Hydroxyl radical induced oxidation of theophylline in water: A kinetic and mechanistic study

Article abstract of DOI:10.1039/c4ob00102h

Oxidative destruction and mineralization of emerging organic pollutants by hydroxyl radicals (OH) is a well established area of research. The possibility of generating hazardous by-products in the case of OH reaction demands extensive investigations on the degradation mechanism. A combination of pulse radiolysis and steady state photolysis (H₂O₂/UV photolysis) followed by high resolution mass spectrometric (HRMS) analysis have been employed to explicate the kinetic and mechanistic features of the destruction of theophylline, a model pharmaceutical compound and an identified pollutant, by OH in the present study. The oxidative destruction of this molecule, for intermediate product studies, was initially achieved by H₂O ₂/UV photolysis. The transient absorption spectrum corresponding to the reaction of OH with theophylline at pH 6, primarily caused by the generation of (T8-OH), was characterised by an absorption band at 330 nm (k₂ = (8.22 ± 0.03) × 10⁹ dm³ mol^-1 s^-1). A significantly different spectrum (λ_max: 340 nm) was observed at highly alkaline pH (10.2) due to the deprotonation of this radical (pK_a ～ 10.0). Specific one electron oxidants such as sulphate radical anions (SO₄^-) and azide radicals (N ₃) produce the deprotonated form (T(-H)) of the radical cation (T⁺) of theophylline (pK_a 3.1) with k₂ values of (7.51 ± 0.04) × 10⁹ dm³ mol^-1 s^-1 and (7.61 ± 0.02) × 10⁹ dm³ mol^-1 s^-1 respectively. Conversely, oxide radicals (O ^-) react with theophylline via a hydrogen abstraction protocol with a rather slow k₂ value of (1.95 ± 0.02) × 10⁹ dm³ mol^-1 s^-1. The transient spectral studies were complemented by the end product profile acquired by HRMS analysis. Various transformation products of theophylline induced by OH were identified by this technique which include derivatives of uric acids (i, iv & v) and xanthines (ii, iii & vi). Further breakdown of the early formed product due to OH attack leads to ring opened compounds (ix-xiv). The kinetic and mechanistic data furnished in the present study serve as a basic frame work for the construction of OH induced water treatment systems as well as to understand the biological implications of compounds of this kind. the Partner Organisations 2014.

Full text of DOI:10.1039/c4ob00102h

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Hydroxyl radical induced oxidation of theophylline
in water: a kinetic and mechanistic study†
M. M. Sunil Paul,a U. K. Aravind,b G. Pramod,c A. Sahad and C. T. Aravindakumar*a,e
Cite this: Org. Biomol. Chem., 2014,
12, 5611
Oxidative destruction and mineralization of emerging organic pollutants by hydroxyl radicals (•OH) is a
•
well established area of research. The possibility of generating hazardous by-products in the case of OH
reaction demands extensive investigations on the degradation mechanism. A combination of pulse radio-
lysis and steady state photolysis (H2O2/UV photolysis) followed by high resolution mass spectrometric
(HRMS) analysis have been employed to explicate the kinetic and mechanistic features of the destruction
•
of theophylline, a model pharmaceutical compound and an identified pollutant, by OH in the present
study. The oxidative destruction of this molecule, for intermediate product studies, was initially achieved
by H2O2/UV photolysis. The transient absorption spectrum corresponding to the reaction of •OH with
theophylline at pH 6, primarily caused by the generation of (T8-OH)•, was characterised by an absorption
band at 330 nm (k2 = (8.22 0.03) × 109 dm3 mol−1 s−1). A significantly different spectrum (λmax: 340 nm)
was observed at highly alkaline pH (10.2) due to the deprotonation of this radical (pKa ∼ 10.0). Specific
•
one electron oxidants such as sulphate radical anions (SO4•−) and azide radicals (N3 ) produce the
deprotonated form (T(−H)•) of the radical cation (T•+) of theophylline (pKa 3.1) with k2 values of (7.51
0.04) × 109 dm3 mol−1 s−1 and (7.61 0.02) × 109 dm3 mol−1 s−1 respectively. Conversely, oxide radicals
(O•−) react with theophylline via a hydrogen abstraction protocol with a rather slow k2 value of (1.95
0.02) × 109 dm3 mol−1 s−1. The transient spectral studies were complemented by the end product profile
•
acquired by HRMS analysis. Various transformation products of theophylline induced by OH were ident-
ified by this technique which include derivatives of uric acids (i, iv & v) and xanthines (ii, iii & vi). Further
breakdown of the early formed product due to •OH attack leads to ring opened compounds (ix–xiv). The
kinetic and mechanistic data furnished in the present study serve as a basic frame work for the construc-
tion of •OH induced water treatment systems as well as to understand the biological implications of com-
pounds of this kind.
Received 14th January 2014,
Accepted 30th April 2014
DOI: 10.1039/c4ob00102h
www.rsc.org/obc
1. Introduction
Hydroxyl radicals (•OH) are the key species responsible for the
oxidative destruction of environmental pollutants in many
aSchool of Environmental Sciences, Mahatma Gandhi University, Kottayam, India.
E-mail: cta@mgu.ac.in
Advanced Oxidation Processes (AOPs) such as TiO2 photocata-
lysis, H2O2/UV photolysis, sonolysis, γ-radiolysis etc.,1–7 and
many biological processes including DNA damage, mutation
and ageing.8–16 The •OH induced destruction of emerging
organic pollutants has been widely experimented1–4,7,17,18
since the presence of pharmaceutically active compounds in
the aquatic environment became an emerging environmental
issue.19–22 Pulse radiolysis studies reveal that •OH reacts with a
large number of organic molecules with a nearly diffusion con-
bAdvanced Centre of Environmental Studies and Sustainable Development, Mahatma
Gandhi University, Kottayam, India
cDepartment of Chemistry, N.S.S. Hindu College, Changanachery, India
dUGC-DAE Consortium for Scientific Research, Kolkata Centre, Kolkata, India
eInter University Instrumentation Centre, Mahatma Gandhi University, Kottayam,
India
†Electronic supplementary information (ESI) available: The spectral and kinetic
•−
•
parameters of the reaction of •OH, SO4
,
N3 and O•− with theophylline
(Table S1); MS and MS/MS spectra of selected transformation products (Fig. S1–
S16); decay (320 nm, 350 nm) traces in the case of reaction of O•− (Fig. S17); tran- trolled rate.5,6,12,15,23,24 Furthermore, the ability to achieve
sient absorption spectrum of theophylline during its reaction with SO4•− and N3
•
complete mineralization2,6 makes techniques of these kinds
at longer time scale (Fig. S18); UV-Vis spectrum of theophylline at pH 6.0 and
acceptable and environmentally friendly.
10.1 (Fig. S19); plot of absorbance of transient at 330 nm obtained by the reac-
tion of theophylline with •OH against pH (Fig. S20); percentage degradation of
Conversely, recent studies using high resolution mass
spectrometric (HRMS) techniques demonstrated that the for-
mation of aromatic intermediates (that are likely toxic) as a
theophylline in N2 purged and aerated conditions as
(Fig. S21). See DOI: 10.1039/c4ob00102h
a function of time
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5611–5620 | 5611

View Article Online
Paper
Organic & Biomolecular Chemistry
result of •OH attack is viable in the case of a number of with oxidizing radicals is very valuable in the case of Advanced
organic compounds.5,6,17,18 This could cause serious health Oxidation Technologies as well as in biological systems mainly
problems on targeted organisms because of their potential due to the anti-oxidant properties of methylxanthines28,36 and
biological activities. The kinetic and mechanistic investigation its structural similarities with purine bases.
of the reaction mechanism, including the identification of the
transient intermediate products as well as the stable trans-
formation products, is thus very important in the case of any
2. Materials and methods
•OH mediated destruction protocol prior to its implementation
Theophylline (1,3-dimethyl-3,9-dihydro-1H-purine-2,6-dione,
CAS no. 58-55-9) was purchased from Sigma Aldrich and was
used as received. Potassium persulfate, 2-methyl-2-propanol,
in a real system.
Theophylline (trade name: Deriphyllin), an extensively used
drug for the therapy of different respiratory diseases such as
sodium azide and sodium hydroxide were obtained from
chronic obstructive pulmonary disease (COPD), asthma
etc.,25–27 is used as a model compound for this study. It is
Fischer Scientific. High purity N2O and Ar gas was used for
pulse radiolysis experiments. HPLC and LC-MS grade solvents
naturally found in a variety of products such as cocoa beans
(∼3 mg g−1), green tea (∼0.5 g kg−1), coffee, chocolate and cola
were used for the HPLC and LC-Q-TOF-MS analysis
beverages in variable amounts.25–27 It is reported as a meta-
respectively.
bolite of caffeine28 and a number of naturally occurring alka-
2.1. Steady state photolysis
loids.29 It enhances the anti-inflammatory effect of steroids by
increasing the histone deacetylase (HDAC) activity.30 Theophyl-
Steady state photolysis experiments in the presence of hydro-
line is moderately toxic to mammals (LD50 > 200 mg kg−1
)
gen peroxide (H2O2/UV photolysis) were carried out on a photo
such as mice, rabbits, rats etc.31 and exhibits a negative (log P
at pH 7 and 20 °C is −0.02) octane–water partition coeffi-
cient.32 It is reported that higher doses of theophylline (>20 µg
ml−1) cause serious health problems (such as cardiac arrest,
arrhythmias and hypotension) in humans.25–27,33 A recent
report by Antoniou and co-workers demonstrated that the toxi-
city of theophylline increases (around double) when it is
administrated with Ciprofloxacin, a widely used fluoroquino-
lone antibiotic.34 Furthermore, the presence of this compound
in the aquatic environment has been recently reported.21,22
Bioaccumulation and specific biological activities of pharma-
ceutically active compounds like theophylline makes them
harmful to targeted organisms especially in the case of the
aquatic environment.21,22 Studies on the removal/degradation
of this compound from aqueous medium is consequently a
very relevant topic of investigation.
reactor supplied by Scientific Aids & Instruments Corporation
(SAIC, Chennai). The reactor utilizes a 125 W medium pressure
deuterium lamp as the light source, which is kept in a quartz
vessel. The lamp emits a continuous light spectrum between
185 nm to 400 nm region. All the experiments were carried out
at near natural pH (∼6) and room temperature.
2.2. HPLC analysis
The variation in the concentration of theophylline after H2O2/
UV photolysis was monitored by using a Shimadzu LC-20AD
Prominence Liquid Chromatography system coupled with a
Diode Array Detector (at 270 nm). An isocratic elution of
methanol and water (25 : 75) at a flow rate of 0.8 mL min−1
against an Enable C18G column (250 mm × 4.6 mm × 5 µm)
was used as mobile phase.
2.3. LC-Q-TOF-MS analysis
The destruction of this compound by one of the AOP tech-
niques (photocatalysis) has been recently reported.35 However,
detailed information on the kinetic and mechanistic aspects
The stable transformation products of theophylline formed
•
during the reaction of OH were analyzed on a Waters Acquity
•
of its reaction with OH is very limited.36 An in-depth under-
H class UPLC system coupled with a Waters Xevo G2 Quadra-
pole-Time-of-Flight (Q-TOF) high resolution mass spectro-
meter (HRMS) in which the electrospray (ESI) technique was
used for the ionization. The samples were introduced into the
mass spectrometer through a BEH C18 column (50 mm ×
2.1 mm × 1.7 µm). A gradient elution of methanol and water
(0.3 mL min−1) was used as the mobile phase. All the spectra
were recorded in both positive and negative ionization modes
in mass-to-charge (m/z) ranges of 50–600 Da.
•
standing of the kinetic and mechanistic aspects of the OH
reaction of this compound is thus very necessary for the
proper application of AOP techniques for the removal of this
compound from water. A systematic investigation of the reac-
tion of theophylline in aqueous medium using photolytically
produced hydroxyl radicals (H2O2/UV) is carried out in the
present study. The kinetics of this reaction and the formation
of possible transient intermediate radicals have been probed
using pulse radiolysis technique. In order to obtain more
insight into the underlying reaction mechanism, the end pro-
2.4. Pulse radiolysis experiments
ducts formed during the reaction were evaluated using a state- Time resolved transient spectroscopic studies of the intermedi-
of-the-art mass spectrometric method. Additionally, a compar- ates were carried out on a 7 MeV linear electron accelerator
ison of •OH reaction of this compound, on the transient (AS & E, USA) connected to an optical absorption detection
spectra and product profile, and with some other specific radi- system (Luzchem, Canada) which consists of a Cermax parallel
cals (SO4•−, N3 and O•−) was also carried out. An in-depth lamp (175 W), a monochromator and Hamamastu photomulti-
•
mechanistic understanding of the reaction of this compound plier (R-7400U-04). The dose per pulse of the electron beam
5612 | Org. Biomol. Chem., 2014, 12, 5611–5620
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
was determined by KSCN dosimetry as per the reported pro-
It is well-known that photolysis of H2O2 yields •OH
cedure and was found to be around 16 Gy per 100 ns pulse.37 (eqn (3)).23 The entire degradation is thus attributed to the oxi-
The transient species generated as a result of pulse irradiation dation reaction initiated by •OH.
was monitored by time resolved UV-Visible spectroscopy. All
H2O2 þ hν ! 2•OH
ð3Þ
other details of the accelerator and detection system have been
published elsewhere.38 Formation processes of primary radicals
along with ionic and molecular species by radiolysis of water
and the creation of an oxidative environment by removing redu-
cing species using N2O gas are given in eqn (1) and (2).23
•
The efficient destruction of this compound induced by OH
is very promising especially for the remediation of this pollu-
tant from the aquatic environment. On the other hand, this
kind of degradation may not lead to a complete mineralisation
of the pollutants. It may also result in a number of intermedi-
ate products which may be hazardous too. Considering these
possibilities, it is proposed that a thorough understanding of
the kinetic and mechanistic aspects of the reaction leading to
the destruction of this compound is very important and
should be evaluated.
H2O
•OH ð0:28Þ; H• ð0:06Þ; eaqꢀð0:27Þ;
ð1Þ
ð2Þ
Hþ ð0:27Þ; H2 ð0:05Þ; H2O2 ð0:07Þ
eaqꢀ þ N2O þ H2O ! N2 þ OH þꢀ OH
•
The values given in parentheses are called G-values which
represent the number of species formed per 100 eV of incident
radiation.18,23
3.2. Product analysis
One convenient method of end product profiling is the use of
a high resolution mass spectrometric (HRMS) method such as
LC-Q-TOF-MS which provides the accurate masses of each
compound. Another advantage of LC-Q-TOF-MS is the ability
to analyze highly complex samples without any pre-purifi-
cation. LC-Q-TOF-MS analyses were carried out on a number of
photo irradiated samples in which nearly 20–60% of the
parent compound was transformed into different products. To
enhance the possibility of detecting minor products that are
expected to be in very low concentrations, an initial concen-
tration of 1 × 10−4 mol dm−3 theophylline is employed for
product studies. The Total Ion Chromatogram (TIC) of theo-
phylline after 6 min of UV irradiation in the presence of H2O2
is shown in Fig. 1.
3. Results and discussion
3.1. Steady state degradation studies
Preliminary experiments were performed to evaluate the
efficiency of •OH mediated destruction of this drug. The initial
concentration of theophylline for these experiments was fixed
as 1 × 10−5 mol dm−3 in order to have a workable sensitivity
for the HPLC-DAD system, however, this is considerably higher
than the reported concentration of pharmaceutical com-
pounds in water.21,22 The photo-irradiation in the presence of
H2O2 results in a complete disappearance of theophylline
within 8 min. Fig. 1 shows the degradation profile of this com-
pound as a function of irradiation time. At this timescale, the
destruction of theophylline by the direct effect of UV radiation
is not significant (data not shown).
The peak with m/z 181 (obs. mass 181.0727; calcd for
C7H9N4O2 is 181.0726) in positive mode (ESI+) and 179 (obs.
mass 179.0565; calcd for C7H7N4O2 is 179.0569) in negative
mode (ESI−) represents the parent molecule (data not shown).
All other peaks in the TIC stand for various oxidation products
of theophylline. To gain further structural features of various
peaks in the TIC, MS/MS studies were carried out. The primary
•
products formed during the reaction of theophylline with OH
comprise
a hydroxylation product (i) and two isomeric
demethylation products (ii & iii). Compound i (1,3-dimethyl-
uric acid), the major product of this reaction, was character-
ized by an intense peak at m/z 197 (obs. mass 197.0675; calcd
for C7H9N4O3 197.0675) in ESI+ and 195 (obs. mass 195.0510;
calcd for C7H7N4O3 195.0518) in ESI−. The demethylation pro-
ducts, 1-methylxanthine (ii) and 3-methylxanthine (iii), show
peaks at m/z 167 (obs. mass 167.0567; calcd for C6H7N4O2
167.0569) in ESI+ and 165 (obs. mass 165.0412; calcd for
C6H5N4O2 165.0413) in ESI−. The substantial difference in the
MS/MS spectra (Fig. S1 and S2, ESI†) permits the identification
of iii and iv using tandem mass spectrometry. Further trans-
formation of the primary products results in isomeric uric
acids like 1-methyluric acid (iv), 3-methyluric acid (v) or
xanthine (vi). The uric acids (iv and v) show peaks at m/z 183
(obs. mass 183.0518; calcd for C6H7N4O3 183.0518) in ESI+
Fig. 1 Photo-degradation profile of theophylline in the presence of
H2O2 as a function of time. [theophylline]0 = 1 × 10−5 mol dm−3; [H2O2]0
=
5 × 10−5 mol dm−3. (Inset) Total ion chromatogram of theophylline after
H2O2/UV photolysis. ([theophylline]0 = 1 × 10−4 mol dm−3; transform-
ation of theophylline = 37.5%).
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5611–5620 | 5613

View Article Online
Paper
Organic & Biomolecular Chemistry
and 181 (obs. mass 181.0360; calcd for C6H5N4O3 181.0362) in
ESI− while the mass spectra of xanthine (vi) exhibits an
intense peak at m/z 153 (obs. mass 153.0410; calcd for
C5H5N4O2 153.0413) in ESI+ and 151 (obs. mass 151.0254;
calcd for C5H3N4O2 151.0256) in ESI−. Compounds i–vi are
previously accounted as microbial metabolites of theophyl-
line.39 The probable mechanism leading to the formation of
these compounds are briefly discussed in section 3.4. In
addition to these products (i–vi), eight more transformation
products (vii–xiv) have been identified. Most of these com-
pounds likely originated from the opening of the imidazole
•
ring of theophylline by OH. The entire list of transformation
products, with molecular weight and elemental composition,
identified by LC-Q-TOF-MS analysis is given in Table 1. The
proposed chemical structures of the identified products are
presented in Scheme 2. The mass spectra and MS/MS patterns
of the transformation products of theophylline are given in the
ESI (Fig. S1–S16†).
Fig. 2 Plot of pH vs. absorbance at 272 nm. (Inset) UV-Vis spectrum of
theophylline.
3.3.1. Reaction with hydroxyl radicals. Pulse radiolysis of
N2O saturated solution of theophylline at pH 5.9 yielded a
transient absorption spectrum having a sharp absorption
maximum at 330 nm and a broad absorption around 500 nm
(Fig. 3). The absorption band at 500 nm completely decays
3.3. Pulse radiolysis studies
In order to obtain more insight into the kinetic and mechanis-
tic aspects of a reaction, the information concerning the short
lived intermediates (i.e., initial stages of reaction) is of utmost
importance. Pulse radiolysis is the ultimate technique for
monitoring the short lived transient species in a nano/micro
second time scale.15 The combination of pulse radiolysis and
mass spectrometric techniques is proven as the ideal metho-
dology for obtaining valuable mechanistic insights.5,6,17 The
pKa value of theophylline was initially determined by plotting
the dependence of absorbance at 272 nm (λmax of theophyl-
line) against the pH of the solution (Fig. 2). Theophylline
thus exhibits two distinct pKa (Fig. 2); at 3.3 and 9.8 corres-
ponding to the protonation and deproptonation of NH group
respectively.
within 100 μs with a first order rate constant of 4.91 × 104 s−1
.
Conversely the absorption band at 330 nm shows a slow decay
(Fig. 3). The bimolecular rate constant corresponding to the
reaction of theophylline with hydroxyl radical was determined
by following the growth of the transient at the 330 nm within a
concentration range of (0.5–1) × 10−4 mol dm−3. A rate con-
stant of (8.22 0.03) × 109 dm3 mol−1 s−1 is obtained for this
reaction (Fig. 4).
Alternatively, the transient absorption spectrum at pH 10.2
(above pKa) shows a red shift (λmax: 340 nm; k2: (7.11 0.07) ×
Table 1 List of transformation products identified by LC-Q-TOF-MS
analysis
Mol.
wt.
Elemental
composition
Sl. no. Product
i
ii
iii
iv
v
vi
vii
1,3-Dimethyluric acid
1-Methylxanthine
3-Methylxanthine
1-Methyluric acid
3-Methyluric acid
Xanthine
1/3-Methyltetrahydro-1H-purine-
2,6-dione
196.16 C7H8N4O3
166.14 C6H6N4O2
166.14 C6H6N4O2
182.14 C6H6N4O3
182.14 C6H6N4O3
152.11 C5H4N4O2
168.15 C6H8N4O2
viii
ix
8-Hydroxy-1/3-methyl-3,7,8,9-
tetrahydro-1H-purine-2,6-dione
5/6-Amino derivative of 5/6-hydroxy-1,3- 171.15 C6H9N3O3
184.15 C6H8N4O3
dimethylpyrimidine-2,4(1H,3H)-dione
x
5/6-Amino derivative of 1/3-
methylpyrimidine-2,4(1H,3H)-dione
141.13 C5H7N3O2
xi
xii
5/6-Aminopyrimidine-2,4(1H,3H)-dione 127.10 C4H5N3O2
5/6-Amino derivative of 5/6-hydroxy-
145.12 C4H7N3O3
Fig. 3 Transient absorption spectrum recorded during the reaction of
dihydropyrimidine-2,4(1H,3H)-dione
1/3-Methylpyrimidine-2,4(1H,3H)-dione 126.11 C5H6N2O2
5,6-Diaminopyrimidine-2,4(1H,3H)-
•OH with theophylline (1 × 10−4 mol dm−3) at pH 5.9 after ( ) 10 and (
●
○
)
xiii
xiv
□
100 μs and at pH 10.2 ( ) after 10 μs of the pulse. (Inset) Decay traces at
142.12 C4H6N4O2
dione
(1) 330 nm (pH 5.9), (2) 500 nm (pH 5.9) and (3) 340 nm (pH 10.2).
5614 | Org. Biomol. Chem., 2014, 12, 5611–5620
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 Plot of kobs against [theophylline] in the case of the reaction of
Fig. 5 Transient absorption spectrum of theophylline (1 × 10−4 mol
•
•−
●
○
▲
OH at pH ( ) 5.9 (330 nm), ( ) 10.2 (340 nm) and ( ) O (350 nm).
(Inset) Transient absorption spectrum recorded during the reaction of
•− with theophylline (1 × 10−4 mol dm−3) after 20 μs of the pulse.
•−
dm−3) recorded during its reaction with SO4 after ( ) 7 μs (pH 6.0) and
●
•
with N3 after (O) 6 μs (pH 6.1). (Inset) Plot of kobs against [theophylline]
O
•−
in the case of reaction of ( ) SO4 (350 nm) and ( ) N3•− (340 nm).
●
○
109 dm3 mol−1 −1) in the UV region lacking any significant
s
absorption in the visible region (Fig. 3).
The reaction of sulfate radical with theophylline at a more
basic pH (9.3) also shows an absorption band at the same
region (data not shown) though the bimolecular rate constant
(k2 = (5.37 0.03) × 109 dm3 mol−1 s−1) obtained in this case is
somewhat lower.
3.3.2. Reaction with oxide radical. The oxide radical anion
(O•−) predominantly undergoes one electron transfer and/or
hydrogen abstraction reactions.23 A comparison of the inter-
mediate spectra from this reaction would help in the identifi-
cation of the intermediate species from •OH reaction. The
transient absorption spectrum obtained from the reaction of
O•− (at pH > 13) is characterized by two absorption bands at
320 nm and 350 nm (Fig. 4) without having any noteworthy
change in the decay profile (Fig. S17, ESI†). At pH > 13, more
than 90% of the hydroxyl radicals are converted into O•− with
a pKa value of 11.9 according to the following equation.
Transient absorption spectra having complementary nature
•
were obtained in the case of the reaction of N3 at pH 6.1
(Fig. 5), 4.0 and 9.5 (data not shown). The bimolecular rate
constant obtained at pH 6.1 (k2 = (7.61 0.02) × 109 dm3 mol−1
s
−1) and 9.6 (k2 = (8.42 0.06) × 109 dm3 mol−1 −1) are very
s
similar. However, the rate constant value obtained at pH 4
(k2 = (4.05 0.02) × 109 dm3 mol−1 s−1) is reduced to nearly
half. It is noteworthy to mention the fast decay of the transi-
ent, with a first order rate constant (kobs) of 3.0 × 104 s−1, at pH
6 though a moderately slow decay is observed in the case of
reaction of SO4•−. The substantial variation in the decay profile
•OH þ ꢀOH $ O•ꢀ þ H2O
ð4Þ
The calculated bimolecular rate constant for the reaction of
O•− with theophylline from the formation kinetics (Fig. 4) at
•−
of the transient in the case of SO4 is accounted by inter-
350 nm is (1.95 0.02) × 109 dm3 mol−1 s−1
.
•−
ference caused by the absorption of SO4 from the UV region.
3.3.3. Reaction with specific one electron oxidants. Being
It is also worth mentioning the similarities of both spectra
•−
•
•−
(SO4•− and N3 ) at longer timescales (less interference of SO4
;
very selective in their reactions, sulfate radical anion (SO4
)
•
and azide radical (N3 ) are ideal for evaluating the contribution
of electron transfer reactions in the case of •OH reaction. Gene-
Fig. S18, ESI†). The spectral and kinetic parameters of
different radicals on its reaction with theophylline are sum-
marized in Table S1 of the ESI.†
•−
•
ration of SO4 and N3 by pulse radiolysis is exemplified in
reactions 5–7.40–42
As a consequence of the structural similarities of theophyl-
line with purine, it is presumed to go through a similar reac-
tion protocol with •OH. Recent works of Chatgilialoglu and co-
workers recommended that the main site of hydroxyl radical
attack at the base part of the guanosine and deoxyguanosine
moiety is the exocyclic amino group (∼65%) whereas the
hydroxylation at the C8 (∼17%) position plays only a minor
role.9,10 The lack of an amino group in theophylline results in
the primary reaction of the hydroxyl radical at carbon 8 to
form the corresponding hydroxyl radical adduct of the form
(T8-OH)• represented in Scheme 1. The analogous stable
•
•
ðCH3Þ3COH þ OH ! CH2ðCH3Þ2COH þ H2O
ð5Þ
ð6Þ
ð7Þ
S2O82ꢀ þ eaq ! SO4•ꢀ þ SO4
ꢀ
2ꢀ
N3ꢀ þ OH ! N3 þ ꢀOH
•
•
•−
Reaction of SO4 at pH 6.0 yielded a transient absorption
spectrum having a broad absorption band around 320–380 nm
region (Fig. 5). A bimolecular rate constant of (7.51 0.04) ×
109 dm3 mol−1 s−1 is obtained for this reaction (Fig. 5).
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5611–5620 | 5615

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Proposed mechanism of the reaction of •OH, SO4•−, N3• and O•− with theophylline.
•
product, 1,3-dimethyluric acid (i), identified in our end that OH predominantly undergoes addition with similar com-
product studies further supports this assignment. The gene- pounds and such adduct radicals exhibit characteristic absorp-
ration of a similar kind of radical is formerly reported in the tion bands in this region.10,12,15,18 On the other hand, the slow
case of the oxidation of caffeine.43
decay observed at 330 nm compared to 500 nm at pH 6 (Fig. 3)
The C8-OH adduct of (deoxy) guanosine generally under- could be due to the interference of the radicals b and c. It is
goes (i) opening of the imidazole ring (yields compounds like also predicted that the weak absorption around 500 nm is the
2,6-diamino-4-hydroxy-5-formamidopyrimidine,
commonly contribution of radical a since there is no information on the
referred as FapyG)10 and (ii) bimolecular transformation to formation of other OH adduct radicals as demonstrated in our
8-oxoguanine.10 Product studies using LC-Q-TOF-MS show analo- LC-Q-TOF-MS results.
gous compounds corresponding to these reactions suggesting
At higher alkaline pH, theophylline exists as its deproto-
the feasibility of both processes in the present case. When the nated form represented as (T−H)− (pKa ∼ 9.8) in Scheme 1.
samples after pulse irradiation were subjected to end product The transient absorption spectrum recorded in the case of
studies, demethylated products like 1- and 3-methylxanthine reaction of •OH with theophylline at pH 10.2 was different
(ii and iii), were detected. Demethylation is supposed to begin from that of pH 6.0 by a red shift in the λmax and a higher
with hydrogen abstraction (see section 3.4.2), the contribution absorbance value at UV region (Fig. 3). A similar shift in λmax
of radical(s) generated by hydrogen abstraction from the and absorbance values was also noticed in the ground state
methyl side chains of theophylline is expected in the transient absorption spectrum of theophylline (Fig. S19, ESI†). The
spectrum. Therefore, it is proposed that the experimental spec- product profile obtained from the high resolution mass
trum observed in the case of reaction of •OH with theophylline spectrometer is nearly the same as that at pH 6.0 apart from
has a contribution from the radicals b and c in addition to the the lack of ring-opened products like ix. Since 1,3-dimethyluric
•
primary radical a. The evidence for the absorbance of the radi- acid is also detected at this pH, the addition of OH at carbon
cals b and c came from the interpretation of the transient 8 is likely responsible for the transient absorption. The plot of
spectrum resulting from the reaction of O•−, which is known absorbance corresponding to this transient (monitored at
to generate hydrogen abstracted species selectively23 (discus- 330 nm) as a function of pH gave a clear pKa curve having an
sion follows). The exact contribution of these individual radi- inflection point around pH 10.0 (Fig. S20, ESI†). Differences in
cals is difficult to predict. However, it is assumed that the the transient spectrum of theophylline on its reaction with
main absorbing species is the radical a since it is well known •OH at pH 10.2 is thus explained on the basis of the deprotona-
5616 | Org. Biomol. Chem., 2014, 12, 5611–5620
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
tion of the (T8-OH)•. A similar mechanism is previously
Another explanation for the existence of radical cations at
reported in the case of a series of pyrimidine derivatives.12 The highly acidic pH is that the protonated form of theophylline
lack of product ix at pH 10.2 indicates the less feasibility of the (pKa ∼ 3.3), represented as TH+, undergoes one electron oxi-
ring opening reaction of (T8-OH)• above the pKa of dation with N3• to form a TH•2+ (f). The TH•2+ is expected to be
theophylline.
highly unstable and instantly releases a proton to yield theo-
•
It is well-known that one electron oxidants such as N3 and phylline radical cation (T•+). Hence, the broad absorption band
•−
SO4 results radical cations on its reaction with purines and obtained between 320–380 nm in the case of the reaction of
•
•−
other aromatic systems.15,44 Because of the high Brønsted N3 and SO4 (Fig. 5) at neutral and basic pH is assigned to
acidity, the radical cation thus formed either reacts with H2O/ the N7 deprotonated neutral radical T(−H)• (Scheme 1).
OH− (depends on pH) and yields a hydroxyl radical adduct or
One electron transfer and/or hydrogen abstraction reactions
is deprotonated into a neutral radical.15,45 If the former is the are reported as the primary route of the reaction of O•− with
case, the corresponding transient spectra should match with aromatic compounds.23 If electron abstraction occurs, an N7
the one obtained with •OH reaction. However, it is not deprotonated radical (e) will be a possible intermediate. Since
observed in the present case (Fig. 3 and 5). In addition, the the reaction of theophylline with specific one electron oxi-
•
end product studies on the photo-irradiated samples in the dants, that is SO4•− and N3 , also generates the same radical, a
•−
presence of K2S2O8 (generates SO4 in the medium)45 did not similar spectrum is presumed. However, this is not observed
provide any evidence for the formation of 1,3-dimethyluric in the present case (Fig. 4 and 5). The only remaining possi-
acid (see section 3.4.2). The remaining possibility is the depro- bility is the abstraction of a hydrogen atom from either of the
tonation of the radical cation in to a neutral radical rep- methyl side chains of theophylline to form another carbon
resented as T(−H)• (e). Previous studies by Santos et al. in the centered T(−H)• radical (b or c). Hence, the absorption bands
case of one-electron oxidized methylxanthines recommended obtained in the present case are likely due to the formation of
the formation of a N7 deprotonated neutral radical with signifi- radicals b or/and c. It is also worth mentioning the low bimole-
cant stability.46
cular rate constant (k2 = (1.95 0.02) × 109 dm3 mol−1 s−1
)
The studies on the fate of radical cations are pertinent in determined for this reaction (k2 values corresponding to the
biological systems because of their involvement in diseases hydrogen abstraction reactions are usually low).23
like cancer.47,48 To gain more insight into mechanistic aspects
3.4. Reaction mechanism
of the deprotonation of the radical cation, the absorbance of
the transient at 340 nm obtained in the case of reaction of The various transformation products corresponding to the
theophylline with N3• were monitored as a function of pH. The reaction of OH with theophylline have already been discussed
•
result is shown in Fig. 6. The inflection point at pH 3.1 in section 3.2. The transformation pathways of theophylline on
•
obtained by this experiment suggests the existence of a radical its reaction with OH are classified into three stages (stage 1–3)
cation, without deprotonation, only below pH 3.1. The deter- and are described in Scheme 2. Stage 1 explains the formation of
mined pKa value of theophylline radical cation is very close to the primary products of theophylline (Schemes 3 and 4) whereas
that of guanine.15,44
stage 2 and 3 explain further transformation of the primary pro-
ducts. The mechanism of formation of the primary products
induced by •OH are separately discussed in Schemes 3 and 4.
3.4.1. Hydroxylation. The primary transformation product
•
of the reaction of OH with theophylline, irrespective of reac-
tion conditions, is 1,3-dimethyluric acid (i). That is the one
corresponding to hydroxylation of the parent molecule at
carbon 8. The nearly diffusion controlled bimolecular rate con-
stant (k2 = (8.22 0.03) × 109 dm3 mol−1 s−1) corresponding to
•
the reaction of theophylline with OH determined during the
pulse radiolysis studies is in accordance with the hydroxylation
possibility (section 3.3.1). Previous studies on structurally
similar molecules, such as caffeine and guanine, also pointed
to a similar protocol.9,15,43
The mechanism of the formation of this compound
involves one electron oxidation of the early intermediate (T8-
OH)• followed by deprotonation (Scheme 3). The OH radical
adduct of guanine also undergoes an analogous mechanism
that leads to the formation of 8-oxo-G,49 a widely used bio-
marker for oxidative stress.9,15
3.4.2. Demethylation. Products corresponding to demethyl-
ation of the methyl side chains of theophylline, that is 1 and
3-methylxanthine (ii and iii, both having M.W. 166.14 but
Fig. 6 Plot of absorbance of the transient at 340 nm against pH in the
•−
case of reaction of N3
.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5611–5620 | 5617

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 2 Proposed degradation pathways of theophylline on its reaction with •OH.
mol dm−3 K2S2O8. Sulfate radical anion (SO4•−), a specific one
electron oxidant, is the reacting species in this case.45 If one
electron oxidation is the fundamental mechanism, ii and iii
are expected in the end product profile. However, the targeted
MS/MS analysis using LC-Q-TOF-MS did not give any infor-
mation towards the formation of ii and iii. The involvement of
one electron oxidation is thus eliminated from this case. The
remaining possibility is hydrogen abstraction. A multi-step
mechanism is proposed by Santos and co-workers for this con-
version.36 The first step of the mechanism involves the hydrogen
abstraction from one of the methyl side chains (for example: N3–
CH3 in Scheme 4) to form a carbon centered radical (b). An
instant electron release from radical b followed by hydrolysis
generates a hydroxymethyl intermediate compound. The liber-
ation of hydroxymethyl groups from this intermediate results in
the corresponding demethylated compound (ii).
Scheme 3 Mechanism of the formation of 1,3-dimethyluric acid from
theophylline.
Scheme 4 Mechanism of the formation of 1-methylxanthine from
3.4.3. Other products. The products iv–xiv, generated by
the further conversion of the initial products by reactive
species, were detected as minor peaks in the TIC. The mecha-
nism of the isomeric uric acids (iv and v) is explained by two
theophylline.
different fragmentation pattern, Fig. S1 and S2, ESI†), are
important minor products in this reaction. This transform-
ation is of biological relevance because of the potent anti-
oxidant activities of methylxanthines.36 Two distinct mecha-
nisms viz. hydrogen abstraction and one electron oxidation are
proposed as the gateway to demethylation.36,50 To account for
the contribution of one electron oxidation in this process, end
product studies were carried out in photo-irradiated samples
of theophylline (1 × 10−4 mol dm−3) in the presence of 1 × 10−3
distinct mechanisms. The first one is
a hydroxylation–
demethylation protocol in which the demethylation of i,
formed as a result of the hydroxylation of theophylline at
carbon 8 (see Scheme 2), at N1 and N3–CH3 groups respectively
results iv and v. Alternately, a demethylation–hydroxylation
protocol is also possible in which hydroxylation of ii and iii at
carbon 8 results in the corresponding products. Likewise,
demethylation of ii and iii results in xanthine (vi) (Scheme 5).
5618 | Org. Biomol. Chem., 2014, 12, 5611–5620
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
4. Conclusions
The kinetic and mechanistic aspects of a representative
methylxanthine drug, theophylline, against hydroxyl radical
(•OH) have been explicated by pulse radiolysis and LC-Q-
TOF-MS techniques. Two pathways – addition and hydrogen
•
abstraction – are demonstrated for the reaction of OH. The
adduct radical (T8-OH)• undergoes a ring opening reaction
and yields products analogous to FapyG (a potential biomarker
for oxidative stress). LC-Q-TOF-MS analysis undoubtedly
demonstrates the role of hydrogen abstraction also in the
initial step of this reaction. This result is rather unusual as
•OH generally undergoes addition reaction with similar com-
pounds. Characterization of various transformation products
including isomeric xanthines (ii & iii) and uric acids (iv & v) by
LC-Q-TOF-MS is an important illustration which is generally
difficult to achieve. The rapid destruction of theophylline by
H2O2/UV photolysis is an excellent hint at the vulnerability of
this compound against oxidizing radicals. This is an important
finding in the context of oxidation technologies for pollutant
degradation. However, the efficiency of oxidizing radicals
against the complete destruction of the parent as well as the
transformed organic compound in natural waters could possi-
Scheme 5 Formation of xanthine and methyluric acids from
theophylline.
As the photo-irradiation proceeds, the concentrations of
some of the transformation products such as i could be equal
(or even exceed) that of the parent compound. Hence the prob-
•
ability of OH attack on the initially formed products rises.
Similar observations were previously reported in the case of
many organic compounds.5,6,17
The opening of the imidazole ring of (T8-OH)• is presumed
to be responsible for the formation of compounds ix–xiv. Ana-
logous reactions are well studied in the case of the hydroxyl
radical adduct of guanine derivatives which leads to the for-
mation of products such as FapyG.9,10,15 However, compounds
like vii and viii, detected at extremely trace level, are not
usually found in oxidative conditions. The mechanism corres-
ponding to the formation of such products, observed in redu-
cing conditions, in the present case is not clear. The possible
side-reactions associated with H2O2 or the direct photo
reduction of some of the initially formed products of theophyl-
line (such as ii–v) by UV radiation is a likely reason for this
conversion. The photo-transformation of aldehydes and
ketones to the corresponding alcohols are familiar in synthetic
chemistry.51 Moreover, the high sensitivity of LC-Q-TOF-MS
allows the detection of such minor products without compro-
mising the mass accuracy.
•
bly be reduced due to scavenging of OH by a variety of in-
organic ions and dissolved organic matter. Therefore,
additional efforts for assessing the efficiency and toxicity/TOC
decline are essential for the practical implementation of these
types of methodologies and are presently in progress.
Acknowledgements
We acknowledge National Centre for Free Radical Research,
University of Pune, Pune for the pulse radiolysis facility. This
work was performed under collaborative research scheme
(Ref no. UGC-DAE-CSR-KC/CRS/09/RC02/1455) of UGC-DAE
Consortium for Scientific Research, Kolkata Centre, Kolkata.
CTA is thankful to DST (under PURSE and FIST scheme) for
partial financial support. GP is thankful to SB College,
Changanasserry for research guideship.
It is worth noting that nearly 20% reduction in the photo
degradation efficiency (at 6 min) is observed in the case of
H2O2/photolysis in N2 purged (that is deaerated) medium
(Fig. S21, ESI†). The high reactivity of (A8-OH)• of N6,N6-
dimethyladenosine towards molecular oxygen is well known.52
A likely reason for the enhanced degradation efficiency in an
aerated medium is the involvement of molecular oxygen in the
reaction mechanism. This result is especially important
because of the presence of molecular oxygen in most of the
biological and environmental systems in which reactions of
•OH occurs. Conversely, the concentration of the major oxi-
dation product, 1,3-dimethyluric acid, formed in the case of N2
purged conditions was significantly higher. It is thus pre-
sumed that the presence of molecular oxygen in the medium
induces further oxidation of the initially formed products like
1,3-dimethyluric acid.
Notes and references
1 E. Brillas, I. Sirés and M. A. Oturan, Chem. Rev., 2009, 109,
6570–6631.
2 P. Calza, C. Medana, E. Raso, V. Giancotti and C. Minero,
Rapid Commun. Mass Spectrom., 2011, 25, 2923–2932.
3 J. De Laat, H. Gallard, S. Ancelin and B. Legube, Chemo-
sphere, 1999, 39, 2693–2706.
4 U. Kalsoom, S. S. Ashraf, M. A. Meetani, M. A. Rauf and
H. N. Bhatti, Chem. Eng. J., 2012, 200–202, 373–379.
5 R. Sreekanth, K. P. Prasanthkumar, M. M. Sunil Paul,
U. K. Aravind and C. T. Aravindakumar, J. Phys. Chem. A,
2013, 117, 11261–11270.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5611–5620 | 5619

View Article Online
Paper
Organic & Biomolecular Chemistry
6 M. M. Sunil Paul, U. K. Aravind, G. Pramod and 29 I. Baranowska and J. Kapinos, Chem. Anal., 2000, 45, 53–58.
C. T. Aravindakumar, Chemosphere, 2013, 91, 295–301.
7 D. C. McDowell, M. M. Huber, M. Wagner, U. von Gunten
and T. A. Ternes, Environ. Sci. Technol., 2005, 39, 8014–8022.
8 J. Cadet, T. Douki and J. L. Ravanat, Free Radicals Biol.
Med., 2010, 49, 9–21.
30 B. G. Cosio, A. Iglesias, A. Rios, A. Noguera, E. Sala, K. Ito,
P. J. Barnes and A. Agusti, Thorax, 2009, 64, 424–429.
31 K. V. Blake, K. L. Massey, L. Hendeles, D. Nickerson and
A. Neims, Ann. Emergency Med., 1988, 17, 1024–1028.
32 I. Kupsáková, A. Rybár, P. Dočolomanský, Z. Drobná,
U. Stein, W. Walther, M. Barančík and A. Breier,
Eur. J. Pharm. Sci., 2004, 21, 283–293.
9 C. Chatgilialoglu, M. D’Angelantonio, G. Kciuk and
K. Bobrowski, Chem. Res. Toxicol., 2011, 24, 2200–2206.
10 C. Chatgilialoglu, M. D’Angelantonio, M. Guerra, 33 H. Yin, X. Meng, H. Su, M. Xu and S. Ai, Food Chem., 2012,
P. Kaloudis and Q. G. Mulazzani, Angew. Chem., Int. Ed.,
2009, 48, 2214–2217.
11 A. Kumar, V. Pottiboyina and M. D. Sevilla, J. Phys. Chem.
B, 2011, 115, 15129–15137.
134, 1225–1230.
34 T. Antoniou, T. Gomes, M. Mamdani and D. Juurlink,
Eur. J. Clin. Pharmacol., 2011, 67, 521–526.
35 R. Liang, A. Hu, W. Li and Y. N. Zhou, J. Nanopart. Res.,
2013, 15, 1–13.
12 T. L. Luke, T. A. Jacob, H. Mohan, H. Destaillats,
V. M. Manoj, P. Manoj, J. P. Mittal, M. R. Hoffmann and 36 P. Santos, S. Silva, G. Justino and A. Vieira, Redox Rep.,
C. T. Aravindakumar, J. Phys. Chem. A, 2002, 106, 2497–2504. 2010, 15, 138–144.
13 V. M. Manoj and C. T. Aravindakumar, Chem. Commun., 37 G. V. Buxton and C. R. Stuart, J. Chem. Soc., Faraday Trans.,
2000, 2361–2362.
1995, 91, 279–281.
14 P. O’Neill, Radiat. Res., 1983, 96, 198–210.
15 S. Steenken, Chem. Rev., 1989, 89, 503–520.
38 K. M. S. P. Yadav, M. B. Shirdhonkar and B. S. M. Rao,
Curr. Sci., 2007, 92, 599–605.
16 J. A. Theruvathu, C. T. Aravindakumar, R. Flyunt, J. von 39 C. L. Yu, T. M. Louie, R. Summers, Y. Kale, S. Gopishetty
Sonntag and C. von Sonntag, J. Am. Chem. Soc., 2001, 123,
9007–9014.
17 S. P. Mezyk, T. J. Neubauer, W. J. Cooper and J. R. Peller,
J. Phys. Chem. A, 2007, 111, 9019–9024.
18 W. Song, W. J. Cooper, S. P. Mezyk, J. Greaves and
B. M. Peake, Environ. Sci. Technol., 2008, 42, 1256–1261.
19 K. Kummerer, Chemosphere, 2009, 75, 417–434.
20 T. Heberer, Toxicol. Lett., 2002, 131, 5–17.
and M. Subramanian, J. Bacteriol., 2009, 191, 4624–4632.
40 P. Manoj, H. Mohan, J. P. Mittal, V. M. Manoj and
C. T. Aravindakumar, Chem. Phys., 2007, 331, 351–358.
41 D. B. Naik and K. Kishore, Res. Chem. Intermed., 2002, 28,
603–617.
42 T. L. Luke, H. Mohan, V. M. Manoj, P. Manoj, J. P. Mittal
and C. T. Aravindakumar, Res. Chem. Intermed., 2003, 29,
379–391.
21 A. J. Ramirez, R. A. Brain, S. Usenko, M. A. Mottaleb, 43 J. P. Telo and A. J. S. C. Vieira, J. Chem. Soc., Perkin Trans. 2,
J. G. O’Donnell, L. L. Stahl, J. B. Wathen, B. D. Snyder, 1997, 1755–1758.
J. L. Pitt, P. Perez-Hurtado, L. L. Dobbins, B. W. Brooks and 44 K. Kobayashi and S. Tagawa, J. Am. Chem. Soc., 2003, 125,
C. K. Chambliss, Environ. Toxicol. Chem., 2009, 28,
2587–2597.
22 Y. Vystavna, F. Huneau, V. Grynenko, Y. Vergeles, H. Celle-
10213–10218.
45 P. Neta, R. E. Huie and A. B. Ross, J.Phys. Chem. Ref. Data,
1988, 17, 1027–1284.
Jeanton, N. Tapie, H. Budzinski and P. Le Coustumer, 46 P. M. P. Santos, J. P. Telo and A. J. S. C. Vieira, Redox Rep.,
Water, Air, Soil Pollut., 2012, 223, 2111–2124. 2008, 13, 123–133.
23 G. V. Buxton, C. L. Greenstock, W. P. Helman and 47 E. Cavalieri and E. Rogan, Environ. Health Perspect., 1985,
A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513–886. 64, 69–84.
24 L. P. Candeias and S. Steenken, Chem. – Eur. J., 2000, 6, 48 A. F. Lehner, J. Horn and J. W. Flesher, Biochem. Biophys.
475–484. Res. Commun., 2004, 322, 1018–1023.
25 A. J. Szentmiklosi, A. Cseppento, R. Gesztelyi, J. Zsuga, 49 L. I. Shukla, A. Adhikary, R. Pazdro, D. Becker and
A. Kortvely, G. Harmati and P. P. Nanasi, Curr. Med. Chem.,
2011, 18, 3695–3706.
26 I. I. Koleva, T. A. van Beek, A. E. M. F. Soffers,
B. Dusemund and I. M. C. M. Rietjens, Mol. Nutr. Food
Res., 2012, 56, 30–52.
M. D. Sevilla, Nucleic Acids Res., 2004, 32, 6565–6574.
50 P. M. P. Santos, A. M. M. Antunes, J. Noronha,
E. Fernandes and A. J. S. C. Vieira, Eur. J. Med. Chem., 2010,
45, 2258–2264.
51 N. Kambe, K. Kondo, S. Murai and N. Sonoda, Angew.
Chem., Int. Ed. Engl., 1980, 19, 1008–1009.
27 B. Stavric, Food Chem. Toxicol., 1988, 26, 541–565.
28 J. R. León-Carmona and A. Galano, Int. J. Quantum Chem., 52 A. J. S. C. Vieira and S. Steenken, J. Am. Chem. Soc., 1987,
2012, 112, 3472–3478.
109, 7441–7448.
5620 | Org. Biomol. Chem., 2014, 12, 5611–5620
This journal is © The Royal Society of Chemistry 2014