Cellular uptake and transfection activity of DNA complexes based on poly(ethylene glycol)-poly(l-glutamine) copolymer with PAMAM G2

Article abstract of DOI:10.1039/c3tb20649a

Poly(ethylene glycol)-poly(l-glutamine) (PEG-PLGA) copolymer EA-G2 (or EA-G1) was prepared by aminolysis of poly(ethylene glycol)-poly(l-benzyl glutamate) (PEG-PBLG) using PAMAM G2 (or G1). The chemical structure of PEG-PLGA was confirmed by FT-IR, ¹H-NMR, DSC and GPC. The performances of the EA-G2 (or EA-G1) were assayed by enzyme degradation, MTT method and agarose gel electrophoresis. The particle size, zeta potential and morphology of EA-G2 (or EA-G1)/pDNA complexes were inspected by DLS and AFM. The cellular uptake mechanism was evaluated by endocytic inhibiting test, cell uptake test and observation of CLSM. The transfection activity was measured by flow cytometry. The EA-G2 (or EA-G1) exhibited good biodegradability, low cytotoxicity and great ability to combine with pDNA. The EA-G2 (or EA-G1) complexes exhibited particle sizes in the range 120-180 nm and zeta potentials in the range 20-40 mV, which were suitable for cell uptake. The cellular uptake of the EA-G2 complexes occurred mainly through clathrin-dependent and caveolin-mediated endocytosis, and at 6 h in 10% FBS and in serum-free media, the percentages of complex uptake reached 89.0% or 72.7%, respectively. EA-G2 complexes could efficiently mediate pEGFP-Cl into the cell nuclei. EA-G2 complexes displayed enhancing transfection efficiency and better serum tolerance. The results suggest that the EA-G2 has potential to be used as a biodegradable, efficient and serum-resistant gene vector.

Full text of DOI:10.1039/c3tb20649a

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
Cellular uptake and transfection activity of DNA
complexes based on poly(ethylene glycol)-poly-
(L-glutamine) copolymer with PAMAM G2†
Cite this: J. Mater. Chem. B, 2013, 1,
114
5
Shirong Pan,* Duanwen Cao, Rong Fang, Wu Yi, Huan Huang, Shouqin Tian
and Min Feng*
Poly(ethylene glycol)-poly(L-glutamine) (PEG-PLGA) copolymer EA-G2 (or EA-G1) was prepared by
aminolysis of poly(ethylene glycol)-poly(L-benzyl glutamate) (PEG-PBLG) using PAMAM G2 (or G1). The
1
chemical structure of PEG-PLGA was confirmed by FT-IR, H-NMR, DSC and GPC. The performances of
the EA-G2 (or EA-G1) were assayed by enzyme degradation, MTT method and agarose gel
electrophoresis. The particle size, zeta potential and morphology of EA-G2 (or EA-G1)/pDNA complexes
were inspected by DLS and AFM. The cellular uptake mechanism was evaluated by endocytic inhibiting
test, cell uptake test and observation of CLSM. The transfection activity was measured by flow
cytometry. The EA-G2 (or EA-G1) exhibited good biodegradability, low cytotoxicity and great ability to
combine with pDNA. The EA-G2 (or EA-G1) complexes exhibited particle sizes in the range 120–180 nm
and zeta potentials in the range 20–40 mV, which were suitable for cell uptake. The cellular uptake of
the EA-G2 complexes occurred mainly through clathrin-dependent and caveolin-mediated endocytosis,
and at 6 h in 10% FBS and in serum-free media, the percentages of complex uptake reached 89.0% or
Received 6th May 2013
Accepted 26th July 2013
7
2.7%, respectively. EA-G2 complexes could efficiently mediate pEGFP-Cl into the cell nuclei. EA-G2
DOI: 10.1039/c3tb20649a
complexes displayed enhancing transfection efficiency and better serum tolerance. The results suggest
that the EA-G2 has potential to be used as a biodegradable, efficient and serum-resistant gene vector.
www.rsc.org/MaterialsB
aggregation or dismantlement, nally leading to in vivo gene
1
Introduction
3,4
transfection in vain.
Nevertheless, polycations are still
Gene therapy is a promising method for the treatment of both
acquired and inherited diseases for which there is presently a
lack of suitable drugs. A premise of successful gene therapy is to
attractive to many researchers because of easy modication for
improvement of properties.
5
PAMAM dendrimers are a class of hyperbranched and
choose gene delivery vectors with safety and effectiveness. Viral narrowly dispersed macromolecules that are composed of
gene vector is a major gene delivery carrier due to the high gene ethylenediamine as a core and outer layer structure (genera-
transfection efficiency, but the high risk of immune reactions tions) formed by alternatively repeating addition and amidation
1
also hinders further clinical applications. Moreover, non-viral reactions. Owing to a higher pK value (about 9.2), the primary
a
gene vector provides advantages such as non-immunogenicity,
amines on the periphery of the PAMAM are easily protonated
and turned into positive charges at neutral pH, which allows the
polycation to combine with negatively charged DNA through
repeat administration, long gene expression, low cost, batch-
2
producing and so on. Some polycations including polylysine,
6,7
chitosan, polyethyleneimine (PEI) and polyamidoamine electrostatic interactions forming nano-scale complexes. Due
PAMAM) have been intensively studied over the past decade. to a lower pK value (about 6.7), the tertiary or second amines
(
a
However, two reasons limit their clinical application: one of inside the PAMAM are protonated only at acidic endosomal pH,
them is the inherent cytotoxicity of the polycations that causes which induces excessive proton uptake in the endosome and
8
massive death of cells in transfection; the other is the extrinsic disruption of the endosomal membrane by osmotic swelling.
serum inhibition, that is a non-specic interaction between
serum and DNA/polycation complexes which causes complex
The strong buffer capacity of the dendrimer facilitates endo-
somal escape. Generally, high-generation PAMAMs such as G4–
G6 exhibit moderate gene transfection efficiency and large
cytotoxicity, but low-generation PAMAMs such as G0–G3 exhibit
The First Affiliated Hospital, Sun Yat-sen University, 58 Zhongshan
2
Road,
9
low gene transfection efficiency and almost no cytotoxicity.
Guangzhou 510080, P.R. China. E-mail: gzpshr@163.com; Fax: +86 20 87330396;
Tel: +86 20 87330757; fengmin@mail.sysu.edu.cn; +86 20 39943119
Although PAMAM G0–G3 cannot be used alone for transfection,
it can oen be conjugated to parent polymers to prepare new
polycations as efficient gene vectors.
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c3tb20649a
1
5
114 | J. Mater. Chem. B, 2013, 1, 5114–5127
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry B
2 2 2
As a biocompatible and biodegradable polymer, poly- 2.73 (t, 28H, 2CH N + 12CH CH CO), 2.80–2.90 (t, 24H,
(
L-benzyl glutamate) (PBLG) is widely used for drug-sustained 4CH CH N + 8CH CH NH ), 3.30–3.40 (t, 24H, 4CH CH N +
2
2
2
2
2
2
2
10,11
delivery.
Further, the aminolysis products of PBLG, i.e. 8CH CH NH ). G2.0 IR (KBr): 3290, 3065, 2941, 2831, 1650,
2 2 2
ꢀ
1 1
poly(L-glutamine) (PLGA), have great potential to be used for 1550, 1450, 1360, 1128, 1033 cm . H NMR (D O): d
¼ 2.55
2
ppm
gene delivery due to the introduction of many amine groups (t, 56H, 28CH
into the side chains. It was reported that the PLGA prepared 2.86–2.98 (t, 56H, 16CH
from oligo-ethyleneimine (OEI) displayed enhanced trans- 16CH CH NH + 12CH CH
2.2.2 Synthesis of g-benzyl-L-glutamate (BLG). BLG was
2
CH
2
CO), 2.75 (t, 60H, 2CH
CH –NH + 12CH
N).
2
N + 28CH
2 2
CH CO),
2
2
2
2
CH N), 3.43 (t, 56H,
2
2
2
2
2
2
12–15
fection efficiency and reduced cytotoxicity.
Similarly, the
17
PLGA prepared from low generation PAMAM such as G2 (or G1) prepared according to the literature. Typically, L-glutamic acid
displayed a higher transfection efficiency, lower cytotoxicity and (29.5 g, 0.20 mol), benzene–methanol (27 g, 0.25 mol) and 60%
3
excellent serum resistant capacity.
sulfuric acid (32.5 g, 0.20 mol) were mixed and stirred in vacuo
ꢁ
In the present study, we prepared the novel PEG-PLGA at 70 C for 4 h. The crude product was poured into 200 mL of
copolymer, EA-G2 (or EA-G1) by an aminolysis reaction of ice water containing sodium bicarbonate (33.5 g, 0.40 mol) and
poly(ethylene glycol)-poly(L-benzyl glutamate) (PEG-PBLG) was allowed to stand at 4 C overnight. Aer being ltered,
ꢁ
using PAMAM G2 (or G1); characterized the chemical structure recrystallized and dried in vacuo, the product, BLG (28.0 g,
of EA-G2 (or EA-G1); evaluated the biodegradability, the cyto- 0.118 mol, yield 59.0%), was obtained.
toxicity and the gene combining capability of EA-G2 (or EA-G1);
2.2.3 Synthesis of g-benzyl-L-glutamate N-carboxyanhy-
measured the particle sizes, zeta potentials and morphology of dride (BLG-NCA). BLG-NCA was prepared according to the
18
EA-G2 complexes; evaluated the endocytic pathway, the cellular literature. Briey, BLG (10.0 g, 0.042 mol) and triphosgene
uptake kinetics and the intracellular trafficking of EA-G2 (6.0 g, 0.020 mol) were added into 100 mL of dried tetrahydro-
ꢁ
complexes; and nally measured the in vitro transfection effi- furan. Aer stirring at 65 C for 30 min, the suspension turned
ciency and the serum tolerance of the complexes.
transparent. The product was precipitated with 500 mL of cold
petroleum ether, ltered and dried in vacuo, then BLG-NCA
(
9.5 g, 0.036 mol, yield 86%) was obtained.
.2.4 Synthesis of PEG-PBLG. MPEG-NH
mmol) as initiator was added to 100 mL of dry dioxane con-
taining BLG-NCA (5.2 g, 20 mmol) at a NCA/NH molar ratio of
2
Materials and methods
2
2
-2000 (0.4 g, 0.2
2
.1 Materials
Methoxy-polyethylene glycol amine (Mn 2000, MPEG-NH
was purchased from Bomei Biological Inc. (Jiaxin, China).
-Hydroxypyridine was purchased from Fluka Inc. (UK). Other
2
-2000)
2
100 : 1. The mixture was stirred at room temperature for 3 d.
The product was precipitated with a 10-fold volume of distilled
water, ltered and vacuum-dried. PEG-PBLG copolymer (4.1 g,
yield 73.2%) was obtained (Table 1). PEG-PLGA IR (KBr): 3294,
2
synthetic materials were obtained from commercially available
companies, and all solvents were distilled to remove any traces of
water before use. Fetal bovine serum (FBS) and Dulbecco's
modied Eagle medium (DMEM) were purchased from Invi-
trogen Gibco (USA). 3-[4,5-Dimethyl-thiazol-2-yl]-2,5-diphenyl-
ꢀ
1
3
065, 2941, 1734, 1650, 1550, 1450, 1120, 1310, 740, 698 cm ;
1
H NMR (CDCl
–
3
): dppm ¼ 5.05 (s, –CH
2
in benzyl group), 4.00 (s,
COCHNH– of BLG), 3.65 (t, CH of PEG), 3.40 (s, CH – of PEG),
2
3
0
2.60–2.00 (m, –CH CH – of side chains of PBLG).
2 2
tetrazolium bromide (MTT), 4 ,6-diami-dino-2-phenylindole
DAPI), PAMAM G5, PEI-25k, Lipofectamine 2000 (Lipo-2k),
2
.2.5 Aminolysis of PEG-PBLG. Aminolysis of PEG-PBLG
(
13
was performed according to the literature. Typically, PEG-
PBLG (0.2 g) was dissolved in 5 mL of dimethyl formamide, then
genistein, amiloride and chlorpromazine were purchased from
Sigma-Aldrich Inc. (St. Louis, MO USA). Label ITꢀTM-rhodamine
kit and Label ITꢀTrackerꢁ Fluorescein kit were purchased from
Mirus (USA). Lab-Tekꢀ Chambered coverglass system was
purchased from Nunc (USA). Dialysis tubing (MWCO 3500) was
purchased from Green bird Inc. (Shanghai China).
2-hydroxypyridine (0.4 g, 5 equiv. of benzyl esters in PBLG block)
and PAMAM G2 (13.5 g, 5 equiv. of benzyl esters in PBLG block)
ꢁ
were added, respectively. The mixture was stirred at 40 C for
2
d, and then precipitated with 100 mL of diethyl ether. The
crude product was dialyzed with dialysis tubing (MWCO 3500)
against distilled water for 48 h and lyophilized. Thus the PEG-
PLGA, EA-G2 (0.33 g), was obtained. At the same time, the EA-G1
2.2 Preparation of EA-G2 and EA-G1
2
.2.1 Synthesis of PAMAM G2 and G1. PAMAM G2 and G1 prepared by PAMAM G1 was obtained by a similar procedure
15,16
were prepared based on our previously published work.
number of terminal primary amines of PAMAM G2 and G1 1450, 1120 cm ; H NMR (D
measured by potential titration was 16.3 or 7.9, respectively. The PLGA), 3.77 (t, CH of PEG), 3.37–2.51 (t, CH
molecular weights of PAMAM G1 and G2 measured by MALDI- –CH CH – of side chains of PLGA). EA-G2 IR (KBr): 3250, 3065,
TOF mass spectra were 1452 and 3274 Da, respectively. FT-IR 2941, 2831, 1650, 1550, 1450, 1120 cm ; H NMR (D
spectra of the polymers were measured using an EQUINOX55 8.13 (s, –COCHNH– of PLGA), 3.78 (t, CH of PEG), 3.36–2.49 (t,
– of side chains of PLGA). DSC
NMR spectra of the polymers were determined using an thermograms were measured on a DSC-200F3 Maia instrument
The (Table 2). EA-G1 IR (KBr): 3250, 3065, 2941, 2831, 1650, 1550,
ꢀ
1 1
2
O): dppm ¼ 8.15 (s, –COCHNH– of
2
2
of PAMAM), 2.4 (t,
2
2
ꢀ1 1
O): dppm
¼
2
2
1
2 2 2
Fourier transform infrared spectrometer (Bruker, Germany). H CH of PAMAM), 2.32 (t, –CH CH
ꢁ
ꢀ1
AVANCE (400 MHz) NMR spectrometer (Bruker, Germany). G1.0 (NETZSCH, Germany) at a rate of 10 C min . The GPC was
IR (KBr): 3290, 3065, 2941, 2831, 1650, 1550, 1450, 1360, 1126, measured on a Agilent 1100HPLC chromatograph (USA), with
036 cm . H NMR (D
ꢀ
1 1
1
2
O): dppm ¼ 2.51 (t, 24H, 12CH
2
CH
2
CO), linear polyethylene glycol used as a standard polymer.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5115

View Article Online
Journal of Materials Chemistry B
Paper
Table 1 Preparation of PEG-PBLG
M
n
of
Feeding molar
a
b
Copolymer
mPEG-NH
2
ratio of NH
2
/NCA
Reaction condition
DP of PBLG block
Yield (%)
73.2
M
n
of PEG-PBLG
4
PEG-PBLG-16
2000
1 : 100
R.T. ꢂ 72h
90.1
2.17 ꢂ 10
a
1
b
Measured by H-NMR spectrum. Calculated from the DP of PBLG block in PEG-PBLG copolymer.
2
.3 Plasmid DNA
The plasmid-enhanced green uorescent protein gene-C1
pEGFP-C1) was obtained from West China University of
(UVITEC, USA). At the same time, PAMAM G1, PAMAM G2,
PAMAM G5 and PEI 25k complexes were used as controls.
(
2
.7 MTT assay
Medical Sciences (Chengdu, China) and contained the early
promoter of CMV. The pEGFP-C1 was amplied by competent
Escherichia coli strain DH5a and puried by a Qiagen kit
ꢀ1
EA-G2 (or EA-G1) solutions at concentrations of 0–200 mg mL
4
were freshly prepared. Hela cells at 2 ꢂ 10 cells per well were
seeded in 96 well plates loaded with 200 mL DMEM with 10%
FBS, and incubated for 20–24 h to achieve 70–80% conuence.
The medium was replaced by polymer solutions without serum
or with 10% FBS medium, and the cells were cultured for 6 h. All
polymer solutions were changed to fresh DMEM, and the cells
continued to be cultured for 48 h. 15 mL of MTT solution was
added and the cells were incubated for another 4 h. The
medium was replaced with 150 mL of DMSO to dissolve the
purple formazan crystal produced by live cells. Finally,
the absorbance in every well was measured at 570/630 nm by an
ELISA micro-plate reader (Bio-Rad, USA). Here, the cytotoxicity
was expressed as “cell viability”, which is a percentage of the
absorbances of complex loading wells to complex-free loading
wells. PAMAM G1, PAMAM G2 and PEI 25k were used as
controls.
(
Chatsworth, USA). The concentration of pEGFP-C1 was
measured at 260 nm on a Nanodrop 2000 UV spectrophotom-
eter (Thermo Fisher scientic, USA).
2
.4 Cell lines
Hek-293 cell line was purchased from American Type Culture
Collection (USA). Bel-7402 and Cos-7 cell lines were purchased
from Shanghai Cell Center (Shanghai, China). Hela cell line was
obtained from the Animal Center of Sun Yetsen University.
2
.5 Biodegradation of EA-G2 and EA-G1
The biodegradations of EA-G2 and EA-G1 in papain solution
were investigated by viscosity method. First, 20 mg of EA-G2 or
EA-G1 and 10 mg of papain were dissolved in 10 mL of PBS (pH
ꢁ
7
.4). Then, the enzyme-degradation was carried out at 37 C for
different times (0, 2, 4, 8, 24 and 48 h, respectively). At every
predetermined time (P), the owing time (t) for an Ubbelohde
2.8 Particle sizes and zeta potential measurement
viscometer was recorded. The intrinsic viscosity was calculated The EA-G2 (or EA-G1)/pEGFP-C1 complexes at w/w ¼ 0.5–70
in the following equation: [h] ¼ (t ꢀ t )/t /C, where t and t were prepared and incubated for 30 min at room temperature.
0
0
0
ꢁ
The particle size and the zeta potential were measured at 25 C
denote the owing times of the pure solvent and polymer
solution, and C denotes the polymer concentration. Finally, the
relationship of [h] /[h] vs. P was plotted, where [h] and [h]
1 0 0 1
with a Zetasizer Nano Series 90 (Malvern Instruments, UK).
PAMAM G1 and PAMAM G2 complexes were used as controls.
denote the intrinsic viscosities at the beginning or the degra-
dation time P, respectively.
2.9 Atomic force microscopy inspection
The EA-G2/pEGFP-C1 complex at w/w ¼ 50 with a nal pDNA
2.6 Agarose gel electrophoresis
ꢀ1
concentration of 20 mg mL
was prepared. A drip of the
The EA-G2 (or EA-G1)/pEGFP-C1 complex solutions at w/w ¼ complex solution was dripped on a uoride crystal mica slice of
–4.0 were freshly prepared before use. 10 mL of the complex 10 ꢂ 10 ꢂ 0.2 mm (Changchun Tianchen Inc, China) and
solution was loaded in 1.0% agarose gel prepared in TAE buffer naturally evaporated to dryness. The specimen was inspected on
0
ꢀ1
containing 0.5 mg mL ethidium bromides, and the electro- a SPM-300HV atomic force microscope (Seiko Instrument Inc
phoresis was run at 100 V for 60 min. Finally, the DNA bands Japan) in the tapping mode at a scan speed of 3 Hz with 512 ꢂ
were photographed using a GAS7001X gel image system 512 pixel resolution.
Table 2 Preparation of EA-G2 and EA-G1
Molar ratio
of BLG/PAMAM
a
b
Polycation
Starting copolymer
Aminolysis agent
Reaction condition
DP of PLGA block
M
n
of product
ꢁ
ꢁ
4
EA-G2
EA-G1
PEG-PBLG-16
PEG-PBLG-16
PAMAM G2
PAMAM G1
1 : 5
1 : 10
40 C ꢂ 48 h
33.9
54.4
11.6 ꢂ 10
8.1 ꢂ 10
4
50 C ꢂ 120 h
a
1
b
Measured by H-NMR spectra. Calculated from the DP of PLGA block in PEG-PLGA copolymer.
5
116 | J. Mater. Chem. B, 2013, 1, 5114–5127
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry B
Scheme 1 Synthetic routes to EA-G1 and EA-G2 (a) and an illustration of the formation of the EA-G2/pDNA complex by self-assembly (b).
2
.10 In vitro transfection efficiency evaluation
to transfection, the Hek293 cells were individually incubated for
h with one of the following inhibitors: chlorpromazine (10 mg
mL ), amiloride (1 mM) or genistein (0.2 mM), respectively.
The subsequent transfection was carried out based on the
procedure outlined in section 2.10. The transfection efficiencies
relative to control were calculated.
1
The EA-G2 (or EA-G1)/pEGFP-C1 complexes at w/w ¼ 10–70 were
ꢀ1
5
prepared. Hek-293 cells (at 10 cells per well) were seeded in
800 mL of DMEM medium with 10% FBS and incubated for 20–
24 h to achieve 70–80% conuence. The medium was replaced by
500 mL of fresh DMEM with or without 10% FBS. 200 mL of
complexes were added to ensure the nal DNA concentration of
ꢀ
1
2
mg well , and the cells were incubated for 6 h. The medium
2
.12 Cellular uptake assay
was replaced with fresh DMEM containing 10% FBS. The cells
continued to be incubated for another 48 h, and were inspected
and photographed by uorescence spectroscopy (Olympus BX51,
Japan). The cells were harvested by trypsin–EDTA treatment, and
The pEGFP-C1 was labeled by uorescein according to the
manufacturer's protocol and the EA-G2/uorescein labeled-
pDNA complexes at w/w ¼ 50 were freshly prepared before use.
5
Hek-293 cells (10 cells per well) were seeded in a 24-well plate

nally suspended in 200 mL of PBS (pH 7.4). To quantify the
and allowed to attach and grow for 20–24 h. The cells were
incubated with the complexes (2 mg of DNA per well) for 0, 0.33,
transfection efficiency, the percentages of cells expressing EGFP
were measured by ow cytometry (BD-FACS-Aria System, USA).
Furthermore, the transfections in Bel-7402 and Cos-7 cells were
carried out as described in the above procedures, and PAMAM
G1, PAMAM G2, PEI-25k and Lipo-2k were also used as controls.
0.67, 1.0, 3.0, 4.0 and 6.0 h, respectively, then the complex
solution was removed. The cells were washed once with cold
PBS, detached by trypsin and centrifuged to collect. The uo-
rescence of the non-internalized complexes was quenched by
20
0.04% trypan blue in PBS solution for 5 min. Aer being
2
.11 Endocytic pathway determination by inhibitors
washed with cold PBS 3 times, the cells were suspended in PBS.
The cell viability of inhibitors at working concentration was The samples were assayed using a FACS Calibur System (Becton-
assayed as described in the above procedures. Endocytic path- Dickinson, San Jose, CA). The cellular uptake was obtained from
19
ways were determined according to the literature. Briey, prior the percentage of cell-internalized uorescein complexes.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5117

View Article Online
Journal of Materials Chemistry B
Paper
1
Fig. 1 Characterization of EA-G2 and related copolymers: FT-IR spectra (a); H-NMR spectra (b) and DSC curves (c).
2.13 Confocal laser scan microscopy observation
cultured for 24 h to achieve 70–80% conuence. 200 mL of
complex solution was added, and the cells were incubated for 1,
The pEGFP-C1 was labeled by rhodamine based on the manu-
facturer's protocol, and the EA-G2/rhodamine labeled pEGFP-
C1 complex at w/w ¼ 50 was freshly prepared before use. The
Hek-293 cells were seeded in a Lab-Tekꢀ chambered coverglass
4
and 8 h, respectively. The cells were xed by 4% para-
ꢀ
1
formaldehyde, and the cell nuclei were labeled by 0.2 mg mL
of DAPI for 10 min. The CLSM observation was performed
21
4
according to the literature. The wells were inspected and
system at 1.2 ꢂ 10 cells per well (8 wells per chamber) and
5
118 | J. Mater. Chem. B, 2013, 1, 5114–5127
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry B
Table 3 Thermal transition of PEG-PBLG, EA-G1 and EA-G2 copolymers
PAMAM
PEG
PBLG
PAMAM
a
g
ꢁ
ꢀ1
b
m
ꢁ
ꢀ1
c
ꢁ
ꢀ1
d
ꢁ
ꢀ1
T
( C)
OH (J g
)
T
( C)
OH (J g
)
T
LC ( C)
OH (J g
)
T
i
( C)
OH (J g )
PEG-PBLG
EA-G1
EA-G2
—
—
0.61
0.45
26.9
32.4
42.6
0.58
0.20
0.27
119.0
—
—
7.3
—
—
—
124.4
137.4
—
179.0
121.8
ꢀ29.2
ꢀ16.3
a
b
c
T
g
is the glass transition temperature of PAMAM blocks.
T
m
is the melt temperature of PEG blocks.
T
LC is low enthalpy change of PBLG blocks
d
from a 7/2 to an 18/5 R-helical conformation.
i
T is the isotropic liquid transition of PAMAM blocks.
photographed on a LSM-510 confocal microscope (Germany).
For rhodamine, the excitation was at 568 nm and the emission
was at 575–640 nm, but for DAPI, the excitation was at 350 nm
and the emission was at 430 nm.
3
Results and discussion
3
.1 Synthesis and characterization of EA-G2 (or EA-G1)
The polymerization of BLG-NCA was carried out by a ring-
opening mechanism using MPEG-NH -2000 as initiator, and
2
the polymerization degree (DP) of the PBLG block was mainly
2.14 Statistical analysis
dependent on BLG-the NCA/NH molar ratio. Furthermore, the
2
The data were summarized as means ꢃ standard deviation. benzyl esters of PBLG block were ammonolyzed by PAMAM G2
Where necessary, the statistical variations (p values) between (or G1) using 2-hydroxypyridine as a catalyst to form the poly-
the group means were evaluated by Student's t-test.
(L-glutamine) (PLGA) block (Scheme 1a).
Fig. 2 Performances of EA-G2 and EA-G1: cell viabilities of polycations with different concentrations in Hela cells in serum-free medium (n ¼ 3) (a), and in 10% FBS
medium (n ¼ 3) (b). Enzyme-degradation curves of EA-G2 and EA-G1 by papain (c). Gel electrophoresis assay of polycation/pDNA complexes at various w/w ratios (d):
EA-G1 (1), EA-G2 (2), PAMAM G1 (3), PAMAM G2 (4), PEI 25k (5), PAMAM G5 (6).
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5119

View Article Online
Journal of Materials Chemistry B
Paper
Table 4 The content of different amines in polycations
to the C–H symmetrical and asymmetrical stretching vibrations,
ꢀ1
respectively; the peaks at 1650 and 1550 cm were attributed to
the C]O stretching vibration of amide (amide I) or N–H
deformation vibration of amide (amide II) of PBLG block; the
a
Mol number of amines per mol polycation
Secondary
ꢀ1
Polycation
Primary amines
amines
Tertiary amines peaks at 740 and 698 cm belonged to the plane bending
23
vibration of the benzene ring in PBLG. In the PAMAM G2
EA-G2
EA-G1
PAMAM G2
PAMAM G1
505.5
380.8
16
0
0
0
0
474.6
326.4
14
ꢀ1
curve, the peaks at 3290 and 3065 cm corresponded to the
N–H symmetrical and asymmetrical stretching vibrations of
8
6
primary amines, respectively; the attributions of the peaks at
ꢀ1
a
2941, 2831, 1660 and 1550 cm were analogous to those of
PEG-PBLG. In the EA-G2 curve, the peaks at 1734, 740 and 698
Calculated from the formula of the polycations.
ꢀ
1
cm all disappeared, but the peaks at 3290, 3065, 2941, 2831,
ꢀ
1
ꢀ1
In the PEG-PBLG curve of Fig. 1a, the peak at 1120 cm was 1650, 1550, 1120 cm still existed, suggesting the removal of all
attributed to the C–O stretching vibration of PEG; the peak at the benzyl esters and the formation of the new PEG-PLGA
ꢀ
1
1
734 cm was attributed to the C]O stretching vibration of copolymer. The IR spectra of PAMAM G1 and EA-G1 (not shown)
22
ꢀ1
benzyl esters; the peaks at 2941 and 2831 cm were attributed were similar to those of PAMAM G2 or EA-G2.
Fig. 3 Particle sizes and zeta potentials of polycation/DNA complexes at different w/w ratios in deionized water measured by DLS: particle sizes (n ¼ 3) (a); zeta
potentials (n ¼ 3) (b); size distribution of EA-G1/pDNA complexes at w/w ¼ 50 (c); size distribution of EA-G2/pDNA complexes at w/w ¼ 50 (d). AFM images of EA-G2/
pDNA complexes at w/w ¼ 50: the 3D view of complexes in the range of 2.0 mm (e); the 2D view of complexes in the range of 2.0 mm (f); the height profiles 1 and
2
corresponding to the sections crossed the A-B or C particles in 2D view, respectively (g).
5
120 | J. Mater. Chem. B, 2013, 1, 5114–5127
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry B
In the PEG-PBLG curve of Fig. 1b, the c and d peaks at 3.65
and 3.39 ppm belonged to the CH chemical shi of the PEG
2
block; the a, b and f peaks at 5.05, 3.94 and 2.26 ppm belonged
to the chemical shis of CH of benzyl ester, CH of peptide
2
2 2
chain and CH –CH of side chain, respectively. In the PAMAM
G2 curve, the l, m, i and h peaks at 3.43, 2.86–2.98, 2.75 and
2
1
CH
.55 ppm belonged to the chemical shis of CH
2
CH
CH
CO on PAMAM G2 frame. In the EA-G2 curve, the peak
2
NH
2
+
2CH
2
CH
2
N, CH
2
CH
2
–NH
2
+ CH
2
CH
2
N, CH
2
N + CH
2
2
CO or
2
CH
2
at 5.05 ppm fully disappeared, but the c, l, m, i, h and f peaks
still existed, representing the complete aminolysis of the PEG-
1
PBLG and the formation of the new PEG-PLGA. The H-NMR
spectra of PAMAM G1 and EA-G1 (not shown) were also similar
influence of endocytosis inhibitors on gene expression: the percentage of trans- to those of PAMAM G2 or EA-G2.
Fig. 4 Cell viability in the presence of the following inhibitors: chlorpromazine
ꢀ1
(
10 mg mL ), amiloride (1 mM) or genistein (0.2 mM), respectively (n ¼ 3) (a);
fection efficiency of EA-G2 and EA-G1 complexes at w/w ¼ 50 to control after
In Fig. 1c and Table 3, the DSC curve of PEG-PBLG covered
two endothermic transitions: the PEG crystalline melting at T
individual incubation with various endocytosis inhibitors (n ¼ 3) (b).
m
ꢁ
24
(26.9 C) and the PBLG (actually the main chains of PLGA)
irreversible low enthalpy change from a 7/2 to an 18/5 R-helical
ꢁ
25
conformation at TLC (around 119.0 C). The DSC curve of EA-
G2 covered three endothermic transitions: the PEG crystalline
ꢁ
melting at T
(
m
(42.6 C), the PAMAM G2 glass transition at T
g
ꢁ
ꢀ16.3 C) and the PAMAM G2 isotropic liquid transition at T
i
ꢁ
26,27
(around 137.4 C).
Because TLC was quite close to T
i
and the
T
LC peak was too weak, the TLC peak might appear on the le
shoulder of the T
i
peak. The EA-G1 curve is similar to that of EA-
ꢁ
G2. The T , T and T values were 32.4, ꢀ29.2 and 124.4 C,
m
g
i
respectively. The DSC analysis proved that the EA-G2 (or EA-G1)
copolymer consists of PAMAM, PEG and PBLG blocks, and
PAMAM G2 (or G1) blocks occupy a signicant proportion, but
PEG and PBLG blocks occupy just a very small proportion in EA-
G2 (or EA-G1). The GPC of EA-G2 (or EA-G1) exhibited a unim-
odal curve (not shown) and no other peak was found, which
proved that EA-G2 (or EA-G1) was a copolymer but not a mixture.
The DP of the PBLG block was calculated through the area
ratio of a peak to c peak in the PEG-PBLG NMR curve, the DP of
the PLGA block in EA-G2 was calculated by the area ratio of f
peak to c peak in the EA-G2 NMR curve. The DP of PLGA in EA-
G1 was calculated in the same way. The data were listed in
Tables 1 and 2. It is noteworthy that the aminolysis of PEG-
PBLG by PAMAM G1 requires a higher temperature and a longer
time than that by PAMAM G2 because the aminolysis rate of
28
PAMAM largely increased with their generation from G0 to G4.
Aer aminolysis, the DP decreased from 90.1 (PBLG) to 33.9
EA-G2) or 54.4 (EA-G1) suggesting the occurrence of consider-
(
able chemical degradation of the PBLG block. Thus, the calcu-
lated average molecular weights of EA-G2 and EA-G1 were
4
4
1
1.6 ꢂ 10 and 8.1 ꢂ 10 Da, respectively.
3.2 Cytotoxicity
With an increase in polymer concentration, the cell viability
decreased, that is cytotoxicity increased. As shown in Fig. 2a,
PAMAM G1 showed the least cytotoxicity. EA-G1 and PAMAM G2
displayed greater cytotoxicity than PAMAM G1 ( p < 0.05, n ¼ 3),
but less cytotoxicity than EA-G2 at all concentrations ( p < 0.01,
Fig. 5 Flow cytometric histograms showing cellular uptake of fluorecein-labeled
pDNA mediated by EA-G2 complexes at different incubation times in 10% FBS
medium (a); in serum-free medium (b); the kinetic curves of cellular uptake rep-
resenting the percentage of fluorescence positive cells as a function of incubation n ¼ 3). PEI-25k exhibited the most severe cytotoxicity, and the
ꢀ
1
time (n ¼ 3) (c).
cell viability was only 4.9% at 200 mg mL . EA-G2, EA-G1,
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5121

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 6 CLMS images in Hek 293 cells at 1 h of transfection for EA-G2/pEGFP-Cl complexes at w/w ¼ 50 (a); at 4 h of transfection for EA-G2/pEGFP-Cl complexes at
w/w ¼ 50 (b); at 8 h of transfection for EA-G2/pEGFP-Cl complexes at w/w ¼ 50 (c); at 8 h of transfection for EA-G1/pEGFP-Cl complexes at w/w ¼ 50 (d); at 8 h of
transfection for PEI-25k/pEGFP-Cl complexes at N/P ¼ 10 (e). The rhodamine-labeled pEGFP-C1 exhibited the red fluorescence, and the DAPI-labeled nuclei exhibited
the blue fluorescence.
PAMAM G2 and PAMAM G1 all exhibited much less cytotoxicity blocks exhibited greater enzyme-degradation rates; however, the
than PEI-25k at all concentrations (p < 0.01, n ¼ 3). Further, in amide bonds on the PAMAM exhibited a slower degradation rate.
the presence of 10% serum medium, all polymers exhibited In the rst 12 h, the degradation mainly occurred on the peptide
smaller cytotoxicity than that that in serum-free medium at all bonds of PLGA, and the subsequent degradation mainly occurred
concentrations (Fig. 2b), suggesting that the serum could help on the amide bonds of PAMAM. The enzyme-degradation of EA-
to mitigate the inuence of polycation on cell membrane.
G1 was faster than that of EA-G2 because the former contained a
greater proportion of PLGA chains than the latter. This is
demonstrated by the fact that the EA-G2 and EA-G1 were both
biodegradable polycations. The good biodegradability facilitates
the safe application in vivo of EA-G2 or EA-G1 as a gene delivery
vector. The degradation products with small molecular weight
are easily eliminated through excretion pathways; hence, the
cytotoxicity will further reduce to a larger extent.
3.3 Biodegradability
The relationship between the intrinsic viscosities [h] and the
molecular weight of polymer M can be expressed as the following
a
equation: [h] ¼ KM , where K and a are constants. The smaller
1 0
the [h] /[h] , the larger the biodegradability. Papain is an
analogue of cathepsin that widely exists in the extracellular
matrix and cell lysosomes, so here, the papain is chosen as an
enzyme for studying biodegradability. As shown in Fig. 2c, a fast
3.4 Combining capacity with pDNA
enzyme-degradation appeared in the rst 12 h, then a relatively The combination of EA-G2 (or EA-G1) with pDNA through
slow degradation occurred. The peptide bonds of the PLGA electrostatic interaction to form complexes was shown in
5
122 | J. Mater. Chem. B, 2013, 1, 5114–5127
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry B
*1
*3
*4
Fig. 7 Transfection efficiencies at different weight ratios of EA-G2/pDNA complexes in serum-free medium or 10% FBS medium (a) ( p < 0.01, p < 0.01, p < 0.01
*
5
*2
and p < 0.05, n ¼ 3, the differences were considered statistically significant; p > 0.05, n ¼ 3, the differences were not considered significant); EA-G1/pDNA
*6
*7
complexes in serum-free medium or 10% FBS medium (b) ( p < 0.01, n ¼ 3, the differences were considered statistically significant; p > 0.05, n ¼ 3, the differences
were not considered significant); the fluorescence images of transfected Hek-293 cells: EA-G2 at w/w ¼ 50 in serum-free medium (c); EA-G2 at w/w ¼ 50 in 10% FBS
medium (d); EA-G1 at w/w ¼ 50 in serum-free medium (e); EA-G1 at w/w ¼ 50 in 10% FBS medium (f); Lipo-2k at w/w ¼ 2 in serum-free medium (g); Lipo-2k at w/w ¼
2
in 10% FBS medium (h); PEI-25k at N/P ¼ 10 in serum-free medium (i); PEI-25k at N/P ¼ 10 in 10% FBS medium (j).
Scheme 1b. The self-assembly of the complexes can be assayed 3.5 Particles and zeta potentials
by gel retardation electrophoresis. As shown in Fig. 2d, the
As shown in Fig. 3a, at w/w ¼ 0.5, the positive charge density of
separated protein bands could be seen in the lanes of naked
DNA or complexes with very small weight ratios. With raising
the weight ratio, the net positive charges in the complexes
increase. When the weight ratio increases to a threshold value,
the polycation can completely combine with DNA, thus the
separated DNA bands are no longer seen. EA-G1, EA-G2, PEI-25k
and PAMAM G5 completely combined with DNA at w/w $ 1.0,
complexes was too small to condense the DNA effectively, so a
loose structure with a relatively large volume was formed. As the
weight ratio was raised to 20 or more, the particle size of
complexes decreased gradually. At w/w ¼ 30–70, the particle
sizes of EA-G2 (or EA-G1) complexes (150–250 nm) were smaller
than those of PAMAM G2 and G1 (200–650 nm), demonstrating
that the former possessed stronger pDNA condensing capacity
than the latter. Fig. 3b showed that at w/w ¼ 0.5, the zeta
potentials of all complexes were negative, implying the
complexes were negatively charged. However, at w/w ¼ 10–70,
the zeta potentials of EA-G2 (or EA-G1) complexes were in the
range of 20–50 mV, indicating the net positively charged
complexes generated. For PAMAM G1 complexes, at w/w ¼ 0.5–
0
.7, 0.5 or 0.5, respectively (Fig. 2d 1–2, 5–6), but PAMAM G1 and
PAMAM G2 completely combined with pDNA at w/w $ 4.0 or
.0, (Fig. 2d 3–4). This is illustrated by the fact that EA-G2, EA-
3
G1, PEI-25k and PAMAM G5 all had a greater DNA combining
ability than PAMAM G2 and G1. The reason is that EA-G2, EA-
G1, PEI-25k and PAMAM G5 contain a greater number of
primary amine groups and higher positive charge density than
PAMAM G2 and G1 (Table 4). The naked pDNA has difficulty
entering cells due to its larger size, enzymatic degradation and
the polyanionic characteristic of DNA etc. The complexes can
not only condense the pDNA to enter into cells, but also prolong
the life cycle of pDNA. In the self-assembly of complexes, no
precipitation was found. The PEG blocks of EA-G2 will extend
outside the complexes to enhance the water solubility and
reduce the cytotoxicity.
20, the zeta potentials were still negative, but at w/w $ 30 the
zeta potentials became positive, demonstrating that the
PAMAM G1 could not completely combine with the DNA. As
shown in Fig. 3c and d the particle size of EA-G2 (or EA-G1)
complexes at w/w ¼ 50 was unimodal distribution. To obtain a
high transfection efficiency the formation of compact
complexes with nano-scale sizes and positive surface charges
(
i.e. positive zeta potentials) is a crucial factor. At w/w ¼ 30–70
the sizes of EA-G2 (or EA-G1) complexes were 150–250 nm and
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5123

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 8 Transfection efficiencies of EA-G2/pDNA complexes at w/w ¼ 50 in different cells (n ¼ 3) (a). The scatter diagrams of transfection for EA-G2 complexes at w/w ¼ 50
measured by flow cytometry: in blank Hek-293 cells (b); in blank Cos-7 cells (c); in blank Bel-7402 cells (d); in Hek-293 cells in serum-free medium (e); in Cos-7 cells in serum-free
medium (f); in Bel-7402 cells in serum-free medium (g); in Hek-293 cells in 10% FBS medium (h); in Cos-7 cells in 10% FBS medium (i); in Bel-7402 cells in 10% FBS medium (j).
Table 5 The rates of transfection efficiency in 10% FBS medium to that in serum-
3.6 Morphology inspected by AFM
free medium under the same conditions (i.e. for the same polymer/pDNA
Further, we also inspected the size and morphology of complexes
using AFM. Fig. 3e vividly showed the compact morphology of the
complexes. Fig. 3f showed that the particle sizes of the complexes
at w/w ¼ 50 were in the range of 120–150 nm, which were slightly
smaller than those measured by DLS. This was possibly because
the sample of AFM was in a dry state, but the sample of DLS was
in a wet state. In Fig. 3g the heights of the A and B peaks of
prole 1 and the C peak of prole 2 were 12, 10.5 or 11 nm,
respectively, which corresponded to Section 1 or 2 in Fig. 3f.
complex, at the same w/w ratio and in the same cells)
a
b
c
d
Cell type
EA-G2
EA-G1
PEI-25k
Lipo-2k
Hek-293
Cos-7
Bel-7402
1.09
1.04
0.94
0.48
—
—
0.33
0.14
0.08
0.28
—
—
14
a
b
c
EA-G2/pDNA at w/w ¼ 50. EA-G1/pDNA at w/w ¼ 50. PEI-25k/pDNA
d
at N/P ¼ 10. Lipo-2k/pDNA, at w/w ¼ 8.
3.7 Endocytosis pathway of complexes
the zeta potentials were 20–50 mV, which was favorable for
internalization through negatively charged glycoprotein and Generally, the complexes are taken up by mammalian cells through
2
9
phospholipids on cell membranes.
endocytosis, which can be classied as phagocytosis or pinocytosis,
This journal is ª The Royal Society of Chemistry 2013
5
124 | J. Mater. Chem. B, 2013, 1, 5114–5127

View Article Online
Paper
Journal of Materials Chemistry B
12
and the latter is further subdivided into clathrin-dependent exhibits blue uorescence. The bright eld images showed the
endocytosis (CME), caveolin-mediated endocytosis (CvME) and morphology of Hek-293 cells. At 1 h of transfection, only a small
macro-pinocytosis. The endocytic pathways of complexes are number of EA-G2 complexes entered into the cells (Fig. 6a). At
dependent not only on their structure and nature (size, charge and 4 h of transfection, many EA-G2 complexes were internalized
30
stability), but also on cell type and cell metabolism. Inhibitors are into the cytoplasm, and some of them had entered into the cell
the commonly used tools to block a specic pathway in order to nuclei (Fig. 6b). At 8 h of transfection, a great number of EA-G2
evaluate whether it plays an important role in the uptake of complexes had entered into the cell nuclei Fig. 6c. As shown in
31
complexes. Here, to determine the endocytic pathways, chlor- Fig. 6b and c, the new purple dots generated were due to an
promazine, amiloride or genistein is chosen as the inhibitor for overlap effect of the red uorescence and the blue uorescence
CME, macro-pinocytosis and CvME, respectively. Fig. 4a showed in the nucleus region (white arrows). In Fig. 6d and e, at 8 h of
when the three inhibitors at the working concentration were used transfection, only a small amount of EA-G1 or PEI-25k
in cells respectively, the cell viabilities all achieved above 90%, complexes entered into the cell nuclei, which was consistent
revealing that the inuence of cytotoxicity on transfection basically with the data of their transfection efficiency (Section 3.10). The
was negligible. As shown in Fig. 4b, when chlorpromazine, ami- clear nucleus localization of pEGFP-Cl suggested that EA-G2
loride or genistein was individually used, the transfection effi- could efficiently mediate the pDNA into the cell nuclei.
ciencies of EA-G2 complexes were 43.6, 90 and 31.8% of the control,
respectively, suggesting that cell uptake occurred mainly through
the CME and CvME pathways. Similarly, in the individual use of
32
3.10 In vitro transfection efficiency
chlorpromazine, amiloride or genistein, the transfection efficien- As shown in Fig. 7a, in 10% FBS or serum-free medium, upon
cies of EA-G1 complexes were 191, 97 and 12.3% of the control, raising the weight ratios from 10 to 70, the transfection effi-
respectively, suggesting that only CvME was the main uptake ciency of the EA-G2 complexes rst increased, and then
pathway. The transfection efficiency of 191% in the presence of decreased. The reason is that raising the weight ratio will
chlorpromazine is due to the increase in the uptake of complexes enhance the positive charge density of complexes which is
non-specically bound to the cell membrane and the internaliza- favorable for condensing the pDNA, but too high a weight ratio
33,34
tion via other uptake pathways.
It was reported that CME and will result in a greater number of cell deaths. In serum-free
CvME were the reliable endocytic pathways leading to efficient medium or 10% FBS medium, the maximum transfection effi-
3
5
transgene expression for nonviral gene vectors.
.8 Kinetics of intracellular uptake
ciency (39.8 or 45.5%, respectively) appeared at w/w ¼ 50. In
serum-free medium, the EA-G2 complexes at w/w ¼ 50 displayed
higher transfection efficiency than PAMAM G2 at w/w ¼ 10
3
*
(
1
p < 0.01, n ¼ 3), and slightly lower transfection efficiency than
The cellular uptake is a critical step of transfection, which is
greatly inuenced by the polycation structure, the cell type and
the culture medium etc. In our study, the cell uptake rate of
complexes is expressed as the percentage of uorescein positive
cells. As shown in Fig. 5a and b, with increasing incubation
time, the uptake curves of EA-G2 complexes in 10% FBS or
serum-free medium moved to the right, implying the number of
*2
Lipo-2k at w/w ¼ 2 ( p > 0.05, n ¼ 3). However, in 10% FBS
medium the EA-G2 complexes at w/w ¼ 30–70 displayed much
higher transfection efficiencies than PAMAM G2 at w/w ¼ 10
*3
*4
and Lipo-2k at w/w ¼ 2 ( p < 0.01 or p < 0.01, n ¼ 3). At the
same weight ratios, the EA-G2 complexes in 10% FBS medium
always displayed higher transfection efficiencies than in serum-
*5
free medium ( p < 0.05, n ¼ 3). As shown in Fig. 7b, in serum-
free medium or 10% FBS medium at all weight ratios, the
transfection efficiencies of EA-G1 were always larger than those

uorescein positive cells increased gradually. Fig. 5c showed in
the rst 1 hour the cellular uptake was much faster; 1 h later the
cellular uptake began to slow down, but in 4–6 h the cellular
uptake gradually reached saturation, that is, the percentage of
*
6
of PAMAM G1 ( p < 0.01, n ¼ 3), and the transfection efficiency
of EA-G1 at w/w ¼ 50 was close to that of PEI-25k at N/P ¼ 10

uorescence positive cells no longer increased. In 10% FBS and
*
7
(
p > 0.05, n ¼ 3). Comparing Fig. 7a with Fig. 7b, the trans-
serum-free medium, the saturation values of the cellular uptake
were 89.0 or 72.7%, respectively. The kinetics of cellular uptake
of EA-G2 complexes can be tted to the rst order exponential
fection efficiencies of EA-G2 at w/w ¼ 50 in both serum-ree and
10% FBS medium were higher than that of EA-G1 ( p < 0.05,
n ¼ 3) due to the larger mol number of primary or tertiary
amines (Table 4). Fig. 7c–j show the uorescence images of
transfected Hek-293 cells. For EA-G2, the green uorescent
spots in 10% FBS medium (Fig. 7d) were greater than that in
serum-free medium (Fig. 7c), but for EA-G1, Lipo-2k and PEI-
ꢀ
0.913t
decay equations: Y ¼ 89.6–89.8e
(in 10% FBS medium,
(in serum-free medium,
2
ꢀ0.907t
R ¼ 0.9952) and Y ¼ 72.7–67.6e
2
R ¼ 0.9837), respectively. In the above equations, Y denotes the
percentage of uorescein positive cells, and t denotes the
incubation time. In the total process, the cell uptake (%) of EA-
G2 in 10% FBS medium was always larger than that in serum-
free medium, which perhaps was one of reasons related to the
higher transfection activity in 10% FBS medium.
25k, the green uorescent spots in 10% FBS medium (Fig. 7f, h
and j) all were much fewer than in serum-free medium (Fig. 7e,
g and i), which was consistent with their transfection efficiency.
Fig. 8a showed that in serum-free or 10% FBS medium the
transfection efficiencies of EA-G2 complexes at w/w ¼ 50
arranged in the following order: in Bel-7402 cells > in Hek-293
3.9 Intracellular trafficking of complexes
On CLSM, the rhodamine-labeled pEGFP-Cl (or complexes) cells > in Cos-7 cells, demonstrating the transfection efficiency
exhibits red uorescence, and the DAPI-labeled nucleus was depended on cell type. In the three different cells, the
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5125

View Article Online
Journal of Materials Chemistry B
Paper
transfection efficiencies in 10% FBS medium were close to those
in serum-free medium. Fig. 8b–j show the scatter diagrams of
transfection in the three cells measured by ow cytometry. In
the diagrams, the dots of the Q4 region (the lower right)
represent the number of cells expressing green uorescence
protein (GFP), and the dots of Q3 region (the lower le) repre-
sent the number of cells not expressing GFP. Fig. 8b–d showed
that in the three blank samples only very few dots in the Q4
region were found, suggesting that almost no transfection
occurred. In Fig. 8e–j, the percentages of the dots in the Q4
region to dots in (Q3 + Q4) regions should be equal to the values
of the transfection efficiencies in Fig. 8a.
2 K. Ma, M. X. Hu, Y. Qi, J. H. Zou, L. Y. Qiu, Y. Jin, X. Y. Ying
and H. Y. Sun, Biomaterials, 2009, 30, 6109–6118.
3 H. M. Wu, S. R. Pan, M. W. Chen, Y. Wu, C. Wang, Y. T. Wen,
X. Zeng and C. B. Wu, Biomaterials, 2011, 32, 1619–
1634.
4 Y. T. Wen, Z. H. Guo, Z. Du, R. Fang, H. M. Wu, X. Zeng,
C. Wang, M. Feng and S. R. Pan, Biomaterials, 2012, 33,
8111–8121.
5 Y. Wang, W. Kong, Y. Song, Y. Duan, L. Wang, G. Steinhoff,
D. Kong and Y. Yu, Biomacromolecules, 2009, 10, 617–622.
6 G. Navarro and C. Tros de ILarduya, Nanomed.: Nanotechnol.,
Biol. Med., 2009, 5, 287–297.
To quantify the serum tolerance, i.e. serum-resistant capacity
for transfection, we measured the rates of transfection effi-
ciency of polymer/pDNA complexes at a xed weight ratio in
7 X. Q. Zhang, X. L. Wang, S. W. Huang, R. X. Zhuo, Z. L. Liu,
H. Q. Mao and K. W. Leong, Biomacromolecules, 2005, 6, 341–
350.
1
0% FBS medium to that in serum-free medium (abbreviated
8 Y. Shoji and H. Nakashima, Curr. Pharm. Des., 2004, 10, 785–
796.
9 N. Shah, R. J. Steptoe and H. S. Parekh, J. Pept. Sci., 2011, 17,
470–478.
as: the rates of transfection efficiency with serum to without
serum). As shown in Table 5, for EA-G2 at w/w ¼ 50 in Hek-293,
Cos-7 and Bel-7402 cells, the rates were 1.09, 1.04 or 0.94. For
PEI-25k at N/P ¼ 10 in the three cells, the rates were only 0.33, 10 W. Tansey, S. Ke, X. Y. Cao, M. J. Pasuelo, S. Wallace and
.14 or 0.08. For EA-G1 at w/w ¼ 50 in Hek-293 cells and Lipo-2k C. Li, J. Controlled Release, 2004, 94, 39–51.
complexes at w/w ¼ 8 in Hek-293 cells, the rates were 0.48 or 11 J. Khandare, A. Mohr, M. Calder ´o n, P. Welker, K. Licha and
.28, respectively. The above results indicated that EA-G2 R. Haag, Biomaterials, 2010, 31, 4268–4277.
complexes displayed greater serum tolerance than PEI-25k, EA- 12 L. Chen, H. Tian, J. Chen, X. Chen, Y. Huang and X. Jing, J.
0
0
G1 and Lipo-2k complexes.
Gene Med., 2010, 12, 64–76.
13 S. K. Tripathi, R. Goyal, K. M. Ansari, K. Ravi Ram, Y. Shukla,
D. K. Chowdhuri and K. C. Gupta, Eur. J. Pharm. Biopharm.,
4
Conclusion
2
011, 79, 473–484.
A novel PEG-PLGA copolymer (EA-G2) was prepared by an ami- 14 Y. T. Wen, S. R. Pan, X. Luo, X. Zhang, W. Zhang and
nolysis of PEG-PBLG copolymer using PAMAM G2. The analysis M. Feng, Bioconjugate Chem., 2009, 20, 322–332.
of FT-IR, H-NMR, DSC and GPC conrmed the formation of EA- 15 X. Zeng, S. R. Pan, J. Li, C. Wang, Y. T. Wen, H. M. Wu,
1
G2. The EA-G2 exhibited considerable biodegradability, low
cytotoxicity and great ability to combine with pDNA. The EA-G2/
C. F. Wang, C. B. Wu and M. Feng, Nanotechnology, 2011,
22, 375102.
pDNA complexes showed particle sizes in the range of 120–180 16 S. R. Pan, C. Wang, X. Zeng, Y. T. Wen, H. M. Wu and
nm and the zeta potentials in the range of 20–40 mV. Cellular M. Feng, Int. J. Pharm., 2011, 420, 206–215.
uptake of the EA-G2 complexes occurred mainly through clathrin- 17 A. Harada and K. Kataoka, Macromolecules, 2003, 36, 4995–
dependent and caveolin-mediated endocytosis. In 10% FBS and 5001.
serum-free media, the percentages of the complex uptake at 6 h 18 S. R. Pan, Q. M. Wang and W. Yi, J. Biomater. Appl., 2007, 22,
reached 89.0 or 72.7%, respectively. EA-G2 could efficiently 181–192.
mediate the plasmid EGFP-Cl into the cell nucleus. EA-G2 19 T. Kurosaki, T. Morishita, Y. Kodama, K. Sato, H. Nakagawa,
complexes displayed enhanced transfection efficiency and much
better serum tolerance. EA-G2 has potential to be used as a
biodegradable, efficient and serum-resistant gene vector.
N. Higuchi, T. Nakamura, T. Hamamoto, H. Sasaki and
T. Kitahara, Mol. Pharmaceutics, 2011, 8, 913–919.
20 D. Y. Olton, J. M. Close, C. S. Sfeir and P. N. Kumta,
Biomaterials, 2011, 32, 7662–7670.
2
1 Y. Shen, H. Peng, S. R. Pan, M. Feng, Y. T. Wen, J. J. Deng,
X. Luo and C. B. Wu, Nanotechnology, 2010, 21, 045102.
Acknowledgements
The authors are thankful to the National Natural Science 22 H. Huang, C. M. Dong and Y. Wei, Comb. Chem. High
Foundation of China (30870618 and 31170918) for nancial Throughput Screening, 2007, 10, 368–376.
support on the projects, and to the Key Laboratory on Assisted 23 R. Qi, Y. Gao, Y. Tang, R. R. He, T. L. Liu, Y. He, S. Sun,
Circulation, Ministry of Health for the cooperation in chemical
preparation, cell culture, gene transfection test and so on.
B. Y. Li, Y. B. Li and G. Liu, AAPS J., 2009, 11, 395–405.
24 N. S. Murthy, W. Wang and J. Kohn, Polymer, 2010, 51, 3978–
3988.
2
5 G. Z. Rong, M. X. Deng, C. Deng, Z. H. Tang, L. H. Piao,
X. S. Chen and X. B. Jing, Biomacromolecules, 2003, 4,
1800–1804.
References
1
T. Kurosaki, Y. Yamashita, K. Aki, H. Harasawa, H. Nakagawa,
Y. Kodama, N. Higuchi, T. Nakamura, T. Kitahara and 26 J. Mijovic, S. Ristic and J. Kenny, Macromolecules, 2007, 40,
H. Sasaki, J. Pharm. Sci., 2011, 100, 4855–4863. 5212–5221.
5
126 | J. Mater. Chem. B, 2013, 1, 5114–5127
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry B
2
2
2
3
3
7 J. M. Rueff, J. Barbera, B. Donnio, D. Guillon, M. Marcos and
J. L. Serrano, Macromolecules, 2003, 36, 8368–8375.
S u¨ ss, G. A. Koning and D. J. Crommelin, Pharm. Res., 2007,
24, 1590–1598.
8 J. L. Burnett, S. H. King Amy, I. K. Martin and L. J. Twyman, 32 Q. L. Hu, J. L. Wang, J. Shen, M. Liu, X. Jin, G. P. Tang and
Tetrahedron Lett., 2002, 43, 2431–2433. P. K. Chu, Biomaterials, 2012, 33, 1135–1145.
9 Y. N. Xue, M. Liu, L. Peng, S. W. Huang and R. X. Zhuo, 33 N. P. Gabrielson and D. W. Pack, J. Controlled Release, 2009,
Macromol. Biosci., 2006, 10, 404–414. 136, 54–61.
0 A. P. Perez, M. L. Cosaka, E. L. Romero and M. J. Morilla, Int. 34 M. J. Reilly, J. D. Larsen and M. O. Sullivan, Mol.
J. Nanomed., 2011, 6, 2715–2728. Pharmaceutics, 2012, 9, 1280–1290.
1 M. A. van der Aa, U. S. Huth, S. Y. H ¨a fele, R. Schubert, 35 C. Y. M. Hsu and H. Uludag, Biomaterials, 2012, 33, 7834–
R. S. Oosting, E. Mastrobattista, W. E. Hennink, R. Peschka- 7848.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. B, 2013, 1, 5114–5127 | 5127