Structure-activity relationship studies for three new asymmetric cis-platinum(II) aminoethanol-based complexes

Article abstract of DOI:10.1016/j.ica.2006.05.011

The design and synthesis of three asymmetrical platinum(II) analogues of cisplatin with substituents on the amine, varying in polarity and steric bulk is presented. Their biological activities, as studied using in vitro cytotoxicity studies in cisplatin sensitive and the corresponding cisplatin resistant cell lines, cellular uptake experiments and in a reaction with model DNA base GMP, are presented. All compounds exhibit promising cytotoxicity in the cisplatin sensitive cell lines albeit lower than cisplatin. On the other hand, the complexes partly overcome cisplatin resistance in the resistant cell lines. A direct correlation between cytotoxicity and cellular uptake was found. Conversely, the rate of reaction of all compounds with the model base GMP was found to be very similar and faster than cisplatin. It was therefore concluded that the difference in activity observed for these complexes is due to differential cellular uptake rather than the reactivity towards the cellular target of platinum complexes, nuclear DNA.

Full text of DOI:10.1016/j.ica.2006.05.011

Inorganica Chimica Acta 359 (2006) 4125–4129
www.elsevier.com/locate/ica
Structure–activity relationship studies for three new asymmetric
cis-platinum(II) aminoethanol-based complexes
Sabine van Rijt, Steven van Zutphen, Hans den Dulk, Jaap Brouwer, Jan Reedijk *
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands
Received 26 April 2006; accepted 7 May 2006
Available online 19 May 2006
Abstract
The design and synthesis of three asymmetrical platinum(II) analogues of cisplatin with substituents on the amine, varying in polarity
and steric bulk is presented. Their biological activities, as studied using in vitro cytotoxicity studies in cisplatin sensitive and the corre-
sponding cisplatin resistant cell lines, cellular uptake experiments and in a reaction with model DNA base GMP, are presented. All com-
pounds exhibit promising cytotoxicity in the cisplatin sensitive cell lines albeit lower than cisplatin. On the other hand, the complexes
partly overcome cisplatin resistance in the resistant cell lines. A direct correlation between cytotoxicity and cellular uptake was found.
Conversely, the rate of reaction of all compounds with the model base GMP was found to be very similar and faster than cisplatin. It was
therefore concluded that the difference in activity observed for these complexes is due to differential cellular uptake rather than the reac-
tivity towards the cellular target of platinum complexes, nuclear DNA.
ꢀ
2006 Elsevier B.V. All rights reserved.
Keywords: Asymmetric platinum complexes; Cisplatin; Cytotoxicity; Antitumour agents
1
. Introduction
located cis to aminoethanol (1) to posses promising cyto-
toxic activity [5]. In the present study we present two
analogues of this complex which have either increased
steric bulk and increased lypophilicity (2), or increased
steric bulk and increased hydrophilicity (3) introduced
to the amino ethanol ligand. The steric bulk is likely
to decrease the reaction rate of the complexes with nucle-
ophiles such as DNA. Conversely the introduction of
more hydroxyl residues is likely to change the nature
and strengthen coordination and hydrogen-bond interac-
tion of the complex with DNA [6–9]. Changes in the
hydrophilicity or lypophilicity are also likely to affect
the cellular uptake of the compounds [10]. Through
studying the effects of these systematic variations on
cytotoxicity, cellular uptake and DNA binding rate we
expect to develop a more profound understanding of
the structure–activity relationships for asymmetric cis-
platin analogues. In addition, complexes with improved
biological activity compared to compound 1 may be
discovered.
The serendipitous discovery of the antitumour activity
of cisplatin by Rosenberg in 1965 [1,2] has led to its world-
wide application in chemotherapy. In attempts to over-
come the drawbacks of cisplatin, such as side effects and
cisplatin resistance, many cisplatin analogues have been
synthesized. So far only a handful of cisplatin analogues
have shown marked improved biological properties and
have subsequently entered the clinic [3,4].
However, based on the work carried out in this field it
can be concluded that even small variations in the amine
ligands of cisplatin analogues have already huge effects
on the activity of the complexes. In a recent combinato-
rial evaluation of asymmetric cisplatin analogues, we
reported about a compound composed of methylamine
*
Corresponding author. Tel.: +31 71 527 4459; fax: +31 71 527 4671.
E-mail address: reedijk@chem.leidenuniv.nl (J. Reedijk).
0
020-1693/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2006.05.011

4126
S. van Rijt et al. / Inorganica Chimica Acta 359 (2006) 4125–4129
2
. Experimental
(27.8 mg, 0.238 mmol) or threonine alcohol (22.6 mg,
.216 mmol) in 3 ml of DMF to the dimer (0.1 g, 0.1 mmol)
0
2
.1. General
in 5 ml of DMF yielded the corresponding asymmetric cis-
iodide complexes after 24 h of reaction. These were cleanly
converted to target compounds 1–3, respectively, by treat-
All NMR spectra were recorded on a 300 MHz Bruker
DPX300 spectrometer with a 5 mm multi-nucleus probe.
MS measurements were taken on a Finnigan AQA instru-
ment equipped with electrospray interface (ESI). The 100-
mm culture and microwell plates were obtained from
NUCLON (Roskilde, Denmark). MTT, [3-(4,5-dimethyl-
thiazol-2-yl)-2,5-diphenyltetrazoliumbromide], was pur-
chased from Sigma Chemical Co. The A2780 and
A2780R cell-lines were a generous gift from Dr. J.M. Perez
ment with AgNO (65 mg, 0.39 mmol) for 24 h in the dark,
3
removal of the precipitated AgI and treatment with KCl
(62 mg, 0.83 mmol) for 48 h. The crude compounds were
lyophilised, redissolved in acetone, filtered and evaporated
to dryness to yield the products as pale yellow powders.
cis-[PtCl (NH CH )(aminoethanol)] (1): Yield 89%.
2
2
3
1
H NMR (acetone-d ) d (ppm): 2.59 (m, 3H; NH CH ),
6
2
3
2.97 (m, 2H; NH CH ), 3.91 (t, J = 12.66 Hz, 2H;
2
2
1
95
(
Universidad Autonoma de Madrid, Spain). The cells were
CH OH);
(calc): 382.9 (358.1) [M+Na ].
Pt NMR (DMF): ꢀ2238 ppm; ESI-MS: m/z
2
+
grown as monolayers in Dulbecco’s modified Eagle’s Med-
ium supplemented with 10% fetal calf serum (Gibco, Pais-
ley, Scotland), penicillin (100 units/ml: Dufecha, the
Netherlands) and streptomycin (100 lg/ml: Dufecha, the
Netherlands). The L1210 and L1210R cell lines were cul-
tured in McCoy’s 5a medium supplemented with 10% fetal
calf serum (Gibco, Paisley, Scotland), penicillin (100 units/
ml: Dufecha, the Netherlands) and streptomycin (100 lg/
ml: Dufecha, the Netherlands). During growth the cells
grew partly in suspension and partly adhered to the flask
walls.
cis-[PtCl (NH CH )(leucine alcohol)] (2): Yield
2
2
3
1
25.5%.
H
NMR (acetone-d6)
d
(ppm): 0.94 (d,
J = 6.3 Hz, 6H; CH(CH ) ), 1.53 (m, 2H; NH CHCH -
3
2
2
2
CH(CH ) ), 1.81 (m, 1H; CH(CH ) ), 2.60 (m, 3H;
3
2
3 2
*
NH CH ), 2.96 (m, 1H; NH CH), 3.70 (d d,
2
3
2
95
1
J = 4.3 * 6.98 Hz, 2H; CH OH);
Pt NMR (DMF):
2
+
ꢀ2224 ppm; ESI-MS: m/z (calc): 415.2 (414.2) [M+H ].
cis-[PtCl (NH CH )(threonine alcohol)] (3): Yield
2
2
3
1
58.7%.
H
NMR (acetone-d6)
d
(ppm): 1.34 (d,
J = 6.48 Hz, 3H; CH(CH )), 1.38 (s, 1H; NH CHCH),
3
2
2
.63 (m, 3H; NH CH ), 2.90 (m, 1H; NH CH), 3.27 (m,
2 3 2
1
95
2
.2. Synthesis of threonine alcohol
2H; CH OH);
Pt NMR (DMF): ꢀ2291; ESI-MS: m/z
+
2
(calc): 402.9 (402.1) [M+H ].
Threonine methyl ester (0.2 g, 1.18 mmol) dissolved in
dry THF (10 ml) was added dropwise to LiAlH in dry
2.4. Cytotoxicity assays
4
THF (12 ml). The reaction mixture was refluxed for 3 h
and allowed to proceed overnight at r.t. The mixture was
cooled to 0 ꢁC, quenched with 0.1 M HCl, filtered and
the solvent was removed in vacuo. Purification on silica
The cellular survival was evaluated using the MTT
method [11]. Cells were plated onto 96-well sterile plates
3
in 100 ll of medium at a density of 2 · 10 cells per well
(
methanol:chloroform (1:1) + 5% ammonia) yielded the
and incubated for 48 h at 37 ꢁC in a 7% CO containing
2
1
product in 75% yield. H NMR: d (ppm): 3.86 (2H, m,
incubator. The compounds were added in final concentra-
tions ranging from 0 to 50 lM. After 72 h, 50 ll MTT in
PBS (5 mg/ ml) was added to each well and the plates were
incubated for 2–3 h at 37 ꢁC. The solution was carefully
removed and the remaining crystals dissolved in 100 ll of
DMSO. Cell survival was evaluated by measuring the
absorbance at 590 nm. All cytotoxicity tests were per-
formed three times in quadruplicate. The IC50 values were
calculated from curves constructed by plotting cell survival
(%) versus compound concentration (in lM).
2
1
6
2.37 Hz), 3.65 (1H, m, 6.48 Hz), 3.03 (1H, m, 10.67 Hz),
.27 (3H, d, 6.34 Hz). C NMR: d (ppm): 65.02 (1H),
2.20 (2H), 59.73 (1H), 20.02 (3H).
1
3
2
.3. Synthesis of complexes 1–3
The symmetric cis-[PtI (NH CH ) ] (4) was formed by
2
2
3 2
treating K PtCl (0.5 g, 1.2 mmol) in water (25 ml) with
2
4
KI (2.0 g, 12 mmol) for 1.5 h, then methylamine (207 ll
0% wt in water, 2.4 mmol) for 3 h. The resultant yellow
4
precipitate was washed with water, methanol and diethyl
2.5. Cellular uptake experiments
ether and dried in a vacuum oven at 40 ꢁC. Yield: 81%,
1
H NMR (acetone-d ): d (ppm): 4.45 (4H, s), 2.65 (6H,
Cells of the A2780 cell line were plated onto 12-well ster-
ile plates in 2 ml of medium at a density of 1 · 10 cells per
6
1
95
5
m).
Pt NMR (acetone-d ): ꢀ3300 ppm. This product
6
was suspended in water (2 ml) and ethanol (9 ml) and trea-
well and incubated for 48 h. The compounds were added to
the wells in a final concentration of 50 lM and incubated
for 2 h. After the incubation time, the cells were washed
twice with PBS solution and were treated with triton + 1%
SDS to lyse the cells. The Pt cellular uptake was measured
using ICP-OES. The experiments were carried out twice in
triplicate.
ted with 17 equiv. of HClO (70%) over a period of 7 days
4
to yield the dimer cis/trans-[PtI (NH CH ) ] (5) as a red-
2
2
3 2 2
brown precipitate which was filtered off, washed with water
1
95
and dried in a vacuum oven at 40 ꢁC. Yield 92%;
Pt
NMR (DMF): ꢀ3967 ppm and ꢀ3981 ppm. Slow addition
of aminoethanol (12.5 ll, 0.21 mmol), leucine alcohol

S. van Rijt et al. / Inorganica Chimica Acta 359 (2006) 4125–4129
4127
0
2
.6. Reactions with 5 -guanosine monophosphate (GMP)
the exchange of the iodido ligands for chlorides to yield
the final complexes, 1–3. This could be monitored by a
1
95
All reactions were carried out in NMR tubes in deuter-
Pt NMR shift of around ꢀ3300 to around ꢀ2200 ppm.
195
ated phosphate buffer (50 mM KD PO , pH 7) Four equiv-
The final complexes were fully characterized using
NMR, H NMR, MS and IR (see Fig. 1).
Pt
2
4
0
1
alents of GMP (5 -guanosine monophosphate) were
incubated with complexes 1–3 at 37 ꢁC. 2% deuterated
DMF was added to fully dissolve the complexes. The final
concentration of the investigated platinum complexes was
kept constant at 4 mM. The reactions were monitored by
3.2. Cytotoxic activity of the asymmetric platinum
compounds
1
recording H NMR spectra at time intervals of 30 min
The cytotoxic activity of the asymmetric Pt(II) com-
pounds was investigated on human ovarian carcinoma cell
line sensitive and resistant to cisplatin (A2780 and A2780R,
respectively) and mouse leukaemia cell line sensitive and
resistant to cisplatin (L1210 and L1210R, respectively).
As a control, cisplatin was included in the assay. The calcu-
lated IC50 values, the concentration of complex needed to
inhibit 50% of cell growth, are summarized in Table 1.
Compound 1 is the least active compound of the series,
showing only moderate activity in the cell lines sensitive
to cisplatin. The other two compounds show promising
activity with IC50 values close to cisplatin in both the sen-
sitive and the resistant cell lines. Moreover, their decrease
in activity from the sensitive to the resistant cell lines is
lower than for cisplatin, as can be seen by the calculated
resistance factors. This is commonly observed for trans
platinum complexes [17,18], but is quite uncommon for
platinum complexes in cis geometry [19]. In particular com-
pound 3 appears to be a promising drug candidate, judging
from these cytotoxicity values.
for 13 h. Additional spectra were taken after 24 h and after
7
2 h. The formation of the products was quantified using
the relative H8 signal intensities of GMP.
3
. Results and discussion
3.1. Synthesis and characterisation of the asymmetric
cis-Pt(II) compounds
The synthesis of the asymmetric platinum complexes 1–3
was accomplished in five steps as depicted in Scheme 1.
Commercially available methylamine, leucine alcohol and
ethanolamine were used. Threonine alcohol was obtained
by reduction of threonine methyl ester using LiAlH [12].
4
By addition of methylamine to K PtI dissolved in water,
2
4
symmetric cis-bis(amine)diiodidooplatinum complex
4
was formed as a yellow precipitate [13]. This could be con-
verted to the iodido-bridged complex 5 through treatment
with perchloric acid [14–16]. This is a slow conversion that
1
95
can be monitored by following the Pt NMR signal of the
starting material at ꢀ3300 ppm disappear with the appear-
ance of two new peaks at ꢀ3967 and ꢀ3981 ppm. These
new peaks correspond to the cis and the trans isomer of
the bridged compound [13]. Reaction of both these isomers
with the ethanol amines will yield the asymmetric cis com-
plexes, so that there is no need to separate the isomers. For
the reaction of the dimer with the ethanol amines, the best
results were obtained when performed in DMF. Although
this may lead to isomerisation of the resultant complexes,
the reaction takes place much faster than in water due to
improved solubility of the reagents. Furthermore, we did
not observe any undesired isomerisation as can be seen in
3
.3. Cellular uptake experiments
To further investigate the effect of the different struc-
tural features of compounds 1–3, the cellular uptake of
the compounds was studied in the A2780 cell line. The
OH
OH
OH OH
H N NH2
H N NH2
H N NH2
2
2
2
Pt
Pt
Pt
Cl
Cl
Cl
Cl
Cl
Cl
1
95
1
2
3
the
Pt NMR chemical shifts and the IR spectra of the
final complexes. The last step of the synthesis involves
Fig. 1. Asymmetric platinum(II) complexes 1–3.
OH
R
2
K
a, b
c
d, e
Cl
Cl
Cl
Cl
H N NH2
I
I
NH2
I
H N NH2
2
2
Pt
Pt
Pt Pt
I
Pt
I
I
H N
Cl
Cl
2
4
5
1 R = H
2
3
R = CH CH(CH )
2
3 2
R = CH(CH )OH
3
Scheme 1. Synthesis of the complexes 1–3. Reagents and conditions: (a) 10 equiv. of KI in H
HClO in H O, 7 days; (d) 2 equiv. of aminoethanol in DMF, 24 h; (e) 1.9 equiv. of AgNO in DMF: H
1:1), 48 h.
2
O, 1.5 h; (b) 2 equiv. methylamine in H
2
O, 3 h; (c) 17 equiv.
4
2
3
2
O (1:1), 24 h then 4 equiv. of KCl in DMF: H O
2
(

4128
S. van Rijt et al. / Inorganica Chimica Acta 359 (2006) 4125–4129
Table 1
The IC50 values in the A2780 human ovarian cancer and the L1210 mouse leukaemia cell lines as well as their resistant analogues, the resistance factor and
cellular uptake in the A2780 cell-line for compounds 1–3 and cisplatin
5
Complex
A2780 IC50 (lmol)
A2780R IC50 (lmol)
rf
L1210 IC50 (lmol)
L1210R IC50 (lmol)
rf
1.7
3.8
1.8
Cellular uptake (lmol/10 cells)
1
4.47
12.9
1.91
1.43
> 50
51.1
13.2
24.8
·
3.9
20.8
5.53
8.52
1.03
35.7
20.9
15.7
15.0
0.023 ± 0.006
0.066 ± 0.008
0.209 ± 0.012
0.022 ± 0.007
2
3
6.9
17.3
Cisplatin
14.6
compounds were added to the cells in a final concentration
of 50 lM and incubated for 2 h. After the incubation time,
the cells were washed and collected in order to determine
their platinum content using ICP-OES. The results are
summarized in Table 1. It can be seen that compound 1
has a cellular uptake close to cisplatin. Compound 2 has
a 3-fold higher uptake and compound 3 a 9-fold higher
uptake. This uptake follows exactly the same trend as
observed in the cytotoxic data for these compounds: the
more cytotoxic compound is found in much higher concen-
tration inside the cells. Clearly, the side-group on the
amino ethanol of compound 3 has the effect of increasing
the transport efficiency across the cellular membranes.
shift of the H8 proton of GMP [21]. The reaction of the
three platinum complexes with a 4-fold excess of GMP
was followed over time by H NMR in a buffered solution
1
at 37 ꢁC, mimicking biological conditions, until no further
changes were observed. For all complexes a new peak
downfield from the H8 signal of free GMP was observed
after 1 h of reaction. This is a clear indication of the coor-
dination of the complexes to the N7 atom of GMP. After
approximately 14 h no further changes to the spectra were
observed, indicating a full conversion of the dichlorido
compounds to their corresponding bis-GMP species. For
compound 1 the final complex showed only one peak at
8.6 ppm (Fig. 2). For both complex 2 and 3 a multiplet
was observed between 8.5 and 8.8 ppm. This is due to the
rotamers formed as a result of restricted rotation of the
GMP on the NMR time scale with these sterically demand-
ing ligands [22]. The mono-adduct, where only one GMP is
coordinated to the central platinum atom, was only
observed for compound 3 (Fig. 2). This may be due to
the additional hydroxyl residue on the ligand, which can
stabilize this intermediate species. Apparently in com-
pound 1 and 2 the substitution of a second molecule of
3
.4. Reactions with GMP
Consecutively, the study of the reaction of compounds
1
–3 with GMP was carried out. This reaction acts as a
model for the reaction of a platinum complex with its ulti-
mate cellular target: guanosine residues on nuclear DNA
[
20]. The formation and disappearance of platinum-GMP
1
adducts can be followed by H NMR, using the chemical
1
0
Fig. 2. H NMR spectra of compounds 1–3 with 4 equiv. of 5 GMP in a buffered solution at 37 ꢁC as a function of time. The symbols show the H8 signals
of the GMP adducts; bisadduct of compound 1 (·), bisadduct of compound 2 (j), mono-adduct of compound 3 (ꢁ) and the bisadduct of compound 3 (d).

S. van Rijt et al. / Inorganica Chimica Acta 359 (2006) 4125–4129
4129
Table 2
700.99.305). The support and sponsorship concerted by
COST Action D20 is kindly acknowledged. The authors
thank Johnson and Matthey (Reading, UK) for their gen-
erous loan of K PtCl .
Half-life times of the formation of the bis-adducts of compounds 1–3 with
GMP
Complex
t1/2 (h)
2
4
1
2
3
2.7
2.5
2.4
References
[
[
1] B. Rosenberg, L. Van Camp, T. Krigas, Nature 205 (1965) 698.
2] B. Rosenberg, L. Van Camp, J.E. Trosko, V.H. Mansour, Nature
2
22 (1969) 385.
GMP occurs too fast for the mono-GMP adduct to be
observed. With half-life times (t1/2, Table 2) between 2.4
and 2.7 h all three complexes react at a comparable rate
with GMP. Cisplatin, in contrast, has a half-life time of
[
[
3] B. Desoize, Anticancer Res. 24 (2004) 1529.
4] M. Galanski, V.B. Arion, M.A. Jakupec, B.K. Keppler, Curr. Pharm.
Design 9 (2003) 2078.
[5] S. van Zutphen, E.A. Stone, S. van Rijt, M.S. Robillard, G.A. van
der Marel, H.S. Overkleeft, H. den Dulk, J. Brouwer, J. Reedijk, J.
Inorg. Biochem. 99 (2005) 2032.
8
h [23–25]. It may be that this relatively high reactivity
causes the complexes to be less than, or as cytotoxic as cis-
platin, even though the cellular uptake of complex 2 and 3
is much higher than that of cisplatin.
[
[
[
[
6] B. Dassarma, S.K. Daley, R.K. Elespuru, Chem.-Biol. Interact. 46
1983) 219.
7] Q.Y. Xu, A.R. Khokhar, J.L. Bear, Inorg. Chim. Acta 178 (1990)
07.
(
1
8] A.R. Khokhar, Q.Y. Xu, R.A. Newman, Z.H. Siddik, J. Inorg.
Biochem. 43 (1991) 57.
9] A.R. Khokhar, Q.Y. Xu, R.A. Newman, Y. Kido, Z.H. Siddik, J.
Inorg. Biochem. 45 (1992) 211.
4
. Conclusion
The importance of the relatively less-studied asymmetric
cisplatin analogues composed of two different amine ligands
in cis position was recently highlighted by our combinato-
rial evaluation [5]. In the current work we have shown that
by further tuning the steric bulk and polarity, asymmetric
complexes of a cytotoxic activity close to cisplatin can be
found. Moreover, we show that these changes lead to com-
plexes which partly overcome cisplatin resistance. In the
current work it can be seen that small changes in the com-
position of one of the two amine ligands have dramatic
effects on the cellular uptake and consequently the cytotox-
icity of the complexes. However, the reactivity of this partic-
ular series of compounds may be too high for them to show
activities surpassing cisplatin in both cisplatin sensitive and
cisplatin resistant cell lines. Our current focus has therefore
been reducing the reactivity of cisplatin analogues [26].
Hopefully a combination of these strategies will lead to
the much needed clinically useful replacements of cisplatin.
[10] M.S. Robillard, B.A.J. Jansen, M. Lochner, H. Geneste, J. Brouwer,
M. Hesse, J. Reedijk, Helv. Chim. Acta 84 (2001) 3023.
[
[
[
11] T. Mosmann, J. Immunol. Meth. 65 (1983) 55.
12] E. Kusters, W. Rahmen, Z. Naturforsch. (B) 43 (1988) 619.
13] F.D. Rochon, V. Buculei, Inorg. Chim. Acta 357 (2004) 2218.
[14] F.D. Rochon, P.C. Kong, Can. J. Chem.-Rev. Can. Chim. 64 (1986)
1894.
[
[
[
[
15] F.D. Rochon, L. Fleurent, Inorg. Chim. Acta 143 (1988) 81.
16] F.D. Rochon, V. Buculei, Inorg. Chim. Acta 358 (2005) 2040.
17] N. Farrell, Cancer Invest. 11 (1993) 578.
18] N. Farrell, Metal Ions Biol. Syst. 32 (1996) 603.
[19] E. Pantoja, A. Alvarez-Valdes, J.M. Perez, C. Navarro-Ranninger, J.
Reedijk, Inorg. Chim. Acta 339 (2002) 525.
[
[
20] J. Reedijk, Proc. Natl. Acad. Sci. USA 100 (2003) 3611.
21] S. Komeda, H. Yamane, M. Chikuma, J. Reedijk, Eur. J. Inorg.
Chem. (2004) 4828.
[
22] S. Moradell, J. Lorenzo, A. Rovira, S. van Zutphen, F.X. Aviles, V.
Moreno, R. de Llorens, M.A. Martinez, J. Reedijk, A. Llobet, J.
Inorg. Biochem. 98 (2004) 1933.
[
[
[
[
23] A. Zenker, M. Galanski, T.L. Bereuter, B.K. Keppler, W. Lindner,
J. Biol. Inorg. Chem. 5 (2000) 498.
24] P. Zollner, A. Zenker, M. Galanski, B.K. Keppler, W. Lindner, J.
Mass Spectrom. 36 (2001) 742.
25] A. Zenker, M. Galanski, T.L. Bereuter, B.K. Keppler, W. Lindner,
J. Chromatogr. B 745 (2000) 211.
26] S. van Zutphen, E. Pantoja, R. Soriano, C. Soro, D.M. Tooke, A.L.
Spek, H. den Dulk, J. Brouwer, J. Reedijk, Dalton Trans. (2006)
1020.
Acknowledgements
This research has been financially supported by the
Council for Chemical Sciences of the Netherlands Organi-
sation for Scientific Research (CW-NWO; Grant No.