4898 Biochemistry 2007, 46, 4898-4911

Denaturation of Protein by Chlorine Dioxide: Oxidative Modification of

Tryptophan and Tyrosine Residues†
Norio Ogata*
Research Institute, Taiko Pharmaceutical Co., Ltd., Suita, Osaka, Japan
ReceiVed September 1, 2006; ReVised Manuscript ReceiVed February 21, 2007

ABSTRACT: Oxychlorine compounds, such as hypochlorous acid (HOCl) and chlorine dioxide (ClO2), have
potent antimicrobial activity. Although the biochemical mechanism of the antimicrobial activity of HOCl
has been extensively investigated, little is known about that of ClO2. Using bovine serum albumin and
glucose-6-phosphate dehydrogenase of Saccharomyces cereVisiae as model proteins, here I demonstrate
that the antimicrobial activity of ClO2 is attributable primarily to its protein-denaturing activity. By solubility
analysis, circular dichroism spectroscopy, differential scanning calorimetry, and measurement of enzymatic
activity, I demonstrate that protein is rapidly denatured by ClO2 with a concomitant decrease in the
concentration of ClO2 in the reaction mixture. Circular dichroism spectra of the ClO2-treated proteins
show a change in ellipticity at 220 nm, indicating a decrease in R-helical content. Differential scanning
calorimetry shows that transition temperature and endothermic transition enthalpy of heat-induced unfolding
decrease in the ClO2-treated protein. The enzymatic activity of glucose-6-phosphate dehydrogenase
decreases to 10% within 15 s of treatment with 10 µM ClO2. Elemental analyses show that oxygen, but
not chlorine, atoms are incorporated in the ClO2-treated protein, providing direct evidence that protein is
oxidized by ClO2. Furthermore, mass spectrometry and nuclear magnetic resonance spectroscopy show
that tryptophan residues become N-formylkynurenine and tyrosine residues become 3,4-dihydroxyphe-
nylalanine (DOPA) or 2,4,5-trihydroxyphenylalanine (TOPA) in the ClO2-treated proteins. Taking these
results together, I conclude that microbes are inactivated by ClO2 owing to denaturation of constituent
proteins critical to their integrity and/or function, and that this denaturation is caused primarily by covalent
oxidative modification of their tryptophan and tyrosine residues.

Proteins are oxidized by many kinds of reactive species, organisms bind to the surface of phagocytes (neutrophils)
such as superoxide anions (O2-•), hydroxyl radicals (•OH), and are incorporated into phagosomes that eventually fuse
singlet oxygen (1∆g, 1O2), ozone (O3), hydrogen peroxide with lysosomes, followed by a burst of oxygen consumption
(H2O2), and hypochlorous acid (HOCl) (1-7). Some of these (15). The oxygen is then converted to highly reactive oxygen
oxidants are generated in the biological systems. For instance, species such as O2-• and H2O2. Myeloperoxidase, together
O2-• is generated during the NADPH oxidase-dependent with H2O2 from lysosomes, is released into the phagosome.
“respiratory burst” in polymorphonuclear neutrophilic leu- The initial product of the myeloperoxidase-H2O2-chloride
kocytes (neutrophils) (8). The oxidative activities of these system in the resulting phagolysosome is HOCl; subsequently
species toward microbial proteins are believed to be a chlorine (Cl2), hydroxyl radical, and singlet oxygen are
protective measure against invading microbes in humans and formed (1, 16-19). Cl2 chlorinates protein and is suggested
other animals, but they are also a major cause of disease to play an important role in host defenses and inflammation
(9). For instance, excessive or misplaced production of these (16). Singlet oxygen is known to kill bacteria in a phagosome
biological oxidants causes host tissue damage, which has of neutrophil (17, 19).
been implicated in human disorders (10) such as atheroscle- HOCl plays a critical role in destroying microbial patho-
rosis, cataract, and inflammatory disease (9). gens (20). In this context, HOCl is considered to be a “natural
HOCl is a highly reactive oxychlorine compound generated innate disinfectant”. Proteins are major targets of HOCl, and
in vivo from H2O2 and chloride ions by the action of the reaction of proteins with HOCl results in modifications
myeloperoxidase (11). It is a major oxidant that is generated to their side chains, backbone fragmentation, and cross-
when neutrophils and macrophages are activated in inflam- linking (21). Because proteins are major targets of HOCl,
matory sites. It is released from activated neutrophils during oxidative reactions of HOCl with amino acids and peptides
the respiratory burst (12, 13) and is implicated in many have been widely studied (22). HOCl readily oxidizes sulfur-
bactericidal and cytotoxic responses (14). Opsonized micro- containing amino acids, namely, cysteine and methionine
(23). HOCl also oxidizes tryptophan, tyrosine, and histidine
†
This work was supported by Taiko Pharmaceutical Co., Ltd. residues of proteins, resulting in protein unfolding (10).
* Correspondence should be addressed to Norio Ogata, MD, PhD, Among the oxychlorine compounds with oxidative ability,
at Research Institute, Taiko Pharmaceutical Co., Ltd., 3-34-14 Uchi-
honmachi, Suita, Osaka 564-0032, Japan. Phone: 81-6-6382-3100. chlorine dioxide (ClO2) is known to be extremely potent.
Fax: 81-6-6382-1152. E-mail: [email protected]. Owing to the presence of one unpaired electron in its
10.1021/bi061827u CCC: $37.00 © 2007 American Chemical Society
Published on Web 03/31/2007
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4899
molecular orbital (24), it is a free radical. While the presence 2 N HCl for 3 min in a 300 mL flask covered by aluminum
of ClO2 in higher organisms, such as human, is not reported foil. The ClO2 (volatile and soluble in water) generated in
in the literature, it is generated in bacteria by a chloroper- the reaction mixture was flushed (bubbled) with a flow of
oxidase-catalyzed dismutation reaction of chlorite (ClO2-), air at 2 L/min for 1.5 min into 50 mL of water chilled on
which forms chloride (Cl-), chlorate (ClO3-), oxygen, and ice. The ClO2 thus recovered in water (∼20 mM) was kept
ClO2 (25). Chlorite is generated by the reduction of either at room temperature in small aliquots in fully filled, tightly
chlorate by chlorate reductase or perchlorate by perchlorate capped amber bottles until use.
reductase (26-28). Chlorite is dismutated to chloride and Iodometric Titration of ClO2. A 25 µL aliquot of the ClO2
oxygen by chlorite dismutase (29). ClO2 is known to have solution to be assayed was mixed with 475 µL of 105 mM
potent antimicrobial activity against bacteria, viruses, and KI. The mixture was titrated with 10 mM Na2S2O3 using a
protozoae (30-38). Despite a wealth of experimental data micropipet until the brownish yellow color of iodine disap-
on the antimicrobial activity of ClO2, however, little is known peared. One mole of ClO2 corresponds to 1 mol of Na2S2O3.
about its biochemical mechanism. The experimental error of this assay was (2%. The validity
The aim of this work was to elucidate the biochemical and accuracy of this assay were confirmed by the absorp-
mechanism of the antimicrobial activity of ClO2, on the basis tion of ClO2 in water, using an absorption maximum, λmax,
of the working hypothesis that ClO2 denatures constituent of 359 nm, and a molar extinction coefficient, max, of
proteins of microbes that are critical for their integrity and/ 1230 M-1‚cm-1 (25).
or function, and thereby destroys their viability. To test this Reactions of Protein, Peptide, and Amino Acid with ClO2.
hypothesis, I used bovine serum albumin (BSA)1 and Unless otherwise specified, standard reactions of protein,
glucose-6-phosphate dehydrogenase (G6PD) of baker’s yeast peptide, and amino acid with ClO2 were set up as follows in
(Saccharomyces cereVisiae) as model proteins. I clearly phosphate-buffered saline (PBS) (20 mM sodium phosphate
demonstrate here that ClO2 indeed denatures proteins as buffer, pH 7.0, 130 mM NaCl). A ClO2 stock solution was
assessed by changes in various physicochemical parameters added (final concentration 1 mM for protein and 0.3 mM
after treatment with ClO2. I further demonstrate that the for peptide or amino acid) to a solution of protein (final
denaturation of the proteins is caused by oxidative modifica- concentration 0.5 mg/mL), peptide or amino acid (each final
tion of their tryptophan and tyrosine residues. More specif- concentration 0.2 mM), and the reaction was carried out at
ically, I show that tryptophan becomes N-formylkynurenine, 25 °C for 2 min. The reaction was terminated by adding a
and that tyrosine becomes 3,4-dihydroxyphenylalanine (DOPA) 2-fold molar excess of 200 mM Na2S2O3. As a “nontreated”
or 2,4,5-trihydroxyphenylalanine (TOPA) owing to oxidative control, ClO2 that had been premixed with a 2-fold molar
covalent modifications. To the best of my knowledge, this excess of Na2S2O3 was added to the protein, peptide, or
is the first report to demonstrate that proteins are oxidatively amino acid solution; ClO2 is reduced to a chlorite anion
modified and denatured by ClO2. (ClO2-) by an equimolar amount of Na2S2O3.
Measurement of Protein Solubility. The solubility of the
MATERIALS AND METHODS protein was measured as described (39) with some modifica-
Materials. Sodium chlorite (NaClO2), trifluoroacetic acid, tions. In brief, the BSA and G6PD solutions (each 0.5 mg/
3,4-dihydroxy-L-phenylalanine (DOPA), sodium thiosulfate mL in PBS, 200 µL) were treated with varying concentrations
(Na2S2O3), sodium azide (NaN3), high-performance liquid of ClO2 at 25 °C for 2 min, and the reaction was terminated
chromatography (HPLC) grade acetonitrile, and a reverse- by the addition of an excess amount of Na2S2O3 as above.
phase HPLC column (Cosmosil 5C18-AR-300, 4.6 × The mixture was then dialyzed for 72 h against 2000 volumes
250 mm) were obtained from Nacalai Tesque (Kyoto, Japan). of 10 mM sodium phosphate buffer (pH 7.0) at 4 °C. The
L-Tryptophan, L-tyrosine, BSA (Cohn fraction V, code A dialyzed solution was centrifuged at 15 000 rpm (22000g)
3059, >83% pure), G6PD (code G 6378, >85% pure), for 30 min at 4 °C in a 1.5 mL conical plastic centrifuge
aminopeptidase M (porcine kidney microsomal, code L tube, and the protein concentration of the supernatant was
0632), proteinase K, subtilisin, superoxide dismutase (from quantified by a Bradford dye-binding assay (40) using BSA
Escherichia coli), Bradford reagent, and L-kynurenine were as a standard. Protein concentrations were corrected for
obtained from Sigma. Trypsin (code V5111) was obtained volume changes caused by the dialysis. As a control, ClO2
from Promega (Madison, WI). 18O2 gas (95% isotopic purity) premixed with excess Na2S2O3 was added to the protein
and H218O (99% isotopic purity) were purchased from Taiyo solution, and the mixture was dialyzed and centrifuged as
Nissan (Tokyo, Japan). All synthetic peptides (>95% pure) above. Oxidation of protein will not affect the assay of
were obtained from Global Peptide Services (Fort Collins, protein concentration, because the assay employed does not
CO). rely on reductive capacity of protein.
Preparation of ClO2. ClO2 was prepared by mixing Calorimetric Measurements of Thermal Denaturation of
50 mL of warmed (∼48 °C) 550 mM NaClO2 with 5 mL of Protein. Differential scanning calorimetry (DSC) measure-
ments of thermal denaturation (unfolding) of protein were
1
Abbreviations: BSA, bovine serum albumin; G6PD, glucose-6- made on a model Q1000 differential scanning calorimeter
phosphate dehydrogenase; rpm, revolution per minute; DSC, differential (TA Instruments, New Castle, DE) using a 15 µL cell for
scanning calorimetry; CD, circular dichroism; HPLC, high-performance
liquid chromatography; MS, mass spectrometry; MALDI-TOF-MS, BSA and G6PD solutions (each 10 mg/mL) treated with or
matrix-assisted laser desorption ionization time of flight MS; ESI, without 15 mM ClO2 in PBS at 25 °C for 2 min. The reaction
electrospray ionization; NMR, nuclear magnetic resonance; COSY, with ClO2 was terminated by 30 mM Na2S2O3. ClO2 that
correlation spectroscopy; NOESY, nuclear Overhouser effect spectros- had been premixed with Na2S2O3 was added to the protein
copy; NFK, N-formylkynurenine; DOPA, 3,4-dihydroxy-L-phenylala-
nine; TOPA, 2,4,5-trihydroxy-L-phenylalanine (3,4,6-trihydroxy-L- solution as a “nontreated” control. The samples were
phenylalanine). degassed before being placed in the calorimeter cells. The
4900 Biochemistry, Vol. 46, No. 16, 2007 Ogata

calorimeter was calibrated using indium (melting point The detection limit of the titrator was 0.01% (w/w). Oxygen
156.61 °C, heat of melting 28.7 J/g). The scans were content was determined by a CHN-O-Rapid oxygen analyzer
performed under a flow (50 mL/min) of nitrogen gas at a (Elementar Analysensysteme GmbH, Hanau, Germany). The
constant rate of 5.0 °C/min from 25 °C to 95 °C. The DSC accuracy of this analysis was (0.15%.
curves obtained were corrected for the instrument baseline, Amino Acid Analysis of Protein. BSA and G6PD (each
which was obtained by heating the solvent alone. The thermal 200 µg) were first treated with 1 mM ClO2 at 25 °C for
transition temperature of unfolding of protein (Tm, the 2 min in 50 µL of PBS. The reaction was terminated with
temperature at which half of the protein molecules are 2 mM Na2S2O3, and the mixtures were lyophilized. The
denatured) was determined for each protein sample by curve lyophilized materials were treated with 1 mL of 30 mM
fitting, assuming a two-state transition of unfolding. The 2-mercaptoethanol containing 0.6 M guanidine hydrochloride
thermal transition enthalpy (∆H) at Tm was obtained from at 25 °C for 2 h, and then with 30 mM 4-vinylpyridine at
the area of the unfolding curve below the baseline. The 25 °C for 30 min to derivatize cysteine residues to S-β-(4-
repeated DSC scans for each sample showed an estimated pyridylethyl)cysteine. The derivatized materials were dia-
error for Tm and ∆H of (0.5 °C and (100 kJ/mol, lyzed against water and lyophilized. The lyophilized materials
respectively. were next hydrolyzed in 100 µL of 6 N HCl with 1% (w/v)
Circular Dichroism (CD) Spectroscopy. CD spectra of phenol at 110 °C for 24 and 72 h for common amino acids,
ClO2 (1 mM)-treated (25 °C, 2 min) and nontreated BSA and in 4.2 N NaOH with 0.5% (w/v) starch at 110 °C for 16
and G6PD proteins (each 3 mg/mL) were acquired in PBS h for tryptophan. The hydrolysates were analyzed by an
on a spectropolarimeter (model J-715, Jasco, Tokyo, Japan). amino acid analyzer (model L-8500, Hitachi, Tokyo, Japan).
Before measurement, the ClO2-treated and nontreated pro- The amount of serine was determined by extrapolating the
teins were dialyzed against PBS at 4 °C for 72 h, and values obtained at 24 and 72 h of hydrolysis to 0 h.
centrifuged at 22000g for 30 min at 4 °C to remove insoluble High-Performance Liquid Chromatography (HPLC). Pep-
materials. The protein concentration of the supernatant was tide or amino acid was loaded on a reverse-phase column
determined by Bradford reagent (40). The CD spectrum of mounted on an HPLC system (model 880, Jasco). The sample
the supernatant was acquired in a quartz cell with a 0.1 cm was eluted at a flow rate of 1.0 mL/min at 25 °C, first using
path length, and the result was normalized to a protein 0.1% (v/v) trifluoroacetic acid for 6 min, and then using a
concentration of 1.0 mg/mL. linear gradient of 0 to 36% (v/v) acetonitrile in 0.1%
Assay of Enzymatic ActiVity. The enzymatic activity of trifluoroacetic acid over 54 min. The eluant was monitored
G6PD was assayed spectrophotometrically as suggested by by absorption at 215 or 270 nm. When a large amount of a
the manufacturer (Sigma) using NADP and glucose-6- peak material was needed for a subsequent analysis, this
phosphate as substrates. The reaction of G6PD (80 µg/mL) HPLC separation was repeated many times on the same scale
with 10 µM ClO2 was done at 25 °C in PBS, and was and the relevant peak fractions were pooled. The pooled
terminated by adding Na2S2O3 to a final concentration of material was lyophilized for subsequent analyses. Care was
20 µM. A control G6PD reaction was set up, in which exercised to protect tyrosine-containing peptide derivatives
10 µM ClO2 was premixed with 20 µM Na2S2O3. from exposure to light.
Consumption of ClO2 by Protein. BSA and G6PD (each HPLC-Mass Spectrometry (MS). An aliquot (about
0.5 mg/mL in PBS, 0.5 mL) were mixed with ClO2 (final 10 nmol) of an amino acid derivative obtained as a peak
concentration 1.05 mM for BSA and 0.86 mM for G6PD) fraction from HPLC was lyophilized, and then dissolved in
for varying time intervals at 25 °C. The reaction was 20 µL of solvent A (0.05% (v/v) formic acid, 50% (v/v)
terminated by adding 2 M KI (final concentration 100 mM). acetonitrile). Next, a 1 µL sample was injected into an
The ClO2 remaining in the mixture was titrated with 10 mM LCMS-IT-TOF mass spectrometer using an LC-20AD pump
Na2S2O3 as described above for the iodometric titration. The (both Shimadzu) at a flow rate of 100 µL/min and solvent
time course curve obtained was fitted to a one-phase A as the mobile phase. The sample was ionized by electro-
exponential decay curve using a computer program (Graph- spray ionization (ESI), and positive ions (primarily [M +
pad Prism 3.02, Hallogram Publishing, Aurora, CO). H]+ and [M + 2H]2+) were acquired at a detector voltage
Absorption and Fluorescence Spectra. Absorption spectra of 1550 V. Tyrosine (molecular weight 181.19) was used to
were acquired by a UVIDEC-505 spectrophotometer (Jasco), calibrate the apparatus. The amino acid sequence of the
and fluorescence spectra (excitation at 280 nm) were acquired peptide was determined by de noVo peptide sequencing by
by a spectrofluorophotometer (model RF-5300PC, Shimadzu, HPLC-ESI-MS/MS using a Bruker Esquire 3000plus mass
Kyoto, Japan). spectrometer, and the fragment ions obtained were classified
Elemental Analysis of Protein. BSA (20 mg) dissolved in according to Biemann’s nomenclature for b-series and
1 mL of PBS was treated with 1 mM ClO2 at 25 °C for y-series ions (41).
2 min; the reaction was then terminated by 2 mM Na2S2O3. Matrix-Assisted Laser Desorption Ionization Time-of-
Next, the mixture was dialyzed extensively against 1000 Flight Mass Spectrometry (MALDI-TOF-MS). Some peptides
volumes of distilled water for a week, changing the water were analyzed by a Voyager-DE mass spectrometer (Per-
three times a day. The dialyzed protein was lyophilized, and Septive Biosystems, Foster City, CA) in linear mode at an
its weight was carefully measured under a stream of argon acceleration voltage of 20 kV, using R-cyano-4-hydroxy-
gas. For measurement of chlorine content, the protein was cinnamic acid (10 mg/mL in 0.05% trifuoroacetic acid, 50%
combusted in an oxygen flask, the combustion gas was acetonitrile) as the matrix. Samples were ionized at a laser
absorbed by a hydrazine solution, and the solution was repetition frequency of 20 Hz.
subjected to potentiometric titration using a potentiometric Nuclear Magnetic Resonance (NMR) Spectroscopy. A
titrator (model AT-118, Kyoto Denshi Kogyo, Kyoto, Japan). derivatized peptide or amino acid (each 2 mg), obtained as
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4901
a peak fraction from HPLC and lyophilized, was dissolved
in 600 µL of D2O, and its 1H NMR signal was acquired in
either a one- or two-dimensional mode using an NMR
spectrometer (model Unity Inova 600, Varian, Palo Alto, CA)
at 25 °C at 599.855 MHz for 1H. Partial deuterium water
(HDO, 4.700 ppm) was used as a standard. Spectra were
acquired under the following conditions: preacquisition delay
time ) 0.000 s, acquisition time ) 3.277 s, pulse width )
100.000 µs, spectral width ) 6000.6 Hz. Two-dimensional
(1H-1H) total correlation spectroscopy (COSY) of amino
acid derivative or derivatized peptide was performed using
a mixing sequence of 100 ms flanked by two 2 ms trim pulses
FIGURE 1: Solubility of protein treated with or without ClO2. BSA
with 256t1 and 2028t2 data points with a pulse repetition time (A) and G6PD (B) (each 0.5 mg/mL) were treated in PBS with
of 6.0 s. After two-dimensional Fourier transformation, the varying concentrations of ClO2 at 25 °C for 2 min, and the mixtures
spectra resulted in 2048 × 2048 data points that were phase- were dialyzed after the addition of Na2S2O3 to a final concentration
and baseline-corrected in both dimensions. A two-dimen- of 2 mM. The dialyzed solutions were centrifuged, and the protein
sional NOESY spectrum was acquired for a spectrum width concentrations in the supernatant were quantified (filled circles).
As a control, ClO2 that had been premixed with 2 mM Na2S2O3
of 6 kHz × 6 kHz and data points of 2000 × 2000. The was added to the protein solution, and then treated as above (open
pulse repetition time was 3.5 s. circles). Each point represents a mean of two experiments.
Aminopeptidase M Digestion of Peptide. ClO2-treated
peptide (∼100 nmol) was first separated by HPLC as meter). This water was next bubbled extensively using 18O2
described above, and a fraction corresponding to a major gas until dissolved oxygen gas concentration became 14 mg/
modified peptide peak was collected. This fraction was L. The peptide treated with ClO2 was then mixed with excess
lyophilized and digested with 2 units of aminoptidase M in Na2S2O3, and loaded on HPLC to isolate modified peptide
200 µL of 10 mM sodium phosphate buffer (pH 7.0) at as described above. The isolated peptide was analyzed by
25 °C for 4 h to release amino acids from the modified HPLC-ESI-MS/MS to measure molecular weight of modi-
peptide. The amino acids released were then separated by fied amino acid residue in the peptide from fragment ions.
HPLC, and a fraction corresponding to a peak of modified In another experiment, peptide was treated with Cl18O2 as
amino acid was collected. This fraction was subjected to above using common distilled water.
further analyses after lyophilization. Edman Degradation of Peptide. Peptide (∼0.5 nmol) was
Proteinase K and Aminopeptidase M Digestion of Protein. subjected to automated Edman degradation by using a Procise
BSA and G6PD (each 1 mg) were treated with 1 mM ClO2 protein sequencer (Applied Biosystems, Foster City, CA)
in 2 mL of PBS at 25 °C for 2 min, and the reaction was according to the manufacturer’s recommended protocol.
terminated by 2 mM Na2S2O3. Next, proteinase K powder
was added to a final concentration of 0.5 mg/mL and the RESULTS
mixture was incubated at 37 °C for 2 h. The mixture was
then subjected to ultrafiltration using a membrane with a Denaturation of Protein by ClO2. When BSA and G6PD
molecular weight cutoff of 10 000. The filtrate was collected were treated with ClO2, their solubility decreased in a
and lyophilized. The lyophilized material was dissolved in concentration-dependent manner (Figure 1). This result
170 µL of distilled water containing 5 units of aminopep- suggests that protein is denatured by ClO2. Next, the
tidase M. The mixture was incubated at 25 °C for 4 h, and physicochemical parameters of the proteins on thermal
separated by HPLC to isolate modified amino acids. denaturation (unfolding) were measured. For BSA, the
Synthesis of Cl18O2. NaCl (0.64 g) and Na2Cr2O7 (12 mg) denaturation temperature (Tm) decreased from 74.5 °C to
were dissolved in 1.7 mL of H218O containing 4.3 µL of 2 70.7 °C upon treatment with ClO2 (Figure 2A, B); in addition
N HCl. The solution was subjected to electrolysis at 60 °C the thermal transition enthalpy (∆H) decreased from 1390
for 3 h at 0.6 A using platinum electrodes. The solution was to 460 kJ/mol after treatment. The Tm of untreated G6PD
next slowly and gently mixed with 187 µL of 10.25 M H2- was 55.6 °C, whereas ClO2-treated G6PD did not show a
SO4, 400 µL of 11.4 N HCl, 850 µL of 10 N NaOH, and clear unfolding temperature, indicating that G6PD had only
510 µL of 30% (w/v) H2O2 in this order at 25 °C. The mixed a marginal amount of higher-order protein structure after
solution was next boiled for 2 min to break down excess treatment with ClO2. The ∆H of G6PD decreased from
H2O2 into oxygen and water. The solution was finally mixed 1090 kJ/mol to 0 kJ/mol (actually not detectable) after ClO2
with 1.2 mL of 2 N HCl and bubbled gently with air to treatment (Figure 2C,D). Taken together, these results
recover Cl18O2 into 5 mL of distilled water chilled on ice. indicate that the higher-order structure of protein becomes
18
O-Labeling Experiment. Peptide (0.2 mM) was treated less stable after treatment with ClO2, and is more easily
with 0.3 mM ClO2 at 25 °C for 2 min in PBS. In this disrupted by an increase in temperature.
experiment, all waters used for the reaction were either H218O This view was further supported by circular dichroism
in place of common distilled water (H216O) or common (CD) spectroscopy, in which the ellipticity of BSA and G6PD
distilled water that had been bubbled beforehand with 18O2 at ∼220 nm (which represents primarily the R-helical content
gas. The 18O2-bubbled water was made first by degassing of protein) became closer to zero when the proteins were
distilled common water under reduced pressure, and then it treated with ClO2 (Figure 3). This reduction indicates that
was bubbled using N2 gas until dissolved oxygen gas the R-helical content in BSA and G6PD decreased after
concentration decreased to 2 mg/L (measured by an oxygen treatment with ClO2. However, the decrease in R-helical
4902 Biochemistry, Vol. 46, No. 16, 2007 Ogata

FIGURE 3: CD spectra of protein treated with or without ClO2. BSA

(A) and G6PD (B) (each 3 mg/mL) were treated with (solid line)
or without (broken line) 1 mM ClO2 in PBS at 25 °C for 2 min,
and the reaction mixtures were dialyzed after the addition of
FIGURE 2: Differential scanning calorimetry of heat-induced Na2S2O3 to a final concentration of 2 mM. CD spectra of the
denaturation of protein. Heat-induced denaturation profiles of BSA dialyzed protein solutions were measured after removal of insoluble
(A, B) and G6PD (C, D) treated with (A, C) or without (B, D) materials by centrifugation. As a control, 1 mM ClO2 that had been
ClO2 were measured by a differential scanning calorimeter. Proteins premixed with 2 mM Na2S2O3 was added to the protein solution,
(each 10 mg/mL) were treated with 15 mM ClO2 in PBS at 25 °C and CD spectra were measured as above. The ellipticity was
for 2 min, and the reaction was terminated by 30 mM Na2S2O3. normalized to a protein concentration of 1.0 mg/mL. Experiments
ClO2 that had been premixed with Na2S2O3 was added to the protein were done twice independently, and similar results were obtained.
solution as a “nontreated” control (B, D). The reaction mixtures The data shown is one of them.
were directly analyzed by differential scanning calorimetry. Scans
were performed from 25 to 95 °C at a constant rate of 5.0 °C/min.
Baselines used for calculation of endothermic transition enthalpy
∆H are shown by broken lines. ∆H was calculated from the area
of the curve below the broken line. Experiments were done twice
independently, and similar results were obtained. The data shown
is one of them.

content, as demonstrated by the ellipticity, was relatively

small in BSA (Figure 3A).
Inhibition of G6PD Enzymatic ActiVity by ClO2. The
protein-denaturing activity of ClO2 was next confirmed by
measuring the enzymatic activity of G6PD. When G6PD was FIGURE 4: Inactivation of the enzymatic activity of G6PD by ClO2.
treated with ClO2, its enzymatic activity decreased markedly G6PD (80 µg/mL) was treated with 10 µM ClO2 in PBS at 25 °C
in both a time- and concentration-dependent manner for the times indicated (filled circles). After the reaction of 4 min,
(Figure 4, filled circles). This inactivation of G6PD enzy- a second aliquot of ClO2 was added to the mixture to a final
matic activity by ClO2 appeared within 15 s of the addition concentration of 10 µM (arrow). The reaction of ClO2 was
terminated by the addition of a 2-fold molar excess of Na2S2O3.
of 10 µM ClO2, and then reached a plateau; the enzymatic The enzymatic activity of G6PD was then assayed. As a control,
activity decreased to 10% within 15 s. The addition of a G6PD was treated with 10 µM ClO2 that had been premixed with
second aliquot of 10 µM ClO2 at 4 min after the first addition 20 µM Na2S2O3, and then its enzymatic activity was assayed (open
(Figure 4, arrow) inactivated the remaining enzymatic circles). The inset shows the enzymatic activity of G6PD (80 µg/
activity. This observation suggests that the first aliquot of mL) treated in PBS at 25 °C for 2 min with varying concentrations
of ClO2 indicated. The reaction was terminated by an excess amount
ClO2 added to the protein solution had been totally “con- of Na2S2O3. Each point represents a mean of two experiments.
sumed” presumably due to its chemical reaction with G6PD
molecules, and consequently it was not able to inactivate consumed. It further suggests that protein is covalently
the remaining enzymatic activity. modified by ClO2.
Chemical Reaction of Protein with ClO2. The above idea The covalent modification of protein by ClO2 was
was further confirmed in an experiment in which the amount examined by elemental analysis of BSA treated with ClO2
of ClO2 remaining after reaction with BSA or G6PD was (Table 1). Whereas the amount of chlorine in BSA did not
quantified. The purpose of this experiment was to determine increase after its reaction with ClO2, the amount of oxygen
how much ClO2 had been consumed by its reaction with increased from 26.90% (w/w) to 28.39% (w/w). This increase
protein. As shown in Figure 5, the ClO2 that had been added is significant, with respect to the accuracy of the method of
to a protein solution decreased exponentially and this elemental analysis used ((0.15%). This increase in the
decrease was temperature-dependent (Figure 5, insets). This oxygen content of the ClO2-treated BSA implies that each
observation suggests that protein is denatured by a chemical molecule of BSA has acquired about 64 oxygen atoms after
reaction involving ClO2, in which ClO2 molecules are its reaction with ClO2. By contrast, not even a single
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4903

Table 2: Amino Acid Composition of ClO2-Treated Proteina

BSA G6PD
amino acid expected found expected found
Asp 54 54.7 58 58.5
Thr 33 32.8 22 21.2
Ser 28 29.8 30 29.5
Glu 79 82.1 53 54.7
Gly 16 18.4 32 33.8
Ala 47 46.7 29 29.4
1/2 Cys 35 37.0 1 1.3
Val 36 35.1 37 36.5
Met 4 2.1 11 9.6
Ile 14 13.5 27 28.3
Leu 61 61.9 45 47.8
FIGURE 5: Consumption of ClO2 by the reaction with protein. ClO2 Tyr 20 13.7 21 14.6
was mixed at 25 °C with 0.5 mg/mL BSA (A) or G6PD (B) in Phe 27 26.7 27 26.7
PBS for the time indicated, and the reaction was terminated by the Lys 59 56.2 43 41.8
addition of KI to a final concentration of 100 mM. The amount of His 17 16.1 10 10.1
ClO2 remaining in the reaction mixture was measured by iodometric Trp 2 0.7 5 2.5
titration. The curve was obtained by fitting the data to a one-phase Arg 23 22.8 25 24.5
exponential decay curve. The inset shows ClO2 concentration Pro 28 30.4 29 29.0
remaining after a 2 min reaction as a function of temperature. Each a
point represents a mean of two experiments. Values indicate moles of amino acid per mole of protein. Each
value, except for tryptophan, represents a mean of two experiments.
The value of tryptophan is a mean of three experiments.
Table 1: Elemental Analyses of BSA Treated with or without ClO2a
wt %
protein chlorine oxygen
BSA 0.01 26.90
ClO2-treated BSA < 0.01 28.39
BSA mixed with ClO2- b 0.01 26.86
BSA mixed with 0.1 mg of NaClc 0.29 ndd
a
Each value represents a mean of two experiments. BSA (20 mg)
was treated with 1 mM ClO2 in 1 mL of PBS at 25 °C for 2 min. The
reaction was terminated by 2 mM Na2S2O3. The reaction mixture was
then dialyzed extensively against distilled water and lyophilized. The
lyophilized protein was subjected to elemental analyses for chlorine
and oxygen. b Treated with 1 mM ClO2 that had been premixed with
2 mM Na2S2O3. ClO2 becomes ClO2- anion in this reaction. c Added
after the dialysis. The theoretical amount of chlorine in this mixture is
0.302% (w/w). d Not determined. FIGURE 6: Absorption and fluorescence spectra of protein treated
with or without ClO2. The absorption and fluorescence (inset,
molecule of chlorine was incorporated into BSA, as assessed excitation at 280 nm) spectra of BSA (A) and G6PD (B) (each
0.5 mg/mL) treated with 1 mM ClO2 in PBS at 25 °C for 2 min
by the detection level of chlorine atoms in the assay (0.01%). are shown (solid lines). The ClO2 reaction was terminated by
Theoretically, if one chlorine atom had been incorporated 2 mM Na2S2O3. As controls, spectra of protein treated with 1 mM
into one molecule of BSA, the chlorine content of BSA ClO2 that had been premixed with 2 mM Na2S2O3 are shown (dotted
should have become 0.053%, which is well above the lines). Each absorption spectrum was obtained after subtracting that
of a mixture of 1 mM ClO2 and 2 mM Na2S2O3 as a background.
detection level of the method used. The fluorescence intensity is shown in an arbitrary unit. Experiments
To confirm that proteins are covalently modified by ClO2, were done twice independently, and similar results were obtained.
the amino acid composition of ClO2-treated BSA and G6PD The data shown is one of them.
was determined. The numbers of tyrosine and tryptophan
residues decreased significantly in both BSA and G6PD after excited at 280 nm. When excited at 280 nm, proteins that
ClO2 treatment (Table 2), suggesting that tyrosine and contain tryptophan commonly show a characteristic fluores-
tryptophan residues are covalently modified by ClO2. The cence emission peak at ∼350 nm. As expected, BSA and
numbers of other amino acids did not decrease significantly G6PD that were not treated with ClO2 showed a sharp
in this analysis. fluorescence emission peak at 348 nm (Figure 6, inset, dotted
Absorption and Fluorescence Spectra of ClO2-Treated line). When these proteins were treated with ClO2, however,
Protein. The absorption spectra of ClO2-treated BSA and this peak disappeared completely (Figure 6, inset, solid line).
G6PD proteins showed that the 280 nm absorption (peak or This result strongly indicates that tryptophan residues in
shoulder) characteristic of the proteins had disappeared proteins are destroyed by ClO2.
(Figure 6, solid line). The spectra showed monotonically Reaction of Model Peptides with ClO2. To substantiate
decreasing absorption that extended up to ∼400 nm. The further the covalent modification of tryptophan and tyrosine
complete disappearance of the 280 nm absorption peak or residues in proteins by ClO2, I used two model peptides
shoulder in the ClO2-treated proteins (Figure 6) strongly derived from BSA, namely, ALKAWSVAR (residues 209-
suggested that the tryptophan residues were destroyed by 217) and GSFLYEYSR (residues 327-335). When these
ClO2. This interpretation was further supported by the peptides were treated with ClO2, new peptide peaks appeared
fluorescence spectra of ClO2-treated BSA and G6PD proteins on the HPLC chromatogram as compared with the nontreated
4904 Biochemistry, Vol. 46, No. 16, 2007 Ogata

FIGURE 8: HPLC analyses of glycine-substituted peptides treated

with or without ClO2. An HPLC profile of 0.2 mM peptide
FIGURE 7: HPLC analyses of peptides treated with or without ClO2.
ALKAGSVAR (100 µL) treated with (B) or without (A) 0.3 mM
An HPLC profile of 0.2 mM peptide ALKAWSVAR (100 µL)
ClO2 in PBS at 25 °C for 2 min and loaded on a C18 reverse-phase
treated with (B) or without (A) 0.3 mM ClO2 in PBS at 25 °C for
column is shown. The reaction was terminated by 0.6 mM Na2S2O3.
2 min and loaded on a C18 reverse-phase column is shown. The
The column was eluted at a flow rate of 1.0 mL/min. Likewise,
reaction was terminated by 0.6 mM Na2S2O3. The column was
0.2 mM peptide GSFLGEGSR treated with (D) or without (C) 0.3
eluted at a flow rate of 1.0 mL/min. Likewise, 0.2 mM peptide
mM ClO2 is shown. Experiments were done twice independently,
GSFLYEYSR treated with (D) or without (C) 0.3 mM ClO2 is
and similar results were obtained. The data shown is one of them.
shown. In B and D, major reaction products collected for subsequent
analyses (W62 and Y78) are shown by arrows. Intact (unreacted)
peptides are shown by arrowheads. Experiments were done five
times independently, and similar results were obtained. The data
shown is one of them.

peptides (Figure 7B,D, arrows). The major peak observed

for the ClO2-treated peptide ALKAWSVAR was named W62
(Figure 7B, arrow, retention time 40.1 min), and that for the
ClO2-treated peptide GSFLYEYSR was named Y78
(Figure 7D, arrow, retention time 41.6 min). For both
peptides, there was also a peak representing unreacted (intact)
peptide on the HPLC chromatogram (retention time 43.3 min
in Figure 7B, arrowhead, and retention time 46.9 min in
Figure 7D, arrowhead). Notably, there were also at least ten
small peaks on the HPLC chromatogram of the ClO2-treated
peptides (Figure 7B,D), indicating that the peptides under-
went some minor reactions with ClO2; however, I did not
pursue these minor reaction products further. When the FIGURE 9: Mass spectrometric analyses of ClO2-treated peptides.
tryptophan (W) and tyrosine (Y) residues in these peptides Peptide ALKAWSVAR or GSFLYEYSR (each 0.2 mM) was
were replaced by glycine (G), and the resulting peptides treated with 0.3 mM ClO2 in PBS at 25 °C for 2 min. The reaction
was terminated by 0.6 mM Na2S2O3. The reaction mixture was then
(ALKAGSVAR and GSFLGEGSR, respectively) were treated loaded on HPLC, and peak compounds that were modified by ClO2
with ClO2, no new peptide peaks appeared (Figure 8B,D), (peak W62 in Figure 7B and peak Y78 in Figure 7D) were collected.
providing direct evidence that the new peaks (Figure 7, W62 They were next analyzed by HPLC-ESI-MS/MS. The y-ion series
and Y78) observed after treatment of the peptides ALKAW- derived from W62 (A) and from Y78 (B) are shown. Modified
tryptophan and tyrosine residues are shown by W* and Y*,
SVAR and GSFLYEYSR with ClO2 were due solely to respectively. The difference of m/z between y5 and y4 ions is shown
covalent modifications of the tryptophan and tyrosine (W + 31.8), where W denotes molecular weight of tryptophan -
residues in these peptides. 18.0, namely, 204.2 - 18.0 (A). The differences of m/z between
To characterize further the kind of modifications of the y5 and y4 ions, and between y3 and y2 ions are shown (each Y +
tryptophan and tyrosine residues in these peptides, fractions 15.9), where Y denotes molecular weight of tyrosine - 18.0,
namely, 181.2 - 18.0 (B). Experiments were done twice indepen-
corresponding to the peptide peaks W62 and Y78 were dently, and similar results were obtained. The data shown is one
isolated by HPLC and were analyzed by HPLC-ESI-MS/ of them.
MS. The fifth residue in the peptide ALKAWSVAR had
become 31.8 atomic mass units heavier than tryptophan heavier than tyrosine (molecular weight 181.19) (Figure 9B),
(molecular weight 204.22) (Figure 9A), indicating that the indicating that each of the two tyrosine residues in this
tryptophan residue had been oxidized by two atoms of peptide had been oxidized by one atom of oxygen.
oxygen. Both of the fifth and the seventh residues of the Reaction Stoichiometry of ClO2. To elucidate the reaction
peptide GSFLYEYSR had become 15.9 atomic mass units stoichiometry between ClO2 and protein, peptide, or amino
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4905

Table 3: Consumption of ClO2 by a Reaction with Amino Acid, Peptide, or Proteina

amino acid, peptide, or protein (mM) ClO2 (mM)
added recoveredb consumedc added recoveredb consumedc stoichiometryd
tryptophan 0.50 0.00 0.50 3.4 2.0 1.4 2.8
tyrosine 0.50 0.00 0.50 3.4 1.8 1.6 3.2
ALKAWSVAR 0.50 0.00 0.50 3.4 1.8 1.6 3.2
GSFLYEYSR 0.50 0.00 0.50 3.4 1.3 2.1 4.2
ALKAGSVAR 0.50 0.49 0.01 3.4 2.9 0.51 nde
GSFLGEGSR 0.50 0.52 0.00 3.4 2.9 0.47 nde
BSA 0.0075 nde nde 3.4 2.5 0.88 nde
G6PD 0.0075 nde nde 3.4 2.4 1.0 nde
a
Amino acid, peptide, or protein of indicated concentration was mixed with ClO2 in PBS at 25 °C for 2 min. It was then mixed with 2 M KI
to a final concentration of 100 mM, and ClO2 concentration remaining in the mixture was quantified by an iodometric titration. b Measured after
the reaction with amino acid, peptide, or protein. ClO2 was measured by the iodometric titration. Amino acid and peptide were measured by HPLC.
Each data is a mean of five experiments, where the error of the assay was within 4%. c Calculated by subtracting the amount recovered from the
amount added. d A molar ratio of ClO2 consumed to amino acid, peptide, or protein consumed. e Not determined.

acid, I quantified the amount of ClO2 remaining in the

reaction mixture of each type of compound with ClO2. Each
molecule of tryptophan reacted with about three molecules
of ClO2 regardless of whether it was free or bound in peptide
(Table 3). Each molecule of tyrosine bound in peptide reacted
with two molecules of ClO2. By contrast, each molecule of
free tyrosine reacted with about three molecules of ClO2,
which strongly indicates that the reaction mechanism of free
tyrosine with ClO2 differs from that of tyrosine bound in
peptide. Notably, the peptides with glycine replacement,
namely, ALKAGSVAR and GSFLGEGSR (Figure 8),
reacted with only a small amount of ClO2 (Table 3). This
result is consistent with the data in Figure 8, showing that
amino acids other than tryptophan and tyrosine in these
peptides do not react with ClO2.
Amino Acid Residues Modified by ClO2. Fractions corre-
sponding to the peptide peaks W62 and Y78 (Figure 7B,D)
were isolated by HPLC, and then treated with aminopeptidase FIGURE 10: Mass spectrometric analyses of modified amino acids
M to release the modified amino acids for further charac- obtained from ClO2-treated peptides. Major reaction products of
terization. The amino acids released by aminopeptidase M peptides ALKAWSVAR and GSFLYEYSR (each 0.2 mM) treated
were separated by a second round of HPLC (data not shown). with 0.3 mM ClO2 (peak W62 in Figure 7B and peak Y78 in Figure
7D) were isolated by HPLC and lyophilized. They were then treated
The amino acid derivatives released from peptides W62 and with aminopeptidase M to enzymatically release modified amino
Y78 were designated W* and Y*, and showed retention times acids (W* and Y* in Figure 9) from the peptides. The aminopep-
of 23.8 min and 12.6 min, respectively, on HPLC. From the tidase M-treated peptides were next loaded on HPLC to isolate the
MS analyses shown in Figure 9, it was already evident that modified amino acids. The modified amino acids collected as single
W* was modified tryptophan and Y* was modified tyrosine. major peaks were next analyzed by HPLC-MS. The mass spectra
derived from modified tryptophan (A) and tyrosine (B) are shown.
Structures of the Modified Amino Acids. Compound W*, Structures of N-formylkynurenine (NFK) (A) and 3,4-dihydrox-
the tryptophan derivative released from peptide W62 by yphenylalanine (DOPA) (B) are shown. Experiments were done
aminopeptidase M and isolated by HPLC (above), was twice independently, and similar results were obtained. The data
further analyzed by HPLC-MS. From W*, an [M + H]+ shown is one of them.
ion of m/z ) 237.1 was generated, which is consistent with
N-formylkynurenine (NFK), whose theoretical value of an from the reaction of a tryptophan-containing peptide with
[M + H]+ ion is m/z ) 237.2 (Figure 10A). A peak at m/z ClO2 is NFK. In addition, I found that NFK was a major
) 209.1 was also observed, which is consistent with NFK reaction product obtained from the reaction of ClO2 with
lacking a formyl residue, i.e., kynurenine, whose theoretical free tryptophan (data not shown).
value of an [M + H]+ ion is m/z ) 209.2. The identification Next, I tried to determine the structure of the modified
of W* was further checked by its absorption spectrum (data tyrosine residue using peptide Y78 (Figure 7D). I first
not shown); the spectrum of W* showed λmax values of 240, isolated peptide Y78 by HPLC, and then treated it with
260, and 321 nm, and λmin values of 248 and 283 nm, which aminopeptidase M to release modified tyrosine (Y*). As
are exactly the same as those reported for NFK (42). The shown by MS (Figure 9B), it was already evident that the
identification of W* as NFK was next confirmed by one- only amino acid modified in peptide Y78 was tyrosine. When
dimensional 1H NMR and two-dimensional 1H-1H COSY peptide Y78 treated with aminopeptidase M was separated
NMR (Figures 11 and 12). In the 1H NMR spectrum, a signal by HPLC, a peak at a retention time of 12.6 min was obtained
at 8.334 ppm was consistent with that of aldehyde. The other (data not shown). The retention time of Y* was exactly same
signals were also consistent with the identification of W* as that of authentic 3,4-dihydroxyphenylalanine (DOPA)
as NFK. From these results, I conclude that W* obtained (12.6 min). In addition, the compound Y* showed a λmax
4906 Biochemistry, Vol. 46, No. 16, 2007 Ogata

FIGURE 11: A one-dimensional 1H NMR spectrum of modified amino acid isolated from ClO2-treated peptide. Peptide ALKAWSVAR
(0.2 mM) was treated with 0.3 mM ClO2 in PBS at 25 °C for 2 min, and the reaction was terminated by 0.6 mM Na2S2O3. The reaction
mixture was then loaded on HPLC, and modified peptide (peak W62 in Figure 7B) was isolated. The isolated peptide was next treated with
aminopeptidase M to release modified amino acid (W* in Figure 9A). The modified amino acid (tryptophan derivative) was next isolated
by HPLC. Its 600 MHz 1H NMR spectrum is shown. The structure of N-formylkynurenine is also shown. Experiments were done twice
independently, and similar results were obtained. The data shown is one of them.

value of 279 nm, which is also the same as that of DOPA. other. Interpretation of the 1H NOESY spectrum suggested
Judging from its structure, DOPA is derived from tyrosine that the 6.847 ppm signal was from the position 2 proton
by the addition of one atom of oxygen. Therefore, DOPA is and the 7.204 ppm signal was from the position 5 proton,
consistent with observed modification of the tyrosine residues because a NOESY correlation signal was found between
in Y78 (Figure 9B). These results suggest that the two these protons. In summary, only the tyrosine residue in the
tyrosine residues in Y78 became DOPA residues after peptide EYSR was modified by ClO2, and it became 2,4,5-
reaction with ClO2; in other words, Y* is DOPA. This idea trihydroxyphenylalanine (TOPA) on the reaction with ClO2.
was further supported by HPLC-MS of Y*; the spectrum To demonstrate directly that tryptophan and tyrosine
showed a peak at m/z ) 198.1 for an [M + H]+ ion residues of proteins are indeed modified by ClO2, G6PD
(Figure 10B), which is almost same as that of DOPA, whose treated with ClO2 was digested first with proteinase K and
theoretical value for an [M + H]+ ion is m/z ) 198.2. Taken then with aminopeptidase M to release the modified amino
together, I conclude that the modified tyrosine residues (Y*) acids from the proteinase K-digested protein. These modified
in Y78 are DOPA. amino acids showed peaks at retention times 11.5 and
I next used a shorter model peptide EYSR (residues 332- 23.3 min on HPLC (data not shown); these retention times
325 of BSA) as a starting material. When EYSR (HPLC are almost same as those of Y* (12.6 min) and W*
retention time 21.6 min) was treated with ClO2, a major (23.8 min), respectively, that were recovered from the ClO2-
reaction product (hereafter called Y93) appeared at a treated peptides Y78 and W62. Fractions corresponding to
retention time of 23.2 min on HPLC (data not shown). Next, the peaks with these retention times were next subjected to
I isolated Y93 by HPLC and analyzed it directly by one- HPLC-MS, which showed a peak at m/z ) 198.1 ([M +
dimensional 1H NMR and two-dimensional 1H NOESY H]+ ion) for the compound with a retention time of
(Figures 13 and 14). In the one-dimensional 1H NMR 11.5 min and a peak at m/z ) 237.1 ([M + H]+ ion) for the
spectrum, aromatic proton signals of 7.204 and 6.847 ppm compound with a retention time of 23.3 min. These m/z
were found. Because these signals were found as singlets, values are also consistent with those observed for DOPA
these protons were interpreted to be in para orientation each (m/z ) 198.2) and NFK (m/z ) 237.2). On the basis of the
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4907

FIGURE 12: A two-dimensional 1H-1H COSY NMR spectrum of modified amino acid isolated from ClO2-treated peptide. Peptide
ALKAWSVAR (0.2 mM) was treated with 0.3 mM ClO2 in PBS at 25 °C for 2 min, and the reaction was terminated by 0.6 mM Na2S2O3.
The reaction mixture was then loaded on HPLC, and modified peptide (peak W62 in Figure 7B) was isolated. The isolated peptide was next
treated with aminopeptidase M to release modified amino acid (W* in Figure 9A). The modified amino acid (tryptophan derivative) was
next isolated by HPLC. Its 1H-1H COSY spectrum is shown. Experiments were done twice independently, and similar results were obtained.
The data shown is one of them.

HPLC retention times and the m/z values, I conclude that atomic mass units were obtained (Table 4, line 10), indicating
DOPA and NFK are indeed formed in proteins after treatment that these amino acid residues were oxidized by 18O that
with ClO2. came from Cl18O2. The phosphate buffer used for the reaction
Origin of the Oxygen Atoms. To elucidate the origin of was not the origin of the oxygen atoms, because the oxidized
the oxygen atoms incorporated in the amino acids in the peptide (Figure 7, W62 and Y78) appeared without the buffer
oxidation reaction by ClO2, an 18O experiment was done. (data not shown).
When water or dissolved oxygen gas in the reaction mixture Sites of Protein Modified by ClO2. To determine the sites
was replaced by H218O or 18O2, respectively, the MS analyses in the proteins modified by ClO2, BSA and G6PD that had
of the reaction product (as in Figure 9) did not show any been treated with ClO2 were digested first with trypsin and
significant change in m/z values of product ions despite the then with subtilisin. The digested peptides were separated
high resolution of the m/z values of the mass spectrometer by HPLC. Edman degradation sequencing of the peptides
used, indicating that 18O atoms are not incorporated in the isolated from BSA identified the peptides EKLGEXG
product oxidized by ClO2 under these conditions (Table 4, (corresponding to residues 395-401), EDXLS (residues
lines 6-9). This result strongly suggests that the oxygen 449-453), and ETXV (residues 494-497). The unknown,
atoms in NFK, DOPA, and TOPA do not come from water and thus presumably modified, residue (X) in these peptides
or dissolved oxygen gas, but rather from ClO2. Direct was found to be tyrosine upon comparison of the amino acid
evidence in support of this interpretation was obtained in an sequences of these peptides with the sequence of BSA,
experiment using Cl18O2. When the peptides were treated suggesting that tyrosine residues Tyr400, Tyr451, and Tyr496
with Cl18O2 in common distilled water, tryptophan and were the sites modified by ClO2. Similarly, peptides EGX-
tyrosine derivatives having 36.4 and 18.2, respectively, LDP (residues 38-43), GNXDT (residues 93-97), AXDD
4908 Biochemistry, Vol. 46, No. 16, 2007 Ogata
HPLC peaks (i.e., reaction products) were seen at all for
ClO2-treated peptides in which the tryptophan and tyrosine
residues had been replaced by glycine (Figure 8B,D). These
results clearly indicate that minor reactions, other than the
formation of NFK, DOPA, and TOPA, occur at the tryp-
tophan and tyrosine residues but not at other amino acid
residues in the peptides treated with ClO2.
Notably, the TOPA formed in peptide EYSR after treat-
ment with ClO2 might be relatively unstable when it is
released from the peptide by aminopeptidase M or it might
resist the digestion of aminopeptidase M, because I did not
detect an amino acid derivative that appeared to be TOPA
on HPLC. It is possible that TOPA is converted to another
compound once it is released from the peptide, which would
explain why TOPA was not present in the digestion product
of protein treated with ClO2. The lack of availability of
authentic TOPA also hindered identification of the compound
on HPLC.
I found many sites of tryptophan and tyrosine residues in
the proteins treated with and modified by ClO2, which
indicates that there is low regiospecificity in protein in its
reaction with ClO2. It is worth noting, however, that whereas
both of the tyrosine residues in the peptide GSFLYEYSR
were modified to DOPA (Figures 9B and 10B), the tyrosine
residue in the peptide EYSR became TOPA (Figures 13 and
14) after treatment with ClO2. This observation clearly
FIGURE 13: A one-dimensional 1H NMR spectrum of peptide indicates that the type of reaction that takes place with ClO2
treated with ClO2. Peptide EYSR (4 mM) was treated with 5.1 mM
ClO2 in PBS at 25 °C for 2 min, and the reaction was terminated at a tyrosine side chain in peptide bonds differs depending
by 10 mM Na2S2O3. The reaction mixture was loaded on HPLC to upon the local fine structure of the peptide.
isolate modified peptide. A 1H NMR spectrum (600 MHz) of the Whereas the CD spectrum of G6PD changed markedly
modified peptide is shown. The 4.700 ppm signal is that of partial after ClO2 treatment, the CD spectrum of BSA changed only
deuterium water (HDO) used as a standard. The structure of the a little (Figure 3). This difference might be due to the number
modified peptide containing a 2,4,5-trihydroxyphenylalanine (TOPA)
residue is shown. Experiments were done twice independently, and of SS (disulfide) linkages in the two proteins: G6PD has
similar results were obtained. The data shown is one of them. no SS linkage, whereas BSA has 17. This large number of
SS linkages in BSA may contribute to stabilization of the
structure of its R-helices, which might account for the small
(residues 304-307), and FXIPE (residue 413-417) were change in the CD spectrum of BSA.
obtained from G6PD. By sequence comparison, I found that Elemental analysis showed that about 64 oxygen atoms
X in FXIPE was tryptophan, and the other unknown residue were found to be incorporated into one molecule of BSA
X was tyrosine in the G6PD sequence, suggesting that the (Table 1), and 48 molecules of ClO2 reacted with one
amino acid residues Tyr40, Tyr95, Tyr305, and Trp414 were molecule of BSA (Table 3), whose molecular weight is
modified by ClO2 in G6PD. 66 414. There are 2 tryptophan and 20 tyrosine residues in
one molecule of BSA. If all of the tryptophan and tyrosine
DISCUSSION
residues react with ClO2 and if two oxygen atoms (in
I have clearly demonstrated that ClO2-treated tryptophan tryptophan) and one or two oxygen atoms (in tyrosine) are
is oxidized by two atoms of oxygen to form NFK regardless incorporated in these amino acid residues to form NFK,
of whether it is free or in a peptide bond. However, I had DOPA, or TOPA, then theoretically the total number of
some difficulty in characterizing the exact reaction products oxygen atoms incorporated in one BSA molecule should be
of free tyrosine. Whereas tyrosine is modified to DOPA or in the range of 24-44, which is 38-69% of the observed
TOPA when it is exposed to ClO2 in a peptide bond, the value (64 oxygen atoms). This difference of oxygen atom
reaction product formed from free tyrosine seems to be incorporation in protein suggests that there might be some
different from them (although complete characterization of other oxygen-involving reactions of the protein with ClO2
ClO2-treated free tyrosine was unsuccessful.) This difference that were not detected by the analyses used in the present
in reactions might be due to the presence of the free amino study.
and/or carboxyl groups in free tyrosine. I speculate that if Mudd and his co-workers have shown that ozone reacts
amino and carboxyl groups of tyrosine are blocked, as they rapidly with tryptophan (43, 44). The absorption spectrum
are in peptide or protein, then no further reaction of the of hen egg-white lysozyme exposed to ozone indicates that
modified tyrosine residue would occur; however, the reaction tryptophan is converted to NFK (45). Meiners and his co-
may proceed further if these groups are free. The peptides workers reported (46) that ozone reacts with indole com-
treated with ClO2 showed several minor peaks on HPLC, as pounds such as tryptophan, 5-hydroxytryptophan, and 5-hy-
well as the major peaks owing to modified tryptophan and droxytryptamine. The reaction is such that approximately
tyrosine (Figure 7B,D). By contrast, neither minor nor major 1 mol of ozone reacts with 1 mol of indole compound when
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4909

FIGURE 14: A two-dimensional 1H-1H NOESY NMR spectrum of peptide treated with ClO2. Peptide EYSR (4 mM) was treated with
5.1 mM ClO2 in PBS at 25 °C for 2 min, and the reaction was terminated by 10 mM Na2S2O3. The reaction mixture was loaded on HPLC
to isolate modified peptide. A 1H-1H NOESY spectrum of the modified peptide is shown. Experiments were done twice independently,
and similar results were obtained. The data shown is one of them.

the concentration of the indole compound is 100 µM. Kuroda presence of superoxide dismutase (final 300 units/mL), the
and his co-workers also demonstrated that a tryptophan oxidation reaction as assessed by the appearance of Y78 and
residue in hen egg-white lysozyme is modified by ozone in W62 on HPLC (Figure 7B,D) occurred to the same extent
an aqueous solution (47), and showed that one of the six as it did without the enzyme (Ogata, N., unpublished data).
tryptophan residues in the enzyme is oxidized to NFK with This suggests that O2-• is not involved in the oxidation of
concomitant loss of enzymatic activity. Taken together, these protein by ClO2.
reports suggest that tryptophan residues in proteins are Modifications of amino acids by HOCl are quite different
converted to NFK by chemicals with oxidizing activity. In from those of ClO2. When protein or peptide is treated with
animals, free tryptophan is enzymatically oxidized to NFK HOCl, tyrosine residues become 3-chlorotyrosine or 3,5-
by an iron heme protein tryptophan pyrrolase (tryptophan dichlorotyrosine (49). Methionine residues become methion-
2,3-dioxygenase (EC 1.13.11.11)). ine sulfoxide (49). Cysteine residues become cysteic acid
Balasubramanian found that tryptophan and tyrosine (49). In a pentapeptide treated with HOCl, tryptophan residue
residues of lens protein crystallin are oxidized by singlet becomes 2-oxoindolone (50) or 3-chloroindolenine (51).
oxygen in a photosensitized oxidation reaction (48), in which HOCl is in equilibrium with Cl2 (16). Therefore, it is unclear
tryptophan residue becomes NFK. Here I found that the whether the above-mentioned modifications by “HOCl” are
reaction of 0.3 mM ClO2 with 0.2 mM peptide ALKAW- actually by HOCl itself or by Cl2. The difference of the
SVAR was markedly inhibited in a concentration-dependent reactions of amino acid residues in protein with ClO2 and
manner by 0.8-20 mM NaN3, a known singlet oxygen with HOCl would be primarily due to the difference that
scavenger (Ogata, N., unpublished data). The result suggests ClO2 is a free radical (ClO2•), but HOCl is not. Since HOCl
that singlet oxygen is involved also in the oxidation of is in equilibrium with Cl2 (16), it would be possible that
tryptophan by ClO2. When the reaction of ClO2 with peptide chlorination of tyrosine or tryptophan residues occurs directly
ALKAWSVAR or GSFLYEYSR was carried out in the by a substitution reaction of H and Cl atoms.
4910 Biochemistry, Vol. 46, No. 16, 2007 Ogata

18 3. Podrez, E. A., Abu-Soud, H. M., and Hazen, S. L. (2000)

Table 4: O-Labeling Experiments of Peptide in the Reaction with
Myeloperoxidase-generated oxidants and atherosclerosis, Free
ClO2a Radical Biol. Med. 28, 1717-1725.
mol wt of modified amino acid 4. Baron, C. P., Refsgaard, H. H., Skibsted, L. H., and Andersen,
M. L. (2006) Oxidation of bovine serum albumin initiated by the
tryptophan in tyrosineb in Fenton reactionseffect of EDTA, tert-butylhydroperoxide and
reaction with ALKAW*SVARc GSFLY*EY*SRc tetrahydrofuran, Free Radical Res. 40, 409-417.
H216O + Cl16O2 204.2 + 32.0 181.2 + 15.9 5. Stadtman, E. R., and Levine, R. L. (2000) Protein oxidation, Ann.
H218O + Cl16O2 204.2 + 31.8 181.2 + 16.2 N.Y. Acad. Sci. 899, 191-208.
16
O2-bubbled H216O + Cl16O2 204.2 + 32.2 181.2 + 15.9 6. Dean, R. T., Fu, S., Stocker, R., and Davies, M. J. (1997)
18
O2-bubbled H216O + Cl16O2 204.2 + 32.2 181.2 + 16.2 Biochemistry and pathology of radical-mediated protein oxidation,
H216O + Cl18O2 204.2 + 36.4 181.2 + 18.2 Biochem. J. 324,1-18.
7. Jung, T., Engels, M., Kaiser, B., Poppek, D., and Grune, T. (2006)
a
Experiments were done twice independently, and similar results Intracellular distribution of oxidized proteins and proteasome in
were obtained. The data shown is one of them. Peptides ALKAWSVAR HT22 cells during oxidative stress, Free Radical Biol. Med. 40,
and GSFLYEYSR (each 0.2 mM) were treated with 0.3 mM ClO2 at 1303-1312.
25 °C for 2 min in PBS, in which either H218O or 18O2-bubbled water 8. Raftery, M. J., Yang, Z., Valenzuela, S. M., and Geczy, C. L.
was used for the reaction. In another experiment, peptides were treated (2001) Novel intra- and inter-molecular sulfinamide bonds in
as above with Cl18O2 in common distilled water. The reaction was S100A8 produced by hypochlorite oxidation, J. Biol. Chem. 276,
terminated by 0.6 mM Na2S2O3, and then the reaction mixture was 33393-33401.
loaded on HPLC to isolate modified peptides (Figure 7, W62 and Y78). 9. Davies, K. J. (1995) Oxidative stress: the paradox of aerobic life,
The isolated modified peptides were analyzed by HPLC-ESI-MS/MS. Biochem. Soc. Symp. 61, 1-31.
Molecular weights of modified amino acids (W* and Y*) in these 10. Hawkins, C. L., and Davies, M. J. (2005) Inactivation of protease
peptides were determined from the difference of fragment ions found inhibitors and lysozyme by hypochlorous acid: role of side-chain
oxidation and protein unfolding in loss of biological function,
on the MS/MS. Molecular weights of tryptophan and tyrosine were
Chem. Res. Toxicol. 18, 1600-1610.
assumed to be 204.2 and 181.2, respectively. b The values indicate those
11. Harrison, J. E., and Schultz, J. (1976) Studies on the chlorinating
of the fifth residue tyrosine in the peptide. The seventh residue tyrosine activity of myeloperoxidase, J. Biol. Chem. 251, 1371-1374.
showed almost the same values. c Modified amino acids are shown by 12. Pattison, D. I., and Davies, M. J. (2005) Kinetic analysis of the
asterisks. role of histidine chloramines in hypochlorous acid mediated
protein oxidation, Biochemistry 44, 7378-7387.
13. Sugiyama, S., Okada, Y., Sukhova, G. K., Virmani, R., Heinecke,
As mentioned above, HOCl can inactivate microbes J. W., and Libby, P. (2001) Macrophage myeloperoxidase
ingested by leukocytes. At very low concentrations ClO2 can regulation by granulocyte macrophage colony-stimulating factor
also inactivate microbes in many in vitro experiments (30- in human atherosclerosis and implications in acute coronary
syndromes, Am. J. Pathol. 158, 879-891.
38). However, it is unclear whether in biological systems of 14. Mutze, S., Hebling, U., Stremmel, W., Wang, J., Arnhold, J.,
higher organisms, such as human, ClO2 will ever be produced Pantopoulos, K., and Mueller, S. (2003) Myeloperoxidase-derived
in sufficient amounts to cause inactivation of invading hypochlorous acid antagonizes the oxidative stress-mediated
microbes just like HOCl produced intracellularly in leuko- activation of iron regulatory protein 1, J. Biol. Chem. 278, 40542-
40549.
cytes does (11-15). In this respect, it would be important 15. Thomas, E. L., Lehrer, R. I., and Rest, R. F. (1988) Human
to note that ClO2 is generated from ClO2- by the reaction neutrophil antimicrobial activity, ReV. Infect. Dis. 10 (Suppl. 2),
with HOCl (52). ClO2 is made also from ClO2- in acidic S450-S456.
environments. It is intriguing to speculate that ClO2 might 16. Hazen, S. L., Hsu, F. F., Mueller, D. M., Crowley, J. R., and
Heinecke, J. W. (1996) Human neutrophils employ chlorine gas
be made from ClO2- in some cells of higher organisms and as an oxidant during phagocytosis, J. Clin. InVest. 98, 1283-
that it works physiologically to inactivate invading microbes. 1289.
If ClO2-, as a precursor of ClO2, is indeed found in cells, 17. Tatsuzawa, H., Maruyama, T., Hori, K., Sano, Y., and Nakano,
M. (1999) Singlet oxygen (1∆g O2) as the principal oxidant in
then it would be possible especially in acidic environments myeloperoxidase-mediated bacterial killing in neutrophil phago-
like inside of lysosome. It would also be possible that, similar some, Biochem. Biophys. Res. Commun. 262, 647-650.
to NO, ClO2 has a functional role in cellssfor example, in 18. Rosen, H., Crowley, J. R., and Heinecke, J. W. (2002) Human
signal transduction or as a local mediator of cellular neutrophils use the myeloperoxidase-hydrogen peroxide-chloride
system to chlorinate but not nitrate bacterial proteins during
functions. Most studies of protein oxidation have focused phagocytosis, J. Biol. Chem. 277, 30463-30468.
on the reactivity of single amino acid side chains and have 19. Arisawa, F., Tatsuzawa, H., Kambayashi, Y., Kuwano, H.,
ignored gross or higher-order protein structures (53). In this Fujimori, K., and Nakano, M. (2003) MCLA-dependent chemi-
luminescence suggests that singlet oxygen plays a pivotal role in
context, the present demonstrationsnamely, that protein side myeloperoxidase-catalyzed bactericidal action in neutrophil pha-
chains are covalently modified by ClO2 while the protein is gosomes, Luminescence 18, 229-238.
simultaneously denaturedswill be very important. 20. Aratani, Y., Koyama, H., Nyui, S., Suzuki, K., Kura, F., and
Maeda, N. (1999) Severe impairment in early host defense against
Candida albicans in mice deficient in myeloperoxidase, Infect.
ACKNOWLEDGMENT Immun. 67, 1828-1836.
21. Pattison, D. I., and Davies, M. J. (2001) Absolute rate constants
The author thanks Koji Koseto for his contribution to this for the reaction of hypochlorous acid with protein side chains and
work and Shigeo Asada for his valuable advice on the Cl18O2 peptide bonds, Chem. Res. Toxicol. 14, 1453-1464.
synthesis. 22. Hazell, L. J., van den Berg, J. J., and Stocker, R. (1994) Oxidation
of low-density lipoprotein by hypochlorite causes aggregation that
is mediated by modification of lysine residues rather than lipid
REFERENCES oxidation, Biochem. J. 302, 297-304.
23. Fu, X., Mueller, D. M., and Heinecke, J. W. (2002) Generation
1. Yi, D., and Perkins, P. D. (2005) Identification of ubiquitin of intramolecular and intermolecular sulfenamides, sulfinamides,
nitration and oxidation using a liquid chromatography/mass and sulfonamides by hypochlorous acid: a potential pathway for
selective detector system, J. Biomol. Tech. 16, 364-370. oxidative cross-linking of low-density lipoprotein by myeloper-
2. Hawkins, C. L., Pattison, D. I., and Davies, M. J. (2003) oxidase, Biochemistry 41, 1293-1301.
Hypochlorite-induced oxidation of amino acids, peptides and 24. Lynch, E., Sheerin, A., Claxson, A. W., Atherton, M. D., Rhodes,
proteins, Amino Acids 25, 259-274. C. J., Silwood, C. J., Naughton, D. P., and Grootveld, M. (1997)
Oxidation of Protein by Chlorine Dioxide Biochemistry, Vol. 46, No. 16, 2007 4911
Multicomponent spectroscopic investigations of salivary antioxi- 39. Middaugh, C. R., Tisel, W. A., Haire, R. N., and Rosenberg, A.
dant consumption by an oral rinse preparation containing the stable (1979) Determination of the apparent thermodynamic activities
free radical species chlorine dioxide (ClO2), Free Radical Res. of saturated protein solutions, J. Biol. Chem. 254, 367-370.
26, 209-234. 40. Bradford, M. M. (1976) A rapid and sensitive method for the
25. Shahangian, S., and Hager, L. P. (1981) The reaction of chlorop- quantitation of microgram quantities of protein utilizing the
eroxidase with chlorite and chlorine dioxide, J. Biol. Chem. 256, principle of protein-dye binding, Anal. Biochem. 72, 248-254.
6034-6040. 41. Biemann, K. (1989) Tandem mass spectrometry applied to protein
26. Kengen, S. W., Rikken, G. B., Hagen, W. R., van Ginkel, C. G., structure problems, Biochem. Soc. Trans. 17, 237-243.
and Stams, A. J. (1999) Purification and characterization of (per)-
chlorate reductase from the chlorate-respiring strain GR-1, J. 42. Makino, K., and Tsuboi, E. (1959) Isolation of formylkynurenine,
Bacteriol. 181, 6706-6711. Biochim. Biophys. Acta 31, 301-304.
27. Steinberg, L. M., Trimble, J. J., and Logan, B. E. (2005) Enzymes 43. Mudd, J. B., Leavitt, R., Ongun, A., and McManus, T. T. (1969)
responsible for chlorate reduction by Pseudomonas sp. are different Reaction of ozone with amino acids and proteins, Atmos. EnViron.
from those used for perchlorate reduction by Azospira sp., FEMS 3, 669-682.
Microbiol. Lett. 247, 153-159. 44. Mudd, J. B., Leh, F., and McManus, T. T. (1974) Reaction of
28. Wolterink, A. F., Schiltz, E., Hagedoorn, P. L., Hagen, W. R., ozone with nicotinamide and its derivatives, Arch. Biochem.
Kengen, S. W., and Stams, A. J. (2003) Characterization of the Biophys. 161, 408-419.
chlorate reductase from Pseudomonas chloritidismutans, J. Bac- 45. Previero, A., Coletti-Previero, M. A., and Jolles, P. (1967)
teriol. 185, 3210-3213. Localization of non-essential tryptophan residues for the biological
29. Hagedoorn, P. L., De Geus, D. C., and Hagen, W. R. (2002) activity of lysozyme, J. Mol. Biol. 24, 261-268.
Spectroscopic characterization and ligand-binding properties of 46. Meiners, B. A., Peters, R. E., and Mudd, J. B. (1977) Effects of
chlorite dismutase from the chlorate respiring bacterial strain GR- ozone on indole compounds and rat lung monoamine oxidase,
1, Eur. J. Biochem. 269, 4905-4911. EnViron. Res. 14, 99-112.
30. Cortezzo, D. E., Koziol-Dube, K., Setlow, B., and Setlow, P.
47. Kuroda, M., Sakiyama, F., and Narita, K. (1975) Oxidation of
(2004) Treatment with oxidizing agents damages the inner
tryptophan in lysozyme by ozone in aqueous solution, J. Biochem.
membrane of spores of Bacillus subtilis and sensitizes spores to
(Tokyo) 78, 641-651.
subsequent stress, J. Appl. Microbiol. 97, 838-852.
31. Young, S. B., and Setlow, P. (2003) Mechanisms of killing of 48. Balasubramanian, D., Du, X., and Zigler, J. S., Jr. (1990) The
Bacillus subtilis spores by hypochlorite and chlorine dioxide, J. reaction of singlet oxygen with proteins, with special reference
Appl. Microbiol. 95, 54-67. to crystallins, Photochem. Photobiol. 52, 761-768.
32. Farr, R. W., and Walton, C. (1993) Inactivation of human 49. Chapman, A. L. P., Hampton, M. B., Senthilmohan, R., Winter-
immunodeficiency virus by a medical waste disposal process using bourn, C. C., and Kettle, A. J. (2002) Chlorination of bacterial
chlorine dioxide, Infect. Control Hosp. Epidemiol. 14, 527-529. and neutrophil proteins during phagocytosis and killing of
33. Moore, N. J., and Margolin, A. B. (1994) Efficacy of nucleic acid Staphylococcus aureus, J. Biol. Chem. 277, 9757-9762.
probes for detection of poliovirus in water disinfected by chlorine, 50. Droźdź, R., Naskalski, J. W., and Sznajd, J. (1988) Oxidation of
chorine dioxide, Ozone, and UV radiation, Appl. EnViron. Mi- amino acids and peptides in reaction with myeloperoxidase,
crobiol. 60, 4189-4191. chloride and hydrogen perodixe, Biochim. Biophys. Acta 957, 47-
34. Alvarez, M. E., and O’Brien, R. T. (1982) Mechanisms of 52.
inactivation of poliovirus by chlorine dioxide and iodine, Appl. 51. Fu, X., Kao, J. L. F., Bergt, C., Kassim, S. Y., Huq, N. P.,
EnViron. Microbiol. 44, 1064-1071. d’Avignon, A., Parks, W. C., Mecham, R. P., and Heinecke, J.
35. Harakeh, M., and Butler, M. (1984) Inactivation of human W. (2004) Oxidative cross-linking of tryptophan to glycine
rotavirus, SA11 and other enteric viruses in effluent by disinfec- restrains matrix metalloproteinase activity, J. Biol. Chem. 279,
tants, J. Hyg. (London) 93, 157-163. 6209-6212.
36. Berman, D., and Hoff, J. C. (1984) Inactivation of simian rotavirus
52. Jia, Z., Margerum, D. W., and Francisco, J. S. (2000) General-
SA11 by chlorine, chlorine dioxide, and monochloramine, Appl.
acid-catalyzed reactions of hypochlorous aicd and acetyl hy-
EnViron. Microbiol. 48, 317-323.
pochlorite with chlorite ion, Inorg. Chem. 39, 2614-2620.
37. Sivaganesan, M., Rice, E. W., and Marinas, B. J. (2003) A
Bayesian method of estimating kinetic parameters for the inactiva- 53. Fu, X., Wang, Y., Kao, J., Irwin, A., d’Avignon, A., Mecham, R.
tion of Cryptosporidium parVum oocysts with chlorine dioxide P., Parks, W. C., and Heinecke, J. W. (2006) Specific sequence
and ozone, Water Res. 37, 4533-4543. motifs direct the oxygenation and chlorination of tryptophan by
38. Griffiths, P. A., Babb, J. R., and Fraise, A. P. (1999) Mycobac- myeloperoxidase, Biochemistry 45, 3961-3971.
tericidal activity of selected disinfectants using a quantitative
suspension test, J. Hosp. Infect. 41, 111-121. BI061827U