2262
Bioconjugate Chem. 2009, 20, 2262–2269
Snail1 Down-Regulation using Small Interfering RNA Complexes Delivered through Collagen Scaffolds Rosa Vin˜as-Castells,†,‡ Carolyn Holladay,† Andrea di Luca,† Victor Manuel Dı´az,‡,§ and Abhay Pandit*,† Network of Excellence for Functional Biomaterials, National University of Ireland, Galway, Ireland, Programa de Recerca en Cancer, IMIM-Hospital del Mar, Barcelona, Spain, and Departament de Cie`ncies Experimentals i de la Salut, Universitat Pompeu Fabra, Barcelona, Spain. Received June 10, 2009; Revised Manuscript Received October 16, 2009
Control of gene expression via small interfering RNA has enormous potential for the treatment of a variety of diseases, including cancer and Huntington’s disease. However, before any therapies can be developed, effective techniques for controlled delivery of these molecules must be devised. In this proof-of-concept study, small interfering RNA was complexed with a polymer and loaded into a biomaterial scaffold. The scaffold was introduced primarily to control the release of the complexes, and the polymer was introduced to improve the transfection efficiency. An optimal dose and complexation ratio were selected, at which more than 50% down-regulation of the target gene Snail1 was observed in two-dimensional culture. Delayed release of the complexes was observed, and significant sustained down-regulation of Snail1 was seen in a three-dimensional scaffold system after 7 days. Thus, the use of the scaffold altered the transfection profile significantly, demonstrating the feasibility of a collagen scaffold as a controlled release system for delivery of small interfering RNA-dendrimer complexes.
INTRODUCTION Cancer manifests through six basic alterations in cell behavior that lead to malignant growth: self-sufficient growth signaling, insensitivity to growth-inhibitory signals, evasion of apoptosis, unlimited reproduction, sustained angiogenesis, and metastasis (1). In epithelial tumor progression, metastasis can occur when loosely attached cancerous cells detach and migrate via the bloodstream to other tissues and organs. E-cadherin (EC), which is localized in the adherens junctions between epithelial cells, has been found to be down-regulated in epithelial tumors that undergo a process termed “epithelial to mesenchymal transition” (EMT). Decreasing EC levels damage the adherens junctions, and therefore low EC is hypothesized to be among the factors promoting metastasis. The transcription factor Snail1 represses EC production (2, 3). Therefore, elevated levels of Snail1 decrease the strength of adherens junctions and thus increase tumor metastasis (4). Down-regulation of Snail1 is thus hypothesized to decrease metastasis of epithelial tumors. RNA interference is a technique used in gene therapy to down-regulate expression of targeted genes. Short (21-23 nucleotides), double-stranded RNA sequences are incorporated into a RNA-induced silencing complex (RISC) and then can bind complementary mRNA strands. The bound mRNA is degraded, thus preventing translation of the gene in question (5, 6). However, to be effective, the small interfering RNA (siRNA) must reach the RISC and selectively bind the targeted sequence. Thus, the siRNA must be internalized by the targeted cell type, escape the endosome or lysosome, avoid incorrect intracellular routing, and still be bioactive when it reaches the RISC (7). Many of the challenges cited here are similar to those encountered for plasmid gene delivery (8). * To whom correspondence should be addressed. Phone: +353 91 492758. E-mail:
[email protected]. † National University of Ireland. ‡ IMIM-Hospital del Mar. § Universitat Pompeu Fabra.
siRNA is not biologically stable, and even in Vitro, chemical modification of the sequence or complexation with a carrier system is required to prevent rapid degradation. While chemical modifications can increase the half-life of the siRNA, they do not significantly increase internalization. Thus, carrier systems are an appealing alternative. Both in Vitro and in ViVo delivery of siRNA using cationic liposomes and polymers has been described (7). Many of these systems are theoretically suitable for conjugation to targeting molecules that can guide them to the targeted tissues or cell types. These carriers may also have the potential to improve endosomal escape and intracellular routing (9). Early transfection studies using the polyamidoamine (PAMAM) dendrimeric system indicated that PAMAM could be effectively complexed with a variety of forms of nucleic acids, and that increased complexation could be achieved by modifying the chemistry of the interaction (10). However, later studies focused on modified PAMAM dendrimers (11-13). In many of these studies, a partially degraded version, SuperFect (dPAMAM), has been shown to be among the most effective transfection reagents on the market (10-20). These compounds have been used both in Vitro and in ViVo with significant success (10, 11, 18, 21, 22). Biomaterials have also been described as gene transfer agents, both for plasmids and for siRNA. Atelocollagen, in particular, has been found to be an effective complexing agent for siRNA (23). Atelocollagen-siRNA complexes were found to be efficiently internalized and resistant to nucleases that would ordinarily destroy exogenous nucleic acids (24). Furthermore, treatment with these atelocollagen-siRNA complexes resulted in effective down-regulation of targeted genes both in Vitro and in ViVo (24, 25). The relative success of dendrimer-siRNA complexes in protecting siRNA, increasing its level of internalization, and assisting its endosomal escape make these agents promising vectors. However, it is well-known that the time scale of siRNA transfection can be problematic; for greater than 90% inhibition of a target gene, biologically active siRNA must be able to act for longer than three times the half-life of the protein in question
10.1021/bc900241w 2009 American Chemical Society Published on Web 11/16/2009
Snail1 Down-Regulation using siRNA
Bioconjugate Chem., Vol. 20, No. 12, 2009 2263
Table 1. Details of Anti-Snail1 siRNA Sequences versions of anti-Sna1 siRNA 1 2 3 4
sequence UGACCUCGCUGUCCGAUGAUU (sense) 5′-PUCAUCGGACAGCGAGGUCAUU (antisense) GAUCUUCAACUGCAAAUAUUU (sense) 5′-PAUAUUUGCAGUUGAAGAUCUU (antisense) CAAACCCACUCGGAUGUGAUU (sense) 5′-PUCACAUCCGAGUGGGUUUGUU (antisense) GCCGGAAGCCCAACUAUAGUU (sense) 5′-PCUAUAGUUGGGCUUCCGGCUU (antisense)
(7). Thus, a combination of dendrimer-siRNA complexes with an atelocollagen scaffold is proposed. It is hypothesized that combining the high transfection efficiency of the dendrimer and the slow release mechanism from the collagen will allow efficient, longer-term gene down-regulation. Optimization of dendrimer-siRNA complexation was required prior to any transfection experiments. Characterization of the chemical interactions between the two components, followed by basic internalization experiments was used to choose the formulations tested for transfection ability. Basic transfection experiments were then used to determine the optimal complex formulation for use in the final set of experiments. The final experiments test the hypothesis that the combination of dendrimer-siRNA complexes with an atelocollagen scaffold can be used to control the release rate of the complexes and the time course of siRNA-mediated protein down-regulation.
EXPERIMENTAL PROCEDURES Anti-Snail1 siRNA. Anti-Snail1 siRNA was purchased from Dharmacon (Chicago, IL). The siRNA was specific to the Snail1 homologue 1 (Drosophila), ID20613 for mouse (SNAI1) ONTARGETplus SMARTpool. It was supplied as a pool of four sequences (Table 1). The dPAMAM (SuperFect) used to complex siRNA for delivery was purchased from Qiagen (Germany). Preparation of Complexes. Control siRNA (siGLO anticyclophilin B siRNA ON-TARGETplus control siRNA, Dharmacon, Chicago, IL) was combined with dPAMAM at a variety of amine to phosphate (N/P) ratios. The dPAMAM concentration was approximately 2.8 mM amine groups (2.8 nmol/µL). Thus, 1 µg of siRNA combined with 1 µL of dPAMAM yielded complexes at an N/P ratio of 1:1. The siRNAs were supplied lyophilized and were resuspended in an RNase-free buffer (20 mM KCl, 6 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) at pH 7.5 and 0.2 mM MgCl2) at a concentration of 20 µM. Appropriate volumes of dPAMAM solution were combined with the siRNA solution in sterile TE buffer. The siRNA-containing solutions were kept on ice to minimize siRNA degradation. Confirmation of Complexation. Complexation was confirmed using two methods. For dye exclusion studies, complexes prepared at N/P ratios of 1-10 were loaded into a polyacrylamide gel. The gel was run at 150 V for approximately 1 h, then removed from the electrophoresis apparatus and stained with SYBRGold nucleic acid gel stain (Invitrogen, Paisley, U.K.) for 45 min. The gel was imaged at 254 nm in a transilluminator (Universal Hood II, BioRad, Hercules, CA). For the TEM studies, complexes were prepared at varying N/P ratios then imaged with a Hitachi H-7500 transmission electron microscope (Tokyo, Japan). Cell Culture. NIH3T3 fibroblast cells were used for all cell culture experiments since they are well-characterized and readily available. Furthermore, they constitutively express high levels of Snail1, which is a good model to use to revert or reduce EMT. Twenty-four hours before siRNA was delivered, 2 × 105 NIH3T3 cells were plated in 6 well plates. After 24 h incubation, which was necessary to allow cells to attach and start proliferating, the cells were washed with phosphate-buffered saline (PBS) and fluorescently tagged control siRNA complexes were deliv-
ered at N/P ratios of 1-10 and at doses of 50, 75, 100, and 200 nM. The transfected plates were then returned to the incubator and left for 24 h. Internalization Studies. The internalization studies included fluorescent microscopy and fluorescence-activated cell sorting (FACS) analysis. In the samples intended for microscopy, the transfected cells were washed, fixed, and stained with 4′,6diamidino-2-phenylindole (DAPI). Images were taken with DAPI and rhodamine filter sets and superimposed over phase contrast images. For FACS analysis, cells were washed twice with PBS, trypsinized, centrifuged, washed again, and resuspended in media. The resuspended cells were analyzed with a FACScalibur (Becton Dickinson, Franklin Lakes, NJ) equipped with CellQuest software, V3.3. The threshold was selected based on unmodified NIH3T3 cells (control). Cytotoxicity Studies. The cytotoxicity of the complexes was evaluated by measuring the metabolic activity of transfected wells. The metabolic activity was assessed with the AlamarBlue (AB) assay (Invitrogen, U.K.). Briefly, 24 h after transfection, cells were washed with PBS, a dilution of 1:10 of AB in PBS was added to the wells, and the plate was returned to the 37 °C incubator for 3 h. The AB solution was then transferred into a 96 well plate, and the fluorescence was read at 590 nm. Semiquantitative Reverse Transcription Polymerase Chain Reaction (RT-PCR). Initial transfection experiments indicated that a single delivery of siRNA complexes, while allowing high levels of internalization, did not measurably decrease protein expression levels. Thus double delivery of the complexes was used in the transfection experiments. The transfection protocol used was similar to that described earlier, except that 24 h after the first transfection, the same dose and N/P ratio of siRNA complexes was added and the cells were left a further 24 h. In this case, the control condition was control siRNA complexed at an N/P of 8. It was also delivered twice. Analysis of mRNA levels was carried out by RT-PCR using 0.25 µg of total RNA per sample. RNA was extracted using PureLink Micro-to-Midi total RNA purification system (Invitrogen, Carlsbad, CA) and quantified using a Nanodrop system (ND-1000 spectrophotometer with the analysis software V3.3.0). PCR analysis was performed using SuperScript OneStep RTPCR kit with Platinum Taq (Invitrogen, Carlsbad, CA) and using oligonucleotides corresponding to the sequence +104/+125 and +290/+31 in order to analyze the RNA levels of Snail1 in the cells. The following oligonucleotides were used for the analysis of Snail1: 5′-GGCGGATCCACCATGCCGCGCTCCTTCCTG3′ for forward transcription and 5′-CAAGATATCCGCCTCCGGAGCA-3′ for reverse transcription. The endogenous housekeeping gene HPRT (hypoxanthine-guanine phosphoribosyl transferase) was used to normalize the RNA quantifications, using the oligonucleotides 5′-GGCCAGACTTTGTTGGATTTG3′ for forward transcription and 5′-TGCGCTCATCTTAGGCTTTGT-3′ for reverse transcription. All primers were obtained from Invitrogen (UK) and the thermocycler was from Biometra (T-Gradient Thermoblock, Go¨ttingen, Germany). Western Blotting. Analysis of protein expression levels was conducted using Western blotting. Twenty-four hours after the second delivery of siRNA, the cells were washed with PBS and lysed using 150 µL of total lysis buffer (20 mM HEPES, pH 7.8, 25% glycerol, 420 mM NaCl, 1% Triton X-100, 1.5 mM MgCl2, 0.2 mM EDTA, and protease inhibitors). The lysate was scraped from the plate, placed in a centrifuge tube and agitated at 4 °C for 10 min. The tubes were centrifuged to precipitate insoluble particles. Total protein quantity was estimated using a standard kit (Bio-Rad Laboratories, Hercules, CA). Bovine serum albumin was used as a standard. Snail1 protein levels were measured via Western blotting. Annexin II was used as
2264 Bioconjugate Chem., Vol. 20, No. 12, 2009
loading control. Anti-Snail1 antibody was monoclonal (26), and antiannexin II was rabbit polyclonal (27). Scaffold Preparation. One milligram cross-linked atelocollagen scaffolds were prepared as described by Holladay et al. (28). Briefly, 333 µL of 3 mg/mL atelocollagen solution was lyophilized, then cross-linked with 1-ethyl-3-(3-dimethylaminopropyl) (EDC) and N-hydroxylsuccinimide (NHS) at an EDC-NHS-collagen carboxyl group ratio of 5:5:1. The scaffolds were then washed thoroughly and freeze-dried again. siRNA-dPAMAM complexes (200 nM) in a total volume of 30 µL were loaded into the dry 1 mg scaffolds. Elution Studies. Release of the siRNA-dPAMAM complexes was evaluated using the fluorescence from the control siRNA, which is labeled with rhodamine. Briefly, complexes were prepared at N/P ) 8 then loaded onto the cross-linked scaffolds in a black 96 well plate. The loaded scaffolds were incubated for 2 h at room temperature to allow the complexes to associate with the scaffold. After this time, cell culture media containing 10% FBS was added to each scaffold. The scaffolds were removed from the media and placed in fresh media. At each time point, this process was repeated. At the end of the experiment, the siRNA content of the solutions was quantified by measuring the fluorescence with a Varioskan Flash spectral scanning multimode reader (Thermo Scientific, Vantaa, Finland) and the cumulative release of siRNA-dPAMAM complexes from the scaffold was calculated. Scaffold-Based Transfection. siRNA-dPAMAM complexes (200 nM) in a total volume of 30 µL were loaded into dry 1 mg scaffolds. The loaded scaffolds were incubated at room temperature for 2 h to allow the complexes to associate with the collagen. NIH3T3 cells (7 × 104) were added to the scaffolds after this incubation. After 24 h, the scaffolds were transferred to a new plate to ensure that all of the cells were attached to the scaffold and not to the plate. New media was added to the scaffolds in the new plate, and the media was changed every 2 days until the end of the experiment. At given time points (24 h, 48 h, and 7 days), the scaffolds were removed from the wells, placed in centrifuge tubes, and centrifuged for 5 min at 3000 rpm, and the supernatant was discarded. The scaffolds were stored at -20 °C. When the samples from all of the experiments were ready, the tubes were thawed, and 25 µL of 2× Laemmli sample buffer was added to each sample. The samples were vortexed for 5 min, then boiled for 5 min at 100 °C. Western blotting, as previously described, was then carried out. Quantification was performed with densitometry using annexin II levels as a normalizing control. Statistics. Graphical results are expressed as mean ( standard deviation. Where applicable, statistical significance was assessed using the analysis of variance (ANOVA). P values of
