.
Angewandte
Communications
formed to monitor the evolution of the nanoparticle surface
charges. MCM-48 MSNs show a zeta potential of À30 mV
owing to anionic silanol groups. M-IP and M-IP-SZ exhibited
similar zeta potential values of À40 mV and À20 mV,
respectively, which is due to the residual silanol groups, with
some modulation of the zeta potential caused by functional-
ization with organic groups. The highly negative zeta potential
means that the particles are stable in suspension (Table S1).
These changes in zeta potential and the particle size increase
for M-IP (230 nm) and M-IP-SZ (285 nm) are both in line
with the results of the EA and TGA analyses. Note that with
a particle size of about 285 nm, M-IP-SZ remains small
enough to be used for targeted delivery applications.[5a]
Additionally, Fourier-transformed infrared (FTIR) spectra
of the native and functionalized hybrid silica nanoparticles
substantiated the proper surface functionalization (Fig-
ure S5). The absorbance bands related to SZ and the
iodopropyl group were found in M-IP-SZ further validating
successful chemical binding.
produced by colonic anaerobic microflora, leading to forma-
tion of primary amines.[23,16] Hence, we hypothesized that both
5-ASA and SP will be released from the silica nanocarriers in
the presence of the enzyme, co-delivering both molecules at
the colonic site (Scheme 1 and Supporting Information,
Reaction S2). To predict the viability of the enzyme-respon-
sive system, in vitro drug release experiments were carried
out in a simulated intestinal fluid with anaerobic bacteria that
release azoreductase (for experimental details, see the
Supporting Information). The release of 5-ASA and SP was
monitored both qualitatively and quantitatively using UV/Vis
spectroscopy and high performance liquid chromatography
(HPLC), respectively (Figures S7 and S8).[24] Figure 2 shows
the percentage release of 5-ASA and SP from M-IP-SZ at
various pH values, with and without the presence of azo-
reductase, confirming the selective process of reduction of
enzyme-treated SZ into SP and 5-ASA. As expected, both
molecules were released (5-ASA 35% and SP 55% release),
which further validates our hypothesis of an attachment from
Solid-state 13C CP NMR and 29Si MAS NMR were
performed to further confirm functionalization and to inves-
tigate the nature of the bonding between the propyl group
and SZ (Figure 1c and Figure S6, respectively). The 3-
iodopropyl functionalized MCM-48 particles (M-IP) show
three distinct signals in the alkane region with chemical shifts
two ends of the SZ molecule through a C N or C O bond.
These results are corroborated by color changes (insert in
Figure 2) wherein the SZ solution changes from yellow to
colorless after bacterial conversion. Similarly, SZ-immobi-
lized silica particles turn from yellow to white after biocon-
version of the covalently attached SZ in M-IP-SZ, with no
color change observed in samples without enzyme treatment,
thus further verifying our hypothesis. Therefore, our study
establishes that SZ remains covalently bound to the MSNs
avoiding premature release in the absence of a trigger.
À
À
À
À
À
À
at 8.5 ppm ( CH2 I), 13.9 ppm (CH2 CH2 CH2), and
À
À
27.6 ppm ( CH2 Si) corresponding to the propyl carbon
atoms, and a fourth weak resonance is observed at 50.9 ppm
indicating a small amount of residual methoxysilane groups
[20a,21]
À À
(Si O CH3).
In addition to the propyl signals in the 5–
The reduction of pure SZ at pH 1.2 (simulated gastric
fluid), pH 7.4 (PBS), and in PBS with bacteria was also tested
to study the effect of pH on the release/degradation of SZ
(Figure S9a). At pH 1.2 and at pH 7.4, no degradation of SZ
was observed in the absence of bacteria. Conversely, more
than 90% of SZ, in the form of pure SZ, was reduced in the
presence of the bacteria. Moreover, no significant leaching of
SZ was observed at pH 1.2 or pH 7.4 from M-IP-SZ,
indicating that the bonding is strong enough to ensure that
the functionalized MSNs can transit through the stomach
30 ppm region, M-IP-SZ displays resonances in the regions:
50–65 ppm owing to the C3 carbon atom of the propyl group,
which forms a covalent bond with the N or O atoms of SZ;
signals around 110–165 ppm can be ascribed to aromatic ring
carbon atoms; and greater than 165 ppm is due to the
carbonyl group. The number of resonances in the upfield
region of the spectrum (less than 65 ppm) clearly indicates
that a mixture of compounds is present, owing to the
nucleophilic N and O atoms of SZ reacting with the 3-
iodopropyl groups. These data confirm that SZ is covalently
À
À
bound to the MSNs by way of C N and C O linkages.
Notably, particles prepared in the absence of triethylamine
(TEA) show no evidence of SZ attachment by 13C CP NMR
(Figure 1c) and elemental analysis (Table S2), indicating that
the reaction occurs only in the presence of base.[20] The 29Si
MAS NMR data of native MCM-48, M-IP, and M-IP-SZ
samples are depicted in Figure S6. The non-functionalized
MCM-48 nanospheres show three signals in the Q region
attributed to different types of silanol groups present on the
surface of the silica. Signals at À93 ppm, À101 ppm, and
À110 ppm correspond to Q2, Q3, and Q4 species, respec-
tively.[22] The spectra of the functionalized M-IP and M-IP-SZ
reveal three signals in the Q region and two additional signals
À
in the T region confirming the presence of Si C bonds
originating from the organosilane grafting, that is, yielding T2
and T3 signals.[13c]
The driving force for the triggered release of drug from M-
IP-SZ is reductive cleavage of the aromatic azo bond of SZ,
which is a substrate of azo-reductase, an extracellular enzyme
Figure 2. Percentage release of 5-ASA and SP from M-IP-SZ at different
pH values and in the presence and absence of bacterial azo-reductase.
Insert: color change owing to bioconversion of SZ into 5-ASA and
sulfapyridine upon treatment with azo-reductase.
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 12486 –12489