Polyamidoamine Dendron-Bearing Lipids as a Nonviral Vector

Feb 29, 2012 - Toward production of highly potent vectors, we examined the influence of the generation of dendron moiety on transfection activity of P...
0 downloads 0 Views 2MB Size
Download PDF
Technical Note pubs.acs.org/bc

Polyamidoamine Dendron-Bearing Lipids as a Nonviral Vector: Influence of Dendron Generation Kenji Kono,* Ryuji Ikeda, Kota Tsukamoto, Eiji Yuba, Chie Kojima, and Atsushi Harada Department of Applied Chemistry, Graduate School of Engineering, Osaka Prefecture University, 1-1 Gakuen-cho, Naka-ku, Sakai, Osaka 599-8531, Japan S Supporting Information *

ABSTRACT: Recently, we demonstrated that octadecyl chains are important as alkyl chain moieties of polyamidoamine (PAMAM) dendron-bearing lipids for their serum-resistant transfection activity [Bioconjugate Chem. 2007, 18, 1349−1354]. Toward production of highly potent vectors, we examined the influence of the generation of dendron moiety on transfection activity of PAMAM dendron-bearing lipids having two octadecyl chains. We synthesized dendron-bearing lipids with PAMAM G1, G2, and G3 dendrons, designated respectively as DL-G1−2C18, DL-G2−2C18, and DL-G3−2C18. The DL-G2− 2C18 and DL-G3−2C18 interacted with plasmid DNA effectively and formed stable lipoplexes with small sizes and spherical shape. However, DL-G1−2C18 interacted with plasmid DNA less effectively and formed tubular-shaped lipoplexes with lower stability and larger size. Cells took up DL-G2−2C18 and DL-G3−2C18 lipoplexes efficiently, but cellular uptake of the DL-G1−2C18 lipoplexes was less efficient. Nevertheless, DL-G1−2C18 lipoplexes achieved 100−10 000 times higher levels of transgene expression, which was evaluated using luciferase gene as a reporter gene. Confocal scanning laser microscopic analysis of intracellular behaviors of the lipoplexes revealed that DL-G1−2C18 lipoplexes generated free plasmid DNA molecules in the cytosol more effectively than other lipoplexes did. Moderate binding ability of DL-G1−2C18 might be responsible for generation of lipoplexes which deliver plasmid DNA into cells, liberate it in the cytoplasm, and induce efficient transgene expression.

■

transfection into cells,11−15 polyamidoamine (PAMAM) dendrimers are one of the most extensively studied.16−18 Indeed, PAMAM dendrimers exhibit efficient gene transfection by generating the proton sponge effect through their tertiary amines.16−18 In addition, attachment of PAMAM dendrimer functional moieties, such as L-argnine,19 L-phenylalanine,20 and hydrophobic alkyl chains,21,22 to the periphery of PAMAM dendrimers was shown to increase transfection activity of the corresponding unmodified dendrimers, indicating that PAMAM dendrimers can be used as a base material for the production of highly efficient gene vectors.13,23 Therefore, we chose PAMAM dendrimers as a polymer-based functional component and derivatized them to generate a new type of lipids that consisted of a PAMAM dendron and two dodecyl chains.24,25 Transfection activity of the PAMAM dendron-bearing lipids was shown to increase concomitantly with increasing dendron generation because of the proton sponge effect increase with dendron generation.24 In addition, we found recently that serum-resistant transfection activity of the PAMAM G3 dendron-based lipid can be improved greatly by increasing their alkyl chain length from dodecyl to octadecyl chains.26 Indeed, we expected that optimization of the generation of

INTRODUCTION Development of nonviral vectors that deliver therapeutic genes to target cells has been demanded for efficient and safe gene therapy and iPS cell-based regenerative medicine.1−3 Among these nonviral vectors, various types of cationic lipids and cationic polymers have been studied intensively, but their activity requires improvement. These vectors can associate with plasmid DNA and form complexes, which are respectively designated as lipoplexes and polyplexes.4−6 These vector− DNA complexes bind to the cell surface through electrostatic interactions and are taken up by cells mainly via endocytosis.6,7 Subsequently, some parts of plasmid DNA contained in the complexes reach the nucleus, where gene transcription occurs. Nevertheless, most complexes are likely to be trapped in the endosome, where they are eventually degraded in the lysosome. Therefore, avoidance of plasmid DNA degradation in the lysosome and its transfer into cytosol is regarded as a key process for efficient transfection.7 To date, various strategies have been explored to promote transfer of plasmid DNA into cytosol, including membrane fusion8−10 and proton sponge effect,11,12 which are, respectively, typical functions of lipid-based and polymer-based vectors. We regarded the combination of these polymer-based and lipidbased functions as an efficient strategy for producing efficient nonviral vectors. With respect to polymer-based vectors, while various types of polymers have been shown to induce gene © 2012 American Chemical Society

Received: July 12, 2011 Revised: December 14, 2011 Published: February 29, 2012 871

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

Figure 1. Structures of DL-G1−2C18 (A), DL-G2−2C18 (B), and DL-G3−2C18 (C).

PAMAM dendron moiety of the octadecyl-type PAMAM dendron-based lipids might generate highly potent vectors, which achieve efficient transfection in the presence of serum. In this study, toward production of highly potent vectors, we examined the influence of the generation of dendron moiety on transfection activity of the PAMAM dendron-bearing lipids having two octadecyl chains. We synthesized dendron-bearing lipids of three kinds with PAMAM G1, G2, and G3 dendrons, which are designated DL-G1−2C18, DL-G2−2C18, and DLG3−2C18 (Figure 1), and examined transfection of cells using these dendron-bearing lipids. Unexpected but marked increases of transfection activity with decreasing generation were found.

■

EXPERIMENTAL SECTION General Methods. Dioctadecylamine was purchased from Fluka (Tokyo, Japan). Methyl acrylate, ethylenediamine, and tris(hydroxymethyl)aminomethane (Tris) were obtained from Kishida Chemical (Osaka, Japan). Sodium cyanide and ethidium bromide were supplied from Wako Pure Chemical Industries (Osaka, Japan). Merck Kieselgel 60 (230−400 mesh) was used for silica gel chromatography. 3-(4,5-Dimethyl-2thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) was obtained from Dojin Laboratories (Kumamoto, Japan). Fetal bovine serum (FBS) was from PAA Laboratories GmbH (Pasching, Austria). Dulbecco’s modified Eagle’s medium (DMEM) was from Nissui Pharmaceutical (Tokyo, Japan). Agarose was purchased from Nacalai Tesque (Kyoto, Japan). Plasmid DNA pCMV-Luc, which contains the cDNA of firefly luciferase driven by a human cytomegalovirus immediate−early promoter, was a gift from Dr. Kazuo Maruyama, Teikyo University, and amplified in E. coli, isolated, and purified using a QIAGEN plasmid Maxi Kit. Dendron-based lipids DL-G1− 2C18, DL-G2−2C18, and DL-G3−2C18 were synthesized as previously reported.26 Preparation of Lipoplexes. To a dry, thin membrane of the dendron-bearing lipid, phosphate-buffered saline (PBS) was added and sonicated for 2 min using a bath-type sonicator to afford a lipid suspension. Plasmid DNA (1 μg) was dissolved in 20 mM Tris-HCl buffer (pH 7.4, 50 μL), mixed with the lipid suspension (50 μL), and incubated for 30 min at room

Figure 2. Agarose gel electrophoretic analysis for complexation of DLG1−2C18 (A), DL-G2−2C18 (B), and DL-G3−2C18 (C), with plasmid DNA at varying N/P ratios. The percentage of free plasmid DNA was evaluated and plotted against N/P ratio (D) for complexation of DLG1−2C18 (diamonds), DL-G2−2C18 (squares), and DL-G3−2C18 (triangles) with plasmid DNA. Each point represents the mean ± SD (n = 2).

temperature to afford a lipoplex with a given ratio of primary amine of DL-G3 to DNA phosphate (N/P ratio). Agarose Gel Electrophoresis. The dendron-bearing lipid−DNA complexes with varying N/P ratios were prepared 872

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

■

by mixing plasmid DNA (1 μg) dissolved in 20 mM Tris-HCl buffer (5 μL) and lipid suspension (5 μL). After 30 min incubation at room temperature, the samples (10 μL) were electrophoresed on 0.6 wt % agarose gel in 40 mM Tris, 20 mM sodium acetate, and 2 mM EDTA buffer (pH 8.0) containing 1 μg/mL ethidium bromide at 100 V for 30 min. The ethidium bromide-stained bands were visualized using a LAS-1000UV mini (Fujifilm, Tokyo, Japan) and analyzed with Science Lab 2003 Multi Gauge software (Fujifilm, Tokyo, Japan). Dynamic Light Scattering (DLS) Measurement. The diameters of lipoplexes were estimated by DLS using an electrophoretic light scattering spectrometer ELS-8000F (Otuka Electronics Co, Ltd., Osaka, Japan). Lipoplexes with various N/P ratios were prepared by mixing plasmid DNA (20 μg) and a given amount of the dendron-bearing lipid and incubated for 30 min as described above. The data were obtained at a detection angle of 90° at 25 °C and were analyzed by the cumulant method. Atomic Force Microscopy (AFM). AFM measurements were performed by SPI3800 probe station and SPA400 unit system of the scanning probe microscopy system (Seiko Instruments Inc., Japan). The cantilever was made of silicon (SI-DF40; Seiko Instruments Inc., Japan), and its spring constant was 16 N/m. Lipoplexes were formed at a total of 100 μg/mL of plasmid DNA concentration. Lipoplexes containing 1 μg of plasmid DNA were applied to freshly cleaved mica and incubated on the mica for 30 min. The measurements were performed in a dynamic force mode (noncontact mode). Transfection. Transfection of CV1 cells or HeLa cells was done according to the following procedures. The cells were seeded in 0.5 mL of DMEM supplemented with 10% FBS in 24well culture plates at 5.0 × 104 cells per well the day before transfection. The cells were washed with PBS containing 0.36 mM CaCl2 and 0.42 mM MgCl2 [PBS(+)] and then covered with DMEM in the presence or absence of 10% FBS (1 mL). The lipoplexes containing plasmid DNA were added gently to the cells and incubated for 4 h at 37 °C. Then, the cells were rinsed with PBS(+), covered with DMEM containing 10% FBS, and incubated at 37 °C. After 40 h, the cells were lysed by adding 50 μL of Luc-PGC-50 detergent (Toyo Ink, Tokyo, Japan). A 20 μL aliquot was taken from each dish and used for one luciferase assay using a kit (Toyo Ink) and a Lumat LB9507 luminometer (Berthold, Bad Wildbad, Germany). The protein content of the lysate was measured by Coomassie Protein Assay Reagent (Pierce, IL, USA) using bovine serum albumin as the standard. Transfection with Lipofectamine LTX was also performed according to the protocol recommended by the manufacturer. Cytotoxicity. HeLa cells (5 × 104 cells per well) were seeded in a 24 well plate and grown in DMEM supplemented with 10% FBS for 20 h. The cells were washed with PBS(+) and covered with DMEM supplemented with 10% FBS (1 mL). Then, the lipoplex suspension (100 μL) containing plasmid DNA (1 μg) at N/P ratio of 2.0 was added to the cells and incubated at 37 °C for 4 h. The cells were rinsed with PBS(+), covered with DMEM containing 10% FCS, and incubated at 37 °C for 40 h and their viability was evaluated using an assay with 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT).27 The cell viability was defined as the ratio to that without treatment.

Technical Note

RESULTS AND DISCUSSION

Lipoplex Formation with Plasmid DNA. The ability of the dendron-bearing lipids with varying dendron generations to form a lipoplex with plasmid DNA was investigated using gel retardation assay. The plasmid DNA (1 μg) was incubated with varying amounts of DL-G1−2C18, DL-G2−2C18, and DL-G3− 2C18 for 30 min and electrophoresed on an agarose gel. Figure 2A−C shows that the amount of free plasmid DNA decreased concomitantly with the increasing N/P ratio for all dendronbased lipids, indicating that these lipids can form a lipoplex with plasmid DNA. No free plasmid DNA was observed at N/P ratios higher than 0.8 for DL-G2−2C18 and DL-G3−2C18, indicating that these lipids formed lipoplexes efficiently with plasmid DNA. However, for DL-G1−2C18, more than 60% of plasmid DNA existed as free after electrophoresis, even at the N/P ratio of 2 (Figure 2D), and the lipoplex formation became complete above the N/P ratio of 8 (see Figure S1 of Supporting Information). It is possible that dissociation of plasmid DNA from DL-G1−2C18 lipoplexes was induced during electrophoresis, but the result might indicate that its lipoplex formation ability is weak. Indeed, electrostatic interaction between the dendron moiety and plasmid DNA are expected to play an important role in their lipoplex formation. The number of terminal amino groups increases concomitantly with increasing dendron generation. Therefore, multivalent interaction of positively charged terminal groups of the dendron moiety of higher generations with negatively charged DNA molecules might result in efficient lipoplex formation.18,24,28 Lipoplex Particle Size. Influence of the generation of the dendron moiety of the dendron-bearing lipids on their lipoplex size was investigated using DLS (Figure 3 and Figure S2).

Figure 3. Mean diameter of DL-G1−2C18 (diamonds), DL-G2−2C18 (squares), and DL-G3−2C18 (triangles) lipoplexes as a function of N/P ratio. The mean diameter was estimated using DLS. The lipid suspension was added to plasmid DNA solution and incubated for 30 min at 25 °C. Each point represents the mean ± SD (n = 2 or 3).

Although DL-G1−2C18 was shown to have a weak ability to form lipoplexes from the electrophoresis experiment (Figure 2), this lipid actually formed lipoplexes whose sizes were generally large. In addition, DL-G2−2C18 formed large lipoplexes of around 2000−3500 nm at the N/P ratio of 2, although their size decreased to about 500 nm above the N/P ratio of 4. A similar N/P ratio dependence was observed for the DL-G3−2C18 lipoplexes, although their sizes were somewhat smaller than those of the DL-G2−2C18 lipoplexes. 873

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

the tubular structures of the DL-G1−2C18 lipoplexes contained plasmid DNA, we prepared the lipoplexes using FITC-labeled plasmid DNA. We observed that the tubular structures displayed rhodamine fluorescence using a confocal laser scanning microscope, indicating that these tubular structures indeed contain plasmid DNA (Figure S3). Transfection Activity of Dendron-Bearing Lipid−DNA Lipoplexes. We investigated the influence of dendron generation of dendron-bearing lipids on their transfection activity. The transfection activity of lipoplexes generally varies depending on their N/P ratio. Therefore, we prepared lipoplexes consisting of the dendron-bearing lipids having different dendron generations and plasmid DNA containing luciferase gene at various N/P ratios. Then, we examined transfection of CV1 cells using these lipoplexes. Figure 5A presents the expression of luciferase in the cells treated with these lipoplexes in the presence of 10% serum. Although both DL-G2−2C18 and DL-G3−2C18 lipoplexes exhibited very low transfection activities at the N/P ratio of 1, their activity increased concomitantly with the increasing N/P ratio from 1 to 2 and reached the maximum value at N/P ratio of 2, where these lipoplexes achieved expression of luciferase of around 1 × 107 RLU/mg protein. However, a further increase in the N/P ratio rather decreased their transfection activity. An increase of the N/P ratio might increase positive charges of the lipoplexes, which causes elevation of their affinity to the cells and increase of their transfection activity. However, a further increase of the N/P ratio might engender cellular toxicity derived from too much positive charge and decreased gene expression in the cells.31 In fact, we have already observed similar N/P ratio dependence of transfection activities for the dendron-based lipids having the same dendron moieties and dodecyl chains instead of octadecyl chains.24−26 In contrast to cases of the G2 and G3 dendron-based lipoplexes, DL-G1−2C18 lipoplexes exhibited extremely high transfection activity irrespective of the N/P ratio. Even the lipoplex with the N/P ratio of 1 induced a very high level of luciferase expression in the cells, at which 4 orders of magnitude higher expression of luciferase gene was achieved compared to the cases of the G2 and G3 dendron-based lipoplexes. An increase in the N/P ratio for the G1-dendron-based lipoplexes increased their transfection activity to some degree and the high level of transfection activity was retained at N/P ratios of 1.5−8, where about 1 × 109 RLU/mg protein of luciferase activity was induced. We also examined the activity of these dendron-bearing lipids for transfection of HeLa cells (Figure 5B). Considerably high activity of DL-G1−2C18 lipoplex was again observed for transfection of HeLa cells, compared to the G2 and G3 dendron-bearing lipid lipoplexes with the same N/P ratio of 2, where these dendron-bearing lipid lipoplexes achieved efficient transfection (Figure 5A). Its activity was of the same level as that of Lipofectamine LTX, which was used as a positive control with high transfection activity. We further checked cytotoxicity for these dendron-based lipid lipoplexes during the transfection. As is seen in Figure 5C, these lipoplexes showed a low level of cytotoxicity under the transfection condition. Dendron-based lipids were originally designed to achieve efficient gene transfection of cells through the synergetic effect of hydrophobic interaction derived from their long alkyl chain moieties and the so-called proton sponge effect derived from the ability of PAMAM dendron moiety to buffer the acidification of endosome, which promotes the transfer of gene from

Charge-neutralized lipoplexes are known to tend to form large aggregates because of their hydrophobic interaction.29 For that reason, these dendron-based lipids might form large lipoplexes around N/P ratio of 2, where their charge density was low. In contrast, at an N/P ratio higher than 4, the lipoplexes might possess sufficient net charge to suppress aggregation of small lipoplexes with G2 and G3 dendron-bearing lipids. In contrast, lipoplexes with the G1 dendron-bearing lipid exhibited a large size of 3000−4000 nm above the N/P ratio of 4, which suggests that the DL-G1−2C18 lipoplexes retain hydrophobic character at high N/P ratios. Because DL-G1− 2C18 molecules have fewer positive charges than the G2 and G3 dendron-bearing lipid molecules, the DL-G1−2C18 lipoplexes contain more octadecyl chains than those of the G2 and G3 dendron-bearing lipids at the same N/P ratio. Therefore, the hydrophobic character of the DL-G1−2C18 lipoplexes might be derived from higher contents of the octadecyl chains of the DL-G1−2C18 lipoplexes than those of the DL-G2−2C18 and DL-G2−2C18 lipoplexes of the same N/P ratio. Morphology of the lipoplexes was also investigated using AFM. As is seen in Figure 4, their morphology is quite hetero-

Figure 4. AFM images of DL-G1−2C18 (A,B), DL-G2−2C18 (C), and DL-G3−2C18 (D) lipoplexes. N/P ratio of lipoplex was 2. (B) A portion of image (A) was enlarged.

geneous, which is consistent with their large size distribution estimated using DLS (Figure S2). The DL-G2−2C18 and DL-G3−2C18 lipoplexes displayed particles with spherical shape, although some particles seem to be formed by fusion of spherical lipoplexes. The DL-G1−2C18 lipoplex also exhibited spherical particles, but these particles contained many tubular structures. The overall dynamic molecular shape of amphiphiles is known to affect their assembly morphology.30 These dendron-based lipids have the same long alkyl chains as the hydrophobic moiety but different generation of dendron as the polar head moiety, which would result in the different molecular shape among these dendron-based lipids. Probably, the DL-G1− 2C18 molecule having a relatively small headgroup may enable formation of complexes with a tubular shape, although correlation between the molecular shape of the dendron-based lipids and their lipoplex morphology is still to be clarified. To confirm that 874

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

Figure 5. (A) Luciferase activities of CV1 cells treated with DL-G1−2C18, DL-G2−2C18, and DL-G3−2C18 lipoplexes with varying N/P ratios. Luciferase activity (B) and viability (C) for HeLa cells treated with DL-G1−2C18, DL-G2−2C18, and DL-G3−2C18 lipoplexes with N/P ratio of 2. Transfection with Lipofectamine LTX was performed as a control. The cells (5 × 104) were treated with the lipoplexes containing 1 μg DNA in the presence of 10% FBS. Each bar represents the mean ± SD (n = 3) for (A) and (B) or (n = 2) for (C).

endosome into cytosol and enables their efficient expression.24 Because the number of tertiary amines of the dendron-bearing lipids, which generate the proton sponge effect, increases in the order of DL-G1−2C18 < DL-G2−2C18 < DL-G3−2C18, we expected that their transfection activity would increase in the same order. In fact, we observed that the dendron-based lipids with dodecyl chains increased their transfection activity with increasing dendron generation (DL-G1−2C12 < DL-G2−2C12 < DL-G3−2C12). In addition, the DL-G1−2C18 lipoplexes have low ability to form lipoplexes (Figure 2). Therefore, the high transfection activity of the G1-based dendron lipid lipoplexes was completely unexpected. Synthetic vectors without the endosome-buffering ability, such as poly(lysine), are known to increase their transfection efficiency in the presence of an endosomotropic agent, chloroquine, because it causes destabilization of endosome and promotes transfer of plasmid DNA into cytosol.32 Therefore, the effect of chloroquine on the dendron-bearing lipid-mediated transfection was examined to estimate their ability to promote the endosome escape of plasmid DNA (Figure 6). HeLa cells

were transfected with the dendron-bearing lipid lipoplexes with the N/P ratio of 2 in the presence or absence of chloroquine. As presented in Figure 6A, the presence of chloroquine decreased luciferase for the cells treated with the G1 and G2 dendronbearing lipids to some degree, probably because cellular toxicity of chloroquine might suppress the gene expression.33 However, for transfection with the G3 dendron-bearing lipid, cellular luciferase expression was 2.5-fold higher in the presence of chloroquine than in the case of transfection in the absence of chloroquine (Figure 6B). This result suggests that G1 and G2 dendron-bearing lipids have sufficient ability to achieve efficient endosome escape of plasmid DNA, whereas the ability of the G3-dendron-bearing lipid is insufficient. In an earlier report, we described that the dendron-bearing lipids having dodecyl chains, such as DL-G1−2C12, DL-G2−2C12, and DL-G3−2C12, exhibited enhancement of transfection in the presence of chloroquine.24 Considering the difference of the alkyl chain length of these dendron-based lipids, more hydrophobic character of octadecyl chains of the dendron-bearing lipids 875

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

Figure 6. Influence of chloroquine on transfection activity of HeLa cells treated with DL-G1−2C18, DL-G2−2C18, and DL-G3−2C18 lipoplexes with N/P ratio of 2. Luciferase activities of the cells treated with the lipoplexes without chloroquine (open bar) or with 100 μM chloroquine (closed bar) in the presence of 10% FBS. Each bar represents the mean ± SD (n = 3). (B) Ratio of cellular luciferase activity in the presence of chloroquine to that in the absence of chloroquine.

10−100 times as high a level of luciferase activity when treated with DOPE-containing lipoplexes of G2- and G3-dendronbearing lipids. In contrast, no fundamental increase of the gene expression induced by the DOPE inclusion was observed for the case of the G1-dendron-bearing lipoplexes. Some reports in the literature describe that DOPE might increase gene transfection by enhancing destabilization and fusion of endosomal membrane, which promotes transfer of plasmid DNA into the cytosol.34,35 Therefore, the results suggest that the G1and G2 dendron-bearing lipid lipoplexes can destabilize the endosomal membrane strongly without support of DOPE. We sought an explanation for why the DL-G1−2C18 lipoplexes achieved markedly higher transgene expression efficiency than the lipoplexes with the larger dendron moieties. We expected that the efficiency of plasmid DNA introduction into the cells might be influenced by the generation of dendron moiety of the lipids. Therefore, to estimate the amount of plasmid DNA introduced into the cells, plasmid DNA labeled with FITC was used for the lipoplex preparation with the dendron lipids. CV1 cells were treated with the lipoplexes for 4 h and the cellular fluorescence was evaluated by flow cytometry (Figure S5). The percentages of cellular association for the G1, G2, and G3 dendron-bearing lipid lipoplexes were estimated to be 77.1 ± 2.3%, 89.5 ± 2.0%, and 88.9 ± 0.8%, respectively, indicating that the lipoplexes were efficiently taken up by the cells. Although these lipoplexes had relatively larger sizes, which might be formed by aggregation of these small lipoplexes, such lipoplexes may be dissociated into lipoplexes of smaller size after adsorption onto the cellular membrane and be taken up by the cells through endocytosis. Figure 8 depicts the average of cellular fluorescence intensity of FITC for cells treated with lipoplexes with various dendron-bearing lipids. Contrary to our expectation, the cellular fluorescence intensities suggested that the amount of plasmid DNA delivered into the cells using DL-G1−2C18 was much smaller than the cases of DL-G2−2C18 and DL-G3−2C18 for DNA delivery, which might result from the low ability of DL-G1−2C18 to form lipoplexes, compared to other dendron lipids with larger dendron moieties (Figure 2). We also expected that intracellular behavior of the lipoplexes might affect their transgene expression efficiency. We prepared lipoplexes from FITC-labeled plasmid DNA, the dendron lipids, and a small aliquot (0.6 mol %) of Rh-PE, which was used for detection of lipid component of lipoplexes in the cell.

might enhance destabilization of endosome membranes through strong hydrophobic interaction. Another strategy to promote the endosomal escape of plasmid DNA introduced with lipid-based vectors is to use fusogenic phospholipids, such as dioleoylphosphatidylethanolamine (DOPE), as an additional component of lipoplexes.34,35 In fact, we observed that the inclusion of DOPE in the dendron-based lipids having dodecyl chains greatly increased their transfection activity.24 Therefore, we examined the effect of DOPE on the transfection activity of the dendron−lipid lipoplexes. The CV1 cells were transfected using dendron− lipid-based lipoplexes with varying DOPE contents and N/P ratios (Figure S4), which showed that the optimum DOPE/ dendron−lipid molar ratio and N/P ratio were, respectively, 0.1/1 and 4 for DL-G1−2C18 lipoplex, 2/1 and 6 for DL-G2− 2C18 lipoplex, and 4/1 and 4 for DL-G3−2C18 lipoplex. We compared the transfection activity of DOPE-containing lipoplexes and DOPE-free lipoplexes with their optimal compositions. As presented in Figure 7, CV-1 cells treated with these lipoplexes containing luciferase gene expressed

Figure 7. Luciferase activities of CV1 cells treated with DL-G1−2C18, DL-G2−2C18, and DL-G3−2C18 lipoplexes without (open bar) or with DOPE (closed bar). The N/P ratio and DOPE/dendron-bearing lipid ratio of the lipoplex are shown in figure. Cells were treated with lipoplexes in the presence of 10% FBS. Each bar represents the mean ± SD (n = 3). 876

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

treated with the DL-G1−2C18 lipoplex (Figure 9A), these cells displayed many spots emitting green fluorescence, which was indicative of the existence of free plasmid DNA, which might be liberated from the lipoplex. Compared to cases of the DL-G1− 2C18 lipoplex-treated cells, in the merge images of the cells treated with the DL-G2−2C18 and DL-G3−2C18 lipoplexes, the quantities of green spots apparently decreased, and yellow fluorescence-emitting spots became predominant (Figure 9B,C). The result indicates that green fluorescence-emitting FITC-labeled plasmid DNA and red fluorescence-emitting RhPE coexisted in the same places, suggesting that plasmid DNA and the dendron lipids were still bound in these cells. Considering that stability of the DL-G1−2C18 lipoplex might be much lower than that of the DL-G2−2C18 and DL-G3−2C18 lipoplexes (Figure 2), the former lipoplex was able to liberate plasmid DNA in the cells readily through interaction with cellular components, such as proteins of various kinds, thereby enabling more efficient transcription of the luciferase gene. The results underscore the importance of the balance of polar head moiety and hydrophobic tail moiety, which reconciles stabilization and destabilization of the lipoplexes, respectively, for efficient delivery of plasmid DNA into cells and effective liberation of plasmid DNA in the intracellular space. In addition, morphology of the lipoplexes may be another factor which affects their high transfection efficiency. Relatively large surface area of the tubular DL-G1−2C18 lipoplexes should enhance interaction with endosomal membrane, resulting in efficient liberation of plasmid DNA from the dendron-based lipid and transfer into cytosol. The lipid having small G1 dendron and two highly hydrophobic octadecyl chains forms moderately stable lipoplexes with a hydrophobic nature, which enables

Figure 8. Fluorescence intensities of CV1 cells treated with DL-G1− 2C18, DL-G2−2C18, and DL-G3−2C18 lipoplexes with N/P ratio of 2 prepared using FITC-labeled plasmid DNA. Cells (1 × 105) were treated with the lipoplexes and the complex containing 2 μg DNA in the presence of 10% FBS. Each bar represents the mean ± SD (n = 3).

HeLa cells were treated with doubly labeled lipoplexes and viewed with a confocal laser scanning microscope (CLSM) (Figure 9). The cells displayed both FITC and Rh fluorescence, indicating that the plasmid DNA component and the lipid components were internalized into the cells. Spots exhibiting fluorescence in the cells are generally smaller than the size of lipoplexes evaluated by DLS (Figure 3), suggesting that the lipoplexes dissociated into smaller particles after internalization in the cell. As presented in the merged image of the cells

Figure 9. Fluorescence microscopic observation of HeLa cells treated with DL-G1−2C18 lipoplexes. Differential interference contrast image (A), green shows the fluorescence of FITC-labeled pDNA (B), red shows the fluorescence of Rhodamine-PE (C), and overlay image (D). Cells (1 × 105) were treated with the lipoplexes containing 2 μg DNA in the presence of 10% FBS. 877

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

biophysics to biological applications. Adv. Drug Delivery Rev. 47, 277−294. (5) Wagner, E., and Kloeckner, J. (2006) Gene delivery using polymer therapeutics. Adv. Polym. Sci. 192, 135−173. (6) Tros de Ilarduya, C., Sun, Y., and Duzgunes, N. (2010) Gene delivery by lipoplexes and polyplexes. Eur. J. Pharm. Sci. 40, 159−170. (7) Pouton, C. W., and Seymour, L. W. (1998) Key issues in nonviral gene delivery. Adv. Drug Delivery Rev. 34, 3−19. (8) Zelphati, O., and Szoka, F. C. Jr. (1996) Mechanism of oligonucleotide release from cationic liposomes. Proc. Natl. Acad. Sci. U.S.A. 93, 11493−11498. (9) Monkkonen, J., and Urtti, A. (1998) Lipid fusion in oligonucleotide and gene delivery with cationic lipids. Adv. Drug Delivery Rev. 34, 37−49. (10) Mok, K. W. C., and Cullis, P. R. (1997) Structural and fusogenic properties of cationic liposomes in the presence of plasmid DNA. Biophys. J. 73, 2534−2545. (11) Boussif, O., Lezoualc’h, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., and Behr, J. P. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: Polyethylenimine. Proc. Natl. Acad. Sci. U.S.A. 92, 7297−7301. (12) Sonawane, N. D., Szoka, F. C., and Verkman, A. S. (2003) Chloride accumulation and swelling in endosomes enhances DNA transfer by polyamine-DNA polyplexes. J. Biol. Chem. 278, 44826− 44831. (13) Dufes, C., Uchegbu, I. F., and Schatzlein, A. G. (2005) Dendrimers in gene delivery. Adv. Drug Delivery Rev. 57, 2177−2202. (14) Miyata, K., Oba, M., Nakanishi, M., Fukushima, S., Yamasaki, Y., Koyama, H., Nishiyama, N., and Kataoka, K. (2008) Polyplexes from poly(aspartamide) bearing 1,2-diaminoethane side chains induce pHselective, endosomal membrane destabilization with amplified transfection and negligible cytotoxicity. J. Am. Chem. Soc. 130, 16287− 16294. (15) Lim, Y., Kim, S., Suh, H., and Park, J. (2002) Biodegradable, endosome disruptive, and cationic network-type polymer as a highly efficient and nontoxic gene delivery carrier. Bioconjugate Chem. 13, 952−957. (16) Haensler, J., and Szoka, F. C. Jr. (1993) Polyamidoamine cascade polymers mediate efficient transfection of cells in culture. Bioconjugate Chem. 4, 372−379. (17) Tang, M., Redemann, C. T., and Szoka, F. C. Jr. (1996) In vitro gene delivery by degraded polyamidoamine dendrimers. Bioconjugate Chem. 7, 703−714. (18) Kukowska-Latallo, J. F., Bielinska, A. U., Johnson, J., Spindler, R., Tomalia, D. A., and Baker, J. R. Jr. (1996) Efficient transfer of genetic material into mammalian cells using Starburst polyamidoamine dendrimers. Proc. Natl. Acad. Sci. U.S.A. 93, 4897−4902. (19) Choi, J. S., Nam, K., Park, J., Kim, J. B., Lee, J. K., and Park, J. (2004) Enhanced transfection efficiency of PAMAM dendrimer by surface modification with L-argnine. J. Controlled Release 99, 445−456. (20) Kono, K., Akiyama, H., Takahashi, T., Takagishi, T., and Harada, A. (2005) Transfection activity of polyamidoamine dendrimers having hydrophobic amino acid residues in the periphery. Bioconjugate Chem. 16, 208−214. (21) Santos, J. L., Oliveira, H., Pandita, D., Rodrigues, J., Pego, A. P., Grnja, P. L., and Tomas, H. (2010) Functionalization of poly(amidoamine) dendrimers with hydrophobic chains for improved gene delivery in mesenchymal stem cells. J. Controlled Release 144, 55−64. (22) Morales-Sanfrutos, J., Megia-Fernandez, A., Hernandez-Mateo, F., Giron-Gonzalez, M. D., Salto-Gonzalez, and Santoyo-Gonzalez, F. (2011) Alkyl sulfonyl derivatized PAMAM-G2 dendrimers as nonviral gene delivery vectors with improved transfection efficiencies. Org. Biomol. Chem. 9, 851−864. (23) Guillot-Nieckowski, M., Eisler, S., and Diederrich, F. (2007) Dendritic vectors for gene transfection. New J. Chem. 31, 1111−1127. (24) Takahashi, T., Kono, K., Itoh, T., Emi, N., and Takagishi, T. (2003) Synthesis of novel cationic lipids having polyamidoamine dendrons and their transfection activity. Bioconjugate Chem. 14, 764− 773.

efficient cellular internalization, efficient endosomal escape, and efficient liberation of plasmid DNA, resulting in efficient gene expression. In conclusion, we demonstrated that the transfection activity of the PAMAM dendron-bearing lipids having two octadecyl chains varied significantly depending on their generation of dendron moiety. The dendron lipid having the smallest dendron DL-G1−2C18 exhibited significantly higher transfection activity among the dendron lipids. Results demonstrated the importance of the dendron moiety of the PAMAM dendron-bearing lipids for their lipoplex formation with plasmid DNA having appropriate stability to reconcile efficient cellular delivery and liberation of plasmid DNA. In this sense, the dendron-bearing lipids are advantageous because their characteristics which are necessary to achieve efficient transfection can be controlled precisely by selecting a backbone structure and generation of dendron moiety. Synthetic vectors of many types are known to need helper lipids, such as DOPE, to achieve efficient transfection. In addition, their transfection activity is often decreased in the presence of serum. Therefore, it is noteworthy that the DL-G1−2C18 achieved efficient gene transfection in the presence of serum without the help of fusogenic lipid DOPE. We expect that structural optimization for alkyl chain moieties, such as introduction of double bonds, and for dendron moiety, such as hydrophobicity, might engender further improvement of transfection activity of the dendron-based lipids. Therefore, dendron-based lipids might be promising for the production of potent synthetic vectors.

■

ASSOCIATED CONTENT

* Supporting Information S

Agarose gel electrophoretic analysis for complexation of DLG1−2C18 with plasmid DNA at varying N/P ratios, size distribution for G1, G2, and G3 dendron-bearing lipid lipoplexes, CLSM analysis of lipoplex containing FITC-labeled DNA, the effects of N/P ratio and DOPE/dendron-lipid ratio on the transfection activity of DL-G1−2C18, DL-G2−2C18, and DL-G3−2C18 lipoplexes, and flow cytometry analysis for CV1 cells treated with DL-G1−2C18, DL-G2−2C18, and DL-G3− 2C18 lipoplexes containing FITC-labeled plasmid DNA. This material is available free of charge via Internet at http://pubs. acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +81-722549330. Fax: +81-722549330. Present Address

C. Kojima, Nanoscience and Nanotechnology Research Center, Research Organization for the 21st Century, Osaka Prefecture University, 1−2 Gakuen-cho, Naka-ku, Sakai, Osaka 599−8570, Japan.

■

REFERENCES

(1) Mintzer, M. A., and Simanek, E. E. (2009) Nonviral vectors for gene delivery. Chem. Rev. 109, 259−302. (2) Okita, K., Nakagawa, M., Hyenjong, H., Ichisaka, T., and Yamanaka, S. (2008) Generation of mouse induced pluripotent stem cells without viral vectors. Science 322, 949−953. (3) Niidome, T., and Huang, L. (2002) Gene therapy progress and prospects: nonviral vectors. Gene Ther. 9, 1647−1652. (4) de Lima, M. C. P., Simoes, S., Pires, P., Faneca, H., and Duzgunes, N. (2001) Cationic lipid-DNA complexes in gene delivery: from 878

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879

Bioconjugate Chemistry

Technical Note

(25) Takahashi, T., Harada, A., Emi, N., and Kono, K. (2005) Preparation of efficient gene carriers using a polyamidoamine dendron-bearing lipid: improvement of serum resistance. Bioconjugate Chem. 16, 1160−1165. (26) Takahashi, T., Kojima, C., Harada, A., and Kono, K. (2007) Alkyl chain moieties of polyamidoamine dendron-bearing lipids influence their function as a nonviral gene vector. Bioconjugate Chem. 18, 1349−1354. (27) Mosmann, T. (1983) Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55−63. (28) Tang, M. X., and Szoka, F. C. (1997) The influence of polymer structure on the interactions of cationic polymers with DNA and morphology of the resulting complexes. Gene Ther. 4, 823−832. (29) Pitard, B., Aguerre, O., Airiau, M., Lachages, A. M., Boukhnikachvili, T., Byk, G., Dubertret, C., Herviou, C., Scherman, D., Mayaux, J. F., and Crouzet, J. (1997) Virus-sized self-assembling lamellar complexes between plasmid DNA and cationic micelles promote gene transfer. Proc. Natl. Acad. Sci. U.S.A. 94, 14412−14417. (30) Israelachvili, J. N., Marcelja, S., and Horn, R. G. (1980) Physical principles of membrane organization. Q. Rev. Biophys. 13, 121−200. (31) Masotti, A., Mossa, G., Cametti, C., Ortaggi, G., Bianco, A., Del Grosso, N., Malizia, D., and Esposito, C. (2009) Comparison of different commercially available cationic liposome-DNA lipoplexes: parameters influencing toxicity and transfection efficiency. Colloids Surf., B 68, 136−144. (32) Brown, M. D., Schatzlein, A. G., and Uchegbu, I. F. (2001) Gene delivery with synthetic (non viral) carriers. Int. J. Pharm. 229, 1−21. (33) Hardy, J. G., Kostianen, M. A., Smith, D. K., Gabrielson, N. P., and Pack, D. W. (2006) Dendrons with spermine surface groups as potential building blocks for nonviral vectors in gene therapy. Bioconjugate Chem. 17, 172−178. (34) Farhood, H., Serbina, N., and Huang, L. (1995) The role of dioleoylphosphatidylethanolamine in cationic liposome mediated gene transfer. Biochim. Biophys. Acta 1235, 289−295. (35) Hui, S. W., Langner, M., Zhao, Y. L., Ross, P., Hurley, E., and Chan, K. (1996) The role of helper lipids in cationic liposomemediated gene transfer. Biophys. J. 71, 590−599.

879

dx.doi.org/10.1021/bc200368b | Bioconjugate Chem. 2012, 23, 871−879