Article pubs.acs.org/ac
Unusual −1 Ribosomal Frameshift Caused by Stable RNA G‑Quadruplex in Open Reading Frame Tamaki Endoh† and Naoki Sugimoto*,†,‡ †
Frontier Institute for Biomolecular Engineering Research (FIBER), Konan University, 7-1-20 Minatojima-minamimachi, Kobe 650-0047, Japan ‡ Faculty of Frontiers of Innovative Research in Science and Technology (FIRST), Konan University, 7-1-20 Minatojima-minamimachi, Kobe 650-0047, Japan S Supporting Information *
ABSTRACT: Tertiary structures formed by mRNAs impact the efficiency of the translation reaction. Ribosomal frameshift is a well-characterized recoding process that occurs during translation elongation. Pseudoknot and stem-loop structures may stimulate frameshifting by causing a translational halt at a slippery sequence. In this study, we evaluated the efficiency of an unusual −1 frameshift caused by a noncanonical RNA Gquadruplex structure in mammalian cells. The reporter gene construct consisting of a fluorescent protein and Luciferase enabled evaluation of apparent and absolute values of the −1 frameshift efficiency and revealed significant increase of the efficiency by G-quadrupex forming potential sequence. In addition, berberine, a small molecule that binds to and stabilizes G-quadruplex structures, further increased the frameshift efficiency. These results indicate that the stable G-quadruplex structure stimulates the unusual −1 frameshift and has a potential to regulate the frameshift with its ligand.
T
of the ribosome at a slippery sequence and the rearrangement of codon−anticodon base pairs between mRNA and tRNAs in the A-site and the P-site.7−9 It has been suggested that the RNA structure functions as a roadblock, halting the ribosome at the slippery sequence and providing tension on the mRNA−tRNA interaction to cause the rearrangement.4,8 As only certain sequences forming pseudoknots or stem-loop structures stimulate frameshifting, the properties of the structures that induce frameshifting are not well understood.7,10 G-quadruplexes are noncanonical structures formed by guanine-rich sequences of single-stranded DNA or RNA.11,12 G-quadruplexes have noteworthy stability under physiological conditions and structural properties different from those of canonical duplexes. Several G-quartets formed by four guanine bases through Hoogsteen base pairs stack in parallel or antiparallel fashion to form the G-quadruplex structures. Both DNA and RNA G-quadruplexes are stabilized under cell-like molecular crowding conditions.13 Recent studies focusing on oncogene mRNAs suggested that G-quadruplex formation in 5′ UTRs suppresses the small ribosomal subunit during the scanning phase.14 In addition to the G-quadruplexes in UTRs, we recently showed that the ribosome in the elongation phase is also suppressed by G-quadruplexes in the ORFs of mRNAs.15
ranslation is one of the most fundamental processes in the central dogma that explains the flow of genetically encoded information of nucleotide sequences into functionally active amino acid sequences. In cells, mRNAs are produced as single-stranded polynucleotides and form complicated secondary and tertiary structures stabilized by interactions such as Watson−Crick and Hoogsteen base pairs and by interactions with the 2′ hydroxyl groups in their ribose sugars that can behave as both hydrogen bond donors and acceptors.1 Several mechanisms involving specific structures of the mRNA, such as internal ribosome entry site and riboswitches, are known to control initiation of translation.2 These structural elements are normally located in the untranslated regions (UTR) of mRNAs. In contrast to the functional mRNA structures in UTRs, some RNA structures in mRNA open reading frames (ORFs) exert elaborate effects on translation. Ribosomal frameshifting is one of the recoding mechanisms known to be endogenously programed in some specific genes in all living organisms, including virus, prokaryote, and eukaryote.3 During frameshifting, the ribosome shifts its reading frame on the mRNA, and this results in expression of more than one protein product from a single transcript. Especially in the case of −1 frameshift, in which the ribosome is forced to shift the reading frame one nucleotide backward, structures in the mRNA, like a pseudoknot or stem-loop, are known to stimulate the frameshift in combination with an upstream slippery sequence.4−6 The key features of this type of frameshift are the translational halt © 2013 American Chemical Society
Received: August 7, 2013 Accepted: November 6, 2013 Published: November 6, 2013 11435
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
In vitro study using reporter mRNAs forming the G-quadruplex at various positions revealed that the G-quadruplex suppressed the translation elongation by halting the ribosome before the G-quadruplex structure with a 5−7 nucleotide spacer sequence.16 The results suggested that the G-quadruplex blocked the further entry of the mRNA into the ribosome. Thus, we hypothesized that the G-quadruplex located at an appropriate distance downstream of the slippery sequence may stimulate frameshifting if a translational halt could be a cause of the −1 frameshift. In this study, we investigated the effect of the G-quadruplex structure on the ORF of mRNAs on the −1 ribosomal frameshift in cultured cells. In previous studies, a pair of reporter mRNAs, one expressed a reporter enzyme through in-frame translation and the other expressed the active reporter enzyme only if a frameshift occurs, were utilized to evaluate the frameshift efficiency inside cells.6,17 The relative frameshift efficiencies caused by nucleotide sequences of interest were evaluated by comparing the relative catalytic activities of the reporter enzyme. To determine the absolute frameshift efficiency (AbFSE), radioisotope labeling of translated products that is confined to in vitro experiments is required.18,19 To investigate the AbFSE inside cells in this study, reporter mRNAs encoding a fluorescent protein (AcGFP) followed by a slippery sequence (U-UUA-AAC), which is one of the most effective slippery sequences in eukaryotic cells,18 a spacer sequence, a quadruplex forming potential (QFP) sequence derived from the E. coli eutE gene,15 and the coding sequence for Renilla luciferase (RL) in the −1 reading frame were designed (Figure 1). AcGFP and AcGFPRL were expressed from the reporter mRNA through in-frame translation and −1 ribosomal frameshift at the slippery sequence, respectively. This reporter construct enables determination of AbFSE by dividing the fluorescence signal of AcGFP-RL by the total fluorescence signal expressed from the reporter mRNA. In addition, measurement of the luminescence signal of RL enables evaluation of the apparent frameshift efficiency (ApFSE) by dividing the luminescence signal by the total fluorescence signal of AcGFP (Figure 1). To evaluate effects of the position of the G-quadruplex on the frameshift efficiency, the QFP sequence derived from the eutE gene was inserted after the slippery sequence with spacer sequences varying from 5 to 8 nucleotides according to the spacer length frequently observed in natural frameshifting mRNAs.20 Amino acid sequences fused to the C-terminus of AcGFP were designed to be identical by locating the UAG or UAA codon within the spacer sequence as differences in the fused amino acids might affect the half-life of AcGFP inside cells and obscure the frameshift signal.21,22 Differences in amino acids sequences at linker region between AcGFP and RL are inevitable depending on the spacer length and might affect specific activity and intracellular stability of AcGFP-RL that result in unexpected fluctuation of RL activity. Thus, mutant QFP sequences, in which the guanine residues in the wild-type QFP sequences were replaced by adenines, were designed as controls. In order to permit direct comparison of frameshift efficiencies between the wild-type and the mutant QFP sequences, the mutant sequences were designed to express AcGFP-RLs of the same amino acid sequences as those of the wild-type (Figure 1). The reporter mRNAs were transcribed in vitro, and formation of the RNA G-quadruplex was evaluated by monitoring fluorescence of N-methyl mesoporphyrin (NMM), which specifically binds to G-quadruplex structures
Figure 1. Reporter mRNAs used to analyze frameshift efficiency inside cells. Sequences of spacer nucleotides and wild-type or mutant quadruplex forming potential (QFP) sequence regions are indicated. The QFP regions of the wild-type sequences are underlined. Amino acids sequences produced with and without −1 frameshift are indicated below the nucleotide sequences. AcGFP and AcGFP fused to Renilla Luciferase (AcGFP-RL) are produced through in-frame translation without frameshift and with −1 frameshift, respectively. Absolute frameshift efficiency (AbFSE) and apparent frameshift efficiency (ApFSE) are calculated from output signals of the reporter proteins.
resulting in an increase in fluorescence (Figure 2a).23,24 All reporter mRNAs containing the wild-type QFP sequences had a higher fluorescence signal in the presence of NMM than did mRNAs containing mutant QFP sequences. The slight signals observed with mutant mRNAs comparing to the control without mRNA were likely due to nonspecific interaction of NMM and mRNA, as previously reported.24 These results indicated that the wild-type QFP sequence formed parallel Gquadruplex as previously shown by circular dichroism analysis using short RNA oligonucleotides of this sequence.15 Plasmid vectors encoding the reporter mRNAs were transfected into MCF7 human breast carcinoma cells. Fluorescence and luminescence signals of AcGFP and AcGFP-RL, respectively, in cell lysates were measured using a multimode microwell plate reader (Varioskan Flash; Thermo Scientific) to evaluate the ApFSEs (Figure 2b). For all the reporter mRNAs except that containing a spacer length of 8 nucleotides, significantly higher ApFSEs were observed with the mRNAs containing the wild-type QFP sequence than those containing the mutant QFP sequence. Since the G-quadruplex halted the ribosome before the G-quadruplex structure with a 5−7 nucleotide spacer sequence in vitro,16 it was considered 11436
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
Figure 2. Ribosomal frameshift stimulated by G-quadruplex formation in cells. (a) Fluorescence intensity of 5 μM N-methyl mesoporphyrin (NMM) mixed with 1 μM reporter mRNAs containing wild-type (blue) or mutant (red) QFP sequences or without mRNA (green). Fluorescence intensities were measured in buffer containing 30 mM HEPES-KOH (pH 6.8), 100 mM KCl, and 0.1% DMSO at 37 °C using 400 nm excitation and 615 nm emission. Error bars indicate SD from triplicate samples. (b) Apparent frameshift efficiencies in MCF7 cells calculated by the equation in Figure 1. MCF7 cells transfected with the plasmid vectors were lysed 24 h after transfection. Fluorescence signal was measured by using 470 nm excitation and 510 nm emission, and luminescence signal was measured after addition of coelenterazine. Error bars indicate SD from triplicate samples. (c) Predicted mechanism of the −1 ribosomal frameshift caused by the RNA G-quadruplex. (d) Fluorescence signals of cell lysate separated on 12% SDS-PAGE were imaged by 473 nm excitation and 510 nm emission. Lysate from nontransfected cells served as a control. Absolute frameshift efficiencies were calculated from triplicate samples by the equation in Figure 1
sequence was 2.4 times larger than that of mutant, which is similar to the relative ApFSE (Figure 2b). These results indicated that the reporter gene construct enabled a straightforward comparison of the ApFSEs using fluorescence and luminescence signals and a quantification of the AbFSEs by fluorescence imaging after separation of cell lysates by SDSPAGE. The AbFSE value of the wild-type QFP sequence (0.069%) was considerably lower than frameshift efficiencies reported by other groups using Luciferases as reporter proteins.6,19 GFP-type fluorescent protein is known to show relatively high stability, in which the intracellular half-life is estimated to be around 26 h.21 This stability might have the effect of lowering the observed frameshift efficiencies because of an accumulation of AcGFP produced without the frameshift. The ApFSE values of both wild-type and mutant QFP sequences 48 h after transfection decreased compared to those of 24 h likely due to the accumulation of AcGFP (data not shown). Since the single-stranded telomere sequence at the ends of chromosomes has the potential to form G-quadruplex structures, and these may contribute to chromosome maintenance and protection, targeting of G-quadruplex structures has been considered as a promising therapeutic strategy.11,26 Small-molecular ligands that specifically bind to and stabilize or destabilize the G-quadruplex have been reported from various research groups.26,27 These molecules possibly function as a trigger of the ribosomal frameshift that enables gene switching mediated by the ribosomal frameshift. Here, we used berberine, which is derived from natural alkaloids and alters gene expression through stabilization of
that the G-quadruplex temporary halted the ribosome at the slippery sequence and provided tension to stimulate the −1 ribosomal frameshift, as a similar mechanism as the pseudoknot structure (Figure 2c). The wild-type mRNA with the 6nucleotide spacer between the slippery sequence and the QFP sequence showed more than 2 times more ApFSE than that of the mutant mRNA. This spacer length is the same as that reported to be optimal for the frameshift caused by a pseudoknot structure.7,25 Almost the same ApFSEs between wild-type and mutant reporter mRNAs containing an 8nucleotide spacer suggested that the ribosome had passed through the slippery sequence when the translation elongation was halted by the G-quadruplex. To quantitatively analyze AbFSE, cells were transfected with the vectors containing wild-type and mutant QFP sequences with 6-nucleotide spacer sequences. Cells were lysed, and samples were separated on 12% SDS-PAGE at room temperature; AcGFP preserves its fluorescence during SDS-PAGE unless it has been denatured before the electrophoresis.24 Gels were imaged using a fluorescence image scanner (FLA5100; GE Healthcare). Although most of the fluorescence signal was that of the product without frameshift, a signal due to AcGFP-RL expressed through the −1 frameshift was observed in the lysate from cells transfected with a vector containing the wild-type QFP sequence, whereas the signal of AcGFP-RL present in lysates of cells transfected with the mutant was obscure (Figure 2d). The AbFSEs were calculated from the fluorescence intensities of AcGFP and AcGFP-RL in cell lysates obtained from three individual cultures (Figure 2d and Figure S1). The AbFSE of the cells transfected with the wild-type QFP 11437
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
DNA G-quadruplexes in cells,28,29 to further stimulate the −1 ribosomal frameshift caused by the wild-type QFP sequence. Although the interaction between the berberine and DNA Gquadruplexes has been well characterized, the effect of the berberine on the stability of the RNA G-quadruplexes is still unclear. Thermal stability of a wild-type oligonucleotide, which contains the guanine-rich region of the wild-type reporter mRNA (Figure 3a), was evaluated by ultraviolet (UV) absorbance at 295 nm in the presence of various concentrations of berberine (Figure 3b). The melting temperature (Tm) of the wild-type oligonucleotide increased depending on the concentration of berberine (Table S1). An RNA oligonucleotide consisting of the mutant QFP sequence did not show a melting transition characteristic of G-quadruplex in either the presence or absence of berberine (data not shown). Circular dichroism (CD) of the wild-type oligonucleotide showed a two-state transition characteristic of a parallel G-quadruplex; the CD spectra had an isosbestic point near 250 nm and linear correlations between signal intensities at 265 and 242 nm during the transition (Figure S2). We assumed that the berberine bound the parallel RNA G-quadruplex through endstacking, as proposed for the binding with DNA Gquadruplex.29,30 Thermodynamic stability of the wild-type oligonucleotide at 37 °C (−ΔG°37) in the presence of various concentrations of berberine were calculated from the UV melting profiles by the van’t Hoff equation (Table S1). The ApFSEs in MCF-7 cells were evaluated in the presence of berberine at different concentrations (Figure 3c). The ApFSE of the mRNA containing the wild-type QFP sequence was increased with the berberine concentration and showed a 4-fold higher value in the presence of 80 μM berberine than in its absence. Although the ApFSE of the mutant reporter was slightly increased in the presence of berberine likely due to a side effect of berberine unrelated to binding to G-quadruplex, the ApFSE values were less than that of the wild-type reporter mRNA in the absence of berberine. Figure 3d shows plots of the ApFSEs of the wild-type reporter mRNA versus −ΔG°37 of the G-quadruplex formed by the wild-type oligonucleotide in the presence of various concentrations of berberine. The ApFSE exponentially depended on the stability of the Gquadruplex. This trend has also been reported in a −1 ribosomal frameshift caused by a pseudoknot structure that the frameshift efficiency exponentially depended on the mechanical strength of the structure.31 Thus, it was considered that the stabilization of the G-quadruplex mediated by the berberine binding was a main factor to further stimulate the ribosomal frameshift caused by the G-quadruplex, although the in vitro and intracellular binding affinity of berberine and stabilization effect would be different. We prepared additional reporter vectors containing wild-type and mutant QFP sequences derived from the E. coli rnlA gene and the human telomeric-repeat-containing RNA (TERRA).15,32 We determined ApFSEs in the presence and absence of berberine. Although the differences in ApFSEs between wild-type and mutant QFP sequences of the rnlA gene and TERRA in the absence of berberine were smaller than that between wild-type and mutant of eutE gene, berberine increased ApFSEs of wild-type QFP sequences significantly (Figure S3). Thus, it was suggested that RNA G-quadruplexes on the ORF of mRNAs potentially stimulate the −1 ribosomal frameshift inside cells, when it was located at appropriate distance downstream of the slippery sequence.
Figure 3. Effect of stabilization of G-quadruplex by berberine on frameshift efficiency. (a) Sequence of wild-type oligonucleotide derived from reporter mRNA containing wild-type QFP sequence of eutE gene. (b) Melting transition of wild-type oligonucleotide (5 μM) in the absence (purple) and presence of 5 (dark blue), 10 (light blue), 20 (dark green), 40 (light green), 60 (orange), and 80 (red) μM berberine in buffer containing 50 mM MES-LiOH (pH 7), 0.1 mM KCl, and 0.2% DMSO. Absorbance at 295 nm was measured at a rate of 0.2 °C min−1 and normalized. (c) ApFSEs in MCF7 cells transfected with plasmid vector containing wild-type (blue) or mutant (red) QFP sequence with 6-nucleotide spacers in the presence of various concentrations of berberine. Cellular medium was replaced by berberine-containing medium at 6 h after transfection, and cells were incubated at 37 °C for 18 h. Error bars indicate SD from five samples. (d) Plots of ApFSEs in MCF7 cells transfected with the plasmid vector containing wild-type QFP sequence vs −ΔG°37 of the G-quadruplex formed by the wild-type oligonucleotide in the presence of various concentrations of berberine.
Ribosomal frameshift have been shown previously to be stimulated by certain pseudoknot and stem-loop structures. One unusual +1 frameshift, which shifts the reading frame one nucleotide forward, has been reported to be caused within a guanine-rich sequence derived from herpes simplex virus.33 This report suggested that an unusual mRNA structure formed by the guanine-rich sequence contributed to the frameshift. In 11438
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
this study, we evaluated stimulation of a −1 frameshift by Gquadruplex structures at different distances from the slippery sequence. The frameshift efficiency was quantitatively evaluated as both apparent and absolute values in cells. Significantly more of the −1 frameshift product was observed from reporters containing the wild-type QFP sequence compared to that from a reporter with a mutant QFP sequence when the spacer was 5 to 7 nucleotides. The spacer lengths corresponded to those between the position of translational halt and the QFP sequence that were revealed in our previous study using in vitro synchronized translation.16,34 Thus, the G-quadruplex stimulated the −1 frameshift by cooperative function with the slippery sequence, which would be located at the A-site of the ribosome when the translation elongation was halted by the Gquadruplex. In addition, a G-quadruplex-stabilizing ligand, berberine, further stimulated the frameshift. These results strongly suggest that a temporary halt in translation elongation due to stable RNA structures can partially stimulate the −1 ribosomal frameshift. However, since the AbFSEs were relatively small compared to the previously reported −1 ribosomal frameshift caused by a pseudoknot, not only the translational halt but also other factors such as tertiary structure and torsional resistant properties may be required for efficient stimulation of the frameshift.10
■
(8) Namy, O.; Moran, S. J.; Stuart, D. I.; Gilbert, R. J.; Brierley, I. Nature 2006, 441, 244−247. (9) Somogyi, P.; Jenner, A. J.; Brierley, I.; Inglis, S. C. Mol. Cell. Biol. 1993, 13, 6931−6940. (10) Plant, E. P.; Dinman, J. D. Nucleic Acids Res. 2005, 33, 1825− 1833. (11) Collie, G. W.; Parkinson, G. N. Chem. Soc. Rev. 2011, 40, 5867− 5892. (12) Williamson, J. R. Annu. Rev. Biophys. Biomol. Struct. 1994, 23, 703−730. (13) Miyoshi, D.; Fujimoto, T.; Sugimoto, N. Top. Curr. Chem. 2012, 330, 87−110. Miyoshi, D.; Karimata, H.; Sugimoto, N. J. Am. Chem. Soc. 2006, 128, 7957−7963. (14) Bugaut, A.; Balasubramanian, S. Nucleic Acids Res. 2012, 40, 4727−4741. Gomez, D.; Guedin, A.; Mergny, J. L.; Salles, B.; Riou, J. F.; Teulade-Fichou, M. P.; Calsou, P. Nucleic Acids Res. 2010, 38, 7187−7198. Halder, K.; Wieland, M.; Hartig, J. S. Nucleic Acids Res. 2009, 37, 6811−6817. Kumari, S.; Bugaut, A.; Huppert, J. L.; Balasubramanian, S. Nat. Chem. Biol. 2007, 3, 218−221. (15) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Angew. Chem., Int. Ed. 2013, 52, 5522−5526. (16) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Methods 2013, 73, 73− 78. (17) Baril, M.; Dulude, D.; Gendron, K.; Lemay, G.; Brakier-Gingras, L. RNA 2003, 9, 1246−1253. Reil, H.; Kollmus, H.; Weidle, U. H.; Hauser, H. J. Virol. 1993, 67, 5579−5584. (18) Brierley, I.; Jenner, A. J.; Inglis, S. C. J. Mol. Biol. 1992, 227, 463−479. (19) Yu, C. H.; Noteborn, M. H.; Pleij, C. W.; Olsthoorn, R. C. Nucleic Acids Res. 2011, 39, 8952−8959. (20) Brierley, I. J. Gen. Virol. 1995, 76, 1885−1892. (21) Corish, P.; Tyler-Smith, C. Protein Eng. 1999, 12, 1035−1040. (22) Li, X.; Zhao, X.; Fang, Y.; Jiang, X.; Duong, T.; Fan, C.; Huang, C. C.; Kain, S. R. J. Biol. Chem. 1998, 273, 34970−34975. (23) Arthanari, H.; Basu, S.; Kawano, T. L.; Bolton, P. H. Nucleic Acids Res. 1998, 26, 3724−3728. (24) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Nucleic Acids Res. 2013, 41, 6222−6231. (25) Brierley, I.; Digard, P.; Inglis, S. C. Cell 1989, 57, 537−547. (26) De Cian, A.; Lacroix, L.; Douarre, C.; Temime-Smaali, N.; Trentesaux, C.; Riou, J. F.; Mergny, J. L. Biochimie 2008, 90, 131−155. (27) Luedtke, N. W. Chimia 2008, 63, 134−139. Yaku, H.; Murashima, T.; Miyoshi, D.; Sugimoto, N. Molecules 2012, 17, 10586−10613. (28) Gu, H. P.; Lin, S.; Xu, M.; Yu, H. Y.; Du, X. J.; Zhang, Y. Y.; Yuan, G.; Gao, W. Endocrinology 2012, 153, 3692−3700. (29) Lin, S.; Gu, H.; Xu, M.; Cui, X.; Zhang, Y.; Gao, W.; Yuan, G. PLoS ONE 2012, 7, e31201. (30) Bazzicalupi, C.; Ferraroni, M.; Bilia, A. R.; Scheggi, F.; Gratteri, P. Nucleic Acids Res. 2013, 41, 632−638. (31) Chen, G.; Chang, K. Y.; Chou, M. Y.; Bustamante, C.; Tinoco, I., Jr. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 12706−12711. (32) Azzalin, C. M.; Reichenbach, P.; Khoriauli, L.; Giulotto, E.; Lingner, J. Science 2007, 318, 798−801. (33) Horsburgh, B. C.; Kollmus, H.; Hauser, H.; Coen, D. M. Cell 1996, 86, 949−959. (34) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Anal. Chem. 2011, 84, 857−861.
ASSOCIATED CONTENT
S Supporting Information *
Thermodynamic stabilities of RNA G-quadruplex, fluorescence signal of cell lysate on SDS-PAGE, CD spectra of wild-type oligonucleotide, and ApFSEs of various QFP sequences in MCF7 cells. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
* E-mail: [email protected]. Fax: (+) 81-78-303-1495. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported in part by Nagase Science and Technology Foundation, by the Grant-in-Aid for Scientific Research, MEXT-Supported Program for the Strategic Research Foundation at Private Universities (2009−2014), Japan, and by the Hirao Taro Foundation of the Konan University Association for Academic Research.
■
REFERENCES
(1) Batey, R. T.; Rambo, R. P.; Doudna, J. A. Angew. Chem., Int. Ed. 1999, 38, 2326−2343. (2) Serganov, A.; Patel, D. J. Curr. Opin. Struct. Biol. 2012, 22, 279− 286. Stoneley, M.; Willis, A. E. Oncogene 2004, 23, 3200−3207. (3) Baranov, P. V.; Gesteland, R. F.; Atkins, J. F. Gene 2002, 286, 187−201. Dinman, J. D. Wiley Interdiscip. Rev. RNA 2012, 3, 661−673. Firth, A. E.; Brierley, I. J. Gen. Virol. 2012, 93, 1385−1409. (4) Giedroc, D. P.; Cornish, P. V. Virus Res. 2009, 139, 193−208. (5) Giedroc, D. P.; Theimer, C. A.; Nixon, P. L. J. Mol. Biol. 2000, 298, 167−185. Marcheschi, R. J.; Staple, D. W.; Butcher, S. E. J. Mol. Biol. 2007, 373, 652−663. (6) Leger, M.; Sidani, S.; Brakier-Gingras, L. RNA 2004, 10, 1225− 1235. (7) Kontos, H.; Napthine, S.; Brierley, I. Mol. Cell. Biol. 2001, 21, 8657−8670. 11439
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Unusual −1 Ribosomal Frameshift Caused by Stable RNA G‑Quadruplex in Open Reading Frame Tamaki Endoh† and Naoki Sugimoto*,†,‡ †
Frontier Institute for Biomolecular Engineering Research (FIBER), Konan University, 7-1-20 Minatojima-minamimachi, Kobe 650-0047, Japan ‡ Faculty of Frontiers of Innovative Research in Science and Technology (FIRST), Konan University, 7-1-20 Minatojima-minamimachi, Kobe 650-0047, Japan S Supporting Information *
ABSTRACT: Tertiary structures formed by mRNAs impact the efficiency of the translation reaction. Ribosomal frameshift is a well-characterized recoding process that occurs during translation elongation. Pseudoknot and stem-loop structures may stimulate frameshifting by causing a translational halt at a slippery sequence. In this study, we evaluated the efficiency of an unusual −1 frameshift caused by a noncanonical RNA Gquadruplex structure in mammalian cells. The reporter gene construct consisting of a fluorescent protein and Luciferase enabled evaluation of apparent and absolute values of the −1 frameshift efficiency and revealed significant increase of the efficiency by G-quadrupex forming potential sequence. In addition, berberine, a small molecule that binds to and stabilizes G-quadruplex structures, further increased the frameshift efficiency. These results indicate that the stable G-quadruplex structure stimulates the unusual −1 frameshift and has a potential to regulate the frameshift with its ligand.
T
of the ribosome at a slippery sequence and the rearrangement of codon−anticodon base pairs between mRNA and tRNAs in the A-site and the P-site.7−9 It has been suggested that the RNA structure functions as a roadblock, halting the ribosome at the slippery sequence and providing tension on the mRNA−tRNA interaction to cause the rearrangement.4,8 As only certain sequences forming pseudoknots or stem-loop structures stimulate frameshifting, the properties of the structures that induce frameshifting are not well understood.7,10 G-quadruplexes are noncanonical structures formed by guanine-rich sequences of single-stranded DNA or RNA.11,12 G-quadruplexes have noteworthy stability under physiological conditions and structural properties different from those of canonical duplexes. Several G-quartets formed by four guanine bases through Hoogsteen base pairs stack in parallel or antiparallel fashion to form the G-quadruplex structures. Both DNA and RNA G-quadruplexes are stabilized under cell-like molecular crowding conditions.13 Recent studies focusing on oncogene mRNAs suggested that G-quadruplex formation in 5′ UTRs suppresses the small ribosomal subunit during the scanning phase.14 In addition to the G-quadruplexes in UTRs, we recently showed that the ribosome in the elongation phase is also suppressed by G-quadruplexes in the ORFs of mRNAs.15
ranslation is one of the most fundamental processes in the central dogma that explains the flow of genetically encoded information of nucleotide sequences into functionally active amino acid sequences. In cells, mRNAs are produced as single-stranded polynucleotides and form complicated secondary and tertiary structures stabilized by interactions such as Watson−Crick and Hoogsteen base pairs and by interactions with the 2′ hydroxyl groups in their ribose sugars that can behave as both hydrogen bond donors and acceptors.1 Several mechanisms involving specific structures of the mRNA, such as internal ribosome entry site and riboswitches, are known to control initiation of translation.2 These structural elements are normally located in the untranslated regions (UTR) of mRNAs. In contrast to the functional mRNA structures in UTRs, some RNA structures in mRNA open reading frames (ORFs) exert elaborate effects on translation. Ribosomal frameshifting is one of the recoding mechanisms known to be endogenously programed in some specific genes in all living organisms, including virus, prokaryote, and eukaryote.3 During frameshifting, the ribosome shifts its reading frame on the mRNA, and this results in expression of more than one protein product from a single transcript. Especially in the case of −1 frameshift, in which the ribosome is forced to shift the reading frame one nucleotide backward, structures in the mRNA, like a pseudoknot or stem-loop, are known to stimulate the frameshift in combination with an upstream slippery sequence.4−6 The key features of this type of frameshift are the translational halt © 2013 American Chemical Society
Received: August 7, 2013 Accepted: November 6, 2013 Published: November 6, 2013 11435
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
In vitro study using reporter mRNAs forming the G-quadruplex at various positions revealed that the G-quadruplex suppressed the translation elongation by halting the ribosome before the G-quadruplex structure with a 5−7 nucleotide spacer sequence.16 The results suggested that the G-quadruplex blocked the further entry of the mRNA into the ribosome. Thus, we hypothesized that the G-quadruplex located at an appropriate distance downstream of the slippery sequence may stimulate frameshifting if a translational halt could be a cause of the −1 frameshift. In this study, we investigated the effect of the G-quadruplex structure on the ORF of mRNAs on the −1 ribosomal frameshift in cultured cells. In previous studies, a pair of reporter mRNAs, one expressed a reporter enzyme through in-frame translation and the other expressed the active reporter enzyme only if a frameshift occurs, were utilized to evaluate the frameshift efficiency inside cells.6,17 The relative frameshift efficiencies caused by nucleotide sequences of interest were evaluated by comparing the relative catalytic activities of the reporter enzyme. To determine the absolute frameshift efficiency (AbFSE), radioisotope labeling of translated products that is confined to in vitro experiments is required.18,19 To investigate the AbFSE inside cells in this study, reporter mRNAs encoding a fluorescent protein (AcGFP) followed by a slippery sequence (U-UUA-AAC), which is one of the most effective slippery sequences in eukaryotic cells,18 a spacer sequence, a quadruplex forming potential (QFP) sequence derived from the E. coli eutE gene,15 and the coding sequence for Renilla luciferase (RL) in the −1 reading frame were designed (Figure 1). AcGFP and AcGFPRL were expressed from the reporter mRNA through in-frame translation and −1 ribosomal frameshift at the slippery sequence, respectively. This reporter construct enables determination of AbFSE by dividing the fluorescence signal of AcGFP-RL by the total fluorescence signal expressed from the reporter mRNA. In addition, measurement of the luminescence signal of RL enables evaluation of the apparent frameshift efficiency (ApFSE) by dividing the luminescence signal by the total fluorescence signal of AcGFP (Figure 1). To evaluate effects of the position of the G-quadruplex on the frameshift efficiency, the QFP sequence derived from the eutE gene was inserted after the slippery sequence with spacer sequences varying from 5 to 8 nucleotides according to the spacer length frequently observed in natural frameshifting mRNAs.20 Amino acid sequences fused to the C-terminus of AcGFP were designed to be identical by locating the UAG or UAA codon within the spacer sequence as differences in the fused amino acids might affect the half-life of AcGFP inside cells and obscure the frameshift signal.21,22 Differences in amino acids sequences at linker region between AcGFP and RL are inevitable depending on the spacer length and might affect specific activity and intracellular stability of AcGFP-RL that result in unexpected fluctuation of RL activity. Thus, mutant QFP sequences, in which the guanine residues in the wild-type QFP sequences were replaced by adenines, were designed as controls. In order to permit direct comparison of frameshift efficiencies between the wild-type and the mutant QFP sequences, the mutant sequences were designed to express AcGFP-RLs of the same amino acid sequences as those of the wild-type (Figure 1). The reporter mRNAs were transcribed in vitro, and formation of the RNA G-quadruplex was evaluated by monitoring fluorescence of N-methyl mesoporphyrin (NMM), which specifically binds to G-quadruplex structures
Figure 1. Reporter mRNAs used to analyze frameshift efficiency inside cells. Sequences of spacer nucleotides and wild-type or mutant quadruplex forming potential (QFP) sequence regions are indicated. The QFP regions of the wild-type sequences are underlined. Amino acids sequences produced with and without −1 frameshift are indicated below the nucleotide sequences. AcGFP and AcGFP fused to Renilla Luciferase (AcGFP-RL) are produced through in-frame translation without frameshift and with −1 frameshift, respectively. Absolute frameshift efficiency (AbFSE) and apparent frameshift efficiency (ApFSE) are calculated from output signals of the reporter proteins.
resulting in an increase in fluorescence (Figure 2a).23,24 All reporter mRNAs containing the wild-type QFP sequences had a higher fluorescence signal in the presence of NMM than did mRNAs containing mutant QFP sequences. The slight signals observed with mutant mRNAs comparing to the control without mRNA were likely due to nonspecific interaction of NMM and mRNA, as previously reported.24 These results indicated that the wild-type QFP sequence formed parallel Gquadruplex as previously shown by circular dichroism analysis using short RNA oligonucleotides of this sequence.15 Plasmid vectors encoding the reporter mRNAs were transfected into MCF7 human breast carcinoma cells. Fluorescence and luminescence signals of AcGFP and AcGFP-RL, respectively, in cell lysates were measured using a multimode microwell plate reader (Varioskan Flash; Thermo Scientific) to evaluate the ApFSEs (Figure 2b). For all the reporter mRNAs except that containing a spacer length of 8 nucleotides, significantly higher ApFSEs were observed with the mRNAs containing the wild-type QFP sequence than those containing the mutant QFP sequence. Since the G-quadruplex halted the ribosome before the G-quadruplex structure with a 5−7 nucleotide spacer sequence in vitro,16 it was considered 11436
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
Figure 2. Ribosomal frameshift stimulated by G-quadruplex formation in cells. (a) Fluorescence intensity of 5 μM N-methyl mesoporphyrin (NMM) mixed with 1 μM reporter mRNAs containing wild-type (blue) or mutant (red) QFP sequences or without mRNA (green). Fluorescence intensities were measured in buffer containing 30 mM HEPES-KOH (pH 6.8), 100 mM KCl, and 0.1% DMSO at 37 °C using 400 nm excitation and 615 nm emission. Error bars indicate SD from triplicate samples. (b) Apparent frameshift efficiencies in MCF7 cells calculated by the equation in Figure 1. MCF7 cells transfected with the plasmid vectors were lysed 24 h after transfection. Fluorescence signal was measured by using 470 nm excitation and 510 nm emission, and luminescence signal was measured after addition of coelenterazine. Error bars indicate SD from triplicate samples. (c) Predicted mechanism of the −1 ribosomal frameshift caused by the RNA G-quadruplex. (d) Fluorescence signals of cell lysate separated on 12% SDS-PAGE were imaged by 473 nm excitation and 510 nm emission. Lysate from nontransfected cells served as a control. Absolute frameshift efficiencies were calculated from triplicate samples by the equation in Figure 1
sequence was 2.4 times larger than that of mutant, which is similar to the relative ApFSE (Figure 2b). These results indicated that the reporter gene construct enabled a straightforward comparison of the ApFSEs using fluorescence and luminescence signals and a quantification of the AbFSEs by fluorescence imaging after separation of cell lysates by SDSPAGE. The AbFSE value of the wild-type QFP sequence (0.069%) was considerably lower than frameshift efficiencies reported by other groups using Luciferases as reporter proteins.6,19 GFP-type fluorescent protein is known to show relatively high stability, in which the intracellular half-life is estimated to be around 26 h.21 This stability might have the effect of lowering the observed frameshift efficiencies because of an accumulation of AcGFP produced without the frameshift. The ApFSE values of both wild-type and mutant QFP sequences 48 h after transfection decreased compared to those of 24 h likely due to the accumulation of AcGFP (data not shown). Since the single-stranded telomere sequence at the ends of chromosomes has the potential to form G-quadruplex structures, and these may contribute to chromosome maintenance and protection, targeting of G-quadruplex structures has been considered as a promising therapeutic strategy.11,26 Small-molecular ligands that specifically bind to and stabilize or destabilize the G-quadruplex have been reported from various research groups.26,27 These molecules possibly function as a trigger of the ribosomal frameshift that enables gene switching mediated by the ribosomal frameshift. Here, we used berberine, which is derived from natural alkaloids and alters gene expression through stabilization of
that the G-quadruplex temporary halted the ribosome at the slippery sequence and provided tension to stimulate the −1 ribosomal frameshift, as a similar mechanism as the pseudoknot structure (Figure 2c). The wild-type mRNA with the 6nucleotide spacer between the slippery sequence and the QFP sequence showed more than 2 times more ApFSE than that of the mutant mRNA. This spacer length is the same as that reported to be optimal for the frameshift caused by a pseudoknot structure.7,25 Almost the same ApFSEs between wild-type and mutant reporter mRNAs containing an 8nucleotide spacer suggested that the ribosome had passed through the slippery sequence when the translation elongation was halted by the G-quadruplex. To quantitatively analyze AbFSE, cells were transfected with the vectors containing wild-type and mutant QFP sequences with 6-nucleotide spacer sequences. Cells were lysed, and samples were separated on 12% SDS-PAGE at room temperature; AcGFP preserves its fluorescence during SDS-PAGE unless it has been denatured before the electrophoresis.24 Gels were imaged using a fluorescence image scanner (FLA5100; GE Healthcare). Although most of the fluorescence signal was that of the product without frameshift, a signal due to AcGFP-RL expressed through the −1 frameshift was observed in the lysate from cells transfected with a vector containing the wild-type QFP sequence, whereas the signal of AcGFP-RL present in lysates of cells transfected with the mutant was obscure (Figure 2d). The AbFSEs were calculated from the fluorescence intensities of AcGFP and AcGFP-RL in cell lysates obtained from three individual cultures (Figure 2d and Figure S1). The AbFSE of the cells transfected with the wild-type QFP 11437
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
DNA G-quadruplexes in cells,28,29 to further stimulate the −1 ribosomal frameshift caused by the wild-type QFP sequence. Although the interaction between the berberine and DNA Gquadruplexes has been well characterized, the effect of the berberine on the stability of the RNA G-quadruplexes is still unclear. Thermal stability of a wild-type oligonucleotide, which contains the guanine-rich region of the wild-type reporter mRNA (Figure 3a), was evaluated by ultraviolet (UV) absorbance at 295 nm in the presence of various concentrations of berberine (Figure 3b). The melting temperature (Tm) of the wild-type oligonucleotide increased depending on the concentration of berberine (Table S1). An RNA oligonucleotide consisting of the mutant QFP sequence did not show a melting transition characteristic of G-quadruplex in either the presence or absence of berberine (data not shown). Circular dichroism (CD) of the wild-type oligonucleotide showed a two-state transition characteristic of a parallel G-quadruplex; the CD spectra had an isosbestic point near 250 nm and linear correlations between signal intensities at 265 and 242 nm during the transition (Figure S2). We assumed that the berberine bound the parallel RNA G-quadruplex through endstacking, as proposed for the binding with DNA Gquadruplex.29,30 Thermodynamic stability of the wild-type oligonucleotide at 37 °C (−ΔG°37) in the presence of various concentrations of berberine were calculated from the UV melting profiles by the van’t Hoff equation (Table S1). The ApFSEs in MCF-7 cells were evaluated in the presence of berberine at different concentrations (Figure 3c). The ApFSE of the mRNA containing the wild-type QFP sequence was increased with the berberine concentration and showed a 4-fold higher value in the presence of 80 μM berberine than in its absence. Although the ApFSE of the mutant reporter was slightly increased in the presence of berberine likely due to a side effect of berberine unrelated to binding to G-quadruplex, the ApFSE values were less than that of the wild-type reporter mRNA in the absence of berberine. Figure 3d shows plots of the ApFSEs of the wild-type reporter mRNA versus −ΔG°37 of the G-quadruplex formed by the wild-type oligonucleotide in the presence of various concentrations of berberine. The ApFSE exponentially depended on the stability of the Gquadruplex. This trend has also been reported in a −1 ribosomal frameshift caused by a pseudoknot structure that the frameshift efficiency exponentially depended on the mechanical strength of the structure.31 Thus, it was considered that the stabilization of the G-quadruplex mediated by the berberine binding was a main factor to further stimulate the ribosomal frameshift caused by the G-quadruplex, although the in vitro and intracellular binding affinity of berberine and stabilization effect would be different. We prepared additional reporter vectors containing wild-type and mutant QFP sequences derived from the E. coli rnlA gene and the human telomeric-repeat-containing RNA (TERRA).15,32 We determined ApFSEs in the presence and absence of berberine. Although the differences in ApFSEs between wild-type and mutant QFP sequences of the rnlA gene and TERRA in the absence of berberine were smaller than that between wild-type and mutant of eutE gene, berberine increased ApFSEs of wild-type QFP sequences significantly (Figure S3). Thus, it was suggested that RNA G-quadruplexes on the ORF of mRNAs potentially stimulate the −1 ribosomal frameshift inside cells, when it was located at appropriate distance downstream of the slippery sequence.
Figure 3. Effect of stabilization of G-quadruplex by berberine on frameshift efficiency. (a) Sequence of wild-type oligonucleotide derived from reporter mRNA containing wild-type QFP sequence of eutE gene. (b) Melting transition of wild-type oligonucleotide (5 μM) in the absence (purple) and presence of 5 (dark blue), 10 (light blue), 20 (dark green), 40 (light green), 60 (orange), and 80 (red) μM berberine in buffer containing 50 mM MES-LiOH (pH 7), 0.1 mM KCl, and 0.2% DMSO. Absorbance at 295 nm was measured at a rate of 0.2 °C min−1 and normalized. (c) ApFSEs in MCF7 cells transfected with plasmid vector containing wild-type (blue) or mutant (red) QFP sequence with 6-nucleotide spacers in the presence of various concentrations of berberine. Cellular medium was replaced by berberine-containing medium at 6 h after transfection, and cells were incubated at 37 °C for 18 h. Error bars indicate SD from five samples. (d) Plots of ApFSEs in MCF7 cells transfected with the plasmid vector containing wild-type QFP sequence vs −ΔG°37 of the G-quadruplex formed by the wild-type oligonucleotide in the presence of various concentrations of berberine.
Ribosomal frameshift have been shown previously to be stimulated by certain pseudoknot and stem-loop structures. One unusual +1 frameshift, which shifts the reading frame one nucleotide forward, has been reported to be caused within a guanine-rich sequence derived from herpes simplex virus.33 This report suggested that an unusual mRNA structure formed by the guanine-rich sequence contributed to the frameshift. In 11438
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
Analytical Chemistry
Article
this study, we evaluated stimulation of a −1 frameshift by Gquadruplex structures at different distances from the slippery sequence. The frameshift efficiency was quantitatively evaluated as both apparent and absolute values in cells. Significantly more of the −1 frameshift product was observed from reporters containing the wild-type QFP sequence compared to that from a reporter with a mutant QFP sequence when the spacer was 5 to 7 nucleotides. The spacer lengths corresponded to those between the position of translational halt and the QFP sequence that were revealed in our previous study using in vitro synchronized translation.16,34 Thus, the G-quadruplex stimulated the −1 frameshift by cooperative function with the slippery sequence, which would be located at the A-site of the ribosome when the translation elongation was halted by the Gquadruplex. In addition, a G-quadruplex-stabilizing ligand, berberine, further stimulated the frameshift. These results strongly suggest that a temporary halt in translation elongation due to stable RNA structures can partially stimulate the −1 ribosomal frameshift. However, since the AbFSEs were relatively small compared to the previously reported −1 ribosomal frameshift caused by a pseudoknot, not only the translational halt but also other factors such as tertiary structure and torsional resistant properties may be required for efficient stimulation of the frameshift.10
■
(8) Namy, O.; Moran, S. J.; Stuart, D. I.; Gilbert, R. J.; Brierley, I. Nature 2006, 441, 244−247. (9) Somogyi, P.; Jenner, A. J.; Brierley, I.; Inglis, S. C. Mol. Cell. Biol. 1993, 13, 6931−6940. (10) Plant, E. P.; Dinman, J. D. Nucleic Acids Res. 2005, 33, 1825− 1833. (11) Collie, G. W.; Parkinson, G. N. Chem. Soc. Rev. 2011, 40, 5867− 5892. (12) Williamson, J. R. Annu. Rev. Biophys. Biomol. Struct. 1994, 23, 703−730. (13) Miyoshi, D.; Fujimoto, T.; Sugimoto, N. Top. Curr. Chem. 2012, 330, 87−110. Miyoshi, D.; Karimata, H.; Sugimoto, N. J. Am. Chem. Soc. 2006, 128, 7957−7963. (14) Bugaut, A.; Balasubramanian, S. Nucleic Acids Res. 2012, 40, 4727−4741. Gomez, D.; Guedin, A.; Mergny, J. L.; Salles, B.; Riou, J. F.; Teulade-Fichou, M. P.; Calsou, P. Nucleic Acids Res. 2010, 38, 7187−7198. Halder, K.; Wieland, M.; Hartig, J. S. Nucleic Acids Res. 2009, 37, 6811−6817. Kumari, S.; Bugaut, A.; Huppert, J. L.; Balasubramanian, S. Nat. Chem. Biol. 2007, 3, 218−221. (15) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Angew. Chem., Int. Ed. 2013, 52, 5522−5526. (16) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Methods 2013, 73, 73− 78. (17) Baril, M.; Dulude, D.; Gendron, K.; Lemay, G.; Brakier-Gingras, L. RNA 2003, 9, 1246−1253. Reil, H.; Kollmus, H.; Weidle, U. H.; Hauser, H. J. Virol. 1993, 67, 5579−5584. (18) Brierley, I.; Jenner, A. J.; Inglis, S. C. J. Mol. Biol. 1992, 227, 463−479. (19) Yu, C. H.; Noteborn, M. H.; Pleij, C. W.; Olsthoorn, R. C. Nucleic Acids Res. 2011, 39, 8952−8959. (20) Brierley, I. J. Gen. Virol. 1995, 76, 1885−1892. (21) Corish, P.; Tyler-Smith, C. Protein Eng. 1999, 12, 1035−1040. (22) Li, X.; Zhao, X.; Fang, Y.; Jiang, X.; Duong, T.; Fan, C.; Huang, C. C.; Kain, S. R. J. Biol. Chem. 1998, 273, 34970−34975. (23) Arthanari, H.; Basu, S.; Kawano, T. L.; Bolton, P. H. Nucleic Acids Res. 1998, 26, 3724−3728. (24) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Nucleic Acids Res. 2013, 41, 6222−6231. (25) Brierley, I.; Digard, P.; Inglis, S. C. Cell 1989, 57, 537−547. (26) De Cian, A.; Lacroix, L.; Douarre, C.; Temime-Smaali, N.; Trentesaux, C.; Riou, J. F.; Mergny, J. L. Biochimie 2008, 90, 131−155. (27) Luedtke, N. W. Chimia 2008, 63, 134−139. Yaku, H.; Murashima, T.; Miyoshi, D.; Sugimoto, N. Molecules 2012, 17, 10586−10613. (28) Gu, H. P.; Lin, S.; Xu, M.; Yu, H. Y.; Du, X. J.; Zhang, Y. Y.; Yuan, G.; Gao, W. Endocrinology 2012, 153, 3692−3700. (29) Lin, S.; Gu, H.; Xu, M.; Cui, X.; Zhang, Y.; Gao, W.; Yuan, G. PLoS ONE 2012, 7, e31201. (30) Bazzicalupi, C.; Ferraroni, M.; Bilia, A. R.; Scheggi, F.; Gratteri, P. Nucleic Acids Res. 2013, 41, 632−638. (31) Chen, G.; Chang, K. Y.; Chou, M. Y.; Bustamante, C.; Tinoco, I., Jr. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 12706−12711. (32) Azzalin, C. M.; Reichenbach, P.; Khoriauli, L.; Giulotto, E.; Lingner, J. Science 2007, 318, 798−801. (33) Horsburgh, B. C.; Kollmus, H.; Hauser, H.; Coen, D. M. Cell 1996, 86, 949−959. (34) Endoh, T.; Kawasaki, Y.; Sugimoto, N. Anal. Chem. 2011, 84, 857−861.
ASSOCIATED CONTENT
S Supporting Information *
Thermodynamic stabilities of RNA G-quadruplex, fluorescence signal of cell lysate on SDS-PAGE, CD spectra of wild-type oligonucleotide, and ApFSEs of various QFP sequences in MCF7 cells. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
* E-mail: [email protected]. Fax: (+) 81-78-303-1495. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported in part by Nagase Science and Technology Foundation, by the Grant-in-Aid for Scientific Research, MEXT-Supported Program for the Strategic Research Foundation at Private Universities (2009−2014), Japan, and by the Hirao Taro Foundation of the Konan University Association for Academic Research.
■
REFERENCES
(1) Batey, R. T.; Rambo, R. P.; Doudna, J. A. Angew. Chem., Int. Ed. 1999, 38, 2326−2343. (2) Serganov, A.; Patel, D. J. Curr. Opin. Struct. Biol. 2012, 22, 279− 286. Stoneley, M.; Willis, A. E. Oncogene 2004, 23, 3200−3207. (3) Baranov, P. V.; Gesteland, R. F.; Atkins, J. F. Gene 2002, 286, 187−201. Dinman, J. D. Wiley Interdiscip. Rev. RNA 2012, 3, 661−673. Firth, A. E.; Brierley, I. J. Gen. Virol. 2012, 93, 1385−1409. (4) Giedroc, D. P.; Cornish, P. V. Virus Res. 2009, 139, 193−208. (5) Giedroc, D. P.; Theimer, C. A.; Nixon, P. L. J. Mol. Biol. 2000, 298, 167−185. Marcheschi, R. J.; Staple, D. W.; Butcher, S. E. J. Mol. Biol. 2007, 373, 652−663. (6) Leger, M.; Sidani, S.; Brakier-Gingras, L. RNA 2004, 10, 1225− 1235. (7) Kontos, H.; Napthine, S.; Brierley, I. Mol. Cell. Biol. 2001, 21, 8657−8670. 11439
dx.doi.org/10.1021/ac402497x | Anal. Chem. 2013, 85, 11435−11439
