Novel splicing factor RBM25 modulates Bcl-x pre-mRNA 5' splice site selection

Abstract

RBM25 has been shown to associate with splicing cofactors SRm160/300 and assembled splicing complexes, but little is known about its splicing regulation. Here, we characterize the functional role of RBM25 in alternative pre-mRNA splicing. Increased RBM25 expression correlated with increased apoptosis and specifically affected the expression of Bcl-x isoforms. RBM25 stimulated proapoptotic Bcl-x(S) 5' splice site (5' ss) selection in a dose-dependent manner, whereas its depletion caused the accumulation of antiapoptotic Bcl-x(L). Furthermore, RBM25 specifically bound to Bcl-x RNA through a CGGGCA sequence located within exon 2. Mutation in this element abolished the ability of RBM25 to enhance Bcl-x(S) 5' ss selection, leading to decreased Bcl-x(S) isoform expression. Binding of RBM25 was shown to promote the recruitment of the U1 small nuclear ribonucleoprotein particle (snRNP) to the weak 5' ss; however, it was not required when a strong consensus 5' ss was present. In support of a role for RBM25 in modulating the selection of a 5' ss, we demonstrated that RBM25 associated selectively with the human homolog of yeast U1 snRNP-associated factor hLuc7A. These data suggest a novel mode for Bcl-x(S) 5' ss activation in which binding of RBM25 with exonic element CGGGCA may stabilize the pre-mRNA-U1 snRNP through interactions with hLuc7A.

FIG. 1.

RBM25 localizes to splicing factor-rich nuclear speckles through the ER domain. (A) Schematic diagram of the structural organization of RBM25. RBM25 is composed of a proline-rich region (P) and an RRM at the amino-terminal end, an RE/RD-rich (ER) domain at the central region, and a PWI domain at the carboxyl-terminal end. (B) Nuclear speckle localization of RBM25 in response to transcriptional inhibition by DRB. HeLa cells were grown in the absence or presence of 100 μM DRB for 2 h at 37°C, processed for immunofluorescence analysis with Abs directed against RBM25 or SC35, and revealed with an Alexa- or fluorescein isothiocyanate-conjugated secondary Ab, respectively. DAPI stains DNA. Bars, 5 μm. (C) The ER domain is critical for RBM25 localization to nuclear speckles. The full length (FL) or the RRM, ER, or PWI domain of RBM25, fused with pEGFP, was transfected into HeLa cells. The expressed EGFP fusion proteins were analyzed for localization relative to splicing factor SC35. Cells were fixed and stained with anti-SC35 Ab and DAPI. Bars, 5 μm.

FIG. 2.

Increased RBM25 expression correlates with induction of apoptosis. HEK293 cells were transfected with EGFP or RBM25/EGFP. The development of apoptosis was assessed by annexin V binding of GFP-positive cells, by activated (cleaved) caspase 3 immunofluorescent staining, and by DAPI staining scores of nuclear fragmentation. (A) Percentage of cells positive for annexin V-phycoerythrin in the GFP-positive populations. Annexin V staining was performed 48 h after transfection. Data are the mean ± SD of three separate experiments. (B) Percentage of cells positive for activated caspase 3 in the GFP-positive population detected by immunofluorescent staining with anti-caspase 3 Ab 48 h after transfection. At least 500 GFP-positive cells for each sample were counted in each experiment. Data are the mean ± SD of three independent experiments. (C) Percentage of cells positive for nuclear fragmentation in the GFP-positive population. Cells were fixed 48 h after transfection, stained with DAPI, and analyzed for DAPI-stained nuclear fragmentation. Data were obtained by analyzing at least 300 GFP-positive cells for each sample and represent the mean ± SD of three separate experiments. (D) Nuclear fragmentation in cells 72 h after transfection as visualized by DAPI staining. The arrows indicate typical apoptotic cells with RBM25 accumulation or fragmented nuclei. Bars, 10 μm. (E) Western blot analysis of exogenously expressed RBM25-GFP protein at 48 h posttransfection in HEK293 cells. Thirty-five micrograms of cell lysate was fractionated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and detected with an anti-RBM25 Ab.

FIG. 3.

Expression and localization of RBM25 during staurosporine-stimulated apoptosis in HEK293 cells. (A) Cellular localization of RBM25 at the indicated time points after staurosporine stimulation. HEK293 cells were mock treated or treated with 500 nM staurosporine and stained with anti-RBM25 Ab and DAPI. Bars, 10 μm. (B) Western blot analysis of RBM25 protein levels in HEK293 cells at the indicated time points after either mock treatment (−) or treatment with staurosporine (+). Thirty-five micrograms of cell lysate was fractionated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and detected with an anti-RBM25 Ab. β-Actin served as a loading control.

FIG. 4.

Effect of RBM25 on alternative splicing of selected apoptotic factors. pCDNA3.1-HA-RBM25-transfected or RBM25 shRNA-depleted HeLa cells were analyzed for alternative splicing patterns of exon 6 of caspase 3, exon 2 of Mcl1, Bcl-x_S and Bcl-x_L of Bcl-x, and exon 6 of Fas by RT-PCR with the respective primer sets. (A) RNA isolated from cells transfected with pCDNA3.1-HA-RBM25 were analyzed for splicing patterns of the indicated genes. Expression of HA-RBM25 was detected in a Western blot assay with an anti-HA Ab. β-Actin served as a loading control. (B) RNA isolated from RBM25 sh-82-depleted cells were analyzed for splicing patterns of the indicated genes. Expression of RBM25 was detected in a Western blot assay with an anti-RBM25 Ab. β-Actin served as a loading control.

FIG. 5.

RBM25 regulates Bcl-x 5′ ss selection. (A) Schematic diagram of the Bcl-x minigene spanning the entire alternatively spliced region from exon 1 to exon 3, with a shortened intron 2. Two splice variants derived from the Bcl-x gene, proapoptotic Bcl-x_S and antiapoptotic Bcl-x_L, are produced via alternative 5′ ss selection within exon 2. (B) Overexpression of RBM25 increased levels of proapoptotic Bcl-x_S in a dose-dependent manner. HeLa cells were transfected with 0.5 μg of Bcl-x minigene and increasing amounts of pcDNA3.1-HA-RBM25 (0 to 1.5 μg) and harvested 24 h after transfection. Semiquantitative RT-PCR was performed to determine the relative abundance of Bcl-x_S and Bcl-x_L mRNAs by densitometric analysis. Mean values ± SD of three independent experiments are shown. Anti-HA Western blotting indicates the expression levels of exogenous RBM25 proteins. β-Actin served as a loading control. The bar graph presents the densitometric analyses of the ratio of Bcl-x_S to Bcl-x_L from the three experiments performed (mean ± SD).

FIG. 6.

Depletion of RBM25 by RNA interference leads to reduction of proapoptotic Bcl-x_S. (A) Schematic of RBM25 regions targeted by RBM25 shRNAs (sh81, sh82, sh83, and sh84). P, proline-rich region; RRM, RNA recognition motif; ER, glutamic acid/arginine-rich domain; PWI, proline-tryptophan-isoleucine domain. (B) HeLa cells were transfected with RBM25 shRNAs or nonsilencing controls. Semiquantitative RT-PCR was performed to determine alternatively spliced forms of Bcl-x 24 h after transfection. RBM25 expression levels in shRNA-treated cells relative to nonsilenced cells are shown at the bottom. Western blotting indicates the expression levels of endogenous RBM25 proteins. β-Actin served as a loading control. The bar graph presents the densitometric analyses of the ratio of Bcl-x_S to Bcl-x_L from the three experiments performed (mean ± SD).

FIG. 7.

A sequence motif within exon 2 is important for regulated splicing of Bcl-x by RBM25. (A) A diagram indicating mutated sequences and nucleotides replaced in the Bcl-x minigene tested in the experiment. (B) Analysis of the effect of mutations on Bcl-x splice site selections in response to RBM25. The minigene and its mutated constructs were transfected into HeLa cells either with a vector alone or with an RBM25-expressing plasmid and analyzed for the expression of Bcl-x isoforms. An anti-HA Ab was used to detect expression levels of transfected RBM25 proteins. β-Actin served as a loading control. The bar graph presents the densitometric analyses of the ratio of Bcl-x_S to Bcl-x_L in the three experiments performed (mean ± SD). −, no RBM25; +, with RBM25.

FIG. 8.

RBM25 binds to Bcl-x RNA through the exonic sequence CGGGCA. (A) Bcl-x RNAs detected by RIP with an anti-RBM25 Ab. HeLa cells were fixed with 1% formaldehyde, and RIP was carried out with the cross-linked cell lysate and anti-RBM25 Ab or control rabbit IgG. PCR was performed with Bcl-x or Mcl1 primers on RNA retrieved by RIP with or without the RT reaction. The input lysate served as a positive control. (B) RNA probes consist of the WT or mutated (Mut) exonic sequences used in gel mobility shift assays. (C) Gel mobility shift assays were performed with the biotinylated WT or mutant probes and purified RBM25 proteins. Increasing amounts of RBM25 were incubated with the RNA probes, fractionated in a native 5% polyacrylamide gel, transferred to Hybond-N+ nylon membrane, and detected with a LightShift chemiluminescent electrophoretic mobility shift assay kit as described in Materials and Methods. Arrow, probe-protein complex. (D) For competition assay, a 1-, 5-, or 20-fold molar excess of unlabeled WT or mutant RNA was added to binding reaction mixtures. −, the probe incubated in the absence of competitors. Arrow, probe-protein complex.

FIG. 9.

Effect of CGGGCA on E1A reporter gene 5′ ss selection in vivo. (A) Schematic diagram of the E1A minigene and its major splicing products. E1A+WT and E1A+Mu4 constructs were created in which the WT (UGGGCA) or the Mu4 (TTAGAG) sequence was inserted at a position 64 to 69 nt upstream of the 9S 5′ ss, respectively. (B) In vivo splicing assays were performed with HeLa cells transfected with the E1A+non, E1A+WT, or E1A+Mu4 construct in the presence or absence of the expression vector pCDNA3.1-HA-RBM25. RNAs were isolated 24 h posttransfection and analyzed for E1A splicing products. Anti-HA Western blotting indicates the expression levels of exogenous RBM25 proteins. β-Actin served as a loading control. (C) The relative abundance of spliced mRNA species was analyzed by densitometric analysis. −, no RBM25; +, with RBM25. Mean values ± SD of three independent experiments are shown. The bar graph presents the densitometric analyses of the expression levels of 13S, 12S, and 9S from the three experiments performed (mean ± SD).

FIG. 10.

RBM25 associates with hLuc7A and promotes U1 snRNP binding to a weak 5′ ss in a CGGGCA-dependent manner. (A) The WT, CGGGCA-to-TTAGAG mutant (Mu4), or consensus 5′ ss mutant (5′cons) Bcl-x minigene with the mutation sequence indicated in the diagram. (B) Weak 5′ ss is central in establishing CGGGCA-dependent recruitment of U1 snRNP by RBM25. Radioactively labeled Bcl-x_S substrates were incubated in HeLa cell nuclear extracts in the absence (−) or presence (+) of psoralen (psor), irradiated with 365-nm UV light, and fractionated on a denaturing polyacrylamide gel. Purified RBM25 was also added to the reaction mixtures as indicated. The position of the U1 snRNA/Bcl-x substrate cross-link is indicated. (C) Effect of RBM25 on WT or consensus Bcl-x_S 5′ ss (5′cons) minigene splicing. The minigenes were cotransfected with the empty vector or RBM25 cDNA into HeLa cells. Bcl-x isoforms were detected by RT-PCR. Anti-HA Ab was used to detect the expression levels of transfected RBM25 proteins. β-Actin served as a loading control. (D) Co-IP of RBM25 with hLuc7A by anti-RBM25 Ab. Western blot analysis of coimmunoprecipitated hLuc7A in the presence (+) or absence (−) of RNase. Preimmune rabbit IgG served as a control. S, IP supernatant; P, IP beads. (E) Association of RBM25 and hLuc7A analyzed in a GST pull-down assay. (Top) Pull-down assay of RBM25 with GST fusion proteins of full-length hLuc7A (7A-FL), its N-terminal half (7A-N), and its C-terminal half (7A-C). In, input lysates. (Bottom) Coomassie blue staining of purified GST-hLuc7A fusion proteins used in the experiment and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. (F) The localization of endogenous RBM25 and pCGT7-hLuc7A was detected with anti-RBM25 and anti-T7 Ab, respectively. Bar, 5 μm.