Synthesis of amphiphilic alternating polyesters with oligo(ethylene glycol) side chains and potential use for sustained release drug delivery

Article abstract of DOI:10.1021/bm200668n

Novel amphiphilic alternating polyesters, poly((N-phthaloyl-l-glutamic anhydride)-co-(2-(2-(2-methoxyethoxy)ethoxy)methyl)oxirane) (P(PGA-co-ME ₂MO)), were synthesized by alternating copolymerization of PGA and ME₂MO. The structures of the synthesized polyesters were characterized by ¹H NMR, ¹³C NMR, FT-IR, and GPC analyses. Because of the presence of oligo(ethylene glycol) (OEG) side chains, the polyesters could self-assemble into thermosensitive micelles. Dynamic light scattering (DLS) showed that these micelles underwent thermoinduced size decrease without intermicellar aggregation. In vitro methyl thiazolyl tetrazolium (MTT) assay demonstrated that the polyesters were biocompatible to Henrietta Lacks (HeLa) cells, rendering their potential for drug delivery applications. Two hydrophobic drugs, rifampin and doxorubicin (DOX), were loaded into the polyester micelles and observed to be released in a zero-order sustained manner. The sustained release could be accelerated in lower pH or in the presence of proteinase K, due to the degradation of the polyester under these conditions. Remarkably, in vitro cell experiments showed that the polyester micelles accomplished fast release of DOX inside cells and higher anticancer efficacy as compared with the free DOX. With enhanced stability during circulation condition and accelerated drug release at the target sites (e.g., low pH or enzyme presence), these novel polyesters with amphiphilic structures are promising to be used in sustained release drug delivery systems.

Full text of DOI:10.1021/bm200668n

ARTICLE
pubs.acs.org/Biomac
Synthesis of Amphiphilic Alternating Polyesters with Oligo(ethylene
glycol) Side Chains and Potential Use for Sustained Release
Drug Delivery
Wei Wang,†,§ Jianxun Ding,†,‡,§ Chunsheng Xiao,†,‡ Zhaohui Tang,† Di Li,† Jie Chen,† Xiuli Zhuang,*,† and
Xuesi Chen†
†Key Laboratory of Polymer Ecomaterials, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences,
Changchun 130022, P. R. China
‡Graduate University of Chinese Academy of Sciences, Beijing 100039, P. R. China
S Supporting Information
b
ABSTRACT: Novel amphiphilic alternating polyesters, poly((N-phthaloyl-L-
glutamic anhydride)-co-(2-(2-(2-methoxyethoxy)ethoxy)methyl)oxirane)
(P(PGA-co-ME2MO)), were synthesized by alternating copolymerization of
PGA and ME2MO. The structures of the synthesized polyesters were
1
characterized by H NMR, 13C NMR, FT-IR, and GPC analyses. Because
of the presence of oligo(ethylene glycol) (OEG) side chains, the polyesters
could self-assemble into thermosensitive micelles. Dynamic light scattering
(DLS) showed that these micelles underwent thermoinduced size decrease
without intermicellar aggregation. In vitro methyl thiazolyl tetrazolium
(MTT) assay demonstrated that the polyesters were biocompatible to
Henrietta Lacks (HeLa) cells, rendering their potential for drug delivery
applications. Two hydrophobic drugs, rifampin and doxorubicin (DOX),
were loaded into the polyester micelles and observed to be released in a zero-
order sustained manner. The sustained release could be accelerated in lower pH or in the presence of proteinase K, due to the
degradation of the polyester under these conditions. Remarkably, in vitro cell experiments showed that the polyester micelles
accomplished fast release of DOX inside cells and higher anticancer efficacy as compared with the free DOX. With enhanced stability
during circulation condition and accelerated drug release at the target sites (e.g., low pH or enzyme presence), these novel polyesters
with amphiphilic structures are promising to be used in sustained release drug delivery systems.
’ INTRODUCTION
poly(L-glutamate)s (PLG) were grafted with thermosensitive poly-
(2-(2-methoxyethoxy)ethyl methacrylate) (PMEO2MA) or oligo-
(2-(2-(2-methoxyethoxy)ethoxy)ethyl methacrylate) (OMEO3MA)
chains through atom transfer radical polymerization.14,15 The resul-
tant copolymers can form micellar aggregates in aqueous solution
with R-helical polypeptide cores and thermosensitive PMEO2MA or
OMEO3MA shells. These micelles showed decreased size at elevated
temperature without intermicellar aggregation, which was attractive
for drug delivery application.
Among all of the above-mentioned amphiphilic assemblies,
the micelles self-assembled in aqueous solution have many potential
applications in biomedical fields, such as controlled drug release16,17
and gene delivery,18,19 because of their biodegradability20,21 and
biocompatibility.22 The self-assembled micelles consist of the hydro-
phobic cores away from the water and the hydrophilic shells extend-
ing into the water. These self-assembled micelles have attracted
significant attention as ideal drug delivery systems,23 which are more
It is well known that amphiphilic block copolymers can self-
assemble into various morphologies,1ꢀ5 such as vesicles6,7 and
micelles,8 depending on the type of selective solvent, the polymer
concentration, and the chain structure. Not only the amphiphilic
block copolymers but also a number of polymers with amphiphilic
structures are capable of providing diverse self-assembled struc-
tures.9ꢀ11 Thayumanavan et al. developed a molecular design based
on homopolymers in which both the hydrophilic and hydrophobic
moieties are incorporated within one structure unit.12 Such homo-
polymers can form both micelle and inverse micelle assemblies
depending on the solvent environment. Amphiphilic graft copolymers
consisting of hydrophobic backbones and hydrophilic side chains or
vice versa are another kind of interesting materials that can self-
assemble into various morphologies. For instance, Huang et al. was
reported to synthesize well-defined amphiphilic copolymers with
hydrophilic poly(acrylic acid) (PAA) backbones and hydrophobic
poly(propylene oxide) (PPO) side chains, which can self-assemble
into micelles.13 The biodegradable and biocompatible polypeptide
can also be introduced as the hydrophobic backbone for its potential
biomedical applications. Chen and coworkers reported that the
Received: January 26, 2011
Revised:
June 3, 2011
Published: June 08, 2011
r
2011 American Chemical Society
2466
dx.doi.org/10.1021/bm200668n Biomacromolecules 2011, 12, 2466–2474
|

Biomacromolecules
ARTICLE
Scheme 1. Synthesis Pathway for PGA Monomer
effective to increase the apparent solubility of the hydrophobic drug
and the benefit of drug and thereby decrease the side effect.24
However, the burst drug release from the micelles limits the appli-
cation of almost all the micelle-based drug delivery systems.25ꢀ27
To overcome this limitation for drug delivery application, the
novel alternating polyesters with hydrophobic N-phthaloyl
groups in the polyester backbone and hydrophilic oligo(ethylene
glycol) (OEG) side chains were synthesized in the present work.
The strong hydrophobic interaction between N-phthaloyl groups
and drugs can weaken the burst drug release from micelles.28 The
OEG side chains can resist nonspecific protein adsorption (i.e.,
nonfouling properties) in the complicated in vivo circumstance.29
The micelles with OEG shells will be advantageous in drug delivery.
In addition, polymers bearing OEG side chains, such as poly(oligo-
(ethylene glycol) acrylate) (POEGMA), are used as a class of
thermosensitive materials.30,31 Accordingly, the self-assembled polye-
ster micelles with OEG shells were also thermosensitive and showed
thermoinduced size decrease without intermicellar aggregation in
aqueous solution. This property would further compress the micelles
at the physiological temperature, and less burst release could be
expected. Therefore, we anticipated that these amphiphilic alternat-
ing polyesters would show promising use in the sustained release
drug delivery. For this purpose, rifampin and doxorubicin (DOX)
were loaded into the micelles, and the release behaviors were
monitored with an ultravioletꢀvisible (UVꢀvis) spectrophotometer
under different conditions. In addition, in vitro cell experiments
revealed that these polyester micelles accomplished fast release of
DOX inside cells and higher anticancer efficacy as compared with the
free DOX control.
Scheme 2. Synthesis Pathway for ME2MO Monomer
acid (20 g, 0.36 mol) and freshly distilled acetic anhydride (40 mL) were
heated to 100 ꢀC until the solution turned clear and then were stirred for an
additional 5 min at 100 ꢀC. The solution was cooled and concentrated under
vacuum. The resulting solid was washed with cold acetic anhydride and dry
ether, and then dried under vacuum (yield: 73.0%). 1H NMR (400 MHz,
DMSO-d6, ppm): 2.54ꢀ2.68 (m, 2H), 2.94ꢀ3.18 (m, 2H), 5.46 (m, 1H),
7.92 (m, 4H). FT-IR (KBr): Absorptions at 1840, 1800, and 1730 cmꢀ1 are
the stretching of carbonyl groups from N-phthaloyl group and anhydride ring
(Supporting Information, Figure S2).
Synthesis of (2-(2-(2-Methoxyethoxy)ethoxy)methyl)oxi-
rane (ME2MO) (Scheme 2). Tetrabutylammonium hydrogen sulfate
(TBAHS, 5 g, 0.02 mol), epichlorohydrin(90 g, 0.97 mol), and aqueous
sodium hydroxide(36 g, 50 wt %) were mixed with vigorous stirring at
0 ꢀC. Then, the 2-(2-methoxyethoxy)ethanol (36 g, 0.3 mol) was added
dropwise. The solution was allowed to warm to room temperature and
stirred for a further 3 h. Chloroform was added to extract the product
and washed with brine to neutrality. The organic phase was dried with
sodium sulfate for 24 h. After filtration, the filtrate was evaporated and
followed by vacuum distillation (80 ꢀC, 10 Pa) (yield: 70.0%). 1H NMR
(400 MHz, CDCl3, ppm): 2.61 (q, 1H), 2.80 (t, 1H), 3.17 (m, 1H), 3.39
(s, 3H), 3.44 (q, 1H), 3.55ꢀ3.58 (m, 2H), 3.63ꢀ3.73 (m, 6H), 3.78,
3.81 (dd, 1H). FT-IR (KBr): 1112 cmꢀ1(υCꢀOꢀC), the stretching
absorptions at 1108 and 910 cmꢀ1 are the characteristic bands of the
epoxide group (Supporting Information, Figure S3).
Synthesis of P(PGA-co-ME2MO) (Scheme 3). As a typical
procedure, PGA (3.67 g, 14.1 mmol), ME2MO (2.5 g, 14.1 mmol), and
AIP (0.04 g, 0.19 mmol) were mixed in a dry flask. The mixture was stirred in
bulk at 120 ꢀC for 24 h. The crude product was dissolved in chloroform and
precipitated with n-hexane/ethyl ether (1/1, v/v). The yellow oily product
was obtained after drying at 70 ꢀC for 24 h under vacuum (yield: 75.0%, Mn,
NMR, Mn,GPC, and Mw/Mn were listed in Table 1).
Preparation of P(PGA-co-ME2MO) Micelles. In brief, 5.0 mg of
polyester was dissolved in 1 mL of DMF, and the solution was allowed to stir
at room temperature for 2 h. Then, to this solution, 2 mL of deionic water
was added with a rate of 1.0 mL minꢀ1 under gentle stirring. After standing
for 3 h at room temperature, DMF was removed by dialysis using a dialysis
bag (MWCO 3500 Da) against deionic water for 24 h to obtain the micelles.
Characterization. 1H NMR and 13C NMR spectra were recorded
on a Bruker AV 400 NMR spectrometer in chloroform-d (CDCl3) or
dimethyl sulfoxide-d6 (DMSO-d6). FT-IR spectra were recorded on a Bio-
Rad Win-IR instrument using the potassium bromide (KBr) method.
Number- and weight-average molecular weights (Mw, Mn) and molecular
weight distributions (PDI = Mw/Mn) were determined by gel permeation
chromatography (GPC) using a series of linear Styragel columns (HT3 and
HT4) and a Waters 515 HPLC pump, with OPTILAB DSP interferometric
refractometer (Wyatt Technology) as the detectors. The eluent was chloro-
form at a flow rate of 1.0 mL minꢀ1 at 40 ꢀC. Monodispersed polystyrene
standards purchased from Waters with a molecular weight range of
’ EXPERIMENTAL SECTION
Materials. L-Glutamic acid (99%, Shanghai Huishi), phthalic anhy-
dride (Beijing Chemical Works), tetrabutylammonium hydrogen sulfate
(99%, Aldrich), and 2-(2-methoxyethoxy)ethanol (98%, Aldrich) were used
without further purification. Aluminum isopropoxide (AIP) and acetic
anhydride (Sinopharm Chemical Reagen) were distilled before used.
Rifampin and doxorubicin hydrochloride (DOX HCl) were purchased
3
from Zhejiang Hisun Pharmaceutical and used as received. All other reagents
and solvents were purchased from Sinopharm Chemical Reagent, China and
used as obtained.
Synthesis of 2-(1,3-Dioxoisoindolin-2-yl)pentanedioic
Acid (Scheme 1). Phthalic anhydride (70 g, 0.47 mol), glacial acetic
acid (150 mL), and L-glutamic acid (74 g, 0.5 mol)) were added to a 500 mL
round-bottomed flask. The mixture was refluxed at 145 ꢀCfor1h. Afterward,
the reaction mixture was filtrated to remove the insoluble substance. The
filtrate was evaporated and then recrystallized from water to afford the
product (yield: 85.0%). 1H NMR (400 MHz, DMSO-d6, ppm): 2.26ꢀ2.50
(m, 4H), 4.91 (m, 1H), 7.75ꢀ7.86 (m, 4H). FT-IR (KBr): Absorptions
at 1777 cmꢀ1, 1712 cmꢀ1 are the stretching of carbonyl groups from
N-phthaloyl group (Supporting Information, Figure S1).
Synthesis of N-Phthaloyl-L-glutamic Anhydride (PGA)
(Scheme 1). In a 250 mL flask, 2-(1,3-dioxoisoindolin-2-yl)pentanedioic
2467
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
Scheme 3. Synthesis Pathway for P(PGA-co-ME2MO)
Table 1. Characterizations of P(PGA-co-ME2MO) Alternating Polyesters
feeding molar ratio of PGA(ME2MO)/-OiPr
DP of P(PGA-co-ME2MO)a
Mn
Mn
PDIb
a
b
polyester
P(PGA-co-ME2MO)23
P(PGA-co-ME2MO)47
25
50
23
47
10100
20500
4800
9900
1.14
1.16
a Determined by 1H NMR. b Determined by GPC (CHCl3 as eluent).
1310ꢀ55 100 were used to generate the calibration curve. Dynamic laser
scattering (DLS) measurements were performed on a WyattQELS instru-
ment with a vertically polarized HeꢀNe laser (DAWN EOS, Wyatt
Technology). The scattering angle was fixed at 90ꢀ. Transmission electron
microscopy (TEM) measurement was performed on a JEOL JEM-1011
transmission electron microscope with an accelerating voltage of 100 KV. A
drop of the micelle solution (0.5 g Lꢀ1) was deposited onto a 230 mesh
copper grid coated with carbon and allowed to dry in air at 25 ꢀC before
measurements. Critical micelle concentration (CMC) was measured by
fluorescence spectroscopy using pyrene as a probe on a Perkin-Elmer LS50B
luminescence spectrometer at the detection wavelength (λem) of 390 nm.
The CMC was obtained from the intersection of the tangent to the
horizontal line of I335/I332 with relative constant values and the diagonal
line with rapidly increased I335/I332 ratio.
Thermal Analyses. Thermal properties of the polyesters were
examined by differential scanning calorimetry (DSC Q100, TA Instruments)
under a N2 atmosphere. The sample was heated from ꢀ35 to 80 ꢀC at a rate
of 10 ꢀC minꢀ1, kept at 80 ꢀC for 5 min and cooled to ꢀ35 ꢀC at a rate of
10 ꢀCminꢀ1. The data collection was carried out on the second heating run,
and the glass-transition temperature (Tg) was taken to be the midpoint (the
temperature corresponding to half of the endothermic shift).
was allowed to stand at room temperature for 2 h. Then, 5 mL of deionic
water was added dropwise to this solution under stirring. The mixture
was stirred at room temperature for 6 h, and the organic solvent was
removed by dialysis against deionic water for 24 h to obtain the drug-
loaded micelles. The solution was filtered and freeze-dried. The drug
loading content (DLC%) and the drug loading efficiency (DLE %) of
drug-loaded micelles were calculated by the following equations:
amount of drug in micelle
DLC% ¼
ꢁ 100%
amount of drug-loaded micelle
amount of drug in micelle
DLE% ¼
ꢁ 100%
total amount of drug for drug loading
In vitro drug release behaviors from the micelles were investigated in
CPBS (pH 7.4). The weighed freeze-dried drug-loaded micelles were
suspended in 5 mL of CPBS and introduced into a dialysis bag (MWCO
3500 Da). We initiated the release experiment by placing the end-sealed
dialysis bag in 60 mL of CPBS at 37 ꢀC with continuous shake at 70 rpm.
At predetermined intervals, 5 mL of CPBS was taken out, and an equal
volume of fresh CPBS was replenished. The amount of released drug was
assayed by spectrophotometry at 480 nm using the standard curve
method. The drug releases were also performed in CPBS (pH 7.4) with
In Vitro Enzymatic Degradation. The in vitro enzymatic degrada-
tion was performed at 37 ꢀC in citric acidꢀdisodium hydrogen phosphate
proteinase K (2 μg mLꢀ1, 0.08 U mLꢀ1) and sodium azide (0.1 g Lꢀ1
)
buffer solution (CPBS, pH 7.4) containing proteinase K (2 μg mLꢀ1
,
and in CPBS at pH 5.5.
0.08 U mLꢀ1) and sodium azide (0.1 g Lꢀ1). The concentration of the
polyester was set at 0.5 g Lꢀ1. At predetermined time intervals, samples
were taken out and freeze-dried for GPC analyses.
Intracellular DOX Release. The cellular uptake and intracellular
release behaviors of DOX-loaded polyester micelles were determined by
confocal laser scanning microscopy (CLSM) using HeLa cells. The cells
were seeded in six-well plates at ∼200 000 cells per well in 2 mL of
complete DMEM and cultured for 24 h, followed by removing culture
medium and adding DOX-loaded polyester micelles at a final DOX
concentration of 10 mg Lꢀ1 (2 mL in complete DMEM medium). Free
DOX was used as the control. The cells were incubated with DOX-
loaded micelles or free DOX for 0.5 or 2 h. Thereafter, the culture media
were removed, and the cells were rinsed four times with PBS and fixed
with 4% formaldehyde for 30 min at room temperature, and the cell
nuclei were stained with 40,6-diamidino-2-phenylindole (DAPI, blue).
CLSM images of cells were obtained through confocal microscope
(Olympus FluoView 1000).
Cytotoxicity Assay. The relative cytotoxicities of polyesters and
drug-loaded polyester micelles were assessed with methyl thiazolyl
tetrazolium (MTT) viability assay against Henrietta Lacks (HeLa) cells.
The cells were seeded in 96-well plates at ∼20 000 cells per well in 100 μL of
complete DMEM containing 10% fetal bovine serum, supplemented with
50 U mLꢀ1 penicillin and 50 U mLꢀ1 streptomycin, and incubated at 37 ꢀC
in 5% CO2 atmosphere for 24 h, followed by removing culture medium and
adding polyester solutions at different concentrations (0 to 0.1 g Lꢀ1) or
drug-loaded polyester micelles (0 to 10 mg Lꢀ1 DOX) (100 μL in complete
DMEM medium). The cells were subjected to MTT assay after being
incubated for another 24 h. The absorbance of the solution was measured on
a Bio-Rad 680 microplate reader at 492 nm. Cell viability (%) was calculated
by (Asample/Acontrol) ꢁ 100, where Asample and Acontrol are denoted as
absorbance of the sample well and control well (without polyester),
respectively. Experiments were performed in triplicate.
’ RESULTS AND DISCUSSION
Synthesis of P(PGA-co-ME2MO). The aliphatic polyesters can
be directly synthesized by living ring-opening polymerization
(ROP) of lactone monomers using covalent metal alkoxides or
carboxylates as the catalysts.33 AIP, one of the most widely used
In Vitro Drug Loading and Release. The polyester (50.0 mg) in
5.0 mL of DMF was mixed with 50.0 mg of rifampin or DOX HCl
3
(neutralized with one molar equivalent of triethylamine). The mixture
2468
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
Figure 2. GPC chromatograms of P(PGA-co-ME2MO)23 (Mn = 4800,
and PDI = 1.14) (a) and P(PGA-co-ME2MO)47 (Mn = 9900, and PDI =
1.16) (b).
Figure 3. DSC thermograms of P(PGA-co-ME2MO).
the copolymerization should be carried out alternatively in a strict
way between PGA and ME2MO monomers.32 The degree of
polymerization (DP, listed in Table 1) of P(PGA-co-ME2MO)
was calculated by comparing the integrated area of peak d or j with
that of the signal at 1.19 ppm (a) assigned to methyl protons of
isopropyloxy end group. The 13C NMR spectrum (Figure 1B)
confirmed the polyester structure. The signal at 171.6 ppm was
assigned to the ester bond from backbone. FT-IR (Figure 1C) result
also verified the generation of polyester based on the appearance of
carbonyl absorption at 1731 cmꢀ1 from ester bond. The disappear-
ance of the epoxide group absorption at 910 cmꢀ1 while remaining
the typical stretching CꢀOꢀC absorption at 1112 cmꢀ1 from
OEG also indicated the ring-opening incorporation of the ME2MO
into the polyester backbone (Figures 1C, S3 in the Supporting
Information). The successful preparations of the polyesters were also
confirmed by the GPC analyses, as shown in Figure 2. The unimodel
peaks with low-molecular-weight distribution were clearly observed,
which may be explained by the living character of the anionic ring-
opening polymerization of PGA and ME2MO monomers catalyzed
by AIP.33 It should be noted that the Mn measured by GPC is
relatively lower than that measured by 1H NMR due to the structural
difference between the resultant polyesters and monodispersed
polystyrene standards, which were used to generate the calibration
curve in GPC analyses.
Figure 1. Typical 1H NMR (in CDCl3) (A), 13C NMR (in CDCl3) (B),
and FT-IR (C) spectra of P(PGA-co-ME2MO)47.
catalysts, has been successfully applied in the ROP of ε-capro-
lactone, lactide, and cyclic phosphoester monomers through a
coordinationꢀinsertion mechanism.33,34 Recently, Hao et al. further
investigated the use of AIP in the ring-opening alternative polymer-
ization of succinic anhydride and functionalized epoxide.32 In this
work, we attempted to introduce natural glutamic acid to the
polyester by using N-phthaloyl-glutamic anhydride (PGA) mono-
mer. First, PGA and ME2MO monomers were synthesized with the
synthetic routes shown in Schemes 1 and 2, respectively.32,35 The
structures of PGA and ME2MO monomers were confirmed by 1H
NMR and FT-IR spectra (Supporting Information, Figures S2 and
S3, respectively). The polymerization was conducted in bulk at
120 ꢀC with PGA/ME2MO monomer molar feed ratio 1:1. The 1H
NMR signals (Figure 1A) clearly verified the composition of
polyester generated from PGA and ME2MO monomers. The molar
composition of the two monomers in the resultant polyester was
calculated to be 1:1 based on the integration of methylene d (2.22 to
2.82 ppm) from glutamic segment and methyl j (3.36 ppm) from
OEG side chain. The homopolymerizations of PGA and ME2MO
were both not able to be initiated by AIP. The results indicated that
Figure 3 shows DSC thermograms of the polyesters. The
glass-transition temperatures (Tg) of the polyesters were obtained
from the DSC curves and listed in Table 2. The Tg exhibited a
slight increase from ꢀ5.0 to ꢀ4.7 ꢀC with the increased degree of
2469
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
Table 2. Properties of P(PGA-co-ME2MO) Alternating Polyesters
polyester
CMC (10ꢀ3 g Lꢀ1
)
Rh (nm) at 25 ꢀC
Tg (ꢀC)a
DLC%b
DLE%b
DLC%c
DLE%c
P(PGA-co-ME2MO)23
P(PGA-co-ME2MO)47
48.1
9.55
100 ( 3.8
40.2 ( 4.7
ꢀ5.0
ꢀ4.7
25.7
27.0
34.5
37.1
32.6
33.8
48.4
51.1
a Determined by DSC. b Rifampin loading properties. c DOX loading properties.
Self-Assembly Behavior of P(PGA-co-ME2MO). It was inter-
esting that the obtained polyesters were amphiphilic because of
the hydrophobic polyester backbone and the hydrophilic OEG
side chains. As shown in Scheme 4, the amphiphilic polyester could
form micelles in water. The self-assembly behavior of the polyester
in aqueous solution was studied using fluorescence spectroscopy
(Figure 5). Pyrene was used as the fluorescence probe, and the
CMC was obtained from the plot of fluorescence intensity ratio of
I335/I332 versus log10 c of the polyester. The CMC values of P(PGA-
co-ME2MO)23 and P(PGA-co-ME2MO)47 were determined to be
4.81 ꢁ 10ꢀ2 and 9.55 ꢁ 10ꢀ3 g Lꢀ1, respectively (Table 2). The
lower CMC value for P(PGA-co-ME2MO)47 was probably attrib-
uted to the increase in polyester backbone length, which enhanced
the probability of polyester backbone winding and the interaction
among the polyester chains. This phenomenon also existed in
amphiphilic star-shaped copolymer.38 Figure 5C shows the typical
TEM micrograph of (PGA-co-ME2MO)23 micelles at 25 ꢀC. The
micelles took spherical shape.
Owing to the presence of OEG side chains in the surface,
the micelles were assessed to have thermosensitive properties.
Dynamic light scattering (DLS) technology was used to determine
the thermoinduced size change of the micelles. Figure 6 shows
the results of the DLS measurements of the polyester micelles in
0.1 g Lꢀ1 with temperature ranging from 25 to 60 ꢀC. The
hydrodynamic radii (Rh) of the P(PGA-co-ME2MO)23 micelles were
measured to decrease from 100 nm at 25 ꢀC to 32 nm at 60 ꢀC. The
decrease in Rh can be attributed to the rapid hydration-to-dehydration
of the OEG chains in the shell with the temperature increase. In
contrast, the Rh of P(PGA-co-ME2MO)47 micelles was only changed
from 40 nm at 25 ꢀC to 23 nm at 60 ꢀC. The lower degree of Rh
variation toward P(PGA-co-ME2MO)47 micelles may be due to the
longer hydrophobic chains in P(PGA-co-ME2MO)47, which form
more compact hydrophobic cores in the micelles. Therefore, these
micelles can be used as stable drug carriers that undergo thermo-
induced size decrease without intermicellar aggregation.
Figure 4. GPC chromatograms of P(PGA-co-ME2MO)47 at different
degradation times cultured at 37 ꢀC in CPBS (pH 7.4) with 2 μg mLꢀ1
(0.08 U mLꢀ1) proteinase K (A) and molecular weights of P(PGA-co-
ME2MO)47 as a function of degradation time (B).
In Vitro Cytotoxicity of P(PGA-co-ME2MO). The in vitro
cytotoxicities of P(PGA-co-ME2MO)23 and P(PGA-co-ME2MO)47
toward HeLa cells were evaluated using MTT assay. The sodium
dodecyl sulfonate (SDS) was used as the positive control. As shown
in Figure 7, the cell viability was treated with polyesters at different
concentrations for 24 h. It was observed that the viabilities of HeLa
polymerization (DP) from 23 to 47. This result was consistent with
the previous reports and could be commonly described by equation:
Tg(N) = Tg(¥) ꢀ C/N, where N is the degree of polymerization,
Tg(¥) is the glass-transition temperature for infinity molecular
weight, and C is the constant.15,36
cells were ∼85 to 100% at all test concentrations up to 100 mg Lꢀ1
,
The in vitro enzymatic degradation of the polyester was
carried out at 37 ꢀC in CPBS (pH 7.4) with 2 μg mLꢀ1 (0.08
U mLꢀ1) proteinase K.37 Figure 4 shows the GPC results after
incubation for different time intervals. As shown in Figure 4B, the
molecular weight (Mn) of incubated polyester decreased significantly
with the extension of incubation time. The Mn of polyester quickly
dropped to 1100 from 9900 within 10 h at the presence of proteinase
K according to GPC analyses. The high degradation rate was mainly
attributable to the high surface-to-volume ratio of the polyester
micelle, which increased significantly the interaction area between
the polyester and proteinase K in the buffer. It indicated that the
amphiphilic alternating polyesters can be used in drug delivery
system controlled by enzymatic degradation.
revealing the low toxicity and good compatibility of the polyesters
to cells.
Drug Loading and Release. To test the possible use of this
kind of micelles as drug carriers, rifampin, a bactericidal antibiotic
drug, was loaded into the micelles (Scheme 4) and the release
behaviors were studied. As listed in Table 2, the DLC% was 25.7 and
27.0%, depending on the compositions of polyesters, whereas the
DLE% was 34.5 and 37.1% for P(PGA-co-ME2MO)23 and P(PGA-
co-ME2MO)47 micelles, respectively. The rifampin release behaviors
were first studied in CPBS at pH 7.4, 37 ꢀC, and the results were
shown in Figure 8. It was interesting to note that the rifampin release
from the micelles followed an approximately zero-order kinetic,
2470
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
Scheme 4. Schematic Illustration of Micellization and Drug-Loading Procedure of Polyester
which was attractive for controlled drug release.39ꢀ41 This should
be ascribed to the hydrophobic interaction between rifampin and
N-phthaloyl groups and the compact structure of the micelles at
37 ꢀC. The hydrophobic interaction and compact structure also
led to lower degree of rifampin released from the micelles (<25%).
In addition, the polyester structure was found to be correlative to
the drug release kinetics. P(PGA-co-ME2MO)47 micelles exhibited
fewer rifampin release than P(PGA-co-ME2MO)23 micelles because
of the more compact hydrophobic cores in P(PGA-co-ME2MO)47
micelles.
To confirm the release behaviors and for further application,
we loaded DOX, an anticancer drug, into the micelles (Scheme 4),
and the release behaviors were also studied. The DLC% were 32.6
and 33.8%, whereas the DLE% were 48.4 and 51.1% for P(PGA-co-
ME2MO)23 and P(PGA-co-ME2MO)47 micelles, respectively. The
DLC% and DLE% of DOX was somewhat higher than that of
rifampin because of the lower molecular weight of DOX and the
chemical structure difference. In a similar way as rifampin, the
cumulative release of the DOX can reach almost 100% under the
presence of proteinase K, whereas lower release was observed at
either pH 7.4 (10%) or pH 5.5 (40%) without enzyme in the test
duration (72 h) (Figure 9). These release profiles are beneficial not
only for minimizing drug loss in blood but also for selective
accumulation in tumor tissue by the EPR effect, which may enhance
the overall therapeutic efficacy in vivo relative to free DOX.
Intracellular DOX Release. The cellular uptake and intracel-
lular release behaviors of DOX-loaded polyester micelles toward
HeLa cells were monitored by CLSM. The HeLa cells were
incubated with DOX-loaded micelles for 0.5 or 2 h with free DOX
as the control (10 mg Lꢀ1 DOX). As shown in Figure 10AꢀC,
results showed stronger DOX fluorescence in the cells after 0.5 h
incubated with free DOX than DOX-loaded polyester micelles,
indicating fast internalization of free DOX. With the incubation time
increased to 2 h, stronger DOX fluorescence was observed in the cells
incubated with DOX-loaded polyester micelles (Figure 10DꢀF). It
has been reported that the incorporation of DOX in the hydrophobic
cores of micelles decreases the DOX fluorescence intensity compared
with free DOX at the same concentration because of the self-
quenching effect of DOX.42,43 Therefore, the enhanced fluorescence
intensity should be the result of the high endocytosis efficiency of
DOX-loaded polyester micelles and enhanced intracellular release of
DOX with the micelles degradation in HeLa cells. It is worth noting
that the DOX fluorescence is stronger in the cells incubated with
DOX-loaded P(PGA-co-ME2MO)23 micelles than with DOX-
loaded P(PGA-co-ME2MO)47 micelles. This is because the DOX
release from DOX-loaded P(PGA-co-ME2MO)23 micelles is quicker
(as shown in Figure 9).
In addition, we further investigated the rifampin release in
vitro at pH 5.5 or in the presence of proteinase K at pH 7.4.
P(PGA-co-ME2MO)47 micelles were used in this study for their
relatively stability at pH 7.4 without enzyme, from which only 10%
of the loaded rifampin was released in the test duration (30 h). As
shown in Figure 8, the rifampin releases were accelerated at pH 5.5
or in the presence of proteinase K at pH 7.4. The release behaviors
both adopted zero-order kinetics for up to 12 h. It was also observed
that the cumulative release of the rifampin can reach almost 100%
under the presence of proteinase K, which was much higher than
that at pH 7.4 (10%) or pH 5.5 (40%) without enzyme. This should
be attributed to the presence of proteinase K, which can cause
effective degradation of the polyester backbone (Figure 4).37 GPC
analyses were used to verify the degradation of the polyester under
these conditions. As shown in Figure S4 (Supporting Information),
the polyester was stable in CPBS at pH 7.4 after performing rifampin
release for 30 h. The lower pH (pH 5.5) and the presence of
proteinase K (2 μg mLꢀ1, 0.08 U mLꢀ1) will partially lead to
degradation of the polyester, demonstrated by the GPC as decrease
in the Mn and broadening of PDI. So, accelerated drug release rate
and increased drug release amount could be observed in these
conditions as compared with that at pH 7.4 without enzyme.
On the basis of the aforementioned data, we can conclude that
these novel amphiphilic alternating polyesters are of great potential
use as smart drug carriers, which are stable under physiological
condition and can undergo sustained release at the targeted sites
with stimulus, such as lower pH or enzyme.
2471
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
Figure 7. In vitro cytotoxicity of P(PGA-co-ME2MO) to HeLa cells
with SDS as the positive control. Data were presented as the average (
standard deviation (n = 3).
Figure 8. Release profiles of rifampin from rifampin-loaded P(PGA-co-
ME2MO)23 micelles in CPBS at 37 ꢀC, pH 7.4 (a), and P(PGA-co-
ME2MO)47 micelles in CPBS at 37 ꢀC, pH 7.4 (b), pH 5.5 (c), and pH
7.4 with 2 μg mLꢀ1 (0.08 U mLꢀ1) proteinase K (d).
Figure 5. Excitation spectra of pyrene in aqueous solution of
P(PGA-co-ME2MO)23 at different concentrations (λem = 390 nm)
(A), the intensity ratio (I335/I332) as a function of concen-
tration (B), and typical TEM micrograph of P(PGA-co-ME2-
MO)23 (C).
Figure 9. Release profiles of DOX from DOX-loaded P(PGA-co-
ME2MO)23 micelles in CPBS at 37 ꢀC, pH 7.4 (a), and P(PGA-co-
ME2MO)47 micelles in CPBS at 37 ꢀC, pH 7.4 (b), pH 5.5 (c), and pH
7.4 with 2 μg mLꢀ1 (0.08 U mLꢀ1) proteinase K (d).
In Vitro Activity of DOX-Loaded P(PGA-co-ME2MO) Micelles.
As discussed above, the DOX-free polyester micelles showed no
noticeable cytotoxicities up to 100 mg Lꢀ1 (Figure 7). In vitro
activity of free DOX and DOX-loaded polyester micelles for HeLa
cells was estimated. As shown in Figure 11, DOX-loaded polyester
micelles showed more effective inhibitory effect on the proliferation
of cancer cells compared with the free DOX. The higher toxicity of
DOX-loaded polyester micelles (IC50 = 3.28, 3.42 mg Lꢀ1 for
Figure 6. Hydrodynamic radii (Rh) of P(PGA-co-ME2MO) in aqueous
solution as a function of temperature.
2472
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
’ CONCLUSIONS
Novel amphiphilic P(PGA-co-ME2MO) polyesters with pen-
dent OEG chains were prepared by alternating ROP of PGA and
ME2MO monomers. The synthesized polyesters were verified to
have precise alternating structure and amphiphilic properties.
They can form micelles in the aqueous solution with hydro-
phobic polyester backbone cores and thermosensitive OEG
shells. The micelles underwent thermoinduced Rh decrease to
form compact micelles without intermicellar aggregation. Rifam-
pin and DOX were loaded into the compact micelles. The
sustained zero-order release was observed in the test duration
at pH 7.4, 37 ꢀC. However, the cumulative release of drug from
the micelles was relatively low. In particular, for P(PGA-co-
ME2MO)47 micelles, only 10% of the drug can be release from
micelles. Fortunately, the sustained release rate can be acceler-
ated at lower pH 5.5 or in the presence of proteinase K at pH 7.4
due to the degradation of the polyesters under these conditions.
The in vitro cell experiments revealed that these polyester
micelles accomplished fast release of DOX inside cells and higher
anticancer efficacy as compared with the free DOX control.
Therefore, it is believed that this kind of polyesters may hold vast
potential for controlled drug delivery systems, which show
stability under physiological condition and accelerated release
at the targeting sites (e.g., low pH or enzyme presence).
’ ASSOCIATED CONTENT
Supporting Information. 1H NMR and FT-IR spectra of
S
b
monomers and GPC chromatograms of polyester after rifampin
release. This material is available free of charge via the Internet at
http://pubs.acs.org.
Figure 10. Representative CLSM images of HeLa cells incubated with
free DOX and DOX-loaded micelles: 0.5 h, incubated with free DOX
(A); 0.5 h, incubatedwith DOX-loaded P(PGA-co-ME2MO)23 micelles (B);
0.5 h, incubated with DOX-loaded P(PGA-co-ME2MO)47 micelles (C);
2 h, incubated with free DOX (D); 2 h, incubated with DOX-loaded P(PGA-
co-ME2MO)23 micelles (E); 2 h, incubated with DOX-loaded P(PGA-co-
ME2MO)47 micelles (F). For each panel, the images from left to right show
differential interference contrast (DIC) image, cell nuclei stained by DAPI
(blue), DOX fluorescence in cells (red), and overlays of the three images.
The bar represents 40 μm.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: zhuangxl@ciac.jl.cn.
Author Contributions
§These two authors are joint first authors.
’ ACKNOWLEDGMENT
This work was supported by grants from National Natural
Science Foundation of China (Key Project 50733003, Project
20904053 and 50973108, A3 Foresight Program 50425309),
Program of Scientific Development of Jilin Province (20090135),
and Knowledge Innovation Program of the Chinese Academy of
Sciences (Grant No. KJCX2-YW-H19). We would like to thank Ms.
Lili Ma from Changchun Institute of Applied Chemisty for her help
with the polymer characterization.
’ REFERENCES
(1) Gaucher, G.; Satturwar, P.; Jones, M. C.; Furtos, A.; Leroux, J. C.
Eur. J. Pharm. Biopharm. 2010, 76, 147–158.
(2) Li, C. H.; Ge, Z. S.; Fang, J.; Liu, S. Y. Macromolecules 2009,
42, 2916–2924.
(3) Lao, S. B.; Zhang, Z. X.; Xu, H. H.; Jiang, G. B. Carbohydr. Polym.
2010, 82, 1136–1142.
Figure 11. In vitro cytotoxicity of DOX-loaded polyester micelle to
HeLa cells with free DOX as the control. Data were presented as the average
( standard deviation (n = 3).
(4) Petersen, M. A.; Yin, L. G.; Kokkoli, E.; Hillmyer, M. A. Polym.
Chem. 2010, 1, 1281–1290.
(5) Wan, W. M.; Pan, C. Y. Polym. Chem. 2010, 1, 1475–1484.
(6) Liu, G. Y.; Hu, X. F.; Chen, C. J.; Ji, J. A. J. Appl. Polym. Sci. 2010,
118, 3197–3202.
P(PGA-co-ME2MO)23, P(PGA-co-ME2MO)47 micelles, respec-
tively) compared with free DOX (IC50 = 3.79 mg Lꢀ1) was
probably due to the effective endocytosis and quick intracellular
DOX release within the endosomes.
2473
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474

Biomacromolecules
ARTICLE
(7) Wang, C.; Chen, Q. S.; Xu, H. P.; Wang, Z. Q.; Zhang, X. Adv.
Mater. 2010, 22, 2553–2555.
(41) Liu, H. B.; Farrell, S.; Uhrich, K. J. Controlled Release 2000,
68, 167–174.
(8) Liu, Z. H.; Tian, Y. M.; Sun, F.; Meng, F. Y.; Zheng, Y. H.; Jiang,
F. M. Colloids Surf., A 2008, 318, 122–124.
(42) Sun, H. L.; Guo, B. N.; Li, X. Q.; Cheng, R.; Meng, F. H.; Liu,
H. Y.; Zhong, Z. Y. Biomacromolecules 2010, 11, 848–854.
(43) Wang, Y. C.; Li, Y.; Sun, T. M.; Xiong, M. H.; Wu, J. A.; Yang,
Y. Y.; Wang, J. Macromol. Rapid Commun. 2010, 31, 1201–1206.
(9) Chen, X. J.; Sanchez-Gaytan, B. L.; Hayik, S. E. N.; Fryd, M.;
Wayland, B. B.; Park, S. J. Small 2010, 6, 2256–2260.
(10) Easwaramoorthi, S.; Kim, P.; Lim, J. M.; Song, S.; Suh, H.;
Sessler, J. L.; Kim, D. J. Mater. Chem. 2010, 20, 9684–9694.
(11) Hung, S. C.; Shiu, S. C.; Chao, J. J.; Lin, C. F. J. Electrochem. Soc.
2010, 157, D496–D499.
(12) Savariar, E. N.; Aathimanikandan, S. V.; Thayumanavan, S.
J. Am. Chem. Soc. 2006, 128, 16224–16230.
(13) Li, Y. G.; Zhang, Y. Q.; Yang, D.; Li, Y. J.; Hu, J. H.; Feng, C.;
Zhai, S. J.; Lu, G. L.; Huang, X. Y. Macromolecules 2010, 43, 262–270.
(14) Ding, J. X.; Xiao, C. S.; Tang, Z. H.; Zhuang, X. L.; Chen, X. S.
Macromol. Biosci. 2011, 11, 192–198.
(15) Ding, J. X.; Xiao, C. S.; Zhao, L.; Cheng, Y. L.; Ma, L. L.; Tang,
Z. H.; Zhuang, X. L.; Chen, X. S. J. Polym. Sci., Part A: Polym. Chem. 2011,
49, 2665–2676.
(16) Qiu, S.; Huang, H.; Dai, X. H.; Zhou, W.; Dong, C. M. J. Polym.
Sci., Part A: Polym. Chem. 2009, 47, 2009–2023.
(17) Gou, P. F.; Zhu, W. P.; Shen, Z. Q. Polym. Chem. 2010,
1, 1205–1214.
(18) Elfinger, M.; Pfeifer, C.; Uezguen, S.; Golas, M. M.; Sander, B.;
Maucksch, C.; Stark, H.; Aneja, M. K.; Rudolph, C. Biomacromolecules
2009, 10, 2912–2920.
(19) Heath, F.; Saeed, A. O.; Pennadam, S. S.; Thurecht, K. J.;
Alexander, C. Polym. Chem. 2010, 1, 1252–1262.
(20) Konda, M.; Kimura, Y. Sen0i Gakkaishi 2003, 59, 319–323.
(21) Hiltunen, K.; Seppala, J. V.; Itavaara, M.; Harkonen, M.
J. Environ. Polym. Degrad. 1997, 5, 167–173.
(22) Duong, H. T. T.; Nguyen, T. L. U.; Stenzel, M. H. Polym. Chem.
2010, 1, 171–182.
(23) Van der Ende, A. E.; Sathiyakumar, V.; Diaz, R.; Hallahan, D. E.;
Harth, E. Polym. Chem. 2010, 1, 93–96.
(24) Taghdisi, S. M.; Abnous, K.; Mosaffa, F.; Behravan, J. J. Drug
Targeting 2010, 18, 277–281.
(25) Choi, J. H.; Jang, J. Y.; Joung, Y. K.; Kwon, M. H.; Park, K. D.
J. Controlled Release 2010, 147, 420–427.
(26) Du, Y. Z.; Weng, Q.; Yuan, H.; Hu, F. Q. ACS Nano 2010,
4, 6894–6902.
(27) Guo, M. A.; Que, C. L.; Wang, C. H.; Liu, X. Z.; Yan, H. S.; Liu,
K. L. Biomaterials 2011, 32, 185–194.
(28) Mikhail, A. S.; Allen, C. Biomacromolecules 2010, 11, 1273–
1280.
(29) Janib, S. M.; Moses, A. S.; MacKay, J. A. Adv. Drug Delivery Rev.
2010, 62, 1052–1063.
(30) Lutz, J. F. J. Polym. Sci., Part A: Polym. Chem. 2008,
46, 3459–3470.
(31) Qiao, Z. Y.; Du, F. S.; Zhang, R.; Liang, D. H.; Li, Z. C.
Macromolecules 2010, 43, 6485–6494.
(32) Feng, L.; Hao, J. Y.; Xiong, C. D.; Deng, X. M. Chem. Commun.
2009, 4411–4413.
(33) Albertsson, A. C.; Varma, I. K. Biomacromolecules 2003,
4, 1466–1486.
(34) Wang, Y. C.; Shen, S. Y.; Wu, Q. P.; Chen, D. P.; Wang, J.;
Steinhoff, G.; Ma, N. Macromolecules 2006, 39, 8992–8998.
(35) Tipson, R. S. J. Org. Chem. 1956, 21, 1353–1357.
(36) Odriscoll, K.; Sanayei, R. A. Macromolecules 1991,
24, 4479–4480.
(37) Zeng, J.; Chen, X. S.; Liang, Q. Z.; Xu, X. L.; Jing, X. B.
Macromol. Biosci. 2004, 4, 1118–1125.
(38) Cheng, J.; Ding, J. X.; Wang, Y. C.; Wang, J. Polymer 2008,
49, 4784–4790.
(39) Hu, Y.; Jiang, X. Q.; Ding, Y.; Zhang, L. Y.; Yang, C. Z.; Zhang,
J. F.; Chen, J. N.; Yang, Y. H. Biomaterials 2003, 24, 2395–2404.
(40) Li, X.; Wu, Q.; Lu, M.; Zhang, F.; Lin, X. F. J. Polym. Sci., Part A:
Polym. Chem. 2008, 46, 2734–2744.
2474
dx.doi.org/10.1021/bm200668n |Biomacromolecules 2011, 12, 2466–2474