Chromium(V) peptide complexes: Synthesis and spectroscopic characterization

Article abstract of DOI:10.1021/ic048322h

A series of stable Cr(V) model complexes that mimic the binding of Cr(V) to peptide backbones at the C-terminus of proteins have been prepared for N,N-dimethylurea derivatives of the tripeptides Aib₃-DMF, AibLAlaAib-DMF, and AibDAlaAib-DMF (Aib = 2-amino-2-methylpropanoic acid, DMF = N,N-dimethylformamide). The Cr(II) precursor complexes were synthesized by the initial deprotonation of the amide and acid groups of the peptide ligands in DMF with potassium tert-butoxide in the presence of CrCl₂. The Cr(II) intermediates thus formed were then immediately oxidized to Cr(V) using tert-butyl hydroperoxide. Spectroscopic and mass-spectrometric analyses of the Cr(V) complexes showed that a new metal-directed organic transformation of the ligand had occurred. This involved a DMF solvent molecule becoming covalently bound to the amine group of the peptide ligand, yielding a urea group, and a third coordinated deprotonated urea nitrogen donor. A metal-directed oxidative coupling has been proposed as a possible mechanism for the organic transformation. The Cr(V/IV) reduction potential was determined for the three Cr(V) complexes using cyclic voltammetry, and in all cases it was quasi-reversible. These are the first isolated and fully characterized Cr(V) complexes with non-sulfur-containing peptide ligands.

Full text of DOI:10.1021/ic048322h

Inorg. Chem. 2005, 44, 1044−1053
Chromium(V) Peptide Complexes: Synthesis and Spectroscopic
Characterization
Peter J. Barnard, Aviva Levina, and Peter A. Lay*
Centre for HeaVy Metals Research, School of Chemistry, UniVersity of Sydney,
New South Wales 2006, Australia
Received November 28, 2004
A series of stable Cr(V) model complexes that mimic the binding of Cr(V) to peptide backbones at the C-terminus
of proteins have been prepared for N,N-dimethylurea derivatives of the tripeptides Aib3-DMF, Aib AlaAib-DMF, and
Aib AlaAib-DMF (Aib 2-amino-2-methylpropanoic acid, DMF N,N-dimethylformamide). The Cr(II) precursor
L
D
)
)
complexes were synthesized by the initial deprotonation of the amide and acid groups of the peptide ligands in
DMF with potassium tert-butoxide in the presence of CrCl2. The Cr(II) intermediates thus formed were then immediately
oxidized to Cr(V) using tert-butyl hydroperoxide. Spectroscopic and mass-spectrometric analyses of the Cr(V)
complexes showed that a new metal-directed organic transformation of the ligand had occurred. This involved a
DMF solvent molecule becoming covalently bound to the amine group of the peptide ligand, yielding a urea group,
and a third coordinated deprotonated urea nitrogen donor. A metal-directed oxidative coupling has been proposed
as a possible mechanism for the organic transformation. The Cr(V/IV) reduction potential was determined for the
three Cr(V) complexes using cyclic voltammetry, and in all cases it was quasi-reversible. These are the first isolated
and fully characterized Cr(V) complexes with non-sulfur-containing peptide ligands.
Introduction
Cr(IV) complexes, free radicals, and reactive oxygen species
(ROS), with the Cr ultimately reduced to Cr(III).
While Cr(VI) is a well-recognized human carcinogen,1
numerous in vitro studies have shown that Cr(VI) does not
interact with, or damage, isolated DNA.2,3 However, in the
presence of reductants, Cr(VI) causes a wide range of DNA
damage similar to that observed in vivo, including DNA-
Cr(III) adducts, DNA-DNA cross-links, and DNA-protein
cross-links.2-5 The uptake-reduction model has been pro-
posed to account for the effects of Cr(VI) in vivo and in
vitro.5,6 In this model, tetrahedral [CrO4]2- enters the cell
readily via general anion transport channels. Once inside the
cell, Cr(VI) is reduced by cellular components producing
reactive intermediates, including Cr(VI) esters, Cr(V) and
Among the most important cellular molecules that are
thought to be involved in the reduction of Cr(VI) to Cr(III)
are the following: glutathione (γ-glutamylcysteinylglycine,
GSH); cysteine; ascorbate; catecholamines such as DOPA.4,5
Formation of relatively stable Cr(V) intermediates is com-
monly detected by EPR spectroscopy during the uptake of
Cr(VI) by cells or tissues,5,7,8 as well as during the in vitro
reduction of Cr(VI) with GSH,9 ascorbate10 and DOPA.11
Several researchers have shown that a direct correlation exists
between the concentration of such intermediates and the level
of DNA damage in vitro.2,3,11,12 The importance of Cr(V) as
a reactive intermediate in Cr(VI)-induced genotoxicity is
further increased in vivo by the formation of relatively stable
Cr(V) complexes with carbohydrates and glycoproteins
within the cells and on the cell surface.4,5,13
* Author to whom correspondence should be addressed. E-mail: p.lay@
chem.usyd.edu.au.
(1) IARC Monographs on the EValuation of the Carcinogenic Risk of
Chemicals to Humans; International Agency for Research on Cancer:
Lyon, France, 1990.
(2) Kortenkamp, A.; Casadevall, M.; Da Cruz Fresco, P.; Shayer, R. O.
J. NATO ASI Ser., Ser. 2 1997, 26, 15-34.
(7) Wetterhahn Jennette, K. J. Am. Chem. Soc. 1982, 104, 874-875.
(8) Liu, K. J.; Shi, X.; Jiang, J.; Goda, F.; Dalal, N.; Swartz, H. M. Ann.
Clin. Lab. Sci. 1996, 26, 176-184.
(3) Stearns, D. M.; Wetterhahn, K. E. NATO ASI Ser., Ser. 2 1997, 26,
55-72.
(4) Codd, R.; Dillon, C. T.; Levina, A.; Lay, P. A. Coord. Chem. ReV.
2001, 216-217, 533-577.
(9) Levina, A.; Zhang, L.; Lay, P. A. Inorg. Chem. 2003, 42, 767-784.
(10) Zhang, L.; Lay, P. A. J. Am. Chem. Soc. 1996, 118, 12624-12637.
(11) Pattison, D. I.; Davies, M. J.; Levina, A.; Dixon, N. E.; Lay, P. A.
Chem. Res. Toxicol. 2001, 14, 500-510.
(5) Levina, A.; Codd, R.; Dillon, C. T.; Lay, P. A. Prog. Inorg. Chem.
2003, 51, 145-250.
(6) Connett, P. H.; Wetterhahn, K. E. Struct. Bond. 1983, 54, 93-124.
(12) Casadevall, M.; Kortenkamp, A. Carcinogenesis 1995, 16, 805-809.
1044 Inorganic Chemistry, Vol. 44, No. 4, 2005
10.1021/ic048322h CCC: $30.25
© 2005 American Chemical Society
Published on Web 01/25/2005

Cr(V) Complexes of Dimethylurea Peptides
Proteins and peptides are the most abundant cellular
constituents and are also one of the most important classes
of biomolecules known to bind metal ions naturally.14 DNA-
Cr(III)-protein cross-links are among the most common
DNA lesions detected following the in vivo exposure to
Cr(VI).2-5 This observation implicates the involvement of
Cr-peptide intermediates in the Cr(VI)-induced DNA dam-
age. Despite this implication, relatively few studies have been
conducted on the coordination chemistry of Cr, particularly
in its higher oxidation states (IV-VI), with peptides (except
for GSH).9,15
Ligands with deprotonated amide nitrogen donors are
known to stabilize high oxidation states of metal ions.16 For
instance, stable Cr(V) complexes of two related macrocyclic
tetraamide ligands, H4mampa and H4mac, have been char-
acterized by X-ray crystallography.17 These square-pyramidal
oxo complexes ([CrV(O)L]-, where L ) mampa or mac)
were synthesized by the reactions of the deprotonated ligands
with CrCl2, followed by oxidation of the Cr(II) intermediates
with an organic peroxide.17
ligands react spontaneously with atmospheric oxygen in
solutions to produce M(III) complexes.20-22 This reaction is
immediately followed by ligand oxidation, leading to divalent
metal complexes of oxidized ligands.20-22 Substitution of
methylene hydrogens in the peptide ligands increases the
stability of the Cu(III) and Ni(III) oxidation states.16,23
Consequently, the ligand Aib3 (where Aib is a synthetic
amino acid, 2-amino-2-methylpropanoic or R-aminoisobu-
tyric acid), in which all the methylene hydrogens were
substituted for methyl groups, was used to prepare relatively
stable Cu(III)- and Ni(III)-peptide complexes,23 and the
crystal structure of one such complex, [CuIII(H-2Aib3)]‚
2H2O‚0.5NaClO4, has been solved.24
In this work, we report the syntheses and definitive
structural characterizations of Cr(V) complexes with trip-
eptide ligands, Aib3, AibLAlaAib, and AibDAlaAib, as well
as unusual covalent transformations of the ligands during
such syntheses.
Experimental Section
Caution! Chromium(V) compounds are mutagenic and possibly
carcinogenic;4,5 contact with skin and inhalation should be aVoided.
Chemicals. The following chemicals of analytical or higher
purity grade were used as received: benzyl alcohol; HCl (37% w/w
in H2O), H2SO4 (98% w/w in H2O), and CaCl2 from Ajax; L-alanine,
D-alanine, 2-amino-2-methylpropanoic acid (R-aminoisobutyric
acid), benzylchloroformate, tert-butyl hydroperoxide (70% w/v in
H2O), L-cysteine, N,N′-dicyclohexylcarbodiimide, potassium tert-
butoxide, 4-toluenesulfonic acid, CH3COOH, HBr in CH3COOH
(30% w/w), NaCNBH3, NaH, NaClO4, and NaOH (all Aldrich);
L-ascorbic acid, ferrocene, CrCl2 (anhydrous), KBr, NaCl, Na2SO4,
NaHCO3 Na2CO3, NaH2PO4‚2H2O, and Pd/C (10% w/w) (all
Merck); N-(2-hydroxyethyl)-piperazine-N′-(2-ethanesulfonic acid)
(HEPES, Research Organics); glutathione (reduced form, Sigma).
Tetra-n-butylammonium tetrafluoroborate (Fluka) was recrystallized
from ethyl acetate. Triethylamine (Aldrich) was freshly distilled
off CaH2 before use. The following solvents (HPLC grade, Aldrich)
were used as received: acetonitrile; acetone; benzene; N,N-
dimethylformamide (DMF); dimethyl sulfoxide (DMSO); hexa-
methylphosphoramide; sulfolane; ethanol; and methanol. Other
solvents, including diethyl ether, hexane, dichloromethane, and ethyl
acetate, were distilled off CaH2 before use. Water was purified by
the Milli-Q technique. Stock solutions used in the preparation of
the buffers (1.0 M of NaH2PO4 or HEPES) were treated with
Chelex-100 ion-exchange resin (BioRad) to remove traces of heavy
metal ions, and the pH values of the buffers were adjusted using
ultrapure NaOH (99.99%, Aldrich).
In addition, Cr(V) complexes of N-deprotonated glycine
or alanine peptides (tri-, tetra-, or pentamers) have been
generated via the reductions of Cr(VI) in methanol solutions
of the ligands18 or via the oxidations of the corresponding
Cr(III)-peptide complexes with Pb(IV) or I(VII) in aqueous
solutions.19 Six-coordinate octahedral structures ([CrV(O)-
L(OR)]2-, where L is a tetradentate ligand, coordinated in
the square plane, and R is H or Me) were assigned to these
complexes, on the basis of EPR spectroscopic and mass-
spectrometric studies.18,19
There has been a considerable work on the stabilization
of higher oxidation states of other metal ions, usually
Cu(III) or Ni(III), by deprotonated peptides or related
ligands.16,20-25 Certain Ni(II) and Cu(II) complexes of such
(13) Codd, R.; Irwin, J. A.; Lay, P. A. Curr. Opinion Chem. Biol. 2003, 7,
213-219.
(14) Hay, R. W. Bio-Inorganic Chemistry; Ellis Horwood: Chichester,
U.K., 1984.
(15) Levina, A.; Lay, P. A. Inorg. Chem. 2004, 43, 324-335.
(16) Margerum, D. W. Pure Appl. Chem. 1983, 55, 23-34.
(17) Collins, T. J.; Slebodnick, C.; Uffelman, E. S. Inorg. Chem. 1990,
29, 3433-3436.
Syntheses. The tripeptide R-aminoisobutyryl-R-aminoisobutyryl-
R-aminoisobutyric acid (Aib3) was prepared using the previously
published method.23 Similar methods were used for the syntheses
of two new tripeptides, R-aminoisobutyryl-L-alanine-R-aminoiso-
butyric acid (AibLAlaAib) and R-aminoisobutyryl-D-alanine-R-
aminoisobutyric acid (AibDAlaAib). These are outlined as follows.
(18) Headlam, H. A.; Lay, P. A. Inorg. Chem. 2001, 40, 78-86.
(19) Headlam, H. A. Ph.D. Thesis, The University of Sydney, 1998.
(20) Paniago, E. B.; Wetherburn, D. C.; Margerum, D. W. Chem. Commun.
1971, 1427-1428.
(21) Bossu, F. P.; Paniago, E. B.; Margerum, D. W.; Kirksey, S. T.; Kurtz,
J. L. Inorg. Chem. 1978, 17, 1034-1042.
(22) Weeks, C. L.; Turner, P.; Fenton, R. R.; Lay, P. A. J. Chem. Soc.,
Dalton Trans. 2002, 931-940.
(23) Kirksey, S. T.; Neubecker, T. A.; Margerum, D. W. J. Am. Chem.
Soc. 1979, 101, 1631-1633.
(24) Diaddario, L. L.; Robinson, W. R.; Margerum, D. W. Inorg. Chem.
1983, 22, 1021-1025.
Inorganic Chemistry, Vol. 44, No. 4, 2005 1045

Barnard et al.
The benzylcarbonyl (Cbz) and benzyl ester (OBz) protected peptide
CbzAlaAibOBz was synthesized from CbzAla and AibOBz by the
dicyclohexylcarbodiimide (DCC) method of amino acid coupling.26
The Cbz protecting group was removed from the dipeptide by
treating with HBr (30% w/w) in glacial CH3COOH.27 The am-
monium nitrogen of the formed AlaAibOBz was then deprotonated
using a saturated aqueous solution of Na2CO3 followed by extraction
of the dipeptide into dichloromethane. The dipeptide was then
coupled to CbzAib by the DCC method, leading to CbzAibAlaAi-
bOBz. The protecting groups were then removed by catalytic
hydrogenation on Pt/C,26 leading to free tripeptides, AibAlaAib.
Details of the syntheses and characterizations of the tripeptides are
given in the Supporting Information, including full assignments of
Na[CrV(O)(H-3AibDAlaAib-DMF)]‚H2O (3). This compound
was prepared using the same procedure as for 2, except that
AibDAlaAib‚H2O (1.0 g, 3.61 mmol) was used in the place of
AibLAlaAib. Yield of brown-black solid: 0.43 g (27.2%). ESMS
(-ve ion; MeCN): m/z 394.5. UV-vis: identical with that of 2.
CD: identical with that of 2 but with opposite signs for the ∆ꢀ
values. IR: identical with that of 2. Anal. Calcd for C14H24N4-
CrNaO7: C, 38.62; H, 5.56; N, 12.87. Found: C, 38.17; H, 5.40;
N, 12.48.
Mass-Spectrometric Characterization of the Ligand from the
Complex 1. A stock solution of 1 in H2O (∼1.0 mM) was freshly
prepared. This solution (0.50 mL) was mixed with aqueous solutions
of ascorbic acid (AsA, 0.50 mL, 40 mM) or L-cysteine (Cys, 0.50
mL, 40 mM). Complete reduction of Cr(V) to Cr(III) by AsA or
Cys (judged from the disappearance of the brown color of Cr(V))
was achieved in ∼1 min or ∼5 h, respectively (22 °C). Mass spectra
of the reaction mixtures were taken at ∼5 h after mixing. The pH
values of the reaction mixtures (measured at 5 h after the mixing)
were as follows: 6.6 for the aqueous solution; 4.5 for the solution
containing AsA; 6.2 for the solution containing Cys.
1
the H and 13C NMR signals (Figure S1 and Tables S1 and S2)
and full IR spectra (Figure S2).
Na[CrV(O)(H-3Aib3-DMF)]‚4.5H2O (1). In a N2-atmosphere
box, anhydrous CrCl2 (0.42 g, 3.1 mmol) was added to a
deoxygenated stirred solution of Aib3 (1.0 g, 3.1 mmol) in DMF
(50 mL). A purple precipitate rapidly formed, and potassium tert-
butoxide (1.1 g, 9.3 mmol) was added after 5 min, resulting in a
color change to gray/black with simultaneous dissolution of the
purple precipitate. At 15 min after dissolution of the precipitate,
aqueous tert-butyl hydroperoxide (70% w/v, 10 mL) was added,
and the mixture was stirred for 12 h at 22 °C. Further manipulations
were performed under ambient atmosphere. The reaction solution
was diluted to 100 mL with water, filtered through a Celite pad,
and loaded onto a Sephadex (DEAE A-25, Amersham) column (2.5
× 20 cm). The column was flushed with water (500 mL) to elute
any nonanionic species and DMF. An aqueous solution of NaCl
(0.20 M) was then used to elute a deep orange band, leaving a
bright yellow band on the column. The latter was eluted with 2.0
M NaCl and contained [CrO4]2- (λmax ) 372 nm, no EPR signals).28
The orange fraction was taken to dryness under reduced pressure,
and the solid was extracted with acetonitrile. The acetonitrile
solution was filtered and again taken to dryness, and the acetonitrile
extraction process was repeated. Yield of deep brown solid: 0.50
g (31.5%). ESMS (-ve ion; MeCN): m/z 408.4. UV-vis (MeCN)
[λ/nm (ꢀ/dm3 mol-1 cm-1)]: 300 (5500), 420 (1900). IR (solid
mixture with KBr; νmax/cm-1): 3400 w (broad), 2977 w, 2931 w,
2872 w, 1686 s, 1622 s, 1531 m, 1454 m, 1373 s, 1308 m, 1195
m, 1022 w, 972 m, 655 w, 479 m. Anal. Calcd for C15H33N4-
CrNaO10.5: C, 35.16; H, 6.49; N, 10.93. Found: C, 34.93; H, 6.57;
N, 10.36.
Analytical Techniques. Elemental analyses (C, H, N) were
performed by the Microanalytical Laboratory at the Australian
National University, Canberra.
EPR-spectroscopic measurements were carried out on a Bruker
EMX X-band spectrometer equipped with an ER 041XG microwave
bridge, an EMX 148T microwave bridge controller, an EMX 035M
NMR gaussmeter, an EMX 032T field controller, and an EMX 120
modulation amplifier. Spectra were recorded at 22 °C from solutions
contained in a quartz flat cell. EPR spectrometer operating
parameters were as follows: operating frequency, ∼9.7 GHz;
microwave power, 6.36 mW; modulation frequency, 100 kHz;
modulation amplitude, 0.5 G; center field, 3480 G; sweep width,
100 G; time constant, 40.96 ms; conversion time, 20.48 ms; receiver
gain, 1 × 104; number of scans, 10. Simulation of the EPR spectra
were performed using the program WinSim, and second-order
corrections were applied in the simulations.29
Electrospray mass spectrometry (ESMS) was performed using
a Finnigan LCQ mass spectrometer; experimental settings were as
follows: sheath gas (N2) pressure, 60 psi; spray voltage, 5.0 kV;
capillary temperature, 200 °C; capillary voltage, 19 V; tube lens
offset, 25 V; m/z range, 100-2000 (both in positive- and negative-
ion modes). Analyzed solutions were injected into a flow of H2O/
MeOH (1:1 v/v, flow rate 0.20 mL min-1). Acquired spectra were
the averages of 10 scans (scan time 10 ms). Simulations of the
mass spectra were performed using IsoPro software.30
Na[CrV(O)(H-3AibLAlaAib-DMF)]‚3.5H2O (2). This com-
pound was prepared using the same procedure as for 1, except that
AibLAlaAib (1.0 g, 3.86 mmol) was used in the place of Aib3. A
longer time (12 h) was required for the reaction of CrCl2 with the
ligand in deoxygenated DMF as, unlike Aib3, this ligand was only
sparingly soluble. Yield of brown-black solid: 0.42 g (22.6%).
ESMS (-ve ion; MeCN): m/z 394.5. UV-vis (MeCN) [λ/nm (ꢀ/
dm3 mol-1 cm-1)]: 289 (5.9 × 103), 398 (2.7 × 103). CD (MeCN)
[λ/nm (∆ꢀ/cm2 mmol-1)]: 237 (15.4), 274 (-7.4), 307 (1.1), 340
Vibrational spectra (IR) were recorded by the diffuse reflectance
technique on a Bio Rad FTS-40 spectrometer with KBr as the matrix
(ν ) 400-4000 cm-1). Electronic absorption (UV-vis) spectra
were recorded on a Varian Cary 5 spectrophotometer using a 1-cm
path length quartz cell (λ ) 200-800 nm). Circular dichroism (CD)
spectra were acquired on a JASCO 710 spectropolarimeter. The
instrument settings were as follows: wavelength range, 200-700
nm; scan rate, 100 nm min-1; response time, 0.25 s. The reported
CD spectra were the averages of five scans.
(2.4), 398 (-6.0), 488 (3.2). IR (solid mixture with KBr; νmax
/
cm-1): 3360 w (broad), 2976 w, 2934 w, 2873 w, 1681 s, 1620 s,
1456 m, 1372 s, 1344 m, 1311 m, 1204 w, 970 w, 900 w, 470 w.
Anal. Calcd for C14H29N4CrNaO9.5: C, 35.00; H, 6.08; N, 11.66.
Found: C, 34.95; H, 5.63; N, 11.18.
Cyclic voltammetry (CV) was performed using a BAS 100B
electrochemical analyzer. The complexes 1-3 were dissolved in
DMF with (nBu4N)BF4 (0.10 M) as the supporting electrolyte. A
three-electrode system with a glassy-carbon working electrode (3.0-
mm diameter), an aqueous Ag/AgCl reference electrode, and a Pt
(25) Hamburg, A.; Chih, H.; Getek, T. A. Inorg. Chem. 1985, 24, 2593-
2594.
(26) Greenstein, J. P.; Winitz, M. Chemistry of the Amino Acids; Wiley:
New York, 1961; Vol. 2.
(27) Ben-Ishai, D.; Berger, A. J. Org. Chem. 1952, 17, 1564-1570.
(28) Brasch, N. E.; Buckingham, D. A.; Evans, A. B.; Clark, C. R. J. Am.
Chem. Soc. 1996, 118, 7969-7980.
(29) Duling, D. R. J. Magn. Reson. 1994, B104, 105-110. The software
is available via the Internet at http://EPR.niehs.nih.gov/.
(30) Senko, M. IsoPro 3.0; Sunnyvale, CA, 1998.
1046 Inorganic Chemistry, Vol. 44, No. 4, 2005

Cr(V) Complexes of Dimethylurea Peptides
wire counter electrode were used. Full iR compensation was applied
for all scans. The ferrocenium/ferrocene (Fc+/0) couple was used
as an internal redox potential standard in all experiments.
Experimental Section) and would be expected to have
displaced the labile DMF ligand, which would be weakly
bound through the carbonyl oxygen.32 For similar reasons,
it is unlikely that DMF would be present as a solvent of
crystallization. Since the results of the elemental analyses
were inconsistent with the coordination of a simple depro-
tonated Aib3 ligand, a detailed spectroscopic analysis was
required to characterize the Cr(V) complexes.
The IR spectra of 1-3 (shown in Figure S2, Supporting
Information, in comparison with those of the corresponding
ligands) were consistent with the formation of pure Cr
complexes (since the spectra were similar for all of the
complexes, Figure S2). The spectra of the complexes were
distinguished from those of the ligands by the absence of
broad signals at ν ) 2500-3500 cm-1 (due to the N-H
stretching vibrations of the amide and amine groups) and
the presence of sharp signals at ∼970 cm-1, characteristic
for CrtO bonds (Figure S2).17,33 The latter feature supports
the formation of a Cr(V) oxo complex.
The negative-ion ESMS spectra of 1-3 displayed es-
sentially only one ion in each case, located at m/z ) -408.4
for 1 and -394.5 for 2 and 3; the isotope distribution patterns
were consistent with the formulations [C15H24N4O6Cr]- for
1 and [C14H22N4O6Cr]- for 2 and 3 (Figure 1). No significant
signals were observed at -m/z > 1000 or in the positive-
ion mode (m/z ) 100-2000). The ESMS data (Figure 1)
are consistent with the observations from ion-exchange
chromatography (Experimental Section), showing that the
isolated Cr(V) complexes are monoanionic species, as well
as with the results of elemental analyses, showing the binding
of a DMF molecule to each of the complexes 1-3. The
formulations of the complexes, [CrV(O)(H-7L-DMF)]- (where
L ) Aib3 or AibAlaAib), deduced from the ESMS data, point
to covalent binding of a DMF molecule to the ligand (see
Discussion), since the charge balance is inconsistent with
DMF binding as a neutral ligand to the Cr(V) center.
Typical EPR spectra of 1-3 (in MeCN, 22 °C) are shown
in Figure 2, a list of EPR spectral parameters (determined
by simulations of the spectra)29 is given in Table 1, and
typical simulation results are shown in Figure S3, Supporting
Information. The main feature in the EPR spectra of all three
compounds (Figure 2) was a seven-line signal centered at
giso ∼ 1.98; this multiplet pattern is consistent with the
coupling from three equivalent 14N nuclei (I ) 1). There
were also weak satellite peaks due to hyperfine splitting of
53Cr (9.55% abundance, I ) 3/2). The magnitudes of the
splitting (the Aiso(53Cr) values in Table 1) were consistent
with a metal-centered radical, confirming that the Cr atom
is d1 and in the formal oxidation state of Cr(V).33 The 14N
superhyperfine coupling constants obtained from the seven-
line signals (the aiso(14N) values in Table 1) observed here
are consistent with those expected from three spectroscopi-
cally equivalent amide 14N atoms.17,18,33 Superhyperfine
Results
Syntheses of the Ligands. Although several modifications
were reported in the literature for the synthesis of the
tripeptide Aib3,25,31 the original procedure of Margerum and
co-workers23 was mainly followed in this work. Important
variations, which improved the yields by ∼20%, were a short
(15 min) reaction time for the removal of the Cbz group
from CbzAibAibOBz and deprotonation of the resultant
AibAibOBz‚HBr prior to its use in the coupling reaction with
CbzAib (see the Supporting Information for details). The
same method was successfully used for the syntheses of new
tripeptides AibLAlaAib and AibDAlaAib.
Syntheses of the Cr(V) Complexes. The synthetic
procedure used in the preparation of the Cr(V) complexes
1-3 was based on the techniques used by Collins and co-
workers for the syntheses of Cr(V) complexes with tetra-
amide macrocyclic ligands H4mampa and H4mac.17 The use
of more polar peptide ligands in this work required the use
of a different solvent/base combination to form a Cr(II)-
ligand complex (DMF/KOBut instead of tetrahydrofuran/
tBuLi).17 Upon addition of CrCl2 to a DMF solution of the
tripeptide ligand, a purple-colored precipitate formed, sug-
gesting the formation of a Cr(II)-peptide intermediate
(possibly involving coordination through the carboxylate and
amine groups). A rapid color change was observed from
purple to deep gray-brown with simultaneous dissolution of
the purple precipitate when KOBut was added to the mixture,
indicating the formation of a deprotonated peptide complex
of Cr(II). Oxidation of the Cr(II) complex was performed
t
using an aqueous solution of BuOOH, which resulted in a
rapid color change from dark gray to deep orange-brown.
The Cr(V) complexes formed in this way were anionic and
were purified by anion-exchange chromatography.
Structures and Spectroscopy of the Cr(V) Complexes.
In the previously studied Cu(III) and Ni(III) complexes of
Aib3,16,23,24 the ligand was coordinated via the amine nitrogen,
two deprotonated amide nitrogens, and an oxygen of the
carboxylate group in a square-planar geometry. It was
anticipated that the Cr(V)-peptide complexes would display
the same ligand coordination with an oxo group in an axial
position.4,17,18 However, the data obtained from elemental
analyses for 1-3 were inconsistent with this expected
structure. Much better agreement between the observed and
calculated elemental compositions (C, H, N) was obtained
if a DMF molecule was incorporated into the chemical
formula, suggesting that a DMF molecule was coordinated
to the Cr center or had become covalently bound to the
peptide ligand or was a solvate of crystallization. It is
improbable that DMF would remain coordinated to Cr(V) if
the final complex was six-coordinate, since water was used
during its synthesis and chromatographic purification (see
(32) Dossing, A.; Engberg, P.; Hazell, R. Acta Chem. Scand. 1997, 51,
849-854.
(33) (a) Farrell, R. P.; Lay, P. A. Comments Inorg. Chem. 1992, 13, 133-
175. (b) Barr-David, G.; Charara, M.; Codd, R.; Farrell, R. P.; Irwin,
J. A.; Lay, P. A.; Bramley, R.; Brumby, S.; Ji, J.-Y.; Hanson, G. R.
J. Chem. Soc., Faraday Trans. 1995, 91, 1207-1216.
(31) Lockwood, M. A.; Blubaugh, T. J.; Collier, A. M.; Lovell, S.; Mayer,
J. M. Angew. Chem., Int. Ed. 1999, 38, 225-227.
Inorganic Chemistry, Vol. 44, No. 4, 2005 1047

Barnard et al.
Figure 1. Typical negative-ion ESMS data for 1 and 2 (∼1 mM in MeCN, ∼5 min after dissolution, 22 °C). The spectra for 3 were not significantly
different from those of 2. Experimental (lines) and simulated (columns, [C15H24N4O6Cr]- for 1 and [C14H22N4O6Cr]- for 2) isotopic distributions of the
main signals are also shown.
Table 1. EPR Spectroscopic Parameters of 1-3 (MeCN solutions, 22
°C, Figure 2)
param
complex 1
Species 1
100
1.9817
1.2
16.9
complex 2
complex 3
abundance, %
54.0
1.9822
1.1
14.8
2.36
0.65
58.6
1.9822
1.2
14.8
2.36
0.66
giso
line width, 104 cm-1
53
A
a
( Cr), 104 cm-1
iso
( N), 104 cm-1
2.36
14
iso
a
iso(1H), 104 cm-1
Species 2
abundance, %
46.0
1.9831
1.3
16.7
2.67
0.72
41.4
1.9830
1.3
16.7
2.69
0.74
giso
line width, 104 cm-1
53
A
a
a
( Cr), 104 cm-1
iso
14
( N), 104 cm-1
iso
iso(1H), 104 cm-1
couplings from amine nitrogens coordinated to Cr(V) are
often not observed and, when they are resolved, the
magnitude of the couplings are different from those due to
amide nitrogens.18,34 These results confirm that the coordina-
tion of the tripeptides to the Cr(V) center was different from
that of the previously reported Cu(III/II) and Ni(III/II)
complexes of Aib3.16,20,21,23,24
An excellent simulation (R2 ) 0.997) was obtained for
the EPR spectrum of 1 assuming coupling from three
equivalent amide nitrogens (14N) coordinated to a single Cr
Figure 2. Typical EPR spectra of freshly prepared (within ∼5-10 min)
solutions of 1 (17.2 mM), 2 (15.6 mM), and 3 (17.5 mM) in MeCN at 22
°C. The inset shows an expanded view of one of the signals due to 53Cr
hyperfine splitting. Spectral parameters are listed in Table 1.
(34) Weeks, C. L.; Levina, A.; Dillon, C. T.; Turner, P.; Fenton, R. R.;
Lay, P. A. Inorg. Chem. 2004, 43, 7844-7856.
1048 Inorganic Chemistry, Vol. 44, No. 4, 2005

Cr(V) Complexes of Dimethylurea Peptides
Figure 4. Cyclic voltammograms of 1-3 ([Cr] ) 1.0 mM, 22 °C) in
DMF with (nBu4N)BF4 (0.10 M) as the supporting electrolyte at a glassy
carbon electrode.
650 nm. The different electronic transitions for 2 and 3 were
more easily observed with the aid of CD spectra. The CD
spectra obtained for 2 and 3 (Figure 3b) were mirror images,
showing that the chirality of the ligands was retained in the
Cr(V) complexes and that 2 and 3 are enantiomeric. The
values of molar extinction coefficients in UV-vis and CD
spectra of 1-3 are listed in the Experimental Section.
Cyclic voltammetry of 1-3 in DMF solutions (in the
presence of (nBu4N)BF4 as a supporting electrolyte) showed
quasi-reversible Cr(V/IV) redox couples for all of the
complexes (Figure 4); the E1/2 values (vs Fc+/0) were -1.23
V for 1, -1.33 V for 2, and -1.34 V for 3, and the respective
peak-to-peak potential separations (∆Ep) were 67, 58, and
57 mV. While the ∆Ep values for 2 and 3 corresponded to
those of reversible redox reactions,35 the values of cathodic
and anodic peak currents for these two compounds were
unequal (ipc > ipa, Figure 4), which indicates that the Cr(IV)
counterparts of 2 and 3 are unstable and some decomposition
or reorganization of the Cr(IV) species occurs on the time
scale of the experiments. Under similar conditions (DMF
solutions, 0.10 M (Bu4N)BF4, 22 °C), 1-3 are significantly
weaker oxidants (E1/2 ) -1.23 to -1.34 V vs Fc+/0) than
the Cr(V) complexes with 2-hydroxycarboxylato or macro-
cyclic amido ligands, which can also be reduced quasi-
reversibly to Cr(IV) (E1/2 ∼ -0.9 V vs Fc+/0).36-38
Figure 3. UV-vis (a) and CD (b) spectra of freshly prepared (within
5-10 min) solutions of 1-3 in MeCN ([Cr] ) 0.35 mM, 22 °C). No
significant CD signals were observed for 1.
center (Figure S3). Three equivalent 14N nuclei were also
used in the EPR simulations for 2 and 3, taking into account
the additional coupling from the R-hydrogen of the Ala. The
1H superhyperfine coupling constants for 2 and 3 (Table 1)
were close to the values observed for Cr(V) diolato com-
plexes.33 Good fits (R2 ) 0.993 or 0.994, respectively) were
obtained only when two separate species were used to
simulate the observed spectra for 2 or 3 (Table 1 and Figure
S3), while significantly lower R2 values (0.941 or 0.952,
respectively) were obtained when only one species was used
(Figure S3). The optimized EPR parameters for the two
species were consistent between the spectra of 2 and 3 (Table
1), which suggests that the improvement of the fit for the
two-species model was not caused simply by an increase in
the number of varied parameters. By contrast, including a
second species into the simulations for 1 did not result in a
significant improvement of the fit (R2 ) 0.998, Figure S3).
The electronic absorption (UV-vis) spectrum for 1 in
MeCN (Figure 3a) was characterized by a relatively feature-
less absorption at 400-800 nm, and two poorly resolved
shoulders were evident at ∼420 nm and ∼300 nm. Identical
electronic absorption spectra were obtained for complexes
2 and 3 (Figure 3a), and these were similar to the spectrum
obtained for 1, although the resolved bands were more
prominent with a distinct maximum at 289 nm due to a
ligand-to-metal charge transfer (LMCT) transition and a
shoulder at 398 nm (d-d transitions) and a weak band at
Stability of the Cr(V) Peptide Complexes. The com-
plexes 1-3 are hygroscopic in the solid state, but samples
stored in a desiccator at 22 °C (protected from light) showed
no signs of decomposition (by UV-vis and EPR spec-
troscopies and ESMS) for at least 3 years. Solutions of 1-3
in DMF or DMSO ([Cr] ) 10 mM) or in MeCN ([Cr] )
1.0 mM) were stable for at least 1 month at 22 °C. Despite
the high stability of 1-3 in nonaqueous media, repeated
(35) Willard, H. H.; Merritt, L. L.; Dean, J. A.; Settel, F. A. Instrumental
Methods of Analysis, 7th ed.; Wadsworth: Belmont, CA, 1988.
(36) Judd, R. J.; Hambley, T. W.; Lay, P. A. J. Chem. Soc., Dalton Trans.
1989, 2205-2210.
(37) Dillon, C. T. Ph.D. Thesis, The University of Sydney, 1995.
(38) Levina, A. Unpublished results, 2003.
Inorganic Chemistry, Vol. 44, No. 4, 2005 1049

Barnard et al.
Table 2. Typical Results of Stability Studies for the Cr(V)-Peptide Complexes
complexa
buffer
pH
addition
% Cr(V) remainingb,c
% Cr(V) decomposd via eq 2c
1
1
1
1
1
2
3
phosphate
phosphate
phosphate
phosphate
HEPES
7.0
7.0
7.0
4.0
7.4
7.0
7.0
23
0
96
100
60
100
95
GSH (5.0 mM)
H2O2 (5.0 mM)
22
32
18
68
71
phosphate
phosphate
82
87
a In all cases, [Cr] ) 0.30 mM. Stock solutions of 1-3 in DMF ([Cr] ) 30 mM, prepared on the day of experiments) were used for the preparation of
the aqueous solutions of the complexes. b Reaction time 7 d at 22 °C. c Determined from the changes in UV-vis spectra of the solutions. The amounts of
remaining Cr(V) species and the Cr(VI) decomposition species were determined from their characteristic absorbances at 534 nm and 350-372 nm, respectively
(Figure S5).28,40,41 The proportion of Cr(V) decomposed by disproportionation (eq 1) was calculated from the amount of Cr(VI) generated.
attempts to grow X-ray-quality crystals of these compounds
have been unsuccessful, as yet.
with the formation of [CrO4]2-, detected from its character-
istic absorbance at λmax ) 372 nm28 (Table 2 and Figure
S5). Reactions of 1 with the most common biological
reductants (Red), glutathione (GSH), L-cysteine (Cys), or
ascorbic acid (AsA), in phosphate buffer (0.10 M, pH 7.0,
[Red] ) 5.0 mM, [Cr] ) 0.30 mM, 22 °C) led to a complete
reduction of Cr(V) to Cr(III) within ∼5 min (for AsA), ∼10
h (for Cys), or ∼30 h (for GSH), as determined from the
disappearance of the Cr(V) EPR signals and the changes in
UV-vis spectra (an example for the reaction with GSH is
shown in Figure S5).
The characteristic seven-line EPR signals of 1-3 (Figure
2 and Table 1) persisted for at least 1 week in aqueous
phosphate buffers (0.10 M, pH 7.0, [Cr] ) 0.30 mM) at 22
°C. A comparison of the EPR spectra for fresh and 1-week-
old aqueous solutions (Figure S4 in Supporting Information)
showed a significant decomposition of 1 during that time,
as evident from a decrease in signal-to-noise ratio and the
appearance of a new EPR signal (broad singlet, giso
)
1.9660). The absence of the 14N coupling for the latter signal
suggests that it is due to a hydrolysis product, where the
N-donors were fully or partially replaced with O-donors.33
The decomposition of 2 or 3 under the same conditions was
much less obvious (Figure S4).
Reduction of 1 by AsA or Cys in aqueous solutions leads
to the release of the free ligand, LH4 ) Aib3-DMF, as shown
-
by ESMS of the reduction products (m/z ) -343.2 for LH3 ,
+345.1 for LH4‚H+, and +367.2 for LH4‚Na+, Figure S6
and Table S3 in Supporting Information). No signals due to
the parent ligand (Aib3, fw ) 273.2) were observed (Figure
S6). Thus, ESMS data for the decomposed solutions of 1
strongly support the formation of a covalent bond between
the peptide ligand and a molecule of solvent (DMF) during
the synthesis of the Cr(V) complex.
The lower stability of 1 compared with 2 or 3 was
confirmed by UV-vis spectroscopy (Table 2; typical spectra
are shown in Figure S5, Supporting Information). Decom-
position in aqueous solutions for most of the known Cr(V)
complexes occurs via parallel disproportionation (eq 1) and
ligand/solvent oxidation (eq 2) routes, with the former
reaction typically predominant in basic media and the latter
process becoming more important with the decrease in the
pH value.9,34,39,40
Discussion
On the basis of the results of spectroscopic and analytical
studies, a molecular structure was assigned for the anionic
complexes in 1-3, in which a DMF solvent molecule has
become covalently bound to the amine group of the ligand,
to form a substituted urea group. The amine nitrogen of a
tripeptide is thus converted to a deprotonated amido nitrogen
that is coordinated to the Cr(V) center.
3Cr(V) f 2Cr(VI) + Cr(III)
(1)
CrVdO + S f Cr(III) + SdO (S ) organic substrate)
(2)
The UV-vis spectrum of the decomposed solution of 1
in aqueous phosphate buffer (0.10 M, pH 7.0, [Cr] ) 0.30
mM) after 7 days at 22 °C shows a general decrease in
absorbance at 300-800 nm compared with that of a fresh
solution (Figure S5), which is indicative of the predominant
formation of Cr(III) under these conditions.4,41 The data
contained in Table 2 suggest that complexes 1-3 decompose
mainly (>80%) via eq 2 (where the ligand acts as a
reductant) at the physiological range of pH values (4.0-
7.4)41 in the absence of added oxidants or reductants.
Addition of H2O2 (5.0 mM) to a solution of 1 (0.30 mM)
at pH 7.0 (0.10 M phosphate buffer, reaction time 7 days at
22 °C) leads to a partial oxidation of the Cr(V) complex
The EPR spectra for solutions of 2 or 3 in MeCN (Figure
2 and Table 1) that show the presence of at least two species
can be explained by the existence of two stereoisomers of 2
or 3, in which the methyl group of the Ala residue is located
(39) Krumpolc, M.; Rocˇek, J. Inorg. Chem. 1985, 24, 617-621.
(40) Levina, A.; Lay, P. A.; Dixon, N. E. Inorg. Chem. 2000, 39, 385-
395.
(41) Lay, P. A.; Levina, A. J. Am. Chem. Soc. 1998, 120, 6704-6714.
1050 Inorganic Chemistry, Vol. 44, No. 4, 2005

Cr(V) Complexes of Dimethylurea Peptides
Scheme 1. Proposed Mechanism for the Oxidative Coupling Reaction
either above or below the equatorial plane with respect to
the oxo ligand (2a,b or mirror images of these structures
for 3).
of DMF and Pyridine43
also act as ligands for the formed Cr(III) ions,41 leads to the
release of the modified peptide ligand, Aib3-DMF, which is
stable for at least several hours in aqueous solutions at pH
4-6 and 22 °C (as shown by ESMS of the reduction products
of 1, Figure S6 and Table S3).
A slight preference for one of the stereoisomers in
solutions, as detected by EPR spectroscopy (∼55% vs ∼45%,
Table 1), may be due to unfavorable steric interactions in
2b, where the oxo group and the methyl group are on the
same side of the ligand plane.
Unsuccessful attempts to generate Cr(V) complexes of
Aib3 and related ligands by oxidation of the corresponding
Cr(II) complexes in polar organic solvents other than DMF
(e.g., DMSO, sulfolane, or hexamethylphosphoramide)44
show that the replacement of the amine group in the peptide
ligands with an amido donor is crucial for the stabilization
of Cr(V). Previously, reduction of Cr(VI) in the presence of
peptide ligands (Glyn or Alan, n ) 3-5) or oxidation of the
corresponding Cr(III)-peptide complexes led to the forma-
tion of Cr(V)-peptide complexes, which were detectable by
EPR spectroscopy but were too unstable to isolate.18,19
Similarly, no stable Cr(V) complexes could be isolated for
a modified tripeptide ligand, N,N-dimethyl-AibAibAib, where
the site for the formation of a DMF adduct was blocked
(synthesis and characterization of this ligand are described
in the Supporting Information).44 It is likely that the
coordination of Cr(V) to the unprotected amino groups of
the peptide ligands was the main reason for the instability
of these complexes.18,19,44
The stability of 1-3 in neutral aqueous solutions (Table
2) is comparable with that of Cr(V) complexes with
macrocyclic tetraamide ligands37,42 and is much higher than
that of well-studied Cr(V) 2-hydroxycarboxylato complexes
(which persist for a few minutes at pH 7.0 and 22 °C).39,40
This high stability of 1-3 is in agreement with the observa-
tion that they are weak oxidants in nonaqueous media, as
shown by electrochemical studies (Figure 4). The higher
stability of 2 and 3, compared with 1, in aqueous media
(Table 2), is consistent with their lower reduction potentials
in DMF solutions (Figure 4). These observations contradict
the suggestion16,23 that the presence of methylene hydrogen
atoms in peptide ligands (such as AibAlaAib) will make them
more susceptible to the reactions with oxidizing metal centers
(such as Cr(V)), compared with the complexes of Aib3, where
such hydrogen atoms are absent. The stability of 1 may be
decreased due to the steric clashes of eight methyl groups
in the ligand (compared with seven such groups in 2 and 3).
To our knowledge, the conversion of the peptide ligands
in 1-3 to substituted urea derivatives represents a novel type
of organic transformation. Probably the closest analogy for
such a reaction is the oxidative coupling of amides (such as
DMF) with heteroaromatic compounds (such as pyridine)
in the presence of organic peroxides and Fe(II) compounds,
which is thought to proceed through the formation of free
carbamoyl radicals (Scheme 1).43
Although the syntheses of 1-3, involving the use of
Cr(II) and organic solvents (see Experimental Section), are
not compatible with biological conditions, the formation of
stable Cr(V) complexes with amido and carboxylato donors
(such as 1-3) is relevant to the proposed role of Cr(V) as a
major reactive species in Cr(VI)-induced genotoxicity.4,5,45
Reactions of Cr(VI) with intracellular reductants are likely
to lead to the formation of stable Cr(III) complexes with
proteins and peptides,5 which can be subsequently reoxidized
to Cr(V/VI) species by cellular oxidants,46 including H2O2-
producing enzymatic systems,47 and peroxides of proteins
or lipids (which are similar to tBuOOH in chemical proper-
ties).48 Such reoxidations are likely to be responsible for the
long-term toxicity caused by the exposure to Cr(VI), which
leads to accumulation of Cr(III) in cells.5,45,46 The possibility
Rapid formation of Cr(II) complexes during the reactions
of the tripeptide ligands with CrCl2 and the relatively high
yields of 1-3 (higher than those expected for a radical chain
reaction, similar to that in Scheme 1)43 suggest that the
oxidative coupling of DMF to the tripeptides is more likely
to occur via a Cr-directed process without the release of free
radical species into the reaction medium. Schemes 2 and 3
illustrate two alternative schemes that are plausible to explain
the formation of the Cr(V) product with the urea/peptide
ligands. Reduction of 1 with L-cysteine or ascorbate, which
(44) Barnard, P. J. Ph.D. Thesis, University of Sydney, 2002.
(45) Levina, A. Lay, P. A. Coord. Chem. ReV. 2005, 249, 281-298.
(46) Dillon, C. T.; Lay, P. A.; Bonin, A. M.; Cholewa, M.; Legge, G. J. F.
Chem. Res. Toxicol. 2000, 13, 742-748.
(47) Mulyani, I.; Levina, A.; Lay, P. A. Angew. Chem., Int. Ed. 2004, 43,
4504-4507.
(48) Knott, H. M.; Baoutina, A.; Davies, M. J.; Dean, R. T. Arch. Biochem.
Biophys. 2002, 400, 223-232.
(42) Dillon, C. T.; Lay, P. A.; Bonin, A. M.; Dixon, N. E.; Collins, T. J.;
Kostka, K. L. Carcinogenesis 1993, 14, 1875-1880.
(43) Arnone, A.; Cecere, M.; Galli, R.; Minisci, F.; Perchinunno, M.; Porta,
O.; Gardini, G. Gazz. Chim. Ital. 1973, 103, 13-29.
Inorganic Chemistry, Vol. 44, No. 4, 2005 1051

Barnard et al.
Scheme 2. Proposed Mechanism for the Oxidative Coupling of DMF with a Tripeptide Ligand during the Syntheses of 1-3 in Which the Coordinated
DMF Radical Reacts with the Amine of the Peptide Ligand
Scheme 3. Proposed Mechanism for the Oxidative Coupling of DMF with a Tripeptide Ligand during the Syntheses of 1-3 in Which the Coordinated
DMF Radical Reacts with the Deprotonated Amine of the Peptide Ligand
of formation of stable (and potentially DNA-damaging)4,5
Cr(V)-peptide complexes as a result of cellular reoxidation
of Cr(III)-peptide complexes is supported by the results of
this work. Synthesis and characterization of the Cr(V)-
peptide complexes 1-3 (which are the first isolated Cr(V)
complexes with non-sulfur-containing peptide ligands)9,49 will
1052 Inorganic Chemistry, Vol. 44, No. 4, 2005

Cr(V) Complexes of Dimethylurea Peptides
Supporting Information Available: Details of syntheses and
characterizations of the ligands Aib3, AibLAlaAib, AibDAlaAib,
and N,N-dimethyl-AibAibAib, IR spectra of 1-3 and the corre-
sponding ligands, typical results of simulations of EPR spectra for
1-3, typical results of stability studies of 1-3 in aqueous solutions
(EPR and UV-vis spectra), and ESMS data for the reduction
products of 1 in aqueous solutions (typical spectra and assignment
of the signals). This material is available free of charge via the
Internet at http://pubs.acs.org.
facilitate the studies of biological activities of such com-
pounds.
Acknowledgment. The Australian Research Council
(ARC) is gratefully acknowledged for the funding for an
ARC Professorial Fellowship to P.A.L., for Large and
Discovery grants, and for RIEF and LIEF equipment grants
for the NMR, EPR, and ESI-MS instrumentation.
(49) Lay, P. A.; Levina, A. In ComprehensiVe Coordination Chemistry II;
McCleverty, J. A., Meyer, T. J., Eds.; Elsevier: Oxford, U.K., 2004;
Vol. 4, pp 313-413.
IC048322H
Inorganic Chemistry, Vol. 44, No. 4, 2005 1053