You are on page 1of 7

Biomaterials 29 (2008) 45544560

Contents lists available at ScienceDirect

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

The targeting of surface modied silica nanoparticles to inamed


tissue in experimental colitis
Brice Moulari a, David Pertuit a, Yann Pellequer a, Alf Lamprecht a, b, *
a
Laboratory of Pharmaceutical Engineering, University Franche-Comte, Besanon, France
b
Pharmaceutical Engineering and Biopharmaceutics, Institute of Pharmacy, University of Bonn, Germany

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

Article history: One aspect in the emerging eld of nanomedicine is site specic drug delivery via nanoparticles. The use
Received 19 June 2008 of nanoparticles allows for increased therapeutic efciency with a lowered risk for and extent of adverse
Accepted 20 August 2008 reactions resulting from systemic drug absorption. 5-Amino salicylic acid (5ASA) loaded silica nano-
Available online 13 September 2008
particles (SiNP) are proposed here as drug delivery system for specic accumulation in inamed colonic
tissues allowing for selective medication delivery to such inammation sites. The drug was covalently
Keywords:
bound to SiNP by a four-step reaction process. In-vitro toxicity of modied SiNP was tested in appropriate
Drug targeting
cell culture systems, while targeting index and therapeutic efciency were evaluated in a pre-existing
Nanoparticles
Drug delivery colitis in mice. Particle diameter was around 140 nm after nal surface modication. In-vitro drug release
Colon targeting demonstrated signicant drug retention inside the NP formulation. Toxicity of the different formulations
Colitis was evaluated in-vitro cell culture exhibiting a lowered toxicity for 5ASA when bound to SiNP. In-vivo,
oral SiNP were found to accumulate selectively in the inamed tissues allowing for signicant amounts
of drug load. SiNP demonstrated their therapeutic potential by signicantly lowering the therapeutically
necessary drug dose when evaluating clinical activity score and myeloperoxidase activity (untreated
control: 28.0  5.0 U/mg; 5ASA-solution (100 mg/kg): 8.2  3.4 U/mg 5ASASiNP (25 mg/kg): 5.2  2.4
U/mg). SiNP allow to combine advantages from selective drug targeting and prodrugs appearing to be
a promising therapeutic approach for clinical testing in the therapy of inammatory bowel disease.
2008 Elsevier Ltd. All rights reserved.

1. Introduction Although many efforts have been made for a higher specicity of
drug release by designing new drug delivery devices [2,3], all
The general principle of drug treatment in inammatory bowel marketed delivery systems still appear to be insufciently selective
disease (IBD) is to induce remission of outbreaks and to prevent [4]. This is due to the fact that the drug release mechanisms are
outbreaks during remission. First-line therapy for patients with IBD based on physiological parameters which are not related to the
has centred on sulfasalazine for decades. The identication of inammation and barely to its location. Drug delivery is usually
mesalamine 5ASA as the active moiety responsible for the luminal either triggered by the luminal pH in the gastrointestinal tract, the
anti-inammatory effects of sulfasalazine, as well as patient allergy, enzymatic activity of colonic bacteria, or the transit time of the drug
intolerance, or unresponsiveness to this agent has led to the carrier normally initiating the drug release in the distal ileum. As
development of multiple 5ASA conjugates, each with its own the inamed areas affect only limited sectors of the intestinal tissue
specic delivery system [1]. It is thus of tremendous importance to and may vary from patient to patient, consequently, pronounced
deliver 5ASA locally in order to reduce inuences by systemic drug drug amounts are still delivered unintentionally to non-inamed
absorption causing adverse effects and drug loss lowering the regions. This misdirected drug release can be hardly avoided by the
probability for a therapeutic success. current state-of-the-art drug delivery systems, thus remaining
a general problem responsible for undesirable adverse effects and
potentially even therapy failure.
Among several new therapeutic approaches, a strategy was
proposed to target the strong cellular immune response occurring
* Corresponding author. Laboratory of Pharmaceutical Engineering, Faculty of
Medicine and Pharmacy, University of Franche-Comte, 25000 Besanon, France.
in the inamed regions, i.e., in general, an increased presence of
Tel.: 33 3 81 66 55 48; fax: 33 3 81 66 52 90. neutrophils, natural killer cells, mast cells, and regulatory T cells,
E-mail address: alf.lamprecht@univ-fcomte.fr (A. Lamprecht). which play an important role in the pathophysiology of

0142-9612/$ see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2008.08.009
B. Moulari et al. / Biomaterials 29 (2008) 45544560 4555

inammatory bowel disease [5,6]. It was proven that particle was added. The mixture was reuxed for 12 h. Final particles were puried by
uptake into those immune-related cells or the disrupted intestinal ltration and washing steps with acetonitrile. Fluorescent SiNP were prepared
accordingly while uoresceinamine was used instead of Me5ASA.
barrier at ulcerated regions [7] allows the selective accumulation of Thin layer chromatography analysis (eluant: ethyl acetate/hexane 7:3 v/v) was
the particulate carrier system in the targeted area. used to conrm coupling of Me5ASASiNP. The structural analyses were performed
The increased adhesion of small particulate drug carriers to the via infrared spectroscopy (Shimadzu FTIR-8201 PC, Japan).
inamed tissue in ulcerative colitis led to a new therapeutic NP were analysed for diameter, size distribution, and zeta potential in 1:15 (v/v)
dilutions with distilled water using a Zetasizer (Zetasizer Nano ZS, Malvern
concept allowing for specic drug targeting in this disease [811].
Instruments, Worcestershire, UK).
This approach is mainly based on two pathophysiological changes Atomic force microscopic imaging was conducted by use of a Nanoscope III with
in the inamed tissue, allowing the higher adhesion of the carriers Bioscope IV controller (Digital Instruments, USA). All experiments were conducted
in the inamed tissue caused by elevated levels of mucus produc- at room temperature in air in Tapping Mode with NSC 16 tips at a resonant
tion and an intensied particle uptake inside the colitis tissue as frequency of approximately 200 kHz. A droplet of the sample dissolved in deionized
water (20 ml) was placed on a freshly cleaned cover slide and was allowed to dry
a result of an enhanced permeability and the presence of a highly before being subjected to AFM investigations.
increased number of immune related cells. This accumulation
phenomenon was observed to be particle size dependent with an 2.2. In vitro drug release
increased adhesiveness for smaller particle diameters [7]. A recent
study on the comparative efciency of this new delivery strategy 50 mg of lyophilized drug-loaded NP were re-suspended in a 15 ml ask con-
and a pH-sensitive approach revealed a major drawback of the new taining either phosphate buffer (pH 7.4) or articial intestinal uid (according to US
Pharmacopoeia 23) and incubated at 37  C under magnetic stirring (200 rpm). At
system, namely the drug leakage during the intestinal passage of
appropriate intervals, 0.5 ml samples were withdrawn, centrifuged at 15,000 g for
nanoparticles [12]. An early uncontrolled release of the active 30 min, and replaced by fresh buffer. Drug loads were determined by dispersing the
molecule reduces the therapeutic drug efciency and may give rise same amount of NP in 1 M NaOH at 37  C until particles were completely degraded.
to adverse effects. Due to the diffusion out of the small particles All experiments were performed in triplicate. The drug release was quantied by use
of an isocratic high performance liquid chromatographic method developed to our
related to the enormous surface, the encapsulated drug is usually
specic requirements allowing simultaneous detection of both, 5ASA and Me5ASA.
leaking easily towards the surrounding aqueous phase [13]. An SG5-ODD1-15QS 150  4.6 mm, 5 mm column (Interchim, France) constituted the
Subsequently, the probability that the drug molecules reach the stationary phase while the mobile phase consisted of acetonitrilecitric acid 0.01 M
colon is limited, especially since this leakage phenomenon is buffer pH 2 in a ratio 2:98 at a ow rate of 1 ml/min. Both, 5ASA and Me5ASA were
further enhanced for hydrophilic drugs such as 5ASA. detected by UV absorbance at 340 nm. Samples of 20 ml were injected into the
column. The limit of detection of 5ASA and of Me5ASA was approximately 1 mg/ml.
The use of silica nanoparticles (SiNP) in the biomedical eld has
The retention time of 5ASA and Me5ASA was circa 3.5 and 7.4 min, respectively.
been progressing for several years with a main focus on cell
recognition [14] for diagnostic purposes as well as drug and gene 2.3. NP toxicity test in cultured cells
delivery [1517]. In cases where SiNP were utilized as medication
carriers, drugs were adsorbed onto the surfaces of the NP. The drug Caco-2 and HEK cells were seeded in 96-well plates (approximately 10,000 cells
release mechanism was triggered by simple desorption kinetics, per well) and grown in Dulbeccos Modied Eagle Medium for one week. Thereafter,
cells were incubated with the various preparations at different dilutions for 24 h.
which are poorly controllable in complex biological liquids (e.g.
After carefully removing the supernatant and two washing steps with KrebsRinger
blood, lymph, and gastrointestinal juices). One major advantage of buffer, cell viability was determined via MTT test. Cytotoxicity was expressed as
SiNP is that they are deemed toxicologically safe, which they have a percentage of controls (untreated cells).
proven in 50 years of use as a pharmaceutical excipient for oral drug
delivery. 2.4. Colitis model
We report here on the design of SiNP with modied surface for
All animal experiments were carried out in accordance with the recommen-
selective drug delivery towards inamed tissue in chronic inam-
dations in the Guide for the Care and Use of Laboratory Animals (Institute of
matory diseases of the intestine. To date, ordinary nanoscale drug Laboratory Animal Resources, National Research Council, National Academy of
delivery systems tend to prematurely release drug compounds that Sciences, US). The TNBS mouse model was chosen as a well-recognized experi-
are physically entrapped or adsorbed. As a solution to that problem, mental model [18] that allows induction of a colitis at an exact location. Male BALB/c
we propose here to bind the anti-inammatory drug covalently to mice (average weight 25 g, n 6/group) were used for the inammation model
where inammation was induced by TNBS via the following procedure: animals
the surface of SiNP. The resulting chemical bond is biodegradable
were catheterized 4 cm intrarectally after light narcotizing with ether. 100 ml of TNBS
and intended to considerably delay drug release when compared to in ethanol were applied in a dose of 160 mg/kg body weight. The mice were housed
conventional physicochemical encapsulation. Minimizing unin- for a day without treatment to establish the full model colitis.
tended drug release during intestinal passage ensures targeted During the treatment period, all animals received orally either 0.1 ml of 5ASA
solution (at a dose of 30 or 100 mg/kg body weight) or Me5ASASiNP suspensions
drug delivery to the site of action.
once daily for six consecutive days. The control groups received saline only (colitis
control) or blank SiNP. The animals were sacriced 24 h after the last drug/particle
administration and their colons were resected.
2. Methods

2.1. DrugSiNP coupling 2.5. Confocal laser scanning microscopy for the qualitative localization of SiNP

The rst step was protective esterication of 5ASA. To a solution of anhydrous A Biorad MRC 1024 Laser Scanning Confocal Imaging System (Hemel Hemp-
methanol (83.4 ml), containing 5ASA (0.033 mol) and cooled at 0  C, thionyl chloride stead, UK), equipped with an argon ion laser (American Laser Corp., Salt Lake City,
(7.7 ml) was added slowly. The mixture was stirred at room temperature for 1 h. The UT, USA) and a Zeiss Axiovert 100 microscope (Carl Zeiss, Oberkochen, Germany),
reaction mixture was then heated under reux for 4 h. Afterwards solvent was was used to qualitatively detect the uorescent particles in the tissue sections. The
removed under reduced pressure. The resulting precipitate was dissolved in ether laser was adjusted in the green uorescence mode, which yielded an excitation
and the solution was ltered. The ltrate was dissolved in a 1:1 (v/v) water/ether wavelength at 488 nm.
mixture at a pH of 9 prior to several washing steps with water and ether followed by
a drying step with sodium sulphate. 2.6. Pathophysiological parameters
Silica nanoparticles (3 g) were dispersed in toluene (100 ml). 3-Aminopropyl- The degree of inammation was quantied by use of a clinical activity index
trimethoxysilane (10 ml) was added to the mixture under stirring and reuxed for assessing weight loss, stool consistency, and rectal bleeding as previously described
12 h. The resulting particles (2.85 g) were puried and dispersed in acetonitrile elsewhere [19].
(100 ml) containing succinic anhydride (10 g). The solution was stirred and reuxed Resected colon tissue samples were opened longitudinally and rinsed with iced
for 12 h and the particles were then puried by washing with acetonitrile. Puried phosphate buffer to remove luminal content. Then tissue wet weight and colon
intermediate particles (2.85 g) were dispersed in acetonitrile containing succinimid length were determined and expressed as a colon weight/length quotient [20]. The
anhydride (10 g) and reuxed for 12 h followed by ltration and washing with measurement of myeloperoxidase activity was performed to quantify the severity of
acetonitrile. Particles were redispersed in acetonitrile (100 ml) and 1 g of Me5ASA the colitis. Myeloperoxidase activity is a reliable index of severity of inammation
4556 B. Moulari et al. / Biomaterials 29 (2008) 45544560

caused by inltration of activated neutrophils into inamed tissue. Enzymatic Me5ASA by high performance liquid chromatography, an amount of
activity was analyzed according to a standard method [21]. 151  62 mg of coupled Me5ASA per gram of SiNP was determined.
2.7. Statistical analysis
The hydrodynamic mean diameter of SiNP increased with every
The results were expressed as mean values  S.D. For the analysis of statistical reaction step from 101 12 nm to a nal 136  22 nm. Atomic force
signicance ANOVA on ranks was applied followed by Dunns test for all pairwise microscopy images showed SiNP as more or less spherically shaped
comparison. In all cases, P < 0.05 was considered to be signicant. (Fig. 2A) while the surface modication did not appear to alter the
overall morphology aspect. A generally negative zeta potential
contributed to excellent stability of particle suspension.
3. Results and discussion Since minimal drug release along the gastrointestinal transit is
crucial in order to accumulate maximal drug amounts at the
Prior to forming a bond between 5ASA and SiNP, the esteri- inammation site, drug leakage experiments were performed in-
cation of 5ASA into methyl-5-ASA (Me5ASA) is necessary in order vitro in phosphate buffer or simulated intestinal uid (Fig. 1B). No
to avoid the dimerisation of 5ASA during its reaction with SiNP. The initial leakage was observed where distinct amounts of drug are
principal coupling reaction consisted of afxing Me5ASA to the usually released similarly to a burst effect. While simple phosphate
surface of SiNP making necessary a prior functionalization of the
SiOH groups. The xation of (3-aminopropyl)-trimethoxysilane on
the SiNP surface was followed by an addition of succinic anhydride.
Finally, after activation by succinimid anhydride Me5ASA was 120
A
coupled to SiNP surfaces. Indirectly dosing free (non-coupled)

100
m
2.0 m
viability [%] 80

1.5
60

Caco-2 cells:
1.0
1 40 5ASA
Me5ASA
SiNP
1.0 1 20 Me5ASA-SiNP

0.5 0
2
1.5 1E-4 1E-3 0.01 0.1 1 10
1.0 formulation conc. [mg/ml]
0.5

B 120
100
pancreatin buffer HEK cells:
phosphate buffer 100 5ASA
80 Me5ASA
SiNP
Me5ASA release [%]

80
Me5ASA-SiNP
viability [%]

60
60

40
40

20
20

0 0
0 12 24 36 48 1E-3 0.01 0.1 1 10
time [h] formulation conc. [mg/ml]
Fig. 1. Atomic force microscopic image of Me5ASASiNP (upper). Cumulated Me5ASA/ Fig. 2. Me5ASASiNP compared with equivalent blank SiNP or 5ASA and Me5ASA
5ASA release versus time of Me5ASASiNP (lower) in phosphate buffer of pH 6.8 or solutions of similar concentrations were tested for their toxicity to Caco-2 (A) and
simulated intestinal uid (pancreatin buffer), respectively (n 3; data are shown as HEK293 (B) cells after incubation for 8 h (n 4; data are shown as mean  S.D.). The
mean  S.D.). NP do not show a burst release, which would be identied by the greater sensitivity of HEK cells led to their slightly higher toxicity. However, the
presence of free Me5ASA. general toxicity of the new system can be considered as minimal.
B. Moulari et al. / Biomaterials 29 (2008) 45544560 4557

buffer systems did not initiate drug release from SiNP in the of drug can be enzymatically released. This is consequently
analyzed interval, enzymatic cleavage released Me5ASA after a lag responsible for the delayed degradation of the peptide bonds
time of around 8 h. This time frame is largely sufcient to reach between SiNP and Me5ASA and thus the drug release.
inamed colon without prior drug loss during intestinal transit. The In the drug release experiments, Me5ASA and 5ASA can be
lag time is potentially based on two phenomena. The chemical found together in the release medium with a relative increasing
modication of the SiNP surface involves the integration of amount of 5ASA with time. This suggests an enzymatic cleavage of
hydrophobic chemical entities which turns the SiNP surface less Me5ASA into 5ASA potentially triggered by esterases which is,
accessible for enzymes. A second aspect is the density of Me5ASA however, relatively slow and needs to be studied in detail. Under in-
on the surface which itself may act as an inhibitor of accessibility to vivo conditions various other mechanisms may have to be
enzymes for steric reasons, and only with time signicant amounts considered.

Fig. 3. Confocal laser scanning microscopy images of colon cross-sections from uorescently labelled SiNP administered to healthy mice (A) and to mice suffering from colitis group
showing low adhesion to the mucus (B) but increased accumulation in ulcerated regions of the inamed tissue (C). Scale bars represent 100 mm.
4558 B. Moulari et al. / Biomaterials 29 (2008) 45544560

In-vitro toxicity was investigated in standard cell lines, namely revealed higher therapeutic effects with Me5ASASiNP at 25 or
human colonic carcinoma cells (Caco-2) and human embryonic 50 mg/kg than 5ASA at 100 mg/kg. Moreover, signicant differ-
kidney cells (HEK 293). Cell culture studies revealed only marginal ences were observed between all Me5ASASiNP groups and 5ASA
differences in toxicity between 5ASA and Me5ASA (Fig. 2). The at 30 mg/kg. In all cases SiNP were without therapeutic effect and
absence of differences between unaltered SiNP and Me5ASA-SiNP results obtained were comparable to untreated colitis.
allows the conclusion that chemical cross-linking reactions have For targeting towards inamed tissues in IBD therapeutic agents
a negligible inuence on toxicity. A lower toxicity of Me5ASASiNP of rst choice (e.g. aminosalicylates, corticosteroids) have been
compared to Me5ASA or 5ASA may be explained by the toxicity developed in special galenic forms to accomplish the delivery of the
limiting availability of bond compared to free drug. active compounds to the terminal ileum and colon [22]. However, it
In histological analyses with uorescently labeled SiNP, they has to be realized that intestinal physiology, severity of IBD as well as
were found to selectively accumulate in inamed tissue while drug disposition demonstrates large inter-individual differences
healthy tissue exhibited limited particle adhesion (Fig. 3). Quanti- resulting in variable clinical response rates. Some of the controlled
tative analyses exhibited a sixfold higher adhesion of SiNP to release preparations release sufcient amounts of 5ASA already in
inamed tissue than in healthy control groups (Fig. 4). the small bowel which could provide some additional benet in
In order to evaluate therapeutic benet, Me5ASASiNP were Crohns disease while colon specic delivery by other galenic
studied on a preexistenting TNBS-colitis model in mice. After formulation and azo-prodrugs will be only effective in patients with
inducing experimental colitis, clinical activity indices increased ulcerative colitis. In the therapy of ulcerative colitis, following the
dramatically within 24 h in response to intestinal inammation administration of azo-prodrugs (e.g. olsalazine), lower plasma
(Fig. 5). The decrease of clinical activity became visible on day 6 for concentrations and higher delivery into the colon of 5ASA were
both Me5ASASiNP and 5ASA treatments. Levels for colitis control found when compared with special galenic formulations of 5ASA
and SiNP treatment were similar during the entire treatment [23]. It should, however, kept in mind that with these systems still
period at constant high values. All 5ASA containing formulations signicant amounts of drug are delivered unintentionally to non-
reduced clinical activity as compared to the untreated colitis inamed regions, since drug release mechanisms are triggered by
control group. Differences became statistically signicant on day 6. physiological parameters which are not directly related to the
Similar results were noted in response to administration of oral inammation nor its location.
5ASA solutions and oral NP formulations although a slightly more As an alternative therapeutic concept, NP administered by
effective treatment was observed with 5ASA at 100 mg/kg. oral route are expected to build up in the inamed tissue to
However, all differences were statistically signicant only in achieve a more local effect through their accumulation at the
comparison to the untreated control group. site of action [7,10,11]. This strategy is mainly based on the
Similar to observations made regarding clinical activity index, recognized correlation between NP accumulation and the strong
colon weight/length ratios after 5ASA treatments were found to cellular immune response occurring at sites of inammation.
be lower in comparison to the colitis control (Fig. 6A). All 5ASA Increased permeability and the presence of high numbers of
containing formulations showed a myeloperoxidase activity immune related cells lead to intense particle uptake inside
signicantly different from the untreated group (Fig. 6B). Myelo- colitis tissue. As shown in earlier studies [10], polymeric NP
peroxidase activity, an enzyme present in inltrating neutrophils provided high selectivity in terms of adhesion to inamed tissue
and thus an excellent indicator for the severity of inammation, sites. However, a major disadvantage was a considerably large
initial drug loss by uncontrolled burst release as well as

30
healthy
5
inflamed
25 non-inflamed treatment
* *
4
* *
adhering SiNP [%]

20
clinical activity score

3
15

10 2
colitis control
SiNP
5ASA 30mg
5
1 5ASA 100mg
TNBS

Me5ASA-SiNP 50mg
Me5ASA-SiNP 25mg
0 Me5ASA-SiNP 25mg rect
blank SiNP Me5ASA-SiNP
0
Fig. 4. Penetrative behavior of SiNP or Me5ASASiNP into healthy, inamed, or non-
0 2 4 6 8
inamed tissue (healthy tissue surrounding gross inammation) samples was analyzed day
(n 6; data are shown as mean  S.D.; *P < 0.05 for differences between healthy and
inamed group tissue samples). Confocal images show an accumulation of uo- Fig. 5. Clinical activity index in the TNBS mouse model during the whole experimental
rescently labeled SiNP inside the inamed tissue while signals from comparable period after oral or rectal drug administration (NP controls similar to the colitis control
healthy areas are negligible. Quantitative analyses of this adhesion phenomenon as well as error bars are not shown for clarity reasons; n 6; *P < 0.05 compared with
conrm the therapeutic selectivity of this strategy. colitis control mice given saline).
B. Moulari et al. / Biomaterials 29 (2008) 45544560 4559

healthy control SiNP Moreover, such particles were found to be present inside the
colitis control Me5ASA-SiNP 50mg inamed tissue for one week or longer permitting a slow drug
5ASA 30mg Me5ASA-SiNP 25mg release with a full therapeutic effect over time [7]. This is sug-
5ASA 100mg Me5ASA-SiNP 25mg rect gesting such a nanoscale system also for an application as a sus-
A 50 tained release formulation lowering the dosing frequency being
more convenient for the patient.

40
colon weight/length [mg/cm]

4. Conclusions

*
SiNP demonstrate possibilities in innovative drug delivery for
* the treatment of IBD through targeted delivery based on factors
30 * specic to the inammation sites. Me5ASASiNP allow selective
adhesion to inamed tissue creating a medication reservoir at that
* site with reduced drug availability at the healthy surrounding
20 * tissue. Additionally, these NP provide selective drug release
towards the inamed tissue and, thus, highly increasing thera-
peutic efciency of the associated amount of medication. This
carrier system successfully combines the advantages from a small
10 scale drug targeting system with those of a prodrug. As this drug
delivery system is transposable to most of the drugs in this thera-
peutic context, this strategy presents a promising alternative for
future innovative treatments of IBD.
0
Acknowledgments
healthy control SiNP
colitis control Me5ASA-SiNP 50mg Alf Lamprecht is grateful to the Institut Universitaire de France
5ASA 30mg Me5ASA-SiNP 25mg for nancial support. The project was co-nanced by the Region
5ASA 100mg Me5ASA-SiNP 25mg rect
B 40 of Franche-Comte (grant VJ000610). We thank V. Weissenborn,
L. Ismaeli and B. Refouvelet for their assistance.

References
30 [1] Head KA, Jurenka JS. Inammatory bowel disease part I: ulcerative colitis
MPO [U/mg tissue]

pathophysiology and conventional and alternative treatment options. Alt Med


2003;8:24783.
[2] Meissner Y, Lamprecht A. Alternative drug delivery approaches for the therapy
of inammatory bowel disease. J Pharm Sci 2008;97:287891.
[3] Friend DR. New oral delivery systems for treatment of inammatory bowel
20 * disease. Adv Drug Deliv Rev 2005;57:24765.
[4] Lamprecht A. Multiparticulate systems in the treatment of inammatory
bowel disease. Curr Drug Targets Inamm Allergy 2003;2:13744.
[5] Allison MC, Cornwall S, Poulter LW, Dhillon AP, Pounder RE. Macrophage
heterogeneity in normal colonic mucosa and in inammatory bowel disease.
* ** Gut 1988;29:15318.
10 * **
** [6] Seldenrijk CA, Drexhage HA, Meuwissen SGM. Dentritic cells and scavenger
* macrophage in chronic inammatory bowel disease. Gut 1989;30:48491.
* [7] Lamprecht A, Schafer U, Lehr CM. Size dependent targeting of micro- and
nanoparticulate carriers to the inamed colonic mucosa. Pharm Res
2001;18:78893.
[8] Nakase H, Okazaki K, Tabata Y, Uose S, Ohana M, Uchida K, et al. Development
0 of an oral drug delivery system targeting immune-regulating cells in experi-
mental inammatory bowel disease: a new therapeutic strategy. J Pharmacol
Fig. 6. Colon length/weight ratio (A) and myeloperoxidase (MPO) activity (B) on day 8 Exp Ther 2000;292:1521.
in TNBS colitis model after oral administration of 5ASA and Me5ASASiNP, respectively [9] Nakase H, Okazaki K, Tabata Y, Uose S, Ohana M, Uchida K, et al. An oral drug
(n 6; data are shown as mean  S.D.; *P < 0.05 compared with colitis control mice delivery system targeting immune-regulating cells ameliorates mucosal injury
given saline; **P < 0.05 compared with mice treated with 5ASA solution 30 mg/kg). in trinitrobenzene sulfonic acid-induced colitis. J Pharmacol Exp Ther
2001;297:11228.
[10] Lamprecht A, Ubrich N, Yamamoto H, Schafer U, Takeuchi H, Lehr CM, et al.
enzymatic particle degradation during the intestinal transit. This Biodegradable nanoparticles for the targeted drug delivery in the treatment of
inammatory bowel disease. J Pharmacol Exp Therap 2001;299:77581.
hypothesis was further supported by the observation that locally [11] Lamprecht A, Yamamoto H, Takeuchi H, Kawashima Y. Nanoparticles enhance
applied NP (circumventing the intestinal passage) showed an therapeutic efciency by selectively increased local drug dose in experimental
enormous therapeutic efciency which was superior to the colitis in rats. J Pharmacol Exp Therap 2005;315:196202.
[12] Meissner Y, Pellequer Y, Lamprecht A. Nanoparticles in inammatory bowel
locally applied drug in solution [11]. disease: particle targeting versus pH-sensitive delivery. Int J Pharm
The general advantage of Me5ASASiNP is a tighter bond 2006;316:13843.
between drug and the carrier system for efcient retention of the [13] Lamprecht A, Ubrich N, Yamamoto H, Schafer U, Takeuchi H, Lehr CM, et al.
Design of rolipram loaded nanoparticles: comparison of two preparation
drug, namely until arrival at sites of inamed colonic tissue.
methods. J Controlled Release 2001;71:297306.
Chemically binding the drug to the SiNP appears to be a key in this [14] Roy I, Ohulchanskyy TY, Bharali DJ, Pudavar HE, Mistretta RA, Kaur N, et al.
context. Me5ASASiNP for drug delivery in IBD combines thera- Optical tracking of organically modied silica nanoparticles as DNA carriers:
peutic approaches of i) passive drug targeting by NP and ii) the a nonviral, nanomedicine approach for gene delivery. Proc Natl Acad Sci
2005;102:27984.
triggered release from a prodrug retaining the entrapped drug until [15] Liu Y, Miyoshi H, Nakamura M. Novel drug delivery system of hollow
selective accumulation in the inamed tissue. mesoporous silica nanocapsules with thin shells: preparation and uorescein
4560 B. Moulari et al. / Biomaterials 29 (2008) 45544560

isothiocyanate (FITC) release kinetics. Colloids Surf B Biointerfaces 2007; [19] Hartmann G, Bidlingmaier C, Siegmund B, Albrich S, Schulze J, Tschoep K, et al.
58:1807. Specic type IV phosphodiesterase inhibitor rolipram mitigates experimental
[16] Li ZZ, Wen LX, Shao L, Chen JF. Fabrication of porous hollow silica nano- colitis in mice. J Pharmacol Exp Ther 2000;292:2230.
particles and their applications in drug release control. J Controlled Release [20] Pellequer Y, Meissner Y, Ubrich N, Lamprecht A. Epithelial heparin delivery via
2004;98:24554. microspheres mitigates experimental colitis in mice. J Pharmacol Exp Ther
[17] Kim S, Ohulchansky TY, Pudavar HE, Pandey RK, Prasad PN. Organically 2007;292:72633.
modied silica nanoparticles co-encapsulating photosensitizing drug and [21] Krawisz JE, Sharon P, Stenson WF. Quantitative assay for acute intestinal
aggregation-enhanced two-photon absorbing uorescent dye aggregates inammation based on myeloperoxidase activity. Gastroenterology 1984;87:
for two-photon photodynamic therapy. J Am Chem Soc 2007;129: 134450.
266975. [22] Klotz U. Colonic targeting of aminosalicylates for the treatment of ulcerative
[18] Neurath MF, Fuss I, Kelsall BL, Stuber E, Strober W. Antibodies to interleukin 12 colitis. Dig Liver Dis 2005;37:3818.
abrogate established experimental colitis in mice. J Exp Med 1995;182: [23] Klotz U, Schwab M. Topical delivery of therapeutic agents in the treatment of
128190. inammatory bowel disease. Adv Drug Deliv Rev 2005;57:26779.