Accelerated gene transfer through a polysorbitol-based transporter mechanism

at SciVerse ScienceDirect

Biomaterials 32 (2011) 9908e9924

Contents lists available

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Accelerated gene transfer through a polysorbitol-based transporter mechanism

Mohammad Ariful Islama,b, Cheol-Heui Yuna,b, Yun-Jaie Choia,b, Ji-Young Shinc, Rohidas Aroted,Hu-Lin Jiangc, Sang-Kee Kanga, Jae-Woon Nahe, In-Kyu Parkf, Myung-Haing Choc,g,**, Chong-Su Choa,b,*aDepartment of Agricultural Biotechnology, Seoul National University, Seoul 151-921, Republic of KoreabResearch Institute for Agriculture and Life Sciences, Seoul National University, Seoul 151-921, Republic of Koreac Laboratory of Toxicology, College of Veterinary Medicine, Seoul National University, Seoul 151-742, Republic of KoreadDepartment of Dentistry, Seoul National University, Seoul 110-749, Republic of KoreaeDepartment of Polymer Science and Engineering, Sunchon National University, Jeonnam 540-742, Republic of KoreafDepartment of Biomedical Sciences, Chonnam National University Medical School, Gwangju 501-746, Republic of KoreagDepartment of Nanofusion Technology, Graduate School of Convergence Science and Technology, Seoul National University, Seoul 151-742, Republic of Korea

a r t i c l e i n f o

Article history:Received 19 August 2011Accepted 6 September 2011Available online 29 September 2011

Keywords:Gene transferPolysorbitolTransporterCOX-2 inhibitorAccelerated transfection

* Corresponding author. Department of AgricuNational University, Seoul 151-921, Republic of Kofax: þ82 2 875 2494.** Corresponding author. Laboratory of Toxicology, CoSeoul National University, Seoul 151-742, Republic offax: þ82 2 873 1268.

E-mail addresses: [email protected] (M.-H. Cho)

0142-9612/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.biomaterials.2011.09.013

a b s t r a c t

Here we report an accelerated gene transfer through a polysorbitol-based osmotically active transporter(PSOAT) that shows several surprising results through interesting mechanisms. The nano-sized andwell-complexed PSOAT/DNA particles are less toxic, stable at serum and show no aggregation afterlyophilization due to their polysorbitol backbone. The transfection is remarkably accelerated both in vitroand in vivo, presumably due to a transporter mechanism of PSOAT in spite of possibility of reduction oftransfection by many hydroxyl groups in the transporter. PSOAT possesses a transporter mechanismowing to its polysorbitol backbone, which enhances cellular uptake by exerting polysorbitol transporteractivity, thus accelerates gene transfer to cells because transfection ability of PSOAT is drastically reducedin the presence of a cyclooxygenase (COX)-2-specific inhibitor, which we have reported as an inhibitor ofthe transporter to cells. Moreover, the gene expression is found to be enhanced by hyperosmotic activityand buffering capacity due to polysorbitol and polyethylenimine backbone of PSOAT, respectively. Thepolysorbitol in PSOAT having polyvalency showed more efficiency in accelerating gene transfer capabilitythan monovalent sorbitol. The above interesting mechanisms display PSOAT as a remarkably potentialsystem to deliver therapeutic (small interfering RNA) and diagnostic agents for effective treatment ofcancer.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

In the nascent years of gene therapy research, gene deliveryvehicles with greater efficacy have been documented with regularfrequency. However, among the investigated reports, the rate ofsignificant and clinically successful advancements has been almostasymptotic, even though the field has matured over the past coupleof decades and the rate of published articles has grown exponen-tially [1]. The role of gene therapy is application oriented to eitherattenuation or overriding of the malfunctioning gene. It can also be

ltural Biotechnology, Seoulrea. Tel.: þ82 2 880 4868;

llege of Veterinary Medicine,Korea. Tel.: þ82 2 880 1276;

, [email protected] (C.-S. Cho).

All rights reserved.

implemented for prevention of disease through vaccination usingan antigenic component encoding a gene for a specific pathogen[1]. In reality, the primary barrier to successful gene therapyremains the lack of a safe and effective gene delivery strategy. Untiltoday, the majority of gene therapy clinical trials have used modi-fied viruses as gene delivery vectors, which, while effective fortransport of DNA to cells, potentially suffer from immunogenicity,severe toxicity, and production problems [2,3]. Therefore, non-viralpolymeric systems for gene delivery have been extensively studieddue to their well defined chemistries, diversity, and tunable phys-icochemical properties, especially cationic polymers, which offerseveral advantages, including effective condensation of anionicDNA, stability of polyplexes, and reduction of immunogenicity, aswell as toxicity and ability to resolve vector size limitations.However, despite these exciting advantages, existing polymericgene carriers are far less efficient, compared with viral vectors,owing to their low level of gene transporting capability [3].Therefore, the hunting of that ‘magic bullet’ for effective gene

mailto:[email protected]

mailto:[email protected]

www.sciencedirect.com/science/journal/01429612

http://www.elsevier.com/locate/biomaterials

http://dx.doi.org/10.1016/j.biomaterials.2011.09.013



M.A. Islam et al. / Biomaterials 32 (2011) 9908e9924 9909

transportation to cells remains the most outstanding challenge tothe scientific community.

Recently, production of molecular transporters has arousedtremendous attention due to their ability to overcome biologicalbarriers, including the cellular plasma membrane, the blood-brainbarrier (BBB), and the nuclear and mitochondrial membranes[4e6]. In this context, peptide-based molecular transporters, suchas cell penetrating peptides and related ones, have been extensivelystudied in recent years for their potential use in protein, nucleicacid, and gene delivery [7e9]. However, they are mostly vulnerableto various endogenous proteases in the body, thereby limiting theirbioavailability. Thus, for successful gene delivery, an effectivetransporter system, which is currently lacking, is a priority demand.For effective gene delivery, the transporters should have several keyfeatures, including biocompatibility, biodegradability, charge/receptor mediated uptake, tissue specificity, endosomal escape,nuclear tropism, and vector unpackaging, all of which contribute tothe canon of requirements. Sorbitol-based molecular transportershave recently been documented in two previous reports. First, Maitiet al. described a guanidine-containing sorbitol-based moleculartransporter [5]. Later, Higashi et al. investigated the potentialapplications of these transporters for DNA and small interferingRNA (siRNA) delivery, where they documented a lipidated sorbitol-based molecular transporter containing guanidine moieties [6].Through these transporters, on one hand, Maiti et al. found a veryinteresting and enhanced cellular uptake, as well as high intracel-lular localization properties, and, on the other hand, Higashi et al.found enhanced transfection activity using pDNA and increasedsilencing effect using siRNA. However, the mechanism behind theabove accelerated cellular uptake by these transporters was notinvestigated. In the present study, we have developed a poly-sorbitol-based osmotically active transporter (PSOAT) based onsorbitol dimethacrylate (SDM) and a low molecular weight linearpolyethylenimine (LMW LPEI), which is entirely different from theaforementioned studies because our transporter is based on poly-sorbitol without guanidine residues. We hypothesized that poly-sorbitol having polyvalency will accelerate cellular uptake morethan sorbitol as a monovalency by its improved transportermechanism; thus, increase transfection activity of the transporterin combination with LPEI. Since our transporter system is based onpolysorbitol, it can be expected that they would exert more accel-erated gene transport capability compared to the previously re-ported transporters based on sorbitol itself [5,6] owing to thepolysorbitol capacity.

Sorbitol (D-glucitol), an organic osmolyte, is widely produced inplants, particularly in those of the Rosaceae family, includingapples, cherries, pears, and others. It is produced commercially byreduction of D-glucose or D-glucono-1, 4-lactone and is usedextensively in the food industry due to its complete water solubilityand lack of any perceptible toxicity [5]. Our transporter systempossesses polysorbitol backbone which contains many hydroxylgroups. It will provide several beneficial properties becausereduced cytotoxicity and stable gene transfer efficiency of genecarriers by inclusion of hydroxyl groups in polycations have beenreported [10], although the hydroxyl groups of polymers such aspoly (3-amino-2-hydroxypropyl methacrylate) (PAHPMA) [11] andpoly(ethylene glycol) (PEG) [12] reduced gene transfection effi-ciency in vitro. Hence, it is interesting to see the transfectioncapability of our PSOAT system. We hypothesized that it willprovide accelerated transfection efficiency by its transportermechanism even though the PSOAT possesses many hydroxylgroups than PAHPMA and PEG.

Polyethylenimine (PEI), a ‘gold standard’ polycation, has beenused extensively for its high DNA complexation ability, and, mostimportantly, its ability to exert ‘proton sponge effects’ for

endosomal escape of polyplexes. Simultaneously, this highlycationic polymer exhibits high cytotoxicity, depending on themolecular weight and type of polycation. In several previousreports, LMW LPEI was found to be significantly less cytotoxic,compared with its high molecular weight counterparts, and moretolerable than branched PEI as well [13,14]. Until today, PEI hasbeen exclusively focused by most of the experts in the field ofpolymer-mediated gene delivery technology [12e21]. However,our novelty in this study is the introduction of a transporter conceptmainly based on polysorbitol which was hypothesized to acceleratetransfection efficiency of the transporter by increasing cellularuptake through an improved transporter mechanism. Because it isworth mentioning that the difference in gene expression dependsmainly on cellular uptake rather than endosomal escape ashypothesized earlier [6].

Here we demonstrate a class of gene transporter, the PSOAT, inorder to provide accelerated transfection efficiency with lowcytotoxicity and also show the insight mechanisms of their hightransfection. Nevertheless, the physicochemical characterizationsof PSOAT, including chemical composition by 1H nuclear magneticresonance (NMR) spectroscopy, molecular weight by gel perme-ation chromatography (GPC), gel retardation assay, protection andrelease assay of DNA, particle size, and zeta potential, wereprecisely examined. We have checked particle size and transfectionefficiency of the polyplexes in the presence of various serumconcentrations, which could be a beneficial first step toward theiruse in future clinical trials. Most important, we have explored andstudied transfection efficiency in various ways and the hightransfection of PSOAT was potentially investigated by various kindsof mechanisms, including one indigenous strategy, which can behelpful for future study in this research area. The potentiality of thisgene transporter system was also explored in vivo.

2. Materials and methods

2.1. Materials

Linear PEI (Mn: 423 Da), branched PEI (Mn: 25 kDa), dimethyl sulfoxide (DMSO),3-(4, 5-dimethyl thioazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT) reagent,bafilomycin A1, and D-sorbitol were purchased from Sigma (St. Louis, Mo, USA).Sorbitol dimethacrylate (SDM)was purchased fromMonomer-Polymer & Dajac Labs,Inc. (Trevose, PA 19053, USA) and was used as received. Luciferase reporter of 1000assay for in vitro transfection study and pGL3-control vector with SV-40 promoter,and enhancer encoding firefly (Photonus pyralis) luciferase were obtained fromPromega (Madison, WI, USA). The pEGFP-N2, which has an early promoter of cyto-megalovirus (CMV) and an enhanced green fluorescent protein (EGFP) gene, wereobtained from Clontoch (Palo Alto, CA, USA). A competent Escherichia coli strain,JM109, was used for amplification of the plasmids and purified with a plasmid DNApurification kit (DNA-spinTM iNtRoN Biotechnology, Inc.). The concentration of thepurified DNA was determined at 260 nm of UV absorbance. Deoxyribonucleic acid(DNA) sodium salt from salmon testes (Sigma) was used for the particle size and zetapotential measurement. All other chemicals used in this study were of analyticalreagent grade.

2.2. Synthesis of polysorbitol-based osmotically active transporter (PSOAT)

The PSOAT was successfully prepared based on SDM and LPEI by simple Michaeladdition reaction with a slight modification. In brief, LPEI (Mn: 423 Da) and SDM(Mn: 318.32) were dissolved separately in DMSO. SDM solutionwas then added dropby drop to LPEI solution while gently stirring at a mole ratio of 1:1 (SupplementaryTable S1). The reaction mixture was kept at 80 �C and maintained for 24 h withcontinuous stirring. Following completion of the reaction, the reaction mixture wasdialyzed using a Spectra/Por� membrane (MWCO: 1000 Da; Spectrum MedicalIndustries, Inc., Los Angeles, CA, USA) for 24 h at 4 �C against distilled water. Finally,the synthesized polymer was lyophilized and stored at �70 �C for later use.

2.3. Characterization of PSOAT

The PSOAT was characterized by 1H nuclear magnetic resonance (1H NMR,Avance� 600, Bruker, Germany) spectroscopy to confirm the synthesis and thecomposition of the synthesized PSOAT. The actual molecular weight of the polymerwas measured using a gel permeation chromatography column coupled with

M.A. Islam et al. / Biomaterials 32 (2011) 9908e99249910

multiangle laser light scattering (GPC-MALLS) using a Sodex OHpack SB-803 HQ(Phenomenox Torrels, CA, USA). The column temperature was kept at 25 �C withaflowrate of 0.5mL/min and 0.5Mammoniumacetatewas used as themobile phase.

2.4. Gel retardation assay

Gel retardation assaywas performed for examination of the condensation abilityof PSOATwith DNA by electrophoresis. PSOAT/DNA (pGL3-control, 0.1 mg) complexeswere prepared at various N/P ratios ranging from 0.5 to 10 by incubating at roomtemperature (RT) for 30min. The final volume of the complexes at each N/P ratiowas12 ml, including the 6x agarose gel loading dye mixture (Biosesang, Korea). Thecomplexes were loaded onto a 1% agarose gel containing ethidium bromide (Et-br)at a concentration of 0.1 mg/ml and run with Tris/borate/EDTA (TBE) buffer at 100 Vfor 40 min. Finally, DNA retardation was observed under ultraviolet illumination.

2.5. Protection and release assay of DNA

Protection and release assay of DNAwas carried out by electrophoresis study. Inbrief, PSOAT/pGL3 complexes (N/P 10) and free pGL3 (0.2 mg) were incubatedseparately with 1 ml of DNase I (50 units) in DNase/Mg2þ digestion buffer containing50 mM Tris-Cl (pH 7.6) and 10 mM MgCl2 at 37 �C for 30 min. Then 4 ml of EDTA(250 mM in 1 N NaOH) was added to each sample and kept 30 min at RT for DNaseinactivation. Finally, 5 ml of 1% sodium dodecyl sulfate (SDS) dissolved in distilledwater was mixed with each sample and incubated for 2 h. The final volumewith theloading dyewas 19 ml and electrophoresis was performedwith TBE running buffer in1% agarose (with 0.1 mg/ml Et-br) gel at 50 V for 1 h.

2.6. Energy-filtering transmission electron microscopy (EF-TEM)

The PSOAT/pGL3 complexes were prepared at an N/P ratio of 20 and themorphology was observed using EF-TEM (LIBRA 120, Carl Zeiss, Germany). Briefly,a single drop of the complexes was placed on a copper grid and stained with 1%uranyl acetate solution for 10 s. The grid was allowed to dry for an additional 10 minandwas then observed under the electronmicroscope. The polyplexes (N/P 20) werealso lyophilized in order to investigate the stability and structural integrity of thecomplexes and compared with lyophilized PEI 25K/pGL3 complexes (N/P 20) by EF-TEM analysis.

2.7. Measurement of particle size and zeta potential

A dynamic light scattering spectrophotometer (DLS 8000, Otsuka Electronics,Osaka, Japan) was used for measurement of the particle size and surface charge ofthe PSOAT/DNA complexes with 90 and 20 scattering angles at 25 �C, respectively.Distilled water was used for preparation of the complexes at various N/P ratios from5 to 30. The volume was 2 ml for each sample with a final DNA concentration of40 mg/ml. The particle sizes of PSOAT/DNA complexes were also examined in thepresence of various serum percentages (0, 10, 20, and 30%) at an N/P ratio of 20,according to the method described above.

2.8. Cell lines, cell culture, and cytotoxicity assay

Human lung adinocarcinoma epithelial cells (A549) and human lung bron-chioalveolar carcinoma cells (H322) were maintained in Roswell Park MemorialInstitute (RPMI)-1640 culture medium and human cervix epithelial carcinoma cells(HeLa) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) containing10% heat-inactivated fetal bovine serum (FBS, Hyclone, Logan, Utah) with 1% peni-cillin/streptomycin. All cells were cultured under standard incubation conditions at37 �C with 5% CO2.

Evaluation of the in vitro cytotoxicity of PSOAT was carried out in three cell lines(A549, HeLa, and H322 cells) and compared with the cytotoxicity of PEI 25K by MTTassay. Cells were seeded at an initial density of 10�104 cells/mL in 24-well plates in1ml growthmedium and cultured to grow to approximately 80% confluency prior toaddition of polyplexes. The growth medium was replaced with polyplexes con-taining fresh medium (supplemented with 10% FBS) at various N/P ratios (5, 10, 20,and 30). After 24 h under standard incubation conditions, 50 ml of MTT reagent (anMTT stock solution of 5 mg/ml) solution were added to each well and incubated foran additional 4 h. The mediumwas then removed and 500 ml of DMSO was added toeach well to dissolve the formed purple formazan crystals. The plates were vigor-ously shaken in order to properly dissolve the crystals and 100 ml of colored solutionfrom each well was transferred to 96-well plates and measured at the absorbance of540 nm using a VERSAmax tunable microplate reader (Sunnyvale, USA).

2.9. In vitro transfection in the absence and presence of serum

In vitro transfection efficiency was studied in three cell lines (A549, HeLa, andH322) in serum free media at various N/P ratios (5, 10, 20, and 30). In brief,10�104 cells/ml were grown to approximately 80% confluency at 37 �C with 5% CO2

incubation conditions. Themedia were then exchangedwith fresh serum freemediacontaining polyplexes (1 mg pGL3) at various N/P ratios and incubated for 6 h at 37 �C

under standard incubation condition. Serum free media were replaced with freshserum containing media and incubated for an additional 48 h. The luciferase assaywas performed according to the manufacturer’s protocol. A Chemiluminometer(Autolumat, LB953; EG&G Berthold, Germany) was used for measurement of therelative light units (RLUs) normalizedwith the estimation of protein concentration inthe cell extract using a BCAprotein assay kit (Pierce Biotechnology, Rockford, IL, USA).

In vitro cell transfection in the presence of different serum percentages wascarried out in A549 cells at an N/P ratio of 30. Approximately 10 x 104 cells/ml wereseeded as an initial cell density in 24-well plates to grow to w 80% confluency. Themedia were then replaced with fresh media with different serum percentages (0, 10,20, and 30%) containing polymer/pGL3 (1 mg) complexes and the transfection effi-ciency was determined as described above. The experiment was carried out intriplicate and transfection activity was evaluated as RLUs/mg of protein.

2.10. Flow cytometry study

To further evaluate the in vitro transfection efficiency, flow cytometry study wasperformed by transfection of A549 cells with the EGFP (1 mg) reporter gene. Aftertransfection, the cells were carefully washed with PBS and harvested with 0.25%Trypsin/EDTA. Transfection efficiency was determined by scoring the percentages ofcells that expressed green fluorescence protein (GFP) through a FACS calibratorsystem from Becton-Dickinson (San Jose, CA, USA). The fluorescence parameterswere acquired from 10,000 cells and the transfection experiment was carried out intriplicate.

2.11. In vivo transfection study

In order to determine the transfection ability of the transporter in vivo, aerosoladministration of PSOAT/tGFP complexes was performed in mice (C57BL/6) lungs.Wild-type male C57BL/6 mice (4 mice/group) were obtained from the breedingcolony of the Human Cancer Consortium National Cancer Institute (Frederick, MD,USA) and kept in the laboratory animal facility with temperature and relativehumidity maintained at 23� 2 �C and 50� 20%, respectively, under a 12 h light/darkcycle. All experimental protocols were reviewed and approved by the Animal Careand Use Committee at Seoul National University (SNU-101025-1). In a typicalexperiment, 8-week old C56BL/6 mice (4 mice/group) were placed in a nose-onlyexposure chamber (NOEC; Dusturbo, Seoul, Korea) and exposed to aerosol admin-istration. Aerosol was prepared using 1 mg of pCMV-AN with turboGFP (tGFP)(Origene, Rockville, USA), which was complexed with polymer at an N/P ratio of 20(optimized from in vitro transfection study) and delivered to mice lungs placedinside an nose-only exposure chamber (NOEC). After 48 h, mice were sacrificed andlungs were slowly flushed with phosphate-buffered-saline through the trachea.

For western blot analysis, frozen lungs were homogenized and proteinconcentrations were measured using a Bradford kit (Bio-Rad, Hercules, CA, USA).Equal amounts of proteins (30 mg) were loaded onto an SDS-gel and separatedaccordingly. Proteins were transferred to a nitrocellulose membrane and pre-blocked with 5% skim milk (w/w 1x TTBS) for 1 h at room temperature. Mono-clonal anti-GFP was produced using a general method described elsewhere anddiluted 1:10 in 5% skim milk. Anti-alpha-tubulin was purchased from Abfrontier(Seoul, Korea), diluted 1:10,000, and incubated overnight at 4 �C. Secondary anti-bodies conjugated with HRP (Invitrogen) were applied according to the manufac-turer’s protocols. Bands of interest were obtained using a luminescence imageanalyzer LAS-3000 (Fujifilm, Tokyo, Japan).

For observation of tGFP expression by fluorescence microscopy analysis, lungswere fixed in 4% paraformaldehyde for 12 h and preserved in 30% sucrose for 48 h at4 �C. Lungs were embedded with OCT compound (Sakura, Torrance, CA, USA)at < 20 �C. Lung sections were prepared at 20 mm thickness using a cryostat (Leica,Wetzlar, Germany) and evaluated under a BX51 light microscope (Olympus, Tokyo,Japan) attached with a fluorescence lamp power supply (Olympus) at an exposuretime of 1/40. Frozen lung sections were prepared at 20 mm thickness on chargedslide glasses (Fisher Scientific, Pittsburgh, PA, USA).

For immunohistochemistry (IHC) analysis, slides were washed with 0.01 M PBSand endogenous peroxidase was quenched with 0.3% H2O2 solution (w/w meth-anol). Slides were pre-blocked with antibody diluent solution (Invitrogen, Carlsbad,CA, USA) and primary anti-GFP was diluted in 1:10 and applied overnight at roomtemperature. Slides were washed, incubated with biotinylated secondary antibodyand streptavidin-enzyme conjugate for 25 min each, and DAB were appliedaccordingly (Invitrogen). They were then viewed under a light microscope(Olympus, Tokyo, Japan) at 200� magnification.

2.12. Mechanistic investigations for high transfection of PSOAT

2.12.1. Effect of polysorbitol activity inhibition on in vitro transfectionSC58236, a COX-2 inhibitor, can be used as a potent osmotic inhibitor because it

was previously reported to prevent organic osmolyte (such as sorbitol) accumula-tion in cultured cells. In order to clearly investigate the mechanism of PSOAT-mediated transfection, SC58236 was used to inhibit the osmotic activity of PSOATin order to determine the inhibition effect on their in vitro transfection in A549 cells.Different concentrations of SC58236 (0, 5, 10, and 30 mM/L) were prepared by


initially dissolving in DMSO and later diluted in serum freemedium and were addedto each respective well containing 10 �104 cells/well in 24-well plates withapproximately 80% cell confluency and kept under standard incubation conditions.After 1, 2, and 4 h of SC58236 incubation, the cells were transfected with serum freemedia containing PSOAT/pGL3 (N/P 20) and PEI 25K/pGL3 (N/P 5) complexes. Thetransfection efficiency was measured as a relative light unit (RLU)/mg of proteinusing the luciferase assay and BCA protein assaymethod, as described earlier. Finally,the effect of SC58236 on in vitro transfection was determined by analysis of thetransfection ability of PSOAT in different SC58236 concentrations.

2.12.2. Study of packed cell volumeFor evaluation of the osmotic activity of polysorbitol backbone in PSOAT,

different concentrations (0.01, 0.03, and 0.05 wt-%) of polysorbitol in PSOAT/DNAcomplexes (N/P 20) were mixed with cell suspension (A549 cells with 10% serumcontaining medium) and the packed cell volume (PCV) was measured in a mini-PCVtube (TPP, Trasadingen, Switzerland). Briefly, the cell suspension containing PSOAT/DNA complexes was incubated at RT for 5 min and then transferred into a mini-PCVtube, which was centrifuged at 5000 rpm for 1 min. The packed cell volume in thegraduated capillary of the PCV tube was estimated as the percentage of PCV, whichwas compared with the identical experiments using similar concentrations of puresorbitol instead of PSOAT/DNA complexes as control.

2.12.3. Effect of bafilomycin A1 on in vitro transfectionThe effect of bafilomycin A1 on in vitro gene transfection was evaluated in A549

cells using 24-well plates (10 x 104 cells/ml/well) with w80% cell confluency. Bafi-lomycin A1, an endosome proton pump inhibitor (a specific inhibitor of vacuolartype Hþ-ATPase), was added to each well for 10 min before transfection ata concentration of 200 nM diluted in DMSO. After 10 min incubation with bafilo-mycin A1, the cells were then transfected with the polyplexes and the luciferaseactivity was measured according to the method described earlier.

2.12.4. Difference in transfection activity between polysorbitol in PSOAT and sorbitolitself

To investigate the superiority of polysorbitol backbone in PSOAT compared tosorbitol itself for acceleration of transfection, we added 1, 3, 5, 10, and 20-foldshigher sorbitol amount with PEI 25K/pGL3 complexes (optimum N/P ratio of 5)compared to the polysorbitol content in PSOAT/pGL3 complexes (selected N/P ratioof 20) and studied their transfection efficiencies through measurement of theluciferase activity as described above.

3. Results

3.1. Synthesis and characterization of the transporter

TheMichael addition reaction has gained considerable attentionas an evergreen strategy for polymer synthesis [15]. In this study,Michael addition reaction chemistry was implemented and wesucceeded to synthesize the PSOAT from SDM and LMW LPEI (PEI423 Da) in anhydrous DMSO at 80 �C. In order to avoid hydrolysisdegradation of PSOAT, an anhydrous organic solvent was usedduring the synthesis [16]. The reaction between SDM and LPEI wascarried out at an increased temperature (80 �C), unlike in theprevious report [17], in order to improve the cross-linking effi-ciency, thus, increasing the molecular weight of the synthesizedPSOAT. The nucleophilic dimethacrylate terminal of SDM wasreacted with amine groups of LPEI for synthesis of the transporter[17,19]. The proposed reaction scheme of the synthesized PSOAT isshown in Fig.1. Each of the different part in the structure representsthe different functional properties of PSOAT as indicated. Thecomposition of the synthesized transporter was determined by 1HNMR spectroscopy and the final molecular weight wasmeasured byGPC (Fig. S1 & Table S1). PSOAT was found to be completely watersoluble, possibly due to the hydrophilic property of polysorbitolbackbone.

Degradation is one of the important characteristics for efficientand safe polymeric gene delivery in vivo [12]. Degradation of thepolymeric gene carrier does not mean the breakdown of thedelivery system immediately after application; rather, degradationshould occur in a slow and controlled fashion. Appropriate degra-dation of the polymeric gene carrier enables circumvention andreduction of cytotoxicity in the organism by generation of smallmolecular weight products in vivo and facilitates easy elimination

through the excretion pathway [12]. Furthermore, degradability ofthe polymeric carrier accelerates its transfection ability byunpackaging the polymer/DNA complexes (polyplexes) and releaseof DNA for appropriate gene expression. However, the location ofpolymer degradation in the cell that would provide the mostbeneficial outcome is still unknown [19]. Nevertheless, it is essen-tial to mention that non-discharged and non-degraded polymericcarrier accumulation causes severe cell cytotoxicity. The mostwidely used polycation, PEI 25K, was reported to induce theiraccumulation in vivo due to lack of degradation, and, thus, pre-vented discharge through excretion pathways, resulting in poten-tial toxicity [20]. In this study, the synthesized PSOAT possessesester linkages (‘degradable part’ in Fig. 1), which are susceptible tohydrolysis in physiological environments to form respective acidand alcohol, thus producing low molecular weight non-toxicbyproducts [17]. The 1H NMR spectrum of PSOAT (after reaction),in comparisonwith SDM (before reaction), was analyzed in order todetermine whether the reaction occurred successfully (Fig. S2). The1H NMR spectrum of SDM shows the peaks of highly reactive vinylgroups at values around 5.6e6.2 ppm, whereas almost no peaksappeared in this range for that of PSOAT, confirming the successfulreaction between the vinyl groups of SDM and amine groups of LPEIand formation of possible ester linkages between them. The aboveanalysis suggests successful occurrence of the reaction betweenSDM and LPEI for synthesis of PSOAT, which possesses ester link-ages, hence degradability.

3.2. Physicochemical characterization of the transporter/DNAcomplexes

3.2.1. Gel retardation, protection and release assay of DNADNA condensation ability is one of the prerequisites for

a successful polymeric gene carrier system. For efficient delivery ofDNA to cells, the cationic polymer should be able to self-assemblewith anionic DNA through electrostatic interaction and condenseit to form positively charged nano-sized polymer/DNA complexes[12,18]. A variety of different analytical strategies can be applied forcharacterization of self-assembled polyplexes, including agarosegel electrophoresis, dynamic light scattering spectroscopy, and zetapotential analysis [21]. The gel retardation study was performed forexamination of DNA (pGL3-control) condensation ability of PSOATby agarose gel electrophoresis at different N/P (amino group ofpolymer per phosphate group of DNA) ratios ranging from 0.5 to 10(Fig. 2A). The result demonstrated that the DNA condensationability of PSOATwas remarkably good, wheremigration of DNAwascompletely retarded, even at the lowest N/P ratio of 0.5. Park et al.reported previously that the LPEI (pure PEI 423 Da) had no ability tocondense DNA, even at an N/P ratio of 50, indicating the inefficientcomplexation capability by the LPEI to DNA [12]. However, ourPSOAT prepared from SDM and PEI 423 Da showed remarkable DNAcondensation ability, possibly due to the presence of hydroxylgroups of polysorbitol, which might tremendously increase thebinding capability to DNA by formation of hydrogen bonds betweenpolymer and DNA [11]. A successful and strong DNA condensationability of PSOAT lead us to investigate the PSOAT/DNA (pGL3-control) complexes for protection of DNA against degradation bynuclease enzyme (DNase I), as previously reported [18,22].Protection and release assay of DNA revealed that PSOAT effectivelyprotected DNA from DNase I, whereas the naked DNA wascompletely degraded in the presence of the same enzyme undersimilar conditions (Fig. 2B), suggesting that PSOAT could formstable and effective complexes with DNA in order to protect themfrom degradation by nucleases.

OH

HO

HO

OH

O

O

O

O

N

H

H

N

Linear Polyethylenimine (LPEI)

+

Michael AdditionDMSO

80 °C

Sorbitol dimethacrylate (SDM)

x

OH

O

HO

HO

OH

O

O

H

N

N

H

O

n

x

Transporter part Degradable

part

Endosomal

escape part

PSOAT prepared from SDM and LPEI

Fig. 1. Proposed reaction scheme for synthesis of PSOAT.


3.2.2. Particle size measurement in the absence and presence ofserum

Transport of the gene into the targeted cells requires nano-sizedtransporters for effective endocytosis by cells. The size requirementof the transporter complexed toDNA is on theorderof 200nmor lessthan that for efficient cellular uptake via endocytosis by most celltypes. Furthermore, a positive charge on the surface of thecomplexeswas shown to be essential for triggering effective cellularuptake [21]. In this regard, particle sizes and zeta potentials ofPSOAT/DNA complexes were investigated at different N/P ratiosusing a dynamic light scattering spectrophotometer (DLS). More-over, particle sizeswere examinedwith andwithout different serumpercentages. Of particular interest, it was found that the sizes ofPSOAT/DNA complexes were not significantly altered, even thoughthe N/P ratio increased from5 to 30, unlike PEI 25K/DNA complexes,whichwere decreasedwith the increase ofN/P ratios (Fig. 3A). In thecase of PEI 25K/DNA complexes, the sizes were small at high N/Pratios, due to the net electrostatic repulsive forces among the highlycationic surface charges of PEI 25K/DNA complexes, as we reportedearlier [12,17]. On the otherhand, the sizes of PSOAT/DNAcomplexeswerenot significantly changed, even at highN/P ratios, probably dueto reduction of repulsive forces, presumably owing to lower surfacecharges investigatedbyzeta potentialmeasurement in thenext part.We also examined the effect of different serum percentages (0, 10,

20, and30) onpolymer/DNAcomplexes (Fig. 3B). The result revealedthat even thehigh serumpercentage of 30didnot significantlyaffectthe particle sizes of PSOAT/DNA complexes, but rather providedslightly smallerparticles aroundor less than200nm, suggesting thatour transporter system effectively complexed with DNA and wasstable enough to retain their size, even in thepresence of high serumpercentages. On the other hand, PEI 25K/DNA complexes were notaffected by 10% of serum; however, at high serum percentages of 20and 30, the particle sizes became significantly larger. This alterationof physicochemical properties might have occurred due to theinteraction of highlycationic PEI 25Kwith the serumprotein, such asalbumin, and others [17].

3.2.3. Zeta potentials of PSOAT/DNA complexesOnce we were convinced by observation of the stability of

PSOAT/DNA complexes, even in the presence of high serumpercentages, we were very much interested in investigation of thesurface charges of these polyplexes at various N/P ratios. Thesurface charges of PSOAT/DNA complexes were measured by zetapotential analysis (Fig. 3C). Of particular interest, unlike in ourprevious report [17,18], the zeta potential of PSOAT/DNA complexesdecreased fromþ34 toþ9.6 mV as the N/P ratio increased from 5 to30. Several previous studies have documented that the existence ofhydroxyl groups can shield the surface charge of the polyplexes due

Fig. 2. (A) Agarose gel electrophoresis of PSOAT/DNA (pGL3) complexes at various N/Pratios ranging from 0.5 to 10. 0.1 mg of pDNA (pGL3) was used to complex with PSOAT.(B) Protection and release assay of DNA. DNA was released by addition of 1% SDS toPSOAT/pGL3 complexes at an N/P ratio of 10. (1) DNA marker, (2) plasmid DNA (pGL3)without DNase I, (3) plasmid DNA (pGL3) with DNase I, (4) PSOAT/pGL3 complexeswithout DNase I, and (5) PSOAT/pGL3 complexes with DNase I.


to formation of stronger intermolecular hydrogen bondingbetween polymers with hydroxyl groups and DNA, which resultedin a lower zeta potential [11,23], Therefore, the sizes of PSOAT/DNAcomplexes were not significantly altered due to reduction of elec-trostatic repulsive forces among polyplexes with lower surfacecharges and were also not significantly affected at different serumpercentages (even at as high as 30% of serum), where lower surfacecharge of the polyplexes might be less attractive to the serumproteins to be adsorbed. Although the zeta potential of PSOAT/DNAcomplexes was lower at high N/P ratios, the polyplexes still havepositive surface charges, which are essential for untargetedattachment to the anionic cell surfaces for cellular uptake [18,24].

3.2.4. Particle morphology by EF-TEM before and afterlyophilization

The polyplexes with uniform spherical shapes, along withnanometer scales, are usually regarded as an important prerequi-site parameter for successful polyplex-mediated gene delivery tocells through endocytosis [25]. TEM images showed that PSOAT/DNA (pGL3-control) complexes were at nanometer scale (barrepresents 200 nm) with uniform spherical and compact shapes,which were appreciably smaller (<200 nm), compared with thoseobtained by DLS, as shown in Fig. 4A. This was probably because thesizes measured by DLS were obtained in a hydrated state in

suspension, while the particles had been dried on carbon-coatedcopper meshes before analysis by TEM [11].

Successful preparation of dry powder formulations could be anessential strategy in polymeric gene delivery technology, whichmight simplify themethodof delivery, aswell as enhance the efficacyof thegene carrierswithout significant loss. Drypowder formulationshave shown potential for nucleic acid delivery to the lung by aerosoladministration [26,27]. With this view, PSOAT/pGL3 complexes werelyophilized and their morphologies were observed by TEM. PEI 25K/pGL3 complexes were used as a control after lyophilization. Thelyophilized PSOAT/pGL3 complexes showed remarkable retention oftheir structural integrity and stability, whereas PEI 25K/pGL3complexes were aggregated after lyophilization, as shown inFig. 4BeD. Several research groups have previously reported that thecombination of a polymeric vectorwith a cryoprotective sugar carriercan maintain the structural stability and integrity of gene deliverycarriers aswell as thebiological activityof plasmidDNA followingdrypowder processing via lyophilization [28,29]. In addition, it isimportant to mention that aggregation of polyplexes reduces theactivityandperformanceof aerosolization, theprocessofdose releaseand subsequent deposition of the polyplexes on the targeted cells[29]. Therefore, PSOAT provided a promising transporter system,which has the capability of improving the dispersibility of thelyophilized polyplex powder and enhancing cellular uptake of gene,thus, gene expression in the targeted cells. All of these promisingphysicochemical properties make PSOAT a remarkable potentialcandidate for use as an effective gene carrier in the future.

3.3. Cytotoxicity assay

A challenging demand in polymeric gene therapy is to providean efficient gene delivery system without significant toxicity[13,14]. Therefore, cytotoxicity is a primary issue in development ofa polymer-mediated gene delivery vector. Several previousresearch works using awide range of polymers have suggested thatcytotoxicity is regulated as a function of polymer structure[21,30,31]. We evaluated the cytotoxicity of PSOAT using the 3-(4,5-dimethyl thioazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT)reagent assay for comparison with the standard cationic polymer,PEI 25K, in three different cell lines at various N/P ratios, from 5 to30, as shown in Fig. 5. The result showed that PSOAT is significantlyless toxic in all cell lines and was found to be remarkably safe, ascompared to the ‘state-of-the-art’ PEI 25K, even at high N/P ratios.In the case of PSOAT, excellent cell viability was observedthroughout the N/P ratios, where almost 80 and 70% cell viabilitywere achieved at high N/P ratios of 20 and 30, respectively. Incontrast, PEI 25K provided around 60% cell viability at the lowest N/P ratio of 5 and dropped by 10e20% as the N/P ratio increased from20 to 30. This high cytotoxicity can be attributed to the fact that PEI25K would be aggregated on the cell surface and impaired cellmembrane function [18,32].

3.4. In vitro transfection activity in the absence and presence ofserum

Non-viral polymeric gene delivery vectors have receivedconsiderable interest due to their non-immunogenicity, and overalldiversity; however, most of the time they suffer from unsatisfactorytransfection efficiency, and have thus remained at the laboratorystage, until now. To investigate the transfection ability of PSOAT, weperformed luciferase expression assays in vitro using the samethree cell lines used for cell viability study at various N/P ratiosranging from 5 to 30, as shown in Fig. 6. The transfection efficiencyof our transporter system was compared with that of the standardpolycation, PEI 25K, and a leading lipid-based transfection reagent,

C

0

10

20

30

40

****

***

3020105

Zeta P

oten

tial (m

V)

N/P ratio

PSOAT/DNA complexes

100

200

300

400

**

305 10 20

N/P ratio

Pa

rtic

le

s

ize

(n

m)

PSOAT/DNA complexes

PEI 25K/DNA complexes

100

200

300

400

* *

Pa

rtic

le

s

ize

(n

m)

Percentage of serum

PSOAT/DNA complexes


10 20 30 40

A B

Fig. 3. Characterization of PSOAT/DNA complexes, (A) particle sizes without serum, (B) particle sizes with serum, and (C) zeta potentials. (n ¼ 3, error bar represents SD) (*p < 0.05;**p < 0.01; ***p < 0.001, one-way ANOVA).


Lipofectamine 2000. All of the cells were transfected using 1 mg ofplasmid DNA complexed to the polymer and the transfection abilitywas estimated by measurement of luciferase enzyme activitynormalized by the total cell protein, as we reported previously [18].The results demonstrated a remarkably high transfection ability ofPSOAT in all three cell lines at various N/P ratios. Of particularinterest, the transfection efficiencies of PSOAT were significantlyhigh at the higher N/P ratios of 20 and 30, compared with PEI 25K.At an N/P ratio of 20, PSOAT provided about 15, 20, and 30-foldshigher luciferase expression levels than those of PEI 25K in H322,A549, and Hela cells, respectively. In addition, the levels of lucif-erase expression achieved by PSOAT were 30-folds higher,compared with PEI 25K at an N/P ratio as high as 30, in all cell lines.We have already reported that the cationic PEI 25K was markedlycytotoxic at high N/P ratios, which might be the reason for its lowtransfection ability at the N/P ratios of 20 or 30, and also itstransfection efficiency was previously reported to decrease withdecreasing cell viability [13]. Transfection was also reported as cellline dependent [12]. As expected, the leading commercial lipid-based transfection agent, Lipofectamine 2000, provided the high-est luciferase expression levels in all cell lines, which were almostsimilar obtained by PSOAT, especially at higher N/P ratios. We alsoevaluated PSOAT-mediated transfection through flow cytometry(FACS) analysis by viewing EGFP expression for comparison with

PEI 25K (Fig. S3). Transfection by PSOAT was significantly higher(over 4-folds) than that of the ‘state-of-the-art’ PEI 25K, whichconfirms our luciferase assay results.

One of the practical and potential problems for polymeric genedelivery in vivo, especially for cationic liposomes, is that geneexpression is markedly inhibited in the presence of serum [33].Hence, for achievement of highly effective gene therapy, develop-ment of novel gene delivery systems for improvement of non-viralvectors that will exhibit stability in serum is essential. Focusing onthis fact, we investigated transfection efficiency of PSOAT/pGL3complexes in various serum percentages (0, 10, 20, and 30) in A549cells at an N/P ratio of 30, as shown in Fig. 7. Results showed that thetransfection efficiencies of PSOAT were not affected at all up to 20percent of serum. According to the previous report, it can bedetermined that the shielding effect of hydroxyl groups of PSOATmight retain and maintain a high transfection efficiency of thecorresponding PSOAT/pGL3 complexes in the presence of variousserum percentages up to 20 [13]. However, the transfection abilityof PSOAT was slightly decreased as the serum percentage increasedto 30. On the other hand, significantly less transfection ability wasobtained by PEI 25K, compared with PSOAT, in serum free mediumdue to their cytotoxicity at a higher N/P ratio of 30, similar to theluciferase assay earlier. However, the transfection ability increasedas the serum percentages increased, up to 20, which was

Fig. 4. TEM images of (A) PSOAT/pGL3, (B) PEI 25K/pGL3, (C) lyophilized PSOAT/pGL3, and (D) lyophilized PEI 25K/pGL3 complexes at an N/P ratio of 20.


presumably due to reduction of cytotoxicity of PEI 25K in thepresence of serum [17]. At a serum percentage of 30, luciferaseexpression was not further increased; rather, it appeared todecrease for PEI 25K. As previously reported, enhancement of genetransfer efficiency in the presence of serum was not due to thepresence of serum components, but rather to enhancement ofcellular function raised by the addition of serum. However, theyalso reported that too high a content of serum further resulted inthe decrease of transfection ability due to cell damage [34]. Thus,our result suggests that at as high as 30 percent of serum, cellactivities might have been reduced, which facilitated the slightreduction of transfection ability by PSOAT, and, in the case of PEI25K, it was not further increased. Lipofectamine-mediated geneexpression was drastically and significantly inhibited, even by 10percent of serum, which is in agreement with the result of theprevious investigation [34]. Therefore, these overall highlightsrevealed that our PSOAT system is remarkably stable in serum andable to provide highly efficient gene transfection in vivo.

3.5. Accelerated in vivo transfection by PSOAT

The significant in vitro results of PSOAT-mediated gene delivery,particularly less cytotoxicity, stability of the polyplexes in serum,as well as after lyophilization, and enhanced transfection effi-ciency encouraged us to investigate the in vivo gene deliveryefficacy of our transporter system by aerosol administration. Ina previous report, it was demonstrated that the gene can bedelivered to the lungs through both intravenous and aerosolroutes; however, aerosol delivery has several advantages overinvasive intravenous administration. Aerosol administrationallows for DNA deposition in the lungs at the pulmonary region,bypassing the systemic distribution, and prolongs the DNAretention time there. On the other hand, intravenous delivery ofthe gene causes quick systemic distribution with a shorter half-lifein the lungs [35]. In addition, aerosol administration of the gene isthree-folds more effective than direct intratracheal delivery [36].Aerosol delivery is a promising and efficient tool for delivery of the

A B

C

0

20

40

60

80

100**

**

**

*

5 10 20 30

% C

ell V

iab

ility

N/P ratio

0

20

40

60

80

100 **

*

*****

5 10 20 30

% C

ell V

ia

bility

N/P ratio

0

20

40

60

80

100 PSOAT/DNA complexes


PSOAT/DNA complexes


PSOAT/DNA complexes


**

*

**

*

N/P ratio

% C

ell V

ia

bility

5 10 20 30

Fig. 5. Cytotoxicity of PSOAT/DNA (pGL3) complexes at various N/P ratios in different cell lines: (A) A549; (B) HeLa; and (C) H322 cells (n ¼ 3, error bar represents SD) (*p < 0.05;**p < 0.01; ***p < 0.001, one-way ANOVA).


target gene to the respiratory system in a non-invasive way, whichwas proven in our previous studies [17,18]. To validate the deliveryefficiency of PSOAT to the lungs in vivo, turboGFP (tGFP) expres-sion was analyzed after aerosol administration of PSOAT/tGFPcomplexes to mice (C57BL/6; n ¼ 4) lungs using a nose-onlyexposure chamber (NOEC). After the estimated time period, thelungs of animals were collected and flushed with 0.01 M PBSthrough the trachea in order to minimize the background GFPsignals from red blood cells and were then analyzed. Lung lysatesfrom each animal were incubated with anti-tGFP. ThroughWestern blot analysis, the PSOAT/tGFP group showed significantlyincreased tGFP expression, compared to the control and PSOAT/control groups (Fig. 8A), indicating that more tGFP was deliveredby the PSOAT/tGFP group. Further confirmations were made withtissue slides of each mouse lung section using fluorescencemicroscopy and immunohistochemistry (IHC) analysis. Onlyminimal background GFP signals were detected in the control andPSOAT/control groups; however, PSOAT/tGFP showed a definiteincrease in tGFP expression, even higher than that of the PEI 25K/tGFP group (Fig. 8B). IHC analysis confirmed that PSOAT isa remarkable gene transporter, which showed more brown signalsthroughout the tissue sections of the target organ (Fig. 8C).

Both the in vitro and the in vivo studies exhibited a remarkablyaccelerated transfection ability of PSOAT even though its trans-fection capability was supposed to decrease due to the presence ofmany hydroxyl groups at its polysorbitol backbone. At this stage,this markedly high transfection ability of PSOAT fascinated us andwe anticipated that our PSOAT system might have a gene trans-porter mechanism; therefore, we investigated the inside mecha-nisms involved in its high transfection.

3.6. PSOAT mechanisms that accelerated the gene transfercapability

3.6.1. Inhibition of polysorbitol activityIn order to mechanistically confirm that the high gene transfer

ability of PSOAT was due to its polysorbitol-based transportermechanism, our first approach was to investigate the effect ofSC58236 on PSOAT-mediated transfection. SC58236, a cyclo-oxygenase (COX)-2-specific inhibitor, was reported to potentiallyreduce osmolyte (such as sorbitol) accumulation in cultured cells[37]. Recent studies have explored the tonicity-dependent regula-tion of COX-2 and correlation of COX-2 activity with the organicosmolyte-dependent adaptation of cells to hyperosmotic stress.

A B

C

****

10

2

10

3

10

4

10

5

10

6

10

7

10

8

PSOAT/DNA complexes


***

RL

U/m

g o

f P

ro

tein

10

3

10

4

10

5

10

6

10

7

10

8

RL

U/m

g o

f P

ro

te

in

10

3

10

4

10

5

10

6

10

7

10

8

RL

U/m

g o

f P

ro

te

in

N/P ratio

DNACells 5 10 20 30 Lipofectamine

N/P ratio


N/P ratio


***

****

PSOAT/DNA complexes


PSOAT/DNA complexes


Fig. 6. Luciferase activity of PSOAT/DNA (pGL3) complexes at various N/P ratios in different cell lines: (A) A549; (B) HeLa; and (C) H322 cells (n ¼ 3, error bar represents SD)(*p < 0.05; **p < 0.01, one-way ANOVA).


They found that COX-2 inhibition by SC58236 resulted in dramat-ically reduced organic osmolyte (such as sorbitol) uptake by cells[37,38]. This fruitful finding prompted us to use this inhibitor toblock polysorbitol activity in PSOAT and to show the necessity ofpolysorbitol backbone for its high transfection mechanism. Evalu-ation of the effect of polysorbitol activity inhibition on PSOAT-mediated gene transfection was carried out in A549 cells, asshown in Fig. 9A. Cells were incubated at 37 �C with 5% CO2 atdifferent concentrations (0, 5,10, and 30 mM/L) of SC58236 inhibitor(dissolved in DMSO and dilutedwith RPMI) for various time periods(1, 2, and 4 h). After each time point, transfection was performedaccording to the earlier procedure. The remarkable result showedthat PSOAT-mediated transfectionwas dramatically decreased withincrease of SC58236 concentrations from 0 to 30 mM/L in a dose-dependent manner at each time point. At the highest inhibitorconcentration of 30 mM/L, there were no significant differences intransfection efficiency between PSOAT/pGL3 complexes and nakedpGL3, suggesting that at a concentration as high as 30 mM/L ofinhibitor, the transfection ability of PSOAT was almost completelylost. On the other hand, SC58236 had no effect on PEI 25K-mediatedtransfection at any estimated time period, even at the highestinhibitor concentration of 30 mM/L, clearly demonstrating that

PSOAT provided a remarkably higher transfection efficiency due toits polysorbitol-based transporter mechanism, which shows thepotential of polysorbitol backbone in PSOAT and novelty of thissystem as a potent transporter for accelerated gene delivery.

3.6.2. Hyperosmotic activity of PSOATA hyperosmotic activity was reported to significantly improve

the transfection capability by increasing the cellular uptake of theosmotically active gene carriers. For measurement and estimationof the osmotic activity of gene delivery vehicles, the packed cellvolume (PCV) assay is a well reported method, which we havealready described in several earlier reports [17,39]. Since ourtransporter system has a polysorbitol backbone, we expected anincreased osmotic property, which might also be synergisticallyresponsible for increasing cellular uptake, thus transfection effi-ciency of PSOAT. Nevertheless, no previous study has reported thatthe polysorbitol moieties in gene delivery vehicles have any kind ofeffect in increasing cellular uptake, and thus increasing trans-fection. Therefore, to elucidate the mechanism behind theenhanced transfection by PSOAT, our second approach was todetermine the effect of osmotic activity of PSOAT/pGL3 complexesin A549 cells by analysis of the reduction in PCV when cells were

10

2

10

3

10

4

10

5

10

6

10

7

10

8

*****

****

Serum percentages

RL

U/m

g o

f P

ro

te

in

DNA Cells 0 10 20 30

PSOAT/DNA complexes


Lipofectamine 2000/DNA complexes

Fig. 7. Luciferase activity of PSOAT/DNA (pGL3) complexes in various serum percent-ages in A549 cells (n ¼ 3, error bar represents SD) (**p < 0.01; ***p < 0.001, one-wayANOVA).


treated with polyplexes. Cells were suspended in RPMI with 10%FBS and PSOAT/DNA complexes with varying polysorbitol content(0.01, 0.03, and 0.05%) in the polymer backbone; we carried out theexperiment using the protocol similar to the reports describedearlier [17,39]. As shown in Fig. 9B, PSOAT/DNA complexes withincreasing polysorbitol contents of 0.01, 0.03, and 0.05% reducedPCV by 2, 8, and 10%, respectively. The pure sorbitol was used asa control at the same amounts of polysorbitol concentrations as inPSOAT and provided 5, 9, and 12% reduction in PCV, respectively.This result suggests that PSOAT/DNA complexes with increasingpolysorbitol content were able to reduce PCV in a fashion similar tothat caused by the respective amount of pure sorbitol, demon-strating that PSOAT has a high osmotic activity which is synergis-tically able to improve the transfection efficiency by increasingcellular uptake of the carrier. In addition, about 22% reduction incell volume was achieved at 1% of pure sorbitol in RPMI, which wasused as a control sorbitol percentage.

3.6.3. Proton sponge-effect by LPEI backbone of PSOATAsaspecificvacuolar typeHþATPase inhibitor, bafilomycinA1was

previously reported to decrease PEI-mediated transfection by inhi-bition of endo-/lysosomal proton sponge activity [39]. To investigatethe mechanism of high transfection by PSOAT, our third and finalapproach was to elucidate the effect of bafilomycin A1 on PSOAT-mediated gene transfection, since our transporter possesses PEImoieties. The result showed that PSOAT-mediated transfection inA549 cells was highly sensitive to bafilomycin A1, as shown in Fig. 9C.After bafilomycin A1 treatment, the transfection efficiency by PSOAT/pGL3complexeswasdrasticallydecreasedbyabout40and30-foldsatN/P ratios of 20 and 30, respectively, undoubtedly suggesting theadvantage of the existence of PEI moieties in PSOAT, which wasrequired for acidification of endosomal vesicles, thereby allowinginflux of water and subsequent swelling and bursting of the endo-some, resulting in enhanced transfection by PSOAT. As expected, PEI25K/pGL3 complexes used as control also showed tremendoussensitivity toward bafilomycin A1. Therefore, the drastic reduction oftransfection ability by PSOAT in the presence of bafilomycin A1strongly suggests involvement of the proton sponge effect in PEI-mediated transfection and the benefits of the PEI moieties in the

PSOAT backbone, which is confirmed as the synergistic physico-chemical mechanism for their high transfection efficiency.

3.6.4. Superiority of polysorbitol over sorbitol itself in accelerationof PSOAT-mediated transfection

We postulated earlier that the polysorbitol backbone in ourPSOAT system having polyvalency would accelerate the trans-fection activity of the transporter more than the monovalentsorbitol by improving the cellular uptake through acceleratedtransporter mechanism. In order to prove this, we investigated thetransfection activity of PEI 25K/DNA complexes (optimum N/P ratioof 5) in the presence of 1, 3, 5, 10, and 20-folds higher sorbitolcontent compared to the polysorbitol amount in PSOAT/DNAcomplexes (N/P ratio of 20). The result showed no significantdifferences among the transfection efficiencies of PEI 25K in thepresence of various increased amounts of sorbitol as shown inFig. 9D. The transfection efficiency of PEI 25K didn’t increase even at20-folds higher sorbitol amount compared to the polysorbitolcontent in PSOAT, revealing that polysorbitol having polyvalency ismore efficient than sorbitol itself having monovalency in acceler-ation of gene transfer.

4. Discussion

Extensive research works have been carried out over the yearsto establish an effective carrier system, especially the non-viralcarriers for successful gene delivery [21e31]. However, all theattempts still remain at the laboratory stage due to the lacking ofa safe and effective gene delivery vector. In the present study, wedemonstrated a highly accelerated gene transporting vehicle, thepolysorbitol-based osmotically active transporter (PSOAT), whichshowed several remarkably promising features for their practicaluse to carry out a successful gene therapy. The most striking featureof our transporter is its polysorbitol backbone which facilitatedmost of the essential properties such as, the excellent condensationof DNA and protection of DNA from nucleases, the formation ofnano-sized transporter with spherical shapes and lower surfacecharges, the stability of the particles in the presence of variousserum percentages, and the ability to formulate dry powder afterlyophilization followed by DNA complexation, which are currentlyconsidered the canon of requirements in the gene delivery tech-nology. Dry powder formulation using our PSOAT system could bean exciting strategy in the polymeric gene delivery technology infuture to improve the efficacy of the gene delivery vectors.

We found that the PSOAT was significantly less cytotoxic since itpossessed polysorbitol backbone with many hydroxyl groups andLMW LPEI coupled with ester linkages. A number of PEI moleculeswith varying molecular weights (e.g. low or high) and structures(e.g. linear or branched) have been studied, and the LMW LPEI wasreported to be substantially less cytotoxic, compared with its highmolecular weight counterparts, as well as branched PEIs, whichwere less efficient and often toxic, especially at high N/P ratios, ascompared with the linear ones [14,40,41]. Until now, many groupshave revealed that introduction of hydroxyl groups to polycationicgene carriers aided in reduction of cytotoxicity [10,11]; thus, PSOATexhibited themselves as a safe gene transporter system owing to anincreased number of hydroxyl groups, which enhanced thebiocompatibility of the carriers, thus, resulting in less cytotoxicity.In addition, the degradability of polymers having ester-linkageswas reported as an important parameter for reduction of cytotox-icity due to formation of non-toxic degradation products appearingfrom the ester backbone [16,17]. Therefore, the combination of theless toxic LMW LPEI, the hydroxyl groups, and the degradable esterlinkages of PSOAT ensures a safe potential gene carrier systemwithout significant toxicity.

Fig. 8. In vivo tGFP expression analysis after aerosol administration to mice (C57BL/6) lungs: (A) western blot analysis of lung lysate (n ¼ 4, error bar represents SD) (*p < 0.05;**p < 0.01, one-way ANOVA) (B) fluorescent microscopic images and (C) immunohistochemistry analysis of the respective groups.


We precisely investigated the transfection ability of PSOATin vitro using different reporter genes such as determination ofluciferase activity and GFP expression. Although there was a defi-nite possibility of reduction of transfection efficiency due to manyhydroxyl groups containing polysorbitol backbone in PSOAT, anaccelerated in vitro gene transfer was achieved by PSOAT using boththe reporter genes (Fig. 6). Transfection was also carried out in theabsence and presence of serum and PSOAT showed stable trans-fection ability even in the presence of various serum percentages(Fig. 7).

The PSOAT/tGFP complexes were highly internalized by alveolarcells through an unclear transportermechanism at the large surfacearea of the pulmonary region, providing increased transfectionefficiency in vivo (Fig. 8). On the contrary, PEI 25K showed lowergene expression due to low cellular uptake by lung cells and alsoowing to its non-degradability, which caused cytotoxicity as well[39]. These remarkable in vivo results strongly suggest our in vitro

findings and undoubtedly provided evidence to support PSOAT asa highly efficient gene transporter system in vitro and in vivo, whichcan be considered for potential use as a gene delivery vehicle forsuccessful clinical application in the future. We found lower zetapotentials of PSOAT at high N/P ratios due to the shielding effect ofhydroxyl groups, which supposed to facilitate lower transfectionability of the transporter. According to our anticipation, however,PSOAT showed accelerated transfection efficiency both in vitro andin vivo even though it possessed many hydroxyl groups containingpolysorbitol backbone, revealing that PSOAT might have animproved transporter mechanism which accelerated the genetransfection.

The first mechanistic strategy for the accelerated PSOAT-mediated transfection was examined using SC58236, a COX-2-specific inhibitor which was reported as an inhibitor for osmo-lytes uptake by cells [37]. The transfection ability of PSOAT wasdrastically reduced in the presence of this inhibitor in a dose-

A

B C

D

Fig. 9. Mechanistic investigations for high transfection of PSOAT: (A) effect of polysorbitol activity inhibition (at 1, 2, and 4 h) by SC58236 on PSOAT-mediated gene transfection, (B)packed cell volume study, (C) bafilomycin A1 study, and (D) difference in transfection efficiency between polysorbitol backbone in PSOAT and sorbitol itself, in A549 cells (n ¼ 3,error bar represents SD) (*p < 0.05, **p < 0.01; ***p < 0.001, one-way ANOVA).


dependent manner, demonstrating the obvious requirement of thepolysorbitol backbone in PSOAT and the mechanism of their highlyaccelerated gene transfection (Fig. 9A). Through this inhibitionstudy, we are proposing an experimental procedure for genedelivery research by which the hindered mechanism of sorbitol-based transporter systems can be determined and explained. Thehighlight of the mechanism study demonstrated the potential ofPSOAT as an extremely promising gene transporter system.However, it is important to determine whether there is (are) anyspecific interaction (s) between the transporter and the cell surfacecomponent/receptor or any cellular signals that remarkablyenhanced the transfection capability of PSOAT by improvingcellular uptake. Nevertheless, activation of epidermal growth factorreceptor (EGFR) and mitogen-activated protein kinase (MAPK)signaling (especially p38 and ERK 1/2) in cells exposed to osmotic

stress that induced COX-2 expression has been previously reported.Under osmotic stimuli, EGFR can be activated by metal-loproteinases (MMPs) through MMP-dependent pro-EGFR ligand(such as EGF, HB-EGF, and TGF-a), which subsequently activatedEGFR and MAPK and was involved in induction of COX-2. Further-more, generation of reactive oxygen species (ROS) is essential forCOX-2 expression under osmotic pressure because MMPs areknown to be activated by ROS, through which, in turn, EGFR can beactivated. Thus, EGFR and associated protein represent amembraneassociated osmosensing complex for COX-2 expression. Moreover,activation of EGFR can occur through ligand dependent as well asindependent mechanisms [42e44]. Further, it is important tomention that production of COX-2, MMPs, and adhesion moleculesunder osmotic stimuli can induce and enhance cellular uptake bydegrading the joint surfacewhere the particles are deposited on the


cell membrane, leading to an efficient infiltration via endocytosisinto the cells [45], resulting in enhanced cellular uptake for hightransfection. Yang et.al reported that cells with higher COX-2expression had higher cellular uptake than those with lowerCOX-2 expression [46]. Several previous reports further revealedthat carcinoma cells displayed constitutive expression of COX-2,whereas COX-2 expression was virtually absent in normal cells,suggesting that our PSOAT system can specifically exert its activity,particularly against cancer cells, which might provide remarkableclinical benefits by delivery of any therapeutic gene (such asa therapeutic siRNA) using this transporter as a potent vehicle fortreatment of cancer [47e49]. In our several previous studies, wealready reported that the hyperosmotic effect exerted from thepolymeric carrier system facilitated efficient cellular uptake, and,thus, provided increased transfection efficiency [17,39]. Grosjeanet al. also reported that transient reporter gene expression in non-synchronized CHO cells was increased as the cell volume decreasedafter post-transfectional osmotic shock [50]. Therefore, the reduc-tion in packed cell volume by PSOAT/DNA complexes can beexplained by the fact that the polysorbitol backbone of PSOATexerted hyperosmotic pressure on the cell surface, which subse-quently labilies the vesicular membrane and improvised cellularuptake by improving membrane permeability (Fig. 9B). The PSOAT-mediated transfection was also found sensitive in the presence ofbafilomycin A1, suggesting the synergistic effect of PEI moieties in

Fig. 10. Schematic presentation of hypothetical mechanism for cellular internalization

PSOAT for improving their transfection ability. Nevertheless, it isimportant to observe that PSOAT accelerated the transfection effi-ciency about 5 and over 15-folds higher compared to PEI 25K(Fig. 9C), even in the presence of bafilomycin A1 at N/P ratios of 20and 30, respectively, suggesting the superiority of polysorbitolbackbone than the PEI moieties in PSOAT for enhancing cellularuptake of the transporter, hence gene expression. Therefore, thisresult confirms the hypothesis that cellular uptake is more impor-tant than endosomal escape to achieve high transfection [6].

Taking together of these mechanism studies, we schematicallyrepresented the hypothetical mechanism of PSOAT for their accel-erated transfection as shown in Fig. 10. We anticipate that the cellshaving sorbitol transporting channels (STC) on their membraneselectively transport sorbitol to maintain osmotic balance [51].However, these STC are not able to transport the polysorbitolcontaining PSOAT/DNA complexes when they interact to cellmembrane owing to size limitations which prevents their trans-portation through STC and interfere with the normal condition formaintenance of osmotic balance. At this situation, the complexesmight stuck on the STC and inactivate them and create a hyper-osmotic environment. This interference with the normal osmoticbalance might lead the cells to form endosome to remove theinactivated STC via rapid endocytosis whichmight be influenced bythe induction of COX-2 expression through an unknown signalingpathway in relation with hyperosmotic condition. After

of PSOAT/DNA complexes (figures not representative of the scale of molecules).

Fig. 11. Speculative representation on the effect of (A) sorbitol itself, and (B) polysorbitol backbone in PSOAT for DNA internalization. (C) Effect of SC58236 inhibitor on PSOAT-mediated DNA internalization (figures not representative of the scale of molecules).


endocytosis, LPEI comes into the act with their proton sponge effectthat leads the endosome swelling which eventually burst beforelysosomal fusion. Finally, DNA release from the complexes afterpolymer degradation and ready to be functioned. Furthermore, weshowed the remarkable benefits of polysorbitol backbone in PSOAThaving polyvalency over the monovalent sorbitol for improvisationof the transfection activity of the transporter as shown in Fig. 9D.We found that the transfection activity of PEI 25K didn’t increaseeven after addition of 20-folds higher sorbitol amount compared tothe polysorbitol content in PSOAT which can be schematicallyrepresented as shown in Fig. 11A. We speculate that the addition ofsorbitol to PEI 25K/DNA complexes even at higher concentrationsdidn’t make any effect on the complex internalization other thanincrease the sorbitol concentrations in cells simply transportingthrough the STC for maintenance the balance of hyperosmoticenvironment. Hence, no additional DNA internalization occurred byPEI 25K-mediated transfection even when higher sorbitol amountswere added. In contrast, Fig. 11B shows the accelerated DNAinternalization mechanism by PSOAT/DNA complexes which mightbe influenced by COX-2 expression as explained schematically inFig. 10. DNA internalization through PSOAT-mediated transfectionis drastically prevented by using SC58236, a COX-2-specific inhib-itor which might inhibit COX-2 expression that cease the osmoticcontrol function in cells, specifically prevents the accumulation ofPSOAT/DNA complexes on the cell membrane, leading to a reducedPSOAT-mediated transfection as shown schematically in Fig. 11C,

demonstrating the most striking mechanism of our PSOAT systeminvolved in acceleration of gene transfer. So far, no other genetransporter has been reported that possesses all the describedattributes as we documented in the present study, making ourPSOAT system remarkably effective for accelerated genetransportation.

5. Conclusion

We have successfully developed a polysorbitol-based osmoti-cally active transporter (PSOAT) which showed markedly enhancedtransfection activity with reduced cytotoxicity. A sharp and strongDNA condensation ability, lower but positive surface charges ofPSOAT/DNA complexes at high N/P ratios, essentially surprisingstability of the nano-sized polyplexes after lyophilization, whichare among themost important and urgent issues in the field of genetherapy technology, have made PSOAT a promising gene carriersystem. In addition, the stability of the polyplexes in the presence ofnegatively charged serum proteins at various concentrations couldbe helpful for further development of gene delivery vectors in vivoand for their clinical trials as well. Despite of having many hydroxylgroups containing polysorbitol backbone, the PSOAT showedaccelerated transfection ability by its improved transporter mech-anisms, which perhaps the most striking features of our PSOATsystem. At this moment, it is not confirmed which specificmembrane functions of cells in relation to the structural features of


our transporter contributed to the observed mechanisms, specifi-cally the inhibition study using a COX-2-specific inhibitor. Wespeculate that PSOATmight have some specific structural affinity inrelation to the cellular membrane and that it might be regulated bya signaling pathway in relation with the induction of COX-2expression, which might enhance the cellular uptake of the trans-porter. Further studies are aimed at elucidation of the cellularsignaling mechanism with regard to how PSOAT works as a highlyefficient gene transporter system and exploration of its therapeuticapplications using siRNA, and are ongoing.

Acknowledgements

This work was supported in part by the R&D Program of MKE/KEIT (10035333, Development of anti-cancer therapeutic agentbased on regulating cell cycle or cell death) as well as by theNational Research Foundation (NRF-2010-0000784), Ministry ofEducation, Science and Technology (MEST) in Korea. This work wasalso supported partially by the National Research Foundation ofKorea Grant funded by the Korean Government (MEST) (NRF-2011-000380). M. A. Islam and J. Y. Shin were supported by the BK21program. We acknowledge the National Instrumental Centre forEnvironmental Management (NICEM), and the support from theVeterinary Research Institute of Seoul National University, Korea.

Appendix. Supplementary information

Supplementary information related to this article can be foundonline at doi:10.1016/j.biomaterials.2011.09.013

References

[1] Putnam D. Polymers for gene delivery across length scales. Nat Mater 2006;5:439e51.

[2] Somia N, Verma IM. Gene therapy: trials and tribulations. Nat Rev Genet 2000;1:91e9.

[3] Anderson DG, Lynn DM, Langer R. Semi-automated synthesis and screening ofa large library of degradable cationic polymers for gene delivery. Angew ChemInt Ed Engl 2003;42:3153e8.

[4] Liu Z, Winters M, Holodniy M, Dai H. siRNA delivery into human T cells andprimary cells with carbon-nanotube transporters. Angew Chem Int Ed Engl2007;46:2023e7.

[5] Maiti KK, Lee WS, Takeuchi T, Watkins C, Fretz M, Kim DC, et al. Guanidine-containing molecular transporters: sorbitol-based transporters show highintracellular selectivity toward mitochondria. Angew Chem Int Ed Engl 2007;46:5880e4.

[6] Higashi T, Khalil IA, Maiti KK, Lee WS, Akita H, Harashima H, et al. Novellipidated sorbitol-based molecular transporters for non-viral gene delivery.J Control Release 2009;136:140e7.

[7] Wadia JS, Stan RV, Dowdy SF. Transducible TAT-HA fusogenic peptideenhances escape of TAT-fusion proteins after lipid raft macropinocytosis. NatMed 2004;10:310e5.

[8] Torchilin VP, Rammohan R, Weissig V, Levchenko TS. TAT peptide on thesurface of liposomes affords their efficient intracellular delivery even at lowtemperature and in the presence of metabolic inhibitors. Proc Natl Acad SciUSA 2001;98:8786e91.

[9] Kogure K, Moriguchi R, Sasaki K, Ueno M, Futaki S, Harashima H. Developmentof a non-viral multifunctional envelope-type nano device by a novel lipid filmhydration method. J Control Release 2004;98:317e23.

[10] Xu FJ, Li H, Li J, Zhang Z, Kang ET, Neoh KG. Pentablock copolymers ofpoly(ethylene glycol), poly((2-dimethyl amino)ethyl methacrylate) andpoly(2-hydroxyethyl methacrylate) from consecutive atom transfer radicalpolymerizations for non-viral gene delivery. Biomaterials 2008;29:3023e33.

[11] Ma M, Li F, Yuan ZF, Zhuo RX. Influence of hydroxyl groups on the biologicalproperties of cationic polymethacrylates as gene vectors. Acta Biomater 2010;6:2658e65.

[12] Park MR, Han KO, Han IK, Cho MH, Nah JW, Choi YJ, et al. Degradablepolyethylenimine-alt-poly(ethylene glycol) copolymers as novel genecarriers. J Control Release 2005;105:367e80.

[13] Breunig M, Lungwitz U, Liebl R, Fontanari C, Klar J, Kurtz A, et al. Gene deliverywith low molecular weight linear polyethylenimines. J Gene Med 2005;7:1287e98.

[14] Wightman L, Kircheis R, Rössler V, Carotta S, Ruzicka R, Kursa M, et al.Different behavior of branched and linear polyethylenimine for gene deliveryin vitro and in vivo. J Gene Med 2001;3:362e72.

[15] Mather BD, Viswanathan K, Miller KM, Long TE. Michael addition reactions inmacromolecular design for emerging technologies. Prog Polym Sci 2006;31:487e531.

[16] Lynn DM, Langer R. Degradable poly (b-amino esters):synthesis, character-ization, and self-Assembly with plasmid DNA. J Am Chem Soc 2000;122:10761e8.

[17] Arote RB, Hwang SK, Yoo MK, Jere D, Jiang HL, Kim YK, et al. Biodegradablepoly (ester amine) based on glycerol dimethacrylate and polyethylenimine asa gene carrier. J Gene Med 2008;10:1223e35.

[18] Arote R, Kim TH, Kim YK, Hwang SK, Jiang HL, Song HH, et al. A biodegradablepoly(ester amine) based on polycaprolactone and polyethylenimine as a genecarrier. Biomaterials 2007;28:735e44.

[19] Forrest ML, Koerber JT, Pack DW. A degradable polyethylenimine derivativewith low toxicity for highly efficient gene delivery. Bioconjug Chem 2003;14:934e40.

[20] Fischer D, Bieber T, Li Y, Elsässer HP, Kissel T. A novel non-viral vector for DNAdelivery based on low molecular weight, branched polyethylenimine: effect ofmolecular weight on transfection efficiency and cytotoxicity. Pharm Res 1999;16:1273e9.

[21] Amiji MM. Polymeric gene delivery: principles and applications. Florida: CRCBoca Raton; 2005. Ch. 14.

[22] Jiang HL, Kim YK, Arote R, Nah JW, Cho MH, Choi YJ, et al. Chitosan-graft-polyethylenimine as a gene carrier. J Control Release 2007;117:273e80.

[23] Liu Y, Reineke TM. Hydroxyl stereochemistry amine number within poly(-glycoamidoamine)s affect intracellular DNA delivery. J Am Chem Soc 2005;127:3004e15.

[24] Mislick KA, Baldeschwieler JD. Evidence for the role of proteoglycans incation-mediated gene transfer. Proc Natl Acad Sci USA 1996;93:12349e54.

[25] Wagner E, Ogris M, Zauner W. Polylysine-based transfection systems utilizingreceptor-mediated delivery. Adv Drug Deliv Rev 1998;30:97e113.

[26] Laube BL. The expanding role of aerosols in systemic drug delivery, genetherapy, and vaccination. Respir Care 2005;50:1161e76.

[27] Li HY, Seville PC, Williamson IJ, Birchall JC. The use of amino acids to enhancethe aerosolisation of spray-dried powders for pulmonary gene therapy. J GeneMed 2005;7:343e53.

[28] Seville PC, Kellaway IW, Birchall JC. Preparation of dry powder dispersions fornon-viral gene delivery by freeze-drying and spray-drying. J Gene Med 2002;4:428e37.

[29] Li HY, Birchall J. Chitosan-modified dry powder formulations for pulmonarygene delivery. Pharm Res 2006;23:941e50.

[30] Lynn DM, Anderson DG, Putnam D, Langer R. Accelerated discovery ofsynthetic transfection vectors: parallel synthesis and screening of a degrad-able polymer library. Am Chem Soc 2001;123:8155e6.

[31] Akinc A, Lynn DM, Anderson DG, Langer R. Parallel synthesis and biophysicalcharacterization of a degradable polymer library for gene delivery. J Am ChemSoc 2003;125:5316e23.

[32] Ryser HJ. A membrane effect of basic polymers dependent on molecular size.Nature 1967;215:934e6.

[33] Goldman CK, Soroceanu L, Smith N, Gillespie GY, Shaw W, Burgess S, et al.In vitro and in vivo gene delivery mediated by a synthetic polycationic aminopolymer. Nat Biotechnol 1997;15:462e6.

[34] Sato T, Ishii T, Okahata Y. In vitro gene delivery mediated by chitosan: effect ofpH, serum, and molecular mass of chitosan on the transfection efficiency.Biomaterials 2001;22:2075e80.

[35] Koshkina NV, Agoulnik IY, Melton SL, Densmore CL, Knight V. Biodistributionand pharmacokinetics of aerosol and intravenously administered DNA-polyethyleneimine complexes: optimization of pulmonary delivery andretention. Mol Ther 2003;8:249e54.

[36] Rudolph C, Schillinger U, Ortiz A, Plank C, Golas MM, Sander B, et al. Aero-solized nanogram quantities of plasmid DNA mediate highly efficient genedelivery to mouse airway epithelium. Mol Ther 2005;12:493e501.

[37] Moeckel GW, Zhang L, Fogo AB, Hao CM, Pozzi A, Breyer MD. COX2 activitypromotes organic osmolyte accumulation and adaptation of renal medullaryinterstitial cells to hypertonic stress. J Biol Chem 2003;278:19352e7.

[38] Yang T, Huang Y, Heasley LE, Berl T, Schnermann JB, Briggs JP. MAPK medi-ation of hypertonicity-stimulated cyclooxygenase-2 expression in renalmedullary collecting duct cells. J Biol Chem 2000;275:23281e6.

[39] Arote RB, Lee ES, Jiang HL, Kim YK, Choi YJ, Cho MH, et al. Efficient genedelivery with osmotically active and hyperbranched poly(ester amine)s. Bio-conjug Chem 2009;20:2231e41.

[40] Bragonzi A, Dina G, Villa A, Calori G, Biffi A, Bordignon C, et al. Biodistributionand transgene expression with nonviral cationic vector/DNA complexes in thelungs. Gene Ther 2000;7:1753e60.

[41] Poulain L, Ziller C, Muller CD, Erbacher P, Bettinger T, Rodier JF, et al. Ovariancarcinoma cells are effectively transfected by polyethylenimine (PEI) deriva-tives. Cancer Gene Ther 2000;7:644e52.

[42] Küper C, Bartels H, Fraek ML, Beck FX, Neuhofer W. Ectodomain shedding ofpro-TGF-alpha is required for COX-2 induction and cell survival in renalmedullary cells exposed to osmotic stress. Am J Physiol Cell Physiol 2007;293:C1971e82.

[43] Küper C, Steinert D, Fraek ML, Beck FX, Neuhofer W. EGF receptor signaling isinvolved in expression of osmoprotective TonEBP target gene aldose reduc-tase under hypertonic conditions. Am J Physiol Renal Physiol 2009;296:F1100e8.



[44] Zhao H, Tian W, Tai C, Cohen DM. Hypertonic induction of COX-2 expressionin renal medullary epithelial cells requires transactivation of the EGFR. Am JPhysiol Renal Physiol 2003;285:F281e8.

[45] O’Neill KD, Chen NX, Wang M, Cocklin R, Zhang Y, Moe SM. Cellular uptake ofbeta2M and AGE-beta2M in synovial fibroblasts and macrophages. NephrolDial Transplant 2003;18:46e53.

[46] Yang DJ, Bryant J, Chang JY, Mendez R, Oh CS, Yu DF, et al. Assessment ofcyclooxygense-2 expression with 99mTc-labeled celebrex. Anticancer Drugs2004;15:255e63.

[47] Dvory-Sobol H, Cohen-Noyman E, Kazanov D, Figer A, Birkenfeld S, Madar-Shapiro L, et al. Celecoxib leads to G2/M arrest by induction of p21 and down-regulation of cyclin B1 expression in a p53-independent manner. Eur J Cancer2006;42:422e6.

[48] Finzi L, Shao MX, Paye F, Housset C, Nadel JA. Lipopolysaccharide initiatesa positive feedback of epidermal growth factor receptor signaling by pros-taglandin E2 in human biliary carcinoma cells. J Immunol 2009;182:2269e76.

[49] Subbaramaiah K, Altorki N, Chung WJ, Mestre JR, Sampat A, Dannenberg AJ.Inhibition of cyclooxygenase-2 gene expression by p53. J Biol Chem 1999;274:10911e5.

[50] Grosjean F, Bertschinger M, Hacker DL, Wurm FM. Multiple glycerol shocksincrease the calcium phosphate transfection of non-synchronized CHO cells.Biotechnol Lett 2006;28:1827e33.

[51] Rolf KHK, Helmut K, Birgit R, Frank W, Stefan HB, Evamaria KS. Organicosmolyte channels in the renal medulla: their properties and regulation. AmerZool 2001;41:728e33.

Accelerated gene transfer through a polysorbitol-based transporter mechanism

Documents

Transcript of Accelerated gene transfer through a polysorbitol-based transporter mechanism