Removable Nanocoatings for siRNA Polyplexes

Published: January 10, 2011

r 2011 American Chemical Society 169 dx.doi.org/10.1021/bc100197e | Bioconjugate Chem. 2011, 22, 169–179

ARTICLE

pubs.acs.org/bc

Removable Nanocoatings for siRNA Polyplexes

Libor Kostka,*,† �Cestmír Ko�n�ak,† Vladimír �Subr,† Milena �Spírkov�a,† Yoseph Addadi,‡ Michal Neeman,‡

Twan Lammers,§,|| and Karel Ulbrich†

†Institute of Macromolecular Chemistry, v. v. i., Academy of Sciences of the Czech Republic, Heyrovsk�y sq. 2, 162 06, Prague 6,Czech Republic‡Department of Biological Regulation, The Weizmann Institute of Science, Rehovot 76100, Israel§Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Sorbonnelaan 16,3584 CA Utrecht, The Netherlands

)Department of Experimental Molecular Imaging, RWTH - Aachen University, Pauwelsstrasse 30, 52074 Aachen, Germany

ABSTRACT: To assist in overcoming the in-herent instability of nucleic acid-containingpolyplexes in physiological solutions, we havehere set out to develop removable nanocoat-ings for modifying the surface of siRNA-basednanoparticles. Here, N-(2-hydroxypropyl)-methacrylamide (HPMA) based copolymerscontaining carbonylthiazolidine-2-thione (TT)reactive groups in their side chains bound viadisulfide spacers to the polymeric backbonewere synthesized, and these copolymers wereused to coat the surface of polyplexes formedby the self-assembly of anti-Luciferase siRNAwith the polycations polyethylene imine (PEI)and poly(HPMA)-grafted poly(L-lysine)(GPL). The coating process was monitoredby analyzing changes in the weight-averagemolecular weight (Mw), the hydrodynamic radius (Rh), and the zeta-potential (ζ) of the polyplexes, using both static (SLS) anddynamic (DLS) light scattering methods. The outlined methods resulted in the attachment of, on average, 28 polymer molecules tothe surface of the polyplexes, forming a ∼5-nm-thick hydrophilic stealth coating. Initial efforts to develop RGD-targeted coatedpolyplexes are also described. Atomic force microscopy (AFM) showed an angular polyplex structure and confirmed the narrow sizedistribution of the coated nanoparticles. The stability of the polymer-coated and uncoated polyplexes was evaluated by gelelectrophoresis and by turbidity measurements, and it was found that modifying the surface of the siRNA-containing polyplexessubstantially improved their stability in physiological solutions. Using polymer-coated GPL-based polyplexes containing anti-Luciferase siRNA, we finally also obtained some initial proof-of-principle for time- and concentration-dependent target-specific genesilencing, suggesting that these systems hold significant potential for further in vitro and in vivo evaluation.

’ INTRODUCTION

Gene therapy provides a possibility for the treatment ofvarious human diseases in situations where conventional ap-proaches are not feasible or ineffective.1However, the expansionof gene therapy is limited by the development of safe, sufficientlystable, specific, and efficient gene delivery systems. The deliverysystems must be able to protect DNA or RNA from degradationduring transport, provide targeted delivery to the pathologicalsite, and promote target cell-specific uptake and intracellulartrafficking.2 The gene delivery vectors, such as polyelectrolytecomplexes (PEC) of DNA,1 RNA,3,4 or modified viruses, havebeen developed to overcome these barriers and efficientlymediate in vivo expression of genes. However, major limitations

that still restrict therapeutic use of nonviral vectors relate to theirinstability in vivo and to nonspecific interactions of PEC withcells of the immune system and blood plasma proteins duringtheir transport to target cells. Both instability and the uptake bymacrophages and the resulting rapid elimination of vectors fromthe bloodstream by cells of the reticuloendothelial system can besuccessfully suppressed by modifying the surface of polyplexeswith covalently linked hydrophilic polymers. Such stealth coat-ings significantly improve stability, they prolong circulation times

Received: April 23, 2010Revised: December 6, 2010

170 dx.doi.org/10.1021/bc100197e |Bioconjugate Chem. 2011, 22, 169–179

Bioconjugate Chemistry ARTICLE

in plasma, and they suppress nonspecific cell entry.5,6 Moreover,the wide tissue distribution of vectors can be restricted by bindingof ligands that are able to selectively interact with receptors ex-pressed on the surface of target cells.7,8 However, although such(targeted) polymeric coatings often sufficiently protect the vec-tors from rapid dissociation and from undesirable interactionswith blood components, and although they generally enableefficient and target cell-specific uptake, they often also preventthe escape of DNA (or RNA) from the delivery system to thecytosol, which results in dramatic decreases of transfectionefficiency.9

In light of this, novel, highly hydrophilic and multivalentcopolymers based on N-(2-hydroxypropyl)methacrylamide(HPMA) containing reactive carbonylthiazolidine-2-thione(TT) side chain groups bound via reductively cleavable disulfidelinkages to the polymer backbone have been developed. Thesepolymers enable surface modification of suitable DNA or RNAnanoparticle carriers with surface nanolayers that are removableby the reductive environment of the cytosol of target cells. Thedisulfide bridge can be cleaved by an interaction with reductiveagents forming two thiol groups, or may interchange theirsubstituents with other thiols or disulfides.10,11 This reductivereaction represents an important function in biological systems,and is especially interesting in the design of drug, protein, RNA,and DNA delivery systems, since the natural reductive agentL-glutathione (L-R-glutamyl-L-cysteinyl glycine, GSH) occurs inconcentrations of 0.1-10.0 mmol/L inside living cells, while theextracellular level of GSH is 3 orders of magnitude lower.12Newreactive copolymers containing reductively degradable spacerscan be used in the design of new polymer-coated gene deliveryvectors that are relatively stable during their transport in thebloodstream, but disintegrate and release siRNA or DNA aftercellular internalization. This process increases gene expressionand the efficacy of RNA interference. The properties of new,highly hydrophilic polymers bearing reactive TT groups attachedvia biodegradable spacers and the efficiency of coating of poly-plexes with the polymers were compared with those of previouslydescribed polymers containing nondegradable spacers, whichwere successfully used for the coating of both viral and nonviralgene delivery vectors.13

In this paper, as a first step toward the development of poly-plexes with improved in vivo stability and prolonged in vivocirculation times, we describe the synthesis and physicochemicalcharacterization of HPMA-based multivalent copolymers con-taining TT reactive groups bound to the polymer backbone viaspacer-containing disulfide bonds. In addition, we also studiedtheir use for the coating of PEC prepared by the self-assembly ofantiluciferase siRNA with polyethylene imine (PEI) or poly-(HPMA)-grafted poly(L-lysine) (GPL) copolymer, and we ex-tensively analyzed their physicochemical properties. The coatingprocess was monitored by changes in the weight-average molec-ular weight (Mw), the hydrodynamic radius (Rh), and the zetapotential (ζ) using either static (SLS) or dynamic (DLS) laserlight scattering methods, and the stability of the coated (siRNA/GPL/PHPMA) or uncoated (siRNA/GPL) siRNA complexeswas examined by gel electrophoresis and by turbidity measure-ments in physiological solutions. Finally, to provide some initialexperimental evidence for biological activity, we also evaluatedthe ability of the coated and uncoated complexes to inducetarget-specific gene silencing; usingMDA-MB-231 human breastcancer cells double-transfected with luciferase and red fluores-cence protein (RFP).

’MATERIALS AND METHODS

Materials. Methacryloyl chloride, 1-aminopropan-2-ol, 3,30-dithio-dipropionic acid (DTDP), 4,5-dihydrothiazole-2-thiol,N,N0-dicyclohexylcarbodiimide (DCC), 4-(dimethylamino)pyr-idine (DMAP), triethylamine (Et3N), 2,2

0-azobis(isobutyro-nitrile) (AIBN), tetrahydrofurane (THF), N,N-dimethylfor-mamide (DMF), dimethyl sulfoxide (DMSO) and branchedpolyethylenimine (PEI, Mw = 25 000, dn/dc = 0.21) were pur-chased from Sigma-Aldrich.N-(3-Aminopropyl)methacrylamidehydrochloride was from Polysciences, Inc. FlucDsiRNA (siRNA,Mw = 17 930, dn/dc = 0.185, mass/charge = 391 g/mol) wasobtained from Integrated DNA Technologies (Leuven, Belgium).Synthesis of cyclic form of the RGD peptide c(RGDfK) and itsscrambled analogue was described previously.14 All other chemi-cals and solvents were of analytical grade.Monomer Synthesis. N-(2-Hydroxypropyl)methacrylamide

(HPMA) was synthesized by a modified reaction of methacryloylchloride with 1-aminopropan-2-ol in dichloromethane in thepresence of sodium carbonate.15 Synthesis of 3-(N-methacry-loylglycylglycyl)thiazolidine-2-thione monomer (Ma-GlyGly-TT)was performed as described in ref 16.N-3-[3-(Methacryloylamino)propyl]-7-(diethylamino)cou-

marine-3-carboxamide. This was prepared by reaction ofN-(3-aminopropyl)methacrylamide (2.5 mg, 13.9 μmol) and7-(diethylamino)coumarine-3-carboxylic acid N-succinimidylester (5 mg, 13.9 μmol) in 300 μL DMF for one hour at roomtemperature. Crude product was purified on a Sephadex LH-20column with methanol as the mobile phase and recrystallizedfrom a mixture of ethyl acetate and diethyl ether. HPLC analysis(Chromolith column, gradient of water in acetonitrile of5-100%) gave a single peak at 340 nm with a retention time of13.6 min.Synthesis of N-1-3-[(3-[3-Oxo-3-(2-thioxo-1,3-thiazolan-

3-yl)propyl]disulfanylpropanoyl)amino]propyl-2-methyla-crylamide (Ma-SS-TT). 2-Methyl-N-[3-[3-[3-oxo-3-(2-thiox-othiazolidin-3-yl)propyl]disulfanylpropanoylamino]propyl]prop-2-enamide (DTDP-(TT)2, 1.77 g, 4.3 mmol; synthesisdescribed previously17) was dissolved in 20 mL of THF. To thissolution, 0.36 g (2 mmol) ofN-(3-aminopropyl)methacrylamidehydrochloride dissolved in a mixture of 5 mL THF and 5 mLmethanol containing 0.33 mL (2.4 mmol) Et3N was addeddropwise, and the reaction mixture was stirred for 30 min atroom temperature. ProductMa-SS-TTwas purified on a silica gelcolumn in a mobile phase of dichloromethane and acetone (1:1),and then recrystallized from a mixture of dichloromethane anddiethyl ether. Yield: 0.34 g (47%). HPLC analysis (Chromolithcolumn, gradient of water in acetonitrile of 5-100%) gave asingle peak at 305 nmwith a retention time of 8.6 min. Elementalanalysis: calcd/found C = 44.11/44.35%; H = 5.78/5.71%;N=9.65/9.45%; S=29.44/28.92%. 1H NMR 300 MHz (CDCl3,296 K) ppm: 1.66-1.69 (m, 2H), 1.97 (s, 3H), 2.61 (t, 2H),2.92-2.99 (m, 4H), 3.27-3.35 (m, 6H), 3.67 (t, 2H), 4.58 (t,2H), 5.34 (s, 1H), 5.76 (s, 1H), 6.37 (s, 1H), 6.67 (s, 1H).Synthesis of Copolymers. All copolymers were prepared by

radical solution copolymerization of HPMA and the respectivecomonomers in DMSO using AIBN as the initiator. Polymeri-zation was carried out under a nitrogen atmosphere in a sealedampule at 60 �C for 6 h.15 Structures of all copolymers are shownin Scheme 1.Polycationic graft copolymer PLL-g-poly(HPMA) (GPL,

Mw=2 �105, dn/dc = 0.18, mass/charge = 344.5 g/mol) was



prepared by reaction of poly(HPMA) grafts (Mn = 4 300)terminated with succinimidyloxycarbonyl end groups with high-molecular-weight PLL (Mw = 130 000).3

Characterization of Reactive Polymers. The content of TTgroups in copolymers was determined spectrophotometricallyusing UV absorption with a Specord 205 (Analytik Jena,Germany) spectrophotometer (ε305 = 9 500 L/mol.cm, methanol).Weight (Mw) and number (Mn) average molecular weights weredetermined by size exclusion chromatography on a HPLCsystem LC-10 (Shimadzu, Japan) equipped with a DAWN 8multiangle light scattering detector, an Optilab rEX (WyattTechnology Corp., USA), a UV detector, and a Superose6 column (300 � 10 mm, GE Healthcare, USA) using 0.3 Msodium acetate buffer, pH 6.5, as a mobile phase at a flow rate of0.5 mL/min.16 Characteristics of reactive polymers are summar-ized in Table 1.Preparation of siRNA/PEI Complexes. Formation of siRNA/

PEI complexes (w/w = 1) was achieved by flash addition of filteredPEI solution (40 μg/mL) to siRNA solution (40 μg/mL) withvigorous stirring. Formed complexes were filtered over 0.22 μmPVDF filters (saturated with PEI).Preparation of siRNA/GPL Complexes. The siRNA/GPL

PECs were prepared with various charge ratios (j) from 20 to 1.We found that the complexes prepared with j = 1.5 are the mostsuitable for subsequent coating reactions and their stability wasreasonable. Since low concentrations of salt may decrease thelevel of aggregation by 2 orders of magnitude,18 all these com-plexes were prepared in the presence of 0.01 M NaCl. ThesePECs were prepared by slow addition from a linear dispenser(rate of 15mL/h) of siRNA solution (4mL, 0.69mg/mL, 0.01M

NaCl) to the filtered solution of GPL (8 mL, 0.584 mg/mL,0.01 M NaCl).Surface Modification of PEC. A freshly prepared aqueous

solution of HPMA copolymer containing TT groups was addedto a stirred solution of PEC prepared with an excess of aminogroups (j = 1.5). Optimal conditions for aminolysis wereachieved by addition of 1 M HEPES buffer (pH 8.7, 40 μLper mL) to reach the final pH of 8.2. The reaction was carried outat room temperature, typically for 0.5 to 1.0 h. An excess ofpolymer TT groups in comparison to free amino groups on thePEC surface was used and the concentrations of the coatingpolymers ranged from 0.5 to 2.0 mg/mL. After the coatingreaction, the excess of the unreacted polymer was removed bycentrifugal filtration, repeated 5 times (Millipore Amicon Ultra-4with Ultracel-100 membrane).PEC Characterization Using Light Scattering Methods.

Weight-average molecular weight was determined by static lightscattering (SLS) carried out with an ALV goniometer. Theinstrument was equipped with an intensity-stabilized 22 mWHe-Ne laser as the light source and a PC for data recording. Theaccuracy of the measurements was found to be approximately1%. The scattering curves were measured 2-3 times after prepa-ration of PEC and 2-3 times after their surface modification.The refractive index increments, dn/dc, of the individual

components of the complexes were taken from the literature.The complex concentrations (cpec) and refractive index incre-

ments (dn/dc)pec of the complexes were calculated as a functionof the molar mixing ratio, j (positive to negative charge ratio),on the basis of the model of complex formation given in ourprevious paper.4

The static light scattering data were analyzed by a Zimm plot

Kcpec

RðqÞ¼

1

Mwpec þ

Rg2q2

3Mwpec ð1Þ

where R(q) is the Rayleigh ratio of the scattering intensity,q = (4π/λ)sin θ/2, λ is wavelength in the medium, θ is the

Scheme 1. Structure of HPMA Copolymers Containing TT Reactive Groups in Side Chains

Table 1. Characteristics of Reactive HPMA Copolymers

Mwcp (g/mol) Mw/Mn TT (mol %)

POL 1 85 300 1.7 9.9

POL 2 56 000 1.9 8.2

POL 3 52 800 1.4 9.0



scattering angle between the incident and the scattered beam,K is a contrast factor containing the optical parameters, cpec is thecomplex concentration, Mw is the weight-average of the molarmass of the complex particles, and Rg is their radius of gyration(calculated from the z-average of the square). The concentrationdependence was neglected, which seemed to be justified becauseof the low concentrations of the PEC solutions (∼10-5 g/mL).Extrapolation to the zero scattering angle was carried out bylinear or quadratic fits of the scattering curves. The experimen-tal error of Mw determination for the complexes was typicallyabout 2%.The hydrodynamic radius of particles (Rh) was determined by

dynamic light scattering (DLS) carried out on the same ALVinstrument in the angular range 30-140�. An ALV 6000 multi-bit, multitau autocorrelator covering approximately 12 decades indelay time (τ) was used for measurements of time autocorrela-tion functions. The autocorrelation functions were analyzedusing the method of cumulants (a quadratic fit). From the firstcumulant, Γ, the diffusion coefficient, D, was calculated from theequation D = Γ(q)/q2, where q is the scattering vector q = 4πnsin(θ/2)/λ, n is the refractive index of a solvent, θ is thescattering angle, and λ is the wavelength of the incident light.The polydispersity index (PDI), which reflects the size poly-dispersity of the complexes, was obtained from the secondmoment (cumulant), μ2, where PDI = μ2/Γh

2. The inverseLaplace transform using the REPES19 method of constrainedregularization, which is similar in many respects to the inversionroutine CONTIN,20 was used for an analysis of the timeautocorrelation functions obtained from multicomponent solu-tions. REPES directly minimizes the sum of the squared differ-ences between the experimental and calculated intensity timecorrelation functions using nonlinear programming. This methoduses an equidistant logarithmic grid with fixed components(here, a grid of 20 components per decade) and determinestheir amplitudes. As a result, a scattered light intensity distribu-tion function, A(τ), of delay times was obtained (τ = 1/Γ) thatcould be easily transformed in a distribution function of hydro-dynamic sizes.The hydrodynamic radius (Rh) was calculated from the

diffusion coefficient (D) using the Stokes-Einstein equation

Rh ¼ kT=6πηD ð2Þ

where k is the Boltzmann constant,T is the absolute temperature,and η (0.894 cP) is the viscosity of water at 25 �C. Theexperimental error of Rh determination for the complexes wastypically about 3%.

Evaluation of the Weight-Average Molecular Weight ofPEC in the Presence of Coating Polymer and ParticleAggregates. In the presence of aggregates and coating polymermolecules in solution, the weight-average molecular weight ofpolyplexes (Mw

pec) was estimated using a combination ofdynamic and static light scattering experiments.21 If several kindsof particles were present in the solution under investigation, thetotal scattered intensity (Ist) could be expressed as the sum oftheir individual contributions. The analysis of dynamic lightscattering data provided, in addition to hydrodynamic sizes ofparticles, corresponding relative scattering amplitudes of thedistinguished particle types present in the tested solution. Thus,the scattering intensity generated by particles of special interest(e.g., PEC) could be extracted from the total scattering intensityof the mixed solution by multiplication of Ist by their relative

scattering amplitude from DLS. This procedure could be suc-cessfully used when the scattered intensities of different particletypes were comparable and their contributions to the distributionof particle sizes (e.g., Rh) were sufficiently distinct. Since both ofthese requirements were fulfilled in the systems under investiga-tion, the weight-average molecular weight of coated PEC (Mw

cc)could be estimated even in the presence of coating polymers andsmall amount of high-molecular-weight PEC aggregates (nanogelsformed by the reaction of excessive PEI and multivalent poly-mer) in the solution.If the light scattering intensity is comparable to that of PEC,

the weight fraction of the aggregates must be on the order of0.01 (1 wt %) and Mw

pec can be calculated using the scatteringintensity generated from coated PEC alone at their originalconcentration, cpec. The weight-averagemolecular weight of PECis then calculated using the regular Zimm plot procedure.ζ-Potential. The ζ-potentials of uncoated and coated PEC

were measured using a Nano-ZS, model ZEN3600 (MalvernInstruments, UK). At least 10measurements of each sample werecarried out to check reproducibility. The measurements of theelectrophoretic mobility were converted to ζ-potentials (mV)using the Smoluchowski approximation. A reference measure-ment using the Malvern ζ-potential standard was run prior toeach sample analysis to check for correct instrument operation.Atomic Force Microscopy (AFM). The nanoparticles (con-

centration 1 mg/mL) were deposited on fresh mica substrate,and the surface morphology (height image) and the sum oftip-sample interactions (phase image) were characterized byAFM. All measurements were performed under ambient condi-tions using a commercial atomic force microscope (NanoScopeDimension IIIa, MultiMode Digital Instruments, Santa Barbara,CA, USA) equipped with a SSS-NCL probe, Super Sharp Silicon -SPM-Sensor (NanoSensors Switzerland; spring constant35 Nm-1, resonant frequency 173 kHz). The tapping modeAFM technique was used for collecting the images. This techni-que allowed two- and/or three-dimensional information to beobtained of both height and material heterogeneity contrastwith high resolution when recording height and phase shiftssimultaneously.Kinetics of Complex Coagulation. The kinetics of coagula-

tion of the siRNA/polycation complexes was measured byturbidity in 0.15 M NaCl solutions. The complexes were preparedand coated with polymer, as described above, at standardconditions in 0.01 M NaCl solutions. The turbidity of the com-plexes was measured spectrophotometrically at λ = 500 nm andstarted immediately after increasing the ionic strength of thecomplex solutions up to 0.15 MNaCl by the fast addition of 4 MNaCl stock solution.Electrophoresis. The ability of coating copolymers to cova-

lently bind on the PEC surface to retard the release of siRNA wasdetermined by horizontal agarose gel electrophoresis. The coat-ed complexes of siRNA/GPL with POL1 or POL3 (siRNA/GPL/PHPMA) were formed as described above at j = 1.5. Thesamples were loaded on E-Gel agarose gel (2%) with SYBRSafe and run for 15 min in an E-Gel iBase from Invitrogen. Thegel images were made by a Canon EOS 400D digital camera at280 nm.Luciferase Assay. MDA-MB-231 human epithelial breast

adenocarcinoma cells expressing both luciferase and red fluores-cent protein (RFP) were kindly provided by Prof. YoramSalomon and were grown in RPMI media supplemented with10% FCS L-glutamin and antibiotics. Cells were plated in 96



well plates (at a density of 55 000 cells per well), and 24 h afterplating, various amounts of siRNA-containing particles equiva-lent to 1, 10, and 20 μg of siRNA were added to a final volume of125 μL per well. In some experiments, siRNA/Lipofectaminecomplex was included as a control, and it was used according tothe instructions of the manufacturer (i.e., Invitrogen). Plateswere analyzed after 48, 72, and 96 h of incubation, using a Victorplate reader (PerkinElmer, USA), and both luminescence andfluorescence were quantified. The luminescent signal was pro-duced by means of the Bright-Glo Luciferase Assay System(Promega, USA), and its reduction (as compared to untreatedcontrols) upon incubation with the complexes was used as ameasure for overall gene silencing. The reduction of RFP-mediated fluorescence (as compared to untreated controls)was used as a measure for nonspecific gene silencing/toxicity.To determine target-specific gene silencing, the percentage ofreduction in fluorescence was subtracted from the percentage ofreduction in luminescence.

’RESULTS AND DISCUSSION

An important limitation of RNAi-based therapy is the diffi-culty in achieving effective delivery of RNAi-bearing vectors totarget cells and tissues in vivo. Many laboratories developingPEC-based delivery vectors for small oligonucleotides, likesiRNA, are designing oligonucleotide delivery vectors usingcommercially available branched polyethylene imine (PEI). Withthe aim of using these complexes as controls in our studies, weprepared these PEI based PECs using standard methodologies

described in the literature.22 We then compared the stability ofthe PEI-based vectors with the stability of poly(HPMA)-graftedpoly(L-lysine) (GPL) vectors, and we evaluated their suitabilityfor further improvement of vector properties by coating theirsurfaces with hydrophilic N-(2-hydroxypropyl)methacrylamide(HPMA) copolymers. It is known that PEI-siRNA complexes arenot suitable for in vivo application,23 especially because of theirimproper stability,24 and coating of this complex does not resultin improvement of its stability (see Figure 1). This result rein-forced our focus on further development of siRNA complexesbased on GPL polycations.

It is well-known, however, that PLL-a polycation routinelyused for DNA delivery-does not form compact complexes withshort oligonucleotides, such as siRNA.24 With the aim of im-proving the ability of PLL to form stable complexes with siRNA,we decided to use a new type of PLL-based polycation, graftedwith short chains of hydrophilic poly(HPMA) developed andstudied in our laboratory.3

In this paper, a comparison of the properties of uncoatedsiRNA/PEI and HPMA copolymer-coated complexes with theproperties of siRNA/GPL complexes and their coated analoguesis presented and discussed.Properties of siRNA/PEI Polyplexes. Basic physical charac-

teristics of siRNA/PEI complexes prepared as reference PECsand models for the study of the coating reaction of PECs withmultivalent reactive copolymers bearing carbonylthiazolidine-2-thione (TT) reactive groups attached to the polymer backbonevia degradable and nondegradable spacers were determined bylight scattering methods (Table 2). The Rh

pec and Mwpec of

complexes coated with various copolymers differed only slightly,which demonstrated a good reproducibility in PEC preparation.The size of the PEI-based PECs was found to be close to thatpublished in the literature.22 The ratio Rh/Rg = 0.8 found forthe PECs corresponded with that predicted for solid spheres(0.78).26The polydispersity of PEI PEC (PDI=0.14) was slightlyhigher than that of DNA polyplexes.13 As the PECs wereprepared with an excess of polycation, their zeta potentials werepositive (Table 2).These PEI-based complexes were then coated with HPMA

copolymers containing nondegradable (POL 1) and degradable(POL 2) spacers terminating with TT reactive groups in buffer atpH 8.2.We incorporated the fluorescent probe coumarin into thecopolymer structure of POL 2, which enabled better monitoringof the coating polymer during the purification step. Particle Rhdistributions found in solutions of uncoated PECs and PECscoated with POL 2 were measured at θ = 150�, and data areshown in Figure 1.While Rh distribution of PECs dominates before coating, as

shown in Figure 1, the distribution of particle size in solutionafter the coating procedure was more complex. The size dis-tribution showed, besides the coated particles (35 nm), new

Figure 1. Rh distributions for uncoated (---) and coated (___) siRNA/

PEI complexes with POL 2 copolymer (ccp = 1� 10-3 g/mL) afterincubation for 220 min; cpec = 2 � 10-5 g/mL, θ = 150�.

Table 2. Effect of Coating on the Characteristics of siRNA/PEI Polyelectrolyte Complexes Coated with the Reactive CopolymersPOL 1 and POL 2

reactive copolymer cpec � 105 (g/mL) ccp � 10-3 (g/mL) Mwa � 10-6 (g/mol) ΔMct � 10-6 (g/mol) Fh g/mL nct Rh (150�) (nm) ζ potential (mV)

- 2 - 4.4 0.065 30 þ12

POL 1 1.85 2 10.7a 3.1 36 >29

POL 1 1.85 0.5 10.0a 2.9 34 >29

POL 2 1.85 1 17.2a 5.1a 0.14 91 35 -5aContribution of the coating polymer and “new particles” to the SLS was subtracted.



particles with Rh of about 120 nm. These particles (nanogel)were probably formed by chemical reaction between aminegroups of free PEI polymers (used in excess) and TT reactivegroups of the coating copolymers. The formation of similarnanogels was also observed in PEC solutions after coating withreactive copolymer POL 1 containing a nondegradable spacer.The small peak at Rh ≈ 8 nm corresponds with the peak of freecoating copolymers, which were used in excess. Unfortunately,purification of complexes from excessive polycations turned outto be impossible, because of low stability of the generatedcomplexes, and disintegration of the PEI-based polyplexes wasobserved after dialysis, column filtration, and centrifugation(results not shown). In the case of such complicated particlemixtures as those obtained after the coating procedure, theweight-average molecular weights of coated polyplexes (Mw

cc)could be estimated using a combination of dynamic and staticlight scattering experiments.21 The increase in the molecularweight of PEC due to coating with reactive copolymer POL 1 andPOL 2, ΔMct, is shown in Table 2.In order to characterize the compactness of complexes, the

structural density (Fh) of PEC was calculated from the hydro-dynamic volume of particles (VH, in cm3) and from theircorresponding molecular weight, Mw

cc (Mwcp þ ΔMct). These

results are shown in Table 2. The value of Fhcc = 0.14 obtained for

coated siRNA/PEI complexes (Table 2) was comparable withthose found for micelles in organic solvents,27 for DNA/PLLcomplexes coated with HPMA copolymer13 and oligonucleotide/(PLL-g-PHPMA) complexes.3,4 Values for Fh

pec of uncoatedsiRNA/PEI complexes were approximately 2-fold smaller (Table 2).This result meant that the coating procedure increased theobserved density of PEC.The coating efficiency was characterized by the number of

molecules attached to the PEC surface, nct, as calculated fromΔMct, using the formula nct = ΔM/Mw

cp, where Mwcp is the

weight-average molecular weight of used coating polymer. Theresults obtained for this parameter for POL 1 and POL 2 areshown in Table 2. The nct was found to be higher for the coatingcopolymer POL 2 due to its lower molecular weight.

On the basis of these findings, it can be concluded that thepurification of polymer-coated siRNA/PEI complexes is impossibledue to their poor stability and the formation of large particles bychemical reaction of an excess of PEIwith reactive coating polymers.This represents a serious problem for the proper characterization ofcoated siRNA/PEI complexes for possible future medical applica-tions. Therefore, we have designed and tested another polyelec-trolyte system (i.e., siRNA/GPL) more suitable for medicinal invivo applications.Properties of siRNA/GPL Polyplexes. First, we studied the

formation of the complexes as a function of the molar mixingratio (j). Typical results are shown in Figure 2, where the Zimmplot of the siRNA/GPL complexes is shown for a variety ofmixing ratios.

Figure 2. Zimm plot of scattering curves of the siRNA/GPL complexesin relation to the mixing ratio, j (top, j = �; bottom, j = 1).

Figure 3. Dependence of molecular weight Mwpec (a) and Rh (b) of

siRNA/GPL complexes on the mixing ratio j. The concentration ofstarting polycation solution was 3.91 � 10-4 g/mL.

Figure 4. Zimm plot for uncoated (9) and coated siRNA/GPLcomplexes with copolymers POL 1, POL 3, POL 3 þ c(RGDfK), andPOL 3þ c(RDGfK) (j = 1.5, ccp = 1� 10-3 g/mL) after an incubationtime of 24 h; cpec = 5.5 � 10-4 g/mL.



The behavior ofMwpec and Rh

pec as a function ofj is shown inFigure 3a,b. TheMw

pec was found to increase with increasing j,with the increase accelerated in the vicinity of the chargecompensation point, j≈ 1. Since the size of PEC was too smallfor reliable determination of the radius of gyration by static lightscattering, the hydrodynamic radius (Rh) of the complexes wasestimated by DLS. The j-dependence of Rh

pec was more com-plex than that ofMw

pec, while Rhpec of complexes was practically

independent of j at j > 2.5. The size of the complexes wasstrongly increased by enhanced hydrophobic aggregation at j≈1. No flocculation was observed in the investigated region of j =10-1.05. The obtained Rh

pec and Mwpec of siRNA/GPL com-

plexes corresponded well with results obtained earlier for otheroligonucleotides.3,4 For all the following experiments, wedecided to use complexes prepared at j = 1.5, which was in theregion of suitable Rh and Mw values of the complexes. Thepresence of a free polycation in the PEC solution was also notobserved at this j; in contrast to PEI, the chemical coatingreaction was not contaminated by the reaction of excessivecoating copolymers with free polycations resulting in formationof nanogel.From the Mw

pec values of the complexes, their aggregationnumber (nagg =Mw

pec/Mws) could be estimated, whereMw

s wasthe molecular weight of the unimolecular siRNA/GPL complex.Assuming full charge compensation of GPL macromolecules bysiRNA at j = 1,Mw

s was calculated using mass/charge data and

data used for Figure 2, with nagg calculated to be 6.7 andthe number of siRNA molecules per PEC was found to benrn ≈ 10.Static and dynamic light scattering methods were also used to

monitor the efficiency of the coating reaction of siRNA/GPLcomplexes. Changes in the weight-average molecular weight(Mw) and hydrodynamic radius (Rh) of the PEC were examined.The sensitivity of the methods is demonstrated in Figures 4and 5, where Zimm plots and the Rh-distribution functions foruncoated and coated siRNA/GPL (siRNA/GPL/PHPMA)complexes are shown.The complexes were prepared at j = 1.5 and were coated

with copolymer POL 1, which was attached to the complexsurface via nondegradable spacers, and polymer POL 3, whichwas connected with degradable linkages (siRNA/GPL/PHPMA)(Scheme 1). To provide some initial proof-of-principle for thefeasibility of developing polymer-coated PECs for targeteddelivery to certain receptor-overexpressing cells, we also incor-porated the standard targeting moiety c(RGDfK) and a controloligopeptide (i.e., c(RDGfK)) into the copolymer structure, bymeans of a polymer analogous reaction (i.e., POL3þc(RGDFK)and POL3þc(RDGfK)), and these copolymers were used forcoating of complexes. Zimm plots of uncoated and coated siRNA/GPL complexes measured after 24 h are shown in Figure 4. Theconcentration of coating copolymer was again 1 mg/mL in theseexperiments. Results of light scattering measurements obtainedfor all copolymer-coated complexes are shown in Table 3. Theefficiency of coating with POL 1 was slightly higher than thatobtained with POL 3 due to its higher molecular weight(Table 1). The number of siRNA molecules incorporated in acomplex was between 3 and 3.5. It was calculated using data onmolecular weight of the complexes, their density, known molec-ular weight of used polycation, and the N/P (j) ratio.In principle, the coating reaction with a multivalent polymer

can result in aggregation of complexes. Inadequate increases inRh of complexes after coating can be used as evidence of suchaggregation. Additional evidence of particle aggregation can berepresented by increased polydispersity (broadening of Rh dis-tribution) of the coated particles, which should be close to that ofuncoated particles. It can be seen in Figure 5 that the width of theRh distribution of the coated complexes was slightly smaller thanthat of the original complexes. Therefore, particle aggregationwas practically negligible. Themaximumof the distribution curveof coated PECs was found to be shifted by 5 nm to higher Rhvalues. Thus, the mean thickness of the coating layer (ΔRhct) wasfound to be 5 nm. The number of molecules attached to the PECsurface (nct) was 28 after 24 h of coating. The results above alsoshow that attachment of targeting oligopeptide does not sig-nificantly influence the coating reaction and such copolymers aresuitable for preparation of coated and oligopeptide-targeteddelivery vectors.

Figure 5. Rh distributions for siRNA/GPL complexes (j = 1.5) andcoated PEC after 24 h; coating copolymer POL 1, ccp = 1� 10-3 g/mLand cpec = 4.2 � 10-4 g/mL.

Table 3. Characteristics of siRNA/GPL PECs Modified by Reactive Copolymers at j = 1.5, Obtained after 24 h

coating polymer

Mwa � 10-6

(g mol-1)

ΔMwct � 10-6

(g mol-1) nct Nrn

Fpec,cc

(g mL-1)

Rh (θ = 0�)

nm

ζ potential

(mV)

- 0.80 3 0.007 37 þ7

POL 1 7.92 2.04 28 0.042 42

POL 3 5.32 1.45 27 0.032 40 -3

POL 3 þ c(RGDfK) 6.69 1.66 0.031 44

POL 3 þ c(RDGfK) 6.65 1.65 0.029 45



Atomic Force Microscopy (AFM). AFM was used as a directmethod for characterization of PEC particle morphology, whichconsisted of size, shape, and size polydispersity. The AFMamplitude images of siRNA/GPL complexes (j = 1.5) obtainedin the tapping mode are shown in Figure 6a,b. The sizes of driedpolyplexes obtained by the analysis of AFM images were onlyslightly higher (the radius was in the range 60-70 nm) thanthe Rh values presented in Table 3. Surprisingly, the uncoatedpolyplexes exhibited a regular angular structure (Figure 6b).Time Stability of PEC. Despite the slow speed of preparation

of siRNA/GPL complexes (15 min), the final complexeswere still in a nonequilibrium state. Therefore, we tested the

medium-term stability of their parameters (Rh,Mw). The resultsof stability for uncoated PECs (siRNA/GPL) and PECs coatedwith copolymer POL 3 (containing a degradable spacer) ob-tained within three weeks are shown in Figure 7a,b.

Figure 6. AFM amplitude images of siRNA/GPL complexes (a,b) (j = 1.5) and their POL 3-coated analogues (c,d) (ccp = 1� 10-3 g/mL): (a,c) fieldof view 2 μm; (b,d) field of view 500 nm.

Figure 7. (a) Time dependence ofMwpec andRh

pec (θ = 0) of PEC prepared atj = 1.5. (b) Time dependence ofMwcc andRh

cc (θ = 0) of PEC preparedat j = 1.5 and coated with POL 3 (ccp = 1 � 10-3 g/mL).

Figure 8. Time dissociation of siRNA/GPL complexes from horizontalelectrophoresis.

Figure 9. Time dependence of the turbidity of uncoated and coatedPEC as investigated in 0.15 M NaCl solvent.



As can be seen, theMw of both types of complexes decreased inthe first four days of observation. After that, coated complexesapproached equilibrium, but theMw of uncoated PEC continuedto decrease slightly until the experiment ended. The Rh

cc ofcoated PEC decreased more rapidly than their Mw

cc, probablydue to their internal restructuring.While the ζ-potential of PEC was positive after the coating

reaction, similar to uncoated complexes, the polarity of theζ-potential of coated complexes switched overnight to becomenegative (see Table 3). The drop in ζ-potential was evidently aresult of hydrolysis of the remaining reactive TT groups notparticipating in the coating reaction (formation of -COOHgroups). If the excessive unreacted coating polymer was notremoved from the PEC solution, the drop in ζ-potential con-tinued to a value of -22 mV after 22 days.SiRNA molecules were fixed inside the complexes by electro-

static interactions, therefore, they could be undesirably displacedfrom complexes by an exchange reaction with polyanions.3 Theremaining coating copolymers were hydrolyzed, and their reac-tive TT groups, which had not been aminolyzed in the coatingreaction, were changed into carboxylic acid groups. SiRNA couldthen be released from the complexes during longer storage timesin their solutions by interaction with hydrolyzed coating copoly-mers. Indeed, the release of siRNA from the unpurified com-plexes was proven by agarose gel electrophoresis, and theresults of the measurements of 22-day-old samples are shownin Figure 8.In order to avoid this problem, we have developed a cleaning

procedure for freshly prepared coated PECs. To purify the PECsfrom excessive coating polymer, the PEC solution was filteredthrough the Millipore Amicon Ultra-4 centrifugation device5 times with an Ultracel-100 membrane from regeneratedcellulose (cutoff 100 kDa) and diluted with 0.01 M NaCl bufferto the original concentration. The concentration and size of thecomplexes was verified by light scattering. The weight losses ofthe PECwere less than 10% after this cleaning procedure. The Rhof complexes was conserved during the cleaning procedure,and results obtained after 22 days of storage of the purifiedcomplexes are shown in Figure 8. Only the purified complexes,therefore, were used for all subsequent experiments and forbiological testing.Stability of Complexes in Saline Solution. To prove the

stability of the studied PECs under conditions mimicking

physiological environments, we measured the time dependenceof turbidity in saline solution (0.15 M NaCl). These measure-ments are related to the kinetics of PEC aggregation followed byphase separation. Results are shown in Figure 9 for uncoated andpolymer-coated complexes. Complexes prepared from PEI poly-cations showed the fastest coagulation rate, while the rate ofcoagulation was lower for PECs prepared by self-assembly ofsiRNA with GPL. Furthermore, both types of polymer-coatedPEC were stable in saline solution; the turbidity did not changewith time. Thus, it can be concluded that coating the PEC withHPMA copolymers substantially increased the colloidal stabilityof the polyplexes, with stability of GPL complexes being superiorto those of PEI, and it seems reasonable to assume that also, invivo, the stability of the coated PEC might be higher than that ofuncoated PEC.Initial in Vitro Evaluation of Gene Silencing Activity. To

provide some initial experimental evidence for the retention ofbiological activity upon coating the siRNA-containing complexes,their silencing efficacy was analyzed using MDA-MB-231 humanepithelial breast adenocarcinoma cells stably expressing lucifer-ase and red fluorescent protein (RFP). To this end, both coatedand uncoated GPL-based antiluciferase PECs were prepared, andtheir ability to specifically reduce the expression of luciferase wascompared to that of the standard transfection agent Lipofect-amine. Here, luciferase knockdown upon exposure to differentconcentrations (i.e., 1, 10, and 20 μg siRNA) and differentincubations times (i.e., 48, 72, and 96 h) was correlated withloss of RFP signal (i.e., nonspecific silencing/toxicity), findingswere normalized to those of untreated controls, and they werecompared to those obtained for Lipofectamine (which accordingthe instructions of the manufacturer was always applied at anoptimal siRNA concentration of 0.5 μg). As shown in Figure 10,it was found in these initial proof-of-principle analyses that, in themajority of cases, luciferase-specific gene silencing induced by thepolymer-coated PEC was more or less comparable to that in-duced by Lipofectamine, and that it in some cases even turnedout to be significantly better. Also, as compared to uncoatedPEC, the silencing efficacy of the polymer-coated PEC turnedout to be superior in the majority of cases (likely mostly becauseof a reduction in nonspecific gene silencing/toxicity, rather thanbecause of an increase in specific gene silencing). These findingsare by no means conclusive, and additional in vitro efficacy, invitro uptake, and in vitro toxicity analyses are necessary before

Figure 10. Concentration (A) and time (B) dependence of target-specific gene silencing in vitro in MDA-MB-231 human epithelial breastadenocarcinoma cells expressing both luciferase (to determine siRNA-specific gene silencing) and RFP (to determine unspecific gene silencing/toxicity)induced by uncoated and POL3-coated antiluciferase siRNA/GPL complexes. Values were compared to those of untreated controls and to thoseinduced by Lipofectamine/siRNA complexes. In B, lipofectamine was used at an optimal dose of 0.5 μg of siRNA (according the manufacturer’sinstructions), while the coated and uncoated polyplexes were used at doses corresponding to 10 μg of siRNA.



transferring these systems to the in vivo situation. Nonetheless,these initial in vitro insights already demonstrate quite clearlythat polymer-coated siRNA-containing PEC are able to specifi-cally induce target-specific gene silencing in vitro, and theythereby exemplify that even systems (physico-) chemicallyoptimized to eventually perform well in vivo compare well tounstable and nonstealth systems primarily designed for genesilencing in vitro.

’CONCLUSIONS

The synthesis and physicochemical properties of reactivecopolymers suitable for the coating of PEC and the physicalproperties of coated and uncoated siRNA-containing PECs aredescribed. Results of stability measurements showed that often-studied PEI-based PECs are not suitable for coating reactionsand consequently are likely suboptimal vectors for in vivo siRNAdelivery. We therefore set out to develop siRNA-containingPECs based on GPL, as well as GPL PECs coated with HPMA-based reactive hydrophilic copolymers. The synthesis and physico-chemical characterization of HPMA-based copolymers contain-ing TT reactive groups bound to the polymer chain via spacerscontaining biodegradable disulfide bonds was described. The useof the HPMA copolymers for siRNA/GPL complex coatingreactions resulted in attachment of (on average) 28 polymermolecules to the nanoparticle surface, forming a 5-nm-thickhydrophilic polymer protective layer that was selectively remo-vable in reducing environments. Coating the siRNA/GPL vec-tors with the stealthy hydrophilic polymer substantially enhancedthe stability of the vectors in aqueous media, and it might therebyimprove their potential for passive targeting to solid tumors bymeans of the EPR effect. Preliminary results on the in vitro genesilencing efficacy of the coated complexes furthermore showedthat they were able to specifically reduce target gene expression inhuman epithelial breast adenocarcinoma cells, thereby suggest-ing that these systems hold significant potential for further invitro and in vivo evaluation.

’AUTHOR INFORMATION

Corresponding Author*To whom correspondence should be addressed. E-mail: [email protected]. Fax: þ420 296809410. Tel.: þ420 296809209.

’ACKNOWLEDGMENT

The authors acknowledge financial support by the Grant Agencyof Academy of Sciences of the Czech Republic (IAAX00500803),Academy of Sciences of the Czech Republic (KAN 200200651)and by the European Union (grant LSHB-CT-2004-512087and MediTrans - Targeted delivery of Nanomedicines).

’ LIST OF SYMBOLSMw - weight-average molecular weight (g/mol)Mwa - apparent weight-average molecular weight (g/mol)Rh - hydrodynamic radius (nm)Rg - radius of gyration (nm)ζ - zeta potential of PEC (mV)F - particle density of PEC (g/mL)c - concentration (g/mL)dn/dc - refractive index incrementq - scattering vector (m-1)j - ratio of positive to negative charges

dct - surface density of coating polymer (g/nm2 mol)nct - number of coating molecules fixed on PEC surfacetct - characteristic reaction time of the slow coating process (min)

’MEANING OF SUBSCRIPTS AND SUPERSCRIPTScp - coating polymerpec - uncoated polyelectrolyte complexescc - coated PECct - coating layer of PEC

’DEFINITIONSΔMw

ct = Mwcc - Mw

pec

ΔRhct = Rh

cc - Rhpec

’REFERENCES

(1) Seymour, L. W., Kataoka, K., and Kabanov, A. V. (1998)(Kabanov, A. V., Felgner, P. L., and Seymour, L. W., Eds.) pp 219,Wiley, New York.

(2) Lechardeur, D., and Lukacs, G. L. (2002) Intracellular barriersto non-viral gene transfer. Curr. Gene Ther. 2, 183.

(3) Dautzenberg, H., Ko�n�ak, �C., Reschel, T., Zintchenko, A., andUlbrich, K. (2003) Cationic graft copolymers as a carriers for deliveryof antisense-oligonucleotides. Macromol. Biosci. 3, 425–435.

(4) Dautzenberg, H., Zintchenko, A., Ko�n�ak, �C., Reschel, T., �Subr,V., and Ulbrich, K. (2001) Polycationic graft copolymers as carriers foroligonucleotide delivery. complexes of oligonucleotides with polycatio-nic graft copolymers. Langmuir 17, 3096–3102.

(5) Green, N. K., Herbert, C. W., Hale, S. J., Hale, A. B., Mautner, V.,Harkins, R., Hermiston, T., Ulbrich, K., Fisher, K. D., and Seymour,L. W. (2004) Extended plasma circulation time and decreased toxicity ofpolymer-coated adenovirus. Gene Ther. 11, 1256–1263.

(6) Oupick�y, D., Ogris, M., Howard, K. A., Dash, P. R., Ulbrich, K.,and Seymour, L. W. (2002) Importance of lateral and steric stabilizationof polyelectrolyte gene delivery vectors for extended systemic circula-tion. Mol. Ther. 5, 463–472.

(7) Fisher, K. D., Stallwood, Y., Green, N. K., Ulbrich, K., Mautner,V., and Seymour, L. W. (2001) Polymer-coated adenovirus permitsefficient retargeting and evades neutralising antibodies. Gene Ther. 8,341–348.

(8) Dash, P. R., Read,M. L., Barrett, L. B.,Wolfert, M. A., and Seymour,L. W. (1999) Factors affecting blood clearance and in vivo distribution ofpolyelectrolyte complexes for gene delivery. Gene Ther. 6, 643.

(9) Walker, G. F., Fella, C., Pelí�sek, J., Fahrmeir, J., Boeckle, S., Ogris,M., and Wagner, E. (2005) Toward synthetic viruses: endosomalpH-triggered deshielding of targeted polyplexes greatly enhances genetransfer in vitro and in vivo. Mol. Ther. 11, 418–425.

(10) Laga, R., Ko�n�ak, �C., �Subr, V., and Ulbrich, K. (2007) Coating ofnanoparticles bearing amino groups on the surface with hydrophilicHPMA-based polymers. Colloid Polym. Sci. 285, 1509–1514.

(11) Bernkop-Schn€urch, A. (2005) Thiomers: A new generation ofmucoadhesive polymers. Adv. Drug Delivery Rev. 57, 1569–1582.

(12) Chawla, R. K., Lewis, F. W., and Kutner, M. H. (1984) Plasmacysteine, cystine, and glutathione in cirrhosis. Gastroenterology 87,770–776.

(13) �Subr, V., Ko�n�ak, �C., Laga, R., and Ulbrich, K. (2006) Coatingof DNA/Poly(l-lysine) complexes by covalent attachment of poly[N-(2-hydroxypropyl)methacrylamide]. Biomacromolecules 7, 122–130.

(14) Pola, R., Studenovsk�y, M., Pechar, M., Ulbrich, K., Hovorka, O.,V�etvi�cka, D., and �Ríhov�a, B. (2009) HPMA-copolymer conjugatestargeted to tumor endothelium using synthetic oligopeptides HPMAcopolymer-based conjugates targeted with specific oligopeptides. J. DrugTargeting 17, 773–786.

(15) Ulbrich, K., �Subr, V., Strohalm, J., Plocov�a, D., Jelínkov�a, M.,and �Ríhov�a, B. (2000) Polymeric drugs based on conjugates of syntheticand natural macromolecules I. Synthesis and physico-chemical charac-terisation. J. Controlled Release 64, 63–79.



(16) �Subr, V., and Ulbrich, K. (2006) Synthesis and properties ofnew N-(2-hydroxypropyl)methacrylamide copolymers containing thia-zolidine-2-thione reactive groups. React. Funct. Polym. 66, 1525–1538.(17) �Subr, V., Kostka, L., Selby-Milic, T., Fisher, K., Ulbrich, K.,

Seymour, L. W., and Carlisle, R. C. (2009) Coating of adenovirus type 5with polymers containing quaternary amines prevents binding to bloodcomponents. J. Controlled Release 135, 152–158.(18) Dautzenberg, H. (1997) Polyelectrolyte complex formation in

highly aggregating systems. 1. Effect of salt: polyelectrolyte complexformation in the presence of NaCl. Macromolecules 30, 7810–7815.(19) �St�ep�anek, P. (1993) (Brown, W., Ed.) p 177, Clarendon Press,

Oxford.(20) Provencher, S. W. (1982) A constrained regularization method

for inverting data represented by linear algebraic or integral equations.Comput. Phys. Commun. 27, 213–227.(21) Ko�n�ak, �C., Ganchev, B., Teodorescu, M., Matyjaszewski, K.,

Kope�ckov�a, P., and Kope�cek, J. (2002) Poly[N-(2-hydroxypropyl)-methacrylamide-block-n-butyl acrylate] micelles in water/DMF mixedsolvents. Polymer 43, 3735–3741.(22) Meyer, M., Philipp, A., Oskuee, R., Schmidt, C., andWagner, E.

(2008) Breathing life into polycations: functionalizationwith pH-responsiveendosomolytic peptides and polyethylene glycol enables siRNA delivery.J. Am. Chem. Soc. 130, 3272–3273.(23) Urban-Klein, B., Werth, S., Abuharbeid, S., Czubayko, F., and

Aigner, A. (2005) RNAi-mediated gene-targeting through systemicapplication of polyethylenimine (PEI)-complexed siRNA in vivo. GeneTher. 12, 461–466.(24) Nakamura, Y., Kogure, K., Futaki, S., and Harashima, H. (2007)

Octaarginine-modified multifunctional envelope-type nano device forsiRNA. J. Controlled Release 119, 360–367.(25) Burchard, W. (1983) Static and dynamic light scattering from

branched polymers and biopolymers, Light Scattering from Polymers,Advances in Polymer Science, Vol 48, pp 1-124, SpringerLink; DOI:10.1007/3-540-12030-0_1.(26) Tuzar, Z., Ple�stil, J., Ko�n�ak, �C., Hlavat�a, D., and Sikora, A.

(1983) Structure and hydrodynamic properties of poly[styrene-b-(ethene-co-butene)-b-styrene] micelles in 1,4-dioxane.Die Makromol.Chem. 184, 2111–2121.(27) Dautzenberg, H. (2001) Physical Chemistry of Polyelectrolytes,

Dekker, New York.

Removable Nanocoatings for siRNA Polyplexes

Documents

Transcript of Removable Nanocoatings for siRNA Polyplexes