J. Photo&em.
Photobiol. B: Biol., I I (1991)
57
57-68
Effects of pH and ionic micelles on the riboflavinsensitized photoprocesses of tryptophan in aqueous solution E. Silva + and V. Riickert Facultad
de Q&mica,
P0ntQX.a
Universidad
Catolica de Chile, CasiUa 6177, Santiago
[Chile)
E. Lissi and E. Abuin Departamento de Quimica, Facultad ok Ciencia, Casi1l.a 307, Ccn-reo 2, Santiago {Chile) (Received
November
Keywwd~.
5, 1990;
accepted
Photosensitization,
Universidad
d.e Santiago
o?e Chile,
February 25, 1991)
riboflavin, tryptophan, pH effect, fluorescence.
Abstract The effects of pH and ionic micelles on the rates of product formation following irradiation of riboflavin in the presence of tryptophan were investigated by absorption and fluorescence spectroscopy. Under anaerobic conditions, formation of riboflavin-tryptophan adducts was inhibited in acid solutions and by the addition of anionic (sodium dodecylsulphate) and cationic (cetyltrimethylammonium bromide) micelles. The oxidation of tryptophan photoinduced by riboflavin was considerably faster in basic solutions.
1. Introduction The role of free and protein-bound tryptophan in the light sensitivity of biological systems has prompted numerous studies of the photoprocesses of this ammo acid, either following the direct absorption of light by the indole group or promoted by compounds acting as sensitizers [ 11. Of the sensitized photoprocesses, those involving the riboflavin-tryptophan system are of particular interest due to the endogenous nature of the sensitizer. Senile nuclear cataractogenesis and aging of the human lens have been associated with a singlet-oxygen-mediated photodynamic process promoted by endogenous sensitizers such as riboflavin (I> and/or formyl kynurenine (II) [ 2, 31. Although tryptophan is an essential amino acid and riboflavin is a vitamin, the simultaneous presence of these two compounds in living organisms has been related to hepatotoxic and cytotoxic effects during parenteral nutrition [ 4,5] and in culture media [ 6, 71 respectively, on exposure ‘Author
to whom correspondence should be addressed.
loll-1344/91/$3.50
0 1991
-
Elsevier
Sequoia,
Lausanne
58
to the action of light. Recently, the generation of an adduct between riboflavin and tryptophan following irradiation under anaerobic conditions has been described [El] and this compound has been proposed to be one of the photoproducts which may be associated with hepatic dysfunction [9] and cytotoxicity in irradiated culture media [ 101. Photoadduct formation by intermolecular photoreduction of riboflavin, promoted by substrates of low oxidation potential, is a common process involving addition to the sensitizer at the C4a, C*” or N5 positions [ll]. The structure proposed for the riboflavin-tryptophan adduct (III) is shown in Scheme 1 [ 12 1. The photo-oxidation of tryptophan sensitized by riboflavin is characterized by higher quantum yields [ 131 than those observed in the presence of sensitizers, such as methylene blue or rose bengal, which involve a type II photo-oxidation mechanism [ 141. This behaviour suggests that riboflavin may act via a different sensitizing mechanism from that accepted for the other sensitizers. Similar conclusions have been obtained in a recent study on the lumiflavin-sensitized photo-oxygenation of indole [ 151. In studies of the photodynamic effect in the lysozyme-riboflavin system, Silva and Gaule [ 161 have demonstrated the formation of a photoinduced complex between the enzyme and the sensitizer concurrent with photooxidation. It has subsequently been shown that the lysozyme-riboflavin complex formed involves tryptophan residues [ 171 and the photoinduced riboflavin-tryptophan adduct has been isolated and characterized [S]. The formation of this adduct has been associated with the higher efficiency of CH20H dHOHj3 I
Riboflavin
CH OH I 2
Photo -adduct
( 111) Scheme
1.
N-formylkynurenine (11)
(1)
59
riboflavin in the sensitization of the photoinactivation of lysozyme in comparison with other sensitizers which do not bind to the enzyme (e.g. methylene blue) [ 181. The evidence given above indicates that flavins may play an important role as sensitizers of tryptophan photoreactions, leading to damage of biological systems, and that the sensitization reactions of flavins are complex. In this work, we report the results of a study on the effects of pH and ionic micelles on the riboflavin-sensitized photoprocesses of tryptophan. If the riboflavin-txyptophan photoreactions take place through a bimolecular interaction between the excited state of riboflavin and tryptophan, the most probable initial step must involve a charge transfer interaction [ 151. The rate of the process should therefore be controlled by the local concentration of the reagents, their protonation states and the feasibility of producing a stable ion pair. In order to evaluate the influence of protonation, the photoprocesses were investigated at different pH values under nitrogen and in oxygensaturated solutions. Local concentration effects and the effect of stabilization of the intermediate ionic species were analysed by carrying out the photolyses in ionic micellar solutions. The presence of strong local electric fields in these solutions should exert noticeable effects on the quantum yields and the nature of the products arising through charge transfer mechanisms [ 19 ].
2. Experimental
details
Riboflavin (Rb) and tryptophan (Trp) were obtained from Sigma Chemical Co. and were used as received. Sodium dodecylsulphate (SDS) (BDH, specially pure) and cetyltrimethylammonium bromide (CTAB) (Fluka) were employed as supplied. The photolyses were performed in a cell (path length, 1 cm) thermostatically controlled at 37 “C. Anaerobic or aerobic conditions were achieved by bubbling nitrogen or oxygen respectively. Light from a 150 W slide projector lamp equipped with an interference filter was employed for irradiation of Rb at 452 run. Samples were irradiated in phosphate buffer at pH 3 and 7 and in Tris-HCl buffer at pH 10. Absorption spectra were recorded on a Varian Super Scan III spectrophotometer. Fluorescence measurements were performed with a Perkin-Elmer 650-10s fluorescence spectrometer. Oxygen uptake measurements were carried out using a YSI Oxygen Monitor model 5300.
3. Results
and discussion
Photolysis of Rb (0.035 mM) with light of 452 nm in the absence of tryptophan leads to noticeable bleaching of the absorption at 450 run and to a reduction in the fluorescence intensity at 520 nm. The rates of both changes increase significantly when the pH of the solution is increased from 3 to 7 to 10. Bleaching of Rb at all pH values takes place without significant
60
alteration of the absorbance in the 300-400 nm region. The only change observed is an increase in absorbance at wavelengths longer than 500 nm which can be assigned to Rb dimerization [20]. The presence of Trp (1.91 n-M), under both aerobic and anaerobic conditions, leads to a significant decrease in the Rb photobleaching rates and to the formation of products with absorption in the 300-400 run region. Figure l(A) shows the temporal changes of the absorption spectra after irradiation under anaerobic conditions. The spectrophotometric alteration observed in the 300-400 run region can be assigned to the formation of the Rb-Trp adduct [8]. Figure l(B) shows the temporal changes of the absorption spectra after irradiation under aerobic conditions. The increase in the absorbance in the 300-400 nm region in the presence of oxygen may have a composite origin: it may be due to adduct formation (as under nitrogen) or it may arise from the formation of tryptophan oxidation products [21]. The observation that the absorbance increases faster under oxygen than under nitrogen indicates the predominance of the oxidative pathway under aerobic conditions. The changes in optical density measured at 325 nm are plotted in Fig. 2 as a function of time. Increasing the solution pH induces an increase in the absorbance in all cases. The effect is not due to changes in the absorption coefficients with pH since spectra generated by photolysis at pH 3 are not modified when the samples
05
w 0
2
a
m
00
LOO
500
LOO
500
a
5:
10
300
Lrn
5m
z&a
5w
x.nm
-
m a
3
Fig. 1. Temporal changes in the absorption spectra measured after irradiation of Rb (0.035 mM; 452 mn) in the presence of Trp (1.91 mM) in aqueous solutions at different pH values (irradiation time between spectra, 15 min): (A) under anaerobic conditions; (B) in oxygensaturated solutions.
61
pH 31) (N,) --x15
30
L5
60
MINUTES
Fig. 2. Time dependence experimental conditions.
of the changes
in optical density measured
at 325 run under different
are made alkaline after irradiation. The dependence of the absorbance on pH indicates that an increase in pH improves the efficiency of adduct formation and Trp photo-oxidation. Further information about the effect of pH on the photoprocesses in the absence and presence of oxygen was obtained by following the temporal changes in the fluorescence spectra and by measuring the rates of tryptophan consumption and oxygen uptake. Photoirradiation of Rb in the absence of Trp, at pH 7, leads to lumichrome (IV) as the main product [ 22 1. Figure 3 shows the temporal changes of the fluorescence spectra obtained at the different pH values in experiments performed in the absence of oxygen (Pig. 3(A)) and in oxygen-saturated solutions (Pig. 3(B)), following excitation with light of 325 nm. This wavelength was selected for excitation since it is not absorbed by Trp and corresponds to a minimum in the Rb absorption spectrum, and thus is suitable for the detection of emission arising from the photoproducts. The spectra obtained at pH 3 and 7 under anaerobic conditions show an increase in the emission intensity of a band at 460 run which may be due to the formation of lumichrome [22, 231. However, at pH 7, the amount of lumichrome formed (as evaluated by the intensity of the fluorescence at these wavelengths) is less than 20% of that obtained when Rb is irradiated alone under similar conditions (data not shown), indicating that most of the Rb is consumed by a different process involving added Trp, in agreement with the conclusion reached from the analysis of the changes in the absorption spectra. The fluorescence spectra measured under anaerobic conditions at pH 10 show the presence of two bands centred at 460 and 510 run. The band at 460 run may be due to hrmichrome, which is one of the main products of Rb photolysis at high pH [23]. However, the total fluorescence observed in this region is larger than that observed in the absence of Trp at total Rb consumption, indicating a contribution of Tr-p-derived photo-
62
A
,H
o
I
3.0
A
570
pH 70
510
>H 10.0
L50
B pH 3.0
L
,I
IO’,
(50
pH 7.0
190
330
,
sm
,
,
,
,
510
El3
390
3M
570
510
I.50
390
330
X,nm
F’ig. 3. Temporal changes in the fluorescence spectra measured after irradiation of Rb (0.035 mM; 452 run) in the presence of Trp (1.91 mM) in aqueous solutions at different pH values (irradiation time between spectra, 15 min): (A) under anaerobic conditions; (B) under aerobic conditions.
products to the recorded emission. Thus, from the absorption and emission spectra, it can be concluded that the presence of Trp leads to the formation of new products which absorb in the 300-400 run region and which present weak fluorescence in the 400-480 nm region. These spectroscopic characteristics most probably correspond to the Rb-Trp adducts. Fluorescence spectra recorded at pH 3 and 7 following irradiation under aerobic conditions (Fig. 3(B)) show an increase in a band at about 430 run which corresponds to the fluorescence of formyl kynurenine [ 241. Concomitant formation of kynurenine is difficult to assess from the fluorescence spectra due to interference by the remaining flavin. At pH 10, fluorescence spectra recorded under aerobic conditions closely resemble those obtained under nitrogen, pointing to the predominance of emission from the flavin-derived products. However, even under these conditions, fluorescence yields in the 400480 run region are larger than those measured in the absence of Trp or under anaerobic conditions, demonstrating the significant contribution of
63
Trp-derived oxidation products to the total fluorescence measured in this region. Trp consumption was measured by following the decrease in the fluorescence intensity at 350 nm after excitation with light of 285 nm. (Measurements were made after dilution of the reaction mixture until the fluorescence intensity was a linear function of tryptophan concentration.) The shape of the fluorescence band observed when the excitation is carried out at 285 run does not change significantly with the time of Rb irradiation, showing that reaction products do not contribute to the fluorescence observed under these conditions. In experiments carried out under anaerobic conditions, no significant Trp decrease is observed even after total Rb consumption. Given the large excess of Trp present, this result is compatible with the formation of Trp-Rb adducts. In contrast, under aerobic conditions, Trp is significantly consumed, showing that its removal must take place via a process which is sensitized by Rb excitation but does not involve a stoichiometric consumption of the sensitizer. Furthermore, the observation that Trp consumption yields are considerably higher than those obtained when other sensitizers are employed [ 13 1, implies that a type I photo-oxidation mechanism is involved [ 151. The rate of the process is dependent on pH: the consumption rates (M mm-‘) are 3.27 X 10e6 at pH 3, 3.26~ lop5 at pH 7 and 6.5 X 1O-5 at pH 10, these correspond to Trp photoconsumption quantum yields of 0.022, 0.22 and 0.43 respectively. The increase in Trp consumption is paralleled by an increase in the rate of oxygen uptake (Fig. 4). The initial rates of oxygen uptake (M mm-‘) are 8 x 10m6 at pH 3, 1.3X 10T4 at pH
MINUTES
Fig.
4.
Effect of pH on the rate of oxygen uptake from oxygen-saturated solutions.
64
7 and 2.2X 10m4 at pH 10. These data show that, irrespective of the pH, nearly three oxygen molecules are consumed by each oxidized Trp molecule. The results show that an increase in pH increases the rate of the process under anaerobic conditions, where the principal process is Trp-Rb adduct formation, and under aerobic conditions, where the main Trp-consuming process is an efficient oxidation which has a quantum yield of nearly unity at pH 10. In order to interpret these results it is necessary to consider the effect of pH on the properties of the Rb excited states and the possible intermediates involved in the process. The Rb excited singlet is quenched by protons and the inter-system crossing quantum efficiency changes from 0.40 at pH 2.2 to 0.7 at pH 7 [22]. Adduct formation will therefore be favoured at higher pH values, irrespective of the Rb excited state involved in the process. Another factor which must be considered is the relative charge of the reaction partners. The ground state pK, value of Rb is extremely low (pK, = 0 [ 251) and that of Trp is 2.8. However, since adduct formation takes place after Rb excitation, the acid-base properties of the excited state, instead of those of the ground state, should be considered. The pK, value of triplet Rb has been reported to be 6.5 [23]. Excitation of Rb at pH 3 therefore gives rise to the protonated triplet which encounters about half of the Trp also in protonated form at this pH. Coulombic repulsion between the reaction partners may inhibit adduct formation in acid solution. Furthermore, it has been shown that the fast interaction of the hnniflavin triplet with indole leads to the formation of the indolyl cation radical and the semiquinone anion radical [ 151. If adduct formation takes place through the interaction of these species, the product yield may be affected by the protonation state of the two species, with p&=4.3 [26] and p&=2.3 [15] respectively. Lumiflavin-sensitized photo-oxygenation of indole in water has been considered to arise mainly from the interaction of the indolyl cation radical with the superoxide anion produced in the oxidation of the semiquinone anion radical by oxygen [ 151. If a similar pathway is assumed for the photooxygenation of Trp in this work, the pH effect observed may also be due to changes in the protonation state of the superoxide anion (pK,= 4.8 (271). However, it is relevant to consider that, given the pK, value of the indolyl cation radical, it is very unlikely that the reaction at pH 10 involves the cation radical and, if superoxide is involved, the reaction should take place between it and the neutral indolyl radical. Micellar effects on the reactions were investigated by carrying out the photolyses in solutions of anionic (SDS) and cationic (CTAR) surfactants. Any study of micellar effects requires an evaluation of the partitioning of the reagents between the micelles and the aqueous pseudophase. Table 1 shows the percentage of incorporation of Rb and Trp into the micellar pseudophases (70 mM concentration of the surfactants) at the three pH values employed. The different capacities of SDS and CTAR micelles for incorporating the reagents, and the effect of pH on the amount incorporated into a given micelle, can easily be rationalized in terms of hydrophobic
65 TABLE
1
Percentage
of Rb and Trp incorporated
mM; temperature,
into micelles in solution (surfactant concentration,
Percentage
Surfactant
Reagent”, b
SDS
Trp Rb
3
>90 70
Trp
7
44 71
10
Photobiol. B: Biol., I I (1991)
57
57-68
Effects of pH and ionic micelles on the riboflavinsensitized photoprocesses of tryptophan in aqueous solution E. Silva + and V. Riickert Facultad
de Q&mica,
P0ntQX.a
Universidad
Catolica de Chile, CasiUa 6177, Santiago
[Chile)
E. Lissi and E. Abuin Departamento de Quimica, Facultad ok Ciencia, Casi1l.a 307, Ccn-reo 2, Santiago {Chile) (Received
November
Keywwd~.
5, 1990;
accepted
Photosensitization,
Universidad
d.e Santiago
o?e Chile,
February 25, 1991)
riboflavin, tryptophan, pH effect, fluorescence.
Abstract The effects of pH and ionic micelles on the rates of product formation following irradiation of riboflavin in the presence of tryptophan were investigated by absorption and fluorescence spectroscopy. Under anaerobic conditions, formation of riboflavin-tryptophan adducts was inhibited in acid solutions and by the addition of anionic (sodium dodecylsulphate) and cationic (cetyltrimethylammonium bromide) micelles. The oxidation of tryptophan photoinduced by riboflavin was considerably faster in basic solutions.
1. Introduction The role of free and protein-bound tryptophan in the light sensitivity of biological systems has prompted numerous studies of the photoprocesses of this ammo acid, either following the direct absorption of light by the indole group or promoted by compounds acting as sensitizers [ 11. Of the sensitized photoprocesses, those involving the riboflavin-tryptophan system are of particular interest due to the endogenous nature of the sensitizer. Senile nuclear cataractogenesis and aging of the human lens have been associated with a singlet-oxygen-mediated photodynamic process promoted by endogenous sensitizers such as riboflavin (I> and/or formyl kynurenine (II) [ 2, 31. Although tryptophan is an essential amino acid and riboflavin is a vitamin, the simultaneous presence of these two compounds in living organisms has been related to hepatotoxic and cytotoxic effects during parenteral nutrition [ 4,5] and in culture media [ 6, 71 respectively, on exposure ‘Author
to whom correspondence should be addressed.
loll-1344/91/$3.50
0 1991
-
Elsevier
Sequoia,
Lausanne
58
to the action of light. Recently, the generation of an adduct between riboflavin and tryptophan following irradiation under anaerobic conditions has been described [El] and this compound has been proposed to be one of the photoproducts which may be associated with hepatic dysfunction [9] and cytotoxicity in irradiated culture media [ 101. Photoadduct formation by intermolecular photoreduction of riboflavin, promoted by substrates of low oxidation potential, is a common process involving addition to the sensitizer at the C4a, C*” or N5 positions [ll]. The structure proposed for the riboflavin-tryptophan adduct (III) is shown in Scheme 1 [ 12 1. The photo-oxidation of tryptophan sensitized by riboflavin is characterized by higher quantum yields [ 131 than those observed in the presence of sensitizers, such as methylene blue or rose bengal, which involve a type II photo-oxidation mechanism [ 141. This behaviour suggests that riboflavin may act via a different sensitizing mechanism from that accepted for the other sensitizers. Similar conclusions have been obtained in a recent study on the lumiflavin-sensitized photo-oxygenation of indole [ 151. In studies of the photodynamic effect in the lysozyme-riboflavin system, Silva and Gaule [ 161 have demonstrated the formation of a photoinduced complex between the enzyme and the sensitizer concurrent with photooxidation. It has subsequently been shown that the lysozyme-riboflavin complex formed involves tryptophan residues [ 171 and the photoinduced riboflavin-tryptophan adduct has been isolated and characterized [S]. The formation of this adduct has been associated with the higher efficiency of CH20H dHOHj3 I
Riboflavin
CH OH I 2
Photo -adduct
( 111) Scheme
1.
N-formylkynurenine (11)
(1)
59
riboflavin in the sensitization of the photoinactivation of lysozyme in comparison with other sensitizers which do not bind to the enzyme (e.g. methylene blue) [ 181. The evidence given above indicates that flavins may play an important role as sensitizers of tryptophan photoreactions, leading to damage of biological systems, and that the sensitization reactions of flavins are complex. In this work, we report the results of a study on the effects of pH and ionic micelles on the riboflavin-sensitized photoprocesses of tryptophan. If the riboflavin-txyptophan photoreactions take place through a bimolecular interaction between the excited state of riboflavin and tryptophan, the most probable initial step must involve a charge transfer interaction [ 151. The rate of the process should therefore be controlled by the local concentration of the reagents, their protonation states and the feasibility of producing a stable ion pair. In order to evaluate the influence of protonation, the photoprocesses were investigated at different pH values under nitrogen and in oxygensaturated solutions. Local concentration effects and the effect of stabilization of the intermediate ionic species were analysed by carrying out the photolyses in ionic micellar solutions. The presence of strong local electric fields in these solutions should exert noticeable effects on the quantum yields and the nature of the products arising through charge transfer mechanisms [ 19 ].
2. Experimental
details
Riboflavin (Rb) and tryptophan (Trp) were obtained from Sigma Chemical Co. and were used as received. Sodium dodecylsulphate (SDS) (BDH, specially pure) and cetyltrimethylammonium bromide (CTAB) (Fluka) were employed as supplied. The photolyses were performed in a cell (path length, 1 cm) thermostatically controlled at 37 “C. Anaerobic or aerobic conditions were achieved by bubbling nitrogen or oxygen respectively. Light from a 150 W slide projector lamp equipped with an interference filter was employed for irradiation of Rb at 452 run. Samples were irradiated in phosphate buffer at pH 3 and 7 and in Tris-HCl buffer at pH 10. Absorption spectra were recorded on a Varian Super Scan III spectrophotometer. Fluorescence measurements were performed with a Perkin-Elmer 650-10s fluorescence spectrometer. Oxygen uptake measurements were carried out using a YSI Oxygen Monitor model 5300.
3. Results
and discussion
Photolysis of Rb (0.035 mM) with light of 452 nm in the absence of tryptophan leads to noticeable bleaching of the absorption at 450 run and to a reduction in the fluorescence intensity at 520 nm. The rates of both changes increase significantly when the pH of the solution is increased from 3 to 7 to 10. Bleaching of Rb at all pH values takes place without significant
60
alteration of the absorbance in the 300-400 nm region. The only change observed is an increase in absorbance at wavelengths longer than 500 nm which can be assigned to Rb dimerization [20]. The presence of Trp (1.91 n-M), under both aerobic and anaerobic conditions, leads to a significant decrease in the Rb photobleaching rates and to the formation of products with absorption in the 300-400 run region. Figure l(A) shows the temporal changes of the absorption spectra after irradiation under anaerobic conditions. The spectrophotometric alteration observed in the 300-400 run region can be assigned to the formation of the Rb-Trp adduct [8]. Figure l(B) shows the temporal changes of the absorption spectra after irradiation under aerobic conditions. The increase in the absorbance in the 300-400 nm region in the presence of oxygen may have a composite origin: it may be due to adduct formation (as under nitrogen) or it may arise from the formation of tryptophan oxidation products [21]. The observation that the absorbance increases faster under oxygen than under nitrogen indicates the predominance of the oxidative pathway under aerobic conditions. The changes in optical density measured at 325 nm are plotted in Fig. 2 as a function of time. Increasing the solution pH induces an increase in the absorbance in all cases. The effect is not due to changes in the absorption coefficients with pH since spectra generated by photolysis at pH 3 are not modified when the samples
05
w 0
2
a
m
00
LOO
500
LOO
500
a
5:
10
300
Lrn
5m
z&a
5w
x.nm
-
m a
3
Fig. 1. Temporal changes in the absorption spectra measured after irradiation of Rb (0.035 mM; 452 mn) in the presence of Trp (1.91 mM) in aqueous solutions at different pH values (irradiation time between spectra, 15 min): (A) under anaerobic conditions; (B) in oxygensaturated solutions.
61
pH 31) (N,) --x15
30
L5
60
MINUTES
Fig. 2. Time dependence experimental conditions.
of the changes
in optical density measured
at 325 run under different
are made alkaline after irradiation. The dependence of the absorbance on pH indicates that an increase in pH improves the efficiency of adduct formation and Trp photo-oxidation. Further information about the effect of pH on the photoprocesses in the absence and presence of oxygen was obtained by following the temporal changes in the fluorescence spectra and by measuring the rates of tryptophan consumption and oxygen uptake. Photoirradiation of Rb in the absence of Trp, at pH 7, leads to lumichrome (IV) as the main product [ 22 1. Figure 3 shows the temporal changes of the fluorescence spectra obtained at the different pH values in experiments performed in the absence of oxygen (Pig. 3(A)) and in oxygen-saturated solutions (Pig. 3(B)), following excitation with light of 325 nm. This wavelength was selected for excitation since it is not absorbed by Trp and corresponds to a minimum in the Rb absorption spectrum, and thus is suitable for the detection of emission arising from the photoproducts. The spectra obtained at pH 3 and 7 under anaerobic conditions show an increase in the emission intensity of a band at 460 run which may be due to the formation of lumichrome [22, 231. However, at pH 7, the amount of lumichrome formed (as evaluated by the intensity of the fluorescence at these wavelengths) is less than 20% of that obtained when Rb is irradiated alone under similar conditions (data not shown), indicating that most of the Rb is consumed by a different process involving added Trp, in agreement with the conclusion reached from the analysis of the changes in the absorption spectra. The fluorescence spectra measured under anaerobic conditions at pH 10 show the presence of two bands centred at 460 and 510 run. The band at 460 run may be due to hrmichrome, which is one of the main products of Rb photolysis at high pH [23]. However, the total fluorescence observed in this region is larger than that observed in the absence of Trp at total Rb consumption, indicating a contribution of Tr-p-derived photo-
62
A
,H
o
I
3.0
A
570
pH 70
510
>H 10.0
L50
B pH 3.0
L
,I
IO’,
(50
pH 7.0
190
330
,
sm
,
,
,
,
510
El3
390
3M
570
510
I.50
390
330
X,nm
F’ig. 3. Temporal changes in the fluorescence spectra measured after irradiation of Rb (0.035 mM; 452 run) in the presence of Trp (1.91 mM) in aqueous solutions at different pH values (irradiation time between spectra, 15 min): (A) under anaerobic conditions; (B) under aerobic conditions.
products to the recorded emission. Thus, from the absorption and emission spectra, it can be concluded that the presence of Trp leads to the formation of new products which absorb in the 300-400 run region and which present weak fluorescence in the 400-480 nm region. These spectroscopic characteristics most probably correspond to the Rb-Trp adducts. Fluorescence spectra recorded at pH 3 and 7 following irradiation under aerobic conditions (Fig. 3(B)) show an increase in a band at about 430 run which corresponds to the fluorescence of formyl kynurenine [ 241. Concomitant formation of kynurenine is difficult to assess from the fluorescence spectra due to interference by the remaining flavin. At pH 10, fluorescence spectra recorded under aerobic conditions closely resemble those obtained under nitrogen, pointing to the predominance of emission from the flavin-derived products. However, even under these conditions, fluorescence yields in the 400480 run region are larger than those measured in the absence of Trp or under anaerobic conditions, demonstrating the significant contribution of
63
Trp-derived oxidation products to the total fluorescence measured in this region. Trp consumption was measured by following the decrease in the fluorescence intensity at 350 nm after excitation with light of 285 nm. (Measurements were made after dilution of the reaction mixture until the fluorescence intensity was a linear function of tryptophan concentration.) The shape of the fluorescence band observed when the excitation is carried out at 285 run does not change significantly with the time of Rb irradiation, showing that reaction products do not contribute to the fluorescence observed under these conditions. In experiments carried out under anaerobic conditions, no significant Trp decrease is observed even after total Rb consumption. Given the large excess of Trp present, this result is compatible with the formation of Trp-Rb adducts. In contrast, under aerobic conditions, Trp is significantly consumed, showing that its removal must take place via a process which is sensitized by Rb excitation but does not involve a stoichiometric consumption of the sensitizer. Furthermore, the observation that Trp consumption yields are considerably higher than those obtained when other sensitizers are employed [ 13 1, implies that a type I photo-oxidation mechanism is involved [ 151. The rate of the process is dependent on pH: the consumption rates (M mm-‘) are 3.27 X 10e6 at pH 3, 3.26~ lop5 at pH 7 and 6.5 X 1O-5 at pH 10, these correspond to Trp photoconsumption quantum yields of 0.022, 0.22 and 0.43 respectively. The increase in Trp consumption is paralleled by an increase in the rate of oxygen uptake (Fig. 4). The initial rates of oxygen uptake (M mm-‘) are 8 x 10m6 at pH 3, 1.3X 10T4 at pH
MINUTES
Fig.
4.
Effect of pH on the rate of oxygen uptake from oxygen-saturated solutions.
64
7 and 2.2X 10m4 at pH 10. These data show that, irrespective of the pH, nearly three oxygen molecules are consumed by each oxidized Trp molecule. The results show that an increase in pH increases the rate of the process under anaerobic conditions, where the principal process is Trp-Rb adduct formation, and under aerobic conditions, where the main Trp-consuming process is an efficient oxidation which has a quantum yield of nearly unity at pH 10. In order to interpret these results it is necessary to consider the effect of pH on the properties of the Rb excited states and the possible intermediates involved in the process. The Rb excited singlet is quenched by protons and the inter-system crossing quantum efficiency changes from 0.40 at pH 2.2 to 0.7 at pH 7 [22]. Adduct formation will therefore be favoured at higher pH values, irrespective of the Rb excited state involved in the process. Another factor which must be considered is the relative charge of the reaction partners. The ground state pK, value of Rb is extremely low (pK, = 0 [ 251) and that of Trp is 2.8. However, since adduct formation takes place after Rb excitation, the acid-base properties of the excited state, instead of those of the ground state, should be considered. The pK, value of triplet Rb has been reported to be 6.5 [23]. Excitation of Rb at pH 3 therefore gives rise to the protonated triplet which encounters about half of the Trp also in protonated form at this pH. Coulombic repulsion between the reaction partners may inhibit adduct formation in acid solution. Furthermore, it has been shown that the fast interaction of the hnniflavin triplet with indole leads to the formation of the indolyl cation radical and the semiquinone anion radical [ 151. If adduct formation takes place through the interaction of these species, the product yield may be affected by the protonation state of the two species, with p&=4.3 [26] and p&=2.3 [15] respectively. Lumiflavin-sensitized photo-oxygenation of indole in water has been considered to arise mainly from the interaction of the indolyl cation radical with the superoxide anion produced in the oxidation of the semiquinone anion radical by oxygen [ 151. If a similar pathway is assumed for the photooxygenation of Trp in this work, the pH effect observed may also be due to changes in the protonation state of the superoxide anion (pK,= 4.8 (271). However, it is relevant to consider that, given the pK, value of the indolyl cation radical, it is very unlikely that the reaction at pH 10 involves the cation radical and, if superoxide is involved, the reaction should take place between it and the neutral indolyl radical. Micellar effects on the reactions were investigated by carrying out the photolyses in solutions of anionic (SDS) and cationic (CTAR) surfactants. Any study of micellar effects requires an evaluation of the partitioning of the reagents between the micelles and the aqueous pseudophase. Table 1 shows the percentage of incorporation of Rb and Trp into the micellar pseudophases (70 mM concentration of the surfactants) at the three pH values employed. The different capacities of SDS and CTAR micelles for incorporating the reagents, and the effect of pH on the amount incorporated into a given micelle, can easily be rationalized in terms of hydrophobic
65 TABLE
1
Percentage
of Rb and Trp incorporated
mM; temperature,
into micelles in solution (surfactant concentration,
Percentage
Surfactant
Reagent”, b
SDS
Trp Rb
3
>90 70
Trp
7
44 71
10
