Journal of Controlled Release 147 (2010) 101–108

Contents lists available at ScienceDirect

Journal of Controlled Release

j ourna l homepage: www.e lsev ie r.com/ locate / jconre l

GENEDELIVERY

Carbonate apatite-facilitated intracellularly delivered siRNA for efficient knockdownof functional genes

Sharif Hossain a, Anthony Stanislaus b, Ming Jang Chua b, Seiichi Tada a, Yoh-ichi Tagawa a,Ezharul Hoque Chowdhury a,b,⁎, Toshihiro Akaike a,⁎a Department of Biomolecular Engineering, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japanb Faculty of Medicine and Health Science, International Medical University (IMU), No. 126, Jalan 19/155B, Bukit Jalil, 57000 Kuala Lumpur, Malaysia

Abbreviations: DMEM, Dulbecco's modified Eagle'sscattering; TEM, transmission electron microscope; GFRNAi, RNA interference; siRNA, small interfering RNA; s⁎ Corresponding authors. Chowdhury andAkaike are to

of Biomolecular Engineering, Graduate School of BioscieInstitute of Technology, 4259 Nagatsuta-cho, Midori-kuTel.: +81 55 928 5267; fax: +81 55 928 5267.

E-mail addresses: ezharul_chowdhury@imu.edu.mytakaike@bio.titech.ac.jp (T. Akaike).

0168-3659/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.jconrel.2010.06.024

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 March 2010Accepted 28 June 2010Available online 8 July 2010

Keywords:Carbonate apatitesiRNAGene silencingCyclin B1ApoptosisCancer cell

Gene therapy through intracellular delivery of a functional gene or a gene-silencing element is a promisingapproach to treat critical diseases. Elucidation of the genetic basis of human diseases with completesequencing of human genome revealed many vital genes as possible targets in gene therapy programs. RNAinterference (RNAi), a powerful tool in functional genomics to selectively silence messenger RNA (mRNA)expression, can be harnessed to rapidly develop novel drugs against any disease target. The ability ofsynthetic small interfering RNA (siRNA) to effectively silence genes in vitro and in vivo, has made themparticularly well suited as a drug therapeutic. However, since naked siRNA is unable to passively diffusethrough cellular membranes, delivery of siRNA remains the major hurdle to fully exploit the potential ofsiRNA technology. Here pH-sensitive carbonate apatite has been developed to efficiently deliver siRNA intothe mammalian cells by virtue of its high affinity interactions with the siRNA and the desirable size of theresulting siRNA/apatite complex for effective cellular endocytosis. Moreover, following internalization bycells, siRNA was found to be escaped from the endosomes in a time-dependent manner and finally, moreefficiently silenced reporter genes at a low dose than commercially available lipofectamine. Knockdown ofcyclin B1 gene with only 10 nM of siRNA delivered by carbonate apatite resulted in the significant death ofcancer cells, suggesting that the new method of siRNA delivery is highly promising for pre-clinical andclinical cancer therapy.

medium; DLS, dynamic lightP, green fluorescence protein;hRNA, short hairpin RNA.be contacted at theDepartmentnce and Biotechnology, Tokyo, Yokohama 226-8501, Japan.

(E.H. Chowdhury),

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Genes encode proteins through messenger RNA (mRNA) to carryout the major functions of a biological system and a disorder eitheracquired or genetic is usually associated with the suppression or theoverexpression of certain genes. Regulation of the gene expressionparticularly through the delivery of exogenous gene(s) or gene-silencing element(s) could assist in restoring the regular physiologicalfunctions for treatment of a genetic or an acquired disease. RNAinterference (RNAi) being one of themechanisms to selectively silencemRNA expression can be harnessed to rapidly develop novel drugsagainst target genes [1–6]. There are two basic ways of implementingRNAi for selective gene inhibition: 1) cytoplasmic delivery of short

interfering RNA (siRNA) and 2) nuclear delivery of gene expressionplasmid to express a short hairpin RNA (shRNA) [7]. Silencing bysynthetic siRNA, an RNA duplex of 21–23 nucleotides, is moreadvantageous than shRNA partly due to the difficulty of constructingshRNA expression systems prior to the selection and verification of theactive sequences [7] and the requirement of the expression system tocross the nuclear membrane for shRNA expression [8]. The ability ofsiRNA to potently, but reversibly, silence genes in vivo has made thema highly promising drug therapeutic with several different clinicaltrials ongoing and more poised to enter the clinic in the future [2,5].However, due to the strong anionic phosphate backbone withconsequential electrostatic repulsion from the anionic cell membrane,siRNA is unable to passively diffuse across the membrane [9]. For theintracellular delivery of siRNA, both viral or non-viral vectors havebeen investigated. Although the viral vectors are highly efficient, theyare limited to shRNA delivery and remain highly immunogenic andcarcinogenic. On the other hand, the non-viral systems are promisingalternatives for siRNA delivery since they are relatively safe and cost-effective. Being usually polycationic, they are able to form complexeswith anionic nucleic acid, protecting it from nuclease attack andfacilitating cellular uptake through electrostatic interactions withnegatively charged plasmamembrane or through specific interactions

102 S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

between the ligand attached to the complex and the receptor on cellmembrane [8]. Among the non-viral vectors, both polyplexes as wellas lipoplexes have been found efficient for siRNA delivery withsignificant gene-silencing effect both in vitro and in vivo [10–38].However, synthetic non-viral systems are inefficient and an increase inperformance is often associated with an increase in cytotoxicity [39].The major obstacle for siRNA delivery in the non-viral route is thedegradation of a significant portion of the internalized siRNA bylysosomal nucleases [40]. The endosomal escape of siRNA is, therefore,a crucial step in successful gene silencing.

We have recently developed an efficient delivery system based onsome fascinating properties of carbonate apatite-ability of preventingcrystal growth for generation of nanoscale particles as needed forefficient endocytosis and fast dissolution kinetics in endosomal acidiccompartments to facilitate the release of delivered therapeutics fromthe particles and endosomes [41–50]. Here, we show that pH-sensitivecarbonate apatite particles havinghigh affinity interactionswith siRNAs,mediate efficient endocytosis and subsequent endosomal escape ofthe siRNAs, leading to the silencing of reporter gene expression moreeffectively than commercially available lipofectamine. Moreover, nano-particle-assisted delivery of validated siRNA against cyclin B1 results insignificant inhibition of cancer cell growth.

2. Materials and methods

2.1. Reagents

Plasmid pGL3 (Promega) containing a luciferase gene under SV40promoter and pEGFP-N2 (CLONTECH Laboratories, Inc.) containinggreen fluorescence protein gene under CMV were propagated in thebacterial strain XL-1 Blue (as described in Molecular Cloning) andpurified by QIAGEN plasmid kits. LysoTracker™ Red DND-99, MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) andDMEM were purchased from Molecular Probes, Sigma and Gibco BRL,respectively. The lipofectamine 2000 transfection reagent was obtainedfrom Invitrogen™ corporation, California, USA. Luciferase GL3 siRNA(Target sequence 5′-AACTTACGCTGAGTACTTCGA-3′ ), GFP-22 siRNA(Target sequence 5′-CGGCAAGCTGACCCTGAAGTTCAT-3′), siRNAagainst cyclin B1 (Target sequence 5′-AACACTTATACTAAGCACCAA-3′)and all Stars Neg. siRNA Fluorescein were purchased from QIAGEN.siRNAs were delivered in the lyophilized form and upon delivery thesiRNAs were diluted to obtain a 20 μM solution using RNAse-free waterprovided by Qiagen. The siRNA solutionwas then allocated intomultiplereaction tubes for storage as repeated thawing might affect siRNA'ssilencing efficiency. The siRNAswere stored at−20 °C as recommendedby Qiagen.

2.2. Cell culture

HeLa cells were cultured in 75-cm2flasks in Dulbecco's modified

Eagle's medium (DMEM, Gibco BRL) supplemented with 10% fetalbovine serum (FBS), 50 μg penicillin ml–1, 50 μg streptomycin ml–1and 100 μg neomycin ml–1 at 37 °C in a humidified 5% CO2-containingatmosphere.

2.3. Formation of siRNA/carbonate apatite particles and transfection ofcells

Cells fromtheexponentially growthphasewere seededat 50,000 cellsper well into 24-well plates the day before transfection. 3–6 μl of 1 MCaCl2 was mixed with 100 pM–100 nM of siRNA in 1 ml of fresh serum-free HCO3

− (44 mM)-buffered DMEM medium (pH 7.5), followed byincubation at 37 °C for 30 min for complete generation of siRNA/carbonate apatite particles. Medium with generated siRNA-containingparticles was addedwith 10% FBS to the rinsed cells. After 4 h incubation,the mediumwas generally replaced with serum-supplemented medium

and the cells were cultured up to 24–72 h depending on the assay. Insome experiments, siRNA/apatite particles were continuously incubatedwith the cells for 48 h. siRNA/lipofectamine formulation and transfectionwere done according to the procedure provided by Invitrogen.

2.4. Particle size measurements

Thedistribution of particle sizewasmeasured using FPAR-1000fiber-optics particle analyzer (Otsuka Electronics, Osaka, Japan) equippedwitha 660 nm diode laser. The measurement was carried out by a scatteringangle of 90° at 25 °C. The size distribution was obtained by CONTINmethod.

2.5. Observation of siRNA/apatite complexes with transmission electronmicroscope (TEM)

siRNA/apatite complexes preparedwith 100 nMof siRNA and 5 mMof CaCl2were observedunderH-7500 transmission electronmicroscope(Hitachi, Tokyo, Japan) at an acceleration voltage of 80 kV. After loadingof the particles on the copper grid with collodion membrane, the gridwas dried in the air.

2.6. Determination of siRNA loading efficiency

The amount of fluorescein-labeled siRNA adsorbed onto apatitenanoparticles was determined from the fluorescence intensity ofthe pellet obtained after centrifugation of siRNA/apatite complexes.Following generation of siRNA/apatite particles as described above,using 5 mM Ca2+ and 1–200 nM of fluorescein siRNA and centrifuga-tion at 15,000 rpm for 3 min, the resulting pellet was washed 3 timeswith the same medium and dissolved in 100 μl of 10 mM EDTA-PBS.The whole dissolved particle solution was taken to an assay plate andquantified for the fluorescence intensity by a fluorescence microplatereader. Free fluorescein-labeled siRNA (1 pM to 200 nM) in PBS wasquantified and plotted to make the calibration curve. siRNA loadingwas quantified with the help of the calibration curve.

2.7. Estimation of the dissolution of carbonate apatite particles in acidicsolution

200 ml of siRNA/apatite complex suspension was prepared fromDMEM containing 100 nM of fluorescein-labeled siRNA and 6 mM ofexogenously added CaCl2. The pH of the suspension was decreased byadding 1 N HCl and 1 ml of the suspension at each pH was taken.Optical density at 320 nm of the collected sample was measured withSmartSpec™ 3000 spectrophotometer (Bio-Rad).

2.8. Cellular uptake and intracellular siRNA measurement in HeLa cellline

siRNA/apatite and siRNA/lipfectamine 2000 particles were pre-pared in the presence of florescein siRNA (100 nM) and were addedonto the HeLa cells with or without (in case of Lipofectamine 2000)10% serum and kept for 4 h in the incubator for cellular internalization.Extracellular particles were removed by EDTA prior to observation ofthe cells by a fluorescence microscope (Olympus-IX71). For quantita-tive analysis, HeLa cells were lysed using the lysis buffer (NP40)following 4 h incubation of siRNA/apatite and siRNA/lipofectamine2000 particle suspensions with the cells and the intracellularfluorescein intensity was determined using a microplate reader (DTX880, Multimode Detector by BECKMAN COULTER). Controls were theintensities for the delivery of free fluorescein siRNA and the cellswithout any particle.

Fig. 1. Binding affinity of siRNA to apatite particles. Particles were prepared by addition ofdifferent concentrations offluorescein siRNAs (1–200 nM) and 5 mMCa 2+ to bicarbonate(44 mM)-buffered DMEM medium (pH-7.5), followed by incubation at 37 °C for 30 min.The generatedparticleswere centrifuged at 15,000 rpm for 10 min and the resulting pelletwaswashedwith the samemedium. Following dissolutionof thepelletwith10 mMEDTA-PBS in 100 μl, the total volumeof thedissolvedparticle solutionwas taken to an assay plateto quantify theflorescence intensity by a spectrophotometer. Thefluorescence intensity offree fluorescein siRNA (1 nM to 200 nM) in PBS was quantified and plotted to make thecalibration curve for subsequent quantitation of siRNA loading into the particles.

103S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

2.9. Intracellular localization of siRNAs with apatite nanoparticles

For confocal microscopic analysis of intracellular localization ofsiRNA, HeLa cells were seeded at 15,000 cells per cm2 on 0.1% (w/v)collagen-coated glass coverslips (24×24 mm) on the day beforeintracellular delivery. Fluorescein siRNA/apatite nanoparticles wereadded onto HeLa cells in the same manner as mentioned above. Afteradding into the cells and removal of residual complexes at 1, 2 or 4 h,

Fig. 2. Morphology and size determination of siRNA/apatite complexes by (A, B) dynamic liformed by mixing 5 mM Ca 2+ in DMEM along with 100 nM of siRNA, followed by incubati

endosomes and lysosomeswere stainedwith LysoTracker™ Red DND-99 (Molecular Probes) according to the manufacturer's protocol andthe cells were fixed with formaldehyde solution. Additionally, nucleiwere stained with 4′, 6-diamino-2-phenylindole (DAPI) and observedwith A1 confocal laser scanning microscope (Nikon, Tokyo, Japan).

2.10. Gene silencing

HeLa cells expressing GFP and luciferase were generated bytransfection with plasmid DNA encoding GFP and luciferase using theapatite and the lipofectamine 2000 according to the manufacturer'sinstructions. Briefly, cells were seeded on a 24-well tissue cultureplate at a density of 0.5×105 cells per well and incubated overnight inDMEM. On the day of transfection, the medium was removed andreplaced with fresh media without serum. Apatite/luciferase plasmidor lipofectamine/luciferase plasmid complexes were added onto thecells in eachwell and incubated for 4 h. Mediawere then removed andthe cells were washed with PBS, followed by replenishing fresh mediacontaining serum. The cells were incubated for 24 h at 37 °C underCO2 atmosphere. In the next day, 1 ml of apatite or lipofectaminesuspensions prepared with 10–100 nM of luciferase siRNA (asdescribed before) were added to each well and incubated at 37 °Cunder a 5% CO2 atmosphere for 24 h. Luciferase gene expression wasmonitored by using a commercial kit (Promega) and photon counting(TD-20/20 Luminometer, USA). Each transfection experiment wasdone in triplicate and transfection efficiency was expressed as meanlight units per mg of cell protein.

ght scattering and (C) transmission electron microscopy. Apatite-siRNA particles wereon at 37 °C for 30 min and addition of FBS (10%). Scale bar, 500 nm.

104 S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

In case of GFP plasmid cotransfection, 2 μg of plasmid DNA and100 pM–100 nM of GFP siRNA were used to prepare the DNA/siRNAcomplexes of apatite and lipofectamine, that were subsequentlyadded onto the cells with either 10% FBS (for apatite complexes) or noFBS (for lipofectamine complexes) for 4 h incubation. The media wasreplaced by fresh DMEM with 10% FBS and incubated for 72 h.

In case of HeLa cells stably expressing GFP, the apatite/GFP siRNAor the lipofectamine/GFP siRNA particle suspensions produced in thesame way as described above, were added to the cells and incubatedfor 72 h. The cells were lysed using the lysis buffer (NP40) and theintracellular fluorescein intensity was determined using a microplatereader (DTX 880, Multimode Detector by BECKMAN COULTER). Incase of cyclin B1 siRNA delivery, apatite nanoparticles with cyclin B1siRNA were incubated with HeLa cells continuously for 48 h and thecellular toxicity level was assessed by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay.

2.11. Cell viability assessment with MTT assay

Following incubation of siRNA/apatite or siRNA/lipofectaminecomplexes either with stable GFP clone of HeLa for 72 h or incotransfection conditions for 4 h at first and after removal of thecomplexes, incubation of HeLa cells for another 72 h, 30 μl of MTTsolution (5 mg/ml) was added to each well and incubated for 4 h at37 °C. 0.5 ml of DMSO was added after removal of media. Afterresolving crystals and incubating for 5 min at 37 °C, absorbance wasmeasured in a micro plate reader at 570 nm with a referencewavelength of 630 nm. Cell viabilities were normalized to theabsorbance of non-treated cells. MTT assay was also performed in asimilar way after 48 h consecutive incubation of HeLa cells with cyclinB1 siRNA/apatite complexes.

3. Results and discussion

3.1. Assessment of binding affinity of siRNA to carbonate apatite

Since siRNA is negatively charged due to its phosphate backbonewhile the carbonate apatite particles are positively charged due to thepresence of calcium ions in the apatite structure, it was presumed thatsiRNAwould bind to the apatite by ionic interactions. To estimate howmuch of the initially added siRNAs were actually associated with theapatite, the particles prepared with different concentrations offluorescein siRNA (1 to 200 nM) were centrifuged and the resultingpellet was washed with the same medium and dissolved with 10 mMEDTA-PBS for quantification of the florescence intensity by a spec-trophotometer. The fluorescence intensity of free fluorescein siRNA(1 nM to 200 nM) in PBS was quantified and plotted to make thecalibration curve for subsequent quantitation of siRNA loading into theparticles. As shown in Fig. 1, the binding affinity of siRNA to the apatiteparticles is increased almost proportionately with increasing concen-

Fig. 3.Optical density of the suspension of carbonate apatite particles and siRNA/apatitecomplexes against pH value.

trations of the available siRNA (1 to 100 nM) in the medium. Amaximum binding of 50% was achieved when the initial siRNAconcentration was 100 nM. No further increase in the % of siRNAbinding was observed by increasing siRNA concentration to 200 nM,suggesting that the anion-binding sites (Ca2+-rich domains) aresaturated by complexation with siRNA molecules.

3.2. Size determination of apatite-siRNA complexes by dynamic lightscattering

The size of siRNA/apatite complex is important as it affects itscellular uptake through endocytosis. Endocytosis of a large complex isusually less efficient than that of a small one [47]. Therefore, the sizedistribution of siRNA/apatite complexes wasmeasured using dynamiclight scattering (DLS). The average diameter of the particles wasestimated at 274.6 nm. As shown in Fig. 2A and B, the relatively small-

Fig. 4. Cellular uptake and intracellular siRNA measurement in HeLa cell line (A) siRNA/apatite particles were prepared (as described in the legend to Fig. 2) in the presence offluorescein siRNA (100 nM) and added onto the HeLa cells in the presence of 10%serum-supplemented and serum deprived medium in case of siRNA/apatite and siRNA/lipofetamine 2000 respectively. The cells with the particles were incubated for 4 h forcellular internalization. Extracellular particles were removed by EDTA prior toobservation of the cells by a fluorescence microscope. Pictures in the first columnand those in the second column were taken under visible and fluorescent light,respectively. Scale bar, 200 μm. (B) HeLa cells were transfected in the same withfluorescein siRNA as described above and extracellular particles were removed by EDTAprior to cell lysis and detection of fluorescein intensity in a microplate reader. Controlwas the intensity for the delivery of free fluorescein siRNA to the cell (without anyparticle). Data represent mean value±SE (n=3).

105S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

size particles constitute themajority of the total particles and accountsfor the maximum amount of the whole particle weight. This suggeststhat the majority of the siRNA/apatite complexes would be able to beefficiently endocytosed by cells [47]. Themorphological observation ofsiRNA/apatite complexes by transmission electron microscopy (TEM)revealed that the particles are compact and spherical shapes withparticle sizes being less than 300 nm (Fig. 2C).

3.3. Dissolution study of apatite-siRNA complexes in acidic condition

Following cellular uptake of siRNA/apatite complexes, the releaseof the siRNA from the apatite carrier is essentially required for bindingwith specific mRNA in cytoplasm for the silencing purpose. Particlesthat can rapidly be dissolved in endosomal acidic pH should becapable of releasing associated siRNA following endocytosis. Fig. 3shows the optical density of siRNA/apatite complexes and apatiteparticles themselves as an indicator of particle existence with lowerturbidity indicating higher dissolution due to the existence of fewerno. of particles. These results indicate that the optical density ofsiRNA/apatite complexes and apatite themselves rapidly decreasedwith a decrease of pH, suggesting that the siRNA/apatite complexeswere almost completely dissolved in the condition below pH 7.0,which might contribute to the destabilization of endosomes followingendocytosis of siRNA/apatite complexes, resulting in the release ofsiRNA to the cytoplasm [45–47].

3.4. Cellular uptake and intracellular siRNA measurement in HeLa cellline

To investigate whether the siRNA/apatite complexes could effec-tively be internalized by cells, 100 nMof thefluorescein siRNAwas used

Fig. 5. Endosomal escape of fluorescein siRNAs following endocytosis by apatite particles.described in the legend to Fig. 2) for 1 to 4 h. After the delivery and removal of residual compwith LysoTracker™ Red DND-99 and nuclei were stained with DAPI. The cells were observe

to separately prepare siRNA/apatite and siRNA/lipofectamine com-plexes which were subsequently incubated with HeLa cells. Following4 h incubation and removal of the extracellular particles by EDTA,cellular fluorescence was observed by a fluorescence microscope. Forquantitative assay, cells were lysed after 4 h incubation and thefluorescence intensity of the cell lysate was estimated by a plate reader.As shown in Fig. 4A,while labelled siRNAuptake by the cellswas low forcommonly used lipofectamine, carbonate apatite particles enhancedcellular siRNA delivery to a significant extent probably due to the strongaffinity of siRNA to the apatite particles and the small sizes of theresulting complexes leading to the efficient endocytosis. The quantita-tive assay indicates that cellular uptake of siRNA is at least 3 timesmore efficient for carbonate apatite than for the liposome formulation(Fig. 4B).

3.5. Endosomal escapes of siRNA carried by apatite nanoparticles

Efficient delivery of siRNA to the cytosol of target cells depends onboth the translocation of the non-viral vectors through the plasmamembrane and their subsequent escape from endosomal/lysosomalcompartments [51,52]. Endocytosis, the vesicularuptake of extracellularmacromolecules, has been established as the main mechanism forthe internalization of non-viral vectors into the cells [53–55]. However,after endocytosis, the internalized molecules tend to be trapped inintracellular vesicles and eventually fusewith lysosomeswhere they aredegraded [56]. It is, therefore, important for a delivery system to avoidthe fate of lysosomal degradation by facilitating release of theinternalized siRNA into the cytoplasm. A confocal microscopic analysiswas performed on HeLa cells after the delivery of fluorescein siRNA/apatite complexes and observed the cellular localization of siRNA ina time-dependent manner from 1 to 4 h. Endosomes/lysosomes were

HeLa cells were incubated with the prepared fluorescein siRNA/apatite complexes (aslexes, the cells were fixed with formaldehyde. Endosomes and lysosomes were stainedd by confocal laser scanning microscope. Scale bar, 10 μm.

106 S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

stained with LysoTracker™ Red. As shown in Fig. 5, most of thefluorescein siRNAs were co-localized in stained endosomes and/orlysosomes after 1 h incubation of HeLa cells with siRNA/apatitecomplexes. However, interestingly after 2 h,most of the siRNAs escapedthe endocytic vesicles and following 4 h, only a few dots of fluoresceinsiRNAs were still associated with the stained endo-lysosomes suggest-ing that the endosomal escape of siRNAwas almost completed after 4 hincubation (Fig. 5). Following 4 h incubation, a homogeneous fluores-cence distribution was observed throughout the cytosol in many cellswhich suggested that most of the intracellularly delivered siRNAescaped endo-lysosomes (data not shown). As mentioned previouslyin the dissolution study of siRNA/apatite complexes (Fig. 3), the highdissolution rate of carbonate apatite particles might contribute tothedestabilization of endosomes releasingsiRNA to the cytoplasm, sincev-ATPase-driven excessive proton transfer into the endosomes todissolve the particles could cause passive chloride influx and createosmotic pressure across the endosomal membrane, leading to endo-some swelling and rupture [45–47].

3.6. Estimation of gene-silencing efficacy

The final effect of siRNA on silencing gene expression depends onthe integrity of the siRNA after release from the endosomes as well asfrom the particles. In other words, the gene-silencing efficiency actuallyreflects the effectiveness of a delivery system that carries and releasessiRNA into the cytosol. The siRNA delivery efficiency was first assessed

Fig. 6. Silencing of GFP and luciferase expression, respectively, by siGFP and anti-luciferaseHeLa cells (protocols have been provided in Materials and methods section). A) Floreslipofectamine 2000 for 72 h. B) Comparison of gene-silencing efficiency (%) of siGFP comestimated by quantitation of green fluorescence using a fluorescence spectrophotometer afteC) Gene-silencing efficiency (%) of anti-luciferase siRNA formulations after 48 h. The level ofusing a luminometer after lysis of the cells. Data represent mean value±SE (n=3).

by cotransfection of GFP gene and siRNA in HeLa cells. The anti-GFPsiRNAs were transfected at varied concentrations (from 100 pM to100 nM) alongwith GFP plasmids using apatite and silencing efficiencywas compared with lipofectamine 2000. At 100 pmol siRNA concentra-tion,more than 80% silencingwas achievedwith apatite particles, beinghigher than that of lipofectamine 2000. With increasing siRNA con-centration to 1 nM, the silencing efficiency for the apatite becamealmost 100% whereas to achieve the similar level of efficacy,lipofectamine required 10 time more concentration of the siRNA (i.e.,10 nM) (Fig. 6A and B).

We also investigated the silencing effects on transiently expressedreporter gene by sequentially transfecting HeLa cells with luciferasegene to express the luciferase mRNA (and protein) and then, withluciferase siRNA to cleave the expressed mRNA. With 10 and 100 nMof siRNA concentrations, almost 72% and 80% silencing were achieved,respectively, with carbonate apatite nanoparticles, that is highlycomparable with lipofectamine 2000 (Fig. 6C).

In HeLa cells stably expressing GFP, the knockdown efficiency ofapatite was much higher than lipofectamine in a variety of siRNAconcentrations. While at 100 pM siRNA concentration only 25% silencingof luciferase expression was achieved, at 10 nM over 50% silencing wasobservedwith carbonate apatite particles as shown in Fig. 7. These resultsclearly suggest that GFP and luciferase knockdown achieved usingcarbonate apatite was significantly higher than commercially availableLipofectamine 2000 due to the higher uptake of siRNA, internalizationthroughendocytosis and its subsequent escape fromtheendocytic vesicle.

siRNA when delivered at different concentrations by apatite and lipofectamine 2000 incence microscopic images of the cells treated with siGFP complexes of apatite andplexes of apatite and lipofectamine 2000 after 72 h. The level of GFP expression wasr lysis of the cells following 72 h of incubation. Data represent mean value±SE (n=3).luciferase expression was estimated by quantitation of photons (relative light unit, RLU)

Fig. 7. Dose dependent silencing efficiency of siGFP (delivered by apatite and lipofectamine2000, as described in Materials and methods section) in HeLa cells stably expressing GFP.

Fig. 9. Cell viability assessment following 48 h incubation of cyclin B1 siRNA/apatitecomplexes with HeLa cells. The preparation of cyclin B1 siRNA/apatite complexes wasdescribed in Materials and methods section. The data represent mean value±SE (n=3).

107S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

3.7. Cell viability assessment for siRNA complexes of apatite andlipofectamine

MTT colorimetric assay was employed to estimate the extent oftoxicity caused by the siRNA complexes of apatite and lipofectaminein both cotransfection and stable clone transfection conditions. ThesiRNA/apatite nanoparticles showed almost no toxicity even after72 h incubation with HeLa cells among which nearly 95% cells wereviable in both experimental conditions while only 52–60% cells wereviable following lipofectamine-mediated siRNA delivery, indicatingthat carbonate apatite is highly biocompatible whereas lipofectamineis extremely cytotoxic for siRNA delivery (Fig. 8).

3.8. Effect of cyclin B1 gene silencing on cell viability

In order to further evaluate the potential of carbonate apatite as avaluable intracellular carrier of siRNAs, we aimed to silence cyclin B1gene whose expression is essentially needed during the onset ofmitosis [57,58] and thus, knockdown of the gene might result inapoptosis of cancer cells. The cyclin B1siRNA–apatite complexes weremade using different concentrations of pre-validated siRNA (0.2 to50 nM) and used for transfection of HeLa cells for 48 h prior to theassessment of cell viability by MTT assay. As shown in Fig. 9, the mosteffective concentration of the siRNA for killing the cancer cells is10 nM, where 54% of the cells were killed, followed by 50 nM whichkilled 46%, 2 nM which killed 33% and 0.4 nM which killed 8% of thecells. The concentration of siRNAs plays a significant role in affectingthe silencing efficiency of a particular gene. At high concentrations,the silencing by siRNAsmight be less specific, leading to the undesiredsilencing of genes. On the other hand, when the siRNA concentrationis too low, the siRNAs might not be enough to silence sufficientamount of the genes. Therefore, when 50 nM is compared to 10 nM of

Fig. 8. Cell viability assessment for the siRNA complexes of apatite and lipofectamine.HeLa cells were transfected with siRNA/apatite or siRNA/lipofectamine 2000 formula-tions according to the protocol (for cotransfection and stable clone transfection) asdescribed in Materials and methods section and finally, cell viability was assessed byMTT cell proliferation assay after 72 h. Data represent mean value±SE (n=3).

the siRNAs, the amount of cells died with 50 nM was lesser than with10 nM. This might be due to the siRNA-mediated silencing of othergenes in addition to cyclin B1 gene, which might also prevent theapoptosis or promote cell survival. At 10 nM, the siRNAs might bemore specific, therefore killing more cells whereas at very lowconcentrations of the siRNAs (2 nM and 0.4 nM), the silencing rate ofcyclin B1 gene might be too low to affect the cell viability.

4. Conclusion

siRNA is highly specific for degradation of the correspondingmRNA in a process where a single strand of the siRNA, complementaryto the target mRNA, is incorporated into an RNA-induced silencingcomplex (RISC) that cleaves the target mRNA and the RISC is recycled[59]. Although RNA interference is highly promising for treatingdifferent diseases through the selective knockdown of the desirablegene, the success of RNAi is highly dependent on the delivery ofexogenous dsRNA or siRNA. In the present study, we have establishedthe intracellular delivery of siRNA using pH-sensitive carbonateapatite taking the advantages of endocytosis-mediated siRNA entryand its subsequent escape from the endocytic vesicle in a time-dependent manner which is a key issue for successful delivery. Ourhighly engineered nanoparticle showed excellent siRNA complexationwith a desirable nano size level suitable for endocytosis. The RNAi effectof carbonate apatite was higher and comparable to that of cytotoxicLipofectamine-2000. Furthermore, nanoparticle-facilitated delivery ofvalidated siRNA against cyclin B1 resulted in significant inhibition ofcancer cell growth. Thus as a powerful tool for intracellular siRNAdelivery, carbonate apatite may have pre-clinical and clinical potentialespecially for cancer treatment.

Acknowledgement

This work was supported by Grants-in-Aid for Scientific Researchfrom the MEXT, Ministry of Education, Science, Sports and Culture ofJapan.

References

[1] J. Winkler, P. Martin-Killias, A. Plückthun, U Zangemeister-Wittke, EpCAM-targeted delivery of nanocomplexed siRNA to tumor cells with designed ankyrinrepeat proteins, Mol. Cancer Ther. 8 (2009) 2674–2683.

[2] A.R. de Fougerolles, Delivery vehicles for small interfering RNA in vivo, Hum. GeneTher. 19 (2008) 125–132.

[3] K. Gao, L. Huang, Nonviralmethods for siRNAdelivery,Mol. Pharm. 6 (2009) 651–658.[4] N.S. Lee, J.J. Rossi, Control of HIV-1 replication by RNA interference, Virus Res. 102

(2004) 53–58.[5] A. Grünweller, R.K. Hartmann, RNA interference as a gene-specific approach for

molecular medicine, Curr. Med. Chem. 12 (2005) 3143–3161.[6] M. Ito, K. Kawano, M. Miyagishi, K. Taira, Genome-wide application of RNAi to the

discovery of potential drug targets, FEBS Lett. 579 (2005) 5988–5995.[7] P.Y. Lu, F. Xie, M.C. Woodle, In vivo application of RNA interference: from

functional genomics to therapeutics, Adv. Genet. 54 (2005) 117–142.[8] E.H. Chowdhury, Nuclear targeting of viral and non-viral DNA, Expert Opin. Drug

Deliv. 6 (2009) 697–703.[9] D. Reischl, A. Zimmer, Drug delivery of siRNA therapeutics: potentials and limits of

nanosystems, Nanomedicine 5 (2009) 8–20.

108 S. Hossain et al. / Journal of Controlled Release 147 (2010) 101–108

GENEDELIVERY

[10] M. Amarzguioui, J.J. Rossi, D. Kim, Approaches for chemically synthesized siRNAand vector-mediated RNAi, FEBS Lett. 579 (2005) 5974–5981.

[11] M. Scherr, M. Eder, Gene silencing by small regulatory RNAs in mammalian cells,Cell Cycle 6 (2007) 444–449.

[12] K. Buyens, M. Meyer, E. Wagner, J. Demeester, S.C. De Smedt, N.N. Sanders,Monitoring the disassembly of siRNA polyplexes in serum is crucial for predictingtheir biological efficacy, J. Control. Release 141 (2009) 38–41.

[13] W. Dobbs, B. Heinrich, C. Bourgogne, B. Donnio, E. Terazzi, M.E. Bonnet, F. Stock, P.Erbacher, A.L. Bolcato-Bellemin, L. Douce, Mesomorphic Imidazolium salts: newvectors for efficient siRNA transfection, J. Am. Chem. Soc. 131 (2009) 13338–13346.

[14] R.K. Delong, U. Akhtar, M. Sallee, B. Parker, S. Barber, J. Zhang, M. Craig, R. Garrad, A.J.Hickey, E. Engstrom, Characterization and performance of nucleic acid nanoparticlescombined with protamine and gold, Biomaterials 30 (2009) 6451–6459.

[15] C. Liu, G. Zhao, J. Liu, N. Ma, P. Chivukula, L. Perelman, K. Okada, Z. Chen, D. Gough,L. Yu, Novel biodegradable lipid nano complex for siRNA delivery significantlyimproving the chemosensitivity of human colon cancer stem cells to paclitaxel, J.Control. Release 140 (2009) 277–283.

[16] H.A. Cho, I.S. Park, T.W. Kim, Y.K. Oh, K.S. Yang, J.S. Kim, Suppression of hepatitis Bvirus-derived human hepatocellular carcinoma by NF-kappaB-inducing kinase-specific siRNA using liver-targeting liposomes, Arch. Pharm. Res. 32 (2009)1077–1086.

[17] O.M. Merkel, A. Beyerle, D. Librizzi, A. Pfestroff, T.M. Behr, B. Sproat, P.J. Barth, T.Kissel, Nonviral siRNA delivery to the lung: investigation of PEG-PEI polyplexesand their in vivo performance, Mol. Pharm. 6 (2009) 1246–1260.

[18] J.A. MacDiarmid, N.B. Amaro-Mugridge, J. Madrid-Weiss, I. Sedliarou, S. Wetzel, K.Kochar, V.N. Brahmbhatt, L. Phillips, S.T. Pattison, C. Petti, B. Stillman, R.M.Graham, H. Brahmbhatt, Sequential treatment of drug-resistant tumors withtargeted minicells containing siRNA or a cytotoxic drug, Nat. Biotechnol. 27(2009) 643–651.

[19] L.S. Mangala, H.D. Han, G. Lopez-Berestein, A.K. Sood, Liposomal siRNA for ovariancancer, Meth. Mol. Biol. 555 (2009) 29–42.

[20] D. Jere, H.L. Jiang, Y.K. Kim, R. Arote, Y.J. Choi, C.H. Yun, M.H. Cho, C.S. Cho,Chitosan-graft-polyethylenimine for Akt1 siRNA delivery to lung cancer cells, Int.J. Pharm. 378 (2009) 194–200.

[21] L. Crombez, M.C. Morris, S. Dufort, G. Aldrian-Herrada, Q. Nguyen, G. Mc Master,J.L. Coll, F. Heitz, G. Divita, Targeting cyclin B1 through peptide-based delivery ofsiRNA prevents tumour growth, Nucleic Acids Res. 37 (2009) 4559–4569.

[22] A. Difeo, F. Huang, J. Sangodkar, E.A. Terzo, D. Leake, G. Narla, J.A. Martignetti,KLF6-SV1 is a novel antiapoptotic protein that targets the BH3-only protein NOXAfor degradation and whose inhibition extends survival in an ovarian cancermodel, Cancer Res. 69 (2009) 4733–4741.

[23] B.C. Ponnappa, siRNA for inflammatory diseases, Curr. Opin. Investig. Drugs 10 (2009)418–424.

[24] Y.K. Oh, T.G. Park, siRNA delivery systems for cancer treatment, Adv. Drug Deliv.Rev. 61 (2009) 850–862.

[25] P. Mu, S. Nagahara, N. Makita, Y. Tarumi, K. Kadomatsu, Y. Takei, Systemic deliveryof siRNA specific to tumor mediated by atelocollagen: combined therapy usingsiRNA targeting Bcl-xL and cisplatin against prostate cancer, Int. J. Cancer 125(2009) 2978–2990.

[26] A.M. Al-Abd, S.H. Lee, S.H. Kim, J.H. Cha, T.G. Park, S.J. Lee, H.J. Kuh, et al.,Penetration and efficacy of VEGF siRNA using polyelectrolyte complex micelles ina human solid tumor model in-vitro, J. Control. Release 137 (2009) 130–135.

[27] M. Koldehoff, A.H. Elmaagacli, Therapeutic targeting of gene expression by siRNAsdirected against BCR-ABL transcripts in a patient with imatinib-resistant chronicmyeloid leukaemia, Meth. Mol. Biol. 487 (2009) 451–466.

[28] X.L. Wang, R. Xu, X. Wu, D. Gillespie, R. Jensen, Z.R. Lu, Targeted systemic deliveryof a therapeutic siRNA with a multifunctional carrier controls tumor proliferationin mice, Mol. Pharm. 6 (2009) 738–746.

[29] M.E. Davis, The first targeted delivery of siRNA in humans via a self-assembling,cyclodextrin polymer-basednanoparticle: from concept to clinic,Mol. Pharm. 6 (2009)659–668.

[30] Y.C. Tseng, S. Mozumdar, L. Huang, Lipid-based systemic delivery of siRNA, Adv.Drug Deliv. Rev. 61 (2009) 721–731.

[31] M. Grzelinski, B. Urban-Klein, T. Martens, K. Lamszus, U. Bakowsky, S. Höbel, F.Czubayko, A. Aigner, RNA interference-mediated gene silencing of pleiotrophinthrough polyethylenimine-complexed small interfering RNAs in vivo exertsantitumoral effects in glioblastoma xenografts, Hum. Gene Ther. 17 (2006)751–766.

[32] J.F. Engelhardt, R.H. Simon, Y. Yang, M. Zepeda, S. Weber-Pendleton, B. Doranz, M.Grossman, J.M. Wilson, Adenovirus-mediated transfer of the CFTR gene to lung ofnonhumanprimates: biological efficacy study, Hum. Gene Ther. 4 (1993) 759–769.

[33] S. Lehrman, Virus treatment questioned after gene therapy death, Nature 401(1999) 517–518.

[34] T.S. Zimmermann, A.C. Lee, A. Akinc, B. Bramlage, D. Bumcrot, M.N. Fedoruk, J.Harborth, J.A. Heyes, L.B. Jeffs, M. John, A.D. Judge, K. Lam, K. McClintock, L.V.Nechev, L.R. Palmer, T. Racie, I. Röhl, S. Seiffert, S. Shanmugam, V. Sood, J.Soutschek, I. Toudjarska, A.J. Wheat, E. Yaworski, W. Zedalis, V. Koteliansky, M.Manoharan, H.P. Vornlocher, I. MacLachlan, RNAi-mediated gene silencing in non-human primates, Nature 441 (2006) 111–114.

[35] E. Fattal, G. Barratt, Nanotechnologies and controlled release systems for thedelivery of antisense oligonucleotides and small interfering RNA, Br. J. Pharmacol.157 (2009) 179–194.

[36] W.J. Kim, S.W. Kim, Efficient siRNA delivery with non-viral polymeric vehicles,Pharm. Res. 26 (2009) 657–666.

[37] S. Akhtar,M.D. Hughes, A. Khan,M. Bibby,M. Hussain, Q. Nawaz, J. Double, P. Sayyed,The delivery of antisense therapeutics, Adv. Drug Deliv. Rev. 44 (2000) 3–21.

[38] Y. Nakamura, K. Kogure, S. Futaki, H. Harashima, Octaarginine-modified multi-functional envelope-type nano device for siRNA, J. Control. Release 119 (3) (2007)360–367.

[39] D. Luo,W.M. Saltzman, Synthetic DNAdelivery systems, Nat. Biotechnol. 18 (2000)33–37.

[40] S. Ganta, H. Devalapally, A. Shahiwala, M. Amiji, A review of stimuli-responsivenanocarriers for drug and gene delivery, J. Control Release 126 (2008) 187–204.

[41] S. Hossain, S. Tada, T. Akaike, E.H. Chowdhury, Influences of electrolytes andglucose on formulation of carbonate apatite nanocrystals for efficient genedelivery to mammalian cells, Anal. Biochem. 397 (2010) 156–161.

[42] E.H. Chowdhury, A. Maruyama, A. Kano, M. Nagaoka, M. Kotaka, S. Hirose, M.Kunou, T. Akaike, pH-sensing nano-crystals of carbonate apatite: effects onintracellular delivery and release of DNA for efficient expression into mammaliancells, Gene 376 (2006) 87–94.

[43] K. Kutsuzawa, T. Akaike, E.H. Chowdhury, The influence of the cell adhesiveproteins E-cadherin and fibronectin embedded in carbonate-apatite DNA carrieron transgene delivery and expression in a mouse embryonic stem cell line,Biomaterials 29 (2008) 370–376.

[44] S. Tada, E.H. Chowdhury, C.S. Cho, T. Akaike, pH-sensitive carbonate apatite as anintracellular protein transporter, Biomaterials 31 (2010) 1453–1459.

[45] E.H. Chowdhury, pH-sensitive nano-crystals of carbonate apatite for smart andcell-specific transgene delivery, Expert Opin. Drug Deliv. 4 (2007) 193–196.

[46] E.H. Chowdhury, T. Akaike, pH-sensitive inorganic nano-particles and theirprecise cell targetibility: an efficient gene delivery and expression system, Curr.Chem. Biol. 1 (2007) 201–213.

[47] E.H. Chowdhury, T. Akaike, High performance DNA nano-carriers of carbonateapatite: multiple factors in regulation of particle synthesis and transfectionefficiency, Int. J. Nanomedicine 2 (2007) 101–106.

[48] E.H. Chowdhury, Self-assembly of DNA and cell-adhesive proteins onto pH-sensitive inorganic crystals for precise and efficient transgene delivery, Curr.Pharm. Des. 14 (2008) 2212–2228.

[49] E.H. Chowdhury, T. Akaike, Bio-functional inorganic materials: an attractivebranch of gene-based nano-medicine delivery for 21st century, Curr. Gene Ther. 5(2005) 669–676.

[50] E.H. Chowdhury, T. Akaike, A bio-recognition device developed onto nano-crystalsof carbonate apatite for cell-targeted gene delivery, Biotechnol. Bioeng. 90 (2005)414–421.

[51] H. Kamiya, H. Akita, H. Harashima, Pharmacokinetic/pharmacodynamic consid-erations in gene therapy, Drug Discov. Today 8 (2003) 990–996.

[52] I.A. Khalil, K. Kogure, H. Akita, H. Harashima, Uptake pathways and subsequentintracellular trafficking in nonviral gene delivery, Pharmacol. Rev. 58 (2006) 32–45.

[53] D.S. Friend, D. Papahadjopoulos, R.J. Debs, Endocytosis and intracellular processingaccompanying transfection mediated by cationic liposomes, Biochim. Biophys.Acta 1278 (1996) 41–50.

[54] F. Labat-Moleur, A.M. Steffan, C. Brisson, H. Perron, O. Feugeas, P. Furstenberger, F.Oberling, E. Brambilla, J.P. Behr, An electronmicroscopy study into themechanismof gene transfer with lipopolyamines, Gene Ther. 3 (1996) 1010–1017.

[55] I.S. Zuhorn, R. Kalicharan, D. Hoekstra, Lipoplex-mediated transfection ofmammalian cells occurs through the cholesterol-dependent clathrin-mediatedpathway of endocytosis, J. Biol. Chem. 277 (2002) 18021–18028.

[56] M.B. Bally, P. Harvie, F.M. Wong, S. Kong, E.K. Wasan, D.L. Reimer, Biological barriersto cellular delivery of lipid-based DNA carriers, Adv. Drug Deliv. Rev. 38 (1999)291–315.

[57] L.A. Porter, D.J. Donoghue, Cyclin B1 and CDK1: nuclear localization and upstreamregulators, Prog. Cell Cycle Res. 5 (2003) 335–347.

[58] T. Miyazaki, S. Arai, Two distinct controls of mitotic cdk1/cyclin B1 activityrequisite for cell growth prior to cell division, Cell Cycle 6 (2007) 1419–1425.

[59] L.J. Scherer, J.J. Rossi, Approaches for the sequence-specific knockdown of mRNA,Nat. Biotechnol. 21 (2003) 1457–1465.