Evaluation of Locked Nucleic Acid–Modified Small Interfering RNA in Vitro and in Vivo

833

Evaluation of locked nucleic acid–modified small interfering RNA in vitro and in vivo

Olaf R. Mook, Frank Baas, Marit B. de Wissel, model. Therefore, LNA-modified siRNA should be pre- and Kees Fluiter ferred over unmodified siRNA. [Mol Cancer Ther 2007; 6(3):833–43] Department of Neurogenetics, Academic Medical Center, Amsterdam, the Netherlands Introduction RNA interference (RNAi) is a natural process that affects Abstract gene silencing in eukaryotic systems at transcriptional, RNA interference has become widely used as an experi- posttranscriptional, and/or translational levels (1). Small mental tool to study gene function. In addition, small interfering RNA (siRNA) molecules are the key intermedi- interfering RNA (siRNA) may have great potential for the ates in this process, which can potentially inhibit the treatment of diseases. Recently, it was shown that siRNA expression of any given target gene. siRNA molecules hold can be used to mediate gene silencing in mouse models. great promise as biological tools and as potential thera- Locally administered siRNAs entered the first clinical trials, peutic agents for targeted inhibition of disease-causing but strategies for successful systemic delivery of siRNA genes. are still under development. Challenges still exist about the To optimize the use of siRNA as a therapeutic therapy, stability, delivery, and therapeutic efficacy of siRNA. In several modes of delivery have been tested in vivo. the present study, we compare the efficacy of two Improved delivery in animals has been achieved by methods of systemic siRNA delivery and the effects of complexation with cationic liposomes (2), polyethyleni- siRNA modifications using locked nucleic acids (LNA) in a mine (3), and Arg-Gly-Asp–polyethylene glycol–polyethy- xenograft cancer model. Low volume tail vein bolus lenimine (4) or by conjugation of cholesterol to siRNA (5). injections and continuous s.c. delivery using osmotic However, Ma et al. (6) has shown that complexation to minipumps yielded similar uptake levels of unmodified cationic liposomes resulted in a potent induction of both siRNA by tumor xenografts. Both routes of administration type I and type II IFN responses in mice. In line, sequence- mediated sequence-specific inhibition of two unrelated specific and Toll-like receptor-7–dependent induction of targets inside tumor xenografts. Previous studies have IFNs has been shown on liposome-mediated siRNA shown that LNA can be incorporated into the sense strand transfection in mice (7). Furthermore, conjugation of of siRNA while the efficacy is retained. Modification of cholesterol to a siRNA against apolipoprotein B could siRNA targeting green fluorescent protein with LNA results specifically degrade its mRNA in vivo (5). in a significant increase in serum stability and thus may be Another approach to improve potency and efficacy of beneficial for clinical applications. We show that minimal siRNA in vivo is by the introduction of other chemical 3¶ end LNA modifications of siRNA are effective in modifications in siRNA. In vitro studies showed that sev- stabilization of siRNA. Multiple LNA modifications in the eral modifications are allowed in functional siRNAs. Modi- ¶ accompanying strand further increase the stability but fications of phosphorothioate (8, 9) 2 -O-methyl (10, 11), ¶ ¶ negate the efficacy in vitro and in vivo. In vivo, LNA- 2 -O-allyl (10), and 2 -deoxy-fluorouridine (8, 9) have been modified siRNA reduced off-target gene regulation com- examined for potential in vivo use. Some of the modified pared with nonmodified siRNA. End-modified siRNA siRNAs were found to exhibit enhanced serum stability (11) ¶ targeting green fluorescent protein provides a good and longer duration of action (10). Modification of the 5 end ¶ trade-off between stability and efficacy in vivo using the of the antisense strand with 2 -O-allyl (10) or chemical ¶ two methods of systemic delivery in the nude mouse blocking of the 5 -hydroxyl group (11) resulted in a dramatic loss in activity consistent with the proposed in vivo requirement for 5¶ end phosphorylation (12). In addition, more substantial modifications, such as total modification by 2¶-O-methyl (8) or phosphorothioate modifications of Received 4/10/06; revised 11/23/06; accepted 1/31/07. every second or all internucleoside linkages (8, 9), increased Grant support: Dutch Cancer Society project no. 2003-2968. cytotoxic effects and resulted in a significant decrease or The costs of publication of this article were defrayed in part by the complete loss of activity. In contrast, totally modified payment of page charges. This article must therefore be hereby marked duplexes containing a combination of 2¶-O-methyl and advertisement in accordance with 18 U.S.C. Section 1734 solely to 2¶-fluoro modifications were shown to be effective in vitro indicate this fact. and in vivo (13). Requests for reprints: Olaf R. Mook, Department of Neurogenetics, Academic Medical Center, Meibergdreef 9, Amsterdam, the Netherlands Locked nucleic acid (LNA) is a novel nucleotide analogue 1105 AZ. Phone: 312-056-64540. E-mail: [email protected] that contains a methylene bridge that connects the 2¶- Copyright C 2007 American Association for Cancer Research. oxygen of the ribose with the 4¶-carbon. The bicyclic doi:10.1158/1535-7163.MCT-06-0195 structure locks the furanose ring of the LNA molecule in

Mol Cancer Ther 2007;6(3). March 2007

a3¶-endo conformation, thereby structurally mimicking the AS 5¶-gaugaacuucagggucagcTT-3¶ and heavily modified standard RNA monomers. LNA induce an increase in siGFP SS 5¶-GCTgacCcuGaagTTcaucTT-3¶ and AS 5¶-gau- ¶ thermal stability (melting temperature, Tm) when bound to gaacuucagggucagcTT-3 (LNA nucleotides are depicted in a matching RNA sequence. Introduction of LNA into capitals) were synthesized by Santaris A/S (Hørsholm, classic antisense oligonucleotides has been shown to Denmark) as described previously (15). increase its serum stability (14). In analogy, Braasch et al. Cell Line (8) showed that LNA can be used to thermally stabilize The pancreatic cancer cell line MiaPaca-II and MiaPaca-II siRNAs without losing their function. Recently, a system- stably expressing eGFP (17) were maintained at 37jC and atic study on LNA containing siRNAs has identified the 5% CO2 by serial passage in DMEM supplemented with number and positions of LNA molecules within the siRNA, 10% FCS, 2 mmol/L L-glutamine, 100 units/mL penicillin, which still allow a functional siRNA (15). Incorporation of and 100 Ag/mL streptomycin. LNA molecules into siRNA significantly increased its Application of Unmodified and LNA-Modified siRNA serum stability, which potentially favors successful in vivo In vivo applications (15). Furthermore, LNA modifications at the S.c. tumors were induced in 8- to 10-week-old NMRI nu/ 5¶ end of the sense strand of siRNA has been reported to nu mice (Charles River, Maastricht, the Netherlands) as favor incorporation of the antisense strand into the RNA- described previously (14). One week after tumor cell induced silencing complex (RISC), thereby reducing se- injection, when tumor take was positive, administration quence related off-target effects (15). A second potential of the siRNA or LNA-modified siRNA started. benefit of LNA-modified siRNA is the protection of the two For POLR2A inhibition, siPOLR2A and the mismatch nt 3¶ overhang of siRNA. Recently, it has been shown that control containing two central mismatches were adminis- in certain cell types blunt-ended double-stranded RNA tered via tail vein injections at a dosage of 0.15 mg/kg (200 (dsRNA; lacking 3¶ overhangs) <30 bp induced dsRNA- ALof1.4Amol/L siPOLR2A solution) twice weekly for 3 mediated signaling (16). Therefore, protection of those weeks as described previously (18). During treatment, overhangs with LNA potentially increases the specificity of tumor growth was monitored as described previously (17). siRNA. In all experiments targeting GFP, administration was In this study, we show that systemic administration of either via tail vein injection as described above or via unmodified siRNA induced specific RNAi effects of two osmotic minipumps (model 1007D; Alzet Corp., Palo Alto, independent targets in tumor xenografts. Different routes CA) dosed at 0.25 mg/kg/d for 7 days. For each treatment, of administration resulted in comparable target knock- five mice per group were used. All animal experiments down. To test whether LNA-modified siRNA contributes to were conducted under the institutional guidelines and the efficacy in vivo, we have designed an end-modified according to the law; they were sanctioned by the animal siRNA targeting green fluorescent protein (siGFP) and a ethics committee. heavily modified siGFP and compared their characteristics Whole-Body Imaging and Tissue Processing with unmodified siGFP both in vitro and in vivo. In the eGFP fluorescence was visualized with whole-body present article, we show that minimal modification of imaging using the GFP fluorescence mode of a LAS3000 siGFP with LNA greatly enhanced its serum stability and (Fuji, Tokyo, Japan). Parts of the tumors were fixed in was compatible with the silencing machinery. End-modi- formaldehyde (4%)/sucrose (20%). Other parts of the fied siGFP effectively lowered its target in vivo after different routes of administration. Target knockdown was not associated with dsRNA-dependent protein kinase activation or induction of the IFN response. Introduction of LNA into siRNA resulted in significantly less off-target regulated genes. These results showthat end-modified siRNA holds promise for in vivo applications.

Materials and Methods siRNA and LNA-Modified siRNA Synthesis siRNA against large subunit of RNA polymerase II (siPOLR2A) SS 5¶-gcugcgcuauggcgaagacgg-3¶ and AS 5¶- gucuucgccauagcgcagctg-3 and the mismatch control siPOLR2A mismatch SS 5¶-gcugcgcuacugcgaagacgg-3¶ and AS 5¶-gucuucgcaguagcgcagctg-3¶ (DNA nucleotides are depicted in italics) were purchased from Proligo (Boulder, Figure 1. Tritium-labeled siRNA was administered via the tail vein at a CO). siRNA against enhanced GFP (eGFP; siGFP) SS 5¶- dose of 0.15 mg/kg (white columns), and the mice were sacrificed after ¶ ¶ 30 min. In a second experiment, tritium-labeled siRNA was administered gcugacccugaaguucauctt-3 and AS 5 -gaugaacuucagggu- black columns ¶ via osmotic minipumps dosed at 0.25 mg/kg/d ( ), and mice cagctt-3 were purchased from MedProbe (Lund, Sweden). were sacrificed after 48 h. After sacrifice, the amount of radioactivity was End-modified siGFP SS 5¶-GcugacccugaaguucaucTT-3¶ and determined in various organs.

MolCancerTher2007;6(3).March2007

Figure 3. Unmodified siGFP, end-modified siGFP, and heavily modified siGFP were incubated in fresh mouse serum. At the indicated time points (hours), aliquots were withdrawn, mixed with formamide loading dye, and stored at À20jC. Then, samples were run on 16% reducing AA gels. siGFPs were visualized with ethidium bromide staining.

lysis buffer (PBS; 1% Triton X-100, 0.01% sodium azide) 72 h posttransfection. Total RNA was isolated with Trizol (Life Technologies, Gaithersburg, MD) according to the manufacturer’s instructions 24 h posttransfection. Western Blots Tumor tissues were homogenized by polytronic disper- sion (400 AL/100 Ag) in lysis buffer (PBS; 1% Triton X-100, 0.01% sodium azide). Cell extracts and tumor homogenates were subjected to 10% SDS-PAGE, and the resolved Figure 2. A, siPOLR2A and siPOLR2A mismatch were administered via the tail vein, dosed at 0.15 mg/kg twice weekly for 3 wks. During proteins were transferred electrophoretically to polyviny- treatment, tumor growth was monitored. B, unmodified siGFP was lidene difluoride membranes (Invitrogen). eGFP was administered either via tail vein injections or via osmotic minipump, and detected with a rabbit anti-GFP polyclonal antibody after sacrifice, eGFP expression was quantified on Western blots and normalized for elongation factor 2a (eEF2) expression. (Molecular Probes, Eugene, OR). Elongation factor 2a (Cell Signaling, Beverly, MA) was used as loading control. Chemiluminescent detection was done on a LAS3000 in accordance with the manufacturer’s instructions. eGFP tumor were used to prepare protein lysates. The rest of the signals were quantified using Aida software version 3.44 tumor tissue was snap frozen in liquid nitrogen and stored and normalized to those of elongation factor 2a. at À80jC until further use. Northern Blots Biodistribution of Unmodified siRNA after Different The RNA isolation was according to the manufacturer’s Routes of Administration procedure. RNA was denatured using glyoxal and sepa- Tissue distribution studies of tritiated siGFP were done rated on 1% agarose gels following standard protocols. according to Bijsterbosch et al. (19). Distribution was RNA was subsequently transferred to Hybond-N+ mem- studied after 30 min of circulation of a bolus injection brane (Amersham, Piscataway, NJ) in 20Â SSC. Following (200 AL) of tritiated siGFP (1.4 Amol/L; 0.15 mg/kg) and transfer, the RNA was UV cross-linked and the membrane after continuous s.c. administration of 0.5 AL/h of tritiated siGFP (40 Amol/L; 0.25 mg/kg/d) for 2 days using osmotic minipumps. Distribution was calculated as disintegrations per minute per gram tissue present at the different organs at the time of sacrifice. Serum Incubations siGFP and LNA-modified siGFP (10 Amol/L) were incubated at 37jC in fresh mouse serum. Aliquots of 4 AL were withdrawn after 2.5, 5.0, 24, 48, 72, and 96 h of incubation, mixed 1:1 with formamide loading dye, and stored at À20jC until used. Samples were analyzed on 16% denaturing polyacrylamide gels. Gels were stained with ethidium bromide, and siGFP was visualized on a LAS3000 and quantified using Aida software version 3.44 (Raytest Benelux, Tilburg, the Netherlands). Transfections Figure 4. MiaPaCa-II cells were transfected with unmodified siGFP, Transfections were done in six-well culture plates with end-modified siGFP, and heavily modified siGFP. A, at 24 h posttransfection, total RNA was isolated and eGFP mRNA was analyzed on Northern LipofectAMINE 2000 (Invitrogen, Carlsbad, CA) as liposo- blots. B, 72 h posttransfection, cells were harvested, protein lysates were mal transfection agent. Protein samples were prepared in prepared, and eGFP protein was analyzed on Western blots.

Mol Cancer Ther 2007;6(3). March 2007

Figure 5. Unmodified siGFP, end- modified siGFP, and heavily modified siGFP were administered at a dosage of 0.25 mg/kg/d via osmotic minipumps. A, 7 d after implantation of the pumps, eGFP fluorescence in the tumors was analyzed using whole- body imaging. B, eGFP protein levels were determined by Western blot and normalized for elongation factor 2a expression. C, eGFP fluorescence in situ was determined with confocal microscopy.

was baked for 4 h at 80jC. For Northern blot analyses, the correction for multiple testing) was used to detect BcgI/HindIII fragment of pEGFP-C1 was used as a probe. significant differences. Genes with an ANOVA P value A 28S probe was used as loading control. Hybridizations <0.05 and a fold change greater than 3 and less than À3 and posthybridization washes were according to Church were considered significantly regulated. and Gilbert (20). GFP Fluorescence In situ Results Tumor tissue was fixed in 4% formaldehyde in PBS containing 20% (w/v) sucrose for 24 h at 4jC and sectioned Comparison of Biodistributions after Different (10 Am thick) at a cabinet temperature of À34jC and stored Routes of Administration at À20jC until use. Before use, sections were washed twice We have done distribution studies of siGFP to test if in PBS and embedded in Vectashield. GFP fluorescence different routes of administration at different dosages was recorded using standardized setting of the confocal could deliver siGFP into various organs and tumors. We laser scanning microscope (TCS SP2) fitted to a DM-IRB have compared these two routes because administration of inverted microscope (Leica, Mannheim, Germany). Excita- antisense oligonucleotides via osmotic pumps is a validat- tion of eGFP was done at 488 nm, and fluorescence was ed route of administration in our model. We compared captured at 500 to 530 nm. GFP fluorescence was quantified that with low volume i.v. bolus injection of siRNA because with ImageJ 1.32 software (W.S. Rasband, ImageJ,1 NIH, it has been shown recently by Duxbury et al. (18) to be Bethesda, MD; 1997–2005). an effective way to mediate RNAi in vivo at very low Expression Profiling dosages. Administration of radiolabeled siGFP dosed at Total RNA from tumors of mice that received saline, 0.25 mg/kg/d via osmotic minipumps delivered siGFP siGFP, or end-modified siGFP (three tumors per group) via into a large number of tissues (Fig. 1). Administration of osmotic pumps dosed at 0.25 mg/kg/d was isolated in 0.15 mg/kg siRNA injected i.v. as a single bolus resulted Trizol according to the manufacturer’s instructions. RNA in high uptake in the kidney. Liver and spleen were organs was further purified with Macherey-Nagel RNA spin with relatively high uptake after systemic administra- columns (Hoerdt, France) and DNase-treated RNA was tion (Fig. 1). Comparing administration via osmotic minipump and bolus injection, it is clear that bolus injection eluted with RNase-free H2O. The quantity and quality of the RNA was assessed with a spectrophotometer (ND1000; results in higher delivery in a large number of organs. NanoDrop Technologies, Rockland, DE) and a bioanalyzer Distribution to skin and muscle did not differ between the (model 2100; Agilent, Palo Alto, CA). HG-U133 Plus 2 two routes of administration. Both routes of administration GeneChips (Affymetrix, Santa Clara, CA) were used for resulted in comparable amounts of siGFP in the tumors mRNA expression profiling. Hybridization and scanning (Fig. 1). In vivo of the chips was done by the MicroArray Department Efficacy of Unmodified siRNA (Amsterdam, the Netherlands). Analysis was done with We have shown previously that antisense oligonucleo- Rosetta Resolver version 5.1.0.1.23. Statistical analysis tides against the POLR2A inhibited tumor growth in vivo (ANOVA with Benjamini-Hochberg false discovery rate (17). Therefore, we have tested if siPOLR2A could inhibit tumor growth. Administration of siPOLR2A via tail vein injection resulted in significant inhibition of tumor growth 1 http://rsb.info.nih.gov/ij/ compared with saline treatment. The absence of growth

MolCancerTher2007;6(3).March2007

inhibition in siPOLR2A mismatch indicated that the growth. siGFP was administered either via osmotic mini- observed effect is a true antisense effect (Fig. 2A). pump or systemically via tail vein injection, and eGFP To further test the effect of different routes of adminis- knockdown in eGFP-positive tumors was evaluated. Both tration and the influence of LNA modifications on the routes of administration were effective in knockdown of in vivo effect, we made use of an eGFP reporter system. In GFP expression in the tumors (Fig. 2B). contrast to POLR2A, the use of GFP as target allows simple LNA Modification Increases Serum Stability readout for gene-specific knockdown, and knockdown of Unmodified siRNA, like our siGFP, is not very stable in GFP expression disconnects RNAi efficacy and tumor serum. Therefore, we designed LNA-modified siGFP to test growth kinetics, allowing simultaneous evaluation of RNAi if these modifications could improve the properties of (GFP knockdown) and sequence unrelated effect on tumor siGFP for in vivo application. Serum stability of unmodified siGFP, end-modified siGFP, and heavily modified siGFP was assessed in vitro by incubation of these molecules in fresh mouse serum. Unmodified siGFP was degraded within 5 h of incubation. In contrast, end-modified siGFP was stable in mouse serum for at least 48 h, where after 96 h, the majority of end-modified siGFP was degraded. The heavily modified siGFP was even more stable and did not showsigns of degradation after 96 h of incubation (Fig. 3). This clearly showed that incorporation of LNA molecules in siGFP contributes to increased half-life in serum. Effect of LNA Moieties on GFP Knockdown In vitro The efficacy and potency of unmodified siGFP and modified siGFPs were tested on cultured MiaPaca-II cells stably expressing GFP. siGFP potently reduced the mRNA levels in a concentration-dependent manner. Knockdown was f90% at unmodified siGFP concentrations of 5 and 25 nmol/L. Levels of knockdown of GFP expression after transfection with end-modified siGFP was 61% and 84% at similar concentrations. Transfection of heavily modified siGFP did not result in knockdown of GFP expression, not even at the highest concentration (Fig. 4A). Decreased mRNA levels were reflected at the protein levels. Unmod- ified siGFP lowered GFP protein levels with 62% and 78% at 5 and 25 nmol/L, respectively (Fig. 4B). End-modified siGFP lowered protein levels with 38% and 64% at the same concentration. Heavily modified siGFP did not affect GFP protein levels at all (Fig. 4B). These results showthat introduction of too many LNA molecules resulted in loss of its ability to knock down GFP expression. However, end modification of siRNA with LNA was compatible with the silencing machinery. Effect of LNA Modifications on GFP Knockdown In vivo To investigate whether eGFP expression could be lowered by modified siGFP in vivo, GFP-positive pancreatic cancer xenografts were induced and mice received LNA- modified siGFP. Target knockdown was compared with that of unmodified siGFP. When unmodified siGFP and modified siGFPs were administered via osmotic minipump, whole-body imaging showed no clear difference between the saline and the heavily modified siGFP-treated group. However, decreased fluorescence in the unmodified siGFP and the end-modified siGFP-treated group was Figure 6. Unmodified siGFP and end-modified siGFP were administered observed (Fig. 5A). GFP protein levels were lowered for via the tail vein, dosed at 0.15 mg/kg twice weekly for 3 wks. After 3 wks f f of treatment, eGFP protein levels were quantified by determination of 70% in the unmodified siGFP-treated group and 55% eGFP fluorescence in the tumors (A) and by Western blot (B). During in the end-modified siGFP-treated group (Fig. 5B). No treatment, tumor volumes were recorded (C). effect was observed in the heavily modified siGFP-treated

Mol Cancer Ther 2007;6(3). March 2007

Table 1. Differentially regulated genes in tumors of mice treated with siGFP

Sequence name (s) Sequence code Accession Sequence description Fold change

1557286_at 1557286_at AK001007 CDMA FLJ10145 fis, clone HEMBA1003322 3,8 1557779_at 1557779_at BC015429 Homo sapiens, clone IMAGE:4400004, mRMA À6,8 1558945_s_at 1558945_s_at BC042451 H. sapiens, clone IMAGE:5164020, mRMA À3,2 1570289_at 1570289_at BC017935 H. sapiens, clone IMAGE:3960940 À13,1 213484_at 213484_at AI097640 Clone 23700 mRNA sequence 4,0 226773_at 226773_at AW290940 MRNA (clone iCRFp507I1077) À3,8 228425_at 228425_at BF056746 7k20h08.x1 NCI_CGAP_Ov18 H. sapiens À3,5 cDNA clone IMAGE:3476199 239294_at 239294_at AA810265 Transcribed locus À5,5 ADAMTSL1,ADAMTSR1,MGC40193 237217_at BF111214 ADAMTS-like 1 À5,6 ADD2,ADDB 205268_s_at NM_017488 Adducin 2 (h)3,7 ADD3,ADDL 201753_s_at NM_019903 Adducin 3 (g)5,6 AKR1C1,C9,DD1,DDH,DDH1,H-37, 204151_x_at NM_001353 Aldo-keto reductase family 1, member C1 À4,4 MBAB,HAKRC ALDH8A1,ALDH12,DJ352A20.2 220148_at NM_022568 Aldehyde dehydrogenase 8 family, member A1 À3,1 AMOTL1,JEAP 225450_at AI433831 Angiomoctin like 1 À3,8 APOBEC3G,ARP9,CEM15,MDS019, 204205_at NM_021822 Apolipoprotein B mRNA editing enzyme, À7,5 FLJ12740 catalytic polypeptide-like 3G AQP1,CO,CHIP28,AQP-CHIP, 209047_at AL518391 Aquaporin 1 (channel-forming integral 6,3 MGC26324 protein, 28 kDa) ARG2 203946_s_at NM_001172 Vesicle transport through interaction with À4,4 t-SNAREs homologue 1B (yeast) ARHGAP15,BM046 218870_at NM_018460 Rho GTPase activating protein 15 À26,9 B3GNT5,B3GN-T5,h3Gn-T5 225612_s_at BE672260 UDP-GlcNAc:hGal h-1,3-N- À3,7 acetylglucosaminyltransferase 5 BCAT1,BCT1,ECA39,MECA39, 225285_at AK025615 Branched chain aminotransferase 1, cytosolic 3,8 DKFZp686E12175 C1orf21,PIG13 223126_s_at AI159874 Chromosome 1 open reading frame 21 4,1 C1orf24,NIBAN 217967_s_at AF288391 Chromosome 1 open reading frame 24 3,0 C20orf100,MGC15880,dJ495O3.1, 228737_at AA211909 Chromosome 20 open reading frame 100 4,4 dJ1108D11.2 C6orf15,STG 221100_at NM_014070 Chromosome 6 open reading frame 15 20,9 C9orf94,FLJ35283 239909_at AI937348 Chromosome 9 open reading frame 94 À6,0 CASP1,ICE,P45,IL1BC 211368_s_at U13700 Caspase-1, apoptosis-related cysteine À5,0 peptidase (interleukin 1, h, con CD24,CD24A 266_s_at L33930 CD24 antigen (small cell lung carcinoma 3,1 cluster 4 antigen) CDH18,CDH14,CDH24,CDH14L, 206280_at NM_004934 Cadherin 18, type 2 À5,3 EY-CADHERIN CLDN11,OSP,OTM 228335_at AW264204 Claudin 11 (oligodendrocyte transmembrane 38,5 protein) CSPG2,VERSICAN,DKFZp686K06110 221731_x_at BF218922 Choncroitin sulfate proteoglycan 2 (versican) 5,0 CTBP2 228949_at AL534095 COOH-terminal binding protein 2 5,0 DSC1,CDHF1.DG2/DG3 207324_s_at NM_004948 Desmocollin 1 À7,7 DST,BPA,BP240,BPAG1,MACF2, 204455_at NM_001723 Dystonin À5,9 CATX-15,KIAA0465 ELF3,ERT,ESX,EPR-1,ESE-1 210627_s_at U73844 E74-like factor 3 (ets domain transcription factor, 3,2 epithelial-specific) ELOVL6,FAE,LCE,MGC5487,FLJ23378 210868_s_at BC0013O5 ELOVL family member 6, elongation of long À3,2 chain fatty acids ELOVL7,FLJ23563 227180_at AWI38767 ELOVL family member 7, elongation of long 4,0 chain fatty acids (yeast) FAS,APT1,CD95,FAS1,APO-1,FASTM, 204780_s_at AA164751 Fas (TNF receptor superfamily, member 6) À5,2 ALPS1A,TNFRSF6 FLJ21075 221172_at NM_025031 Hypothetical protein FLJ21075 À5,8 FRMD4B,GRSP1,KIAA1013, 213056_at AU145019 FERM domain containing 4B À5,9 6030440G05Rik

(Continued on the following page)

MolCancerTher2007;6(3).March2007

Table 1. Differentially regulated genes in tumors of mice treated with siGFP (Cont’d)

Sequence name (s) Sequence code Accession Sequence description Fold change

FZD8,FZ-8,hFZ8 227405_s_at AW340311 Frizzled homologue 8 (Drosophila)4,4 IGAS1 204456_s_at AW611727 Growth arrest-specific 1 3,7 GAS2,MGC32610 205848_at NM_005256 Growth arrest-specific 2 À4,8 GJB2,HID,KID,PPK,CX26,DFNA3, 223278_at M86849 H. sapiens connexin 26 (GJB2) mRNA, complete cds À19,3 DFNB1,NSRD1 GLIPR1,GLIPR,RTVP1,CRISP7 204222_s_at NM_006851 GLI pathogenesis-related 1 (glioma) À3,7 GPM6B,M6B,MGC17150,MOC54284 209167_at AI419030 Glycoprotein M6B À3,3 GPR87,GPR95,FKSG78,KPG_002 219936_s_at NM_023915 G protein–coupled receptor 87 4,1 GRAMD3,NS3TP2,FLJ21313 218706_s_at NM_023927 HCV NS3-transactivated protein 2 À3,2 GRIK2,EAA4,GLR6,GLUR6,MGC74427 213845_at AL355532 Human DMA sequence from clone RP11-487F5 À3,1 on chromosome 6 GULP1,CED6,GULP,CED-6 204235_s_at AF200715 GULP, engulfment adaptor PTB domain containing 1 À4,0 HES1,HHL,HRY,HES-1,FLJ20408 203394_s_at BE973687 Hairy and enhancer of split 1, (Drosophila)3,9 HK2,HKII,HXK2,DKFZp686M1669 202934_at AI761561 Hexokinase 2 À4,4 KCNQ2,EBN,BFNC,EBN1,ENB1, 205737_at NM_004518 Potassium voltage-gated channel, KQT-like 4,3 HNSPC,KV7.2,KCNA11 subfamily, member 2 KIAA1211 227230_s_at BE855799 KIAA1211 protein 4,2 KIAA1958,FLJ39294 235112_at AA088388 KIAA1958 4,2 LOC154092 237731_at AW665570 Hypothetical protein LOC154092 À3,6 LOC158257 227793_at AA969238 Hypothetical protein LOC158257 3,6 LOC284072 227452_at AI832118 Hypothetical protein LOC284072 3,3 LOC286334 228590_at AA910497 Hypothetical protein LOC286334 5,8 LOC401398 244705_at AW470690 Hypothetical protein LOC401398 À4,5 MAP2,MAP2A,MAP2B,MAP2C 225540_at BF342661 Microtubule-associated protein 2 À3,4 MDFIC,HIC 211675_s_at AF054589 MyoD family inhibitor domain containing À5,4 MEIS4,MRG2,MEIS3,Meis (mouse) 214077_x_at H15129 Meis1, myeloid ecotropic viral integration À3,7 homologue 3,Meis1 site 1 homologue 4 (mouse) MGC16037 229C02_at AI095583 Hypothetical protein MGC16037 5,1 MPDZ,MUPP1,FLJ25909, 213306_at AA917899 Multiple PDZ domain protein À3,3 DKFZp781P216 MPPE1 213924_at BF476502 Metallophosphoesterase 1 15,6 MRPS6 213167_s_at BF982927 602306318F1 NIH_MGC_88 H. sapiens cDNA À3,4 clone IMAGE: 4397525 MYBPH 206304_at NM_004997 Myosin binding protein H 5,3 NAV3,POMFIL1,unc53H3, 204323_at NM_014903 Ribosomal protein S16 family protein À4,7 KIAA0938,STEERIN3 NRP1,NRP,VEGF165R, 212298_at BE620457 Neuropilin 1 4,5 DKFIp781F1414 NT5E,eN,NT5,NTE,eNT, 1553995_a_at BC015940 5¶-Nucleotidase, ecto (CD73) À3,1 CD73,E5NT OLFML2A,FLJ00237,PRO34319 213075_at AL050002 Olfactomedin-like 2A 3,5 PMAIP1,APR,NOXA 204286_s_at NM_021127 Phorbol 12-myristate 13-acetate–induced À3,2 protein 1 PRSS23,SIG13,SPUVE,ZSIG13, 202458_at NM_007173 Protease, serine, 23 6,4 MGC5107 ROBO1,SAX3,DUTT1,FLJ21882 213194_at BF059159 Roundabout, axon guidance receptor, À5,3 homologue 1 (Drosophila) RTN1,NSP 210222_s_at BC000314 Reticulon 1 36,1 SCNN1A,ENaCa,SCNEA,SCNN1, 203453_at NM_001038 Sodium channel, nonvoltage-gated 1a À5,6 ENaCa SEPP1,SeP,SELP 201427_s_at NM_005410 Selenoprotein P. plasma, 1 17,2 SFN 33323_r_at X57348 H. sapiens mYRNA (clone 1912) À9,0 SLC2A3,GLUT3 202497_x_at AI631159 Solute carrier family 2 (facilitated glucose À4,6 transporter), member 3 SLC5A3 1553313_s_a NM_006933.1 gb:NM_006933.1 /DB_XREF=gi: 5902091/ À3,8 GEN=SLC5A3 /TID=Hs2.421196

(Continued on the following page)

Mol Cancer Ther 2007;6(3). March 2007

Table 1. Differentially regulated genes in tumors of mice treated with siGFP (Cont’d)

Sequence name (s) Sequence code Accession Sequence description Fold change

SLC6A12,BGT1,BGT-1 206058_at U27699 Solute carrier family 6 (neurotransmitter 4,1 transporter, betaine/GABA) SLCO1B3,OATP8,OATP1B3, 206354_at NM_019844 Solute carrier organic anion transporter 4,3 SLC21A8,LST-3TM13 family, member 1B3 SLITRK6,MGC119595,MGC119596, 235976_at AI680986 SLIT and NTRK-like family, member 6 À4,1 MGC119597 STC1,STC 230746_s_at AW003173 Stanniocalcin 1 6,5 SYTL2,SLP2,SGA72M,CHR11SYT, 225496_s_at N21426 Synaptotagmin-like 2 3,3 KIAA1597,exophilin-4 TANC,KIAA1728 225308_s_at AB051515 TPR domain, ankyrin-repeat and coiled-coil- 4,6 containing TGF-hllR a,unnamed 208944_at D50683 H. sapiens mRNA for TGF-hllRa, complete cds 7,1 TLE2,ESG,ESG2,GRG2 204431 _at NM_003260 Transducin-like enhancer of split 2 [E(sp1) 7,6 homologue, Drosophila] TMEM45A,DERP7,FLJ10134 219410_at NM_018004 Transmembrane protein 45A À5,7 TMEPAI,STAG1,PMEPA1 222450_at AL035541 Human DNA sequence from clone RP4-718J7 3,5 TSPAN18,TSPAN 227307_at AL565381 Tetraspanin 18 13,3 TXNRD2,TR,TR3,SELZ,TRXR2, 211177_s_at AB019695 Thioredoxin reductase 2 3,4 TR-BETA ZNF488,FLJ32104 229901_at AI056483 Zinc finger protein 488 25,3

NOTE: Fold changes are calculated as ratio average saline treatment / average siGFP. Abbreviations: TNF, tumor necrosis factor; GABA, g-aminobutyric acid.

group (Fig. 5B). In sections of unmodified siGFP and end- scription (STAT) 1 protein involved in dsRNA-dependent modified siGFP but not heavily modified siGFP-treated protein kinase response and IFN response showed no tumors, eGFP fluorescence was lowered compared with the significant difference between saline-treated, unmodified saline control (Fig. 5C). siGFP or end-modified siGFP-treated animals (data not Unmodified siGFP and the end-modified siGFP admin- shown). istered via tail vein injection significantly lowered the Effect of LNA Modifications on ‘‘Off-Target’’ Gene GFP fluorescence in the tumors (Fig. 6A). In the unmo- Regulation dified siGFP and the end-modified siGFP-treated groups, Introduction of LNA molecules in siRNA increases the f GFP protein levels were reduced by 50% in both Tm of siRNA toward its target, which increases the groups (Fig. 6B), showing that unmodified siGFP and potential of ‘‘off-target‘‘ effects. To test this assumption, end-modified siGFP are effective in both administration we did expression profiling of tumors of mice shown in routes. Fig. 4A. Administration of end-modified siGFP resulted in During treatment, there was no significant effect of either only 7 differentially regulated genes, whereas administra- siGFP on tumor growth (Fig. 6C). Furthermore, protein tion of siGFP resulted in 93 differentially regulated genes levels of phosphorylated eukaryotic initiation factor-2a and (Tables 1 and 2). There was no overlay between the phosphorylated signal transducer and activator of tran- differentially regulated genes in the two experiments. This

Table 2. Differentially regulated genes in tumors of mice treated with end-modified siGFP

Sequence name(s) Sequence code Accession Sequence description Fold change

ATN1,B37,NOD,DRPLA,D12S755E 40489_at D31840 Atrophin 1 9,3 HBE1 217683_at AA115963 Hemoglobin, e1 À3,2 LENG9 224673_at AI613244 Leukocyte receptor cluster member 9 5,1 NPAS2,MOP4,PASD4,FLJ23138,MGC71151 1557690_x_at BU683708 Neuronal PAS domain protein 2 4,1 SERPINB6,CAP,PI6,PTI.MSTP057,MGC111370, 231628_s_at AW262311 Serpin peptidase inhibitor, clade B À18,2 DKFZp686I04222 (ovalbumin), member 6 TBC1D3,PRC17,TRE17,TBC1D3A 209403_at AL136860 TBC1 domain family, member 3 3,1 TncRNA,trophoblast STAT utron,trophoblast MHC 227062_at AU155361 Trophoblast-derived noncoding RNA 3,2 class II suppressor

NOTE: Fold changes are calculated as ratio average saline treatment/average end-modified siGFP.

MolCancerTher2007;6(3).March2007

Table 3. Ratios of expression levels of genes involved in signaling on dsRNA recognition and a selection IFN-related genes

Sequence name(s) Sequence code Accession Sequence description Fold change Fold change saline vs saline vs siGFP end-modified siGFP

IFNA1,IFL,IFN,IFNA@,IFNA13 208375_at NM_024013 IFN, a1 À1,2 1,1 IFNA10,MGC119878,MGC119879 208261_x_at NM_002171 IFN, a10 1,0 À1,1 IFNA13 208344_x_at NM_006900 IFN, a13 À1,6 À2,1 IFNA14,MGC125756,MGC125757 208182_x_at NM_002172 IFN, a14 À1,1 À1,5 IFNA16 208448_x_at NM_002173 IFN, a16 À1,1 À1,5 IFNA17,IFNA 211405_x_at M38289 IFN, a17 À1,1 À1,2 IFNA2,IFNA,INFA2,cytokine 211338_at M54886 IFN, a21,2À1,1 IFNA21 211145_x_at M12350 IFN, a21 1,2 1,2 IFNA4,INFA4 207964_x_at NM_021068 IFN, a4 1,0 1,1 IFNA5,INFA5 214569_at NM_002169 IFN, a5 1,0 1,3 IFNA6 208548_at NM_021002 IFN, a61,4À1,1 IFNA7,IFNA-J 208259_x_at NM_021057 IFN, a7 À1,2 À1,1 IFNA8 2D7932_at NM_002170 IFN, a8 À1,3 1,0 IFNB1,IFB,IFF,IFNB,MGC96956 208173_at NM_002176 IFN, h1, fibroblast 1,0 1,7 IRF1,MAR,IRF-1 202531_at NM_002198 IFN regulatory factor 1 1,1 1,4 IRF3 202621_at NM_001571 IFN regulatory factor 3 1,0 1,3 OAS1,OIAS,IFI-4,OIASI 205552_s_at NM_002534 2¶,5¶-Oligoadenylate synthetase 1 À1,1 1,1 OAS2,MGC78578 204972_at NM_016817 2¶-5¶-Oligoadenylate synthetase 2 1,4 À1,2 PRKRA,RAX,PACT,HSD14 228714_at AI681888 Protein kinase, IFN-inducible 1,1 1,1 dsRNA-dependent activator RNASEL,HPC1,RNS4,PRCA1 221287_at NM_021133 RNase L À1,1 1,2 STAT1,ISGF-3,STAT91 209969_s_at BC002704 Signal transducer and activator 1,1 1,2 of transcription 1 STAT2,P113,ISGF-3,STAT113 225636_at H98105 Signal transducer and activator 1,0 À1,1 of transcription 2 TLR3 206271_at NM_003265 Toll-like receptor 3 À1,2 À1,3 TLR7 222952_s_at AF245702 Toll-like receptor 7 À1,2 À1,5 TLR8 220832_at NM_016610 Toll-like receptor 8 1,2 1,4 TLR9 223903_at AB045180 Toll-like receptor 9 À1,2 À1,6

NOTE: Fold changes are calculated as ratio average saline treatment / average siGFP and average saline treatment/end-modified GFP.

suggests that none of these genes are regulated due to an tumors was significantly lowered after systemic delivery of effect on GFP expression. Therefore, we consider these siRNA directed against eGFP. These results agree with genes off-target effects. Expression of genes involved in the three other independent studies. First, it was shown that IFN response and dsRNA-induced signaling did not differ systemically administered siRNA induced silencing of between the different groups (Table 3). CXCR4 and reduced breast cancer metastasis (21). Second, silencing of unmodified siRNA against CEACAM6 reduced pancreatic tumor growth (18). Furthermore, daily i.p. Discussion injections of siRNA directed against Bcl-2 resulted in We have shown that unmodified and uncomplexed siRNA growth inhibition of pancreatic cancer xenografts (22). In can be delivered to multiple tissues, including tumor our study, knockdown was also achieved irrespective of xenografts, in vivo by at least two different ways of the route of administration because continuous delivery of administration. Systemic delivery via tail vein injections siGFP via osmotic minipumps resulted in comparable of siRNA designed against POLR2A, a target previously levels of knockdown compared with delivery via tail vein shown suitable for tumor growth inhibition using antisense twice weekly. Using the tail vein method, the amount of oligonucleotides (14), resulted in tumor growth inhibition. siRNA needed to do in vivo experiments is f6-fold lower To further optimize in vivo application of siRNA, we compared with pump delivery and therefore can be done choose to target the exogenously expressed eGFP gene cost effective. because it allows simple readout for gene-specific knock- Our studies on LNA-modified siGFP have shown that down. Knockdown of GFP expression should not affect LNA offers the means to improve the serum half-life of tumor growth kinetics allowing simultaneous evaluation of siRNAs. These results agree with the findings of Elmen RNAi (GFP knockdown) and sequence unrelated effect on et al. (15) who also show that only end modifications are tumor growth. In line with tumor growth inhibition due to needed to obtain this increase in half-life and is in line with siPOLR2A, eGFP expression in eGFP-positive MiaPaca-II the predominant degradation of DNA oligonucleotides due

Mol Cancer Ther 2007;6(3). March 2007

to 3¶ exonuclease activity (23). Thus, introduction of a few We have no evidence for a dsRNA-dependent protein LNA moieties into siRNA provides an excellent means to kinase and IFN response in our study. Phosphorylated protect siRNA against degradation. STAT1 and phosphorylated eukaryotic initiation factor-2a In vitro studies showed that introduction of LNA were not elevated in the tumors after siRNA treatment. molecules in end-modified siGFP lowered its efficacy in Microarray data also did not give an indication of an IFN target knockdown. Introduction of additional LNA mod- response. This agrees with findings of Heidel et al. (30) who ifications in the sense strand as in the heavily modified showed no induction of IFN on systemic administration of siGFP resulted in loss of GFP knockdown. This may be uncomplexed siRNA. explained by the fact that RISC incorporates dsRNA and In conclusion, unmodified naked siRNA-mediated target becomes active after sense strand degradation in the RISC knockdown in vivo is possible at very lowconcentrations. complex (24, 25). Heavily modified siGFP may prevent sense Modification of siRNA with LNA molecules greatly strand degradation and therefore RISC activation could not enhanced its half-life in serum. However, it is clear that occur. However, not all sense strand modifications inhibit LNA modifications should be kept at a minimum to allow RISC. Effective siRNAs with modified sense strands compatibility with the RNAi machinery. The levels of composed of a combination DNA and 2¶-fluoro–modified target knockdown achieved in vivo by very low-dosage nucleotides have been described (13). Alternatively, the unmodified siGFP and end-modified siGFP suggest that duplex dissociates without sense strand degradation. As lowered efficacy was compensated for by increased opposed to LNA, DNA and 2¶-fluoro modifications do not stability. LNA-modified siRNA should be preferred over affect Tm values dramatically. Therefore, the observed unmodified molecules because introduction of LNA in the difference between LNA-modified and DNA/2¶-fluoro– end-modified siGFP strongly reduced off-target gene modified siRNA could also be explained by the differences regulation. in Tm. We conclude that only a fewLNA modifications, like our end-modified siGFP, result in a dramatic increased half- References life in serum, whereas the efficacy was only marginally 1. Meister G, Tuschl T. Mechanisms of gene silencing by double-stranded RNA. Nature 2004;431:343 – 9. reduced. These characteristics potentially improve this 2. Yano J, Hirabayashi K, Nakagawa S, et al. Antitumor activity of small molecule for therapeutic application. interfering RNA/cationic liposome complex in mouse models of cancer. In agreement with target knockdown of unmodified Clin Cancer Res 2004;10:7721 – 6. siGFP after different routes of administration, end-modi- 3. Urban-Klein B, Werth S, Abuharbeid S, Czubayko F, Aigner A. RNAi- mediated gene-targeting through systemic application of polyethylenimine fied siGFP was effective on tail vein delivery or via osmotic (PEI)-complexed siRNA in vivo. Gene Ther 2005;12:461 – 6. minipumps. Furthermore, both routes of administration 4. Schiffelers RM, Ansari A, Xu J, et al. Cancer siRNA therapy by tumor resulted in comparable target knockdown due to unmod- selective delivery with ligand-targeted sterically stabilized nanoparticle. ified siGFP and end-modified siGFP. This suggests that Nucleic Acids Res 2004;32:e149. lowered efficacy as observed with in vitro assays is 5. Soutschek J, Akinc A, Bramlage B, et al. Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs. Nature compensated for by increased in vivo stability. 2004;432:173 – 8. Off-target effects on gene regulation have to be 6. Ma Z, Li J, He F, Wilson A, Pitt B, Li S. Cationic lipids enhance siRNA- considered when using siRNA for functional genomics mediated interferon response in mice. Biochem Biophys Res Commun and therapeutic applications. Recently, it has been shown 2005;330:755 – 9. 7. Hornung V, Guenthner-Biller M, Bourquin C, et al. Sequence-specific that complementarity of the seed region (bases 2–7) of a potent induction of IFN-a by short interfering RNA in plasmacytoid siRNA is sufficient for down-regulation of untargeted dendritic cells through TLR7. Nat Med 2005;11:263 – 70. genes (26–28). In line with these observations, it was 8. Braasch DA, Jensen S, Liu Y, et al. RNA interference in mammalian shown that 2-O-methyl modification of the second cells by chemically-modified RNA. Biochemistry 2003;42:7967 – 75. nucleotide in the antisense strand reduced silencing of 9. Harborth J, Elbashir SM, Vandenburgh K, et al. Sequence, chemical, and structural variation of small interfering RNAs and short hairpin RNAs most off-target transcripts with complementarity to the and the effect on mammalian gene silencing. Antisense Nucleic Acid Drug seed region of the siRNA, whereas target knockdown was Dev 2003;13:83 – 105. not affected (29). In our study, end-modified siGFP 10. Amarzguioui M, Holen T, Babaie E, Prydz H. Tolerance for mutations and showed a strong reduction in off-target gene regulation, chemical modifications in a siRNA. Nucleic Acids Res 2003;31:589– 95. 11. Czauderna F, Fechtner M, Dames S, et al. Structural variations and whereas on-target knockdown was not affected to that stabilising modifications of synthetic siRNAs in mammalian cells. Nucleic extent, indicating that this effect is not due to lowered Acids Res 2003;31:2705 – 16. potency of end-modified siGFP. This agrees with findings 12. Nykanen A, Haley B, Zamore PD. ATP requirements and small interfering of Elmen et al. (15) who showed that reduced off-target RNA structure in the RNA interference pathway. Cell 2001;107:309–21. effects on placement of a LNA molecule at the 5¶ end of 13. Morrissey DV, Lockridge JA, Shaw L, et al. Potent and persistent in vivo anti-HBV activity of chemically modified siRNAs. Nat Biotechnol the sense strand. Two mechanisms could be responsible 2005;23:1002 – 7. for this phenomenon. (a) Introduction of LNA into the 5¶ 14. Fluiter K, ten Asbroek AL, de Wissel MB, et al. In vivo tumor growth end of the sense strand results in preferential loading of inhibition and biodistribution studies of locked nucleic acid (LNA) antisense oligonucleotides. Nucleic Acids Res 2003;31:953 – 62. the antisense strand into RISC. (b) If, despite guided RISC 15. Elmen J, Thonberg H, Ljungberg K, et al. Locked nucleic acid (LNA) loading, sense strand is incorporated into RISC, this mediated improvements in siRNA stability and functionality. Nucleic Acids design likely prevents 5¶ phosphorylation and the forma- Res 2005;33:439– 47. tion of active RISC. 16. Marques JT, Devosse T, Wang D, et al. A structural basis for

MolCancerTher2007;6(3).March2007

discriminating between self and nonself double-stranded RNAs in kinetics of degradation of antisense oligonucleotides by a 3¶ exonuclease mammalian cells. Nat Biotechnol 2006;24:559– 65. in plasma. Antisense Res Dev 1991;1:141 – 51. 17. Fluiter K, ten Asbroek AL, van Groenigen M, Nooij M, Aalders MC, 24. Matranga C, Tomari Y, Shin C, Bartel DP, Zamore PD. Passenger- Baas F. Tumor genotype-specific growth inhibition in vivo by antisense strand cleavage facilitates assembly of siRNA into Ago2-containing RNAi oligonucleotides against a polymorphic site of the large subunit of human enzyme complexes. Cell 2005;123:607 – 20. RNA polymerase II. Cancer Res 2002;62:2024 – 8. 25. Rand TA, Petersen S, Du F, Wang X. Argonaute2 cleaves the anti- 18. Duxbury MS, Matros E, Ito H, Zinner MJ, Ashley SW, Whang guide strand of siRNA during RISC activation. Cell 2005;123:621 – 9. EE. Systemic siRNA-mediated gene silencing: a new approach to targeted therapy of cancer. Ann Surg 2004;240:667 – 74; discussion 26. Jackson AL, Burchard J, Schelter J, et al. Widespread siRNA ‘‘off- 75 – 6. target’’ transcript silencing mediated by seed region sequence complementarity. RNA 2006;12:1179– 87. 19. Bijsterbosch MK, Manoharan M, Rump ET, et al. In vivo fate of phosphorothioate antisense oligodeoxynucleotides: predominant uptake 27. Lin X, Ruan X, Anderson MG, et al. siRNA-mediated off-target gene by scavenger receptors on endothelial liver cells. Nucleic Acids Res 1997; silencing triggered by a 7 nt complementation. Nucleic Acids Res 2005; 25:3290 – 6. 33:4527 – 35. 20. Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci U S A 28. Birmingham A, Anderson EM, Reynolds A, et al. 3¶ UTR seed 1984;81:1991 – 5. matches, but not overall identity, are associated with RNAi off-targets. 21. Liang Z, Yoon Y, Votaw J, Goodman MM, Williams L, Shim H. Silencing Nat Methods 2006;3:199 – 204. of CXCR4 blocks breast cancer metastasis. Cancer Res 2005;65:967– 71. 29. Jackson AL, Burchard J, Leake D, et al. Position-specific chemical 22. Ocker M, Neureiter D, Lueders M, et al. Variants of bcl-2 specific modification of siRNAs reduces ‘‘off-target’’ transcript silencing. RNA siRNA for silencing antiapoptotic bcl-2 in pancreatic cancer. Gut 2005;54: 2006;12:1197 – 205. 1298 – 308. 30. Heidel JD, Hu S, Liu XF, Triche TJ, Davis ME. Lack of interferon 23. Eder PS, DeVine RJ, Dagle JM, Walder JA. Substrate specificity and response in animals to naked siRNAs. Nat Biotechnol 2004;22:1579– 82.

Mol Cancer Ther 2007;6(3). March 2007

Olaf R. Mook, Frank Baas, Marit B. de Wissel, et al.

Mol Cancer Ther 2007;6:833-843.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/6/3/833

Cited articles This article cites 30 articles, 7 of which you can access for free at: http://mct.aacrjournals.org/content/6/3/833.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/6/3/833.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/6/3/833. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.