Probing lipid peroxidation by using linoleic acid and benzophenone

Article abstract of DOI:10.1002/chem.201100983

A thorough mechanistic study has been performed on the reaction between benzophenone (BZP) and a series of 1,4-dienes, including 1,4-cyclohexadiene (CHD), 1,4-dihydro-2-methylbenzoic acid (MBA), 1,4-dihydro-1,2-dimethylbenzoic acid (DMBA) and linoleic acid (LA). A combination of steady-state photolysis, laser flash photolysis (LFP), and photochemically induced dynamic nuclear polarization (photo-CIDNP) have been used. Irradiation of BZP and CHD led to a cross-coupled sensitizer-diene product, together with 6, 7, and 8. With MBA and DMBA as hydrogen donors, photoproducts arising from cross-coupling of sensitizer and diene radicals were found; compound 7 was also obtained, but 6 and o-toluic acid were only isolated in the irradiation of BZP with MBA. Triplet lifetimes were determined in the absence and in the presence of several diene concentrations. All three model compounds showed similar reactivity (k _q≈10⁸ M^-1 s^-1) towards triplet excited BZP. Partly reversible hydrogen abstraction of the allylic hydrogen atoms of CHD, MBA, and DMBA was also detected by photo-CIDNP on different timescales. Polarizations of the diamagnetic products were in full agreement with the results derived from LFP. Finally, LA also underwent partly reversible hydrogen abstraction during photoreaction with BZP. Subsequent hydrogen transfer between primary radicals led to conjugated derivatives of LA. The unpaired electron spin population in linoleyl radical (LA^.) was predominantly found on H(1-5) protons. To date, LA-related radicals were only reported upon hydrogen transfer from highly substituted model compounds by steady-state EPR spectroscopy. Herein, we have experimentally established the formation of LA^. and shown that it converts into two dominating conjugated isomers on the millisecond timescale. Such processes are at the basis of alterations of membrane structures caused by oxidative stress. Copyright

Full text of DOI:10.1002/chem.201100983

FULL PAPER
DOI: 10.1002/chem.201100983
Probing Lipid Peroxidation by Using Linoleic Acid and Benzophenone
Inmaculada Andreu,[a] Dmytro Neshchadin,[b] Enrique Rico,[a] Markus Griesser,[b]
Abdelouahid Samadi,[a] Isabel M. Morera,[a] Georg Gescheidt,*[b] and
Miguel A. Miranda*[a]
Abstract:
A
thorough mechanistic
only isolated in the irradiation of BZP
with MBA. Triplet lifetimes were deter-
mined in the absence and in the pres-
ence of several diene concentrations.
All three model compounds showed
similar reactivity (kq ꢀ108 mÀ1 sÀ1) to-
wards triplet excited BZP. Partly rever-
sible hydrogen abstraction of the allylic
hydrogen atoms of CHD, MBA, and
DMBA was also detected by photo-
CIDNP on different timescales. Polari-
zations of the diamagnetic products
were in full agreement with the results
derived from LFP. Finally, LA also un-
derwent partly reversible hydrogen ab-
straction during photoreaction with
BZP. Subsequent hydrogen transfer be-
tween primary radicals led to conjugat-
ed derivatives of LA. The unpaired
electron spin population in linoleyl rad-
study has been performed on the reac-
tion between benzophenone (BZP)
and a series of 1,4-dienes, including
1,4-cyclohexadiene (CHD), 1,4-dihy-
dro-2-methylbenzoic acid (MBA), 1,4-
C
dihydro-1,2-dimethylbenzoic
acid
ical (LA ) was predominantly found on
HACHTUNGRNET(NUGN 1-5) protons. To date, LA-related
(DMBA) and linoleic acid (LA). A
combination of steady-state photolysis,
laser flash photolysis (LFP), and photo-
chemically induced dynamic nuclear
polarization (photo-CIDNP) have been
used. Irradiation of BZP and CHD led
radicals were only reported upon hy-
drogen transfer from highly substituted
model compounds by steady-state EPR
spectroscopy. Herein, we have experi-
mentally established the formation of
C
to
a
cross-coupled sensitizer–diene
LA and shown that it converts into
product, together with 6, 7, and 8. With
MBA and DMBA as hydrogen donors,
photoproducts arising from cross-cou-
pling of sensitizer and diene radicals
were found; compound 7 was also ob-
tained, but 6 and o-toluic acid were
two dominating conjugated isomers on
the millisecond timescale. Such pro-
cesses are at the basis of alterations of
membrane structures caused by oxida-
tive stress.
Keywords: dienes · hydrogen ab-
straction · photochemistry · photol-
ysis · transient absorption spectros-
copy
Introduction
triplets.[2] After oxygen trapping, the fatty acid free-radical
forms hydroperoxide, which in turn can initiate the peroxi-
dation chain cycle through the generation of hydroxyl and
alkoxyl radicals.[3]
The reaction of the benzophenone (BZP) triplet excited
state with allylic hydrogen atoms has attracted considerable
interest during recent decades. Hydrogen abstraction and p
quenching have been found to be the most important deacti-
vation processes.[4]
Herein, we build up the knowledge of lipid peroxidation
by starting with 1,4-cyclohexadiene (CHD) and its deriva-
tives as simple models and then move on to linoleic acid
(LA), which serves as a realistic paradigm for lipid peroxi-
dation. Studies performed by combining HPLC and laser
flash photolysis (LFP) on tiaprofenic acid, which is a benzo-
phenone-derived, nonsteroidal anti-inflammatory drug, have
demonstrated that LA photoperoxidation is mediated by
both type I and type II mechanisms through hydrogen ab-
straction and singlet oxygen production, respectively.[5] Irra-
diation of BZP in the presence of 1,4-cyclohexadienes
(CHDs) as models for LA leads to a complex mixture of
photoproducts arising from hydrogen abstraction, via triplet
radical pairs.[6] Interestingly, with enantiomerically pure
BZP--CHD model dyads, a significant stereodifferentation
Peroxidation of polyunsaturated fatty acids plays a crucial
role in many oxidative processes and is responsible for food
deterioration and numerous diseases, including Alzheimerꢀs
disease, atherosclerosis, and cancer rheumatoid arthritis.[1] It
is generally accepted that a key step in lipid peroxidation
(type I mechanism) is hydrogen abstraction of the allylic hy-
drogen atoms from fatty acids by reactive free radicals, such
as hydroxyl, alkyl, alkoxyl, peroxyl radicals, and carbonyl
[a] Dr. I. Andreu, E. Rico, Dr. A. Samadi, Dr. I. M. Morera,
Prof. Dr. M. A. Miranda
Departamento de Quꢁmica-Instituto de Tecnologꢁa Quꢁmica UPV-
CSIC
Universidad Politꢂcnica de Valencia, Camino de Vera s/n
46022, Valencia (Spain)
Fax : (+34)96-387-9349
E-mail: mmiranda@qim.upv.es
[b] Dr. D. Neshchadin, M. Griesser, Prof. Dr. G. Gescheidt
Institute of Physical and Theoretical Chemistry
Technical University Graz, Technikerstrasse 4/I, 8010 Graz (Austria)
E-mail: g.gescheidt-demner@tugraz.at
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100983.
Chem. Eur. J. 2011, 17, 10089 – 10096
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10089

has been found in the intramolecular hydrogen-abstraction
process, as revealed by the triplet lifetimes.[7]
In addition to LFP and photoproduct studies, chemically
induced dynamic nuclear polarization (CIDNP) is a power-
ful mechanistic tool for the study of free-radical reactions.[8]
Thus, CIDNP allows reactions involving radical/radical ion
pairs, which initially reside in a spin-correlated state, to be
followed selectively. To form a product from the two free
radicals, their electron spins have to form an overall singlet.
Since the nuclear spin states, observable by NMR spectros-
copy, are coupled to the spin states of the electron, electron-
spin alterations can be detected by NMR spectroscopy. The
consequence of the spin-selected reaction pathways is a
non-Boltzmann population of spin states within “cage” and
“escape” products, resulting in enhanced absorption (EA,
“positive peaks”) or emission (EE, “negative peaks”) in the
NMR spectrum.[9] Accordingly, CIDNP allows very short-
lived intermediates to be detected at very low concentra-
tions due to large enhancement factors.
The intensity of such polarized NMR signals of the prod-
ucts mirrors, for example, the spin distribution in the inter-
mediate radicals. Thus, CIDNP provides information about
initially formed radicals (with lifetimes of a few ns) and
products formed selectively by radical pathways.
With this background, the aim of the present work was to
perform a thorough mechanistic study on the reaction be-
tween BZP and a series of 1,4-dienes, including CHD, 1,4-
dihydro-2-methylbenzoic acid (MBA), 1,4-dihydro-1,2-dime-
thylbenzoic acid (DMBA), and LA. All of these dienes con-
tain double allylic hydrogen atoms and may be suitable to
provide conclusive evidence for the generation and behavior
of caged radical pairs. For this purpose, a combination of
steady-state photolysis, LFP, and photo-CIDNP appeared to
be ideal.[10]
Figure 1. Photoreactivity of BZP in the presence of different dienes;
a) UV spectra of a solution of BZP (1.0ꢄ10À4 m) and CHD (1.5ꢄ10À3 m)
in acetonitrile after increasing irradiation times (t=0, 1, 3, 5, 10 min)
under argon. b) Decrease in the absorbance at l254 for solutions of BZP
~
&
*
with CHD ( ), MBA ( ), and DMBA ( ) as a function of the irradiation
time.
of BZP with MBA and DMBA, although BZP was clearly
more photoreactive in the presence of CHD (Figure 1b).
The photochemical reaction of BZP with CHD has been
previously studied. Using LFP, Scaiano and Encinas ob-
served quenching of the benzophenone triplet excited state
(3BZP*) by CHD to produce the corresponding radical
pairs.[4a] Furthermore, we found that irradiation of oxygenat-
ed solutions of methanol containing BZP and CHD led to a
complex mixture.[6] Analysis of the mixture by HPLC, GC-
1
MS, and H NMR spectroscopy showed the formation of dif-
ferent photoproducts. They were tentatively identified on
the basis of their MS data; however, isolation and structural
elucidation was not attempted. On the other hand, no re-
ports have been found about the photochemical reaction of
BZP with MBA and DMBA.
Thus, to investigate formation of the photoproducts,
steady-state irradiation of concentrated acetonitrile solu-
tions containing BZP and the different dienes was carried
out under nitrogen, through Pyrex, with a 400 W medium-
pressure mercury lamp.
Results and Discussion
Steady-state photolysis: Irradiations were performed at
lmax =350 nm (Gaussian distribution) in acetonitrile under
an argon atmosphere. They were monitored by UV spectro-
photometry following the disappearance of the typical p–p*
BZP absorption band at 254 nm. As shown in Figure 1a,
BZP reacts with CHD in a few minutes under anaerobic
conditions. Similar spectra were obtained for the irradiations
10090
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10089 – 10096

Probing Lipid Peroxidation
FULL PAPER
The complex photomixture obtained by irradiation of
BZP and CHD was analyzed by GC-MS and then submitted
to silica gel column chromatography. Spectral analysis of the
separated fractions revealed the formation of the cross-cou-
pled sensitizer–diene product 1 (ca. 17%), together with the
reduced (6, ca. 13%) and dimeric (7, ca. 4%) forms of the
sensitizer (Scheme 1). Careful GC-MS analysis of the mix-
Scheme 1. Product formation in the photoreaction of BZP with CHD,
MBA, and DMBA.
Figure 2. Crystal structures of photoproducts 2 (top) and 4 (bottom). El-
ture also revealed the presence of benzene (8) and several
other minor products obtained by dimerization and further
oxidation of the cyclohexadienyl radical. The structure of
compound 1 was unambiguously assigned on the basis of
NMR spectroscopic data (1H, 13C). The structures of 6 and 7
were assigned by comparison with pure compounds.
Likewise, irradiation of BZP in the presence of MBA led
to a complex photomixture that was resolved by reverse-
phase HPLC. The photoproducts arising from cross-coupling
of sensitizer and diene radicals (2, 3) were obtained with
yields of 6.3 and 13%, respectively (Scheme 1). Similar re-
sults were obtained by irradiation of BZP with DMBA; in
this case, compounds 4 (15%) and 5 (28%) were isolated as
the major photoproducts. Furthermore, compound 7 was ob-
tained in both irradiations (17–23%). On the other hand,
compound 6 (23%) and o-toluic acid (9, 30%) were only
isolated from the photomixture obtained from the irradia-
tion of BZP with MBA.
lipsoids are drawn at the 50% probability level.
Transient absorption spectroscopy: To investigate the mech-
anism involved in the photoreactions of BZP with CHD,
MBA, and DMBA, LFP studies were performed on deareat-
ed solutions in acetonitrile, using 355 nm laser excitation
(Nd:YAG). In all cases, the characteristic spectrum of the
ketone triplet excited state (3BZP*; lmax =330 and
525 nm)[2c] was observed immediately after the laser pulse.
This species was clearly quenched by added CHD, MBA, or
DMBA (Figure 3).
The triplet lifetimes were determined in the absence and
in the presence of several quencher concentrations. The rate
constants, kq, were obtained from Stern–Volmer plots
(Figure 3, insets). They were found to be 1.2ꢄ108, 1.1ꢄ108,
and 8.8ꢄ107 mÀ1 sÀ1 for CHD, MBA, and DMBA, respective-
ly; these values are in reasonable agreement with the value
of 2.9ꢄ108 mÀ1 sÀ1 reported for the BZP/CHD system in ben-
zene.[4a]
The structures of photoproducts 2–5 were unambiguously
assigned based on their NMR spectroscopic data (1H, 13C)
and IR spectra. Mass spectrometry supported the assign-
ments. X-ray analysis of crystals obtained from compounds
2 and 4 (Figure 2) clearly supported the cis configuration of
these photoproducts.
On the other hand, transient absorption spectra obtained
upon excitation at 355 nm of a deareated acetonitrile solu-
tion containing BZP (3.5ꢄ10À3 m) and a large excess of
Chem. Eur. J. 2011, 17, 10089 – 10096
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10091

G. Gescheidt, M. A. Miranda et al.
Figure 4. Transient absorption spectra for BZP (3.5ꢄ10À3 m) and CHD
(2.0ꢄ10À1 m) in acetonitrile obtained 0.4 ( ), 1.0 ( ), 4.0 ( ), 10.5 ( ),
~
!
&
*
^
21.5 ( ), 32.2 (3), and 38.3 ms (hexagon) after the laser pulse (355 nm).
[D3]acetonitrile. A Nd:YAG laser at 355 nm and a Hg–Xe
high-pressure lamp were used as the irradiation sources.
Model systems CHD, MBA, and DMBA: Figure 5a shows
1
the H NMR spectrum of the (non-irradiated) mixture (1/1)
1
of CHD/BZP as the reference and the H CIDNP spectra
3
Figure 3. Normalized BZP* decay traces at 525 nm in the presence of in-
creasing amounts (0.25–6 mm) of CHD (a), MBA (b), and DMBA (c)
after laser excitation at 355 nm. Insets: Stern–Volmer plots of 3BZP*
quenching by CHD, MBA, and DMBA.
CHD (2ꢄ10À1 m) are shown in Figure 4. Under these condi-
C
tions, only the ketyl radical (BZPH , lmax =545 nm) was de-
tected, even at short delay times after the laser pulse.
Photo-CIDNP: To bridge the gap between steady-state pho-
tolysis and the time-resolved investigation, we performed
photo-CIDNP experiments with CHD, MBA, DMBA, and
LA using BZP as the hydrogen-abstracting agent in
Figure 5. 1H NMR (a) and CIDNP spectra obtained after photolysis by a
Nd:YAG laser (1 ms) (b) and Hg–Xe high-pressure lamp (300 ms) (c). All
experiments were performed with BZP/CHD in [D3]acetonitrile at room
temperature.
10092
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10089 – 10096

Probing Lipid Peroxidation
FULL PAPER
obtained after its photolysis by a laser and a Hg–Xe lamp
(Figure 5b and c, respectively) representing different time-
scales. The H NMR spectrum (Figure 5a) presents an over-
However, in the present case, we most likely observe a vio-
lation of Kapteinꢀs rules due to the large hfcH(1a,b) and small
1
[12]
C
hfcH(2a,b) in the CHD radical.
lay of the signals from the two parent compounds. The back-
ground-free (Figure 5b, see the Experimental Section) 1H
CIDNP spectrum taken 1 ms after (time delay to the center
of the radio frequency (RF) pulse) the laser pulse (8 ns) re-
veals resonances in enhanced emission and absorption.
After this short delay between irradiation and the observ-
ing RF pulse, predominantly geminate (“cage”) radical pairs
led to the formation of CIDNP effects.
The most prominent polarized signals (enhanced emis-
sion) were observed for H(1) (d=2.64 ppm) and H(2) (d=
5.68 ppm) of CHD, indicating that hydrogen abstraction by
the triplet excited state of BZP from CHD was reversible at
this short timescale. Although H(1) and H(2) are equivalent
in CHD, they do not retain
Apart from the geminate polarizations of the parent
CHD, a number of additional polarized CIDNP signals are
detected. The emissive resonances at d=4.15 and 5.90 ppm,
together with the enhanced absorption at d=5.75 ppm, are
assigned to H(1), H(3), and H(2) of tertiary alcohol 1, which
is the coupling product of the radicals formed by hydrogen
C
C
atom transfer between CHD and BZPH , respectively. The
opposite signs of the polarizations of H(2) and H(3) of 1 are
in agreement with Kapteinꢀs rules and DFT calculations. Po-
larization of CH and OH protons of corresponding 6 are
found at d=5.80 and 3.80 ppm. As the result of hydrogen
C
transfer from CHD to the ketyl radical, we observed a po-
larized signal attributed to 8 at d=7.37 ppm (enhanced
emission). Thus, compounds 6 and 1 are produced in cages
on the timescale of this experiment. Compound 7, which
might be formed by coupling of two ketyl radicals, is silent
C
their equivalence in CHD . In
this case, an average value of
the coupling constants can be
used to predict the sign of
CIDNP polarization. Calcula-
tions (DFT) shown in Figure 6
1
in H CIDNP due to the small hfcs of the phenyl protons in
the ketyl radicals.
Figure 5c shows the H CIDNP spectrum of a CHD/BZP
1
mixture acquired 300 ms after the flash of a Hg–Xe high-
pressure lamp. While indicating an almost identical number
of polarized signals, this spectrum possesses quite a different
polarization pattern from that of Figure 5b. Most of the po-
larizations appear in opposite directions than those of the
1 ms laser-CIDNP spectrum (Figure 5b). This is due to the
much longer delay time (300 ms) between the lamp flash
and the observing RF pulse. On this timescale, signals of
“escape” products dominate. According to Kapteinꢀs rules,
cage and escape products must carry opposite polarizations.
Thus, products that can be formed by both geminate and
indicate that HACTHNUGRTNEUNG(1a,b) and H-
C
ACTHNUGNRTE(NUG 2a,b) in CHD possess aver-
aged isotropic hyperfine cou-
pling constants (hfcs) with op-
Figure 6. Calculated (B3LYP/
TZVP) 1H hfcs (in gauss) of
posite signs, that is, hfcH(1a,b)
=
C
protons in the CHD radical.
(hfcH(1a) ꢄ2+hfcH(1b))/3= +
87.7 G, hfcH(2a,b) =(hfcH(1a) ꢄ2+
hfcH(2b))/4=À2.8 G. According
to Kapteinꢀs rules,[11] this should lead to diverging CIDNP
polarizations after the reversible hydrogen abstraction.
1
Scheme 2. Product formation in the photoreaction of BZP and CHD as determined by H CIDNP.
Chem. Eur. J. 2011, 17, 10089 – 10096
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10093

G. Gescheidt, M. A. Miranda et al.
escape reactions (1, 6, and 8; Scheme 2) appear at inverted
polarizations at the longer 300 ms timescale than in the 1 ms
CIDNP spectra.
The reactivity of MBA and DMBA essentially reflects
that of CHD. Since the reactivity of the escape channel is at-
tenuated by steric factors, a smaller amount of follow-up
products can be established for MBA and DMBA. Com-
pounds 2, 3, 4, and 5 as well as 6 were observed being polar-
ized in CIDNP spectra. The 1H CIDNP spectra recorded
after photoexcitation of mixtures of MBA and DMBA are
presented in the Supporting Information.
Linoleic acid (LA): The principal reaction routes established
for the simple models CHD, MBA, and DMBA are the
basis for gaining insight into the reactivity of LA. Unexpect-
1
edly, H CIDNP experiments with mixtures of BZP and LA,
using the laser, did not yield sufficiently intense signals to
allow clear-cut analysis. However, the application of the
high-pressure Hg–Xe lamp as the light source led to distin-
guishable CIDNP polarizations. It has to be considered that
both geminate and escape polarizations may appear in the
spectrum.[13] Under these conditions, geminate and escape
polarizations may be cancelled by each other to a certain
extent. Moreover, nuclear spin relaxation does not allow ef-
ficient saturation of the background NMR spectroscopy sig-
nals after 300 ms, thus complicating analysis of the spec-
trum.
To avoid misinterpretations, we concentrate on new, addi-
tionally emerging signals. Figure 7 shows 1H NMR and
CIDNP spectra of LA in the presence of BZP. Theoretical
(DFT) calculations predict that predominant spin popula-
C
tion in the linoleyl radical LA resides at the penta-1,4-diene
moiety mirrored by the resonances of H(1)–H(5) in the
CIDNP spectra (Scheme 3). Therefore, products formed via
Figure 7. 1H NMR (a) and CIDNP spectra obtained after photolysis by a
Hg–Xe high-pressure lamp (b). All experiments were performed with
BZP/CHD in [D3]acetonitrile at room temperature.
C
LA in the CIDNP spectrum must inherit polarizations
1
Scheme 3. Product formation in the photoreaction of BZP and LA as determined by H CIDNP.
10094
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10089 – 10096

Probing Lipid Peroxidation
FULL PAPER
based on H(1)–H(5). Indeed, the multiplet at d=6.31 ppm,
appearing in enhanced absorption, can be straightforwardly
assigned to the vinylic H(3) protons of both isomers of con-
jugated derivatives of LA (CLA1, CLA2). Additionally, po-
larizations of the protons H(2) and H(4) of CLA1 of CLA2,
respectively, appear at d=5.82 ppm. Those conjugated acids
are formed by hydrogen abstraction of one of the double al-
lylic hydrogen atoms H(3) of LA by the triplet excited state
of BZP. Reverse hydrogen transfer, that is, hydrogen trans-
Experimental Section
General: BZP, CHD, and MBA were commercially available, whereas
DMBA was synthesized through Birch reduction and subsequent methyl-
ation of 9. The 1H and 13C NMR spectra were recorded at 300 and
75 MHz, respectively, on a Bruker AC-300 spectrometer in CDCl3; chem-
ical shifts are reported in ppm downfield from an internal solvent signal.
GC-MS analysis was performed on an Agilent mass selective detector
coupled to an HP GC system. High-resolution mass spectra (EI, FAB)
were recorded on a VG Auto Spec Fisons spectrometer instrument. All
reactions were monitored by analytical TLC with silica gel 60 F254
(Merck), revealed with cerium ammonium sulfate/ammonium molybdate
reagent. Column chromatogaphy was performed on silica gel 60 (0.063–
0.2 mm) and HPLC was carried out by using a C-18, Kromasil column.
C
C
fer from the ketyl radical (H(1) of BZPH ) to LA leads to
the formation of a conjugated diene system. This is in agree-
ment with DFT calculations and EPR spectroscopy experi-
ments.[14] The background CIDNP spectrum indicates strong
polarization of H(1)–H(5) of parent LA, which points to the
reversibility of hydrogen abstraction: H(1) of the ketyl radi-
cal is transferred back to regenerate parent LA as well as its
trans isomer. The signal corresponding to trans-LA is found
at d=5.25 ppm (Figure 7b). The triplet at d=4.51 ppm (J=
9.9 Hz) is attributed to the tertiary proton of LA–BZP, pro-
General irradiation procedure
Procedure A: Solutions of CHD, MBA or DMBA (1.5ꢄ10À3 m) and BZP
(1.0ꢄ10À4 m) in acetonitrile were irradiated in a multilamp photoreactor
at lmax =350 nm (Gaussian distribution) under an argon atmosphere.
They were monitored by UV spectrophotometry, following the disappear-
ance of the absorption band at 254 nm
Procedure B: Deaerated solutions of CHD, MBA or DMBA (1.3ꢄ
10À1 m) and BZP (3.3ꢄ10À2 m) in acetonitrile were irradiated through
Pyrex with a 400 W medium-pressure mercury lamp. Reactions were
monitored by TLC and NMR spectroscopy. After 10 h, the reaction mix-
tures were concentrated under reduced pressure and submitted to chro-
matography. The CHD and BZP reaction mixture was purified by silica
gel column chromatography, using hexane/ethyl acetate (90:10 v/v) as the
eluent. The reaction mixtures obtained by irradiation of MBA or DMBA
and BZP were separated by reverse-phase HPLC chromatography, using
acetonitrile/water/acetic acid (70:29.7:0.3 v/v/v) as the eluent.
C
duced by coupling of primary ketyl and LA radicals. The
mechanism of the photoreaction of LA with BZP is present-
ed in Scheme 3.
LA also undergoes partly reversible hydrogen abstraction
during photoreaction with BZP. Follow-up hydrogen transfer
between primary radicals leads to the formation of conjugat-
ed derivatives of LA. On the other side, reverse hydrogen
transition can direct cis–trans isomerization of the parent
Laser flash photolysis measurements: A pulsed Nd:YAG laser (SL404G-
10 Spectrum Laser Systems) was used for the excitation at 355 nm. The
single pulses were about 10 ns in duration and the energy was 16 mJ per
pulse. The laser flash photolysis apparatus consisted of the pulsed laser, a
Xe lamp, a monochromator, and a photomultiplier made up of a tube,
housing, and power supply. The output signal from the oscilloscope was
transferred to a personal computer for data analysis. All experiments
were carried out at room temperature. The sensitizer (BZP) dissolved in
acetonitrile (3.5 mm) had an absorbance of about 0.3 at 355 nm. Solutions
were deaerated by bubbling nitrogen through the solution. The rate con-
stants of triplet excited-state quenching by CHD, MBA, and DMBA
were determined by the Stern–Volmer equation (1/t=1/t0 +kq-
C
LA. The unpaired electron spin population in LA is pre-
dominantly located on protons H(1)–H(5).
Conclusion
The photoreaction between BZP and model systems CHD,
MBA, and DMBA proceeds by hydrogen abstraction of the
allylic hydrogen by the triplet excited state of BZP. In neat
acetonitrile, all three model compounds show similar reac-
AHCTUNGTRENNUNG
3
tivity towards BZP*, the lifetimes of which in the presence
1
Photo-CIDNP experiments: H NMR and CIDNP spectra were recorded
on a 200 MHz Bruker AVANCE DPX spectrometer. Irradiation was car-
ried out by using a frequency-tripled Spectra Physics Nd:YAG INDI
laser (355 nm, ca. 40 mJ per pulse, ca. 10 ns) and a Hamamatsu (Japan)
Hg–Xe lamp (SP4, L8252 lamp, 150 W, 300 ms). The following pulse se-
quence was used: presaturation laser/lamp flash–308–RF–detection pulse
(2.2 ms)–free induction decay. The concentrations of BZP, CHD, MBA,
DMBA, and LA were 0.01m. The hyperfine coupling constants (hfc) of
the free radicals were calculated by using the Gaussian 03 package.[15] All
calculations (geometry optimizations and single-point calculations) were
conducted at the B3LYP[16] level of theory with the basis set TZVP.[17]
of dienes were found to be in good agreement with previous
studies on CHD in benzene.[4a]
Partly reversible hydrogen abstraction of the allylic hydro-
gen atoms of CHD, MBA, and DMBA was also detected by
photo-CIDNP on different timescales. CIDNP polarizations
of the diamagnetic products were in a full agreement with
the results derived by LFP.
Similar processes have been found for LA, yet LA-related
radicals were only reported when generated by hydrogen
transfer from highly substituted model compounds, allowing
their observation by steady-state EPR spectroscopy.[14]
Herein, we have experimentally established the formation
Compound 1: 1H NMR (300 MHz; CDCl3, TMS): d=7.6–7.1 (m, 10H),
5.9 (m, 2H), 5.5 (m, 2H), 4.1 (m, 1H), 2.7 ppm (m, 2H); 13C NMR
(75 MHz; CDCl3, TMS): d=146.1, 133.0–125.0, 79.5, 44.4, 27.2 ppm;
HRMS: m/z calcd for C19H19O: 263.14359 [M H]+; found: 263.14379.
Compound 2: 1H NMR (300 MHz; CDCl3, TMS): d=7.6–7.1 (m, 10H),
5.8 (m, 1H), 5.6 (m, 1H), 5.3 (m, 1H), 4.1 (m, 1H), 3.6 (m, 1H), 1.6 ppm
(s, 3H); 13C NMR (75 MHz; CDCl3, TMS): d=175.2, 144.7, 144.3, 127.3,
127.3, 127.0, 125.7, 125.6, 124.6, 124.5, 121.5, 78.5, 47.7, 44.4, 21.2 ppm;
HRMS: m/z calcd for C21H18O2: 302.13068 [MÀ18]+; found: 302.12991.
C
C
of LA and shown that LA converts into two dominating
conjugated isomers on the ms timescale. Such processes are
the basis of alterations of membrane structures caused by
oxidative stress.
1
Compound 3: H NMR (300 MHz; CDCl3): d=7.6–7.1 (m, 10H), 5.9 (m,
1H), 5.7 (m, 1H), 5.5 (m, 1H), 4.1 (m, 1H), 3.6 (m, 1H), 1.7 ppm (s,
3H); 13C NMR (75 MHz; CDCl3): d=177.4, 145.4, 145.3, 133.1, 128.4,
Chem. Eur. J. 2011, 17, 10089 – 10096
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10095

G. Gescheidt, M. A. Miranda et al.
127.8, 126.8, 126.7, 125.7, 125.5, 125.4, 122.7, 79.5, 47.4, 44.7, 22.2 ppm;
[6] A. Samadi, L. A. Martꢁnez, M. A. Miranda, I. M. Morera, Photo-
chem. Photobiol. 2001, 73, 359–365.
HRMS: m/z calcd for C21H20O3: 320.14124 [M]+; found: 320.14007.
Compound 4: M.p. 186–1888C (from CH3CN); 1H NMR (300 MHz;
CDCl3): d=7.6–7.2 (m, 10H), 5.7 (m, 1H), 5.6 (m, 1H), 5.4 (m, 1H), 4.1
(m, 1H), 1.7 (s, 3H), 1.4 ppm (s, 3H); 13C NMR (75 MHz; CDCl3): d=
177.6, 145.6, 145.3, 137.8, 132.4, 128.4, 128.4, 126.7, 126.7, 126.0, 125.6,
125.5, 122.0, 80.2, 48.7, 45.1, 22.9, 19.9 ppm; HRMS: m/z calcd for
C22H23O3: 335.16472 [M+H]+; found: 335.16509.
[7] a) M. A. Miranda, L. A. Martꢁnez, A. Samadi, F. Boscꢅ, I. M.
Morera, Chem. Commun. 2002, 280–281; b) F. Boscꢅ, I. Andreu,
I. M. Morera, A. Samadi, M. A. Miranda, Chem. Commun. 2003,
1592.
[8] a) M. Goez, I. Frisch, J. Phys. Chem. A 2002, 106, 8079–8084; b) M.
Goez, G. Eckert, Phys. Chem. Chem. Phys. 2006, 8, 5294; c) H. D.
Roth, Photochem. Photobiol. Sci. 2008, 7, 540–546; d) D. Neshcha-
din, R. Levinn, G. Gescheidt, S. N. Batchelor, Chem.Eur. J. 2010, 16,
7008–7016.
[9] a) G. L. Closs, Adv. Magn. Reson. 1974, 7, 157–229; b) K. M. M. Sal-
ikhov, Yu. N. Sagdeev, R. Z. Buchachenko, Studies in Physical and
Theoretical Chemistry, Vol. 22: Spin Polarization and Magnetic Ef-
fects in Radical Reactions, Elsevier, Amsterdam, 1984, p. 419.
[10] M. Goez, Concepts Magn. Reson. 1995, 7, 69–86.
1
Compound 5: H NMR (300 MHz; CDCl3): d=7.6–7.2 (m, 10H), 5.7 (m,
1H), 5.6 (m, 1H), 5.4 (m, 1H), 4.2 (m, 1H), 1.7 (s, 3H), 1.3 ppm (s, 3H);
13C NMR (75 MHz; CDCl3): d=179.4, 145.4, 145.2, 137.7, 132.5, 128.3,
126.8, 126.7, 125.7, 125.6, 125.5, 121.8, 79.2, 48.0, 45.0, 23.9, 20.0 ppm;
HRMS: m/z calcd for C22H22O3: 334.15689 [M]+; found: 334.15627.
X-ray crystallography: X-ray structures were measured on a Bruker
Appe-II CCD instrument. CCDC-819279 (2) and 819278 (4) contain the
supplementary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
[11] a) G. L. Closs, J. Am. Chem. Soc. 1969, 91, 4552–4554; b) R. Kap-
tein, Chem. Phys. Lett. 1969, 4, 214–216.
[12] K. Salikhov, Chem. Phys. 1982, 64, 371–379.
[13] M. Goez, Chem. Phys. Lett. 1992, 188, 451–456.
[14] H. Kitaguchi, K. Ohkubo, S. Ogo, S. Fukuzumi, J. Am. Chem. Soc.
2005, 127, 6605–6609.
[15] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
Acknowledgements
Financial support from the MICINN (grants CTQ2009-13699 and
CTQ2010-14882), from the Generalitat Valenciana (GV/2009/104) and
from Carlos III Institute of Health (grant RIRAAF, RETICS program)
is gratefully acknowledged. We thank also COST project CM603 for fa-
cilitating our collaboration.
[1] a) A. W. Girotti, Photochem. Photobiol. 1990, 51, 497–509; b) A.
Sevanian, F. Ursini, Free Radical Biol. Med. 2000, 29, 306–311;
c) T. J. Montine, M. D. Neely, J. F. Quinn, M. F. Beal, W. R. Mar-
kesbery, L. J. Roberts, J. D. Morrow, Free Radical Biol. Med. 2002,
33, 620–626; d) F. Shahidi, Y. Zhong, Chem. Soc. Rev. 2010, 39,
4067–4079.
[2] a) N. A. Porter, Methods Enzymol. 1984, 105, 273–282; b) N. A.
Porter, Acc. Chem. Res. 1986, 19, 262–268; c) F. Bosca, M. Miranda,
J. Photochem. Photobiol. B 1998, 43, 1–26.
[3] A. C. Velosa, W. J. Baader, C. V. Stevani, C. M. Mano, E. J. H. Be-
chara, Chem. Res. Toxicol. 2007, 20, 1162–1169.
[4] a) M. V. Encinas, J. C. Scaiano, J. Am. Chem. Soc. 1981, 103, 6393–
6397; b) W. M. Nau, F. L. Cozens, J. C. Scaiano, J. Am. Chem. Soc.
1996, 118, 2275–2282; c) W. Adam, J. N. Moorthy, W. M. Nau, J. C.
Scaiano, J. Org. Chem. 1997, 62, 8082–8090.
[16] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) P. J. Stephens, F. J.
Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98,
11623–11627.
[17] A. Schꢆfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829.
[5] F. Boscꢅ, M. A. Miranda, I. M. Morera, A. Samadi, J. Photochem.
Photobiol. B 2000, 58, 1–5.
Received: March 30, 2011
Published online: July 26, 2011
10096
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10089 – 10096