Dual stimuli-responsive poly(2-hydroxyethyl methacrylate-co-methacrylic acid) microgels based on photo-cleavable cross-linkers: PH-dependent swelling and light-induced degradation

Article abstract of DOI:10.1021/ma201706r

Dual stimuli-responsive p(HEMA-co-MAA) microgels were prepared in a facile way by inverse miniemulsion copolymerization of 2-hydroxyethyl methacrylate (HEMA) with methacrylic acid (MAA) and two kinds of newly synthesized photodegradable cross-linkers. The pH-dependent swelling behavior induced by the protonation/deprotonation of the methacrylic acid groups in the network-forming polymer was investigated by means of the particle volume change as determined by DLS measurements. Photolytic degradation experiments were conducted by irradiation with UV light which led to particle disintegration caused by cleavage of the photolabile cross-linking points. The degradation behavior of the microgels was investigated with respect to degradation rates and changes in the degree of swelling. Those parameters were found to depend on the pH value of the solvent, the light intensities, and the irradiation wavelengths applied. For similar conditions, the degradation profile was demonstrated to strongly depend not only on the molecular structure of the cross-linking molecule but also on the molecular weight of the network-forming copolymers. The particular combination of the stimuli described is designed as a new strategy to two different swelling/degradation profiles. This dual stimuli-responsiveness was shown to enable the efficient loading and subsequent release of myoglobin as a model protein. Here a slow diffusion controlled release (induced by changes of the pH) was combined with a fast degradation controlled (induced by irradiation) on-demand release. This novel two-step release profile is proposed to bear great potential for delivery applications.

Full text of DOI:10.1021/ma201706r

ARTICLE
pubs.acs.org/Macromolecules
Dual Stimuli-Responsive Poly(2-hydroxyethyl
methacrylate-co-methacrylic acid) Microgels Based on Photo-Cleavable
Cross-Linkers: pH-Dependent Swelling and Light-Induced Degradation
Daniel Klinger and Katharina Landfester*
Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
S Supporting Information
b
ABSTRACT: Dual stimuli-responsive p(HEMA-co-MAA) mi-
crogels were prepared in a facile way by inverse miniemulsion
copolymerization of 2-hydroxyethyl methacrylate (HEMA)
with methacrylic acid (MAA) and two kinds of newly synthe-
sized photodegradable cross-linkers. The pH-dependent swel-
ling behavior induced by the protonation/deprotonation of the
methacrylic acid groups in the network-forming polymer was
investigated by means of the particle volume change as deter-
mined by DLS measurements. Photolytic degradation experi-
ments were conducted by irradiation with UV light which led to
particle disintegration caused by cleavage of the photolabile cross-linking points. The degradation behavior of the microgels was
investigated with respect to degradation rates and changes in the degree of swelling. Those parameters were found to depend on the
pH value of the solvent, the light intensities, and the irradiation wavelengths applied. For similar conditions, the degradation profile
was demonstrated to strongly depend not only on the molecular structure of the cross-linking molecule but also on the molecular
weight of the network-forming copolymers. The particular combination of the stimuli described is designed as a new strategy to two
different swelling/degradation profiles. This dual stimuli-responsiveness was shown to enable the efficient loading and subsequent
release of myoglobin as a model protein. Here a slow diffusion controlled release (induced by changes of the pH) was combined with
a fast degradation controlled (induced by irradiation) on-demand release. This novel two-step release profile is proposed to bear
great potential for delivery applications.
’ INTRODUCTION
One interesting class of stimuli-responsive microgels for
release applications is based on changing the physicochemical
properties of the network-forming polymers upon application
of an external trigger. This concept is mainly realized by either
temperature-16,17 or pH-sensitive gel particles14,18ꢀ20 and can be
applied for controlling the diffusion of a payload by several
mechanisms: (i) triggered swelling and deswelling13,21,22 allows
the adjustment of mesh sizes relative to the hydrodynamic radii
of the functional substances (e.g., proteins, enzymes, catalytic
metal nanoparticles);23 (ii) ionization/deionization of a network-
forming polyelectrolyte allows the adjustment of electrostatic
interactions of e.g. small molecules with the polymeric matrix24,25
and (iii) embedding compounds into initially collapsed networks26—
thus excluding water from the gel interior—prevents diffusion
prior to a triggered swelling-induced release.
Since the discovery of microgels as a unique class of polymeric
nanoscale materials, this research area has gained increasing
attention. In general, the exceptional properties of microgels
stem from the combination of their colloidal nature with their
internal network structure.1 These 3-dimensional networks in
the nanoscale represent highly functional materials exhibiting
bespoken properties for a tremendous variety of applications
such as sensors,2 optics,3,4 colloidal crystals,4,5 and release appli-
cations in materials science6ꢀ8 and biomedical fields.9,10
Regarding the development of microgels for release applica-
tions in general, one of the main tasks is to combine an efficient
loading process with a well-defined release profile.11 Here, it is of
high interest to prevent leakage of the embedded functional
compound until the targeted site or time point is reached. One
interesting approach concept to realize this aim is based on the
ability to control the diffusion of compounds in the network.12,13
Since the latter is characterized by parameters such as the mesh
size, the polymer volume fraction or the interaction of embedded
functional compounds with the network, the ability to control
these factors by the application of external triggers represents the
underlying concept to stimuli-responsive microgels.14,15
An alternative approach to achiev e controlled release out of
gels is based on the use of cleavable cross-linking points allow-
ing the complete decomposition of the network architecture
by using external stimuli. In the context of such degradation
Received: July 25, 2011
Revised:
November 4, 2011
Published: November 28, 2011
r
2011 American Chemical Society
9758
dx.doi.org/10.1021/ma201706r Macromolecules 2011, 44, 9758–9772
|

Macromolecules
ARTICLE
Figure 1. Schematic representation of the double stimuli-responsive character of the photodegradable p(HEMA-co-MAA) microgels.
controlled delivery systems, e.g., enzyme-sensitive microgels
containing dextran-based cross-linkers,27,28 reducing agent sen-
sitive cross-linkers based on disulfide bonds,29 or acid-degradable
microgels synthesized with cross-linkers based on tertiary esters30
or acetals,31,32 have been investigated for delivery systems in
organic and aqueous media. Here, relatively large functional
compounds—such as, e.g., DNA31 or proteins33—are primarily
entrapped in a network due to initial mesh sizes smaller than the
hydrodynamic diameters of the respective compounds to be
delivered. Upon the degradation of cross-linking points, the
cross-linking density decreases which results in increased mesh
sizes. Thus, the release of functional compounds is enabled.
Regarding particulate delivery systems solely based on pH-
sensitive materials, either degradation or diffusion controlled
release occurs as a response to the H+ concentration of the local
environment, thus rendering those particles “smart”. Such con-
cepts are investigated for drug delivery applications by exploiting
the acidic pH of malignant tissue and lysosomes.23,34 Never-
theless, these systems require the addition of protons or the
localization in acidic milieus.
In comparison, inducing the release of a payload by applying
an external trigger without a change in the chemical composition
of the surrounding media bears the advantage of an on-demand
delivery. Concerning this point, light represents an outstanding
position as it can be applied in a very precise manner by selecting
suitable wavelengths, polarization directions and intensities in a
noncontact approach. In addition, light offers the possibility to
change the polymer properties in very confined spaces and time
scales. Exploiting these advantages for triggered release applica-
tions can in general be realized by the utilization of a broad
variety of chromophores. Among those, o-nitrobenzyl moieties
are very attractive due to the fast and quantitative reaction of the
chromophore upon irradiation with UV light.35,36 The good
photolytic properties in various solvents and even the solid37 or
macromolecular38,39 state render those moieties highly interest-
ing for a broad variety of applications.40,41 In particular, the
tunable36 enhanced absorption for wavelengths of λ > 300 nm
enables photoreactions to be carried out in a controlled manner
and under mild conditions suitable for biological applications.42,43
Recent investigations based on light-sensitive materials for
release applications include photoswitchable block copolymer
micelles for the delivery of hydrophobic compounds44ꢀ47 as well
as photoresponsive nanocarriers based on capped dendrimers48
or hyperbranched polyglycerol nanocapsules.49 Additionally, we
recently reported on the formation of light-sensitive PMMA
microgels containing photodegradable cross-linking points based
on o-nitrobenzyl moieties.50 Since these materials exhibited well
pronounced light-triggered swelling and degradation profiles in
organic solvents, it is highly interesting to transfer this concept to
aqueous media by the incorporation of such cross-linking molecules
into hydrogel nanoparticles and to investigate the potential of these
materials for loading and release applications.
On the basis of these considerations, the aim of this work is
the preparation of dual stimuli-responsive microgels to exhibit a
pH-dependent volume phase transition due to ionizable metha-
crylic acid (MAA) groups in poly(2-hydroxyethyl methacrylate)
(PHEMA) as network-forming polymer and the ability to com-
plete particle disintegration upon irradiation due to light-
cleavable cross-linking points. The particular combination of the
mentioned stimuli is designed to exhibit two different degradation
profiles as shown in Figure 1. In both cases, particles are proposed
to be highly swollen for pH values higher than the pKa of the
microgels, whereas decreasing the pH leads to a deswelling of the
networks.24,25 The disintegration of the collapsed particles by
direct irradiation represents a one-step degradation profile. In
contrast, by transferring the collapsed particles to phosphate buffered
saline (PBS), slightly swollen particles are obtained, which can
subsequently be degraded by the application of UV light, therefore
characterizing a two-step swelling/degradation profile.
’ EXPERIMENTAL SECTION
Materials. All chemicals were purchased from Sigma-Aldrich and
used without further purification unless otherwise stated. 2-Hydroxy-
ethyl methacrylate (HEMA) and methacrylic acid (MAA) were freshly
distilled under reduced pressure prior to use.
Instrumentation. 1H (300 MHz) and 13C NMR (75 MHz) spectra
were measured using a Bruker spectrometer. Particle size distributions
were determined by dynamic light scattering (DLS) using a NICOMP
Zetasizer measuring at a fixed scattering angle of 90°. The measurements
were carried out at 25 °C on diluted dispersions in the respective
solvents. A Perkin-Elmer Lambda 25 UVꢀvis spectrometer was used to
measure UVꢀvis spectra. Irradiations were either carried out by using a
365 nm emitting UV-LED with a maximum power of 150 W or an
OSRAM HBO 100 W/2 mercury short arc lamp combined with a UG-11
and a W-320 filter resulting in an output of wavelengths of λ = 315ꢀ390 nm.
HPLC measurements were conducted using an Agilent quarternary gradient
pump (series 1100) combined with an Agilent photodiode array detector
(DAD series 1200). A Gemini 1530 (Carl Zeiss AG, Oberkochem,
9759
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
Scheme 1. Synthesis of Photolabile Cross-Linkers: Reagents and Conditionsa
a Key: (a) methyl 4-bromobutyrate, K2CO3/DMF (anhyd.), 25 °C, 16 h, quant.; (b) acetic anhydride/nitric acid (1:2, v/v), 0 °C, 3 h, 53%; (c,d)
NaBH4/MeOH/THF, 25 °C, 16 h, NaOH, 25 °C, 7 h, 89%; (e-1) BH3*THF/THF, 0°C, 1 h then 40 °C, 18 h, 86%; (e-2) 2-isocyanatoethyl
methacrylate, DBTDL (cat.)/THF, 65 °C, 48 h, 63%; (f-1) methacryloyl chloride, NEt3/DCM, 0 to 25 °C, 15 h, 42%; (f-2) ethylenediamine, DCC,
DMAP (cat.)/DCM, 0 to 25 °C, 15 h, 53%.
Germany) with an InLens detector was used to take scanning electron
micrographs (SEM). The samples were prepared by drop-casting of diluted
dispersions on a silicon wafer. UVꢀvis measurements for the loading and
release experiments were performed on a Tecan plate reader.
THF for 18 h at 40 °C. The destruction of excess borane by adding water
was followed by extracting the mixture with ethyl acetate. The dried organic
layers were recrystallized from chloroform yielding pure HEMNPB (6) in
86% yield.
1H NMR (CDCl3): δ = 7.55 (s, 1H), 7.29 (s, 1H), 5.55 (q, 1H, J =
6.3 Hz), 4.10 (t, 2H, J = 6.1 Hz), 3.97 (s, 3H), 3.73 (t, 2H, J = 6.2 Hz),
1.98 (m, 2H), 1.77 (m, 2H), 1.55 (d, 3H, J = 6.3 Hz). 13C NMR
(CDCl3): δ = 153.67, 146.38, 138.87, 137.94, 108.62, 108.30, 68.91,
64.65, 61.40, 55.96, 28.89, 25.23, 24.60. MS (FD): m/z 285.0 (M+).
2-((1-(4-(4-(2-(Methacryloyloxy)ethylcarbamoyloxy)bu-
toxy)-5-methoxy-2-nitrophenyl)ethoxy)carbonylamino)-
ethyl methacrylate (CL-A). Synthesis was performed according
to the literature50 by the reaction of 2-isocyanatoethyl methacrylate
with HEMNPB (6) and dibutyltin dilaurate in anhydrous THF.
Purification by column chromatography over silica using CHCl3/
MeOH (10:1) as eluent yielded CL-A as slightly yellow colored oil;
yield 63%.
Synthesis of Photo-Degradable Cross-Linkers. On the basis
of the photolabile molecules of either 4-(4-(1-hydroxyethyl)-2-methoxy-5-
nitrophenoxy)butan-1-ol (HEMNPB) or 4-(4-(1-hydroxyethyl)-2-meth-
oxy-5-nitrophenoxy)butanoic acid (HEMNPBA), two different cross-linkers
were synthesized. The synthetic concept is presented in Scheme 1.
4-(4-(1-Hydroxyethyl)-2-methoxy-5-nitrophenoxy)buta-
noic acid (HEMNPBA) (5). As described in a previous publication,50
the product was prepared based on the synthetic protocols by
Holmes.36,51 Briefly, acetovanillone was reacted with methyl 4-bro-
mobutyrate in the presence of potassium carbonate to form the
corresponding keto ester. After removing the inorganic salts by
extraction with ethyl actetate, subsequent nitration of the crude
product was achieved by the reaction with acetic anhydride and nitric
acid (1:2 v/v). In the next step, the reduction of the keto group with
excess borohydride was conducted and directly followed by inducing
the ester cleavage by the addition of NaOH in water. Purification by
recrystallization from ethyl acetatet/hexane afforded HEMNPBA (5)
in 89% yield as a pale yellow solid.
1H NMR (CDCl3): δ = 7.56 (s, 1H), 6.98 (s, 1H), 6.37 (q, 1H, J = 6.3
Hz), 6.10 (d, 2H, J = 7.7 Hz), 5.59 (m, 2H), 5.06 (s, 1H), 4.96 (t, 1H),
4.13 (m, 8H), 3.93 (s, 1H), 3.48 (m, 4H), 1.93 (d, 6H, J = 4.6 Hz), 1.81
(m, 4H), 1.81 (m, 1H),1.59 (d, 3H, J = 6.3 Hz). 13C NMR (CDCl3): δ =
167.29, 156.47, 155.20, 153.97, 147.20, 139.69, 135.96, 133.81, 126.05,
108.92, 69.10, 68.85, 64.48, 63.70, 63.65, 56.30, 40.18, 25.62, 25.46,
22.17, 18.27. MS (FD): m/z 595.5 (M+).
4-(4-(1-(Methacryloyloxy)ethyl)-2-methoxy-5-nitrophe-
noxy)butanoic acid (7). To an ice-cooled suspension of 5 (1.750 g,
5.85 mmol) and triethylamine (1.775 g, 17.54 mmol) in 70 mL of
anhydrous DCM, methacryloyl chloride (1.746 g, 14.62 mmol) was
added dropwise at 0 °C under argon atmosphere. The reaction
mixture was allowed to warm to room temperature while stirring
1H NMR (CDCl3): δ = 7.41 (s, 1H), 7.24 (s, 1H), 5.37 (q, J = 6.2 Hz,
1H), 3.97 (t, J = 6.4 Hz, 2H), 3.82 (s, 3H), 2.37 (t, J = 7.2 Hz, 2H), 2.00
(p, J = 6.8 Hz, 3H), 1.35 (d, J = 6.3 Hz, 1H). 13C NMR (CDCl3):
δ = 174.92, 154.05, 146.56, 139.12, 138.33, 108.97, 108.89, 68.24, 64.95,
56.23, 30.18, 24.84, 24.18. MS (FD): m/z 299.2 (M+).
4-(4-(1-Hydroxyethyl)-2-methoxy-5-nitrophenoxy)butan-1-
ol (HEMNPB) (6). The photolabile molecule HEMNPB was ob-
tained by reduction of the carboxy group with 2.5 equiv of BH3 THF in
3
9760
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
overnight and the resulting solution was washed with Na2CO3 (5%),
HCl (1%), and water. The organic solvent was evaporated, the residue
was dissolved in acetone/water and stirred overnight, insoluble
components were removed by filtration and the mixture was extracted
with DCM. After washing the resulting solution with HCl (1%) and
water, the organic phase was dried over MgSO4 and the solvent was
removed. Purification by column chromatography over silica using
MeOH/CHCl3 (1:5 v/v) as eluent yielded pure 7 in 43% yield.
1H NMR (CDCl3): δ = 7.59 (s, 1H), 7.02 (s, 1H), 6.52 (q, 1H), 6.17
(s, 1H), 5.61 (s, 1H), 4.12 (t, 2H), 3.92 (s, 3H), 2.61 (t, 2H), 2.19 (qnt,
2H), 1.95 (s, 3H), 1.66 (d, 3H). 13C NMR (CDCl3): δ = 178.21, 166.08,
154.01, 147.12, 139.76, 136.35, 133.58, 125.73, 109.10, 108.11, 68.76,
68.04, 56.20, 30.21, 23.99, 22.04, 18.32.
1,10-(4,40-(4,40-(Ethane-1,2-diylbis(azanediyl))bis(4-oxobu-
tane-4,1-diyl))bis(oxy)bis(5-methoxy-2-nitro-4,1-phenylene))-
bis(ethane-1,1-diyl)bis(2-methyl acrylate) (CL-B). A solution of 7
(0.847 g, 2.30 mmol), DMAP (0.054 g, 0.44 mmol) and ethylene diamine
(0.065 g, 1.10 mmol) in 40 mL of anhydrous DCM was cooled to 0 °C and
stirred at this temperature for 30 min. DCC (0.520 g, 2.52 mmol) was
dissolved in 10 mL of anhydrous DCM and added to the reaction mixture
which was allowed to warm to room temperature while stirring overnight.
Filtration to remove the produced urea was followed by washing the organic
solution with saturated NH4Cl, saturated Na2CO3, and brine. The organic
phase was dried over MgSO4 and the solvent was evaporated under reduced
pressure. The resulting residue was purified by column chromatography
using silica and MeOH/CHCl3 as eluent yielding pure CL-B as a pale yellow
solid (0.422 g, 0.56 mmol). Yield 51%.
Table 1. Synthetic Details for the Miniemulsion Polymeri-
zations of p(HEMA-co-MAA) Microgels
cross-linker
sample
DMSO (m/mg)
type
---
amount, m/mg amount, mol %
MG-10
MG-1A
MG-20
MG-2A
MG-2B
---
---
---
185
---
---
2.5
---
2.5
2.5
CL-A
---
375
375
375
CL-A
CL-B
185
235
phase consisting of a solution of the amphiphilic block copolymer
P(E/B)-b-PEG (75 mg) as nonionic surfactant in 12.5 g of cyclohexane.
The miniemulsion was formed by first stirring the mixture at 1750 rpm
for 1.5 h and then homogenizing the obtained pre-emulsion by ultra-
sonication for 2 min at 90% intensity (Branson sonifier W450 Digital, 0.5
in. tip) at 0 °C. 2,20-Azobis(2-methyl-butyronitrile) (V-59) (50 mg,
0.3 mmol) as thermal initiator was added and polymerizations were then
carried out overnight in an oil bath set at 70 °C. The used amounts
of cross-linkers and DMSO for the different reactions are summarized
in Table 1.
After the polymerization, coagulates were removed by filtration over
kimwipe paper and the resulting dispersions were centrifuged at
10 000 rpm for 30 min to collect the particles. The supernatant was
removed and replaced by cyclohexane. Redispersion was carried out
using a vortex. In order to remove excess surfactant, the dispersions were
further washed four times with cyclohexane, following the procedure
described above. The purified dispersions of the MGs were examined
with regard to the particle size distributions by means of dynamic light
scattering (DLS) and scanning electron microscopy (SEM). Freeze-
drying removed unreacted monomer and yielded the cross-linked
p(HEMA-co-MAA) gel particles as white to slightly yellow colored
powders. The freeze-dried particles were swollen overnight at 5% (w/v)
in water adjusted to pH 12 at room temperature. Three additional
washing steps by centrifugation at 17 500 rpm for 90 min at 15 °C and
redispersion in deionized water were performed to remove the sol
content which consisted of all the soluble polymers and non-cross-linked
particles. The purified particles were freeze-dried and redispersed in the
respective media by simple swelling at room temperature for 6 h at the
desired concentration.
1H NMR (CDCl3): δ = 7.58 (s, 2H), 7.01 (s, 2H), 6.50 (q, 2H), 6.46
(s, 2H), 6.17 (s, 2H), 5.61 (s, 2H), 4.09 (t, 4H), 3.92 (s, 6H), 3.38 (dd,
4H), 2.40 (t, 4H), 2.16 (qnt, 4H), 1.95 (s, 6H), 1.65 (d, 6H). 13C NMR
(CDCl3): δ = 173.44, 166.23, 154.02, 147.23, 139.96, 136.46, 133.68,
125.89, 109.21, 108.24, 68.87, 68.70, 56.39, 40.38, 32.90, 24.88, 22.14,
18.43. MS (FD): m/z 759.6 (M + 1+), 1518.1 (2M+).
Photo-Degradation Studies of Cross-linkers in Solution
via 1H NMR and Mass Spectroscopy. Cross-linkers were dissolved
in THF with a concentration of 3.0 ꢁ 10ꢀ3 mol/L. Irradiations were
performed in a quartz cuvette for 10 h using a mercury short arc lamp
combined with the described filters and an intensity of I = 17 mW/cm2
for the wavelengths of λ = 315ꢀ390 nm. The reaction mixture was
divided into two fractions and both were evaporated to dryness under
reduced pressure. One fraction was investigated by 1H NMR spectros-
copy, the other fraction was investigated by field desorption mass
spectrometry.
pH-Dependent Swelling Studies of Cross-Linked p(HEMA-
co-MAA) Microgels. Particle size distributions were determined by
DLS in water as described above. SEM analysis was conducted to
determine the morphologies of the particles.
Photo-Degradation Studies by UVꢀVis Measurements. A
quartz cuvette containing a solution (c = 4.5 ꢁ 10ꢀ4 mol/L) of
the respective cross-linker in THF was placed under a UV lamp
(λ = 315ꢀ390 nm, I = 17 mW/cm2) and the samples were irradiated
for fixed time intervals. UVꢀvis measurements were performed
subsequently.
Photo-Degradation Studies of Cross-Linked Gel Particles
in Water. Samples of 0.125% (w/v) MG in phosphate buffer saline
(PBS) or water adjusted to the desired pH were placed in a quartz
cuvette and irradiated with the respective UV light source. At prede-
termined times, samples were collected and turbidity measurements
were conducted by using a NICOMP Zetasizer. The volume of the
irradiated sample was retained by returning the withdrawn samples to
the cuvette after every measurement. The turbidity was obtained by
calculating the ratio of the scattering intensity at 90° of the irradiated
samples relative to the one of the nonirradiated samples. Particle
size distributions of the irradiated particle dispersions were obtained
by DLS in the respective solvent. Furthermore, SEM images were taken
to investigate the morphology of the irradiated particles.
Kinetic HPLC Measurements. Cross-linker solutions in THF
were prepared and irradiated analogously to those used for the 1H NMR
spectroscopic measurements. Samples were taken at fixed time intervals
and directly measured by HPLC using a solvent gradient of THF/water
from 40/60 to 30/70. Photolytic conversion of the cross-linkers was
determined by the decrease of the peak area of the starting compound.
Preparation of Cross-Linked Photo-Degradable p(HEMA-
co-MAA) Gel Particles by Inverse Miniemulsion Polymeriza-
tion. Cross-linked p(HEMA-co-MAA) microgels (MG) were synthe-
sized by free radical copolymerization in inverse miniemulsion.
Microgels MG-10 and MG-1A were prepared by first mixing HEMA
(1.275 g, 9.8 mmol), MAA (0.225 g, 2.6 mmol), water (70 mg,
3.9 mmol), and 2.5 mol % of the respective cross-linker. In case of the
microgels MG-20, MG-2A, and MG-2B, additional DMSO (0.375 g) as
solvent was added. Afterward, the mixture was added to the continuous
Loading of p(HEMA-co-MAA) Microgels with Myoglobin
and Subsequent Release Experiments. To 18 mL of a stock
solution of myoglobin (c = 1 mg/mL) in water with a pH adjusted to pH
6.75 were added 2 mL of a dispersion of MG-2B p(HEMA-co-MAA)
microgels in water (c = 1 mg/mL). The pH of the mixture was adjusted
9761
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
Scheme 2. Photo-Reactions of the Cleavable Cross-Linkers: (a) CL-A and (b) CL-B
to pH 6.75 by the addition of diluted NaOH and HCl. After equilibration
at room temperature overnight, the pH was adjusted to pH 4.5 by the
addition of diluted HCl and the particles were collected by centrifuga-
tion (10 000 rpm; 30 min). The supernatant (20 mL) was examined by
UVꢀvis spectroscopy. The pellet of the centrifugation was redispersed
in 20 mL of water adjusted to pH 4.5 using a vortex. The loaded
microgels were washed four more times following the described pro-
cedure and the supernatants of every washing step were investigated by
UVꢀvis spectroscopy. Subsequently to the last centrifugation step,
the microgels were redispersed in PBS with pH 7.4 (20 mL). A sample
(2 mL) was taken and investigated by UVꢀvis spectroscopy. The
dispersion of the loaded microgels in PBS was shaken at 37 °C and
samples were taken after predetermined time intervals. The release of
myoglobin into the surrounding medium was investigated by removing
the microgels from each sample by centrifugation and comparing the
UVꢀvis spectra of the supernatants to a calibration curve of myoglobin
in PBS. Irradiation-induced release experiments were conducted by
irradiating samples of 4 mL in a quartz cuvette under stirring for 90 min
with UV light of the wavelengths of λ = 315ꢀ390 nm and an intensity of
17 mW/cm2. The loading and release experiments were performed
in duplicate and UVꢀvis measurements were conducted on 2 different
dilutions of the samples. Every measurement was performed in tripli-
cate using a Tecan plate reader. Reported values represent the average
of all measurements and errors correspond to the respective standard
deviation.
induced swelling/collapsing and the subsequent light-induced
degradation of the microgels as shown in Figure 1.
This dual stimuli-responsive character was shown to enable
the efficient loading and release of myoglobin as a model protein.
A new two-step—pH-induced swelling followed by light-induced
degradation—release profile was demonstrated.
Synthesis and Characterization of Photo-Labile Cross-
Linkers. Two different types of photolabile cross-linkers were
synthesized in order to investigate the influence of the molecular
structure on the degradation behavior of the hydrogel particles.
The synthetic pathway and the reaction conditions are shown in
Scheme 1.
Both structures are based on 4-(4-(1-hydroxyethyl)-2-
methoxy-5-nitrophenoxy)butanoic acid (HEMNPBA) which was
synthesized accordingly to a previous publication.50 Briefly, aceto-
vanillone was reacted with methyl 4-bromobutyrate to the
corresponding keto ester (2) under basic conditions. A subse-
quent nitration of the crude product was achieved by the reaction
with acetic anhydride and nitric acid. In the next step, the
reduction of the keto group of 3 with excess borohydride was
conducted and directly followed by inducing the ester cleavage
by the addition of NaOH in water yielding HEMNPBA (5). CL-
A was synthesized by first reducing the acid group of 5 to the
corresponding alcohol and then reacting 6 with 2 equiv of
2-isocyanatoethyl methacrylate in the presence of catalytic
amounts of DBTDL to form the carbamate. In the case of CL-
B, HEMNPBA (5) was reacted with an excess amount of
methacryloyl chloride to form the o-nitrobenzyl ester of 7.
The symmetric cross-linker CL-B was then prepared by DCC
coupling of 2 equiv of 7 with ethylene diamine.
Both cross-linkers are based on nitroveratryloxycarbonyl
derivatives as the photoreactive moieties but vary in the way of
the attachment of the polymerizable methacrylate units to the
chromophore core. The o-nitrobenzyl group in general is char-
acterized by its chemical stability and rapid cleavage upon expo-
sure to UV light. The α-methyl group on the benzylic carbon of
the o-nitrobenzyl core is known to increase the rate of photolysis
significantly52 and was therefore introduced into the molecules.
Considering the abstraction of a benzylic proton by the photo-
activated nitro group as the rate-limiting step in the photolysis of
the o-nitrobenzyl group, the additional methyl group increases
’ RESULTS AND DISCUSSION
The concept of novel light- and pH-sensitive p(HEMA-co-
MAA) microgels was realized by combining two orthogonal
approaches. The incorporation of MAA into the microgels
resulted in a pH-dependent swelling behavior of the particles.
Protonation/deprotonation of the carboxylic acid groups changed
the physicochemical properties of the network-forming co-
polymers and therefore caused the collapse at low pH and the
expansion at pH values higher than the apparent pKa of the
respective microgels. Light-sensitivity of the hydrogel parti-
cles was achieved by the utilization of two different photolabile
cross-linking molecules. Irradiation of the microgels led to the
degradation of the cross-linking points and therefore to the dis-
integration of the particle structure. This dual stimuli-responsive
performance was designed to enable the independent pH-
9762
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
the acidity of this proton. Furthermore, the introduction of alkoxy
substituents in the o-nitrobenzylic core results in a considerably
increased UV absorption for λ > 315 nm.39
Generally, photolysis of the cross-linkers only requires the
irradiation wavelength to overlap with the absorption band of the
photolabile chromophore. However, degradation of cross-linkers
and the resulting p(HEMA-co-MAA) gel particles should be
performed in a controlled manner under mild conditions. Thus,
Norrish-type side reactions, which primarily take place for
irradiations with wavelengths of λ < 300 nm,53,54 should be
avoided. In addition, to evade undesirable side effects in biolo-
gical applications, irradiations with UV-A light of wavelengths
λ > 315 nm is favored. UVꢀvis spectroscopy of the cross-linkers
CL-A and CL-B showed in every case an absorption maximum at
300 nm and an additional absorption maximum at 342 nm (see
Supporting Information). Thus, both cross-linkers exhibit a high
absorption coefficient in the targeted photolysis wavelength
region of λ > 315 nm.
Photolysis of the Cleavable Cross-Linkers in Solution.
Although either type of cross-linker consists of one or two central
o-nitrobenzylic group(s), the variation of the molecular attach-
ment of the radically polymerizable methacrylate moieties onto
the photolabile chromophore results in different photoproducts
with specific functional groups. The respective photoreactions
are depicted in Scheme 2.
The carbamate derivative of cross-linker CL-A degrades into
an amine and a ketone (CL-A-UV), whereas irradiation of the
ester of CL-B generates two carboxylic acid groups and a
molecule containing two ketone moieties (CL-B-UV-2). Earlier
investigations on the photolysis of CL-A revealed that upon
longer irradiation times with UV light of the wavelengths of
λ > 300 nm, dimeric azoxybenzene compounds are formed from the
primary photoproducts (CL-A-UV) in THF solution. As shown
for photodegradable PMMA microgels, the covalent attachment
of those side products to the polymer backbone hinders com-
plete particle degradation due to the formation of new cross-
linking points in the respective gels.50 Even though a dramatic
increase in the degree of swelling for those particles could be
detected for the mentioned irradiation conditions, complete
microgel disintegration enhances the potential for delivery appli-
cations. Therefore, the symmetric molecular structure of CL-B
was designed to generate only carboxylic acid groups on the
polymer backbone by photolytic cleavage of the o-nitrobenzyl
ester bonds. The separated middle part (CLꢀB-UV-2) of the
cross-linking molecule contains both of the chromophore cores
and is not connected to the gel-forming polymers any more
(compare Scheme 2b). Thus, a possible formation of dimeric side
products of the photoproduct is assumed to result in liberated
molecules not hindering the particle degradation. In addition, the
formed methacrylic acid groups for CL-B in the polymer back-
bone are supposed to enhance the swelling of the anionic gels
whereas the amine groups generated upon photolysis of CL-A
are—due to electrostatic interactions and dependent on the
pH—able to form physical cross-links with the methacrylic acid
groups in the p(HEMA-co-MAA) copolymer.
Figure 2. Investigations on the photolytic cross-linker degradation.
Irradiation time-dependent HPLC measurements of (a) CL-A and (b)
CL-B.
absorption maxima to λmax = 385 nm. These light-induced
changes of the absorption spectra point toward successful
photoreactions. However, the formation of possible side prod-
ucts absorbing light in the same spectral region could not be
excluded. In the case of CL-A, no further changes in the
absorption spectra were detected after an irradiation time of
5 min indicating no further conversion of the photoreaction (see
Supporting Information, Figure S1a). In comparison, more than
10 min of irradiation were necessary to complete the light-
induced cleavage of the CL-B molecule (see Supporting Infor-
mation, Figure S1b). Comparing the molecular structure of CL-A
to CL-B, it becomes obvious that for the same molar concentra-
tion in THF, in the latter case two chromophores per molecule
have to be cleaved. As a result, the required irradiation time is
increased.
In order to perform a quantitative time-dependent degrada-
tion analysis of the cross-linkers, HPLC measurements were
conducted to examine the composition of the reaction mixtures
at various irradiation times (Figure 2, parts a and b). The kinetics
of the photolysis of the photolabile molecules was therefore
investigated by monitoring the rate of disappearance of the start-
ing compounds during irradiation. Kinetic plots of ꢀln[CL]t/
[CL]0 versus time show all excellent linearity, indicating the
In comparison to the studies performed on CL-A,50 the
photodegradation behavior of the cross-linkers in equimolar
THF solutions was first investigated by irradiation time-dependent
UVꢀvis measurements. The respective spectra are shown in
Figure S1 in the Supporting Information. Irradiation of either the
CL-A or the CL-B cross-linker with UV light of the wavelengths
λ = 315ꢀ390 nm resulted in every case in a red shift of the
9763
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
expected first order kinetic with respect to the chromophore
concentration (see Figure 3). Figure 2 presents the time depen-
dent HPLC elution curves of the photolysis of CL-A in compar-
ison to those of CL-B. The kinetic plots for the calculation of the
rate constants for both cross-linkers and obtained half-life times
and rate constants are listed in Figure 3.
Both cross-linkers were found to degrade completely upon
irradiation with UV light of wavelengths with λ = 315ꢀ390 nm.
As shown before,50 the elugram of cross-linker CL-A (Figure 2a)
exhibits the direct formation of one main photoproduct (peak
at 4.25 min) accompanied by the generation of the dimeric
azoxybenzene side product (peak at 4.5 min). In comparison, the
elugram of CL-B (Figure 2b) displays the formation of an
intermediary compound (peak at 6.6 min) during irradiation in
addition to the main photoproduct (peak at 5.0 min). The
intermediate was assigned to the compound generated by the
unilateral cleavage of one o-nitrobenzylic ester group of the
symmetric bifunctional molecule. Further irradiation leads to
the complete cleavage of both chromophore units and there-
fore results in one main photoproduct, which was assigned to the
peak at 5.0 min. It was demonstrated that the variation of the
molecular structures of the cross-linkers CL-A and CL-B results
in different degradation profiles whereas CL-A reacts around
1.5 times faster than CL-B. Taking into account absorption of the
(intermediary) photoproducts in the same spectral region as the
wavelengths used for the cleavage of the cross-linkers, the
photoreaction of the two chromophores containing CL-B is
significantly slowed down compared to CL-A. Nevertheless, even
a unilateral cleavage of the CL-B molecule is assumed to already
enable the degradation of a polymeric network. In addition, the
complete cleavage of CL-B for longer irradiation times does not
generate any polymer bound dimeric side products, which is
expected to facilitate the complete particle degradation under
mild conditions.
Preparation and Characterization of Cross-Linked Photo-
Degradable p(HEMA-co-MAA) Hydrogel Nanoparticles by
Inverse Miniemulsion Copolymerization. Regarding the pre-
paration of polymeric nanoparticles by copolymerization of
different monomers and cross-linkers, the miniemulsion poly-
merization technique is highly advantageous compared to other
techniques for microgel preparation such as, e.g., precipitation
polymerization. As the latter is strongly dependent on a specific
hydrophobicity of the materials prepared, hydrophilic (co)mono-
mers are only incorporable to a certain extent.55 Moreover, it is
limited to the utilization of water-soluble monomers.20 In con-
trast, (inverse) miniemulsions are characterized by suppression of
diffusion of compounds between droplets, rendering those as
nanoreactors.56ꢀ58 As a result the composition of the latex
particles resembles the composition of the monomer phase in
the droplets. In particular, for the copolymerization of different
Figure 3. Kinetic investigations on the photolytic cross-linker degrada-
tion: Kinetic plots of ꢀln[CL]t/[CL]0 versus irradiation time for CL-A
and CL-B, respective rate constants and half times.
Figure 4. Representative SEM images of p(HEMA-co-MAA) microgels drop cast from cyclohexane dispersions and hydrodynamic diameters obtained
from DLS measurements in cyclohexane.
9764
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
Figure 5. Representative SEM images of photodegradable p(HEMA-co-MAA) microgels drop cast from aqueous dispersions at pH 8.5.
particle dispersions were repeatedly washed with cyclohexane
after polymerization yielding stable dispersions. Hydrodynamic
diameters of the nonswollen MGs were determined by DLS
measurements in cyclohexane and were found to be in the size
range of around 130 nm for MG-10 and MG-1A particles and
about 85 nm for MG-20, MG-1A, and MG-2A microgels. For the
miniemulsions including DMSO as solvent for the dispersed
phase, smaller particles were obtained after polymerization. This
effect might be either explained by a changed surface tension or a
decreased viscosity in this system. Figure 4 shows representative
SEM pictures of the drop cast cyclohexane dispersions on silica
wafers together with the hydrodynamic diameters obtained from
DLS measurements.
In order to transfer the prepared microgels to the aqueous
phase, the freeze-dried samples were allowed to swell in water
(solid content 5.0% (w/v)) adjusted to pH 12. Here, complete
deprotonation of the carboxylic acid groups was achieved as
could be monitored by the decrease of the pH to about pH 8 for
all samples. After titration with diluted hydrochloric acid to pH 7,
the swollen microgels were washed 3 times with deionized water
to remove formed sodium chloride, unreacted monomers and
non-cross-linked free polymer chains and oligomers. Freeze-
drying of the purified dispersions yielded the microgels as slightly
yellow colored powders which can be stored in the dried state
and transferred to water by simple swelling. Resulting aqueous
dispersions are stable without any additional surfactant due to
sterical stabilization by dangling chains of the swollen outer layer
of the microgels.
The retention of the particulate structure for the cross-linked
particles dispersed in aqueous media of pH 8.5 was confirmed by
SEM analysis. Figure 5 shows representative pictures. Because of
the highly swollen state of the microgels prior to drying, the
particles can be identified as flattened spheres on the wafer. No
drastic morphological deviations from the samples dropcast from
cyclohexane dispersion were observed. This is assumed to be the
result of drying effects on the relatively hydrophobic silica
surface: whereas in regions of high particle concentration inter-
penetration of the outer particle layers is visible, single or
separated particles exhibit only slightly flattened morphologies
of marginally increased diameters. non-cross-linked reference
particles MG-10 and MG-20 completely dissolved in water and
SEM analysis only showed polymer films but no spherical
structures.
Figure 6. Investigations on the pH-dependent swelling behavior of
photodegradable p(HEMA-co-MAA) microgels: degrees of swelling in
dependency on pH and additional values for PBS as medium.
monomers and cross-linkers, this method allows the equal
distribution of all functionalities incorporated. In addition, the
preparation of microgels is not limited to water-soluble mono-
mers since stable miniemulsions can also be obtained for dis-
persed phases consisting of organic solvents immiscible with the
organic continuous phase.
Photodegradable p(HEMA-co-MAA) microgels were pre-
pared by the copolymerization of HEMA, MAA, and the re-
spective cross-linkers in inverse miniemulsion. In order to ensure
comparability between the individual gel particles regarding
their swelling and degradation behavior, 2.5 mol % of cross-
linkers were used in every case. For the MG-1A microgels, the
cross-linking molecule CL-A was readily soluble in the mono-
mer mixture, so no further solvent was used to form the dis-
persed phase prior to polymerization. Solubility restrictions of
CL-B in the same monomer composition hindered analogous
reaction conditions. Therefore, in order to form MG-2B micro-
gels, 20 w% of DMSO were added to ensure complete cross-
linker solubility. Analogously, MG-2A gel particles were prepared
using cross-linker CL-A. In addition, particles containing no
cross-linking molecules were prepared as reference particles for
the swelling and degradation experiments. P(HEMA-co-MAA)
MG-10 nanoparticles were polymerized analogously to MG-1A
gel particles without any additional solvent for the dispersed
phase. In contrast, MG-20 particles, polymerized using 20 w%
DMSO as solvent, represent the reference particles to the MG-
2A and MG-2B microgels. To remove excess surfactant, all
Investigations on the pH-Dependent Swelling Behavior
of the Microgels. The prepared microgels are designed to exhibit
a pH-dependent swelling behavior due to the anionic network-
forming p(HEMA-co-MMA) copolymer. Deprotonation of the
carboxylic acid groups increases the hydrophilicity of the network,
9765
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
Table 2. Results from Titration Experiments on Photo-
Degradable p(HEMA-co-MAA) Microgel Dispersionsa
sample
Q / 10‑3 mol/g
IE (MAA)/%
pKa
MG-1A
MG-2A
MG-2B
1.26
1.17
1.16
81
77
75
6.30
6.28
6.27
a Obtained values for the content of ionizable groups Q, the incorpora-
tion efficiency IE of MAA and the pKa values.
thus inducing an enlargement of the degree of swelling of the
particles. This effect is even enhanced by electrostatic repulsion of
the generated anionic groups at higher pH values.
With the aim to investigate the described behavior, the pH
values of microgel dispersions in water were varied from pH 4 to
pH 12 by adding an aqueous NaOH solution, and hydro-
dynamic diameters (dh) of the particles were measured by
DLS for every pH value. The resulting graphs of plotting dh
against pH are shown in Figure S2 in the Supporting Informa-
tion. Figure 6 shows the resulting degrees of swelling calculated
as DGS = Vswollen/Vnonswollen, where Vnonswollen represents the
values obtained for the particle dispersions in cyclohexane.
As expected, all microgels exhibited pH-dependent swelling
properties with critical swelling transitions from the collapsed to
the swollen state at pH values between 6.0 and 8.5. A detailed
examination and discussion can be found in the Supporting
Information. Comparing the actual relative volumes of the
different particles, it is noteworthy that all microgels dispersed
in phosphate buffer saline (PBS) with pH 7.4 showed a lower
swelling degree compared to the identical particles in water of the
same pH (see Figure 6). Cationic electrolytes present in the
buffer solution shield the generated anionic groups in the
network, thus resulting in a reduced electrostatic repulsion.
Furthermore, it becomes obvious that the DGS values of MG-
2A and MG-2B enlarged with increasing pH to a similar steady
value of DGSmax = 11 at pH 8.5. In contrast, at the same pH, MG-
1A is swollen to a much greater extend, as characterized by a
higher maximum degree of swelling of DGSmax = 17. As MG-1A
and MG-2A only differ in their preparation, namely their initial
composition of the dispersed phase before the polymerization, it
is assumed that this parameter either influences the amount of
included cross-linker or the incorporation ratio of HEMA/MAA
in the final network.
UVꢀvis measurements of microgel dispersions in DMSO
were conducted to determine the amount of incorporated cross-
linking molecules (see Supporting Information: Figure S3 and
Figure S4). The relative quantity of incorporated cross-linker was
determined with respect to HEMA and MAA amounts before
polymerization as mol % of theory. Whereas in the case of MG-
2A and MG-2B 73 and 82 mol % of CL-A and CL-B were
incorporated into the respective microgels, for MG-1A a value of
117 mol % integration of CL-A was determined. Considering the
pH-dependent swelling profile of the different microgels, an
increased DGSmax of MG-1A in combination with a higher
amount of incorporated cross-linker (>100 mol % of theory) is
assumed to base on an enhanced incorporation of MAA into the
network. Since the similar cross-linker incorporation efficiencies
into the microgels, polymerized using additional DMSO as
solvent (MG-2A and MG-2B), were found to vary significantly
from the value for MG-1A, it is assumed that the composition of
the dispersed phase during polymerization influences both
Figure 7. Investigations on the light-induced particle degradation at pH
8.5 by turbidity measurements for different light intensities and wave-
lengths for (a) MG-1A, (b) MG-2A, and (c) MG-2B.
parameters mentioned above: the cross-linker incorporation
and the HEMA/MAA ratio in the network forming polymer.
In general, nonquantitative conversions of the polymerization
reactions yielded sol contents consisting of unreacted mono-
mers, cross-linkers and non-cross-linked oligomer or polymer
chains. Removal of the sol during gel purification therefore
resulted in cross-linker incorporation efficiencies depending on
the composition of the respective sol content and differing from
the theoretical amount.
In order to investigate the proposed enhanced incorporation
of MAA into MG-1A microgels, potentiometric titration experi-
ments were conducted and yielded the contents of ionizable
groups Q and the pKa values of the microgels (see Supporting
Information, Figure S5).59 From the calculated values for Q, the
MAA incorporation efficiency IE (MAA) was determined relative
to the theoretical amount assuming 100% incorporation of
monomers and cross-linkers into the latex particles. The ob-
tained values are listed in Table 2.
It was demonstrated that for all samples the amount of
integrated MAA groups in the microgels differs from the theo-
retical amount of 100%. Keeping in mind that the final composi-
tion of the polymeric gels is influenced by the conversion of the
polymerization, this effect can be explained by removal of the sol
content during the purification procedure which resulted in the
observed deviations from the theoretical values. Nevertheless,
MG-1A with Q = 1.26 ꢁ 10ꢀ3 mol/g clearly contains a higher
percentage of acid groups than microgels MG-2A and MG-2B
with Q = 1.17 ꢁ 10ꢀ3 mol/g and 1.16 ꢁ 10ꢀ3 mol/g, thus result-
ing in a higher degree of swelling despite an also enhanced
9766
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
incorporation of cross-linking points. The observed pKa values
are in all cases in good agreement with the observed swelling
profiles.
Photo-Degradation Experiments of p(HEMA-co-MAA) Mi-
crogels in Aqueous Media. Photodegradation of the different
microgels in water was examined by irradiating 0.125% (w/v)
dispersions in various aqueous media with UV light. The light-
induced swelling/degradation was monitored by turbidity of the
samples. Analogously to irradiation studies of photodegradable
PMMA particles, turbidity was measured as the relative scattering
intensity at 90°.50 Here, the cleavage of cross-linking points leads
to a looser network structure, thus decreasing the contrast in the
refractive indices between solvent and particle.60,61
In a first attempt, dispersions of highly swollen microgels at pH
8.5 were irradiated with UV light of varying wavelengths and
intensities. Figure 7 shows the resulting turbidity curves.
As can be seen in Figure 7a, irradiation of MG-1A microgels
with UV light of the wavelengths of 315ꢀ390 nm and an
intensity of 17 mW/cm2 does only result in a decay of the
turbidity down to a constant level of about 21%, therefore
indicating only increased particle swelling due to incomplete
degradation. In contrast, MG-2A and MG-2B both exhibit a
decrease in scattering intensity to almost 0% of the initial value
being a sign of complete particle disintegration. Furthermore,
using a discrete irradiation wavelength of 365 nm and increasing
the UV light intensity up to 30 and 60 mW/cm2 results in a faster
degradation for all microgels but does not influence the constant
turbidity level >0% of MG-1A. In general, this behavior allows
controlling the degradation rate via the light intensities used.
The differences of the degradation profiles of MG-1A and
MG-2A microgels, both containing cross-linker CL-A, are based
on their respective polymerization conditions. Whereas in the
case of MG-1A no additional solvent was used during polymer-
ization, MG-2A microgels were prepared by adding DMSO to
the dispersed phase. As discussed above, this parameter influ-
ences the composition of the final gels in terms of amount of
MAA groups in the network-forming copolymer and incorpo-
rated cross-linker molecules. As a higher cross-linker content was
determined for MG-1A particles compared to MG-2A gels, the
observed retardation of light induced particle disintegration can
be explained. Nevertheless, increasing the irradiation light in-
tensity does not affect the constant turbidity level observed for
MG-1A latexes, therefore indicating the influence of another
parameter. Irradiation of CL-A cross-linking molecules is known
to form dimeric azoxybenzene side products. As shown for
PMMA-gel particles, those side products are bound to the
polymer backbone, therefore forming new cross-linking points
hindering complete particle disintegration.50 It is assumed that
this effect also occurs in MG-1A microgels but is suppressed in
the case of MG-2A. It is proposed that a higher degree of chain
entanglement in combination with a higher amount of incorpo-
rated cross-linker in MG-1A microgels retards the diffusion of
cleaved polymer chains in the gels, thus rendering the formation
of dimers by the encounter of two polymer bound chromophores
more likely. In this context, the use of additional DMSO as
solvent for the dispersed phase of MG-2A is assumed to influence
the resulting molecular weight of the gel forming polymers by
restraining the Trommsdorff effect due to a higher dilution of the
polymerization mixture. Even though reference particles MG-10
and MG-20 were completely soluble and disintegrated at pH 8.5,
the determination of the molecular weights of the polymers by
GPC analysis yielded values of Mn = 235 000 g/mol (PDI = 4.3)
Figure 8. Investigations on the light-induced (λ = 365 nm; I =
60 mW/cm2) particle degradation by turbidity measurements at differ-
ent pH values for (a) MG-1A, (b) MG-2A, and (c) MG-2B.
and Mn = 53 000 g/mol (PDI = 2.0) for MG-10 and MG-20,
respectively (see Figure S6 in the Supporting Information). The
about five times increased molecular weight of the polymers
formed without additional DSMO as well as the dramatically
increased polydispersity point toward the occurrence of a gela-
tion effect during the polymerization. It is concluded that
incomplete particle degradation of MG-1A originates in the
distinct photolysis behavior of CL-A in correlation with polymer
entanglement of the network-forming copolymer.
Regarding possible release applications for the double stimuli-
responsive microgels, a direct degradation of the collapsed
particles at pH 4.5 would enable a one-step release, whereas
the degradation of slightly swollen microgels in PBS would
combine a primary slow diffusion-based release with a subse-
quent irradiation-based degradation, thus representing a two-
step release profile. In order to examine the dependency of the
light-triggered particle disintegration on the pH, irradiation
experiments with UV light of λ = 365 nm and I = 60 mW/cm2
were conducted in different aqueous media and reactions were
monitored again by turbidity measurements. The obtained
curves are shown in Figure 8.
The irradiation of MG-1A microgels in aqueous media at
different pH resulted in every case in a decay of turbidity down to
a constant level higher than 0%. Comparing the results obtained
for pH 8.5 and 10, no differences in the turbidity curves were
observed, thus confirming that incomplete particle degradation is
a result of newly formed cross-links as side products of the
photolysis of CL-A. Physical cross-links between the formed
amine groups of cleaved CL-A and the carboxylic acid groups of
9767
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
the network-forming copolymer can be excluded due to the
absence of protonated amine moieties at this pH value. Measure-
ments in PBS at pH 7.4 and aqueous medium at pH 4.5 resulted
in an increase of the constant turbidity levels up to 32 and 60%
respectively. Here, analogously to the experiments in basic
media, irradiation of the microgels causes the cleavage of CL-A
molecules and the formation of new cross-links as side reactions
resulting in swollen gel particles rather than free polymer chains.
The final degree of swelling after irradiation correlates to the
constant turbidity level and is influenced by two parameters: (a)
the ratio of CL-A cleavage to dimeric side product formation and
(b) the pH-dependent swelling properties of the gels. An initial
low degree of swelling due to a lower pH favors in this context
the formation of new cross-links. A more confined space en-
hances the reaction probability of two polymer bound groups
thus hindering particle swelling due to newly formed dimeric
cross-links.
Experiments conducted on MG-2A microgels confirm the
conclusions derived above. While complete degradation was
observed for a high initial degree of swelling at pH 8.5, irradia-
tions in PBS or pH 4.5 resulted in incomplete particle disinte-
gration, characterized by final constant turbidity values of about
60%. At pH 8.5, in contrast to MG-1A latexes, reduced entangle-
ment of the shorter polymer chains enables the generated
photolysis products to diffuse apart sufficiently fast enough not
to form dimeric side products. By lowering the pH and thereby
reducing the initial degree of swelling, the spatial concentration
of the photolysis products in the less swollen gels is increased,
thus enabling the formation of new dimeric cross-links and
resulting in swelling of the particles or formation of macroscopic
aggregates rather than complete degradation.
Regarding MG-2B gel particles cross-linked with CL-B—
independent of the pH of the media—the turbidity of the
samples decreased upon irradiation in every case down to almost
0%, therefore indicating complete particle degradation. In con-
trast to MG-1A and MG-2A particles cross-linked with CL-A,
CL-B—used to cross-link MG-2B—was designed to be cleaved
upon irradiation without generating polymer bound chromo-
phores. As a result, photolytic generation of dimeric side
products prohibits the formation of new cross-links. The ex-
cellent and fast degradation behavior of MG-2B independent of
the pH of the aqueous phase renders those particles highly
interesting for possible release applications.
Investigations on the Double Stimuli-Responsive Beha-
vior of the p(HEMA-co-MAA) Hydrogel Nanoparticles. De-
pending on the specific properties of the active compound to be
delivered, the loading and release mechanisms due to an ex-
ternally induced volume transition can vary significantly. While
collapsing a microgel containing low molecular weight com-
pounds electrostatically adsorbed to the former swollen network
can induce its release,20 a similar volume transition can be used to
entrap high molecular weight compounds such as, e.g., proteins
in the gel.62 Here, the swollen state enables the diffusion of active
compounds with a hydrodynamic diameter smaller than the
mesh size of the swollen network in and out of the gel. Deswelling
of the network is characterized by an effective decrease of the
mesh size, therefore entrapping the active compound due to
hindered diffusion. Subsequent swelling results then in the
release of the captured compound.
Figure 9. Investigations on the double stimuli-responsive behavior of
photodegradable p(HEMA-co-MAA) microgels: pH-induced swelling
and light-triggered (partial) degradation in different media. Plots of DGS
in dependency on different stimuli and surrounding media for (a) MG-
1A, (b) MG-2A, and (c) MG-2B.
with subsequent photodegradation. Figure 9 demonstrates the
investigated responses in terms of the degree of swelling of the
different microgel particles. It becomes obvious that all microgel
particles can easily be transferred from the cyclohexane phase to
aqueous media by swelling freeze-dried samples at pH values
higher than pH 8. Lowering the pH to around 4.5 resulted in every
case in deswelling of the microgels, therefore giving rise to a
potential entrapment of active compounds in the network. Col-
lapsed particles can then follow either a direct degradation or a
Regarding the potential of the dual stimuli-sensitive photo-
degradable p(HEMA-co-MAA) microgels for release applications,
pH-dependentswelling/deswellingof the gels shouldbe combined
9768
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
Figure 10. Schematic illustration of the loading and release strategy for p(HEMA-co-MAA) microgels. (1ꢀ2) Loading of large cationic functional
compounds into the anionic gel: (i) entrapment by pH-induced deswelling. (3) Diffusion controlled release: (ii) reswelling in PBS. (4) Degradation
controlled release: (iii) irradiation in PBS.
two-step swelling/degradation profile. Whereas the direct case
represents the light-triggered degradation of the collapsed particles
in acidic medium, the two-step profile can be realized by slightly
swelling the particles in PBS which is then followed by (complete)
particle degradation upon irradiation.
medium is lowered in order to induce a deswelling of the gel
and thereby a decrease of the pore sizes of the network
(Figure 10(i)). As a result, for substances of a hydrodynamic
diameter larger than the pore sizes of the collapsed gel, their
diffusion from the network is hindered even though the electro-
static interactions are also weakened. This approach is assumed
to result in an entrapment of the respective compound (see
Figure 10(2)). In this case, leakage of the payload from the
collapsed microgels at low pH values is prevented.
Regarding the release of the embedded compounds, a two-
step release profile based on the demonstrated two-step swelling
and degradation profile (see Figure 10) is proposed. First, by
increasing the pH to a value at which both the functional
compound and the carboxylic acid groups in the network are
negatively charged (Figure 10(ii)), the reduced electrostatic
interactions between embedded substance and network in
combination with the increased pore sizes of the swollen gel
induce the release (see Figure 10(3)). Second, this first pH-
dependent diffusion controlled release profile can be combined
with either a subsequent or substitutional degradation-induced
release upon irradiation (see Figure 10, parts iii and 4).
In order to examine the proposed method, investigations on
the loading of MG-2B microgels with myoglobin and its sub-
sequent release were carried out. Myoglobin was chosen as a
model protein since it fulfills certain important criteria for the
successful loading. The isoelectric point of myoglobin of Ip = 7.0
was shown to enable the successful loading into anionic micro-
gels containing methacrylic acid groups at pH 6.75.24 At this pH
value it was demonstrated that the MG-2B microgels are highly
swollen due to the presence of anionic charges in the network-
forming copolymer. In addition, this particular pH value lower
than the isolectric point of the protein renders the latter
positively charged, thus enabling an efficient loading due to
electrostatic interactions with the negatively charged gel. More-
over, the relatively large hydrodynamic diameter of around
4.4 nm of the globular protein (molecular weight of approxi-
mately 17 kDa) is assumed to facilitate the entrapment in the
microgels upon deswelling. Another beneficial characteristic is
It was demonstrated that in the case of microgel particles MG-
1A and MG-2A cross-linked with CL-A, the direct photolytic
degradation of the collapsed particles in pH 4.5 is hindered by the
formation of new cross-links and results only in slightly swollen
particles or macroscopic coagulates. Nevertheless, swelling of
these nonirradiated microgels in PBS is characterized by an initial
small increase in the degree of swelling up to a DGS of about 3.
Subsequent irradiation leads then to dramatic increases in the
particle volumes to a DGS of about 8. This pronounced dif-
ference is assigned to increased mesh sizes, thus increasing the
diffusion coefficients of potential embedded compounds in the
swollen state. In contrast, due to the specific designed photolytic
behavior of CL-B cross-linking molecules, MG-2B microgels
can be fully degraded upon UV light irradiation either in acidic
aqueous phase or in PBS.
Investigations on the Loading and Release of Myoglobin.
Having demonstrated the successful combination of a well-
defined pH-dependent swelling profile with the photolytic
degradability of p(HEMA-co-MAA) microgels, their potential
utilization for the loading and release of functional compounds
is of high interest. A proposed loading and release mechanism
of anionic p(HEMA-co-MAA) microgels is based on the fol-
lowing considerations which are schematically illustrated in
Figure 10.
The gel particles are highly swollen in aqueous medium of pH
values higher than pH 6.0 and consist of negatively charged
networks of large pore sizes. Immersing the microgels in a
solution of a substance with a hydrodynamic diameter smaller
than the mesh sizes of the gel and positively charged at the
respective pH, enables the diffusion of the functional compound
into the network. Loading due to electrostatic interactions
between the different ionic groups can therefore be achieved
(see Figure 10(1)). In a next step the pH of the surrounding
9769
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
Following a procedure described by Eichenbaum et al.,25
loading of anionic p(HEMA-co-MAA) MG-2B microgels with
myoglobin was performed by immersing the gel particles over-
night in a solution of the protein at pH 6.75. After equilibration,
the pH of the dispersion was adjusted to pH 4.5 in order to
collapse the microgels and entrap the embedded protein. After-
ward, the microgel particles were separated by centrifugation,
the supernatant was removed and the particles were redispersed
in water with pH 4.5. After repeatedly washing the loaded
particles, the microgels were finally redispersed in PBS. The
loading process was monitored by UVꢀvis spectroscopy (see
Figure 11a). Comparing the UVꢀvis spectrum of a myoglobin
reference solution (Mb-I) (of the same concentration as used for
the loading experiments) at pH 4.5 to the spectrum of the
supernatant (Mb-II) (after removal of the loaded and collapsed
microgels by centrifugation), it can be seen that the intensity of
the absorbance at λmax = 410 nm decreased significantly. The
observed difference can be attributed to a removed amount of
myoglobin from the solution therefore indicating successful
loading of the protein into the microgels by entrapment. The
subsequent washing steps were conducted to remove excess
myoglobin. By adjusting the pH value of the surrounding
medium to pH 4.5, carboxylic acid groups are protonated and
the electrostatic interaction with the positively charged myoglo-
bin is weakened. While protein molecules in the microgel interior
are entrapped due to the decreased pore sizes of the gel network,
loosely adsorbed myoglobin molecules on the surface of the
particles are assumed to be liberated. Therefore, multiple wash-
ing steps with water of pH 4.5 were conducted to remove
physically adsorbed proteins. The UVꢀvis spectra of the super-
natants of the performed washing steps did not show a significant
absorbance in the wavelength region of λmax of myoglobin after
the second washing cycle. Hence, the absence of myoglobin in
the washing phases points toward an efficient entrapment
accompanied by the successful suppression of leakage at the
respective low pH value of 4.5. After the last centrifugation step,
the microgels were redispersed in PBS.
Calculation of loading- (LE) and embedding efficiency (EE)
(see Supporting Information) afforded the values of LE = 53.2 (
7.4% and EE = 6.3 ( 0.9%. Regarding the high loading efficiency
of 53.2% it can be concluded that the investigated approach of
loading myoglobin upon entrapment of the protein in collapsed
MG-2B microgels is a very efficient process. The relatively low
embedding efficiency of 6.9% can be assigned to a huge excess of
myoglobin with respect to the amount of microgels. Never-
theless, the examined loading strategy allows the recycling of not
incorporated myoglobin since the microgels were easily removed
from the solution by centrifugation.
Having demonstrated the efficient loading of myoglobin in
MG-2B microgels, in a next step the subsequent diffusion
controlled release of the protein was investigated by transferring
the loaded microgels from water of pH 4.5 to PBS of pH 7.4. The
resulting microgel dispersion was shaken at 37 °C and samples
were taken after predetermined time intervals. After removal of
the microgels by centrifugation of each sample, the released
amount of myoglobin in the supernatant was determined by com-
paring the UV absorbance at λmax = 410 nm to a calibration curve
of myoglobin in PBS (see Supporting Information, Figure S8).
Here, performed control experiments ensured that centrifugation
did not result in sedimentation of the pure protein (see Support-
ing Information, Figure S9).
Figure 11. Investigations on the loading of myoglobin and its subse-
quent release from light-sensitive p(HEMA-co-MAA) microgels: (a)
UVꢀvis spectroscopic examination of the loading process of MG-2B
microgels with myoglobin; (b) time-dependent UVꢀvis spectra of the
release of myoglobin from MG-2B microgels in PBS at 37 °C; (c)
cumulative release dependency on the incubation and irradiation time.
The dotted line is a guide to the eyes.
the UV absorption maximum at λmax = 410 nm, which permits
both the determination of loading efficiencies and the monitor-
ing of the release by UVꢀvis spectroscopy. Furthermore, the
comparably low absorption of the protein in the spectral region
used for the irradiation is beneficial (see Figure S7 in the
Supporting Information).
9770
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
In order to investigate the photodegradation-induced release
of the protein, after 4 and 24 h of incubation, samples were taken
and irradiated (1.5 h, λ = 315ꢀ390 nm). To ensure complete
particle disintegration, irradiations were carried out until the
relative turbidity of the samples dropped below 5% of their initial
values. The release of the protein was then calculated from the
absorption spectra of the resulting clear solutions. Here, it is
noteworthy that the UV absorbance of the photoproducts of the
cleavable cross-linker CL-B can have a potential influence on the
spectroscopic quantification of the amount of free protein.
Therefore, the respective spectra were calculated by subtracting
a baseline of irradiated “empty” microgels in PBS (of the same
concentration). Furthermore, it is known that irradiation of oxy-
myoglobin (MbO2) leads to photodissociation to (Mb) and
photo-oxidation to metmyoglobin (MMb).63ꢀ65 Since these
compounds exhibit different absorption spectra, irradiation-
induced changes in the absorbance of the protein have to be
considered in order to enable an accurate quantification of
the released protein. For this, a control sample of pure myoglobin
(of the same concentration as in the loaded microgel dispersion)
in PBS was irradiated and the change in absorbance at 410 nm was
calculated relative to the nonirradiated sample. The amount of
released myoglobin was then obtained by taking this parameter
into account. Figure 11b shows the measured time-dependent
UVꢀvis spectra of the diffusion controlled release experiment in
PBS and Figure 11c depicts the cumulative release of the protein
in dependency on the incubation and irradiation time.
The increasing absorbance at λmax = 410 nm in the time-
dependent UVꢀvis spectra clearly demonstrates an increase of
the concentration of myoglobin in the supernatants after cen-
trifugation. By transferring the loaded, collapsed particles from
pH 4.5 to PBS of pH 7.4, the loaded myoglobin is assumed to be
released due to a combination of different factors. First, the
particles exhibit an increased degree of swelling in the medium of
increased pH, therefore consisting of larger pore sizes which
enable a diffusion of the protein from the network. Second, even
though the carboxylic acid groups in the network forming
copolymer are assumed to be protonated at pH 7.4, this pH
value is higher than the isoelectric point of myoglobin (Ip = 7.0).
Thus, the protein is also mainly negatively charged. The resulting
electrostatic repulsion between the anionic groups in the net-
work and on the protein results in the release from the gel. Third,
the high ionic strength of the buffer solution induces a shielding
of anionic groups in the network. Moreover, the sodium coun-
terions are assumed to displace electrostatic interactions between
remaining positive charges of the protein and anionic carboxy-
late groups of the gel network therefore minimizing attractive
physical interactions.
the concentration of myoglobin in the local environment of the
microgels.
Regarding the light-induced release of myoglobin by photo-
degradation of the microgel networks, it can be seen from the
graph in Figure 11c that this method enables a quantitative
release at any desired time point. In comparison to the
diffusion controlled profile, it was shown that the amount of
released protein can hereby be increased by around 65% at
5.5 h and by 40% at 25.5 h of incubation thus enabling a very
efficient light-induced release. In summary, it can be stated
that a novel two-step release profile was realized by combining
the pH-dependent swelling and the light-induced degradation
of these materials.
’ CONCLUSION
Novel photodegradable p(HEMA-co-MAA) microgels were
prepared by free radical inverse miniemulsion copolymerization
of HEMA and MAA with two newly synthesized photolabile
cross-linking molecules. The pH-dependent swelling behavior
induced by the protonation/deprotonation of the methacrylic
acid groups in the network-forming polymer, was found to
exhibit large differences of the particle volumes in acidic and
basic media. UV light irradiation led to particle disintegration
caused by cleavage of the photolabile cross-linking points. The
particular combination of the stimuli described is designed as a
new strategy to two different swelling/degradation profiles. By
exploiting these mechanisms, p(HEMA-co-MAA) MG-2B mi-
crogels were successfully loaded with myoglobin as a model
protein by electrostatic interactions of the positively charged
myoglobin molecules with the negatively charged polymeric gel
in combination with large pore sizes of the swollen network.
Subsequent entrapment of the compound by pH-induced des-
welling of the particles resulted in a high loading efficiency, thus
rendering this newly developed approach highly effective. The
observed diffusion controlled release profile upon swelling the
particles under physiological conditions followed an initial burst
release in combination with a slow release over a prolonged
period of time.
In addition, it was demonstrated that a subsequent fast and
quantitative on-demand release could be realized by the applica-
tion of UV light thereby representing a novel two-step release
profile.
While the observed purely diffusion controlled release profile
is very promising for a fast provision of a medium concentration
of the compound in combination with a subsequent steady
supply for a longer period of time, especially the additional
ability to induce a fast and quantitative on-demand release upon
the application of UV light in a noncontact approach, renders this
new concept very interesting for delivery applications, e.g., in
materials science or biomedical fields.
On the basis of the UVꢀvis spectroscopic investigations, the
cumulative release was calculated. The observed diffusion con-
trolled release profile (see Figure 11c) was found to follow an
initial burst release of around 40% of myoglobin during the first
3 h of incubation in combination with a slow additional release of
about 20% of embedded protein over a 22.5 h time span. Since
the maximum release after 25.5 h of incubation was determined
to be 62% of the loaded amount, it can be stated that a quanti-
tative release was not achieved by the purely diffusion controlled
mechanism in this time span. It is assumed that this retardation
behavior is based on the equilibration between the release of
myoglobin and the repenetration/diffusion of myoglobin into
the highly swollen networks. Since the surrounding medium was
not changed, it is proposed that the release profile depends on
’ ASSOCIATED CONTENT
S
Supporting Information. UVꢀvis investigations on the
b
photolysis of cleavable cross-linkers in solution, pH-dependent
DLS measurements and titration curves of microgels, UVꢀvis
spectroscopic determination of cross-linker incorporation, GPC
measurements, and UVꢀvis spectroscopic investigations for the
loading and release experiments. This material is available free of
charge via the Internet at http://pubs.acs.org/.
9771
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772

Macromolecules
ARTICLE
’ AUTHOR INFORMATION
(30) Jhaveri, S. B.; Carter, K. R. Macromolecules 2007, 40, 7874–7877.
(31) Goh, S. L.; Murthy, N.; Xu, M. C.; Frechet, J. M. J. Bioconjugate
Chem. 2004, 15 (3), 467–474.
Corresponding Author
*E-mail: landfester@mpip-mainz.mpg.de.
(32) Bulmus, V.; Chan, Y.; Nguyen, Q.; Tran, H. L. Macromol. Biosci.
2007, 7 (4), 446–455.
(33) Murthy, N.; Xu, M. C.; Schuck, S.; Kunisawa, J.; Shastri,
N.; Frechet, J. M. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100 (9), 4995–5000.
(34) Standley, S. M.; Kwon, Y. J.; Murthy, N.; Kunisawa, J.; Shastri,
N.; Guillaudeu, S. J.; Lau, L.; Frechet, J. M. J. Bioconjugate Chem. 2004,
15 (6), 1281–1288.
’ ACKNOWLEDGMENT
D.K. acknowledges the International Max Planck Research
School (IMPRS) for financial support. We thank Sebastian
Goerres for conducting titration experiments.
(35) Reichmanis, E.; Smith, B. C.; Gooden, R. J. Polym. Sci., Part A:
Polym. Chem. 1985, 23 (1), 1–8.
’ REFERENCES
(1) Zha, L.; Banik, B.; Alexis, F. Soft Matter 2011, 7 (13), 5908–
5916.
(36) Holmes, C. P. J. Org. Chem. 1997, 62 (8), 2370–2380.
(37) del Campo, A.; Boos, D.; Spiess, H. W.; Jonas, U. Angew. Chem.,
Int. Ed. 2005, 44 (30), 4707–4712.
(38) Wong, W. K.; Schupp, H.; Schnabel, W. Macromolecules 1989,
22 (5), 2176–2181.
(2) Karg, M.; Pastoriza-Santos, I.; Perez-Juste, J.; Hellweg, T.; Liz-
Marzan, L. M. Small 2007, 3, 1222–1229.
(3) Sorrell, C. D.; Carter, M. C. D.; Serpe, M. J. Adv. Funct. Mater.
2011, 21 (3), 425–433.
(39) Barzynski, H.; Sanger, D. Angew. Makromol. Chem. 1981, 93
(FEB), 131–141.
(4) Debord, J. D.; Lyon, L. A. J. Phys. Chem. B 2000, 104 (27),
6327–6331.
(5) Kawaguchi, H.; Fujimoto, K. Bioseparation 1998, 7 (4ꢀ5),
253–258.
(6) Welsch, N.; Ballauff, M.; Lu, Y., In Chemical Design of Responsive
Microgels; Pich, A. R. W., Ed.; Springer: Berlin, 2010; Vol. 234, pp 129ꢀ163.
(7) Ballauff, M.; Lu, Y. Polymer 2007, 48 (7), 1815–1823.
(8) Lu, Y.; Mei, Y.; Drechsler, M.; Ballauff, M. Angew. Chem., Int. Ed.
2006, 45 (5), 813–816.
(40) Katz, J. S.; Doh, J.; Irvine, D. J. Langmuir 2006, 22 (1), 353–359.
(41) Piggott, A. M.; Karuso, P. Tetrahedron Lett. 2005, 46 (47),
8241–8244.
(42) Kloxin, A. M.; Kasko, A. M.; Salinas, C. N.; Anseth, K. S. Science
2009, 324 (5923), 59–63.
(43) Dvir, T.; Banghart, M. R.; Timko, B. P.; Langer, R.; Kohane,
D. S. Nano Lett. 2010, 10 (1), 250–254.
(44) Babin, J.; Pelletier, M.; Lepage, M.; Allard, J. F.; Morris, D.;
Zhao, Y. Angew. Chem., Int. Ed. 2009, 48 (18), 3329–3332.
(45) Jiang, J. Q.; Tong, X.; Morris, D.; Zhao, Y. Macromolecules 2006,
39 (13), 4633–4640.
(9) Hamidi, M.; Azadi, A.; Rafiei, P. Adv. Drug Delivery Rev. 2008,
60 (15), 1638–1649.
(10) Vinogradov, S. V. Curr. Pharm. Des. 2006, 12 (36), 4703–4712.
(11) Pich, A.; Richtering, W. In Chemical Design of Responsive Microgels;
Pich, A., Richtering, W., Eds.; Springer: Berlin, 2010; Vol. 234, pp 1ꢀ37.
(12) Malmsten, M.; Bysell, H.; Hansson, P. Curr. Opin. Colloid
Interface Sci. 2010, 15 (6), 435–444.
(13) Peppas, N. A.; Huang, Y.; Torres-Lugo, M.; Ward, J. H.; Zhang,
J. Annu. Rev. Biomed. Eng. 2000, 2, 9–29.
(14) Das, M.; Zhang, H.; Kumacheva, E. Annu. Rev. Mater. Res. 2006,
36, 117–142.
(46) Jiang, J. Q.; Tong, X.; Zhao, Y. J. Am. Chem. Soc. 2005, 127 (23),
8290–8291.
(47) Zhao, Y. Chem. Rec. 2007, 7, 286–294.
(48) Li, Y.; Xinru, J.; Min, G.; Kuang, G.; Wei, Y. J. Polym. Sci., Part A:
Polym. Chem. 2010, 48, 551–557.
(49) Burakowska, E.; Zimmerman, S. C.; Haag, R. Small 2009, 5 (19),
2199–2204.
(50) Klinger, D.; Landfester, K. Soft Matter 2011, 7 (4), 1426–1440.
(51) Holmes, C. P.; Jones, D. G. J. Org. Chem. 1995, 60 (8),
2318–2319.
(52) Buhler, S.; Lagoja, I.; Giegrich, H.; Stengele, K. P.; Pfleiderer, W.
Helv. Chim. Acta 2004, 87 (3), 620–659.
(15) Lin, C.-C.; Metters, A. T. Adv. Drug Delivery Rev. 2006,
58 (12ꢀ13), 1379–1408.
(16) Nayak, S.; Lee, H.; Chmielewski, J.; Lyon, L. A. J. Am. Chem. Soc.
2004, 126 (33), 10258–10259.
(53) Braun, D.; Berger, J. Kolloid Z. Z. Polym. 1972, 250 (2),
142–147.
(17) Nayak, S.; Lyon, L. A. Angew. Chem., Int. Ed. 2005, 44 (47),
7686–7708.
(18) Wells, L. A.; Sheardown, H. Eur. J. Pharm. Biopharm. 2007,
65 (3), 329–335.
(19) Dou, H.; Jiang, M. Polym. Int. 2007, 56 (10), 1206–1212.
(20) Das, M.; Mardyani, S.; Chan, W. C. W.; Kumacheva, E. Adv.
Mater. 2006, 18 (1), 80–83.
(54) Conforti, P. F.; Garrison, B. J. Chem. Phys. Lett. 2005, 406 (4ꢀ6),
294–299.
(55) Nayak, S.; Lyon, L. A. Angew. Chem., Int. Ed. 2005, 44 (47),
7686–7708.
(56) Landfester, K.; Musyanovych, A. In Chemical Design of Respon-
sive Microgels; Pich, A., Richtering, W., Eds.; Springer: Berlin, 2010; Vol.
234, pp 39ꢀ63.
(21) Qiu, Y.; Park, K. Adv. Drug Delivery Rev. 2001, 53 (3), 321–339.
(22) Lin, C. C.; Metters, A. T. Adv. Drug Delivery Rev. 2006,
58 (12ꢀ13), 1379–1408.
(57) Landfester, K. Angew. Chem., Int. Ed. 2009, 48 (25), 4488–4507.
(58) Landfester, K.; Willert, M.; Antonietti, M. Macromolecules
2000, 33 (7), 2370–2376.
(23) Murthy, N.; Thng, Y. X.; Schuck, S.; Xu, M. C.; Frechet, J. M. J.
J. Am. Chem. Soc. 2002, 124 (42), 12398–12399.
(24) Eichenbaum, G. M.; Kiser, P. F.; Dobrynin, A. V.; Simon, S. A.;
Needham, D. Macromolecules 1999, 32 (15), 4867–4878.
(25) Eichenbaum, G. M.; Kiser, P. F.; Shah, D.; Simon, S. A.;
Needham, D. Macromolecules 1999, 32 (26), 8996–9006.
(26) Sankaranarayanan, J.; Mahmoud, E. A.; Kim, G.; Morachis,
J. M.; Almutairi, A. ACS Nano 2010, 4 (10), 5930–5936.
(27) Franssen, O.; Stenekes, R. J. H.; Hennink, W. E. J. Controlled
Release 1999, 59 (2), 219–228.
(59) Kawaguchi, S.; Yekta, A.; Winnik, M. A. J. Colloid Interface Sci.
1995, 176 (2), 362–369.
(60) Al-Manasir, N.; Zhu, K. Z.; Kjoniksen, A. L.; Knudsen, K. D.;
Karlsson, G.; Nystrom, B. J. Phys. Chem. B 2009, 113 (32), 11115–11123.
(61) Lecner, M. D. J. Serb. Chem. Soc. 2005, 70 (3), 361–369.
(62) Zhang, Y.; Zhu, W.; Wang, B. B.; Ding, J. D. J. Controlled Release
2005, 105 (3), 260–268.
(63) Ma, J. Y.; Zheng, X. F.; Tang, Q.; Ma, J.; Gao, D.; H., H. J. Sci.
China, Ser. B 2008, 51 (5), 414–419.
(64) Hutchinson, J. A.; Noe, L. J. J. Quantum Electron. 1984, 20 (12),
1353–1362.
(28) De Geest, B. G.; De Koker, S.; Demeester, J.; De Smedt, S. C.;
Hennink, W. E. Polym. Chem. 2010, 1 (2), 137–148.
(29) Chong, S.-F.; Chandrawati, R.; Staedler, B.; Park, J.; Cho, J.;
Wang, Y.; Jia, Z.; Bulmus, V.; Davis, T. P.; Zelikin, A. N.; Caruso, F. Small
2009, 5 (22), 2601–2610.
(65) Bertelsen, G.; Skibsted, L. H. Meat Sci. 1987, 19, 243–251.
9772
dx.doi.org/10.1021/ma201706r |Macromolecules 2011, 44, 9758–9772