Photoinduced formation of reactive oxygen species from the acid form of 6-(hydroxymethyl)pterin in aqueous solution

Article abstract of DOI:10.1002/hlca.200690107

The photochemistry of 6-(hydroxymethyl)pterin (HPT; 1) in aqueous solution (pH 5-6) was investigated by irradiation at 350 nm at room temperature. The photochemical reactions of the acidic form 1a were followed by UV/VIS spectrophotometry, thin-layer chromatography (TLC), high-performance liquid chromatography (HPLC), and enzymatic methods for the determination of the superoxide anion radical (O₂^.-) and hydrogen peroxide (H₂O₂). When 1a is exposed to UV-A radiation, the intermediates 4 and 4′ are formed reacting with O₂ to yield 6-formylpterin (FPT; 5) and 6-carboxypterin (CPT; 6). under formation of O ₂^.- and H₂O₂ (Scheme 3). The quantum yields of the disappearance of HPT (1a) and of the formation of the photoproducts 5 and 6 were determined. HPT was investigated for its efficiency in singlet-oxygen (¹O₂) production in acidic aqueous solution. The corresponding quantum yield of ¹O₂ production (Φ_Δ) was 0.15±0.02, as measured by the ¹O₂ luminescence in the near-IR (1270 nm) upon continuous excitation of the sensitizer. However, ¹O₂ does not participate in the actual photooxidation of HPT (1a) to FPT (5) and CPT (6).

Full text of DOI:10.1002/hlca.200690107

1090
Helvetica Chimica Acta – Vol. 89 (2006)
Photoinduced Formation of Reactive Oxygen Species from the Acid Form of
6-(Hydroxymethyl)pterin in Aqueous Solution
by Andrés H. Thomasa), Romina Cabrerizoa), Mariana Vignonia), Rosa Erra-Balsellsb), Franco M.
Cabrerizo*b), and Alberto L. Capparelli*a)
a) Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas (INIFTA), Departamento de Quí-
mica, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, CONICET, Casilla de Correo 16,
Sucursal 4, 1900-La Plata, Argentina (e-mail: alcappa@dalton.quimica.unlp.edu.ar)
b) CHIDECAR – CONICET, Departamento de Química Orgánica, Facultad de Ciencias Exactas y
Naturales, Universidad de Buenos Aires, Pabellón 2, 3p, Ciudad Universitaria, 1428-Buenos Aires,
Argentina (e-mail: fcabrerizo@qo.fcen.uba.ar)
The photochemistry of 6-(hydroxymethyl)pterin (HPT; 1) in aqueous solution (pH 5–6) was inves-
tigated by irradiation at 350 nm at room temperature. The photochemical reactions of the acidic form 1a
were followed by UV/VIS spectrophotometry, thin-layer chromatography (TLC), high-performance liq-
uid chromatography (HPLC), and enzymatic methods for the determination of the superoxide anion rad-
ical (Oꢀ2ꢁ) and hydrogen peroxide (H2O2). When 1a is exposed to UV-A radiation, the intermediates 4
and 4’ are formed reacting with O2 to yield 6-formylpterin (FPT; 5) and 6-carboxypterin (CPT; 6)
under formation of Oꢀ2ꢁ and H2O2 (Scheme 3). The quantum yields of the disappearance of HPT (1a)
and of the formation of the photoproducts 5 and 6 were determined. HPT was investigated for its effi-
ciency in singlet-oxygen (1O2) production in acidic aqueous solution. The corresponding quantum
yield of 1O2 production (FD) was 0.15ꢂ0.02, as measured by the 1O2 luminescence in the near-IR
(1270 nm) upon continuous excitation of the sensitizer. However, 1O2 does not participate in the actual
photooxidation of HPT (1a) to FPT (5) and CPT (6).
Introduction. – Pterins, i.e., 2-amino-4-hydroxypteridine derivatives, occur in a wide
range of biological systems [1][2]. Although pterins are present in only minute amounts
in living systems, they are known to participate in important biological functions [1–3]
including various photobiological processes such as photoreception in plants [4][5]. For
example, the folic acid derivative ‘5,10-methenyltetrahydrofolate’ is present as the
light-harvesting antenna in DNA photolyases [6][7], which, in turn, are involved in
DNA repair after UV irradiation. Pterins are also interesting as photosensitizers in
photochemical processes taking place in organisms exposed to UV-A radiation, and
are known to participate as sensitizers in photochemical reactions that induce DNA
damage [8–10].
Many pterin derivatives are good singlet-oxygen (1O2) sensitizers in aqueous solu-
tions [11–13]. Singlet oxygen, the lowest electronic excited state of molecular oxygen,
is an important reactive oxygen species (ROS) and one of the main species responsible
for the damaging effects of light on biological systems (photodynamic effects) [14][15].
The 1O2 excited state is much more reactive than the triplet ground state (3O2). Thereby,
photosensitization is primarily responsible for the production of 1O2 in vivo [16].
© 2006 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 89 (2006)
1091
Biomedical studies have suggested that pterins are involved in vitiligo, a depigmen-
tation disorder [17][18]. Moreover, different pterin derivatives such as 6- and 7-bio-
pterin are accumulated in the epidermis of affected patients, which leads to a typical
fluorescence under Wood’s light (351 nm) [17]. Hydrogen peroxide (H2O2), another
ROS, participates in the pathogenesis of vitiligo, and is also accumulated in the epider-
mis of affected patients [19].
Pterin proper and its 6-substituted congeners, the most-common derivatives found
in nature, behave as weak acids in aqueous solutions. As reported by Albert [20], for
several pterin derivatives, the dominant equilibrium at pH>5 involves an amide
(acid form) and a phenolate (basic form), as shown in Scheme 1 for the acid/base equi-
librium of 6-(hydroxymethyl)pterin (HPT; 1). The pKa of the acid form 1a and of sim-
ilarly substituted analogs is ca. 8. Note that, in aqueous solution, the protonated 2-
amino group or the ring N-atoms have pKa values below 2 [20].
Scheme 1
The mechanism involved in the photooxidation of 6-substituted pterins strongly
depends on the nature of the substituent at C(6), as well as on the pH. In other
words, the photochemistry of the acid and base forms of pterin derivatives are different.
Some pterins exhibit a very characteristic photochemistry. For instance, when solutions
of biopterin (2) and neopterin (3), previously purged with Ar gas, are exposed to UV-A
radiation, red-colored compounds are generated (Scheme 2) [21][22]. It has been pro-
posed that these compounds, with low-intensity long-wavelength absorption bands at
ca. 480 nm, correspond to 6-acyl or 6-formyl-5,8-dihydropterins such as 4 [21][22].
This would imply reduction of the pterin moiety and oxidation of the substituent via
an intramolecular redox reaction. Compounds of type 4 are then easily oxidized
upon exposure to air or O2 to yield the corresponding oxidized pterins, e.g., 6-formylp-
terin (FPT; 5), together with H2O2 (Scheme 2).
Scheme 2. Photooxidation of Biopterin and Neopterin under UV-A Radiation in Aqueous Solution
[21][22]

1092
Helvetica Chimica Acta – Vol. 89 (2006)
We have recently reported [23] that the photochemical behavior of the basic form
1b of HPT in aqueous solution is similar to that shown in Scheme 2. The photooxidation
of HPT to FPT (5) occurs via at least two steps. In the first step, a photon is absorbed in
the absence of O2, which generates a red-colored intermediate, probably 6-formyl-5,8-
dihydropterin. In the second step, this colored intermediate reacts with molecular O2 to
yield FPT and H2O2.
As part of our studies of the photochemistry of pterins, we herein report in detail
the photochemistry of the acid form 1a of HPT under UV-A irradiation (350 nm).
The results are summarized in Scheme 3. We have identified the resulting photoprod-
ucts and determined the corresponding quantum yields. The role of molecular O2 on
the mechanisms of these photochemical reactions was analyzed, and the capability of
HPT to yield ROS is discussed. In particular, the production of 1O2, O2ꢀꢁ, and H2O2 dur-
ing the photooxidation of HPTwas studied, and the results compared with those descri-
bed earlier for the basic form 1b and similar 6-substituted pterins.
Scheme 3. Proposed Reaction Mechanism for the Photooxidation of 6-(Hydroxymethyl)pterin (1a) in
Acidic Aqueous Solution
Experimental. – General. 6-(Hydroxymethyl)pterin (HPT; 1), 6-formylpterin (FPT; 5), 6-carboxyp-
terin (CPT; 6), and other pterins were purchased from Schircks Laboratories (Switzerland), and used
without further purification. Cytochrome c (Cyt) from horse heart, and superoxide dismutase (SOD)
from bovine erythrocytes were obtained from Sigma Chemical Co. The pH of aq. solns. was adjusted

Helvetica Chimica Acta – Vol. 89 (2006)
1093
by adding drops of 0.1–0.2M aq. NaOH or HCl solns. with a micropipette. The ionic strength was ca. 10ꢁ3
M in all experiments. In experiments with singlet oxygen (1O2), D2
O (>99.9%; Euriso-top), DCl (99.5%;
Aldrich) in D2O, and NaOD (CEA) in D2O were employed.
ACHTREUNG
A
ACHTREUNG
Photolysis. Continuous photolysis of HPT (1) solns. was performed at pH 5.5 in 1-cm quartz cells at
r.t. with a Rayonet RPR lamp emitting at 350 nm (Southern N.E. Ultraviolet Co.). The experiments were
performed in the presence and absence of air. O2-Free solns. were obtained by bubbling with Ar gas for
20 min.
UV/VIS Analysis. Electronic spectra were recorded on a Varian Cary-3 spectrophotometer. Meas-
urements were made in quartz cells of 1 cm optical-path length and double cells of 1 cm (for irradiation)
and 0.2 cm (for absorbance) length, resp. The absorption spectra of the solns. were recorded at regular
intervals of irradiation time, and the signals were averaged and smoothed with the Varian software.
Experimental-difference (ED) spectra were obtained by subtracting the spectrum at time t=0 from
the subsequent spectra recorded at different times t. Each ED spectrum was normalized rel. to the max-
imum abs. value of the absorbance difference, yielding the normalized experimental-difference (NED)
spectrum. Reference-difference (RD) spectra and normalized-reference-difference (NRD) spectra
were obtained from aq. solns. of commercial standards. The comparison between NED and NRD spectra
allows the characterization of the major photolysis products [24][25].
Thin-Layer Chromatography (TLC). TLC Experiments were performed on DEAE cellulose plates
(100 mm) eluting with 0.3% aq. NH4Cl soln. Irradiated HPT (1) solns. (1.5×10ꢁ4 M) and aq. solns. of
pterin-based standards were developed by exposure to UV/VIS radiation (350 nm), which gave rise to
blue fluorescent spots. The corresponding Rf values obtained for pterin proper and several pterin deriv-
atives have been reported before [24].
High-Performance Liquid Chromatography (HPLC). A System Gold HPLC setup (Beckman Instru-
ments) was used to monitor and quantify the photolysis reaction and photoproducts. A Pinnacle-II C18
column (250×4.6 mm, 5 mm; Restek) was used for product separation, eluting with aq. 20 mM potassium
phosphate buffer (pH 5.5) containing 2.5 mM EDTA. HPLC Runs were monitored by UV/VIS at 340 nm.
Determination of Quantum Yields. The quantum yields of disappearing HPT (1a) and of formed FPT
(5) and CPT (6) were determined in experiments performed under different conditions. Aberchrome 540
(Aberchromics Ltd.) was used as actinometer for the measurements of the incident photon flux P0 at the
excitation wavelength (350 nm). The method for the determination of P0 has been described in detail
elsewhere [26]. Values of the photon flux absorbed by 1a, i.e., Pa, were calculated from P0 according
to the Lambert–Beer law:
Pa =P0 (1ꢁ10ꢁA
)
(1)
where A is the absorbance of 1a at the excitation wavelength. The evolution of the concentrations of reac-
tant and photoproducts during irradiation was followed by HPLC. Aq. solns. of commercial standards
were employed for obtaining the corresponding calibration curves.
Determination of the Concentration of O2 and H2O2. The O2 consumption during photolysis was
measured with an O2-selective electrode (Orion 37-08-99). The exper. setup for these measurements
was described elsewhere [27]. For the determination of H2O2, a Cholesterol Kit (Wiener Laboratorios
S.A.I.C.) was used. H2O2 was quantified after reaction with 4-aminophenazone and phenol [28][29].
Briefly, 400 ml of irradiated HPT (1a) soln. was added to 2 ml of reagent. The absorbance at 505 nm of
the resulting mixture was measured after 30 min at r.t., using the reagent as blank. Aq. H2O2 solns. pre-
pared from commercial standards were employed for obtaining the corresponding calibration curves.
Superoxide-Anion Assay. Photogenerated superoxide anion radicals (Oꢀ2ꢁ) were detected with the
aid of the SOD-inhibitable Cyt reduction [30]. Air-equilibrated HTP (1a) solns. were irradiated in the
presence of 14 mM Cyt. The extent of Cyt reduction was determined by monitoring the increase in the
UV/VIS absorbance at 550 nm. The absorption spectra of oxidized and reduced Cyt are very character-
istic, and this spectral change is a sensitive assay for electron-transfer reactions. This assay has been used
to detect not only Oꢀ2ꢁ in aerated soln., but also electron transfer from excited sensitizers under anaerobic
conditions [31].

1094
Helvetica Chimica Acta – Vol. 89 (2006)
1
Determination of Quantum Yields of O2 Production. The quantum yield of 1O2 production (FD) was
determined by direct analysis of the 1O2 near-IR luminescence at 1270 nm [32–34] occurring during con-
tinuous irradiation (lex =367 nm) of the aq. soln. of HPT (1a). The main features of the method and the
equipment have been described in detail before [35–37]. Briefly, a sample soln. in a quartz cuvette is irra-
diated with a Xe/Hg arc through a H2O filter, focusing optics, and a monochromator. The 1O2 lumines-
cence is collected with a mirror, chopped, and, after passing through a focusing lens, a cut-off filter (1000
nm), and an interference filter (1271 nm), detected at 908 (with respect to the incident beam) using a
cooled NIR photomultiplier (Hamamatsu). For determining FD, the luminescence signals Se of solns.
of a sensitizer (SSe ) and a reference sensitizer (SeR) were measured. The ratio SeS/SRe is given by the follow-
ing equation:
SeS
SeR
PS0 FD
PR0 FRD
kd
PS0 FaDpp
PR0 FRD
¼
¼
(2)
kd þ kS ½SensŠ
t
where PS0 /PR0 is the ratio of the incident photon flux at the excitation wavelength of the sensitizer (S)
investigated, and of the reference (R). When kd, the non-radiative deactivation rate constant, and kSt ,
1
the rate constant of O2 total (physical and chemical) quenching by the sensitizer itself, are known for
a given solvent, FD may be calculated from Eqn. 2 by measuring the luminescence signals of the sensitizer
and the reference solns., as well as the corresponding incident photon flux. Further, when kSt is unknown,
app
1
1
an apparent quantum yield of O2 production (FD ), not corrected for potential O2 quenching by the
1
sensitizer itself, is obtained. Because of the short lifetime of O2 in H2O (tD =1/kd =3.8 ms), D2
used as the solvent in all experiments (tD =62 ms) [38][39]. Measurements with HPT (1a)
(e345 =6.3×103 ꢁ1 cmꢁ1) were carried out at an excitation wavelength of 367 nm in D2
O, where the ref-
erence sensitizer phenalen-1(1H)-one has a FRD value of 0.97 [37][40].
Continuous Photolysis in H2O vs. D2O. Solns. of HPT (1a) were prepared at the same concentration
in both H2O and D2O, and irradiated under identical conditions. Solvent effects were then evaluated by
A
M
ACHTREUNG
AHCTREUNG
comparing the results of the UV/VIS spectrophotometric analyses.
Calculation of Molecular Properties. The ground-state geometry of the acidic form of HPT (1a), 6-
formyl-5,8-dihydropterin (4) and its enol tautomer 4’ (see Scheme 3) were calculated with the semiempir-
ical PM3 method, as implemented in HyperChem, vers. 7.01, for Windows [41]. The following parameters
were used: restricted Hartree–Fock (RHF) formalism; total charge, 0; spin multiplicity, 1; convergence
limit, 1e-008; iteration limit, 50; accelerate convergence, on; CI method, none; Polak–Ribiere algorithm;
RMS gradient, 0.006 kcal Åꢁ1 molꢁ1; in vacuo. PM3 has proven to be more-effective in studies on mol-
ecules containing hetero atoms compared to other methods such as MINDO/3 or MNDO. Heats of for-
mation (DHf), charge densities, and HOMO and LUMO energies were also calculated with the PM3
method (after transferring the optimized geometries): restricted RHF formalism; total charge, 0; spin
multiplicity, 1; convergence limit, 1e-008; iteration limit, 50; accelerate convergence, on; CI method, sin-
1
gle excited; orbital criterion: 6 :6. The UV/VIS transition 1S0 ! S1 (lmax), and the corresponding oscil-
lator strength (f) were calculated after geometry optimization (PM3) with the ZINDO/S (CI, 6 :6)
method.
Results and Discussion. – 1. Photolysis in the Presence of O2. The spectrum of air-
equilibrated solutions of the acidic form of HPT (1a) changed significantly with irradi-
ation time, as shown in Fig. 1,a. An isosbestic point at 279 nm was observed for solu-
tions irradiated for at least 40 min. No further changes were detected for irradiated sol-
utions stored in the dark. NRD Spectra (see Experimental) were obtained by subtract-
ing the spectrum of a standard solution of 1a from those of standard solutions contain-
ing FPT (5) and CPT (6) in different ratios. As shown in Fig. 1,b, the NED spectra were
similar to the NRD spectra obtained for a solution containing 15% of 5 and 85% of 6.
This suggests that, under these experimental conditions, CPT (6) is the main photo-
product, FPT (5) being the side product.

Helvetica Chimica Acta – Vol. 89 (2006)
1095
Fig. 1. UV/VIS Spectroscopic behavior of 1a upon irradiation. a) Evolution of the absorption spectra
of irradiated aqueous, aerobic solutions of 1a (180 mM; pH 5.5) as a function of irradiation time. The
three spectra were recorded in intervals of 20 min; arrows indicate the direction of the observed
changes at different wavelengths. b) NED Spectrum (—) obtained by subtracting the initial UV/VIS
spectrum of 1a from the corresponding spectrum recorded after 20 min of photolysis, and NRD spec-
trum (- - -) obtained from standard solutions of 1a, 5, and 6. For details, see Experimental.
TLC Analysis showed the presence of three fluorescent substances in irradiated
HPT (1a) solutions at t=0, 10, 30, and 55 min. The Rf values were in agreement with
those for HPT, FPT and CPT. The presence of FPT and CPT in irradiated solutions
of HPTwas further confirmed by HPLC experiments. The concentrations of these com-
pounds were determined as a function of irradiation time t. Experiments performed at
pH 5.5 with a solution of 1a at an initial concentration of 200 mM showed that, within the
first 15 min of irradiation, the evolution of the concentrations of HPT, FPT and CPT
followed a zero-order rate law (Fig. 2). Under these conditions, the initial flux of pho-
tons absorbed by 1a (Pa) was (8.2ꢂ0.4)×10ꢁ4 einstein lꢁ1 minꢁ1, which corresponds to
ca. 95% of the incident UV-A radiation. The rate of HPT (1a) disappearance was

1096
Helvetica Chimica Acta – Vol. 89 (2006)
~
&
Fig. 2. HPLC-Based change in the solution concentrations of 1a (l), 5 ( ), and 6 ( ) upon irradiation
of 1a (200 mM) at pH 5.5 under aerobic conditions as a function of irradiation time. The mathematical
sum of the different experimental concentrations (^) is also shown.
1.9ꢂ0.1 mM minꢁ1, and the rates of FPT (5) and CPT (6) formation were 0.4ꢂ0.1 and
1.5ꢂ0.1 mM minꢁ1, respectively, the sum of the latter two values being equal to the rate
of HPT disappearance. This indicates that FPT and CPT were the main photoproducts,
with no other substances being formed in significant amounts. The rates of formation of
5 and 6 were 21 and 79%, respectively, of the rate of the disappearance if 1a. Therefore,
this result was in agreement with the UV/VIS analysis (see above). The quantum yield
associated with the disappearance of HPT (1a) was calculated as (2.3ꢂ0.2)×10ꢁ3; the
values for FPT and CPT formation were (0.5ꢂ0.1)×10ꢁ3 and (1.8ꢂ0.3)×10ꢁ3, respec-
tively.
The above results are surprising, because the only product in the photooxidation of
the basic form 1b of HPT is FPT (5) [23]. In contrast, the photooxidation of the acid
form 1a yields both FPT and CPT (6) (see Scheme 3 above). At a first glance, this
may lead to the conclusion that CPT is formed by photooxidation of FPT. However,
we were able to discard this hypothesis because a high percentage of CPTwas observed
already during the first few minutes of irradiation, when the absorption of HPT at 350
nm is so high that only a small proportion of the incident radiation can be absorbed by
FPT. Moreover, experiments performed at higher initial concentrations of HPT showed
similar results.
The evolution of O2 concentration during irradiation was monitored with an O2
electrode in a closed cell (see Experimental). A significant decrease in [O2] with irradi-

Helvetica Chimica Acta – Vol. 89 (2006)
1097
ation time was observed. This result confirms that O2 present in solution was consumed
during photolysis.
H2O2 was detected in irradiated solutions of HPT (1a), indicating that photooxida-
tion generates this ROS. Its formation was monitored as a function of irradiation time
of an aqueous (pH 5.5) 150 mM soln. of 1a (Fig. 3). The estimated quantum yield of H2O2
formation was (2.2ꢂ0.2)×10ꢁ3. Comparison of the quantum yield of H2O2 formation
and that of HPT consumption suggested that, for each molecule of HPT consumed,
one molecule of H2O2 is generated.
Fig. 3. Evolution of the H2O2 concentration in an irradiated solution of 1a (150 mM) at pH 5.5 as a
function of irradiation time
The generation of H2O2 during the photooxidation of other pterin derivatives (bio-
pterin, sepiapterin, FPT, 6-methylpterin, pterin) was previously reported [13][18][42].
The basic form of HPT (1a) also produces H2O2 under UV-A irradiation [43] in the
presence of O2. Generation of H2O2 during the photolysis of pterins is important
from a biomedical point of view, and is particularly relevant for skin diseases [19].
2. Photolysis in the Absence of O2. Anaerobic photolysis of 1a led to the formation
of at least one compound with a long-wavelength band and a low molar-absorption
coefficient at 480 nm. As soon as air (O2) was let into the cell, a thermal reaction in
the dark was observed, and the mentioned UV/VIS absorption band disappeared. Tak-
ing into account the studies carried out by Pfleiderer and co-workers on the photochem-
istry of biopterin and neopterin [21][22][44], and considering the spectroscopic fea-
tures of related compounds like 5,8-dihydrolumazines [45] and 1,4-dihydropyrazines
[46], our results suggest that, as in the case of anaerobic irradiation of the basic form
1b of HPT [23], an intermediate such as 6-formyl-5,8-dihydropterin is formed. How-
ever, the spectroscopic changes at wavelengths above 420 nm were lower than those

1098
Helvetica Chimica Acta – Vol. 89 (2006)
observed during the irradiation of 1b. Therefore, the amount of the intermediate accu-
mulated in the irradiated acidic solutions seems to be very low.
Fast spectroscopic changes were observed when solutions of 1a, irradiated in the
absence of O2, were aerated. The NED spectrum, obtained by subtracting the spectrum
of a non-irradiated solution of 1a from that of an irradiated and then immediately aer-
ated solution, matched with the NRD spectrum obtained in the same way as that
obtained for the photolysis in the presence of O2 (e.g., for a solution containing 15%
of FPT and 85% of CPT). This observation suggests that, under these conditions,
FPT and CPT are final, stable compounds generated in the same proportion as under
aerobic conditions.
TLC Analysis of the irradiated HPT (1a) solutions under anaerobic conditions, fol-
lowed by aeration, showed three fluorescent spots, with Rf values identical to those for
HPT, FPT and CPT. The intensity of the spot of HPT decreased with irradiation time.
No other product could be detected by this procedure.
The results obtained by HPLC analysis agreed with those obtained by TLC and
UV/VIS analyses. FPT (5) and CPT (6) were the only photoproducts found in the
absence of O2, followed by immediate aeration (Fig. 4). During the first 60 min of irra-
diation, the change in concentration of HPT, FPT and CPT followed a zero-order rate
law (Fig. 4). Under these conditions, Pa was determined as (1.1ꢂ0.1)×10ꢁ4 einstein lꢁ1
minꢁ1). The quantum yield associated with the disappearance of HPT (1a) was
(2.0ꢂ0.3)×10ꢁ3, and those due to the formation of FPT (5) and CPT (6) were
(0.2ꢂ1)×10ꢁ3 and (1.8ꢂ0.3)×10ꢁ3, respectively. These values are quite similar to
the ones obtained in experiments performed under aerobic conditions, which suggests
that the underlying mechanisms are the same.
The determination of H2O2 in solutions of 180 mM HPT at pH 5.5, immediately aer-
ated after different times of irradiation under anaerobic conditions, revealed that this
ROS is also generated. The obtained value for the quantum yield of H2O2 production
was (2.5ꢂ0.2)×10ꢁ3 for the first 50 min of irradiation. This value is similar, within
experimental error, to the quantum yield associated with the disappearance of HPT
under similar conditions ((2.0ꢂ0.3)×10ꢁ3), and to the quantum yield of H2O2 produc-
tion in experiments performed under aerobic conditions ((2.2ꢂ0.2)×10ꢁ3).
3. The Role of the Superoxide Anion. When air-equilibrated solutions containing
100 mM of HPT (1a) and 14 mM of cytochrome c (Cyt) at pH 6.7 were irradiated, an
increase in the UV/VIS absorbance at 550 nm was observed. The difference in absorb-
ance, DA550, steadily increased with irradiation time, reaching a plateau after ca. 6 min
(Fig. 5). The corresponding ED spectra in the range 500–600 nm were in good agree-
ment with those reported in the literature for the reduction of Cyt [47]. Photoreduction
of Cyt was not observed in similar experiments performed in the presence of superox-
ide dismutase (SOD; 200 U/ml), as can be seen from Fig. 5. These results show that,
under the present conditions, Oꢀ2ꢁ, generated photochemically by HPT, is the actual
species that transfers an electron to Cyt.
When the same assay was carried out under anaerobic conditions, photoreduction
of Cyt was also observed, even in the presence of SOD (200 U/ml; Fig. 5). Moreover,
the extent of Cyt reduction was higher than in the presence of air, i.e., the plateau
was reached at a higher DA550 value. Thus, the reduction of Cyt, photosensitized by
HPT (1a), can occur in the absence of O2 by means of a direct electron transfer. The

Helvetica Chimica Acta – Vol. 89 (2006)
1099
Fig. 4. Evolution of the HPLC-based concentrations of HTP (1a), FTP (5), and CPT (6). The initial
concentration of 1a was 180 mM. The aq. solutions (pH 5.5) were irradiated in the absence of O2, and
aerated in the dark immediately after irradiation. The mathematical sum (^) of the different experi-
mental concentrations is also shown.
photoreduction of Cyt in the absence of O2 has been observed before for other hetero-
cyclic compounds in aqueous solution [31].
Control experiments were performed to investigate thermal or photochemical pro-
cesses that might interfere with the assay. No reduction of Cyt was observed in solutions
containing 100 mM of 1a and 14 mM of Cyt at pH 6.7, when kept in the dark for more than
2 h. When solutions containing only Cyt were exposed to UV-A radiation, no changes
were observed in the spectra.
Taking into account the studies reported in the literature on the photooxidation of
different pterin derivatives (see Introduction and Scheme 2), in combination with the
results presented in the previous sections, it can be concluded that the intermediate(s)
generated after irradiation of HPT (1a) is (are) able to transfer one electron either to
O2 to yield Oꢀ2ꢁ, or directly to Cyt. In air-equilibrated solutions, where there is a large
excess of O2 over Cyt, the first process takes place. However, in the absence of O2, the
latter occurs. Thereby, the observed H2O2 is produced from Oꢀ2ꢁ by the following reac-
tions:
H3
A
(3)
(4)
H
ACHTREUNG

1100
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 5. Photoreduction of cytochrome c (Cyt; 14 mM) sensitized with 1a (100 mM) in aqueous solution
(pH 6.7). The experiment was performed in the presence and absence of O2, and in the presence and
absence of superoxide dismutase (SOD).
When taking into account the mechanism proposed in Scheme 3 and the fact that
Oꢀ2ꢁ can abstract H+ from very weak acids, then one can conclude that the H-atoms
in H2O2 may come either from H2O (Eqns. 3 and 4) or from the pterinic intermediates.
4. Singlet-Oxygen Studies. The value of the apparent quantum yield of 1O2 produc-
tion, FaDpp, by the acid form 1a of HPT was determined in O2-saturated and air-equili-
brated D2
O solutions by monitoring the near-IR 1O2 luminescence (see Experimental).
G
Most pterins give rise to a relatively high fluorescence quantum yield FF [49]. There-
fore, control experiments in Ar-saturated solutions were carried out to check possible
tailing of the fluorescence emission of 1a in the near-IR. However, no luminescence was
detected at 1270 nm under the experimental conditions.
Experiments for determining FDapp were performed at pD 5.5. Now, a significant 1O2
emission was observed, and a value of 0.15ꢂ0.02 was determined for air-equilibrated
1
solutions. Taking into aꢁc1count the rate constant of O2 total quenching by the basic
form 1b (kt =3.1×106
M
sꢁ1) [23], and assuming that this value does not drastically
change for the acidic form 1a, FaDpp may be considered as FD.
It is worth mentioning that the value of FaDpp for the acid form 1a did not increase
when the solutions were saturated with O2. Thus, the excited triplet state was com-
pletely quenched by O2 in the air-equilibrated solutions. The value of Fapp obtained
D
for the acid form 1a of HPT (0.15ꢂ0.02) is a little lower than that previously found
for the corresponding basic form 1b (FD =0.21ꢂ0.01) [23], an observation made before
for other pterin derivatives [13].

Helvetica Chimica Acta – Vol. 89 (2006)
1101
1
Taking into consideration that the acidic form 1a of HPT is able to generate O2
under UV-A irradiation, it is necessary to check if this ROS is involved in the mecha-
nism of the photoreaction of HPT. Thus, photolysis experiments were performed in
D2
N
N
were irradiated, and UV/VIS spectra were recorded as a function of irradiation time
under both experimental conditions. Spectroscopic analysis showed that the photooxi-
dation of HPT (1a) to FPT (5) and CPT (6) is not accelerated in D2O. This result indi-
U
cates that 1O2 does not actively participate in the main mechanism of photooxidation of
HPT, in agreement with the results presented above.
5. Mechanistic Considerations. Taking into account the results discussed in the pre-
vious sections, we propose the reaction mechanism shown in Scheme 3 (see above). We
suggest that irradiation of the acid form 1a of HPT gives rise to 6-formyl-5,8-dihydrop-
terin (4) by means of a process similar to that already reported for the basic form 1b
[23]. The same holds for other pterin derivatives [22], where 6-acyl-5,8-dihydropterin
intermediates were formed. However, under our specific experimental conditions, 4
might also yield another intermediate through tautomerization (keto/enol equili-
brium). This alternative intermediate, the enol tautomer 4’ in Scheme 3, could then
react with O2 to yield CPT (6).
6. Theoretical Considerations. To compare the stability of HPT (1a) and the photo-
products 4 and 4’, their ground-state geometries, heats of formation (DHf), and other
properties were calculated by means of the semiempirical parametrized PM3 method
(Table 1). The loss of the aromatic character of the pterin moiety on going from 1a
to 4 and 4’ agrees with the different calculated DHf values and the experimental obser-
vation that both 4 and 4’ are stable only under inert atmosphere, both being formed
from the electronic excited state of 1a.
Table 1. Calculated Heats of Formation (DHf), HOMO and LUMO Energies (E), Maximum-Absorption
Wavelengths (lmax), and Oscillator Strengths (f) of Selected Compounds
DHf
EHOMO
[eV]
ELUMO
[eV]
DE (LUMOꢁHOMO)
lmax
[nm]a)
f
[kcal molꢁ1
]
[eV]
1a
4
4’
ꢁ31.43
ꢁ30.34
ꢁ20.82
ꢁ9.29
ꢁ7.76
ꢁ7.84
ꢁ1.24
ꢁ0.85
ꢁ0.95
8.05
6.91
6.89
308
366
393
0.632
0.011
0.226
a) Calculated lmax values for 5 and 6: 327 and 317 nm, resp.
The HOMO and LUMO energies were calculated to compare the relative reactiv-
ities of these three compounds towards O2. As shown in Table 1, the energy of the
HOMO orbital, EHOMO of 4 and 4’ are higher than that of HPT (1a). Since the
HOMO energy is related to nucleophilicity, this finding indicates that the two isomers
4 and 4’ are more nucleophilic than the starting material 1a.
Next, the charge density at each C-atom was evaluated (Table 2). We found consid-
erably higher charge densities for the C(6)=CH2 moiety in 4 and 4’ compared to 1a,
indicating a higher nucleophilic character of this double bond and, thus, a higher reac-
tivity towards O2. Therefore, the C(a)ꢁOOH (hydroperoxide) intermediate can be
easily formed yielding CPT (6) as the final product.

1102
Helvetica Chimica Acta – Vol. 89 (2006)
Table 2. PM3-Calculated Charge Densities for Atoms C(6) and C(a)
Position
1a
4
4’
5
6
C(6)
C(a)
ꢁ0.182
ꢁ0.271
ꢁ0.229
ꢁ0.035
ꢁ0.255
ꢁ0.174
0.133
0.335
0.334
0.445
Finally, the UV/VIS absorption spectra of 1a, 4, and 4’ were simulated with the
ZINDO/S method. A clear bathochromic shift of lmax for 4 and 4’ was predicted, in
agreement with the bathochromic shift experimentally observed during irradiation of
1a under inert atmosphere.
Conclusions. – Photooxidation of the acid form 1a of 6-(hydroxymethyl)pterin
(HPT; Scheme 1) shows a different pattern to that observed in the photooxidation of
HPT in alkaline medium (basic form 1b) [23]. The quantum yields associated with
the disappearance of 1a, determined in the presence and in the absence of O2, are sim-
ilar (2.3×10ꢁ3), but much lower than that determined for 1b (1.8×10ꢁ2) [23]. The pho-
toreaction occurs through at least two steps. In the first, after absorption of UV-A radi-
ation and without participation of O2, two intermediates (4 and 4’) are generated
(Scheme 3). In the second step, these intermediates react with O2 to yield FPT (5)
and CPT (6), the proportion of the latter being much higher than that of the former.
During photooxidation of HPT (1a), the superoxide anion (Oꢀ2ꢁ) is generated, prob-
ably as a consequence of electron transfer from these intermediates to O2 present in
solution. This result is very important from a biological point of view, because Oꢀ2ꢁ,
along with other reactive oxygen species (ROS), is implicated in the etiology of
many pathological conditions. To the best of our knowledge, the photoinduced gener-
ation of Oꢀ2ꢁ by oxidized pterins has not been reported before. One molecule of H2O2 is
generated for each molecule of HPT consumed, H2O2 probably being a product of the
thermal dismutation of Oꢀ2ꢁ.
The acidic form 1a of HPT is a singlet-oxygen (1O2) sensitizer under UV-A irradi-
ation in aqueous solution (FaDpp =0.15ꢂ0.02). This result is in good agreement with pre-
vious studies reported on other pterin derivatives [12]. However, experiments in D2
indicate that O2 does not participate in the main pathway of the photooxidation of
A
1
HPT to FPT (5) and CPT (6).
The present work was partially supported by Consejo Nacional de Investigaciones Cientificas y Técni-
cas (CONICET-PIP 02470/00), Agencia de Promoción Científica y Tecnológica (ANPCyT Grants No.
PICT 06-12610/02 and 06-12312), Universidad Nacional de La Plata, and Universidad de Buenos
Aires (X022). R. E-B. and A. L. C. thank the Fundación Antorchas (Grant No. 14116-207/03); F. M. C.
also thanks the Fundación Antorchas (grant No. 4248-70). A. H. T. thanks the Deutscher Akademischer
Austauschdienst (DAAD) for a research fellowship. F. M. C. wishes to thank CONICET for a research
post-doctoral fellowship. We deeply acknowledge Prof. A. M. Braun and Prof. E. Oliveros for offering
their facilities to A. H. T. and F. M. C. during their stays at the Engler-Bunte Institute, University of
Karlsruhe, Germany. We also thank Dr. Carolina Lorente and Gabriela Petroselli for help in singlet-oxy-
gen and superoxide-detection experiments, resp. R. E-B., A. L. C., and A. H. T. are research members of
CONICET.

Helvetica Chimica Acta – Vol. 89 (2006)
1103
REFERENCES
[1] W. Pfleiderer, in ‘Chemistry and Biology of Pteridines and Folates’, Ed. J. E. Ayling, M. G. Nair,
C. M. Baugh, Plenum Press, New Nork, 1993, p. 1.
[2] C. A. Nichol, G. K. Smith, D. S. Duch, Annu. Rev. Biochem. 1985, 54, 729.
[3] R. L. Blakley, ‘The Biochemistry of Folic Acid and Related Pteridines’, North-Holland Publishing
Co., Amsterdam, 1969.
[4] R. C. Fuller, G. W. Kidder, N. A. Nugent, V. C. Dewey, N. Rigopoulos, Photochem. Photobiol. 1971,
14, 359.
[5] P. Galland, H. Senger, Photochem. Photobiol. 1988, 48, 811.
[6] P. Heelis, S. T. Kim, T. Okamura, A. Sancar, J. Photochem. Photobiol., B 1993, 17, 219.
[7] J. E. Hearst, Science 1995, 268, 1858.
[8] K. Ito, S. Kawanishi, Biochemistry 1997, 36, 1774.
[9] C. Lorente, A. H. Thomas, L. S. Villata, D. Hozbor, A. Lagares, A. L. Capparelli, Pteridines 2000,
11, 100.
[10] K. Hirakawa, H. Suzuki, S. Oikawa, S. Kawanishi, Arch. Biochem. Biophys. 2003, 410, 261.
[11] K. V. Neverov, E. A. Mironov, T. A. Lyudnikova, A. A. Krasnovsky, M. S. Kritsky, Biokhim. 1996,
61, 1627.
[12] A. H. Thomas, C. Lorente, A. L. Capparelli, C. G. Martínez, A. M. Braun, E. Oliveros, Photochem.
Photobiol Sci. 2003, 2, 245.
[13] F. M. Cabrerizo, C. Lorente, M. Vignoni, R. Cabrerizo, A. H. Thomas, A. L. Capparelli, Photochem.
Photobiol. 2005, 81, 793.
[14] A. A. Frimer, ‘Singlet Oxygen’, CRC Press, Boca Raton, 1985, Vols. I–IV.
[15] A. M. Braun, M. T. Maurette, E. Oliveros, ‘Photochemical Technology’, John Wiley & Sons, Chi-
chester, 1991, Chapt. 11, and refs. cit. therein.
[16] E. Cadenas, Annu. Rev. Biochem. 1989, 58, 79.
[17] K. U. Schallreuter, J. M. Wood, M. R. Pittelkow, M. Gütlich, K. R. Lemke, W. Rödl, N. N. Swanson,
K. Hitzemann, I. Ziegler, Science 1994, 263, 1444.
[18] H. Rokos, W. D. Beazley, K. U. Schallreuter, Biochem. Biophys. Res. Commun. 2002, 292, 805.
[19] K. U. Schallreuter, J. Moore, J. M. Wood, W. D. Beazley, E. M. J. Peters, L. K. Marles, S. C. Behrens-
Williams, R. Dummer, N. Blau, B. Thöny, J. Invest. Dermatol. 2001, 116, 167.
[20] A. Albert, Biochem. J. 1953, 54, 646.
[21] R. Baur, M. Kappel, R. Mengel, W. Pfleiderer, in ‘Chemistry and Biology of Pteridines‘, Eds. R. L.
Kisliuk, G. M. Brown, Elsevier, North Holland, New York, 1979, p. 13.
[22] W. Pfleiderer, M. Kappel, R. Baur, in ‘Biochemical and Clinical Aspects of Pteridines‘, Eds. W.
Pfleiderer, H. Wachter, H. C. Curtius, Walter de Gruyter, Berlin, New York, 1984, Vol. 3, p. 3.
[23] F. M. Cabrerizo, A. H. Thomas, C. Lorente, M. L. Dántola, G. Petroselli, R. Erra-Balsells, A. L.
Capparelli, Helv. Chim. Acta 2004, 87, 349.
[24] A. H. Thomas, G. Suárez, F. M. Cabrerizo, R. Martino, A. L. Capparelli, J. Photochem. Photobiol.,
A 2000, 135, 147.
[25] A. H. Thomas, G. Suárez, F. M. Cabrerizo, F. S. García Einschlag, R. Martino, C. Baiocchi, E. Pra-
mauro, A. L. Capparelli, Helv. Chim. Acta 2002, 85, 2300.
[26] A. M. Braun, M. T. Maurette, E. Oliveros, ‘Photochemical Technology‘, John Wiley & Sons, Chi-
chester, 1991, Chapt. 2.
[27] F. García Einschlag, M. R. Féliz, A. L. Capparelli, J. Photochem. Photobiol., A 1997, 110, 235.
[28] C. C. Allain, L. S. Poon, C. S. G. Chan, W. Richmond, P. C. Fu, Clin. Chem. 1974, 20, 470.
[29] H. M. Flegg, Ann. Clin. Biochem. 1973, 10, 79.
[30] J. M. McCord, J. Fridovich, J. Biol. Chem. 1969, 244, 6049.
[31] D. E. Moore, J. Wang, J. Photochem. Photobiol., B 1998, 43, 175.
[32] A. A. Krasnovsky Jr., Biophysics 1979, 24, 769.
[33] A. U. Kahn, Chem. Phys. Lett. 1980, 72, 112.
[34] P. Murasecco-Suardi, E. Oliveros, A. M. Braun, H.-J. Hansen, Helv. Chim. Acta 1988, 71, 1005.

1104
Helvetica Chimica Acta – Vol. 89 (2006)
[35] T. Aminian-Saghafi, G. Nasini, T. Caronna, A. M. Braun, E. Oliveros, Helv. Chim. Acta 1992, 75,
531.
[36] A. M. Braun, E. Oliveros, Pure Appl. Chem. 1990, 62, 1467.
[37] E. Oliveros, P. Suardi-Murasecco, T. Aminian-Saghafi, A. M. Braun, Helv. Chim. Acta 1991, 74, 79.
[38] L. A. Martínez, C. G. Martínez, B. B. Klopotek, J. Lang, A. Neuner, A. M. Braun, E. Oliveros, J.
Photochem. Photobiol., B 2000, 58, 94.
[39] C. S. Foote, E. L. Clennan, in ‘Active Oxygen in Chemistry‘, Eds. C. S. Foote, J. S. Valentine, A.
Greenberg, J. F. Liebman, Chapman & Hall, New York, 1995, Vol. 2, Chapt. 4.
[40] R. Schmidt, C. Tanielian, R. Dunsbach, C. Wolff, J. Photochem. Photobiol., A 1994, 79, 11.
[41] HyperChem, vers. 7.01 for Windows, Hypercube Inc., Ontario, 2002.
[42] F. M. Cabrerizo, M. L. Dántola, A. H. Thomas, C. Lorente, A. M. Braun, E. Oliveros, A. L. Cappar-
elli, Chem. Biodiv. 2004, 1, 1800.
[43] A. H. Thomas, G. Suárez, F. M. Cabrerizo, A. L. Capparelli, Helv. Chim. Acta 2001, 84, 3849.
[44] R. Mengel, W. Pfleiderer, W.-R. Knappe, Tetrahedron Lett. 1977, 32, 2817.
[45] W. Pfleiderer, R. Gottlieb, Heterocycles 1980, 14, 1603.
[46] W. Lown, M. H. Akhtar, R. S. McDaniel, J. Org. Chem. 1974, 39, 1998.
[47] R. Kuciel, A. Mazurkiewicz, Biochem. Mol. Biol. Educ. 2004, 32, 183.
[48] B. Bielski, D. Cabelli, R. Arudi, J. Phys. Chem. Ref. Data 1985, 14, 1041.
[49] A. H. Thomas, C. Lorente, A. L. Capparelli, M. R. Pokhrel, A. M. Braun, E. Oliveros, Photochem.
Photobiol. Sci. 2002, 1, 421.
Received February 16, 2006