Psoralens sensitize glutathione photooxidation in vitro

Biochimica et Bwph):sica Ac:a. 993 (1989) 143-147

143

Elsevier BBAGEN23200

Psoralens sensitize glutathione photooxidation in vitro Marco d'Ischia, Alessandra N a p o l i t a n o and G i u s e p p e Prota Department of Organic and Biological Chemistry. University of Naples, Naples ( llaly)

(Received26 April1989)

Key words: Furocoumadn;Psoralen; Angelicin;GlutatbJorle;Fhotooxldation;Melaninpigmentauon; Ultravioletirradiation; Oxygenradical In vitro experiments are reported showing that pseralens and other furocoumarins of current pharmacological interest, e-g., angeliein and 4,6,4'-trimethylangelicin, all have, to a variable extent, the abilit), to sensitize the photooxidation of ghitathiene in ethanol/phosphate buffer with pyrex-filtered ultraviolet light. Besides substrate concentration and the nature of the furueoumarin used, the rate of the sensitized reaction is markedly dependent on the partial pressure of oxygen and the pH of the medium, being progressively faster on passing from pH 5 to oH 8.5. Scavengers of superoxide inns (superoxide dismntase), hydrogen peroxide (catalase) and singlet oxygen (sedium azide, diazabieyclooctane, sorbic acid) have little or no inhibitory effect on the reaction rate. These and other data suggest that furocoumarins can directly sensitize glutathlone photooxidafion by forming a charge transfer complex which is driven to the oxidized products in the presence o[ oxygen. The possible ~elevance of these results ¢o the mechanisms of skin melanin hyperpigmentafion induced by furocoumarins and u!travialet light is discussed.

Introduction The remarkable ability of psoralens and structurally related furocoumarins to enhance the pigmenting response of human skin to sunlight or ultraviolet radiation has provided much of the incentive for the investigation of the photochemical properties of these ~:otent cutaneous photosensitizers [l 7]. Yet, in spite of a lnzssive body of research carded out over more t h n two decades, the precise molecular mechanism of this synergistic effect on skin pigmentation remains largely unknown. According to one view, increased pigment production would result from a furocoumarin-sensitized type 1I photooxidation process targeted to tyrosine and dopa [8,9], which involves the generation of ringlet oxygen and superoxide ions as the actual oxidizing agents [10]. It should be noted, however, that when observed in vitro this photosensitized prccess is rather poor, especially in the case of tyrosine, and seems therefore unlikely to provide more than a nfinor eontri-

Abbreviations: GSH. reduced glutallfione;PS, psoralen; 8-MOP, 8-methoxypsoralen; 5-MOP, 5-methoxypsoralen; TMP, 4.5',8-trimeth~'lpsora!en;AN, angelicin;TMA, 6,4,4'-tdmethylangelicin;SOD, superordde disrnutase; DABCO,1.4-diazabicyelooctane. Correspondence: G, Prota, Department of Organic and Biological Chemistry, Universityof Naples, Via MeTzocannone 16, 1-80134 aples, Italy.

button to the psoralen-induced skin hyperpigmentation. More plausible targets, in this respect are the 5,6-dihydroxyindole melanin intermediates, which have recently been shown to be highly susceptible to pbotooxidation [11-13]. In any case, such a process may play a role in immediate pigment darkening, but can hardly account for delayed tanning [14], which is the most significant phenomenon accompanying furocoumarin skin photosensitizatiop. An interesting clue to the understanding of the pigmentary effects of furocoumarins is provided by the early observation of Goldblum et al. [15], who found that 8-methoxypsoralen can inhibit the rise of serum SH groups caused by ultraviolet radiation. Along this line, Knox and Ogura [16] reported that administration of the drug at high concentrations induces varying changes in epidermal and serum SH levels in rats exposed to ultraviolet (UV) radiation, More recently, Connor and Wheeler [17] found that irradiation of hairless mice skirt with ultraviolet-A light combined with oral administration of psoraleus leads to an extensive and prolonged depletion of both epidermal and dermal ghitathione (GSH), the severity of which is dependent on the psoralen dose and may last for several days. Overall, tllese studies point to an important effect of psoralens and UV light on .q,'. GSH system in vivo. However, whether the observed depletion of GSH is a secondary metabolic event elicited by excited psora[ens or follows from the direct photooxidation of GSH is not yet clear.

{;304-4165/89/$03.50© 1989ElsevierSciencePublishersB.V.(BiomedicalDiwsion)

144

~cH~

ocH~

cH~

cH3

TMP

AN

OH3

TMA

In this paper we report the results of an in vitro study providing evidence that under physiologically relevant conditions a number of furocoumarins of current pharmacologieM interest share to a variable extent the ability to accelerate the kinetics of photooxidation of GSH with pyrex-filtered UV light. Materials and Methods

Materials Psoraien (PS), 8-methoxypsoralen (8-MOP), 5methoxypsoralen (5-MOP) and 4,5',8-trimethylpsoralen (TMP) were from Fluka. Angelicin (AN) and 6,4,4'-trimethylangelicin (TMA) (structures given above) were kindly provided by Professor F. Dall'Acqua (University of Padua). Reduced ghitathione (GSH), sodium azide, diazabicyclooctane, sorbic acid and 5,5"-dithiobis(2nitrobenzoic acid) were from Fluka. Superoxide dismutase (SOD, 3000 U/rag) from bovine erythrocytes and catalase (2500 U / m g ) from bovine liver were from Sigma. All other chemicals were of the highest purity available. Distilled water and buffers were passed over Chelex 100 resin (Na ÷ form) before use, to remove metal impurities. Glassware was washed with 1 M HCI "0.01 M EDTA and 10 15 rinses of deionized water. Absolute ethanol was from Carlo Erba. Ellman's reagent was prepared according to Riddles et al. [18]. Glutathione photooxidation In a typical experiment, a solution of GSH made up to the desired concentration in ethanol/0.02 M phosphate buffer 6 : 4 (v/v) (2.5 mi) was irradiated in a cuvette with a 500 W Helios halquartz high-pressure mercury_ lamp equipped with a pyrex water cooling jacket at a distance of 10 cm. The system produced UV irradiance of 7.0 m'V/cm 2 at a 10 cm distance, as determined with an EG & G photometer-radiometer model 550-1. Stock solutions of furocoumarins (1 m g / m l in ethanol) were prepared fresh before use. When necessary, aliquots of the psoralen solutions were added to the reaction mixture up to the Jesired concentration.

Addition of sodium azide, sorbic acid or DABCO for inhibition experiments significantly altered the pH of the medium and adjustment to the initial value was required before irradiation. Stock solutions of SOD (1 mg/ml) and catalase (2 mg/ml) were prepared freshly in deionized water, and 100-FI aliquots were added when necessary. The extent of GSH photooxidation was determined by periodically withdrawing a 100 ~1 aliquot of the reaction mixture via a syringe and adding it to Ellman's reagent (900 pA), care being taken to ensure complete mixing. After 1 rain equilibration, the absorbance at 412 nm was read against a blank containing Ellman's reagent. Each measure was made at least in triplicate. Control experiments showed that the SH assay is not affected by the furocoumarin or any of the additives used. Absorbance measurements were carried out with a Perkin-Elmer model 550S spectrophotometer having the cell compartment controlled at 25 ° C. Results Fig. 1 shows the effect of various furocoumarins (0.35 mM) on the UV light-induced autoxidation of 1 mM GSH in ethanol/phosphate buffer (pH 6.8) in air. While direct irradiation with pyre~-filtered UV light induced only ,~ modest decomposition of GSH, the addition of furocoumarins led to a marked enhancement of the reaction rate, to an extent depending on the structure of the furocoumarin added. No influence of furocoumalins on GSH autoxidation could be observed in the dark. Among the compounds tested, PS, TMP and TMA were the most effective in inducing GSH decomposition, followed by AN, 8-MOP, and 5-MOP in that order. Control experiments revealed that, upon irradiation under the reaction conditions, PS, TMA and TMP undergo partial photodecomposition, which is not affected by 1 mM GSH, while 8-MOP and 5-MOP are quite stable over the whole duration of the irradiation. Accordingly, g-MOP was chosen throughout this study as the model compound to investigate the mechanism of furoeumarin induced photooxidation of GSH. In further experiments it was found that the extent of GSH decomposition depends on the partial pressure of oxygen, no decomposition occurring under nitrogen atmosphere. Under aerobic conditions, ana!:,'sis of the reaction mixture showed that the disappearance of GSH was due to oxidative conversion to the corresponding disulphide (GSSG) and not to chemical reaction with psoralens. As expected, the rate of GSH oxidation was found to depend on the concentration of 8-MOP. Thus, increasing the psorulen concentration from 0.08 to 0.35 mM with 1 mM GSH led to a 30% increase of the reaction rate. Fig. 2 shows the photooxidation of GSH both in the presence and in the absence of 8-MOP, as a function of pH. A marked increase in the rate of both

145 ,oo

n

2o

lc / Fig. l. Effect of furocoumarins (0,35 raM) on the aerial photooxidation of GSH (1 mM) in ethanol/002 M phosphate buffer (pH 6.8) 6:4 (v/v). 0 , dark; c3, UV light; o, UV light+5-MOP; z~, UV light +8-MOP; ~,. UV light+AN; m. UV light +TMP; [3. UV light+ TMA; v, UV lighl + PS. Each value is the mean of four expenments_+ S.E. (designated by tit: ~x:rtical bar). sensitized a n d unsensitized reactions can be observed with increasing p H , suggesting that the ionized sulphydryl g r o u p of G S H is more susceptible to oxidation than the undissociated form (PKsH of G S H = 9.65) [191. A t a concentration of 110 U / r o t S O D , an efficient scavenger o f superoxide ions, was found to induce a b o u t a 10% inhibition of the 8-MOP-sensitized p h o t o o x i d a tion of G S H . Catatase, both alone (100 U / m l ) or c o m b i n e d with S O D , was totally ineffective in inhibiting G S H p h o t o o x i d a t i o n . S o d i u m azide, generally regarded as a most efficient quencher of singlet oxygen, at

'°°-1

~:

nl

Fig. 3, Plot of the reciprocal of the concentration of photooxidized GSH vs. the reciprocal of the initial concentration of GSH. Irradiation was carried out in ethanol/0 02 M phosphate buffer (pH 6.8) in the presence of 0.35 mM 8-MOP. The yields of photooxidized GSH were estimated from the concentration of unreacted GSH after an irradiation time of 2 h. Each value is the mean of three experiments:c S.E. (designated by the vertical bar).

a c o n c e n t r a t i o n as high as 10 m M also had n o effect on G S H p h o t o o x i d a t i o n . Q u i t e unexpectedly, another well-known singlet oxygen quencher, D A B C O , caused a more than 2-fold increase in the reaction rate. T h i s effect was f o u n d to d e p e n d on the presence of psoralen, since in control e x p e r i m e n t s n o e n h a n c e m e n t of G S H p h o t o o x i d a t i o n by D A B C G alone occurred. A d d i t i o n of 10 m M sorbic acid, b o t h a triplet quencher a n d a singlet oxygen scavenger [9,20], led to a m o d e s t (not greater than 30%) inhibitory effect on the kinetics of the 8 - M O P sensitized reaction. In s u b s e q u e n t experiments it was found that the rate of G S H d e c o m p o s i t i o n varies significantly with the inidal c o r c e n t r a t i o n of the peptide within the range 5- 10 -4 1 • 10 -2 M ( 8 - M O P concentration - 0 . 3 5 raM). T h i s is a p p a r e n t f r o m Fig. 3, which shows a plot of the reciprocal of the concentration of photooxidized G S H vs. the reciprocal of the initial concentration of G S H . T h e d a t a exhibit a g o o d linear relationship (c.c - 0.98) with a slope of 1.40 and a n intercept of 5.87. A n interpretation of this plot is given in the Discussion. Discussion

J ca pH Fig. 2. pH-dcpcndence of the phot)oxidation nf GSH (l mMI in ethanol/0.02 M phosphate buffer. Each value is the mean of three e~pcr;mcnts±S.E. (designated b'/ tbe vertical bar). The +rradiatitln time was 2 h. O, dark; zx, UV light; A, UV light +0.35 mM 8+MOP. 5

6

•

b

Acute skin photosensitization reactions induced by f u r o c o u m a r i n s a n d U V - A irradiation are generally believed to result f r o m two m a i n photochemical processes, anoxic type I, which gives rise to covalent b i n d i n g product s with nucleic acids, proteins, and polyunsaturated fatty acids, and oxygen-dcpendcnt type IL media-

146 ted by acive oxygen species, such as singlet oxygen and superoxide ions, targeted :o aromatic amino acids, lipids, etc. [1,2,21]. Our finding that furocoumarins can directly sensitize the in vitro plaotooxidation of GSH with UV light points to a new interesting property of these compounds, which may account for a variety of shortterm effects associated with psoralen therapy, including hyperpigrnentation. Mechanistically. the reaction is mediated by molecular oxygen and occurs mainly at the expenses of the thiolate anion of GSH, as the marked pH-dependence indicates. Interestingly, active oxygen species appear to play only a minor role, as evidenced by the lack of inhibitory effect of sodium azide and catalase, and the low effect of superoxide dismutase. Most likely, superoxide ions are not the primary oxidizing species towards the sulphydryl group, and are only involved in side oxidative processes. The unexpected ability of DABCO to enhance the 8-MOP-induced pho;coxidation of GSH can be ascribed to the basic nature of the compound, which facilitates proton abstraction from the SH group, in a general base-catalyzed reaction. A related but opposite effect, clue to the presence of the carboxyl group suppressing the ionization of the SH group, could perhaps contribute to the partial inhibition induced by sorbic acid. In the light of the foregoing, it appears that photoexe2ed furocoumarins would directly promote GSH oxidation by forming an initial intermediate exciplex, most reasonably a charge transfer complex, which collapses under anaerobic conditions but is susceptible to rapid oxidation in the presence of oxygen, leading to GSH thiyl radical and then to GSSG. With regard to the nature of the furocoumarin species responsible for GSH oxidation, available evidence does not allow a decision between a singlet or a triplet intermediate. In this connection, it is noteworthy that psoralen triplets can give efficient chargetransfer complexation with a variety of electron rich molecules, including nucleic acid bases and amino acids [22]. In any case, whatever the actual mechanism, the existence of a linear relationship between the reciprocal of the concentration of oxidized GSH and the reciprocal of the initial concentration of GSH (Fig. 3) indicates that a single excited psoralen species is mainly inv,~;ved in the photoactivation process, at least within the range of GSH concentrations examined. This relationship can be obtained from the following equations [23], in wlfich F* denotes the excited furocoumarin species directly stimulating GSH oxidation (either singlet or triplet), GS is the first formed oxidized product of GSH (e.g., GSH thiyl radical), and lab s iS the number of mole quanta per liter absorbed during the irradiation time:

k: GSH ~ s = ~v' kl + k, GSH

(a)

1/¢~s = O/q'v')'(~ +(k~/X'D'O/GSta))

(5)

1/GS= (1/~',F.)./,~,.(1 + (kl/k2).(l/OSH))

(6)

From the ratio of the slope to the intercept on the ordinate axis, a k J k 2 value of 0.24 M can be obtained, which points to a quite efficient interaction between excited 8-MOP and GSH. Clearly, tb: above equations are based on an oversimplified mechanistic scheme which does not take into account all the possible species involved and their mutual interactions. However. for our purposes, they pro~ide a useful basis for the treatment of the experimental data. It is difficult to assess to what extent the proposed sensitizatioa mechanism is responsible for psoralen-induced depletion of GSH in vivo. Skin photosensitization by furocoumarins involves a variety of biochemical processors and it may well be that long-term lowering of the levels of cutaneous GSH is the result of a profound modification of the enzymic systems controlling the biosynthesis and the redox state of this tripeptide. Since the studies of Rothman in the 1940"s it has been known that GSH plays a key regulator3' role in melanin pigmentation [24,25]. Comparative studies on man and animal models with different types of pigmentation showed that GSH and GSH reductase levels are significantly higher in white or light pigmented skins than in dark or black skins [26,27]. Evidence accumulated in recent years indicates that the inhibitory effect of GSH on melanogenesis depends on the ability of this peptide to chemically react with the melanin precursor dopaquinone to give the colourless glutathionyldopa adduets, which cannot be converted to melanin pigments [25]. In the light of this mechanism. the finding that furocoumarins can induce the photooxidative depletion of GSH provides a plausible explanation to the skin pigmentogenie response, since the perturbation of the GSH-GSSG equilibrium towards the oxidized form would expectedly result in an increased level of dopaquinone avzilable for melanogenesis [27]. Acknowledgements

(t)

This work was supported in part by grants from MPI (Rome) and CNR (Rome). The authors thank Dr. Alexander Chan (Clairol Research laboratories. Stanford, U.S.A.) for helpful discussions and reading the manuscript.

v'~r

(2)

References

F* 4 GSH ~GS

(3)

v~v"

1 Song,P.-S. (]979) Pholocllem,PhotobioL29, 1177-1|97 2 Parson. B.J. (1980lPhotochcm. Photobiol.32, 813-821.

147 3 Redighiero, G., Dall'Acqua, F. and Pathak, M.A. (1984) in Topics in Phntomedieine (Smith, K.C., ed.), pp. 319-398, Plenum Press, New York. 4 Honigsmann. H. (1987) in Light in Biology and Medicine (DougI n . R.H., Moan, I., Dall'Acqua, F., eds.), Vol. 1, pp. 315-319, Plenum Pres.% New York. 5 For;at, P. (1989) in Psoralens, Past, Present and Future of Photochemoprotocfion and Other Biological Activities (Fitzpatrick, T.B., Forlot, P., Pathak, M.A. and Urbach, F., eds.), pp. 63-71, John Libbey Eurotext, Paris. 6 Pathak, M.A., Dalle C~rbonare. M. (1989) in Psoralens. Past, Present and Future of Photochemopr~tecBon and other Bia!ogical Activities (Fitzpatrick. T.B.. Forlot, P. Pathak, M.A. aiad Urbach. F., eds.), pp. 87-101. John Libbey Eurotexh Pads. 7 DalPAcqua, F. (1989) in Psoralens. Past, Present and Future of Photochemoprotection and other Binloglcal Activities (Fitzpatrick, T.B., Forlot, P., Pathak, M.A. mad Urhnch, F., eds.), pp. 237-250, John Libbey Eurotext, Paris. 8 P e Mol, N.J. and Beijersbergen van Henegouwen, G.M.J. (1981) Photochem. Photobiol. 33, 815-819. 9 JoshJ. P.C., Carraro. C. ~-ad Pathak. M.A. (1987) Biochem. Bigphys. Res. Commun. 142, 265-274. 10 Pathak. M.A. and Joshi. P.C. (1984) Biochim. Biophys. Acta 798. 118-126. 11 D'Ischia, M. and Prota. G. (1986) Gazz. Chim. Ital. 116, 407-410. 12 D'IschJa. M. and Prota, G. (1987) Tetrahndron 43. 431-434. 13 Prota. G. (1987) in Light in Biology and Medicine (Douglas. R.H., Moan, l., Dall'Acqua, F., eds.) Voi. •. pp. 329-336, Plenum Press. New York. 14 Pathak, M.A. (1967) in Advances in Biology of Skin (Montagna. W., Hu, F., eds.), "COL~;. pp. 397-420, Pergamon Press, Oxford. 15 Goldblum, R.W., Piper, W.N. and Olsen, C.J. (1955) J. Invest. Dermatok 25.139.

16 Knox, J.M. and Ogura, R. (1964) J. Invest. DermatoL 42. 9 5 - 9 9 17 Connor, M.L and Wheeler, L.A (1987) Photcmhem. Photobiul. 46, 239-245 18 Riddles, P.W., Blakeley, R.L. and Zerner, B. (1979) Anal. B i b ehnm. 94, 75-81. 19 Dawson, R.M.C., Elliott, D.C., Elhntt, W H . and Jones. K.M, (eds.) (1969) Data for Biochemical Research, gnd Ed., Oxford University Press. Oxford. 20 Beehara, E.J.H., Farla Oliveira, O.M.M., Duran, N , Casadei de Baptista, R. and Cilento, G. (1979) Photochem. Phntobiol. 30, 101 110. 21 Czffied, S., Vedaldi, 0,, Daga, A , DalPAcqua, F. (19891 in Psoralens. Past. Present and Future in Photochemoprotection and other Biological Activities (Filzoatriek, T.B., Forhn. P.. Pathak. M.A. and Urhz h, F., eds.), pp. 137-145, John Libbey Eurotext, Paris. 22 Land, E.J. and Truscott. T.(5". (1979) Photochem. Pholobiol. 29. 861-866. 23 Matsuura. T.. Yoshimura. N., Nishinaga, A. and Saito, I. (1972) Tetrahedron, 28. 4933-4938. 24 Prola. G.. D'lschia. M. and Napolitano. A. (1988) Pigment Cell Research Supplement 1.48-53. 25 Prota, G. (1988) Medicin. Res. Rev. 8, 525-556. 26 Halpdn, K.M. and Ohkawara, A. (1967) in Advances in the Biology of Skin (Montagna, W. and Hu, F., eds.), pp. 241 251, Pergamon Press. Oxford. 27 Prota. (3. (1989) in Psoralens. Past, Present and Future in Photochemoprotection and other Biological Acti'nties (Fitzpatrick, TB., Forlot. P., Pathak, M.A. and Urbach, F., eds.), pp. 13 23, John Libbey Eurotext, Paris.