Bioconjugate Chem. 2005, 16, 1375−1381
1375
Structural Characterization and Buffering Capacity in Relation to the Transfection Efficiency of Biodegradable Polyurethane S-ja Tseng,†,‡ Shiue-cheng Tang,‡ Min-da Shau,*,† Yi-fang Zeng,† Jong-yuh Cherng,§ and Mei-fen Shih⊥ Department of Applied Chemistry, and Department of Pharmacy, Chia-Nan University of Pharmacy and Science, 60 Erh-Jen Rd., Sec 1, Jen-Te, Taiwan, R.O.C., Department of Chemical Engineering, National Tsing-Hua University, Hsinchu 300, Taiwan, R.O.C., and Department of Chemistry and Biochemistry, National Chung Cheng University, 168 University Rd., Min-Hsiung Chia-Yi, Taiwan, R.O.C. Received January 11, 2005; Revised Manuscript Received October 13, 2005
Inefficient release of polymer/DNA complexes from endocytic vesicles into the cytoplasm and the cytotoxic nature of cationic polymers are two of the primary causes of poor gene delivery. EGpolyurethane [poly(ethylene glycol)-PU, Poly 1], EGDM-polyurethane [poly(ethylene glycol), 2-(dimethylamino)ethylamine-PU, Poly 2], and MDEADM-polyurethane [N-methyldiethanolamine, 2-(dimethylamino)ethylamine-PU, Poly 3] were designed in this study to overcome these obstacles. The structural characteristics of polyurethanes and physicochemical properties of their formed complexes with DNA were determined to correlate their transfection efficiency. The results revealed that Poly 2 and Poly 3 could bind with plasmid DNA and yield positively charged complexes with a size required for transfection. Poly 3 showed the best in buffering capacity and its formed complexes with DNA could transfect COS-7 cells better than those of Poly 2 and Poly 1. This study reveals that the amine groups in the polymeric structure and the buffer capacity of a polymeric transfectant would affect its potential in DNA delivery. Also the size and binding properties of DNA and polymeric transfectants can be in correlation to the transfection efficiency of resulting DNA/polymer complexes.
INTRODUCTION
Gene therapy (the introduction of an extraneous gene into a cell with the aim of replacing a lost cellular utility or introducing a new functionality) requires a gene delivery system that is efficient and has no or low cytotoxic side effects. Several polycations have been proposed for the delivery of DNA into mammalian cells (1-8). These polycations not only can condense DNA into structures small enough to enter cells through endocytosis but also afford protection from nuclease degradation. It has been demonstrated that the presence of polycations mediates transfection, but they were also associated with a considerable degree of cytotoxicity (5, 7, 9). As a result, the successful design of polycations for a gene vector application generally requires a balance between transfection efficiency and short-term or longterm cytotoxicity. Polyurethane is a class of biodegradable polymer with urethane linkages in the backbone. It has been studied as a biomaterial in tissue engineering (10-12). Recently, we have first demonstrated that cationic polyurethane can introduce DNA into cells (13). Besides the evaluation of the buffer capacity of transfectants for better transfection efficiency, we also systemically investigated the correlations between the structure of polymers, the physicochemical characteristics of polymer/DNA complexes, and their transfection efficiency. To study these crucial factors of polymers and complexes efficient for * Corresponding author. Tel +886-6.2664911, ext 245. Fax +886-6.2667319. E-mail address:
[email protected]. † Department of Applied Chemistry, Chia-Nan University of Pharmacy and Science. ‡ National Tsing-Hua University. § National Chung Cheng University. ⊥ Department of Pharmacy, Chia-Nan University of Pharmacy and Science.
gene delivery, for example, size distribution of polyurethane/plasmid DNA complexes and structural effect of polyurethane, three polymers with different structures of the polyurethane (e.g., tertiary amine groups in side chain or backbone) were prepared in this article. This insight of correlation can shed light on the preferred in vitro properties of DNA-polymer particles to achieve optimum transfection efficiency. EXPERIMENTAL PROCEDURES
Materials. N-Methyldiethanolamine (MDEA), glutaraldehyde, 2-dimethylaminoethylamine (DMAE), and poly(ethylene glycol) (PEG, Mw ) 200), which were vacuum-dried before use, were obtained from Fluka. L-Lysine methyl ester diisocyanate (LDI) was from Kyowa Hakko Kogyo co. (Japan). Polyethylenimine (branched PEI, Mw ) 25 000), poly(L-lysine) (PLL, Mw ) 25 600), ethidium bromide, bovine serum albumin, 5-bromo-4chloro-3-indoyl-β-galacto-pyranoside (X-Gal), Triton X-100, dibutyltin dilaurate, and N-[2-hydroxyethyl] piperazineN′-[2-ethaneslfonic acid] (HEPES) were purchased from Sigma co. (USA). N-Methyl dibenzopyrazine methyl sulfate (electron-coupling reagent) and sodium (2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide) (XTT) were purchased from Roche co. (USA). KpnI was purchased from Promega co. (USA). Solvent N,N-dimethylformamide (DMF, Tedia co., USA) was dried over calcium hydride and distilled just before use. Polymer Characterization. The structures of polymers were characterized by nuclear magnetic resonance (NMR, Bruker AMX-400 spectrometer) and Fourier transform infrared (FT-IR, Mattson Galerxy Series 5000 spectroscopy). All of the chemical shifts in 1H NMR and 13 C NMR spectra were reported in parts per million (ppm). A 99.8% pure DMSO-d6 was used as the solvent to characterize polymers. IR spectra were recorded on a
10.1021/bc050005r CCC: $30.25 © 2005 American Chemical Society Published on Web 11/01/2005
1376 Bioconjugate Chem., Vol. 16, No. 6, 2005
spectrometer as KBr pellets. The molecular weights of polymers were determined by high-performance liquid chromatography/gel permeation chromatography analysis (GPC, Waters model LC-2410). THF was used as the eluent and polystyrene as the reference. The weight- and number-average molecular weights (Mw and Mn, respectively) were calibrated with standard polystyrene samples. The sample concentration in the THF was 8.0 mg/mL, and the flow rate was 1.0 mL/min. Preparation of pCMV-βgal and Cell Culture. The plasmid pCMV-LacZ (pCMV-βgal), which contained a CMV promoter to drive the β-galactosidase (LacZ) gene expression, was used (7, 9, 13, 14). The plasmid DNA was amplified in Escherichia coli (DH5R strain) and purified by column chromatography (Qiagen Plasmid Mega kit, Germany). The purified plasmid DNA was dissolved in tris(hydroxymethyl) methylamine-ethyldiaminetetraacetic acid (Tris-EDTA) buffer (pH 8.0) and determined by the ratio of UV absorbance at 260 nm. Agarose gel electrophoresis analysis using restriction enzymes showed that the plasmid was in supercoiled form. The same analysis further indicated that a size of 7.8 kb of DNA was visible by using a restriction enzyme. The cell line COS-7 (SV 40 virus transformed African green monkey cell line, ATCC CRL-1651) was cultured in Dulbecco’s modified Eagle’s medium (DMEM, GibcoBRL Co., Ltd.) with 4 mM L-glutamine adjusted to 10% fetal serum albumin, 1.5 g/L sodium bicarbonate, and 4.5 g/L glucose at 37 °C in a humidified atmosphere of 5% CO2. Polymer Synthesis. The LDI and MDEA or PEG with a NCO/OH molar ratio of 1.1/1 were mixed in anhydrous DMF solvent within the three-necked reaction flask under dry nitrogen purge and then heated to 70 °C to react for 6 h with 0.5 wt % dibutyltin dilaurate included as a catalyst. And then an excess amount of methanol was slowly added into the reaction mixture until no unreacted isocyanate groups were detected. The Poly 1 was precipitated in ethyl ether and then dried at 40 °C under vacuum. Poly 2 and Poly 3 were synthesized using the aminolysis reaction (13). The polymers were characterized by FT-IR, 1H NMR, and 13C NMR. EG-Polyurethane (Poly 1). 1H NMR(400 MHz, DMSOd6, ppm) δ: 1.3-1.8 (6H, -CH2CH2CH2-), 2.9 (2H, -CH2NHCOO-), 3.1 (1H, -CHNHCOO-), 3.3 (3H, -NHCOOCH3), 3.6 (-CH2- of PEG), 4.05 (3H, -COOCH3), 5.2 (1H, -CH2NHCOO-), 5.5 (1H, -CHNHCOO-). 13C NMR (400 MHz, DMSO-d6, ppm) δ: 22.3, 29.1, 31.4 (-CH2CH2CH2-), 40.5 (-CH2NHCOO-), 52.4 (-NHCOOCH3), 56.3 (-CHNHCOO-), 60.9(-COOCH3), 63.670.1 (-CH2- of PEG), 156.6 (-CH2NHCOO-), 158.1 (-CHNHCOO-), 173.8 (-COOCH3). EGDM-Polyurethane (Poly 2). 1H NMR(400 MHz, DMSO-d6, ppm) δ: 1.3-1.8 (6H, -CH2CH2CH2-), 2 (6H, -N(CH3)2), 2.5 (2H, -CH2CH2N(CH3)2), 2.9 (2H, -CH2NHCOO-), 3.1 (1H, -CHNHCOO-), 3.3 (3H, -NHCOOCH3), 3.5 (2H, -CH2NHCO-), 3.6 (-CH2- of PEG), 5.2 (1H, -CH2NHCOO), 5.5 (1H, -CHNHCOO-), 7.0 (1H, -NHCO-). 13C NMR (400 MHz, DMSO-d6, ppm) δ: 22.3, 29.1, 31.4 (-CH2CH2CH2-), 38.9 (-CH2NHCO-), 40.5 (-CH2NHCOO-), 45.8 (-N(CH3)2), 52.4 (-NHCOOCH3), 56.1 (-CH2CH2N(CH3)2), 56.3 (-CHNHCOO-), 63.6-70.1 (-CH2- of PEG), 156.6 (-CH2NHCOO-), 158.1 (-CHNHCOO-), 162.8 (-NHCO-). MDEADM-Polyurethane (Poly 3). 1H NMR(400 MHz, DMSO-d6, ppm) δ: 1.3-1.8 (6H, -CH2CH2CH2-), 2 (6H, -N(CH3)2), 2.3 (3H, -CH2CH2N(CH3)CH2CH2-), 2.5-2.8 (2H, -CH2CH2N(CH3)2; 4H, -CH2CH2N(CH3)CH2CH2), 2.9 (2H, -CH2NHCOO-), 3.1 (1H, -CHNHCOO-), 3.3
Tseng et al.
(3H, -NHCOOCH3), 3.5 (2H, -CH2NHCO-), 4.1 (4H, -CH2CH2N(CH3)CH2CH2-), 5.2 (1H, -CH2NHCOO-), 5.5 (1H, -CHNHCOO-), 7.0 (1H, -NHCO-). 13C NMR (400 MHz, DMSO-d6, ppm) δ: 22.3, 29.1, 31.4 (-CH2CH2CH2-), 38.9 (-CH2NHCO-), 40.5 (-CH2NHCOO-), 44.1 (-CH2CH2N(CH3)CH2CH2-), 45.8 (-N(CH3)2), 52.4 (-NHCOOCH3), 56.1 (-CH2CH2N(CH3)2), 56.3 (-CHNHCOO-), 56.8 (-CH2CH2N(CH3)CH2CH2-), 59.1 (-CH2CH2N(CH3)CH2CH2-), 156.6 (-CH2NHCOO-), 158.1 (-CHNHCOO-), 162.8 (-NHCO-). Acid-Base Titration. Acid-base titration was used to evaluate the buffering capacity of synthesized cationic polyurethanes. In this assay, 10 mg of polymer was dissolved in 10 mL of 150 mM NaCl, and then 100 µL of 1 N NaOH was added to the solution to adjust the pH to the alkaline range at 11.6. HCl (0.1 N) was used as the titratant to lower the pH to acidic conditions at around 2-2.5. Titration increment size ) 100 µL. Preparation and Characterization of Polymer/ Plasmid DNA Complexes. Formation of Polymer/ Plasmid DNA Complexes. Poly 1, 2, and 3 (10 mg/mL) were dissolved in the 20 mM HEPES buffer (pH 7.4), and serial dilutions were made in which the mass ratio of polymer/DNA (w/w) was from 1/2 to 300/1. Particle Size, ζ-Potential, and Atomic Force Microscopy (AFM) Measurements of Polymer/Plasmid DNA Complexes. The particle sizes and surface charges of the polymer/DNA complexes were determined by dynamic light scattering (Nicomp 380 system, USA) at 25 °C using a 5-mW He-Ne laser (λ ) 633 nm) as the incident beam at a scattering angle of 90° and electrophoretic mobility with a ζ-potential system (Nicomp Instrument, USA) at 25 °C. The image of the Poly 3/DNA complexes coated on the ruby mica surface was observed on a nanoscope II (Digital Instruments, Santa Barbara, CA) as described previously (13). Stability of the Polymer/DNA Complexes in the Presence of Bovine Serum Albumin. The stability studies of the complexes in the presence of bovine serum albumin were evaluated by agarose gel retardation. The complexes were prepared at a mass ratio of 150/1 for 30 min, and bovine serum albumin was added into the solution of complexes to make final albumin concentrations of 5, 25, and 50 mg/mL. The DNA released from complexes was determined after 12 h at room temperature and 37 °C. DNA Gel Retardation and Restriction Endonuclease Protection Assay. The polymer/DNA complexes were loaded into a 0.7% agarose gel containing ethidium bromide (0.3 µg/mL) in a tris-acetate-EDTA (TAE) buffer and electrophoresed at 100 V for 45 min. After electrophoresis, the DNA bands were visualized by UVirradiation. The MDEA-PU/DNA polyplexes at ratios of 1/1, 5/1, and 50/1 (w/w) were incubated with KpnI at a concentration of 10 U/µL at 37 °C for 90 min in the provided reaction buffer. After the restriction endonuclease digests, samples were analyzed by 0.7% agarose gel electrophoresis in the same manner as described above. Transfection Studies and β-Galactosidase Assay. The COS-7 cells were used to evaluate the transfection efficiency of the polymer/DNA complexes. The cells were seeded in a 96-well plate (1.0 × 104 cells per well) in complete DMEM and incubated for 24 h before transfection. The polymer/DNA complexes (volume 200 µL) then were added to the cells. The DNA concentration was kept constant at 5 µg/mL (1.0 µg/well), and the amounts of polymer/DNA were expressed as mass ratios. After 1 h at 37 °C, the transfection mixture was removed and replaced with complete DMEM for an additional 48 h.
Transfection Efficiency of Biodegradable Polyurethane
Bioconjugate Chem., Vol. 16, No. 6, 2005 1377
Figure 2. Acid-base titration profile of various polymers with 0.1 N HCl solution. Figure 1. The structures of Poly 1 (EG-polyurethane; EGPU), Poly 2 (EGDM-polyurethane; EGDM-PU), and Poly 3 (MDEADM-polyurethane; MDEADM-PU).
Then a series of tests were performed for reporter gene expression (β-galactosidase) by histochemical staining with X-Gal, essentially as described previously (7, 9, 13, 15). Determination of the Cytotoxic Effect of Polymers and Polymer/DNA Complexes. Cytotoxicity of Poly 1, Poly 2, and Poly 3 in comparison with a commonly used gene carrier, PEI, was evaluated using the XTT assay (7, 9, 13, 16). In a 96-well plate, COS-7 cells were cultured in complete DMEM then seeded at a density of 1.0 × 104 cells/well. The cells were incubated at 37 °C and 5% CO2 in humidified atmosphere for 24 h. Subsequently, the cells were incubated for 1 h in 200 mL of FBS-free DMEM containing a polymer only (Poly 1, Poly 2, Poly 3, and PEI) or complexes at various concentrations. The cells incubated in DMEM were only for a negative control. After 1 h, the cells were washed with 200 µL of PBS solution and replaced by complete DMEM. After 48 h incubation, 50 µL of XTT labeling mixture was added to each well and incubated at 37 °C for 1 h. Results were expressed as the relatively cell viability (%) with respect to control wells containing culture medium cells. RESULTS AND DISCUSSION
Structural Characterizations of Polymers. The structures of EG-polyurethane, EGDM-polyurethane, and MDEADM-polyurethane (Poly 1, Poly 2, and Poly 3, respectively) were shown in Figure 1. The resulting polyurethane (Poly 3) with tertiary amines (-CH2CH2N(CH3)CH2CH2-) and (-N(CH3)2) in the backbone and side chain was isolated in up to 90% yield with weight average molecular weight of 37 000 and a polydispersity of 1.8 relative to polystyrene standards in THF. Poly 2, with tertiary amines (-N(CH3)2) in the side chain, and Poly 1 showed that the weight-averaged molecular weights were 35 000 and 39 000 with a polydispersity of 1.7 and 1.8, respectively. The structures of Poly 1, Poly 2, and Poly 3 were confirmed by FT-IR, 1 H NMR, and 13C NMR. From the FT-IR spectra, the peaks at 1720 cm-1 (CdO stretching, urethane), 1651 cm-1 (CdO stretching, amide), 1553 cm-1 (N-H bending, amide), and 3320 cm-1 (N-H stretching, urethane) represent the absorptions of urethane links in Poly 1,
Figure 3. Sizes and ζ-potentials of polymer/pCMV-βgal complexes prepared at different mass ratios. Results are presented as mean ( SD (n ) 3).
Poly 2, and Poly 3 (data not shown). In the 1H NMR spectra of Poly 1, Poly 2, and Poly 3, two distinct peaks at 5.9 and 5.2 ppm were assigned to the NH protons derived from the R- and -amine groups of L-lysine respectively (13). Chemical shifts of 2.0 and 7.0 ppm, observed in the Poly 2 and Poly 3, derived from the methyl and amide protons of the pendant group, respectively. In 13C NMR spectra, there was an accompanied disappearance of chemical shift at 173.8 ppm, which represented the ester carbon of the pendent group in the Poly 1. The FT-IR, 1H NMR, and 13C NMR characterizations of the synthesized polymers provided clear evidence that Poly 1, Poly 2, and Poly 3 had been successfully synthesized. Buffering Capacity of Polymers. Titration studies were performed to determine the buffering capacities of the various polymers within the endosomal/lysosomal compartments of the cell (Figure 2). All of the polycation solutions had pH 11.5-11.8 after addition of 100 µL of 1.0 N NaOH. PEI showed the best buffering capacity with a wide pH range. Poly 1, containing no amine group in the structure, demonstrated a poor buffering range from pH 11.5 and 10. The buffering properties of the polymers were improved through the incorporation of tertiary amine groups into the backbone or side chain. Poly 2, bearing tertiary amines in the side chain, and Poly 3, with tertiary amines in the backbone and side chain, showed buffering capacities from pH 11.5-7.0 and 11.7-
1378 Bioconjugate Chem., Vol. 16, No. 6, 2005
Tseng et al.
Figure 4. The AFM images of Poly 3/pCMV-βgal complexes (w/w), 150/1. The height of the AFM image is represented by a graded brown-yellow scale with the yellow color indicating a height more than 10 nm above the ruby mica. The x and y dimensions are scaled as shown.
5.0, respectively. It was found that Poly 3 possessed excellent properties in DNA condensing and buffering abilities. Self-Assembly of Polymer with Plasmid DNA and Restriction Endonuclease Protection Assay of Polymer/DNA Complexes. Figure 3 shows the size and ζ-potential of the Poly 1/DNA complexes, Poly 2/DNA complexes, and Poly 3/DNA complexes at various mass ratios ranging from 1/2 to 300/1. The results show that the average diameter (100-110 nm) of the Poly 3/DNA complexes and the average diameter (170-180 nm) of the Poly 2/DNA complexes fall within the general size required for cellular endocytosis (
