Effects of incorporation of poly (g-glutamic acid) in chitosan / DNA complex nanoparticles on cellular uptake and transfection efficiency

Biomaterials 30 (2009) 1797–1808
Contents lists available at ScienceDirect
Biomaterials
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / b i o m a t e r i a l s
Effects of incorporation of poly(g-glutamic acid) in chitosan/DNA complex
nanoparticles on cellular uptake and transfection ef?ciency
Shu-Fen Peng a, Mei-Ju Yang a,b, Chun-Jen Su a, Hsin-Lung Chen a, Po-Wei Lee a,
Ming-Cheng Wei a, Hsing-Wen Sung a,*
a Department of Chemical Engineering, Bioengineering Program, National Tsing Hua University, Hsinchu 30013, Taiwan, ROC
b Biomedical Engineering Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan, ROC
a r t i c l e
i n f o
a b s t r a c t
Article history:
Chitosan (CS)/DNA complex nanoparticles (NPs) have been considered as a vector for gene delivery.
Received 3 October 2008
Although advantageous for DNA packing and protection, CS-based complexes may lead to dif?culties in
Accepted 7 December 2008
DNA release once arriving at the site of action. In this study, an approach through modifying their
Available online 24 December 2008
internal structure by incorporating a negatively charged poly(g-glutamic acid) (g-PGA) in CS/DNA
complexes (CS/DNA/g-PGA NPs) is reported. The analysis of small angle X-ray scattering results revealed
Keywords:
that DNA and g-PGA formed complexes with CS separately to yield two types of domains, leading to the
Chitosan
formation of ‘‘compounded NPs’’. With this internal structure, the compounded NPs might disintegrate
Poly(g-glutamic acid)
into a number of even smaller sub-particles after cellular internalization, thus improving the dissociation
Transfection ef?ciency
Cellular uptake
capacity of CS and DNA. Consequently, after incorporating g-PGA in CS/DNA complexes, a signi?cant
increase in their transfection ef?ciency was found. Interestingly, in addition to improving the release of
DNA intracellularly, the incorporation of g-PGA in CS/DNA complexes signi?cantly enhanced their
cellular uptake. We further demonstrated that besides a non-speci?c charged-mediated binding to cell
membranes, there were speci?c trypsin-cleavable proteins involved in the internalization of CS/DNA/g-
PGA NPs. The aforementioned results indicated that g-PGA played multiple important roles in enhancing
the cellular uptake and transfection ef?ciency of CS/DNA/g-PGA NPs.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
[8] and developed alternative methods of DNA packaging, adsorp-
tion and encapsulation [11]. It has been shown that DNA adsorbed
Chitosan (CS), a cationic polysaccharide, is biodegradable, non-
onto the surface of CS/alginate nanoparticles (NPs) reveals
toxic and tissue compatible [1–4]. It has the potential to condense
a signi?cant improvement in transfection ef?ciency [10]. However,
anionic DNA into a compact structure through electrostatic inter-
surface presentation of DNA may render it susceptible to enzymatic
actions and has been considered to be a good candidate as non-viral
degradation. Therefore, an ideal gene-delivery system should
vectors [5–7]. CS/DNA complexes can be readily prepared to
provide an adequate protection of loaded DNA in the course of
provide an effective protection against DNase [8,9]. CS/DNA
delivery and release it when appropriately.
complexes generally transfect cells more ef?ciently than naked
In this study, an approach for the enhancement of cellular
DNA but less than commercially available liposome formulations. It
uptake and transfection ef?ciency of CS/DNA complexes through
has been suggested that the strength of electrostatic interactions
modifying their internal structure by incorporating a negatively
between CS and DNA prevents their dissociation within cells, thus
charged poly(g-glutamic acid) (g-PGA) is reported. We demon-
precluding transcription of DNA and resulting in low transfection
strated that mixing CS, DNA and g-PGA in aqueous media led to the
[10].
formation of ‘‘compounded NPs’’ containing domains of CS/DNA
To improve the transfection ef?ciency of CS/DNA complexes,
and CS/g-PGA complexes. With this internal structure, the com-
recent studies have examined the use of low molecular weight CS
pounded NPs might produce a number of even smaller CS/DNA
complex NPs after disintegration within cells, thus enhancing the
dissociation capacity of CS and DNA due to a large speci?c surface
area (i.e., surface area per unit volume). g-PGA, a naturally occur-
* Corresponding author. Tel.: þ886 3 574 2504; fax: þ886 3 572 6832.
E-mail address: hwsung@che.nthu.edu.tw (H.-W. Sung).
ring peptide, is water-soluble, biodegradable and non-toxic. g-PGA-
0142-9612/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2008.12.019

---

1798
S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
a
b
DNA Ladder
pEGFP-N2
N/P=5/1
N/P=10/1
N/P=15/1
DNA Ladder
pEGFP-N2
N/P/C=10/1/0
N/P/C=10/1/0.5
N/P/C=10/1/1
N/P/C=10/1/2
N/P/C=10/1/4
N/P/C=10/1/6
Fig. 1. (a) Gel retardation analyses of CS/DNA complex nanoparticles prepared at different N/P ratios. Samples were run on a 0.8% agarose gel and subsequently stained using
ethidium bromide. Complete complexation of DNA was noted at an N/P ration of 10/1; (b) gel retardation analyses of CS/DNA/g-PGA nanoparticles prepared at different N/P/C ratios.
based NPs have been used as a carrier for oral delivery of insulin
internalization ef?ciency was examined using a confocal laser
[12,13] and been employed to deliver protein vaccines and
scanning microscope (CLSM) and a ?ow cytometer.
appeared to have a great potential as an adjuvant [14].
The study was to examine characteristics of the compounded
2. Materials and methods
NPs containing CS, DNA and g-PGA by dynamic light scattering
(DLS), transmission electron microscopy (TEM) and small angle
2.1. Plasmid DNA
X-ray scattering (SAXS). The potential of gene expression and
The plasmid DNAs used in the study were pEGFP-N2 (4.7 kb, coding an
transfection ef?ciency of test NPs was evaluated by ?uorescence
enhanced green ?uorescence protein reporter gene, Clontech, Palo Alto, CA, USA)
and luminance spectrometry and ?ow cytometry, while their
and pGL4.13 (4.6 kb, coding a ?re?y luciferase reporter gene, Promega, Madison, WI,
Fig. 2. TEM micrographs of CS/DNA/g-PGA nanoparticles prepared at different N/P/C ratios.

---

S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
1799
Table 1
1.0
Size, polydispersity index (PI), zeta potential and encapsulation ef?ciency (EE) of
a
pEGFP-N2 of test nanoparticles prepared at distinct N/P/C ratios (n ¼ 5).
N/P/C=10/0/6
N/P/C=10/1/0
N/P/C ratio
Size (nm)
PI
Zeta potential (mV)
EE (%)
0.9
N/P/C=10/1/1
N/P/C=10/1/2
10/1/0
140.2 Æ 7.7
0.33 Æ 0.04
31.7 Æ 0.8
97.7 Æ 0.4
10/1/0.5
135.5 Æ 3.2
0.21 Æ 0.01
35.3 Æ 0.3
97.3 Æ 0.9
0.8
N/P/C=10/1/4
N/P/C=10/1/6
10/1/1
130.8 Æ 1.3
0.20 Æ 0.01
34.5 Æ 0.2
96.1 Æ 0.6
10/1/2
132.8 Æ 4.9
0.22 Æ 0.03
33.3 Æ 1.1
94.2 Æ 0.4
10/1/4
152.5 Æ 5.1
0.16 Æ 0.02
28.7 Æ 1.2
99.5 Æ 0.1
0.7
10/1/6
204.5 Æ 3.7
0.11 Æ 0.01
18.7 Æ 0.2
99.5 Æ 0.1
0.6
USA). pEGFP-N2 and pGL4.13 were ampli?ed using DH5a and puri?ed by Qiagen
Plasmid Mega Kit (Valencia, CA, USA) according to the manufacturer’s instructions.
0.5
The purity of plasmids was analyzed by gel electrophoresis (0.8% agarose), while
their concentration was measured by UV absorption at 260 nm (Jasco, Tokyo, Japan).
0.4
2.2. Preparation of test NPs
Intensity (q)(a.u.)
The charge ratio (N/P/C) of test NPs was expressed as the ratio of moles of the
0.3
amino groups (N) on CS to the phosphate groups (P) on DNA and the carboxyl groups
(C) on g-PGA. Test NPs at various known N/P/C molar ratios (10/1/0, 10/1/0.5, 10/1/1,
10/1/2, 10/1/4 or 10/1/6) were prepared by an ionic-gelation method [9]. Brie?y, an
0.2
aqueous DNA (pEGFP-N2 or pGL4.13, 33 mg) was mixed with an aqueous g-PGA
(20 kDa, Vedan, Taichung, Taiwan) at different molar ratios (0, 6.5, 12.8, 25.6, 51.2 or
76.8 mg) with a ?nal volume of 100 ml. Test NPs were obtained upon addition of the
0.1
mixed solution, using a pipette, into an aqueous CS (15 kDa, with a degree of
deacetylation of 85%, 0.2 mg/ml, 100 ml, pH 6.0, Challenge Bioproducts, Taichung,
Taiwan) and then thoroughly mixed for 30–60 s by vortexer and left for at least 1 h at
0.0
room temperature.
0.8
1.2
1.6
2.0
2.4
2.8
3.2
3.6
4.0
2.3. Characterization of test NPs
q (nm-1)
The encapsulation ef?ciency of DNA in each studied group was estimated by
measuring the amount of DNA left in the supernatant after centrifugation [15]. The
Fig. 3. (a) SAXS pro?les of CS/g-PGA, CS/DNA and CS/DNA/g-PGA complex nano-
size distribution and zeta potential of test NPs were investigated using DLS (Zeta-
particles. The arrow marks the excess intensity arising from the CS/g-PGA domains in
sizer 3000HS, Malvern Instruments Ltd., Worcestershire, UK). The morphology of
complexes formed in the ternary system (CS/DNA/g-PGA nanoparticles); (b) schematic
test NPs was examined by TEM (JEOL, Tokyo, Japan) [16]. The retardation of DNA in
illustrations of the internal structures of CS/DNA and CS/DNA/g-PGA complex nano-
NPs prepared at various N/P/C ratios was evaluated by electrophoresis.
particles; (c) agarose gel electrophoresis of the DNA released from CS/DNA/g-PGA
The internal structure of test NPs was probed by SAXS. Aqueous suspensions of
nanoparticles prepared at different N/P/C ratios. Prior to electrophoresis, test nano-
NPs were directly introduced into the sample cell comprising two ultralene
particles were exposed in a phosphate buffered saline at pH 7.2 and then treated with
windows for SAXS measurements. SAXS experiments were performed using
restriction enzymes [BamHI (B) and HindIII (H)], simulating the pH environments in
a Bruker NanoSTAR SAXS instrument, which consisted of a Kristallo?ex K760 1.5 kW
the cytoplasms and nuclei.
X-ray generator (operated at 40 kV and 35 mA), cross-coupled Go¨bel mirrors for Cu
Ka-radiation (l ¼ 1.54 Å) resulting in a parallel beam of about 0.05 mm2 in cross
2.5. Percentage of cells transfected
section at the sample position and a Siemens multiwire type area detector with
1024 Â 1024 resolution mode. All data were corrected by the empty beam scattering
The percentage of cells transfected was quantitatively assessed at 48 h after
and the sensitivity of each pixel of the area detector. The area scattering pattern had
transfection by ?ow cytometry. Cells were detached by 0.05% collagenase. Cell
been circularly averaged to increase the ef?ciency of data collection. The intensity
suspensions were then transferred to microtubes, ?xed by 4% paraformaldehyde and
pro?le was output as the plot of the scattering intensity (I) vs. the scattering vector,
determined the transfection ef?ciency by a ?ow cytometer (Beckman Coulter,
q ¼ 4p/l sin(q/2) (q ¼ scattering angle) [17].
Fullerton, CA, USA) equipped with a 488 nm argon laser for excitation. For each
The dissociation of DNA from its vector within cells was investigated by
sample, 10,000 events were collected and ?uorescence was detected. Signals were
exposing test NPs (encapsulated with pGL4.13) in a phosphate buffered saline (PBS)
ampli?ed in logarithmic mode for ?uorescence to determine the EGFP-positive
at pH 7.2 and then treating with restriction enzymes (BamHI and HindIII, New
events by a standard gating technique. The percentage of positive events was
England Biolabs, Ipswich, MA, USA), simulating the pH environments in the cyto-
calculated as the events within the gate divided by the total number of events,
plasms and nuclei.
excluding cell debris.
2.4. In vitro transfection
2.6. Gene expression level
HT1080 (human ?brosarcoma) cells were cultured in DMEM media supple-
The gene expression levels of cells were assayed by quantifying the expressions
mented with 2.2 g/l sodium bicarbonate and 10% fetal bovine serum (FBS). Cells
of EGFP or luciferase. The expression level of EGFP was quanti?ed by comparing
were subcultured according to ATCC recommendations without using any antibiotic.
average ?uorescence of 1 Â106 cells. Brie?y, cells were treated with test NPs
For transfection, cells were seeded on 12-well plates at 2 Â 105 cells/well and
encapsulated with pEGFP-N2 or naked DNA. After 48 h, cells were detached as
transfected the next day at 50–80% con?uency. Prior to transfection, the media were
described in Section 2.5. Aliquots of 50 ml were transferred to 96-well black plates
removed and cells were rinsed twice with transfection media (DMEM without FBS,
and the ?uorescence intensity was analyzed using a multi-detection microplate
pH 6.0). Cells were replenished with 0.6 ml transfection media containing test NPs
reader (Molecular Devices, Sunnyvale, CA, USA) and normalized to the total cell
or naked DNA at a concentration of 2 mg DNA/well.
number of each sample.
At 2 h post-transfection, the transfection media containing NPs were removed,
For the expression of luciferase, cells were plated on 24-well plates and trans-
the cells rinsed twice with transfection media and re?lled with FBS-containing
fected as described in Section 2.4 with the exception that 1 mg pGL4.13 was used. The
media until analysis at 48 h after transfection. Cells were then observed under
cells transfected were lysed by 100 ml of passive lysis buffer (Promega). The cell
a ?uorescence microscope (Carl Zeiss Optical, Chester, VA, USA) to monitor any
lysate was transferred into a 1.5-ml microtube, while the cell debris was separated
morphological changes and to obtain an estimate of the transfection ef?ciency. Cells
by centrifugation (14,000 rpm, 5 min). Subsequently, a 100 ml of the luciferase assay
transfected with LipofectamineÔ 2000 (Invitrogen, Carlsbad, CA, USA) were used as
reagent (Promega) was added to a 20 ml of the supernatant and the relative lumi-
a positive control and those without any treatment were used as a negative control.
nescence of the sample was determined by a microplate luminometer (Berthold
Transfection ef?ciencies were presented by two numeric indicators: percentage of
Technologies, Bad Wildbad, Germany) and normalized to the total cell protein
cells transfected and gene expression level [18].
concentration by the Bradford method.

---

1800
S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
b
dDNA=2.24 nm
CS/DNA Complexes
Deprotonation of
CS Chains
CS/DNA
Shortly after
Domain
Endocytosis
dDNA=2.10 nm
CS/DNA/?-PGA Complexes
(Compounded NPs)
CS/DNA
Domain
Deprotonation of
Bridged CS Chains
CS/DNA and CS/?-PGA Domains
DNA
Bridged by Pendant CS Chains
CS
CS/?-PGA
Dissociation of CS/DNA and
?-PGA
Domain
CS/?-PGA Domains
DNA Ladder
pGL4.13
pGL4.13+B/H
N/P/C=10/1/0+B/H
N/P/C=10/1/0.5+B/H
N/P/C=10/1/1+B/H
N/P/C=10/1/2+B/H
N/P/C=10/1/4+B/H
N/P/C=10/1/6+B/H
c
Fig. 3. (Continued).

---

S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
1801
a
b 80
60
70
50
60
40
50
30
40
30
20
% Cells Transfected
20
10
Relative Fluorescence Intensity 10
0
0
NC
NK
10/1/0
10/1/0.5
10/1/1
10/1/2
10/1/4
10/1/6
LF
NC
NK
10/1/0
10/1/0.5
10/1/1
10/1/2
10/1/4
10/1/6
LF
N/P/C Ratio
N/P/C Ratio
Normalized Luciferase/mg Protein
c
1E+12
1E+9
1E+11
1E+10
1E+8
1E+9
1E+6
1E+5
1E+3
1E+4
1E+3
1E+2
1E+2
Normalized Luciferase/mg Protein 1E+1
1E+1
NC
NK
10/1/0
10/1/0.5
10/1/1
10/1/2
10/1/4
10/1/6
LF
N/P/C Ratio
Fig. 4. Ef?ciencies of cell transfection: (a) percentages of cells that were transfected; (b) relative ?uorescence intensities of transfected cells that expressed EGFP protein; (c)
normalized luciferase activities of transfected cells that expressed the luciferase. Cells were transfected in vitro using CS/DNA/g-PGA nanoparticles prepared at different N/P/C ratios
(n ¼ 5). NC: negative control (the group without any treatment); NK: naked DNA; LF: LipofectamineÔ 2000.
2.7. Fluorescent NP preparation, CLSM visualization and ?ow-cytometry analysis
To determine whether cell-surface proteins were involved in the uptake of test
NPs, cells were treated with trypsin (0.01%, 0.025% or 0.05% by w/v in Hanks’
Cy3-labeled CS (Cy3-CS) and FITC-labeled CS (FITC-CS) were synthesized as per
balanced salt solution) for 5 min prior to transfection [21]. Cells were then treated
the methods described in the literature [19,20]. To remove the unconjugated Cy3
with FITC-labeled NPs and analyzed by ?ow cytometry as described above.
and FITC, the synthesized Cy3-CS and FITC-CS were dialyzed in the dark against
deionized (DI) water and replaced on a daily basis until no ?uorescence was
2.8. MTT assay
detected in the supernatant. The resultant Cy3-CS and FITC-CS were lyophilized in
a freeze dryer. Cy3- and FITC-labeled NPs were then prepared as described in Section
The cytotoxicity of NPs was evaluated in vitro using the MTT assay [22]. HT1080
2.2 to track the internalization of NPs by CLSM and to quantify their cellular uptake
cells were seeded on 24-well plates at 5 Â 104 cells/well, allowed to adhere over-
by ?ow cytometry, respectively.
night and transfected by test NPs containing 1 mg DNA. After 2 h, test samples were
To track the internalization of NPs, cells were seeded on 12-well plates with
aspirated and cells were incubated for another 46 h. Subsequently, cells were
a sterile glass coverslip at 2 Â 105 cells/well and incubated overnight. Subsequently,
incubated in a growth medium containing 1 mg/ml MTT reagent for an additional
cells were rinsed twice with transfection media and transfected with Cy3-labeled
4 h; a 500 ml of dimethyl sulfoxide was added to each well to ensure solubilization of
NPs. After incubation for 2 h, test samples were aspirated. Cells were then washed
formazan crystals. Finally, the optical density readings were performed using
twice with pre-warmed PBS before they were ?xed in 4% paraformaldehyde. Finally,
a multiwall scanning spectrophotometer (Dynatech Laboratories, Chantilly, VA, USA)
the ?xed cells were examined under a CLSM (TCS SL, Leica, Germany).
at a wavelength of 570 nm.
To quantify the cellular uptake of NPs, cells were plated on 12-well plates and
transfected with FITC-labeled NPs at a concentration of 2 mg DNA/well for 1 h. After
transfection, cells were detached by 0.05% collagenase and transferred to micro-
2.9. Statistical analysis
tubes. Subsequently, cells were resuspended in PBS containing 1 mM EDTA and ?xed
in 4% paraformaldehyde. Finally, the cells were introduced into a ?ow cytometer
Comparison between groups was analyzed by the one-tailed Student’s t-test
equipped with a 488-nm argon laser.
(SPSS, Chicago, IL). All data are presented as a mean value with its standard deviation

---

1802
S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
a
Bright Field
Cy3-CS
EGFP
Merged
N/P/C
=10/1/0
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
N/P/C
=10/1/0.5
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
N/P/C
=10/1/1
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
Bright Field
Cy3-CS
EGFP
Merged
N/P/C
=10/1/2
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
N/P/C
=10/1/4
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
N/P/C
=10/1/6
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
Fig. 5. (a) Confocal images of cells transfected with Cy3-labeled CS/DNA/g-PGA nanoparticles prepared at different N/P/C ratios for 2 h; (b) EGFP expressions of cells transfected
with Cy3-labeled CS/DNA/g-PGA nanoparticles prepared at different N/P/C ratios for 48 h; (c) percentages of cellular uptake of FITC-labeled CS/DNA/g-PGA nanoparticles prepared
at different N/P/C ratios, analyzed by ?ow cytometry (n ¼ 3); (d) intracellular ?uorescence intensities of CS/DNA/g-PGA nanoparticles prepared at different N/P/C ratios determined
by ?ow cytometry. NC: negative control (the group without any treatment).

---

S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
1803
Bright Field
Cy3-CS
EGFP
Merged
b
N/P/C
=10/1/0
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
N/P/C
=10/1/0.5
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
N/P/C
=10/1/4
20.00 ?m
20.00 ?m
20.00 ?m
20.00 ?m
c 100
128
d
Negative Control
N/P/C=10/1/0
N/P/C=10/1/0.5
95
N/P/C=10/1/1
N/P/C=10/1/2
N/P/C=10/1/4
N/P/C=10/1/6
Events
90
Cell Uptake (%)
85
20
0
10
100
101
102
103
104
0
FL1 Fluorescence
NC
10/1/0
10/1/0.5
10/1/1
10/1/2
10/1/4
10/1/6
N/P/C Ratio
Fig. 5. (Continued).
indicated (mean Æ SD). Differences were considered to be statistically signi?cant
the results are shown in Fig. 1a. The CS/DNA complex with an
when the P values were less than 0.05.
N/P ratio of 5/1 was physically unstable, resulting in the
partial dissociation of plasmids. In contrast, as the N/P ratio
3. Results
was increased to 10/1, the migration of DNA was retarded
completely. Therefore, preparation of test NPs was carried out
3.1. Agarose gel retardation
using an N/P ratio of 10/1 in the subsequent experiments. As
shown in Fig. 1b, by incorporating g-PGA in NPs (N/P/C ratios
The binding capacity of CS with DNA prepared at various
of 10/1/0.5 to 10/1/6), no signi?cant DNA release was
N/P ratios was evaluated using the gel retardation assay and
observed.

---

1804
S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
128
b
a
Negative Control
120
Negative Control
N/P/C=10/1/0
Nanoparticles (NPs)
N/P/C=10/1/0+0.001% Trypsin
0.001% Trypsin Treated + NPs
110
N/P/C=10/1/0+0.025% Trypsin
0.025% Trypsin Treated + NPs
N/P/C=10/1/0+0.05% Trypsin
0.05% Trypsin Treated + NPs
100
Events
90
80
70
0
101
102
103
100
60
FL1 Fluorescence
c 128
Cell Uptake (%)
50
Negative Control
N/P/C=10/1/4
40
N/P/C=10/1/4+0.001% Trypsin
N/P/C=10/1/4+0.025% Trypsin
N/P/C=10/1/4+0.05% Trypsin
30
20
Events
10
0
NC
N/P/C
N/P/C
=10/1/0
=10/1/4
0
101
102
103
100
FL1 Fluorescence
Fig. 6. Results of intracellular uptake of CS/DNA and CS/DNA/g-PGA (N/P/C ratio of 10/1/4) nanoparticles after the cells being treated with different concentrations of trypsin,
determined by ?ow cytometry: (a) percentages of intracellular uptake of test nanoparticles (n ¼ 3); (b) ?uorescence intensities after intracellular uptake of CS/DNA nanoparticles;
(c) ?uorescence intensities after intracellular uptake of CS/DNA/g-PGA nanoparticles.
3.2. Morphology, size, zeta potential and encapsulation
3.3. Effects of g-PGA on internal structure of NPs
ef?ciency of test NPs
and their release of DNA
TEM was used to examine the morphology of test NPs
SAXS was used to examine the internal structure of test NPs.
prepared at various N/P/C ratios (Fig. 2). As shown, the CS/DNA
Fig. 3a shows the SAXS pro?les of CS/g-PGA (N/P/C ratio of 10/0/6),
complex (N/P/C ratio of 10/1/0) had a heterogeneous size distri-
CS/DNA (N/P/C ratio of 10/1/0) and CS/DNA/g-PGA (N/P/C ratios of
bution with a donut, rod or pretzel shape. In contrast, with the
10/1/1 to 10/1/6) NPs. The scattering pro?le of CS/g-PGA NPs dis-
incorporation of g-PGA, test NPs (N/P/C ratios of 10/1/0.5 to 10/1/
played a featureless monotonic decay, revealing a disordered
6) were spherical in shape with a relatively homogeneous size
internal structure. In this case, the SAXS intensity might stem from
distribution.
the characteristic concentration ?uctuations of the CS/g-PGA
The size distribution and zeta potential of the prepared NPs in
complex within NPs. The SAXS pro?le of CS/DNA NPs was found to
aqueous environment were investigated by DLS. As shown in
display a peak at 2.8 nmÀ1 associated with the spatial correlation of
Table 1, with an increase in the amount of g-PGA incorporated,
DNA in CS/DNA complexes. The characteristic spacing between
the size of NPs increased appreciably while their polydispersity
DNA calculated from the peak position (qDNA) via dDNA ¼ 2p/qDNA
index and zeta potential value decreased noticeably. The diame-
was 2.24 nm. Therefore, the DNA chains in CS/DNA complexes were
ters of NPs measured by DLS were relatively larger than those
packed densely to form a tight bundle phase, as the positively
observed by TEM. This is because the diameters of NPs obtained
charged CS chains wrapped around DNA for charge matching
by DLS re?ected their hydrodynamic diameters swelled in
caused a signi?cant aggregation of DNA.
aqueous solution, while those observed by TEM were the diam-
The DNA correlation peak associated with CS/DNA complexes was
eters of dried NPs. The encapsulation ef?ciencies of DNA in NPs
also observable in the SAXS pro?les of NPs containing CS, DNA and
prepared at distinct N/P/C ratios were about the same and
g-PGA. Interestingly, this peak was located at a higher q (3.0 nmÀ1)
approached 100%.
than that associated with the binary CS/DNA complex, showing that

---

S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
1805
?ow cytometry at 48 h post-transfection. As shown in Fig. 4a, only
110
up to 15% of the cells produced EGFP when transfected with the NPs
100
containing no g-PGA (N/P/C ratio of 10/1/0, i.e., CS/DNA NPs). By
90
incorporating g-PGA in NPs (N/P/C ratios of 10/1/0.5 to 10/1/6),
a signi?cant increase in the percentage of cells that expressed EGFP
80
was found. Transfection was increased approximately fourfold
70
(55%) for the cells transfected with the NPs with an N/P/C ratio of
60
10/1/4 compared to those treated with the NPs containing no
g-PGA (P < 0.05).
50
The results of expression levels of EGFP or luciferase of cells are
40
given in Fig. 4b and c, respectively. As shown, the EGFP expression
30
levels of cells transfected with the NPs incorporating g-PGA were
signi?cantly higher than those treated with the NPs containing no
20
Cell Viability (% of Control)
g-PGA (P < 0.05). The luciferase gene expression of cells transfected
10
by the NPs with an N/P/C ratio of 10/1/4 was about 10-fold
0
increased in comparison with those treated with the NPs contain-
ing no g-PGA (P < 0.05). These results indicated that the trans-
fection ef?ciency of NPs was signi?cantly enhanced after the
incorporation of g-PGA. Among all studied groups, the cells trans-
NC
NK
?
-PGA
CS
10/1/0
10/1/0.5
10/1/1
10/1/2
10/1/4
10/1/6
LF+DNA
fected with the NPs with an N/P/C ratio of 10/1/4 had the highest
gene expression level.
N/P/C Ratio
Fig. 7. Results of the cell viability after being treated with CS/DNA/g-PGA nanoparticles
prepared at different N/P/C ratios determined by the MTT assay (n ¼ 5). NC: negative
3.5. Cellular uptake
control (the group without any treatment); NK: naked DNA; g-PGA: poly(g-glutamic
acid); CS: chitosan; LF: LipofectamineÔ 2000.
CLSM was used to visualize the cellular uptake of Cy3-labeled
NPs and their EGFP expression. The results of ?uorescence images
the incorporation of g-PGA induced a more compact packing of DNA
of cells after exposure to NPs prepared at different N/P/C ratios are
chains (with a reduction of dDNA to 2.10 nm). Moreover, an excess
shown in Fig. 5a and b. At 2 h after transfection, accumulation of
intensity (marked by the arrow in Fig. 3a) was identi?ed at lower q for
Cy3-lableded NPs was observed in most of the incubated cells in all
the ternary system (CS/DNA/g-PGA) and it became more signi?cant
studied groups (Fig. 5a). The ?uorescence intensity observed in
at a higher content of g-PGA. The presence of this intensity contri-
cells increased notably with increasing the amount of g-PGA
bution implied the existence of a heterogeneity or domains with
incorporated in NPs. At this time, no EGFP expression was observed
a characteristic length scale larger than dDNA in NPs.
for all of the studied groups. At 48 h after transfection, it appeared
The features of SAXS patterns indicated that the NPs formed by
that the numbers of cells that expressed EGFP in the groups
the ternary system were composed of two types of domains con-
transfected with the NPs incorporating g-PGA were more than the
taining CS/DNA and CS/g-PGA complexes, as schematically illus-
group treated with the NPs containing no g-PGA (Fig. 5b).
trated in Fig. 3b. This type of NPs is called ‘‘compounded NPs’’ here.
After a 2-h transfection, the percentage of cells that internalized
The dense packing of DNA chains in CS/DNA domains gave rise to
FITC-labeled NPs and their ?uorescence intensity were quanti?ed
the scattering peak, while the CS/g-PGA domains surrounding the
by ?ow cytometry. As shown in Fig. 5c and d, the percentage of
CS/DNA domains contributed to the excess intensity observed at
?uorescent cells and their ?uorescence intensity upon internali-
lower q. In this case, the CS/g-PGA domains might create a relatively
zation of NPs were signi?cantly enhanced with increasing the
rigid environment to suppress the positional distortion of the DNA
amount of g-PGA incorporated (P
chains induced by thermal ?uctuations; as a result, the DNA chains
< 0.05). However, there were no
statistically signi?cant differences between test NPs with N/P/C
in CS/DNA domains exhibited a more compact packing. In addition,
ratios of 10/1/4 and 10/1/6 (P
we expected the presence of unbounded (or pendant) segments of
> 0.05).
To further elucidate differences in the uptake mechanism, the
CS chains emanating from the surface of CS/g-PGA complexes;
interaction of NPs with cell membranes was investigated by
these pendant CS segments might subsequently bridge with the
treating cells with trypsin at different concentrations prior to
neighboring DNA chains to from two types of domains containing
transfection. As shown in Fig. 6a, trypsinization resulted in
CS/g-PGA and CS/DNA complexes within the compounded NPs
a signi?cant decrease in the internalization of FITC-labeled NPs
(Fig. 3b).
with or without g-PGA (P
After being exposed at pH 7.2 and subsequently treated by
< 0.05). However, trypsin did not induce
a concentration-dependent effect on the uptake of the NPs con-
restriction enzymes (BamHI and HindIII), simulating the environ-
taining no g-PGA (Fig. 6a and b, P
ments in the cytoplasms and nuclei, test NPs were analyzed by gel
> 0.05), while it caused
a concentration-dependent decrease in the internalization of the
electrophoresis. As shown in Fig. 3c, there were three DNA bands
NPs incorporating g-PGA (Fig. 6a and c, P
(4.6, 2.7 and 1.9 kb, respectively, digested by restriction enzymes)
< 0.05). These results
implied that by incorporating g-PGA, test NPs might be internalized
identi?ed in gel for each studied group. With increasing the
by cells via a speci?c protein-mediated endocytosis.
amount of g-PGA incorporated in NPs, the intensities of DNA bands
observed in gel were stronger. These results implied that incorpo-
ration of polyanionic g-PGA might enhance the release of DNA
3.6. MTT assay
from NPs within cells (i.e., at pH w 7.2).
Fig. 7 shows the viability of cells cultured in the media treated
3.4. Percentage of cells transfected and gene expression level
with varying test samples. As shown, the cytotoxicity of naked DNA,
g-PGA and CS was quite low. The viability of the cells treated with
To determine the percentage of cells that actually expressed the
test NPs decreased relatively with increasing the amount of g-PGA
transgene, we counted the number of EGFP-positive cells using
incorporated, due to their high expression of EGFP.

---

1806
S.-F. Peng et al. / Biomaterials 30 (2009) 1797–1808
a
Trypsin Treated
: Test NPs (N/P/C=10/1/0)
Trypsin Treated
: Non-Specific Proteoglycans
(low concentration)
(high concentration)
: Specific Proteins
: Trypsin Treatment
: DNA
Cytoplasm
Nucleus
b
: Test NPs (N/P/C=10/1/4)
Trypsin Treated
Trypsin Treated
: Non-Specific Proteoglycans
(low concentration)
(high concentration)
: Specific Proteins
: Trypsin Treatment
: DNA
Cytoplasm
Nucleus
Fig. 8. Schematic illustrations of potential uptake pathways: (a) CS/DNA nanoparticles. The internalization of CS/DNA nanoparticles might be mainly via a non-speci?c charge-
mediated interaction between the nanoparticles (positively charged) and the components of cell membranes (negatively charged proteoglycans); (b) CS/DNA/g-PGA nanoparticles.
Besides a non-speci?c charged-mediated binding to cell membranes, there were speci?c trypsin-cleavable proteins involved in the internalization of CS/DNA/g-PGA nanoparticles.
4. Discussion
ionized. The ionized CS, DNA and g-PGA can form polyelectrolyte
complexes (CS/DNA/g-PGA NPs) by electrostatic interactions
CS/DNA complex NPs have been considered as a candidate for
between the positively charged amino groups (–NHþ
3 ) on CS and the
gene delivery [23–25]. Although advantageous for DNA packing
negatively charged phosphate groups (–POÀ
4 ) on DNA and carboxyl
and protection, CS-based complexes may lead to dif?culties in DNA
groups (–COOÀ) on g-PGA.
release once they arrive at the site of action [11], thus limiting their
The SAXS results revealed that DNA and g-PGA formed
transfection ef?ciency. In the study, we demonstrated that after the
complexes with CS separately to yield two types of domains bridged
incorporation of g-PGA in CS/DNA NPs, the percentage of cells
by pendant CS chains in the compounded NPs (Fig. 3b) [9,26]. After
transfected and their gene expression level were signi?cantly
internalization into cells, the compounded NPs would be expected
enhanced (Fig. 4a–c and 5b).
to disintegrate into a number of even smaller sub-particles
The pKa values of CS and g-PGA are 6.5 and 2.9, respectively [25].
composing CS/DNA and CS/g-PGA complexes, due to deprotonation
When prepared in DI water (pH 6.0), CS, DNA and g-PGA are
of the bridged CS chains. The subsequent release of DNA through the

---

Document Outline

• Effects of incorporation of poly(gamma-glutamic acid) in chitosan/DNA complex nanoparticles on cellular uptake and transfection efficiency
  • Introduction
  • Materials and methods
    • Plasmid DNA
    • Preparation of test NPs
    • Characterization of test NPs
    • In vitro transfection
    • Percentage of cells transfected
    • Gene expression level
    • Fluorescent NP preparation, CLSM visualization and flow-cytometry analysis
    • MTT assay
    • Statistical analysis
  • Results
    • Agarose gel retardation
    • Morphology, size, zeta potential and encapsulation efficiency of test NPs
    • Effects of gamma-PGA on internal structure of NPs and their release of DNA
    • Percentage of cells transfected and gene expression level
    • Cellular uptake
    • MTT assay
  • Discussion
  • Conclusions
  • Acknowledgments
  • flink6
  • References

---