Synthesis of cycloruthenated compounds as potential anticancer agents

Article abstract of DOI:10.1002/ejic.200601149

A library of 19 cycloruthenated derivatives is constructed by making use of the well-known cyclometalation reaction. Their geometries are modified in a straightforward manner by addition of either mono- or bidentate ligands, such as bipyridine, phenanthroline, 1,2-bis(diphenylphosphanyl)ethane, dimethylphenyiphosphane, triphenylphosphane, and 1,3,5-triaza-7- phosphatricyclo[3.3.1.1]decane (PTA) ligands, to cationic cycloruthenated centers. The antitumor properties of the compounds thus obtained are investigated in order to compare them with recently reported ruthenium complexes and cisplatin. IC₅₀ values against mammalian cells (A-172, HCT-116, and RDM-4) are determined for the library compounds and some of them, such as those derived from orthoruthenated phenylpyridine and a bidentate N,N ligand, display activity of the same order of magnitude as cisplatin. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejic.200601149

FULL PAPER
DOI: 10.1002/ejic.200601149
Synthesis of Cycloruthenated Compounds as Potential Anticancer Agents
Lida Leyva,[a] Claude Sirlin,*[a] Laura Rubio,[b] Cecilia Franco,[b] Ronan Le Lagadec,[b]
John Spencer,[c] Pierre Bischoff,[d] Christian Gaiddon,[e] Jean-Philippe Loeffler,[e] and
Michel Pfeffer*[a]
Keywords: Bioinorganic chemistry / Ruthenium / Metallacycles / Antitumor agents
A library of 19 cycloruthenated derivatives is constructed by
making use of the well-known cyclometalation reaction.
Their geometries are modified in a straightforward manner
by addition of either mono- or bidentate ligands, such as bi-
pyridine, phenanthroline, 1,2-bis(diphenylphosphanyl)eth-
ane, dimethylphenylphosphane, triphenylphosphane, and
1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane (PTA) ligands,
to cationic cycloruthenated centers. The antitumor properties
of the compounds thus obtained are investigated in order to
compare them with recently reported ruthenium complexes
and cisplatin. IC50 values against mammalian cells (A-172,
HCT-116, and RDM-4) are determined for the library com-
pounds and some of them, such as those derived from ortho-
ruthenated phenylpyridine and a bidentate N,N ligand, dis-
play activity of the same order of magnitude as cisplatin.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
In this respect, special attention has been paid to the ap-
plication of ruthenium compounds in chemotherapy since
such complexes are often able to display properties similar
to those of platinum and iron in that they have analogous
ligand-exchange abilities to platinum complexes and are less
toxic than platinum, presumably because they can mimic
iron in the course of binding to serum transferrin or albu-
min. Since rapidly dividing cells have a greater requirement
for iron, these cells will increase the number of transferrin
receptors at their surface, thereby sequestering more metal-
loaded receptors. As the drug becomes targeted toward can-
cer cells, its toxicity should be reduced.[3] Some noticeable
progress has been made recently whereby several teams have
developed new organometallic or coordination complexes
based on ruthenium(III) for this purpose.[4] Recently, [ImH]+-
[Ru(Im)(Me2SO)Cl4]– (NAMI-A; Im = imidazole) and
[IndH]+[Ru(Ind)2Cl4]– (KP1019; Ind = 1H-indazole) have
successfully completed phase I clinical trials as antimeta-
static drugs[5] and have been shown to be precursors of RuII
complexes in vivo. Following on from these results a grow-
ing number of research groups have studied the biological
activities of related ruthenium(II) complexes. Thus, Sadler
et al. have found that some areneruthenium(II) complexes
exhibit interesting in vitro and in vivo anticancer activi-
ties,[6] and Dyson et al. have studied related complexes and
determined their antibiotic and antiviral activities.[7] Simi-
larly, Reedijk et al. have established the high cytotoxicity of
a series of bis(2-phenylazopyridine)ruthenium(II) com-
plexes against A2780 human ovarian carcinoma cell lines.[8]
Some of us have recently reported that organometallic
cycloruthenated compounds, such as those formed by the
direct cyclometalation (involving a CH activation reaction)
Introduction
It is now well established that metal complexes can be
considered as pharmaceuticals for therapeutic or diagnostic
use.[1] Since the fortuitous discovery of their anticancer
properties in the 1960s, platinum compounds have been rec-
ognized as very powerful anticancer drugs that have helped
to revolutionize cancer therapy.[2] Consequently, cisplatin is
currently one of the three most widely used drugs in chemo-
therapy and is highly effective in treating ovarian and testic-
ular cancers. It also contributes to the treatment of many
other cancers. Despite these facts, it presents two major dis-
advantages, namely its severe toxicity, especially nephrotox-
icity, neurotoxicity, and emetogenesis, and its limited appli-
cability to a narrow range of tumors as several tumors are
naturally resistant or have developed resistance.[2c] In order
to improve cancer therapies, new platinum and non-plati-
num containing entities such as metallocenes, titanium(IV)
and gold(I) complexes, and gallium(III) salts have been con-
sidered as alternatives to cisplatin.[1b,1c]
[a] Institut de Chimie de Strasbourg, CNRS – Université Louis
Pasteur,
4, rue Blaise Pascal, 67000 Strasbourg, France
[b] Instituto de Quimica, Universidad National Autonoma de
Mexico,
Circuito Exterior, Ciudad Universitaria, Coyoacan, 04510
Mexico DF, Mexico
[c] School of Science, University of Greenwich,
Chatham Maritime, Kent ME4 4TB, UK
[d] Laboratoire d’Altérations Génetiques des Cancers et Réponse
thérapeutique, UPRES EA-3430, IRCAD, Hôpitaux Universit-
aires,
1, place de l’Hôpital, Strasbourg, France
[e] Laboratoire de Signalizations Moléculaires et Neurodégéneresc-
ence, U692, INSERM – Université Louis Pasteur,
11, rue Human, Strasbourg, France
Eur. J. Inorg. Chem. 2007, 3055–3066
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3055

C. Sirlin, M. Pfeffer et al.
FULL PAPER
of a nitrogen-containing ligand, can interact with biological early stage of the procedure. The cycloruthenation of N,N-
molecules. Indeed, these organometallic species can be used dimethyl(phenylmethyl)amine and the 2-phenylpyridine li-
as very efficient mediators of electron transfer to or from gands to afford the starting materials 1 and 2 has been de-
oxidized or reduced active sites of redox enzymes.[9] We thus scribed elsewhere.[13]
reasoned that these molecules might have potential antican-
Complexes 3 and 4 were synthesized more than a decade
cer applications, as was shown recently for related cyclopal- ago by the same cyclometalation reaction, although a trans-
ladated or cycloplatinated complexes.[10] Given that most metalation reaction using orthomercuriated derivatives of
metal complexes that have been previously tested for antitu- N,N-dimethylbenzylamine and (R)- or (S)-N,N-dimethyl(1-
mor activity contain ligands that are only weakly bound to phenylethyl)amine, respectively, have been shown to be
the metal via a heteroatom (N, O, S) to metal coordination more efficient for obtaining these chloride derivatives.[14]
bond, it is very likely that at some point in the process these
Compounds 5–8 were obtained by adding one equivalent
ligands can dissociate from the metal in an in vivo context. of the required phosphane ligand {PMe2Ph, PPh3, or 1,3,5-
In contrast to previous work, we were interested in as- triaza-7-phosphatricyclo[3.3.1.1]decane (PTA)[15]} at room
sessing the behavior towards tumor cells of related com- temperature to solutions of 1 (to afford 5) or 2 (to afford
plexes in which a ligand is bound to a metal by strong co- 6, 7, and 8). Similar yields of product were obtained in
valent bonds such as, for instance, a C–M σ bond stabilized dichloromethane, acetonitrile, and methanol.
by intramolecular coordination of an heteroatom, a bond-
The compounds were characterized by elemental analysis
ing scheme that is typical of cyclometalated compounds. We as well as by H and 31P NMR spectroscopy. Thus, a JP,H
reasoned that these cyclometalated ligands might remain at- coupling constant was observed for the signal of the η6-
tached to the metal throughout the biological process, benzene ligand in 5, thereby providing evidence for the trans
whereas more labile ligands such as halides or solvents can geometry of the phosphane and the metalated arene. A chi-
dissociate and enable DNA binding. These stable ligands ral version of this compound has been reported previously
1
4
may additionally impart useful physical properties to the along with its crystal structure.[16] The H and 31P NMR
1
organometallic moiety (fluorescence or phosphorescence spectra of 6, 7, and 8 reveal that these compounds exist as
for instance), thus enabling the metal and ligand to be only one regioisomer. The acetonitrile ligands gave rise to
5
traced in the cells and in vivo. Other reasons for studying two signals in a 1:2 ratio and both of them show a JP,H
compounds with monoanionic bidentate ligands strongly coupling of 1.3–1.8 Hz. This clearly indicates the presence
bound to the metal center include: (i) the complexes ob- of two nonequivalent acetonitrile molecules located trans to
tained with only neutral ligands will be monocationic each other at the ruthenium atom; the third one and the
whereas the corresponding compounds having a neutral bi- phosphane ligand are in the same plane as the orthoruthen-
dentate chelate will lead to dicationic species under similar ated phenylpyridine unit. However, spectroscopic investi-
conditions, and (ii) the electronic behavior (electrophilicity gations were not helpful in determining the position of the
or nucleophilicity) of the ruthenium center will be signifi- phosphane ligand with respect to the C–N chelate in these
cantly modified as compared to that of RuII complexes con- complexes. The single-crystal X-ray diffraction analysis of
taining neutral bidentate ligands. This might shed light 7 and 8 (see Figures 1 and 2, respectively, for ORTEP draw-
upon its mechanism of interaction with cell-based macro- ings of the cations) revealed that the geometry of the com-
molecules such as DNA or proteins, for instance. Previous plexes was as predicted from the NMR spectroscopic data.
studies of the reactivity of cycloruthenated complexes
The identification of the atoms of the C–N chelate bound
towards various nucleophilic or electrophilic reagents[11] to Ru from the X-ray structure was, however, ambiguous,
have shown that, depending on the reaction conditions, the although the analysis of the lengths of the Ru–N bond of
metallacyclic unit of these compounds is rather inert and the Ru-NCMe units allowed us to establish unambiguously
that the existence of the cycloruthenated ligand is able to that the phosphorus atom is trans to the carbon atom of
stabilize the whole complex. We thus decided to evaluate the chelate in both 7 and 8. The Ru–N4 bond (2.029 Å) in
the biological activity of some of these compounds and 7 and Ru–N3 bond (2.024 Å) in 8 are typical of acetonitrile
were delighted to observe that they indeed display interest- ligands bound to Ru trans to N, as can be verified from
ing properties as antitumor agents in vitro.[12] In this paper X-ray diffraction data of related compounds (see Table 1).
we wish to disclose the synthesis and characterization to- Indeed, when the acetonitrile ligand is trans to C, the Ru–
gether with some further biological results that we have es- N distance is usually around 0.1 Å longer (see compounds
tablished in this area with cycloruthenated compounds.
2 and 9). Consequently, the phosphane ligands must be lo-
cated trans to the carbon atom in 7 and 8, and this fact is
further highlighted by the longer Ru–P bonds [2.458(1) and
2.395(1) Å, respectively] than in any other RuII complex
studied in this paper [average: 2.367(1) Å]. This lengthening
of the Ru–N and Ru–P bonds is expected for such bonds
Synthesis and Characterization of the
Cycloruthenated Compounds
The library of organoruthenium compounds employed trans to a carbon atom, which is known to exert a larger
in this study is shown in Scheme 1. Most starting cycloru- trans influence than a nitrogen atom.[17]
thenated materials were obtained by the well-known cyclo-
As the positions trans to the carbon and trans to the
metalation reaction involving a C–H bond activation at an nitrogen atom are identical here as far as their steric con-
3056
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 3055–3066

Cycloruthenated Compounds as Potential Anticancer Agents
FULL PAPER
Scheme 1.
straints are concerned, the location of the phosphanes is
When two equivalents of phosphane were added to 2 in
very likely to be the result of a larger trans effect[17] of the refluxing acetonitrile and the mixture stirred for 24 h a dif-
phenyl group of the C–N chelate bonded to Ru through a ferent reaction took place as we observed the formation of
C–Ru σ bond as compared to that of the pyridine unit of 9 due to the substitution of two acetonitrile ligands on the
the same chelate. This result is in line with what was found ruthenium atom by two triphenylphosphanes. The positions
recently when bidentate bipyridine or phenanthroline li- occupied by these latter ligands seemed to be symmetrical
gands were coordinated to the same compound 2 (in 12 and as only one signal is observed for the two phosphorus
11) and for which one nitrogen atom was also coordinated atoms in the 31P NMR spectrum. In contrast to the pre-
trans to the Ru–C bond. The position of the PPh3 and the vious case, the signals of the acetonitrile protons did not
PTA ligands trans to the carbon atom was, however, not display any JH,P coupling with the phosphorus atoms,
expected as there are plenty of examples in the literature of thereby suggesting that these ligands are located trans to
the coordination of phosphane ligands in similar cyclomet- the N and the C atoms of the phenylpyridine chelate,
alated compounds of palladium for which it has been respectively. The structure of 9 was ascertained by an X-
shown that the phosphane is indeed very reluctant to coor- ray diffraction study on a single crystal.[21] The results are
dinate to the palladium atom trans to the carbon atom.[20] displayed in Figure 3, which shows an ORTEP view of the
It was thus not unreasonable to expect the phosphane li- cationic part of 9. It is at once apparent that the two phos-
gands to be bound to Ru in any position except the one phane ligands are located in a mutual trans arrangement.
trans to C. Complexes 7 and 8 are likely to be formed under The Ru–P, Ru–C, and Ru–N bond lengths are in the ex-
kinetic control.
pected range for such bonds.
Eur. J. Inorg. Chem. 2007, 3055–3066
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3057

C. Sirlin, M. Pfeffer et al.
FULL PAPER
Table 1. Ru–N bond lengths of Ru–NCMe units in various cycloru-
thenated complexes.
Ru–N [Å]
Atom or ligands
trans to N
Ref.
[18]
[19]
1
2.058(2)
η6-benzene
N (PhPy)[a]
C (PhPY)
N (MeCN)
N (MeCN)
N (PhPy)
2.055(6), 2.034(4)
2.154(6), 2.162(6)
2.019(5), 2.009(4)
2.015(5), 2.009(4)
2.026(3)
2[b]
7
2.039(3)
2.009(3)
2.024(2)
2.013(2)
2.022(2)
2.026(3)
2.134(3)
N (MeCN)
N (MeCN)
N (PhPy)
N (MeCN)
N (MeCN)
N (PhPy)
this work
this work
this work
8
9
C (PhPY)
10
11
16
2.001(6)
2.001(6)
1.989(5)
2.002(4)
2.005(7)
2.045(7)
N (MeCN)
N (MeCN)
N (PhPy)
this work
[19]
N(phen)[c]
Figure 1. ORTEP view of compound 7. Selected bond distances
[Å] and angles [°]: Ru1–N2 2.009(3), Ru1–N(4) 2.026(3), Ru1–N3
2.039(3), Ru1–C7 2.049(4), Ru1–N1 2.088(3), Ru1–P1 2.458(1);
N2–Ru1–N4 87.89(11), N2–Ru1–N3 172.84(11), N4–Ru1–N3
91.65(11), N2–Ru1–C7 87.63(12), N4–Ru1–C7 92.85(13), N3–Ru1–
C7 85.26(12), N2–Ru1–N1 87.46(10), N4–Ru1–N1 171.53(11).
N [3,4-(OMe)2dmba][d]
N(bipy)[e]
[18]
18–19 2.017(5), 2.009(5)
N [(R,S)-1-PEA][f]
N(bipy)
2.001(5), 2.012(4)
this work
[a] PhPy = orthometalated 2-phenylpyridine. [b] This compound
exists in two different forms, although the molecular structure of
each is identical. [c] phen = phenanthroline. [d] dmba = ortho-
metalated N,N-dimethylbenzylamine. [e] bipy = 2,2Ј-bipyridine. [f]
PEA = orthometalated phenylethylamine.
Figure 2. ORTEP view of the cation of 8. Selected bond distances
[Å] and angles [°]: Ru1–N2 2.013(2), Ru1–N4 2.022(2), Ru1–N3
2.024(2), Ru1–C11 2.068(3), Ru1–N1 2.087(2), Ru1–P1 2.395(1);
N2–Ru1–N4 175.34(8), N2–Ru1–N3 88.70(8), N4–Ru1–N3
87.91(7), N2–Ru1–C11 90.75(8), N4–Ru1–C11 90.75(8), N3–Ru1–
C11 95.46(9), N2–Ru1–N1 89.66(7), N4–Ru1–N1 93.41(7).
Adding one equivalent of 1,2-bis(diphenylphosphanyl)-
ethane (dppe) to 2 in acetonitrile under the same conditions
as for the synthesis of 9 afforded 10 in yields of up to 60%.
Whereas the 31P NMR spectrum of 10 shows an AB-type
Figure 3. ORTEP view of the cationic part of compound 9. Se-
lected bond distances [Å] and angles [°]: Ru1–N2 2.026(3), Ru1–
N3 2.065(3), Ru1–C7 2.058(4), Ru1–N1 2.134(3), Ru1–P1
2.3765(9), Ru1–P2 2.3781(9); N2–Ru1–P1 92.94(8), N2–Ru1–N3
177.20(10), P2–Ru1–N3 88.73(7), N2–Ru1–C7 101.57(12), N1–
Ru1–C7 171.72(11), N3–Ru1–C7 79.22(12), N2–Ru1–N1 86.19(10).
pattern for two nonequivalent phosphorus atoms (3JP,P
=
8 Hz), the signal of the acetonitrile protons is a singlet with
no JH,P coupling visible. This situation clearly indicates that
the acetonitrile ligands are located trans to each other. This
was confirmed by the X-ray diffraction study of a single
crystal of 10, an ORTEP view of which is shown in Fig-
The remaining compounds of our library of cycloruthen-
ure 4. We note that the Ru–C(8) and Ru–N(1) bond lengths ated compounds (see Scheme 1) that do not contain a phos-
are elongated by 5% (approx. 0.1 Å) relative to related com- phane ligand (11–19) have either been described earlier
pounds due to the large trans influence of the phosphorus (11[19] and 13–17[18]) or were synthesized by following sim-
atoms. The distances and angles of the other parts of the ilar procedures (12 as 11 and 18, 19 as 13) Compounds
molecules are as expected.
18 and 19 are chiral versions of 13 with the two opposite
3058
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 3055–3066

Cycloruthenated Compounds as Potential Anticancer Agents
FULL PAPER
of this compound did not cause any change in its de, even
after several days at room temperature. Moreover, the com-
position of the initial mixture of diastereomers in CD3CN
did not vary with temperature in the range –40 to 55 °C.
These data allowed us to conclude that these diastereomers,
in contrast to their pseudo-tetrahedral counterpart [(η6-
C6H6)Ru{C6H4-2-(R)-CHMeNMe2}(NCMe)]PF6, are very
likely to be rigid in solution such that their epimerization,
if it takes place at all, should occur at much higher tempera-
tures. An X-ray diffraction study was performed for each of
the complexes. An ORTEP of 18, together with some typi-
cal distances and angles, is given in Figure 5; the data for
19, which are obviously identical to those of 18, have been
deposited.
Figure 4. ORTEP view of the cationic part of compound 10. Se-
lected bond distances [Å] and angles [°]: Ru1–N3 2.001(6), Ru1–
N2 2.001(6), Ru1–C8 2.096(6), Ru1–N1 2.118(5), Ru1–P1 2.343(2),
Ru1–P2 2.373(2); N3–Ru1–N2 175.9(2), P1–Ru1–P2 83.08(6), C8–
Ru1–N1 77.6(2), C8–Ru1–P1 100.42(17), N1–Ru1–P2 98.86(18).
configurations at the benzylic fragment of the cycloruthen-
ated ligand. The starting materials for the synthesis of 18
and 19, namely [(η6-C6H6)Ru{C6H4-2-(R)-CHMeNMe2}-
(NCMe)]PF6 and [(η6-C6H6)Ru{C6H4-2-(S)-CHMeNMe2}-
(NCMe)]PF6 respectively, were obtained by a direct cyclo-
metalation of the corresponding (R)- or (S)-1,1-N,N-di-
methyl(phenylethyl)amine, respectively.[12] Due to the chi-
rality at the ruthenium center these two complexes were a
mixture of diastereomers with a diastereomeric excess (de)
of 48% in favor of the (RC,SRu) and (SC,RRu) isomers
rather than the (RC,RRu) and (SC,SRu) isomers, respectively.
We will only describe the synthesis and characterization
of 18 from [(η6-C6H6)Ru{C6H4-2-(R)-CHMeNMe2}-
(NCMe)]PF6 as 19 displays exactly the same behavior. The
η6-benzene ligand was removed by adding one equivalent
of bipyridine in acetonitrile and stirring at room tempera-
Figure 5. ORTEP view of the cationic part of compound 18. Se-
lected bond distances [Å] and angles [°]: Ru1–N3 2.175(5), Ru1–
N4 2.018(5), Ru1–N5 2.001(5), Ru1–C13 2.027(6), Ru1–N1
2.057(5), Ru1–N2 2.148(5); N2–Ru1–N1 78.16(19), N2–Ru1–N3
96.69(18), N3–Ru1–C13 78.4(2), N4–Ru1–N5 84.5(2), N4–Ru1–N3
172.4(2).
1
It is immediately clear that the structure is similar to that
of the N,N-dimethylbenzylamine derivatives as one of the
bipyridine nitrogen atoms is linked to Ru trans to the C
atom of the C–N chelate. The configuration at the ruthe-
nium metal is such that the major diastereomer has a
lambda (Λ)[22] configuration at the ruthenium atom associ-
ated with an R configuration at the carbon atom, whereas
a delta (∆) configuration was found for the other derivative
which has an S configuration at the benzylic carbon atom.
In order to verify that the crystal selected for the X-ray
diffraction study was indeed the major isomer, we recorded
its 1H NMR spectra. The 2D NOESY spectrum shows that
there is an interaction between the benzylic proton and the
ture for 12 h, which led to good yields of 18. The H NMR
spectra indicated that this complex also exists as dia-
stereomers, with the de having dropped to 22%. Thus, 18 is
a mixture of two diastereomers (18maj and 18min).
Crystallization from CH2Cl2/Et2O afforded crystals that ortho H [on C(6)] of the bipyridine [they are around 2.53 Å
consisted mainly of the major isomer 18maj, with a de value apart; see HC(19)···HC(6) in Figure 5]. This cannot be the
of 98.5% (note that the mother liquor from which these case in the minor isomer as the C–N chelate must have ro-
crystals were obtained now has a de = 0% as a result of the tated by 180° around the Ru–C bond, which places the ben-
removal of part of the major diastereomer). Redissolution zylic proton away from the bipyridine one.
Eur. J. Inorg. Chem. 2007, 3055–3066
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3059

C. Sirlin, M. Pfeffer et al.
FULL PAPER
prone to oxidation as we observed a UV/Vis spectrum that
is very similar to that described in MeOH. We also checked
that a solution used for in vitro tests irradiated as above
had a lower antitumor activity than the nonirradiated one
(see Table 2 below).
Solubility and Stability of the Complexes
Solubilities
Except for 8, most of the cycloruthenated compounds
are poorly soluble in pure water. For instance, complex 11
has a solubility in water of only 0.1 m. However, all of
them are sufficiently soluble in DMSO, MeCN, or alcohols
(methanol, ethanol). The samples for in vitro tests were ob-
tained from 50.0 m solutions of the ruthenium complexes
in pure DMSO or MeCN, which were then sequentially di-
luted with the required amount of cell culture media in or-
der to obtain the solutions to be studied, whose concentra-
tion ranged from 0.2 to 50 µ. These solutions were assimi-
lated to water solutions of our compounds. In marked con-
trast to the other compounds, 8 has a sufficiently high solu-
bility in water (3.2 m) that its characterization by 1H
NMR spectroscopy in D2O proved possible. This latter
study showed that the MeCN ligands are not displaced by
water since a 5JP,H coupling was observed for the signals of
the acetonitrile ligands.
As compound 11 gave some of the best results with re-
spect to in vitro activity, we studied the behavior of solu-
tions of this complex in water. Because of its insufficient
aqueous solubility we could not follow the behavior of such
solutions by NMR spectroscopy. Nevertheless, we found
that the MeCN ligands were not displaced by CD3OD or
[D6]acetone.
Biological Effects
In Vitro Cell Growth Inhibition
To evaluate the antitumor potential of the various ruthe-
nium-derived compounds we analyzed their effect on cell
proliferation. A-172 cells derived from a human glioblas-
toma were treated with different doses of complexes from
our library or with cisplatin and the cells’ viability was de-
termined by measuring a specific cellular enzymatic activity
of the remaining living cells after 48 h. We next studied the
effect of these compounds on other cell lines derived from
adenocarcinoma (HCT-116) and lymphoma (RDM-4) and
evaluated their IC50 values. The results obtained with the
three different cell lines are summarized in Table 2.
Table 2. IC50 [µ] values of the cycloruthenated complexes com-
pared with those of reference compounds (cisplatin, 20, and 21) for
three cell lines.
Ref.
Names
used
Tumor cell line
A-172
HCT-116 RDM-4
in ref.[12]
[12]
[12]
[12]
cisplatin 3.9Ϯ0.2 3Ϯ2
3Ϯ2
Ͼ 50
Ͼ 50
Ͼ 50
30Ϯ10
10Ϯ5
–
2
Ͼ50
Ͼ50
Ͼ50
4.8Ϯ0.2 3Ϯ2
1.7Ϯ03
Ͼ50
Ͼ50
Ͼ50
Ͼ50
RDC8
RDC3
Stabilities
3
We initially verified that the UV/Vis spectrum of a 10–4

4
[12]
[12]
5
RDC6
RDC9
solution of 11 in pure CH3CN did not change with time
(after 48 h). This somewhat predictable result proved that
the C–N chelate is not labile in this medium. The same
spectrum and the same stability over a 24 h period were
observed for 10–4  solutions of 11 obtained from either a
CH3CN solution or a DMSO solution of this compound
to which water, buffered at pH 7, was added. This result
highlighted the fact that the behavior of solutions of 11 in
water is not dependent upon the solvent used to obtain
their mother solutions. No changes were observed in the
UV/Vis spectra when the pH of the solution was increased
to 9 or when sodium chloride (150 m) was added to the
water solution at pH 7. Moreover, no significant changes
were observed for any of these solutions after 48 h in the
dark.
6
3Ϯ2
Ͼ50
5Ϯ2
7Ϯ2
8
this work
this work
9
15Ϯ2
3Ϯ2
–
–
10
11[b]
12
13
14
15
16
17
18
19
20
21
this work
[12]
1.9Ϯ0.2 3Ϯ2
10Ϯ5
10Ϯ5
10Ϯ5
–
RDC11
3Ϯ2
12Ϯ2
6Ϯ2
9Ϯ2
Ͼ50
15Ϯ2
Ͼ50
Ͼ50
5Ϯ2
Ͼ50
3Ϯ2
10Ϯ5
6Ϯ2
3Ϯ2
Ͼ50
15Ϯ5
20Ϯ5
20Ϯ5
20Ϯ5
Ͼ50
this work[a] RDC12
this work
this work
–
this work
30Ϯ10 this work
30Ϯ10 this work
–
–
–
–
this work
this work
this work
this work
[a] The IC50 values reported for this compound in ref.[12] are errone-
ous. [b] The IC50 values obtained for solutions of 11 irradiated with
visible light are about twice as high as those for solutions protected
from light (the data given here are for solutions of 11 protected
from light).
We also found that the stability of aqueous solutions of
11 changed dramatically when they were irradiated with an
halogen lamp (approx. 150 W) for up to 20 minutes as we
observed the disappearance of the absorptions at 400 and
450 nm together with the appearance of a new absorption
We first tested the complexes 3–5 that are structurally
at around 366 nm. Ryabov et al. have recently reported a related to 20,[6] which is known to inhibit the growth of A-
related study of the behavior of a methanol solution of 11, 2780 human ovarian cancer cells. The IC50 values for 3 and
which they found to be photosensitive,[19] and they sug- 4 show that these species are inactive, whereas 5 displayed
gested that the MeCN ligand is substituted by methanol a good activity. According to the above classification, 20
upon irradiation. It is very likely that our water solution of displays a moderate IC50 value of 5–20 µ with our cell
11 behaves similarly to give an aquo species that is more lines.
3060
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 3055–3066

Cycloruthenated Compounds as Potential Anticancer Agents
FULL PAPER
In order to mimic the anticancer behavior of cisplatin we imate IC50 value) led to a marked increase in the number
then turned our attention to ruthenium-containing species of cells in the G0/G1 phase. On the other hand, treatment
that have two mutually cis and potentially aquatable coordi- with 45 µ of 6 or 11 (total inhibiting concentration) led
nation sites for binding to DNA. We thus selected alterna- to the formation of hypodiploid particles that created an
tive cycloruthenated compounds containing from two to important sub-G1 phase. The number of cells accumulated
four MeCN ligands on Ru. Complex 2, which has four ace- in the sub-G1 phase was lower than 10% at 24 h, but ex-
tonitrile ligands bound to the ruthenium atom, was found ceeded 50% at 48 h and reached 60% at 72 h. These results
to be inactive. However, substitution of these MeCN li- show that cycloruthenated compounds are able to both in-
gands with either 2,2Ј-bipyridine (bipy), phenanthroline duce a G1 cell-cycle arrest and DNA fragmentation, which
(phen), or phosphane or diphosphane ligands significantly suggests induction of apoptosis.
improved its activity. This increased activity for the bipy or
To further analyze the characteristics of cell death in-
phen derivatives might be related to the fact that ruthenium duced by cycloruthenated compounds we performed immu-
complexes containing bipyridine or phenanthroline are nocytochemistry experiments to assess the status of the nu-
known to be cytotoxic due to their potential for intercal- cleus and the possible activation of caspase 3, a protease
ation.[1g] Compounds 5–15 thus displayed good to medium induced by, and involved in, the apoptotic process. A-172
activities, with the exception of 8, which displayed a disap- cells were treated with cisplatin, 6, or 11 for 24 or 48 h.
pointingly low activity with two cell lines (see Table 2). This After fixation, the cells were labeled with an antibody that
low activity is not correlated to the instability of the com- recognizes the active fragment of caspase 3. The nuclei were
plex towards oxygen as we verified that only 15% of the stained with Hoechst colorant. The cells treated with cispla-
compound is oxidized in water solution after 48 h [1H and tin or cycloruthenated compounds showed nuclei alter-
31P NMR spectra indicated the presence of O=P(CH2-N- ations, such as fragmentation or condensation. Moreover,
CH2)3[23]]. Note, however, that the two MeCN ligands do we observed that cycloruthenated compounds stimulate
not have to be mutually cis, as 9 and 10, which have two Bax protein levels, which suggests that these compounds
MeCN ligands that are cis and trans to each other, respec- induce apoptosis by a mitochondrial pathway.[12] However,
tively, displayed similarly good activities
it is possible that other apoptotic pathways involving Fas
As this stage, we checked whether the presence of a Ru– and/or caspase 8 might still operate.
C bond has a significant effect on the activity of such com-
pounds. Indeed, we found that 21,[24] which is structurally
Conclusions
closely related to 11, displayed no activity. We note, how-
ever, that 21 has a double positive charge on the Ru atom
whereas 11 has only a single charge. The increased activity
on going from 21 to 11 might thus be due to the presence
of the Ru–C bond but also to a larger lipophilicity for 11
as compared to 21.
A series of cycloruthenated complexes, some of which
have been synthesized and characterized for the first time
here, have been tested for their biological activities. Some
of these compounds induce cytostatic and cytotoxic effects
on mammalian tumor cells at least as effectively as cisplatin.
Taking into account all of the results we have not yet been
able to determine the nature of the particular pharmaco-
phore. Cellular studies have shown that several ruthenium-
derived compounds lead to G1 arrest and induce apoptosis
in various tumor lines of glioblastoma, neuroblastoma, and
lymphoma. Several investigations are currently underway to
improve the potential use of these substances as a cancer
treatment. These include the determination of their in vivo
toxicity and efficiency and the evaluation of the DNA-drug
interaction at the molecular level.
The putative target of the cycloruthenated compounds is
DNA, which is a chiral molecule. As a consequence we pre-
pared and tested the two enantiomeric complexes 18 and
19, which both consist of a mixture of diastereomers with
a de of 98.5 (see above), in order to observe potential chiral
recognition. Unfortunately, these two complexes were found
to be only poorly active and no conclusion could be drawn.
We were also interested in examining the effect of the
electronic properties of the metalated phenyl unit upon the
activity. The trifluoromethyl-substituted compound 14 and
the dimethyl-substituted complex 15 were found to be as
efficient as 13, whereas the dimethoxy and nitro derivatives
(16 and 17, respectively) are less active. These data did not
allow us to correlate the activity of the organoruthenium
compounds to the electronic nature of the cyclometalated
ligand.
Experimental Section
Chemicals: All reactions were performed in Schlenk tubes under
argon. Further workup, chromatography over standardized alu-
mina, and crystallization were also performed under argon. We
note, however, that our compounds are stable in air as solids once
isolated, whereas solutions are stable for up to 48 h. Solvents were
dried and distilled under argon prior to use: diethyl ether and n-
hexane over sodium/benzophenone, dichloromethane and acetoni-
trile over calcium hydride. Mass spectra were recorded with a JEOL
JMS-SX102A instrument with m-nitrobenzyl alcohol as the matrix.
1H and 13C NMR spectra were recorded with FT-Bruker AC 300
Induction of G1 Arrest and Apoptosis
To understand the effect of our compounds on cell
growth we examined their effect on the cell cycle by FACS
(Fluorescence Activated Cell Sorter) analysis. Treatment of
RDM-4 cell lines with 15 µ of 6 or 11 (near to the approx- and ARX 500 spectrometers operating at 300.13 and 500.14 MHz
Eur. J. Inorg. Chem. 2007, 3055–3066
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3061

C. Sirlin, M. Pfeffer et al.
FULL PAPER
for H and 75.47 and 125.77 MHz for 13C, respectively. 2D COSY
1
1
and H/13C HSQC sequences were used to help the assignments of
1
the H and 13C spectra. The chemical shifts are referenced to the
residual solvent peaks; chemical shifts (δ) and coupling constants
(J) are expressed in ppm and Hz, respectively. Elemental analyses
were performed by the Service de Microanalyse de lЈInstitut de Chi-
mie, Strasbourg (France); the presence of crystallization solvent in
1
some compounds was ascertained by H NMR spectroscopy.
[Ru(η6-C6H6){2-(CH2NMe2-κN)-C6H4-κC1}(PMe2Ph)]PF6 (5): A
yellow solution of complex 1 (0.08 g, 0.156 mmol) and PMe2Ph
(0.023 mL, 0.162 mmol) was stirred in CH2Cl2 (10 mL) for 3 h at
room temperature. The reaction mixture was concentrated in vacuo
and washed with n-hexane. The yellow residue was dissolved in a
minimum amount of CH2Cl2 (1 mL) and a yellow solid was precipi-
tated by the addition of n-hexane (0.09 g, 95% yield).
C23H23F6NP2Ru·1/2CH2Cl2 (632.91): calcd. C 44.13, H 4.69, N
trans-[Ru(C6H4-2-C5H4N)(PPh3)(NCMe)3]PF6 (7): Complex
2
(0.200 g, 0.35 mmol) and PPh3 (0.093 g, 0.35 mmol) were dissolved
in 30 mL of acetonitrile and stirred at room temperature for 72 h.
The solvent was then removed in vacuo and the product was puri-
fied by column chromatography over Al2O3 using dichloromethane
as eluent. A yellow fraction was collected and the solvent was re-
moved in vacuo. The solid was dissolved in a mixture of CH3CN/
CH2Cl2 (1:1) and slow diffusion of Et2O into this solution afforded
7 as yellow crystals, which were filtered off, washed three times with
Et2O, and dried in vacuo. Yield: 0.172 g (62%). C35H32F6N4P2Ru
(785.66): calcd. C 53.51, H 4.11, N 7.13; found C 53.62, H 4.33, N
1
2.19; found C 44.30, H 4.59, N 2.16. H NMR (CD3CN): δ = 7.75
3
3
4
(dt, J = 7.5 Hz, 1 H, H6), 7.39 (tdd, J = 7.5, J = 1.7 Hz, 1 H,
Hp), 7.22 (td, 3J = 8.0, 4J = 2.0 Hz, 2 H, Hm), 7.08 (tdd, 3J =
7.1 Hz, 1 H, H4 or H5), 7.00–6.88 (m, 3 H, Ho and H4 or H5), 6.72
(d, 3J = 7.5 Hz, 1 H, H3), 5.74 (d, 3JH,P = 1.1 Hz, 6 H, C6H6), 2.85
2
4
and 2.43 (AB, J = 14.5 Hz, 2 H, CH2), 2.77 (d, JH,P = 1.1 Hz, 3
3
7.11. 1H NMR (CD3CN): δ = 8.50 (d, J = 5.8 Hz, 1 H, H12), 8.10
H, NMe), 2.66 (s, 3 H, NMe), 1.99 (d, 2J = 9.3 Hz, 3 H, PMe),
3
4
5
3
(ddd, J = 7.4, J = 4.6, J = 1.2 Hz, 1 H, H5), 7.94 (d, J = 8 Hz,
1.50 (d, J = 9.7 Hz, 3 H, PMe) ppm. 31P{1H} NMR (CD3CN): δ
2
1 H, H9), 7.85 (dd, J = 7.7, J = 1.3 Hz, 1 H, H8), 7.70 (ddd, J
3
4
3
1
= 6.37 (s, 1 P, PMe2Ph), –142.97 (sept, JP,F = 711 Hz, 1 P, PF6)
4
5
= 8.1, J = 7.4, J = 1.4 Hz, 1 H, H10), 7.64–7.59 (m, 6 H, PPh3),
ppm.
7.49–7.44 (m, 9 H, PPh3), 7.24 (tt, J = 7.4, J = 1.4 Hz, 1 H, H6
3
4
or H7), 7.10 (td, J = 7.4, J = 1.4 Hz, 1 H, H6 or H7), 6.76 (ddd,
3
4
3J = 7.4, J = 5.8, J = 1.4 Hz, 1 H, H11), 2.00 (s, JH,P = 1.3 Hz,
4
5
5
3 H, CH3CN), 1.82 (s, JH,P = 1.3 Hz, 6 H, CH3CN) ppm. 31P
5
1
NMR (CD3CN): δ = 27.35 (s, PPh3), –144.03 (sept, J = 706 Hz,
PF6) ppm. MS (FAB+): m/z (%) 641 (13) [M + H]+, 600 (3) [M +
H – MeCN]+, 559 (12) [M + H – 2MeCN]+, 518 (68) [M + H –
3MeCN]+. 379 (100) [M + H – PPh3]+, 338 (60) [M + H – PPh3 –
MeCN]+, 297 (65) [M + H – PPh3 – 2MeCN]+, 256 (28) [M + H –
PPh3 – 3MeCN]+.
trans-[Ru(C6H4-2-C5H4N)(PTA)(NCMe)3]PF6 (8): PTA was syn-
thesized as described in the literature.[23] One equivalent of PTA
(0.35 g, 2.2 mmol) was added to an orange solution of 2 (1.26 g,
2.16 mmol) in MeOH (300 mL). The yellow solution thus obtained
was stirred at room temperature for about 4 h and then dried in
vacuo. The resulting yellow powder was dissolved in CH2Cl2. This
solution was washed two times with half its volume of water and
dried with MgSO4. Caution: this procedure is aimed at removing
the impurities that are soluble in water; however, this should be
done rapidly in order to avoid oxidation of the product and to limit
its dissolution in water. The CH2Cl2 was then removed in vacuo
[Ru(C6H4-2-C5H4N)(PMe2Ph)(NCMe)3]PF6 (6): Dimethylphen-
ylphosphane (0.031 mL, 0.22 mmol) was added to a solution of
compound 2 (0.124 g, 0.22 mmol) in MeCN (5 mL) and the solu-
tion was stirred at room temp. for 18 h. The yellowish-green solu-
tion was then filtered through Al2O3 with MeCN as eluent. A yel-
low fraction was collected and concentrated in vacuo. The powder
thus formed was dissolved in a minimum amount of MeCN/Et2O.
After addition of n-hexane 6 was obtained as a yellow powder
(0.122 g, 84% yield). C25H28F6N4P2Ru (662.07): calcd. C 45.39, H
1
4.27, N 8.47; found C 45.35, H 4.49, N 8.33. H NMR (CD3CN): and the powder thus formed dissolved in CH3CN to which Et2O
δ = 8.39 (ddd, 3J = 5.8, 4J = 1.6, 5J = 0.8 Hz, 1 H, H12), 8.08
was added. This afforded yellow microcrystals of 8 (80 mg, 22%
3
4
4
5
(dddd, J = 7.1, JH,P = 4.7, J = 1.3, J = 0.5 Hz, 1 H, H5), 7.89
yield). C23H29F6N7P2Ru (680.53): calcd. C 40.59, H 4.30, N 14.41;
3
3
1
(d, J = 8.1 Hz, 1 H, H9), 7.80 (d, J = 7.8 Hz, 1 H, H8), 7.71–7.63
found C 40.68, H 4.38, N 14.22. H NMR (D2O, 400 MHz): δ =
3
3
3
4
(m, 3 H, H10 and Ho), 7.53–7.49 (m, 2 H, Hm), 7.44 (t, 3J = 7.4 Hz,
8.55 (d, J = 5.6 Hz, 1 H, H12), 8.17 (ddd, J = 7, JH,P = 4.9, J
1 H, Hp), 7.20 (td, J = 7.3, J = 1.4 Hz, 1 H, H6), 7.06 (td, J = = 1 Hz, 1 H, H5), 8.03 (d, J = 8 Hz, 1 H, H9), 7.98 (d, J = 8 Hz,
3
4
3
3
3
7.5, 4J = 1.4 Hz, 1 H, H7), 6.92 (ddd, 3J = 7.3, 3J = 5.8, 4J =
1 H, H8), 7.85 (td, J = 8, J = 1.2 Hz, 1 H, H10), 7.41 (dd, J ≈
3
4
3
5
1
3
4
1.5 Hz, 1 H, H11), 2.33 (d, JH,P = 1.7 Hz, 3 H, NCMe), 1.97 (d,
7.3, J = 1 Hz, 1 H, H6), 7.26 (td, J = 7.3, J = 1 Hz, 1 H, H7),
5JH,P = 1.8 Hz, 6 H, 2 NCMe), 1.86 (d, 2JH,P = 5.8 Hz, 6 H, PMe2) 7.19 (ddd, J = 7.2, J = 5.6, J = 1.2 Hz, 1 H, H11), 4.60 (s, 6 H,
3
3
4
ppm. 13C{1H} NMR (CD3CN): δ = 185.5 (C4), 170.2 (C2), 156.6 NCH2N), 4.42 (d, JH,P = 3.2 Hz, 6 H, PCH2), 2.6 (d, JH,P
=
2
5
5
(C12), 147.1 (C3), 138.0 (C5), 137.5 (C10), 131.0 (Co), 129.6 and
129.7 (Cm), 129.3 and 129.2 (C6 and Cp), 124.4 (C8 and Cipso), 123.6
(NCMe), 122.9 (C7), 122.3 (C11), 119.4 (C9), 13.55 and 13.41
(PMe2), 4.31 and 3.99 (NCMe) ppm. 31P{1H} NMR (CD3CN): δ
1.6 Hz, 3 H, NCCH3 equatorial), 2.03 (d, JH,P = 1.6 Hz, 6 H,
NCCH3 apical) ppm. 13C{1H} NMR: δ = 157.5 (C12), 137.5 (C10),
137.2 (C5), 130 (C6), 125 (C8), 124.4 (C7), 123.3 (C11), 119.8 (C9),
72 (d, JC,P = 7 Hz, NCH2N), 49 (s, JC,P = 6 Hz, PCH2), 3.8
(NCCH3), 3.4 (NCCH3) ppm. 31P{1H} NMR (CD3CN, 300 MHz):
δ = –68.4 (s, 1 P, PTA), –143.4 (sept, 1JP,F = 706 Hz, 1 P, PF6) ppm.
1
= –7.08 (s, 1 P, PMe2), –144.40 (sept, JP,F = 704.6 Hz, 1 P, PF6)
ppm.
3062
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 3055–3066

Cycloruthenated Compounds as Potential Anticancer Agents
FULL PAPER
trans-[Ru(C6H4-2-C5H4N)(PPh3)2(NCMe)2]PF6 (9): Complex
2
ppm. 31P NMR (CD3CN): δ = 68.41 (s, PPh2), 43.64 (s, PPh2),
(0.200 g, 0.35 mmol) and PPh3 (0.186 g, 0.71 mmol) were dissolved –143.3 (sept, JP,F = 706 Hz, PF6) ppm. MS (FAB+): m/z (%) 736
in 30 mL of acetonitrile and refluxed for 24 h. The solvent was
then removed in vacuo and the product was purified by column
chromatography over Al2O3 using dichloromethane as eluent. A
yellow fraction was collected and the solvent was removed in vacuo.
The solid was dissolved in a mixture of CH3CN/CH2Cl2 (1:1) and
slow diffusion of Et2O into this mixture afforded 9 as yellow crys-
tals, which were filtered off, washed three times with Et2O, and
dried in vacuo. Yield: 0.190 g (54%). C31H44F6N3P3Ru (766.68):
[M]+, 654 (100) [M – 2MeCN]+.
1
calcd. C 61.15, H 4.40, N 4.17; found C 61.15, H 4.59, N 4.47. H
NMR (CD3CN): δ = 7.95 (dd, 3J = 5.9, 4J = 0.7 Hz, 1 H, H12),
3
7.26 (d, J = 7.4 Hz, 1 H, H5), 7.4–7.0 (m, 33 H, PPh3, H8, H9 and
3
3
4
H10), 6.75 (td, J = 7.4 Hz, 1 H, H6 or H7), 6.67 (td, J = 7.4, J
= 1.5 Hz, 1 H, H7 or H6), 6.47 (td, 3J = 5.8, 4J = 1.4 Hz, 1 H, H11),
2.14 (s, 3 H, CH3CN), 1.96 (s, 3 H, CH3CN) ppm. 13C{1H} NMR
(CD3CN): δ = 180.9 (C4), 167.1 (C2 or C3), 152.5 (C12), 146.5 (C3
or C2), 141.4 (C5), 135.4 (C10 or C8), 134.3 (t, JC,P = 5.2 Hz, Co),
132.7 (JC,P = 19 Hz, Cipso), 130.2 (s, Cp), 128.8 (t, JC,P = 8 Hz, Cm),
128.7 (C6), 123.7 (C8 or C10), 122.15 (C11), 120.9 (C7), 119.09 (C9),
30.9 (MeNC) ppm. 31P NMR (CD3CN): δ = 35.3 (s, PPh3), –143.4
[Ru(C6H4-2-C5H4N)(4,4Ј-diMe-2,2Ј-bipy)(NCMe)2]PF6 (12): 4,4Ј-
Dimethyl-2,2Ј-bipyridine (4,4Ј-diMe-2,2Ј-bipy; 0.033 g, 0.181
mmol) was added to a solution of 2 (0.102 g, 0.181 mmol) in
CH2Cl2 (13 mL) and the solution was stirred at room temp. for
1
2 days. The course of the reaction was followed by H NMR spec-
troscopy. The solvent was then removed in vacuo and the product
was dissolved in a minimum of MeCN/Et2O. Hexane (30 mL) was
added and the mixture was left to stand for 3 days to give a dark-
brown powder (yield: 0.116 g, 96%). C27H26F6N5PRu (666.56):
calcd. C 48.65, H 3.93, N 10.51; found C 48.24, H 4.07, N 10.60.
1
(sept, JP,F = 710 Hz, PF6) ppm. MS (FAB+): m/z (%) 821 (20)
[M – MeCN]+, 780 (95) [M – 2MeCN]+, 518 (100) [M – 2MeCN –
PPh3]+, 256 (9) [M – 2MeCN – 2PPh3]+.
3
1H NMR (CD3CN): δ = 9.18 (d, J = 5.5 Hz, 1 H, H6Ј), 8.30 (s, 1
3
4
5
H, H3Ј), 8.21 (ddd, J = 7.4, J = 1.3, J = 0.5 Hz, 1 H, H12), 8.09
(s, 1 H, H3ЈЈ), 7.85–7.81 (m, 2 H, H5 and H9), 7.68–7.66 (m, 2 H,
H5Ј and H6ЈЈ), 7.52 (ddd, 3J = 8.2, 3J = 7.4, 4J = 1.6 Hz, 1 H, H10),
3
4
5
3
7.44 (ddd, J = 5.7, J = 1.6, J = 0.8 Hz, 1 H, H8), 7.24 (td, J =
7.3, 4J = 1.3 Hz, 1 H, H11), 7.05 (ddd, 3J = 7.7, 3J = 7.2, 4J =
1.3 Hz, 1 H, H6), 6.85 (dd, J = 5.9, J = 1.8 Hz, 1 H, H5ЈЈ), 6.73
(ddd, 3J = 7.3, 3J = 5.7, 4J = 1.5 Hz, 1 H, H7), 2.64 and 2.34 (2ϫs,
6 H, CH3), 2.21 and 2.19 (2ϫs, 6 H, NCMe) ppm. 13C{1H} NMR
(CD3CN): δ = 300 MHz: 193.5 (C4), 169.5 (C2), 159.95, 155.9,
154.5 (C5Ј or C6ЈЈ), 151.5 (C8), 150.4 (C6Ј), 149.5, 148.1, 146.7,
138.8 (C12), 136.2 (C10), 129.0 (C11), 128.5 (C6ЈЈ or C5Ј), 126.9
(C5ЈЈ), 124.6–124.3 (C3Ј + C3ЈЈ + C9 or C5), 121.8 (C7), 121.2 (C6),
118.6 (C5 or C9), 21.27 (Me), 20.75 (Me), 4.2 (MeCN), 3.9 (MeCN)
ppm.
3
4
cis-[Ru(C6H4-2-C5H4N)(dppe)(NCMe)2]PF6 (10): Complex
2
(0.200 g, 0.355 mmol) and dppe (Ph2PC2H4PPh2; 0.141 g,
0.355 mmol) were dissolved in 30 mL of methanol to yield a yellow
solution. This solution was refluxed for 19 h. The solvent was then
evaporated in vacuo and the product filtered through Al2O3 using
dichloromethane as eluent. A yellow fraction was collected and
concentrated. Diethyl ether was then added to precipitate the prod-
uct as an amorphous yellow solid. This solid was filtered off,
washed three times with diethyl ether, and dried in vacuo. Yellow
crystals were obtained by slow diffusion of Et2O into a concen-
trated solution of 10 in CH2Cl2/CH3CN (1:1) at room temperature.
Yield: 0.143 g (46%). C41H38F6N3P3Ru·CH2Cl2 (955.67): calcd. C
52.17, H 4.40, N 4.55; found C 52.31, H 4.55, N 4.71. 1H NMR
3
(CD3CN): δ = 8.07 (d, J = 8.2 Hz, 1 H, H12), 7.95–7.85 (m, 3 H,
H8, H5 and H9), 7.65–7.35 (m, 21 H, PPh2 + H10), 7.07–7.01 (m, 2
H, H6 and H7), 6.88 (ddd, 3J = 7.3, 3J = 6.0, 4J = 1.3 Hz, 1 H, [Ru(2,2Ј-bipyridine){(R)-2-(CHMeNMe2-κN)-C6H4-κC1}-
5
H11), 2.69 (m, 4 H, CH2), 1.50 (d, JH,P = 1.1 Hz, 6 H, CH3CN) (NCMe)2 ]PF6 (18): A solution of [(η6 -C6 H6 )Ru{(R)-2-
ppm. 13C{1H} NMR (CD3CN): δ = 167.9 (d, JC,P = 6 Hz, C4),
(CHMeNMe2)C6H4}(NCMe)]PF6 (0.100 g, 0.19 mmol) and 2,2Ј-
bipyridine (0.030 g, 0.19 mmol) in 15 mL of acetonitrile was stirred
156.4 (d, JC,P = 7.4 Hz, C12), 148.4 (s, C2 or C3), 144.7 (d, JC,P
=
4.7 Hz, C9), 138.9 (s, C8 or C6),134.8 (d, JC,P = 26 Hz, Cipso), 134.2 at room temp. for 12 h. The resulting deep purple solution was
(d, JC,P = 40 Hz, Cipso), 133.7 (d, JC,P = 9 Hz, Co), 133.5 (d, JC,P
= 10 Hz, Co), 131.1 (d, JC,P = 6.7 Hz, Cp), 129.9 (d, JC,P = 8 Hz,
evaporated to dryness and the residue (whose de was 22%) was
purified by column chromatography over Al2O3 using dichloro-
Cm), 129.4 (d, JC,P = 9 Hz, Cm), 129.1 (d, JC,P = 5.7 Hz, C10), 125.5 methane as an eluent. The purple band was collected and the sol-
(s, C6 or C8), 125.1 (d, JC,P = 3 Hz, C3 or C2), 123.5 (s, C7 or C11),
123.2 (s, C11 or C7), 120.2 (s, C5), 32.6 (dd, JC,P = 30, JC,P = 18 Hz, concentrated solution of the purple solid in CH2Cl2/MeCN (1:1)
CH2), 29.2 (dd, JC,P = 25, JC,P = 10 Hz, CH2), 4.02 (s, MeNC) gave dark purple crystals of 18 with a de of 98.5%. The de of
vents evaporated to dryness. Slow diffusion of diethyl ether into a
Eur. J. Inorg. Chem. 2007, 3055–3066
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3063

C. Sirlin, M. Pfeffer et al.
FULL PAPER
18maj vs. 18min was determined by comparing the intensity of the
benzylic proton of the minor isomer with the intensity of the satel-
lite signal of the same proton coupled to 13C of the major isomer.
Yield: 0.064 g (53%). C24H28F6N5PRu (632.55): calcd. C 45.57, H
4.46, N 11.07; found C 45.59, H 4.51, N 10.93. 1H NMR (CD3CN):
(CHMe), 46.5 and 45.7 (NMe2), 9.9 (CHCH3), 3.8 and 3.1
(NCCH3) ppm. MS (FAB+): m/z (%) 633 (8) [M+ + PF6 + H]+,
–
488 (24) [M]+, 447 (12) [M – MeCN]+, 406 (88) [M – 2MeCN]+.
3
4
5
18maj: δ = 9.33 (ddd, J = 5.3, J = 1.5, J = 0.7 Hz, 1 H, H6ЈЈ),
8.44 (d, 3J = 8.2 Hz, 1 H, H3ЈЈ), 8.31 (d, 3J = 7.9 Hz, 1 H, H3Ј),
3
4
5
3
8.18 (ddd, J = 5.6, J = 1.5, J = 0.6 Hz, 1 H, H6Ј), 8.08 (ddd, J
= 9.1, 3J = 7.6, 4J = 1.5 Hz, 1 H, H4ЈЈ), 7.79 (m, 2 H, H4Ј and H6),
3
3
4
3
7.73 (ddd, J = 7.5, J = 5.2, J = 1.1 Hz, 1 H, H5ЈЈ), 7.17 (td, J
= 6.5, J = 1 Hz, 1 H, H5Ј), 7.07 (m, 1 H, H5), 6.92 (m, 2 H, H3
4
and H4), 3.40 (q, 3J = 6.7 Hz, 1 H, CH), 2.41 (s, 3 H, CH3CN),
2.16 (s, 3 H, NCH3), 2.06 (s, 3 H, CH3CN), 1.49 (s, 3 H, NCH3),
1.18 (d, 3J = 6.7 Hz, 3 H, CH3) ppm; 18min: δ = 9.35 (m, 1 H,
H6ЈЈ), 8.71 (dd, H6Ј), 8.43 (d 1 H, H3ЈЈ), 8.31 (d, 1 H, H3Ј), 8.08
[Ru(2,2Ј-bipyridine){(S)-2-(CHMeNMe2-κN)-C6H4-κC1}(NCMe)2]-
PF6 (19): The synthesis and spectroscopic data were the same as
for 18. C24H28F6N5PRu (632.55): calcd. C 45.57, H 4.46, N 11.07;
found C 45.65, H 4.54, N 10.91
4
(ddd, 1 H, H4ЈЈ), 7.90 (dd, 3J = 7.2, J = 1.2 Hz, 1 H, H6), 7.79 (m,
2 H, H4Ј), 7.73 (ddd, 1 H, H5ЈЈ), 7.23 (ddd, 1 H, H5Ј) 7.17 (td, 1 H,
3
H5), 6.94 (m, 1 H, H4), 6.86 (d, 1 H, H3), 4.09 (q, J = 6.7 Hz, 1
H, CH), 2.46 (s, 3 H, CH3CN), 2.31 (s, 3 H, NCH3), 2.07 (s, 3 H, X-ray Diffraction Studies of Compounds 7–10, 18, and 19: Crystals
CH3CN), 1.27 (d, 3J = 6.7 Hz, 3 H, CH3), 1.13 (s, 3 H, NCH3) of the various compounds were obtained according to the crystalli-
ppm. 13C{1H} NMR (CD3CN): 18maj: δ = 183.6, 159.7, 155.9,
zation procedure described for each compound (Table 3). Diffrac-
153.4 (C6ЈЈ), 151.0, 150.4 (C6Ј), 137.0 (C4Ј or C6), 136.0 (C4ЈЈ), tion intensities data were collected with a SMART Apex dif-
134.9 (C4Ј or C6), 126.7 (C5Ј), 125.5 (C5ЈЈ), 124.8 (C5), 122.9 fractometer equipped with a graphite-monochromated Mo-Kα ra-
(C3Ј), 122.8 (C3 or C4), 120.4 (C3 or C4), 120.3, 118.0, 70.1
diation source and CCD area detector, at room temperature. The
Table 3. Crystal data and structure refinement.
7
8
9
Empirical formula
Formula weight
Crystal system
Space group
C36H32F6N4OP2Ru
813.67
monoclinic
P21/n
C25H32F6N8P2Ru
721.60
monoclinic
P21/c
C51H44F6N3P3Ru
1006.87
monoclinic
P21/c
a [Å]
b [Å]
c [Å]
α [°]
12.4740(8)
24.6128(15)
13.9723(8)
90
11.8140(2)
15.6510(3)
16.7910(4)
90
16.796(1)
15.938(1)
17.377(1)
90
β [°]
γ [°]
111.242(1)
90
102.598(1)
90
90.677(1)
90
Z
4
4
4
Flack’s parameter
Final R indices [I Ͼ 2σ(I)]
R1 = 0.0413
wR2 = 0.0917
R1 = 0.0610
wR2 = 0.0965
0.906
R1 = 0.0428
wR2 = 0.0978
R1 = 0.0796
wR2 = 0.1115
1.003
R1 = 0.0424
wR2 = 0.0969
R1 = 0.0564
wR2 = 0.1017
0.951
R indices (all data)
Goodness-of-fit on F2
Largest diff. peak and hole [eÅ–3]
0.627 and –0.229
0.901 and –0.767
0.644 and –0.437
10
18
19
Empirical formula
Formula weight
C42H40Cl2F6N3P3Ru
965.63
C24H28F6N5PRu
632.55
C24H28F6N5PRu
632.55
Crystal system
Space group
monoclinic
P21/n
tetragonal
P43
tetragonal
P41
a [Å]
b [Å]
c [Å]
α [°]
11.7961(12)
16.3996(16)
22.226(2)
90
8.4853(3)
8.4853(3)
37.630(3)
90
8.4835(3)
8.4835(3)
37.615(3)
90
β [°]
γ [°]
90.341(2)
90
90
90
90
90
Z
4
4
4
Flack’s parameter
Final R indices [I Ͼ 2σ(I)]
–0.03(4)
R1 = 0.0423
wR2 = 0.0803
R1 = 0.0485
wR2 = 0.0825
1.059
–0.02(5)
R1 = 0.0460
wR2 = 0.1083
R1 = 0.0508
wR2 = 0.1114
1.020
R1 = 0.0607
wR2 = 0.0905
R1 = 0.1468
wR2 = 0.1064
0.784
R indices (all data)
Goodness-of-fit on F2
Largest diff. peak and hole [eÅ–3]
0.882 and –0.484
0.710 and –0.520
0.603 and –0.533
3064
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 3055–3066

Cycloruthenated Compounds as Potential Anticancer Agents
FULL PAPER
data collected were processed to produce conventional intensity
data with the SAINT plus program. The intensity data were cor-
rected for Lorentz and polarization effects, no absorption correc-
tion was applied. The structures were solved by direct methods,
completed by a subsequent difference Fourier synthesis map, and
refined by full-matrix least-squares procedures on F2. All non-hy-
drogen atoms were refined anisotropically. The crystal structures
have solvents of crystallization, which were refined isotropically in
two or three major contributors. In the case of compound 7, the
molecule of methanol was refined in three positions with major
contributors. Hydrogen atoms could not be refined due to the high
level of disorder. The PF6 anions are in special positions and the
fluorine atoms are distorted and were modeled and refined aniso-
tropically in two major contributors. Hydrogen atom positions
were calculated and included in the final cycle of refinement. All
calculations were performed with the SHELXTL (6.10) program
package.[25] CCDC-617638 to -617642 and CCDC-629738 (for 7,
9, 10, 18, 19, and 8, respectively) contain the supplementary crys-
tallographic data for this paper. These data can be obtained free of
charge from the Cambridge Crystallographic Data Center at
http://www.ccdc.ac.uk.data/cif.
[1]
a) C. Orvig, M. J. Abrams, Chem. Rev. 1999, 99, 2201–2204; b)
M. J. Clarke, F. Zhu, D. R. Frasca, Chem. Rev. 1999, 99, 2511–
2534; c) Z. Guo, P. J. Sadler, Angew. Chem. Int. Ed. 1999, 38,
1512–1531; d) Y. K. Yan, M. Melchart, A. Habtemariam, P. J.
Sadler, Chem. Commun. 2005, 4764–4776; e) U. Schatzschne-
ider, N. Metzler-Nolte, Angew. Chem. Int. Ed. 2006, 45, 1504–
1507; f) A. R. Timerbaev, C. G. Hartinger, S. S. Aleksenko,
B. K. Keppler, Chem. Rev. 2006, 106, 2224–2248; g) W. H. Ang,
P. J. Dyson, Eur. J. Inorg. Chem. 2006, 4003–4018; h) P. J. Dy-
son, G. Sava, Dalton Trans. 2006, 1929–1933; i) D. Wang, S. J.
Lippard, Nat. Rev. Drug Discov. 2005, 4, 307–320; j) Bioorgano-
metallics, Biomolecules, Labelling, Medicine (Ed.: G. Jaouen),
John Wiley & Sons, 2005; k) Metallotherapeutic Drugs and
Metal-Based Diagnostic Agents: The Use of Metals in Medicine
(Eds.: M. Gielen, E. R. T. Tiekink), John Wiley & Sons, 2005.
a) Platinum and Other Metal Coordination Compounds in Can-
cer Chemotherapy (Eds.: H. M. Pinto, J. H. Schornagel), Ple-
num, New York, 1996; b) Cisplatin: Chemistry and Biochemis-
try of a Leading Anticancer Drug (Ed.: B. Lippert), Wiley-
VCH, Weinheim, 1999; c) E. Wong, C. M. Giandomenico,
Chem. Rev. 1999, 99, 2451–2466; d) E. R. Jamieson, S. J. Lip-
pard, Chem. Rev. 1999, 99, 2467–2498.
a) C. S. Allardyce, P. J. Dyson, Platinum Met. Rev. 2001, 45,
62–69; b) M. J. Clarke, Coord. Chem. Rev. 2003, 236, 209–233.
a) H. Depenbrock, S. Schmelcher, R. Peter, B. K. Keppler, G.
Weirich, T. Block, J. Rastetter, A. R. Hanauske, Eur. J. Cancer
1997, 33, 2404–2410; b) M. A. Jakupec, E. Reisner, A. Eich-
inger, M. Pongratz, V. B. Arion, M. Galanski, C. G. Hartinger,
B. K. Keppler, J. Med. Chem. 2005, 48, 2831–2837 and refer-
ences cited therein.
a) G. Sava, A. Bergamo, Int. J. Oncol. 2000, 17, 353; b) J. M.
Rademaker-Lakhai, D. van den Bongard, D. Pluim, J. H. Be-
ijnen, J. H. Schellens, Clin. Cancer Res. 2004, 10, 3717–3727;
c) C. G. Hartinger, S. Zorbas-Seifried, M. A. Jakupec, B. Kyn-
ast, H. Zorbas, B. K. Keppler, J. Inorg. Biochem. 2006, 100,
891–904.
a) R. E. Morris, R. E. Aird, P. del S. Murdoch, H. Chen, J.
Cummings, N. D. Hughes, S. Parsons, A. Parkin, G. Boyd, D. I.
Jodrell, P. J. Sadler, J. Med. Chem. 2001, 44, 3616–3621; b)
R. E. Aird, J. Cummings, A. A. Ritchie, M. Muir, R. E. Morris,
H. Chen, P. J. Sadler, D. I. Jodrell, Br. J. Cancer. 2002, 86,
1652–1657.
C. S. Allardyce, P. J. Dyson, D. J. Ellis, P. A. Salter, R. Scopel-
liti, J. Organomet. Chem. 2003, 668, 35–42.
A. C. G. Hotze, M. Baca, A. H. Velders, B. A. J. Jansen, H.
Kooijman, A. L. Spek, J. G. Haasnoot, J. Reedijk, J. Med.
Chem. 2003, 46, 1743–1750.
a) A. D. Ryabov, V. Sukharev, L. Alexandrova, R. Le Lagadec,
M. Pfeffer, Inorg. Chem. 2001, 40, 6529–6532; b) A. D. Ryabov,
V. Sukharev, L. Alexandrova, R. Le Lagadec, M. Pfeffer, Inorg.
Chem. 2003, 42, 6598–6600; c) A. D. Ryabov, V. S. Kurova,
E. V. Ivanova, R. Le Lagadec, L. Alexandrova, Anal. Chem.
2005, 77, 1132–1139; d) R. Le Lagadec, L. Alexandrova, H.
Estevez, M. Pfeffer, V. Laurinavicius, J. Razumiene, A. D. Rya-
bov, Eur. J. Inorg. Chem. 2006, 2735–2738.
[2]
[3]
[4]
Cell-Proliferation Assays: The ruthenium samples for in vitro tests
were obtained from 50.0 m solutions of the ruthenium complexes
in pure DMSO, which were then sequentially diluted with the re-
quired amount of cell culture media in order to obtain study solu-
tions with a concentration ranging from 0.2 to 50 µ. A-172 and
HCT-116 cells were obtained from American Type Cell Culture
Collection. RDM-4, a murine T lymphoma, was obtained from
Dr. D. Oth (Institut Armand Frappier, Laval des Rapides, Québec,
Canada). Cells were grown in 96-well plates and treated at 70%
confluence. The medium was removed after 48 h and MTT
(0.5 mgmL–1) in DMEM (Dulbecco’s Modified Eagle Medium)
was added for 1 h. The medium was removed again and 0.04%
HCl in 2-propanol was added to solubilize the crystals. Absorption
differences were quantified with an Elisa plate reader (Metertech
USA) at 490–650 nm. The experiments were repeated between two
and five times, and the mean deviation was determined by consider-
ing the extreme values found over all experiments.
[5]
[6]
Cell-Cycle Analysis and Apoptosis Assays: Hypodiploid DNA was
measured as described according to Nicoletti.[26] Briefly, 106 cells
were centrifuged and fixed in 1 mL of cold 70% ethanol at 4 °C
for one hour, washed once with PBS (Phosphate Buffered Saline)
and EDTA (2 m), and re-suspended in 1 mL of PBS containing
0.25 mg Rnase A, 2 m EDTA and 0.1 mg of propidium iodide.
Cells were analyzed after incubation at 37 °C for 30 min in the
dark. The fluorescence of 10,000 cells was analyzed using a FACS
scan flow cytometer and CellQuest software (Becton Dickinson,
San José, CA).
[7]
[8]
[9]
Acknowledgments
[10]
[11]
a) A. Gómez-Quiroga, C. Navarro-Ranninger, Coord. Chem.
Rev. 2004, 248, 119–133 and references cited therein b) G. L.
Edwards, D. S. C. Black, G. B. Deacon, L. P. G. Wakelin, Can.
J. Chem. 2005, 83, 969–979; c) C. Bincoletto, I. L. S. Tersariol,
C. R. Oliveira, S. Dreher, D. M. Fausto, M. A. Soufen, F. D.
Nascimento, A. C. F. Caires, Bioorg. Med. Chem. 2005, 13,
3047–3055 and references cited therein.
a) H. C. L. Abbenhuis, M. Pfeffer, J. P. Sutter, A. de Cian, J.
Fischer, L. H. Ji, J. H. Nelson, Organometallics 1993, 12, 4464–
4472; b) W. Ferstl, I. K. Sakodinskaya, N. Beydoun-Sutter, G.
Le Borgne, M. Pfeffer, A. D. Ryabov, Organometallics 1997, 16,
411–418; c) M. Pfeffer, J. P. Sutter, E. P. Urriolabeitia, Bull. Soc.
Chim. Fr. 1997, 134, 947–954; d) V. Ritleng, M. Pfeffer, C. Sir-
lin, Organometallics 2003, 22, 347–354.
The Université Louis Pasteur and the Ministère de l’Education Na-
tionale are thanked for a student fellowship (to L. L.), the Centre
National de la Recherche Scientifique (CNRS), Institut National
de la Santé et de la Recherche Médicale (INSERM), the Consejo
Nacional de Ciencia y Tecnología (CONACYT) (project numbers
34293-E and 40135-Q), and the Agence Nationale de la Recherche
(ANR-05-EMPB-019-01/02) are thanked for partial support of the
work. We are grateful to Ms. B. Wieczoreck and K. Küpfer for
their help in the synthesis and characterization of 11. X-ray diffrac-
tion studies were performed by Dr. Simon Hernandez and Dr.
Ruben Alfredo Toscano (UNA Mexico; compounds 7, 9, 10, 18,
and 19) and by Dr. A. de Cian (ULP Strasbourg, compound 8).
Eur. J. Inorg. Chem. 2007, 3055–3066
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3065

C. Sirlin, M. Pfeffer et al.
FULL PAPER
[12] C. Gaiddon, P. Jeannequin, P. Bischoff, M. Pfeffer, C. Sirlin,
J. P. Loeffler, J. Pharmacol. Exp. Ther. 2005, 315, 1403–1411.
[13] S. Fernandez, M. Pfeffer, V. Ritleng, C. Sirlin, Organometallics
1999, 18, 2390–2394.
[14] S. Attar, J. H. Nelson, J. N. Fischer, A. de Cian, J. P. Sutter, M.
Pfeffer, Organometallics 1995, 14, 4559–4569.
Frankland, M. C. Ramírez de Arellano, Chem. Eur. J. 1999, 5,
3066–3075 and references cited therein.
[21] The X-ray diffraction study was performed on single crystals
obtained from a sample that was synthesized in MeOH instead
of MeCN. This explains why one disordered molecule of
MeOH was found in this structure.
[15] C. Scolaro, A. Bergamo, L. Brescacin, R. Delfino, M. Cochi-
etto, G. Laurenczy, T. J. Gelbach, G. Sava, P. J. Dyson, J. Med.
Chem. 2005, 48, 4161–4171.
[16] V. Ritleng, P. Bertani, M. Pfeffer, C. Sirlin, J. Hirschinger, In-
org. Chem. 2001, 40, 5117–5122.
[17] Advanced Inorganic Chemistry, 5th ed. (Eds.: F. A. Cotton, G.
Wilkinson), Wiley-Interscience, New York, 1988, p. 1299–1300.
[18] R. Le Lagadec, L. Rubio, L. Alexandrova, R. A. Toscano, E. V.
Ivanova, R. Meskys, V. Laurinavicius, M. Pfeffer, A. D. Rya-
bov, J. Organomet. Chem. 2004, 689, 4820–4832.
[19] A. D. Ryabov, R. Le Lagadec, H. Estevez, R. A. Toscano, S.
Hernandez, L. Alexandrova, V. S. Kurova, A. Fischer, C. Sirlin,
M. Pfeffer, Inorg. Chem. 2005, 44, 1626–1634.
[22] See IUPAC, Nomenclature of Inorganic Chemistry, Recommen-
dations (Ed.: G. J. Leigh), Blackwell Scientific Publications, Ox-
ford, 1990.
[23] D. J. Aigle, Inorg. Synth. 1998, 32, 40–45.
[24] G. M. Brown, R. W. Callahan, T. J. Meyer, Inorg. Chem. 1975,
14, 1915–1921.
[25] G. M. Sheldrick, SHELXTL (Version 6.10), Bruker AXS Inc.,
Madison, Wisconsin, USA, 2000.
[26] I. Nicoletti, G. Migliorati, M. C. Pagliacci, F. Grignani, C. Ric-
cardi, J. Immunol. Methods 1991, 139, 271.
Received: December 5, 2006
Published Online: May 10, 2007
[20] a) J. Dehand, J. Jordanov, M. Pfeffer, M. Zinsius, C. R. Acad.
Sci. Paris 1975, 281, 651–654; b) J. Vicente, J. A. Abad, A. D.
3066
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 3055–3066