Hyperbranched cationic amylopectin derivatives for gene delivery

Hyperbranched cationic amylopectin derivatives for gene delivery

Q1 Yanfang Zhou a,c,2, Bin Yang b,2, Xianyue Ren a, Zhenzhen Liu b, Zheng Deng b, Luming Chen d,Yubin Deng a,1,**, Li-Ming Zhang b, Liqun Yang b,1,*

aDepartment of Pathophysiology, Zhongshan School of Medicine, Sun Yat-sen University, Guangzhou 510080, Chinab Institute of Polymer Science, School of Chemistry and Chemical Engineering, DSAPM Lab and PCFM Lab, Sun Yat-sen University, Guangzhou 510275, ChinacDepartment of Pathophysiology, Guangdong Medical College, Dongguan 523808, Chinad Yale University, New Haven 06511,USA

a r t i c l e i n f o

Article history:Received 21 January 2012Accepted 4 March 2012Available online xxx

Keywords:AmylopectinOligoamineGene deliveryTransfection efficiency

a b s t r a c t

A series of hyperbranched cationic amylopectin derivatives conjugated with 1,2-ethylenediamine,diethylenetriamine and 3-(dimethylamino)-1-propylamine residues, named as EDA-Amp, DETA-Ampand DMAPA-Amp, were synthesized by the N,N’-carbonyldiimidazole activation method at roomtemperature. Their structures were characterized by FTIR and 1H NMR analyses, and their bufferingcapability was assessed by acid-base titration. The amylopectin derivatives exhibited better bloodcompatibility and lower cytotoxicity when compared to branched polyethyleneimine (bPEI) in thehemolysis and MTT assays. Atomic force microscopy and optical microscopy confirmed that theamylopectin derivatives exhibited lower damage for erythrocytes than bPEI. The amylopectin derivativescould bind and condense plasmid DNA (pDNA) to form the complexes with the size ranging from 100 to300 nm. The resultant complexes showed higher transfection efficiency in 293T cells than in A549 cells.The DMAPA-Amp derivative-mediated gene transfection for Forkhead box O1 exhibited higher proteinexpression than that of the EDA-Amp and DETA-Amp derivatives in 293T cells, which was analyzed bywestern blot, flow cytometry and Hoechst staining assay. On the basis of these data, amylopectinderivatives exhibit potential as nonviral gene vectors.

� 2012 Published by Elsevier Ltd.

1. Introduction

Successful gene therapy relies on the development of genedelivery vectors with high transfection efficiency and minimalcytotoxicity [1]. Nonviral gene delivery vectors, especially cationicpolymers and liposomes, have attractedmuch attention in the genetherapy. Unlike viral vectors, these nonviral vectors presentadvantages such as low immune response, capacity to deliver largeDNAmolecules, and large-scale production at low cost [2,3]. Recentwork has established that the architecture of cationic polymersaffects the stability, delivery and transfection efficiency of thecomplex with DNA [4e6]. In contrast to linear cationic polymers,hyperbranched cationic polymers exhibit higher gene expressiondue to presence of compact and globular structures in combinationwith a great number of various amine groups, which are

advantageous for the binding and condensation of negativelycharged DNA into compact particles [5e10]. However, the synthetichyperbranched cationic polymers, such as branched poly-ethyleneimine (bPEI) and poly(amidomine) (PAMAM) dendrimer,are still associated with problems of cytotoxicity, biocompatibilityand biodegradability, which need to be overcome for in vivoapplication [11,12].

Natural polysaccharides, due to their outstanding nontoxicity,good biocompatibility and biodegradability, have been the recipi-ents of increasing attention in the biomaterial field [13]. Followingamination, various cationized polysaccharides have been designedand synthesized as gene delivery vectors with high transfectionefficiency, e.g., cationized cycloamylose [4], pullulan [14,15],schizophyllan [16], pectin [17], dextran [18], chitosan [19], andcellulose [20]. Possible acceleration of cell internalization by cellsurface sugar-recognition receptors, low cytotoxicity and immu-nogenicity, good biodegradability, and high water solubility are alladvantages of cationic polysaccharides-based vectors over othercationic polymers [14e17]. As a result of these advantages,polysaccharide-based hyperbranched polycations are currentlyemerging as a generation of nonviral gene delivery vectors [6].

* Corresponding author. Tel.: þ86 20 84110934; fax: þ86 20 84112245.** Corresponding author. Tel.: þ86 20 87331698; fax: þ86 20 37616451.

E-mail addresses: [email protected] (Y. Deng), [email protected](L. Yang).

1 Both corresponding authors contributed equally to this work.2 Both authors contributed equally to this work.

Contents lists available at SciVerse ScienceDirect

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

0142-9612/$ e see front matter � 2012 Published by Elsevier Ltd.doi:10.1016/j.biomaterials.2012.03.014

Biomaterials xxx (2012) 1e10

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

JBMT13897_proof ■ 13 March 2012 ■ 1/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

Amylopectin, found in starch granules with amylose, is a naturalhyperbranched polysaccharide. The hyperbranched structure ofamylopectin is illustrated in Scheme 1. Approximately every 20e25a-D-(1/4) glucose units of the linear chain of amylopectin areinterlinked by a-D-(1/6) glycosidic linkage, forming the branch-on-branch topological structure [21]. The hydroxyl groups presenton amylopectin are available for simple chemical modificationssuch as amination. In this work, the hyperbranched cationicamylopectin derivatives conjugated with various oligoamine resi-dues were assessed for their potential as nonviral gene deliveryvectors, and they were synthesized by the N,N’-carbon-yldiimidazole (CDI) activation method at room temperature(Scheme 1). The blood compatibility of these derivatives wasquantified by the hemolysis test, and the erythrocyte damagecaused by the amylopectin derivatives was investigated by atomicforce microscopy (AFM) and optical microscopy. The bufferingcapacity of the amylopectin derivatives, the size and zeta potentialof the particles prepared from the cationic amylopectin derivativesand plamid DNA (pDNA) were characterized. The cytotoxicity of theamylopectin derivatives and their pDNA complexes in the 293T(Human embryonic kidney) cell line and the A549 (human lungcancer) cell line was evaluated using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay.The pDNA transfection efficiency mediated by the cationic amylo-pectin derivatives was evaluated in the 293T and A549 cell lines.The transfection induced protein expression of Forkhead box O1(FoxO1) and cell apoptosis mediated by the cationic amylopectinderivatives was further assessed in the 293T cell line.

2. Materials and methods

2.1. Materials

Amylopectin (from maize) and bPEI were purchased from SigmaeAldrich(Missouri, USA), of which the weight average molecular weight (Mw) values weredetermined to be 3.1�107 and 2.5�104 g/mol by static light scattering, respectively.CDI, 1,2-ethylenediamine (EDA), diethylenetriamine (DETA), 3-(dimethylamino)-1-propylamine (DMAPA) and dimethyl sulfoxide (DMSO) were acquired fromAladdin Reagent Company (Shanghai, China). DMSO was dried by soaking in

molecular sieves and calcium hydride for a week before used. Fetal bovine serum(FBS), dulbecco’s modified eagle’s medium (DMEM), penicillin-streptomycin, MTTand trypsin were purchased from Gibco Co. (Carlsbad, CA, USA). The reporter pDNA,plamid EGFP-N1, was supplied by Promega Co. (Madison, WI, USA). Triton X-100 wasobtained from Sigma Chemical Co. (St. Louis, MO, USA). FoxO1 was purchased fromCell Signaling Technology (Danvers, MA, USA). Horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (1:3000 dilution) was procured from Cell SignalingTechnology (Danvers, MA, USA). All other reagentswere of analytical grade andwereused without further purification.

2.2. Synthesis and structural characterization of the hyperbranched cationicamylopectin derivatives conjugated with oligoamine residues

Various oligoamine compounds were introduced to the hydroxyl groups ofamylopectin by the CDI activation method [4,14]. Amylopectin (0.1000 g, 0.62 mmolglucose units) was dissolved in 10mL of water-free DMSO, and thenwas activated byadding CDI (0.9045 g, 5.58 mmol) and stirred for 1 h in a nitrogen atmosphere atroom temperature. EDA (2.2357 g, 37.2 mmol), DETA (3.8379 g, 37.2 mmol) andDMAPA (3.8011 g, 37.2 mmol) were added to the amylopectin reaction solution inthe order listed. The reaction was allowed to proceed for 24 h in a nitrogen atmo-sphere at room temperature. The reaction solutionwas dialyzed against the distilledwater in a dialysis bag (molecular weight cutoff 14,000) for 3 days, and lyophilized toyield the solid products named as EDA-Amp, DETA-Amp and DMAPA-Amp,respectively.

FTIR measurement was performed with an FTIR Analyzer (Nicolet/Nexus 670,Thermo Nicolet Corporation,Wisconsin, USA) at a resolution of 4 cm�1 using the KBrmethod. 1H NMR analysis was carried out on an NMR Spectrometer (Mercury-Plus300, Varian, USA) at 50 �C using D2O as a solvent. The signal at d 4.50 ppm for HDOwas used as the internal standard [22]. The degree of substitution (DS) of oligoamineresidues on the amylopectin, which is defined as the number of oligoamine residuesper glucose unit of amylopectin, was determined by the integration of H1 ofamylopectin and protons of oligoamine residues. EDA-Amp: 1H NMR (D2O, ppm):d ¼ 5.40 (glucose unit, H1); d ¼ 4.50e3.50 (glucose unit, H2-H6); d ¼ 3.23(eCONHeCH2e); d ¼ 2.79 (eCH2eNH2). DS ¼ 0.88.

DETA-Amp: 1H NMR (D2O, ppm): d ¼ 5.40 (glucose unit, H1); d ¼ 4.50e3.50(glucose unit, H2-H6); d ¼ 3.25 (eCONHeCH2e); d ¼ 2.73 (eCH2eNH2).DS ¼ 1.00. DMAPA-Amp: 1H NMR (D2O, ppm): d ¼ 5.40 (glucose unit, H1);d ¼ 4.50e3.50 (glucose unit, H2-H6); d ¼ 3.16 (eCONHeCH2e); d ¼ 2.36(eCH2eN<); d ¼ 2.21 [eN(CH3)2]. DS ¼ 2.88.

2.3. Buffer capacity

The buffer capacity of the cationic amylopectin derivatives and bPEI wasdetermined by acid-base titration in accordance with the literature [2,19]. Eachsample solution (0.2 mg/mL) was prepared in 30 mL of aqueous NaCl solution

Scheme 1. Synthesis of the hyperbranched cationic amylopectin derivatives conjugated with oligoamine residues: (a) chemical reaction and (b) proposed structures of amylopctinand its hyperbranched cationic derivatives.

Y. Zhou et al. / Biomaterials xxx (2012) 1e102

111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175

176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240

JBMT13897_proof ■ 13 March 2012 ■ 2/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

(0.15 mol/mL). The sample solutions were firstly titrated by a 0.1 mol/mL NaOHsolution to a pH of 10.0, and then a 0.1 mol/mL HCl solutionwith particular volumeswere added to obtain the mixture solutions with different pH values, which werethen measured by a microprocessor pH meter (Shanghai Shengke Instrument andEquipment Co., China).

2.4. Hemolysis assay

The hemolytic activity of the cationic amylopectin derivatives and bPEI wasinvestigated according to previous work [23]. SpragueeDawley rats (weighting from100 to 120 g) were provided by the Center of Experimental Animals of Sun Yat-senUniversity and approved by the University Animal Study Committee. The erythro-cytes were isolated from blood of SpragueeDawley rats by centrifugation at1000 rpm and were washed in 0.9% NaCl solution until the supernatant was clear.Then the erythrocytes were diluted in 0.9% NaCl solution to 5 � 109 cells/mL. Theobtained suspension (100 mL) was incubated with 1 mL of the solution of amylo-pectin derivatives and bPEI at different concentrations (31e500 mg/mL) at 37 �C for90 min, respectively. Following centrifugation at 1000 rpm for 5 min, the super-natant was applied to a 96-well plate, and hemoglobin release was measured as theabsorbance at 560 nm using a microplate reader (Wellscan MK3, LabsystemsDragon, Helsinki, Finland). A negative control solution (0% hemolysis) was preparedby adding 0.9% NaCl solution to the erythrocyte suspension, and a positive controlsolution (100% hemolysis) was prepared by adding 10% Triton X-100 to the eryth-rocyte suspension. The hemolysis rate was calculated as: Hemolysis rate(%) ¼ (A � A0)/(A100 � A0) � 100, where A, A0 and A100 represent the absorbancevalues of the hemoglobin released solution incubated with the polymers and thevalues of the negative and positive control solutions. The samples were analyzed intriplicate.

2.5. Morphology of erythrocytes

The morphology of erythrocytes, incubated with the amylopectin derivativesand bPEI (500 mg/mL) as mentioned above, was observed on an Olympus IX71microscope (Melville, NY, USA). For AFM observation, erythrocytes were incubatedwith 125 mg/mL cationic amylopectin derivatives and bPEI at 37 �C under gentleshaking, fixed by addition of 1% glutaraldehyde, and then 25 mL of each sample wasapplied a corresponding microscope slide. After air drying, the samples were gentlyrinsed with deionized water three times and then air dried again before analysis. AllAFM studies were carried out with an atomic force microscope (Autoprobe CPResearch, Thermomicroscopes, USA). The mean surface roughness of the obtainedimages from erythrocytes surfaces was analyzed.

2.6. Preparation of the cationic amylopectin derivative/pDNA complexes

The cationic amylopectin derivatives (10 mg) were dissolved in phosphatebuffered saline solution (PBS, pH 7.4) with a concentration of 2 mg/mL and thenfiltered through membrane filters (nominal pore sized 0.22 mm). The pDNA stocksolution (250 ng/mL) was prepared in PBS (pH 7.4). The pDNA solution (4 mL) wasthen added to the amylopectin derivative solution at various amylopectin derivative/pDNA weight ratios, followed by gentle agitation using a vortex agitator for 5 s andincubation at 37 �C for 30 min before use.

2.7. Agarose gel electrophoresis

Electrophoresis was performed to assess pDNA condensation ability of theamylopectin derivatives and protection against DNase I degradation. DNA wasvisualized and photographed using a Gel Doc XR gel image machine (Syngene,Cambridge, UK).

2.7.1. pDNA condensation ability of the amylopectin derivativesTwenty microliter of amylopectin derivative/pDNA complexes with different

amylopectin derivative/pDNAweight ratios ranging from 0.2 to 80 were loaded onto0.8% agarose gels with ethidium bromide (0.1 mg/mL) and runwith Tris-acetate (TAE)running buffer at 90 V for 1 h. DNA retardation was then observed by agarose gelelectrophoresis.

2.7.2. Protection against DNase I degradationThe method used to measure DNase I protection was based on that of previous

work [24]. DNase I (10 units, 2 mL) was added to 1 mg of naked plasmid DNA oramylopectin derivative/pDNA complexes (weight ratios 5, 10, and 20) and incubatedat 37 �C while shaking at 100 rpm for 30min. Subsequently, an EDTA (4 mL, 250 mM)and sodium dodecyl sulfate (SDS) solution (4 mL, 10%, w/v) was added and themixture was incubated at room temperature for 1 h. The samples were loaded ontothe gel and electrophoresed to examine the integrity of DNA.

2.8. Measurements of particle size and zeta potential

The amylopectin derivative/DNA complexes (2 mL, weight ratios 5, 10, and 20)were prepared in the distilled water and with a final concentration of 1 mg/mL for

Fig. 1. (A) FTIR spectra of (a) Amylopectin, (b) EDA-Amp, (c) DETA-Amp and (d)DMAPA-Amp; (B) 1H NMR spectra of (a) EDA-Amp, (b) DETA-Amp and (c) DMAPA-Ampin D2O.

Y. Zhou et al. / Biomaterials xxx (2012) 1e10 3

241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305

306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370

JBMT13897_proof ■ 13 March 2012 ■ 3/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

plasmid DNA. The sample mixtures were gently agitated by a vortex agitator for 5 s,incubated at 37 �C for 30 min, and then filtered through membrane filters (nominalpore sized 0.45 mm). Particle sizes and zeta potentials of amylopectin derivative/pDNA complexes were measured using ZetaPALS (Brookhaven Instruments Corpo-ration) at 25 �C with 90� scattering angle. Each measurement was performed intriplicate.

2.9. Scanning electron microscopy (SEM) observation

The solutions of amylopectin derivative/pDNA complexes (0.1 mL, w/w ¼ 20)were dropped onto a tinfoil, and then dried at room temperature. Themorphology ofthe complexes was observed on an S-4800 scanning electron microscope (HI-9056-0003, Hitachi, Japan), after the samples were sputter-coated with gold in an E-1045ion sputter (Hitachi, Japan).

2.10. Cytotoxicity assay

Cells from the 293T and A549 cell lines293T and A549 cells were seeded in 96-well plates at a density of 1�104 cells/well in 200 mL DMEM containing 10% FBS andcultured at 37 �C for 24 h. After the medium was replaced by fresh serum-freemedium, solutions of the amylopectin derivatives with different concentrationswere added to the culture medium of each well to assess their cytotoxicity. Inaddition, 200 mL of amylopectin derivative/pDNA complexes with different weightratio were added for the cytotoxicity assay, and the content of pDNA was 0.2 mg foreach complex. The bPEI and the bPEI/pDNA complex were used as the positivecontrol. After cells were incubated for an additional 24 h, 20 mL of MTT (5 mg/mL) inPBS solution (pH7.4) was added to each well, yielding a final concentration of0.5 mg/mL. The cells were further incubated for 4 h, and 150 mL of DMSO was addedto dissolve the formazan crystals formed in live cells. Using a spectrometer, theabsorbance values of the samples were measured at 560 nm. Cell viability (%) wascalculated according to the following equation: cell viability (%) ¼ (A560-sample/A560-control) � 100, where A560-sample and A560-control were obtained in the presenceand absence of the amylopectin derivatives, respectively. All experiments wereconducted in five times.

2.11. In vitro transfection

Cells from the 293T and A549 cell lines were plated in 24-well plates at 1 �104

cells/well, and were incubated for 12 h under the same conditions mentioned above.The media were replaced with serum-free media containing the amylopectinderivative/pDNA complexes and the bPEI/pDNA complex at different weight ratios,2 mg of pDNA. The bPEI/pDNA complex was used as the positive control and nakedpDNA was used as the negative control. After incubation for 4 h, the serum-freemedia were changed with fresh media containing 10% serum. After further incu-bation for 48 h, cells were directly observed on an Olympus IX71 fluorescencemicroscope. Subsequently, the transfected cells were washed once with PBS (pH7.4)and detached with 0.25% trypsin. Using an FACS Aria flow cytometer (BD Biosci-ences, Sparks, MD, USA), the gene transfection efficiency was evaluated by scoringthe percentage of cells expressing green fluorescence protein (GFP).

2.11.1. Incubation of 293T cells with the polymer/FoxO1 complexesCells from the 293Tcell linewere plated in 6-well plates at 1�105 cells/well, and

were incubated for 24 h under the same conditions mentioned above. The mediawere then replaced with serum-free media containing the amylopectin derivative/FoxO1 complexes (w/w ¼ 10) and the bPEI/FoxO1 complex (N/P ¼ 10), in which thecontent of FoxO1 was 4 mg for each complex. Cell were washed with PBS 24 h post-transfection and then assayed for FoxO1 gene expression and apoptosis.

2.11.2. Evaluation of FoxO1 gene expression by western blotProtein samples (40 mg) were separated using SDS-PAGE and transferred onto

polyvinylidene difluoride membranes. The membranes were blocked in 5% bovineserum albumin solution for 1 h, and then incubated with the primary antibodies(1:2000 dilution) against the FoxO1. HRP-conjugated anti-rabbit IgG (1:3000 dilu-tion) was used as the second antibody. The blot was developed using SuperenhancedChemiluminescence Detection Kit (Applygen Technologies Inc., Beijing, China). Theresults were normalized by the b-actin signal.

2.11.3. Apotosis assay by Annexin V/propidium iodide staining assayCells (293T) transfected by FoxO1 gene were quickly trypsinized, detached from

plastic plate, and washed with PBS for two times. The cells were then suspended

Fig. 2. (A) Influence of hemolysis rate on the concentration of amylopectin derivatives and bPEI, (B) Visual observation of hemolysis caused by the amylopectin derivatives and bPEI,(C) AFM images of erythrocytes in (a) 0.9% NaCl solution, incubated with cationic polymers (500 mg/mL): (b) EDA-Amp, (c) DETA-Amp, (d) DMAPA-Amp, (e) bPEI, and (f) incubatedwith 10% Trion-100 (I: phase mode image, scale bar: 5 mm; II: height mode image).

Y. Zhou et al. / Biomaterials xxx (2012) 1e104

371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435

436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500

JBMT13897_proof ■ 13 March 2012 ■ 4/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

with HEPES buffer and stained with Annexin V and propidium iodide for 15 min asdirected by the manufacturer’s protocol (Lianke, China). The stained cells wereassessed with an FACS Aria flow cytometer.

2.11.4. Apotosis assay by Hoechst staining assayCells (293T) transfected with FoxO1 gene were treated with 0.1 mg/ml Hoechst

33342 for 10 min at 37 �C in the dark. Hoechst-stained nuclei were observed usingan Olympus IX71 fluorescence microscope.

2.12. Statistical analysis

Statistical analysis was performed by one-factor analysis of variance (SPSSsoftware, version 13.0, SPSS Inc). The results were expressed as mean � SD, anda value of P < 0.05 was considered statistically significant.

3. Results and discussion

3.1. Synthesis of hyperbranched cationic amylopectin derivatives

The hyperbranched conformation of amylopectin used in thisstudy was characterized by static and dynamic light scatteringanalyses in dimethyl sulfoxide (Supplementary Data). The radius ofgyration (Rg) and the dynamic radium (Rh) values of amylopectinwere determined to be 137 and 158 nm, respectively (Fig. S1). Theparameter of r (Rg/Rh) was then calculated to be 0.87, and this resultindicates that amylopectin exists as a compact and hyperbranchedmacromolecule, which is consistent with previous work [25].

Since amylopectin itself has no cationic charges, cationicamylopectin derivatives were synthesized for the complexationwith negatively charged pDNA. As shown in Scheme 1, differentoligoamine compounds including EDA, DETA and DMAPA wereintroduced to the hydroxyl groups of amylopectin by the conven-tional CDI activation method [4,14].

From the FTIR spectra shown in Fig. 1A, it was found that thestrong absorption peak at 1709 cm�1 of amylopectin derivativeswas different from that of amylopectin, which could be assigned tothe eC]O vibration of carbamate groups [26e28]. Furthermore,the absorption peaks at 1549 and 1261 cm�1 were attributed to theeNHe and CeN vibrations of carbamate groups, and aliphaticamines, respectively [28]. The results indicate that the oligoamineresidues were conjugated with amylopectin with carbamate link-ages. The 1H NMR spectra of EDA-Amp, DETA-Amp, and DMAPA-Amp are shown in Fig. 1B. The proton peaks of oligoamine resi-dues appear at 1.5e3.3 ppm, further confirming that the oligo-amine residues were conjugated with amylopectin.

3.2. Buffer capacity

It is well known that the high buffer capacity of cationic poly-mers result in relatively high gene-transfer activity because theyare postulated to cause an increase swelling of endocytic vesicles tohelp DNA complexes escape into the cytoplasm [2,29]. The buffercapacity of cationic amylopectin derivatives in aqueous NaCl solu-tion was assessed by acid-base titration. As shown in Fig. S2, thebuffer capability of amylopectin derivatives was higher than that ofamylopectin. However, the amylopectin derivatives exhibited lowerbuffer capability than that of bPEI. The lower buffer capability of theamylopectin derivatives is likely due to the presence of relativelyfewer amine groups on the derivatives.

3.3. Blood compatibility assay

The instability of delivery vehicles in the blood was consideredas one of the serious limitations in the therapeutic of cationicpolymers [10,30]. The nonspecific interactions of cationic polymerswith blood components could severely diminish the half-life andtargetability of complexes as well as the reproducibility of such

therapies [10]. The blood compatibility of the cationic amylopectinderivatives was assessed by spectrophotometric measurement ofhemoglobin release from erythrocytes after polymer treatment. Asshown in Fig. 2A, the hemolysis rate of the EDA-Amp and TETA-Amp groups were less than 5%, a rate suggestive of low hemolyticactivity. In contrast, the hemolysis rate of the DMAPA-Amp groupincreased. The increase in the hemolysis rate of the DMAPA-AMPgroup is likely caused by the presence of a large number oftertiary amine groups. Mainly as a result of the erythrocytemembrane disruption, bPEI caused serious hemolysis in a concen-tration-dependent manner. This hemolytic phenomenon wasclearly observed in Fig. 2B and corresponds to the quantitativehemolytic activity shown in Fig. 2A.

The detailed morphological changes and specific surface char-acteristics of erythrocytes were analyzed, with the aid of the liter-ature, by AFM [31]. As shown in Fig. 2C-a, the majority oferythrocytes in 0.9% NaCl solution appeared as normal cells in theshape of biconcave discs with smooth surfaces. In contrast, eryth-rocytes in 10% Trin-100 were completely destroyed, showingserious hemolysis, and a large quantity of erythrocyte debris andhemoglobin (Fig. 2C-f). The morphology of erythrocytes showed noobvious change after the erythrocytes were incubated with theEDA-Amp and DETA-Amp derivatives (Fig. 2C-b and c), comparedwith those in 0.9% NaCl solution (Fig. 2C-a). After they were incu-bated with the DMAPA-Amp derivative, the surface of the

Fig. 3. (A) Agarose gel electrophoresis retardation assay of amylopectin derivative/pDNA complexes at different weight ratios: (a) EDA-Amp/pDNA, (b) EDA-Amp/pDNA,and (c) DMAPA-Amp/pDNA; (B) Protection and release assay of pDNA. Lanes 1 and2: naked pDNA, lanes 3e5: EDA-Amp/pDNA complexes (w/w ¼ 5, 10 and 20), lanes6e8: DETA-Amp/pDNA complexes (w/w ¼ 5, 10 and 20), lanes 9e11: DMAPA-Amp/pDNA complexes (w/w ¼ 5, 10 and 20).

Y. Zhou et al. / Biomaterials xxx (2012) 1e10 5

501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565

566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630

JBMT13897_proof ■ 13 March 2012 ■ 5/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

erythrocytes became rough, and the cells deformed and aggregated(Fig. 2C-d). The deformation and aggregation became more seriousfor erythrocytes incubated with bPEI (Fig. 2C-e), because ofstronger interaction between the erythrocytes and bPEI. The meansurface roughness of erythrocytes was quantitatively evaluated byAFM as shown in Fig. S3. The surface roughness of EDA-Amp andDETA-Amp derivatives was less than 100 nm, however, that of theDMAPA-Amp derivative and bPEI increased to 590 and 770 nm,respectively. The overall morphology of erythrocytes was alsoobserved in optical images as shown in Fig. S4. The results of AFMand optical morphology are consistent with those of the hemolysisassay and the hemolytic phenomenon, indicating that the bloodcompatibility of the EDA-Amp and DETA-Amp derivatives is betterthan that of the DMAPA-Amp derivative and in particular, that ofbPEI.

3.4. Complex formation of cationic amylopectin derivative withpDNA

Using pDNA as a control, the formation of cationic amylopectinderivative/pDNA complexes was examined by agarose gel electro-phoresis. As shown in Fig. 3A, the migration of pDNA wascompletely retarded when the amylopectin derivative/pDNAweight ratio exceeded 5. The result suggests that the amylopectinderivatives have good pDNA binding ability due to cationic oligo-amine residues. The protection effect of complexes against DNAdegradation by DNase I is shown in Fig. 3B. While naked pDNAwascompletely digested, EDA-Amp and DETA-Amp at all weight ratios(5, 10 and 20) exhibited distinct protective effects against DNase I.Likely due to the presence of large amounts of tertiary aminegroups, the DMAPA-Amp/DNA complex could not be replaced with10% SDS.

Previous work reported that positive surface charge and properparticle size of cationic polymers/DNA complexes were importantfor efficient gene delivery [2,19,31]. Complexes with the sizesbetween 50 to several hundred nanometers were suitable for celluptake [32]. As shown in Fig. 4A, it was found that the zeta potentialof amylopectin derivative/pDNA complexes increased withincreasing weight ratios. For example, the zeta potential of DMAPA-Amp/pDNA complex increased from þ7 mv to þ32 mv when theweight ratio increased from 1 to 30. The positive surface charge ofthe complexes is amenable to the effective condensation of DNA. Asexpected, the average particle size of amylopectin derivative/pDNAcomplexes decreased with increasing weight ratios (Fig. 4b).Additionally, it was found that the type and content of oligoamineresidues of amylopectin derivatives had an effect on the DNAcondensation. As shown in Fig. 4a and b, the positive surface chargeof the complexes of the same weight ratio were generally in theorder of DMAPA-Amp/pDNA > DETA-Amp/pDNA > EDA-Amp/pDNA, while the size changes were in the reverse order. This isattributed to the structural differences of oligoamine residuesconjugated to amylopectin. The EDA-Amp, DETA-Amp and DMAPAderivatives respectively contain primary amines, primary andsecondary amines, and tertiary amines besides carbamate groups(Scheme 1). Furthermore, the oligoamine residue content of theDMAPA derivative was higher than that of the EDA-Amp and DETA-Amp derivatives. As a result, the DMAPA-Amp derivative possessedbetter DNA condensation ability than the other derivatives due toits higher positive surface charge.

The morphology of amylopectin derivative/pDNA complexeswas investigated by SEM. As shown in Fig. 4cee, the complexes atweight ratio 20 were observed as particles with irregular shape inthe sizes ranging from 100 to 300 nm. Moreover, the size of thecomplexes decreased with increasing the surface charge in the

Fig. 4. Properties of amylopectin derivative/pDNA complexes at various weight ratios: (a) zeta potential, (b) average particle size, and SEM images of (c) EDA-Amp/pDNA, (d) DETA-Amp/pDNA and (e) DMAPA-Amp/pDNA (w/w ¼ 20).

Y. Zhou et al. / Biomaterials xxx (2012) 1e106

631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695

696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760

JBMT13897_proof ■ 13 March 2012 ■ 6/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

order of EDA-Amp/pDNA, DETA-Amp/pDNA and DMAPA-Amp,which corroborates the results from Fig. 4b.

3.5. In vitro cytotoxicity

Along with blood compatibility, in vitro cytotoxicity of cationicpolymers is also considered an important factor of biocompatibility[23]. The in vitro cytotoxicity of amylopectin derivatives and bPEIwith concentrations varied from 16 to 1000 mg/mL was assessed in293T and A549 cells by MTT assay. As shown in Fig. 5a and b, it wasfound that the cell viability of the amylopectin derivatives washigher than that of bPEI, indicating the amylopectin derivativesexhibited lower cytotoxicity. This could be attributed to thebiocompatibility of amlopectin. Furthermore, it was found that thecell viability of EDA-Amp and DETA-Amp derivatives was generallyhigher than that of DMAPA-Amp, indicating that cytotoxicity isrelated to the type and content of the oligoamine residues conju-gated to each amylopectin derivative. The EDA-Amp and DETA-Amp derivatives, which are comprised by low-density primaryand secondary amines, exhibited lower cytotoxicity than theDMAPA-Amp derivative, which is comprised by high-densitytertiary amines.

The cytotoxicity of the amylopectin derivative/pDNA complexeswith various weight ratios were further evaluated in 293Tand A549cells by MTT assay. As a result of lower cytotoxicity of the amylo-pectin derivatives, the cell viability of the amylopectin derivative/

pDNA complexes was higher than that of the bPEI/pDNA complex atweight ratios ranging from 2.5 to 80 (Fig. 5c and d). Themorphology of 293Tand A549 cells incubatedwith the amylopectinderivative/pDNA and bPEI/pDNA complexes is shown in Fig. S5.Both cell lines incubated with the bPEI/pDNA complex weredestroyed or shrunken, suggesting strong cytotoxicity. In contrastwith the control, the morphology of cells incubated with the EDA-Amp/pDNA and EDA-Amp/pDNA complexes showed no obviouschange. The DMAPA-Amp/pDNA complex exhibited higher cyto-toxicity than the EDA-Amp/pDNA and EDA-Amp/pDNA complexes,but lower cytotoxicity than the bPEI/pDNA complex.

3.6. In vitro transfection of pDNA

The gene transfer capacity of the cationic amylopectin deriva-tives was evaluated in 293T and A549 cell lines. The transfectionefficiencywas quantitatively determinedbyusingflowcytometry toassess the actual percentage of GFP-expressing cells. As shown inFig. 6A, the gene transfection efficiency of the cationic amylopectinderivative/pDNA complexes depends on theweight ratio values, thecell types, and the oligoamine residues of the amylopectin deriva-tives. At the weight ratio of 5, the complexes exhibited low genetransfer capacity in the two cell lines. However, the transfectionefficiency of the complexes increased with increasing weight ratios.Furthermore, it was found that the gene transfer capacity of thecationic amylopectin derivatives in 293T cells was higher than that

Fig. 5. Viability of (a) 293T cells and (b) A549 cells incubated with the amylopectin derivative and bPEI at various concentrations for 24 h (n ¼ 5), and viability of (c) 293T cells and(d) A549 cells incubated with the amylopectin derivative/pDNA and bPEI/pDNA complexes at various weight ratios for 24 h (n ¼ 5).

Y. Zhou et al. / Biomaterials xxx (2012) 1e10 7

761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825

826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890

JBMT13897_proof ■ 13 March 2012 ■ 7/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

in A549 cells. The higher gene transfer capacity is likely due todistinct endocytosis ability of 293T cells. The transfection of nakedpDNAwas less than0.3% in the two cell lines, further confirming thatthe cationic amylopectin derivatives have the capability to protectpDNA and improve GFP-expression. Noticeably, the transfectionefficiency of the DMAPA-Amp/pDNA complex (w/w ¼ 20) in 293Tcells increased to about 85%, which is equal to that of the bPEI/pDNAcomplex. The high transfection efficiency is likely related to the highDNA condensation ability of DMAPA-Amp.

GFP-expressing cells were observed using an inverted fluores-cence microscope in order to further visualize the transfectionefficiency of the amylopectin derivative/pDNA complexes (w/w ¼ 10). As shown in Fig. 6B, a strong green fluorescent signal wasobserved in 293T cells transfected with the amylopectin deriva-tives, endorsing the high transfection efficiency found by flowcytometry. Additionally, it was found that the transfection effi-ciency of the amylopectin derivatives was higher in 293T cells thanin A549 cells.

Fig. 6. (A) Transfection efficiency of amylopectin derivative/pDNA complexes at various weight ratio (w/w ¼ 5, 10, and 20) and the bPEI/pDNA complex (N/P ¼ 10) determined byflow cytometry in 293T cells and A549 cells; (B) Fluorescence micrographs and light inverted micrographs of 293T cells and A549 cells transfected by the amylopectin derivative/pDNA complexes (w/w ¼ 10): (a) EDA-Amp/pDNA, (b) DETA-Amp/pDNA, (c) DMAPA-Amp/pDNA, and (d) the bPEI/pDNA complex (N/P ¼ 10) in the presence of serum (�100).

Y. Zhou et al. / Biomaterials xxx (2012) 1e108

891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955

956957958959960961962963964965966967968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999

100010011002100310041005100610071008100910101011101210131014101510161017101810191020

JBMT13897_proof ■ 13 March 2012 ■ 8/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

3.7. Analysis of FoxO1 gene expression and induced apoptosis

FoxO1, a forkhead transcription factors, can bind and regulatethe expression of cell cycle regulators and apoptosis [33].The FoxO1transfection activity mediated by the amylopectin derivatives in293T cells was evaluated using western blot analysis, which candetect specific protein expression and is more sensitive than fluo-rescence intensity determination. As shown in Fig. 7A, cells treatedby the amylopectin derivative-mediated gene transfection wereable to express the FoxO1 protein. The DMAPA-Amp derivative-mediated gene transfection, likely due to its higher delivery effi-ciency, exhibited higher protein expression than the gene trans-fection mediated by the EDA-Amp and DETA-Amp derivatives.Although the protein expression levels were high for the amylo-pectin derivatives, the FoxO1 protein expressed by bPEI-mediatedgene transfection remains high as well.

The delivery efficiency of the FoxO1 gene mediated by theamylopectin derivatives was further evaluated in situ by using flowcytometry to detect FoxO1-induced apoptotic cells. As shown inFig. 7B-a, most cells were distributed in the Q3 region, indicatinggood survival status under the cell culture conditions. The quantityof cells in the Q2 and Q4 regions increased after 293T cells wereincubated with the amylopectin derivative/FoxO1 and bPEI/FoxO1complexes, with increasing cell counts in the order EDA-Amp < DETA-Amp < DMAPA-Amp < bPEI (Fig. 7B-b,c,d,e). Thisresult indicates that the level of cell apoptosis among all theamylopectin derivatives treated cell goups is highest for theDMAPA-Amp/FoxO1 complex, and relatively lower than that of thebPEI/FoxO1 complex.

The FoxO1 delivery efficiency was also evaluated by detection ofFoxO1-induced apoptotic cells using Hoechst staining assay. Asshown in Fig. S6, cells undergoing apoptosis were distinguishedfrom nonapoptotic cells because their nuclei were stained blue ordisintegrated. Apoptosis is negligible in the normally culturedcontrol cells (cells marked with white arrow in Fig. S6-a). Incontrast, more blue Hoechst-stained nuclei were observed inapoptotic cells transfected with the amylopectin derivatives/FoxO1

complexes (cells marked with white arrow in Fig. S6-b, c, d), indi-cating the high delivery efficiency. This result is in agreement withthe results from flow cytometry.

4. Conclusions

A series of hyperbranched cationic amylopectin derivativesconjugated with various oligoamine residues were synthesized bythe CDI activation method at room temperature. They exhibitedbetter blood compatibility and lower cytotoxicity when comparedto bPEI. The amylopectin derivatives could bind pDNA and FoxO1 toform complexes that showed high gene delivery capability andtransfection efficiency in 293T cells. Amylopectin derivatives andthe complexes formed using these derivatives possess potential asnonviral gene vectors for future gene therapy applications.

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (20974130, 20574089, 30901547, 30973093,C81171711), the Natural Science Foundation of Guangdong Provincein China (2009B020313001, 2009B050200010) and the Funda-mental Research Funds for the Central Universities of China(09lgpy14).

Appendix A. Supplementary data

Light scattering analyses of amylopectin, acid-base titrationprofiles of the amylopectin derivatives, mean surface roughnessand morphology of erythrocytes, morphology of 293T and A549cells incubated with the amylopectin derivatives, and morphologyof Hoechst staining assay after 293T cells incubated with theamylopectin derivative/FoxO1 complexes and the bPEI/FoxO1complex. Supplementary data related to this article can be foundonline at doi:10.1016/j.biomaterials.2012.03.014.

Fig. 7. (A) Western blot analysis of FoxO1 protein expression in 293T cells incubated with the amylopectin derivative/FoxO1 complexes (w/w ¼ 10) and the bPEI/FoxO1 complex (N/P ¼ 10), and (B) apotosis analysis by flow cytometry after 293T cells incubated with the amylopectin derivative/FoxO1 complexes (w/w ¼ 10) and the bPEI/FoxO1 complex (N/P ¼ 10): (a) the control, (b) EDA-Amp/FoxO1, (c) DETA-Amp/FoxO1, (d) DMAPA-Amp/FoxO1, and (e) bPEI/FoxO1.

Y. Zhou et al. / Biomaterials xxx (2012) 1e10 9

10211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085

10861087108810891090109110921093109410951096109710981099110011011102110311041105110611071108110911101111111211131114111511161117111811191120112111221123112411251126112711281129113011311132113311341135113611371138113911401141114211431144114511461147114811491150

JBMT13897_proof ■ 13 March 2012 ■ 9/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014

References

[1] Varkouhi AK, Scholte M, Storm G, Haisma HJ. Endosomal escape pathways fordelivery of biological. J Control Release 2011;151:220e8.

[2] Lu B, Wang CF, Wu DQ, Li C, Zhang XZ, Zhuo RX. Chitosan based oligoaminepolymers: synthesis, characterization, and gene delivery. J Control Release2009;137:54e62.

[3] Buyens K, De Smedt SC, Braeckmans K, Demeester J, Peeters L, Grunsven LA,et al. Liposome based systems for systemic siRNA delivery: stability in bloodsets the requirements for optimal carrier design. J Control Release; 2011.doi:10.1016/j.jconrel.2011.10.009.

[4] Toita S, Morimoto N, Akiyoshi K. Functional cycloamylose as a polysaccharide-based biomaterial: application in a gene delivery system. Biomacromolecules2010;11:397e401.

[5] Wang R, Zhou L, Zhou Y, Li G, Zhu X, Gu H, et al. Synthesis and gene delivery ofpoly(amido amine)s with different branched architecture. Biomacromolecules2010;11:489e95.

[6] Malmo J, Varum KM, Strand SP. Effect of chitosan chain architecture on genedelivery: comparison of self-branched and linear chitosans. Bio-macromolecules 2011;12:721e9.

[7] Fischer D, Harpe A, Kunath K, Petersen H, Li Y, Kissel T. Copolymers ofethylene imine and N-(2-hydroxyethyl)-ethylene imine as tools to studyeffects of polymer structure on physicochemical and biological properties ofDNA complexes. Bioconjug Chem 2002;13:1124e33.

[8] Wang X, He Y, Wu J, Gao C, Xu Y. Synthesis and evaluation of phenylalanine-modified hyperbranched poly(amido amine)s as promising gene carriers.Biomacromolecules 2010;11:245e51.

[9] Schallon A, Jerome V, Walther A, Synatschke CV, Muller AHE, Freitag R.Performance of three PDMAEMA-based polycation architectures as genedelivery agents in comparison to linear and branched PEI. React Funct Polym2010;70:1e10.

[10] Kainthan RK, Gnanamani M, Ganguli M, Ghosh T, Brooks DE, Maiti S, et al.Blood compatibility of water soluble hyperbranched polyglycerol-basedmultivalent cationic polymers and their interaction with DNA. Biomaterials2006;27:5377e90.

[11] Zintchenko A, Philipp A, Dehshahri A, Dehshahri E. Simple modifications ofbranched PEI lead to highly efficient siRNA carriers with low toxicity. Bio-conjug Chem 2008;19:1448e55.

[12] Kim T, Seo HJ, Choi JS, Hang JH, Baek J, Kim K, et al. PAMAM-PEG-PAMAM:novel Triblock copolymer as a biocompatible and efficient gene deliverycarrier. Biomacromolecules 2004;5:2487e92.

[13] Liu Z, Jiao Y, Wang Y, Zhou C, Zhang Z. Polysaccharides-based nanoparticles asdrug delivery systems. Adv Drug Deliv Rev 2008;60:1650e62.

[14] Jo J, Ikai T, Okazaki O, Yamamoto M, Hirano Y, Tabata Y. Expression profile ofplasmid DNA by spermine derivatives of pullulan with different extents ofspermine introduced. J Control Release 2007;118:389e98.

[15] Juan AS, Hlawaty H, Chaubet F, Letourneur D, Feldman LJ. Cationized pullulan3D matrices as new materials for gene transfer. J Biomed Mater Res 2007;82A:354e62.

[16] Nagasaki T, Hojo M, Uno A, Satoh T, Koumoto K, Mizu M, et al. Long-termexpression with a cationic polymer derived from a natural polysaccharide:schizophyllan. Bioconjug Chem 2004;15:249e59.

[17] Katav T, Liu L, Traitel T, Goldbart R, Wolfson M, Kost J. Modified pectin-basedcarrier for gene delivery: cellular barriers in gene delivery course. J ControlRelease 2008;130:183e91.

[18] Wang ZH, Li WB, Ma J, Tang GP, Yang WT, Xu FJ. Functionalized nonionicdextran backbones by atom transfer radical polymerization for efficient genedelivery. Macromolecules 2011;44:230e9.

[19] Deng J, Zhou Y, Xu B, Mai K, Bin Y, Zhang L. Dendronized chitosan derivative asa biocompatible gene delivery carrier. Biomacromolecules 2011;12:642e9.

[20] Song Y, Sun Y, Zhang X, Zhou J, Zhang L. Homogeneous quaternization ofcellulose in NaOH/urea aqueous solutions as gene carriers. Bio-macromolecules 2008;9:2259e64.

[21] Miranda JA, Cacita N, Okano LT. Evaluation of amylopectin clusters and theirinteraction with nonionic surfactants. Colloid Surf B: Biointerfaces 2007;60:19e27.

[22] Gottlieb HE, Kotlyar V, Nudelman A. NMR chemical shifts of common labo-ratory solvents as trace impurities. J Org Chem 1997;62:7512e5.

[23] Fischer D, Li Y, Ahlemeyer B, Krieglstein J, Kissel T. In vitro cytotoxicity testingof polycations: influence of polymer structure on cell viability and hemolysis.Biomaterials 2003;24:1121e31.

[24] Srinivasachari S, Fichter KM, Reineke TM. Polycationic b-cyclodextrin “clickclusters”: monodisperse and versatile scaffolds for nucleic acid delivery. J AmChem Soc 2008;130:4618e27.

[25] Yang L, Fu S, Zhu X, Zhang L, Yang Y, Yang X, et al. Hyperbranched acidicpolysaccharide from green tea. Biomacromolecules 2010;11:3395e405.

[26] Engelmann G, Jobmann M, Raflter G. Dextran carbamates - materials formicroencapsulation. Ind Crop Prod 2004;20:37e48.

[27] Yin C, Li J, Xu Q, Peng Q, Liu Y, Shen S. Chemical modification of cottoncellulose in supercritical carbon dioxide: synthesis and characterization ofcellulose carbamate. Carbohyr Polym 2007;67:147e54.

[28] Wu J. Techniques and application of modern FTIR spectroscopy. Beijing:Science Press; 1994. 603e624.

[29] Pack DW, Hoffman AS, Pun S, Stayton PS. Design and development of poly-mers for gene delivery. Nat Rev Drug Discov 2005;4:581e93.

[30] Yang J, Liu Y, Wang H, Liu L, Wang W, Wang C, et al. The biocompatibility offatty acid modified dextran-agmatine bioconjugate gene delivery vector.Biomaterials 2012;33:604e13.

[31] Girasole M, Pompeo G, Cricenti A, Longo G, Boumis G, Bellelli A, et al. The how,when, and why of the aging signals appearing on the human erythrocytemembrane: an atomic force microscopy study of surface roughness. NanomedNanotechnol Biol Med 2010;6:760e8.

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

[33] Jia G, Aggarwal A, Tyndall SH, Agrawal DK. Tumor necrosis factor-a regulatesp27kip expression and apoptosis in smooth muscle cells of human carotidplaques via forkhead transcription factor O1. Exp Mol Pathol 2011;90:1e8.

Y. Zhou et al. / Biomaterials xxx (2012) 1e1010

115111521153115411551156115711581159116011611162116311641165116611671168116911701171117211731174117511761177117811791180118111821183118411851186118711881189

119011911192119311941195119611971198119912001201120212031204120512061207120812091210121112121213121412151216121712181219122012211222122312241225122612271228

JBMT13897_proof ■ 13 March 2012 ■ 10/10

Please cite this article in press as: Zhou Y, et al., Hyperbranched cationic amylopectin derivatives for gene delivery, Biomaterials (2012),doi:10.1016/j.biomaterials.2012.03.014