Thermoresponsive Graft Copolymers of ‑Isopropylacrylamide and Hyperbranched Polyglycerol as Thermally Induced Drug Delivery and Release Nanoformulation Systems for Curcumin with High Colloidal Stability and Enhanced Anticancer Effect.

Introduction
Polymer-based delivery systems are undoubtedly
one of the most
extensively investigated classes of materials for drug formulations.
Considering their ability to adjust the concentration of the drug
in the body by increasing its solubility, controlling the rate of
drug release, and specifically targeting the site of action, they
can positively influence the bioactivity, bioavailability, biodistribution,
and stability of the drugs. Over the years, various polymer-based
drug delivery systems have been investigated, including well-defined
functionalized polymers for conjugation with drugs and targeting moieties
−

and especially amphiphilic polymers that exhibit self-assembling
properties to form nanocarriers (e.g., micelles,
−

nanoparticles,
−

polymersomes,
,
etc.) for drug encapsulation in a noncovalent manner. Among such macromolecular assemblies, responsive
(smart, intelligent, adaptive) polymers have gained notable interest
in drug formulation due to the abrupt changes in their physical or
chemical properties caused by external stimuli, which can trigger
drug uptake and release.
−

These include polymers responding to changes in the
temperature, light, electromagnetic fields, pH, or ultrasound. Nevertheless,
thermoresponsive polymers have been the most extensively investigated
in this field due to the ease of stimuli by heating or cooling.
−

The thermoresponsive nature of polymer solutions is characterized
by LCST or UCST phenomena, corresponding to the temperatures at which
the critical solution temperature (CST) reaches its minimum or maximum,
respectively, in the polymer/solvent phase diagram.
,
Several (co)­polymers exhibit thermoresponsive behavior,
,
such as poly­(2-alkyl-2-oxazoline)­s,
,−

various poly­(meth)­acrylamides,
−

oligo- and poly­(ethylene glycol)-(meth)­acrylate-based
polymers,
−

poly­(methyl vinyl ether), poly­(N,N-dimethylaminoethyl methacrylate),
−

and so on. However, poly­(N-isopropylacrylamide)
(PNiPAAm) is still the most widely studied thermoresponsive polymer.
−

The CST of an aqueous solution of PNiPAAm is close to that of the
human body, at approximately 32–34 °C, allowing its use
in biomedical applications, particularly in drug delivery. At and
above the CST, PNiPAAm chains undergo a reversible coil-to-globule
(hydrated-dehydrated) transition and simultaneous aggregation to form
colloidal suspensions or macroscopic precipitates.
,
To enhance the colloidal stability and to tune the thermoresponsive
and drug delivery properties, PNiPAAm-based copolymers, including
block,
−

random,
,
and graft copolymers,
−
,
have been intensively investigated.
Poly­(ethylene
glycol)­s (PEGs) are biocompatible materials with
prolonged blood circulation time, and therefore, copolymerization
of PNiPAAm with PEGs is commonly implemented to adjust the CST.
,
Above the CST, PEG-containing copolymers demonstrate enhanced colloidal
stability.
,
Comparatively, PNiPAAm copolymers
with PEG side chains show a more pronounced effect on CST than those
with PEG incorporated as a block.
,,
Although PEG is one of the most extensively investigated
synthetic polymers for bioapplications, it has several known disadvantages,
−

such as possible accumulation in the body, hypersensitivity, formation
of anti-PEG antibodies, accelerated blood clearance, and nonbiodegradability,
indicating an increasing need for PEG alternatives with more advantageous
properties than that of PEG.
According to our concept, hyperbranched
polyglycerol (HbPG), which
possesses numerous advantageous properties, including biocompatibility,
excellent water solubility, and high colloidal stability,
−

can be applied as a PEG alternative to obtain hydrophilic side-chain-grafted
thermoresponsive copolymers. Additionally, the biomedical application
possibilities of HbPG are being widely investigated in several directions,
such as drug delivery, inhibition of viral infections, and antifouling
coatings.
,
Recently, thermoresponsive acrylated HbPG-functionalized
β-cyclodextrin was copolymerized with NiPAAm leading to gels
with enhanced topical drug delivery, indicating
the advantage of the presence of HbPG in such biomedically viable
macromolecular constructs. A variety of hypergrafted copolymers of
HbPG have also been studied, in which the HbPG was grafted onto linear
polymer backbones
−

or thermoresponsive polymers were grafted onto the HbPG core
−

to obtain copolymers with beneficial characteristics. Biocompatible
HbPG-grafted poly­(N,N-diethylacrylamide)
(PDEAAm) thermoresponsive copolymers, indicating the advantages of
HbPG grafts in such macromolecular assemblies, have been recently
reported by us.

This study deals
with PNiPAAm-g-HbPG graft copolymers
with composition-dependent thermoresponsive behavior as novel carriers
with high colloidal stability and in vitro susceptibility
for drug encapsulation and release of cargo with low water solubility.
The targeted PNiPAAm-g-HbPG-s were obtained by utilizing
a well-defined amine monofunctional HbPG for highly efficient postpolymerization grafting onto copolymers
of NiPAAm with N-acryloxysuccinimide (NAOS), as a
reactive comonomer, which has already been applied to obtain PNiPAAm-g-PEG brush copolymers. Subsequently,
thermally induced encapsulation and release, cell internalization,
and anticancer effect of curcumin, a natural antioxidant, anti-inflammatory,
antibacterial, and anticancer agent,
−

with the thermoresponsive PNiPAAm-g-HbPG graft copolymers as drug carrier nanoformulations were investigated
and are reported herein.

Experimental Section

Materials
Glycidol (from Sigma-Aldrich) was freshly
distilled under reduced pressure before use. N-isopropylacrylamide
(NiPAAm from Sigma-Aldrich) was purified by recrystallization in n-hexane. Azobis­(isobutyronitrile) (AIBN, Sigma-Aldrich)
was recrystallized in methanol two times before use. The following
reagents were utilized directly as supplied: acrylic acid N-hydroxysuccinimide ester (N-acryloxysuccinimide,
NAOS, >90%, Sigma-Aldrich), phthalimide (≥99%, Sigma-Aldrich),
potassium phthalimide (≥99%, Sigma-Aldrich), hydrazine monohydrate
(reagent grade, ≥97%, Sigma-Aldrich), phosphate-buffered saline
(PBS) tablets (Sigma-Aldrich), Tween 80 (Sigma-Aldrich), lithium bromide
(99.99%, Alfa Aesar), and curcumin (97%, Apollo Scientific). Solvents
including methanol (MeOH), ethanol (EtOH), diethyl ether, and n-hexane (all from Molar Chemicals Ltd.), as well as N,N-dimethylformamide (DMF, HPLC grade,
VWR Chemicals) were employed without further purification. Tetrahydrofuran
(THF, ≥99.5%, Molar Chemicals Ltd.) was refluxed over potassium
hydroxide and distilled prior to use as the polymerization medium
or applied directly for precipitation. For the in vitro assays, RPMI-1640 medium, PBS (pH = 7.4), fetal bovine serum (FBS), l-glutamine, trypsin, and penicillin–streptomycin were
purchased from Thermo Fisher Scientific (Gibco; Waltham, MA, USA).
AlamarBlue reagent was from Merck (Budapest, Hungary).

Synthesis Methods

Synthesis
of Thermoresponsive Poly­(N-isopropylacrylamide)
(PNiPAAm) Homopolymer and Poly­(N-isopropylacrylamide)-co-(N-acryloxysuccinimide) (P­(NiPAAm-co-NAOS)) Copolymers
Free radical polymerization
was employed to synthesize the PNiPAAm homopolymer as well as the
P­(NiPAAm-co-NAOS) copolymers, following the procedure
described recently by us. The initiator/monomer
molar ratio was 1:100 in all cases, and the NiPAAm/NAOS comonomer
ratios were 20:1, 10:1, and 5:1. As an example, the synthesis of P­(NiPAAm-co-NAOS) with a NiPAAm/NAOS ratio of 20:1 is described below.
NiPAAm (1.9002 g, 16.8 mmol) and NAOS (0.1420 g, 0.84 mmol) were dissolved
in 16 mL of distilled THF, and the resulting solution was heated to
60 °C under nitrogen. Subsequently, 2.76 mL of AIBN in a THF
stock solution (10 mg/mL) was introduced to the reaction mixture,
and it was stirred under a nitrogen atmosphere at 60 °C for 16
h. After cooling, the reaction mixture was poured in hexane, and the
resulting precipitate was isolated by centrifugation. The polymer
was then dried under vacuum at 60 °C until a constant mass was
obtained. The yields of the P­(NiPAAm-co-NAOS) copolymers
were >90% in all cases.

Synthesis of Amine-Monofunctional HbPG (NH2–HbPG)
The NH2–HbPG was
synthesized by a two-step method. Briefly,
phthalimide monofunctional HbPG (PhthIm-HbPG)
was produced by phthalimide/potassium phthalimide-initiated ring-opening
multibranching polymerization of glycidol. Phthalimide (1.4999 g,
10.2 mmol) and potassium phthalimide (0.2154 g, 1.2 mmol) were measured
into a three-necked round-bottomed flask equipped with a mechanical
stirrer and connected to a vacuum line, while the third neck was sealed
with a septum. The reaction flask was heated to 95 °C and was
purged with a stream of nitrogen. Subsequently, 9 mL (10.065 g, 136
mmol) of glycidol was added dropwise (flow rate: 2.5 mL/h) to the
reaction flask via a syringe pump. The reaction mixture was stirred
for an additional 18 h to ensure complete incorporation of the monomer.
The crude product was cooled and dissolved in ethanol, passed through
a column filled with a cation exchange resin, and precipitated in
diethyl ether. It was then dried under a vacuum at 60 °C (yield
= 89%). Subsequently, 1.45 mL of hydrazine monohydrate (1.5043 g,
30.1 mmol) was added to the PhthIm-HbPG (M
n = 1180 g/mol determined by 1H NMR, see Figure S1, 1.6527 g, 1.5 mmol) dissolved in 20 mL of EtOH
and the reaction mixture was stirred overnight at room temperature.
After filtration, the crude product was obtained by precipitation
in a large excess of a THF/diethyl ether mixture (2:1 V/V) followed by centrifugation. The precipitation
step was performed twice to remove residual impurities, after which
the amine-functional HbPG was dried under vacuum at 45 °C (yield:
75%; M
n = 1050 g/mol from 1H NMR; Figure S2). The gel permeation
chromatography (GPC) chromatograms and MWD curves of the phthalimide-
and amine-monofunctional HbPGs are presented in Figure S3.

Synthesis of PNiPAAm-g-HbPG
Copolymers by Grafting
of P­(NiPAAm-co-NAOS) Copolymers with the NH2–HbPG
The grafting of the HbPG onto the NAOS-containing
PNiPAAm copolymers was performed according to our recently presented
method. In a typical grafting procedure,
100 mg of P­(NiPAAm-co-NAOS) was dissolved in MeOH,
and a methanolic solution containing NH2–HbPG in
an equimolar amount relative to the NAOS units was added. After the
polymer concentration was adjusted to 10 mg/mL, the solution was stirred
at ambient temperature overnight. Thereafter, the products were purified
by dialysis (SpectraPor 6, MWCO 1 kDa) against distilled water for
3 days to remove NHS byproduct and unreacted HbPG and then lyophilized.

Characterizations

Gel Permeation Chromatography
GPC
analyses were performed
on a setup consisting of a Waters 515 pump and a JASCO AS-4150 autosampler,
with Styragel HR1, HR4, and HR6 columns connected in series and a
JASCO 4035 refractive-index detector. The eluent was DMF containing
LiBr of 1 g/L concentration with a flow rate of 0.3 mL/min at 50 °C.
The average molecular weights and dispersity values were obtained
by calibration using PEG standards (PSS, 194–217,000 Da).

NMR Spectroscopy

1H NMR analyses were performed
on a Varian Inova spectrometer at a 500 MHz resonance frequency. The
polymers were analyzed in deuterated dimethyl sulfoxide (DMSO-d
6) and deuterium oxide (D2O) at 30
°C.

Thermoresponsive Behavior Investigations
Turbidimetric
analyses were conducted to evaluate the thermoresponsive behavior
of the homopolymer and copolymer samples. Measurements were performed
on a JASCO V-650 spectrophotometer equipped with a temperature-control
system (JASCO MCB-100 circulation unit and a Peltier thermostat).
For the NAOS-containing copolymers, transmittance was recorded at
1 g/L in water, while the HbPG-grafted derivatives were examined in
both water and PBS at polymer concentrations of 1, 5, and 10 g/L.
All experiments were carried out at 488 nm wavelength over a temperature
range of 20–80 °C using a heating/cooling rate of 0.5
°C min–1, with a 1 min equilibration at each
temperature step. The cloud point (T
CP) and clearing point (T
CL) were determined
as the inflection points of the corresponding transmittance–temperature
curves during the heating and cooling cycles, respectively.
,

Encapsulation of Curcumin
A stock solution of curcumin
(0.5 mL, 20 g/L in acetone) was added dropwise to 1 mL of an aqueous
solution of PNiPAAm-g-HbPG with varying polymer concentrations
(from 0.1 to 10 mg/mL in distilled water under vigorous stirring at
25 and 37 °C). Subsequently, the stirring was continued for an
additional 4 h to permit the evaporation of the acetone. The polymer
concentration was maintained by the addition of water, if necessary.
The suspensions were filtered through a 0.45 μm syringe filter
to remove the insolubilized curcumin. Subsequently, 50 μL of
the filtered solution was taken and diluted with 0.95 mL of ethanol,
and the solution was analyzed by spectrophotometry (JASCO V-650) at
25 °C. The encapsulated drug contents were evaluated by a curcumin
calibration curve by the absorbance at 424 nm in absolute EtOH, i.e.,
at a wavelength having no absorbance for the PNiPAAm-g-HbPG graft copolymers (see Figures S14 and S15). All the measurements were performed in triplicate. The encapsulation
was expressed in the drug loading content (DLC),
which was determined by the following equation

Statistical analysis
was performed using a two-sample t-test, statistical
significance:
*p < 0.05, ***p < 0.001.
All data are presented as the mean ± SD, N =
3.

Colloidal Stability
The colloidal stability of the
PNiPAAm homopolymer, the PNiPAAm-g-HbPG graft copolymers,
and curcumin in both native and nanoformulated forms was analyzed
in PBS by using a JASCO V-650 spectrophotometer equipped with a JASCO
MCB-100 circulation unit and a Peltier-based temperature controller.
Polymer concentrations were set at 10 g/L for the homopolymer and
copolymers, while the drug-containing nanoformulations were prepared
at 5 g/L with the maximum curcumin loading. After 10 min of equilibration
at 37 °C under constant stirring at 600 rpm, stirring was stopped,
and the transmittance at 600 nm was monitored over 24 h with PBS serving
as the reference medium.

Dynamic Light Scattering Measurements
The hydrodynamic
diameter and the size distribution of the PNiPAAm-g-HbPG copolymers and the curcumin-loaded nanoformulations were investigated
by dynamic light scattering (DLS) using a Zetasizer Nano ZS (Malvern
Instruments, Malvern, UK) with an operating wavelength of 633 nm and
a measurement angle of 173°. Measurements of the pure graft copolymers
were performed at a polymer concentration of 1 g/L in water without
filtration. Before each measurement at varying temperatures, the samples
were equilibrated for 30 min. The curcumin-loaded nanoparticles were
prepared by the same method described in the previous section, using
a polymer concentration of 1 g/L in PBS and a 100 μM curcumin
concentration. Subsequently, the samples were filtered through a 0.45
μm membrane filter and subsequently measured at 37 °C after
30 min of equilibration time. The particle size distribution of the
samples was determined using statistical data analysis methods (cumulant
analysis and the CONTIN algorithm) of the resulting autocorrelation
data. The data presented in this paper represent the mean distributions
calculated from three independent replicates.

Drug Release
The release of curcumin from the PNiPAAm-g-HbPG
nanoformulations prepared at 25 and 37 °C was
examined using the well-established dialysis method
,,
and compared to the release
of free curcumin (1 g/L curcumin in DMSO/water = 40/60 V/V %). Immediately after preparation, 1 mL of each
curcumin-loaded nanoformulation (c
polymer = 5 mg/mL) was transferred into a glass tube fitted with 3.5 kDa
SpectraPor 6 dialysis membranes and the other end was sealed with
a cap. The tubes were then immersed in 40 mL of PBS containing 3%
Tween 80, stirred at 150 rpm, and maintained at 37 ± 0.5 °C.
At predetermined intervals, 50 μL of the samples were withdrawn
and diluted with 0.95 mL of EtOH, and the drug content was quantified
spectrophotometrically (424 nm, 25 °C, JASCO V-650). An additional
experiment was carried out similarly, with the exception that the
temperature of the medium was reduced to 32 °C after 1 h. All
the measurements were conducted in triplicate.
The effect of
the temperature reduction on the curcumin content of the graft copolymer-based
formulations was investigated. After preparing the drug-loaded aggregates
at 37 °C, the solutions were cooled to 32 and 25 °C. Any
precipitated curcumin was separated via filtering with a 0.45 μm
syringe filter. The concentration of curcumin was then determined
by using a JASCO V-650 spectrophotometer at 424 nm and ambient temperature.
All measurements were carried out in triplicate.

Hemolysis
Assay
The hemolytic activity of the polymers
was tested on human red blood cells (RBCs), obtained from healthy
volunteers, by the Hungarian National Blood Transfusion Service (Budapest,
Hungary). First, RBCs were washed twice with PBS, then diluted with
PBS (2% V/V), and plated on a 96-well,
round-bottom plate (Sarstedt, Nümbrecht, Germany) at a volume
of 50 μL. Solutions were prepared at a volume of 50 μL
with concentrations of 1.25, 2.5, and 5 g/L and added to the erythrocytes.
The final RBC concentration was 1% V/V, and the final volume was 100 μL in the wells. Plates were
incubated at 37 °C, 5% CO2 for 2 h, centrifuged at
2000 rpm (5 min, 4 °C), and then 50 μL of the supernatants
were carefully transferred to a new plate containing 100 μL
of H2O in each well. Optical density was measured at 414
and 450 nm by a Synergy 2 multimode microplate reader (BioTek, Winooski,
VT, USA), and the percentage of hemolysis compared to a positive control
(bee venom Melittin, 10 μM, in PBS) was determined. All data
are presented as the mean ± SD, N = 3.

Cytostasis
Assay on HT29 Cells
First, viability of
HT-29 colorectal adenocarcinoma cells was tested after treatment with
the polymers at 2, 0.67, and 0.22 g/L concentrations. Cells were seeded
1 day prior to the experiment in a 96-well tissue culture plate (Sarstedt)
at a density of 10,000 cells in 100 μL per well in RPMI-1640
medium. The day after, cells were treated with polymers for 24 h (37
°C in 5% CO2), then washed 3 times, and incubated
for a further 2 days in 10% FBS-containing RPMI-1640 medium. Percentage
of viability, compared to medium-treated cells, was measured by using
the (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
colorimetric assay described by Mossmann and modified by Horváti. All
data are presented as the mean ± SD, N = 4.
The cytostatic effect of curcumin and curcumin-loaded polymers was
tested, as described above. To avoid possible precipitation of free
curcumin, which could hamper the readability of the assay, the final
curcumin concentration was adjusted to 7 μM in all formulations,
including both free and encapsulated samples, while the polymer concentration
was 0.67 g/L.
Statistical analysis was performed using one-way
analysis of variance
(ANOVA), followed by Tukey’s post hoc test. Statistical significances
were denoted as ***p < 0.0001.

Internalization
Study
Internalization of the curcumin
and curcumin-loaded polymers was measured in the HT-29 cells. One
day before the experiment, cells were seeded in a 24-well tissue culture
plate (Sarstedt, 100,000 cells/well in 1 mL of RPMI-1640 medium).
The next day, cells were treated with 5 μM curcumin, with or
without the polymers. The polymer concentration was 0.5 g/L. After
incubation for 3 h, cells were washed with serum-free RPMI-1640 medium.
Then, the supernatant was removed, and 100 μL of 0.25% trypsin
was added to the wells. After 10 min of incubation, 0.8 mL of 10%
FBS/RPMI-1640 medium was added and then transferred to 5 mL FACS tubes.
Cells were washed two times with PBS, and then the intracellular fluorescence
intensity of the cells was measured on a CytoFLEX (Beckman Coulter,
Brea, California, US) on an FITC fluorescence channel (excitation
λ = 488 nm, emission at λ = 525/40 nm). All measurements
were performed in triplicate, and the mean fluorescent intensity together
with SD was presented. Statistical analysis was performed using one-way
ANOVA, followed by Tukey’s post hoc test. Statistical significance
was denoted as ***p < 0.001.

Results
and Discussion

The Synthesis of P­(NiPAAm-co-NAOS) and the
PNiPAAm-g-HbPG Copolymers

Scheme
shows the synthetic route
of the preparation of PNiPAAm-g-HbPG graft copolymers.
First, NiPAAm-based copolymers containing reactive NAOS comonomer
units were prepared via free radical copolymerization using three
distinct comonomer ratios. The obtained comonomer ratios in the resulting
P­(NiPAAm-co-NAOS) copolymers were evaluated from
the integral ratios of the methine protons of the NiPAAm isopropyl
groups and the methylene protons of NAOS (2.88–3.13 ppm) in
the 1H NMR spectra (Figure S1). It was found that the amount of NAOS incorporated was slightly
less than that of the feed ratio. Specifically, the molar contents
of NAOS in the copolymers are 4.3, 7.8, and 15.7 mol % for samples
with 4.8, 9.1, and 16.7 mol % NAOS in the feed, respectively. Afterward,
the measured comonomer contents, that is, 4.3, 7.8, and 15.7, will
be used for sample identification. The amino-monofunctional HbPG (NH2–HbPG, M
n,NMR = 1050 g/mol,
PDI∼1.2; see Figures S2–S4) was obtained via a two-step procedure described in our prior work. Specifically, monofunctional phthalimide-HbPG
was synthesized by anionic ring-opening multibranching polymerization
of glycidol, employing a phthalimide/phthalimideK initiator system,
followed by hydrazinolysis to obtain the amino-functional polymer,
as illustrated in Scheme
, step 2. Subsequently, the obtained amino-HbPG was grafted
onto the P­(NiPAAm-co-NAOS) precursors in MeOH at
room temperature, as shown in Scheme
, step 3.
Based on the GPC curves (Figure S5),
the P­(NIPAAm-co-NAOS) copolymers are formed within
a similar molecular weight range with a PDI of approximately 2. It
has to be noted that the average molecular weights are relative values
due to the differences in the hydrodynamic volumes of the linear homopolymer
standards used to calibrate GPC and those of the prepared copolymers.
Comparing the GPC curves of the PNiPAAm-g-HbPG graft
copolymers to those of the initial P­(NiPAAm-co-NAOS)
copolymers indicates that the relative amount of the small molecular
weight fraction decreases. At the same time, the curves broaden, and
a shoulder appears in the high molecular weight range for all graft
copolymers, suggesting the successful grafting of the HbPG. The appearance
of the shoulder in the higher molecular weight fraction could be explained
by the different solvation, thereby the different hydrodynamic properties
of this fraction containing more hydrophilic HbPG. The conjugation
reaction between the P­(NiPAAm-co-NAOS) and the NH2–HbPG is also supported by the 1H NMR spectroscopy
results (see Figure S1). The characteristic
signal of the succinimidyl functional groups of NAOS in the 2.8–3.1
ppm range does not appear in the spectra of the PNiPAAm-g-HbPG products, and the appearance of the broad peaks of the methylene
and methine protons next to the ether groups in HbPG in the range
of 3.4–4.1 ppm confirms the presence of HbPG in the resulting
polymers, i.e., the formation of PNiPAAm-g-HbPG graft
copolymers by grafting of HbPG onto the P­(NiPAAm-co-NAOS) copolymers. In this work, the term grafting density refers
to the number of HbPG chains attached per 100 monomer units, which
is directly determined from the mol % NAOS content obtained by 1H NMR analysis.

The Thermoresponsive Behavior of the P­(NiPAAm-co-NAOS) and the PNiPAAm-g-HbPG Copolymers
The thermoresponsive properties of both the P­(NiPAAm-co-NAOS) copolymers and the PNiPAAm-g-HbPG graft copolymers
were studied by turbidimetric measurements. In our study, turbidity
is defined and measured via transmittance at 488 nm using a UV–Vis
spectrophotometer. The recorded Tr-T curves of the
P­(NiPAAm-co-NAOS) copolymer precursors and the HbPG-grafted
PNiPAAm-based copolymers in water are presented in Figure
A,B, respectively (for their
first derivative curves, see Figure S6).
The turbidimetry results of the PNiPAAm homopolymer are displayed
in Figure S7. The determined T
CP and T
CL values for all
of the investigated polymers are listed in Table S1.

Figure
A shows
that the transmittance of the P­(NiPAAm-co-NAOS) copolymers
and PNiPAAm decreases to zero within a narrow temperature range and
exhibits reversibility in heating–cooling cycles. It has to
be noted that despite the fact that the synthesis and application
of P­(NiPAAm-co-NAOS) copolymers have already been
reported,
,−

the impact of the NAOS comonomer
on the thermoresponsive behavior has not been investigated so far.
As shown in Figure
C, the values of both the T
CP and the T
CL decrease linearly with increasing NAOS content
in the P­(NiPAAm-co-NAOS) copolymers, which is a consequence
of the hydrophobic nature of the NAOS comonomer, in line with the
effect of other hydrophobic comonomers.

This finding indicates that the CST values decrease by approximately
0.5 °C/mol % as the mole fraction of NAOS increases. As can be
seen in Figure
A,C,
there is a notable hysteresis between the T
CP and T
CL values for the PNiPAAm homopolymer and also in the case of the P­(NiPAAm-co-NAOS) copolymers due to the well-known fact that PNiPAAm
forms intermolecular and intramolecular hydrogen bonds in the precipitated
stage above the CST. However, this hysteresis is not observed for
the P­(DEAAm-co-NAOS) copolymers, which are free from
hydrogen bonding as reported recently. Nevertheless, the increase of the NAOS content does not lead to
a decrease in hysteresis, which suggests that this monomer is also
involved in the formation of secondary interactions, i.e., hydrogen
bonding with NiPAAm monomer units, and consequently, responsible also
for the appearance of hysteresis. The CST values were also determined
for the PNiPAAm-g-HbPG graft copolymers at a 1 g/L
polymer concentration (Figure
B). Contrary to the P­(NiPAAm-co-NAOS) copolymers,
a linear increase in CST values was observed with increasing the HbPG
content of the copolymers as shown in Figure
C. This result unambiguously demonstrates
the successful grafting of HbPG to the PNiPAAm-co-NAOS copolymers. The molar effect of HbPG on the CST value is more
pronounced than that of the NAOS. Specifically, the CST value of the
PNiPAAm-g-HbPG copolymers increases approximately
by 1.15 °C/mol % of grafting side chains, which is attributed
to the high hydrophilicity of the HbPG chains. This trend is in good
agreement with our previous observations on PDEAAm-g-HbPG copolymers, where an analogous increase in CST with increasing
HbPG content was also detected. Surprisingly,
the hysteresis between the T
CP and T
CL values measured for the PNiPAAm-g-HbPG copolymers is smaller than that for the P­(NiPAAm-co-NAOS) copolymers, indicating that the bulky hydrophilic HbPG grafts
suppress hydrogen bonding between the PNiPAAm chain segments to some
extent.
The Tr-T curves of the PNiPAAm homopolymer
and
PNiPAAm-g-HbPG copolymer solutions were measured
at varying polymer concentrations (1, 5, 10 g/L) in water and a physiological
PBS buffer, as shown in Figure
D–I. The transmittance change is less sharp in the
case of the PNiPAAm-g-HbPG graft copolymers than
that for the PNiPAAm homopolymer at the lowest polymer concentration
(1 g/L), occurring within a range from 40 to 60 °C, from 45 to
70 °C, and from 45 to 80 °C for the PNiPAAm-g-HbPG4.3, PNiPAAm-g-HbPG7.8, and PNiPAAm-g-HbPG15.7 samples, respectively.
The corresponding T
CP and T
CL values are plotted against the grafting density, i.e.,
the molar fraction of the HbPG in the copolymers, and against the
polymer concentration, as shown in Figure
J,K. The data shown indicate that both the
cloud point (T
CP) and clearing point (T
CL) decrease with increasing polymer concentration,
independent of copolymer composition or the solvent used. Notably, at the same polymer concentration,
the transmittance changes for the PNiPAAm-g-HbPG
copolymers appear at higher temperatures compared with the homopolymer.
This trend demonstrates that both T
CP and T
CL rise substantially with increasing HbPG graft
density in water and PBS (Figure
J,K). As expected, the presence of salt, i.e., the
use of PBS, has a notable impact on the thermoresponsive behavior.
Namely, similar to the recently reported PDEAAm-g-HbPG copolymers, the CST values in
PBS are about 2–3 °C lower than in water, although there
is a slight decline in this effect with increasing polymer concentration.
When using PNiPAAm-g-HbPG graft copolymer-based drug
delivery systems, it is important to consider that increasing the
HbPG content while decreasing the polymer concentration may limit
the ability to exploit the thermoresponsive nature, as the CST values
of graft copolymers with more HbPG at lower concentrations in PBS
exceed body temperature.
The thermoresponsive behavior of the
graft copolymers was also
investigated by DLS measurements. Figure
A–C presents the size distribution
of the PNiPAAm-g-HbPG graft copolymers with a 1 g/L
polymer concentration at varying temperatures. As observed, the graft
copolymers exhibit a mean diameter of approximately 5–10 nm
at room temperature, which increases markedly around T
CP and levels off over 50 °C in the 350–550
nm size range. Upon returning to room temperature, the mean diameter
exhibits the same small values observed at the initial stage of the
heating process, indicating the reversible nature of the thermally
induced transition. Figure
D–F shows the hydrodynamic diameter values as a function
of the temperature, overlaid with the turbidimetric curves measured
at the same concentration. Although the initial temperature of the
size change is often used in the literature to characterize the CST, the coil-to-globule and globule-to-aggregate
transitions take place simultaneously, i.e., at this temperature,
the aggregation of partially dehydrated chains can commence instantaneously
before the entire polymer undergoes the coil-to-globule transition.
Consequently, similar to the turbidimetric studies,
the temperature
of the coil-to-globule-to-aggregate transition can be identified as
the temperature at the inflection point of the sigmoid curve fitted
to the size versus temperature data. The temperatures at the inflection
points (T
inf) are consistent with the
CST values determined by turbidimetric measurement and exhibit a similar
trend as displayed in Figure
G. The T
inf increases with increasing
HbPG graft density, and the difference between these values and the T
CP determined by turbidimetry is about 1.7–2.4
°C. This discrepancy between these values can be attributed to
the sensitivity of DLS and/or the differences in the applied concentrations,
heating and cooling rates, and the wavelength used in turbidimetry
and DLS measurements.
,
As presented in Figure
D–F, in addition to T
inf, the initial temperature of size change
(T
onset) and the maximum mean diameter
are also affected by the composition of the PNiPAAm-g-HbPG graft copolymers. Increasing the HbPG content results in higher
onset temperatures and smaller maximum average sizes of the aggregates
above the CST, i.e., 550, 410, and 370 nm for copolymers with 4.3,
7.8, and 15.7 mol % of HbPG, respectively. This indicates hindered
aggregation by the hydrophilic HbPG side chains leading to smaller
aggregates above the CST, which is in accordance with the turbidimetric
results. At the same concentration, the transmittance of the copolymer
solutions with a higher HbPG content does not reach zero even at 80
°C. Specifically, the transmittance values were approximately
8% and 35% for the PNiPAAm-g-HbPG7.8 and
PNiPAAm-g-HbPG15.7 samples, respectively.
These findings indicate that in addition to the CST, the extent of
copolymer aggregation, which leads to the formation of nanoparticles
with PNiPAAm-enriched hydrophobic core and a well-hydrated HbPG shell,
can be regulated by the HbPG content of the copolymers, i.e., the
grafting density, as depicted in Figure
H.

Drug Delivery Properties of the PNiPAAm-g-HbPG
Graft Copolymers
The results presented in the previous section
on the thermoresponsive behavior of PNiPAAm-g-HbPG
graft copolymers have revealed that these copolymers form nanosized
aggregates composed of hydrophobic PNiPAAm and hydrophilic HbPG segments
at specified concentrations above their CST. Consequently, the amphiphilic
nature of these nanoassemblies may enable the encapsulation of hydrophobic
drugs. Therefore, the drug delivery properties of the graft copolymers
were investigated by determining the drug uptake and release profile
of curcumin.
The amount of curcumin encapsulated within PNiPAAm-g-HbPG aqueous solutions was measured by UV–vis spectroscopy
across a range of polymer concentrations at 25 and 37 °C (Figure
A,B). Figure
shows that increasing the
polymer concentration results in an increase in the concentration
of encapsulated curcumin. At higher polymer concentrations, the concentration
of curcumin using the PNiPAAm-g-HbPG copolymers is
2 or 3 orders of magnitude higher than its reported water solubility
of 0.6 mg/L at room temperature.
,
Importantly,
the control experiment performed with pure HbPG (10 g/L) showed no
detectable increase in the curcumin concentration. At higher PNiPAAm-g-HbPG concentrations (c
polymer > 2 g/L), the concentration of the encapsulated curcumin increases
even at room temperature, i.e., below the CST of the graft copolymers.
This finding suggests a moderate interaction between the hydrophobic
curcumin and NiPAAm segments of the graft copolymers, while the HbPG-enriched
hydrophilic outer shell remains intact.

Figure
C shows
that the composition has a considerable influence on the encapsulation
capabilities, as evidenced by the comparison of the drug loading content
(DLC) at a polymer concentration of 10 g/L. As obtained, while there
is no significant difference in the DLC at room temperature, drug
loading contents are consistently higher at 37 °C, and the increasing
HbPG content reduces the relative encapsulated drug content. This
finding is related to the relative hydrophobic nature of the copolymer
aggregates; that is, increasing the HbPG content of the copolymer
increases the CST values and decreases the apparent hydrophobicity
of the PNiPAAm segments at 37 °C, resulting in a reduction of
the encapsulation capacity. Consequently, a higher degree of curcumin
encapsulation can be achieved at elevated temperatures and a higher
NiPAAm content of the graft copolymer nanocarrier. As recently reported,
the PDEAAm-g-HbPG copolymers exhibit higher curcumin
encapsulation even at room temperature due to the more hydrophobic
nature of PDEAAm compared to that of PNiPAAm. Therefore, the temperature-dependent drug uptake property of the
PNiPAAm-based graft copolymers, which is illustrated in Figure
D, suggests the possibility
of a thermally induced release profile, in contrast to the PDEAAm-containing
analogs for which the absence of this ability was obtained.

The drug release profile of curcumin from
the polymer aggregates
(c
polymer = 5 mg/mL), and also the free
form as control to demonstrate the applicability of the utilized measurement
technique (1 g/L curcumin in DMSO/water = 40/60 V/V %), was investigated
by using the dialysis method in PBS (pH = 7.4) in the presence of
3% Tween 80 at 37 °C.
,,
The drug release is expressed as the cumulative release as a function
of the release time (Figure
A). The measured curcumin concentration as a function of the
release time can be seen in Figure S9.
As shown, free curcumin in DMSO/water exhibits rapid diffusion through
the dialysis membrane, resulting in over 90% release within approximately
5 h. However, an increase in the HbPG content in the PNiPAAm-g-HbPG graft copolymers results in a slower sustained release
of curcumin. This effect can be attributed to the fact that a higher
HbPG content results in a denser and well-hydrated surface region
of the particles, which suppresses the transfer of hydrophobic curcumin
to the medium. A similar trend was observed in the case of PDEAAm-g-HbPG graft copolymer, as reported recently. These findings are also consistent with those
of Wu et al., who demonstrated that
high-molecular-weight HbPG delays the release of hydrophobic guest
molecules.
One of the most significant advantages
of thermoresponsive
polymers
is their ability to facilitate the controlled release of drugs in
response to changes in the temperature. Consequently, the effect of
cooling on the curcumin content and release properties was investigated.
The concentration of the encapsulated curcumin was evaluated at 37
°C and after cooling to 32 and 25 °C (Figure
B), revealing a notable decrease in drug
concentration as the temperature is lowered.
The data in Figure
B also indicate that
increasing the HbPG grafting density results
in a less pronounced loss of the encapsulated curcumin content upon
cooling. This finding aligns with the encapsulation results; i.e.,
the lower HbPG content of the graft copolymers leads to enhanced encapsulation
ability at elevated temperatures, and following cooling, the encapsulated
surplus of curcumin is fully released. The release is less temperature-dependent
when the HbPG content is higher, due to the fact that the CST of the
copolymer containing the highest amount of HbPG at the applied polymer
concentration (5 g/L) is above 37 °C. Nevertheless, these findings
clearly prove that cooling the environment can trigger drug release
from the PNiPAAm-g-HbPG graft copolymers. Consequently,
to reveal the impact of modifying the temperature of the medium on
the release profile, drug release experiments were conducted with
a reduction in temperature from 37 to 32 °C after 1 h. Figure
C displays the initial
release of curcumin during the first 6 h, while the subsequent decrease
in curcumin concentration and the cumulative release over 24 h are
depicted in Figure S10. Following the cooling
process, the release of curcumin increased in all graft copolymers
compared to the release measured at 37 °C, with a notable dependence
on the composition. In the case of PNiPAAm-g-HbPG4.3, the cumulative drug release was increased by approximately
15% after 1 h of cooling. Similarly, in the case of PNiPAAm-g-HbPG7.8, an increase of approximately 20% was
observed over the 2–3 h period. However, a substantial difference
was not observed in the release profile of the highest HbPG-containing
graft copolymer in the first hours, and the increase in the release
was lower than 10% after 3 h following the temperature change. These
observations were further supported by the kinetic evaluation of the
curcumin release according to the Higuchi model (Figure S11), which revealed a predominantly diffusion-controlled
release mechanism for all cases (R
2 >
0.94). The obtained Higuchi rate constants (K
H) increase for all the graft copolymer-based nanoformulations
when the temperature is reduced from 37 to 32 °C, indicating
an acceleration of the curcumin release upon cooling. Specifically,
the K
H values are 37.2, 16.6, and 10.6%
h–1/2 at 37 °C and 39.0, 22.5, and 13.0% h–1/2 at 32 °C for the PNiPAAm-g-HbPG graft copolymer nanocarriers with increasing HbPG grafting
density, respectively. Moreover, the amount of drug released after
24 h increased upon cooling, rising from 90.5% to 93.6%, 44.6% to
55.1%, and 27.6% to 33.6% with increasing HbPG content. While the
released drug fraction for the PNiPAAm-g-HbPG4.3 sample is close to complete, the other two formulations
still retain a portion of curcumin. These findings indicate that both
the drug uptake and the release of the loaded drug can be induced
by changing the temperature. In brief, the release of curcumin from
the PNiPAAm-g-HbPG-based nanoformulations is sustained
and continuous, and its rate can be decreased by increasing the HbPG
content. Conversely, a lower HbPG content enables triggering of the
release by decreasing the temperature, which induces the disaggregation
of the nanoformulation, as illustrated in Figure
D. As a result, the PNiPAAm-g-HbPG graft copolymers represent promising candidates for temperature-sensitive
drug carrier and release systems.

Colloidal Stability of
the PNiPAAm-g-HbPG Graft
Copolymers and the Curcumin-Loaded Formulations
The colloidal
stability of drug-loaded polymeric nanoparticles at the temperature
of administration in the body is crucial for medical applications.
To further assess their behavior under physiologically relevant conditions,
the colloidal stability of the PNiPAAm-g-HbPG graft
copolymers, the PNiPAAm homopolymer, and curcumin in its free and
nanoformulated forms was analyzed using turbidimetry.
As shown
in Figure
, increasing
the temperature results in the formation of highly turbid systems
above their CSTs. In contrast to the PNiPAAm, for which rapid aggregation
results in macroscopic precipitation and sedimentation within approximately
100 min, the turbidity of the HbPG-grafted copolymers remains unchanged
over the course of the measurement (see also Figure S12). These findings indicate that the stable nanoassemblies
formed above the CSTs possess a hydrophobic central core coated with
a HbPG-enriched hydrophilic shell, which ensures steric stabilization
of the colloidal dispersion, as illustrated in Figure
B. As illustrated in Figure
A, free curcumin underwent rapid aggregation
and sedimentation as indicated by the sharply increasing transmittance,
highlighting the challenges associated with hydrophobic drugs, particularly
their limited bioavailability. These processes can be slightly delayed
by the presence of the PNiPAAm homopolymer, but the sedimentation
remains fully complete at a rate very similar to that of the pure
homopolymer. In contrast, the curcumin-containing PNiPAAm-g-HbPG copolymer dispersions, which form aggregates with
mean diameters of 250–320 nm (Figure S13) and relatively narrow dispersity (PDI 0.05–0.07), demonstrate
high and nearly constant turbidity over the investigated period. This
finding indicates that the stability of the PNiPAAm-g-HbPG nanoaggregates predominantly governed by steric hindrance from
the hydrophilic HbPG shell, with negligible electrostatic effects,
i.e., HbPG as a part of the outer shell of the thermally formed nanoparticles
can hinder further aggregation and sedimentation, as depicted schematically
in Figure
B.

In Vitro Susceptibility and Bioapplicability
of the PNiPAAm-g-HbPG Graft Copolymers
To
ascertain the suitability of the prepared PNiPAAm-g-HbPG graft copolymers as drug carriers in biological systems, their
hemolytic activity and cytotoxicity were evaluated by using human
red blood cells (RBC) and the HT-29 human cell line with epithelial
morphology, respectively. It was found that none of the polymers tested
impaired cell viability, even at the highest tested concentration
of 2 or 2.5 g/L (Figure
A,B). These findings suggest sufficient in vitro susceptibility of the PNiPAAm-g-HbPG graft copolymers,
which is in line with our observations with PDEAAm-g-HbPG. Subsequently, the biological
applicability of the PNiPAAm-g-HbPG graft copolymers
as a drug delivery system was investigated by examining the cellular
uptake and cytostatic effect of curcumin-loaded copolymer formulations
on the HT-29 human colorectal adenocarcinoma cell line, which is frequently
utilized to identify novel chemotherapeutic agents and more efficient
anticancer treatments.
,
Numerous studies in
the past decade have demonstrated that curcumin possesses both antiproliferative
and proapoptotic activities toward several human cancer cell lines,
including those derived from colon cancer. As illustrated in Figure
C, curcumin exhibits a considerable cytostatic effect of approximately
60% when administered in its free form at a concentration of 7 μM.
Furthermore, similar to the case of PDEAAm-g-HbPG, the cytostatic effect is notably enhanced by
utilizing PNiPAAm-g-HbPG graft copolymers for formulation.
The improved anticancer activity can be explained by the results of
the cellular uptake measurements (Figure
D). In the case of the curcumin-loaded graft
copolymers, the mean fluorescence intensity is 2-fold that of the
unformulated curcumin, indicating that the graft copolymers significantly
enhance the drug uptake of curcumin by cancer cells. These results
provide clear evidence of the advantageous applicability of the HbPG-grafted
PNiPAAm copolymers as drug delivery and release systems.

Conclusions
In
this study, HbPG-grafted poly­(N-isopropylacrylamide)
(PNiPAAm-gHbPG) thermoresponsive copolymers with
varying grafting density were successfully synthesized through the
grafting of monofunctional amino-HbPG onto reactive N-hydroxyl succinic amide active ester-containing functional PNiPAAm
copolymers. The thermoresponsive behavior of the copolymers was investigated
using turbidimetry and DLS measurements, which demonstrated that the
CST of the PNiPAAm-g-HbPG copolymers increases with
increasing HbPG content at a given polymer concentration, conversely
to the P­(NiPAAm-co-NAOS) copolymer precursors, where
a decrease in CST was observed with increasing hydrophobic NAOS content.
The reversible aggregation–disaggregation transition of the
PNiPAAm-g-HbPG graft copolymers was demonstrated
by a heating/cooling cycle between the small, coiled particles (∼5
nm) at room temperature and larger copolymer aggregates with a 300–500
nm mean diameter above the CST.
The encapsulation experiments
confirmed that curcumin was effectively
loaded into the nanoscale aggregates formed by the graft copolymers.
It was found that the encapsulated curcumin concentration increases
with increasing polymer concentration up to 2–3 orders of magnitude
higher than that of the water solubility of curcumin. An increase
in temperature or a reduction in HbPG content in the PNiPAAm-g-HbPG graft copolymers enhanced the curcumin loading, which
can be attributed to the increased hydrophobic character of the PNiPAAm
polymer backbone. Another significant consequence of HbPG grafting
onto PNiPAAm is the sustained release of curcumin from the drug formulations.
The extent of sustained drug release was found to be lower in graft
copolymers with higher HbPG content, resulting in a denser and more
hydrated surface region of the polymer nanoparticles, which may hinder
the release of hydrophobic curcumin to some extent. Temperature variations
can also affect the release profile of curcumin from the graft copolymers;
i.e., lowering the release temperature by 5 °C results in a higher
release rate, especially for copolymers with lower grafting density.
This enables site-specific drug release induced by localized cooling.
Due to the hydrophilic branched grafted side chain, exceptional colloidal
stability of the PNiPAAm-g-HbPG copolymers and the
drug-loaded formulations were obtained under physiologically relevant
conditions, suggesting that the formed nanoassemblies have a stabilizing
outer shell enriched with the hydrophilic HbPG side chains.
The graft copolymers were found to be nontoxic to epithelial cells
and did not cause hemolysis of human RBC even at relatively high concentrations,
i.e., 2 and 2.5 g/L, thereby indicating their in vitro susceptibility as an essential requirement. The bioactivity study
on the HT-29 colon cancer cell line shows that the nanoformulation
of curcumin by the graft copolymers enhances the internalization rate,
resulting in a significantly higher anticancer activity for the curcumin-loaded
copolymer nanoparticles than that of free curcumin. The tested cell
line can effectively model other colorectal cancer cell lines, such
as HC-T15, HCT-116, or SW-620, and this cell line is also often used
as a model system for studying multiple aspects of cancer biology,
including cytostasis, cell proliferation, differentiation, and drug
resistance. In particular, the HT-29 cell line is used in numerous
protocols involving multicellular intestinal systems and organoids, and HT-29 xenograft mouse model (subcutaneous
and metastatic) is also available, enabling
preclinical therapeutic studies. Considering all of these aspects,
our results show promising translational potential. Overall, the PNiPAAm-g-HbPG graft copolymers exhibit several favorable characteristics,
including tunable thermoresponsiveness, excellent colloidal stability,
and in vitro susceptibility. Due to the thermally
triggered encapsulation and sustained drug release properties with
enhanced bioactivity, the PNiPAAm-g-HbPG graft copolymers
have significant potential as advanced drug delivery and release systems.

Supplementary Material