Biomaterials 145 (2017) 138e153

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Clickable and imageable multiblock polymer micelles with

magnetically guided and PEG-switched targeting and release property
for precise tumor theranosis
Jing Wei a, Xiaoyu Shuai a, Rui Wang a, Xueling He a, b, Yiwen Li a, Mingming Ding a, *,
Jiehua Li a, Hong Tan a, **, Qiang Fu a
a
College of Polymer Science and Engineering, State Key Laboratory of Polymer Materials Engineering, Sichuan University, Chengdu, 610065, China
b
Laboratory Animal Center of Sichuan University, Chengdu, 610041, China

a r t i c l e i n f o a b s t r a c t

Article history: Targeted delivery of therapeutics and diagnostics using nanotechnology holds great promise to minimize
Received 22 February 2017 the side effects of conventional chemotherapy and enable specific and real-time detection of diseases. To
Received in revised form realize this goal, we report a clickable and imageable nanovehicle assembled from multiblock poly-
1 July 2017
urethanes (MPUs). The soft segments of the polymers are based on detachable poly(ethylene glycol)
Accepted 1 August 2017
Available online 18 August 2017
(PEG) and degradable poly(ε-caprolactone) (PCL), and the hard segments are constructed from lysine-
and cystine-derivatives bearing reduction-responsive disulfide linkages and click-active alkynyl moieties,
allowing for post-conjugation of targeting ligands via a click chemistry. It was found that the cleavage of
Keywords:
Click chemistry
PEG corona bearing a pH-sensitive benzoic-imine linkage (BPEG) could act as an on-off switch, which is
Biodegradable multiblock polyurethane capable of activating the clicked targeting ligands under extracellular acidic condition, followed by
Multifunctional polymer micelles triggering the core degradation and payload release within tumor cells. In combination with super-
Magnetic resonance imaging paramagnetic iron oxide nanoparticles (SPION) clustered within the micellar core, the MPUs exhibit
Tumor targeting excellent magnetic resonance imaging (MRI) contrast effects and T2 relaxation in vitro, as well as
Cancer theranosis magnetically guided MR imaging and multimodal targeting of therapeutics to tumor precisely, leading to
significant inhibition of cancer with minimal side effect. This work provides a safe and versatile platform
for the further development of smart theranostic systems for potential magnetically-targeted and
imaging-guided personalized medicine.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
The effective treatment and early detection of cancers still
remain great challenges, since the transport of therapeutic and
diagnostic agents to the tumors have been seriously hampered by
the complicated body environments and various physiological
barriers [1]. To address these challenges, targeted delivery system
based on nanotechnology holds enormous potential to enable
engineered nanomedicines to travel through the body in a specific
way [2]. On the one hand, drug-loaded nanoparticle systems can
overcome the problems brought by conventional free drugs,
* Corresponding author.
** Corresponding author.
E-mail addresses: dmmshx@scu.edu.cn, dmmshx@163.com (M. Ding), hongtan@
scu.edu.cn (H. Tan).
including poor solubility, low stability, rapid clearance and lack of
selectivity, which have exerted adverse effects on healthy cells and
prevented the dose escalation necessary to eliminate malignant
cells [3]. On the other hand, the encapsulation of diagnostic agents
into polymeric nanovehicles can remedy the inadequate sensitivity
and specificity of most commercialized contrast agents for imaging
(e.g., Endorem, Resovist, etc.) [4], resulting in an improved spatial
revolution and remarkable contrast against surrounding tissues by
clustering these agents and localizing them at the sites of interest
[5].
Relying on the so-called enhanced permeability and retention
(EPR) effect ascribable to the presence of leaky vasculature and
impaired lymphatic drainage in most rapidly growing tumors,
nanomedicines show higher accumulation at tumor sites than the
conventional drugs and contrast compounds [6]. However, EPR
effect has been regarded recently as an immensely heterogeneous
phenomenon, which varies with the types and stages of diseases

http://dx.doi.org/10.1016/j.biomaterials.2017.08.005
0142-9612/© 2017 Elsevier Ltd. All rights reserved.
 J. Wei et al. / Biomaterials 145 (2017) 138e153
and differs from person to person [7]. Therefore, the EPR-based
passive targeting seems not sufficient to match the changeable
and complicated tumor microenvironment. As a result, it is only
effective for limited patient subpopulations [8]. The cooperation of
magnetic nanoparticles is expected to offer improved specificity via
magnetic drug targeting (MDT) [9,10]. As a physical force, the
magnetic stimulus is independent of the complicated physiological
conditions. Hence, MDT employing an external field gradient to
manipulate magnetically responsive particles can expedite the
extravasation and transport of nanoparticles into tumors, and
thereby break through the physical barriers and overcome the
intrinsic limitations of EPR effect [9,11]. Furthermore, magnetic
nanoparticles can also act as excellent contrast agents for magnetic
resonance imaging (MRI), which provides great opportunity to
construct theranostic platform for specific disease detection, real-
time monitoring and potential imaging-guided therapy [11,12].
Although nanomedicines can passively or magnetically target to
tumor tissues, the scarcity of cell-specific interaction and insuffi-
cient uptake of nanoparticles may decrease the therapeutic efficacy
and even induce drug expulsion and multiple drug resistance
(MDR) [13]. To overcome this problem, nanocarriers can be
equipped with a good variety of targeting ligands, such as anti-
bodies, folic acid (FA) and peptides that can actively bind to anti-
gens and receptors highly overexpressed by tumor cells [14].
Unfortunately, most of the actively targeted delivery systems
contain ligands on their surface [15]. The exposure of active surface
may promote nonspecific interactions with endothelial and other
non-cancerous cells and result in opsonization-mediated clearance
of nanomedicines from the body [16]. To address this dilemma,
coating nanocarrier with hydrophilic polyethylene glycol (PEG)
corona (known as PEGylation) can increase the circulation time by
reducing interactions with serum proteins and mitigating uptake
by phagocytic cells [17], whereas such a strategy in turn compro-
mises the targeting specificity [18].
Another approach to targeting is to take advantage of the special
microenvironments of tumors to achieve on-demand delivery,
which in principle allow for tailored release profiles with excellent
temporal, spatial and dosage control [19]. It is known that most
solid tumors develop unique microenvironments, such as lowered
interstitial pH, elevated glutathione (GSH) concentration and
increased level of certain enzymes [20]. In particular, the GSH
concentration in some tumors is several times higher than that in
normal tissues, and the intracellular level of GSH (1e11 mM in the
cytoplasm) is 2e3 orders higher than that outside the cells
(2e20 mM) [21,22]. Therefore, bio-responsive formulations con-
taining disulfide functionality can facilitate tumor-specific and
intracellular transport of therapeutics by the cleavage of disulfide
bond through thiol-disulfide exchange reactions triggered by GSH
[23]. Nonetheless, recent in vivo studies have shown that disulfide
could be potentially cleaved under circulation due to the presence
of reducing species in the blood [24], thus decreasing the specificity
of redox-responsive delivery systems [25]. On the other hand, the
current studied reduction-responsive systems are commonly pre-
pared by incorporating non-native segments containing disulfide
linkages, which may raise a safety concern [21,26].
As mentioned above, although the conventional targeting and
responsive systems taking advantages of EPR effect, targeting li-
gands and tumor microenvironments allow for specific delivery of
theranostics, each of these approaches is still suffering from some
limitations. Hence, the development of versatile nanomedicines to
overcome all the limitations and realize multimodal targeting and
precise therapy is highly desirable. To fulfill this need, multiblock
polyurethane (MPU) appears as a promising platform due to their
good biocompatibility and high molecular tunability [27,28]. MPUs
could be molecularly engineered to integrate various desired
139
functions into a single macromolecule in a smart and coordinated
way, as demonstrated recently by our group [29]. However, the role
of MPUs in the field of imaging and personalized nanomedicine
remains largely unexplored [30]. To date, the theranostic property
of MPUs have yet to be demonstrated. In addition, to make MPUs
more clinically approvable, a further improvement of PU versatility
and optimization of polymer structure is clearly warranted [27].
In this study, we report a clickable and imageable multiblock
polyurethane (MPU) system with switchable tumor targeting and
triggered drug release properties for precise tumor therapy and
specific MR imaging (Fig. 1). The soft segments of MPUs comprise
PEG and PCL, and the hard segments contain L-lysine ethyl ester
diisocyanate (LDI) (Fig. 1A). To eliminate the need of multiple chain
extenders and complicated synthetic procedures for the construc-
tion of multifunctional MPUs [29], an L-cysteine-derived versatile
chain extender (Cys-PA) was incorporated to endow polymers with
a number of reduction-cleavable disulfide linkages in the backbone
and clickable alkyne sites on the side chains (Fig. 1). Subsequently,
as a proof-of-concept, a post-conjugation of targeting ligand was
performed using a facile click chemistry after the formation of
polymer micelles (Fig. 1C). We found lately that FA-clicked polymer
micelles exhibit relatively lower targeting efficacy compared with
other FA-based delivery systems [31]. Herein, it is interesting to
note that the cleavage of PEG segment bearing a pH-sensitive
benzoic-imine linkage (BPEG) could act as a switch, which is
capable of activating cell targeting under extracellular condition,
followed by triggering the cleavage of disulfide and accelerating the
release of payloads within tumor cells (Fig. 1E). Furthermore, super-
paramagnetic iron oxide nanoparticles (SPIONs) and doxorubicin
(DOX) as model contrast and anticancer agents were efficiently co-
encapsulated into the micellar core to enable magnetophoretic
enhance of targeting and theranostic capability of MPUs. The
multiple stimuli-responsive property and multimodal targeting
effect of MPUs were systematically demonstrated, and the
magnetically guided cancer therapy and MR imaging were also
evaluated in vivo.
2. Materials and methods
2.1. Materials
N,N-Dimethylacetamide (DMAc) was purchased from Adamas
Reagent Co., Ltd., China, and dried over CaH2 overnight and vacuum
distilled. Methoxyl-poly(ethylene glycol) (MPEG, MW 1900, Alfa
Aesar) and PCL (MW 2000, Dow Chemical) were dehydrated under
reduced pressure at 90
C for 2 h before use. LDI, BPEG, Cys-PA, and
azido modified FA (AzFA) were synthesized according to previous
reports [31,32]. Iron (III) chloride hexahydrate (FeCl3$6H2O) was
purchased from Tianjin Damao Chemical Co., Ltd., China.
CuSO4.5H2O was attained from Tianjin Bodi Chemical Co., Ltd.,
China. Sodium ascorbate was purchased from Aladdin Chemical Co.
Ltd., China. Iron (II) chloride tetrahydrate (FeCl2$4H2O), ammonium
hydroxide solution (NH3$H2O), oleic acid, propanediol (PDO,
98.0%) and dimethyl sulfoxide (DMSO) were supplied by Chengdu
Kelong Chemical Co., Ltd., China. Doxorubicin hydrochloride
(DOX$HCl) was obtained from Tecoland Co., USA. 3-(4,5-dimethyl
thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) was pur-
chased from Sigma-Aldrich, USA. 2-(4-amidinophenyl)-6-
indolecarbamidine dihydrochloride (DAPI) was obtained from
Roche Diagnostics, Germany. Other reagents were purchased from
commercial suppliers and used without further purification.
2.2. Characterization
Proton nuclear magnetic resonance (1H NMR, 400 MHz)
140 J. Wei et al. / Biomaterials 145 (2017) 138e153

Fig. 1. Design of clickable and imageable multiblock polymer micelles. Schematic chemical structure (A) and molecular architecture (B) of clickable MPUs. The waved lines in (A)
represent PCL soft segments. (C) Self-assembly of MPU micelles and post-conjugation of folic acid via click chemistry. (D) Schematic illustration of FA residues on the interface of
polymer micelles. (E) Illustration of magnetic-guided and PEG-switched targeting and release properties of MPU nanocarriers.

spectroscopy was performed on a Bruker AV II-400 MHz spec-
trometer using DMSO-d6 as a solvent and TMS as an internal
standard at room temperature.
Fourier transform infrared (FTIR) spectroscopy was performed
on a Nicolet 6700 spectrometer (Thermo Electron Corporation,
U.S.A). The solution samples were cast on KBr plates before
measurement.
Gel permeation chromatography (GPC) was performed on
Waters-1515 (U.S.A) with DMF as an eluent at 40
C. Samples were
run at 2.0 mg mL1 with a flow rate of 1.0 mL min1.
Size measurement was performed on a Zetasizer Nano ZS dy-
namic light-scattering (DLS) instrument (Malvern, UK) at 25
C at
an angle of 90
. Each sample was determined for three times.
Transmission electron microscopy (TEM, Hitachi model H-600-
4, Japan) was recorded with an accelerating voltage of 75 KV. A drop
of micellar solution was placed on a copper grid with Formvar film,
and stained with 1% (w/v) phosphotungstic acid. Then the solution
was absorbed by a filter paper and dried naturally before mea-
surement. For the observation of magnetic nanoparticles, no
staining was needed.
High-resolution TEM (HRTEM, Tecnai G2 F20 S-TWIN, FEI, USA)
was operated with an accelerating voltage of 200 KV. A drop of
SPION solution in hexane was dropped on a copper grid with For-
mvar film and air-dried before measurement.
 J. Wei et al. / Biomaterials 145 (2017) 138e153
Thermogravimetric analyzer (TGA Q500, TA Instruments, USA)
was used to obtain the TGA curves of samples. The fully dried
powder was heated from room temperature to 800
C at a rate of
10
C min1 under argon atmosphere.
Crystallographic analysis was carried out on an X-ray diffrac-
tometer (XRD, Phlips X'Pert PRO, XL-30) using Cu Ka radiation. The
scattering angle (2q) was in the range of 20
e90
. The structure and
composition of the sample were determined by comparing the
spectrogram with standard powder diffraction files from the Joint
Committee for Powder Diffraction Studies (JCPDS).
Vibrating sample magnetometer (VSM, Lakeshore 7410, USA)
was used at 298 K under an applied magnetic field varying
from 15 to15 kOe.
2.3. Construction and characterization of clickable and imageable
multiblock polymer micelles
2.3.1. Synthesis of clickable and multistimuli-responsive MPUs
The clickable MPUs were synthesized through a solution poly-
merization as described previously [31]. In brief, PCL and LDI were
copolymerized in anhydrous DMAC at 60
C under a dry nitrogen
atmosphere for 1 h in the presence of 1‰ stannous octoate as a
catalyst. After cooling to room temperature, multifunctional chain
extender Cys-PA was added to react for 1 h at room temperature
and another 2 h at 60
C. Then different ratios of MPEG and BPEG
were added, and the reaction was completed after 6 h. Finally, the
polymers were precipitated in a mixture of methanol and anhy-
drous diethyl ether and dried under vacuum at 60
C.
2.3.2. Preparation of clickable MPU micelles
MPU micelles were prepared by a dialysis method. Typically, a
solution of MPU in DMAc (10 mg mL1) was dropped into phos-
phate buffered saline (PBS, pH 7.4) under ultrasonic condition. Then
the solution was transferred into a dialysis bag (MWCO 3500) and
dialyzed against PBS for 3 d, refreshing the dialysate every 3 h.
Afterward, the micellar solution was diluted with PBS to 1 mg mL1
and centrifugalized at 3000 r min1 for 15 min. Finally, the obtained
micelles were passed through a 0.45 mm pore-sized syringe filter
(Milipore, Carrigtwohill, Co. Cork, Ireland) and stored at 4
C.
2.3.3. Stability and corona-detachment behavior
To evaluate the stability and stimuli-responsiveness of clickable
and imageable polymer micelles, the samples were incubated in pH
7.4 or pH 6.5 at 37
C with shaking (100 r min1) for 24 h, and then
measured by DLS and TEM.
To investigate the corona-detachment profiles, 10 mL of SSPHPU
micellar solution (2 mg mL1) was transferred into a dialysis bag
(MWCO 6500) and incubated at pH 6.5 and pH 7.4 for different
times at 37
C with shaking (110 r min1). At proper time intervals,
3 mL of release media was taken out and replenished with an equal
volume of fresh media. The release of PEG derivatives was deter-
mined by an L5 ultravioletevisible (UVeVis) spectrometer
(Shanghai INESA Analytical Instrument Co., Ltd.) at the wavelength
of 256 nm. The concentration of released PEG segments was
calculated according to the calibration curve established from
MPEG-CBA standard solutions in release media at different con-
centrations. Values were reported as the means for each triplicate
samples. For structural characterization, solutions inside and
outside the dialysis bag were collected and lyophilized for 1H NMR
measurements.
To determine the change of molecular weight before and after
stimuli-responsive degradation, 10 mL of SSPHPU micellar solution
(2 mg mL1) was incubated under different conditions (pH 7.4; pH
5.0; 100 mM GSH; pH 5.0 and 100 mM GSH) for 48 h with shaking
(110 r min1). The solutions were freeze-dried for GPC analysis.
141
2.3.4. Preparation of FA-clicked MPU micelles
50 mL of MPU micellar solution (1 mg mL1) was mixed with
10 mg of AzFA, and 1 mg of sodium ascorbate and 0.5 mg of
CuSO$45H2O were added as catalysts. The solution was stirred for
24 h at room temperature, then transferred into a dialysis bag
(MWCO 3500) and dialyzed against PBS (pH 7.4) for 5 d. Thereafter,
the solution was centrifugalized at 3000 r min1 for 20 min and
filtrated by a 0.45 mm pore-sized syringe filter and stored at 4
C.
The grafting ratio of FA on PU micelles was measured using UVevis
spectroscopy at the wavelength of 280 nm, and calculated ac-
cording to a calibration curve generated from AzFA solutions at
different concentrations.
2.3.5. Synthesis of SPIONs
SPIONs were synthesized using a modified chemical co-
precipitation method [30,33]. Briefly, FeCl3$6H2O (4.1 g) and
FeCl2$4H2O (1.8 g) were dissolved in 45 mL of distilled water with
stirring under nitrogen. Then 12 mL of NH3$H2O was quickly added
for 1 h of reaction at 80
C. Thereafter, 1.8 g of oleic acid was added
dropwise, and the reaction was allowed to continue for 1 h. When
the mixture was cooled to ambient temperature, the SPION fluid
was collected using a magnet and washed with water and ethanol
for several times. The obtained product was re-dispersed in hexane
and storage at 4
C.
2.3.6. Preparation of SPION and drug-loaded MPU micelles
To prepare SPION-loaded MPU micelles (SPION@micelles),
MPUs and SPIONs were co-dissolved in THF/DMAc (1/9, v/v) and
dropped into deionized water under ultrasonic condition. The so-
lution was dialyzed against deionized water for 3 d, centrifugalized
at 3000 r min1 for 15 min, and filtered through a 0.45 mm pore-
sized syringe filter and stored at 4
C. To determine the loading
content of SPIONs in MPU micelles, the SPION@micelle solution was
lyophilized and characterized with TGA. To assess the dispersibility
of magnetic loaded nanocarriers, the SPION@micelle solution was
mixed with an equal volume of hexane. The mixture was vigorously
shaken and allowed to separate into two layers, then imaged with a
color digital camera.
To load drugs, DOX$HCl was firstly dissolved in DMAc and
desalted in the presence of excess amount of trimethylamine (TEA)
with stirring for 2 h. The solution was mixed with the above MPU
solution in DMAc or MPU/SPIONs solution in THF/DMAc, respec-
tively, for the preparation of drug-loaded micelles (DOX@micelles)
or drug and SPION co-encapsulated micelles (DOX/SPION@mi-
celles). Afterward, the mixtures were dropped into PBS (pH 7.4)
under ultrasonic condition, and dialyzed against PBS for 3 d to
remove the organic solvents. The micellar solution was cen-
trifugalized for 15 min at 3000 r min1 and filtered as described
above except that the whole process was operated in the dark. The
loading content of DOX in MPU micelles was determined using
UVevis spectroscopy.
2.4. PEG-switched and GSH-accelerated drug release
To study the controlled drug release behavior of multifunctional
micelles, 2 mL of DOX-loaded micelles prepared in PBS was trans-
ferred into a dialysis bag (MWCO 3500), and dialyzed against 30 mL
of PBS (pH 7.4 or 6.5) with or without 10 mM of GSH at 37
C with
shaking. The release media also contains 100 mM of sodium sa-
licylate to solubilize DOX and maintain a sink condition. At pre-
determined frequencies, 3 mL of release media was taken out and
replenished with an equal volume of fresh media. The content of
DOX in the media was measured by UVeVis spectrometer. The
release experiments were conducted in triplicate.
To evaluate the release of SPIONs from MPU micelles, 2 mL of
142 J. Wei et al. / Biomaterials 145 (2017) 138e153
SPION@SSPHPU was incubated at pH 5.0 or in the presence of
100 mM GSH with shaking at 37
C for 48 h. The solution was
allowed to stand for 12 h. Then images were recorded with a color
digital camera, and the supernatant was collected for turbidity
measurement by UVeVis spectrometer at 650 nm.
2.5. Cellular uptake and intracellular delivery
HeLa cells obtained from West China Medical Center of Sichuan
University were cultured in RPMI 1640 medium (HyClone, U.S.A.)
containing 10% fetal bovine serum (FBS, Gibco Life, U.S.A.) and 5%
penicillin-streptomycin (Gibco Life, U.S.A.) in 5% CO2 at 37
C. To
evaluate the effect of corona-detachment behavior on cellular up-
take, HeLa cells were cultured in a six-well plate (with a clean
coverslip placed in each well before use) at a density of 1  105 cells
per well for 24 h. The pH of the culture media was adjusted to 6.5 or
7.4 before treatment. Then 1 mL of DOX-loaded MPU micelles
containing 10 mg mL1 DOX were added and cultured for 4 h. The
supernatant was carefully removed and the cells were washed with
PBS for three times. The cells were fixed with 1 mL of 4% formal-
dehyde for 30 min and stained with DAPI for 10 min. Finally, the
coverslip was mounted with 50% glycerol solution and observed
with a confocal laser scanning microscope (CLSM, Leica TCS CP5).
The mean signal intensity (SI) of DOX fluorescence in cytoplasm
and nucleus was calculated and normalized with ImageJ 1.50i
(Wayne Rasband, National Institutes of Health, USA). For flow
cytometry, HeLa cells were cultured in a culture dish at a density of
1  106 cells per dish for 24 h. After pre-adjusting the culture media
to pH 6.5 or 7.4, 1 mL of DOX-loaded micelles containing 10 mg mL1
DOX were added and incubated for 4 h. The suspension was dis-
carded and the cells were washed with PBS for three times. Then
the cells were digested by trypsin, centrifuged and re-suspended in
0.5 mL of PBS, and analyzed using a flow cytometer (Beckman
cytoflex).
To further study the influence of redox-responsiveness on
intracellular delivery, DOX-loaded MPU micelles (SSPHPU or PHPU)
were incubated with HeLa cells for 24 h. Then the medium was
removed, and the cells were washed three times with PBS. The
samples were fixed, stained, mounted and tested as described
above.
2.6. MTT assay
The cytotoxicity of DOX-loaded micelles and blank micelles
against HeLa cells were evaluated in vitro by an MTT assay. HeLa
cells were incubated in 96-well plates at a density of 1  105 cells
per well in RPMI 1640 media for 24 h, followed by replacing the
media with 100 mL of micellar solutions with different concentra-
tions of DOX. The treated cells were incubated for 24 h and 72 h.
Then 20 mL of MTT solution (5 mg mL1) was added into each well.
After another 4 h of incubation, the medium was replaced by 200 mL
of DMSO. The plates were shaken for 10 min to dissolve the for-
mazan crystals. The absorption intensity at 490 nm was recorded
on a microplate reader (DNM-9602, Nanjing Perlove Medical
Equipment Co., Ltd., China).
2.7. In vitro MR imaging
The MRI experiments were carried out at 25
C using a 1.5 T
clinical MRI instrument (Siemens, Munich, Germany). The
following parameters were set in the measurement: repetition time
(TR) was 5000 ms, multiple echo times (TE) were 10, 20, 30, 40, 50,
60, 70, 80, 90, 100, 130, 150, 170 or 200 ms, slice thickness was
1.5 mm, field of view (FOV) was 150 mm  300 mm. The transverse
relaxation time (T2) was obtained from signal intensity in regions of
interest (ROI), and the relaxivity values were calculated by linear
least-squares fitting of 1/T2 versus the iron concentration (mM Fe).
2.8. Animals and tumor model
All animal procedures were approved by the Ethics Committee
of Sichuan University, in conformity with the Principles of Labo-
ratory Animal Care of the National Institutes of Health for the care
and use of laboratory animals. 5e6-week-old female athymic nude
mice were purchased from the Laboratory Animal Center of Sichuan
University (Sichuan, China). All the animals were kept under
pathogen-free conditions according to AAALAC guidelines and
acclimatized for 1 week prior to any experiments. Each mouse
received subcutaneous injections of 1  107 HeLa cells in the upper
flank of right hind leg. To evaluate magnetic-guided targeting, mice
were inoculated with two tumors in the left and right back of the
hind leg region.
2.9. In vivo MR imaging
When the tumor had grown to approximately 100 mm3 in
volume, the nude mice bearing HeLa tumors were intravenously
injected with SPION@SSPHPU-FA at a dose of 4.3 mg Fe kg1. Af-
terward, the right tumor was exposed to an external magnet
(Grade: N38) during the experiment while the left one was not. T2-
weighted MR images were acquired under 3% isoflurane anesthesia
before and after micelle administration for 1, 3 and 7 h using a
Bruker BioSpec 7.0 T MRI scanner (Bruker BioSpin, Billerica, MA,
USA). The mice were fixed with a surface coil in the MRI scanner.
MR images were obtained using a fast spin echo sequence (TURBO-
RARE) with the following parameters: TR/TE ¼ 1500/33 ms, slice
thickness ¼ 1 mm, flip angle ¼ 180
, matrix size ¼ 256  256 mm2,
FOV ¼ 50  50 mm2, NEX ¼ 4. T2-weighted datasets were processed
with ImageJ 1.50i (Wayne Rasband, National Institutes of Health,
USA). The signal intensity of tumor was normalized to that of
muscle at each time point. Relative MRI SI was defined as:
Normalized SI of tumor (post-injection)/ Normalized SI of tumor
(pre-injection)  100%, and T2 signal contrast effect enhancement
was defined as [Normalized SI of tumor (pre-injection) - Normal-
ized SI of tumor (post-injection)]/ Normalized SI of tumor (pre-
injection)  100%. Following MR imaging, mice were sacrificed and
tumors were removed and stained with Prussian blue for histo-
logical examination of iron distribution. Quantitative analysis of the
Prussian blue staining images was conducted with Image-Pro Plus
6.0 software (Media Cybernetics, Silver Spring, US) based on the
mean optical density.
2.10. In vivo antitumor effect and safety studies
When the average tumor volume approximately reached
30e50 mm3, nude mice bearing single HeLa tumor models were
randomly divided into four groups (n ¼ 5), and treated with free
DOX, DOX/SSPHPU-FA, and DOX/SSPU-FA at a DOX dose of
4 mg kg1 via tail vein every 3 d for 12 d. Mice injected with saline
solution were set as negative controls. To demonstrate magnetic
targeting, mice bearing two HeLa tumors were divided into two
groups, and each group was injected with DOX/SPION@SSPHPU-FA
and DOX/SPION@PHPU-FA, respectively, with an external magnet
(Grade: N38) placed on the right tumor for 1 h immediately after
every injection. The body weights and tumor growth of the mice
were monitored every 3 d. The length (A) and width (B) of tumors
were measured by a digital calliper, and the tumor volume was
calculated from the following equation: AB2/2. On the day of 15, all
the mice were sacrificed and the tumors were excised and weighed.
The tumor weight inhibition rate (TWI) was calculated through the
 J. Wei et al. / Biomaterials 145 (2017) 138e153
formula: (the average tumor weight of control group e the average
tumor weight of experiment group)/ the average tumor weight of
control group  100%.
Thereafter, the tumor tissues and main organs including heart,
liver, spleen, lung and kidney of the mice were harvested and used
for histological analysis. To analyze the tumor tissue apoptosis, the
terminal deoxynucleotidyl transferased dUTP nick end labeling
(TUNEL) assay was conducted. Briefly, the removed tumor tissues
were fixed in 4% paraformaldehyde solution and embedded and
sliced, which were then incubated with the TUNEL reaction
mixture freshly prepared according to the manufacturer's protocol
(Roche, Penzberg, Germany) for 60 min at 37
C in a humidified
atmosphere in the dark. The tissue sections were labeled with DAPI
and observed with fluorescence microscopy. The apoptotic rate of
tumor sections was calculated with Image-Pro Plus 6.0 software
(Media Cybernetics, Silver Spring, US) based on the mean optical
density measured from three random fields per slice. Furthermore,
the tumor tissues and the main organs were stained with hema-
toxylin and eosin (H&E) for histological examination.
To evaluate the safety of MPU formulations, the blood samples
(0.8 mL) were collected from each group of mice at the end of the
experiment. Aliquots of plasma were obtained by centrifugation at
3000 r min1 at 4
C for 10 min. The serum biochemical parameters
including alanine aminotransferase (ALT), blood urea nitrogen
(BUN) and creatinine kinase (CK) were assessed as function
markers of liver, kidney and heart, respectively. All these parame-
ters were measured with an IDEXX Biochemical Analyzer (West-
brook, ME, USA) using the kits according to the manufacturer's
instructions.
2.11. Statistics
The quantitative data collected were expressed as
means ± standard deviations (SD). Statistical analysis was per-
formed using the Statistical Package for the Social Sciences (IBM
SPSS Statistics software, Version 19, IBM, New York, USA). Student's
t-test or one-way analysis of variance (ANOVA) was conducted to
evaluate the statistical significance within the data at 95% confi-
dence levels (p < 0.05).
3. Results and discussion
3.1. Synthesis of clickable MPUs
A series of clickable MPUs with controlled responsiveness were
successfully synthesized from biodegradable PCL, L-lysine derivat-
ized LDI, cleavable BPEG bearing a pH-sensitive benzoic-imine
linkage, and a clickable and multifunctional chain extender
generated from L-cystine. The structure of the polymers is shown in
Fig. 1A. By appropriately varying the feed ratios, the amounts of
labile disulfides and benzoic-imine linkages in the polymers were
well adjusted. Typically, MPU containing both disulfide and
benzoic-imine linkages is named as SSPHPU, while those including
only benzoic-imine or reductive disulfide bonds are denoted as
PHPU or SSPU, respectively. The polymers possess moderate mo-
lecular weight (Mw 14600e18800), with relatively narrow molec-
ular weight distributions (Fig. S1). The structure of MPU was
verified by 1H NMR. As depicted in Fig. S2, the characteristic peaks
of PCL (3.94, 2.25, 1.52, 1.27 ppm), PEG (3.51e3.30 ppm) and LDI
(4.05, 1.15 ppm) can be observed for all the samples. The signals of
Cys-PA (3.13, 2.78 ppm) and BPEG (8.30e7.88 ppm) can be found in
the spectra of SSPU and PHPU, respectively, confirming that the
designed functional moieties have been successfully introduced
into the multiblock polymers.
143
3.2. Construction of clickable and imageable multiblock polymer
micelles
The prepared MPUs could self-assemble into a core-shell
nanostructure, where insoluble PCL soft segment constitutes a
hydrophobic core, surrounded by a cleavable corona formed by
BPEG. The hard segments comprising LDI and Cys-PA tend to
distribute on the subsurface of nanoparticles due to their relatively
higher polarity compared with PCL and the poor compatibility
between the soft segments and hard segments [34]. The nano-
vehicles appear as dispersed spherical particles with diameters
around 57 nm, as determined by DLS and TEM (Fig. 2A and B). The
formation of micelles was first confirmed by using the fluorescence
of pyrene as a probe. As the concentration of MPU increases, the
fluorescent intensity of pyrene rises, and the (0, 0) band shifts from
331 nm to 335 nm (Fig. S3). The result demonstrates the transfer of
pyrene molecules from a polar environment to a nonpolar micro-
environment of micellar core [35]. The micellar structure was also
characterized by 1H NMR. As seen in Fig. S4, the chemical shifts of
all the components are evident in DMSO-d6 (Fig. S4A), while those
of PCL are significantly weakened and those of BPEG increase
remarkably in D2O (Fig. S4B), suggesting the formation of core-shell
assemblies. To visually clarify the structure of polymer micelles,
computational simulation was carried out using a dissipative par-
ticle dynamics (DPD) model (Supporting Information). The results
present a spherical assembly structure with a PCL core and a BPEG
corona, and the hard segments were found distributed mainly on
the interface (Fig. 2A inset), which is in agreement with our pre-
vious finding [36]. Such an architecture is attractive for integration
of targeting and stealth properties into a single nanosystem. One
the one hand, the clickable alkyne sites located on the subsurface of
micelles enable post-conjugation of targeting ligands via a facile
click chemistry. On the other hand, the cleavable PEG corona could
provide temporary protection for the hard segments bearing di-
sulfide and targeting moieties against destabilization and clearance
from the body (Fig. 1).
To construct imageable multiblock polymer micelles, SPIONs
were prepared by a facile and efficient chemical co-precipitation
method and encapsulated into the micellar core through a dial-
ysis process. Due to the hydrophobic nature of oleic acid coating,
the SPIONs are well dispersed in hexane with spherical morphology
and an average size of 16.3 nm, as determined by DLS and TEM
(Fig. 2C and D). HRTEM image presents a single crystallite of SPION
(Fig. 2D inset and Fig. S5), where the lattice planes are clearly
visible. The measured d-spacing of 2.53 and 2.96 Å correspond to
the (311) and (220) planes of magnetite, respectively. After loading
into the core of polymer micelles, the SPIONs are transferred into
aqueous phase with high stability due to the surface protection
from amphiphilic polymers (Fig. 2C). The average hydrodynamic
particle size of SPION@micelles is 95 nm with a polydispersity in-
dex of 0.19 (Fig. 2C), which is nearly twice the diameter of blank
micelles, probably because the encapsulation of solid SPIONs in-
creases the volume of micellar core. The morphology of SPION@-
micelles observed by TEM is presented in Fig. 2D. The
SPION@micelles display a spherical shape consisting of Fe3O4
clusters, which is highly beneficial to the enhancement of imaging
contrast [37].
To further investigate the structure and composition of
SPION@micelles, FTIR, XRD and TGA analysis were conducted tak-
ing SPION and blank micelles as controls. As shown in Fig. S6, a
strong absorption at 563 cm1 in the FTIR spectra of SPION is
attributed to FeeO vibration in Fe3O4. This peak can be observed in
the spectra of SPION@micelles and appears greatly weakened,
which may be due to the encapsulation of SPIONs in polymer mi-
celles. Fig. S7 shows the TGA thermograms of SPIONs and
144 J. Wei et al. / Biomaterials 145 (2017) 138e153

Fig. 2. Construction and characterization of clickable and imageable MPU micelles. Size distributions (A, C, E) and TEM micrographs (B, D, F) of SSPHPU micelles (A, B), SPIONs
(C, D) and SPION@micelles (E, F). The bars in (B, D, F) are 100 nm. The inset in (A) shows a representative computational simulation image of MPU micelles. The insets in (C) and (E)
present SPIONs and magnetic micelles dispersed in hexane and water, respectively. The inset in D indicates HRTEM image of a single crystallite of SPION showing the (311) and (220)
lattice fringe at 2.53 and 2.96 Å, respectively (the bar is 2 nm). A high resolution figure of HRTEM image is depicted in Fig. S5.

SPION@micelles. The weight loss of 12% in the range of 200e400
C
is associated with the decomposition of oleic acid on Fe3O4. For
SPION@micelles, 75.7% weight loss in the range of 200e450
C can
be attributed to the degradation of oleic acid and MPUs (Fig. S7).
Based on the residual weight percentages, the SPION loading con-
tent in the magnetic micelles was calculated to be 27.6%
(250 mg mL1), with a loading efficiency near 60%. The XRD pattern
of SPION and SPION@micelles are depicted in Fig. S8. The SPION
exhibits diffraction peaks at 2q of 30.2
, 35.7
, 43.3
, 54.2
, 57.2
,
62.9
, 74.5
, corresponding to the (220), (311), (400), (422), (511),
(440), (533) reflection planes of the face-centered cubic (fcc) Fe3O4
crystal [38]. The XRD pattern of the magnetic loaded polymer mi-
celles is similar to that of SPIONs, further indicating that the pre-
pared micelles are composed of Fe3O4 nanoparticles.
3.3. Corona-detachment profiles of multiblock polymer micelles
The MPU micelles exhibit good responsiveness to tumor mi-
croenvironments. With the change of solution pH from 7.4 to 6.5,
the micellar diameter increases from 57 nm to approximately
130 nm, and the size distribution is significantly broadened
(Fig. 3A). Moreover, the particle surface turns somewhat rough
(Fig. 3B), which is consistent with a previous report [31]. This
phenomenon is expectable since corona detachment may decrease
the density of PEG on micellar surface, and thus reduce the steric
repulsive forces against van der Walls interactions between
exposed micellar cores [39]. As a result, a secondary aggregation
would be favored in order to minimize the interfacial energy [40].
The aggregation of polymer micelles and increase of particle size
 J. Wei et al. / Biomaterials 145 (2017) 138e153 145

were also visually proved by computational simulation results,
where a core-shell structure of micelles is still retained with a
larger diameter after partial detachment of outer corona (Fig. 3A
inset).
To further demonstrate the impact of corona-detachment on
micellar structure, the morphology change of SPION-loaded MPU
micelles in response to an extracellular acidic environment was
investigated. It was found that the size and size distribution of
SPION@micelles increase moderately after acid treatment (Fig. 3C).
Interestingly, the magnetic nanoparticles still form stable clusters
and remain well dispersed in an aqueous phase (Fig. 3D), further
confirming that the MPU micelles could maintain their core-shell
structure to accommodate SPIONs after partial shell-detachment.
This result is potentially helpful for nanomedicines to expose
their targeting ligands while avoiding premature drug release at
tumor sites.
To study the kinetics of corona-detachment, SSPHPU micelle
solution was transferred into a dialysis bag and incubated in a weak
acid media (pH 6.5) with shaking. As the labile benzoic-imine
bonds are cleaved in weak acid, the yielded BPEG residue, i.e.
MPEG-CBA, is capable of penetration through the dialysis mem-
branes, which shows a characteristic UV absorption at 256 nm and
allows for a facile quantitative determination with UVeVis spec-
trometer (Fig. 4A and Fig. S9). The determined cumulative PEG
concentration in acid release media increases remarkably over
time. By contrast, there is no noticeable PEG derivative detected in
neutral media (Fig. 4B). To further verify the corona-detachment,
the solutions outside and inside the dialysis bag were collected
and lyophilized for 1HNMR analysis. As shown in Fig. 4C, the
resonance peaks corresponding to the protons of PEG (3.51,eCH2e)
and benzene ring (7.80e8.10,eCHe) are clearly observed in the
1
HNMR spectra of release media, while these signals decline or
even disappear for samples inside the dialysis bag (Figs. S4 and
S10). These results demonstrate that the MPU micelles display
high stability under physiological conditions and pH-triggered
shell-detachment behavior in response to tumor extracellular
environment.
3.4. Triggered release of payloads from multiblock polymer micelles
Since the change of micellar structure can affect the drug release
behavior, we investigated the release of a model drug, DOX, from
MPU micelles under different stimuli in vitro. The micelles can
encapsulate DOX efficiently, with a loading content up to 23%
(230 mg mL1). The DOX-loaded SSPHPU micelles display a slow and
sustained drug release rate in PBS solution (pH 7.4), with approx-
imately 40% of the total DOX released within 96 h. While in a weak
acid PBS solution (pH 6.5) mimicking the extracellular environ-
ment, the release of DOX from SSPHPU micelles increases slightly
(Fig. 4D), owning to the detachment of PEG corona and increase of
particle diameter (Fig. 3). Nevertheless, the total amounts of
released DOX is still less than 50%. A possible reason is that the
micelles may preserve a core-shell architecture to accommodate
drugs after partial detachment of outer corona, as has been
demonstrated above. This phenomenon is advantageous to evasion
of premature extracellular release of therapeutics [29].
On the other hand, the release of DOX from SSPHPU in the
presence 10 mM of GSH was found accelerated mildly (P ¼ 0.053),

Fig. 3. Corona-detachment property of MPU micelles. Size distributions (A, C) and TEM micrographs (B, D) of SSPHPU micelles (A, B) and SPION@micelles (B, D) after incubation at
pH 6.5 for 24 h. The bars in both TEM images are 100 nm. The inset in (A) shows a representative computational simulation image of MPU micelles after partial corona detachment.
The insets in (C) presents SPION@micelles dispersed in hexane and water after acidic treatment.
146 J. Wei et al. / Biomaterials 145 (2017) 138e153

Fig. 4. Corona-detachment and triggered release profiles of MPU micelles. (A) UVevis spectra of MPEG-CBA in aqueous solutions at different concentrations. (B) Time-
dependent cumulative release of PEG outer corona from SSPHPU micelles at different pH values. (C) 400 MHz 1H NMR spectrum of lyophilized acidic release media in DMSO-
d6. (D) The release profiles of DOX from SSPHPU micelles in different media with or without acidic (pH 6.5) and reductive (10 mM GSH) stimuli.

with nearly 65% of drug released in 96 h (Fig. 4D). Such an
enhancement of drug release should not be as significant as ex-
pected, considering the dramatic increase of micellar diameter in
response to GSH [31] and relatively higher responsivity reported for
other disulfide-based drug delivery systems [41]. From a structural
perspective, the disulfide bonds are mainly distributed in the hy-
drophobic hard segments of PUs, which are further shielded by the
soft segments comprising PEG and PCL during assembly. The
presence of soft segments may promote a steric repulsion against
the penetration of GSH and inhibit the responsive efficiency of
polymers [42]. In addition, the steric hindrance of propargyl amide
groups should also be taken in to account [43]. This hypothesis
could be partially confirmed by GPC test, where SSPHPU does not
show significant decrease in molecular weight after GSH treatment,
whereas it can be greatly degrade as acid and GSH exist simulta-
neously (Fig. S11). To further understand the result, the release
experiment was conducted in an acid media containing 10 mM of
GSH. As anticipated, the release of DOX from SSPHPU was greatly
accelerated (P ¼ 0.003), with almost 100% of the loaded drugs
released (Fig. 4D).
In addition, the release of SPIONs from MPU micelles was also
evaluated using a turbidity measurement, considering the poor
permeability of solid Fe3O4 nanoparticles through the dialysis bag.
As seen from Fig. S12, the absorption of SPION@SSPHPU decreases
moderately under acidic or reductive conditions. By contrast, in the
presence of both acid and GSH stimuli, the turbidity drops
dramatically and visible participates appear at the bottom of vials
(Fig. S12). This result is in good agreement with DOX release pro-
files, suggesting a high specificity of SPION release from MPU mi-
celles in responsive to intracellular microenvironments. Since the
possibility of disulfide breakage occurred in the blood circulation
has been noticed elsewhere [24,25], the “PEG switch” strategy is
promising for remitting undesirable polymer degradation and drug
leakage on the way of nanomedicines to tumor site, and triggering
the cleavage of redox-sensitive moieties and release of payloads
within tumor cells. To better understand the pH-activated
responsiveness and release behavior, further work is needed and
being carried out in our group.
3.5. Cellular uptake of FA-clicked MPU micelles
To improve the specificity of clickable MPU micelles, FA as a
model targeting ligand was chemically attached onto the micellar
interface via a copper catalyzed alkyne-azide cycloaddition
(CuAAC). The FA-conjugated MPU micelles (SSPHPU-FA) hold an FA
amount of 3.1 w/w%, as determined by UVeVis spectrometer. The
critical micelle concentration (CMC) of MPU increases slightly after
FA conjugation (Fig. S13). To investigate the effect of “PEG-switch”
on the interaction between MPU micelles and cancer cells, the DOX-
loaded targeted and untargeted formulations were incubated with
HeLa cells, and the cellular uptake of nanocarriers was determined
using CLSM and flow cytometry. As shown in Fig. 5, SSPHPU mi-
celles exhibit a low degree of internalization into tumor cells
cultured at pH 7.4, with weak fluorescence observed predomi-
nantly in the cytoplasm after 4 h of incubation (Fig. 5A, E). After
conjugation of FA, the nanocarriers show a mild increase of uptake,
owning to folate receptor-mediated endocytosis causing more
micelles to be taken up by tumor cells [44]. However, it is note-
worthy that such a promotion of cellular uptake is relatively lower
than those attained for other FA-based nanomedicines [45]. We
 J. Wei et al. / Biomaterials 145 (2017) 138e153 147

previously speculate that the short spacer of FA residues and the
long chain PEG corona should be responsible for the decreased
activity [31]. Herein, to validate this hypothesis, the cellular uptake
experiments were carried out under a simulated extracellular
acidic conditions (pH 6.5). Remarkably, the targeted MPU micelles
(SSPHPU-FA) present significantly increased DOX fluorescence in
tumor cells (Fig. 5D and E). This result could be explained by the
fact that the deshielding of PEG shell can expose FA-decorated
micellar surface to receptor-enriched cells, thus turning the tar-
geting switch “on” and facilitating subsequent cell internalization

Fig. 5. Cellular uptake of FA-clicked MPU micelles. CLSM images of HeLa cells incubated with SSPHPU-DOX (A, B) and FA-clicked MPU micelles (SSPHPU-FA-DOX) (C, D) at pH 7.4
(A, C), and pH 6.5 (B, D) for 4 h. Nuclei of cells were stained with DAPI. The scale bars are 25 mm. (E) The mean SI of DOX fluorescence in cytoplasm and nucleus calculated and
normalized with an ImageJ Software. Statistical significance: *P < 0.05; **P < 0.01; ***P < 0.005. (F) Flow cytometry histograms of HeLa cells incubated with various formulations,
taking untreated cells as controls. (a), (b), (c) and (d) in (E, F) represent cells treated with SSPHPU-DOX at pH 7.4, SSPHPU-DOX at pH 6.5, SSPHPU-FA-DPX at pH 7.4, and SSPUPU-FA-
DOX at pH 6.5, respectively.
148 J. Wei et al. / Biomaterials 145 (2017) 138e153

(Fig. 1E). On the other hand, unconjugated polymer nanocarriers
(SSPHPU) are also able to remove their corona in acid media,
nonetheless, the lack of targeting groups results in an insufficient
uptake, as evidenced by the negligible enhancement of fluores-
cence detected in tumor cells (Fig. 5B). The same phenomenon was
also found in flow cytometry result, where the cells treated by
SSPHPU-FA under pH 6.5 display the highest fluorescence intensity
(Fig. 5F). These results demonstrate that the targeting-clicked MPU
nanovehicles are excellent candidates for achieving on-demand
tumor cell targeting and controlled drug delivery.
3.6. Intracellular drug delivery and antitumor activity
It is known that tumor cells develop unique microenvironments
such as slightly acidic (pH 4e6) [46] and markedly reductive (up to
10 mM of GSH) intracellular space [47]. Therefore, the clickable

Fig. 6. Intracellular drug delivery and antitumor activity in vitro. CLSM images of HeLa cells incubated with DOX-loaded redox-responsive SSPHPU (B) and insensitive PHPU (A)
for 24 h. Nuclei of cells were stained with DAPI. The scale bars are 25 mm. (C) The mean SI of DOX fluorescence in cytoplasm and nucleus calculated and normalized with an ImageJ
Software. Cytotoxicity of redox-responsive SSPHPU-DOX and insensitive MPU micelles against HeLa cells for 24 h (D) and 72 h (E) of incubation. Free DOX was used as a positive
control. (F) Cytotoxicity of SPION@micelles without drug loading at different concentrations.
 J. Wei et al. / Biomaterials 145 (2017) 138e153 149

multifunctional PU micelles equipped with a “PEG switch” are
promising for triggered breakage of disulfide bonds and maximal
release of payloads within tumor cells. To prove this potential, the
MPU formulations were incubated with HeLa cells for 24 and 72 h.
As seen in Fig. 6AeC, the cells incubated with redox-responsive
SSPHPU show a strong DOX fluorescence located in the nucleus.
In contrast, the signal in cells cultured with insensitive PHPU
formulation was detected mainly in the cytoplasm. This phenom-
enon can be attributed to the triggered release of DOX in response
to the acidity and GSH inside tumor cells. The free drugs released
from the carriers could be readily transported into the cell nucleus,
where it can interfere the process of DNA replication and cause cell
death [48]. With this in mind, we further assessed the antitumor
effects of DOX-loaded MPU micelles using an MTT assay. It was
found that all the formulations display a time and dose dependent
cytotoxicity toward HeLa cells (Fig. 6D and E). Apparently, the drug
efficacy of SSPHPU micelles is significantly higher compared with
insensitive PHPU and SSPU formulations (P < 0.05), with IC50 (50%
inhibitory concentration) values of 2.2 mg mL1 and 1.7 mg mL1
after incubation for 24 h and 72 h, respectively, which are two fold
lower than those of PHPU (13.0 and 4.0 mg mL1) and SSPU (4.6 and
3.1 mg mL1). Besides, the blank SPION@micelles do not show
significant toxic effect at different concentrations (Fig. 6F). These
results demonstrate that the MPU micelles with switchable release
capacity hold great promise in multifunctional intracellular drug
delivery applications.
3.7. Magnetic property and MRI contrast effects of MPU micelles
The superparamagnetic property is important to biomedical
applications of SPION@micelles. Under an external magnetic field,
the SPION-loaded micelles can be manipulated and guided to
specific sites (Fig. 7A). On the other hand, the SPION@micelles lose
magnetism and re-disperse when the magnetic field is turned off.
This phenomenon can be exploited in magnetic targeted drug de-
livery, cell sorting and separation, as well as imaging agents for
early detection of several pathologies. The magnetic hysteresis
loops of SPIONs and SPION@micelles were determined using a
vibrating-sample magnetometer. As shown in Fig. 7A, both
SPION@micelles and SPIONs are superparamagnetic at room tem-
perature without magnetic remnants observed at a small applied
magnetic field. The saturation magnetization value of SPION-
loaded micelles is 21.0 emu g1, smaller than that of Fe3O4 nano-
particles (80 emu g1). This can be attributed to the nonmagnetic

Fig. 7. Magnetic property and MR imaging of MPU micelles in vitro and in vivo. (A) Magnetic hysteresis loops of SPIONs and SPION@micelles. The insets show magetic loaded
MPU micelles in response to an external magnetic field for 0 h (upper) and 6 h (lower). (B) T2-weight MRI images of SPIONs (I) and SPION@micelles (II) recorded on a 1.5 T clinical
MRI instrument at different Fe concentrations (mM): a, 0; b, 0.05; c, 0.1; d, 0.2; e, 0.3; f, 0.4; g, 0.5. (C) T2 relaxation rates as a function of Fe concentrations. In vivo T2-weighed MR
images of HeLa tumor-bearing mice before (D) and after intravascular injection of MPU micelles loading with SPIONs for (E) 1 h, (F) 3 h, and (G) 7 h. (H) Optical images of a
representative mouse inoculated with two tumors for MRI. The tumors in the left and right flanks are marked with green dots and red circles, respectively. The magnetic force was
applied to the right tumor. (I) Relative T2 signal intensity of tumors as a function of time. The standard deviation of signal intensity is derived from five random ROI in tumor
sections. Statistical significance: *P < 0.05; **P < 0.01; ***P < 0.005. Histological analysis of the right tumor (J) and the left tumor (K) after Prussion blue staining. The scale bars are
10 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
150 J. Wei et al. / Biomaterials 145 (2017) 138e153

polymer coating reducing the total magnetization. However, the
absence of coercivity and remanence reveals that SPIONs remain
their superparamagnetic nature after encapsulation into MPU
micelles.
MR imaging is one of the most useful clinical diagnostic tech-
niques in providing non-invasive and real-time detection of various
diseases including tumors. The capabilities of MRI have been
greatly enhanced by the introduction of SPIONs, which are known
to shorten the transverse (spin-spin) relaxation time (T2) of water
protons and can exert negative contrast in the interfered area [49].
To evaluate the MRI contrast effect rendered by SPION@micelles,
the transverse relaxation rates (1/T2) and T2-weighted MR images
of SPION@micelles with different iron concentrations were ob-
tained on a clinical 1.5 T MRI instrument. As presented in Fig. 7B

Fig. 8. Multimodal targeting and antitumor effect in vivo. (A) Optical images of representative mice inoculated with two tumors, where the left and right tumors are marked
with green dots and red circles, respectively. The right tumors were exposed to an external magnetic field for magnetically guided therapy (as shown by red curves). (B) Growth
curves of the left tumors (a, c) and right tumors (b, d) after intravenous injection of DOX/SPION@SSPHPU (a, b) and DOX/SPION@SSPU in HeLa tumor-bearing mice (c, d) at a DOX
dose of 4 mg kg1. (C) The average weights of tumors separated from animals receiving different treatments. The insets show representative images of tumors. Statistical sig-
nificance: *P < 0.05; **P < 0.01; ***P < 0.005. (D) Ex vivo histological analyses of tumor sections after treatment with DOX/SPION@SSPHPU. H&E staining (a, b) and TUNEL
immunofluorescence staining (c, d) of slices from the left tumors (a, c) and right tumors (b, d). In TUNEL analysis, the nuclei were labeled blue, and the apoptotic cells were stained
green. The scale bars are 500 mm. (E) H&E staining analysis of the major organs (heart, liver, spleen, lung, kidney) of the mice after different treatments. The scale bars are 100 mm.
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
 J. Wei et al. / Biomaterials 145 (2017) 138e153
and C, the signal intensity of T2-weighted MR images deceases with
increasing of iron concentration. Such a concentration-dependent
darkening phenomenon demonstrates a negative contrast
enhancement in MRI. Particularly, the magnetic micelles display
higher efficiency in signal darkening than SPIONs (Fig. 7B). As an
indicator of the ability of contrast agent to shorten T2, the specific
relaxivity rate (r2) of SPION@micelles (89.5 mM1 s1) is much
higher than that of SPIONs (54.6 mM1 s1) (Fig. 7C), owning to the
fact that the physical encapsulation of clustered SPIONs into
confined polymeric micelles can increase the T2 relaxivity and
contrast effect of MRI [37,50]. This clustering effect has also been
demonstrated in many other carrier systems [51e53].
3.8. Magnetically guided targeting and MR imaging in vivo
The incorporation of magnetic nanoparticles is beneficial for
magnetically targeted drug delivery as well as excellent contrast
effect for MR imaging [11]. To evaluate the MDT and MRI ability of
MPU formulations, nude mice were inoculated with two tumors in
the left and right back of the hind leg region, and administrated
with SPION-loaded SSPHPU-FA micelles through the tail vein at a
dose of 4.3 mg Fe kg1, with the right tumor exposed to an external
magnetic field after injection. The T2-weighted axial MR images of
the mice were acquired on a 7.0 T MRI instrument before and after
treatment at different time points. As shown in Fig. 7DeG,
remarkable signal attenuation was observed at both the left and
right tumors after 1 h, and the MRI signal intensity further de-
creases over time. In particular, compared with the left tumors with
passive EPR effect and FA-mediated active targeting for MPU de-
livery, the right tumors with additional magnetic exposure exhibit
much faster and higher degree of signal darkening. The T2 signal
contrast effect enhancement at 1 h in the right tumors (60%) is
more than two times higher than that in the left tumors (27%)
(Fig. 7I).
To confirm the accumulation of MPU formulations and assess
the concentration of intratumoral Fe, the excised tumor tissues
were stained with Prussian blue for observation. As indicated in
Fig. 7J, numerous dark blue regions are observed in the magneti-
cally treated tumors, in good agreement with the MR images. By
contrast, the contralateral tumors without magnet exposure
display much weaker blue signals (Fig. 7K). Quantitative analysis
reveals that the Prussian signal intensity in the right tumors is
approximately 4.5 times higher than that in the left tumors. These
results demonstrate that magnetically-guided delivery of MPUs
into tumors could largely complement the targeting efficacy
mediated by EPR effect, and show great potential to overcome the
tumor heterogeneity and complicated physiological environment.
Further advances in the development of high-strength magnetic
field gradients and novel magnetic nanocomposites with multi-
modal contrast capacity would enable the manipulation of MPU
nanomedicines at higher depths with greater efficiency [9,54].
3.9. Multimodal targeting and antitumor effect in vivo
To investigate the targeting effect and antitumor efficacy of MPU
formulations in vivo, nude mice bearing HeLa tumors were treated
with DOX loaded SSPHPU-FA and SSPU-FA micelles, taking free
DOX and saline as controls. It was found that both DOX@SSPHPU-FA
and DOX@SSPU-FA show remarkable inhibitory effects on tumor
growth in comparison with saline solution after intravenous in-
jection (Fig. S14). Particularly, the TWI of DOX@SSPHPU-FA (89.7%)
is much higher than that of DOX@SSPU-FA (57.9%). The results
confirms the significant role of “PEG switcher” in SSPHPU-FA,
which enables improved cellular drug targeting and effective
anticancer activity (Fig. 1E).
151
To further test magnetic guided targeting and therapy in vivo,
SPIONs and DOX were co-encapsulated into SSPHPU-FA and SSPU-
FA micelles, and injected into the nude mice bearing two tumors. As
expected, a much greater suppression on the growth of right tu-
mors was observed for both SSPU-FA and SSPHPU-FA formulations,
with the aid of an external magnet (Fig. 8A and B). Moreover, the
PEG-switched targeting effect is still obvious in the presence of
MDT, as evidenced by the higher therapeutic efficacy of DOX/
SPION@SSPHPU-FA compared with that of SSPU-FA formulation
(Fig. 8). Excitingly, a synergistic targeting effect could be antici-
pated for PEG-triggered active targeting in combination with
magnetically-guided targeting, considering the excellent thera-
peutic efficacy of FA-clicked SSPHPU micelles toward the right tu-
mor with magnetic exposure (TWI is up to 98.6%, Fig. 8C).
The enhanced anticancer activity of MPUs was further evaluated
by histological analysis with H&E and TUNEL staining. It was
observed that the DOX and SSPU-FA groups exhibit moderate
cancer cell remission in comparison with saline treated groups
(Fig. S14). The sections from targeted SSPHPU-FA group display
greater extent of necrosis, and the area of the damaged regions was
found much larger for right tumors receiving magnetic treatment
(Fig. 8D). Furthermore, the quantitative analysis of apoptotic cells
by TUNEL staining assay also confirms the above observations. The
percentage of apoptotic cells in the tumor treated by DOX/
SPION@SSPHPU-FA with magnetic treatment is about 92.1%,
which is much higher than those of other groups without MDT
(54.0%) and PEG trigger (26.2%) (Fig. S14). In view of the exciting
delivery efficiency and therapeutic effect, further study using a
single injection is warranted to demonstrate the potential of MPU
micelles.
It is worth noting that no mice died during the treatment period
due to the relative low dose (Fig. S15), nonetheless, severe weight
loss was noticed in mice treated with DOX (Fig. S14B), indicating
systemic toxicity of free drug without any targeting ability. To
determine the potential toxic effect, we further tested the
biochemistry parameters ALT, BUN and CK as function markers of
liver, kidney and heart, respectively. Even though all the parameters
measured fell within the normal range (Fig. S16), the level of liver
enzyme ALT was found much higher in mice receiving DOX treat-
ment, compared with other groups (Fig. S16A). Furthermore, cell
nuclei shrinkage was observed by H&E staining of liver tissues
(Fig. 8E), revealing moderate hepatotoxicity in the mice treated
with free DOX. In contrast, no significant dropping in body weights,
change of biochemistry parameters and abnormality of major or-
gans was discovered in MPU groups during the experiment (Fig. 8E,
Figs. S14 and S16), suggesting that the MPU formulations are
favorably biocompatible. These results demonstrate that the click-
able and imageable polymer micelles with smart switches and
multimodal synergistic targeting ability would provide a novel
route for improving the specificity, efficacy and safety of cancer
theranosis. To further evaluate the efficacy and safety of MPU mi-
celles in vivo, dose response experiments and drug analysis in
different organs are being carried out in our laboratory.
4. Conclusion
In summary, a rationally designed clickable and imageable MPU
was successfully constructed for on-demand tumor targeting and
controlled drug release. The polymeric nanovehicles possess
attractive core-shell architecture, appropriate size, high loading
capacity for DOX and SPIONs and clickable sites for functionaliza-
tion. Furthermore, they present time-dependent detachment of
outer corona under extracellular acidic condition, which can switch
on the clicked FA ligands for activated tumor targeting and cellular
uptake, followed by facilitating the cleavage of reducible core
152 J. Wei et al. / Biomaterials 145 (2017) 138e153
segments as well as triggered drug release within tumor cells. In
addition, the versatile nanoformulations demonstrate not only
potent antitumor activity and effective MRI contrast enhancement
in vitro, but also magnetically guided and tumor-activated syner-
gistic targeting ability in vivo, thus resulting in precise anticancer
therapy and specifically enhanced MR imaging. Our work provides
a promising platform for the development of smart theranostic
systems for potential imaging-guided cancer treatment and real-
time detection of diseases.
Acknowledgment
This work was supported by the National Natural Science
Foundation of China (21474064, 51203101); the National Science
Fund for Distinguished Young Scholars of China (51425305); the
Project of State Key Laboratory of Polymer Materials Engineering
(sklpme2015-3-02); and the Youth Science and Technology Inno-
vation Team of Sichuan Province (2015TD0001).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.biomaterials.2017.08.005.
References
[1] a) F. Alexis, E. Pridgen, L.K. Molnar, O.C. Farokhzad, Factors affecting the
clearance and biodistribution of polymeric nanoparticles, Mol. Pharm. 5
(2008) 505e515;
b) F.M. Kievit, M. Zhang, Cancer nanotheranostics: improving imaging and
therapy by targeted delivery across biological barriers, Adv. Mater. 23 (2011)
H217eH247.
[2] a) O.C. Farokhzad, R. Langer, Impact of nanotechnology on drug delivery, ACS
Nano 3 (2009) 16e20;
b) J.A. Barreto, W. O Malley, M. Kubeil, B. Graham, H. Stephan, L. Spiccia,
Nanomaterials: applications in cancer imaging and therapy, Adv. Mater. 23
(2011) H18eH40;
c) A. Schroeder, D.A. Heller, M.M. Winslow, J.E. Dahlman, G.W. Pratt, R. Langer,
et al., Treating metastatic cancer with nanotechnology, Nat. Rev. Cancer 12
(2012) 39e50;
d) Y. Zhang, H.F. Chan, K.W. Leong, Advanced materials and processing for
drug delivery: the past and the future, Adv. Drug Deliv. Rev. 65 (2013)
104e120;
e) H. Hosoya, A.S. Dobroff, W.H.P. Driessen, V. Cristini, L.M. Brinker,
F.I. Staquicini, et al., Integrated nanotechnology platform for tumor-targeted
multimodal imaging and therapeutic cargo release, Proc. Nat. Acad. Sci. U. S.
A. 113 (2016) 1877e1882.
[3] a) C.E. Ashley, E.C. Carnes, G.K. Phillips, D. Padilla, P.N. Durfee, P.A. Brown, et
al., The targeted delivery of multicomponent cargos to cancer cells by nano-
porous particle-supported lipid bilayers, Nat. Mater. 10 (2011) 389e397;
b) X. Guo, X. Wei, Y. Jing, S. Zhou, Size changeable nanocarriers with nuclear
targeting for effectively overcoming multidrug resistance in cancer therapy,
Adv. Mater. 27 (2015) 6450e6456.
[4] a) Y.J. Wang, S.M. Hussain, G.P. Krestin, Superparamagnetic iron oxide contrast
agents: physicochemical characteristics and applications in MR imaging, Eur.
Radiol. 11 (2001) 2319e2331;
b) P. Reimer, T. Balzer, Ferucarbotran (Resovist): a new clinically approved
RES-specific contrast agent for contrast-enhanced MRI of the liver: properties,
clinical development, and applications, Eur. Radiol. 13 (2003) 1266e1276.
[5] a) J. Xie, S. Lee, X. Chen, Nanoparticle-based theranostic agents, Adv. Drug
Deliv. Rev. 62 (2010) 1064e1079;
b) X. Wang, D. Niu, Q. Wu, S. Bao, T. Su, X. Liu, et al., Iron oxide/manganese
oxide co-loaded hybrid nanogels as pH-responsive magnetic resonance
contrast agents, Biomaterials 53 (2015) 349e357.
[6] H. Maeda, H. Nakamura, J. Fang, The EPR effect for macromolecular drug de-
livery to solid tumors: improvement of tumor uptake, lowering of systemic
toxicity, and distinct tumor imaging in vivo, Adv. Drug Deliv. Rev. 65 (2013)
71e79.
[7] a) U. Prabhakar, H. Maeda, R.K. Jain, E.M. Sevick-Muraca, W. Zamboni,
O.C. Farokhzad, et al., Challenges and key considerations of the enhanced
permeability and retention effect for nanomedicine drug delivery in oncology,
Cancer Res. 73 (2013) 2412;
b) H. Maeda, Toward a full understanding of the EPR effect in primary and
metastatic tumors as well as issues related to its heterogeneity, Adv. Drug
Deliv. Rev. 91 (2015) 3e06.
[8] Y. Huang, S. He, W. Cao, K. Cai, X. Liang, Biomedical nanomaterials for
imaging-guided cancer therapy, Nanoscale 4 (2012) 6135e6149.
[9] J. Park, N.R. Kadasala, S.A. Abouelmagd, M.A. Castanares, D.S. Collins, A. Wei, et
al., Polymereiron oxide composite nanoparticles for EPR-independent drug
delivery, Biomaterials 101 (2016) 285e295.
[10] H. Jeon, J. Kim, Y.M. Lee, J. Kim, H.W. Choi, J. Lee, et al., Poly-paclitaxel/
cyclodextrin-SPION nano-assembly for magnetically guided drug delivery
system, J. Control. Release 231 (2016) 68e76.
[11] W. Chen, G. Luo, Q. Lei, F. Cao, J. Fan, W. Qiu, et al., Rational design of
multifunctional magnetic mesoporous silica nanoparticle for tumor-targeted
magnetic resonance imaging and precise therapy, Biomaterials 76 (2016)
87e101.
[12] a) Y. Chen, K. Ai, J. Liu, X. Ren, C. Jiang, L. Lu, Polydopamine-based coordination
nanocomplex for T1/T2 dual mode magnetic resonance imaging-guided
chemo-photothermal synergistic therapy, Biomaterials 77 (2016) 198e206;
b) J. Chen, M. Shi, P. Liu, A. Ko, W. Zhong, W. Liao, et al., Reducible
polyamidoamine-magnetic iron oxide self-assembled nanoparticles for
doxorubicin delivery, Biomaterials 35 (2014) 1240e1248.
[13] M.M. Gottesman, T. Fojo, S.E. Bates, Multidrug resistance in cancer: role of
ATP-dependent transporters, Nat. Rev. Cancer 2 (2002) 48e58.
[14] a) Y. Yan, G.K. Such, A.P.R. Johnston, J.P. Best, F. Caruso, Engineering particles
for therapeutic delivery: prospects and challenges, ACS Nano 6 (2012)
3663e3669;
b) S. Lee, H. Chen, T.V. O Halloran, S.T. Nguyen, “Clickable” polymer-caged
nanobins as a modular drug delivery platform, J. Am. Chem. Soc. 131 (2009)
9311e9320.
[15] C. Shi, X. Guo, Q. Qu, Z. Tang, Y. Wang, S. Zhou, Actively targeted delivery of
anticancer drug to tumor cells by redox-responsive star-shaped micelles,
Biomaterials 35 (2014) 8711e8722.
[16] S. Wang, P. Huang, X. Chen, Stimuli-responsive programmed specific targeting
in nanomedicine, ACS Nano 10 (2016) 2991e2994.
[17] a) K. Knop, R. Hoogenboom, D. Fischer, U.S. Schubert, Poly(ethylene glycol) in
drug delivery: pros and cons as well as potential alternatives, Angew. Chem.
Int. Ed. 49 (2010) 6288e6308;
b) X. Du, J. Wang, W. Liu, J. Yang, C. Sun, R. Sun, et al., Regulating the surface
poly(ethylene glycol) density of polymeric nanoparticles and evaluating its
role in drug delivery in vivo, Biomaterials 69 (2015) 1e11.
[18] a) E. Gullotti, Y. Yeo, Extracellularly activated nanocarriers: a new paradigm of
tumor targeted drug delivery, Mol. Pharm. 6 (2009) 1041e1051;
b) X. Yang, J. Du, S. Dou, C. Mao, H. Long, J. Wang, Sheddable ternary nano-
particles for tumor acidity-targeted siRNA delivery, ACS Nano 6 (2011)
771e781.
[19] S. Mura, J. Nicolas, P. Couvreur, Stimuli-responsive nanocarriers for drug de-
livery, Nat. Mater. 12 (2013) 991e1003.
[20] a) D. Roy, J.N. Cambre, B.S. Sumerlin, Future perspectives and recent advances
in stimuli-responsive materials, Prog. Polym. Sci. 35 (2010) 278e301;
b) Y. Lu, A.A. Aimetti, R. Langer, Z. Gu, Bioresponsive materials, Nat. Rev.
Mater. 1 (2016) 16075.
[21] F. Meng, W.E. Hennink, Z. Zhong, Reduction-sensitive polymers and bio-
conjugates for biomedical applications, Biomaterials 30 (2009) 2180e2198.
[22] H. Wei, R. Zhuo, X. Zhang, Design and development of polymeric micelles with
cleavable links for intracellular drug delivery, Prog. Polym. Sci. 38 (2013)
503e535.
[23] a) J. Li, X. Yu, Y. Wang, Y. Yuan, H. Xiao, D. Cheng, et al., A reduction and pH
dual-sensitive polymeric vector for long-circulating and tumor-targeted
siRNA delivery, Adv. Mater. 26 (2014) 8217e8224;
b) M. Huo, J. Yuan, L. Tao, Y. Wei, Redox-responsive polymers for drug de-
livery: from molecular design to applications, Polym. Chem. 5 (2014)
1519e1528.
[24] G.K. Such, Y. Yan, A.P.R. Johnston, S.T. Gunawan, F. Caruso, Interfacing mate-
rials science and biology for drug carrier design, Adv. Mater. 27 (2015)
2278e2297.
[25] L. Brülisauer, M.A. Gauthier, J. Leroux, Disulfide-containing parenteral delivery
systems and their redox-biological fate, J. Control. Release 195 (2014)
147e154.
[26] a) X. Guo, C. Shi, G. Yang, J. Wang, Z. Cai, S. Zhou, Dual-responsive polymer
micelles for target-cell-specific anticancer drug delivery, Chem. Mater. 26
(2014) 4405e4418;
b) X. Wang, J. Zhang, R. Cheng, F. Meng, C. Deng, Z. Zhong, Facile synthesis of
reductively degradable biopolymers using cystamine diisocyanate as a
coupling agent, Biomacromolecules 17 (3) (2016) 882e890.
[27] M. Ding, J. Li, H. Tan, Q. Fu, Self-assembly of biodegradable polyurethanes for
controlled delivery applications, Soft Matter. 8 (2012) 5414e5428.
[28] a) Y. Wang, G. Wu, X. Li, Y. Wang, H. Gao, J. Ma, On-off switchable drug release
from multi-responsive degradable poly(ether urethane) nanoparticles, Bio-
mater. Sci. 1 (2013) 614e624;
b) S. Yu, J. Ding, C. He, Y. Cao, W. Xu, X. Chen, Disulfide cross-linked poly-
urethane micelles as a reduction-triggered drug delivery system for cancer
therapy, Adv. Healthc. Mater. 3 (2014) 752e760;
c) F. Huang, R. Cheng, F. Meng, C. Deng, Z. Zhong, Micelles based on acid
degradable poly(acetal urethane): preparation, pH-sensitivity, and triggered
intracellular drug release, Biomacromolecules 16 (2015) 2228e2236;
e) M. He, A. Potuck, J.C. Kohn, K. Fung, C.A. Reinhart-King, C. Chu, Self-
assembled cationic biodegradable nanoparticles from pH-responsive amino-
acid-based poly(ester urea urethane)s and their application as a drug delivery
vehicle, Biomacromolecules 17 (2016) 523e537.
[29] a) M. Ding, N. Song, X. He, J. Li, L. Zhou, H. Tan, et al., Toward the next-
 J. Wei et al. / Biomaterials 145 (2017) 138e153
generation nanomedicines: design of multifunctional multiblock poly-
urethanes for effective cancer treatment, ACS Nano 7 (2013) 1918e1928;
b) M. Ding, J. Li, X. He, N. Song, H. Tan, Y. Zhang, et al., Molecular engineered
super-nanodevices: smart and safe delivery of potent drugs into tumors, Adv.
Mater. 24 (2012) 3639e3645.
[30] a) M. Ding, X. Zeng, X. He, J. Li, H. Tan, Q. Fu, Cell internalizable and intra-
cellularly degradable cationic polyurethane micelles as a potential platform
for efficient imaging and drug delivery, Biomacromolecules 15 (2014)
2896e2906;
b) M. Ding, J. Wei, R. Wang, X. Shuai, J. Li, H. Tan, et al., Multifunctional
cystine-installed clickable micelles for efficient drug delivery and magnetic
resonance imaging, J. Control. Release 259 (2017) e96.
[31] N. Song, M. Ding, Z. Pan, J. Li, L. Zhou, H. Tan, et al., Construction of targeting-
clickable and tumor-cleavable polyurethane nanomicelles for multifunctional
intracellular drug delivery, Biomacromolecules 14 (2013) 4407e4419.
[32] M. Ding, J. Li, X. Fu, J. Zhou, H. Tan, Q. Gu, et al., Synthesis, degradation, and
cytotoxicity of multiblock poly(ε-caprolactone urethane)s containing gemini
quaternary ammonium cationic groups, Biomacromolecules 10 (2009)
2857e2865.
[33] X. Li, H. Li, G. Liu, Z. Deng, S. Wu, P. Li, et al., Magnetite-loaded fluorine-
containing polymeric micelles for magnetic resonance imaging and drug de-
livery, Biomaterials 33 (2012) 3013e3024.
[34] M. Ding, Z. Qian, J. Wang, J. Li, H. Tan, Q. Gu, et al., Effect of PEG content on the
properties of biodegradable amphiphilic multiblock poly(ε-caprolactone
urethane)s, Polym. Chem. 2 (2011) 885e891.
[35] I. Astafieva, X.F. Zhong, A. Eisenberg, Critical micellization phenomena in
block polyelectrolyte solutions, Macromolecules 26 (1993) 7339e7352.
[36] M. Ding, X. He, Z. Wang, J. Li, H. Tan, H. Deng, et al., Cellular uptake of poly-
urethane nanocarriers mediated by gemini quaternary ammonium, Bio-
materials 32 (2011) 9515e9524.
[37] H. Ai, C. Flask, B. Weinberg, X.T. Shuai, M.D. Pagel, D. Farrell, et al., Magnetite-
loaded polymeric micelles as ultrasensitive magnetic-resonance probes, Adv.
Mater. 17 (2005) 1949e1952.
[38] H. Zhu, J. Tao, W. Wang, Y. Zhou, P. Li, Z. Li, et al., Magnetic, fluorescent, and
thermo-responsive Fe3O4/rare earth incorporated poly(St-NIPAM) coreeshell
colloidal nanoparticles in multimodal optical/magnetic resonance imaging
probes, Biomaterials 34 (2013) 2296e2306.
[39] a) C. Allen, D. Maysinger, A. Eisenberg, Nano-engineering block copolymer
aggregates for drug delivery, Colloids Surfaces B 16 (1999) 3e27;
b) M. Ding, L. Zhou, X. Fu, H. Tan, J. Li, Q. Fu, Biodegradable gemini multiblock
poly(ε-caprolactone urethane)s toward controllable micellization, Soft Matter.
6 (2010) 2087e2092.
[40] J. Zhang, L. Wang, H. Wang, K. Tu, Micellization phenomena of amphiphilic
block copolymers based on methoxy poly(ethylene glycol) and either crys-
talline or amorphous poly(caprolactone-b-lactide), Biomacromolecules 7
(2006) 2492e2500.
[41] a) H. Sun, B. Guo, R. Cheng, F. Meng, H. Liu, Z. Zhong, Biodegradable micelles
with sheddable poly(ethylene glycol) shells for triggered intracellular release
of doxorubicin, Biomaterials 30 (2009) 6358e6366;
b) W. Lu, X. Wang, R. Cheng, C. Deng, F. Meng, Z. Zhong, Biocompatible and
bioreducible micelles fabricated from novel a-amino acid-based poly(disulfide
urethane)s: design, synthesis and triggered doxorubicin release, Polym. Chem.
6 (2015) 6001e6010.
[42] a) C. Wu, C. Belenda, J. Leroux, M.A. Gauthier, Interplay of chemical micro-
environment and redox environment on thioledisulfide exchange kinetics,
153
Chem-Eur J. 17 (2011) 10064e10070;
b) J. Kerr, J.L. Schlosser, D.R. Griffin, D.Y. Wong, A.M. Kasko, Steric effects in
peptide and protein exchange with activated disulfides, Biomacromolecules
14 (2013) 2822e2829.
[43] D.J. Phillips, M.I. Gibson, Redox-sensitive materials for drug delivery: targeting
the correct intracellular environment, tuning release rates, and appropriate
predictive systems, Antioxid. Redox Sign 21 (2013) 786e803.
[44] H.S. Yoo, T.G. Park, Folate receptor targeted biodegradable polymeric doxo-
rubicin micelles, J. Control. Release 96 (2004) 273e283.
[45] a) J. Lin, J. Chen, S. Huang, J. Ko, Y. Wang, T. Chen, et al., Folic acidePluronic
F127 magnetic nanoparticle clusters for combined targeting, diagnosis, and
therapy applications, Biomaterials 30 (2009) 5114e5124;
b) M. Guo, C. Que, C. Wang, X. Liu, H. Yan, K. Liu, Multifunctional super-
paramagnetic nanocarriers with folate-mediated and pH-responsive targeting
properties for anticancer drug delivery, Biomaterials 32 (2011) 185e194;
c) L. Ao, B. Wang, P. Liu, L. Huang, C. Yue, D. Gao, et al., A folate-integrated
magnetic polymer micelle for MRI and dual targeted drug delivery, Nanoscale
6 (2014) 10710e10716;
d) J. Eliezar, W. Scarano, N.R.B. Boase, K.J. Thurecht, M.H. Stenzel, In vivo
evaluation of folate decorated cross-linked micelles for the delivery of plat-
inum anticancer drugs, Biomacromolecules 16 (2015) 515e523.
[46] Y. Huang, Z. Tang, X. Zhang, H. Yu, H. Sun, X. Pang, et al., pH-triggered charge-
reversal polypeptide nanoparticles for cisplatin delivery: preparation and
in vitro evaluation, Biomacromolecules 14 (2013) 2023e2032.
[47] F. Meng, Z. Zhong, J. Feijen, Stimuli-responsive polymersomes for pro-
grammed drug delivery, Biomacromolecules 10 (2009) 197e209.
[48] H. Sun, B. Guo, X. Li, R. Cheng, F. Meng, H. Liu, et al., Shell-sheddable micelles
based on dextran-SS-poly(ε-caprolactone) diblock copolymer for efficient
intracellular release of doxorubicin, Biomacromolecules 11 (2010) 848e854.
[49] a) S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L. Vander Elst, et al., Magnetic
iron oxide nanoparticles: synthesis, stabilization, vectorization, physico-
chemical characterizations, and biological applications, Chem. Rev. 108 (2008)
2064e2110;
b) H.B. Na, I.C. Song, T. Hyeon, Inorganic nanoparticles for MRI contrast agents,
Adv. Mater. 21 (2009) 2133e2148.
[50] H.Y. Yoon, G. Saravanakumar, R. Heo, S.H. Choi, I.C. Song, M.H. Han, et al.,
Hydrotropic magnetic micelles for combined magnetic resonance imaging
and cancer therapy, J. Control. Release 160 (2012) 692e698.
[51] X. Yang, S. Pilla, J.J. Grailer, D.A. Steeber, S. Gong, Y. Chen, et al., Tumor-tar-
geting, superparamagnetic polymeric vesicles as highly efficient MRI contrast
probes, J. Mater. Chem. 19 (2009) 5812e5817.
[52] H. Su, Y. Liu, D. Wang, C. Wu, C. Xia, Q. Gong, et al., Amphiphilic starlike
dextran wrapped superparamagnetic iron oxide nanoparticle clsuters as
effective magnetic resonance imaging probes, Biomaterials 34 (2013)
1193e1203.
[53] C.E. Smith, D. Ernenwein, A. Shkumatov, N.E. Clay, J. Lee, M. Melhem, et al.,
Hydrophilic packaging of iron oxide nanoclusters for highly sensitive imaging,
Biomaterials 69 (2015) 184e190.
[54] a) J. Zhu, J. Wang, X. Wang, J. Zhu, Y. Yang, J. Tian, et al., Facile synthesis of
magnetic core-shell nanocomposites for MRI and CT bimodal imaging,
J. Mater. Chem. B 3 (2015) 6905e6910;
b) Y. Sarigiannis, A. Kolokithas-Ntoukas, N. Beziere, R. Zboril, E. Papadimitriou,
K. Avgoustakis, et al., Synthesis and evaluation of condensed magnetic
nanocrystal clusters with in vivo multispectral optoacoustic tomography for
tumour targeting, Biomaterials 91 (2016) 128e139.