| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Research Articles: Therapeutics, Targets, and Development
Evaluation of locked nucleic acid–modified small interfering RNA in vitro and in vivo
Department of Neurogenetics, Academic Medical Center, Amsterdam, the Netherlands
Requests for reprints: Olaf R. Mook, Department of Neurogenetics, Academic Medical Center, Meibergdreef 9, Amsterdam, the Netherlands 1105 AZ. Phone: 312-056-64540. E-mail: o.r.mook{at}amc.uva.nl
Abstract
RNA interference has become widely used as an experimental tool to study gene function. In addition, small interfering RNA (siRNA) may have great potential for the treatment of diseases. Recently, it was shown that siRNA can be used to mediate gene silencing in mouse models. Locally administered siRNAs entered the first clinical trials, but strategies for successful systemic delivery of siRNA are still under development. Challenges still exist about the stability, delivery, and therapeutic efficacy of siRNA. In the present study, we compare the efficacy of two methods of systemic siRNA delivery and the effects of siRNA modifications using locked nucleic acids (LNA) in a xenograft cancer model. Low volume tail vein bolus injections and continuous s.c. delivery using osmotic minipumps yielded similar uptake levels of unmodified siRNA by tumor xenografts. Both routes of administration mediated sequence-specific inhibition of two unrelated targets inside tumor xenografts. Previous studies have shown that LNA can be incorporated into the sense strand of siRNA while the efficacy is retained. Modification of siRNA targeting green fluorescent protein with LNA results in a significant increase in serum stability and thus may be beneficial for clinical applications. We show that minimal 3' end LNA modifications of siRNA are effective in stabilization of siRNA. Multiple LNA modifications in the accompanying strand further increase the stability but negate the efficacy in vitro and in vivo. In vivo, LNA-modified siRNA reduced off-target gene regulation compared with nonmodified siRNA. End-modified siRNA targeting green fluorescent protein provides a good trade-off between stability and efficacy in vivo using the two methods of systemic delivery in the nude mouse model. Therefore, LNA-modified siRNA should be preferred over unmodified siRNA. [Mol Cancer Ther 2007;6(3):833–43]
Introduction
RNA interference (RNAi) is a natural process that affects gene silencing in eukaryotic systems at transcriptional, posttranscriptional, and/or translational levels (1). Small interfering RNA (siRNA) molecules are the key intermediates in this process, which can potentially inhibit the expression of any given target gene. siRNA molecules hold great promise as biological tools and as potential therapeutic agents for targeted inhibition of disease-causing genes.
To optimize the use of siRNA as a therapeutic therapy, several modes of delivery have been tested in vivo. Improved delivery in animals has been achieved by complexation with cationic liposomes (2), polyethylenimine (3), and Arg-Gly-Asp–polyethylene glycol–polyethylenimine (4) or by conjugation of cholesterol to siRNA (5). However, Ma et al. (6) has shown that complexation to cationic liposomes resulted in a potent induction of both type I and type II IFN responses in mice. In line, sequence-specific and Toll-like receptor-7–dependent induction of IFNs has been shown on liposome-mediated siRNA transfection in mice (7). Furthermore, conjugation of cholesterol to a siRNA against apolipoprotein B could specifically degrade its mRNA in vivo (5).
Another approach to improve potency and efficacy of siRNA in vivo is by the introduction of other chemical modifications in siRNA. In vitro studies showed that several modifications are allowed in functional siRNAs. Modifications of phosphorothioate (8, 9) 2'-O-methyl (10, 11), 2'-O-allyl (10), and 2'-deoxy-fluorouridine (8, 9) have been examined for potential in vivo use. Some of the modified siRNAs were found to exhibit enhanced serum stability (11) and longer duration of action (10). Modification of the 5' end of the antisense strand with 2'-O-allyl (10) or chemical 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 every second or all internucleoside linkages (8, 9), increased cytotoxic effects and resulted in a significant decrease or complete loss of activity. In contrast, totally modified duplexes containing a combination of 2'-O-methyl and 2'-fluoro modifications were shown to be effective in vitro and in vivo (13).
Locked nucleic acid (LNA) is a novel nucleotide analogue that contains a methylene bridge that connects the 2'-oxygen of the ribose with the 4'-carbon. The bicyclic structure locks the furanose ring of the LNA molecule in a 3'-endo conformation, thereby structurally mimicking the standard RNA monomers. LNA induce an increase in thermal stability (melting temperature, Tm) when bound to a matching RNA sequence. Introduction of LNA into classic antisense oligonucleotides has been shown to increase its serum stability (14). In analogy, Braasch et al. (8) showed that LNA can be used to thermally stabilize siRNAs without losing their function. Recently, a systematic study on LNA containing siRNAs has identified the number and positions of LNA molecules within the siRNA, which still allow a functional siRNA (15). Incorporation of LNA molecules into siRNA significantly increased its serum stability, which potentially favors successful in vivo applications (15). Furthermore, LNA modifications at the 5' end of the sense strand of siRNA has been reported to favor incorporation of the antisense strand into the RNA-induced silencing complex (RISC), thereby reducing sequence related off-target effects (15). A second potential benefit of LNA-modified siRNA is the protection of the two nt 3' overhang of siRNA. Recently, it has been shown that in certain cell types blunt-ended double-stranded RNA (dsRNA; lacking 3' overhangs) <30 bp induced dsRNA-mediated signaling (16). Therefore, protection of those overhangs with LNA potentially increases the specificity of siRNA.
In this study, we show that systemic administration of unmodified siRNA induced specific RNAi effects of two independent targets in tumor xenografts. Different routes of administration resulted in comparable target knockdown. To test whether LNA-modified siRNA contributes to the efficacy in vivo, we have designed an end-modified siRNA targeting green fluorescent protein (siGFP) and a heavily modified siGFP and compared their characteristics with unmodified siGFP both in vitro and in vivo. In the present article, we show that minimal modification of siGFP with LNA greatly enhanced its serum stability and was compatible with the silencing machinery. End-modified 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 show that 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, CO). siRNA against enhanced GFP (eGFP; siGFP) SS 5'-gcugacccugaaguucauctt-3' and AS 5'-gaugaacuucagggucagctt-3' were purchased from MedProbe (Lund, Sweden). End-modified siGFP SS 5'-GcugacccugaaguucaucTT-3' and AS 5'-gaugaacuucagggucagcTT-3' and heavily modified siGFP SS 5'-GCTgacCcuGaagTTcaucTT-3' and AS 5'-gaugaacuucagggucagcTT-3' (LNA nucleotides are depicted in capitals) were synthesized by Santaris A/S (Hørsholm, Denmark) as described previously (15).
Cell Line
The pancreatic cancer cell line MiaPaca-II and MiaPaca-II stably expressing eGFP (17) were maintained at 37°C and 5% CO2 by serial passage in DMEM supplemented with 10% FCS, 2 mmol/L L-glutamine, 100 units/mL penicillin, and 100 µg/mL streptomycin.
Application of Unmodified and LNA-Modified siRNA In vivo
S.c. tumors were induced in 8- to 10-week-old NMRI nu/nu mice (Charles River, Maastricht, the Netherlands) as described previously (14). One week after tumor cell injection, when tumor take was positive, administration of the siRNA or LNA-modified siRNA started.
For POLR2A inhibition, siPOLR2A and the mismatch control containing two central mismatches were administered via tail vein injections at a dosage of 0.15 mg/kg (200 µL of 1.4 µmol/L siPOLR2A solution) twice weekly for 3 weeks as described previously (18). During treatment, tumor growth was monitored as described previously (17).
In all experiments targeting GFP, administration was either via tail vein injection as described above or via osmotic minipumps (model 1007D; Alzet Corp., Palo Alto, CA) dosed at 0.25 mg/kg/d for 7 days. For each treatment, five mice per group were used. All animal experiments were conducted under the institutional guidelines and according to the law; they were sanctioned by the animal ethics committee.
Whole-Body Imaging and Tissue Processing
eGFP fluorescence was visualized with whole-body imaging using the GFP fluorescence mode of a LAS3000 (Fuji, Tokyo, Japan). Parts of the tumors were fixed in formaldehyde (4%)/sucrose (20%). Other parts of the tumor were used to prepare protein lysates. The rest of the tumor tissue was snap frozen in liquid nitrogen and stored at –80°C until further use.
Biodistribution of Unmodified siRNA after Different Routes of Administration
Tissue distribution studies of tritiated siGFP were done according to Bijsterbosch et al. (19). Distribution was studied after 30 min of circulation of a bolus injection (200 µL) of tritiated siGFP (1.4 µmol/L; 0.15 mg/kg) and after continuous s.c. administration of 0.5 µL/h of tritiated siGFP (40 µmol/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 µmol/L) were incubated at 37°C in fresh mouse serum. Aliquots of 4 µL 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 –20°C 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
Transfections were done in six-well culture plates with LipofectAMINE 2000 (Invitrogen, Carlsbad, CA) as liposomal transfection agent. Protein samples were prepared in 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 dispersion (400 µL/100 µg) 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 proteins were transferred electrophoretically to polyvinylidene difluoride membranes (Invitrogen). eGFP was detected with a rabbit anti-GFP polyclonal antibody (Molecular Probes, Eugene, OR). Elongation factor 2
(Cell Signaling, Beverly, MA) was used as loading control. Chemiluminescent detection was done on a LAS3000 in accordance with the manufacturer's instructions. eGFP signals were quantified using Aida software version 3.44 and normalized to those of elongation factor 2.
Northern Blots
The RNA isolation was according to the manufacturer's procedure. RNA was denatured using glyoxal and separated on 1% agarose gels following standard protocols. RNA was subsequently transferred to Hybond-N+ membrane (Amersham, Piscataway, NJ) in 20x SSC. Following transfer, the RNA was UV cross-linked and the membrane was baked for 4 h at 80°C. For Northern blot analyses, the BcgI/HindIII fragment of pEGFP-C1 was used as a probe. A 28S probe was used as loading control. Hybridizations and posthybridization washes were according to Church and Gilbert (20).
GFP Fluorescence In situ
Tumor tissue was fixed in 4% formaldehyde in PBS containing 20% (w/v) sucrose for 24 h at 4°C and sectioned (10 µm thick) at a cabinet temperature of –34°C and stored at –20°C until use. Before use, sections were washed twice in PBS and embedded in Vectashield. GFP fluorescence was recorded using standardized setting of the confocal laser scanning microscope (TCS SP2) fitted to a DM-IRB inverted microscope (Leica, Mannheim, Germany). Excitation of eGFP was done at 488 nm, and fluorescence was captured at 500 to 530 nm. GFP fluorescence was quantified with ImageJ 1.32 software (W.S. Rasband, ImageJ,1 NIH, Bethesda, MD; 1997–2005).
Expression Profiling
Total RNA from tumors of mice that received saline, siGFP, or end-modified siGFP (three tumors per group) via osmotic pumps dosed at 0.25 mg/kg/d was isolated in Trizol according to the manufacturer's instructions. RNA was further purified with Macherey-Nagel RNA spin columns (Hoerdt, France) and DNase-treated RNA was eluted with RNase-free H2O. The quantity and quality of the RNA was assessed with a spectrophotometer (ND1000; NanoDrop Technologies, Rockland, DE) and a bioanalyzer (model 2100; Agilent, Palo Alto, CA). HG-U133 Plus 2 GeneChips (Affymetrix, Santa Clara, CA) were used for mRNA expression profiling. Hybridization and scanning of the chips was done by the MicroArray Department (Amsterdam, the Netherlands). Analysis was done with Rosetta Resolver version 5.1.0.1.23. Statistical analysis (ANOVA with Benjamini-Hochberg false discovery rate correction for multiple testing) was used to detect significant differences. Genes with an ANOVA P value <0.05 and a fold change greater than 3 and less than –3 were considered significantly regulated.
Results
Comparison of Biodistributions after Different Routes of Administration
We have done distribution studies of siGFP to test if different routes of administration at different dosages could deliver siGFP into various organs and tumors. We have compared these two routes because administration of antisense oligonucleotides via osmotic pumps is a validated route of administration in our model. We compared that with low volume i.v. bolus injection of siRNA because it has been shown recently by Duxbury et al. (18) to be an effective way to mediate RNAi in vivo at very low dosages. Administration of radiolabeled siGFP dosed at 0.25 mg/kg/d via osmotic minipumps delivered siGFP into a large number of tissues (Fig. 1
). Administration of 0.15 mg/kg siRNA injected i.v. as a single bolus resulted in high uptake in the kidney. Liver and spleen were organs with relatively high uptake after systemic administration (Fig. 1). Comparing administration via osmotic minipump and bolus injection, it is clear that bolus injection results in higher delivery in a large number of organs. Distribution to skin and muscle did not differ between the two routes of administration. Both routes of administration resulted in comparable amounts of siGFP in the tumors (Fig. 1).
|
|
LNA Modification Increases Serum Stability
Unmodified siRNA, like our siGFP, is not very stable in serum. Therefore, we designed LNA-modified siGFP to test if these modifications could improve the properties of 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 show signs 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.
|
90% 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. Unmodified 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 show that 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.
|
70% in the unmodified siGFP-treated group and 55% in the end-modified siGFP-treated group (Fig. 5B). No effect was observed in the heavily modified siGFP-treated group (Fig. 5B). In sections of unmodified siGFP and end-modified siGFP but not heavily modified siGFP-treated tumors, eGFP fluorescence was lowered compared with the saline control (Fig. 5C).
|
50% in both groups (Fig. 6B), showing that unmodified siGFP and end-modified siGFP are effective in both administration routes.
|
and phosphorylated signal transducer and activator of transcription (STAT) 1 protein involved in dsRNA-dependent protein kinase response and IFN response showed no significant difference between saline-treated, unmodified siGFP or end-modified siGFP-treated animals (data not shown).
Effect of LNA Modifications on "Off-Target" Gene Regulation
Introduction of LNA molecules in siRNA increases the Tm of siRNA toward its target, which increases the potential of "off-target" effects. To test this assumption, we did expression profiling of tumors of mice shown in Fig. 4A. Administration of end-modified siGFP resulted in only 7 differentially regulated genes, whereas administration of siGFP resulted in 93 differentially regulated genes (Tables 1
and 2
). There was no overlay between the differentially regulated genes in the two experiments. This suggests that none of these genes are regulated due to an effect on GFP expression. Therefore, we consider these genes off-target effects. Expression of genes involved in the IFN response and dsRNA-induced signaling did not differ between the different groups (Table 3
).
|
|
|
We have shown that unmodified and uncomplexed siRNA can be delivered to multiple tissues, including tumor xenografts, in vivo by at least two different ways of administration. Systemic delivery via tail vein injections of siRNA designed against POLR2A, a target previously shown suitable for tumor growth inhibition using antisense oligonucleotides (14), resulted in tumor growth inhibition. To further optimize in vivo application of siRNA, we choose to target the exogenously expressed eGFP gene because it allows simple readout for gene-specific knockdown. Knockdown of GFP expression should not affect tumor growth kinetics allowing simultaneous evaluation of RNAi (GFP knockdown) and sequence unrelated effect on tumor growth. In line with tumor growth inhibition due to siPOLR2A, eGFP expression in eGFP-positive MiaPaca-II tumors was significantly lowered after systemic delivery of siRNA directed against eGFP. These results agree with three other independent studies. First, it was shown that systemically administered siRNA induced silencing of CXCR4 and reduced breast cancer metastasis (21). Second, silencing of unmodified siRNA against CEACAM6 reduced pancreatic tumor growth (18). Furthermore, daily i.p. injections of siRNA directed against Bcl-2 resulted in growth inhibition of pancreatic cancer xenografts (22). In our study, knockdown was also achieved irrespective of the route of administration because continuous delivery of siGFP via osmotic minipumps resulted in comparable levels of knockdown compared with delivery via tail vein twice weekly. Using the tail vein method, the amount of siRNA needed to do in vivo experiments is 6-fold lower compared with pump delivery and therefore can be done cost effective.
Our studies on LNA-modified siGFP have shown that LNA offers the means to improve the serum half-life of siRNAs. These results agree with the findings of Elmen et al. (15) who also show that only end modifications are needed to obtain this increase in half-life and is in line with the predominant degradation of DNA oligonucleotides due to 3' exonuclease activity (23). Thus, introduction of a few LNA moieties into siRNA provides an excellent means to protect siRNA against degradation.
In vitro studies showed that introduction of LNA molecules in end-modified siGFP lowered its efficacy in target knockdown. Introduction of additional LNA modifications in the sense strand as in the heavily modified siGFP resulted in loss of GFP knockdown. This may be explained by the fact that RISC incorporates dsRNA and becomes active after sense strand degradation in the RISC complex (24, 25). Heavily modified siGFP may prevent sense strand degradation and therefore RISC activation could not occur. However, not all sense strand modifications inhibit RISC. Effective siRNAs with modified sense strands composed of a combination DNA and 2'-fluoro–modified nucleotides have been described (13). Alternatively, the duplex dissociates without sense strand degradation. As opposed to LNA, DNA and 2'-fluoro modifications do not affect Tm values dramatically. Therefore, the observed difference between LNA-modified and DNA/2'-fluoro–modified siRNA could also be explained by the differences in Tm. We conclude that only a few LNA modifications, like our end-modified siGFP, result in a dramatic increased half-life in serum, whereas the efficacy was only marginally reduced. These characteristics potentially improve this molecule for therapeutic application.
In agreement with target knockdown of unmodified siGFP after different routes of administration, end-modified siGFP was effective on tail vein delivery or via osmotic minipumps. Furthermore, both routes of administration resulted in comparable target knockdown due to unmodified siGFP and end-modified siGFP. This suggests that lowered efficacy as observed with in vitro assays is compensated for by increased in vivo stability.
Off-target effects on gene regulation have to be considered when using siRNA for functional genomics and therapeutic applications. Recently, it has been shown that complementarity of the seed region (bases 2–7) of a siRNA is sufficient for down-regulation of untargeted genes (26–28). In line with these observations, it was shown that 2-O-methyl modification of the second nucleotide in the antisense strand reduced silencing of most off-target transcripts with complementarity to the seed region of the siRNA, whereas target knockdown was not affected (29). In our study, end-modified siGFP showed a strong reduction in off-target gene regulation, whereas on-target knockdown was not affected to that extent, indicating that this effect is not due to lowered potency of end-modified siGFP. This agrees with findings of Elmen et al. (15) who showed that reduced off-target effects on placement of a LNA molecule at the 5' end of the sense strand. Two mechanisms could be responsible for this phenomenon. (a) Introduction of LNA into the 5' end of the sense strand results in preferential loading of the antisense strand into RISC. (b) If, despite guided RISC loading, sense strand is incorporated into RISC, this design likely prevents 5' phosphorylation and the formation of active RISC.
We have no evidence for a dsRNA-dependent protein kinase and IFN response in our study. Phosphorylated STAT1 and phosphorylated eukaryotic initiation factor-2 were not elevated in the tumors after siRNA treatment. Microarray data also did not give an indication of an IFN response. This agrees with findings of Heidel et al. (30) who showed no induction of IFN on systemic administration of uncomplexed siRNA.
In conclusion, unmodified naked siRNA-mediated target knockdown in vivo is possible at very low concentrations. Modification of siRNA with LNA molecules greatly enhanced its half-life in serum. However, it is clear that LNA modifications should be kept at a minimum to allow compatibility with the RNAi machinery. The levels of target knockdown achieved in vivo by very low-dosage unmodified siGFP and end-modified siGFP suggest that lowered efficacy was compensated for by increased stability. LNA-modified siRNA should be preferred over unmodified molecules because introduction of LNA in the end-modified siGFP strongly reduced off-target gene regulation.
Footnotes
Grant support: Dutch Cancer Society project no. 2003-2968.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 4/10/06; revised 11/23/06; accepted 1/31/07.
References
by short interfering RNA in plasmacytoid dendritic cells through TLR7. Nat Med 2005;11:263–70.[CrossRef][Medline] Braasch DA, Jensen S, Liu Y, et al. RNA interference in mammalian cells by chemically-modified RNA. Biochemistry 2003;42:7967–75.[CrossRef][Medline] Harborth J, Elbashir SM, Vandenburgh K, et al. Sequence, chemical, and structural variation of small interfering RNAs and short hairpin RNAs and the effect on mammalian gene silencing. Antisense Nucleic Acid Drug Dev 2003;13:83–105.[CrossRef][Medline] Amarzguioui M, Holen T, Babaie E, Prydz H. Tolerance for mutations and chemical modifications in a siRNA. Nucleic Acids Res 2003;31:589–95.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Cell Growth & Differentiation |
