THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 282, NO. 21, pp. 15645–15651, May 25, 2007 © 2007 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.
The Interaction between Cap-binding Complex and RNA Export Factor Is Required for Intronless mRNA Export*□ S
Received for publication, January 23, 2007, and in revised form, March 15, 2007 Published, JBC Papers in Press, March 15, 2007, DOI 10.1074/jbc.M700629200
Takayuki Nojima‡§, Tetsuro Hirose¶储, Hiroshi Kimura**, and Masatoshi Hagiwara‡§1 From the ‡Laboratory of Gene Expression, School of Biomedical Science, the §Department of Functional Genomics, Medical Research Institute, Tokyo Medical and Dental University, Yushima 1-5-45, Bunkyo-ku, Tokyo 113-8510, the ¶Biological Information Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Aomi 2-42, Koto-Ku, Tokyo 135-0064, 储 PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama, and the **Nuclear Function and Dynamics Unit, HMRO, Graduate School of Medicine, Kyoto University, Yoshida Konoe-cho, Sakyo-ku, Kyoto 606-8501, Japan
In eukaryotic cells, pre-mRNAs are primarily transcribed by RNA polymerase II (RNAPII); this process is followed by complex RNA processing steps that include capping, splicing, and polyadenylation to produce mature mRNAs. Recent studies have shown that individual events occurring during eukaryotic gene expression are coupled together under more elaborate regulatory controls than previously imagined (1–5). Coupling stimulates the rate and specificity of enzymatic reactions by tethering mechanisms to each other and to their substrates. Following processing, mRNA is exported as a large mRNA-protein complex (mRNP)2 through the nuclear pore to the cytoplasm for subsequent translation.
* This work was supported by grants-in-aid (to M. H.) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan, the National Institute of Biomedical Innovation (NIBI), and the 21st COE Program for Frontier Research on Molecular Destruction and Reconstruction of Tooth and Bone, represented by Masaki Noda. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. □ S The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1 and S2. 1 To whom correspondence should be addressed. Tel.: 81-3-5803-5836; Fax: 81-3-5803-5853; E-mail:
[email protected]. 2 The abbreviations used are: mRNP, mRNA-protein complex; CBC, cap-binding complex; CBP20, cap-binding protein 20; EJC, exon-junction complex; AdMLP, adenovirus major late promoter; REF, RNA and export factor-binding proteins; RNAPII, RNA polymerase II; TREX, transcription/export (complex); nt, nucleotide(s); FITC, fluorescein isothiocyanate; RNP, ribonucleoprotein; WT, wild type.
MAY 25, 2007 • VOLUME 282 • NUMBER 21
The transport of mRNA from the nucleus to the cytoplasm is linked to pre-mRNA splicing, especially in metazoans (6). Exon junction complexes (EJCs), which are deposited on mRNAs at specific sites relative to the exon junction as a consequence of splicing, form the basis of this connection (7, 8). The EJC consists of four core proteins, eIF4A3 (9 –11), Y14 (7), Magoh (12, 13), and MLN51 (14), plus other auxiliary proteins including REF (7), UAP56 (15), RNPS1 (7), SRm160 (7), Pinin (16, 17), Acinus L (18), SAP18 (18), and hUpf3 (8, 19, 20). The recruitment of REF during mRNA biogenesis is thought to be responsible for the increased export of spliced mRNA (21). A DEAD box RNA helicase, UAP56, is required for the recruitment of REF to mRNA (15, 22). Subsequently, UAP56 is displaced from REF by the mRNA export factor TAP (23). TAP forms a heterodimer with p15 that then directly interacts with the nuclear pore to facilitate mRNP transport into the cytoplasm (24). Although the above model explains the apparent link between splicing and RNA export in metazoans, the question of how intronless mRNAs, which lack EJC deposition, are exported to the cytoplasm naturally arises. Some intronless transcripts (e.g. histone H2A) have been reported to contain specific sequences that recruit export factors independently of splicing (25). SRp20 and 9G8, which belong to members of the evolutionarily conserved SR (serine/arginine-rich) protein family, specifically bind to a sequence in intronless mRNA and greatly facilitate the export of mRNA by recruiting TAP (26). However, the intronless mRNAs coding Ftz, dihydrofolate reductase, and -globin, which lack such cis-acting sequences, can be effectively exported regardless of whether splicing has occurred (27–29). The injection experiments of ␣-REF antibody into the nuclei of Xenopus oocytes indicated that REF stimulates directly the export of these intronless mRNAs (27). Mass spectrometry and Western blotting of purified spliceosomes revealed that REF is a component of H complex (30), suggesting that REF can associate with mRNAs in a splicing-independent manner. In situ analysis of green fluorescent protein-tagged REF showed its accumulation at sites of transcription (31), suggesting that REF binds to mRNA co-transcriptionally. To clarify the recruiting mechanism of RNPs on mRNAs, we primarily developed a coupled in vitro transcription-splicing system. This in vitro system led us to discover a novel mechanism through which REF can associate with mRNA in a manner that is independent of splicing, instead of via the cap structure JOURNAL OF BIOLOGICAL CHEMISTRY
15645
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
RNA export factor (REF) is a component of the exon junction complex (EJC) that is deposited on mRNA in a splicing-dependent manner, and targets spliced mRNA for export. In this study, analysis of the RNA-binding protein complexes revealed that REF associates with -globin mRNA at the region other than the EJC deposition site. Comparison between RNA polymerase II and T7 transcription and further analysis showed that the deposition of REF apart from the EJC is dependent on the 5ⴕ cap structure, but not splicing. Excess amounts of m7GpppG cap analog reduced REF binding to intronless mRNA, and a co-immunoprecipitation experiment revealed that REF interacts with the cap-binding protein CBP20. The export of Cy3-labeled intronless -globin mRNA from nuclei of HeLa cells was enhanced by co-injection of CBP20 and REF. Thus, REF recruited by CBP20 may play a stimulatory role to export the capped intronless mRNAs.
Interaction between Cap-binding Complex and RNA Export Factor that is created during RNAPII transcription. The cap structure at the 5⬘ end of mRNA is associated with the nuclear cap-binding complex (CBC), consisting of CBP20 and CBP80 (32). The microinjection of mRNAs into the living cell nuclei indicated that interaction between CBP20 and REF is required for the promotion of intronless mRNA export.
EXPERIMENTAL PROCEDURES
15646 JOURNAL OF BIOLOGICAL CHEMISTRY
RESULTS REF Binds to mRNA Independently of the EJC Deposition— To analyze transitions in mRNP composition during mRNA biogenesis in the nucleus, we developed a coupled in vitro system for examining transcription-splicing reactions. Psoralenmodified and biotinylated DNA were employed as transcription templates to capture RNP complexes containing nascent transcripts (Fig. 1A, see “Experimental Procedures ” for details). As the EJC proteins (REF, Y14, SRm160, UAP56, RNPS1, and Magoh) are accumulated at sites of transcription (31), we immunoprecipitated in vitro transcribed and spliced -globin mRNA using antibodies against the EJC components and examined whether RNAPII transcription enhances the EJC deposition or not. The ␣-REF antibody has been reported to favorably precipitate spliced mRNAs (21). Our immunoprecipitation experiments revealed that the ␣-REF antibody efficiently precipitated the spliced form of the RNAPII transcripts but not the spliced form of the T7 transcripts (Fig. 1B, lanes 5 and 6). On the other hand, the ␣-heterogeneous nuclear ribonucleoprotein A1 antibody comparably precipitated the unspliced and spliced RNAs as well as the transcripts of RNAPII and T7 (Fig. 1B, lanes 13 and 14). These data coincide with a VOLUME 282 • NUMBER 21 • MAY 25, 2007
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
Antibodies and Plasmids—The ␣-heterogeneous nuclear ribonucleoprotein A1 (4B10; Abcam), ␣-Y14 (4C4; Abcam), ␣-Aly/REF (11G5; Abcam), ␣-FLAG (M2; Sigma), and ␣-m3G/ m7G cap (H20; Synaptic Systems) antibodies were purchased. The ␣-UAP56 and ␣-CBP20 antibodies were kindly provided by M. Green and E. Izaurralde, respectively. The -globin ⌬6 and pAd-SX (Eco) DNA templates were kindly provided by A. Krainer and K. Mizumoto, respectively. The DNA template for the intronless mRNA was constructed using the QuikChange site-directed mutagenesis kit (Stratagene). In Vitro Splicing and Coupled Transcription/Splicing Assay— The in vitro splicing reactions were performed as described by Krainer et al. (33). For the coupled reaction, the immobilized and psoralen-modified PCR product (see below, 100 ng), containing an AdMLP promoter fused to two exons and a single intron of the -globin gene, was incubated in a 25-l reaction mixture containing 40 mM HEPES-KOH (pH 7.9), 0.5 mM dithiothreitol, 3 mM MgCl2, 30 mM KCl, 5 mM phosphocreatine, 0.5% polyvinyl alcohol, 200 M ATP, CTP, and GTP; 15 M UTP, 40 Ci of [␣-32P]UTP, and HeLa nuclear extract (90 g). The PCR product was amplified from the region 300 bp upstream of the AdMLP promoter to 143 nucleotides (nt) downstream of exon 2. The reaction was performed at 30 °C for the indicated times. The remaining RNAs were purified by protein removal and ethanol precipitation. The purified RNAs were analyzed using denaturing PAGE and imaged using a phosphorimage analyzer (FLA-3000G; FUJIFILM). Preparation of Immobilized Template DNA Modified by Psoralen—To pause the transcriptional machinery on the DNA template at a point containing a triplex targeting sequence (5⬘-AAAAGAAAAGGGGGG-3⬘) in -globin exon 2, the biotinylated PCR products were incubated with an excess (500-fold) of psoralen-modified oligonucleotide probes, P15 (5⬘-[PsoralenC2]TTTT[5Me-dC]TTTTGGGGGG-3⬘), as described by Wang and Rana (34). The biotinylated PCR products containing P15 were then mixed with streptavidin beads (M280, Dynal) and incubated in a mixture containing 10 mM Tris-HCl (pH 7.5), 1 mM EDTA, and 1 M NaCl at room temperature for 2 h. After immobilization, the magnetic beads were washed with and suspended in BC100 (20 mM HEPES-KOH (pH 7.9), 0.5 mM EDTA, 100 mM KCl, 20% glycerol) containing 0.05% Nonidet P-40. RNP Release and RNA Immunoprecipitation—Deoxyoligos (final concentration, 3 M) were added to in vitro transcription/ splicing reactions and incubated at 30 °C for 15 min to enable site-specific RNA digestion by endogenous RNase H as previously described by Hirose et al. (35). Oligo R (5⬘-CACTCAGTGTGGCAA-3⬘) was used to release the nascent mRNA from RNAPII. The RNA fragments released from the beads were
immunoprecipitated with the indicated antibodies, and analyzed on 8% denaturing PAGE. Pulldown Assay Using m7GTP-Sepharose—7-Methyl-GTP (m7GTP)-Sepharose 4B (Amersham Biosciences, 8 nmol of ligand) was incubated with HeLa nuclear extract (⬃0.6 mg) in NET-2 buffer containing 3 mM MgCl2 at 4 °C for 3 h and then washed eight times with NET-2 buffer containing 3 mM MgCl2. The proteins attached to the washed Sepharose were then subjected to 12% SDS-PAGE followed by Western blotting. Co-immunoprecipitation Analysis—Immunoprecipitation was performed using ␣-FLAG M2-agarose (Sigma), as described in the legend to Fig. 3C. Whole cell extracts from HEK293 cells were prepared using a PARIS kit (Ambion). The extracts containing FLAG tag proteins were incubated with RNase A (5 g/ml) at 30 °C for 15 min prior to immunoprecipitation. For immunoblot analysis, polyclonal ␣-FLAG antibody (Sigma) was used as the primary antibody. Nuclear Microinjection Analysis—The microinjection was performed as described previously (36). Human -globin RNA, lacking an intron and containing 5⬘ m7GpppG cap and poly(A) sequence (25 nt), was transcribed by T7 RNA polymerase. The RNA was labeled with the TransIT Cy3 labeling kit (Mirus). The amount of Cy3 coupled with mRNA was measured at an excitation wavelength of 550 nm. The labeled RNA (0.8 M), purified FLAG-tagged protein (⬃1 M), and lysine-fixable FITC-conjugated 70-kDa dextran (1.5 mg/ml, Molecular Probe) were injected into HeLa cell nuclei, using FemtoJet and InjectMan NI2 (Eppendorf) at the conditions of injection pressure 75 hectopascal (hPa), compensation pressure 24 hPa, and injection duration 0.3 s. The injected cells were incubated at 37 °C under 5% CO2 for 10 h and analyzed under a confocal microscope (Olympus, FV1000; confocal aperture 679 m) with UPLSAPO ⫻40 NA:0.90 objective lens. The signals of FITC-dextran and Cy3-RNA were sequentially collected by excitation with 488 and 543 nm lasers, respectively.
Interaction between Cap-binding Complex and RNA Export Factor
recent report stating that the H complex containing heterogeneous nuclear ribonucleoprotein A1 was easily formed on the T7 transcripts (37). These immunoprecipitation results raise the possibility that RNAPII transcription stimulates the recruitment of REF to the spliced transcripts. To determine whether REF is recruited as an EJC component, we examined the EJC deposition on spliced mRNA originating from T7 and RNAPII transcripts. The EJC is specifically formed 20 –24 nt upstream of the exon-exon junction of spliced mRNAs (7). When a deoxy-oligo covering ⫺34 to ⫺19 was added after 90 min of incubation, during which time the in vitro transcriptionMAY 25, 2007 • VOLUME 282 • NUMBER 21
JOURNAL OF BIOLOGICAL CHEMISTRY
15647
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
FIGURE 1. Modified in vitro transcription/splicing coupling system. A, schematic representation of two constructs: unmodified and psoralenmodified DNA. When the unmodified DNA template is used, RNA polymerase II (RNAPII) reads through and dissociates with the nascent mRNA. If the psoralen-modified DNA template is used, transcription is paused at the psoralen site. Furthermore, the DNA template is immobilized by the streptavidin-biotin interaction at the 5⬘ end. The nascent RNA is released from RNAPII by the addition of deoxy-oligo (hereafter referred to as oligo R), accompanied by RNase H digestion. B, both the immobilized/psoralen T7--globin and the MLP--globin DNA templates were used to generate the RNA. Each RNA was transcribed by the supplied T7 RNAP or RNAPII in HeLa nuclear extracts for 90 min under transcription/splicing coupling conditions. Oligo R was added to the 90-min reactions. Following the release of RNPs by RNase H, the nascent RNA species were co-immunoprecipitated (IP) with antibodies against REF (lanes 5 and 6), heterogeneous nuclear ribonucleoprotein (hnRNP) A1 (lanes 13 and 14), and with control mouse IgG (lanes 7 and 8, 15 and 16). The RNAs were resolved on denaturing polyacrylamide gel. Ten percent of the input RNAs are indicated in lanes 1– 4 and 9 –12. The asterisk shows an unknown band that appeared during the immunoprecipitation reaction. C, the immunoprecipitation efficiencies were determined by dividing the amount of immunoprecipitated RNAs by the amount of input RNA; the error bars represent the standard deviation of at least three independent experiments. The white and black bars indicate the immunoprecipitation efficiency of the unspliced and spliced transcripts, respectively.
splicing reaction was being driven either by the T7 promoter or MLP, the spliced -globin mRNA appeared to be resistant to RNase H, suggesting that the EJC assembly comparably occurs on both T7- and MLP-derived spliced mRNAs (data not shown). Next, the association between the EJC and the protected region of the spliced mRNA was confirmed by the immunoprecipitation of RNA species synthesized from three -globin gene constructs in the coupled in vitro transcription-splicing reaction (Fig. 2A). EJC assembly reportedly requires at least 38 nt of the upstream exon (8). Thus, we expected that the EJC would be able to assemble on the /177- (/FL: full-length) and /38spliced RNAs but not on the /17-spliced RNA. The co-immunoprecipitation of RNA species synthesized from the respective three -globin templates in vitro revealed that each REF or FLAG-RNPS1 (Fig. 2B) co-precipitated /38-spliced RNA more efficiently (2.7–5.9-fold) than /17-spliced RNA (compare /38 and /17 in Fig. 2B). At least three EJC components were confirmed to associate at the predicted EJC binding site. Unexpectedly, /FL-spliced RNAs were precipitated more efficiently (5.6-fold) than /38-spliced RNA when the REF was pulled down (compare /FL and /38 in Fig. 2C), whereas almost the same amounts of /FL and /38 were precipitated with FLAG-RNPS1 (1.2-fold). FLAG-UAP56 also co-precipitated /FL more efficiently (2.8-fold) than /38 (supplementary Fig. S1). These data indicate that REF and UAP56 have additional binding site(s) located further upstream of ⫺38 (hereafter referred to as the UP site), relative to the exon junction (⫹1). Splicing-independent Association of REF with mRNA—REF efficiently binds to mRNA independently of splicing (27). To examine whether the binding of REF to the UP site is splicingdependent, a -globin construct lacking introns (intronless) was used in the coupled in vitro transcription-splicing reaction and RNA fragments digested by RNase H were immunoprecipitated by an ␣-REF antibody. RNase H mapping revealed that REF binds to intronless mRNA upstream of the EJC deposition site (supplementary Fig. S2). These data strongly suggest the existence of two independent mechanisms of REF recruitment: 1) a splicing-dependent mechanism that acts just upstream of the exon junction (⫺24 to ⫺20) and that involves the EJC, and 2) a splicing-independent mechanism acting at the UP site. Cap-dependent Association of REF with RNA—Capping is a post-transcriptional RNA processing event that is coupled to RNAPII transcription. Our observations that 1) ␣-REF antibody preferentially precipitated RNAPII transcripts, rather than T7 transcripts (Fig. 1, B and C), and 2) REF associated with both spliced and intronless transcripts prompted us to examine the formation of the cap structure in our transcription/splicing coupling system. As expected, RNAPII transcripts were immunoprecipitated with the ␣-m7G cap antibody (H20) 7.0-fold more efficiently than the T7 transcripts (Fig. 3A, lane 3). This finding was consistent with previous reports stating that capping is coupled to RNAPII transcription (38, 39). Next, to examine the mechanism by which the cap structure is linked to REF recruitment, we utilized a canonical in vitro splicing system and pre-synthesized m7G-capped -globin pre-mRNA. Immunoprecipitation using an antibody against REF revealed that REF associates with both m7G-capped spliced and intronless
Interaction between Cap-binding Complex and RNA Export Factor
15648 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 282 • NUMBER 21 • MAY 25, 2007
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
at similar levels in the presence of RNase A, suggesting that no RNA molecule bridges the two proteins and that REF interacts with CBC and/or the cap structure (Fig. 3B, lanes 3 and 4). Our trial with m7GTP-Sepharose failed to detect the pulldown of FLAG-Y14 from an HEK293 whole cell extract. On the other hand, FLAG-REF, -CBP20, and -CBP80 were efficiently precipitated with m7GTP-Sepharose, confirming that REF and the CBC associate with the cap structure (data not shown). CBC-dependent REF Recruitment to Intronless mRNA—Next, we performed a co-immunoprecipitation analysis to examine the interaction between CBC and REF. FLAG-REF was transfected into HEK293 cells and then immunoprecipitated using an ␣-FLAG antibody from RNasetreated whole cell extract. A Western blot of the immunoprecipitated fraction revealed that CBP20 (Fig. 3C lane 4), but not Y14 (data not shown), was co-immunoprecipitated with FLAG-REF. As a negative control, FIGURE 2. Association of REF with spliced RNA at a region other than the EJC deposition site. A, the FLAG-U2B⬙ was utilized for the coimmobilized/psoralen MLP--globin DNA templates containing 177, 38, and 17 nt of exon 1 (named /FL-DNA, immunoprecipitation experiment; /38-DNA, and /17-DNA, respectively) were used to generate each RNA. Le Hir et al. (8) showed that the EJC is not deposited on /17-spliced RNA in vitro (EJC, /38, ⫹; /17, ⫺). B, oligo R was added to the 90-min tran- however, neither CBP20, REF, nor scription/splicing coupling reactions in the HeLa alone (lanes 1– 8) and the HeLa/HEK293 nuclear extract mix- Y14 was detected in the immunotures (lanes 9 –16) containing FLAG-vector (control) and FLAG-RNPS1; the mixtures were incubated at 30 °C for precipitated fraction (data not 15 min to release each mRNP from RNAPII. The /FL, /38, and /17 nascent RNA species were co-immunoprecipitated (IP) using ␣-REF (lanes 5– 8) or ␣-FLAG M2 (lanes 13–16) antibodies. These RNAs were resolved on shown). All these co-immunopredenaturing polyacrylamide gels. Ten percent of the input RNAs is shown in lanes 1– 4 and 9 –12. C, the immu- cipitation experiments showed that noprecipitation efficiencies were determined by dividing the amount of immunoprecipitated RNAs by the amount of each spliced RNA input; the error bars represent the standard deviation of at least three independent REF interacts with the CBC in experiments. snRNP, small nuclear ribonucleoprotein. HEK293 cells. To test whether the cap structure mRNAs under in vitro splicing conditions, even if transcription is essential for the intronless RNA binding of REF, we next is not coupled (27). To obtain additional evidence that UP site performed a competition experiment using a cap analog recruitment of REF is reproduced in the in vitro splicing reac- (m7GpppG). As shown in Fig. 3D, excess amounts of the cap tion, RNase H mapping was performed to map the REF-binding analog (100 M) impaired the co-immunoprecipitation of the site. This analysis revealed that REF bound to the UP site of the capped intronless RNA with REF, suggesting that the cap strucspliced and the intronless m7G-capped mRNA even in the ture is required for interaction between REF and intronless absence of coupling with RNAPII transcription (data not mRNA. We next examined whether the cap structure was shown). These observations support our hypothesis that the required for association of REF with spliced mRNA. Even cap structure, rather than RNAPII transcription itself, plays a though the in vitro splicing reaction was partially impaired by critical role in REF recruitment at the UP site. the addition of the cap analog (10 M m7GpppG), as reported 7 The m G cap structure is recognized by a heterodimeric previously (40), the spliced -globin RNAs were efficiently conuclear CBC that consists of CBP20 and CBP80 (32). To inves- immunoprecipitated with ␣-REF antibody (Fig. 3E, lane 3). The tigate the interaction between the cap structure and REF, we relative efficiency of the co-immunoprecipitation of spliced performed a pulldown assay using Sepharose beads conjugated RNA in the presence or absence of the cap analog was deterwith an m7GTP cap analog (m7GTP-Sepharose). The precipi- mined (Fig. 3E, lane 2, 100%; lane 3, 60%). The 40% diminution tated m7GTP-Sepharose from HeLa nuclear extract contained in immunoprecipitation can be explained by the impairment of CBP20 and REF but did not contain either of the EJC compo- cap-dependent REF recruitment resulting from the addition of nents (Y14 and UAP56) (Fig. 3B) or the TATA box-binding the cap analog, but the splicing-dependent recruitment of REF protein (data not shown). This interaction remained detectable into the EJC was not affected. We observed the same effect of
Interaction between Cap-binding Complex and RNA Export Factor
MAY 25, 2007 • VOLUME 282 • NUMBER 21
DISCUSSION Nascent RNAPII transcripts are immediately packed with various RNA-binding proteins. Recently, the idea that mRNP packaging during transcription may be mechanistically linked to the progression of mRNA processing has emerged. In this study, we observed that RNA nuclear export factor REF was recruited to RNAPII transcripts more efficiently than to T7 RNAP transcripts. Co-immunoprecipitation analysis attributed this observation to the association of REF with capped mRNAs through CBC. Our findings are consistent with the previous study showing that REF can interact with pre-mRNA prior to spliceosome assembly, whereas other EJC components (Y14, Magoh, RNPS1, UAP56, and SRm160) are found in spliantibody. After the reaction, each RNA was co-immunoprecipitated by antibodies against REF (lanes 1– 6) or FLAG M2 (lanes 7–12). Ten percent of the input RNA is shown in the top panel. The indicated immunoprecipitation efficiencies were determined by dividing the amount of each co-immunoprecipitated RNA by the amount of each RNA input and normalized by that in the absence of the cap analog (lanes 2 for 3, 5 for 6, 8 for 9, and 11 for 12).
JOURNAL OF BIOLOGICAL CHEMISTRY
15649
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
FIGURE 3. REF recruitment by CBC to spliced and intronless RNAs. A, both the immobilized/psoralen T7--globin and MLP--globin cDNA templates were used to generate intronless transcripts over a period of 90 min. After protein removal and ethanol precipitation, each RNA was immunoprecipitated with ␣-m7G cap antibody (H20). The RNAs were resolved on denaturing polyacrylamide gel and imaged using a phosphorimage analyzer. Ten percent of the input RNA is shown in lane 1. The immunoprecipitation (IP) efficiencies were determined by dividing the amount of immunoprecipitated RNA with the ␣-m7GpppG cap antibody by the amount of each RNA input. The mean efficiencies of at least three independent experiments are shown. B, either 7-methyl GTP (m7GTP, lanes 3 and 4) or ␣-FLAG M2 (lanes 5 and 6) Sepharose 4B was incubated with HeLa nuclear extract at 4 °C for 3 h. The incubated Sepharose was washed eight times with NET-2 buffer. The proteins bound to each Sepharose were analyzed by Western blotting using the antibodies indicated on the right. HeLa nuclear extract was preincubated in either the absence (⫺) or presence (⫹) of RNase A for 15 min at 30 °C. A Western blot of 2.5% of the input HeLa nuclear extract is shown in lanes 1 and 2 as a control. C, expression plasmids containing FLAG-vector (control) or FLAG-REF (F-REF) were transfected into HEK293 cells. These cells were lysed using a non-ionic detergent buffer and incubated with RNase A for 15 min at 30 °C. Immunoprecipitation was performed using ␣-FLAG M2-agarose, followed by elution with the FLAG peptide. The resulting immunoprecipitants were analyzed by Western blotting using the antibodies indicated to the left of the panel. A Western blot of 25% of the input HEK293 whole extract is shown in lanes 1 and 2 as a control. D, HeLa nuclear extract was pre-treated with the indicated concentration (0, 1, 10, or 100 M) of the m7GpppG cap analog at 30 °C for 30 min. Following the incubation of this extract with RNA pre-synthesized by T7 RNAP (containing an m7GpppG cap and lacking an intron) in a 90-min in vitro splicing reaction, each RNA was co-immunoprecipitated with ␣-REF antibody. Ten percent of the input RNA is shown in the top panel. The indicated immunoprecipitation efficiencies were determined by dividing the amount of co-immunoprecipitated RNA by the amount of each RNA input and normalized by that in the absence of the cap analog (lane 2). E, HeLa nuclear extract was pretreated with (100 M; lanes 3, 6, 9, and 12) or without (lanes 1, 2, 4, 5, 7, 8, 10, and 11) m7GpppG cap analog at 30 °C for 30 min. The preincubated extract was mixed with the capped RNA pre-synthesized by T7 RNAP and either containing (lanes 1–3 and 7–9) or lacking (lanes 4 – 6 and 10 –12) an intron in 90-min in vitro splicing reactions. The nuclear extract from HEK293 cells expressing FLAG-vector (control) and FLAG-UAP56 (F-UAP56) were added to that from HeLa cells for immunoprecipitation using the ␣-FLAG M2
the cap analog using another RNA template (IgM C3-C4) (data not shown). The association of UAP56 with intronless RNA was also impaired by the cap analog (Fig. 3E, lane 12, 4%), but the association with spliced mRNA remained at a substantial level (Fig. 3E, lane 9, 58%). Interestingly, UAP56 did not bind to the m7GTP cap analog (Fig. 3B, lanes 3 and 4), even though the binding of UAP56 to intronless RNA was impaired by the cap analog (Fig. 3E, lane 12). These results suggest that UAP56 recruitment to intronless RNA may require the cap-dependent binding of REF to RNA. The CBP20-REF Interaction-mediated mRNA Export—To identify the CBP20-binding region of REF, we prepared expression vectors of FLAG-REF mutants, which lack one of highly conserved domains referred as REF-N, -RNP2, -RNP1, and -C motifs (Fig. 4A) (23). The co-immunoprecipitation experiment from the HEK293 cell extracts transfected with the expression vectors showed RNP consensus sequences to be RNA-binding regions that are critical for the interaction with CBP20 (Fig. 4A). REF-RNP2 and -1 consensus sequences are not involved in the interactions with CBP80, UAP56, and TAP (data not shown). These observations indicate that RNP consensus sequences are important for CBP20, but not other interaction proteins tested. Next, to examine the functional role of the CBP20-REF interaction in mRNA export, we prepared affinity purified wild-type REF (REF-WT) and a mutant REF lacking CBP20-binding region (REF-⌬RNP1), and used them for microinjection into HeLa cell nuclei combining with Cy3-labeled intronless -globin mRNA, CBP20, and FITC-dextran as an injection marker. The purity and amounts of CBP20, REF-WT, and REF-⌬RNP1 were estimated on the Coomassie Brilliant Blue-stained gel (Fig. 4B). Cy3-labeled intronless -globin mRNA injected with CBP20 and REF-WT was transported to cytoplasm in more than 90% of cells within 10 h (n ⫽ 26, Fig. 4C, panel b), whereas the Cy3-labeled RNA with CBP20 and REF-⌬RNP1 was retained in nuclei in more than 60% of injected cells during the same period (n ⫽ 33, Fig. 4C, panel f), indicating that the interaction between CBP20 and REF is required for the promotion of intronless mRNA export from the nucleus.
Interaction between Cap-binding Complex and RNA Export Factor
ceosomes (30). And this may explain that REF proteins mediate the export of both spliced and intronless mRNAs from the nucleus (27). Interestingly, REF was inefficiently precipitated at early time points in our transcription/splicing coupling reaction, although the transcripts possess a cap structure. The ␣-REF antibody may be inaccessible to REF in the CBC-REF complex under this condition. The 5⬘ cap structure, a target for CBC, stimulates pre-mRNA splicing (28, 29) and polyadenylation (41) in the nucleus. A chromatin immunoprecipitation assay in yeast has shown that the CBC is necessary for the correct co-transcriptional assembly of the spliceosome (42) and that CBC depletion reduced the recognition of cap-proximal 5⬘ splice sites but did not affect that of cap-distal splice sites (43). These observations suggest that the CBC forms a unique complex on cap-proximal exons.
15650 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 282 • NUMBER 21 • MAY 25, 2007
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
FIGURE 4. CBP20-REF complex-mediated mRNA export. A, expression plasmids encoding FLAG-REF wild-type (WT) or mutants (⌬N, ⌬RNP2, ⌬RNP1, and ⌬C) were transfected into HEK293 cells. The constructs of mutants are shown in the upper panel. Whole cell extracts were incubated with RNase A for 15 min at 30 °C, followed by immunoprecipitation using ␣-FLAG antibody. Thirty percent of the input extracts (lanes 1– 6) and the resulting immunoprecipitates (lanes 7–12) were analyzed by Western blotting using the antibodies indicated at the right. B, FLAG-CBP20 (lane 1), REF-WT (lane 2), and REF-⌬RNP1 (lane 3) were affinity purified, and separated by SDS-PAGE, and the amounts of proteins were estimated on the gel. An asterisk indicates nonspecific protein that is precipitated by ␣-FLAG antibody. C, HeLa cells were microinjected into nuclei with RNP containing Cy3-intronless -globin mRNA, FLAG-CBP20, and FLAG-REF (WT, panels a– d; or ⌬RNP, panels e– h), and incubated for 10 h. Representative photos (panels a–h) and statistical summary (the bottom table) are shown. Presented are single confocal sections of FITC-dextran (a and e), Cy3-intronless RNA (b and f), differential interference contrast (DIC) (c and g), and merged images (d and h). Ten hours after nuclear microinjection, FITC-dextran remained in nuclei (a and e). Cy3-RNA injected with CBP20 and REF (WT) was exported into cytoplasm (b), whereas that with CBP20 and REF (⌬RNP) remained in nuclei in most cells (f). The bar indicates 10 m. Among cells showing nuclear FITC signal, the number of cells is scored by the preferential localization of Cy3-RNA, in the nucleus, cytoplasm, or equally in both. The actual number of cells with nuclear Cy3-RNA retention and the percentage of the total cells are shown in the table.
Our observations showing an interaction between REF and the CBC and the inhibitory effect of a cap analog on the interaction between REF and capped RNA suggest that the cap-dependent complex contains at least CBC and REF. The CBC is essential for U small nuclear RNA export in metazoans (44). Uridine-rich small nuclear RNA export requires CRM1 that is the importin  receptor family but does not require the mRNA export receptor TAP (45). PHAX (phosphorylated adaptor for RNA export) acts as an adaptor that links the CBC-capped RNA complex to the CRM1-RanGTP complex (46). Because the export pathway of U1 small nuclear RNA can be switched to a TAP-mediated pathway by the insertion of a 300-nt Ftz mRNA sequence into U1 small nuclear RNA (28), the CBC-RNA complex may use either PHAX-CRM1 or REF-TAP as its mRNA export machinery, depending on the length (or structure) of the RNA. A model for the REF-TAP-dependent export of intronless mRNAs was initially proposed based on research examining the export of herpes simplex virus-1 (HSV-1) intronless mRNAs (47, 48). HSV-1 encodes the trans-acting protein ICP27 that is involved in the export of viral mRNAs. The interaction of ICP27 with REF was demonstrated in a yeast two-hybrid system and confirmed in virus-infected cells (47, 48). The injection of ICP27 into Xenopus oocytes dramatically stimulated the export of intronless viral mRNAs, whereas a mutant that did not interact with REF was inactive during RNA export (48), indicating that the recruitment of REF is critical for intronless mRNA export. The involvement of REF in the export of intronless mRNAs was also suggested by experiments involving the injection of ␣-REF antibody into the nuclei of Xenopus oocytes (28). These previous observations could be explained by our finding that REF was recruited by CBP20. Actually the CBP20-REF interaction was essential for the promotion of the intronless mRNA export, according to our microinjection experiments with wild-type and mutant REF proteins. In Drosophila cells, RNA interference experiments showed that REF1/Aly are dispensable for bulk mRNA export (49), suggesting that introncontaining mRNAs do not require REF for their export. As the EJC is not formed on intronless mRNA, CBP20-mediated REF recruitment may play a central role for the export of intronless mRNAs. In yeast, the THO complex consisting of Tho2p, Hpr1p, Mft1p, and Thp2p (50) interacts genetically and physically with components of mRNA export machineries (51, 52). Yra1p/REF and Sub2p/UAP56 are stoichiometrically associated with the heterotetrameric THO complex, and is recruited to transcription-activated genes with or without introns and designated the transcription/export (TREX) complex (53, 54). In the case of Drosophila, gene expression profiling in S2 cells depleted of THO2, UAP56, and REF showed that they play differential roles in mRNA export, suggesting these proteins do not act as units of a single protein complex (49, 55, 56). In mammals, GSTUAP56 reportedly pulled down hTho2, fSAP79, hHpr1, hTex1, fSAP35, fSAP24, and REF in RNase-treated HeLa nuclear extracts (57). Masuda et al. (57) referred to the complex containing REF and UAP56 as the human TREX. However, pre-mRNA splicing and the 5⬘ cap structure are required for the recruitment of human TREX complex to mRNA, and this recruitment did not depend on RNAPII (58). Thus, in mam-
Interaction between Cap-binding Complex and RNA Export Factor mals, three distinct mechanisms of REF recruitment to mRNA may exist: 1) splicing-dependent recruitment involving the EJC, 2) cap- and splicing-dependent recruitment involving the TREX complex on spliced mRNA, and 3) cap-dependent recruitment to intronless mRNA. Although further studies are required, they may play differential roles in selective mRNA export in mammalian cells.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24.
Hirose, Y., and Manley, J. L. (2000) Genes Dev. 14, 1415–1429 Howe, K. J. (2002) Biochim. Biophys. Acta 1577, 308 –324 Maniatis, T., and Reed, R. (2002) Nature 416, 499 –506 Neugebauer, K. M. (2002) J. Cell Sci. 115, 3865–3871 Kornblihtt, A. R., Mata, M., Fededa, J. P., Munoz, M. J., and Nogues, G. (2004) RNA (Cold Spring Harbor) 10, 1489 –1498 Luo, M. L., and Reed, R. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 14937–14942 Le Hir, H., Izaurralde, E., Maquat, L. E., and Moore, M. J. (2000) EMBO J. 19, 6860 – 6869 Le Hir, H., Gatfield, D., Izaurralde, E., and Moore, M. J. (2001) EMBO J. 20, 4987– 4997 Chan, C. C., Dostie, J., Diem, M. D., Feng, W., Mann, M., Rappsilber, J., and Dreyfuss, G. (2004) RNA (Cold Spring Harbor) 10, 200 –209 Palacios, I. M., Gatfield, D., St. Johnston, D., and Izaurralde, E. (2004) Nature 427, 753–757 Shibuya, T., Tange, T. O., Sonenberg, N., and Moore, M. J. (2004) Nat. Struct. Mol. Biol. 11, 346 –351 Le Hir, H., Gatfield, D., Braun, I. C., Forler, D., and Izaurralde, E. (2001) EMBO Rep. 2, 1119 –1124 Kataoka, N., Diem, M. D., Kim, V. N., Yong, J., and Dreyfuss, G. (2001) EMBO J. 20, 6424 – 6433 Degot, S., Le Hir, H., Alpy, F., Kedinger, V., Stoll, I., Wendling, C., Seraphin, B., Rio, M. C., and Tomasetto, C. (2004) J. Biol. Chem. 279, 33702–33715 Luo, M. L., Zhou, Z., Magni, K., Christoforides, C., Rappsilber, J., Mann, M., and Reed, R. (2001) Nature 413, 644 – 647 Li, C., Lin, R. I., Lai, M. C., Ouyang, P., and Tarn, W. Y. (2003) Mol. Cell. Biol. 23, 7363–7376 Sakashita, E., Tatsumi, S., Werner, D., Endo, H., and Mayeda, A. (2004) Mol. Cell. Biol. 24, 1174 –1187 Tange, T. O., Shibuya, T., Jurica, M. S., and Moore, M. J. (2005) RNA (Cold Spring Harbor) 11, 1869 –1883 Lykke-Andersen, J., Shu, M. D., and Steitz, J. A. (2000) Cell 103, 1121–1131 Kim, V. N., Kataoka, N., and Dreyfuss, G. (2001) Science 293, 1832–1836 Zhou, Z., Luo, M. J., Strasser, K., Katahira, J., Hurt, E., and Reed, R. (2000) Nature 407, 401– 405 Strasser, K., and Hurt, E. (2001) Nature 413, 648 – 652 Stutz, F., Bachi, A., Doerks, T., Braun, I. C., Seraphin, B., Wilm, M., Bork, P., and Izaurralde, E. (2000) RNA (Cold Spring Harbor) 6, 638 – 650 Katahira, J., Stra¨ßer, K., Podtelejnikov, A., Mann, M., Jung, J. U., and Hurt, E. (1999) EMBO J. 18, 2593–2609
MAY 25, 2007 • VOLUME 282 • NUMBER 21
JOURNAL OF BIOLOGICAL CHEMISTRY
15651
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
Acknowledgments—We are grateful to Micheal Green and Amy Virbasius for providing ␣-UAP56 antibody and His-REF expression vector H4pRSETC, and Elisa Izaurralde for providing the ␣-CBP20 antibody. We also thank Takako Oshiro and Hiroto Nakanoya for technical assistances, Takashi Ideue and Naoyuki Kataoka for thoughtful discussion, Tokio Tani for helpful information about the mRNA microinjection assay, and members of the Hagiwara Laboratory for helpful discussions.
25. Huang, Y., Wimler, K. M., and Carmichael, G. G. (1999) EMBO J. 18, 1642–1652 26. Huang, Y., and Steitz, J. A. (2001) Mol. Cell. 7, 899 –905 27. Rodrigues, J. P., Rode, M., Gatfield, D., Blencowe, B. J., Carmo-Fonseca, M., and Izaurralde, E. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 1030 –1035 28. Ohno, M., Segref, A., Kuersten, S., and Mattaj, I. W. (2002) Mol. Cell 9, 659 – 671 29. Masuyama, K., Tanigushi, I., Kataoka, N., and Ohno, M. (2004) Genes Dev. 18, 2074 –2085 30. Reichert, V. L., Le Hir, H., Jurica, M. S., and Moore, M. J. (2002) Genes Dev. 16, 2778 –2791 31. Custodio, N., Carvalho, C., Condado, I., Antoniou, M., Blencowe, B. J., and Carmo-Fonseca, M. (2004) RNA (Cold Spring Harbor) 10, 622– 633 32. Izaurralde, E., Lewis, J., McGuigan, C., Jankowska, M., Darzynkiewicz, E., and Mattaj, I. W. (1994) Cell 78, 657– 668 33. Krainer, A. R., Maniatis, T., Ruskin, B., and Green, M. R. (1984) Cell 36, 993–1005 34. Wang, Z., and Rana, T. M. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 6688 – 6693 35. Hirose, T., Shu, M. D., and Steitz, J. A. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 17976 –17981 36. Tokunaga, K., Shibuya, T., Ishihama, Y., Tadakuma, H., Ide, M., Yoshida, M., Funatsu, T., Ohshima, Y., and Tani, T. (2006) Genes Cell 11, 305–317 37. Das, R., Dufu, K., Romney, B., Feldt, M., Elenko, M., and Reed, R. (2006) Genes Dev. 20, 1100 –1109 38. Chiu, Y. L. C., Ho, K., Saha, N., Schwer, B., Shuman, S., and Rana, T. M. (2002) Mol. Cell 10, 585–597 39. Moteki, S., and Price, D. (2002) Mol. Cell. 10, 599 – 609 40. Konarska, M. M., Padgett, R. A., and Sharp, P. A. (1984) Cell 38, 731–736 41. Flaherty, S. M., Fortes, P., Izaurralde, E., Mattaj, I. W., and Gilmartin, G. M. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 11893–11898 42. Gornemann, J., Kotovic, K. M., Hujer, K., and Neugebauer, K. M. (2005) Mol. Cell 19, 53– 63 43. Lewis, J. D., Izaurralde, E., Jarmolowski, A., McGuigan, C., and Mattaj, I. W. (1996) Genes Dev. 10, 1683–1698 44. Izaurralde, E., Lewis, J., Gamberi, C., Jarmolowski, A., McGuigan, C., and Mattaj, I. W. (1995) Nature 376, 709 –712 45. Fornerod, M., Ohno, M., Yoshida, M., and Mattaj, I. W. (1997) Cell 90, 1051–1060 46. Ohno, M., Segref, A., Bachi, A., Wilm, M., and Mattaj, I. W. (2000) Cell 101, 187–198 47. Koffa, M. D., Clements, J. B., Izaurralde, E., Wadd, S., Wilson, S. A., Mattaj, I. W., and Kuersten, S. (2001) EMBO J. 20, 5769 –5778 48. Chen, I. H., Sciabica, K. S., and Sandri-Gordin, R. M. (2002) J. Virol. 76, 12877–12889 49. Gatfield, D., and Izaurralde, E. (2002) J. Cell Biol. 159, 579 –588 50. Cha´vez, S., Garcı´a-Rubio, M., Prado, F., and Aguilera, A. (2001) Mol. Cell. Biol. 21, 7054 –7064 51. Jimeno, S., Rondon, A. G., Luna, R., and Aguilera, A. (2002) EMBO J. 21, 3526 –3535 52. Zenklusen, D., Vinciguerra, P., Wyss, J. C., and Stutz, F. (2002) Mol. Cell. Biol. 22, 8241– 8253 53. Strasser, K., Masuda, S., Manson, P., Pfannstiel, J., Oppizzi, M., RodriguezNavarro, S., Rondon, A. G., Aguilera, A., Struhl, K., Reed, R., and Hurt, E. (2002) Nature 417, 304 –308 54. Abruzzi, K. C., Lacadie, S., and Rosbash, M. (2004) EMBO J. 23, 2620 –2631 55. Herold, A., Teixeira, L., and Izaurralde, E. (2003) EMBO J. 22, 2472–2483 56. Rehwinkel, J., Herold, A., Gari, K., Ko¨cher, T., Rode, M., Ciccarelli, F. L., Wilm, M., and Izaurralde, E. (2004) Nat. Struct. Mol. Biol. 11, 558 –566 57. Masuda, S., Das, R., Cheng, H., Hurt, E., Dorman, N., and Reed, R. (2005) Genes Dev. 19, 1512–1517 58. Cheng, H., Dufu, K., Lee, C. S., Hsu, J. L., Dias, A., and Reed, R. (2006) Cell 127, 1389 –1400
Supplementary Figure S1. UAP56 association with spliced RNA at the region other than the EJC deposition site. A.
Oligo R was added to 90-minute transcription/splicing coupling reactions in a HeLa/HEK293 nuclear extract mixture containing FLAG-vector (control) and FLAG-UAP56. The mixture was then incubated at 30°C for 15 minutes to release each mRNP from RNAPII. The β/FL, β/38 and β/17 nascent RNA species were co-immunoprecipitated with α-FLAG M2 antibody. These RNAs were resolved on a denaturing polyacrylamide gel. Ten percent of the input RNAs is shown in lanes 1-4.
B. The immunoprecipitation efficiencies were determined by dividing the amount of immunoprecipitated RNAs by the amount of each spliced RNA input; the error bars represent the standard deviation for at least three independent experiments. Supplementary Figure S2. RNase H mapping of REF association site on intronless RNA. A. Schematic representation of intronless mRNA and fragments digested by RNase H. Four kinds of deoxy-oligos (short bar, 1-4) were added to coupling reactions in HeLa nuclear extract and incubated for 15 minutes at 30°C. The 5’ fragments (a, b, c and d) were generated by RNase H using each oligo (1, 2, 3 and 4, respectively), as described. B. The immobilized/psoralen MLP-β-globin DNA template was used to generate the nascent RNA.
The
90-minute
RNAPII
transcripts
were
produced
under
in
vitro
transcription/splicing coupling conditions and incubated with oligo R plus one of the four oligos. The digested RNA species were then co-immunoprecipitated with α-REF antibody (lanes 7 - 12). Each RNA was resolved on denaturing gel. Ten percent of the input RNAs are shown in lanes 1 - 6.
The Interaction between Cap-binding Complex and RNA Export Factor Is Required for Intronless mRNA Export Takayuki Nojima, Tetsuro Hirose, Hiroshi Kimura and Masatoshi Hagiwara J. Biol. Chem. 2007, 282:15645-15651. doi: 10.1074/jbc.M700629200 originally published online March 15, 2007
Access the most updated version of this article at doi: 10.1074/jbc.M700629200
Click here to choose from all of JBC's e-mail alerts
Supplemental material: http://www.jbc.org/content/suppl/2007/03/16/M700629200.DC1.html This article cites 58 references, 30 of which can be accessed free at http://www.jbc.org/content/282/21/15645.full.html#ref-list-1
Downloaded from http://www.jbc.org/ at Radcliffe Science Lib, Oxford Univ on February 6, 2016
Alerts: • When this article is cited • When a correction for this article is posted
