RNA helicase A (RHA) is a highly conserved DEAD-box protein that activates transcription, modulates RNA splicing and binds the nuclear pore complex. The life cycle of typical mRNA involves RNA processing and translation after ribosome scanning of a relatively unstructured 5′ untranslated region (UTR). The precursor RNAs of retroviruses and selected cellular genes harbor a complex 5′ UTR and use a yet-to-be-identified host post-transcriptional effector to stimulate efficient translation. Here we show that RHA recognizes a structured 5′-terminal post-transcriptional control element (PCE) of a retrovirus and the JUND growth-control gene. RHA interacts with PCE RNA in the nucleus and cytoplasm, facilitates polyribosome association and is necessary for its efficient translation. Our results reveal a previously unidentified role for RHA in translation and implicate RHA as an integrative effector in the continuum of gene expression from transcription to translation.
Translational control regulates expression of 30% of the eukaryotic proteome and has a pivotal role in effecting rapid response to changes in cellular growth conditions1. Initiation is the rate-limiting step in translation, and the prevailing model describes efficiently translated RNA as monocistronic, containing a m7G5′ppp5′N cap structure at the 5′ terminus and a 5′ UTR that is less than 100 nucleotides (nt) in length and is relatively unstructured2,3. Precursor mRNA (pre-mRNA) splicing stimulates translation, and recent findings imply that the deposition of an exon-junction complex (EJC) in the nucleus constitutes an RNA-surveillance checkpoint that is necessary for efficient translation in the cytoplasm4,5,6. Naturally unspliced mRNA templates that contain a long and highly structured 5′ UTR represent exceptions to this paradigm; these mRNAs require alternative post-transcriptional regulatory circuits for efficient protein synthesis. Examples of such templates include the pre-mRNAs of retroviruses and some naturally intronless cellular genes, which contain a complex 5′ UTR and a m7G5′ppp5′N cap structure and are expected to lack an EJC7,8.
In the case of human immunodeficiency virus type 1 (HIV-1) and other complex retroviruses, the viral regulatory protein Rev, in conjunction with the Rev-responsive element (RRE), trans-activates efficient post-transcriptional expression of gag from the viral pre-mRNA9. However, genetically simpler retroviruses lack a viral post-transcriptional regulatory protein and are reliant on a host post-transcriptional regulatory protein. Bioinformatic searches identify approximately 200 human naturally intronless genes that share features with retrovirus gag mRNA. One example is the JUND growth-control gene, of the AP-1 family of transcription factors, which contains a highly structured ∼200-nucleotide (nt) 5′ UTR and initiates translation by a cap-dependent mechanism7. It is possible that JUND and selected retroviral genes recruit a common host post-transcriptional regulatory protein to stimulate their efficient translation.
Recently, a unique 5′-terminal orientation-dependent PCE has been identified in the avian spleen necrosis virus (SNV) and Mason-Pfizer monkey virus that facilitates translation of unspliced RNA containing a complex 5′ UTR10,11. The PCE functions in concert with a yet-to-be-identified host-cell effector protein to stimulate Rev-RRE–independent expression of HIV-1 unspliced gag RNA by facilitating polyribosome association11,12. Incongruously, however, the PCE and adjacent cis-acting replication sequences present structural barriers to ribosome scanning13,14. Combined results of site-directed mutagenesis and enzymatic mapping experiments have demonstrated that the PCE is composed of two redundant stem-loop structures that interact with the host-cell effector protein15. Studies with bicistronic reporter RNAs have determined that the PCE does not function as an internal ribosome entry site to stimulate cap-independent internal initiation12. Quantitative RNA and protein analyses of deletion and point mutants has shown that elimination of Gag protein production does not correlate with a change in steady-state abundance, splicing efficiency or cytoplasmic accumulation of gag RNA containing the PCE (PCE-gag). Instead, loss-of-function mutations eliminate use of the cytoplasmic RNA as template for Gag protein production10,15. Results of RNA transfection and competition studies using Rev-RRE have suggested that PCE interaction with a cellular effector in the nucleus is necessary for translational stimulation16.
We hypothesized that the PCE effector is a nucleocytoplasmic shuttle protein with RNA helicase activity that neutralizes barriers to efficient ribosome scanning. A good candidate was RHA, a member of the DEAD-box family of RNA helicases17. DEAD-box proteins catalyze rearrangements of RNA–RNA and RNA–protein complexes in virtually all steps of RNA processing and metabolism, including transcription, splicing, association of RNA with the nuclear pore complex and export18. Here we set out to identify and characterize the PCE effector protein. Our findings reveal a new and essential role for RHA in efficient translation of selected mRNAs by recognition of features of their complex 5′ UTR.
RNA helicase A recognizes structural features of PCE
PCE RNA–protein complexes were isolated by RNA-affinity chromatography using biotinylated RNA and concentrated HeLa nuclear extracts. In RNA electrophoretic mobility shift assays (EMSAs), the complexes were selectively competed by excess PCE RNA (Fig. 1a). They were not competed by antisense PCE RNA, which is similar in length but dissimilar in secondary structure and lacks PCE activity15. RNA-affinity chromatography and SDS-PAGE results from three replicate experiments consistently detected a ∼150-kDa protein bound exclusively to PCE RNA (Fig. 1b). The lack of detectable binding to antisense RNA indicates a concordance between binding and PCE activity. Two other proteins of ∼100 and 60 kDa consistently eluted from both PCE and antisense RNA affinity columns, indicating lack of concordance between binding of these proteins and PCE activity. Eluants of four replicate PCE RNA affinity columns were pooled and the 150-kDa protein was evaluated by MALDI-TOF MS. Bioinformatic analysis identified the ∼150-kDa protein as human DExH-box protein 9 (DHX9), also known as RNA helicase A. Western blotting verified that RHA was present exclusively in the 2 M eluant of the PCE RNA affinity column and not that of the antisense RNA affinity column (Fig. 1c).
EMSAs were used to examine the specificity of interaction of RHA with the PCE. EMSA reactions were performed with the PCE and the AC′Aall point mutant, which contains substitutions that eliminate the complementary base-pairing between the A and C stems and alter the unpaired A-loop nucleotides that are necessary for PCE activity15. Addition of RHA antiserum to a PCE EMSA reaction produced a subtle shift in the mobility of the high–molecular weight PCE–protein complexes (Fig. 1d). A shift was not observed upon addition of RHA antiserum to the AC′Aall and antisense EMSA reactions. Similar results were observed in EMSA reactions performed with cell extracts prepared from COS cells that had been transfected with plasmid expressing Flag epitope–tagged RHA (pFlag-RHA) (Fig. 1e). Addition of Flag antibody to the PCE EMSA produced the subtle shift in the mobility of the PCE–protein complexes; addition of Flag antibody to AC′Aall and the antisense EMSA reactions did not produce a shift. These results demonstrated that RHA is a component of the PCE RNA–protein complexes.
The specificity of PCE interaction with RHA was further investigated by a coimmunoprecipitation assay. COS cells were cotransfected with HIV-1 gag test plasmids that contained wild-type PCE, the antisense sequence or PCE point mutants (Fig. 2a), together with pFlag-RHA. Cell extracts were prepared 2 d later, assayed for PCE activity by Gag ELISA (Fig. 2a) and immunoprecipitated with Flag antibody. Western blotting of the cell lysate before the immunoprecipitation showed that similar levels of Flag-RHA were present in the cells and that Flag-RHA was not detectable in control cells transfected with empty plasmid (Fig. 2b, 'L' lanes). Western blotting of the supernatant after immunoprecipitation showed that Flag-RHA had been efficiently depleted (Fig. 2b, 'S' lanes). Reverse-transcription (RT)-PCR on total cellular RNA with gag-specific primers demonstrated that gag RNA was expressed in the cells (Fig. 2c). The RT-PCR product was present in reactions that contained reverse transcriptase and absent in reactions that lacked reverse transcriptase. RNA was harvested from the immunoprecipitates and subjected to RT-PCR with gag-specific primers. An RT-PCR product was detectable in samples from cells that expressed gag with wild-type PCE, but not gag with an antisense or PCE deletion mutant (Delta). Flag-tagged RHA did not coprecipitate with gag RNA containing the AC′ or AC′Aall structural mutation15 (Fig. 2d). The AA′CC′ compensatory mutation rescued PCE activity and coprecipitation with Flag-RHA (Fig. 2a,d). The AallCall mutant showed partial PCE activity and coprecipitated with Flag-RHA. RT-PCR with gene-specific primers showed that RHA did not coprecipitate with endogenous GAPDH or with MYC RNA, which contains a highly structured 5′ UTR (Fig. 2d). These results indicate that RHA interaction correlates with PCE activity and that RHA selectively recognizes structural characteristics presented by A and C stem-loops.
RHA is necessary for PCE activity
To investigate the possibility that RHA is necessary for PCE translational enhancement, the effect of downregulating endogenous RHA was examined using small interfering RNAs (siRNAs). Immunoprecipitation assays with RHA antibody showed that the half-life of RHA protein in COS cells is approximately 30 h and established that RHA is a feasible target for downregulation by RNA-mediated interference (data not shown). The cells were transfected with either RHA siRNAs (RHA) or scrambled siRNAs (Sc) and the downregulation of RHA was monitored by northern and western analyses. Transfection with RHA siRNAs reduced endogenous RHA mRNA to <15% of control, whereas no effect was observed on GAPDH RNA levels (Fig. 3a) and cell viability remained similar during the 4 consecutive days after transfection. RHA protein was reduced to <30% at day 3 and day 4 post-treatment (Fig. 3b). To evaluate PCE activity, the cells were transfected 2 d after siRNA treatment with the PCE HIV-1 gag test plasmid and the pGL3 reference plasmid that produces firefly luciferase. Cell extracts were harvested on day 4 and Gag and luciferase levels were measured by ELISA and enzymatic assay, respectively. Results of five independent transfection experiments indicated that RHA downregulation significantly reduced Gag protein production without reducing luciferase production (P = 0.003) (Table 1). Gag activity was reduced to within 17%–40% of control and was proportional to the downregulation of RHA by the siRNAs.
Results of real-time RT-PCR on total cellular RNA showed that the steady-state abundance of HIV-1 gag RNA was not changed by RHA downregulation (Fig. 3c). The copy numbers of HIV-1 gag RNA remained similar (40 × 102 and 43 × 102 copies per μg), despite reduction of Gag protein production by a factor of 5, from 53 to 10 ng ml−1. To confirm that the reduction in Gag production was not attributable to a lower steady-state abundance of gag RNA, RNase protection assays (RPAs) were performed on nuclear and cytoplasmic RNA. RPAs demonstrated that RHA downregulation did not reduce the steady-state abundance of PCE-gag RNA containing the PCE (Fig. 3d), despite the five-fold reduction in Gag production. Furthermore, RHA downregulation did not affect splicing efficiency or reduce cytoplasmic accumulation. Immunoblotting with histone H1 and β-actin on an aliquot of the nuclear and cytoplasmic proteins verified effective fractionation of the nucleus from the cytoplasm and confirmed that equivalent amounts of protein were analyzed (Fig. 3e). Immunoblotting with RHA showed that RHA downregulation was effective and that RHA is abundant in the nucleus (Fig. 3e).
The observation that RHA activity is not attributable to increased PCE-gag RNA in the cytoplasm is consistent with the hypothesis that RHA facilitates translation of this mRNA. If RHA facilitates translation, RHA present in the cytoplasm may preferentially associate with polyribosomes. Sucrose gradients and RHA immunoblotting were used to assess the distribution of RHA in the cytoplasm. RHA present in the cytoplasm was associated with the high–molecular weight fractions and was not observed in lower–molecular weight messenger ribonucleoprotein particles (mRNPs) (Fig. 3f). Dissociation of the ribosomal subunits by treatment with EDTA disrupted the polyribosome association of RHA. Furthermore, reduction of polyribosomes by treatment with puromycin to induce premature translation termination also redistributed RHA to lighter fractions. These data indicate that RHA in the cytoplasm is preferentially associated with actively translating polyribosomes. The polyribosomal distribution of RHA is consistent with the hypothesis that RHA is necessary for efficient translation of PCE-gag RNA.
Another approach to examine the effect of RHA on PCE activity used supplementation of endogenous RHA with recombinant RHA. COS cells were cotransfected with the PCE-gag test plasmid, the pGL3 reference plasmid and increasing amounts of pFlag-RHA expression plasmid. Equivalent amounts of transfected DNA were maintained by supplementation with empty plasmid. Flag immunoblotting verified a gradient of expression of recombinant RHA (Fig. 4). A Gag ELISA revealed a positive correlation between Flag-RHA expression and PCE activity, with PCE activity increasing from 8 ng ml−1 in the absence of Flag-RHA to 32 ng ml−1 at the highest concentration of Flag-RHA. In the absence of Flag-RHA, Gag production from the antisense-gag test plasmid was less than the minimum detectable and was not stimulated by expression of Flag-RHA. RPAs verified that Flag-RHA overexpression did not alter the steady-state abundance, splicing efficiency or nuclear export of PCE-gag RNA (data not shown), consistent with RHA facilitating the translation of PCE-gag RNA. These results indicate that overexpression of RHA increases PCE activity.
RHA is necessary for efficient translation of PCE-gag RNA
Metabolic labeling experiments were used to assess the effect of RHA downregulation on translational efficiency of PCE-gag RNA and cellular RNAs. COS cells were treated with siRNAs for 2 d and downregulation of RHA was verified by western blotting. The cells were then cotransfected with the same siRNAs, the PCE-gag test plasmid and the pGL3 reference plasmid. After 2 d, we used western blotting on an aliquot of the cells to verify downregulation of RHA to <30%, and additional aliquots of the cells were labeled with [35S]cysteine and [35S]methionine for 15-min intervals. Cell extracts were prepared and subjected to immunoprecipitation with Gag antibody, trichloroacetic acid (TCA) precipitation or both. The Gag immunoprecipitation demonstrated that RHA downregulation severely reduced the rate of Gag protein synthesis (Fig. 5a). For cells treated with scrambled siRNAs, the incorporation of [35S]cysteine/methionine into Gag increased from 20 × 103 units at 15 min after labeling to 130 × 103 units at 60 min after labeling. By contrast, for cells treated with RHA siRNAs, the level of incorporation of [35S]cysteine/methionine into Gag remained approximately 10 × 103units at each time point. Evaluation in parallel of TCA-precipitable counts showed that global cellular protein synthesis was not altered by RHA downregulation (Fig. 5b). Similarly, metabolic labeling with [3H]uridine and a TCA-precipitation assay verified that RHA downregulation did not alter global cellular mRNA synthesis (Fig. 5cFootnote 1). These results, together with the lack of effect of RHA downregulation on PCE-gag RNA metabolism, indicate that RHA is necessary for efficient translation of PCE-gag RNA.
RHA is necessary for efficient translation of JUND mRNA
Because the 5′ UTR of the naturally intronless JUND shares features with the SNV PCE7,15, derivatives of the PCE-gag test plasmid were created to test the hypothesis that JUND contains a cellular PCE. The 5′-terminal 119-nt sequence of rat Jund was substituted in place of the SNV PCE in the sense and antisense orientations. Results of three replicate transient transfections in 293 cells showed that PCE activity is conferred by Jund in the sense, but not in the antisense orientation (30 ± 5 ng ml−1 Gag as opposed to less than the minimum detectable). The metabolic labeling experiment was used to assess the effect of RHA downregulation on translational efficiency of endogenous JUND RNA. COS cells were treated with siRNAs for 2 d, aliquots of the cells were labeled with [35S]cysteine/methionine and cell extracts were prepared and subjected to immunoprecipitation with the JunD antibody. Immunoprecipitation demonstrated that RHA downregulation substantially reduced JunD protein synthesis (Fig. 6a). For cells treated with scrambled siRNAs, the incorporation of [35S]cysteine/methionine into JunD increased from 15 × 103 units at 30 min after labeling to 130 × 103 units at 360 min after labeling (Fig. 6b). In contrast, for cells treated with RHA siRNAs, the level of incorporation of [35S]cysteine/methionine into JunD was limited to 30 × 103 units at 360 min. By comparison, immunoprecipitation with GAPDH antibody showed that the rate of GAPDH protein synthesis was unaffected by downregulation of RHA (Fig. 6a). Northern blot analysis of total cellular RNA showed that RHA downregulation did not alter the steady-state abundance of JUND or GAPDH mRNA, but reduced RHA mRNA to a barely detectable level (Fig. 6c). To further evaluate the effect of RHA downregulation on translational efficiency of JUND, sucrose gradients were prepared and RT-PCR used to assess JUND RNA association with polyribosomes. Downregulation of RHA did not reduce the cytoplasmic abundance of JUND or GAPDH mRNA, but it selectively reduced the polyribosome association of JUND mRNA (Fig. 6d).
The coimmunoprecipitation assay was used to assess Flag-RHA interaction with endogenous JUND RNA in the cells cotransfected with the PCE test plasmids (Fig. 2d). RT-PCR with JUND-specific primers showed that JUND RNA coprecipitated with Flag-RHA. The RT-PCR products were not detected in reactions that lacked reverse transcriptase. The RNA immunoprecipitation assay was performed on fractionated nuclear and cytoplasmic proteins to evaluate interaction with RHA in the nucleus or the cytoplasm. COS cells were transfected with Flag-RHA, PCE-gag test plasmid and the pGL3 reference plasmid. Cells were harvested 2 d later and nuclear and cytoplasmic proteins were isolated and subjected to immunoprecipitation with Flag antibody. RNA isolated from the immunoprecipitates produced an RT-PCR product with JUND primers and gag primers, but not with GAPDH primers (Fig. 7a). The JUND and PCE-gag RNAs were detectable in both the nucleus and cytoplasm. Control cells that lacked Flag-RHA and reactions that lacked reverse transcriptase did not produce the RT-PCR product. Immunoblotting with histone H1 or β-tubulin antiserum verified effective fractionation of nucleus and cytoplasm (Fig. 7b). Immunoblotting of the samples before immunoprecipitation confirmed that levels of Flag-RHA were similar (Lysate), and immunoblotting of the supernatants after immunoprecipitation showed that the Flag-tagged RHA was effectively depleted (Sup). Immunoblotting with β-actin antiserum verified similar protein loading. The combined results of the immunoprecipitation and the metabolic labeling assays indicate that RHA interacts with JUND mRNA and PCE-gag in the nucleus and cytoplasm and selectively stimulates their rates of protein synthesis.
Previous to this report, RHA had been shown to activate transcription19,20,21,22, modulate RNA splicing23,24,25 and bind the nuclear pore complex26. Our results demonstrate that RHA is necessary for efficient translation of selected RNAs that contain a highly structured 5′ UTR. The RHA-PCE interaction stimulates polyribosome incorporation and the rate of protein synthesis. RNA immunoprecipitation experiments with PCE point mutants indicate that RHA selectively recognizes structural characteristics of the PCE. These structural characteristics are not shared with GAPDH, which contains a relatively short 5′ UTR, or with MYC RNA, which contains a long and highly structured 5′ UTR, or PCE point mutants with disrupted base-pairing of two functionally redundant stem-loops (designated A and C) that are necessary for PCE activity.
Previous RNA structure–mapping studies have shown that point mutations that disrupt complementary base-pairing within the redundant A and C stem-loops eliminate PCE activity15. The loss of function is not attributable to reduction in the steady-state abundance or cytoplasmic accumulation of the RNA15. Instead, the cytoplasmic RNA is translationally silent, reminiscent of translationally silent mRNAs sequestered in stress granules or processing bodies27. Compensatory mutations that restore the base-pairing of the A and C stems rescue PCE activity. The present results show that RHA and PCE RNAs coprecipitate from the nucleoplasm, and previous results show that interaction with a nuclear protein is necessary for PCE activity16. Together, these results support the hypothesis that RHA interaction early in the post-transcriptional expression of PCE RNA satisfies an RNA-surveillance checkpoint necessary for efficient translation in the cytoplasm. Because the JUND 5′ UTR forms redundant stem-loops that are remarkably similar to the PCE7,15, we expect that mutations that disrupt base-pairing within the stem-loops in the JUND PCE will likewise eliminate interaction with RHA and efficient translation. JunD is a member of the Jun family of transcription factors28, which dimerize with Fos or other Jun family members to form activator protein-1 (AP-1). AP-1 is a central signaling component in cell-growth regulation and provides a key pathway for early response to cellular stress29. The expression of AP-1 family members is tightly regulated, and dysregulation of AP-1 family members contributes to cancer and metabolic disease30,31. It is possible that future studies will identify JUND PCE to be an important component in effective regulation of JUND and responsiveness to changes in cell growth conditions. Bioinformatic searches have identified additional naturally intronless genes that likewise have a highly structured 5′ UTR. It is possible that RHA interaction with structural features of these pre-mRNAs operates a conserved post-transcriptional control axis to overcome barriers presented by conserved structural motifs in the 5′ UTR.
A long-standing issue has been to understand the observation that nuclear interactions profoundly effect the ultimate translational use of spliced RNAs32,33,34,35,36. Several recent studies have identified protein mediators that promote translation and RNA surveillance in coordination with intron removal. For example, RNA-protein tethering experiments have determined that the EJC component RNPS1 and Y14 can stimulate translation4,5. Naturally unspliced mRNA templates are expected to require alternative post-transcriptional regulatory circuits for efficient protein synthesis. Our results indicate that RHA is necessary for translation of two unspliced mRNAs. It is possible that recruitment of RHA satisfies an RNA-surveillance checkpoint that would otherwise impede translation of these unspliced mRNAs.
Model for RNA helicase A translation stimulation
We propose a model in which RHA recognition of structural features of PCE stimulates RNA-RNA and/or RNA-protein rearrangements that are necessary for efficient translation (Fig. 8). PCE-containing RNA that does not interact with RHA early in its post-transcriptional expression is incompetent for translation. This translationally silent mRNA is detectable in the cytoplasm and may be sequestered in the form of nonpolysomal mRNP complexes, such as stress granules or processing bodies27. RHA-PCE interaction may stimulate ribosome scanning directly by promoting RNA-protein rearrangements that are necessary for translation initiation. A non-mutually exclusive possibility is that RHA stimulates ribosome recycling by securing circularization of the mRNA template37. A recent study of bovine diarrheal virus has found RHA to bind redundant 5′ and 3′ UTR structures, and the authors speculate that circularization of viral RNA is secured by RHA and promotes productive viral replication38. An important issue to be addressed is whether RHA activity on PCE is constitutive or regulated. We speculate that regulation of RHA activity occurs by post-translational modification of RHA, which provides a conformational change that is necessary for interaction with PCE or an essential cofactor.
Previous studies have shown that translation of mRNAs harboring a complex 5′ UTR can be increased by overexpression of eIF4E, and this activity has been attributed to improved recruitment of eIF4A to the 5′ RNA terminus3. eIF4A is a DEAD-box RNA helicase that has RNA chaperone activity, defined as RNA-unwinding activity and RNPase activity1. The observed requirement for eIF4A in proportion to the degree of secondary structure implies that increased RNA-chaperone activity is engaged during efficient scanning of a complex 5′ UTR39. It is possible that RHA supplements eIF4A RNA chaperone to neutralize the complex 5′ UTR of PCE-containing RNA. RHA may serve to rearrange RNA secondary structures characteristic of the PCE substrates to facilitate eIF4A function. Notably, in in vitro assays, RHA is limited in its ability to unwind stable duplexes and has been shown to require a 3′ overhang40. Another non-mutually exclusive possibility is that RHA serves to stimulate PCE RNA–protein rearrangements. Recent studies on spliced mRNAs reinforce the point that RNA-protein rearrangements operate sequential RNA-surveillance checkpoints that are necessary for efficient post-transcriptional gene expression41.
Convergence of DEAD-box helicases
DEAD/DEXH-box proteins are a versatile superfamily of proteins that have roles in all processes involving RNA18. Recent studies suggest that selected family members recognize specific subsets of RNA targets. For example, Dbp5 provides RNA-chaperone activity to spliced mRNAs exported by a nuclear export pathway mediated by TAP (also called NXF1) during release from the nuclear pore complex42,43. The nuclear isoform of eIF4A, eIF4AIII, preferentially associates with spliced RNAs and functions in nonsense-mediated decay44,45. Another family member, DED1, was previously identified to modulate translation in yeast46. Recent findings connect the human homolog of DED1, DDX3, with nuclear export of Rev-RRE–dependent HIV-1 mRNA by CRM-1 nuclear export receptor47. The possibility that Dbp5 and DDX3 can substitute for RHA to confer PCE activity was ruled out in overexpression experiments (T.R.H. and K.B.-L., unpublished data). It has recently been demonstrated that yeast DEAD-box helicase Dhh1p is a general translation repressor and facilitator of processing-body formation48. Dhh1p modulates the conversion of an active mRNA template to a translationally silent form that is targeted to processing bodies. The results with Dhh1p identify a general mechanism of translation repression that is applicable to the majority of cytoplasmic mRNAs. By contrast, our results identify a selective mechanism of translation stimulation for a specific class of mRNAs that contain a complex 5′ UTR.
DNA templates for in vitro transcription were PCR products that contained the T7 promoter and either the wild-type PCE (pYW100)11, the indicated mutants15 or the antisense PCE (pTR147)12 (Supplementary Table 1 online). After transcription with 15 mM biotinylated UTP or CTP (ENZO) and 135 mM UTP using the Megashortscript kit (Ambion), RNA was precipitated, resuspended in diethylpyrocarbonate-treated water and applied to G25 exclusion columns. Biotinylated RNA (2 μg) was incubated with 50 μl streptavidin coated beads and 120 μl binding buffer (40 mM KCl, 10 mM HEPES (pH 7.6), 5 mM EDTA (pH 8.0), 3 mM MgCl2, 5% (v/v) glycerol, 0.5% (v/v) NP40, 2 mM DTT) for 1 h at 4 °C, centrifuged at 1,200g and incubated with 250 μl biotin blocking solution at room temperature for 5 min. Pelleted beads were washed three times with 250 μl binding buffer and added to HeLa nuclear extract (600 μg) in 100 μl binding buffer with 1,000 μg heparin and 50 μg tRNA, incubated for 2 h at 4 °C, centrifuged and washed four times. Proteins were eluted in 100 μl of progressive concentrations of KCl. The 2-M elution products were dialyzed overnight at 4 °C. Samples were separated on 8% (w/v) polyacrylamide SDS gels and stained with Coomassie blue. Proteins in gel slices were digested with trypsin and microsequenced at the Ohio State University Proteomics Facility.
RNA electromobility shift assays.
Probes were prepared as above but with [α-32P]UTP and the MAXIscript kit (Ambion), and 200,000 c.p.m. was incubated on ice for 30 min with 5 μg heparin and HeLa nuclear lysates or total cell extracts of COS cells 2 d after transfection with pFlag-RHA. Preincubation of lysate was with competitor RNA or RHA antiserum or Flag M2 monoclonal antibody for 10 min and was followed by electrophoresis on 6.5% (w/v) polyacrylamide gels and PhosphorImager analysis. HeLa nuclear extract was prepared from 10 l cell pellet purchased from the National Cell Culture Center. Sucrose-gradient analysis was performed as described12. Puromycin treatment (200 μg ml−1) was for 30 min before cell harvest. Proteins were precipitated in 20% (w/v) trichloroacetic acid for 30 min on ice, centrifuged for 20 min at 13,400g at 4 °C, washed with acetone five times and dried in a Speedvac for 2 h. Pellets were separated on 8% (w/v) polyacrylamide SDS gels and subjected to western blot analysis.
Plasmids, transfections and protein analysis.
PCE-gag reporter plasmids have been described11,12,15. Cotransfections with pGL3 luciferase expression plasmid and pcDNA-Flag-RHA (C.G. Lee, University of Medicine and Dentistry of New Jersey) were performed in triplicate with Lipofectamine (Invitrogen) as described10. Cocktails of RHA siRNAs and scrambled RHA siRNAs that target Homo sapiens RHA (GenBank accession code NM_001357) were provided by Dharmacon and transfected with Oligofectamine at a final concentration of 100 nM per 1 × 105 cells. A Bradford assay was used to measure 50 μg of protein for immunoblots, which were performed as described11 with antiserum including polyclonal rabbit RHA (C.G. Lee), histone H1 and β-tubulin mouse monoclonal antibody (Abcam), β-actin rabbit polyclonal antibody (Novus Biologicals) and Flag M2 monoclonal antibody (Sigma). Metabolic labeling, TCA precipitation assays and immunoprecipitation were performed as described49. Briefly, COS cells were plated at 2 × 105 per 35-mm well and transfected with siRNA in two 24-h intervals. After transfection with pYW100 and incubation for 24 h, the cells were incubated with 10 μCi ml−1 of [35S]cysteine/methionine or [3H]uridine for the indicated periods.
Nuclear and cytoplasmic fractions of COS cells were prepared and RNAs were isolated in TriReagent and treated at least twice with DNase as described10. RPA and probes are described in ref. 15, and northern blots were performed by standard protocols with 32P-labeled DNA probes prepared by random priming (Random Primers DNA Labeling System, Invitrogen). For RNA immunoprecipitation, 4 × 106 COS cells were transfected with pYW100 with pcDNA-Flag-RHA or empty plasmid. Cells were harvested 48 h post-transfection, and either total cellular protein or nuclear and cytoplasmic fractions were prepared as described50 and subjected to either western blotting or RNA immunoprecipitation as described49 using Flag M2 monoclonal antibody. For RNA isolation, the sample was extracted in Trizol and subjected to four DNase treatments. We used either random hexamer or JUND-specific antisense primer and Sensiscript reverse transcriptase (Qiagen) to generate cDNA; one-tenth of the 20-μl reaction was used for PCR with the primers complementary to HIV-1 gag, JUND, MYC or GAPDH (Supplementary Table 1). For real-time PCR assays to quantify steady-state RNA, 50-ng aliquots of total RNA were reverse transcribed and one-tenth of the 20-μl reaction was subjected to Quantitect SYBR Green PCR (Qiagen) using a Lightcycler (Roche).
Note: Supplementary information is available on the Nature Structural & Molecular Biology website.
* NOTE: In the version of this article initially published online, the label for the x-axis in Figure 5c was incorrect. This error was introduced during the production process for the article. The correct label should read "Time 3H metabolic labeling (min).” The error has been corrected for all versions of the article. We apologize for any inconvenience this may have caused.
Merrick, W.C. & Hershey, J.W.B. The pathway and mechanism of eukaryotic protein synthesis. in Translational Control (eds. Hershey, J.W.B., Mathews, D.H. & Sonenberg, N.) 31–69 (Cold Spring Harbor Laboratory Press, Plainview, New York, USA, 1996).
Kozak, M. An analysis of 5′-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15, 8125–8148 (1987).
Ehrenfeld, E. Initiation of translation by picornavirus RNAs. in Translational Control (eds. Hershey, J.W.B., Mathews, M. & Sonenberg, N.) 549–573 (Cold Spring Harbor Laboratory Press, Plainview, New York, USA, 1996).
Wiegand, H.L., Lu, S. & Cullen, B.R. Exon junction complexes mediate the enhancing effect of splicing on mRNA expression. Proc. Natl. Acad. Sci. USA 100, 11327–11332 (2003).
Nott, A., Le Hir, H. & Moore, M.J. Splicing enhances translation in mammalian cells: an additional function of the exon junction complex. Genes Dev. 18, 210–222 (2004).
Gudikote, J.P., Imam, J.S., Garcia, R.F. & Wilkinson, M.F. RNA splicing promotes translation and RNA surveillance. Nat. Struct. Mol. Biol. 12, 801–809 (2005).
Short, J.D. & Pfarr, C.M. Translational regulation of the JunD messenger RNA. J. Biol. Chem. 277, 32697–32705 (2002).
Yilmaz, A., Bolinger, C. & Boris-Lawrie, K. Retrovirus translation initiation: issues and hypotheses derived from study of HIV-1. Curr. HIV Res. 3, 13 (2006).
Boris-Lawrie, K., Roberts, T.M. & Hull, S. Retroviral RNA elements integrate components of post-transcriptional gene expression. Life Sci. 69, 2697–2709 (2001).
Hull, S. & Boris-Lawrie, K. RU5 of Mason-Pfizer monkey virus 5′ long terminal repeat enhances cytoplasmic expression of human immunodeficiency virus type 1 gag-pol and nonviral reporter RNA. J. Virol. 76, 10211–10218 (2002).
Butsch, M., Hull, S., Wang, Y., Roberts, T.M. & Boris-Lawrie, K. The 5′ RNA terminus of spleen necrosis virus contains a novel posttranscriptional control element that facilitates human immunodeficiency virus Rev/RRE-independent Gag production. J. Virol. 73, 4847–4855 (1999).
Roberts, T.M. & Boris-Lawrie, K. The 5′ RNA terminus of spleen necrosis virus stimulates translation of nonviral mRNA. J. Virol. 74, 8111–8118 (2000).
Pelletier, J. & Sonenberg, N. Insertion mutagenesis to increase secondary structure within the 5′ noncoding region of a eukaryotic mRNA reduces translational efficiency. Cell 40, 515–526 (1985).
Miele, G., Mouland, A., Harrison, G.P., Cohen, E. & Lever, A.M. The human immunodeficiency virus type 1 5′ packaging signal structure affects translation but does not function as an internal ribosome entry site structure. J. Virol. 70, 944–951 (1996).
Roberts, T.M. & Boris-Lawrie, K. Primary sequence and secondary structure motifs in spleen necrosis virus RU5 confer translational utilization of unspliced human immunodeficiency virus type 1 reporter RNA. J. Virol. 77, 11973–11984 (2003).
Dangel, A.W., Hull, S., Roberts, T.M. & Boris-Lawrie, K. Nuclear interactions are necessary for translational enhancement by spleen necrosis virus RU5. J. Virol. 76, 3292–3300 (2002).
Zhang, S. & Grosse, F. Multiple functions of nuclear DNA helicase II (RNA helicase A) in nucleic acid metabolism. Acta Biochim. Biophys. Sin.(Shanghai) 36, 177–183 (2004).
Linder, P. Molecular biology. The life of RNA with proteins. Science 304, 694–695 (2004).
Nakajima, T. et al. RNA helicase A mediates association of CBP with RNA polymerase II. Cell 90, 1107–1112 (1997).
Anderson, S.F., Schlegel, B.P., Nakajima, T., Wolpin, E.S. & Parvin, J.D. BRCA1 protein is linked to the RNA polymerase II holoenzyme complex via RNA helicase A. Nat. Genet. 19, 254–256 (1998).
Aratani, S. et al. Dual roles of RNA helicase A in CREB-dependent transcription. Mol. Cell. Biol. 21, 4460–4469 (2001).
Fujii, R. et al. A role of RNA helicase A in cis-acting transactivation response element-mediated transcriptional regulation of human immunodeficiency virus type 1. J. Biol. Chem. 276, 5445–5451 (2001).
Kernan, M.J., Kuroda, M.I., Kreber, R., Baker, B.S. & Ganetzky, B. napts, a mutation affecting sodium channel activity in Drosophila, is an allele of mle, a regulator of X chromosome transcription. Cell 66, 949–959 (1991).
Kuroda, M.I., Kernan, M.J., Kreber, R., Ganetzky, B. & Baker, B.S. The maleless protein associates with the X chromosome to regulate dosage compensation in Drosophila. Cell 66, 935–947 (1991).
Li, J. et al. A role for RNA helicase A in post-transcriptional regulation of HIV type 1. Proc. Natl. Acad. Sci. USA 96, 709–714 (1999).
Gatfield, D. et al. The DExH/D box protein HEL/UAP56 is essential for mRNA nuclear export in Drosophila. Curr. Biol. 11, 1716–1721 (2001).
Kedersha, N. et al. Stress granules and processing bodies are dynamically linked sites of mRNP remodeling. J. Cell Biol. 169, 871–884 (2005).
Ryder, K., Lanahan, A., Perez-Albuerne, E. & Nathans, D. jun-D: a third member of the jun gene family. Proc. Natl. Acad. Sci. USA 86, 1500–1503 (1989).
van Dam, H. & Castellazzi, M. Distinct roles of Jun: Fos and Jun: ATF dimers in oncogenesis. Oncogene 20, 2453–2464 (2001).
Eferl, R. & Wagner, E.F. AP-1: a double-edged sword in tumorigenesis. Nat. Rev. Cancer 3, 859–868 (2003).
Ricci, R. et al. Distinct functions of junD in cardiac hypertrophy and heart failure. Genes Dev. 19, 208–213 (2005).
Braddock, M. et al. Intron-less RNA injected into the nucleus of Xenopus oocytes accesses a regulated translation control pathway. Nucleic Acids Res. 22, 5255–5264 (1994).
Matsumoto, K., Wassarman, K.M. & Wolffe, A.P. Nuclear history of a pre-mRNA determines the translational activity of cytoplasmic mRNA. EMBO J. 17, 2107–2121 (1998).
Kim, V.N., Kataoka, N. & Dreyfuss, G. Role of the nonsense-mediated decay factor hUpf3 in the splicing-dependent exon-exon junction complex. Science 293, 1832–1836 (2001).
Nott, A., Meislin, S.H. & Moore, M.J. A quantitative analysis of intron effects on mammalian gene expression. RNA 9, 607–617 (2003).
Lu, S. & Cullen, B.R. Analysis of the stimulatory effect of splicing on mRNA production and utilization in mammalian cells. RNA 9, 618–630 (2003).
Mangus, D.A., Evans, M.C. & Jacobson, A. Poly(A)-binding proteins: multifunctional scaffolds for the post-transcriptional control of gene expression. Genome Biol. 4, 223 (2003).
Isken, O. et al. Members of the NF90/NFAR protein group are involved in the life cycle of a positive-strand RNA virus. EMBO J. 22, 5655–5665 (2003).
Svitkin, Y.V. et al. The requirement for eukaryotic initiation factor 4A (elF4A) in translation is in direct proportion to the degree of mRNA 5′ secondary structure. RNA 7, 382–394 (2001).
Lee, C.G. & Hurwitz, J. A new RNA helicase isolated from HeLa cells that catalytically translocates in the 3′ to 5′ direction. J. Biol. Chem. 267, 4398–4407 (1992).
Moore, M.J. From birth to death: the complex lives of eukaryotic mRNAs. Science 309, 1514–1518 (2005).
Tseng, S.S. et al. Dbp5p, a cytosolic RNA helicase, is required for poly(A)+ RNA export. EMBO J. 17, 2651–2662 (1998).
Snay-Hodge, C.A., Colot, H.V., Goldstein, A.L. & Cole, C.N. Dbp5p/Rat8p is a yeast nuclear pore-associated DEAD-box protein essential for RNA export. EMBO J. 17, 2663–2676 (1998).
Chan, C.C. et al. eIF4A3 is a novel component of the exon junction complex. RNA 10, 200–209 (2004).
Ferraiuolo, M.A. et al. A nuclear translation-like factor eIF4AIII is recruited to the mRNA during splicing and functions in nonsense-mediated decay. Proc. Natl. Acad. Sci. USA 101, 4118–4123 (2004).
Chuang, R.Y., Weaver, P.L., Liu, Z. & Chang, T.H. Requirement of the DEAD-Box protein ded1p for messenger RNA translation. Science 275, 1468–1471 (1997).
Yedavalli, V.S., Neuveut, C., Chi, Y.H., Kleiman, L. & Jeang, K.T. Requirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function. Cell 119, 381–392 (2004).
Coller, J. & Parker, R. General translational repression by activators of mRNA decapping. Cell 122, 875–886 (2005).
Butsch, M. & Boris-Lawrie, K. Translation is not required to generate virion precursor RNA in human immunodeficiency virus type 1-infected T cells. J. Virol. 74, 11531–11537 (2000).
Hull, S. & Boris-Lawrie, K. Analysis of synergy between divergent simple retrovirus posttranscriptional control elements. Virology 317, 146–154 (2003).
We are grateful to K. Green-Church and the Ohio State University CCIC proteomics core for mass spectrophotometry, W.C. Merrick for valuable discussion, K. Hayes, I. Younis and members of the K.B.-L. laboratory for comments on the manuscript and T. Vojt for figure preparation. This work was supported by grants from the US National Institutes of Health (P01CA16058 and P30CA100730).
The authors declare no competing financial interests.
About this article
Cite this article
Hartman, T., Qian, S., Bolinger, C. et al. RNA helicase A is necessary for translation of selected messenger RNAs. Nat Struct Mol Biol 13, 509–516 (2006). https://doi.org/10.1038/nsmb1092
Virus Research (2021)
Future Science OA (2020)
Scientific Reports (2020)
The mRNA encoding the JUND tumor suppressor detains nuclear RNA-binding proteins to assemble polysomes that are unaffected by mTOR
Journal of Biological Chemistry (2020)