Endosomal pH-activatable magnetic nanoparticle-capped mesoporous silica for intracellular controlled release

Article abstract of DOI:10.1039/c2jm32020g

Endosomal pH-driven linkage-disintegration is a promising strategy to achieve intracellular delivery and controlled drug release. In this paper, a rapid endosomal pH-sensitive MSNs ensemble (i.e., MCM-TAA-Fe₃O ₄) with magnetic nanopa

Full text of DOI:10.1039/c2jm32020g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 15960
www.rsc.org/materials
PAPER
Endosomal pH-activatable magnetic nanoparticle-capped mesoporous silica
for intracellular controlled release†
b
b
b
Qi Gan,ab Xunyu Lu, Wenjie Dong, Yuan Yuan, Jiangchao Qian,a Yongsheng Li,b Jianlin Shib
and Changsheng Liu
*
ab
*
Received 31st March 2012, Accepted 12th June 2012
DOI: 10.1039/c2jm32020g
Endosomal pH-driven linkage-disintegration is a promising strategy to achieve intracellular delivery and
controlled drug release. In this paper, a rapid endosomal pH-sensitive MSNs ensemble (i.e., MCM-TAA-
Fe3O4) with magnetic nanoparticle caps was developed by anchoring superparamagnetic Fe3O4
nanoparticles on the pore openings with an acid-labile substituted 1,3,5-triazaadamantane (TAA) group.
The functionalized Fe3O4 nanoparticles served as a nanogate to regulate the release pattern and/or dosage
of payload. The in vitro release experiment with model dexamethasone showed that the MCM-TAA-
Fe3O4 ensembles exhibited quick release at pH 5.0–6.0 and zero release in physiological environment (pH
¼ 7.4). Demonstrated with a MC3T3-E1 model cell line, this hybrid nanomaterial could successfully be
endocytosed into cells and then release the encapsulated exogenous cargos into the cytosol. The new rapid
endosomal pH-sensitive Fe3O4-capped-MSNs could serve as efficient carriers for intracellular controlled
release of therapeutic agents in live cells, and may be potentially applied in clinical disease therapy,
especially therapeutics and the metabolic manipulation of cells.
non-functionalized, negatively charged MSNs can not only be
1. Introduction
efficiently internalized by cells through a nonspecific adsorptive
endocytosis, but are also able to escape from endosomal
entrapment via the so-called ‘‘Proton Sponge’’ effect within
several hours.13 These outstanding properties enable MSNs to be
an ideal intracellular delivery carrier with great advantages over
their counterparts. As a matter of fact, several kinds of ‘‘gate-
like’’ MSN-based intracellular delivery systems have been
designed and reported in the past decade.14–17 Particularly, from
the viewpoint of in vivo biological applications, autonomous
pH-responsive MSN intracellular delivery systems, which can
hold the entrapped cargos in the physiological pH value of body
fluid (or the extracellular environment) and release the drug in
the acidic intracellular early endosomes (pH around 6.0) and late
endosomes/lysosomes (pH 5.0), appears to be the most prom-
ising.18 Most available pH-responsive MSNs ensembles use
either pH-sensitive supramolecules or acid-labile linkers.19–21
Despite the efforts given in endosomal pH-responsive MSNs,
however, these gated molecules used show a lack of targeting.
Also, the endosomal pH-triggered reaction is often slow22,23 and
most of the entrapped drug cannot be released before the MSNs
escape from the endosomal counterpart (within 4–6 hours).2,13
Hence, there is great need to develop an alternative endosomal
pH-responsive MSN ensemble with better targeting efficiency
and rapid release behavior in mild acidic endosomes.
Due to the great potential therapeutic applications, intracellular
delivery of genes and membrane-impermeable drugs/proteins has
become a hot research topic in recent years.1–3 The key to success
in intracellular delivery is a carrier that can selectively and
efficiently deliver the genes or therapeutic agents to the target
cells with ‘‘zero release’’ and minimal toxicity, and then breach
the cell-membrane barriers and release the entrapped cargos into
the cytosol or nucleus.4–8 To achieve this, various cell-penetrating
molecular agents and materials, including peptides, polymers,
and nano-scale particles, have been developed as carriers with
different degrees of success.
Mesoporous silica nanoparticles (MSNs) with unique parallel
open pores through the entire ensembles and tunable caps on the
pore ends can easily offer ‘‘zero-release’’ (when the pore gates are
fully closed) and ‘‘controlled-release’’ (when the pore gates are
open) patterns simultaneously in a single system.9–12 Moreover,
aThe State Key Laboratory of Bioreactor Engineering, East China
University of Science and Technology, Shanghai 200237, PR China
bKey Laboratory for Ultrafine Materials of Ministry of Education,
Engineering Research Center for Biomedical Materials of Ministry of
Education, East China University of Science and Technology, Shanghai
200237, PR China. E-mail: yyuan@ecust.edu.cn; liucs@ecust.edu.cn;
Fax: +86 21-54283420; Tel: +86 21-64251358
† Electronic supplementary information (ESI) available: Schematic
N2
adsorption/desorption isotherms, EDS spectra and magnetic properties
test of our materials and the dexamethasone released profile from
MCM-TAA-Fe3O4 with two pH stimuli. See DOI: 10.1039/c2jm32020g
In this respect, we recently developed a rapid endosomal pH-
responsive ensemble with magnetic nanogates–Fe3O4 nano-
particles which were grafted on the pore openings of MSNs via
synthesis
outline
of
MCM-TAE
and
Fe3O4-CBA,
15960 | J. Mater. Chem., 2012, 22, 15960–15968
This journal is ª The Royal Society of Chemistry 2012

View Article Online
an acid-labile substituted 1,3,5-triazaadamantane (TAA)
group (MCM-TAA-Fe3O4). Such TAA linkers are stable in
neutral/basic conditions, while they can be quickly hydrolyzed
into tri(aminomethyl)ethane (TAE) when the pH is # 6.0.24,25
Therefore, it is our hypothesis that, as shown in Scheme 1, at a
neutral pH (7.4), the linkers between the MSNs and Fe3O4
remain intact and the pore entries on the MSNs are tightly
blocked by Fe3O4 nanoparticles. However, at pH # 6.0 (e.g., at
the acidifying disease sites or cellular compartments), the Fe3O4
caps are removed with the destruction of TAA and thus the
encapsulated molecules diffuse out freely. The Fe3O4 nano-
particles here were chosen because they can not only be used as a
nanogate, but also can be targeted to the disease site under the
external magnetic field. In vitro functional pH-driven controlled
release of these MCM-TAA-Fe3O4 nanoparticles was investi-
gated with hydrophobic dexamethasone as the model drug. With
MC3T3-E1 cells, immature osteogenic cells which displayed a
fibroblastic morphology in the active growing phase, the
biocompatibility and cellular uptake were studied. Also, the
autonomous activation intracellular release behavior of these
MCM-TAA-Fe3O4 ensembles without invasive control was
investigated with doxorubicin as a model drug.
Sigma-Aldrich (CA, America). Pentaerythritol tribromide was
purchased from TCI Co. Ltd. (Tokyo, Japan) and ammonium
hydroxide was from Sinopharm Chemical Reagent Co. Ltd.
(Shanghai, China). Tetraethylorthosilicate (TEOS) and cetyl-
trimethyl ammonium bromide (CTAB) were from Shanghai
Lingfeng Chemical Reagent Co. Ltd. (Shanghai, China). Solvents
like dimethyl formamide (DMF), toluene, and methanol were from
Shanghai Chemical Reagent Co. Ltd. (Shanghai, China). DAPI
and Lyso-Tracker Red were purchased from Beyotime Biotech Co.
Ltd. (Jiangsu, China). All cell culture related reagents were
purchased from Gibco (Grand Island, NY).
2.2. Synthesis of 2-hydroxy-1,1,1-tris(aminomethyl)ethane
trihydrochloride (TAE$3HCl)
Sodium azide (NaN3, 107.7 mmol) was added into a solution of
pentaerythritol tribromide (10.2 mmol) in DMF (60 mL) under the
protection of nitrogen and incubated at 363 K for 36 h.26 The
solution was then cooled and diluted with 500 mL water, followed
by aether extraction (45 mL ꢀ 4). The organic fractions were
combined and dried with MgSO4. 1,4-Dioxane (100 mL) was
added and the volume of solvent was reduced to 40 mL. Addi-
tional 1,4-dioxane (125 mL) was added under stirring conditions,
together with the addition of triphenyl phosphine (PPh3,
50.7 mmol) and NH3 (aq, 30%, 100 mL). After 24 h, the product
was filtered, re-suspended in CHCl3 (200 mL), and extracted with
HCl (aq, 2.5 M, 35 mL ꢀ 5). The aqueous fractions were
combined, washed with CHCl3 (10 mL ꢀ 4) and then concentrated
to 25 mL. After that, 5 mL of concentrated HCl (37.5%) was
added and the solution was cooled to 277 K. The white crystal
powders, 2-hydroxy-1,1,1-tris(aminomethyl)ethane trihydro-
chloride, obtained after filtration were washed with cold HCl
(37.5%, 2 mL), EtOH (2 mL), and Et2O (10 mL ꢀ 5) in turn. The
product was dried under vacuum. The feature peaks of 2-hydroxy-
2. Experimental section
2.1. Materials
Sodium azide, 1,2-hexadecanediol, Fe(acac)3, oleylamine, oleic
acid, 1,4-dioxane, triethylamine, 4-carboxybenzaldehyde, 3-iso-
cyanatopropyltriethoxysilane, 1,8-diazabicyclo[5.4.0]undec-7-ene,
triphenyl phosphine, dexamethasone, doxorubicin, 3-(4,5-dime-
thylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and
fluorescein isothiocyanate isomer I (FITC) were all purchased from
1
1,1,1-tris(aminomethyl)ethane trihydrochloride in H NMR (200
MHz, D2O) were d 3.85 (s, 2H, –CH2O–), 3.29 (s, 6H, –CH2N–).
2.3. Synthesis of 7-(hydroxymethyl)-2,4,6-triphenyl-1,3,5-
triazaadamantane (TAA)
Triethylamine (1.65 mL, 11.8 mmol) was added to a suspension
of TAE$3HCl (0.8 g, 3.3 mmol) in MeOH (10 mL) under the
protection of nitrogen. Benzaldehyde (1.05 mL, 10.4 mmol) was
then added to the solution and refluxed for 30 min. After that, the
solution was cooled and the solvent was removed. The residue
1
was slurried with water (10 mL) and then filtered. H NMR
(400 MHz, CDCl3): d 7.21–7.89 (m, 15H, Ar), 5.62 (s, 1H,
PhCHeq), 5.42 (s, 2H, PhCHax), 3.53 (d, J ¼ 13.2 Hz, 2H,
–CH2N–), 3.22 (d, J ¼ 12.9 Hz, 2H, –CH2N–), 3.07 (s, 1H, OH),
2.93 (s, 4H, –CH2N–, –CH2O–).
2.4. Synthesis of 4-carboxybenzaldehyde dimethoxyacetal
(CBA-DMA)
A suspension of 4-carboxybenzaldehyde (4 g, 2.66 mmol) in dry
MeOH (80 mL) was added into ammonium chloride (8 g,
150 mmol) and the reaction was heated under reflux for 20 h.
After the solvent was removed under reduced pressure, a white
solid was obtained by recrystallizing from boiling hexane.
Scheme 1 Schematic of the reversible pH-responsive Fe3O4-capped
mesoporous silica ensembles with an acid-labile substituted tri-
azaadamantane linker.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 15960–15968 | 15961

View Article Online
2.5. Synthesis of tri(aminomethyl)ethane functionalized MCM
(MCM-TAE)
150 mg). The mixture was mixed at room temperature for 10 h.
The produced nanoparticles (Fe3O4-CBA-DMA) were then
centrifuged, washed with DCM, water and ethanol and dried
under vacuum. Fe3O4-CBA-DMA nanoparticles (20 mg) were
dispersed in 30 mL DCM and trifluoroacetic acid was then added
(1.2 mL) and the mixture was allowed to react for another 2 h,
The Fe3O4-CBA nanoparticles were obtained by magnetic
decantation, DCM, water and ethanol washing, and drying
under vacuum.
The MCM-TAE was prepared according to Scheme S1a† with
the following three major steps.
Synthesis of MCM-41: MCM-41 was prepared by a typical
CTAB-templated, base-catalyzed sol–gel method.27 The pH
value of deionized water (500 mL) was adjusted to approximately
11 with 26.4 mL ammonium hydroxide (29 wt% NH3 in water).
The liquid was heated to 323 K and then 0.56 g CTAB and
2.9 mL TEOS were added sequentially under strong agitation.
After 2 h, the sample was centrifuged and washed thoroughly
with DI water and ethanol. The surfactant templates were then
extracted by acidic methanol (9 mL of HCl/400 mL of methanol,
36 h) at 343 K, then further centrifuged and washed several times
with ethanol, and dried under vacuum for 20 h.
2.7. Drug loading and Fe3O4-CBA nanoparticles capped
MCM-TAE nanoparticles (MCM-TAA-Fe3O4)
200 mg MCM-TAE was dispersed in 10 mL anhydrous ethanol,
followed by the addition of 100 mg dexamethasone. The mixture
was allowed to react at room temperature for 24 h. After that,
50 mg Fe3O4-CBA nanoparticles and 25 mL of NaOH (1 M) were
added and the reaction continued at room temperature for
another 6 h. Dexamethasone-entrapped MCM-TAA-Fe3O4
nanoparticles were separated via magnetic decantation, and
washed thoroughly with ethanol, then dried under vacuum.
Doxorubicin-entrapped MCM-TAA-Fe3O4 nanoparticles were
prepared by the same method, except dispersed in DI water
instead. After the loading process, the amount of dexametha-
sone/doxorubicin loading was measured.
Synthesis of 3-isocyanatopropyltriethoxysilane functionalized
mesoporous silica nanoparticles (MCM-NCO): Chemical
modification of the silica surfaces was carried out by reacting
500 mg of the synthesized MCMs with 0.2 mL of 3-iso-
cyanatopropyltriethoxysilane (IPTS) in 90 mL toluene. The
reactions were performed under room temperature for 16 h and
the products were centrifuged and washed with toluene. The
IPTS-functionalized mesoporous silica nanoparticles (MCM-
NCO) were dried at 343 K for 12 h under vacuum.
Synthesis of MCM-TAE: 200 mg of IPTS-functionalized
MCMs were mixed with 150 mg of 7-(hydroxymethyl)-2,4,6-tri-
phenyl-1,3,5-triazaadamantane (TAA) in 100 mL toluene. Then,
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU, 0.05 mL) was added
to the solution and the reaction was carried out at 333 K for 12 h
with agitation. The products were then centrifuged and washed
with toluene. The TAA-functionalized mesoporous materials
(MCM-TAA) were dried at 343 K for 12 h under vacuum.
MCM-TAA (200 mg) nanoparticles were dispersed in 15 mL
H2O and 50 mL HCl (1 M) was then added to the solution. After
90 min, the benzaldehyde-removed materials (MCM-TAE) were
centrifuged, washed with water and dried at 298 K under
vacuum.
2.8. Synthesis of fluorescein isothiocyanate-labeled Fe3O4-
nanoparticle-capped MSNs (FITC–MCM-TAA-Fe3O4)
FITC–MCM was prepared by reacting 40 mg FITC with 0.2 mL
(3-aminopropyl)trimethoxysilane (APTMS) in 10 mL DMF for
3 h. The reaction product was then co-condensed with 5.8 mL
tetraethylorthosilicate (TEOS), 1.12 g CTAB, 1000 mL water,
and 52.8 mL ammonium hydroxide (29 wt% NH3 in water). The
reaction mixture was heated to 323 K under vigorous stirring
agitation for 2 h, and held for another 2 h. The surfactant
templates were then removed by extracting with acidic methanol.
Then FITC–MCM-TAE and FITC–MCM-TAA-Fe3O4 were
synthesised according to the same methods as for preparation
of MCM-TAE and MCM-TAA-Fe3O4 but not involving
dexamethasone.
2.6. Synthesis of carboxybenzaldehyde functionalized Fe3O4
nanoparticles (Fe3O4-CBA)
2.9. Characterization
The Fe3O4-CBA was prepared according to Scheme S1b†. The
process included the synthesis of Fe3O4-NH2 nanoparticles and
Fe3O4-CBA.
1H NMR spectra was recorded on Varian Gemini 200 or Unity
400 NMR spectrometers at the indicated frequencies. The FTIR
spectrum was measured by ATR on a Nicolet 380 (Thermo,
USA). X-Ray diffraction patterns were obtained in a RINT2000
vertical goniometer (Rigaku, Japan) using Cu Ka irradiation.
Surface analysis of the nanoparticles was performed by nitrogen
isothermal adsorption in a Micromeritics ASAP2010 sorp-
tometer (Micromeritics, USA). Their surface area was calculated
by the Brunauer–Emmett–Teller (BET) approach and the pore
size distributions were obtained by the Barrett–Joyner–Halenda
method. TEM and high-resolution TEM (HRTEM) were per-
formed on a transmission electron microscope (JEM-2100,
Japan) with a field emission gun operating at 200 kV. EDS was
performed on an X-ray energy dispersive analysis system
(EDAX, USA) installed on a transmission electron microscope.
Magnetic properties of the MCM-TAA-Fe3O4 nanoparticles
Synthesis of Fe3O4-NH2 nanoparticles: 200 mg of synthesized
5 nm Fe3O4 nanoparticles was prepared by an approach
described elsewhere.15,28 They were then dispersed in 400 mL
ethanol and 3-aminopropyltriethoxysilane (AMPES) was added
into the resulting suspension and mixed at room temperature
overnight under the protection of nitrogen. The obtained nano-
particles (Fe3O4-NH2) were separated, washed with water and
ethanol, and dried under vacuum.
Synthesis of carboxybenzaldehyde functionalized Fe3O4 nano-
particles: 100 mg Fe3O4-NH2 was dispersed in 150 mL anhydrous
dichloromethane (DCM) with agitation. Then, 4-carboxy-
benzaldehyde dimethoxyacetal (70 mg) was added into the
suspension, together with N-hydroxysuccinimide (NHS, 200 mg)
and 1-ethyl-3-(3-dimethylamino-propyl)-carbodiimide (EDC,
15962 | J. Mater. Chem., 2012, 22, 15960–15968
This journal is ª The Royal Society of Chemistry 2012

View Article Online
were measured under an external magnetic field and MPMS
(SQUID) VSM (Quantum Design, USA). The UV-visible spec-
troscopy was carried out with a UV/Vis spectrometer (Sunny-
vale, CA).
2.14. Flow cytometry analysis
MC3T3-E1 cells were plated at a density of 1.5 ꢀ 105 cells per
well in 35 mm Corningꢀ dishes and incubated for 24 h. After
attachment, the cells were cultured with a medium containing
50 mg mLꢂ1 of FITC–MCM-TAA-Fe3O4 for 1, 2, 4 and 14 h.
After that, these cells were washed with PBS (2ꢀ) and then
harvested by trypsinization. The cells were centrifuged, rinsed
with PBS and resuspended in PBS at a concentration of 1 ꢀ 106
cell mLꢂ1. Then the cells were immediately analyzed with a flow
cytometer at an emission wavelength of 530 nm for FITC.
2.10. In vitro drug release
50 mL PBS with different pH values was used to measure the
dexamethasone release profile. A bag filter loaded with 200 mg
dexamethasone-encapsulated MCM-TAA-Fe3O4 nanoparticles
was placed into a ZTY Auto Temperature sink (Jiangsu, China)
filled with PBS solutions at a certain pH. The system was then
maintained in circulation and at 310 K for 240 min. The release
of dexamethasone from the pore voids to the PBS was monitored
via UV-Vis spectrum measurement with the band of the drug
centered at 242 nm.
2.15. Drug release measurement from Fe3O4-capped MSNs in
cells
MC3T3-E1 cells were plated at a density of 5.0 ꢀ 104 cells per
well and incubated for 24 h. In order to track drug release from
the FITC-labeled hybrid materials, a red dye, doxorubicin, was
selected as a model drug. The loading efficiency of doxorubicin
was nearly 18% and the concentration of the doxorubicin-
entrapped hybrid materials added to the cells was 50 mg mLꢂ1
.
2.11. Cell culture
After each data collecting time, the cells were fixed and stained.
Confocal images were captured with a Leica TCS SP5 confocal
fluorescence microscope system.
MC3T3-E1 osteoblastic cells, derived from spontaneously
immortalized calvarial cells and representing immature osteogenic
cells, were purchased from the American Type Culture Collection
(ATCC). MC3T3-E1 cells were cultured in 37.5 cm2 flasks at 37 ꢁC
in a humidified atmosphere of 5% CO2/95% air with a MEM
medium containing 10% fetal bovine serum, 100 U mLꢂ1
penicillin-G, and 100 mg mLꢂ1 streptomycin.
2.16. Statistical analysis
The results were expressed as mean ꢃ SD. All data were gener-
ated in three or four independent experiments.
3. Results
2.12. Cytotoxicity assay
3.1. Characterization of the MCM-TAA-Fe3O4 and its
precursors
The cytotoxicity was evaluated with MTT (3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide) assay. MC3T3-E1 cells
were plated into 24-well plates at a density of 1.0 ꢀ 104 cells per
well in 1.0 mL culture medium and incubated for 24 h. Afterwards,
the growth media was removed and the cells were washed with
PBS (2ꢀ). To each well was then added 1 mL MEM medium
containing different materials (Fe3O4-CBA, MCM-TAE, MCM-
TAA-Fe3O4) at the same concentration of 100 mg mLꢂ1. The MTT
assay was then conducted following the standard protocol after
incubating cells for 24 h, 48 h, and 72 h.
Fig. 1a shows the TEM images of the Fe3O4 nanoparticles used
as nanogates for the MSN ensembles. The particle size is ꢄ5 nm.
As shown in Fig. 1b, the synthesized MCM-41 nanoparticles
exhibited a highly ordered hexagonal array and streaky struc-
tural features. Such typical mesoporous channels (black and
white stripes) could also be found in the TEM image of the final
MCM-TAA-Fe3O4 nanoparticles (see Fig. 1c). The Fe3O4-CBA
2.13. Cellular uptake of FITC–MCM-TAA-Fe3O4
To investigate the cellular uptake efficiency, the MC3T3-E1 cells
were plated at a density of 5.0 ꢀ 104 cells per well in 35 mm
Corningꢀ dishes with coverslips at the bottom of the dishes and
incubated for 24 h. Then, FITC–MCM-TAA-Fe3O4 suspensions
(50 mg mLꢂ1) were immediately added into the wells and left for
1 h or 13 h incubation. The growth media was then replaced with
2 mL of fresh medium including Lyso-Tracker Red (75 nM,
Beyotime). After 1 h incubation, the cells were washed with PBS
(pH ¼ 7.4) and re-incubated for 30 min in PBS solution con-
taining 3.7% formaldehyde and 57.0 mM DAPI. After incuba-
tion, the PBS solution was removed and the coverslips were
covered with 50% glycerol in water. The DAPI-stained coverslips
were placed on microscope slides and examined under a Leica
TCS SP5 confocal fluorescence microscope system.
Fig. 1 TEM images of synthesized Fe3O4 (a), MCM-41 (b), and MCM-
TAA-Fe3O4 (c). The black arrows in micrograph (c) point to Fe3O4
nanoparticles. The scale bars are 50 nm.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 15960–15968 | 15963

View Article Online
NPs were clearly seen in the MCM-TAA-Fe3O4 (the dark zones
on the outside of the mesopores indicated by arrows in the
figure), but not in the corresponding solid MCM-TAE. This
suggests that the drug loading and the later functionalization
processes did not cause obvious damage to the mesoporous
structure of MCM-41.
The small-angle X-ray patterns of MCM-41, MCM-TAE and
MCM-TAA-Fe3O4 are shown in Fig. 2a. It can be found that the
MSN nanoparticles exhibited typical MCM-41-type hexagonal
ordered frameworks with the characteristic (100) peak at 2.50ꢁ
(2q) and a d100 spacing of 3.9 nm. After functionalization by
tri(aminomethyl)ethane (MCM-TAE), a relatively weak peak at
2.74 degrees could be found and the d100 spacing was also
decreased to 3.2 nm, accompanied by the concomitant loss of the
(110) and (200) reflections. In the case of MCM-TAA-Fe3O4, all
peaks nearly disappeared. Fig. 2b shows the high-angle powder
XRD patterns of the Fe3O4 nanoparticles, Fe3O4-CBA nano-
particles and the MCM-TAA-Fe3O4. It can be seen that the
synthesized Fe3O4 and Fe3O4-CBA all had typical reflection
patterns of magnetite,28 which can be indexed as (220), (311),
(400), (422), (511) and (440), respectively. This confirmed that
Fe3O4 nanoparticles were successfully synthesized and the car-
boxybenzaldehyde functionalization process did not have an
obvious effect on the phase composition of the Fe3O4 nano-
particles. For MCM-TAA-Fe3O4, the main diffraction peaks for
Fe3O4-CBA were still visible, suggesting the presence of Fe3O4-
CBA on the surface of the Fe3O4-capped-MSNs.
Fig. 3 The FTIR spectra of (a) MCM-41 and MCM-41 materials after
IPTS modification (as MCM-NCO), and further tri(aminomethyl)ethane
functionalization (as MCM-TAE), and (b) Fe3O4 nanoparticles after
carboxybenzaldehyde modification (as Fe3O4-CBA) and MCM-TAA-
Fe3O4.
the typical IV isotherm curve disappeared; meanwhile, the
surface area and pore volume measured with BET were also
greatly decreased (Table 1). The EDS pattern (Fig. S1b†) further
confirmed the presence of Fe3O4 on the surface of the silica
ensembles. Fig. S1c† shows the magnetic response of the MCM-
TAA-Fe3O4 under an external magnetic field. The hysteresis loop
(Fig. S1d†) of the prepared hybrid materials at room temperature
showed that there was no hysteresis. These results indicated that
the particles are superparamagnetic at room temperature,
meaning that the thermal energy can overcome the anisotropy
energy barrier of a single particle, and the net magnetization of
the particle assemblies in the absence of an external field is zero.
In the FTIR spectra of the MCM-TAA-Fe3O4 nanoparticles
and their precursors (MCM-41), a band at 2280 cmꢂ1, corre-
sponding to the group of NCO, was present in the MCM-NCO
spectrum (Fig. 3).29 This confirmed that IPTS had been grafted
on MCM-41. However, after TAE grafting, this –NCO peak
nearly disappeared in the FTIR spectra of Fe3O4-CBA, MCM-
TAE, and MCM-TAA-Fe3O4 nanoparticles (Fig. 3b). Instead,
the absorption band at ꢄ626 cmꢂ1, the featured stretching
vibration band of the Fe–O bond, emerged.30 This confirmed the
existence of Fe3O4 nanoparticles on the surface of the meso-
porous particles.
3.2. In vitro functional pH-driven controlled release
To investigate the pH-responsive regulation efficiency of this
hybrid MCM-TAA-Fe3O4, dexamethasone release was
measured in PBS solution at different pH values. The drug
loading capacity calculated by division of the final weight of
MCM-TAA-Fe3O4 and the entrapped drug by the weight of the
entrapped drug is about 10.2%. As shown in Fig. 4, almost no
dexamethasone release was observed for MCM-TAA-Fe3O4 at
pH 7.4. This also confirmed no leakage of the entrapped drug
and a good gate efficiency of the Fe3O4 nanoparticles. However,
when the pH value was lowered to 6 and 5, a significant amount
of dexamethasone was released from the MCM-TAA-Fe3O4
ensembles. In particular at pH 5, nearly 80% drug was released
within 140 min. We also demonstrated that repeatable regulation
of the opening/closing of the nanogates by pH value stimuli is
feasible for MCM-TAA-Fe3O4 (Fig. S2†). The TEM image
(Fig. S3†) of the MCM-TAA-Fe3O4 nanoparticles at pH 5.0
further confirmed the separation of the Fe3O4 nanoparticles from
the MSNs with the destruction of the TAA at acidic pH. This
may provide a convenient means for pulsatile delivery of exog-
enous cargos.
N2 adsorption–desorption isotherms of MCM-41, MCM-
TAE and MCM-TAA-Fe3O4 nanoparticles are shown in
Fig. S1a†. As shown, the MCM-41 and MCM-TAE samples
showed a classical IV curve with a single, strong, and sharp
adsorption step at intermediate relative partial pressure values
around 0.25–0.3. However, after anchoring Fe3O4 nanoparticles,
Table 1 BET specific surface values, pore volume and pore size calcu-
lated from the nitrogen adsorption/desorption isotherms for the materials
used in this research
BET Surface
area (m2 gꢂ1
Pore volume
(cm3 gꢂ1
)
Pore size
(nm)
)
Fig. 2 (a) The small-angle powder X-ray diffraction spectra of MCM-
41, tri(aminomethyl)ethane-functionalized MCM-41 (MCM-TAE) and
MCM-TAA-Fe3O4; (b) the powder X-ray diffraction of the Fe3O4
nanoparticles, the Fe3O4 nanoparticles after carboxybenzaldehyde
modification (Fe3O4-CBA), and MCM-TAA-Fe3O4.
MCM-41
MCM-TAE
MCM-TAA-Fe3O4
1360.0
900.6
78.4
0.82
0.59
0.22
3.4
2.6
—
15964 | J. Mater. Chem., 2012, 22, 15960–15968
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 6 Flow cytometry analyses of MC3T3-E1 cells incubated with
FITC–MCM-TAA-Fe3O4 of 50 mg mLꢂ1 for 1, 2, 4 and 14 h.
Fig. 4 In vitro pH-responsive release profiles of dexamethasone from
MCM-TAA-Fe3O4 nanoparticles in PBS at pH 7.4 and pH 5.0–6.0.
after 1 h incubation. These results demonstrated that cellular
uptake of MCM-TAA-Fe3O4 materials mainly occurred in the
initial 4 h.
3.3. Biocompatibility of the MCM-TAA-Fe3O4 nanoparticles
and their hydrolysis products
Next, the cellular internalization and subsequent localization
of the hybrid nanoparticles were further investigated by confocal
fluorescence microscopy. This objective was accomplished by co-
incubating the FITC-labeled MCM-TAA-Fe3O4 with MC3T3-
E1 cells for 2 and 14 h. At the end of the time, the endosomal and
lysosomal compartments were stained with a red fluorescent
Lyso-Tracker and the cell nuclei were stained with DAPI. As
shown in Fig. 7, after 2 h incubation, the FITC–MCM-TAA-
Fe3O4 particles (in green) were found in the endosomes (in red)
(in yellow color in the merged images). It indicates that the MSN
particles could be internalized by the cells and stayed mainly in
the endosomal compartments in the initial 2 h uptake. However,
after 14 h incubation, the majority of the FITC-labeled MCM-
TAA-Fe3O4 had left the endosomes and entered into the cytosol.
In the experiment, we found that the escape of the FITC-labeled
MCM-TAA-Fe3O4 mainly occurred in 4–6 h. This result is in
agreement with a previous report.13
The cytotoxicities of the MCM-TAA-Fe3O4 systems and their
hydrolysis products, Fe3O4-CBA and MCM-TAE, were
measured on MC3T3-E1 cells at a concentration of 100 mg mLꢂ1
by MTT assay.31 As shown in Fig. 5, the values of the cell
viabilities for TAE functionalized MSNs, Fe3O4-CBA, and
MCM-TAA-Fe3O4 were all greater than 90%, even after 3 days
incubation, meaning a good level of biocompatibility associated
with TAE functionalized MSNs, Fe3O4-CBA and MCM-TAA-
Fe3O4 under standard physiological conditions.
3.4. Cellular uptake and intracellular controlled release of
FITC–MCM-TAA-Fe3O4 nanoparticles
In vitro quantitative cellular uptake of these hybrid materials was
performed via flow cytometry analysis. As shown in Fig. 6, with
the prolongation of the incubation time, the fluorescent intensity
profiles were shifted in the right direction. Specifically, the extent
of cellular uptake of FITC–MCM-TAA-Fe3O4 for 2 h, 4 h and
14 h were about 1.8-fold, 2.9-fold and 3.1-fold greater than that
In order to further determine whether the intracellular endo-
somal pH can trigger the release of the entrapped drug in the
Fig. 7 The confocal microscopy images of the uptake of FITC–MCM-
TAA-Fe3O4 by MC3T3-E1 cells after incubation for 2 h and 14 h. The
images on the left column indicate the green FITC–MCM-TAA-Fe3O4
materials taken up by cells. The images in the middle column are the red
fluorescent Lyso-Tracker Red-labeled lysosomes. The images on the right
column show the merging of the green and red fluorescent and blue
DAPI-stained nuclei.
Fig. 5 The in vitro cytotoxicity of Fe3O4-CBA nanoparticles, MCM-TAE,
and MCM-TAA-Fe3O4 materials at a concentration of 100 mg mLꢂ1 in
MC3T3-E1 cells, determined by MTT assay.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 15960–15968 | 15965

View Article Online
MSNs, confocal microscopy was used to follow the kinetics of
doxorubicin release from the FITC-labeled MCM-TAA-Fe3O4
in MC3T3-E1 cells at different time points. The fluorescence of
doxorubicin was significantly quenched when loaded into the
MSNs, and the appearance of fluorescence signals inside the cells
indicated the actual release of doxorubicin from MCM-TAA-
Fe3O4. The results are shown in Fig. 8. At the beginning of
internalization (1 h), the doxorubicin was retained in MSN-based
particles (shown as fluorescent yellow dots) with no leakage of
the entrapped doxorubicin from the MSN delivery system,
indicating that the ‘‘gate-like’’ TAA can hold the encapsulated
drugs under extracellular conditions. Whereas, the TAA
dissolved within the intracellular endosomes to initiate drug
release from the FITC–MCM-TAA-Fe3O4 after 4 h and a
remarkable amount of doxorubicin was observed throughout the
entire cells after 14 h incubation (an increasing red fluorescence
signal).
Therefore, in this study, 100–200 nm MSNs were used to fabri-
cate this new intracellular delivery system. Superparamagnetic
iron oxide nanoparticles (slightly bigger than the pore size of the
MSNs) used here act not only as nanogates, but also as a
magnetic species. The monodisperse Fe3O4 nanoparticles with
diameters smaller than 20 nm proved to be a promising candidate
for highly sensitive magnetic nanodevices.28,33,34 Moreover, the
magnetic particles with diameters of less than 10 nm can be
rapidly removed by extravasations and renal clearance, and show
little toxicity.35 With these considerations, ꢄ5 nm Fe3O4 nano-
particles were used in this study. As expected, after functionali-
zation by CBA and grafting onto the MSNs, the Fe3O4
nanoparticles could effectively prevent the leakage of the
entrapped drug. In addition, the iron oxide nanoparticles
anchored onto the MSNs still retained better magnetic
properties.
In order to achieve successful anchorage of Fe3O4 onto the
pore outlets of MSNs, the amino group of the TAE molecule was
first protected by benzaldehyde to form TAA and then
grafted onto the external surface of the mesoporous materials.
Also, the Fe3O4 nanoparticles were functionalized by dime-
thoxyacetal-protecting benzaldehydes to produce Fe3O4-CBA-
DMA (Scheme S1†). Prior to the capping reaction, the amino-
and aldehyde-protecting agents were removed from the MCM-
TAA and Fe3O4-CBA-DMA to produce amine-bearing MCM-
TAE and aldehyde-bearing Fe3O4-CBA, which quickly reacted
to form the final MCM-TAA-Fe3O4. The results from TEM,
XRD and FTIR spectra all confirmed the successful immobili-
zation of the CBA-functionalized Fe3O4 nanoparticles onto the
MCM-TAE matrix. The strong value and intensity of the d100
peak in the XRD pattern and TEM observation of the final
MCM-TAA-Fe3O4 are evidence that even after the loading
process with the drug and the further functionalization with
Fe3O4 nanoparticles, the mesoporous 3D MCM-41 scaffolding
still existed. Additionally, it is worthwhile to note that after
Fe3O4 capping, an obvious decrease in the intensity of the small-
angle powder XRD peaks and the BET specific surface and pore
volume were observed. Such decreases, we believe, may be
related to the pore-filling effect induced by both the entrapped
drug and the Fe3O4 nanoparticles.
4. Discussion
Due to the great potential applications in clinical imaging and
therapy, the dual intracellular release with ‘‘zero release’’ before
reaching the desired delivery sites and an autonomously triggered
release inside target cells has attracted great attention nowadays.
As we all know, the external delivery vectors are often internal-
ized by cells via endocytic pathways and thus are entrapped in
VATPase-positive acidic endosomes and lysosomes.6 From this
viewpoint, endosomal pH-sensitive dual intracellular controlled
release is very easy to realize in vivo. In this study, we designed
novel endosomal pH-activatable magnetic nanoparticle-capped
MSNs for intracellular controlled release with better target effi-
cacy and rapid response to endosomal pH environment.
According to a previous study, 100–200 nm MSNs exhibited
the best cellular uptake efficiency and biocompatibility.32
In vitro release data obtained here showed that the synthesized
MCM-TAA-Fe3O4 worked efficiently to prevent drug molecules
from undesired leaching at pH 7.4 and to achieve a rapid release
at pH 5.0–6.0. Moreover, even after internalization into
the acidifying endosomal environment, the MCM-TAA-Fe3O4
ensembles are still pH-responsive and the red-dye doxorubicin,
entrapped in the MCM-TAA-Fe3O4 nanoparticles could release
and then diffuse into the cytosol. Hence, it can be deduced that
rapid disintegration of pore blockers results in the opening of the
pores by the weak acidity in the endosomal counterparts, thereby
triggering drug release. The potential is not only favorable for
minimizing drug loss during the delivery process but also for
selective accumulation in targeted tissue by the enhanced
permeation and retention (EPR) effect. It is worthwhile to note
that the reaction between Fe3O4-CBA and MCM-TAE is
reversible (Fig. S2†). Therefore, it can be inferred that this MSN-
based nanogated container is much in favor of sustained and
pulsatile drug release, which is greatly needed for the treatment
of chronic and resistant diseases.
Fig. 8 Doxorubicin release from the FITC-labeled MCM-TAA-Fe3O4
nanoparticles in MC3T3-E1 cells after incubation for 1 h, 4 h, and 14 h.
The doses of the doxorubicin-entrapped MCM-TAA-Fe3O4 materials
added to the cells were 50 mg mLꢂ1. The left panels show the green
fluorescent FITC–MCM-TAA-Fe3O4 materials taken up by the MC3T3-
E1 cells. The middle panels show the distribution of doxorubicin in cells.
The right panels show the merging of the green fluorescent materials and
red dye doxorubicin and blue DAPI-stained nuclei images.
15966 | J. Mater. Chem., 2012, 22, 15960–15968
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Combining these drug release results with the acid-labile
properties associated with TAA, it can be strongly deduced that
the formation of TAA at neutral pH and the rapid hydrolysis of
these TAA groups under acidic conditions plays an important
role in the opening/closing of the pores. In addition, the elec-
trostatic interactions and the hydrogen bonding may also have
effects on the opening/closing of the nano-gates. The zeta-
potentials of the Fe3O4-CBA and MCM-TAE nanoparticles are
all positive at pH 5.0, but are negative at pH 7.4 (Table S1†). It
can be inferred that the electrostatic interactions between Fe3O4-
CBA and MCM-TAE may be conducive to the opening of the
‘‘nano-gate’’ at pH 5.0, and preventing the closing of the ‘‘nano-
gate’’ at pH 7.4. This may be the reason why the drugs continue
to release for several minutes when the pH value transferred back
to 7.4 in Fig. S2†. However, on the contrary, the hydrogen bonds
between Fe3O4-CBA and MCM-TAE might facilitate the
formation of the ‘‘nano-gate’’ of TAA at pH 7.4, and inhibit the
dissociation of the ‘‘nano-gate’’ at pH 5.0.
MCM-TAA-Fe3O4 nanoparticles, further experiments are
underway in our lab and the results will be addressed in the near
future.
Because of their improved therapeutic efficiency and the easy
triggering features, endosomal pH-sensitive intracellular delivery
systems are gaining importance. The MCM-TAA-Fe3O4 nano-
gated ensemble developed here, with good magnetic properties,
biocompatibility, and a pH-stimulated gate guard, can effectively
inhibit drug release under some conditions (e.g., pH ¼ 7.0–8.0).
While in the acidic endolysosomal environment, the Fe3O4
nanogate can be rapidly opened and thus the entrapped drug is
released. With these results, we envisioned that the MCM-TAA-
Fe3O4 ensembles developed here could serve as efficient targeted,
intracellular delivery carriers to deliver drugs/proteins into the
cytosol, especially for tumor and inflammatory cells with lower
endosomal pH values. Also, due to the inherent reversible
properties of the reaction between TAE and CBA at neutral pH
nearby and in acidic conditions, this MSN-based nanogated
container is very favorable for sustained and pulsatile drug
release.
The results obtained here indicated that the MCM-TAA-
Fe3O4 developed here could be internalized by the MC3T3-E1
cells within 4 h. Also, based on the previous reports and our
experiments, the MSNs could escape from the endosomal
counterpart in about 6 h. That is to say, for the endosomal pH-
triggered MSN-based release, the entrapped drug may be
released in the initial approximately 6 h. Otherwise, once the
MSNs escape from the endosomes with lower pH, the gate
cannot be opened and the encapsulated cargos cannot be released
in the cytosol with normal pH. That is to say, the time for the
disintegration of the linkage between the MSNs and the gate
molecules under the endosomal pH environment should be
shorter than that of the escape of the MSNs. From Fig. 4, it can
be found that at pH 5.0, 90% entrapped drug was released in the
initial 3 h, and the release rate was relatively faster than that at
pH 6.0. Moreover, Fig. 8 confirmed a rapid gate-opening and
drug release with this new MCM-TAA-Fe3O4 system in normal
MC3T3-E1 osteogenic cells. As we know, many diseased cells,
such as tumor cells and inflammatory cells often exhibit a lower
pH environment than normal cells.36 So, it can be inferred that
more rapid gate-opening would occur in the diseased cells.
Biocompatibility is a major concern when external substances,
such as NPs, are introduced into cells and tissues. As a new pH-
sensitive delivery system with great potential applications, the
Fe3O4-CBA, MCM-TAE and MCM-TAA-Fe3O4 nanoparticles
at a concentration of 100 mg mLꢂ1 all had no significant cyto-
toxicity. Such results are consistent with other researchers’
findings, in which MSNs and surface aminopropyl functionalized
MSNs exhibited little toxicity to HeLa cells.13 As far as the
biocompatibility of Fe3O4 is concerned, controversial findings
have been obtained. Uncoated Fe3O4 has been reported to
exhibit significant toxic effects at 400 mM iron concentration
after 72 h incubation, but after coating with PVA, the cell
viability was greatly improved.34 In our previous study, after
coating with a saccharide derivative, the Fe3O4 nanoparticles
enhanced the viability of MC3T3-E1 cells.15 In this study, the
Fe3O4-CBA and MCM-TAA-Fe3O4 nanoparticles all showed
minimal toxicity to the MC3T3-E1 cells. We believe such a
desirable response in cell viability was probably as a result of the
biological activity of carboxybenzaldehyde. To have a better
understanding of the in vivo toxicity and the metabolism of the
5. Conclusions
In summary, we have successfully developed a magnetic, endo-
somal pH-sensitive Fe3O4 nanoparticle-capped MSNs ensemble
by using a reversible acid-labile TAA unit for intracellular
delivery. In this drug delivery system, Fe3O4 nanoparticles were
anchored onto the pore openings through a reversible reaction of
the tris(aminomethyl)ethane entities on MSNs and the carbox-
ybenzaldehyde grafted on Fe3O4. The results showed that the
encapsulated molecules were held in the hybrid ensembles in a
physiological environment (pH 7.4) while quickly released in an
acidic environment (pH 5.0–6.0). These hybrid MCM-TAA-
Fe3O4 nanoparticles not only retain their good magnetic prop-
erties and high cell biocompatibility, but also allow efficient
endocytosis and endosomal escape in the MC3T3-E1 cell line.
These results demonstrate the great potential for future in vivo
intracellular delivery and controlled-release applications.
Acknowledgements
We greatly thank the National Basic Research Program of China
(973 Program, no. 2012CB933600), the National Natural Science
Foundation of China (no. 31070850 and 31100679), the Shanghai
Science and Technology Commission (no. 10540709800) and the
Program for Changjiang Scholars and Innovative Research Team
in University (IRT0825) for financial support. We also thank Dr
Shengnian Wang of Louisiana Tech University for his good advice
and English modification.
Notes and references
1 C. J. Ke, T. Y. Su, H. L. Chen, H. L. Liu, W. L. Chiang, P. C. Chu,
Y. N. Xia and H. W. Sung, Angew. Chem., Int. Ed., 2011, 50, 8086–
8089.
2 I. I. Slowing, B. G. Trewyn and V. S. Y. Lin, J. Am. Chem. Soc., 2007,
129, 8845–8849.
3 F. Torney, B. G. Trewyn, V. S. Y. Lin and K. Wang, Nat.
Nanotechnol., 2007, 2, 295–300.
4 B. A. Chen, M. Liu, L. M. Zhang, J. Huang, J. L. Yao and
Z. J. Zhang, J. Mater. Chem., 2011, 21, 7736–7741.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 15960–15968 | 15967

View Article Online
5 R. Cheng, F. H. Meng, S. B. Ma, H. F. Xu, H. Y. Liu, X. B. Jing and
Z. Y. Zhong, J. Mater. Chem., 2011, 21, 19013–19020.
6 N. V. Cuong, Y. L. Li and M. F. Hsieh, J. Mater. Chem., 2012, 22,
1006–1020.
7 N. Murthy, Y. X. Thng, S. Schuck, M. C. Xu and J. M. J. Frechet,
J. Am. Chem. Soc., 2002, 124, 12398–12399.
21 M. Xue, X. Zhong, Z. Shaposhnik, Y. Q. Qu, F. Tamanoi,
X. F. Duan and J. I. Zink, J. Am. Chem. Soc., 2011, 133, 8798–
8801.
22 R. Casasus, E. Climent, M. D. Marcos, R. Martinez-Manez,
F. Sancenon, J. Soto, P. Amoros, J. Cano and E. Ruiz, J. Am.
Chem. Soc., 2008, 130, 1903–1917.
8 A. K. Varkouhi, M. Scholte, G. Storm and H. J. Haisma,
J. Controlled Release, 2011, 151, 220–228.
23 R. Liu, Y. Zhang, X. Zhao, A. Agarwal, L. J. Mueller and P. Y. Feng,
J. Am. Chem. Soc., 2010, 132, 1500–1501.
9 J. E. Lee, D. J. Lee, N. Lee, B. H. Kim, S. H. Choi and T. Hyeon,
J. Mater. Chem., 2011, 21, 16869–16872.
24 A. M. Balija, R. E. Kohman and S. C. Zimmerman, Angew. Chem.,
Int. Ed., 2008, 47, 8072–8074.
10 J. Rosenholm, C. Sahlgren and M. Linden, J. Mater. Chem., 2010, 20,
2707–2713.
25 R. E. Kohman and S. C. Zimmerman, Chem. Commun., 2009, 794–
796.
11 I. I. Slowing, J. L. Vivero-Escoto, B. G. Trewyn and V. S. Y. Lin,
J. Mater. Chem., 2010, 20, 7924–7937.
26 B. G. Davis, K. Khumtaveeporn, R. R. Bott and J. B. Jones, Bioorg.
Med. Chem., 1999, 7, 2303–2311.
12 T. W. Kim, I. I. Slowing, P. W. Chung and V. S. Y. Lin, ACS Nano,
2011, 5, 360–366.
13 I. I. Slowing, B. G. Trewyn and V. S. Y. Lin, J. Am. Chem. Soc., 2006,
128, 14792–14793.
14 V. Cauda, C. Argyo, A. Schlossbauer and T. Bein, J. Mater. Chem.,
2010, 20, 4305–4311.
27 R. I. Nooney, D. Thirunavukkarasu, Y. M. Chen, R. Josephs and
A. E. Ostafin, Chem. Mater., 2002, 14, 4721–4728.
28 S. H. Sun, H. Zeng, D. B. Robinson, S. Raoux, P. M. Rice,
S. X. Wang and G. X. Li, J. Am. Chem. Soc., 2004, 126, 273–279.
29 S. Stankovich, R. D. Piner, S. T. Nguyen and R. S. Ruoff, Carbon,
2006, 44, 3342–3347.
15 E. Aznar, M. D. Marcos, R. Martinez-Manez, F. Sancenon, J. Soto,
P. Amoros and C. Guillem, J. Am. Chem. Soc., 2009, 131, 6833–6843.
16 Q. Gan, X. Y. Lu, Y. Yuan, J. C. Qian, H. J. Zhou, X. Lu, J. L. Shi
and C. S. Liu, Biomaterials, 2011, 32, 1932–1942.
17 K. C. F. Leung, T. D. Nguyen, J. F. Stoddart and J. I. Zink, Chem.
Mater., 2006, 18, 5919–5928.
18 H. A. Meng, M. Xue, T. A. Xia, Y. L. Zhao, F. Tamanoi,
J. F. Stoddart, J. I. Zink and A. E. Nel, J. Am. Chem. Soc., 2010,
132, 12690–12697.
30 P. P. Yang, Z. W. Quan, Z. Y. Hou, C. X. Li, X. J. Kang, Z. Y. Cheng
and J. Lin, Biomaterials, 2009, 30, 4786–4795.
31 J. S. Chang, K. L. B. Chang, D. F. Hwang and Z. L. Kong, Environ.
Sci. Technol., 2007, 41, 2064–2068.
32 Q. Gan, D. W. Dai, Y. Yuan, J. C. Qian, S. Sha, J. L. Shi and
C. S. Liu, Biomed. Microdevices, 2011, 14, 259–270.
33 R. Hao, R. J. Xing, Z. C. Xu, Y. L. Hou, S. Gao and S. H. Sun, Adv.
Mater., 2010, 22, 2729–2742.
34 M. Mahmoudi, A. Simchi and M. Imani, J. Phys. Chem. C, 2009, 113,
9573–9580.
19 S. Angelos, N. M. Khashab, Y. W. Yang, A. Trabolsi, H. A. Khatib,
J. F. Stoddart and J. I. Zink, J. Am. Chem. Soc., 2009, 131, 12912–
12914.
35 A. K. Gupta and M. Gupta, Biomaterials, 2005, 26, 3995–
4021.
20 C. Park, K. Oh, S. C. Lee and C. Kim, Angew. Chem., Int. Ed., 2007,
46, 1455–1457.
36 S. H. Tang, X. Q. Huang, X. L. Chen and N. F. Zheng, Adv. Funct.
Mater., 2010, 20, 2442–2447.
15968 | J. Mater. Chem., 2012, 22, 15960–15968
This journal is ª The Royal Society of Chemistry 2012