synthesis of mesoporous silica nanoparticle-encapsulated alginate microparticles for sustained...

Synthesis of mesoporous silica nanoparticle-encapsulated alginatemicroparticles for sustained release and targeting therapy

Yu-Te Liao, Kevin C.-W. Wu, Jiashing Yu

Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan

Received 24 January 2013; revised 24 June 2013; accepted 8 July 2013

Published online 30 August 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33007

Abstract: This study reports the synthesis of mesoporous

silica nanoparticle-encapsulated alginate microparticles

(MSN@Alg) for sustained release and targeting therapy. The

MSN@Alg was synthesized by air dynamical atomization,

and the effects of several critical factors including concentra-

tion of alginate solution, flow rate of alginate solution, flow

rate of air, the distance between nozzle and calcium bath,

and stirring rate of calcium on the particle size of the synthe-

sized MSN@Alg were investigated. For studying the sus-

tained release properties of the MSN@Alg, rhodamine 6G

(R6G) was used as a model drug, and we compared the

release properties of R6G/MSN and R6G/MSN@Alg using dif-

ferent concentrations of alginate, concentrations and volumes

of phosphate-buffered saline (PBS) buffer solutions. The sus-

tained release behavior of the R6G/MSN@Alg system can be

prolonged to 20 days with an optimal condition of 1 mg R6G/

MSN@Alg to 2 mL PBS (10 mM). To achieve targeting ther-

apy, an anticancer drug, doxorubicin (Dox), was loaded into

MSN@Alg, and a arginine, glycine, and aspartic acid (RGD)-

based peptide was functionalized onto the surface of

MSN@Alg for the purpose of specific targeting. The results

showed that the intracellular drug delivery efficiency was

greatly enhanced (i.e., 3.5-folds) for the Dox/MSN@Alg-RGD

drug delivery system. VC 2013 Wiley Periodicals, Inc. J Biomed

Mater Res Part B: Appl Biomater, 102B: 293–302, 2014.

Key Words: alginate, mesoporous silica nanoparticle, sus-

tained release, cell-specific targeting

How to cite this article: Y. T. Liao, K. C.-W. Wu, J. Yu. 2014. Synthesis of mesoporous silica nanoparticle-encapsulated alginatemicroparticles for sustained release and targeting therapy. J Biomed Mater Res Part B 2014:102B: 293–302.

INTRODUCTION

Due to the defense of immune system and metabolic sys-tem, the circulating time of drug decreases apparently. Drugnot only has effect on pathological cells but also has sideeffect on normal cells. To enhance the efficiency of drug andprotect healthy cells, distribution of a good drug deliverysystem is important and drug carrier is a good method toenhance the efficiency of drug. Good drug carriers normallyhave the features as followed1,2: (a) they are synthesizedwith regular shape and uniform particle size; (b) theyshould possess of more stable structure to prevent prema-ture release of entrapped drug molecules before reaching tospecific sites; (c) these drug carriers are made of materialswith high biocompatibility and low antigenic properties;and (d) the vehicles must have the high loading capacity/encapsulation of desired drug molecules. Moreover, the abil-ities of sustained release and specific targeting are alsoimportant for drug carriers. Many researchers have reportedthat once the drugs loaded into carrier, the cytotoxicitycould be reduced apparently. The carrier can penetrate intocell via different endocytosis, the drugs inside carrier couldpass into cell easily and the efficiency could also increase.The ability of sustained release enables the drug releasefrom carrier slowly and keeps the concentration of drug

higher than effective concentration for a long period, whichmeans that the circulating time of drug can be improvedeffectively. The ability of specific targeting can limit the dis-tribution of carriers to pathological cells and minimize theside effects to healthy cells. Ability of specific targeting canalso be used for intracellular delivery of protein-baseddrugs or macromolecular drugs.

In 1950s, Jatzkewitz3 used polyvinylpyrrolidone, mesca-line and peptide to form the first drug carrier via coacerva-tion. After that, polyethylene glycol4- and dextran5-basedpolymer drug carriers have gotten lots of attention.Polymer-based drug carrier had many types, such as micro-particle,6 colloid,7 and micelle.8 Every type had its ownadvantages and disadvantages. The user can choose anytype depending on the purpose of carrier. For example,micelle can enhance the loading capacity of hydrophobicdrug. Micelle encapsulated hydrophobic drugs in oil phaseand delivered them into water phase. The process of encap-sulating drug for micelle was usually in organic phase,which may destroy the drug. The encapsulated drugs wouldbe limited by the oil phase. Colloid had high diffusibilityand mobility.6 It also had changeable hydrodynamic radius,which can reduce the chance of clogging in blood vessel;however, after drying, colloid would have chance to lose the

Correspondence to: J. Yu (e-mail: [email protected])

VC 2013 WILEY PERIODICALS, INC. 293

ability of dispersion and may aggregate together. Micropar-ticles had high surface area and unchanged size and struc-ture to prevent being extruded from surrounding.6 Kimet al.9 used chitosan modified with cholanic acid and a dyecalled Cy5.5 to encapsulate cisplatin. The author claimedthat this carrier had high loading capacity of hydrophobicdrug, suitable size and ability of targeting. Yun et al.10 usedhyaluronic acid modified with monoclonal as carrier forgene delivery. The author mentioned that their processreduced the use of organic solvent and their carrier hadability of controlled release and site-specific targeting. Nanj-wade et al.11 synthesized alginate as nanoparticle to encap-sulate carboplatin. This carrier had long term releasebehavior and high loading capacity and accumulated on spe-cific organs, therefore enhanced the efficiency ofcarboplatin.

Alginate and silica are two well-known materials thathave been used as biomaterials for many years. Alginate isa kind of polysaccharide existing in brown algae. Generally,alginate can chelate with most of divalent cations to formrigid structure.12 The structure of alginate could be in manydifferent types; such as gel,13 particle.14 fiber,15 bulk,16 andfilm17 depended on the function of alginate. There are threecommon methods to synthesize alginate carriers; coacerva-tion,18 layer by layer,19 and atomization.20 Some advantagesof alginate are adequate of source, relatively stable environ-ment, operation under room temperature without organicsolvent, tunable size and structure, low immunity, biocom-patible, and adequate of carboxylic group for functionaliza-tion. Silica is rich in earth. Kato and coworkers21

synthesized first silica structure with regular pores namedFSM in 1990. Mobile corporation22 also synthesized silicastructure with regular pores named M41S in 1992. Afterthat time, porous silica structure has been researched in dif-ferent fields. The morphology of silica structure can be sep-arated into bulk,23 particle, sphere,24 and film.25 Also, thepore of silica structure can be divided into regular andirregular categories. Here, we introduced a kind of silicastructure named mesoporous silica nanoparticle (MSN).MSN was synthesized from surfactant and silicate precursor.The structure of MSN was spherical with regular pore.According to IUPAC, the pore size of MSN was between2 nm and 50 nm. MSN has some advantages as followed:adequate of source, high surface area, tunable and regularpore size, biocompatible, and adequate of hydroxyl groupfor functionalization.

In this article, we introduced a fashionable organic/inor-ganic hybrid composite with behavior of sustained releaseand ability of specific targeting. The organic part came fromCa21-alginate and inorganic part came from MSN. Most ofsilica/alginate hybrid composites were alginate structurecoated with silica26,27 or coacervation28 of alginate andsilica together. MSNs were encapsulated into alginate toform Ca21-alginate microparticles via air-dynamic atomiza-tion. This kind of nano-in-micro structure29,30 was an inno-vative drug carrier. Air-dynamically atomization was aknown method to synthesize microparticles. This methodcan produce microparticles with high generating rate and

without organic solvent. On the other hand, MSN was aninorganic core which can enhance the loading efficiency ofdrugs. The matrix of alginate microparticles could slowdown the release rate and change the release behavior intosustained release. In the last part of the study, the surfaceof mesoporous silica nanoparticle-encapsulated alginatemicroparticles (MSN@Alg) was labeled with arginine, gly-cine, and aspartic acid (RGD)-based peptide as ligand to tar-get on cancer cell with amb3 receptor. The results showedthat RGD not only targeted on specific cell but alsoenhanced the efficiency of drug about 3.5-fold.

MATERIALS AND METHODS

MaterialsAll materials used for cell culture [fetal bovine serum (FBS),phosphate-buffered saline (PBS), modified Eagle’s medium(MEM), etc.] were purchased from Thermo Scientific. Allmaterials used for synthesis of MSN were sourced fromAldrich. Rhodamine 6G (R6G), fluorescein isothiocyanate(FITC), doxorubicin (Dox) hydrochloride, Brij-97, cetyltrime-thylammonium bromide [C16H33N(CH3)3Br], N-(3-dimethyla-minopropyl)-N0-ethylcarbodiimide hydrochloride (EDC),N-hydroxysuccinimide (NHS), 3-aminopropyltrimethoxysilane(APTMS, 97%), tetraethoxysilane (TEOS), and dimethylphthalate (DOP, >99%) were purchased from Sigma-Aldrich.RGD-based peptide, RGDYK4, was purchased from Kelowna(Taiwan). All other materials not mentioned were also pur-chased from Sigma-Aldrich.

Synthesis of MSNThe process for synthesis of MSN was referred to Wu.31

Briefly, 6.92 mL of Brij-97 was added into 180 mL of deion-ized water and stirred for 1 h in room temperature. Aftertotally dissolution of Brij-97, 0.3 mL of APTMS and 0.08 mLof DOP were added to Brij-97 aqueous solution, respectively.After stirring for 30 min, 6.7 mL of TEOS was added intoabove solution and stirred for 24 h in room temperature.After 24 h, the mixed solution was stirred with reflux at100�C for another 24 h. The precipitation was collected byfiltration and washed with methanol several times toremove the surfactant, Brij-97. Finally, the collected precipi-tation was dried in vacuum and stored for using.

Synthesis of MSN@AlgMSN@Alg was prepared with air-dynamical atomizer asshown in Scheme 1. The air-dynamically atomizer used inthis paper was VAR J30 (Nisco, Zurich, Switzerland). First,MSN was suspended into alginate aqueous solution; afterwell suspended, the solution was loaded into the syringe.The solution was extruded out by a syringe pump cooperatedwith air flow and dropped into calcium bath for cross-linkingof alginate. The as-synthesized MSN@Alg was collected bycentrifugation and dried out for store. There were severalinstrumental parameters (Nozzle diameter, flow rate of solu-tion, pressure, distance of cross-linker solution from nozzle,rotating speed of cross-linker solution) and sample parame-ters (concentration of alginate, concentration of calcium bath,concentration of MSN) for optimization. Several steps of

294 LIAO ET AL. SYNTHESIS OF MESOPOROUS SILICA NANOPARTICLE

optimization were carried out to fix the suitable instrumentaland sample parameters. To produce ideal size of MSN@Alg,the suitable instrumental parameters are: 0.35 mm for diam-eter of nozzle, 0.5 mL min21 for flow rate of solution,500 mbar for pressure and 6 cm for cross-linker solutionfrom nozzle; the suitable sample parameters are: 1 wt % forconcentration of alginate, 0.1 M for concentration of calciumbath and 0.3 wt % for concentration of MSN.

Preparation of Drugs-Loaded MSN and Drug-LoadedMSN@AlgIn this article, there are two kinds of cargo loaded inMSN@Alg, respectively. R6G was chosen as model drug toshow the release behavior of MSN and MSN@Alg. Dox waschosen as chemotherapy drug to treat with breast cancer cell(BT20). For R6G/MSN, 30 mg of MSN was suspended into1025 M R6G aqueous solution with 10 mL in volume andstirred for 24 h. For Dox/MSN, 30 mg of MSN was suspendedinto 1024 M Dox aqueous solution with 10 mL in volumeand stirred for 24 h. Cargo/MSN was collected by centrifugeand dried in vacuum. We did not have washing step aftercentrifuge. The reason is that after washing step the egg-boxstructure of alginate would collapse, and the cargo would alldiffuse out. We are aware of that there are some cargosremaining on the surface of MSN@alg microspheres. To elimi-nate the effect of cargo on the surface, we regarded therelease amount of cargo in the first refresh time as the cargoon the surface and deducted the amount when we drew therelease profile. For cargo/MSN@Alg, the process of loadingwas the same with cargo/MSN; however, the particle was notcollected but mixed with 100 mg alginate in used cargo solu-tion. After alginate was totally dissolved in solution, the solu-tion was loaded into syringe to produced cargo/MSN@Alg

with atomizer. The parameters of atomizer were the samewith “Synthesis of MSN@Alg” section.

Functionalization of Alginate With RGDYK4 PeptideThere are adequate of carboxyl group on the chain of algi-nate and amine group on the peptide. It is common to useEDC and NHS to conjugate carboxyl group and amine groupto form amide. At first, EDC and NHS were dissolved into10 mM pH 7.4 PBS with concentration of 0.27 mM and 0.14mM, respectively. Alginate was put into above solution withthe ratio of 10 mg alginate to 1 mL PBS and stirred for 1 hin 4�C. After 1 h, 12 lg of RGDYK4 was added into solutionand the solution was stirred for 4 h in 4�C. Finally, MSNwas suspended into alginate-RGD solution for atomization.

To make sure that Dox/MSN@Alg was labeled with pep-tide, Fourier transform infrared spectroscopy (FTIR; PerkinEimer, Spectrum 100) and ultraviolet–vis spectrum werechosen to analyze sample.

Particle CharacterizationThe structure and surface area of MSN were observed withfield-emission scanning electron microscopy (SEM; NovaTM,NanoSEM 230) and Nitrogen adsorption–desorption (Microme-ritics, ASAP2010). The surface potential of MSN was observedwith zeta-potential (Malvern, Nano-ZS). The distribution of MSNinside of alginate beads was observed with confocal laser scan-ning microscope (CLSM; Leica TCS SP5 II). The particle uptakein cell was traced by CLSM. The particle size distribution wasobserved with particle size analyzer (Beckman, Coulter LS230).

In Vitro Release of Drugs From MSN and MSN@AlgA 10 mg of R6G/MSN and 43 mg of R6G/MSN@Alg wereput into several volumes and concentrations, pH 7.4 PBS

SCHEME 1. An illustration showing the preparation of MSN-encapsulated alginate microspheres by air dynamically atomizer. [Color figure can

be viewed in the online issue, which is available at wileyonlinelibrary.com.]

ORIGINAL RESEARCH REPORT

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | FEB 2014 VOL 102B, ISSUE 2 295

wileyonlinelibrary.com

solution without stirring in 37�C. After specific time, thesolution was sucked up 20% and centrifuged. 50% of super-natant was moved into cuvette and remained 50% of solidwith supernatant was moved back to original solution. Atthe same time, New PBS was added into the solution tokeep total volume in same level. The concentration of R6Gwas analyzed with photoluminescence (PL; Hitachi, F-7000).At first hour of release, the solution was analyzed after5 min, 15 min, and 30 min. Between 1 h to 12 h, the solu-tion was analyzed every hour. Between 12 h and 36 h, thesolution was analyzed every 2 h. After 36 h, the solutionwas analyzed every day for 20 days. The concentration ofPBS was from 10 mM to 50 mM and the volume of PBS wasfrom 10 mL to 30 mL.

The release of Dox/MSN and Dox/MSN@Alg wassimilar with the process of R6G/MSN and R6G/[email protected] volume and concentration of PBS were 20 mL and10 mM.

R6G was exited at 526 nm and emission at 551 nm andthe photomultiplier was tuned to 400 V. Dox was exited at470 nm and emission at 558 nm, and the photomultiplierwas tuned to 700 V.

Specific Targeting and Uptake Amount of MSNTo trace the position and the amount of MSN, MSN waslabeled with FITC. FITC-MSN was encapsulated into alginateto form FITC-MSN@Alg and FITC-MSM@Alg-RGD. For spe-cific targeting, BT20 were cultured in four-well slides andthe cell density was 5 3 104 well21. Each well was treatedwith two carriers, respectively, for 4 h. After 4 h, the slideswere observed with CLSM.

For uptake amount of MSN, BT20 cancer cells were cul-tured in six-well plates and there were 2 3 105 cells well21.Each well was then treated with two carriers, respectively,for 4 h. After 4 h, trypan blue was added into plates for 10min to quench FITC outside cells. The cell membrane wasdecomposed by lysis buffer. The mixed solution was ana-lyzed with PL.

Cell Culture and Viability AssayBT20 cancer cells were routinely cultured in MEM contain-ing 10% FBS and incubated at 37�C in a humidified atmos-phere with 5% CO2.

For (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo-lium bromide (MTT) assay, BT20 cells were cultured in24-well microplate (105 cells well21) with 0.5 mL of MEMcontaining 10% FBS for 1 day. After 1 day, the mediumwas replaced with medium without FBS and cultured foranother 1 day to let the cells in starvation. Dox,MSN@Alg, Dox/MSN@Alg, and Dox/MSN@Alg-RGD withvarious concentrations were suspended into serum-freeMEM. The suspension was loaded into 24-well microplateseparately. After 24 h, MTT stock solution (0.5 mg mL21)was added into 24-well microplate. After 4 h, 0.4 mL ofDMSO was added into 24-well microplate to dissolve pur-ple crystal. Finally, the optical density was analyzed byenzyme-linked immunosorbent assay Reader at wave-length of 570 nm.

RESULTS AND DISCUSSIONS

Synthesis and Characterization of MSN@AlgMicroparticlesSEM image [Figure 1(a)] displayed the structure of MSN.The particle was sphere-like structure and the diameter wasabout 500 nm. The results of N2 adsorption–desorption ofMSN [Figure 1(b)] cooperated with methods of BET and BJHshowed that the specific area of MSN was 285 m2 g21, thepore size was about 30 nm, and the specific pore volume ofMSN was 1.08 cm3 g21. In addition for loading small molec-ular drugs, the pore diameter (30 nm) of MSN was largeenough to load macromolecular drug, such as protein, gene,and si-RNA. With 30 nm of pore size and 285 m2 g21 ofspecific pore size, MSN synthesized here was a good candi-date as vehicle to transport different cargos.

Herrero et al.32 have synthesized alginate microbeadsvia atomization and concluded that three important factors,concentration of alginate, pressure of air flow and flow rateof solution, are critical. To synthesize MSN-encapsulatedalginate beads with suitable size and structure, we sug-gested that two more factors should be considered, the dis-tance of cross-linker solution from nozzle and rotating

FIGURE 1. (a) SEM image of MSN and (b) N2 adsorption and desorp-

tion of MSN. [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com.]



speed of cross-linker solution. Table I listed the combinationof five parameters. Concentration of alginate, pressure, flowrate, distance of cross-linker solution from nozzle and rotat-ing speed of cross-linker solution are chosen as five impor-tant parameters to affect the size and structure of alginatebead. In this part, it was not necessary to suspend MSN intoalginate aqueous solution; therefore, composition of thebeads produced in Table I was alginate only. Nozzle diame-ter of 350 lm and concentration of cross-linker solution of0.1 M were kept the same from whole experiment. Therewere three variations in each parameter and totally 11groups in Table I. The particle size and structure of particleof each group were also summarized in Table I.

The solution was not viscous enough to form a sphericalstructure and the structure of alginate turns into sheet asthe concentration of alginate was below 1 wt % [0.5 wt %in Table I trail (a)]. The structure was spherical and thediameter of particle was about 20 lm as the concentrationof alginate was 2 wt % [Table I trail (b)]; the structure wasspherical and the diameter of particle was also 20 lm asthe concentration of alginate solution is 4 wt % but the algi-nate solution was too viscous and clogged the nozzle easily[Table I trail (c)]. Although it is easy to hypothesize that thediameter of particle should increase when the concentrationof alginate increased, we did not found this phenomenonand we suggested that alginate oligomers cross-link witheach other tightly and thus inhibit the swelling.

Regarding to the effect of pressure on particle size, themorphology of particle was spherical when the pressure was300 mbar and 400 mbar but the diameter of particle becamelarger when the pressure was lower [Table I trails (d) and(e)]. Only when the pressure was 500 mbar it is possible toproduce MSN@Alg microparticles with diameter in 20 lm.The distribution of particle size would be boarder when thepressure was higher than 400 mbar. Rizt and Lefebvre33

examined the mechanism of producing microparticlesthrough atomizer, and they thought that as the flow rateincreased, the flow of solution disintegrated earlier and thedrop also formed earlier. Therefore, the higher the flow ratethe smaller the drop, which makes the microbeads smaller.

To narrow the distribution of particle size, a good choicewas to tune the flow rate of solution. When the flow ratewas 1.5 mL min21, the pressure driven force cannot dispersethe alginate uniformly [Table I trail (g)]. Therefore, the resultwas that distribution of particle size was in a board rangewith varied structure. The distribution of particle becamenarrow with spherical structure when the flow rate wastuned to 0.25 mL min21 [Table I trail (f)]; however, the pro-duction efficiency was low and the flow also jams the nozzleeasily when the flow rate was tuned to 0.25 mL min21. Con-sidered with production efficiency, distribution and structure,the optimized flow rate was 0.5 mL min21. Dombrowski andJohns34 demonstrated that slow flow rate made the film ofsolution thinner and thinner liquid film broke into smallerdrops. They also suggested that the diameter of drops wasroughly proportional to the square root of the film thickness.

When the droplets were extruded out from nozzle, thedroplets were not spherical but teardrop like. The dropletsneeded enough distance to arrive terminal velocity. Afterreaching to a terminal velocity, the velocity inside dropletwas uniform. At that time, the structure of droplet wasshaped into sphere to reach the lowest free energy. As thedistance between cross-linker solution and nozzle was 2 cm,the droplet has not reached terminal velocity yet. The struc-ture of particles was sheet with 2 cm in distance [Table Itrail (h)]. As the distance was 11 cm, the droplets hadhigher chance to combine with others in the surrounding.After combination of droplet, the diameter of particle waslarger than before [Table I trail (i)].

After the droplet dropped into cross-linker solution, algi-nate started the cross-linking process. The particles aggre-gated together easily when the rotating speed was as slowas 150 rpm [Table I trail (j)]. When the rotating speed wasraised to 500 rpm to prevent particles from aggregation, theshear stress shaped the particle into olive but not sphere.

Finally, trail (b) was chosen as the most suitable param-eters: hereafter 2 wt % for concentration of alginate,500 mbar for pressure, 0.5 mL min21 for flow rate of solu-tion, 6 cm for the distance from cross-linker solution to noz-zle and 300 rpm for rotating speed of cross-linker solution.

TABLE I. Several Parameters for Controlling Particle Size of MSN@Alg Samples. [Color table can be viewed in the online

issue, which is available at wileyonlinelibrary.com.]

The shading indicates the 3 different values of each parameters (ex: only 0.5%, 2% and 4%). *means the structures of the beads are not

constant in shape.



TrailConcentration

of AlginatePressure(mbar)

Flow Rate(mL/min) Height (cm)

RotatingSpeed (rpm) Structure

ParticleSize (lm)

(a) 0.5wt% 500 0.5 6 350 Sheet N/A(b) 2 wt% 500 0.5 6 350 Sphere 20(e) 4 wt% 500 0.5 6 350 Sphere 20(d) 2 wt% 300 0.5 6 350 Sphere 50(e) 2wt% 400 0.5 6 350 Sphere 70(f) 2 wt% 500 0.25 6 350 Sphere 15(g) 2wt% 500 1.5 6 350 Mix* N/A(h) 2 wt% 500 0.5 2 350 Sheet N/A(i) 2 wt% 500 0.5 11 350 Sphere 30(i) 2wt% 500 0.5 6 150 Sphere 31(k) 2 wt% 500 0.5 6 500 Olive 20

The diameter of particle counted by optical microscope wasabout 20 lm to 30 lm. Counted with particle size analyzer,most of particles had diameter in 20 lm (Figure 2). Withsuspension of MSN, alginate solution in 2 wt % was too vis-cous and jammed the nozzle, too. To avoid the interruptedjam, the concentration of alginate was changed to 1 wt %.The concentration of MSN was chosen as 0.3 wt %, whichwas maximum amount of MSN that could be suspendedwell in alginate solution. Not only stabilizing the particlesize, MSNs can also enhance the structure of alginate beads.To make sure that the structure and diameter of particle

did not changed after adding MSN into alginate solution,MSN@Alg microparticles were counted with particle sizeanalyzer, and the diameter was about 20 lm, too.

FITC was used as a tracer and labeled on the surface ofMSN. FITC-MSN-encapsulated alginate microbeads (FITC-MSN@Alg) were synthesized to analyze the distribution ofMSN inside the alginate microbeads. From various imageswith different z-axis, the distribution of FTIC-MSNs on z-axis can be analyzed clearly. Figure 3 was a supposedFITC-MSN@Alg divided into six planes based on the heightof z-axis. From Figure 3(c), some green light appeared inupper plane and disappeared in lower plane, which meansthat FITC-MSNs suspended well inside the alginate bead.There was no green spot appearing in uppermost and low-ermost plane [Figure 3(a,f), respectively]. It indicates thatMSNs were encapsulated inside alginate microbeads butnot on the surface of alginate microbeads. MSNs did notaggregate together but suspended as tridimension insidethe alginate beads.

To target the cancer cell, RGD-based peptide, RGDYK4,was labeled on the surface of MSN@Alg microbeads as aligand. The ratio of alginate:EDC:peptide is 1000:5:2 mol.Shimoda35 reported that the concentration of RGD shouldnot be higher than 100 nM because high concentration ofRGD would not enhance attachment but induce oppositeeffect.35 The results of FTIR (Figure 4) showed that therewas a signal representing amine I from lysine (1680cm21),36 which demonstrates that RGD were labeled onMSN@Alg beads successfully. Alginate had a very strong sig-nal28 at 1600 cm21 and low ratio of RGDYK4 to alginate

FIGURE 2. A typical optical microscopy of alginate beads followed by

parameters in Table I trail (b).

FIGURE 3. Distribution of MSN inside alginate bead by scanning on z-axis. Image (a) is the highest surface and image (f) is the lowest surface.

[Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]



(1:500 mol) made the amine I signal (1680 cm21) in FTIRnot sharp.

Release Behavior of ParticlesThe amount of R6G and Dox in MSN was previously calcu-lated to be 3.1E–9 mol mg21 and 4.7E–9 mol mg21, respec-tively. The equation of loading efficiency is as the followingequation:

Loading efficiency5total amount of durg2the rest of drug

total amount of drug

3volume of solution

weight of MSN3100%

(1)

The adsorbing efficiency of R6G and Dox was 29% and4.7%, respectively. The loading efficiency of Dox was low(4.7%). That was because Dox was more hydrophobic thanR6G, the affinity between MSN and Dox was weaker thanMSN and R6G. Only a few amount of Dox was adsorbed onMSN, and most of Dox still dissolved in solution.

Srivastava and coworkers30 has used gelatin nanopar-ticles as core and encapsulated nanoparticles by alginatemicroparticles via air dynamic atomizer. After that, the syn-thesized nano-in-micro particles were used for drug releasestudies. The author proved that without encapsulating themicroparticles, the release of gelatin nanoparticles had 25%of burst and stopped after 5 days. The nano-in-micro par-ticles, however, decreased the burst to 17% and prolongedthe time to 10 days.

Generally, release time was used as x-axis but the meas-urements were taken in different intervals and kept in longtime (20 days) in our experiment. The release behaviorwould not be obvious if release time were used as x-axis.Therefore, refresh times were used to replace time as x-axis.The 28th refresh represented the measurement at 48 h, andthe 46th refresh represented measurement at 20 days. The

release behavior of R6G from MSN@Alg included threesteps: diffusion of cargo from MSN to alginate, diffusion ofcargo from alginate to medium, transference of MSN fromalginate to medium and the erosion of Ca21-alginate. Therelease rate mentioned later would all relative value insteadof absolute release rate.

Figure 5(a) illustrated that the release rate of R6G fromR6G/MSN@Alg was slower than R6G/MSN, and the releasebehavior could be kept up to 20 days in 20 mM PBS. Therewere more phosphate molecules to erode Ca21-alginate.Compared with releasing from R6G/MSN, R6G releasedfrom R6G/MSN@Alg with slower rate in PBS. Although therelease amount and release rate of R6G from R6G/MSN@Algin 20 mM, 10 mL PBS was fewer and slower than that fromR6G/MSN [Figure 5(a)], the amount of R6G released fromR6G/MSN@Alg was close to that from R6G/MSN. Theobserved release behavior was ideal for sustained release.With the same concentration, the amount of phosphate mol-ecules could be increased by expanding the volume of PBS.The release rate of R6G from R6G/MSN@Alg was alsoslower than R6G/MSN, and the release behavior also couldbe kept for 20 days even in 20 mL of PBS [Figure 5(b)].The rationale was also that more phosphate molecules erod-ing Ca21-alginate taken place at the same time. As men-tioned, the release behavior of R6G/MSN@Alg in 10 mM, 10mL PBS was not suitable for sustained release. As the vol-ume of PBS was increased to 20 mL [Figure 5(a)], the

FIGURE 5. Release behavior of R6G loaded in MSN and MSN@Alg

from (a) 20 mM, 10 mL PBS and (b) from 10 mM, 20 mL PBS. [Color

figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

FIGURE 4. FTIR of materials. Alg-RGD represents the signal of RGD-

labeled alginate. RGD represents the signal of pure RGD peptide. Alg

represents the signal of pure alginate. EDC represents the signal of

EDC and NHS. [Color figure can be viewed in the online issue, which

is available at wileyonlinelibrary.com.]





difference of release behavior between MSN and MSN@Algwas more obvious. Although there was same amount ofphosphate molecules in 20 mM, 10mL PBS and 10 mM, 20mL PBS, MSN@Alg microbeads had better behavior of sus-tained release in 10 mM, 20 mL PBS.

Both the concentration and the volume of PBS affectedthe release behavior of MSN@Alg but only the volume ofPBS would affect the release behavior of MSNs. It wasimportant to control the ratio of the amount of MSN@Alg tothe volume of PBS to induce sustained release. The opti-mized ratio of them was 0.5 mg of MSN@Alg (1 wt %) to1 mL PBS (10 mM). With this optimized ratio, the releasebehavior turned into sustained release and kept going in20 days. The cargo, then, was changed to Dox, and the fac-tors of release were selected with 10 mM, 20 mL PBS and1 wt % Dox/MSN@Alg. Dox stopped releasing from Dox/

MSN after 5 h but still released from Dox/MSN@Alg after48 h (data not shown). Although the release behavior ofsustained release was not obvious as R6G/MSN@Alg, Dox/MSN@Alg also could lead to sustained release in this opti-mized ratio.

Internalization of ParticlesBreast cancer cells BT-20 were fed with FITC-MSN@Alg-RGD and FITC-MSN@Alg. The images of CLSM illustratedthat there were less FITC-MSNs swallowed by cells fed withFITC-MSN@Alg [Figure 6(a)] than that fed with FITC-MSN@Alg-RGD [Figure 6(b)]. Generally, material can beinternalized into cell by clathrin-mediated endocytosis,which was also known as receptor-mediated endocytosis(RME). Park37 used hyaluronic acid as carrier and CD44 asligand to deliver drug into cell. Corre38 mentioned that RGDcan enhance the amount of uptake about twofolds. Theauthor also mentioned that compared with nanoparticle,microparticle had higher payload. Thus, functionalizedmicroparticle had higher efficiency than functionalizednanoparticle. Shimoda35 mentioned that RGD not onlyinduced RME but also induced macropinocytosis. And Desi-mone39 stated that the particle with 2 lm had the best effi-ciency of uptake by macropinocytosis. We constructed ahypothesis. At first, RGD ligand combined with integrinreceptor; in other words, FITC-MSN@Alg microbeadsattached on the surface of cell. Then, cell started to progressof RME as Ca21-alginate on FITC-MSN@Alg microbeadswere eroded by phosphate molecules. The cell swallowedFITC-MSNs when FITC-MSNs were dropped out to the sur-face of cell.

To proof our hypothesis, we used lysis buffer to decom-pose cell membrane and analyzed the amount of FITC-MSNsinside two cell lines with PL. Because we have alreadyquenched the fluorescence outside membrane and on thesurface of membrane by trypan blue,40 we could make surethat the fluorescence was from FITC-MSN inside cell. Figure7 showed that the intensity of fluorescence from FITC-MSN@Alg-RGD was higher than that from FITC-MSN@Alg.

FIGURE 6. Confocal microscopy images of BT20 cancer cells treated

with (a) FITC-MSN@Alg and (b) FITC-MSN@Alg-RGD. 1: Merged

image, 2: transmission image, 3: DAPI channel, and 4: FITC channel.

[Color figure can be viewed in the online issue, which is available at


FIGURE 7. Uptake amount of FITC-MSN@Alg and FITC-MSN@Alg-

RDG in cancer cell BT-20. [Color figure can be viewed in the online

issue, which is available at wileyonlinelibrary.com.]




In Vitro CytotoxicityA good drug carrier material should have higher biocompat-ibility to reduce the chance of harming normal cell. BT20cells were treated with MSN@Alg for the test of cytotoxicity.Figure 8 showed that even at high concentration ofMSN@Alg,10 mg mL21, there were still at least 80% viable.Generally, when the diameter of particles is smaller than100 nm, the cytotoxicity of particles would increase obvi-ously; on the other hand, when the diameter of particles islarger than 500 nm, the cytotoxicity of particles woulddecrease.41 Because the diameter of Alg was about 20 lmand the diameter of MSN was about 500 nm, the cell couldbe alive when the concentration of carrier was as high as10 mg mL21. It was confirmed again that MSN@Alg was agood drug carrier.

Most of anticancer drugs are membrane impermeable.The lethal concentration 50 (LC50) of Dox to BT20 after 24 hwas 2.3 lM. Figure 9 showed that after Dox was loaded intoMSN to form Dox/MSN@Alg and Dox/MSN@Alg-RGD, thecarriers slightly decreased cell viability. After calculation, theLC50 of Dox only decreased from 2.1 lM to 1.8 lM. Whencarrier was modified with RGD, it can enhance the efficiencyof drug to cancer cells. Consequently, alginate microparticles

modified with RGD can target on cell specifically and alsoenhance the deliver efficiency of anticancer drug.

CONCLUSIONS

We have used atomizer to synthesize MSN@Alg micropar-ticles successfully. By controlling five parameters, differentkinds of composites were synthesized with different applica-tions. The spherical MSN@Alg microparticles with 20 lmwere used for sustained release. The suitable ratio ofMSN@Alg to PBS for sustained release was 1 mg ofMSN@Alg to 2 mL PBS (10 mM) and the release could bekept for 20 days in this situation. The cell viability test evi-denced that the fabricated MSN@Alg microparticles hadexcellent biocompatibility. BT20 were treated with Dox/MSN@Alg and Dox/MSN@Alg-RGD, and the results showedthat when carrier was labeled with RGD ligand, it could tar-get to cell specifically with enhanced deliver efficiency. TheMSN@Alg microparticles prepared in this study would be agood candidate featured with behavior of sustained releaseand specific targeting in drug delivery system.

REFERENCES

1. Slowing II, Vivero-Escoto JL, Wu CW, Lin VSY. Mesoporous silica

nanoparticles as controlled release drug delivery and gene trans-

fection carriers. Adv Drug Deliv Rev 2008;60(11):1278–1288.

2. Zhang CM, Li CX, Huang SS, Hou ZY, Cheng ZY, Yang PP, Peng

C, Lin J. Self-activated luminescent and mesoporous strontium

hydroxyapatite nanorods for drug delivery. Biomaterials 2010;

31(12):3374–3383.

3. Jatzkewitz H. An ein kolloidales blutplasmaersatzmittel (polyvinyl-

pyrrolidon) gebundenes peptamin (glycyl-L-leucyl-mezcalin) als

neuartige depotform fur biologisch aktive primare amine (mezca-

lin). Z Naturforsch 1955;10:27–31.

4. Chiou WL, Smith LD. Solid dispersion approach to the formula-

tion of organic liquid drugs using polyethylene glycol 6000 as a

carrier. J Pharm Sci 1971;60(1):125–127.

5. Molteni L. Dextran and insulin conjugates as drug carriers. Meth-

ods Enzymol 1985;112:285–298.

6. Kawaguchi H. Functional polymer microspheres. Progr Polym Sci

2000;25(8):1171–1210.

7. Jones MC, Leroux JC. Polymeric micelles—A new generation of

colloidal drug carriers. Eur J Pharm Biopharm 1999;48(2):101–111.

8. Peppas NA, Keys KB, Torres-Lugo M, Lowman AM. Poly(ethylene

glycol)-containing hydrogels in drug delivery. J Controlled

Release 1999;62(1-2):81–87.

9. Kim JH, Kim YS, Park K, Kang E, Lee S, Nam HY, Kim K, Park JH,

Chi DY, Park RW, Kim IS, Choi K, Chan Kwon I. Self-assembled

glycol chitosan nanoparticles for the sustained and prolonged

delivery of antiangiogenic small peptide drugs in cancer therapy.

Biomaterials 2008;29(12):1920–1930.

10. Yun YH, Goetz DJ, Yellen P, Chen WL. Hyaluronan microspheres

for sustained gene delivery and site-specific targeting. Biomateri-

als 2004;25(1):147–157.

11. Nanjwade BK, Singh J, Parikh KA, Manvi FV. Preparation and

evaluation of carboplatin biodegradable polymeric nanoparticles.

Int J Pharm 2010;385(1-2):176–180.

12. Clark AH, Rossmurphy SB. Structural and mechanical-properties

of bio-polymer gels. Adv Polym Sci 1987;83:57–192.

13. Novikova LN, Mosahebi A, Wiberg M, Terenghi G, Kellerth JO,

Novikov LN. Alginate hydrogel and matrigel as potential cell car-

riers for neurotransplantation. J Biomed Mater Res Part A 2006;

77A(2):242–252.

14. Tan WH, Takeuchi S. Monodisperse alginate hydrogel microbeads

for cell encapsulation. Adv Mater 2007;19(18):2696–2701.

15. Knill CJ, Kennedy JF, Mistry J, Miraftab M, Smart G, Groocock

MR, Williams HJ. Alginate fibres modified with unhydrolysed and

FIGURE 8. Cytotoxicity of MSN@Alg microparticles to cancer cell BT-

20. [Color figure can be viewed in the online issue, which is available

at wileyonlinelibrary.com.]

FIGURE 9. Cell viability of BT-20 treated with Dox-loaded MSN@Alg

microparticles. LC50 of Dox/MSN@Alg was 2.1 lM. [Color figure can

be viewed in the online issue, which is available at






hydrolysed chitosans for wound dressings. Carbohydr Polym

2004;55(1):65–76.

16. West TP, Preece JE. Bulk alginate encapsulation of Hibiscus

moscheutos nodal segments. Plant Cell Tiss Organ Cult 2009;

97(3):345–351.

17. Srivastava R, McShane MJ. Application of self-assembled ultra-

thin film coatings to stabilize macromolecule encapsulation in

alginate microspheres. J Microencapsulation 2005;22(4):397–411.

18. Motwani SK, Chopra S, Talegaonkar S, Kohl K, Ahmad FJ, Khar

RK. Chitosan-sodium alginate nanoparticles as submicroscopic

reservoirs for ocular delivery: Formulation, optimisation and in

vitro characterisation. Eur J Pharm Biopharm 2008;68(3):513–525.

19. Hu Y, Hu N. pH-dependent behaviors of electroactive myoglobin

loaded into layer-by-layer films assembled with alginate and

hydroxyethyl cellulose ethoxylate. J Phys Chem B 2008;112(31):

9523–9531.

20. Chan ES, Lim TK, Ravindra P, Mansa RF, Islam A. The effect of

low air-to-liquid mass flow rate ratios on the size, size distribution

and shape of calcium alginate particles produced using the atom-

ization method. J Food Eng 2012;108(2):297–303.

21. Yanagisawa T, Shimizu T, Kuroda K, Kato C. The preparation of

alkyltrimethylammonium-kanemite complexes and their conver-

sion to microporous materials. Bull Chem Soc Jpn 1990;63(4):

988–992.

22. Beck JS, Vartuli JC, Roth WJ, Leonowicz ME, Kresge CT, Schmitt

KD, Chu CTW, Olson DH, Sheppard EW, McCullen SB. A new fam-

ily of mesoporous molecular-sieves prepared with liquid-crystal

templates. J Am Chem Soc 1992;114(27):10834–10843.

23. Nakahira A, Nagata H, Onoki T, Yamasaki Y. Evaluation of micro-

structures and properties of bulk mesoporous silica. Res Chem

Intermed 2008;34(4):347–352.

24. Hu Y, Zhi Z, Zhao Q, Wu C, Zhao P, Jiang H, Jiang T, Wang S. 3D

cubic mesoporous silica microsphere as a carrier for poorly solu-

ble drug carvedilol. Microporous Mesoporous Mater 2012;147(1):

94–101.

25. Grosso D, Balkenende AR, Albouy PA, Ayral A, Amenitsch H,

Babonneau F. Two-dimensional hexagonal mesoporous silica thin

films prepared from black copolymers: Detailed characterization

amd formation mechanism. Chem Mater 2001;13(5):1848–1856.

26. Haufova P, Knejzlik Z, Jaroslav H, Zadrazil A, Stepanek F. Reversi-

ble buckling and diffusion properties of silica-coated hydrogel

particles. J Colloid Interface Sci 2011;357(1):109–115.

27. Lim SY, Kim KO, Kim DM, Park CB. Silica-coated alginate beads

for in vitro protein synthesis via transcription/translation machin-

ery encapsulation. J Biotechnol 2009;143(3):183–189.

28. Xu SW, Jiang ZY, Lu Y, Wu H, Yuan WK. Preparation and catalytic

properties of novel alginate-silica-dehydrogenase hybrid biocom-

posite beads. Ind Eng Chem Res 2006;45(2):511–517.

29. Coradin T, Livage J. Mesoporous alginate/silica biocomposites for

enzyme immobilisation. C R Chim 2003;6(1):147–152.

30. Joshi A, Keerthiprasad R, Jayant RD, Srivastava R. Nano-in-micro

alginate based hybrid particles. Carbohydr Polym 2010;81(4):790–

798.

31. Chang RHY, Jang J, Wu KCW. Cellulase immobilized mesoporous

silica nanocatalysts for efficient cellulose-to-glucose conversion.

Green Chem 2011;13(10):2844–2850.

32. Herrero EP, Del Valle EMM, Galan MA. Development of a new

technology for the production of microcapsules based in atomiza-

tion processes. Chem Eng J 2006;117(2):137–142.

33. Rizk NK, Lefebvre AH. The influence of liquid-film thickness on

airblast atomization. J Eng Power Trans ASME 1980;102(3):706–

710.

34. Dombrowski N, Johns WR. The aerodynamic instability and disin-

tegration of viscous liquid sheets. Chem Eng Sci 1963;18(3):203–

214.

35. Shimoda A, Sawada S, Akiyoshi K. Cell specific peptide-

conjugated polysaccharide nanogels for protein delivery. Macro-

mol Biosci 2011;11(7):882–888.

36. Morgan AW, Roskov KE, Lin-Gibson S, Kaplan DL, Becker ML,

Simon CG Jr. Characterization and optimization of RGD-

containing silk blends to support osteoblastic differentiation. Bio-

materials 2008;29(16):2556–2563.

37. Choi KY, Yoon HY, Kim JH, Bae SM, Park RW, Kang YM, Kim IS,

Kwon IC, Choi K, Jeong SY, Kim K, Park JH. Smart nanocarrier

based on PEGylated hyaluronic acid for cancer therapy. ACS

Nano 2011;5(11):8591–8599.

38. Brandhonneur N, Chevanne F, Vie V, Frisch B, Primault R, Le

Potier M-F, Le Corre P. Specific and non-specific phagocytosis of

ligand-grafted PLGA microspheres by macrophages. Eur J Pharm

Sci 2009;36(4-5):474–485.

39. Petros RA, DeSimone JM. Strategies in the design of nanopar-

ticles for therapeutic applications. Nat Rev Drug Discovery 2010;

9(8):615–627.

40. Rosenholm JM, Meinander A, Peuhu E, Niemi R, Eriksson JE,

Sahlgren C, Linden M. Targeting of porous hybrid silica nanopar-

ticles to cancer cells. ACS Nano 2009;3(1):197–206.

41. Yu KO, Grabinski CM, Schrand AM, Murdock RC, Wang W, Gu B,

Schlager JJ, Hussain SM. Toxicity of amorphous silica nanopar-

ticles in mouse keratinocytes. J Nanopart Res 2009;11(1):15–24.


synthesis of mesoporous silica nanoparticle-encapsulated alginate microparticles for sustained...

Documents