Rational design of highly cytotoxic η6-arene β-diketiminato - Ruthenium complexes

Article abstract of DOI:10.1021/om900991b

A series of ruthenium-benzene complexes with β-diketiminate ligands modified with electronwithdrawing groups were prepared and characterized by NMR spectroscopy, mass spectrometry, and single-crystal X-ray diffraction. The complexes are stable in air and undergo controlled hydrolysis in water. The complexes were evaluated for anticancer activity in vitro, and two of them proved to be highly cytotoxic, comparable or even superior to cisplatin. This work shows the potential utility of the βdiketiminate ligand in the rational design of new anticancer metal-containing drugs. A related complex with a η⁶-C₆H₅CF₃ ligand was prepared and found to undergo a nucleophilic addition reaction at the coordinated arene ring to afford a substituted η⁵-cyclohexadienyl derivative.

Full text of DOI:10.1021/om900991b

Organometallics 2010, 29, 417–427 417
DOI: 10.1021/om900991b
Rational Design of Highly Cytotoxic η6-Arene
β-Diketiminato-Ruthenium Complexes
Andrew D. Phillips,*,†,‡ Olivier Zava,‡ Rosario Scopelitti,‡ Alexey A. Nazarov,‡ and
Paul J. Dyson*,‡
†School of Chemistry and Chemical Biology, University College Dublin, Belfield, Dublin 4, Ireland and
ꢀ ꢀ ꢀ
Institut des Sciences et Ingenierie Chimiques, Ecole Polytechnique Federale de Lausanne (EPFL), CH-1015
‡
Lausanne, Switzerland
Received November 13, 2009
A series of ruthenium-benzene complexes with β-diketiminate ligands modified with electron-
withdrawing groups were prepared and characterized by NMR spectroscopy, mass spectrometry, and
single-crystal X-ray diffraction. The complexes are stable in air and undergo controlled hydrolysis in
water. The complexes were evaluated for anticancer activity in vitro, and two of them proved to be highly
cytotoxic, comparable or even superior to cisplatin. This work shows the potential utility of the β-
diketiminate ligand in the rational design of new anticancer metal-containing drugs. A related complex
with a η6-C6H5CF3 ligand was prepared and found to undergo a nucleophilic addition reaction at the
coordinated arene ring to afford a substituted η5-cyclohexadienyl derivative.
Introduction
acetylacetones, imidazoles, maltol, ethylenediamines, and
others.10-21 It seems that many different supporting ligands
can be used in combination with the ruthenium(II)-arene
motif to provide different biological functions, including
biologically active ligands,22-24 and functionalization of
the arene is also possible;25,26 however, at present very few
structure-activity relationships have been developed,
Organometallic compounds are under intensive investiga-
tion as putative chemotherapeutic compounds.1-7 This re-
search is being driven by a continuing search for new types of
ligands that endow a metal with specific biological functions
while providing relevant stability and pharmacokinetic
properties.8,9 While many different classes of organometallic
compounds are under evaluation, the η6-arene ruthenium
complexes have proven to be a highly versatile class of
metallopharmaceutical. A number of different types of
auxiliary supporting ligands have been used in conjugation
with the η6-arene ligand including water-soluble phosphines,
(14) Berger, I.; Hanif, M.; Nazarov, A. A.; Hartinger, C. G.; John,
R. O.; Kuznetsov, M. L.; Groessl, M.; Schmitt, F.; Zava, O.; Biba, F.;
Arion, V. B.; Galanski, M.; Jakupec, M. A.; Juillerat-Jeanneret, L.;
Dyson, P. J.; Keppler, B. K. Chem.-Eur. J. 2008, 14 (29), 9046–9057.
(15) Renfrew, A. K.; Phillips, A. D.; Egger, A. E.; Hartinger, C. G.;
Bosquain, S. S.; Nazarov, A. A.; Keppler, B. K.; Gonsalvi, L.; Peruzzini,
M.; Dyson, P. J. Organometallics 2009, 28 (4), 1165–1172.
(16) Renfrew, A. K.; Phillips, A. D.; Tapavicza, E.; Scopelliti, R.;
Rothlisberger, U.; Dyson, P. J. Organometallics 2009, 28 (17), 5061–
5071.
*Corresponding authors. E-mail: andrew.phillips@ucd.ie; paul.dyson@
epfl.ch.
(1) Jaouen, G. Bioorganometallics; VCH-Wiley: Weinheim, 2006.
(2) Hartinger, C. G.; Dyson, P. J. Chem. Soc. Rev. 2009, 38 (2), 391–
401.
(3) Allardyce, C. S.; Dorcier, A.; Scolaro, C.; Dyson, P. J. Appl.
Organomet. Chem. 2005, 19 (1), 1–10.
(4) Allardyce, C. S.; Dyson, P. J. Bioorganomet. Chem. 2006, 17, 177–
210.
(17) Vock, C. A.; Ang, W. H.; Scolaro, C.; Phillips, A. D.; Lagopoulos,
L.; Juillerat-Jeanneret, L.; Sava, G.; Scopelliti, R.; Dyson, P. J. J. Med.
Chem. 2007, 50 (9), 2166–2175.
(18) Vock, C. A.; Renfrew, A. K.; Scopelliti, R.; Juillerat-Jeanneret,
L.; Dyson, P. J. Eur. J. Inorg. Chem. 2008, (10), 1661–1671.
(19) Vock, C. A.; Scolaro, C.; Phillips, A. D.; Scopelliti, R.; Sava, G.;
Dyson, P. J. J. Med. Chem. 2006, 49 (18), 5552–5561.
(5) Fish, R. H.; Jaouen, G. Organometallics 2003, 22 (11), 2166–2177.
(6) van Staveren, D. R.; Metzler-Nolte, N. Chem. Rev. 2004, 104 (12),
5931–5986.
(7) Leyva, L.; Sirlin, C.; Rubio, L.; Franco, C.; Lagadec, R. L.;
Spencer, J.; Bischoff, P.; Gaiddon, C.; Loeffler, J.-P.; Pfeffer, M. Eur. J.
Inorg. Chem. 2007, (19), 3055–3066.
(20) Scolaro, C.; Chaplin, A. B.; Hartinger, C. G.; Bergamo, A.;
Cocchietto, M.; Keppler, B. K.; Sava, G.; Dyson, P. J. Dalton Trans.
2007, 5065–5072.
(21) Ang, W. H.; Daldini, E.; Scolaro, C.; Scopelliti, R.; Juillerat-
Jeannerat, L.; Dyson, P. J. Inorg. Chem. 2006, 45 (22), 9006–9013.
(22) Schmid, W. F.; John, R. O.; Arion, V. B.; Jakupec, M. A.;
Keppler, B. K. Organometallics 2007, 26 (26), 6643–6652.
(8) Dyson, P. J. Chimia 2007, 61 (11), 698–703.
€
(9) Ang, W. H.; Dyson, P. J. Eur. J. Inorg. Chem. 2006, 20, 4003–4018.
(10) Ferreira, A. P.; da Silva, J. o. L. F.; Duarte, M. T.; da Piedade,
M. F. t. M.; Robalo, M. P.; Harjivan, S. G.; Marzano, C.; Gandin, V.;
Marques, M. M. Organometallics 2009, 28 (18), 5412–5423.
(11) Peacock, A. F. A.; Melchart, M.; Deeth, R. J.; Habtemariam, A.;
Parsons, S.; Sadler, P. J. Chem.-Eur. J. 2007, 13 (9), 2601–2613.
(12) Kandioller, W.; Hartinger, C. G.; Nazarov, A. A.; Kuznetsov,
M. L.; John, R. O.; Bartel, C.; Jakupec, M. A.; Arion, V. B.; Keppler,
B. K. Organometallics 2009, 28 (15), 4249–4251.
(23) Schmid, W. F.; John, R. O.; Muhlgassner, G.; Heffeter, P.;
Jakupec, M. A.; Galanski, M.; Berger, W.; Arion, V. B.; Keppler, B.
K. J. Med. Chem. 2007, 50 (25), 6343–6355.
(24) Ang, W. H.; De Luca, A.; Chapuis-Bernasconi, C.; Juillerat-
Jeanneret, L.; Lo Bello, M.; Dyson, P. J. ChemMedChem 2007, 2 (12),
1799–1806.
(25) Ang, W. H.; Daldini, E.; Juillerat-Jeanneret, L.; Dyson, P. J.
Inorg. Chem. 2007, 46 (22), 9048–9050.
(26) Ang, W. H.; Parker, L. J.; De Luca, A.; Juillerat-Jeanneret, L.;
Morton, C. J.; Lo Bello, M.; Parker, M. W.; Dyson, P. J. Angew. Chem.,
Int. Ed. 2009, 48 (21), 3854–3857.
(13) Peacock, A. F. A.; Sadler, P. J. Chem.-Eur. J. 2008, 3 (11), 1890–
1899.
r
2009 American Chemical Society
Published on Web 12/22/2009
pubs.acs.org/Organometallics

418 Organometallics, Vol. 29, No. 2, 2010
Phillips et al.
Figure 1. Substitution pattern of the β-diketiminato ligands with electron-withdrawing substituents employed in this study.
Chart 1
although some trends are emerging. In any case, compared to
platinum-based drugs, ligand-based studies are limited, and
there is a sustained need for further ligand screening to
facilitate a more rational drug design approach.
Results and Discussion
A useful feature of β-diketiminate ligands is the ability to
modulate their electronic and steric properties by changing
the nature of the flanking aryl groups and the substituent
at the R-positions of the ligand backbone. Previously, we
reported the chloro-β-diketiminato-ruthenium complex 2,
with electron-donating substituents, i.e., 2,6-dimethylphenyl
and R-methyl groups, which displays limited stability in
solution when exposed to water or oxygen.32 In order to
deter reactivity with O2 and control the rate of hydrolysis,
deactivation of the nucleophilic properties associated with
the central β-carbon position of the β-diketiminate ligand is
essential. Deactivation may be accomplished by introducing
strongly electron-withdrawing CF3 groups onto the β-dike-
timinate ligand. Accordingly, three different ligands, 4, 5,
and 6 (see Figure 1), were synthesized using established
literature procedures. Furthermore, a simple unsubstituted
β-diketiminate ligand with phenyl flanking groups was pre-
pared as a reference compound to assist in comparing the
stability of complexes bearing electron-withdrawing groups.
This β-diketiminate was obtained in high yield according to
the procedures described by Holm and McGeachin;35,36
however, decomposition is observed if this compound is
not protected from moisture. Ligand 4 was prepared under
conditions similar to those used for 2,6-dimethylphenyl-
substituted β-diketiminate. The synthesis of ligand 5 relied
on titanium chloride-assisted thermocoupling of substituted
anilines with hexafluoroacetylacetone as described pre-
viously,37,38 whereas compound 6 utilized the thermal-in-
duced method of NdC coupling, using the ylid Ph3PdN(3,5-
(CF3)2C6H3) and CF3C(O)CH2C(O)CF3, with elimination
of Ph3PO as described by Sadighi.39
Recently our research has focused on an interesting family
of anionic ligands, namely, the diazo-chelating β-diketimi-
nates 1, which have traditionally found widespread use in the
isolation of a wide variety of unusual and interesting main-
group-, lanthanide-, and actinide-centered complexes.27 De-
spite a growing number of catalytic applications of metal-β-
diketiminate complexes reported in the past decade, as far as
we are aware, no biological applications of β-diketiminate-
containing complexes have been reported. Instead, within
the field of bioinorganic chemistry, the synthesis and char-
acterization of low-coordinate complexes supported by
β-diketiminates have focused on the binding of small-mole-
cule substrates in unusual bonding modes, and a number of
complexes have been identified as models for the catalyti-
cally active sites in metalloenzymes.28-31
Previously, we reported the first examples of β-diketiminato-
ruthenium complexes that also contain a η6-C6H6 ligand, both
with (2) and without (3) a chloride co-ligand; see Chart 1.32,33
The bifunctional nature of these compounds proved valuable
in metal-ligand-mediated catalysis.34 Through the rational
modulation of the β-diketiminato ligand we report herein on a
series of new arene β-diketiminato-ruthenium complexes that
display potent in vitro anticancer activity. Indeed, the results
lead us to hypothesize that the ability of the β-diketiminato
ligand to induce the facile loss of the chloride ligand may
facilitate reactions with potential biomolecular targets.
(27) Bourget-Merle, L.; Lappert, M. F.; Severn, J. R. Chem. Rev.
2002, 102 (9), 3031–3066.
(28) Holland, P. L. Can. J. Chem. 2005, 83 (4), 296–301.
(29) Tolman, W. B., Ed. Activation of Small Molecules: Organome-
tallic and Bioinorganic Perspectives; 2006; p 363.
To facilitate binding of the β-diketiminate ligand to the
η6-arene-ruthenium fragment, deprotonation of the azo cen-
ter was accomplished using n-butyllithium in hydrocarbon
(30) Tolman, W. B.; Spencer, D. J. E.; Reynolds, A. M.; Holland,
P. L.; Jazdzewski, B. A.; Young, V. G. J. Inorg. Biochem. 2001, 86 (1),
107–107.
(31) Inosako, M.; Kunishita, A.; Shimokawa, C.; Teraoka, J.; Kubo,
M.; Ogura, T.; Sugimoto, H.; Itoh, S. Dalton Trans. 2008, 44, 6250–6256.
(32) Phillips, A. D.; Laurenczy, G.; Scopelliti, R.; Dyson, P. J.
Organometallics 2007, 26 (5), 1120–1122.
(35) McGeachin, S. G. Can. J. Chem. 1968, 46 (11), 1903–1912.
(36) Parks, J. E.; Holm, R. H. Inorg. Chem. 1968, 7 (7), 1408–1416.
(37) Carey, D. T.; Cope-Eatough, E. K.; Vilaplana-Mafe, E.; Mair,
F. S.; Pritchard, R. G.; Warren, J. E.; Woods, R. J. Dalton Trans. 2003, 6,
1083–1093.
(33) Moreno, A.; Pregosin, P. S.; Laurenczy, G.; Phillips, A. D.;
(38) Li, Y.; Jiang, L.; Wang, L.; Gao, H.; Zhu, F.; Wu, Q. Appl.
Dyson, P. J. Organometallics 2009, 28 (22), 6432-6441.
Organomet. Chem. 2006, 20 (3), 181–186.
(39) Laitar, D. S.; Mathison, C. J. N.; Davis, W. M.; Sadighi, J. P.
Inorg. Chem. 2003, 42 (23), 7354–7356.
€
(34) Grutzmacher, H. Angew. Chem., Int. Ed. 2008, 47 (10), 1814–
1818.

Article
Organometallics, Vol. 29, No. 2, 2010 419
Scheme 1. General Synthetic Route for the η6-Arene β-Diketiminato-Ruthenium Complexes 7-10
solvent at -78 °C to prevent possible decomposition. The
resulting β-diketiminato-lithium complexes precipitated from
solution;40 see Scheme 1. We have previously reported the
synthesis of the arene chloro-β-diketiminato-ruthenium com-
plex 2, in which the corresponding β-diketiminato-lithium
conjugate is added to the [(η6-C6H6)RuCl2]2 dimer in dichlor-
omethane and subsequent recrystallization affords the pro-
duct.32 Following this procedure, complexes 7-10 were
prepared in moderate to high yield. Complexes 8-10 are air
stable, and evidence for decomposition in a saturated dichlor-
omethane solution was detectable by 19F NMR only after
several weeks of storage without precautions to exclude air. In
contrast, the phenyl-substituted complex 7 displays signifi-
cantly increased reactivity toward oxygen and water, and
manipulation of this compound required the use of inert
conditions.
Scheme 2. Proposed Mechanism for the Formation of Complex
11, Involving Initial Nucleophilic Attack by the β-Carbon Posi-
tion of a β-Diketiminato-Lithium Conjugate on the Coordinated
η6-Arene of the Chloro-β-diketiminato-Ruthenium Complex 12,
Followed by Elimination of LiCl from 13, Yielding Complex 11,
Which Features a Hybrid β-Diimine-η5-cyclohexadienyl Ligand
Previously, we have shown that ruthenium complexes
bearing fluorinated η6-arenes can markedly alter the rate
of hydrolysis in aqueous solutions.16 Therefore, in an at-
tempt to isolate a β-diketiminato-ruthenium complex bear-
ing an electron-withdrawing η6-C6H5CF3 arene using the
above-mentioned procedure, an unexpected side reaction
was observed. The resulting product 11, obtained in 42%
yield, featured two sets of aryl-imine-type resonances in
the 1H NMR spectrum. The structure of 11 was established
by X-ray crystallography, revealing the presence of a β-
diketiminato-ruthenium moiety; however, a β-diimine li-
gand was attached through the β-carbon to the para-position
of the η6-arene of species 12 in Scheme 2, forming a com-
pound bearing an anionic η5-cyclohexadienyl ligand, which
is very rare for ruthenium.41-45 Presumably, the reaction
proceeds via nucleophilic attack by the β-carbon position of
the β-diketiminato-lithium conjugate, and similar reactions
of this type of lithium complex with other substrates have been
reported.37,46 Accordingly, the intermediate precursor of 13
features two anionic ligands, and hence the chloride ligand is
eliminated as LiCl, forming 11. Nucleophilic addition to
coordinated η6-arene ligands has been observed for other types
of ruthenium complexes.41-43,47,48
All the complexes are soluble in chlorinated solvents,
methanol, acetone, and DMSO and are slightly soluble in
water. The structures of the complexes were confirmed by
1
solution H, 13C, and 19F NMR spectroscopy, and conclu-
sive assignment of all resonances was performed using 2D
NMR techniques (gCOSY, HSQC, and HMBC). In the case
of the fluorinated compounds, 1D 13C{19F} NMR spectros-
copy was used to simplify the spectra by suppressing the
C-F coupling, which otherwise resulted in overlapping
quartets. Table 1 lists the various chemical shift values for
diagnostic protons and carbons related to the η6-arene and
β-diketiminate ligands. For complex 2, the ortho-methyl
groups are inequivalent, indicating that the aryl group is
not able to rotate freely about the Cipso-N bond. In 9,
however, the ortho-methyl group resonances are observed
as a broad singlet (w1/2=30 Hz), and although rotation of the
aryl group is sterically restricted, the origin of this dynamic
behavior probably results in a dynamic bending of the
β-diketiminate ligand along the N,N0-vector. Moreover,
(40) Stender, M.; Wright, R. J.; Eichler, B. E.; Prust, J.; Olmstead, M.
M.; Roesky, H. W.; Power, P. P. J. Chem. Soc., Dalton Trans. 2001, 23,
3465–3469.
(41) Robertson, D. R.; Robertson, I. W.; Stephenson, T. A. J.
Organomet. Chem. 1980, 202 (3), 309–318.
(42) Robertson, D. R.; Stephenson, T. A. J. Organomet. Chem. 1977,
142 (2), C31–C34.
(43) Bhambri, S.; Tocher, D. A. J. Organomet. Chem. 1996, 507
(1-2), 291–293.
(44) Yang, S.-M.; Chan, M. C.-W.; Peng, S.-M.; Che, C.-M. Orga-
nometallics 1998, 17 (2), 151–155.
(45) Huckett, S. C.; Angelici, R. J. Organometallics 2002, 7 (7), 1491–
1500.
(46) Burford, N.; D’Eon, M.; Ragogna, P. J.; McDonald, R.;
(47) Le Bozec, H.; Touchard, D.; Dixneuf, P. H. Organometallic
Chemistry of Arene Ruthenium and Osmium Complexes; Academic Press:
London, 1989; Vol. 29, pp 163-248.
Ferguson, M. J. Inorg. Chem. 2004, 43 (2), 734–738.
(48) Bennett, M. A. Coord. Chem. Rev. 1997, 166, 225–254.

420 Organometallics, Vol. 29, No. 2, 2010
Phillips et al.
Table 1. Selected 1H and 13C NMR Chemical Shifts of the η6-C6H6 and β-Diketiminate Ligands in 2 and 7-10
δ(1H)
β-CH
δ(13C)
β-CH
δ(13C)
R-CCH3
δ(1H)
C6H6
δ(13C)
C6H6
δ(19F)
Ar-CF3
δ(19F)
R-CF3
aryl
R
2
7
8
9
10
2,6-Me2C6H3
C6H5
3,5-(CF3)2C6H3
2,6-Me2C6H3
3,5-(CF3)2C6H3
Me
Me
Me
CF3
CF3
5.77
4.45
4.45
5.52
5.54
99.53
94.42
94.41
90.93
86.90
160.81
159.85
160.10
151.35
150.85
4.84
4.52
4.52
4.57
4.67
85.38
86.32
86.30
87.93
87.49
-59.2
-63.3
-64.2
-59.7
Figure 2. ORTEP diagrams of 7 (top left), 8 (top right), 9 (bottom left), and 10 (bottom right). Thermal ellipsoids are drawn with 50%
probability. Solvates and η6-C6H6 disorder have been omitted for clarity.
the fluorine atoms of the R-CF3 group also exhibit broad-
ening in the 19F NMR spectrum (see below). In contrast,
complexes 8 and 10 with CF3-substitution on the 3- and
5-positions of the flanking aryl groups feature only a single
resonance in the 19F NMR spectra with the equivalence
being due to free rotation about the N-Cipso bond.
[M - Cl - H]þ. Fragmentation of the [M - Cl - H]þ peak
by collision-induced dissociation (CID) revealed that the
η6-benzene ligand is lost in preference to the β-diketiminate
ligand. Further fragmentation did not result in loss of the
β-diketiminate ligand, but instead at high CID energies, loss
of the flanking aryl groups was observed. The ESI-MS for
the phenyl-substituted complex 7 also revealed an additional
peak at 16 m/z greater than the cationic parent ion, 429.0 m/z.
This species features a single oxygen, presumably bound to
the Ru center, as observed in a number of reported ruthe-
nium porphyrin species.49,50 Furthermore, if the ESI-MS is
performed with degassed solvent, this additional signal is
suppressed. Only 2 and 7 demonstrated this behavior; species
bearing electron-withdrawing groups in the aryl or R-posi-
tion showed no indication of a Ru-O containing species.
Structural Characterization of 7-11 in the Solid State.
Suitable single crystals of 7-10 were obtained using solvent
diffusion methods (see Experimental Section) and were
analyzed by X-ray diffraction, and the resulting structures
are shown in Figure 2. All complexes have a piano-stool-type
geometry with the Cl-Ru vector bisecting the chelating
β-diketiminate ligand, with the latter bond being parallel to
the β-CH position. A comparison of the pertinent bonding
parameters for the complexes reveals that the most striking
difference concerns the Ru-Cl bond length. Previously we
A comparison of the NMR data of 2 with 7 reveals
considerable electronic differences between the two com-
pounds. In particular, the δ(1H) and δ(13C) of the β-CH
position are more deshielded in 2, suggesting a greater
buildup of electron density in this region of the β-diketimi-
nate ligand. The structural dichotomy between the com-
plexes is strongly evident in the solid-state structure (see
below). Moreover, the δ(1H) and δ(13C) values associated
with the η6-C6H6 ring provide an indication of the relative
strength of metal-benzene bonding in the series of com-
plexes and importantly the influence exerted by different
β-diketiminate ligands; complex 2, with the most strongly
donating ligand of the series, has a η6-benzene with the
highest amount of distortion in planarity, as shown by a
more shielded δ(1H) value. In contrast, the fluorinated
β-diketiminate complexes, especially those with CF3 R-sub-
stitution, i.e., 8 and 10, have the least distorted η6-C6H6 ring,
suggesting relatively weak σ-donation by these electron-
withdrawing β-diketiminato ligands.
Complexes 7-10 were analyzed by ESI-MS in the positive
ion mode in CH2Cl2 or MeOH, which in general afforded a
strong peak envelope corresponding to the molecular species
in which the chloride ligand and a proton had been lost, viz.,
(49) Meunier, B. Chem. Rev. 2002, 92 (6), 1411–1456.
(50) Groves, J. T.; Shalyaev, K.; Lee, J. Oxometalloporphyrins in
Oxidative Catalysis; Academic Press: London, 2000; Vol. 4, pp 17-40.

Article
Organometallics, Vol. 29, No. 2, 2010 421
˚
Table 2. Selected Bond Lengths (A) and Angles (deg) for
Complexes 2 and 7-10
a completely planar β-diketiminate ligand incorporating the
metal center. Consequently, the η6-benzene is tilted in the
opposite direction away from the chloride ligand, and the
Ru-Cl bond vector is almost at a right angle with the ligand
plane. This type of geometry is very similar to that observed
in η6-arene chloro-acetylacetone complexes featuring ruthe-
nium.52
2
7
8a
9a
10
Ru-Cl
Ru-N
2.521(1) 2.461(1)
2.099(2) 2.105(2)
2.102(2)
2.464(2) 2.463(1) 2.414(1)
2.115(5) 2.127(2) 2.108(2)
2.103(4) 2.115(2) 2.106(2)
1.431(7) 1.460(4) 1.436(3)
1.435(7) 1.457(3) 1.433(3)
1.332(7) 1.326(3) 1.319(3)
1.335(7) 1.323(3) 1.314(3)
1.398(8) 1.400(4) 1.391(3)
1.395(7) 1.394(4) 1.391(3)
1.525(8) 1.533(4) 1.530(3)
1.529(7) 1.536(4) 1.537(3)
Cipso-N
N-CR
1.450(4) 1.440(3)
1.439(3)
1.335(3) 1.330(3)
1.330(4)
Complex 9, with R-CF3 groups, adopts folded β-diketimi-
nate geometry, but in solution, there appears to be a rapid
equilibrium between both types of ligand geometries, as
exemplified by complexes 2 and 7; see Figure 3. Interestingly,
all of the complexes retain near identical Cl-Ru-Nβ bond
angles (see Table 2). The ortho-CH3 substitution on the
flanking aryls in 9 also promotes a repulsive steric effect,
which is demonstrated by wider Ru-N-Cipso angles com-
pared to 7, 8, and 10. A stronger steric effect is observed in
the R-substitution of the CF3 versus CH3 group in the
β-diketiminate ligand. Specifically, the CR-C(F3) bonds in
9 and 10 are longer than the corresponding CR-C(H3) bonds
in 2, 7, and 8. This is somewhat counterintuitive, since it is
expected that the strongly electron withdrawing groups
should shorten bonds and enhance the transfer of electron
density. However, it is apparent that the repulsive steric
effects of the larger size fluorine are dominant, and thus
the group as a whole imparts a greater ionic polar component
to the C-CF3 bond, i.e., δþC;δ-CF3.53 A comparison of 2
and 9 with ortho-methyl substitution (see Figure 2) reveals
that the most significant changes are larger N-CR-C(H/F)
bond angles and a slight decrease in the Ru-N-Cipso bond
angle in 9. All these features can be traced to greater repulsive
effects of the trifluoromethyl group. Further discussion
regarding the solid-state characterization of complexes
7-10 is given in the Supporting Information.
The structure of 11 is shown in Figure 4, with selected
structural parameters given in Table 3. The β-diketimina-
to-ruthenium fragment has metric parameters that are
identical to those of the cationic η6-arene β-diketimina-
to-ruthenium species, 3, previously described.33 The
Ar-NdC(CH3) (Ar = 2,6-dimethylphenyl) arms of the
β-diimine-η6-cyclohexadienyl ligand are disordered over
two sites, with the common linkage site being the β-carbon,
which is attached to the η6-cyclohexadienyl moiety. The
elongated N-C bonds have distances that are consistent
with imine-type bonds. The η6-cyclohexadienyl ligand con-
sists of a methylene CH2 group, which is puckered away from
the metal center; however, this particular ligand is more
highly distorted than the η6-C6H7 ligand found in the
ruthenium-1,4,7-triazacyclononane complex, which is prob-
ably due to the presence of the highly electron withdrawing
CF3 and β-diimine groups.
CR-Cβ
1.394(3) 1.398(4)
1.406(4)
CR-C(H/F)3
Ru-areneb
1.518(4) 1.525(4)
1.522(4)
1.687(1) 1.690(13) 1.715(4) 1.673(5) 1.703(1)
1.672(6) 1.734(7)
areneb-Ru-Cl 122.3(1) 125.6(1)
areneb-Ru-Nc 154.1(1) 150.7(8)
123.2(1) 122.8(2) 126.0(1)
129.1(1) 127.1(3)
153.2(2) 155.5(2) 154.0(1)
147.3(2) 151.2(3)
83.5(1)
88.3(3)
114.5(3) 118.2(2) 113.8(1)
115.6(3) 116.8(2) 113.5(1)
118.4(4) 119.0(2) 120.0(2)
117.5(4) 120.7(2) 120.4(2)
120.0(5) 119.3(2) 126.1(2)
119.2(5) 119.0(2) 126.4(2)
129.5(5) 126.0(3) 126.6(2)
177.3(3) 155.1(1) 170.5(1)
Cl-Ru-Nc
N-Ru-N
Cipso-N-Ru
83.6(1)
86.6(1)
118.5(2) 115.0(2)
115.0(2)
83.5(1)
88.5(1)
81.7(1)
88.1(1)
80.0(1)
87.8(1)
Cipso-N-CR
N-Cβ-CR
116.8(2) 118.3(2)
118.0(2)
118.9(2) 119.2(2)
120.5(2)
CR-Cβ-CR
Ru-Nc-CR
126.8(3) 128.6(3)
157.8(2) 178.2(1)
a The η6-C6H6 ring is disordered between two sites in these complexes.
b Refers to the centroid position of the η6-C6H6 ligand. c Refers to the
midpoint of the vector connecting both nitrogen centers.
have reported that 2 features the longest Ru-Cl bond for all
known η6-arene ruthenium-containing complexes.32 The
weakness of this bond is confirmed by facile removal or
substitution of the chloride ligand. The other newly reported
complexes feature significantly shorter Ru-Cl distances and
˚
average about 2.46 A, which is close to the median value
reported for all compounds featuring the η6-arene-Ru
fragment in the CSD,51 except for the highly substituted
fluorinated complex 10, which has the shortest Ru-Cl bond
of the series reported in this paper. The Ru-N and Ru-C-
(arene) distances are largely consistent across the series and
do not change significantly with different β-diketiminate
substitution patterns. Within the β-diketiminato ligand
structure bond distances and angles are also invariant, as
indicated by identical N-CR bond lengths and CR-Cβ-CR
bond angles; see Table 2.
Two distinctive geometric forms are observed in the
structures, which are typified by 2 and 9; see Figures 2
and 3. In complexes featuring ortho-positioned methyl
flanking aryls, 2 and 9, folding occurs in the β-diketiminate
ligand along the N-N vector. This folding effectively divides
and separates the π-bonding of the chelating ligand into two
distinctive regions, N-Ru-N and CR-Cβ-CR. A folding of
the flanking aryl groups toward the benzene ligand ensures
that both nitrogen atoms are planar and engaged in
π-bonding. The tilting angle of the aza groups toward the
Ru center indicates this π-delocalized allylic-type interaction
involves the metal d-orbitals. Moreover, the orientation of
the η6-benzene is orthogonal to the central β-diketiminate
ligand plane. The N,N0 ligand folding in 2 and 9 is exempli-
fied by the narrower Ru-Ncentriod-CR angles of approxi-
mately 150° compared to ca. 180° in 7, 8, and 10, indicative of
Computational Analysis of 2 and 7-10. Density functional
theory was used to optimize structures 7-10 to an energetic
minimum, and vibration analysis confirmed that the struc-
tures were the lowest points on the local potential energy
surface. Geometrical comparison indicated that all calcu-
lated structures have overestimated bond distances as com-
pared to the experimentally observed distances in the solid
state, cf. Tables 2 and 3. However, these results are typical for
(52) Habtemariam, A.; Melchart, M.; Fernandez, R.; Parsons, S.;
Oswald, I. D. H.; Parkin, A.; Fabbiani, F. P. A.; Davidson, J. E.;
Dawson, A.; Aird, R. E.; Jodrell, D. I.; Sadler, P. J. J. Med. Chem. 2006,
49 (23), 6858–6868.
(53) Crampton, M. R.; Emokpae, T. A.; Isanbor, C. Eur. J. Org.
(51) Allen, F. Acta Crystallogr. Sect. B 2002, 58 (3 Part 1), 380–388.
Chem. 2007, (8), 1378–1383.

422 Organometallics, Vol. 29, No. 2, 2010
Phillips et al.
Figure 3. Comparison of the two major structural variants observed. Complexes with the 2,6-dimethylphenyl flanking aryl groups, 2
and 9 (left), feature a fold along the N,N0 vector and divide the β-diketiminate into two regions. Compounds 7, 8, and 10 have no ortho
substitution on the flanking aryl (right) and feature a planar β-diketiminate ligand including the Ru center. Also these complexes
contain a tilted η6-arene group, while complexes 2 and 9 (left) have a perpendicular-oriented arene.
binding, i.e., MBI(Ru-N) = 0.503, also has the most
strongly bound η6-benzene of the series. This result is con-
sistent with the established bonding model of η6-arenes with
metal centers, where a stronger electron donation into the π*
MOs of the arene induces a greater π-back-bonding interac-
tion with the coordinated metal.
The calculated models also provided useful information
regarding the nucleophilicity of the β-carbon atom. Our
impetus for employing the fluorinated β-diketiminate ligand
was to deter reactivity with oxygen to prevent complex
deactivation. Previous work has demonstrated that the
cationic complex 2 promotes heterolytic cleavage of H2
through a cooperative binding of the metal and β-carbon
position. Therefore, assuming a similar reactivity of η6-arene
β-diketiminato-ruthenium with dioxygen, reducing the
charge of the β-carbon position is essential for obtaining
stability in a physiological environment. By introducing
electron-withdrawing groups to the β-diketiminate, induc-
tive σ-based mesomeric effects should remove electron den-
sity from the central core section of the ligand, including the
β-carbon site. To test this assumption, for each complex an
electrostatic potential map was generated using a fixed
electric-static charge range; see Figure 5. As predicted, the
total atomic charge, calculated by the natural bonding order
method (NBO), of both the Cl atom and the β-carbon
position is reduced with trifluoromethyl substitution irre-
spective of the substitution pattern of the β-diketiminate; see
Table 4. Moreover, replacing CF3 by CH3 at the R-positions
of the β-diketiminate ligand is more effective at removing
electron density from the β-carbon position than using aryl
groups with meta-substituted CF3 groups. However, from
the ESP map (see Figure 5), strengthening of the Ru-Cl
bond is predicted when the flanking aryl groups contain
strongly electron withdrawing groups. This effect is additive,
and consequently 10 has the least electronegative Cl and
β-carbon center.
Biological Evaluation of 8-10. Prior to the in vitro bio-
logical evaluation of 8-10 their aqueous stability was stu-
died. Related ruthenium-arene complexes with chelating N-
and O-donor ligands have been shown to undergo exchange
of the chloride ligand with a water ligand with the rate of this
process depending on the electronic nature of the chelating
ligand and the reaction conditions (pH, chloride concentra-
tion, etc).11,56 Moreover, for clinically used Pt-based drugs
Figure 4. ORTEP diagram of 11 with thermal ellipsoids drawn
with 50% probability. The disorder associated with the η5-
cyclohexadienyl β-diimine ligand has been omitted for clarity.
geometry optimizations that employ the hybrid functional
BYL3P.54,55 Importantly, the calculated structures repro-
duce accurately the observed structural trends between the
different complexes, including the Ru-Cl and Ru-N bond
lengths and the ruthenium-η6-arene centroid distances; see
Table 4. A comparison of the calculated and observed
structures for all the complexes is provided in the Supporting
Information, Figures S4-S7. For each system, the Mayer
bond indices (MBI) for selected diagnostic bonds within the
complexes were calculated and give a useful indication of the
relative bond strengths among the series (see Table 4). The
MBI values highlight a number of trends among the com-
plexes, specifically for 10, featuring the heavily fluorinated
β-diketiminate ligand. In agreement with the observed solid-
state Ru-Cl bond distances, 10 has the shortest and stron-
gest Ru-Cl bond of the series. However, the MBI values also
reveal that 10 has the weakest metal coordination to the
η6-benzene. Conversely, 2, which has the strongest electron
donating β-diketiminate ligand, as indicated by strong metal
(56) Mendoza-Ferri, M. G.; Hartinger, C. G.; Mendoza, M. A.;
Groessl, M.; Egger, A. E.; Eichinger, R. E.; Mangrum, J. B.; Farrell,
N. P.; Maruszak, M.; Bednarski, P. J.; Klein, F.; Jakupec, M. A.;
Nazarov, A. A.; Severin, K.; Keppler, B. K. J. Med. Chem. 2009, 52
(4), 916–925.
(54) Sholl, D. S.; Steckel, J. A., Density Functional Theory: A
Practical Introduction; John Wiley & Sons: Hoboken, NJ, 2009.
(55) Koch, W.; Holthausen, M. C., A Chemist’s Guide to Density
Functional Theory, 2nd ed.; Wiley-VCH: Weinheim, 2001.

Article
Organometallics, Vol. 29, No. 2, 2010 423
˚
Table 3. Selected Bond Distances (A) and Angles (deg) for 11
N(1)-Ru(1) = 2.050(5)
N(2)-Ru(1) = 2.054(6)
N(1)-C(1) = 1.460(9)
N(2)-C(14) = 1.447(10)
N(1)-C(9) = 1.340(10)
N(2)-C(12) = 1.343(10)
C(9)-C(11) = 1.390(10)
C(11)-C(12) = 1.398(10)
Ru(1)-C(23) = 2.205(6)
Ru(1)-C(24) = 2.166(7)
Ru(1)-C(25) = 2.120(7)
Ru(1)-C(26) = 2.157(7)
Ru(1)-C(27) = 2.233(7)
C(40)-N(4) = 1.229(12)
C(37)-N(3) = 1.236(14)
N(3B)-C(37B) = 1.38(5)
C(39)-C(37B)-N(3B) = 125(3)
C(38)-C(37)-N(3) =127.6(11)
C(38B)-C(37B)-N(3B) = 107(3)
C(38)-N(3)-C(29) = 87.4(5)
C(38B)-N(3B)-C(29B) = 92.2(18)
C(39)-C(40)-N(4) = 119.3(7)
C(39)-C(40)-N(4B) =109.9(14)
C(40)-N(4)-C(42) = 123.7(9)
C(40)-N(4B)-C(42B) = 110(2)
C(41)-N(4)-C(42) = 91.8(6)
C(41B)-N(4B)-C(42B) = 90.4(19)
N(3B)-C(36B) = 2.91(5)
N(3B)-C(29B) = 1.41(4)
C(40)-N(4B) = 1.37(4)
N(4B)-C(42B) = 1.45(3)
N(1)-Ru(1)-N(2) = 88.5(2)
C(1)-N(1)-Ru(1) = 116.8(4)
C(14)-N(2)-Ru(1) = 116.6(4)
Ru(1)-N(1)-C(9) = 127.2(4)
Ru(1)-N(2)-C(12) = 127.7(5)
N(1)-C(9)-C(10) = 119.4(6)
N(2)-C(12)-C(13) = 119.8(7)
N(1)-C(9)-C(11) = 124.7(6)
N(2)-C(12)-C(11) = 123.7(7)
C(39)-C(37)-N(3) = 117.2(10)
Figure 5. Graphical comparison of the calculated electrostatic potential (ESP) of complexes 7 (top left), 8 (top right), 9 (bottom left),
and 10 (bottom right). Each of the 3D ESP maps is drawn with a fixed charge range, where red represents the areas of highest
electronegativity and blue represents areas of lowest electronegativity.
Table 4. Comparison of Calculated Ru-Cl, Ru-N, and Ru-C-
(η6-C6H6) Bond Distances, along with the Associated Bond
Strength, as Indicated by the Mayer Bond Indices (MBI) for
Energy-Minimized Gas Phase Complexes 2 and 7-10a
hydrolysis is believed to be critical for biological activity,
although for ruthenium(II)-arene systems direct reaction
involving an associative pathway is also possible.8 Thus, the
hydrolysis of 8-10 (7 was not included in this study due to
instability; see above) was studied by UV-vis spectropho-
2
7
8
9
10
1
tometry and H NMR spectroscopy. The study shows that
˚
Ru-Cl (A)
calc 2.483
2.462
2.449
expt 2.521(1) 2.461(1) 2.464(2)
2.464
2.438
in the three complexes the chloride ligand is exchanged by
water quite rapidly (<1 h). Over longer periods the η6-C6H6
ligand may also be displaced. This observation is in agree-
ment with the ESI-MS study (see above) that showed loss of
chloride, followed by loss of the benzene (at modest collision
energies) without loss of the β-diketiminato ligand observed
(even at high collision energies).
The cytotoxicities of 8-10, and cisplatin included as a
control, were established on ovarian cancer cells, grown for
three days at 37 °C in the presence of the complexes, and
following incubation, cell survival was monitored using the
MTT assay; see Table 5 and Figure 6. Compound 8 is only
sparingly cytotoxic, whereas 9 and 10, with R-CF3 substitu-
tion in the β-diketiminate backbone, are highly cytotoxic
compounds. Their cytotoxicity is equivalent, if not slightly
2.463(1)
0.82
2.138
2.414(1)
0.86
2.122
MBI(Ru-Cl)
Ru-N (A)
0.788
calc 2.13
0.802
2.109
0.829
2.118
˚
expt 2.099(2) 2.105(2) 2.115(5)
2.102(2) 2.103(4)
2.127(2)
2.115(2)
0.497
2.108(2)
2.106(2)
0.483
MBI
0.509
0.504
1.816
expt 1.687(1) 1.69(1)
0.494
1.823
b
˚
Ru-C (A) calc 1.804
1.819
1.842
1.715(3)c 1.673(5)c 1.703(1)
1.672(5)c 1.734(7)c
MBI(Ru-Cb)
β (NBO)
Cl (NBO)
0.44
-0.421
-0.412
0.421
-0.431
-0.4
0.423
-0.423
-0.374
0.433
-0.405
-0.37
0.413
-0.411
-0.355
C
a The charge of the β-carbon position of the β-diketiminate ligand of
the complexes is given, determined by the natural bond order (NBO)
method. b Refers to the distance between Ru and the centroid point of
the η6-C6H6 arene. c Experimental accuracy is reduced due to positional
disorder of the η6-C6H6 arene.

424 Organometallics, Vol. 29, No. 2, 2010
Phillips et al.
Table 5. IC50 Values (μM) of Compounds 8-10 and Cisplatin
Tested on A2780 and A2780cisR Ovarian Cancer Cells
Experimental Section
Synthesis of the starting materials and complexes was carried
out under a purified N2 atmosphere with standard Schlenk
techniques,66 and manipulations of complexes 2 and 7 were
performed in a drybox with a N2 atmosphere containing less
than 1 ppm of O2 and H2O. Complexes 8-10 could be handled
in air without decomposition. All solvents were dried by passage
through aluminum oxide columns or, in some cases, degassed
by passage through a copper column (Innovative Technologies)
and then stored in Schlenk flasks. Celite (545 grade, Merck Co.)
was dried at 160 °C for two days. The ruthenium precur-
sors [(η6-C6H6)RuCl2]2 and [(η6-C6H5CF3)RuCl2]2 and all
β-diketiminates were prepared according to literature pro-
cedures.16,35,39,67-69 The corresponding β-diketiminato-
lithium complexes were prepared by treating the appropriate
protonated β-diketiminate dissolved in pentane with a stoicio-
metric quantity of n-BuLi (1.6 M in hexanes) at -70 °C.40 The
precipitated, highly air and moisture sensitive solids were col-
lected and washed with small amounts of pentane. All other
reagents were purchased from commercial sources and used as
received. NMR spectra were recorded using either Bruker
Avance 200 or 400 instruments. 1H (COSY, NOE) and 13C
(HMBC and HSQC) one- and two-dimensional spectra were
used to assign molecular connectivity and conformation in
solution. Deuterated dichloromethane was distilled over CaH2
complex
A2780
A2780cisR
8
9
10
91 ((3)
1.8 ((0.3)
<0.5
108 ((3)
1.9 ((0.5)
<0.5
cisplatin
4.3 ((0.5)
18.2 ((1)
superior to cisplatin in the A2780 cell line and markedly more
effective than cisplatin in the cisplatin-resistant A2780cisR
variant.
Indeed, complexes 9 and 10 are among the most active
ruthenium-arene compounds reported to date. While com-
parison should be taken with caution, due to the intrinsic
variance between cells (even within the same cell line), most
ruthenium-arene compounds tend to be at least an order of
magnitude less cytotoxic than 9 and 10.2 The increased
cytotoxicity may be correlated with the presence of CF3
groups, with 9 containing R-CF3 groups on the β-diketimi-
nate ligand and the most cytotoxic compound (IC50 < 0.5
μM), i.e., 10, having R-CF3 groups and CF3 groups at the
3,5-positions of the flanking aryl rings. It is unlikely that
the increased cytotoxicity emanates from the release of the
β-diketiminate ligand, with the ligand being the cytotoxic
fragment since loss of the β-diketiminate ligand has not been
observed under any conditions. Moreover, the free β-diketi-
minate ligands are highly insoluble in water, and it was not
possible to determine their in vitro activity. Therefore, if the
β-diketiminate ligand is released, then this process only
occurs once the complex has penetrated the cell membrane.
Concluding Remarks. Modification of the β-diketiminate
ligand with electron-withdrawing groups provides oxygen-
stable ruthenium(II)-arene complexes that hydrolyze in a
controlled fashion when dissolved in water. On the basis of
considerable interest in the cytotoxic properties of organo-
metallic compounds,2,57-64 and in particular of ruthenium-
(II)-arene half-sandwich complexes,15-17,19,22,52,65 the
β-diketiminato complexes were evaluated for anticancer
activity in vitro. Two of these compounds proved to be
highly cytotoxic, comparable or even superior to cisplatin.
Such strong cytotoxicity is, as far as we are aware, unprece-
dented for this class of compounds. While further rational
modification of the β-diketiminate structure is required,
together with further biological studies, this work shows
the potential utility of this ligand in the rational design of
new anticancer metal-containing drugs.
1
˚
and stored over 4 A molecular sieves. Chemicals shifts for H
and 13C spectra were referenced to Me4Si, and 19F spectra were
referenced to CF3Cl. ATR FT-IR spectra were recorded on a
Perkin-Elmer Spectrum-One instrument using freshly ground
samples pressed on top of a diamond anvil window. Sample
preparation and spectral recording (in air) were performed
within 2 min. Elemental microanalyses were obtained using an
Exeter Analytical CE-440 elemental analyzer. Mass spectra
were recorded using either electrospray or nanoelectrospray
techniques on a ThermoFinnigan LCQ DECA XP Plus quadru-
pole ion trap instrument set in positive mode (flow rate: 5 μL per
min; spray voltage: 5 kV; capillary temperature: 100 °C; capil-
lary voltage: 20 V). Conditions were used as described pre-
viously.70 Spectrophotometric measurements were performed
in a 9:1 solution of methanol/water in quartz Suprasil curvettes
(1 cm path length) on either a Perkin-Elmer Lambda 850 or a
Jasco V-550 spectrometer at 25 °C.
Synthesis of (η6-C6H6)RuCl[(C6H5NC(CH3)]2CH, 7. To a 50
mL Schlenk flask was added 0.125 g of [(η6-C6H6)RuCl2]2
(orange powder) along with CH2Cl2 (10 mL). Subsequently, 2
equiv of the corresponding β-diketiminato-lithium complex
Li[(C6H5NC(CH3)]2CH (0.142 g, 0.507 mmol) was dissolved in
10 mL of CH2Cl2. This solution was slowly added to the reaction
mixture over a period of approximately 0.5 h. The flask was
capped, and the reaction mixture was stirred for 14 h. After-
ward, the resulting dark red or purple solution was filtered
through a 1 cm Celite-Schlenk frit combination, and the
solution was reduced under vacuum to a volume of ca. 1 mL.
While stirring, 25 mL of pentane was added, causing the
formation of a purple or red precipitate. This solid was collected
on a Schlenk frit and washed three times with 5 mL of n-pentane.
For recrystallization, the complex was dissolved in a minimum
amount of acetone or chloroform, and n-pentane was slowly
added until the majority of the compound had crystallized. This
purple microcrystalline solid was filtered and washed with
(57) Nguyen, A.; Top, S.; Pigeon, P.; Vessieres, A.; Hillard, E. A.;
Plamont, M. A.; Huche, M.; Rigamonti, C.; Jaouen, G. Chem.-Eur. J.
2009, 15 (3), 684–696.
(58) Top, S.; Thibaudeau, C.; Vessieres, A.; Brule, E.; Le Bideau, F.;
Joerger, J.-M.; Plamont, M.-A.; Samreth, S.; Edgar, A.; Marrot, J.;
Herson, P.; Jaouen, G. Organometallics 2009, 28 (5), 1414–1424.
(59) Strohfeldt, K.; Tacke, M. Chem. Soc. Rev. 2008, 37 (6), 1174–
1187.
ꢁ
ꢀ
(60) Hogan, M.; Claffey, J.; Pampillon, C.; Tacke, M. Med. Chem.
2008, 4 (2), 91–99.
(61) Kirin, S. I.; Ott, I.; Gust, R.; Mier, W.; Weyhermuller, T.;
Metzler-Nolte, N. Angew. Chem., Int. Ed. 2008, 47 (5), 955–959.
(62) Gross, A.; Metzler-Nolte, N. J. Organomet. Chem. 2009, 694
(7-8), 1185–1188.
(63) Zobi, F.; Blacque, O.; Sigel, R. K. O.; Alberto, R. Inorg. Chem.
2007, 46 (25), 10458–10460.
(66) Shriver, D. F.; Drezdzon, M. A., The Manipulation of Air-
Sensitive Compounds, 2nd ed.; Wiley-Interscience: Weinheim, 1986.
(67) Bennett, M. A.; Smith, A. K. J. Chem. Soc., Dalton Trans. 1974,
2, 233–241.
(64) Xavier, C.; Giannini, C.; Gano, L.; Maiorana, S.; Alberto, R.;
Santos, I. J. Biol. Inorg. Chem. 2008, 13 (8), 1335–1344.
(65) Romerosa, A.; Saoud, M.; Campos-Malpartida, T.; Lidrissi, C.;
Serrano-Ruiz, M.; Peruzzini, M.; Garrido, J. A.; Garcia-Maroto, F. Eur.
J. Inorg. Chem. 2007, (18), 2803–2812.
(68) Gao, W.; Mu, Y.; Li, G.-H.; Liu, X.-M.; Su, Q.; Yao, W.; Feng,
S.-H. Gaodeng Xuexiao Huaxue Xuebao 2005, 26 (1), 134–137.
(69) Budzelaar, P. H. M.; De Gelder, R.; Gal, A. W. Organometallics
1998, 17 (19), 4121–4123.
(70) Dyson, P. J.; McIndoe, J. S. Inorg. Chim. Acta 2003, 354, 68–74.

Article
Organometallics, Vol. 29, No. 2, 2010 425
Figure 6. Cell viability using the MTT test: (A) viability of A2780 and A2780cisR cells after 72 h of incubation with 8; (B) viability of
A2780 and A2780cisR cells after 72 h of incubation with 9 and 10.
2
10 mL of n-pentane and dried for at least 10 h under high
vacuum, yield 0.159 g (0.336 mmol, 67.3%). Spectroscopic
analysis reveals residual CH2Cl2 is present. Anal. Calcd for
7: C, 59.54; H, 5.00; N, 6.04. Found: C, 58.52; H, 4.94; N, 5.91.
ESI-MS (25 °C, dry CH2Cl2), (m/z) positive mode: 429.0 ([M -
Cl - H]þ 60%, calcd 429.1), 351.1 ([M - H - Cl - C6H6], 40%,
151.35 (q, JCF = 30.93 Hz, R-CCF3)3), 154.12 (s, Ar i-C).
19F NMR (25 °C, 188.1 MHz, CD2Cl2) δ(ppm): -64.2 (s,
1JFC=283 Hz, R-CF3). FT-IR (25 °C, solid) ν(cm-1): 3183(w),
3060(w), 3008(w), 2928(w), 1593(w), 1570(w), 1551(w), 1471(m),
1455(m), 1384(w), 1311(m), 1298(m), 1268(w), 1256(w),
1215(m), 1206(s), 1168(s), 1138(s), 1093(s), 1035(w), 1011(w),
982(w), 958(m), 920(w), 888(w), 833(s), 796(w), 768(s), 737(m),
715(w), 709(w), 692(w), 666(w).
1
calcd 351.0). H NMR (25 °C, 400.1 MHz, CD2Cl2) δ(ppm):
1.65 (s, 6H, R-CH3), 4.45 (s, 1H, β-CH), 4.52 (s, 6H, η6-C6H6),
7.40 (m, 3JHH=7.46 Hz, 4H, Ar o-CH), 7.46 (m, 3JHH=7.45 Hz,
2H, Ar p-CH), 7.63 (m, 3JHH=7.45, 3JHH=7.46, 4H, Ar m-CH).
13C NMR (25 °C, 100.1 MHz, CD2Cl2) δ(ppm): 24.69 (s,
R-CH3), 86.32 (s, η6-C6H6), 94.42 (s, β-CH), 125.24 (s, Ar
p-CH), 126.92 (s, Ar m-CH), 127.63 (s, Ar m-CH), 129.68 (s,
Ar o-CH), 159.85 (s, R-CCH3), 160.12 (s, Ar i-C). FT-IR (25 °C,
solid): ν(cm-1): 3391(br), 3056(w), 2922(w), 1590(w), 1558(m),
1529(m), 1482(m), 1457(s), 1432(m), 1401(s), 1349(w), 1277(w),
1203(m), 1164(w), 1152(w), 1136(w), 1070(w), 1039(w), 1033(w),
1022(w), 977(w), 955(w), 922(w), 884(w), 860(w), 844(w),
821(m), 821(m), 760(m), 753(m), 712(s), 704(s), 665(w).
Synthesis of (η6-C6H6)RuCl((3,5-(CF3)2C6H3NC(CF3))2CH,
10. This intense red colored compound was prepared with a
method identical to that used for complex 7. [(η6-C6H6)RuCl2]2
(0.125 g, 0.500 mmol) (orange powder) with 0.321 g (0.504
mmol) of Li((3,5-(CF3)2C6H3NC(CF3))2CH was employed.
Anal. Calcd for 10: C, 38.43; H, 1.55; N, 3.32. Found: C,
39.06; H, 1.84; N, 3.24. ESI-MS (25 °C, CH2Cl2), (m/z) positive
mode: 809.0 ([M - Cl - H]þ, 77%, calcd 809.0), 730.9 ([M -
C6H6 - Cl - H]þ, 23%, calcd 730.9). H NMR (25 °C, 400.1
1
MHz, CD2Cl2) δ(ppm): 4.70 (s, 6H, η6-C6H6), 5.54 (s, 1H,
β-CH), 7.90 (s, 2H, Ar p-CH), 8.07 (s, 4H, Ar o-CH). 13C
NMR (25 °C, 100.6 MHz, CD2Cl2) δ(ppm): 86.96 (s, η6-
Synthesis of (η6-C6H6)RuCl((3,5-(CF3)2C6H3NC(CH3))2CH,
8. The magenta-colored complex was prepared using a method
identical to that employed for complex 7. [(η6-C6H6)RuCl2]2
(0.125 g, orange powder) and 0.280 g (0.507 mmol) of Li
((3,5-(CF3)2C6H3NC(CH3))2CH (white powder) were used.
Yield of 8: 0.298 g (0.392 mmol, 78.5%). Spectroscopic analysis
reveals residual CH2Cl2 is present. Anal. Calcd for 8: C, 44.06;
H, 2.60; N, 3.81. Found: C, 44.94; H, 2.51; N, 3.59. ESI-MS
(25 °C, CH2Cl2), (m/z) positive mode: 700.9 ([M - Cl - H]þ,
100%, calcd 701.0). 1H NMR (25 °C, 400.1 MHz, CD2Cl2)
δ(ppm): 1.65 (s, 9H, R-CH3), 4.45 (s, 1H, β-CH), 4.52 (s, 6H, η6-
C6H6), 8.06 (s, 2H, Ar p-CH), 8.15 (s, 4H, Ar o-CH). 13C NMR
(25 °C, 100.6 MHz, CD2Cl2) δ(ppm): 24.68 (s, R-CCH3), 86.30
(s, η6-C6H6), 94.41 (s, β-CH), 123.18 (q, 1JCF=316 Hz, Ar CF3),
125.24 (s, Ar p-CH), 126.90 (s, Ar o-CH), 129.03 (s, Ar o-CH’),
1
C6H6), 86.90 (s, β-CH), 119.07 (q, JCF = 285 Hz, R-CF3),
120.09 (q, 3JCF=3.69 Hz, Ar p-CH), 122.99 (q, 1JCF=258 Hz,
Ar m-CF3), 126.43 (s, Ar o-CH), 131.63 (m, 2JCF=33.9 Hz, Ar
m-CCF3), 150.85 (q, 2JCF=26.9 Hz, R-CCF3), 156.52 (s, Ar i-C).
19F NMR (25 °C, 188.1 MHz, CD2Cl2) δ(ppm): -63.7 (s, 1JFC
=
258 Hz, Ar m-CF3), -59.7 (s, 1JFC=285 Hz, R-CF3). FT-IR (25
°C, solid) ν(cm-1): 2160(w), 1584(w), 1558(w), 1466(m), 1364(s),
1312(w), 1277(s), 1222(m), 1172(s), 1132(s), 1117(s), 1104(s),
1011(w), 983(w), 962(m), 938(w), 900(m), 894(m), 874(m),
829(m), 805(w), 779(m), 743(w), 722(m), 708(m), 682(s).
Synthesis of {η6-1,4-CF3C6H5[2,6-(CH3)2C6H3NC(CH3)]2-
CH}Ru[2,6-(CH3)2C6H3NC(CH3)]2CH, 11. In a 50 mL Schlenk
flask, 0.200 g of [(η6-C6H5CF3)RuCl2]2 (orange powder)
was dissolved in 25 mL of CH2Cl2. Two equivalents of
the β-diketiminato-lithium complex Li((2,6-(CH3)2C6H3NC-
(CH3))2CH was dissolved in 10 mL of CH2Cl2 and slowly added
to the reaction mixture over a period of approximately 10 min.
The flask was capped and the reaction mixture stirred for 36 h.
Afterward, the resulting pink-colored solution was filtered
through a 1 cm Celite-Schlenk frit combination, and the
solvent removed under vacuum to dryness. The resulting residue
was extracted with 30 mL of pentane and filtered through a 1 cm
Celite-Schlenk frit combination. Slow evaporation of a satu-
rated pentane solution in a glovebox resulted in the deposition
of pink crystals. The remaining solution was decanted, and the
product was dried for at least 2 h under high vacuum, yield 42%.
The compound is highly sensitive to oxygen. Anal. Calcd for 11:
C, 68.59; H, 6.46; N, 6.53. Found: C, 67.82; H, 6.40; N, 6.44.
Atom labels for the NMR assignment of 11 are given in Figure 7.
1H NMR (25 °C, 400.1 MHz, C6D6) δ(ppm): 1.45 (s, 6H, m-
CH3), 1.70 (s, 6H, R-CH3), 1.93 (s, 6H, b-CH3), 2.02 (s, 6H,
2
133.19 (q, JCF = 34.0 Hz, Ar m-CCF3), 159.82 (s, Ar i-C),
160.10 (s, R-CCH3). 19F NMR (25 °C, 188.1 MHz, CD2Cl2)
1
δ(ppm): -59.2 (s, JFC = 316 Hz, 3,5-(CF3)2C6H3). FT-IR
(25 °C, solid): 3401(br w), 1618(w), 1558(w), 1536(m),
1456(m), 1437(w), 1401(w), 1359(s), 1275(s), 1223(w), 1167(s),
1127(s), 1114(s), 1039(w), 1010(w), 985(m), 938(w), 930(w),
915(w), 895(m), 873(w), 845(w), 823(m), 768(w br), 717(w),
703(m), 682(s).
Synthesis of (η6-C6H6)RuCl((2,6-(CH3)2C6H3NC(CF3))2CH,
9. The dark pink colored compound was prepared using an
identical method to that for complex 7. [(η6-C6H6)RuCl2]2
(0.125 g, 0.500 mmol) (orange powder) with 0.210 g (0.500
mmol) of Li((2,6-(CH3)2C6H3NC(CF3))2CH was employed.
Yield: 0.246 g (0.392 mmol, 78.4%). Anal. Calcd for 9: C,
51.64; H, 4.01; N, 4.46. Found: C, 51.08; H, 3.93; N, 4.40.
ESI-MS (25 °C, CH2Cl2), (m/z) positive mode: 701.1 ([M - Cl]þ,
100%, calcd 701.0). 1H NMR (25 °C, 400.1 MHz, CD2Cl2)
δ(ppm): 2.52 (br s, 6H, o-CH3), 4.59 (s, 6H, η6-C6H6), 5.55 (s,
1H, β-CH), 7.52 (m, 6H, Ar m-,p-CH). 13C NMR (25 °C, 100.1
MHz, CD2Cl2) δ(ppm): 19.66 (s, Ar o-CH3), 87.93 (s, 6H,
f-CH3), 2.25 (s, 6H, n-CH3), 2.37 (s, 6H, r-CH3), 2.72 (d, 3JHH
=
10.7 Hz, k-CH), 2.81 (dd, 3JHH=6.48 Hz, 3JHH=6.30 Hz, 2H,
i-CH), 3.45 (dd, 3JHH=6.30 Hz, 3JHH=10.7 Hz, 1H, j-CH), 3.49
(d, 3JHH=6.48 Hz, 2H, h-CH), 5.47 (s, 1H, β-CH), 7.04 (m, 8H,
d-,o-,p-,r-CH), 7.13 (m, 2H, e-CH), 7.19 (m, 2H, c-CH). 13C
1
η6-C6H6), 90.93 (s, β-CH), 119.44 (q, JCF =283 Hz, R-CF3),
127.01 (s, Ar p-CH), 128.75 (s, Ar m-CH), 133.86 (s, Ar o-C),

426 Organometallics, Vol. 29, No. 2, 2010
Phillips et al.
ABSSCALE program in CrysAlis RED.73 Space group deter-
mination was performed with the XPREP program.76 A struc-
ture solution based on the direct-method algorithm was
employed with SHELXS-97.77 Afterward, anisotropic refine-
ment of all non-hydrogen atoms was completed on the basis of a
least-squares full-matrix method against F2 data using
SHELXL-97.77 Hydrogen atoms were added through geome-
trically calculated positions and refined as a riding model using a
scaled thermal parameter to the connecting atom. In structures 8
and 9, a positional η6-ring was treated by splitting the atoms
over two positions, the indivdual occupancy of the disordered
groups were allowed to freely refine, while the total site occu-
pancy was set to 1.0 (details described in the accompanying
CIFs). A similar procedure was employed for the disordered η5-
cyclohexadienyl β-diimine component in complex 11. A small
2
number of reflections in some cases were removed when Δ(Fo
-
2
Fc )/σ exceeded 10.0. Selected crystallographic data for all
structures are given in the Supporting Information. Drawings
were produced with the program ORTEP-3.78
Figure 7. Labeling diagram for the positions of hydrogen and
carbon atoms in complex 11.
Computational Studies. In silico studies were performed using
the Gaussian 03 program79 employing density functional theory
with the three-parameter hybrid B3LYP method, which incor-
porates the exchange functional developed by Lee, Parr, and
Yang.80,81,82 For all nonmetal atoms, the contracted 6-31G(d,p)
basis set was selected, which includes diffuse functions.83 For the
ruthenium center, the double-ζ basis set, LANL2dz, was used in
conjunction with a pseudopotential representing the core set of
electrons.84 All structures were geometrically optimized to an
energy minimum. The position on the local potential energy
surface was confirmed through vibration analysis using second
derivatives, where no imagery frequencies were observed. The
electrostatic potential maps were generated from the calculated
electron density using Gaussview 3.85
Cells and Cell Treatment. Human A2780 and A2780cisR cells
were obtained from the European Centre of Cell Cultures
(ECACC, Porton down, Salisbury, UK). All cell culture re-
agents were obtained from Gibco-BRL (Basel, Switzerland).
The cells were grown in RPMI 1640 medium containing 10%
fetal calf serum (FCS) and antibiotics. The complexes were
dissolved in DMSO as 40 mM for stock solution and then
diluted in complete medium to the required concentration.
DMSO at comparable concentrations did not show any effects
on cell cytotoxicity (results not shown).
NMR (25 °C, 100.1 MHz, C6D6) δ(ppm): 18.09 (s, n-CH3), 18.63
(s, b-CH3), 18.77 (s, f-CH3), 18.89 (s, l-CH3), 19.00 (s, r-CH3),
22.82 (s, R-CH3), 40.50 (s, j-CH), 47.35 (s, i-CH), 70.73 (s,
2
k-CH), 73.59 (s, h-CH), 80.49 (q, JCF = 35.4 Hz, g-CH),
100.25 (s, β-CH), 122.87 (s, p-CH), 124.84 (s, f-CCH3), 125.34
(s, q-CH), 125.37 (s, o-CH), 127.68 (s, b-CCH3), 128.08 (s,
d-CH), 128.13 (s, c-CH), 128.28 (s, e-CH), 129.64 (s, r-CCH3),
1
130.65 (q, JCF=271 Hz, CF3), 132.22 (s, n-CCH3), 149.28 (s,
a-C), 154.03 (s, m-C), 159.65 (s, R-CCH3), 167.43 (s, l-CCH3).
19F NMR (25 °C, 188.1 MHz, CD2Cl2) δ(ppm): -60.2 (s,
1JFC = 271 Hz, CF3). FT-IR (25 °C, Nujol mull) ν(cm-1):
1652(m), 1624(w), 1594(w), 1541(m), 1521(w), 1319(m),
1286(w), 1262(w), 1248(w), 1209(w), 1185(m), 1166(m),
1159(m), 1118(m), 1092(w), 1053(w), 1030(w), 985(w), 961(w),
935(w), 919(w), 881(w), 853(w), 828(w), 798(w), 789(w), 722(w),
698(w), 689(w), 634(w), 621(w), 541(w).
Crystallographic Details. Suitable single crystals were re-
moved from the sample vial under a flow of N2 and manipulated
in a perfluoropolyalkyl ether oil matrix (F06206K, ABCR
company) in a specially constructed Dewar partially filled with
liquid nitrogen. The crystals were mounted on the end of a glass
fiber (diameter at least 0.1 mm) attached to a metal pin fixed to a
goniometer head, which was placed in the Euler cradle, while
maintaining a cold blanket of N2 gas. For structures 7-9 and 11,
a Nonius Kappa-CCD diffractometer equipped with a Bruker-
Apex II CCD area detector and an Enraf-Nonius FR590 X-ray
generator was used, while for 10, an Oxford-Diffraction Kuma
Kappa diffractometer with a Sapphire CCD area detector was
employed. Both instruments utilize a graphite-monochromated
(76) Sheldrick, G. M. XPREP, A Reciprocal Space Exploration
Program, 2.06; Bruker-AXS: Madison, WI, 2003.
(77) Sheldrick, G. Acta Crystallogr., Sect. A 2008, 64 (1), 112–122.
(78) Farrugia, L. J. Appl. Crystallogr. 1997, 30 (5 Part 1), 565.
(79) Frisch, M. J. T., G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.;Cheeseman, J. R.;Montgomery, J. A., Jr.;Vreven, T.;Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; and Pople, J. A. Gaussian 03,
Revision E.01; Gaussian Inc.: Wallingford CT, 2004.
˚
Mo KR radiation source with λ=0.71073 A. The crystals were
kept under a 140 or 100 K gaseous flow of N2 during the
collection procedure. The unit cell and orientation matrix were
determined by indexing reflections measured from phi-chi
scans and analyzed with the program DIRAX,71,72 or in the
case of 10 the unit cell was determined from the entire data set
using CrysAlis RED.73 All data sets are based on collecting
reflections using an optimized scanning strategy utilizing
the programs CollectCCD and CrysAlis CCD (for 10 only).74
After data integration with either EvalCCD75 or CrysAlis RED
(for 10 only),73 a multiscan absorption correction based on a
semiempirical method was applied using the SADABS8 or
(71) Duisenberg, A. J. M. J. Appl. Crystallogr. 1992, 25, 92–96.
(72) Duisenberg, A. J. M.; Hooft, R. W. W.; Schreurs, A. M. M.;
Kroon, J. J. Appl. Crystallogr. 2000, 33 (2), 893–898.
(73) CrysAlis Pro RED, 1.71; Oxford Diffraction Ltd.: Yarnton,
Oxfordshire, U.K., 2006.
(74) CrysAlis Pro CCD, 1.71; Oxford Diffraction Ltd.: Yarnton,
Oxfordshire, U.K., 2006.
(75) Duisenberg, A. J. M.; Kroon-Batenburg, L. M. J.; Schreurs, A.
(80) Becke, A. D. J. Chem. Phys. 1993, 98 (7), 5648–5652.
(81) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58 (8),
1200–1211.
(82) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98 (45), 11623–11627.
(83) Petersson, G. A.; Allaham, M. A. J. Chem. Phys. 1991, 94 (9),
6081–6090.
(84) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82 (1), 299–310.
(85) Gaussview, 3; Gaussian Inc.: Wallingford, CT, 2003.
M. M. J. Appl. Crystallogr. 2003, 36, 220–229.

Article
Determination of Cytotoxicity. Cells were grown in 96-well
Organometallics, Vol. 29, No. 2, 2010 427
Experiments were conducted in quadruplicate wells and re-
peated at least twice.
cell culture plates (Corning, NY) at the density of ca. 25 ꢀ 103
cells per well. The culture medium was replaced with fresh
medium containing complexes 8-10 at concentrations varying
respectively from 0 to 160 μM for 8 and from 0 to 10 μM for 9
and 10, with an exposure time of 72 h. Thereafter, the medium
was replaced by fresh medium, and cell survival was measured
using the MTT test as previously described.86 Briefly, 3-(4,5-
dimethyl-2-thiazoyl)-2,5-diphenyltetrazolium bromide (MTT,
Merck) was added at 250 μg/mL, and incubation was continued
for 2 h. Then the cell culture supernatants were removed, the cell
layer was dissolved in DMSO, and absorbance at 540 nm
was measured in a 96-well multiwell plate reader (iEMS
Reader MF, Labsystems, Bioconcept, Switzerland) and com-
pared to the values of control cells incubated without complexes.
Acknowledgment. This research was supported by
grants from the European Commission Marie Curie
Action (A.D.P., MEIF-CT-2005-025287, CARCAS; A.
A.N., MEIF-CT-220890-SuRuCo), the Swiss National
Science Foundation, the EPFL (A.D.P., O.Z., A.A.N.,
and P.J.D.), and University College Dublin. A.D.P.
thanks Science Foundation Ireland (SFI) for a Stokes
Lectureship. Computational resources were provided by
the Irish Center for High-End Computing (ICHEC),
Dublin, Ireland.
Supporting Information Available: This material is available
free of charge via the Internet at http://pubs.acs.org.
(86) Berger, Y.; Greppi, A.; Siri, O.; Neier, R.; Juillerat-Jeanneret, L.
J. Med. Chem. 2000, 43 (25), 4738–4746.