GENE DELIVERY

Journal of Controlled Release 156 (2011) 212222

Contents lists available at SciVerse ScienceDirect

Journal of Controlled Release

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l

Nanoparticle mediated delivery of pure P53 supercoiled plasmid DNA for

gene therapy
Vtor M. Gaspar a, Ildio J. Correia a, ngela Sousa a, Filomena Silva a, Catarina M. Paquete b,
Joo A. Queiroz a, Fani Sousa a,
a
b

CICS-UBI Centro de Investigao em Cincias da Sade, Universidade da Beira Interior, Covilh, Portugal
ITQB-UNL Instituto de Tecnologia Qumica e Biolgica, Universidade Nova de Lisboa, Oeiras, Portugal

a r t i c l e

i n f o

Article history:
Received 2 May 2011
Accepted 5 August 2011
Available online 12 August 2011
Keywords:
Gene therapy
Nanoparticles
p53 Tumor suppressor
Supercoiled plasmid DNA

a b s t r a c t
The translation of non-viral gene replacement therapies for cancer into clinical application is currently
hindered due to known issues associated with the effectiveness of plasmid DNA (pDNA) expression vectors
and the production of gene delivery vehicles. Herein we report an integrative approach established on the
synthesis of nanoparticulated carriers, in association with the supercoiled (sc) isoform purication of a p53
tumor suppressor encoding plasmid, to improve both delivery and transfection. An arginine-based
chromatographic matrix with specic recognition for the different topoisoforms was used to completely
isolate the biologically active sc pDNA. Our ndings showed that the sc topoisoform is recovered under mild
conditions with high purity and structural stability. In addition, to further enhance protection and transfection
efciency, the naked sc pDNA was encapsulated within chitosan nanoparticles by ionotropic gelation. The
mild conditions for particle synthesis used in the former technique allowed the attainment of a high
encapsulation efciency for sc pDNA (N 75%). Moreover, in vitro transfection experiments conrmed the
reinstatement of the p53 protein expression and most importantly, the sc pDNA transfected cells exhibited
the highest p53 expression levels when compared to other formulations. Overall, given the fact that sc pDNA
topoisoform indeed enhances transgene expression rates this approach might have a profound impact on the
development of a sustained nucleic acid-based therapy for cancer.
2011 Elsevier B.V. All rights reserved.

1. Introduction
The development of novel cancer gene therapy approaches based
on the re-establishment of tumor suppressor regulated molecular
pathways is raising new prospects on the outcome of an effective anticancer therapy [1].
The p53 protein is a unique tumor suppressor, procient in the
selective induction of growth arrest and apoptosis in response to
oncogenic or damage signaling, acting as a prevailing guardian against
malignant cell transformation [2]. On the other hand, it is estimated
that the p53 gene is mutated or deleted in approximately 50% of all
human cancers and its ubiquitous loss of function contributes as one
of the underling events that trigger and sustain tumorigenesis [3]. It
becomes therefore reasonable that the reinstatement of the wild-type
p53 expression and consequent reactivation of its downstream
effector pathways has impact on cancer therapy, as recently reported
[4].

Corresponding author at: CICS-UBI Centro de Investigao em Cincias da Sade,

Universidade da Beira Interior, Av. Infante D. Henrique, 6200-506 Covilh, Portugal.
Tel.: + 351 275 329 002; fax: + 351 275 329 099.
E-mail address: fani.sousa@fcsaude.ubi.pt (F. Sousa).
0168-3659/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jconrel.2011.08.007

Non-viral delivery, arises as the most suitable approach for gene

therapy due to markedly safety advantages over its viral counterpart [5].
However, an effective application of a p53 DNA-based cancer therapy
has been hampered so far by issues associated with the intracellular
delivery, transfection efciency, and the purication of plasmid DNA
(pDNA) expression vectors [5,6]. Purication of pharmaceutical-grade
pDNA is a challenging process since the downstream processing must
not be approached in an individual basis, but in an integrative
perspective that accounts for cell impurities and contaminants such as
RNA, genomic DNA (gDNA) and endotoxins that derive from the
upstream stages, causing deleterious side effects if delivered to the host
[7]. Moreover, despite the fact that pDNA is a very stable biomolecule,
during its production and recovery, it can undergo several types of stress
that may disrupt its structural stability [8]. This event must be accounted
for, since it mainly affects the sc topoisoform, the only one considered
intact and undamaged [7,8]. Nonetheless, it is important to point out
that the sc pDNA topoisoform possesses the highest transfection
efciency both in vitro [9], and in vivo [8], when compared to the open
circular (oc) or linear variants, rendering itself as a valuable alternative
to improve transgene expression at the pDNA vector level.
Considering the ever-growing need to meet the preemptive requirements of purity and structural stability, our group has recently
developed a high throughput arginine afnity chromatography-based

213

from Amersham Biosciences, (Uppsala, Sweden). Chitosan low molecular

weight (LMW), pentasodium tripolyphosphate (TPP), Anti-VE Cadherin
antibody, uorescein isothiocyanate isomer I (FITC) and cell culture
reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA). 2(4-aminophenyl ethylamine) was purchased from Acros Organics (Geel,
Belgium). (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)2-(4-sulfophenyl)-2H-tetrazolium) (MTS) was obtained from Promega
(Madison, WI, USA). A549 (non-small lung carcinoma cell line) and HeLa
(Human negroid cervix epitheloid carcinoma) cells were purchased from
ATCC, Middlesex, UK. Hoechst 33342, AlexaFluor 546, AlexaFluor
647, and transfection reagents including Lipofectamine 2000, were
obtained from Invitrogen (Carlsbad, CA, USA). Analytical grade salts were
used throughout this work.

approach to recover sc pDNA for gene therapy [10]. Afnity chromatography supports based on immobilized amino acids possess unique
features since they take advantage of naturally occurring biological
interactions between DNA and specic amino acids, purifying biomolecules from the standpoint of their biological function or chemical
structure [10,11]. In addition, the intrinsic separation selectivity that
also occurs between pDNA and contaminants, render it a noteworthy
approach to purify plasmid biopharmaceuticals that comply with the
strict regulatory directives issued for gene-based therapies [10].
However, despite these major improvements regarding vector transfection efciency and safety, its delivery towards and into the
intracellular compartment still remains a key setback, since naked
pDNA is vulnerable to degradation (e.g. by serum nucleases and shear
forces) [8], is rapidly cleared from systemic circulation (sc pDNA plasma
half-life of 0.15 min) [12], and is rather inefcient in transposing
extracellular barriers and consequently in attaining therapeutic expression levels [13,14]. These limitations regarding intracellular access,
protection and bioavailability may be overcome if pDNA is packaged and
protected in nanocarriers that can selectively encapsulate and deliver it
in the cytoplasm [15].
Non-viral nanoparticulated carriers usually include cationic liposomes and cationic polymers such as chitosan [16]. Chitosan is a
biocompatible and biodegradable polymer [17], with a cationic charged
backbone that is responsible for its higher DNA loading capacity [16,18].
In fact, the charge density is its most important physicochemical feature,
since it allows the polymer backbone not only to complex with DNA but
also to instantly gel upon contact with oppositely charged crosslinkers
and surfactants [14,19]. Taking this into account, we recently reported
the synthesis of chitosan-pDNA loaded nanoparticles formulated by
gelation with a counter polyanionic crosslinker [20].
As schematized in Fig. 1, herein, we report the development of an
integrative approach that gathers the recent progresses regarding
purication and transfection efciency of plasmid biopharmaceuticals
and the novel generation of gene delivery vehicles, effectively covering
the key issues that currently hinder the translation of non-viral gene
therapy into clinical applications.

The pcDNA3FLAGp53 plasmid was amplied in a bacterial cell

culture of Escherichia coli (E. coli) DH5 and recovered as previously
described [20].
Chromatography was conducted in a fast protein liquid chromatography system (FPLC), (Amersham Biosciences, Sweden) using a column
packed with an argininesepharose 4B gel. All experiments were
performed under controlled temperature conditions (4 C) by using a
water cooling system. The experiments were performed using two
buffer solutions: (i) 10 mM TrisHCl (pH 8.0) solution; (ii) 300 mM
NaCl solution in 10 mM TrisHCl (pH 8.0). The recovered pDNA isoforms
were subsequently precipitated using one volume isopropanol and
centrifuged for 30 min, 15,000 g, 4 C, to remove the salt present in the
pDNA samples. Afterwards, the pDNA pellet was then resuspended in
10 mM TrisHCl (pH 8.0).

2. Materials and methods

2.4. Agarose gel electrophoresis

2.1. Materials

The agarose gel electrophoresis experiments were performed using a

1% agarose gel with ethidium bromide (0.5 g/mL). Electrophoresis was
carried out at 100 V for 45 min in TrisAcetateEthylene Diamine (TAE)
buffer. The agarose gels were revealed under UV light. Lane density
measurements were performed in the software Bio-Rad Quantity One,
(Hercules, USA).

The 6.07 kbp pcDNA3FLAGp53 plasmid was purchased from

Addgene (Cambridge, MA, USA). Arginine Sepharose 4B gel was obtained

2.2. Methods
A general description of all of the used methodologies is provided
as following. Nevertheless, a more detailed description is made
available as supplementary material.
2.3. Plasmid biosynthesis and purication by afnity chromatography

2.5. Synthesis of chitosan and formulation of pDNA loaded nanoparticles

LMW chitosan with a deacetylation degree (N98%) was puried as
formerly reported [20]. Plasmid DNA loaded chitosan nanoparticles
were synthesized by the ionotropic gelation technique. For this
synthesis a 0.75 mg/ml (pH 4.9) chitosan solution and a 0.5 mg/mL
(pH 5.5) solution of the anionic crosslinker, pentasodium tripolyphosphate (TPP), were prepared. All the solutions were then ltered with
a 0.22 m lter to remove traces of solid particles. In order to promote
encapsulation, pDNA (2 mg/mL) was added to the TPP solution prior
to particle formation. Afterwards 1 mL of the pDNATPP solution was
added dropwise to 4 mL of chitosan solution, under magnetic stirring
(300 50 rpm), at room temperature, for 30 min. The formulated
nanoparticles were then pelleted by centrifugation at 17,000 g for
30 min.

Fig. 1. Schematics of an integrative approach for non-viral cancer gene therapy. (I.)
Purication of plasmid biopharmaceuticals and recovery of the sc pDNA topoisoform
using arginine afnity chromatography; (II.) Nanoparticle mediated delivery and
transfection; (III.) Expression of the p53 tumor suppressor.

2.6. Morphology of the nanoparticles

Particle morphology was analyzed by scanning electron microscopy
(SEM) using a Hitachi S-2700 (Tokyo, Japan) electron microscope

GENE DELIVERY

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

GENE DELIVERY

214

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

working at 20 kV and with different magnications. Prior to image

acquisition one drop of nanoparticles was dispersed through the surface
of a cover glass and vacuum dried at 37 C overnight. Afterwards the
samples were mounted in microscope stubs and sputter coated with
gold using a sputter coater.
2.7. Nanoparticle size
Nanoparticle size was determined by dynamic light scattering (DLS).
To determine the hydrodynamic diameter by DLS, chitosan nanoparticles were diluted in 800 l of milliQ ultrapure water as a dispersant
medium, followed by 30 min incubation at room temperature prior to
analysis. Size measurements were then performed in a Zetasizer Nano
ZS instrument (Malvern Instruments, Worcestershire, UK), in automatic
mode and with a scattered light detection angle of 173. All the
measurements were performed in triplicate and measured 40 times. The
reported particle size was obtained as an intensity distribution by
cumulative analysis performed in the zetasizer software (version 6.20).
2.8. Zeta potential
For zeta potential experiments chitosan nanoparticles were prepared
as previously described for size measurements. The determination of the
zeta potential of the different naked pDNA topoisoforms was also
executed, with the pDNA samples dispersed in TrisHCl 10 mM (pH 8.0).
Zeta potential quantication was carried out in a Zetasizer Nano ZS
instrument using a zeta dip cell. The experiments were performed
in triplicate and an average of 100 measurements was acquired individually for each sample.
2.9. Encapsulation efciency of pDNA and loading capacity
To determine pDNA encapsulation efciency (EE) nanoparticle
samples were isolated by centrifugation, and the supernatant recovered
for further analysis. The concentration of unbound pDNA was measured
by UVvis analysis (Shimadzu UVvis spectrophotometer, Shimadzu
Inc, Japan) as reported in the literature [20]. Loading capacity (LC) was
determined by weighing blank tubes before the experiment and after
nanoparticle recovery.
2.10. Fluorescence labeling of pDNA
Plasmid DNA was uorescently labeled using a FITCdiazonium
salt as described elsewhere [21].
2.11. In vitro transfection and cell culture
A549, HeLa cells and rat skin Fibroblasts, isolated as previously
described [17], were cultured in 25 cm2 T-asks in Ham's F12K,
Dulbecco's Modied Eagle's Medium (DMEM) and DMEM-F12 respectively, supplemented with 10% fetal bovine serum (FBS) and 1%
antibioticsantimicotics, at 37 C, in an humidied atmosphere with
5% CO2. One day prior to transfection malignant cells were seeded in 24
well plates (4 104 cells/well). On the day of transfection, cells at 90
95% conuence were transfected with either Lipofectamine 2000 or
nanoparticles.
2.12. Cellular uptake of nanoparticles
To characterize the amount of chitosan/pDNA nanoparticles
uptake by the cells ow cytometry experiments were conducted by
using uorescently labeled FITCpDNA conjugates at different
transfection times (2 h and 6 h).The experiments were performed
on a BD FACSCalibur ow cytometer and the acquisition of the data
was made in the CellQuest TM Pro software where 1 10 4 events were
counted.

Initially, to determine the optimal gating parameters and the

possible auto-uorescence of the malignant cells a screening with
non-labeled cells was performed by accumulating the signals
corresponding to forward and side scatter measurements (FSC and
SSC) and dening the region of interest (ROI) according to a given
threshold level. After the establishment of the ROI the resulting
uorescent signals of 2 10 4 events were recorded with the FL-1
(530/30 nm) band pass lter. Flow cytometry dot plots were analyzed
in FCS Express version 4 Research Edition (De Novo Software, LA,
USA).

3. Immunouorescence and confocal laser scanning microscopy

(CLSM)
Fluorescence experiments were performed with conuent cells.
After transfection with either Lipofectamine 2000 or nanoparticles,
the cells were xed in 4% paraformaldehyde (PFA) in PBS for 20 min,
permeabilized and blocked for 3 h at room temperature. The
cells were then incubated with the primary anti-VE cadherin
monoclonal antibody (dilution 1:250) for 1 h and subsequently rinsed
10 times with PBS-Tween 20 (PBS-T) solution. To stain the nucleus of
the cells a Hoechst 33342 molecular probe was then added and
incubated for 15 min followed by 10 washing steps with PBS-T.
After nuclear staining, the cells were incubated with an AlexaFluor
546 goat anti-(rabbit IgG) conjugate for 1 h at room temperature.
The samples were then visualized using a Zeiss AX10 microscope
(Carl Zeiss, USA). Axio Vision Real 4.6 software was used for image
analysis.
To address the intracellular localization and movement of pDNA,
CLSM was performed. Conuent cell monolayer's were xed, blocked
and stained similarly to immunouorescence however with the
exception of the secondary antibody incubation, which was conducted
with a far-red AlexaFluor 647 goat anti-(rabbit IgG) antibody.
Confocal images were obtained with a Zeiss LSM 710 laser
scanning confocal microscope (Carl Zeiss SMT Inc., USA) equipped
with a plane-apocromat 63/DIC objective. To obtain enough data
for 3D reconstruction, a series of sequential slices, with different slices
thickness (m), were acquired along the Z-axis using optimized
pinhole parameters in order to comply with the Nyquist-Shannon
sampling theorem and minimize image aliasing during acquisition. All
of the acquired Z-stacks were open as a merged le in the LSM 710
software (Carl Zeiss SMT Inc., USA) where subsequent 3D reconstruction was performed. Fast maximum intensity projections (MIP) of the
reconstructed images were obtained using Huygens Essential software (Scientic Volume Imaging, Hilversum, The Netherlands).

3.1. Reinstatement of p53 expression in malignant cells

The re-establishment of the expression of the p53 tumor
suppressor was determined by indirect ow cytometry. Twenty four
hours after transfection, cells were xed with 4% PFA for 15 min at
room temperature, permeabilized with 1% Triton X-100 for 5 min, and
trypsinized using Trypsin/EDTA. The cell suspension was then
pelleted and resuspended in ice cold PBS, 5% FBS, followed by
incubation with rabbit anti-p53 antibody (10 g/mL) (sc-6243, Santa
Cruz Biotechnology, CA, USA) for 1 h at room temperature. After
incubation the cells were spun down and resuspended in PBS 3 times
to remove unbound anti-p53. Afterwards the cells were incubated
with an anti-rabbit Alexa 647 secondary antibody for 1 h and washed
as formerly described. Fluorescence was then detected with the FL-4
(661/16) band pass lter on a BD FACSCalibur ow cytometer where
1 10 4 events were recorded. Data analysis was performed in FCS
Express version 4 Research Edition (De Novo Software, LA, USA) and
Treestar FlowJo software version 7.6.1 (Ashland, OR, USA).

215

3.2. Quantication of apoptosis

P53 mediated apoptosis in transfected A549 and HeLa malignant
cells was assessed by the Annexin VFITC and PI kit (Calbiochem,
USA) according to the manufacturer's instructions. Annexin VFITC/PI
labeled cells were excited with a 15 mW laser at 488 nm and the
resulting uorescent signals of 2 10 4 events were recorded with FL-1
(530/30 nm) and FL-2 (585/42) band pass lters. Non transfected
cells were used as controls. Flow cytometry dot plots were analyzed in
FCS Express version 4 Research Edition (De Novo Software, LA,
USA).
3.3. Cytotoxicity assays
The cellular toxicity of the different formulations of nanoparticles
(native, oc and sc) was determined by the MTS assay, which was
performed both in HeLa cells and rat skin Fibroblasts, according to the
manufacturer instructions. Twenty four hours prior to the experiment
the cells were seeded at a density of 2 10 4 cells per well into 96-well
at bottom culture plates with 200 L of cell culture medium
supplemented with 10% FBS, without antibiotics. On the day of the
experiment the culture medium was aspirated and replaced by fresh
medium. The cells were then incubated with 30 L of nanoparticle
formulations for 24 and 72 h. All the formulations of nanoparticles were
resuspended in pre-warmed culture medium containing 10% FBS and
then added to each well. A total of ve replicates were considered for
each formulation.
3.4. Statistical analysis
Comparison between multiple plasmid formulations was performed using one-way analysis of variance (ANOVA), with the
StudentNewmanKeuls test. Comparisons between plasmid formulations were determined using a two-tailed Student's t-test. A value of
P b 0.05 was considered statistically signicant.
4. Results

Fig. 2. Selective separation of oc and sc topoisoforms by arginine afnity chromatography. (A) Chromatographic prole. Peak 1: weakly bound species; Peak 2: tightly
bound species. (B) Electrophoresis of peaked fractions. Lane MW: molecular weight
marker; Lane N: native pDNA sample. Lane 1: Peaked fraction 1; Lane 2: Peaked fraction
2. (C) Density analysis of the corresponding lanes in agarose gel electrophoresis.

4.1. Purication of sc pDNA by arginine afnity chromatography

Purication of pDNA by arginine afnity chromatography presents
signicant advantages over the existing strategies because sc pDNA is
recovered with high yield, structural stability and in a single
purication step as it was recently reported for a model plasmid
(pVAX1LacZ) [9]. However, given the fact that specic and reversible
bioafnity interactions are involved in the recognition of the different
topological isoforms, inherent particularities in argininepDNA interactions arise for different pDNA expression vectors. For this reason,
the adjustment of the conditions that promote the recovery of the
different topoisoforms is essential to achieve high resolution and
selectivity proles that in turn inuence the overall recovery yield and
purication degree of the plasmid of interest.
The obtained chromatographic prole when a native pDNA sample
(oc + sc) was loaded onto the arginine support is shown in Fig. 2A.
Under the reported conditions, two resolved peaks eluting at 112 mM
NaCl (peak 1) and 300 mM NaCl (peak 2) were detected. In order to
establish a correlation between the different plasmid topoisoforms
and the eluting peaks in the chromatogram, an agarose gel
electrophoresis was performed (Fig. 2B). The results revealed that
the elution of the oc topoisoform occurs in the rst peak (Fig. 2B, lane
1), at low ionic strength. Whereas, elution of the sc plasmid bound
topoisoform only takes place at higher ionic force (Fig. 2B, lane 2),
implying that the interaction with the arginineagarose matrix is
stronger. Additionally, the results presented in Fig. 2B also suggest
that sc pDNA samples are recovered with 100% homogeneity (i.e.
without any traces of either oc or linear variants). Hence, to further

support this conception, lane density was also determined, and as the
result in Fig. 2C depicts, the density peak indeed corresponds only to
the sc topoisoform, that is therefore recovered with maximum purity
and preserved structural characteristics.

4.2. Synthesis and characterization of pDNA loaded nanoparticles

All formulations of pDNA loaded nanocarriers (i.e. with native, oc
and sc samples) were synthesized by the ionotropic gelation
technique which is based on the electrostatic interactions that occur
between the positively charged polymer backbone, pDNA and a
counter anionic crosslinker, TPP, that is responsible for the spontaneous gelation of chitosan. Taking into account the possible
disturbance of the tertiary superhelical DNA structure and consequent
topoisoform conversion (sc to oc or linear conformation), during
particle synthesis, a stability study was performed prior to the
production of the different particle formulations. Our results demonstrated that the submission of pDNA to the pH range and stirring
parameters used does not promote signicant topoisoform conversion since the sc pDNA content remained higher than 95% (Supplementary Fig. S1), a value that is approximately the one issued in
regulatory directives for the use of plasmid DNA biopharmaceuticals
in therapeutic applications (sc pDNA content N 97%) [8]. It is important
to point out that the topoisoform conversion in other nanoparticle
manufacturing techniques is markedly higher [29].

GENE DELIVERY

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

GENE DELIVERY

216

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

Fig. 4. Encapsulation efciency of the nanoparticles obtained for the different pDNA
preparations; native pDNA, oc and sc. Each bar represents the mean s.d. (n = 3).
*P b 0.05, **P b 0.01.
Fig. 3. Characterization of nanoparticle morphology by SEM. (A) Nanoparticles
formulated with native pDNA (Scale bar 500 nm). (B) Nanoparticles formulated
with oc pDNA (Scale bar 500 nm). (C) Nanoparticles formulated with sc pDNA (Scale
bar 1 m).

The morphology of the different nanocarriers obtained is presented in Fig. 3. Nanoparticulated systems formulated from native and
oc pDNA preparations showed randomly shaped particles with
spherical, rod and oval-like morphologies as shown in Fig. 3A and B.
However, to address the possible inuence of the chitosan/ oc pDNA
ratio in particle morphology, different formulation ratios were also
investigated. Our ndings demonstrate that at higher ratios (5:1; 6:1)
the synthesized particles possess the same random morphological
characteristics, and at lower ratios no particles are formed under the
given formulation conditions (Supplementary Fig. S2). Whereas, the
nanoparticles obtained from the sc pDNA samples (Fig. 3C) demonstrated well dened spherical shapes.
Other relevant characteristics such as particle size, zeta potential,
and loading capacity were also determined for the different
formulations. As shown in Table 1, all of the nanocarriers demonstrated narrow size distribution with particle sizes suited for delivery
in tumoral microenvironments. Moreover, the chitosan nanoparticles
formulated with the different topological pDNA isoforms exhibit a
strong positive charge on their surface, an important feature that not
only inuences particlecell membrane interactions but also particle
colloidal stability.
Regarding the results of the loading efciency of the nanocarrier
systems it should also be pointed out that all the formulations yielded
particles comprised by a signicant content of genetic material (4458%)
when compared to the content of chitosanTPP (4256%), meaning that
the delivery vehicle not only upholds its important chitosan associated
features but is also formed by a considerable amount of the therapeutic
transgene. The ndings related with the process yield are similar for the
different formulations and in accordance with those reported in the
literature [22].

Table 1
Characterization of the different formulations of nanocarriers. Data is presented as the
mean s.d. .(n = 3).
Sample

Ratio

Particle size
DLS (nm)

Zeta
potential
(mV)

Particle loading
capacity (%)

Yield (%)

pDNA
(sc + oc)
Oc
Sc

4:1

157.6197.4

+ 20.24 18.17

51.18 6.71

26.54 4.29

251.3272.0
109.8162.5

+ 34.55 14.63
+ 22.6 4.94

41.44 0.51
42.80 2.54

21.42 0.33
20.93 1.24

4.3. Nanocarrier pDNA encapsulation ability

Fig. 4 presents the EE for the nanoparticulated systems synthesized
with the different plasmid preparations. Our results showed that the
encapsulation of sc pDNA is signicantly higher than that of native or oc
topologies. Moreover, it is important to point out as well that oc based
nanocarrier formulations possess the lowest EE of all tested samples.
Interestingly, the results are suggestive of a pattern associated
encapsulation, since the presence of the sc topoisoform (both in native
and pure sc samples) increases encapsulation. These ndings are
important for the overall formulation process and may be indicative
that condensed pDNA topology plays somehow an important role in
positive/negative, polymerDNA interactions.
In fact, to further explore this possibility the surface charge of the
pDNA biomolecules was determined. The results presented in Table 2
indeed illustrate a distinct difference regarding the electrostatic
characteristics of the various topoisoforms. These striking results will
be further discussed.
4.4. Cellular uptake of nanoparticles
The results presented in Fig. 5 characterize the cellular internalization of the different nanoparticles with either native (oc + sc), oc
and sc pDNA topological isoforms. The mean uorescence intensity
values obtained reveal a slightly increased cellular uptake for sc pDNA
nanoparticles in comparison to the other formulations at 2 h (Fig. 5).
In addition, this difference in the internalization of the sc pDNA
nanoparticles is increased after 6 h of transfection (Fig. 5). Regarding
the cellular uptake of native pDNA and oc nanoparticles, the amount
of DNA transported to the intracellular compartment is higher for the
nanoparticles synthesized with native pDNA.
4.5. In vitro delivery and intracellular trafcking
Analyzing the in vitro transfection results depicted in Fig. 6 it is clear
that either nanoparticulated carriers or commercial Lipofectamine are
capable of transporting the genetic material to the cell as previously
demonstrated in cellular uptake experiments. However, a thorough
analysis of Fig. 6B and F reveals that the dispersion of the green

Table 2
Characterization of the charge distribution of the different plasmid topological
isoforms.
Sample

Solution

Zeta potential (mV)

pDNA (oc + sc)

Oc
Sc

10 mM TrisHCl (pH = 8.0)

16.4
7.31
9.33

217

Fig. 5. Cellular uptake of the nanocarrier systems formulated with different pDNA
topoisoforms. White bars represent cellular internalization after 2 h of transfection.
Black bars represent cellular internalization after 6 h of transfection.

uorescent dots (corresponding to pDNA) is clearly increased when

nanocarrier mediated delivery takes place. The intracellular movement
and localization of pDNA can be illustrated by the Figures obtained at
2 h, 4 h and 6 h of transfection. As demonstrated in the CLSM images sc
pDNA-based delivery systems are localized within the cytoplasmic
compartment, perinuclear space and nucleus (Fig. 6O, P and Q) at 2 h, 4
and 6 h respectively. Whereas, native pDNA-based nanocarriers are only
localized in the perinuclear space, not reaching the nucleus as fast as sc

Fig. 6. Immunouorescence and CLSM of A549 cells. (A, E) Nuclear staining with
Hoechst 33342 (blue); (B, F) FITC labeled sc pDNA (green); (C, G) Staining with AntiVE cadherin-Alexa 546 antibody (red); (D, H) Merged images; (I, J, K) MIP images of
chitosan/native pDNA mediated transfection at different time-frames, 2 h, 4 h and 6 h
respectively; (L, M, N) MIP images of chitosan/oc pDNA mediated transfection at
different time-frames, 2 h, 4 h and 6 h respectively; (O, P, Q) MIP images of chitosan/oc
pDNA mediated transfection at different time-frames, 2 h, 4 h and 6 h respectively.
White arrows indicate FITCpDNA.

Fig. 7. Flow cytometry analysis of the reinstatement of p53 expression in A549 cells. (A,
D) Half overlay histograms of the re-establishment of p53 with the different pDNA
topological isoforms; (B, E) Mean uorescence intensity (MFI) values of p53 labeled
Alexa 647; (C, F) P53 relative expression mediated by the different pDNA topoisoforms
with either nanoparticles and Lipofectamine, respectively.

GENE DELIVERY

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

GENE DELIVERY

218

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

pDNA (Fig. 6K). In turn oc-based nanoparticles at 6 h post transfection

are localized in the cytoplasmic compartment.
4.6. Reinstatement of the expression of p53 tumor suppressor
As the results depicted in Fig. 7C and F demonstrate, both
nanoparticle and Lipofectamine 2000 mediated transfection lead to
the restoration of the expression of the p53 tumor suppressor.
Notwithstanding, it is important to denote that different expression
levels were obtained according to the different plasmid formulations
(native, oc or sc). In fact, this is also observed in the histograms and
mean uorescence values depicted in Fig. 7A, D and B, E respectively,
where sc pDNA mediated transgene expression levels are higher than
those of native and oc formulations, both in nanoparticle and
Lipofectamine transfected cells. Additionally the same expression
prole associated with the different pDNA topological isoforms is also
attained in western blot (Supplementary Fig. S3), thus, further
evidencing the existence of a topoisoform-associated transgene
expression prole.
4.7. P53 mediated apoptosis in malignant cells
The results in Fig. 8 demonstrate the quantication of p53
mediated apoptosis in HeLa cells. As depicted by the dot plots
(Fig. 8B, C and D), it is clear that programmed cell death occurred to a
much higher extent in the sc pDNA transfected cells in comparison to
the other formulations. Regarding the apoptosis assay in A549 cells,
much lower apoptosis was observed (4% for sc pDNA) when put
side by side with that attained for HeLa cells. However, it is important
to point out that our values were in accordance with those obtained
for p53 transfection with adenovirus [23], and hence demonstrate
that the nanoparticles are also quite effective delivery vehicles. These
results are likely a consequence of the fact that A549 cells are resistant
to p53 mediated apoptosis as recently reported [24]. It is important to
underline that despite the low apoptosis values in A549 cells the same
apoptosis pattern obtained in HeLa cells for the different formulations
was preserved, with the sc pDNA transfected cells yielding the higher
percentage of apoptotic/late apoptotic cells.
4.8. In vitro characterization of the cytotoxic prole of nanoparticles
The cellular cytotoxicity prole of all the chitosan/pDNA formulations was characterized to address whether p53 dependent
apoptosis is only correlated with the activity of the tumor suppressor
protein and not to a cytotoxic effect of the synthesized formulations.
As presented in Fig. 9A and B, at 24 h, cellular viability is clearly not
affected by the existence of the native, oc or sc pDNA nanoparticles
since the majority of cells remained viable (N95%) both in malignant
and non-malignant cells. However, in order to provide additional
insights onto the inuence of the therapeutic approach the cytotoxic
index was also determined after 72 h. This approach brought forth
signicant cell viability differences between nanocarrier formulations.
In fact, analyzing the results for malignant cells, those that were
transfected with the sc pDNA nanoparticles exhibit a much lower
viability in comparison to their topological counterparts. This is an
interesting result since this pattern is not observed in non-malignant
cells where cell viability remains unchanged and even increases at
72 h, hence, suggesting that the drop off in tumoral cells viability

Fig. 8. P53 tumor suppressor mediated apoptosis in HeLa cells induced by different
pDNA topoisoforms. Representative dot plots of Annexin VFITC (x-axis)/PI (y-axis)
double staining of (A) Untransfected cells (control); (B) Cells transfected with native
pDNA. (C) Cells transfected with oc pDNA. (D) Cells transfected with sc pDNA. (E)
Percentage of apoptotic/late apoptotic HeLa cells.

Fig. 9. MTS cytotoxicity index of the nanoparticles formulated with the different pDNA
topoisoforms, native, oc, sc, blank nanoparticles (without pDNA) in: (A) HeLa
malignant cells. (B) Rat skin broblasts. White bars represent cell viability at 24 h.
Black bars represent cell viability at 72 h. Non-transfected cells were used as negative
controls for cytotoxicity (K). Ethanol treated cells were used as positive controls for
cytotoxicity (K+).

might be correlated with the expression of the p53 therapeutic

transgene.
5. Discussion
Nowadays several approaches are being devised in order to
translate non-viral cancer gene delivery into clinical applications [1].
However, known issues associated with insufcient transfection
efciency of both pDNA vectors and delivery vehicles, has restrained
the outcome of a widespread and effective anti-cancer therapy.
In the present study, we propose a novel approach that successfully
covers the limitations associated with non-viral gene therapy, not only
improving transfection and delivery of the genetic material, but also
enhancing purity and transgene expression efciency of plasmid
vectors.
To bring about the potential of this approach, native pDNA
preparations were initially processed using a high throughput arginine
afnity chromatography support, in order to recover the plasmid
topoisoform with higher biological activity and transfection efciency
(sc pDNA) [9]. In fact, our results demonstrate that the arginine support
was indeed able to selectively recognize the different pDNA topological
conformations under the established retention/elution conditions,
since sc pDNA remained bound to the support, at low salt concentration, whilst oc pDNA was totally eluted, as depicted in the chromatographic prole of Fig. 2A. This unique selectivity feature of the
chromatographic support is a consequence of the interaction with the
nucleotides bases and pairing preference between particular amino
acids and nucleotide bases, namely arginineguanine [25]. Indeed as
previously reported by our group, arginineguanine pairing via Hbonding is the prevalent interaction when selective separation of both
topoisoforms occurs [26]. In addition, we also showed that these
interactions are stronger for sc topologies due to nucleotide base
exposure that occurs as a consequence of torsional strain deformations
[26]. Our results for the pcDNA3FLAGp53 plasmid are then in

219

accordance with those previously reported for pVAX1LacZ [27], since

the sc pDNA topoisoform is eluted at higher salt concentrations for both
plasmids, suggesting that the chromatographic support specically
recognizes and strongly binds the sc topology regardless of the used
plasmid. More importantly, as our results demonstrate the recovered sc
pDNA samples presented 100% homogeneity and up kept all of their
distinctive structural characteristics (Fig. 2B and C), that in turn largely
inuence the transgene expression efciency and thus the therapeutic
outcome.
Although sc pDNA percentage is considered the most important
quality attribute of a plasmid preparation [28], in order to translate
nucleic acid-based cancer gene therapy into clinical applications,
pharmaceutical-grade pDNA must also comply with current good
manufacturing practice (cGMP) guidelines regarding protein, gDNA,
and endotoxin levels. The latter is extremely important since the
presence of endotoxins may cause deleterious side effects such as
fever and complement system cascade which may elicit a severe
immunological response [28]. Regarding these issues, we recently
showed that the presence of contaminants in the sc pDNA sample
recovered from the arginineagarose was signicantly reduced not
only regarding the gDNA content (up to 117-fold reduction), but also
endotoxins (95% removal) and proteins (undetectable levels) [9],
therefore complying with cGMP for gene therapy applications.
From this stand point it seems undoubtedly clear that major
improvements regarding the issues associated with pDNA expression
vectors may be outdone if arginine chromatography is employed in
preparation of pharmaceutical grade sc pDNA.
However, other key issues arise since pDNA is susceptible to
degradation in the extracellular environment and rather unable to
transpose cellular barriers in its naked form [14]. Therefore, its
encapsulation in nanoparticulated delivery systems that are able to
protect and deliver the transgene of interest into the intracellular
compartment is a valuable approach. Hence, in order to synthesize
nanocarrier systems that were capable of covering the previous issues
and at the same time preserve the topological characteristics of pDNA
vectors we selected ionotropic gelation technique for nanocarrier
production because DNA encapsulation and particle formation both
take place under milder conditions than other nanoparticle formulation techniques (spray-drying, sonication, complex coacervation)
[29]. Most importantly since minor sc pDNA topoisoform conversion
occurs, the bulk of the available DNA that interacts with the cationic
polymer backbone during particle production is the biologically active
sc pDNA conformation. Regarding particle formulation our results
showed that nanocarriers with native and oc DNA samples formed
particles with random shapes (Fig. 3). On the other hand, particles
formed from sc pDNA interestingly presented dened spherical
morphologies, most likely due to the compact form of the plasmid
expression vector, suggesting that although pDNA molecular weight
remains the same, condensed pDNA topology may inuence the
thermodynamics of the prevalent electrostatic interactions, and
thereof the formation of spherical morphologies. Our results are
consistent with those obtained by Dunlap et al., 1997, that reported
the synthesis of round and globular shaped polyplexes using sc pDNA
preparations, as opposed to those with rather undened morphology,
formed from less compact topologies [30]. These ndings assume
further signicance, since as shown by Chithrani and Chan, 2007,
nanoparticles with spherical shapes are wrapped promptly by the cell
membrane (due to surface-to-volume ratio), and as a result exhibit
greater uptake when compared with those with rod-like morphology
[31]. Other relevant characteristic of the delivery system that
denitely inuences the overall transfection efciency is nanoparticle
size. The results herein obtained reveal that all of the nanoparticulated
carriers have sizes that promote their accumulation in the microenvironment that encloses tumor cells, since the blood vessels that
surround tumors possess fenestrations that range between 100 and
600 nm [32]. Nevertheless, it is also important to point out that as

GENE DELIVERY

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

GENE DELIVERY

220

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

reported by Perrault et al., 2009, smaller sized particles have the

ability to rapidly diffuse through the tumor matrix and therefore
possibly reach the target cell, increasing the therapeutic effect [33].
However, not only particle size but also the surface charge density of the
carrier system inuences gene delivery, due to the fact that nanoparticles
must also possess the ability to interact with the key extracellular barrier
to transfection, the cell membrane, and hence be internalized via
different uptake pathways [34]. This interaction is mainly governed by
electrostatic forces involving negatively charged proteoglycans in the cell
membrane and positively charged particles. In respect, to particle surface
charge our results demonstrate that all the nano delivery system
formulations possess a positive zeta potential above 20 mV and
consequently colloidal dispersion stability. Analyzing the surface charge
data present in Table 1 is quite interesting to denote that the nanocarriers
formulated with the sc pDNA topoisoform hold a lower overall value of
surface charge in comparison to the other formulations, a nding that will
be further explored.
Despite the inuence of morphology, size and zeta potential, in the
overall transfection efciency, the ability of the carrier system to
encapsulate as much genetic material as possible, is also an essential
parameter. For this reason, we also determined the EE of the
nanocarriers, in order to address if the presence of different topological
isoforms would also inuence the carrier's ability to transport pDNA. In
fact, as the results in Fig. 4 depict, EE of the sc pDNA isoform is indeed
considerably greater than that of the oc pDNA. On the other hand,
although the EE attained for nanocarriers formulated with native
plasmid preparations is smaller than that of the sc pDNA, when
compared to oc encapsulation, a signicant difference is observed. These
results are likely a consequence of the presence of both oc and sc
topoisoforms in native samples. Interestingly, akin to the differences
observed in morphology, also encapsulation is inuenced, and in this
particular case, also improved with the presence of a densely packed
DNA topology. A possible explanation for this event has been previously
proposed by Turro et al., 1991, that mentioned that the compact
geometrical distribution inuences the outer DNA electric double-layer
(Gouy-Chapman layer), mostly composed by ionic phosphate head
groups [35]. Therefore, the enhancement in sc pDNA encapsulation
levels might be explained by the torsional deformations associated with
sc topology that overexposes nucleotide bases, inuencing the negative
charge of the electrical double-layer and the extent of the interaction
with the positively charged polymer backbone. In fact, it is recognized
that sc pDNA possesses higher charge density in comparison to its
conformational variables (oc or linear) [36]. Nevertheless this issue
remains rather uncertain and to our knowledge the surface charge
density of a pure sc pDNA biomolecule has yet to be determined. Taking
this into account and in an attempt to further characterize the
supercoiled molecular assembly characteristics, the zeta potential of
the different pDNA topoisoforms was determined. Remarkably, our
results show the existence of a signicant difference in the zeta potential
between the relaxed (oc) and compact (sc) isoforms, with the latter
presenting an extra negative charge. As a result, larger amounts of
counterions are condensed by the superhelical backbone and can
therefore be released upon complexation with polycationic polymers,
promoting the establishment of favored interactions between sc pDNA
and chitosan [37]. This remarkable phenomenon is descriptive of some
biorecognition of this topoisoform from behalf of the polymer during
particle synthesis [37], and may explain the lower zeta potential value
for the sc pDNA nanoparticles. Interestingly, the lowest values of LC, EE
as well as highest particle size obtained for oc pDNA formulations might
also be correlated with the negative charge spatial distribution of the
relaxed oc biomolecules, since as mentioned for sc pDNA, the density of
the double-layer largely inuences the interactions established with the
polymer. Indeed, due to the fact that oc pDNA possesses smaller
counterion displacement capacity it has less ability to condense the
cationic chitosan backbone which results in the synthesis of larger
particles with smaller amounts of pDNA, in comparison to the other

formulations. Taking also into account the occurrence of favorable crossinteractions between sc pDNA and polycations, it is important to point
out that these interactions are responsible for the maintenance of sc
isoform structural stability after its encapsulation in the nanocarriers
systems when an excessive amount of positive charges from the
polymer are present, as reported by Bronich et al., 2000, [38]. In our
particular case since the synthesis of chitosan/pDNA nanoparticles is
accomplished in positive overcharging conditions, thermodynamic
stabilization of the sc topological conformation upon complexation is
therefore promoted [38].
Regarding the cellular uptake of the different nanoparticle formulations the results obtained are in accordance both with the EE and the
morphological characteristics of the different carriers. Particularly, sc
pDNA nanoparticles exhibited slightly higher cellular internalization
(Fig. 5). This fact is likely a consequence of the combination of the
physicochemical characteristics of the sc pDNA synthesized nanoparticles, since particle size, zeta potential and morphology markedly
inuence the overall particle uptake [39].
The encapsulation capacity of the nanoparticulated systems is also
emphasized in Fig. 6 since the amount of uorescently labeled pDNA
that reaches the intracellular compartment is superior when chitosan
based nanoparticulated systems are employed as delivery vehicles, in
detriment of cationic lipid-based nanocarriers. In fact as previously
reported, chitosan is able to condense higher amounts of pDNA than
cationic lipids [14]. All the results shown in Fig. 6 not only further
evidence that the nanocarriers formulated with the different topological
isoforms efciently transpose the extracellular and intracellular barriers
but also that the transport of the genetic material does not affect the
carrier transfection capacity. Actually, our group recently reported that
chitosan-based delivery systems also protect pDNA from the deleterious
action of DNA degrading enzymes present in serum, this is an important
feature since as mentioned earlier, the material must upkeep its
structural stability [20]. Upon intracellular localization the genetic
material is expected to follow the routes depicted in the schematics in
Fig. 1. In fact, as the MIP reconstruction images at different time-frames
depict, initially the pDNA is internalized in cell vesicles (Fig. 6I, O)
Afterwards since vesicle disruption is promoted by the nanocarrier
systems (proton sponge effect) the pDNA is localized in the cytoplasm
and then the intracellular trafcking to the nuclear periphery is initiated
(Fig. 6J and P), a transport that is thought to be essentially controlled by
microtubules [40]. Finally the genetic material is then shuttled into the
nucleus by the cell endogenous import machinery [41], (Fig. 6Q) where
it will ultimately be expressed. However, it is important to denote that
the intracellular transit of the oc pDNA /chitosan particles is relatively
slower than that of the native or sc, a fact clearly noticeable at 6 h where
these particles are still localized in the cytoplasm, whilst the sc pDNA
nanoparticles are extensively localized in the nucleus and the native
pDNA nanoparticles in the perinuclear space. These ndings may
partially explain the increased biological activity of the sc pDNA
formulation due to the fact that the kinetics of gene expression may
be different among the studied formulations.
Altogether, as the results of Fig. 7 demonstrate, p53 gene expression
is indeed re-established in malignant cells. As illustrated by Fig. 7C and F,
higher p53 protein levels are attained with sc pDNA mediated
transfection either in nanoparticulated or liposome transfected cells,
in comparison to other topologies. It should also be emphasized that
despite transfection levels of nanoparticles and Lipofectamine 2000 are
comparable, the slightly higher levels of protein expression obtained
for liposome transfected cells might be correlated with the differences
in pDNA vector unpacking properties of the nanocarriers at these
transfection times [32]. Taking this into account, it is also essential to
point out that Lipofectamine mediates transient transfection, whilst, the
transgene expression promoted by chitosan nanoparticles can achieve
quantiable protein levels in an extended time-frame as recently
reported [19], thus, rendering it a more powerful and valuable approach
for a future p53 cancer-based therapy. In accordance with our p53 sc

pDNA dependent transgene expression, Chandok et al., 2010 also

reported that gene expression in mammalian cells is enhanced by the
superhelicity of DNA, demonstrating that this topological conformation
is also responsible for the localized unwinding of the DNA molecule and
effective recruitment of the cell replication machinery [42]. Apart from
this, it is relevant to point out that this was a key experimental nding
since it emphasizes the impact of the plasmid topological conformation
not only in the synthesis of the nanoparticulated carriers but also in the
expression of the therapeutic transgene. Indeed, the analysis performed
for the p53 mediated apoptosis has revealed that the sc pDNA transfected cells were those where higher apoptosis was attained, when
compared to the other formulations (Fig. 8). The latter results are further
supported by the cytotoxic prole of the different isoforms, since
malignant cell viability decreased over time with sc pDNA-mediated
transgene expression (Fig. 9). Albeit, to exclude the possibility that the
cellular viability and consequently apoptosis of malignant cells would be
correlated with cytotoxicity of sc pDNA topoisoform non-malignant
cells were also assayed for viability and as our results demonstrate cell
viability has even increased over time, suggesting that p53-sc pDNA
mediated transgene expression indeed instigates a therapeutic response.
In summary, we have demonstrated that pharmaceutical-grade sc
pDNA is successfully recovered in a one step process, with high yield,
and topological stability using arginine afnity chromatography, a cost
effective and simple approach for purifying pDNA that can subsequently
be encapsulated in biocompatible nanocarriers. These carriers have
shown the ability to encapsulate pDNA and mediate its delivery to the
location where it will ultimately exert its therapeutic effect. To our
knowledge this was the rst time that a pure sc pDNA sample was
encapsulated onto chitosan-based nanoparticulated systems, hence this
strategy may therefore also provide the foundations for novel research
developments in non-viral gene therapy.
In conclusion, our collective approach provides valuable improvements regarding the delivery of genetic material as well as vector
associated transfection and will in a near future be used for a translational
p53-based anti-cancer therapy.
Acknowledgments
The authors would like to thank to Ana Martinho for her help with
uorescence and western blot experiments and Dr. Thomas Roberts
for providing the pcDNA3FLAGp53 construct trough Addgene, ref:
10838. The authors would also like to thank Dr. Olga Loureno for her
advice in ow cytometry experiments. This work was supported by
the Portuguese Foundation for Science and Technology (FCT), (PTDC/
EME-TME/103375/2008 and PTDC/EBB-BIO/114320/2009).
Appendix A. Supplementary data
Supplementary data to this article can be found online at doi:10.
1016/j.jconrel.2011.08.007.
References
[1] J. Roth, J. Nemunaitis, L. Ji, R. Ramesh, Tumor suppressor gene therapy, GeneBased Therapies for Cancer, Springer, London, 2010, p. 275.
[2] M. Junttila, G. Evan, p53a Jack of all trades but master of none, Nat. Rev. Cancer 9
(2009) 821829.
[3] G. Evan, K. Vousden, Proliferation, cell cycle and apoptosis in cancer, Nature 411
(2001) 342348.
[4] A.S. Azmi, M. Millard, D. Pathania, F. Grande, S. Xu, N. Neamati, H. Shen, C.G. Maki,
P. Secchiero, R. Bosco, Hot topic pharmaceutical reactivation of p53 pathways in
cancer, Curr. Pharm. Des. 17 (2011).
[5] D. Glover, H. Lipps, D. Jans, Towards safe, non-viral therapeutic gene expression in
humans, Nat. Rev. Genet. 6 (2005) 299310.
[6] P.H. Oliveira, K.J. Prather, D.M.F. Prazeres, G.A. Monteiro, Structural instability of
plasmid biopharmaceuticals: challenges and implications, Trends Biotechnol. 27
(2009) 503511.
[7] F. Sousa, D. Prazeres, J. Queiroz, Afnity chromatography approaches to overcome
the challenges of purifying plasmid DNA, Trends Biotechnol. 26 (2008) 518525.

221

[8] S.G.L. Quaak, J.H.v. den Berg, K. Oosterhuis, J.H. Beijnen, J.B.A.G. Haanen, B. Nuijen,
DNA tattoo vaccination: effect on plasmid purity and transfection efciency of
different topoisoforms, J. Control. Release 139 (2009) 153159.
[9] F. Sousa, D. Prazeres, J. Queiroz, Improvement of transfection efciency by using
supercoiled plasmid DNA puried with arginine afnity chromatography, J. Gene
Med. 11 (2009) 7988.
[10] F. Sousa, D. Prazeres, J. Queiroz, Binding and elution strategy for improved
performance of arginine afnity chromatography in supercoiled plasmid DNA
purication, Biomed. Chromatogr. 23 (2009) 160165.
[11] A. Cheng, W. Chen, C. Fuhrmann, A. Frankel, Recognition of nucleic acid bases and
base-pairs by hydrogen bonding to amino acid side-chains, J. Mol. Biol. 327 (2003)
781796.
[12] Z. Parra-Guilln, G. Gonzlez-Aseguinolaza, P. Berraondo, I. Trocniz, Gene
therapy: a pharmacokinetic/pharmacodynamic modelling overview, Pharm. Res.
(2010) 14871497.
[13] Y. Seow, M. Wood, Biological gene delivery vehicles: beyond viral vectors, Mol.
Ther. 17 (2009) 767777.
[14] R. Jayakumar, K.P. Chennazhi, R.A.A. Muzzarelli, H. Tamura, S.V. Nair, N.
Selvamurugan, Chitosan conjugated DNA nanoparticles in gene therapy, Carbohydr. Polym. 79 (2010) 18.
[15] M. Davis, Nanoparticle therapeutics: an emerging treatment modality for cancer,
Nat. Rev. Drug Discov. 7 (2008) 771782.
[16] S. Mao, W. Sun, T. Kissel, Chitosan-based formulations for delivery of DNA and
siRNA, Adv. Drug Deliv. Rev. 62 (2009) 1227.
[17] M. Ribeiro, A. Espiga, D. Silva, P. Baptista, J. Henriques, C. Ferreira, J. Silva, J. Borges,
E. Pires, P. Chaves, Development of a new chitosan hydrogel for wound dressing,
Wound Repair Regen. 17 (2009) 817824.
[18] M. Jean, F. Smaoui, M. Lavertu, S. Mthot, L. Bouhdoud, M. Buschmann, A.
Merzouki, Chitosanplasmid nanoparticle formulations for IM and SC delivery of
recombinant FGF-2 and PDGF-BB or generation of antibodies, Gene Ther. 16
(2009) 10971110.
[19] N. Csaba, M. Kping-Hggrd, M. Alonso, Ionically crosslinked chitosan/
tripolyphosphate nanoparticles for oligonucleotide and plasmid DNA delivery,
Int. J. Pharm. 382 (2009) 205214.
[20] V.M. Gaspar, F. Sousa, J.A. Queiroz, I.J. Correia, Formulation of chitosantpppDNA
nanocapsules for gene therapy applications, Nanotechnology 22 (2011) 015101.
[21] T. Ishii, Y. Okahata, T. Sato, Facile preparation of a uorescence-labeled plasmid,
Chem. Lett. 29 (2000) 386387.
[22] A. Krauland, M. Alonso, Chitosan/cyclodextrin nanoparticles as macromolecular
drug delivery system, Int. J. Pharm. 340 (2007) 134142.
[23] S. Kagawa, J. Gu, S.G. Swisher, L. Ji, J.A. Roth, D. Lai, L.C. Stephens, B. Fang,
Antitumor effect of adenovirus-mediated Bax gene transfer on p53-sensitive and
p53-resistant cancer lines, Cancer Res. 60 (2000) 1157.
[24] E. Komlodi-Pasztor, S. Trostel, D. Sackett, M. Poruchynsky, T. Fojo, Impaired p53
binding to importin: a novel mechanism of cytoplasmic sequestration identied
in oxaliplatin-resistant cells, Oncogene 28 (2009) 31113120.
[25] N. Luscombe, R. Laskowski, J. Thornton, Amino acidbase interactions: a threedimensional analysis of proteinDNA interactions at an atomic level, Nucleic
Acids Res. 29 (2001) 2860.
[26] F. Sousa, T. Matos, D. Prazeres, J. Queiroz, Specic recognition of supercoiled
plasmid DNA in arginine afnity chromatography, Anal. Biochem. 372 (2008)
432434.
[27] A. Sousa, F. Sousa, J. Queiroz, Selectivity of arginine chromatography in promoting
different interactions using synthetic oligonucleotides as model, J. Sep. Sci. 32
(2009) 16651672.
[28] Y. Cai, S. Rodriguez, R. Rameswaran, R. Draghia-Akli, R.J. Juba Jr., H. Hebel,
Production of pharmaceutical-grade plasmids at high concentration and high
supercoiled percentage, Vaccine 28 (2010) 20462052.
[29] J.S. Blum, W.M. Saltzman, High loading efciency and tunable release of plasmid
DNA encapsulated in submicron particles fabricated from PLGA conjugated with
poly-L-lysine, J. Control. Release 129 (2008) 6672.
[30] D. Dunlap, A. Maggi, M. Soria, L. Monaco, Nanoscopic structure of DNA condensed
for gene delivery, Nucleic Acids Res. 25 (1997) 3095.
[31] B. Chithrani, W. Chan, Elucidating the mechanism of cellular uptake and removal
of protein-coated gold nanoparticles of different sizes and shapes, Nano Lett. 7
(2007) 15421550.
[32] E. Gullotti, Y. Yeo, Extracellularly activated nanocarriers: a new paradigm of
tumor targeted drug delivery, Mol. Pharm. 6 (2009) 10411051.
[33] S.D. Perrault, C. Walkey, T. Jennings, H.C. Fischer, W.C.W. Chan, Mediating tumor
targeting efciency of nanoparticles through design, Nano Lett. 9 (2009)
19091915.
[34] M. Morille, C. Passirani, A. Vonarbourg, A. Clavreul, J. Benoit, Progress in
developing cationic vectors for non-viral systemic gene therapy against cancer,
Biomaterials 29 (2008) 34773496.
[35] N. Turro, J. Barton, D. ToMALIA, Molecular recognition and chemistry in restricted
reaction spaces. Photophysics and photoinduced electron transfer on the surfaces
of micelles, dendrimers, and DNA, Acc. Chem. Res. 24 (1991) 332340.
[36] G.S. Manning, The molecular theory of polyelectrolyte solutions with applications
to the electrostatic properties of polynucleotides, Q. Rev. Biophys. 11 (1978)
179246.
[37] T.K. Bronich, H.K. Nguyen, A. Eisenberg, A.V. Kabanov, Recognition of DNA
topology in reactions between plasmid DNA and cationic copolymers, J. Am.
Chem. Soc. 122 (2000) 83398343.
[38] B.A. Lobo, S.A. Rogers, S. Choosakoonkriang, J.G. Smith, G. Koe, C.R. Middaugh,
Differential scanning calorimetric studies of the thermal stability of plasmid DNA
complexed with cationic lipids and polymers, J. Pharm. Sci. 91 (2002) 454466.

GENE DELIVERY

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

GENE DELIVERY

222

V.M. Gaspar et al. / Journal of Controlled Release 156 (2011) 212222

[39] Q. Gan, T. Wang, C. Cochrane, P. McCarron, Modulation of surface charge, particle

size and morphological properties of chitosanTPP nanoparticles intended for
gene delivery, Colloids Surf. B Biointerfaces 44 (2005) 6573.
[40] A.M. Miller, D.A. Dean, Tissue-specic and transcription factor-mediated nuclear
entry of DNA, Adv. Drug Deliv. Rev. 61 (2009) 603613.

[41] A. Lam, D. Dean, Progress and prospects: nuclear import of nonviral vectors, Gene
Ther. 17 (2010) 439447.
[42] G. Chandok, K. Kapoor, R. Brick, J. Sidorova, M. Krasilnikova, A distinct rst
replication cycle of DNA introduced in mammalian cells, Nucleic Acids Res.
(2010).