Lipid-Conjugated 27-Nucleotide Double-Stranded RNAs with Dicer

Apr 11, 2012 - Department of Life Science, Faculty of Pharmacy, Yasuda Women's University, Hiroshima, Japan. ‡. Department of Biochemistry, Graduate...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/molecularpharmaceutics

Lipid-Conjugated 27-Nucleotide Double-Stranded RNAs with Dicer-Substrate Potency Enhance RNAi-Mediated Gene Silencing Takanori Kubo,*,† Kazuyoshi Yanagihara,† Yoshifumi Takei,‡ Keichiro Mihara,§ Yuichiro Sato,∥ and Toshio Seyama† †

Department of Life Science, Faculty of Pharmacy, Yasuda Women’s University, Hiroshima, Japan Department of Biochemistry, Graduate School of Medicine, Nagoya University, Nagoya, Japan § Department of Hematology and Oncology, Research Institute for Radiation Biology and Medicine, Hiroshima University, Hiroshima, Japan ∥ Department of Medical Pharmacy, Faculty of Pharmacy, Yasuda Women's University, Hiroshima, Japan ‡

S Supporting Information *

ABSTRACT: Short interfering RNAs (siRNAs), used in RNA interference (RNAi) technology, are powerful tools for targetgene silencing in a sequence-specific manner. In this study, Dicer-substrate 27-nucleotide (nt) double-stranded RNAs (dsRNAs), which are known to have a highly potent RNAi effect, were conjugated with palmitic acid at the 5′-end of the sense strand to enhance intracellular delivery and RNAi efficacy. The palmitic acid-conjugated 27-nt dsRNAs (C16ds27RNAs) were prepared by our simple synthesis strategy in good yield. The C16-ds27RNAs were cleaved by a Dicer enzyme, leading to the release of 21-nt siRNAs. The high level of stability in serum using C16-ds27RNAs was also confirmed. The C16-ds27RNAs showed enhanced RNAi potency targeted to both an exogenous luciferase and an endogenous vascular endothelial growth factor (VEGF) gene in the presence or absence of a transfection reagent, such as Lipofectamine 2000. In addition, the C16-ds27RNAs had a more potent genesilencing activity than the other lipid-conjugated 21-nt siRNAs and 27-nt dsRNAs. The C16-ds27RNAs also exhibited significant membrane permeability. These results suggested that the C16-ds27RNAs will be useful for next-generation RNAi molecules that can address the problems of RNAi technology. KEYWORDS: palmitic acid conjugates, Dicer substrate, 27-nt dsRNA, potent gene silencing, intracellular delivery, nuclease resistance

■

INTRODUCTION RNA interference (RNAi) technology, first discovered by Fire et al. as a powerful tool for suppressing the expression of specific genes,1 is one of the most attractive candidates for the treatment of various diseases.2−5 In RNAi, the long doublestranded RNAs (dsRNAs) are cleaved by Dicer, a member of the RNase III family of ribonucleases, to generate 21-nucleotide (nt) short interfering RNAs (siRNAs). The siRNAs are converted into a single-stranded form as they assemble into RNAinduced silencing complexes (RISCs). Antisense strands of siRNAs incorporated into RISCs are used to target perfectly complementary mRNA species, and RISCs cleave the target mRNA. 6−10 The 21-nt siRNAs have been chemically synthesized and have shown potent RNAi effects.11,12 However, they still have the problems of poor membrane permeability and nuclease resistance, which limit their therapeutic © 2012 American Chemical Society

applicability. Various chemically modified-21-nt siRNAs were developed to address these problems.13−25 The 2′-modifications, such as 2′-O-methyl, 2′-fluoro, and locked nucleic acid (LNA) of 21-nt siRNA, as well as modifications to the phosphate backbone (e.g., phosphorothioate and boranophosphate), demonstrated high nuclease resistance.13−17 Direct conjugations of 21-nt siRNA with functional molecules, including polymers,18,19 peptides,20−22 and lipids,23−26 have been reported to enhance nuclease stability or the cell membrane permeability of siRNAs. Although such chemical modifications and direct conjugations to the siRNAs could solve some of the Received: Revised: Accepted: Published: 1374

December 7, 2011 April 6, 2012 April 11, 2012 April 11, 2012 dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

problems of RNAi, most of them weaken gene-silencing efficacy. Recently, 27-nt dsRNAs have been found to exhibit much stronger gene-silencing effects than 21-nt siRNAs.27,28 The 27-nt dsRNAs are cleaved by a Dicer enzyme, leading to the release of 21-nt siRNAs, and thus are incorporated into the RISC. We previously reported that 27-nt dsRNAs modified with amine at the 5′-end of the sense strand exhibited much better RNAi potency compared with the nonmodified 27-nt dsRNAs.29−31 In the present study, we prepared lipid-conjugated 27-nt dsRNAs, including palmitic acid, lauric acid, and cholesterol, in which those lipids modified the 5′-end of the sense strand, with the aim of facilitating cellular uptake and adding a potent genesilencing effect. All conjugates were investigated with respect to Dicer cleavage, stability in serum, cellular uptake, and RNAi efficacies. Among these conjugates, palmitic acid-conjugated 27-nt dsRNA (C16-ds27RNA) exhibited the best potential for development into a new generation of RNAi molecules.

■

EXPERIMENTAL SECTION Design and Synthesis of dsRNAs. The 21-nt siRNAs and 27-nt dsRNAs were designed as targets to the Renilla luciferase and vascular endothelial growth factor (VEGF) genes (Figure 1A). The 21-nt and 27-nt single-strand RNAs (ssRNAs; antisense and sense strand) including amino modifications were purchased from Integrated DNA Technologies, Inc. (IDT, Coralville, IA, USA). Amino-modified sense-stranded ssRNAs were designed so as to be amine-assembled on only the 5′-end. The molecular weights of all ssRNAs were confirmed by matrixassisted desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (Ultraflex; Bruker Daltonics, Bremen, Germany) using saturated solutions of 2,4,6-trihydroxyacetophenone (Sigma-Aldrich, St. Louis, MO, USA) in 50 mg/mL diammonium hydrogen citrate (Wako, Osaka, Japan) in 50% acetonitrile as a matrix.32 The concentrations of all ssRNAs were calculated using their absorbance at 260 nm as detected spectrophotometrically (V-670 spectrophotometer; Jasco, Tokyo, Japan). The sense and antisense strands of RNAs were annealed in annealing buffer to prepare dsRNAs (20 μM concentrations as stock solutions) following the manufacturer's instructions. The dsRNA formation was confirmed by 20% polyacrylamide gel electrophoresis (PAGE; 30 mA, 70 min) and visualized by silver staining (DNA Silver Stain Kit; GE Healthcare, Piscataway, NJ). Synthesis of Lipid-Conjugated dsRNA. In the synthesis of 21-nt and 27-nt ssRNAs conjugated with palmitic acid (C16ss21RNA and C16-ss27RNA) and lauric acid (C12-ss21RNA and C12-ss27RNA), the amino-modified ssRNAs (4 nmol in 20 μL of water) were reacted with 40 nmol of palmitic acid N-hydroxysuccinimide ester (Sigma-Aldrich) and lauric acid 4-nitrophenyl ester (Tokyo Chemical Industry, Tokyo, Japan), respectively, dissolved in 10 μL of N,N-dimethylformamide (DMF; Sigma-Aldrich) containing 0.7 μL of N,N-diisopropylethylamine (DIEA; Sigma-Aldrich) in 100 μL of isopropanol/ water (1:1) mixture solutions, for 12 h at room temperature. These lipid-conjugated ssRNAs were purified by reversed-phase high-performance liquid chromatography (RP-HPLC) using an ODS column (4.6 × 150 mm, 5 μm) under a linear gradient condition of acetonitrile while shifting the concentrations from 7% to 70% for 40 min in 20 mM TEAA (pH 7.0) (see Figure S1A−D in the Supporting Information). The cholesterolconjugated 21-nt and 27-nt ssRNAs at the 5′-end (Chol-ss21RNA

Figure 1. Sequences, synthesis, and structures of lipid-conjugated dsRNAs. (A) Sequences of 21-nt siRNAs and 27-nt dsRNAs. The dsRNAs were designed to target Renilla luciferase (si21Luc and ds27Luc) and VEGF (si21VEGF and ds27VEGF). (B) Simple synthesis of C16-dsRNAs. (C) Structures of lipid-conjugated dsRNAs. Palmitic acid, lauric acid, and cholesterol were covalently conjugated to the 21-nt siRNAs and 27-nt dsRNAs at the 5′-end of the sense strand.

and Chol-ss27RNA) were purchased from Hokkaido System Science (HSS, Sapporo, Japan). The molecular weights of the conjugates were confirmed by MALDI-TOF mass spectrometry (Ultraflex; Bruker Daltonics; see Figure S2A−H in the Supporting Information). The yields of the conjugates were spectrophotometrically calculated on the basis of absorbance at a 260 nm wavelength. To prepare the lipid-conjugated 21-nt siRNAs and 27-nt dsRNAs, antisense 21-nt and 27-nt ssRNAs were annealed with the 21-nt and 27-nt ssRNA conjugations. Palmitic acid-conjugated 21-nt and 27-nt dsRNAs (C16-si21RNA and C16ds27RNA), lauric acid-conjugated 21-nt and 27-nt dsRNAs (C12-si21RNA and C12-ds27RNA), and cholesterol-conjugated 21-nt and 27-nt dsRNAs (Chol-si21RNA and Cholds27RNA) were prepared. The quality of the lipid-conjugated 21-nt siRNAs and 27-nt dsRNAs was confirmed by 20% PAGE. In Vitro Cleavage of Lipid-Conjugated dsRNAs by Dicer. The 21-nt siRNAs and 27-nt dsRNAs (20 pmol), including the lipid conjugates, were mixed with 1 U of recombinant Dicer (Gene Therapy Systems Inc., San Diego, CA) in 10 μL of 20 mM Tris-HCl (pH 8.0) containing 150 mM NaCl and 2.5 mM MgCl2. The mixtures were incubated at 37 °C, and the reaction was stopped by adding 2 μL of the stop solution 1375

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

C16 conjugates (100 pmol), each of which was mixed with LF2000 in 100 μL of Opti-MEM, were added to each well of a 24-well microplate containing the cells in 900 μL of fresh culture medium. After 6 h of incubation, the medium was replaced with fresh medium, and the cells were cultured for 48 h. To investigate the RNAi in the absence of LF2000, siRNAs including C16 conjugates (5 nmol) were added to each well of a 24well microplate containing the cells in 900 μL of medium and cultured for 48 h. Gene Silencing of Renilla Luciferase. The efficacy of RNAi against Renilla luciferase was evaluated by the Dual-Glo Luciferase Assay System (Promega). To detect Firef ly luciferase activity as an intraplasmid control, 50 μL of Dual-Glo Luciferase reagent-1 (beetle luciferin) was added to each well of a 96-well multiplate containing siRNAs. A plate was incubated in the dark for 10 min at room temperature. Luminescence emitted from the Firef ly luciferase catalytic reaction was measured for 1 s for each well on a microplate reader (Wallac 1420 ARVO MX; PerkinElmer, Waltham, MA). To measure the Renilla luciferase activity and to quench the luminescence from the Firef ly luciferase catalytic reaction, 50 μL of Dual-Glo Stop and Glo reagent-2 (containing coelenterazine) was added to each well. The multiplates were then incubated in the dark for 10 min at room temperature. The luminescence arising from the Renilla luciferase catalytic reaction was measured in the same way as described above for Firef ly luciferase activity and was normalized by the luminescence of Firef ly luciferase activity in each well of the 96-well multiplates. The RNAi of 21-nt siRNA (si21Luc) and 27-nt dsRNA (ds27Luc), including lipid conjugates (C16-si21Luc, C16-ds27Luc, C12-si21Luc, C12-ds27Luc, Chol-si21Luc, and Chol-ds27Luc), toward the Renilla luciferase was assessed as a percentage of the control (dsRNA nontreated) sample. Gene Silencing of VEGF. The level of RNAi activity directed toward VEGF was evaluated by measuring VEGF mRNA quantitatively using RT-PCR. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA was also measured as an intrinsic control. The total RNA was extracted by an RNeasy Plus Mini Kit (Qiagen, Valencia, CA), and the yield was determined by the optical density at 260 nm wavelength on a UV spectrometer. RT-PCR analysis was conducted using 100 ng of extracted total RNA. The primers used for the target mRNAs were as follows: VEGF forward primer, 5′CCCTGATGAGATCGAGTACATCTT-3′; VEGF reverse primer, 5′-ACCGCCTCGGCTTGTCAC-3′; GAPDH forward primer, 5′-GGAAAGCTGTGGCGTGATG-3′; GDPDH reverse primer, 5′-CTGTTGCTGTAGCCGTATTC-3′. RTPCR was carried out using a One-step RT-PCR Kit (Qiagen) according to the manufacturer's protocol. Briefly, the RT step was carried out at 50 °C for 30 min, followed by heat denaturation at 94 °C for 10 min. The PCR step consisted of 25 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s. The PCR products were electrophoresed on 2% agarose gel, and the predicted size of amplified molecules was certified with the aid of an LAS4000 imaging system (Fujifilm). The RNAi potency of 21-nt siRNA (si21VEGF) and 27-nt dsRNA (ds27VEGF), including C16 conjugates (C16-si21VEGF and C16-ds27VEGF), toward the VEGF was assessed as a percentage of the control (nontreated dsRNA) sample after each sample was calibrated with the intrinsic control. Intracellular Delivery of C16-21-nt siRNA and C16-27nt dsRNA. To prepare fluorescence (FAM)-labeled 21-nt siRNA and 27-nt dsRNA, antisense 21-nt and 27-nt ssRNA

(Gene Therapy Systems Inc.) after 12 h of incubation. The reaction products were electrophoresed on 20% PAGE (30 mA, 70 min) and visualized by silver staining (DNA Silver Stain Kit; GE Healthcare). The signals of the cleaved products were photographed with an LAS4000 imaging system (Fujifilm, Tokyo, Japan). Stability against Nuclease Degradation in Cell Culture Medium. A portion of 10 μL of the 21-nt siRNA and 27-nt dsRNA (200 pmol), including the palmitic acid (C16) conjugates, was added to 90 μL of Dulbecco's modified Eagle's medium (DMEM; Wako) containing 10% heat-inactivated FBS (Invitrogen, La Jolla, CA). The samples were incubated for different time periods (0, 0.5, 1, 2, 4, 6, 8, 12, 24, and 48 h) at 37 °C. An aliquot (10 μL) was taken from each sample. The samples were frozen in liquid nitrogen (to stop the nuclease reaction) and were kept at −20 °C. The RNA products of nuclease degradation were detected and analyzed by 20% PAGE and RP-HPLC as described in the preceding section. Cell Culture and Transfections. The human cervical carcinoma cell line (HeLa) and human lung carcinoma cell line (A549) were cultured in DMEM (Wako), and the human gastric carcinoma cell line (SH-10-TC) was cultured in RPMI1640 (Wako), supplemented with 10% heat-inactivated FBS (Invitrogen), 100 U/mL penicillin (Wako), and 100 μg/mL streptomycin (Wako). All cell lines were obtained from Riken Bioresource Center Cell Bank (Tsukuba, Japan). To evaluate the RNAi potency of dsRNAs, including lipid conjugates, that targeted the Renilla luciferase gene, we used the psiCHECK-2 Vector (Promega, Madison, WI) as a reporter gene, which contains both the Firef ly and Renilla luciferase genes, in HeLa cells. HeLa cells were seeded at 5 × 104 cells/ mL in 100 μL of medium in each well of a 96-well multiplate and cultured under a 100% humidified atmosphere (5% CO2, 37 °C). Twelve hours later, 0.02 μg of psiCHECK-2 Vector, which was mixed with 0.2 μL of Lipofectamine 2000 (LF2000; Invitrogen) in 10 μL of Opti-MEM (Invitrogen), was added to each well of a 96-well multiplate containing the cells in 90 μL of fresh culture medium without antibiotics. To investigate RNAi in the presence of LF2000, the dsRNAs, including lipid conjugates, at two different concentrations (1.0 and 0.2 nM) were preincubated with 0.2 μL of LF2000 in 10 μL of OptiMEM. Four hours after the vector transfection, 10 μL of the preincubated mixtures of dsRNAs, including lipid conjugates, with LF2000 was added to each well containing 90 μL of fresh culture medium. After another 8 h incubation, the culture medium was replaced with 100 μL of fresh medium, and the cells were cultured for 48 h to assess RNAi. To investigate RNAi effect in the absence of any transfection reagents, the psiCHECK-2 Vector was first transfected using LF2000 by the same procedure as described above. The cells were washed three times with culture medium after 4 h of vector transfection, and then 10 μL of samples of the dsRNAs, including lipid conjugates, at two different concentrations (600 and 50 nM) without any transfection reagents was added to each well containing the cells in 90 μL of fresh culture medium. The cells were cultured for 48 h. The luciferase activity was analyzed 48 h after the dsRNAs were transfected. For the RNAi analysis of the 21-nt and 27-nt dsRNAs, including C16 conjugates, against exogenous VEGF gene expression in the HeLa, A549, and SH-10-TC cell lines, each cell line was adjusted to 5 × 104 cells in 1 mL of medium in each well of a 24-well multiplate and cultured. To investigate RNAi in the presence of LF2000, 21-nt siRNAs and 27-nt dsRNAs including 1376

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

Table 1. Characterization of ssLuc Conjugated with Lipid Molecules name

target gene

conjugated molecule

HPLC retention timea (min)

MALDI-TOF MSb found/calcd

yieldc (%)

ss21Luc ss27Luc C16-ss21Luc C16-ss27Luc C12-ss21Luc C12-ss27Luc Chol-ss21Luc Chol-ss27Luc ss21VEGF ss27VEGF C16-ss21VEGF C16-ss27VEGF

luciferase luciferase luciferase luciferase luciferase luciferase luciferase luciferase VEGF VEGE VEGF VEGF

none none palmitic acid palmitic acid lauric acid lauric acid cholesterol cholesterol none none palmitic acid palmitic acid

11.9 12.5 32.1 34.1 27.3 27.4 31.9 30.4 12.5 12.1 31.7 30.2

6569.8/6569.9 8463.0/8465.1 6984.6/6986.4 8880.5/8879.8 6930.6/6930.3 8824.5/8823.7 7324.7/7325.0 9222.1/9222.1 6662.8/6663.1 8584.8/8584.2 7080.7/7080.5 8999.7/9001.1

d d 52.80 64.28 46.8 49.02 d d d d 46.76 45.02

a

A linear gradient condition of CH3CN shifting the concentrations from 7% to 70% during 40 min in 20 mM TEAA (pH 7.0) using an ODS column. bA saturated solution of 2,4,6-trihydroxyacetophenone in 50 mg/mL diammonium hydrogen citrate in 50% acetonitrile was used as a matrix. c Overall yields of the products were determined by measuring the absorbance at 260 nm after HPLC purification. dThe purified ssRNAs were purchased.

were labeled with 5′-fluorescein phosphoramidite (6-FAM; Glen Research, Sterling, VA) at the 5′-end. The FAM-labeled antisense 21-nt and 27-nt ssRNAs were annealed with C16conjugated and unconjugated sense 21-nt and 27-nt ssRNA, respectively, in annealing buffer. To deliver the prepared dsRNA intracellularly to the HeLa, A549, and SH-10-TC cell lines, 100 pmol of siRNA and dsRNA, including C16 conjugates labeled with FAM, was incubated with 2 μL of LF2000 in 100 μL of Opti-MEM diluted twice for 30 min at room temperature, or 5 nmol of siRNA and dsRNA, including C16 conjugates labeled with FAM, was dissolved in 100 μL of Opti-MEM. Then, 100 μL of each sample was added to 900 μL culture medium of each cell line (5 × 104 cells) and incubated for 6 h in the dark under a humidified atmosphere (5% CO2, 37 °C). The cells were washed several times with fresh medium, and the intracellularly incorporated quantity of dsRNAs labeled with FAM in cells was examined under a fluorescent confocal microscope (IX70; Olympus, Tokyo, Japan). Another approach was used to investigate the intracellular delivery of 21-nt siRNA and 27-nt dsRNA, including C16 conjugates labeled with FAM, by flow cytometry (FACSAria; BD Biosciences, Franklin Lakes, NJ). The forward- and sidescatter parameters were adjusted to accommodate the inclusion of each of the dissociated cell lines with the aid of FAM as a marker. A total of 5000 cells were analyzed, and no cells were excluded from the analysis. Data were collected and analyzed using FACSDiva software (BD Bioscience).

■

Figure 2. Dicer-substrate dsRNAs including lipid conjugates. si21Luc (A), C16-si21Luc (B), C12-si21Luc (C), Chol-si21Luc (D), ds27Luc (E), C16-ds27Luc (F), C12-ds27Luc (G), and Chol-ds27Luc (H) were reacted with the recombinant Dicer enzyme for 12 h at 37 °C. The reaction products were electrophoresed on 20% PAGE and visualized by silver staining.

RESULTS Synthesis of Lipid-Conjugated 21-nt siRNAs and 27-nt dsRNAs. The ssRNAs (21-nt and 27-nt) modified with amine at the 5′-end were condensed with palmitic acid Nhydroxysuccinimide ester in solution (Figure 1B). The amino-modified 21-nt and 27-nt ssRNAs were also reacted with lauric acid p-nitrophenyl. The crude C16-ss21RNA, C16ss27RNA, C12-ss21RNA, and C12-ss27RNA were purified by RP-HPLC. Purified C16-ss21RNA, C16-ss27RNA, C12ss21RNA, and C12-ss27RNA were obtained in 52.80%, 64.28%, 46.8%, and 49.02% overall yields, respectively. Purified Chol-ss21RNA and Chol-ss27RNA were purchased from HSS. The molecular weights of all lipid-conjugated ssRNAs were confirmed by MALDI-TOF mass spectrometry. The RP-HPLC

retention times of the lipid-conjugated 21-nt and 27-nt ssRNAs were found to be slower than those of the nonconjugated 21-nt and 27-nt ssRNAs under the purified conditions (see the Experimental Section). All lipid-conjugated 21-nt and 27-nt ssRNAs were obtained at high purity with a single one-product by RP-HPLC analysis (see Figure S1A−D in Supporting Information). Table 1 summarizes the characteristics of the lipid-conjugated ssRNAs targeted to the Renilla luciferase 1377

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

Figure 3. Comparative RNAi efficacy of dsRNAs including lipid conjugates to detect the luminescence of Renilla luciferase activity in HeLa cells at concentrations of 1 nM (A) and 0.2 nM (B) in the presence of LF2000. The controls were given only PBS(−). RNAi efficacies of the dsLuc including lipid conjugates were evaluated to detect the luminescence of Renilla luciferase activity, which was normalized by the luminescence of Firef ly luciferase activity, after 48 h of incubation. The mean and SD values are from three independent experiments. *, P < 0.05; ns = not significant (t test).

Figure 4. Comparative RNAi efficacy of dsRNAs including lipid conjugates to detect the luminescence of Renilla luciferase activity in HeLa cells at concentrations of 600 nM (A) and 50 nM (B) in the absence of LF2000. The controls were given only PBS(−). RNAi efficacies of the dsLuc including lipid conjugates were evaluated to detect the luminescence of Renilla luciferase activity, which was normalized by the luminescence of Firef ly luciferase activity, after 48 h of incubation. The mean and SD values are from three independent experiments. *, P < 0.05; **, P < 0.01; ns = not significant (t test).

produce the released 21-nt siRNA products. In contrast, all lipidconjugated si21RNAs showed the same mobility shift on PAGE in either the presence or absence of the Dicer enzyme. RNAi Efficacy of Lipid-Conjugated dsRNAs Targeted to the Exogenous Luciferase. We performed a genesilencing study of the lipid-conjugated si21RNAs and ds27RNAs targeted to the Renilla luciferase gene in either the presence or absence of LF2000. In the presence of LF2000, Renilla luciferase gene expression was suppressed in lipid conjugates with high potency (Figure 3). The C16 conjugates had a higher inhibitory effect than the other lipid conjugates (Figure 3A). In particular, C16-ds27Luc exhibited a more

(ssLuc) and VEGF (ssVEGF) gene. All lipid-conjugated 21-nt and 27-nt ssRNAs were annealed with the antisense 21-nt and 27-nt ssRNAs, respectively, to make lipid-conjugated si21RNAs and ds27RNAs (Figure 1C). All lipid conjugates were confirmed to be double-stranded by using 20% PAGE, and each lipid conjugate showed a different mobility shift on PAGE in comparison with nonmodified 21-nt siRNA and 27-nt dsRNA. Dicer Processing of Lipid-Conjugated dsRNAs. We investigated whether or not the lipid conjugates are substrates of the Dicer enzyme cleaved to the 21-nt siRNA (Figure 2). The Dicer enzyme recognized all lipid-conjugated ds27RNAs as a substrate to 1378

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

potent gene-silencing efficacy than C16-si21Luc at a concentration of 0.2 nM (Figure 3B). In the absence of LF2000, the C16-ds27Luc also exhibited a stronger RNAi efficacy compared with all of the other lipid conjugates (Figure 4). Si21Luc and ds27Luc did not inhibit Renilla luciferase activity at any concentration, even at a concentration as high as 600 nM. C16-si21Luc and Cholds27Luc were verified to have stronger RNAi efficacy at a concentration of 600 nM (Figure 4A), but both of them showed lower gene-silencing activity than C16-ds27Luc at a concentration of 50 nM (Figure 4B). C12-siLuc and C12ds27Luc did not show any remarkable inhibitory effect at any concentration. These results indicate that the C16-ds27RNA is one of the most attractive molecules in terms of its application for RNAi among the lipid-conjugated dsRNAs used in this study. Nuclease Resistance of C16-Conjugated dsRNAs. We investigated the nuclease resistance of dsRNAs, including C16 conjugates, in a cell culture medium containing 10% FBS at different incubation times (0, 0.5, 1, 2, 4, 6, 8, 12, 24, and 48 h; Figure 5A). Si21Luc was degraded immediately and completely disappeared after about 5 h of incubation. In contrast, C16si21Luc showed greater stability against nuclease than si21Luc. Although ds27Luc was highly stable against nuclease degradation, C16-ds27Luc showed excellent stability. To investigate the RNAi efficacy of survived dsRNAs, including C16 conjugates, after nuclease treatment each aliquot sample described above was examined in a gene-silencing study targeted to the Renilla luciferase gene (Figure 5B). The aliquot samples at different incubation times (0, 1, 4, 8, 24, and 48 h) were prepared to 1 nM concentrations of dsRNAs, which was attributed to the concentration before nuclease treatment, and the RNAi efficacy of each was investigated in the presence of LF2000 in the same way as described above for the Renilla luciferase assay system. Si21Luc exhibited a dramatic attenuation of RNAi efficacy with time. In contrast, the aliquot samples of ds27Luc, C16-si21Luc, and C16-ds27Luc maintained an inhibitory effect even though the sample was incubated for a long time. Particularly, C16-ds27Luc maintained strong genesilencing efficacy throughout a long incubation time and exhibited 80% knockdown of Renilla gene expression using an aliquot sample after 48 h treatment. There was a linear correlation between high nuclease resistance and the longevity of the RNAi efficacy of lipid-conjugated dsRNAs. RNAi Efficacy of C16-Conjugated dsRNAs Targeted to the Endogenous VEGF Gene. To evaluate RNAi efficacy against the VEGF gene, we newly designed and synthesized si21VEGF, ds27VEGF, C16-si21VEGF, and C16-ds27VEGF. VEGF plays a critical role in angiogenesis and has been linked to tumor growth and metastasis.33,34 We investigated the RNAi efficacy of si21VEGF and ds27VEGF, including C16 conjugates, in either the presence or absence of LF2000 in HeLa, A549, and SH-10-TC cells, in each of which the gene is constitutively active. The resultant VEGF mRNA expression was analyzed by RT-PCR assay after 48 h of incubation. In the presence of LF2000, the VEGF mRNA contents were reduced by all of the dsRNAs including C16 conjugates in all three cell lines (Figure 6). The C16 conjugates had stronger inhibitory effects than nonmodified dsRNAs. In particular, the gene-silencing efficacy of the C16ds27VEGF was much more intensified in all three cell lines compared with that of the C16-si21VEGF. In the absence of LF2000, the VEGF mRNA expression was not remarkably

Figure 5. Nuclease resistance and its relationship to the RNAi efficacy of C16-conjugated dsRNAs. (A) Nuclease resistance of C16conjugated dsRNAs. dsLuc, including C16 conjugates, was incubated in culture medium containing 10% FBS at 37 °C for 0, 0.5, 1, 2, 4, 6, 8, 12, 24, or 48 h, and aliquots were analyzed on 20% PAGE and RPHPLC. The mean and SD values are from three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 at 48 h of incubation time (t test). (B) RNAi efficacy of C16-conjugated dsRNAs. The aliquots containing dsRNAs (1 nM) after treatment with 10% FBS for different time intervals (0, 1, 4, 8, 24, or 48 h) were investigated for RNAi efficacy in the presence of LF2000. The mean and SD values are from three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (t test).

inhibited by si21VEGF or ds27VEGF even at a concentration as high as 5 μM. In contrast, C16-si21VEGF and C16ds27VEGF significantly reduced VEGF mRNA expression at a concentration of 5 μM (Figure 7). C16-ds27VEGF especially exhibited strong gene-silencing efficacy in all cell lines. Membrane Permeability of C16-Conjugated dsRNAs. The membrane permeability of dsRNAs, including C16 conjugates, was investigated in HeLa, A549, and SH-10-TC cell lines in the presence or absence of LF2000 using confocal microscopy and flow cytometry. In the observation by confocal microscopy, the cells treated with dsRNAs including C16 conjugates, all of which were labeled with FAM at the 5′-end of the antisense strand, exhibited bright fluorescence in the presence of LF2000 (Figure 8). Among them, both C16si21RNA and C16-ds27RNA labeled with FAM exhibited extremely high fluorescence intensity in all cell lines. We also 1379

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

Figure 7. RNAi efficacy of dsRNAs including C16 conjugates (5 μM) targeting VEGF in HeLa (A), A549 (B), and SH-10-TC (C) cell lines in the absence of LF2000 evaluated by measuring VEGF mRNA quantitatively using RT-PCR. GAPDH mRNA was also measured as an intrinsic control. dsVEGFs including C16 conjugates were transfected into HeLa cells and then incubated for 48 h. The mean and SD values are from three independent experiments. *, P < 0.05; **, P < 0.01 (t test).

Figure 6. RNAi efficacy of dsRNAs, including C16 conjugates (100 nM), targeting VEGF in HeLa (A), A549 (B), and SH-10-TC (C) cell lines in the presence of LF2000 evaluated by measuring VEGF mRNA quantitatively using RT-PCR. GAPDH mRNA was also measured as an intrinsic control. dsVEGFs including C16 conjugates were transfected into HeLa cells with LF2000 and then incubated for 48 h. The mean and SD values are from three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (t test).

treated with C16 conjugates labeled with FAM exhibited higher fluorescence-intensity populations than the histograms of the cells treated with nonconjugated dsRNAs. Especially, high percentages of cells treated with C16ds27RNA labeled with FAM were distributed at highfluorescence-intensity populations in the histograms in both the presence and absence of LF2000. Accordingly, the C16 conjugates showed enhanced cell membrane permeability in culture cells, as predicted by the properties of palmitic acid.

checked the membrane permeability of dsRNAs, including C16 conjugates, in the absence of LF2000. Neither si21RNA nor ds27RNA clearly show fluorescence signals, whereas C16si21RNA and C16-ds27RNA showed weak fluorescence signals in each cell line (Figure 9). The results of flow cytometric analysis showed the same tendency as the microscope observations. In the flow cytometric analysis, the area of the histogram is separated into four populations (a−d) along with the fluorescence intensity. The histograms of the cells 1380

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

Figure 8. Confocal microscopy images and flow cytometric analysis of HeLa (A), A549 (B), and SH-10-TC (C) cells incubated for 6 h with dsRNAs, including C16 conjugates (100 nM) labeled with FAM, in the presence of LF2000. FL, fluorescence image; Merge, merged image of FL and Trans; FC, flow cytometric analysis. In the flow cytometric analysis, the logarithm of the fluorescence intensity is shown on the horizontal axis, and the number of cells is shown on the longitudinal axis. (D) Cell population is separated into four parts (a−d) along with the fluorescence intensity.

Figure 9. Confocal microscopy images and flow cytometric analysis of HeLa (A), A549 (B), and SH-10-TC (C) cells incubated for 6 h with dsRNAs, including C16 conjugates (5 μM) labeled with FAM, in the absence of LF2000. FL, fluorescence image; Merge, merged image of FL and Trans; FC, flow cytometric analysis. In the flow cytometric analysis, the logarithm of the fluorescence intensity is shown on the horizontal axis, and the number of cells is shown on the longitudinal axis. (D) Cell population is separated into four parts (a−d) along with the fluorescence intensity.

■

DISCUSSION High RNAi potency of lipid-conjugated siRNA molecules, such as cholesterol and fatty acid, has been reported both in vitro and in vivo.23−26 The lipids, including cholesterol, bile acids, and long-chain fatty acids conjugated with siRNAs,

facilitated cellular penetration of the siRNAs through their interaction with the lipid bilayer of the cell membrane, lipoprotein particles, lipoprotein receptors, and transmembrane proteins.23−25 However, in the literature, the direct conjugation of lipids to the siRNAs has been limited by the conjugating 1381

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

RNAi efficacy of C16-dsRNAs against the endogenous VEGF gene in three cell lines (HeLa, A549, and SH-10-TC) in the presence or absence of LF2000. Both C16-si21VEGF and C16ds27VEGF suppressed VEGF gene expression with high efficacy in all three cell lines. In particular, C16-ds27VEGF showed the most enhanced RNAi efficacy in both the presence and absence of LF2000. Hence, the results suggested that the C16-ds27RNAs were promising for RNAi, because they exhibited powerful gene silencing against not only the exogenous gene but also the endogenous gene. Efficient cellular uptake of FAM-labeled C16-dsRNAs was observed in three different cell lines (HeLa, A549, and SH-10TC) in the presence of LF2000. Weak fluorescence signals were observed in the cells treated with FAM-labeled C16-dsRNAs in the absence of LF2000, whereas the cells treated with nonmodified dsRNAs did not show any fluorescence signals. This efficient membrane permeability of C16-dsRNAs is attributable to its satisfactory interaction with both the cellular membrane and palmitic acid. These results suggested that the properties of efficient cellular uptake were one of the reasons why C16ds27RNAs enhanced RNAi efficacy. In conclusion, the C16-ds27RNAs prepared by our simple synthesis strategy exhibited powerful gene-silencing activity to promote the RNAi mechanisms, by virtue of their enhanced membrane permeability and accelerated RISC selection through the Dicer cleavage. Although C16-si21RNAs have excellent properties as RNAi molecules, C16-ds27RNAs excel in all things, such as recognition by Dicer, nuclease resistance, cell membrane permeability, and RNAi efficacy, compared with C16-si21RNA. We are certain that C16-ds27RNAs can be useful as a new generation of RNAi molecules that overcome some of the limitations of RNAi technology and that are applicable in vivo.

position to the siRNA, mostly to the 3′-end using lipophilebearing solid supports24 or to the 5′-end using phosphoramidite derivatives;25 both of these processes involve complicated syntheses. We previously reported a simple and efficient synthesis and demonstrated potent RNAi efficacy of lipidconjugated 21-nt siRNAs.35 However, lipid-conjugated 21-nt siRNA may, in part, obstruct certain RISC functions. In this study, we focused on the properties of blunt-ended 27-nt dsRNA, and we developed lipid-conjugated 27-nt dsRNAs to overcome the problems of siRNAs. We expected the 27-nt dsRNAs conjugated with lipids to have high membrane permeability and a Dicer substrate cleaved to the 21-nt siRNA, leading to the release of lipid molecules. The conjugation of the lipids to dsRNAs was restricted to the 5′-end of the sense strand in our previous reports.29−31 We initially performed comparative studies of Dicer cleavage and gene silencing using lipid-conjugated si21RNAs and ds27RNAs. The lipid-conjugated ds27RNAs exhibited the 21nt siRNA product after reaction with Dicer, whereas the lipidconjugated si21RNAs were not cleaved by Dicer. These results suggested that the lipid molecules conjugated to the 5′-end of 27-nt dsRNA did not obstruct Dicer recognition. Kim et al. reported that the cleavage of a 27-nt dsRNA by Dicer inside cells could result in a variety of distinct 21-nt siRNA products and that a mix of these possible 21-nt siRNAs is significantly more potent than the specific 21-nt siRNA.27 They also discussed that a 27-nt dsRNA had a stronger RNAi efficacy than the seven different synthesized 21-nt siRNA sets that could be derived from the cleavage of a 27-nt dsRNA by Dicer. Thus, the recognition of Dicer to the siRNAs is an essential element for obtaining potent RNAi efficacy. To evaluate the gene-silencing efficacy of lipid conjugates, we selected an exogenous Renilla luciferase as a target gene, because the luciferase reporter assays were among the conventional approaches for RNAi. Although C16-si21Luc has a potent RNAi efficacy, C16-ds27Luc was determined to have the highest RNAi potential in HeLa cells in the presence or absence of LF2000. Although the literature describes excellent RNAi potency of the cholesterol-conjugated siRNAs both in vitro and in vivo,23 our data suggested that the C16-ds27RNAs used in this study are certain to become a new generation of RNAi therapeutic agents rather than cholesterol-conjugated siRNAs. These intensified inhibitory effects of C16-ds27Luc are attributed to their high stability against nuclease degradation (Figure 5). Another report also described that the lipidconjugated siRNAs survived to a much greater extent than nonmodified siRNAs as a result of their interaction with lipoprotein particles.24 In our results, C16-ds27Luc exhibited enhanced nuclease resistance and RNAi efficacy compared with that of C16-si21Luc. In addition, C16-ds27Luc has stable genesilencing efficacy with high potency even after long-term treatment with nuclease. Thus, the strong resistance of siRNA against nuclease degradation is one of the essential elements for longevity in RNAi therapy. Therefore, the Dicer-cleavable lipidconjugated dsRNAs, such as C16-ds27RNAs in this study, could be more prominent than Dicer-uncleavable lipidconjugated siRNAs, such as C16-si21RNAs in this study. For further improvements, we evaluated the RNAi potency of C16-dsRNAs against the endogenous VEGF gene. Although the matter of gene silencing against an exogenous gene by siRNA is important, the investigation of gene-silencing efficacy against the endogenous gene is essential for evaluating siRNA molecules as therapeutic tools. Accordingly, we investigated the

■

ASSOCIATED CONTENT

S Supporting Information *

Figures S1 and S2: HPLC profiles and molecular weight confirmation by MALDI-TOF mass spectrometry. This information is available free of charge via the Internet at http://pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Author

*Yasuda Women’s University, Faculty of Pharmacy, Department of Life Sciences, 6-13-1 Yasuhigashi, Asaminami-ku, Hiroshima 731-0153, Japan. Tel: +81-82-878-9473. Fax: +8182-878-9540. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported in part by a Grant-in-Aid for Young Scientists (B) from the Japan Society for the Promotion of Science (JSPS).

■

REFERENCES

(1) Fire, A.; Xu, S.; Montgomery, M. K.; Kostas, S. A.; Driver, S. E.; Mello, C. C. Potent and specific genetic interference by doublestranded RNA in Caenorhabditis elegans. Nature 1998, 391, 806−811. (2) Ryther, R. C.; Flynt, A. S.; Phillips, J. A.; Patton, J. G. siRNA therapeutics: big potential from small RNAs. Gene Ther. 2005, 12, 5− 11. (3) Kim, D.; Rossi, J. RNAi mechanisms and applications. Biotechniques 2008, 44, 613−616. 1382

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383

Molecular Pharmaceutics

Article

(4) Honma, K.; Iwao-Koizumi, K.; Takeshita, F.; Yamamoto, Y.; Yoshida, T.; Nishio, K.; Nagahara, S.; Kato, K.; Ochiya, T. RPN2 gene confers docetaxel resistance in breast cancer. Nat. Med. 2008, 14, 939− 948. (5) Geisbert, T. W.; Lee, A. C.; Robbins, M.; Geisbert, J. B.; Honko, A. N.; Sood, V.; Johnson, J. C.; Jong, S.; Tavakoli, I.; Judge, A.; Hensley, L. E.; MacLachlan, I. Postexposure protection of non-human primates against a lethal Ebola virus challenge with RNA interference: a proof-of-concept study. Lancet 2010, 375, 1896−1905. (6) Myers, J. W.; Jones, J. T.; Meyer, T.; Ferrell, J. E. Recombinant Dicer efficiently converts large dsRNAs into siRNAs suitable for gene silencing. Nat. Biotechnol. 2003, 21, 324−328. (7) Macrae, I. J.; Zhou, K.; Li, F.; Repic, A.; Brooks, A. N.; Cande, W. Z.; Adams, P. D.; Doudna, J. A. Structural basis for double-stranded RNA processing by Dicer. Science 2006, 311, 195−198. (8) Silhavy, D.; Molnar, A.; Lucioli, A.; Szittya, G.; Hornyik, C.; Tavazza, M.; Burgyan, J. A viral protein suppresses RNA silencing and binds silencing-generated, 21- to 25-nucleotide double-stranded RNAs. EMBO J. 2002, 21, 3070−3080. (9) Hornung, V.; Guenthner-Biller, M.; Bourquin, C.; Ablasser, A.; Schlee, M.; Uematsu, S.; Noronha, A.; Manoharan, M.; Akira, S.; Fougerolles, A.; Endres, S.; Hartmann, G. Sequence-specific potent induction of IFN-alpha by short interfering RNA in plasmacytoid dendritic cells through TLR7. Nat. Med. 2005, 11, 263−270. (10) Elbashir, S. M.; Harborth, J.; Lendeckel, W.; Yalcin, A.; Weber, K.; Tuschl, T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001, 411, 494−498. (11) Amarzguioui, M.; Holen, T.; Babaie, E.; Prydz, H. Tolerance for mutations and chemical modifications in a siRNA. Nucleic Acids Res. 2003, 31, 589−595. (12) Braasch, D. A.; Jensen, S.; Liu, T.; Kaur, K.; Arar, K.; White, M. A.; Corey, D. R. RNA interference in mammalian cells by chemicallymodified RNA. Biochemistry 2003, 42, 7967−7975. (13) Sipa, K.; Sochacka, E.; Kazmierczak-Baranska, J.; Maszewska, M.; Janicka, M.; Nowak, G.; Nawrot, B. Effect of base modifications on structure, thermodynamic stability, and gene silencing activity of short interfering RNA. RNA 2007, 13, 1301−1316. (14) Czauderna, F.; Fechtner, M.; Dames, S.; Aygun, H.; Klippel, A.; Pronk, G. J.; Giese, K.; Kaufmann, J. K. Structural variations and stabilising modifications of synthetic siRNAs in mammalian cells. Nucleic Acids Res. 2003, 31, 2705−2716. (15) Chiu, Y. L.; Rana, T. M. RNAi in human cells: basic structural and functional features of small interfering RNA. Mol. Cell 2002, 10, 549−561. (16) Grunweller, A.; Wyszko, E.; Bieber, B.; Jahnel, R.; Erdmann, V. A.; Kurreck, J. A. Comparison of different antisense strategies in mammalian cells using locked nucleic acids, 2′-O-methyl RNA, phosphorothioates and small interfering RNA. Nucleic Acids Res. 2003, 12, 3185−3193. (17) Hall, A. H.; Wan, J.; Spesock, A.; Sergueeva, Z.; Shaw, B. R.; Alexander, K. A. High potency silencing by single-stranded boranophosphate siRNA. Nucleic Acids Res. 2006, 34, 2773−2781. (18) Jung, S.; Lee, S. H.; Mok, H.; Chung, H. J.; Park., T. G. Gene silencing efficiency of siRNA-PEG conjugates: Effect of PEGylation site and PEG molecular weight. J. Controlled Release 2010, 144, 306− 313. (19) Oishi, M.; Nagasaki, Y.; Itaka, K.; Nishiyama, N.; Kataoka, K. Lactosylated poly(ethylene glycol)-siRNA conjugate through acidlabile beta-thiopropionate linkage to construct pH-sensitive polyion complex micelles achieving enhanced gene silencing in hepatoma cells. J. Am. Chem. Soc. 2005, 127, 1624−1625. (20) Moschos, S. A.; Jones, S. W.; Perry, M. N.; Williams, A. E.; Erjefalt, J. S.; Turner, J. J.; Barnes, P. J.; Sproat, B. S.; Gait, M. J.; Lindsay, M. A. Lung delivery studies using siRNA conjugated to TAT(48−60) and Penetratin reveal peptide induced reduction in gene expression and induction of innate immunity. Bioconjugate Chem. 2007, 18, 1450−1459.

(21) Turner, J. J.; Jones, S.; Fabani, M. M.; Ivanova, G.; Arzumanov, A. A.; Gait, M. J. RNA targeting with peptide conjugates of oligonucleotides, siRNA and PNA. Blood Cells Mol. Dis. 2007, 38, 1−7. (22) Detzer, A.; Overhoff, M.; Wunsche, W.; Rompf, M.; Turner, J. J.; Ivanova, G. D.; Gait, M. J.; Sczakiel, G. Increased RNAi is related to intracellular release of siRNA via a covalently attached signal peptide. RNA 2009, 15, 627−636. (23) Soutschek, J.; Akinc, A.; Bramlage, B.; Charisse, K.; Constien, R.; Donoghue, M.; Elbashir, S.; Geick, A.; Hadwiger, P.; Harborth, J.; John, M.; Kesavan, V.; Lavine, G.; Pandey, P. K.; Racie, T.; Rajeev, K. G.; Röhl, I.; Toudjarska, I.; Wang, G.; Wuschko, S.; Bumcrot, D.; Koteliansky, V.; Limmer, S.; Manoharan, M.; Vornlocher, H. P. Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature 2004, 432, 173−178. (24) Nishina, K.; Unno, T.; Uno, Y.; Kubodera, T.; Kanouchi, T.; Mizusawa, H.; Yokota, T. Efficient in vivo delivery of siRNA to the liver by conjugation of alpha-tocopherol. Mol. Ther. 2008, 16, 734−740. (25) Wolfrum, C.; Shi, S.; Jayaprakash, K. N.; Jayaraman, M.; Wang, G.; Pandey, R. K.; Rajeev, K. G.; Nakayama, T.; Charrise, K.; Ndungo, E. M.; Zimmermann, T.; Koteliansky, V.; Manoharan, M.; Stoffel, M. Mechanisms and optimization of in vivo delivery of lipophilic siRNAs. Nat. Biotechnol. 2007, 25, 1149−1157. (26) Wong, N.; Liu, Z.; Dai, H. Functionalization of carbon nanotubes via cleavable disulfide bonds for efficient intracellular delivery of siRNA and potent gene silencing. J. Am. Chem. Soc. 2005, 127, 12492−12493. (27) Kim, D. H.; Behlke, M. A.; Rose, S. D.; Chang, M. S.; Choi, S.; Rossi, J. J. Synthetic dsRNA Dicer substrates enhance RNAi potency and efficacy. Nat. Biotechnol. 2005, 23, 222−226. (28) Marques, J. T.; Devosse, T.; Wang, D.; Zamanian-Daryoush, M.; Serbinowski, P.; Hartmann, R.; Fujita, T.; Behlke, M. A.; Williams, B. R. A structural basis for discriminating between self and nonself double-stranded RNAs in mammalian cells. Nat. Biotechnol. 2006, 24, 559−565. (29) Kubo, T.; Zhelev, Z.; Ohba, H.; Bakalova, R. Modified 27-nt dsRNAs with dramatically enhanced stability in serum and long-term RNAi activity. Oligonucleotides 2007, 17, 445−464. (30) Kubo, T.; Zhelev, Z.; Bakalova, R.; Ohba, H. Enhancement of gene silencing potency and nuclease stability by chemically modified duplex RNA. Nucleic Acids Symp. Ser. 2007, 51, 407−408. (31) Kubo, T.; Zhelev, Z.; Bakalova, R.; Ohba, H. Highly Efficient Gene Suppression by Chemically Modified 27 Nucleotide DoubleStranded RNAs. Jpn. J. Appl. Phys. 2008, 47, 1346−1350. (32) Turner, J. J.; Jones, S. W.; Moschos, S. A.; Lindsay, M. A.; Gait, M. J. MALDI-TOF mass spectral analysis of siRNA degradation in serum confirms an RNAse A-like activity. Mol. Biosyst. 2007, 3, 43−50. (33) Kim, K. J.; Li, B.; Winer, J.; Armanini, M.; Gillett, N.; Phillips, H. S.; Ferrara, N. Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo. Nature 1993, 362, 841−844. (34) Takei, Y.; Kadomatsu, K.; Yuzawa, Y.; Matsuo, S.; Muramatsu, T. A small interfering RNA targeting vascular endothelial growth factor as cancer therapeutics. Cancer Res. 2004, 64, 3365−3370. (35) Kubo, T.; Yanagihara, K.; Takei, Y.; Mihara, K.; Morita, Y.; Seyama, T. Palmitic acid-conjugated 21-nt siRNA enhances genesilencing activity. Mol. Pharm. 2011, 8, 2193−2203.

1383

dx.doi.org/10.1021/mp2006278 | Mol. Pharmaceutics 2012, 9, 1374−1383