Journal of Controlled Release 128 (2008) 255–260

NANOTECHNOLOGY

Contents lists available at ScienceDirect

Journal of Controlled Release

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

Stable nanocolloids of poorly soluble drugs with high drug content prepared usingthe combination of sonication and layer-by-layer technology

Anshul Agarwal a, Yuri Lvov a, Rishikesh Sawant b,c, Vladimir Torchilin b,c,⁎a Institute for Micromanufacturing, Louisiana Tech University, Ruston, LA 71272, United Statesb Department of Pharmaceutical Sciences, Northeastern University, Boston, MA 02115, United Statesc Center for Pharmaceutical Biotechnology and Nanomedicine, Northeastern University, Boston, MA 02115, United States

⁎ Corresponding author. Department of PharmaceUniversity, 360 Huntington Ave, Mugar Building RoomStates. Tel.: +1 617 373 3206; fax: +1 627 373 8886.

E-mail address: v.torchilin@neu.edu (V. Torchilin).

0168-3659/$ – see front matter © 2008 Published by Edoi:10.1016/j.jconrel.2008.03.017

A B S T R A C T

A R T I C L E I N F O

Article history:

Stable nanocolloids of inso Received 27 November 2007Accepted 11 March 2008Available online 26 March 2008

Keywords:Poorly soluble drugsLayer-by-layer self-assemblyColloidsNanoparticles

luble drugs with very high drug content (up to 90% wt) can be easily andreproducibly prepared through the application of the layer-by-layer (LbL) technology, alternate adsorption ofoppositely charged polyelectrolytes on the surface of drug nanoparticles produced by ultrasonication oflarger drug crystals. Such polymeric coating prevents drug nanoparticle aggregation and creates a firmpolymeric shell on their surface. Drug release rate from such nanocolloidal particles can be easily controlledby assembling multilayer shells with variable shell density and thickness. Various additional functions, suchas specific targeting ligands, can be easily attached to the surface on nanocolloidal particles of poorly solubledrugs by using a polymer with free reactive groups for the “outer” coating. This may represent a novelapproach to preparing convenient dosage forms of poorly soluble drugs.

© 2008 Published by Elsevier B.V.

1. Introduction

The desired features of pharmaceutical drug delivery systems (DDS)for parenteral (intravenous) administration include their small size,biodegradability, high content of a drug in afinal preparation, prolongedcirculation in the blood, and, ideally, the ability to target required areaspassively (via the enhanced permeability and retention— EPR— effect)or actively (via specific ligands, such as monoclonal antibodies) [1].While these requirements are reasonably well met by a variety of DDS(liposomes, microcapsules, nanoparticles) for water-soluble drugs [2,3],the development of nanoparticulate DDS displaying all of theseproperties for poorly soluble pharmaceuticals still represents achallenge.

Low solubility in water, however, tends to be an intrinsic propertyof many drugs, including powerful anti-cancer agents [4]. This is notsurprising, since the membrane permeability and efficacy of variousdrugs increase with increasing hydrophobicity [4–6]. At the sametime, intravenous administration of those intrinsically hydrophobicagents could be associated with serious safety problems. One of themis that the intravenous administration of relatively large aggregates/crystals of insoluble drug that form in an aqueous media mayembolize blood capillaries. Additionally, the low solubility of hydro-

utical Sciences, Northeastern312, Boston, MA 02115, United

lsevier B.V.

phobic drugs in combination with excretion and metabolic degrada-tion often does not allow for achieving therapeutically significantsystemic concentrations. As a result, many promising drug candidatesdo not enter further development because of solubility problems [7,8].

Currently, the most popular approach to solubilize poorly solubledrugs and prepare their dosage forms with sufficiently high bioavail-ability is the use of micellar drug carriers, primarily, various polymericmicelles [9,10]. Micelles are colloidal particles, usually spherical, withthe size in a nanometer range, into which many amphiphilicmolecules self-assemble spontaneously. In an aqueous environment,hydrophobic fragments of amphiphilic molecules form the core of amicelle, which is segregated from the environment by hydrophilicparts of the molecules forming the micelle corona. It was clearlyshown on multiple occasions that micelles can serve as drug deliverysystems for poorly soluble pharmaceuticals [11–13]. The hydrophobiccore of micelles may be used as a cargo space for encapsulation of avariety of sparingly soluble therapeutic and diagnostic agents. Suchencapsulation substantially increases the bioavailability of pharma-ceuticals, protects them from destructive factors upon parenteraladministration, and beneficially modifies their pharmacokinetics andbiodistribution [11–15] including the target accumulation via the EPReffect [16–18].

However, there exist a set of serious problems associated withmicellar carriers, which include low loading efficacy of the drug into themicelles (usually well below 5% by wt); impossibility to use the sameprotocol for making solubilized forms of different drugs since each drugrequires its own specific conditions for solubilization; problems with

NANOTECHNOLOGY

256 A. Agarwal et al. / Journal of Controlled Release 128 (2008) 255–260

controlling the release rate of the drug from micelles; difficulties withthe scaling up of the technology; insufficient storage stability ofmicellesand their frequent instability in the body, to name just a few.

On the other hand, there exists an interesting approach toassemble polyelectrolyte multilayer shells on drug particles withfew nanometer wall thicknesses through a layer-by-layer assembly[19–21]. A layer-by-layer nanoassembly technique (LbL) is based on analternate adsorption of oppositely charged polyelectrolytes ontocertain surfaces [22–25]. These polycation/polyanion multilayerscould be built with a required composition. Among many otherapplications [23], the formation of LbL polyelectrolyte capsules wasapplied to prepare sustained release (up to several hours) formula-tions of some water-soluble drugs, such as ibuprofen, furosemide,nifedipine, and insulin, which under normal circumstances comple-tely dissolve in few minutes [26,27].

Here, we suggest a novel application for the LbL coating technologytomake stable aqueous nanocolloids of poorly soluble drugs with very

Fig. 1. A — A principal scheme of LbL nanocolloidal particles formation from insoluble drupaclitaxel nanoparticles via free amino groups of the “outer” polymeric layer.

high content of the active drug and controllable drug release rate. Toachieve this, aqueous suspensions of poorly soluble drugs withmicronrange particles are subjected to the ultrasonic treatment in order todecrease the size of individual drug particles to the nano level(between 100 and 200 nm), and while still keeping the nanoparticlesformed under the sonication to prevent their fast agglomeration,stabilize them in solution by applying the LbL coating (alternatingaddition of polycations and polyanions to the system) and assemblingthin polyelectrolyte shells on their surface (Fig. 1A). In the process ofthe assembly, the highly charged polymeric layer is formed on thedrug particle surface after the first polymer application, and this layerprevents drug particle aggregation after terminating the sonication. Atthe end of the process, stable coated nanocolloidal drug dispersionsare formed with high drug content in each particle (more than 50% bywt and up to 90% by wt).

LbL assembly can utilize a broad variety of polyiones, includingmany biocompatible ones. Polycations can be represented by poly

gs. B — Schematics of the conjugation of an antibody (or any other ligand) to the LbL-

Fig. 2. The dependence of drug particle size on the duration of sonication. Experimentwith tamoxifen and paclitaxel nanocrystals (2 mg/mL). PAH was present in solution tofor the first polymeric monolayer on the particle surface and prevent particleaggregation after the removal of ultrasound.

NANOTECHNOLOGY

257A. Agarwal et al. / Journal of Controlled Release 128 (2008) 255–260

(ethyleneimine) (PEI), poly(allylamine) (PAA), polylysine, chitosan,gelatin B, amino-dextran, and protamine sulfate; polyanions — by poly(styrene sulfonate) (PSS), poly(acrylic acid), dextran sulfate, carbox-ymethyl cellulose, sodium alginate, hyaluronic acid, gelatin A, chon-droitin, and heparin. Some of these polymers are considered GRAS andused as pharmaceutical excepients in parenteral dosage forms. After, forexample, the first polycation layer is deposited on the surface of drugnanoparticle, it is followed by the addition of an oppositely chargedpolyanion and results in formation of a stable interpolyelectrolytecomplex shell (3–30 nm) around each drug nanoparticle. This polyelec-trolyte multilayer shell can be easily and reproducibly formed on thesurface of any drug particle. By varying the charge density on eachpolymer or the number of coating cycles, particles with a differentsurface charge and different composition of the polymeric coat can beprepared. This provides away to control drug release from suchparticlesby designing the shell architecture at nanometer level. The use of apolymer containing reactive groups (such as amino or carboxylicgroups) for the last “outer” surface layer will allow for the attachmentof specific ligands, or reporter groups, and other moieties of interest todrug nanoparticles (see example in Fig. 1B).

We present here the results of our studies using the LbL assemblyfor the preparation of stable nanocolloids of poorly soluble drugs withvery high drug content. By nanoencapsulation of such anti-cancerdrugs as tamoxifen and paclitaxel, we demonstrate the generalapplicability of this approach to prepare non-targeted and targeteddrug nanoparticles with very high drug content and controlled drugrelease rate.

2. Materials and methods

2.1. Materials and instruments

Poorly soluble and potent anti-cancer drugs tamoxifen (TMF) andpaclitaxel (PCT) have been used for our experiments (solubility below1 μg/mL). Polyelectrolyte used for the LbL assembly included positivelycharged poly(allylamine hydrochloride) (PAH), FITC-labeled PAH, andpoly(dimethyldiallylamide ammonium chloride) (PDDA); all at theconcentration of 2 mg/mL; and negatively charged sodium poly(styrene sulphonate) (PSS); also at the concentration of 2 mg/mL.Deionized water and PBS, pH 7.4, were used as the solvents. UltraSonicator 3000 (Misonix Inc, Farmingdale NY) was used for drugcrystals disintegration at the power of 3–18 W and sonication time of10–30 min. To prevent sample overheating during the sonication andkeeping the temperature in the range of 20–30 °C, the cooling ofsample tubes with liquid nitrogen was used. Quartz Crystal Micro-balance (9 MHz QCM, USI-System, Japan) was used to measure thethickness of the polyelectrolyte multilayer.

Surface potential (zeta-potential) and particle size measurements(by light scattering) were performed using ZetaPlus Microelectro-phoresis (Brookhaven Instruments).

Field Emission Scanning Electron Microscope (Hitachi, 2006) wasused for particle imaging. The SEM images were obtained by washingthe samples with deionized water to get rid of excess polyelectrolyteand after the overnight drying, tiny drops of approx. 1 µL of thesampleswere applied on bare siliconewafers at room temperature. Nocontrasting by coating with gold or other means under vacuum wasused prior to the imaging. The silicon wafer with the samples wasmounted on a metal stub with an adhesive.

Laser Scanning Confocal Microscope, Leica TCS SP2 (Leica Micro-systems Inc) was also used to control the shell formation and to followthe colloid stability.

2.2. LbL assembly and some properties of nanocolloids

Initially, for disintegration, all drug samples were ultra sonicatedwith cooling at 18 W for up to 30 min in 1 mL volume before any

polyelectrolyte was added. The size of drug particles formed wasperiodically measured. Prior to the addition of the first layer ofpolyelectrolyte, the zeta-potential reading was also taken. Polycationswere used to form the first surface layer, since drug nanoparticles ofboth drugs were found to bear the intrinsic negative charge. Drugsamples were then centrifuged at 14,000 rpm for 7 min, washed andre-suspended in water or PBS to remove the excess polyelectrolytebefore further zeta-potential readings were taken. Then, the coatingprocess was repeatedwithout the ultrasound applied andwith the useof the oppositely charged polymer (polyanion, in this case). Zeta-potential measurements were taken after each subsequent layeraddition. The images of the colloidal particles formed were takenimmediately and 48 h after the LbL assembly was completed to checkfor the stability of the colloids formed. Dry samples were prepared forSEM imaging using 5–10 µL of the colloidal suspension obtained.Sample droplets on bare silicon wafers were dried by heating them at50 °C for 1 h or, by keeping overnight at room temperature. Drugcolloids were kept in a low volume of saturated solution to preventdrug release.

2.3. Drug release from colloidal particles at sink conditions

To study the release rate of different drugs from the colloidalparticles prepared using the LbL assembly, the samples prepared usinga different number of coating cycles were placed in 1 mL horizontaldiffusion chambersmade of cellulose acetate membrane and stirred inlarge volume of PBS, pH 7.2, to mimic sink conditions expected in vivo.To mimic the biological surrounding, 3% solution of bovine serumalbumin (BSA) was also used as a release medium. The concentrationof the released drugs was measured by the HPLC described in detailbelow.

2.4. Attachment of ligand moieties to the LbL nanocolloids of poorlysoluble drugs

To prepare nanocolloids with the “reactive” surface suitable for thecovalent attachment of various ligands to their surface, PAH contain-ing free amino groups was used to form the outer layer on drugparticles. Paclitaxel was used in this series of experiments. Toconjugate the monoclonal nucleosome-specific 2C5 antibody (mAb2C5) recognizing a broad variety of cancer cells via the cancer cellsurface-bound nucleosomes [28,29], the reaction was carried out in 2steps (Fig. 2). In the first step, the carboxylate groups on the mAb 2C5

Fig. 3. Changes on drug particle zeta-potential in the process of the LbL assembly.(A) LbL assembly of tamoxifen (2 mg/mL) resulting in the product coatedwith 3 bilayerscomposed of PDDA and PSS. (B) LbL assembly of paclitaxel (4 mg/mL) resulting in theproduct coated with two bilayers composed of PAH and PSS.

NANOTECHNOLOGY

258 A. Agarwal et al. / Journal of Controlled Release 128 (2008) 255–260

were activated using 1-ethyl-3-carbodiimide hydrochloride (EDC) andN-hydroxysulfosuccinimide (sulfo-NHS) to make it amine-reactive. Inthe second step, the activated antibody was added to LbL-paclitaxelnanoparticles coated with polyamino-containing PAH polymer. Allreactions were carried out in HBS, pH 7.4 at 4 °C with continuousstirring in presence of argon gas. The modified particles werecentrifuged at 12 rpm for 10 min and re-suspended twice using PBSto remove the unconjugated antibody.

The amount of paclitaxel in the nanoparticle preparations wasmeasured by the reversed phase-HPLC. The particles were dissolvedwith themobile phase prior to applying onto theHPLC column.D-7000HPLC system equipped with a diode array (Hitachi, Japan) andSpherisorb ODS2 column, 4.6 mm×250 mm (Waters, Milford, MA,USA) was used. The column was eluted with acetonitrile/water(65:35%, v/v) at 1.0 mL/min. Paclitaxel peak was detected at 227 nm.Injection volume was 50 µL; all samples were analyzed in triplicate.

2.5. Antibody activity preservation on the surface of LbL drug nanocolloid

To verify the preservation of mAb 2C5 specific activity after theconjugation with LbL-paclitaxel nanoparticles, a standard ELISA wasperformed. Briefly, ELISA plates pretreated with 40 µg/mL polylysinesolution in TBS, pH 7.4, were coated with 50 µL of 40 µg/mLnucleosomes (thewater-soluble fraction of calf thymus nucleohistone,Worthington Biochemical, Lakewood, NJ) and incubated for 1 h atroom temperature. The plates were rinsed with 0.2% casein, 0.05%Tween 20 in TBS (casein/TBS), pH 7.4. To these plates, serial dilutionsof mAb 2C5-containig samples were added and incubated for 1 h atroom temperature. The plates were extensively washed with casein/TBS and coated with horseradish peroxidase goat antimouse IgGconjugate (ICN Biomedical, Aurora, OH) diluted according to themanufacturer's recommendation. After 1 h incubation at roomtemperature, the plates were washed with casein/TBS. Boundperoxidase was quantified by the degradation of its substrate,diaminobenzidine supplied as a ready-for-use solution, Enhanced K-Blue TMB substrate (Neogen, Lexington, KY). The intensity of the colordeveloped was analyzed using an ELISA reader at the wavelength of492 nm, Labsystems Multiscan MCC/340 (Labsystems and LifeSciences International, UK).

2.6. Cytotoxicity of targeted paclitaxel LbL nanocolloid

The cytotoxicity of various concentrations of LbL-paclitaxelnanoparticles against and MCF-7 and BT-20 cells was studied usingaMTT test. A ready-for-use CellTiter 96® Aqueous One solution of MTT(Promega, Madison, WI) was used according to a protocol suggestedby themanufacturer. Formulations with paclitaxel concentration of upto 200 ng/mL dispersed in Hanks' buffer were added to cells grown in96-well plates to about 40% confluence. After 48 h or 72 h ofincubation at 37 °C, 5% CO2, plates were washed three times withHanks' buffer followed by the addition of 100 µL of media and 20 µL ofCellTiter 96® Aqueous One solution. After 1 h of incubation at 37 °C, 5%CO2, the cell survival rate was estimated by measuring the colorintensity of the MTT degradation product at 492 nm using an ELISAplate reader. Untreated cells were considered as 100% growth.

3. Results and discussion

3.1. LbL-stabilized drug nanocolloids and surface zeta-potential

To find optimal sonication condition, we have performed our initialexperiments with tamoxifen crystals at drug concentration in thesuspension of 2 mg/mL. Fig. 2 clearly shows that particle size asexpected strongly depends on the sonication time. After 30 min ofsonication at 18 W, particles with the size of about 100 nm wereobtained (with the addition of polycationic PDDA prior to the size

measurement to prevent particle re-aggregation). When similarsonication conditions were applied to paclitaxel, particle sizes ofabout 100 nm were also obtained. Further increase in the sonicationtime did not result in significant decrease in drug particle size.

Fig. 3A gives values of the zeta-potential during the process ofsequential polycation/polyanion adsorption on tamoxifen cores. Afterthe addition of the PDDA polycation to initially negatively chargedtamoxifen particles under sonication, drug nanoparticles wererecharged to the positive potential of ca. +45 mV and formed stablecolloidal solution when the sonication is terminated. Then, weproceeded with the LbL assembly on tamoxifen nanoparticles throughthe PSS polyanion adsorption adding onemore monolayer to the shell,and again reversed the surface potential to the negative value of−17 mV. By adding the PDDA polycation again, we again madetamoxifen particles positively charged at ca. +80 mV. The fourthpolymer layer (negative PSS this time) made tamoxifen particlesnegative again, and so on. At the end, we obtained tamoxifennanoparticles coated with the organized multilayer shell with thecomposition (PDDA/PSS)3.

Fig. 3B demonstrates similar changes in zeta-potential valuesduring the LbL shell assembly process for paclitaxel. Again, we haveinitially negatively charged bare drug nanoparticles, which after thesonication with the addition of another polycation (PAH) becamepositive and formed stable colloids. Further assembly with thecorresponding changes in zeta-potential values allowed for theformation of the final shell with (PAH/PSS)2 composition.

In separate experiments with quartz Crystal Microbalance (QCM)monitoring of the PDDA/PSS or PAH/PSS assembly on quartz resonator,we found that a single polycation/polyanion bilayer has a thickness of1.5 nm in dry state. Polyelectrolyte multilayer thickness doubles in

Fig. 4. A— SEM images: (a–b) Tamoxifen nanoparticles of ca. 120 nm after 30 min sonication in the presence of 2 mg/mL PAH. (c) Paclitaxel nanoparticles after 30 min sonication andassembly of the LbL coating with (PAH-PSS)2 composition. B — Confocal fluorescent image of an aqueous dispersion of LbL tamoxifen nanocolloid coated with FITC-labeled PAH.

NANOTECHNOLOGY

259A. Agarwal et al. / Journal of Controlled Release 128 (2008) 255–260

water [22,23]. Therefore, one could estimate our (PDDA/PSS)3 shellthickness as ca. 4.5 nm in dry state and 9 nm in an aqueous solution,and (PAH/PSS)2 shell thickness as ca. 3 nm in dry state and 6 nm inwater.

3.2. Nanoparticle imaging and some properties

The particle size of all the drug samples formulated by the LbLtechnology into nanocolloidal state was confirmed by scanningelectron microscopy and confocal fluorescent microscopy. Fig. 4Ashows SEM images of prepared colloidal drug particles. Particles oftamoxifen demonstrate mainly spherical shape and have a diameter of120±30 nm. LbL-coated paclitaxel has an elongated rod-like shapewith approximate measurements of 50×120 nm. Although thereasons for the differences in the final shape of tamoxifen andpaclitaxel nanoparticles are not immediately evident, onemay assumethat they are associated with different patterns of melting andsolidification of two drugs in the field of powerful sonication.

The SEM images were obtained after drying the samples, andduring the drying process the nanoparticles become partiallyaggregated as can be seen in the Fig. 4Aa. To prove that thisaggregation is the result of SEM sample preparation and does notproceed in aqueous suspension, we made fluorescence confocalimages of our samples. With this in mind, we have used FITC-labeledPAH to prepare LbL-coated drug particles following same protocol asfor the samples shown on Fig. 4Aa,b. Fig. 4B, showing the fluorescenceimage of LbL-coated tamoxifen particles in suspension does not revealany aggregation. One can see multiple individual fluorescent greendots (the color is due to FITC labeling), although the confocalmicroscope resolution of ca. 100 nm does not allow for visualizationthe detailed structure of individual nanocapsules; and particle size

Fig. 5. Controlled drug release from the LbL nanocolloidal particles. Dissolution rate offree tamoxifen (as drug crystals without sonication — 1, and nanoparticles of sonicatednon-coated drug — 2) and tamoxifen release form the ca. 125 nm LbL nanocolloidalparticles with different coating composition: PDDA coating — 3, and (PDDA/PSS)3coating — 4. Sink conditions; PBS buffer, pH 7.4.

measurement using the light scattering technique also confirmed theparticle size between 100 and 150 nm.

Taking into account that the thickness of a single polymeric layer isapprox. 1.5 nm in the dry state, we can easily calculate that the drugcontent in stable nanocolloidal particles of poorly soluble drugs isfrom 85% wt (in case of tamoxifen particles with a triple bilayercoating) to approx. 90%wt (in case of paclitaxel particles with a doublelayer coating), which is dramatically higher that is the case of anyother solubilization method. Colloidal suspensions of both drugs werecompletely stable during 2 weeks of observation.

3.3. Drug release from LbL nanocolloids

LbL technology allows for the easy control of drug release rate frompolymer-stabilized colloidal nanoparticles by simple changes ofcoating thickness or composition. Fig. 5 presents the release curvesfor tamoxifen with different coating thickness in standard sinkconditions at the same drug concentration of 2 mg/mL. The curveswere produced from the experimental points using Peppas' model ofexponential approximation [30]. As one could expect, we observed aslower release rate as the number of polyelectrolyte layers in the shellincreases. Under sink conditions, non-coated tamoxifen crystals (bothwithout and with sonication) were solubilized within approx. 2 h,while LbL coating extended this time to approx. 10 h (extrapolation).Similar results were obtained for paclitaxel (data not shown).Certainly, slower release rates are also possible depending on thearchitecture of the LbL coating.

Although, it is hard to imagine any specifically increased sensitivityof the polymeric coatings used to biological media, we haveperformed some preliminary experiments on the behavior of LbL-coated drug nanoparticles in 3% solution of bovine serum albumin

Fig. 6. ELISA for 2C5-LbL-paclitaxel immuno-nanoparticles (the values are average±SD).

NANOTECHNOLOGY

260 A. Agarwal et al. / Journal of Controlled Release 128 (2008) 255–260

(BSA, mimicking blood plasma). It was found that the incubation ofpaclitaxel nanoparticles coated with (PAH-PSS)2 at 1 mg/mL concen-tration for 24 h in the BSA solution does not change particle size andsize distribution pattern (light scattering), and the drug release rateremains the same as in PBS buffer, pH 7.4 (sink conditions)— ca. 90% ofthe drug was released within 14 h in both media. More experimentson the stability of nanocolloids and drug release in the blood andblood-simulating media are now in progress.

3.4. Surface modification of LbL-coated drug nanoparticles

To confirm that LbL-coated drug nanoparticles can be easilyderivatized on the surface in order to impart various additionalproperties including targetability, we have attached tumor-specificmAb 2C5 to the paclitaxel nanoparticles via free amino groupsbelonging to the surface layer of PAH (see Materials and methods).

ELISA with the nucleosome monolayer (specific antigen for mAb2C5), the results of which are presented in Fig. 6, clearly confirms that2C5-modified LbL-coated paclitaxel nanoparticles acquired the abilityto specifically recognize the target antigen, i.e. became targeted.

3.5. Increased cytotoxicity of tumor cell-targeted LbL drug nanocolloidsin vitro

Our preliminary in vitro experiments with MCF-7 and BT-20 cancercell lines clearly confirmed that targeted drug nanoparticles demon-strate higher cytotoxicity than the non-targeted counterparts. Atpaclitaxel concentration of 100 ng/mL in case of MCF-7 cells and30 ng/mL in case of BT-20 cells, when virtually no cytotoxic effect canbe observed with non-targeted LbL-coated paclitaxel nanoparticles(around 95% of cancer cells remains alive after the incubation for 48 or72 h), 2C5-targeted LbL-coated paclitaxel particles kill at least 30% ofcancer cells. This result confirms that LbL technology allows fordecorating the surface of stable colloidal drug particles with very highdrug content with various additional functions as needed.

4. Conclusions

Thus, we have clearly demonstrated the following: 1) Stablenanocolloids of insoluble drugs with very high drug content can beeasily and reproducibly prepared through the application of the LbLtechnology, i.e. combination of sonication and alternate adsorption ofoppositely charged polyelectrolytes, resulting in nanoparticles withthe content of the drug far exceeding other known systems; 2) Drugrelease rate from such nanocolloidal particles can be easily controlledby assembling organized multilayer shells with required wallcomposition, density and thickness; 3) Various additional functions,such as specific targeted ligands, can be easily attached to the surfaceon nanocolloidal particles of poorly soluble drugs by using a polymerwith free reactive groups for the “outer” coating and preserve theirspecific properties upon the attachment; 4) Since drugs are notmodified in any way in the process of solubilization and release as freedrug molecules, there is no concern regarding possible changes indrug activity in vivo, however, to deliver a desired dose of a poorlysoluble drug in the body, a very small quantity of a polymer (polymericcarrier) is required compared with other protocols currently used foradministration of poorly soluble pharmaceuticals, thus simplifying theadministration procedure and decreasing the possibility of carrier-related side-reactions. This may represent a promising approach toobtaining convenient dosage forms of poorly soluble drugs.

References

[1] V.P. Torchilin, Multifunctional nanocarriers, Adv. Drug Deliv. Rev. 58 (2006)1532–1555.

[2] V.P. Torchilin (Ed.), Nanoparticulates as Pharmaceutical Carriers, Imperial CollegePress, London, UK, 2006.

[3] A.J. Domb, Y. Tabata, M.N.V. Ravi Kumar, S. Farber (Eds.), Nanoparticles forPharmaceutical Applications, American Scientific Publishers, Stevenson Ranch, CA,2007.

[4] B.A. Shabner, J.M. Collings (Eds.), Cancer Chemotherapy: Principles and Practice,Lippincott Williams & Wilkins, Philadelphia, PA, 1990.

[5] K. Yokogawa, E. Nakashima, J. Ishizaki, H. Maeda, T. Nagano, F. Ichimura,Relationships in the structure-tissue distribution of basic drugs in the rabbit,Pharm. Res. 7 (1990) 691–696.

[6] A. Hageluken, L. Grunbaum, B. Nurnberg, R. Harhammer, W. Schunack, R. Seifert,Lipophilic beta-adrenoceptor antagonists and local anesthetics are effective directactivators of G-proteins, Biochem. Pharmacol. 47 (1994) 1789–1795.

[7] C.A. Lipinski, Drug-like properties and the causes of poor solubility and poorpermeability, J. Pharmacol. Toxicol. Methods 44 (2000) 235–249.

[8] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and computa-tional approaches to estimate solubility and permeability in drug discovery anddevelopment settings, Adv. Drug Deliv. Rev. 46 (2001) 3–26.

[9] M. Jones, J. Leroux, Polymeric micelles — a new generation of colloidal drugcarriers, Eur. J. Pharm. Biopharm. 48 (1999) 101–111.

[10] V.P. Torchilin, Micellar nanocarriers: pharmaceutical perspectives, Pharm. Res. 24(2007) 1–16.

[11] D.D. Lasic, Mixed micelles in drug delivery, Nature 355 (1992) 279–280.[12] Y. Masuda, H. Yoshikawa, K. Takada, S. Muranishi, The mode of enhanced enteral

absorption of macromolecules by lipid-surfactant mixed micelles. I, J. Pharmaco-bio-Dyn. 9 (1986) 793–798.

[13] S. Muranishi, Absorption enhancers, Crit. Rev. Ther. Drug Carr. Syst. 7 (1990) 1–33.[14] B. Saletu, P. Anderer, K. Kinsperger, J. Grunberger, W. Sieghart, Comparative

bioavailability studies with a new mixed-micelles solution of diazepam utilizingradioreceptor assay, psychometry and EEG brain mapping, Int. Clin. Psychophar-macol. 3 (1988) 287–323.

[15] A.V. Kabanov, V.P. Chekhonin, V. Alakhov, E.V. Batrakova, A.S. Lebedev, N.S. Melik-Nubarov, S.A. Arzhakov, A.V. Levashov, G.V. Morozov, E.S. Severin, et al., Theneuroleptic activity of haloperidol increases after its solubilization in surfactantmicelles. Micelles as microcontainers for drug targeting, FEBS Lett. 258 (1989)343–345.

[16] H. Maeda, T. Sawa, T. Konno, Mechanism of tumor-targeted delivery ofmacromolecular drugs, including the EPR effect in solid tumor and clinicaloverview of the prototype polymeric drug SMANCS, J. Control. Release 74 (2001)47–61.

[17] H. Maeda, J. Wu, T. Sawa, Y. Matsumura, K. Hori, Tumor vascular permeability andthe EPR effect in macromolecular therapeutics: a review, J. Control. Release 65(2000) 271–284.

[18] G. Kwon, S. Suwa, M. Yokoyama, T. Okano, Y. Sakurai, K. Kataoka, Enhanced tumoraccumulation and prolonged circulation times of micelle-forming poly(ethyleneoxide-aspartate) block copolymer-adriamycin conjugates, J. Control. Release 29(1994) 17–23.

[19] F. Caruso, R.A. Caruso, H. Mohwald, Nanoengineering of inorganic and hybridhollow spheres by colloidal templating, Science 282 (1998) 1111–1114.

[20] Y. Lvov, A. Antipov, H. Mohwald, G.B. Sukhorukov, Urease encapsulation innanoorganized microshells, Nano Lett. 1 (2001) 125–128.

[21] A. Antipov, G.B. Sukhorukov, E. Donath, H. Möhwald, Sustained release propertiesof polyelectrolyte multilayer capsules, J. Phys. Chem. 105 (2001) 2281–2284.

[22] G. Decher, Fuzzy nanoassemblies: toward layered polymeric multicomposites,Science 227 (1997) 1232–1237.

[23] G. Decher, J. Schlenoff (Eds.), Multilayer Thin Films: sequential assembly ofnanocomposite materials, Wiley-VCH, Weinheim, Germany, 2003.

[24] Y. Lvov, G. Decher, H. Möhwald, Assembly, structural characterization and thermalbehavior of layer-by-layer deposited ultrathin films, Langmuir 9 (1993) 481–486.

[25] Y. Lvov, K. Ariga, I. Ichinose, T. Kunitake, Assembly of multicomponent proteinfilms by means of electrostatic layer-by-layer absorption, J. Am. Chem. Soc. 117(1995) 6117–6122.

[26] X. Qui, S. Leporatti, E. Donath, H. Mohwald, Studies of the ibuprofen releaseproperties of polysaccharide encapsulated ibuprofen, Langmuir 17 (2001)5375–5380.

[27] H. Ai, S.A. Jones, M.M. de Villiers, Y.M. Lvov, Nano-encapsulation of furosemidemicrocrystals for controlled drug release, J. Control. Release 86 (2003) 59–68.

[28] L.Z. Iakoubov, V.P. Torchilin, A novel class of antitumor antibodies: nucleosome-restricted antinuclear autoantibodies (ANA) from healthy aged nonautoimmunemice, Oncol. Res. 9 (1997) 439–446.

[29] L.Z. Iakoubov, V.P. Torchilin, Nucleosome-releasing treatment makes survivingtumor cells better targets for nucleosome-specific anticancer antibodies, CancerDetect. Prev. 22 (1998) 470–475.

[30] N.A. Peppas, Analysis of Fickian and non-Fickian drug release from polymers,Pharm. Acta Helv. 60 (1985) 110–112.