Phenylboronate-diol crosslinked glycopolymeric nanocarriers for insulin delivery at physiological pH

Article abstract of DOI:10.1039/c3sm52485j

Research into polymers with glucose-sensitivity in physiological conditions has expanded recently due to their therapeutic potential in diabetes. Herein, to explore the glucose-responsive properties of a new polymer under physiological conditions, we synthesized an amphiphilic block glycopolymer based on phenylboronic acid and a carbohydrate, which was named poly(d- gluconamidoethyl methacrylate-block-3-acrylamidophenylboronic acid) (p(AAPBA-b-GAMA)). Based on the cross-linking between the diol groups of the carbohydrates and phenylboronic acid, the glycopolymers self-assembled to form nanoparticles (NPs). The glucose-sensitivity was revealed by the swelling behavior of the NPs at different glucose concentrations and was found to be dependent on the glucose level. The morphology of the NPs revealed by transmission electron microscopy showed that the NPs were spherical in shape with good dispersity. The cell viability of the NPs investigated by MTT assay was more than 90%, indicating that the glycopolymers had good cytocompatibility. Insulin could be loaded onto the glycopolymer NPs with high efficiency (up to 10%), and insulin release increased with enhancement of the glucose level in the medium. Such a glucose-responsive glycopolymer is an excellent candidate that holds great potential in the treatment of diabetes.

Full text of DOI:10.1039/c3sm52485j

Soft Matter
View Article Online
View Journal | View Issue
PAPER
Phenylboronate-diol crosslinked glycopolymeric
nanocarriers for insulin delivery at
physiological pH†
Cite this: Soft Matter, 2014, 10, 911
a
b
a
Qianqian Guo, Zhongming Wu, Xinge Zhang, Lei Suna and Chaoxing Lia
*
Research into polymers with glucose-sensitivity in physiological conditions has expanded recently due to
their therapeutic potential in diabetes. Herein, to explore the glucose-responsive properties of a new
polymer under physiological conditions, we synthesized an amphiphilic block glycopolymer based on
phenylboronic acid and a carbohydrate, which was named poly(D-gluconamidoethyl methacrylate-
block-3-acrylamidophenylboronic acid) (p(AAPBA-b-GAMA)). Based on the cross-linking between the
diol groups of the carbohydrates and phenylboronic acid, the glycopolymers self-assembled to form
nanoparticles (NPs). The glucose-sensitivity was revealed by the swelling behavior of the NPs at different
glucose concentrations and was found to be dependent on the glucose level. The morphology of the
NPs revealed by transmission electron microscopy showed that the NPs were spherical in shape with
good dispersity. The cell viability of the NPs investigated by MTT assay was more than 90%, indicating
that the glycopolymers had good cytocompatibility. Insulin could be loaded onto the glycopolymer NPs
with high efficiency (up to 10%), and insulin release increased with enhancement of the glucose level in
the medium. Such a glucose-responsive glycopolymer is an excellent candidate that holds great
potential in the treatment of diabetes.
Received 21st September 2013
Accepted 14th November 2013
DOI: 10.1039/c3sm52485j
www.rsc.org/softmatter
expedient strategy is to develop glucose-responsive materials to
be utilized in a self-regulated insulin delivery system. Among
three kinds of typical glucose-responsive materials,7–9 phenyl-
boronic acid and its derivatives are versatile enough to be used
in different designs and are more stable than protein-based
systems (glucose oxidase and concanavalin A),10 have attracted
great interest.
1. Introduction
Insulin-dependent diabetes is a serious disease affecting many
people in the world. Currently, the treatment for diabetes
requires persistent monitoring of blood glucose levels and
multiple daily insulin injections.1 This provides some control of
the diabetic state, but this is far from the physiologic regulation
achieved by a normally functioning pancreas and results in
long-term complications, including cardiovascular disease,
nephropathy, cataracts, retinopathy, and skin ulcers.2,3 Great
efforts have been devoted to investigating a system that is
capable of consistently delivering accurate dose of insulin in
response to glucose level changes.4–6 For example, Gu et al.
developed an articial “closed-loop” system that could mimic
pancreas activity and release insulin in response to glucose level
changes. The self-regulating system could release insulin at
basal release rates under normoglycemic conditions and at
higher rates under hyperglycemic conditions.5,6 Thus, one
It is well known that phenylboronic acid and its derivatives
have a specic covalent binding interactions with diol units,
such as glucose,11 and an important property of phenylboronic
acid compounds in aqueous solution is that they are equilib-
rium between an uncharged and a charged form.12,13 The
charged phenylboronic acid is well known to form a stable
complex with glucose, which lays the foundation for the
glucose-responsiveness of phenylboronic acid-containing
materials. However, phenylboronic acid and its derivatives are
kinds of weak acids with a pKa of 8.2 ꢀ 8.6 (ref. 14 and 15) and
most of the phenylboronic acid-based materials reported
demonstrate maximum glucose-sensitivity at pH ¼ 9 ꢀ 10.
Thus, signicant efforts have been focused on decreasing the
pKa of phenylboronic acid-based glucose-responsive materials
and enhancing their sensitivity to glucose at physiological
pH.16–28 Based on this object, Wang et al. described glucose-
responsive micelles from the self-assembly of poly(ethylene
glycol)-b-poly(acrylic acid-co-acrylamidophenylbornic acid)
that could be applied to control insulin release at pH 7.4.16
Matsumoto et al. introduced novel phenylborate derivatives,
aKey Laboratory of Functional Polymer Materials of Ministry Education, Institute of
Polymer Chemistry, Nankai University, Tianjin 300071, China. E-mail: zhangxinge@
nankai.edu.cn; Fax: +86-22-23505598; Tel: +86-22-23501645
b2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics, Key
Laboratory of Hormones and Development (Ministry of Health), Metabolic Diseases
Hospital, Tianjin Medical University, Tianjin 300070, China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3sm52485j
This journal is © The Royal Society of Chemistry 2014
Soft Matter, 2014, 10, 911–920 | 911

View Article Online
Soft Matter
Paper
4-(1,6-dioxo-2,5-diaza-7-oxamyl)phenylboronic acid, with a pKa and thionyl chloride at 75 ꢁC for 8 h and freshly distilled before
of approximately 7.8 into a polymer as a glucose-sensing moiety, used, and methacryloyl chloride was obtained in a similar
and the polymer displayed glucose-sensitivity at physiological method at 90 ꢁC. 2-Dodecylsulfanylthiocarbonylsulfanyl-2-
pH.23 Wu et al. also prepared a multifunctional hybrid nanogel methylpropionic acid (DMP) was synthesized according to
with optical glucose sensing and controlled insulin release at the method as previously reported.38,39 D-Gluconolactone and
physiological pH.26 Recently, Kim et al. studied a mono- D-glucose were supplied by Tianjin no. 1 Chemical Reagent
saccharide-responsive system with the block copolymers self- Factory (Tianjin, China). 2,20-Azobisisobutyronitrile (AIBN) was
assembling into a variety of nanostructures that exhibited recrystallized twice from ethanol and dried in vacuum. Pure
monosaccharide-responsive disassembly in
a neutral-pH crystalline porcine insulin was purchased from Xuzhou Wan-
medium and encapsulated insulin could be released from the bang Biochemical Co. Ltd. (Xuzhou, China). 3-[4,5-Dime-
polymersomes only in the presence of sugar under physiologi- thylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) was
cally relevant pH conditions.28
obtained from Beyotime Institute of Biotechnology (Nantong,
However, a disadvantage of boronic acid-containing com- China). Other reagents were of analytical grade and used as
pounds is that they have some degree of cytotoxic activity.29,30 To received.
overcome the problem, it is necessary to introduce biocompat-
ible compounds. Our previous work demonstrated that such a
copolymer system composed of 3-acrylamidophenylboroniccid
2.2. Synthesis of glycopolymer based phenylboronic acid
(AAPBA) and carbohydrate was devoted to decreasing the pKa
2.2.1. Synthesis of AAPBA. AAPBA was prepared using the
and improving the biocompatibility of phenylboronic acid.31–37 modied method described by Lee et al.40 Briey, 3-amino-
The initial strategy involved the utilization of amino glucose to phenylboronic acid monohydrate (5.0 g, 32.2 mmol) and
lower the pKa of AAPBA via the free-radical polymerization, and sodium bicarbonate (5.0 g, 59.5 mmol) were dissolved in 45 mL
the results showed that this copolymer, as glucose-sensitive of a water and tetrahydrofuran (THF) mixed solution (v/v, 2 : 1),
matrix, had excellent potential for controlled insulin release.31 and cooled in an ice bath. Acryloyl chloride (5 mL, 62.5 mmol)
Additionally, to enhance the loading capability of drug, we was added drop-wise into the mixture solution with stirring over
prepared a kind of core–shell hybrid nanoparticles with pH- a period of 1 h. Then the reaction mixture was preformed for
gated and glucose-sensitivity that was utilized in controlling another 2 h at room temperature. The obtained solution was
insulin release.35 To further explore the in vivo application of the extracted with acetic ether and the organic phase was concen-
glucose-sensitive polymer, we designed and prepared injectable trated to obtain a dry crude product. Subsequently, the crude
ꢁ
and glucose-sensitive nanogels based on poly(N-isopropyla- product was recrystallized in 80 mL of hot water (90 C). The
crylamide), dextran and poly(3-acrylamidophenylboronic acid) resulting crystals were ltered, washed with water, dried under
and found that the nanogels decreased the blood glucose levels vacuum, and isolated in 52% yield.
in diabetic rats and maintained 51% of the baseline level for
2.2.2. Synthesis of poly(3-acrylamidophenylboronic acid)
almost 2 h.36 Based on above-mentioned results, the introduc- (pAAPBA). pAAPBA was synthesized by RAFT polymerization
tion of the carbohydrate into polymers containing phenyl- using a previously method reported.41 The polymerization was
boronic acid could not only improve the hydrophilicity and carried out by using AAPBA as the monomer, DMP as the RAFT
biocompatibility of the polymer but also decrease the pKa of agent, AIBN as the initiator, and N,N-dimethylformamide
phenylboronic acid-containing materials.
(DMF)–water (v/v, 19 : 1) as the solvent. The mixture solution
To further investigate the structure–property relationship, was sealed in a 50 mL reaction tube and purged with nitrogen
herein, we synthesized new glucose-responsive diblock for 30 min. Then the tube was transferred to a preheated oil
glycopolymers p(AAPBA-b-GAMA), comprised of AAPBA and bath at 70 ꢁC. The reaction was quenched aer 10 h by cooling
D-gluconamidoethyl methacrylate (GAMA) by reversible addi- in ice water for 5 min. The resultant polymer was isolated by
tion-fragmentation chain transfer (RAFT) polymerization. The removing the solvent, precipitating into diethyl ether, washing
AAPBA endowed the glycopolymers with glucose-sensitivity, with acetone and drying under vacuum. By changing the feed
while GAMA, a kind of synthetic carbohydrate monomer, was a molar ratio of AAPBA to DMP (100 : 1, 150 : 1 and 200 : 1), three
renewable source and introduced to reduce the toxicity and different homopolymers were obtained, marked as pAAPBA100,
decrease the pKa of AAPBA. Since the amphiphilic polymers can pAAPBA150 and pAAPBA200, respectively.
self-assemble into nanoparticles with a core–shell structure, the
formation of the nanoparticles was investigated, and the in vitro
2.2.3. Synthesis of GAMA
2.2.3.1. Synthesis of ethanolamine hydrochloride. The etha-
release of insulin in response to glucose changes was further nolamine hydrochloride was synthesized according to
evaluated.
a
previous report.42 A 37% (w/w) hydrochloric acid solution
(71 mL, 0.875 mol) was added during 2 h to a mixture of etha-
nolamine (53 mL, 0.875 mol) and water (100 mL) under stirring
at 90 ꢁC. Aer being incubated for 1 h at 105 ꢁC, the mixture was
2. Materials and methods
2.1. Materials
concentrated, precipitated in anhydrous ether and dried under
ꢁ
3-Aminophenylboronic acid monohydrate was purchased from vacuum at 40 C with a yield of 95%.
Nanjing Kangmanlin Chemical Industry Co. Ltd (Nanjing,
China). Acryloyl chloride was prepared by reuxing acrylic acid was synthesized using
2.2.3.2. Synthesis of 2-aminoethyl methacrylate (AMA). AMA
a
previously reported protocol.43
912 | Soft Matter, 2014, 10, 911–920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Soft Matter
Ethanolamine hydrochloride (65.0 g, 0.67 mol), methacryloyl the insulin-loaded NPs suspension was centrifuged at 12 000
chloride (100 mL, 0.96 mol), and hydroquinone (0.5 g) were rpm for 20 min. The amount of free insulin in the supernatant
added to a three-necked round-bottomed ask tted with a was monitored by the Bradford method using a UV spectrom-
condenser. The mixture was heated in an oil bath to 93 ꢁC under eter (Shimadzu UV-2550) at 595 nm.44 Insulin encapsulation
a nitrogen atmosphere for 1 h with vigorously stirring, and then efficiency (EE) and loading capacity (LC) were calculated using
further maintained at 70 ꢁC for 2 h. The mixture was cooled to the following equations:
40 ꢁC and 150 mL of THF was added. The solution was then
total insulin ꢂ free insulin
EE% ¼
LC% ¼
ꢃ 100
ꢃ 100
(1)
(2)
dropped slowly to cold 600 mL of n-pentane and the formed
creamy white precipitate was isolated by ltration, washed well
with n-pentane, and dried under vacuum. The crude product
was recrystallized using an ethyl acetate and 2-propanol mixture
(v/v, 7 : 3) and isolated in its hydrochloride salt form. The yield
was 60%.
total insulin
total insulin ꢂ free insulin
NPs weight
2.5. Characterization of NPs
2.2.3.3. Synthesis of GAMA. GAMA was prepared according
to the previous report.43 Briey, D-gluconolactone (10 g, 56.2
mmol) was dissolved in 100 mL of methanol. AMA (16.0 g, 96.6
mmol), hydroquinone (0.1 g) and triethylamine were added,
and the reaction solution was stirred at 20 ꢁC for 5 h. The
obtained mixture was then concentrated and precipitated into
either excess dichloromethane or 2-propanol. Aer ltered off,
washed, and dried under vacuum, the pure product was
obtained with the yield of 45%.
2.2.4. Synthesis of p(AAPBA-b-GAMA). RAFT copolymeriza-
tion was performed at [GAMA] : [pAAPBA macroRAFT] : [AIBN] ¼
50 : 1 : 1. By using GAMA as a monomer, the homopolymer
pAAPBA as a macroRAFT agent, AIBN as the initiator, and DMF–
water (v/v, 19 : 1) as the solvent, the reaction system was placed
in a preheated oil bath at 70 ꢁC for 12 h following nitrogen
purging for 30 min. The polymerization was quenched by
cooling in ice water for 5 min. The resulting polymer was iso-
lated by washing with water and acetone three times, respec-
tively, and drying under vacuum. By changing the ratio of
pAAPBA to GAMA, three distinct glycopolymers were prepared
and named as p(AAPBA2-b-GAMA1), p(AAPBA3-b-GAMA1) and
p(AAPBA4-b-GAMA1).
The hydrodynamic diameter (DH) and size distribution of the
NPs were characterized by dynamic light scattering (DLS) using
a Malvern Zetasizer Nano S apparatus equipped with a 4.0 mV
laser operating at l ¼ 636 nm. All measurements were per-
formed at a scattering angle of 90ꢁ. The morphological char-
acteristics of the NPs were determined by transmission electron
microscopy (TEM, Philips EM400ST). Samples were placed on
copper grids coated with Formvar lms for observing under
TEM.
2.6. Reversible glucose sensitivity of the NPs
p(AAPBA4-b-GAMA1) NPs were chosen to be the representative
nanoparticles to evaluate the reversible glucose sensitivity. The
blank NPs were treated with 3 mg mLꢂ1 glucose in PBS followed
with dialysis against water for 24 h. Subsequently, the NPs aer
dialysis were further exposed to 3 mg mLꢂ1 glucose. The size of
the NPs was measured by DLS.
2.7. Cell viability
Cell viability of p(AAPBA-b-GAMA) was evaluated by using NIH
3T3 cells. The cell line grew at 5% CO2, 95% O2 in Dulbecco's
modied Eagle's medium (DMEM, Gibco). The cells were
seeded into 96-well plates at 10 000 cells per well and were
incubated for 24 h. 100 mL of glycopolymer NPs suspension with
a range of concentrations from 25 to 500 mg mLꢂ1 was added
into the plates for further incubation aer removal of the cell
culture medium. Aer 24 h, the culture medium and 20 mL of
MTT were added to replace the mixture in each well and the
cells were incubated for a further 4 h. Aer removal of the
culture medium and MTT, 100 mL of DMSO was added to
each well to dissolve the formazan crystals under low-speed
shaking for 10 min, the absorbance was measured on a
microplate reader at 495 nm. The cell viability was determined
as a percentage of the negative control; untreated cells were
used as the negative control and their proliferation rate was
set as 100%.
2.3. Characterization of the glycopolymers
1H NMR spectra were recorded at room temperature using a
Varian Unity-plus 400 NMR spectrometer. FT-IR spectra were
recorded on a Fourier Transform Infrared Spectrometer (FTS-
6000, Bio-Rad Co.) using a KBr tablet containing the sample
powder at a resolution of 8 cmꢂ1
.
2.4. Preparation of p(AAPBA-b-GAMA) NPs
p(AAPBA-b-GAMA) NPs were prepared according to a previously
reported method.31 10 mg of p(AAPBA-b-GAMA) was dissolved in
2 mL of a mixed solvent containing dimethyl sulfoxide (DMSO)
and water (v/v, 1 : 1) and 20 mL of water was then added
dropwise to the solution under stirring. The resulting opales-
cent solution was transferred to a dialysis tube (MWCO 3500)
and dialyzed against water for 24 h. The organic solvent was
removed by replacing water every 3 h during the dialyzing
process. Aer freeze-drying the suspension, blank NPs were
obtained.
2.8. In vitro release study
In vitro release of different NPs was investigated by incubating 6
ꢁ
The insulin-loaded NPs were prepared with a similar method mg of insulin-loaded NPs at 37 ꢄ 0.5 C with 2 mL of pH 7.4
by using 500 mg mLꢂ1 of insulin solution. Aer dialysis for 24 h, phosphate buffer solution (PBS, 0.1 M) containing different
This journal is © The Royal Society of Chemistry 2014
Soft Matter, 2014, 10, 911–920 | 913

View Article Online
Soft Matter
Paper
glucose concentrations (0, 1 and 3 mg mLꢂ1) under shaking process, we found that the glycopolymer was difficult to
(100 rev per min). Insulin release from p(AAPBA4-b-GAMA1) NPs dissolve as the ratio of GAMA to AAPBA increased. Importantly,
was also achieved by altering the glucose concentration from GAMA was introduced to decrease the pKa of AAPBA (8.2) to 7.5.
0 to 3 mg mLꢂ1 in the same medium. The NPs were incubated in The pKa of the glycopolymer was determined using a pH meter
free glucose for the rst 9 h and then glucose was added to and the results are shown in the Supporting Information.†
obtain a nal concentration of 3 mg mLꢂ1. At a predetermined Fig. 1 showed the FT-IR spectra of pAAPBA and p(AAPBA-b-
time, samples were centrifuged at 12 000 rpm for 5 min and 100 GAMA). In comparison with pAAPBA, p(AAPBA-b-GAMA)
mL of supernatant was withdrawn and fresh buffer solution was maintained the typical absorption of pAAPBA described as
added. The amount of free insulin was monitored using the follows. There was a broad absorption band in the range of
Bradford method. Each sample was analyzed in triplicate and 3200 to 3600 cmꢂ1, which was due to the hydrogen bonds
the results were reported as mean ꢄ standard deviation (n ¼ 3). formed between hydroxyls of the carbohydrate moieties and N–
H stretching of amide groups. The peak band at 1666 cmꢂ1 was
assigned to C]O stretching (amide I band) while the amide II
2.9. Circular dichroism spectroscopy
band attributed to N–H bending vibration of a secondary amide
The stability of the released insulin was determined by analysis
resulting in an absorption band of 1539 cmꢂ1. The band in
of the conformation of released insulin using circular
the range of 1200 to 1000 cmꢂ1 corresponded to C–O stretching
dichroism (CD) and the resulting spectrum was compared to
of the alkoxy bond in the carbohydrate moieties. Besides, a
that of standard insulin. The standard insulin solution was
prepared in pH 7.4 PBS to a nal concentration of 0.1 mg mLꢂ1
.
CD measurements were performed on a Jasco J-715 CD spec-
tropolarimeter at 25 ꢁC with a cell length of 0.1 cm. For the far-
UV CD spectra, samples were scanned from 190 to 260 nm and
accumulated 10 times, at a resolution of 1.0 nm and scanning
speed of 100 nm minꢂ1. All CD data are expressed as mean
residue ellipticity.45
3. Results and discussion
3.1. Synthesis and characterization of p(AAPBA-b-GAMA)
glycopolymers
p(AAPBA-b-GAMA) glycopolymers were synthesized via
a
sequential RAFT polymerization method (Scheme 1). First,
pAAPBA homopolymers were prepared by RAFT of AAPBA using
DMP as initiator. Then, p(AAPBA-b-GAMA) diblock glycopol-
ymers were prepared by subsequent RAFT using pAAPBA as
macroRAFT. By changing the ratio of pAAPBA to GAMA, three
distinct glycopolymers were obtained. During the synthetic
Fig. 1 FT-IR spectra of p AAPBA and p(AAPBA-b-GAMA).
Scheme 1 Synthesis of the block amphiphilic glycopolymer p(AAPBA-b-GAMA).
914 | Soft Matter, 2014, 10, 911–920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Soft Matter
new typical absorption appeared at 1720 cmꢂ1 attributing to the
ester groups stretching vibration in GAMA units. Except for the
broad absorption band and C–O stretching band that reected
the presence of carbohydrate moieties, typical peaks of the
phenyl ring in AAPBA in the 1500 to 1300 cmꢂ1 region and
absorption peaks at 794 and 706 cmꢂ1 were observed. 1H NMR
spectra of AAPBA, pAAPBA, and p(AAPBA-b-GAMA) are shown in
Fig. 2. Compared with the spectrum of AAPBA, it was observed
in the spectrum of p(AAPBA-b-GAMA) that the resonance
signals of phenyl-ring protons at 7.0 ꢀ 8.0 ppm were retained,
but the resonance signals assigned to protons on the double
bond completely disappeared. And new resonance signals
also appeared in the range of 0.5 ꢀ 1.7 and 3.2 ꢀ 3.6 ppm,
which were assigned to the protons from the sugar units,
indicating that p(AAPBA-b-GAMA) copolymer was synthesized
successfully.
Scheme 2 Schematic representation of the p(AAPBA-b-GAMA) NPs
formation.
3.3. The glucose sensitivity
3.2. Characterization of p(AAPBA-b-GAMA) NPs
The introduction of AAPBA groups endowed the resultant NPs
with glucose sensitivity. It can be seen in Table 1 that all of the
NPs appeared larger in size aer treatment with 1 or 3 mg mLꢂ1
glucose. The results in Fig. 3B show that p(AAPBA4-b-GAMA1)
NPs increased to approximately 120 nm in size aer treatment
with 3 mg mLꢂ1 glucose. Obviously, aer glucose treatment, the
size of the NPs became larger than that without glucose
The p(AAPBA-b-GAMA) NPs were prepared by nanopreci-
pitation method coupled with dialysis. p(AAPBA-b-GAMA) self-
assembled to form NPs with a hydrophilic pGAMA corona and
a hydrophobic pAAPBA core (Scheme 2). TEM photographs of
p(AAPBA4-b-GAMA1) NPs in Fig. 3A clearly illuminate that the
NPs were spherical shape with good dispersion. The results
determined by DLS are shown in Fig. 4 and 5 and Table 1.
Generally, the size of blank p(AAPBA-b-GAMA) NPs was in
the range of 125 to 190 nm and exhibited a narrow size
distribution, with
a polydispersity index (PDI) between
0.090 and 0.186. The average DH for the NPs p(AAPBA2-b-
GAMA1), p(AAPBA3-b-GAMA1) and p(AAPBA4-b-GAMA1) were
129.2, 156.9 and 188.6 nm, respectively. It is shown in Fig. 5
that the size of the blank NPs had little change over 6 weeks,
indicating that p(AAPBA-b-GAMA) NPs were stable in aqueous
solution.
Fig. 2 1H NMR spectra of (a) AAPBA (DMSO-d6); (b) pAAPBA (Meth-
anol-d4); and (c) p(AAPBA4-b-GAMA1) (the mixed solvent of DMSO-d6 Fig. 3 TEM of p(AAPBA4-b-GAMA1) NPs before (A) and after (B)
and D2O).
treatment with 3 mg mLꢂ1 glucose.
This journal is © The Royal Society of Chemistry 2014
Soft Matter, 2014, 10, 911–920 | 915

View Article Online
Soft Matter
Paper
an increase in the DH for the three NPs. Noted that the decre-
ment of the I/I0 value was more in 3 mg mLꢂ1 glucose than that
in 1 mg mLꢂ1, which was attributed to more dissociation of NPs
in higher glucose concentration. That the DH increased with an
increase of the glucose concentration signicantly affected the
swelling of NPs. This is due to the boronic ester linkages
between the boronic acid-containing block copolymers with
diols being dynamic-covalent, and the dynamic-covalent nature
of the boronic ester allowed the linkage to recongure their
structure in the presence of other diols that competed for
bonding with the boronic acid competent. These results were
agreement with those reported by the Sumerlin group.46,47 Thus,
the linkages could be induced to dissociate via competitive
exchange reactions.
Fig. 7 showed the stimulation-responsiveness of p(AAPBA4-b-
GAMA1) NPs in the presence of glucose. The NPs in the glucose
free medium remained relatively stable during the determined
time span. The results in Fig. 7A show that the light scattering
intensity gradually reduced and subsequently levelled off with
increasing glucose concentration over 24 h, that is, aggregate
dissociation was highly dependent on the concentration of
glucose, with higher concentrations leading to faster and more
efficient dissociation. The glucose-responsiveness of the NPs
with a lower concentration is shown in Fig. 7B. Compared to
Fig. 7A with a higher NP concentration, aggregate dissociation
was rapid and efficient for all concentrations of glucose. Aer
the addition of glucose, I/I0 reduced with the increment of
glucose concentration due to the swelling and dissociation of
NPs inuenced by glucose concentration. Aer 16 h, the I/I0
curves reached a plateau over zero which was ascribed to the
cross-linking48,49 and the homeostasis of the dissociation and
association16,50 between boronic acid derivatives and glucose in
the solution.
Fig. 4 Size distribution of blank p(AAPBA-b-GAMA)NPs in pH 7.4 PBS.
The result of DLS monitoring in Fig. 8 showed the glyco-
polymers with reversible glucose sensitivity. The p(AAPBA4-b-
GAMA1) swelled from 191.7 to 217.7 nm aer the rst treatment
with 3 mg mLꢂ1 glucose for 4 h, and then contracted to 193.0
nm, which was near to its original size aer dialysis against
water for 24 h. When treated with 3 mg mLꢂ1 glucose for the
second time, the NPs swelled to 220.1 nm, which was close to
the rst swelling size.
Fig. 5 The stability of blank p(AAPBA-b-GAMA) NPs in pH 7.4 PBS.
treatment. The swelling phenomenon indicates that the glucose
exposure inuenced the morphology of the NPs and p(AAPBA-b-
GAMA) was responsive to glucose. Fig. 6 shows the values of I/I0
versus time, and the inset showed the values of DH versus
glucose concentrations in pH 7.4 PBS. It is clearly shown that
3.4. Insulin encapsulation and in vitro release
the values of I/I0 decreased gradually as glucose concentration Insulin-loaded p(AAPBA-b-GAMA) NPs were prepared by adding
increased from 0 to 3 mg mLꢂ1, which was in accordance with insulin solution into p(AAPBA-b-GAMA) solution. The self-
Table 1 DH and PDI of the NPs at various glucose concentrations measured by DLS
Glucose concentration (mg mLꢂ1
0
)
1
3
Samples
DH (nm)
PDI
DH (nm)
PDI
DH (nm)
PDI
p(AAPBA2-b-GAMA1)
p(AAPBA3-b-GAMA1)
p(AAPBA4-b-GAMA1)
129.2 ꢄ 1.243
156.9 ꢄ 1.546
188.6 ꢄ 2.041
0.186 ꢄ 0.015
0.162 ꢄ 0.017
0.090 ꢄ 0.024
148.3 ꢄ 1.942
168.7 ꢄ 1.408
201.7 ꢄ 1.967
0.167 ꢄ 0.009
0.154 ꢄ 0.024
0.095 ꢄ 0.027
160.9 ꢄ 1.212
181.6 ꢄ 2.076
220.8 ꢄ 2.328
0.124 ꢄ 0.052
0.111 ꢄ 0.024
0.097 ꢄ 0.041
916 | Soft Matter, 2014, 10, 911–920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Soft Matter
Fig. 7 Light scattering intensity of the glucose-responsiveness of two
concentrations of p(AAPAB4-b-GAMA1) NPs in pH 7.4 PBS with
different glucose concentrations in terms of light scattering intensity
as a function of time: (A) 1 mg mLꢂ1 and (B) 0.1 mg mLꢂ1
.
Fig. 6 Light scattering intensity of the glucose-response of 1 mg mLꢂ1
of p(AAPBA-b-GAMA) NPs: (A) p(AAPBA2-b-GAMA1); (B) p(AAPBA3-b-
GAMA1); (C) p(AAPBA4-b-GAMA1) and their light scattering intensity as
a function of time in pH 7.4 PBS.
assembly process was affected by some kinds of interactions
such as hydrophobic–hydrophilic interaction, electrostatic
interaction, hydrogen bonding and other intermolecular
Fig. 8 Reversible glucose sensitivity of p(AAPBA4-b-GAMA1) NPs.
Soft Matter, 2014, 10, 911–920 | 917
This journal is © The Royal Society of Chemistry 2014

View Article Online
Soft Matter
Paper
interactions. Insulin, a protein containing kinds of amino acids,
has hydrophilic and hydrophobic residues and can form stable
complexes due to these interactions.51–53 Thus, when insulin
was added into p(AAPBA-b-GAMA) solution and coupled with
the amphiphilic glycopolymer, insulin can be loaded to the
glycopolymer NPs with interactions of hydrogen bonding, van
der Waals force and the hydrophilic–hydrophobic interaction.
The results in Table 2 show that insulin was easily encapsulated
into p(AAPBA-b-GAMA) NPs, and the LC of insulin was about
10% for all NPs. What's more, the EE of insulin increased from
53.2% to 63.8% with an increase of pAAPBA moieties. The
possible reason for the high efficiency was the hydrophobic
interaction between pAAPBA and insulin.
Fig. 9A exhibited the release proles of insulin from drug-
loaded p(AAPBA-b-GAMA) NPs upon exposure to 3 mg mLꢂ1
glucose. The release displayed a burst release phase, indicating
that some insulin was adsorbed onto the particle surface during
the preparation of the NPs, and then rapidly diffused when the
NPs came into contact with the release medium. Moreover, the
burst insulin was higher, which was 20%, 16% and 12% for
p(AAPBA2-b-GAMA1), p(AAPBA3-b-GAMA1) and p(AAPBA4-b-
GAMA1), respectively. The release rate of insulin gradually
decreased and the amount of insulin released aer 48 h was
68%, 57% and 47% for p(AAPBA2-b-GAMA1), p(AAPBA3-b-
GAMA1) and p(AAPBA4-b-GAMA1), respectively. As AAPBA
content in p(AAPBA-b-GAMA) increased, more glucose-mediated
association and relatively less dissociation occurred leading to a
release rate of insulin from the NPs in the following order:
p(AAPBA2-b-GAMA1)
> p(AAPBA3-b-GAMA1) > p(AAPBA4-b-
GAMA1), which was in accordance with the reported result.16,50
Fig. 9B showed the release prole of insulin in response to
different glucose concentrations from drug-loaded p(AAPBA4-b-
GAMA1) NPs. There was approximately 18.7% of insulin
released in the glucose-free medium within 48 h, resulting from
the insulin adsorbed onto the particle surface. As the glucose
concentration increased from 0 to 3 mg mLꢂ1, the amount of
insulin release enhanced signicantly from 26% to 47%, which
was ascribed to a greater association for 3 mg mLꢂ1 glucose to
the AAPBA moiety which is consistent with the results displayed
in Fig. 6C. Further investigation on the release prole of
p(AAPBA4-b-GAMA1) NPs was carried out by altering the glucose
concentration from 0 to 3 mg mLꢂ1. Aer insulin release
reached a plateau within the rst 9 h in the glucose-free
medium, glucose was added to obtain a nal concentration of 3
mg mLꢂ1. The result in Fig. 9B shows that there was a signi-
cant increase in the amount of insulin release, which was
dependent on the glucose concentration, suggesting the
responsivity of the glycopolymers to glucose and the rapid
insulin release upon exposure to glucose. Generally, insulin is
Fig. 9 In vitro cumulative release of insulin in pH 7.4 PBS from (A)
p(AAPBA-b-GAMA) NPs at the glucose concentration of 3 mg mLꢂ1; (B)
p(AAPBA4-b-GAMA1) NPs at various glucose concentrations (0, 1 and
3 mg mLꢂ1) and medium only for the first 12 h and then 3 mg mLꢂ1
.
injected into the patients before dinner so that the insulin can
decrease the glucose level and avoid the occurrence of high
blood sugar aer dinner. In the present work, the time scales for
insulin delivery are compatible with the patient's needs. Since
the glycopolymer NPs deliver insulin in less than 30 min, they
facilitate a decrease in the postprandial blood sugar of the
patients.54 Importantly, the glucose-sensitive carrier could
control insulin release responding to the glucose levels in the
blood, which avoids glucose uctuations, and subsequently
decreases the occurrence and development of complications.
These self-regulated insulin delivery in vitro studies may have
potential application for the treatment of diabetes.
Table 2 EE and LC of the p(AAPBA-b-GAMA) NPs
3.5. Stability of the released insulin
Samples
EE (%)
LC (%)
CD spectroscopy was used to evaluate the conformational
changes of insulin.55,56 The far-UV-CD band at 208 nm primarily
arises from an a-helix structure, while the band at 223 nm is for
the b-structure. The ratio between both bands ([F]208/[F]223) can
p(AAPBA2-b-GAMA1)
p(AAPBA3-b-GAMA1)
p(AAPBA4-b-GAMA1)
54.20 ꢄ 0.68
57.19 ꢄ 1.17
63.76 ꢄ 1.53
9.90 ꢄ 3.94
13.38 ꢄ 1.88
11.65 ꢄ 4.47
918 | Soft Matter, 2014, 10, 911–920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Soft Matter
be used to generate a qualitative measure of the overall
conformational structure of insulin. In our study, the
[F]208/[F]223 ratio for standard insulin and released insulin was
1.29 and 1.14, respectively. As indicated by the CD spectra
(Fig. 10), no signicant conformational change was observed for
the insulin released from the NPs at pH 7.4 in comparison with
standard insulin. Furthermore, the spectral characteristics
indicate that the tertiary structure of released insulin has not
been distorted.
3.6. Cell viability
It is important to verify the cytotoxicity of synthetic glycopol-
ymers as phenylboronic acid and its derivatives have had cyto-
toxic activity in multiple cell lines. Fig. 11 shows the cytotoxicity
of p(AAPBA-b-GAMA) NPs on NIH3T3 cells determined by the
MTT method. The cells were exposed to the NPs suspensions
from 25 to 500 mg mLꢂ1 and the cells without pretreatment were
used as a negative control group. Aer incubating for 24 h, the
relative cell viability was higher than 90% for all of the glyco-
polymers, indicating that the presence of the glycopolymer did
not negatively impact cell viability. The cell viability almost
increased with decreasing AAPBA content in the glycopolymer.
Overall, these results suggest that the carbohydrate moieties in
the glycopolymers reduced the cytotoxicity and enhanced the
glucose-sensitivity of the glycopolymers bearing phenylboronic
acid, and that these NPs have a potential application for in vivo
insulin delivery.
Fig. 11 Cell viability as
a
function of p(AAPBA-b-GAMA) NPs
concentration by the MTT assay at 37 ꢁC after the incubation for 24 h.
Each value represents the mean ꢄ SD (n ¼ 5).
glucose sensitivity of the glycopolymer NPs was reversible.
Insulin was encapsulated into the NPs with high loading
capacity of approximately 10%. The release of insulin in
response to glucose levels was revealed by the insulin release
proles. The biocompatibility of the glycopolymers bearing
phenylboronic acid was improved by introduction of carbohy-
drate moieties, which gives the glycopolymers potential to be
used in biomedical elds. In conclusion, the data indicate that
the glycopolymers have the potential for use in a self-regulated
insulin delivery system in the future.
4. Conclusions
In this study, a new amphiphilic block glycopolymer based on
phenylboronic acid and carbohydrate was synthesized. The
glycopolymers could self-assemble to form stable spherical NPs
with phenylboronate-diol crosslinked bond and exhibited
glucose-sensitive behavior at physiological pH. The particle size
increased in correlation with the glucose concentration, and the
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (Grant no. 51173085 and 81000683) and
PCSIRT (IRT 1257).
References
1 D. R. Owens, B. Zinman and G. B. Bolli, Lancet, 2001, 358,
739–746.
2 K. M. Bratlie, R. L. York, M. A. Invernale, R. Langer and
D. G. Anderson, Adv. Healthcare Mater., 2012, 1, 267–284.
3 V. Ravaine, C. Ancla and B. Catargi, J. Controlled Release,
2008, 132, 2–11.
4 K. Kumareswaran, M. L. Evans and R. Hovorka, Expert Rev.
Med. Devices, 2009, 6, 401–410.
5 Z. Gu, A. A. Aimetti, Q. Wang, T. T. Dang, Y. L. Zhang,
O. Veiseh, H. Cheng, R. S. Langer and D. G. Anderson, ACS
Nano, 2013, 7, 4194–4201.
6 Z. Gu, T. T. Dang, M. L. Ma, B. C. Tang, H. Cheng, S. Jiang,
Y. Z. Dong, Y. L. Zhang and D. G. Anderson, ACS Nano,
2013, 7, 6758–6766.
7 T. Miyata, T. Uragami and K. Nakamae, Adv. Drug Delivery
Rev., 2002, 54, 79–98.
Fig. 10 UV-CD spectra of standard insulin and released insulin.
This journal is © The Royal Society of Chemistry 2014
Soft Matter, 2014, 10, 911–920 | 919

View Article Online
Soft Matter
Paper
8 W. Qi, X. H. Yan, J. B. Fei, A. H. Wang, Y. Cui and J. B. Li, 33 C. Cheng, X. G. Zhang, Y. X. Wang and C. X. Li, J. Controlled
Biomaterials, 2009, 30, 2799–2806. Release, 2011, 152(suppl. 1), 267–269.
9 K. Makino, E. J. Mack, T. Okano and W. K. Sung, J. Controlled 34 Y. X. Wang, X. G. Zhang, Y. C. Han, C. Cheng and C. X. Li,
Release, 1990, 12, 235–239. Carbohydr. Polym., 2012, 89, 124–131.
10 V. Lapeyre, I. Gosse, S. Chevreux and V. Ravaine, 35 L. Sun, X. G. Zhang, C. Zheng, Z. M. Wu and C. X. Li, J. Phys.
Biomacromolecules, 2006, 7, 3356–3363. Chem. B, 2013, 117, 3852–3860.
11 K. Kataoka, H. Miyazaki, T. Okano and Y. Sakurai, 36 Z. M. Wu, X. G. Zhang, H. L. Guo, C. X. Li and D. M. Yu,
Macromolecules, 1994, 27, 1061–1062. J. Mater. Chem., 2012, 22, 22788–22796.
12 A. M. Horgan, A. J. Marshall, S. J. Kew, K. E. Dean, 37 L. Sun, X. G. Zhang, C. Zheng, Z. M. Wu, X. T. Xia and
C. D. Creasey and S. Kabilan, Biosens. Bioelectron., 2006,
21, 1838–1845.
13 V. Lapeyre, C. Ancla, B. Catargi and V. Ravaine, J. Colloid
Interface Sci., 2008, 327, 316–323.
C. X. Li, RSC Adv., 2012, 2, 9904–9913.
38 C. J. Ferguson, R. J. Hughes and D. Nguyen, Macromolecules,
2005, 38, 2191–2204.
39 A. Aqil, H. Qiu and J. F. Greisch, Polymer, 2008, 49, 1145–
1153.
14 D. Shiino, Y. Murata, A. Kubo, Y. J. Kim, K. Kataoka,
Y. Koyama, A. Kikuchi, M. Yokoyama, Y. Sakurai and 40 M. C. Lee, S. Kabilan, A. Hussain, X. P. Yang, J. Blyth and
T. Okano, J. Controlled Release, 1995, 37, 269–276. C. R. Lowe, Anal. Chem., 2004, 76, 5748–5755.
15 A. Matsumoto, S. Ikeda, A. Harada and K. Kataoka, 41 D. Roy, J. N. Cambre and B. S. Sumerlin, Chem. Commun.,
Biomacromolecules, 2003, 4, 1410–1416. 2008, 21, 2477–2479.
16 B. L. Wang, R. J. Ma, G. Liu, Y. Li, X. J. Liu, Y. L. An and L. Q. 42 Q. Yang, Z. K. Xu, Z. W. Dai, J. L. Wang and M. Ulbricht,
Shi, Langmuir, 2009, 25, 12522–12528. Chem. Mater., 2005, 17, 3050–3058.
17 T. Hoare and R. Pelton, Macromolecules, 2007, 40, 670–678. 43 R. Narain and S. P. Armes, Biomacromolecules, 2003, 4, 1746–
18 T. Hoare and R. Pelton, Biomacromolecules, 2008, 9, 733–740. 1758.
19 V. Lapeyre, C. Ancla, B. Catargi and V. Ravaine, J. Colloid 44 M. M. Bradford, Anal. Biochem., 1976, 72, 248–254.
Interface Sci., 2008, 327, 316–323.
20 Y. J. Zhang, Y. Guan and S. Q. Zhou, Biomacromolecules,
2006, 7, 3196–3201.
45 M. Sato, T. Furuike, R. Sadamoto, N. Fujitani, T. Nakahara,
K. Niikura, K. Monde, H. Kondo and S. I. Nishimura,
J. Am. Chem. Soc., 2004, 126, 14013–14022.
21 Y. J. Zhang, Y. Guan and S. Q. Zhou, Biomacromolecules, 46 A. P. Bapat, D. Roy, J. G. Ray, D. A. Savin and B. S. Sumerlin,
2007, 8, 3842–3847. J. Am. Chem. Soc., 2011, 133, 19832–19838.
22 V. Lapeyre, I. Gosse, S. Chevreux and V. Ravaine, 47 J. N. Cambre, D. Roy and B. S. Sumerlin, J. Polym. Sci., Part A:
Biomacromolecules, 2006, 7, 3356–3363. Polym. Chem., 2012, 50, 3373–3382.
23 A. Matsumoto, S. Ikeda, A. Harada and K. Kataoka, 48 S. Boduroglu, J. M. E. Khoury, D. V. Reddy, P. L. Rinaldi and
Biomacromolecules, 2003, 4, 1410–1416.
24 A. Matsumoto, R. Yoshida and
Biomacromolecules, 2004, 5, 1038–1045.
J. Hu, Bioorg. Med. Chem. Lett., 2005, 15, 3974–3977.
Kataoka, 49 S. Q. Li, E. N. Davis, J. Anderson, Q. Lin and Q. Wang,
Biomacromolecules, 2009, 10, 113–118.
K.
25 Y. J. Lee, S. A. Pruzinsky and P. V. Braun, Langmuir, 2004, 20, 50 R. J. Ma, H. Yang, Z. Li, G. Liu, X. C. Sun, X. J. Liu, Y. L. An
3096–3106. and L. Q. Shi, Biomacromolecules, 2012, 13, 3409–3417.
26 W. T. Wu, N. Mitra, E. C. Y. Yan and S. Q. Zhou, ACS Nano, 51 E. R. Ballister, A. H. Lai, R. N. Zuckermann, Y. Chenq and
2010, 4, 4831–4839.
27 A. Matsumoto, T. Ishii, J. Nishida, H. Matsumoto, K. Kataoka
J. D. Mouqous, Proc. Natl. Acad. Sci. U. S. A., 2008, 105,
3733–3738.
and Y. Miyahara, Angew. Chem., Int. Ed., 2012, 51, 2124–2128. 52 R. A. Miller, A. D. Presley and M. B. Francis, J. Am. Chem. Soc.,
28 H. Kim, Y. J. Kang, S. Kang and K. T. Kim, J. Am. Chem. Soc.,
2012, 134, 4030–4033.
29 C. Cheng, X. G. Zhang, Y. X. Wang, L. Sun and C. X. Li, J. Mol.
Catal. A: Chem., 2012, 36, 1413–1421.
2007, 129, 3104–3109.
53 T. Scheibel, R. Parthasarathy, G. Sawicki, X. M. Lin, H. Jaeqer
and S. L. Lindquist, Proc. Natl. Acad. Sci. U. S. A., 2003, 100,
4527–4532.
30 C. Zheng, Q. Q. Guo, Z. M. Wu, L. Sun, Z. P. Zhang, C. X. Li 54 W. T. Wu, N. Mitra, E. C. Y. Yan and S. Q. Zhou, ACS Nano,
and X. G. Zhang, Eur. J. Pharm. Sci., 2013, 49, 474–482. 2010, 4, 4831–4839.
31 X. J. Jin, X. G. Zhang, Z. M. Wu, D. Y. Teng, X. J. Zhang, 55 K. Hinds, J. J. Koh, L. Joss, F. Liu, M. Baudys and S. W. Kim,
Y. X. Wang, Z. Wang and C. X. Li, Biomacromolecules, 2009,
10, 1337–1345.
32 C. Cheng, X. G. Zhang, J. X. Xiang, Y. X. Wang, C. Zheng,
Z. T. Lu and C. X. Li, So Matter, 2012, 8, 765–773.
Bioconjugate Chem., 2000, 11, 195–201.
56 S. Lee, K. Kim, T. S. Kumar, J. Lee, S. K. Kim, D. Y. Lee,
Y. K. Lee and Y. Byun, Bioconjugate Chem., 2005, 16, 615–
620.
920 | Soft Matter, 2014, 10, 911–920
This journal is © The Royal Society of Chemistry 2014