THE JOURNAL OF GENE MEDICINE RESEARCH ARTICLE J Gene Med 2008; 10: 101–109. Published online 20 November 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jgm.1124
Sequence-specific inhibition of RNA polymerase III-dependent transcription using Zorro locked nucleic acid (LNA)
Rongbin Ge Mathias G. Svahn Oscar E. Simonson Abdalla J. Mohamed Karin E. Lundin C. I. Edvard Smith* Clinical Research Center, Department of Laboratory Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, SE-141 86 Stockholm, Sweden *Correspondence to: C. I. Edvard Smith, Clinical Research Center, Department of Laboratory Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, SE-141 86 Stockholm, Sweden. E-mail:
[email protected]
Abstract Background RNA polymerase III (pol III)-dependent transcripts are involved in many fundamental activities in a cell, such as splicing and protein synthesis. They also regulate cell growth and influence tumor formation. During recent years vector-based systems for expression of short hairpin (sh) RNA under the control of a pol III promoter have been developed as gene-based medicines. Therefore, there is an increasing interest in means to regulate pol III-dependent transcription. Recently, we have developed a novel anti-gene molecule ‘Zorro LNA (Locked Nucleic Acid)’, which simultaneously hybridizes to both strands of super-coiled DNA and potently inhibits RNA polymerase II-derived transcription. We have now applied Zorro LNA in an attempt to also control U6 promoter-driven expression of shRNA. Methods In this study, we constructed pshluc and pshluc2BS plasmids, in which U6 promoter-driven small hairpin RNA specific for luciferase gene (shluc) was without or with Zorro LNA binding sites, respectively. After hybridization of Zorro LNA to pshluc2BS, the LNA-bound plasmid was cotransfected with pEGFPluc into mammalian cells and into a mouse model. In cellular experiments, cotransfection of unhybridized pshluc2BS, Zorro LNA and pEGFPluc was also performed. Results The results showed that the Zorro LNA construct efficiently inhibited pol III-dependent transcription as an anti-gene reagent in a cellular context, including in vivo in a mouse model. Conclusions Thus, this new form of gene silencer ‘Zorro LNA’ could potentially serve as a versatile regulator of pol III-dependent transcription, including various forms of shRNAs. Copyright 2007 John Wiley & Sons, Ltd. Keywords
locked nucleic acid; shRNA; RNA polymerase III; transcription
Introduction
Received: 14 July 2007 Revised: 20 September 2007 Accepted: 21 September 2007
Copyright 2007 John Wiley & Sons, Ltd.
RNA polymerase III (pol III) is the largest RNA polymerase with the greatest number of subunits [1]. All of its products are short untranslated transcripts, which rarely exceed 300 nucleotides in length. Apart from 5S rRNA and MRP RNA, they include tRNA, which is required for translation, and the 7SL RNA, which is needed to introduce proteins into membranes as part of the signal recognition particle. Other essential pol III
102
products include the U6 and H1 RNAs, which are involved in processing mRNA and tRNA, respectively. Pol III transcription occurs at unique sites within the nucleoplasm, each of which contains on average five molecules of active polymerase [2]. Since pol IIIdependent transcription is involved in many fundamental activities in a cell, such as splicing and protein synthesis, and, moreover, regulates cell growth and influences tumor formation, knowledge about transcriptional repressors is important. While one such inhibitor is p53 [3], this molecule has multiple other functions and, furthermore, is not selective for particular pol III-dependent promoters. RNA interference (RNAi) is a biological, strongly conserved post-transcriptional gene silencing mechanism mediated by double-stranded RNA molecules [4]. It was first found and applied in Caenorhabditis elegans [5], but since the breakthrough discovery that RNAi can also be used for gene knockdown in mammalian cells [6], it has become a useful tool in biological studies at large [7]. Chemically synthesized siRNAs permit transient gene repression but preclude inhibition of stable gene products as well as long-term phenotypic analyses. Alternatively, siRNAs can be constitutively transcribed as stem-loop precursors, short hairpin (sh) RNA by pol III from promoters, such as the U6 or H1, and stable cell lines can be established in which the shRNA is expressed in a constitutive manner [8–13]. These approaches based on pol III promoters, however, have a major limitation: inhibition cannot be controlled in a time- or tissue-specific manner. Conditional suppression of genes will also be important for therapeutic applications by permitting termination of treatments at the onset of unwanted side effects. For these reasons, it will be of importance to generate new, alternative forms of gene silencers which could potently control pol III-dependent transcription. A locked nucleic acid (LNA) is a nucleic acid analog containing a 2 -O,4 -C-methylene bridge in the ribose moiety [14–18]. This conformational restriction allows better stacking and higher thermal stability in hybridization to perfectly matched DNA/RNA sequences. Moreover, LNAs possess a number of attractive features. Thus, LNAs have also been used as agents to attach functional moieties to the plasmid DNA [19], as DNA correcting agents [20] as well as in the form of antigene reagents [21,22]. Recently, we generated a novel sequence-specific anti-gene reagent ‘Zorro LNA’, in which a 14-mer LNA oligonucleotide binds to the coding strand, whilst a connected 16-mer LNA binds to the opposite template strand. Our data suggested that the novel antigene reagent induced potent strand invasion into the DNA duplex and powerful inhibition of RNA polymerase II-dependent transcription in a sequence-specific way, also in a cellular context [23]. In order to expand the application of this novel LNA construct we have now explored whether Zorro LNA can also control U6 pol III promoter-driven expression of shRNA. Our results demonstrate that the Zorro LNA construct efficiently inhibited pol III-dependent transcription in a cellular context, including in vivo, in Copyright 2007 John Wiley & Sons, Ltd.
R. Ge et al.
a mouse model. Thus, this new form of gene silencer could potentially serve as a versatile regulator of pol III-dependent transcription, including various forms of shRNAs.
Materials and methods Cell culture The NIH 3T3 and 293T cells were obtained from the American Type Culture Collection (ATCC, HTB-96, Rockville, MD, USA). Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) heat-inactivated fetal calf serum and 100 µg/ml of penicillin-streptomycin (Invitrogen, Stockholm, Sweden) at 37 ◦ C in a humidified 5% CO2 incubator.
Plasmid constructs and siRNA The plasmid pEGFPLUC was constructed by standard molecular cloning procedures [24], in which the EGFPluc reporter gene was driven by human cytomegalovirus (CMV) early promoter/enhancer elements. In order to generate pshluc and pshluc2BS plasmids, DNA fragments including U6 promoter-driven, small hairpin RNA specific for luciferase gene (shluc) 5 ggattccaattcagcgggagccacctgatgaagcttgatcgggtggctctcgctgagttggaatccattttt-3 , with or without two Zorro LNA binding sites, were ordered from Genscript (NJ, USA). The construct was inserted into SpeI and ApaI sites of the pEGFPLUC plasmid, respectively. Then, the fragment including the CMV promoter-EGFPluc reporter gene, was deleted by VspI and HpaI, followed by self-ligation of the vector.
Transfection Cells were seeded in six-well plates at a density of 1 × 106 cells per well and allowed to attach for 18 h prior to transfection. The plasmids pshluc and pshluc2BS (1.6 µg) were incubated with 10-fold molar excess of Zorro LNA (1 µM) per mole of binding site (BS) in 20 mM phosphate buffer (pH 6.8) at 37 ◦ C overnight. Immediately before transfection, 0.4 µg of target plasmid pEGFPLUC was mixed with respective LNA-bound plasmids. Fugene 6 reagent (Roche Molecular Biochemicals, Stockholm, Sweden) was used to deliver the plasmid mixture. Transfection solution was prepared according to the manufacturer’s protocol with serumfree medium. Cells were harvested for analysis of protein expression 48 or 72 h after transfection. In cotransfection experiments, 1 µg of plasmid pshluc2BS and 0.25 µg of target plasmid pEGFPLUC were mixed with increasing amounts of Zorro LNA, (50-fold, 100-fold, 150-fold, 200-fold molar excess per BS), respectively, immediately before transfection, and then the mixture J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
103
Zorro LNA for Sequence-Specific Inhibition of RNA pol III
of LNAs and plasmid (50 µl) was transfected into U-2 OS cells (2.5 × 105 cells per well in 24-well plates in 450 µl of serum-containing medium) by Fugene 6. Cells were harvested for analysis of protein expression 48 h after transfection. All of the experiments were repeated at least three times. Values are presented as the mean of a triplicate ([±] standard deviation (SD)) from a representative experiment.
4–20% polyacrylamide TBE gels (Invitrogen). All of the experiments were repeated at least three times. Relative intensities are presented as the mean of a triplicate ([±] SD) from a representative experiment.
Oligonucleotides and LNA
Total RNA was prepared using the Qiagen RNA/DNA mini kit (Qiagen, Stockholm, Sweden), and RT-PCR was performed using the Qiagen Onestep RT-PCR Kit (Qiagen) and gene-specific primers according to the manufacturer’s protocol. Forward and reverse primers for EGFPluc were 5 -CTTCTTCAAGTCCGCCATG-3 and 5 -GAACCTCTTGGCAACCGCT-3 , respectively. GAPDH primers were 5 -GGGTGTGGGCAAGGTCATCC-3 and 5 TCCACCACCCTGTTGCTGTA-3 , respectively [25]. RTPCR products were analyzed on Novex 4–20% polyacrylamide TBE gels (Invitrogen). The EGFPluc primers amplified a fragment of 1 499 bp. For detection of EGFPluc and GAPDH transcripts in the exponential phase of amplification, initial experiments were performed to optimize assay conditions (i.e. number of cycles, primer concentration and amount of RNA template).
Oligonucleotides containing Zorro LNA binding sites (5 -CAGCGCATGGGTGCCCCTCCTCTTTCTTCA-3 and 5 and TGAAGAAAGAGGAGGGGCACCCATGCGCTG-3 ) primers used in this study were synthesized by DNA Technology A/S (Aarhus, Denmark) or Cybergene AB (Huddinge, Sweden) and purified with cartridge purification. A series of Cy3 and Cy5 end-labeled, HPLC-purified LNAs, shown in Figure 1, were purchased from Proligo SAS (Paris, France).
LNA binding assay Cy3/Cy5-labeled LNAs binding to the target sites of the plasmid were detected using Molecular Imager FX equipment (Bio-Rad Laboratories, Sundbyberg, Sweden). The Zorro LNA was prepared by mixing both LNA389 and LNA390 at a ratio of 1 : 1 and incubating the solution at 95 ◦ C for 5 min and then cooling it down to room temperature slowly. An amount of 5 µg of plasmid DNA, pshluc or pshluc2BS, was mixed with 10-fold molar excess per BS of LNA oligomers in 20 mM phosphate buffer (pH 6.8) and incubated at 37 ◦ C overnight, respectively. This reaction was analyzed by electrophoresis in Novex
RNA isolation and reverse-transcription polymerase chain reaction (RT-PCR)
Western blot analysis Cells were lysed in boiling lysis buffer (2% SDS, 10 mM Tris-HCl, pH 6.8), and protein lysate was fractionated by 4–20% Novex sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels (Invitrogen) and transferred to nitrocellulose membranes (Advantec MFS, Dublin, CA, USA). The nitrocellulose membrane was incubated with blocking buffer (5% dried milk in phosphate-buffered saline (PBS) and 0.1% Tween-20), probed with a 1 : 5000 dilution of rabbit anti-GFP antibody (BD Biosciences, Stockholm, Sweden) followed by goat anti-rabbit IgG conjugated to horseradish peroxidase (1 : 2000 dilution). Immune complexes were detected with the Supersignal West Femto Chemiluminescence Western blotting detection system (Pierce, Rockford, IL). anti-γ -Tubulin antibody (Sigma, Stockholm, Sweden) was used to detect γ -tubulin as internal control. antiNeomycin phosphotransferase II antibodies (Biosite, Stockholm, Sweden) were applied to detect neomycin phosphotransferase.
Immunofluorescence Figure 1. Schematic representation of two target sites (2BS) and sequence bound by LNA. (a) Schematic representation of plasmid pshluc, pshluc2BS. (b) Zorro LNA that binds to one of the two target sites. The Zorro LNA consists of a 14-mer (LNA389) bound to the coding strand, while another 16-mer arm (LNA390) is bound to the template strand with the two arms connected by a bridge containing seven base pairs Copyright 2007 John Wiley & Sons, Ltd.
Cells were pre-seeded in six-well plates at a density of 1 × 106 cells per well. Transfection of plasmids was as described in the ‘Transfection section’. Cells were visualized 48 h post-transfection using a fluorescence microscope (Carl Zeiss, Oberkochen, Germany). J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
104
Hydrodynamic transfection Hydrodynamic transfections of plasmids in PBS were carried out as described [26,27]. Briefly, 1.8 mL of DPBS (with no MgCl2 or CaCl2 ) were mixed with 1 µg of pEGFPluc plasmid, 1 µg of pEGFPluc plasmid and 4 µg of shRNA plasmids alone or prehybrdized with Zorro LNA, respectively. This mixture was introduced by tail vein injection over a period of 5 s to inbred 25- to 30-g female NMRI mice. Live, anesthetized mice were imaged for 10 s to 5 min using an intensified CCD camera (IVIS Imaging System, Xenogen) at day 1, day 3 and day 6 after injection, respectively. This image is comprised of a pseudocolor image representing intensity of emitted light (red most intense and blue least intense) superimposed on a grayscale reference image (for orientation). All animal experiments were approved by the local ethical committee at Karolinska Institutet.
Intramuscular injection Physiological saline solution (50 µl) was mixed with 5 µg of pEGFPluc plasmid, 5 µg of pEGFPluc plasmid and 20 µg of shRNA plasmid alone or prehybridized with Zorro LNA, respectively. This mixture was injected into the tibialis anterior muscle (sin et dex). Injection pressure and time were constant between the injections and dorsal flexation of the talocrural joint was observed as a sign for correct injection. Live, anesthetized mice were imaged for 1 min using an intensified CCD camera (IVIS Imaging System, Xenogen, MA, USA) at 24 h post-injection. All animal experiments were approved by the local ethical committee at Karolinska Institutet.
R. Ge et al.
were run on 4–20% polyacrylamide TBE gels (Figure 2a). Measurement of Cy5 fluorescence suggested that strand invasion into super-coiled DNA induced by Zorro LNA was much more powerful than that induced by only one of the linear Zorro arms, LNA390 (16-mer, lower half of Zorro LNA as depicted in Figure 1), whereas on control plasmid pshluc, lacking binding sites, only very weak unspecific binding by Zorro LNA was seen (Figure 2a, panel 1). Detection of Cy3 gave a similar result, namely that binding efficiency induced by Zorro LNA was also stronger than that induced by the other arm, LNA389 (14-mer, upper half of Zorro LNA). Likewise, only very weak unspecific binding of Zorro LNA to the control plasmid was observed when Cy3 was documented (Figure 2a, panel 2). Supercoiled plasmid was stained by SYBR Gold to determine that the samples were equally loaded (Figure 2a, panel 3). Densitometer measurements revealed that binding induced by Zorro LNA was 9.2-fold more efficient than that induced by linear LNA390 and 8.7-fold more efficient than that induced by linear LNA389 (Figure 2b). These results suggest that plasmid binding of a 14-mer or a 16mer linear LNA is quite limited, whereas the Zorro LNA construct was considerably more potent.
Results Effect of binding Zorro LNAs to double-stranded DNA targets in a pol III promoter Recently, we presented a novel construct designated ‘Zorro LNA’ [23], which simultaneously binds to both DNA strands and blocks pol II transcription. In order to investigate if this construct also can induce inhibition of pol III-dependent transcription, a plasmid construct was designed, in which two Zorro LNA binding sites (BS) were cloned between the U6 promoter and a small hairpin RNA specific for luciferase RNA (shluc) (Figure 1a). Because hybridization of LNA to DNA causes only modest retardation in gel mobility shift assays, all of the LNAs used in this study were end-labeled with Cy3 or Cy5 fluorophores to simplify their detection. The plasmids pshluc and pshluc2BS (5 µg) (Figure 1a) were mixed with 10-fold molar excess of Zorro LNA, formed by the 14-mer LNA389 together with the 16-mer LNA390 (Figure 1b); linear LNA389 and LNA390 were included as controls. LNA-bound plasmids and control plasmids Copyright 2007 John Wiley & Sons, Ltd.
Figure 2. DNA duplex binding of Zorro LNA. (a) Comparison of binding efficiency of Zorro LNA and other linear LNA constructs. (1) Signals from LNA labeled by Cy5 probe. (2) Signals from LNA labeled by Cy3 probe. (3) Signals from super-coiled plasmid stained with SYBR Gold. (b) Densitometer analysis of binding efficiency induced by Zorro LNA and linear LNAs J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
Zorro LNA for Sequence-Specific Inhibition of RNA pol III
Inhibition of RNA pol III-dependent transcription using Zorro LNA In order to expand the potential application of Zorro LNA, we applied this novel construct in an attempt to control pol III-mediated transcription by studying U6 promoterdriven expression of shRNA. The pshluc2BS plasmid (1.6 µg) was pre-hybridized with 10-fold molar excess of Zorro LNA, linear LNA389, or linear LNA390, respectively, in 20 mM phosphate buffer at 37 ◦ C overnight.
105
The LNA-bound plasmid (pshluc2BS) and control plasmid (pshluc) were mixed with pEGFPluc reporter plasmid (0.4 µg), respectively, prior to transfection into preseeded 293T cells. After 48 h, cells were harvested and EGFPluc fusion protein expression was measured by Western blot of cellular lysates (Figure 3a). Densitometer measurements revealed an approximately 80% inhibition of EGFPluc expression caused by pshluc2BS plasmid. Binding of Zorro LNA to pshluc2BS plasmid compromised the inhibition of EGFPluc expression induced
Figure 3. Inhibition of pol III-dependent transcription caused by Zorro LNA. (a) Expression of EGFPluc reporter gene after co-transfection of pEGFPluc plasmid and pre-hybridized plasmid pshluc or pshluc2BS with different LNA constructs into 293T cells, respectively, measured by densitometer analysis. (b) Imaging of cells transfected as described in (a). 1. pEGFPluc, 2. pEGFPluc+pshluc, 3. pEGFPluc+pshluc+Zorro LNA, 4. pEGFPluc+pshluc2BS, 5. pEGFPluc+pshluc2BS+Zorro LNA, 6. pEGFPluc+pshluc2BS+LNA389, 7. pEGFPluc+pshluc2BS+LNA390 (c) EGFPluc mRNA level as determined by RT-PCR. Densitometer analysis of EGPFluc mRNA level Copyright 2007 John Wiley & Sons, Ltd.
J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
106
by pshluc2BS, whereas binding of linear LNA389 or LNA390 to plasmid pshluc2BS did not detectably inhibit U6 promoter-dependent transcription. When added to the control plasmid pshluc, lacking binding sites, Zorro LNA did not inhibit U6 promoter-dependent transcription (Figure 3b). In order to define the specific effect of Zorro LNA on transcription, we also measured the expression level of the neomycin resistance gene located in cis and driven by the SV40 early promoter (Figure 1a). This analysis demonstrated that Zorro LNA did not affect transcription of the adjacent neomycin resistance gene. Because both the pEGFPluc and the pshluc/pshluc2BS plasmids carry the neomycin resistance gene, when pshluc/pshluc2BS plasmid was co-transfected with EGFPluc reporter plasmid at the ratio of 4 : 1, the expression of the neomycin resistance gene was 5 times higher than the expression in the sample where only the pEGFPluc reporter plasmid was transfected. Expression of the γ -tubulin protein was used as internal control to demonstrate that the samples were equally loaded. The same results were achieved when analyzing transfected cells by a fluorescence microscope. Plasmid pshluc2BS induced significant down-regulation of EGFPluc fusion protein expression. Again, the binding of Zorro LNA to pshluc2BS inhibited the U6 promoter-derived transcription, thus compromising the down-regulation of EGFPluc protein expression. No significant inhibition of U6 promoterderived transcription was observed, using either control plasmid pshluc, or any of the other oligonucleotides, i.e. linear LNA constructs (Figure 3b). In order to verify that the elevated protein expression was due to increased RNA levels and not just due to changes in translation efficiency of the EGFPluc, the mRNA level was determined by RT-PCR. Densitometer measurements revealed that the mRNA level correlated with the expression level of the EGFPluc fusion protein (Figure 3c). Similar results were obtained in the fibroblast cell line, NIH-3T3 (data not shown).
Zorro LNA induces inhibition of pol III transcription in a cellular context In this study, we co-transfected 1 µg of unhybridized plasmid pshluc2BS with increasing amounts of Zorro LNA, respectively, as well as 0.25 µg EGFPluc reporter plasmid. After 48 h of culture, the EGFPluc protein expression was monitored by Western blot analysis (Figure 4a). Densitometer measurements revealed that pshluc2BS induced 80% inhibition of EGFPluc fusion protein. When 50-fold molar excess of Zorro LNA was co-transfected with pshluc2BS plasmid, the inhibition was reduced to 54%, and, with an increasing amount of Zorro LNA, the degree of inhibition was further reduced (100-fold molar excess, 33% inhibition, 150-fold and 200-fold molar excess almost completely compromised the inhibition of EGFPluc). On BS-lacking plasmid, pshluc, 200-fold molar excess of Zorro LNA did not inhibit U6-dependent transcription. Copyright 2007 John Wiley & Sons, Ltd.
R. Ge et al.
Figure 4. Cellular effect of Zorro LNA on pol III-dependent transcription. (a) Expression of EGFPluc fusion protein after co-transfection of EGFPluc target gene and unhybridized plasmids pshluc, pshluc2BS as well as Zorro LNA into 293T cells, respectively. (50) 50-fold molar excess, (100) 100-fold molar excess, (150) 150-fold molar excess, (200) 200-fold molar excess. (b) Densitometer analysis of inhibition of EGFPluc expression induced by Zorro LNA
Effect of Zorro LNA on pol III-dependent transcription in a mouse model To test the effects of Zorro LNA on inhibition of pol III-dependent transcription in vivo, we utilized hydrodynamic delivery to co-inject EGFPluc reporter plasmids and Zorro LNA-bound pshluc2BS plasmid/control pshluc2BS plasmid or irrelevant shBtk plasmid. High-pressure injection of naked DNA into the tail vein of rodents leads to efficient transgene expression in the liver [26,27]. Luciferase expression was monitored in living animals using quantitative, whole-body imaging [28]. The Zorro LNA was pre-hybridized with 4 µg pshluc2BS plasmid, and then co-injected with 1 µg EGFPluc reporter plasmid into the tail vein of mice, resulting in rapid cytoplasmic delivery J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
Zorro LNA for Sequence-Specific Inhibition of RNA pol III
107
Figure 6. Effect of Zorro LNA on pol III-dependent transcription in mice following intramuscular injection. EGFPluc reporter gene plasmid (5 µg) was co-injected into the tibialis anterior muscle of mice with 20 µg of pshluc2BS pre-hybridized with Zorro LNA or other controls. On day 1 post-injection, luciferin was administered and the mice were studied using the IVIS in vivo imaging system
Figure 5. Effect of Zorro LNA on pol III-dependent transcription in mice using hydrodynamic transfection. (a) EGFPluc reporter gene plasmid (1 µg) was co-injected at constant high pressure (1.8 ml over 4–5 s) into the tail veins of mice with 4 µg of pshluc2BS pre-hybridized with the Zorro LNA or other controls. On day 1 post-injection, luciferin was administered intraperitoneally, and the mice were studied using the IVIS in vivo imaging system. shBtk, irrelevant shluc. (b) Luciferase expression in live mice on days 1, 3 and 6 after hydrodynamic injection
of nucleic acids. As seen in Figure 5a, injection of pshluc2BS plasmid reduced luciferase expression by greater than 90% in vivo. However, injection of Zorro LNA-bound pshluc2BS plasmid blocked pol III-derived transcription, thus compromising inhibition of luciferase expression. Luciferase expression was analyzed in living mice on days 1, 3 and 6 after hydrodynamic transfection. At all time points tested, pshluc2BS plasmid devoid of Zorro LNA robustly blocked EGFPluc expression, compared to mice injected with Zorro LNA-bound pshluc2BS plasmid. There was no significant down-regulation detected in mice injected with the unrelated, control shBtk plasmid with or without Zorro LNA (Figure 5b). Similar results were achieved following intramuscular injection of plasmids (Figure 6).
Discussion Recently, we have generated a novel sequence-specific anti-gene reagent ‘Zorro LNA’, in which a 14-mer LNA Copyright 2007 John Wiley & Sons, Ltd.
oligonucleotide binds to the coding strand, whilst a connected 16-mer LNA binds to the opposite template strand. We have verified that this takes place in a sequence-specific manner and that potent inhibition of pol II-dependent transcription in cultured mammalian cells is induced [23]. In order to further study the effect of Zorro LNA as a transcriptional blocker, we investigated the effect on pol III. For this purpose two Zorro LNA binding sites were inserted between a U6 promoter and a shRNA cassette against the luciferase gene (Figure 1a). Two sites were used because, in our previous studies, there was no significant inhibition detected after transfection of Zorro LNA-bound plasmid carrying a single binding site into cells. It is highly likely that this is due to cooperative binding, since this is known to occur from other studies [24,29,30]. We are currently generating extended versions of Zorro LNAs in order to address this issue further. While the current need for two adjacent sites necessitates the use of two different Zorro LNAs when non-repetitive sequences are targeted, this could also be advantageous owing to the fact that it increases sequence specificity. In spite of that we have not as yet combined different Zorro target sequences; for shRNAencoding plasmids, the target sequences used in this study could presumably be grafted onto any such plasmid, irrespective of the composition of its shRNA. Extending the region between the promoter and the shRNA normally decreases expression, but the construct which we have generated shows that sufficient expression is obtained to allow regulation by Zorro LNAs. An alternative to LNA could be the synthetic oligonucleotide peptide nucleic acid (PNA). However, although bisPNA also induces pronounced strand invasion into a DNA duplex, it is known to only recognize homopurine sequences [31,32], thus limiting the versatility of PNA. Moreover, our previous data in a pol II system showed that if one arm of Zorro LNA was substituted by a PNA J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
108
oligonucleotide, the effect of binding to a DNA duplex decreased significantly [23]. Our in vitro data showed a strong dsDNA binding induced by the Zorro LNA construct; the binding efficiency was almost 10-fold higher than the control linear LNA389 or LNA390 oligomers (Figure 2). In order to evaluate the biological effect of the Zorro LNA construct on pol III-dependent transcription in cells, the two Zorro LNA binding sites were inserted between the U6 promoter and the shRNA cassette (Figure 1a). Because a stretch of four thymidines (TTTT) is the shortest termination signal recognized by pol III in mammals [33], the inserted LNA binding sites including (TTT) in the coding strand did not terminate transcription, as presented in Figures 3 and 4. In this study, we found that pshluc2BS induced 80% down-regulation of EGFPluc fusion protein expression, whereas binding of Zorro LNA inhibited pol III-derived transcription, thus compromising the inhibition of EGFPluc expression. There was no evidence suggesting that pol III could be suppressed by the control, linear LNA389 or 390 constructs (Figure 3). We also observed that the efficiency of EGFPluc reporter gene down-regulation induced by the pshluc plasmid was more pronounced (greater than 90%) than the one induced by the pshluc2BS plasmid (approximately 80%). We believe that this phenomenon is because the pol III-dependent transcription becomes weaker with an increasing distance between the pol III promoter and the shRNA cassette. The inhibition of transcription was limited to the gene carrying the Zorro LNA target site, since the expression of the plasmid selection marker, neomycin phosphotransferase, was unaffected. Similar to the approach used for the pol II promoter [23], we also tested the effect of peptide nucleic acids (PNAs) with an identical sequence as described [23] on a pol III promoter. However, also for the pol III promoter, PNA was unable to significantly block transcription, although binding to the target site was clearly demonstrable (data not shown). We also evaluated the biological effect of the novel Zorro LNA in the cellular context in the absence of prehybridization. Previously we have shown that short-term incubation was not enough to saturate the binding sites; overnight incubation (16 h) was required to saturate binding [23], i.e. during co-transfection of LNA and plasmids carrying the specific binding sites, the Zorro LNAs cannot find their own targets in the plasmid during a transfection procedure of only 15 min. The fact that as much as 150- to 200-fold molar excess of Zorro LNA was required to completely compromise the inhibition of EGFPluc expression (Figure 4) might be due to the chemical nature of the molecules, compromising DNA targeting in a cellular context. In comparison to PNAs, linear LNAs have been shown to perform less well as anti-gene reagents under certain conditions [34]. In this report two consecutive transfections with LNA were required in order to inhibit cellular gene expression. Second, the intracellular route of LNAs may be subject to more roadblocks, limiting access to the nucleus. Copyright 2007 John Wiley & Sons, Ltd.
R. Ge et al.
Adding membrane-penetrating peptides and/or nuclear localization signals might be investigated in order to increase the cellular uptake [35,36]. LNA oligonucleotides display a high intracellular stability [37,38] allowing prolonged incubation times, although high doses of LNAmodified oligonucleotides have in some instances induced liver toxicity in mice [39], while others have been given high doses without any detectable toxicity [40]. Finally, we extended the ability of Zorro LNA to inhibit target gene expression in two mouse model systems using either hydrodynamic delivery of the nucleic acids to mouse liver or intramuscular injection. We observed that the pshluc2BS plasmid robustly blocked EGFPluc expression, compared to mice injected with Zorro LNA-bound pshluc2BS plasmid. Moreover, the degree of inhibition of pol III-dependent transcription in mice was similar on days 1, 3 and 6 (Figure 5). As previously reported, the luciferase activity in the liver decreases with time, most likely due to loss of DNA or to promoter silencing [41]. Nevertheless, this study clearly demonstrates that Zorro LNA inhibits pol III-derived transcription in mammals. Thus, this new form of gene silencer could potentially serve as a versatile regulator of pol III-dependent transcription, including various forms of shRNAs in vivo. However, the development of this new form of gene silencer is still in its infancy, and, similar to siRNA, adequate delivery systems are needed in order to efficiently target endogenous genes.
Acknowledgements This work was supported by the European Union grants NMP4CT-2004-013775 and ILSHB-CT-2005-018716. Moreover, we also appreciate the support from The Wallenberg Foundation, the Swedish Science Council, Aroseniusfonden, Sigurd and Elsa Golje Memorial Foundation as well as the Swedish Foundation for Strategic Research Bio-X grant.
References 1. Schramm L, Hernandez N. Recruitment of RNA polymerase III to its target promoters. Genes Dev 2002; 16: 2593–2620. 2. Pombo A, Jackson DA, Hollinshead M, et al. Regional specialization in human nuclei: visualization of discrete sites of transcription by RNA polymerase III. EMBO J 1999; 18: 2241–2253. 3. Cairns CA, White RJ. p53 is a general repressor of RNA polymerase III transcription. EMBO J 1998; 17: 3112–3123. 4. Meister G, Tuschl T. Mechanisms of gene silencing by doublestranded RNA. Nature 2004; 431: 343–349. 5. Fire A, Xu S, Montgomery MK, et al. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998; 391: 806–811. 6. Elbashir SM, Harborth J, Lendeckel W, et al. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001; 411: 494–498. 7. Hannon GJ, Rossi JJ. Unlocking the potential of the human genome with RNA interference. Nature 2004; 431: 371–378. 8. Brummelkamp TR, Bernards R, Agami R. A system for stable expression of short interfering RNAs in mammalian cells. Science 2002; 296: 550–553. 9. McManus MT, Sharp PA. Gene silencing in mammals by small interfering RNAs. Nat Rev Genet 2002; 3: 737–747. J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
Zorro LNA for Sequence-Specific Inhibition of RNA pol III 10. Miyagishi M, Taira K. U6 promoter-driven siRNAs with four uridine 3 overhangs efficiently suppress targeted gene expression in mammalian cells. Nat Biotechnol 2002; 20: 497–500. 11. Paddison PJ, Caudy AA, Bernstein E, et al. Short hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells. Genes Dev 2002; 16: 948–958. 12. Paul CP, Good PD, Winer I, et al. Effective expression of small interfering RNA in human cells. Nat Biotechnol 2002; 20: 505–508. 13. Yu JY, DeRuiter SL, Turner DL. RNA interference by expression of short-interfering RNAs and hairpin RNAs in mammalian cells. Proc Natl Acad Sci U S A 2002; 99: 6047–6052. 14. Braasch DA, Corey DR. Locked nucleic acid (LNA): fine-tuning the recognition of DNA and RNA. Chem Biol 2001; 8: 1–7. 15. Koshkin AA, Wengel J. Synthesis of novel 2 ,3 -linked bicyclic thymine ribonucleosides. J Org Chem 1998; 63: 2778–2781. 16. Obika S, Nanbu D, Hari Y, et al. Stability and structural features of the duplexes containing nucleoside analogues with a fixed N-type conformation, 2 -O,4 -C-methyleneribonucleosides. Tetrahedron Lett 1998; 39: 5401. 17. Singh SK, Kumar R, Wengel J. Synthesis of novel bicyclo[2.2.1] ribonucleosides: 2 -amino- and 2 -thio-LNA monomeric nucleosides. J Org Chem 1998; 63: 6078–6079. 18. Vester B, Wengel J. LNA (locked nucleic acid): high-affinity targeting of complementary RNA and DNA. Biochemistry 2004; 43: 13233–13241. 19. Hertoghs KM, Ellis JH, Catchpole IR. Use of locked nucleic acid oligonucleotides to add functionality to plasmid DNA. Nucleic Acids Res 2003; 31: 5817–5830. 20. Parekh-Olmedo H, Drury M, Kmiec EB. Targeted nucleotide exchange in Saccharomyces cerevisiae directed by short oligonucleotides containing locked nucleic acids. Chem Biol 2002; 9: 1073–1084. 21. Brunet E, Alberti P, Perrouault L, et al. Exploring cellular activity of locked nucleic acid-modified triplex-forming oligonucleotides and defining its molecular basis. J Biol Chem 2005; 280: 20076–20085. 22. Brunet E, Corgnali M, Perrouault L, et al. Intercalator conjugates of pyrimidine locked nucleic acid-modified triplex-forming oligonucleotides: improving DNA binding properties and reaching cellular activities. Nucleic Acids Res 2005; 33: 4223–4234. 23. Ge R, Heinonen JE, Svahn MG, et al. Zorro locked nucleic acid induces sequence-specific gene silencing. FASEB J 2007; 21: 1902–1914. 24. Lundin KE, Ge R, Svahn MG, et al. Cooperative strand invasion of supercoiled plasmid DNA by mixed linear PNA and PNApeptide chimeras. Biomol Eng 2004; 21: 51–59. 25. Catapano CV, McGuffie EM, Pacheco D, et al. Inhibition of gene expression and cell proliferation by triple helix-forming oligonucleotides directed to the c-myc gene. Biochemistry 2000; 39: 5126–5138.
Copyright 2007 John Wiley & Sons, Ltd.
109 26. Liu F, Song Y, Liu D. Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther 1999; 6: 1258–1266. 27. Zhang G, Budker V, Wolff JA. High levels of foreign gene expression in hepatocytes after tail vein injections of naked plasmid DNA. Hum Gene Ther 1999; 10: 1735–1737. 28. Contag CH, Spilman SD, Contag PR, et al. Visualizing gene expression in living mammals using a bioluminescent reporter. Photochem Photobiol 1997; 66: 523–531. 29. Lundin KE, Hasan M, Moreno PM, et al. Increased stability and specificity through combined hybridization of peptide nucleic acid (PNA) and locked nucleic acid (LNA) to supercoiled plasmids for PNA-anchored ‘‘Bioplex’’ formation. Biomol Eng 2005; 22: 185. 30. Kurakin A, Larsen HJ, Nielsen PE. Cooperative strand displacement by peptide nucleic acid (PNA). Chem Biol 1998; 5: 81–89. 31. Peffer NJ, Hanvey JC, Bisi JE, et al. Strand-invasion of duplex DNA by peptide nucleic acid oligomers. Proc Natl Acad Sci U S A 1993; 90: 10648–10652. 32. Cherny DY, Belotserkovskii BP, Frank-Kamenetskii MD, et al. DNA unwinding upon strand-displacement binding of a thyminesubstituted polyamide to double-stranded DNA. Proc Natl Acad Sci U S A 1993; 90: 1667–1670. 33. Braglia P, Percudani R, Dieci G. Sequence context effects on oligo(dT) termination signal recognition by Saccharomyces cerevisiae RNA polymerase III. J Biol Chem 2005; 280: 19551–19562. 34. Beane RL, Ram R, Gabillet S, et al. Inhibiting gene expression with locked nucleic acids (LNAs) that target chromosomal DNA. Biochemistry 2007; 46: 7572–7580. 35. Abes R, Arzumanov AA, Moulton HM, et al. Cell-penetratingpeptide-based delivery of oligonucleotides: an overview. Biochem Soc Trans 2007; 35: 775–779. 36. Branden LJ, Mohamed AJ, Smith CI. A peptide nucleic acidnuclear localization signal fusion that mediates nuclear transport of DNA. Nat Biotechnol 1999; 17: 784–787. 37. Elmen J, Thonberg H, Ljungberg K, et al. Locked nucleic acid (LNA) mediated improvements in siRNA stability and functionality. Nucleic Acids Res 2005; 33: 439–447. 38. Schmidt KS, Borkowski S, Kurreck J, et al. Application of locked nucleic acids to improve aptamer in vivo stability and targeting function. Nucleic Acids Res 2004; 32: 5757–5765. 39. Swayze EE, Siwkowski AM, Wancewicz EV, et al. Antisense oligonucleotides containing locked nucleic acid improve potency but cause significant hepatotoxicity in animals. Nucleic Acids Res 2007; 35: 687–700. 40. Roberts J, Palma E, Sazani P, et al. Efficient and persistent splice switching by systemically delivered LNA oligonucleotides in mice. Mol Ther 2006; 14: 471–475. 41. McCaffrey AP, Meuse L, Karimi M, et al. A potent and specific morpholino antisense inhibitor of hepatitis C translation in mice. Hepatology 2003; 38: 503–508.
J Gene Med 2008; 10: 101–109. DOI: 10.1002/jgm
