Uric Acid: A Less-than-Perfect Probe for Singlet Oxygen

Article abstract of DOI:10.1111/php.12971

Uric acid and/or its monoanion has long been used as chemical-trapping agents to demonstrate the presence of singlet oxygen, O₂(a¹Δ_g), in aqueous systems. “Oxidative bleaching” of uric acid, generally monitored through changes in the uric acid absorption spectrum, is often used in support of claims for the intermediacy of O₂(a¹Δ_g). The bleaching of uric acid has also been used to quantify photosensitized O₂(a¹Δ_g) yields in selected systems. Unfortunately, experiments performed to these ends often neglect processes and phenomena that can influence the results obtained. For the present study, we experimentally examined the behavior of uric acid under a variety of conditions relevant to the photoinitiated creation and subsequent removal of O₂(a¹Δ_g). Although the oxidative destruction of uric acid can indeed be a useful tool in some cases, we conclude that caution must be exercised such as not to incorrectly interpret the data obtained.

Full text of DOI:10.1111/php.12971

DR. PETER R. OGILBY (Orcid ID : 0000-0003-0165-5159)
Article type : Special Issue Research Article
Uric Acid: A Less-than-Perfect Probe for Singlet Oxygen†
Mikkel Bregnhøj, Lea Dichmann,† Ciaran K. McLoughlin,† Michael Westberg, and Peter R.
Ogilby*
Department of Chemistry, Aarhus University, DK-8000 Aarhus, Denmark
† These authors contributed equally to this work
* Corresponding author’s email: progilby@chem.au.dk (Peter Ogilby)
†This article is part of a Special Issue celebrating Photochemistry and
Photobiology's 55th Anniversary.
ABSTRACT
Uric acid and/or its mono anion have long been used as chemical trapping agents to
demonstrate the presence of singlet oxygen, O2(a1g), in aqueous systems. “Oxidative
bleaching” of uric acid, generally monitored through changes in the uric acid absorption
spectrum, is often used in giort of claims for the intermediacy of O2(a1g). The bleaching of
uric acid has also been used to quantify photosensitized O2(a1g) yields in selected systems.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/php.12971
This article is protected by copyright. All rights reserved.

Unfortunately, experiments performed to these ends often neglect processes and phenomena
that can influence the results obtained. For the present study, we experimentally examined
the behavior of uric acid under a variety of conditions relevant to the photo-initiated creation
and subsequent removal of O2(a1g). Although the oxidative destruction of uric acid can
indeed be a useful tool in some cases, we conclude that caution must be exercised such as not
to incorrectly interpret the data obtained.
INTRODUCTION
Uric acid, UA, has long been used as a hydrophilic probe to assess whether or not
singlet oxygen, O2(a1g), is involved in a given reaction. To this end, the “oxidative
bleaching” of UA is commonly monitored through changes in the absorption spectrum of the
sample. Indeed, this practice is so prevalent, as reflected in a plethora of published papers,
that it is arguably inappropriate to support this statement with only one or two references to
the literature. Unfortunately, in many of these published experiments, the data obtained are
treated rather superficially leaving ample room for misinterpretation. In short, UA is
assumed to be the consummate quencher of O2(a1g) and much is often taken for granted.
For example, in photosensitized systems, UA could also influence the production of O2(a1g)
by quenching the sensitizer as well. Although some substantive studies of UA chemistry in
photo-initiated oxidizing environments have been published (1-3), many of the fundamental
concepts reported therein are often overlooked by the community. As such, we feel it is
appropriate to better clarify aspects of uric acid behavior under conditions in which O2(a1g)
is produced.
This article is protected by copyright. All rights reserved.

The fully protonated neutral form of uric acid, UAn, present under acidic conditions,
is shown in Figure 1. Reported pKa values for the production of the more hydrophilic mono-
anion, UAma, range from 5.4 to 5.8, with some disagreement as to which proton is actually
removed (1, 4-7). In Figure 1, we show the structures of the two mono-anions for which
there appears to be substantive support.
>Figure 1<
Once formed in an aqueous system that contains uric acid, O2(a1g) can be
removed/deactivated by several kinetically competing processes (Figure 2). Over the years, it
has been established that electron-rich compounds are particularly adept at
deactivating/removing O2(a1g) (8, 9). In general, among other things, the magnitude of rate
constants for chemical reaction, krxn, reflect the electrophilicity of O2(a1g) and can be as
large as ~ 109 s-1 M-1 (8, 10). Independently, the magnitude of rate constants, kq, for the
-
quenching of O2(a1g) to the ground state, O2(X3g ), are generally larger for molecules Q
-•
that can readily donate charge and facilitate mixing with the Q+•O2 charge-transfer state (8,
9). Likewise, values of kq can be as large as ~ 109 s-1 M-1. On this basis, one would expect
n
that the rate constants krxnn and kq for UAn would be smaller than krxnma and kqma for the more
electron-rich UAma. Specifically, as one changes the pH of an aqueous solution of uric acid,
bracketing the UA pKa value of ~ 5.6, one will change the relative concentrations of UAn and
UAma and, as such, should change measured values for the rate constants of O2(a1g)
removal/deactivation. This expectation has been confirmed by Montaña, et al. (1), who
showed that measured values of krxn and kq for the removal and deactivation of O2(a1g),
respectively, indeed increase as a solution of UA is made more alkaline. Given that the pKa
value of ~5.6 is sufficiently close to the range of pH values found in a mammalian cell (11),
This article is protected by copyright. All rights reserved.

this phenomenon can have significant ramifications in the interpretation of data recorded
from a host of biological systems.
> Figure 2 <
A variety of oxygenated products can be envisioned as a consequence of reactions
between O2(a1g) and UAma and UAn, certainly when one takes secondary and tertiary
reactions into account (12, 13). In Figure 1, we show one molecule, parabanic acid, that has
recently been identified as a specific product of the reaction between O2(a1g) and UA (12).
With respect to the use of UA as a probe for the presence of O2(a1g), a key point in this
regard is that these reactions change the absorption spectrum of the given sample over the
range ~ 200 – 350 nm. This is a spectral range over which many molecules M have
appreciable absorption that likewise changes as a result of oxidation reactions with O2(a1g)
to yield oxygenated products, MO2 (Figure 2).
Carrying this latter point further, it is important to recognize that UA will also react
with other reactive oxygen species (ROS) often found in O2(a1g)-containing systems (1, 2,
12, 14). In short, UA is not a O2(a1g)-selective probe, and these competing reactions can
appreciably complicate the system being examined. This point is often overlooked to the
detriment of mechanistic conclusions that may be drawn from a given system.
This article is protected by copyright. All rights reserved.

There is a great need for molecules that can selectively probe for O2(a1g) under a
variety of conditions. The advantages of using an optical spectroscopic approach in this
regard should be apparent. Although activity to this end is as old as O2(a1g) itself, it remains
an active field of research (15-20), certainly with respect to the goal of elucidating the
intracellular behavior of O2(a1g) in time- and spatially-resolved experiments. Sadly, it is our
impression that UA has acquired the stature of an “ideal” water-soluble probe for O2(a1g),
likely by virtue of longevity in its use. As such, UA is often used in indiscriminate ways
without concern for potential complications that could adversely influence data interpretation.
On the basis of these points, we set out to perform selected experiments to further
illustrate and establish guidelines for the effective use of uric acid as a probe for O2(a1g).
MATERIALS AND METHODS
Materials. Uric acid (UA), 9,10-anthracenediyl-bis(methylene)dimalonic acid (ADA), L-
tryptophan, meso-Tetrakis(4-sulfonatophenyl)porphine (TPPS), Flavin mononucleotide
(FMN), Rose Bengal (RB), and Methylene Blue (MB) were obtained from Sigma-Aldrich
and used as received. The sulfonated derivative of 1H-phenalenone (PNS) was synthesized as
described by Nonell et al. (21). D2O (99.9% D) was obtained from EurisoTop and H2O was
purified by filtration (Milli-Q system by Millipore Corporation).
H2O- and D2O-based phosphate buffer solutions (PBS) were prepared using
commercially available PBS tablets (Sigma-Aldrich). (Note: pD = pH + 0.4) The pD of D2O-
based solutions was adjusted with deuterochloric acid (DCl) and sodium deuteroxide
(NaOD). The pH of H2O-based solutions was similarly adjusted using HCl and NaOH.
This article is protected by copyright. All rights reserved.

Methods. In previous publications (22, 23), we describe the time-resolved instruments and
approaches used to (a) generate and optically detect O2(a1g), and (b) monitor sensitizer
excited states. All reported quantum yields of O2(a1g) production were measured using PNS
in D2O-PBS as the standard (Δ = 0.97 ± 0.06) (24).
Selected experiments were performed in D2O-based solutions, instead of H2O-based
solutions, to exploit the fact that the lifetime of O2(a1g) in D2O is appreciably longer than
that in H2O (8, 25). For time-resolved O2(a1g) phosphorescence experiments, the use of
D2O as a solvent thus makes it easier to accurately discern quencher-dependent kinetics on
the rising portion of the phosphorescence signal (i.e., events that influence the decay of the
O2(a1g) precursor).
RESULTS AND DISCUSSION
Absorption and Fluorescence Spectra
The absorption and fluorescence spectra of UA dissolved in a phosphate-buffered
H2O solution (pH = 7.4) are shown in Figure 3. Note that, at this pH, the spectra reflect
contributions from both UAn and UAma. This point is substantiated by the pH-dependent data
in Figure 4, where the absorption spectra obtained are consistent with a system of two species
in equilibrium.
> Figure 3 <
This article is protected by copyright. All rights reserved.

> Figure 4 <
As seen in Figure 3, the absorption spectra show three clearly resolved bands. This is
consistent with spectra recorded by Montaña, et al. (1), but differs from spectra recorded by
Fischer et al. (14), likely because the latter did not extend their measurement down to 200 nm
using a more dilute solution.
The recorded absorbance values scale linearly over the UA concentration range of ~ 0
– 300 M and pH range of ~ 2 – 10. Thus, under these conditions, solute aggregation does
not appear to be a problem. The data yield a molar extinction coefficient for the band with
λmax = 292 nm of (1.17 ± 0.07) × 104 M-1cm-1 at pH = 7.4.
The fluorescence spectrum we record has a band maximum at ~ 375 nm (Figure 3).
Using tryptophan in H2O-PBS (f = 0.12 ± 0.01) (26) as a fluorescent standard, we estimate
that the quantum efficiency of the fluorescent signal recorded from the UA solution at this pH
is < 0.0005.
The comparatively small fluorescence quantum yield from UA is consistent with the
fact that, over the years, the photo-oxidative bleaching of UA has commonly been monitored
in absorption experiments. The latter are generally performed using the UA band centered at
~ 292 nm. As we outline further below, this can be complicated given that many molecules
apt to be present in the system of interest likewise absorb light in this spectral region.
This article is protected by copyright. All rights reserved.

UA-Sensitized Production of O2(a1g)
Upon irradiation of an air-saturated phosphate buffered solution of UA at 295 nm, we
were not able to detect a time-resolved phosphorescence signal at 1275 nm characteristic of
-
the O2(a1g) - O2(X3g ) transition. For this experiment, we specifically prepared our buffer
solution using heavy water (pD = 7.8) to exploit the fact that the lifetime of O2(a1g) in D2O
is appreciably longer than that in H2O and, as such, the quantum efficiency of O2(a1g)
phosphorescence is greater (8, 25). Using the corresponding O2(a1g) phosphorescence
signal from a PNS sensitized reaction as a standard, we estimate that the quantum yield of
O2(a1g) production sensitized by UA under these conditions is < 0.001. These results,
combined with the weak fluorescence signal, lead us to infer that non-radiative deactivation
of UA excited states is comparatively rapid and efficient. This inference is consistent with
the results of recent computations (27).
Overall Rate Constant for UA-Mediated Removal of O2(a1g)
We performed experiments to quantify the rate constant, ktotal, for the overall
deactivation/removal of O2(a1g) mediated by UA. With reference to Figure 2, ktotal is equal
to the sum of the reactive and physical deactivation channels of O2(a1g) (i.e., ktotal = krxn + kq)
. As outlined above, krxn and kq will be different for UAn and UAma, and the measured value
of ktotal at any given pH will thus depend on the respective fractions, χn and χma, of UAn and
UAma present in the solution (eq 1).
(1)
This article is protected by copyright. All rights reserved.

We performed these experiments in the standard way using the time-resolved 1275
nm phosphorescence of O2(a1g) as a probe. Specifically, we recorded the O2(a1g) lifetime,
, as a function of the UA concentration in both D2O-based and H2O-based phosphate
buffer solutions and, independently, as a function of pD. In separate experiments, O2(a1g)
was produced using RB, MB, TPPS, PNS, and FMN as the sensitizers. Representative data
are shown in Figure 5.
> Figure 5 <
In D2O-PBS at pD = 7.8, our experiments yield ktotal = (3.6 ± 0.2) × 108 M-1s-1. This
number is in excellent agreement with the number reported by Kanofsky for experiments also
performed in a D2O-based buffer at pD = 7.8 (ktotal = (3.6 ± 0.2) × 108 M-1s-1) (28). A
corresponding measurement in H2O-PBS yields an identical value of ktotal = (3.5 ± 0.3) × 108
M-1s-1 confirming the expected absence of a solvent isotope effect on ktotal. These numbers
differ only slightly from the value 5.1 × 108 M-1s-1 published by Montaña, et al. (1) for a
reported pH value of 7 in H2O using a different experimental technique. At pD ~ 3, where
UA is present as UAn, ktotal drops to a value of (2.3 ± 0.3) × 107 M-1s-1 (Figure 5c). The value
of ktotal reported by Montaña, et al. (1), 1.8 × 108 M-1s-1, for an experiment at pH = 5 (i.e., pD
= 5.4) is somewhat larger. Nevertheless, this number of Montaña, et al. is consistent with our
titration curve shown in Figure 5c.
The comparatively large rate constant of ktotalma = (3.6 ± 0.2) × 108 M-1s-1 for neutral
and moderately alkaline solutions indicates that, as a probe added in modest concentrations,
UA will indeed kinetically compete with many other molecules in processes that
remove/deactivate O2(a1g) (8).
This article is protected by copyright. All rights reserved.

Rate Constants for the Removal of O2(a1g) by Reaction with UA: Sensitized
Experiments
As established above, monitoring UA disappearance in an absorption experiment can
be complicated because UA absorbs light at wavelengths where many other molecules also
absorb. For example, in protein-containing systems, tryptophan has a distinct absorption band
with λmax at ~ 298 nm, and the intensity of this band likewise changes upon the facile reaction
of tryptophan with O2(a1g) (29). Many O2(a1g) sensitizers also absorb at these same
wavelengths, and these molecules may likewise “bleach” upon irradiation.
However, even if other molecules in the system remain stable over the reaction
period, the spectral changes due to the reactions of UA alone can render analysis difficult.
We illustrate this point with the data shown in Figure 6. In these experiments, we monitored
the photosensitized bleaching of UA in both D2O- and H2O-based buffer solutions. We
ascertained that, over the course of the reaction, the sensitizer remained stable. As such, we
could subtract the absorption spectrum of the sensitizer to reveal the dynamic changes related
to UA alone.
> Figure 6 <
On the basis of the kinetic scheme in Figure 2, and the knowledge that the solvent-
mediated O2(a1g) deactivation channels are less efficient in D2O than in H2O (8, 9, 25), one
would expect that the rate of O2(a1g)-mediated UA change should be greater in the D2O-
based buffer. The data in Figure 6 confirm this expectation. Most importantly, however, it is
not just the reaction rates that distinguish the D2O-based data from the H2O-based data; the
spectra themselves show pronounced differences as a function of elapsed photolysis time.
This article is protected by copyright. All rights reserved.

We infer that these spectral differences reflect the rate at which the reaction products appear
and subsequently disappear. Specifically, the D2O/H2O isotope effect will principally affect
the rate of O2(a1g)-mediated UA disappearance and, hence, the rate of initial product
formation (e.g., hydroperoxides). However, it is reasonable to assume that many secondary
and tertiary oxidation products will be formed via thermal reactions that do not depend on the
solvent isotopic composition (e.g., hydroperoxide decomposition and subsequent
rearrangement/fragmentation reactions) (1, 12, 13). We indeed find that, after an initial
period of irradiation, samples stored in the dark continue to show changes in the absorption
spectrum. These subsequent reactions will contribute differently to the spectra in the D2O-
and H2O-based experiments because the observation times are different (Figure 6). We
observed no changes in the absorption spectrum for an unirradiated control sample kept in the
dark for 24 h (data not shown).
On this basis, we argue that any attempt to use data such as those in Figure 6 to
quantify krxn for O2(a1g) and UA can be subject to error. Indeed, in such an attempt, we
obtain appreciably different results upon analysis of the spectra in Figures 6a and 6b. In this
regard, and to our knowledge, there is no precedence for a solvent isotope effect on krxn for a
reaction involving O2(a1g). As outlined in subsequent sections, more accurate information
can only be obtained from such absorption spectra under the following conditions:
(a) Ensure that the spectra principally reveal changes in UA concentration alone.
(b) Ensure that the rate and amount of O2(a1g) production is known and that UA
does not also quench the photosensitizer.
This article is protected by copyright. All rights reserved.

(c) Ensure that, in the nascent reactions of a photoinitiated process, O2(a1g) is
selectively produced at the expense of other ROS.
Rate Constants for the Removal of O2(a1g) by Reaction with UA: Sensitizer-Free
Experiments
As indicated above, if one wants to accurately quantify O2(a1g)-dependent changes
in UA concentration, one should ideally perform the experiment under conditions where
O2(a1g) is exclusively produced in controlled amounts at the expense of other ROS.
Because many sensitizers are known to concomitantly produce more than one ROS upon
excitation (i.e., O2(a1g) via energy transfer and the superoxide ion via electron transfer) (9,
30), one would ideally want an independent sensitizer-free approach to make O2(a1g).
Moreover, removing the sensitizer decreases the potential complexity of the experiment; even
if the sensitizer does not immediately react with O2(a1g), it may respond to secondary
reactions (e.g., radical propagation processes) and still influence the spectra recorded.
We have recently shown that experimentally useful amounts of O2(a1g) can be
selectively produced in sensitizer-free experiments by excitation of oxygen to its second
+
-
+
excited state, O2(b1g ), through the O2(X3g ) → O2(b1g ) transition at 765 nm (23, 31, 32).
+
O2(b1g ) subsequently decays with near unit efficiency to O2(a1g).
This article is protected by copyright. All rights reserved.

In Figure 7, we show absorption spectra of a sensitizer-free sample of UA in D2O-
-
PBS recorded as a function of elapsed irradiation into the spectrally narrow O2(X3g ) →
+
O2(b1g ) band at 765 nm. Identical samples that were irradiated at 800 nm, or kept in the
dark, did not show detectable changes in the absorption spectrum over the course of 24 hours.
Thus, we infer that our results from the samples irradiated at 765 nm are not the result of
thermal reactions or multi-photon processes involving UA. The data in Figure 7 also show
that, for sensitizer-free samples irradiated over a comparatively long time period at 765 nm,
oxidation products contribute much less to the absorption spectra.
> Figure 7 <
Using the extinction coefficient of oxygen at 765 nm in water ((13.4 ± 1.6) × 10-4 M-
1cm-1) (31), and the concentration of oxygen in air-saturated D2O-PBS at 23 °C (0.29 ± 0.01
mM) (33), the data in Figure 7 yield a rate constant for the reaction of O2(a1g) with UA of
(1.5 ± 0.2) × 108 M-1s-1. Control experiments in which we monitored the O2(a1g)-mediated
bleaching of UA relative to the bleaching of ADA (krxn = 5.6 ± 0.3 × 107 M-1s-1) (31), gave
identical results in both H2O- and D2O-based buffered solutions. This latter observation is
consistent with our expectations regarding a solvent isotope effect on krxn (vide supra).
> Table 1 <
Our values of krxn thus obtained (Table 1) are similar to those reported by Montaña, et
al. (1) for UA in H2O at pH = 5 (1 × 107 M-1s-1) and pH = 7 (1.2 × 108 M-1s-1). The data of
Montaña, et al. were obtained from a RB-sensitized system using an approach based on
oxygen consumption with furfuryl alcohol as a reference reactant.
This article is protected by copyright. All rights reserved.

The Quenching of Sensitizer Triplet States by Uric Acid
Under ideal circumstances, the optimal probe for O2(a1g) will selectively quench
and/or react with O2(a1g), and O2(a1g) alone. Thus, in a photosensitized reaction, where the
sensitizer triplet state is the immediate precursor of O2(a1g), an ideal probe for O2(a1g) will
not quench the sensitizer triplet state. If such quenching occurs, it will influence both the
yield and kinetics of O2(a1g) formation.
We performed a series of time-resolved O2(a1g) phosphorescence studies to
investigate the effects of UA on the O2(a1g) precursor. We were particularly interested in
the effect of UA on the kinetics of O2(a1g) formation (i.e., the decay rate of the sensitizer
triplet state). In Figure 8, we show pertinent O2(a1g) phosphorescence data obtained for a
PNS-sensitized system in a D2O-based buffer solution at pD = 7.8. We recorded data as
function of the incident laser power and the concentration of UA in the solution. In all cases,
we could fit the phosphorescence trace using a difference of two exponential functions (25):
one function describing the rise of the O2(a1g) signal and a second describing the fall of the
signal. As such, the decay of the O2(a1g) precursor (i.e., the sensitizer triplet state) followed
first order kinetics, and the decay of O2(a1g) itself likewise followed first order kinetics.
Using standard procedures (34, 35), we also used the integral of these traces to obtain the
quantum yield of photosensitized O2(a1g) production, , as a function of UA concentration.
> Figure 8 <
In these studies, we ascertained that the kinetics of O2(a1g) formation and decay did
not change upon increasing the incident laser power. Furthermore, the absorption spectrum of
the sample did not change appreciably over the course of the experiment (i.e., the change in
This article is protected by copyright. All rights reserved.

absorbance at the UA λmax of 292 nm was always less than 2%). As such, transport of UA in
and out of the irradiated volume was sufficiently facile in our stirred cuvette to maintain a
stable concentration of UA throughout the experiment.
We plotted the kinetic parameters obtained from the data in Figure 8 against the
concentration of UA (Figure 9). From these plots, we obtain (a) the rate constant for total
quenching of O2(a1g) by UA, ktotal, from the UA-dependent changes in the fall of the
O2(a1g) signal (e.g., Figure 5b), (b) the rate constant for the quenching of the sensitizer
triplet state by UA, kT, from the UA-dependent changes in the rise of the O2(a1g) signal
(Figure 9a), and (c) the concomitant changes in the quantum yield of O2(a1g) production, Δ,
from the integrated signal intensity normalized by the O2(a1g) lifetime (Figure 9b).
> Figure 9 <
As expected for a system in which UA quenches the sensitizer triplet state, the yield
of O2(a1g) drops appreciably as the concentration of added UA is increased. Specifically, Δ
for PNS drops from ~ 1 to ~ 0.5 in the presence of just 200 µM of UA. Moreover, the
measured rate constant for quenching of the sensitizer triplet state (just like the rate constant
for O2(a1g) quenching) varies considerably as a function of pD (Figure 9c). Thus, we infer
that UAn and UAma show different quenching capabilities toward 3PNS. From the data in
n
Figure 9c, we find kT = (2.7 ± 0.3) × 109 M-1s-1 and kTma = (1.5 ± 0.2) × 109 M-1s-1.
We performed experiments analogous to those with PNS using other common water-
soluble sensitizers. Data obtained from these experiments are shown in Table 2.
This article is protected by copyright. All rights reserved.

> Table 2 <
Flavins are important O2(a1g) photosensitizers with a host of biological ramifications
(6). It has been shown that UA quenches the riboflavin triplet state, and that this process
involves electron transfer from UA to riboflavin (3). Among other things, the resultant
-
riboflavin radical anion will reduce O2(X3g ) to the superoxide radical anion, which, in turn,
can react with UA (1). Thus, in a riboflavin-sensitized experiment, the use of UA as a probe
for O2(a1g) will not only disturb the yields and kinetics of O2(a1g), it will also result in the
production of other oxidizing species whose reactions with UA will complicate the system
even further.
We obtain a value of kT for FMN at pD = 7.8 of (1.6 ± 0.2) × 109 M-1s-1. Given that
the chromophore in FMN is identical to that in riboflavin, it is reassuring to note that our
number is similar to that reported by Cardoso, et al. (3), (2.90 ± 0.14) × 109 M-1s-1, for
riboflavin at pH = 6.4. It is important to note that the mechanism by which UA quenches
3FMN (i.e., electron transfer) yields radical ions that can ultimately result in the production of
-.
other ROS (specifically O2 ) (1, 3). As such, at least in FMN photosensitized systems, care
-.
must be exercised in interpreting relative UA-dependent yields of O2(a1g) and O2 (38).
Of the five O2(a1g) sensitizers examined, we find that kT for TPPS is uniquely small
(Table 2). Although we were unable to find literature data for the triplet state energy, ET, of
UA, this value of kT could nevertheless reflect an energy mismatch between values of ET for
UA and TPPS (39). In contrast, kT for MB is comparatively large, despite an ET similar to
that of TPPS. In this case, it is reasonable to speculate that the unique redox properties of
MB (40, 41) play a more dominant role in the interaction with UA. The important point in
this regard is that the rate of UA-mediated triplet state deactivation depends on the sensitizer.
This article is protected by copyright. All rights reserved.

Our data clearly show that care must be exercised if one intends to obtain the quantum
yield of photosensitized O2(a1g) production using a procedure that depends on quenching by
UA. In this regard, we note that such a procedure, in fact, has been promoted (14, 42, 43).
However, our results indicate that one should expect such a procedure to yield values of Δ
that depend on (a) the rate constant for sensitizer triplet state quenching by UA, (b) the initial
concentration of UA used, and (c) the pH of the solution.
Uric Acid as a Probe for O2(a1g)
UA shows properties that, at a first glance, appear to make it a suitable water-soluble
probe for O2(a1g): it has a high reactivity toward O2(a1g), is generally biocompatible, and
does not itself sensitize the production of O2(a1g).
However, as we have established, UA has a number of undesirable features:
First, with the common approach of monitoring UA disappearance via absorption
spectroscopy, the data obtained can be appreciably influenced by absorbance changes due to
the time-dependent composition of UA oxidation products and/or other molecules in the
system. As such, one may have to revert to other, more elaborate, quantification techniques.
Second, UA is not a probe specific for O2(a1g); it reacts with a host of other species,
including many common ROS (1, 2, 12, 14). Thus, the bleaching of UA alone cannot be used
to claim the intermediacy of O2(a1g) in a given reaction. This is a major drawback, certainly
This article is protected by copyright. All rights reserved.

if one intends to use UA in a complicated biological system where other ROS are apt to be
present.
Third, UA can perturb and quench the excited states of other molecules present in the
system. Specifically, when UA is present, the photophysics of selected sensitizers may not be
well defined, and care must be exercised to obtain an accurate quantum yield of
photosensitized O2(a1g) production.
Finally, with all of the above points in mind, the extent to which UA perturbs a given
system depends on pH. Since the pertinent pKa value of UA is ~ 5.6, this can have
appreciable ramifications for experiments performed “indiscriminately” in the biologically
relevant pH range of ~ 5-7.
CONCLUSIONS
Uric acid is often regarded as one of the consummate water soluble quenchers of
O2(a1g). As such, it is frequently used to assess whether or not O2(a1g) is involved as a
reactive intermediate. Moreover, the rate of uric acid removal, as recorded through changes
in its absorption spectrum, has been touted as a convenient method to quantify yields of
O2(a1g) in selected systems. Despite the level of certainty now often taken for granted in this
regard, we hope to have established that care must be exercised when using uric acid as a
quantitative probe for O2(a1g), certainly in selected photosensitized reactions.
This article is protected by copyright. All rights reserved.

Although other water-soluble probes for O2(a1g) are available (16, 17, 19, 20, 31),
these molecules likewise have selected disadvantages. As such, there is still a need to
develop and characterize a wider range of O2(a1g) probes.
ACKNOWLEDGEMENTS
Grants from the Danish National Research Foundation (Center for Oxygen
Microscopy and Imaging), the Danish Research Council, and the Danish Ministry of Higher
Education in Science (LaserLab.dk) supported the work described herein. The authors thank
Jan Thøgersen (Aarhus University) for technical assistance.
REFERENCES
1. Montaña, M. P., W. A. Massad, F. Amat-Guerri and N. A. Garcia (2008) Scavenging of
Riboflavin-Photogenerated Oxidative Species by Uric Acid, Xanthine or Hypoxanthine:
A Kinetic Study. J. Photochem. Photobiol. A. Chem. 193, 103-109.
2. Sueishi, Y., M. Hori, M. Ishikawa, K. Matsu-ura, E. Kamogawa, Y. Honda, M. Kita and
K. Ohara (2014) Scavenging Rate Constants of Hydrophilic Antioxidants against
Multiple Reactive Oxygen Species. J. Clin. Biochem. Nutr. 54, 67-74.
3. Cardoso, D. R., P. Homem-de-Mello, K. Olsen, A. B. F. Da Silva, D. W. Franco and L. F.
Skibsted (2005) Deactivation of Triplet-Excited Riboflavin by Purine Derivatives:
Important Role of Uric Acid in Light-Induced Oxidation of Milk Sensitized by
Riboflavin. J. Agric. Food Chem. 53, 3679-3684.
4. Simic, M. G. and S. V. Jovanovic (1989) Antioxidation Mechanisms of Uric Acid. J. Am.
Chem. Soc. 1989, 5778-5782.
This article is protected by copyright. All rights reserved.

5. Jimenez, V. and J. B. Alderete (2005) Theoretical Calculations on the Tautomerism of
Uric Acid in Gas Phase and Aqueous Solution. J. Mol. Struct.: THEOCHEM 755, 209-
214.
6. Halliwell, B. and J. M. C. Gutteridge (2015) Free Radicals in Biology and Medicine. 5th
Edition. Oxford University Press, Oxford.
7. Voet, D. and J. G. Voet (2004) Biochemistry. John Wiley and Sons, New York.
8. Wilkinson, F., W. P. Helman and A. B. Ross (1995) Rate Constants for the Decay and
Reactions of the Lowest Electronically Excited Singlet State of Molecular Oxygen in
Solution. An Expanded and Revised Compilation. J. Phys. Chem. Ref. Data 24, 663-
1021.
9. Schweitzer, C. and R. Schmidt (2003) Physical Mechanisms of Generation and
Deactivation of Singlet Oxygen. Chem. Rev. 103, 1685-1757.
10. Clennan, E. L. and A. Pace (2005) Advances in Singlet Oxygen Chemistry. Tetrahedron
61, 6665-6691.
11. Casey, J. R., S. Grinstein and J. Orlowski (2010) Sensors and Regulators of Intracellular
pH. Nature Reviews 11, 50-61.
12. Iida, S., Y. Ohkubo, Y. Yamamoto and A. Fujisawa (2017) Parabanic Acid is the Singlet
Oxygen Specific Oxidation Product of Uric Acid. J. Clin. Biochem. Nutr. 61, 169-175.
13. Matsuura, T. and I. Saito (1968) Photoinduced Reactions XXI: Photosensitized
Oxygenation of N-Substituted Hydroxypurines. Tetrahedron 24, 6609-6614.
14. Fischer, F., G. Graschew, H.-J. Sinn, W. Maier-Borst, W. J. Lorenz and P. M. Schlag
(1998) A Chemical Dosimeter for the Determination of the Photodynamic Activity of
Photosensitizers. Clinica Chimica Acta 274, 89-104.
This article is protected by copyright. All rights reserved.

15. Setsukinai, K., Y. Urano, K. Kakinuma, H. J. Majima and T. Nagano (2003)
Development of Novel Fluorescence Probes that can Reliably Detect Reactive Oxygen
Species and Distinguish Specific Species. J. Biol. Chem. 278, 3170-3175.
16. Lemp, E. and A. L. Zanocco (2016) Singlet Oxygen Chemical Acceptors. In Singlet
Oxygen: Applications in Biosciences and Nanosciences 2 (Edited by S. Nonell and C.
Flors), pp 83-101. Royal Society of Chemistry, Cambridge.
17. Ruiz-Gonzalez, R. and A. L. Zanocco (2016) Singlet Oxygen Fluorescent Probes. In
Singlet Oxygen: Applications in Biosciences and Nanosciences 2 (Edited by S. Nonell
and C. Flors), pp 103-120. Royal Society of Chemistry, Cambridge.
18. Pimenta, F. M., J. K. Jensen, M. Etzerodt and P. R. Ogilby (2015) Protein-encapsulated
bilirubin: paving the way to a useful probe for singlet oxygen. Photochem. Photobiol.
Sci. 14, 665-677.
19. Pedersen, S. K., J. Holmehave, F. H. Blaikie, A. Gollmer, T. Breitenbach, H. H. Jensen
and P. R. Ogilby (2014) Aarhus Sensor Green: A Fluorescent Probe for Singlet Oxygen.
J. Org. Chem. 79, 3079-3087.
20. Frausto, F. and S. W. Thomas (2017) Ratiometric Singlet Oxygen Detection in Water
Using Acene-Doped Conjugated Polymer Nanoparticles. ACS Appl. Mater. Interfaces 9,
15768-15775.
21. Nonell, S., M. Gonzalez and F. R. Trull (1993) 1H-Phenalen-1-one-2-sulfonic acid: An
Extremely Efficient Singlet Molecular Oxygen Sensitizer for Aqueous Media. Afinidad
448, 445-450.
22. Arnbjerg, J., M. Johnsen, P. K. Frederiksen, S. E. Braslavsky and P. R. Ogilby (2006)
Two-Photon Photosensitized Production of Singlet Oxygen: Optical and Optoacoustic
Characterization of Absolute Two-Photon Absorption Cross Sections for Standard
Sensitizers in Different Solvents. J. Phys. Chem. A 110, 7375-7385.
This article is protected by copyright. All rights reserved.

23. Bregnhøj, M., A. Blázquez-Castro, M. Westberg, T. Breitenbach and P. R. Ogilby
(2015) Direct 765 nm Optical Excitation of Molecular Oxygen in Solution and in Single
Mammalian Cells. J. Phys. Chem. B 119, 5422-5429.
24. Marti, C., O. Jürgens, O. Cuenca, M. Casals and S. Nonell (1996) Aromatic Ketones as
Standards for Singlet Molecular Oxygen O2(1g) Photosensitization. Time-Resolved
Photoacoustic and Near-IR Emission Studies. J. Photochem. Photobiol., A. Chem. 97,
11-18.
25. Ogilby, P. R. (2010) Singlet Oxygen: There is Indeed Something New Under the Sun.
Chem. Soc. Rev. 39, 3181-3209.
26. Brouwer, A. M. (2011) Standards for photoluminescence quantum yield measurements
in solution (IUPAC Technical Report). Pure Appl. Chem. 83, 2213-2228.
27. Yamazaki, S., S.-H. Urashima, H. Saigusa and T. Taketsugu (2014) Ab Initio Studies on
the Photophysics of Uric Acid and its Monohydrates: Role of the Water Molecule. J.
Phys. Chem. A 118, 1132-1141.
28. Kanofsky, J. R. (1990) Quenching of Singlet Oxygen by Human Plasma. Photochem.
Photobiol. 51, 299-303.
29. Jensen, R. L., J. Arnbjerg and P. R. Ogilby (2012) Reaction of Singlet Oxygen with
Tryptophan in Proteins: A Pronounced Effect of the Local Environment on the Reaction
Rate. J. Am. Chem. Soc. 134, 9820-9826.
30. Silva, E. F. F., C. Serpa, J. M. Dabrowski, C. J. P. Monteiro, S. J. Formosinho, G.
Stochel, K. Urbanska, S. Simoes, M. M. Pereira and L. G. Arnaut (2010) Mechanisms of
Singlet Oxygen and Superoxide Ion Generation by Porphyrins and Bacteriochlorins and
their Implications in Photodynamic Therapy. Chem. Eur. J. 16, 9273-9286.
This article is protected by copyright. All rights reserved.

31. Bregnhøj, M., M. V. Krægpøth, R. J. Sørensen, M. Westberg and P. R. Ogilby (2016)
-
+
Solvent and Heavy-Atom Effects on the O2(X3g ) - O2(b1g ) Absorption Transition. J.
Phys. Chem. A 120, 8285-8296.
32. Bregnhøj, M., M. Westberg, F. Jensen and P. R. Ogilby (2016) Solvent-Dependent
Singlet Oxygen Lifetimes: Temperature Effects Implicate Tunneling and Charge-
Transfer Interactions. Phys. Chem. Chem. Phys. 18, 22946-22961.
33. Westberg, M., M. Bregnhøj, M. Etzerodt and P. R. Ogilby (2017) Temperature Sensitive
Singlet Oxygen Photosensitization by LOV-Derived Fluorescent Flavoproteins. J. Phys.
Chem. B 121, 2561-2574.
34. Scurlock, R. D., D. O. Mártire, P. R. Ogilby, V. L. Taylor and R. L. Clough (1994)
Quantum Yield of Photosensitized Singlet Oxygen (a1g) Production in Solid
Polystyrene. Macromolecules 27, 4787-4794.
35. Wilkinson, F., W. P. Helman and A. B. Ross (1993) Quantum Yields for the
Photosensitized Formation of the Lowest Electronically Excited Singlet State of
Molecular Oxygen in Solution. J. Phys. Chem. Ref. Data 22, 113-262.
36. Montalti, M., A. Credi, L. Prodi and M. T. Gandolfi (2006) Handbook of
Photochemistry. Third. CRC Press, Boca Raton.
37. Schmidt, R., C. Tanielian, R. Dunsbach and C. Wolff (1994) Phenalenone, a Universal
Reference Compound for the Determination of Quantum Yields of Singlet Oxygen
Sensitization. J. Photochem. Photobiol., A: Chem. 79, 11-17.
38. Barnett, M. E., T. M. Baran, T. H. Foster and A. P. Wojtovich (2018) Quantification of
Light-Induced miniSOG Superoxide Production using the Selective Marker, 2-
Hydroxyethidium. Free Rad. Biol. Med. 116, 134-140.
39. Sandros, K. (1964) Transfer of Triplet State Energy in Fluid Solutions. III. Reversible
Energy Transfer. Acta Chem. Scand. 18, 2355-2374.
This article is protected by copyright. All rights reserved.

40. Junqueira, H. C., D. Severino, L. G. Dias, M. S. Gugliotti and M. S. Baptista (2002)
Modulation of Methylene Blue Photochemical Properties based on Adsorption at
Aqueous Micelle Interfaces. Phys. Chem. Chem. Phys. 4, 2320-2328.
41. Jahnke, L. S. and A. W. Frenkel (1978) Photooxidation of Epinephrine Sensitized by
Methylene Blue - Evidence for the Involvement of Singlet Oxygen and of Superoxide.
Photochem. Photobiol. 28, 517 - 523.
42. Rabello, B. R., A. P. Gerola, D. S. Pellosi, A. L. Tessaro, J. L. Aparicio, W. Caetano and
N. Hioka (2012) Singlet Oxygen Dosimetry using Uric Acid as a Chemical Probe:
Systematic Evaluation. J. Photochem. Photobiol. A. Chem. 238, 53-62.
43. Gerola, A. P., J. Semensato, D. S. Pellosi, V. R. Batistela, B. R. Rabello, N. Hioka and
W. Caetano (2012) Chemical Determination of Singlet Oxygen from Photosensitizers
Illuminated with LED: New Calculation Methodology Considering the Influence of
Photobleaching. J. Photochem. Photobiol. A. Chem. 232, 14-21.
This article is protected by copyright. All rights reserved.

Table 1. Measured rate constants for the interaction of UAn and UAma with O2(a1g) in D2O.
Total O2(a1g)
deactivation
ktotal (M-1s-1)
Reactive O2(a1g)
deactivation
Physical O2(a1g)
deactivation
kq (M-1s-1)b
Compound
krxn (M-1s-1)a
UAn (pD = 4)
(2.3 ± 0.3) × 107
(3.6 ± 0.2) × 108
(2.3 ± 0.3) × 107
(1.5 ± 0.2) × 108
~ 0
UAma (pD = 7.8)
~ 2 × 108
aObtained in sensitizer-free experiments.
bObtained from ktotal – krxn.
Table 2. Sensitizer triplet state energy, ET, rate constant for sensitizer triplet state quenching
by UA, kT, and the quantum yield of O2(a1g) production, Δ, at two different UA
concentrations.
Quantum yield of O2(a1g)
Sensitizer
Triplet state Rate constant of triplet
a
energy,
ET (kJ/mol)
209b
state quenching by UA,
kT (109 M-1s-1)a
1.6 ± 0.2
production, φΔ
[UA] = 0 µM
0.67 ± 0.05
[UA] = 200 µM
0.32 ± 0.03
FMN
PNS
RB
182c
164d
138d
138e
1.5 ± 0.2
1.4 ± 0.2
3.9 ± 0.3
< 0.05
0.97 ± 0.06
0.76 ± 0.05
0.52 ± 0.05
0.73 ± 0.06
0.51 ± 0.04
0.52 ± 0.04
0.26 ± 0.03
0.72 ± 0.06
MB
TPPS
aData were recorded in D2O-PBS at pD = 7.8 and thus reflect the effects of UAma.
bData for riboflavin (36).
cData for unsulfonated 1H-phenalenone (PN) (37).
dFrom reference (36).
eData for unsulfonated tetraphenyl-porphyrin (TPP) (36).
This article is protected by copyright. All rights reserved.

Figure Captions:
Figure 1. Chemical structures of the neutral form of uric acid (left), the most likely mono-
anions (right), and the reported principal product of reaction with O2(a1g),
parabanic acid (bottom).
Figure 2. Scheme illustrating the kinetically competing pathways for the
removal/deactivation of O2(a1g) in a solvent, S, that contains both a reactive
molecule, M, the neutral form of uric acid, UAn, and the uric acid mono-anion,
UAma. Note that the bimolecular rate constants for physical quenching, kq, and
chemical reaction, krxn, depend on the molecule involved. In most solvents,
certainly in H2O, the solvent-mediated channel for non-radiative O2(a1g)
deactivation dominates over the channel for radiative deactivation (i.e., knr >> kr).
Figure 3. Normalized absorption and fluorescence spectra recorded from a phosphate
buffered H2O-solution of uric acid (pH = 7.4). The fluorescence spectrum has a

max of ~ 375 nm (excitation wavelength = 265 nm).
Figure 4. Absorption spectra of uric acid in H2O recorded as a function of pH over a range
that brackets the first pKa value of ~ 5.6. The spectrum of UAma is red-shifted
relative to the spectrum of UAn, and isosbestic points are clearly visible. The latter
are consistent with an equilibrium between UAn and UAma.
Figure 5. (a) Representative time-resolved O2(a1g) phosphorescence traces recorded at
1275 nm upon 400 nm pulsed laser irradiation of FMN in D2O-PBS at pD = 7.8.
-1
(b) The inverse lifetime of O2(a1g), τΔ , obtained from the time-resolved traces
plotted as a function of UA concentration (Note: τΔ-1 = (kr + knr)[solvent] +
ktotal[UA], see text). (c) The rate constant for total O2(a1g) removal, ktotal, as a
function of pD in D2O-based solutions. The solid line is a numerical fit to eq 1. The
dashed lines indicate the values of ktotal for the neutral and mono-anionic forms of
UA, respectively. The inflection point in this titration curve yields a pKa of 5.7 ±
0.2. The data point at pD = 9.6 is likely influenced by the start of the titration
curve for the second pKa, and was excluded from the fit.
This article is protected by copyright. All rights reserved.

Figure 6. Spectra of phosphate buffered aqueous solutions of UA. Spectra were recorded as a
function of the elapsed irradiation time of the O2(a1g) sensitizer methylene blue
(MB). We independently ascertained that MB does not bleach under these
conditions. Hence, the MB spectrum was subtracted to yield the data shown. (a)
Experiment performed in D2O-PBS (pD = 7.8). (b) Experiment performed in H2O-
PBS (pH = 7.4).
Figure 7. Absorption spectra of UA in D2O-PBS as a function of elapsed irradiation into the
-
+
O2(X3g ) → O2(b1g ) band at 765 nm (fs laser spectral bandwidth of ~ 8 nm
fwhm). The incident laser power was 2.52 W.
Figure 8. Representative time-resolved O2(a1g) phosphorescent traces recorded upon
pulsed laser excitation of PNS in D2O-PBS. (a) Traces recorded at six different
concentrations of added UA. (b) Expanded portion of the traces in Figure 8a to
more clearly show the UA-dependent changes on the rising portion of the signal
(i.e., the part of the trace that principally reflects the lifetime of the O2(a1g)
precursor).
Figure 9. Pertinent parameters for the UA-mediated quenching of 3PNS in D2O. (a) The
inverse of the PNS triplet state lifetime, τT-1 = kT, obtained from the rise of the
O2(a1g) phosphorescence signal in Figure 8. (b) The quantum yield of O2(a1g)
production, Δ, obtained from the integral of the O2(a1g) phosphorescence signal
normalized by the O2(a1g) lifetime. (c) The rate constant of triplet state
quenching, kT, obtained from the slope of plots such as the one shown in panel (a),
as a function of pD. The data point at pD = 9.6 is likely influenced by the start of
the titration curve for the second pKa, and was excluded from the fit.
This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.