Redox-sensitive disassembly of amphiphilic copolymer based micelles

Article abstract of DOI:10.1021/la904437u

Amphiphilic polymers of different hydrophilic-lipophilic ratios were prepared by free radical polymerization using two monomers consisting of triethylene glycol as the hydrophilic part and an alkyl chain connected by disulfide bond as the hydrophobic part. These polymers form micelle-like nanoassemblies in aqueous media and can encapsulate hydrophobic drug molecules up to 14% of their mass. In a reducing environment, these polymeric micelles disassemble and dissolve in water, since the amphiphilic polymers are converted into hydrophilic polymers upon cleavage of the disulfide bond. This disassembly event results in the release of hydrophobic molecules that had been encapsulated inside the micelle, the rate of which was found to be dependent on the concentration of the reducing agent, glutathione (GSH). In vitro experiments also show that the GSH-dependent release of the doxorubicin can be used to effect cytotoxicity in MCF-7 cells.

Full text of DOI:10.1021/la904437u

pubs.acs.org/Langmuir
© 2010 American Chemical Society
Redox-Sensitive Disassembly of Amphiphilic Copolymer Based Micelles
Ja-Hyoung Ryu, Raghunath Roy, Judy Ventura, and S. Thayumanavan*
Department of Chemistry, University of Massachusetts, Amherst, Massachusetts 01003
Received November 23, 2009. Revised Manuscript Received December 23, 2009
Amphiphilic polymers of different hydrophilic-lipophilic ratios were prepared by free radical polymerization using
two monomers consisting of triethylene glycol as the hydrophilic part and an alkyl chain connected by disulfide bond as
the hydrophobic part. These polymers form micelle-like nanoassemblies in aqueous media and can encapsulate hydro-
phobic drug molecules up to 14% of their mass. In a reducing environment, these polymeric micelles disassemble and
dissolve in water, since the amphiphilic polymers are converted into hydrophilic polymers upon cleavage of the disulfide
bond. This disassembly event results in the release of hydrophobic molecules that had been encapsulated inside the
micelle, the rate of which was found to be dependent on the concentration of the reducing agent, glutathione (GSH).
In vitro experiments also show that the GSH-dependent release of the doxorubicin can be used to effect cytotoxicity
in MCF-7 cells.
Introduction
being focused on developing targeted delivery using a specific
ligand for cancer cells5,6 or systems responsive to a specific
stimulus such as pH,7,8 redox,9 ultrasound,10 or temperature.11
Most of the stimuli-responsive systems use a prodrug concept, in
which the drug is conjugated to the polymer using labile func-
tionalities that are cleaved upon encountering a stimulus. Thus, a
necessary prerequisite in these systems is that the drug has a
specific functional group that serves as the handle to attach the
linkers. Thus, the chemistry that is developed for this purpose
needs to bespecific for each drug. A promising alternate approach
involves the noncovalent incorporation of the hydrophobic guest
(drug) molecules in an assembly, which is capable of releasing its
content in response to a stimulus. Chemically cross-linked poly-
mer nanoparticles and liposomes for redox-sensitive delivery
have been designed for this purpose.12-14 However, cross-linked
Amphiphilic assemblies are promising vehicles for drug deli-
very because of their ability to noncovalently encapsulate insol-
uble hydrophobic drugs in aqueous media.1 Polymeric scaffolds,
which form these assemblies, are particularly interesting due to
their greater stability and low critical micelle concentrations
(CMCs) compared to their small molecule counterparts.2,3 If
these were nanosized assemblies, then they have the added
advantage for being employed in targeted drug delivery to cancer
cells. This is because it has been shown that the nanosized particles
tend to selectively accumulate in tumor tissue due to the so-called
enhanced permeability and retention (EPR) effect.4 While these
properties outline the advantages gained in sequestering the drug
molecules, it is also important that these supramolecular assem-
blies be capable of unpacking their contents upon reaching their
target site. For this purpose, it is important that these assemblies
are designed in a manner that they lose their container property in
response to a specific stimulus. Moreover, delivering drug mole-
cules nonselectively leads to the toxic side effects observed in
traditional chemotherapeutics. Therefore, research efforts are
(7) (a) Henry, S. M.; El-Sayed, M. E. H.; Pirie, C. M.; Hoffman, A. S.; Stayton,
P. S. Biomacromolecules 2006, 7, 2407–2414. (b) Koa, J.; Park, K.; Kim, Y.-S.; Kim,
M. S.; Han, J. K.; Kim, K.; Park, R.-W.; Kim, I.-S.; Song, H. K.; Lee, D. S.; Kwon, I. C.
J. Controlled Release 2007, 123, 109–115. (c) El-Sayed, M. E. H.; Hoffman, A. S.;
Stayton, P. S. J. Controlled Release 2005, 101, 45–58. (d) Chan, Y.; Wong, T.; Byrne, F.;
Kavallaris, M.; Bulmus, V. Biomacromolecules 2008, 9, 1826–1836.
(8) (a) Bae, Y.; Fukushima, S.; Harada, A.; Kataoka, K. Angew. Chem., Int. Ed.
2003, 42, 4640–4643. (b) Gillies, E. R.; Jonsson, T. B.; Frechet, J. M. J. J. Am. Chem.
Soc. 2004, 126, 11936–11934. (c) Gillies, E. R.; Frechet, J. M. Bioconjugate Chem.
2005, 16, 361–368.
*To whom correspondence should be addressed. E-mail: thai@chem.
umass.edu.
(1) (a) Soussan, E.; Cassel, S.; Blanzat, M.; Rico-Lattes, I. Angew. Chem., Int.
Ed. 2009, 48, 274–288. (b) Kataoka, K.; Harada, A.; Nagasaki, Y. Adv. Drug Delivery
Rev. 2001, 47, 113–131. (c) Ganta, S.; Devalapally, H.; Shahiwala, A.; Amiji, M.
J. Controlled Release 2008, 126, 187–204.
(9) (a) Bulmus, V.; Woodward, M.; Lin, L.; Murthy, N.; Stayton, P.; Hoffman,
A. J. Controlled Release 2003, 93, 105–120. (b) Sun, K. H.; Sohn, Y. S.; Jeong, B.
Biomacromolecules 2006, 7, 2871–2877. (c) Cerritelli, S.; Velluto, D.; Hubbell, J. A.
Biomacromolecules 2007, 8, 1966–1972.
(10) Kabanov, A. V.; Batrakova, E. V.; Alakhov, V. Y. J. Controlled Release
2002, 82, 189–212.
(11) (a) Liua, S. Q.; Tong, Y. W.; Yang, Y.-Y. Biomaterials 2005, 26, 5064–5074.
(b) Nakayama, M.; Okano, T.; Miyazaki, T.; Kohori, F.; Sakai, K.; Yokoyama, M.
J. Controlled Release 2006, 115, 46–56. (c) Chung, J. E.; Yokoyama, M.; Okano, T.
J. Controlled Release 2000, 65, 93–103.
(12) (a) West, K. R.; Otto, S. Curr. Drug Discovery Technol. 2005, 2, 123–160.
(b) Ganta, S.; Devalapally, H.; Shahiwala, A.; Amiji, M. J. Controlled Release 2008,
126, 187–204. (c) Shen, W.-C.; Ryser, H. J. P.; LaManna, L. J. Biol. Chem. 1985, 260,
10905–10908. (d) Hamann, P. R.; Hinman, L. M.; Hollander, I.; Beyer, C. F.; Lindh, D.;
Holcomb, R.; Hallett, W.; Tsou, H. R.; Upeslacis, J.; Shochat, D.; Mountain, A.; Flowers,
D. A.; Bernstein, I. Bioconjugate Chem. 2002, 13, 47–58.
(2) (a) Kataoka, K.; Kwon, G. S.; Yokoyama, M.; Okano, T.; Sakurai, Y.
J. Controlled Release 1993, 24, 119–132. (b) Gaucher, G.; Dufresne, M.-H.; Sant, V. P.;
Kang, N.; Maysinger, D.; Leroux, J.-C. J. Controlled Release 2005, 109, 169–188.
(c) Davis, M. E.; Chen, Z.; Shin, D. M. Nat. Rev. Drug Discovery 2008, 7, 771–782.
(3) (a) Kabanov, A. V.; Batrakova, E. V.; Miller, D. W. Adv. Drug Delivery Rev.
2003, 55, 151–164. (b) van Dongen, S. F. M.; de Hoog, H.-P. M.; Peters, R. J. R. W.;
Nallani, M.; Nolte, R. J. M.; van Hest, J. C. M. Chem. Rev. 2009, 109(11), 6212–6274.
(c) Haag, R.; Kratz, F. Angew. Chem., Int. Ed. 2006, 45, 1198–1215.
(4) (a) Maeda, H.; Wua, J.; Sawaa, T.; Matsumurab, Y.; Horic, K. J. Controlled
Release 2000, 65, 271–284. (b) Brannon-Peppas, L.; Blanchete, J. O. Adv. Drug
Delivery Rev. 2005, 56, 1649–1659. (c) Baban, D. F.; Seymour, L. W. Adv. Drug
Delivery Rev. 1998, 34, 109–119.
(5) (a) Yoo, H. S.; Park, T. G. J. Controlled Release 2004, 100, 247–256. (b) Arap,
W.; Pasqualini, R.; Ruoslahti, E. Science 1998, 279, 377–380. (c) Nasongkla, N.; Shuai,
X.; Ai, H.; Weinberg, B. D.; Pink, J.; Boothman, D. A.; Gao, J. Angew. Chem., Int. Ed.
2004, 43, 6323–6327.
(13) (a) Bhavane, R.; Karathanasis, E.; Annapragada, A. V. J. Controlled
Release 2003, 93, 15–28. (b) Zhang, J. X.; Zalipsky, S.; Mullah, N.; Pechar, M.; Allen,
T. Pharmacol. Res. 2004, 49, 185–198.
(14) (a) Shu, X. Z.; Liu, Y.; Luo, Y.; Roberts, M. C.; Prestwich, G. D.
Biomacromolecules 2002, 3, 1304–1311. (b) Oh, J. K.; Siegwart, D. J.; Matyjaszewski,
K. Biomacromolecules 2007, 8, 3326–3331. (c) Oh, J. K.; Siegwart, D. J.; Lee, H.-i.;
Sherwood, G.; Peteanu, L.; Hollinger, J. O.; Kataoka, K.; Matyjaszewski, K. J. Am.
Chem. Soc. 2007, 129, 5939–5945.
€
(6) (a) Demirgoz, D.; Garg, A.; Kokkoli, E. Langmuir 2008, 24, 13518–13524.
(b) Kim, S. H.; Jeong, J. H.; Chun, K. W.; Park, T. G. Langmuir 2005, 21, 8852–8857.
(c) Gao, X.; Cui, Y.; Levenson, R. M.; Chung, L. W. K.; Nie, S. Nat. Biotechnol. 2004,
22, 969–976.
7086 DOI: 10.1021/la904437u
Published on Web 01/14/2010
Langmuir 2010, 26(10), 7086–7092

Ryu et al.
Article
Figure 1. Schematic representation of drug release via disassembly of amphiphilic polymer micelle triggered by GSH.
polymer nanoparticles and liposomes often require that the drug
molecules be water-soluble for effective encapsulation. Consider-
ing that most chemotherapeutic drugs are lipophilic, alternate
strategies are desirable. Stimuli-sensitive polymeric micellar assem-
blies are promising alternates that have the potential to overcome
these issues, which is the topic of this Article.
Nanoscopic stimuli-sensitive drug delivery vehicles have been
the focus of intense research, because they offer unique advan-
tages, such as the ones mentioned above for targeted delivery to
cancer cells. Redox-sensitive nanoscopic vehicles are particularly
interesting, because of the presence of higher glutathione con-
centrations incancer cells. Disulfide functionality containing drug
delivery vehicles have been explored for this purpose.15 Disulfide
bonds can be cleaved in the presence of reducing agents such as
glycol monomethyl ether (95%), and the other conventional
reagents were used as received from commercial sources. Com-
pounds were synthesized according to the procedure described in
Scheme 1 and then purified by silica gel column chromatography.
Compound 1 was prepared by following the reported proce-
dure.18
Characterization Techniques. 1H NMR spectra were recor-
ded on a 400 MHz Bruker NMR spectrometer using the residual
protonresonanceofthesolventasthe internalstandard. Chemical
shifts are reported in parts per million (ppm). When peak multi-
plicities are given, the following abbreviations are used: s, singlet;
d, doublet; t, triplet; m, multiplet; and b, broad. 13C NMR spectra
were proton decoupled and recorded on a 100 MHz Bruker
spectrometer using the carbon signal of the deuterated solvent
as the internal standard. Molecular weights of the polymers were
estimated by gel permeation chromatography (GPC) using a
poly(methyl methacrylate) (PMMA) standard with a refractive
index detector. Dynamic light scattering (DLS) measurements
were performed using a Malvern Nanozetasizer instrument. The
fluorescence spectra were obtained from a JASCO FP-6500
spectrofluorimeter. Transmission electron microscopy (TEM)
images were taken from a JEOL 100CX instrument at 100 KV.
Toxicity and Cell Viability Assays. The cell viability of the
polymer and toxicity of doxorubicin (Dox)-encapsulated polymer
micelles were tested against the MCF-7 cell line. MCF-7 cells were
cultured in T75 cell culture flasks containing Dulbecco’s modified
Eagle’s medium/nutrient mixture F-12 (DMEM/F12) with 10%
fetal bovine serum (FBS) supplement. For cell viability and
toxicity experiments, the cells were seeded at 10000 cells/well/
100 μL in a 96-well plate and allowed to grow for 24 h at 37 °C in a
5% CO2 incubator. The cells were then treated with either
polymers or Dox-encapsulated polymer micelles of different
concentrations and were incubated again for 72 h. The cell
viability was measured by Alamar Blue assay in triplicate. To
study the effect of release of Dox on cell death at different
concentrations of glutathione, the cells were first incubated with
an aqueous solution of glutathione monoester (GSH-OEt) for
24 h after seeding cells. The cells were then washed two times with
PBS (phosphate-buffered saline). The Dox-encapsulated polymer
solutions were finally added to the cells of different amounts of
GSH and were incubated again for another 72 h, and cell death was
measured by Alamar Blue assay in triplicate.
L-cysteine and glutathione (GSH). In humans, micromolar con-
centrations of GSH are found in the blood plasma, whereas it is
around 10 mM in the cytosol.16 Interestingly, the cytosolic GSH
concentration in some tumor cells has been found to be about
seven times higher than that in normal cells.17 Therefore, drug
delivery based on disulfide functionality can facilitate intracellu-
lar release of the encapsulated drug molecules by cleavage of this
bond. Recently, our group reported that a disulfide-based surfac-
tant that forms a micelle can be disassembled by a reducing agent,
dithiothreitol (DTT).18 This cleavage converts the amphiphilic
molecule to its hydrophilic and hydrophobic constituents, and
therefore, the micelle is disassembled. Small molecule surfactants
based on micelles are likely to beinefficient for drugdelivery, since
the critical micellar concentrations of these systems are too high to
be practical. Micellar assemblies based on polymers can circum-
vent this shortcoming. Herein we report that redox-sensitive
micelles from amphiphilic polymers, containing disulfide bonds
in their hydrophobic part, can disassemble in response to an
increase in biologically relevant GSH concentrations, with a con-
comitant release of the hydrophobic drug molecules (see Figure 1
for a schematic illustration).
Experimental Section
Materials. 1-Undecanethiol (98%), 2,20-dithiodipyridine (98%),
2-mercaptoethanol (99%), acryloyl chloride (97%), triethylene
Synthesis. The synthetic procedures used in the preparation of
amphiphilic copolymers (P1-P4) are described in Scheme 1.
Synthesis of Compound 2. Compound 1 (4.00 g, 13.47 mmol)
was dissolved in 50 mL of methanol, and a catalytic amount
of acetic acid was added. To this solution was added dropwise
2-mercaptoethanol, and the reaction mixture was stirred for 6 h at
room temperature. After completion of the reaction, the solvent
was evaporated and the crude product was purified by column
chromatography using silica gel as the stationary phase and a
mixture of ethyl acetate/hexane (1:4 v/v ratio) as eluent to yield
(15) (a) Koo, A. N.; Lee, H. J.; Kim, S. E.; Chang, J. H.; Park, C.; Kim, C.; Park,
J. H.; Lee, S. C. Chem. Commun. 2008, 6570–6572. (b) Zhang, L.; Liu, W.; Lin, L.;
Chen, D.; Stenzel, M. H. Biomacromolecules 2008, 9, 3321–3331.
(16) (a) Saito, G.; Swanson, J. A.; Lee, K. D. Adv. Drug. Delivery Rev. 2003, 55,
199–215. (b) Lee, Y.; Koo, H.; Jin, G.-W.; Mo, H.; Cho, M. Y.; Park, J.-Y.; Choi, J. S.;
Park, J. S. Biomacromolecules 2005, 6, 24–26. (c) Balendiran, G. K.; Dabur, R.; Fraser,
D. Cell Biochem. Funct. 2004, 22, 343–352.
(17) Russo, A.; DeGraft, W.; Friedman, N.; Mitchell, J. B. Cancer Res. 1986, 46,
2845–2848.
(18) Ghosh, S.; Irvin, K.; Thayumanavan, S. Langmuir 2007, 23, 7916–7919.
Langmuir 2010, 26(10), 7086–7092
DOI: 10.1021/la904437u 7087

Article
Ryu et al.
Scheme 1. Synthesis of Amphiphilic Copolymers
2.3 g (65%) of 2 as a colorless liquid. 1H NMR (CDCl3, ppm) 3.85
(t, 2H), 2.80 (t, 2H), 2.67(t, 2H), 1.65 (t, 2H), 1.36-1.23 (m, 16H),
0.87 (t, 3H); 13C NMR (CDCl3, ppm) 60.3, 41.1, 39.1, 31.9, 29.6,
29.5, 29.3, 29.2, 29.2, 28.8, 28.5, 22.7, 14.1.
Table 1. Amphiphilic Polymer with Various Hydrophobic and
Hydrophilic Ratios
x (hydrophobic): y (hydrophilic)
Mn
Mw
CMC
Synthesis of Compound 3. To a solution of compound 2
(1.3 g, 4.92 mmol) in dichloromethane (50 mL) was added tri-
ethylamine (1.0 mL, 7.62 mmol). This solution was cooled to 0 °C,
and acryloyl chloride (0.56 g, 6.18 mmol) was added dropwise.
The reaction mixture was allowed to stir for 4 h at room
temperature, which was then extracted with CH2Cl2 and the
combined organic layers were concentrated with rotary evapora-
tor. The crude product was purified by column chromatography
using silica gel as the stationary phase and a mixture of ethyl
acetate/hexane (1:4 v/v ratio) as eluent to yield 1.4 g (89%) of 3 as
a colorlessliquid. 1H NMR (CDCl3, ppm) 6.41(d, 1H), 6.13-6.09
(m, 1H), 5.80 (d, 2H), 4.38 (t, 2H), 2.90 (t, 2H), 2.68(t, 2H), 1.64 (t,
2H), 1.36-1.23 (m, 16H), 0.85 (t, 3H); 13C NMR (CDCl3, ppm)
165.6, 130.6, 128.1, 60.4, 41.3, 39.2, 31.9, 29.6, 29.5, 29.3, 29.2,
29.2, 28.8, 28.5, 22.7, 14.1; HRMS (FAB) calcd for C16H30O2S2:
318.1687, found: 138.1687.
expected
observeda
(g/mol)b (g/mol)b PDIb (mg/L)c
P1
P2
P3
P4
1:9
2:8
3:7
4:6
1.0:9.0
1.4:8.6
2.8:7.2
4.0:6.0
11.0K 11.0K 1.9
65
50
25
20
9.4K
9.3K
10.0K 28.6K 2.3
9.4K 2.1
9.3K 2.2
a Determined by NMR. b From GPC using PMMA standards.
c Determined using pyrene emission spectra.
Result and Discussion
Synthesis and Characterization of Amphiphilic Polymers.
Amphiphilic polymers (P1-P4) were synthesized by a simple free
radical polymerization of monomers 3 and 4 in different ratios
(Scheme 1). Monomer 3 is lipophilic due to the undecyl alkyl
chain, while 4 is relatively hydrophilic due to the ethyleneglycol-
based functionality. We expect that a combination of these two
monomers can provide the necessary hydrophilic-lipophilic
balance (HLB) for forming amphiphilic assemblies in the aqueous
phase. In our polymers, since the disulfide bond is placed in the
lipophilic side chain, a redox-sensitive cleavage of the disulfide
bond will cause the release of the undecyl group from the polymer
backbone. We envisaged that the remaining polymeric scaffold
will be significantly more hydrophilic and thus will not sustain an
amphiphilic assembly. This redox-induced change from an amphi-
philic polymer to a hydrophilic polymer should thus result in
a disassembly of the micelles and a concurrent release of the
sequestered guest molecules from the interiors of the amphiphilic
assembly.
The monomers were synthesized from the reaction of the
corresponding alcohols with acryloyl chloride, as shown in
Scheme 1. Polymers were synthesized by mixing different ratios
of the monomers and subjecting the mixture to a free radical
polymerization using AIBN as the initiator. The synthesized
polymers were characterized by NMR and GPC as shown in
Table 1. The experimental ratio of the hydrophilic and lipophilic
components of the polymer was calculated by the relative inte-
Synthesis of Compound 4. Compound 4 was prepared by a
similar procedure which was previously reported.19 Briefly, to a
solution of triethyleneglycol monomethyl ether (10.00 g, 60.90
mmol) in dichloromethane (100 mL) was added triethylamine
(11.10 mL, 79.17 mmol). This solution was cooled to 0 °C, and
acryloyl chloride (6.60 g, 73.08 mmol) was added dropwise and
allowed to stir for 4 h at room temperature. The reaction mixture
was extracted with CH2Cl2, and the combined organic layer was
concentrated with rotary evaporator. The crude product was
purified by column chromatography using silica gel as the sta-
tionary phase and ethyl acetate as eluent to yield 10.7 g (81%) of 4
as a colorless liquid. 1H NMR (CDCl3, ppm) 6.11 (d, 1H),
5.90-5.83 (m, 1H), 5.57 (d, 2H), 4.02 (t, 2H), 3.47 (t, 2H),
3.38-3.34 (m, 6H), 3.25 (t, 2H), 3.09 (s, 3H); 13C NMR (CDCl3,
ppm) 165.6, 130.6, 128.1, 71.7, 70.3, 70.3, 68.8, 63.4, HRMS (FAB)
calcd for C10H18O5þH: 219.1232, found: 219.1232.
Polymerization. All polymers were synthesized using similar
methods. In the case of P1, compound 3 (110 mg, 0.354 mmol),
compound 4 (700 mg, 3.19 mmol), and azobisisobutyronitrile
(AIBN) (6.00 mg, 0.035 mmol) were dissolved in anisole and
purged with argon for 10 min. The mixtures were transferred
to a preheated oil bath at 65 °C and stirred for 24 h. The
polymerization was stopped by cooling to room temperature,
and the polymer was purified by precipitation with diethylether.
1
gration of the H NMR peaks corresponding to the methoxy
proton at the end of triethylene glycol and the methyl proton at
the end of the undecyl chain. All polymers exhibited similar
molecular weights and polydispersities.
(19) Carlise, J. R.; Kriegel, R. M.; Rees, W. S., Jr.; Weck, M. J. Org. Chem. 2005,
70, 5550–5560.
7088 DOI: 10.1021/la904437u
Langmuir 2010, 26(10), 7086–7092

Ryu et al.
Article
Figure 2. (a) Size distribution and TEM image (inset) of P2 (0.1 wt %) in water. (b) Emission spectra of Nile Red sequestered in polymeric
micelles.
Self-Assembly and Hydrophobic Dye Encapsulation of
Amphiphilic Polymers. The amphiphilic assemblies formed by
the polymers in the aqueous phase were characterized by light
scattering and transmission electron microscopy (TEM). Dyna-
mic light scattering (DLS) experiments were performed to inves-
tigate the self-assembly behavior in water. The polymers aggre-
gate into several tens of nanometer sized micellar structures.20 The
size distribution of P2 in water is shown in Figure 2a. Its average
hydrodynamic radius (RH) was observed to be approximately
32 nm. Sizes obtained from TEM closely correlate with those from
the DLS experiments. Critical micelle concentrations (CMCs) of
these polymers were estimated using pyrene as the spectroscopic
probe (Table 1).21 We found that increasing the hydrophobic
composition of the polymer lowers the relative CMC.
We also investigated the encapsulation characteristics of the
polymeric assemblies with a hydrophobic dye. Nile Red is not
soluble in water by itself, but it can be sequestered within the
hydrophobic interiors of the polymer assemblies. Since Nile Red
also exhibits an environmentally sensitive fluorescence beha-
vior, this is an interesting guest molecule to study. As shown in
Figure 2b, we found that the emission intensity of Nile Red
sequestered in micelles gradually increases upon increasing the
ratio of the hydrophobic block. In addition, we also observed that
the emission wavelength maximum of Nile Red is systematically
blue-shifted from P1 to P4. Considering that Nile Red is a
solvatochromic probe that shows an environment-dependent
emission wavelength,22 the core of P4 was found to be most
hydrophobic, which is understandable as the percentage of the
lipophilic units is the highest in this copolymer. This observation
indicates that the hydrophobicity of the micelle core can be simply
controlled by tuning the monomer ratio, and this result is
consistent with the CMC trend of the polymers (Table 1).
its hydrophobic core. To examine the dependence of disassembly
and dye release upon the difference in GSH concentration, Nile
Red containing polymer (0.1 wt %) in pH 7.4 phosphate saline
buffer solution was treated with various concentrations of GSH
(0, 10 μM, 10 mM, 30 mM, and 70 mM) and the emissionintensity
of the Nile Red was monitored for every 12 h for 4 days. These
experiments were performed only on polymers P1 and P2,
because P3 and P4 precipitated out in the buffer. Figure 3 shows
the change of the dye emission intensity of P2 solution with
various GSH concentrations. In the control experiment, wherein
no GSH was added, the intensity remained relatively steady
throughout the time of the experiment (Figure 3a). With a 10 μM
concentration of GSH, the emission intensity was essentially
unchanged. This concentration is interesting, as it corresponds
to the GSH concentration in the plasma of the human body.23
This result can therefore be taken to suggest that this micellar
solution is likely to be stable in the plasma without releasing the
hydrophobic guest molecule. On the other hand, inthepresenceof
10 mM GSH, there was a slight indication of guest molecule
release at very long time scales (Figure 3c). Interestingly, at even
higher concentrations of GSH (30 and 70 mM), the emission
intensity of the dye molecules significantly decreased with time
(Figure 3d and e). The release of the dye molecule is also
illustrated in Figure 3f, where the Nile Red encapsulated polymer
solution becomes completely colorless after 96 h in the presence of
70 mM GSH. The disassembly of the polymeric micelles was
further confirmed by DLS.20 The correlation function in the DLS
experiments, which was excellent before addition of GSH, turned
poor after 96 h. This clearly indicates the lack of nanoassembly in
solution. Note that a significant enhancement in the responsive
nature of this polymer micelle was achieved with about seven
times differenceinthe GSH concentrationinthe millimolar range.
Althoughthe viability of this polymer in a clinical settingis several
steps removed, we were gratified to note that our polymers could
respond to the differences in the GSH concentrations that cor-
respond to the difference between healthy cells and cancer cells.17
This phenomenon of the GSH-dependent disassembly and
guest release can be understood based on the destabilization of
the hydrophobic core due to the disulfide bond cleavage resulting
in a change in the HLB of the polymer. To further investigate the
effect of the hydrophobic core upon the release, we performed
GSH-dependent release experiments with P1. Figure 4 shows the
Disassembly of Polymeric Micelles Triggered by Gluta-
thione. We were next interested in testing our hypothesis that
these assemblies will be stimuli-sensitive and would be dis-
assembled in the presence of a redox stimulus, such as glutathione
(GSH). The disassembly of a micellar aggregate can be tested by
monitoring the release of the guest molecules encapsulated within
(20) See Supporting Information for details.
(21) (a) Wilhelm, M.; Zhao, C.-L.; Wang, Y.; Xu, R.; Winnik, M. A.; Mura,
J.-L.; Riess, G.; Croucher, M. D. Macromolecules 1991, 24, 1033–1040. (b) Kwon,
G.; Naito, M.; Yokoyama, M.; Okano, T.; Sakurai, Y.; Kataoka, K. Langmuir 1993, 9,
945–949. (c) See the Supporting Information.
(22) Stuart, M. C. A.; van de Pas, J. C.; Engberts, J. B. F. N. J. Phys. Org. Chem.
(23) Jones, D. P.; Carlson, J. L.; Samiec, P. S.; Sternberg, P.; Mody, V. C.; Reed,
2005, 18, 929–934.
R. L.; Brown, L. A. S. Clin. Chim. Acta 1998, 275, 175–184.
Langmuir 2010, 26(10), 7086–7092
DOI: 10.1021/la904437u 7089

Article
Ryu et al.
Figure 3. Dye release from P2 depending on various times at different GSH concentrations: (a) 0, (b) 10 μM, (c) 10 mM, (d) 30 mM, and
(e) 70 mM. (f) Photograph of polymeric solution containing dye before and after adding 70 mM GSH.
release profile of Nile Red with P1- and P2-based assemblies for
up to 96 h. There seems to be a burst release with P1 during the
first 10 h, which might be due to the inherently less lipophilic
nature of P1 relative to P2, as similar burst release was observed
even in the absence of GSH. The only discernible difference
between P1 and P2 systems is that, at 10 mM concentration of
GSH, there seems to be a finite amount with P1 while there is
negligible release with P2.
Cell Viability of Polymeric Micelles Containing Doxo-
rubicin (Dox). Considering the results that we obtained in the
buffer, we were interested in testing the translation of this work
with chemotherapeutic molecules. Thus, we have carried out in
vitro cell viability assays for these polymers. First, we measured
the cell viability of P2 with the MCF-7 cell line. MCF-7 cells were
treated with different concentrations of polymeric micellar solu-
tion and were incubated for 72 h. The cytotoxicity of these cells
was measured using the Alamar Blue assay. As shown in Figure 5a,
the polymeric micelles of P2 were found to exhibit greater than
80% cell viability even at a concentration of about 1 mg/mL.
To investigate the possibility of utilizing this polymer as a drug
delivery vehicle, we encapsulated the chemotherapeutic cytotoxic
drug molecule doxorubicin (Dox). The loading capacity ofP2 was
found to be 14% of the polymer w/w. When the micelle (0.2 mg/
mL) loaded with Dox (10 μg/mL) was added to MCF-7 cells, the
extent of cell death was 22% after 72 h. Considering that the GSH
concentration in MCF-7 cells is about 3 mM,24 this cell death
value is reasonable. We hypothesized that since the disulfide bond
in the hydrophobic part of the polymeric micelle could bepartially
disconnected, it is likely that only a small amount of Dox was
released and thus the cytotoxic effect of Dox was limited. To
investigate if higher concentrations of GSH would result in higher
release of Dox and thus better cytotoxicity of the polymer-Dox
complex, we artificially enhanced the intracellular GSH concen-
tration using a well-established procedure.25 In this case, the cells
were incubated with glutathione monoester (GSH-OEt), which is
known to be rapidly taken up by the cell and hydrolyzed to
produce GSH. Thus, a measured amount of GSH-OEt was added
to the MCF-7 cells and incubated for 1 h. This solution was then
washed twice with PBS buffer to remove GSH-OEt in the media,
which was not taken up by the cells. To these cells, we added the
polymeric micelle solution (0.2 mg/mL) containing 10 μg/mL
Dox, and the cells were incubated for 72 h. The Alamar Blue assay
results of these experiments are shown in Figure 5b. After
treatment with 5 mM GSH-OEt, the cell death doubled and there
€
(24) Boubakari, B. K.; Neumann, C.; Grunert, R.; Bednarski, P. J. Arch. Pharm.
(Weinheim, Ger.) 2004, 337, 668–671.
(25) Anderson, M. E.; Meister, A. Anal. Biochem. 1989, 183, 16–20.
7090 DOI: 10.1021/la904437u
Langmuir 2010, 26(10), 7086–7092

Ryu et al.
Article
Figure 4. Dye release at different GSH concentration with (a) P1 and (b) P2.
Figure 5. In vitro toxicity of (a) empty micelles and (b) Dox-loaded micelles of P2 with MCF-7 cells after 72 h incubation with various
GSH-OEt concentrations (Alamar Blue assay). (c) In vitro toxicity of Dox-loaded P5 micelles and the chemical structure of P5.
was a clear correlation between the concentration of the GSH-
OEt and the extent of cytotoxicity. In a control experiment,
wherein no Dox was included, P2 polymeric micelles did not show
any cell death with different GSH-OEt concentrations.20 These
results show that Dox can be released in the presence of higher
concentrations of GSH to effect cell death.
To further confirm that cell death is because of the Dox
released by the disassembly of polymeric micelles due to di-
sulfide bond cleavage, we investigated the cytotoxicity of Dox-
encapsulated polymeric micelles of P5. This polymer is structurally
analogous to P2, but it lacks the disulfide bond in its hydrophobic
chain. The polymeric micelle solution (0.2 mg/mL) of P5 contains
the same amount of Dox, 10 μg/mL, as that with P2. As shown in
Figure 5c, the polymer P5 does not show significant cell death
indicating that the drug was indeed released due to the disassem-
bly of the micelle.
Conclusions
In summary, we have synthesized a series of amphiphilic
polymers with varying hydrophilic-hydrophobic ratios. These
polymers form micelle-type nanostructures in water and can
encapsulate hydrophobic drug molecules. Since the hydrophobic
component of the polymer is linked using a disulfide bond, the
amphiphilic polymer is converted to a more hydrophilic polymer
when glutathione reacts with the disulfide functionality and
releases the lipophilic alkyl chain. This conversion causes
the amphiphilic supramolecular nanoassembly to disassemble,
which results in the release of the hydrophobic guest molecule
Langmuir 2010, 26(10), 7086–7092
DOI: 10.1021/la904437u 7091

Article
Ryu et al.
sequestered within the polymer assembly. The kinetics of guest
release is shown to be dependent on the concentration of GSH.
We have then utilized these polymeric vehicles to sequester
hydrophobic chemotherapeutic drug molecules and release them
intracellularly, in a concentration-dependent manner, to effect
drug-induced cytotoxicity. We clearly show that the observed
glutathione-dependent cytotoxicity with our drug-encapsulated
polymeric vehicles is due to the cleavage of the disulfide bond and
the ensuing release of the drug molecule. Considering that the
GSH concentrations used in our studies are biologically relevant,
our amphiphilic polymer design could be useful in drug delivery
applications to cancer cells.
Acknowledgment. We thank DARPA and the NSF Center for
Hierarchical Manufacturing for support. We also thank the NSF-
MRSEC facility at the University of Massachusetts for the
infrastructure used in these characterizations. J.-H.R thanks the
Korea Research Foundation Grant funded by the Korean
Government (MOEHRD) for support (KRF-2007-357- C00078).
Supporting Information Available: 1H NMR, CMC plots,
DLS and TEM of amphiphilic polymers, dye-release profile
of P1, correlation function of P2 before/after addition of
GSH, and synthesis of P5. This material is available free of
charge via the Internet at http://pubs.acs.org/.
7092 DOI: 10.1021/la904437u
Langmuir 2010, 26(10), 7086–7092