Micelles based on gold-glycopolymer complexes as new chemotherapy drug delivery agents

Article abstract of DOI:10.1039/c2cc30510k

Polymeric versions of deacetylated auranofin, a gold complex with a sugar ligand, were prepared by post-modifying RAFT glycopolymers. Micellisation of a block copolymer containing pendant Au(i) units produced nanoparticles with an increased anti-proliferative effect against OVCAR-3 human ovarian carcinoma cells.

Full text of DOI:10.1039/c2cc30510k

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 4695–4697
www.rsc.org/chemcomm
COMMUNICATION
Micelles based on gold-glycopolymer complexes as new chemotherapy
drug delivery agentsw
Samuel Pearson, Wei Scarano and Martina H. Stenzel*
Received 22nd January 2012, Accepted 9th March 2012
DOI: 10.1039/c2cc30510k
Polymeric versions of deacetylated auranofin, a gold complex
with a sugar ligand, were prepared by post-modifying RAFT
glycopolymers. Micellisation of a block copolymer containing
pendant Au(I) units produced nanoparticles with an increased
anti-proliferative effect against OVCAR-3 human ovarian
carcinoma cells.
Auranofin has been identified as one of the most potent and
selective inhibitors of the mitochondrial protein thioredoxin
reductase,3 an important new drug target central to the
progression of several chronic diseases including rheumatoid
arthritis and some cancers.4 Increased expression of thioredoxin
reductase is linked to proliferant tumour cell growth, suppressed
apoptosis, and resistance to conventional chemotherapy.5
Conversely, inhibition of thioredoxin reductase causes a
build-up of hydrogen peroxide and oxidized thioredoxins
which act on a variety of targets within the mitochondria to
induce cell death.6 Direct stimulation of this mitochondrial cell
death pathway reduces the cells’ ability to establish resistance,
potentially overcoming a key challenge presented by conven-
tional chemotherapy.7 For example, both normal and cisplatin-
resistant human ovarian cancer cells have been shown to be
highly susceptible to auranofin, whereas cisplatin was ineffective
against the resistant cell line.8 Auranofin therefore displays
promising attributes as a potential chemotherapy drug able to
overcome the resistance and selectivity problems which plague
conventional chemotherapy agents.
Auranofin (1, Scheme 1) is a gold(I) complex which has been
used extensively to treat rheumatoid arthritis for the past four
decades,1 but more recently its activity against several tumour
cell lines has generated great interest for its potential use as a
chemotherapy agent.2
Despite its pronounced antitumour activity in vitro,9
however, auranofin is highly reactive towards protein thiols
and undergoes rapid ligand displacement reactions in vivo,10
primarily with serum albumin.11 This rapid ligand displacement
severely limits the applicability of auranofin as an antitumour
agent in vivo. A range of alternative Au(I) complexes with thiol-
free ligands such as chelating diphosphines have demonstrated
stability in the presence of thiols12 and high activity against
several human cisplatin-resistant ovarian cancer cell lines, but
activity closely correlated with in vivo hepatotoxicity13 and
reduced solubility.14
Rather than seeking alternative complexes, we propose that
a micellar system containing pendant auranofin-like groups in
the core and a hydrophilic shell will offer the dual advantages
of protecting the auranofin from interacting with serum
proteins, and accumulating preferentially in tumour tissue
via the enhanced permeability and retention effect15 which will
aid selective uptake by tumour cells.
Scheme 1 (a) Vinyl acrylate, Novozym 435, acetone, 5 A molecular
sieves, 50 1C, 5 days; (b) Chlorobenzene, AIBN, RAFT agent 5, 60 1C,
7 h; (c) DMSO, DTT, 0 1C, 24 h; (d) DMAc, NEt3, AuPEt3, 0 1C, 24 h;
(e) RAFT agent 5, DMF, AIBN, 60 1C, 70 min; (f) DMF, AIBN, 4, 60 1C,
9 h; (g) DMAc, DTT, 0 1C, 24 h; (h) DMAc, TEA, AuPEt3Cl, 0 1C, 24 h.
Inset; TEM image of the self-assembled poly(HEA)-b-poly(4-AuPEt3).
We have designed a glycomonomer which mimics the
structure of deacetylated auranofin (2, Scheme 1) whose
antitumour potency is comparable to that of auranofin.16
The monomer (4, Scheme 1) was synthesized in 6 steps from
D-glucose with the anomeric thiol protected with a pyridyl disulfide
group, and an enzymatic approach chosen to regioselectively
Centre for Advanced Macromolecular Design, University of
New South Wales, Sydney, Australia. E-mail: M.Stenzel@unsw.edu.au;
Fax: + 61-2-93856250; Tel: + 61-2-93854344
w Electronic supplementary information (ESI) available: Syntheses,
NMR and TGA characterisations, details on chain transfer experiment,
complexation efficiency calculations. See DOI: 10.1039/c2cc30510k
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4695–4697 4695

View Online
functionalise the 6-position of the sugar with an acrylate
group.17 This reaction using vinyl acrylate was catalysed by
Novozym 435, a lipase derived from Candida antarctica, and
5 A molecular sieves were used to absorb the acetaldehyde
byproduct and thereby remove it from the equilibrium to
maximise the yield (see ESIw).18 RAFT polymerization was
identified as a suitable technique to polymerize monomer 4
due to its robustness in the presence of many functional groups
and the non-toxicity of most RAFT agents.19 Polymerization
of 4 was performed using the trithiocarbonate RAFT agent
3-benzylsulfanylthiocarbonylsulfanylpropionic acid (5, Scheme 1)
to generate a homopolymer (poly(4), Scheme 1) capable of
forming a polymeric analogue of deacetylated auranofin. DMF
and DMAc were effective solvents for 4, but some yellowing of
the polymerization solution at longer reaction times encouraged
a switch to chlorobenzene given that the solubility of 4 was
limited in other more commonly used polymerization solvents
such as methanol, ethanol, isopropanol, acetone, and DMSO.
After a polymerization time of 7 h ([M]0 = 0.8 M) a conversion
of around 40% was reached (Fig. S1, ESIw).
The resulting homopolymer of 4 was reasonably well defined
(Mn,theor = 14.1 kDa, PDI 1.49) with slight broadening
attributed to side reactions involving the pyridyl disulfide
protecting group. A chain transfer experiment using its
disulfide precursor (3, Scheme 1) gave a chain transfer constant
of 2.8 Â 10À3 for the disulfide group (Fig. S2–S4, ESIw), which
is significant enough to influence the breadth of the molecular
weight distributions. However, a marked narrowing of the
molecular weight distribution to a PDI of 1.23 and a corres-
ponding shift to higher molecular weight (Fig. 1) was achieved
when the pyridyl disulfide groups were cleaved using
D,L-dithiothreitol (DTT) to generate free thiols on the sugar
units, indicating that some broadening of poly(4) might also
have been the result of disulfide bridge formation during the
polymerization. The shift to higher molecular weight suggests
that the thiol-containing polymer possesses a higher hydro-
dynamic radius than its precursor.
AuPEt3Cl in DMAc was added slowly under a nitrogen
atmosphere. The complexed polymer poly(4-AuPEt3) was
purified by three precipitations into cold EtOAc (in which
AuPEt3Cl is soluble). The complexation was accompanied by
a shift towards higher molecular weight as determined by SEC
(Fig. 1).
1H NMR spectroscopy (Fig. 2) showed the disappearance
of pyridyl proton peaks in the final complex, demonstrating
that the cleavage of the pyridyl disulfide protecting groups
occurred almost quantitatively, and the emergence of multi-
plets at 1.87 ppm and 1.11 ppm confirmed the presence of the
ethyl groups of the triethylphospine-containing Au(I) units.
Proton peaks were assigned with the aid of {1H–13C}HSQC
correlation measurements (Fig. S5–S6, ESIw). In addition, the
31P NMR spectrum shows a single broad peak at 38.92 ppm
characteristic of the triethyl phosphine ligands distributed along
the polymer (Inset of Fig. 2b), with the breadth arising due to
the slightly different electronic environments experienced by
each pendant group. With a phosphorous chemical shift very
close to that of 2 (Inset of Fig. 2c; the slight difference in
chemical shift attributable to the differing NMR solvents) and
an absence of impurity peaks, it is confirmed that the complexa-
tion affords the desired product efficiently with no side reactions.
Thermogravimetric analysis (TGA) of poly(4-AuPEt3)
using air as the furnace gas (Fig. S7, ESIw) generated a residual
mass of 29% at 800 1C corresponding to elemental gold
derived from the polymeric complex. An identical analysis of
the poly(4-SH) precursor confirmed that all mass contributed
by the polymer was oxidized to leave zero residual mass.
Assuming that every pendant group in poly(4-SH) contained
a free thiol to which Au(I) could complex, the efficiency of the
complexation step was 72% (see ESIw for calculations).
After a successful proof of concept in complexing AuPEt3Cl
to the thiol-bearing poly(4-SH) to generate a polymeric form
of 2, the strategy was extended to a block copolymer system in
which a trithiocarbonate macroRAFT agent of 2-hydroxy-
ethyl acrylate (HEA) was chain extended with 4 to produce a
poly(HEA)-b-poly(4) block copolymer (Scheme 1).
The thiol-containing polymer poly(4-SH) was purified by
two precipitations into cold EtOAc before the isolated powder
was immediately dissolved in DMAc and purged with nitrogen to
prevent disulfide formation before the subsequent complexation.
It was also important to ensure no DTT was present prior to the
addition of AuPEt3Cl to avoid side reactions in the following
step. Triethylamine (TEA) was added to deprotonate the anomeric
thiols, and a degassed solution containing 1.2 equivalents of
The block copolymer displayed a slight broadening of the
molecular weight distribution compared to the poly(HEA)
Fig. 1 SEC traces of (a) Poly(4), Mn,SEC = 15.0 kDa, PDI = 1.49;
(b) Poly(4-SH), Mn,SEC = 28.2 kDa, PDI = 1.23; and (c) Poly(4-AuPEt3),
Mn,SEC = 40.2 kDa, PDI = 1.24.
Fig. 2 1H NMR spectra of (a) Poly(4) in d6-DMSO; (b) Poly(4-AuPEt3)
in d6-DMSO; and (c) Deacetylated auranofin 2 in D2O; 31P NMR spectra
of (b) and (c) are inset.
c
4696 Chem. Commun., 2012, 48, 4695–4697
This journal is The Royal Society of Chemistry 2012

View Online
the mitochondria, higher activity in our micellar system is very
promising for the potential development of this polymer-gold
system into a chemotherapy treatment.
The increased activity can be explained via the different
route of cell entry. Whilst low molecular weight drugs such as
2 diffuse through the cell membrane, nanoparticles are taken
up by endocytosis. This route not only provides a more
efficient entry pathway, but also offers the possibility of
by-passing the drug resistance mechanism induced by small
molecules.20
It has been demonstrated that glycopolymers bearing pro-
tected thiosugar units can be efficiently converted to polymeric
Au(I) complexes. In micellar form, the block copolymer system
displayed higher activity against OVCAR-3 cells than its small
molecule analogue, offering a promising alternative delivery
mechanism which may overcome the stability and toxicity
issues faced by discrete Au(I) based chemotherapeutics.
We are currently investigating the protection offered by the
micellar system in preventing ligand displacement with serum
proteins, and the controlled synthesis of related polymer-gold
systems.
Fig. 3 Proliferation of OVCAR-3 human ovarian carcinoma cells in
the presence of 2 (closed squares) and poly(HEA)-b-poly(4-AuPEt3)
(open triangles) as a function of Au(I) content.
macroRAFT agent, but after purification by dialysis and
cleavage of the pyridyl disulfide groups using DTT, the
molecular weight distribution narrowed and shifted to higher
molecular weight in a similar fashion to the homopolymer
system (Fig. S8, ESIw). Complexation of AuPEt3Cl to the
purified block copolymer using the same approach generated
an amphiphilic block copolymer with the AuPEt3 complexed
to the thiol units of the pendant sugars. Based on TGA
analysis (Fig. S9, ESIw), the block copolymer contained 20%
gold by mass which corresponds to a loading efficiency of
approximately 74%. It was hoped that the introduction of
the AuPEt3 group would render the Au(I)-containing block
hydrophobic relative to the readily water soluble poly(HEA)
block and thereby encourage self-assembly into a core-shell
structure in an aqueous environment. Indeed, controlled addition
of water to a DMSO solution of the pure block copolymer
followed by dialysis against water produced micelles with a mean
diameter of approximately 75 nm by dynamic light scattering
(DLS, Fig. S10, ESIw). Transmission electron microscopy (TEM)
confirmed the presence of spherical structures containing distinct
gold-rich domains which strongly scattered the electron beam
and generated excellent contrast in the images (Inset, Scheme 1).
The activity of the micellar poly(HEA)-b-poly(4-AuPEt3)
system against tumour cells was evaluated by measuring the
proliferation OVCAR-3 human ovarian carcinoma cells in the
presence of either the discrete drug 2 or its micellar analogue at
increasing concentrations (Fig. 3). It was not deemed appropriate
to use either poly(HEA)-b-poly(4) or poly(HEA)-b-poly(4-SH)
precursors as negative controls, because neither the pyridyl
disulfide groups nor the free thiols in each would constitute a
valid reference in the same way that an unloaded micelle serves as
the negative control for a micelle system physically loaded with a
drug. The dose-response curves of cell proliferation as a percentage
of the control (cells incubated with medium only) demonstrate
similar profiles for 2 and poly(HEA)-b-poly(4-AuPEt3), with
poly(HEA)-b-poly(4-AuPEt3) actually exhibiting slightly higher
activity compared to 2. Indeed, regression analysis of the two
curves indicates that the IC50 inhibitory concentration for
poly(HEA)-b-poly(4-AuPEt3) is 0.94 mM (based on the
concentration of Au(I) in the system) and the value for 2 is
1.51 mM. Considering that 2 is one of the most potent known
compounds for inducing cell death via direct interaction with
MS thanks the Australian Research Council for financial
support.
Notes and references
1 C. Gaujoux-Viala, J. S. Smolen, R. Landewe, M. Dougados,
´
T. K. Kvien, E. M. Mola, M. Scholte-Voshaar, P. van Riel and
L. Gossec, Ann. Rheum. Dis., 2010, 69, 1004.
2 C. F. Shaw, Chem. Rev., 1999, 99, 2589.
3 (a) S. Gromer, L. D. Arscott, C. H. Williams, R. H. Schirmer and
K. Becker, J. Biol. Chem., 1998, 273, 20096; (b) S. Urig, K. Fritz-
Wolf, R. Reau, C. Herold-Mende, K. Toth, E. Davioud-Charvet
´ ´
and K. Becker, Angew. Chem., Int. Ed., 2006, 45, 1881.
4 K. Becker, S. Gromer, R. H. Schirmer and S. Muller, Eur. J.
Biochem., 2000, 267, 6118.
¨
5 A. Burke-Gaffney, M. E. J. Callister and H. Nakamura, Trends
Pharmacol. Sci., 2005, 26, 398.
6 M. P. Rigobello, A. Folda, M. C. Baldoin, G. Scutari and
A. Bindoli, Free Radical Res., 2005, 39, 687.
7 V. R. Fantin and P. Leder, Oncogene, 2006, 25, 4787.
8 C. Marzano, V. Gandin, A. Folda, G. Scutari, A. Bindoli and
M. P. Rigobello, Free Radical Biol. Med., 2007, 42, 872.
9 C. K. Mirabelli, R. K. Johnson, D. T. Hill, L. F. Faucette,
G. R. Girard, G. Y. Kuo, C. M. Sung and S. T. Crooke,
J. Med. Chem., 1986, 29, 218.
10 S. Ahmad, A. A. Isab, S. Ali and A. R. Al-Arfaj, Polyhedron, 2006,
25, 1633.
11 J. R. Roberts, J. Xiao, B. Schliesman, D. J. Parsons and
C. F. Shaw, Inorg. Chem., 1996, 35, 424.
12 S. J. Berners-Price and A. Filipovska, Aust. J. Chem., 2008, 61, 661.
13 J. J. Liu, P. Galettis, A. Farr, L. Maharaj, H. Samarasinha,
A. C. McGechan, B. C. Baguley, R. J. Bowen, S. J. Berners-Price
and M. J. McKeage, J. Inorg. Biochem., 2008, 102, 303.
14 M. J. McKeage, S. J. Berners-Price, P. Galettis, R. J. Bowen,
W. Brouwer, L. Ding, L. Zhuang and B. C. Baguley, Cancer
Chemother. Pharmacol., 2000, 46, 343.
15 A. K. Iyer, G. Khaled, J. Fang and H. Maeda, Drug Discovery Today,
2006, 11, 812.
16 E. R. T. Tiekink, Crit. Rev. Oncol. Hematol., 2002, 42, 225.
17 O.-J. Park, D.-Y. Kim and J. S. Dordick, Biotechnol. Bioeng.,
2000, 70, 208.
18 Y. Li and D. G. Rethwisch, Biotechnol. Bioeng., 2002, 79, 15.
19 A. Gregory and M. H. Stenzel, Expert Opin. Drug Delivery, 2011,
8, 237.
20 J. Zastre, J. Jackson, M. Bajwa, R. Liggins, F. Iqbal and H. Burt,
Eur. J. Pharm. Biopharm., 2002, 54, 299.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4695–4697 4697