Superoxide dismutase activity of ternary copper complexes of sulfathiazole and imidazole derivatives. Synthesis and properties of [CuL2(R-Him)2] [HL = 4-amino-N-(thiazol-2-yl)benzenesulfonamide, R-Him = 4-methylimidazole, 4,4-dimethylimidazoline or 1,2-dimethylimidazole]. Crystal structure of [CuL2(4,4-dimethylimidazoline)2]

Article abstract of DOI:10.1016/S0020-1693(00)00080-3

New ternary copper(II) complexes of sulfathiazole (4-amino-N-(thiazol-2-yl)benzenesulfonamide)(HL) and methyl imidazole derivatives have been synthesised and characterised. The crystal structure of the complex [CuL₂(4,4-dmHim)₂] (1) [4,4-dmHim = 4,4-dimethylimidazoline] has been determined. The copper centre has a quasi regular square planar environment with Cu-nitrogen bond lengths ranging from 1.952 to 2.010 ?. From the spectroscopic properties of the complexes [CuL₂(1,2-dmHim)₂] (2) [1,2-dmHim = 1,2-dimethylimidazole] and [CuL₂(4-mHim)₂] (3) [4-mHim = 4-methylimidazole] a distorted tetragonal octahedral geometry is deduced. The compounds showed SOD mimetic activity in fact, a low concentration of the complexes catalyses the dismutation of superoxide at biological pH. This SOD activity is correlated with their structural properties. Using Extended Huckel Molecular Orbital Calculations the one-electron energy levels of the CuN₄ chromophore in complex 1 are reported and correlated with the data of the CuN₆ and CuN₅ chromophores of the previously reported [CuL₂(Him)₂]·MeOH and [CuL₂(mim)₂]·H₂O compounds. In addition, the influence of the geometry distortion on the composition and energy of the molecular orbitals is described using idealised models. (C) 2000 Elsevier Science S.A.

Full text of DOI:10.1016/S0020-1693(00)00080-3

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 304 (2000) 170–177
Superoxide dismutase activity of ternary copper complexes of
sulfathiazole and imidazole derivatives. Synthesis and properties of
[CuL2(R-Him)2] [HL=4-amino-N-
(thiazol-2-yl)benzenesulfonamide, R-Him=4-methylimidazole,
4,4-dimethylimidazoline or 1,2-dimethylimidazole]. Crystal
structure of [CuL2(4,4-dimethylimidazoline)2]
Javier Casanova, Gloria Alzuet, Sacramento Ferrer, Julio Latorre,
Jose Antonio Ram´ırez, Joaqu´ın Borra´s *
Departamento de Qu´ımica Inorga´nica, Facultad de Farmacia, Uni6ersidad de Valencia, A6enida Vicent Andre´s Estelle´s s/n, 46100-Burjassot,
Valencia, Spain
Received 21 October 1999; accepted 26 January 2000
Abstract
New ternary copper(II) complexes of sulfathiazole (4-amino-N-(thiazol-2-yl)benzenesulfonamide)(HL) and methyl imidazole
derivatives have been synthesised and characterised. The crystal structure of the complex [CuL2(4,4-dmHim)2] (1) [4,4-dmHim=
4,4-dimethylimidazoline] has been determined. The copper centre has a quasi regular square planar environment with Cu-nitrogen
,
bond lengths ranging from 1.952 to 2.010 A. From the spectroscopic properties of the complexes [CuL2(1,2-dmHim)2] (2)
[1,2-dmHim=1,2-dimethylimidazole] and [CuL2(4-mHim)2] (3) [4-mHim=4-methylimidazole] a distorted tetragonal octahedral
geometry is deduced. The compounds showed SOD mimetic activity in fact, a low concentration of the complexes catalyses the
dismutation of superoxide at biological pH. This SOD activity is correlated with their structural properties. Using Extended
Hu¨ckel Molecular Orbital Calculations the one-electron energy levels of the CuN4 chromophore in complex 1 are reported and
correlated with the data of the CuN6 and CuN5 chromophores of the previously reported [CuL2(Him)2]·MeOH and
[CuL2(mim)2]·H2O compounds. In addition, the influence of the geometry distortion on the composition and energy of the
molecular orbitals is described using idealised models. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Sulfathiazole complexes; Imidazole complexes; Copper complexes; Crystal structures; Superoxide-dismutase
integral part of the active site. Copper ions, as centres
of the active site of various metalloproteins, play an
1. Introduction
essential role in biological processes like electron trans-
A unique feature of metalloproteins is their capability
fer, oxidation and dioxygen transport. Except for the
to discriminate between various metal ions which are
Cu(I) metallothioneins, all copper proteins studied con-
present in their natural surroundings. The active site of
tain one or more imidazole residues of histidine bound
a metalloprotein is built up in such a way that only
to copper ion. Among these metalloproteins the super-
specific metal ions fit in, thus making the metal ions an
oxide-dismutase has the metal ion bound to three imi-
dazole from histidines and one imidazolato ion that
acts as a bridging ligand between the copper and zinc
ions [1]. In a previous paper [2], we reported ternary
* Corresponding author. Tel.: +34-96-386 4530; fax: +34-96-386
4960.
E-mail address: joaquin.borras@uv.es (J. Borra´s)
0020-1693/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0020-1693(00)00080-3

J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
171
2.3. Synthesis
copper complexes of sulfathiazole with imidazole (Him)
and with its N-methyl derivative (mim). From their
crystal structures we deduced the importance of the
substituents of the imidazole ring in the coordination
polyhedron of the copper ion likewise in the chro-
mophore. While in the [CuL2(Him)2]·MeOH the metal
ion is in a very distorted octahedron the stereochem-
istry of the [CuL2(mim)2]·H2O is intermediate between
the square pyramidal and trigonal bipyramidal. In the
first compound the geometry adopted by the Cu(II) is
probably due to the stacking interactions between the
phenyl rings from the sulfathiazole ligands and the
imidazole rings. In the latter, the methyl group linked
to the N-imidazole avoids the intramolecular stacking
interactions observed in the [CuL2(Him)2]·MeOH com-
plex so that, a five coordinate geometry is adopted by
the copper ion. Both compounds have high superoxide
dismutase mimetic activity. The parameters involved in
the higher or lower SOD mimetic activity of the copper
complexes in vitro, are in study. Among these factors, it
seems important a limited steric hindrance to the ap-
proach of the superoxide anion, likewise the presence of
coordination sites belonging to nitrogen heteroatomic
rings such as imidazole or pyridine. In the cited paper,
we proposed a correlation between the axial bond
distance and the superoxide mimetic activity. In order
to obtain a better understanding about the possible
correlation between structural factors and the SOD
mimetic activity likewise the influence of the sub-
stituents of the imidazole ring in the coordination poly-
hedron of the sulfathiazole copper complexes, we have
synthesised and characterised several complexes with
different methyl imidazole derivatives that are reported
in the present paper.
2.3.1. [CuL2(4,4-dmHim)2] (1)
CuCl2·2H2O (0.17 g, 1 mmol) was added to a solu-
tion of HL (0.5 g, 2 mmol) in hot methanol (75 cm3).
Then, to the resulting yellowish mixture 4,4-dmHim
(0.98 g, 10 mmol) was added and the solution turned
dark green. It was stirred for about 15 min and then
propan-2-ol (35 cm3) was added. The final solution was
left to stand at 4°C. After 2 months prismatic brown
crystals suitable for X-ray diffraction measurements
were formed. Anal. Calc. for C28H36CuN10O4S4 requires
C, 43.8; H, 4.7; N, 18.2; Cu, 8.3; S, 16.7. Found: C,
43.5; H, 4.5; N, 18.0; Cu, 8.3; S, 16.9%. IR: wmax/cm−1
3440 and 3350 (N-Hsulfathiazole); 3120 (N-Himida-
zole); 1460 (thiazole ring); 1320, 1150 and 560 (SO2);
930 (SꢀN).
2.3.2. [CuL2(1,2-dmHim)2] (2)
The synthesis of the 1,2-dmHim complex was similar
but 1,2-dmHim (0.96 g, 10 mmol) was added instead of
4,4-dmHim. After 3 days a green precipitate was ob-
tained. Anal. Calc. for C28H32CuN10O4S4 requires C,
44.0; H, 4.2; N, 18.3; Cu, 8.3; S, 16.8. Found: C, 43.7;
H, 4.3; N, 18.2; Cu, 8.1; S, 17.0%. IR: wmax/cm−1 3440
and 3350 (N-Hsulfathiazole); 3120 (N-Himidazole);
1450 (thiazole ring); 1320, 1150 and 560 (SO2); 940
(SꢀN).
2.3.3. [CuL2(4-mHim)2] (3)
A solution of HL (0.255 g, 1 mmol) in 50 ml of
methanol was added to 25 ml of a methanolic solution
containing 4-mHim (0.328 g,
Cu(OAc)2·H2O (0.2 g, 1mmol) in an ice bath. Under
these conditions the resulting mixture was stirred for 8
h and then a green solid was obtained. Anal. Calc. for
C26H28CuN10O4S4 requires C, 42.4; H, 3.8; N, 19.0; Cu,
8.6; S, 17.4. Found: C, 41.8; H, 3.9; N, 18.8; Cu, 8.7; S,
17.7%. IR: wmax/cm−1 3440 and 3390 (N-Hsulfathia-
zole); 3120 (N-Himidazole); 1480 (thiazole ring); 1320,
1150 and 560 (SO2); 940 (SꢀN).
4
mmol) and
2. Experimental
2.1. Materials
2.4. X-ray crystallography of complex 1
Copper chloride dihydrate, 4-methylimidazole (4-
mHim), 1,2-dimethylimidazole (1,2-dmHim), 4,4-
dimethylimidazole (4,4-dmHim) and sulfathiazole (HL)
were reagents of grade and used without purification.
Xanthine, xanthine oxidase, superoxide dismutase,
bovine serum albumine and nitrobluetetrazolium were
purchased from Sigma.
2.4.1. Crystal data
C28H36CuN10O4S4, M=767.54, space group=P1,
a=8.849(2), b=9.2945(9), c=11.1876(8) A, h=
(
,
92.491(9)°, i=95.041(12)°, k=109.780(9)°, Z=2,
3
−3
,
V=859.9(2) A , Dc=1.61 g cm
.
2.4.2. Data collection and processing
Analysis on single crystals of C28H36CuN10O4S4 (ap-
proximate size 0.15×0.20×0.20 mm) were carried out
with an Enraf–Nonius CAD-4 single crystal diffrac-
2.2. Physical measurements
Analytical data (C, H, N, S), IR, UV–Vis, EPR
spectra and magnetic measurements were carried out as
described previously [3].
,
tometer (u=0.71073 A). The unit cell dimensions were
measured form the angular setting of 25 reflections with
q between 15 and 25°.

172
J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
parameters of the non-hydrogen atoms were refined.
The reflections were measured in the hkl range (0,
The hydrogen atoms were refined with a common ther-
mal parameter. The final conventional agreement fac-
tors were R1=0.0633 and wR2=0.1693. The maximum
shift of the e.s.d ratio in the last full matrix least-
−11, −13) to (10, 11, 13) between limits: 1.83°BqB
24.97°. The ꢀ–2q scan technique and a variable scan
rate with a maximum scan time of 60 s per reflection
were used. The intensity of the primary beam was
checked throughout the data collection by monitoring
three standard reflections every 3600 s. The final drif
corrections factors were in the range of 0.98 and 1.02.
Profile analysis was performed on all reflections [4,5] a
semiempirical absorption correction, „-scan based, was
performed [6]. In total, there were 3019 reflections, of
which 2295 had Fo\4|Fo. Lorentz and polarisation
corrections were applied and the data were reduced to
Fo values. The structure was solved by the Patterson
method using the program SHELXS-86 [7] running on an
IBM PENTIUM II 300 computer. Isotropic least-
squares refinement was performed by means of the
program SHELXL-93 [8]. Hydrogen atoms were placed
in calculated positions.
squares cycle was 0.001. The final difference Fourier
−3
,
map showed no peaks higher than 1.293 e A and not
deeper than −2.096 e A−3. Atomic scattering factors
,
were taken from the International Tables for X-ray
crystallography [9]. The molecular plots were produced
by program ORTEP [10].
2.5. Superoxide-dismutase assay
The superoxide dismutase activity of the metal com-
plexes was determined according to Oberley and Spitz
[11] with minor modifications [3].
2.6. EHMO calculations
During the final stages of the refinement the posi-
tional parameters and the anisotropical thermal
All calculations were performed by using the Package
of Programs for Molecular Orbital Analysis by Mealli
and Proserpio [12] based on CDNT (atom Cartesian
coordinate calculations), ICON (extended Hu¨ckel
method with the weighted Hij formula) and FMO
(fragment molecular orbital), including in the drawing
program CACAO (computer aided composition of
atomic orbitals).
The extended Hu¨ckel parameters are as follows: Hij:
Cu 4s, −11.40 eV; Cu 4p, −6.06 eV; Cu 3d, −14.00
eV; N 2s, −26.00 eV; N 2p, −13.401.950 eV; S 3s,
−20.30 eV; S 3p, −14.00; O 2s, −32.30 eV; O 2p,
−14.80 eV; C 2s, −21.40 eV; C 2p, −11.40 eV; H 1s,
−13.60 eV. Orbital exponents (Contraction Coeffi-
cients in Double-x Expansion given in Parentheses): Cu
4s, 4p, 2.20; Cu 3d, 5.95(0.5933), 2.30(0.5744); N 2s, 2p,
1.950; S 3s, 3p, 1.900; O 2s, 2p, 2.275; C 2s, 2p, 1.625;
H 1s, 1.300.
Table 1
,
Selected bond lengths (A) and bond angles (°) for [CuL2(4,4-
dmHim)2] a
CuꢀN(2)
CuꢀN(1)c1
1.952(3) CuꢀN(2)c1
2.010(3) CuꢀN(1)
1.952(3)
2.010(3)
N(2)ꢀCuꢀN(2)c1
N(2)ꢀCuꢀN(1)c1
N(2)c1ꢀCuꢀN(1)c1
180.0
N(1)c1ꢀCuꢀN(1) 180.0
88.0(1) N(2)c1ꢀCuꢀN(1)
88.0(1)
91.9(1)
91.9(1) N(2)ꢀCuꢀN(1)
a Symmetry transformations used to generate equivalent atoms:
c1 −x,−y,−z.
3. Results and discussion
3.1. Crystal structure of [CuL2(4,4-dmHim)2] (1)
Relevant bond distances and angles are given in
Table 1. Fig. 1 shows the numbering scheme and
thermal ellipsoids drawn at the 30% probability level.
The copper ion is bound centrosymmetrically by four
ligands forming a square planar arrangement that con-
sists of two imidazole nitrogen atoms and two thiazole
nitrogen atoms of the sulfathiazolato anions in a trans
,
disposition. The CuꢀNthiazole distances are 2.010(3) A
which are similar to those determined for related Cu(II)
coordination compounds [2,3]. The CuꢀNimidazole dis-
,
tances are 1.954(3) A, such short CuꢀN distances are
also present in previously reported Cu complexes with
imidazole ligands [13,14].
Fig. 1. ORTEP drawing of the complex [CuL2(4,4-dmHim2] (1) (ther-
mal ellipsoids drawn at the 30% probability level.

J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
173
3.2. Infrared spectra
The IR spectra of the compounds have in common
the bands for the stretching and bending modes of the
NꢀH groups from sulfathiazole and imidazole. As it
was found for other copper–sulfathiazole complexes
[2,3] the bands at 3500 and 3380 cm−1 assigned to
stretching vibrations of the NꢀH groups of the sul-
fathiazole are shifted to higher frequencies respect to
those of the free ligand. The band corresponding to
the stretching NꢀH imidazole ring vibration appears at
approximately 3230 cm−1. In all the complexes the
bands attributed to the SO2 vibrations remain un-
changed. Evidence of the deprotonation of the sul-
fathiazole ligand and its coordination through the
thiazole N atom arises from the significant changes of
the bands assigned to the thiazole ring and the SꢀN
stretching vibrations at 1550 and at 920 cm−1, respec-
tively, which appear at 1480–1450 and at 950–940
cm−1 in the complexes.
Fig. 2. Reflectance spectra. (a) [CuL2(4,4-dmHim)2] (1); (b)
[CuL2(1,2-dmHim)2] (2); (c) [CuL2(4-mHim)2] (3); (d)
[CuL2(Him)2]·MeOH; (e) [CuL2(mim)2]·H2O.
3.3. Electronic absorption spectra
The coordination geometry of the metal in the
complex is virtually square planar with NꢀCuꢀN
bond angle deviations from 90° of less than 2°.
The Cu(II) is in the N4 plane. Tetrahedrality for
any tetracoordinate copper complex can be deter-
mined from the angle subtended by two planes,
each encompassing the copper and two adjacent atoms
[15]. For strictly square planar complexes with D4h
symmetry, the tetrahedrality is 0°. For tetrahedral
complexes with D2d symmetry, the tetrahedrality
equals 90°. The dihedral angle for [CuL2(4,4-dmHim)2]
complex of 0° agrees with its regular square planar
geometry.
The sulfathiazolato anion acts as a monodentate
ligand through the N atom of the thiazole ring. It
should be noted that this coordination behaviour, that
has been also observed in the [CuL(py)3Cl] complex
[3], differs from that exhibited by the sulfathiazole
ligand in the [CuL2(Him)2]·MeOH and [CuL2-
(mimH)2]·H2O compounds [2]. In contrast to the semi-
coordination of the sulfonamidate N atom found in
Fig. 2 shows the diffuse reflectance spectra of the
complexes 1, 2 and 3 and the corresponding ones to
the previously reported [CuL2(Him)2]·MeOH and
[CuL2(mim)2]·H2O. Several interesting aspects related
with the nature of the chromophore of the complexes
can be deduced from comparison of these diffuse
reflectance spectra. The solid spectrum of [CuL2(4,4-
dmHim)2] (1) exhibits three maxima at 638, 455 and
340 nm. The first one corresponds to a d–d band in a
square planar environment around Cu(II) which is in
agreement with the crystal structure of the compound.
The second one could be attributed to a d–d or to a
charge transfer ligand-to-metal transition. Square pla-
nar copper (II) complexes with a regular CuN4 chro-
mophore show a characteristic d–d band at 500 nm
[16]. Since the crystal structure of [CuL2(4,4-dmHim)2]
(1) reveals a quasi regular square planar CuN4 entity
it seems reasonable to suggest that the intense band at
455 nm includes both transitions. The third band is
assigned to a ligand-to-ligand transition of the sul-
fathiazolato anion. The [CuL2(1,2-dmHim)2] (2) and
the [CuL2(4-mHim)2] (3) compounds display spectral
features different from those of [CuL2(4,4-dmHim)2].
Both spectra have a similar pattern to that of the
[CuL2(Him)2]·MeOH suggesting the same chro-
mophore. On the basis of the crystal of the Him
complex, that showed a very distorted tetragonal elon-
gated octahedral CuN6 entity, we propose a similar
coordination polyhedra for compounds 2 and 3.
,
these complexes (CuꢀN bond lengths, 2.68 and 2.57 A,
respectively), in the [CuL2(4,4-dmHim)2] such interac-
tion is not presented, being the CuꢀNꢀsulfonamidate
,
distance of 3.146 A. Though there is no obvious expla-
nation, the lack of bonding may be due to crystal
packing requirements owing to the steric hindrance of
the two methyl substituents of the imidazole ring.
Moreover, this fact leads to a different copper (II)
coordination polyhedra in the three complexes. While
the [CuL2(4,4-dmHim)2] presents a square planar envi-
ronment, the [CuL2(Him)2]·MeOH is a very distorted
3.4. Magnetic measurements and EPR spectra
octahedra and the [CuL2(mim)2]·H2O exhibits
slightly distorted square pyramidal geometry.
a
Magnetic moments values at room temperature (r.t.)
of 1.80 MB for all the complexes are consistent with the

174
J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
presence of one unpaired electron in monomeric
Cu(II) coordination compounds.
octahedral geometry around the Cu(II) which is in
good agreement with the electronic spectra. The lower
gꢀꢀ for complex 1 compared to those of the complexes
2 and 3 is consistent with the square planar stereo-
chemistry of the former.
The polycrystalline X-band EPR spectra of the com-
plexes 1, 2, and 3 at room temperature are shown in
Fig. 3. These EPR spectra are axial with gꢀꢀ\gÞ and
are consistent with a distorted tetragonal geometry.
The EPR parameters were extracted by simulation
[17]. The EPR of complex 1 has a parallel region
poorly resolved, being the EPR parameters: gꢀꢀ:2.20
and gÞ=2.07. The lack of copper(II) hyperfine cou-
pling in this complex is likely due to dipole–dipole
interactions between copper atoms of neighboring
molecules. The EPR of the complexes 2 and 3 show
the hyperfine structure in the gꢀꢀ region. The EPR
parameters are gꢀꢀ=2.28 and 2.25; gÞ=2.08 for both
complexes and Aꢀꢀ=165 and 170 G, respectively. Ac-
cording to Sakaguchi and Addison [18] the value of
the gꢀꢀ/Aꢀꢀ quotient of 138 cm for compound 2 and of
132 cm for complex 3 suggests a distorted tetragonal
3.5. Superoxide-dismutase assays
The superoxide dismutase activity of the complexes
has been measured. All the compounds exhibit cata-
lytic activity toward the dismutation of the superoxide
anions. Fig. 4 shows percentage inhibition of the re-
duction of nitrobluetetrazolium plotted against the
concentration of the complex [CuL2(4,4-dmHim)2] (1).
Superoxide was supplied enzymatically from hypoxan-
thine–xanthine oxidase reaction to the evaluating sys-
tem. The mode of generation of superoxide anions
must be emphasized: it has been shown through the
use of pulse radiolysis techniques for supplying super-
oxide anions that most of the copper compounds cata-
lyze the dismutation with similar efficiency to that of
the native SOD [19]. This is due to the fact that pulse
radiolysis techniques provided O2− in high concentra-
tion (10−5 M). When the superoxide is provided from
a biological source, the concentration is about 10−15
M and only few copper complexes exhibit SOD activ-
ity because of the reversibility of the first step of the
dismutation reaction.
In the experimental conditions used to evaluate the
SOD like activity, the phosphate ion is present in a
very large excess with respect to the complexes. It has
been demonstrated that the species [Cu(HPO4)] is
much less active than some dipeptide copper (II) com-
plexes and obviously less than the active enzyme itself
[20]. So, the SOD mimetic activity of the complexes
reported here could not be due to the formation of
some amount of [Cu(HPO4)] species.
Fig. 3. EPR spectra. (a) [CuL2(4,4-dmHim)2] (1); (b) [CuL2(1,2-
dmHim)2] (2); (c) [CuL2(4-mHim)2] (3).
The IC50 value of 0.742 mM for complex 1, 1.03 mM
for complex 2 and 0.586 mM for complex 3 indicates
that they are strong mimetic superoxide dismutase
compounds.
The percentage of inhibition of NBT reduction was
tested in solutions containing 0.75 mM (complex 1), 1.0
mM (complex 2) and 0.6 mM (complex 3) and 1.5 mM of
BSA (bovine serum albumin). The inhibition of NBT
(%) was 49.6 for complex 1, 49.8 for complex 2 and
51.5 for complex 3. The same solutions without BSA
give inhibition values of 51.2, 47.3 and 52.6%, respec-
tively. As consequence, the activity of the complexes
was not significantly influenced by the physiological
chelator albumin. Recently it was proposed that only
flexible Cu(II)/Cu(I) chelates are genuine active site
analogues of SOD [21]. In fact, SOD mimicking sys-
tems should permit the coordination of both Cu(II) and
Cu(I) ions. The SOD activity is related to the equatorial
Fig. 4. Percentage inhibition of nitroblue tetrazolium reduction plot-
ted against the concentration of the [CuL2(4,4-dmHim)2] complex.
Each point represents the mean9standard deviation of triplicate
determination.

J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
175
Fig. 5. Energy levels of d orbitals for the CuN6, CuN5 and CuN4 idealised models.
field experienced by the copper(II) ions [20,22–25]. A
strong equatorial field opposes the interaction of the
complexes with superoxide radical, disfavouring the
probable formation of the intermediate copper-superox-
ide adduct [26]. Moreover, a strong field is a factor
which disfavours the reduction process from Cu(II) to
Cu(I). The higher IC50 value and therefore the lower
SOD activity for complex 1 compared with complex 3
and the previously reported [CuL2(Him)2]·MeOH
(IC50=0.664 mM) and [CuL2(mHim)2]·H2O (IC50=
0.429 mM) compounds could be correlated with the
strong field experienced by the Cu(II) in a square
planar geometry. Considering this factor, it must be
expected that complex 2 would have a superoxide dis-
mutase activity similar to that of complex 3 and higher
than the corresponding one to complex 1. However, its
IC50 value indicates the lowest activity. Although there
is not an obvious explanation it could be attributed to
the higher steric hindrance of the two methyl groups
that could avoid the approach of the superoxide anion
in the coordination sphere of the metal ion.
derived from the crystal structures, so it is possible to
calculated the complete molecular correlation diagrams
for the three complexes. From these, it must be noted
that in the SOMO’s of the three complexes, besides the
contribution of the metal orbitals, the only atomic
orbitals of the ligands that participate are the Nimida-
zole and Nthiadiazole donor atoms, being their relative
contribution around the 25%. As it can expected from
the similar CuꢀN bond distances observed in the crystal
structures there is not significative difference in the
participation of both Nimidazole and Nthiazole atoms.
With respect to the d copper orbitals involved in the
SOMO’s, the main contribution corresponds to the dxy
(27%) and dz (9%) for the [CuL2(4,4-dmHim)2] (1)
2
complex, to the dx
(16%); dyz (10%) and dz (8%)
2
2
2
−y
for the [CuL2(mim)2]·H2O complex and to the dyz (26%)
and dxz (8%) for the [CuL2(Him)2]·MeOH complex.
The systems are so complicated that it is very difficult
to obtain results that can be correlated to the electronic
properties of the complexes. To clarify these calcula-
tions we have designed three idealised models substitut-
ing the six, five and four donor atoms of the ligands in
the three copper complexes by ammonia molecules
using the coordinates of the N donor atoms of the
ligands obtained from the crystallographic data. In this
way, in the idealised models the copper ion and the N
atoms retain the sites found in the crystal structures
and the distortions observed in the coordination poly-
hedron of the three complexes remain. The influence of
the distortion in the energy and the nature of the
molecular orbitals involving the copper d orbitals can
be established from the diagram deduced from the
EHMO calculations for the three idealised models
where that for a regular octahedral Cu(NH3)62+ is also
3.6. Molecular orbital calculations
We have obtained crystal structures of copper-sul-
fathiazolato complexes with Him, mim and 4,4-dmHim
whose coordination polyhedron ranges from a very
irregular distorted octahedral through a distorted
square pyramidal to a quasi regular square planar
geometries. To investigate the electronic structures of
the three complexes we have carried out Extended
Hu¨ckel Molecular Orbital (EHMO) calculations by
means of the CACAO program [12]. This program
permits us to use the crystallographic coordinates

176
J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
included (Fig. 5). From this we can deduced several
important aspects:
Acknowledgements
The SOMO’s present a participation of the dxy or-
bital of 43% in all the models. Furthermore, the energy
of the SOMO’s is slightly different, being the corre-
sponding to the regular octahedron the most stable
(−10.934 eV) and that of the distorted one the most
unstable (−10.745 eV).
We greatly appreciate financial support from
Comision Interministerial de Ciencia y Tecnolog´ıa
(PM97-0105-C02-01) and the Generalitat Valenciana
(GV-D-CN-09-145-96).
In all the models the dz orbital is the main com-
2
ponent of the HOMO. At the same time is the orbital
that have more energy variation when go from the
regular octahedron (−12.106 eV) to the distorted one
(−12.890 eV), to the distorted square pyramid
(−12.999 eV) and to the quasi regular square planar
(−13.793 eV).
The influence of the distortion in the coordination
polyhedron is clearly observed in the HOMO-1,
HOMO-2 and HOMO-3. In the regular geometries they
are constituted by individual d orbitals, being the en-
References
[1] I. Bertini, H.B. Gray, S.J. Lippard, J.S. Valentine, Bioinorganic
Chemistry, University Science Books, Mill Valley, USA, 1994,
pp. 298–310.
[2] J. Casanova, G. Alzuet, J. Borra´s, O. Carugo, J. Chem. Soc.,
Dalton Trans. (1996) 2239.
[3] J. Casanova, G. Alzuet, J. Borra´s, J. Latorre, M. Sanau, S.
Garc´ıa-Granda, J. Inorg. Biochem. 60 (1996) 219.
[4] M.S. Lehman, F.K. Larsen, Acta Crystallogr., Sect. A 30 (1974)
580.
ergy order dx
2 \dxz]dyz while in the two distorted
2
[5] D.F Grant, E.F. Gabe, J. Appl. Crystallogr., Sect A 11 (1978)
114.
[6] A.C.T. Nort, D.C. Philips, F.S. Mathews, Acta Crystallogr.,
Sect. A 24 (1968) 351.
[7] G.M. Sheldrick, C. Kruger, R. Goddard (Eds.), Crystallographic
Computing, Clarendon, Oxford, UK, 1985, p. 175.
[8] G.M. Sheldrick, SHELXL-93: Program for Crystal Structure
Refinement, Institute fu¨r Anorgannishe Chemie der Universitat
Gottingen, Go¨ttingen, Germany, 1993.
[9] D.T. Cromer, J.T. Waber, International Tables for X-ray Crys-
tallography, vol. IV, Kynoch, Birmingham, UK, 1974.
[10] C.K. Johnson, ORTEP, Report ORNL-3794, Oak Ridge National
Laboratory, Oak Ridge, TN, 1971.
models they a−rey a mixture of the dx
2, dxz and dyz
2
−y
orbitals because of the distortion leads to a significant
interaction of the dxz and dyz orbitals with the ligand
ones.
According from these calculations in the electronic
spectrum of the complexes it can be expected only
one band at about 415 nm for the [CuL2(4,4-
dmimH)2] (1) complex, two bands at approximately
595 and 425 nm for the [CuL2(mim)2]·H2O complex,
three bands at about 580, 475 and 405 nm for the
[CuL2(Him)2]·MeOH complex and two bands near 500
and 422 nm for the idealised hexacoordinate
[11] L.W. Oberley, D.R. Spitz, in: R.A. Greenwald (Ed.), Handbook
of Methods for Oxygen Radicals Research, CRC Press, Boca
Raton, FL, 1986, pp. 217–220.
Cu(NH3)62+
.
If we compare these theoretical electronic transitions
with those of the reflectance spectra of the complexes
(Fig. 2) we observe a good agreement although an
accurate assignment is difficult to made due to the
width of the bands of the experimental spectra and the
trouble to individualise the d metal atomic orbitals
contribution to the molecular orbital of the complexes
as a consequence of the distortion of the coordination
polyhedron and of the known limitations of this type of
calculations.
Although several attempts to interpret the visible
spectrum of the transition metal complexes using the
deconvolution curves have been reported, from the
results obtained in this paper it seems very difficult to
distinguish the energy levels of the individual d orbitals
when some distortion takes places.
[12] C. Mealli, D. Proserpio, Computer Aided Composition of
Atomic Orbitals (CACAO program) PC version, July 1992. See
also: J. Chem. Educ. 67 (1990) 399.
[13] L. Antolini, L.P. Battaglia, A. Bonamartini-Corradi, G. Marcot-
rigiano, L. Menabue, M. Saladini, Inorg. Chem. 21 (1982) 1391.
[14] L.P. Battaglia, A. Bonamartini-Corradi, L. Menabue, G.C.
Pellacani, R. Prampolini, M. Saladini, J. Chem. Soc. (1982)
781.
[15] L.P. Battaglia, A. Bonamartini-Corradi, G. Marcotrigiano, L.
Menabue, G.C. Pellacani, Inorg. Chem. 18 (1979) 148.
[16] V. Ravichandran, K.K. Chacko, A. Aoki, H. Yamazaki, J.
Ruiz-Sanchez, J. Suarez-Varela, J.D. Lopez-Gonzalez, J.M.
Salas-Peregrin, E. Colacio-Rodriguez, Inorg. Chim. Acta 173
(1990) 107.
[17] WINEPR-Simfonia.1.25. Bruker Analytik GmbH, Karlsruhe,
FRG, 1994–1996.
[18] U. Sakaguchi, A.W. Addison, J. Chem. Soc., Dalton Trans.
(1979) 600.
[19] J.L. Pierre, P. Chautemps, S. Refaif, C. Bequin, A. El Marzouki,
G. Serratrice, E. Saint-Aman, P. Rey, J. Am. Chem. Soc. 117
(1995) 1965.
[20] L.L. Constanzo, G. de Guidi, S. Giufrida, E. Rizarelli, G.
Vechio, J. Inorg. Biochem. 50 (1993) 273.
4. Supplementary material
Atomic coordinates, thermal parameters, bond
lengths and angles and structure factors are available
from the authors upon request.
[21] M. Linss, U. Weser, Inorg. Chim. Acta 138 (1987) 163.
[22] R.P. Bonomo, E. Conte, R. Marchelli, A.M. Santoro, G. Tabbi,
J. Inorg. Biochem. 53 (1994) 127.

J. Casano6a et al. / Inorganica Chimica Acta 304 (2000) 170–177
177
[23] R.P. Bonomo, A. De Flora, E. Rizarelli, G. Vecchio, J. Inorg.
Biochem. 59 (1995) 773.
[25] S. Kubota, J.T. Yang, Proc. Natl. Acad. Sci USA 84 (1984)
3283.
[24] R.P. Bonomo, R. Marchelli, G. Tabbi, J. Inorg. Biochem. 60
(1995) 205.
[26] G. Condorelli, L.L. Constanzo, G. De Guidi, S. Giuffrida, E.
Rizarelli, G. Vecchio, J. Inorg. Biochem. 54 (1994) 257.
.