A novel role for shuttling SR proteins in mRNA translation

Abstract

The Ser-Arg-rich (SR) proteins comprise a large family of nuclear phosphoproteins that are required for constitutive and alternative splicing. A subset of SR proteins shuttles continuously between the nucleus and the cytoplasm, suggesting that the role of shuttling SR proteins in gene expression may not be limited to nuclear pre-mRNA splicing, but may also include unknown cytoplasmic functions. Here, we show that shuttling SR proteins, in particular SF2/ASF, associate with translating ribosomes and stimulate translation when tethered to a reporter mRNA in Xenopus oocytes. Moreover, SF2/ASF enhances translation of reporter mRNAs in HeLa cells, and this activity is dependent on its ability to shuttle from the nucleus to the cytoplasm and is increased by the presence of an exonic-splicing enhancer. Furthermore, SF2/ASF can stimulate translation in vitro using a HeLa cell-free translation system. Thus, the association of SR proteins with translating ribosomes, as well as the stimulation of translation both in vivo and in vitro, strongly suggest a role for shuttling SR proteins in translation. We propose that shuttling SR proteins play multiple roles in the posttranscriptional expression of eukaryotic genes and illustrate how they may couple splicing and translation.

Figure 1.

Shuttling SR proteins are associated with the translation machinery. (A) HeLa cell cytosolic extracts were fractionated across a 10%-50% sucrose gradients and analyzed by Western blotting with antibodies against SF2/ASF, SRp20, poly(A)-binding protein (PABP), and the ribosomal protein, rpS6. (Top) UV absorbance (254 nm) profile of cytosolic ribonucleoprotein complexes. (B) Association of SF2/ASF with polyribosomes requires intact ribosomal particles. Western blot analysis of RNA-binding proteins associated with polyribosomes, purified by pelleting through a 10%-50% sucrose gradient from mock treated (-) or EDTA-treated (+) HeLa cytosolic extracts. Unfractionated cytosolic extract is shown in lanes designated by C.

Figure 2.

Tethered SR proteins stimulate translation in Xenopus oocytes. (A) mRNAs encoding either the MS2 protein alone, or fusions between MS2-U1A, MS2-SF2/ASF, and MS2-PABP were coinjected into the cytoplasm of Xenopus oocyctes with a luciferase reporter mRNA containing MS2p-binding sites within its 3′UTR and a β-galactosidase mRNA lacking MS2-binding sites. The injected fusion proteins did not significantly affect β-galactosidase levels and nonspecific effects of the fusion proteins on translation were taken into account by normalizing luciferase activity to β-galactosidase activity. These data represent the average stimulation from three independent experiments. (B) Northern blot analysis showing constant levels of mRNA reporter upon injection of MS2-U1A (lanes 1,2) and MS2-SF2/ASF (lanes 3,4). (t) Time after injection (0 and 16 h, respectively). (C) The effect of the MS2-SF2/ASF fusion protein is specific to reporter mRNAs containing cognate MS2-binding sites. Microinjection experiments were performed using a luciferase reporter mRNA containing (white bars) or lacking (black bars) functional MS2-binding sites. (D) The RS domain of SR proteins is sufficient to stimulate translation in vivo. mRNAs encoding MS2, MS2-U1A, MS2-SF2/ASF, MS2-SF2 FF-DD, MS2-RS SF2/ASF, or MS2-RS SC35 were injected into the cytoplasm of Xenopus oocytes along with reporter mRNAs as described above. These data represents the average stimulation from three independent experiments.

Figure 3.

SF2/ASF stimulates translation of a luciferase reporter mRNA in HeLa cells. (A) Schematic diagrams of the pLCS reporter system. The fibronectin EDA ESE or a mutant version were inserted in frame and upstream of the Firefly luciferase ORF. After cotransfection into HeLa cells with a Renilla luciferase reporter (driven by the thymidine kinase promoter), the pLCS reporters are transcribed under the control of the SV40 promoter and translation leads to the synthesis of the reporter enzymes. The promega Dual Luciferase Reaction (DLR) system is used to assay levels of the Firefly (translational reporter) and Renilla (to control for transfection efficiency) luciferase levels. (B) The presence of an ESE stimulates expression of a Firefly Luciferase reporter mRNA. HeLa cells were transiently transfected with pLCS-Stop, pLCS-EDA, or pLCS-EDA^mt, which contains a mutation in the EDA sequence along with the nonspecific Renilla luciferase reporter. The data are expressed as a ratio of Firefly luciferase activity to Renilla luciferase activity, and have been normalized to changes in nuclear and cytoplasmic distribution of the reporter mRNAs. (C) Overexpression of SF2/ASF stimulates expression of reporter mRNAs in vivo. HeLa cells were transiently transfected with pLCS-Stop, pLCS-EDA, or pLCS-EDA^mt along with the nonspecific Renilla luciferase reporter and pCGT7-SF2/ASF and analyzed as described above. (D) Analysis of the nucleocytoplasmic distribution of reporter mRNAs (top) and of the cytosolic levels of reporter mRNA relative to the endogenous rpS8 mRNA (bottom) by RT-PCR. (E) Table summarizing the effects of the EDA ESE and overexpression of SF2/ASF on LCS reporter mRNA distribution, cytoplasmic accumulation, and reporter enzyme expression.

Figure 4.

SF2/ASF stimulates the translation of both intronless and intron-containing reporter mRNAs. (A) Schematic diagrams of the intronless and intron-containing LCS reporter mRNAs. The fibronectin EDA ESE or a mutant version were inserted in-frame and upstream of the Firefly luciferase ORFs. (Right) Note the presence of an heterologous intron in the 5′UTR. (B) The presence of an ESE (EDA ESE) stimulates translation of both intronless and intron-containing Firefly Luciferase reporter mRNAs. HeLa cells were transiently transfected with pLCS-Stop, pLCS-EDA, or pLCS-EDA^mt, which contains a mutation in the EDA sequence along, lacking or containing an intron, with the nonspecific Renilla luciferase reporter. The data are expressed as a ratio of Firefly luciferase activity to Renilla luciferase activity. (C) Overexpression of SF2/ASF dramatically increases the expression of both intronless and intron-containing LCS reporter mRNAs in HeLa cells. HeLa cells were transiently transfected with pLCS-Stop, pLCS-EDA, or pLCS-EDA^mt with or without an intron along with the nonspecific Renilla luciferase reporter and pCGT7-SF2/ASF and analyzed as described above.

Figure 5.

Nucleocytoplasmic shuttling of SF2/ASF is important for stimulation of translation in vivo. (A) HeLa cells were cotransfected with pLCS-EDA and empty vector (black bar), pCGT7-SF2/ASF (gray bar), pCGT7-SF2/ASF-NRS (white bar), or pCGT7-SC35 (hatched bar). Dual Luciferase assays measured the effect of shuttling on the translation of the reporter enzymes. These data represent the average stimulation from three independent experiments. (B) Analysis of the nucleocytoplasmic distribution of the reporter mRNAs (top) or of the cytosolic levels of reporter mRNA relative to the endogenous rpS8 mRNA (bottom) by RT-PCR. (C) Table summarizing the effects of overexpression of SF2/ASF, SF2/ASF-NRS, and SC35 on LCS reporter mRNA distribution, cytoplasmic accumulation, and reporter enzyme expression.

Figure 6.

Effect of shuttling SR proteins on the activation of translation of a luciferase reporter mRNA in HeLa cells. (A) Schematic diagrams of the pLCS reporter constructs containing in-frame ESEs recognized by SF2/ASF and 9G8, SRp20 and SC35. (B) The presence of the EDA ESE, but not the SRp20 ESE nor the SC35 ESE, stimulates expression of a Firefly Luciferase reporter mRNA. HeLa cells were transiently transfected with pLCS-EDA, with pLCS-EDA^mt, which contains a mutation in the EDA sequence, or with pLCS SRp20 ESE and pLCS SC35 ESE along with the nonspecific Renilla luciferase reporter. The data are expressed as a ratio of Firefly luciferase activity to Renilla luciferase activity. (C) Overexpression of SF2/ASF, but not of 9G8, SRp20, SC35, or hnRNP A1, stimulate expression of the LCS EDA reporter mRNA in vivo. HeLa cells were cotransfected with pLCS-EDA and empty vector (control), pCGT7-SF2/ASF, pCGT7-9G8, pCGT7-SRp20, pCGT7-SC35, and pCGT7-hnRNP A1 along with the nonspecific Renilla luciferase reporter and pCGT7-SF2/ASF and analyzed as described above. (D) Table summarizing the effects of overexpression of SR proteins and hnRNP A1 on LCS reporter mRNA distribution, cytoplasmic accumulation, and reporter enzyme expression.

Figure 7.

Overexpression of SF2/ASF enhances translation of reporter mRNAs. RT-PCR analysis of LCS EDA reporter mRNA levels from HeLa cells transfected with pLCS EDA and either empty expression vector (control) or pCGT7-SF2/ASF (SF2/ASF) fractionated across a 10%-50% sucrose gradient as described in Figure 1. Each transfection also contained the Renilla luciferase reporter.

Figure 8.

ESEs and recombinant SF2/ASF stimulate translation of reporter mRNAs in vitro. (A,B) Insertion of the EDA ESE into the luciferase ORF stimulates translation in vitro. Constructs described in Figure 3A were transcribed in vitro. Reporter mRNAs (200 ng) containing either one or three copies of a wild-type or mutant version of the EDA ESE were incubated in a HeLa translation extract. Following incubation at 37°C, luciferase assays were performed (Promega). (C) Recombinant SF2/ASF stimulates expression of LCS reporter mRNAs in vitro. In vitro translation reactions were performed in the presence (black bars) or absence (white bars) of recombinant SF2/ASF (200 ng), as described above. (D) Insertion of the EDA ESE into the 3′UTR enhances translation of the luciferase reporter mRNA in vitro. Reporter mRNAs (100 ng) containing either six copies of a wild-type or mutant EDA ESE in the 3′UTR were incubated in HeLa translation extract. Following incubation at 37°C, luciferase assays were performed.