polycation ternary polyplexes

Acta Biomaterialia 8 (2012) 3014–3026

Contents lists available at SciVerse ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Influence of the polyanion on the physico-chemical properties and biological activities of polyanion/DNA/polycation ternary polyplexes Cuifeng Wang a, Xin Luo a, Yuefang Zhao a, Lina Han a, Xin Zeng a, Min Feng a,⇑, Shirong Pan b, Hui Peng c,⇑, Chuanbin Wu a a b c

Department of Pharmacy, School of Pharmaceutical Sciences, Sun Yat-sen University, University Town, Guangzhou 510006, People’s Republic of China Cardiovascular Laboratory, The First affiliated Hospital, Sun Yat-sen University, 74 Zhongshan Road II, Guangzhou 510080, People’s Republic of China Zhongshan School of Medicine, Sun Yat-sen University, 80 Zhongshan Road II, Guangzhou 510080, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 16 December 2011 Received in revised form 26 March 2012 Accepted 20 April 2012 Available online 27 April 2012 Keywords: DNA ternary polyplex Biocompatible polyanions Transfection efficiency Loosening of polycation–DNA association Interactions between serum proteins and DNA polyplex

a b s t r a c t It was recently reported that polyanion/DNA/polycation ternary polyplexes markedly improve gene transfection activity in comparison with the original DNA/polycation binary polyplexes. In this study to explore the influence of the polyanion on the physico-chemical properties and biological activity of polyanion/pDNA/polycation ternary polyplexes four types of biocompatible polyanions were selected, mainly based on the acid strength of the anionic functional groups and the molecular rigidity on forming ternary polyplexes with 25 kDa polyethyleneimine and DNA. Polyanion loosening of the DNA polyplex, weakening of the adsorption of serum proteins and improving of cellular uptake, which are thought to be important factors leading to a high transfection efficiency of DNA ternary polyplexes, were specifically investigated. Electrophoresis retardation analysis indicated that the loosening capacity of polyanions depended on the pKa value of the functional anion groups as well as the flexibility of the polyanion. The low pKa and flexible structure of the polyanions tended to loosen the compact DNA polyplexes. Thermodynamic analysis by isothermal titration calorimetry provided direct evidence about the serum protein–DNA ternary polyplex interactions. The polyanion/pDNA/polycation ternary polyplexes exhibited obviously lower binding affinities and less adsorption to serum proteins compared with the original DNA/ polycation binary polyplexes. These relatively stable DNA ternary polyplexes maintained high levels of cellular uptake and intracellular accumulation in serum-containing medium that correlated with their high transfection efficiency. In contrast, the original pDNA/polycation binary polyplexes became clustered by strong adsorption of large amounts of serum proteins, leading to a sharp reduction in cellular uptake and intracellular accumulation, and thus low gene transfer efficiency. These results provide a basis for the development of polyanion/DNA/polycation ternary polyplexes for polyfection. Ó 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Polycationic polymers have shown significant promise as nonviral gene delivery vectors because of their safety and versatility in comparison with viral vectors. However, until now most gene therapy clinical trials have relied on viral vectors to deliver the desired gene. A relatively low efficiency is the main reason that polycationic vectors have not been widely used in clinical trials [1]. Today non-viral gene delivery vector studies focus on how to achieve a pronounced improvement in the transfection efficiency of cationic polymer vectors. The main strategies include the chemical modification of current polycation-based vectors and the synthesis of novel polycations [2]. ⇑ Corresponding authors. Tel.: +86 20 39943119 (M. Feng). E-mail addresses: [email protected] (M. Feng), [email protected] (H. Peng).

Recently a simple, novel approach to improve the transfection efficiency and reduce the cytotoxicity of polycations has been developed, adding biocompatible polyanions to DNA polyplex formulations [3–5]. Polyanion/DNA/polycation ternary polyplexes can form through electrostatic interactions resulting in adsorption of oppositely charged polyelectrolytes in aqueous media. Two types of biocompatible polyanions containing carboxyl groups have been widely investigated. One is polysaccharides such as alginic acid [6] and hyaluronic acid [7,8], the other polypeptides such as poly (glutamic acid) [3,9,10]. Yeo’s research showed that a DNA/crosslinked polyethyleneimine (PEI)/hyaluronate ternary polyplex achieved remarkably higher transfection efficiency than other polycation systems, especially under serum-containing conditions [7]. Urtt’s investigations revealed that coating a DNA/PEI polyplex with low molecular weight anionic hyaluronan (<10 kDa) facilitated CD44 receptor-mediated uptake, leading to increased transfection efficiency [8]. Sasaki’s research group found that coating a DNA/

1742-7061/$ - see front matter Ó 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2012.04.034

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

polycation polyplex with poly(c-glutamic acid) (c-PGA) achieved high levels of transgene expression both in vitro and in vivo, which was attributed to an increase in cellular uptake by a c-PGA-specific receptor-mediated energy-dependent process [3,9]. Our previous studies also revealed that poly(a-glutamic acid) (a-PGA) was able to significantly improve the transfection efficiency of a DNA polyplex in the presence of serum compared with a DNA polyplex lacking a-PGA [10]. Koyama’s research group demonstrated that the addition of anionic polyampholyte polyethyleneglycol (PEG) derivatives or hyaluronic acid derivatives to DNA cationic polyplexes avoided adverse interactions with blood components and assisted dissociation of the DNA polyplexes to enhance the transcriptional efficiency [11,12]. Until now protecting the DNA polyplex against unfavorable interactions with serum components and loosening the DNA polyplex within cells have been thought of as the main roles that polyanions play in improving the transfection efficiency of DNA polyplexes [8,13], although the mechanism is not yet fully understood. How DNA polyplexes interact with serum proteins and whether the particle size of the DNA polyplex can change in the presence of serum is still unknown. Furthermore, the anionic polysaccharides and polypeptides employed in previous studies contained carboxyl groups with various degrees of ionization and having obviously different structures. Whether different polyanions have the same positive effects on the transfection efficiency of DNA polyplexes have not been reported. In this study we have explored the influence of polyanion on the physico-chemical properties and transfection efficiency of polyanion/DNA/polycation ternary polyplexes. Four polysaccharides and polypeptides containing carboxyl groups or sulfonic acid groups were used as biocompatible polyanions to form ternary polyplexes with 25 kDa PEI and DNA. 25 kDa PEI was selected as a model polycation because it is well-established as an efficient non-viral gene vector under serum-free conditions [14]. All the optimal DNA ternary polyplexes formed with the four polyanions showed significantly higher transfection activities and lower cytotoxicities than their original DNA binary polyplexes. Interactions among the polyanion, PEI and DNA in the ternary polyplex and the ability of different polyanions to loosen the DNA polyplex were investigated. Changes in the DNA polyplex in the presence of serum were evaluated by comparing the stability and cellular uptake under serum-free and serum-containing conditions. Isothermal titration calorimetry was used as a real time tracking method to evaluate serum protein–DNA polyplex affinity.

(FBS) and penicillin/streptomycin were purchased from Gibco. Agarose was purchased from Biowest. Dimethyl sulfoxide (DMSO) and 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT) were obtained from Sigma–Aldrich. All other reagents were analytical grade. 2.2. Preparation of plasmid pEGFP-C1 The plasmid pEGFP-C1 (4.8 kbp) encoding the enhanced green fluorescent protein under control of the human cytomegalovirus (CMV) promoter as a model plasmid DNA (pDNA) was replicated in competent high copy strain DH5-a Escherichia coli grown in Luria–Bertani medium containing kanamycin (50 lg ml 1). Endotoxin-free pDNA was purified using an EndoFree Plasmid Kit according to the manufacturer’s protocol. The concentration of pDNA was determined from the UV absorbance at 260 nm using an extinction coefficient of 0.02 mg 1 cm 1 ml. Additionally, the purity was evaluated by UV spectroscopy (E 260/280 nm ratio). pDNA size and homogeneity were assayed by 1.0% agarose gel electrophoresis. 2.3. Preparation of the pEGFP-C1/polyanion/polycation ternary polyplexes 25 kDa PEI is considered one of the most potent synthetic gene carriers in vitro. It has been used as a typical polycation in DNA ternary polyplex formulations. The polyanion/pDNA/PEI ternary polyplexes were prepared as shown in Scheme 1. Briefly, 5 lg of pEGFP-C1 (50 lg ml 1, 100 ll) was mixed by gentle vortexing with various amounts of a 200 lg ml 1 solution of the polyanion (ALG, HEP), PASP or PGA) in phosphate-buffered saline, pH 7.4 (PBS), based on the A/N ratio, i.e. the molar ratio of anionic groups, including carboxyl and sulfonic acid groups, of the polyanion to

2. Materials and methods 2.1. Materials Heparin sodium salt (HEP) with an activity >170 units mg 1 and average molecular weight of 11.8 kDa was purchased from J&K Chemical Ltd. Alginic acid sodium salt (ALG) (low viscosity) was obtained from Sigma–Aldrich. Poly(aspartic acid) (PASP) with average an molecular weight of 10 kDa and poly(glutamic acid) (PGA) with an average molecular weight of 8 kDa were synthesized in our laboratory according to the procedure detailed in Wang et al. [10]. PEI an with average molecular weight of 25 kDa was purchased from Sigma–Aldrich. The pEGFP-C1 plasmid was a gift from the West China University of Medical Sciences. The EndoFree Plasmid Kit was purchased from Tiangen. HeLa cells were obtained from the American Type Culture Collection. Lipofectamine 2000 and ethidium bromide were obtained from Invitrogen Corp. Fluorescein isothiocyanate (FITC) and paraformaldehyde were obtained from Sigma. The Label ITÒ Cy5 Labeling Kit was purchased from Mirus. 4,6-Diamidine-2-phenylindole dihydrochloride (DAPI) was purchased from Roche. RPMI-1640 medium, fetal bovine serum

3015

Scheme 1. Formation of the polyanion/pDNA/PEI ternary polyplex.

3016

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

O

H O S O

*

O

HOOC O

HO COOH

m

OH O

O O

O

O n

S

HO

O O

O

OH

Alginic acid (ALG)

Heparin (HEP)

O

O

H N

COOH *

NH S

OH O

*

HO

O

O

H O O C OH O

OH O

O

n

NH

H N

*

CH2 CH

COOH m

O

p

Poly (aspartic acid) (PASP)

2

COOH

Poly (glumatic acid) (PGA)

Scheme 2. Chemical structure of polyanions in the DNA ternary polyplex.

amine groups of PEI. According to the polyanion structures shown in Scheme 2 the number of functional anionic groups in the repeat units of ALG, HEP, PASP, and PGA are 1, 4, 1, and 1, respectively. The amount of polyanion added to the DNA ternary polyplex was calculated using the formula: Wpolyanion = (A/N ratio  (WPEI/43)  molecular weight of the repeat units of the polyanion) Then 6.6 lg of 25 kDa PEI (200 lg ml 1, 33 ll) at the optimal N/ P ratio of 10 (the molar ratio of amine groups of PEI to phosphate groups of DNA) was added to the pEGFP-C1/polyanion mixture and rapidly mixed by pipetting. The resulting mixtures were incubated for 15 min at ambient temperature to yield the polyanion/pEGFPC1/PEI ternary polyplexes. 2.4. Size and zeta potential measurements The particle size and size distribution of the polyanion/pDNA/ PEI ternary polyplexes were measured by photon correlation spectroscopy (PCS) in a Malvern Zetasizer NS90 (Malvern Instruments, Malvern, UK). The instrument was equipped with a 10 mW helium–neon (He-Ne) laser producing light at a wavelength of 633 nm. The DNA polyplexes were dispersed in 5% glucose solution, commonly used as a diluting solution for injection. Measurements were carried out in cuvettes at 25 °C with a fixed scattering angle of 90° through a 400 lm pinhole. Each data point is comprised of at least three independent experiments. Zeta potential values were also obtained using a Zetasizer NS90 with a He-Ne laser beam. All measurements were done at a wavelength of 633 nm at 25 °C with a scattering angle of 90°. Samples were dispersed in 10 mM NaCl solution and zeta potentials were calculated from the mean electrophoretic mobility by applying the Smoluchowski equation. The results are the mean of five measurements ± standard deviation. 2.5. Agarose gel electrophoresis experiments To investigate the interactions between the polyanion, PEI and DNA in the ternary polyplexes and the ability of the polyanions to loosen the DNA polyplexes a series of gel retardation assays was performed by electrophoresis. The polyanion/pDNA/PEI ternary polyplexes were formulated at the optimal N/P ratio of 10 and

varying A/N ratios. Each DNA polyplex sample was loaded into the wells of a 1.0% agarose gel prepared in Tris–acetate–EDTA (TAE) buffer containing 0.5 lg ml 1 ethidium bromide. Samples were subsequently electrophoresed in a 110 V electric field for 60 min. Bands corresponding to pDNA were visualized on a UV transilluminator (UVP GDS 8000 Bioimaging System) and photographed. 2.6. In vitro transfection study HeLa cells were seeded into 24-well plates for 24 h at a density of 1.5  104 cells per well in 1.0 ml of culture medium. Prior to the experiment the cells were rinsed twice with warm PBS. 0.1 ml of polyanion/pDNA/PEI ternary polyplex and 0.4 ml of serum-free or serum-containing medium were added. The final pDNA concentration was 1.0 lg well 1. After transfection for 4 h at 37 °C in a 5% CO2 atmosphere at 90% humidity the cells were rinsed with warm PBS and placed in 0.5 ml of culture medium. After 48 h incubation to enhance green fluorescent protein expression the cells were rinsed twice with PBS. The green fluorescence was observed by fluorescence spectroscopy (Olympus IX71 fluorescence spectroscope). Subsequently the cells were treated with trypsin/EDTA for 2 min, collected by centrifugation, suspended in 0.3 ml of PBS and kept on ice until analysis. The percentage GFP-expressing cells was employed to quantify transfection efficiency by flow cytometry using a FACS-Calibur Instrument (Becton–Dickinson) equipped with an argon laser with an excitation wavelength of 488 nm. The filter setting for emission was 530/30 nm bandpass. Data was acquired in linear mode and visualized in linear mode. 2.7. Cell viability test MTT colorimetric assays were performed to evaluate whether biocompatible polyanions reduced the cytotoxicity of polymerbased vectors. Briefly, HeLa cells were seeded in 96-well plates at a density of 5000 cells well 1. After 24 h the culture medium was replaced with different concentrations of PEI/polyanion combined vector solution in serum-free or serum-containing medium. The concentrations of PEI were 50 and 2.56 lg ml 1, while the concentration of polyanion depended on the A/N ratio. After incubation for another 24 h 20 ll of MTT in 5 mg ml 1 PBS was added to each sample. After incubation for 4 h the supernatant was aspirated and

3017

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

A

30 pDNA/PEI ALG/pDNA/PEI HEP/pDNA/PEI PASP/pDNA/PEI PGA/pDNA/PEI

20

Zeta Potential (mV)

10 0 -10 -20 -30 -40 -50 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

B

0.9

1.0

pDNA/PEI ALG/pDNA/PEI HEP/pDNA/PEI PASP/pDNA/PEI PGA/pDNA/PEI

1800 1600 1400

Particle Size (nm)

0.8

A/N Ratio

1200 1000 800 600 400 200 0 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

A/N Ratio

Fig. 2. Agarose gel electrophoresis to assay the loosening effect of the polyanion on the compact DNA/PEI polyplex.

Fig. 1. (A) Zeta potential and (B) particle size of the polyanion/pDNA/PEI ternary polyplex at a N/P ratio of 10 and varying A/N ratios.

2.8. Fluorescence microscopic observation PEI was labeled with FITC at a molar ratio of 100:1 (PEI: FITC) [15]. Polyanion/pDNA/FITC-labeled PEI ternary polyplexes at the optimal A/N ratio for transfection efficiency and the original pDNA/FITC-labeled PEI binary polyplexes were prepared by the above mentioned method. The samples were incubated in 10% FBS-containing aqueous solution for 30 min at 37 °C. The final concentration of DNA was 0.5 lg ml 1. Fluorescence microscopic observation was carried out using an Olympus IX71 microscope and images were taken with an Olympus DP71 digital camera. Particle size and size distribution were measured by PCS with a Malven Zetasizer NS90. 2.9. Cellular uptake study Cellular uptake of the polyanion/pDNA/PEI ternary polyplexes was monitored by fluorescence-activated cell sorting (FACS). The

70 60

GFP Positive Cells (%)

the formazan crystals dissolved in 150 ll of DMSO. Absorption was measured at 570 nm with background correction using a Bio-Tek ELX800 ELISA reader. The results of eight measurements were normalized to the control group (exposure to full medium) (100%). Cells without addition of MTT were used as a blank to calibrate the spectrophotometer to zero absorbance. The turnover of the substrate relative to control cells was expressed as relative cell viability, and was calculated using the formula A(test)/A(control)  100%.

50

Serum Free

**

*

**

**

**

**

**

**

** **

40

10% FBS

*

** **

** ALG/pDNA/PEI HEP/pDNA/PEI PASP/pDNA/PEI PGA/pDNA/PEI pDNA/PEI pDNA/Lipo 2K

30 20 10 0

-- =0.1 =0.2 =0.3 =0.4 25K 2K -- =0.1 =0.2 =0.3 =0.4 25K 2K N N N N I po N N N N I po A/ A/ A/ A/ PE Li A/ A/ A/ A/ PE Li

--

Hela Cells Transfection Fig. 3. Effect of the polyanion on DNA polyplex-mediated gene delivery. HeLa cells were transfected with the polyanion/pEGFP-C1/PEI ternary polyplex at a N/P ratio of 10 and varying A/N ratios of polyanion to PEI. The percentage of GFP-positive cells was assessed 48 h post-transfection. Significant differences are marked with asterisks (⁄P < 0.05, ⁄⁄P < 0.01).

3018

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

pDNA in the polyplex was labeled with Cy5 for fluorescent detection. HeLa cells were seeded 24 h prior to the experiments in 24-well plates at a density of 1.5  104 cells per well in 1.0 ml of culture medium. The Cy5-labeled pDNA ternary and binary polyplexes were prepared and then added to the cells. After 4 h incubation with the Cy5-labeled pDNA polyplexes in serum-free or serum-containing medium the cells were rinsed twice with PBS, trypsinized with trypsin/EDTA for 2 min, washed with PBS, and fixed with fresh 4% paraformaldehyde for 5 min at room temperature. The fixed cells were washed once and resuspended in PBS and stored at 4 °C prior to FACS measurement. In all FACS analyses cell debris and aggregates were excluded by setting a gate on the plot of side-scattered light (SSC) vs. forward-scattered light (FSC). The cell fluorescence intensity was measured with a Becton–Dickinson FACS-Calibur equipped with a He-Ne ion laser (633 nm) and using the FL4 = 675 nm band path. A total of 10,000 gated cells were analyzed using Win MDI software (Joe Trotter, Scripps Institute). Non-treated cells were analyzed in parallel as a negative control.

ratios (the molar ratio of anionic groups, carboxyl and sulfonic acid groups, of the polyanion to amine groups of PEI), as portrayed in Scheme 1. The particle size and f-potential of the polyanion/ pDNA/PEI ternary polyplexes are shown in Fig. 1. The original DNA/PEI binary polyplexes had a diameter of about 128.3 ± 7.6 nm and a f-potential cationic charge of about 29.6 ± 1.9 mV. Upon addition of polyanion (ALG, HEP, PGA, or PASP, as shown in Scheme 2) at an A/N ratio in the range 0.3–0.5 the particle size of the DNA ternary polyplexes increased sharply (to more than 1 lm) and the f-potential was reduced to near zero, most likely due to ternary polyplex aggregation induced by charge neutralization. However, The DNA ternary polyplexes still formed small particles of about 200 nm diameter on addition of more polyanion to an A/N ratio of 1.0, with the positive surface charge being reversed to highly negative. These results suggest that the DNA ternary polyplexes formed polyelectrolyte complexes, with the stability mainly dependent on a charge repulsion effect. 3.2. Capacity of the polyanion to loosen the DNA polyplexes

2.10. Confocal laser scan microscopy study

2.11. Isothermal titration calorimetry (ITC) assay The serum protein binding affinity to the DNA polyplexes was determined by ITC studies with a MicroCal Auto-iTC200 isothermal titration calorimeter (MicroCal Inc.) at 25 °C. 450 ll of DNA polyplex (DNA concentration 5 lg ml 1) was added to a cell. To determine the binding affinity 250 ll of FBS stock solution was added step-wise. Typically 15 additions of a 2 ll volume were made at intervals of 150 s between each addition. The first 2 ll FBS titration in each experiment was subsequently deleted from the data analysis. The titration data were integrated and analyzed using Origin using the MicroCal software provided. The ITC fitting curve was obtained from the volume ratio of FBS to DNA complex solution.

The capacity of the polyanions to loosen the DNA polyplexes were studied using a gel retardation assay. Fig. 2 shows the retardation assay images of the polyanion/pDNA/PEI ternary polyplexes at an N/P ratio of 10 and varying A/N ratios. On adding polyanion to the DNA/PEI binary polyplexes above a critical A/N ratio a free DNA band was detected for all four polyanion/pDNA/PEI ternary

A 120 100

Cell Viability (%)

The cellular entry of Cy5-labeled pDNA/FITC-labeled PEI binary polyplexes and Cy5-labeled pDNA/polyanion/FITC-labeled PEI ternary polyplexes under 10% FBS-containing conditions was monitored by confocal laser scanning microscopy (CLSM) (Zeiss LSM710). The nuclei of cells were stained with DAPI. The cells transfected with DNA binary and ternary polyplexes were fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. After washing with PBS the coverslip was mounted on a glass microscope slide using a drop of diazabicyclooctan antifade solution (2.5% w/v). The samples were observed by CLSM with a He-Ne laser (650 nm excitation) to induce the red fluorescence of Cy5 and the emission was observed at 670 nm. The FITC fluorescence was observed at 520 nm with an argon laser (492 nm excitation) to induce green fluorescence. The blue fluorescence of DAPI was induced by 350 nm excitation with a UV laser and detected at a wavelength of 430 nm.

80

ALG/PEI HEP/PEI PASP/PEI PGA/PEI

60 40 20 0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

A/N ratio (Serum Free)

B

2.12. Statistical analysis Data are expressed as the mean ± the standard error of the mean. Statistical comparisons were performed using one-way analysis of variance (ANOVA). P values <0.05 or <0.01 were taken as statistically significant and highly statistically significant, respectively. 3. Results 3.1. Formation of the polyanion/pDNA/polycation ternary polyplexes The polyanion/pDNA/PEI ternary polyplexes were formed through self-assembly of negatively charged pEGFP-C1 and polyanions with positively charged PEI at a N/P ratio of 10 and varying A/N

Fig. 4. (A) Cytotoxicity of PEI/polyanion combined vectors at a PEI concentration of 50 lg ml 1 and varying A/N ratios in serum-free medium. (B) PEI/polyanion combined vectors at the optimal concentration (PEI concentration 2.65 lg ml 1) for the transfection experiments.

3019

Volume (%)

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

Size Distribution by Volume

40

d1=7.0 nm

30 20

d2=27.6 nm

10 0 1

10

100

1000

10000

Size (d.nm)

10%FBS

Size Distribution by Volume Volume (%)

Volume (%)

PEI

15 10

d=124.6 nm

5 0

Size Distribution by Volume

15

20

10

d1=155.3 nm

5

d2=806.2 nm

0 1

10

100

1000

1

10000

10

Volume (%)

ALG

Volume (%)

Size Distribution by Volume 12 10 8 6 4 2 0

d=188.3 nm

1

10

100

100

1000

10000

Size (d.nm)

Size (d.nm)

1000

25 20 15 10 5 0

Size Distribution by Volume

d=361.9 nm

1

10

100

1000

10000

Size (d.nm)

10000

Size (d.nm)

Volume (%)

Volume (%)

HEP

15 10

d=879.4 nm

5 0

20

10

100

1000

d=398.6 nm

10 0

1

Size Distribution by Volume

30

Size Distribution by Volume 20

1

10

100

1000

10000

Size (d.nm)

10000

Size (d.nm)

Volume (%)

Volume (%)

20

PASP

15 10

d=950.9 nm

5 0

20 10 0

1

10

100

1000

Size Distribution by Volume

30

Size Distribution by Volume

d=409.5 nm 1

10

100

1000

10000

Size (d.nm)

10000

25 20 15 10 5 0

Size Distribution by Volume Volume (%)

PGA

Volume (%)

Size (d.nm)

d=830.4 nm

1

10

100

1000

10000

25 20 15 10 5 0

Size Distribution by Volume

d=394.5 nm

1

10

100

1000

10000

Size (d.nm)

Size (d.nm)

Before incubation

After incubation

Fig. 5. Effect of the polyanion-protected DNA polyplex on FBS-induced aggregation. Fluorescence microscopic images (200) and the size distribution of the polyanion/ pDNA/PEI ternary polyplex before and after incubation in 10% FBS in deionized water.

polyplexes. It is suggested that the polyanions caused DNA dissociation from the polyplex by competitive complexing with PEI. Competitive binding of the polyanions to PEI rather than DNA should result in a weakening of the interaction between DNA and polycation and loosening of the compact DNA polyplex. Among the four DNA polyplexes that containing HEP showed the lowest critical A/ N ratio of 0.5, while that containing ALG had the highest critical A/N ratio of 2.0, demonstrating that HEP and ALG possessed the strongest and weakest abilities to loosen the DNA polyplex among the four polyanions.

3.3. In vitro transfection efficiency of the polyanion/pDNA/polycation ternary polyplexes To investigate whether the polyanion/pDNA/polycation ternary polyplex formulations could increase the efficiency of polymermediated gene transfer the polyanion/pEGFP-C1/PEI ternary polyplex was assessed for in vitro transfection efficiency by green fluorescent protein (GFP) assay. The optimal A/N ratios of polyanion to PEI in the presence of serum were 0.1, 0.2, 0.3 and 0.3 for the ALG/pDNA/PEI, HEP/pDNA/PEI, PASP/pDNA/PEI and PGA/pDNA/PEI

3020

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

Fig. 6. (A) Percentages of Cy5-positive cells after 4 h incubation in serum-free and serum-containing media. (B) Intracellular accumulation levels of DNA polyplex. (C) A representative flow cytometry histogram showing the intracellular accumulation of the DNA polyplex after 4 h incubation in 10% FBS in deionized water.

ternary polyplexes, respectively. We found that the gene transfection activities of the four polyanion/pEGFP-C1/PEI ternary polyplex samples at their optimal A/N ratios were significantly higher than that of a pEGFP-C1/PEI binary polyplex at its optimal N/P ratio of 10 under both serum-free and serum-containing conditions (P < 0.01). They were also greatly higher than that of a pEGFP-C1/ Lipofectamine 2000 complex at a pDNA: lipid ratio of 1:2.5 (w/ w) in serum-containing medium (P < 0.01), as shown in Fig. 3. The four polyanion/pDNA/PEI ternary polyplexes also achieved high levels of GFP expression compared with the original binary polyplexes (Supplementary Fig. S1). The transfection results confirmed that the polyanion/pDNA/polycation ternary polyplexes could improve gene transfection activity in comparison with the original pDNA/polycation binary polyplexes. The following experiments were performed to further explore the influence of these polyanions on the polyanion/pDNA/polycation ternary polyplexes at their optimal A/N ratios of polyanion to PEI to achieve high transfection efficiency.

3.4. Cytotoxicity of the PEI/polyanion combined vectors Polycations can induce cytotoxicity due to their high positive surface charges having non-specific membrane destabilizing effects [16]. The oppositely charged polyanions complexed with polycations could neutralize the high positive surface charges of polycations. Colorimetric MTT assays were performed to assess the

cytotoxicity of the PEI/polyanion combined vectors. Fig. 4A shows that all of the PEI/polyanion combined vectors at A/N ratios from 0.1 to 3.0 were less toxic than 25 kDa PEI. The HEP/PEI, PASP/PEI and PGA/PEI combined vectors at A/N ratios above 0.3 and the ALG/PEI combined vector at A/N ratios above 0.5 were not toxic to HeLa cells and cell viability was more than 95%. Conversely, only 12.5 ± 3.0% of HeLa cells incubated with 25 kDa PEI (50 lg ml 1) remained viable at the same concentration of PEI as in the PEI/polyanion combined vectors. The ALG/PEI combined vector resulted in less viability compared with the other three polyanion/PEI combined vectors at the same A/N ratio, which could be explained by the results of gel retention experiments that showed that ALG had a weak ability to neutralize the highly positive charges of PEI compared with the other three polyanions tested. However, the cell viability profiles showed little difference between 25 kDa PEI and PEI/polyanion combined vectors at the optimal A/N ratios and at the working concentration for transfection experiments (Fig. 4B). These results suggest that the reduced cytotoxicity of the polycationic vectors due to the polyanions was not a critical factor in improving the gene transfection efficiency of DNA polyplexes in vitro.

3.5. Effect of the polyanion/pDNA/PEI ternary polyplexes on FBSinduced aggregation Stability of the DNA complex is a prerequisite for efficient gene transfection for in vivo gene therapy. Fluorescence microscopic

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

3021

Fig. 7. Confocal microscope images of HeLa cells incubated with the DNA polyplex for 4 h in 10% FBS in deionized water. Nuclei were stained blue with DAPI. Cy5-labeled pDNA and FITC-labeled PEI are shown as red and green fluorescence, respectively. The focal plane of the samples was the middle of the nuclei.

observations and particle size distribution measurements were performed to investigate whether the polyanions could inhibit DNA polyplex aggregation in serum-containing medium. In the presence of serum the small-sized pDNA/PEI binary polyplex formed large aggregates. The corresponding size distribution exhibited two distinct peaks (one at 155.3 nm and the other at 806.2 nm) rather than the single peak before FBS addition, as shown in Fig. 5. This confirmed that aggregates of the pDNA/PEI binary polyplex were induced by FBS. In contrast, the four polyanion/pDNA/FITC-labeled PEI ternary polyplexes treated with 10% FBS did not show obvious aggregation in comparison with the same samples prior to the addition of FBS. Interestingly, it was noted that all the polyanion/pDNA/PEI ternary polyplexes except

for the ALG/pDNA/PEI ternary polyplex tended to form smaller particles with narrower size distributions when incubated with 10% FBS than those ternary polyplexes without added FBS. This result demonstrated that the polyanions could effectively protect DNA polyplexes against FBS-induced aggregation. 3.6. Cellular uptake and intracellular accumulation of the polyanion/ pDNA/PEI ternary polyplexes Cellular uptake efficiency and intracellular accumulation levels of the polyanion/pDNA/PEI ternary polyplexes were measured to determine whether they contributed to the high transfection efficiency. When the Cy5-labeled DNA polyplexes were incubated

3022

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

0.00

B 0.00

-1.00

-1.00

µcal/sec

-2.00

µcal/sec

A

0 -2000

-3.00

-4000

-2.00 0

-3.00

-2000 -4000

-4.00

-4.00 -6000

-6000

-5.00

pDNA/PEI

-6.00 0.00

-8000 0

10.00

10

20.00

20

-5.00

30

ALG/pDNA/PEI -8000 0

0.00

30.00

10.00

Time (min) 0.00

D 0.00

-1.00

-1.00

-2.00

-2.00

0

-2000

-3.00

20.00

-4000

-4.00 -8000 0

10.00

30.00

0

-3.00

-4000

-4.00 -6000

HEP/pDNA/PEI

-6.00 0.00

30

-2000

-6000

-5.00

20

Time (min)

µal/sec

µal/sec

C

10

10

20.00

20

-5.00

30

PASP/pDNA/PEI

-6.00 0.00

30.00

10.00

Time (min)

-8000 0

10

20.00

20

30

30.00

Time (min)

E 0.00 -1.00 -2.00

µal/sec

0

-2000

-3.00 -4000

-4.00 -5.00

-6000

PGA/pDNA/PEI

-6.00 0.00

10.00

-8000 0

10

20.00

20

30

30.00

Time (min) Fig. 8. (A–E) Typical ITC data recorded for DNA polyplex and FBS interactions. 15 injections of FBS solution were added to the DNA polyplex dispersion in a ITC cell. The area underneath each injection peak (main panels) was equal to the total heat released for that injection. When this integrated heat was plotted against the volume of FBS added to the DNA polyplex in the cell a complete binding isotherm for the interaction was obtained (insets).

with HeLa cells for 4 h under the same conditions as in the transfection experiments all of the DNA ternary polyplexes tested showed 2–3-fold higher levels of intracellular accumulation than the original DNA binary polyplexes. The cellular uptake efficiencies of the four DNA ternary polyplexes showed little difference from each other in serum-free medium (Fig. 6A and B). In the presence of serum the Cy5-positive fraction of cells transfected by the polyanion/pDNA/PEI ternary polyplexes decreased only slightly and the intracellular accumulation was maintained at high levels. In contrast, the cellular uptake efficiency and intracellular accumulation levels of the original pDNA/PEI binary polyplex was reduced by more than half in serum-containing medium, compared with in serum-free medium. In addition, transfection with Lipofectamine 2000 resulted in a high percentage of Cy5-positive cells in both serum-free and serum-containing medium, but the intracellular accumulation levels were significantly lower than those of the polyanion/pDNA/PEI ternary polyplexes (Fig. 6B and C). These results suggest that high levels of intracellular accumulation induced by the DNA ternary polyplex-

es are one important factor improving the gene transfection efficiency of DNA polyplexes. Cellular uptake of the DNA polyplexes was further observed using CLSM. The fluorescence signal emitted from FITC-labeled PEI was green, while emission from the Cy5-labeled pDNA was red. The overall emission from the polyanion/pDNA/PEI ternary polyplex and DNA/PEI binary polyplex was yellow or orange (a combination of overlapping green and red fluorescence). Fluorescent signal spots were detected around the nuclei of cells for all samples by 4 h post-transfection in 10% serum-containing medium, as shown in Fig. 7. Furthermore, the number of fluorescent spots was much greater and their intensity was also obviously stronger in cells transfected by the four polyanion/pDNA/PEI ternary polyplexes, compared with the original pDNA/PEI binary polyplex. These images confirmed that cellular entry of the four DNA ternary polyplexes was more efficient than that of the DNA binary polyplex. Considering that the only different component in the formulations of the DNA ternary polyplex and binary polyplex was the polyanion it is reasonable to assume that the

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

Maximal exothermic energy (Cal/mol)

50000 6500 40000 30000

5500

20000 4500 10000

Total exothermic energy (Cal/mol)

60000

7500

0

A

PG

A

/p D

N

N

/P EI

A/ PE I

/P EI A N

SP /p D PA

EP /p D H

A

LG

pD

/p D

N

N

A

/P EI

A/ PE I

3500

Fig. 9. The maximal (red bar) and total (blue square) exothermic energy released in the interaction between the DNA polyplex and FBS. The N/P ratio of the PEI/DNA polyplex was 10. The optimal A/N ratios for the ALG, HEP, PASP and PGA ternary polyplexes were 0.1, 0.2, 0.3 and 0.3, respectively.

polyanion overcame to some degree serum inhibition of polycation-mediated gene delivery.

3023

shown in Fig. 8. The main curves in Fig. 8 show the raw data obtained during each addition. Each peak represents the heat variation related to the addition of a small aliquot of FBS to the ITC reaction cell containing the DNA polyplex sample. As increasing amounts of FBS were added to the ITC cell the heat released was directly proportional to the level of protein binding. As the system reached saturation heat release decreased until only the heat of dilution was observable. The insets correspond to the integrated calorimetric response plotted against the total volume of titrate added. There was a cleared difference in the titration profiles of the DNA binary and ternary polyplexes. The initial exothermic release measured by ITC usually represents the interaction strengths of the titrant and the test substance [17]. Interaction between the pDNA/PEI binary polyplex and FBS during early titration showed a value of more than 7000 cal mol 1, which was significantly higher than those of the polyanion/pDNA/PEI ternary polyplexes (4122–5828 cal mol 1) (Fig. 9). These results demonstrate that polyanion addition to the DNA/polycation polyplex formulations decreased their affinities for FBS. The total exothermic heat release by the samples was calculated to evaluate the amount of FBS adsorbed onto the DNA polyplex. The corresponding total exothermic heat release for the pDNA/PEI binary polyplex was 51132 cal mol 1. The lower total exothermic heat release by the polyanion/pDNA/PEI ternary polyplex in the range 3291–45914 cal mol 1 implies that FBS binding to the polyanion/pDNA/PEI ternary polyplex was less than to the original pDNA/PEI binary polyplex. This result could also explain the fluorescence microscopic observation of polyanion protection of the DNA polyplex against FBS-induced aggregation.

3.7. Thermodynamic analysis of the interactions between the DNA polyplexes and FBS 4. Discussion ITC was used to investigate the thermodynamic affinities of FBS and the DNA polyplexes. The ITC responses recorded during titration of FBS with the DNA binary and ternary polyplexes are

Safe and versatile cationic polymers holding the promise of replacing viral vectors have been widely investigated in recent

Scheme 3. Schematic illustration of the adsorption of serum proteins in culture medium resulting in dramatic changes in stability of the polyanion/pDNA/PEI ternary polyplex and pDNA/PEI binary polyplex.

3024

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

decades. However, 70% of clinical gene therapy trials to date have relied on viral vectors to deliver the desired gene. A major limitation of the currently used cationic polymers as non-viral vectors is their low gene transport and expression efficiency compared with viral vectors [18]. Several research groups have reported that addition of polyanions such as PGA, hyaluronic acid and anionic polyampholyte PEG derivatives to a polycation system, forming polyanion/pDNA/polycation ternary polyplexes, significantly improved the gene delivery efficiency in comparison with their original pDNA/polycation binary polyplexes [7–12]. It is a versatile and effective method to overcome the limitations of cationic polymerbased gene delivery systems. However, the mechanism by which polyanions result in high gene transport efficiency is not yet fully understood. Whether general polyanions can improve the transfection efficiencies of polycation-based gene delivery systems is still controversial [9,19]. Furthermore, what happens during polyanion/pDNA/polycation ternary polyplex cell transfection under serum-containing conditions has not been reported. To better understand previous experimental findings our present work was aimed at revealing the influence of biocompatible polyanions on the physico-chemical properties and transfection efficiency of polyanion/pDNA/polycation ternary polyplexes. Four types of biocompatible polyanions, including two polysaccharides (alginate and heparin) and two polypeptides (poly(aspartic acid) and poly(glutamic acid)), were selected to form polyanion/pDNA/PEI ternary polyplexes with DNA and 25 kDa PEI. Loosening of the DNA polyplex is thought to be an important role that polyanions play in improving the transfection efficiency of DNA polyplexes. Urtti’s work showed that excess anionic glosaminoglycans (heparan sulfate, chondrotin sulfate B, chondroitin sulfate C and hyaluronic acid) could enhance DNA release from DNA/PEI polyplexes [20]. Our experiments have revealed that the polyanions, pDNA and PEI only successfully formed ternary polyplexes at A/N ratios below a critical value for dissociation, while DNA dissociated from the polyplex above the critical value (Fig. 2). The competitive binding of polyanions with PEI rather than DNA in the ternary polyplex could result in loosening of a compact DNA/PEI polyplex. The ability of polyanions to loosen DNA polyplexes based on the critical A/N ratio for dissociation can be ranked as follows: HEP > PASP = PAG > ALG. HEP had the lowest critical A/ N ratio of 0.5, most likely due to HEP containing not only carboxyl groups but also a larger number of N- and O-sulfate groups. The sulfate groups in HEP are a strong acid with a pKa of approximately 3.1 [21], which is the lowest among the four polyanions that show very strong electrostatic interactions with cationic PEI and electrostatic competition with DNA. The other three polyanions (ALG, PASP and PGA) only carry carboxyl groups on the side-chains, which are weak acids, resulting in relatively weak electrostatic bonds with PEI. It should also be noted that although the pKa of ALG (3.38 and 3.65 for guluronic acid and mannuronic acid, respectively [22]) is lower than that of PASP (about 4.0 [23]) and PGA (about 4.3 [24]), the critical A/N ratio for ALG-induced dissociation of DNA from the polyplex was higher than those of the two polypeptides. We presume that compared with the flexible polypeptides PASP and PGA the six-membered ring structure of the monomeric units of ALG is more rigid, weakening the interactions between ALG and PEI. Therefore, the capacities of polyanions to loosen DNA polyplexes were found to depend on the pKa value of the polyanions as well as their flexibility. In vitro transfection results confirmed that the incorporation of biocompatible polyanions, including polysaccharides and polypeptides, was indeed favorable for exogenous gene delivery. The optimal A/N ratios for the four polyanion/pDNA/PEI ternary polyplexes were ranked as follows: ALG < HEP < PASP = PGA, which was not completely consistent with the ability of polyanions to loosen DNA polyplexes. These results motivated us to further explore

additional important properties of polyanions, which can achieve high transfection efficiencies using DNA ternary polyplexes. Cationic polymers like PEI are highly cytotoxic to many cell lines, which limits their potential application in vivo [16]. In this study four types of biocompatible polyanion sharply reduced the polyanion/PEI combined vector toxicity at PEI high concentrations (0.05 lg ll 1, Fig. 4A). However, there were no significant differences in cytotoxicity between PEI and the polyanion/PEI combined vectors at the PEI concentrations used in the transfection experiments (Fig. 4B). These results suggest that the improved biocompatibility of the polyanion/PEI combined vectors is not a critical factor contributing to their high gene delivery efficiency. Cellular uptake is considered to be a crucial step in improving the transfection efficiency of non-viral gene delivery vectors [25]. Generally, positively charged DNA polyplexes are assumed to be taken up well by cells because cationic molecules are attracted by the negatively charged cell surface [26,27]. Our results indicate that all the slightly positively charged polyanion/pDNA/PEI ternary polyplexes and the highly positively charged pDNA/PEI binary polyplex achieved almost the same percentage cellular uptakes (>95%) in serum-free medium, whereas the cells endocytosed larger amounts of the sub-micron DNA ternary polyplexes compared with the small DNA binary polyplex. These findings confirmed earlier studies that the cellular uptake pathway of particulate system depends on both their surface properties and particle size [26]. They also suggest that the high intracellular accumulation levels of the large sized DNA ternary polyplexes were important in improving the transfection efficiency. It is well known that the presence of serum can reduce the transfection efficiency of DNA polyplexes [28]. In this study all the DNA binary and ternary polyplex samples showed a decrease in cellular uptake and intracellular accumulation in the presence of serum. The original pDNA/PEI binary polyplex showed an approximately 60% reduction in cellular uptake and intracellular accumulation, which was proportional to the magnitude of the reduction in transfection efficiency. However, the reduction in cellular uptake of the polyanion/pDNA/PEI ternary polyplexes was much less than that of the original DNA binary polyplex (Fig. 6). What happened to the DNA ternary polyplexes and the original binary polyplex when they were incubated in serum-containing medium was investigated further. We found that the size and size distribution of the polyanion/pDNA/PEI ternary polyplexes in serum-containing medium were obviously smaller and narrower compared with in serum-free medium, except for ALG/pDNA/PEI (Fig. 6). In contrast, some of the pDNA/PEI binary polyplex formed large particles with sizes of more than 0.8 lm under 10% serumcontaining conditions. Although serum proteins (7–27.6 nm) may interfere with the particle size measurements, the size change could be confirmed by making measurements under the same conditions. Furthermore, the fluorescence observations also indicated that the original pDNA/PEI binary polyplex formed extensive aggregates in the presence of serum, leading to clustered DNA/ PEI binary polyplexes (Fig. 6). In contrast, the addition of serum to the polyanion/pDNA/PEI ternary polyplex dispersions did not result in obvious aggregation. These results are consistent with the size change measured by PCS. Based on the above experimental data we conjecture that the original stable DNA/PEI binary polyplex, which has a high surface energy (a high surface charge and a small particle size), adsorbed a lot of oppositely charged serum proteins from the culture medium, resulting in neutralization of the pDNA/PEI binary polyplex charges and polyplex cluster formation. In contrast, submicron sized polyanion/pDNA/PEI ternary polyplexes with near neutral surface charges in the metastable state became relatively stable in serum-containing medium due to relatively strong repulsive electric forces between each DNA ternary polyplex with relatively fewer adsorbed serum proteins. The

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026

repulsive electric forces overcome the attractive forces among submicron sized DNA ternary polyplexes and prevent their aggregation, with the formation of even smaller particles, as shown in Scheme 3. In general, electrostatic repulsion and steric hindrance are the key factors in maintaining the physical stability of this particulate system [29]. The metastable state of the near stoichiometric polyanion/pDNA/PEI ternary polyplexes changed to a relatively stable state when they were incubated in serum-containing medium, implying that interaction of the DNA polyplexes, as polyelectrolyte particles, with oppositely charged biological macromolecules that exist in vivo is an important factor in their physical stability. The conjecture that polyanion/pDNA/PEI ternary polyplexes have lower affinities for serum components and tend to adsorb fewer serum proteins from the culture medium than DNA/PEI binary polyplexes was also supported by the ITC results. ITC is an accurate, real time thermodynamic tracking technique that has been successfully used to directly measure protein–protein interactions and protein–nanoparticle and protein–ligand complexation [30]. It avoids the measurement errors in gel filtration and density gradient centrifugation separation methods such as SDS–PAGE and BCA assay. In all ITC experiments the interactions between FBS and the DNA polyplexes were exothermic (Fig. 8), indicating that serum proteins were indeed adsorbed onto both the DNA binary polyplex and four types of DNA ternary polyplex. The interaction strengths between serum proteins and the DNA polyplexes were evaluated on the basis of the exothermicity during the initial stage of interaction [31]. The original pDNA/PEI polyplex has been shown to bind serum proteins exceptionally strongly, whereas the four polyanion/pDNA/PEI ternary polyplexes had relatively weak affinities for serum proteins (Fig. 9). Furthermore, the pDNA/PEI binary polyplex showed much higher total exothermic heat release than the four polyanion/pDNA/PEI ternary polyplexes, reaching an adsorption saturation that is in direct proportion to the number of bound proteins. The results provide direct evidence confirming our above conjecture that the low affinities of the polyanion/ pDNA/PEI ternary polyplexes for serum proteins and the fewer adsorbed proteins is one important factor in their high transfection efficiencies under serum-containing conditions.

5. Conclusion Polyanions/DNA/polycation ternary polyplexes showed significantly higher transfection efficiencies than the original DNA/polycation binary polyplex especially under serum containing conditions in early studies. Four types of biocompatible polyanions were selected to explore their influence on the physico-chemical properties and transfection efficiencies of polyanion/DNA/polycation ternary polyplexes. The ability of the polyanions to loosen DNA polyplexes depended on the pKa values of their functional anion groups as well as the flexibility of the polyanions. The low pKa and flexible structure of the polyanions tended to loosen the compact DNA polyplexes at a relatively low A/N ratio. The polyanion/ pDNA/polycation ternary polyplexes exhibited obviously lower binding affinities and less serum protein adsorption than the original DNA/polycation binary polyplex. These relatively stable DNA ternary polyplexes retained high levels of cellular uptake and intracellular accumulation in serum-containing medium, which was correlated with their high transfection efficiency. In contrast, the original pDNA/polycation binary polyplex formed clustered polyplexes by strong adsorption of large amounts of serum proteins, leading to a sharp reduction in cellular uptake and intracellular accumulation, and a low gene transfer efficiency. These results provide a basis for the development of polyanion/DNA/ polycation ternary polyplexes for polyfection.

3025

Acknowledgements The authors gratefully acknowledge the National Natural Science Foundation of China (project no. 30600789) and the Guangdong Natural Science Fund (project no. 9151063301000008) for their financial support of this research. Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figs. 1, 3–7 and 9, are difficult to interpret in black and white. The full colour images can be found in the on-line version, at http://dx.doi.org/10.1016/ j.actbio.2012.04.034. Appendix B. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.actbio.2012. 04.034. References [1] Li S, Huang L. Nonviral gene therapy: promises and challenges. Gene Ther 2000;7:31–4. [2] Park TG, Jeong JH, Kim SW. Current status of polymeric gene delivery systems. Adv Drug Deliv Rev 2006;58:467–86. [3] Ito T, Yoshihara C, Hamada K, Koyama Y. DNA/polyethyleneimine/hyaluronic acid small complex particles and tumor suppression in mice. Biomaterials 2010;31:2912–8. [4] Kurosaki T, Kitahara T, Kawakami S, Higuchi Y, Yamaguchi A, Nakagawa H, et al. Gamma-polyglutamic acid-coated vectors for effective and safe gene therapy. J Control Release 2010;142:404–10. [5] Ito T, Abe K, Yoshihara C, Tanaka CI, Koyama Y. Efficient in vivo gene expression by DNA/polycation/polyanion ternary complexses. Mol Ther 2006;13:S72. [6] Jiang G, Min SH, Oh EJ, Hahn SK. DNA/PEI/alginate polyplex as an efficient in vivo gene delivery system. Biotechnol Bioproc Eng 2007;12:684–9. [7] Xu P, Quick GK, Yeo Y. Gene delivery through the use of a hyaluronateassociated intracellularly degradable crosslinked polyethyleneimine. Biomaterials 2009;30:5834–43. [8] Hornof M, de la Fuente M, Hallikainen M, Tammi RH, Urtti A. Low molecular weight hyaluronan shielding of DNA/PEI polyplexes facilitates CD44 receptor mediated uptake in human corneal epithelial cells. J Gene Med 2008;10:70–80. [9] Kurosaki T, Kitahara T, Fumoto S, Nishida K, Nakamura J, Niidome T, et al. Ternary complexes of pDNA, polyethylenimine, and c-polyglutamic acid for gene delivery systems. Biomaterials 2009;30:2846–53. [10] Wang C, Feng M, Deng J, Zhao Y, Zeng X, Han L, et al. Poly(a-glutamic acid) combined with polycation as serum-resistant carriers for gene delivery. Int J Pharm 2010;398:237–45. [11] Ito T, Iida-Tanaka N, Niidome T, Kawano T, Kubo K, Yoshikawa K, et al. Hyaluronic acid and its derivative as a multi-functional gene expression enhancer: protection from non-specific interactions, adhesion to targeted cells, and transcriptional activation. J Control Release 2006;112:382–8. [12] Koyama Y, Yamashita M, Iida-Tanaka N, Ito T. Enhancement of transcriptional activity of DNA complexes by amphoteric PEG derivative. Biomacromolecules 2006;7:1274–9. [13] Yoshihara C, Shew CY, Ito T, Koyama Y. Loosening of DNA/polycation complexes by synthetic polyampholyte to improve the transcription efficiency: effect of charge balance in the polyampholyte. Biophys J 2010;98:1257–66. [14] Demeneix B, Behr JP. Polyethylenimine (PEI). Adv Genet 2005;53:217–30. [15] Gosselin MA, Guo W, Lee RJ. Efficient gene transfer using reversibly crosslinked low molecular weight polyethylenimine. Bioconjug Chem 2001;12:989–94. [16] Fischer D, Li Y, Ahlemeyer B, Krieglstein J, Kissel T. In vitro cytotoxicity testing of polycations: influence of polymer structure on cell viability and hemolysis. Biomaterials 2003;24:1121–31. [17] Cedervall T, Lynch L, Lindman S, Berggard T, Thulin E, Nilsson H, et al. Understanding the nanoparticle-protein corona using methods to quantify exchange rates and affinities of proteins for nanoparticles. Proc Natl Acad Sci USA 2007;104:2050–5. [18] Niidome T, Huang L. Gene therapy progress and prospects: nonviral vectors. Gene Ther 2002;9:1647–52. [19] Ruponen M, Arkko S, Urtti A, Reinisalo M, Ranta VP. Intracellular DNA release and elimination correlate poorly with transgene expression after non-viral transfection. J Control Release 2009;136:226–31. [20] Ruponen M, Ylä-Herttuala S, Urtti A. Interactions of polymeric and liposomal gene delivery systems with extracellular glycosaminoglycans:

3026

[21]

[22]

[23]

[24]

[25]

C. Wang et al. / Acta Biomaterialia 8 (2012) 3014–3026 physicochemical and transfection studies. Biochim Biophys Acta 1999;1415: 331–341. Wang H, Loganathan D, Linhardt R. Determination of the pKa of glucuronic acid and the carboxy groups of heparin by carbon-13 nuclear magnetic resonance spectroscopy. Biochem J 1991;278:689–95. Kerchove J, Elimelech M. Structural growth and viscoelastic properties of adsorbed alginate layers in monovalent and divalent salts. Macromolecules 2006;39:6558–64. Joshi MD, Hedberg A, Mcintosh LP. Complete measurement of the pKa values of the carboxyl and imidazole groups in Bacillus circulans xylanase. Protein Sci 1997;6:2667–70. Kono K, Nishii H, Takagishi T. Fusion activity of an amphiphilic polypeptide having acidic amino acid residues: generation of fusion activity by alpha-helix formation and charge neutralization. Biochim Biophys Acta 1993;1164:81–90. Konate K, Crombez L, Deshayes S, Decaffmeyer M, Thomas A, Brasseur R, et al. Insight into the mechanism of the peptide-based gene delivery system MPG: implications for delivery of siRNA into mammalian cells. Biochemistry 2010;49:3393–402.

[26] Okuda-Shimazaki J, Takaku S, Kanehira K, Sonezaki S, Taniguchi A. Effects of titanium dioxide nanoparticle aggregate size on gene expression. Int J Mol Sci 2010;11:2383–92. [27] Nishikawa M, Takakura Y, Hashida M. Pharmacokinetics of plasmid DNAbased non-viral gene medicine. Adv Genet 2005;53:47–68. [28] Chen QR, Zhang L, Stass SA, Mixson AJ. Co-polymer of histidine and lysine markedly enhances transfection efficiency of liposomes. Gene Ther 2000;7:1698–705. [29] Chen CC, Kuo PL, Cheng YC. Spherical aggregates composed of gold nanoparticles. Nanotechnology 2009;20:055603. [30] Gourishankar A, Shukla S, Ganesh KN, Sastry M. Isothermal titration calorimetry studies on the binding of DNA bases and PNA base monomers to gold nanoparticles. J Am Chem Soc 2004;126:13186–7. [31] Hooper C, Pinteaux-Jones F, Fry VA, Sevastou IG, Baker D, Heales SJ, et al. Differential effects of albumin on microglia and macrophages: implications for neurodegeneration following blood–brain barrier damage. J Neurochem 2009;109:694–705.