Towards multifunctional antioxidants: Synthesis, electrochemistry, in vitro and cell culture evaluation of compounds with ligand/catalytic properties

Article abstract of DOI:10.1039/b503282m

Numerous human diseases are linked to a biochemical condition known as oxidative stress (OS). Antioxidants are therefore becoming increasingly important as potential disease prevention and therapeutic agents. Since OS is a multi-stressor event, agents combining a range of different antioxidant properties, such as redox catalysis and metal binding, might be more effective and selective than mono-functional agents. Selenium derivatives of aniline and pyridine combine redox activity with metal binding properties. These multifunctional agents have a distinct electrochemical profile, and exhibit good catalytic activity in the glutathione peroxidase mimic and metallothionein assays. They also show antioxidant activity in a skin cell model of UVA-induced stress. These compounds might therefore provide the basis for novel agents combining two or more distinct antioxidant properties. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b503282m

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Towards multifunctional antioxidants: synthesis, electrochemistry,
in vitro and cell culture evaluation of compounds with
ligand/catalytic properties
Catriona A. Collins,a Fiona H. Fry,a,b Andrea L. Holme,a Anthie Yiakouvaki,c
Abdullah Al-Qenaei,c Charareh Pourzandc and Claus Jacob*a,b,d
a The Biocatalysis Centre, School of Biological and Chemical Sciences, University of Exeter,
Stocker Road, Exeter, UK EX4 4QD. E-mail: C.Jacob@ex.ac.uk; Fax: +44 (0)1392 263434;
Tel: +44 (0)1392 263462
b Exeter Antioxidant Therapeutics Ltd., The Innovation Centre, Rennes Drive, Exeter, UK
c Department of Pharmacy and Pharmacology, University of Bath, Bath, UK BA2 7AY
d Fachbereich 8.2 Pharmazeutische und Medizinische Chemie, Universita¨t des Saarlandes,
Postfach 151150, 66041 Saarbru¨cken, Germany
Received 4th March 2005, Accepted 4th March 2005
First published as an Advance Article on the web 18th March 2005
Numerous human diseases are linked to a biochemical condition known as oxidative stress (OS). Antioxidants are
therefore becoming increasingly important as potential disease prevention and therapeutic agents. Since OS is a
multi-stressor event, agents combining a range of different antioxidant properties, such as redox catalysis and metal
binding, might be more effective and selective than mono-functional agents. Selenium derivatives of aniline and
pyridine combine redox activity with metal binding properties. These multifunctional agents have a distinct
electrochemical profile, and exhibit good catalytic activity in the glutathione peroxidase mimic and metallothionein
assays. They also show antioxidant activity in a skin cell model of UVA-induced stress. These compounds might
therefore provide the basis for novel agents combining two or more distinct antioxidant properties.
Introduction
Results
Oxidative stress (OS) is a biochemical condition associated with
numerous human diseases, ranging from rheumatoid arthritis
and neurodegenerative diseases to inflammation and cancer.1,2
It is characterised by a significantly increased concentration of
intracellular oxidising species, such as reactive oxygen species
(ROS), which is often accompanied by the simultaneous loss
of antioxidant defence capacity. Effective antioxidants able to
counteract OS are therefore becoming increasingly important
in disease prevention and therapy. Among them, compounds
with glutathione peroxidase (GPx)-like activity are particularly
interesting, since they catalytically remove oxidative stressors,
and can therefore be applied in small quantities. In addition,
catalysts are able to respond to their redox environment, hence
providing selectivity for cells under OS. Not surprisingly, one
such catalyst, ebselen, has already made its way into clinical
trials.3
OS is not, however, a single stressor event, and a wide range
of stressors, such ROS (e.g. oxygen-based radicals, peroxides),
nitric oxide, peroxynitrite and free metal ions are known to form
part of its biochemical makeup. To respond to this complex
multi-stressor environment more effectively, compounds can be
envisaged that combine a range of antioxidant activities in one,
chemically simple molecule. Since ROS and adventitious, free
metal ions, such as iron and copper, play a major role in OS,
antioxidants able to interact with several of these species might
be particularly active. Ultimately, such compounds might not
only be very effective, but also quite selective for the particular
OS stress environment they are tailored to.
Selection and synthesis of multifunctional agents
Sulfur and selenium compounds are known to play a major role
in antioxidant reactions. While sulfur agents can serve as electron
donors, organoselenium agents frequently possess GPx-like
activity. In addition, sulfur and nitrogen agents have excellent
metal binding properties, and can bind adventitious metal
ions such as copper and iron. As a consequence, dichalcogen
derivatives of known metal binding agents were chosen for this
study, since they promise to be redox active under physiological
conditions, form thiols and selenols upon reduction in vivo,
catalyse the reduction of ROS, and also interact with redox
active metal ions. The chemical structures of these compounds
are shown in Fig. 1.
Compounds 1 and 3 were successfully synthesised according
to the method of Bhasin et al. (see Experimental section).5
The other compounds were commercially available. These
compounds were then evaluated for the following antioxidant
activities: (a) electron donor/acceptor properties, (b) catalytic
activity, (c) metal binding properties and (d) protection of cells
from UVA-radiation. These studies required a combination of
chemical and biochemical techniques, such as electrochemistry,
protein and metal binding assays, and ultimately cell culture
studies. In the following, the results are presented according to
the potential antioxidant properties investigated.
Redox properties of dichalcogens
Here we present initial results from in vitro and cell culture
studies of chemically simple agents that combine redox, catalytic
and metal binding sites, and might therefore act as multifunc-
tional antioxidants.4 We are able to show that the dichalcogen
derivatives of either pyridine, aniline or quinoline combine the
redox activity of the chalcogen moiety with the metal binding
properties characteristic of their nitrogen and chalcogen groups.
Generally, each of the compounds studied by Cyclic Voltam-
metry (CV) on the mercury electrode exhibited one characte-
ristic, reversible electron transfer typical of a thiol/disulfide or
selenol/diselenide couple. In line with previous findings, poten-
tials for the latter were around 100 to 200 mV more negative
compared to the thiol/disulfide couple. The redox couples had
an anodic and cathodic peak (Epa and Epc, respectively) in the
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 4 1 – 1 5 4 6
1 5 4 1
©

View Article Online
Fig. 1 (a) Chemical structures of compounds used in this study. (b)
Synthetic pathway for diselenides.5
biologically relevant range between 0 and −900 mV vs. SSE,
approximately corresponding to a one electron transfer process
(DEp,1/2 = 3.53 RT/nF = 90.6/n mV, see Table 1). In all cases,
the current vs. scan rate plots for the dichalcogen reduction
processes were linear (R2 between 0.973 and 0.999 for scan rates
between 100 and 500 mV/s), characteristic of electrochemically
active species adsorbed on the mercury electrode surface.
When the three classes of compounds, i.e., pyridine, aniline
and quinoline derivatives, were compared, the pyridine agents
(1 and 5) exhibited the most positive E0ꢀ (i.e., (Epa + Epc)/2)
values (−511 mV for selenium and −305 mV for the sulfur
analogue). In contrast, the aniline derivatives (2 and 7) had the
most negative values (−851 mV for selenium and −741 mV for
the sulfur analogue), implying that the aniline derivatives were
the most ‘reducing’ of the compounds, and therefore the most
promising candidates for antioxidant activity.6,7 Interestingly,
ebselen, used here as a benchmark control, exhibited irreversible
electrochemical behaviour under these conditions (i.e., at a
mercury electrode), with an Epa value of −703 mV.
Overall, these studies confirmed that all dichalcogens and
their reduced analogues were redox active, and able to undergo
reversible oxidation and reduction reactions within the phys-
iologically relevant potential range. Additionally, the results
obtained by CV hinted at a ranking order of good reducing
agents, with 2 topping this list.
Peroxidase-like catalytic activity
Compounds were then evaluated in in vitro assays indicative
of potential antioxidant activity, with focus on catalytic activity
and interactions with pro-oxidant metal ions (Table 1). Catalytic,
GPx-like activity was first measured in the thiophenol (PhSH)
peroxidation assay.8 The latter is a well established and easy to
use assay measuring the formation of PhSSPh from PhSH in
the presence of H2O2. As expected, the selenium, but not the
sulfur compounds were active in this assay.6,7 While the former,
once reduced, undergo selenol/selenenic acid catalysis, sulfur
1 5 4 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 4 1 – 1 5 4 6

View Article Online
compounds generally evade the sulfenic acid state and tend
to remain in—or return to—the disulfide state, which is not
catalytically active.9 There was no clear trend in activity among
the selenium compounds 1–4, with the quinoline derivative
3 being the most active, and the others having an activity
comparable to the one of ebselen (Table 1).
give results. These studies relied on changes in the compounds’
UV/VIS absorption spectra upon binding to Cu2+ (Fig. 3).
Catalytic activity was therefore further investigated in the
metallothionein (MT) assay that measures catalytic zinc release
from the zinc/sulfur protein in the presence of peroxide.10 This
assay is biologically more relevant than the PhSH assay (choice
of protein, buffer, pH) and is generally also more robust and
reliable.11 As in the PhSH assay, the selenium compounds were
more active than the sulfur ones. Among the former, the pyridine
derivative 1 was the most, and the quinoline derivative 3 the
least active (78 and 42% zinc release, respectively). Interestingly,
selenium compounds 1–3 were significantly more active than
ebselen in this assay (28% zinc release), confirming good catalytic
activity in vitro.
Interactions with copper ions
In addition to ROS, free, adventitious metal ions able to generate
oxygen radicals are frequently associated with OS. Among them,
non-protein bound copper plays a major role because of its
ability to convert peroxide into hydroxyl radicals in a Fenton-
type reaction.12 Chelators such as desferrioxamine are therefore
frequently used as ‘indirect’ antioxidants, i.e. compounds that
are not redox active themselves but interfere with the redox
reactions of others.13
Interactions between the compounds and Cu2+ were therefore
studied by differential pulse polarography (DPP) and UV/VIS
spectrophotometry.14 DPP of the copper ion detects changes in
the metal’s redox behaviour upon complexation, and is indicative
of the ligand’s indirect antioxidant properties. Changes in
UV/VIS spectra of ligands can be used to calculate an apparent
metal binding constant.
A summary of the results is shown in Table 1. Among the
compounds tested, the pyridine derivatives (1 and 5) caused the
largest shift in the Cu2+ reduction potential (DE1/2 = +298 mV
for 1 and +344 mV for 5). This effect is illustrated for 1 in
Fig. 2. Importantly, the shift in the copper reduction potential to
more positive values was accompanied by shifts in the oxidation
and reduction potentials of the ligand to more negative values
(DE0ꢀ = −50 mV for 1, −225 mV for 5). In contrast, the
aniline derivatives showed the smallest shift in copper reduction
potential (DE1/2 = +97 mV for 2 and +135 mV for 7), and also
the smallest change in the electrochemical potential of the ligand
(DE0ꢀ = −3 mV for 2 and −40 mV for 7).
Fig. 3 Changes in the UV/VIS absorption spectra of (a) compound
6 and (b) compound 10 upon addition of increasing Cu2+ (CuSO4)
concentrations; (a–f) 0–50 lM Cu2+ in 10 lM increments. Spectra were
recorded in 10 mM MOPS buffer (pH 7.0), at 25 ◦C. These changes
in UV/VIS absorption form the basis for the estimation of copper
binding stoichiometry and binding constants. Insert: mole-ratio plots to
determine Cu2+ binding to (a) compound 6 and (b) compound 10. Both
compounds bind the copper ion with a 2 : 1 stoichiometry, indicating
tetrahedral coordination around the metal.
For 10 (100 lM), a peak shift from 447 to 410 nm was observed
in the presence of Cu2+ (5–50 lM), and the absorbance at 410 nm
was used to determine the stoichiometic ratio by the ‘mole-
ratio’ method (an extension of Job’s method). For 6 (100 lM), a
decrease in peak absorbances at 272 and 343 nm was observed
in the presence of Cu2+ (5–50 lM), indicating the formation of
an UV-transparent copper complex. For 8 (100 lM) a peak shift
from 305 to 308 nm was observed upon addition of increasing
concentrations of Cu2+ (5–25 lM), and was accompanied by
decreases in peak absorbance at 224 and 261 nm. Unfortunately,
8 rapidly oxidises to the disulfide to give a peak at 328 nm,
resulting in a more complex spectrum that although confirming
Cu2+ binding to 8, could not be quantitatively analysed by the
‘mole-ratio’ method.
The mole-ratio plots derived from the spectra of 6 and 10 are
shown in Fig. 3 (insert). They indicate that 6 and 10 bind Cu2+
with a 2 : 1 ligand : metal stoichiometry, hinting at a tetrahedrally
coordinated copper ion. Using changes in the UV/VIS spectra
of the compounds upon addition of copper, copper binding
constants were estimated to be 1.5 × 108 M−1 for 6 and 1.2 ×
108 M−1 for 10 at 25 ◦C.
Fig. 2 (a) Cyclic voltammogram of 1 in the absence (solid) and presence
(dashed) of Cu2+. (b) Differential pulse polarogram of Cu2+ in the
absence (solid) and presence (dashed) of1. Experimental details are given
in the text. The voltammogram display follows the IUPAC convention.
Binding constants
In the above experiments, the reduced and oxidised forms of
the sulfur compounds had similar effects on the Cu2+ potential.
Cu2+ binding was therefore independently confirmed for 6, 8
and 10, the compounds stable enough in their reduced states to
Considered together, the electrochemical and in vitro studies
confirm a range of possible interactions of compounds such
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 4 1 – 1 5 4 6
1 5 4 3

View Article Online
as 1–3 with various, chemically very different oxidative stressors
present during OS. They also hint at a complex set of ‘responses’
these compounds show in the presence of stressors, such as
shifts in electrochemical potentials, which might allow activation
of antioxidant properties in the presence of OS. The most
promising compounds (1–3) were therefore tested in a human
skin fibroblast cell line (FCP7).
Protection against UVA radiation
An ultraviolet A (UVA, 320–400 nm) based assay was chosen
since radiation-induced OS is associated with both, ROS and
adventitious metal ions.15 In addition, skin is easily accessible
from a pharmacological point of view and provides a prime
target for antioxidant use.
Fig. 4 shows the results obtained for 1–3 in the cell culture
studies. A decrease in cell viability to approximately 60% of
control cells was observed when cells were irradiated with a
dose of 500 kJ m−2 UVA compared to that of control cells.
Pretreatment with 1 and 2 significantly reduced this loss, most
likely due to the antioxidant activity of these compounds
(24%
5.6 and 27%
7.6 noted absolute increase in cell
viability at maximum protective concentrations, respectively).
Activity of both compounds clearly exceeded that of ebselen
(16% 3.5 increase). In contrast to the cytoprotective effects
offered by 1 and 2, the quinoline derivative 3, although slightly
protective at 1 lM (9% 5), behaved as a pro-oxidant at higher
concentrations (5 lM) when exposed to UVA irradiation, and
dramatically enhanced UVA cell damage (reducing overall cell
viability by another 42% 8.5). Such phototoxic responses are
frequently reported for quinoline based drugs.16
Discussion
The use of antioxidants in disease prevention and therapy is
often limited by the complex network of biochemical redox
interactions which is characteristic of OS. Antioxidants able to
neutralise just one (or a few) stressors are therefore unlikely to
be either very effective or selective in cells under OS. In contrast,
multi-functional agents, the redox and metal binding properties
of which are tailored to closely mirror the redox environment
posed by a specific form of OS, such as UVA-induced stress,
promise to be considerably more effective.
Taken together, the results obtained in this study demonstrate
that it is possible to design and synthesise chemically simple
molecules that combine a range of important antioxidant
properties. While the individual activities, such as GPx-like
catalysis, already point towards good antioxidant behaviour, it
is the combination of these different activities which provides
the extraordinary protection from OS-related damage seen in
the skin cell culture.
Compound 2, in particular, is active in all of the various
antioxidant assays employed. Compared to ebselen, it is also
considerably more active in the skin cell culture and can therefore
be considered as a leading compound for the development of
more effective, multi-functional antioxidants. Interestingly, this
compound has recently been shown to be protective against lipid
peroxidation in brain tissue extracts from mice.17 While these cell
culture studies support our results, they provide little insight into
the chemical basis for such high antioxidant activity, and do not
hint at potential design criteria for improved antioxidant activity.
In contrast, our electrochemical and in vitro results indicate that
this activity might be due to a combination of redox catalysis
and metal binding.
Fig. 4 Activity of 1–3 in a human skin fibroblast cell line. Cells were
irradiated with UVA (500 kJ m−2) following an 18 h pre-incubation
period with and without 1–3 (5 lM for 1 and 3, 1 lM for 2) at 37 ◦C
in a 5% CO2 atmosphere. Control cells were treated similarly except
that they were not irradiated. Cell viability was determined 4 h post
irradiation using a standard colourimetric MTT assay. Results represent
3 independent experiments, each of which was performed in triplicate;
*P < 0.05: significant difference from untreated control; †P < 0.05:
significant difference from UVA treated control.
therefore employed CV on a mercury electrode to focus on the
thiol/disulfide and selenol/diselenide redox couples. The results
obtained by CV show that 2 has exceptionally low oxidation
and reduction potentials, even when compared to structurally
similar selenides 1 and 3. The reduced form of the compound
can therefore be expected to be a good reducing agent, one
that would be readily reactive towards intracellular ROS such as
hydrogen peroxide. In addition, reversibility of electron transfer
also points towards the compound’s ability to ‘redox cycle’, i.e.,
to participate in redox catalysis.
We have used CV in the past to gain a deeper understand-
ing of the redox properties of chalcogen compounds.6,7 The
electrochemical behaviour of sulfur, selenium and tellurium
compounds is somewhat more complex than that of metals
and metal complexes, and can be performed on carbon as
well as mercury electrodes.6,7,9 The studies presented here have
1 5 4 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 4 1 – 1 5 4 6

View Article Online
These electrochemical findings are reflected in the catalytic
assays. Not surprisingly, the selenium compounds are more
active than the sulfur compounds, a trend already observed
earlier.7 Interestingly, compound 1, the most oxidizing of the
selenium agents studied, is also the most active in the MT
assay. This is hardly surprising, since the catalytic, GPx-like
peroxidation cycle involves a slow thiol/selenosulfide exchange
step, when the selenosulfide is reduced back to selenol by a thiol
(such as PhSH, MT or GSH). More oxidising selenium species
might therefore result in a faster exchange reaction and might
well be catalytically more active than more reducing ones.
MT is also known to lose zinc ions in the presence of metal
binding agents.18 Since compounds such as 1 can also interact
with metal ions, their activity in the MT assay could be the result
of a combination of different protein/compound interactions.
This point obviously needs further investigation (see also below).
In any case, it is unlikely that peroxidation catalysis alone
explains why 2 is a good antioxidant in skin cell culture under
UVA-induced stress, whilst 3 actually acts as a pro-oxidant.
The notion that compounds such as 1–3 act via a more com-
plex mechanism is supported by the copper binding experiments.
As mentioned in the introduction, OS is frequently characterised
by increased concentrations of ROS and free, adventitious metal
ions. While compounds such as GSH, vitamin E and many
selenium compounds (such as ebselen) might defend against
ROS, chelators like desferrioxamine have been used to neutralise
the threat posed by redox-active metal ions participating in
Fenton-type reactions. The results obtained here show that
compounds like 2 integrate both of these antioxidant activities.
As far as copper binding is concerned, the studies have
revealed a number of interesting trends. Firstly, almost all of the
compounds studied are able to shift the copper reduction poten-
tial to more positive values, and this shift is most pronounced for
ligands with more positive oxidation and reduction potentials.
A positive shift in the Cu2+ potential upon ligand binding
has been observed before, and agrees with the stabilisation of
the Cu+ oxidation state by the sulfur, selenium and nitrogen
ligands.19 The importance of the chalcogen group for metal
interactions is underlined by the shifts observed for the control
selenium compound 4, while pyridine, aniline or quinoline did
not interfere with the Cu2+ reduction potential. The shift of the
copper potential implies a less reducing Cu+ ion, and this might
be biochemically beneficial, since Cu+ is acting as an electron
donor in the Fenton reaction.
Secondly, the copper/ligand interaction also leads to sig-
nificant changes in the ligand’s electrochemical properties,
with shifts to more negative values most pronounced for the
pyridine derivatives, which also cause the largest shifts of
copper reduction potential. The fact that copper binding would
therefore make the ligands more reducing is rather interesting,
since it would (at least in theory) imply antioxidant activation
by the pro-oxidant metal ion. This effect might ultimately
provide the basis for increased efficiency and selectivity of
compounds such as 1 and 2. While efficiency would be increased
by multifunctional properties, selectivity would be the result of
copper-activated reduction by the ligand. The binding constants
estimated for 6 and 10 support this notion. Compared to
conventional chelators, such as desferrioxamine (copper binding
constant 1.3 × 1014 M−1) and EDTA (copper binding constant
5 × 1018 M−1), metal binding to these chalcogen compounds is
weak enough to allow metal sensing and exchange in a cellular
environment.
Although the relationship between ligand redox behaviour,
complexation and peak shifts is therefore only just emerging,
the results obtained so far for 1 and 2 are rather promising.
They support the notion of chemically simple, multifunctional
agents, and underline the usefulness of techniques such as CV
and DPP play in their analysis. Together with the cell culture
studies, the in vitro results hint at a class of multifunctional
agents whose antioxidant activity might well surpass the one of
‘mono-functional’ selenium or tellurium agents studied in the
past.6,7
Conclusion
The study of compounds combining a number of distinct antiox-
idant properties provides the opportunity to refine and enhance
antioxidant activity, whilst also potentially increasing selectivity
for cells under OS. These are clear advantages over previously
used agents, since they no longer require the administration of
combinations of redox active and metal binding agents, and
might also exploit the emerging synergy between redox and
metal binding properties.
Future studies will obviously have to focus on the interplay
of redox processes and metal binding and evaluate the precise
underlying redox and metal exchange mechanisms. At the
same time, complementary biochemical studies will also have
to address the activity of such compounds in relevant cell
models and on human skin. Nevertheless, it seems feasible that
such chemically simple, yet multi-functional and biochemically
active compounds might play a major role in future antioxidant
research.
Experimental
Materials
Compounds 2, 9 and 11 were purchased from Fisher Scientific
(Leicestershire, UK). 2-Bromopyridine and 8-bromoquinoline
were purchased from Sigma-Aldrich (Poole, UK). Selenium
powder and hydrazine hydrate were purchased from Lancaster
Synthesis UK (Lancashire, UK). Cell culture materials were ob-
tained from Life Sciences Technologies (Paisley, UK), with the
exception of foetal calf serum (FCS), obtained from PAA (Aus-
tria). All other reagents were analytical grade and purchased
from Sigma-Aldrich (Poole, UK). Cd,Zn-metallothionein was
purchased from Sigma, and Zn7–MT was reconstituted and
purified according to a standard method.20
Methods
Synthesis of compounds 1 and 3. Compound 1 was syn-
thesised from selenium and 2-bromopyridine according to
Bhasin et al., and analytical results were found to be in
accordance with published values.5 Compound 3 was synthe-
sised according to this procedure replacing 2-bromopyridine
with 8-bromoquinoline and refluxing for 3 d. This compound
has previously been synthesised following a different synthetic
pathway. Analytical results agreed with literature values.21
Cyclic voltammetry (CV) and differential pulse polarogra-
phy (DPP). CV and DPP were performed on a 100B/W
workstation (BAS) at 25 ◦C. Cyclic voltammograms of the
compounds (25 lM) were recorded in MOPS buffer (10 mM,
pH 7) containing potassium nitrate (50 mM), in the presence
and absence of Cu(NO3)2 (25 lM), using a mercury working
electrode, an Ag/AgCl reference electrode (SSE) and a platinum
wire counter electrode at a potential range between 0 and −800
mV. This buffer was chosen since it is known not to interact with
metal ions.22
Future studies will have to consider the precise nature of the
emerging redox/metal binding relationship. For example, metal
binding to the diselenides might well be triggered or enhanced by
dichalcogen reduction, a process possible in cells with sufficient
GSH concentrations. In addition, the metal complexes might
exhibit their own catalytic behaviour, possibly even similar to
metal-based dismutase mimics. These topics require further
chemical and biochemical investigations.
Compounds (75 lM) were studied by DPP in the presence and
absence of Cu(NO3)2 (75 lM), using the above buffer system
and a glassy carbon working electrode (cleaned and polished
with Al2O3 after each scan), at a potential range between +600
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 4 1 – 1 5 4 6
1 5 4 5

View Article Online
and −1000 mV, pulse amplitude 80 mV, sample width 45 ms
and pulse width 100 ms. Quinoline derivatives required 50%
methanol due to limited solubility in aqueous media. Buffers
were purged with nitrogen for 30 min prior to use. Potentials are
given vs. SSE and are standardised against ferrocene.23 For the
sulfur compounds, the disulfide and the reduced thiol analogues
were studied.
oxidation potential; Epc, cathodic reduction potential; E0ꢀ, stan-
dard potential of redox couple; GCE, glassy carbon electrode;
GPx, glutathione peroxidase; GSH, glutathione; HEPES, N-(2-
hydroxyethyl)piperazine-Nꢀ-(2-ethanesulfonic acid); H2O2, hy-
drogen peroxide; OS, oxidative stress; PAR, 4-(2-pyridylazo)-
resorcinol monosodium salt; PBS, phosphate buffered saline;
PhSH, thiophenol; PhSSPh, diphenyldisulfide; MT, met-
allothionein; MOPS, 3-[N-morpholino]propanesulfonic acid;
MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bro-
mide; ROS, reactive oxygen species; SSE, standard silver elec-
trode.
Thiophenol and MT assays. The thiophenol (PhSH) and
MT assays are spectrophotometric assays that can be used to
determine the activity of compounds as peroxidation catalysts.
While the PhSH assay directly measures the oxidation of PhSH
in the presence of H2O2 in methanol, the MT assay monitors
oxidative zinc release from the two zinc/sulfur clusters of MT (20
cysteine residues holding 7 zinc ions) in the presence of tBuOOH
and a chromophoric dye. Both assays have their own advantages,
and together, they provide a good indication of whether a new
compound, when measured against known benchmark activity
(e.g., ebselen), is catalytically active or not.
For the PhSH assay, 100 lM of the sulfur or selenium
compound was added to a 1 mM methanolic solution of PhSH
and the reaction initiated by addition of 2 mM H2O2. PhSSPh
formation was monitored at 305 nm for 10 min at 25 ◦C. Initial
velocities (V0) were calculated from the first 5–10% of the
reaction.
Acknowledgements
This work was financially supported by the Wellcome Trust, the
Leverhulme Trust, DAART and the University of Exeter. The
authors are grateful to Sandra Pariagh, James Tucker and Yann
Molard (Exeter) for helpful discussions and to Rex Tyrrell for
allowing access to the UVA radiation facility in his laboratory.
References
1 B. Halliwell, Am. J. Med., 1991, 91, S14–S22.
2 H. Sies, Eur. J. Biochem., 1993, 215, 213–219.
3 M. Parnham and H. Sies, Expert Opin. Invest. Drugs, 2000, 9, 607–
619.
4 S. Pariagh, K. M. Tasker, F. H. Fry, A. L. Holme, C. A. Collins, N.
Okarter, N. Gutowski and C. Jacob, Org. Biomol. Chem., 2005, 3,
975–980.
5 K. K. Bhasin and J. Singh, J. Organomet. Chem., 2002, 658, 71–76.
6 G. I. Giles, K. M. Tasker, R. J. K. Johnson, C. Jacob, C. Peers and
K. N. Green, Chem. Commun., 2001, 2490–2491.
7 G. I. Giles, F. H. Fry, K. M. Tasker, A. L. Holme, C. Peers, K. N.
Green, L.-O. Klotz, H. Sies and C. Jacob, Org. Biomol. Chem., 2003,
1, 4317–4322.
For the MT assay, MT (0.5 lM) was incubated with tBuOOH
(500 lM) and PAR (100 lM) in 20 mM HEPES–Na+ buffer
(pH 7.0, 25 ◦C) in the absence and presence of sulfur or selenium
compounds (1 lM). Catalytic activity was assessed based on the
total extent of zinc release after 60 min. Maximum zinc release
occurs when all protein thiols (10 lM) are oxidised. This was
measured using an excess ebselen, and activities are expressed as
a percentage of this value.
Cell culture. Fibroblast monolayers (FCP7) were cultured
routinely in Minimum Essential medium with Earle’s salts
(EMEM) supplemented with 2 mM L-glutamine, 50 lg ml−1
penicillin, 50 lg ml−1 streptomycin, 0.2% sodium bicarbonate
and 15% FCS at 37 ◦C in 5% CO2. Cells were used between
passages 11 and 15. A standard colourimetric MTT (3-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay
was used to assess cell viability following UVA irradiation.24
For this, cells were seeded in 3 cm dishes (5 × 104 cells/dish)
in supplemented EMEM medium (2 ml) and the fibroblast
monolayers were allowed to grow for 3 d in order to reach
80% confluency. Cells were pretreated with compounds (0.1–
10 lM dissolved in DMSO) 18 h prior to the UVA irradiation
process. The concentration of DMSO was kept constant at 0.1%
of the medium in order to avoid cellular effects of the vehicle, and
control cells were pretreated with DMSO alone. UVA irradiation
was performed using a broad spectrum 4 kW UVA lamp (350–
400 nm, Sellas, Germany) whereby cells were irradiated with
500 kJ m−2 UVA in Ca2+–Mg2+ PBS according to an established
literature procedure.25 Following irradiation, cells were washed
with PBS (1 ml) and the conditioned medium (1.5 ml) was
added back to each dish. Cells were incubated further for 1 h at
37 ◦C, after which the MTT assay was performed as previously
described.26
8 M. Iwaoka and S. Tomoda, J. Am. Chem. Soc., 1994, 116, 2557–2561.
9 C. Jacob, G. I. Giles, N. M. Giles and H. Sies, Angew. Chem., Int.
Ed., 2003, 42, 4742–4258.
10 C. Jacob, W. Maret and B. L. Vallee, Proc. Natl. Acad. Sci. U. S. A.,
1999, 96, 1910–1914.
11 C. Jacob, W. Maret and B. L. Vallee, Biochem. Biophys. Res.
Commun., 1998, 248, 569–573.
12 S. I. Dikalov, M. P. Vitek and R. P. Mason, Free Radical Biol. Med.,
2004, 36, 340–347.
13 G. J. Kontoghiorghes, K. Pattichis, K. Neocleous and A. Kolnagou,
Curr. Med. Chem., 2004, 11, 2161–2183.
14 W. C. Vosburgh and G. R. Cooper, J. Am. Chem. Soc., 1741, 63, 437.
15 J. L. Zhong, A. Yiakouvaki, P. Holley, R. M. Tyrrell and C. Pourzand,
J. Invest. Dermatol., 2004, 123, 771–780.
16 G. G. Aloisi, A. Barbafina, M. Canton, F. Dall’Acqua, F. Elisei, L.
Facciolo, L. Latterini and G. Viola, Photochem. Photobiol., 2004, 79,
248–258.
17 F. C. Meotti, E. C. Stangherlin, G. Zeni, C. W. Nogueira and J. B.
Rocha, Environ. Res., 2004, 94, 276–282.
18 C. Jacob, W. Maret and B. L. Vallee, Proc. Natl. Acad. Sci. U. S. A.,
1998, 95, 3489–3494.
19 J. R. Hartman and S. R. Cooper, J. Am. Chem. Soc., 1986, 108,
1202–1208.
20 M. Vasˇa´k, Methods Enzymol., 1991, 205, 41–44.
21 E. Sekido and I. Fujiwara, Talanta, 1972, 19, 647–655.
22 Q. Yu, A. Kandegedara, Y. Xu and D. B. Rorabacher, Anal. Biochem.,
1997, 253, 50–56.
23 D. T. Sawyer, A. Sobkowiak and J. L. Roberts, Electrochemistry for
Chemists, John Wiley and Sons, New York, 1995.
24 T. Mosmann, J. Immunol. Methods, 1983, 65, 55–63.
25 C. Pourzand, R. D. Watkin, J. E. Brown and R. M. Tyrrell, Proc.
Natl. Acad. Sci. U. S. A., 1999, 96, 6751–6756.
Abbreviations
tBuOOH, tert-butyl hydroperoxide; Cu(NO3)2, copper nitrate;
CuSO4, copper sulfate;CV, cyclicvoltammetry; DPP, differential
pulse polarography; E1/2, half-wave potential; Epa, anodic
26 S. Basu-Modak, M. J. Gordon, L. H. Dobson, J. P. Spencer, C. Rice-
Evans and R. M. Tyrrell, Free Radical Biol. Med., 2003, 35, 910–921.
1 5 4 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 5 4 1 – 1 5 4 6