P

C

MGa

b

c

d

a

ARRAA

KCSDS

1

iefRSMapttpce2

cf

0h

International Journal of Pharmaceutics 436 (2012) 778– 785

Contents lists available at SciVerse ScienceDirect

International Journal of Pharmaceutics

journa l h omepa g e: www.elsev ier .com/ locate / i jpharm

harmaceutical Nanotechnology

arboxylic modified spherical mesoporous silicas аs drug delivery carriers

argarita D. Popovaa,∗, Ágnes Szegedib, Iliyan N. Kolevc, Judit Mihályb, Borislav S. Tzankovd,eorgi Tz. Momekovd, Nikolai G. Lambovd, Krassimira P. Yonchevad

Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, Sofia, BulgariaResearch Centre for Natural Sciences, Hungarian Academy of Sciences, 1025 Budapest, Pusztaszeri út 59-67, HungaryInstitute of Molecular Biology, Bulgarian Academy of Sciences, Sofia, BulgariaFaculty of Pharmacy, Medical University, 2 Dunav Str., 1000 Sofia, Bulgaria

r t i c l e i n f o

rticle history:eceived 13 June 2012eceived in revised form 26 July 2012ccepted 27 July 2012vailable online 4 August 2012

ey words:arboxylic modificationpherical mesoporous silicarug release

a b s t r a c t

The present study deals with the development and functionalization of mesoporous silica nanoparti-cles as drug delivery platforms. Spherical MCM-41 and SBA-15 silicas with different pore sizes (2.7 nmand 5.5 nm, respectively) were post-synthesis modified applying a new, two step process. The initialstep was the modification with 3-amino-propyltriethoxysilane, and the next was the reaction withsuccinic anhydride in toluene in order to obtain carboxylic modified mesoporous carriers. The carboxylic-functionalized mesoporous materials were characterized by XRD, nitrogen physisorption, TEM, ATR FT-IRspectroscopy. The successful carboxylic functionalization was proved by the changes of the zeta poten-tial of the mesoporous materials before and after modification. The parent and the carboxylic-modifiedMCM-41 and SBA-15 materials showed high adsorption capacity (approximately 50 wt.%, except for non-

ulfadiazine functionalized MCM-41) for sulfadiazine that possesses amino functional groups. Mesoporous structurepeculiarities lead to different adsorption capacities on the carriers. In vitro release studies showed slowerrelease rate of sulfadiazine from carboxylic modified MCM-41 and SBA-15 mesoporous particles com-pared to the non modified ones. Both non loaded and drug-loaded silica materials demonstrated nocytotoxicity on Caco-2 cell line. The functionalized mesoporous systems are appropriate drug deliveryplatforms due to their biocompatibility and the possibility to modify drug release.

. Introduction

In recent years, there has been a constantly increasing interestn the application of mesoporous silica materials as drug carri-rs for controlled drug release in order to meet the requirementsor prolonged and better control of drug administration (Vallet-egí et al., 2001; Horcajada et al., 2006; Vallet-Regí et al., 2006;lowing et al., 2008; Wang, 2009; Izquierdo-Barba et al., 2010).esoporous nanosized silicas, characterized by high BET surface

reas, large pore volumes, uniform porosity, stable aqueous dis-ersion, excellent biocompatibility, in vivo biodegradability, andheir facile functionalization with different organic groups, makehem very attractive candidates for a wide range of biomedical pur-oses, such as controlled drug delivery, bone tissue regeneration,

ell tracking and immobilization of proteins or enzymes (Trewynt al., 2007; Manzano et al., 2008; Yang et al., 2008a,b; He et al.,010; Huang et al., 2010). They have been proposed for the first

∗ Corresponding author at: Institute of Organic Chemistry with Centre of Phyto-hemistry, Acad. G. Bonchev str., bl. 9, 1113 Sofia, Bulgaria. Tel.: +359 2979 39 61;ax: +359 28700 225.

E-mail address: mpopova@orgchm.bas.bg (M.D. Popova).

378-5173/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.ijpharm.2012.07.061

© 2012 Elsevier B.V. All rights reserved.

time as carriers for drug delivery in 2001 by Vallet-Regí et al. (2001).Mesoporous silica materials also fulfil the conditions for homoge-nous distribution of the drug through the matrix in contrast to theconventionally used polymeric materials. The appropriate chemicalmodification of the surface of the mesoporous matrix is essentialbecause a mesoporous silica surface covered with silanol groupsis not selective enough to adsorb drug molecules with differentfunctionalities (Anderson et al., 2004; Yang et al., 2008a,b; Caudaet al., 2010; Kato et al., 2010). Organophil surface modification canenhance the adsorption capacity of drug molecules, and it shouldalso allow modulating their release (Huh et al., 2003; Qu et al., 2006;Zeng et al., 2005; Balas et al., 2006; Katiyar et al., 2006; Munoz et al.,2003; Tang et al., 2006a; Szegedi et al., 2011). MCM-41 and SBA-15mesoporous silica materials, with uniform pore sizes up to 30 nm,are excellent matrices for encapsulation of drug molecules withdifferent sizes (Yang et al., 2001; Lei et al., 2004; Vinu et al., 2004;Qiao et al., 2006; Cho et al., 2008; Rosenholm et al., 2010; Lin et al.,2012).

The carboxylic-modified mesoporous materials are of great

interest in nanomedicine applications because of the possibility ofpH-response delivery for controlled release of drug molecules withfree amino functional groups (Tang et al., 2006b). Carboxylic mod-ification can be realized by one pot or by post synthesis methods.

rnal of

YrApe(iegSwcmcctTlocmcs

aItatpvoioptCsp

2

2

4S

2

Hwcwim

M.D. Popova et al. / International Jou

ang and co-workers (Yang et al., 2003) reported a co-condensationoute to directly synthesize carboxyl-modified SBA-15 materials.n intermediate example between the two methods is the tem-late ion-exchange by 11-triethoxysilanylundecanoicacid ethylster and its transformation to carboxylic groups by acidificationYang et al., 2005). In most of the cases post synthesis grapht-ng by organosylanes can be achieved in a one step procedure,.g. by amino-propyl triethoxysilane (APTES) for anchoring aminoroups (Horcajada et al., 2006; Shen et al., 2008; Wang, 2009;zegedi et al., 2011). However, modification of the silica surfaceith carboxylic groups is a multi-step procedure. Until now the

arboxylic functionalized mesoporous materials were obtained byodification with cyanoethylsilane followed by its hydrolysis to

arboxylic group by means of acidic media. This procedure can bearried out by post synthesis method or by direct synthesis, wherehe template is removed by strong acids (Yang et al., 2008a,b).he procedure includes a treatment with hazardous compounds,ike cyanosilanes, and it should also be noted that the efficiencyf the modification is low. The reaction of ring opening of suc-inic anhydride or maleic anhydride with APTES was applied forodification of SiO2 (An et al., 2007). The latter reaction was

arried out only for amorphous silica in N,N-dimethylformamideolvent.

In the present study, this approach was modified, namely bypplying a more appropriate solvent for the mesoporous system.n order to avoid the presence of water during the modifica-ion procedure and because of the good solubility of succinicnhydride in toluene, the latter was chosen as a solvent. In addi-ion, our previous investigations showed that the mesoporousarticle dispersion is better in toluene than that in the other sol-ents. Here we report a novel route for COOH functionalizationf MCM-41 and SBA-15 materials by means of the ring open-ng reaction with succinic anhydride in toluene and its effectn sulfadiazine loading and release. Sulfadiazine was chosen asrobe molecule (1) because of the available free amine func-ional group, still taking into account an eventual binding withOOH functional groups of the modified materials. Beside, the drugize was calculated (1.0 nm × 0.5 nm) and considered as appro-riate for encapsulation in the pores of our mesoporous carriers.

. Materials and methods

.1. Materials

C16TMABr and PEO43PPO15PEO43 (Pluronic F-38; MW700 g mol−1) were purchased by Aldrich and BASF, respectively.ulphadiazine (>99.5%) was provided from Aldrich.

.2. Synthesis of MCM-41 and SBA-15 spherical silica materials

Spherical MCM-41 was prepared according to the procedure ofuh et al. (2003). This sol–gel procedure is carried out at 353 Kithout co-solvent, only in water solution, and with NaOH as a

atalyst. The relative molar composition of the reaction mixtureas: 1.0 TEOS:0.12 C16TMABr:0.31 NaOH:1190 H2O, where TEOS

s tetraethyl orthosilicate and C16TMABr is hexadecyltrimethylam-onium bromide. The formed gel was aged at 353 K for 2 h, then

Pharmaceutics 436 (2012) 778– 785 779

washed with distilled water until neutral pH, and dried at ambient.Template removal of MCM-41 materials was carried out in air at823 K with 1 K/min rate for 5 h.

Spherical SBA-15 was synthesized according to the proce-dure of Katiyar et al. (2006). Pluronic P123 (triblockcopolymer(PEO20PPO70PEO20)), and hexadecyltrimethylammonium bromide(C16TMABr) were used as templates and tetraethyl orthosili-cate (TEOS) as silica source. The synthesis media also containedethanol besides the usual HCl solution. Synthesis mixturewith the following relative molar composition was applied:1.0 TEOS:0.012 PEO20PPO70PEO20: 0.036 C16TMABr:2 HCl:7.65EtOH:44.8·H2O. The mixture was stirred at 308 K for 45 minthan put in a Teflon lined autoclave and aged at 348 K for 10 hand subsequently at 393 K for 36 h. The product was washedchloride free with distilled water and dried at ambient. Tem-plate removal was carried out in air at 723 K with 1 K/min ratefor 5 h.

2.3. Modification of the materials by APTES

Modification of the spherical MCM-41 and SBA-15 withamino groups was accomplished by reaction with 3-amino-propyltriethoxysilane (APTES) in ethanol (5 h, 50 ◦C) and inanhydrous toluene (24 h, 60 ◦C), respectively. Samples werewashed with several portions of solvent, and finally with waterand dried at room temperature. The reaction of amino modified sil-icas with succinic anhydride was carried out in toluene. To removeadsorbed water azeotropic drying of amino modified silicas wasmade at 115 ◦C by mixing 1 g of silica with 20 ml of anhydroustoluene. 6.6 mmol of succinic anhydride (assuming 2 wt.% of aminocontent on silica) was added to the mixture at 60 ◦C and treated for24 h. The samples were dried by vacuum evaporation (0.04 Pa) atroom temperature for 6 h.

The parent mesoporous spherical silica materials were desig-nated as MCM-41 and SBA-15 and the carboxylic modified ones asMCM-41COOH and SBA-15COOH, respectively.

2.4. Characterization

X-ray diffractograms were recorded by a Philips PW 1810/3710diffractometer with Bragg–Brentano parafocusing geometry apply-ing monochromatized CuK� (� = 0.15418 nm) radiation (40 kV,35 mA) and a proportional counter.

Nitrogen physisorption measurements were carried out at 77 Kusing Quantachrome NOVA Automated Gas Sorption Instrument.The pore-size distributions were calculated from the desorptionbranch of the isotherms with the BJH method. Samples were pre-treated at 353 K for 5 h before measurements.

TEM images were taken using a MORGAGNI 268D TEM(100 kV; W filament; point-resolution = 0.5 nm).

The thermogravimetric measurements were performed by aSetaram TG92 instrument with a heating rate of 5 K/min in nitrogenflow.

Attenuated total reflection infrared (ATR-FTIR) spectra wererecorded by means of a Varian Scimitar 2000 FTIR spectrome-ter equipped with a MCT (mercury–cadmium–tellur) detector anda single reflection ATR unit (SPECAC “Golden Gate”) with dia-mond ATR element. In general, 128 scans and 4 cm−1 resolutionwas applied. For all spectra ATR-correction was performed (VarianResPro 4.0 software).

Determination of nanoparticle size, polydispersity index and

zeta potential was performed using a Zetasizer (Zetasizer Nano ZS,Malvern Instruments, Worcestershire, UK). The samples were dis-persed in distilled water and measured at scattering angle of 90◦

at 25 ◦C.

780 M.D. Popova et al. / International Journal of Pharmaceutics 436 (2012) 778– 785

ylic fu

2

tsaov

w(ddtwlr

2

lpfiwCmwvTMdwaocetAbPtw(mvcw

2

db

for adsorbed water in the channels. The modification of MCM-41and SBA-15 by APTES resulted in the appearance of bands at 1597,1547 and 1634 cm−1. The first one can be associated with primaryamino groups, whereas the two others are with the protonated form

Table 1Physicochemical properties of the parent, amino and the carboxylic modified spher-ical MCM-41 and SBA-15 samples.

Samples a0a (nm) BET (m2/g) PDb (nm) Zeta-potential (mV)

MCM-41 4.4 1175 2.7 −12.4 ± 0.4MCM-41NH2 4.2 570 2.2 –MCM-41COOH 3.7 412 1.9 −22.7 ± 0.5SBA-15 10.2 878 5.5 −24.9 ± 0.8

Scheme 1. Process of preparation of the carbox

.5. Sulfadiazine loading and in vitro release measurements

Sulfadiazine (SD) loading was carried out in a mixture of ace-one and methanol (1:1) at 37 ◦C applying 1 g of SD per 1 gupport. After incubation for 24 h, the mixtures were centrifugedt 15,000 rpm, rinsed with distilled water, separated by a sec-nd centrifugation, and finely dried at room temperature underacuum.

For in vitro release studies, 10 mg of the drug-loaded particlesere incubated in 200 ml distilled water at 37 ◦C under stirring

100 rpm). At appropriate time intervals, 5 ml samples were with-rawn, centrifuged at 15,000 rpm for 15 min, and replaced by freshistilled water. The concentration of the released sulfadiazine inhe samples was determined by UV–vis spectrophotometry at aavelength of 262 nm. The concentration of sulfadiazine was calcu-

ated according to the standard curve prepared in the concentrationange of 1.9–18.8 �g/ml (r = 0.996).

.6. In vitro cytotoxicity assay

Stock solutions of sulfadiazine or dispersions of empty andoaded carriers were freshly prepared in DMSO and thereafterromptly diluted with fresh growth medium, to yield the desirednal concentrations. Caco-2 cells (human colon adenocarcinoma)ere supplied from German Collection of Microorganism and Cellultures (DSMZ, Germany). They were maintained as adherentonolayer cultures in 80% MEM (with Earle’s salts) supplementedith 20% fetal bovine serum and non-essential amino acids. The cell

iability was assessed using the standard MTT-dye reduction assay.he method is based on the reduction of the yellow tetrazolium saltTT to a violet formazan product via the mitochondrial succinate

ehydrogenase in viable cells. In brief, exponentially growing cellsere seeded in 96-well flat-bottomed microplates (100 �l/well) at

density of 1 × 105 cells/ml and allowed to attach to the surfacef the well for 24 h, at 37 ◦C. Thereafter the medium was dis-arded and cells were exposed to different dilutions of the testedmpty and loaded silica carriers, or to sulfadiazine. At the final dilu-ions obtained the concentration of DMSO never exceeded 0.5%.t least 8 wells were used for each concentration. After the incu-ation with the test compounds 10 �l MTT solution (10 mg/ml inBS) aliquots were added to each well. The microplates were fur-her incubated for 4 h at 37 ◦C and the formazan crystals formedere dissolved through addition of 100 �l/well 5% formic acid

in 2-propanol). The absorption was measured using a multimodeicroplate reader (Beckman Coulter DTX-880) at 580 nm. Cell sur-

ival fractions were calculated as percentage of the solvent-treatedontrol (0.5% DMSO in growth medium) (GraphPad Prizm Soft-are).

.7. Statistical analysis

All data were arranged as mean ± standard deviation. Significantifferences were determined by t-test. P < 0.05 was considered toe significant.

nctionalized mesoporous silica nanoparticles.

3. Results and discussion

The present study develops a new method for post-syntheticcarboxylic functionalization of spherical MCM-41 and SBA-15silicas. Thus, the initial step was the modification with 3-amino-propyltriethoxysilane, and the next was the reaction with succinicanhydride in toluene in order to obtain carboxylic modified meso-porous carriers (Scheme 1).

3.1. Materials characterization

XRD patterns of the carboxylic modified MCM-41 and SBA-15samples at low 2 theta degree region show the preservation ofthe hexagonal structure (Fig. 1). Compared to the parent materialssome broadening, intensity decrease, and structure deteriorationcan be observed, especially for the modified MCM-41 samples. XRDpatterns (not shown) at higher angles show the presence of smallamount of separate sulfadiazine phase. These observations weresupported also by the N2 adsorption data (Fig. 2).

The nitrogen physisorption isotherms of amino and carboxylicmodified MCM-41 materials (Fig. 2) show the typical capillary con-densation step of nitrogen between 0.2 and 0.3 relative pressures,but the adsorbed volume is much lower than in the parent MCM-41. The N2 adsorption isotherms of SBA-15COOH exhibit a H2 typehysteresis loop, also with lower adsorbed volume than the parentone. Textural parameters are summarized in Table 1. The modifica-tion by amino and carboxylic groups did not influence the sphericalmorphology of MCM-41 and SBA-15 materials as evidenced byTEM investigations (Fig. 3). The two step modification procedureresulted in a significant specific surface area decrease for both typesof materials. Also, narrower pore sizes were detected. This effectwas more pronounced for MCM-41COOH because of the smallerpore size of the parent material.

Formation of carboxylic groups was evidenced by ATR FT-IRmeasurements. Fig. 4 shows the FT-IR spectra of the parent, theamino and the carboxylic modified silica materials. The parent sil-ica materials show only a band at around 1630 cm−1 characteristic

SBA-15NH2 10.1 477 5.1 –SBA-15COOH 8.2 300 5.1 −29.1 ± 2.1

a Unitcell parameter (a0 = 2d1 0 0(3)−1/2).b Pore diameter calculated by the BJH method.

M.D. Popova et al. / International Journal of Pharmaceutics 436 (2012) 778– 785 781

oxyli

ossi1Ytpiattcwc

Fig. 1. XRD patterns of the parent, amino and carb

f amino groups. The appearance of these bands is the evidence ofuccessful modification of silica materials by amino groups. Theymmetric and asymmetric stretching vibration bands character-stic for carboxylate ( COO−) groups can be found at 1555 and411 cm−1 in both samples, respectively (Horcajada et al., 2006;ang et al., 2008a,b). The band at 1626 cm−1 is registered only inhe spectrum of MCM-41COOH and it can be associated with theresence of amid groups ( NH C O) strongly hydrogen bonded

n the network. In the spectrum of the MCM-41COOH sample twodditional bands appear. The band at 1697 cm−1 can be attributedo the strongly hydrogen bonded COOH groups. This is an indica-

ion of the fact that the number of carboxylic groups is higher, theyan be found in close vicinity and can interact with each other orith the neighboring silanol groups. The shoulder at 1717 cm−1 is

haracteristic of non dissociated, “free” carbonyl groups. The weak

Fig. 2. Nitrogen adsorption/desorption isotherms of the parent, amino

c modified MCM-41 (A) and SBA-15 (B) materials.

band at 1770 cm−1 can be associated with the unreacted succinicanhydride in the sample. The CH2 deformation vibration of hydro-carbon chains can also be detected at the 1350–1450 cm−1 region.

The amount of grafted organic groups was determined by TGanalysis. Our result show that the amount of COOH groups cor-responds to the total amount of amino groups formed during themodification by APTES. Taking into account the weight loss on TGcurve (not shown) between 200 and 500 ◦C we found that theamount of amino groups was 2.3 wt.% and 3.1 wt.% in MCM-41and in SBA-15, respectively. Considering the theoretical increase ofmolecular weight due to carboxylic modification the correspond-

ing values were 2.3 and 3.0 wt.%. In our previous work based onour FTIR results we have found that almost all the surface silanolgroups reacted with APTES (Szegedi et al., 2012), therefore the for-mation of the same amount of carboxylic groups points toward

and carboxylic modified MCM-41 (A) and SBA-15 (B) materials.

782 M.D. Popova et al. / International Journal of Pharmaceutics 436 (2012) 778– 785

ical M

tTtgato

a(c

Fig. 3. TEM images of carboxylic modified spher

hat all the amino groups were reacted in the modification process.he presence of amino groups was also approved and quantita-ively determined by the reaction of ninhydrin with primary aminoroups as suggested in our previous paper. The ninhydrin reactionpplied for the carboxylic modified samples did not show forma-ion of Ruhemann’s purple which is due to the total transformationf amino groups to carboxylic ones.

The zeta-potential measurements of both types of non-modified

nd carboxylic modified materials demonstrate negative valuesTable 1). As shown, the zeta potential is more negative for thearboxylic modified silicas than the values for the parent ones.

Fig. 4. ATR FT-IR spectra of the parent, amino and carb

CM-41 (A and B) and SBA-15 (C and D) samples.

This observation can be also considered as an indication for theefficient carboxylic functionalization of the silica carriers. Particlesize determination results of sulfadiazine loaded samples showedthat the MCM-41 materials are less agglomerated upon carboxylicmodification and drug loading procedure than the SBA-15 ones(Table 2). Moreover, the polydispersity index did not increasesignificantly upon modification of MCM-41 material, the narrowparticle size distribution observed on the parent MCM-41 material

was preserved. The average particle size of SBA-15 materials signif-icantly increased after modification, indicating a higher tendencyfor agglomeration. Polydispersity measurements showed that both

oxylic modified MCM-41 and SBA-15 materials.

M.D. Popova et al. / International Journal of Pharmaceutics 436 (2012) 778– 785 783

Table 2Physico-chemical properties of the sulfadiazine loaded parent and carboxylic modified MCM-41 and SBA-15 materials.

Samples Size (nm) Polydispersity index Zeta-potential (mV) Drug loadinga (%)

MCM-41-SD 480.5 0.43 −10.4 ± 2.0 39.05MCM-41COOH-SD 603.9 0.55 +3.5 ± 0.1 49.78SBA-15-SD 938.7 0.97 −20.3 ± 2.5 53.60SBA-15COOH-SD 1740.3 0.95 +1.8 ± 0.7 52.40

a Calculated from TG analysis, n = 3.

and ca

Sdtd

3

twomMaitmsStomt1m1cidic

i(

spectrum like free sulfadiazine. Typical bands of sulfadiazine at1652 (� NH2), 1577 (�CC of aromatic ring), 1490 (�CCH of aro-matic ring), and at 1443 and 1411 cm−1 (�CCH and �HCN ofpyrimidine ring, respectively) appear (Ogruc-Ildiz et al., 2009),

Fig. 5. TG curves of the sulfadiazine loaded parent

BA-15 materials included size fractions from 0.9 to 5 �m. The poly-ispersity index for parent SBA-15 is about two times higher thanhat of for MCM-41 and the modification resulted only in a slightecrease of it.

.2. Adsorption of sulfadiazine

Successful modification was also evidenced by our drug adsorp-ion experiments (Table 2). The amount of adsorbed sulfadiazineas measured by TG analysis (Fig. 5) corrected by the amount

f adsorbed water and carboxylic groups. It was found that non-odified MCM-41 can adsorb 39.05 wt.% of sulfadiazine (Table 2).odification of MCM-41 with carboxylic groups resulted in higher

mount of adsorbed sulfadiazine (49.78 wt.%). In contrast, mod-fication by carboxylic groups did not have a positive effect onhe adsorption capacity of sulfadiazine in the case of SBA-15

aterial. The initial and the carboxylic modified SBA-15 pos-ess similar adsorption capacity for sulfadiazine (53.63 wt.% forBA-15 and 52.40 wt.% for SBA-15COOH). Analyzing the N2 adsorp-ion isotherms by the �s plot method, the microporous volumef the parent SBA-15 sample was 0.05 cm3/g and practically noicropores were detected for functionalized SBA-15COOH. Hence,

he expected higher loading capacity of the functionalized SBA-5COOH was not registered because of the total blocking of theicropores. The close adsorption capacity of both samples (SBA-

5 and SBA-15COOH) is due to the presence of micropores in thease of non-functionalized SBA-15 and the stronger adsorption sitesn the case of the functionalized SBA-15COOH. Such effect wasescribed in our previous paper, where the adsorption capacity for

buprofen on the amino modified SBA-15 was found to be lower in

omparison to that of the parent one (Szegedi et al., 2011).

The adsorption capacity of SBA-15COOH for sulfadiazines higher (52.4 wt.%) compared to MCM-41COOH (49.78 wt.%)Table 2), but not as much as could be expected owing to the

rboxylic modified MCM-41 and SBA-15 materials.

bigger pore size of the SBA-15 carrier. The reason for the highadsorption in the case of MCM-41COOH is the electrostatic bind-ing between sulfadiazine and COOH groups on the outer surfaceof MCM-41COOH sample and additional binding in the secondarymesopores between the silica particles. The textural mesoporosityis formed only for MCM-41 silica.

The presence of electrostatically bonded sulfadiazine was alsoevidenced by the formation of COO− NH3

+ groups in the ATR FT-IR spectra of sulfadiazine loaded samples at 1545 cm−1 (Fig. 6).Non-modified, sulfadiazine loaded MCM-41 silica shows similar

Fig. 6. ATR FT-IR spectra of the sulfadiazine loaded MCM-41 and MCM-41COOHsamples.

784 M.D. Popova et al. / International Journal of Pharmaceutics 436 (2012) 778– 785

Fs

itcapbtdaobato2

omtM

3

pFm4ThprCspatmims

Acknowledgement

ig. 7. In vitro release of sulfadiazine from the parent and the carboxylic modifiedpherical MCM-41 (A) and spherical SBA-15 (B) in distilled water (mean ± SD, n = 3).

ndicating that sulfadiazine can be found physically adsorbed inhe channels or on the outer surface of MCM-41 material. Bandsharacteristic for sulfadiazine and surface carboxylic modificationppeared also in the spectrum of MCM-41COOH/SD sample. Theresence of electrostatically bonded sulfadiazine was evidencedy the formation of COO− NH3

+ groups at 1550 cm−1. Parallel,he intensive stretching vibration attributed to carbonyl in nonissociated carboxylic groups of MCM-41COOH at 1717 cm−1 islmost disappeared upon SD adsorption. Similar phenomenon wasbserved by Tang et al. (2006b) adsorbing famotidine on car-oxylic modified MCM-41. The intensity of 1577, 1490 cm−1 bands,ssigned to aromatic ring vibrations, are also decreased comparedo that of free SD, indicating a weak interaction of aromatic ringf SD drug with the narrow channels of MCM-41 (Braschi et al.,010).

The electrostatic binding of sulfadiazine with carboxylic groupsf modified carriers was also proved by zeta-potential measure-ents. Significant increase of zeta-potential was observed for both

ypes of drug-loaded modified materials, +3.5 and +1.8 mV forCM-41COOH and SBA-15COOH, respectively (Table 2).

.3. In vitro release

The results of the sulfadiazine release from the parent meso-orous silicas and their carboxylic-modifications are plotted inig. 7. As shown, sulfadiazine was rapidly released from non-odified silica carriers (MCM-41 and SBA-15). In particular, in

h approximately the whole amount of sulfadiazine was released.his fact could be explained by taking into account that theydrogen bonding between sulfadiazine and the parent meso-orous silica materials is relatively weak. In contrast, slower releaseate was observed from carboxylic-modified samples (MCM-41-OOH and SBA-15-COOH). The formed COO− NH3

+ bond betweenulfadiazine and the carboxylic groups of the modified meso-orous samples is stronger than that of between sulfadiazinend silanol groups of the parent silicas. This effect can explainhe slower release rates of sulfadiazine from both carboxylic-

odified MCM-41 and SBA-15 samples in comparison to the

nitial materials. Thus, the drug release process was governed

ainly by the functionality of the carrier rather than the poreize.

Fig. 8. In vitro cytotoxicity of MCM-41, SBA-15 and their carboxylic modifiedanalogs (MCM-41COOH and SBA-15COOH) (A) and the sulfadiazine-loaded materi-als (B) on Caco-2 cell line (mean ± SD, n = 3).

3.4. Cytotoxicity of the carriers and the drug loaded silicaparticles

The cytotoxic activity of the non-modified and the COOH-functionalized carriers was evaluated in Caco-2 cell line after72 h continuous exposure using the MTT dye-reduction assay.The results are shown in Fig. 8. It is clearly evident that bothplacebo carrier suspensions are practically devoid of cytotoxic andantiproliferative effects in the model cellular population. Further,a comparative assessment of the cytotoxic effects of free vs. sul-fadiazine loaded samples was performed. As evident from the cellsurvival data (Fig. 8), the free sulfadiazine shows low cytotoxicityand the applied mesoporous carriers do not increase the cytotoxiceffect of the loaded samples.

4. Conclusions

We have demonstrated a novel route to synthesize MCM-41and SBA-15 mesoporous silica materials functionalized with car-boxylic groups. The reaction of ring opening of amino-modifiedmesoporous silicas with succinic anhydride in toluene was devel-oped. The method allowed applying mild conditions combinedwith less toxic reagents. Surface modification with carboxylicgroups resulted in a higher degree of sulfadiazine loading forthe carboxylic modified MCM-41COOH (∼50 wt.% versus 39% forMCM-41), whereas no difference was observed for SBA-15 andSBA-15COOH. In vitro release studies showed that the functional-ization of mesoporous MCM-41 and SBA-15 silicas with carboxylicgroups make them appropriate carriers for prolonged release ofsulfadiazine. The new method for carboxylic modification of themesoporous carriers did not increase cytotoxicity on Caco-2 cellline suggesting that the obtained modified carriers were biocom-patible.

Financial support by the Bulgarian-Hungarian Inter-AcademicExchange Agreement is greatly acknowledged.

rnal of

R

A

A

B

B

C

C

H

H

H

H

I

K

K

L

L

M

M

O

M.D. Popova et al. / International Jou

eferences

n, Y., Chen, M., Xue, Q., Liu, W., 2007. Preparation and self-assembly of carboxylicacid-functionalized silica. J. Colloid Interface Sci. 311, 507–513.

nderson, J., Rozenholm, J., Areva, S., Linden, M., 2004. Influences of material char-acteristics on ibuprofen drug loading and release profiles from ordered micro-and mesoporous silica matrices. Chem. Mater. 16, 4160–4167.

alas, F., Manzano, M., Horcajada, P., Vallet-Regi, M., 2006. Confinement andcontrolled release of bisphosphonates on ordered mesoporous silica-basedmaterials. J. Am. Chem. Soc. 128, 8116–8117.

raschi, I., Gatti, G., Paul, G., Gessa, C.E., Cossi, M., Marchese, L., 2010. Sulfon-amide antibiotics embedded in high silica zeolite: a combined experimentaland theoretical study of host–guest and guest–guest interactions. Langmuir 26,9524–9532.

auda, V., Schlossbauer, A., Bein, Th., 2010. Bio-degradation study of colloidalmesoporous silica nanoparticles: effect of surface functionalization withorgano-silanes and poly(ethylene glycol). Micropor. Mesopor. Mater. 132,60–71.

ho, Y., Shi, R., Borgens, R.B., Ivanisevic, A., 2008. Functionalized mesoporous silicananoparticle-based drug delivery system to rescue acrolein-mediated cell death.Nanomedicine 3, 507–519.

e, Q., Shi, J., Chen, F., Zhu, M., Zhang, L., 2010. An anticancer drug delivery systembased on surfactant-templated mesoporous silica nanoparticles. Biomaterials31, 3335–3346.

orcajada, P., Ramila, A., Ferey, G., Vallet-Regí, M., 2006. Influence of superficialorganic modification of MCM-41 matrices on drug delivery rate. Solid State Sci.8, 1243–1249.

uang, X., Teng, X., Chen, D., Tang, F., He, J., 2010. Engineered aprotinin for improvedstability of fibrin biomaterials. Biomaterials 32, 430–438.

uh, S., Wiench, J.W., Yoo, J.-Ch., Pruski, M., Lin, V.S.-Y., 2003. Organic function-alization and morphology control of mesoporous silicas via a co-condensationsynthesis method. Chem. Mater. 15, 4247–4256.

zquierdo-Barba, I., Colilla, M., Manzano, M., Vallet-Regi, M., 2010. In vitro stabil-ity of SBA-15 under physiological conditions. Micropor. Mesopor. Mater. 132,442–452.

atiyar, A., Yadav, S., Smirniotis, P.G., Pinto, N.G., 2006. Synthesis of ordered largepore SBA-15 spherical particles for adsorption of biomolecules. J. Chromatogr.A 1122, 13–20.

ato, N., Ishii, T., Koumoto, Sh, 2010. Synthesis of monodisperse mesoporous sil-ica hollow microcapsules and their release of loaded materials. Langmuir 26,14334–14344.

ei, J., Fan, J., Yu, C.Z., Zhang, L.Y., Jiang, S.Y., Tu, B., Zhao, D.Y., 2004. Immobilizationof enzymes in mesoporous materials: controlling the entrance to nanospace.Microp. Mesopor. Mater. 73, 121–128.

in, Ch.H., Cheng Sh, H., Liao, W.-N., Wei, P.-R., Sung, P.-J., Weng, Ch.-F., Lee, Ch.-H.,2012. Mesoporous silica nanoparticles for the improved anticancer efficacy ofcis-platin. Int. J. Pharm. 429, 138–147.

anzano, M., Aina, V., Arean, C.O., Balas, F., Cauda, V., Colilla, M., Delgado, M.R.,Vallet-Regí, M., 2008. Studies on MCM-41 mesoporous silica for drug deliveryeffect of particle morphology and amine functionalization. Chem. Eng. J. 137,

30–37.

unoz, B., Ramila, A., Pariente, J.P., Diaz, I., Vallet-Regí, M., 2003. MCM-41 organicmodification as drug delivery rate regulator. Chem. Mater. 15, 500–503.

gruc-Ildiz, G., Akyuz, S., Ozle, A.E., 2009. Experimental, ab initio and density func-tional theory studies on sulfadiazine. J. Mol. Struct. 924–926, 514–522.

Pharmaceutics 436 (2012) 778– 785 785

Qiao, S.Z., Djojoputro, H., Hu, Q.H., Lu, G.Q., 2006. Synthesis and lysozyme adsorptionof rod-like large-pore periodic mesoporous organosilica. Solid State Chem. 34,249–256.

Qu, F.Y., Zhu, G.S., Huang, S.Y., Li, S.G., Qiu, S.L., 2006. Effective controlled release ofcaptopril by silylation of mesoporous MCM-41. Chem. Phys. Chem. 7, 400–406.

Rosenholm, J.M., Sahlgren, C., Lindén, M., 2010. Towards multifunctional, targeteddrug delivery systems using mesoporous silica nanoparticles – opportunitiesand challenges. Nanoscale 2, 1870–1883.

Shen, Sh., Chow, P.Sh., Kim, S., Zhu, K., Tan, R.B.H., 2008. Small-angle X-ray scatteringof silica dispersions at low particle concentrations. J. Colloid Interface Sci. 331,365–371.

Slowing, I.I., Viverro-Escoto, J.L., Wu Ch, W., Lin, V.S.-Y., 2008. Mesoporous silicananoparticles as controlled release drugdelivery and gene transfection carriers.Adv. Drug Deliv. Rev. 60, 1278–1288.

Szegedi, A., Popova, M., Goshev, I., Mihály, J., 2011. Effect of amine functionalizationof spherical MCM-41 and SBA-15 on controlled drug release. J. Solid State Chem.184, 1201–1207.

Szegedi, A., Popova, M., Goshev, I., Klébert, S., Mihály, J., 2012. Controlleddrug release on amine functionalized spherical MCM-41. J. Sol. State Chem.,http://dx.doi.org/10.1016/j.jssc.2012.05.030.

Tang, Q.L., Xu, Y., Wu, D., Sun, Y.H.,.Wang, J.Q., Xu, J., Deng, F., 2006a. Studies on anew carrier of trimethylsilyl-modified mesoporous material for controlled drugdelivery. J. Control. Release 114, 41–46.

Tang, Q.L., Xu, Y., Wu, D., Sun, Y.H., 2006b. A study of carboxylic-modified meso-porous silica in controlled delivery for drug famotidine. J. Solid State Chem. 179,1513–1520.

Trewyn, B.G., Slowing, I.I., Giri, S., Chen, H.-T., Lin, V.S.-Y., 2007. Synthesis and func-tionalization of a mesoporous silica nanoparticle based on the sol–gel processand applications in controlled release. Acc. Chem. Res. 40, 846–853.

Vallet-Regí, M., Ramila, A., del Real, R.P., Perez-Pariente, J., 2001. A new property ofMCM-41: drug delivery system. Chem. Mater. 13, 308–311.

Vallet-Regí, M.A., Ruiz-Gonzalez, L., Gonzalez-Calbet, I.J.M., 2006. Revisiting silicabased ordered mesoporous materials: medical applications. J. Mater. Chem. 16,26–31.

Vinu, A., Murugesan, V., Hartmann, M., 2004. Adsorption of lysozyme over meso-porous molecular sieves MCM-41 and SBA-15: influence of pH and aluminumincorporation. J. Phys. Chem. B 108, 7323–7330.

Wang, Sh, 2009. Ordered mesoporous materials for drug delivery. Micropor. Meso-por. Mater. 117, 1–9.

Yang, C.M., Sheu, H.S., Chao, K.J., 2001. Templated synthesis and structural studyof densely packed metal nanostructures in MCM-41 and MCM-48. Adv. Funct.Mater. 12, 143–148.

Yang, C.M., Zibrowius, B., Schüth, F., 2003. A novel synthetic route for negativelycharged ordered mesoporous silica SBA-15. Chem. Commun., 1772–1773.

Yang, Q., Wang, S.C., Fan, P.W., Wang, L.F., Di, Y., Lin, K.F., Xiao, F.S., 2005. pH-responsive carrier system based on carboxylic acid modified mesoporous silicaand polyelectrolyte for drug delivery. Chem. Mater. 17, 5999–6003.

Yang, J., Lee, J., Kang, J., Lee, K., Suh, J.-S., Yoon, H.-G., Huh, Y.-M., Haam, S., 2008a. Hol-low silica nanocontainers as drug delivery vehicles. Langmuir 24, 3417–3421.

Yang, P., Quan, Z., Lu, L., Huang Sh Lin, J., 2008b. Luminescence functionalization of

mesoporous silica with different morphologies and applications as drug deliverysystems. Biomaterials 29, 692–702.

Zeng, W., Qian, X.F., Zhang, Y.B., Yin, J., Zhu, Z.K., 2005. Organic modified mesoporousMCM-41 through solvothermal process as drug delivery system. Mater. Res. Bull.40, 766–772.