Photochemistry and Photobiology, 2005, 81 : 206-21 1

Enhancement of Riboflavin-mediated Photo-Oxidation of Glucose 6-phosphate Dehydrogenase by Urocanic AcidT

Eduardo Silva*', Leonard0 Herrera', Ana Maria Edwards', Julio de la Fuente' and Eduardo L i d 3 'Facultad de Quimica, Pontificia Universidad Catolica de Chile, Santiago, Chile 2Facultad de Ciencias Quimicas y Farmaceuticas, Universidad de Chile, Santiago, Chile 3Facultad de Quimica y Biologia, Universidad de Santiago de Chile, Santiago, Chile

Received 14 July 2004; accepted 21 October 2004

ABSTRACT

We have investigated the riboflavin (RF)-sensitized inactivation of glucose 6-phosphate dehydrogenase (G6PD) in the presence and absence of trans-urocanic acid (UCA). The inactivation of the enzyme results from its direct oxidation by the excited triplet RF in a Type-I-photosensitized reaction whose efficiency increases at low oxygen concentration. The addition of histidine to the system produced no change in the inactivation rate, discarding the participation of singlet oxygen in the reaction. On the other hand, the presence of UCA results in its bleaching, with a significant enhancement of RF-mediated inactivation of G6PD. Both the consumption of UCA and G6PD are faster at low oxygen concentrations. UCA also produced a decrease in the sensitizer photodecomposition yield. These results indicate that the enhancement of the RF-mediated G6PD inactivation observed in the presence of UCA is not a singlet oxygen- mediated process. It is proposed that UCA consumption and its effect on G6PD inactivation are due to a complex reaction sequence initiated by a direct oxidation of UCA by the excited sensitizer triplet. The oxidation of the semireduced flavin gives rise to reactive oxygen species (ROS) responsible for the increased rate of the process. This is supported by the protection afforded by several additives with ROS removal capacity: benzoate, superoxide dismutase and catalase.

INTRODUCTION trans-Urocanic acid (UCA), (E)-3-(1H-imidazol-4-yl)propenoic acid, is synthesized in vivo from the amino acid histidine by the enzyme histidase (I). The photobiology of UCA is a matter of considerable interest because of its high concentration in the human skin (2), its potential role as natural photoprotecting agent (3,4), its interaction with biomolecules (5-7) and its role in photoinduced immunosuppression (8-12), currently associated with its light-

TPosted on the website on 29 December 2004. *To whom correspondence should be addressed: Departamento de Quimica

Fisica, Facultad de Quimica, Pontificia Universidad Catdica de Chile, Casilla 306, Correo 22, Santiago, Chile. Fax:56-2-686-4744; e-mail: esilva@puc.cl

Ahhreviutions: G6PD, glucose 6-phosphate dehydrogenase; NAD+, nicotinamide adenine dinucleotide; N B F 2 , p-nitro blue tetrazolium chloride; RF, riboflavin; ROS, reactive oxygen species; SOD, superoxide dismutase; T-T, triplet-triplet; UCA, urocanic acid.

0 2005 American Society for Photobiology 003 1 -8655/05

promoted trans to cis photoisomerization. In vivo photoprocesses mediated by UCA light absorption are restricted to the UV-A and UV-B range; hence, their rates are limited by the relatively low irradiance at these wavelengths (13). In this context, it could be important the occurrence of sensitized processes after light absorption by endogenous substances with significant absorbances in the visible range (14), such as riboflavin (RF). These processes could take place by mechanisms implying sensitizer-UCA reactions (Type I), singlet oxygen production (Type II) or triplet-triplet energy transfer. The triplet-triplet energy transfer should be important for sensitizers whose triplet energy is higher than 55 kcal (14). On a molar basis, and measuring the relative singlet oxygen-mediated abilities of various imidazole compounds to form endoperoxides and bleach N , N-dimethyl-4-nitrosoaniline, it was determined that UCA reacts with singlet oxygen two- to four-fold faster than free L-

histidine (15). The generation of a trans-annular peroxide has been proposed as responsible of the enhancement of RF-mediated photooxidation of doxorubicin by histidine and UCA (1 6). The ability of UCA to act as an electron donor, using as probe the highly electrophilic reagent, p-nitro blue tetrazolium chloride (NBP2), under conditions wherein N B p 2 absorbs light, has been studied (17). The reaction between the NBTf2 excited state and UCA gives rise to NBT- and the UCA radical-cation (17). Studies of RF- sensitized photooxygenation of indoles and proteins have revealed that the Type-I process is more efficient than the competing Type-I1 process (18-20). The electron transfer reaction of 3RF plays a crucial role in the flavin-sensitized oxidation reaction, where the one- electron-reduced RF (RF'-) and one-electron-oxidized substrates are formed. The purpose of this article is to study the mechanism of RF-sensitized oxidation of UCA and its effect on the RF-sensitized inactivation of glucose 6-phosphate dehydrogenase (G6PD). This enzyme was selected because of the simplicity of its activity determination, its presence in skin and its relevance in processes associated with reactive oxygen species (ROS) detoxification (21).

MATERIALS AND METHODS Chemicds. G6PD, sodium benzoate, catalase, glucose 6-phosphate, nicotinamide adenine dinucleotide (NAD+), RF, superoxide dismutase (SOD), UCA and all other chemicals were purchased from Sigma (St. Louis, MO), unless otherwise indicated.

G6PD activity assay. G6PD activity was determined before and after irradiation. The enzymatic activity was determined by the method of Langdon (22). The total reaction mixture (3 mL) consisted of 2 mL of a solutioncontaining 0.625 Uofenzyme in0.55 mMNAD+and 1 mLof a 1:l mixture of MgC12 (6.7 mM) and glucose 6-phosphate (1 mM). The activity was estimated from the increase in the absorption band at 340 nm (Hewlett

206

Photochemistry and Photobiology, 2005, 81 207

h > c ._ ._ c 9

8 a a

s

-13- 5% 0

-0- 100% 0,

00 05 1 0 1 5 20 2 5 30

Time (min)

Figure 1. Percentage of activity lost of aG6PD solution irradiated with visible light at different oxygen concentrations (5%, 20% and 100%) in the presence and absence of UCA (1.23 X lo4 M), using RF (3.5 X 44) as sensitizer.

Packard 8456 spectrophotometer, Colorado Springs, CO) over a 2 min period because of the formation of reduced form of nicotinamide adenine dinucleotide. Measurements were performed in triplicate and the results expressed as percentage of the activity measured at zero time of irradiation.

Irrudiation conditions. SolutionsofUCA(1.23X 104M),His(1.23X lo4 M), G6PD (2.8 X lop8 M) and RF (3.5 X M) were irradiated in a quartz cuvette (path length, 1 cm) with a 150 W, 24 V Osram Halogen Bellaphot (Munich, Germany) lamp from a slide projector. The light coming from the opticd system of the projector shows a negligible intensity in the near-W region (below 400 nm). During the irradiation, the solutions were bubbled with 5%, 20% or 100% oxygen or nitrogen. Both UCA and RF consumption were determined from the change in the sample absorption at 280 and 450 nm, respectively. UCA was determined at 280 nm to minimize the spectral interference of RF. Nevertheless, the estimated contribution of RF to the absorbance at 280 nm was subtracted in the UCA consumption determinations.

Laser flash photolysis und triplet quenching. Experiments were performed with the transient detection system described previously (19,20). For quenching of triplet RF, optimal results were obtained with RF in phosphate buffer at pH 7.4, with absorbances 0.4-0.6 at the excitation wavelength (355 nm). Solutions (2 or 3 mL) of RF, in phosphate buffer at pH 7.4, were bubbled with Nz for 20 min in a 10 mm path fluorescence quartz cell sealed with a septum. Immediately after purging, aliquots of UCA solutions were added through the septum for the quenching experiments. In these experiments, absorbances at the excitation wave- lengths of 355 and 446 nm (the maximum for RF) were checked after a few laser pulses to ensure the integrity of the solutions. The averaged decay of four traces was used to estimate the triplet lifetime.

Singlet quenching experiments. The fluorescence quenching of a 3.5 X 10" M RF solution in the presence of increasing amounts of UCA was determined in a Perkin Elmer LS 55 Luminescence Spectrometer (Beaconsfield, UK). The excitation and emission wavelengths were 450 and 520 nm, respectively. The experiments were performed in phosphate buffer at pH 7.4 and at room temperature.

Effect of ROS scavengers. Solutions of UCA (1.23 X lo4 M) and RF (3.5 X M). in phosphate buffer at pH 7.4 and 5% or 100% oxygen, were irradiated in the presence of either 300 U of catalase or 100 U of SOD, and the photobleaching of UCA was determined by the decrease in the absorption at 280 nm. The same experiments were also performed in the presence of the same amount of the denatured enzymes.

A solution of UCA (1.23 X lo4 M ) , RF (3.5 X lo-' M) and G6PD (2.8 X lo-* M) was irradiated in the absence and presence of 5 mM sodium benzoate, and the inactivation of the enzyme was determined in both conditions. The experiments were carried out in phosphate buffer at pH 7.4 and under 5% or 100% oxygen.

RESULTS RF-sensitized photoinactivation of GBPD in the presence and absence of UCA

Figure 1 shows the loss of activity suffered by a solution of G6PD irradiated with polychromatic visible light in the presence of RF

-.- 100% 0, -0- 1 OOSb 0, + GGPD

0 I I I 1 00 05 1 0 1 5 2 0

Time (min)

Figure 2. RF (3.5 X M)-sensitized decomposition of a UCA solution (1.23 X lo4 M) irradiated with visible light at 5%, 20% and 100% oxygen, both in the presence and absence of G6PD. The irradiations were carried out in 0.01 M phosphate buffer at pH 7.4.

(3.5 X lop5 M) under different oxygen concentrations (5%, 20% and 100% 0 2 ) . It can be observed that the efficiency of the inactivation increases when the present oxygen in the reaction medium decreases. However, the rate of the process is negligible in nitrogen-purged solutions. Similar response regarding the oxygen concentration was found when experiments were performed in the presence of UCA (1.23 X lo4 M) (Fig. 1). When both sets of results (with and without UCA) are compared at the same oxygen concentrations, it is concluded that the presence of UCA significantly enhances the RF-mediated photoinactivation of the enzyme. Because both, UCA and histidine, have been reported to enhance the RF-mediated photobleaching of doxorubicin per- formed in open cell culture dishes (16), it was of interest to test whether histidine caused such effect on the RF-sensitized inactivation of GBPD. The presence of histidine (1.23 X lo4 M) did not produce any effect on the RF-mediated inactivation, irrespective of the oxygen atmosphere used (data not shown). This lack of effect indicates that, in this system, histidine neither enhances the inactivation nor protects the enzyme against its oxidation, despite its great reactivity toward singlet oxygen (23).

RF-mediated photodecomposition of UCA irradiated alone or in the presence of G6PD

Figure 2 shows the decomposition of a UCA solution (1.23 X lo4 M) irradiated with polychromatic visible light in the presence of RF (3.5 X lop5 M) and under different oxygen concentrations (5%, 20% and 100% 02). On the other hand, there was no UCA consumption when the irradiation was performed in the absence of the sensitizer. Furthermore, as the data in Fig. 2 show, the effectiveness of the RF-photosensitized UCA decomposition increases with the decrease of oxygen concentration present in the reaction media. A slower rate of UCA consumption was observed when the RF-containing solution was irradiated under the same conditions but in the presence of G6PD (Fig. 2).

No decomposition of UCA was observed when irradiated in the presence of 0.1 mM K3Fe(CN)6, using RF as sensitizer and different oxygen concentrations (5%, 20% and 100%). No photobleaching of RF was observed under these experimental conditions.

To assess the contribution of reactive oxygen species to the UCA-photosensitized consumption, experiments were performed

Eduardo Silva et a/. 208

6 0 3 5

1 -0-,5% 0, , I . , 0

O 0 0.5 ' ' 2.0 0.0 0.5 1.0 1 5 2.0

Time (min) Time (min)

Figure 3. Left: RF (3.5 X M)-sensitized decomposition of UCA (1.23 X lo4 M ) when irradiated with visible light at 5% of oxygen in the absence and presence of catalase (300 U) and SOD (100 U). Right: Percentage of activity lost of a G6PD (2.8 X lo-* M) and UCA (1.23 X lo4 M ) solution irradiated with visible iight at 5% oxygen in the resence and absence of 5 mM sodium benzoate, using RF (3.5 X 10- M) as sensitizer.

t:

in the presence of catalase and SOD. These enzymes exerted no protective effect when they were added in denatured form to the irradiated system. On the other hand, the results obtained using the native enzymes (Fig. 3) show that both the additives significantly reduce the rate of RF-photosensitized UCA consumption. Similar results (data not shown) were obtained at higher oxygen pressures. The role of hydroxyl radicals in the inactivation of G6PD was assessed by experiments carried out in the presence of 5 mM benzoate. The results shown in Fig. 3 indicate that this additive partially protects the enzyme from its photosensitized inactivation. Similar effects were observed at higher oxygen concentrations (data not shown). The moderate effect observed can be interpreted in terms of a site-specific hydroxyl radical production at the protein surface (24).

Photobleaching of RF solutions irradiated with visible light under nitrogen atmosphere

RF is particularly sensitive to light, and its photobleaching can be experimentally determined by measuring the decrease of the absorption band in the 450 nm region. When RF is irradiated in the absence of other compounds, its bleaching is due to an intra- molecular photoreduction of the isoalloxazine ring system in which the ribityl side chain acts as electron donor (25). During oxidation of the side chain, fragmentation may occur to produce several photoproducts. The presence of an appropriate external reductant can also induce the photoreduction of the isoalloxazine ring, with the corresponding photobleaching. However, in this situation the ribityl chain is conserved unmodified. In the presence of oxygen, the externally mediated semi- or totally reduced form of RF can be reoxidized with a concomitant recovery of its color.

RF was irradiated with polychromatic visible light in the absence of oxygen, with the aim to maximize its photobleaching (Fig. 4). The addition of G6PD to the reaction mixture produced only a slight increase in the RF decomposition. When the same experiments were performed in the presence of UCA, or the mixture of UCA and G6PD, a significant decrease of the RF photobleaching was observed. Figure 5 shows the amount of UCA photobleached and RF consumed as a consequence of the exposure

60

50 -.- GGPD \ l lx \x 40 -A- GGPD + UCA -.---lB

-x- RF alone 30' ' I " " ' I ' " "

0 20 40 60 80 100 120

Time (s)

Figure 4. Photobleaching of an RF solution (3.5 X visible light in the presence of different compounds (2.8 X 1.23 X lo4 M UCA) and alone, in a N2 atmosphere.

M) irradiated with M G6PD,

of this mixture to visible light in the absence of oxygen. These data show that the amount of photodecomposed UCA is of the same order of magnitude but slightly smaller than that of the sensitizer. This would indicate that, under nitrogen, UCA is consumed in a near one to one process; hence, its total consumption would be limited by the RF concentration. The faster rate of RF consumption could be associated to the remaining self-reaction of the sensitizer.

Quenching assays

Figure 6A,B show the Stem-Volmer type plots for the 'RF and 3RF quenching by UCA, respectively. Bimolecular quenching rate constants, kq, obtained from the slope of these plots are 6.7 X lo9 (k,rs = 33.3 K' and 7s = 5 X lo-' s [26]) and 2.0 X lo* K' s-', respectively. At the UCA concentration used, 1.3 X lo4 M , these values imply that singlet quenching is negligible (less than l%), whereas triplet quenching is competitive to 3RF interaction with oxygen. In fact, from the 3RF intrinsic lifetime ( T ~ = 0.34 X lo4 s) and the quenching rate constant by oxygen (12 X 10' K1 sf'), Eq. (1) allows to evaluate the fraction of triplets quenched by UCA [ f ( 3 R F ) ~ ~ ~ ] under the different conditions used in this study.

-.- RF -0- UCA

00 50 100 150 200 250 300

Time (s)

Figure 5. RF and UCA photoconsumption when a solution of these compounds (3.5 X M RF and 1.23 X lo4 M UCA) is irradiated with visible light in a N2 atmosphere. The irradiations were performed in 0.01 M phosphate buffer at pH 7.4.

Photochemistry and Photobiology, 2005, 81 209

[UCA]x104 M 0.0 2.5 5.0 7.5 10.0

l ' l - l ~ l ' l

0 4 a 12 16

[UCA]x105 M Figure 6. (A) RF (3.5 X lo-' M) fluorescence quenching by UCA. (B) Lifetime of Rf triplet as a function of UCA concentration. Triplet lifetime estimated from the decay of triplet Rf absorption at 660 nm by UCA.

f(3RF),c, = ~ I C A ( ~ T ' + ~ucA[UCA] + ko2[02])-'[UCA] (1)

where kUCA and ko, represent the rate constant for the quenching of 3RF by UCA and 02, respectively. Table 1 shows the f(3RF)ucA values and the initial rates of UCA decomposition when the experiments are performed at different oxygen concentrations.

K3Fe(CN)6 has the property to quench 3RF with a k, = 1.74 X lo9 K1 s-', and when the quenching effect of 0.1 mM K3Fe(CN)6

> less favored process at low oxygen concmtr&mn + more favored process ai low oxygen concmtrabon

Table 1. Fraction of triplet RF quenched by UCA (Y3RF)uc~) and initial decomposition rate of UCA (V [Mlmin]) when solutions of UCA (1.23 X lo4 M) are irradiated with polychromatic light in the presence of RF (3.5 X

M) at different oxygen concentrations

0 5

20 100

0.45 0.22 0.09 0.02

6.0 X 8.6 X lo-' 5.6 X lo-' 2.5 X lo-'

is included in Eq. 1, the fraction of 3RF quenched by UCA is reduced to 0.091, 0.054 and 0.018 at 5%, 20% and 100% of oxygen, respectively.

DISCUSSION Visible light photosensitization by vitamin B2, RF, can take place by a Type-I mechanism, Type-11 mechanism involving singlet oxygen, or T-T energy transfer. In the present system, this last process can be disregarded because RF triplet lays 49.8 kcal above its ground state, making energy transfer to UCA energetically unfavored (14). The main results obtained in this study regarding RF-sensitized decomposition of UCA or the sensitized inactivation of G6PD (or both) can be summarized as follows. (1) UCA reduces the rate of RF photodecomposition initiated by intramolecular redox reactions. (2) UCA is consumed and G6PD is inactivated by RF photosensitization. RF-sensitized inactivation of G6PD is reduced in the presence of benzoate (a hydroxyl radical scavenger). These processes require oxygen, but their rate decrease when the oxygen concentration increases. (3) UCA-sensitized photocon- sumption is reduced in the presence of highly electrophilic compounds (&Fe(cN),) or ROS scavengers (catalase and SOD). (4) UCA addition increases the RF-photosensitized rate of G6PD inactivation. (5) G6PD addition reduces the RF-photosensitized rate of UCA decomposition.

These results can be explained in terms of Scheme 1. In this Scheme we have not included the interaction with RF-excited singlet because of the low UCA concentration used. Also, it is considered that a Type-11 mechanism involving singlet oxygen produced in 3 ~ - 0 2 interaction is not the main pathway for UCA

hv "2 RF ___+ I R F ------+ 3 R F - 1 RF + '0, > UCA,, 06PD

p' G6PD

0, RF +ucA+- prO&ctS

+RF -, 1 ,,,UCA(-H)OO 0,i '02

Fu-0, OHI H20,1 0,

H,02+RF 4- RFH2+RF 4- RFH +ucA(-H)

Scheme 1.

210 Eduardo Silva et a/.

consumption and G6PD inactivation. This conclusion is based in the decrease of the rate of these processes observed at higher oxygen pressures.

The results included in Table 1 clearly show that minimal concentrations of oxygen are necessary to promote UCA consump- tion. This is accounted for by the role of oxygen in secondary reactions (Scheme 1). On the other hand, higher oxygen concen- trations decrease the rate of the process, emphasizing the importance of the direct interaction between the excited RF triplet and UCA. However, it is interesting to note that the decrease in rate elicited by increasing the oxygen concentration is less than that expected by the diminution in 'RF-UCA interaction, in agreement with the role of oxygen in the secondary reactions.

A reduction of both UCA and RF decomposition when the experiments are performed in the presence of K3Fe(CN)6 can be partially explained as due to its quenching effect on 3RF. However, the total protection afforded must be attributed to the additional property of this compound to interfere with electron transfer processes (18). This result is also a proof of the initial direct chemical interaction between the excited flavin and UCA. Quenching experiments indicate that UCA can interact both with the singlet and triplet state of the flavin. The action of visible light on RF conduces to its irreversible photodecomposition initiated by an intramolecular redox reaction between the ribityl side chain and the isoalloxazine ring. During oxidation of the side chain, fragmentation may occur, leading to several photoproducts (25) . The protective effect of UCA on the photodecomposition of the flavin (Fig. 4) is due to the competitive direct oxidation of UCA by the excited flavin that gives rise to the semireduced flavin radical, which in turn can be reconverted to its native form in the presence of oxygen (see Scheme).

It is considered that, in the conditions of the present study, oxidized products produced in singlet oxygen-UCA reactions are not involved in the mechanism leading to an enhanced G6PD inactivation by UCA. The enhancement of RF-mediated photoox- idation of doxorubicin by histidine and UCA has been reported previously (16), and it has been proposed to be due to the formation of a trans-annular peroxide of the imidazole ring. In the RF-mediated inactivation of G6PD studied in this article, the amino acid and UCA showed different behaviors. The presence of histidine seems to have no effect on the G6PD inactivation, whereas the addition of UCA generates a significant enhancement that was still higher when the experiments were performed at lower oxygen concentrations. The lack of effect in the presence of histidine allows us to discard an important participation of singlet oxygen on the RF-mediated oxidation of the enzyme. The in- creased enhancement of the G6PD inactivation in the presence of UCA and the fact that it is favored at low oxygen concentrations is in this case also incompatible with a singlet oxygen-mediated reaction. This result points out to a mechanism that involves the direct interaction between UCA and the excited flavin, giving rise to the generation of the RF semireduced radical and to the corresponding UCA radical cation. This reaction path is similar to the reported photolytic reduction of N B F 2 excited state by either UCA or its ethyl esters (17) that gives rise to the corresponding radicals. As can be seen in Scheme 1, the RF anion radical can react with oxygen-producing anion superoxide and two molecules of the neutral RF radical can give rise to one molecule of RF in the reduced form and one RF in the oxidized form. Reduced RF in the presence of oxygen is immediately reoxidized with the corre- sponding production of hydrogen peroxide. Under these reaction

conditions, the anion superoxide, together with hydrogen peroxide, can give rise to 'OH through a Haber-Weiss reaction and, in consequence, three ROS can simultaneously be present in the medium. The generation of these ROS in the photosensitized decomposition of UCA is supported by the protection afforded by catalase or SOD. The high oxidative capacity of these ROS can also explain the enhancement of the RF-mediated photoinactiva- tion of G6PD in the presence of UCA.

Acknowledgements-This work was supported by FONDECYT (Grant I030033l

REFERENCES

1. Baden, H. P. and M. A. Pathak (1967) The metabolism and function of urocanic acid in skin. J . Investig. Dermatol. 48, 11-17.

2. Norval, M., T. J. Simpson, E. Bordshin and J. Crosby (1989) Quantification of urocanic acid isomers in human stratum comeum. Photodermatology 6, 142-145.

3. Momson, H. (1985) Photochemistry and photobiology of urocanic acid. Photodermatology 2, 158-165.

4. Kavanagh, G., J. Crosby and M. Norval (1995) Urocanic acid isomers in human skin: analysis of site variation. Br. J . Dermatol. 133, 728-731.

5. Deibel, R. M., H. Morrison and W. M. Baird (1987) Urocanic acid photobiology. Photochemical binding of urocanic acid to bovine serum albumin. Photochem. Photobiol. 45, 20&211

6. Morrison, H., B. Mauclair, R. M. Deibel, G. Pandey and W. M. Baird (1985) Urocanic acid photobiology. Photochemical binding to calf thymus DNA. Photochem. Photobiol. 41, 251-257.

7. Farrow, S . J., T. Mohammad, W. Baird and H. Morrison (1990) Photochemical covalent binding of urocanic acid to polynucleic acids. Interact. 75, 105-1 18.

8. Norval, M., T. J. Simpson and J. A. Ross (1989) Urocanic acid and immunosuppression. Photochem. Photobiol. 50, 267-275.

9. Noonan, F. P. and E. C. De Fabo (1992) Immunosuppression by ultraviolet B radiation: initiation by urocanic acid. Immunol. Today 13, 250-254.

10. Norval, M., N. K. Gibbs and J. Gilmour (1995) The role of urocanic acid in UV-induced immunosuppression: recent advances (1992-1994). Photochem. Photobiol. 62, 209-217.

11. Norval, M. (1996) Chromophore for UV-induced immunosuppression: urocanic acid. Photochem. Photobiol. 63, 386-390.

12. Sleijffers, A,, A. Kammeyer, F. R. De Gruijl, G. J. Boland, J. van Hattum, W. A. van Vloten, H. van Loveren, M. B. M. Teunissen and J. Garssen (2003) Epidermal cis-urocanic acid levels correlate with lower specific cellular immune responses after hepatitis B vaccination of ultraviolet B-exposed humans. Photochem. Photobiol. 77, 271-275.

13. Haralampus-Grynaviski, N., C. Ransom, T. Ye, M. Rozanowska, M. Wrona, T. Sama and J. D. Simon (2002) Photogeneration and quenching of reactive oxygen species by urocanic acid. J . Am. Chem.

14. Mohammad, T., H. Morrison and H. HogenEsch (1999) Urocanic acid photochemistry and photobiology. Photochem. Photobiol. 69, 115-135.

15. Hartmant, P. E., 2. Hartmant and K. T. Ault (1990) Scavenging of singlet molecular oxygen by imidazole compounds: high and sustained activities of carboxy terminal histidine dipeptides and exceptional activity of imidazole-4-acetic acid. Photochem. Photobiol. 51, 59-66.

16. Ratnu, A., M. M. Mehta, T. Leaseburg and A. Aleksic (2001) The enhancement of riboflavin-mediated photo-oxidation of doxorubicin by histidine and urocanic acid. Cancer Chemother. Pharmocol. 47,

17. Morrison, H. and R. M. Deibel (1988) Urocanic acid photobiology. Photooxidation and superoxide formation. Photochem. Photobiol. 48, 153-156.

18. Silva, E., R. Ugarte, A. Andrade and A. M. Edwards (1994) Riboflavin- sensitized photoprocesses of tryptophan. J . Photochem. Photobiol. B: Biol. 23, 43-48.

SOC. 124, 3461-3468.

338-346.

19. Silva, E., M. Jopia, A. M. Edwards, E. Lernp, J. R. De la Fuente and E. Lissi (2002) Protective effect of boldo and tea infusions on the visible light-mediated pro-oxidant effects of vitamin B2, riboflavin. Photo- chem. Photohiol. 75, 585-590.

20. Viteri, G. , A. M. Edwards, J. De la Fuente and E. Silva (2003) Study of the interaction between triplet riboflavin and a-, PH- and pLcrystallins of the eye lens. Photochem. Photobiol. 77, 535-540.

21. Balin, A. K., L. Pratt and R. G. Allen (2002) Effects of ambient oxygen concentration on the growth and antioxidant defenses of human cell cultures established from fetal and postnatal skin. Free Radic. Biol. Med. 32, 257-267.

Photochemistry and Photobiology, 2005, 81 21 1

22. Langdon, R. (1996) Glucose-6-phosphate dehydrogenase from eryth- rocytes. Methods Enzymol. 9, 126-131.

23. Davies, M. J. (2004) Reactive species formed on proteins exposed to singlet oxygen. Photochem. Photobiol. Sci. 3, 17-25.

24. Edwards, A. M., M. Ruiz, E. Silva and E. Lissi (2002) Lysozyme modification by the Fenton reaction and gamma radiation. Free Radic. Res. 36, 277-284.

25. Cairns, W. L. and D. E. Metzler (1971) Photochemical degradation of flavins. VI. A new photoproduct and its use in studying the photolytic mechanism. J . Am. Chem. Soc. 93, 2772-2777.

26. Heelis, P. F. (1982) The photophysical and photochemical properties of flavins (isoalloxazines). Chem. Soc. Rev. 11, 15-39.