Addition of (pyrazol-1-yl)acetyl and (pyridin-2-yl)acetyl groups to the terminal amino group of a Phe-Gly dipeptide affords ATCUN-like copper(II) binding sites

Article abstract of DOI:10.1002/ejic.200400437

(Pyrazol-1-yl)acetic acid (HL¹, 1) and (pyridin-2-yl)acetic acid have been successfully condensed with the terminal amino group of a Phe-Gly dipeptide to generate the derivatives PzCH₂-Phe-Gly (H ₃L², 7) and PyCH₂-Phe-Gly (H₃L ³, 8) respectively, which contain ATCUN-like metal ion binding sites. The crystal structures of 1 and the copper(II) complexes Cu(L¹) ₂(H₂O)₂ (2) and [nBu₄N][Cu(L ²)] (9) are described. Complex 2 has a tetragonally-elongated, Jahn-Teller distorted octahedral geometry and complex 9 is square-planar, The structural similarity of the copper(II) complexes of 7 and 8 is demonstrated by their almost identical electronic absorption and EPR spectra. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejic.200400437

FULL PAPER
Addition of (Pyrazol-1-yl)acetyl and (Pyridin-2-yl)acetyl Groups to the
Terminal Amino Group of a PheϪGly Dipeptide Affords ATCUN-Like
Copper(II) Binding Sites
Andrew N. Boa,*[a] Jonathan D. Crane,*[a] Radoslaw M. Kowalczyk,[b] and
Nayer H. Sultana[a]
Keywords: Bioinorganic chemistry / Copper / Peptides / Tetradentate ligands
(Pyrazol-1-yl)acetic acid (HL1, 1) and (pyridin-2-yl)acetic
acid have been successfully condensed with the terminal
amino group of a PheϪGly dipeptide to generate the deriva-
tives PzCH2–PheϪGly (H3L2, 7) and PyCH2–PheϪGly (H3L3,
8) respectively, which contain ATCUN-like metal ion binding
sites. The crystal structures of 1 and the copper(II) complexes
Cu(L1)2(H2O)2 (2) and [nBu4N][Cu(L2)] (9) are described.
Complex 2 has a tetragonally-elongated, Jahn–Teller dis-
torted octahedral geometry and complex 9 is square-planar.
The structural similarity of the copper(II) complexes of 7 and
8 is demonstrated by their almost identical electronic absorp-
tion and EPR spectra.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
CUN motif are able to mediate oxidative DNA damage,
including cleavage and cross-linking, and (ii) even simple
ATCUN-like complexes display selective binding to, and
cleavage of, DNA and RNA.[1,2] Indeed this has inspired
the synthesis and study of ATCUN-containing proteins as
selective DNA binding/cleavage systems.[2,5,6] In addition,
polydentate ligands predominantly based upon the AT-
CUN motif have been demonstrated to act as highly selec-
tive, fluorescent chemosensors for copper(ii).[8,9]
We report herein the synthesis of novel copper(ii) binding
peptides (Scheme 1) that do not contain a His residue
within the amino-terminal protein sequence. In the peptide
systems (7 and 8) the heterocyclic N-donor ligand for the
metal ion is provided by the straightforward condensation
of (pyrazol-1-yl)acetic acid (1) or (pyridin-2-yl)acetic acid
with the amino terminal of the peptide. In principle, this
approach allows the generation of an ATCUN-like site at
the amino terminal of any dipeptide. In the specific exam-
ples described here the chosen starting dipeptide was
PheϪGly and in the resulting tetradentate systems the ter-
minal carboxylate group acts as the fourth ligand to the
bound, square-planar copper(ii) ion.
Introduction
The amino terminal copper(ii) and nickel(ii)-binding pro-
tein motif (H2N)–Xaa–Xaa–His (ATCUN), where Xaa is
an α-amino acid other than Pro, occurs naturally in albu-
mens (human, bovine and rat), histatins, human sperm
protamine P2a, and neuromedins (C and K).[1] In the albu-
mens its prominent role is proposed to be the binding and
transport of metal ions, usually copper(ii) and nickel(ii).
These ions have been the most extensively studied for AT-
CUN and ATCUN-like systems.[2–6] The presence of the
ATCUN motif in the other proteins raises the possibility
that the binding of metal ions may have some influence,
presumably adverse, upon the normal function of these sys-
tems. It is also worthy of note that although the copper(ii)
binding sites identified in the prion protein of Creutzfeld–
Jakob disease (CJD) are not terminal, they are nevertheless
structurally very similar to the ATCUN motif, and it has
been proposed that this may indicate the function of the
normal cellular form of this protein is copper(ii) binding,
storage and/or transport.[7]
The investigation of the role of the ATCUN and related
motifs in biological systems is well established, but of par-
ticular chemical importance are: (i) The observations that
both the copper(ii) and nickel(ii) bound forms of the AT-
Results and Discussion
(Pyrazol-1-yl)acetic acid HL1 (1) was prepared in good
yield by the alkylation of pyrazole with bromoacetic acid,
and X-ray quality crystals were obtained by recrystallis-
ation from water. The structure of 1 (Figure 1) reveals that
the pyrazole group is not sufficiently basic for the com-
pound to crystallise in a zwitterionic form; the freely refined
acidic proton H(1) is clearly located on the carboxylic acid
[a]
Department of Chemistry, University of Hull,
Cottingham Road, Kingston-upon-Hull, HU6 7RX, UK
Fax: +44-1482-466410
E-mail a.n.boa@hull.ac.uk
[b]
EPSRC Multi-frequency EPR Service Centre, Department of
Chemistry, The University of Manchester,
Manchester M13 9PL
Supporting information for this article is available on the
WWW under http://www.eurjic.com or from the author
872
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400437
Eur. J. Inorg. Chem. 2005, 872–878

Formation of ATCUN-Like Copper(ii) Binding Sites
FULL PAPER
similar to that reported for the substituted analogue 3,5-
dimethyl-(pyrazol-1-yl)acetic acid.[10]
The complexation of 1 with copper(ii) yields the Jahn–
Teller distorted, centrosymmetric, octahedral structure
Cu(L1)2(H2O)2 (2), the molecular structure of which is
shown in Figure 2. The ligand bite angle is 90.58(7)° and
the neutral pyrazole Cu(1)–N(1) bond length of 1.954(2) Å
is significantly shorter than the Cu(1)–O(1) bond length of
1.980(2) Å involving the anionic carboxylate ligand. This is
consistent with the delocalisation of the carboxylate nega-
tive charge over both oxygen atoms as evinced by the sim-
ilar C(5)–O(1) and C(5)–O(2) bond lengths of 1.274(3) and
1.246(2) Å respectively [∆d = 0.028(4) Å], and this delocalis-
ation is stabilised by the presence of two hydrogen-bonds to
the non-coordinated carboxylate oxygen atom O(2). Upon
complexation the dihedral angle between the least-squares
planes of the pyrazole and carboxylic acid groups has
changed from 80.76(6)° in the free ligand (1) to 32.35(5)° in
2. Due to the non-planarity of the ligand the dihedral angle
between the pyrazole group and the copper square-plane is
22.55(4)°. The electronic absorption spectrum of 2 in aque-
ous solution is typical of axially elongated octahedral cop-
per(ii) with a single broad absorption band at 14700 cm–1
(ε = 32 m–1 cm–1). There are no reported complexes of either
(pyrazol-1-yl)acetic acid (1) or its simple derivatives in the
literature, however several transition metal complexes
(VO2+, Mn2+, Fe2+, Co2+, Ni2+ and Cu2+) of the similar
ligand imidazol-4-yl-acetic acid have been described.[11–16]
The crystal structures of many of these have been re-
ported,[17] but surprisingly not for Cu2+. Furthermore, there
is only one unequivocally characterised transition metal
complex (Pt2+) of (pyridin-2-yl)acetic acid or its simple de-
rivatives.[18] To date, attempts to prepare and characterise
other first row transition metal complexes of 1 have been
unsuccessful, thus at present it is difficult to compare di-
Scheme 1
group with a strong intermolecular hydrogen bond to N(1)
of a neighbouring molecule. Thus 1 crystallises as parallel,
1-D, hydrogen-bonded chains that are helical and homochi-
ral (non-centrosymmetric space group P212121). The least-
squares planes of the pyrazole and carboxylic acid groups
are almost mutually perpendicular with a dihedral angle of
80.76(6)° and N(2) lies only 0.0276(14) Å out of the plane
of the carboxylic group. In all these regards the structure is
Figure 2. ORTEP[25] view of the molecular structure of 2 with ther-
mal ellipsoids shown at 50 %. Selected bond lengths [Å] and angles
[°]: Cu(1)–O(1), 1.980(2); Cu(1)–N(1), 1.954(2); Cu(1)–O(3),
2.503(2); O(1)–Cu(1)–N(1), 90.52(4); O(1)–Cu(1)–O(3), 92.57(4);
N(1)–Cu(1)–O(3), 93.60(4); O(3)–H(3d), 0.80(2); H(3d)···O(2)i,
Figure 1. ORTEP[25] view of the molecular structure and hydrogen-
bonding of 1 with thermal ellipsoids shown at 50 %. Selected bond
lengths [Å] and angles [°]: O(1)–C(5), 1.309(2); O(2)–C(5), 1.207(2); 2.11(2); O(3)···O(2)i, 2.886(1); O(3)–H(3d)···O(2)i, 167(2); O(3)–
O(1)–H(1), 0.95(3); H(1)···N(1)i, 1.69(3); O(1)···N(1)i, 2.628(2); H(3e), 0.77(2); H(3e)···O(2)ii, 2.03(2); O(3)···O(2)ii, 2.801(1); O(3)–
O(1)–H(1)···N(1)i, 170(3). Symmetry transformations: [i] –x, y + H(3e)···O(2)ii, 176(2). Symmetry transformations: [i] x + ½, –y +
½, –z + 1½; [ii] x, y + 1, z.
1½, z + ½; [ii] –x + 1½, y – ½, –z – ½.
Eur. J. Inorg. Chem. 2005, 872–878
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
873

A. N. Boa, J. D. Crane, R. M. Kowalczyk, N. H. Sultana
FULL PAPER
rectly the coordination behaviour of these related ligand pale violet and the absorption band shifts to 18000 cm–1.
systems.
This shift of 1500 cm–1 to lower energy is similar to that
The two tetradentate square planar ligands H3L2 (7) and observed for the 1,4-diazacycloheptane (LL) complexes
H3L3 (8) were prepared in two steps, from 1 and (pyridin- [Cu(LL)2]2+ and [Cu(LL)2(H2O)]2+ [19]
and may therefore
,
2-yl)acetic acid, respectively, by standard peptide coupling indicate the coordination of one water ligand to give a five
with the dipeptide H–PheϪGlyϪOBn, followed by hydroly- coordinate, square pyramidal complex in aqueous solution.
sis of the benzyl ester. The complexation of 7 with cop-
per(ii) yields the deep pink, square-planar complex [Cu(L2)]–
as the tetra-n-butylammonium salt 9. The crystal structure
of the complex anion of 9 (Figure 3) reveals that the cop-
per(ii) centre is close to planar; the sum of the angles sub-
tended at the metal is 360.0° and the maximum deviation
from the least-squares coordinate plane for Cu(1) and its
four donor atoms is only 0.017(2) Å for N(4). There is a
very slight tetrahedral distortion of the Cu(1) coordination
geometry with the pairs of opposite donor atoms lying on
opposite sides of the least-squares plane and N(1)–Cu(1)–
N(4) and N(3)–Cu(1)–O(3) angles of 178.79(6) and
168.71(5)° respectively. As expected, the shortest bond
lengths in the copper coordination sphere involve the de-
protonated amide nitrogen atoms. The strength of these
bonds is further illustrated by the short C=O [average
1.244(2) Å] and long C–N [average 1.324(2) Å] bond
lengths within the amide groups showing very little delocal-
isation of the formal negative charge onto the oxygen atom,
and hence relatively little contribution to the electronic
structure by the oxy-imine [RN=C(RЈ)–O–] resonance form.
Due to the square-planar geometry of the copper centre in
9 and the consequent absence of any axial ligand interac-
9 with thermal ellipsoids shown at 50 %. Selected bond lengths
Figure 3. ORTEP[25] view of the structure of the complex cation of
tions, the neutral pyrazole Cu(1)–N(1) bond length of
1.9321(13) Å is shorter than in 2, despite being trans to the
shortest Cu(1)–N(amide) bond. The carboxylate Cu(1)–
O(3) bond length is the longest in the coordination sphere,
yet this is also shorter than the similar bond in 2. In ad-
dition the carboxylate C(8)–O bond lengths are more dis-
similar [∆d = 0.067(3) Å] than in 2, consistent with the ab-
[Å] and angles [°]: Cu(1)–O(3), 1.9606(12); Cu(1)–N(1), 1.9321(13);
Cu(1)–N(3), 1.9299(13); Cu(1)–N(4), 1.8654(13); O(1)–C(5),
1.246(2); N(3)–C(5), 1.325(2); O(2)–C(7), 1.242(2); N(4)–C(7),
1.323(2); O(3)–C(9), 1.298(2); O(4)–C(9), 1.231(2); N(1)–Cu(1)–
N(3), 94.42(5); N(3)–Cu(1)–N(4), 84.66(6); O(3)–Cu(1)–N(4),
84.09(6); O(3)–Cu(1)–N(1), 96.85(5).
The (pyridin-2-yl)acetyl analogue, 10, could be prepared
sence of any stabilisation of negative charge on O(4) by hy- as a pink complex by the same method as used for 9. Unfor-
drogen bonding. Due to the non-planar nature of the (pyr- tunately, a pure or crystalline sample of 10 could not be
azol-1-yl)acetyl group the dihedral angle between the pyr- obtained due to its slow decomposition during attempted
azole group and the copper square-plane is 17.81(5)°, sim- recrystallisation, resulting in a pale green solution. Elec-
ilar to that found in 2. The phenyl substituent of the ligand tronic absorption spectra of 10 in dichloromethane/meth-
is positioned above one face of the copper square plane in anol (3:1) and water were recorded for freshly prepared
the solid state and the angle between their respective least- samples and were found to be almost identical to the corre-
squares planes is 61.59(5)°. Of the three possible staggered sponding spectra of 9, consistent with the formation of a
conformations about the C(6)–C(10) bond, this is the one structurally very similar complex. Additionally, the negative
that minimises the unfavourable steric interactions between ion electrospray mass spectrum of complex 10 in dichloro-
the phenyl ring and the carbonyl oxygen atoms O(1) and methane solution revealed the desired molecular ion. For
O(2). The Flack parameter determined from the structure further confirmation of the structure of 10, and for com-
refinement confirms the known (S)-chirality of the ligand parison with 9, the EPR spectra of both were measured at
at C(6), being derived from (S)-phenylalanine. The pink col- X-band. Complexes 9 and 10 gave near identical spectra in
our of 9 is retained in dichloromethane solution and arises the same solvents systems (Figure 4).
from a single absorption band at 19500 cm–1 in the visible
At room temperature the Cu hyperfine (to both iso-
region of the electronic absorption spectrum. There is no topes), as well as ligand superhyperfine splittings to 14N
1
significant change to the spectrum either in dichlorometh- and H can be resolved, in both aqueous and in CH2Cl2
ane/methanol (1:1) or upon the addition of ten equivalents solution. In frozen solution at ca. 110 K (adding ca. 10 %
of a good N-donor ligand (N-methylimidazole), indicating glycerol to H2O, or ca. 10 % toluene to CH2Cl2, to aid
that the square-planar geometry is retained under these glassing) we observed well resolved spectra which are axial
conditions. In aqueous solution however, the complex is within the resolution of the X-band experiment. The 14N
874
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 872–878

Formation of ATCUN-Like Copper(ii) Binding Sites
FULL PAPER
Figure 4. Fluid solution EPR spectra at 290 K and 9.773 GHz (left) and frozen solution EPR spectra at 120 K and 9.457 GHz (right).
(a) 9 in H2O/glycerol; (b) 10 in H2O/glycerol; (c) 9 in CH2Cl2/toluene; (d) 10 in CH2Cl2/toluene.
Table 1. Simulation of EPR parameters for complex 9 (A in Gauss). Complex 10 gave nearly identical spectra in the same solvent systems.
Cu
3N
2/3H
Cu
Cu
3N
Solvent
giso
Aiso
Aiso
Aiso
gz
gxy
Az
Axy
Axy
H2O
CH2Cl2
2.104
2.088
79
90
14
16
8
8
2.206
2.165
2.040
2.050
190
216
20
35
16
16
Table 2. Comparison of visible and EPR parameters for complexes 9 and related CuII complexes (data taken from reference 1. *pH =
6.5; HSA, human serum albumin; DSA, dog serum albumin; A values in Gauss).
Cu
Complex*
λmax. (nm)
εmax. (m–1 cm–1)
gz
gxy
Az
9 (H2O)
555
513
525
600 (br)
525
110
135
101
Ϫ
103
103
2.206
2.165
2.166
2.256
2.170
2.167
2.040
2.050
2.051
2.059
2.051
2.050
190
216
214
163
211
211
9 (CH2Cl2)
HSAϪCuII
DSAϪCuII
GlyϪGlyϪHisϪCuII
AspϪAlaϪHisϪCuII
525
coupling in the “perpendicular” region of the spectra is also switch to a square pyramidal structure for complex 9 in
resolved. The frozen solution EPR spectra were modeled water is clearly revealed with the expected changes in gz,
Cu
assuming coupling of the unpaired electron with Cu and to Az and λmax. Dog serum albumin does not show the AT-
three equivalent in-plane coordinated 14N nuclei (equivalent CUN motif as it does not contain histidine as the third
within the resolution of the EPR experiment). These simu- residue from the N terminus. The EPR and visible absorp-
lations were sensitive to both the 14N and Cu hyperfine tion data for this albumin is clearly divergent from the other
coupling in the “perpendicular” region (the latter having complexes.
been estimated initially from the values of AzCu and AisoCu).
In conclusion, the condensation of (pyrazol-1-yl)acetic
In order to simulate the isotropic spectra it was also neces- acid and (pyridin-2-yl)acetic acid with the terminal amino
1
sary to include coupling to H (this is not resolved in the group of a dipeptide has been shown to generate copper(ii)
anisotropic spectra). Simulation parameters are in Table 1. ion binding sites. The extension of this methodology to
Cu
The large values of Az are consistent with the N3O in- larger oligopeptides, the investigation of the chemical reac-
plane coordination, and the AN values are typical for (in- tivity of the metal centre, the cytotoxicity of these com-
plane) N-bound Cu complexes. The important trends are plexes and their interaction/reaction with DNA are in pro-
that gz increases in switching from CH2Cl2 to H2O solution gress.
Cu
and Az (and AxyCu) decreases in going from CH2Cl2 to
H2O solution. These trends are consistent with going from
a square-planar complex in dichloromethane to square py- Experimental Section
ramidal species in water and thus support the conclusions
drawn from the electronic spectra. The EPR spectroscopic
data from 9 in dichloromethane compared with data from
human serum albumin, GlyϪGlyϪHisϪCuII and AspϪAl-
aϪHisϪCuII are very similar, and are consistent with
The protected dipeptide [(S)-2-tert-butoxycarbonylamino-3-phenyl-
propanoylamino]acetic acid benzyl ester (BocϪPheϪGlyϪOBn, 3)
was prepared using standard procedures.[20] Elemental analyses
were obtained with a Fisons EA1108 CHNS analyser. IR spectra
were recorded as KBr disks with a Perkin–Elmer Paragon-100 FT-
square-planar coordination in these systems (Table 2). The IR spectrophotometer. Electronic absorption spectra were recorded
Eur. J. Inorg. Chem. 2005, 872–878
www.eurjic.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 875

A. N. Boa, J. D. Crane, R. M. Kowalczyk, N. H. Sultana
FULL PAPER
in the range 1100–300 nm with a Perkin–Elmer Lambda-40 spec-
trophotometer. NMR spectra were recorded with a Jeol JNM-
LA400 spectrometer at room temperature (20 °C) and are reported
relative to tetramethylsilane. Mass spectra were recorded with a
Finnigan MAT-1020 or Thermo Finnigan LCQ ion trap spectrom-
eter. EPR spectra were measured with a Bruker EMX X-band spec-
trometer.
7.3 Hz, 2 H, PhCH2), 3.98 (ABX, 2JH,H = 18.3, 3JH,H = 5.2, 5.6 Hz,
3
2 H, CH2NH), 4.68 (br. q, JH,H Ϸ 7.3 Hz, 1 H, CHCH2), 4.75
2
(AB, JH,H = 16.8 Hz, 2 H, CH2Pz), 5.14 (s, 2 H, CH2O), 6.30 (t,
3
3JH,H Ϸ 2 Hz, 1 H, PzH), 6.63 (br. t, JH,H Ϸ 5.6 Hz, 1 H, NH),
3
6.77 (br. d, JH,H Ϸ 7.8 Hz, 1 H, NH), 7.04–7.06 (m, 2 H, PhH),
7.17–7.26 (m, 3 H, PhH), 7.32–7.39 (m, 6 H, PhH + PzH), 7.58 (d,
3JH,H Ϸ 2 Hz, 1 H, PzH) ppm. 13C NMR (100.6 MHz, CDCl3): δ
= 37.7, 41.3, 54.3, 54.7, 67.2, 106.8, 127.0, 128.4, 128.5(7), 128.6(3),
128.6(5), 129.2, 131.1, 135.1, 136.1, 141.3, 167.5, 169.2, 170.5 ppm.
MS (EI): m/z (%) = 420 (1) [M+]. HRMS found: [M]+ 420.1798;
C23H24N4O4 requires [M]+ 420.1798.
(Pyrazol-1-yl)acetic Acid (HL1, 1): Sodium hydroxide (16.2 g,
0.404 mol) was dissolved in water (185 mL) and pyrazole (12.5 g,
0.184 mol) was added. Next bromoacetic acid (28.1 g, 0.202 mol)
was added in portions with stirring and the resulting mixture then
heated at reflux for 2 hours. After cooling, the mixture was care-
fully acidified (2 m HCl) to about pH 3 whereupon the product
crystallized from the solution. After drying under vacuum, 1 was
isolated as a pale yellow crystalline solid. Yield: 15.26 g (67 %). 1H
NMR (400 MHz, CDCl3): δ = 4.0–4.5 (br. s, 1 H, OH), 4.95 (s, 2
Benzyl
{(S)-3-Phenyl-2-[(2-pyridin-2-yl)acetylamino]propionyl-
amino}acetate (2-PyCH2CO–PheϪGlyϪOBn, 6): A sample of the
dipeptide salt 4 (5.0 g, 12.0 mmol) was coupled to (pyridin-2-yl)
acetic acid hydrochloride (2.04 g, 12.0 mmol) as described above to
1
give the title compound as a white solid. Yield: 3.40 g (67 %). H
3
2
3
H, CH2), 6.25 [dd, J(H,H) = 2.3 and 1.8 Hz, 1 H, PzH], 7.42 (d,
NMR (400 MHz, CDCl3): δ = 3.10 (ABX, JH,H = 14.0, JH,H
=
3
3JH,H = 1.4 Hz, 1 H, PzH), 7.70 (d, JH,H = 2.3 Hz, 1 H, PzH)
5.9, 7.9 Hz, 2 H, PhCH2), 3.66 (AB, 2JH,H = 15.7 Hz, 2 H, CH2Py),
4.01 (ABX, JH,H = 18.2, JH,H = 5.0, 5.9 Hz, 2 H, CH2NH), 4.77
(br. q, JH,H Ϸ 8.0 Hz, 1 H, CHCH2), 5.14 (s, 2 H, CH2O), 7.09–
7.20 (m, 8 H, ArH), 7.30–7.38 (m, 5 H, ArH), 7.60 (td, JH,H
ppm. IR (KBr): ν = 3423, 2509, 1733, 1071, 780 cm–1. MS (EI):
2
3
˜
m/z (%) = 126 (20) [M+].
3
3
=
Synthesis of the Copper(II) Complex Cu(L1)2(H2O)2 (2): A solution
of copper(ii) acetate (0.39 g, 2 mmol) in water (15 mL) was added
to a solution of (pyrazol-1-yl)acetic acid (1, 0.50 g, 4 mmol) in
water (25 mL) with stirring. The blue solution was filtered and the
solvent was evaporated slowly. Large, blue, X-ray quality crystals
of the complex 2 deposited. Yield: 0.39 g (56 %). C10H14CuN4O6
(349.79): calcd. C 34.34, H 4.03, N 16.02; found C 34.11, H 3.98,
N 15.86. UV/Vis (H2O): λ = 14700 cm–1 (32 m–1 cm–1). IR (KBr):
4
3
7.6, JH,H = 1.7 Hz, 1 H, ArH), 7.76 (br. d, JH,H Ϸ 7.9 Hz, 1
H, NH), 8.41 (br. d, JH,H Ϸ 4.8 Hz, 1 H, ArH) ppm. 13C NMR
3
(100.6 MHz, CDCl3): δ = 37.4, 41.4, 44.7, 54.4, 67.1, 122.1, 124.0,
126.7, 128.3, 128.5 (two peaks), 128.6, 129.2, 135.2, 136.6, 137.2,
149.0, 155.1, 169.4, 169.6, 171.3 ppm. MS (EI): m/z (%) = 431 (2)
[M+], 340 (20), 239 (45), 120 (100).
{(S)-3-Phenyl-2-[(2-pyrazol-1-yl)acetylamino]propionylamino}acetic
Acid (H3L2, 7): The modified tripeptide 5 (1.0 g, 2.38 mmol) was
dissolved in H2O/ethanol (1:1, 8.0 mL), and to this solution was
added 1 m NaOH (7.2 mL, 7.2 mmol). The mixture was stirred at
room temperature overnight, then carefully acidified to pH Ϸ 2–3
with 2 m HCl causing precipitation of a white fluffy solid. After
removal by filtration the solid obtained was dried in vacuo to yield
7. Yield: 0.546 g (69 %). C16H18N4O4 (330.34): calcd. C 58.17, H
5.49, N 16.96; found C 58.15, H 5.53, N 16.74. 1H NMR
ν = 3466, 1655, 1415, 1369, 1338, 1284, 1076, 926, 809, 755, 722,
˜
614, 577 cm–1.
(S)-[1-(Benzyloxycarbonylmethylcarbamoyl)-2-(phenylethyl)]ammo-
nium Trifluoroacetate (TFA·H–PheϪGlyϪOBn, 4): The dipeptide
3 (2.51 g 6.09 mmol) was dissolved in dichloromethane (10 mL) to
which was added trifluoroacetic acid (3 mL). The mixture was
stirred in an open flask for 2–3 h, and then the solvent and excess
trifluoroacetic acid were evaporated in vacuo. The resulting oil was
triturated thoroughly with diethyl ether/toluene to give 4 as a white
solid. Yield: 2.26 g (87 %). 1H NMR (400 MHz, CDCl3/[D6]
2
3
(400 MHz, [D6]DMSO): δ = 2.90 (ABX, JH,H = 13.7, JH,H = 4.3,
3
9.5 Hz, 2 H, PhCH2), 3.77 (d, JH,H = 6.1 Hz, 2 H, CH2NH), 4.57
(br. dt, JH,H Ϸ 9, 4.2 Hz, 1 H, CHCH2), 4.75 (AB, JH,H
16.0 Hz, 2 H, CH2Pz), 6.20 (br. t, JH,H Ϸ 2 Hz, 1 H, PzH), 7.16–
3
2
2
3
=
DMSO): δ = 3.18 (ABX, JH,H = 13.9, JH,H = 6.2, 6.8 Hz, 2 H,
3
3
3
PhCH2), 3.98 (d, JH,H = 5.6 Hz, 2 H, CH2NH), 4.31 (br. t, JH,H
Ϸ 6.7 Hz, 1 H, CHCH2), 5.10 (s, 2 H, CH2O), 7.21–7.47 (m, 10 H,
PhH), 8.80 (br. t, JH,H Ϸ 5.6 Hz, 1 H, NH) ppm.
3
7.27 (m, 5 H, PhH), 7.39 (d, JH,H = 1.7 Hz, 1 H, PzH), 7.55 (d,
3JH,H = 2.5 Hz, 1 H, PzH), 8.32 (br. d, JH,H Ϸ 8.5 Hz, 1 H, NH),
3
3
8.49 (br. t, 3JH,H Ϸ 5.8 Hz, 1 H, NH) ppm. 13C NMR (100.6 MHz,
[D6]DMSO): δ = 37.9, 40.9, 53.6(8), 53.7(1), 106.3, 126.3, 128.1,
129.2, 131.2, 137.5, 138.8, 166.4 (2 peaks), 171.0 ppm. MS (EI):
m/z (%) = 330 (2) [M+].
Benzyl
{(S)-3-Phenyl-2-[(2-pyrazol-1-yl)acetylamino]propionyl-
amino}acetate (PzCH2CO–PheϪGlyϪOBn, 5): A sample of dipep-
tide 3 (3.0 g, 7.3 mmol) was deprotected as above and the resulting
crude oil dissolved in dry dimethylformamide (50 mL). Next (pyr-
azol-1-yl)acetic acid (1, 1.08 g, 8.0 mmol) hydroxybenzotriazole
(1.18 g, 8.8 mmol) and diisopropylethylamine (1.52 mL, 9.6 mmol)
{(S)-3-Phenyl-2-[(2-pyridin-2-yl)acetylamino]propionylamino}acetic
Acid (H3L3, 8): The modified tripeptide ester 6 (3.0 g, 6.96 mmol)
were added to the solution followed by 1-[3-(dimethylamino)pro- was hydrolysed by the method described for 7 to give three crops
pyl]-3-ethylcarbodiimide hydrochloride (1.54 g, 8.8 mmol). After
stirring at room temperature overnight the dimethylformamide was
removed by rotary evaporation under high vacuum and the oil dis-
solved in dichloromethane/water. After separation, the dichloro-
methane layer was washed with water, dried with anhydrous so-
dium sulfate, filtered then the solvent removed in vacuo. The re-
sulting buff-coloured solid was dissolved in the minimum volume
of hot ethyl acetate, and upon cooling an amorphous white solid
precipitated. Filtration and thorough drying (to remove large
amounts of ethyl acetate associated with the solid) under vacuum
gave 5. Yield: 2.34 g (76 %). C23H24N4O4 (420.18): calcd. C 65.70,
H 5.75, N 13.33; found C 65.50, H 5.85, N 13.31. 1H NMR
of a white solid. Total yield: 1.57 g (66 %). 1H NMR (400 MHz,
2
[D6]DMSO + CDCl3): δ = 2.77 (dd, JH,H = 13.9, 3JH,H = 10.4 Hz,
2
3
1 H, PhCHHCH), 3.10 (dd, JH,H = 13.9, JH,H = 3.9 Hz, 1 H,
PhCHHCH), 3.59 (AB, 2JH,H = 14.6 Hz,, 2 H, CH2Py), 3.80 (ABX,
2JH,H = 13.9, JH,H = 14.6, 14.6 Hz, 2 H, CH2NH), 4.58 (br. dt,
3
3JH,H = 8.7, 4.2 Hz, 1 H, CHCH2), 7.07 (d, JH,H = 7.9 Hz, 1 H,
3
3
4
ArH), 7.15–7.24 (m, 6 H, ArH), 7.62 (td, JH,H = 7.6, JH,H
=
1.7 Hz, 1 H, ArH), 8.42–8.46 (m, 2 H, NH + ArH), 8.52 (br. t,
3JH,H Ϸ 5.9 Hz, 1 H, NH), 12.6 (br. s, 1 H, CO2 H) ppm. 13C NMR
(100.6 MHz, [D6]DMSO + CDCl3): δ = 37.5, 40.7, 44.6, 53.8,
121.6, 123.4, 126.1, 127.9, 129.1, 136.4, 137.8, 148.6, 156.1, 168.8,
171.0, 171.4 ppm. MS (EI): m/z (%) = 341 (1) [M+], 250 (15), 239
(20), 120 (90).
2
3
(400 MHz, CDCl3): δ = 3.02 (ABX, JH,H = 13.9, JH,H = 6.6,
876
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 872–878

Formation of ATCUN-Like Copper(ii) Binding Sites
FULL PAPER
Table 3. Crystal data and structure refinement for 1, 2 and 9.
Compound
1
2
9
Empirical formula
Mr [g mol–1]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
C5H6N2O2
126.12
orthorhombic
P212121
4.9166(4)
7.7232(9)
15.8610(15)
90
602.27(10)
4, 1.391
C10H14CuN4O6
349.79
monoclinic
C2/c
12.4705(16)
10.2272(10)
10.3266(12)
99.298(10)
1299.7(3)
C32H51CuN5O4
633.32
orthorhombic
P212121
10.2652(3)
17.4034(6)
18.8660(8)
90
β [°]
V [Å3]
3370.4(2)
4, 1.248
0.689
Z, Dcalcd. [gcm–3]
4, 1.788
1.716
µ (Mo-Kα) [mm–1]
0.110
F(ooo
)
264
716
1356
Crystal dimensions [mm3]
θmin., θmax. [°]
0.60 × 0.20 × 0.20
2.57, 34.81
–7/6, –12/10, –25/25
8870/1525
0, 87
Ϫ
0.0389
0.30×0.20×0.20
2.59, 30.00
–17/17, –12/14, –14/14
6426/1862
0, 106
0.6075, 0.7546
0.0195
0.50×0.40×0.30
2.46, 34.77
–16/14, –27/27, –30/30
50653/14515
0, 380
0.6962, 0.8495
0.0380
h, k, l ranges
Collected/unique reflections
Restraints, parameters
Tmin., Tmax.
Rint
Goodness-of-fit on F2
R indices [I Ͼ 2σ(I)]
R indices (all data)
Flack parameter
1.088
1.025
1.085
R1 = 0.0429, wR2 = 0.1082
R1 = 0.0529, wR2 = 0.1148
not determined
0.219/–0.249
R1 = 0.0212, wR2 = 0.0563
R1 = 0.0260, wR2 = 0.0571
–
R1 = 0.0392, wR2 = 0.0961
R1 = 0.0475, wR2 = 0.1029
0.001(7)
Larg. diff. peak/hole [e·Å–3]
0.427/–0.362
0.696/–0.732
Synthesis of the Copper(II) Complex [Bu4N][Cu(L2)] (9): The modi-
fied tripeptide 7 (200 mg, 0.606 mmol) was added to potassium hy-
droxide (102.1 mg, 1.808 mmol) in water (2 mL) to give a pale yel-
low solution. To this was added dropwise a solution of cupric ace-
tate monohydrate (120.7 mg, 0.606 mmol) in water (2 mL) and then
the resulting royal blue solution was filtered through a cotton wool
plug. (Tetra-n-butyl)ammonium bromide (194.3 mg, 0.606 mmol)
was added and the solution evaporated to dryness giving a dark
blue solid. Addition of chloroform (10 mL) resulted in the forma-
tion of a cherry red solution and white solid, the latter being re-
moved by filtration. The red solution was evaporated and thor-
oughly dried in vacuo before crystallization from dichloromethane/
ethyl acetate (1:6). Yield: 300 mg (78 %). Growth of crystals suit-
able for single-crystal X-ray diffraction was achieved by slow evap-
oration of dichloromethane from a dichloromethane/ethyl acetate
solution, resulting in the isolation of red needle-like crystals.
C32H51CuN5O4 (633.33): calcd. C 60.69, H 8.12, N 11.06; found C
60.48, H 7.84, N 11.05. MS (APCI): m/z (%) = 631.3 [M – H]+.
UV/Vis (CH2Cl2): λ = 19500 cm–1 (135 M–1 cm–1); (H2O): 18000
Cu65). EPR: see Figure 4 for the near identical spectra of complexes
9 and 10. This data is available as Supporting information.
Crystal Structure Determinations: Data for 1, 2 and 9 were collected
at 150(2) K with graphite-monochromated Mo-Kα radiation (wave-
length of 0.71073 Å) with a Stoe IPDS II image plate dif-
fractometer (Table 3). Space groups were determined by an exami-
nation of the systematic absences in the data, and their correct
identification was confirmed by the successful solution and refine-
ment of the structures. Solutions were provided via direct methods
using SHELXS-97 and refined by full-matrix least-squares on F2
using SHELXL-97.[21] All non-hydrogen atoms were refined aniso-
tropically. All hydrogen atoms on carbon were placed in calculated
positions with Uiso set at 1.2 times the Ueq of the parent atom, and
the positional and isotropic displacement parameters of all O–H
groups were freely refined. Numerical absorption corrections were
applied to 2 and 9 using X-RED/X-SHAPE.[22] The analysis of the
structures was carried out with PLATON[23] and WinGX.[24]
Supplementary data for 1, 2 and 9 (CCDC-239525, -239527 and
-239526, respectively) are available free of charge from the
Cambridge Crystallographic Data Centre (CCDC) via
www.ccdc.cam.ac.uk/data_request/cif.
(110). IR (KBr): ν = 3090 (w), 2964 (m), 2935 (w), 2875 (m), 1623
˜
(m), 1598 (s), 1387 (m), 1338 (m), 1081 (w), 1073 (w), 756 (w), 709
(w), 536 (w), 500 (w) cm–1. EPR: See Table 1 and Figure 4. These
spectra are available as Supporting Information (see also the foot-
note on the first page of this article).
Acknowledgments
Synthesis of the Copper(II) Complex [Bu4N][Cu(L3)] (10): The prep-
aration of the copper(ii) complex of the modified tripeptide 8
(200 mg, 0.606 mmol) was attempted using the method described
above for 9. Neither a pure sample nor X-ray quality crystals of 10
could be obtained due to the slow decomposition of the complex
in solution over several days to give a pale green solution. The
EPR and electronic absorption spectra were determined for freshly
prepared solutions and the extinction coefficients were calculated
with the assumption that all the ligand was successfully complexed
to copper(ii). UV/Vis [CH2Cl2/methanol (3:1)]: λ = 19200 cm–1
(125 m–1 cm–1); (H2O): 18000 (110). MS (negative ion electrospray)
[CH2Cl2]: m/z (%) = 401.0 (100) ([M]– Cu63), 403.0 (45) ([M]–,
We acknowledge the University of Hull for financial support of
this work and Dr E. J. L. McInnes, of the EPSRC Multi-frequency
EPR Service Centre at The University of Manchester, for his con-
tribution to the analysis of the EPR data.
[1] C. Harford, B. Sarkar, Acc. Chem. Res. 1997, 30, 123–130.
[2] E. C. Long, Acc. Chem. Res. 1999, 32, 827–836.
[3] L. W. Donaldson, N. R. Skrynnikov, W. Y. Choy, D. R. Mu-
handiram, B. Sarkar, J. D. Forman-Kay, L. E. Kay, J. Am.
Chem. Soc. 2001, 123, 9843–9847.
Eur. J. Inorg. Chem. 2005, 872–878
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
877

A. N. Boa, J. D. Crane, R. M. Kowalczyk, N. H. Sultana
FULL PAPER
[4] W. Bal, G. N. Chmurny, B. D. Hilton, P. J. Sadler, A. Tucker, J.
Am. Chem. Soc. 1996, 118, 4727.
[5] R. Nagane, T. Koshigoe, M. Chikira, J. Inorg. Biochem. 2003,
93, 204–212.
[6] R. Nagane, T. Koshigoe, M. Chikira, E. C. Long, J. Inorg. Bi-
ochem. 2001, 83, 17–23.
[7] G. L. Millhauser, Acc. Chem. Res. 2004, 37, 79–85.
[8] A. Torrado, G. K. Walkup, B. Imperiali, J. Am. Chem. Soc.
1998, 120, 609–610.
[17] The United Kingdom Chemical Database Service: D. A.
Fletcher, R. F. McMeeking, D. Parkin, J. Chem. Inf. Comput.
Sci. 1996, 36, 746–749.
[18] U. Bierbach, M. Sabat, N. Farrell, Inorg. Chem. 2000, 39, 1882–
1890.
[19] W. K. Musker, M. S. Hussain, Inorg. Nucl. Chem. Lett. 1967,
3, 271–274.
[20] M. Bodansky, A. Bodansky, The Practice of Peptide Synthesis;
Springer, New York, 1984.
[9] Y. Zheng, K. M. Gattás-Asfura, V. Konka, R. M. Leblanc, [21] G. M. Sheldrick, SHELXS-97 and SHELXL-97, Programs for
Chem. Commun.2002, 2350–2351.
[10] A. N. Boa, J. D. Crane, Acta Crystallogr. Sect. E 2004, 60,
O966–O967.
the Solution and Refinement of Crystal Structures; University
of Göttingen, 1997.
[22] Stoe & Cie, X-AREA and X-RED, Stoe & Cie GmbH, Darm-
stadt, 2001.
[23] A. L. Spek, PLATON - A Multipurpose Crystallographic Tool;
Utrecht University, 2000.
[24] WinGX - An Integrated System of Windows Programs for the
Solution, Refinement and Analysis of Single Crystal X-ray Dif-
fraction Data: L. J. Farrugia, J. Appl. Crystallogr. 1999, 32,
837–838.
[11] K. Kurdziel, T. Głowiak, S. Materazzi, J. Jezierska, Polyhedron
2003, 22, 3123–3128.
[12] P. Drozdzewski, B. Pawlak, T. Głowiak, J. Mol. Struct. 2003,
˙ ˙
654, 111–118.
[13] S. Materazzi, E. Vasca, Thermochim. Acta 2001, 373, 7–12.
[14] K. Fukui, H. Ohya-Nishiguchi, H. Kamada, Inorg. Chem.
1998, 37, 2326–2327.
[15] D. Sanna, G. Micera, P. Buglyó, T. Kiss, T. Gajda, P. Surdy, [25] ORTEP-3 for Windows: L. J. Farrugia, J. Appl. Crystallogr.
Inorg. Chim. Acta 1998, 268, 297–306.
[16] M.-A. Martinez-Lorente, J.-P. Tuchagues, V. Pétrouléas, J. M.
Savariault, R. Poinsot, M. Drillon, Inorg. Chem. 1991, 30,
3587–3589.
1997, 30, 565.
Received: May 25, 2004
878
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 872–878