Drug encapsulation and release by mesoporous silica nanoparticles: The effect of surface functional groups

Article abstract of DOI:10.1002/chem.201403551

Mesoporous silica nanoparticles (MSNPs) have been widely used as drug carriers for stimuli-responsive drug delivery. Herein, a catalysis screening technique was adopted for analyzing the effects of chain length, terminal group, and density of disulfide-appended functional ligands on the surface of MSNPs on drug-loading capacity and glutathione-triggered drug-release kinetics. The ligand with an intermediate length (5 carbon atoms) and a bulky terminal group (cyclohexyl) that complexes with theβ-cyclodextrin ring showed the highest drug loading capacity as well as good release kinetics. In addition, decreasing the surface coverage of the functional ligands led to an enhancement in drug release. In vitro drug-delivery experiments on a melanoma cell line (B16-F10) by using the functionalized MSNPs further supported the conclusion. The results obtained may serve as a general guide for developing more effective MSNP systems for drug delivery. Maximizing drug release from drug carriers is important to achieve better therapeutic efficiency. Using glutathione (GSH)-triggered drug-release systems as a case study, the effect of the nature of surface functional groups on mesoporous silica nanoparticles on drug-loading and drug-release capabilities was investigated.

Full text of DOI:10.1002/chem.201403551

DOI: 10.1002/chem.201403551
Communication
&
Drug Delivery
Drug Encapsulation and Release by Mesoporous Silica
Nanoparticles: The Effect of Surface Functional Groups
Si Yu Tan,[a] Chung Yen Ang,[a] Peizhou Li,[a] Qi Ming Yap,[a] and Yanli Zhao*[a, b]
Chem. Eur. J. 2014, 20, 1 – 8
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
ed. Amorꢁs and co-workers, on the other hand, investigated
the relationship between different alkyl chain lengths and drug
Abstract: Mesoporous silica nanoparticles (MSNPs) have
been widely used as drug carriers for stimuli-responsive
drug delivery. Herein, a catalysis screening technique was
adopted for analyzing the effects of chain length, terminal
group, and density of disulfide-appended functional li-
gands on the surface of MSNPs on drug-loading capacity
and glutathione-triggered drug-release kinetics. The
ligand with an intermediate length (5 carbon atoms) and
a bulky terminal group (cyclohexyl) that complexes with
theb-cyclodextrin ring showed the highest drug loading
capacity as well as good release kinetics. In addition, de-
creasing the surface coverage of the functional ligands led
to an enhancement in drug release. In vitro drug-delivery
experiments on a melanoma cell line (B16-F10) by using
the functionalized MSNPs further supported the conclu-
sion. The results obtained may serve as a general guide
for developing more effective MSNP systems for drug de-
livery.
release behavior.[9] In a recent work by Gaberscek and co-work-
ˇˇ
ers, redox-responsive systems with different level of hindrance
on the disulfide linkage were studied, showing that an increase
in hindrance results in slower release.[10] Although some studies
based on GSH-triggered drug release in vitro and in vivo have
been reported,[11] the effect of terminal group, chain length,
and amount of disulfide unit on the MSNP surface on drug
loading and release is still not well understood.
Herein, we report a catalysis screening method to study the
effect of chain length, terminal group, and disulfide amount
on the surface of MSNPs on drug-loading capacity and release
kinetics (Figure 1). Supramolecular complexes involving b-cy-
clodextrin (b-CD) and the terminal groups of different function-
al ligands were employed in this study owing to the ease of
functionalizing the surface of MSNPs. A systemic approach was
adopted, whereby ligands with different chain lengths were
first tested. Thereafter, the chain length that gave the highest
release capability was used for the following tests that in-
volved different types of terminal units. Then, the effect of
varying the amount of disulfide unit on the surface was investi-
gated by using the functional ligand that showed the highest
release ability. Finally, in vitro drug-delivery experiments were
conducted on a melanoma cell line (B16-F10) by using the
functionalized MSNPs screened. We here selected doxorubicin
(DOX) as the cargo within MSNPs, as DOX is a common anti-
cancer drug that is loaded into MSNPs for drug-delivery appli-
cations.
Mesoporous silica nanoparticles (MSNPs) have shown various
potential biomedical applications, particularly as nanocarriers
for drug delivery.[1] Their ease of synthesis and tunable proper-
ties make MSNPs highly customizable according to desired
purpose.[2] Furthermore, its high surface area allows the graft-
ing of a large number of functional groups on the surface so
as to impart various interesting properties to MSNPs.[3] One of
the properties that can be imparted onto MSNPs is stimulated
release of loaded cargos through the grafting of capping
groups. Stimulated cargo release has been widely utilized as
a key strategy for the specific release of drugs in a controlled
manner[4] in order to reduce side effects brought about by the
use of anticancer drugs for chemotherapy. One common stim-
ulation method is redox activation based on glutathione (GSH)
in cancer cells,[5] as it has been shown that some cancer cells
express a significant amount of intracellular GSH compared to
healthy cells.[6] Under this strategy, the surface of MSNPs are
bridged with capping agents through a disulfide linkage.[7] In
this way, intracellular GSH would induce reductive cleavage of
the disulfide bond, thereby removing the capping agents and
releasing the loaded drugs. In a study by Zink and co-work-
ers,[8] the effect of complexation of a-cyclodextrin (a-CD) with
aniline units on hollow MSNPs on drug release was investigat-
We begin the studies by first synthesizing MCM-41 MSNPs
by using a surfactant-directed technique with hexadecyltrime-
thylammonium bromide (CTAB) as the surfactant. The as-syn-
thesized MSNPs were characterized by transmission electron
microscopy (TEM), and the results showed that the obtained
MSNPs were spherical in shape with diameter of 70–80 nm.
The Brunauer–Emmett–Teller (BET) measurements showed
a type IV adsorption/desorption isotherm, which is a character-
istic feature of mesoporous materials. The BET measurements
also show that the MSNPs possessed a high BET surface area of
921 m2 gÀ1 with a Barrett–Joyner–Halenda (BJH) pore size of
3.03 nm, which is slightly larger than the pore size (3.56 nm)
measured using powder X-ray diffraction (XRD). These charac-
terizations confirmed the successful synthesis of the MSNPs.
After grafting the surface with a mercaptan group, X-ray pho-
toelectron spectroscopic (XPS) analysis of mercaptan-grafted
MSNPs, that is, MSNPs-3(1), shows the presence of an S 2p
peak at 163.6 eV, indicating successful conjugation of the mer-
captan group onto the surface. The FT-IR spectrum presents
clear alkyl CÀH peaks, corresponding to the propyl groups of
the grafted 3-mercaptopropyltrimethoxysilane.
[a] S. Y. Tan,+ C. Y. Ang,+ P. Li, Q. M. Yap, Prof. Dr. Y. Zhao
Division of Chemistry and Biological Chemistry
School of Physical and Mathematical Sciences
Nanyang Technological University
21 Nanyang Link, Singapore 637371 (Singapore)
E-mail: zhaoyanli@ntu.edu.sg
Homepage: http://www.ntu.edu.sg/home/zhaoyanli/
The next step was the synthesis of a series of thiol-activated
ligands (Figure 2) for conjugating with the mercaptan-grafted
MSNPs. Thus, the conjugation of the n-bromo-terminated car-
boxylic acids of different chain lengths with a variety of alkyl
(R) amines was carried out, followed by the conversion of the
bromide groups into the thiol groups. The obtained ligands
were then coupled onto the surface of MSNPs prior to DOX
[b] Prof. Dr. Y. Zhao
School of Materials Science and Engineering
Nanyang Technological University
Singapore 639798 (Singapore)
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403551.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
terminal group were used. After
conjugation of the respective li-
gands onto the surface of
MSNPs, XPS was adopted to
measure the amount of func-
tional ligands on the surface of
MSNPs, that is, MSNPs-4(1)-1D,
MSNPs-4(1)-1E, and MSNPs-
4(1)-1F. The results (Table 1)
show that the sulfur content of
MSNPs-3(1),
MSNPs-4(1)-1D,
MSNPs-4(1)-1E
ca. 12.6–14.9%.
are
similar,
However,
MSNPs-4(1)-1F showed a slightly
higher sulfur content (ca. 15.3%)
upon conjugation of ligand 1F
to the surface of MSNPs-3(1). As
ligand 1F is longer and thus
having
a lower steric effect,
more ligand would conjugate
onto the surface of MSNPs-3(1).
Thus, the MSNPs-4(1) series
was employed for evaluating
drug-loading capacity. Table 1
summarizes the DOX loading ca-
pacity of these functionalized
MSNPs. The order of DOX load-
ing capacity was MSNPs-4(1)-
1E>MSNPs-4(1)-1F>MSNPs-
4(1)-1D. This observation was
surprising as we initially expect-
ed that MSNPs-4(1)-1F would
have the highest drug-loading
capacity owing to it having the
largest number of ligands func-
tionalized on the surface of
MSNPs. The results, however,
showed that ligand 1E, having 5
carbon atoms, gave the highest
Figure 1. a) Synthesis of different types of functional MSNPs under different conditions. First set of conditions
screened involved the use of ligands with different chain lengths. The second set of conditions tested was the
effect of the terminal groups with different sizes. Lastly, different amounts of the same ligand were conjugated
onto the surface of MSNPs to investigate the effects towards the surface coverage, drug loading capacity and re-
lease kinetics. m indicates the volume (mL) of 3-mercaptopropyltrimethoxysilane (MPS) used per 1 g of MSNPs.
b) Synthetic representation for the release mechanism from b-CD-capped and DOX-loaded MSNPs triggered by
GSH.
DOX loading capacity, 13.9%. The reason for such a high load-
ing capacity may be that ligand 1E possesses the ideal length
for the drug to enter into the mesopores and the ability to
retain the drug within the mesopores after capping with b-CD.
MSNPs-4(1)-1F exhibited a loading capacity of 13.2%. After
capping ligand 1F with b-CD, b-CD is a bit far away from the
surface of MSNPs, thus not being able to efficiently block the
mesopores for drug storage. The slight leakage of the loaded
drug during the washing process may in turn explain the
lower drug-loading capability of MSNPs-4(1)-1F compared
with MSNPs-4(1)-1E. Although MSNPs-4(1)-1D has approxi-
mately the same amount of ligands on the surface as MSNPs-
4(1)-1E, the cyclohexyl group is much closer to the mesopores
of MSNPs. This makes the drug-loading process difficult and
thus results in a low drug-loading capacity compared with
MSNPs-4(1)-1E.
Figure 2. Ligands conjugated onto the surface of MSNPs for investigating
the effect of ligand length and bulkiness.
loading and complexed with b-CD. The complexation of b-CD
with these (R) ending groups could keep the loaded drug
within the mesopores without leakage. Thereafter, the DOX
loading capacity and release profiles of functionalized MSNPs
were measured accordingly.
The first set of conditions tested was the effect of ligand
length on the drug-capping and release efficiency. Ligands
with varying lengths (3, 5, and 7 carbon atoms) and cyclohexyl
Then, the release profiles were obtained to investigate the
effect of chain length. Figure 3 shows that MSNPs-4(1)-1E
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Table 1. XPS data indicating the sulfur content after the conjugation of
respective ligands with different chain lengths on the surface of MSNPs.
In addition, the DOX loading capacity with standard deviation was ob-
tained for MSNPs-4(1)-1D, MSNPs-4(1)-1E, and MSNPs-4(1)-1F.
Table 2. Binding constants for the different terminal groups with b-CD.
Terminal group Binding constant
calculated from
region 1
Binding constant
calculated from
region 2
(R2 value)
Average
MSNPs
% Si
% S[a]
Loading capacity
(R2 value)
MSNPs-3(1)
87.4
85.1
85.3
84.6
12.6
14.9
14.7
15.4
–
adamantamine 2044 (0.9725)
cyclohexylamine 265 (0.9973)
3072 (0.9981)
322 (0.9727)
1766 (0.7339)
2873
293
–
MSNPs-4(1)-1D
MSNPs-4(1)-1E
MSNPs-4(1)-1F
12.2Æ0.8
13.9Æ0.3
13.2Æ0.9
[a]
n-butylamine
1545 (0.7989)
[a] Binding constant of n-butylamine with b-CD was also determined.
Since the R2 value was very big, the binding constant was considered to
be very low.
[a] The sulfur content corresponds to the amount of ligands being conju-
gated onto the surface of MSNPs.
MSNPs was studied using XPS. The XPS data of functionalized
MSNPs show different sulfur contents, thus representing differ-
ent amounts of ligands on the surface of functionalized
MSNPs. MSNPs-4(1)-1K exhibits the highest sulfur content fol-
lowed by MSNPs-4(1)-1B and lastly MSNPs-4(1)-1E. This order
is due to the differences in the flexibility and size of the end
groups. The small and flexible n-butyl group facilitates the con-
jugation of ligand 1K to the MSNP surface, resulting in the
highest sulfur content. The adamantyl and cyclohexyl groups,
on the other hand, are much larger than the n-butyl group,
thus leading to a low density of functional groups on the sur-
face. The rigid adamantyl group may be able to arrange more
neatly on the surface of MSNPs as compared with the cyclo-
hexyl group, leading to conjugation of more ligand 1B on the
surface than ligand 1E. Thus, the structural nature of these li-
gands could be used to explain the results obtained in Table 3.
Figure 3. DOX release kinetics for MSNPs-4(1)-1D, MSNPs-4(1)-1E, and
MSNPs-4(1)-1F when GSH (10 mm) was added at the 10th minute.
gave the highest release followed by MSNPs-4-1F and lastly
MSNPs-4(1)-1D. These results are consistent with the DOX
loading capacity, indicating that the interaction of GSH with
the disulfide bond on the MSNP surface leads to the uncap-
ping of the b-CD complexes for drug release. The complexa-
tion of b-CD did not have an obvious influence on the cleav-
age of the disulfide bond. Because MSNPs-4(1)-1E showed the
highest level of release, this system was chosen for the next
set of experiments.
Table 3. XPS data indicating the sulfur content upon conjugation of dif-
ferent ligands on the surface of MSNPs. In addition, the DOX loading ca-
pacity was obtained for MSNPs-4(1)-1B, MSNPs-4(1)-1E, and MSNPs-
4(1)-1K.
MSNPs
% Si
% S[a]
Loading capacity
MSNPs-3(1)
87.4
85.4
85.3
83.1
12.6
15.6
14.7
16.9
–
MSNPs-4(1)-1B
MSNPs-4(1)-1E
MSNPs-4(1)-1K
11.7Æ0.2
13.9Æ0.3
13.1Æ0.1
After the effect of chain length was investigated, we varied
the terminal groups for complexation with b-CD. As the cyclo-
hexyl end group was used in ligand 1E, two other end groups,
adamantyl and n-butyl, were selected for comparison studies.
The adamantyl end group serves as a more bulky counterpart
of the cyclohexyl group, while the n-butyl group only consists
of 4 carbon atoms. To investigate the effect of these terminal
groups on the DOX loading capacities and release profiles,
their binding constants involving complexation with b-CD
were determined. Table 2 shows the average binding constants
[a] The sulfur content corresponds to the amount of ligands being conju-
gated onto the surface of MSNPs.
The order of loading capacity for this series of functionalized
MSNPs was MSNPs-4(1)-1E>MSNPs-4(1)-1K>MSNPs-4(1)-1B
(Table 3). The loading capacity was mainly dependent on two
factors: firstly, the amount of the ligand on the surface of
MSNPs in relation to access to the mesopores, and secondly,
the binding of the ligand with b-CD to prevent the loaded
drug from escaping. MSNPs-4(1)-1B gave the lowest loading
capacity, although XPS showed that there was a substantial
amount of the ligand on the surface of MSNPs. The low level
of loading capacity might be due to the steric effect of the
adamantyl group, which prevents the drug from entering the
mesopores of MSNPs during the loading process. Compared
with MSNPs-4(1)-1K with the n-butyl end group, MSNPs-4(1)-
1E showed higher loading capacity on account of the higher
1
obtained from the Hildebrand plot of H NMR chemical shifts
of the different terminal groups with b-CD. As expected, the
adamantyl group gave the highest binding constant with b-
CD, followed by the cyclohexyl group. Because the R2 value of
the binding constant between n-butylamine and b-CD was
very large, the binding constant was considered to be very
low.
After obtaining the different binding constants, the conjuga-
tion of the different ligands (1B, 1E, 1K) onto the surface of
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
binding constant of the cyclohexyl end group with b-CD to
better keep the loaded drug within the mesopores. The DOX
release behavior of these systems was then tested, indicating
that the trend of the release profiles matches the trend of the
DOX loading capacities (Figure 4). The observation further sup-
ports that the interaction of GSH with the disulfide bond was
consistent for all the functionalized MSNPs. By comparing the
release profiles for this set of conditions, MSNPs-4(1)-1E again
gave the highest level of release and was chosen for the last
set of experiments.
crease in sulfur content after conjugation was proportional to
the type of MSNPs, making the comparison of the drug-load-
ing capacity easier.
The loading capacity of various functionalized MSNPs is pre-
sented in Table 4, and the results reveal that the loading ca-
pacity of MSNPs-4(1)-1E is higher than that of MSNPs-4(2)-1E
and MSNPs-4(0.5)-1E. Although MSNPs-4(2)-1E showed the
highest amount of ligands on the surface of MSNPs, its loading
capacity was lower, as the excessive amount of cyclohexyl
groups on the surface could block the mesopores for drug
loading. MSNPs-4(0.5)-1E also gave a low loading capacity be-
cause having less ligands on the surface cannot well retain
a large amount of DOX within the mesopores. Thus, MSNPs-
4(1)-1E has a reasonable amount of ligands on the surface to
allow sufficient loading of DOX into the mesopores without
obvious leakage. The level of drug release of these different
samples gave the following trend: MSNPs-4(0.5)-1E>MSNPs-
4(1)-1E>MSNPs-4(2)-1E (Figure 5). This order was unexpected
Figure 4. Release kinetics for MSNPs-4(1)-1B, MSNPs-4(1)-1E, and MSNPs-
4(1)-1K when GSH (10 mm) was added at the 10th minute.
The last set of conditions to be explored would be the
effect of different amount of the ligands on the surface to-
wards the DOX loading capacity and release efficiency. Firstly,
XPS was again used to quantify the amount of thiol units as
well as the final ligand on the surface of MSNPs. From the XPS
data in Table 4, an increase in the sulfur content was observed
when more mercaptan groups were grafted on the surface,
giving the following order in terms of the sulfur content:
MSNPs-3(2)>MSNPs-3(1)>MSNPs-3(0.5). After conjugation
of the final ligands onto the surface through disulfide bond
formation, the same order in sulfur content was observed:
MSNPs-4(2)-1E>MSNPs-4(1)-1E>MSNPs-4(0.5)-1E. The in-
Figure 5. Release kinetics for MSNPs-4(0.5)-1E, MSNPs-4(1)-1E, and MSNPs-
4(2)-1E when GSH (10 mm) was added at the 10th minute.
because the loading capacity was proportional to the release
profile in above two sets of conditions tested. In this experi-
ment, although MSNPs-4(0.5)-1E did not give the highest
loading capacity, the amount of DOX released from MSNPs-
4(0.5)-1E was the most significant. This is probably because
the low amount of ligands on the surface enables GSH to
easily interact with the disulfide bond, thus removing the cap-
ping agents for drug release. This explanation could be sup-
ported by the release profile of MSNPs-4(2)-1E, where the
lowest amount of release was observed. The surface of
MSNPs-4(2)-1E is highly hindered owing to the presence of
a large amount of ligands capped with b-CD, thus preventing
GSH from reacting with the disulfide bond.
Table 4. XPS data indicating the sulfur content after grafting different
amount of mercaptan groups on the surface of MSNPs as well as after
further conjugation of different amount of ligand 1E on the surface. In
addition, the DOX loading capacity was obtained for MSNPs-4(0.5)-1E,
MSNPs-4(1)-1E, and MSNPs-4(2)-1E.
MSNPs
% Si
% S[a]
Loading capacity
To investigate the applicability of these functional MSNPs, in
vitro drug-delivery experiments were carried out on the B16-
F10 melanoma cell line. B16-F10 cells were incubated with
MSNPs-4(0.5)-1E, MSNPs-4(1)-1E, and MSNPs-4(2)-1E, respec-
tively, and the cell viability data as well as fluorescence micros-
copy images were then obtained. The cell viability (Figure 6a)
involving these functional MSNPs at concentrations of
41.7 mgmLÀ1 and 20.8 mgmLÀ1 exhibited a proportional rela-
MSNPs-3(0.5)
MSNPs-4(0.5)-1E
MSNPs-3(1)
MSNPs-4(1)-1E
MSNPs-3(2)
88.1
86.9
87.4
85.3
85.6
82.7
11.9
13.1
12.6
14.7
14.4
17.3
–
11.2Æ0.1
–
13.9Æ0.2
–
11.3Æ0.7
MSNPs-4(2)-1E
[a] The sulfur content corresponds to the amount of ligands being conju-
gated onto the surface of MSNPs.
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Acknowledgements
This research is supported by the National Research Founda-
tion (NRF), Prime Minister’s Office, Singapore under its NRF Fel-
lowship (NRF2009NRF-RF001-015) and Campus for Research Ex-
cellence and Technological Enterprise (CREATE) programme—
Singapore Peking University Research Centre for a Sustainable
Low-Carbon Future, and the NTU-A*Star Centre of Excellence
for Silicon Technologies (A*Star SERC No.: 112 351 0003).
Keywords: drug delivery · ligand effects · mesoporous silica
nanoparticles · release kinetics · surface functionalization
[1] a) D. Tarn, C. E. Ashley, M. Xue, E. C. Carnes, J. I. Zink, C. J. Brinker, Acc.
Chem. Res. 2013, 46, 792–801; b) M. W. Ambrogio, C. R. Thomas, Y.-L.
Zhao, J. I. Zink, J. F. Stoddart, Acc. Chem. Res. 2011, 44, 903–913; c) M.
Vallet-Regꢂ, F. Balas, D. Arcos, Angew. Chem. 2007, 119, 7692–7703;
Angew. Chem. Int. Ed. 2007, 46, 7548–7558; d) J. L. Vivero-Escoto, I. I.
Slowing, B. G. Trewyn, V. S. Y. Lin, Small 2010, 6, 1952–1967; e) Z. Li,
J. C. Barnes, A. Bosoy, J. F. Stoddart, J. I. Zink, Chem. Soc. Rev. 2012, 41,
2590–2605.
Figure 6. a) Percentage cell viability of B16-F10 after incubation with
MSNPs-4(0.5)-1E, MSNPs-4(1)-1E, and MSNPs-4(2)-1E for 4 h before chang-
ing the culture media. b) Fluorescence microscopy images of the B16-F10
cell line after incubation with MSNPs-4(0.5)-1E, MSNPs-4(1)-1E, and MSNPs-
4(2)-1E.
[2] a) N. K. Mal, M. Fujiwara, Y. Tanaka, Nature 2003, 421, 350–353; b) C.
Wang, Z. Li, D. Cao, Y.-L. Zhao, J. W. Gaines, O. A. Bozdemir, M. W. Am-
brogio, 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; c) Z.
Li, J. L. Nyalosaso, A. A. Hwang, D. P. Ferris, S. Yang, G. Derrien, C. Char-
nay, J.-O. Durand, J. I. Zink, J. Phys. Chem. C 2011, 115, 19496–19506;
d) A. Bernardos, L. Mondragon, E. Aznar, M. D. Marcos, R. Martinez-
Manez, F. Sancenon, J. Soto, J. M. Barat, E. Perez-Paya, C. Guillem, P.
Amoros, ACS Nano 2010, 4, 6353–6368.
[3] a) I. Slowing, B. G. Trewyn, V. S. Y. Lin, J. Am. Chem. Soc. 2006, 128,
14792–14793; b) A. Nieto, M. Colilla, F. Balas, M. Vallet-Regi, Langmuir
2010, 26, 5038–5049; c) J. C. Doadrio, E. M. B. Sousa, I. Izquierdo-Barba,
A. L. Doadrio, J. Perez-Pariente, M. Vallet-Regi, J. Mater. Chem. 2006, 16,
462–466.
[4] a) E. Aznar, L. Mondragꢁn, J. V. Ros-Lis, F. Sancenꢁn, M. D. Marcos, R.
Martꢂnez-MꢃÇez, J. Soto, E. Pꢄrez-Pꢃya, P. Amorꢁs, Angew. Chem. 2011,
123, 11368–11371; Angew. Chem. Int. Ed. 2011, 50, 11172–11175; b) Z.
Luo, K. Cai, Y. Hu, B. Zhang, D. Xu, Adv. Healthcare Mater. 2012, 1, 321–
325; c) Y.-L. Zhao, Z. Li, S. Kabehie, Y. Y. Botros, J. F. Stoddart, J. I. Zink, J.
Am. Chem. Soc. 2010, 132, 13016–13025; d) C. R. Thomas, D. P. Ferris, J.-
H. Lee, E. Choi, M. H. Cho, E. S. Kim, J. F. Stoddart, J.-S. Shin, J. Cheon,
J. I. Zink, J. Am. Chem. Soc. 2010, 132, 10623–10625; e) K. Patel, S. An-
gelos, W. R. Dichtel, A. Coskun, Y.-W. Yang, J. I. Zink, J. F. Stoddart, J. Am.
Chem. Soc. 2008, 130, 2382–2383; f) H. Yan, C. Teh, S. Sreejith, L. Zhu, A.
Kwok, W. Fang, X. Ma, K. T. Nguyen, V. Korzh, Y. Zhao, Angew. Chem.
2012, 124, 8498–8502; Angew. Chem. Int. Ed. 2012, 51, 8373–8377;
g) C. Park, K. Lee, C. Kim, Angew. Chem. 2009, 121, 1301–1304; Angew.
Chem. Int. Ed. 2009, 48, 1275–1278; h) C. Y. Ang, S. Y. Tan, X. Wang, Q.
Zhang, M. Khan, L. Bai, S. T. Selvan, X. Ma, L. Zhu, K. T. Nguyen, N. S. Tan,
Y. Zhao, J. Mater. Chem. B 2014, 2, 1879–1890.
[5] a) J. Lai, B. P. Shah, E. Garfunkel, K.-B. Lee, ACS Nano 2013, 7, 2741–
2750; b) B. A. Kellogg, L. Garrett, Y. Kovtun, K. C. Lai, B. Leece, M. Miller,
G. Payne, R. Steeves, K. R. Whiteman, W. Widdison, H. Xie, R. Singh,
R. V. J. Chari, J. M. Lambert, R. J. Lutz, Bioconjugate Chem. 2011, 22,
717–727; c) P. E. Thorpe, P. M. Wallace, P. P. Knowles, M. G. Relf, A. N. F.
Brown, G. J. Watson, D. C. Blakey, D. R. Newell, Cancer Res. 1988, 48,
6396–6403; d) A. N. Koo, H. P. Rim, D. J. Park, J.-H. Kim, S. Y. Jeong, S. C.
Lee, Macromol. Res. 2013, 21, 809–814.
tionship with the corresponding release profiles (Figure 5). This
trend is reasonable because a higher level of drug released
would translate to higher cell death, thus causing lower cell vi-
ability. At concentrations lower than 20.8 mgmLÀ1, the differ-
ence between the three types of functional MSNPs became in-
significant. Therefore, we employed the concentration of
20.8 mgmLÀ1 for in vitro imaging studies. From the fluores-
cence images shown in Figure 6b, it was observed that
MSNPs-4(0.5)-1E gave the strongest red fluorescence intensity
compared to the other two samples. This red fluorescence is
attributed to DOX, and the intensity observed is proportional
to the amount of DOX released into the cells. Hence, it could
be concluded that MSNPs-4(0.5)-1E showed the highest level
of DOX release followed by MSNPs-4(1)-1E and lastly MSNPs-
4(2)-1E. This observation once again supports the results ob-
tained from the cell viability assay as well as the release profile
studies.
In conclusion, this work has adopted a supramolecular ap-
proach in analyzing the effects of the nature of functional
groups on the surface of MSNPs on drug-loading capacity and
release kinetics. A catalysis screening method has been em-
ployed by first varying chain length, then terminal group, and
lastly the amount of thiol units on the surface. The ligand with
an intermediate length (5 carbon atoms) and a bulky terminal
group (cyclohexyl) shows the highest drug loading and release
capacity under the same conditions. To achieve a high drug-re-
lease capability, the surface coverage of the ligands should be
relatively low. This observation has been further supported by
in vitro studies, proving the applicability of the screening
method. Guided by the results obtained from this work, better
drug delivery systems may be designed for more efficient
cancer treatment.
[6] a) J. A. Cook, H. I. Pass, S. N. Iype, N. Friedman, W. Degraff, A. Russo, J. B.
Mitchell, Cancer Res. 1991, 51, 4287–4294; b) M. J. Allalunisturner, F. Y. F.
Lee, D. W. Siemann, Cancer Res. 1988, 48, 3657–3660.
[7] a) X. Ma, K. T. Nguyen, P. Borah, C. Y. Ang, Y. Zhao, Adv. Healthcare Mater.
2012, 1, 690–697; b) Y. Cui, H. Dong, X. Cai, D. Wang, Y. Li, ACS Appl.
Mater. Interfaces 2012, 4, 3177–3183; c) Z. Luo, K. Cai, Y. Hu, L. Zhao, P.
Liu, L. Duan, W. Yang, Angew. Chem. 2011, 123, 666–669; Angew. Chem.
Int. Ed. 2011, 50, 640–643; d) Q. Zhang, F. Liu, K. T. Nguyen, X. Ma, X.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
Wang, B. Xing, Y. Zhao, Adv. Funct. Mater. 2012, 22, 5144–5156; e) H.
Kim, S. Kim, C. Park, H. Lee, H. J. Park, C. Kim, Adv. Mater. 2010, 22,
4280–4283.
[11] a) Z. Luo, X. Ding, Y. Hu, S. Wu, Y. Xiang, Y. Zeng, B. Zhang, H. Yan, H.
Zhang, L. Zhu, J. Liu, J. Li, K. Cai, Y. Zhao, ACS Nano 2013, 7, 10271–
10284; b) Q. Zhang, X. Wang, P.-Z. Li, K. T. Nguyen, X.-J. Wang, Z. Luo, H.
Zhang, N. S. Tan, Y. Zhao, Adv. Funct. Mater. 2014, 24, 2450–2461.
[8] L. Du, S. Liao, H. A. Khatib, J. F. Stoddart, J. I. Zink, J. Am. Chem. Soc.
2009, 131, 15136–15142.
[9] E. Aznar, F. Sancenꢁn, M. D. Marcos, R. Martꢂnez-MꢃÇez, P. Stroeve, J.
Cano, P. Amorꢁs, Langmuir 2012, 28, 2986–2996.
[10] P. Nadrah, U. Maver, A. Jemec, T. Tisler, M. Bele, G. Drazic, M. Bencina, A.
Pintar, O. Planinsek, M. Gaberscek, ACS Appl. Mater. Interfaces 2013, 5,
3908–3915.
Received: May 16, 2014
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
COMMUNICATION
&
Drug Delivery
S. Y. Tan, C. Y. Ang, P. Li, Q. M. Yap,
Y. Zhao*
&& – &&
Drug Encapsulation and Release by
Mesoporous Silica Nanoparticles: The
Effect of Surface Functional Groups
Maximizing drug release from drug
carriers is important to achieve better
therapeutic efficiency. Using glutathione
(GSH)-triggered drug-release systems as
a case study, the effect of the nature of
surface functional groups on mesopo-
rous silica nanoparticles on drug-load-
ing and drug-release capabilities was in-
vestigated.
Controlled Drug Delivery
Maximizing drug release from drug carriers is important to
achieve better therapeutic efficiency. In their
Communication on page && ff., Y. Zhao et al. show how
a catalysis screening technique can be adopted for
analyzing the effects of chain length, terminal group, and
density of functional ligands on the surface of mesoporous
silica nanoparticles on drug-loading capacity and
glutathione-triggered drug-release kinetics. The results
obtained may serve as a general guide for developing more
effective carriers for drug delivery.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!