UVA irradiation of fatty acids and their oxidized products substantially increases their ability to generate singlet oxygen

Article abstract of DOI:10.1039/c3cp51399h

UVA radiation plays an important role for adverse reactions in human tissue. UVA penetrates epidermis and dermis of skin being absorbed by various biomolecules, especially endogenous photosensitizers. This may generate deleterious singlet oxygen (¹O₂) that oxidizes fatty acids in cell membranes, lipoproteins, and other lipid-containing structures such as the epidermal barrier. Indications exist that fatty acids are not only the target of ¹O₂ but also act as potential photosensitizers under UVA irradiation, if already oxidized. Five different fatty acids in ethanol solution (stearic, oleic, linoleic, linolenic and arachidonic acid) were exposed to UVA radiation (355 nm, 100 mW) for 30 seconds. ¹O ₂ luminescence was detected time-resolved at 1270 nm and confirmed in spectrally-resolved experiments. The more double bonds fatty acids have the more ¹O₂ photons were detected. In addition, fatty acids were continuously exposed to broadband UVA for up to 240 min. During that time span, UVA absorption and ¹O₂ luminescence substantially increased with irradiation time, reached a maximum and decreased again. HPLC-MS analysis showed that the amount of peroxidized fatty acids and the ¹O₂ generation increased and decreased in parallel. This indicates the high potential of peroxidized fatty acids to produce ¹O₂ under UVA irradiation. In conclusion, fatty acids along with peroxidized products are weak endogenous photosensitizers but become strong photosensitizers under continuous UVA irradiation. Since fatty acids and their oxidized products are ubiquitous in living cells and in skin, which is frequently and long-lasting exposed to UVA radiation, this photosensitizing effect may contribute to initiation of deleterious photooxidative processes in tissue.

Full text of DOI:10.1039/c3cp51399h

PCCP
View Article Online
View Journal | View Issue
PAPER
UVA irradiation of fatty acids and their oxidized
products substantially increases their ability to
generate singlet oxygen
Cite this: Phys. Chem. Chem. Phys., 2013,
15, 17672
Johannes Regensburger,* Tim Maisch, Alena Knak, Anita Gollmer,
Ariane Felgentraeger, Karin Lehner and Wolfgang Baeumler
UVA radiation plays an important role for adverse reactions in human tissue. UVA penetrates epidermis and
dermis of skin being absorbed by various biomolecules, especially endogenous photosensitizers. This may
generate deleterious singlet oxygen (1O2) that oxidizes fatty acids in cell membranes, lipoproteins, and
other lipid-containing structures such as the epidermal barrier. Indications exist that fatty acids are not only
the target of 1O2 but also act as potential photosensitizers under UVA irradiation, if already oxidized. Five
different fatty acids in ethanol solution (stearic, oleic, linoleic, linolenic and arachidonic acid) were exposed
to UVA radiation (355 nm, 100 mW) for 30 seconds. 1O2 luminescence was detected time-resolved at
1270 nm and confirmed in spectrally-resolved experiments. The more double bonds fatty acids have the
more 1O2 photons were detected. In addition, fatty acids were continuously exposed to broadband UVA for
up to 240 min. During that time span, UVA absorption and 1O2 luminescence substantially increased with
irradiation time, reached a maximum and decreased again. HPLC-MS analysis showed that the amount of
peroxidized fatty acids and the 1O2 generation increased and decreased in parallel. This indicates the high
potential of peroxidized fatty acids to produce 1O2 under UVA irradiation. In conclusion, fatty acids along
with peroxidized products are weak endogenous photosensitizers but become strong photosensitizers
under continuous UVA irradiation. Since fatty acids and their oxidized products are ubiquitous in living cells
and in skin, which is frequently and long-lasting exposed to UVA radiation, this photosensitizing effect may
contribute to initiation of deleterious photooxidative processes in tissue.
Received 3rd April 2013,
Accepted 28th August 2013
DOI: 10.1039/c3cp51399h
www.rsc.org/pccp
Charge transfer may generate other reactive oxygen species
such as radicals. All these species are frequently responsible
Introduction
Solar radiation on earth comprises UVA (B95%) and UVB (B5%)
radiation. UV radiation can induce DNA damage, oxidative stress
or immune-suppression.1 UV radiation affects cellular and tissue
integrity2–10 and is a major cause of human skin cancer.11,12
Furthermore, statistical investigations in USA provide evidence
that skin cancer is the most frequent diagnosed cancer in humans
(40% of all cancer diseases).13 Due to a change of lifestyle and an
increased expectation of life, exposure to UVA-radiation should
even have a higher impact on skin damage.14
UVA-radiation (320–400 nm) deeply penetrates skin and can
be absorbed by various endogenous molecules. In case that
such an endogenous molecule represents a photosensitizer,
energy or charge can be transferred via an excited triplet state to
oxygen. Energy transfer may excite an oxygen molecule in its
triplet ground state to its first singlet state (singlet oxygen).
for deleterious cellular effects, in which singlet oxygen plays
a major role.15–17
Singlet oxygen is intentionally produced in photodynamic
therapy to kill cancer cells via photooxidation of cellular con-
stituents. In the case of UVA radiation, it has already been
shown that keratinocytes or living skin can generate singlet
oxygen in skin without an additional exogenous photo-
sensitizers.18 In that case, photosensitizers are endogenous cell
components like flavines, lipids or oxidized fatty acids.19,20
Lipids and fatty acids are ubiquitous inside skin cells and also
outside cells in the epidermis.
We already detected the luminescence of singlet oxygen, which
was generated by different fatty acids and lipids in solution.20 The
signal intensities from these molecules are usually low and
it remains questionable whether and to what extent singlet
oxygen generation by fatty acids and lipids might contribute to
deleterious effects of UVA on skin. A reason for the small signal
was the low absorption of UVA radiation in fatty acids or lipids.
However, the structure of such endogenous photosensitizers
Department of Dermatology, University of Regensburg, 93042 Regensburg, Germany.
E-mail: johannes.regensburger@klinik.uni-regensburg.de; Fax: +49-941-944-8943;
Tel: +49-941-944-8944
c
17672 Phys. Chem. Chem. Phys., 2013, 15, 17672--17680
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
may change under UVA irradiation, which may in turn change 1400 nm (4000 photons). The irradiation of fresh solutions of
the UV absorption and hence singlet oxygen production. linoleic (29 000 photons) and arachidonic (46 000 photons)
In the present investigation, we exposed different fatty acids acids yielded a clear luminescence signal at 1270 nm. In spite
(stearic, oleic, linoleic, linolenic and arachidonic acid), which of the act a fresh linolenic acid solution was used, a total number
are main constituents of cell lipids,21,22 to UVA radiation, the of 126000 luminescence photons (about 3 times higher than the
change in absorption was measured and singlet oxygen was number of arachidonic acid) were detected, which might be
detected by time- and spectrally resolved luminescence at caused by a higher content of already oxidized linolenic acid in
regular time intervals. The irradiated samples were additionally the fresh solution.
investigated by HPLC, because UVA radiation can induce lipid
Singlet oxygen generation in fatty acid solutions with UVA
peroxidation,23–25 even in the skin of mice26 or cause mutations pre-irradiation. Since all fatty acid solutions showed a clear
and DNA damage.27
singlet oxygen signal under UVA irradiation, the generated
singlet oxygen could in turn peroxide the unoxidized fatty acids
and hence could change their molecular constitution.28 Con-
sequently, the absorption of fatty acid solutions could be
changed and oxygen could be consumed. Therefore, the differ-
ent fatty acid solutions were irradiated with the broadband
UVA-lamp for long time irradiation.
Changes in absorption spectra. The absorption spectra of
the five fatty acid solutions after different irradiation times
were recorded. Changes in the absorption spectra depended on
the UVA irradiation time and are shown only for the excitation
wavelength at 355 nm in Fig. 1. The maximal change in
absorption was detected for arachidonic acid solution upon
an irradiation time of 75 minutes, whereas the absorption of
arachidonic acid solution increased up to 87%. During irradiation,
the oxygen concentration was kept constant by bubbling the
solution with oxygen because the rate and the mechanism of such
reactions depend on oxygen concentration. Due to the observed
singlet oxygen generation and oxygen consumption during irradia-
tion, the latter is discussed in a following paragraph, a change in
the chemical molecule structure was supposed.
Results
Singlet oxygen generation in fatty acid solutions without UVA
pre-irradiation
Absorption at 355 nm. Laser radiation at 355 nm was used to
excite the investigated molecules. All fatty acids (50 mmol Lꢀ1
)
absorb partially UV radiation in the range from 200 to 250 nm
(data not shown). Stearic acid solution without double bonds
and oleic acid solution with only one double bond showed fairly
low absorption at 355 nm, and less than 0.5% of irradiation was
absorbed at a concentration of 50 mmol Lꢀ1 for each fatty
acids. For the other three fatty acid solutions the absorption
ranged up to 5% (data not shown).
Singlet oxygen luminescence. The luminescence signals of
the fatty acid solutions were measured time resolved at
1270 nm as well as spectrally resolved (1150–1400 nm). The
signals were recorded in the range from 1150 nm to 1400 nm
and all luminescence photons after the excitation laser pulse
and noise (120 ms) were summed up. Thus, a clear maximum of
luminescence photons between 1270 and 1280 nm was detected for
all fatty acids (data not shown). For singlet oxygen the transition to
the ground state in ethanol is centered at 1274 nm, this corre-
sponds to an energy of 0.975 eV.
Changes in luminescence signals at 1270 nm. The lumines-
cence signals (at 1270 nm) of the five fatty acid solutions were
detected at regular time intervals during UVA irradiation.
The integrated singlet oxygen luminescence changed during
irradiation with UVA, in particular for linolenic (3 double bonds)
and arachidonic acid (4 double bonds) solution (Fig. 2A). The
total count of luminescence photons increased for linoleic,
linolenic and arachidonic acid solutions. For longer irradiation
For stearic and oleic acid solution only a weak luminescence
signal at 1270 nm was measured with a total photon number
of 7700 and 11 000 photons, respectively. These values are
clearly higher than the measured luminescence at 1150 and
Fig. 1 The absorption value for all tested fatty acids (50 mmol Lꢀ1) at 355 nm after different irradiation times with 1 W cmꢀ2 UVA radiation (red: stearic acid; green:
oleic acid; dark blue: linoleic acid; purple: linolenic acid; cyan: arachidonic acid).
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 17672--17680 17673

View Article Online
PCCP
Paper
Fig. 2 The integrated signals of singlet oxygen luminescence at 1270 nm for all five fatty acids after different irradiation times with UVA radiation (A) (red: stearic
acid; green: oleic acid; dark blue: linoleic acid; purple: linolenic acid; cyan: arachidonic acid; the values of stearic and oleic acid are magnified by a factor 10).
Luminescence signals of 50 mmol Lꢀ1 arachidonic acid after 0 (red) and 50 (black) minutes continuous broadband UVA irradiation (B). Solid line is an exponential fit
with Mathematica 5.2. The singlet oxygen decay time is in the range of 14 ꢁ 2 ms, the amount of photons increases with more than a factor of 12.
times the count of luminescence photons decreased. For stearic (46 000 vs. 580 000 photons). Table 1 shows the luminescence
and oleic acid solution the luminescence integral is pronounced counts of 1O2 photons detected at the beginning of UV irradiation
for short irradiation times and decreases after 5 minutes of for all fatty acid solutions, at the time of the maximal lumines-
UVA-irradiation (Fig. 2A).
cence intensity and at the end of UV irradiation. The results show
Fig. 2B exemplarily shows the luminescence signal of that the more double bonds the fatty acids have the more singlet
50 mmol Lꢀ1 arachidonic acid solution before broadband irradia- oxygen luminescence is detected. The maximum values of
tion and at the maximal luminescence after 50 minutes (black) luminescence increase from stearic acid solution (8600 counts), oleic
UVA irradiation (1 W cmꢀ2). The decay time slightly decreased but acid solution (11 600 counts), linoleic acid solution (152 000 counts),
was still in the range of 14 ꢁ 2 ms. The total number of detected linolenic acid solution (517 000 counts), to arachidonic acid solution
luminescence photons increased by a factor of more than 12 (583 000 counts).
c
17674 Phys. Chem. Chem. Phys., 2013, 15, 17672--17680
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
Table 1 Singlet oxygen luminescence while longtime irradiation
irradiation. The rate constants of the oxygen depletion of the five
fatty acid solutions are shown in Table 2.
Luminescence photonsa
HPLC-MS measurement. Liquid chromatography analysis
with mass spectroscopy detection was performed to confirm the
observed results. The results showed that the molecular weight of
fatty acids (except for stearic acid) increased by 32 Dalton that
clearly indicates a binding of oxygen molecule to the fatty acids
(hydroperoxides). Furthermore, dehydrated molecules could be
detected (loss of H2O).
Maximum of
Beginning of luminescence and End of UV
UV irradiation time of maximum irradiation (time)
Fatty acid
Stearic acid
Oleic acid
Linoleic acid
Linolenic acid
7700
11 000
29 000
126 000
8600 (5 min)
11 600 (5 min)
152 000 (105 min) 94 500 (240 min)
517 000 (60 min) 114 000 (240 min)
583 000 (50 min) 137 000 (240 min)
5800 (120 min)
5100 (120 min)
Arachidonic acid 46 000
For example, HPLC showed that the amount of arachidonic acid
decreased with irradiation time (Fig. 6, red part). In contrast, the
amount of hydroperoxides increased during the first 60 minutes and
decreased afterwards (Fig. 6, blue part). HPLC-MS measurement
a
All detected photons within the measurement time span of 1O2
luminescence (120 ms).
Changes in the spectrally and time resolved luminescence showed that nearly no arachidonic acid or respective hydroperoxides
signals. To clearly assign the detected signal at 1270 nm to were present after 240 min of irradiation time, but some other
singlet oxygen luminescence during the long time of UV irradiation, oxidized products (Fig. 6, green part).
the luminescence signal was measured spectrally resolved in parallel
to time resolution. Both results were combined to a three dimen-
sional plot that clearly shows the increase and decrease of singlet
Discussion
oxygen luminescence over time. For arachidonic acid solution a clear At first view, the absorption of radiation by fatty acids should be
maximum at 1270 nm after 50 minutes is noticeable (Fig. 3A). The very low for wavelengths longer than 220 nm.30 However, under
noise level at 1150 and 1400 nm is nearly constant over time. For atmospheric conditions, autooxidation of molecules can occur
oleic acid solution a clear maximum at 5 minutes UVA irradiation is that cannot be avoided in the experimental setting.31,32 Auto-
identifiable, after that the luminescence decreases to a noise level oxidized products of fatty acids and impurities may absorb
after 120 minutes of irradiation (Fig. 3B). Stearic acid solution radiation in the ultraviolet spectral range up to 400 nm.29 The
showed a similar course of luminescence as oleic acid solution, results of luminescence detection clearly showed that UVA
whereas the luminescence intensity was small for all irradiation along with oxidized fatty acids generate singlet oxygen and
times (data not shown).
The maximum of luminescence for linoleic acid or linolenic acid increasing double bonds of fatty acids solution.33
solution was detected after 105 minutes or 60 minutes of irradiation, Skin is continuously targeted by UVA radiation, mainly due
the concentration of produced singlet oxygen increased with
respectively, and for both, the signals decreased for irradiation times to solar radiation. In Germany, the annual terrestrial radiant
up to 120 minutes. As shown in Fig. 1 and 2A both the absorption exposure of UVA is about 22 000 J cmꢀ2 which is equivalent to
and the luminescence increased to a maximum value at nearly the 60 J cmꢀ2 daily on average.34 A major constituent of the skin
same time of broadband irradiation. This is consistent to the fact and its cells are various lipids containing the different fatty
that absorption of photons is necessary to generate singlet oxygen. acids. In addition to living cells in the epidermis and dermis,
Based on the theory of singlet oxygen luminescence, that the total the stratum corneum contains a mixture of ceramides and free
amount of luminescence clearly correlates with the concentration of fatty acids that forms the permeability barrier.35 Thus, fatty
singlet oxygen [1O2].29 Therefore, a correlation between [1O2] and acids are a target of continuous UVA of solar radiation, which
UVA absorption was made to prove whether the generation of singlet penetrates the skin down to the mid of the dermis at least.
oxygen was caused by UVA absorption in fatty acid solutions (Fig. 4).
Consequently, the five fatty acid solutions were exposed
The slope in Fig. 4 shows, within experimental accuracy, a good to continuous, broadband UVA radiation to investigate the impact
correlation of the increase of [1O2] with increase of UVA absorption of singlet oxygen generation. The results revealed an important
in fatty acids. The decrease of [1O2] also correlates with UVA process. The application of the continuous UVA radiation increased
absorption but shows also a deviation for long UVA exposure the absorption of fatty acid solutions including the wavelength
times, indicating that the luminescence signal decreased faster 355 nm (Fig. 1). This wavelength is used in the experimental setup
than the absorption. This leads to the suggestion that molecules for luminescence detection of singlet oxygen. The change in absorp-
in the solution appear which absorb UVA but fail to generate tion was fairly small for stearic and oleic acid, but substantially
singlet oxygen.
increased with irradiation time for fatty acid solutions with two or
Oxygen consumption. To clarify the reason for the change in more double bonds. The absorption of linolenic or arachidonic
absorption the oxygen concentration in the fatty acid solutions acid solution reached values of about 20% within 5 min, which
was measured during broadband irradiation without oxygen is equivalent to a radiant exposure of 300 J cmꢀ2, the fivefold
bubbling. When starting with 100% oxygen saturation in the daily solar UVA radiation in Germany.34
cuvette ([O2] = 1850 mmol Lꢀ1 for ethanol), oxygen concentration
Along with the increase of radiation absorption, the generation
clearly decreased with irradiation time and the decrease was of singlet oxygen also substantially increased with irradiation time
correlated to the number of double bonds (Fig. 5). The more for linoleic, linolenic and arachidonic acid solution (Fig. 2). Singlet
double bonds the faster was the oxygen decrease under UVA oxygen was continuously produced by UVA radiation up to 240 min
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 17672--17680 17675

View Article Online
PCCP
Paper
Fig. 3 Wavelength- and irradiation time-resolved three dimensional diagrams of arachidonic (A) and oleic acid (B). Values are determined by integration of the
luminescence photons for each crossing point.
(e.g. for arachidonic acid), which was confirmed by spectral be due to the less singlet oxygen production (oleic acid) or to the
resolution (Fig. 3). At the same time, the oxygen concentration lack of double bounds (stearic acid). The consumption in stearic
in solutions of fatty acid decreased, whereas the rate constant of acid solution is most likely caused by impurities of the fatty acids
oxygen consumption correlated to the number of double bonds (purity at about 99%).
(Fig. 5, Table 2). These results led to the assumption that singlet
oxygen, produced by an interaction of UVA and peroxidized fatty reaction of singlet oxygen with unsaturated fatty acids.28,30 It is
acids, can react with other non-oxidized fatty acids. known that those oxidized fatty acids are the typical products being
The oxygen consumption is obviously caused by the well-known
Oxygen consumption occurred even for oleic acid solution and responsible for the so-called photo-oxidative stress signal in living
stearic acid solution but on a much longer time scale. This should cells.26,36 Our investigations with HPLC/MS showed a clear increase
c
17676 Phys. Chem. Chem. Phys., 2013, 15, 17672--17680
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
Fig. 4 Luminescence at 1270 nm vs. absorption at 355 nm for all measured points from 0 minutes up to 240 minutes of broadband UVA irradiation. For increase of
absorption and luminescence a linear correlation could be shown.
Fig. 5 Oxygen concentration in ethanol solutions of the five investigated fatty acids against irradiation time with 1 W cmꢀ2 (red: stearic acid; green: oleic acid; dark
blue: linoleic acid; purple: linolenic acid; cyan: arachidonic acid).
The singlet oxygen peroxidized fatty acids37 increased, which in
Table 2 Oxygen consumption of fatty acids under UV irradiation
turn led to higher UVA absorption and so to a higher amount of
Fatty acid
Oxygen consumption rate (mmol s–1
)
singlet oxygen. The 3D-plot shows that singlet oxygen lumines-
cence decreases over time after having reached a maximum,
which represents a decrease of the generated singlet oxygen due
to further oxidation reactions with hydroperoxides.
To prove a correlation of singlet oxygen production and
absorption of UVA radiation in fatty acid solutions, the total
Stearic acid 18 : 0
Oleic acid 18 : 1
Linoleic acid 18 : 2
Linolenic acid 18 : 3
0.15
0.42
3.48
8.32
9.55
Arachidonic acid 20 : 4
of molecular weight by 32 Dalton (molecular weight of O2) that counts of luminescence were plotted against the UVA absorption at
is exemplarily shown for arachidonic acid solution in Fig. 6. 355 nm for the different times of UVA irradiation (up to 240 min).
With increasing irradiation time, the peak of peroxidized fatty Except for stearic and oleic acid solution (low singlet oxygen
acids in the diagram decreased. Based on this knowledge, the time production), the total count of luminescence was correlated
course of UVA radiation in Fig. 3 should be explained as follows. to UVA absorption for the other fatty acid solutions. Thus, it can
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 17672--17680 17677

View Article Online
PCCP
Paper
Fig. 6 HPLC analysis of arachidonic acid (red part: unoxidized arachidonic acid; blue part: molecules with 32 Da (O2), 16 Da (O) and14 Da (O2–H2O)) more molecular
weight compared to arachidonic acid; green part: molecules with different weight ratios of oxygen content.
be assumed that the absorbing molecules (oxidized fatty acids) are
responsible for the singlet oxygen luminescence. Furthermore for
long exposure times, other oxidized products are generated in fatty
acid solutions, which absorb UVA radiation but without being able
to generate singlet oxygen (Fig. 4). Consequently, UV radiation is
still absorbed but yielding less singlet oxygen.
It should be emphasized that in the case of UVA and fatty
acid solutions, singlet oxygen is most likely not produced via
normal photosensitizing because the typical chemical structures of
photosensitizers (cyclic structures, conjugated double bonds) are
lacking in fatty acids. The underlying mechanisms of singlet oxygen
production in fatty acid solutions should be related to fast chemical
reactions. This is confirmed by the fact that the time-resolved
luminescence signals showed a rise time, which was close to the
time resolution of the detection system and hence substantially
shorter than for any known photosensitizer (>0.2 ms). The decay
time showed values of 14 ms, which is typical for singlet oxygen in
ethanol solutions.38 Miyamoto et al. have already shown that lipid
Fig. 7 Mechanism of singlet oxygen generation by UVA irradiation of lipid
hydroperoxides (LOOH) can generate singlet oxygen and that these hydroperoxides and further oxidation of fatty acids.
can decompose by heat, UV light and by the addition of transition
metals.39 Furthermore Pratt et al. have shown that unstable hydro-
link together to form a tetra-oxide. Consequently, ketones or
aldehydes and singlet oxygen can be produced by cleavage of such
tetra-oxides (Fig. 7). That description of singlet oxygen in those
solutions under UVA exposure is not necessarily complete and may
comprise other radical mechanisms.
peroxides can lead to a propagation of lipid hydroperoxides as long
as unoxidized fatty acids (LH) and oxygen are available.40 With the
assumption that even in fresh solutions oxidized products of fatty
acids are already present, which was confirmed by HPLC data in
this study, the process actually start with a mixture of substances,
even in fatty acids with a high purity. When exposed to UVA
radiation, those molecules, particularly the oxidized ones, will
undergo two different mechanisms. Firstly, these molecules might
decompose to radicals and will react with surrounding molecules,
Experimental
Preparation of solutions
to form new lipid radicals and lipid hydroperoxides. Both can in The following substances were always freshly used in each
turn increase the overall number of lipid hydroperoxides again and single experiments and always dissolved in ethanol. All
again. Secondly, the generation of singlet oxygen is performed by substances were purchased from Sigma-Aldrich (Steinheim,
the so-called Russell mechanism,41 whereby two hydroperoxides Germany) with a purity of about 99%. Concentrations of stearic
c
17678 Phys. Chem. Chem. Phys., 2013, 15, 17672--17680
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
acid (18 : 0), oleic acid (18 : 1), linoleic acid (18 : 2), or linolenic added up from 100 ns after the laser pulse to the end of the
acid (18 : 3) were 50 mmol Lꢀ1 each. The brackets show the ratio luminescence signal.
of the number of carbon atoms to the number of double bonds
UVA irradiation
of the respective fatty acids.
Continuous, broadband UVA irradiation (320–400 nm) was
Transmission spectra
performed using a high pressure mercury lamp (OmniCure
Series 2000, IGB-Tech GmbH, Friedelsheim, Germany) at an
intensity of 1 W cmꢀ2. During irradiation the solutions were
magnetically stirred and saturated with oxygen. The fatty acids
were irradiated maximum for 240 min, which is corresponding
Transmission spectra from 200–400 nm of each probe were
recorded at room temperature with a Beckman DU640 spectro-
photometer (Beckman Instruments GmbH, Munich, Germany)
using a quartz cuvette with an optical path of 1 cm (QS-101,
Hellma Optik, Jena, Germany). The absorption values are
calculated to be A = 100%ꢀT, where T is the transmission value.
to a dose of 14.4 kJ cmꢀ2
.
Liquid chromatography analysis and mass spectroscopy
(HPLC-MS)
Oxygen concentration in solution
The irradiated fatty acid samples were filtered using a PTFE
filter (Chromafil, O-20/15, organic, pore size 0.2 nm, Machery-
Nagel, Du¨ren, Germany). A 100 mL sample was analyzed using a
1100 HPLC (Agilent Technologies, Waldbronn, Germany) fitted
with a C18 analytical column (Phenomenex Luna, particle size
3 mm, 150 ꢂ 2.00 mm, Aschaffenburg, Germany) and a diode
array detector (DAD). The data-files were analysed using a
HPLC-3D-ChemStation Rev. B.03.01. For chromatography a
gradient elution with water [0.0059 wt% trifluoroacetic acid]
(solvent A) and acetonitrile (solvent B) at a constant flow rate of
1.0 mL minꢀ1 was done. A gradient profile with the following
proportions of solvent B was applied [t (min), % B]: (0, 5), (30, 95).
The chromatograms were monitored at 220 nm. Mass spectroscopy
was done using an Agilent 6100 Series Single Quadrupole Superior
Line LC/MSD (Agilent Technologies, Waldbron, Germany) and
an ESI/APCI Multimode Source.
The oxygen concentration in solution was measured in the cuvette
using a needle sensor that was placed in the cuvette (MICROX TX,
PreSens GmbH, Regensburg, Germany). Thereby the cuvette was
completely filled, hermetically closed and magnetically stirred
during irradiation. The oxygen sensor continuously measured
the oxygen concentration during irradiation.
Luminescence experiments
Fatty acids were transferred into a quartz cuvette (QS-101, Hellma
Optik, Jena, Germany). They were excited using a frequency-tripled
Nd:YAG laser (PhotonEnergy, Ottensoos, Germany) with a repeti-
tion rate of 2.0 kHz (wavelength 355 nm, pulse duration 70 ns). The
laser pulse energy for luminescence experiments was 50 mJ. At
regular time intervals, the singlet oxygen luminescence at 1270 nm
was detected with using an IR-sensitive photomultiplier (R5509-42,
Hamamatsu Photonics Deutschland GmbH, Herrsching, Germany)
with a rise time of about 3 ns. Additional details of the setup
were described elsewhere.42 The number of laser pulses for
excitation was 60 000. The luminescence signal was detected at
wavelengths of 1150, 1200, 1250, 1260, 1270 (emission maximum
of singlet oxygen), 1280, 1300, 1350, 1400 nm using appropriate
interference filters in front of the photomultiplier. All luminescence
experiments are done in oxygen saturated solutions guaranteed
by constant oxygen bubbling.
Conclusion
We showed that fatty acids (with isolated double bonds) can
generate singlet oxygen, although these molecules are not normal
photosensitizers with conjugated double bonds or cyclic structures
that offer a defined triplet T1 state. When pre-irradiated with UVA,
the absorption of ultraviolet radiation substantially increased and
in turn the production of singlet oxygen also increased. This vicious
cycle rapidly consumes oxygen in solution leading to the produc-
tion of peroxidized fatty acids.
Determination of singlet oxygen luminescence decay and rise
time
These results in solutions can also occur inside cells, which
are frequently exposed to ultraviolet irradiation as for the skin and
the eye. Fatty acids are ubiquitously present in human cells but also
exhibit the so-called ‘‘acid-mantle of the stratum corneum’’ that is
important for both permeability barrier formation and cutaneous
As shown in ref. 43 the luminescence intensity is given by
ꢂ
ꢀ
ꢁ
ꢀ
ꢁꢃ
C
t
t
IðtÞ ¼
exp ꢀ
ꢀ exp ꢀ
(1)
ꢀ1
ꢀ1
tR ꢀ tD
tD
tR
The constant C was used to fit the luminescence signal. tD and antimicrobial defense.44,45 Consequently, the generation of singlet
tR are the decay and rise times, respectively. To determine the oxygen may affect cellular and skin integrity leading to the
rise and decay times of singlet oxygen luminescence, we used the frequently reported disturbance of the skin barrier, skin ageing
least square fit routine of Mathematica 5.2 (Wolfram Research, and even skin malignancies.
Berlin, Germany). The experimental error of the fit was estimated
The photosensitized generation of singlet oxygen by endogenous
to be between 15 and 25% of the values that were determined by photosensitizers such as flavins and its role for those adverse
the fit. The low signal level in some samples yielded a higher reactions are well documented.19 In this study it could be shown
error of 25%. The integral of eqn (1) from t = 0 - N yields the that under UVA irradiation the amount of generated singlet oxygen
luminescence energy. To compare the luminescence energy of depends on the number of double bonds of fatty acids and the
the different probes, the photomultiplier detected photons were amount of respective hydroperoxides. The extent of contribution
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 17672--17680 17679

View Article Online
PCCP
Paper
19 J. Baier, T. Maisch, M. Maier, E. Engel, M. Landthaler and
W. Baumler, Biophys. J., 2006, 91, 1452–1459.
20 J. Baier, T. Maisch, J. Regensburger, C. Pollmann and
W. Baumler, J. Biomed. Opt., 2008, 13, 044029.
21 M. A. Lampe, M. L. Williams and P. M. Elias, J. Lipid. Res.,
1983, 24, 131–140.
to those adverse effects of singlet oxygen generation upon UVA
exposure, in particular in the so-called ‘‘acid-mantle of the
skin’’, is worth to be investigated in the near future.
Acknowledgements
The excellent technical assistance of Francesco Santarelli is gratefully
acknowledged. Francesco Santarelli was funded by a grant of Dr
August Wolff GmbH & Co. KG. The work of Johannes Regensburger
is supported by a grant of the ‘Deutsche Forschungsgemeinschaft’
(DFG), grant number RE3323/2-1. The work of Anita Gollmer is
supported by a grant of the ‘Deutsche Forschungsgemeinschaft’
(DFG), grant number GO2340/1-1. The work of Alena Knak is
22 M. A. Lampe, A. L. Burlingame, J. Whitney, M. L. Williams,
B. E. Brown, E. Roitman and P. M. Elias, J. Lipid. Res., 1983,
24, 120–130.
23 M. Wrona, W. Korytowski, M. Rozanowska, T. Sarna and
T. G. Truscott, Free Radical Biol. Med., 2003, 35, 1319–1329.
24 T. Kriska, W. Korytowski and A. W. Girotti, Free Radical Biol.
Med., 2002, 33, 1389–1402.
¨
supported by a grant of the ‘Bayerische Eliteforderung’. All
authors declare no conflict of interest.
25 N. Bando, H. Hayashi, S. Wakamatsu, T. Inakuma,
M. Miyoshi, A. Nagao, R. Yamauchi and J. Terao, Free
Radical Biol. Med., 2004, 37, 1854–1863.
26 Y. Minami, K. Yokoyama, N. Bando, Y. Kawai and J. Terao,
Free Radical Res., 2008, 42, 197–204.
Notes and references
1 S. Seite, A. Fourtanier, D. Moyal and A. R. Young, Br. J. 27 T. M. Runger and U. P. Kappes, Photodermatol., Photoimmunol.
Dermatol., 2010, 163, 903–914. Photomed., 2008, 24, 2–10.
2 M. Berneburg and J. Krutmann, J. Photochem. Photobiol., B, 28 A. W. Girotti, Photochem. Photobiol., 1990, 51, 497–509.
2000, 54, 87–93.
29 J. Baier, T. Maisch, J. Regensburger, M. Loibl, R. Vasold and
3 M. Berneburg, H. Plettenberg, K. Medve-Konig, A. Pfahlberg,
W. Baumler, J. Biomed. Opt., 2007, 12, 064008.
H. Gers-Barlag, O. Gefeller and J. Krutmann, J. Invest. Dermatol., 30 O. Sergent, I. Morel, P. Cogel, M. Chevanne, M. Beaugendre,
2004, 122, 1277–1283.
P. Chillard and J. Cillard, Anal. Biochem., 1993, 211, 219–223.
31 N. A. Porter, R. A. Wolf, E. M. Yarbro and H. Weenen,
Biochem. Biophys. Res. Commun., 1979, 89, 1058–1064.
32 K. E. Fygle and T. B. Melo, Chem. Phys. Lipids, 1996, 79,
39–46.
33 W. Baumler, J. Regensburger, A. Knak, A. Felgentrager and
T. Maisch, Photochem. Photobiol. Sci., 2012, 11, 107–117.
34 U. Feister, E. Jakel and K. Gericke, Photochem. Photobiol.,
2002, 76, 281–293.
4 A. M. Bode and Z. Dong, Sci. STKE, 2003, 2003, RE2.
5 L. O. Klotz, N. J. Holbrook and H. Sies, Curr. Probl. Dermatol.,
2001, 29, 95–113.
6 L. O. Klotz, K. D. Kroncke and H. Sies, Photochem. Photobiol.
Sci., 2003, 2, 88–94.
7 T. Schwarz, J. Photochem. Photobiol., B, 1998, 44, 91–96.
8 T. Schwarz, Keio. J. Med., 2005, 54, 165–171.
9 J. Wenk, P. Brenneisen, C. Meewes, M. Wlaschek, T. Peters,
R. Blaudschun, W. Ma, L. Kuhr, L. Schneider and 35 K. C. Madison, J. Invest. Dermatol., 2003, 121, 231–241.
K. Scharffetter-Kochanek, Curr. Probl. Dermatol., 2001, 29, 36 A. W. Girotti, Free Radical Biol. Med., 2008, 44, 956–968.
83–94.
37 A. W. Girotti, J. Photochem. Photobiol., B, 2001, 63, 103–113.
10 M. Wlaschek, I. Tantcheva-Poor, L. Naderi, W. Ma, 38 F. Wilkinson, W. P. Helman and A. B. Ross, J. Phys. Chem.
L. A. Schneider, Z. Razi-Wolf, J. Schuller and K. Scharffetter-
Kochanek, J. Photochem. Photobiol., B, 2001, 63, 41–51.
11 H. Ikehata and T. Ono, Photochem. Photobiol., 2007, 83, 196–204.
12 R. M. Tyrrell, Biochem. Soc. Symp., 1995, 61, 47–53.
13 S. L. Parker, T. Tong, S. Bolden and P. A. Wingo, Ca–Cancer
J. Clin., 1997, 47, 5–27.
Ref. Data, 1995, 24, 663–1021.
39 S. Miyamoto, G. E. Ronsein, F. M. Prado, M. Uemi,
T. C. Correa, I. N. Toma, A. Bertolucci, M. C. Oliveira,
F. D. Motta, M. H. Medeiros and P. D. Mascio, IUBMB Life,
2007, 59, 322–331.
40 D. A. Pratt, K. A. Tallman and N. A. Porter, Acc. Chem. Res.,
2011, 44, 458–467.
14 A. P. Schuch and C. F. Menck, J. Photochem. Photobiol., B,
2010, 99, 111–116.
41 G. A. Russell, J. Am. Chem. Soc., 1957, 79, 3871–3877.
15 E. Kvam and R. M. Tyrrell, Carcinogenesis, 1997, 18, 42 J. Baier, M. Maier, R. Engl, M. Landthaler and W. Baumler,
2379–2384. J. Phys. Chem. B, 2005, 109, 3041–3046.
16 G. M. Halliday, N. S. Agar, R. S. Barnetson, H. N. Ananthaswamy 43 D. Baumer, M. Maier, R. Engl, R. Markus Szeimies and
and A. M. Jones, Photochem. Photobiol., 2005, 81, 3–8. W. Baumler, Chem. Phys., 2002, 285, 309–318.
17 K. Scharffetter-Kochanek, M. Wlaschek, P. Brenneisen, 44 J. W. Fluhr, J. Kao, M. Jain, S. K. Ahn, K. R. Feingold and
M. Schauen, R. Blaudschun and J. Wenk, Biol. Chem.,
1997, 378, 1247–1257.
18 J. Baier, T. Maisch, M. Maier, M. Landthaler and W. Baumler,
J. Invest. Dermatol., 2007, 127, 1498–1506.
P. M. Elias, J. Invest. Dermatol., 2001, 117, 44–51.
45 Y. Zheng, H. Yin, W. E. Boeglin, P. M. Elias, D. Crumrine,
D. R. Beier and A. R. Brash, J. Biol. Chem., 2011, 286,
24046–24056.
c
17680 Phys. Chem. Chem. Phys., 2013, 15, 17672--17680
This journal is the Owner Societies 2013