Gene Therapy (2003) 10, & 2003 Nature Publishing Group All rights reserved /03 $25.00

Transcription

1 (23) 1, & 23 Nature Publishing Group All rights reserved /3 $25. RESEARCH ARTICLE Water-soluble lipopolymer as an efficient carrier for gene delivery to myocardium M Lee 1, J Rentz 2, S-O Han 1, DA Bull 2 and SW Kim 1 1 Department of Pharmaceutics and Pharmaceutical Chemistry, Center for Controlled Chemical Delivery, University of Utah, USA; and 2 Department of Surgery, Division of Cardiothoracic Surgery, University of Utah Health Sciences Center, UT, USA Water-soluble lipopolymer (WSLP), which consisted of polyethylenimine (, 18 Da) and cholesterol, was characterized as a gene carrier to smooth muscle cells and myocardium. Acid base titration showed that WSLP had a proton-buffering effect. The size of WSLP/plasmid DNA (pdna) complex was around 7 nm. WSLP/pDNA complex was transfected to A7R5 cells, a smooth muscle cell line. WSLP showed the highest transfection at a 4/1 N/P ratio. WSLP has higher transfection efficiency than (18 and 25 Da), SuperFect, and lipofectamine. In addition, WSLP has less cytotoxicity than (25 Da), SuperFect, and lipofectamine. Since WSLP has cholesterol moiety, it may utilize cellular cholesterol uptake pathway, in which lowdensity lipoprotein (LDL) is involved. An inhibition study with free cholesterol or low-density lipoprotein (LDL) showed that transfection was inhibited by cholesterol or LDL, suggesting that WSLP/pDNA complex is transfected to the cells through the cholesterol uptake pathway. To evaluate the transfection efficiency to myocardium, WSLP/pDNA complex was injected into the rabbit myocardium. WSLP showed higher transfection than and naked pdna. WSLP expressed the transgene for more than 2 weeks. In conclusion, WSLP is an efficient carrier for local gene transfection to myocardium, and useful in in vivo gene therapy. (23) 1, doi:1.138/sj.gt Keywords: cholesterol; gene delivery; myocardium; receptor-mediated endocytosis; smooth muscle cells; water-soluble lipopolymer Introduction Development of a safe and efficient gene carrier is one of the main requirements for success of gene therapy. Although viral-based gene delivery is currently the most effective way to transfer genes to cells, nonviral vectors are increasingly being considered for in vivo gene delivery. The advantages of nonviral gene therapy are lack of specific immunogenecity, simplicity of use, and ease of large-scale production. 1,2 In addition, the simple conjugation of a targeting moiety to nonviral gene carrier can facilitate tissue-targeting gene delivery To use it for cardiovascular disease gene therapy, we have developed a new gene carrier system, TerplexD- NA TerplexDNA is composed of plasmid DNA (pdna), stearyl-poly-l-lysine (stearyl-pll), and lowdensity lipoprotein (LDL). LDL binds to many types of cells such as vascular endothelial cells, vascular smooth muscle cells, hepatocytes, and macrophages, and initiates receptor-mediated endocytosis Therefore, LDL of TerplexDNA facilitates receptor-mediated endocytosis, and TerplexDNA showed high transfection efficiency to smooth muscle cells in vitro and myocardium in vivo. 14,18 Correspondence: Dr SW Kim, Department of Pharmaceutics and Pharmaceutical Chemistry, Center for Controlled Chemical Delivery, University of Utah, 3 S 2 East RM 21, Salt Lake City, UT , USA Received 22 April 22; accepted 15 October 22 Polyethylenimine () has been reported as an effective gene carrier because can deliver pdna to the cytoplasm via endosomes due to the proton-sponge effect. 22,23 However, high molecular weight (25 Da), which is commonly used for gene delivery, has high toxicity. 24 Although low molecular weight (18 Da) has low cytotoxicity, it also has low transfection efficiency. 23 To reduce cytotoxicity and enhance transfection efficiency of, we developed another gene carrier, water-soluble lipopolymer (WSLP), in which cholesterol was conjugated to (18 Da). 25 Owing to the balance between hydrophilicity of and hydrophobicity of cholesterol, WSLP can form small size micelles in an aqueous solution. WSLP has high transfection efficiency to 293T or CT26 cells with negligible cytotoxicity to the cells. In this research, WSLP was characterized in various methods as a gene carrier to smooth muscle cells in vitro and myocardium in vivo. Acid base titration was performed to confirm the proton-buffering effect of WSLP, and mean particle size of WSLP/pDNA complex was measured at various concentrations by dynamic light scattering (DLS). WSLP/pDNA complex was transfected to A7R5 cells, a smooth muscle cell line, to evaluate the transfection efficiency. The inhibition study with free cholesterol or LDL was carried out to prove that WSLP utilizes cellular cholesterol uptake pathway. Finally, the transfection efficiency to rabbit myocardium was evaluated. The results suggest that WSLP can be applied to in vivo gene therapy for the treatment of ischemic heart disease.

2 586 Results Physical characterization of WSLP WSLP was synthesized as described previously. 25 has been known to have a proton-buffering effect, resulting in disruption of endosome in the transfection process. WSLP has a moiety and may have a protonbuffering effect. To confirm the buffering effect of WSLP, acid base titration was performed (Figure 1). The acid base titration profile was obtained with PLL, (18 Da), and WSLP. The initial ph points of these polymers were close ranging from to When.1 N HCl was added to the polymer solutions from to 1. ml, there was no significant change in the ph of PLL,, and WSLP, probably because of the protonation of primary amine groups. However, the titration curve trend was significantly different after that point. Titration curve of PLL was nearly vertical, suggesting little buffering capacity of PLL. Unlike PLL, WSLP and showed considerable buffer capacity. The tertiary and secondary amine groups of WSLP and were protonated at the ph range from 9. to 4.. The data showed that the buffer capacity of WSLP was slightly lower than that of. The size of WSLP/pCMV-Luc was measured at various concentrations (Figure 2a). To formulate WSLP/pCMV-Luc complexes, WSLP was mixed with pcmv-luc at a 4/1 N/P (nitrogen atoms of polymer/ phosphate of pdna) ratio, because WSLP showed the highest transfection efficiency at this N/P ratio in the following in vitro study. The concentration of pcmv-luc was in the range from to 2 mg/ml. As a result, the complex size was around 7 nm at mg of pcmv- Luc/ml. The size was not changed up to 25 mg of pcmv-luc/ml. However, at 2 mg of pcmv-luc/ml, the size increased abruptly to nm, suggesting aggregation. This suggested that WSLP/pDNA complexes should be formulated below 25 mg/ml pdna concentration. Therefore, in the following experiments, 5 mg/ml pdna concentration was used for in vitro experiments and 2 mg/ml pdna concentration for in vivo experiments. ph Volume of.1 N HCl (ml) NaCl PLL WSLP Figure 1 Acid base titration. In total, 1 mg of WSLP, 18, or PLL was dissolved in 1 ml of 15 mm NaCl. In all, 1 ml of 1 N NaOH was added to the polymer solution. The polymer solution was titrated with increasing volume of.1 N HCl. a Particle size (nm) b Zeta potential (mv) Concentration of pdna (µg/ml) Concentration of pdna (µg/ml) Figure 2 (a) Size of WSLP/pCMV-Luc depending on the concentration. (b) z potential of WSLP/pCMV-Luc depending on the concentration. Zeta potential of WSLP/pCMV-Luc complexes was measured at various pcmv-luc concentrations (Figure 2b). N/P ratio for the complex was fixed at 4/1. z potential of WSLP/pCMV-Luc complex was not significantly variable, even though there was a slight decrease with increasing pcmv-luc concentration. This may be because of the increase in the particle size, which is likely to slow down the mobility of the WSLP/pCMV-Luc complex. In vitro transfection assay of WSLP to smooth muscle cells Previous reports showed that WSLP showed the highest transfection efficiency at a 2/1 N/P ratio to 293T cells or CT26 cells. 25 However, the N/P ratio that shows the highest transfection would be variable depending on cell type. WSLP/luciferase plasmid (pcmv-luc) complexes were prepared at various N/P ratios and transfected to A7R5 smooth muscle cells. Transfection of pcmv-luc to A7R5 cells was increased with increasing N/P ratio up to 4/1 (Figure 3). Above a 4/1 N/P ratio, the transfection was saturated. WSLP was compared with low molecular weight (18), high molecular weight (25 ), Super- Fect (starburst dendrimer), and lipofectamine in terms of transfection efficiency to A7R5 cells. Optimal N/P ratio for 18 has not been reported. Therefore, to determine an optimal N/P ratio for 18, 18/

3 2.5e+8 2.e+8 1.5e+8 1.e+8 5.e+7 1/1 15/1 2/1 25/1 3/1 35/1 4/1 45/1 5/1 N/P ratio Figure 3 Effect of N/P ratio of WSLP/pCMV-Luc complex on transfection to A7R5 cells. WSLP/pCMV-Luc complexes were prepared at various N/P ratios. Transfection efficiency was measured by luciferase assay. The highest transfection was obtained at a 4/1 N/P ratio. The data are expressed as mean values (7s.d.) of four experiments. Po.1 as compared with the N/P ratios of 1/1, 15/1, 2/1, 25/1, and 3/1, but no statistical significance as compared with the N/P ratios of 35/1, 45/1, and 5/1. 4.5e+8 4.e+8 3.5e+8 3.e+8 2.5e+8 2.e+8 1.5e+8 1.e+8 5.e+7. WSLP 25 Lipofect -amine SuperFect 18 Figure 4 Transfection efficiency of WSLP, 25, lipofectamine, SuperFect, or 18 to A7R5 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 are expressed as mean values (7s.d.) of four experiments. Po.1 as compared with lipofectamine, SuperFect, and 18, but no statistical significance as compared with pcmv-luc complexes were prepared at various N/P ratios. As a result, 18 showed the highest transfection efficiency at a 4/1 N/P ratio (data not shown). In addition, transfection condition of lipofectamine or SuperFect was optimized. The results showed that lipofectamine and SuperFect showed the highest transfection efficiency to A7R5 cells at a 5/1 weight ratio and 6 ml of SuperFect/mg DNA, respectively (data not shown). Therefore, WSLP/pCMV-Luc and 18/ pcmv-luc complexes were prepared at a 4/1 N/P ratio. Lipofectamine/pCMV-Luc and SuperFect/pCMV- Luc complexes were formulated at a 5/1 weight ratio and 6 ml of SuperFect/mg DNA, respectively. 25 / pcmv-luc complex was prepared at a 5/1 N/P ratio, based on the previous reports. 22,26 29 After transfection to A7R5 cells, transgene expression was evaluated by luciferase assay. As a result, WSLP showed higher transfection efficiency than 18, SuperFect, or lipofectamine to A7R5 cells (Figure 4). showed a similar transfection efficiency to WSLP. Cytotoxicity of WSLP To evaluate the cytotoxicity of WSLP, WSLP/pCMV-Luc complex was transfected to A7R5 cells. The cytotoxicity of WSLP was compared with that of 18, 25, lipofectamine, and SuperFect. Polymer/pDNA complexes were prepared at their optimal transfection condition as described above. The complexes were transfected to A7R5 cells, and the cytotoxicities of the complexes were evaluated by MTT assay. The results showed that WSLP and 18 had low cytotoxicity (Figure 5). WSLP and 18 showed about 9% cell viability after the transfection. However, 25, lipofectamine, and SuperFect were highly toxic to the cells (Figure 5). In vitro transfection assay: cholesterol or LDL inhibition Since WSLP contains cholesterol moiety, it is possible that the WSLP/pDNA complex is internalized by the Cell viability (%) WSLP 25 Lipofect -amine SuperFect 18 Figure 5 Cytotoxicity of WSLP, 25, lipofectamine, SuperFect, or 18 to A7R5 cells. A7R5 cells were seeded in 96-well microassay plates. 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 371C. After the incubation, the transfection mixture was replaced with 1 ml of fresh DMEM medium supplemented with 1% FBS. The cells were incubated for an additional 2 h at 371C. After the incubation, cell viability was measured by MTT assay. The data are expressed as mean values (7s.d.) of five experiments. Po.1 as compared with 25, lipofectamine, SuperFect, but no statistical significance as compared with 18. cells through a cellular cholesterol uptake pathway. To confirm this possibility, A7R5 cells were incubated with free cholesterol for 3 min before the transfection to saturate the cholesterol uptake pathway. After the incubation of cholesterol, WSLP/pCMV-Luc complex was added to the cells. The results showed that incubation with free cholesterol inhibited WSLPmediated transfection (Figure 6a). However, 18- mediated transfection was not inhibited by incubation with free cholesterol (Figure 6b). In addition, the cells were incubated with LDL for 3 min before the transfection. After the incubation of the cells with LDL, the transfection efficiency by WSLP was decreased (Figure 7a). However, LDL did not decrease the 18-

4 588 a b 3.5e+8 3.e+8 2.5e+8 2.e+8 1.5e+8 1.e+8 5.e+7 1.6e+7 1.4e+7 1.2e+7 1.e+7 8.e+6 6.e+6 4.e+6 2.e Cholestrol (µg/well) Cholestrol (µg/well) Figure 6 Inhibition of transfection by the incubation of cholesterol. A7R5 cells were seeded in six-well microassay plates. A7R5 cells were incubated with various amounts of cholesterol for 3 min before transfection. WSLP/ pcmv-luc complex (a) or 18/pCMV-Luc (b) was added to the cells. After the transfection, the cells were harvested and luciferase activity was measured. The data are expressed as mean values (7s.d.) of four experiments. Po.1 as compared with 4, 8, 16, and 32 mg of cholesterol. mediated transfection (Figure 7b). Therefore, it is suggested that WSLP can utilize a cellular cholesterol uptake pathway, which is related to LDL. In vivo transfection assay with WSLP in rabbit myocardium To evaluate the in vivo transfection efficiency of WSLP, WSLP/pCMV-Luc complex was injected to the left ventricle of a rabbit heart at various N/P ratios. The heart was harvested and luciferase activity measured 3 days after the injection. Unlike the in vitro transfection, WSLP showed the best transfection efficiency at a 1/1 N/P ratio in vivo (Figure 8). It may be because of the interaction of WSLP with serum proteins, and high N/P ratio may decrease the transfection efficiency. RT-PCR was performed to detect the expressed VEGF mrna. For RT-PCR, the primers were designed to encompass the intron to exclude the amplification of the contaminated pdna or the endogenous VEGF mrna. RT-PCR data showed that the VEGF mrna was expressed at the highest level in the 1/1 N/P ratio complex injected animals (Figure 9). Therefore, these a b 4.e+8 3.5e+8 3.e+8 2.5e+8 2.e+8 1.5e+8 1.e+8 5.e+7. 5e+6 4e+6 3e+6 2e+6 1e LDL (µg/well) LDL (µg/well) Figure 7 Inhibition of transfection by the incubation of LDL. A7R5 cells were seeded in six-well microassay plates. A7R5 cells were incubated with various amounts of LDL for 3 min before transfection. After the incubation, WSLP/pCMV-Luc complex (a) or 18/pCMV-Luc (b) was added to the cells. After the transfection, the cells were harvested and luciferase activity was measured. The data are expressed as mean values (7s.d.) of four experiments. Po.1 as compared with 4, 8, 16, and 32 mg of LDL. results confirmed that WSLP had the highest transfection efficiency to myocardium at a 1/1 N/P ratio. To compare the in vivo transfection efficiency of WSLP with that of 18, 25, and naked pdna, the polymer/pdna complexes or naked pdna were injected to rabbit myocardium. Since lipofectamine and SuperFect were highly cytotoxic to smooth muscle cells, they were excluded in in vivo experiments. WSLP showed higher transfection efficiency than 18, 25, and naked pdna (Figure 1). Persistence of gene expression by WSLP in rabbit myocardium To evaluate the persistence of gene expression, WSLP/ pcmv-luc complex was injected into rabbit myocardium at a 1/1 charge ratio. Naked pdna was injected into myocardium as a control. Luciferase activity was measured at 3, 7, or 14 days after injection (Figure 11). WSLP showed higher transgene expression than naked pdna. The luciferase activity by naked pdna

5 /1 1/1 N/P ratio Figure 8 Effect of N/P ratio of WSLP/pCMV-Luc complex on transfection to myocardium. WSLP/pCMV-Luc complexes were prepared at various N/ P ratios. WSLP/pCMV-Luc complex was injected to the left ventricles of the hearts of New Zealand white rabbits. Transgene expression was evaluated 3 days after the injection by luciferase assay. The highest transfection was obtained at a 1/1 N/P ratio. The data are expressed as mean values (7s.d.) of three experiments. Po.5 as compared with the N/P ratios of 5/1, 2/1, and 4/1. 2/1 4/ Naked DNA WSLP Figure 1 Transfection efficiency of WSLP, 18, 25, and naked pdna to rabbit myocardium. WSLP/pCMV-Luc complex and 18/pCMV-Luc complexes were prepared at a 1/1 N/P ratio. 25 /pcmv-luc complexes were formulated at a 5/1 N/P ratio. WSLP/pCMV-Luc, 18/pCMV-Luc, 25 /pcmv-luc complex, or naked pdna was injected to the left ventricles of the hearts of New Zealand white rabbits. Transgene expression was evaluated 3 days after the injection by luciferase assay. The data is expressed as mean values (7s.d.) of three experiments. Po.1 as compared with naked pdna, 18, and decreased rapidly and the activity reached background level at 7 days after the injection. WSLP/pCMV-Luc complex showed a longer gene expression profile than naked DNA. The WSLP/pCMV-Luc complex showed a significant level of the luciferase expression 14 days after the injection (Figure 11). Discussion NP ratio - 1/1 2/1 4/1 M bp 4 bp 3 bp Figure 9 RT-PCR of the VEGF mrna after myocardial injection. WSLP/ psg5-vegf complexes were prepared at various N/P ratios. WSLP/ pdna complex was injected to the left ventricles of the hearts of New Zealand white rabbits. Transgene expression was evaluated 3 days after the injection by RT-PCR. High molecular weight (25 ) has been reported to be effective for gene delivery, because it has a proton-buffering effect and facilitates endosomal escape. However, 25 has high cytotoxicity to cells and cannot be applied to a clinical trial. 24 In addition, low molecular weight (18) has relatively low cytotoxicity and low transfection efficiency. In this study, WSLP was synthesized by the conjugation of cholesterol to low molecular weight 18. Therefore, WSLP has the advantages of and cholesterol. First, WSLP has moiety and, therefore, it has the proton-buffering effect. To prove this, we performed an acid base titration (Figure 1). Although WSLP has a lower buffering effect than, the effect of WSLP is much higher than PLL Naked DNA WSLP/pDNA complex days Figure 11 Persistence of the transgene expression by WSLP in rabbit myocardium. WSLP/pCMV-Luc complex was at a 1/1 N/P ratio. WSLP/ pcmv-luc complex or naked pcmv-luc was injected to the left ventricles of the hearts of New Zealand white rabbits. Transgene expression was evaluated 3, 7, or 14 days after the injection by luciferase assay. The data are expressed as mean values (7s.d.) of three experiments. Po.1 as compared with naked DNA. This buffering effect may facilitate endosome escape, resulting in high transfection efficiency. Second, low molecular weight 18 has low cytotoxicity to cells. The cytotoxicity of WSLP was similar to that of 18 (Figure 5). Previously, starburst polyamidoamine dendrimers were evaluated as a gene carrier into murine cardiac grafts. 3 In addition, the transfection efficiency of the dendrimer was evaluated to smooth muscle cells and endothelial cells. 27 The dendrimer was effective in that it had higher transfection efficiency than naked pdna and prolonged gene expression. However, as proved in our study, the dendrimer (SuperFect) is highly toxic to smooth muscle cells, suggesting that WSLP has an advantage against the dendrimer in terms of safety.

6 59 Because of low toxicity, WSLP/pDNA complex can be administrated repeatedly. Third, the transfection efficiency of WSLP was enhanced by the cholesterol moiety. Cholesterol is taken up by the cells through receptormediated endocytosis or transfer from the lipoprotein to the exoplasmic leaflet of the plasma membrane bilayer. 31 In this study, the transfection efficiency of WSLP/pDNA complex was inhibited by free cholesterol or LDL. This suggests that WSLP utilizes the cellular cholesterol uptake pathway to internalize the complex in smooth muscle cells. Fourth, WSLP does not require co-lipid. To date, several cholesterol-based cationic lipids have been used for gene transfer. 3b[N,(N,N -dimethylaminoethane) carboneyl cholesterol (DC-chol) was synthesized by Gao and Huang. 32 DC-chol was mixed with colipid DOPE to form cationic liposome, which has high transfection efficiency to mammalian cells Spermine cholesteryl carbamate and spermidine cholesteryl carbamate have been shown to be more effective than DCchol. 35 Unlike these cholesterol based cationic lipids, WSLP has proton sponge effect of, and destabilizes endosomal membrane without the requirement of a colipid, resulting in high transfection efficiency. In the previous reports, WSLP had high transfection efficiency to human kidney 293T cells, CT26 colon carcinoma cells, and mouse insulinoma MIN6 cells. 25,36 In addition, WSLP can protect pdna from nuclease. 36 In this study, we evaluated WSLP as a gene carrier to smooth muscle cells in vitro and myocardium in vivo. The transfection efficiency of WSLP was higher than that of 18, lipofectamine, or SuperFect to A7R5 smooth muscle cells. In addition, WSLP has negligible cytotoxicity to A7R5 cells. The inhibition studies suggest that WSLP utilizes the cholesterol uptake pathway. In Figure 6a, the incubation with free cholesterol decreased the transfection efficiency of WSLP. The efficiency of myocardium injection is proved in Figures 8 1. Unlike the in vitro transfection, WSLP has the highest transfection efficiency to rabbit myocardium at a 1/1 N/P ratio in luciferase assay (Figure 8). In addition, RT-PCR data showed that WSLP had the highest transfection efficiency at a 1/1 N/P ratio (Figure 9). The difference between in vitro transfection and in vivo transfection may be because of the interaction of WSLP with serum proteins. At a higher N/P ratio, WSLP/pDNA complex has higher positive charge and may interact with serum proteins more strongly, resulting in the decrease of transfection efficiency. Unlike in vitro transfection, 25 showed similar transgene expression to 18 in myocardial transfection. This low gene expression may be because of the high cytotoxicity of 25. In addition, 25 forms a complex with pdna more tightly than 18, suggesting that the release of pdna in myocardium may be delayed compared to 18. The delay of the pdna release from the complex may decrease transfection efficiency. As shown in Figure 8, WSLP had lower gene expression at a 2/1 or 4/1 N/P ratio than 5/1 or 1/1 N/P ratio. Besides serum protein effect, the release of pdna from the complex may be delayed at a high N/P ratio, resulting in the decrease of the transfection efficiency. At a 1/1 N/P ratio, WSLP has higher transfection efficiency than 18, 25, and naked pdna (Figure 1). The higher transfection efficiency of WSLP to myocardium may be related to the interaction with a myocardial LDL receptor. In addition, WSLP expressed the transgene longer than naked pdna (Figure 11). WSLP/pDNA complex expressed the transgene for more than 14 days in myocardium. The longer gene expression by WSLP may be because of the prolonged half-life of pdna, since WSLP protects pdna from nuclease. In summary, WSLP has high transfection efficiency and low cytotoxicity to smooth muscle cells in vitro. WSLP may use cellular cholesterol uptake pathway to internalize the complex into the cells. In addition, the injection of WSLP/pDNA to rabbit myocardium also showed that WSLP had higher transfection efficiency than (18 Da) and longer gene expression than naked pdna. Therefore, WSLP is an efficient carrier to myocardium. In addition, WSLP s characteristics of low toxicity and high transfection efficiency may allow for the application of in vivo gene therapy in the treatment of heart disease. Materials and methods Synthesis of WSLP WSLP was synthesized as described previously. 25 Briefly, (18 Da) was stirred on ice in a mixture of anhydrous methylene chloride and triethylamine for 3 min. Cholesteryl chloroformate was dissolved in methylene chloride and slowly added to the solution. The mixture was stirred for 12 h on ice and the resulting product was dried on a rotary evaporator. The powder was dissolved in.1 N HCl. The aqueous solution was extracted with methylene chloride and then filtered through a glass microfiber filter. The product was concentrated by solvent evaporation, precipitated with large excess acetone, and dried under vacuum. Acid base titration The ability of WSLP to protonate and obtain a positive charge was determined by acid base titration. 37,38 Briefly, 1 mg of WSLP, (18 Da), or PLL was dissolved in 1 ml of 15 mm NaCl. A total of 1 N NaOH was added to this solution and the ph was measured. The polymer solution was titrated with increasing volume of.1 N HCl and the ph was measured at every point. Preparation of plasmid pcmv-luc was constructed by inserting the HindIII/ XbaI firefly luciferase cdna fragment from pgl3- control vector into the HindIII/XbaI site of pcdna3. 25 psg5-vegf was kindly provided by Jozef Dulak (Department of Cardiology, University of Innsbrook, Innsbrook, Austria). The pdnas were transformed in Escherichia coli DH5a and amplified in terrific broth media at 371C overnight at 225 rpm. The amplified plasmid DNA was purified by using the Maxi plasmid purification kit (Qiagen, Valencia, CA, USA). Purified plasmid DNA was dissolved in Tris-EDTA (TE) buffer, and its purity and concentration were determined by ultraviolet (UV) absorbance at 26 nm. The optical density ratios at nm of these plasmid preparations were in the range of To confirm that there is no rearrangement of the gene during cloning and propagation of the plasmid DNA, restriction enzyme assay was done followed by 1% agarose gel electrophoresis.

7 Particle size and z potential z potential and particle size of WSLP/pDNA complexes were measured as described by Maheshwari et al. 39 Briefly, polymer/pdna complexes were prepared as discussed above and were diluted 4 times in the cuvette and the electrophoretic mobility was determined with zpals (Brookhaven Instruments Corp., Holtsville, NY, USA). All experiments were performed at 251C, ph 7. and 677 nm wavelength at a constant angle of 151. The z potential was automatically calculated from the electrophoretic mobility based on Smoluchowski s formula. Following the determination of electrophoretic mobility, the samples were subjected to mean particle size measurement by the same equipment using the same light source and wavelength. The particle size was reported as effective mean diameter. In vitro transfection A7R5 cells, a smooth muscle cell line, were maintained in DMEM medium supplemented with 1% FBS in a 5% CO 2 incubator. For the transfection studies, A7R5 cells were seeded at a density of cells/well in sixwell flat-bottomed microassay plates (Falcon Co., Becton Dickenson, Franklin Lakes, NJ, USA) 24 h before transfection. After 24 h, the cells were washed twice with serum-free DMEM medium, and then 2 ml of fresh serum-free medium was added. WSLP/pDNA and 18/pDNA complexes were prepared at various N/P ratios. Lipofectamine/pDNA complex was prepared at a 5/1 weight ratio (lipofectamine/pdna). SuperFect/pDNA and 25/pDNA complexes were prepared at 6 ml of SuperFect/mg DNA and a 5/1 N/P ratio, respectively. The pdna concentrations of the complex solutions were fixed at 5 mg/ml. The polymer/pdna complexes were added to the cells. The cells were then incubated for 4 h at 371C in a 5% CO 2 incubator. After 4 h, the transfection mixtures were removed and 2 ml of fresh DMEM medium containing FBS was added. The cells were incubated for an additional 44 h at 371C. For the inhibition study, the cells were preincubated for 3 min with various amounts of cholesterol or LDL before transfection. For the cholesterol inhibition study, cholesterol was dissolved in ethanol at various concentrations. In total, 5 ml of each cholesterol solution was added to the cells. As a control, 5 ml of ethanol was added to the cells. For LDL inhibition study, LDL was prepared at various concentrations in PBS buffer. In total, 5 ml each LDL solution was added to the cells. As a control, 5 ml of PBS was added to the cells. Luciferase assay After transfection, the cells were washed with PBS twice, and 2 ml of reporter lysis buffer (Promega, Madison, WI, USA) was added to each well. After 15 min of incubation at room temperature, the cells were harvested and transferred to microcentrifuge tubes. After 15 s of vortexing, the cells were centrifuged at 11 rpm for 3 min. The extracts were transferred to fresh tubes and stored at 71C until use. The protein concentrations of the extracts were determined by using the BCA protein assay kit (Pierce, Iselin, NJ, USA). Luciferase activity was measured in terms of relative light units (RLU) using a 96-well plate Luminometer (Dynex Technologies Inc., Chantilly, USA). The luciferase activity was monitored and integrated over a period of 3 s. The final values of luciferase were reported in terms of RLU/mg total protein. Cytotoxicity assay Evaluation of cytotoxicity was performed by the 3-[4, 5- dimetylthiazol-2, 5-diphenyltetrazolium bromide (MTT) assay. 4 A7R5 cells were seeded at a density of cells/well in 96-well microassay plates (Falcon Co., Becton dickenson, Franklin Lakes, NJ, USA), and incubated for 24 h before transfection. WSLP/pDNA and 18/pDNA complexes were prepared at a 4/1 N/P ratio and lipofectamine/pdna was formulated at a 5/1 weight ratio (lipofectamine/pdna). 25/ pdna complex was prepared at a 5/1 N/P ratio and SuperFect/pDNA was formulated at 6 ml of SuperFect/ mg DNA. The amount of pdna was fixed at.2 mg/well. Polymer/pDNA was added to the cells and incubated for 4 h at 371C. After the incubation, the transfection mixture was replaced with 1 ml of fresh DMEM medium supplemented with 1% FBS. The cells were incubated for an additional 2 h at 371C. After the incubation, 24 ml of 2 mg/ml MTT solution in PBS was added. The cells were incubated for an additional 4 h at 371C and then MTT-containing medium was aspirated off, and 15 ml of DMSO was added to dissolve the formazan crystal formed by live cells. Absorbance was measured at 57 nm. The cell viability (%) was calculated according to the following equation: Cell viabilityð%þ ¼ðOD 57ðsampleÞ =OD 57ðcontrolÞ Þ1 here the OD 57(sample) represents the measurement from the wells treated with polymer/pdna complex and the OD 57(control) represents the measurements from the wells treated with 5% glucose solution only. In vivo transfection WSLP/pDNA complexes were prepared at various N/P ratios. 18/pDNA and 25 /pdna complexes were prepared at 1/1 and 5/1 N/P ratios, respectively. WSLP/pDNA, 18/pDNA, 25 /pdna complexes, and naked pdna were injected to the left ventricles of hearts of New Zealand white rabbits. The anterior descending, circumflex, and posterolateral coronary arteries were used as landmarks for the left ventricular injection site in the rabbit model. All injections were performed over 1 min in a subepicardial location with a fixed amount of pdna (2 mg) in injectate volumes of 1 ml. Transgene expression in the rabbit model was evaluated 3, 7, or 14 days after gene transfer by luciferase assay or RT-PCR. The hearts were harvested 3, 7, or 14 days after the injections and homogenized in reporter lysis buffer (Promega, Madison, WI, USA). The homogenate was incubated on ice for 1 h and spun for 5 min to pellet debris. Luciferase activity was measured as described above. RT-PCR Total RNA was prepared by acid guanidium thiocyanate phenol chloroform extraction from the injected myocardium, using RNAwiz (Ambion, Austin, TX, USA). The total RNA was treated with RNase-free DNase I (Promega, Madison, WI, USA) to eliminate the 591

8 592 contaminated DNA. The concentration of RNA was measured by the absorbance at 26 nm, and the integrity of RNA was confirmed by formaldehyde formamide denatured agarose gel electrophoresis. Total RNA (2 mg) was hybridized to the backward primer and reverse transcribed using AMV reverse transcriptase (Promega, Madison, WI). The sequences of the primers were as follows: forward primer, 5 -TCGATCCTGAGAACTT- CAG-3, backward primer, 5 -GGATGGCTTGAAGATG- TACT-3. The primers encompassed the intron sequence of psg5-vegf to exclude the amplification of the endogenous VEGF or the contaminated pdna. The reverse-transcribed samples were amplified by PCR using Taq polymerase (Promega, Madison, WI, USA). The sequences of the specific oligonucleotide primers were same as the primers for PCR. The primers encompassed the intron sequence of psg5-vegf to exclude the amplification of the contaminated pdna or the endogenous VEGF mrna. The PCR reaction consisted of 941C for 3 min, 35 cycles at 941C for 3 s, 551C for 3 s, and 721C for 1 min, followed by an extension of 1 min at 721C. The PCR products were separated by electrophoresis in 1.5% agarose gels. The length of the expected product was 376 bp. Statistical analysis The statistical significance of the results of cytotoxicity and transfection was evaluated by Student s t-test. P value under.5 was thought to be statistically significant. Acknowledgements We thank Thomas B Skidmore and Michael D Skidmore for technical assistance. This work was supported by NIH grant HL65477 and Expression Genetics Inc. References 1 Kabanov AV, Kabanov VA. DNA complexes with polycations for delivery of genetic material into cells. Bioconjug Chem 1995; 6: Han S, Mahato RI, Sung YK, Kim SW. Development of biomaterials for gene therapy. Mol Ther 2; 2: Wu GY, Wu CH. Receptor-mediated in vitro gene transformation by a soluble DNA carrier system. J Biol Chem 1987; 262: Wu GY, Wu CH. Receptor-mediated gene delivery and expression in vivo. J Biol Chem 1988; 263: Wu CH, Wilson JM, Wu GY. Targeting genes: delivery and persistent expression of a foreign gene driven by mammalian regulatory elements in vivo. J Biol Chem 1989; 264: Chiou HC et al. Enhanced resistance to nuclease degradation of nucleic acids complexed to asaloglycoprotein polylysine carriers. Nucleic Acids Res 1994; 22: Wagner E et al. Transferrin polycation conjugates as carriers for DNA uptake into cells. Proc Natl Acad Sci USA 199; 87: Wagner E et al. Influenza virus hemagglutinin HA-2 N-terminal fusogenic peptides augment gene transfer by transferrin polylysine DNA complexes: toward a synthetic virus-like gene-transfer vehicle. Proc Natl Acad Sci USA 1992; 89: Trubetskoy VS, Torchilin VP, Kennel SJ, Huang L. Use of N- terminal modified poly(l-lysine)-antibody conjugated as a carrier for targeted gene delivery in mouse lung endothelial cells. Bioconjug Chem 1992; 3: Midoux P et al. Specific gene transfer mediated by lactosylated poly-l-lysine into hepatoma cells. Nucleic Acids Res 1993; 21: Martinez-Fong D et al. Nonenzymatic glycosylation of poly-llysine: a new tool for targeted gene delivery. Hepatology 1994; 2: Erbacher P et al. Gene transfer by DNA/glycosylated polylysine complexes into human blood monocyte-derived macrophages. Hum Gene Ther 1996; 7: Choi YH et al. Characterization of a targeted gene carrier, lactosepolyethylene glycol-grafted poly-l-lysine and its complex with plasmid DNA. Hum Gene Ther 1999; 1: Kim J-S, Maruyama A, Akaike T, Kim SW. In vitro gene expression on smooth muscle cells using a terplex delivery system. J Control Rel 1997; 47: Kim J S et al. A new non-viral DNA delivery vector: the terplex system. J Control Rel 1998; 53: Kim J S, Maruyama A, Akaike T, Kim SW. Terplex DNA delivery system as a gene carrier. Pharm Res 1998; 15: Yu L, Nielsen M, Han S, Kim SW. TerplexDNA gene carrier system targeting artery wall cells. J Control Rel 21; 72: Affleck DG et al. Augmentation of myocardial transfection using TerplexDNA: a novel gene delivery system. Gene Ther 21; 8: Havel RJ. Receptor and non-receptor mediated uptake of chylomicron remnants by the liver. Atherosclerosis 1998; 141(Suppl 1): S1-S7. 2 Shih IL, Lees RS, Chang MY, Lees AM. Focal accumulation of an apolipoprotein B-based synthetic oligopeptide in the healing rabbit arterial wall. Proc Natl Acad Sci USA 199; 87: Lougheed M, Moore ED, Scriven DR, Steinbrecher UP. Uptake of oxidized LDL by macrophages differs from that of acetyl LDL and leads to expansion of an acidic endolysosomal compartment. Arterioscler Thromb Vasc Biol 1999; 19: Abdallah B et al. A powerful nonviral vector for in vivo gene transfer into the adult mammalian brain: polyethylenimine. Hum Gene Ther 1996; 7: Godbey WT, Wu KK, Mikos AG. Size matters: molecular weight affects the efficiency of poly(ethylenimine) as a gene delivery vehicle. J Biomed Mater Res 1999; 45: Benns JM et al. Folate-PEG-folate-graft-polyethylenimine-based gene delivery. J Drug Target 21; 9: Han S, Mahato RI, Kim SW. Water-soluble lipopolymer for gene delivery. Bioconjug Chem 21; 12: Lemkine GF et al. Optimisation of polyethyleneimine-based gene delivery to mouse brain. J Drug Target 1999; 7: Turunen MP et al. Efficient adventitial gene delivery to rabbit carotid artery with cationic polymer plasmid complexes. Gene Ther 1999; 6: Whitman L, Patzelt E, Wagner E, Kircheis R. Development of transferring-polycation/dna based vector for gene delivery to melanoma cells. J Drug Target 1999; 7: Nguyen H-K et al. Evaluation of polyether polyethyleneimine graft copolymers as transfer agents. Gene Ther 2; 7: Qin L et al. Efficient transfer of genes into murine cardiac grafts by Starburst polyamidoamine dendrimers. Hum Gene Ther 1998; 9: Simons K, Ikonen E. How cells handle cholesterol. Science 2; 29: Gao X, Huang L. A novel cationic liposome reagent for efficient transfection of mammalian cells. Biochem Biophys Res Commun 1991; 179: Nabel GH et al. Direct gene transfer with DNA liposome complexes in melanoma: expression, biologic activity, and lack

9 of toxicity in humans. Proc Natl Acad Sci USA 1993; 9: Gill DR et al. A placebo-controlled study of liposome-mediated gene transfer to the nasal epithelium of patients with cystic fibrosis. Gene Ther 1997; 4: Lee ED et al. Detailed analysis of structures and formulations of cationic lipids for efficient gene transfer to the lung. Hum Gene Ther 1996; 7: Lee M, Han S, Ko KS, Kim SW. Cell type specific and glucose responsive expression of interleukin-4 by using insulin promoter and water soluble lipopolymer. J Control Rel 21; 75: Benns JM et al. ph-sensitive cationic polymer gene delivery vehicle: N-Ac-poly (L-histidine)-graft-poly(L-lysine) comb shaped polymer. Bioconjug Chem 2; 11: Tang MX, Szoka FC. The influence of polymer structure on the interactions of cationic polymers with DNA and morphology of the resulting complexes. Gene Ther 1997; 4: Maheshwari A et al. Soluble biodegradable polymer-based cytokine gene delivery for cancer treatment. Mol Ther 2; 2: Lee M et al. Water soluble and low molecular weight chitosan based gene delivery. Pharm Res 21; 18: