Ternary copper(II) complexes with N-carboxymethyl-l-prolinato(2-) ion and imidazole or creatinine: A comparative study of the interligand interactions influencing the molecular recognition and stability

Article abstract of DOI:10.1016/j.jinorgbio.2005.03.012

The compounds {[Cu(CMP)(Him)]·H₂O}_n (I) and [Cu(CMP)(crea)H₂O]·3H₂O (II) were synthesized and characterized by X-ray diffraction, thermal, spectral and magnetic methods (CMP = N-carboxymethyl-;l-prolinato(2-) ion, Him = imidazole and crea = creatinine). Appropriate structural comparison with other compounds such as {[Cu(CMP)(H₂O)]·H₂O}_n, [Cu(crea)₂Cl₂] and [Cu(dipeptide)(crea)(H₂O)_x]·nH₂O (x = 0 or 1) have been made in order to prove that crea can act as an imidazole-like ligand (because it is able to promote the same fac- to mer-CMP tridentate conformational change in copper(II) complexes) as well as to discuss the interligand interactions which control the 'Cu(CMP) complex-crea, molecular recognition processes. In contrast to that found in related ternary complexes, we have concluded that direct CMP-crea interligand interactions are missing in the Cu-CMP-crea complex due to the inappropriate correspondence between the donor and/or acceptor H-bonding properties of these ligands. CMP can only act as H-acceptor by its two terminal carboxylate group, and crea can display H-donor and H-acceptor roles by its exocyclic -NH₂ and O moieties, respectively. That promotes the reinforcement of the Cu-N(crea) bond by a bridge -N-H(crea)...O(aqua) (2.867(3) ?, 176.4°).

Full text of DOI:10.1016/j.jinorgbio.2005.03.012

JOURNAL OF
Inorganic
Biochemistry
Journal of Inorganic Biochemistry 99 (2005) 1424–1432
www.elsevier.com/locate/jinorgbio
Ternary copper(II) complexes with N-carboxymethyl-L-prolinato(2ꢀ)
ion and imidazole or creatinine: A comparative study of the
interligand interactions influencing the molecular
recognition and stability
a
b
a
a
M. Tribet , B. Covelo , A.G. Sicilia-Zafra , R. Navarrete-Casas ,
c
a
c
a,
*
´
´
´
´
D. Choquesillo-Lazarte , J.M. Gonzalez-Perez , A. Castineiras , J. Niclos-Gutierrez
˜
a
Department of Inorganic Chemistry, Faculty of Pharmacy, University of Granada, E-18071Granada, Spain
b
Department of Inorganic Chemistry, Faculty of Sciences, University of Vigo, E-36200 Vigo, Spain
Department of Inorganic Chemistry, Faculty of Pharmacy, University of Santiago de Compostela, E-15706 Santiago de Compostela, Spain
c
Received 31 January 2005; received in revised form 14 March 2005; accepted 25 March 2005
Available online 5 May 2005
´
´
In memoriam to Prof. Francisco Gonzalez Garcıa, University of Sevilla, Spain
Abstract
The compounds {[Cu(CMP)(Him)] Æ H2O}n (I) and [Cu(CMP)(crea)H2O] Æ 3H2O (II) were synthesized and characterized by X-
ray diffraction, thermal, spectral and magnetic methods (CMP = N-carboxymethyl-;L-prolinato(2ꢀ) ion, Him = imidazole and
crea = creatinine). Appropriate structural comparison with other compounds such as {[Cu(CMP)(H2O)] Æ H2O}n, [Cu(crea)2Cl2]
and [Cu(dipeptide)(crea)(H2O)x] Æ nH2O (x = 0 or 1) have been made in order to prove that crea can act as an imidazole-like ligand
(because it is able to promote the same fac- to mer-CMP tridentate conformational change in copper(II) complexes) as well as to
discuss the interligand interactions which control the ꢀCu(CMP) complex-creaꢁ molecular recognition processes. In contrast to that
found in related ternary complexes, we have concluded that direct CMP–crea interligand interactions are missing in the Cu–CMP–
crea complex due to the inappropriate correspondence between the donor and/or acceptor H-bonding properties of these ligands.
CMP can only act as H-acceptor by its two terminal carboxylate group, and crea can display H-donor and H-acceptor roles by its
exocyclic –NH2 and O moieties, respectively. That promotes the reinforcement of the Cu–N(crea) bond by a bridge –N–
˚
H(crea)ꢁ ꢁ ꢁO(aqua) (2.867(3) A, 176.4ꢀ).
ꢁ 2005 Elsevier Inc. All rights reserved.
Keywords: Copper(II); N-carboxymethyl-L-proline; Creatinine; Imidazole; Interligand interactions; Molecular recognition; Crystal structure
1. Introduction
organic chemists had been applying inorganic concepts
and models to rationalize the metal–ligand binding affin-
ities in biological systems. As far as concerns to the low
molecular weigh fraction, amino acids, peptides, nucleo-
bases, nucleosides, nucleotides and other biomolecules
(and also many drugs) are able to give diverse ternary
metal complexes, a kind of species which formation in
solution is favoured by a statistical factor [2]. However,
there are abundant evidences that the dominant affinity
In plasma, a large variety of biomolecules, from low
to high molecular weigh, are potential ligands for metal
ions [1]. In this connection, since several decades bioin-
*
Correspondence author. Tel.: +34 95 8243855; fax: +34 95
8246219.
´
´
E-mail address: jniclos@ugr.es (J. Niclos-Gutierrez).
0162-0134/$ - see front matter ꢁ 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2005.03.012

M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
1425
between metal ions and ligands is strongly influenced by
the ꢀhard and soft acid and baseꢁ criterion [3,4], mainly
developed by Pearson. Moreover, in recent years an
additional effort is carried out in order to deepen on
the influence of intra-molecular interligand interactions
in the stability of the ternary and related complex spe-
cies. Among all these bioligands, a limited attention
has been paid to creatinine (2-amino-1,5-dihydro-1-
methyl-4H-imidazol-4-one, hereafter crea) which coor-
dination properties were broadly reviewed by Mitewa
[5] in a rather comprehensive report, some years ago.
Since then, some new contributions have been made
on the metal binding properties of crea because its rele-
vant versatility as ligand, which can act in various tauto-
meric forms (Scheme 1), and due to its important roles
in the kidney clearance as a relevant final product for
the catabolism of the nitrogen biomolecules.
The crystal structure determinations of creatinine [6–
8], four organic derivatives (salts [9,10] or clathrates [8])
as well as metal complexes of Ni(II) [11], Cu(II) [12–15],
Zn(II) [16,17], Cd(II) [18,19], Ph–Hg(I) [20,21], Ag(I)
[22], Pd(II) [23,24], Pt(II) [25–30] and Ir [31] have been
reported. A survey of this literature reveals the following
coordination roles for crea:
Scheme 2.
(1) In most of these cases neutral crea binds the metal
through the N heterocyclic donor (Scheme 2(a))
[11,13–20,22–30].
(5) The anion crea-H also acts as bridging ligand for
two phenyl-mercury(I) by both N-heterocyclic
and N-exocyclic imino groups (Scheme 2(e)) [20].
(2) Other unidentate mode for neutral crea is known
via the O-exocyclic coordination as for the
2þ
NiðenÞ2 complex unit (Scheme 2(b)) [11].
Crea complexes of known structure include a series of
compounds with general formula [MII(crea)2L2] where
M = Cu [13], Zn [16], Cd [18], Pd [23,24] or Pt
[23,25,26,28] and Lꢀ = Clꢀ [13,15,18,23–26], NOꢀ2 [28]
or Iꢀ [23]. All the [MII(crea)2Cl2] compounds are trans
isomers. Four crystallographic polymorphs have been
studied for trans-[PtII(crea)2Cl2], whereas the analogue
complexes with Lꢀ ¼ NOꢀ2 [28] or Iꢀ [23] are the cis-
isomers. It is interesting to note that most metal ions
for which crea complexes have been structurally charac-
terized fall in the borderline or the soft groups of Pear-
son acids, with a clear preference for N-heterocyclic donor
atoms (borderline base of Pearson) as those of pyridine
or imidazole and related ligands. Indeed, this structural
chemistry reveals a trend of crea to exhibits ligand
behaviour close to that of imidazole-like ligands. As a
part of our program on mixed-ligand complexes with
amino acid derivatives and N-heterocyclic donors, we
have carried out a series of studies involving the nucleo-
base adenine [32–35]. These and many other studies re-
veal that structure determinations are not only the
most rigorous method to investigate mixed-ligand com-
plexes [2] but also constitute a suitable way to elucidate
interligand interactions involved in the molecular recog-
nition processes which enable a binary complex to react
(3) It is also known the chelating role involving the N-
heterocyclic and the O-exocyclic donors of crea to
give a four-membered ring (Scheme 2(c)) [31].
(4) In one organometallic derivative of iridium, crea
acts as unidentate by the N-exocyclic imino group
(Scheme 2(d)) [31].
Scheme 1.

1426
M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
was added. The reaction mixture was heated at 80 ꢀC
and this temperature was maintained for 4 h. A solution
of KOH (0.05 mol, 2.8 g) in 50 ml of water was dropwise
added to the reaction mixture during this time. The
mother liquor was cooled at room temperature and neu-
tralized with HCl 6 N. The solution was concentrated
under reduced pressure at 40 ꢀC until a white precipitate
of KCl appeared. The mixture was then cooled in an
ice–salt bath and the KCl (by-product) removed by fil-
tration. This operation was repeated several times until
ꢂ25 g of KCl were removed (approximately a half of
the estimated amount of this by-product). The pH of
the partially de-salted solution was adjusted to 2–2.5
with HCl 6 N, to produce the precipitation of the de-
sired product. The crude acid was dissolved in water
by addition of a less amount of NaHCO3 and re-precip-
itated by addition of the equivalent amount of HCl 6 N.
The desired product was collected by filtration, washed
with cool water and air-dried. Small samples of the
product were re-crystallized from a water:ethanol mix-
ture (70:30) or from a water solution and subsequent
addition of isopropanol, which produced needless
shaped crystals, but non-suitable for X-ray crystallogra-
phy. Yield: 49% (37% in [39]). Anal. Calc. for
C7H11NO4: C 48.55, H 6.40, N 8.09. Found: C 48.34,
H 6.77, N 8.16. Melting point (dec.) = 217.5–219.5 ꢀC
in [39]. Relevant FT-IR spectrum bands (cmꢀ1) and ten-
tative assignments [41]: m(OH) 3427 mb, m(C@O) 1703 s,
m(N–H) 1529 m, mas(COO) 1630 mb, combination mode
of –COOH group m(C–O) + d(O–H) 1414 m, ms(COO)
1360 m, out-of-plane deformation mode of –COOH
group p(O–H) 845 m. H2CMP pyrolyzes in three steps
(220–300, 300–430 and 430–500 ꢀC) giving a residue be-
low 1.5%. The first step is a decarboxylation and pyro-
lysis (with production of CO2, H2O and CO), whereas
the other two steps produce N-containing gases (NH3
alone or with N2O and NO). 1H NMR spectrum
(300 MHz, d6-DMSO; d, ppm; J, Hz; see Scheme 4):
1.67–1.98 (m, 2H on C7, 1H on C6), 1.99–2.12 (m, 1H
on C6), 2.70 (c, 1H on C5, J 8.2), 3.07–3.13 (m, 1H on
C5), 3.39 (d, 1H, J 17 on C4), 3.50 (d, 1H on C3, J
3.7 and 5.1), 3.57 (d, 1H on C4, J 17). 13C NMR spec-
trum (75 MHz, d6-DMSO; d, ppm): 23.7 (C7), 29.1
(C6), 56.3 (C5), 56,9 (C4), 67.0 (C3), 168.8 (C1), 171.5
(C2). This assignations [42] were also based on a DEPT
analysis (results available from authors upon request).
Scheme 3.
with an auxiliary ligand to give a ternary complex.
Indeed these general ideas apply to most biological sys-
tems where ternary species should be formed (metallo-
enzyme–substrate complexes, protein–metal–nucleic
acid interactions, and diverse low-weight ternary metal
complexes in plasma). In this broad context we found
instructive to carry out comparative studies on ternary
metal complexes with an amino acidato ligand and imid-
azole (Him) or creatinine as auxiliary ligands. This work
deals with (N-carboxymethyl-L-prolinato)-copper(II)
complexes with Him or crea, which are synthesized
and studied by X-ray diffraction, thermal, spectral and
magnetic methods (see Scheme 3). In this sense, it is rel-
evant to known the structural reports of the closely re-
lated polymeric compounds {[Cu(CMP)(H2O)] Æ H2O}n
[36] and {[Cu(CEP)(Him)]}n [37] (CEP = N-(2-carboxy-
ethyl)-L-prolinato) (see Scheme 3).
2. Experimental
All chemicals used in this study were purchased from
Aldrich, Acros or other commercial sources with the
best analytical grade, and were used as received.
2.1. Synthesis of N-carboxymethyl-L-proline (H2CMP)
Korman and Clarke [38] described the synthesis of
this amino diacid long time ago by reaction of bromo-
acetic acid and L-Pro in alkaline medium and Miyazawa
[39] has reported similar results on a broad work to ob-
tain a series of N-carboxymethyl amino acids. The acid-
ity constants at 25 ꢀC and I = 0.1 (KNO3) are
pK0 = 1.47, pK1 = 2.32 and pK2 = 10.14 [40]. Stability
constants for M(CMP) and M(CMP)2 chelates with
MII = Co, Ni, Cu, and Cd were also reported [40]. We
have prepared this ligand by reaction of potassium salts
of L-proline and chloroacetic acid in water. To a cooled
solution of chloroacetic acid (0.22 mol, 20.8 g) in 100 ml
of water, kept in a three neck reactor on an ice–salt bath
(ꢀ18 ꢀC), a solution of KOH (0.44 mol, 24.7 g) in
100 ml was added under a flow of N2. To this alkaline
solution, an aqueous solution of potassium L-prolinate
(previously prepared with 0.2 mol of amino acid and a
cooled solution of 0.2 mol of KOH in 50 ml of water)
Scheme 4.

M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
1427
2.2. Synthesis of the copper(II) complexes
155 ꢀC) prior to the pyrolysis of the organic ligands in
three steps (155–280, 280–415 and 415–480 ꢀC) to give
CuO as final residue. Electronic spectrum: mmax
14185 cmꢀ1. ESR spectrum: axial, with gi 2.26 and g^
2.05. lCu = 1.86 B.M. at r.t.
2.2.1. Synthesis and physical properties of poly-imidazole
(l-N-carboxymethyl-L-prolinato)copper(II)
monohydrate, {[Cu(CMP)(Him)] Æ H2O}n
(compound I)
Cu2CO3(OH)2 (221 mg, 1 mmol), H2CMP (346 mg,
2 mmol) and imidazole Him (136 mg, 2 mmol) were re-
acted in water (125 ml) in a Kitasato device with heating
(50–60 ꢀC), stirring and vacuum until a dark blue solu-
tion was obtained. Heating and stirring was continued
for half an hour and the solution was left to cool at
room temperature and slowly filtered with vacuum on
a crystallization device. After a week of evaporation,
dark blue parallelepiped crystals appeared, many of
them suitable for X-ray diffraction studies. The product
was collected in successive fractions until a good yield
(85%). Anal. Cald. for C10H15CuN3O5: C 37.44, H
4.71, N 13.10. Found: C 37.62, H, 4.80, N 13.24. Rele-
vant FT-IR spectrum bands (cmꢀ1) and tentative assign-
ments [41]: mas(H2O) ꢂ 3400 sb, ms(H2O) and m(N–H,
Him) 3255 mb, mas(COO, CMP) and d(H2O) 1617 s,
d(N–H, Him) 1546 w, ms(COO, CMP) 1389 m, p(C–H,
Him) 658 m. The compound slowly loss the non-coordi-
nated water (50–210 ꢀC) and then pyrolyses the organic
ligands in two steps (210–320 and 320–450 ꢀC) to give
2.3. Crystal structure determinations
Suitable crystals of I and II were selected for X-ray
single crystal diffraction experiments and mounted on a
glass fiber and used for data collection on a Bruker
SMART CCD 1000. The data were processed with
SAINT [43] and corrected for absorption using SAD-
ABS [44]. The structures were solved by direct methods
using the program SHELXS-97 [45] and refined by full-
matrix least-squares techniques against F2 using
SHELXL-97 [46]. Positional and anisotropic atomic dis-
placement parameters were refined for all non-hydrogen
atoms. The hydrogen atoms were located unambigu-
ously from difference Fourier maps (N–H and O–H
hydrogen atoms) and included as fixed contributions rid-
ing on attached atoms or included in geometrically ideal-
ized positions employing appropriate riding models, with
isotropic displacement parameters constrained to 1.2
times those of their carrier atoms (1.5 for methyl groups).
Molecular graphics and geometric calculations were ob-
tained from SCHAKAL [47] and SHELXTL [48].
CuO as residue. Electronic spectrum: mmax 15150 cmꢀ1
.
ESR spectrum: Isotropic, with gav ꢂ 2.10. lCu = 1.79
Crystal data for I: C10H15CuN3O5, Mw = 320.79,
orthorrombic, a = 8.3030(6), b = 8.5660(6), c =
B.M. at r.t.
˚
17.3550(13) A,
a = 90ꢀ,
b = 90ꢀ,
c = 90ꢀ,
3
˚
2.2.2. Synthesis and physical properties of aqua-
(creatinine)(N-carboxymethyl-L-prolinato)copper(II)
V = 1234.35(15) A , T = 100(2) K, space group P212121,
Z = 4, l(Mo Ka) = 1.791 mmꢀ1, 7970 reflections mea-
three-hydrate, [Cu(CMP)(crea)(H2O)] Æ 3H2O
(compound II)
sured,
0.0273.
2918
unique.
R1 = 0.0245,
wR2 =
Cu2CO3(OH)2 (221 mg, 1 mmol) and H2CMP
(346 mg, 2 mmol) were reacted in water (100 ml) in a
Kitasato device with heating (50–60 ꢀC), stirring and
vacuum until a clear blue clear solution was obtained.
Heating and stirring was continued for half an hour
and the solution was left to cool at room temperature
and slowly filtered on an Erlenmeyer flask. Crea
(226 mg, 2 mmol) were added and the mixture stirred
until a solution was obtained. This solution was filtered
on a crystallization device and EtOH was added. The
crystallization was slowly produced keeping the device
covered by a plastic film at ꢀ18 ꢀC. This procedure pro-
duced a little amount of crystals suitable for X-ray dif-
fraction purposes. Yield: 55%. Anal. Cald. for
C11H24CuN4O9: C 31.47, H 5.76, N 13.40. Found: C
31.87, H, 6.08, N 13.35. Relevant FT-IR spectrum
bands (cmꢀ1) and tentative assignments [41]: mas(H2O)
3446 sb, ms(H2O) and m(N–H, Him) 3291 mb, m(C@O,
crea) 1669, mas(COO, CMP), d(H2O) and d(NH2, crea)
1617 s, ms(COO, CMP) 1389 m. A part of the water
content is lost in air-dry flow. The remaining non-coor-
dinated water and the aqua ligand were easily lost (40–
Crystal data for II: C11H24CuN4O9, Mw = 419.88,
orthorrombic,
c = 19.6073(9) A,
V = 1840.59(16) A ,
a = 9.0084(5),
a = 90ꢀ,
b = 10.4206(5),
˚
b = 90ꢀ,
c = 90ꢀ,
group
3
˚
T = 293(2) K,
space
P212121, Z = 4, l(Mo Ka) = 1.237 mmꢀ1, 3627 reflec-
tions measured, 3627 unique. R1 = 0.0249, wR2 =
0.0268.
2.4. Other physico-chemical studies
Elemental analysis, thermal, spectral and magnetic
studies were carried out as described elsewhere [32].
3. Results and discussion
3.1. Molecular and crystal structures
The crystal of compound I consists in a polymer with
chains extending along the b-axis. The Cu(II) atom
exhibits a square base pyramidal coordination, type
4 + 1 (Fig. 1 and Table 1). The N amino and two

1428
M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
Fig. 1. Asymmetric unit in the crystal of {[Cu(CMP)(Him)] Æ H2O}n (I).
Table 1
Selected structural data for {[Cu(CMP)(Him)] Æ H2O}n (I) and [Cu(CMP)(crea)(H2O)] Æ 3H2O (II)
Compound I
Compound II
Coordination polyhedron
Cu–O(1)
Cu–O(3)
Cu–N(2)
Cu–N(1)
Cu–O(2)#1
Coordination polyhedron
Cu(1)–N(2)
Cu(1)–O(21)
Cu(1)–O(11)
Cu(1)–N(1)
˚
˚
1.954(2) A
1.955(2) A
˚
˚
1.959(2) A
1.965(2) A
˚
˚
1.962(2) A
1.968(2) A
˚
˚
1.987(2) A
1.976(2) A
˚
˚
2.586(2) A
Cu(1)–O(1)
2.342(2) A
O(1)–Cu–O(3)
N(2)–Cu–N(1)
168.2(1)ꢀ
176.8(1)ꢀ
O(21)–Cu(1)–O(11)
N(2)–Cu(1)–N(1)
167.4(1)ꢀ
161.4(1)ꢀ
˚
˚
H-bonding interactions
N(3)–H(3)ꢁ ꢁ ꢁO(5)#2
O(5)–H(5A)ꢁ ꢁ ꢁO(2)#3
O(5)–H(5B)ꢁ ꢁ ꢁO(3)#4
d(Dꢁ ꢁ ꢁA) (A)
\DHA (ꢀ)
174.4
145.8
H-bonding interactions
O(1)–H(1A)ꢁ ꢁ ꢁO(12)#1
O(1)–H(1B)ꢁ ꢁ ꢁO(22)#2
N(3)–H(3A)ꢁ ꢁ ꢁO(2)#3
N(3)–H(3B)ꢁ ꢁ ꢁO(1)
d(Dꢁ ꢁ ꢁA) (A)
\DHA (ꢀ)
167.7
151.4
176.4
168.9
164.1
156.4
151.9
160.2
157.3
2.745(3)
2.776(3)
2.794(2)
2.765(2)
2.649(2)
2.867(3)
2.861(3)
2.777(3)
2.876(2)
2.894(3)
2.921(5)
2.883(6)
153.7
O(2)–H(20A)ꢁ ꢁ ꢁO(10)#3
O(2)–H(20B)ꢁ ꢁ ꢁO(11)
O(3)–H(30A)ꢁ ꢁ ꢁO(12)#4
O(3)–H(30B)ꢁ ꢁ ꢁO(4)
O(4)–H(40B)ꢁ ꢁ ꢁO(2)#4
Symmetry transformations
Symmetry transformations
#1 ꢀx, y ꢀ 1/2, ꢀz + 1/2
#2 ꢀx, y + 1/2, ꢀz + 1/2
#3 x ꢀ 1/2, ꢀy + 3/2, ꢀz
#4 x, y ꢀ 1, z
#1 = A = ꢀx + 1, y + 1/2, ꢀz + 1/2
#2 x ꢀ 1, y + 1, z
#3 ꢀx + 1, y ꢀ 1/2, ꢀz + 1/2
#4 x, y ꢀ 1, z
˚
carboxyl O donors from CMP and the N-pyridine-like
plane P(1) of 0.01 A). The Cu(II) atom is displaced
˚
donor of the Him ligand define the square basal set (with
0.065 A towards the fifth donor atom, that is the O(2A)
carboxyl from the adjacent CMP ligand (symmetry
˚
bond length close to 2 A and deviations from the mean

M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
1429
transformation A = ꢀx + 2, y ꢀ 1/2, ꢀz + 3/2).
Thus, the amino diacidato ligand displays tridentate
and bridging roles. It should be noted that CMP adopts
a mer-conformation, with two nearly coplanar chelate
rings (dihedral angle between mean ring planes
u = 15.3ꢀ). That is in contrast to the fac-conformation
of CMP in the parent compound {[Cu(CMP)(H2O)] Æ
H2O}n (u = 84.7ꢀ). A similar conformational change
have been previously observed for the polymer {[Cu(I-
DA)(H2O)2]}n (IDA = iminodiacetato(2ꢀ) or N-carb-
oxymethyl-glycinato (2ꢀ) ligand) and a large series of
mixed-ligand complexes with Cu(II)/(IDA or IDA-
like)/(N-heterocyclic donor atom) = 1/1/1 [49]. In addi-
tion, it could be appreciate that the bridging carboxylate
group (with a roughly anti–syn configuration) belongs to
the N-carboxymethyl arm, which will be less steric hin-
dered than that of the L-proline moiety. Indeed, the
packing. In this connection, it should be moted that
the internal cohesion of the crystal is also reinforced
by a very weak interaction between each copper(II)
atom and a sixth potential O-donor (O4B, B = x ꢀ 1/
2, ꢀy + 3/2, ꢀz + 1) from an adjacent CMP ligand.
˚
The Cu–O4B distance (2.799 (2) A) is large enough
to be considered as an ꢀeffectiveꢁ coordination bond,
because it is very close to the sum of the Van der
˚
˚
˚
Waals radii (2.90 A) of the O(1.40 A) and Cu(1.50 A)
atoms.
Compound II has a molecular structure (Fig. 2 and
Table 1) and the copper(II) atom also exhibits a square
base pyramidal coordination, type 4 + 1. In this mole-
cule the N-heterocyclic donor of crea and the three
donor atoms of the mer-tridentate CMP define a
square base. The coordination polyhedron is accom-
plished by the apical aqua ligand. It is interesting to
note that Cu–N(crea) bond is the shortest coordination
˚
bond distances Cu–N(Him) (1.962(2) A and Cu–
˚
˚
N(CMP) 1.987(2) A) fall in the normal ranges previ-
bond in this molecule (1.955(2) A). This value falls
ously reported for closely Cu–IDA derivatives. Other
interesting structural features of this compound are the
distortion of the 4 + 1 polyhedron towards a bipyrami-
dal coordination (as measured by the parameter of
Addison [50], s = [(trans-basal angles)/60) · 100] =
14.28% and the dihedral angle between the basal coordi-
nation plane and the Him plane (27.7ꢀ). This last value is
consistent with the absence of direct intra-molecular int-
erligand Him-CMP interaction (as for example, a non-
conventional C–H(Him)ꢁ ꢁ ꢁ O(carboxyl) interaction). In
below the Cu–N(crea) distances reported for the
˚
related compounds [Cu(crea)2Cl2] (1.982(2) A) [13],
˚
[Cu(GlyGly)(crea)(H2O)] Æ 3/2H2O (2.014(2) A) [14],
˚
[Cu(Gly-L-Trp)(crea)] Æ 1.25 H2O (1.996(9) A) [15] and
˚
[Cu(L-Ala-Gly)(crea)] Æ 2H2O (average 2.008(9) A) [15].
The most striking structural feature is the perpendicular
disposition of the mean crea and the basal coordination
(P(1)) planes (89.9ꢀ) which is explained by the formation
of a rather strong intra-molecular interaction N–
˚
H(crea)ꢁ ꢁ ꢁO(aqua apical) (2.86 A, 169ꢀ), which to the
˚
fact, only inter-molecular N–Hꢁ ꢁ ꢁO(water) (2.75 A,
best of our knowledge represents a novel insight in
molecular recognition between crea and a binary com-
plex to form a ternary specie.
˚
174ꢀ) and O(water)–Hꢁ ꢁ ꢁ O!(carboxyl) (2.78 0.01 A,
149 4ꢀ) interactions contribute actively to the crystal
Fig. 2. Asymmetric unit in the crystal of [Cu(CMP)(crea)(H2O)] Æ 3H2O (II).

1430
M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
3.2. Structural comparisons
angles between the mean heterocycle and basal
coordination planes, a = Him/P(1) or crea/P(1).
Both Him complexes have a values below 28ꢀ,
whereas this angle represents a perpendicular dis-
position for the crea derivative II (a = 89.9ꢀ). This
finding could be responsible of the different molec-
ular recognition process between the Cu–CMP
and Cu–CEP chelates with Him or between the
Cu–CMP chelate and crea.
It is instructive to carry out structural comparisons to
evaluate the behaviour of Him and crea as auxiliary li-
gands in ternary copper(II) complexes. To this purpose,
the Table 2 summarizes a series of structural data for
Cu(II)–CMP complexes and a related ternary complex
derived of CEP. From N-carboxyalyl-L-prolinato(2-)-
copper(II) complexes we can deduce the following
features:
(1) In these complexes, the Cu-L-Pro chelate ring is the
most strained. Thus, in absence of Him or crea the
Cu-L-Pro ring is in the basal plane of the distorted
Cu(II) coordination polyhedron. That should be
attributed to the steric factor of the N,C-n-propyl
side chain in the skeleton of this amino acid.
(2) The conformation of CMP is fac-tridentate in
absence of an auxiliary N-heterocyclic donor
ligand (u = 84.7ꢀ), whereas CMP and CEP ligands
exhibits a mer-tridentate conformation in the com-
plexes having Him or crea as secondary ligand. To
this respect, there is a similar role between Him
and crea ligands.
(3) The change from fac-CMP to mer-CMP represents
a shortening of the Cu–N(CMP) bond and an
increase of the distortion from the square base
pyramidal coordination towards a trigonal bipyra-
midal geometry, as reflect the increase in the s
value.
3.3. Molecular recognition and interligand interactions
The structures of only five crea-copper(II) derivatives
have been previously reported: the salt (Hcrea)[CuCl4]
[12], [Cu(crea)2Cl2] [13] and three mixed-ligand com-
plexes with the dipeptides glycyl-glycine (Gly-Gly) [14],
glycyl-L-tryptophane (Gly-Trp) [15] or L-alanyl-glycine
(Ala-Gly) [15]. The molecule [Cu(crea)2Cl2] has a
roughly tetrahedral coordination and two N–H(exocy-
clic)ꢁ ꢁ ꢁCl intra-molecular interligand interactions con-
tributing to this molecular geometry, which is typical
of all known [MII(crea)2Cl2] complexes. Compounds of
general formula [Cu(dipeptide)(crea)(H2O)x] Æ nH2O
(x = 0 or 1) have a distorted square planar (type
4 + 0) or square pyramidal (type 4 + 1) topology and
crea is nearly coplanar to the square or basal coordina-
tion plane. In these complexes there are two intra-molec-
ular interligand H-bonding interactions, between the
amino-terminal of the dipeptide and the O-exocyclic
acceptor of crea and the exocyclic amino group of crea
and the coordinated O-carboxyl donor of the dipeptide.
That means an optimized Cu(dipeptide)-crea molecular
(4) The main difference between the crea complex II
and the Him analogues, I and {[Cu(CEP)(Him)]}n
in Table 2, concerns to the values of the dihedral
Table 2
A comparison of structural parameters of the studied Cu–CMP compounds with those of related CMP [24] and CEP [25] derivatives reported in the
literature
Data
Compound
{[Cu(CMP)(H2O)] Æ H2O}n {[Cu(CMP)Him] Æ H2O}n (I) [Cu(CMP)(crea)(H2O)] Æ {[Cu(CEP)(Him)]}n
3H2O (II)
Cu(II) coordination type
CMP or CEP ligand roles
Cu–N(CMP or CEP) (A)
4 + 1
4 + 1
4 + 1
4 + 1
fac-tridentate + bridge
2.037(2)
2.312(2)
1.929(2)
1.954(2)
1.935(2)
None
173.28
172.15
1.88
0.083(1)
0.03(1)
mer-tridentate + bridge
1.986(2)
1.955(2)
1.958(2)
2.586(2)
None
1.962(2)
168.4
176.86
mer-tridentate
1.955(2)
1.965(2)
1.968(2)
None
2.342(2)
1.976(2)
167.4
161.4
9.93
0.156(1)
0.15(1)
15.0(1)
mer-tridentate + bridge
2.002(4)
1.943(3)
1.921(3)
2.509(2)
None
1.975(4)
176.9
165.4
˚
˚
Cu–O(Gly or b-Ala arm) (A)
Cu–O(Pro arm)
0
˚
Cu–O (Gly/b-Ala-bridging) (A)
˚
Cu–O(aqua) (A)
˚
Cu–N(Him) (A)
Trans-angle O–Cu–O (ꢀ)
Trans-angle (O,N)–Cu–N (ꢀ)
Distortion parameter s (%)
14.28
0.065(1)
0.01(1)
19.16
0.095(1)
0.14(1)
˚
d(Cu)–P(1)* (A)
˚
Mean deviation from P(1)* (A)
Dihedral angle / formed by
mean planes of chelate rings (ꢀ)
Dihedral angle a between the
mean heterocycle and P(1)
planes (ꢀ)
84.7(1)
15.3(1)
17.3(1)
None
27.7
89.9
14.8

M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
1431
recognition process in which both moieties display do-
nor and acceptor roles to feature two intra-molecular
H-bonding interligand interactions. In compound II,
the binding of crea to the Cu–CMP chelate could pro-
motes a conformational change in CMP (IDA-like)
from fac- to mer-tridentate. In this case, the CMP ligand
has two terminal carboxylate groups. We can assume
that by one side the binding of crea to the copper(II)
atom by its N-heterocyclic donor enables an intra-
molecular N–H(crea)ꢁ ꢁ ꢁO(CMP) interaction. However,
on the other side, the carboxyl group of CMP (H-accep-
tor) results unappropriate to build an H-bonding
interaction with the O-exocyclic atom of crea (also
H-acceptor). As a result, a rotation around the single
Cu–N(crea) to enable an intra-molecular interligand
interaction with the apical aqua ligand can be expected.
Because this aqua ligand can play both donor and
acceptor H-bonding roles, we can assume that two pos-
sibilities of aqua-crea interligand interactions could be
expected, as depicted in the Scheme 5(a) and (b). In this
connection, it is interesting to note that both possibilities
consist in the formation of a six-membered intra-molec-
ular ring. The crystal structure determination of com-
pound II reveals that the option 5-a is preferred. In
order to explain this feature, we could keep in mind that
the ꢀapicalꢁ coordination of the aqua ligand (Cu–O
metal ions as copper(II) by the N-heterocyclic donor. In
mixed-ligand copper(II) complexes as those of [Cu-
(crea)2Cl2] [13], [Cu(dipeptide)(crea) (H2O)x] Æ nH2O
[14,15] or [Cu(CMP)(crea)(H2O)] Æ 3H2O (compound
II), the coordination number and polyhedron as well
as the orientation of the crea mean plane seems to be
dominated by the best option in order to favour the
strongest intra-molecular H-bonding interligand inter-
action. In the referred Cu(II) derivatives, different
modes for such H-bonding interactions have been found
and all of them comprise a role of H-donor for the exo-
cyclic amino group of crea. Until now all known struc-
tures of metal–crea complexes correspond to the
tautomeric forms having the O-keto exocyclic atom
((a) and (b) in Scheme 1), with a clear preponderance
of those corresponding to the A tautomer.
An additional support to this arguments can be ob-
tained considering the molecular recognition pathway
recently
reported
for
[Zn(IDA)(AdeH)(H2O)2]
(AdeH = adenine) [35]. In this case it was argued that
the polarizing effect of the Zn(II) atom on the aqua li-
gands enhances its H-donor properties and promotes
the formation of an O(aqua)-Hꢁ ꢁ ꢁN3(AdeH) bond to
reinforce the Zn–N9(AdeH) metal–nucleobase coordina-
tion bond. This compound approaches to the option 5-b,
with the aqua ligand as donor and the nucleobase as
acceptor of the intra-molecular H-bonding interaction.
In this sense, it should be noted that the formation of a
metal(II)–N9(AdeH) bond can promote a tautomerism
of the H atom usually in N9(AdeH) to N3 (the less basic
N-heterocyclic atom of adenine). Indeed, that occurs in a
dinuclear Cu(II)-derivative with bridging l-N7,N9–
AdeH and N-benzyl–IDA (NBzIDA) as metal-chelator
[34]. In this interesting compound, the Cu–N9(AdeH)
bond is effectively reinforced by an O(NBzIDA)ꢁ ꢁ ꢁH–
N3(AdeH) intra-molecular interaction, in a rather fash-
ion to the 5-a mode, here reported for compound II.
˚
2.586(2) A) do not enhance significantly the polarity of
the water molecule (and hence its H-donor properties,
as correspond to the option 5-b). Consequently, it seems
most reasonably that compound II prefers the option
5-a in which crea acts as H-donor and the apical aqua
ligand assume the H-acceptor role.
3.4. Concluding remarks
Our results indicate that creatinine is a borderline
base of Pearson and consequently, binds borderline acid
Acknowledgements
Financial support from ERDF-EC, MEC-Spain
(research project BQU2002-04523-C02) and FIS-MSC-
Spain (research project PI021029) are gratefully acknowl-
edged. DChL thanks to Prof. Castineiras for a research
˜
stayinhisGroupatUniversityofSantiagodeCompostela.
Appendix A. Supplementary information
Further details of crystal structure determination has
been deposited at the Cambridge Crystallographic Data
Centre (12 Union Road, Cambridge, CB2 1EZ, UK)
and can be obtained free of charge by quoting the depo-
sition numbers CCDC-260699 (compound I), CCDC-
260700 (compound II).
Scheme 5.

1432
M. Tribet et al. / Journal of Inorganic Biochemistry 99 (2005) 1424–1432
[28] J. Macicek, O. Angelova, G. Gencheva, M. Mitewa, P.R.
References
Bontchev, J. Crystallogr. Spectrosc. Res. 18 (1988) 651–658.
[29] M. Mitewa, G. Gencheva, P.R. Bontchev, O. Angelova, J.
Macicek, Polyhedron 7 (1988) 1273–1278.
[30] J. Macicek, G. Gencheva, M. Mitewa, Acta Crystallogr. C 52
(1996) 2429–2432.
[1] R.M. Roat-Malone, Bioinorganic Chemistry, John Wiley & Sons
Inc., Hoboken, New Jersey, 2002.
[2] H. Sigel (Ed.), Metal Ions in Biological Systems: Mixed Ligand
Complexes, vol. 2, Marcel Dekker, New York, 1973.
[3] J.E. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry:
Principles of Structure and Reactivity, fourth ed., Harper Collins,
New York, 1993 (Chapter 9).
[4] R.B. Martin, Inorg. Chim. Acta 339 (2002) 27–33.
[5] M. Mitewa, Coord. Chem. Rev. 140 (1995) 1–25.
[6] H. Banerjee, Indian J. Phys. 12 (1938) 183.
[7] S. du Pre, H. Mendel, Acta Crystallogr. 8 (1955) 311–313.
[8] T.W. Bell, Z. Hou, Y. Luo, M.G.B. Drew, E. Chapoteau, B.P.
Czech, A. Kumar, Science 269 (1995) 671–674.
[9] I.L. Karle, K.S. Dragonette, S.A. Brenner, Acta Crystallogr. 19
(1965) 713.
[31] J.M. OꢁConnor, K. Hiibner, A.L. Rheingold, L.M. Liable-Sands,
Polyhedron 16 (1997) 2029–2035.
´
[32] E. Bugella-Altamirano, D. Choquesillo-Lazarte, J.M. Gonzalez-
´
´
´
´
´
Perez, M.J. Sanchez-Moreno, R. Marın-Sanchez, J.D. Martın-
´
Ramos, B. Covelo, R. Carballo, A. Castineiras, J. Niclos-
˜
´
Gutierrez, Inorg. Chim. Acta 339 (2002) 160–170.
´
´
[33] M.J. Sanchez-Moreno, D. Choquesillo-Lazarte, J.M. Gonzalez-
´
´
´
Perez, R. Carballo, A. Castineiras, J. Niclos-Gutierrez, Inorg.
˜
Chem. Commun. 5 (2002) 800–802.
´
´
´
[34] P.X. Rojas-Gonzalez, A. Castineiras, J.M. Gonzalez-Perez, D.
˜
´
´
Choquesillo-Lazarte, J. Niclos-Gutierrez, Inorg. Chem. 41 (2002)
6190–6192.
[10] G. Smith, J.M. White, Aust. J. Chem. 54 (2001) 97–100.
[11] A. Panfil, J.J. Fiol, M. Sabat, J. Inorg. Biochem. 60 (1995) 109–
122.
[35] A.C. Morel, D. Choquesillo-Lazarte, C. Alarco´n-Payer, J.M.
´
´
´
´
Gonzalez-Perez, A. Castineiras, J. Niclos-Gutierrez, Inorg. Chem.
˜
Commun. 6 (2003) 1354–1357.
[12] M.R. Udupa, B. Krebs, Inorg. Chim. Acta 33 (1979) 241–244.
[36] I.F. Burshtein, A.L. Poznyak, L.V. Stopolyanskaya, Kris-
tallografiya 45 (2000) 259 (in Russian).
[37] I.F. Burshtein, A.L. Poznyak, L.V. Stopolyanskaya, Russ. J.
Inorg. Chem. 44 (1999) 1179–1184.
´
[13] B.S. Parajon-Costa, E.J. Baran, O.E. Piro, Polyhedron 16 (1997)
3379–3383.
´
´
[14] A. Garcıa-Raso, A. Terron, J.J. Fiol, E. Molins, C. Miravitlles,
Polyhedron 14 (1995) 2537–2545.
[38] S. Korman, H.T. Clarke, J. Biol. Chem. 221 (1956) 113–131.
[39] T. Miyazawa, Bull. Chem. Soc. Jpn. 53 (1980) 2555–2565.
[40] P. Mirti, Anal. Chim. Acta 69 (1974) 69–77.
[41] L.J. Belamy, third ed.The Infrared Spectra of Complex Molecules,
vol. 1, Chapman and Hall, London, 1975.
´
[15] A. Garcıa-Raso, J.J. Fiol, B. Adrover, P. Tauler, A. Pons, I.
´
Mata, E. Espinosa, E. Molıns, Polyhedron 22 (2003) 3255–
3264.
[16] N. Okabe, Y. Kohyama, K. Ikeda, Acta Crystallogr. C 51 (1995)
222–224.
[17] W.T.A. Harrison, R.M. Yeates, M.L.F. Phillips, T.M. Nenoff,
Inorg. Chem. 42 (2003) 1493–1498.
[18] N. Okabe, K. Ikeda, Y. Kohyama, Y. Sasaki, Acta Crystallogr. C
51 (1995) 224–226.
[19] R. Chotalia, T.W. Hambley, D.D. Ridley, P. Turner, Acta
Crystallogr. C 52 (1996) 3043–3045.
[20] A.J. Canty, M. Fyfe, B.M. Gatehouse, Inorg. Chem. 17 (1978)
1467–1471.
[21] A.J. Canty, N. Chaichit, B.M. Gatehouse, Acta Crystallogr. B 35
(1979) 592–596.
[22] M.R. Udupa, B. Krebs, Inorg. Chim. Acta 55 (1981) 153–156.
[23] P.T. Beurkens, A. Perales, F.J. Martın-Gil, J. Martın-Gil,
[42] E. Pretsch, P. Buhlmann, C. Affolter, Structure Determination
¨
of Organic Compounds, third ed., Springer-Verlag, New York,
2000.
[43] Bruker, SMART, SAINT, Area Detector Control Integration
Software, Bruker Analytical X-ray Instruments Inc., Madison,
WI, USA, 1999.
[44] G.M. Sheldrick, SADABS. Program for Empirical Absorption
Correction of Area Detector Data, University of Go¨ttingen,
Germany, 1997.
[45] G.M. Sheldrick, Acta Crystallogr. A 46 (1990) 467–473.
[46] G.M. Sheldrick, SHELXL-97. Program for the Refinement of
Crystal Structures, University of Go¨ttingen, Germany, 1997.
[47] E. Keller, SCHAKAL-99. A Computer Program for the Graphic
Representation of Molecular and Crystallographic Models, Uni-
versity of Freiburg, Germany, 1999.
´
´
Monatsh. Chem. 119 (1988) 1189–1199.
[24] M. Mitewa, T. Todorov, G. Gencheva, O. Angelova, T.N.
Bakalova, J. Coord. Chem. 55 (2002) 1367–1372.
[25] A.M. Beja, J.A.C. Paixao, J.M. Gil, M.A. Salgado, Acta
Crystallogr. C 47 (1991) 2333–2336.
[26] M. Ramos Silva, J.A. Paixao, A. Matos Beja, L. Alte da Veiga, J.
Mart´ın-Gil, F.J. Mart´ın-Gil, Acta Crystallogr. C 52 (1996) 2450–
2452.
[27] G. Gencheva, M. Mitewa, P.R. Bontchev, G. Gochev, J.
Macicek, E. Zhecheva, N.D. Yordanov, Polyhedron 11 (1992)
365–370.
[48] Bruker, SHELXTL. Integrated System for the Determination of
Crystal Structures, Bruker AXS Inc., Madison, WI, USA, 2000.
´
´
[49] M.J. Roman-Alpiste, J.D. Martın-Ramos, A. Castineiras-Cam-
˜
´
pos, E. Bugella-Altamirano, A.G. Sicilia-Zafra, J.M. Gonzalez-
´
´
´
Perez, J. Niclos-Gutierrez, Polyhedron 18 (1999) 3341–3351, and
references therein.
[50] A. Addison, T. Rao, J. Reedijk, G. Verschoor, J. Chem. Soc.,
Dalton Trans. (1984) 1349–1356.