
OCH3
O2, H2O
O O S N R1
O2 O2 OCH3
OH
H2O2 O S N R1
1487
OCH3 Scheme 1. Mechanism proposed for the formation of LPZSO.
S N R1
OCH3
1488
~ero-Santiago et al. Luis E. Pin
aerobic conditions is LPZSO, although the amount is considerably higher in the aqueous environment. The characterization of the corresponding main photoproduct in MeCN could not be achieved with the available separation methods. Nevertheless, only traces of LPZSO and no other oxidation products were detected for this solvent.
CONCLUSIONS To elucidate its photooxidation mechanism, the photophysics and photochemistry of LPZ and LPZ-MS were studied in different solvents, under aerobic and anaerobic conditions. The results of this work are summarized in the reaction mechanism given in Scheme 1. This scheme shows that LPZSO does not form as a result of a reaction of singlet oxygen with ground state LPZ as previously proposed by Hovey (12). Nonetheless, consistent with this author’s findings, our results show that the LPZ+ is actually formed by an electron transfer process between 3LPZ* and molecular oxygen upon irradiation with 313 nm. This proves that the cation radical of the promazine derivatives can form in three different ways: (a) in a biphotonic process with k > 300 nm, as proposed by Chignell et al. (55) and Kochevar et al. (47) for CPZ; (b) in a monophotonic process with k < 300 nm; and (c) in monophotonic process at wavelengths higher than 300 nm, provided there is an acceptor that serves as a catalyst for the electron transfer process. However, no LPZSO is formed by a singlet oxygen reaction in PBS. Moreover, the oxidation photoproducts in MeOH (N-oxide and N-oxide sulfoxide) may form by a reaction between singlet oxygen and the ground state of LPZ, but they are only minor products. The cation radical, in turn, reacts with ground-state triplet oxygen, producing a peroxyl free radical, as previously proposed by Iwaoka and Kondo for the photooxidation of CPZ in water with 254 nm (56). This last radical is then responsible for the formation of LPZSO. Acknowledgements—This work has been supported in part by NIH– MBRS grant SO6GM08216 to UPR–Humacao and the UPRH-FOPI Program (CGR). Financial support by the Spanish Government (Ramon y Cajal contract to VLV) is gratefully acknowledged. We especially thank the High Performance Computing Facility at UPR Río Piedras for assistance with the theoretical calculations and Elizabeth Hodges for the revision of the manuscript.
REFERENCES 1. Nogrady, T. (1988) Medicinal Chemistry: A biochemical approach, 2nd edn. Oxford University Press, New York. 2. Oyola, R. (2001) Design, construction and use of a nanosecond laser transient absorption spectrokinetic system for the study of the photochemical and photophysical properties of relevant biological molecules: dTpA, dApT and tricyclic antidepressive drugs. Ph. D. Thesis, University of Puerto Rico, Río Piedras. 3. Dahl, S. G. and H. Hall (1981) Binding affinity of levomepromazine and two of its major metabolites to central dopamine and a-adrenergic receptors in the rat. Psychopharmacology 74, 101–104. 4. Sivaraman, P., R. Rattehalli and M. Jayaram (2012) Levomepromazine for schizophrenia. Schizophrenia Bull. 38, 219–220. 5. Dietz, I., A. Schmitz, I. Lampey and C. Schulz (2013) Evidence for the use of levomepromazine for symptom control in the palliative care setting: A systematic review. BMC Palliat. Care 12, 1–11. 6. Vargas, F., K. Carbonell and M. Camacho (2003) Photochemistry and in vitro phototoxicity studies of levomepromazine (methotrimeprazine), a phototoxic neuroleptic drug. Pharmazie 5, 315–319.
7. Morliere, P., J. Haigle, K. Aissani, P. Filipe, J. N. Silva and R. Santus (2004) An insight into the mechanisms of the phototoxic response induced by cyamemazine in cultured fibroblasts and keratinocyte. Photochem. Photobiol. 79(2), 163–171. 8. Rasheed, A., M. Javed, S. Nazir and O. Khawaja (1994) Interaction of chlorpromazine with tricyclic anti-depressants in schizophrenic patients. J. Pakistan Med. Asoc. 44, 233–234. 9. Karpinska, J., A. Soko, A. Bernatowicz, A. Szul^ecka and U. Kotowska (2012) Studies on photodegradation of levomepromazine and olanzapine under simulated environmental conditions. Photochem. Photobiol. Sci. 11, 1575–1584. 10. Dahl, S. G. and H. Refsum (1976) Effects of levomepromazine, chlorpromazine and their sulfoxides on isolated rat atria. Eur. J. Pharm. 37, 241–248. 11. Morliere, P., F. Bosca, M. A. Miranda, J. V. Castell and R. Santus (2004) Primary photochemical processes of the phototoxic neuroleptic cyamemazine: a study by laser flash photolysis and steady-state irradiation. Photochem. Photobiol. 80, 535–541. 12. Hovey, M. C. (1982) Micellar photochemistry. Photooxidations with intramicellar-generated singlet oxygen. J. Am. Chem. Soc. 104, 4196–4202. 13. Thankamoniamma, M., M. Narayanapillai, A. M. Braun and E. Oliveros (2012) Self sensitized photooxidation of n-methyl phenothiazine: acidity control of the competition between electron and energy transfer mechanisms. Photochem. Photobiol. Sci. 11, 1744– 1755. 14. Xiang, B. and L. Kevan (1994) Photooxidation of phenothiazine derivatives in silicas of different pore sizes. Langmuir 10, 2688–2693. 15. Nath, S. and A. V. Sapre (2001) Photoinduced electron transfer from chloropromazine and promethazine to chloroalkanes accompanied by cleavage of C-Cl bond. Chem. Phys. Lett. 344, 138–146. 16. Cheng, H. Y., P. H. Sackett and R. L. McCreery (1978) Kinetics of chlorpromazine cation radical decomposition in aqueous buffers. J. Am. Chem. Soc. 100, 962–967. 17. Rosenthal, I. and T. Bercovici (1977) Dye-sensitized photooxidation of chlorpromazine. J. Heterocyclic Chem. 14, 355–357. 18. Galzigna, L., M. P. Schiappelli and M. Scarpa (1997) A peroxidase-catalyzed sulfoxidation of promethazine. Free Radic. Res. 27, 501–504. 19. Baciocchi, E., T. Del Giacco, O. Lanzalunga, A. Lapi and D. Raponi (2007) The singlet oxygen oxidation of chlorpromazine and some phenothiazine derivatives. Products and reaction mechanisms. J. Org. Chem. 72, 5912–5915. 20. Baciocchi, E., T. Del Giacco and A. Lapi (2004) Oxygenation of benzyldimethylamine by singlet oxygen. Products and mechanism. Org. Lett. 6, 4791–4794. 21. Baciocchi, E., T. Del Giacco and A. Lapi (2006) Quenching of singlet oxygen by tertiary aliphatic amines. Structural effects on rates and products. Helv. Chim. Acta 89, 2273–2280. 22. Hall, R. D., G. R. Buettner, A. Motten and C. F. Chignell (1987) Near-infrared detection of singlet molecular oxygen produced by photosensitization with promazine and chlorpromazine. Photochem. Photobiol. 46, 295–300. 23. Petrushenko K. B., A. I. Vokin, V. K. Turchanino an and Y. L. Frolov (1983) Energy degradation processes of the electron excitation of the phenothiazine molecule. Bull. Acad. Sci. USSR Div. Chem. Sci.(Engl. Transl.) 32, 2151–2152. 24. Demas, J. N. and G. Crosby (1971) The measurement of photoluminescence quantum yields: a review. J. Phys. Chem. 75, 991–1024. 25. Bosca, F., P. Moliere, M. A. Miranda, J. Castell and R. Santus (2006) Primary steps of the photochemical reactions of 2-cyano-10(3-[dimethylamino, n-oxide]-2-methylpropyl)-5-oxide-phenothiazine, the photoproduct of cyamemazine, a phototoxic neuroleptic: comparison with the sulfoxide. Photochem. Photobiol. Sci. 5, 336–342. 26. Loennechen, T. and S. G. Dahl (1990) High-performance liquid chromatography of levomepromazine (methotrimprazine) and its main metabolites. J. Chromatography A 503, 205–215. 27. Pi~nero, L. E. (2006) Synthesis, characterization, photophysics, and photochemistry of 2-chlorinated phenothiazine derivatives in several solvents. Master Thesis, UPR-Río Piedras, Puerto Rico. 28. García, C., L. Pi~nero, R. Oyola and R. Arce (2009) Photodegradation of 2-chloro substituted phenothiazines in alcohols. Photochem. Photobiol. 85, 160–170. 29. Tosa, M., C. Paizs, C. Majdik, L. Poppe, P. Kolonits, J. A. Silberg, L. Novak and F. Irimie (2001) Selective oxidation methods for
Photochemistry and Photobiology, 2013, 89
30.
31. 32.
33.
34. 35.
36.
37. 38. 39. 40.
41.
42. 43.
44.
preparation of N-alkylphenothiazine sulfoxide and sulfones. Heterocycl. Commun. 7(3), 277–282. Tosa M., C. Paizs, C. Majdik, L. Novak, P. Kolonits, F. Irimie and L. Poppe (2002) Synthesis of optically active 3-substituted-10-alkyl10H-phenothiazine-5-oxide by enantioselective biotransformations. Tetrahedron: Asymmetry 13, 211–221. Jouyban, A. and B. H. Yousefi (2003) A quantitative structure property relationship study of electrophoretic mobility of analytes in capillary zone electrophoresis. Comp. Biol. Chem. 27, 297–303. Amovillia C., V. Barone, R. Cammic, E. Cancesd, M. Cossib, B. Mennuccia, C. S. Pomellie and J. Tomasi (1998) Recent advances in the description of solvent effects with the polarizable continuum model. Adv. Quantum Chem. 32, 227–261. Charaf-Eddin A., A. Planchat, B. Mennucci, C. Adamo and D. Jacquemin (2013) Choosing a functional for computing absorption and fluorescence band shapes with TD-DFT. J. Chem. Theory Comput. 9, 2709–2760. Kuhn, H. J., S. E. Braslavsky and R. Schmidt (1989) Chemical actinometry. Pure Appl. Chem. 61(2), 187–210. García, C., R. Oyola, L. E. Pi~nero, R. Arce, J. Silva and V. Sanchez (2005) Substitution and solvent effect on the photophysical properties of several series of 10-alkylated phenothiazine derivatives. J. Phys. Chem. A 109, 3360–3371. Lhiaubet-Vallet, V., N. Belmadoui, M. J. Climent and M. A. Miranda (2007) The long-lived triplet excited state of an elongated ketoprofen derivative and its interactions with amino acids and nucleosides. J. Phys. Chem. B 111, 8277–8282. Bensasson, R. and E. J. Land (1971) Triplet-triplet extinction coefficient via energy transfer. Trans. Faraday Soc. 67, 1904–1915. Cosa, G. (2004) Laser techniques in the study of drug photochemistry. Photochem. Photobiol. 80, 159–174. Merkel, P. and J. P. Dinnocenzo (2008) Thermodynamic energies of donor and acceptor triplet states. J. Photochem. Photobiol. A 193, 110–121. Bensasson, R., C. R. Goldschmidt, E. J. Land and T. G. Truscott (1978) Laser intensity and the comparative method for determination of the triplet quantum yield. Photochem. Photobiol. 28, 277–281. Allonas, X., C. Ley, C. Bibaut, P. Jacques and J. Fouassier (2000) Investigation of the triplet quantum yield of thioxanthone by timeresolved thermal lens spectroscopy: solvent and population lens effects. Chem. Phys. Lett. 322, 483–490. Yip, R. W., A. G. Szabo and P. K. Tolg (1973) Triplet state of ketones in solutions. Quenching rate studies of thioxanthenone triplets by flash absorption. J. Am. Chem. Soc. 95(13), 4471–4472. Aaron, J., M. Maafi, C. Kersebet, C. Parkanyi, A. M. Shafik and N. Motohashi (1996) A solvatochromic study of new benzo [a] phenothiazines for the determination of dipole moments and specific solute-solvent interactions in the first excited singlet state. J. Photochem. Photobiol. A 101, 127–136. Karpinska, J., A. Sokol and M. Skoczylas (2008) An application of UV-derivative spectrophotometry and bivariate calibration algorithm
45.
46. 47.
48.
49. 50. 51.
52. 53.
54. 55.
56. 57. 58.
1489
for study of photostability of levomepromazine hydrochloride. Spectrochimica Acta Part A 71, 1562–1564. Barra, M., G. S. Calabrese, M. T. Allen, R. W. Redmond, R. Sinta, A. A. Lamola, R. D. Small and J. Scaiano (1991) Photophysical and photochemical studies of phenothiazine and some derivatives: explanatory studies of novel photosensitizers for photoresist technology. Chem. Mater. 3, 610–616. Pi~nero, L., X. Calderon, J. Rodriguez, I. Nieves, R. Arce, C. García and R. Oyola (2008) Spectroscopic and electrochemical properties of 2-aminophenothiazine. J. Photochem. Photobiol. A 195, 85–91. Garcia, C., G. A. Smith, M. Grant, I. E. Kochevar and R. W. Redmond (1995) Mechanism and solvent dependence for the photoionization of promazine and chlorpromazine. J. Am. Chem. Soc. 117, 10871–10878. Elisei, F., L. Latterini, G. G. Aloisi, U. Mazzucato, G. Viola, G. Miolo, D. Vedaldi and F. Dall’Acqua (2002) Excited state properties and in vitro phototoxicity studies of three phenothiazine derivatives. Photochem. Photobiol. 75(1), 11–21. Saucin, M. and A. Van der Vorst (1980) Photodynamic potentialities of some phenothiazine derivatives. Radiat. Environ. Biophys. 17, 159–168. Madej, E. and P. Wardman (2006) Pulse radiolysis and cyclic voltammetry studies of redox properties of phenothiazine radicals. Rad. Phys. Chem. 75, 990–1000. Kelder, P. P., N. J. DeMol and L. H. Janssen (1991) Mechanistic aspects of the oxidation of phenothiazine derivatives by methemoglobin in the presence of hydrogen peroxide. Biochem. Pharmacol. 42, 1551–1559. Levy, L. (1972) Stability of some phenothiazine free radicals. J. Med. Chem. 15, 898–905. Vazquez, A., J. Tudela, R. Varon and C. F. García (1992) Determination of the molar absorptivities of phenothiazine cation radicals generated by oxidation with hydrogen peroxide/peroxidase. Anal. Biochem. 202, 245–248. Bancirova, M. (2011) Sodium azide as a specific quencher of singlet oxygen during chemiluminescent detection by luminol and cypridina luciferin analogues. Luminescence 26, 685–688. Motten, A., G. R. Buettner and C. F. Chignell (1985) Spectroscopic studies of cutaneous photosensitizing agents - VIII. A spin-trapping study of light induced free radical from chlorpromazine and promazine. Photochem. Photobiol. 42(1), 9–15. Iwaoka, T. and M. Kondo (1974) Mechanistic studies on the photooxidation of chlorpromazine in water and ethanol. Bull. Chem. Soc. Jap. 47, 980–986. Maroi, Y., A. M. Braun and M. Graetzel (1979) Light-initiated electron transfer in functional surfactant assemblies. 1. Micelles with transition metal counterions. J. Am. Chem. Soc. 101, 567–572. Alkaitis, S., G. Beck and M. Gratzel (1975) Laser photoionization of phenothiazine in alcoholic and aqueous micellar solution. Electron transfer from triplet states to metal ion acceptors. J. Am. Chem. Soc. 97, 5723–5729.
