New coordination modes of L-ascorbic acid and dehydro-L-ascorbic acid as dianionic chelating ligand for platinum

Article abstract of DOI:10.1002/ejic.200700808

A variety of coordination modes of L-ascorbic acid as an anionic bidentate ligand has been exploited to prepare platinum(II) complexes 1-7 that contain phosphanes or R,R-dach (1R,2R-diaminocyclohexane) as neutral ligands in which O2, O3, O5, O6 and C2 act as anionic donating functionalities. An alternative synthetic route to known O2,O3 complexes is proposed, and their solubility in water has been enhanced by introducing PTA (1,3,5-triaza-7-phosphaadamantane) as a neutral ligand. A new coordination mode of ascorbic acid (O2 and O3 protected) as an O5,O6-diolate chelating ligand has been characterised in solution by NMR spectroscopy and in the solid state by X-ray crystallography. The first example of a platinum complex that contains dehydroascorbic acid, 7, has also been prepared and its X-ray crystal structure has been determined. The antiproliferative activity in vitro of complexes 1-7 has been tested, and the best values were obtained for the DHA complex 7, which was found to be more active than cisplatin on both a cisplatin-sensitive and a cisplatin-resistant cell line. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejic.200700808

FULL PAPER
DOI: 10.1002/ejic.200700808
New Coordination Modes of
L
-Ascorbic Acid and Dehydro-
L
-ascorbic Acid as
Dianionic Chelating Ligand for Platinum
Paola Bergamini,*[a] Elena Marchesi,[a] Valerio Bertolasi,[a] Marco Fogagnolo,[a]
Luca Scarpantonio,[a] Stefano Manfredini,[b] Silvia Vertuani,[b] and Alessandro Canella[c]
Keywords: Ascorbic acid / Dehydroascorbic acid / Platinum / Antitumour agents / Drug delivery / Drug design
A variety of coordination modes of
L
-ascorbic acid as an an-
the solid state by X-ray crystallography. The first example of
a platinum complex that contains dehydroascorbic acid, 7,
has also been prepared and its X-ray crystal structure has
been determined. The antiproliferative activity in vitro of
complexes 1–7 has been tested, and the best values were ob-
tained for the DHA complex 7, which was found to be more
active than cisplatin on both a cisplatin-sensitive and a cis-
platin-resistant cell line.
ionic bidentate ligand has been exploited to prepare plati-
num(II) complexes 1–7 that contain phosphanes or R,R-dach
(1R,2R-diaminocyclohexane) as neutral ligands in which O2,
O3, O5, O6 and C2 act as anionic donating functionalities.
An alternative synthetic route to known O2,O3 complexes is
proposed, and their solubility in water has been enhanced
by introducing PTA (1,3,5-triaza-7-phosphaadamantane) as a
neutral ligand. A new coordination mode of ascorbic acid (O2
and O3 protected) as an O5,O6-diolate chelating ligand has
been characterised in solution by NMR spectroscopy and in
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
the action of ascorbic acid as a cellular reducing agent of
anticancer PtIV complexes.[5]
Introduction
The lack of efficacy of chemotherapies for brain tumours
arises mainly because of the difficulty in crossing the blood-
brain barrier. This is a cause of major frustration, but it
also represents a great challenge in the search of new ap-
proaches for delivering drugs inside the brain.[1]
Among the strategies for the delivery of drugs to the cen-
tral nervous system (CNS), there is now a wide interest in
the so-called “trojan horse tactic”, which is performed by
attaching an active drug molecule to a vector that accesses
a specific transport mechanism.
An example of this strategy was reported by Manfredini
et al. who illustrated the effectiveness of the anticonvulsant
drug nipecotic acid in inhibiting inducted convulsions when
conjugated to ascorbic acid, by exploiting ascorbate specific
transporters.[2]
The interaction of transition metals with -ascorbic acid,
a natural, low-cost molecule, has been extensively studied,[3]
mainly because ascorbic acid is widely used as a nontoxic
reducing agent. A few coordination modes to platinum(II)
have been described, and the potential anticancer activity
of the resulting complexes has been mentioned,[4] as well as
Nevertheless, the above-mentioned discovery of the abil-
ity of ascorbic acid to act as a carrier for drugs across the
brain-blood barrier made us reconsider its binding ability
to metal ions with known medicinal activity with the per-
spective of developing metal-containing drugs that are able
to take advantage of ascorbic acid specific transporters
(carrier-mediated transport). Moreover, also the oxidised
form, dehydroascorbic acid (DHA), can be taken into con-
sideration as a carrier; in fact, it was reported that DHA
crosses the blood-brain barrier through the glucose trans-
porters GLUT1 and GLUT2.[6]
The application of this approach to platinum-based
drugs would be of paramount importance because it is
known that cisplatin, one of the most successful anticancer
drugs in clinical use, is unable to pass the brain-blood bar-
rier, and it is therefore inactive against brain tumours;[7] for
this reason, we present in this paper a complete picture of
the coordination chemistry of ascorbic acid to platinum and
the first example of a DHA–metal complex.
[a] Dipartimento di Chimica e Centro di Strutturistica Diffratto-
metrica dell’Università di Ferrara
Results and Discussion
Via L. Borsari, 46, 44100 Ferrara, Italy
[b] Dipartimento di Scienze Farmaceutiche dell’Università di Fer-
rara,
Following the most successful pattern of efficacious plati-
num anticancer drugs, i.e. two neutral and two anionic li-
gands in a cis disposition, we have considered several pos-
sibilities of introducing -ascorbic acid and its oxidised
form, DHA, as a dianionic bidentate ligand on platinum.
Via Fossato di Mortara, 17–19, 44100 Ferrara, Italy
[c] Dipartimento di Biochimica e Biologia Molecolare dell’Univer-
sità di Ferrara,
Via L. Borsari, 46, 44100 Ferrara, Italy
Eur. J. Inorg. Chem. 2008, 529–537
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
529

P. Bergamini et al.
FULL PAPER
All the reported complexes were obtained from Pt pre-
cursors bearing amines or phosphanes as neutral ancillary
ligands, in addition to easily replaceable anionic ligands.
We describe here three coordination modes of -ascorbic
acid as a dianionic chelating ligand: (i) through deproton-
ated atoms O2 and O3, (ii) through deprotonated atoms O5
and C2, (iii) through deprotonated atoms O5 and O6 (this
requires the protection of the atoms O2 and O3).
O2,O3 Coordination – A New Route to Known Complexes
Scheme 1.
The coordination of an alcoholic oxygen atom (hard do-
nor) to platinum (soft acid) is thermodynamically disfav-
oured and occurs exclusively when it is supported by an
additional driving force like the precipitation of an insolu-
ble side product or the development of a volatile side prod-
uct.
When ascorbic acid reacts with a platinum precursor un-
der such conditions, the reacting groups are selectively O2–
H and O3–H, because the ene–diolic system makes them
more nucleophilic than other groups in the molecule. The
formation of O2,O3-coordinated platinum complexes has
already been obtained through the reaction of phosphane–
nitrato precursors[4] or through the reaction of a dihydroxy
Me3PPt complex[8] and ascorbic acid. We obtained O2,O3-
coordinated ascorbate–platinum complexes (Scheme 1),
with better yields and with an easier work up procedure, by
treating ascorbic acid in dichloromethane with Pt–carbon-
ate precursors, which are known to react with vicinal diols,
as described by Andrews.[9]
The complex [Pt(O2,O3-asc)(dppe)] [2, dppe = 1,2-bis(di-
phenylphosphanyl)ethane], obtained from [Pt(CO3)(dppe)],
1
shows two doublets at δ = 32.8 and 29.2 ppm with JPt,P
=
3364 and 3664 Hz, respectively, in its 31P NMR spectrum.
Further, in this case, the phosphorus atom trans to O3 is
1
assigned by the value of the coupling constant JPt,P, which
is 300 Hz larger than the value of the other coupling con-
1
stant. The H NMR spectrum in CDCl3 is similar to that
of 1, with the exception of the presence of a multiplet at δ
= 2.40 ppm, which is assigned to the CH2 protons of dppe.
In the IR spectrum, a broad band between 3500 cm–1 and
3100 cm–1 confirms the presence of OH at positions 5 and
6.
Complexes 1 and 2 are stable in the solid state and in
organic solvent solutions. Isomerisation to the C2,O5 coor-
dination has never been observed.
The complex [Pt(O2,O3-asc)(PPh3)2] (1) was recovered
after solvent evaporation as the only product, and it was
characterised by NMR spectroscopy. The 31P NMR spec-
trum in CDCl3 shows two doublets with satellites assigned
Complex 3 is another example of the O2,O3 coordina-
tion mode that we obtained, which contains the water-solu-
ble phosphane PTA (triazaphosphaadamantane). There is
an increasing interest in coordination compounds of this
phosphane and in their use in medicinal chemistry because
of their favourable properties like hydrophilicity.[10]
to the two nonequivalent phosphanes that are trans to the
2
two oxygen atoms at δ = 9.7 and 6.9 ppm with JP,P
=
1
20 Hz. The coupling constants JPt,P of 3490 Hz and
3776 Hz, respectively, are typical of a PPh3 group trans to
an oxygen atom.[4] By comparison with the data of complex
5 (see below), we assign the larger Pt–P coupling constant
to a P atom trans to O3. The IR spectrum shows a broad
signal between 3500 and 3100 cm–1, which can be assigned
to the hydroxy groups O5–H and O6–H. The 1H NMR
spectrum in CDCl3 shows a doublet at δ = 4.30 ppm for the
proton 4-H, while the multiplets at δ = 3.60 and 3.30 ppm
Complex 3 was prepared from the known precursor cis-
are assigned to the proton 5-H and to the two dia- [PtCl2(PTA)2],[11] subsequently converted in the nitrato spe-
stereotopic protons 6-H. The chemical shifts of these sig- cies by treatment with silver nitrate in water. Sodium as-
nals are very similar to the corresponding signals in free corbate was then added directly to the solution, and com-
ascorbic acid, because they belong to protons that are lo- plex 3 was promptly formed, as confirmed by 31P NMR
cated far away from the coordination site.
analysis (Scheme 2).
530
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 529–537

New Coordination Modes of Dianionic Chelating Ligand for Platinum
Table 1. H NMR spectroscopic data in D2O.
1
4-H
5-H
6-H
Complex 4
4.27 ppm (s)
4.08 ppm (t)
JH,H = 6.5 Hz
3.42 ppm (d)
3JH,H = 6.5 Hz
3
3JH,H = 1.5 Hz 3JH,H = 6.4 Hz; JH,H = 6.4 Hz
Ascorbic acid
4.90 ppm (d)
3.78 ppm (td)
3.72 ppm (d)
3
3JH,H = 1.5 Hz 3JH,H = 6.4 Hz; JH,H = 6.4 Hz
3JH,H = 1.5 Hz
Sodium ascorbate
4.41 ppm (d)
3.90 ppm (td)
3.73 ppm (d)
3
3JH,H = 1.6 Hz 3JH,H = 7.5 Hz; JH,H = 7.5 Hz
3JH,H = 1.6 Hz
Scheme 2.
The infrared spectrum of 4 presents a broad band be-
tween 3550 and 2860 cm–1 assigned to the O2–H, O6–H
and NH2 stretching modes, a strong signal at 1716 cm–1 as-
signed to the C=O stretching mode and two signals at 1630
and 1590 cm–1 for the C2=C3 stretching mode.
In the 31P NMR spectrum of 3 in D2O, two doublets
1
with satellites at δ = –58.34 and –58.71 ppm with JPt,P
=
3302 and 3343 Hz, respectively, are observed. The 1H NMR
spectrum consists of a doublet at 4.42 for the proton 4-H
3
with JH,H = 2 Hz, while the protons 5-H and 6-H give rise
3
O5,O6 Coordination of Protected Ascorbic Acid – A New
Coordination Mode
to a triplet of doublets at δ = 3.93 ppm with JH,H = 7 Hz
and 4JH,H = 2 Hz and a doublet of doublets at δ = 3.65 ppm
3
4
with JH,H = 10.5 Hz and JH,H = 2 Hz, respectively.
The aqueous solution of 3 does not show any sign of
decomposition in 10 d, but it is reconverted to the
dichlorido complex after 48 h in the presence of an excess
(4 equiv.) of added chloride.
From the information on the interaction between -
ascorbic acid and its specific transporter through the blood-
brain barrier, the integrity of the ene–diolic side seems to
be crucial.[14] It is therefore of interest to drive the coordi-
nation of Pt elsewhere, for example, towards the diolic O5
and O6 atoms.
The coordination of the oxygen atoms of O5–H and O6–
H of ascorbic acid to a metal has never been reported be-
fore. Because of the more acidic character of O2–H and
O3–H, it is clear that protection of the groups is required
in order to avoid their coordination with high specificity as
shown above.
All the steps for the protection of O2–H and O3–H with
benzyl groups have been reported previously[15] and are de-
scribed in Scheme 3.
C2,O5 Coordination – Known Complex
It has been reported that amine–Pt precursors such as
[Pt(NH3)2(OH2)2]2+ or [Pt(R,R-dach)(OH2)2]2+ react in
water with ascorbate giving C2,O5 coordination com-
plexes.[4,12] This coordination mode seems to be precluded
with phosphanic precursors, probably because of steric in-
teractions.[4] On the contrary, any attempts to obtain the
O2,O3 coordination with amine–Pt precursors invariably
gave C2,O5 complexes. The complex [Pt(C2,O5-asc){trans-
(R,R-dach)}] (4) was prepared as reported by Hollis.[13]
The reaction between O2Bn,O3Bn ascorbic acid and the
carbonate precursor [Pt(CO3)(PPh3)2] gave complex 5. Also
in this case, the 31P NMR spectrum in acetone shows two
close signals with satellites at δ = 13.7 ppm (1JPt,P
=
3512 Hz) and 12.8 ppm (1JPt,P = 3372 Hz) coupled each
2
other with JP,P = 21.4 Hz. The NMR spectroscopic obser-
vation over a long period indicates that 5 is unchanged for
20 d in acetone, while it is rapidly converted to the dichlo-
ride complex cis-[PtCl2(PPh3)2] in the presence of added
Cl–. Comparison of the 31P NMR spectroscopic data of 5
(O5,O6 complex) and 2 (O2,O3 complex) shows that, al-
though the patterns appear to be very similar, it is possible
to distinguish between the two coordination modes by care-
Although the solubility of 4 in water is less then 1 mg/ ful analysis of the 31P NMR spectroscopic data.
mL, it is possible to acquire the 1H NMR spectrum in D2O,
The analogue dppe complex 6 was obtained in the same
which is diagnostic for the coordination mode by careful way and characterised by NMR spectroscopy. The 31P
comparison of the signals of protons 4-H, 5-H and 6-H in NMR spectrum in [D6]acetone shows two doublets with
1
free, salified and coordinated ascorbic acid, as reported in satellites at δ = 29.7 and 29.4 ppm with JPt,P = 3280 and
1
2
Table 1. The H NMR spectrum in D2O does not reveal 3229 Hz, respectively, and JP,P = 9.0 Hz. The 1H NMR
any sign of decomposition of complex 4 after 10 d, even spectrum of 6 in [D6]acetone shows four signals at 5.39,
when NaCl is added.
5.22, 4.77 and 4.72 ppm assigned to the CH2Ph protons, a
Eur. J. Inorg. Chem. 2008, 529–537
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
531

P. Bergamini et al.
FULL PAPER
Scheme 3.
Table 2. Selected bond lengths [Å] and angles [°] for complex 6.
Bond lengths
Pt1–P1
Pt1–O5
C5–O5
C5–C6
C4–C3
C1–O4
C1–C2
C2–O2
2.222(1)
2.028(3)
1.410(5)
1.533(7)
1.484(7)
1.366(8)
1.436(10)
1.387(8)
Pt1–P2
Pt1–O6
C6–O6
C5–C4
C4–O4
C1–O1
C2–C3
C3–O3
2.220(1)
2.028(3)
1.412(6)
1.524(7)
1.421(7)
1.221(10)
1.330(9)
1.335(8)
multiplet at δ = 4.44, 4.14, 3.89 and 3.66 ppm (for 4-H,
5-H, 6-H and 6Ј-H, respectively) together with the signal
assigned to the CH2 of dppe at δ = 2.80 ppm.
The recrystallisation of crude 6 from acetone and diethyl
ether gave crystals suitable for X-ray structure determi-
nation (Figure 1). This is the first reported X-ray crystal
structure for an O5,O6-coordinated ascorbic acid deriva-
tive. In Table 2 a selection of bond lengths and angles for 6
is reported.
Angles
P1–Pt1–P2
P1–Pt1–O6
P2–Pt1–O6
Pt1–O5–C5
O5–C5–C6
85.8(1)
94.3(1)
179.6(1)
108.9(2)
108.0(3)
P1–Pt1–O5
P2–Pt1–O5
O5–Pt1–O6
Pt1–O6–C6
O6–C6–C5
177.5(1)
96.6(1)
83.3(1)
109.0(3)
108.7(4)
C22–P2 exhibit twisted (3T4) and envelope (E4) conforma-
tions, respectively.
In order to exploit the ene–diolic system of ascorbic acid
with respect to the interaction with its specific transporter,
it is necessary to deprotect the O2 and O3 atoms in com-
plexes 5 and 6 after coordination to platinum.
Unfortunately, every attempt to remove the benzylic
groups with known procedures induced decomposition of
the complex. In particular, the reductive route with H2/Pd
produced a mixture of several Pt-containing species, proba-
bly Pt-hydride. The alternative debenzylation with BCl3 at
–78 °C[17] did not also work, because the oxygen donors are
replaced by chloride ions in the complex as a result of the
HCl formed in situ, and only cis-[PtCl2(PPh3)2] was iden-
tified in the reaction mixture.
Figure 1. ORTEP[16] view of complex 6. The thermal ellipsoids are
at 30% level of probability.
Coordination of Dehydro-
L-ascorbic Acid
The structure of complex 6 shows that the coordination
The commercially available DHA is a crystalline solid
around platinum is a slightly distorted square planar – the and is available in the monomeric form (a) or the dimeric
Pt1 atom is displaced from the mean plane passing through form (c), whereas in solution, there is an equilibrium be-
the basal atoms P1, P2, O5 and O6 by 0.0024(2) Å. The tween the monomeric diketonic form (a), the monomeric
C1–C2–C3–C4–O4 ring is essentially planar and forms an hemiacetalic form (b) and the dimeric form (c). The ratio
angle of 85.7(2)° with the coordination plane. The two co- between the three forms depends on the solvent and on the
ordination rings Pt1–O5–C5–C6–O6 and Pt1–P1–C21– preparation procedure (Scheme 4).[18]
532
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 529–537

New Coordination Modes of Dianionic Chelating Ligand for Platinum
Scheme 4.
The reaction of DHA with [Pt(CO3)(PPh3)2] was per-
formed because we wanted to check whether the major re-
activity of O6–H with Pt would have shifted the equilibrium
towards the monomeric form (a) to give the complex in
which O5 and O6 are bonded to Pt. Unexpectedly, the reac-
tion gave a single product 7, which corresponds to the coor-
dination of form (b) in addition to one molecule of water.
The 31P NMR spectrum of 7 in CDCl3 shows two dou-
Figure 2. ORTEP view of complex 7. The thermal ellipsoids are at
2
blets with satellites at δ = 12.1 ppm (1JPt,P = 3617 Hz, JP,P
30% level of probability.
2
= 24 Hz) and 8.3 ppm (1JPt,P = 3553 Hz, JP,P = 24 Hz).
1
The H NMR spectrum in CDCl3 presents a singlet at δ =
Table 3. Selected bond lengths [Å] and angles [°] for complex 7.
4.72 ppm for the proton 4-H, a doublet at δ = 4.20 ppm
(2JH,H = 10 Hz) for the proton 5-H and two doublets at δ
Bond lengths
Pt1–P1
2.255(1)
2.044(3)
1.379(6)
1.537(7)
1.534(7)
1.562(6)
1.518(7)
1.196(6)
1.412(7)
Pt1–P2
Pt1–O3Ј
C3–O3Ј
C2–O2
C3–O3
C4–O4
C1–O4
C5–C6
C6–O3
2.235(1)
2.027(3)
1.365(5)
1.401(6)
1.437(5)
1.450(6)
1.359(7)
1.524(7)
1.426(6)
= 3.95 and 3.90 ppm for the diastereotopic protons 6-H
and 6Ј-H.
Pt1–O2Ј
C2–O2Ј
C2–C3
C2–C1
C3–C4
C4–C5
C1–O1
C5–O5
Complex 7 is stable in various solvents (in CH2Cl2,
CHCl3 and DMSO no decomposition was detected by 31P
NMR spectroscopy after 3 d). In DMSO in the presence
of NBu4Cl (3 equiv.), only traces of cis-[PtCl2(PPh3)2] were
detected by 31P NMR spectroscopy after 18 h. Complex 7
was recrystallised from acetone and ether to give crystals
suitable for X-ray diffraction. An ORTEP projection of the
molecule is shown in Figure 2. Table 3 reports a selection
of bond lengths and angles for 7.
The projection presented in Figure 2 indicates that the Pt
has an almost distorted square-planar coordination geome-
try, and the Pt1 atom is displaced from the mean plane
passing through the basal atoms P1, P2, O2Ј and O3Ј by
0.0089(2) Å. The conformation of the coordination ring
Pt1–O2Ј–C2–C3–O3Ј is twisted (4T5), and the other two
five-membered rings, C1–C2–C3–C4–O4 and C3–C4–C5–
C6–O3, assume twisted (3T2) and envelope (E5) conforma-
tions, respectively. The O2–H2 and O5–H5 hydroxy groups
give rise to intermolecular hydrogen bonds whose param-
eters are reported in Table 4.
Angles
P1–Pt1–P2
P1–Pt1–O3Ј
P2–Pt1–O3Ј
Pt1–O2Ј–C2
O2Ј–C2–C3
99.9(1)
170.7(1)
89.4(1)
112.7(3)
109.9(4)
P1–Pt1–O2Ј
P2–Pt1–O2Ј
O2Ј–Pt1–O3Ј
Pt1–O3Ј–C3
O3Ј–C3–C2
88.2(1)
171.9(1)
82.6(1)
107.1(2)
113.8(4)
Table 4. Structural parameters of the intermolecular hydrogen
bonds of complex 7.
D–H···A
D–H [Å]
D···A [Å]
H···A [Å]
D–H···A [°]
O2–H2···O5i 1.08(8)
2.924(6)
2.808(5)
1.97(8)
1.81(4)
145(6)
175(3)
O5–H5···O3ii 1.00(4)
i: x – 1/2, –y – 1/2, –z; ii: x + 1/2, –y – 1/2, –z
Eur. J. Inorg. Chem. 2008, 529–537
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
533

P. Bergamini et al.
FULL PAPER
Complex 7 is the first reported example of the coordina-
tion of the hemiketalic form of dehydroascorbic acid. This
coordination locks this form in a very stable mode. This
molecule could be of great interest for structural studies on
the various forms of dehydroascorbic acid involved in the
redox mechanism that makes ascorbic acid one of the most
diffuse antioxidant in biological and synthetic systems.[19]
We propose a hypothesis for the formation of 7, which is
depicted in Scheme 5. Form (a) is in equilibrium with its
hydrated hemiketalic form, which is intercepted by
[Pt(CO3)(PPh3)2] to give complex 7, with concomitant re-
lease of CO2 and water. The formation of 7 is probably
driven by the development of CO2 and by the formation of
three condensed five-membered rings. This reaction gives
rise to two new chiral centres. Although several isomeric
forms of 7 are possible, the reaction product always appears
as a single species (NMR observation). The same product
is obtained when the dimeric form of DHA (c) is used, but
in this case, the process is much slower (10 d versus 3 h).
Figure 3. T2 Cell line growth inhibition.
Table 6. Percentage (ϮSD) of growth inhibition of SKOV3 cells.
50 µ
10 µ
2 µ
1
81.7
15.5
10.9
3.6
86.2
35.5
84.3
20.9
6.7
3.3
6.2
3.3
17.4
11.5
73.7
13.7
0.4
0.4
4.6
1.2
2.1
4.1
18.1
8.1
2
3
4
5
6
7
cisplatin
Scheme 5.
Figure 4. SKOV3 Cell line growth inhibition.
Cell Growth Inhibition of Complexes 1–7
It can be seen that complexes 1, 3 and 5 show an antipro-
liferative activity on T2 that is similar to that of cisplatin at
high and medium concentrations, whereas at 2 µ they are
less active than the reference compound. Complexes 2, 4
and 6 present low activity, probably because of their low
hydrophilicity. Complex 7 appears much more active than
all the others. This trend is maintained for the cisplatin-
resistant cell line SKOV3, where 1 and 5 show some activity
limited to the highest doses; complexes 2, 3 and 4 show no
activity and complex 7 has a remarkable activity at 50 and
10 µ.
In order to check if platinum complexes containing
ascorbic acid are promising candidates for the development
of anticancer drugs, we tested their antiproliferative activity
in vitro on two human tumoural cell lines, the cisplatin-
sensitive T2 and the cisplatin-resistant SKOV3 cell lines.
The results of the cell growth inhibition obtained with com-
plexes 1–7 for the cisplatin-sensitive T2 cell line, relative to
cisplatin, are reported in Table 5 and Figure 3, and for the
cisplatin-resistant SKOV3 cell line, in Table 6 and Figure 4.
Activity measurements at lower doses (Table 7), limited
to the most-active complex 7 and cisplatin as reference
compound, allowed us to estimate an IC50 value of 1.2 µ
for both 7 and cisplatin on the T2 cell line, whereas on
cisplatin-resistant SKOV3, we found a value of 6.9 µ for 7
and 86.2 µ for cisplatin. The subsequently estimated val-
ues for the resistance factors (ratio of the IC50 on resistant
cells/IC50 on parental cells),[20] 5.75 for 7 and 71.8 for cis-
platin, seem to support the hypothesis of a different mecha-
nism of action for the two complexes.
Table 5. Percentage (ϮSD) of growth inhibition of T2 cells.
50 µ
10 µ
2 µ
1
87.1
73.4
69.5
6.6
87.4
54.5
86.2
90.0
83.7
9.8
69.3
5.8
85.4
4.7
86.4
80.0
16.3
6.2
43.2
0.0
9.9
0.0
2
3
4
5
6
7
85.4
65.7
cisplatin
534
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 529–537

New Coordination Modes of Dianionic Chelating Ligand for Platinum
75% yield. C32H30O6P2Pt (768): calcd. C 50.04, H 3.94; found C
Table 7. Percentage of growth inhibition of T2 and SKOV3 cells for
7 and cisplatin.
49.82, H 3.86 IR (KBr): ν = 3500–3100 (νOH), 1700 (νC=O), 1620
˜
(νC=C), 1480–1430 (aromatic νC–C) cm–1. 1H NMR (200 MHz,
50 µ 10 µ 2 µ 0.4 µ 0.08 µ 0.016 µ IC50
3
CDCl3, 25 °C): δ = 8.0–7.5 (m, 20 H, Ph), 4.49 (d, JH,H = 7.1 Hz,
7 (T2)
Cisplatin (T2)
86.2
90.0
86.4 85.4 0.0
80.0 65.7 22.0 0.0
0.0
0.0
0.0
1.2
1.2
1 H, 4-H), 3.71 (m, 3 H, 5-H, 6-H), 3.43 (m, 1 H, OH), 2.40 (m, 4
H, dppe) ppm. 31P NMR (81.15 MHz, CDCl3, 25 °C): δ = 32.8
(1JPt,P = 3364, 2JP,P = 9 Hz), 29.2 (1JPt,P = 3664, 2JP,P = 9 Hz) ppm.
7 (SKOV3)
Cisplatin (SKOV3)
84.3
20.9
73.7 18.1 8.5
13.7 8.1 6.7
4.7
6.3
4.7
0.0
6.9
86.2
[Pt(O2,O3-asc)(PTA)2] (3): A 0.2- solution of cis-[Pt(NO3)2-
(PTA)2] was prepared from cis-[PtCl2(PTA)2] (1.85 mmol, 1.07 g)
and AgNO3 (3.7 mmol, 0.62 g) in H2O (20 mL). The mixture was
left to stir for 90 min at 45 °C. AgCl was then filtered with filter
paper, and sodium ascorbate (3.7 mmol, 0.73 g) was added to the
filtrate whilst stirring under nitrogen. The mixture was left to stir
for 24 h at room temperature. 31P NMR inspection of a solution
aliquot showed the complete formation of 5. The pure product was
precipitated with ethanol. Yield 84%. C18H30N6O6P2Pt·3H2O
(737): calcd. C 29.33, H 4.88, N 11.40; found C 29.29, H 4.87, N
Conclusions
In this work, we have shown that ascorbic acid is a versa-
tile ligand for platinum, where the O2, O3, O5, O6 and C2
atoms can act as anionic donating functionalities. Alterna-
tive synthetic routes to known complexes of -ascorbic acid
with platinum(II) have been proposed, and the hydrophi-
licity of O2,O3 complexes has been enhanced by introduc-
ing PTA as a neutral ligand (complex 3). New modes of
coordination of ascorbic acid and dehydroascorbic acid to
platinum(II) have been characterised in solution by NMR
spectroscopy and in the solid state by X-ray crystallography.
The antiproliferative activity in vitro of complexes 1–7 was
tested, and the best results were obtained for the DHA
complex 7, which was found to be more active than cispla-
tin towards both a cisplatin-sensitive and a cisplatin-resist-
ant cell line.
11.37. IR (KBr): ν = 3300 (br. νOH), 1730 (νC=O), 1641 (νC=C) cm–1.
˜
1H NMR (200 MHz, D2O, 25 °C): δ = 4.50 (m, 12 H, NCH2N),
3
2
4.42 (d, JH,H = 2 Hz, 1 H, 4-H), 4.29 (d, JH,P = 13 Hz, 12 H,
3
4
NCH2P), 3.93 (td, JH,H = 7, JH,H = 2 Hz, 1 H, 5-H), 3.65 (dd,
3JH,H = 10.5, 4JH,H = 2 Hz, 2 H, 6-H) ppm. 31P NMR (81.15 MHz,
D2O, 25 °C): δ = –58.34 (1JPt,P = 3302, JP,P = 23.3 Hz), –58.71
2
(1JPt,P = 3343, JP,P = 23.3 Hz) ppm.
2
[Pt(C2,O5-asc){trans-(R,R-dach)}] (4): [Pt{trans-(R,R-dach)}I2]
(1.5 g, 2.6 mmol) was placed in water (13 mL), and a second solu-
tion containing AgNO3 (0.88 g, 5.2 mmol) in H2O (13.3 mL) was
added. The mixture was left to stir for 90 min at 45 °C, and the
precipitate of AgI was then removed by filtration with filter paper.
The clear filtrate, containing the aqua-species [Pt{trans-(R,R-
dach)}(OH2)2](NO3)2, was degassed and kept under nitrogen, and
then sodium ascorbate (1.02 g, 5.2 mmol) was added. The mixture
was left to stir for 24 h at room temperature, and then at 4 °C
overnight. Crystals slowly formed, which were collected, and fil-
tered with filter paper. Yield 46%. C12H20N2O6Pt·3H2O (537):
calcd. C 26.81, H 4.88, N 5.21; found C 26.45, H 4.75, N 5.27. IR
Experimental Section
[11]
PTA
as well as the Pt complexes [Pt(CO3)(PPh3)2],[9] [Pt(CO3)-
(dppe)],[9] cis-[PtCl2(PTA)2][11] and [Pt{trans-(R,R-dach)}I2][13] were
prepared as reported. All the other chemicals and solvents were
used as purchased (reagent grade). Elemental analyses (C, H, N)
were performed by using a Carlo Erba instrument model EA1110.
FT-IR spectra were recorded with a Nicolet 510P FT-IR instru-
ment (4000–300 cm–1). NMR spectra were recorded with a Bruker
AM spectrometer: 200 MHz (1H NMR), 81.15 MHz (31P NMR).
Peak positions are relative to tetramethylsilane and were calibrated
against the residual solvent resonance (1H) and measured relative
to external 85% H3PO4 with downfield values taken as positive
(31P).
(KBr): ν = 3550–2860 (νOH), 1716 (νC=O), 1630 and 1590 (νC=C
)
˜
cm–1. 1H NMR (200 MHz, D2O, 25 °C): δ = 4.27 (s, 1 H, 4-H),
1
1
4.08 (t, JH,H = 6.5 Hz, 1 H, 5-H), 3.42 (d, JH,H = 6.5 Hz, 2 H, 6-
H), 2.32, 1.91, 1.50, 1.10 (m, 10 H, dach) ppm.
[Pt(O5,O6-(O2,O3dibenz)-asc)(PPh3)2] (5): The carbonate complex
[Pt(CO3)(PPh3)2] (0.15 g, 0.19 mmol) was dissolved in dichloro-
methane (25 mL). To this solution dibenzyl ascorbic acid (68 mg,
0.19 mmol) was added, and the mixture was left overnight whilst
stirring. It was then concentrated to dryness, and the solid residue
was washed with diethyl ether to give pure 5 (0.113 g, 0.12 mmol,
63% yield). C56H48O6P2Pt (1073): calcd. C 62.62, H 4.47; found C
Synthesis and Characterisation of Complexes 1–7
62.61, H 4.40. IR (nujol): ν = 1754 and 1667 (νC=O) cm–1. 1H NMR
[Pt(O2,O3-asc)(PPh3)2] (1): The carbonate complex [Pt(CO3)-
(PPh3)2] (0.15 g, 0.19 mmol) was dissolved in dichloromethane sat-
urated with water (25 mL), and -ascorbic acid (0.034 g,
0.19 mmol) was added. The solution was left to stir whilst shielded
from light for 18 h. It was then concentrated to dryness, and the
solid residue was washed with diethyl ether to give 1 (0.14 g,
0.15 mmol, yield 81%). C42H36O6P2Pt (894): calcd. C 56.37, H
˜
(200 MHz, C3D6O, 25 °C): δ = 7.5–8.0 (m, 40 H, Ph), 5.25 (d, 2JH,H
2
= 12 Hz,1 H, PhCH2), 5.12 (d, JH,H = 12 Hz, 1 H, PhCH2), 4.75
2
2
(d, JH,H = 11 Hz, 1 H, PhCH2), 4.65 (d, JH,H = 11 Hz, 1 H,
3
PhCH2), 4.38 (d, JH,H = 11 Hz, 1 H, 4-H), 4.26 (m, 1 H, 5-H),
4.04 (m, 1 H, 6-H), 3.64 (m, 1 H, 6Ј-H) ppm. 31P NMR
(81.15 MHz, C3H6O, 25 °C): δ = 13.7 (1JPt,P = 3512, JP,P
=
2
21.4 Hz), 12.8 (1JPt,P = 3372, JP,P = 21.4 Hz) ppm.
2
4.03; found C 56.25, H 4.01. IR (KBr): ν = 3500–3100 (νOH), 1780–
˜
1630 (νC=O + νC=C), 1480–1430 (aromatic νC···C) cm–1. 1H NMR
[Pt(dppe)(O5,O6-(O2,O3dibenz)-asc)] (6): The preparation of com-
plex 6 was carried out in the same manner as that for complex 5.
Yield 75%. C46H42O6P2Pt (948): calcd. C 58.22, H 4.43; found C
2
(200 MHz, CDCl3, 25 °C): δ = 7.5–7 (m, 30 H, Ph), 4.30 (d, JH,H
= 7.6 Hz, 1 H, 4-H), 3.60 (m, 1 H, 5-H), 3.30 (m, 2 H, 6-H) ppm.
31P NMR (81.15 MHz, CDCl3, 25 °C): δ = 9.7 (1JPt,P = 3490, JP,P
2
58.21, H 4.45. IR (nujol): ν = 3500–2800 (νOH), 1754 (νC=O), 1670
˜
= 20 Hz), 6.9 (1JPt,P = 3776, JP,P = 20 Hz) ppm.
2
(νC=C), 1480–1430 (νC–C aromatic) cm–1. 1H NMR (200 MHz,
2
[Pt(dppe)(O2,O3-asc)] (2): Complex 2 was prepared in the same
way as complex 1 from the corresponding carbonate precursor in
C3D6O, 25 °C): δ = 8.25 and 7.35 (2m, 30 H, Ph), 5.39 (d, JH,H
=
2
12 Hz, 1 H, PhCH2), 5.22 (d, JH,H = 12 Hz, 1 H, PhCH2), 4.77
Eur. J. Inorg. Chem. 2008, 529–537
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
535

P. Bergamini et al.
FULL PAPER
2
2
(d, JH,H = 11 Hz, 1 H, PhCH2), 4.72 (d, JH,H = 11 Hz, 1 H, Growth Inhibition Assays: Cell growth inhibition assays were car-
PhCH2), 4.44 (m, 1 H, 4-H), 4.14 (m, 1 H, 5-H), 3.89 (m, 1 H, 6-
ried out with the cisplatin-sensitive T2 human cell line and the cis-
H), 3.66 (m, 1 H, 6Ј-H), 2.80 (m, 4 H, dppe) ppm. 31P NMR platin-resistant SKOV3 cell line. T2 is a cell hybrid obtained by the
(81.15 MHz, C3D6O, 25 °C): δ = 29.7 (1JPt,P = 3280, 2JP,P = 9.0 Hz), fusion of the human lynphoblastoid line 174 (B lynphocyte trans-
2
29.4 (1JPt,P = 3229 Hz, JP,P = 9.0 Hz) ppm.
formed by the Epstein-Barr virus) with the CEM human cancer
line (leukaemia T), whereas SKOV3 is derived from a human ovar-
ian tumour. The cells were seeded in triplicate in 96-well trays at a
density of 50ϫ103 in 50 µL of AIM-V medium for T2 and 25ϫ103
in 50 µL of AIM-V medium for SKOV3. Stock solutions (10 m)
of the PtII complexes were made in DMSO and diluted in AIM-V
medium to give final concentration of 2, 10 and 50 µ. Cisplatin
was employed as a control for the cisplatin-sensitive T2 cell line
and for the cisplatin-resistant SKOV3 cell line. Untreated cells were
placed in every plate as a negative control. The cells were exposed
to the compounds for 72 h, after which a 4,5-(dimethylthiozol-2-
yl)2,5-diphenyltetrazolium bromide solution (25 µL, 12 m) was
added. After 2 h of incubation, 100 µL of lysing buffer (50% DMF
+ 20% SDS, pH 4.7) was added to convert 4,5-dimethylthiozol-2-
yl-2,5-diphenyltetrazolium bromide into a brown formazane. After
an additional 18 h, the solution absorbance, proportional to the
number of live cells, was measured by spectrophotometry at 570 nm
and converted to % of growth inhibition.[27]
[Pt(O2,O3-DHA)(PPh3)2] (7): The complex [Pt(CO3)(PPh3)2]
(0.1 g, 0.13 mmol) was dissolved in anhydrous dichloromethane
(15 mL), and dehydroascorbic acid (22 mg, 0.13 mmol) was added.
The reaction mixture was left to stir for 60 h and concentrated to
dryness, and the solid residue recrystallised from acetone and ether
(0.06 g, 0.067 mmol). Yield 52.5%. C42H36O7P2Pt (910): calcd. C
55.44, H 3.96; found C 55.15, H 3.97. IR (nujol): ν = 3500–2800
˜
(br., νOH), 1795 (νC=O), 1680 (νC=C), 1480 –1430 (νC–C aromatic)
cm–1. 1H NMR (200 MHz, CDCl3, 25 °C): δ = 7.60–7.05 (m, 30 H,
Ph), 4.72 (s, 1 H, 4-H), 4.20 (d, 2JH,H = 10 Hz 1 H, 5-H), 3.95 (d,
2JH,H = 10 Hz, 1 H, 6-H), 3.90 (d, 2JH,H = 10 Hz, 1 H, 6Ј-H) ppm.
31P NMR (81.15 MHz, CDCl3, 25 °C): δ = 12.1 (1JPt,P = 3617,
2
2JP,P = 24 Hz), 8.3 (1JPt,P = 3553, JP,P = 24 Hz) ppm.
Crystal-Structure Determination: The crystal data of compounds 6
and 7 were collected at room temperature by using a Nonius Kappa
CCD diffractometer with graphite monochromated Mo-Kα radia-
tion. The data sets were integrated with the Denzo-SMN pack-
age[21] and corrected for Lorentz, polarisation and absorption ef-
fects (SORTAV[22]). The structures were solved by direct methods
(SIR97[23]) and refined by using full-matrix least-squares with all
non-hydrogen atoms refined anisotropically and with hydrogen
atoms included on calculated positions, riding on their carrier
atoms, except for the hydroxy hydrogen atoms of 7, which were
refined isotropically. All calculations were performed with
SHELXL-97[24] and PARST[25] implemented in the WINGX[26] sys-
tem of programs. The crystal data are given in Table 8. CCDC-
655518 and -655519 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgments
The authors thank Consorzio Interuniversitario di Ricerca in Chi-
mica dei Metalli nei Sistemi Biologici for a grant to E. M.
[1]
[2]
W. J. Pardridge, Nat. Rev. Drug Discovery 2002, 1, 131–139.
S. Manfredini, B. Pavan, S. Vertuani, M. Scaglianti, D. Com-
pagnone, C. Biondi, A. Scatturin, S. Manganelli, L. Ferraro, P.
Prasad, A. Dalpiaz, J. Med. Chem. 2002, 45, 559–562.
[3]
[4]
M. B. Davies, Polyhedron 1992, 11, 285–321.
G. K. Anderson, M. J. Arendse, N. P. Rath, Inorg. Chem. 1999,
38, 5864–5869.
Table 8. Crystallographic data of complexes 6 and 7.
Compound
6
7
[5]
[6]
Formula
M
Space group
Crystal system
a [Å]
C46H42O6P2Pt
947.83
C42H36O7P2Pt
909.74
a) M. J. Arendse, G. K. Anderson, R. N. Majola, N. P. Rath,
Inorg. Chim. Acta 2002, 65–69; b) R. N. Bose, E. W. Weaver, J.
Chem. Soc. Dalton Trans. 1997, 1797–1799.
a) B. Agus, S. S. Gamghir, W. M. Pardridge, C. Spielholz, J.
Baselga, J. Vera, D. W. Golde, J. Clin. Invest. 1997, 100, 2842–
2848; b) S. C. Rumsey, O. Kwon, G. W. Xu, C. F. Burant, I.
Simpson, M. Levine, J. Biol. Chem. 1997, 18982–18989.
a) T. Minami, J. Okazaki, A. Kawabata, R. Kuroda, Y. Oka-
zaki, Toxicology 1998, 130, 107–113; b) T. Anda, H. Yamash-
ita, H. Khalid, K. Tsutsumi, H. Fujita, Y. Tokunaga, S. Shib-
ata, Neurological Res. 1997, 19, 369–376.
a) T. K. Miyamoto, H. Yuge, Chem. Lett. 1996, 375–376; b) H.
Yuge, T. K. Miyamoto, Inorg. Chim. Acta 2002, 329, 66–70.
P212121
orthorhombic
10.2000(1)
18.9649(3)
21.8290(4)
4222.6(1)
4
295
1.491
1896
34.45
P212121
orthorhombic
10.7247(1)
17.7069(2)
19.8126(2)
3672.44(7)
4
295
1.606
1808
38.65
b [Å]
c [Å]
U [Å3]
Z
[7]
T [K]
Dc [gcm–3
]
F(000)
[8]
[9]
M(Mo-Kα) [cm–1
Measured reflections
Unique reflections
Rint
]
29342
10157
0.0502
7777
39562
9060
0.0550
7674
a) M. A. Andrews, E. J. Voss, G. L. Gould, W. T. Klooster, T. F.
Koetzle, J. Am. Chem. Soc. 1994, 116, 5730–5740; b) M. A.
Andrews, G. L. Gould, W. T. Klooster, K. S. Koenig, E. J. Voss,
Inorg. Chem. 1996, 35, 5478–5483.
Observed reflections
[IՆ2σ(I)]
[10]
[11]
θmin–θmax [°]
hkl ranges
2.34–28.00
3.62–28.00
P. J. Dyson, G. Sava, Dalton Trans. 2006, 1929–1933.
D. J. Darensbourg, T. J. Decuir, N. W. Stafford, J. B. Robertson,
J. D. Draper, J. H. Reibenspies, Inorg. Chem. 1997, 36, 4218.
a) L. S. Hollis, A. R. Amundsen, E. W. Stern, J. Am. Chem.
Soc. 1985, 107, 274–276; b) L. S. Hollis, E. W. Stern, A. R.
Amundsen, A. V. Miller, S. L. Doran, J. Am. Chem. Soc. 1987,
109, 274–276; c) L. S. Hollis, E. W. Stern, Inorg. Chem. 1988,
27, 2826–2831.
–11,13; –25,24; –28,28 –14,14; –23,23; –26,26
R(F2) (obs.reflections)
wR(F2) (all reflections)
No. variables
0.0356
0.0785
483
1.023
–0.025(6)
1.93; –1.25
0.0318
0.0673
476
1.038
–0.031(5)
1.10; –1.27
[12]
Goodness-of-fit
Flack parameter
∆ρmax; ∆ρmin [eÅ–3
]
536
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 529–537

New Coordination Modes of Dianionic Chelating Ligand for Platinum
[13] L. S. Hollis, S. L. Doran, A. R. Amundsen, E. W. Stern, Inorg.
Synth. 1990, 27, 283–286.
[14] S. Manfredini, S. Vertuani, B. Pavan, F. Vitali, M. Scaglianti,
F. Bortolotti, C. Biondi, A. Scatturin, P. Prasad, A. Dalpiaz,
Bioorg. Med. Chem. 2004, 12, 5453.
[15] a) K. Kato, S. Terao, N. Shimamoto, M. Hirata, J. Med. Chem.
1988, 31, 793–798; b) M. G. Kulgami, S. R. Thopate, Tetrahe-
dron 1996, 52, 1293–1302.
[20] A. M. Bergman, C. M. Kuiper, D. A. Voorn, E. M. Comijn, F.
Myhren, M. L. Sandvold, H. R. Hendriks, G. J. Peters, Bio-
chemical Pharmacology 2004, 67, 503–511.
[21] Z. Otwinowski, W. Minor in Methods in Enzymology (Eds.:
C. W. Carter, R. M. Sweet Editors), Vol. 276, Part A, Academic
Press, London, 1997, p. 307.
[22] R. H. Blessing, Acta Crystallogr. Sect. A 1995, 51, 33.
[23] SIR97:A. Altomare, M. C. Burla, M. Camalli, G. L. Cas-
carano, C. Giacovazzo, A. Guagliardi, A. G. Moliterni, G. Pol-
idori, R. Spagna, J. Appl. Crystallogr. 1999, 32, 115.
[24] G. M. Sheldrick, SHELX-97, Program for Crystal Structure
Refinement, University of Göttingen, Germany, 1997.
[25] M. Nardelli, J. Appl. Crystallogr. 1995, 28, 659.
[26] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837.
[27] M. B. Hansen, S. E. Nielsen, K. Berg, J. Immunol. Methods
1989, 119, 203.
[16] M. N. Burnett and C. K. Johnson, ORTEP III, Report ORNL-
6895, Oak Ridge National Laboratory, Oak Ridge, TN, 1996.
[17] S. R. Malic´, D. Svedruzic, T. Gazivoda, A. Marunovic´, A. H.
ˇ
Brundic´, A. Nagl, J. Balzarini, E. De Clerq, M. Mintas, J. Med.
Chem. 2000, 43, 4806–4811.
[18] G. M. Tsivgoulis, P. A. Afroudakis, P. V. Ioannou, J. Inorg. Bi-
ochem. 2004, 98, 649–656.
[19] T. Gazidova, K. Fittine, I. Lovric´, D. Makuc, J. Plavec, M.
ˇ
Cetina, D. Mrvosˇ-Sermek, L. Suman, M. Kralj, K. Pavelic´, M.
Received: July 31, 2007
Mintas, S. Raic´-Malic´, Carbohydr. Res. 2006, 433–442.
Published Online: November 30, 2007
Eur. J. Inorg. Chem. 2008, 529–537
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
537