Efficient Delivery of Bcl-2-Targeted siRNA Using Cationic Polymer

Dec 15, 2008 - Imperial College London, Hammersmith Hospital Du Cane Road London, W120NN, United Kingdom. Received June 27, 2008; Revised ...
0 downloads 0 Views 4MB Size
Biomacromolecules 2009, 10, 41–48

41

Efficient Delivery of Bcl-2-Targeted siRNA Using Cationic Polymer Nanoparticles: Downregulating mRNA Expression Level and Sensitizing Cancer Cells to Anticancer Drug Cyrus W. Beh,† Wei Yang Seow,†,§ Yong Wang,† Ying Zhang,† Zhan Yuin Ong,‡ Pui Lai Rachel Ee,‡ and Yi-Yan Yang*,† Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, Singapore 138669, Singapore, Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore, 117543, Singapore, and Department of Immunology, Faculty of Medicine Imperial College London, Hammersmith Hospital Du Cane Road London, W120NN, United Kingdom Received June 27, 2008; Revised Manuscript Received November 1, 2008

In this study, cationic nanoparticles self-assembled from the amphiphilic copolymer poly(N-methyldietheneamine sebacate)-co-[(cholesteryl oxocarbonylamido ethyl) methyl bis(ethylene) ammonium bromide] sebacate) (P(MDSco-CES) were synthesized and used to deliver Bcl-2 targeted siRNA into HepG2, HeLa and MDA-MB-231 cell lines, and downregulate Bcl-2 mRNA expression levels. Confocal microscopic studies show that the nanoparticles were able to complex with siRNA and deliver it inside the cells efficiently, but siRNA was easily dissociated from the complexes in the cytoplasm for its biological functions. Bcl-2 mRNA expression levels as low as 10% were achieved after treatment with nanoparticle/siRNA complexes. The downregulation efficiency of Bcl-2 mRNA level was similar to that mediated by Lipofectamine but higher than that induced by PEI. PEG was also conjugated to siRNA via a cleavable disulfide bond, and nanoparticle/siRNA-PEG complexes showed no significant protein adsorption as compared with 26 and 17% for blank nanoparticles and nanoparticle/siRNA complexes, respectively. The presence of serum caused slight aggregation of nanoparticle/siRNA or nanoparticle/siRNA-PEG complexes. However, the size of the complexes was still below 250 nm after being incubated in PBS containing 10% serum for 4 h. On the other hand, PEGylated siRNA delivered by the nanoparticles downregulated Bcl-2 mRNA expression level in the cells as efficiently as unmodified siRNA. Bcl-2 protein was also downregulated efficiently by nanoparticle/siRNA complexes in all cell lines tested. The downregulation of Bcl-2 mRNA or Bcl-2 protein did not show significant cell death in the tested siRNA and polymer concentration range. However, the delivery of siRNA sensitized HeLa cells to paclitaxel treatment, yielding significant improvement over the untreated cells (p < 0.05). These cationic nanoparticles may be potentially employed to downregulate Bcl-2 expression and sensitize cancer cells to anticancer drugs for more efficient chemotherapy.

Introduction RNA interference (RNAi) is a phenomenon discovered by Andrew Fire and Chris Mello less than a decade ago, in which post-transcriptional gene silencing was achieved in Caenorhabditis elegans worms by the introduction of double-stranded RNA (dsRNA) molecules.1 It has been confirmed that the dsRNA molecules are processed in vivo to short-interfering RNA duplexes (siRNA) of around 21-23bp. That the two have received the Nobel Prize for Physiology or Medicine 2006 is testimony of the importance of this discovery. The potency of gene silencing is such that even a few molecules of siRNA are able to result in extensive silencing, and this silencing can even be spread to other tissues.2 Besides its utility as a tool for molecular biology (achieving easy gene knockdowns), RNAi has obvious applications in cancer treatment.3 Indeed, several groups have targeted various oncogenes and shown the efficacy of RNAi in tumor suppression, either through the triggering of apoptosis or by sensitizing cells to conventional chemotherapy.3-6 * To whom correspondence should be addressed. Tel.: 65-68247106. Fax: 65-64789084. E-mail: [email protected]. † Institute of Bioengineering and Nanotechnology. ‡ National University of Singapore. § Imperial College London.

The existing strategies to cause RNAi can be broadly divided into two categories, namely, expressed short-hairpin RNAs (shRNAs), with precursors introduced into cells using viruses as well as delivered siRNAs. The use of viruses, though effective, results in immunogenic response, which limits its potential for clinical applications. Therefore, as of now, the majority of proposed clinical applications of RNAi employ siRNA duplexes, delivered into cells by nonviral vectors.7 Bcl-2 protein is an attractive target for gene therapy because it exhibits antiapoptotic activity and is overexpressed in several types of cancers. Overexpression of Bcl-2 is also implicated in the resistance of cancer cells to a variety of anticancer agents.8-10 Briefly, the Bcl-2 protein is postulated to block the release of cytochrome C after the initiation of apoptosis, which prevents the downstream propagation of the death signal, thereby promoting cell survival.11 It has been shown that cancer cell apoptosis or chemosensitization can be effected by silencing this gene.4-6 The chemosensitization of the cancer cells by RNAi has been examined using paclitaxel due to its excellent therapeutic effects.12 Different methods of delivery have recently been proposed by various investigators, including conjugating siRNA to cholesterol, encapsulating within functionalized liposomes, various aptamer-based strategies, as well as antibodies targeting.7 However, cholesterol and liposomes are nonselective, while

10.1021/bm801109g CCC: $40.75  2009 American Chemical Society Published on Web 12/15/2008

42

Biomacromolecules, Vol. 10, No. 1, 2009

Beh et al.

Figure 1. Chemical structure of cationic amphiphilic copolymer P(MDS-co-CES).

aptamers are relatively new and limited in terms of the targets available. Finally, although antibody-based targeted therapy is both mature and selective, it compromises on either stability (such as Fab) or safety (as in whole antibodies).13 On the other hand, cationic polymers and lipids have received increasing attention as vectors for siRNA delivery because they not only deliver high payloads, but are also amenable to modifications to yield suitable sizes, as well as selectivities using receptor ligands. For example, in vitro and in vivo delivery of siRNA molecules have been attempted using folate-modified polyethyleneimine (PEI),14 chitosan,15 cholesteryl oligoarginine,16 and cationic liposomes.17 In addition, PEGylation of siRNA via disulfide bond has been reported to provide improved stability in a serum-containing environment when compared to unmodified siRNA, leading to an enhanced gene silencing effect when complexed with a cationic fusogenic peptide (KALA) or PEI.18,19 Lactosylated PEG-siRNA (RecQL1 targeted) conjugate has also been synthesized, and poly(L-lysine)/lactosylated PEG-siRNA complexes have demonstrated efficient growth inhibition of hepatic multicellular tumor spheroids.20 To our knowledge, only one paper was reported on use of cationic micellar nanoparticles for siRNA delivery.21 We have recently developed novel cationic polymer micelles (i.e., core/shell structured nanoparticles) self-assembled from poly(N-methyldietheneamine sebacate)-co-[(cholesteryl oxocarbonylamido ethyl) methyl bis(ethylene) ammonium bromide] sebacate) (P(MDSco-CES)) and demonstrated that these nanoparticles induced high gene transfection efficiency both in vitro and in vivo and were able to deliver drug and DNA simultaneously.22,23 In addition, they can effectively deliver biologically active proteins into various cancer cell lines.24 In this study, Bcl-2 targeted siRNA and PEG-conjugated siRNA were delivered into HepG2 human liver carcinoma, HeLa human cervical, and MDA-MB-231 human breast cancer cell lines using these cationic nanoparticles. Bcl-2 mRNA expression levels in the cells were studied using real-time reverse-transcription polymerase-chain reaction (RTPCR) in comparison with Lipofectamine and PEI, and downregulation of Bcl-2 protein was also investigated via western blot assays. Viability of cells was analyzed by MTT assay after incubation with nanoparticle/siRNA or nanoparticle/siRNA-PEG complexes. The combined delivery of Bcl-2 targeted siRNA and paclitaxel was explored to determine the efficiency of chemosensitization after treatment with siRNA. In addition, protein adsorption on the nanoparticle/siRNA complexes after PEGylation, and the effect of PEGylation on siRNA delivery were evaluated. Aggregation of nanoparticle/siRNA or nanoparticle/siRNA-PEG complexes was studied by monitoring particle size change as a function of time.

Experimental Section Materials. 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) sodium salt buffer solution, N-succimidyl 3-(2-pyridyldithio)propionate (SPDP), paclitaxel, polyethyleneimine (PEI, branched, Mw ) 25 kDa), dithiothreitol (DTT), and 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide solution (MTT) were purchased from Sigma-Aldrich. Amine-terminated polyethylene glycol (PEG, Mw ) 2 kDa) was purchased from Boehringer Ingelheim. Thiol-modified siRNA was ordered from Dharmacon (U.S.A.), with the following sequence: sense 5′-thiol-GUA CAU CCA UUA UAA GCU G-dTdT; antisense CAG CUU AUA AUG GAU GUA C-dTdT. Spectra/Por dialysis membranes with molecular weight cutoff (MWCO) of 1 and 2 kDa were purchased from Biotech. RNeasy kits and QiaShredder spin columns, and AllStars negative control siRNA with and without Alexa Fluor 546 label were bought from Qiagen. Superscript III First-Strand Synthesis kits, diethylpyrocarbonate (DEPC)-treated water, Lipofectamine RNAiMax, Dulbecco’s modified Eagle’s medium (DMEM) and Leibovitz’s L-15 medium were obtained from Invitrogen. Desalting columns (PD-10) were purchased from Fluka. RNasin RNase inhibitor, luciferase substrate, and 5× lysis buffer from Promega, RNase Zap from Ambion and Bcl-2 PCR primers from Sigma-Aldrich were used as received. SYBR Green I Mastermix (2×) was purchased from Stratagene and used according to manufacturer’s instructions. The BCA protein assay kit was purchased from Pierce. Custom oligonucleotides were purchased from Research Biolabs Ltd. and used as PCR primers. Mouse monoclonal primary antibodies for Bcl-2 (C-2) and R-tubulin (B-5-1-2) were purchased from Santa Cruz Biotechnology and Sigma-Aldrich respectively. Goat antimouse IgG conjugated to horseradish peroxidase (HRP) was purchased from Bio-Rad Laboratories Pte. Ltd. HepG2, HeLa and MDA-MB-231 human cancer cell lines were purchased from ATCC. The polymer P(MDS-co-CES) was freshly synthesized for the experiments. N-Methyldiethanolamine (98%), sebacoyl chloride (97%), triethylamine (TEA), p-toluenesulfonyl chloride (99%), cholesteryl chloroformate (98%), and 2-bromoethylamine hydrobromide (98%) were purchased from Sigma-Aldrich. N-Methyldiethanolamine and sebacoyl chloride were purified by vacuum distillation and stored for subsequent use. TEA was first treated with p-toluenesulfonyl chloride and distilled. Both TEA and tetrahydrofuran (THF) were then further dried with sodium and always freshly distilled under nitrogen prior to reaction. Dichloromethane (DCM) was stored with activated molecular sieves (3 Å) until use. All other reagents required in the polymer synthesis were used as received. Methods. Synthesis of P(MDS-co-CES). The synthetic scheme for P(MDS-co-CES) has been elaborated elsewhere.22 However, some modifications were made during the synthesis of this current batch of polymer. Briefly, the main chain, poly(N-methyldietheneamine sebacate) (PMDS), was first produced by condensation polymerization between N-methyldiethanolamine and sebacoyl chloride. Excess triethylamine was used to remove hydrochloride and limit protonation of the tertiary amine. Next, cholesteryl chloroformate was allowed to react with

Efficient Delivery of Bcl-2-Targeted siRNA

Biomacromolecules, Vol. 10, No. 1, 2009

43

Figure 2. MALDI-TOF mass spectrum of PEGylated siRNA. The guide and sense strands unwind during the ionization, giving the unmodified guide fragment with a mass of 6675 g/mol, and the modified strand centered on around 9000 g/mol, with a ∆m ∼ 45 Da (see inset), corresponding to the mass of an ethylene glycol repeat unit.

Figure 3. Electrophoretic mobility of siRNA in the nanoparticle/siRNA complexes (A) and siRNA-PEG in the nanoparticle/siRNA-PEG complexes (B), and PEI/siRNA complexes prepared at various N/P ratios as specified (C). Lipofectamine was added to 22 nM siRNA at the stated volumes (1-2 µL) in 1 mL of Opti-MEM (D).

2-bromoethylamine hydrobromide in an amidation reaction. The resulting hydrophobic N-(2-bromoethyl) carbarmoyl cholesterol (Bechol) was then grafted onto the hydrophilic poly(N-methyldietheneamine sebacate) main chain through a quaternization reaction to obtain the final product P(MDS-co-CES) (Figure 1). The final reaction was conducted in anhydrous toluene and refluxed for 24 h under argon instead of 48 h, as reported previously.22 The structures of PMDS, Bechol, and P(MDS-co-CES) were confirmed with 1H NMR (Bruker Avance 400 Spectrometer, 400 MHz). The molecular weight (Mw) and distribution of the polymer was estimated by gel permeation chromatography (GPC) (Waters 2690, MA, U.S.A.) with a differential refractometer detector (Waters 410, MA, U.S.A.). The mobile phase

used was THF with a flow rate of 1 mL/min. Weight average molecular weight as well as polydispersity index were calculated from a calibration curve using a series of polystyrene standards (Polymer Laboratories Inc., MA, U.S.A., with molecular weight ranging from 1350 to 151700). Conjugation of siRNA with PEG. The siRNA-PEG conjugate was formed via the reducible heterobifunctional cross-linker, SPDP. SPDP (6.2 mg, 2 µmol) was reacted with 10 times molar excess amine-terminated PEG (Mw ) 2000, 20 µmol) at room temperature for 1 h. PEG-NH2 was dissolved in 4.5 mL of phosphate-buffered saline (PBS, pH 7.4), while SPDP was dissolved in 0.5 mL of absolute ethanol and added dropwise to the PEG-NH2 solution with stirring under argon. The PEG-SPDP intermediate was dialysed against deionized (DI) water with a continuous flow to remove the unreacted SPDP in a dialysis membrane with MWCO of 1000 Da for approximately 5 h. Unreacted amine-terminated PEG was removed in the next step. To determine the efficiency of the first step of the reaction, the products were reacted with a large excess of DTT in PBS (pH 7.4), a small thiol-containing compound that reacts rapidly with the PEG-SPDP through thiol-exchange, releasing pyridine-2-thione. Pyridine-2-thione has an extinction coefficient of 8.08 × 103 M-1 cm-1 at 343 nm25 and can thus be used to determine the concentration of PEG-SPDP in the solution. siRNA (Mw ) 13 kDa, 5 mg, 0.385 µmol) was reacted with 1 mg of PEG-SPDP (Mw ) 2.3 kDa, 0.435 µmol) in 5 mL of RNase-free water. An aliquot sample (0.5 mL) was taken at 1 h intervals, and the absorbance at 343 nm was tested to determine the concentration of pyridine-2-thione and, hence, the extent of disulfide exchange. This is possible because the siRNA does not absorb at this wavelength.26 The conjugated product was then purified using a PD-10 desalting column

44

Biomacromolecules, Vol. 10, No. 1, 2009

Beh et al.

Figure 4. Protein adsorption of blank nanoparticles and nanoparticle/ (siRNA+siRNA-PEG) complexes fabricated at various molar percentages of siRNA-PEG. (N/P 50).

Figure 5. Average size of blank nanoparticles, nanoparticle/siRNA, and nanoparticle/siRNA-PEG complexes fabricated at N/P 50 after being incubated in PBS with 10% serum for different periods of time.

(Mw cutoff: 5000 Da), eluting with 1× HEPES buffer to remove unreacted amine-terminated PEG, SPDP, and byproducts. The eluted fractions were tested for purity and concentration using a NanoDrop spectrometer (ND1000, NanoDrop Technologies, DE, U.S.A.). Successful conjugation of siRNA with PEG was evidenced by matrix-assisted laser desorption ionization of time-of-flight (MALDI/ TOF) mass spectrometry (Autoflex II, Bruker Daltronics), as described previously.26 Complex Formation. Cationic nanoparticles were formed by dialyzing 3 mg/mL solutions of P(MDS-co-CES) in dimethylformamide (DMF) against 0.02 M sodium acetate buffer (pH 4.6). The nanoparticle/ siRNA complexes were formed by mixing siRNA and nanoparticle solutions gently. The solutions were then allowed to stand for 30 min before use. High N/P ratios (defined as the molar ratio of nitrogen content in the polymer to the phosphorus content in the siRNA) were achieved by decreasing siRNA content but fixing polymer concentration at 15 or 60 mg/L or below because the polymer caused high cytotoxicity at high concentrations. Fixing the polymer concentration also allowed us to isolate the cytotoxic effect of downregulation from that contributed by the polymer. Polyethyleneimine (PEI) was complexed with siRNA at N/P 50, while Lipofectamine was used according to manufacturer’s instructions (1-2 µL in 1 mL of Opti-MEM at siRNA concentration of 1-50 nM). Analyses of Particle Size and Zeta Potential. The particle size and zeta potential of the freshly prepared nanoparticles were measured by ZetaSizer Nano ZS90 equipped with a He-Ne laser beam at 633 nm (scattering angle, 90°; Malvern Instruments Ltd., U.K.) at 25 °C. Each measurement was repeated five times. An average value was obtained from the five measurements. Gel Retardation Assay. Various formulations of nanoparticle or PEI/ siRNA complexes were prepared with N/P ratio ranging from 1 to 50 (fixing siRNA concentration at 350 nM), whereas Lipofectamine/siRNA complexes were fabricated with the recommended volumes of 1, 1.6, and 2 µL of Lipofectamine (in 1 mL of Opti-MEM) at siRNA concentration of 22 nM. Postequilibration, the complexes were electrophoresed on 1% agarose gel (stained with 4 µL of 0.5 µg/mL ethidium bromide per 50 mL of agarose solution) in 0.5× TBE buffer at 80 mV for 30 min. Naked siRNA was used as control. The gel was then analyzed on a UV illuminator (Chemi Genius, Geneflow Ltd., U.K.) to indicate the extent of complexation. Cell Culture. HepG2 and HeLa cells were cultured using DMEM (high glucose with L-glutamine) under a 5% CO2 environment at 37 °C. MDAMB-231 cells were cultured in L-15 medium (with L-glutamine), in nonCO2-equilibrated environment at 37 °C. Both media were supplemented with 10% fetal bovine serum and penicillin-streptomycin. HepG2, HeLa, and MDA-MB-231 cells were seeded in 12-well plate (density: 2 × 105, 1 × 105, and 3 × 105 in 1 mL per well, respectively) or 96-well plate (density: 8 × 103, 5 × 103, and 1.5 × 104 in 200 µL per well, respectively). Cells were treated 1 day after seeding. Protein Adsorption Assay. The protein adsorption assay was performed as previously reported.23 Briefly, bovine serum albumin (BSA) was dissolved in PBS (pH 7.4), giving a 0.5 mg/mL solution

before addition to 0.4 mg/mL nanoparticle solutions. All nanoparticle/ siRNA or siRNA-PEG complexes used contained 0.4 mg/mL nanoparticles in the sodium acetate buffer and were prepared at N/P 50. BSA solution (0.5 mL), PBS (1 mL), and complex solution (0.5 mL) were mixed together in a microcentrifuge tube. The pH of the test solution was around 7. After incubation for 72 h at 37 °C on an orbital shaker (100 rpm), 200 µL of each sample was taken and vortexed for 10 s to ensure homogeneity. The samples were then centrifuged at 10000 × g for 15 min to precipitate the nanoparticles. After centrifugation, the supernatant was carefully removed to prevent the resuspension of the pellet and tested using the BCA kit. An uncentrifuged BSA solution was used to normalize the results. The level of protein adsorption was calculated for nanoparticle/siRNA-PEG complexes, nanoparticle/siRNA complexes, and blank nanoparticles. Stability Assay. Blank polymeric nanoparticles, nanoparticle/siRNA complexes and nanoparticle/siRNA-PEG complexes were incubated in PBS supplemented with 10% FBS up to 4 h at 37 °C on an orbital shaker at 100 rpm, at an N/P ratio of 50, and a final polymer concentration of 80 mg/L. The sizes of the particles were observed using ZetaSizer Nano ZS90, as described above. Intracellular Distribution of Alexa Fluor 546-Labeled siRNA, P(MDS-co-CES) Nanoparticles, and their Nanocomplexes. P(MDS-coCES) nanoparticles (15 mg/L) were complexed with 22 nM of Alexa Fluor 546-labeled scrambled siRNA (siRNA-AF546). HeLa cells were plated on a glass-bottomed well and incubated with DMEM for 1 day. Phenol red-free DMEM was supplemented with 10% FBS and used to rinse the cells. The nanoparticle/DNA complexes were added to the cells in the completed colorless DMEM and imaged using a confocal laser scanning microscope (ZEISS LSM 500 LIVE) with 5% CO2 at 37 °C using a C-Apochromat 63×/1.2 water immersion lens. After 4 h of introduction, the cells were washed using colorless DMEM and were further incubated for another 20 h. The cells were then treated with FM4-64 lipophilic dye according to manufacturer’s instruction before imaging. siRNA Transfection and Cytotoxicity. Nanoparticle/siRNA or nanoparticle/siRNA-PEG complexes were prepared at different N/P ratios as described above. After incubating the cells (seeded in 12-well plates) with the complexes for 4 h, the media in the wells were changed with fresh media to remove the free complexes. The cells were further incubated for another 44 h, and the cellular levels of Bcl-2 mRNA and Bcl-2 protein were tested using real-time RT-PCR and Western blot, respectively, while cytotoxicity was determined via MTT test with the cells seeded in 96-well plates. The experiments were conducted by holding polymer concentration constant (i.e., 15 mg/L), as the polymer caused cytotoxicity that varied depending on polymer concentration and cell type. More than 80% cells were viable after incubation with the polymer at the concentrations tested. Real-Time RT-PCR. Each sample was derived from a single well from a 12-well plate. Each condition was performed in duplicate to ensure the reproducibility of the results. Total RNA was isolated from each well using the RNeasy kit and QiaShredder spin columns. The total RNA was reverse-transcribed using Superscript III First-Strand

Efficient Delivery of Bcl-2-Targeted siRNA

Biomacromolecules, Vol. 10, No. 1, 2009

45

Figure 6. Distribution of fluorescent-labeled siRNA in HeLa cells at 24 h after the treatment of nanoparticle/siRNA-AF546 complexes. Blue regions: siRNA-AF546 (B); cell membrane was labeled with FM 4-64 (red, A). C: Transmitted channel. D: Overlay of A and B. Scale bar: 20 µm.

Synthesis kit, using oligo-dTs as primers, according to manufacturer’s instructions. The products were then amplified using primers for Bcl-2 (target gene) and β-Actin (endogeneous control) cDNAs (cDNAs). Custom primers were purchased for both Bcl-2 and β-Actin, with sequences as follows: Bcl-2, sense 5′-CGACG ACTTC TCCCG CCGCT ACCGC-3′, antisense 5′-CCGCA TGCTG GGGCC GTACA GTTCC-3′; β-Actin, sense 5′-GCTCG TCGTC GACAA CGGCT C-3′, antisense 5′-CAAAC ATGAT CTGGG TCATC TTCTC-3′. β-Actin primer sequence is based on Invitrogen primers provided with the Superscript III reverse-transcription kit, and yields a 353-bp product. Custom Bcl-2 primer is based on the literature.6 All primers were used at a concentration of 10 µM, with 1 µL of primer set (sense plus antisense) in 25 µL of PCR reaction solution containing 12.5 µL of 2× SYBR Green I Mastermix, 9.5 µL of DEPC-treated water, and 2 µL of sample. Real-time RT-PCR was performed using Rotorgene 6000 (Corbett Research), with cycling conditions as follows: 1 cycle of 95 °C for 10 min (to “hot-start” reagents in Mastermix); 45 cycles of 95 °C for 45 s, 55 °C for 45 s, 72 °C for 90 s; 1 cycle of 72 °C for 7 min for final extension; ramp from 72 to 95 °C, 1 degree change per step, 5 s interval between steps. The sizes of the respective PCR products were then estimated using agarose gel electrophoresis, with a Fermentas 1 kb DNA ladder, to ensure purity. Downregulation of Bcl-2 mRNA was then determined by comparison of the ratio between Bcl-2 and β-Actin mRNA concentrations for the treated samples against that of the untreated sample. Western Blot Analysis. To assay changes in Bcl-2 protein levels, 30 µg of total protein lysate extracted from nanoparticle/siRNA transfected cells were resolved electrophoretically on 10% SDS-polyacrylamide gel under denaturing conditions and transferred onto PVDF membrane. Membranes were blocked for 1 h and probed with primary monoclonal anti-Bcl-2 (1:500 dilution) and antitubulin (1:5000 dilution) antibodies

Figure 7. Downregulation of Bcl-2 mRNA in MDA-MB-231, HeLa, and HepG2 cells across different siRNA concentrations. Polymer concentration: 15 mg/L and N/P ratio: 1-50 with siRNA concentration of 1100-22 nM.

at 4 °C overnight. Following incubation with HRP-conjugated antimouse secondary antibodies, proteins were detected using Supersignal West Pico Chemiluminescent Reagent and visualized on exposure to film. Band intensities were quantified using Quantity One (Bio-Rad Laboratories Pte. Ltd.), normalized to R-tubulin levels, and then expressed relative to the nontransfected controls. MTT Assay. The cytotoxicities of both the blank nanoparticles and the complexes were tested with the standard cell cytotoxicity test. The test was performed by adding 20 µL of MTT (5 mg/mL in PBS) to 200 µL of medium in wells containing cells that had been incubated with the nanoparticles or siRNA complexes. Each condition was performed with eight replicates. After a 4 h incubation period with MTT, the medium was removed from the wells, and dimethylsulfoxide (DMSO) was added. The value OD550-OD690 was detected using a

46

Biomacromolecules, Vol. 10, No. 1, 2009

Figure 8. Effect of PEGylation on Bcl-2 mRNA downregulation in HeLa cells and comparison in suppressing mRNA expression level among nanoparticles, Lipofectamine, and PEI. Nanoparticles (1, 2), Lipofectamine (3, 4), and PEI (5) were complexed with siRNA at concentrations of 7.3 (1, 3) and 20 nM (2, 4, 5; N/P 50 for nanoparticles and PEI).

microplate reader after mixing. The values were then normalized against control samples that had been treated with 0.02 M pH 4.6 sodium acetate buffer to yield viability figures. Combined DeliVery of siRNA and Paclitaxel. Paclitaxel incorporation into the nanoparticles was performed as reported previously.22 The final mass ratio of polymer to paclitaxel was 15:1. HeLa cells were treated with nanoparticle/siRNA or nanoparticle/siRNA-PEG complexes for 4 h. After 24 or 48 h, the cells were incubated with paclitaxel-loaded nanoparticles mixed in varying proportions with blank nanoparticles (to keep polymer concentration constant) for 4 h and the growth media were then replaced with fresh media. MTT assays were conducted after 24 h. In all cases, siRNA was applied at N/P 50, with the total polymer concentration fixed at 15 mg/L. All viability values were normalized using results from cells treated with nanoparticle/scrambled siRNA complexes followed by the treatment of blank nanoparticles after 24 or 48 h.

Results and Discussion Polymer Characterization. P(MDS-co-CES) had a nitrogen content of 4.14%, determined by elemental analysis. Degree of cholesterol grafting was 30-40%, and the weight average molecular weight of the polymer was around 3.5 kDa (PDI ) 1.57). The nanoparticles formed in 20 mM sodium acetate buffer (pH 4.6) had an average size of 90-130 nm and a zeta potential value of +70 mV. PEGylation of siRNA and siRNA Binding of Nanoparticles. A thiol group was built into the sense strand of siRNA through which siRNA was conjugated with PEG using SPDP. The guide strand of siRNA retained its native structure to conserve its biological functions. The disulfide bond, which links siRNA and PEG together, can be easily cleaved in the reducing environment of the cytoplasm,27 ensuring that the siRNA will be available to be incorporated into the RISC once it is freed from the endosomes. In the MALDI-TOF spectrum of PEGylated siRNA, no unreacted siRNA sense strand peak (6.8 kDa) was observed. However, the characteristic peak of PEGylated siRNA appeared (Figure 2), indicating successful PEGylation of siRNA. As reported previously,22 the nanoparticles condensed plasmid DNA efficiently and complete retardation of DNA mobility in the agarose gel was achieved at N/P 2. However, the nanoparticles provided weaker siRNA binding ability, and complete retardation of siRNA was not observed until N/P 20 and 50 for siRNA and siRNA-PEG, respectively (Figure 3A and B). This is possibly because the small size of siRNA led to less electrostatic interaction with the nanoparticles. Therefore, siRNA

Beh et al.

Figure 9. Viability of MDA-MB-231, HeLa, and HepG2 cells after being treated with nanoparticle/siRNA complexes containing different siRNA concentrations. Polymer concentration fixed at 15 mg/L. Viability expressed as a percentage of nanoparticle-treated cells without siRNA treatment. N/P 1-50 with siRNA concentration of 1100-22 nM.

might be more easily released from the cationic surface as compared with the much larger DNA. The higher N/P ratio needed to condense siRNA-PEG reflects the steric hindrance that the PEG moiety introduces. PEI had stronger siRNA binding, and the complete retardation of siRNA mobility was observed at N/P 20 (Figure 3C). In contrast, the siRNA binding of Lipofectamine was much weaker than that of the nanoparticles and PEI, and the complete retardation of siRNA was not observed even when 2 µL of Lipofectamine was used (Figure 3D). The siRNA-PEG binding of PEI and Lipofectamine was observed to be as efficient as their siRNA binding. Effect of PEGylation on Protein Adsorption. As can be seen from Figure 4, after being complexed with siRNA, the amount of protein adsorbed on the nanoparticles was significantly reduced due to the neutralization of cationic charge by siRNA. For example, 26 and 17% protein was adsorbed onto the nanoparticles and the nanoparticle/siRNA complexes prepared at N/P 50, respectively. In addition, the amount of protein adsorbed on the nanoparticle/siRNA complexes was significantly reduced when siRNA-PEG was mixed with siRNA due to shielding of cationic charge by PEG molecules. For example, when 20% siRNA-PEG was used, the amount of protein adsorbed on the complexes decreased from 26% to less than 5%. In addition, when 100% siRNA-PEG was used, the amount of protein adsorbed was further reduced to less than 3%. These findings illustrate the important role that neutralization and shielding of cationic charge take in preventing protein adsorption. The results also suggest that the use of PEG-modified siRNA may result in a delivery carrier having a prolonged blood circulation property. It was reported that less than 10% PEG (Mw ) 2 kDa) on the surface could adequately prevent macrophage phagocytosis.28 Therefore, the mixture of siRNA and siRNA-PEG may be used to prolong blood circulation of the nanoparticle/siRNA complexes. This approach can also be applied to conjugate other biomolecules such as proteins with PEG when nanoparticulate delivery systems are employed to prevent protein adsorption since the conjugation of PEG can be performed under mild conditions. Stability of Blank Polymeric Nanoparticles, Nanoparticle/ siRNA, or siRNA-PEG Complexes. The size of blank polymeric nanopartilces and siRNA/siRNA-PEG complexes increased after being incubated with serum-containing PBS for 1 h due to the aggregation of nanoparticles or siRNA complexes (Figure 5). However, there was no further increase in size at hour 4, and the size was still below 250 nm for all the samples tested. Bcl-2 Silencing. As shown in Figure 6, siRNA was successfully delivered inside the cells using the nanoparticles as

Efficient Delivery of Bcl-2-Targeted siRNA

Biomacromolecules, Vol. 10, No. 1, 2009

47

Figure 10. Viability of HeLa cells after being treated with nanoparticle/scrambled siRNA, siRNA, or siRNA-PEG complexes followed by paclitaxelloaded nanoparticles, relative to the cells treated with scrambled siRNA/nanoparticles followed by blank nanoparticles. Cells were treated with paclitaxel-loaded nanoparticles either at day 1 (A) or at day 2 (B) after the treatment of nanoparticle/scrambled siRNA, siRNA, or siRNA-PEG complexes. Polymer: 15 mg/L; siRNA: 22 nM. Nonfunctional scrambled siRNA was used as a control. Vertical axis represents blank polymer instead of 0.001 mg/L paclitaxel.

evidenced by the blue regions, interior to the membrane stained with FM 4-64. Therefore, the nanoparticle/siRNA complexes were effective in causing Bcl-2 downregulation at the mRNA level especially in MDA-MB-231 and Hela cells, and the mRNA expression levels were lower than 30% of the untreated MDAMB-231 and HeLa cells at all siRNA concentrations tested (Figure 7). The downregulation of the mRNA level in HepG2 cells was not as efficient as in the other two cell lines. A higher siRNA concerntraion led to a lower mRNA expression level. In a control experiment where scramble siRNA was used, no mRNA downregulation was observed. For example, mRNA expression levels in HepG2 and HeLa cells were 97 ( 11% and 105 ( 8%, respectively, when siRNA concentration of 22 nM and polymer concentration of 15 mg/L were employed. Consistent with the downregulation of Bcl-2 mRNA levels, Bcl-2 protein expression was downregulated by 36-66% in cells treated with nanoparticle/siRNA complexes. It is not surprising that the reduction in protein levels was not as drastic as that of mRNA given that mRNA knock-down is more immediate, albeit transient, than protein knock-down.29 In addition, the magnitude in protein reduction following siRNA treatment is highly dependent on the abundance and turnover (half-life) of the target protein. However, a similar trend was observed for both mRNA and protein downregulation with a more prominent reduction in levels in both MDA-MB-231 and HeLa cells (Figure S1). The use of higher siRNA concentrations resulted in a lower Bcl-2 protein expression level. For example, Bcl-2 protein level in MBA-MB-231 cells was downregulated by about 36 and 66% at siRNA concentrations of 22 (i.e., N/P 50) and 1100 nM (i.e., N/P 1), respectively. RNA interference of Bcl-2 was also achieved using PEGmodified siRNA, and the efficiency was not significantly different from that provided by the unmodified siRNA (Figure 8). It is expected that when delivered via a systemic administration route, the reduced clearance of nanoparticle/siRNA-PEG complexes from the body may in fact result in better effect due to increased availability.30 From Figure 8, it can also be seen that the efficiency of siRNA delivery induced by the nanoparticles was slightly lower than that mediated by Lipofectamine, but it was greater than that yielded by PEI under the tested

conditions. In addition, this siRNA was highly potent in suppressing the mRNA expression, and a concentration of 7.3 nM was sufficient to downregulate the mRNA expression level in HeLa cells to 10% or below when the nanoparticles or Lipofectamine was used as a carrier. Cell Viability after Being Treated with siRNA. The cells appeared to be resistant to siRNA-induced cell death over all siRNA concentrations tested (Figure 9). This finding led to our interest in studying chemosensitization by using HeLa as a model cell line and paclitaxel as a model anticancer drug. Because the cells essentially showed no change in viability with siRNA treatment, it is possible to isolate the effects of paclitaxel before and after Bcl-2 mRNA downregulation. Chemosensitization of HeLa Cells to Paclitaxel Treatment. Although the siRNA treatment was reported to be able to sensitize various types of cancer cells to chemotherapeutics,4 sensitization of HeLa cells was not found in the literature. Simultaneous delivery of Bcl-2-targeting siRNA and paclitaxel to HeLa cells was found to be ineffective in achieving a synergistic effect (data not shown). Similarly, a one-day delay between siRNA and paclitaxel treatments did not yield significant improvement to drug efficacy (Figure 10A). However, the introduction of paclitaxel treatment after a delay of 2 days was able to cause the cells to become more susceptible to paclitaxel at all paclitaxel concentrations tested (scrambled siRNA curve versus siRNA curve in Figure 10B). In other words, the functional phenotype of Bcl-2 knock-down by siRNA seemed to be apparent only after 2 days in this study. Although mRNA degradation could be detected shortly after siRNA treatment, this does not necessarily translate to a functional effect in the cells within the same time frame. Given the relatively long halflife of the Bcl-2 protein,31 it will take considerable time for the effect of protein reduction due to mRNA degradation by siRNA treatment to be obvious. As we observed, cells became more susceptible to paclitaxel-induced apoptosis only after a period of 48 h, when existing antiapoptotic Bcl-2 protein has degraded appreciably and new protein production slowed due to siRNA knockdown. The sensitization of the cells by siRNA-PEG was as efficient as that by siRNA. This was expected as both siRNA and siRNA-PEG down-regulated the mRNA expression to a

48

Biomacromolecules, Vol. 10, No. 1, 2009

similar level under the same conditions. Because the polymer was slightly toxic to the cells (more than 80% cells remained viable at the tested concentration), the sensitization also resulted in increased cytotoxicity of the blank nanoparticles (comparing the first data points in Figure 10B). These findings further demonstrate the importance of the fundamental role Bcl-2 proteins play in the maintenance of cellular health and rescuing cells from apoptosis. This in turn suggests a potential application of the findings to conventional cancer treatment by compromising cancer cell apoptosis rescue functions, followed by timely administration of anticancer drugs.

Conclusions It has been shown that the cationic nanoparticles selfassembled from P(MDS-co-CES) successfully delivered siRNA into HepG2, HeLa, and MDA-MB-231 cell lines, and downregulated Bcl-2 mRNA and Bcl-2 protein expression levels. The efficiency of siRNA delivery was similar to that mediated by Lipofectamine but higher than that induced by PEI. PEGylation of siRNA prevented the protein adsorption of nanoparticle/ siRNA complexes. Although the presence of serum caused slight aggregation of nanoaprticle/siRNA or siRNA-PEG complexes, the size of the complexes was still below 250 nm after 4 h of incubation. siRNA-PEG downregulated Bcl-2 mRNA expression as efficiently as unmodified siRNA. These findings demonstrate the possibility of delivering siRNA systemically using cationic nanoparticulate vectors via PEGylation of siRNA. Although the downregulation of Bcl-2 mRNA and Bcl-2 protein expression did not show significant cytotoxicity, it sensitized HeLa cells to paclitaxel treatment applied at day 2 after the siRNA treatment. Such a polymer nanoparticle system has good potential in clinical applications, improving on current chemotherapeutic regimens. Acknowledgment. Financial support from Institute of Bioengineering and Nanotechnology, Agency for Science, Technology and Research, Singapore is gratefully acknowledged. Supporting Information Available. Western blot results. This material is available free of charge via the Internet at http:// pubs.acs.org.

References and Notes (1) Fire, A.; Xu, S. Q.; Montgomery, M. K.; Kostas, S. A.; Driver, S. E.; Mello, C. C. Nature 1998, 391, 806–811. (2) Dudley, N. R.; Amin, A. Z.; Goldstein, B. In RNA Interference Technology; Appasani, K., Ed.;Cambridge University Press: New York, NY, 2005; pp 55-68. (3) Gartel, A. L.; Kandel, E. S. Biomol. Eng. 2006, 23, 17–34.

Beh et al. (4) Feng, L. F.; Zhong, M.; Lei, X. Y.; Zhu, B. Y.; Tang, S. S.; Liao, D. F. J. Drug Targeting 2006, 14, 21–26. (5) Ocker, M.; Neureiter, D.; Lueders, M.; Zopf, S.; Ganslmayer, M.; Hahn, E. G.; Herold, C.; Schuppan, D. Gut 2005, 54, 1298–1308. (6) Huang, S.; Wu, Y.; Yu, H.; Zhang, X. Q.; Ying, L.; Zhao, H. F. Acta Pharm. Sin. 2006, 27, 242–248. (7) Kim, D. H.; Rossi, J. J. Nat. ReV. Genet. 2007, 8, 173–184. (8) Berardo, M. D.; Elledge, R. M.; de Moor, C.; Clark, G. M.; Osborne, C. K.; Allred, D. C. Cancer 1998, 82, 1296–1302. (9) Tanabe, K.; Kim, R.; Inoue, H.; Emi, M.; Uchida, Y.; Toge, T. Int. J. Oncol. 2003, 22, 875–881. (10) Webb, A.; Cunningham, D.; Cotter, F.; Clarke, P. A.; di Stefano, F.; Ross, P.; Corbo, M.; Dziewanowska, Z. Lancet 1997, 349, 1137–1141. (11) Walensky, L. D. Cell Death Differ. 2006, 13, 1339–1350. (12) Feng, S. S.; Chien, S. Chem. Eng. Sci. 2003, 58, 4087–4114. (13) Peer, D.; Karp, J. M.; Hong, S.; Farokhzad, O. C.; Margalit, R.; Langer, R. Nat. Nanotechnol. 2007, 2, 751–760. (14) Kim, S. H.; Jeong, J. H.; Cho, K. C.; Kim, S. W.; Park, T. G. J. Controlled Release 2005, 104, 223–232. (15) Howard, K. A.; Rahbek, U. L.; Liu, X.; Damgaard, C. K.; Glud, S. Z.; Andersen, M. Ø.; Hovgaard, M. B.; Schmitz, A.; Nyengaard, J. R.; Besenbacher, F.; Kjems, J. Mol. Ther. 2006, 14, 476–484. (16) Kim, W. J.; Christensen, L. V.; Jo, S.; Yockman, J. W.; Jeong, J. H.; Kim, Y. H.; Kim, S. W. Mol. Ther. 2006, 14, 343–350. (17) Santel, A.; Aleku, M.; Keil, O.; Endruschat, J.; Esche, V.; Fisch, G.; Dames, S.; Lo¨ffler, K.; Fechtner, M.; Arnold, W.; Giese, K.; Klippel, A.; Kaufmann, J. Gene Ther. 2006, 13, 1222–1234. (18) Lee, S. H.; Kim, S. H.; Park, T. G. Biochem. Biophys. Res. Commun. 2007, 357, 511–516. (19) Kim, S. H.; Jeong, J. H.; Lee, S. H.; Kim, S. W.; Park, T. G. J. Controlled Release 2006, 116, 123–129. (20) Oishi, M.; Nagasaki, Y.; Nishiyama, N.; Itaka, K.; Takagi, M.; Shimamoto, A.; Furuichi, Y.; Kataoka, K. ChemMedChem 2007, 2, 1290–1297. (21) Sun, T. M.; Du, J. Z.; Yan, L. F.; Mao, H. Q.; Wang, J. Biomaterials 2008, 29, 4348–4355. (22) Wang, Y.; Gao, S. J.; Ye, W. H.; Yoon, H. S.; Yang, Y. Y. Nat. Mater. 2006, 5, 791–796. (23) Wang, Y.; Wang, L. S.; Goh, S. H.; Yang, Y. Y. Biomacromolecules 2007, 8, 1028–1037. (24) Lee, A.; Wang, Y.; Ye, W. H.; Yoon, H. S.; Chan, S. Y.; Yang, Y. Y. Biomaterials 2008, 29, 1224–1232. (25) Stuchbury, T.; Shipton, M.; Norris, R.; Malthouse, J. P. G.; Brocklehurst, K.; Herbert, J. A. L.; Suschitzky, H. Biochem. J. 1975, 151, 417–432. (26) Turner, J. J.; Jones, S. W.; Moschos, S. A.; Lindsay, M. A.; Gait, M. J. Mol. BioSyst. 2007, 3, 43–50. (27) Cumming, R. C.; Andon, N. L.; Haynes, P. A.; Park, M.; Fischer, W. H.; Schubert, D. J. Biol. Chem. 2004, 279, 21749–21758. (28) Moghimi, S. M.; Hamad, I.; Andresen, T. L.; Jørgensen, K.; Szebeni, J. FASEB J. 2006, 20, 2591–2593. (29) Holen, T.; Amarzguioui, M.; Wiiger, M. T.; Babaie, E.; Prydz, H. Nucleic Acids Res. 2002, 30, 1757–1766. (30) Mok, H.; Palmer, D. J.; Ng, P.; Barry, M. A. Mol. Ther. 2006, 11, 66–79. (31) Celli, A.; Que, F. G.; Gores, G. J.; LaRusso, N. F. Am. J. Physiol. 1998, 275, 749–757.

BM801109G