Interaction of VIVO, VVO2 and CuII with a peptide analogue SalGly-L-Ala

Article abstract of DOI:10.1002/ejic.200200511

The formation of complexes of a tripeptide analogue, salicylglycyl-L-alanine [HOC₆H₄C(O)NHCH₂C(O)NHCH(CH₃) -COOH, H₂SalGly-L-Ala], was studied with Cu^II, V^IVO and V^VO₂ in aqueous solution using pH-potentiometric and spectroscopic (UV/Vis, CD, EPR and ⁵¹V NMR) techniques. The results demonstrated the ambidentate character of the ligand. The metal ion-induced deprotonation and subsequent coordination of the two neighbouring amide groups was shown to occur in a cooperative way in the pH range 5-6. At pH ≈ 6.5 a complex with an (O^-, 2 x CON^-, COO^-) binding set becomes predominant for both Cu^II, and V^IVO. The affinity of the phenolate-O^- for V^IVO is high and the ligand is bound to this metal ion even at pH values greater than 10. Conversely, the affinity for V^VO2 is significantly lower and no interaction between this metal oxo-ion and the ligand could be detected in the pH range 2-12. In contrast, with the dipeptide analogue H2SalGly, chelation involving the deprotonated amide-N-could be unambiguously detected. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200200511

FULL PAPER
Interaction of VIVO, VVO2 and CuII with a Peptide Analogue SalGly-
-Ala
L
[a]
[b]
[c]
[d]
[e]
´
´
´
´
´
Tamas Jakusch, Agnes Dörnyei, Isabel Correia, Lıgia M. Rodrigues, Gabor K. Toth,
[a]
[c]
[c]
´
˜
˜
Tamas Kiss,* Joao Costa Pessoa,* and Susana Marcao
Keywords: Vanadium / Copper / Peptides / Speciation / Spectroscopy / Bioinorganic chemistry
The formation of complexes of a tripeptide analogue, salicyl- predominant for both CuII and VIVO. The affinity of the
glycyl-
COOH, H2SalGly-
L
-alanine
[HOC6H4C(O)NHCH2C(O)NHCH(CH3)- phenolate-O− for VIVO is high and the ligand is bound to this
-Ala], was studied with CuII, VIVO and metal ion even at pH values greater than 10. Conversely, the
L
VVO2 in aqueous solution using pH-potentiometric and spec- affinity for VVO2 is significantly lower and no interaction be-
troscopic (UV/Vis, CD, EPR and 51V NMR) techniques. The
results demonstrated the ambidentate character of the ligand.
tween this metal oxo-ion and the ligand could be detected in
the pH range 2−12. In contrast, with the dipeptide analogue
The metal ion-induced deprotonation and subsequent coor- H2SalGly, chelation involving the deprotonated amide-N−
dination of the two neighbouring amide groups was shown to could be unambiguously detected.
occur in a cooperative way in the pH range 5−6. At pH ഠ 6.5 a ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
complex with an (O−, 2 × CON−, COO−) binding set becomes
Germany, 2003)
Introduction
the systems VIVO2ϩϪGSH and VIVO2ϩϪGSSG have been
recently studied.[4Ϫ5]
In order that simple di- or tripeptides bind metal ions
strongly enough, the amide-NϪ should be involved in coor-
dination, but in some cases suitable anchoring donors
should be available to promote amide-NH deprotonation
and subsequent coordination. Various metal ions were
found to engage in this type of bonding, such as PtII, PdII,
Vanadium is an important trace metal with a great variety
of physiological effects, including its interference with vari-
ous phosphate-metabolizing reactions.[1] This suggests inter-
actions between vanadium ions and protein side-chain
donors. Peptides are not the best molecules for modeling
metal ion binding to proteins, since the specially arranged
side-chain donor groups of the amino acid units, determined
by the tertiary structure of the proteins, are usually the pri-
mary bindings sites of proteins, while in the oligopeptides
the N- and C- terminal donors are the basic sites for metal
ion coordination. However, peptides are the most closely re-
lated models for proteins, and in some cases proteins may
have specific terminal metal ion bindings sites as in the case
of albumin.[2] For VIVO one of the known biologically im-
portant interactions with peptides is that with the important
cell constituent and natural tripeptide, glutathione (GSH).
In fact it is believed that GSH may play a role in the redox
transformations of vanadium in the cells.[3] The equilibria in
[7]
CuII, NiII,[6] VVO2 and VIVO.[8] The usual anchoring do-
nor for binding metal ions, including CuII, and inducing am-
ide deprotonation is the terminal NH2 group,[6] although in
some cases the terminal COOϪ or the side chain thiolate-
SϪ or imidazol-N also proved to be efficient at inducing this
process.[6] For VIVO2ϩ, the terminal NH2 is not a suitable
anchor,[9] but phenolate proved to be an efficient anchoring
donor in cases such as H2salGly (2-OH-hippuric acid) where
the pK for the amide-NH deprotonation/coordination of
4.76 was found to be similar to values seen for VIVO and
CuII.[8,10]
In this work the binding capability of the tripeptide ana-
logue H2SalGly--Ala (1) containing a phenolate anchoring
donor has been studied, and the solution speciations and
binding modes of VIVO, VVO2 and CuII complexes of 1 are
reported. In some cases complexes were obtained in the
solid state and characterized.
[a]
Biocoordination Chemistry Research Group of the Hungarian
Academy of Sciences,
P. O. Box 440, 6701 Szeged, Hungary
[b]
University of Szeged, Department of Inorganic and Analytical
Chemistry,
P. O. Box 440, 6701 Szeged, Hungary
[c]
´
´
Centro Quımica Estrutural, Instituto Superior Tecnico,
1049-001 Lisboa, Portugal
[d]
[e]
´
Departamento de Quımica, Universidade do Minho,
Results and Discussion
Largo do Paco, 4709 Braga, Portugal
¸
University of Szeged, Department of Medical Chemistry,
´
´
Dom ter 8, 6720 Szeged, Hungary
The tripeptide analogue H2SalGly--Ala (1) was synthe-
sized by the procedure described in the Exp. Sect. Its for-
Supporting information for this article is available on the WWW
under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122
DOI: 10.1002/ejic.200200511
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2113

T. Kiss, J. Costa Pessoa et al.
FULL PAPER
mula together with that of H2SalGly are included in
Scheme 1.
Scheme 1
Interactions of SalGly-L
-Ala with CuII and VIVO
The protonation and VIVO and CuII complex formation
constants are in Table 1. H2SalGly--Ala contains two dis-
sociable protons: on the phenolic OH and the terminal
Figure 1. Species distribution curves as a function of pH for the
COOH groups. The log KOH value of 7.81 can be assigned CuIIϪSalGly--Ala system at a 1:2 ratio and cCu ϭ 0.002 
to the less acidic phenolic OH and the log K value of 3.37
to the terminal COOH group. These data are similar to
those of the corresponding dipeptide analogue, H2SalGly
(log KOH ϭ 8.16 and log KCOOH ϭ 3.37).[8]
Table 1. Protonation (log K) and CuII and VIVO complex formation
constants (log β with 3 S.D. values in parentheses) of the complexes
of SalGly--Ala at T ϭ 25 °C and I ϭ 0.2  (KCl)
Species
CuII
VIV
O
log K(phen-OϪ)
log K(COOϪ)
MLH
ML
MLHϪ1
7.81(1)
3.37(2)
9.49(8)
9.98(11)
6.43(5)
1.06(8)
Ϫ4.31(6)
245
0.0032
3.55
4.45(5)
Ϫ0.49(6)
Ϫ6.64(5)
297
0.0017
5.03
MLHϪ2
No. of points
Fitting[a] (mL)
pK(MLH)
pK(ML)
Figure 2. Species distribution curves as a function of pH for the
VIVOϪSalGly--Ala system at a 1:2 ratio and cVO ϭ 0.002 .
Dashed lines indicate a higher uncertainty and thus only the tend-
ency of the speciation as a result of slow pH equilibration (see
Exp. Sect.)
4.92
5.37
pK(MLHϪ1
)
6.16
5.37
[a]
Average difference between the experimental and calculated ti-
tration curves expressed in mL of the titrant.
the formation of the protonated species [MLH]ϩ is indicated
by the speciation curves. It can also be seen in Figure 3 that
Based on earlier speciation studies of the corresponding as complex formation proceeds, the d-d transition of the
dipeptide H2SalGly[8,10] and also Gly-like tripeptides,[6] the CuII-ligand system is shifted to higher energies and becomes
pH-metric titration curves measured for the CuIIϪ and constant at a pH of around 7.5, indicating the exclusive for-
VIVOϪSalGly--Ala systems were evaluated by assuming mation of a single species [CuLHϪ2 2Ϫ (λmax ϭ 545 nm, ε ϭ
]
1:1 complex formation at the different states of protonation. 240 Ϫ1·cmϪ1). At the same time in the corresponding VIVO
Whenever 2:1 complexes were included in addition to the system the visible spectra remain almost constant in the pH
1:1 complexes, they were always rejected by the PSEQUAD range 6.5Ϫ8.0 ([VOLHϪ2
]
2Ϫ: λmax ϭ 960 nm (ε ϭ 120
computer program. The best fit between the experimental

Ϫ1·cmϪ1) and 690 nm (ε ϭ 115 Ϫ1·cmϪ1). However, at
and the calculated titration curves was obtained with the set higher pH the intensity of the absorption decreases as the
of species listed in Table 1. The concentration distribution ligand is displaced by OHϪ. The CD spectra depicted in
curves of the complexes formed in the CuIIϪSalGly--Ala Figures 4 and 5 are in full agreement with these obser-
and VIVOϪSalGly--Ala systems are depicted in Figures 1 vations.
and 2.
At pH Ͻ 5 the CD spectra of the VIVOϪSalGly--Ala
In agreement with the speciation curves the visible spectra system have relatively low intensities (species [VOLH]ϩ and
of the metal-ligand systems (e.g. Figure 3 for the CuII sys- [VOL]). At pH Ͼ 5 their intensities increase (parallel with
tem) start to deviate from that of the corresponding aqua the increase in the visible absorption), and three peaks de-
ions [Cu(H2O)6]2ϩ or [VO(H2O)5]2ϩ only at pH Ͼ 3, where velop in the CD spectra at λmax 760 nm (∆ε Ͻ 0), 610 nm
2114
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122

Interaction of VIVO, VVO2 and CuII with a Peptide Analogue
FULL PAPER
at λ ϭ 545 nm which is apparent by a pH of approximately
7 (see Figure 3), corresponds to two bands in the CD spectra
at 490 nm (∆ε ϭ Ϫ0.3 Ϫ1·cmϪ1) and at 610 nm (∆ε ϭ
Ϫ0.28 Ϫ1·cmϪ1). The significant blue shift in the absorp-
tion spectra and the intense CD bands indicate coordination
of the amide NϪ in the corresponding complex. In the CD
spectrum of the GlyGly--AlaϪCuII solutions, a band with
λmax at 475 nm (∆ε ϭ Ϫ0.8 Ϫ1·cmϪ1) was observed.[11] The
different CD spectra obtained here confirm that in the CuII
complexes formed with 1, which predominate at pH Ͼ 5,
the phenolate OϪ is bound to the metal ion.
The visible and CD spectra obtained at different pH val-
ues were used to calculate the spectrum of each individual
species formed in the CuIIϪ and VIVOϪSalGly--Ala sys-
tems, using the PSEQUAD computer program. Hydrolysis
and observable increases in the absorbances of the very
slightly cloudy solutions, especially in the case of the
VIVOϪSalGly--Ala system, precluded the use of visible
spectroscopic data to determine individual spectra precisely.
CD data were more suitable for such a treatment. The fact
that reasonable CD spectra could be simulated for each
species (see Figure 6) confirms that the speciation models
proposed and the stability constants calculated from the pH-
metric measurements are correct.
Figure 3. Visible absorption spectra of the CuIIϪSalGly--Ala sys-
tem at a 1:2 metal ion to ligand ratio at several pH values and cCu ϭ
0.006 
It is worth noting here that significant changes and in-
creases in the CD intensities can be observed only upon ap-
pearance of the [MLHϪ2
is reflected in the calculated CD spectra of [MLHϪ2
]
2Ϫ species in both systems, and this
]
2Ϫ. This
indicates that coordination of the closest donor atom to the
chiral centre (the first amide NϪ from the C terminus) pre-
sumably occurs only in this species.
The EPR spectra help further to elucidate the binding
modes of the complexes formed in the two systems. For the
VIVO systems, Chasteen[12] introduced an additivity rule to
est
est
Figure 4. CD spectra of the CuIIϪSalGly--Ala system at a 1:2
metal ion to ligand ratio at several pH values and cCu ϭ 0.006 
estimate the hyperfine coupling constants A (A ϭ ΣA||,i),
||
||
based on the contributions A||,i of each of the four equatorial
donor groups, its estimated uncertainty being Ϯ 3 ؋ 10Ϫ4
cmϪ1. Most of the A||,i parameters were presented by Chas-
teen[12] while the others were presented by Cornman et al.
[A||(NϪ) ϭ 34Ϫ35 ؋ 10Ϫ4 cmϪ1],[13] Tasiopoulous et al.
[A||(NϪ) ϭ 34Ϫ43 ؋ 10Ϫ4 cmϪ1],[14] Hamstra et al. [A||(Oamide
)
ϭ 43Ϫ44 ؋ 10Ϫ4 cmϪ1],[15] and Costa Pessoa et al.
[A||(NϪ) dipeptides ϭ 34Ϫ36 ϫ 10Ϫ4 cmϪ1
and
[16]
A||(COOϪ) ϭ 42 ؋ 10Ϫ4 cmϪ1].[17] As can be seen, there is
a rather high uncertainty in the reported A||(NϪ) values:
36Ϯ4 ؋ 10Ϫ4 cmϪ1. We chose the value 39 ؋ 10Ϫ4 cmϪ1
for CONϪ, which is within the uncertainty range. This was
chosen without any theoretical precedent but because it gave
a better agreement with the experimental coupling constants
of the present system. Although a similar additivity rule has
not been developed for CuII complexes, the large amount of
EPR data available in the literature also allows us to make
similar estimates for the coordinating donors in the corre-
sponding CuII complexes. A rigorous comparison of the
EPR spectra with the pH-metrically determined speciation
Figure 5. CD spectra of the VIVOϪSalGly--Ala system at a 1:2
metal ion to ligand ratio at several pH values and cVO ϭ 0.0045 
(∆ε Ͼ 0) and at 415 nm (∆ε Ͼ 0) (Figure 5), indicating coor- curves may indicate some differences in the relative amount
dination of the amide NϪ group close to the chiral centre. of some of the species at given pH values. This may be ex-
Similarly, in the CuIIϪSalGly--Ala system the visible band plained by the differences in the temperature, which was
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2115

T. Kiss, J. Costa Pessoa et al.
FULL PAPER
Figure 7. Low-field region of frozen EPR spectra (X-band) re-
corded at 77 K on the CuIIϪSalGly--Ala system at a 1:2 metal
ion to ligand ratio at different pH values, cCu ϭ 0.006 
Figure 6. Simulated CD spectra for each individual species formed
in (a) the CuIIϪSalGly--Ala system and (b) the VIVOϪSalGly--
Ala system, using the program PSEQUAD
25 °C for pH-metry and ca. Ϫ3 °C for EPR (the approxi-
mate freezing point of the sample), and also in the concen-
tration of the metal ion. For this reason EPR is used only
in a quantitative way to confirm the formation of each of
the species assumed from the pH-metric speciation measure-
ments. Figures 7 and 8 depict the low-field region (high-field
in the case of VIVO) of some of the frozen solution EPR
spectra of the two systems at different pH values. Table 2
summarizes the EPR parameters obtained from simulation
of the spectra by using the computer program of Rocken-
bauer,[18] and also the most likely binding modes of the com-
plexes. For the VIVO system, CD and EPR spectra were also
recorded at pH ϭ 5.5 and cL ϭ 10 m, with a variable cVO
concentration always confirming the presence of the
MLHϪ1 species (see Figure S1, Supporting Information, see
also the footnote on the first page of this article).
Figure 8. High-field region of frozen EPR spectra (X-band) re-
corded at 77 K on the VIVOϪSalGly--Ala system at a 1:2 metal
ion to ligand ratio at different pH values, cVO ϭ 0.0045 
The pH-metric speciation curves (Figures 1 and 2) indi-
cate that complex formation starts at a pH of about 3 in
both systems with a protonated species [MLH]ϩ. In this ture 1, Scheme 2). This binding mode explains why the ob-
species, like the corresponding complexes of the dipeptide served EPR and visible spectra at pH around 3 show only
analogue H2SalGly, coordination at the terminal COOϪ minor differences from those of the aqua ions. In fact no
through (COOϪ, CONH) chelation is most likely (see struc- significant differences are expected in the EPR and visible
2116
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122

Interaction of VIVO, VVO2 and CuII with a Peptide Analogue
Table 2. Calculated EPR parameters (g|| and A||) for the complexes formed in the CuIIϪ and VIVOϪSalGly--Ala systems
FULL PAPER
Species
CuII
g||
VIV
g||
O
Equatorial binding set
est
A||
A||
A
||
]
104 [cmϪ1
]
104 [cmϪ1
MLH
ML
MLHϪ1
MLHϪ2
2.380
2.358
2.270
2.209
147
148
178
192
1.934
1.937
1.945
1.958
180
176
165
159
178
171Ϫ174
164
(COOϪ, CONH)
(OϪ, 2 ϫ CONH) or (OϪ, CONH, COOϪ) or (OϪ, CONH)
(OϪ, CONϪ, CONH, COOϪ)
159
(OϪ, 2 ϫ CONϪ, COOϪ)
spectra for such binding modes when compared with the measurements in the wide L:M range of 20Ϫ1 were also
aqua ions.[16,19]
carried out (see Exp. Sect.). Under such conditions the spec-
ies was clearly detected by the appearance of a distinct signal
(see Figure S1, Supporting Information). As mentioned
above, the fact that significant increases in the |∆ε| CD par-
ameters are only observed for the fully deprotonated com-
2Ϫ
plex [MLHϪ2
]
indicates the following sequence of amide
deprotonation: first that is closer to the phenolate OϪ then
that is closer to the COOϪ, the latter being closer to the
chiral centre. Interestingly, the opposite sequence was ob-
served for the organotin()Ϫpeptide complexes,[16] which
pointed to the C-terminal COOϪ group being the primary
anchor. With VIVO and CuII, the phenolate OϪ seems to be
the primary anchor. In the case of Gly-type tripeptides, the
N-terminal site is the primary binder for CuII.[6] At pH Ͼ
10, a significant decrease in the intensities of the EPR sig-
nals was observed in the VIVOϪSalGly--Ala system (to
compensate for this, the receiver gain and the modulation
were increased) indicating the partial hydrolysis of the metal
ion through formation of the EPR silent hydroxo-bridged
species (represented by [(VO)2(OH)5]Ϫ in the speciation
diagram).
Scheme 2
Three protons are liberated from these complexes with
stepwise pK values of 5.03, 4.92 and 6.16 for the CuII com-
plexes, and 3.55, 5.37 and 5.37 for the VIVO complexes. The
first deprotonation process, leading to the species [ML] in
both systems, is attributable to the phenolic OH group of
the metal-bound ligand. This indicates a rearrangement to
(OϪ, 2 ϫ CONH) or (OϪ, CONH, COOϪ) binding modes
(see structure 2, Scheme 2), as supported by the spectral
Compounds in the Solid State
The tripeptide analogue H2SalGly--Ala 1 was synthe-
changes. The EPR parameters (see Table 2) are character- sized by the procedure described in the Exp. Sect. Complex
istic of pure O-coordination, and agree well with those in the 2, formulated as VO(SalGly--Ala)·3.2H2O·0.2EtOH, was
corresponding complex formed with H2SalGly[8] and other prepared from solutions containing VIVO and H2SalGly--
aromatic hydroxycarboxylic acids.[20]
Ala at a pH of around 5.5, i.e. where we expect the [VOL]
The next two pK values can be assigned to the depro- stoichiometry to be predominant. Elemental analysis (see
tonation/coordination of the two amides, yielding (OϪ, Exp. Sect.) indicates that the amide is not deprotonated, and
CONϪ, CONH, COOϪ) and (OϪ, CONϪ, CONϪ, COOϪ) the CD spectrum of 2 in a KBr disk shows a band pattern
joint chelate systems (structures 3 and 4, Scheme 2) respec- and a λmax value similar to those of the VOL species shown
tively. As the simultaneous equatorial coordination of the in Figure 6b. The binding in the solid state presumably in-
second amide-O and the terminal COOϪ may be sterically volves OϪ, CONH, COOϪ and H2O, and eventually another
hindered, the tridentate (OϪ, CONϪ, COOϪ) binding mode amide O in the axial position. In frozen solution the exper-
of the ligand (see Scheme 2) is also possible in species imental A|| is around 175Ϫ176 ؋ 10Ϫ4 cmϪ1, which corre-
VOLHϪ1. Due to the coordination of the two amide nitro- sponds either to (i) (OϪ, CONH, COOϪ, H2O)equatorial coor-
est
gens, superhyperfine coupling (five lines with shf ANЌϭ15.5 dination (A ϭ 38.9 ϩ 43.7 ϩ 42.1 ϩ 45.6 ϭ 170 ؋ 10Ϫ4
||
G) can be observed in the perpendicular region of the EPR cmϪ1) or to (OϪ, CONH, 2 ϫ H2O)equatorial coordination
est
spectrum of the CuLHϪ2 species. In the case of VIVO, the (A ϭ 38.9 ϩ 43.7 ϩ 45.6 ϩ 45.6 ϭ 174 ؋ 10Ϫ4 cmϪ1).
||
EPR parameters estimated by the additivity rule are 164
Complex 3, formulated as K[Cu(SalGly--Ala)]·3.2H2O,
؋10Ϫ4 cmϪ1 and 159 ؋ 10Ϫ4 cmϪ1 which are in reasonable was also prepared from aqueous/ethanolic solutions at
agreement with the experimental hyperfine coupling con- pH ϭ 6. The elemental analysis agrees with the formulation
stants (Table 2). Since the species [VOLHϪ1]Ϫ, with one de- of CuLHϪ1 for this complex and the CD spectrum of 3 in
protonated amide NϪ, could hardly be detected in the pH a KBr disk shows a band pattern and a λmax value close to
range 5Ϫ6 at a comparable metal ion to ligand ratio, EPR those of the CuLHϪ1 species (see Figure S2, Supporting
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2117

T. Kiss, J. Costa Pessoa et al.
FULL PAPER
Table 3. IR bands [cmϪ1] and assignments of H2SalGly--Ala and its VO and Cu complexes prepared in this work (vs: very strong, s:
strong, m: medium, w: weak, vw: very weak; *: bands emerging from a broad band)
Ligand 1
Complex 2
Complex 3
Hydrogen bonded OϪH
and NϪH
ν(NϪH)
Broad band (2500Ϫ3500)
Broad band (2500Ϫ3600) Broad band (2400Ϫ3700 with
a few weak peaks emerging from it
3325 vs*
3346 m*
Ϫ
Aromatic CϪH strech
Aliphatic CϪH strech
Other amide bands
3000 w*
3020 w*
2990 w*
A few very weak bands
ഠ3000 vw*
ഠ2950 vw*
2932 w*
3072 vw* ,[23][24] 2613 vw*[a]
,
2499 vw*[a], 2411 vw*[a], 2290 vw*[a]
ν(CϭO)COOH
Amide I (CϭO)
Amide II
1730 s
1661 s, 1632 m (?)
1547 s
Ϫ
Ϫ
1605 s*
1534 s*
ഠ1600 s*
1572 s*
1400 m-s (?)
1256 s
[22]
1638 m-s* [22,p.72]
1541 s
1593 s*
1574 s* [22,p.72Ϫ73]
1370 m-s
1236 m-s
966s
CϭC Ring stretch
1593 s
νas(COOϪ)
νs(COOϪ)
ν(CϪOPhe
)
1344 m-s (?)
Ϫ
ν(VϭO)
Ϫ
CϪH Out-of-plane deformation 758 m-s
758 m-s
762 s
[a]
These are multiple combination bands due to symmetrical, asymmetrical and overtone NϪH bands.[21Ϫ24]
Information). The elemental analysis and spectra of this AlaGly and AlaHis. A single complex species of moderate
CuII complex are therefore consistent with a binding of the stability (at approx. tenfold excess of ligand ca. 50% of the
ligand through OϪ, COOϪ and CONϪ. No individual vanadate was bound to the peptide) was formed with the
ν(NϪH) band could be detected emerging from the broad above binding mode in fairly slow processes (completion of
band in the 2400Ϫ3700 cmϪ1 range.
complex titration took several hours). The chemical shift
A comparison of the FTIR spectroscopic data for com- values were as follows: δ ϭ Ϫ495 ppm for ProAla, δ ϭ
pounds 1؊3 is summarised in Table 3. The data for 2 and Ϫ512 ppm for AlaGly and δ ϭ Ϫ518 ppm for AlaHis.[27,28]
3 agree with the binding modes proposed. The assignments
are based on Refs. 21Ϫ25, and by comparison with some
VIVO and CuII complexes formed with the reduced Schiff
base ligands derived from the reaction of salicylaldehyde
and GlyGly or GlyGlyGly.[26] All compounds present broad
bands in the wavenumber range corresponding to hydrogen-
bonded OϪH and NϪH. In the case of 1 a strong ν(NϪH)
band emerges from this broad band as well as several other
weak or very weak bands (Table 3). This is characteristic of
compounds containing secondary amide groups.[23,24] The
strong ν(CϭO)COOH band at 1730 cmϪ1 in the FTIR of 1
is absent in its VIVO and CuII complexes. For 2, a narrow
band emerges at 3346 cmϪ1 and a second weaker one at
3454 cmϪ1, presumably also due to ν(NϪH) of the amide
groups.[21,23] All our efforts to characterise other solid com-
pounds from these metal-ligand systems, particularly those
of oxovanadium with a deprotonated amide, were unsuccess-
ful. Either no solid could be isolated or when precipitation
was achieved, no reasonable formulation could be found.
As phenolate proved to be a more efficient anchoring do-
nor to promote amide deprotonation with VIVO, we were
also interested in studying this interaction with vanadate().
The 51V NMR spectrum of the vanadate()ϪSalGly system
(cVϭ 0.003 ) displayed an extra peak (δ ϭ Ϫ508.5 ppm)
as well as the resonances of the V1 (δ ϭ Ϫ561.8 ppm), V2
(δ ϭ Ϫ574.8 ppm), V4 (δ ϭ Ϫ579.2 ppm) and V5 (δ ϭ
Ϫ593.7 ppm) oxoanions. At a pH of around 7 the maxi-
mum amount of the complex was around 54 % with a ten-
fold excess of ligand (see Figure 9). The spectrum was ident-
ical after 24 h. This is in agreement with the earlier litera-
ture findings (vide supra), and an (OϪ, CONϪ,COOϪ)
binding mode has been suggested for the complex
detected.[7,27Ϫ29] Interestingly, no new peaks were observed
in the spectrum of the vanadate(V)ϪSalGly--Ala system
which also had a tenfold excess of ligand (see Figure S3,
Supporting Information), whether at pH 7 or 8. In both
cases the polyoxoanions predominated in the system and
the spectra were practically unchanged after 24 h.
In the case of tripeptides, amide deprotonation seems to
occur at higher pH values when compared with dipeptides
(e.g. for the corresponding CuII systems pK(CuA) ϭ 4.40
Interactions of H2SalGly and H2SalGly-L
-Ala with VVO2
Vanadate()-promoted deprotonation of the amide NH for CuIIϪSalGly and pK(CuA) ϭ 4.92 and pK(CuAHϪ1) ϭ
and the formation of stable complexes with dipeptides 6.16 for SalGly--Ala), where the likelihood of hydrolysis
through (NH2, CONϪ, COOϪ) coordination was first de- of the copper() is higher. This sensitive equilibrium be-
tected by Rehder[7] using 51V NMR spectroscopy (δ ϭ tween the metal-peptide complex and the binary (hydro-
Ϫ505Ϫ510 ppm). Pettersson et al.[27,28] gave complete speci- xo)metal complex is finely shifted to the direction of peptide
ation descriptions of VVO2 with dipeptides such as ProAla, coordination with dipeptides, but to the direction of
2118
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122

Interaction of VIVO, VVO2 and CuII with a Peptide Analogue
FULL PAPER
detected in the pH range 7Ϫ8. This is in full accordance
with earlier literature reports concerning the interaction of
vanadateV with various dipeptides.[7,27Ϫ28] However, the
moderate affinity of vanadateV for amides, even with the
more hard and probably more efficient phenolate OϪ
anchor, is reflected in the fact that even with a tenfold ex-
cess of ligand only around 50% of the metal ions are bound
to SalGly, the rest being in the various oxoionic forms.
The considerably weaker ability of vanadium() to induce
amide deprotonation, compared with vanadium(), can
ϩ
probably be explained by electrostatic (VIVO2ϩ vs. VVO2
towards LHϪ24Ϫ) and steric arguments (there are 4 equa-
torial positions free around VIVO2ϩ, while only 3 in the
ϩ
case of VVO2 because of the cis arrangement of the two
oxo groups).
Experimental Section
Synthesis of H2SalGly-L-Ala (1): The compound was synthesized
Figure 9. 51V NMR spectra measured at pH ഠ 7 of a solution of
the VVϪSalGly system at a 1:10 metal ion to ligand ratio. (a) in
freshly prepared solution, (b) 24 h later, cV ϭ 0.003 
by two different routes: (i) through the stepwise build up of the
pseudopeptide from the individual components via the classical
liquid phase method, (ii) through the Merrifield solid-phase ap-
proach.
Route 1
oxoanion formation in the presence of tripeptides. Similar
observations were made with organotin()Ϫdi and tripep-
tide systems.[29]
Z-GlyOH: To an ice-cold solution of glycine (7.5 g, 100 mmol) in
1  NaOH (100 mL) was added benzyl chloroformate (16 mL,
100 mmol), dropwise with stirring. The pH of the solution was
maintained at 10Ϫ10.5 by addition of 1  NaOH. The mixture was
stirred for 2 h and then extracted with diethyl ether. The aqueous
layer was acidified to a pH ഠ 1 and the solid which precipitated
was filtered, washed with water and dried. Yield (17.9 g, 86%), m.p.
119Ϫ120 °C (ref.[30] m.p. 120Ϫ121 °C). NMR ([D6]DMSO): δ ϭ
3.66 (d, J ϭ 6 Hz, 2 H, CH2Gly), 5.03 (s, 2 H, CH2Ar), 7.34 (m, 5
H, Ar), 7.55 ( t, J ϭ 6 Hz, 1 H, NH), 12.25 (br. s, 1 H, COOH)
ppm.
Conclusions
The speciation and solution structural investigations dis-
cussed above strongly indicate that both the terminal
COOϪ and the phenolate OϪ can behave as anchoring do-
nors to chelate this tripeptide-analogue ligand. Coordi-
nation at a pH of around 3 starts at the terminal car-
boxylate through chelation with the neighbouring amide
carbonyl. At a pH of approximately 4.5 the binding mode
also involves the terminal phenolate and this induces depro-
tonation of the two amide-NH groups in a consecutive but
rather cooperative way in the pH range 5Ϫ6. With the phe-
nolate OϪ anchor VIVO seems to be slightly more efficient in
promoting deprotonation than CuII (the value of ΣpKamides
is 10.74 for VIVO, and 11.08 for CuII), MLHϪ2 being
the most important stoichiometry in the pH range 6Ϫ10,
corresponding to an (OϪ, 2 ϫ CONϪ, COOϪ)equatorial bind-
L-AlaOEt·HCl: Thionyl chloride (12.5 mL) was added dropwise to
stirred and cooled (Ϫ10 °C) ethanol (50 mL), followed by the ad-
dition of -Ala (4.46 g, 50 mmol). The temperature of the solution
was increased and maintained at 40 °C for 4 h. The solvent was
removed and crystallization from methanol/ethyl ether yielded
4.65 g (86%) of white crystals melting at 76 °C (ref.[31] m.p. 76 °C).
NMR ([D6]DMSO): δ ϭ 1.22 (t, J ϭ 7.2 Hz, 3 H, CH2CH3), 1.40
(d, J ϭ 7 Hz, 3 H, CH3 Ala), 4.00 (q, J ϭ 7.2 Hz, 2 H, CH2CH3),
4.17 (apparent q, J ϭ 7.2 Hz, 1 H, αCH Ala), 8.66 (br. s, 3 H,
NH3ϩ) ppm.
Z-Gly-L-AlaOEt: Z-GlyOH (4.2 g, 20 mmol) was dissolved in ethyl
ing set. While CuII is able to promote amide deprotonation acetate (50 mL) and DCC (4.3 g, 21 mmol) was added after cooling
to 0 °C. The mixture was stirred for 30 min whereupon -
AlaOEt·HCl (3.1 g, 20 mmol) and triethylamine (2.8 mL, 20 mmol)
were added. The reaction mixture was stirred at room temperature
overnight. The insoluble material was filtered off and the solution
was washed successively with 1  NaOH, 1  HCl, and saturated
NaCl. After drying with MgSO4 the solvents were removed under
reduced pressure. The oil obtained was crystallized from ethyl acet-
ate/petroleum ether. The white crystals were filtered and dried
(4.52 g, 73%), m.p. 60Ϫ64 °C; (ref.[31] m.p. 60Ϫ62 °C). NMR
(CDCl3): δ ϭ 1.28 (t, J ϭ 8 Hz, 3 H, CH2CH3), 1.40 (d, J ϭ 7 Hz,
with normal tripeptides with about the same efficiency
(ΣpKamides is 11.82[8]), VIVO cannot induce deprotonation
of the amide because of its low affinity for primary amine
donors and the metal ion precipitates as VO(OH)2. Interest-
ingly, phenolate OϪ did not prove to be such an efficient
anchoring donor with vanadate() or VVO2 as with VIVO.
Amide deprotonation could not be detected by 51V NMR
spectroscopy in the case of the tripeptide analogue
H2SalGly--Ala. With the dipeptide analogue H2SalGly, a
species with an (OϪ, NϪ, COOϪ) binding mode could be 3 H, CH3 Ala), 3.89Ϫ3.92 (m, 2 H, CH2Gly), 4.19 (q, J ϭ 7.2 Hz,
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122
www.eurjic.org  2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 2119

T. Kiss, J. Costa Pessoa et al.
FULL PAPER
2 H, CH2CH3), 4.52Ϫ4.60 (m, 1 H, αCH Ala), 5.13 (s, 2 H,
Amino acid or salicylic acid incorporation was monitored using
CH2Ar), 5.57 (br. s, 1 H, NH), 6.71 (br. s, 1 H, NH), 7.36Ϫ7.33 the ninhydrin test. The completed peptide resins were treated with
(m, 5 H, Ar) ppm.
TFA/dichloromethane/water (35:60:5, vol/vol), at 0 °C for 1 h. The
solvents were removed and the resultant free product was dissolved
in 10% aqueous acetic acid, filtered and lyophilised (798 mg). The
crude peptide derivative (100 mg) was purified by reverse-phase
HPLC using a Nucleosil 7C-18 column (16 ϫ 250 mm). The sol-
vent system used was the following: 0.1% TFA in water, 0.1% TFA,
80% acetonitrile in water, gradient: 0% Ǟ 0% B over 10 min, then
10% Ǟ 30% over 50 min, flow 3.5 mL/min, detection at 226 nm.
The appropriate fractions were combined and lyophilised. Peptide
purity was above 97% (HPLC) and the measured Mw (227) value
was in good agreement with the calculated one. NMR (D2O): δ ϭ
1.38 (d, J ϭ 7.2 Hz, 9 Hz, 3 H, CH3 Ala), 4.11 (s, 2 H, CH2 Gly),
4.34 (m, 1 H, αCH Ala), 6.99 (tap, J ϭ 7.8 Hz, 2 H, 3-H and 5-
H), 7.46 (tap, J ϭ 7.8 Hz, 1 H, 4-H) 7.50 (dd, J ϭ 1.5 Hz, 2 and
9.46 Hz, 1 H, 6-H) ppm.
HGly- -AlaOEt·HBr: Z-Gly--AlaOEt (3.08 g, 10 mmol) was
L
treated with 40% HBr/acetic acid (15 mL) at room temperature for
1 h with occasional stirring and protection from moisture. Diethyl
ether was added and after cooling for a few hours the precipitated
solid was filtered, washed with diethyl ether and dried. The chrom-
atographically pure compound was used in the next step without
further purification. Yield (2.42 g, 94%). NMR ([D6]DMSO): δ ϭ
1.20 (t, J ϭ 7.2 Hz, 3 H, CH2CH3), 1.30 (d, J ϭ 7.2 Hz, 3 H, CH3
Ala), 3.57 (d, J ϭ 3.3 Hz, 2 H, CH2 Gly), 4.10 (q, J ϭ 7.2 Hz, 2
H, CH2CH3), 4.23Ϫ4.32 (m, 1 H, αCH Ala), 8.01 (br. s, 3 H,
NH3ϩ), 8.70 (d, J ϭ 7.2 Hz, 1 H, NH Ala) ppm.
HSalGly-L-AlaOEt: Salicylic acid (0.69 g, 5 mmol) was dissolved
in ethyl acetate (20 mL) and the solution cooled to 0 °C. DCC
(1.08 g, 5.2 mmol), HGlyAla·OEt·HBr (1.28 g, 5 mmol) and tri-
ethylamine (0.9 mL, 5 mmol) were then added successively. The
reaction mixture was stirred at room temperature overnight. The
insoluble material was filtered off and the solvents removed under
reduced pressure. The residue was dissolved in a minimum amount
of acetone and stored at 0 °C for 4 h. The precipitated solid was
filtered off (dicyclohexylurea) and the solution concentrated. The
oil obtained was crystallized from ethyl acetate/petroleum ether.
(0.808 g, first crop, 55%, m.p. 104Ϫ106 °C). NMR ([D6]DMSO):
δ ϭ 1.28 (d, J ϭ 7.5 Hz, 3 H, CH3 Ala), 1.17 (t, J ϭ 7.2 Hz, 3 H,
CH2CH3), 4.07 (q, J ϭ 7.2 Hz, 2 H, CH2CH3), 3.94 (m, 2 H, CH2
Gly), 4.20 (m, 1 H, αCH Ala), 6.91 (tap, J ϭ 7.8 Hz, 2 H, 3-H and
5-H), 7.40 (tap, J ϭ 8 Hz, 1 H, 4-H) 7.86 (dd, J ϭ 1.2 Hz and
8 Hz,1 H, 6-H), 8.46,(d, J ϭ 8.4 Hz, 1 H, NH Ala), 9.04 (t, J ϭ
6 Hz, 1 H, NH Gly), 12.25 (br. s, 1 H, OH) ppm. C14H18N2O5
(294.3): calcd. C 57.14, H 6.16, N 9.52; found C 57.6, H 6.1, N 9.7.
Synthesis of the Complexes
VO(SalGly-L-Ala)·3.2H2O·0.2EtOH (2): To
a
solution of
H2SalGly--Ala (1 mmol) and sodium acetateϪtrihydrate (2 mmol)
in ethanol/water (2.5:5 mL) was slowly added a solution of
VOSO4·5H2O (1 mmol) in water (1 mL). The mixture was divided
in two equal parts (A and B). The pH of solution A was adjusted
to 5.5. A blue-green solid precipitated which was collected by fil-
tration and washed with ethanol/water (1:2), diethyl ether
and dried. Yield: 40% (80 mg). C12H12N2O6V·3.2H2O·
0.2EtOH: calcd. C 37.42, H 4.86, N 7.04; found C 37.4, H 5.3,
N 7.00.
K[Cu(SalGly-L-Ala)]·3.2H2O (3): To a solution of H2SalGly--
Ala (0.5 mmol) and KOH (0.5 mmol) in water (10 mL), was slowly
added Cu(CH3COO)2·H2O (0.5 mmol) in ethanol (5 mL). The pH
was adjusted to 6 with 2  KOH and kept at ഠ 45 °C for 1 h. A
green precipitate was collected by filtration, washed with ethanol/
water (1:2), ethanol and diethyl ether and dried. Yield: 42% (90
mg). C12H11CuKN2O5·3.2H2O: calcd. C 34.03, H 4.14, N 6.61;
found C 33.7, H 3.7, N 6.2.
H2Sal-L-GlyAla (1): HSalGly--AlaOEt (0.734 g, 2.5 mmol) was
dissolved in methanol (5 mL) and 1  NaOH (5.5 mL, 5.5 mmol)
was added. The mixture was stirred at room temperature for 4 h
and 1  HCl (2.5 mL, 2.5 mmol) was then added. The methanol
was removed under reduced pressure and the solution thus ob-
tained cooled in an ice bath and acidified with 1  HCl (3 mL)
with vigorous stirring. The white solid precipitated was filtered off,
washed with water and dried (0.581 g, 87%), m.p. 197Ϫ199 °C.
NMR ([D6]DMSO): δ ϭ 1.28 (d, J ϭ 7.5 Hz, 3 H, CH3 Ala), 3.94
(m, 2 H, CH2 Gly), 4.20 (m, 1 H, αCH Ala), 6.91 (tap, J ϭ 7.8 Hz,
2 H, 3-H and 5-H), 7.39 (tap, J ϭ 7.8 Hz, 1 H, 4-H) 7.86 (dd, J ϭ
1.2 Hz and 8 Hz, 1 H, 6-H), 8.34 (d, J ϭ 7.5 Hz, 1 H, NH Ala),
9.03 (t, J ϭ 6 Hz, 1 H, NH Gly), 12.20(s, 1 H, OH), 12.40 (br. s, 1
H, OH) ppm. C12H14N2O5 (266.3): calcd. C 54.13, H 5.30, N 10.52;
found C 54.17, H 5.41, N 10.42.
Physical and Spectroscopic Studies
Melting points were recorded by using a Gallenkamp melting
point apparatus and are uncorrected. NMR spectroscopic data
were recorded either with a Varian Unity Plus 300 Spectrometer
(for the several steps of the synthesis of 1) in the solvent indicated,
or with a Varian Unity-500 Spectrometer operating at frequencies
of 131.404 MHz and 499.824 MHz, respectively, using a 5-mm
broad-band probe at 25.0Ϯ0.5 °C. 51V NMR chemical shifts were
referenced to an external VOCl3 solution at δ ϭ 0 ppm. The 51V
NMR acquisition parameters were as follows: 33 kHz spectral
width, 30-µs pulse width, 1 s acquisition time and 10 Hz line broad-
ening. IR spectra were recorded with a BioRad FTS-3000 MX
FTIR spectrometer. CD spectra were recorded with a JASCO-720
spectropolarimeter with a red-sensitive photomultiplier (EXWL-
308). Visible spectra were recorded with a Hitachi U-2000 spectro-
photometer. X-band (9.44 GHz) EPR spectra were usually re-
corded at 77 K (on glasses made by freezing solutions in liquid
nitrogen) with a Bruker ESP 300E X-band spectrometer. Mass
spectra were recorded with a Finnigan TSQ-7000 tandem quadru-
pole mass spectrometer equipped with an electrospray ion source.
The HPLC apparatus was made by Knauer (Berlin, Germany).
Route 2
The dipeptide chain was elongated on an HMP resin (1.1 mmol/g).
In the first step, Fmoc-AlaϪOH (5.47, 17.6 mmol) was coupled to
the resin (4 g, 4.4 mmol) in the presence of HOBt (2.38 g,
17.6 mmol), DCC (3.63 g, 17.6 mmol) and dimethylaminopyridine
(1.07 g, 8.8 mmol). After removing the Fmoc protecting group by
treatment with piperidine, the resultant free amino function was
acylated with FmocϪGlyϪOH (2.97 g, 10 mmol). After com-
pletion of the dipeptide unit, the N-terminal Fmoc-protecting
group was removed as previously described and the resultant pri-
mary amino group was acylated with salicylic acid (2.76 g,
20 mmol). The phenolic hydroxy group of the salicylic acid residue
was unprotected. All the couplings were performed with N,NЈ-di-
cyclohexylcarbodiimide in the presence of 1-hydroxybenzotriazole.
All samples for the pH-potentiometric and spectrophotometric
measurements were prepared under inert conditions (under high
purity nitrogen or argon). For potentiometric measurements the
2120
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122

Interaction of VIVO, VVO2 and CuII with a Peptide Analogue
FULL PAPER
ionic strength was adjusted to 0.2  KCl and the temperature main-
tained at 25.0Ϯ0.1 °C. For the spectral measurements the tempera-
ture was maintained at 25.0Ϯ0.3 °C with circulating water.
Acknowledgments
The authors are grateful to Profs C. Geraldes and M. Castro (Uni-
versity of Coimbra, Portugal) for help in recording the NMR spec-
tra. This work was carried out in as part of a COST D21 project,
and was financially supported by the National Research Fund
(OTKA T31896/2000), the Hungarian Academy of Sciences, the
pH Measurements for Spectroscopic Studies: For preparation of the
solutions and pH calibrations a special double-walled glass vessel
was used with entrances for the combined electrode (Radiometer
’’Red Rod’’ pHC2015Ϫ8), thermometer, nitrogen and reagents (e.g.
base). A computerized system developed locally (IBM-PCXT 286
computer) was used to control the titration conditions for pH cali-
brations. The E.M.F measurements were made with a Denver
Model 15 pH meter.
Fundo Europeu para o Desenvolvimento Regional, Fundaca˜o para
¸
ˆ
a Ciencia e Tecnologia, the POCTI Programme (project POCTI/
35368/QUI/2000), and the Hungarian-Portuguese Intergovern-
mental S & T Co-operation Programme for 2000Ϫ2001.
pH-Potentiometric Titrations: Stock solutions of VIVO and vana-
date() for the pH-metric titrations were prepared and standardised
as described earlier.[32] The purity of the ligand was checked and
the exact concentration of its solutions was determined by the ap-
propriate Gran plot.[33] The stability constants of the proton and
metal complexes of the ligand were determined by pH-metric ti-
tration of 10.0 cm3 samples. The ligand concentrations were in the
range of 0.001 to 0.002  and the metal ion to ligand molar ratio
varied from 1:1 to 1:3. Titrations were performed from pH 2.0 up
to pH 11 with KOH solution of known concentration (ca. 0.2 )
under purified argon unless very extensive hydrolysis or slow equili-
bration was observed. The reproducibility of titration points in-
cluded in the evaluation was within 0.005 pH units in the whole
pH range. The pH was measured with an Orion-710A precision
digital pH meter equipped with an Orion Ross 8103BN type com-
bined glass electrode, calibrated for hydrogen ion concentration as
described earlier.[34] The ionic product of water, pKw is 13.76. The
concentration stability constants βpqr ϭ [MpLqHr] / [M]p[L]q[H]r
were calculated with the aid of the PSEQUAD computer pro-
gram.[35] The formation of the hydroxo complexes of VIVO was
taken into account. The following species were assumed:
[1]
D. C. Crans, R. L. Bunch, L. A. Theisen, J. Am. Chem. Soc.
1989, 111, 7597Ϫ7607.
[2]
G. Rakhit, B. Sarkar, J. Inorg. Biochem. 1981, 15, 233Ϫ241.
[3]
H. Sakurai, S. Shimomura, F. Fukazawa, K. Ishizu, Biochem.
Biophys. Res. Commun. 1980, 90, 293Ϫ298.
[4]
´
J. Costa Pessoa, I. Tomaz, T. Kiss, E. Kiss, P. Buglyo, J. Biol.
Inorg. Chem. 2002, 7, 225Ϫ240; T. Armas, A. Mederos, P. Gili,
S. Dominguez, R. Hernandez-Molina, P. Lorenzo, E. J. Baran,
M. L. Araujo, F. Brito, Polyhedron 2001, 20, 799Ϫ804.
[5]
[6]
´
J. Costa Pessoa, I. Tomaz, T. Kiss, P. Buglyo, J. Inorg. Biochem.
2001, 84, 259Ϫ270.
´ ´ ´
I. Sovago, in Biocoordination Chemistry, (Ed.: K. Burger), Ellis
Horwood, New York 1990, p. 135Ϫ184.
[7]
[8]
D. Rehder, Inorg. Chem. 1988, 27, 4312Ϫ4316.
´
´
T. Kiss, K. Petrohan, P. Buglyo, D. Sanna, G. Micera, J. Costa
Pessoa, C. Madeira, Inorg. Chem. 1998, 37, 6389Ϫ6391.
J. Costa Pessoa, S. M. Luz, R. Duarte, J. J. G. Moura, R. D.
Gillard, Polyhedron 1993, 23, 2857Ϫ2867; J. Costa Pessoa, S.
M. Luz, R. D. Gillard, J. Chem. Soc., Dalton Trans. 1997,
569Ϫ576.
[9]
[10]
E. B. Gonzalez, N. N. Daeid, K. B. Nolan, E. Farkas, Poly-
hedron 1994, 13, 1495Ϫ1499.
R. B. Martin, Met. Ions Biol. Syst. 1974, 1, 129Ϫ157.
N. D. Chasteen, in Biological Magnetic Resonance, (Eds.: J.
Lawrence, L. J. Berliner, J. Reuben), Plenum Press, New York
1981, vol. 3, p. 53Ϫ119.
C. R. Cornman, E. P. Zovinka, Y. D. Boyajian, K. M. Geiser-
Bush, P. D. Boyle, P. Singh, Inorg. Chem. 1995, 34, 4213Ϫ4219.
A. J. Tasiopoulos, A. N. Troganis, A. Evangelou, C. P. Rapto-
poulou, A. Terzis, Y. G. Deligiannakis, T. A. Kabanos, Chem.
Eur. J. 1999, 5, 910Ϫ921.
B. J. Hamstra, A. L. P. Houseman, G. J. Colpas, J. W. Kampf,
R. LoBrutto, W. D. Frasch, V. L. Pecoraro, Inorg. Chem. 1997,
36, 4866Ϫ4874.
[VOIV(OH)]ϩ (log β10Ϫ1 ϭ Ϫ5.94), [(VIVO)2(OH)2]2ϩ (log β20Ϫ2
ϭ
Ϫ6.95). Stability constants were calculated from the data of Henry
[11]
[12]
et al.[36] and corrected for the different ionic strengths by using the
Davies equation. [(VIVO)2(OH)5]Ϫ (log β20Ϫ5
[VIVO(OH)3]Ϫ (log β10Ϫ3 ϭ Ϫ18.0).[37]
ϭ Ϫ22.5) and
[13]
[14]
Spectroscopic Measurements: Unless otherwise stated, by visible
(Vis) and circular dichroism (CD) spectra we mean a representation
of εm or ∆εm values vs. λ [εm ϭ absorption/(bcM) and ∆εm
ϭ
differential absorption/(bcM) where b ϭ optical path and cMϭtotal
VIVO or CuII concentration]. The spectral range covered is nor-
mally 400Ϫ900 (Vis) and 400Ϫ1000 nm (CD). All measurements
and operations of the spectropolarimeter were computer con-
trolled. For the CD spectra a Fast Fourier-Transform noise-re-
duction routine (JASCO) was used without affecting peak shapes.
For the VIVO system, the EPR, Vis and CD spectra were recorded
by varying the pH with an approximately fixed total vanadium and
ligand concentration, at L/M ratios of 1.76 (cVO ഠ 0.003 ) and
2.00 (cVO ഠ 0.0045 ), and at pH ϭ 5.5 and cligand ഠ 0.010 m .
[15]
[16]
[17]
[18]
[19]
[20]
J. Costa Pessoa, S. M. Luz, R. D. Gillard, J. Chem. Soc., Dalton
Trans. 1997, 569Ϫ576.
´
´ ´
P. Buglyo, E. Kiss, I. Fabian, T. Kiss, D. Sanna, E. Garriba,
G. Micera, Inorg. Chim. Acta 2000, 306, 174Ϫ183.
A. Rockenbauer, L. Korecz, Appl. Magn. Resonance 1996, 10,
29Ϫ43.
J. Costa Pessoa, L. F. Vilas Boas, R. D. Gillard, R. J. Lanca-
shire, Polyhedron 1988, 7, 1245Ϫ1262.
L/M ratios were varied in the range of 20Ϫ1 by addition of VIV
O
stock solution. For the CuII system, the EPR, Vis and CD spectra
M. Jezowska-Bojczuk, H. Kozlowski, A. Zubor, T. Kiss, M.
Branca, G. Micera, A. Dessi, J. Chem. Soc., Dalton Trans.
1990, 2903Ϫ2907; G. Micera, A. Dessi, D. Sanna, T. Kiss, P.
were recorded varying the pH with cCu ഠ 0.006  and L/M ϭ 2.
CD spectra of Solid Complexes: Samples of compounds 2 and 3
were prepared as described previously,[38] dispersed in KBr disks,
and placed between two microscope slides. One or two such paired
microscope slides were placed in the sample compartment. Each
final spectrum is the average of 4 to 6 individual spectra, recorded
as described above.
Supporting Information: Additional EPR, CD and 51V NMR spec-
tra for the VIVOϪSalGly--Ala and VVO2ϪSalGly--Ala systems
are available (see also the footnote on the first page of this article).
´
Buglyo, Gazz. Chim. It. 1993, 123, 573Ϫ577.
[21]
[22]
D. H. Williams, I. Fleming, Spectroscopic Methods in Organic
Chemistry, 5th ed., McGraw-Hill, London 1995, p. 42, 49.
K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, 5th ed., John Wiley & Sons, New
York, 1997, p. 62Ϫ73, 271.
[23]
[24]
E. Silverstein, G. C. Bassler, T. C. Morril, Spectrometric Identi-
fication of Organic Compounds, 5th ed., John Wiley & Sons,
New York 1991, p. 122Ϫ125.
G. Socrates, Infrared and Raman Characteristic Group Fre-
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2121

T. Kiss, J. Costa Pessoa et al.
FULL PAPER
[34]
[35]
quencies. Tables and Charts, 3rd ed., John Wiley & Sons,
Chichester 2001, p. 51, 107Ϫ113, 143Ϫ145.
A. J. Tasiopoulos, E. J. Tolis, J. M. Tsangaris, A. Evangelou, J.
H. M. Irving, M. G. Miles, L. D. Pettit, Anal. Chim. Acta 1967,
38, 575Ϫ481.
[25]
´ ´
´
L. Zekany, I. Nagypal, PSEQUAD: A Comprehensive Program
for the Evaluation of Potentiometric and/or Spectrophotometric
Equilibrium Data Using Analytical Derivatives’’ in Compu-
tational Methods for the Determination of Formation Constants,
(Ed.: D. J. Leggett), Plenum Press, New York, 1991.
R. P. Henry, P. C. H. Mitchell, J. E. Prue, J. Chem. Soc., Dalton
Trans. 1973, 1156Ϫ1159.
A. Komura, M. Hayashi, H. Imanaga, Bull. Chem. Soc. Jpn.
1977, 50, 2927Ϫ2931; L. F. Vilas Boas, J. Costa Pessoa, in
Comprehensive Coordination Chemistry (Eds.: G. Wilkinson, R.
D. Gillard, J. A. McCleverty), Pergamon Press, Oxford 1987,
vol. 3, p. 453Ϫ583.
D. Woollins, A. M. Slawin, J. Costa Pessoa, I. Correia, T. A.
Kabanos, J. Biol. Inorg. Chem. 2002, 7, 363Ϫ374.
T. Jakusch, I. Correia, J. Costa Pessoa, T. Kiss, J. Inorg. Bi-
[26]
ochem. 2001, 86, 278.
K. Elvingson, M. Fritzsche, D. Rehder, L. Pettersson, Acta
[27]
[36]
[37]
Chem. Scand. 1994, 48, 878Ϫ885.
M. Fritzsche, K. Elvingson, D. Rehder, L. Pettersson, Acta
[28]
Chem. Scand. 1997, 51, 483Ϫ491.
[29]
T. Gajda, personal communication.
[30]
George R. Pettit, Synthetic Peptides, Elsevier Scientific, Am-
sterdam 1980, vol. 5.
George R. Pettit, Synthetic Peptides, Academic Press, New
[31]
[38]
I. Cavaco, J. Costa Pessoa, D. Costa, M. T. L. Duarte, P. M.
Matias, R. D. Gillard, J. Chem. Soc., Dalton Trans. 1994,
149Ϫ157.
York 1975, vol. 3.
[32]
´
´ ´
I. Nagypal, I. Fabian, Inorg. Chim. Acta 1982, 61, 109Ϫ113.
[33]
G. Gran, Acta Chem. Scand. 1950, 4, 559Ϫ575.
Received September 16, 2002
2122
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2113Ϫ2122