Selective inactivation of serine proteases by nonheme iron complexes

Article abstract of DOI:10.1021/ic102320j

Oxidative inactivation of the serine proteases trypsin and chymotrypsin by nonheme iron complexes is described. The nonheme ligands N4Py (1) and derivative 3CG-N4Py (2), which contains a pendant guanidinium group, were used as ligands for iron. Ferryl (Fe

Full text of DOI:10.1021/ic102320j

ARTICLE
pubs.acs.org/IC
Selective Inactivation of Serine Proteases by
Nonheme Iron Complexes
Jai Prakash and Jeremy J. Kodanko*
Department of Chemistry, Wayne State University, 5101 Cass Avenue, Detroit, Michigan 48202, United States
S Supporting Information
b
ABSTRACT: Oxidative inactivation of the serine proteases
trypsin and chymotrypsin by nonheme iron complexes is
described. The nonheme ligands N4Py (1) and derivative
3CG-N4Py (2), which contains a pendant guanidinium group,
were used as ligands for iron. Ferryl (FeIVO) species derived
from these ligands, [FeIV(O)(N4Py)]2þ (7) and [FeIV(O)-
(3CG-N4Py)]3þ (8), inactivate trypsin and chymotrypsin by
the oxidation of amino acid side chains. Ferryl 8 is most effective
with chymotrypsin (IC50 value of 26 μM for 8 vs 119 μM for 7).
IC50 values of 71 and 54 μM were obtained for trypsin with 7
and 8, respectively. Amino acid analysis confirmed that residues cysteine, tyrosine, and tryptophan are oxidized under these
conditions. Trypsin is inactivated preferentially over chymotrypsin under catalytic conditions, where the enzyme was pulsed with
H2O2 in the presence of ferrous complexes [FeII(OH2)(N4Py)]2þ(5) and [FeII(Cl)(3CG-N4Py)]2þ (6). Control experiments
support the action of a unique oxidant, other than ferryls or hydroxyl radicals, under these conditions, where tyrosine residues are
targeted selectively.
’ INTRODUCTION
palladium(II) complexes, which cleave peptides and proteins
adjacent to histidine residues.13ꢀ16 Metal complexes that inactivate
proteins by oxidation can be separated into two classes: those that
Enzyme inhibitors play a pivotal role in life sciences today, as
reagents for chemical biology and as therapeutics for the treat-
ment of human diseases. To achieve inhibition, molecules that
bind selectively to targets and diminish enzyme activity are deliv-
ered. This inhibition can be reversible or irreversible. In the case of
reversible inhibition, the inhibitor molecule binds tightly to the
target using noncovalent interactions. Although this strategy is most
common, it is reversible and can require high concentrations of
inhibitor to achieve the desired effect, which can lead to off-target
activities. Irreversible inactivation of the target enzyme, or suicide
inhibition, involves the formation of a covalent bond between the
inhibitor molecule and target enzyme. Because of issues with toxicity
and immunogenicity, suicide inhibition is less common in pharma-
ceuticals. Importantly, both of these strategies require a stoichio-
metric amount of the inhibitor. To circumvent these issues, many
have recognized that catalytic inactivation of enzymes would be an
ideal approach for halting protein function because it can lead to
irreversible inactivation of the target protein with low concentrations
of inhibitor.1ꢀ6
Metal catalysts are ideal for inactivating proteins because of
their reactivity. These catalysts can be classified as hydrolytic
complexes, which attack the amide bonds of the peptide backbone,
or oxidative complexes, which can cleave the peptide backbone or
oxidize amino acid side chains. The general strategy for either
approach is the same and involves the tethering of metal catalysts to
protein-affinity ligands for achieving target specificity. Examples of
hydrolytic complexes include copper(II) and cobalt(III) complexes,
which hydrolyze amide bonds near physiological pH,3,7ꢀ12 and
1
generate diffusible reactive oxygen species (ROS), such as O2,
superoxide, or hydroxyl radicals, and those that generate metal-
based oxidants, where the oxygen atom is bound to a metal center
and does not diffuse from the catalytic center.
Damage to proteins by ROS has been well studied because of
its significance in aging, neurodegenerative diseases, and radia-
tion therapy.17ꢀ21 Early examples of catalysts that relied on the
formation of ROS included FeIIIEDTA-based reagents for targeting
calmodulin1 and streptavidin2 as well as a copper(I) complex of
1,10-phenanthroline for targeting carbonic anhydrase.22 Later work
with chromophore-assisted light inactivation showed that proteins
1
can be inactivated selectively using O2 generated by the metal-
based chromophore [RuII(bpy)3]2þ 23,24
which showed an advan-
,
tage over more traditional organic photosensitizers,25ꢀ27 which are
susceptible to photobleaching. Appending [RuII(bpy)3]2þ to pep-
toid derivatives provided a straightforward way to increase potency
without resorting to extensive optimization of the inhibitor
structure.24 Importantly, selective targeting of intra- and extracellular
proteins was demonstrated with this approach.
When compared with ROS, inactivation of proteins by metal-
based oxidants represents an underexplored area of chemistry.
Nickel(II) and copper(II) complexes derived from the tripeptide
GlyꢀGlyꢀHis have proven applications in protein oxidation,
Received: November 19, 2010
Published: April 06, 2011
r
2011 American Chemical Society
3934
dx.doi.org/10.1021/ic102320j Inorg. Chem. 2011, 50, 3934–3945
|

Inorganic Chemistry
ARTICLE
where control experiments indicate that the action of a metal-
based oxidant is responsible for the inactivation event. However,
the identities of the metal-based oxidants have not been estab-
lished. With nickel(II) complexes, site-specific cleavage of cal-
modulin was demonstrated,28,29 whereas with copper(II) com-
plexes, inactivation of angiotensin-converting enzyme, endothe-
lin-converting enzyme 14, and human carbonic anhydrase-I29
occurred by oxidizing the amino acid side chains of the target
proteins rather than cleaving the peptide backbone.
Figure 1. Ligands N4Py (1) and derivative 3CG-N4Py (2) containing a
propylguanidinium group.
Considering that there were only a few examples of targeting
proteins with metal-based oxidants, we found interest in applying
iron complexes developed for small-molecule oxidation toward
this purpose. Since studies with iron complexes first began,
ethylenediaminetetraacetic acid (EDTA) has been the ligand of
choice, especially for footprinting applications.30ꢀ35 However,
others established that moving from carboxylate-rich ligands,
such as EDTA, to more nitrogen-rich ligands resulted in better
control and higher levels of selectivity in the oxidation of small
molecules36 because of the action of metal-based oxidants rather
than ROS, specifically HO• through the Fenton reaction.37 Work
in this area proved that iron-based oxidants, such as the well-
characterized high-valent iron(IV) oxo species [FeIV(O)-
’ EXPERIMENTAL SECTION
General Considerations. All reagents were purchased from
commercial suppliers and used as received. NMR spectra were recorded
on a Varian FT-NMR Mercury 400 MHz spectrometer. Mass spectro-
metry (MS) spectra were recorded on a Micromass Quattro LC Triple-
Quad mass spectrometer using an electrospray ionization source. IR spectra
were recorded on a Nicolet FT-IR spectrophotometer. Trypsin from bovine
pancreas and its substrate, N-R-benzoyl-DL-arginine-4-nitroanilide hydro-
chloride, were obtained from Sigma and Acros Organics, respectively, and
were used as received. Chymotrypsin and its substrate N-succinyl-Ala-Ala-
Pro-Phe-pNA were obtained from Sigma. Absorbance data were collected
on a GENios Pro, microplate reader, TECAN, using 96-well microplates.
The compounds 1,1-dipyridin-2-yl-N,N-bis(pyridin-2-ylmethyl)methana-
mine (N4Py, 1),55 methyl 6-[[(dipyridin-2-ylmethyl)(pyridin-2-ylmethyl)-
amino]methyl]nicotinate (3),61 and N-acetyltyrosine (9)64 were synthe-
sized using literature procedures.
Ligand Synthesis. Preparation of tert-Butyl 1-{6-[[(Dipyridin-2-yl-
methyl)(pyridin-2-ylmethyl)amino]methyl]pyridin-3-yl}-11,11-dimethyl-
1,9-dioxo-10-oxa-2,6,8-triazadodecan-7-ylidenecarbamate (4). A solution
of N-(3-aminopropyl)-6-[[(dipyridin-2-ylmethyl)(pyridin-2-ylmethyl)-
amino]methyl]nicotinamide (100 mg, 0.21 mmol) in a mixture of
tetrahydrofuran (THF; 0.3 mL) and H2O (20 μL) was treated with a
solution of 1,3-bis(tert-butoxycarbonyl)-2-methyl-2-thiopseudourea (61.8 mg,
0.21 mmol) in THF (0.2 mL) dropwise at room temperature. The reaction
mixture was heated to 50 °C for 1 h and then concentrated in vacuo. The
resulting crude reaction mixture was partitioned between CHCl3 (5mL) and
a saturated aqueous solution of NaHCO3 (5 mL). The organic layer was
separated, dried over anhydrous Na2SO4, filtered, and concentrated to obtain
4 as a pale-yellow solid (131 mg, 88%). 1H NMR (CDCl3): δ 11.50 (s, 1H),
9.07 (d, J = 1.6 Hz, 1H), 8.55ꢀ8.52 (m, 3H), 8.48 (d, J = 4.0 Hz, 1H),
8.29ꢀ8.26 (m, 1H), 8.16 (dd, J = 8.1 and 2.4 Hz, 1H), 7.71ꢀ7.56 (m, 7H),
7.24ꢀ7.07 (m, 3H), 5.29 (s, 1H), 3.99 (s, 2H), 3.93 (s, 2H), 3.47ꢀ3.41 (m,
4H), 1.74ꢀ1.72 (m, 2H), 1.48 (s, 9H), 1.28 (s, 9H). 13C NMR (CDCl3): δ
165.5, 163.2, 162.9, 159.8, 157.5, 153.2, 149.3, 149.1, 148.4, 136.4, 136.2,
135.5, 128.4, 123.9, 122.7, 122.2, 122.1, 121.9, 83.5, 79.7, 71.3, 57.2, 56.9,
36.9, 35.5, 30.3, 28.2, 28.0. IR (cmꢀ1): 3324, 3055, 2978, 2931, 2359, 2239,
1723, 1642, 1589, 1569, 1474, 1433, 1368, 1325, 1223, 1135, 1049, 1026,
995, 912, 855, 732, 645, 616. HRMS (ESI). Calcd for C38H47N9O5 [(M þ
Na)þ]: m/z 732.3598. Found: m/z 732.3589.
(N4Py)]2þ 38ꢀ40
have spectacular applications in the selective
,
oxidation of organic molecules38,40ꢀ47 and in mimicking non-
heme protein active sites found in nature,48ꢀ50 including en-
zymes that are used to oxidize amino acids.51,52 We were further
motivated to investigate these complexes as protein-targeting
reagents because they mimic the action of bleomycin, an example
of an iron-based catalytic drug that has been used for decades in
cancer chemotherapy, which selectively oxidizes and cleaves
DNA catalytically using bioavailable oxidants such as O2 or
H2O2.53ꢀ61 Yet, the targeting of proteins with these reagents
remained unexplored.
As a starting point to investigate protein targeting with iron-
based oxidants, our group demonstrated that the well-character-
ized ferryl [FeIV(O)(N4Py)]2þ oxidizes side chains and cleaves
the backbone of protected amino acids.62,63 However, questions
remained regarding the reactivity of iron-based oxidants toward
full proteins, where the reactivities of residues could be depen-
dent on the local protein structure and could differ from the
model systems. Would a ferryl, which is typically more selective
than HO•, be reactive enough to inactivate a protein, and if so,
could this be done selectively? Could the parent iron(II) com-
plexes oxidize and inactivate enzymes in a catalytic and selective
fashion using bioavailable oxidants, and if so, would protein
cleavage or side-chain oxidation be observed? In order to address
these questions, the reactivity of two ferryl complexes, the known
ferryl [FeIV(O)(N4Py)]2þ and a derivative containing an acyl-
propylguanidinium group for protein binding named [FeIV(O)-
(3CG-N4Py)]3þ for 3-carbon guanidinium (see Figure 1 for
ligands), were evaluated for their ability to inactivate serine
proteases trypsin and chymotrypsin under single-turnover con-
ditions. In addition, the starting ferrous complexes were com-
bined with enzyme and treated with oxidants to determine if
catalysis was possible. Results disclosed herein prove that ferryls
can act as selective reagents for inactivation of serine proteases
and that selectivity can be observed under catalytic conditions,
although evidence supports that a metal-based oxidant other than
ferryl is at play. These results confirm for the first time that the
new class of iron complexes that have extensive applications in
the oxidation of small molecules can be applied to the selective
targeting of proteins.
Preparation of 6-[[(Dipyridin-2-ylmethyl)(pyridin-2-ylmethyl)amino]-
methyl]-N-(3-guanidinopropyl)nicotinamide Hydrochloride (2). Com-
pound 4 (100 mg, 0.141 mmol) was maintained with 4 M HCl in 1,4-
dioxane (2 mL) for 3 h at room temperature. The reaction mixture was
concentrated to furnish 2 (100 mg) as a hydroscopic pale-yellow solid in
quantitative yield as its hydrochloride salt. Stock solutions of 2 in H2O
(4 mM) were prepared from this solid for the following studies based on the
quantitative yield in the conversion of 4 to 2. 1H NMR (DMSO-d6): δ 9.18
(s, 2H), 8.82(d, J = 4.1 Hz, 1H), 8.62 (d, J = 4.1 Hz, 2H), 8.41 (d, J= 7.3 Hz,
1H), 8.27 (t, J = 7.3 Hz, 1H), 8.18ꢀ8.14 (m, 2H), 8.01 (br s, 1H), 7.92 (d,
J = 8.1 Hz, 2H), 7.82 (d, J = 7.3 Hz, 2H), 7.69ꢀ7.62 (m, 4H), 6.06 (s, 1H),
4.44 (s, 2H), 4.26 (s, 2H), 3.37ꢀ3.33 (m, 2H), 3.23ꢀ3.21 (m, 2H),
1.76ꢀ1.73 (m, 2H). 13C NMR (CD3OD): δ 167.3, 161.4, 161.2, 161.1,
156.1, 156.0, 148.7, 147.5, 146.6, 143.3, 143.0, 138.5, 131.0, 127.8, 127.3,
3935
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
126.7, 126.4, 124.8, 118.7, 115.8, 58.1, 55.9, 40.0, 38.2, 29.8. IR (cmꢀ1):
3105, 2359, 1674, 1539, 1469, 1435, 1320, 1200, 835, 798, 721. HRMS
(ESI). Calcd for C28H32N9O [(M þ H)þ]: m/z 510.2730. Found: m/z
510.2733.
pulsing, protein samples were precipitated with 20% (w/v) trichloroa-
cetic acid, centrifuged (5 min, 25 °C, 6600 rpm), the supernatant was
removed, and the precipitate was treated with 2,4-dinitrophenylhydra-
zine in 2 M HCl (10 mM, 500 μL). The samples were allowed to stand at
room temperature for 1 h with vortexing every 10ꢀ15 min. Enzymes
were precipitatedwith20%trichloroaceticacid(500μL) and centrifuged (5
min, 25 °C, 6600 rpm), the supernatant was discarded, and the pellet was
washed three times with 1 mL of ethanol/ethyl acetate (1:1) to remove
any free reagent. The sample was allowed to stand 10 min before
centrifugation (5 min, 25 °C, 6600 rpm), and the supernatant was
discarded each time. The precipitated enzyme was dissolved in 0.6 mL of
a 6 M guanidine solution (pH = 2.3, 20 mM potassium phosphate), and
the carbonyl contents were deteremined by UVꢀvis spectroscopy using
Metal Complexation and Ferryl Generation. The ironꢀligand
complexes [FeII(OH2)(N4Py)]2þ (5) and [FeII(Cl)(3CG-N4Py)]2þ (6)
were prepared by adding 1 equiv of FeII(ClO4)2 xH2O (100 mM stock
3
solution in H2O) to H2O solutions of 1 and 2, respectively. Ferryls
[FeIV(O)(N4Py)]2þ (7) and [FeIV(O)(3CG-N4Py)]3þ (8) were gener-
ated by adding 8 μL of a 100 mM solution of peracetic acid in H2O
(2 equiv) to solutions of 5 and 6 (200 μL, 2 mM).
Magnetic Susceptibility Measurements and NMR Studies.
The solution-phase magnetic moments of iron complexes 5 and 6 were
the known molar extinction coefficient, ε360 = 22 000 Mꢀ1 cmꢀ1 66
.
1
determined by H NMR in a D2O solvent following Evan’s method65
Synthesis of 10. A solution of Ac-Tyr-OH (9; 56 mg, 0.25 mmol)
and [FeII(N4Py)(CH3CN)](ClO4)2 (17 mg, 25 μmol) in a 10 mM
acetate buffer (25 mL, pH = 6.0, 150 mM NaCl) was pulsed with H2O2
(50 μL ꢁ 5, 10 mM total, 2 mM per pulse) every 10 min over the course
of 1 h. The solution was frozen and lyophilized. The lyophilized powder
was dissolved in water and filtered through Dowex 50WX4-100 ion-
exchange resin to remove ligand and iron. The filtrate was frozen and
lyophilized. The product 10 was purified by HPLC (HPLC column
Zorbax XDB-C18, 21.2 ꢁ 150 mm, 5 μm equipped with a guard column
Zorbax XDB-C18, 21.2 mm, 5 μm, 0.1% TFA/MeOH 87:13, flow rate =
using tert-butyl alcohol as the reference. A small capillary tube (closed at
one end) was filled with a solution of D2O and tert-butyl alcohol [50ꢀ75
μL, 7:1 (v/v)] and sealed. Solutions of 5 (4 mM, 0.016 mmol, 4 mL) and
6 (5 mM, 0.015 mmol, 3 mL) were made by mixing ligands 1 and 2 with
1 equiv of Fe(ClO4)2, respectively, in D2O containing tert-butyl alcohol.
The sealed capillary tubes were placed inside the NMR tubes, iron
1
complex solutions (800 μL) were added, and H NMR spectra were
recorded. For solution NMR studies, 10 mM solutions of ligands 1 and 2
were prepared in a D2O solvent. Their iron complexes 5 and 6 were
generated in situ by mixing 800 μL of ligand solutions 1 and 2, with 10
μL of a 0.8 M Fe(ClO4)2 solution (203.8 mg, 0.8 mmol, 1 mL, 1 equiv)
in D2O, respectively. The samples were analyzed by 1H NMR spectros-
copy after the addition of ligand, iron, and CH3CN (21 μL, 50 equiv;
Figures S16ꢀS19 in the Supporting Information).
1
20 mL minꢀ1; TR = 6.24 min); H NMR and MS data of isolated 10
matched the literature data for the known compound.67
’ RESULTS
Enzyme Inactivation Assay. The enzymatic assays were initiated
by adding the substrate (N-R-benzoyl-DL-arginine-4-nitroanilide hydro-
chloride for trypsin and N-succinyl-Ala-Ala-Pro-Phe-pNA for
chymotrypsin) to the solutions of enzyme containing varied concentra-
tions of inhibitor (0ꢀ1000 μM). The initial velocities, obtained from the
A405 versus time plot, were converted to the percentage of enzyme
activity (% enzyme activity) with respect to the control reaction
containing no inhibitor. For each assay, % enzyme activity was computed
using the average of three runs, with the error equal to the standard
deviation, and plotted as a function of the inhibitor concentration. Data
were fit using a sigmoidal equation to obtain the IC50 value.
Pulse Experiments with H2O2. The enzyme (1 μM) was incubated
with FeII(ClO4)2, or 5or 6(20 μM) (total volume 600 μL), and H2O2 (5ꢁ
40 μM) was added every 10 min. Enzyme activities were determined on
aliquots removed from the solution as described above, 10 min after each
addition of H2O2. Activities were adjusted for dilution prior to their con-
version to % enzyme activity, with 100% activity equal to the activity of the
blank reaction in the absence of inhibitor at t = 0. In control experiments,
ROS scavengers D-mannitol, imidazole, and NaN3 (see Figure 8) were added
as aqueous solutions before the addition of H2O2.
Liquid Chromatography/Mass Spectrometry (LCMS)
Analysis of Protein Samples. LCMS was performed on a Micro-
mass QuattroLC triple quadrupole mass spectrometer with an electro-
spray/APCI source and Waters Alliance 2695 liquid chromatograph.
Approximately 100 pmol of the sample was injected onto a Jupiter 5 μm
C18 300 Å 50 ꢁ 2.0 mm column (Phenomenex) using an autosampler.
Separation was achieved using a linear gradient of 20ꢀ30% (trypsin)
and 20ꢀ40% (chymotrypsin) CH3CN in water and a 0.1% HCO2H
solution in water. A total of 80 μL of H2O (blank) was injected after each
sample to minimize any carryover from the previous sample. High-
performance liquid chromatography (HPLC) was carried out at a flow
rate of 0.6 mL minꢀ1 with the column heated to 45 °C.
Synthesis of Ligand 2. The N4Py derivative 2, which contains
a pendant acylpropylguanidinium group, was synthesized in
three steps starting from the known ester 3 (Scheme 1).61
Following a literature procedure, refluxing the methanolic solu-
tion of 3 and 1,3-diaminopropane in the presence of a catalytic
amount of NaCN under a N2 atmosphere gave the aminated
product in 77% yield.61 The resultant primary amine was treated
with 1,3-bis(tert-butoxycarbonyl)-2-methyl-2-thiopseudourea in
THF/H2O at 50 °C for 1 h, furnishing 4 in 88% yield.
Compound 4 was deprotected by treatment with 4 M HCl in
1,4-dioxane for 3 h at room temperature, giving the ligand 3CG-
N4Py (2) as a hydrochloride salt in quantitative yield.
Generation and Characterization of Ferryl Species. The
ferrous complexes 5 and 6 were generated in situ by adding 1
equiv of FeII(ClO4)2 to aqueous solutions of ligands 1 and 2,
respectively. The orange complexes 5 and 6 were characterized
1
by UVꢀvis and H NMR spectroscopies and MS. UVꢀvis spec-
troscopy indicated similar data for 5 and 6 (ε380 = 1700 Mꢀ1 cmꢀ1
with a shoulder at 460 nm and ε375 = 1700 Mꢀ1 cmꢀ1 with a
shoulder at 470 nm for 5 and 6, respectively).68 1H NMR spectra for
5and 6in D2O were consistent with the complexes bearing high-spin
S = 2 FeII centers. Broad resonances ranging from 150 to 20 ppm,
well outside of the diamagnetic region, were observed for 5. Complex
6 showed resonances within the same region, except that downfield
resonances were split in two, which is consistent with the lower
symmetry of 6 (C1 symmetry) with respect to 5 (Cs symmetry)
because of the acylpropylguanidinium group on the ligand 3CG-
N4Py. Magnetic moments for 5 and 6 in D2O were 4.4 and 4.5 μB,
respectively, as determined by Evan’s method, which are close to the
expected spin-only values for a high-spin iron(II) complex.69 Inter-
estingly, treatment of 5 or 6 in D2O with excess CH3CN (50 equiv)
caused a spin transition from high to low spin, as judged by 1H NMR
spectroscopy, in which resonances between 200 and 20 ppm were no
longer observed. In addition, a sharp drop in the magnetic moment,
Carbonyl Assay. The carbonyl contents of the oxidized enzyme
samples were determined using the following literature method.66
Enzymes (20 μM) were treated with 5 or 6 (50 μM) and pulsed with
varying amounts of H2O2 (100ꢀ500 μM, 100 μM per pulse). After
3936
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
Scheme 1. Synthesis of Ligand 2a
a (a) 1,3-Diaminopropane, MeOH, NaCN, reflux, 24 h; (b) 1,3-bis(tert-butoxycarbonyl)-2-methyl-2-thiopseudourea, THF/H2O, 50 °C, 1 h; (c) 4 M
HCl in 1,4-dioxane, 3 h.
Figure 2. UVꢀvis spectra of ferrous and ferryl complexes in H2O: (a) 5 (red) and 6 (blue); (b) 7 (red) and 8 (blue).
<0.2 μB for both samples, was observed by Evan’s method. MS data
for 5 in H2O matched the formula [FeII(OH2)(N4Py)]2þ as des-
cribed previously,70 and data for 6 in H2O showed a prominent
molecular ion at m/z 300.5860, along with an isotope pattern that
agreed with the molecular formula [FeII(Cl)(3CG-N4Py)]2þ
(Figure S3 in the Supporting Information), where chloride presum-
ably binds to the iron center in place of H2O because it is present as
the counterion to the guanidinium salt. Treatment of 5with peracetic
acid at 25 °C gave a pale-green species, [FeIV(O)(N4Py)]2þ (7), in
∼20 min with a maximum absorption wavelength at λmax = 680 nm
IC50 values were calculated (Table 1). In these studies, trypsin (1 μM)
was treated with each reagent (0ꢀ1000 μM) in a acetate buffer
(pH = 6.0, 150 mM NaCl) at room temperature. After incubation for
1 h, the enzyme activities were determined by addition of the
chromogenic substrate BAPNA. These experiments indicated that
ligands 1and 2inhibit trypsin in the mid-micromolar range (entries 1
and 2). Ligand 2 containing the propylguanidinium group was the
more potent inhibitor. Despite the fact that FeII(ClO4)2 had no effect
on the trypsin activity, the ferrous complexes 5 and 6 showed more
potent inhibition than ligands alone (entries 3 and 4). Although
peracetic acid inhibited trypsin, ferryls 7 and 8 were more potent oxi-
(ε = 170 Mꢀ1 cmꢀ1), which agrees with the published data.71
A
similar procedure was used to prepare the ferryl [FeIV(O)(3CG-
N4Py)]3þ (8; λmax = 675 nm, ε = 170 Mꢀ1 cmꢀ1; Figure 2) from 6,
where it took ∼10 min for complete generation. Ferryl 8 was char-
acterized by high-resolution electrospray ionization MS (ESMS),
which displayed a prominent molecular ion at m/z 194.0682, along
with a suitable isotopic pattern, consistent with a trication derived
from the molecular formula [FeIV(O)(3CG-N4Py)]3þ (Figure S1
in the Supporting Information).
Enzyme Inactivation with Ferryl Compounds. In order to
evaluate the ability of the ferrous and ferryl complexes 5ꢀ8 to inac-
tivate trypsin, the enzyme activities were determined in the presence
of varied concentrations of these reagents along with controls, and
dants (entries 5ꢀ7). Between the ferryls (Figure 3a), 8 (IC50
=
54 μM) was found to be a better oxidant than 7 (IC50 = 71 μM).
When inactivation studies were performed on chymotrypsin,
different results were obtained (Table 1). Neither the ligands 1
and 2 nor the ferrous complexes 5 and 6 (entries 1ꢀ4) inhibited
chymotrypsin in the range from 0 to 1000 μM, with the exception
of 6, which inhibited chymotrypsin only at very high concentra-
tions (IC50 = 790 μM). This observation is consistent with the
loss of binding at the active site because the S1 pocket of
chymotrypsin prefers hydrophobic groups as opposed to the
charged propylguanidinium group of 2. Despite the fact that
ligands 1 and 2 and their iron complexes show only weak
3937
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
inhibition of chymotrypsin, both ferryls 7 and 8 (entries 6 and 7)
inactivate chymotrypsin in a concentration-dependent fashion
(Figure 3b). Between the two ferryls, again 8 was found to be a
better oxidant, and a higher level of selectivity was observed
between the ferryls with chymotrypsin versus trypsin. Binding of
ferryl 8 near the active site is not likely because the S1 pocket prefers
hydrophobic groups. Instead, the lower IC50 value of 8 compared to
7 may be attributed to favorable electrostatic interactions between
the charged ferryl 8and carboxylates on the surface of chymotrypsin,
which are present in higher abundance than with trypsin (Figure S4
in the Supporting Information).72
Characterization of Products. To gain insight into the mode
of inactivation, enzyme products were analyzed by SDS-PAGE
(Figure 4), which confirmed that trypsin was oxidized but not
cleaved extensively after treatment with ferryls 7 and 8. Incuba-
tion of the protein alone for 1 h resulted in a major loss of
intensity for the parent band at 23 kD and the formation of new
bands with lower molecular weight, consistent with cleavage of
the protein by autoproteolysis (Figure 4a). In contrast, enzyme
samples treated with peracetic acid or ferryls 7 and 8 showed little
or no cleavage of the protein, consistent with an inactivation
event being faster than autoproteolysis of the enzyme. In the case
of trypsin treated with peracetic acid (lane 3), a sharp, well-
defined band was observed at 23 kD, whereas samples treated
with the ferryl complexes showed a broad, smeared band (lanes 4
and 5) near the trypsin region (∼23 kD), indicating a distribu-
tion of molecular weights near the parent mass of the enzyme.73
Similar results were obtained with chymotrypsin. In the control
sample with enzyme alone, fragments with lower molecular weights
were observed, which is consistent with autoproteolysis (Figure 4b,
lane 2). Autoproteolysis was inhibited by peracetic acid, but the
enzyme appeared as a sharp band (not shown). In contrast, broad
smeared bands near the 25 kD region were observed for chymo-
trypsin treated with ferryls 7 and 8 (lanes 3 and 4). Fragments of
lower molecular weight were not observed in lanes 3 and 4, indicating
that inactivation of the enzyme was rapid and complete before
autoproteolysis began. Because chymotrypsin is composed of three
subunits, R (1254 kD), β (13924 kD), and γ (10 067 kD), which
are linked by disulfides, the same three samples were treated with
dithiothreitol (DTT) to determine if internal cross-links had formed
between the subunits. Lane 5 with enzyme plus DTT showed three
major fragments, plus new bands that were expected because of
autoproteolysis. Only three bands were observed in samples treated
with ferryls 7 and 8 followed by DTT (lanes 6 and 7), which are
consistent with the expected molecular weights for the subunits (β
and γ) plus an additional band resulting from incomplete cleavage of
the disulfide bond between the R and β subunits. Although the band
at 14 kD could represent a covalent cross-link that was not cleaved by
DTT, the sample of enzyme only treated under the same conditions
shows the same pattern, suggesting that incomplete reduction of the
disulfide bond between the R and β subunits occurred. However,
bands at 14 and 11 kD in samples treated with ferryls 7 and 8 appear
broad with respect to the same bands observed in lane 5, as would be
expected if the individual subunits were oxidized at the side chains
and present as a distribution of masses near the parent region.
LCMS chromatograms of trypsin (20 μM, not incubated), and
trypsin incubated with the ferryls 7 and 8 (1000 μM) for 1 h at
room temperature, show different retention times and product
distributions for enzyme (control) versus enzymes treated with
ferryls 7 and 8 (Figure 5, left panels). Integration of the peaks and
deconvolution of the MS spectra gave a single peak for the
Table 1. IC50 Values (μM) for the Inhibition of Trypsin and
Chymotrypsin by Ligands 1 and 2 and Their Respective
Ferrous (5 and 6) and Ferryl (7 and 8) Complexes
entry
compound
trypsina
chymotrypsina
1
2
3
4
5
6
7
8
N4Py (1)
493
354
301
217
236
71
>1000
1024
>1000
787
3CG-N4Py (2)
[FeII(OH2)(N4Py)]2þ (5)
[FeII(Cl)(3CG-N4Py)]2þ (6)
peracetic acid
[FeIV(O)(N4Py)]2þ (7)
[FeIV(O)(3CG-N4Py)]3þ (8)
FeII(ClO4)2
128
119
54
26
>1000
>1000
a The % enzyme activities at different concentrations were determined as
the average from three independent experiments, with 100% activity
equal to the activity of the blank reaction in the absence of inhibitor and
the error equal to the standard deviation of the data set. Activities were
plotted against log [inhibitor] and fit to a sigmoidal curve to calculate the
IC50 values. The enzyme (trypsin or chymotrypsin) concentration was
1.0 μM; the trypsin substrate N-R-benzoyl-DL-arginine-4-nitroanilide
hydrochloride and chymotrypsin substrate N-succinyl-Ala-Ala-Pro-Phe-
pNA concentrations were 1.0 mM. The reactions were conducted at
pH = 6.0 in a 10 mM acetate buffer containing 150 mM NaCl.
Figure 3. IC50 plots with ferryl species 7 (red 2) and 8 (blue [) for trypsin (a) and chymotrypsin (b). See Table 1 footnote a for conditions and other details.
3938
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
Figure 4. (a) SDS-PAGE analysis (16%), stained with Coomassie blue, of trypsin (20 μM) incubated at room temperature for 1 h at pH = 6.0, in a
10 mM acetate buffer containing 150 mM NaCl. A smeared band near 23 kD in lanes 3ꢀ5 showed oxidized enzyme. Lane 1: MW marker. Lane 2: enzyme only.
Lane 3: peracetic acid (1000 μM). Lane 4: 7 (1000 μM). Lane 5: 8 (1000 μM). (b) SDS-PAGE analysis (16%), stained with Coomassie blue, of chymotrypsin
(20 μM) incubated at room temperature for 1 h at pH = 6.0, in a 10 mM acetate buffer containing 150 mM NaCl. A smeared band near 25 kD in lanes 3 and 4
showed oxidized enzyme. Lane 1: MW marker. Lane 2: enzyme only. Lane 3: 7(500 μM). Lane 4: 8(500 μM). Lane 5: enzymewith DTT (100 mM). Lane 6:7
(500 μM) with DTT (100 mM). Lane 7: 8 (500 μM) with DTT (100 mM). Equal amounts of protein were loaded in each lane.
Figure 5. Left panels: LC chromatograms (top) and deconvoluted MS spectra (bottom) of trypsin (spectrum a) and after treatment with 8 (spectrum
b) or 7 (spectrum c). Reactions were conducted at room temperature by incubating trypsin (20 μM) with 7 or 8 (1000 μM) for 1 h before desalting and
subjecting the samples to ESMS analysis. Right panels: LCMS chromatograms (top) and deconvoluted MS spectra (bottom) of chymotrypsin
(spectrum a) and after treatment with 8 (spectrum b) or 7 (spectrum c). Reactions were conducted at room temperature by incubating chymotrypsin
(20 μM) with 7 or 8 (1000 μM) for 1 h before desalting and subjecting the samples to ESMS analysis.
control sample with enzyme only, which corresponds to the MW
of bovine trypsin (MW = 23 293) and complex spectra for both
samples treated with ferryls. For the treated samples, a distribution
of molecular weights larger than that of trypsin was observed,
consistent with oxidation of the enzyme by the addition of mult-
iple oxygen atoms to the protein, as evidenced by the difference
between masses corresponding to 16 units. In the case of the sample
treated with ferryl 8 (left panel, spectrum b), a second group of
3939
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
Figure 6. Trypsin (graph a) and chymotrypsin (graph b) activities as a function of added H2O2. Samples containing enzyme only (brown b,1 μM) and
enzyme treated with FeII(ClO4)2 (green 9), 5 (red 2), or 6 (blue [) (20 μM) were pulsed with H2O2 every 10 min (40 μM per pulse). The reaction
was performed in an acetate buffer of pH = 6.0 containing 150 mM NaCl. The activities were converted to % enzyme activity, with 100% activity equal to
the activity of the blank reaction in the absence of inhibitor at t = 0. Data points are averages from three independent experiments, where errors are
reported as standard deviations.
peaks was observed below the parent mass of trypsin, which is
consistent with oxidation of trypsin and cleavage of a small
fragment, approximately 2 kD in weight, most likely from the N
terminus (vide infra). Similar results were observed for chymo-
trypsin. LCMS chromatograms of chymotrypsin (20 μM) incu-
bated with the ferryls 7 and 8 (1000 μM) for 1 h at room
temperature are shown in Figure 5 (right panels). The sample
with enzyme only showed a sharp peak that was integrated and
deconvoluted to a mass of 25 449, whereas broad peaks were
observed for samples treated with ferryls 7 and 8. Again integration
of these peaks gave complex spectra with peaks higher in masses
than the chymotrypsin itself consistent with oxidation of protein.
Enzyme Inactivation with Ferrous Complexes and H2O2.
After it was established that trypsin and chymotrypsin could be
inactivated selectively with ferryls 7 and 8, experiments were
performed in which the enzymes were treated with ferrous com-
plexes 5 and 6 prior to the addition of oxidant. These conditions
were examined to determine if catalytic inactivation of the proteases
was possible. When enzymes were incubated in the presence of 5 or
6 under an aerobic atmosphere, the enzyme activity did not change
within reasonable time periods (<6 h). Similar results were obtained
in the presence of the reductant ascorbate or DTT, which did not
accelerate enzyme inactivation. Thus, these results confirm that
enzyme inactivation of trypsin and chymotrypsin using O2 as the
oxidant, either in the presence or absence of reductant, is a relatively
slow process with these iron complexes.74
More promising results were obtained when H2O2 was used as
the oxidant. For these experiments, solutions of trypsin (1 μM)
and several iron species [FeII(ClO4)2, 5 or 6, 20 μM] were pulsed
with H2O2 (40 μM per pulse). Enzyme activities were deter-
mined by removing aliquots after each pulse (10 min between
pulses) and plotted against the total amount of peroxide added
(Figure 6a). Negligible changes in the activities were observed
with trypsin alone and trypsin treated with FeII(ClO4)2. In the
case of the sample treated with FeII(ClO4)2, the trypsin activity
dropped immediately by ∼15% after the first addition of H2O2
but did not drop further after the addition of more H2O2, which is
consistent with a Fenton reaction carried out by an iron species
that was no longer viable after the first addition of the oxidizing
reagent. In contrast, enzyme activities were lost in a dose-
dependent fashion with respect to H2O2 when pulsing was
performed in the presence of 5 or 6. In both cases, the activities
were diminshed by roughly 80% after the addition of 200 μM
H2O2, 10 equiv with respect to the iron complexes. These data
confirm that 6, which contains the propylguanidinium group,
inactivates trypsin more effectively than 5. For example, after three
additions of H2O2 (120 μM total), the trypsin activity dropped by
80% with 6 compared to 60% with 5. The more efficient inactivation
of trypsin with 6 compared to 5 confirms that selectivity can be
gained between two iron complexes and is consistent with the
protein-affinity group (propylguanidinium) of 6 directing the com-
plex toward the protein, presumably the S1 pocket where Asp 189
may form a salt bridge with the guanidinium group.
When chymotrypsin was treated with H2O2 under the
same conditions, controls with enzyme alone or enzyme plus
FeII(ClO4)2 showed no significant changes after the addition of
200 μM H2O2 (Figure 6b). Again, the enzyme activity was lost in
a dose-dependent fashion when chymotrypsin was pulsed with
H2O2 in the presence of 5 and 6. However, the enzyme activities
dropped only by ∼35% in both cases after the addition of
200 μM H2O2, as opposed to 80% with trypsin. Selectivity was
not observed between 5 and 6. These data confirm that selective
inactivation of trypsin over chymotrypsin can be observed with
the iron complexes 5 and 6 in the presence of H2O2. Differences
in reactivity for 5 and 6 with trypsin versus chymotrypsin parallel
data from inhibition studies (Table 1, entries 3 and 4), where the
complexes 5 and 6 show more a potent inhibition of trypsin than
chymotrypsin. The trend observed is consistent with the active
reagents derived from 5 or 6 and H2O2 possessing a higher
effective molarity for trypsin over chymotrypsin and causing
more efficient oxidative damage.
Analysis by LCMS and SDS-PAGE was used in order to
characterize the enzyme products from trypsin and chymotryp-
sin. SDS-PAGE analysis of the enzymes (1 μM), pulsed with
H2O2 (40 μM per pulse) every 10 min, in the presence of 6
showed an increase in the smearing of the band near the parent
trypsin band (23 kD, Figure 7a) and the chymotrypsin band
region (25 kD, Figure 7b) with added H2O2, although smearing
was less extensive than in the samples treated with ferryls 7 and 8.
This observation is consistent with a distribution of molecular
weights after oxidation with 6 and H2O2. Because other bands
were not observed in either case, the oxidation with 6 and H2O2
3940
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
Figure 7. SDS-PAGE analysis (16%), stained with silver, of (a) trypsin and (b) chymotrypsin (1 μM), pulsed with H2O2 (40 μM) every 10 min in the
presence of 6 (20 μM) at room temperature at pH = 6.0, in a 10 mM acetate buffer containing 150 mM NaCl. Smeared bands near 23 kD in part a and 25
kD in part b in lanes 3ꢀ7 showed oxidized enzymes. Lane 1: MW marker. Lane 2: enzyme only. Lane 3: H2O2 (40 μM). Lane 4: H2O2 (80 μM). Lane 5:
H2O2 (120 μM). Lane 6: H2O2 (160 μM). Lane 7: H2O2 (200 μM).
does not lead to cleavage of the enzymes. The chro-
matograms of trypsin (1 μM) and chymotrypsin pulsed with
H2O2 (40 μM per pulse) in the presence of 5 or 6 (20 μM) at
room temperature show different retention times and product
distributions for enzyme (control) versus treated samples (Figure
S5 in the Supporting Information). Integration of the peaks and
deconvolution of the MS spectra gave single peaks for the control
samples with enzyme only, while complex spectra were obtained for
samples where enzymes were pulsed with H2O2 in the presence of 5
or 6, confirming that oxidation of the proteins occurred.
Mechanistic Studies. Additional experiments were carried
out to gain further insight into the nature of protein inactivation
under single-turnover (ferryls 7 and 8) and catalytic conditions
(5 or 6 plus H2O2). In particular, the identity of the active oxidant
under catalytic conditions was in question. Both HO• (Fenton
Figure 8. Inactivation of trypsin in the presence of ROS scavangers
chemistry) and iron-based oxidants have been implicated in
oxidation reactions when ferrous complexes such as 5 are treated
with H2O2, and the outcome varies according to the ligand set
and the conditions of the experiment (vide infra).75ꢀ77 To
distinguish between Fenton chemistry (HO•) and iron-based
oxidants, control experiments were performed in the presence of
reagents that scavenge ROS. When the inactivation of trypsin
with 6 was performed in the presence of D-mannitol, a potent
hydroxyl radical scavenger, inactivation was not inhibitited with
0.1 mM D-mannitol and was only slightly inhibited with 20 mM
D-mannitol, confirming that HO• can act as only a minor com-
ponent of the inactivation pathway if at all (Figure 8 for trypsin;
see Figure S6 in the Supporting Information for chymotrypsin).
Thus, another oxidant was likely at play. Similar results were
obtained in the presence of imidazole, a scavenger of the hydroxyl
(azide, D-mannitol, and imidazole). The reactions were conducted at
pH = 6.0, in a 10 mM acetate buffer containing 150 mM NaCl. Trypsin
(1 μM) with 6 (20 μM) was pulsed with H2O2 (40 μM) in the presence
of a buffer as a blank (blue 9) and ROS scavengers D-mannitol (0.1 mM,
green 2; 20 mM, black O), imidazole (0.1 mM, purple 0), and NaN3
(0.1 mM, maroon [; 1.0 mM, red b). Enzyme activities, where 100%
activity is equal to the activity of the blank reaction in the absence of
inhibitor at t = 0, were determined by removing aliquots before each
pulse (10 min between pulses) and plotted against the total amount of
peroxide added. Data points are averages from three independent
experiments; errors were (5%.
inactivation of trypsin and chymotrypsin in the presence of 5 or 6
and H2O2.
The presence of ketones and aldehydes in proteins, known as
the carbonyl content, is a key indicator of oxidative damage
observed in vitro and in vivo during aging and under conditions
of oxidative stress.66 Because normal proteins do not contain
ketone or aldehyde carbonyls that react electrophilically with
hydrazine nucleophiles, the reagent 2,4-dinitrophenylhydrazine
can be used to determine the overall protein carbonyl content
resulting from protein oxidation. Following a literature
method,66 the amount of carbonyl groups formed with trypsin
and chymotrypsin (20 μM) was determined as a function of the
pulsed H2O2 concentration in the presence of 6 (50 μM;
Figure 9). Similar results were obtained with 5 (not shown). In
the case of both enzymes, an increase in the carbonyl content was
observed that plateaued at 500 μM H2O2 to approximately
20 nmol/mg of protein. Because nearly identical results were
1
radical and O2. Interestingly, inactivation was inhibited some-
what in the presence of NaN3 (0.1ꢀ1 mM), again in a concen-
tration-dependent fashion. Although azide is a scavenger of 1O2,
it is known to inhibit tyrosine nitration catalyzed by the heme
enzyme myeloperoxidase.78 In this case, rather than quenching
1O2, competitive binding of azide to the peroxidase iron center
may poison the heme by blocking access to H2O2.79 In support of
this mode of inhibition, the binding of azide to 6 was confirmed
by UVꢀvis titration of 6 with NaN3 (Figure S7 in the Supporting
Information), which suggests that the action of peroxide with the
iron catalyst can be inhibited by ligands that bind competively to
the sixth site of the iron center. Together, these results indicate
1
that an oxidant other than HO• or O2 is responsible for the
3941
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
obtained with trypsin and chymotrypsin, these data indicate that
the formation of protein carbonyls is not coupled to the
inactivation event imparted by 6 and H2O2 (Figure 6a,b). In
fact, using the molecular weight of the proteins and the observed
content of protein carbonyls, one can estimate that only four or
five carbonyls are formed per protein out of 223 (trypsin) and
241 (chymotrypsin) residues when the proteins (20 μM) are
treated with 6 (50 μM) and 500 μM H2O2. Assuming that a
single carbonyl is formed per oxidized residue, this modification
accounts for less than 3% of all residues.
The aforementioned results confirm that a unique oxidant
must give preferential inactivation of trypsin over chymotrypsin
under catalytic conditions. Attempts to analyze the oxidized
proteins by LCMS/MS analysis were inconclusive due to low
sequence coverage because the oxidized proteins did not respond
well to digestion. Although LCMS/MS analysis can reveal
information about the location of oxidized residues in proteins,
it is difficult to obtain quanitative measurements of protein
oxidation using this method. Therefore, amino acid analysis
was performed on the oxidized samples, in order to determine
which types of residues were affected by treatment of trypsin and
chymotrypsin under the single-turnover (ferryls 7 and 8) and
catalytic conditions (ferrous complexes 5 and 6 plus H2O2)
(Figure 10). Strikingly, catalytic conditions produced a major
modification in the levels of the amino acid tyrosine with respect
to the control, while leaving other amino acid levels unchanged
within the margin of error. Different results were obtained with
the ferryls, where the levels of tyrosine, cysteine, and tryptophan
were diminished significantly with respect to controls. These data
are consistent with published model studies that indicated that
cysteine, tyrosine, and tryptophan were the most reactive amino
acids with 7.63 Results obtained in the single-turnover and
catalytic conditions were not highly dependent on the ligand
structure, suggesting that only a modest level of directablility was
achieved by attaching the propylguanidinium group to the N4Py
ligand. These results confirm that catalytic conditions result in a
milder form of oxidation than oxidation by the ferryls and
support the hypothesis that an oxidant other than ferryl is acting
in the presence of H2O2.
In order to gain insight into the reaction of the iron catalysts
with tyrosine in the presence of H2O2, the oxidation of 9 was
examined. Under the same conditions as those used in the
enzyme experiments, the catechol product Ac-DOPA-OH (10)
was observed as a major product by HPLC analysis (Scheme 2
and Figure S8 in the Supporting Information). Dityrosine, a
common product of tyrosine oxidation resulting from oxidative
coupling of the two phenol rings, was detected by fluorescence
analysis of the crude reaction mixture but not by ESMS,
suggesting that this product was present in only trace amounts.
Importantly, chlorotyrosine was not detected by ESMS or HPLC
analysis of the crude reaction mixture, which confirmed that
Figure 9. Carbonyl content (nmol/mg of protein) as a function of
pulsed H2O2 (0ꢀ500 μM) for enzymes (20 μM) trypsin (blue b) and
chymotrypsin (red 2) in the presence of 6 (50 μM). Data points are
averages from three independent experiments, where the errors are
reported as standard deviations.
Figure 10. Bar graphs showing the compositions of five natural amino acids present in the oxidized proteins samples isolated from single-turnover
(ferryls 7 and 8 in green and gold, respectively) and catalytic conditions (5 or 6 plus H2O2 in red and blue, respectively) for trypsin (a) and chymotrypsin
(b). Compositions are reported as percentages normalized with respect to the control experiments with no iron or oxidant added. Enzymes trypsin and
chymotrypsin (20 μM) were treated with ferryl 7 or 8 (1000 μM) or with 5 or 6 (50 μM) and pulsed with H2O2 (500 μM total, 100 μM per pulse).
Standard errors are 10%. The results indicate that tyrosine was modified selectively in the case of catalytic conditions with 5 or 6 plus H2O2, whereas
cysteine, tyrosine, and tryptophan were modified by ferryls 7 and 8. Full tables of calculated and experimental compositions for all 20 natural amino acids
and cystic acid are presented in the Supporting Information (Tables S1ꢀS4).
3942
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
Scheme 2
oxidation of 9 was not due to the action of HOCl generated from
H2O2 and chloride ion.80,81 Taken together, these observations
are diagnostic and significant because catechol products are not
formed when tyrosine derivatives are treated with the ferryl 7,
confirming that a different mechanism of action takes place when
H2O2 is used as the oxidant. In other words, ferryl 7 was not the
active oxidant. To probe this further, the reaction was monitored
by UVꢀvis spectroscopy. When the ferrous complex 5 was
treated with 0.5 equiv of H2O2 under these conditions, bleaching
occurred, which consumed the signature absorbances for FeII at
380 and 464 nm (Figure S9 in the Supporting Information).
However, ferryl 7 was not observed, suggesting that a ferric
complex was formed instead (vide infra). When this mixture was
treated with 9, the yellow color of the starting ferrous complex
returned and reached completion after approximately 10 min, as
judged by UVꢀvis spectroscopy. If this reaction solution was
treated again with peroxide, bleaching occurred again, and after
ca. 10 min, the UV spectrum returned to its original intensity,
suggesting that turnover was possible without significant loss of
the starting ferrous complex (Figure S10 in the Supporting
Information). In conclusion, these results are consistent with
the phenol ring of tyrosine being oxidized by an oxidant other
than ferryl 7, which is discussed in the following section.
Figure 11. Surface models of trypsin and chymotrypsin with tyrosine
residues near the active site highlighted in red. Trypsin contains four
residues near the active site, whereas chymotrypsin contains three.
oxidative stress, ferryls such as ferrylmyoglobin can oxidize pro-
teins.21,82ꢀ88 Observations reported herein reveal useful informa-
tion about how proteins are modified and inactivated by ferryls. For
example, when tyrosine and trytophan residues are oxidized, as our
results confirm, this type of oxidation is often irreversible.17 In
contrast, oxidative modification of cysteine residues can be reversed
if disulfides are formed.89 However, when higher oxidation states of
sulfur are obtained, such as sulfenic and sulfonic acid derivatives,
oxidative damage becomes irreversible. The levels of cystic acid did
not change in these studies between control and treated samples of
enzyme, consistent with the oxidants presented herein forming
products other than cystic acid.
Resonable levels of selectivity were observed in the oxidation
of trypsin and chymotrypsin catalyzed by 5 and 6 in the presence
of H2O2. However, as opposed to the single-turnover conditions
where ferryl 8 was the most effective at inactivating chymotryp-
sin, catalytic conditions provided the most efficient inactivation
of trypsin with both 5 and 6. The advantage of incorporating iron
and oxidant with the inhibitors 1 and 2 was clear. Upon addition
of these two reagents, a ligand concentration of 20 μM becomes
more effective than 500 μM inhibitor alone, which amplifies the
effects of the ligand by almost 2 orders of magnitude. The results
obtained from amino acid analysis prove that tyrosine residues of
both proteins were selectively targeted under these conditions,
with residues of trypsin being diminished to a greater extent
(60% of control) than chymotrypsin (80%). These results
suggest that iron complexes derived from N4Py mimic the action
of myeloperoxidase, a heme enzyme produced by neutrophil
granulocytes (white blood cells) that is known to oxidize tyrosine
residues of proteins using H2O2 as the oxidant.90
The selective inactivation of trypsin observed under catalytic
conditions may be due to the active iron reagent obtained with
H2O2, giving a higher effective molarity with trypsin than with
chymotrypsin. Alternatively, this selectivity could be due to an
inherent sensitivity of trypsin over chymotrypsin to oxidation of
its Tyr residues. Both enzymes have numerous tyrosine residues
on the surface near the active site (Figure 11). The fact that
ligands 1 and 2 were more effective at inhibiting trypsin than
chymotrypsin, in the absence and presence of FeII(ClO4)2,
supports the hypothesis that the active reagents derived from 5
and 6 bind tighter to trypsin than chymotrypsin. Furthermore,
previous studies performed with 1O2 confirmed that trypsin and
chymotrypsin have similar sensitivity to oxidation, although the
types of residues targeted in this case were likely different and
were not confirmed.91 More studies in this area are needed,
specifically over a larger group of enzymes such as the family of
’ DISCUSSION
Our results confirm that ferryls are potent and selective
oxidants that inactivate serine proteases. The ferryls 7 and 8
were more powerful oxidants than peracetic acid. Selectivity was
observed for a single ferryl between two enzymes (trypsin and chy-
motrypsin; Table 1, entry 7 vs entry 8) but also for a single enzyme
between two ferryls (Table 1, entries 7 and 8, trypsin vs chymo-
trypsin). Ferryl 8 was most effective in the inactivation of chymo-
trypsin with an IC50 value of 24 μM. With both serine proteases, the
oxidation of protein side chains occurred. Major cleavage products
were not observed by SDS-PAGE or LCMS, confirming that
cleavage of the protein backbone is a slow process relative to side-
chain oxidation. In the oxidation of trypsin by 8, LCMS analysis
indicated a minor cleavage product consistent with a loss of
approximately 2 kD, presumably at a glysine residue located near
the active site, 20 residues from the N terminus of the enzyme
(Figure S11 in the Supporting Information). Therefore, these
results are in good agreement with model studies that focused on
protected amino acids,63 which confirmed that activation of the R-
CH bond of amino acids and resultant oxidative cleavage of the
backbone are slow relative to side-chain oxidation with ferryl
complexes. Futhermore, the results from the amino acid analysis
of enzyme samples oxidized by the ferryls prove a direct correlation
between the most reactive amino acids (cysteine, tyrosine, and
tryptophan) in the model studies with Ac-AA-NHtBu substrates
and the residues modified in the proteins trypsin and chymotrypsin.
These results have relevance to biology because, under conditions of
3943
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
serine proteases, to discern between these two possibilities.
Importantly, these studies suggest that incorporating protein
affinity groups that bind tighter to enzyme targets and place iron
catalysts in the vicinity of tyrosine residues has the potential to
achieve protein inactivation at low concentrations.
’ ACKNOWLEDGMENT
We thank Mary Kay Pflum for helpful discussion, Brian Shay
and Lew Hryhorczuk for assistance with MS and protein LCMS
analysis, the University of California at Davis Molecular Structure
Facility for amino acid analysis, and Wayne State University for
its generous funding of this research.
The results with iron complexes and H2O2 indicate that a
unique oxidant was responsible for the selective protease in-
activation. Control experiments ruled out ROS such as HO•
1
(Fenton chemistry) or O2. The action of ferryls under these
’ REFERENCES
conditions is also unlikely because ferryls were not observed
when the ferrous complexes 5 or 6 were treated with peroxide.92
Furthermore, studies in Table 1 prove that chymotrypsin is
inactivated at lower ferryl concentrations than trypsin, so the
opposite trend in selectivity would be expected under catalytic
conditions if ferryls were operating. The action of a different
oxidant is also supported by the fact that catechol is observed in
the oxidation of Ac-Tyr-OH 9 with H2O2. Catechols are not
observed when tyrosine derivatives are treated with ferryl 7;
instead, hydrogen-atom transfer occurs to generate phenoxyl
radicals that decompose, presumably by polymerization.63 Tak-
ing into account our results and data in the literature,93 we can
conclude that another oxidant is likely to be generated under
catalytic conditions. However, further mechanisitc studies in this
area are needed to differentiate between the potential pathways
and to determine if there is an analogy between the chemistry
reported herein and the action of the nonheme iron enzyme
tyrosine hydroxylase52,94,95 or the heme enzyme myeloperoxi-
dase.90 In any case, the effect of the ligand N4Py is clear in the
inactivation experiments because FeII(ClO4)2 did not give the
same level of dose-dependent inactivation with H2O2.
(1) Schepartz, A.; Cuenoud, B. J. Am. Chem. Soc. 1990, 112, 3247–3249.
(2) Hoyer, D.; Cho, H.; Schultz, P. G. J. Am. Chem. Soc. 1990,
112, 3249–3250.
(3) Jeon, J. W.; Son, S. J.; Yoo, C. E.; Hong, I. S.; Song, J. B.; Suh, J.
Org. Lett. 2002, 4, 4155–4158.
(4) Gokhale, N. H.; Cowan, J. A. J. Biol. Inorg. Chem. 2006, 11,
937–947.
(5) Rokita, S. E.; Burrows, C. J. Met. Ions Biol. Syst. 2001, 38,
289–311.
(6) Brown, K. C.; Kodadek, T. Met. Ions Biol. Syst. 2001, 38,
351–384.
(7) Hegg, E. L.; Burstyn, J. N. J. Am. Chem. Soc. 1995, 117,
7015–7016.
(8) Hegg, E. L.; Burstyn, J. N. Coord. Chem. Rev. 1998, 173, 133–165.
(9) Polzin, G. M.; Burstyn, J. N. Met. Ions Biol. Syst. 2001,
38, 103–143.
(10) Suh, J.; Yoo, S. H.; Kim, M. G.; Jeong, K.; Ahn, J. Y.; Kim, M.-s.;
Chae, P. S.; Lee, T. Y.; Lee, J.; Lee, J.; Jang, Y. A.; Ko, E. H. Angew. Chem.,
Int. Ed. 2007, 46, 7064–7067.
(11) Chae, P. S.; Kim, M.-s.; Jeung, C.-S.; Lee, S. D.; Park, H.; Lee, S.;
Suh, J. J. Am. Chem. Soc. 2005, 127, 2396–2397.
(12) de Oliveira, M. C. B.; Scarpellini, M.; Neves, A.; Terenzi, H.;
Bortoluzzi, A. J.; Szpoganics, B.; Greatti, A.; Mangrich, A. S.; de Souza,
E. M.; Fernandez, P. M.; Soares, M. R. Inorg. Chem. 2005, 44, 921–929.
(13) Zhu, L.; Qin, L.; Parac, T. N.; Kostic, N. M. J. Am. Chem. Soc.
1994, 116, 5218–5224.
’ CONCLUSIONS
Nonheme iron complexes inactivate serine proteases selec-
tively. Ferryls are potent oxidants that inactivate trypsin and
chymotrypsin by oxidation of amino acid side chains rather than
backbone cleavage, where the residues cysteine, tyrosine, and
tryptophan are the most susceptible to oxidation. Oxidation of
the proteases by ferrous complexes in the presence of hydrogen
peroxide leads to the preferential inactivation of trypsin over
chymotrypsin. In this case tyrosine residues are targeted, and data
support action of a pathway unique from that of an FeII/FeIV
cycle involving ferryl compounds. Importantly, these results
suggest a promising future direction for the method presented
herein where complexes that cycle between accessible oxidation
states are used to target proteins. Although the scope of such a
method may be limited with respect to ROS for attack on
tyrosine residues, there are many proteins, including important
medicinal targets, that contain crucial tyrosine residues near their
active sites.
(14) Karet, G. B.; Kostic, N. M. Inorg. Chem. 1998, 37, 1021–1027.
(15) Milovic, N. M.; Kostic, N. M. J. Am. Chem. Soc. 2002,
124, 4759–4769.
(16) Stoffregen, S. A.; Griffin, A. K. K.; Kostic, N. M. Inorg. Chem.
2005, 44, 8899–8907.
(17) Berlett, B. S.; Stadtman, E. R. J. Biol. Chem. 1997, 272,
20313–20316.
(18) Garrison, W. M.; Jayko, M. E.; Bennett, W. Radiat. Res. 1962,
16, 483–502.
(19) Davies, M. J. Biochim. Biophys. Acta, Proteins Proteomics 2005,
1703, 93–109.
(20) Stadtman, E. R.; Berlett, B. S. Chem. Res. Toxicol. 1997, 10,
485–494.
(21) Stadtman, E. R. Free Radical Biol. Med. 1990, 9, 315–325.
(22) Gallagher, J.; Zelenko, O.; Walts, A. D.; Sigman, D. S. Biochem-
istry 1998, 37, 2096–2104.
(23) Lee, J.; Yu, P.; Xiao, X.; Kodadek, T. Mol. BioSyst. 2008,
4, 59–65.
(24) Lee, J.; Udugamasooriya, D. G.; Lim, H.-S.; Kodadek, T. Nat.
Chem. Biol. 2010, 6, 258–260.
(25) Jay, D. G. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 5454–5458.
(26) Beck, S.; Sakurai, T.; Eustace, B. K.; Beste, G.; Schier, R.;
Rudert, F.; Jay, D. G. Proteomics 2002, 2, 247–255.
(27) Yan, P.; Xiong, Y.; Chen, B.; Negash, S.; Squier, T. C.; Mayer,
M. U. Biochemistry 2006, 45, 4736–4748.
’ ASSOCIATED CONTENT
S
Supporting Information. Amino acid analysis data,
b
UVꢀvis spectra, HRMS and LCMS data, HPLC chromato-
grams, H and 13C NMR data, and three-dimensional figures
1
of trypsin and chymotrypsin. This material is available free of
charge via the Internet at http://pubs.acs.org.
(28) Cuenoud, B.; Tarasow, T. M.; Schepartz, A. Tetrahedron Lett.
1992, 33, 895–898.
(29) Gokhale, N. H.; Bradford, S.; Cowan, J. A. J. Am. Chem. Soc.
2008, 130, 2388–2389.
(30) Heyduk, T.; Baichoo, N.; Heyduk, E. Met. Ions Biol. Syst. 2001,
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: jkodanko@chem.wayne.edu.
38, 255–287.
(31) Heyduk, E.; Heyduk, T. Biochemistry 1994, 33, 9643–9650.
3944
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945

Inorganic Chemistry
ARTICLE
(32) Cheal, S. M.; Ng, M.; Barrios, B.; Miao, Z.; Kalani, A. K.;
Meares, C. F. Biochemistry 2009, 48, 4577–4586.
(67) Smith, K. C.; White, R. L.; Le, Y.; Vining, L. C. J. Nat. Prod.
1995, 58, 1274–1277.
(33) Traviglia, S. L.; Datwyler, S. A.; Yan, D.; Ishihama, A.; Meares,
C. F. Biochemistry 1999, 38, 15774–15778.
(34) Greiner, D. P.; Hughes, K. A.; Gunasekera, A. H.; Meares, C. F.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 71–75.
(68) Absorbances were significantly larger when the ferrous salt
[FeII(CH3CN)(N4Py)](ClO4)2 was used to make aqueous solutions of
5. This effect was due to the presence of CH3CN, consistent with
formation of the low-spin iron(II) species [FeII(CH3CN)(N4Py)]2þ
,
(35) Heilek, G.; Marusak, R.; Meares, C. F.; Noller, H. F. Proc. Natl.
Acad. Sci. U.S.A. 1995, 92, 1113–1116.
which has larger extinction coefficients than the high-spin species
[FeII(OH2)(N4Py)]2þ Conversion of [FeII(OH2)(N4Py)]2þ to
.
(36) Chen, K.; Que, L., Jr. J. Am. Chem. Soc. 2001, 123, 6327–6337.
(37) Fenton, H. J. H. J. Chem. Soc., Trans. 1894, 65, 899–910.
(38) Kaizer, J.; Klinker, E. J.; Oh, N. Y.; Rohde, J.-U.; Song, W. J.;
Stubna, A.; Kim, J.; M€unck, E.; Nam, W.; Que, L., Jr. J. Am. Chem. Soc.
2004, 126, 472–473.
[FeII(CH3CN)(N4Py)]2þ was confirmed by titration with CH3CN.
(69) Drago, R. S. Physical Methods for Chemists, 2nd ed.; Saunders:
Orlando, FL, 1992.
(70) Campanali, A. A.; Kwiecien, T. D.; Hryhorczuk, L.; Kodanko,
J. J. Inorg. Chem. 2010, 49, 4759–4761.
(39) Klinker, E. J.; Kaizer, J.; Brennessel, W. W.; Woodrum, N. L.;
Cramer, C. J.; Que, L., Jr. Angew. Chem., Int. Ed. 2005, 44, 3690–3694.
(40) Sastri, C. V.; Seo, M. S.; Park, M. J.; Kim, K. M.; Nam, W. Chem.
Commun. 2005, 1405–1407.
(41) Yoon, J.; Wilson, S. A.; Jang, Y. K.; Seo, M. S.; Nehru, K.;
Hedman, B.; Hodgson, K. O.; Bill, E.; Solomon, E. I.; Nam, W. Angew.
Chem., Int. Ed. 2009, 48, 1257–1260.
(42) Nehru, K.; Jang, Y.; Oh, S.; Dallemer, F.; Nam, W.; Kim, J. Inorg.
Chim. Acta 2008, 361, 2557–2561.
(43) Jeong, Y. J.; Kang, Y.; Han, A.-R.; Lee, Y.-M.; Kotani, H.;
Fukuzumi, S.; Nam, W. Angew. Chem., Int. Ed. 2008, 47, 7321–7324.
(44) Fukuzumi, S.; Kotani, H.; Lee, Y.-M.; Nam, W. J. Am. Chem. Soc.
2008, 130, 15134–15142.
(45) Nehru, K.; Seo, M. S.; Kim, J.; Nam, W. Inorg. Chem. 2007,
46, 293–298.
(46) Nehru, K.; Jang, Y. K.; Seo, M. S.; Nam, W.; Kim, J. Bull. Korean
Chem. Soc. 2007, 28, 843–846.
(47) de Visser, S. P.; Oh, K.; Han, A.-R.; Nam, W. Inorg. Chem. 2007,
46, 4632–4641.
(71) Jabre, N. D.; Hryhorczuk, L.; Kodanko, J. J. Inorg. Chem. 2009,
48, 8078–8080.
(72) Louie, A. Y.; Meade, T. J. Chem. Rev. 1999, 99, 2711–2734.
(73) Brown, K. C.; Yang, S.-H.; Kodadek, T. Biochemistry 1995,
34, 4733–4739.
(74) Aqueoussolutionsof5and 6decomposeoverthecourseof24hat
room temperature under an aerobic atmosphere, which indicates that the
oxidation of these ferrous complexes with O2 is a relatively slow process.
(75) Roelfes, G.; Lubben, M.; Hage, R.; Que, L., Jr.; Feringa, B. L.
Chem.—Eur. J. 2000, 6, 2152–2159.
(76) Li, F.; England, J.; Que, J. L. J. Am. Chem. Soc. 2010, 132,
2134–2135.
(77) Makhlynets, O. V.; Das, P.; Taktak, S.; Flook, M.; Mas-Balleste, R.;
Rybak-Akimova, E. V.; Que, L., Jr. Chem.—Eur. J. 2009, 15, 13171–13180.
(78) Hazen, S. L.; Zhang, R.; Shen, Z.; Wu, W.; Podrez, E. A.;
MacPherson, J. C.; Schmitt, D.; Mitra, S. N.; Mukhopadhyay, C.; Chen,
Y.; Cohen, P. A.; Hoff, H. F.; Abu-Soud, H. M. Circ. Res. 1999, 85, 950–958.
(79) Wever, R.; Bakkenist, A. R. J. Biochim. Biophys. Acta, Protein
Struct. Mol. Enzymol. 1980, 612, 178–184.
(48) Krebs, C.; Fujimori, D. G.; Walsh, C. T.; Bollinger, J. M., Jr. Acc.
(80) Fu, S.; Wang, H.; Davies, M.; Dean, R. J. Biol. Chem. 2000,
Chem. Res. 2007, 40, 484–492.
275, 10851–10858.
(49) Nam, W. Acc. Chem. Res. 2007, 40, 522–531.
(50) Que, L. Acc. Chem. Res. 2007, 40, 493–500.
(81) Bovonsombat, P.; Khanthapura, P.; Krause, M. M.; Leykajar-
akul, J. Tetrahedron Lett. 2008, 49, 7008–7011.
(51) Hoffart, L. M.; Barr, E. W.; Guyer, R. B.; Bollinger, J. M., Jr.;
Krebs, C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 14738–14743.
(52) Eser, B. E.; Barr, E. W.; Frantom, P. A.; Saleh, L.; Bollinger, J. M.,
Jr.; Krebs, C.; Fitzpatrick, P. F. J. Am. Chem. Soc. 2007, 129, 11334–11335.
(53) Hecht, S. M. Anticancer Agents Nat. Prod. 2005, 357–381.
(54) Burger, R. M. Chem. Rev. 1998, 98, 1153–1169.
(55) Roelfes, G.; Lubben, M.; Leppard, S.; Schudde, E. P.; Hermant,
R. M.; Hage, R.; Wilkinson, E. C.; Que, L., Jr.; Feringa, B. L. J. Mol. Catal.
A: Chem. 1997, 117, 223–227.
(82) Xiong, Y. L.; Park, D.; Ooizumi, T. J. Agric. Food. Chem. 2009,
57, 153–159.
(83) Ouellet, H.; Ranguelova, K.; LaBarre, M.; Wittenberg, J. B.;
Wittenberg, B. A.; Magliozzo, R. S.; Guertin, M. J. Biol. Chem. 2007,
282, 7491–7503.
(84) Reeder, B. J.; Svistunenko, D. A.; Sharpe, M. A.; Wilson, M. T.
Biochemistry 2002, 41, 367–375.
(85) Irwin, J. A.; Ostdal, H.; Davies, M. J. Arch. Biochem. Biophys.
1999, 362, 94–104.
(56) Guajardo, R. J.; Tan, J. D.; Mascharak, P. K. Inorg. Chem. 1994,
33, 2838–2840.
(86) Mikkelsen, A.; Skibsted, L. H. Z. Lebensm.-Unters. Forsch. 1998,
206, 199–202.
(57) Guajardo, R. J.; Hudson, S. E.; Brown, S. J.; Mascharak, P. K.
J. Am. Chem. Soc. 1993, 115, 7971–7977.
(87) Miura, T.; Muraoka, S.; Ogiso, T. Biochem. Mol. Biol. Int. 1995,
36, 587–594.
(58) Mialane, P.; Nivorojkine, A.; Pratviel, G.; Azema, L.; Slany, M.;
Godde, F.; Simaan, A.; Banse, F.; Kargar-Grisel, T.; Bouchoux, G.;
Sainton, J.; Horner, O.; Guilhem, J.; Tchertanova, L.; Meunier, B.;
Girerd, J.-J. Inorg. Chem. 1999, 38, 1085–1092.
(59) Decker, A.; Chow, M. S.; Kemsley, J. N.; Lehnert, N.; Solomon,
E. I. J. Am. Chem. Soc. 2006, 128, 4719–4733.
(88) Cooper, C. E. New Compr. Biochem. 1994, 28, 67–111.
(89) Mieyal, J. J.; Gallogly, M. M.; Qanungo, S.; Sabens, E. A.;
Shelton, M. D. Antioxid. Redox Sign. 2008, 10, 1941–1988.
(90) Bian, K.; Gao, Z.; Weisbrodt, N.; Murad, F. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 5712–5717.
(91) Suto, D.; Iuchi, Y.; Ikeda, Y.; Sato, K.; Ohba, Y.; Fujii, J. Arch.
(60) Stubbe, J.; Kozarich, J. W. Chem. Rev. 1987, 87, 1107–1136.
(61) Roelfes, G.; Branum, M. E.; Wang, L.; Que, L., Jr.; Feringa, B. L.
J. Am. Chem. Soc. 2000, 122, 11517–11518.
(62) Ekkati, A. R.; Kodanko, J. J. J. Am. Chem. Soc. 2007, 129,
12390–12391.
(63) Abouelatta, A. I.; Campanali, A. A.; Ekkati, A. R.; Shamoun, M.;
Kalapugama, S.; Kodanko, J. J. Inorg. Chem. 2009, 48, 7729–7739.
(64) Chenault, H. K.; Dahmer, J.; Whitesides, G. M. J. Am. Chem.
Soc. 1989, 111, 6354–6364.
(65) Sur, S. K. J. Magn. Reson. 1989, 82, 169–173.
(66) Levine, R. L.; Garland, D.; Oliver, C. N.; Amici, A.; Climent, I.;
Lenz, A. G.; Ahn, B. W.; Shaltiel, S.; Stadtman, E. R. Methods Enzymol.
1990, 186, 464–478.
Biochem. Biophys. 2007, 461, 151–158.
(92) It has been noted that treatment of 5 with H2O2 leads to the
ferric compound [FeIII(OH)(N4Py)]2þ in a MeOH solvent but gives rise to
7 in an aqueous solvent.40 Bleaching of 5 was observed with H2O2 in an
acetate buffer using 0.5 equiv. Larger amounts of H2O2 (>10 equiv) led to
the formation of the hydroperoxo compound [FeIII(OOH)(N4Py)]2þ as
judged by UVꢀvis spectroscopy (λmax = 515 nm) and ESMS. Ferryl 7 was
not detected.
(93) Das, P.; Que, L. Inorg. Chem. 2010in press.
(94) Fitzpatrick, P. F. Biochemistry 2003, 42, 14083–14091.
(95) Kitajima, N.; Ito, M.; Fukui, H.; Morooka, Y. J. Am. Chem. Soc.
1993, 115, 9335–9336.
3945
dx.doi.org/10.1021/ic102320j |Inorg. Chem. 2011, 50, 3934–3945