Dipeptide-based models of nickel superoxide dismutase: Solvent effects highlight a critical role to Ni-S bonding and active site stabilization

Article abstract of DOI:10.1021/ic2016462

Nickel superoxide dismutase (Ni-SOD) catalyzes the disproportionation of the superoxide radical to O₂ and H₂O₂ utilizing the Ni(III/II) redox couple. The Ni center in Ni-SOD resides in an unusual coordination environment that is distinct from other SODs. In the reduced state (Ni-SOD_red), Ni(II) is ligated to a primary amine-N from His1, anionic carboxamido-N/thiolato-S from Cys2, and a second thiolato-S from Cys6 to complete a NiN₂S₂ square-planar coordination motif. Utilizing the dipeptide N₂S^2- ligand, H ₂N-Gly-l-Cys-OMe (GC-OMeH₂), an accurate model of the structural and electronic contributions provided by His1 and Cys2 in Ni-SOD _red, we constructed the dinuclear sulfur-bridged metallosynthon, [Ni₂(GC-OMe)₂] (1). From 1 we prepared the following monomeric Ni(II)-N₂S₂ complexes: K[Ni(GC-OMe)(SC ₆H₄-p-Cl)] (2), K[Ni(GC-OMe)(S^tBu)] (3), K[Ni(GC-OMe)(SC₆H₄-p-OMe)] (4), and K[Ni(GC-OMe)(SNAc)] (5). The design strategy in utilizing GC-OMe^2- is analogous to one which we reported before (see Inorg. Chem.2009, 48, 5620 and Inorg. Chem. 2010, 49, 7080) where Ni-SOD_red active site mimics can be assembled at will with electronically variant RS^- ligands. Discussed herein is our initial account pertaining to the aqueous behavior of isolable, small-molecule Ni-SOD model complexes (non-maquette based). Spectroscopic (FTIR, UV-vis, ESI-MS, XAS) and electrochemical (CV) measurements suggest that 2-5 successfully simulate many of the electronic features of Ni-SOD_red. Furthermore, the aqueous studies reveal a dynamic behavior with regard to RS^- lability and bridging interactions, suggesting a stabilizing role brought about by the protein architecture.

Full text of DOI:10.1021/ic2016462

ARTICLE
pubs.acs.org/IC
Dipeptide-Based Models of Nickel Superoxide Dismutase: Solvent
Effects Highlight a Critical Role to NiꢀS Bonding and Active Site
Stabilization
Eric M. Gale, Darin M. Cowart, Robert A. Scott, and Todd C. Harrop*
Department of Chemistry, The University of Georgia, 1001 Cedar Street, Athens, Georgia 30602, United States
S Supporting Information
b
ABSTRACT: Nickel superoxide dismutase (NiꢀSOD) cata-
lyzes the disproportionation of the superoxide radical to O2 and
H2O2 utilizing the Ni(III/II) redox couple. The Ni center in
NiꢀSOD resides in an unusual coordination environment that
is distinct from other SODs. In the reduced state (NiꢀSODred),
Ni(II) is ligated to a primary amine-N from His1, anionic
carboxamido-N/thiolato-S from Cys2, and a second thiolato-S
from Cys6 to complete a NiN2S2 square-planar coordination
motif. Utilizing the dipeptide N2S2ꢀ ligand, H2N-Gly-L-Cys-
OMe (GC-OMeH2), an accurate model of the structural and
electronic contributions provided by His1 and Cys2 in Niꢀ
SODred, we constructed the dinuclear sulfur-bridged metallosyn-
thon, [Ni2(GC-OMe)2] (1). From 1 we prepared the following monomeric Ni(II)ꢀN2S2 complexes: K[Ni(GC-OMe)(SC6H4-p-
Cl)] (2), K[Ni(GC-OMe)(StBu)] (3), K[Ni(GC-OMe)(SC6H4-p-OMe)] (4), and K[Ni(GC-OMe)(SNAc)] (5). The design
strategy in utilizing GC-OMe2ꢀ is analogous to one which we reported before (see Inorg. Chem. 2009, 48, 5620 and Inorg. Chem.
2010, 49, 7080) where NiꢀSODred active site mimics can be assembled at will with electronically variant RSꢀ ligands. Discussed
herein is our initial account pertaining to the aqueous behavior of isolable, small-molecule NiꢀSOD model complexes (non-
maquette based). Spectroscopic (FTIR, UVꢀvis, ESI-MS, XAS) and electrochemical (CV) measurements suggest that 2ꢀ5
successfully simulate many of the electronic features of NiꢀSODred. Furthermore, the aqueous studies reveal a dynamic behavior
with regard to RSꢀ lability and bridging interactions, suggesting a stabilizing role brought about by the protein architecture.
’ INTRODUCTION
Scheme 1. Chemistry Catalyzed by SOD Split into Half
Reactions
The cytotoxic superoxide anion radical (O2•ꢀ) is an inevitable
byproduct of aerobic metabolism that is capable of effecting sig-
nificant cellular damage if not tightly regulated.1ꢀ3 In fact, increased
production of superoxide has been associated with carcinogenesis,4,5
post-ischemic tissue damage,3,6 and neurodegenerative diseases
such as Parkinson’s7,8 and Alzheimer’s.3,9 To maintain safe con-
centrations of intracellular O2•ꢀ (∼10ꢀ10 M), aerobic organisms
employ superoxide dismutases (SODs), enzymes that catalyze
amine of His1, the anionic carboxamido-N from Cys2, and the
cysteine thiolates of Cys2 and Cys6 to complete an N2S2 square-
planar geometry (Chart 1).18,19 In its oxidized Ni(III) state, the
coordination geometry is square pyramidal with the imidazole-N
of His1 occupying the apical position (Chart 1).18,19 Although a
growing body of protein,15,16,20ꢀ26 polypeptide maquette,27ꢀ34
and small-molecule35ꢀ41 literature has shed some light upon the role of
this unusual donor set, many outstanding questions remain as to the
interplay between the Ni ion, ligands, secondary coordination sphere,
substrate, and products during SOD catalysis.20ꢀ30,35ꢀ39,42ꢀ44
the diffusion-controlled (109 Mꢀ1 ꢀ1) disproportionation of
s
O2•ꢀ to O2 and H2O2.10 All SODs possess a transition metal in
their active site, which catalyzes the disproportionation reaction
by toggling between oxidized and reduced states as demonstrated
in Scheme 1. Over the years, a rich knowledge of SODs employ-
ing redox-active Fe,11,12 Mn,13 and Cu14 has accumulated. However,
less is understood about the newly discovered NiꢀSOD, which is
isolated from certain Streptomyces microbial strains15,16 and also
believed to be utilized by cyanobacteria.17
While several redox-active Ni-containing enzymes are known,45ꢀ49
coordination of cysteine thiolates appears to be a reoccurring
The NiꢀSOD enzyme is strikingly different from other
known SODs. Crystallographic characterization reveals the Ni
ion within a nine residue (HCDLPCGVY) “Ni-hook motif”. In
its reduced Ni(II) state, the metal is bound by the N-terminal
Received: July 29, 2011
Published: September 20, 2011
r
2011 American Chemical Society
10460
dx.doi.org/10.1021/ic2016462 Inorg. Chem. 2011, 50, 10460–10471
|

Inorganic Chemistry
ARTICLE
Chart 1. (Top) Active Site of NiꢀSODox (left), NiꢀSODred
(middle), and NiꢀSODred Analogues Described in the Pre-
sent Work, [Ni(GC-OMe)(SR)]ꢀ (right). (Bottom) R groups
used in this study
suitable water soluble construct to prepare the dinuclear metal-
losynthon, [Ni2(GC-OMe)2] (1). Complex 1 was then used as a
direct precursor toward the NiꢀSOD model complexes, K[Ni-
(GC-OMe)(SC6H4-p-Cl)] (2), K[Ni(GC-OMe)(StBu)] (3),
K[Ni(GC-OMe)(SC6H4-p-OMe)] (4), and K[Ni(GC-OMe)-
(SNAc)] (5) (Chart 1). These monomeric Ni(II) complexes
were prepared via methodologies previously described by us.38,39
The present study is among the few regarding the characteriza-
tion of discrete isolable NiꢀSOD analogues in an aqueous
environment (low MW containing only one peptide linkage;
non-maquette based) providing for the use of small-molecule
characterization techniques in aqueous media, which has been
utilized exclusively for tripeptide33,34 and polypeptide maquette
NiꢀSOD analogues.27ꢀ30 The synthesis, structure, and proper-
ties of 1ꢀ5 are discussed in this work. The results obtained shed
light on the behavior of species electronically analogous to the
NiꢀSODred active site under pseudo-physiological conditions
when stripped of their macromolecular surroundings.
feature, presumably in order to depress the redox couples
required to achieve both high Ni(III) and low Ni(I) oxidation
states toward physiologically accessible potentials.50,51 Intrigu-
ingly, the majority of these enzymes catalyze reactions of great
environmental and economical importance, such as the equili-
bration of CO2 and CO for utilization as one-carbon equivalents47
and consumption and generation of H+ and H2.45 In NiFeꢀ
hydrogenase and NiꢀSOD, the thiolate ligands have been
implicated as active players in the catalytic cycle performing
roles beyond redox modulation, including H+ storage and trans-
port.26,52ꢀ54 The unusual coordination observed at the NiꢀSOD
active site has provided a novel platform for synthetic chemists to
model and gain a basic understanding of such interactions. The
fundamental role of the atypical donor set, in conjunction with
the obvious relevance of SOD to health and medicine,4ꢀ9 has
provided motivation for modeling studies pertaining to this unique
metalloenzyme.
Previously, we reported a general and facile route toward
NiꢀSODred model complexes utilizing the tridentate N2S
ligand nmp2ꢀ (where nmp2ꢀ is the dianion of N-(2-mercapt-
oethyl)picolinamide) that models the basal plane contributions
of His1 and Cys2.38,39 The exogenously added monodentate
thiolate ligand in these [Ni(nmp)(SR)]ꢀ complexes (where RSꢀ
models Cys6) can be systematically substituted via a variety of
synthetic protocols. This feature was exploited to obtain a library
of [Ni(nmp)(SR)]ꢀ complexes with electronically variant mono-
dentate thiolates, which incorporate second-sphere functional-
ities tethered via the exogenously added thiolate. This study was
fruitful in that it allowed for an in-depth analysis as to the effects
manifested by intramolecular hydrogen bonding/thiolate pro-
tonation on the structural, spectroscopic, and reactivity proper-
ties of NiN2S2 complexes related to NiꢀSODred. For example,
we demonstrated that H-bonding to Ni-coordinated thiolates
promotes more metal-based reactivity via reduction of S-con-
tributions to the HOMO (relative to Ni) as well as by decreasing
the electrostatic potentials of the coordinated thiolates.38 Un-
fortunately, the [Ni(nmp)] series of complexes suffer from poor
solubility and stability in aqueous solution, thus limiting reactiv-
ity studies to organic solvents.
’ EXPERIMENTAL SECTION
General Information. All reagents were purchased from commer-
cial suppliers and used as received unless otherwise noted. Acetonitrile
(MeCN), methylene chloride (CH2Cl2), tetrahydrofuran (THF), di-
ethyl ether (Et2O), and pentane were purified by passage through
activated alumina columns using an MBraun MB-SPS solvent purifica-
tion system and stored under a dinitrogen (N2) atmosphere before use.
N,N0-dimethylformamide (DMF) was purified with a VAC Solvent
Purifier containing 4 Å molecular sieves and stored under similar con-
ditions. Triethylamine (TEA) was dried over Na2SO3. Anhydrous MeOH
was obtained by distilling from Mg(OMe)2 under an N2 atmosphere. All
reactions were performed under an inert atmosphere of N2 using
standard Schlenk line techniques or in an MBraun Unilab glovebox
under an atmosphere of purified N2. Potassium salts of the monodentate
thiolate ligands were prepared according to a modified literature
procedure55 via addition of 1 mol-equiv of K(0) to the appropriate
thiol in dry MeOH.
Physical Methods. FTIR spectra were collected on a Thermo-
Nicolet 6700 spectrophotometer running the OMNIC software; sam-
ples were prepared as pressed KBr pellets. Electronic absorption spectra
were run at 298 K using a Cary-50 spectrophotometer containing a
Quantum Northwest TC 125 temperature control unit. The UVꢀvis
samples were prepared in gastight Teflon-lined screw cap quartz cells
with an optical pathlength of 1 cm equipped with a rubber septum.
Cyclic voltammetry (CV) measurements were performed with a PAR
model 273A potentiostat using a Ag/Ag+ (0.01 M AgNO3/0.1 M
nBu4NPF6 in MeCN) reference electrode, Pt counter electrode, and
glassy carbon millielectrode (2 mm) as the working electrode when
measuring in DMF. An Ag/AgCl reference electrode was used when
measuring the aqueous samples. Measurements were performed at
ambient temperature using 5.0 mM analyte in the appropriate Ar-purged
n
solvent containing 0.1 M Bu4NPF6 as the supporting electrolyte in
DMF or 0.5 M KNO3 for aqueous measurements. The “Maximize
Stability” mode was utilized in the PAR PowerCV software utilizing a
low-pass 5.3 Hz filter. To ensure accuracy in the measured CVs, these
experiments were performed in triplicate while polishing the working
electrode between each run, and we report an average Eox. Additionally,
potentials were checked and corrected by recording the CV of a
ferrocene standard in DMF or a potassium ferricyanide standard in
aqueous conditions (referenced to the ferro/ferricyanide couple in
50 mM pH 7.0 phosphate buffer)56 under the same conditions as the
complexes before each run. NMR spectra were recorded in the listed
deuterated solvent on a 400 MHz Bruker BZH 400/52 NMR
The objective of the present work was to prepare small-
molecule [Ni(N2S)(SR)]ꢀ complexes of variant RSꢀ (similar
to the Ni(nmp) system) as NiꢀSOD models that could be studied
in a biologically relevant aqueous environment. To this end, we
utilized thedipeptideN2Sligand, H2N-Gly-L-Cys-OMe (abbreviated
as GC-OMeH2, where H represents dissociable protons), as a
10461
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
spectrometer or 500 MHz Varian Unity INOVA NMR spectrometer at
ambient temperature with chemical shifts referenced to TMS or residual
protio signal of the deuterated solvent. Low-resolution ESI-MS data
were collected using a Perkin-Elmer Sciex API I Plus quadrupole mass
spectrometer and high-resolution ESI-MS data were collected using a
Bruker Daltonics 9.4 T APEX Qh FT-ICR-MS. Elemental analysis for C,
H, and N was performed at QTI-Intertek in Whitehouse, NJ.
solvent protio signal, thus difficult to integrate, SH). 13C NMR (100.6
MHz, CD3CN, δ from solvent): 222.35 (CdOTFA), 171.02 (CdOester),
167.09 (CdOpeptide), 55.68, 53.21, 41.52, 31.47, 26.74. 1H NMR (500
MHz, D2O, δ from protio solvent): 4.82 (m, 1H), 3.94 (s, 2H), 3.82
(s, 3H), 3.03 (m, 2H). FTIR (KBr pellet), νmax (cmꢀ1): 3325 (m, NꢀH),
3138 (br w, NꢀH), 1753 (s, CdOester), 1677 (vs, CdOpeptide), 1631
(w) 1563 (w), 1552 (m), 1536 (w), 1513 (w), 1493 (w), 1434 (m), 1350
(m), 1323 (w), 1208 (s, CꢀF), 1186 (s, CꢀF), 1129 (s), 1043 (w), 965
(w), 906 (m), 841 (m), 802 (m), 726 (m), 572 (w). LRMS-ESI (m/z):
[M + H]+ calcd for C6H13N2O3S, 193.2; found, 193.0.
Synthesis of H2N-Gly-L-Cys-OMe TFA (GC-OMe TFA). Synth-
3
3
esis of the ligand comprised the following steps.
Step 1. H2N-L-Cys(STrit)-OMe. To a batch of 13.701 g (52.629 mmol)
of triphenyl methanol dissolved in 50 mL of trifluoroacetic acid (TFA)
was added 9.095 g (52.99 mmol) of L-cysteine methyl ester hydro-
chloride forming a red-orange solution. The resultant solution was left to
stir at room temperature for 1 h before it was concentrated to an orange
oil, which was partitioned between 300 mL of CH2Cl2 and 300 mL of
H2O, resulting in bleaching of the orange color. Portions of K2CO3 were
then slowly added to the aqueous layer until a basic pH was sustained for
up to 1 h. The organic layer was then separated, washed with saturated
NaHCO3(aq) and brine, dried over MgSO4, and concentrated to a pale
oil which solidified after overnight stirring in hexanes. The white solids
thus obtained were isolated via vacuum filtration (16.104 g, 42.660
[Ni2(GC-OMe)2] (1). Procedure 1 (TFA Adduct). To a batch of
0.174 g (0.699 mmol) of Ni(OAc)2 4H2O stirring in 12 mL of MeOH
3
was added a 4 mL MeOH solution of GC-OMeH2 TFA (0.217 g, 0.709
3
mmol) followed by a 3 mL MeOH solution of NaOAc (0.060 g, 0.731
mmol). The resultant deep orange solution was stirred at room
temperature for 16 h, after which the reaction was concentrated to
dryness to afford an orange residue. This residue was then stirried in
20 mL of MeCN, which resulted in 0.153 g (0.250 mmol, 72%) of
1
product as peach-colored solids. H NMR (400 MHz, D2O, δ from
solvent): 4.28 (m, 1H), 3.96 (s, 3H), 3.49 (dd, 1H), 3.32 (dd, 1H), 2.51
(br, s, 1H), 2.31 (d, 1H), 2.08 (s, 1H), 1.94 (s, 1H). FTIR (KBr pellet),
νmax (cmꢀ1): 3420 (w, br, OH), 3246 (m, br, NꢀH), 3119 (m, br,
NꢀH), 2950 (w), 1729 (s, CdOester), 1694 (m, CdOTFA), 1601
(vs, CdOpeptide), 1437 (m), 1409 (s), 1338 (w), 1273 (w), 1205
(s, CꢀF), 1169 (s, CꢀF), 1138 (m), 1019 (w), 973 (w), 944 (w),
843 (w), 801 (w), 722 (w), 669 (w), 650 (w), 625 (w), 584 (w), 459 (w).
HRMS-ESI (m/z): [M + H]+ calcd for C12H21N4Ni2O6S2 (relative
abundance), 496.9604 (100), 497.9637 (13), 498.9559 (88), 499.9591
(11), 500.9515 (22); found, 496.9614 (100), 497.9646 (11), 498.9570
(88), 499.9604 (10), 500.9522 (18). UVꢀvis (pH 7.5, 50 mM PIPES,
298 K) λmax, nm (ε, Mꢀ1 cmꢀ1): 338 (4,470), 458 (570). Anal. Calcd for
C12H20N4Ni2O6S2 TFA 2H2O: C, 25.95; H, 3.89; N, 8.65. Found: C,
1
mmol, 81%). Mp: 57ꢀ58 °C. H NMR (400 MHz, CDCl3, δ from
TMS): 7.48 (d, 6H), 7.32 (t, 6H), 7.25 (t, 3H), 3.68 (s, 3H), 3.24
(m, 1H), 2.64 (dd, 1H), 2.52 (dd, 1H), 1.54 (br s, 2H, NH2). 13C NMR
(100.6 MHz, CDCl3, δ from TMS): 174.26 (CdO), 144.64, 129.67,
128.04, 126.87, 66.94, 53.89, 52.22, 37.01. FTIR (KBr pellet), νmax
(cmꢀ1): 3391 (m, NꢀH), 3318 (w, NꢀH), 3062 (w), 2997 (w), 2966
(w), 2948 (w), 2920 (w), 1725 (vs, CdOester), 1593 (m), 1487 (s), 1446
(s), 1434 (s), 1371 (m), 1348 (m), 1316 (m), 1260 (m), 1205 (s), 1186
(s), 1172 (s), 1095 (m), 1031 (m), 1020 (m), 1001 (m), 952 (m), 892
(w), 854 (m), 838 (m), 810 (m), 771 (m), 747 (s), 705 (s), 676 (m), 665
(m), 630 (m), 615 (s), 530 (w), 508 (m), 494 (m), 474 (m). LRMS-ESI
(m/z): [M + H]+ calcd for C23H24NO2S, 378.2; found, 378.2.
Step 2. Boc-Gly-L-Cys(STrit)-OMe. A batch of 4.089 g (10.83 mmol) of
H2N-L-Cys(STrit)-OMe and 2.947 g (10.82 mmol) of Boc-Gly-OSu
was combined in 200 mL of CH2Cl2 and stirred at room temperature
overnight. The solution was then washed with saturated NaHCO3(aq)
and brine and dried over MgSO4 before concentration to a white foam
solid (4.934 g, 9.228 mmol, 85%). Mp: 59ꢀ62 °C. 1H NMR (400 MHz,
CDCl3, δ from TMS): 7.44 (d, 6H), 7.33 (t, 6H), 7.26 (t, 3H), 6.80
(br s, 1H, NHpeptide), 5.38 (br, s, 1H, NHcarbamate), 4.59 (m, 1H), 3.82
(m, 2H), 3.72 (s, 3H), 2.71 (m, 2H), 1.49 (s, 9H). 13C NMR (100.6
MHz, CDCl3, δ from TMS): 170.70 (CdOester), 169.31 (CdOpeptide),
155.95 (CdOcarbamate), 144.27, 129.51, 128.03, 126.92, 80.11, 66.98,
52.64, 51.20, 33.67, 28.32. FTIR (KBr pellet), νmax (cmꢀ1): 3320 (br s,
NꢀH), 3057 (m), 3030 (m), 2977 (s), 2952 (m), 2931 (m), 1746 (vs,
CdOester), 1720 (vs, CdOcarbamate), 1689 (vs, CdOpeptide), 1595 (w),
1514 (s), 1493 (s), 1434 (m), 1391 (w), 1367 (m), 1248 (m), 1210 (m),
1168 (s), 1083 (w), 1050 (m), 1031 (m), 1001 (w), 985 (w), 941 (w),
862 (w), 766 (m), 744 (s), 701 (s), 675 (m), 621 (m), 544 (w), 505 (w).
LRMS-ESI (m/z): [M + K]+ calcd for C30H34KN2O5S, 573.2;
found, 573.0.
3
3
26.13; H, 3.94; N, 8.62.
Procedure 2. To a batch of 1.292 g (5.192 mmol) of Ni(OAc)2 4H2O
3
and 0.428 g (5.22 mmol) of NaOAc dissolved in 80 mL of MeOH was
added 1.584 g (5.172 mmol) of GC-OMeH2 TFA in 10 mL of MeOH.
3
The reaction mixture instantly became a deep red-orange color upon
addition of ligand and was subsequently heated to 50 °C for 16 h, which
resulted in precipitation of product as pink solids. The solids were
isolated via vacuum filtration to afford 1.100 g (2.210 mmol, 85%) of
product. The isolated product is not readily soluble as is the TFA adduct
(see above), but prolonged stirring in D2O or buffer provided orange
solutions with identical spectral properties (1H NMR, UVꢀvis, ESI-
MS). FTIR revealed no TFA peaks. FTIR (KBr pellet), νmax (cmꢀ1):
3326 (w, NꢀH), 3228 (w, NꢀH), 3084 (br w, NꢀH), 2978 (w), 2962
(w), 2929 (w), 1755 (s, CdOester), 1578 (vs, CdOpeptide), 1433 (m),
1404 (m), 1360 (m), 1312 (w), 1265 (w), 1200 (m), 1177 (m), 1150
(m), 1114 (m), 1055 (w), 1030 (w), 993 (w), 969 (w), 946 (w), 853
(w), 801 (w), 717 (w), 660 (w), 603 (w), 573 (w), 483 (w), 424 (m).
Anal. Calcd for C12H20N4Ni2O6S2 0.5H2O: C, 28.44; H, 4.18; N, 11.05.
3
Found: C, 28.47; H, 3.94; N, 10.78.
[Ni2(GC-OMe)2] (1) Prepared in Situ. To a freshly dissolved
Step 3. H2N-Gly-L-Cys-OMe TFA (GC-OMeH2 TFA). A batch of 3.748
batch of 18 mg (0.06 mmol) of GC-OMeH2 TFA in pH 7.5 buffer
3
3
3
g (7.010 mmol) of Boc-Gly-L-Cys(STrit)-OMe was dissolved in 40 mL
of a 1:1 CH2Cl2/TFA solution resulting in a bright-yellow color. After
30 min, 1.092 g (9.391 mmol) of Et3SiH was added dropwise to the
solution, which bleached the bright-yellow solution to a clear and pale-
yellow color. After 90 min of stirring, the solution was concentrated to
∼50% of its original volume and the resulting insoluble triphenyl methane
byproduct filtered off and washed with TFA. The solution was then
concentrated to a pale oil, which solidified upon stirring in Et2O (1.601
g, 5.228 mmol, 75%). Mp: 74ꢀ77 °C. 1H NMR (500 MHz, CD3CN, δ
from protio solvent): 7.86 (br s, 3H, NH3), 7.62 (s, NHpeptide), 4.73
(m, 1H), 3.80 (m, 2H), 3.72 (s, 3H), 2.93 (m, 2H), 1.92 (br s, coincides with
(50 mM PIPES) was added 15 mg (0.06 mmol) of Ni(OAc)2 4H2O to
3
form a richly colored orange solution. All spectral properties are in
accordance with those of solid batches of isolated 1 upon dissolution.
K[Ni(GC-OMe)(SC6H4-p-Cl)] (2). To an 8 mL DMF solution
containing 0.142 g (0.777 mmol) of KSC6H4-p-Cl was added a 5 mL
DMF slurry of 1 (0.202 g, 0.406 mmol). The resulting purple hetero-
geneous mixture was heated to 45 °C and stirred at this temperature for
16 h to form a rich mostly homogeneous violet-colored solution. The
reaction was subsequently cooled to room temperature and filtered to
remove any unreacted 1, and the violet mother liquor was concentrated
to dryness. The resultant purple residue was then taken up in 2 mL of
10462
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
THF, which was then precipiated out by addition of 12 mL of Et2O. The
product was collected as dull violet solids via vacuum filtration (0.316 g,
0.732 mmol, 94%). 1H NMR (500 MHz, D2O, as isolated product in the
presence of 1 extra mol-equiv of KSC6H4-p-Cl, δ from protio solvent):
7.53 (d, 1.4H), 7.30 (d, 1.7H, free KSC6H4-p-Cl), 7.14 (d, 1.4H), 7.08
(d, 1.7H, free KSC6H4-p-Cl), 4.15 (m, 1H), 3.79 (s, 3H), 3.31 (dd, 1H),
3.15 (dd, 1H), 2.60 (m, 1H), 2.24 (d, 1H). 1H NMR (400 MHz,
CD3CN, as isolated solid, δ from protio solvent): 7.68 (d, 1.3H), 6.93
(d, 1.1H), 3.84 (m, 1H), 3.61 (s, 2.4H), 3.01 (m, 1H), 2.91 (m, 1H), 2.36
(m, 1H), 2.15 (s, 1H, NH), 2.13 (m, 1H), 1.87 (br s, 1H, NH). FTIR
(KBr pellet), νmax (cmꢀ1): 3325 (w, NH), 3195 (br w, NꢀH), 3108
(br w, NꢀH), 2948 (m), 2924 (m), 1722 (s, CdOester), 1586
(vs, CdOpeptide), 1468 (s), 1437 (m), 1410 (m), 1336 (w), 1265 (w),
1207 (m), 1168 (m), 1090 (s), 1033 (m), 1010 (m), 934 (w), 818 (m),
740 (w), 701 (w), 662 (w), 543 (m), 499 (w), 425 (w). HRMS-ESI (m/
z): [M ꢀ K]ꢀ calcd for C12H14ClN2NiO3S2 (relative abundance),
390.9482 (100), 391.9515 (13), 392.9439 (48), 393.9470 (6), 394.9408
(21); Found, 390.9487 (100), 391.9520 (11), 392.9442 (41), 393.9472
(4), 394.9416 (16). UVꢀvis (DMF, 298 K) λmax, nm (ε, Mꢀ1 cmꢀ1):
481 (390), 560 (230). UVꢀvis (pH 7.5, 50 mM PIPES, 298 K) λmax, nm
(ε, Mꢀ1 cmꢀ1): 471 (430), 560 (240). Eox (DMF): 220 mV. Eox (pH
KSC6H4-p-OMe, 3H from OMe of free KSC6H4-p-OMe), 3.23 (dd,
1H), 3.07 (dd, 1H), 2.53 (m, 2H). H NMR (500 MHz, CD3CN, as
1
isolated solid, δ from protio solvent): 7.40 (d, 2H), 6.50 (d, 0.3H), 6.41
(d, 1.7H), 3.75 (t, 1H), 3.55 (s, 3H), 3.44 (s, 3H), 2.93 (m, 2H), 2.17 (m,
1H), 2.08 (m, 1H and br s, 1H, NH), 1.85 (br s, 1H, NH). FTIR (KBr
pellet), νmax (cmꢀ1): 3319 (w, NꢀH), 3190 (br w, NꢀH), 3090 (br w,
NꢀH), 2946 (w), 2834 (w), 1720 (s, CdOester), 1663 (m, CdODMF),
1587 (vs, CdOpeptide), 1486 (s), 1463 (w), 1439 (m), 1409 (m), 1336
(w), 1273 (m), 1235 (s), 1169 (s), 1099 (m), 1028 (m), 933 (w), 826
(w), 796 (w), 638 (w), 625 (w), 561 (w), 525 (w), 470 (w), 425 (w).
HRMS-ESI (m/z): [M ꢀ K]ꢀ calcd for C13H17N2NiO4S2 (relative
abundance), 386.9978 (100), 388.0011 (14), 388.9933 (48), 389.9965
(7); found, 386.9980 (100), 388.0011 (14), 388.9936 (46), 389.9983
(11). UVꢀvis (DMF, 298 K) λmax, nm (ε, Mꢀ1 cmꢀ1): 480 (500), 565
(200). UVꢀvis (pH 7.5, 50 mM PIPES, 298 K) λmax, nm (ε,
Mꢀ1 cmꢀ1): 467 (400), 550 (230). Eox (DMF): 170 mV. Eox (pH
7.4): 315 mV. Anal. Calcd for C13H17KN2NiO4S2 0.5H2O: C, 35.79; H,
3
4.16; N, 6.42. Found: C, 35.84; H, 4.46; N, 6.48.
K[Ni(GC-OMe)(S-NAc)] (5). To a 5 mL DMF solution containing
0.094 g (0.249 mmol) of 3 was added a 2 mL DMF solution of N-acetyl-
L-cysteine methyl ester (0.045 g, 0.254 mmol). The solution rapidly
changed from violet to orange-brown and was left to stir at room
temperature for 16 h. The solution was then concentrated to a brick-red
colored residue and stirred in 15 mL of a 2:1 Et2O/THF mixture to
afford free-flowing solids. The solids were collected via vacuum filtration
7.4): 285 mV. Anal. Calcd for C12H14ClKN2NiO3S2 0.5H2O: C, 32.71;
3
H, 3.43; N, 6.36. Found: C, 32.63; H, 3.56; N, 6.50.
K[Ni(GC-OMe)(StBu)] (3). To 115 μL (0.092 g, 1.02 mmol) of tert-
butyl thiol dissolved in 3 mL of DMF was added 0.040 g (0.997 mmol) of
KH as a 4 mL DMF slurry to generate KStBu. Immediate effervescence
was observed, and a pale yellow homogeneous solution formed over the
course of 20 min. To this solution was added a 5 mL DMF slurry of
1 (0.278 g, 0.558 mmol), resulting in a purple heterogeneous mixture.
This solution was then heated to 45 °C and stirred for 16 h to form a rich
violet-colored solution. The reaction was subsequently cooled to room
temperature and filtered to remove any unreacted 1, and the violet
mother liquor was concentrated to dryness. The resultant purple residue
was taken up in 6 mL of THF, which was treated with 25 mL of Et2O to
form free-flowing solids. The product was collected as pink solids via
1
to yield 0.105 g (0.226 mmol, 91% yield) of product. H NMR (400
MHz, D2O, as isolated product in the presence of one extra mol-equiv of
KS-NAc, δ from protio solvent): 4.45 (t, 1H, free KSNAc), 4.32 (t, 1H),
4.11 (d, 1H), 3.73 (s, 6H, coincidental peaks: 3H each from OMe of
coordinated [GC-OMe]2ꢀ and [S-NAc]ꢀ, 3H from OMe of free KS-
NAc), 3.31 (dd, 1H), 3.17 (dd, 1H), 2.86 (m, 2.5H, free KSNAc), 2.63
(m, 1H), 2.26 (d, 1H), 2.14 (m, 2H), 2.08 (s, 3H), 2.02 (s, 4H, free
KSNAc). 1H NMR (500 MHz, CD3CN, as isolated solid, δ from protio
solvent): 8.62 (br s, 1H, NH), 4.05 (br s, 1H), 3.90 (br s, 1H), 3.60
(s, 8H, coincidental peaks from: OMe of both coordinated [GC-OMe]2ꢀ
and [SN-Ac]ꢀ, residual THF from reaction workup), 3.32 (br s, 1H),
3.05 (m, 1H), 2.97 (m, 1H), 2.51 (m, 1H), 2.14 (m, 10H, integrates
slightly high due to overlap with residual H2O peak). FTIR (KBr pellet),
νmax (cmꢀ1): 3246 (br w, NꢀH), 2950 (w), 2919 (w), 2848 (w), 1731
(s, CdOester), 1662 (s, CdOSNAC peptide), 1589 (vs, CdOGCOMe peptide),
1436 (m), 1412 (m), 1372 (w), 1337 (w), 1302 (w), 1268 (w), 1212
(m), 1167 (m), 1122 (w), 1032 (w), 938 (w), 901 (w), 843 (w), 800
(w), 661 (w), 567 (w), 488 (w), 423 (w). HRMS-ESI (m/z): [M ꢀ K]ꢀ
calcd for C12H20N3NiO6S2 (relative abundance), 424.0142 (100),
425.0174 (13), 426.0097 (48), 427.0129 (6), 428.0072 (5); Found,
424.0144 (100), 425.0180 (12), 426.0101 (48), 427.0135 (6), 428.0081
(4). LRMS-ESI (m/z): [M ꢀ K]ꢀ calcd for C12H20N3NiO6S2, 424.0;
found, 424.0. UVꢀvis (DMF, 298 K) λmax, nm (ε, Mꢀ1 cmꢀ1): 358 sh
(1,200), 463 (350), 545 (160). UVꢀvis (pH 7.5, 50 mM PIPES, 298 K)
λmax, nm (ε, Mꢀ1 cmꢀ1): 334 (1,200), 464 (300), 545 (140).
Eox (DMF): 310 mV. Eox (pH 7.4): 550 mV. Anal. Calcd for
C12H20KN3NiO6S2 0.33THF 0.33H2O: C, 32.40; H, 4.76; N, 8.50.
1
vacuum filtration (0.328 g, 0.870 mmol, 87%). H NMR (400 MHz,
D2O, as isolated product in the presence of 1 extra mol-equiv of KStBu, δ
from protio solvent): 4.01 (d, 1H), 3.71 (s, 3H), 3.30 (dd, 1H), 3.14
(dd, 1H), 2.53 (m, 1H), 2.21 (d 1H), 1.26 (s, 9H). 1H NMR (500 MHz,
CD3CN, as isolated solid, δ from protio solvent): 3.70 (m, 1H), 3.59
(s, 3H), 3.04 (m, 1H), 2.95 (m, 1H), 2.30 (m, 1H), 2.13 (m, 1H), 1.84
(s, 2H, NH), 1.31 (s, 9H). FTIR (KBr pellet), νmax (cmꢀ1): 3343 (br w,
NꢀH), 3222 (br w, NꢀH), 3107 (br w, NꢀH), 2948 (w), 2887 (w),
2848 (w), 1721 (m, CdOester), 1584 (vs, CdOpeptide), 1439 (w), 1410
(m), 1355 (w), 1273 (w), 1207 (m), 1166 (m), 932 (w), 742 (w), 646
(w), 586 (w), 465 (w). UVꢀvis (DMF, 298 K) λmax, nm (ε, Mꢀ1 cmꢀ1):
358 (1400), 484 (440), 570 (230). Eox (DMF): 80 mV. Anal. Calcd for
C10H19KN2NiO3S2 H2O: C, 30.39; H, 5.36; N, 7.09. Found: C, 30.08;
3
H, 5.02; N, 7.16.
K[Ni(GC-OMe)(SC6H4-p-OMe)] (4). To a 3 mL DMF solution
containing 0.046 g (0.258 mmol) of KSC6H4-p-OMe was added a 3 mL
DMF slurry of 1 (0.073 g, 0.147 mmol). The purple heterogeneous
reaction mixture that formed was then heated to 45 °C and stirred for
16 h to form a violet-brown mostly homogeneous solution. The reaction
was subsequently cooled to room temperature and filtered to remove
any unreacted 1, and the violet-brown mother liquor was concentrated
to dryness. To the resultant red-brown residue was added 5 mL each of
THF and Et2O, and the residue was scraped to form free-flowing solids.
The product was collected as red-brown solids via vacuum filtration
(0.096 g, 0.225 mmol, 87% yield). 1H NMR (400 MHz, D2O, as isolated
product in the presence of 1 extra mol-equiv of KSC6H4-p-OMe, δ from
protio solvent): 7.37 (d, 2H), 7.21 (d, 2H, free KSC6H4-p-OMe),
6.72 (m, 4H, overlap of bound and free KSC6H4-p-OMe), 4.08
(d, 1H), 3.71 (s, 9H, coincidental peaks from: 3H from OMe of coordinated
3
3
Found: C, 32.46; H, 4.80; N, 8.67.
Bulk Oxidation of 2. To a batch of 52 mg (0.12 mmol) of 2 in 3 mL
of DMF was added 40 mg (0.12 mmol) of ferrocenium hexafluoropho-
sphate in 3 mL of DMF. Instantaneously, the violet solution developed
an orange-brown color and was left to stir at room temperature for 2 h.
The reaction mixture was then concentrated to an orange-brown
residue, which solidified upon stirring in Et2O. Brown solids (45 mg)
were separated via vacuum filtration, and the yellow mother liquor was
concentrated to a pale yellow residue (40 mg). FTIR (KBr), 1H NMR
(D2O), and ESI-MS (positive ion mode) revealed the brown insoluble
solids to be comprised of 1 and KPF6 (87% recovery), while 1H NMR in
CDCl3 of the yellow Et2O-soluble residue was revelead to be comprised
10463
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
Scheme 2. Synthesis of GC-OMeH2
solely of ferrocene and the disulfide of the thiol HSC6H4-p-Cl (100%
recovery).
NiN2S2 species.38,39 In this work we utilized design methodol-
ogies from i and v (see previous) resulting in the Gly-Cys N2S
ligand that we denote as GC-OMeH2. We hypothesized that GC-
OMeH2 would provide water-soluble small-molecule NiꢀSOD
biomimetics that model the exact basal plane contributions of His1
and Cys2 in the enzyme. Simultaneously, via strategies developed
by our group, a fourth and exchangeable coordination site
amenable to synthetic fine tuning is left available by which
monodentate thiolate ligands (modeling the Cys6 contribution
in NiꢀSOD) can be appended.
X-ray Absorption Studies. Samples were loaded under anaerobic
conditions into 0.5 mm thick XAS aluminum sample holders with Mylar-
tape windows and quickly frozen in liquid nitrogen prior to XAS data
collection. Nickel K-edge XAS data were collected at the Stanford
Synchrotron Radiation Lightsource (SSRL) on beamline 7ꢀ3 with
the SPEAR storage ring operating at 3.0 GeV, 345ꢀ350 mA. X-ray
absorption spectra for the solid samples were recorded with the sample
at 10 K using a 1 mm ꢁ 4 mm vertical aperture beam incident on a fully
tuned Si[220] double-crystal monochromator and collected using
transmission mode. Three ionization chambers with N2 gas were used
with the first defining I0. A nickel foil was placed between the second and
the third ionization chambers I1 and I2 to serve as an internal calibration
standard with the first inflection point for the nickel foil assigned at
8331.6 eV. k values were calculated using a threshold (k = 0) energy of
8340 eV. The averaged XAS data for a solid sample represent 5ꢀ6 scans,
each of 27 min duration; the averaged spectra were calculated as
log(I0/I1). Data reduction and analysis were performed with EXAFSPAK
software (www-ssrl.slac.stanford.edu/exafspak.html) according to stan-
dard procedures as described before.57 Fourier transforms of the EXAFS
spectra were generated using sulfur-based phase correction. Phase and
amplitude functions used in the curve fitting were calculated using FEFF
version 8.0.58
Synthesis of GC-OMeH2 occurred in three steps by standard
peptide coupling routes that provided the final ligand in high
yield (Scheme 2). Protection of the thiol group of L-cysteine
methyl ester occurred in 81% yield by reaction of the amino acid
with triphenylmethanol (trityl protection) in TFA. Formation of
the pre-ligand is achieved by coupling of the protected cysteine
methyl ester with BOC-Gly-OSu, resulting in the S-trityl-pro-
tected ligand in 85% yield. Finally, deprotection of the trityl and
BOC groups with Et3SiH in a TFA/CH2Cl2 (1:1) mixture
furnished the TFA adduct of GC-OMeH2 in 74% yield. The
ligand is a white solid material that is thermally stable even in
aerobic conditions.
The monomeric Ni(II)ꢀN2S2 complexes originate from the
dimeric species, [Ni2(GC-OMe)2] (1), which is isolated in high
yield by reaction of Ni(OAc)2 4H2O with ligand and 1 mol-
3
’ RESULTS AND DISCUSSION
equiv of NaOAc in MeOH at room temperature, resulting in
orange-colored solutions. TFA still seems to be associated with 1
depending on the reaction conditions, which has been confirmed
by FTIR and elemental analysis (see Experimental Section).
These peach-colored solids have been formulated as an aquated
TFA species, namely, 1 TFA 2H2O, according to microanalysis.
Synthesis. In constructing accurate structural and functional
models of NiꢀSOD, certain design elements must be consid-
ered. First, the ligand frame must contain a mixed N/S donor set
to mimic the coordination sphere observed in the enzyme.
Second, and beyond a mere stoichiometric replication of the
donor atoms, is the electronic nature of the N/S frame which
should contain one peptide-N (carboxamide-N is the more
general terminology), one primary amine-N, and two cis-thio-
lates situated in a planar N2S2 arrangement. An additional
imidazole-N ligand would be required to replicate the four-
coordinate to five-coordinate Ni(II/III)ꢀSOD redox conver-
sion. Third, the ligand construct should be amenable to a variety
of straightforward modifications. This important last require-
ment will permit for electronic and structural “fine tuning” of the
biomimetic for its desirable function. Several groups have
achieved some or all of these criteria by employing (i) small
peptides or peptide maquettes typically 3ꢀ12 amino acids
long,27ꢀ30,33,34 (ii) symmetric N2S2 frames that approximate
the electronic nature of the NiꢀSOD donors,40,41,59 (iii) elec-
tronically accurate N2S2 frames,35,36 (iv) asymmetric N2S ligands
that approximate the Ni active site with an open coordination site
where exogenous N-donors bind (NiN3S),60,61 and (v) asym-
metric N2S ligands with NiꢀSOD accurate donors containing an
open coordination site where exogenous S-ligands bind to afford
3
3
Heating MeOH solutions of 1 TFA 2H2O to 50 °C for 16 h
3
3
resulted in precipitation of analytically pure [Ni2(GC-OMe)2]
(1) as pink solids with no evidence of TFA peaks in the IR
spectrum. Complex 1 is generally insoluble in both polar and
non-polar solvents, a property consistent with previously isolated
dimeric S,S0-bridged Ni2(N2S)2 complexes;38,39 however, stir-
ring in water resulted in slow dissolution to afford orange solu-
tions with a spectral profile identical to that obtained via the
spontaneous metalation of GC-OMeH2 TFA with Ni(II) in pH
3
7.5 PIPES buffer. ESI-MS+ analysis of 1 also revealed the
presence of species corresponding to [Ni2(GC-OMe)2 + H]+
(1 + H), [Ni3(GC-OMe)3 + H]+, and [Ni4(GC-OMe)4 + H]+,
suggesting that various permutations of 1:1 Ni/GC-OMe stoi-
chiometry may exist upon dissolution (Figures S17ꢀ19, Sup-
1
porting Information). Regardless, H NMR studies in D2O
revealed a chemical equivalency throughout any species which
may be present (vide infra). It should be noted that deprotona-
tion of ligand protons with a strong base such as NaH in an
aprotic solvent (MeCN) resulted in formation of intractable
10464
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
brown solids upon addition of Ni(II); the exact nature of such
species is unknown at present, but FTIR revealed loss of the GC-
OMeH2 ester functionality, suggesting that the coordination
due to light scattering as these solutions became cloudy. The
dimeric complex 1 is air stable in solution over 16 h as evidenced
by 1H NMR and UVꢀvis, consistent with the stability brought
about by S,S0-bridging. Collectively, solutions of 2ꢀ5 are rela-
tively air stable and can be handled aerobically without significant
decomposition onthehour timescale. However, prolonged aerobic
storage resulted in color bleaching and cloudiness for periods of
extended exposure (>6 h).
chemistry of GC-OMeH2 TFA with Ni(II) is relatively facile in
3
polar protic solvents and a strong base is not required.
Despite its unfavorable solubility properties, complex 1 is a
good metallosynthon for preparation of the monomeric Ni(II)
complexes 2ꢀ5 via addition of 2 mol-equiv of the appropriate
potassium thiolate salts to DMF solutions of 1 heated to 45 °C
(Scheme 3) (where SC6H4-p-Cl = thiolate of p-chlorobenzene
thiol; StBu = thiolate of tert-butylthiol; SC6H4-p-OMe = thiolate
of p-methoxybenzenethiol; SNAc = thiolate of N-acetyl L-
cysteine methyl ester). The S,S0-bridge-splitting reactions to
form 2ꢀ4 resulted in high yields of the complexes (87ꢀ90%).
As reported by our group, complexes of general formula
[Ni(N2S)(SR)]ꢀ can be synthesized several ways: (i) S,S0-
bridge-splitting of Ni dimer (similar protocols have been utilized
by Holm62 and Darensbourg60); (ii) pKa-driven thiol exchange;
and (iii) a disulfide/coordinated thiolate redox exchange
procedure.38,39 These alternate methodologies have been suc-
cessfully applied to this generation of water soluble NiꢀSOD
mimics. For example, complex 5 was obtained in 90% yield via
addition of 1 mol-equiv of HSNAc to 3 in DMF. Complex 2 was
also prepared from 3 utilizing 0.5 mol equiv of the disulfide of
HSC6H4-p-Cl. These facile reactions to obtain monomeric Ni-
(II) species demonstrate the efficacy of these preparations when
applied to different ligand systems that may be applied to future
NiꢀSOD models.
Infrared. FTIR spectra of metalꢀpeptide complexes serve as
good indicators for peptido-N f M binding from the observed
red-shift in νCO of the carbonyl peptide. This noted shift is
observed in 1ꢀ5with νCO frequencies between 1578 and 1589 cmꢀ1
(KBr matrix), which red-shifted ∼100 cmꢀ1 from the free ligand
(νCO 1677 cmꢀ1 in KBr) and confirmed formation of the
peptido-N f M bond. In general, νCO appeared to be invariant,
regardless of the donor strength of the monodentate thiolate
ligand trans to the peptido-N. In fact, the lowest energy νCO
stretch was observed for 1 despite the bridging nature of the
coordinated thiolates. As expected, νNH of the peptide-N was
absent in the Ni(II) complexes. Coordination of the ꢀNH2 func-
tionality in 1ꢀ5 was also evident as the νNH stretches appeared at
values from 3042 (5) to 3108 cmꢀ1 (2) compared to νNH of the
free ligand (3318 cmꢀ1). In general, νNH of organic primary
amines occurs at values g3250 cmꢀ1 64
.
1
1H NMR and ESI-MS. H NMR spectra of 1ꢀ5 reveal that
these Ni(II) complexes are d8 square-planar systems in solution
(D2O and CD3CN at 298 K) as evident by their spectra, which
are typical for diamagnetic (S = 0) molecules. For example, the
D2O 1H NMR spectrum of dimer 1 revealed only one signal per
non-equivalent proton, indicative of an electronically identical
environment per Ni(II) with respect to GC-OMe2ꢀ protons. In
’ PROPERTIES
1
General. Complexes 1ꢀ5 are stable in anaerobic aqueous
solutions for up to 16 h as monitored by 1H NMR. No evidence
of ligand protonation, dissociation, or metal-mediated ester
hydrolysis is observed which is in contrast to other reported
Ni(II)N2S2 complexes.63 UVꢀvis spectral monitoring of com-
plexes 2, 4, and 5 in pH 7.5 buffer (PIPES) revealed slight
increases in the intensity of λmax over prolonged time periods (>6 h)
contrast, the D2O H NMR spectra of as-isolated 2ꢀ5 display
broad, ill-resolved signals indicative of an equilibrium mixture of
Ni(II) species likely due to loss of the exogenous RSꢀ ligand and
formation of [Nix(GC-OMe)x(SR)] or [Ni(GC-OMe)(OH2)x]-
type species (Scheme 4). The precise nature of the species present
upon dissolution of the as-isolated complexes is unknown, but
the spectra suggest more than one which may be oligomeric in
nature. When 1 mol-equiv of the corresponding thiolate is added
to these D2O solutions, however, the NMR spectra became well
resolved with defined peaks that are consistent with monomeric
(S = 0) [Ni(GC-OMe)(SR)]ꢀ complexes (Figure 1 for 2, see the
Supporting Information for 3ꢀ5). ESI-MSꢀ analysis of as-
isolated 2 and 4 dissolved in protic solvents revealed ions corre-
sponding to [Ni2(GC-OMe)2(SR)]ꢀ, [Ni3(GC-OMe)3(SR)]ꢀ,
and [Ni4(GC-OMe)4(SR)]ꢀ in addition to those of the parent
ion (Figures S20ꢀ25, Supporting Information). Similarly, [Ni2-
(GC-OMe)2(SNAc)]ꢀ is observed along with the parent ion for
5. The tBuSꢀ ligand of complex 3 dissociated completely at pH
Scheme 3. Synthesis of Ni(II) Complexes 2ꢀ5
Scheme 4. Equilibrium Distribution of Speculative [Nix(GC-OMe)x(SR)] Species (where x = 2ꢀ4) Present upon Dissolution of
As-Isolated 2ꢀ5 in pH 7.5 PIPES Buffer at 298 Ka
a L is a generic ligand potentially representative of solvent or other [Ni(GC-OMe)] fragments.
10465
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
of such behavior and display clean spectra consistent with the
diamagnetic square-planar [Ni(GC-OMe)(SR)]ꢀ formulation
(Supporting Information). Thus, the as-isolated Ni(II) mono-
meric complexes remain fully intact in polar aprotic solvents,
which further support the notion that the equilibrium is pro-
moted by interaction with solvent protons and thiolate dissociation.
Electronic Absorption Spectroscopy. The UVꢀvis spectra
of 2ꢀ5 in DMF or pH 7.5 buffer (PIPES) at 298 K display two
predominant features in the visible region (∼470 and 550 nm)
(Figure 2, Table 1), which are consistent with the red-violet color
23
of these solutions and similar in nature to NiꢀSODred
.
In
DMF, the most intense visible band is broad and ranges from
463 nm (ε = 350 Mꢀ1 cmꢀ1) for 5 to 484 nm (ε = 440
Mꢀ1 cmꢀ1) for 3; similar visible transitions are observed in
NiꢀSODred at 450 nm (ε = 480 Mꢀ1 cm
)
ꢀ1 23 and other square-
planar Ni(II)N2S2 complexes.35,36,65ꢀ71 These absorptions are
proposed to be ligand-field (dꢀd) transitions containing a minor
charge-transfer (CT) contribution.23 A second less intense
ligand-field band appears as a shoulder between 545 nm (ε =
160 Mꢀ1 cmꢀ1) for 5 and 570 nm (230 Mꢀ1 cmꢀ1) for 3 and
mirrors a similar feature in NiꢀSODred at 543 nm (ε = 150
Mꢀ1 cmꢀ1). The high-energy band at 361 nm (ε = 880
Mꢀ1 cmꢀ1) in NiꢀSODred is also observed in 3 (λmax
=
358 nm; ε = 1400 Mꢀ1 cmꢀ1) and 5 (λmax = 358 nm; ε =
1200 Mꢀ1 cmꢀ1) further underscoring the electronic structural
similarities between our models and the enzyme (Figure S1,
Supporting Information). The same high-energy feature is
difficult to elucidate in 2 and 4 due to the presence of an
overwhelming transition in the UV, presumably πꢀπ* in nature.
UVꢀvis spectra of 2, 4, and 5 in buffer do not differ much from
those in DMF (Figure 2, Table 1), suggesting that the various
Ni(II) species present retain GC-OMe2ꢀ ligation. For example,
λmax = 471 nm (ε = 430 Mꢀ1 cmꢀ1) for 2 and 458 nm (ε = 570
Figure 1. 1H NMR spectrum of 2 recorded under different conditions
at 298 K. (Top) As-isolated 2 in D2O (signals at 3.01 and 2.85 ppm
correspond to associated DMF). (Middle) Complex 2 in D2O after
addition of 1 mol-equiv of KSC6H4-p-Cl (doublets at 7.41 and 7.19
correspond to non-coordinated KSC6H4-p-Cl). (Bottom) As-isolated 2
in CD3CN (signal at 3.42 corresponds to Et2O present in the sample;
2.89 and 2.77 correspond to associated DMF; 1.94 arises from residual
protio solvent).
M
ꢀ1 cmꢀ1; 285 Mꢀ1 cmꢀ1/Ni) for 1. Similarly, the less intense
feature at 545 nm (ε = 140 Mꢀ1 cmꢀ1) for 5 and 560 nm (ε = 240
ꢀ1 cmꢀ1) for 2 is somewhat less pronounced than in the DMF
M
samples. Higher energy transitions at λmax ≈ 338 nm are also
present for 1 and 5 (Figure S2, Supporting Information).
Interestingly, dissolution of 3 in buffer provides a spectral profile
identical to dimer 1, suggestive of protonation and loss of the
more basic tBuSꢀ ligand.
7.5, affording orange-colored solutions that are consistent with
reformation of dimer 1. However, addition of 1 mol-equiv of
KStBu in D2O reestablished the violet monomer (vide infra).
Furthermore, neither the parent ion of 3 or any species of
formula [Nix(GC-OMe)x(StBu)]ꢀ is observed in the ESI-MS.
These results suggest a dynamic situation involving potential
S-bridging interactions, loss of monodentate thiolate, and possi-
ble aquation/solvation for 2ꢀ5 in H2O and other protic media
(Scheme 4). It is not too surprising that an equilibrium exists
with these peptide ligands as a nonapeptide NiꢀSOD maquette
analogue has been shown to exist in equilibrium with a 2:1 Ni/
peptide and 1:1 Ni/peptide species in solution via UVꢀvis
titrations.31 However, ESI-MS measurements reveal a 1:2 Ni/
peptide ratio with a considerable amount of free peptide. These
seemingly contradictory results clearly establish that in the
absence of the NiꢀSOD protein matrix the probability of
multiple NiꢀL species (where L is a generic representation of
any peptido-N/thiolato-S ligand of denticity e 3) is high at least
in water. In fact, site-directed mutagenesis of the Cys-S ligands of
NiꢀSOD suggest that even mutation of one cysteine to serine
promotes a high-spin (S = 1) octahedral Ni(II) site with no
S-ligands (replaced by H2O).22,24 To support this proposal, the
1H NMR spectra of as-isolated 2ꢀ5 in CD3CN show no evidence
Since the aqueous electronic absorption spectra of as-isolated
2, 4, and 5 represent an equilibrium mixture of Ni(II) species we
also recorded their spectra in the presence of excess thiolate to
generate one monomeric species (see NMR discussion above).
When 1 mol-equiv of KSC6H4-p-Cl, KSC6H4-p-OMe, and
HSNAc are added to as-isolated buffered solutions of 2, 4, and
5, respectively, only subtle changes in the visible region of the
UVꢀvis spectra resulted (Figures S26ꢀ28, Supporting In-
formation). For example, the lowest energy ligand-field shoulder
between 545 and 560 nm becomes more pronounced for all
complexes. As discussed above, 3 forms dimer 1 when dissolved
in buffer. Even addition of up to 5 mol equiv of KStBu to as-
isolated 3 afforded no change in the UVꢀvis. Interestingly,
addition of 1 mol equiv of HSNAc to 5 resulted in loss of
the 334 nm band (Figure S28, Supporting Information)
that likely originates from 1 or a species of high nuclearity,
suggesting a similar dissociation process may also occur in 5.
Collectively, these results suggest that the Ni(II) species present
upon dissolution of as-isolated 2, 4, and 5 in buffer are electro-
nically similar to the formally monomeric species that form in the
presence of excess RSꢀ.
10466
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
Figure 2. Electronic absorption spectra for 2 (black), 3 (green), 4 (blue), and 5 (red) in DMF (left) and of 2, 4, and 5 (same color scheme) in pH 7.5
PIPES buffer (right) recorded at 298 K.
Table 1. Spectroscopic (UVꢀvis) and Electrochemical (CV) Properties of 2ꢀ5 in the Listed Solvents in Comparison to
NiꢀPeptide Maquettes and NiꢀSODred
complex
λmax, nm (ε, Mꢀ1 cmꢀ1) in DMF λmax, nm (ε, Mꢀ1 cmꢀ1) in buffer E (V vs Ag/AgCl) in DMF E (V vs Ag/AgCl) in buffer
ref
2a
3a
4a
5a
481 (390)
560 (230)
484 (440)
570 (230)
480 (500)
565 (200)
463 (350)
545 (160)
N/A
471 (430)
560 (240)
N/Ae
0.220
0.080
0.170
0.310
N/A
N/A
N/A
0.285
N/Ae
0.315
0.550
0.700
0.520
0.487
this work
this work
this work
this work
467 (400)
550 (230)
464 (300)
545 (140)
458 (510)
552 (240)
457 (345)
548 (130)
450 (480)
543 (150)
[Ni(SODM1)]b
[Ni(SODM2)]c
28
29
N/A
N/A
d
22,24
NiꢀSODred
a UVꢀvis and CV (Eox reported) recorded at 298 K; spectra recorded in pH 7.5 PIPES buffer; CV recorded in pH 7.4 phosphate buffer. b M1 = H2N-
HCDLPCGVYDPA-CO2H;28 UVꢀvis and CV recorded at room temperature in pH 7.2 phosphate buffer (quasi-reversible E1/2 reported on a sample
immobilized as a thin film; 0.1 M NaCl electrolyte). c M2 = H2N-HCDLPCG-CO2H;30 UVꢀvis and CV recorded at room temperature in pH 7.4 NEM
buffer (quasi-reversible E1/2 reported on a sample immobilized as a thin-film; 0.1 M NaClO4 electrolyte).30 d Redox titration recorded at room
temperature in pH 7.5 phosphate buffer (E1/2 originally referenced versus NHE = 0.290 V).25 e Complex 3 reverts to dimer 1 in aqueous solutions
(see the main text).
The similarities between the electronic absorption spectra of
1ꢀ5 and NiꢀSODred suggest that our models successfully
reproduce the electronic environment of the NiN2S2 unit in
the enzyme active site (Figure 2, Table 1). One interesting point
to note when comparing spectra obtained in DMF and buffer is
that the ligand-field transitions of 2 and 4 blue shift ∼10 nm in
buffer; however, complex 5 does not shift at all. An attractive
explanation is that the peptide-NH of the SNAcꢀ ligand forms an
intramolecular H-bond with the thiolato-S ligands in DMF in a
manner approximating aqueous H-bonding interactions.38 The
observation that δNH shifts by >1 ppm downfield in the 1H NMR
(8.62 ppm in CD3CN) of the tethered peptide of SNAcꢀ in 5
(Supporting Information) compared to δNH of free HSNAc
(7.43 ppm) supports the presence of intramolecular H-bonding
interactions. Prior experimental and theoretical investigations
pertaining to SNAcꢀ bound to the Ni(nmp) synthon also
corroborate this notion.38 Furthermore, spontaneous loss of
the tBuSꢀ ligand from 3 in buffer provides supporting evidence
for interaction of the coordinated thiolates with solvent protons
in these systems.
Electrochemistry. Cyclic voltammetry (CV) measurements
in DMF show irreversible oxidation (Eox) events for 2ꢀ5
(Figure 3, Table 1). Irreversible CVs have been typical for a
number of Ni(II)ꢀN2S2 SOD models35,36,38,39 and indicate an
overall change in the coordination environment after oxidation,
most likely a result of a thiolateꢀdisulfide conversion, which is
prevented in the enzyme. Comparison of the potentials demon-
strated patterns which can be roughly correlated to the donor
strength of the monodentate RSꢀ ligand. For example, Eox for the
reported complexes increase (more difficult to oxidize) in the
following order: 3 (0.080 V vs Ag/AgCl) < 4 (0.170 V) < 2
(0.220 V) < 5 (0.310 V). This observation corresponds well with
the trend established using the Ni(nmp) synthon where redox
potentials of [Ni(N2S)(SR)]ꢀ complexes increase as the donor
strength of the exogenous RSꢀ decreases when separately
analyzing aryl- and alkyl-SR ligands.38 As observed in the UVꢀvis,
complexes featuring alkyl RSꢀ donors tend to be more sensitive
to electronic perturbations such as intramolecular H-bonding
and solvent environment. In pH 7.4 buffer (phosphate) Eox
follows 2 (0.285 V vs Ag/AgCl) < 4 (0.315 V) < 5 (0.550 V)
10467
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
Figure 3. (Left) Cyclic voltammograms (CVs) of 5 mM solutions of 2 (black), 3 (green), 4 (blue), and 5 (red) (vs Ag/Ag+ in DMF, 0.1 M nBu4NPF6
supporting electrolyte, glassy carbon working electrode, scan rate 100 mV/s, room temperature). (Right) CVs of 5 mM solutions of 2, 4, and 5 (vs Ag/
AgCl in pH 7.4 phosphate buffer, 0.5 M KNO3 supporting electrolyte, glassy carbon working electrode, scan rate 100 mV/s, room temperature). The CV
of 5 in buffer was scaled down (33%). Arrow indicates direction of scan.
(Figure 3, Table 1). Comparing 2 and 4, it is evident that the
more positive redox event belongs to the complex with the more
strongly donating RSꢀ. Among the alkyl RSꢀ no comparison is
available due to spontaneous dissociation of tBuSꢀ from 3 in buffer
(1 shows no redox waves up to +1.00 V). In fact, a second Eox is
observed in the CV of 4 at 0.480 V, which was found to match
that of KSC6H4-p-OMe dissolved in the same buffer and suggests
that the facile RSꢀ dissociation observed in 3 occurs in 4 as well
albeit to a lesser magnitude. The RSꢀ protonation/dissociation
observed in buffer points toward strong interactions between
RSꢀ and solvent protons. The appearance of such an interaction
is expected to increase with increasing RSꢀ basicity and serves to
rationalize the reverse trend in Eox of 2 and 4 as well as the
observed dissociation of the more basic thiolates. In accordance
with this hypothesis, shifts of 0.065, 0.145, and 0.220 V to more
positive potentials are observed for 2, 4, and 5, respectively, when
comparing the potentials obtained in buffer to those in DMF; a
trend which also correlates to relative RSꢀ basicity.
The Eox values for 2ꢀ5 arise at potentials which are distinct
from those obtained for free ligand and the K+ salts of their
respective RSꢀ ligands in both DMF and buffer, supporting the
notion that the redox-active species present are indeed the Ni(II)
complexes and not dissociated RSꢀ. It should also be noted that
addition of 1 mol-equiv of the appropriate RSꢀ or RSH ligand to
solutions of the as-isolated complexes provided no change in Eox,
consistent with the aforementioned observation (UVꢀvis,
above) that the species present upon dissolution of as-isolated
2, 4, and 5 in aqueous buffer are electronically similar to the
corresponding discrete monomeric species. The recorded Eox
values also fall between the 0.04 and 1.09 V (vs Ag/AgCl)
potential window defined by the two SOD half reactions72 and
imply that these complexes are thermodynamically poised for at
least one turnover. However, the irreversible nature of the CV
demonstrates that the Ni(III) state obtained during catalysis in
the enzyme is not stabilized in these models. Not surprisingly,
few NiꢀSOD peptide maquettes or small-molecule analogues
have ever been isolated in the corresponding Ni(III) state and
attempts at chemical oxidation resulted in complex degradation,
most likely via disulfide formation.29 In fact, disulfide formation
of the coordinated monodentate thiolate in 2, 4, and 5 appear to
be responsible for the reported Eox. Bulk oxidation of 2 with
1 mol-equiv of a ferrocenium salt afforded 1 and the disulfide of
p-chlorobenzene thiol in quantitative yield. Accordingly, these
complexes provide no protection against superoxide reduction of
p-nitroblue tetrazolium chloride (NBT) to its ring-open forma-
zan form in the presence of 25 mol-equiv of KO2.73 It should be
noted that addition of up to 10 mol-equiv of an exogenous
N-donor such as imidazole (Im) imparted no SOD activity,
electrochemical reversibility as measured by CV, or changes in
the UVꢀvis spectrum, suggesting no affinity of the Ni center in
these systems for Im.
X-ray Absorption Spectroscopy (XAS). Although single
crystals of the Niꢀpeptide complexes were not obtained, we
were able to obtain limited structural information on these sys-
tems through Ni K-edge X-ray absorption spectroscopy (XAS).
The XAS acquired for complexes 2 and 5 reveal Ni(II) systems
with an approximately square-planar coordination geometry
about the metal center. The XANES area of the spectrum
confirmed this assignment as each complex displays an intense
pre-edge peak corresponding to a 1s f 4pz transition at ∼8336 eV
(see Figure 4 and Table 2), which is characteristic of Ni(II) in
D4h (square-planar) symmetry.74,75 For comparison, XAS were
acquired on members of the crystallographically characterized
[Ni(nmp)(SR)]ꢀ family of complexes, which also afford edge
features similar to 2 and 5 as further confirmation of the square-
planar nature of the systems reported here. The EXAFS region of
2 and 5 are best fit with two non-equivalent NiꢀN scatterers at
1.83(3) and 1.96(3) Å, which we assign as NiꢀNpeptide and
NiꢀNamine, respectively, based on EXAFS of the Ni(nmp)
complexes (see Table 2 and the Supporting Information). To
complete the NiN2S2 coordination sphere, two NiꢀS scatterers
are also observed at an average distance of 2.16(2) Å. These
distances compare well with other NiꢀN2S2 SOD models,35,36,38,39
peptide maquettes,29 and the enzyme itself.18ꢀ20 The XAS thus
1
complement the UVꢀvis, ESI-MS, and H NMR studies and
reveal that the reported Ni(II)ꢀpeptide complexes are excellent
structural and spectroscopic analogues of NiꢀSODred.
Relationship to NiꢀSOD. Ligand substitution in square-
planar d8 complexes generally follow an associative or interchange
pathway where a five-coordinate (5C) trigonal bipyramidal inter-
mediate is invoked in the transition state.66 In the present systems,
one may assume a 5C intermediate is traversed (and may be
stabilized) during spontaneous RSꢀ dissociation in aqueous
buffer. Formation of a higher coordination number species
1
would explain the broadness in the H NMR spectra due to
potential Ni(II) S = 1 solvated derivatives that form once the
10468
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
Figure 4. Ni K-edge X-ray absorption data for K[Ni(GC-OMe)(S-p-C6H4Cl)] (2) (left) and K[Ni(GC-OMe)(S-NAc)] (5) (right). (Top) Edge
spectra displaying the Ni (1s f 4pz) transition consistent with square-planar Ni(II). (Middle) EXAFS data. (Bottom) FT k3 EXAFS data. Experimental
data are in black with simulations in red.
Table 2. NiꢀK-Edge X-ray Absorption Parameters for Complexes 2 and 5 and [Ni(nmp)(SR)]ꢀ Complexes38,39 a
pre-edge
b
complex
n
shell
Ras (Å)
σas2 (Å2)
f0
Eo (eV)
peak (eV)
intensity
0.31
K[Ni(GC-OMe)(S-p-C6H4Cl)] (2)
1
1
2
1
1
2
1
1
2
1
1
2
1
1
2
NiꢀN
NiꢀN
NiꢀS
NiꢀN
NiꢀN
NiꢀS
NiꢀN
NiꢀN
NiꢀS
NiꢀN
NiꢀN
NiꢀS
NiꢀN
NiꢀN
NiꢀS
1.83
1.99
2.16
1.83
1.99
2.17
1.85
1.97
2.16
1.86
1.96
2.17
1.85
1.97
2.16
0.0024
0.0024
0.0059
0.0024
0.0024
0.0052
0.0024
0.0024
0.0045
0.0024
0.0024
0.0028
0.0024
0.0024
0.0051
0.088
8340
8335.8
K[Ni(GC-OMe)(SNAc)] (5)
K[Ni(nmp)(S-p-C6H4Cl)]
(Et4N)[Ni(nmp)(S-o-babt)]
K[Ni(nmp)(StBu)]
0.100
0.075
0.061
0.076
8340
8340
8340
8335.7
8337.6
8336.5
8336.4
0.34
0.42
0.42
0.35
8340
2
a Shell is the chemical unit defined for the multiple scattering calculation. Ras is the metal-scatterer distance. σas is a mean square deviation in Ras. b f0 is a
normalized error (chi squared): f0 = ({∑i[k3(χoi bs ꢀ χci alcd)]2/N}1/2)/([(k3χobs
)
max ꢀ (χcalcd
)min)]).
monodentate thiolate is lost. Recent mutagenesis studies by
Maroney and Brunold on NiꢀSOD at Cys2, Cys6, and Cys2/
Cys6 (double mutant) have shown that even the absence of one
cysteine thiolate promotes a high-spin (S = 1) aquated Ni(II)
center at the NiꢀSOD active site with no evidence of the remaining
NiꢀSCys bond in the single mutants.22,24 This finding, in combi-
nation with the results reported here, suggests more than
redox-modulation/H-storage roles for cysteine in NiꢀSOD. We
propose that both Cys6 and Cys2 ligands in NiꢀSOD are crucial
for proper active site assembly and stabilization of the low-spin
10469
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
square-planar Ni(II) state. This explanation is similar to the
advocated H-bond directionality and positioning that Glu17 and
Arg47 impart on His1 especially toward formation of the Ni-
(III)ꢀNHis bond in NiꢀSODox. Perhaps this hypothesis also
explains the various Ni/peptide stoichiometries observed in
other NiꢀSOD maquettes31 as well as the chiral inversion seen
in the Ni(NCC) tripeptide complex34 over time. While the
hexameric quaternary structure of NiꢀSOD likely precludes
Cys6 dissociation, cysteinate protonation may be sufficient to
trigger weak binding of the His1 imidazole (becoming 5C) to
further destabilize the Ni(II) center. The promiscuous nature of
these SOD models in buffer underscores the importance of the
immediate environment surrounding the active site and points to
a role for secondary and tertiary sphere residues in the overall
structure and in preserving the integrity and activity of the
mononuclear assembly. Various mutagenesis studies on the
enzyme have confirmed this notion.25 Accordingly, the six Ni
centers of the NiꢀSOD hexamer comprise the vertices of an
octahedron with a large interior cavity and are separated by a
minimum distance of 23 Å.18,19 It should be noted that NiꢀS
complexes are notorious for oligomerization in protic media51,76
and those containing monodentate thiolate ligands are known to
be particularly labile.55 In fact, to the best of our knowledge, there
remains a paucity of literature precedence regarding square-
planar Ni complexes featuring monodentate thiolates62 outside
of our work and few pertaining to the characterization of such
species in water.
protic media. The NMR data in D2O coupled with ESI-MS analysis
revealed various species of general formula [Ni1ꢀ4(GC-OMe)1ꢀ4
-
(SR)]ꢀ and suggest that the active site fragment of NiꢀSOD
may be a rather dynamic species with a propensity to oligomerize
outside of the protein matrix. Thus, the surrounding environ-
ment may be crucial to maintaining coordinative integrity. At
present, we are undertaking studies pertaining to the role played
by the axial imidazole-N donor77 and exploring ligand modifica-
tions with the aim of achieving electrochemical reversibility and
ultimately SOD activity.
’ ASSOCIATED CONTENT
S
Supporting Information. Full UVꢀvis spectra of 3 and
b
5, full-scale 1H NMR in D2O and CD3CN, high-resolution ESI-
MS, UVꢀvis of 2, 4, and 5 in buffer with and without added thiolate,
and X-ray absorption data for Ni(nmp) complexes. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: tharrop@uga.edu.
’ ACKNOWLEDGMENT
T.C.H. acknowledges support from the National Science
Foundation (NSF CAREER grant CHE-0953102). R.A.S. ac-
knowledges support from the National Institutes of Health (NIH
GM042025).
’ CONCLUSIONS
In summary, the metallopeptide-derived complexes 1ꢀ5
provide one of the few aqueous studies pertaining to isolable,
small-molecule NiꢀSOD biomimetics (non-maquette based).
The dimeric complex 1 serves as a suitable metallosynthon to
generate several unique Ni(II)ꢀN2S2 complexes with one vari-
able basal plane coordination site. The electronic absorption
spectra of the resulting monomers 2ꢀ5 in both DMF and pH 7.5
buffer compare well with that obtained for NiꢀSODred and
suggest that our [Ni(GC-OMe)(SR)]ꢀ models have a similar
electronic structure to NiꢀSODred. CV measurements in DMF
and buffer display irreversible oxidation events within the poten-
tial window defined by the SOD half reactions. The CV data
suggests that while our mimics are electrochemically capable of
promoting SOD chemistry, the Ni(III) state is not accessible,
which is crucial for catalysis. Apparently the N2S2 planar ligand
lacks electrochemical stabilization needed to support a compe-
tent SOD mimetic. The failure of our models to provide protec-
tion to NBT from superoxide reduction corroborates this notion.
This lack of electrochemical reversibility and SOD activity may
be attributed to the absence of an axial N-donor or the protective
protein shell to stabilize a higher coordinate structure. In fact,
flooding these complexes with excess imidazole does not impart
any electrochemical reversibility or other spectroscopic changes.
The 1H NMR spectra of as-isolated 2ꢀ5 in D2O provide broad,
ill-resolved signals which may correspond to a variety of species
present, perhaps in exchange with one another or with solvent.
However, addition of 1 mol-equiv of the appropriate thiolate
ligand afforded well-resolved spectra with splitting patterns
consistent with a monomeric diamagnetic square-planar species.
Dissolution of as-isolated 2ꢀ5 in CD3CN afforded neat, readily
discernible 1H NMR spectra consistent with discrete species and
suggesting further speciation is assisted via interaction with
’ REFERENCES
(1) Beyer, W.; Imlay, J.; Fridovich, I. Prog. Nucleic Acid Res. Mol. Biol.
1991, 40, 221.
(2) Maritim, A. C.; Sanders, R. A.; Watkins, J. B., III J. Biochem. Mol.
Toxicol. 2003, 17, 24.
(3) Valentine, J. S.; Wertz, D. L.; Lyons, T. J.; Liou, L.-L.; Goto, J. J.;
Gralla, E. B. Curr. Opin. Chem. Biol. 1998, 2, 253.
(4) Kumar, B.; Koul, S.; Khandrika, L.; Meacham, R. B.; Koul, H. K.
Cancer Res. 2008, 68, 1777.
(5) Ishikawa, K.; Takenaga, K.; Akimoto, M.; Koshikawa, N.;
Yamaguchi, A.; Imanishi, H.; Nakada, K.; Honma, Y.; Hayashi, J.-I.
Science 2008, 320, 661.
(6) Fortunato, G.; Pastinese, A.; Intrieri, M.; Lofrano, M. M.; Gaeta,
G.; Censi, M. B.; Boccalatte, A.; Salvatore, F.; Sacchetti, L. Clin. Biochem.
1997, 30, 569.
(7) Kocat€urk, P. A.; Akbostanci, M. C.; Tan, F.; Kavas, G. O.
Pathophysiology 2000, 7, 63.
(8) Riley, D. P. Chem. Rev. 1999, 99, 2573.
(9) De Leo, M. E.; Borrello, S.; Passantino, M.; Palazzotti, B.;
Mordente, A.; Daniele, A.; Filippini, V.; Galeotti, T.; Masullo, C.
Neurosci. Lett. 1998, 250, 173.
(10) Auchꢀere, F.; Rusnak, F. J. Biol. Inorg. Chem. 2002, 7, 664.
(11) Tierney, D. L.; Fee, J. A.; Ludwig, M. L.; Penner-Hahn, J. E.
Biochemistry 1995, 34, 1661.
(12) Miller, A.-F. Curr. Opin. Chem. Biol. 2004, 8, 162.
(13) Borgstahl, G. E. O.; Parge, H. E.; Hickey, M. J.; Beyer, W. F.,
Jr.; Hallewell, R. A.; Tainer, J. A. Cell 1992, 71, 107.
(14) Tainer, J. A.; Getzoff, E. D.; Richardson, J. S.; Richardson, D. C.
Nature 1983, 306, 284.
(15) Youn, H.-D.; Youn, H.; Lee, J.-W.; Yim, Y.-I.; Lee, J. K.; Hah,
Y. C.; Kang, S.-O. Arch. Biochem. Biophys. 1996, 334, 341.
(16) Youn, H.-D.; Kim, E.-J.; Roe, J.-H.; Hah, Y. C.; Kang, S.-O.
Biochem. J. 1996, 318, 889.
10470
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471

Inorganic Chemistry
ARTICLE
(17) Palenik, B.; Brahamsha, B.; Larimer, F. W.; Land, M.; Hauser,
L.; Chain, P.; Lamerdin, J.; Regala, W.; Allen, E. E.; McCarren, J.;
Paulsen, I.; Dufresne, A.; Partensky, F.; Webb, E. A.; Waterbury, J.
Nature 2003, 424, 1037.
(51) Fox, S.; Wang, Y.; Silver, A.; Millar, M. J. Am. Chem. Soc. 1990,
112, 3218.
(52) Pavlov, M.; Siegbahn, P. E. M.; Blomberg, M. R. A.; Crabtree,
R. H. J. Am. Chem. Soc. 1998, 120, 548.
(18) Barondeau, D. P.; Kassmann, C. J.; Bruns, C. K.; Tainer, J. A.;
Getzoff, E. D. Biochemistry 2004, 43, 8038.
(53) Niu, S.; Thomson, L. M.; Hall, M. B. J. Am. Chem. Soc. 1999,
121, 4000.
(19) Wuerges, J.; Lee, J.-W.; Yim, Y.-I.; Yim, H.-S.; Kang, S.-O.;
Carugo, K. D. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 8569.
(20) Choudhury, S. B.; Lee, J. -W.; Davidson, G.; Yim, Y.-I.; Bose, K.;
Sharma, M. L.; Kang, S.-O.; Cabelli, D. E.; Maroney, M. J. Biochemistry
1999, 38, 3744.
(21) Bryngelson, P. A.; Arobo, S. E.; Pinkham, J. L.; Cabelli, D. E.;
Maroney, M. J. J. Am. Chem. Soc. 2004, 126, 460.
(22) Ryan, K. C.; Johnson, O. E.; Cabelli, D. E.; Brunold, T. C.;
Maroney, M. J. J. Biol. Inorg. Chem. 2010, 15, 795.
(23) Fiedler, A. T.; Bryngelson, P. A.; Maroney, M. J.; Brunold, T. C.
J. Am. Chem. Soc. 2005, 127, 5449.
(54) Amara, P.; Volbeda, A.; Fontecilla-Camps, J. C.; Field, M. J.
J. Am. Chem. Soc. 1999, 121, 4468.
(55) Rosenfield, S. G.; Armstrong, W. H.; Mascharak, P. K. Inorg.
Chem. 1986, 25, 3014.
(56) O’Reilly, J. E. Biochim. Biophys. Acta 1973, 292, 509.
(57) Scott, R. A. In Physical Methods in Bioinorganic Chemistry:
Spectroscopy and Magnetism; Que, L., Jr., Ed.; University Science Books:
Sausalito, CA, 2000; p 465.
(58) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. Phys.
Rev. B 1998, 58, 7565.
(59) Gennari, M.; Orio, M.; Pꢀecaut, J.; Neese, F.; Collomb, M.-N.;
Duboc, C. Inorg. Chem. 2010, 49, 6399.
(24) Johnson, O. E.; Ryan, K. C.; Maroney, M. J.; Brunold, T. C.
J. Biol. Inorg. Chem. 2010, 15, 777.
(60) Jenkins, R. M.; Singleton, M. L.; Almaraz, E.; Reibenspies, J. H.;
Darensbourg, M. Y. Inorg. Chem. 2009, 48, 7280.
(61) Jenkins, R. M.; Singleton, M. L.; Leamer, L. A.; Reibenspies,
J. H.; Darensbourg, M. Y. Inorg. Chem. 2010, 49, 5503.
(62) Huang, D.; Deng, L.; Sun, J.; Holm, R. H. Inorg. Chem. 2009,
48, 6159.
(63) Baidya, N.; Olmstead, M. M.; Mascharak, P. K. Inorg. Chem.
1991, 30, 3967.
(64) Silverstein, R. M.; Webster, F. X.; Kiemle, D. J. Spectrometric
Identification of Organic Compounds; Wiley: Hoboken, NJ, 2005.
(65) Mills, D. K.; Reibenspies, J. H.; Darensbourg, M. Y. Inorg. Chem.
1990, 29, 4364.
(25) Herbst, R. W.; Guce, A.; Bryngelson, P. A.; Higgins, K. A.; Ryan,
K. C.; Cabelli, D. E.; Garman, S. C.; Maroney, M. J. Biochemistry 2009,
48, 3354.
(26) Szilagyi, R. K.; Bryngelson, P. A.; Maroney, M. J.; Hedman, B.;
Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2004, 126, 3018.
(27) Neupane, K. P.; Shearer, J. Inorg. Chem. 2006, 45, 10552.
(28) Shearer, J.; Neupane, K. P.; Callan, P. E. Inorg. Chem. 2009,
48, 10560.
(29) Shearer, J.; Long, L. M. Inorg. Chem. 2006, 45, 2358.
(30) Neupane, K. P.; Gearty, K.; Francis, A.; Shearer, J. J. Am. Chem.
Soc. 2007, 129, 14605.
(31) Tietze, D.; Breitzke, H.; Imhof, D.; Kothe, E.; Weston, J.;
Buntkowsky, G. Chem.—Eur. J. 2009, 15, 517.
(32) Schmidt, M.; Zahn, S.; Carella, M.; Ohlenschl€ager, O.; G€orlach,
M.; Kothe, E.; Weston, J. ChemBioChem 2008, 9, 2135.
(33) Krause, M. E.; Glass, A. M.; Jackson, T. A.; Laurence, J. S. Inorg.
Chem. 2010, 49, 362.
(66) Smee, J. J.; Miller, M. L.; Grapperhaus, C. A.; Reibenspies, J. H.;
Darensbourg, M. Y. Inorg. Chem. 2001, 40, 3601.
(67) Farmer, P. J.; Reibenspies, J. H.; Lindahl, P. A.; Darensbourg,
M. Y. J. Am. Chem. Soc. 1993, 115, 4665.
(68) Grapperhaus, C. A.; Maguire, M. J.; Tuntulani, T.; Darensbourg,
M. Y. Inorg. Chem. 1997, 36, 1860.
(34) Krause, M. E.; Glass, A. M.; Jackson, T. A.; Laurence, J. S. Inorg.
(69) Colpas, G. J.; Kumar, M.; Day, R. O.; Maroney, M. J. Inorg.
Chem. 2011, 50, 2479.
Chem. 1990, 29, 4779.
(35) Mathrubootham, V.; Thomas, J.; Staples, R.; McCraken, J.;
Shearer, J.; Hegg, E. L. Inorg. Chem. 2010, 49, 5393.
(36) Shearer, J.; Zhao, N. Inorg. Chem. 2006, 45, 9637.
(37) Shearer, J.; Dehestani, A.; Abanda, F. Inorg. Chem. 2008, 47, 2649.
(38) Gale, E. M.; Narendrapurapu, B. S.; Simmonett, A. C.; Schaefer,
H. F., III; Harrop, T. C. Inorg. Chem. 2010, 49, 7080.
(39) Gale, E. M.; Patra, A. K.; Harrop, T. C. Inorg. Chem. 2009,
48, 5620.
(70) Hanss, J.; Kr€uger, H.-J. Angew. Chem., Int. Ed. 1998, 37, 360.
(71) Maroney, M. J.; Choudhury, S. B.; Bryngelson, P. A.; Mirza,
S. A.; Sherrod, M. J. Inorg. Chem. 1996, 35, 1073.
(72) Sawyer, D. T.; Valentine, J. S. Acc. Chem. Res. 1981, 14, 393.
(73) Tabbi, G.; Driessen, W. L.; Reedijk, J.; Bonomo, R. P.; Veldman,
N.; Spek, A. L. Inorg. Chem. 1997, 36, 1168.
(74) Gu, Z.; Dong, J.; Allan, C. B.; Choudhury, S. B.; Franco, R.;
Moura, J. J. G.; Moura, I.; LeGall, J.; Przybyla, A. E.; Roseboom, W.;
Albracht, S. P. J.; Axley, M. J.; Scott, R. A.; Maroney, M. J. J. Am. Chem.
Soc. 1996, 118, 11155.
ꢁ
(40) Ma, H.; Chattopadhyay, S.; Petersen, J. L.; Jensen, M. P. Inorg.
Chem. 2008, 47, 7966.
(41) Ma, H.; Wang, G.; Yee, G. T.; Petersen, J. L.; Jensen, M. P. Inorg.
Chim. Acta 2009, 362, 4563.
(42) Pelmenschikov, V.; Siegbahn, P. E. M. J. Am. Chem. Soc. 2006,
128, 7466.
(75) Colpas, G. J.; Maroney, M. J.; Bagyinka, C.; Kumar, M.; Willis,
W. S.; Suib, S. L.; Baidya, N.; Mascharak, P. K. Inorg. Chem. 1991,
30, 920.
(76) Tremel, W.; Kriege, M.; Krebs, B.; Henkel, G. Inorg. Chem.
(43) Mullins, C. S.; Grapperhaus, C. A.; Kozlowski, P. M. J. Biol.
Inorg. Chem. 2006, 11, 617.
(44) Prabhakar, R.; Morokuma, K.; Musaev, D. G. J. Comput. Chem.
1988, 27, 3886.
(77) Gale, E. M.; Simmonett, A. C.; Telser, J.; Schaefer, H. F., III;
Harrop, T. C. Inorg. Chem. 2011, 50, in press (http://dx.doi.org/
10.1021/ic201822f).
2006, 27, 1438.
(45) Volbeda, A.; Charon, M. H.; Piras, C.; Hatchikian, E. C.; Frey,
M.; Fontecilla-Camps, J. C. Nature 1995, 373, 580.
(46) Ermler, U.; Grabarse, W.; Shima, S.; Goubeaud, M.; Thauer,
R. K. Science 1997, 278, 1457.
(47) Dobbek, H.; Svetlitchnyi, V.; Gremer, L.; Huber, R.; Meyer, O.
Science 2001, 293, 1281.
(48) Doukov, T. I.; Iverson, T. M.; Seravalli, J.; Ragsdale, S. W.;
Drennan, C. L. Science 2002, 298, 567.
(49) Darnault, C.; Volbeda, A.; Kim, E. J.; Legrand, P.; Vernede, X.;
Lindahl, P. A.; Fontecilla-Camps, J. C. Nat. Struct. Biol. 2003, 10, 271.
(50) Kr€uger, H.-J.; Peng, G.; Holm, R. H. Inorg. Chem. 1991, 30, 734.
10471
dx.doi.org/10.1021/ic2016462 |Inorg. Chem. 2011, 50, 10460–10471