Polyethylenimine with acid-labile linkages as a biodegradable gene carrier

www.elsevier.com/locate/jconrel

Polyethylenimine with acid-labile linkages as a biodegradable gene carrier Young Heui Kima, Jeong Hyun Parka,b, Minhyung Leec, Yong-Hee Kima, Tae Gwan Parkd, Sung Wan Kima,* a

Center for Controlled Chemical Delivery, University of Utah, Pharmaceutics and Pharmaceutical Chemistry, 30S 2000 E, RM 201, Salt Lake City, UT 84112-5820, USA b Division of Endocrinology and Metabolism, Department of Internal Medicine, Pusan Paik Hospital, College of Medicine, Inje University, Busan, South Korea c Clinical Research Center, College of Medicine, Inha University, Inchon 400-711, South Korea d Department of Biological Science, Korea Advanced Institute of Science and Technology, Taejon 305-701, South Korea Received 9 August 2004; accepted 4 November 2004 Available online 15 December 2004

Abstract Polyethylenimine (PEI) is a gene carrier with high transfection efficiency. However, PEI has high cytotoxicity, which depends on its molecular weight. To reduce the cytotoxicity, degradable PEIs with acid-labile imine linkers were synthesized with low molecular weight PEI1.8K (1.8 kDa) and glutadialdehyde. The molecular weights of the synthesized acid-labile PEIs were 23.7 and 13 kDa, respectively. The half-life of the acid-labile PEI was 1.1 h at pH 4.5 and 118 h at pH 7.4, suggesting that the acid-labile PEI may be rapidly degraded into nontoxic low molecular weight PEI in acidic endosome. In a gel retardation assay, plasmid DNA (pDNA) was completely retarded at a 3:1 N/P (nitrogen of polymer/phosphate of DNA) ratio. The zeta potential of the polyplexes was in the range of 46.1 to 50.9 mV and the particle size was in the range of 131.8 to 164.6 nm. In vitro transfection assay showed that the transfection efficiency of the acid-labile PEIs was comparable to that of PEI25K. In toxicity assay, the acid-labile PEI was much less toxic than PEI25K, due to the degradation of acid-labile linkage. Therefore, the acid-labile PEIs may be useful for the development of a nontoxic polymeric gene carrier. D 2004 Elsevier B.V. All rights reserved. Keywords: Acid-labile polymer; Cytotoxicity; Gene delivery; Polyethylenimine (PEI); Transfection

1. Introduction

* Corresponding author. Tel.: +1 801 581 6834; fax: +1 801 581 7848. E-mail address: [email protected] (S.W. Kim). 0168-3659/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2004.11.008

Gene therapy is of great interest since their studies can improve not only the treatment of diseases with genetic defects, but also the treatment and prevention of acquired diseases such as cardiovascular disease

GENE DELIVERY

Journal of Controlled Release 103 (2005) 209 – 219

GENE DELIVERY

210

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

and diabetes. However, naked therapeutic genes are so rapidly degraded by nucleases and showed poor cellular uptake that development of a safe and efficient gene carrier is one of the prerequisites for the success of gene therapy. Nonviral vectors for gene therapy have drawn much attention for its safety as well as ease of manufacturing and handling [1,2]. Among the cationic polymers, polyethylenimine (PEI), a commercially available cationic polyamine first introduced by Boussif et al. [3], is one of the most successful and widely studied gene delivery polymers [4]. PEI effectively buffers the endosomal environment due to the so-called dproton sponge effectT, thus facilitating endosomal escape to the cytoplasm [3]. In addition, due to its high transfection efficiency, branched PEI25K (25 kDa) has been used as a standard reference when it compared with other newly designed polymers [5]. A variety of PEIs have been studied for their transfection efficiency and several groups have reported that PEI has high cytotoxicity in many cell lines. Transfection efficiency and cytotoxicity of PEI depends on the molecular weight and it is generally accepted that PEI with a higher molecular weight (i.e. 25 kDa) shows high transfection efficiency and cytotoxicity. On the other hand, PEI with a lower molecular weight (i.e. 1.8 kDa) shows low transfection and cytotoxicity [6,7]. Several biodegradable PEIs have been investigated as gene carriers to reduce the cytotoxicity due to high molecular weight [8–10]. The major shortcoming of these polymers is mainly due to their ester-linkages as cross-linkers, providing undesirably long half-life. It was hypothesized that high molecular weight polymer composed of cross-linked PEI1.8K with acid-labile linkages would show transfection efficiencies comparable to that of high molecular weight PEI such as PEI25K. Furthermore, when the transfection complexes were exposed to low endosomal pH (~4–5), the acid-labile PEI would break down into low molecular weight counterparts, thus decreasing cytotoxicity. In this study, acid-labile PEIs with imine linkages was synthesized and their transfection efficiencies to 293T and A7R5 cell lines were evaluated. These acid-labile PEIs showed negligible cytotoxicity in both cell lines, suggesting their safety in the application to nonviral gene delivery.

2. Materials and methods 2.1. Materials PEI25K and chemicals were purchased from Aldrich (Milwaukee, WI, USA). PEI1.8K and PEI10K were purchased from Polysciences (Milwaukee, WI, USA). RQ1 RNase-free DNase I was purchased from Promega (Madison, WI). Ethidium bromide was purchased from ISC Bioexpress (Kaysville, UT). Medium and heat-inactivated fetal bovine serum were purchased from Hyclone (Logan, UT). Albumin standard and BCA protein assay kit were obtained from Pierce (Rockford, IL). The Endofree Maxi Plasmid Purification Kit was purchased from QIAGEN (Valencia, CA). Spectra/Por Regenerated Cellulose membranes (MWCO 2000) were purchased from Spectrum Laboratories (Rancho Dominquez, CA). Chemicals were purchased from Sigma (St. Louis, MO). Solvents were purified in common methods [11]. Other commercially available reagent chemicals were used as received. 2.2. Synthesis of polymers In a two-necked flask (250 or 500 ml) equipped with a dropping funnel, a measured amount of PEI1.8K was introduced in a measured amount of anhydrous CH2Cl2 (0.05 M) and stirred vigorously to dissolve. After PEI became clearly dissolved in the reaction solution, a measured amount of glutadialdehyde solution in anhydrous CH2Cl2 (0.005 M) was dropwise added through the dropping funnel with vigorous stirring over 2 h at 25 8C. After stirring for another 4 h, the solution was evaporated to remove the solvent. The viscous residue was dissolved again in double-distilled water and dialyzed through a cellulose membrane of molecular weight cut off of 2000 for 24 h. Lyophilization over 2 days gave yellowish products as solid or high viscous liquid. 2.3. Molecular weight determination and degradation study Capillary viscosity measurements were carried out to determine molecular weight and study degradation profile [9,12]. Polymers were dissolved in 0.5 M

sodium nitrate (NaNO3) aqueous solution to concentrations from 0.5 to 2.0 g/dl. Viscosity measurements were carried out using a Cannon-Fenske capillary viscometer (Cannon, Viscometer No. 75 N956) at 25 8C. The molecular weight of cross-linked polymers was determined using the following relationship (known as Mark-Houwink-Sakurada equation) by comparison with unconjugated polyethylenimine standard: ½g ¼ Kv M a where, K and a are Mark-Houwink parameters. For degradation study, polymers were dissolved in water to concentration of 1.0 g/dl and pH was adjusted to 4.5, 5,4 and 7.4. The polymer solution in viscometer was incubated in water bath at 37 8C. Decreasing viscosities were plotted as a function of time. The half-life of polymer in each pH condition was calculated on the assumption of 1st-order exponential decay. 2.4. Particle size and zeta potential measurement Polymer/pCMV-Luc complexes were measured at several N/P ratios for particle size and zeta potential on a MALVERN Instruments (Worcestershire, UK) Zetasizer 3000HSA. Zeta potential measurements were carried out in a standard capillary electrophoresis cell at 25 8C. The sampling time was set to automatic. Average values were calculated and reported for particle size as effective mean diameter (n=8). 2.5. Gel retardation assay Plasmid DNA (pDNA) condensation by acidlabile PEI conjugates was evaluated by a gel retardation assay. Acid-labile PEI/pCMV-Luc complexes were prepared at various N/P ratio (nitrogen atoms of PEI/phosphate of pDNA) and left for 20 min at room temperature for complex formation. The acid-labile PEI/pCMV-Luc complexes were electrophoresed on a 1% (w/v) agarose gel pretreated with 0.5 mg/ml ethidium bromide in 1 Tris-boric acid-EDTA (TBE) buffer at 80 V. The gel was analyzed on a UV illuminator to show the location of DNA.

211

2.6. DNase I protection assay It was reported previously that PEI1.8K demonstrated maximum transfection efficiency at a 40:1 N/P ratio [13]. PEI25 K had optimum transfection efficiency at a 5:1 N/P ratio [13]. Therefore, PEI25K/pDNA complex and acid-labile PEI/pDNA complex were formed at 5:1 and 40:1 N/P ratios, respectively. DNase I protection assay was carried out as described previously [14]. Ten micrograms of pCMV-Luc was mixed with 52 Ag of acid-labile PEI in 500 Al of PBS (pH 7.3). After complex formation, DNase I (12 Units; Promega) was added to the complex solution and the reaction mixture was incubated at 37 8C. One hundred microliters of the sample was taken at 0, 30, 60, 90 and 120 min after incubation, mixed with 20 Al of stop solution (0.4 M NaCl, 80 mM EDTA, and 2% SDS) and placed on ice. To dissociate the pDNA from acid-labile PEI, 80 Al of heparin solution (2 mg/ml) was added and the mixtures were incubated at 65 8C overnight [15]. Twenty microliters of the solution was characterized by 1% agarose gel electrophoresis. The polymer concentrations were calculated from the desired N/P ratio and the amount of plasmid, assuming that 43.1 g/ mol corresponded to each repeating unit of PEI containing one nitrogen atom and 330 g/mol corresponded to each repeating unit of DNA containing one phosphorus atom. 2.7. Transfection assay 293 Tcells and A7R5 cells were seeded separately in six-well tissue culture plates at 2105 and 9104 cells per well, respectively, in 10% FBS containing DMEM media. The plates were incubated at 37 8C and humidified 5% CO2 until cell confluency reached ~70% after which they were transfected with acidlabile PEI/pCMV-Luc complexes prepared at different N/P ratios ranging from 1:1 to 50:1. The total amount of plasmid DNA loaded was maintained constant at 2.0 Ag/well and transfection was carried out in the absence of serum. The cells were allowed to incubate at 37 8C in the presence of complexes for 4 h in CO2 incubator followed by replacement of 2 ml of DMEM containing 10% FBS. Thereafter, the cells were incubated at 37 8C for an additional 44 h. Cells were lysed using 1 lysis buffer after washing with PBS.

GENE DELIVERY

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

The luciferase activity was monitored and integrated over a period of 30 s. The final values of luciferase were reported in terms of RLU/mg total protein. In all the above experiments, both naked DNA as well as untreated cultures were used as positive and negative controls, respectively. 2.8. Cytotoxicity assay A Cell Counting Kit-8 (CCK-8) (Dojindo Molecular Technologies, Gaithersburg, MD) was used to evaluate the cytotoxicity of the acid-labile PEI/ pCMV-Luc complexes. Five thousand cells (A7R5) were seeded on a 96-well plate with DMEM+10% FBS and incubated at 37 8C and humidified 5% CO2 until confluency reached ~70%. Acid-labile PEI/ pCMV-Luc complexes were prepared at various N/P ratios ranging from 5:1 to 50:1. A7R5 cells were transfected with 0.2 Ag of pCMV-Luc, in the absence of 10% FBS for 4 h after which the media was changed, and transfection proceeded for an additional 44 h in the presence of 10% FBS. Ten microliters of thawed CCK-8 solution was added to each well. Plates were incubated for 4 h at the same incubator conditions after which the absorbance was read at 450

Fig. 1. Reaction scheme for the copolymerization of PEI and glutadialdehyde.

46.077

51.037

71.287

77.849

41.496

49.688 48.115

39.880

55.1.69 54.171 53.511

Total protein assays were carried out using BCA protein assay kit. Luciferase activity was measured in terms of relative light units (RLU) using 96-well plate Luminometer (Dynex Technologies, Chantilly, VA).

167.136

GENE DELIVERY

212

160

150

140

130

Fig. 2.

13

120

110

100

90

80

70

C NMR spectra of acid-labile PEI (No. 3).

60

50

ppm

213

nm with a reference wavelength of 600 nm. Cell viability was calculated as

Table 2 Relative ratio of different amino functions in PEI calculated from 13 C NMR analysis


Cell viability ð%Þ¼ OD450ðsampleÞ =OD450ðcontrolÞ 100

Polymer

Amines (%) Primary (18)

Secondary (28)

Tertiary (38)

28/38

where OD450(sample) is the absorbance at 450 nm of the transfected cells and OD450(control) is the absorbance at 450 nm of the negative control (nontransfected cells).

PEI PEI PEI PEI PEI

35 31 43 41 33

37 39 24 28 37

28 30 33 31 30

1.32 1.30 0.56 0.90 1.23

3. Results and discussion

and reaction temperature (Table 1). The results in Table 1 show that low concentration was required to obtain a water-soluble copolymer and the molecular weight increased. Harpe et al. assigned each signals in the 13C NMR spectrum and introduced a new method to calculate the ratio of different amino groups in the PEI [12,16]. Relative ratio of different amino functions in synthesized copolymer was determined with the same method. Inversely gated decoupling pulse sequence was used in 13C NMR experiments to avoid the influence of the Nuclear Overhauser effect on the signal intensities [12]. The results are shown in Table 2. From the integral results of secondary (28) to tertiary (38) amines, the relative ratio of linear to branched structure was determined. The results in Table 2 show that synthesized polymer (No. 3 or No. 4) has a higher degree of branching than starting low molecular weight PEI 1.8K.

Biodegradable polyethylenimines having imine linkages as acid-labile linkers were synthesized through the reaction of low molecular weight PEI 1.8 K and glutadialdehyde (Fig. 1). In the 1H NMR spectra at 400 MHz, all PEI CH2 signals resonate between 2.5 and 2.7 ppm (data not shown). Unfortunately, methylene groups with different amine substituents cannot be sufficiently assigned for quantitative analysis. Following synthesis and purification, conjugation was confirmed by the chemical shift change of the peak for C from 197.3 to 167.1 ppm due to the disappearance of aldehyde carbonyl peak (CHMO) at 197.3 ppm and the appearance of signals for the imine carbon (CMNH) at 167.1 ppm in the 13C NMR spectrum (Fig. 2). Because the structure of PEI has many primary amino groups, the reaction between PEI and glutadialdehyde generally resulted in an insoluble crosslinked polymer. It was possible to make water-soluble copolymer by several careful controls over reaction conditions such as molar ratios of reacting materials, concentration of the reaction mixture, rate of addition, Table 1 Copolymerization results of PEI and glutadialdehyde No. Initial concentration (mol/l)

1 2 3 4 5

Dialdehyde

PEI Dialdehyde 1.8k

0.01 0.01 0.05 0.05 0.05

0.04 0.02 0.005 0.005 0.005

1 1 2 4 10

4 2 1 1 1

5000 4500 4000 3500

pH 7.4 pH 4.5

3000

pH 5.4

2500

Initial molar ratio Solubility M v in water

PEI 1.8k

5500

Molecular weight

3.1. Synthesis of polymers

1.8k 25k (No. 3) (No. 4) (No. 5)

2000 1500

No No Yes Yes Yes

n.a. n.a. 23,000 13,000 6800

0

10

20

30

40

50

time (h) Fig. 3. Degradation of acid-labile PEI (No. 5). Decreasing molecular weights due to hydrolysis of polymer as a function of time at pH 7.4 (n), pH 5.4 (4), and pH 4.5 (.).

GENE DELIVERY

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

GENE DELIVERY

214

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

Cont.

1

2

3

4

5

molecular interactions of the charges bound to the polymer backbone [17]. The measured molecular weights of acid-cleavable PEI were 23,000, 13,000 and 6800 (Table 1). The degradation of gene carrier is important to reduce the cytotoxicity of the carrier. The commonly used branched PEI25K is known to be cytotoxic in many cell lines. Also, accumulation of nonbiodegradable PEI may be problematic in vivo, since there is neither a known degradation pathway nor an excretion mechanism. However, acid-labile PEIs will be less cytotoxic, since they are degraded into nontoxic low molecular weight PEI. The degradation of the acidlabile PEI (No. 5) was measured at pH 7.4, 5.4 and 4.5. Decreasing viscosities were plotted as a function of time and then converted to corresponding molecular weights (Fig. 3). The half-life of the polymer in each pH condition was calculated on the assumption of 1st-order exponential decay. In acidic condition, presumable relative to endosomal pH, hydrolysis was rapid with a half-life of 1.1 h (at pH 4.5) and 2.5 h (at pH 5.4), whereas at pH 7.4, the half-life was 118 h. The half-life of imine linkages was shorter than those of ester linkage which was previously reported [9]. These rapid degradations may lower the cytotoxicity to the cells.

A.

B.

Fig. 4. Gel retardation assay. Acid-labile PEI/pDNA complexes were prepared at various N/P ratios. The complexes were analyzed by agarose gel electrophoresis: (a) polymer (No. 3) and (b) polymer (No. 4). Cont indicates pDNA in the absence of polymer. Lane numbers indicate N/P ratios.

3.2. Degradation of polymers Although size exclusion chromatography (SEC) is often used for nonionic polymers, SEC is considered to be problematic for polyelectrolytes, due to their interactions with column packing materials and lack of suitable calibration standards. Other methods such as MALDI-MASS and vapor pressure osmometry (VPO) are also not suitable for polyelectrolytes. In order to determine molecular weight and study the degradation of the polymer, capillary viscosity measurements were used as described previously [12]. An abnormal behavior of polyelectrolyte solution, socalled dpolyelectrolyte behaviorT was overcome by adding salt in sufficient concentrations to the polymer solution. This behavior is typically due to intra-

3.3. Characterization of polymer complex Gel retardation assay was performed to investigate whether the synthesized acid-labile PEIs (Nos. 3 and

Acid-labile PEI (No. 3) /pDNA complex

PEI25K/pDNA complex DNase I incubation time (min)

0

30

60

90

120

0

30

60

90

120

1

2

3

4

5

6

7

8

9

10

Fig. 5. DNase I protection assay. Acid-labile PEI/pDNA complexes were prepared as described in Materials and methods. The complex solution was incubated with DNase I for 30, 60, 90 and 120 min. After incubation, DNA was analyzed by 1% (w/v) agarose gel electrophoresis.

Particle size (nm)

A.

200

150

100 PEI 25 kDa PEI 1.8 kDa Polymer (No.3) Polymer (No.4)

50

0 0

10

30

20

40

N/P ratio

B. 50

Zeta potential (mV)

4) form complexes with pCMV-Luc. The band of pCMV-Luc was retarded as the amount of acid-labile PEI increased, suggesting that acid-labile PEI forms a complex with pCMV-Luc (Fig. 4). pCMV-Luc was completely retarded at a 3:1 N/P and higher N/P ratio. This result indicates that DNA forms complex with acid-labile PEI at a 3:1 N/P ratio and that the negative charges of DNA are completely shielded by the polymer. DNase I protection assay was carried out to confirm that acid-labile PEI protected DNA from DNase I. After complex formation, acid-labile PEI/ pDNA complex was incubated with DNase I for 30, 60, 90 and 120 min. Naked pDNA was completely degraded by DNase I after 20 min of incubation (data not shown). However, the synthesized polymer (No. 3) successfully protected pDNA for over 120 min (Fig. 5, lanes 6–10). The results were quite similar to those of PEI25K (Fig. 5, lanes 1–5). The particle size distribution of acid-labile PEI/ pDNA complexes was determined by dynamic light scattering (DLS). The mean particle size of the complexes was in the range of 131.8 to 164.6 nm and remained relatively constant over the range of 5:1–40:1 N/P ratios (Fig. 6A). Zeta potential of acidlabile PEI/pDNA complexes prepared at different N/P ratios was in the range of 46.1 to 50.9 mV and remained relatively constant also over those ranges (Fig. 6B). The zeta potentials of the synthesized polymers were quite similar to that of PEI25K. However, the zeta potentials of acid-labile PEIs were different from that of PEI1.8K.

215

40

30

20 PEI 25 kDa PEI 1.8 kDa Polymer (No.3) Polymer (No.4)

10

0 0

10

20

30

40

N/P ratio Fig. 6. (A) Particle size of acid-labile PEI/pCMV-Luc complexes depending on N/P ratio. (B) Zeta potential of acid-labile PEI/pCMVLuc complexes depending on N/P ratio. The data were expressed as mean values (Fstandard deviations) of four experiments.

3.4. In vitro transfection To evaluate the transfection efficiency of acidlabile PEIs, acid-labile PEI/pCMV-Luc complexes were formulated at various N/P ratios in 5% (w/v) glucose solution and transfected to 293T or A7R5 cells. In 293T cells, the transfection efficiency increased with the N/P ratios and showed an almost saturated level at a 50:1 N/P ratio (Fig. 7A). In A7R5 cells, transfection efficiency of acid-labile PEI (No. 3) was saturated at a 20;1 N/P ratio (Fig. 7B), suggesting that the transfection efficiency of acid-labile PEI may be dependent on cell lines. Acid-labile PEIs were compared with PEI25K and PEI1.8K in terms of transfection efficiency. Acid-

labile PEIs (Nos. 3 and 4) showed much higher transfection efficiency than PEI1.8K and slight lower transfection efficiency than PEI 25K (Fig. 8). 3.5. Cytotoxicity assay To evaluate the cytotoxicity of the acid-labile PEIs, acid-labile PEI/pCMV-Luc complexes were transfected to A7R5 cells. The cytotoxicity of the acid-labile PEI was compared with that of PEI25K or PEI1.8K. Polymer/pDNA complexes were prepared at their optimum transfection conditions. A Cell Counting Kit-8 (CCK-8) was used to evaluate the cytotoxicity of the PEI/pCMV-Luc complexes.

GENE DELIVERY

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

A. Luciferase activity (RLU/mg protein)

1.0E+09 1.0E+08 1.0E+07 1.0E+06 1.0E+05 1.0E+04 1.0E+03 1.0E+02 1/1

5/1

10/1

20/1

30/1

40/1

50/1

30/1

40/1

50/1

N/P ratios

B. 1.0E+06

Luciferase activity (RLU/mg protein)

GENE DELIVERY

216

1.0E+05

1.0E+04

1.0E+03

1.0E+02 1/1

5/1

10/1

20/1

N/P ratios Fig. 7. Effect of N/P ratio of acid-labile PEI (No. 3)/pCMV-Luc complex on transfection to 293T cells (A) and A7R5 cells (B). Acid-labile PEI (No. 3)/pCMV-Luc complexes were prepared at various N/P ratios. Transfection efficiency was measured by luciferase assay. The data were expressed as mean values (Fstandard deviations) of five experiments.

The results showed that acid-labile PEIs and PEI1.8K have low cytotoxicities compared to PEI25K (Fig. 9). Acid-labile PEIs and PEI1.8K showed above 90% of cell viability at a 40:1 N/P ratio. In addition, acid-labile PEIs (Nos. 3 and 4) showed above 85% of cell viability at a 50:1 N/P ratio. Low cytotoxicity is one of the requirements for in vivo application of a polymeric carrier. Low cytotoxicity of acid-labile PEI suggests that acid-

labile PEI is a safer gene carrier than PEI25K for in vivo gene delivery. Therefore, acid-labile PEI is effective in that it has characteristics of high molecular weight PEI in transfection and that of low molecular weight PEI in cytotoxicity. In nonviral gene delivery, most protocols generally use net cationic complexes with an excess of polymer to maintain equilibrium between the complexed and dissociated form in solution. God-

217

Luciferase activtiy (RLU/mg protein)

1.0E+08

1.0E+07

1.0E+06 PEI 25kDa

PEI 1.8kDa

No.4 (50/1)

No.3 (50/1)

Carriers Fig. 8. Transfection efficiency of acid-labile PEIs to 293T cells. Polymer/pCMV-Luc complexes were prepared as described in Materials and methods. Transfection efficiency of each complex was measured by luciferase assay. The data were expressed as mean values (Fstandard deviations) of five experiments.

100

Cell viability (%)

80

60

40

20

0 l

tro

n Co

5K I2 PE

I PE

1.

8K No

.

40 3(

/1

) No

.

50 3(

/1

)

1)

0/

No

.

4 4(

No

.

5 4(

0/

1)

Fig. 9. Cytotoxicity of acid-labile PEIs to 293T cells. Polymer/pCMV-Luc complexes were prepared as described in Materials and methods. Polymer/pCMV-Luc complexes were added to the cells and incubated for 4 h at 37 8C. After the incubation, the transfection mixture was replaced with 100 Al of fresh DMEM medium containing 10% FBS. The cells were incubated for an additional 44 h at 37 8C. After the incubation, cell viability was measured by CCK-8 assay. The data were expressed as mean values (Fstandard deviations) of four experiments.

GENE DELIVERY

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

GENE DELIVERY

218

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219

bey et al. [18] reported that there are at least two types of cytotoxicity associated with PEI-mediated cell transfection. One is an immediate toxicity associated with free PEI, while the other is a delayed toxicity associated with cellular processing of PEI/DNA complexes [19]. Free PEI binds on the outer surface of the plasma-membrane and precipitates in huge clusters adhering to the cell surface [20]. This effect destabilizes the plasma-membrane and induces the immediate toxicity. However, when PEI was complexed with DNA, the complex showed minimal morphological effects on cells and the immediate toxicity was decreased [19]. The delayed toxicity by PEI/DNA complex is closely related to the release of DNA from PEI [19]. When PEI separates from DNA, free PEI is restored. The free PEI interacts with cellular components and inhibits normal cellular process. Acid-labile PEI degrades rapidly in acidic endosome condition, producing less toxic low molecular weight PEI. After release of DNA, low molecular weight PEI produced from acid-labile PEI may have lower toxicity than nondegradable PEI. Rapid degradation of acid-labile PEI may decrease the delayed toxicity. The cytotoxicity assay was performed in the same method as the transfection assay. After the transfection, the cells were maintained for 48 h. The delayed toxicity is induced by the internalized polymer/DNA complex and starts at several hours after transfection [19]. The half-life of acid-labile PEI at 4.5 and 5.4 was in the range of 1–2 h. Therefore, the time in cytotoxicity assay was sufficient to get degradation of acid-labile PEI.

4. Conclusion Biodegradable polyethylenimine with imine linkages as acid-labile moieties were synthesized and investigated for pDNA delivery. The acid-labile PEIs formed complexes with pDNA at a 3:1 or higher N/P ratios and protected pDNA from DNase I over 2 h. The acid-labile PEIs were rapidly degraded in acidic conditions. Acid-labile PEIs showed close transfection efficiency to PEI25K, but much less toxicity due to the degradation of acid-labile linkage. Therefore, acid-labile PEIs may be useful for development of nontoxic polymeric gene carriers.

Acknowledgment The authors thank Jay Olsen for his assistance with NMR studies and we thank Dr. Dong Hoon Choi for assistance with the preparation of pCMV-Luc. This work was supported by Expression Genetics.

References [1] W.F. Anderson, Human gene therapy, Nature 392 (1998) 25 – 30. [2] S. Han, R.I. Mahato, Y.K. Sung, S.W. Kim, Development of biomaterials for gene therapy, Molec. Ther. 2 (4) (2000) 302 – 317. [3] O. Boussif, F. Lezoualc’h, M.A. Zanta, M.D. Mergny, D. Scherman, B. Demeneix, J.P. Behr, A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 7297 – 7301. [4] W.T. Godbey, K.K. Wu, A.G. Mikos, Poly(ethylenimine) and its role in gene delivery, J. Control Release 60 (1999) 149 – 160. [5] K. Kunath, A.V. Harpe, D. Fisher, H. Petersen, U. Bickel, K. Voigt, T. Kissel, Low-molecular-weight polyethylenimine as a non-viral vector for DNA delivery: comparison of physicochemical properties, transfection efficiency and in vivo distribution with high-molecular-weight polyethylenimine, J. Control Release 89 (2003) 113 – 125. [6] W.T. Godbey, K.K. Wu, A.G. Mikos, Size matters: molecular weight affects the efficiency of poly(ethylenimine) as a gene delivery vehicle, J. Biomed. Mater. Res. 45 (1999) 268 – 275. [7] D. Fischer, T. Bieber, Y. Li, H. Elsasser, T. Kissel, A novel non-viral vector for DNA delivery based on low molecular weight, branched polyethylenimine: effect of molecular weight on transfection efficiency and cytotoxicity, Pharm. Res. 16 (1999) 1273 – 1279. [8] C. Ahn, S.Y. Chae, Y.H. Bae, S.W. Kim, Biodegradable poly(ethyleneimine) for plasmid DNA delivery, J. Control Release 80 (2002) 273 – 282. [9] M.L. Forrest, J.T. Koerber, D.W. Pack, A degradable polyethylenimine derivative with low toxicity for highly efficient gene delivery, Bioconjug. Chem. 14 (2003) 934 – 940. [10] M.A. Gosselin, W. Guo, R.J. Lee, Efficient gene transfer using reversibly cross-linked low molecular weight polyethylenimine, Bioconjug. Chem. 12 (2001) 989 – 994. [11] W.L.F. Armarego, D.D. Perrin, Purification of Laboratory Chemicals, 4th edition, Butterworth-Heinemann, Jordan Hill, Oxford, 1996. [12] A.V. Harpe, H. Petersen, Y. Li, T. Kissel, Characterization of commercially available and synthesized polyethylenimines for gene delivery, J. Control Release 69 (2000) 309 – 322. [13] M. Lee, J. Rentz, S. Han, D.A. Bull, S.W. Kim, Water soluble lipopolymer as an efficient carrier for gene delivery to myocardium, Gene Ther. 10 (2003) 585 – 593.

[14] M. Lee, S. Han, K.S. Ko, S.W. Kim, Cell type specific and glucose responsive expression of interleukin-4 by using insulin promoter and water soluble lipopolymer, J. Control Release 75 (2001) 421 – 429. [15] I. Moret, J. Esteban Peris, V.M. Guillem, M. Benet, F. Revert, F. Dasi, A. Crespo, S.F. Alino, Stability of PEI-DNA and DOTAP-DNA complexes: effect of alkaline pH, heparin and serum, J. Control. Release 76 (2001) 169 – 181. [16] D. Fisher, A.V. Harpe, K. Kunath, H. Petersen, Y. Li, T. Kissel, Copolymer of ethylene imine and N-(2-Hydroxyethyl)-ethylene imine as tools to study effects of polymer structure on physicochemical and biological properties of DNA complexes, Bioconjug. Chem. 13 (2002) 1124 – 1133. [17] M.J. Han, S.M. Park, J.Y. Park, S.H. Yoon, Polynucleotide analogues: 3.1. Synthesis, characterization, and physicochem-

219

ical properties of poly(thymidy1ic acid) analogues, Macromolecules 25 (1992) 3534 – 3539. [18] W.T. Godbey, K.K. Wu, G.J. Hirasaki, A.G. Mikos, Improved packing of poly(ethyleneimine)/DNA complexes increases transfection efficiency, Gene Ther. 6 (1999) 1380 – 1388. [19] W.T. Godbey, K.K. Wu, A.G. Mikos, Poly(ethyleneimine)mediated gene delivery affects endothelial cell function and viability, Biomaterials 22 (2001) 471 – 480. [20] D. Fischer, T. Bieber, Y. Li, H.P. Elsasser, T. Kissel, A novel non-viral vector for DNA delivery based on low molecular weight, branched polyethylenimine: effect of molecular weight on transfection efficiency and cytotoxicity, Pharm. Res. 16 (1999) 1273 – 1279.

GENE DELIVERY

Y.H. Kim et al. / Journal of Controlled Release 103 (2005) 209–219