Targeted minicircle DNA delivery using folate-poly ... · Targeted minicircle DNA delivery using folateepoly ... The investigation of the endocytosis pathway of the polyplexes showed

lable at ScienceDirect

ARTICLE IN PRESS

Biomaterials xxx (2010) 1e12

Contents lists avai

Biomaterials

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

Targeted minicircle DNA delivery using folateepoly(ethylene glycol)epolyethylenimine as non-viral carrier

Chao Zhang a, Shijuan Gao a, Wei Jiang a, Song Lin c,d, Fusheng Du c, Zichen Li c,**,1, Wenlin Huang a,b,*,1

aCAS Key Laboratory of Pathogenic Microbiology and Immunology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, PR Chinab State Key Laboratory of Oncology in South China, Cancer Center, Sun Yat-sen University, Guangzhou 510060, PR ChinacBeijing National Laboratory for Molecular Sciences, Key Laboratory of Polymer Chemistry and Physics of Ministry of Education, College of Chemistry & Molecular Engineering,Peking University, Beijing 100871, PR Chinad Institute of Medical Equipment, The Academy of Military Medical Sciences, Tianjin 300161, PR China

a r t i c l e i n f o

Article history:Received 2 March 2010Accepted 21 April 2010Available online xxx

Keywords:FolateePEGePEIReceptor-mediated gene deliveryMinicircle DNATumor targeting

* Corresponding author. CAS Key Laboratory of PImmunology, Institute of Microbiology, Chinese A100101, PR China. Tel.: þ86-10-64807806; fax: þ86-1** Corresponding author. Tel.: +86-10-62755543.

E-mail addresses: [email protected] (Z. Li), hwenl@m1 These authors contributed equally to this work.

0142-9612/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.biomaterials.2010.04.042

Please cite this article in press as: Zhang C, etviral carrier, Biomaterials (2010), doi:10.101

a b s t r a c t

Targeted gene delivery systems have attracted great attention due to their potential in directing thetherapeutic genes to the target cells. However, due to their low efficiency, most of the successfulapplications of polymeric vectors have been focused on genes which can achieve robust expression.Minicircle DNA (mcDNA) is a powerful candidate in terms of improving gene expression and prolongingthe lifespan of gene expression. In this study, we have combined folate/poly(ethylene glycol) modifiedpolyethylenimine and mcDNA as a new tumor gene delivery system. We found that folate-labeled pol-yplexes were homogenous, with a size ranging from 60 to 85 nm. mcDNA increased folate-labeled vectorbased gene expression 2e8 fold in folate receptor-positive cells. Results of folic acid competition assayindicated that mcDNA mediated by folate-labeled vector were internalized into cells through receptor-mediated endocytosis. The investigation of the endocytosis pathway of the polyplexes showed thata large portion of them escaped from endo/lysosome and the polyplexes were associated before beingseparated in the nucleus. Furthermore, in vivo optical imaging and luciferase assays demonstrated thatsystemic delivery of the folate-labeled polyplexes resulted in preferential accumulation of transgenes infolate receptor-positive tumors, and mcDNA mediated approach achieved 2.3 fold higher gene expres-sions in tumors than conventional plasmid. Cytotoxicity assays showed that PEG-shielding of the poly-plexes reduced the toxicity of PEI.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Non-viral vectors are attractive gene delivery systems for tumorgene therapy. Compared with viral vectors, they are safe, simple toprepare and modify, and have larger gene encapsulation capability[1,2]. In the past decade, a number of non-viral vectors, such asliposomes or cationic polymers have been widely used in genetherapy [3e5].

Polyethylenimine (PEI) is one of the most successful and effi-cient non-viral vectors. It is well documented that PEI can condenseplasmid DNA or siRNA to polyplexes and promote endosomalescape through the well-known “proton sponge” effect [6e8]. In

athogenic Microbiology andcademy of Sciences, Beijing0-64807808.

ail.sysu.edu.cn (W. Huang).

All rights reserved.

al., Targetedminicircle DNA d6/j.biomaterials.2010.04.042

addition, it has been reported that PEI promotes transgene deliveryto the nucleus [9,10]. However, PEI itself is still limited as a genecarrier because of its high cytotoxicity and non-specific interactionsin vivo [11]. In order to improve the biocompatibility of PEI,hydrophilic polymers such as poly(ethylene glycol) (PEG) wereintroduced. It is thought that PEG chains out of polyplexes arecapable of reducing the cytotoxicity of PEI and its interactions withblood and extracellular components [12,13]. PEGylation extendscirculation time and reduces in vivo toxicity of PEI-based polyplexeswhen used in systemic gene delivery [6,14,15]. Unfortunately, mostof these PEGylated polyplexes are lacking, in terms of reducedassociation with cells, diminished cellular uptake and/or endo-somal escape, and inefficient cell transfection [16,17]. An efficientstrategy for overcoming this limitation would be to add targetingligands to polyplexes which would enhance their cell-specific genedelivery via receptor-mediated cellular uptake [18,19]. One of thebest-characterized targeting ligands for tumor treatment is folate,since folate receptors (FR) exhibit limited expression on healthycells, but are over-expressed in certain cancer cells [20]. Folate/PEG

elivery using folateepoly(ethylene glycol)epolyethylenimine as non-

mailto:[email protected]

mailto:[email protected]

www.sciencedirect.com/science/journal/01429612

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

C. Zhang et al. / Biomaterials xxx (2010) 1e122

ARTICLE IN PRESS

modified PEI has previously been shown to promote target-specificgene delivery both in vitro and in vivo [21e24], showing superiorperformance compared to PEI [21e23,25]. In a recent study, whenfolateePEGePEI was used for the delivery of therapeutic genes bydirect injection into tumors of glioma-xenografted rats, a signifi-cant antitumor effect was achieved [26].

Minicircle DNA (mcDNA) is a form of supercoiled DNA con-taining only a gene expression cassette and lacking extraneousplasmid sequences [27]. It is generated by site-specific recombi-nation in E. coli [28,29]. mcDNA is superior to conventional plas-mids in terms of improvement of gene expression efficiency andprolonging the time span of gene expression [27,28,30]. Previousresearch has demonstrated the advantages of mcDNA overconventional plasmids in mouse skeletal muscle, liver, humancarcinoma xenograft tumor, and iPS cells, [28,31e33]. Furthermore,mcDNA increases PEI-based transfection efficiency both in vitro andin vivo [34,35]. However, tumor-targeted delivery of mcDNA bynon-viral vectors has not been studied before.

In this study, we used folate/PEG-modified PEI as a deliverysystem for tumor-target transfer of mcDNA. The folate-labeledpolyplexes containing mcDNA exhibit strong tumor-targetingcapability and high levels of gene expression both in vitro and invivo.

2. Materials and methods

2.1. Reagents

Branched PEI (MW 25 kDa) was obtained from SigmaeAldrich (St. Louis, MO,USA). Folic acid was purchased from Sinopharm Chemical Reagent Co., Ltd.(Shanghai, China). MAL-PEG3500-NHSwith an averagemolecular weight of 3500 Dawas provided by Jenkem Technology Co., Ltd. (Beijing, China). FITC-labeled PEI andEMA-labeled mcDNA were synthesized as described in the Supplementarymaterials.

2.2. Plasmids

Plasmid p2FC31 (9.7 kb) was a generous gift from Dr. Zhiying Chen (StanfordUniversity, Stanford, CA, USA). pUC-EGFP carrying gfp gene and pShuttle-luciferase(pShuttle-luc) carrying luciferase gene were constructed in our lab. p2FC31-EGFP(1.1 kb) and p2FC31-luciferase (p2FC31-luc, 1.2 kb) were constructed for producingminicircle-GFP (mc-GFP) and minicircle-luciferase (mc-luc), respectively.

2.3. Synthesis of (FAePEG)2.3ePEI conjugates

(FAePEG)2.3ePEI conjugates were synthesized as shown in Scheme 1. Fordetailed information, please see Supplementary materials.

2.4. Physicochemical characterization of polyplexes

Polyplexes were prepared at specified N/P ratios (molar ratio of polymernitrogen atoms to DNA phosphates) and evaluated by agarose gel retardation assay[34]. Particle size and z-potential of polyplexes were measured by a dynamic lightscattering instrument (DynaPro Tian, Wyatt Technology Inc, CA) and a z-PALSanalyzer (Brookhaven Instruments Corporation, BIC), respectively. DNase protectionof polyplexes was performed. Details of the experiments are given in theSupplementary materials.

2.5. Cell culture and transfection

SKOV3 (human ovarian carcinoma) and HeLa (human adenocarcinoma) cells arefolate receptor-positive cells, while A431 (human epidermoid carcinoma of thevulva), HepG2 (human hepatoblastoma), and H22 (Murine hepatoma) cells arefolate receptor-negative cells. SKOV3 cell line is a kind gift from Dr. Cui Heng(People’s Hospital, Peking University). A431, H22, HeLa and HepG2 cell lines fromATCC (Manassas, VA, USA) were maintained by our lab. Cells were maintained inRPMI 1640 medium supplemented with 10% fetal bovine serum (Invitrogen, Carls-bad, CA, USA) at 37 �C, in 5% CO2. When confluence reached 80e90%, cells weretransfected with polyplexes containing 0.3 mg DNA at the specified N/P ratios andincubated in RPMI 1640 (folate-free) medium for 6 h. Themediumwas then changedwith fresh RPMI 1640 growth medium and cells were incubated for the reportergene assays.

Please cite this article in press as: Zhang C, et al., Targetedminicircle DNA dviral carrier, Biomaterials (2010), doi:10.1016/j.biomaterials.2010.04.042

2.6. MTT assay

The viability of transfected cells was measured using the Methylthiazolete-trazolium method (MTT). Details are given in the Supplementary materials.

2.7. Luciferase and GFP reporter gene assays

Luciferase activity was analyzed with a Luciferase Assay System (E1500, Prom-ega, WI, USA), according to the manufacturer’s instructions. Luciferase activity wasexpressed as RLU/mg protein. GFP expression was observed with a fluorescencemicroscope (Axiovert200, ZEISS, Germany). GFP (Excitation ¼ 488 nm,emission ¼ 507 nm).

2.8. Flow cytometry

HeLa cells were seeded into 12-well plates at a density of 1 �105 cells per well.When cells were at 90% confluence, they were treated with FITC-labeled polyplexescontaining 1.0 mg mcDNA in 1 mg/L folate medium or folate-free medium, respec-tively. 30 min post-transfection, the cells were washed three times with cold PBS,detached using 0.25% trypsin and 0.2% EDTA and collected in PBS. Subsequently, thecells were analyzed by flow cytometry (FACS Aria, BD, USA). The transfection effi-ciency was calculated based on measurements from three individual experiments.

2.9. Tracking the intracellular pathway of FPP/PEI/mcDNA polyplexes

Hela cells cultured in RPMI1640 (folate-free) growthmedium on glass coverslipswere transfectedwith FITC-labeled polyplexes containing 1 mg EMA-labeledmcDNA.At the time points indicated, the medium was removed and cells were rinsed threetimes with PBS before staining with the nuclear stain DAPI (1 mg/mL), and fixingwith 4% (v:v) paraformaldehyde in PBS. The pathway of the polyplexes was deter-mined using a confocal microscope (LS CMFV500, Olympus, Japan). The excitation/emission wavelengths were 340 nm/488 nm for DAPI, 490 nm/520 nm for FITC, and488 nm/600 nm for EMA [36]. To determine whether polyplexes were present inlysosomes, HeLa cells were first treated with FITC-labeled FPP/PEI/mcDNA poly-plexes containing 1 mg DNA. After 2e3 h, cells were then treated with 50 nM LysoTraker Red DND-99 (Invitrogen, Carlsbad, CA, USA) for 30 min to visualize thelysosomes. The excitation/emission wavelengths for Lyso Tracker were 577 nm/590 nm. Polyplex location was determined by scanning sequentially along thevertical axis from top to bottom.

2.10. Delivery of polyplexes in vivo

Female BAlB/c nude mice (4e5 weeks old) were purchased from the LaboratoryAnimal Center of Peking University Health Science Center. All animal experimentswere performed at the Institute of Genetics and Developmental Biology, ChineseAcademy of Sciences, in accordance with protocols approved by the InstitutionalAnimal Care and Use Committee. Nude mice were injected s.c. in the subaxillaryregion with 1 � 107 HeLa cells or 2 � 105 H22 cells. Experiments were carried outwhen tumors reached approximately 200e300 mm3.

(1) A total of 4 groups of 2 mice per group were used for in vivo fluorescenceimaging. 400 mL of the warmed polyplexes containing 40 mg EMA-labeled mcand 5% glucose were injected into each mouse via the tail vein. Two to 4 h later,mice were anesthetized and injected intraperitoneally with pentobarbitalsodium (45 mg/1 kg body weight) and the whole-body distribution of poly-plexes was monitored by an in vivo fluorescence imaging system (Maestro,Cambridge Research and Instrumentation Inc. MA, USA) or IVIS ImagingSystem (Xenogen, California, USA). Photons emitted from the mouse andtransmitted through the tissues were collected using a cooled charge coupleddevice (CCD) camera (Maestro or Xenogen IVIS). Images were obtained usingLiving Image software (Maestro or Xenogen IVIS).

(2) A total of 4 groups of 3 mice per group were used for luciferase expressionanalysis in the xenograft model. All mice were treated with 60 mg DNA/miceand sacrificed by cervical vertebra dislocation. The heart, liver, spleen, kidneys,brain, and tumors were collected, washed with PBS and homogenized in lysisbuffer (Promega, WI, USA) on ice. The homogenate was centrifuged at12,000�g for 10 min at 4 �C, and supernatants were used in the luciferaseassay.

2.11. Statistical analysis

Data are presented as mean values� stand deviation (S.D.). Statistical tests wereperformed with the Student’s t test or the ManneWhitney test. The ManneWhitneyU test was used to determine the difference between independent groups (SPSSversion 11.5, SPSS Inc). All statistical tests were two-tailed tests and the differencesbetween variants were considered to be statistically significant if p < 0.05.


N

HN

N

N

NH

N

H

CO2H

H2N

O

O CO2H

N

O

O

O

C

O

N

H

C

O

SH

N

O

O

O

C

O

O

O

N

H

O

N

O

O

N

H

C

O

S

n

*N

H

C

O

O

O

N

H

O

N

O

O

N

H

C

O

S

n

m

CO2H

i ii iii

iv

CO2H *= = bPEI 25K

(FA-PEG)m-PEI

FA

FA-NHS FA-SH

FA-PEG

Scheme 1. (i) NHS, DCC, (C2H5)3N, DMSO; (ii) cysteamine, (C2H5)3N, DMSO; (iii) MalePEGeNHS, pH 7.4 PBS; (iv) bPEI 25k, (C2H5)3N, DMSO.

C. Zhang et al. / Biomaterials xxx (2010) 1e12 3

ARTICLE IN PRESS

3. Results

3.1. Polymer synthesis and FPP/PEI/pDNA ternary polyplexscreening

Folate-containing PEGylatedPEI, (FAePEG)mePEI, was preparedas shown in Scheme 1. The average ratio (m) of FAePEG to PEI wase2.3 as determined by 1H NMR (Supplementary Data, Fig. S2). Forconvenience, (FAePEG)2.3ePEI will hereafter be abbreviated as FPPin this paper. Our preliminary results showed that the transfectionefficiency of FPP was greatly decreased compared to intact PEI(Supplementary Data, Fig. S3). In order to improve transfectionefficiency while preserving the shielding effect of PEG and thetargeting effect of folate, ternary polyplexes composed of pDNA,FPP and PEI were prepared at different FPP/PEI molar ratios. GFPexpressionwas used to evaluate these polyplexes in folate receptor-positive (FRþ) HeLa cells and folate receptor-negative (FR�) HepG2cells. Results demonstrated that transfection efficiency wasenhanced by increasing the ratio of PEI. Optimized transfectionefficiency was achieved in HeLa cells, but not in HepG2 cells, whenthe ratio of FPP to PEI reached 15:85 (data not shown). Ternarypolyplexes formed by DNA and FPP/PEI at this ratio were used insubsequent experiments, unless otherwise stated.

3.2. Physicochemical characteristics of polyplexes

The size and topology of nucleic acids may affect the physico-chemical and biological properties of polyplexes. It has beenreported that the optimum molecular weight and topology ofcationic polymers are different for the delivery of plasmid DNA andsiRNA [37,38]. In the present study, the size of mcDNA was smallerthan its original plasmid, as it was devoid of extraneous plasmidsequences. Thus, we investigated the effect of this size difference onthe physicochemical properties of polyplexes formed by mcDNA orplasmids (Fig. 1).


Electrophoretic images of FPP/PEI-containing plasmids ormcDNA at various N/P ratios (Fig. 1A) show that complete retar-dation was achieved for both polyplexes when the N/P ratio wasequal to or above 4, indicating that there were no obvious differ-ences in the binding capability of cationic polymers betweenmcDNA and plasmids. The size at various N/P ratios was measuredby dynamic light scattering, using PEI or the FPP/PEI mixture as thecationic polymer (Fig. 1B). All polyplexes analyzed had a mono-modal size distributionwith diameters of 60e85 nm at N/P ratios of10 to 20. The size of polyplexes containing mcDNA was slightlylarger than those containing plasmids at each N/P ratio analyzed. Inline with the findings given by Park et al. [34], mcDNA-containingpolyplexes in this study showed slightly higher z-potentials thanpolyplexes containing plasmids at an N/P ratio of 10. However, athigher N/P ratios, such as 20, the difference in z-potential almostdisappeared (Fig. 1C).

Protection of DNA against DNase degradation in the plasma isessential for efficient gene delivery in vivo. Fig. 1D shows thatnaked mcDNA can be degraded very quickly by DNase. Whencomplexed with cationic polymers (either PEI or the FPP/PEImixture) at an N/P ratio of 10, no degradation of mcDNA wasobserved. Similar results were obtained for polyplexes withplasmid DNA (data not shown).

3.3. In vitro biological characterization of mcDNA-containingpolyplexes

Four cell lines were used to characterize mc-containing poly-plexes. HeLa and SKOV3 cells are FRþ cells which over-expressfolate receptors [39,40]; A431 and HepG2 are FR� cells which aredeficient in folate receptors [41,42]. The expression of FR in thesecells was verified by RT-PCR (Supplementary Data, Fig. S4). In orderto clarify the functionalities of PEG and folate, both PEI and the FPP/PEI mixture (molar ratio of 15:85) were applied to condensemcDNA.


Fig. 1. Physicochemical characterization of the polyplexes. (A) Agarose gel electrophoresis of (a) pUC-EGFP and (b) mc-GFP complexed with FPP/PEI at different N/P ratios. (B)Diameters of the polyplexes formed by pUC-EGFP and mc-GFP with PEI or the FPP/PEI mixture at various N/P ratios, measured by DLS. (C) Zeta (z) potential of the polyplexes formedby pShuttle-luc and mc-luc with PEI or the FPP/PEI mixture at specified N/P ratios. (D) Protection of mc-GFP against DNase degradation. For mc-GFP in the polyplexes, the N/P ratiowas 10. Results are presented as mean values � SD.


ARTICLE IN PRESS

Cytotoxicity of the polyplexes was evaluated using the MTTassay (Fig. 2). The cytotoxicity of both PEI and FPP/PEI polyplexesincreased with increasing N/P ratios. The FPP/PEI/mcDNA poly-plexes exhibited lower cytotoxicity than PEI/mcDNA polyplexesregardless of cell lines. This can be attributed to the PEG shieldingeffect. In general, PEGylated polyplexes had lower cytotoxicity thannon-PEGylated polyplexes. This is in agreement with the results ofother scientists [23,43,44].

To evaluate the potential of using FPP/PEI/mcDNA polyplexesin targeted tumor therapy, in vitro gene transfection was inves-tigated by detection of luciferase (mc-luc) and GFP (mc-GFP)expression in FRþ and FR� cells, with PEI/mcDNA polyplexes ascontrols (Fig. 3 and Supplementary Data, Fig. S5). Luciferaseactivity revealed that gene expression depended on both celltypes and N/P ratios. Luciferase activity was highest in both


PEI/mcDNA and FPP/PEI/mcDNA polyplexes at N/P ratio of 15 inall the cell lines studied, and this ratio was used in subsequentexperiments. In addition, PEI/mc-luc showed higher transfectionefficiency than FPP/PEI/mc-luc. These results can be attributed tothe effects of PEG shielding in reducing cellular associations andsubsequent internalization of polyplexes, and in retardingendosomal escape [16,17,45,46]. However, differences in trans-fection efficiency between PEI/mc-luc and FPP/PEI/mc-luc poly-plexes were significantly larger for FR� cell lines, i.e., HepG2 andA431 (Fig. 3C, D). In FRþ cells, the transfection efficiency of FPP/PEI/mc polyplexes was 1.7e6.7 times lower than that of PEI/mcpolyplexes. In the case of FR� cells, the efficiency of FPP/PEI/mcDNA polyplexes was 4.5e47 times lower than PEI/mcDNApolyplexes. A similar phenomenon was also observed in the GFPassay (Supplementary Data, Fig. S5). These results can be attributed


Fig. 2. Cytotoxicity evaluated with the MTT assay in (A) HeLa, (B) SKOV3, (C) HepG2, and (D) A431 cells. Results are presented as mean values � SD.


ARTICLE IN PRESS

to the involvement of the folate receptor in cellular association andendocytosis of FPP/PEI/mcDNA polyplexes in FRþ cells.

Folate competition assays were conducted to confirm that thecellular uptake of FPP/PEI/mcDNA polyplexes was a target-specificevent. FRþ HeLa and SKOV3 cells were incubated in media con-taining different folate concentrations and then treated with mc-luc formed polyplexes. Results demonstrated that the transfectionefficiency of FPP/PEI/mc-luc polyplexes depends on the folateconcentration of the medium. Luciferase expression level markedlydecreased with increasing folate concentration. However, folateconcentration had little effect on the luciferase expression of PEI/mc-luc polyplexes in both cell lines studied (p > 0.05) (Fig. 4). Tofurther confirm the effect of the folate ligand on FPP/PEI/mcDNApolyplexes, we used flow cytometry tests. FITC-labeled PEI wasused to form PEI/mcDNA polyplexes or FPP/PEI/mcDNA polyplexes.HeLa cells were incubated in media with or without folate, andtreated with these FITC-labeled polyplexes. The percentage of FITC-positive cells and themean fluorescence intensity were 2.3 fold and1.2 fold higher, respectively, in medium without folate (Fig. 5, andSupplementary Data, Table S1).

3.4. Tracking the intracellular pathway of the FPP/PEI/mcpolyplexes

The above results show that FPP/PEI/mcDNA polyplexes wereinternalized through the FR- mediated pathway. Using confocalmicroscopy we further analyzed the intracellular pathway of thepolyplexes. Polymers and mcDNA were labeled with FITC (green)and EMA (red), respectively. The double-labeled polyplexes wereincubated with HeLa cells and the intracellular location of thepolyplexes was analyzed at the time points indicated. At 15 minpost-transfection, the polyplexes were poorly incorporated into


HeLa cells as only a few fluorescent spots being observed in thecytoplasm. Larger fluorescent granules first appeared at 30 minpost-transfection and only appeared in the nucleus at 2 h post-transfection. The number and intensity of fluorescent granules inboth the cytoplasm and nucleus gradually increased with timeduring the first 3 h. At 4 h, most of the polyplexes were located inthe nucleus but still in the complex form. At 6 h, nuclear aggregatesappeared disrupted and fluorescence began to fade and disappear,indicating dissociation of the polyplexes and release of free DNA(Fig. 6A, and Supplementary Data, Fig. S6).

To better understand whether FPP/PEI/mcDNA polyplexes effi-ciently escape from endo/lysosomes, Lyso Tracker Red was used tolabel late endosomes or lysosomes of HeLa cells. 2.5 h post-trans-fection with FITC-labeled FPP/PEI/mcDNA polyplexes, most greenspots (FITC) were clearly separated from the red ones (LysoTracker), indicating that the polyplexes efficiently escaped fromendo/lysosome, probably due to the “proton sponge” effect of PEI[8,47,48].

3.5. Delivery of polyplexes in vivo

To determine whether the targeted polyplexes are able to reachspecific tumors while in the circulation system, we investigated thedistribution of the polyplexes in the bodies of living animals byoptical imaging and by examining luciferase activity in primarytissues.

First, we evaluated the whole-body distribution of non-targetedand targeted polyplexes in FRþ HeLa xenografted mice aftersystemic injection of polyplexes containing EMA-labeled mcDNA.The main signals in mice treated with FPP/PEI/mcDNA polyplexeswere from tumor sites (Fig. 7A and Supplementary Data, Fig. S8B).In contrast, mcDNA delivered by non-targeted particles was mainly


Fig. 3. Gene transfection efficiency detected by the luciferase activity test in (A) HeLa, (B) SKOV3, (C) HepG2, and (D) A431 cells at various N/P ratios of polyplexes. The results arepresented as mean values � SD.


ARTICLE IN PRESS

found in the lung, and only trace amounts in tumors (Fig. 7A andSupplementary Data, Fig. S8A). This was consistent with resultsfrom luciferase assays. Gene expression of targeted polyplexes intumors was about 3.6 fold higher than that of non-targeted poly-plexes (p < 0.05), whereas gene expression in normal organs wasgreatly decreased, especially in the lung, heart and liver (Fig. 7C).

To further investigate the tumor-specific targeting capability ofFPP/PEI/mcDNA polyplexes, the distribution of polyplexes in micewith FRþ and FR� tumors was studied. Mice bearing FRþ HeLatumors under two sides, and those bearing FRþ HeLa tumors underthe left axillary fossa and FR� H22 tumors under the right axillaryfossa, were injected with FPP/PEI/EMA-mcDNA polyplexes via thetail vein. We found that the signal intensity from FRþ tumors wasstronger than that from FR� tumors (Fig. 7B bottom). For the micebearing FRþ HeLa tumors in both sides of the axillary fossa, noobvious difference in the signal intensity between tumors in the leftand right axillary fossa was observed (Fig. 7B top). These results


indicate that the targeted polyplexes are preferentially accumu-lated in FRþ tumors. In addition, luciferase activity assay revealedthat gene expression in FRþ tumors was 2.2 times higher than thatof FR� tumors (p < 0.1) (Fig. 7D).

3.6. Gene expression with mcDNA and conventional plasmids

To further evaluate the transfection properties of mcDNAcompared with conventional plasmids, in vitro transfection effi-ciency was measured in HeLa and SKOV3 cells. We compared mc-luc and its derived plasmid DNA (pShuttle-luc), both of whichcontained the same luciferase gene and cytomegalovirus promoter.Cells treated with mc-luc were divided into three groups: mc-A(equal weight of plasmid DNA and mcDNA), mc-B (weight ofplasmid DNA and mcDNA adjusted with stuffer DNA to give anequal molar ratio of the luciferase gene), mc-C (equal molar ratiosof the luciferase gene, but weight of plasmid DNA and mcDNA was


Fig. 4. Folic acid competition study. (A) HeLa and (B) SKOV3 cells were incubated in RPMI 1640 medium with different folic acid concentrations and treated with FPP/PEI/mc-luc orPEI/mc-luc polyplexes at an N/P ratio of 15. Luciferase expression was detected at 36 h post-transfection. RLU values are presented as mean values � SD. *p < 0.05, **p < 0.01.


ARTICLE IN PRESS

not adjusted with stuffer DNA). All treatments were carried out byFPP/PEI at a N/P ratio of 15. The results indicate that 72 h aftertransfection, luciferase expression in mc-A, mc-B, or mc-C treatedHeLa and SKOV3 cells were significantly higher than that in cells

Fig. 5. Analysis of endocytosis mediated by folate-receptors. HeLa cells were treated with (A30 min in 1 mg/L folate and folate-free medium, respectively. Cells were collected and the inas controls.


treatedwith pShuttle-luc (5.4, 4.4, and 5.8 fold in HeLa cells; 4.6, 2.6and 8.4 fold in SKOV3 cells, respectively) (Fig. 8A, B). The highexpression efficiency of mcDNA is in agreement with earlierresearch [30,32]. Luciferase expression of mc-C was higher than

) FITC-labeled FPP/PEI/mcDNA polyplexes or (B) FITC-labeled PEI/mcDNA polyplexes forternalization efficiency was measured using flow cytometry. Untreated cells were used


Fig. 6. Tracking the intracellular pathway of double-labeled FPP/PEI/mcDNA polyplexes. (A) Application of FITC-labeled FPP/PEI and EMA-labeled mcDNA polyplexes to HeLa cells.After the times indicated, cells were stained with DAPI and fixed, and images were observed by confocal microscopy. I: Green shows FITC-labeled FPP/PEI; II：Red shows EMA-labeled mcDNA; III: Blue indicates DAPI-stained nucleus; IV: Yellow indicates polymers associated with mcDNA; V: Bright-field images. (B) Tracking FITC-labeled FPP/PEI/mcDNApolyplexes in HeLa cells with labeled lysosomes. Cells were treated with FITC-labeled FPP/PEI/mcDNA and Lysotracker Red DND-99, and examined by confocal microscopy. I: FITC-labeled FPP/PEI (green); II: Lyso Tracker Red labeled lysosomes (red); III: Overlay of I and II yellow indicates the colocalization of polymers and mcDNA; IV: Bright-field images.


ARTICLE IN PRESS

that of mc-A and mc-B, may be in part due to the lower cytotoxicityinduced by the polyplexes in this group [30,32].

In order to compare the expression levels of mcDNA andconventional plasmids in vivo, HeLa cell-xenografted mice were


treated with FPP/PEI/mc-luc polyplexes and FPP/PEI/pShuttle-lucpolyplexes injected via the tail vein. After 7 days, luciferaseexpression in the tumors of mice treated with FPP/PEI/mc-lucpolyplexes was 2.3 times higher than that in those treated with


Fig. 7. Distribution of polyplexes in tumor-bearing mice. (A) Representative images of mice bearing FRþ HeLa tumors after injection of non-targeted or targeted polyplexes. (B)Representative images of mice bearing FRþ HeLa tumors under two sides, and bearing FRþ HeLa tumors under the left axillary fossa and FR� H22 tumors under the right axillaryfossa, after injection of the targeted polyplexes. Polyplexes containing EMA-labeled mcDNA were injected via the tail vein, and mice were anesthetized 4 h after the treatment,optical images were recorded using an in-vivo imaging system. Tumor sites are shown in red boxes. Fluorescence intensity is expressed by different colors with blue representingthe lowest intensity and red representing the highest intensity. The bottom panel shows luciferase activity detected in (C) mice bearing FRþ HeLa tumors after injection of non-targeted or targeted polyplexes, (D) mice bearing FRþ HeLa tumors under the left axillary fossa and FR� H22 tumors under the right axillary fossa, after injection of the targetedpolyplexes. Data are presented as mean values � SD (n ¼ 3). **p < 0.01.


ARTICLE IN PRESS

FPP/PEI/pShuttle-luc polyplexes (p < 0.01) (Fig. 8C), indicating, inagreement with previous research [32,34,35], that mcDNA couldachieve higher expression efficiencies than conventional plasmids.

4. Discussion

We combined folate/PEG modified PEI and mcDNA to developa new tumor-targeted systemic delivery system. We demonstratethat FPP/PEI/mcDNA ternary polyplexes exhibit specific tumor-targeting capability and higher levels of transgene expression


compared with the polyplexes containing conventional plasmids,indicating the FPP/PEI/mcDNA polyplexes possess qualities that arepromising in tumor therapy.

The (FAePEG)2.3ePEI conjugate (FPP) was synthesized bymodifying an approach that was used for the preparation of fola-teePEGechitosan conjugates [49]. FPP alone shows very lowtransfection efficiency compared to branched PEI 25K. It is probablethat the highly-dense PEG layer shielding the polyplexes reducedtheir cellular association and internalization. In addition, the PEGlayer shielding the polyplexes may have inhibited endosomal


Fig. 8. Comparison of luciferase gene expression using different vehicles. Cells treated with mc-luc were divided into three groups: mc-A (equal weight of plasmid DNA and mcDNA), mc-B (weight of plasmid DNA and mc DNA adjusted with stuffer DNA to give an equal molar ratio of the luciferase gene), mc-C (equal molar ratios of the luciferase gene, butweight of plasmid DNA and mc DNA was not adjusted with stuffer DNA). Luciferase activity was detected in (A) HeLa and (B) SKOV3 cells treated with FPP/PEI/mc-luc polyplexes atan N/P ratio of 15 for the durations indicated. (C) In vivo luciferase expression detected 7 days after intravenous administration of targeted polyplexes. The results are presented asmean values � SD. *p < 0.05; **, p < 0.01, when compared with the pShuttle-luc group.


ARTICLE IN PRESS

escape of the polyplexes or release of their DNA [16,17,45,46]. Inorder to solve this problem, FPP was mixed with intact PEI 25K atvarious ratios to condense DNA. On the basis of our preliminaryscreening results, an FPP/PEI/DNA ternary systemwith a ratio of FPPto PEI of 15:85 was used here. At this ratio, the ternary polyplexesformed still showed a moderately positive surface charge whichfacilitated cellular association and uptake, as well as endosomalescape. In addition, appropriate PEGylation and folate modificationwere helpful for circulation in blood and accumulation in tumorsites in a receptor-mediated way. This is consistent with resultsreported by other scientists [12,19,46,50]. PEGylation also played animportant role in maintaining cell viability (Fig. 2) [23,43]. Theeffect of folate ligand was confirmed by various comparativeexperiments including a folate competition assay (Fig. 4) and flowcytometry (Fig. 5, Supplementary Data, Table S1), as well as lucif-erase comparative assays in FRþ and FR� cell lines (Fig. 3 andSupplementary Data, Fig. S5). These results are in agreement withreport on receptor-mediated cellular uptake of folate-conjugatedfluorescent nanodiamonds [51].

Results from our investigation of the endocytosis pathway ofpolyplexes showed that FPP/PEI/mcDNA polyplexes were efficientlyinternalized into all cells and a large portion of polyplexes escapedfrom the endo/lysosome, thereby escaping lysosomal degradationand promoting transfection [47,48]. Endo/lysosome escape isa critical step for transfection process, probably due to the capacityof FPP/PEI to buffer the endosomal environment, prompting theosmotic swelling of the vesicle and finally its rupture [52] or thedirect interaction of aggregates with the inside of the endosomalmembrane, inducing local membrane damages [47,48]. Thoughsome investigations demonstrated that only DNA reached thenucleus [53e55] and cells with nuclear labeling were necrotic cells[47,56], we observed FPP/PEI were associated with mcDNA beforebeing separated inside the nucleus and these cells still maintaineda normal state. This is in line with conclusions published by Godbeyet al. and Pollard et al. [9,36]. Bieber et al. reported that thedissociation of plasmid DNA from polyplexes is not necessary fortranscription [47], but our experiment demonstrates that disrup-tion of the polyplexes occurred in nucleus. It is might be interpretedthat mcDNA is easier to dissociate from polyplexes as the interac-tion of mcDNA with positively charged FPP/PEI is not as strong asthat of plasmid. Taken together, FPP/PEI/mcDNA polyplexes are anefficient delivery system to transport transgene from the extracel-lular space to the nucleus.


FPP/PEI/mc polyplexes have high transfection efficiency andexcellent targeting capability in vitro. Further experiments werecarried out to evaluate targeting efficacy of this delivery system intumor-bearing mice. The polyplexes were administered to mice viathe tail vein. In general, the DNA concentration of polyplexes for invivo tests is 200 mg/mL [46], much higher than those applied in vitro(20 mg/mL). Large aggregates were formed at such high DNAconcentrations as was the case in a previously published protocol[8] (Supplementary Data, Fig S7c). These large particles are easilyremoved by macrophages [57], and may cause obstruction in thefine capillary beds of the lung and possibly induce severe toxicity[12]. In order to prevent polyplex aggregation at high concentra-tions, we prepared polyplexes by amodified procedure as describedin the Supplementary data. Briefly, the polyplexes were prepared insalt free buffer with 5% glucose, followed by being concentratedfrom 20 mg/mL to 200 mg/mL using centrifugal filter. Results showedthat the polyplexes formed were uniform, without aggregation(Supplementary Data, Fig. S7) and were well tolerated withoutvisible toxicity in animal experiments.

In vivo data further supported our approach and showed thatthe use of targeted polyplexes resulted in preferential accumulationin FRþ tumors compared with non-targeted polyplexes, andenhanced their internalization by cancer cells. It was safer as itminimized adverse side-effects on healthy organs and cells, andwas more efficient as it maximized therapeutic effects on tumorcells (Fig. 7C) [12]. Relatively high accumulation of modified poly-plexes was also detected in FR� tumors. This may be due to a so-called “enhanced permeation and retention”, resulting in theaccumulation of polyplexes in the 50e100 nm size range at tumorsites [58,59]. However, targeted polyplexes demonstrated prefer-ential accumulation in FRþ tumors compared to FR� tumors, sug-gesting that folate ligands conjugated to the polyplexes playeda vital role in enhancing cellular uptake into cancer cells [60]. Thuswe have successfully delivered mcDNA to tumors over-expressingfolate receptors specifically using FPP/PEI. The data is consistentwith the results for plasmid vectors used in targeted polyplexes[12,19,61].

With respect to biological properties, mc-A, mc-B and mc-Cdemonstrated 2e8 fold increased luciferase activity compared withthe plasmid, consistent with Darquet’s reports [27,31]. Higherluciferase expression suggests that transgenes are either trans-fected into cells at higher levels or that the genes are capable ofhigher levels of expression [62]. Here, evenwhen equal numbers of



ARTICLE IN PRESS

moles of pShuttle-luc and mc-luc were transfected into cells, geneexpression of mcDNAwas still superior to that of plasmid DNA. Thiscan be attributed to the robust expression of mcDNA. According toa recent report, this may be due to the fact that mcDNA does notcontain extraneous plasmid sequences which could cause tran-scriptional repression via the formation of repressive heterochro-matin which can spread and then inactivate transgenes [63]. Ina previous study, systemic delivery of mcDNA was mediated byhydrodynamic injection or PEI, and high level of gene expressionwas achieved in liver and blood, respectively [28,29,34]. However,the study on mcDNA in a tumor-targeted gene therapy system hasnot been addressed. Here, we assessed the potential of mcDNA inthe system after intravenous administration. Consistent withprevious research [31,32], our experiments showed that luciferaseexpression mediated by mcDNA in tumors was 2.3 times higherthan that mediated by plasmid DNA 7 days after systemic treat-ment. In vitro and in vivo results both suggest that mcDNA can resultin higher levels of gene expression compared with conventionalplasmids in this tumor-targeted delivery system [27,28,30].

5. Conclusion

In this study, we described the targeted delivery of mcDNA totumors and tumor specific cells using FPP/PEI. This approachresulted in enhanced tumor specific gene expression of targetedpolyplexes compared to conventional plasmid systems. This systemis a promising approach for tumor gene therapy, since the mcDNAcan be universally used for systemic tumor gene therapies by tar-geted non-viral vectors.

Acknowledgements

We thank Dr. Zhiying Chen (Stanford University, Stanford, CA)for his generous gift of p2FC31 and advice on this work; Mr. YangWang (Peking University, Beijing, PR China) for his support of FITC-PEI; Dr. Christopher Vavricka for his comments on the manuscript.This work was supported by CAS Innovation Program (No. KSCX1-YW-R-10), the National Natural Science Foundation of China (No.30901751, No. 30973448, and No. 20874001), and National BasicResearch Program 973 (2010CB529904).

Appendix. Supplementary data

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.biomaterials.2010.04.042.

Appendix

Figures with essential colour discrimination. Figs. 1, 5e7 of thisarticle may be difficult to interpret in black and white. The fullcolour images can be found in the on-line version at doi:10.1016/j.biomaterials.2010.04.042.

References

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

[2] Niidome T, Huang L. Gene therapy progress and prospects: nonviral vectors.Gene Ther 2002;9:1647e52.

[3] Putnam D. Polymers for gene delivery across length scales. Nat Mater2006;5:439e51.

[4] De Laporte L, Cruz Rea J, Shea LD. Design of modular non-viral gene therapyvectors. Biomaterials 2006;27:947e54.

[5] Mintzer MA, Simanek EE. Nonviral vectors for gene delivery. Chem Rev2009;109:259e302.

[6] Gref R, Lück M, Quellec P, Marchand M, Dellacherie E, Harnisch S, et al.‘Stealth’ corona-core nanoparticles surface modified by polyethylene glycol(PEG): influences of the corona (PEG chain length and surface density) and of


the core composition on phagocytic uptake and plasma protein adsorption.Colloids Surf B Biointerfaces 2000;18:301e13.

[7] Sonawane ND, Szoka Jr FC, Verkman AS. Chloride accumulation and swellingin endosomes enhances DNA transfer by polyamine-DNA polyplexes. J BiolChem 2003;278:44826e31.

[8] Boussif O, Lezoualc’h F, Zanta MA, Mergny MD, Scherman D, Demeneix B, et al.A versatile vector for gene and oligonucleotide transfer into cells in cultureand in vivo: polyethylenimine. Proc Natl Acad Sci U S A 1995;92:7297e301.

[9] Pollard H, Remy JS, Loussouarn G, Demolombe S, Behr JP, Escande D. Poly-ethylenimine but not cationic lipids promotes transgene delivery to thenucleus in mammalian cells. J Biol Chem 1998;273:7507e11.

[10] Neu M, Fischer D, Kissel T. Recent advances in rational gene transfer vectordesign based on poly(ethylene imine) and its derivatives. J Gene Med2005;7:992e1009.

[11] Kunath K, von Harpe A, Fischer D, Petersen H, Bickel U, Voigt K, et al. Low-molecular-weight polyethylenimine as a non-viral vector for DNA delivery:comparison of physicochemical properties, transfection efficiency and in vivodistribution with high-molecular-weight polyethylenimine. J Control Release2003;89:113e25.

[12] Ogris M, Brunner S, Schuller S, Kircheis R, Wagner E. PEGylated DNA/trans-ferrin-PEI complexes: reduced interaction with blood components, extendedcirculation in blood and potential for systemic gene delivery. Gene Ther1999;6:595e605.

[13] Oupicky D, Ogris M, Seymour LW. Development of long-circulating poly-electrolyte complexes for systemic delivery of genes. J Drug Target2002;10:93e8.

[14] Neu M, Germershaus O, Behe M, Kissel T. Bioreversibly crosslinked polyplexesof PEI and high molecular weight PEG show extended circulation times invivo. J Control Release 2007;124:69e80.

[15] Kircheis R, Schuller S, Brunner S, Ogris M, Heider KH, Zauner W, et al. Poly-cation-based DNA complexes for tumor-targeted gene delivery in vivo. J GeneMed 1999;1:111e20.

[16] Nguyen HK, Lemieux P, Vinogradov SV, Gebhart CL, Guerin N, Paradis G, et al.Evaluation of polyetherepolyethyleneimine graft copolymers as gene transferagents. Gene Ther 2000;7:126e38.

[17] Choi YH, Liu F, Kim JS, Choi YK, Park JS, Kim SW. Polyethylene glycol-graftedpoly-L-lysine as polymeric gene carrier. J Control Release 1998;54:39e48.

[18] Fernandez-Megia E, Novoa-Carballal R, Quinoa E, Riguera R. Conjugation ofbioactive ligands to PEG-grafted chitosan at the distal end of PEG. Bio-macromolecules 2007;8:833e42.

[19] Ogris M, Walker G, Blessing T, Kircheis R, Wolschek M, Wagner E. Tumor-targeted gene therapy: strategies for the preparation of ligand-polyethyleneglycolepolyethylenimine/DNA complexes. J Control Release 2003;91:173e81.

[20] Weitman SD, Lark RH, Coney LR, Fort DW, Frasca V, Zurawski Jr VR, et al.Distribution of the folate receptor GP38 in normal and malignant cell linesand tissues. Cancer Res 1992;52:3396e401.

[21] Benns JM, Mahato RI, Kim SW. Optimization of factors influencing thetransfection efficiency of folateePEGefolate-graft-polyethylenimine. J ControlRelease 2002;79:255e69.

[22] Benns JM, Maheshwari A, Furgeson DY, Mahato RI, Kim SW. Fola-teePEGefolate-graft-polyethylenimine-based gene delivery. J Drug Target2001;9:123e39.

[23] Cheng H, Zhu JL, Zeng X, Jing Y, Zhang XZ, Zhuo RX. Targeted gene deliverymediated by folateepolyethylenimine-block-poly(ethylene glycol) withreceptor selectivity. Bioconjug Chem 2009;20:481e7.

[24] Kim SH, Mok H, Jeong JH, Kim SW, Park TG. Comparative evaluation of target-specific GFP gene silencing efficiencies for antisense ODN, synthetic siRNA,and siRNA plasmid complexed with PEIePEGeFOL conjugate. Bioconjug Chem2006;17:241e4.

[25] Kim SH, Jeong JH, Cho KC, Kim SW, Park TG. Target-specific gene silencing bysiRNA plasmid DNA complexed with folate-modified poly(ethylenimine).J Control Release 2005;104:223e32.

[26] Liang B, He ML, Chan CY, Chen YC, Li XP, Li Y, et al. The use of folateePEG-grafted-hybranched-PEI nonviral vector for the inhibition of glioma growth inthe rat. Biomaterials 2009;30:4014e20.

[27] Darquet AM, Cameron B, Wils P, Scherman D, Crouzet J. A new DNA vehicle fornonviral gene delivery: supercoiled minicircle. Gene Ther 1997;4:1341e9.

[28] Chen ZY, He CY, Ehrhardt A, Kay MA. Minicircle DNA vectors devoid ofbacterial DNA result in persistent and high-level transgene expression in vivo.Mol Ther 2003;8:495e500.

[29] Chen ZY, He CY, Kay MA. Improved production and purification of minicircleDNA vector free of plasmid bacterial sequences and capable of persistenttransgene expression in vivo. Hum Gene Ther 2005;16:126e31.

[30] Bigger BW, Tolmachov O, Collombet JM, Fragkos M, Palaszewski I, Coutelle C.An araC-controlled bacterial cre expression system to produce DNA minicirclevectors for nuclear and mitochondrial gene therapy. J Biol Chem2001;276:23018e27.

[31] Darquet AM, Rangara R, Kreiss P, Schwartz B, Naimi S, Delaere P, et al. Min-icircle: an improved DNA molecule for in vitro and in vivo gene transfer. GeneTher 1999;6:209e18.

[32] Wu J, Xiao X, Zhao P, Xue G, Zhu Y, Zhu X, et al. Minicircle-IFN gamma inducesantiproliferative and antitumoral effects in human nasopharyngeal carci-noma. Clin Cancer Res 2006;12:4702e13.

[33] Jia F, Wilson KD, Sun N, Gupta DM, Huang M, Li Z, et al. A nonviral minicirclevector for deriving human iPS cells. Nat Methods 2010;7:197e9.


http://dx.doi.org/10.1016/j.biomaterials.2010.04.042




ARTICLE IN PRESS

[34] Park JH, Lee M, Kim SW. Non-viral adiponectin gene therapy into obese type 2diabetic mice ameliorates insulin resistance. J Control Release 2006;114:118e25.

[35] Chang C-W, Christensen LV, Lee M, Kim SW. Efficient expression of vascularendothelial growth factor using minicircle DNA for angiogenic gene therapy. JControl Release 2008;125:155e63.

[36] Godbey WT, Wu KK, Mikos AG. Tracking the intracellular path of poly(eth-ylenimine)/DNA complexes for gene delivery. Proc Natl Acad Sci U S A1999;96:5177e81.

[37] Stevenson M, Ramos-Perez V, Singh S, Soliman M, Preece JA, Briggs SS, et al.Delivery of siRNA mediated by histidine-containing reducible polycations.J Control Release 2008;130:46e56.

[38] Shim MS, Kwon YJ. Controlled cytoplasmic and nuclear localization of plasmidDNA and siRNA by differentially tailored polyethylenimine. J Control Release2009;133:206e13.

[39] Atkinson SF, Bettinger T, Seymour LW, Behr JP, Ward CM. Conjugation offolate via gelonin carbohydrate residues retains ribosomal-inactivatingproperties of the toxin and permits targeting to folate receptor positive cells.J Biol Chem 2001;276:27930e5.

[40] Campbell IG, Jones TA, Foulkes WD, Trowsdale J. Folate-binding protein isa marker for ovarian cancer. Cancer Res 1991;51:5329e38.

[41] Shiokawa T, Hattori Y, Kawano K, Ohguchi Y, Kawakami H, Toma K, et al. Effectof polyethylene glycol linker chain length of folate-linked microemulsionsloading aclacinomycin A on targeting ability and antitumor effect in vitro andin vivo. Clin Cancer Res 2005;11:2018e25.

[42] Theti DS, Bavetsias V, Skelton LA, Titley J, Gibbs D, Jansen G, et al. Selectivedelivery of cb300638, a cyclopenta[g]quinazoline-based thymidylate synthaseinhibitor into human tumor cell lines overexpressing the alpha-isoform of thefolate receptor. Cancer Res 2003;63:3612e8.

[43] Liang B, He ML, Xiao ZP, Li Y, Chan CY, Kung HF, et al. Synthesis and char-acterization of folateePEG-grafted-hyperbranched-PEI for tumor-targetedgene delivery. Biochem Biophys Res Commun 2008;367:874e80.

[44] Sung SJ, Min SH, Cho KY, Lee S, Min YJ, Yeom YI, et al. Effect of polyethyleneglycol on gene delivery of polyethylenimine. Biol Pharm Bull2003;26:492e500.

[45] Oupicky D, Ogris M, Howard KA, Dash PR, Ulbrich K, Seymour LW. Importanceof lateral and steric stabilization of polyelectrolyte gene delivery vectors forextended systemic circulation. Mol Ther 2002;5:463e72.

[46] Kursa M, Walker GF, Roessler V, Ogris M, Roedl W, Kircheis R, et al. Novelshielded transferrin-polyethylene glycol-polyethylenimine/DNA complexesfor systemic tumor-targeted gene transfer. Bioconjug Chem 2003;14:222e31.

[47] Bieber T, Meissner W, Kostin S, Niemann A, Elsasser HP. Intracellular routeand transcriptional competence of polyethylenimine-DNA complexes.J Control Release 2002;82:441e54.


[48] Godbey WT, Barry MA, Saggau P, Wu KK, Mikos AG. Poly(ethylenimine)-mediated transfection: a new paradigm for gene delivery. J Biomed Mater Res2000;51:321e8.

[49] Chan P, Kurisawa M, Chung JE, Yang YY. Synthesis and characterization ofchitosan-g-poly(ethylene glycol)efolate as a non-viral carrier for tumor-tar-geted gene delivery. Biomaterials 2007;28:540e9.

[50] Tang GP, Zeng JM, Gao SJ, Ma YX, Shi L, Li Y, et al. Polyethylene glycol modifiedpolyethylenimine for improved CNS gene transfer: effects of PEGylationextent. Biomaterials 2003;24:2351e62.

[51] Zhang B, Li Y, Fang CY, Chang CC, Chen CS, Chen YY, et al. Receptor-mediatedcellular uptake of folate-conjugated fluorescent nanodiamonds: a combinedensemble and single-particle study. Small 2009;5:2716e21.

[52] Boussif O, Zanta MA, Behr JP. Optimized galenics improve in vitro genetransfer with cationic molecules up to 1000-fold. Gene Ther 1996;3:1074e80.

[53] Remy-Kristensen A, Clamme JP, Vuilleumier C, Kuhry JG, Mely Y. Role ofendocytosis in the transfection of L929 fibroblasts by polyethylenimine/DNAcomplexes. Biochim Biophys Acta 2001;1514:21e32.

[54] Jiang G, Park K, Kim J, Kim KS, Oh EJ, Kang H, et al. Hyaluronic acidepolyethyleneimine conjugate for target specific intracellular delivery of siRNA.Biopolymers 2008;89:635e42.

[55] Jeong JH, Kim SH, Christensen LV, Feijen J, Kim SW. Reducible poly(amidoethylenimine)-based gene delivery system for improved nucleus trafficking ofplasmid DNA. Bioconjug Chem 2010;21:296e301.

[56] Feng M, Lee D, Li P. Intracellular uptake and release of poly(ethyleneimine)-co-poly(methyl methacrylate) nanoparticle/pDNA complexes for genedelivery. Int J Pharm 2006;311:209e14.

[57] Yao H, Ng SS, Tucker WO, Tsang YK, Man K, Wang XM, et al. The genetransfection efficiency of a folate-PEI600-cyclodextrin nanopolymer. Bioma-terials 2009;30:5793e803.

[58] Shchors K, Evan G. Tumor angiogenesis: cause or consequence of cancer?Cancer Res 2007;67:7059e61.

[59] Davis ME, Chen ZG, Shin DM. Nanoparticle therapeutics: an emerging treat-ment modality for cancer. Nat Rev Drug Discov 2008;7:771e82.

[60] Nie S, Xing Y, Kim GJ, Simons JW. Nanotechnology applications in cancer.Annu Rev Biomed Eng 2007;9:257e88.

[61] Kircheis R, Blessing T, Brunner S, Wightman L, Wagner E. Tumor targetingwith surface-shielded ligandepolycation DNA complexes. J Control Release2001;72:165e70.

[62] Wang Y, Gao S, Ye WH, Yoon HS, Yang YY. Co-delivery of drugs and DNA fromcationic core-shell nanoparticles self-assembled from a biodegradablecopolymer. Nat Mater 2006;5:791e6.

[63] Chen ZY, Riu E, He CY, Xu H, Kay MA. Silencing of episomal transgeneexpression in liver by plasmid bacterial backbone DNA is independent of CpGmethylation. Mol Ther 2008;16:548e56.


Documents

Targeted minicircle DNA delivery using folate-poly ... · Targeted minicircle DNA delivery using folateepoly ... The investigation of the endocytosis pathway of the polyplexes showed