Human RNPS1 and its associated factors: a versatile alternative pre-mRNA splicing regulator in vivo

Abstract

Human RNPS1 was originally purified and characterized as a pre-mRNA splicing activator, and its role in the postsplicing process has also been proposed recently. To search for factors that functionally interact with RNPS1, we performed a yeast two-hybrid screen with a human cDNA library. Four factors were identified: p54 (also called SRp54; a member of the SR protein family), human transformer 2 beta (hTra2 beta; an exonic splicing enhancer-binding protein), hLucA (a potential component of U1 snRNP), and pinin (also called DRS and MemA; a protein localized in nuclear speckles). The N-terminal region containing the serine-rich (S) domain, the central RNA recognition motif (RRM), and the C-terminal arginine/serine/proline-rich (RS/P) domain of RNPS1 interact with p54, pinin, and hTra2 beta, respectively. Protein-protein binding between RNPS1 and these factors was verified in vitro and in vivo. Overexpression of RNPS1 in HeLa cells induced exon skipping in a model beta-globin pre-mRNA and a human tra-2 beta pre-mRNA. Coexpression of RNPS1 with p54 cooperatively stimulated exon inclusion in an ATP synthase gamma-subunit pre-mRNA. The RS/P domain and RRM are necessary for the exon-skipping activity, whereas the S domain is important for the cooperative effect with p54. RNPS1 appears to be a versatile factor that regulates alternative splicing of a variety of pre-mRNAs.

FIG. 1.

In vitro binding assays of RNPS1 to pinin, hTra2β, and p54 proteins. (A) The indicated recombinant GST-tagged fusion proteins coupled to glutathione Sepharose beads were incubated with in vitro-translated ³⁵S-labeled RNPS1 or luciferase (Luc) as a control, and the pelleted beads were washed and analyzed by 10% SDS-PAGE (GST pull-down assays). The upper panel shows the autoradiograph. A portion (one fifth) of each ³⁵S-labeled protein used in the pull-down reactions was loaded as the input control (Input). Marker protein sizes are indicated on the left. The lower panel shows a 10% SDS-PAGE gel (stained with Coomassie blue) that was used for autoradiography. The corresponding proteins are indicated with arrowheads. (B) Expression of FLAG-tagged constructs (GFP, p54, and hTra2β) in HeLa whole-cell extracts. The extracts (5 μl each) were separated by 10% SDS-PAGE and detected by immunoblotting with anti-FLAG antibody. (C) Immunoprecipitation analyses of ³⁵S-labeled RNPS1 and luciferase (Luc) as a control with the indicated expressed FLAG-fused proteins (GFP as a control). Anti-FLAG antibody immobilized on protein G-Sepharose was used for the immunoprecipitation. Immunoprecipitated ³⁵S-labeled RNPS1 protein was detected by autoradiography. A portion (one fifth) of each ³⁵S-labeled protein used in the immunoprecipitation was loaded as control (Input).

FIG. 2.

Protein interaction domains of RNPS1 analyzed by yeast two-hybrid assays. (A) Yeast host strain EGY48 was cotransformed with the indicated combinations of RNPS1 deletion constructs (in pLexA; see Table 1) and selected clones (p54, hTra2β, and pinin in pB42AD). As a negative interaction control, the empty vector pB42AD (vec.) was used. Protein interactions were detected by β-galactosidase activity and auxotrophy for leucine. (B) A model of protein interactions of RNPS1 with p54, pinin, and hTra2β is shown.

FIG. 3.

Subcellular localization of WT RNPS1 and deletion mutants, p54, and hTra2β. (A) EGFP-fused mouse RNPS1 was transiently transfected into living Ehrlich ascites tumor cells and analyzed by fluorescence microscopy (24 h after transfection). A phase contrast micrograph taken at the same time is shown below. (B) The GFP fusion constructs with either NT or RS/P domain (GFP alone as a control) were transiently transfected into HeLa cells. The cells were fixed and permeabilized at 12 h after transfection and analyzed by fluorescence microscopy. (C) The DsRed fusion proteins of human RNPS1 (WT) and two deletion mutants, RNPS1-ΔS and RNPS1-ΔRS/P, were transiently cotransfected into HeLa cells with FLAG-tagged p54 and FLAG-tagged hTra2β. The cells were fixed and permeabilized at 15 h after transfection, treated with an anti-FLAG antibody, and analyzed by confocal microscopy. Both images were also superimposed, and the yellow color indicates colocalization of the two proteins in nuclear speckles (Merge). Overexposed images (black and white) are also shown to highlight the cytoplasmic localization that is specifically observed in RNPS1-ΔRS/P protein. Bar, 10 μm.

FIG. 4.

Effect of transiently overexpressed RNPS1, p54, and hTra2β on three kinds of alternatively spliced pre-mRNAs. (A) In vivo splicing of a model β-globin DUP51 pre-mRNA. (B) In vivo splicing of an hTra2β minigene pre-mRNA (C) In vivo splicing of a human F₁γ minigene pre-mRNA. The structures of the pre-mRNAs and the spliced products are schematically illustrated on the top of each panel (numbers denote length in nucleotides). HeLa cells were transfected with a constant amount of individual pre-mRNA reporter plasmid along with a constant (indicated by rectangles) or increasing (indicated by triangles) amounts of protein expression plasmids pCMV-3×FLAG-RNPS1 (wild type), pCMV-3×FLAG-RNPS1-ΔS (ΔS), pCMV-3×FLAG-RNPS1-ΔRS/P (ΔRS/P), pCMV-3×FLAG-RNPS1-mtRRM (mtRRM), pCMV-3×FLAG-RNPS1-RS/P (RS/P), pFLAG-p54 (p54), and pFLAG-hTra2β (hTra2β). An empty vector, pFLAG-CMV-2, was used to keep the total amount of transfected DNA constant. Spliced products were analyzed by RT-PCR with specific DNA primers (indicated by arrows), followed by agarose gel electrophoresis. The length (in nucleotides) of splicing products (A) or specific alternatively spliced isoforms (B and C) are indicated on the right of the panels. The bands above the beta 1 (lanes 4, 5, and 12 to 16 in B) and non-muscle type (* in C) were found as nonspecific amplified products unrelated to splicing (data not shown) (42).