Photodegradation of prednisolone under UVB solar irradiation. Role of photogenerated ROS in the degradation mechanism

Article abstract of DOI:10.1039/c7pp00200a

The use of biologically active substances with anti-inflammatory properties such as corticosteroids has increased considerably in the last few decades. Particularly, the compound we are interested in, prednisolone (Predn), is a glucocorticoid with high biological activity. This compound absorbs UV radiation and may participate in photochemical processes, which can result in its own decomposition. These processes could result in the formation of free radicals or reactive oxygen species (ROS). On these grounds, the kinetic and mechanistic aspects of the direct photodegradation of Predn have been studied in aqueous medium under different atmospheric conditions by stationary and time-resolved techniques. The mechanism involved in the photodegradation has been elucidated. Predn is capable of generating the excited triplet state ³Predn? as a result of UVB light absorption. In the presence of oxygen, ³Predn? allows the formation of ROS, of which O₂(¹Δ_g) (Φ_Δ = 0.014), H₂O₂ and the radical OH stand out. The latter is generated from the spontaneous dismutation of O₂^- and subsequent homolytic cleavage, photochemically promoted, of H₂O₂. Predn undergoes unimolecular photodegradation reactions under an inert argon atmosphere. In this study we found that in the presence of oxygen, the Predn photo-consumption is improved. This implies that the attack by ROS involves a very important additional contribution to the photodegradation of Predn under aerobic conditions.

Full text of DOI:10.1039/c7pp00200a

View Article Online
View Journal
Photochemical &
Photobiological
Sciences
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: D. Cacciari, E.
Reynoso, H. A. Montejano and A. Biasutti, Photochem. Photobiol. Sci., 2017, DOI: 10.1039/C7PP00200A.
Volume 15 Number
3
March 2016 Pages 311–458
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Photochemical &
Photobiological
Sciences
An international journal
www.rsc.org/pps
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1474-905X
rsc.li/pps

Page 1 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
Photodegradation of Prednisolone under UVB solar irradiation. Role of ROSꢀ
photogenerated in the degradation mechanism.
R. Daniel Cacciari, Eugenia Reynoso, Hernán A. Montejano*and M. Alicia Biasutti*
Departamento de Química, Facultad de Ciencias Exactas, FísicoꢀQuímicas y Naturales,
Universidad Nacional de Río Cuarto, Ruta Nacional 36 Km 601, X5804BYA Río Cuarto,
Córdoba, Argentina.

Photochemical & Photobiological Sciences
Page 2 of 35
View Article Online
DOI: 10.1039/C7PP00200A
Abstract
The use of biologically active substances with antiꢀinflammatory properties such as
corticosteroids has increased considerably in the last decades. Particularly, the compound
we are interested in, Prednisolone (Predn), is a glucocorticoid with high biological activity.
This compound absorbs UV radiation and may participate in photochemical processes,
which can result in its own decomposition. These processes could result in the formation of
free radicals or reactive oxygen species (ROS).On these grounds, the kinetic and
mechanistic aspects of the direct photodegradation of Predn have been studied in aqueous
medium under different atmospheric conditions by stationary and timeꢀresolved techniques.
The mechanism involved in the photodegradation has been elucidated. Predn is capable of
3
generating the excited triplet state Predn* as a result of UVB light absorption. In the
3
*
1
presence of oxygen, Predn allows the formation of ROS, of which the O ( ꢁ ) (Φ =0.014),
2
g
ꢁ
•
H O and the radical OH highlight. The latter is generated from spontaneous dismutation of
2
2
•ꢀ
O2 and subsequent homolytic cleavage, photochemically promoted, of H O . Predn
2
2
undergoes unimolecular photodegradation reactions in inert argon atmosphere.
In this study we found that in the presence of oxygen, the Predn photo consumption is
improved. This implies that the attack by ROS involves a very important additional
contribution to the photodegradation of Predn in aerobic conditions.
Keywords: Prednisolone, corticosteroids, photodegradation mechanism, reactive oxygen
species.

Page 3 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
1
.
Introduction
Any substrate possessing the ability to absorb photons can participate in a photochemical
process; i.e. may be photoreactive, which can lead to its own degradation or degradation of
substances associated to it/or present in the medium, among which compounds of
biological origin such as proteins, sugars, lipids, or nucleic acids, are the most important.
Moreover, a drug product can be exposed to natural or artificial light during production,
storage and use. The great majorities of pharmaceutical substances absorb UV radiation
and may participate in a photochemical process which can result in its own decomposition.
These processes could cause the formation of free radicals or reactive oxygen species
(ROS) [1].
On the other hand, the use of biologically active substances with antiꢀ
inflammatory properties has increased considerably in the last decades. In the present
contribution, we are interested in the photochemical behavior of corticosteroids. These
compounds are the main antiꢀinflammatory drugs currently used, they are synthetic
derivatives of hydrocortisone, a steroid hormone produced in the adrenal gland [2]. They
are highly used in many fields of modern medicine [3] because they possess the ability to
inhibit immunological, inflammatory and allergic processes that occur in response to
external harmful stimuli [4]. Particularly, the compound we are interested in, Prednisolone
(Predn)is a glucocorticoid with high biological activity. It is part of the list of "Essential
Medicines" created by the World Health Organization (WHO) [5], which nucleates all
drugs that are considered important in health care of most of the population. Predn is the
biologically active metabolite of prednisone, which is obtained by dehydrogenation of the

Photochemical & Photobiological Sciences
Page 4 of 35
View Article Online
DOI: 10.1039/C7PP00200A
molecule of Cortisone in positions 1 and 2 of the cyclopentaneꢀperhydroꢀphenantrene
nucleus [6].
A very important and exhaustive compilation about photostability of drugs, has
been done by H.H. Tønnesen [7]. Particularly, the photoreactivity of glucocorticosteroides
have been investigates in solution as well in solid state and it has been demonstrated that
these compounds must to be protected from light. Glucocorticosteroides cross conjugates
such as prednisolone, prednisone betamethasone, triamacinolone and others result to be
quite photoreactive. This fact may expect taking into account that the efficient
photorearrangement of cyclohexadienones to bicycle hexanones is well known [8].
Another aspect that should be considered when studying the photochemical
behavior of drugs is the new approach that has been given to the study of these compounds
as environmental pollutants, specifically as water pollutants [9,10]because when these
drugs are consumed by humans, they are not fully metabolized and are eliminated virtually
unchanged through the urine or feces. Only some of these substances can be removed by
wasteꢀwater treatment plants; consequently, the drugs reach water reservoirs [11,12],
accumulate and cause ecological damage, as well as hormonal and immunological changes
on aquatic populations[13]. These substances are part of the group of emergent
contaminants and their impact on the environment is not fully understood.
Even when the toxicity of Predn and its photochemical derivatives on aquatic
organisms have been investigated by other authors [14], the mechanism and kinetic aspects
of the photodegradation process has not yet been elucidated.
Several studies related to photochemical behavior of Predn and structural related
compounds under an inert atmosphere have been done. Beginning pioneering works
related to the photochemistry of prednisone acetate and related substances [15], Williams

Page 5 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
et al. decided to revisit the issue [16,17]. They propose to reꢀinvestigate the
photochemistry of various steroids, finding differences in the photoproducts obtained in
deoxygenated organic solvents and mixtures of acetic acid/water, and concluding that
photoisomerizations of crossꢀconjugated cyclohexadienones take place via an nꢀπ* excited
triplet state and show a strong solvent dependence.
More recently, Ricci et al. [18] returned to the subject of photochemistry (in
oxygen free atmosphere) of Pregnaꢀ1,4ꢀdienꢀ3,20ꢀdiones and related molecules in which
fluorine or chlorine atoms were incorporated. They found different reaction products
depending on the wavelength at which the compounds (dissolved in acetonitrile) were
irradiated.
On the other hand, a study [19] of the photocatalytic transformation of
dexamethasone using titanium dioxide as a photocatalyst have demonstrated that the
photodegradation occurs through the formation of some hydroxyl derivatives. The main
•
oxidation step occurs from the attack by one OH radical.
Despite this, very little is known about its photochemistry in aqueous medium
under aerobic conditions. Here we present a study on the photochemistry of this emergent
pollutant under advantageous conditions to achieve its natural degradation in the
environment, i.e. solar irradiation of watercourses with dissolved oxygen. The focus was
placed on the influence of ROS as intermediaries for the Predn degradation under such
conditions.
The importance of studying the mechanism of photodegradation of Predn, which
are of great pharmacological and environmental importance, is that the elucidation of
pathways which are part of the mechanism, will allow access to important information,
such as: a) the feasibly of these events occur under certain environmental condition, b) the

Photochemical & Photobiological Sciences
Page 6 of 35
View Article Online
DOI: 10.1039/C7PP00200A
speed and efficiency of the process, c) the values of controllable variables to avoid the
photodegradation, d) the possible photoproducts formed, to mentioned only a few.
The aims of this study were to identify the behavior of Predn under UVB sunlight
irradiation and to determine the influence of oxygen on the degradation reactions of the
drug. To accomplish these goals, the study of the photophysical and photochemical
processes of the glucocorticoid Predn, under different experimental conditions, has been
done by using different techniques in order to evaluate the participation of the different
ROS and to propose a possible mechanism which accounts for the processes and species
involved in the degradation of the compound.
2
.
Experimental
2.1.Materials
Prednisolone 99% (Predn), superoxide dismutase from bovine erythrocytes
Biuret>95% (SOD) and catalase from bovine liver (CAT) Biuret> 95%. Sodium azide99%
(
NaN ) and isopropanol 99.7% (isoOH) were purchased from Merck. Furfuryl acetate 99%
3
(FFAc), Perinaphthenone 97% (PN) and Rose Bengal 95%were provided by Aldrich
Chemical Company. Acetonitrile (MeCN), HPLC quality, were provided by Sintorgan.
All these compounds were used as received. The water used was triply distilled. All
measurements were carried out with freshly prepared solutions at 298 K, except when other
temperature is indicated. In order to increase the solubility of Predn, 1% MeCN was added.
Where otherwise indicated, “aqueous solution” refers to water/MeCN 99/1% v/v mixture.
2.2.Methods
2
.2.1. Stationary photolysis experiment

Page 7 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
Stationary direct photolysis experiments were carried out by irradiation of aqueous
ꢀ5
solutions containing Predn, ([Predn]=4x10 M), which was irradiated with monochromatic
light at the maximum of the absorption band (248 nm, see Figure 1) at room temperature.
The equipment used consists on a Photon Technology International (PTI) unit provided
with a high pass monochromator, and equipped with a 150 W Xenon lamp UVꢀvisible
absorption spectra which were recorded on a Shimadzu UVꢀ2401PC spectrometer
(Shimadzu Corporation, Tokyo, Japan).
Experiments in airꢀequilibrated aqueous solutions were carried out in a quartz
spectrophotometric cell, open to the atmosphere and under constant magnetic stirring. In
the experiments in which it was necessary to ensure a high concentration of oxygen, the
reaction cell was continuously bubbled with this gas. Experiments under anaerobic
conditions were made in a quartz cell provided with a hermetic seal and bubbled with argon
for 15 min.
For the determination of the direct photodegradation quantum yields of Predn
Predn
(ΦR
), measurements were performed according to the method described by Rahn et al.
[
20] using a KI/KIO mixture in a borates buffer as actinometer.
3
To establish the possible involvement of different ROS in the photodegradation of
Predn, direct photolysis experiments were carried out in the presence and in the absence of
specific ROS scavengers.
1
In order to corroborate the ability of Pred to generate O ( ꢁ ), the photooxidation of
2
g
FFAc in the presence of the corticoid was evaluated in oxygen and argon atmosphere
1
following the decrease of the absorption band of FFAc at 215 nm. FFAc react with O ( ꢁ )
2
g
7
−1 −1
with a rate constant value of 5.5×10 M s [21].

Photochemical & Photobiological Sciences
Page 8 of 35
View Article Online
DOI: 10.1039/C7PP00200A
2
.2.2. Laser flash photolysis experiment
Transient absorption measurements were performed using laserꢀflash photolysis
equipment. A Spectron SL400 Nd:YAG laser generating 266 nm laser pulses (~18 ns pulse
width) was used for sample excitation. The laser beam was defocused in order to cover all
the path length (10 mm) of the analyzing beam from a 150 W Xenon lamp. The detection
system comprises a PTI monochromator coupled to a redꢀextended Hamamatsu R666 PM
tube. The signals initially captured by a HP54504 digitizing oscilloscope were averaged
and then transferred to a computer for storage and analysis. The experiments were
determined in argonꢀsaturated MeCN solutions.
2
.2.3. Timeꢀresolved singlet oxygen phosphorescence detection
1
The timeꢀresolved O ( ꢀ ) phosphorescence detection (TRPD) has been previously
2
g
described [22]. Briefly, it consisted of a Nd:YAG laser (Spectron) as the excitation source.
The output at 266 nm was employed to excite Predn and PN. The emitted radiation (1270
nm) was detected at right angles using an amplified Judson J16 ꢂ 8Sp Germanium detector
after passing through appropriate filters. The output of the detector was coupled to a digital
oscilloscope and to a personal computer for processing the signal. Sixteen shots were
averaged to get a good signalꢀtoꢀnoise ratio, from which the decay times were calculated
[23].
1
The O ( ꢀ ) production by Predn (φ ) was determined in MeCN by the comparative
2
g
ꢀ
1
method already described [24]. The dependence of O ( ꢀ ) phosphorescence intensity at
2
g
1
270 nm, extrapolated to zeroꢀtime (I ), on the laser fluence (E ) was measured for Predn
0 L
and the reference solutions, changing the laser fluency by employing filters of neutral

Page 9 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
density. φ value was calculated by comparison of the slopes of the linear plots of I vs.E
ꢀ
0
L
for reference and sample.
It is known that the slope can be considered as proportionally constant, which
accounts for electronic and geometric characteristics of the detection system and solvent
properties. Hence, a reference is needed for calibration. The reference was PN in MeCN
(
φꢀ=0.95) [25]. Measurements of sample and reference under identical experimental
conditions, constant characteristics of irradiation wavelength and matched absorbances
allow the evaluation of φꢀ.
2
.2.4. Semi-empirical molecular orbital calculations
Semiꢀempirical molecular calculations (AM1) were carried out using CS Chem3D Pro
software from Cambridge Soft Corporation. AM1 was employed in structure geometry
optimization and calculations for the solute dipole moment.
3. Results and Discussion
3.1.Photolytic kinetic and quantum yield of photodegradation
Although UVB radiation range comprises from 280 nm to 315 nm, the excitation of the
sample was carried out in the maximum of the absorption band, in order to obtain a greater
sensitivity in the observed changes.
The photodegradation of Predn was studied in aqueous solution by direct irradiation
with monochromatic light at the maximum of the absorption band (248 nm) as a function of
time, following the changes in the absorption spectra of the corticoid.
In order to ensure the absence of substrate aggregation effects in the solvent used in this
study, the absorption spectra of Predn in the water/MeCN 99/1% v/v mixture were

Photochemical & Photobiological Sciences
Page 10 of 35
View Article Online
DOI: 10.1039/C7PP00200A
ꢀ5
ꢀ3
determined at different substrate concentrations from 1.5x10 to 3.0x10 M. As shown in
Figure 1ꢀa (supplementary material) no spectral changes although the change in
concentration is more than two orders of magnitude.
In addition, to confirm the absence of aggregation phenomena, absorption spectra were
determined in different solvents, including MeCN and MeOH, for which the solubility of
Predn is much higher than that corresponding to water. As can be seen in Figure 1ꢀb
(supplementary material) there is only a characteristic band shift for the variation of
polarity of the medium, without hint of new bands or shoulders.
The solubility of the corticoid in the solvent mixture used is consistent with the
moderate dipole moment calculated for Predn (5.2 D, see section 2.2.4.).
In order to evaluate the influence of the oxygen concentration in the process,
equivalent experiments were made at different atmospheric condition: argonꢀsaturated
solution, oxygenꢀsaturated solution and airꢀbalanced solution. The irradiation at 248 nm
produces changes in the Predn absorption spectra; the result obtained for oxygen and argon
atmospheres are shown in Figure 1. In all cases, a decrease in the intensity of the absorption
band with the increase of irradiation time was observed as a consequence of Predn
photodegradation. Although the consumption of the corticoid is independent of the nature
of the atmosphere, the spectral modifications in inert atmosphere conditions are different
from those in the presence of oxygen. In oxygen atmosphere, a decrease of the band is
observed without noticeable shifts in peak position as well as two clear isosbestic points,
while in inert atmosphere, decreased absorption is observed with a progressive blue shift of
the maximum. These results are a clear signal that the photoproducts formed in the absence
of oxygen are different in nature from those obtained in the presence of this gas.

Page 11 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
It must be noted; that the photoproducts of prednisolone, dexamethasone and
prednisone have been already identified under sunlight irradiation in air saturated aqueous
solution, through different techniques [14,26]. However, the mechanistic aspects have not
been elucidated. In this context, we carry out several experiments with the aim to have
evidence to propose a possible mechanism.
Based on the observed changes in the absorption spectra, the photodegradation
Predn
quantum yield of Predn (ΦR
) was determined by actinometrical measurements using the
equation below:
ꢅ
ꢃꢄ ꢆꢃ
ꢀ
ꢁ ꢂꢃꢄꢅꢇ ꢈ =ꢉ±Φ ꢋ ꢌꢀ × ꢍ
ꢊ ꢄ
ꢆꢃ
Equation 1
where the + or ꢀ sign depends on whether we consider the formation of a product or
disappearance of a reagent, respectively.
From the slope of the plot of the logarithmic terms vs. irradiation time in equation
Act
(
1) and, knowing the actinometer quantum yield (ΦR ) and its molar absorption coefficient
ꢀ1 ꢀ1
(
ε =26400 M s ) [15] in the experimental conditions, it is possible to determine the value
3
52
of the incident radiation intensity (I ). Then, with the I value and knowing Predn molar
0
0
ꢀ1
ꢀ1
absorption coefficient (ε =18500 M cm ,determined in the present work) in the same
2
48
Predn
experimental condition that the actinometer, it is possible to calculate ΦR
from the slope
of the plot of the logarithmic terms vs. irradiation time in equation (1). The respective
Predn
values of ΦR
at different atmospheric condition are summarized in Table 1.
ꢀ5
Considering that the substrate concentration is 3x10 mol/L, this represents the
upper limit for the concentrations of all the reactive species formed (if its generation

Photochemical & Photobiological Sciences
Page 12 of 35
View Article Online
DOI: 10.1039/C7PP00200A
efficiencies were unitary).Since the concentration of oxygen in air saturated water at 298 K
is 0.27 mmol/L [27], this value represents a large excess compared to the initial substrate
concentration, and for this reason no differences are observed when the concentration of
dissolved oxygen is increased. Then it reaches a saturation situation (plateau) which
Predn
justifies that the ΦR
determined in the atmospheres of air and oxygen are practically the
same. Experiments were performed in triplicate, always obtaining the same values.
On the other hand, considering that Predn photodegradation obeys a firstꢀorder
kinetic behavior and reordering equation 1, the apparent firstꢀorder rate constant (k ) can
app
be determined from the semiꢀlogarithmic plot of A /A as a function of irradiation time
t
0
(equation (2)) for each experimental condition, based on the analysis of the decrease in the
absorbance at 248 nm.
ꢎꢏꢐ
ꢑ
ꢀ
ꢁ ꢂꢎꢏꢐ ꢈ = −ꢉ2.303ꢉℓꢒΦ ꢋ ꢔꢌ × ꢍꢉ =ꢉ−ꢕ
× ꢍ
ꢖꢗꢗ
ꢓ ꢄ
ꢓ
ꢇ
Equation 2
The firstꢀorder plots are shown in Figure 2 and the kapp values are displayed in Table
. Under complete or partial oxygenation conditions, it is observed that the
1
photodegradative kapp of Predn is higher than the one found in anaerobic conditions.
This provides evidence that molecular oxygen is involved, in some way in
photochemical reactions thus allowing inferring Predn can generate ROS as a result of the
UVꢀabsorption. Selfꢀsensitization capacity for the photochemical generation of ROS by
steroid compounds has been lately demonstrated in the case of the steroid estrogens, 17αꢀ
ethynylestradiol[28].
Furthermore, it is very important to note that Predn also undergoes photochemical

Page 13 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
reactions in the absence of oxygen. This behavior is probably due to typical unimolecular
photochemical reactions, mainly lumiketone rearrangement following the UV light
absorption[17,29].
However, other authors have done a very interesting research for two synthetic steroids
with a great anabolic potency: Boldenone and Trenbolone [30]. They found a different
behavior, observing that the rates of the photo consumption process are identical in the
absence and in the presence of oxygen, in the reaction mixture. This fact implies that the
ROS are not involved in the photodegradation mechanism of such steroids.
3.2.Interaction of Predn with photogenerated ROS
According to what was discussed above, molecular oxygen plays an important role
in the process of direct photodegradation of Predn, and this process could be the result of
the participation of photogenerated ROS, obtained by selfꢀsensitization from the corticoid.
In order to elucidate the nature of ROS involved in the process, photolysis experiments
with monochromatic light irradiation in the presence and in the absence of additives with
ROSꢀscavenging capacity were carried out in aqueous solution. For SOD, CAT and isoOH
the wavelength irradiation was 248 nm, while for NaN the experiment was carried out at
3
2
60 nm to avoid the absorption of this salt.
As shown in Figure 3, no change was observed in the rate of photoꢀconsumption of
Predn in the presence of 10 mM NaN as compared to the respective photoꢀconsumption in
3
1
the absence of the salt. This salt is an exclusive physical quencher of O ( ꢀ ) with a rate
2
g
8
−1 −1
constant of 3×10 M s in aqueous solution (process 1) and has been used previously, in
1
similar concentration, to ascertain whether O ( ꢀ ) participates in a given reaction
2
g

Photochemical & Photobiological Sciences
Page 14 of 35
View Article Online
DOI: 10.1039/C7PP00200A
1
mechanism [31,32]. Then, these results suggest that O ( ꢀ ) is not involved in Predn
2
g
photodegradation under this work experimental conditions. Section 3.3 will show that the
photoproduction of singlet oxygen is very low with a Φ of 0.014 ± 0.002
ꢀ
.
1
To confirm that O ( ꢀ )does not contribute to the consumption of Predn, a photolysis
2
g
1
experiment sensitized by Rose Bengal (which has a quantum yield of O ( ꢀ )formation of
2
g
1
0
.7 in water [33]) was carried out. The purpose of this experiment was to generate O ( ꢀ )
2
g
at much higher concentration than that produced by energy transfer from the triplet excited
3
*
state of Predn ( Predn ) to the ground state of oxygen. In this experiment (using cutꢀoff
filters to avoid direct excitation of the substrate) no changes in the concentration of Predn
were observed even after prolonged irradiation. (See figure 2 in supplementary material).
1
This result confirms clearly that there is no reaction of Predn with O ( ꢀ ).
2
g
ꢃ
ꢆ
ꢞ
ꢆ ∗
ꢞ
ꢘ ꢚ ∆ ꢜ +ꢉꢝ ꢉ→ꢉ ꢒꢝ ꢔ +ꢉꢘ ꢚ Σ ꢜ
process 1
ꢙ
ꢛ
ꢞ
ꢞ
ꢙ
ꢛ
A similar experiment was performed in the presence of 1 mg/100 ml SOD. This
•−
enzyme, which catalyzes the dismutation of O2 through process 2,has been utilized
•−
previously as a quencher of O2 mediated photooxidation, in similar concentrations to those
employed in this work [34ꢀ36].
ꢡꢢꢣ
∙
ꢆ
ꢠ
ꢞ
2
ꢘ + 2ꢟ ꢤꢥꢦ ꢘ ꢚ Σ ꢜ +ꢉꢟ ꢘ
process 2
ꢙ
ꢙ
ꢛ
ꢙ ꢙ
The increase in the rate of Predn photoꢀconsumption induced by the presence of the
•–
enzyme, supports the idea of the involvement of the O2 species. It is well known [37,38],

Page 15 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
•–
that in an O mediated mechanism, SOD can either preclude or stimulate the oxidation of a
2
given substrate, depending on the particular mechanistic role. That is, the latter effect will
depend on whether ROS acts as a final oxidative species or as an intermediate one in the
•
subsequent generation of other ROS. Due to the fact that eventually other ROS such as OH
•−
and H O can be generated from O , equivalent experiments in the presence of specific
2
2
2
•
scavengers for OH and H O (0.2 M isoOH and 1 mg/100 ml CAT, respectively) have
2
2
been performed. CAT and isoOH had been used previously for the detection of ROS
•
[
28,39ꢀ41]. The species H O is decomposed by CAT according to process 3 whereas OH
2 2
reacts with isoOH by process4.
ꢧꢏꢨ
ꢞ
2
ꢟ ꢘ ꢤ ꢥꢦ 2ꢟ ꢘ + ꢘ ꢚ Σ ꢜ
process 3
process 4
ꢙ
ꢙ
ꢙ
ꢙ
ꢛ
∙
ꢒꢩꢟ ꢔ ꢩꢟꢘꢟ + 2ꢘꢟ →ꢉꢒꢩꢟ ꢔ ꢩꢘ +ꢉ2ꢟ ꢘ
ꢞ ꢙ
ꢞ ꢙ
ꢙ
In Figure 3, a decrease in the Predn photoꢀconsumption in the presence of both
ROSꢀscavengers is observed. These results could indicate that, initially, the
•
abovementioned species (H O and OH ) are involved in the Predn photodegradation
2
2
•−
mechanism. The H O , generated from O dismutation, undergoes homolytic cleavage
2
2
2
•
giving OH . The decomposition reaction of H O photochemically catalyzed (process 5) has
2
2
been extensively studied [42ꢀ44]. The irradiation conditions used in this work could
•
facilitate the photogeneration of OH from H O .
2
2

Photochemical & Photobiological Sciences
Page 16 of 35
View Article Online
DOI: 10.1039/C7PP00200A
ꢪꢫ
∙
ꢟ ꢘ ꢤꢦ ꢉ2ꢘꢟ
process 5
ꢙ
ꢙ
Similar results were found for 17αꢀethynylestradiol upon addition of isoOH into
•
solution. The degradation rate decreased. This indicates that the OH , photogenerated from
H O , is involved in the selfꢀsensitization photodegradation of this steroid [28].
2
2
To reinforce the participation of the radical OH• in the photolysis of Predn, a similar
experiment (to that carried out with isoOH) was done in the presence of mannitol, which is
another typical OH• radical trap. The rate of consumption of Predn in the presence of
mannitol decreases with respect to similar experience in the absence of this OH• scavenger.
Although photolysis experiments in the presence of NaN do not present significant
3
1
changes in Predn photodegradation followed by absorption, it is possible that O ( ꢀ ) could
2
g
be generated in the medium, but does not react with the corticoid. To corroborate this,
photooxidation experiments of a reference compound, furfuryl acetate (FFAc), were made
in oxygenꢀsaturated and argonꢀsaturated aqueous solutions in the presence of Predn. FFAc
1
selectively reacts with O ( ꢀ ) [45] through process 6.
2
g
ꢃ
ꢬꢬꢭꢮ +ꢉꢘ ꢚ ∆ ꢜꢉ→ ꢯꢰꢱꢲꢳꢮꢍꢴ
process 6
ꢙ
ꢛ
The decrease in FFAc absorbance with the increase of irradiation time in oxygenꢀ
saturated condition (Figure 4), as well as the absence of changes in inert atmosphere (inset,
1
Figure 4) allows us to conclude that O ( ꢀ ) is generated. However, this ROS does not
2
g
interact with Predn, which justifies the consumption profile of the corticoid in the presence
of NaN3.

Page 17 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
1
•−
All the above evidence indicates that neither O ( ꢀ ), nor O , nor H O react with
2
g
2
2
2
•
Predn, which allows us to assume that OH is the species responsible for the degradation of
Predn. This agrees with the previously photodegradation reported for other related
corticosteroid (dexamethasone) [19], in which the photoinduced transformation also was
•
initiated only by OH radical attack.
3.3.Singlet oxygen generation by Predn
To corroborate the effects observed in the photolysis experiments with FFAc in the
1
presence of Predn, the ability of the corticoid as a O ( ꢀ ) generator was quantified through
2
g
1
TRPD as described in the experimental section. Quantum yield of O ( ꢀ ) generation by
2
g
Predn (Φ ) was determined in MeCN and by using PN as a reference. Results are shown in
ꢀ
Figure5 and the value obtained: 0.014 ± 0.002 was determined based on equation 3:
ꢷ
ꢸꢹꢗꢺꢉꢒꢻꢼꢺꢽꢾꢔ
Φ ꢒꢯꢰꢶꢲꢁꢔꢉ=ꢉ
ꢉ×ꢉΦ ꢒꢯꢝꢔ
Equation 3
ꢵ
ꢵ
ꢷꢸꢹꢗꢺꢉꢒꢻꢿꢔ
1
Although Φ has a low value, Predn generates O ( ꢀ ) photochemically, as it was
ꢁ
2
g
demonstrated by this experiment as well as by the FFAc photodegradation.
Other authors [46,47] have done a comprehensive investigation for other corticoids,
1
flumethasone and fluocinoloneacetonide. They found that these substrates generate O ( ꢀ )
2
g
under UVB irradiation with a remarkable efficiency for fluocinoloneacetonide than for
flumethasone.
1
However,O ( ꢀ ) is not involved in the photodegradation process of Predn as it was
2
g
evidenced in the photolysis experiments with the ROSꢀscavenger NaN3.

Photochemical & Photobiological Sciences
Page 18 of 35
View Article Online
DOI: 10.1039/C7PP00200A
3.4.Transient species generation
Because the carbonyl photochemical reactions and the photogeneration of ROS can
3
occur from Predn*, laser flash photolysis measures were performed with the aim to detect
the presence of this transient species.
The transient (difference) absorption spectrum of Predn in (deoxygenated) MeCN is
shown in Figure 6.Thespectrum shows positive absorptions between 300ꢀ500 nm, where an
intense band with a maximum at 330 nm and a shoulder near 350 nm is highlighted. Two
minor bands centered near 400 nm and 480 nm can be clearly identified too.
The absorption kinetic profiles determined between 300 and 500 nm followed firstꢀ
order kinetics with identical lifetimes (4 µs). From these results, it is possible to conclude
that the absorption bands observed in the transient spectrum correspond to a unique
transient species.
On the other hand, it is important to note that a possible intermolecular electron
transfer is discarded due to the absence of new bands formed concomitantly to the
disappearance of the triplet state of Predn (Figure 6). This is a good argument to infer the
absence of intermolecular auto quenching processes. That is, the triplet state does not react
with the substrate in the ground state neither with the solvent.
It should be noted that in the presence of oxygen, no significant transient absorption
signals could be detected. This is probably due to the fact that with the dissolved oxygen in
the solvent used the triplet state of Predn is completely deactivated (diffusional quenching),
and that the products of this process (step 4, scheme 2) does not show significant
absorption within the spectral and temporary range accessible to our equipment.

Page 19 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
These observations, together with the tripletꢀtriplet absorption spectrum previously
reported for other steroid (17ꢀβꢀhydroxyandrostaꢀ4,6ꢀdienꢀ3ꢀone (∆6ꢀtestosterone))[48]
which also has a dienone functionality, strongly suggest that the observed absorption
3
*
spectrum corresponds to the triplet excited state of Predn ( Predn ). To the best of our
knowledge, the tripletꢀtriplet absorption spectrum of Predn has not been reported yet in the
literature. A possible reason for this is the apparently very low efficiency of the intersystem
crossing process in these kinds of molecules.
3.5.Photodegradation mechanism proposed
Considering all the evidence presented, the mechanism shown in Scheme 2 is
hereby proposed, which explains the photodegradation of Predn in aqueous solution.
The absorption of UVB light promotes Predn, in ground state, to electronically
1
3
excited singlet and triplet states, Predn* and Predn*, respectively (reaction1 and 2).
It is assumed that the fluorescence quantum yield of Predn in aqueous solution is
negligible, since its emission could not be detected.
3
*
In aerobic conditions (pathway I) Predn may react with ground state oxygen
3
1
(
O ( Σ )) forming the species singlet oxygen (O ( ꢀ )) by energy transfer (reaction 3) or
2
g
2
g
•−
•−
superoxide radical anion (O ) by electron transfer (reaction4). O
could suffer
2
2
spontaneous dismutation generating hydrogen peroxide (H O ) by reaction 5. This species
2
2
may experience homolytic photoꢀcleavage of the OꢀO bond and generate hydroxyl radical
•
(
OH ) by reaction 6. This highly reactive ROS is also involved in the photodegradation of
Predn (reaction 7). It must be noted that H O , does not react with the corticoid (see Figure
2
2
3
).

Photochemical & Photobiological Sciences
Page 20 of 35
View Article Online
DOI: 10.1039/C7PP00200A
On the other hand, under inert atmosphere, the pathway II occurs, leading to the
3
unimolecular photodegradation of Predn. This reaction occurs very likely from the Predn*
by intramolecular reorganization (reaction 8) [14]. It is important to note that in the
presence of oxygen, both reactive pathways occur simultaneously.
In the literature, a photodegradation study of ibuprofen (IBP) under UV irradiation
in aqueous media has been carried out [41], and there it was proved that the mechanism
included direct photolysis and selfꢀsensitization via ROS. However, in that case, the photo
consumption of IBP is less effective in the presence of oxygen. This fact is indicating that
the direct photolysis way is preponderant in the global mechanism.
Contrary to what was observed in the photolysis of IBP, in this study we found that
in the presence of oxygen, the Predn photo consumption is enhanced (see Figure 2). This
implies that the attack by ROS (via 1) involves a significant additional contribution to the
photo degradation of the steroid in aerobic conditions. This fact is very important in order
to take into account this emerging contaminant photodegradation in aqueous wastes by
sunlight.
Conclusions
The main finding in this work is that the biologically active glucocorticoid Predn
undergoes photoꢀinduced degradation by irradiating with solar UVB light. This degradation
includes two reactive ways: direct photolysis and selfꢀsensitization via ROS.
The molecular oxygen has an important role in the Predn photodegradation
mechanism in aqueous medium, making the formation of highly oxidizing ROS possible.
3
*
On the other hand, Predn is capable of generating the excited triplet state, Predn
as a result of UVB light absorption. The spectrum of this transient species is reported here

Page 21 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
3
*
1
for the first time. Predn allows the formation of ROS, making O ( ꢀ ) (Φ = 0.014), H O
2
g
ꢁ
2
2
•
and the radical OH stand out.
Likewise, photodegradation of Predn in aqueous medium is controlled by the
•
1
radical OH , since the species O ( ꢁ ) and H O do not interact chemically with the drug.
2
g
2
2
Predn undergoes unimolecular photodegradation reactions in inert argon atmosphere.
In this study we found that in the presence of oxygen, the Predn photo
consumption is improved. This implies that the attack by ROS involves a very important
additional contribution to the photodegradation of Predn in aerobic conditions. These
results are essential for the purpose of considering the photodegradation of this emerging
contaminant in aqueous wastes by sunlight.

Photochemical & Photobiological Sciences
Page 22 of 35
View Article Online
DOI: 10.1039/C7PP00200A
Acknowledgements
Financial support from Consejo Nacional de Investigaciones Científicas y Técnicas
(
CONICET) and Secretaría de Ciencia y Técnica de la Universidad Nacional de Río Cuarto
SECyT UNRC), both from Argentine, is gratefully acknowledged.
(

Page 23 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
Figures, Schemes and captions
Scheme 1. Chemical structure of the corticoid Prednisolone (Predn)
Predn
(
(
1) hυ
1Predn*
2) kISC
3Predn*
I
II
(8) Predn
kET
3
3
1
3
*
(
3) Predn*+ O ( Σ )
Predn + O ( ꢀ )
Products
2
g
2
g
keT
.+ + O2.ꢀ
3
3
(
(
4) Predn*+ O ( Σ )
Predn
2
g
.
ꢀ
+
3
5) 2O + 2H
O ( Σ ) + H O
2
2
g
2
2
.
2 OH
hυ
(
6) H O
2
2
.
(
7) OH + Predn
Products
Scheme 2.Proposed pathway for photodegradation of Predn

Photochemical & Photobiological Sciences
Page 24 of 35
View Article Online
DOI: 10.1039/C7PP00200A
Predn
Table 1.Quantum yield (ΦR
) and rate apparent constant (k ) of photodegradation of
app
Predn, upon monochromatic irradiation at 248 nm, in different atmospheres condition.
argon
oxygen
air
Predn
ΦR
(0.019 ± 0.002)
(0.034 ± 0.004) (0.034 ± 0.005)
ꢀ1
ꢀ4
ꢀ4
ꢀ4
kapp (s )
(0.7 ± 0.2) x10
(1.4 ± 0.5) x10
(1.4 ± 0.4) x10

Page 25 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
0
0
0
0
0
.8
.6
.4
.2
.0
0.8
0.6
0.4
0.2
0.0
0
1
2
3
4
0
0
0
0
50
0
1
2
3
4
5
0
0
0
0
0
200
250
300
λ
(nm)
2
00
250
300
350
λ
(nm)
ꢀ5
Figure 1. Spectral evolution of Predn (4x10 M) in aqueous solution (water/MeCN 99/1%
v/v) irradiated at 248 nm, in oxygenꢀsaturated condition. Inset: Spectral evolution of Predn
ꢀ5
(
4x10 M) in aqueous solution (water/MeCN 99/1% v/v) irradiated at 248 nm, argonꢀ
saturated condition. Insert numbers indicate photoirradiation time in min.

Photochemical & Photobiological Sciences
Page 26 of 35
View Article Online
DOI: 10.1039/C7PP00200A
0
,0
-0,2
-0,4
-0,6
-0,8
0
2
4
6
8
10
t (min)
ꢀ5
Figure 2.Predn photodegradation (4x10 M) at different atmosphere conditions: (ꢀ) argon
saturatedꢀsolution; (ꢁ) air saturatedꢀsolution; () oxygen saturatedꢀsolution. Error bars
indicate one standard deviation.

Page 27 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
0,0
-
-
0,3
0,6
0
10
20
30
40
t (min)
ꢀ5
Figure 3. Effects of ROSꢀscavengers on Predn photoconsumption. (ꢁ) 4x10 M Predn;
ꢀ5
ꢀ5
(
ꢂ) 4x10 M Predn dark control; (ꢀ) 4x10 M Prend + 10 mM H O dark control; (ꢀ) 10
2 2
mM NaN ; () 1 mg% SOD; (ꢀ) 1 mg% CAT; (ꢁ) 0.2 M IsoOH. Error bars indicate one
3
standard deviation. All measurements were carried out at 298 K, except for CAT
experience (310 K).

Photochemical & Photobiological Sciences
Page 28 of 35
View Article Online
DOI: 10.1039/C7PP00200A
0
0
.9
.6
0
0
0
0
.9
.6
.3
.0
0
2
0
0
4
9
1
2
2
0.3
5
0
5
0
.0
200
225
250
λ
(nm)
2
00
220
240
260
λ (nm)
ꢀ4
ꢀ5
Figure 4. FFAc (1x10 M) photoconsumption in the presence of Predn (4x10 M) in
aqueous solution (water/MeCN 99/1% v/v) upon 248 nm irradiation in airꢀsaturated
ꢀ4
ꢀ5
conditions. Inset: FFAc (1x10 M) photo consumption in the presence of Predn (4x10 M)
in aqueous solution (water/MeCN 99/1% v/v) upon 248nm, irradiation in argonꢀsaturated
conditions. Numbers on the spectra represent the irradiation time in minutes.
ꢀ5
As blank solution it was used Predn (4x10 M) in aqueous solution photolyzed at the same
ꢀ
4
ꢀ5
times as the FFAc (1x10 M) / Predn (4x10 M) sample.

Page 29 of 35
Photochemical & Photobiological Sciences
View Article Online
DOI: 10.1039/C7PP00200A
1
Figure 5. Laser fluence (E ) dependence on the amplitude of O ( ꢁ ) phosphorescence
0
2
g
emission at zero time (I ) in MeCN. (●) PN; (○) Predn.
0
0
0
0
0
0
0
.0025
.0020
.0015
.0010
.0005
.0000
300
350
400
450
500
550
600
λ
(nm)
Figure 6. Transient absorption spectra of Predn in argonꢀsaturated solution in MeCN,
determined (●) 1ꢁs, (○) 4 ꢁs and ( ) 8 ꢁs after laser flash excitation at 266 nm.
■