Article pubs.acs.org/molecularpharmaceutics
High Gene Transfer by the Osmotic Polysorbitol-Mediated Transporter through the Selective Caveolae Endocytic Pathway Quynh-Phuong Luu,†,‡,§ Ji-Young Shin,§,∥ You-Kyoung Kim,†,⊥ Mohammad Ariful Islam,†,‡ Sang-Kee Kang,†,‡ Myung-Haing Cho,∥ Yun-Jaie Choi,*,†,‡ and Chong-Su Cho*,†,‡ †
Department of Agricultural Biotechnology, Seoul National University, Seoul 151-921, Republic of Korea Research Institute for Agriculture and Life Sciences, Seoul National University, Seoul 151-921, Republic of Korea ∥ Laboratory of Toxicology, College of Veterinary Medicine, Seoul National University, Seoul 151-742, Republic of Korea ⊥ Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing 210009, China ‡
S Supporting Information *
ABSTRACT: Cationic polymers have been the subject of intense research as nonviral gene delivery systems due to several advantages in comparison with viral vectors. However, the nonsimultaneous combination of high transfection efficiency and low cytotoxicity of nonviral vectors for gene delivery has long been an issue for scientists looking into ways to deliver genes into cells. Toward this goal, we designed, synthesized, and evaluated a safe and accelerated gene transfer system through polysorbitol-mediated transporter (PSMT) based on sorbitol diacrylate (SDA) and low molecular weight polyethylenimine (LMW PEI). The PSMT formed stable complexes with plasmid DNA in serum. The nano sizes and spherical shapes of PSMT/ DNA complexes are not toxic, even at a high concentration of PSMT. The higher transfection efficiency of PSMT compared to PEI 25K was observed both in vitro, despite the existence of many hydroxyl groups, and in vivo. These improvements presumably stem from the osmotic property of polysorbitol and endosomal buffer capacity of PEI in PSMT. Most importantly, we confirmed that the selective cavaeolae endocytic pathway played a role in high transfection efficiency by osmotic PSMT-mediated gene delivery. We propose that PSMT is a promising nonviral carrier for the effective gene delivery to cancer cells via synergistic effects derived from rapid cellular uptake through the caveolae endocytic pathway and a high endosomal buffering capacity. KEYWORDS: gene transfer, polysorbitol, transporter, caveolae, endocytic pathway
1. INTRODUCTION The introduction of genes into appropriate cells and tissues has been reported as an efficient new method for treating diseases such as cystic fibrosis, Parkinson’s disease, and cancers.1−3 Treatment of cells with naked DNA is ineffective due to rapid degradation by serum nucleases in the cytoplasm.4−6 Therefore, delivery of genes into the human body requires safe and efficient carriers to protect the therapeutic genes from degradation in the bloodstream and to transfer them to target cells without affecting the immune system. Viral vectors have shown high transfection efficiencies; however, the risk of introducing immune responses, the vector size limitation, and the potential for mutagenesis limit their use in clinical trials.7 Therefore, nonviral vectors have been considered attractive alternatives to viral vectors for gene therapy. Among nonviral gene carriers, polyethylenimine (PEI), with a protonable amino group in every third position and high cationic charge density,7−9 has been widely used. It is highly effective in complex formations with plasmid DNA to © 2012 American Chemical Society
form small nanoparticles that can cross the cell membrane through the endocytic pathway. In addition to its DNA condensing property, this polymer has a significant endosomolytic activity due to the strong buffering capacity in acidic endosomal pH, the so-called proton sponge effect. However, high molecular weight PEI (HMW PEI) induced cell membrane damage because of the high density of positive charges on the surface. Furthermore, the nondegradability of HMW PEI gives rise to high toxicity that reduces the transfection efficiency. To overcome this issue, we studied the combination of low molecular weight PEI (LMW PEI) with degradable cross-linkers such as ester, phosphoester and disulfide for intracellular degradation through hydrolysis at low endosomal pH, enzymatic Received: Revised: Accepted: Published: 2206
February 8, 2012 May 18, 2012 June 18, 2012 June 18, 2012 dx.doi.org/10.1021/mp300072r | Mol. Pharmaceutics 2012, 9, 2206−2218
Molecular Pharmaceutics
Article
anhydrous DMSO. Then, SDA was slowly dropped to PEI solution in a three-neck reaction flask under dry nitrogen purge at a stoichiometric ratio of 1:1 of PEI to SDA. The reaction was performed at 80 °C for 24 h with magnetic stirring. Then, the reaction mixture was dialyzed against distilled water (MWCO: 3500 Da) at 4 °C for 48 h and lyophilized. The final products were stored at −70 °C for the steps. The composition of PEI in PSMT was estimated by dissolving in deuterized DMSO and running proton nuclear magnetic resonance (Avance TM 600, Bruker, Germany). The molecular weight of the copolymer was measured using gel permeation chromatography with multiangle laser light scattering (GPC-MALLS) and 690 nm laser wavelength (Dawn EOS, Wyatt, Santa Barbara, CA, USA). A Shodex OHpak SB-803 HQ (Phenomenex, Torrance, CA, USA) column was used. The column temperature was maintained at 25 °C. The flow rate was 0.5 mL/min, and the mobile phase was 0.5 M ammonium acetate (pH 5.5). 2.3. Physicochemical Characterization of PSMT/DNA Polyplexes. Complexes between PSMT and DNA were prepared in distilled water by gently mixing pGL3 (0.1 μg) with PSMT solution at different N/P ratios. The DNA condensation ability of the PSMT copolymer was confirmed by electrophoresis. The morphology of PSMT/DNA complexes was observed by EF-TEM (JEM 1010, JEOL, Japan). Dynamic light scattering (ELS 8000, Otsuka Electronics, Osaka, Japan) at 90° and 20° scattering angles was used to measure the particle size and surface charge of polyplexes at various N/P ratios. The protection and release of DNA in complexes were measured using electrophoresis. 2.4. Cell Line and Cell Culture. Human cervical carcinoma (HeLa) and lung adenocarcinoma epithelial cell lines (A549) were cultured in Roswell Park Memorial Institute medium (RPMI). Human embryonic kidney (293T) cells were incubated in Dulbecco’s modified Eagle’s medium (DMEM). Both of these media contain 10% fetal bovine serum (FBS), 100 μg/mL streptomycin, and 100 U/mL penicillin. Cells were cultured using standard incubation conditions of 5% CO2 at 37 °C. 2.5. Cytotoxicity Study. The cytotoxicity study was performed in HeLa, A549, and 293T cell lines. The cells were seeded at a density of 10 × 104 cells/well in 1 mL of growth medium containing 10% FBS in 24-well tissue culture plates and incubated for 18−20 h to obtain 80% confluence at treatment. Growth media were removed and exposed to 500 μL of fresh serum-free medium containing PSMT at different polymer concentrations. After 24 h of incubation, cells were treated with 50 μL of MTT solution and incubated for an additional 4 h at 37 °C to form MTT formazan, and were then dissolved in DMSO to measure mitochondrial activity of the cells at 540 nm by ELISA plate reader (GLR 1000, Genelabs Diagnostics, Singapore). 2.6. In Vitro Transfection Efficiency. 2.6.1. Luciferase Activity. PEI 25K and Lipofectamine 2000 were used as positive controls in transfection studies to compare with PSMT. Three different cell lines, A549, HeLa, and 293T, were used for in vitro transfection studies. Cells were grown in their respective media as mentioned earlier. The PSMT/pGL3 (1 μg of pGL3) complexes were prepared at various N/P ratios for 30 min. The cells were transfected with the polyplexes with and/or without serum-containing medium. Following 4 h of transfection with PSMT/pGL3 complexes, the old medium was replaced by 10% serum-containing media and incubated for an additional 24 h. Luciferase activity was measured using a chemiluminometer (Autolumat, LB953; EG&G Berthold, Germany). Relative light units were normalized with the estimation of protein
degradation, and cytosol-specific reductive degradation by glutathione in the cells.10−14 Biological barriers such as binding to the cell surface, traversing the plasma membrane, escaping from endosomes prior to fusing with lysosomes, release to the cytoplasm, and transport to the nucleus limit the transfection efficiency of nonviral vectors. Some molecular transporters based on cellular penetrating peptides (CPP) have been used to accelerate the cellular internalization of biomolecules associated with them and to increase transfection efficiency. Nevertheless, the exact mechanisms underlying their cellular uptake remain poorly understood and are still the object of some controversy.15,16 The sorbitol-based transporter reported by Matti et al. has shown efficient cell penetration and unique intracellular selectivity toward mitochondria.17 Furthermore, high transfection efficiency of DNA and gene silencing of siRNA were obtained using sorbitol as a transporter, which was reported by Higashi et al.18 However, the cellular uptake mechanism of these peptides was not investigated. Our group recently reported a transporter, polysorbitol-based osmotically active transporter (PSOAT), which was synthesized from LMW linear PEI and sorbitol dimethacrylate (SDM) through the Michael addition reaction. Interestingly, the results exhibited accelerated transfection ability with an interesting cellular internalization mechanism. Although the presence of hydroxyl groups in gene carriers such as poly(3-amino-2hydroxypropyl methacrylate) (PAHPMA) and poly(ethylene glycol) (PEG) resulted in the reduction of their transfection efficiency in vitro, the PSOAT actually accelerated transfection efficiency in vitro, even though it had more hydroxyl groups than PAHPMA due to a rapid internalization mechanism.19 The hyperosmotic activity by polysorbitol and proton sponge effect by PEI also revealed a synergistic effect on the improvement of the transfection capacity of PSOAT. In this study, we aimed to determine a clear mechanism for the rapid and selective internalization by polysorbitol-mediated transporter (PSMT) using endocytic pathway inhibitors, because transfection efficiency is closely related to the endocytic pathway20,21 and its osmotically active property affects the endocytosis of nanoparticles.22,23 In the present study, sorbitol diacrylate (SDA), instead of the previously used sorbitol dimethacrylate (SDM), was used as a cross-linker of LMW branched PEI, instead of LMW linear PEI, to reduce the cytotoxicity of the hydrophobic methyl groups in SDM.
2. MATERIALS AND METHODS 2.1. Materials. Branched PEI (MW, 1200 Da; purity, 97%), PEI 25K (MW, 25000 Da; purity, 97%), sorbitol diacrylate (SDA) (MW, 290.3 Da; purity, 98%), MTT reagent [3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide], and anhydrous dimethyl sulfoxide (DMSO) were purchased from Sigma Aldrich (St. Louis, MO, USA). The luciferase reporter 1000 assay system for in vitro transfection and pGL3 control vector with SV-40 promoter and enhancer encoding firefly luciferase were obtained from Promega (Madison, WI, USA). Plasmid pEGFP-N2 with early CMV promoter was obtained from Clontech (Palo Alto, CA, USA). The plasmids were amplified with the competent Escherichia coli strain JM109 and purified using a Qiagen kit. The concentration of purified DNA was determined by measuring UV absorbance at 260 nm. 2.2. Synthesis and Characterization. PSMT composed of LMW branched PEI and SDA was synthesized via the Michael addition reaction as previously reported.22 In brief, LMW branched PEI and SDA were separately dissolved in a vial with 2207
dx.doi.org/10.1021/mp300072r | Mol. Pharmaceutics 2012, 9, 2206−2218
Molecular Pharmaceutics
Article
biotinylated secondary antibody and streptavidin-enzyme conjugate for 25 min each, and DAB was applied accordingly (Dako, Denmark). The slides were then viewed under a light microscope at 200× magnification. 2.8. Screening of Toxicity in Vivo. Twelve week old male C57BL/6 mice were randomly divided into 3 groups: control, PSMT/pDNA, and PEI25K/pDNA. Complexes were prepared as described previously, and inhalation was performed 6 times in 2 weeks (SNU-120426-5). Animals were sacrificed 2 days after last exposure. For the bronchoalveolar lavage fluid analysis, the tracheas were cannulated and the lungs were lavaged 5 times with 0.6 mL (total volume of 3 mL) of steriled 0.9% NaCl following anesthetization with zolazepam/tiletamine/xylazine solution. First and second aliquots were combined and centrifuged (400g, 10 min, 4 °C), and the cell-free supernatants were stored at −70 °C for further analysis. ELISA kits for lactate dehydrogenase (LDH) and Nacetyl-beta-D-glucosaminidase (NAG) were purchased from MyBioSource (San Diego, CA), and samples were analyzed following the manufacturer’s protocol. Other aliquots were combined, and supernatants were discarded after centrifugation. All pellets were resuspended in 200 μL of modified Eagle’s medium (MEM), and the total bronchoalveolar lavage fluid (BALF) inflammatory cell numbers were counted with a hemocytometer (Marienfeld, Lauda-Königshofen, Germany) after trypan blue staining (Invitrogen, Carlsbad, CA). Differential cell counts (minimum of 5,000 cells/slide) were performed on cytospin-prepared slides (FisherScientific, Milwaukee, WI) with cytocentrifuge (2000 rpm, 5 min, 4 °C) and stained with DiffQuik (Sysmex, Kobe, Japan). Three replicate slides were used for analysis. For the histopathological analysis, lungs were fixed with 10% neutral formalin after lavage and sections were made in 4 μm thickness. Slides were stained with hematoxylin and eosin, and inflammatory lesions were graded into 4 groups: control, mild, moderate, and severe.38 2.9. Mechanistic Study. 2.9.1. Endocytic Mechanism. A549 cells were seeded in 24-well plates at a density of 1 × 105cells/well and incubated for 20 h to reach 80% confluency. PSMT/pGL3 and PEI/pGL3 polyplexes were prepared at a pGL3 concentration of 1 μg/mL with a N/P ratio of 20, according to an earlier procedure. Then, the cells were preincubated with various endocytic pathway inhibitors such as chlorpromazine, β-methyl cyclodextrin, genistein, and wortmannin at various inhibitor concentrations in serum-free medium for 1 h before transfection. After 4 h, the serum-free medium was replaced with 10% serum containing medium and incubated for an additional 24 h. Transfection efficiency was assayed by measuring the luciferase activity as described earlier. The cytotoxicities of the investigated endocytic pathway inhibitors were also checked by a MTT assay. 2.9.2. Proton Sponge Effect. Similar to the endocytic pathway inhibition study, A549 cells were seeded in a 24-well plate at a density of 1 × 105cells/well and cultured for 20 h to reach 80% confluency. The cells were incubated with bafilomycin A1 (a specific inhibitor of vacuolar type H+ ATPase) for 10 min before transfection with polymer/DNA complexes. The effect of bafilomycin on transfection efficiency was evaluated using the luciferase assay, as described above.
concentration in the cell extract using a BCA protein assay kit (Pierce Biotechnology, Rockford, IL, USA). 2.6.2. Flow Cytometry. A549 cells were seeded in 6-well plates at an initial density of 30 × 104 cells/well in 2 mL of growth medium and incubated in standard incubation conditions. After 6 h of transfection with different N/P ratios of PSMT/GFP (3 μg of GFP) complexes, the old medium was replaced with fresh serum-containing medium and cells were further incubated for 24 h at 37 °C in 5% CO2. The transfection efficiency was checked by scoring the percentage of cells expressing GFP using Fluorescence Activated Cell Sorter (FACS) Calibur system (San Jose, CA, USA). 2.7. In Vivo Transfection Efficiency. Wild-type male C57BL/6 mice (4 mice/group) were obtained from the breeding colony of Human Cancer Consortium National Cancer Institute (Frederick, MD, USA) and kept in the laboratory animal facility with temperature and relative humidity maintained at 23 ± 2 °C and 50 ± 20%, respectively, under a 12 h light/dark cycle. All of the experimental protocols were reviewed and approved by the Animal Care and Use Committee at Seoul National University (SNU-110729-1). PSMT/tGFP complexes were prepared by combining 0.6 mg of pCMV-AN with turbo GFP (Origene, Rockville, MD, USA) and 0.288 mg of PSMT at a N/P ratio of 20, optimized from the in vitro study. Under gentle vortexing, prepared DNA was added into the polymer solution dropwise and the total volume was 30 mL. The complexes were then incubated at room temperature for 30 min before use. For inhalation, animals were randomly divided into 4 groups; control, PSMT, PSMT/tGFP, and PEI 25K/tGFP (N/P 5). The animals were placed into a nose-only exposure-chamber (NOEC) for 30 min. After 48 h, the mice were sacrificed and their lungs were slowly flushed with 3 mL of PBS through the trachea to minimize the background signal of GFP from red blood cells. For the Western blot assay, frozen lung tissues were homogenized and protein concentrations were measured with a Bradford kit (Bio-Rad, Hercules, CA, USA). Equal amounts of protein (30 μg) were loaded onto the SDS gels and separated accordingly. Proteins were transferred to nitrocellulose membrane and preblocked with 5% skim milk (w/w 1× TTBS) for an hour at room temperature. Monoclonal anti-GFP was produced using a general method described elsewhere and diluted 1:10 at 5% skim milk. Anti-beta-actin was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA), diluted 1:5000, and incubated overnight at 4 °C. Secondary antibodies conjugated to horseradish peroxidase (HRP) (Invitrogen, Carlsbad, CA, USA) were applied according to the manufacturer’s protocols. Bands of interest were obtained with a luminescent image analyzer LAS3000 (Fujifilm, Tokyo, Japan). For fluorescence microscopy observations, the lungs were fixed in 4% paraformaldehyde for 12 h and preserved in 30% sucrose for 48 h at 4 °C. Lungs were embedded with OCT compound (Sakura, Torrance, CA, USA) at
