Hybrid mesoporous silica nanoparticles with pH-operated and complementary H-bonding caps as an autonomous drug-delivery system

Article abstract of DOI:10.1002/chem.201402864

Mesoporous silica nanoparticles (MSNPs) are functionalized with molecular-recognition sites by anchoring a triazine or uracil fragment on the surface. After loading these MSNPs with dyes (propidium iodide or rhodamine B) or with a drug (camptothecin, CPT)

Full text of DOI:10.1002/chem.201402864

DOI: 10.1002/chem.201402864
Full Paper
&
Nanoparticles
Hybrid Mesoporous Silica Nanoparticles with pH-Operated and
Complementary H-Bonding Caps as an Autonomous Drug-
Delivery System
Christophe Thꢀron,[a, c] Audrey Gallud,[b] Carole Carcel,*[a] Magali Gary-Bobo,*[b]
Marie Maynadier,[b] Marcel Garcia,[b] Jie Lu,[d] Fuyuhiko Tamanoi,[d] Jeffrey I. Zink,*[c] and
Michel Wong Chi Man*[a]
Abstract: Mesoporous silica nanoparticles (MSNPs) are func-
tionalized with molecular-recognition sites by anchoring a tri-
azine or uracil fragment on the surface. After loading these
MSNPs with dyes (propidium iodide or rhodamine B) or with
a drug (camptothecin, CPT) they are capped by the comple-
mentary fragments, uracil and adenine, respectively, linked
to the bulky cyclodextrin ring. These MSNPs are pH-sensitive
and indeed, the dye release was observed at acidic pH by
continuously monitored fluorescence spectroscopy studies.
On the other hand, no dye leakage occurred at neutral pH,
hence meeting the non-premature requirement to minimize
side effects. In vitro studies were performed and confocal
microscopy images demonstrate the internalization of the
MSNPs and also dye release in the cells. To investigate the
drug-delivery performance, the cytotoxicity of CPT-loaded
nanoparticles was tested and cell death was observed. A re-
markably lower amount of loaded CPT in the MSNPs (more
than 40 times less) proved to be as efficient as free CPT.
These results not only demonstrate the drug release after
pore opening under lysosomal pH, but they also show the
potential use of these MSNPs to significantly decrease the
amount of the administered drug.
Introduction
ate material because they can be synthesized with adjustable
size and pore diameter. They are extremely effective solid sup-
ports and are suitable for drug storage and delivery due to
their stability and noncytotoxicity.[8] Additionally, they can be
modified by organic groups and have successfully been used
as efficient cell markers[9] and also for regulated drug-delivery
rates.[10] Interestingly, the modification by organic functions
can be achieved both onto the walls of the pores and on the
external surface[11] and in this context, mechanized hybrid
MSNPs have recently been devised with triggering functions
and used as nanocarriers for cargo release upon suitable stim-
ulus.[12] Hence, several thermal-,[13] redox-,[14] photo-,[15] bio-[16]
and pH-responsive[17] capping molecules and polymers have al-
ready been used to control the opening of the mesopores to
release the cargo molecules on demand,[12a] for example in
tumor and in plant cells.[18] In the case of tumor cells, a main
critical point is to prevent leakage and to avoid premature re-
lease of the drug before reaching the target cells. Therefore
a key challenge is to develop an autonomous MSNP with
a self-triggered gate-opening system liable to occur inside the
infected cells without external control. Because MSNP can
enter infected cells[19] due to the enhanced permeability and
retention (EPR) effect and through the endocytosis route to
the acidic lysosome compartment, a mild pH operating strat-
egy may afford such autonomy to MSNP. Capped pH-sensitive
nanocarriers thus appear as suitable systems to achieve these
goals. Several MSNP caps have already been explored, includ-
ing polyelectrolytes,[20] polymer coatings,[21] hydrazone,[22] cy-
In the early 1990s, porous and high-surface-area silicas became
easily accessible[1] and are since being actively developed for
catalysis,[2] sensors,[3] separation,[4] light-harvesting, and charge-
transport materials.[5] Recently, the nanofabrication of hybrid
silica-based materials turned out to be a main target that con-
tributes to advances in applications including energy, optics,
microelectronics,[6] and bio- and nanomedicine.[7] Mesoporous
silica nanoparticles (MSNPs) therefore appeared as an appropri-
[a] Dr. C. Thꢀron, Dr. C. Carcel, Dr. M. Wong Chi Man
Institut Charles Gerhardt Montpellier UMR-5253 CNRS-UM2-ENSCM-UM1
Ecole Nationale Supꢀrieure de Chimie de Montpellier
8 rue de l’ꢀcole normale, 34296 Montpellier Cedex 05 (France)
E-mail: carole.carcel@enscm.fr
michel.wong-chi-man@enscm.fr
[b] A. Gallud, Dr. M. Gary-Bobo, Dr. M. Maynadier, Dr. M. Garcia
Institut des Biomolꢀcules Max Mousseron UMR 5247 CNRS
UM 1; UM 2 - Facultꢀ de Pharmacie
15 Avenue Charles Flahault, 34093 Montpellier Cedex 05 (France)
E-mail: magali.gary-bobo@inserm.fr
[c] Dr. C. Thꢀron, Prof. J. I. Zink
Department of Chemistry and Biochemistry
University of California Los Angeles
405 Hilgard Avenue, Los Angeles, CA 90095 (USA)
E-mail: zink@chem.ucla.edu
[d] Dr. J. Lu, Prof. F. Tamanoi
Microbiology, Immunology and Molecular Genetics
University of California Los Angeles
405 Hilgard Avenue, Los Angeles, CA 90095 (USA)
Chem. Eur. J. 2014, 20, 9372 – 9380
9372
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
clodextrins,[23] base pairing and/or DNA[17e,24] systems. To date,
very few of these MSNPs have been tested in vitro and in
vivo[24c] and they usually do not comply with non-premature
drug release.
first loaded, with fluorescent dyes (PI or RhB) to first monitor
the self-opening of the pores owing to the weak acidic
medium and to assess the internalization of the MNSPs in the
cells. In the next study, the MNSPs were loaded with an anti-
cancer drug (CPT) to examine the cytotoxicity through in vitro
studies.
In this paper, we report two stable multi-H-bonded and
capped MSNPs based on base-pairing systems, which undergo
lysosomal release of the cargo molecules (propidium iodide
(PI), rhodamine B (RhB), and camptothecin (CPT)). Besides
meeting the non-premature release requirement, these MSNPs
can potentially be applied to significantly decrease the concen-
tration of the administered drug and hence minimizing the
side effects of extra drug delivered on premature release.
We previously reported the synthesis of a pH-sensitive
bridged silsesquioxane in which the silica-anchored melamine
fragments hold cyanuric acid molecules tightly through hydro-
gen bonds (molecular recognition) and we demonstrated that
these H-bonds can be cleaved under mild acidic conditions,
successfully and completely releasing the templating cyanuric
molecules.[25] Based on these results, we envisioned two
MSNPs endowed with molecular-recognition groups as auton-
omous pH-operating drug nanocarriers and delivery systems.
The strategy and pH-controlled release of cargo from MSNPs
are outlined in Figure 1. Pores should be closed at blood pH
(7.4) to avoid premature release and should be open at the
acidic pH of lysosomes (5.5–4.5) to deliver the drug.
Results and Discussion
Precursors and materials synthesis
The synthesis of the complementary cap based on b-CD cou-
pled with a nucleic base was carried out through a known
two-step protocol[26] with tosylation of native b-CD followed by
nucleic-base substitution as described in Figure 2. Cap 1 and
Cap 2 were obtained with an overall yield of 4.7 and 5%, re-
spectively.
To achieve such systems two separate syntheses were re-
quired: 1) the functionalization of MCM-41[1c] to provide graft-
ed stalks with selective H-bonding recognition sites (this was
done either with a triazine or with an uracil fragment) on the
external surface of the resulting nanocarriers, and 2) the syn-
thesis of b-cyclodextrin (b-CD) derivatives of the complementa-
ry bases (respectively, uracyl and adenine as Cap 1 and Cap 2).
By base-pairing with the corresponding stalks, the bulky cyclo-
dextrin would then cap the pores. To investigate their viability,
we studied these MSNPs towards human breast cancer cells
(MCF-7). In a first study, the functionalized nanoparticles were
Figure 2. Synthesis of Cap 1 and 2.
MCM-41 nanoparticles were functionalized with two organo-
silanes with recognition sites to provide the stalks to the
MSNPs. The syntheses of these organosilane precursors Stalk 1
and Stalk 2 are given in Figure 3. Precursor Stalk 1 was pre-
pared according to literature[25a] and Stalk 2 was synthesized
by reacting 5-aminouracil with 3-isocyanatopropyltriethoxysi-
lane with a yield of 78%.
These stalks were separately and covalently attached to the
MSNP surfaces by means of a post-synthesis protocol. In both
cases, the attachment was confirmed by 29Si-CPMAS NMR with
the appearance of T sites (Figure 4a,c) with a higher condensa-
tion rate for Stalk 1 (T3 site only and Q3 +Q4) than for Stalk 2
(T2 +T3 sites and only Q3). In addition, the functionalization
was also demonstrated by the IR spectra, both MSNPs exhibit-
ing strong and specific bands (mainly within the 1300–
Figure 1. Schematic representation of pH-sensitive MSNPs.
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9373
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. Synthesis of Stalk 1 and 2. DIPEA=N,N-diisopropylethylamine.
1800 cmÀ1 range) similar to those of the corresponding precur-
sors Stalks 1 and 2; these were not observed in the IR spec-
trum of the nonfunctionalized MCM-41 (Figure 4b,d).
Thermogravimetric analyses (TGA) were also performed on
the materials (Figure 5). All materials exhibit a slight weight
loss up to about 2008C. This is attributed to the adsorbed
water molecules as well as residual released hydroxyl and
ethoxyl groups upon the condensation process during heat
treatment. In the next segment (200–5008C), significant weight
losses were observed for the functionalized materials (about
36 and 25% for Stalks 1 and 2, respectively) due to the degra-
dation of the organics during pyrolysis contrarily to the initial
MCM-41 material with no organic stalk (less than 5% weight
loss). This means that the average grafting on the nanoparti-
cles is 1.5 times higher for Stalk 2 than for the Stalk 1. This is
expected because the former has two triethoxy anchoring
groups.
Figure 4. 29Si-CPMS NMR of a) MCM-41+Stalk 1 and c) MCM-41+Stalk 2
and FT-IR of b) MCM-41+Stalk 1 and d) MCM-41+Stalk 2.
The morphology of these materials was then investigated
through TEM (Figure 6) experiments, which confirmed the uni-
form particle size as well as the well-ordered mesostructure of
the pores of the functionalized MSNPs.
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9374
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. TGA analyses of MCM-41 (black), MCM-41+Stalk 1 (dotted) and
MCM-41+Stalk 2 (gray).
Figure 7. DLS of NP1PI and NP2PI.
periments, the NPs were left at neutral pH 7 for 1 h before ad-
justing the solution to the chosen acidic pH as represented by
the vertical line in Figure 8. No PI was detected in the NP1PI
and NP2PI solutions at pH 7, demonstrating that the pores re-
mained closed. Indeed, H bonds are stable at neutral pH and
thus, as desired, no premature release was observed with
NP1PI and NP2PI. Upon decreasing the pH to 6 (Figure 8), PI
was released, indicating the potential of these systems for au-
tonomous drug delivery in cells. For each NP type, the delivery
rate depends directly on the pH value with higher release rates
occurring at higher acidity. However, the release profiles of the
two systems are different (Figure 8). Considering that the maxi-
mum release is attained at pH 3, NP1PI can release between
90–100% of the dye against only 45–60% for NP2PI at endoso-
mal pH after 15 h (see the Supporting Information, Figure S2).
It appeared that the release is faster with NP1 with an instan-
taneous release within 30 min and reaching progressively
nearly 80% release in less than 3 h in all cases, whereas for
NP2, the fast step lasted 2 h before reaching a plateau at 40,
50, and 60% at pH 6, 5, and 4, respectively. This result is in per-
fect agreement with the fact that the unpairing of nucleic base
pairs (NP2) is known to be complete at lower pH values than
in the case of melamine complexes (NP1).
Figure 6. TEM of MCM-41+Stalk 1.
Release experiments
In order to demonstrate the efficiency of the system at the tar-
geted pH conditions in water, the functionalized MSNPs, were
first loaded with PI as fluorescent dye. The loading was per-
formed by soaking the MSNPs in an aqueous solution of PI
(5 mm) during 12 h before filtration. Then the pores of the
modified MSNPs were capped through hydrogen bonding in
a concentrated solution of Cap 1 or Cap 2 in DMSO for three
days. For accurate release experiments the nanoparticles were
carefully washed with DMSO several times to remove any
physisorbed PI; this afforded NP1PI and NP2PI. The sizes of
these loaded NPs are quite close to those of the initial MCM-
41 (around 130 nm) as observed by dynamic light scattering
(DLS) experiments (Figure 7).
The release of PI was tested in H2O at several pH values: 4
to 6 to mimic the pH of endosomal compartments, and pH 3
to evaluate 100% release of the dye, followed by continuously
monitored fluorescence spectroscopy (see the Supporting In-
formation, Figure S1).[17b] NP1PI or NP2PI were placed in the
corner of a cuvette and the surrounding liquid was stirred
gently. A probe beam was used to illuminate the liquid above
the nanoparticles and to follow the release of the PI molecules
in the supernatant as they escaped the nanoparticles. To
ensure that no PI release occurred at the beginning of the ex-
In vitro studies
Based on these successful operations of the pH-controlled cap
of both NP1 and NP2 in water, in vitro studies were carried
out by using confocal microscopy techniques to determine
whether these pH-responsive MSNPs can function similarly in
the cells. Through these studies we wanted to demonstrate
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9375
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 9. Cellular uptake of PI-loaded nanoparticles. Human breast cancer
cells (MCF-7) were incubated at 378C at different times with NP1PI or NP2PI
(40 mgmLÀ1). Fluorescence intensities were quantified on living cells by flow
cytometry. Values represent the meanÆSD calculated from three independ-
ent experiments.
RhB), were incubated with MCF-7 living cells, and analyzed by
confocal laser scanning microscopy (Figure 10a). Costaining
was performed with Hoechst, a blue fluorescent marker of the
nuclei. After 20 h incubation, the amount of both nanoparticles
in the living cells is quite similar. These data are in agreement
with the kinetic study. Because PI cannot cross the membranes
of living cells, it should remain confined in the cell compart-
ment as observed in Figure 10a. In contrast, RhB crosses cell
membranes and diffuses into the cytoplasm. Figure 10a shows
a full staining of the cytoplasm by RhB, demonstrating the re-
lease from both NP1RhB and NP2RhB. The efficiency of the two
constructions appears to be equivalent. These results support
the use of these pH-sensitive MSNPs to display an intracellular
delivery of a molecule of interest.
Figure 8. Dye release at acidic pH of NP1PI (top) and NP2PI (bottom).
two main points: first, that the MSNP were taken up by cells
and localized in the acidic lysosome and, second, that the
pores were opened and the dye released in the cytoplasm
(Figure 10b). Two fluorescent dyes were used: PI and RhB with
a similar loading procedure in both cases as described above.
Similarly to the PI-loaded nanoparticles (NP1PI and NP2PI),
The next step was to prove the application of these pH-sen-
sitive MSNPs in intracellular controlled delivery of drugs. In
order to investigate drug delivery, the nanoparticles were
loaded with a chemotherapeutic agent, CPT, and were named
NP1CPT and NP2CPT, containing, respectively, 1.2 and 1.1 mg of
CPT per mg of mesoporous silica as determined by UV/Vis. The
cytotoxicity of these nanoparticles was established on MCF-7
cell cultures treated by increasing doses of nanoparticles with
or without CPT (NP1CPT and NP1RhB, respectively), in order to
distinguish the part of the toxicity induced by the nanoparti-
cles from that induced by the drug. In parallel, the cells were
treated with increasing doses of free CPT. Strikingly, we ob-
served a better efficiency of both CPT-containing nanoparticles
(with a much lower CPT concentration) compared with free
CPT. With the latter, a concentration that was more 40 times
higher than that of loaded CPT in nanoparticles was needed to
reach similar toxicity (see the Supporting Information,
Table S1). Indeed the results revealed that nanoparticles
(NP1CPT) and CPT induced a cell death of 31 and 33%, respec-
tively, at the higher concentration (60 mgmLÀ1; Figure 11a). For
the empty nanoparticles, used as control (NP1RhB), we ob-
served a slight cell death of merely 8%. The second construc-
tion (NP2CPT) was evaluated under the same conditions and
the data demonstrated a decrease in cell viability of 27%
those loaded with RhB will be noted NP1RhB and NP2RhB
.
Hence nanoparticles were loaded with PI (NP1PI and NP2PI)
to assess whether they can achieve internalization in living
cells. PI and 4’,6-diamidino-2-phenylindole dihydrochloride
(DAPI) are fluorescent dyes that are currently used because
they cannot pass through intact plasma membranes of living
cells. Therefore, to exclude dead cells that could retain dyes,
a costaining with DAPI was realized. The aim of this experi-
ment is to evaluate by flow cytometry analysis, the uptake of
nanoparticles in living cells. This kinetic study was performed
on MCF-7 breast cancer cells incubated with nanoparticles
(40 mgmLÀ1). The data showed a rapid cellular entry (Figure 9)
during the first 7 h. The slight difference in time of internaliza-
tion observed could be attributed to the nature of the cap.
However, both constructions have an equivalent maximal
uptake with a plateau that is reached after 7 h of incubation.
Having demonstrated the internalization of nanoparticles,
we then investigated whether the caps are sensitive to pH
acidification, as in the lysosome compartment. In order to
assess the intracellular delivery nanoparticles, NP1PI, NP2PI,
NP1RhB, and NP2RhB loaded with red fluorescent probes (PI or
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9376
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 10. a) Confocal microscopy images of living cells. MCF-7 cells were incubated during 20 h with nanoparticles (40 mgmLÀ1) loaded with PI (NP1PI and
NP2PI) or RhB (NP1RhB and NP2RhB) and costained with a nuclear marker, Hoescht 33342. RhB and PI are visualized in red at 620 nm and Hoechst is visualized
in blue at 460 nm. Images are representative of at least three independent experiments. Bar: 5 mm. b) Illustration of cellular uptake of and drug release from
NPPI and NPRhB..Only released RhB from NPRhB (not PI) is able to pass through the pores of lysosome membrane and stain the cytoplasm .
whereas the corresponding empty nanoparticles (NP2) did not
induce any toxicity (Figure 11b).
unit to the bulky b-cyclodextrin. The pH sensitivity of these
new nanomachines was proven by release of the dye at acidic
pH, as demonstrated by continuously monitored fluorescence
spectroscopy studies. In addition, these nanomachines were
stable at neutral pH, because not the slightest dye leakage
was detected. These systems obviously comply with the re-
quired conditions for therapeutic applications, namely, a non-
premature release (the drug remains in the material at neutral
pH) and a controlled release triggered by an acid stimulus (pH
close to the weak acidic media of the cell compartments). The
acid-triggered self-release behavior of the two systems was
similar in cancer cells. Indeed, confocal microscopy studies
confirmed the internalization of the MSNPs and also dye re-
lease. Moreover, the cytotoxicity experiments on human breast
cancer cells by using MSNPs loaded with CPT showed a de-
crease in cell viability similar to that of the drug alone, confirm-
ing the release of the active compound after pore opening in
the acidic lysosomes. Interestingly, a much smaller amount of
the chemotherapeutic agent CPT (less than 40 times) loaded in
these MNPs, compared with free CPT, is needed to reach simi-
lar efficacy towards the MCF-7 cells; this approach is therefore
appropriate to decrease side-effects due to excessive drug ad-
ministration.
The cytostatic effect of CPT-loaded nanoparticles was also
studied by flow cytometry analysis (Figure 11c). After 24 h in-
cubation, the cells were fixed with ethanol and stained with PI
to assess cell-cycle distribution. NP1CPT and NP2CPT were able to
trigger, respectively, 67 and 71% cell-cycle arrest in the G0/G1.
Interestingly, CPT treatment also induces accumulation of cells
in G0/G1 in the same range (73%). These data are in agree-
ment with cell-viability tests and demonstrate that CPT-loaded
nanoparticles, which despite of being loaded with much less
CPT (<40 times), are as efficient as free CPT. It implies the re-
lease of CPT after pore opening at lysosomal pH.
Conclusion
In this work, a new type of nanomachine was prepared. For
this, silylated precursors with molecular recognition sites based
on triazine (Stalk 1) and uracil (Stalk 2) fragments were grafted
on MSNPs surfaces. After the loading with cargo molecules,
the nanoparticles were successfully capped through hydrogen
bonds with the complementary caps. These were prepared by
covalently coupling an uracil (Cap 1) or an adenine (Cap 2)
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9377
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Experimental Section
General methods
Reagents were used as received. Solvents were dried by using an
MB SPS-800 apparatus. FTIR data were obtained on a PerkinElmer
FT-IR system Spectrum BX spectrophotometer equipped with Glad-
iaATR. Liquid 1H and 13C NMR spectra were recorded on a Bruker
AC-400 spectrometer at room temperature with CDCl3 and DMSO
as solvents and tetramethylsilane (TMS) as an internal reference.
29Si solid-state NMR spectra were obtained on a Bruker DSX
300 MHz spectrometer by using cross-polarization and magic-angle
spinning techniques (CP-MAS). High-resolution mass spectra (Q-
TOF ES+) were measured on a JEOL MS-DX 300 mass spectrome-
ter. Discontinuous nitrogen sorption isotherms were measured at
77.15 K on a Micromeritics 2010 apparatus. Specific surface areas
were derived with the BET transform of the sorption isotherms,
taking 0.162 nm2 as the cross section area for nitrogen. No micro-
porosity could be detected with the t-plot method, taking Aero-
sil200 as reference.
Synthesis
Mono-6-deoxy-6-(p-tosylsulfonyl)-b-CD[26b] (1): b-Cyclodextrin
(20.0 g, 17.62 mmol) was suspended in water (170 mL), and NaOH
(2.19 g in 7 mL of water) was added dropwise over 6 min. The sus-
pension became homogeneous and slightly yellow before the ad-
dition was complete. p-Toluenesulfonyl chloride (3.36 g,
17.62 mmol) in acetonitrile (10 mL) was added dropwise over
8 min, causing immediate formation of a white precipitate. After
2 h of stirring at 238C the precipitate was removed by suction fil-
tration and the filtrate refrigerated during 72 h at 48C. The result-
ing white precipitate was recovered by suction filtration and after
drying for 12 h under reduced pressure to afforded 9% (2.12 g,
1.64 mmol) of a pure white solid. 1H NMR (DMSO, 400 MHz): d=
7.74 (d, 2H), 7.42 (d, 2H), 5.87–5.58 (m, 14H), 4.82 (brs, 4H), 4.76
(brs, 3H), 4.55–4.13 (m, 6H), 3.74–3.43 (m, 28H), 3.42–3.18 (m,
overlaps with HOD), 2.42 ppm (s, 3H); 13C NMR (DMSO, 400 MHz):
d=144.9 (s), 132.8 (s), 129.8 (d), 127.8 (d), 102.1 (m), 81.8 (d), 73.3–
71.4 (m), 70.0, 68.7, 59.5 (t), 21.1 ppm (4); HRMS (ESI+): m/z: calcd
for C49H77O37S (M+H): 1289.3786; found: 1289.3784.
Cap 1 (mono(6-ura-6-deoxy)-b-CD)[26a]: Anhydrous K2CO3 (0.204 g,
1.47 mmol) was added to a solution of uracil (0.165 g, 1.47 mmol)
and 1 (1.25 g, 0.97 mmol) in dry DMF (60 mL). The mixture was
heated at 808C for 24 h under nitrogen atmosphere. Then, the un-
reacted K2CO3 was removed by filtration, and the filtrate was
evaporated under a reduced pressure to dryness. The residue was
dissolved in water, and the aqueous solution was poured into vigo-
rously stirred acetone (300 mL) to give a slightly yellow precipitate.
This operation was repeated twice. The crude product obtained
was dried in vacuum and purified by column chromatography over
Sephadex G-25 with distilled deionized water as an eluent to give
Cap 1 in 52% yield. 1H NMR (D2O, 400 MHz): d=3.47–3.91 (m,
42H), 4.94–5.05 (m, 7H), 5.76 (d, 1H), 7.53 ppm (d, 1H); HRMS
(ESI+): m/z: calcd for C46H73N2O36 (M+H): 1228.3865; found:
1227.40.
Figure 11. a,b) Cell viability after a 72 h incubation with nanoparticles or
CPT. MCF-7 breast cancer cells were treated with increasing doses of nano-
particles (0–60 mgmLÀ1) or CPT (0–3 mgmLÀ1) and investigated with the MTT
assay as described in the Experimental Section. The percentage of living
cells treated with NP is compared to control cells (treated with vehicle
alone) which is considered to be 100% of living cells. NP1CPT and NP2CPT con-
tained 5 mg of CPT per mg of mesoporous silica. Values represent meanÆSD
of three experiments. c) Cell cycle analysis of MCF-7 treated with nanoparti-
cles or CPT. Cells were treated with nanoparticles (60 mgmLÀ1) or CPT (with
3 mgmLÀ1) for 24 h and then stained with PI. The DNA content was analyzed
by means of flow cytometry. G0/G1, G2/M, and S indicate cell-cycle phases.
Bar graphs are from three representative experiments (+SD).
Cap 2 (mono(6-ade-6-deoxy)-b-CD)[26a]: Anhydrous K2CO3 (0.139 g,
1.01 mmol) was added to
a solution of adenine (0.136 g,
1.01 mmol) and 1 (1.30 g, 1 mmol) in dry DMF (60 mL). The mixture
was heated at 808C for 24 h under nitrogen atmosphere. Then, the
unreacted K2CO3 was removed by filtration, and the filtrate was
evaporated under a reduced pressure to dryness. The residue was
dissolved in water, and the aqueous solution was poured into vigo-
rously stirred acetone (300 mL) to give a slightly yellow precipitate.
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9378
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
This operation was repeated twice. The crude product obtained
was dried in vacuum and purified by column chromatography over
Sephadex G-25 with distilled deionized water as an eluent to give
Cap 2 in 55% yield. 1H NMR (D2O, 400 MHz): d=3.00–3.80 (m,
42H), 4.86 (s, 7H), 7.92 (s, 1H), 7.95 ppm (s, 1H); HRMS (ESI+): m/
z: calcd for C47H74N5O34 (M+H): 1251.4137; found: 1250.48.
Determination of loaded CPT in NP1CPT and NP2CPT
To evaluate the quantity of CPT loaded in the nanoparticles, NP1
and NP2 (5 mgmLÀ1) were dissolved in an acidic solution of DMSO
(pH 2). After 15 min of sonication at room temperature, the nano-
particles were centrifuged (10 min, 10,000 rpm). Absorbances of
the supernatants were read at 340 nm in a microplate reader. The
titration gave 1.2 and 1.1 mgmgÀ1 of CPT loaded in NP1CPT and
NP2CPT, respectively (see Table S1 in the Supporting Information).
Stalk 1 and 2: Stalk 1 was prepared according to a reported pro-
cedure.[25a] In a Schlenk under inert atmosphere, 5-aminouracil
(0.5 g, 3.93 mmol) was dissolved in anhydrous DMSO (25 mL). The
solution was stirred at 608C until complete dissolution of 5-amino-
uracil. Then, 3-isocyanatopropyltriethoxysilane (1.0 mL, 4.04 mmol)
was added dropwise and the solution was stirred overnight under
nitrogen at 608C. After removal of DMSO under reduced pressure,
dry CH2Cl2 (20 mL) then dry pentane (20 mL) were added to obtain
a precipitate. The resulting yellow solid was washed with dry pen-
tane (three times, 50 mL). Stalk 2 was obtained in 78% yield
Cell culture conditions
Human breast cancer cells (MCF-7) were purchased from ATCC
(American Type Culture Collection, Manassas, VA). MCF-7 cells were
cultured in DMEM-F12 culture medium supplemented with fetal
bovine serum (10%) and gentamycin (50 mgmLÀ1). These cells
grew in a humidified atmosphere at 378C under CO2 (5%).
1
(1.15 g, 3.07 mmol) as a yellow powder. H NMR (DMSO, 400 MHz):
d=0.53 (t, 2H); 1.14 (t, 9H); 1.43 (m, 2H); 2.99 (q, 2H); 3.74 (q,
6H); 6.76 (s, 1H); 7.67 (s, 1H); 7.83 (s, 1H); 10.40 (s, 1H);
11.37 ppm (s, 1H); 13C NMR (DMSO, 400 MHz): d=7.31 (s); 18.33
(s); 23.3 (s); 41.76 (s); 57.82 (s); 115.13 (s); 123.83 (s); 149.41 (s);
155.27 (s); 160.98 ppm (s); HRMS (ESI+): m/z: calcd for
C14H27N4O6Si (M+H): 375.1622; found: 375.1682.
Cellular uptake kinetics of nanoparticles
For cellular uptake kinetics of the nanoparticles, MCF-7 cells were
seeded into a six-well plate and allowed to grow for 24 h. The PI-
loaded nanoparticles, NP1PI, NP2PI (40 mgmLÀ1), were added and
incubated for 4–24 h. The cells were then washed with phosphate-
buffered saline (PBS), trypsinized and suspended in Dulbecco’s
Modified Eagle Medium (DMEM)/F-12phenol redfree medium
(Gibco, Saint-Aubin, France). The costaining viability assay was per-
formed with DAPI (0.5 mgmLÀ1) and cells were subsequently intro-
duced into a FACSCanto II flow cytometer (Becton Dickinson,
France). The data were analyzed with Win MDI software v2.8.
Preparation of porous MCM-41 silica particles: Cetyltrimethylam-
monium bromide (CTAB; 0.315 g) was added to deionized H2O
(150 mL). NaOH (2m, 1.1 mL) was added to the solution causing
the pH to increase to above 12 and inducing the complete dissolu-
tion of CTAB. The solution was heated to 808C while stirring to
create
a homogeneous solution. Tetraethylorthosilicate (TEOS;
1.4 mL) was added to the solution dropwise. The reaction was
completed after 2 h. Then the suspension was centrifuged and the
nanoparticles were washed with ethanol. The CTAB surfactant was
removed by a solution of HCl (9 mL,12m) in ethanol (160 mL) for
6 h at 608C.
Visualization of fluorescent nanoparticles in living MCF-7 cells
MCF-7 cells were seeded onto glass dishes (Word Precision Instru-
ment, Stevenage, UK) at a density of 106 cellscmÀ2. One day after
seeding, the cells were incubated for 20 h with fluorescent-loaded
nanoparticles NP1PI, NP1RhB, NP2PI, and NP2RhB (40 mgmLÀ1). Then,
the cells were counterstained with Hoechst 33342 (5 mgmLÀ1, Invi-
trogen, Cergy-Pontoise, France). Before visualization, the cells were
washed gently with phenol redfree DMEM/F-12 to remove free
nanoparticles. Representative images were obtained under a Zeiss
Axio Observer LSM780 confocal microscope.
NP1 and NP2: Step 1—grafting silylated precursor on MCM-41:
Stalk 1 or 2 was added to a suspension of MCM-41 in dry toluene.
The solution was heated at 808C during one day. Then the nano-
particles were collected by centrifugation and washed twice with
ethanol and five times with water. For NP1: Stalk 1 (80 mg,
0.1 mmol), MCM-41 (80 mg), and dry toluene (10 mL). 29Si CP MAS
solid-state NMR: d=À57.0, À102.4, À108.9 ppm. For NP2: Stalk 2
(150 mg, 0.4 mmol), MCM-41 (300 mg), and dry toluene (30 mL).
29Si CP MAS solid-state NMR: d=À57.1, À102.2 ppm. Step 2—load-
ing: Nanoparticles were added to a solution of cargo molecules
(5 mm) in water. To maximize dispersion, the suspension was soni-
cated during 20 min and then stirred overnight to allow the cargo
to enter inside the pores. After centrifugation, the particles were
washed three times with water and then twice with DMSO.
In vitro cytotoxicity experiments
MCF-7 cells were seeded into 96-well plates at 2ꢂ104 cells per well
in culture medium (200 mL). The nanoparticles were freshly dis-
solved in ethanol (5 mgmLÀ1) and put in an ultrasonic bath until
complete dissolution. One day after seeding, the cells were incu-
bated 72 h with different concentrations of nanoparticles (0, 10,
30, and 60 mgmLÀ1). CPT was freshly dissolved in ethanol
(0.5 mgmLÀ1) and put in an ultrasonic bath until complete dissolu-
tion and the cells were treated with increasing concentrations of
CPT (0.5–3 mgmLÀ1). Then, a MTT assay was performed to evaluate
the toxicity.[27] Briefly, cells were incubated for 4 h with MTT
NP1: no loading; NP1PI: PI (45 mg) in water (13.5 mL), and nano-
particles (90 mg); NP1RhB: RhB (32 mg) in water (13.5 mL), and
nanoparticles (90 mg); NP1CPT: CPT (15 mg) in DMSO (15 mL), and
nanoparticles (90 mg).
NP2: no loading; NP2PI: PI (45 mg) in water (13.5 mL), and nano-
particles (90 mg); NP2RhB: RhB (32 mg) in water (13.5 mL), and
nanoparticles (90 mg); NP2CPT: CPT (15 mg) in DMSO (15 mL), and
nanoparticles (90 mg).
(0.5 mgmLÀ1
,
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide; Promega) in media. The MTT/media solution was then re-
moved and the precipitated crystals were dissolved in ethanol/
DMSO (1:1). The solution absorbance was read at 540 nm.
Step 3—capping: Nanoparticles were dispersed in a DMSO solu-
tions of Cap 1 or 2 (245 mg in 12 mL of DMSO) and stirred for
three days at room temperature. After centrifugation, the nanopar-
ticles were washed with DMSO until the washing solution was col-
orless. The nanoparticles were finally dried overnight under
vacuum at 808C.
Cell-cycle analysis
Flow cytometric analysis was performed on 350000 cells seeded in
culture dishes (60 mm diameter) and allowed to grow for 24 h. The
cells were then treated with nanoparticles (60 mgmLÀ1) or CPT
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9379
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(3 mgmLÀ1). The cells were harvested 24 h after treatment and
fixed with 70% ethanol overnight. The fixed cells were then incu-
bated with RNase A (1 mgmLÀ1) and PI (40 mgmLÀ1) for 48 h, in the
dark at 48C. Finally, the DNA content of the cells was analyzed
with a FACS Calibur flow cytometer with a cell-quest software
(Becton Dickinson, France).
[11] a) A. Stein, B. J. Melde, R. C. Schroden, Advd. Mater. 2000, 12, 1403–
1419; b) D. S. Shephard, W. Zhou, T. Maschmeyer, J. M. Matters, C. L.
Roper, S. Parsons, B. F. G. Johnson, M. J. Duer, Angew. Chem. 1998, 110,
2847–2851; Angew. Chem. Int. Ed. 1998, 37, 2719–2723.
[12] a) K. K. Cotꢄ, M. E. Belowich, M. Liong, M. W. Ambrogio, Y. A. Lau, H. A.
Khatib, J. I. Zink, N. M. Khashab, J. F. Stoddart, Nanoscale 2009, 1, 16–
39; b) M. W. Ambrogio, C. R. Thomas, Y.-L. Zhao, J. I. Zink, J. F. Stoddart,
Acc. Chem. Res. 2011, 44, 903–913; c) S. Angelos, M. Liong, E. Choi, J. I.
Zink, Chem. Eng. J. 2008, 137, 4–13; d) Z. X. Li, J. L. Nyalosaso, A. A.
Hwang, D. P. Ferris, S. Yang, G. Derrien, C. Charnay, J. O. Durand, J. I.
Zink, J. Phys. Chem. C 2011, 115, 19496–19506; e) C. Wang, Z. Li, D. Cao,
Y.-L. Zhao, J. W. Gaines, O. A. Bozdemir, M. W. Ambrogio, M. Frasconi,
Y. Y. Botros, J. I. Zink, J. F. Stoddart, Angew. Chem. 2012, 124, 5556–
5561; Angew. Chem. Int. Ed. 2012, 51, 5460–5465; f) Z. X. Li, J. C. Barnes,
A. Bosoy, J. F. Stoddart, J. I. Zink, Chem. Soc. Rev. 2012, 41, 2590–2605.
[13] J. Croissant, J. I. Zink, J. Am. Chem. Soc. 2012, 134, 7628–7631.
[14] T. D. Nguyen, Y. Liu, S. Saha, K. C. F. Leung, J. F. Stoddart, J. I. Zink, J. Am.
Chem. Soc. 2007, 129, 626–634.
Acknowledgements
We are grateful to the CNRS and the French Higher Education
and Research Ministry. This work was funded with a young re-
searcher grant from the French National Research Agency
JCJC-(ANR-2010–1006–01), from the French American Cultural
Exchange Partner University Fund (Grant FACE-PUF 20091853)
and from a NIH grant (NIH grant CA133697). Montpellier Rio
Imaging platform was used for FACS analyses.
[15] Y. A. Lau, B. L. Henderson, J. Lu, D. P. Ferris, F. Tamanoi, J. I. Zink, Nano-
scale 2012, 4, 3482–3489.
[16] L. Chen, Y. Wen, B. Su, J. Di, Y. Song, L. Jiang, J. Mater. Chem. 2011, 21,
13811–13816.
Keywords: drug delivery
sensitivity · mesoporous hybrid silica nanoparticles · sol–gel
· molecular recognition · pH-
[17] a) S. Angelos, N. M. Khashab, Y. W. Yang, A. Trabolsi, H. A. Khatib, J. F.
Stoddart, J. I. Zink, J. Am. Chem. Soc. 2009, 131, 12912; b) D. Tarn, M.
Xue, J. I. Zink, Inorg. Chem. 2013, 52, 2044–2049; c) M. Xue, D. Cao, J. F.
Stoddart, J. I. Zink, Nanoscale 2012, 4, 7569–7574; d) Y. L. Zhao, Z. X. Li,
S. Kabehie, Y. Y. Botros, J. F. Stoddart, J. I. Zink, J. Am. Chem. Soc. 2010,
132, 13016–13025; e) Y. L. Choi, J. H. Lee, J. Jaworski, J. H. Jung, J.
Mater. Chem. 2012, 22, 9455–9457.
[1] a) T. Yanagisawa, T. Shimizu, K. Kuroda, C. Kato, Bull. Chem. Soc. Jpn
1990, 63, 988–992; b) J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz,
C. T. Kresge, K. D. Schmitt, C. T. W. Chu, D. H. Olson, E. W. Sheppard, S. B.
McCullen, J. B. Higgins, J. L. Schlenker, J. Am. Chem. Soc. 1992, 114,
10834–10843; c) C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli,
J. S. Beck, Nature 1992, 359, 710–712.
[18] F. Torney, B. G. Trewyn, V. S. Y. Lin, K. Wang, Nat. Nanotechnol. 2007, 2,
295–300.
[19] a) I. Slowing, B. G. Trewyn, V. S. Y. Lin, J. Am. Chem. Soc. 2006, 128,
14792–14793; b) J. M. Rosenholm, A. Meinander, E. Peuhu, R. Niemi,
J. E. Eriksson, C. Sahlgren, M. Linden, Acs Nano 2009, 3, 197–206.
[20] Q. Yang, S. Wang, P. Fan, L. Wang, Y. Di, K. Lin, F.-S. Xiao, Chem. Mater.
2005, 17, 5999–6003.
[21] J.-T. Sun, C.-Y. Hong, C.-Y. Pan, J. Phys. Chem. C 2010, 114, 12481–12486.
[22] C.-H. Lee, S.-H. Cheng, I. P. Huang, J. S. Souris, C.-S. Yang, C.-Y. Mou, L.-
W. Lo, Angew. Chem. 2010, 122, 8390–8395; Angew. Chem. Int. Ed.
2010, 49, 8214–8219.
[23] a) C. Park, K. Oh, S. C. Lee, C. Kim, Angew. Chem. 2007, 119, 1477–1479;
Angew. Chem. Int. Ed. 2007, 46, 1455–1457; b) H. A. Meng, M. Xue, T. A.
Xia, Y. L. Zhao, F. Tamanoi, J. F. Stoddart, J. I. Zink, A. E. Nel, J. Am. Chem.
Soc. 2010, 132, 12690–12697.
[24] a) C. Chen, F. Pu, Z. Huang, Z. Liu, J. Ren, X. Qu, Nucleic Acids Res. 2011,
39, 1638–1644; b) L. Chen, J. Di, C. Cao, Y. Zhao, Y. Ma, J. Luo, Y. Wen,
W. Song, Y. Song, L. Jiang, Chem. Commun. 2011, 47, 2850–2852; c) D.
He, X. He, K. Wang, M. Chen, Y. Zhao, Z. Zou, J. Mater. Chem. B 2013, 1,
1552–1560.
[2] a) A. Monge-Marcet, R. Pleixats, X. Cattoen, M. W. C. Man, Catal. Sci.
Technol. 2011, 1, 1544–1563; b) A. Zamboulis, N. Moitra, J. J. E. Moreau,
X. Cattoꢃn, M. Wong Chi Man, J. Mater. Chem. 2010, 20, 9322–9338;
c) U. Dꢄaz, D. Brunel, A. Corma, Chem. Soc. Rev. 2013, 42, 4083–4097.
[3] a) J. Lei, L. Wang, J. Zhang, Chem. Commun. 2010, 46, 8445–8447;
b) W. S. Han, H. Y. Lee, S. H. Jung, S. J. Lee, J. H. Jung, Chem. Soc. Rev.
2009, 38, 1904–1915; c) T.-H. Tran-Thi, R. Dagnelie, S. Crunaire, L.
Nicole, Chem. Soc. Rev. 2011, 40, 621–639.
[4] F. Hoffmann, M. Frçba, Chem. Soc. Rev. 2011, 40, 608–620.
[5] a) N. Mizoshita, T. Tani, S. Inagaki, Chem. Soc. Rev. 2011, 40, 789–800;
b) M. ꢅlvaro, B. Ferrer, V. Fornes, H. Garcia, Chem. Commun. 2001,
2546–2547.
[6] a) L.-C. Hu, K. J. Shea, Chem. Soc. Rev. 2011, 40, 688–695; b) C. Sanchez,
P. Belleville, M. Popall, L. Nicole, Chem. Soc. Rev. 2011, 40, 696–753.
[7] a) A. Baeza, E. Guisasola, E. Ruiz-Hernandez, M. Vallet-Regi, Chem. Mater.
2012, 24, 517–524; b) S. Angelos, E. Johansson, J. F. Stoddart, J. I. Zink,
Advd. Funct. Mater. 2007, 17, 2261–2271; c) Y. Chen, H. Chen, J. Shi,
Advd. Mater. 2013, 25, 3144–3176.
[25] a) L. Fertier, C. Thꢀron, C. Carcel, P. Trens, M. Wong Chi Man, Chem.
Mater. 2011, 23, 2100–2106; b) S. Giret, C. Thꢀron, A. Gallud, M. May-
nadier, M. Gary-Bobo, M. Garcia, M. Wong Chi Man, C. Carcel, Chem. Eur.
J. 2013, 19, 12806–12814.
[26] a) Y. Liu, Q. Zhang, Y. Chen, J. Phys. Chem. B 2007, 111, 12211–12218;
b) R. C. Petter, J. S. Salek, C. T. Sikorski, G. Kumaravel, F. T. Lin, J. Am.
Chem. Soc. 1990, 112, 3860–3868.
[8] a) Y. S. Lin, S. H. Wu, Y. Hung, Y. H. Chou, C. Chang, M. L. Lin, C. P. Tsai,
C. Y. Mou, Chem. Mater. 2006, 18, 5170–5172; b) M. Liong, J. Lu, M. Ko-
vochich, T. Xia, S. G. Ruehm, A. E. Nel, F. Tamanoi, J. I. Zink, Acs Nano
2008, 2, 889–896; c) H. Zhang, D. R. Dunphy, X. Jiang, H. Meng, B. Sun,
D. Tarn, M. Xue, X. Wang, S. Lin, Z. Ji, R. Li, F. L. Garcia, J. Yang, M. L. Kirk,
T. Xia, J. I. Zink, A. Nel, C. J. Brinker, J. Am. Chem. Soc. 2012, 134, 15790–
15804.
[27] E. Chelebaeva, J. Larionova, Y. Guari, R. A. S. Ferreira, L. D. Carlos, A. A.
Trifonov, T. Kalaivani, A. Lascialfari, C. Guerin, K. Molvinger, L. Datas, M.
Maynadier, M. Gary-Bobo, M. Garcia, Nanoscale 2011, 3, 1200–1210.
[9] Y. S. Lin, C. P. Tsai, H. Y. Huang, C. T. Kuo, Y. Hung, D. M. Huang, Y. C.
Chen, C. Y. Mou, Chem. Mater. 2005, 17, 4570–4573.
[10] a) B. Munoz, A. Ramila, J. Perez-Pariente, I. Diaz, M. Vallet-Regi, Chem.
Mater. 2003, 15, 500–503; b) M. A. Vallet-Regꢄ, L. Ruiz-Gonzalez, I. Iz-
quierdo-Barba, J. M. Gonzalez-Calbet, J. Mater. Chem. 2006, 16, 26–31;
c) M. Vallet-Regꢄ, M. Colilla, B. Gonzalez, Chem. Soc. Rev. 2011, 40, 596–
607.
Received: March 31, 2014
Published online on July 1, 2014
Chem. Eur. J. 2014, 20, 9372 – 9380
www.chemeurj.org
9380
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim