Mn(II) Binding and Subsequent Oxidation by the Multicopper Oxidase MnxG Investigated by Electron Paramagnetic Resonance Spectroscopy

Article abstract of DOI:10.1021/jacs.5b04331

The dynamics of manganese solid formation (as MnO_x) by the multicopper oxidase (MCO)-containing Mnx protein complex were examined by electron paramagnetic resonance (EPR) spectroscopy. Continuous-wave (CW) EPR spectra of samples of Mnx, prepared in atmosphere and then reacted with Mn(II) for times ranging from 7 to 600 s, indicate rapid oxidation of the substrate manganese (with two-phase pseudo-first-order kinetics modeled using rate coefficients of: k_1obs = 0.205 ± 0.001 s^-1 and k_2obs = 0.019 ± 0.001 s^-1). This process occurs on approximately the same time scale as in vitro solid MnO_x formation when there is a large excess of Mn(II). We also found CW and pulse EPR spectroscopic evidence for at least three classes of Mn(II)-containing species in the reaction mixtures: (i) aqueous Mn(II), (ii) a specifically bound mononuclear Mn(II) ion coordinated to the Mnx complex by one nitrogenous ligand, and (iii) a weakly exchange-coupled dimeric Mn(II) species. These findings provide new insights into the molecular mechanism of manganese mineralization.

Full text of DOI:10.1021/jacs.5b04331

Article
pubs.acs.org/JACS
Mn(II) Binding and Subsequent Oxidation by the Multicopper
Oxidase MnxG Investigated by Electron Paramagnetic Resonance
Spectroscopy
†
†
§
§
∥
Lizhi Tao, Troy A. Stich, Cristina N. Butterfield, Christine A. Romano, Thomas G. Spiro,
§
†,‡
,†
Bradley M. Tebo, William H. Casey, and R. David Britt*
†
‡
Department of Chemistry and Department of Geology, University of California, One Shields Avenue, Davis, California 95616,
United States
§
Division of Environmental and Biomolecular Systems, Institute of Environmental Health, Oregon Health & Science University,
Portland, Oregon 97239, United States
∥
Department of Chemistry, University of Washington, Box 351700, Seattle, Washington 98195, United States
*
S Supporting Information
ABSTRACT: The dynamics of manganese solid formation
as MnO ) by the multicopper oxidase (MCO)-containing
(
x
Mnx protein complex were examined by electron paramagnetic
resonance (EPR) spectroscopy. Continuous-wave (CW) EPR
spectra of samples of Mnx, prepared in atmosphere and then
reacted with Mn(II) for times ranging from 7 to 600 s, indicate
rapid oxidation of the substrate manganese (with two-phase
pseudo-first-order kinetics modeled using rate coefficients of:
−1
−1
k1obs = 0.205 ± 0.001 s and k2obs = 0.019 ± 0.001 s ). This
process occurs on approximately the same time scale as in vitro solid MnO formation when there is a large excess of Mn(II). We
x
also found CW and pulse EPR spectroscopic evidence for at least three classes of Mn(II)-containing species in the reaction
mixtures: (i) aqueous Mn(II), (ii) a specifically bound mononuclear Mn(II) ion coordinated to the Mnx complex by one
nitrogenous ligand, and (iii) a weakly exchange-coupled dimeric Mn(II) species. These findings provide new insights into the
molecular mechanism of manganese mineralization.
1
. INTRODUCTION
oxidized to the Cu(II) oxidation state, T1 sites are
characterized by an intense absorption feature at ≈590 nm
A key step of the global redox cycling of manganese is the
oxidation of aqueous Mn(II) to yield solid MnO , with the rates
2
2
that is assigned to a ligand-to-metal (Cys S → Cu 3d
)
π
x −y
x
1
,2
charge-transfer transition. Oxidized T1 sites are also known to
give rise to electron paramagnetic resonance (EPR) signals with
a small 6 Cu hyperfine coupling (A|| ≈ 4−10 mT). These
spectroscopic properties are diagnostic of the highly covalent
Cu−Cys S bond. Oxidized type 2 (T2) copper (with square-
depending sensitively on solution pH. In seawater (pH =
.16), abiotic Mn(II) oxidation proceeds very slowly with a
8
3,65
−5
−1
3
pseudo-first-order rate constant ca. 6.7 × 10 h . However,
microorganisms (bacteria and fungi)
process by 3−5 orders of magnitude. In marine Bacillus
bacteria, a cluster of seven genes (mnxA through mnxG) was
identified as being responsible for Mn(II) oxidation. One of
these genes mnxG encodes a putative multicopper oxidase
4
−8
can speed up this
2
63,65
planar geometry) centers exhibit larger
Cu hyperfine
9
,10
splitting value (A ≈ 20 mT). Oxidized type 3 (T3) copper
||
sites are antiferromagnetically coupled binuclear Cu(II) clusters
bridged by oxygen ligands (either hydroxido, oxido, or bis
2
,9−11
(MCO),
similar to the laccase, ascorbate oxidase, human
12,13
oxido) and tend to be EPR silent.
Some MCOs have
ceruloplasmin (hCp), and Fet3p enzymes which oxidize a
variety of organic and inorganic substrates coupled to reduction
of O to H O.
additional copper centers: for example, hCp has two additional
1
2,13
T1 copper sites (giving a total of six copper ions per
The MnxG protein was recently isolated as
2
2
15,16
protein).
One of these T1 sites appears to be redox active
part of a multiprotein complex along with gene products MnxE
16
during the ferroxidase reaction, implying a mechanistic role.
and MnxFa complex (denoted as “Mnx” in the following)
14
The MCO-catalyzed reduction of O is well stud-
2
with a molecular weight of ≈230 kDa.
12,13,17−19
ied.
In the simplest case, the resting oxidized state
MCO enzymes contain three types of copper cofactors. Type
(T1), or blue copper sites, have the copper ion coordinated in
of MCO is known to have all four copper ions in the 2+
1
oxidation state, and a μ -OH bridge exists between the two
a trigonal-planar-type geometry by two histidine and one
cysteine side chains. An axial ligand sits above this plane and is
generally a weakly bound methionine residue, but this amino
2
Received: April 27, 2015
13
acid is not strictly conserved across all T1 sites. When
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
copper centers of the T3 site. The T1 copper accepts electrons
from substrate molecules and funnels them via intramolecular
transferred by glass pipet to mix with Mnx protein. The mixing was
achieved by first bringing the tip of the pipet filled with MnCl2
aqueous solution to the bottom of the EPR tube prefilled with 100 μL
−1
17
electron transfer (IET, kIET ≈ 0.11 s at 4 °C) over ≈13 Å to
the trinuclear copper site (TNC), consisting of one T2 and a
T3 site). Upon the accumulation of four electrons from
substrate, the MCO is fully reduced. At this stage, exogenous
Mnx protein buffer solution and then by repeatedly pipetting the
solution up and down in the tip for complete mixing. After being
allowed to react for set times (7−600 s) at room temperature (22 °C),
samples were frozen and stored in liquid N .
2
6
−1 −1
O binds to and rapidly oxidizes (k ≈ 10 M s ) the TNC
2
eff
Fully reduced Mnx protein was prepared anaerobically in a glovebox
by two electrons leading to formation of the peroxy
intermediate (PI). PI then rapidly decays to the so-called
native intermediate (NI) that spectroscopic evidence shows to
that had been purged with N . A 5-fold excess of sodium dithionate
2
was added to Mnx and allowed to react for 30 min during which the
blue color from the T1 copper center (λmax = 590 nm) bleached.
Excess sodium dithionate was removed by desalting the solution using
a spin column (Thermo Scientific, #89849) that had been pre-
equilibrated with deoxygenated HEPES-buffered solution. A 3-fold
exchange of the buffer was accomplished using 0.5 mL centrifugal
filters (Amicon Ultra, #UFC510024). The resulting reduced protein
was checked by X-band CW EPR spectroscopy for residual signals
from Cu(II)-containing species. No such signals were observed,
indicating that all the Cu(II) centers in Mnx were fully reduced by the
treatment. This reduced protein solution is denoted as “RedMnx”. All
the Mnx-containing samples prepared for this study used either as-
isolated oxidized Mnx protein complex or RedMnx.
have a μ -oxido bridging the three copper centers of the TNC,
3
while the μ -OH remains bound between the two copper
2
18
centers of the T3 site. In the absence of additional O , the
native intermediate will slowly decay (≈0.007 s at 4 °C) back
to the resting, fully oxidized state.
2
−1
17
In the present paper, we examine the binding of Mn(II)
substrate to the Mnx protein complex using pulse EPR
spectroscopic methods. We find that one proteinaceous
nitrogen ligand and two water ligands coordinate to a
specifically bound mononuclear Mn(II) center. Continuous-
wave (CW) EPR studies suggest that a dimeric Mn(II) species
is formed early in the reaction cycle. CW EPR studies are
further employed to monitor the kinetics of manganese
oxidation catalyzed by the MCO and reveal that Mn(II)
oxidation occurs on a time scale that is approximately equal to
that for bulk MnO2 formation. Collectively, these results
provide new insights into the molecular mechanism of
biological Mn(II) oxidation.
Under anaerobic conditions, 100 μL aliquots of 50 μM RedMnx
were pre-incubated with 4 equiv of MnCl before being mixed with air-
2
saturated ([O ] ≈ 260 μM) HEPES buffer solution (as above). These
2
mixtures were allowed to react for different times (23, 60, 180, and 300
s) before being hand-frozen in liquid N . The sample denoted ‘0 s’ was
2
obtained by mixing with deoxygenated buffer instead so that no
oxidation occurred.
Samples of RedMnx that were anaerobically incubated with varying
amounts of MnCl
([Mn(II)]:[RedMnx] = 0.2, 0.8, 4.0, and 6.0 equiv)
2
were prepared in a glovebox purged with N . To make these samples,
2
1
00 μL RedMnx was mixed with 100 μL of the appropriate MnCl2
2
. EXPERIMENTAL PROCEDURES
solution and allowed to incubate for 30 min in an X-band EPR tube
before being frozen and stored in liquid N2.
2
.1. Sample Preparation. Protein was purified by native methods
20 14,21
or by Strep-tag affinity chromatography as described previously.
In short, MnxEFG protein was expressed from a mnxDEFG gene
construct and loaded with Cu by incubating the static Escherichia coli
Mn(II) bound to bovine serum albumin (BSA) was prepared by
mixing 100 μL 326 μM BSA (Sigma-Aldrich, CAS: 9048-46-8) in
HEPES buffer solution (pH = 7.8) with 100 μL 256 μM MnCl2
aqueous solution. The mixing procedure in the X-band EPR tube was
identical to that of aerobically hand-quenched samples of Mnx protein
BL21 (DE3) culture with CuSO to allow for microaerobic in vivo
4
22
uptake of Cu. The protein was purified by a series of native
chromatographic separations or on a Strep-tactin affinity column.
Excess Cu was removed with extensive dialysis in 20 mM HEPES, 50
mM NaCl, pH 7.8. The protein was then concentrated and flash
and MnCl . After mixing for 5 min at room temperature (22 °C), the
sample was frozen and stored in liquid N2.
2
A sample of RedMnx was anaerobically incubated with 0.8 equiv of
frozen in liquid N . Since protein yield was low (approximately 3 mg
MnCl in D O that had been prepared in a glovebox purged with N in
2
2
2
2
protein per L culture), samples were thawed and pooled to carry out
the same procedure as sample “RedMnx + 0.8Mn(II)”, except that
spectroscopic studies.
HEPES buffer in D O (20 mM HEPES, 20 mM NaCl, 20% ethylene
2
In order to determine possible ligands to the T1 copper site, a
comparison of the Mnx amino acid sequence against that of other
MCOs was performed. His340 was identified as a potential ligand to
glycol, pD = 8.2) was used for buffer-exchange procedure for three
times and mixed with 0.8 ratio MnCl in D O instead. In addition,
2
2
standard samples of 237 μM MnCl in both H O and D O as well as
2
2
2
9
,10
the T1 copper.
extension PCR.
The H340A mutant was generated by overlap
220 μM MnDTPA (DTPA = diethylene triamine pentaacetic acid) in
2
3
The inner primers used were 5′-
both H O and D O were prepared (with 20% ethylene glycol) as
2
2
GTATCCGgcCTTCGGC-3′ (forward) and 5′-GCCGAAGgcCGGA-
TAC-3′ (reverse). The nucleotides in lowercase bold mark the
location of a CAC → GCC mutation that encodes “A” instead of “H”.
The outer primers used were 5′-GTTAATACGTTCGGGGACCA-3′
and 5′-TATGCGGCATGCGATTAGTA-3′, which flank the HindIII
and SbfI sites within the mnxG sequence. These restriction sites were
used to clone the mutant H340A fragment into the pTXB1
overexpression plasmid, and the resulting construct was transformed
into E. coli BL21 (DE3). The expression and purification procedure
standards for water counting by electron spin−echo envelope
modulation (ESEEM) spectroscopy.
2.2. Spectroscopy. X-band CW EPR spectra were recorded using
a Bruker (Billerica, MA) Biospin EleXsys E500 spectrometer.
Cryogenic temperatures were achieved and controlled using an
ESR900 liquid helium cryostat in conjunction with a temperature
controller (Oxford Instruments ITC503) and gas flow controller. For
perpendicular-mode EPR (B0 ⊥ B ), a superhigh Q resonator
1
(ER4122SHQE) was employed, while all parallel-mode EPR experi-
14
was similar to that of wild type Mnx protein complex.
ments (B || B ) made use of a dual-mode resonator (ER4116DM). All
0
1
Reaction mixtures of Mnx with aqueous Mn(II) and O2 were
prepared both aerobically and anaerobically. Aerobically as-isolated
CW-EPR data were collected under slow-passage, nonsaturating
conditions. Spectrometer settings were as follows: conversion time =
40 ms, modulation amplitude = 0.8 mT, and modulation frequency =
100 kHz; other settings are given in corresponding figure captions.
X-band (9.47−9.53 GHz) electron-spin−echo (ESE)-detected field-
swept (FS) EPR (π/2-τ-π-τ-echo) and three-pulse ESEEM spectra (π/
2-τ-π/2-T-π/2-τ-echo) were collected using a Bruker EleXsys E580
spectrometer equipped with a cylindrical dielectric cavity (MD5,
Bruker). Experimental parameters are given in the corresponding
oxidized Mnx was reacted with 4 equiv of MnCl and frozen by hand
2
after different reaction times. First, at room temperature (22 °C), 100
μL Mnx (50 μM) in a pH-buffered solution (20 mM HEPES, 20 mM
NaCl, pH = 7.8, 20% ethylene glycol and saturated with O from air)
2
was transferred to the bottom of quartz X-band EPR tube (Wilmad
PQ-706, ID = 2.8 mm, OD = 3.8 mm) using a glass pipet. Then 100
μL MnCl (200 μM) aqueous solution saturated with O from air was
2
2
B
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
figure captions. Spectral simulations were performed using the
amino acid sequence of the mnx gene products, this molecular
24,25
Easyspin 4.0 toolbox
Mathworks Inc., Natick, MA).
.3. Quantification of Aqueous Mn(II). In order to determine
within the MatLab software suite (The
weight can be obtained from oligomerization of a combination
14,21
of between 6 and 8 copies of MnxE and F with one MnxG.
2
The Mnx protein has been analyzed for metal content
previously using inductively coupled plasma-optical emission
spectroscopy, and these analyses reveal ≈10 copper ions per
the concentration of aqueous Mn(II) in reaction mixture EPR samples,
a set of standard MnCl2 samples was made with concentrations
ranging from 25 to 200 μM in HEPES buffer (as described above).
The X-band CW EPR is characterized by a sextet of derivative-shaped
peaks split by ≈9.0 mT. These features arise from the central EPR
14
protein complex. Amino acid sequence analysis and other
biochemical assays suggest that there is only one T1 copper site
and one dinuclear T3 copper site in the complex, leaving
transition (|m = +1/2, m > ↔ |m = −1/2, m >) of the high-spin (S =
S
I
S
I
55
9
,10
5
/2) Mn(II) split by the hyperfine interaction with the I = 5/2 Mn
26,27
approximately seven copper centers of the T2 variety.
nucleus (100% natural abundance).
The peak-to-peak amplitude
The as-isolated, oxidized Mnx protein was analyzed by X-
band (9.68 GHz) CW EPR spectroscopy, and the correspond-
ing spectrum (blue trace, top of Figure 1) is dominated by
contributions from at least two classes of T2 copper sites as
well as the T1 site. The oxidized T3 site is expected to be EPR-
silent (vide supra). EPR spectra of T2 coppers are axial with g >
of the sixth, highest-field hyperfine peak was correlated to the
concentration of aqueous/high-symmetry Mn(II) after scaling to
account for incident microwave power and the quality factor of the
resonator. The resulting linear equation, [Mn(II)] = 8.718 ×
intensitypeak‑to‑peak amplitude, (R2 = 0.995), was used to determine the
concentration of aqueous Mn(II) in reaction mixture samples (see
below).
||
g⊥ and possess a quartet of 6 Cu hyperfine lines (both copper
3,65
2
.4. Counting Bound Waters. The number of water ligands
bound to a Mn(II) center in Mnx can be determined by evaluating the
solvent-exchangeable deuterium modulation depth in three-pulse
28
ESEEM spectra. The interpulse delay value τ was set to a value
1
equal to three times the H precessional period at each magnetic field
2
9
in order to suppress modulation from weakly coupled protons. The
ESEEM data of Mn(II) bound to Mnx were scaled to a value of 1.0 at
the first observable maximum in the spectrum. The resultant ESEEM
spectrum of the sample prepared in buffered D O was divided by that
2
of an identical sample in H O in order to cancel out all modulations
2
from nonexchangeable protons and from other kinds of hyperfine-
1
4
coupled nuclei (i.e., N). The resultant quotient (or “D O/H O-
2
2
ratioed”) spectrum consists only of modulations arising from solvent-
exchangeable deuterons. These deuterons are typically found in the
first- and second-coordination sphere of the Mn(II) ion (red trace,
shown in Figure S1A). In order to obtain signals from only those first-
coordination-sphere deuterons, the quotient spectrum was subse-
quently divided by the D O/H O-ratioed ESEEM spectrum of
2
2
Mn(II)DTPA. Mn(II)DTPA possesses no inner-sphere exchangeable
deuterons, therefore the D O/H O-ratioed spectrum (red trace,
2
2
shown in Figure S1B) represents modulation arising from only
second-coordination-sphere deuterons that are weakly coupled to the
Mn(II) spin center. Ideally, this procedure of dividing all experimental
D O/H O-ratioed ESEEM data by the D O/H O spectrum of the
2
2
2
2
Mn(II)DTPA complex leaves only modulation from inner-sphere-
coordinated deuterons. This procedure applied to Mn(II) in Mnx can
be summarized by the equation:
Mn(II) + Mnx(D O)/Mn(II) + Mnx(H O)
2
2
Mn(II)DTPA(D O)/Mn(II)DTPA(H O)
2
2
It is most likely that the inner-sphere deuterons detected this way
are part of ligand waters. In order to determine how many waters ligate
the Mn(II) center, we compare the resultant first-coordination-sphere
deuterium-modulation depth to that for aqueous Mn(II) (aq) ion
made by dissolving MnCl in either H O or D O buffers, yielding
Figure 1. X-band CW EPR spectra of Mnx protein samples mixed with
Mn(II) and hand-quenched in air after being allowed to react for times
ranging from 7 to 600 s. The ratio of MnCl to Mnx protein was 4 at
2
2
2
2
2+
2+
either [Mn(OH ) ] or [Mn(OD ) ] ions (shown in Figure S1C).
time = 0 s; 100 μL of 50 μM Mnx protein mixed with 100 μL of 200
2
6
2 6
Standard modulation patterns for the case where there are less than six
μM MnCl (total volume: 200 μL); both solutions were saturated with
2
water ligands can be generated from the MnCl data by taking the 1/6,
O from air. Spectra of as-isolated Mnx protein (topmost blue trace)
2
2
2
/6, 3/6, 4/6, and 5/6 root of the inner-sphere deuterium modulation
and 100 μM MnCl in buffer (bottommost blue trace) are presented
for comparison. Inset shows the low field spectral region around g =
2
for the D O/H O-ratioed of aqueous Mn(II) (Figure S1D). As an
2
2
added check for contributions from paramagnetic contaminants (e.g.,
Cu(II) from Mnx), this procedure was performed at two magnetic
field positions, 327.3 and 363.0 mT, the latter being one at which
Cu(II) resonances do not contribute.
4.3 (B = 155 mT). Experimental parameters: temperature = 15 K;
0
microwave frequency = 9.68 GHz; microwave power = 0.6325 mW;
conversion time = 40 ms; modulation amplitude = 0.8 mT;
modulation frequency = 100 kHz. Simulation (red trace) of the as-
isolated Mnx complex was achieved using three spectroscopically
distinct Cu(II) species: g = [2.065 2.050 2.305] and A = [45 45 210]
3
. RESULTS
.1. EPR Characterization of Copper in Mnx Protein
1
1
MHz for T1 Cu; g = [2.035 2.055 2.280] and A = [27 27 600] MHz
2
2
3
for T2 Cu−A; g = [2.055 2.050 2.210] and A = [30 30 600] MHz for
3
3
Complex. The large molecular weight ≈230 kDa of the Mnx
complex was validated by gel filtration. Based on the predicted
T2 Cu−B; with arbitrary ratio of 1:1:2 for T1 Cu:T2 Cu−A: T2 Cu−
B.
C
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
63
65
isotopes, Cu, 69% abundant, and Cu, 31% abundant, are I =
/2) centered at the field position corresponding to g with
3
||
splittings of approximately 20 mT (560 MHz). One class of T2
signals is characterized by g = 2.28 A = 600 MHz (denoted as
|
|
||
T2 Cu−A) and the other has g = 2.21 A = 600 MHz (denoted
|
|
||
as T2 Cu−B); results consistent with those observed
2
1
previously.
As up to seven T2 copper sites contribute to the EPR
spectrum, it is difficult to unambiguously assign the signal
arising from the lone T1 copper center. Sequence analysis
identified a possible histidine ligand (His340) to the T1 site of
9
,10
the MnxG protein.
H340A), the blue color of the Mnx protein was lost (Figure
In mutating this residue to alanine
(
S2A), signaling that the electronic structure of T1 copper site
was significantly perturbed. The EPR spectrum of this H340A
mutant (green trace shown in Figure S2B) shows loss of two
narrower signals at 275.0 to 286.0 mT in the spectrum of the
oxidized wild-type (WT) Mnx, which we assign to T1 Cu with
g = 2.30 A = 210 MHz. This H340A mutant can still bind
Figure 2. Aqueous Mn(II) concentration measured by CW EPR in
freeze-quenched mixtures of Mnx and MnCl as a function of reaction
2
|
|
||
time. Red-filled circles correspond to data in Figure 1 from samples
prepared aerobically. Black diamonds correspond to data in Figure 3
from samples prepared anaerobically before being mixed with an O2-
copper at the T1 binding site, however, the modified ligand set
leads to a change of the copper ion electronic structure
3
0
properties. Canters and co-workers reported that an
analogous His-to-Ala mutation to the T1 site ligand set in
nitrite reductase altered the redox properties of this site such
that it could only be oxidized in the presence of exogenous
ligands (either imidazole or chloride). Therefore, one reason
for the loss of the T1 Cu EPR signal from the spectrum of the
H340A mutant of Mnx could be due to it being locked into the
containing buffer solution. Dashed lines correspond to fits obtained
using pseudo-first-order kinetics. For the aerobic data set, two phases
−
1
−1
were employed with k1obs = 0.205 s and k2obs = 0.019 s . For the
−1
anaerobic data set, one phase was employed with kobs = 0.015 s . Free
Mn(II) concentrations were determined using the relation [Mn(II)] =
2
8
.71 × intensitypeak‑to‑peak amplitude, (R = 0.995) as described in the
Experimental Procedures section.
3
1
reduced form.
This signal loss is attributed to either oxidation of the
substrate to the 3+ or 4+ oxidation state or coordination of the
formerly aqueous Mn(II) ion in a binding site that induces a
significantly larger zero-field splitting (ZFS) interaction, which
would broaden the monitored EPR feature possibly beyond
Some of the T2 copper centers are also likely found in the
MnxE and MnxF proteins; whether these are involved in
manganese oxidation is unknown. While future spectroscopic
investigations will explore more precisely the distribution of
coppers about the Mnx protein complex, for the present study,
we have established that both T1 and T2 copper sites are
spectroscopically observable, consistent with all other MCOs.
This EPR spectrum of Mnx in its as-isolated, oxidized state will
serve as the starting point for our analysis of spectral changes
incurred during Mnx-catalyzed Mn(II) oxidation described
next.
26,27,32,33
detection.
Given these two possibilities, the data (red
circles) in Figure 2 were fit using a biphasic kinetic model (red
circles) dashed line in Figure 2. The major component
(
0
0
≈68.7%) had an effective rate constant of k
= 0.205 ±
1obs
.001 s− and the minor component (≈31.3%) had k
1
=
obs
2
−1
.019 ± 0.001 s .
We tentatively assign the more rapid process to the binding
3
.2. Kinetics of Mn(II) Oxidation. As-isolated, oxidized
of Mn(II) to a specific site within the Mnx complex, although
Mnx protein was mixed in air with a solution containing 4 equiv
the exact K is unknown. The slower process is assigned to
d
of Mn(II) in atmosphere-equilibrated buffer ([solvated O ] ≈
2
substrate oxidation as the observed rate of Mn(II) oxidation is
260 μM). The aging time of the reaction mixtures was varied
quite similar to the reported rate Fe(II) oxidation by Rhus
from 7 to 600 s after which the reaction was quenched by
freezing the samples in liquid nitrogen. X-band (9.68 GHz)
CW EPR spectra were then collected (Figure 1, black traces).
In the spectra of the early freeze-quenched samples (frozen
within 73 s of mixing), we observe the iconic sextet of
derivative-shaped peaks for Mn(II)-containing species (cf.
−1
34
vernicifera laccase (0.029 s at 25 °C, pH = 7.0). We note,
however, that faster single turnover kinetics are reported for
Fe(II) oxidation catalyzed by other MCO metallo-oxidases, e.g.,
the reduction rate constant for the T1 Cu in Fet 3p is >1200
−1
35
−1
s
(4 °C, pH = 7.0) and in hCp it is >150 s (25 °C, pH =
16,34
7
.0).
Both of these processes are on the same time scale as
bottommost trace corresponding to aqueous MnCl in Figure
2
the rate of formation of bulk solid MnOx measured by
1
4
1) in addition to signals for the T1 and T2 copper sites
electronic absorption spectroscopy (λmax ≈ 380 nm).
described in Section 3.1 above (top trace in Figure 1). This
aqueous Mn(II) EPR signal will be denoted as the “class i”
signal for the remainder of this work. As described in the
Experimental Procedures section, we used the peak-to-peak
intensity of the sixth, highest-field EPR feature of this Mn(II)
sextet (peak at 369.0 mT, vertical dotted line in Figure 1) as an
indicator of aqueous Mn(II) concentration during turnover.
This signal rapidly decreases in intensity with increasing aging
time (Figure 2, red circles). Samples frozen after 73 s of
reaction are completely devoid of aqueous Mn(II) signals.
Though the Mn(II) signal decreases monotonically with
time, the signals for T2 copper sites are evident at all times
(
Figure 1). During the reaction, subtle changes in relative
intensities of the various Cu(II) signals do occur, though a
complete analysis of these changes is beyond the scope of this
study.
In addition to the loss of the aqueous Mn(II) signal, there is
a new multiplet apparent at lower field that appears in the
spectra of early reaction mixtures and eventually decays away
(see inset of Figure 1). In the spectrum of the sample freeze-
D
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
quenched at 13 s, one can clearly see a regular splitting of
of O , this multiplet is only observed in spectra of samples
2
approximately 4.5 mT between at least five lines ca. 136.0 mT
frozen after an aging times of 13 s but disappears from those
aged longer than 180 s. This behavior suggests that the putative
dimer takes some time to form from aqueous Mn(II) ions
before it becomes spectroscopically silent againpresumably
as a result of manganese-centered oxidationconsistent with
the biphasic kinetics of Mn(II) signal loss that we observe
(Figure 2).
(
we call this the “class iii” signal, vide infra). This field position,
corresponding to g ≈ 5.0, is typically where the most intense
features of high-spin, S = 3/2, Mn(IV)-containing species are
found.
3
6−40
However, we disfavor this assignment as similar
features are also present in the spectrum of a sample of
anaerobically prepared, dithionite-reduced Mnx that had been
incubated with Mn(II) (see top trace in inset of Figure 3).
To test this hypothesis, we investigated the oxidation of
Mn(II) by RedMnx protein that was pre-incubated with 4 equiv
Mn(II) and then reacted with O -saturated buffer solution, yet
2
maintained in a N atmosphere. As mentioned previously, by
2
pre-incubating Mnx with Mn(II), we assume that the substrate-
binding site is already loaded with Mn(II) before the reaction
begins. The aging times after mixing were varied from 0 to 300
s. The corresponding EPR spectra of these reaction mixtures
are shown in Figure 3. For the first time point (no O added), a
2
typical Mn(II) sextet centered at g ≈ 2.0 is observed as well as
the multiplet tentatively assigned to a Mn(II,II) species
centered at 136.0 mT. In addition to this signal at 136.0 mT
from the Mn(II,II) dimer, there is a sextet of EPR features
centered at 170 mT and split by 9.0 mT (top spectrum, Figure
3
inset) that we assign to the half-field transition of a Mn(II)-
containing species with a modest ZFS value (D ≈ 1000
26,27
MHz).
signal is achieved using D = 1080 MHz and E = 356 MHz
Figure S3). Higher-field EPR spectroscopic studies are
Indeed, an acceptable preliminary simulation of the
(
underway to determine the ZFS parameters more assuredly.
We will refer to this manganese center as the “class ii”
mononuclear Mn(II) species for the remainder of this work.
ZFS arises from the introduction of angular momentum in the
ground electronic state caused by molecular distortions and
Figure 3. X-band CW EPR spectra of anaerobically prepared samples
of RedMnx protein pre-incubated with 4 equiv Mn(II) and hand-
leads to the breaking of the degeneracy of the M = | ±1/2⟩, |
s
quenched after being allowed to react with an O -containing solution
±3/2⟩, | ±5/2⟩ Kramers doublets of the S = 5/2 Mn(II) ion in
2
for times ranging from 0 to 300 s. 100 μL of 50 μM RedMnx protein
and 200 μM Mn(II) was mixed with 100 μL buffered solution
the absence of an applied magnetic field. The energy spacing
2
s
between these levels goes as DM (where D is the axial ZFS
saturated with O from air (total volume: 200 μL). Topmost, blue
2
parameter) and further by a function of E (the rhombic term).
This rhombic distortion for Mn(II) centers leads to mixing spin
trace is the spectrum of a sample that was not reacted with O . Inset
2
shows the low field spectral region around g = 4.3 (B = 155 mT).
0
states differ by M = ± 2, providing possibilities for transitions
s
Experimental parameters: temperature = 15 K; microwave frequency =
2
6,27
between levels that formally differ by Δm = ± 2.
This will
s
9
.37 GHz; microwave power = 0.6325 mW; conversion time = 40 ms;
give rise to EPR features at half the resonant field position for a
modulation amplitude = 0.8 mT; modulation frequency = 100 kHz.
4
6−52
free electron.
We see some evidence of these half-field
features in the data presented in Figure 1 (see traces
corresponding to 13 and 30 s freezing times), but clearly the
concentration of the corresponding Mn(II)-containing species
is quite low.
Absent from the spectrum of this anaerobic preparation are any
signals from T1 or T2 sites in the Cu(II) oxidation state. With
no oxidized copper centers and no O , there is no way for Mnx
2
to oxidize Mn(II) to Mn(IV).
Once the samples are reacted with O , Cu(II) EPR signals
2
Instead, we suggest that the multiplet at 136.0 mT arises
from two spatially close, weakly exchange-coupled Mn(II) ions.
The typical EPR features for such a spin system contain 11 lines
appear, and all three Mn(II) signals start to diminish in
5
5
intensity monotonically. Again, by monitoring the sixth Mn
hyperfine peak at 360 mT (vertical dotted line in Figure 3), we
have determined aqueous Mn(II) concentration as a function
of mixing time (presented in Figure 2, black diamonds). This
signal loss is well-modeled (black dashed line) by a single
(
with a 1:2:3:4:5:6:5:4:3:2:1 intensity pattern) split by ≈4.5
55
mT due to equivalent hyperfine interactions with both Mn
3
3,41
nuclei.
Isolated Mn(II) ions generally possess an isotropic
Mn hyperfine coupling constant between 8.5 and 9.0 mT, so
the observed coupling of 4.5 mT in this case is strongly
5
5
−1
decaying exponential with kobs = 0.015 ± 0.001 s . That not all
of the Mn(II) signal is gone at long reaction times suggests that
42
indicative of a dimeric Mn(II,II) species. That we observe just
five peaks with roughly equal intensity could be due to our only
seeing the central five features of the undecaplet with 4:5:6:4:5
the dissolved O in the starting mixture is exhausted before
2
complete Mn(II) oxidation. This kobs is comparable to that
found for the slower process causing loss of the Mn(II) EPR
signal in the presence of atmospheric O (k = 0.019 ± 0.001
43,44
intensity, and the other features are simply not resolved.
2
2obs
5
5
−1
Similar Mn hyperfine patterns are also observed from
s ).
Mn(II,II) species in manganese-bound DNA-binding protein
from nutrient starved cells. For the reactions of Mn(II) with
as-isolated, oxidized Mnx that were performed in the presence
We also looked for EPR spectroscopic signatures of higher
oxidation states of manganese in either mononuclear or
multinuclear species. At no point did we observe parallel-
45
E
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
mode EPR signals that would possibly correspond to a
mononuclear S = 2 Mn(III) species (data not shown).
The samples with substoichiometric manganese0.2 equiv
(red trace, Figure 4) and 0.8 (green trace, Figure 4) equiv
give rise to EPR signals centered at approximately 160 mT
ascribed to class ii and class iii Mn(II) species. The spectra of
samples made with 4 equiv of substrate or more (blue and black
traces, Figure 4) possess two additional narrow peaks (ca. 165.0
and 174.0 mT). They could be part of an undecaplet arising
from another two weakly exchange-coupled Mn(II) ions whose
binding sites are populated only at these higher Mn:Mnx ratios
(4:1 and 6:1). This could, to some extent, confirm our
hypothesis that multiple Mn(II) binding sites exist from nearby
to the T1 copper site to the surface of Mnx complex. Three sets
of undecaplets with varying effective g values arising from
different Mn(II)−Mn(II) species in manganese-bound DNA-
binding protein from nutrient starved cells are reported by
52−54
However, the zero-field splitting parameters of a Mn(III) spin
system could be too large or too heterogeneous to see the
corresponding signals. Also, Mn(III) is a potent oxidant and
disproportionates rapidly unless strongly ligated to a compound
55−57
such as pyrophosphate.
We also saw no evidence of
multiline signals that would result from two antiferromagneti-
cally coupled Mn ions giving rise to a net S = 1/2 electron spin,
5
8
44,59
e.g., a Mn(II,III) or a Mn(III,IV) species.
As for possible
mononuclear Mn(IV) complexes, again we observe no obvious
signals in the g ≈ 4.0 region that would be diagnostic of such S
=
3/2 species (see above discussion). However, we know that a
Mn(IV)-containing species is ultimately produced and
incorporated into the product, solid manganese oxide MnOx,
although it is not clear how long this species exists as a
monomer (if at all) since the Mn(IV) oligomerizes extensively
at all but very low pH conditions. Experiments in our
laboratories will continue to target these putative higher-
45
Hayden et al. Alternatively, the peaks at 165.0 and 174.0 mT
could arise from just a different EPR transition of the class iii
60
41,44,62−65
dimer
and happen to be evident only at high
concentrations of Mn(II) relative to the Mnx complex.
Preliminary EPR spectral simulations are provided in Figure
S4 to illustrate the sensitivity of the spectra of these spin
systems to modest changes in various spin Hamiltonian
parameters.
oxidation-state intermediates that lead to MnO (s) generation
x
from Mn(II).
3
.3. EPR Characterization of Mn(II) Bound To Mnx.
Fully reduced Mnx protein that was anaerobically incubated
with 0.8 equiv of Mn(II) yields the CW EPR spectrum
The temperature dependence of the EPR spectral features for
the class ii and iii Mn(II)-containing species is presented in
Figure 5. The intensity of the half-field transitions assigned to
the class ii Mn(II) center (red filled diamonds, Figure 5B)
increases with increasing temperature (over the temperature
range explored, 4.5−60 K) until approximately 20 K at which
time it then behave according to Curie law, indicating that this
feature arises from EPR transitions between spin levels of an
(“RedMnx + 0.8 Mn(II)”) shown in Figure 4 (green traces).
6
6
energetically excited spin manifold. This behavior is
consistent with an S = 5/2 spin center experiencing a modest
positively signed ZFS interaction (D > 1000 MHz) suggesting
32
pseudo-octahedrally coordinated Mn(II) ion. For the signals
assigned to the binuclear Mn(II,II) site (blue filled circles,
Figure 5B), as the temperature is increased to 40 K, the
intensity of the class iii multiplet increases, indicating that the
donor spin levels operative in the EPR transition become more
3
3,41
populated.
This behavior is consistent with weak
antiferromagnetic coupling between the two ions, i.e., the
isotropic-exchange-coupling term J is negative in the Heisen-
berg−Dirac−van Vleck Hamiltonian (H = −2J S S ). Increasing
1
2
the temperature further leads to a decrease in the intensity of
the class iii signal at 135 mT, suggesting that this multiplet
could arise from low-lying spin manifold which is depopulated
at higher temperatures.
We further interrogated the coordination environment of the
class ii Mn(II) center using ESEEM spectroscopy. The X-band
Figure 4. X-band CW EPR spectra (normalized to total [Mn]) of
RedMnx protein incubated anaerobically with varying ratios (0.2, 0.8,
4
, and 6) of MnCl . Inset shows the low field spectral region around g
2
=
4.3 (B = 155 mT). Experimental parameters: temperature = 15 K;
0
microwave frequency = 9.37 GHz; microwave power = 0.6325 mW;
conversion time = 40 ms; modulation amplitude = 0.8 mT;
modulation frequency = 100 kHz.
(
9.47 GHz) three-pulse ESEEM spectrum of RedMnx treated
with 0.8 equiv of Mn(II) is shown in Figure 6A (black trace;
corresponding echo-detected EPR spectrum is shown in Figure
S5). These data were collected in resonance with the lowest-
field member of the central sextet (314.0 mT). The
corresponding Fourier transform is given in Figure 6B and
shows a set of sharp peaks at 0.4, 1.4, and 2.0 MHz as well as a
3 MHz broad feature centered at 4.4 MHz. These features are
Again, no Cu(II) signals are evident, confirming that the Mnx
complex is completely reduced. Even at this relatively low
manganese loading, all three classes of Mn(II) signals described
above are seen in the spectrum: namely, those that are
diagnostic of a (i) mononuclear, likely aqueous, Mn(II) species
experiencing a weak ZFS interaction; (ii) a mononuclear
Mn(II) species with a modest ZFS interaction; and (iii) a
weakly exchange-coupled Mn(II,II) dimeric species. There is
also a small amount of contaminating Fe(III) species giving rise
assigned, respectively, to the ν , ν , ν nuclear quadrupole
0
−
+
transitions and to ν , a double quantum transition, of a
dq
14
hyperfine-coupled N-containing (I = 1) ligand to manga-
6
7,68
1
nese.
The peak at 13.4 MHz (≈ Larmor frequency of H at
314.0 mT) is attributed to weakly hyperfine-coupled protons.
61
to the signal at g ≈ 4.3.
Similar spectra have been observed for several Mn(II)-
F
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. (A) Temperature dependence of EPR signals that appear around g = 4.3 (B = 155 mT) for RedMnx mixed with 0.2 equiv Mn(II).
0
Experimental parameters: microwave frequency = 9.37 GHz; conversion time = 100 ms; modulation amplitude = 0.8 mT; modulation frequency =
1
00 kHz. (B) Temperature dependence of the average peak-to-peak amplitudes of the EPR features centered at 128.3, 136.0, and 141.0 mT (blue
circles) and the EPR features centered at 155.3, 181.6, and 190.0 mT (red diamonds) that correspond to the class iii and class ii Mn(II) species,
respectively.
Figure 6. (A) Three-pulse ESEEM spectra of RedMnx mixed with 0.8 equiv Mn(II) (black), Mn(II) bound to BSA (red), and Mn(II) bound to CP
(
blue). (B) Corresponding cross-term averaged Fourier transforms of data in A. Experimental parameters (top to bottom): temperature = 10 K,
microwave frequency = 9.47 GHz, B = 314.0 mT, π/2 = 16 ns, τ = 140 ns, T = 80 ns; temperature = 10 K, microwave frequency = 9.37 GHz, B =
0
0
3
1
14.0 mT, π/2 = 16 ns, τ = 140 ns, T = 80 ns; temperature = 10 K, microwave frequency = 9.55 GHz, B = 342.5 mT, π/2 = 8 ns, τ = 100 ns, T =
36 ns. Sixth root of Mn(II)-CP ESEEM time domain spectrum (blue dashed line) and fourth root of Mn(II)-BSA ESEEM time domain spectrum
0
(
red dashed line) are shown in A for comparison.
6
7−69
containing proteins,
in particular, concanavalin A (ConA),
Mnx. The two systems we chose for this comparison are
71
which is known to have a single histidine ligand to the metal
Mn(II) bound to BSA and Mn(II) bound to calprotectin
7
0
72,73
center.
(CP).
BSA contains a typical amino-terminal Cu(II)- and
In our case, we do not know the number of nitrogenous
ligands and cannot determine this information from the Fourier
transformed ESEEM spectrum (Figure 6B). Instead, by
analyzing the depth of the N modulations in the time-domain
spectrum (Figure 6A) and comparing this depth to similar
ESEEM traces for mononuclear Mn(II) bound to a known
number of nitrogen ligands, we can roughly estimate the
number of nitrogens bound to the class ii Mn(II) center in
Ni(II)-binding motif (ATCUN) that consists of one α-amino
nitrogen, two peptide nitrogens, and one imidazole nitrogen of
7
1,74−76
histidine, four metal-binding nitrogen ligands in all.
This
and Ni(II)
1
4
77,78
ATCUN motif binds Cu(II) (log K ≈12.0)
f
79
(log K ≈ 10.0) with high affinity and with somewhat lower
f
80,81
affinity for Mn(II) (log K ≈4.0).
CP is a manganese
f
sequestration protein and in the presence of 40 equiv Ca(II)
binds Mn(II) tightly (K = 194 ± 203 nM). X-ray crystal
72
d
G
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 7. Three-pulse ESEEM spectra of RedMnx protein incubated with 0.2 (red traces), 0.8 (green traces), and 6 (blue traces) equiv Mn(II): (A)
Raw time domain spectra; (B) same data as in A after subtracting a decaying exponential function to account for electron spin relaxation and
unmodulated ESEEM from aqueous Mn(II). Experimental parameters: temperature = 10 K; microwave frequency = 9.53 GHz, B = 363.0 mT, π/2
0
=
16 ns, τ = 128 ns, T = 80 ns.
Figure 8. (A) Simulation (black traces) of three-pulse ESEEM spectrum (blue traces) of RedMnx mixed with 0.8 equiv Mn(II) and corresponding
14
cross-term averaged Fourier transform. (B) Simulation parameters: one N nucleus was employed with Aiso = +2.38 ± 0.01 MHz, Aaniso = [0.01
2
0
.20−0.21] MHz, e Qq/h = 2.33 ± 0.01 MHz, η = 0.22 ± 0.01, and Euler angles for the NQI frame relative to the principal axes of hyperfine tensor
are α = 0 ± 5°, β = 90 ± 5°, γ = 0 ± 5°. Experimental parameters: temperature = 10 K, microwave frequency = 9.53 GHz, B = 327.3 mT, π/2 = 12
ns, τ = 144 ns, T = 80 ns.
0
structures of CP show Mn(II) is bound by six histidine side
hyperfine tensor and nuclear quadrupole interactions (collec-
7
3
chains. The ESEEM traces of substoichiometric Mn(II)
bound to each of these two protein systems are shown in
Figure 6A,B (time and frequency domain, respectively) and
tively, this branching is described by the so-called Mims
84
matrix). More specifically, for three-pulse ESEEM spectros-
copy, the modulation observed in the time-domain spectrum is
82,83
α
β
compared to those for RedMnx + 0.8 equiv Mn(II).
given by V (τ,T) ∝ ∑ (∏ V (q,θ) + ∏ V (q,θ)). The term
3
3p
θ
q
3p
q 3p
α(β)
Qualitatively, one can see that the modulation depth evident in
the ESEEM spectrum of Mn(II)-BSA (red trace) is deeper than
that for Mn(II)-Mnx (black trace), and the modulation depth
measured for Mn(II)-CP (blue trace) is deeper still. This trend
suggests that Mnx provides less than four nitrogen ligands to
the class ii Mn(II) center.
In general, the depth of ESEEM modulation is governed the
number of hyperfine-coupled nuclei and the facility of nuclear
spin level “branching” due to the nonsecular contribution to the
V
(q,θ) describes the contribution to the modulation by
p
nucleus q with orientation θ relative to the external static
magnetic field. The modulations from all like-oriented nuclei
contributing via the same coherence transfer pathway (α or β)
are multiplied according to the product rule; then the products
from each coherence transfer pathway are added. Finally, this
result is summed over all possible orientations to give the total
observed echo envelope modulation. In practice, the previous
equation essentially provides a “product rule” description of the
H
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Measured HFI and NQI Parameters of 14N Nuclei Coupled to Mn(II)
species
14N
imino
Aiso (MHz)
Aaniso (MHz)
e2Qq/h (MHz)
η
ref
imidazole
−
−
−
−
2.30
3.00
2.10
2.50
3.19
2.38
−
−
−
−
−
3.3
0.13
0.99
0.13
0.92
0.40
0.30
0.40
0−1
87
87
87
87
67
68
68
88
89
amino
imino
amino
histidine
guanosine
guanosine
−
1.43
L-histidine
3.36
1.44
Mn(II)-RNA
2.90
a
Mn(II)GMP
0.65
3.20
0.40
2.90
Mn(II)ADP
Mn(II)Im6
Mn(II)-Mnx
−
2.60−3.10
3.0
−
n.a.
0.63
0.1−0.3
0.22 ± 0.01
[0.01 0.20 −0.21]
2.33 ± 0.01
this work
a
Mn(II)GMP =: guanosine 5′-monophosphate.
Figure 9. D O/H O ratioed ESEEM spectrum of RedMnx mixed with 0.8 equiv Mn(II) (red traces) measured at two resonance field positions (A)
2
2
2
+
3
27.3 and (B) 363.0 mT. From this residual deuterium modulation, 20% of the residual deuterium modulation from [Mn(D/H O) ] was
2 6
subtracted to account for the small amount of class i, aqueous Mn(II) that contributes to the spectrum. This resultant trace is compared to that
obtained for a various number of water ligands to the Mn(II) center (black traces) that were calculated from the D O/H O ratioed ESEEM spectra
2
2
2+
of [Mn(D/H O) ] (Figure S1D) as described in the Experimental Procedures section.
2
6
observed ESEEM modulation, whereby the modulation
significant increase in the exponential decay of the three-pulse
ESEEM. However, fitting and subtracting this exponential
function from the data yields modulation patterns that are
nearly identically deep (Figure 7B; corresponding Fourier
transforms of these data can be found in Figure S6). This
behavior indicates that two classes of Mn(II) species contribute
to the ESEEM spectra: one is the aqueous/adventitiously
patterns for each isolated hyperfine-coupled nucleus can be
85
multiplied together yielding the net modulation pattern. For
multiple nuclei whose nuclear quadrupole and hyperfine
interaction parameters are similar, e.g., the six histidine
nitrogens coordinating the manganese center in CP are
83
magnetically identical, the experimental modulation pattern
can be thought of as the ESEEM spectrum of one of these
nitrogens raised to the sixth power.
14
bound Mn(II) which possesses no N ligands and thus
contributes only to the exponential decay of the modulation
pattern and one which is specifically bound to Mnx by just one
Therefore, by taking the fourth and sixth roots of ESEEM
14
spectra of Mn(II) bound to BSA and CP, respectively, we can
N ligand as described above.
1
4
visualize the modulation depth of a single N ligand to
manganese (see dashed red and blue traces, Figure 6A). We
find that the arithmetically scaled spectra of Mn(II)-CP and
Mn(II)-BSA give a nearly identical modulation depth to that
observed for Mn(II) bound to Mnx spectrum (black solid line)
The ESEEM spectrum of RedMnx with 0.8 equiv of Mn(II)
1
4
is well-simulated assuming a single hyperfine-coupled N atom
(based on above analysis) to the metal center (blue traces in
Figure 8). The 14N magnetic parameters determined from the
fit are as follows: Aiso = +2.38 ± 0.01 MHz and Aaniso = [0.01,
1
4
2
indicating that just one N-containing ligand in Mnx binds
0.20, − 0.21] MHz with nuclear quadrupole parameters e Qq/h
8
6
Mn(II).
= 2.35 ± 0.01 MHz and η = 0.22 ± 0.01. Euler angle for
quadrupole tensor relative to hyperfine tensor is α = 0 ± 5°, β =
To confirm that this 14N modulation is from a specifically
bound and not an adventitiously bound Mn(II) center, we
conducted ESEEM studies on samples of RedMnx containing
−Γt
90 ± 5°, γ = 0 ± 5°. An additional damping function (e , Γ ≈
−1
0.11 μs ) was applied to the simulation to model the loss of
nuclear coherence due to relaxation. Both hyperfine and nuclear
quadrupole parameters are in the range of reported values for
0.2, 0.8, and 4.0 equiv of Mn(II) (Figure 7A). As the
concentration of Mn(II) relative to Mnx is increased, there is a
I
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
other 14N-coordinated Mn(II) systems (Table 1).67,68,87−89
The magnitude of the hyperfine interaction measured for the
Mnx-bound Mn(II) center (class (ii) is similar to the value for
analogous systems with a histidine imidazole serving as a ligand
to manganese (cf. Table 1), indicating the 1 N-containing
ligand could be a imidazole side chain of an active site histidine.
The number of water ligands to this Mnx-bound Mn(II)
center was determined using three-pulse ESEEM spectroscopy
performed at 327.3 and 363.0 mT as described in the
Experimental Procedures section. The ratioed D O/H O
guide (E° = +1.5 V vs NHE). These values, of course,
correspond to conditions much different than those in effect in
the MCO. Alternatively, Mnx could also avoid the reactive and
potentially deleterious Mn(III) ion and proceed directly to
4
MnO (s) from Mn(II) (E° = +1.23 V vs NHE), however, this
2
reaction also requires a stronger oxidizing potential from the T1
copper than is likely available. Therefore, the binding site of the
substrate must aid in the tuning the redox potential of the
Mn(II) ion into an accessible range much as the trigonal
52,56,92
bipyramidal active site of Mn-superoxide dismutase does.
2
2
spectrum that contains only modulation from exchangeable
deuterons that are hyperfine-coupled to spin centers resonant
at g ≈ 2 is presented in Figure S1A). In this sample, those spin
centers include the both the class i and class ii mononuclear
Mn(II) species described above. While the sextet observed ca.
Our EPR spectroscopic studies of fully reduced Mnx treated
anaerobically with Mn(II) reveal at least two classes of
mononuclear Mn(II)-containing species. One is simply the
hexa-aquo Mn(II) substrate that gives rise to the sextet
centered at g = 2. The other mononuclear Mn(II) species is
identified by the appearance of a sextet of transitions at “half-
field” (g ≈ 4.0) that are split by 9.0 mT. Such features are
typically diagnostic of a Mn(II) center experiencing modest
3
30 mT in the X-band CW EPR spectrum arises solely from the
aqueous Mn(II) in the sample “RedMnx + 0.8Mn(II)”, the
echo-detected EPR spectrum possess significant intensity from
both types of manganese sites. This is due to the larger ZFS for
−1
26,27
ZFS interactions (D < 0.1 cm ),
indicating that the
the class ii site broadening the central field transitions (Δm =
substrate is bound by the Mnx complex with five or six ligands
(more likely six). Our ESEEM spectroscopic investigations help
to more fully characterize the coordination environment of this
bound Mn(II) center. We detect modulations from a single
nitrogenous ligand (Figure 6). Based on a comparison between
S
±
1), allowing them to go undetected in the field-modulated
CW experiment (see preliminary simulation in Figure S3).
In the ratioed D O/H O ESEEM spectrum, we have
2
2
accounted for the amount of modulation from class i, aqueous
Mn(II) in the following way: Using the [Mn(II)] vs EPR signal
intensity standard curve described in the Experimental
Procedures section, we found that 20% of the total Mn in
the sample was of the class i, aqueous Mn(II) type. Preliminary
simulations of the CW EPR spectrum achieved using two
components (aqueous Mn(II) and bound Mn(II) with ratio of
14
the N magnetic parameters for the Mnx-bound Mn(II)
derived from simulation of the ESEEM spectra (Figure 8) and
14
those from other studies of Mn(II)− N-containing spin
systems (cf. values in Table 1), we assign this nitrogen as
coming from the imidazole side chain of an active site histidine.
Based on our analysis of the ratioed D O/H O ESEEM data
2
2
0.2:0.8) are provided in Figure S4B. We then subtracted a 20%
of Mnx-bound Mn(II) (Figure 9), we hypothesize that there
are two coordinated water molecules in addition to the
histidine ligand. This coordination environment is strikingly
similar to that reported for the Mn(II) center in ConA; the X-
ray crystal structure and EPR data show that there are one
of the residual deuterium modulation for aqueous Mn(II) from
that for the “RedMnx + 0.8Mn(II)” sample yielding the red
traces in Figure 9. This trace is compared to hypothetical
modulation patterns for 1 through 6 water ligands obtained by
taking the N/6 (N = 1−6) root of the D O/H O ratioed
2
2
histidine and two water ligands as well as three O donors from
2
+
93,94
ESEEM spectrum of [Mn(D O) ] (black traces in Figure 9).
2
6
proteinacious carboxylate side chains.
Similarly the MCO
The residual deuterium modulation depth evident in the
ratioed spectrum of Mnx-bound Mn(II) corresponds most
closely to that for two water ligands (≈1.6 at 327.3 mT and
hCp binds its substrate Fe(II) with one histidine and three
15
carboxylate ligands. These parallels lead us to propose the
Mnx binding environment for the mononuclear Mn(II) center
that is depicted in Figure 10. This manganese binding
environment is retained for mutant H340A, which gives rise
to an almost identical ESEEM spectrum as WT Mnx (Figure
S7).
≈1.9 at 363.0 mT).
4
. DISCUSSION
The Bacillus exosporium multicopper oxidase (MCO) Mnx
protein complex plays a central role in manganese biominer-
alization by catalyzing Mn(II) oxidation to form solid
1
1,14,57
Mn(IV)Ox.
While several MCOs that possess metallo-
oxidase activity have been characterized previously, all of these
systems perform merely one-electron chemistry. Thus, the two-
electron oxidation of Mn(II) to Mn(IV) catalyzed by Mnx is
mechanistically intriguing.
This process begins with Mn(II) binding to the Mnx
complex, presumably in the vicinity of the T1 copper site, in
analogous fashion to Fe(II) binding to ceruloplasm (hCp).
Results from previous studies of MCOs showed that the
reduction potential of the T1 copper site can vary widely from
Figure 10. Proposed binding environment of the class ii Mn(II) center
bound to the Mnx protein complex. Red colored atoms represent
those that were detected from ESEEM measurements. Gray colored
atoms are assumed by analogy to hCp and ConA.
15
4
00 < E < 1000 mV vs NHE and that this variation is usually
h
13,90,91
influenced by the nature of the T1 Cu axial ligand.
While
A third Mn(II) signal was observed in the EPR spectra of
reduced Mnx + Mn(II) that consisted of at least five but
perhaps 11 peaks split by a 55Mn hyperfine interaction of
approximately 4.5 mT (inset Figure 3). This magnitude of the
hyperfine splitting of this feature is very diagnostic of a weakly
exchange-coupled Mn(II,II) dimer. This signal is also apparent
we do not yet know the redox potential of the T1 Cu site in
Mnx, if it were as high as +1000 mV, the largest known value
for a multicopper oxidase, this would still be insufficient
electromotive force to oxidize an aqueous Mn(II) ion to the 3+
oxidation state if the standard-state potentials are used as a
J
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 11. Proposed mechanism for Mn(II) binding and oxidation by the MCO in the Mnx protein complex aerobically.
4
4,59
in spectra of oxidized Mnx mixed with Mn(II) and then rapidly
frozen (Figure 1) but requires some time to develop (>10 s).
The Mn(II,II) dimer then appears to be oxidized as the signal
dimer
that would yield spin system with an S = 1/2
55
ground electron spin state and give rise to Mn hyperfine-split
resonances centered at g = 2. However, our spectroscopic
observation of a Mn(II,II) dimer that disappears in the
disappears after 70 s of reaction with O . While no mixed-
2
59
95
valence dimers were observed, if Mn(III,III) or Mn(IV,IV)
presence of excess O suggests that higher oxidation-state
2
dimers were generated, their corresponding EPR spectra are
notoriously broad and ill-resolved; thus, it is not surprising that
they are non-obvious if present.
manganese dimeric species are possible.
Furthermore, it is reasonable that no mononuclear Mn(III)
or Mn(IV) species ever accumulate as both are potent oxidants
which need to be stabilized by strong donor ligands.
Instead, polynuclear Mn(III) or Mn(IV) complexes could be
3
6−40,96,97
The X-band CW EPR spectra show that the aqueous Mn(II)
concentration decreases with reaction time via pseudo-first-
order kinetics and possibly by two distinct pathways, having
stabilized by oxido-bridging similar to what is prevalent in
−
1
98−100
rate coefficients of: k
±
= 0.205 ± 0.001 s and k2obv = 0.019
proteins, such as photosystem II
catalases.
and manganese
1
obv
−
1
101,102
0.001 s . Previous investigations using either the
exosporium from Bacillus sp. SG-1 spores
Therefore, we tentatively expect oxidized
11,57
or purified
manganese species in the Mnx protein to oligomerize quickly,
giving rise to homovalent dimers (or higher order multimers).
Given all these data, we propose a possible mechanism for
the oxidation of Mn(II) by the Mnx enzyme similar to (or
consistent with) that proposed by Soldatova et al. (Figure
14
Mnx protein complex indicate that Mn(III) is generated
transiently as the complex Mn(III)pyrophosphate (PP) was
observed when enzymatic manganese oxidation was performed
in the presence of this strong Mn(III) chelator. These results
suggest that Mn(II) oxidation involves two discrete one-
electron oxidation steps to generate the Mn(IV) oxide.
However, it should be noted that the presence of PP retards
11
11). The mononuclear Mnx-bound Mn(II) is assumed to be
adjacent to the T1 copper site (again, by analogy to several
other MCOs). When the MCO is fully oxidized, this Mn(II)
site gets rapidly oxidized to the 3+ state. This resultant Mn(III)
could then abstract electrons from the nearby binuclear Mn(II)
complex, driven by the formation of hydroxido or oxido-bridges
11
Mn(IV) oxide formation by a factor of >100.
Under all of the conditions we employed (notably, always in
the absence of PP), we never detected any higher oxidation-
state manganese complexes. These included mononuclear
Mn(III) or Mn(IV) species which should give rise to diagnostic
features in parallel- and perpendicular-polarized EPR spectra,
respectively. We also saw no evidence for the generation of
antiferromagnetically coupled mixed-valence manganese
1
1
from coordinated water molecules. In this way, the
mononuclear Mn(II) center serves as an electron shuttle
from the binuclear site to the coppers of the MCO. Upon
completion of four electron transfers, the nascent [Mn-
2+
(IV) O ] unit will be formed. However, it is difficult to
2
2
5
8
dimerseither a Mn(II,III) dimer
or a Mn(III,IV)
imagine that this happens as described, as solid MnO would
2
K
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
soon clog the substrate channel into the MCO active site.
Therefore, it is interesting to think about extending this model
to include multiple Mn(II) binding sites in Mnx that represent
a so-called manganese wire. At the termination of this wire, in
the extracellular space, the resultant Mn O synthon can
(13) Solomon, E. I.; Szilagyi, R. K.; George, S. D.; Basumallick, L.
Chem. Rev. 2004, 104, 419.
(
14) Butterfield, C. N.; Soldatova, A. V.; Lee, S. W.; Spiro, T. G.;
Tebo, B. M. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 11731.
15) Lindley, P. F.; Card, G.; Zaitseva, I.; Zaitsev, V.; Reinhammar,
B.; Selin-Lindgren, E.; Yoshida, K. J. Biol. Inorg. Chem. 1997, 2, 454.
16) Machonkin, T. E.; Solomon, E. I. J. Am. Chem. Soc. 2000, 122,
2547.
17) Heppner, D. E.; Kjaergaard, C. H.; Solomon, E. I. J. Am. Chem.
(
x
2y
migrate and nucleate at a nearby site to begin synthesis of the
(
1
(
extended solid MnO . This mechanism is rather similar to that
2
employed by the iron storage protein ferritin in which Fe(II) is
oxidized in binuclear sites to form a dinuclear Fe(III) species
which is then transferred along a protein channel to the
Soc. 2013, 135, 12212.
(18) Lee, S. K.; George, S. D.; Antholine, W. E.; Hedman, B.;
Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2002, 124, 6180.
1
03−105
nucleation sites.
(19) Solomon, E. I.; Augustine, A. J.; Yoon, J. Dalton Trans. 2008, 30,
3
(
(
921.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b04331.
■
20) Schmidt, T. G.; Skerra, A. Nat. Protoc. 2007, 2, 1528.
21) Butterfield, C., Ph.D., Oregon Health & Science University,
*
S
Portland, OR, 2014.
(22) Durao, P.; Chen, Z.; Fernandes, A. T.; Hildebrandt, P.; Murgida,
D. H.; Todorovic, S.; Pereira, M. M.; Melo, E. P.; Martins, L. O. J. Biol.
Inorg. Chem. 2008, 13, 183.
(
̃
Standard spectra for water-counting via three-pulse
ESEEM; electronic absorption spectrum and CW EPR
data and simulations of H340A Mnx; echo-detected X-
band field swept EPR spectrum of RedMnx + 0.8
Mn(II); and Fourier transform of three-pulse ESEEM
data from Figure 7B (PDF)
23) Vallejo, A. N.; Pogulis, R. J.; Pease, L. R. CSH Protoc. 2008
1
0.1101/pdb.prot4861.
(
(
24) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42.
25) Stoll, S.; Britt, R. D. Phys. Chem. Chem. Phys. 2009, 11, 6614.
(26) Reed, G. H.; Markham, G. D. Bio. Magn. Reson. 1984, 6, 73.
27) Stich, T. A.; Lahiri, S.; Yeagle, G.; Dicus, M.; Brynda, M.; Gunn,
A.; Aznar, C.; DeRose, V. J.; Britt, R. D. Appl. Magn. Reson. 2007, 31,
21.
(
AUTHOR INFORMATION
Corresponding Author
rdbritt@ucdavis.edu
■
*
3
(
28) Hoogstraten, C. G.; Britt, R. D. RNA 2002, 8, 252.
(29) Mims, W. B.; Peisach, J. Biol. Magn. Reson. 1981, 3, 213.
(
30) Wijma, H. J.; Boulanger, M. J.; Molon, A.; Fittipaldi, M.; Huber,
M.; Murphy, M. E. P.; Verbeet, M. P.; Canters, G. W. Biochemistry
003, 42, 4075.
31) Prudencio, M.; Sawers, G.; Fairhurst, S. A.; Yousafzai, F. K.;
Notes
The authors declare no competing financial interest.
2
(
ACKNOWLEDGMENTS
■
Eady, R. R. Biochemistry 2002, 41, 3430.
(32) Hayden, J. A.; Farquhar, E. R.; Que, L.; Lipscomb, J. D.;
Hendrich, M. P. J. Biol. Inorg. Chem. 2013, 18, 717.
We thank Prof. Leone Spiccia of Monash University for useful
discussions. This work was supported by the National Science
Foundation: award numbers EAR-1231322 to WHC, CHE-
213699 to RDB, CHE-1410688 to BMT, and an NSF
Postdoctoral Research Fellowship in Biology Award ID: DBI-
202859 to C.A.R. The EPR spectrometers at the CalEPR
(
33) Pierce, B. S.; Elgren, T. E.; Hendrich, M. P. J. Am. Chem. Soc.
003, 125, 8748.
34) Machonkin, E. T.; Quintanar, L.; Palmer, A. E.; Hassett, R.;
Severance, S.; Kosman, D. J.; Solomon, E. I. J. Am. Chem. Soc. 2001,
23, 5507.
35) Quintanar, L.; Gebhard, M.; Wang, T. P.; Kosman, D. J.;
Solomon, E. I. J. Am. Chem. Soc. 2004, 126, 6579.
36) Rajendiran, T. M.; Kampf, J. W.; Pecoraro, V. L. Inorg. Chim.
Acta 2002, 339, 497.
37) Weyhermuller, T.; Paine, T. K.; Bothe, E.; Bill, E.; Chaudhuri, P.
Inorg. Chim. Acta 2002, 337, 344.
38) Paine, T. K.; Weyhermuller, T.; Bothe, E.; Wieghardt, K.;
Chaudhuri, P. Dalton Trans. 2003, 3136.
39) Parsell, T. H.; Behan, R. K.; Green, M. T.; Hendrich, M. P.;
Borovik, A. S. J. Am. Chem. Soc. 2006, 128, 8728.
40) Gupta, R.; Taguchi, T.; Borovik, A. S.; Hendrich, M. P. Inorg.
Chem. 2013, 52, 12568.
41) Mathur, P.; Crowder, M.; Dismukes, G. C. J. Am. Chem. Soc.
2
(
1
1
1
(
facility used in this study were funded by the National Institutes
of Health (S10-RR021075) and the NSF (CHE-1048671).
(
REFERENCES
■
(
(
1) Tebo, B. M.; Bargar, J. R.; Clement, B. G.; Dick, G. J.; Murray, K.
J.; Parker, D.; Verity, R.; Webb, S. M. Annu. Rev. Earth Planet. Sci.
(
̈
2
(
004, 32, 287.
2) Spiro, T. G.; Bargar, J. R.; Sposito, G.; Tebo, B. M. Acc. Chem.
Res. 2010, 43, 2.
3) Morgan, J. J. Geochim. Cosmochim. Acta 2005, 69, 35.
4) Nealson, K. H.; Tebo, B. M. Origins Life 1980, 10, 117.
5) Geszvain, K.; McCarthy, J. K.; Tebo, B. M. Appl. Environ.
(
(
(
(
(
(
Microbiol. 2013, 79, 357.
6) Ridge, J. P.; Lin, M.; Larsen, E. I.; Fegan, M.; McEwan, A. G.; Sly,
L. I. Environ. Microbiol. 2007, 9, 944.
7) Corstjens, P. L. A. M.; de Vrind, J. P. M.; Goosen, T.; Jong, E. W.
d. V. d. Geomicrobiol. J. 1997, 14, 91.
8) Francis, C. A.; Tebo, B. M. Appl. Environ. Microbiol. 2002, 68,
74.
9) Dick, G. J.; Torpey, J. W.; Beveridge, T. J.; Tebo, B. M. Appl.
Environ. Microbiol. 2008, 74, 1527.
10) Waasbergen, L. G.; Hildebrand, M.; Tebo, B. M. J. Bacteriol.
996, 178, 3517.
11) Soldatova, A. V.; Butterfield, C.; Oyerinde, O. F.; Tebo, B. M.;
Spiro, T. G. J. Biol. Inorg. Chem. 2012, 17, 1151.
12) Solomon, E. I.; Sundaram, U. M.; Machonkin, T. Chem. Rev.
996, 96, 2563.
(
1987, 109, 5227.
(42) Preliminary spectral simulations in support of this assignment
are provided in Figure S4. However, given that the only feature we
observe experimentally is multiplet at 136.0 mT, we caution the reader
from using the given spin Hamiltonian parameters in any meaningful
way. They are presented here merely to show that reasonable values
for these parameters can give rise to the observed spectral feature.
(43) Similar features have been observed for the Mn(II,II) form of
ribonucleotide reductase from E. coli.
(44) Cotruvo, J. A.; Stich, T. A.; Britt, R. D.; Stubbe, J. J. Am. Chem.
Soc. 2013, 135, 4027.
(45) Hayden, J. A.; Hendrich, M. P. J. Biol. Inorg. Chem. 2010, 15,
729.
(46) Reed, G. H.; Ray, W., Jr. Biochemistry 1971, 10, 3190.
(
(
8
(
(
1
(
(
1
L
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(
47) Haffner, P. H.; Goodsaid-Zalduondo, F.; Coleman, J. E. J. Biol.
Chem. 1974, 249, 6693.
48) Chang, C. H.; Svedruzic, D.; Ozarowski, A.; Walker, L.; Yeagle,
G.; Britt, R. D.; Angerhofer, A.; Richards, N. G. J. Biol. Chem. 2004,
79, 52840.
(83) Gagnon, D. M.; Brophy, M. B.; Bowman, S. E. J.; Stich, T. A.;
Drennan, C. L.; Britt, R. D.; Nolan, E. M. J. Am. Chem. Soc. 2015, 137,
3004.
(
(
(
84) Mims, W. B. Phys. Rev. B 1972, 5, 2409.
85) Schweiger, A.; Jeschke, G. Principles of Pulse Electron
2
Paramagnetic Resonance; Oxford University Press: New York, 2001.
(
49) Reed, G. H.; Morgan, S. D. Biochemistry 1974, 13, 3537.
(86) In particular, we are focusing on the first observed beat in the
(
50) Whiting, A. K.; Boldt, Y. R.; Hendrich, M. P.; Wackett, L. P.;
modulation pattern as the different nuclear relaxation behavior for
each Mn(II)-containing protein should have negligible effects on the
depth of this first modulation.
Que, L. Biochemistry 1996, 35, 160.
51) Burrell, M. R.; Just, V. J.; Bowater, L.; Fairhurst, S. A.; Requena,
L.; Lawson, D. M.; Bornemann, S. Biochemistry 2007, 46, 12327.
52) Sheng, Y.; Stich, T. A.; Barnese, K.; Gralla, E. B.; Cascio, D.;
Britt, R. D.; Cabelli, D. E.; Valentine, J. S. J. Am. Chem. Soc. 2011, 133,
0878.
53) Campbell, K. A.; Force, D. A.; Nixon, P. J.; Dole, F.; Diner, B.
A.; Britt, R. D. J. Am. Chem. Soc. 2000, 122, 3754.
54) Campbell, K. A.; Yikilmaz, E.; Grant, C. V.; Gregory, W.; Miller,
A.; Britt, R. D. J. Am. Chem. Soc. 1999, 121, 4714.
(
(
87) Ashby, C. I. H.; Cheng, C. P.; Brown, T. L. J. Am. Chem. Soc.
978, 100, 6057.
88) Astashkin, A. V.; Nesmelov, Y. E. J. Phys. Chem. B 2012, 116,
3655.
89) Garcia-Rubio, I.; Angerhofer, A.; Schweiger, A. J. Magn. Reson.
007, 184, 130.
90) Quintanar, L.; Stoj, C.; Taylor, A. M.; Hart, P. J.; Kosman, D. J.;
Solomon, E. I. Acc. Chem. Res. 2007, 40, 445.
91) Xu, F.; Berka, R. M.; Wahleithner, J. A.; Nelson, B. A.; Shuster, J.
R.; Brown, S. H.; Palmer, A. E.; Solomon, E. I. Biochim. Biophys. Acta
998, 334, 63.
92) Miller, A. F. FEBS Lett. 2012, 586, 585.
(
1
(
1
(
2
(
2
(
(
(
(
55) Davies, G. Coord. Chem. Rev. 1969, 4, 199.
56) Whittaker, J. W.; Whittaker, M. M. J. Am. Chem. Soc. 1991, 113,
(
5
528.
57) Webb, S. M.; Dick, G. J.; Bargar, J. R.; Tebo, B. M. Proc. Natl.
Acad. Sci. U. S. A. 2005, 102, 5558.
58) Teutloff, C.; Schafer, K. O.; Sinnecker, S.; Barynin, V.; Bittl, R.;
1
(
(
(93) Sanz-Aparicio, J.; Hermoso, J.; Grangeiro, T. B.; Calvete, J. J.;
Cavada, B. S. FEBS Lett. 1997, 405, 114.
(
Wieghardt, K.; Lendzian, F.; Lubitz, W. Magn. Reson. Chem. 2005, 43,
S51.
(94) Rozwarski, D. A.; Swami, B. M.; Brewer, C. F.; Sacchettini, J. C.
J. Biol. Chem. 1998, 273, 32818.
(
59) Retegan, M.; Collomb, M. N.; Neese, F.; Duboc, C. Phys. Chem.
Chem. Phys. 2013, 15, 223.
60) Baral, S.; Lume-Pereira, C.; Janata, E.; Henglein, A. J. Phys.
Chem. 1986, 90, 6025.
(95) Caudle, M. T.; Kampf, J. W.; Kirk, M. L.; Rasmussen, P. G.;
Pecoraro, V. L. J. Am. Chem. Soc. 1997, 119, 9297.
(96) Weiss, R.; Gold, A.; Trautwein, A. X.; Terner, J. In The porphyrin
(
handbook; Kadish, K. M., Smith, K. M., Guilard, R., Ed.; Academic:
Boston, 2002.
(
(
61) Rogers, P. A.; Ding, H. J. Biol. Chem. 2001, 276, 30980.
62) Epel, B.; Schafer, K. O.; Quentmeier, A.; Friedrich, C.; Lubitz,
(
(
97) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1988, 110, 8628.
98) Ferreira, K. N.; Iverson, T. M.; Maghlaoui, K.; Barber, J.; Iwata,
̈
W. J. Biol. Inorg. Chem. 2005, 10, 636.
S. Science 2004, 303, 1831.
99) Loll, B.; Kern, J.; Saenger, W.; Zouni, A.; Biesiadka, J. Nature
005, 438, 1040.
100) Yeagle, G. J.; Gilchrist, M. L.; McCarrick, R. M.; Britt, R. D.
Inorg. Chem. 2008, 47, 1803.
101) Barynin, V. V.; Whittaker, M. M.; Antonyuk, S. V.; Lamzin, V.
S.; Harrison, P. M.; Artymiuk, P. J.; Whittaker, J. W. Structure 2001, 9,
25.
102) Stich, T. A.; Whittaker, J. W.; Britt, R. D. J. Phys. Chem. B 2010,
14, 14178.
(
(
63) Golombek, A. P.; Hendrich, M. P. J. Magn. Reson. 2003, 165, 33.
64) Khangulov, S. V.; Barynin, V. V.; Voevodskaya, N. V.; Grebenko,
(
2
(
A. I. Biochim. Biophys. Acta 1990, 1020, 305.
(
(
65) Mathur, P.; Dismukes, G. C. J. Am. Chem. Soc. 1983, 105, 7093.
66) Golynskiy, M. V.; Gunderson, W. A.; Hendrich, M. P.; Cohen,
(
S. M. Biochemistry 2006, 45, 15359.
67) Morrissey, S. R.; Horton, T. E.; Grant, C. V.; Hoogstraten, C.
G.; Britt, R. D.; DeRose, V. J. J. Am. Chem. Soc. 1999, 121, 9215.
68) Hoogstraten, C. G.; Grant, C. V.; Horton, T. E.; DeRose, V. J.;
Britt, R. D. J. Am. Chem. Soc. 2002, 124, 834.
69) Espe, M. P.; Hosler, J. P.; Ferguson-Miller, S.; Babcock, G. T.;
McCracken, J. Biochemistry 1995, 34, 7593.
70) McCracken, J.; Peisach, J.; Bhattacharyya, L.; Brewer, F.
Biochemistry 1991, 30, 4486.
(
7
(
1
(
(103) Liu, X.; Theil, E. C. Acc. Chem. Res. 2005, 38, 167.
(
(104) Ha, Y.; Shi, D.; Small, G. W.; Theil, E. C.; Allewell, N. M. J.
Biol. Inorg. Chem. 1999, 4, 243.
105) Turano, P.; Lalli, D.; Felli, I. C.; Theil, E. C.; Bertini, I. Proc.
Natl. Acad. Sci. U. S. A. 2010, 107, 545.
(
(
(
(
71) Peters, T.; Blumenstock, F. A. J. Biol. Chem. 1967, 242, 1574.
72) Hayden, J. A.; Brophy, M. B.; Cunden, L. S.; Nolan, E. M. J. Am.
Chem. Soc. 2013, 135, 775.
73) Damo, S. M.; Kehl-Fie, T. E.; Sugitani, N.; Holt, M. E.; Rathi, S.;
(
Murphy, W. J.; Zhang, Y.; Betz, C.; Hench, L.; Fritz, G.; Skaar, E. P.;
Chazin, W. J. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 3841.
(74) Harford, C.; Sarkar, B. Acc. Chem. Res. 1997, 30, 123.
(75) Flanagan, H. L.; Singel, D. J. J. Chem. Phys. 1987, 87, 5606.
(76) Kozlowski, H.; Potocki, S.; Remelli, M.; Rowinska-Zyrek, M.;
Valensin, D. Coord. Chem. Rev. 2013, 257, 2625.
77) Sankararamakrishnan, R.; Verma, S.; Kumar, S. Proteins: Struct.,
Funct., Genet. 2005, 58, 211.
78) Rozga, M.; Sokolowska, M.; Protas, A. M.; Bal, W. J. Biol. Inorg.
Chem. 2007, 12, 913.
79) Masuoka, J.; Hegenauer, J.; Dyke, B. R.; Saltman, P. J. Biol.
Chem. 1993, 268, 21533.
(
(
(
(
80) Mildvan, A. S.; Cohn, M. Biochemistry 1963, 2, 910.
(
81) Liang, H.; Tu, C.; Zhang, H.; Shen, X.; Zhou, Y.; Shen, P. Chin.
J. Chem. 2000, 18, 35.
82) The data for CP-bound Mn(II) are reproduced from a recently
published EPR study.
(
M
DOI: 10.1021/jacs.5b04331
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX