RESEARCH PAPER
ice | science
Key words: Biopolymer, Biodegradable, Drug delivery, Click
chemistry, Self-assembly
.
ICE Publishing: All rights reserved
Cyclodextrin-based Biodegradable Polymer
Stars: Synthesis & Fluorescence Studies
Gayan K. Tennekone, MSc
Department of Chemistry, University of Prince Edward Island, Charlottetown, PE, Canada
*Brian D. Wagner, PhD
Department of Chemistry, University of Prince Edward Island, Charlottetown, PE, Canada, bwagner@upei.ca
*Michael P. Shaver, PhD
Department of Chemistry, University of Prince Edward Island, Charlottetown, PE, Canada and School of Chemistry, University of Edinburgh,
Edinburgh, UK
Polymer stars built using aliphatic polyester arms and a β-cyclodextrin (CD) core are prepared by two synthetic
methodologies. The CD core stars offer the intriguing potential of loading molecules of interest into two zones by exploiting
the host-guest chemistry of the hydrophobic cyclodextrin core and by physical trapping in the polymer arms. Core-first
syntheses were achieved through the ring-opening polymerization of rac-lactide,
L-lactide,
β-butyrolactone and
lactide/glycolide monomers generating 7-armed stars with a heptakis(2,6-di-O-methyl)-β-cyclodextrin core. Arm-first
syntheses were achieved through the copper-catalyzed azide-alkyne cycloaddition of alkyne-terminated poly(lactic acid)s,
poly(3-hydroxybutyrate) and a heptakis-azido-β-cyclodextrin. For both synthetic strategies use of the industry standard
catalyst, Sn(Oct)2, gave polymers with broadened dispersities (1.3-1.7) compared to aluminum complexes supported by
salen ligands (< 1.2). Synthetic strategies were compared by both measures of reaction control (molecular weight, dispersity,
conversion) and controlled release (fluorescence spectroscopy of hydrolytic and enzymatic degradation), each offering
benefits to the synthetic polymer chemist, variable sustainability and scalability and a clear direction for further star design.
Fluorescence-based controlled release studies were performed in water or methanol, releasing the encapsulated 7-MC
fluorescent probe through both hydrolytic and enzymatic degradation. The release was shown to be strongly accelerated in
the presence of the enzyme.
1.
Introduction
Polymer stars have a robust history in biomedicine and nanotechnology. The unique star macrostructure adopts the physical properties of
the polymer arms whilst maintaining a much higher total molecular weight.1,2 Combined with higher end-group concentrations these
systems have been used as novel and tailorable platforms in applied polymer research. 1,2 One particular area of application is polymeric
drug delivery systems with potentially superior release properties and controlled degradation timeframes.3,4
A key component of controlled release polymer stars is a green polymer with a tunable degradation profile. Our team and many others1
are particularly interested in aliphatic polyesters as sustainable, biodegradable and hydrolytically degradable arms in star polymers. Past
efforts have focused on poly(lactic acid) (PLA,), poly(3-hydroxybutyrate) (PHB) and poly(lactic-co-glycolic acid) (PLGA) as
biodegradable and biocompatible polymers.5-7 PLA, prepared readily via ring-opening polymerization (ROP) of the racemic, rac, and
stereoregular, L, forms of lactide, is extensively used in commodity and biomedical applications8-10 while PHB is produced by bacteria
and algae11 and derived synthetically from the ROP of β-butyrolactone.6,12 Polymer properties can be further modified through
copolymerization. In particular, the ratio of lactic acid to glycolic acid linkages in PLGA can vary (bio)degradation rates.7 Building these
polymers into larger macromolecular frameworks can offer a second variable for application-specific polymer development.
We have previously explored the ROP of lactide to synthesize polymer stars with polyol, rigid and dendritic cores. 13-16 Of interest,
however, is the potential of cyclodextrins, cyclic oligosaccharides that consist of 6 (α), 7 (β) and 8 (γ) α-1,4-linked D-glucopyranose
units,17,18 to act as cores. These non-toxic, biocompatibile and inexpensive cores combined with biodegradable and biocompatible arms
are potentially ideal materials for sustainable biomedical, fragrance and flavor applications where controlled release is essential.
Cyclodextrins have an internal nanocavity that is accessible to a wide array of organic substrates. 19 Inclusion of a smaller guest molecule
within the CD cavity results in the formation of a supramolecular host-guest inclusion complex.20 The ability of CDs to form inclusion
complexes has inspired the development of new biomaterials in polymer chemistry which have been used as drug,21 protein22 and other
compound23 delivery vectors, stimuli-responsive hydrogels24 and to solubilize monomers and chain transfer agents25,26 The presence of a
host cavity and a polymer arm corona gives CD-polymer stars two regions to encapsulate molecules of interest, potentially allowing for
increased loadings, monitoring of drug delivery or complex multi-drug delivery profiles built from both a green polymer and sustainable
core.
The field of PLA-decorated polymer stars is robust. Functionalization of cyclodextrins with polymers has been achieved through atom
transfer radical polymerization, anionic polymerization and arm grafting.17,21-23,27 ROP of cyclic esters in the presence and absence of a
co-catalyst28 can access these important star macrostructures and utilize sustainable polymers. Coupled with the sustainable core these
materials can be viewed as entirely biodegradable, bioassimilable polymer systems. However, obtaining materials of low dispersity
remains an issue.28a Our goal is to contribute to this area by systematically investigating both core-first and arm-first synthetic strategies
for CD-star polymer synthesis while expanding the scope of catalysts and monomers. Core-first stars are synthesized using ROP and a
methylated core: heptakis(2,6-di-O-methyl)-β-cyclodextrin (Figure 1). Arm-first stars were prepared via copper-catalyzed azide-alkyne
cycloaddition (CuAAC) coupling a heptakis-azido-β-cyclodextrin core (Figure 1) and linear acetylene-terminated polymer arms. In this
work, modified β-cyclodextrin derivatives were used as initiators for the synthesis of polymer stars based on poly(lactic acid), poly(3hydroxybutyrate) and poly(lactic-co-glycolic acid). Loading and release studies were performed using the 7-methoxycoumarin (7-MC)
fluorophore, which exhibits fluorescence suppression upon inclusion into the hydrophobic environment of CD cavities, 29 and thus will
also show fluorescence suppression upon encapsulation into the hydrophobic environment of the polymer star in these materials.
Degradation methodologies (hydrolytic and enzymatic) were compared to develop a protocol compatible with fluorescence detection.
Figure 1. Structure of heptakis(2,6-di-O-methyl)-β-cyclodextrin and a heptakis-azido- β-cyclodextrin.
2.
Experimental
2.1
Materials.
HPLC-grade toluene, dichloromethane and methanol were purchased from Fisher Scientific. Deuterated chloroform and dimethyl
sulfoxide were purchased from Cambridge Isotopes. Spectroscopic-grade dimethylformamide was obtained from Sigma Aldrich.
Anhydrous toluene was obtained by passing the solvent through an MBraun solvent purification system consisting of columns of alumina
and a copper catalyst and was degassed prior to use. Chemicals for catalyst synthesis including trimethylaluminum (2.0M solution in
hexanes), 3,5-di-tert-butyl-2-hydroxybenzaldehyde, 1,3-diaminopropane and paraformaldehyde were purchased from Aldrich Chemicals
and used as received. Tin(II) ethylhexanoate was purchased from Aldrich Chemicals and distilled under reduced pressure prior to use. βCyclodextrin and heptakis(2,6-di-O-methyl)-β-cyclodextrin were purchased from Aldrich Chemicals. The monomers rac- and L-lactide
were purchased from PURAC Biomaterials and were purified by three successive vacuum sublimations prior to use. Glycolide were
purchased from Aldrich Chemicals and was used as received. β-Butyrolactone and ε-caprolactone were purchased from Aldrich
Chemicals, dried over CaH2 and distilled under reduced pressure before use. Aluminum catalyst tBu[salen]AlMe, where tBu[salen] is
N,N’-bis(3,5-di-tert-butylsalicylidene)-1,3-propanediamine, was synthesized according to previously reported procedures. 30 5-Hexyn-1-ol
was purchased from Aldrich chemicals and was distilled under vacuum at 80 °C prior to use. Copper(II) sulfate was purchased from
Aldrich Chemicals and sodium ascorbate from Alfa Aesar. Heptakis-azido-β-cyclodextrin was synthesized according to a previously
reported literature procedure.31
2.2
General Considerations.
All experiments involving moisture and air-sensitive compounds were performed under a nitrogen atmosphere using an MBraun
LABmaster sp glovebox system equipped with a -35 °C freezer and [H2O] and [O2] analyzers. All GPC analyses were performed in THF
(flow rate: 1 mL min-1) at 50 °C with a Polymer Laboratories PL-GPC 50 Plus integrated GPC system with two 300 × 7.8 mm Jordi Gel
DVB mixed bed columns coupled to a Wyatt miniDAWN Treos Light Scattering system using dn/dc values for PLA, PHB, PLGA and
PCL of 0.0558, 0.0670, 0.0540 and 0.0650 respectively. 32-35 All fluorescence experiments were performed using a Photon Technology
International RF-M2004 luminescence spectrophotometer equipped with a PTI LPS-220 lamp power supply, a xenon 75 watt arc lamp, a
PTI-610 photomultiplier tube detector, and a PTI MD-5020 motor driver. All samples were excited at 320 nm and the emission was
scanned from 330-630 nm using a step size of 1 nm and integration time of one second. 1H NMR spectra were recorded with a Bruker
Avance Spectrophotometer (300 MHz) in CDCl3.
2.3
Core-First Polymer Stars.
All polymerization experiments were performed in sealed ampules under an inert nitrogen atmosphere at 120 °C in toluene using a
varying monomer ratio (70, 140, 210, 280 and 350 eq.) and a core:catalyst ratio of 1:0.5 for Sn(Oct)2 and 1:1 for tBu[salen]AlMe.
Monomers used were rac-lactide, L-lactide, β-butyrolactone and glycolide, with heptakis(2,6-di-O-methyl)-β-cyclodextrin as the initiator.
Poly(lactic acid)-co-(glycolic acid) was prepared using a 1:1 mixture of lactide and glycolide monomers. The reactions were carried out
for 3 hours and then quenched with a 10:1 v/v mixture of CH2Cl2/MeOH solution and stirred for 30 minutes. The resulting mixture was
precipitated into cold pentane, filtered and dried in vacuo prior to analysis. A specific exemplary procedure for the case of stars with 30
monomer length arms is as follows: rac- or L-lactide (0.500 g, 3.5 mmol), heptakis(2,6-di-O-methyl)-β-cyclodextrin (0.022 g, 0.02
mmol), Sn(Oct)2 (0.005 g, 0.01 mmol) or tBu[salen]AlMe (0.010 g, 0.02 mmol) and toluene were added to an ampule equipped with a
magnetic stir bar in the glovebox. The ampule was sealed and placed into an oil bath at 120 °C for 3 hours. The ampule was then
removed from the oil bath and cooled to room temperature. The vessel was opened to the atmosphere and quenched with 5 mL of a 10:1
v/v mixture of CH2Cl2 and MeOH. The resulting solution was stirred for 30 minutes and then precipitated by dropwise addition into 100
mL of cold pentane. The polymer was then filtered and dried in vacuo for 24 hours. Percent conversion was obtained from 1H-NMR
spectroscopy of the crude material and was determined to be 72%. 1H NMR (300 MHz, CDCl3, 25°C) δ: 5.16 (m, PLA-CH), 3.63 (s, 21H,
OCH3), 3.40 (s, 21H, OCH3), 1.53 (m, PLA-CH3).
2.3.1 5-Hexyn-1-ol polymer synthesis.
The same general procedure was followed for all 5-hexyn-1-ol terminated polymer chains with varying arm lengths. 5-Hexyn-1-ol (0.011
g, 0.12 mmol), Sn(Oct)2 (0.025 g, 0.06 mmol) and the desired monomer (loadings at 1.20, 2.40, 3.60, 4.80, 5.00 mmol) in toluene were
added to an ampule under an N2 atmosphere. The ampule was then placed in an oil bath at 70 °C and the reaction was set to stir for 3
hours. The reaction was then quenched with 5 mL of a 10:1 v/v mixture of CH 2Cl2 and MeOH. The resulting solution was stirred for 30
minutes and then precipitated by dropwise addition into 100 mL of cold pentane. The polymer was then filtered and dried in vacuo for 24
hours prior to analysis by NMR and GPC.
2.3.2
Arm-First Polymer Stars.
All CuAAC reactions were performed at room temperature in DMF using heptakis-azido- β-cyclodextrin (0.013 g, 0.01 mmol), 5-hexyn1-ol terminated polymer, CuSO4 (0.002 g, 1.2 × 10-3 mmol) and sodium ascorbate (0.010 g, 005 mmol). Reactions were carried out for 24
hours and then precipitated into deionized H2O, filtered and dried at 70 °C in a vacuum oven prior to analysis. A specific exemplary
procedure, for the case of PLA stars with 30 monomer length arms, is as follows: PLA (0.187 g, 0.07 mmol), β-CD-(N3)7 (0.013, 0.01
mmol), CuSO4 (0.002 g, 1.2 × 10-3 mmol) and sodium ascorbate (0.010 g, 0.05 mmol) were dissolved in DMF (2 mL). The reaction
mixture was stirred for 24 hours at room temperature and then precipitated into 50 mL of deionized H2O, filtered and dried at 70 °C in a
vacuum oven. Percent conversion by 1H NMR spectroscopy of crude samples was determined to be 57%. 1H NMR (300 MHz, CDCl3,
25°C) δ: 5.90 (br, 7H, β-CD-(N3)7), 5.75 (br, 7H, , β-CD-(N3)7), 5.22 (m, PLA-CH), 4.21 (m, 7H, β-CD-(N3)7), 4.11 (m, 7H, β-CD-(N3)7),
1.55 (m, PLA-CH3).
2.3.3
Degradation and Release Experiments.
All controlled release experiments were performed at room temperature. The polymer samples (100.0 mg) were pressed at 2000 psi into
thin pellets. 3.0×10-5 M-1 and 1.5×10-5 M-1 stock solutions of 7-MC were prepared by dissolving 0.20 mg or 0.10 mg of 7-MC in 100 mL
of a 1:1 solution of H2O/MeOH and 100 mL of H2O respectively. The uncatalyzed pellets were then placed in 20 mL of a 1:1 mixture of
H2O and MeOH or pure H2O in order for hydrolytic degradation to occur. For enzymatic degradation 8 mL of a 1:1 H 2O/MeOH solution
containing 16 mg (2 mg/mL) of proteinase K with about 5 mg of CaCl2 was used. The degradation was monitored at desired time
intervals by agitating the solution and then measuring the fluorescence spectrum of a 3 mL aliquot of the solution and comparing the
measured emission to that of the corresponding standard 7-MC stock solution. Both fluorescence spectra were corrected by subtracting
the measured spectrum of the corresponding solvent blank.
3.
Results and Discussion
3.1
Synthesis of Core-First Polymer Stars
Initial studies using unsubstituted β-CD as a core initiator showed low polymer yields, inconsistent initiation and poor correlation
between molecular weight and monomer:core ratios. Building from a hypothesis of poor core solubility we investigated the
commercially-available heptakis(2,6-di-O-methyl)-β-CD cyclodextrin. The core exhibits excellent binding with 7-MC (Figures S1 to S3,
Supporting Information) and increased solubility in organic media compared to an unmodified β-cyclodextrin. The seven alcohol
functionalities on the 3 carbon served as macroinitiators for the ring-opening polymerization of lactide, β-butyrolactone and glycolide.
Sn(Oct)2 (1) and tBu[salen]AlMe (2) (Figure 2) were chosen as catalysts because of their ubiquitous use and excellent control over both
linear and star PLA polymer macrostructures.1 Catalyst 1 is the most widely used industrial catalyst for PLA synthesis, yielding atactic
polymer with narrow dispersities.
Figure 2. Catalyst systems Sn(Oct)2 (1) and tBu[salen]AlMe (2).
Standard reaction conditions used for polyol stars were first explored (Scheme 1). 13 A catalyst:CD ratio of 1:0.071 provided one Sn
catalyst for every two alcohol functionalities (or a 1:2 ratio of catalyst to CD-alcohols). Variable monomer:initiator ratios determined the
total molecular weight of the star and length of the polymer arms. Polymerization of rac-lactide by 1 at 70 °C in toluene for 3 hours gave
no polymeric product, suggesting no initiation. Increasing reaction temperatures to 120 °C produced poly(lactic acid) with monomodal
GPC traces. Reactions were quenched with a 10:1 mixture of CH2Cl2 and MeOH and then precipitated, filtered and dried in vacuo for 24
hours prior to analysis. Results are summarized in Table 1.
Scheme 1. CD poly(lactic acid) polymer stars based on rac-lactide and Sn(Oct)2 (1).
[M]:[C]:[I]
% Conv.a
Mn,thb
Mn,GPCc
Mn, NMRd
Đc
Đarme
70:0.5:1
57
7082
29310
17210
1.25
3.11
140:0.5:1
60
13438
37900
25700
1.37
3.78
210:0.5:1
72
23307
54150
49070
1.28
3.35
280:0.5:1
70
29581
70790
52590
1.35
3.54
350:0.5:1
78
40679
87040
71330
1.41
3.96
Polymerizations were conducted in 2 mL of toluene with 0.01 mmol of Sn(Oct)2 at 120 °C for 3 hours, using heptakis(2,6-di-O-methyl)-β-CD as an initiator. a %
Conv. determined by gravimetric analysis. b Mn,th = [M](per arm) × 7(Initiating Groups) × MW(Lactide) × (% conv.) + MW(core)
c Determined
by GPC using
refractive index coupled with light scattering detectors. d Mn from 1H NMR spectroscopy. e Calculated from the Szymanski method.36
Table 1. Polymerization data for CD poly(lactic acid) polymer stars based on rac-lactide and Sn(Oct)2 (1).
Polymers were analyzed using GPC and 1H NMR spectroscopy. Characteristic resonances in the 1H NMR spectra were located at δ 3.63
and 3.39 ppm, corresponding to the methoxy groups of the cyclodextrin core located on the 2 and 6 carbons. Integration relative to the
methyl and methine protons of the PLA chains gave Mn,NMR values notably different to the molecular weights determined directly from
GPC. The disappearance of the original resonances confirmed that all initiation sites were substituted with polymer arms. Monomer
conversion was determined gravimetrically and used to calculate the expected theoretical molecular weights (Mn,th) for a living
polymerization. Both Mn,GPC and Mn,NMR values were significantly higher than the Mn,th values for all entries suggesting a lack of control.
The Szymanski method,36 often used to analyze complex macromolecular architectures and determine arm uniformity (Đarm), confirmed
only moderate levels of control. Coupled with slightly broadened dispersities (Đs >1.25) either poor initiation or transesterification at the
high polymerization temperature prevented optimal star formation.
Similar trends were observed when the modified CD core was employed in L-lactide polymerization (Table 2). A better agreement
between the experimental and theoretical molecular weights are observed from these isotactic derivatives at low monomer loadings, but
high Mn stars showed significant inconsistencies, potentially due to the insolubility of these semi-crystalline polymers in organic media.37
Molecular weights do not correlate to monomer:initiator ratio. In contrast, molecular weights obtained by 1H NMR spectroscopy, which
assumes a pseudo-living polymerization, showed a linear increase in molecular weights with monomer:initiator ratio, suggesting that
transesterification may be reducing the length of the L-lactic acid arms. Regardless of the initiator, reaction conditions or characterization
technique tested, 1 offered inadequate control over the synthesis of β-CD polymer stars.
[M]:[C]:[I]
% Conv.a
Mn,thb
Mn, GPCc
Mn, NMRd
Đc
Đarme
70:0.5:1
50
7082
19630
13670
1.27
3.31
140:0.5:1
60
13438
19270
25780
1.29
3.34
210:0.5:1
64
20702
14900
52160
1.41
4.46
280:0.5:1
68
28774
16080
54270
1.40
4.33
350:0.5:1
80
41688
16130
68880
1.31
3.59
Polymerizations were conducted in 2 mL of toluene with 0.01 mmol of Sn(Oct)2 at 120 °C for 3 hours, using heptakis(2,6-di-O-methyl)-β-CD as an initiator. a %
Conv. determined by gravimetric analysis. b Mn,th = [M](per arm) × 7(Initiating Groups) × MW(Lactide) × (% conv.) + MW(core)
refractive index coupled with light scattering detectors. Mn from
d
1H
NMR spectroscopy.
e Calculated
c Determined
by GPC using
from the Szymanski method.
Table 2. Polymerization data for CD poly(lactic acid) polymer stars based on L-lactide and Sn(Oct)2 (1).
Catalyst 2 offered potential improvements to the control problems observed with the tin mediator. We have recently shown that 2 has a
wider monomer scope for both the living and immortal polymerization of cyclic esters than previously reported, with little observed
transesterification, and thus may offer improved control in star synthesis. 12a Polymerizations were performed under the optimized
conditions of 120 °C in toluene for 3 hours using rac-lactide to afford isospecific PLA arms: the bulky tBu salen derivative supports a
catalyst that provides excellent control over molecular weight, Đ and tacticity.30,38 Results obtained using 2 are summarized in Table 3.
Although significant deviations between theoretical molecular weight and M n values obtained from both GPC and 1H NMR spectroscopy
are still observed, control over the polymerization is excellent with Đs of <1.15, Đarm values <1.60, and well correlated relationships
between molecular weights and monomer:initiator ratios. The deviations in molecular weights, especially in light of the experimental
levels of control, are suggestive of challenges in gravimetric conversion determinations. Loss of low molecular weight fractions would
both lower conversion, and thus Mn,th, and increase Mn,GPC. Another possibility is that not all CD cores are initiating due to their poor
solubility in organic media. Once activated, the core becomes much more soluble and polymer chains propagate, therefore any uninitiated
cores result in lower conversions and higher molecular weight deviations would be observed. However, we have not observed the
presence of unreacted cyclodextrin in any of our crude reaction mixtures.
[M]:[C]:[I]
% Conv.a
Mn,thb
Mn, GPCc
Mn, NMRd
Đd
Đarme
70:1:1
62
7081
24490
13880
1.06
1.47
140:1:1
59
12691
28890
25130
1.07
1.53
210:1:1
59
18643
53340
34880
1.15
1.51
280:1:1
69
32933
55470
41760
1.07
1.53
350:1:1
65
35208
76600
44680
1.06
1.44
Polymerizations were conducted neat with 0.02 mmol of tBu[salen]AlMe at 120 °C for 3 hours, using heptakis(2,6-di-O-methyl)-β-CD as an initiator. a % Conv.
determined by gravimetric analysis. b Mn,th = [M](per arm) × 7(Initiating Groups) × MW(Lactide) × (% conv.) + MW(core)
c Determined
by GPC using refractive
index coupled with light scattering detectors. d Mn from 1H NMR spectroscopy. e Calculated from the Szymanski method.
Table 3. Polymerization data for CD poly(lactic acid) polymer stars based on rac-lactide and tBu[salen]AlMe (2).
The promising results obtained with 2 and lactide led us to expand our scope to polymer stars with PHB and PLGA arms (Table 4). With
our aluminum-based catlysts the PHB stars displayed excellent control with Đs as low as 1.01 and Đarm values <2.70, representing the
best control reported for PHB core-first star synthesis.1 Percent conversions for the synthesized stars were also very high, with >90 %
monomer conversion determined by 1H NMR spectroscopy after 3 hours. Likewise, a library of PLGA stars were synthesized using a 1:1
ratio of rac-lactide and glycolide with 2 at 120 °C in toluene for 3 hours. These copolymer stars displayed moderate levels of control for
PLGA, with Đs <1.31 and Đarm values <3.70. Correlation between molecular weights from GPC and 1H NMR spectroscopy was
improved, although at higher monomer loadings a greater deviation was observed potentially due to mass transport issues in the
extremely viscous solution. Molecular weights, especially when obtained from 1H NMR spectroscopy, showed excellent correlation with
expected molecular weights: poor solubility in THF may have affected obtained GPC molecular weights. Percent conversions, which
were determined by gravimetric analysis, were moderate to high. The synthesis of these stars highlights the versatility of both CD cores
and aluminum salen complexes to initiate polymer and copolymer synthesis.12a
Polymer
[M]:[C]:[I]
% Conv. a
Mn,thb
Mn,GPCc
Mn,NMRd
Đc
Đarme
PHB
70:1:1
98
7211
15310
12130
1.19
2.60
PHB
140:1:1
96
12902
29440
22980
1.09
1.69
PHB
210:1:1
94
19410
28710
24900
1.01
1.07
PLGA
70:1:1
68
7096
10760
8320
1.17
2.55
PLGA
140:1:1
81
15827
18750
21060
1.31
3.51
PLGA
210:1:1
41
19410
18360
20270
1.30
3.44
PLGA
280:1:1
45
16987
21760
26490
1.18
2.43
PLGA
350:1:1
69
31411
24530
29000
1.16
2.25
Polymerizations were conducted neat with 0.02 mmol of tBu[salen]AlMe at 120 °C for 3 hours, using heptakis(2,6-di-O-methyl)-β-CD as an initiator. a % Conv.
determined by gravimetric analysis. b Mn,th = [M](per arm) × 7(Initiating Groups) × MW(Monomer) × (% conv.) + MW(core) c Determined by GPC using refractive
index coupled with light scattering detectors. d Mn from 1H NMR spectroscopy. e Calculated from the Szymanski method.
Table 4. Polymerization data for CD poly(3-hydroxybutyrate) and poly(lactic-co-glycolic acid) polymer stars based on tBu[salen]AlMe (2).
3.2
Core-First Polymer Star Fluorescence Studies
An important property of these new star polymers, and in fact all aliphatic polyesters as a class of green materials, is their degradation.
While hydrolytic degradation rates are slow under environmental conditions, enzymatic degradation has made these materials interesting
potential candidates for biocompatible applications.3,4 We explored the use of 7-methoxycoumarin as a fluorescent sensor that would
illuminate in an aqueous environment upon release. Using our poly(lactic acid) star with ca. 30 monomer units per arm we explored
degradation conditions and loading regimes to develop a fluorescence method to measure analyte release from the star.
Our fluorescent probe, 7-MC, is highly fluorescent in polar media but exhibits relatively rare fluorescence suppression upon entering the
hydrophobic environment of the host. We expected that, with an intact star, loading after star formation would embed 7-MC within the
arms of the star, not necessarily accessing the sterically crowded CD cavity. In order to embed 7-MC into the polymer network, the star
was dissolved in CH2Cl2 and 100 equivalents of 7-MC was added to the solution. The solution was then refluxed for 24 hours and the
polymer was precipitated and washed with MeOH to remove surface fluorophore, confirmed by a constant ratio of 7-MC to CD in 1H
NMR spectra. The polymer was dried in vacuo prior to hydrolysis. PLA pellets were formed using a manual press and were degraded in
vials containing either a 1:1 mixture of nano-pure H2O and MeOH or nano-pure H2O. Initial degradation experiments suggested that acidor base-catalyzed hydrolytic degradation was incompatible with our fluorescence measurements: the 7-MC fluorescence is quenched in
even moderately alkaline or acidic environments. Further hydrolytic degradation studies were therefore conducted at neutral pH.
In order to measure the fluorescence, standard stock solutions of 7-MC were prepared using H2O/MeOH and H2O of 1.5×10-5 and 3.0×105 M solutions respectively, to provide a fixed 7-MC concentration reference for the fluorescence intensity. Hydrolytic degradations
proceeded at a slow rate without catalysis, as observed by visual inspection of the pellets and confirmed by gravimetric analysis of the
polymer samples. However, aliquots of the solution in which the pellets were degrading were tested over 135 days in order to observe 7MC release and suggested that release of the 7-MC plateaued after only 20 days. Plotting F/F0 versus time captures this release, where
F/F0 is the ratio of the integrated fluorescence of the pellet solution to that of the standard stock solution of 7-MC, as determined by
Equation 1. Individual plots of PLA-star-CD release in H2O and H2O/MeOH mixtures can be found in the supporting information
(Figures S4 and S5) whilst a normalized plot (such that the plateau of the plot is scaled to 1) showing the two release profiles for
comparison is shown in Figure 3.
F/Fo = IFd, HG - IFd, Blank
IFd, G - IFd, Blank
1.
Figure 3. Normalized F/F0 plot for 7-MC release in H2O/MeOH (blue) and pure H2O (red).
In both solvent systems no initial fluorescence is observed. However, a burst release after 24 hours is noted by the significant increase in
fluorescence intensity due to leaching of the 7-MC into solution as the alcohol terminated polymer chains gain mobility in solution. A
faster release is noted in the H2O/MeOH solvent system due to increased alkalinity provided by MeOH. The fluorescence plateaued after
20 days and 80 days respectively for the mixed and aqueous solvent systems, although visible degradation of the pellets did not occur.
This suggests that release was diffusion controlled out of the polymer arms, with the MeOH mobilizing chain ends and improving
diffusion-controlled release. GPC and gravimetric analysis of the polymer pellets following hydrolysis experiments confirmed that
polymer arms maintained >90% of their original weight. Without the ability to use acid or base catalysis, understanding the complete
release profile was challenging. We turned to enzymatic biodegradation to improve our methodology, exploiting proteinase K for this
purpose: secreted by the fungus Trirachium album, it has been reported to enzymatically degrade PLA.39 The release profile of 7-MC
from PLA stars using proteinase K, compared to purely hydrolytic degradation (i.e. in the absence of enzyme), is shown in Figure 4. In
this figure, the values of F/F0 have been normalized to 1 at t = 1 hour, to remove the differences in the initial 7-MC concentration (burst
release), and thus allow for direct comparison (a complementary unscaled plot for the enzymatic process over an 80- hour time frame is
shown in the Supporting Information, Figure S6). This enzymatic degradation gave a much more controlled release profile over an 80
hour time frame. The release profile suggests that the initial diffusion controlled burst release is complemented by a continuous release
promoted by degradation of the polymer chains. This degradation is confirmed by visual inspection of the pellets and GPC and
gravimetric analyses (Mn,original = 53,340; Mn,final = 22,840).
Figure 4. Normalized F/F0 plot for 7-MC release in with proteinase K (green) and without (purple).
The PHB stars were not analyzed for controlled release using this technique: their thermoplastic properties meant embedding and pellet
pressing proved unfeasible. PLGA degradation in the presence of proteinase K was performed (Supporting information, Figure S7). As
expected, the PLGA pellets degraded faster than the PLA pellets due to the higher accessibility of ester linkages in the PLGA polymer
chains.40 Higher rates of 7-MC release were observed relative to the other degradations, with F/F0 increasing from 0.06 to 0.90 over the
period of 8 hours. In general, these fluorescence studies confirmed that these systems could be used for controlled release, trapping
fluorescent molecules or bioactives within the intertwined polymer arms and promoting release slowly through simple diffusion control
or at faster rates through degradation of the polymer arms.
The core-first technique does provide a method to synthesize a library of diverse polymer stars with varying molecular weights and
excellent control. Utilising aluminum catalyzed ROP reactions significantly improved reaction control and star dispersity and increased
monomer scope. However, molecular weights did not always correlate with those expected from monomer conversions. To contrast these
polymers we also investigated the preparation of arm-first stars in the hope of achieving more predictable molecular weights. Our strategy
was to employ acetylene terminated polymer arms “clicked” onto an azide-modified cyclodextrin using copper-catalyzed azide-alkyne
cycloaddition (CuAAC).
3.3
Synthesis of Acetylene-terminated polymers
Schubert et al. first reported the synthesis of acetylene terminated poly(ε-caprolactone)s utilizing the unprotected 5-hexyn-1-ol as an
initiator and Sn(Oct)2 as a catalyst achieving moderate levels of control over molecular weight and low Đs.41 Building from this report we
utilized 5-hexyn-1-ol as an initiator for the ROP of rac-lactide, L-lactide and β-butyrolactone, accessing a library of linear polymers for
future click chemistry reactions.
Scheme 2. Synthesis of acetylene terminated linear polymers.
Synthesis of acetylene-terminated rac-lactide polymers using Sn(Oct)2, 1, as a catalyst gave broad dispersity, poorly controlled polymer
products at high temperatures (120 °C, Supporting Information, Table S1) due to transesterification and excellent control over molecular
weight and dispersity in toluene at 70 °C (Table 5). Five polymer arms were prepared with 10-50 monomer units per arm. Similar levels
of control were obtained using L-lactide to prepare semi-crystalline poly(L-lactic acid) polymer arms of controlled length.
[M] : [C] : [I]
Lactide
% Conv.a
Mn,thb
Mn, GPCc
Đc
10 : 0.5 : 1
rac
>99
833
790
1.11
20 : 0.5 : 1
rac
>99
2635
1350
1.10
30 : 0.5 : 1
rac
99
2433
1650
1.10
40 : 0.5 : 1
rac
98
5748
3050
1.13
50 : 0.5 : 1
rac
98
7161
3610
1.14
10 : 0.5 : 1
L
>99
1539
1400
1.18
20 : 0.5 : 1
L
>99
2391
2390
1.15
30 : 0.5 : 1
L
99
4280
6410
1.34
30 : 0.5 : 1
L
99
4422
2660
1.08
40 : 0.5 : 1
L
98
5748
3930
1.04
Polymerizations were conducted in 2 mL of toluene with 0.06 mmol of Sn(Oct)2 at 70 °C, using 5-hexyn-1-ol as an initiator for 3 h. a % Conv. determined by 1H
NMR spectroscopy. b Mn,th = [M] × MW(Lactide) × (% conv.) + MW(5-hexyn-1-ol)
detectors.
Table 5. Polymer data for acetylene terminated PLA from Sn(Oct)2 (1).
c
Determined by GPC using refractive index coupled with light scattering
The success achieved with rac-PLA and L-PLA prompted the exploration of other acetylene-terminated polymers. Initial trials with βBL
catalyzed by 1 yielded no polymer products at timeframes of 1-24 hours, confirming that Sn(Oct)2 is not an effective catalyst for the ROP
of βBL. Using 2, a capable catalyst for the living and immortal polymerization of βBL, 12a yielded acetylene-terminated PHB with
excellent levels of control over molecular weights and Đ values as low as 1.04 even at high conversion (Table 6). At higher monomer
loadings, molecular weights deviated from expected values, suggesting a limitation to arm length through this method and the potential
problem of impurities complicating PHB synthesis.
[M] : [C] : [I]
% Conv.a
Mn,thb
Mn, GPCc
Đc
10 : 1 : 1
>99
1539
1400
1.18
20 : 1 : 1
>99
2391
2390
1.15
30 : 1 : 1
>99
2680
1910
1.04
40 : 1 : 1
>99
5748
3930
1.04
50 : 1 : 1
>99
7088
3960
1.05
Polymerizations were conducted in neat with 0.12 mmol of tBu[salen]AlMe at 70 °C for 3 hours, using 5-hexyn-1-ol as an initiator. a % Conv. determined by 1H
NMR spectroscopy. b Mn,th = [M] × MW(βBL) × (% conv.) + MW(5-hexyn-1-ol)
c
Determined by GPC using refractive index coupled with light scattering detectors.
Table 6. Polymer data for acetylene terminated PHB from tBu[salen]AlMe (2).
3.4
Synthesis and Analysis of Arm-First Polymer Stars
The reaction of β-CD-(N3)7 with 7 eq. of the acetylene terminated rac-PLA in the presence of CuSO4 and C6H7NaO6 afforded the desired
polymer stars (Scheme 2). Reactions were performed in DMF as the β-CD-(N3)7 was found to be highly insoluble in other organic
solvents. After 24 hours the polymer stars were precipitated by addition of H 2O to the reaction mixture. Polymers were dried in a vacuum
oven for 24 hours at 75 °C to remove any traces of DMF/H2O. Characterization details are provided in Table 7 along with similarly
prepared polymer stars of L-lactide and β-butyrolactone.
Scheme 3. Synthesis of arm-first polymer stars.
Monomer
[M]/armf
% Yielda
Mn,thb
Mn, GPCc
Mn, NMRd
Đc
Đarme
rac-LA
10
40
4749
12010
14170
1.05
1.42
rac-LA
20
38
10774
16540
16760
1.04
1.35
rac-LA
30
57
23472
24680
26990
1.03
1.23
rac-LA
40
21
22646
23430
29230
1.04
1.31
rac-LA
50
48
26573
31700
35310
1.05
1.38
L-LA
10
52
11117
24470
8830
1.05
1.39
L-LA
20
74
18047
23150
15870
1.02
1.15
L-LA
30
75
19909
23310
21470
1.01
1.07
L-LA
40
81
28838
29800
26290
1.04
1.31
L-LA
50
80
29121
31650
26810
1.04
1.30
βBL
10
46
5294
15260
13200
1.04
1.34
βBL
20
63
9160
14260
17710
1.02
1.17
βBL
30
86
14659
21710
19800
1.02
1.16
βBL
40
73
20763
26520
30740
1.15
1.16
βBL
50
69
23521
24470
31390
1.01
1.07
Polymerizations were conducted in 2 mL of DMF with 0.001 mmol of CuSO4 at room temperature for 24 hours. a % Conv. determined by GPC. b Mn,th = MW(arm)
× 7(Azide Groups) + MW(core) c Determined by GPC using refractive index coupled with light scattering detectors. d Determined by 1H NMR spectroscopy. e
Calculated from the Szymanski method. f [M]/arm is the number of monomer units per arm.
Table 7. Polymer data for rac-PLA arm-first stars.
GPC analysis of these polymer stars showed excellent Đ values and improved control over corresponding core-first star syntheses. As no
ROP is performed, star synthesis depends upon the efficiency of the click reaction, thus a product yield and not a monomer conversion.
Good correlation between the molecular weights obtained from GPC and 1H NMR spectroscopy and theoretical molecular weights was
observed. No longer plagued by the potential for transesterification reactions, the arm-first method helps access uniform polymer stars, a
potentially important component to green polymers in biomedical applications. Unfortunately, yields were variable and, in fact, deviated
from batch to batch (three runs were performed, with the average yields reported in Table 7). As no stars with fewer than seven arms
were observed by NMR or GPC it is expected that the low solubility of the β-CD-(N3)7 leads to cores not initiating despite the efficiency
of the CuAAC reaction. However, from a synthetic perspective, the arm-first method offers a synthetic route to aliphatic polyester stars
with a cyclodextrin core.
The arm first strategy also allowed for a different inclusion strategy for our fluorescence studies. Inclusion of 7-MC into the arm-first
stars was achieved through a one-pot synthesis, using excess 7-MC as a component in the CuAAC reaction. After the 24 hour reaction
time, the product was precipitated and washed using MeOH and dried in vacuo. Analysis by 1H NMR spectroscopy confirmed the
presence of 7-MC in the polymeric network. Integration of 7-MC versus core peaks showed that loading of the 7-MC by this strategy did
not alter the capacity of the star. Enzymatic degradation of a rac-PLA star mediated by proteinase K is shown in Figure 5. A burst release
followed by controlled release of the fluorescent probe was observed. Unfortunately, the mimicking of the trends observed in core-first
star release suggest that there is no selective segregation of the 7-MC in the cyclodextrin core. This is likely due to the poor binding
constant of 7-MC in the required DMF solvent. Second generation water-soluble, azide-functionalized cores are being designed by our
team to circumvent this and potentially permit the desired two-stage release.
Figure 5. Release of 7-MC from arm-first PLA star pellet in the presence of proteinase K.
An important final note concerns a comparison of the two synthetic methodologies. In evaluating the greenness of core-first versus armfirst strategies it is important to note that if the application permits a small decrease in control, sustainability is significantly improved and
waste reduced through core-first methods: the polymerization can be conducted neat, in the absence of solvent, albeit at a higher
temperature. Core-first stars can be prepared in a one-pot reaction and, due to the inherent increased solubility of the starting core, avoid
toxic DMF use. Beyond these reaction conditions, isolated star yields are higher for core-first stars. Whilst all stars are hydrolytically and
enzymatically degradable, important for both their reduced environmental impact and candidacy as potential controlled release drug
delivery platforms, it is important to recognize that the polymerization process should also be considered when evaluating green choices
for potential biomedical applications.
4.
Conclusions
PLA-based cyclodextrin-cored polymer stars were successfully prepared and characterized using two distinct approaches: core-first and
arm-first. Improved control for the core-first approach was achieved using lactide with 2,6-di-O-methyl-β-CD and a tBu[salen]AlMe
catalyst. The same catalyst also facilitated the preparation of stars with either PHB or PLGA arms. Loading of 7-MC into these core-first
polymer stars allowed for controlled release fluorescence studies. In water, these polymer stars degraded slowly and 7-MC release was
diffusion controlled. Degradation rates could be significantly enhanced with the addition of a proteinase K enzyme, with controlled
release observed over the course of 2 days. A complementary arm-first synthetic strategy was developed using synthesized -CD-(N3)7 in
a click chemistry approach. These arm-first polymer stars showed similar degradation properties to the core-first versions. The arm-first
strategy resulted in lower dispersity, but lower yields, as compared to the core-first strategy, with the strategy of choice depending on the
desired properties, end application and desired sustainability. These results provide a firm foundation for the exploration of new
biomedical materials based on cyclodextrin cores, with potential applications in the field of controlled drug release.
Acknowledgments
We thank the University of Edinburgh, the University of Prince Edward Island and the Natural Sciences and Engineering Research
Council of Canada for financial support, the Canadian Foundation for Innovation and the Atlantic Canada Opportunities Agency for
infrastructure support and Innovation PEI for the support of G.K.T. with a Graduate Student Fellowship.
References
1) Cameron, D. J. A.; Shaver, M. P. Aliphatic polyester polymer stars: synthesis, properties and applications in biomedicine and
nanotechnology. Chem. Soc. Rev. 2011, 40, 1761-1776.
2) Colson, Y.L.; Grinstaff, M.W. Biologically responsive polymeric nanoparticles for drug delivery. Adv. Mater. 2012, 24, 3878-3886.
3) (a) Kumari, A.; Yadav, S.K.; Yadav, S.C. Biodegradable polymeric nanoparticles based drug delivery systems. Colloids Surf., B. 2010,
75, 1-18; (b) Siepmann, J.; Siepmann, F. Mathematical modeling of drug delivery. Int. J. Pharm. 2008, 364, 328-343.
4) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M. Polymeric systems for controlled drug release. Chem. Rev. 1999,
99, 3181-3198.
5) Lim, L.-T.; Auras, R.; Rubino, M. Processing technologies for poly(lactic acid). Prog. Polym. Sci. 2008, 33, 820-852.
6) Rieth, L.R.; Moore, D.R.; Lobkovsky, E.B.; Coates, G.W. Single-site β-diiminate zinc catalysts for the ring-opening polymerization of
β-butyrolactone and β-valerolactone to poly(3-hydroxyalkanoates). J. Am. Chem. Soc, 2002, 124, 15239-15248.
7) Danheir, F.; Ansorena, E.; Silva, J.M.; Coco, R.; Le Breton, A.; Préat, V. PLGA-based nanoparticles: An overview of biomedical
applications. J. Control. Release 2012, 161, 505-522.
8) Hu, W.; Huang, Z-M. Biocompatibility of braided poly(L-lactic acid) nanofiber wires applied as tissue sutures. Polym. Int. 2010, 59,
92-99.
9) Kumbar, S.G.; Nukavarapu, S.P.; James, R.; Nair, L.S.; Laurencin, C.T. Electrospun poly(lactic acid-co-glycolic acid) scaffolds for
skin tissue engineering. Biomaterials 2008, 29, 4100-4107.
10) Dev, A.; Binulal, N.S.; Anitha, A.; Vair, S.V.; Furuike, T.; Jayakumar, R. Preparation of poly(lactic acid)/chitosan nanoparticles for
anti-HIV drug delivery applications. Carbohydr. Polym. 2010, 80, 833-838.
11) Ajellal, N.; Bouyahyi, M.; Amgoune, A.; Thomas, C.M.; Bondon, A.; Pillin, I.; Grohens, Y.; Carpentier, J-F. Syndiotactic-enriched
poly(3-hydroxybutyrate) via stereoselective ring-opening polymerization of racemic β-butyrolactone with discrete yttrium catalysts.
Macromolecules 2009, 42, 987.
12) (a) Cross, E. D.; Allan, L. E. N.; Decken, A.; Shaver, M. P. Aluminum salen and salan complexes in the ring-opening polymerization
of cyclic esters: Controlled immortal and copolymerization of rac-beta-butyrolactone and rac-lactone. J. Polym. Sci., Part A: Polym.
Chem. 2013, 51, 1137-1146; (b) Barud, H. S.; Souza, J.S.; Santos, D.B.; Crespi, M.S.; Ribeiro, C.A.; Messaddeq, Y.; Ribeiro, S.J.L.
Bacterial cellulose/poly(3-hydroxybutyrate) composite membranes. Carbohydr. Polym. 2011, 83, 1279.
13) Shaver, M.P.; Cameron, D.J.A. Tacticity control in the synthesis of poly(lactic acid) polymer stars with dipentaerythritol cores.
Biomacromolecules 2010, 11, 3673-3679.
14) Perry, M.R.; Shaver, M.P. Flexible and rigid core molecules in the synthesis of poly(lactic acid) star polymers. Can. J. Chem. 2011,
89, 499-505.
15) Cameron, D. J. A.; Shaver, M. P. Control of thermal properties and hydrolytic degradation in poly(lactic acid) polymer stars through
control of isospecificity of polymer arms. J. Polym. Sci., Part A: Polym. Chem. 2012, 50, 1477-1484.
16) Keefe, G. E.; Twibanire, J. K.; Grindley, T. B.; Shaver, M. P. Poly(lactic acid) polymer stars built from early generation dendritic
polyols. Can. J. Chem. 2013, 91, 392-397.
17) Zhang, Q.; Li, G-Z.; Becer, R.; Haddleton, D. M. Cyclodextrin-centred star polymers synthesized via a combination of thiol-ene click
and ring opening polymerization. Chem. Commun. 2012, 48, 8063-8065.
18) Astray, G.; Gonzalez-Berreiro, C.; Mejuto, J.C.; Rial-Otero, R.; Simal-Gándara, J. A review on the use of cyclodextrins in foods.
Food Hydrocolloids 2009, 23, 1631-1640.
19) Jansook, P.; Kurkov, S. V.; Loftsson, T. Cyclodextrins as solubilizers: formation of complex aggregates. J. Pharm. Sci. 2010, 99,
719-729.
20) Cruz, J.R.; Becker, B.A.; Morris, K.F.; Larive, C.K. NMR characterization of the host–guest inclusion complex between βcyclodextrin and doxepin. Magn. Reson. Chem. 2008, 46, 838-845.
21) (a) Hoare, T. R.; Kohane, D.S. Hydrogels in drug delivery: Progress and challenges. Polymer 2008, 49, 1993-2007; (b) Shen, J.; Du,
G.; Zhang, H. Sun, H. A convenient preparation of 6-oligo(lactic acid)cyclomaltohepaose as kinetically degradable derivative for
controlled release of amoxicillan. Carbohydr. Res. 2008, 343, 2517-2522.
22) (a) Gao, H.; Wang, Y.N.; Fan, Y.G. Ma, J.B. Synthesis of biodegradable tadpole-shaped polymer via the coupling reaction of
polylactide onto momo(6-(2-aminoethyl)amino-6-deoxy)--cyclodextrin and its properties as the new carrier of protein delivery
system. J. Contr. Release 2005, 107, 158-173; (b) Gao, H.; Wang, Y.N.; Fan, Y.G. Ma, J.B. Conjugates of poly(DL-lactide-coglycolide) on amino cyclodextrins and their nanoparticles as protein delivery system. J. Biomed. Mater. Res. 2006, 80A, 111-122.
23) Adeli, M.; Zarnegar, Z.; Kabiri, R. Amphiphic star copolymers containing cyclodextrin core and their application as nanocarrier. Eur.
Polym. J. 2008, 44, 1921-1930.
24) Cai, T.; Yang, W.J.; Zhang, Z.; Zhu, X.; Neoh, K-G.; Kang, E-T. Preparation of stimuli-responsive hydrogel networks with threaded
-cyclodextrin end-capped chains via combination of controlled radical polymerization and click chemistry. Soft Matter 2012, 8,
5612-5620.
25) Hashidzume, A.; Harada, A. Recognition of polymer side chains by cyclodextrins. Polym. Chem. 2011, 2, 2146-2154.
26) Schmidt, B.V.K.J.; Rudolph, T.; Hetzer, T.; Ritter, H.; Schacher, F.H.; Barner-Kowollik, C. Supramolecular three-armed star
polymers via cyclodextrin host–guest self-assembly. Polym. Chem. 2012, 3, 3139-3145.
27) Selected additional recent and pioneering references include: (a) Huin, C.; Eskandani, Z.; Badi, N.; Farcas, A.; Bennevault-Celton,
V.; Guegan, P. Anionic ring-opening polymerization of ethylene oxide in DMF with cyclodextrin derivatives as new initiators.
Carbohydr. Polym. 2013, 94, 323-331; (b) Durmaz, Y. Y.; Lin, Y-L.; El Sayed, M. E. H. Development of degradable, pH-sensitive
star vectors for enhancing the cytoplasmic delivery of nucleic acids. Adv. Funct. Mat. 2013, 23, 3885-3895; (c) Huan, X.; Wang, D.;
Dong, R.; Tu, C.; Zhu, B.; Yan, D.; Zhu, X. Supramolecular ABC miktoarm star terpolymer based on host−guest inclusion
complexation. Macromolecules 2012, 45, 5941-5947; (d) Li, J.; Guo, Z.; Xin, J.; Zhao, G.; Xiao, H. 21-Arm star polymers with
different cationic groups based on cyclodextrin core for DNA delivery. Carbohydr. Polym. 2010, 79, 277-283; (e) Ohno, K.; Wong,
B.; Haddleton, D. M. Synthesis of well-defined cyclodextrin-core star polymers. J. Polym. Sci., Part A: Polym.
Chem. 2001, 39, 2206-2214; (f) Feng, J.-X.; Su, S.-P.; Zhu. J. An intumescent flame retardant system using-cyclodextrin as a
carbon source in polylactic acid (PLA). Polym. Adv. Technol. 2011, 22, 1115-1122.
28) (a) Miao, Y.; Zinck, P. Ring-opening polymerization of cyclic esters initiated by cyclodextrins. Polym. Chem. 2012, 3, 1119; (b)
Normand, M.; Kirillov, E.; Carpentier, J.-F.; Guillaume, S.F. Cyclodextrin-centered polyesters: controlled ring-opening
polymerization of cyclic esters from b-cyclodextrin –diol. Macromolecules 2012, 45, 1122-1130.
29) Wagner, B.D.; Fitzpatrick, S.J.; McManus, G.J. Fluorescence suppression of 7-methoxycoumarin upon inclusion into cyclodextrins. J.
Incl. Phenom. Macro. 2003, 47, 187-192.
30) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; Pugh, R. I.; White, A. J. P. Study of ligand substituent effects on the rate and
stereoselectivity of lactide polymerization using aliminum sale-type initiators. Proc. Nat. Acad. Sci. 2006, 103, 15343-15348.
31) Ashton, P. R.; Königer, R.; Stoddart, J. F.; Alker, D.; Harding, V. D. Amino Acid Derivatives of -Cyclodextrin. J. Org. Chem. 1996,
61, 903-908.
32) Yu, I.; Acosta-Ramirez, A.; Mehrkhodavandi, P. Mechanism of living lactide polymerization by dinuclear indium catalysts and its
impact on isoselectivity. J. Am. Chem. Soc. 2012, 134, 12758-12773.
33) Xu, C.; Yu, I.; Mehrkhodavandi, P. Highly controlled immortal polymerization of β-butyrolactone by a dinuclear indium catalyst.
Chem. Commun. 2012, 48, 6806-6608.
34) Körber, M. PLGA Erosion: Solubility- or Diffusion-Controlled? Pharm. Res. 2010, 27, 2414-2420.
35) Corbin, P. S.; Webb, M. P.; McAlvin, J. E.; Fraser, C. L. Biocompatible polyester macroligands: New subunits for the assembly of
star-shaped polymers with luminescent and cleavable metal cores. Biomacromolecules 2001, 2, 223-232.
36) Szymanski, R. On the reshuffling of polymer segments in star polymer systems. Macromolecules 2002, 35, 8239-8242.
37) Qian, F.; Huang, J.; Hussain, M. A. Drug–polymer solubility and miscibility: stability consideration and practical challenges in
amorphous solid dispersion development. J. Pharm. Sci. 2010, 99, 2941-2947.
38) Zhong, Z.; Dijkstra, P. J.; Feijen, J. [(salen)Al]-Mediated, controlled and stereoselective ring-opening polymerization of lactide in
solution and without solvent: Synthesis of highly isotactic polylactide stereocopolymers from racemic d,l-lactide. Angew. Chem.
2002, 114, 4692-4695.
39) Lenglet, S.; Li, S.; Vert, M. Lipase-catalysed degradation of copolymers prepared from -caprolactone and DL-lactide. Polym.
Degrad. Stab. 2009, 94, 688-692.
40) Park, T. G. Degradation of poly(lactic-co-glycolic acid) microspheres: effect of copolymer composition. Biomaterials 1995, 16, 1123.
41) Hoogenboom, R.; Moore, B. C.; Schubert, U. S. Synthesis of star-shaped poly(ε-caprolactone) via ‘click’ chemistry and
‘supramolecular click’ chemistry. Chem. Commun. 2006, 4010-4012.