Superior chemotherapeutic benefits from the ruthenium-based anti-metastatic drug NAMI-A through conjugation to polymeric micelles

Article abstract of DOI:10.1021/ma402078d

Macromolecular ruthenium complexes are a promising avenue to better, and more selective, chemotherapeutics. NAMI-A is a ruthenium(III) drug in Phase II clinical trials that has low cytotoxicity and is inactive against primary tumors. However, it displays both antiangiogenic and anti-invasive properties and has been shown to specifically target tumor metastases, preventing both development and growth. To increase the cytotoxicity and cell uptake of this promising drug, we designed a biocompatible amphiphilic block copolymer capable of self-assembling into polymeric micelles. An appropriate method for the synthesis of a macromolecular NAMI-A drug was identified - the polymerization of vinyl imidazole and subsequent addition of a ruthenium(III) precursor complex. The cytotoxicity of these polymeric moieties was tested on ovarian cancer A2780 and Ovcar-3 and pancreatic AsPC-1 cancer cell lines. On average, across the tested cell lines, a 1.5 times increase in toxicity was found for the NAMI-A copolymer micelles when compared to the NAMI-A molecule. Furthermore, the antimetastatic potential was assessed by evaluating the inhibitory effects on the migration and invasion of cells against three cell lines characterized by differing degrees of malignancy (MDA-MB-231 > MCF-7 > CHO). The NAMI-A micelles were shown to have an improved antimetastatic potential in comparison to NAMI-A.

Full text of DOI:10.1021/ma402078d

Article
pubs.acs.org/Macromolecules
Superior Chemotherapeutic Benefits from the Ruthenium-Based
Anti-Metastatic Drug NAMI‑A through Conjugation to Polymeric
Micelles
Bianca M. Blunden,†,‡ Aditya Rawal,§ Hongxu Lu,† and Martina H. Stenzel†,*
†Centre for Advanced Macromolecular Design, University of New South Wales, Sydney, NSW 2052, Australia
‡Cooperative Research Centre (CRC) for Polymers, 8 Redwood Drive, Notting Hill, Victoria 3618, Australia
§NMR Facility, Mark Wainwright Analytical Centre, University of New South Wales, Sydney, NSW 2052, Australia
S
* Supporting Information
ABSTRACT: Macromolecular ruthenium complexes are a
promising avenue to better, and more selective, chemo-
therapeutics. NAMI-A is a ruthenium(III) drug in Phase II
clinical trials that has low cytotoxicity and is inactive against
primary tumors. However, it displays both antiangiogenic and
anti-invasive properties and has been shown to specifically
target tumor metastases, preventing both development and
growth. To increase the cytotoxicity and cell uptake of this
promising drug, we designed a biocompatible amphiphilic
block copolymer capable of self-assembling into polymeric
micelles. An appropriate method for the synthesis of a
macromolecular NAMI-A drug was identifiedthe polymer-
ization of vinyl imidazole and subsequent addition of a ruthenium(III) precursor complex. The cytotoxicity of these polymeric
moieties was tested on ovarian cancer A2780 and Ovcar-3 and pancreatic AsPC-1 cancer cell lines. On average, across the tested
cell lines, a 1.5 times increase in toxicity was found for the NAMI-A copolymer micelles when compared to the NAMI-A
molecule. Furthermore, the antimetastatic potential was assessed by evaluating the inhibitory effects on the migration and
invasion of cells against three cell lines characterized by differing degrees of malignancy (MDA-MB-231 > MCF-7 > CHO). The
NAMI-A micelles were shown to have an improved antimetastatic potential in comparison to NAMI-A.
chloride to enhance the stability in the infusion solution, when
given to patients.8
INTRODUCTION
■
Over the past 25 years, numerous ruthenium complexes have
been investigated for medicinal applications. Two RuIII
compounds have entered phase II clinical trials: KP1019 has
been shown to have anticancer activity while NAMI-A is an
antimetastatic agent. Antimetastatic agents are extremely
important in the treatment of cancer, since the majority of
cancer patients die from these secondary cancers.
NAMI-A has low cytotoxicity and is inactive against primary
tumors and thus failed the usual screens of putative anticancer
agents.1 However, it has been shown to specifically target tumor
metastases,2−4 preventing both development and growth,1,3,5
and has significantly greater activity than Cisplatin on these cell
types.6 The NAMI-A effect appears to be independent of the
type of primary tumor or the stage of growth of metastases.3 It
displays both antiangiogenic and anti-invasive properties on
tumor cells and blood vessels.1 It modifies important
parameters of the metastasis such as tumor invasion, matrix
metallo-proteinases activity, and cell cycle progression.7
It has previously been demonstrated that the therapeutic
benefits of anticancer drugs, for example platinum-centered,9
gold-centered,10 or ruthenium-centered,11,12 can be enhanced
by encapsulation in, or conjugation to, a polymer matrix. The
surrounding polymer protects the drug, increases solubility and
often increases cell uptake efficiency due to the cell entry
process being altered from a diffusion mechanism to
endocytosis.13 The creation of a drug carrier in the nanosize
range enables fast endocytosis. Micelles present a unique size
that leads to increased circulation times,14 and the potential to
exploit the EPR effect,15 which are important properties for
therapeutic moieties. Covalent attachment of a therapeutic
agent is a useful avenue to delay drug release until the micelle
reaches a target site.16 Micelles, wherein a drug is covalently
bound have been investigated, for example, PEG-b-poly(ε-
caprolactone) with chemically conjugated docetaxel17 and a
However, the hydrolytic stability of NAMI-A in phosphate
buffer (pH 7.4) at 37 °C is a limiting factor for administration
since it has a half-life of less than 30 min. Thus, NAMI-A is
administered with a physiological concentration of sodium
Received: October 9, 2013
Revised: February 6, 2014
Published: February 26, 2014
© 2014 American Chemical Society
1646
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
(ethylene glycol) methyl ether acrylate (PEGMEA; Aldrich) were
obtained as indicated. 2,2-Azobis(isobutyronitrile) (AIBN; 98%,
Fluka) was purified by recrystallization from methanol. Milli-Q water
was obtained from an UltraPure water purification system. The RAFT
agent 2-(((dodecylthio)carbonothioyl)thio)-2-methylpropanoic acid
was prepared according to a literature procedure.33
Analyses. Size Exclusion Chromatography (SEC). SEC was
implemented using a Shimadzu modular system comprising a DGU-
12A degasser, LC-10AT pump, SIL-10AD automatic injector, CTO-
10A column oven, and a RID-10A refractive index detector. An Agilent
aquagel−OH guard column and two linear columns (Agilent PL
aquagel−OH Mixed-H and Mixed-M 8 μm particle size) were used for
the analyses. A Milli-Q water/acetic acid/methanol (54:23:23)
solution with a flow rate of 0.8 mL min−1 and a constant temperature
of 30 °C was used as the mobile phase with an injection volume of 100
μL. The samples were filtered through 0.45 μm filters. The unit was
calibrated using commercially available linear poly(ethylene oxide)
standards (1900 − 909 500 g·mol−1, Polymer Laboratories).
Chromatograms were processed using Cirrus 2.0 software (Polymer
Laboratories).
Nuclear Magnetic Resonance (NMR) Spectrometry. NMR general
characterization was conducted using a Bruker Avance DPX 300
spectrometer (1H, 300.2 MHz). Further characterization was
conducted using a Bruker DMX 600 MHz spectrometer (1H, 600.13
MHz) with a QNP probe and an Avance III 500 MHz spectrometer
(1H, 500.13 MHz) with a TBI probe. The 600 and 500 MHz
spectrometer parameters are: 8390 Hz sweep width, 1.95 s acquisition
time, 2 s recycle delay; 6009 Hz sweep width, 5.45 s acquisition time, 5
s recycle delay; respectively. The pulse program WATERGATE was
used for water suppression.34 Samples were analyzed in the solvents
DMSO-d6 and D2O. All chemical shifts are stated in ppm (δ) relative
to tetramethylsilane (δ = 0 ppm), referenced to the chemical shifts of
residual solvent resonances (1H and 13C).
number of poly(ethylene oxide) based polymers containing
cisplatin.18−24
NAMI-A comprises an imidazole ligand and counterion. The
imidazole ring is biocompatible, antimicrobial, and anti-
inflammatory and can regulate blood pressure.25 It has many
uses, which often center around its ability to bond to metals as
a ligand and its ability to hydrogen bond with drugs and
proteins.25 Imidazoles also play a vital role in the inhibition of
post-translational farnesylationa key step for RAS proteins
that influence cancer proliferation.25
Importantly, for this work, the imidazole ligand provides an
avenue for polymerization. Poly(vinyl imidazole)s have long
been investigated for their use as catalysts, pH-sensitive DNA
carriers,26 complex coacervates,27 nonviral gene delivery
therapeutics,28,29 and oxygen transport membranes.25 Also,
imidazolium salts are used to extract metal ions from aqueous
media,30 coat metal nanoparticles, dissolve carbohydrates, and
create polyelectrolyte brushes on surfaces.25 Imidazolium
analogues offer electrostatic interactions, aggregation, and self-
assembly. Further, imidazole-based polymers readily associate
with biological molecules through hydrogen-bonding.25 Poly-
(N-vinyl imidazole) has been investigated as an agent to
selectively bind metal ions in order to isolate them from
wastewater. It was found that complexation occurred through
the basic nitrogen atoms at position three of the imidazole
ring.30 Thus, we propose that a polymeric form of NAMI-A can
also be synthesized using the reactive nitrogen on the imidazole
ring.
Imidazole-containing polymers are typically obtained using
free radical methods, which does not offer control of molecular
weight and polymer architecture.29 Dual stimuli-responsive
block copolymers containing 1-vinylimidazolium monomers
have been synthesized in a controlled manner using the RAFT/
MADIX process,31 and more recently Allen et al.32 successfully
polymerized 4-vinyl imidazole via RAFT, using a trithiocar-
bonate RAFT agent and glacial acetic acid as a distinctive
solvent.
In this publication, we demonstrate for the first time that
conjugation of NAMI-A to a nanoparticle can significantly
enhance the drug performance, with respect to cytotoxicity and
also the antimetastatic potential, creating an avenue for the
better treatment of metastatic cancer. The aim of this work is to
explore a synthetic avenue to a macromolecular version of
NAMI-A, and also the synthesis of an amphiphilic block
copolymer incorporating NAMI-A, that allows for the
formation of micelles. 4-Vinyl imidazole was polymerized via
RAFT and the formation of NAMI-A was examined. The
NAMI-A micelles were then evaluated against different
cancerous cell lines, and contrasted with the cytotoxicity of
the NAMI-A drug alone. Finally, it was shown that the
antimetastatic ability of NAMI-A was improved by incorporat-
ing it into a nanoparticle. The nanosized drug conjugate was
significantly better at inhibiting the invasion of cells and also
better at inhibiting the migration of cells, than NAMI-A alone.
The solid-state NMR measurements were carried out on a Bruker
Avance III 300 MHz spectrometer (1H, 299.8 MHz). The samples
were center packed in 4 mm zirconia rotors with Kel-F caps and spun
up to 12 kHz in a Bruker 4 mm H-X double resonance wide bore
1
solids probe head with variable temperature capability. The H NMR
T1H measurements were done using a saturation recovery experiment
involving a saturation comb made of 50 to a hundred delays of 30 μs
to 1 ms duration each. The recycle delays were set to 100 ms with 256
to 512 transients detected for signal averaging.
X-ray Crystallography. Crystals were grown from a DCM/heptane
layered solution in the glovebox. The single crystal of the compound
for analysis was mounted on a Bruker APEXII CCD single crystal
diffractometer equipped with graphite monochromated Mo K
radiation.
Solid-State Elemental Analysis. Polymer samples were sent to The
Campbell Microanalytical Laboratory at the University of Otago in
New Zealand for elemental analysis.
Dynamic Light Scattering (DLS). Particle sizes were determined
using a Brookhaven Zetaplus particle size analyzer (laser = 35 mW, λ =
632 nm, angle = 90°) and a solution of 1 mg mL−1 polymer in distilled
water at 25 °C. Five measurements, with three runs consisting of 2
scans of 2 min in each measurement, were taken of each sample.
Samples were purified from dust using a micro filter (0.45 μm) before
analysis. The mean diameter was obtained from the arithmetic mean
using the relative number or volume intensity of each particle size.
Transmission Electron Microscopy (TEM). The TEM micrographs
were obtained using a JEOL1400 transmission electron microscope,
consisting of a dispersive X-ray analyzer interfaced to the column and a
Gatan CCD facilitating the acquisition of digital images. It was
operated at an accelerating voltage of 80 kV. Samples were prepared by
casting a 1 mg mL−1 polymeric micelle solution onto a copper grid.
The grids were air-dried and then negatively stained with
phosphotungstic acid for 2 min and air-dried again.
EXPERIMENTAL SECTION
■
Chemicals. Urocanic acid (99%, Aldrich), 4,4′-azobis-
(cyanopentanoic acid) (ACPA; 98%, Fluka), glacial acetic acid
(>99.5%, Aldrich), methanol (MeOH; HPLC grade, APS), ethanol
(EtOH; absolute, Fluka), dimethyl sulfoxide (DMSO; >98.9%, Ajax
Finechem), hydrochloric acid (HCl; 37%, Aldrich), acetone (HPLC
grade, Aldrich), diethyl ether (99%, Ajax Finechem), imidazole (>95%,
Fluka), ruthenium trichloride hydrate (RuCl3·H2O; Aldrich), poly-
Thermo Gravimetric Analysis (TGA). Thermal decomposition
properties of products were recorded using a TG5000 Thermogravi-
metric Analyzer. The sample (<1 mg) was placed on a thermo-balance
1647
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
and heated from 25 to 1000 °C at 20 °C min−1 and held isothermally
for 10 min under an air atmosphere.
UV−vis Spectrometry. Analyzed using a Varian Cary 300 UV−vis
Spectrophotometer, fitted with a single cell chamber. Samples were
analyzed at concentrations of 1−5 mg mL−1 in MeOH and water, from
200 to 800 nm at a scan rate of 600 nm min−1.
Fourier-Transform Near-Infrared (FT-NIR) Spectroscopy. FT-NIR
spectroscopy was used to determine monomer conversions by
following the decrease of the vinylic stretching overtone of the
monomer. A Bruker IFS 66\S Fourier transform spectrometer
equipped with a tungsten halogen lamp, a CaF2 beam splitter, and a
liquid nitrogen cooled InSb detector was used. The sample was placed
in a FT-NIR quartz cuvette (1 cm or 2 mm) and polymerized at 70 °C.
Traces were elaborated with OPUS software.
stirring, and acetone (50 mL) and diethyl ether (5 mL) were added.
The solution was transferred into a conical flask and placed in a dark
cupboard to allow for crystal formation. Large red-orange crystals were
washed with diethyl ether, dried under vacuum, and confirmed by X-
ray crystallography to be consistent with literature.36 1H NMR (600.13
MHz, D2O, 25 °C): δ (ppm) = 2.74 (s, 6H, [DMSO2H]); −16.62
(very broad, 6H, [trans-RuCl4(DMSO)2]).
Synthesis of (ImH)[RuIIICl4(Im)(S-DMSO)] (NAMI-A). (ImH)-
[RuIIICl4(Im)(S-DMSO)] (NAMI-A) was synthesized following a
literature procedure.37 [DMSO2H][trans-RuCl4(DMSO)2] (0.1 g, 1.8
× 10−4 mol) was crushed to a mustard-yellow powder and suspended
in acetone (2.0 mL) with stirring. Imidazole (50 mg, 7.4 × 10−4 mol)
was then slowly added while rapidly stirring. The reaction was left
rapidly stirring for 4 h, after which time the product was filtered,
washed with acetone (3.0 mL) and diethyl ether (3.0 mL), and dried
under vacuum, to give a mustard yellow-orange solid (54 mg, 65%).
Syntheses. Synthesis of 4-Vinyl Imidazole (VIm). 4-Vinyl
imidazole (VIm) was synthesized following a modified literature
procedure.32 Urocanic acid (1.5 g, 1.1 × 10−2 mol) was heated to 240
°C under reduced pressure (2.2 × 10−2 mbar) for 6 h. The product
distilled as a clear liquid, which solidified upon cooling to give 4-vinyl
1
The compound was confirmed by H NMR to be consistent with
1
literature. H NMR (600.13 MHz, D2O, 25 °C): δ (ppm) = 8.71 (s,
1H, ImH); 7.50 (s, 1H, ImH); −3.43 (broad, 1H, Im); −5.13(broad,
1H, Im); −6.54(broad, 1H, Im); −15.1 (very broad, 6H, (S-DMSO).
Synthesis of Macromolecular NAMI-A [P(NAMI-A)]. In a typical
preparation, [DMSO2H][trans-RuCl4(DMSO)2 (15 mg, 2.7 × 10−5
mol) was crushed to an orange powder and dissolved in ethanol or
methanol (1 mL). PVIm (10 mg, 1.1 × 10−4 mol) was suspended in
the same solvent (1 mL). The solutions were combined and stirred to
give a mustard yellow suspension. For cytotoxicity assays, the solution
was diluted with water and used immediately. For UV−vis analysis, the
solution was diluted with methanol. For elemental, NMR, and TGA
analyses, the product was filtered, washed with diethyl ether (2 mL),
and dried under vacuum, to give a mustard brown solid.
1
imidazole (4-VIm) as a waxy colorless solid (0.7 g, 66%). H NMR
(300.30 MHz, DMSO-d6, 25 °C): δ (ppm) = 12.12 (s, 1H, NH), 7.61
(s, 1H, Hd); 7.06 (s, 1H, Hc); 6.57 (dd, 1H, Hb); 5.60 (dd, 1H, Ha);
4.99 (dd, 1H, Ha). 13C NMR (300.30 MHz, DMSO-d6, 25 °C): δ
(ppm) = 136 (Cd); 135 (Cc); 128 (Cb); 120 (Cq); 110 (Ca).
Polymerization of 4-Vinyl Imidazole (PVIm) via RAFT Polymer-
ization. VIm was polymerized following a literature procedure.32 VIm
(0.25 g, 2.7 × 10−3 mol), RAFT Agent (2-(((dodecylthio)-
carbonothioyl)thio)-2-methylpropanoic acid) (5.0 mg, 1.3 × 10−5
mol) and ACPA (0.9 mg, 3.3 × 10−6 mol) as initiator, were dissolved
in glacial acetic acid (3.3 mL) to give [VIm]:[RAFT]:[ACPA] =
200:1:0.25. The solution was transferred to a 10 mL Schlenk vial and
deoxygenated by four freeze−pump−thaw cycles, and then placed in
an oil bath at 70 °C. The polymerization was stopped after 24 h by
cooling the vial and opening to air. The solution was dialyzed
(MWCO = 3500 g mol−1) against Milli-Q water and dried under
vacuum to give poly(4-vinyl imidazole) (PVIm) as a pale yellow low
Synthesis of Amphiphilic Block Copolymer P(NAMI-A)-PPEGMEA.
In a typical preparation, [DMSO2H][trans-RuCl4(DMSO)2 (4.6 mg,
7.9 × 10−6 mol) was crushed to an orange powder and dissolved in
methanol (1 mL). PVIm-PPEGMEA (5.1 mg, 3.2 × 10−5 mol) was
also dissolved in MeOH (1 mL). The solutions were combined to give
a bright yellow solution and analyzed via UV−vis immediately. For
cytotoxicity assays, the solution was slowly diluted (1 h) and then
dialyzed against MQ water for 1 h and used immediately. For solid-
state NMR, the solution was vacuum-dried to give an orange solid.
In Vitro Cytotoxicity Assay. Human ovarian carcinoma A2780 and
Ovcar-3, and pancreatic AsPC-1 cells were cultured in 75 cm2 tissue
culture flasks with RPMI 1640 medium supplemented with 10% fetal
bovine serum, 4 mM glutamine, 100 U/mL penicillin, 100 μg mL−1
streptomycin, and 1 mM sodium pyruvate at 37 °C under an
atmosphere of 5% CO2. After reaching 70% confluence, the cells were
washed with phosphate buffered saline (PBS) and collected by
trypsin/EDTA treatment. The cells were seeded in 96-well cell culture
plates at 4000 cells per well and cultured at 37 °C for one day. The
medium in the cell culture plate was discarded and 100 μL fresh 2 ×
concentrated RPMI 1640 serum medium was added. The samples
were added into the plate at 100 μL per well for 72 h. Before loading
onto the cells, solutions were sterilized by UV irradiation for 15 min in
a biosafety cabinet and then serially diluted (2 × dilution) with sterile
water and incubated for 2 h at room temperature. The working
concentration of ethanol in the culture medium was adjusted to
1 v/v % in case of PVIm and P(NAMI-A) samples. One v/v % ethanol
did not show significant influence on the viability of A2780, Ovcar-3
and AsPC-1 cells.
The cell viability was measured using a WST-1 assay (Roche
Diagnostics). This is a colorimetric assay for the quantification of cell
viability and proliferation that is based on the cleavage of a tetrazolium
salt (WST-1) by mitochondrial dehydrogenases in viable cells.
Increased enzyme activity leads to an increase in the amount of
formazan dye, which is measured with a microplate reader. After
incubation for three days, the culture medium was removed and 100
μL fresh medium was added along with 10 μL WST-1. The plates were
then incubated for an additional 4 h at 37 °C. After incubation, the
absorbance of the samples against the background control on a
Benchmark Microplate Reader (Bio-Rad) was obtained at a wave-
length of 440 nm with a reference wavelength of 650 nm. Four wells
density solid (0.16 g). Total reaction time = 24 h, xNMR = 76%, xMass
=
1
65%, Mn,theo = 14 300 g mol−1. H NMR (300.30 MHz, MeOD, 25
°C): δ (ppm) = 7.52 (broad, 1H, Hd); 6.35 (broad, 1H, Hc); 2.40−
1.90 (broad, 1H, Hb); 1.62 (broad, 2H, Ha).
Chain Extension of 4-Vinyl Imidazole with Poly(ethylene glycol)
Methyl Ether Acrylate (PVIm-PPEGMEA) via RAFT Polymerization.
PEGMEA (0.1 g, 2.3 × 10−4 mol), VIm MacroRAFT (40 mg, 4.6 ×
10−6 mol), and AIBN (0.2 mg, 9.3 × 10−7 mol) as initiator, were
dissolved in MeOH (0.45 mL) to give [PEGMEA]:[PVIm]:[AIBN] =
50:1:0.2. The solution was transferred to a 10 mL Schlenk vial and
deoxygenated by five freeze−pump−thaw cycles. The solution was
subsequently transferred to a 2 mm FT-IR quartz cuvette in a
glovebox, and placed in the preheated FT-IR cell at 65 °C. The
polymerization was monitored over time and stopped after 4 h by
cooling the cuvette and opening to air. The solution was dialyzed
(MWCO = 3500 g mol−1) against Milli-Q water and dried under
vacuum to give poly(4-vinyl imidazole)-b-poly(poly(ethylene glycol)
methyl ether acrylate) (PVIm-PEGMEA) as a pale yellow rubbery
solid (96 mg). Total reaction time = 4 h, xFT‑NIR = 48%, xMass = 63%,
1
Mn,theo = 28 700 g mol−1. H NMR (300.30 MHz, MeOD, 25 °C): δ
(ppm) = 7.54 (broad, 1H, Hd); 6.42 (broad, 1H, Hc); 4.23 (broad, 2H,
Hg), 3.63 (broad, PEG), 3.53 (broad, 2H, Hh), 3.32 (broad, 3H, H1),
2.79 (broad, 1H, Hf), 2.40−1.90 (broad, 1H, Hb); 1.60 (broad, 4H, Ha
and He).
Synthesis of [DMSO2H][trans-RuCl4(DMSO)2] (Ru Precursor).
[DMSO2H][trans-RuCl4(DMSO)2] (Ru precursor) was synthesized
following a literature protocol.35,36 Ruthenium trichloride hydrate
(RuCl3.H2O) (1.5 g, 7 × 10−3 mol) was combined with dimethyl
sulfoxide (DMSO) (7.0 mL) in a 100 mL round-bottom flask.
Hydrochloric acid (HCl) (32%) (1.2 mL) was added and the solution
heated to 80 °C for 20 min with vigorous stirring. The solution turned
a deep red color and the temperature was increased to 100 °C and
kept at this temperature for 20 min, at which time the solution was an
orange-red color. The solution was cooled to room temperature with
1648
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
under each condition were used for the measurement to calculate the
means and standard deviations. All cytotoxicity data are reported as
mean standard deviation. A two-tailed student’s t test was executed
to reveal the statistical differences. A p-value less than 0.05 was
considered statistically significant.
RESULTS AND DISCUSSION
Synthesis and Polymerization of 4-Vinyl Imidazole.
■
Consistent with the observations by Allen et al.32 we were
unable to synthesize the water-soluble28 poly(N-vinyl imida-
zole) in a controlled manner by RAFT polymerization since it
forms a highly reactive and unstable propagating radical due to
the absence of resonance stabilization. In contrast, 4-vinyl
imidazole (VIm) could be polymerized via RAFT, using the
trithiocarbonate RAFT Agent (2-(((dodecylthio)-
carbonothioyl)thio)-2-methylpropanoic acid) and glacial acetic
acid as a distinctive solvent, due to the increased radical
stability. VIm was synthesized and subsequently polymerized
(Scheme S1, Supporting Information) following a modified
literature procedure.32 Urocanic acid was heated under reduced
pressure to decarboxylate and distill as a clear liquid, which
solidified upon cooling to give a waxy colorless solid. The
Antimetastatic Assay. The highly invasive MDA-MB-231 human
breast cancer cell line was obtained from the Australia Cell Bank. The
MCF-7 human breast cancer cell line was kindly supplied by the Lowy
Cancer Research Centre (UNSW, Sydney, Australia). The non-
tumorigenic Chinese hamster ovary (CHO) cells were kindly supplied
by St. George Hospital. The three cell lines were cultured in 75 cm2
tissue culture flasks with a complete medium composed of Dulbecco’s
Modified Eagle Medium (DMEM) supplemented with 10% FBS, 4
mM glutamine, 100 U mL−1 penicillin, 100 μg·mL−1 streptomycin, 1
mM sodium pyruvate and 1% MEM nonessential amino acid at 37 °C
under an atmosphere of 5% CO2. After reaching 70% confluence, the
cells were washed with PBS and collected by trypsin/EDTA treatment.
The cells were seeded in 6-well cell culture plates and incubated for 1
day before P(NAMI-A)-PPEGMEA, NAMI-A, and PVIm-PPEGMEA
treatment. The Ru concentration in P(NAMI-A)-PPEGMEA and
NAM-A was 5 μM. The polymer concentrations in both P(NAMI-A)-
PPEGMEA and PVIm-PPEGMEA was 3.5 μg·mL−1.
Chemotaxis and Haptotaxis. The migratory ability resulting from
a haptotactic or a chemotactic stimulus was measured in 24-well
Millicell hanging cell culture inserts (Millipore, Billerica, MA) by the
methods of Bergamo et al. with some modifications.38 In the
haptotaxis assay the lower surface of a polyethylene terephthalate
(PET) filter (8-μm pore size) was coated with 10 μg·mL−1 fibronectin
and left in a humidified cell culture chamber at 4 °C overnight, then
washed with PBS before cell seeding. In the chemotaxis assay, inserts
were used without coating. Cells were treated for 1 h with all the
samples in the complete medium. Then the cells were washed with
PBS three times, collected by trypsinisation and centrifugation, and
resuspended in serum-free medium supplemented with 0.1% w/v
bovine serum albumin (BSA). A 0.2 mL cell suspension at 5 × 105
cells·mL−1 was seeded in the upper compartment of each insert. The
lower compartment was filled with serum free medium supplemented
with 0.1% w/v BSA and with complete medium for the haptotaxis and
the chemotaxis assay, respectively. Cells were incubated at 37 °C for
24 h, and then the cells on the upper surface of the filters were
removed with a cotton swab. The migrated cells in the lower surface
were washed with PBS, trypsinized with 0.3 mL of trypsin/EDTA, and
counted by a hemacytometer.
1
product was confirmed by H NMR spectroscopy (Figure S1).
The monomer was subsequently polymerized via RAFT
polymerization (Scheme 1 and Figure S2, Supporting
Scheme 1. Synthesis of the Amphiphilic Block Copolymer
P(NAMI-A)-PPEGMEA Using 2-
(((Dodecylthio)carbonothioyl)thio)-2-methylpropanoic
Acid RAFT Agent and Micellization in Water
Invasion Assay. 24-well Millicell hanging cell culture inserts (8-μm
pore size) were coated with 50 μL of 1 mg.mL−1 Matrigel (diluted in
serum-free DMEM medium; BD Sciences, Franklin Lakes, NJ) and air-
dried overnight at room temperature. The filters were reconstituted
with serum free medium immediately before use. Then, 2.5 × 105 cells
in 0.2 mL of serum free medium (with 0.1% BSA) were added to the
upper chamber and the lower compartment was filled with complete
medium. The cells were allowed to invade for 48 h at 37 °C in a CO2
incubator. Then the cells on the upper surface of the filters were
removed with a cotton swab. The invading cells on the lower surface
were washed with PBS, trypsinized with 0.3 mL trypsin/EDTA and
counted by a hemacytometer.
Cell Viability Assay. The effects on the cell viability of P(NAMI-A)-
PPEGMEA, NAMI-A and PVIm-PPEGMEA were evaluated by a
WST-1 assay. The cells were seeded on 96-well plates at 4 000 cells/
well and incubated for 24 h. The samples were loaded to the cell with
the complete medium and cultured for 72 h at 37 °C in the incubator.
The viability was measured by the WST-1 assay as described above.
Cells without sample treatment were used as controls. Each assay
was done in triplicate. Data represent means SD. A one way analysis
of variance was used to reveal the statistical difference among different
groups. A Dunnett’s post hoc test was used to compare the difference
between samples and the control. A p-value less than 0.05 was
considered a statistically significant difference.
Information) to give a colorless solid that was soluble in
methanol, DMSO and a water/ethanol mixture. The polymer-
ization was monitored using water-based SEC (Figure S3,
Supporting Information) and shown to be consistent with
previous observations by Allen et al.32 as the molecular weight,
shown by the change in retention time, of the polymer
increased over time.
The [DMSO2H][trans-RuIIICl4(DMSO)2] (Ru Precursor)
complex (Scheme 1 and Scheme S2, Supporting Information)
1649
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
that is used in the synthesis of the antimetastatic coordination
compound NAMI-A was synthesized following a literature
procedure35,36 and confirmed by X-ray crystallography to be
consistent with literature (Table S1, Supporting Information).36
Because of the unfavorable occurrence of two trans S-bonded
DMSO’s in the precursor complex, substitution of at least one
of them with a stronger σ-donor ligand is a relatively simple
task.36 Alessio et al.35 reported that one DMSO ligand can
rapidly be replaced in an organic solvent at room temperature
using a slight excess of a nitrogen ligand. Hence, NAMI-A
(Scheme 2) was synthesized following a literature procedure,37
counterion imparts more favorable chemical properties on
NAMI-A than other counterions, for example Na+).5
Consistent with the synthesis of NAMI-A, the Ru precursor
was first dissolved in the selected solvent (methanol for
P(NAMI-A) and acetone for M-NAMI-A) giving a bright
orange solution. The imidazole (PVIm or 4-methyl imidazole)
was subsequently added while vigorously stirring, immediately
producing a yellow-orange suspended precipitate. The solid was
washed to remove any unreacted materials and dried under
vacuum. The UV−vis absorbance maxima of the products in
methanol were compared (Figure 1). A clear shift from the Ru
Scheme 2. Synthesis of (ImH)[RuIIICl4(Im)(S-DMSO)]
(NAMI-A) and M-NAMI-A in Acetone at Room
a
Temperature
Figure 1. UV−vis spectrum in methanol at 25 °C. The attachment of
imidazole to RuIII elicits a clear shift in absorbance maxima from 375
to 400 nm.
precursor at 375 nm to NAMI-A, M-NAMI-A and P(NAMI-A)
at 400 nm, indicated complete substitution of one DMSO
ligand with an imidazole, producing the desired products.
Thermogravimetric analysis was used to confirm the amount
of ruthenium in each sample (Table 1). A dry polymer sample
Table 1. Thermogravimetric Analysis of NAMI-A
a
Analogues
Ru
precursor
c
sample
NAMI-A M-NAMI-A P(NAMI-A)
residue (%) using N2
residue (%) using O2
21
23
21
13
21
16
−
24
29
23
b
Ru (%)
18
22
17
a
P(NAMI-A) was prepared in ethanol. The photos depict the orange
d
Ru (% theo)
18.16
22.11
20.83
colored crystals of the precursor and the mustard-yellow-colored
product.
a
Samples were analyzed using both nitrogen and oxygen to degrade
b
the samples. Ru fraction = (mass residue using O2 × 0.759)/initial
mass. Ru precursor = [DMSO2H][trans-RuCl4(DMSO)2]. Calcu-
c
d
lated from the structural formula.
1
using an excess of imidazole, and characterized via H NMR
spectroscopy (Figure S4) and confirmed to be consistent with
literature due to the very broad DMSO methyl peak at −15.2
ppm.
was loaded onto a TGA pan and heated to 1000 °C at a rate of
20 °C·min−1. Since ruthenium oxidizes to RuO2 at 600 °C,
75.9% of the residual mass can be ascribed to ruthenium,
assuming that the residue is pure RuO2.
In addition, a dry P(NAMI-A) polymer sample was
submitted for solid-state elemental analysis. The number of
imidazole repeating units was calculated from the NMR
conversion (n = 152). A spread sheet was then constructed
to determine the number of imidazole units that had ruthenium
attached, thus forming NAMI-A, assuming that for each
imidazole with ruthenium there is another imidazole that acts
as a counterion. It was found that 49% of the polymer units
were NAMI-A units (Table S2, Supporting Information). This
A polymeric form of NAMI-A [P(NAMI-A)] was synthe-
sized by combining PVIm (Mn,theo = 14 300 g·mol−1) and the
Ru Precursor in ethanol or methanol. A methylated version of
NAMI-A (using 4-methyl imidazole) was also synthesized as a
control for comparison, since the homopolymer was synthe-
sized from 4-vinyl imidazole. Following a literature procedure
for the synthesis of NAMI-A,37 the imidazole was added in a
4:1 excess to the Ru precursor, to ensure complete conjugation
of ruthenium to imidazole, and the formation of imidazole
counterions (previous research has shown that the imidazole
1650
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
is consistent with the complete conjugation (within error) of all
added ruthenium since the Ru precursor was combined with a
4:1 excess of imidazole units and two imidazole units form
NAMI-A (one conjugated to ruthenium and the other forming
the positively charged counterion). The fraction of ruthenium
calculated from TGA is also consistent with the calculated
percentage. The remaining error is due to the nature of
polymer entities and the variability between polymer chains.
The resulting structure with the calculated repeating unit is
depicted in Scheme 2.
The P(NAMI-A) polymer is barely soluble in any solvent
(including water). This low solubility in water is advantageous
as it provides an avenue to the formation of nanoparticles. One
of the primary motivations for incorporating anticancer drugs
into polymers is to increase cell uptake in the body. Since
P(NAMI-A) is insoluble in water, it can act as the hydrophobic
component of polymeric micelles, which have been shown to
have better cell uptake than linear polymers.39 Therefore, a
suitable comonomer, namely poly(ethylene glycol) methyl
ether acrylate (PEGMEA), was chosen and PVIm was used as a
macroRAFT agent and chain extended with PEGMEA (Figure
S5). The monomodal SEC trace (Figure 2) shows a clear shift
the UV−vis absorbance maxima for P(NAMI-A)-PPEGMEA,
comparable to that of NAMI-A and P(NAMI-A).
Solution NMR was used as a final analysis method to confirm
the complex formation of NAMI-A. However, this was not
possible for the macromolecular forms of NAMI-A − neither
homopolymer nor copolymer. Paramagnetism of the RuIII
nucleus combined with the inherent broadening of polymer
signals prevented a detailed solution NMR study. Solid-state
NMR was thus used to determine the homogeneity of NAMI-A
within the P(NAMI-A) and P(NAMI-A)-PPEGMEA polymers.
The presence of paramagnetic RuIII causes fast relaxation
with concomitant line broadening and paramagnetic contact
shift in the 1H and 13C solid state NMR of these materials. As a
result using chemical shift to measure incorporation and
bonding of the RuIII species is challenging. An alternative
1
method is to look at the spin−lattice relaxations of the H
(T1H) where the presence of a paramagnetic species would
1
cause enhanced relaxation of the H nuclei. The T1H was
measured by a saturation recovery experiment where the
equilibrium magnetization was first suppressed by a comb of
saturation pulses and delays after which a recovery delay was
1
introduced for the H magnetization to recover. The total
1
integrated area of the H NMR signal (normalized against M0,
the signal area for the fully recovered magnetization) was
plotted against the recovery time (ms) as seen in Figure 3. The
Figure 2. Water SEC trace of the polymerization of poly(ethylene
glycol) methyl ether acrylate in methanol at 65 °C using P(4-vinyl
imidazole) MacroRAFT agent (Mn,theo = 14 300 g mol−1). [PEGMEA]
= 0.4 M, [PEGMEA]:[PVIm]:[AIBN] = 50:1:0.2. Total reaction time
= 4 h, xFT‑NIR = 48%, Mn,theo = 28 700 g mol−1. The eluent was a Milli-
Q water/acetic acid/methanol (54:23:23) solution.
Figure 3. 1H saturation recovery experiment of NAMI-A, P(NAMI-A),
and P(NAMI-A)-PPEGMEA via solid-state NMR.
in retention time, indicative of a molecular weight change and a
successful chain extension. The absence of any low molecular
weight shoulder indicates that the PVIm MacroRAFT is quite
efficient in mediating the polymerization of PEGMEA. Since
the Mn values obtained from the water SEC traces cannot be
used due to the difference in the polymer and the PEO
calibration standards, the molecular weights used in further
discussions were calculated using conversions determined from
FT-NIR.
The Ru precursor was reacted with the copolymer P(NAMI-
A)152-PPEGMEA19 in a similar fashion to the homopolymer −
the difference being that the copolymer remained soluble in
methanol after conjugation, and it could be dried and
redissolved. Because of the difference in solubilities, the
homopolymeric drug was purified by washing with methanol,
whereas the copolymeric drug was purified by dialysis. UV−vis
was used to confirm the complexation. Figure 1 shows a shift in
data was fitted by a single or double exponential of the form
M/M0 = 1 − A exp−t/T − B exp−t/T , where A and B are the
fractions of slow and fast relaxing components with spin−lattice
relaxation times of T1A and T1B. It should be noted that in the
case of paramagnetic driven relaxation processes, the saturation
recovery behavior can be nonexponential (stretched exponen-
tial) particularly in the case of low natural abundance NMR
active nuclei such as 13C.40 However in the case of nuclei such
1A
1B
1
as H in the current materials with fast nuclear spin diffusion
and moderate MAS (12 kHz), the saturation recovery behavior
is better described by a single or double exponential curve.
For example, the 1H saturation recovery of the neat NAMI-A
complex is well fitted by a single exponential with a T1H of 0.9
ms. The extremely fast T1H is consistent with the presence of
the paramagnetic RuIII in the solid material. In the case of the
1
P(NAMI-A), the overall fast spin−lattice relaxation of the H
species, confirmed the attachment of a paramagnetic species,
1651
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
Figure 4. P(NAMI-A)-PPEGMEA Micelles were prepared by dialyzing a MeOH solution against water, and analyzed by TEM and DLS. Sample a
was drop-loaded onto a grid and air-dried. Samples b and c were drop-loaded onto the grid, air-dried, and stained with phosphotungstic acid. Scale
bar: a = 100 nm; b = 2 μm; c = 0.2 μm;.
and thus it can be inferred that ruthenium is still present as
RuIII. However the saturation recovery curve P(NAMI-A)
followed a dual exponential behavior. The faster relaxing
component with a very fast T1H relaxation time of 0.5 ms
corresponded to 45% of the polymer that is at an average
distance of ca. 1 nm from the RuIII complex, while the
comparatively slower T1H relaxation time of 12 ms
corresponded to ca. 55% of the polymer beyond nanometer
proximity of the RuIII complex. This is consistent with the
elemental analysis and TGA results, which showed that
approximately 50% of the imidazole units formed NAMI-A.
Finally, the absence of any slow relaxing components with a
T1H in the order of 100 ms to several seconds, as would be
expected for diamagnetic, amorphous, organic polymers,
implies that on a conservative 10 nm scale there is a uniform
distribution of the RuIII complex within the P(NAMI-A)
polymer matrix.
(a) The overall fast relaxation indicates that the complex is
incorporated into the polymer.
(b) The comparatively longer relaxation time (12 ms)
indicates that on the 10 nm scale, there is homogeneous
distribution of the complex.
(c) The short relaxation time (0.5 ms) corresponds to the
region within nanometer scale proximity of the complex.
It also indicates that the complex is dispersed within the
polymer and has not aggregated.
1
(d) The copolymer H spectrum consists of a broad and
narrow component with very different relaxation proper-
ties, indicative that RuIII is incorporated uniformly in a
single block of the polymer, i.e., P(NAMI-A).
After drug conjugation, the methanol solution was
immediately dialyzed against water to remove methanol and
simultaneously form micelles in solution. The dialyzed yellow
solutions were analyzed using TEM and DLS (Figure 4 and 5)
1
In the case of the P(NAMI-A)-PPEGMEA the H spectrum
of the copolymer yields two peaksone broad and one narrow.
The broad component corresponds to the rigid P(NAMI-A)
block containing the RuIII complex with a fast T1H of ca. 1.5 ms
relaxation. The narrow component corresponds to the mobile
PPEGMEA block which, as expected, has a distinctly longer
relaxation time (28 ms) than the P(NAMI-A) block. It is likely
that the PPEGMEA block is strongly coupled to the faster-
relaxing proton pool of the P(NAMI-A) block accounting for a
T1H of ca. 28 ms. It is highly unlikely that the PPEGMEA
contains RuIII since there are no binding sites in this copolymer
block to allow for the formation of the ruthenium complex.
Additionally, the UV−vis spectrum (Figure 1) confirmed the
attachment of the RuIII complex to the imidazole units in the
PVIm block, as shown by a clean shift of a single peak, and
there was no indication of RuIII interacting with the PPEGMEA
block in the polymer. However, approximately 6% of the
P(NAMI-A) block has a longer relaxation time of ca. 20 ms. We
hypothesize that this can be attributed to the region where both
blocks are connected and that the proximity to the PEG
induces motional mobility to a fraction of the P(NAMI-A)
which can enhance the relaxation times. Considering the
relatively small fraction of the P(NAMI-A) block with the
longer relaxation time, it is also possible that it has a lower RuIII
concentration. However, given the synthesis method, it is
unlikely that the RuIII would distribute heterogeneously within
the PVIm copolymer block, since it is homogeneous in the
homopolymer on the 10 nm scale. Overall, these results are
consistent with the distinct PPEGMEA and P(NAMI-A)
phases, where the predominant fraction of the PPEGMEA
phase is further from the paramagnetic centers.
Figure 5. Stability of micelles as determined by DLS (volume
distribution) in phosphate buffered saline (pH 7.4) at 37 °C over 3
days.
to determine micelle size and homogeneity. TEM shows
micelles to be a reasonable size, although a size distribution is
evident. This distribution was also apparent in the DLS
measurements, indicated by the PDI. The micelles are round in
shape, but probably slightly distorted due to the drying process
for TEM preparation. The unstained sample in Figure 4a
appears black due to the conjugation of ruthenium, which has a
high electron density. Closer inspection of Figure 4a reveals the
darker appearance of the center of the micelle due to the
location of ruthenium in the core of the micelle.
In summary, these results clearly suggest the following:
1652
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
The size of the micelle in PBS buffer was in addition
confirmed using DLS (Figure 5). Directly after dialysis of the
methanol solution against aqueous solution the size distribution
was found to be broad as indicated by the large error bars. After
an intermittent size increase, probably due to aggregate
formation, the micelle size was stable over several days with
an average hydrodynamic volume of approximately 90 nm with
a low distribution. However, the degradation of the drug during
that long period of time cannot be avoided and color changes
are visible.
In Vitro Cytotoxicity and Antimetastatic Potential. The
NAMI-A polymeric micelles were tested against ovarian A2780
and Ovcar-3 and pancreatic AsPC-1 cancer cell lines, and
compared with the drug NAMI-A. The homopolymer and
block copolymer prior to drug conjugation were also tested for
toxicity (Table 2, Figure 6). PVIm was found to be toxic, which
A statistically significant decrease in the IC50 values for
P(NAMI-A)-PPEGMEA micelles compared with NAMI-A
alone at the same ruthenium concentration, was found for all
cell lines (Table 3, Figure 7). The micelles were found to be
Table 3. IC50 Values with Respect to the Ruthenium
Concentration, against Ovarian A2780 and Ovcar-3 and
Pancreatic AsPC-1 Cancer Cell Lines
IC50 ([Ru], μM)
A2780
>50043
AsPC-1
OVCAR-3
NAMI-A literature
NAMI-A
−
−
595.6 28.5
438.7 14.8
601.7 13.5
737.8 38.0
P(NAMI-A)-
PPEGMEA
397.6 24.7
433.4 26.8
Table 2. IC50 Values with Respect to the Polymer
Concentration, against Ovarian A2780 and Ovcar-3 and
Pancreatic AsPC-1 Cancer Cell Lines
IC50 ([P], μM)
A2780
AsPC-1
OVCAR-3
PVIm
0.37
>15.37
9.77
0.51
>15.37
8.86
0.70
>15.37
9.65
PVIm-PPEGMEA
P(NAMI-A)-PPEGMEA
a
a
Polymer concentrations at IC50 in Table 3.
Figure 7. Cytotoxicity of NAMI-A and P(NAMI-A)-PPEGMEA
against ovarian A2780 and Ovcar-3 and pancreatic AsPC-1 cancer cell
lines. For P(NAMI-A)-PPEGMEA, the polymer concentration at [Ru]
= 449 and 224.5 μM is 10 and 5.0 μM respectively. Mean SD, n = 4,
(∗∗) significantly different p < 0.01, (∗∗∗) significantly different p <
0.001.
∼1.5 times better than NAMI-A at inhibiting cancer cell
growth, across the tested cell lines. Most notably, they were
active on the highly aggressive pancreatic cancer cell line.
However, IC50 values cannot be solely relied upon to determine
the efficacy of a particular drug, since there are many
determining factors for a drug to enter into clinical trials, for
example the drug’s ability to reach the target.
Figure 6. Cytotoxicity of PVIm and PVIm-PPEGMEA against ovarian
A2780 and Ovcar-3 and pancreatic AsPC-1 cancer cell lines.
It has previously been established by Sava and co-workers
that, for NAMI-A-type compounds, in vitro inhibition of tumor
invasion correlates with the in vivo inhibition of metastasis
formation.44−46 Bergamo et al.38 subsequently developed a
series of experiments that simulate in vitro the main steps of
metastatic progression: detachment from the primary tumor,
degradation of the extracellular matrix, ability to migrate, invade
and adhere to a new organ. These protocols were adapted to
assess the antimetastatic potential of P(NAMI-A)-PPEGMEA
micelles compared with NAMI-A.
The migration and invasion assays were chosen to assess the
antimetastatic potential of the ruthenium conjugated P(NAMI-
A)-PPEGMEA micelles in comparison to the antimetastatic
drug NAMI-A. The effects of NAMI-A, the block copolymer
without ruthenium, and the micelles incorporating NAMI-A
were evaluated against three cell lines characterized by differing
degrees of malignancy (MDA-MB-231 > MCF-7 > CHO).
makes P(NAMI-A) an undesirable therapeutic agent, since the
toxicity is derived from the polymer itself and not the drug.
This can be attributed to the positive charges of the imidazole
units, as positively charged polymers have previously shown
toxicity.41 The copolymer, PVIm-PPEGMEA, was completely
nontoxic to all cell lines up to a concentration of 10 μM. This
was surprising since the copolymer still contains unmodified
PVIm. The reason for this complete lack of toxicity can be
attributed to the morphology of the block copolymer, since the
homopolymer PVIm is insoluble in water. This may lead to the
micelle formation of PVIm-PPEGMEA, with positively charged
PVIm forming the core of the micelle. The sudden
disappearance of apparent cytotoxicity has been observed
earlier when a cationic polymer formed the core of a cross-
linked micelle resulting in the loss of toxicity in this assay.42
1653
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
A ruthenium concentration of 5 μM was chosen as a suitable
concentration to be used for the following experiments since
the cell viability was unaffected by the compounds at this
concentration. Higher ruthenium concentrations would lead to
significant cell deaths and the measurement of the migration
would therefore be inaccurate. Figure S6 (Supporting
Information) shows that the ruthenium compound induces
only slight and nonsignificant oscillations ( 10%) in cell
viability as compared to the relevant controls. This is necessary
in order to exclude the interference of the cytotoxicity of the
compounds in the following assays.
The influence of the compounds on the migration process of
the three cell lines MDA-MB-231 (invasive cancerous), MCF-7
(noninvasive, cancerous) and CHO (noncancerous) was
evaluated when a chemical (chemotaxis) and a contact
(haptotaxis) stimulus was applied to promote cell movement
(Figure 8 and Figure 9). The chemotactic and haptotactic
Figure 9. Effect of NAMI-A, PVIm-PPEGMEA, and P(NAMI-A)-
PPEGMEA Micelles on the haptotactic migration of cells through
polycarbonate filters. MDA-MB-231, MCF-7, and CHO cell were
treated for 1 h with the drugs where [Ru] = 5 μM. The cells were then
removed from the flasks, collected, resuspended and seeded on the
insets of Transwell cell culture chambers. Data represent cells that
after 24 h have migrated and are present on the lower surface of the
filter. Data are the percent of variation vs controls calculated from the
mean SD of one experiment performed in triplicate.
Figure 8. Effect of NAMI-A, PVIm-PPEGMEA, and P(NAMI-A)-
PPEGMEA Micelles on the chemotactic migration of cells through
polycarbonate filters. MDA-MB-231, MCF-7 and CHO cell were
treated for 1 h with the drugs where [Ru] = 5 μM. The cells were then
removed from the flasks, collected, resuspended and seeded on the
insets of Transwell cell culture chambers. Data represent cells that
after 24 h have migrated and are present on the lower surface of the
filter. Data are the percent of variation vs controls calculated from the
mean SD of one experiment performed in triplicate.
Figure 10. Effect of NAMI-A, PVIm-PPEGMEA, and P(NAMI-A)-
PPEGMEA Micelles on the invasion of cells through Matrigel. MDA-
MB-231, MCF-7, and CHO cells were treated for 1 h with the drugs
where [Ru] = 5 μM. The cells were then removed from the flasks,
collected, resuspended and seeded on inserts. Data represent cells that
after 96 h have invaded and are present on the lower surface of the
filter. Data are the percent of variation vs controls calculated from the
migration of both the MDA-MB-231 and MCF-7 cells was
inhibited more than that of the CHO cells. However, no
statistically significant effect was observed with respect to the
controls.
mean
SD of one experiment performed in triplicate. Key: (∗)
significant difference vs control, p < 0.05.
The effect on the invasive ability of these cells through
Matrigel was also assessed (Figure 10). The invasive ability of
both the MDA-MB-231 and MCF-7 cells was inhibited more
than that of the CHO cells, with respect to the controls. A
statistically significant inhibitory effect was observed for the
invasive MDA-MB-231 cell line (Figure 10).
NAMI-A and the micelles display cell type specificity,
characterized by a more pronounced effect on tumor cells,
than on the nontumorigenic CHO cells. The NAMI-A micelles
did not display any statistically significant effect for the
migration assay, although they had a more pronounced
inhibitory effect than NAMI-A. However, since the NAMI-A
micelles significantly inhibited the invasion of the highly
invasive MDA-MB-231 cells, it can be inferred that these drugs
also interfere more selectively with tumor cells with the highest
inclination to invade and metastasise. Thus, it is probable that
the antimetastatic ability of NAMI-A has been enhanced by
incorporation in polymeric nanoparticles. It would be valuable
to evaluate the antimetastatic potential of these nanoparticles at
a higher ruthenium concentration and also in vivo.
CONCLUSIONS
■
An amphiphilic block copolymer capable of self-assembling into
polymeric micelles was identified as an appropriate drug carrier
for NAMI-A. A suitable method for the synthesis of a
macromolecular NAMI-A drug was identifiedthe polymer-
ization of vinyl imidazole and subsequent addition of a
ruthenium(III) precursor complex. A water-soluble block
copolymer was designed to increase biocompatibility and cell
uptake through the formation of micelles. On average, across
the tested cell lines, a 1.5 times increase in toxicity was found
1654
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655

Macromolecules
Article
(20) Nishiyama, N.; Yokoyama, M.; Aoyagi, T.; Okano, T.; Sakurai,
Y.; Kataoka, K. Langmuir 1999, 15, 377−83.
for the NAMI-A block copolymer micelles when compared to
the NAMI-A molecule. Furthermore, the micelles were shown
to have an improved antimetastatic potential than NAMI-A.
Further work will entail a detailed assessment of the
antimetastatic effects of the macromolecular NAMI-A chemo-
therapeutic in vivo.
(21) Nishiyama, N.; Kataoka, K. J. Controlled Release 2001, 74, 83−
94.
(22) Nishiyama, N.; Okazaki, S.; Cabral, H.; Miyamoto, M.; Kato, Y.;
Sugiyama, Y.; Nishio, K.; Matsumura, Y.; Kataoka, K. Cancer Res. 2003,
63, 8977−83.
(23) Bontha, S.; Kabanov, A. V.; Bronich, T. K. J. Controlled Release
2006, 114, 163−74.
ASSOCIATED CONTENT
* Supporting Information
■
(24) Matsumoto, A.; Matsukawa, Y.; Suzuki, T.; Yoshino, H. J.
Controlled Release 2005, 106, 172−80.
S
Schemes showing the synthesis and polymerization of VIm and
the synthesis of the Ru precursor, figures showing NMR
spectra, SEC traces, and the viablility of cells, and tables of
crystallographic and reaction data. This material is available free
of charge via the Internet at http://pubs.acs.org.
(25) Anderson, E. B.; Long, T. E. Polymer 2010, 51, 2447−54.
(26) Hakamatani, T.; Asayama, S.; Kawakami, H. Nucleic Acids Symp.
Ser. 2008, 52, 677−8.
(27) Srivastava, A.; Waite, J. H.; Stucky, G. D.; Mikhailovsky, A.
Macromolecules 2009, 42, 2168−76.
(28) Asayama, S.; Sekine, T.; Kawakami, H.; Nagaoka, S. Bioconjugate
Chem. 2007, 18, 1662−7.
AUTHOR INFORMATION
Corresponding Author
*(M.H.S.) E-mail: M.stenzel@unsw.edu.au. Fax: +61-2-
93856250. Telephone: +61-93854344.
■
(29) Green, M. D.; Allen, M. H.; Dennis, J. M.; La Cruz, D. S.; Gao,
R.; Winey, K. I.; Long, T. E. Eur. Polym. J. 2011, 47, 486−96.
(30) Pekel, N.; Gu, O. Colloid Polym. Sci. 1999, 573, 570−3.
(31) Yuan, J.; Schlaad, H.; Giordano, C.; Antonietti, M. Eur. Polym. J.
2011, 47, 772−81.
Notes
The authors declare no competing financial interest.
(32) Allen, M. H.; Hemp, S. T.; Smith, A. E.; Long, T. E.
Macromolecules 2012, 45, 3669−76.
(33) Skey, J.; O’Reilly, R. K. Chem. Commun. (Cambridge, U.K.)
2008, 44, 4183−5.
ACKNOWLEDGMENTS
■
We would like to thank Dr. Andrew Gregory for preparing the
RAFT Agent, Dr. Mohan Bhadbhade at the Solid State and
Elemental Analysis Unit, UNSW, for the X-ray crystallography
analysis of the complexes and The Campbell Microanalytical
Laboratory at the University of Otago in New Zealand for the
solid-state elemental analysis of the polymer.
(34) Piotto, M.; Saudek, V.; Sklenar
661−5.
́ ̌
, V. J. Biomol. NMR 1992, 2,
(35) Alessio, E.; Balducci, G.; Lutman, A.; Mestroni, G.; Calligaris,
M.; Attia, W. M. Inorg. Chim. Acta 1993, 203, 205−17.
(36) Alessio, E.; Balducci, G.; Calligaris, M.; Costa, G.; Attia, W. M.;
Mestroni, G. Inorg. Chem. 1991, 30, 609−18.
(37) Mestroni, G.; Alessio, E.; Sava, G. Salts of anionic complexes of
RU (III), as antimetastatic and antineoplastic agents. US006221905B1,
2001.
REFERENCES
■
(1) Dyson, P. J.; Sava, G. Dalton Trans. 2006, 35, 1929−33.
(2) Morris, R. R. E.; Aird, R. E.; del Socorro Murdoch, P.; Chen, H.;
Cummings, J.; Hughes, N. D.; Parsons, S.; Parkin, A.; Boyd, G.;
Jodrell, D. I.; Sadler, P. J. J. Med. Chem. 2001, 44, 3616−21.
(3) Antonarakis, E. S.; Emadi, A. Cancer Chemother. Pharm. 2010, 66,
1−9.
(38) Bergamo, A.; Masi, A.; Dyson, P. J.; Sava, G. Int. J. Oncol. 2008,
33, 1281−1289.
(39) Barz, M.; Luxenhofer, R.; Zentel, R.; Kabanov, A. V. Biomaterials
2009, 30, 5682−90.
(40) Panich, A. M.; Furman, G. B. Diamond Relat. Mater. 2012, 23,
157−161.
(4) Allardyce, C. S.; Dyson, P. J. Platinum Met. Rev. 2001, 45, 62−9.
(5) Sava, G.; Capozzi, I.; Clerici, K.; Gagliardi, G.; Alessio, E.;
Mestroni, G. Clin. Exp. Metastasis 1998, 16, 371−9.
(6) Bergamo, A.; Gaiddon, C.; Schellens, J. H. M.; Beijnen, J. H.;
Sava, G. J. Inorg. Biochem. 2012, 106, 90−9.
(41) Lv, H.; Zhang, S.; Wang, B.; Cui, S.; Yan, J. J. Controlled Release
2006, 114, 100−9.
(42) Zhang, L.; Nguyen, T. L. U.; Bernard, J.; Davis, T. P.; Barner-
Kowollik, C.; Stenzel, M. H. Biomacromolecules 2007, 8, 2890−901.
(43) Groessl, M.; Zava, O.; Dyson, P. J. Metallomics 2011, 3, 591−9.
(44) Alessio, E.; Mestroni, G.; Bergamo, A.; Sava, G. Curr. Top. Med.
Chem. 2004, 4, 1525−35.
(7) Kostova, I. Curr. Med. Chem. 2006, 13, 1085−1107.
(8) Groessl, M.; Hartinger, C. G.; Dyson, P. J.; Keppler, B. K. J. Inorg.
Biochem. 2008, 102, 1060−5.
(45) Bratsos, I.; Jedner, S.; Gianferrara, T.; Alessio, E. Chimia
(Aarau) 2007, 61, 692−7.
(46) Bergamo, A.; Gagliardi, R.; Scarcia, V.; Furlani, A.; Alessio, E.;
Mestroni, G.; Sava, G.; Foundation, C.; Sciences, B. J. Pharm. Exp.
Ther. 1999, 289, 559−64.
(9) Huynh, V. T.; Quek, J. Y.; de Souza, P. L.; Stenzel, M. H.
Biomacromolecules 2012, 13, 1010−23.
(10) Pearson, S.; Scarano, W.; Stenzel, M. H. Chem. Commun.
(Cambridge, U.K.) 2012, 48, 4695−7.
(11) Blunden, B. M.; Thomas, D. S.; Stenzel, M. H. Polym. Chem.
2012, 3, 2964−75.
(12) Blunden, B. M.; Lu, H.; Stenzel, M. H. Biomacromolecules 2013,
14, 4177−88.
(13) Savic,
343−55.
́
R.; Eisenberg, A.; Maysinger, D. J. Drug Target 2006, 14,
(14) Stenzel, M. H. Chem. Commun. (Cambridge, U.K.) 2008, 44,
3486−503.
(15) Matsumura, Y.; Maeda, H. Cancer Res. 1986, 46, 6387−92.
(16) Ambade, A. V.; Savariar, E. N.; Thayumanavan, S. Mol.
Pharmaceutics 2005, 2, 264−72.
(17) Mikhail, A. S.; Allen, C. Biomacromolecules 2010, 11, 1273−80.
(18) Cabral, H.; Nishiyama, N.; Okazaki, S.; Koyama, H.; Kataoka, K.
J. Controlled Release 2005, 101, 223−32.
(19) Cabral, H.; Nishiyama, N.; Kataoka, K. J. Controlled Release
2007, 121, 146−55.
1655
dx.doi.org/10.1021/ma402078d | Macromolecules 2014, 47, 1646−1655