Biochem. J. (2008) 416, 441–452 (Printed in Great Britain)
441
doi:10.1042/BJ20070941
Tryptophan residues are targets in hypothiocyanous acid-mediated
protein oxidation
Clare L. HAWKINS1 , David I. PATTISON, Naomi R. STANLEY and Michael J. DAVIES
The Heart Research Institute, 114 Pyrmont Bridge Road, Camperdown, Sydney, NSW 2050, Australia
Myeloperoxidase, released by activated phagocytes, forms
reactive oxidants by catalysing the reaction of halide and pseudohalide ions with H2 O2 . These oxidants have been linked to tissue
damage in a range of inflammatory diseases. With physiological
levels of halide and pseudo-halide ions, similar amounts of
HOCl (hypochlorous acid) and HOSCN (hypothiocyanous acid)
are produced by myeloperoxidase. Although the importance of
HOSCN in initiating cellular damage via thiol oxidation is
becoming increasingly recognized, there are limited data on the reactions of HOSCN with other targets. In the present study, the
products of the reaction of HOSCN with proteins has been studied.
With albumin, thiols are oxidized preferentially forming unstable
sulfenyl thiocyanate derivatives, as evidenced by the reversible
incorporation of 14 C from HOS14 CN. On consumption of the HSA
(human serum albumin) free thiol group, the formation of stable
14
C-containing products and oxidation of tryptophan residues
are observed. Oxidation of tryptophan residues is observed on
reaction of HOSCN with other proteins (including myoglobin,
lysozyme and trypsin inhibitor), but not free tryptophan, or
tryptophan-containing peptides. Peptide mass mapping studies
with HOSCN-treated myoglobin, showed the addition of two
oxygen atoms on either Trp7 or Trp14 with equimolar or less oxidant, and the addition of a further two oxygen atoms to the
other tryptophan with higher oxidant concentrations (2-fold).
Tryptophan oxidation was observed on treating myoglobin with
HOSCN in the presence of glutathione and ascorbate. Thus
tryptophan residues are likely to be favourable targets for the
reaction in biological systems, and the oxidation products formed
may be useful biomarkers of HOSCN-mediated protein oxidation.
INTRODUCTION
infections (reviewed in [1]). In contrast, the role of SCN− -derived
oxidants in disease is not well understood, mainly due to a lack of
specific biomarkers for SCN− -mediated damage. The importance
of SCN− -derived oxidants produced by MPO in atherosclerosis has been highlighted by the detection of elevated levels of
carbamylated proteins in plaques [8]. Moreover, plasma concentrations of SCN− , which are elevated in smokers [9], correlate
significantly with early markers of atherosclerosis, including
deposition of oxidized lipoproteins in the artery wall and the
formation of lipid-laden macrophages (foam cells) [10,11].
It has been proposed that the oxidation product responsible
for the antibacterial activity of the LPO/H2 O2 /SCN− system is
HOSCN (hypothiocyanous acid) [12–18]. The proposed mechanism of SCN− oxidation by the peroxidase and H2 O2 are outlined
in eqns (1–3).
Peroxidase enzymes play an important role in mammalian defence
mechanisms by catalysing the formation of oxidants that act
as potent antibacterial agents (reviewed in [1]). MPO (myeloperoxidase) and EPO (eosinophil peroxidase) are released by
activated phagocytes [2,3], whereas LPO (lactoperoxidase) is
an antimicrobial agent in milk, saliva and tears [4]. However,
excessive or misplaced generation of oxidants by peroxidases,
particularly MPO and EPO, is believed to contribute to the
progression of a number of diseases, including atherosclerosis,
chronic inflammation, asthma and some cancers (reviewed in [1]).
It is well-established that MPO and EPO utilize chloride ions
(Cl− ) and bromide ions (Br− ) respectively, to produce HOCl
(hypochlorous acid) and HOBr (hypobromous acid) [2,3]. However, oxidation of SCN− (thiocyanate ions) by MPO and EPO
is also important, as SCN− is the preferred substrate for these
peroxidases at physiological halide ion concentrations (100–
140 mM Cl− , 20–100 μM Br− , < 1 μM I− and 120 μM
SCN− ) [3,5,6]. It has been estimated that approx. 50 % of the
H2 O2 consumed by MPO oxidizes SCN− under physiological
conditions, with most of the remaining H2 O2 (approx. 45 %) being
used to oxidize Cl− , based on the specificity constants of 1:60:730
for Cl− , Br− and SCN− respectively [5].
HOCl is known to play an important role in the oxidative
damage observed in various pathologies, as evidenced by the
detection of a specific biomarker for this oxidant, 3-chlorotyrosine, in diseased tissue (e.g. [7]). Similarly, detection of 3-bromotyrosine implies a critical role of HOBr in the tissue damage
observed in asthma, allergic reactions, malignancies and
Key words: hypochlorous acid (HOCl), hypothiocyanous acid
(HOSCN), myeloperoxidase, protein oxidation, thiocyanate.
H2 O2 + 2SCN− + 2H+ → 2H2 O + (SCN)2
(1)
(SCN)2 + H2 O HOSCN + H+ + SCN−
(2)
OSCN− + H+
HOSCN (3)
The pK a of HOSCN is 5.3 [14], therefore a mixture of both the
protonated form and anion OSCN− will exist at physiological
pH. HOSCN may not be the only SCN− -derived oxidant, as
HOSCN decomposes readily in aqueous solution to form other
reactive species [14,16,17]. HOSCN is used to represent this
mixture hereon. The nature of these decomposition products
remains controversial, with evidence for the production of (SCN)2
(thiocyanogen) [19,20], HO2 SCN (cyanosulfurous acid) [13],
Abbreviations used: ANS, 8-anilino-1-napthalenesulfonic acid; DTNB, 5,5 -dithio-2-nitrobenzoic acid; DTT, dithiothreitol; EPO, eosinophil peroxidase;
LPO, lactoperoxidase; MPO, myeloperoxidase; MS/MS, tandem MS; NFK, N -formylkynurenine; SRM, selective reaction monitoring; STI, soya bean trypsin
inhibitor; TCA, trichloroacetic acid; TFA, trifluoroacetic acid; TNB, 5-thio-2-nitrobenzoic acid; XC, cross correlation score.
1
To whom correspondence should be addressed (email [email protected]).
c The Authors Journal compilation c 2008 Biochemical Society
442
C. L. Hawkins and others
HO3 SCN (cyanosulfuric acid), (SCN)3 − (trithiocyanate) [21] and
CN− (cyanide) [16,17]. OCN− (cyanate) is a product of the EPO
and MPO/H2 O2 /SCN− systems [8,22], and radical species, such
as SCN , OSCN− or (SCN)2 − may also play a role in biological
damage [23–25]. The yield of SCN− -derived oxidants may be
increased in vivo, by the reaction of SCN− with HOCl and HOBr
[26,27].
Although it is well-established that HOSCN exerts potent
bactericidal and bacteriostatic effects, the mechanism(s) of this
process are unclear, although there is evidence to suggest that
thiol oxidation and protein modification play an important
role [15,28,29]. Similarly, oxidation of cellular thiols plays
an important role in HOSCN-mediated damage to mammalian
cells [3,30–32]. Although protein thiol groups are known to be
particularly susceptible to HOSCN-induced oxidation [29,33],
there are a significant number of proteins that lack this functional
group. Currently, there is little information available regarding the
reactivity of other protein sites, or the consequences of HOSCNinduced protein oxidation.
In the present study, the reactivity of HOSCN, generated enzymatically by the LPO/H2 O2 /SCN− system, with different thioland non-thiol containing proteins and plasma has been investigated. With HSA (human serum albumin) and plasma, it is shown
that HOSCN reacts preferentially with the free thiol group.
However, in the absence of the thiol group, or after consumption
of this species, tryptophan residues are important targets for
HOSCN-mediated oxidation. The HOSCN-induced oxidation of
tryptophan residues results in the formation of several stable
products, and results in a loss of protein integrity via unfolding.
䊉
䊉
䊉
MATERIALS AND METHODS
Reagents
Nanopure water was filtered in a four-stage Milli-Q system
(Millipore). pH control was achieved using 0.1 M sodium
phosphate buffer (pH 7.4) pre-treated with Chelex resin (BioRad) to remove contaminating trace metal ions. HSA (essentially
fatty-acid free), STI (soya bean trypsin inhibitor, type 1-S),
lysozyme (type C from chicken egg) and myoglobin (from horse
heart) were obtained from Sigma–Aldrich. H2 18 O was obtained
from Novachem. All other chemicals were of analytical reagent
grade. H2 O2 [30 % (v/v) solution; Merck] was quantified by UV
absorbance at 240 nm using a molar absorption coefficient (ε)
of 39.4 M−1 · cm−1 [34]. Plasma was obtained from heparinized
blood donated by multiple volunteers (males and females), with
full written consent in accordance with local ethical guidelines,
by centrifugation at 1000 g for 20 min at 4 ◦C. Plasma was divided
into aliquots and stored at − 80 ◦C immediately after separation.
Generation and quantification of HOSCN
HOSCN was produced enzymatically using LPO (from bovine
milk; Calbiochem) as described previously [30]. Catalase (from
bovine liver, 140 units; Sigma–Aldrich) was added to remove
unreacted H2 O2 before filtration by centrifugation (10 000 g for
5 min at 4 ◦C) through nanosep devices (Pall Life Sciences) with
a 10 000 Da molecular-mass cut-off to remove catalase and LPO.
The concentration of HOSCN (and other oxidizing decomposition
products) was assessed immediately by the use of TNB (5-thio-2nitrobenzoic acid). HOSCN is used to represent this potential
mixture of species hereon. TNB was prepared as described
previously [35]. The concentration of TNB consumed after the
reaction with HOSCN for 5 min was determined at 412 nm using
a molar absorption coefficient (ε) of 14 150 M−1 · cm−1 [36].
c The Authors Journal compilation c 2008 Biochemical Society
Quantification of protein thiols
The concentration of protein thiol groups was assessed by reaction
with DTNB (5,5 -dithio-2-nitrobenzoic acid). HSA (1.5 mM) and
plasma (diluted 2.5-fold) samples were added to DTNB (500 μM)
at pH 7.4 in sodium phosphate buffer (0.1 M). Samples were
incubated for 30 min at 20 ◦C in the dark, before measuring the
gain in absorbance at 412 nm and quantification of the thiols with
a molar absorption coefficient (ε) of 13 600 M−1 · cm−1 [36].
S14 CN incorporation studies with HOS14 CN
HOS14 CN was prepared as above except that KS14 CN (3.75 mM;
American Radiolabeled Chemicals) was used and 1.875 mM
H2 O2 . Control experiments to assess non-specific binding of
S14 CN− to the protein were performed using solutions prepared
in the absence of LPO. HOS14 CN was added to HSA (25 μM)
and incubated for 15 min–24 h before precipitation with 10 %
(w/v) TCA (trichloroacetic acid). Protein was pelleted by
centrifugation (10 000 g for 5 min at 4 ◦C), and the free S14 CN−
label was removed by washing the protein pellets twice with
5 % (w/v) TCA and twice with ice-cold acetone. Protein pellets
were air-dried and re-suspended in 100 μl of formic acid before
addition to 5 ml of Ultima Gold scintillant (PerkinElmer) and
liquid scintillation counting.
Protein hydrolysis and amino acid analysis by HPLC
Proteins (5 μM) and plasma (diluted 10–200-fold) were incubated with HOSCN before precipitation with 0.015 % sodium deoxycholate and 5 % (w/v) TCA, pelleting, washing, solubilization
in 4 M methanesulfonic acid containing 0.2 % tryptamine
(Sigma–Aldrich), and hydrolysis as described previously [37].
The amino acids were separated by HPLC after pre-column derivatization with o-phthaldialdehyde reagent (with 2-mercaptoethanol added) for 1 min before injection on to an Ultrasphere
ODS column (4.6 mm × 250 mm, 5 μm particle size; Beckman–
Coulter) and separation on a Shimadzu HPLC system, at 30 ◦C
and 1 ml · min−1 , as described previously [37]. Amino acid
derivatives were quantified by fluorescence at λEx 340 nm and
λEm 440 nm, with amino acid standards [Sigma–Aldrich; with
methionine sulfoxide added].
Preparation of apomyoglobin
Haem-depleted apomyoglobin was prepared by dropwise
addition of ice-cold, acidified [with 1 % (v/v) HCl] acetone
containing 0.1 % 2-mercaptoethanol to a solution of myoglobin
(0.1 g · ml−1 ) while vortex mixing, as described previously [38].
The protein was pelleted by centrifugation at 500 g for 15 min,
washed twice with ice-cold acetone, and allowed to air dry.
Complete removal of the haem group was confirmed by the
absence of the Soret band (400–420 nm) by visible spectroscopy
(results not shown).
LC/MS peptide mass mapping studies
Myoglobin (50 μM) was treated with HOSCN (0–500 μM) for
30 min before addition of 5 % (w/v) TCA to precipitate the
protein, and pelleting by centrifugation (7500 g for 5 min at 4 ◦C).
Pellets were solubilized (in 8 M urea and 0.4 M NH4 HCO3 ) after
washing twice with ice-cold acetone. Proteins were reduced with
45 mM DTT (dithiothreitol) for 15 min at 50 ◦C and alkylated
with 100 mM iodoacetamide for 15 min at 21 ◦C prior to 4-fold
dilution with H2 O and digestion with trypsin (sequencing grade,
Hypothiocyanous acid-mediated protein oxidation
443
Promega; 235 units for 18 h at 37 ◦C). Digestion was terminated
by the addition of 0.02 % TFA (trifluoroacetic acid).
The resulting peptides were analysed by LC/MS in the positiveion mode with a Finnigan LCQ Deca XP ion-trap instrument
coupled to a Finnigan Surveyor HPLC system. Peptides were
separated on a Zorbax ODS column (3.0 mm × 250 mm, 5 μm
particle size; Agilent Technologies) at 30 ◦C with a flow rate of
0.4 ml · min−1 using the following gradient: 5–10 % B (solvent
B is 0.1 % TFA in acetonitrile) over 5 min, 10–20 % B over
10 min, 20–50 % B over 20 min, 50–100 % B over 5 min, with a
15 min wash at 100 % B before 10 min re-equilibration to 5 % B
with solvent A (0.1 % TFA in H2 O). The electrospray needle was
held at 4500 V. Nitrogen, the sheath and sweep gas, were set to
80 and 10 units respectively. The collision gas was helium. The
temperature of the heated capillary was 250 ◦C.
Quantification of protein unfolding by fluorescence binding
The fluorescence of the hydrophobic probe, ANS (8-anilino1-napthalenesulfonic acid), on addition to native and oxidized
STI was monitored at λEx 387 nm and λEm 479 nm as previously
described [35]. Samples contained 5 μM protein and ANS to a
final concentration of 90 μM.
Statistical analyses
Statistical analyses to compare the effect of HOSCN treatment
with the untreated control were carried out using one-way
ANOVA with Dunnett’s post-hoc test. A two-way ANOVA with
Bonferroni post-hoc testing was employed to compare the incorporation of the 14 C isotope in experiments with HSA and
HOS14 CN under different experimental conditions. The correlation of STI tryptophan oxidation with protein unfolding was
carried out by linear regression analysis. All statistical analyses
were performed using GraphPad Prism 4 (GraphPad Software;
http://www.graphpad.com), with P < 0.05 taken as significant.
RESULTS
Reactivity of protein thiol groups with HOSCN
Reaction of HSA (1.5 mM) with various concentrations of
HOSCN (0–600 μM) at pH 7.4, resulted in a dose-dependent
increase in the consumption of thiol groups, quantified by the
DTNB assay (Figure 1a). In each case, the concentration of protein
was in excess (at least 2.5-fold) of the oxidant to minimize any
confounding reaction of the HOSCN with the TNB generated on
the reaction of DTNB with the protein thiols. Similar results were
obtained in experiments with plasma (Figure 1b). With HSA and
plasma, the reaction of HOSCN with protein thiols occurred with a
stoichiometry of one HOSCN molecule consumed per thiol group.
The presence of residual HOSCN remaining after the reaction
with HSA or plasma was examined using the TNB assay. With
HSA, 85–90 % of the added HOSCN was consumed, even at
the highest concentration of HOSCN added (600 μM), which
was in excess of the protein thiol concentration (results not
shown). A similar extent of oxidant consumption (75–80 %) was
observed in the corresponding experiments with plasma (results
not shown). It is likely that these data are an underestimation of
the concentration of oxidant consumed, as any protein-derived
reactive intermediates (e.g. RN-SCN) will also react with TNB.
Reaction of HOSCN with other protein sites
The potential reactivity of HOSCN with other sites on proteins
was investigated initially by quantifying the incorporation of 14 C
Figure 1
Reactivity of protein thiol groups with HOSCN
(a) HSA (1.5 mM) treated with HOSCN (0–600 μM) and (b) human plasma (diluted 2.5-fold)
with HOSCN (0–1 mM) for 5 min at pH 7.4, before the addition of DTNB (500 μM) to assess the
concentration of thiol groups. The concentration of thiols in native HSA and untreated plasma
+
+
(diluted 2.5-fold) was 340 +
− 18 μM and 215 − 20 μM respectively. Values are means − S.D.
(n 4).
from HOS14 CN. Studies were carried out with HSA (25 μM)
treated with increasing concentrations of HOS14 CN (1.25–
125 μM) for 30 min before precipitation of the protein, washing
to remove free S14 CN− , and scintillation counting. This treatment
resulted in significant incorporation of the 14 C label (Figure 2a). In
contrast, low incorporation was observed in control experiments
where HSA was treated with S14 CN− /H2 O2 that had been incubated in the absence of LPO (Figure 2a). Under these conditions,
approx. 2 % of the HOS14 CN added was incorporated into the
protein, consistent with the modification of approx. 1 in 50 HSA
molecules. The extent of reversible compared with irreversible
incorporation of 14 C into the protein was examined by addition
of DTT (1.25 mM) prior to the precipitation of HOS14 CNtreated samples. Addition of DTT resulted in a significant
decrease in the level of 14 C incorporated into HSA to control
values, in experiments with low concentrations ( 5 μM, 0.2-fold
molar excess) of HOS14 CN (Figure 2a). However, a significant
(P < 0.01, two-way ANOVA) proportion of the 14 C label remained
associated with the HSA on treatment with higher concentrations
( 25 μM, equimolar or greater oxidant to protein) of HOS14 CN,
after treatment with DTT, compared with the LPO-free control
(Figure 2a). The incorporation of 14 C was lower in the presence
of DTT with approx. 1 in 500 and 1 in 100 molecules of HSA
modified in experiments with equimolar and 5-fold molar excess
HOS14 CN respectively. This suggests that in addition to the
reaction with thiol groups, HOS14 CN reacts with other protein
sites to generate products containing the 14 C label that are not
able to be reduced by DTT.
The incubation time of HSA (25 μM) with HOS14 CN (25 μM)
before precipitation of the protein was varied (5 min–24 h),
to investigate the stability of the 14 C-labelled products. In the
absence of added DTT, the extent of 14 C incorporation into
HSA was dependent on the incubation time with HOS14 CN. The
amount of 14 C-label incorporated into HSA decreased rapidly over
60 min, showing that the DTT-reducible products were unstable
(Figure 2b). In contrast, the level of 14 C-label incorporation
c The Authors Journal compilation c 2008 Biochemical Society
444
C. L. Hawkins and others
Figure 3
14
C incorporation on treating poly-lysine with HOS14 CN
Poly-lysine (10 mg · ml−1 , pH 7.4) was treated with HOS14 CN (125 μM) or oxidant-free
S14 CN− solutions for 30 min with and without DTT (1.25 mM). Values are means +
− S.E.M.
(n 3). * represents a significant increase (P < 0.05) in 14 C incorporation compared with that
observed on addition of DTT, or in the absence of HOS14 CN by one-way ANOVA with
Neuman–Keuls post-hoc testing.
Figure 2
14
C incorporation on treating HSA with HOS14 CN
(a) HSA (25 μM) treated with HOS14 CN (1.25–125 μM) for 30 min, with no DTT added (white
bars) and after addition of DTT (1.25 mM; grey bars) or oxidant-free S14 CN− solutions for
30 min with no DTT added (black bars) and after addition of DTT (1.25 mM; hatched bars).
(b) HSA (25 μM) treated with HOS14 CN (25 μM) (䊉, solid line) or oxidant-free S14 CN−
solutions (䊊, broken line) for various times with no DTT added. (c) HSA (25 μM) treated with
HOS14 CN (25 μM) (䊉, solid line) or oxidant-free S14 CN− solutions (䊊, broken line) for various
times before addition of DTT (1.25 mM). (d) HSA (25 μM) treated with freshly prepared HOS14 CN
(125 μM) (white bars), HOS14 CN aged for 3 h (black bars) and HOS14 CN aged for 48 h (hatched
bars) or the respective oxidant-free S14 CN− solutions for 30 min, with and without addition of
DTT (1.25 mM). In (a) letters ‘a’ and ‘b’ over bars represent a significant increase (P < 0.05) in the
incorporation of 14 C compared with the control treated with oxidant-free S14 CN− solutions
in the absence and presence of DTT respectively, using two-way ANOVA with Bonferroni
post-hoc testing. In (c) * represents a significant increase (P < 0.05) in 14 C incorporation
compared with earlier time points by one-way ANOVA with Neuman–Keuls post-hoc testing. In
(d) * represents a significant increase (P < 0.05) in 14 C incorporation compared with freshly
prepared oxidant, by two-way ANOVA with Bonferroni post-hoc testing. In all cases, values are
means +
− S.E.M. (n 3).
observed in the experiments with added DTT was not dependent
on the incubation time of the protein with HOS14 CN over 6 h
(Figure 2c). However, a small, but significant, increase in the HSA
14
C incorporation was observed after 24 h incubation when DTT
was added (Figure 2c). This suggests that 14 C-labelled products
c The Authors Journal compilation c 2008 Biochemical Society
are also formed to a minor extent via reactions of HOS14 CN
decomposition products, or via secondary reactions mediated by
protein-derived reactive intermediates. Analogous experiments
were performed with aged HOS14 CN solutions that had been
pre-incubated for 48 h, to allow decomposition of the HOS14 CN,
prior to addition to the HSA. This treatment resulted in the loss
of approx. 80 % of the TNB-reactive species (results not shown).
Reaction of HSA (25 μM) with aged HOS14 CN solutions for
30 min resulted in a significant increase in the formation of reducible 14 C-products, but a decrease in the concentration of nonreducible 14 C-products, compared with experiments with freshly
prepared oxidant (Figure 2d).
Carbamylation (addition of -CONH2 ) of lysine residues has
been reported on reaction of MPO/H2 O2 /SCN− with isolated
proteins [8]. Thus the formation of stable, 14 C-labelled material on
reaction of poly-lysine (10 mg · ml−1 ) with HOS14 CN (125 μM)
for 30 min at pH 7.4 was studied. In this case, the 14 C incorporation
observed was completely reversible on addition of DTT, with no
evidence obtained for stable products (Figure 3). This suggests
that the stable products observed with HSA are not due to
carbamylation or other lysine-derived modifications. Experiments
were also performed to assess whether these lysine-derived products were able to oxidize TNB, in accordance with previous
experiments with HOCl and HOBr, where TNB-reactive chloramines and bromamines are formed (e.g. [35,39]). Reaction of
poly-lysine (5 mg · ml−1 ) with HOSCN (250 μM) for 30 min
at pH 7.4 resulted in the formation of polymer-derived, TNBreactive species (33.8 +
− 2.9 μM). Care was taken to remove any
residual HOSCN using a Sephadex PD-10 column before quantification of the polymer-derived species using the TNB assay.
The relative susceptibilities of the different amino acids in
HSA to damage by HOSCN was investigated by amino acid
analysis using HPLC. HSA (5 μM) was treated with various
concentrations of HOSCN (5–750 μM, equimolar to 150-fold
molar excess) for 30 min before precipitation of the protein,
hydrolysis with methanesulfonic acid, derivatization and analysis
by HPLC with fluorescence detection. This treatment resulted
in a dose-dependent decrease in the concentration of tryptophan
residues with increasing amounts of HOSCN (Figure 4a). No loss
of any other amino acid quantified by this method (methionine,
histidine, tyrosine, lysine, arginine, serine, glycine, threonine,
alanine, valine, isoleucine, leucine and phenylalanine) was observed. Similarly, there was no increase in the formation of the oxidation product methionine sulfoxide (results not shown). The
extent of tryptophan oxidation was significant (P < 0.05, oneway ANOVA) on reaction of HSA with equimolar HOSCN,
Hypothiocyanous acid-mediated protein oxidation
445
modification of the haem group, was observed after treating
the protein with HOSCN (5- and 10-fold molar excess) (results
not shown). The relative proportions of ferricmyoglobin (MbIII )
and ferrylmyoglobin (MbIV ) were determined on the reaction
of myoglobin (25 μM) with HOSCN (250–500 μM), using the
Whitburn equations (eqns 4 and 5) [40].
[MbIII ] = 146 A490 − 108 A560 + 2.1 A580
(4)
[Mb ] = −62 A490 + 242 A560 − 123 A580
(5)
IV
The spectral changes observed were consistent with a loss
(approx. 15 %) of MbIII , with no significant formation of
MbIV observed. Control experiments with SCN− and myoglobin
resulted in different spectral changes, with a new shoulder peak
observed at 545 nm, consistent with the formation of SCNmyoglobin [41]. Thus the nature of the haem product formed
on treating myoglobin with HOSCN is not certain.
LC/MS peptide mass mapping studies with HOSCN-treated
myoglobin
Figure 4
HOSCN
Oxidation of tryptophan residues on treatment of proteins with
(a) HSA (5 μM) treated with HOSCN (5–750 μM equimolar, 150-fold molar excess).
(b) Human plasma treated with HOSCN (0.25–500 μmol of plasma · ml−1 ). (c) Myoglobin
(white bars), apomyoglobin (striped bars), lysozyme (black bars) and STI (hatched bars) (all
5 μM) treated with HOSCN (5–325 μM equimolar, 65-fold molar excess) for 30 min, at pH 7.4,
before protein precipitation and hydrolysis. * represents a significant decrease (P < 0.05) in
tryptophan compared with the native protein or plasma by one-way ANOVA with Dunnetts
post-hoc testing. Values are means +
− S.E.M. (n 3).
where complete depletion of the protein thiol groups was observed
(Figure 1). Significant tryptophan oxidation was also observed on
reaction of plasma with HOSCN under conditions where depletion
of plasma thiols was observed ( 2.5 μmol of HOSCN per ml of
undiluted plasma; Figure 4b). Varying the incubation time (5 min–
6 h) of the protein or plasma with HOSCN had no significant effect
on the extent of tryptophan oxidation (results not shown).
Tryptophan oxidation was also observed on treating non-thiol
proteins, including myoglobin, lysozyme and STI with HOSCN
(Figure 4c). However, no evidence for loss of tryptophan was
obtained on treating free tryptophan, N-acetyl-tryptophan methyl
ester, Gly-Trp-Gly or Gly-Lys-Arg-Trp-Gly with HOSCN (results
not shown). These results indicate that the location or environment
of the tryptophan residues is critical to the reaction. With myoglobin, as the loss of tryptophan was particularly striking, the effect
of the haem group was investigated in experiments with the haemdepleted protein, apomyoglobin. With apomyoglobin, the extent
of tryptophan oxidation was less marked, with a significant
loss of tryptophan observed on reaction of the protein with a
20-fold molar excess of HOSCN, compared with a 5-fold
molar excess of oxidant with native myoglobin (Figure 4c).
Moreover, a small shift in the UV-visible absorbance spectrum
of the myoglobin in the 450–650 nm range, consistent with
As the loss of tryptophan observed with myoglobin was particularly marked (Figure 4c), the structure of the tryptophan-derived
oxidation products formed on HOSCN-treated myoglobin was
investigated in LC/MS peptide mass mapping studies. Reaction
of myoglobin (50 μM) with HOSCN (10–500 μM) for 30 min before trypsin digestion, resulted in the loss of the peptide that contains both tryptophan residues (GLSDGEWQQVLNVWGK, m/z
1816.8) (Figure 5a). In contrast, there was no significant change
in the concentration of the other peptides detected, including the
methionine-containing peptides (HPGDFGADAQGAMTK, m/z
1502.6 and HLKTEAEMK, m/z 1086.6), in accordance with the
amino acid analysis results reported above. The peptides were
identified by searching the NCBI database (http://www.ncbi.
nlm.nih.gov) using the Sequest search engine. The loss of the
tryptophan-containing peptide corresponded with the detection
of four new species: two major product peptides with m/z 1848.8
(Figure 5b) and 1880.8 (Figure 5c), and two minor product
peptides with m/z 1832.8 and 1864.8 (results not shown). These
mass-to-charge ratios correspond to the addition of 32, 64, 16 and
48 atomic mass units respectively, relative to the native peptide
(m/z 1816.8), and are consistent with the addition of two, four, one
and three oxygen atoms. These products were only observed on
addition of freshly prepared HOSCN, and were not detected with
aged (decomposed) solutions of HOSCN (Figure 5d). Similarly,
no loss of the tryptophan-containing peptide or product formation,
was observed on addition of solutions prepared in the absence of
either LPO, H2 O2 or SCN− (Figure 5d). No evidence for the
incorporation of SCN− , OCN− or other C-containing functional
group was obtained.
Subsequent data-dependent MS/MS (tandem MS) analyses
were performed with the major product peptides (m/z 1848.8
and 1880.8). Comparison of the resulting spectra with the peptide database Sequest, revealed that for the [M + 32 + H]+ peptide
(m/z 1848.8), the series of both y and b fragment ions observed
corresponding to cleavages along the peptide backbone, supported
the assignment of the addition of two atoms of oxygen to either
Trp7 or Trp14 (Figure 6 and Table 1). It appears that the peptide
modified by + 32 at Trp14 elutes slightly earlier than the corresponding peptide with modification at Trp7 . Thus the XC (cross
correlation score) values on comparing the spectra from the
Trp14 + 32 product were higher at a retention time of 27.4 +
−
0.2 min, whereas the Trp7 + 32 product had higher XC values
at a retention time of 27.9 +
− 0.2 min (Table 1). In each case,
these values were markedly higher than the XC observed for
c The Authors Journal compilation c 2008 Biochemical Society
446
Figure 5
C. L. Hawkins and others
Loss of tryptophan peptide and formation of products with tryptic digests of HOSCN-treated myoglobin
Myoglobin (50 μM) treated with HOSCN (10–500 μM) for 30 min at pH 7.4 before trypsin digestion and LC/MS/MS analysis. Results represent (a) loss of the tryptophan-containing peptide (m /z
1816.8) assessed by SRM of the y 12 fragment ion (m /z 1444.6), (b) formation of the + 32 product peptide (m /z 1848.8) assessed by SRM of the b 14 + 32 fragment ion (m /z 1645.8) and (c) formation
of the + 64 product peptide (m /z 1880.8) assessed by SRM of the b 14 + 64 fragment ion (m /z 1677.3). Results shown in (d) represent the concentration of the tryptophan-containing peptide (m /z
1816.8) observed with buffer alone, solutions prepared in the absence of either LPO, H2 O2 or SCN− and aged HOSCN solutions. In each case, the area of the peptide peak is standardized to the
area of a methionine-containing peptide (m /z 1502.6). * represents a significant change in peak area (P < 0.05) compared with the non-treated control sample, by one-way ANOVA with Dunnetts
post-hoc testing. Values are means +
− S.E.M. (n = 6).
Trp7 + 16/Trp14 + 16 (Table 1). With the [M + 64 + H]+ product
peptide (m/z 1880.8), the y and b fragment ions observed were
consistent with the addition of two oxygen atoms to Trp14 and two
oxygen atoms to Trp7 (Figure 6c and Table 1). Data-dependent
MS/MS analyses were not performed with the minor product
peptides (m/z 1832.8 and 1864.8) owing to the low abundance of
these species.
There was no difference in the nature or concentration of the
products observed on treating myoglobin (50 μM) with HOSCN
(50–250 μM) in the presence or absence of oxygen (results not
shown). However, a series of product peptides containing one
or more 18 O atoms were observed on treating myoglobin with a
5-fold molar excess of HOSCN in the presence of 50 % (v/v)
H2 18 O (Figure 7). In this case, the doubly charged species
[M + 32 + 2H]2 + (m/z 924.6) and [M + 64 + 2H]2 + (m/z 940.6)
were monitored for improved sensitivity. With the + 32 peptide,
products were observed with m/z 924.6, 925.6 and 926.6, which
corresponded to the doubly charged species with m/z 1848.8
(16 O,16 O), 1850.8 (18 O,16 O) and 1852.8 (18 O,18 O) respectively
(Figure 7). With the + 64 peptide, products were observed with
m/z 940.6, 941.6, 942.6, 943.6 and 944.6, which corresponded
to the doubly charged species with m/z 1880.8, 1882.8, 1884.8,
1886.8 and 1888.8 owing to the incorporation of one, two, three
or four 18 O atoms respectively (Figure 7). This demonstrates that
the tryptophan products contain oxygen atoms from H2 O rather
than O2 .
The concentration of the parent peptide (m/z 1816.8) and two
major product peptides (m/z 1848.8 and 1880.8) was determined
by SRM (selective reaction monitoring) (Table 1). The two
c The Authors Journal compilation c 2008 Biochemical Society
[M + 32 + H]+ peptides were quantified together, owing to the
poor resolution of these species and formation of an intense,
common, fragment ion (b14 , m/z 1645.8). As the methioninecontaining peptide (m/z 1502.6) is not oxidized by HOSCN (see
above), this peptide was employed as an internal standard to
account for differences in peptide recovery and any variation in
MS ionization efficiencies between samples. The concentration of
the product peptides formed was dependent on the molar excess
of HOSCN added. With equimolar HOSCN, the [M + 32 + H]+
peptides (m/z 1848.8) were the major species formed (Figure 5);
however, the relative concentration of this product decreased, and
an increase in the concentration of the [M + 64 + H]+ peptide (m/z
1880.8) was observed with > 2-fold molar excess of HOSCN
(Figure 5). This is consistent with the further oxidation of the
+ 32 peptide to the + 64 peptide in the presence of a higher
concentration of HOSCN.
The susceptibility of the tryptophan residues on myoglobin
(50 μM) to oxidation by HOSCN (250 μM) in the presence of
GSH (0–500 μM) was also investigated. A less marked decrease
in the concentration of parent tryptophan peptide was observed
on addition of HOSCN to myoglobin in the presence of a 5fold molar excess GSH (compared with protein) (Figure 8a).
Similarly, a corresponding decrease in the concentration of the
[M + 64 + H]+ peptide was observed (Figure 8b). In contrast,
an increase in the concentration of the [M + 32 + H]+ peptides
was observed with 0.5–5-fold molar excess of GSH (Figure 8b).
However, only low levels of the [M + 32 + H]+ peptides were
observed with a 10-fold molar excess of GSH (Figure 8b).
Ascorbate ( 5-fold molar excess compared with protein) also
Hypothiocyanous acid-mediated protein oxidation
Figure 6
447
Fragmentation ion spectra of [M + 32 + H]+ and [M + 64 + H]+ peptides from tryptic digests of HOSCN-treated myoglobin
Myoglobin (50 μM) treated with 50 μM (a and b) and 250 μM (c) HOSCN for 30 min at pH 7.4 before trypsin digestion and data-dependent MS/MS analyses. Data shown in (a) are (left-hand
side) an extracted ion chromatogram obtained from MS/MS of the doubly charged species m /z 924.6 ( + 32 product peptide, m /z 1848.8) with the fragment ions specific for the Trp14 + 32 product
(y 5 + 32, y 6 + 32, y 9 + 32 and b 13 ) and (right-hand side) the corresponding MS/MS spectrum. Equivalent data shown in (b) are (left-hand side) obtained from monitoring the specific fragment
ions from the Trp7 + 32 product (y 5 , y 9 and b 7 + 32). Data shown in (c) represent (left-hand side) a total ion chromatogram obtained from MS/MS of the doubly charged species m /z 940.6 ( + 64
product peptide, m /z 1880.8) and (right-hand side) the corresponding MS/MS spectra. Data are representative of at least six independent experiments.
prevented HOSCN-mediated myoglobin tryptophan oxidation
but, in this case, a greater concentration (25-fold molar excess)
of the scavenger was required to completely prevent loss of
the tryptophan peptide (Figure 8c). Again, an increase in the
concentration of the [M + 32 + H]+ peptides was observed in
experiments with lower amounts of ascorbate (1–10-fold molar
excess) (Figure 8d). With > 10-fold molar excess of ascorbate, a
significant reduction in both the [M + 32 + H]+ and [M + 64 + H]+
peptides was observed.
Effect of HOSCN on protein integrity
Protein unfolding was investigated using the fluorescent hydrophobic probe ANS and STI, which has been examined previously
c The Authors Journal compilation c 2008 Biochemical Society
448
Table 1
C. L. Hawkins and others
MS and HPLC features of the different myoglobin peptides observed after HOSCN treatment on tryptic digestion
XC represents the cross correlation score obtained on comparison of the experimental spectra with theoretical fragmentation ions using the Sequest software. This value represents the average of at
least six individual data files. Within the peptide sequences, # represents the addition of 32 atomic mass units to tryptophan and * represents the addition of 16 atomic mass units to tryptophan. The
tryptophan residues are indicated in bold.
Sequence
m / z ( M H+ )
Retention time (min)
SRM parameters
XC
Assignment
GLSDGEWQQVLNVWGK
GLSDGEWQQVLNVW#GK
GLSDGEW#QQVLNVWGK
GLSDGEW*QQVLNVW*GK
GLSDGEWQQVLNVW#GK
GLSDGEW#QQVLNVWGK
GLSDGEW*QQVLNVW*GK
GLSDGEW#QQVLNVW#GK
1816.8
1848.8
31.0
27.5
1442–1447a
1643–1648b
1848.8
28.0
1643–1648b
1880.8
24.1
1675–1680c
4.30
3.70
1.46
1.02
2.13
3.43
1.12
3.95
Parent peptide
Trp14 + 32
Trp7 + 32
Trp7 + 16/Trp14 + 16
Trp14 + 32
Trp7 + 32
Trp7 + 16/Trp14 + 16
Trp7 + 32/Trp14 + 32
a
SRM of y 12 fragment ion m /z 1444.6.
SRM of b 14 + 32 fragment ion m /z 1645.8.
c
SRM of b 14 + 64 fragment ion m /z 1677.3.
b
fluorescence, consistent with protein unfolding (Figure 9a). A
significant correlation (P = 0.0047) was obtained on comparison
of the extent of protein unfolding induced by HOSCN with the
level of tryptophan oxidation, as assessed by amino acid analysis,
using linear regression and correlation analysis (Figure 9b).
DISCUSSION
Figure 7 Formation of products containing
HOSCN in the presence of H2 18 O
18
O in myoglobin treated with
Spectra of the doubly charged ions observed on reaction of myoglobin (50 μM) with
HOSCN (250 μM) for 30 min in the presence of approx. 50 % (v/v) H2 18 O before trypsin
digestion and LC/MS/MS analyses. Data in (a–c) represent the ions observed with the
parent tryptophan-containing peptide, + 32 product peptide and the + 64 product peptide
respectively in the absence of H2 18 O. Data in (d–f) represent the ions observed with the parent
tryptophan-containing peptide, + 32 product peptides and the + 64 product peptide respectively
in the presence of H2 18 O. In (e), three products are observed with m /z 924.6, 925.6 and 926.6
and in (f), five products are observed with m /z 940.6, 941.6, 942.6, 943.6 and 944.6 consistent
with the incorporation of 16 O, 18 O or a combination of 16 O and 18 O. The traces shown as a
thinner line in (b), (c), (e) and (f) represent the background signal observed on analysis of the
equivalent region in the non-oxidant-treated control.
in studies with HOCl and HOBr [35,39]. Reaction of STI (5 μM)
with HOSCN (5–500 μM, equimolar to 100-fold molar excess)
for 30 min prior to addition of ANS, resulted in an increase in ANS
c The Authors Journal compilation c 2008 Biochemical Society
It is well-established that excessive or misplaced production of
HOCl from MPO results in tissue damage that is important in
disease, particularly in inflammatory disorders [42,43]; however,
it is predicted that similar amounts of HOCl and HOSCN will
be produced under physiological conditions, as the favoured
substrate for MPO is SCN− [5]. HOSCN may also play a role
in the initiation of cellular damage, as although HOSCN is a less
powerful oxidant than HOCl, it is more selective, particularly
for thiol groups, and as such may cause more extensive damage
to susceptible sites [22]. In the present study, the reaction of
HOSCN with both thiol-containing proteins, and proteins that
lack this functional group, has been examined. HOSCN-induced
protein modifications have been examined previously, using a
LPO/H2 O2 /SCN− system [12,16,28,33]. However, the presence
of the peroxidase enzymes in these studies prevents contributions
arising from direct oxidation of the target by activated haem
being discerned from those of HOSCN [12,16,28,33]. Tyrosine
binds at the active site of LPO, which may facilitate the reaction
between tyrosine and products of SCN− oxidation in experiments
carried out in the presence of the peroxidase [28]. It has also
been suggested that, in the presence of the complete LPO system,
an enzyme-bound, oxidized form of SCN− such as SCN+ is a
precursor for formation of (SCN)2 [12]. Hence, in the present
study, the LPO (and catalase that is added to remove any residual
H2 O2 ) was removed by filtration prior to reaction of HOSCN with
the proteins. The results reported here can therefore be directly
attributed to HOSCN and materials derived therefrom, and not
from direct peroxidase reactions.
With HSA and plasma, HOSCN reacts preferentially with
the free thiol group, with an apparent stoichiometry of one
HOSCN molecule consumed per thiol group. However, it
is possible that residual HOSCN may react with the TNB
generated on reaction of the thiols with DTNB during thiol
quantification; this would result in an underestimation of the
extent of thiol oxidation. In order to minimize this confounding
reaction, care was taken to ensure the protein was present in
Hypothiocyanous acid-mediated protein oxidation
Figure 8
449
Effect of GSH and ascorbate on the oxidation of myoglobin tryptophan residues
Myoglobin (50 μM) was reacted with HOSCN (250 μM) in the presence of (a and b) GSH (10–500 μM) or (c and d) ascorbate (Asc, 25 μM–5 mM) for 30 min before trypsin digestion and
LC/MS/MS analyses. Data in (a and c) represent the loss of the tryptophan peptide (m /z 1816) in the absence (white bar) and presence (grey bars) of HOSCN. (b and d) show the peak areas of the
+ 32 product peptide (m /z 1848; white bars) and the + 64 product peptide (m /z 1880; grey bars). Peak areas were quantified as outlined in Figure 5. * Represents a significant increase (P < 0.05),
and ‘a’ shows a significant decrease (P < 0.05), in the peptide peak area (concentration) compared with the area observed in the absence of GSH or ascorbate as assessed by one-way ANOVA with
Neuman–Keuls post-hoc testing. Values are means +
− S.E.M. (n = 6).
at least a 2.5-fold molar excess compared with HOSCN. Moreover, in each case, typically > 80 % of the initial oxidant added
was consumed on quantification by the TNB assay. However, this
may be an underestimate of the extent of oxidant consumption
if protein-derived oxidizing intermediates, such as RN-SCN
derivatives, are formed.
The 1:1 stoichiometry is consistent with previous studies [33],
and is attributed to the formation of sulfenyl thiocyanate (RSSCN) derivatives (eqn 6), which subsequently undergo further
oxidation or decomposition. There is evidence that RS-SCN
derivatives can react with another thiol to form a disulfide bond,
which results in the consumption of two thiol groups per molecule
of HOSCN (eqn 7):
R-SH + HOSCN → RS-SCN + H2 O
(6)
RS-SCN + R1 -SH → RS-SR1 + SCN− + H+
(7)
Eqn (7) is slow in comparison with eqn (6), probably as a result of
steric constraints to disulfide bond formation, which is greater in
the case of proteins compared with small substrates such as GSH
[33,44]. The RS-SCN adducts generated on reaction of HOSCN
with protein thiols are significantly more stable than the sulfenyl
chloride (RS-Cl) adducts formed in analogous experiments with
HOCl [45].
The reaction of HOSCN with other residues was investigated
by determining the ability of the proteins to incorporate a 14 C
label on treatment with HOS14 CN. With HSA, 14 C from HOS14 CN
was incorporated in both a reversible and non-reversible manner.
Treating the protein with low concentrations of HOS14 CN, resulted in completely reversible 14 C incorporation on addition of the
reductant DTT, consistent with the formation of RS-S14 CN
adducts [33]. With higher concentrations of HOS14 CN, a significant amount of the incorporated 14 C label (10 % and 30 %
with equimolar and 5-fold molar excess oxidant respectively) remained associated with the protein after DTT
treatment. However, these products are minor species,
typically accounting for 1 % of the added HOS14 CN. The
increase in reducible 14 C-containing HSA products observed with aged HOS14 CN compared with freshly prepared
solutions is attributed to the reaction of cyanate (O14 CN− ) with
protein thiols to give thiocarbamate (RS-CONH2 ) derivatives.
Thiol groups are known to be carbamylated by cyanate more
readily than other protein side-chains, but the resulting S-carbamyl
derivatives are not stable at physiological pH [46].
The EPO/H2 O2 /SCN− and MPO/H2 O2 /SCN− systems both
generate OCN− as a minor product [8,22]. Reaction of OCN− with
the free amino groups of lysine residues results in carbamylation
[46]. It is possible that OCN− may also be produced in LPOmediated oxidation of SCN− and that protein carbamylation could
contribute to the non-reversible incorporation of 14 C into HSA.
However, reaction of HSA with aged HOS14 CN resulted in a
decrease in the concentration of stable 14 C products, and no
evidence for loss of lysine residues was obtained in amino acid
analysis experiments. Moreover, only DTT-reversible incorporation of 14 C was observed on treating poly-lysine with HOS14 CN.
This reversible incorporation of 14 C is attributed to the formation
c The Authors Journal compilation c 2008 Biochemical Society
450
Figure 9
C. L. Hawkins and others
Protein unfolding observed on treating proteins with HOSCN
(a) The extent of protein unfolding observed on treating STI (5 μM) with HOSCN (5–500 μM) for
30 min at pH 7.4 as assessed by the fluorescence (λEx 387 nm, λEm 479 nm) of the hydrophobic
probe ANS. *Represents a significant increase (P < 0.05) in fluorescence (unfolding) compared
with the native protein by one-way ANOVA with Dunnetts post-hoc testing. Values are
means +
− S.E.M. (n = 6). (b) Linear correlation analysis plot displaying the level of tryptophan
oxidation, quantified by HPLC amino acid analysis, compared with protein unfolding, as assessed
by ANS fluorescence, of STI treated with various concentrations of HOSCN. A significant
correlation was observed with P = 0.0047 and R 2 = 0.7617 by linear regression analysis.
of amino thiocyanate (R-NH-SCN) derivatives (eqn 8). This
assignment is supported by the detection of high-molecular-mass
TNB-reactive products on reaction of poly-lysine with HOSCN.
This was unexpected, as it has been reported previously that RNHSCN derivatives formed on reaction of free amine groups with
HOSCN are readily hydrolysed in aqueous solutions (eqn 9) [12]:
RNH2 + HOSCN → RNH-SCN + H2 O
(8)
RNH-SCN + H2 O RNH2 + HOSCN
(9)
Non-reversible incorporation of S14 CN− and 35 SCN− into proteins
has been observed previously in studies with the LPO system
and BSA [28]. This incorporation was attributed to modification
of tyrosine, histidine and tryptophan residues based on the
results from studies with isolated poly-amino acids [28]. It was
suggested that (SCN)2 was produced by LPO-catalysed oxidation
of SCN− , which modified the proteins in a similar manner to
halogenation, as similar results were obtained in studies with
chemically generated (SCN)2 . However, as these studies were
carried out in the presence of LPO, it is also possible that
the oxidation of the aromatic amino acids was due to direct
peroxidase-mediated chemistry.
Reaction of HSA with HOSCN resulted in the loss of tryptophan residues, in contrast with experiments with HOCl and
HOBr where histidine, tyrosine, methionine, lysine and arginine
residue oxidation occurs in addition to reaction with tryptophan
(e.g. [35,39]). Tryptophan oxidation was observed in experiments
with plasma, suggesting that these residues are able to compete
effectively with other plasma components, such as ascorbate, for
HOSCN. Tryptophan oxidation was also observed on reaction
of HOSCN with other proteins, including myoglobin, apomyoglobin, lysozyme and STI. A greater extent of tryptophan loss
c The Authors Journal compilation c 2008 Biochemical Society
was observed with myoglobin compared with the haem-depleted
apomyoglobin, suggesting that the haem group plays a role in
the HOSCN-mediated tryptophan oxidation in this case. This is
supported by the detection of shifts in the myoglobin UV-visible
spectrum, and loss (approx. 15 %) of MbIII . The identity of the
haem product is not certain, as the spectral shifts observed are
not consistent with the formation of MbIV , MbO2 (oxymyoglobin) or SCN-myoglobin. However, significant tryptophan
oxidation is still observed on reaction of HOSCN in the absence
of the haem group (with apomyoglobin), with the extent of
tryptophan loss similar to experiments with non-haem proteins.
In contrast, tryptophan oxidation was not observed in experiments with HOSCN and the free amino acid, or small
peptides, including N-acetyl-tryptophan-methyl ester, Gly-TrpGly or Gly-Lys-Arg-Trp-Gly. This may be due to the tertiary
structure of the protein, making the interaction of HOSCN with
tryptophan residues more favourable. Alternatively, the oxidation
of tryptophan residues may be mediated by a protein-derived,
reactive intermediate, such as RNH-SCN. Protein-derived chloramines (RNH-Cl) and bromamines (RNH-Br) can readily oxidize
tryptophan residues [35,39], and evidence has been obtained in
the present study for the formation of RNH-SCN species on polylysine. Moreover, histidine and imidazole are able to enhance
the stability of HOSCN/OSCN− through the formation of more
stable, oxidizing, RN-SCN derivatives [12]. The formation of such
histidine-derived species was not investigated in the present study
owing to the insolubility of poly-histidine at pH 7.4. However,
generation of protein-derived TNB-reactive products on treatment
with HOSCN was not observed, and no loss of tryptophan was
observed with Gly-Lys-Arg-Trp-Gly.
The identity of the tryptophan-derived products was
investigated in LC/MS experiments with tryptic digests of
HOSCN-treated myoglobin, as the tryptophan residues of this
protein were particularly sensitive to oxidation by HOSCN. Both
myoglobin tryptophan residues (Trp7 and Trp14 ) are located on the
same peptide, GLSDGEWQQVLNVWGK (residues 1–16), after
digestion with trypsin. Evidence was obtained for the formation
of four product peptides, consistent with the addition of one to
four oxygen atoms, together with the parent tryptophan peptide.
The major product peptides observed were + 32 and + 64 atomic
mass units greater than the mass of the parent peptide. The
decrease in the concentration of the [M + 32 + H]+ peptides and
the corresponding increase in the [M + 64 + H]+ peptide observed
with 5-fold HOSCN, is consistent with HOSCN mediating
further oxidation of the former peptide. By LC/MS, complete
loss of the parent tryptophan peptide is observed on reaction
of myoglobin with a 10-fold molar excess of HOSCN. This is
consistent with the amino acid analysis results, where approx.
50 % loss of tryptophan is observed under these conditions, as
the oxidation of one tryptophan residue will result in the loss of
the parent tryptophan peptide (m/z 1816.8) by LC/MS analysis.
The [M + 32 + H]+ peptides and the [M + 64 + H]+ peptide
have been assigned to the formation of a mixture of Trp7 –O2
and Trp14 –O2 and Trp7 –O2 /Trp14 –O2 respectively, on the basis
of MS/MS analyses, and comparison of the fragmentation spectra
with the peptide database Sequest. The oxygen atoms are believed
to be incorporated from H2 O rather than O2 on the basis of the
multiple products formed in experiments with H2 18 O, and the lack
of a significant difference in reactivity observed on performing
the experiments in the absence of O2 . The relative intensity of
these product peaks is slightly different to the predicted 1:2:1
(for the + 32 products) and 1:4:6:4:1 (for the + 64 products),
which may be associated with the presence of multiple additional
products containing (natural abundance) 13 C. No evidence was
obtained for the incorporation of carbon atoms (e.g. OCN, CN,
Hypothiocyanous acid-mediated protein oxidation
Scheme 1
451
Proposed mechanism of the reaction of HOSCN with protein-derived tryptophan residues
SCN), as expected from the observation of stable 14 C-containing
products in the HSA/HOS14 CN experiments. This may be due to
a low abundance of these products, as quantification of 14 C by
scintillation counting is highly sensitive, or instability under the
MS conditions employed.
The major product peptides observed are consistent with the
addition of two oxygen atoms to tryptophan residues. Reaction
of HOCl with tryptophan is reported to result in the formation of
oxygenated products containing mainly a single oxygen atom,
including oxindolyalanine or 2-hydroxytryptophan, which are
generated via a 3-chloroindolenine or N-chloroindole species
[47]. It has also been proposed that HOCl induces the formation of
5,7-dihydroxy-2-indolone (addition of three oxygen atoms), but
this assignment was made solely on the basis of UV absorbance
spectra [48]. Other studies of the oxidation of tryptophan using
chlorine dioxide (ClO2 ) show evidence for the formation of NFK
(N-formylkynurenine) [49]. This latter product was also observed
on treating proteins with Fe(II)-EDTA/H2 O2 , but not HOCl
[50]. NFK, together with other oxygenated derivatives of tryptophan, including oxindolyalanine and dioxindolyalanine, are also
observed during photo-oxidation reactions [51].
It is believed that the dioxygenated tryptophan products
observed on treating myoglobin with HOSCN are due to the
formation of dioxindolyalanine or NFK which, by analogy with
HOCl, may occur via the formation of an indole-SCN intermediate
(Scheme 1). Thus related alkyl-SCN derivatives, including tertbutyl-SCN can be readily synthesized and isolated [52]. The
stability of these species is not certain, although it is possible
that the formation of tryptophan-SCN intermediates may account
for a proportion of the 14 C incorporation observed in experiments
with HOS14 CN. It may be that the dioxygenated products are
favoured with HOSCN rather than HOCl owing to the more ready
displacement of SCN− compared with Cl− by H2 O. However, the
reaction mechanism and assignment of these products is tentative,
and further experiments are in progress to determine the exact
structure of these species.
Loss of the tryptophan parent peptide and formation of the
+ 32 and + 64 product peptides was also observed on treating
myoglobin with HOSCN in the presence of excess (compared with
protein or oxidant) GSH and ascorbate. A significant reduction
in tryptophan oxidation to control levels was only observed
when the GSH was present in a 10-fold molar excess compared
with protein, or a 2-fold molar excess compared with HOSCN.
Ascorbate was not as effective at protecting tryptophan from
HOSCN-mediated oxidation, with a reduction in oxidation to
control levels observed with a 25-fold molar excess of ascorbate
compared with protein, or a 5-fold molar excess compared with
HOSCN. These results suggest that tryptophan residues compete
effectively with low-molecular-mass scavengers, such as GSH
and ascorbate, for HOSCN, and are consistent with the plasma
studies. These results appear to be in contrast with the HSA
results, where significant tryptophan oxidation is only observed
on consumption of the protein thiol group. This may be associated
with the presence of the haem group in myoglobin increasing the
relative reactivity of the tryptophan residues in this case. GSH
and ascorbate also reduced the HOSCN-mediated oxidation of
the + 32 product peptides to the + 64 product peptide. These
compounds may be acting by either directly scavenging HOSCN,
or by quenching potential protein-derived reactive intermediates.
It is well-established that GSH reacts readily with HOSCN [44],
but no information is available regarding the reaction of this
oxidant with ascorbate.
HOSCN also induced protein unfolding, which correlated
significantly (P < 0.0047) with the level of tryptophan oxidation,
in accordance with previous studies with HOCl and HOBr [35,39].
The ability of HOSCN to mediate unfolding is of potential
significance, as HOCl and HOBr can induce inactivation of
proteinase inhibitors via the oxidation of key aromatic residues,
with resultant protein unfolding compromising the ability of
the inhibitor to form a complex with the protease substrate
[35,39].
In summary, protein thiol groups are oxidized preferentially by
HOSCN on reaction with HSA and plasma. However, tryptophan
residues are also important targets for HOSCN. The HOSCNmediated oxidation of tryptophan residues appears to be specific
for proteins, and may occur via the generation of RN-SCN derivatives. The reaction of HOSCN with tryptophan is competitive,
and occurs in the presence of excess amounts of the antioxidants
ascorbate and GSH. The tryptophan-derived oxidation products
were characterized with myoglobin in LC/MS/MS experiments
where evidence was obtained for the formation of Trp7 –O2 , Trp14 –
O2 and Trp7 –O2 /Trp14 –O2 . These products were stable, and may
have the potential to act as specific biomarkers of HOSCNmediated reactions in vivo.
We would like to thank the National Health and Medical Research Council (Australia), the
National Heart Foundation (Australia) and the Australian Research Council (through
the Centres of Excellence Scheme) for financial support, Dr Philip Morgan and Mr
Mitchell Lloyd for helpful discussions and Dr Andrea Szuchman-Sapir for technical
assistance.
c The Authors Journal compilation c 2008 Biochemical Society
452
C. L. Hawkins and others
REFERENCES
1 Davies, M. J., Hawkins, C. L., Pattison, D. I. and Rees, M. D. (2008) Mammalian heme
peroxidases: from molecular mechanisms to health implications. Antiox. Redox Signaling
10, 1199–1234
2 Kettle, A. J. and Winterbourn, C. C. (1997) Myeloperoxidase: a key regulator of neutrophil
oxidant production. Redox Rep. 3, 3–15
3 Slungaard, A. and Mahoney, J. R. (1991) Thiocyanate is the major substrate for
eosinophil peroxidase in physiological fluids: implications for cytotoxicity. J. Biol. Chem.
266, 4903–4910
4 de Wit, J. N. and van Hooydonk, A. C. M. (1996) Structure, functions and applications of
lactoperoxidase in natural antimicrobial systems. Nethl. Milk Dairy J. 50, 227–244
5 van Dalen, C. J., Whitehouse, M. W., Winterbourn, C. C. and Kettle, A. J. (1997)
Thiocyanate and chloride as competing substrates for myeloperoxidase. Biochem. J. 327,
487–492
6 Thomas, E. L. and Fishman, M. (1986) Oxidation of chloride and thiocyanate by isolated
leukocytes. J. Biol. Chem. 261, 9694–9702
7 Winterbourn, C. C. (2002) Biological reactivity and biomarkers of the neutrophil oxidant,
hypochlorous acid. Toxicology 181, 223–227
8 Wang, Z., Nicholls, S. J., Rodriguez, E. R., Kummu, O., Horkko, S., Barnard, J., Reynolds,
W. F., Topol, E. J., DiDonato, J. A. and Hazen, S. L. (2007) Protein carbamylation links
inflammation, smoking, uremia and atherogenesis. Nature Med. 13, 1176–1184
9 Rehak, N. N., Cecco, S. A., Niemela, J. E. and Elin, R. J. (1997) Thiocyanate in smokers
interferes with the Nova magnesium ion-selective electrode. Clin. Chem. 43, 1595–1600
10 Scanlon, C. E. O., Berger, B., Malcom, G. and Wissler, R. W. (1996) Evidence for more
extensive deposits of epitopes of oxidized low density lipoproteins in aortas of young
people with elevated serum thiocyanate levels. Atherosclerosis 121, 23–33
11 Botti, T. P., Amin, H., Hiltscher, L. and Wissler, R. W. (1996) A comparison of the
quantitation of macrophage foam cell populations and the extent of apolipoprotein E
deposition in developing atherosclerotic lesions in young people: high and low serum
thiocyanate groups as an indication of smoking. Atherosclerosis 124, 191–202
12 Thomas, E. L. (1981) Lactoperoxidase-catalyzed oxidation of thiocyanate: equilibria
between oxidized forms of thiocyanate. Biochemistry 20, 3273–3280
13 Pruitt, K. M. and Tenovuo, J. (1982) Kinetics of hypothiocyanite production during
peroxidase-catalyzed oxidation of thiocyanate. Biochim. Biophys. Acta 704, 204–214
14 Tenovuo, J., Pruitt, K. M., Mansson-Rahemtulla, B., Harrington, P. and Baldone, D. C.
(1986) Products of thiocyanate peroxidation: properties and reaction mechanisms.
Biochim. Biophys. Acta 870, 377–384
15 Aune, T. M. and Thomas, E. L. (1977) Accumulation of hypothiocyanite ion during
peroxidase-catalyzed oxidation of thiocyanate ion. Eur. J. Biochem. 80, 209–214
16 Thomas, E. L. (1985) Products of the lactoperoxidase-catalysed oxidation of thiocyanate
and halides. In The Lactoperoxidase System: Chemistry and Biological Significance
(Pruitt, K. M. and Tenovuo, J. O., eds), pp. 31–53, Marcel Dekker, Inc., New York
17 Modi, S., Deodhar, S. S., Behere, D. V. and Mitra, S. (1991) Lactoperoxidase-catalyzed
oxidation of thiocyanate by hydrogen peroxide: 15 N nuclear magnetic resonance and
optical spectral studies. Biochemistry 30, 118–124
18 Nagy, P., Algunindigue, S. S. and Ashby, M. T. (2006) Lactoperoxidase-catalysed
oxidation of thiocyanate by hydrogen peroxide: a reinvestigation of hypothiocyanite by
nuclear magnetic resonance and optical spectroscopy. Biochemistry 45, 12610–12616
19 Figlar, J. N. and Stanbury, D. M. (2000) Thiocyanogen as an intermediate in the oxidation
of thiocyanate by hydrogen peroxide in acidic aqueous solution. Inorg. Chem. 39,
5089–5094
20 Nagy, P., Lemma, K. and Ashby, M. T. (2007) Kinetics and mechanism of the
comproportionation of hypothiocyanous acid and thiocyanate to give thiocyanogen in
acidic aqueous solution. Inorg. Chem. 46, 285–292
21 Barnett, J. J. and Stanbury, D. M. (2002) Formation of trithiocyanate in the oxidation of
aqueous thiocyanate. Inorg. Chem. 41, 164–166
22 Arlandson, M., Decker, T., Roongta, V. A., Bonilla, L., Mayo, K. H., MacPherson, J. C.,
Hazen, S. L. and Slungaard, A. (2001) Eosinophil peroxidase oxidation of thiocyanate:
characterization of major reaction products and a potential sulfhydryl-targeted cytotoxicity
system. J. Biol. Chem. 276, 215–224
23 van Dalen, C. J. and Kettle, A. J. (2001) Substrates and products of eosinophil
peroxidase. Biochem. J. 358, 233–239
24 Exner, M., Hermann, M., Hofbauer, R., Hartmann, B., Kapiotis, S. and Gmeiner, B. (2004)
Thiocyanate catalyzes myeloperoxidase-initiated lipid oxidation in LDL. Free Radical Biol.
Med. 37, 146–155
25 Lovaas, E. (1992) Free-radical generation and coupled thiol oxidation by
lactoperoxidase/SCN− /H2 O2 . Free Radical Biol. Med. 13, 187–195
26 Ashby, M. T., Carlson, A. C. and Scott, M. J. (2004) Redox buffering of hypochlorous acid
by thiocyanate in physiologic fluids. J. Am. Chem. Soc. 126, 15976–15977
Received 17 July 2007/24 July 2008; accepted 24 July 2008
Published as BJ Immediate Publication 24 July 2008, doi:10.1042/BJ20070941
c The Authors Journal compilation c 2008 Biochemical Society
27 Nagy, P., Beal, J. L. and Ashby, M. T. (2006) Thiocyanate is an efficient endogenous
scavenger of the phagocytic killing agent hypobromous acid. Chem. Res. Toxicol. 19,
587–593
28 Aune, T. M., Thomas, E. L. and Morrison, M. (1977) Lactoperoxidase-catalyzed
incorporation of thiocyanate ion into a protein substrate. Biochemistry 16, 4611–4615
29 Thomas, E. L. and Aune, T. M. (1978) Lactoperoxidase, peroxide, thiocyanate
antimicrobial system: correlation of sulfhydryl oxidation with antimicrobial action.
Infect. Immun. 20, 456–463
30 Lloyd, M. M., Van Reyk, D. M., Davies, M. J. and Hawkins, C. L. (2008) HOSCN is a more
potent inducer of apoptosis and protein thiol depletion in murine macrophage cells than
HOCl or HOBr. Biochem. J. 414, 271–280
31 Grisham, M. B. and Ryan, E. M. (1990) Cytotoxic properties of salivary oxidants.
Am. J. Physiol. 258, C115–C121
32 Wang, J. G., Mahmud, S. A., Thompson, J. A., Geng, J. G., Key, N. S. and Slungaard, A.
(2006) The principal eosinophil peroxidase product, HOSCN, is a uniquely potent
phagocyte oxidant inducer of endothelial cell tissue factor activity: a potential mechanism
for thrombosis in eosinophilic inflammatory states. Blood 107, 558–565
33 Aune, T. M. and Thomas, E. L. (1978) Oxidation of protein sulfhydryls by products of
peroxidase-catalyzed oxidation of thiocyanate ion. Biochemistry 17, 1005–1010
34 Nelson, D. P. and Kiesow, L. A. (1972) Enthalpy of decomposition of hydrogen peroxide
by catalase at 25 ◦C (with molar extinction coefficent of H2 O2 solution in the UV).
Anal. Biochem. 49, 474–478
35 Hawkins, C. L. and Davies, M. J. (2005) Inactivation of protease inhibitors and lysozyme
by hypochlorous acid: role of side-chain oxidation and protein unfolding in loss of
biological function. Chem. Res. Toxicol. 18, 1600–1610
36 Eyer, P., Worek, F., Kiderlen, D., Sinko, G., Stuglin, A., Simeon-Rudolf, V. and Reiner, E.
(2003) Molar absorption coefficients for the reduced Ellman reagent: reassessment.
Anal. Biochem. 312, 224–227
37 Pattison, D. I., Hawkins, C. L. and Davies, M. J. (2007) Hypochlorous acid-mediated
protein oxidation: how important are chloramine transfer reactions and protein tertiary
structure? Biochemistry 46, 9853–9864
38 Chapman, A. L. P., Winterbourn, C. C., Brennan, S. O., Jordan, T. W. and Kettle, A. J.
(2003) Characterisation of non-covalent oligomers of proteins treated with hypochlorous
acid. Biochem. J. 375, 33–40
39 Hawkins, C. L. and Davies, M. J. (2005) The role of aromatic amino acid oxidation,
protein unfolding and aggregation in the hypobromous acid-induced inactivation of
trypsin inhibitor and lysozyme. Chem. Res. Toxicol. 18, 1669–1677
40 Whitburn, K. D., Shieh, J. J., Sellers, R. M., Hoffman, M. Z. and Taub, I. A. (1982) Redox
transformations in ferrimyoglobin induced by radiation-generated free radicals in
aqueous solutions. J. Biol. Chem. 257, 1860–1869
41 Smith, D. W. and Williams, R. J. P. (1968) Analysis of the visible spectra of some sperm
whale ferrimyoglobin derivatives. Biochem. J. 110, 297–301
42 Heinecke, J. W. (1999) Mechanisms of oxidative damage by myeloperoxidase in
atherosclerosis and other inflammatory disorders. J. Lab. Clin. Med. 133, 321–325
43 Weitzman, S. A. and Gordon, L. I. (1990) Inflammation and cancer: role of
phagocyte-generated oxidants in carcinogenesis. Blood 76, 655–663
44 Ashby, M. T. and Aneetha, H. (2004) Reactive sulfur species: aqueous chemistry of
sulfenyl thiocyanates. J. Am. Chem. Soc. 126, 10216–10217
45 Davies, M. J. and Hawkins, C. L. (2000) Hypochlorite-induced oxidation of thiols:
formation of thiyl radicals and the role of sulfenyl chlorides as intermediates. Free Radical
Res. 33, 719–729
46 Stark, G. R., Stein, W. H. and Moore, S. (1960) Reactions of the cyanate present in
aqueous urea with amino acids and proteins. J. Biol. Chem. 235, 3177–3181
47 Fu, X., Wang, Y., Kao, J., Irwin, A., d’Avignon, A., Mecham, R. P., Parks, W. C. and
Heinecke, J. W. (2006) Specific sequence motifs direct the oxygenation and chlorination
of tryptophan by myeloperoxidase. Biochemistry 45, 3961–3971
48 Drozdz, R., Naskalski, J. W. and Sznajd, J. (1988) Oxidation of amino acids and peptides
in reaction with myeloperoxidase, chloride and hydrogen peroxide. Biochim. Biophys.
Acta 957, 47–52
49 Ogata, N. (2007) Denaturation of protein by chlorine dioxide: oxidative modification of
tryptophan and tyrosine residues. Biochemistry 46, 4898–4911
50 Manzanares, D., Rodriguez-Capote, K., Lui, S., Haines, T., Ramos, Y., Zhao, L.,
Doherty-Kirby, A., Lajoie, G. and Possmayer, F. (2007) Modification of tryptophan and
methionine residues is implicated in the oxidative inactivation of surfactant protein B.
Biochemistry 46, 5604–5615
51 Savige, W. A. (1971) Isolation and identification of some photo-oxidation products of
tryptophan. Aust. J. Chem. 24, 1285–1293
52 Kharasch, M. S., May, E. M. and Mayo, F. R. (1937) The addition of thiocyanic acid to
olefinic double bonds. J. Am. Chem. Soc. 59, 1580