High antitumor activity of highly resistant salan-titanium(IV) complexes in nanoparticles: An identified active species

Full text of DOI:10.1002/anie.201205973

Angewandte
Chemie
DOI: 10.1002/anie.201205973
Drug Discovery
High Antitumor Activity of Highly Resistant Salan–Titanium(IV)
Complexes in Nanoparticles: An Identified Active Species**
Sigalit Meker, Katrin Margulis-Goshen, Ester Weiss, Shlomo Magdassi,* and Edit Y. Tshuva*
Titanium(IV)-based anticancer complexes were the first to
enter clinical trials after platinum compounds.[1] In particular,
budotitane ([(bzac)2Ti(OEt)2]; bzac = benzoylacetonate) and
titanocene dichloride ([Cp2TiCl2]; Cp = cyclopentadienyl)
demonstrated high antitumor activity with reduced toxicity
toward a range of cancer cells; however, use of these
complexes, which both bear two labile ligands, was limited
by their instability in water.[2] Therefore, mechanistic aspects
remained unresolved, including the nature of the active
species and its identification from the multiple hydrolysis
products that were formed. We have recently introduced
cytotoxic salan–TiIV complexes,[3] which are: a) substantially
more hydrolytically stable than known TiIV complexes, and
b) markedly more active than [(bzac)2Ti(OiPr)2], [Cp2TiCl2],
and cis-platin toward a variety of cancer-derived cell lines.
Structure–activity-relationship studies based on both salan[3a,-
solubility and cell-penetration of which are improved through
the reduction of their particle size to the nanoscale dimension.
Reduction of the particle size to the nanometric region
accelerates intercellular permeability and increases the sol-
ubility and dissolution rate.[7] Nanoparticles of salan–TiIV
complexes were obtained by a rapid conversion of a volatile
oil-in-water microemulsion into a dry powder composed of
nanoparticles. Rapid evaporation of the volatile droplets that
contain the complex gave a powder[8] that was easily
dispersible in aqueous media to form a dispersion of stable
nanoparticles. Notably, the surfactants used are approved by
the Food and Drug Administration (FDA) for incorporation
into pharmaceutical dosage forms.
Ti3L1 (m2-O)3 is the crystallographically characterized
3
trinuclear hydrolysis product of the cytotoxic TiL1(OiPr)2
(Scheme 1) and was previously reported as inactive.[3g] This
and labile ligand[3g,4c,d] variations revealed that
b,e–h,4a,b]
reduced steric bulk is favored for cytotoxicity. Additionally,
all cytotoxic complexes slowly gave defined oxo-bridged
polynuclear hydrolysis products.[3b,e–g] Particularly, N-methy-
lated complexes produced well-defined trinuclear clusters,[3f,g]
which were stable for weeks in water. Several observations
indicated that the hydrolysis products play a meaningful role
in the cytotoxicity mechanism;[3] nevertheless, direct meas-
urements on the isolated clusters showed no activity.[3f,g]
Herein we address the hypothesis that cellular penetration,
which is size-dependent, and/or impaired solubility were
limiting factors, and that labile ligands may actually not be
required for cytotoxicity of TiIV complexes, unlike for cis-
platin.[6] This paper presents the high activity of a hydrolysis
product and other particularly resistant TiIV complexes, the
Scheme 1. Salan complex TiL1(OiPr)2 and its trinuclear hydrolysis
product Ti3L1 (m2-O)3.
3
trimer was incorporated into a microemulsion of n-butyl
acetate in water, ultimately forming nanoparticles with
a mean size of (17.6 Æ 1.7) nm for 0.2 wt% of the powder in
water. Cytotoxicity of the clear and stable aqueous dispersion
was measured on colon HT-29 and ovarian OVCAR-1 cancer
cell lines by the methylthiazolyldiphenyl-tetrazolium bromide
(MTT) assay (Figure 1, Table 1). The nanoformulated trimer
exhibited high cytotoxicity, with IC50 values comparable to
those of its monomeric precursor, as previously recorded
without formulations.[3g,h] Control measurements showed no
activity for empty nanoparticles. Thus, inactivity of the non-
formulated trimer resulted from insufficient solubility and/or
inhibited cellular penetration, rather than impaired interac-
tion with the biological target. Apparently, any activity
observed previously for partly hydrolyzed salan–TiIV com-
plexes with some labile ligands does not necessarily identify
the active species, but rather may be a consequence of
additional hydrolysis steps.[9] A previous report on diketonato
derivatives also presented cytotoxicity for a tetrameric
hydrolysis product encapsulated in a liposome.[10] It is thus
plausible that the hydrolysis product serves as a cellular active
[*] S. Meker, E. Weiss, Prof. E. Y. Tshuva
Institute of Chemistry, The Hebrew University of Jerusalem
Jerusalem, 91904 (Israel)
E-mail: edit.tshuva@mail.huji.ac.il
Homepage: http://chem.ch.huji.ac.il/~tshuva
K. Margulis-Goshen, Prof. S. Magdassi
Casali Institute of Applied Chemistry, Institute of Chemistry, The
Center for Nanoscience and Nanotechnology, The Hebrew Univer-
sity of Jerusalem
Jerusalem, 91904 (Israel)
E-mail: magdassi@mail.huji.ac.il
[**] We thank Dr. Shmuel Cohen for crystallography work. Funding was
received from the European Research Council under the European
Community’s Seventh Framework Programme (FP7/2007-2013)/
ERC Grant agreement (No.239603), and partially from the Israel
Science Foundation (grant No.124/09) and Lower Saxony Ministry
of Science.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201205973.
Angew. Chem. Int. Ed. 2012, 51, 1 – 4
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 2. TiIV complexes of reduced lability.
Figure 1. Dependence of OVCAR-1 cell viability on concentration of
TiL1(OiPr)2 (previously reported without nanoformulation[3g,h]) and its
hydrolysis product Ti3L1 (m2-O)3 administered in a nanoformulated and
3
non-nanoformulated[3 g] manner following an incubation period of
three days (concentration of polynuclear compounds was determined
per entire cluster); green: TiL1(OiPr)2 non-nanoformulated; red: Ti3L1 -
3
(m2-O)3 non-nanoformulated; blue: Ti3L1 (m2-O)3 nanoformulated.
3
Table 1: Mean particle size measured for 0.2 wt% dispersion in water
and IC50 (mM) values toward OVCAR-1 and HT-29 cancer cells for the
nanoformulated complexes.
Complex[a]
Particle size [nm]
OVCAR-1 [mm]
HT-29 [mm]
TiL1(OiPr)2
–
12Æ1
14Æ1
[b],[3h]
Ti3L1 (m2-O)3
17.6Æ1.7
9.0Æ0.6
110Æ12.7
5.3Æ0.3
10Æ3
13Æ3
3
TiL2(OiPr)
70Æ22
110Æ50
14Æ4
54Æ16
200Æ30
12Æ2
Ti2L2 (m2-O)
2
TiL3
[a] Ti3L1 (m2-O)3, TiL2(OiPr), and TiL3 were all inactive when administered
3
directly in a non-nanoformulated manner; [b] non-formulated reference.
species as a general mechanistic pathway for TiIV complexes,
thus further questioning the requirement of labile ligands for
the cytotoxicity of TiIV centers.[6]
Under the notion that labile ligands are not essential for
cytotoxicity, pre-designed, inert, and hydrolytically stable
complexes should be cytotoxic as are, thus eliminating the
hydrolysis step and the accompanying undesired release of
side products, such as free labile ligands. Thus, tris- and
tetrakis(phenolato) ligands (L2H3 and L3H4, respectively)
were prepared,[11] and afforded the monomeric octahedral
complexes TiL2(OiPr) and TiL3, respectively (Scheme 2).
1H NMR spectroscopy confirmed that a single product of
each complex had formed and the X-ray crystal structure of
TiL3 featured an octahedral C2-symmetrical complex
(Figure 2).
Figure 2. ORTEP drawings of a) TiL3 and b) Ti2L2 (m2-O) with thermal
ellipsoids at 50% probability; hydrogen atoms and solvent molecule
were omitted for clarity.
2
Although no hydrolysis products could be detected for
TiL3, TiL2(OiPr) slowly hydrolyzed to give a singly oxo-
bridged dimeric product.[3e] The X-ray structure showed a Ci-
symmetrical complex, in which each TiIV center was bound to
the tris(phenolato) ligand (Figure 2).
Comparative hydrolysis measurements by 1H NMR spec-
troscopy were carried out as previously described[3f,g] by
monitoring the integration of selected signals over time
following addition of 10% D2O to a solution of the complexes
TiL2(OiPr), TiL3, and the hydrolysis product Ti2L2 (m2-O)
2
were all inactive when measured directly on HT-29 and
OVCAR-1 cells. However, when formulated into nanoparti-
cles, all three were cytotoxic (Figure 3, Table 1). Particularly,
the most stable complex TiL3 exhibited the highest cytotox-
in [D8]THF. The t = value for isopropoxo hydrolysis of TiL2-
1
2
(OiPr) was approximately 100 hours, markedly higher than
the value previously reported for TiL1(OiPr)2 (ca. 5 h), which
was obtained under similar conditions.[3g] TiL3 demonstrated
even higher stability, where no substantial hydrolysis was
observed for over a week. As expected, the decrease in the
number of labile ligands dramatically increased the hydrolytic
stability of the complexes.
icity, with IC50 values that are comparable to those of the non-
[3g,h]
nanoformulated TiL1(OiPr)2
and its nanoformulated
hydrolysis product Ti3L1 (m2-O)3 (Table 1).
3
To conclude, this work showed that defined hydrolysis
products from salan–TiIV complexes serve as active species
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 4
These are not the final page numbers!

Angewandte
Chemie
Canales, J. Tanski, M. Rossi, Eur. J. Inorg. Chem. 2003, 3221 –
3232; c) F. Caruso, M. Rossi, Mini-Rev. Med. Chem. 2004, 4, 49 –
60; d) F. Caruso, M. Rossi, C. Pettinari, Expert Opin. Ther. Pat.
2001, 11, 969 – 979; e) C. V. Christodoulou, A. G. Eliopoulos,
L. S. Young, L. Hodgkins, D. R. Ferry, D. J. Kerr, Br. J. Cancer
1998, 77, 2088 – 2097; f) G. Kelter, N. J. Sweeney, K. Strohfeldt,
H.-H. Fiebig, M. Tacke, Anti-Cancer Drugs 2005, 16, 1091 – 1098;
g) B. K. Keppler, C. Friesen, H. G. Moritz, H. Vongerichten, E.
Vogel, Struct. Bonding (Berlin) 1991, 78, 97 – 127; h) P. Kçpf-
Maier, H. Kçpf, Chem. Rev. 1987, 87, 1137 – 1152; i) P. Kçpf-
Maier, H. Kçpf, Struct. Bonding (Berlin) 1988, 70, 103 – 185; j) E.
Melꢀndez, Crit. Rev. Oncol. Hematol. 2002, 42, 309 – 315; k) K.
Strohfeldt, M. Tacke, Chem. Soc. Rev. 2008, 37, 1174 – 1187;
l) J. H. Toney, T. J. Marks, J. Am. Chem. Soc. 1985, 107, 947 – 953.
[3] a) C. M. Manna, G. Armony, E. Y. Tshuva, Chem. Eur. J. 2011,
17, 14094 – 14103; b) C. M. Manna, G. Armony, E. Y. Tshuva,
Inorg. Chem. 2011, 50, 10284 – 10291; c) C. M. Manna, O.
Braitbard, E. Weiss, J. Hochman, E. Y. Tshuva, ChemMedChem
2012, 7, 703 – 708; d) C. M. Manna, E. Y. Tshuva, Dalton Trans.
2010, 39, 1182 – 1184; e) S. Meker, C. M. Manna, D. Peri, E. Y.
Tshuva, Dalton Trans. 2011, 40, 9802 – 9809; f) D. Peri, S. Meker,
C. M. Manna, E. Y. Tshuva, Inorg. Chem. 2011, 50, 1030 – 1038;
g) D. Peri, S. Meker, M. Shavit, E. Y. Tshuva, Chem. Eur. J. 2009,
15, 2403 – 2415; h) M. Shavit, D. Peri, C. M. Manna, J. S.
Alexander, E. Y. Tshuva, J. Am. Chem. Soc. 2007, 129, 12098 –
12099; i) E. Y. Tshuva, J. A. Ashenhurst, Eur. J. Inorg. Chem.
2009, 2203 – 2218; j) E. Y. Tshuva, D. Peri, Coord. Chem. Rev.
2009, 253, 2098 – 2115; k) H. Glasner, E. Y. Tshuva, J. Am.
Chem. Soc. 2011, 133, 16812 – 16814.
Figure 3. Dependence of OVCAR-1 cell viability on concentration TiL2-
(OiPr), TiL3, and the hydrolysis product Ti2L2 (m2-O) administered in
2
a nanoformulated (green: TiL2(OiPr), red: TiL3, blue: Ti2L2 (m2-O)) and
2
non-nanoformulated (pink: TiL2(OiPr), purple: TiL3, black: Ti2L2 (m2-O))
2
manner following an incubation period of three days (concentration of
polynuclear compounds was determined per entire cluster).
inside the cell. This conclusion is enabled by the particular
advantage of the salan–TiIV system, which, in addition to
being highly cytotoxic, gives defined hydrolysis products over
long periods that may be more easily monitored for mech-
anistic evaluation. Formulating the hydrolysis products into
nanoparticles enabled overcoming limitations regarding sol-
ubility and cellular penetration of these bulky compounds,
which apparently had inhibited their activity when adminis-
tered directly. Furthermore, the results presented herein
attest to the ability of biologically friendly TiIV to form
complexes that are stable for weeks in an aqueous environ-
ment and are highly cytotoxic. It is now well-established that,
unlike for platinum compounds, ligand lability is not a pre-
requisite for cytotoxicity of TiIV complexes,[6] which serves as
a particular advantage in this case because of the rich aquatic
chemistry of TiIV compounds. Thus, stable TiIV complexes are
certainly attainable and may lead to high activity in a con-
trolled manner without the accompanying release of unde-
sired products. For this reason, administering the stable active
species directly is advantageous over receiving it through
a hydrolysis step. Future design of TiIV-based antitumor drugs
should probably focus on the isolation of small, water-soluble,
coordinatively saturated complexes, with the fewest polyden-
tate ligands possible, with an advantage for a single hexa-
dentate strongly bound ligand. The development of such
systems is underway in our group, as well as the mechanistic
investigation of possible biological targets for these bulky and
inert compounds.
[4] a) A. D. Tinoco, H. R. Thomas, C. D. Incarvito, A. Saghatelian,
A. M. Valentine, Proc. Natl. Acad. Sci. USA 2012, 109, 5016 –
5021; b) T. A. Immel, M. Debiak, U. Groth, A. Burkle, T. Huhn,
ChemMedChem 2009, 4, 738 – 741; c) T. A. Immel, U. Groth, T.
Huhn, Chem. Eur. J. 2010, 16, 2775 – 2789; d) T. A. Immel, M.
Grutzke, A.-K. Spate, U. Groth, P. Ohlschlager, T. Huhn, Chem.
Commun. 2012, 48, 5790 – 5792.
[5] D. Peri, PhD Thesis, The Hebrew University of Jerusalem
(Israel), 2011.
[6] a) M. Shavit, D. Peri, A. Melman, E. Y. Tshuva, J. Biol. Inorg.
Chem. 2007, 12, 825 – 830; b) F. Caruso, M. Rossi, C. Opazo, C.
Pettinari, Bioinorg. Chem. Appl. 2005, 3, 317 – 329; c) M.
Ravera, C. Cassino, E. Monti, M. Gariboldi, D. Osella, J.
Inorg. Biochem. 2005, 99, 2264 – 2269.
[7] a) A. A. Noyes, W. R. Whitney, J. Am. Chem. Soc. 1987, 109,
930 – 934; b) L. F. Knapp, Trans. Faraday Soc. 1922, 17, 457 – 465;
c) A. P. Simonelli, S. C. Mehta, W. I. Higuchi, J. Pharm. Sci. 1970,
59, 633 – 638.
[8] a) K. Margulis-Goshen, H. D. Netivi, D. T. Major, M. Gradziel-
ski, U. Raviv, S. Magdassi, J. Colloid Interface Sci. 2010, 342,
283 – 292; b) K. Margulis-Goshen, B. F. B. Silva, E. F. Marques,
S. Magdassi, Soft Matter 2011, 7, 9359 – 9365; c) K. Margulis-
Goshen, E. Kesselman, D. Danino, S. Magdassi, Int. J. Pharm.
2010, 393, 231 – 238; d) K. Margulis-Goshen, S. Magdassi, Nano-
medicine 2009, 5, 274 – 281.
Received: July 26, 2012
Published online: && &&, &&&&
[9] T. A. Immel, M. Grutzke, E. Batrogg, U. Groth, T. Huhn, J.
Inorg. Biochem. 2012, 106, 68 – 75.
[10] F. Caruso, M. Rossi, J. Tanski, R. Sartori, R. Sariego, S. Moya, S.
Diez, E. Navarrete, A. Cingolani, F. Marchetti, C. Pettinari, J.
Med. Chem. 2000, 43, 3665 – 3670.
Keywords: antitumor agents · cis-platin · ligand design ·
nanoformulation · titanium
.
[11] C. S. Higham, D. P. Dowling, J. L. Shaw, A. Cetin, C. J. Ziegler,
J. A. Farrell, Tetrahedron Lett. 2006, 47, 4419 – 4423.
[12] CCDC 893691 (TiL3) and 893692 (Ti2L2 (m2-O)) contain the
2
[1] K. M. Buettner, A. M. Valentine, Chem. Rev. 2012, 112, 1863 –
1881.
[2] a) P. M. Abeysinghe, M. M. Harding, Dalton Trans. 2007, 3474 –
3482; b) F. Caruso, L. Massa, A. Gindulyte, C. Pettinari, F.
Marchetti, R. Pettinari, M. Ricciutelli, J. Costamagna, J. C.
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Angew. Chem. Int. Ed. 2012, 51, 1 – 4
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Communications
Drug Discovery
S. Meker, K. Margulis-Goshen, E. Weiss,
S. Magdassi,*
E. Y. Tshuva*
&&&& — &&&&
High Antitumor Activity of Highly
Resistant Salan–Titanium(IV) Complexes
in Nanoparticles: An Identified Active
Species
A nanoformulated trinuclear hydrolysis
product of a bis(alkoxo) salan–TiIV com-
plex shows high antitumor activity (see
picture), which identifies it as an active
species in cells. Additional highly stable
mononuclear derivatives also show high
activity, when formulated into nanoparti-
cles, thus evincing that biologically
friendly TiIV can provide high cytotoxicity
with controlled biological function.
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 4
These are not the final page numbers!