Roles of hnRNP A1, SR proteins, and p68 helicase in c-H-ras alternative splicing regulation

Abstract

Human ras genes play central roles in coupling extracellular signals with complex intracellular networks controlling proliferation, differentiation, and apoptosis, among others processes. c-H-ras pre-mRNA can be alternatively processed into two mRNAs due to the inclusion or exclusion of the alternative exon IDX; this renders two proteins, p21H-Ras and p19H-RasIDX, which differ only at the carboxy terminus. Here, we have characterized some of the cis-acting sequences and trans-acting factors regulating IDX splicing. A downstream intronic silencer sequence (rasISS1), acting in concert with IDX, negatively regulates upstream intron splicing. This effect is mediated, at least in part, by the binding of hnRNP A1. Depletion and add-back experiments in nuclear extracts have confirmed hnRNP A1's inhibitory role in IDX splicing. Moreover, the addition of two SR proteins, SC35 and SRp40, can counteract this inhibition by strongly promoting the splicing of the upstream intron both in vivo and in vitro. Further, the RNA-dependent helicase p68 is also associated with both IDX and rasISS1 RNA, and suppression of p68 expression in HeLa cells by RNAi experiments results in a marked increase of IDX inclusion in the endogenous mRNA, suggesting a role for this protein in alternative splicing regulation.

FIG. 1.

Intron D1 splicing is repressed by a downstream ISS. (A) Map of the region around alternative exon IDX, with exons shown as boxes and introns shown as lines. Restriction sites indicate the start or end of the different constructs. Pre-mRNA 1N, pre-mRNA 2N, and pre-mRNA N encompass the NcoI/PstI region, the NheI/NotI region, and the whole NcoI/NotI fragment, respectively. (B) Left, in vitro splicing assays of the constructs depicted in the upper diagram; odd lanes are negative controls without ATP. Band identity was assessed by extensive RT-PCR and/or primer extension of all lariats and products (drawings in the margins). % spliced, ratio of spliced RNA to initial pre-mRNA. Right, spliceosome complex formation on the 1N and 1NΔD2 substrates. The H, A, and B/C complexes are indicated on the right of the gel. Lanes 1 and 2, construct 1N in the presence of 0.8 and 1.6 ¨g of heparin (hep)/¨l as unspecific competitor. Lanes 3 and 4, construct 1NΔD2 in the presence of 0.8 and 1.6 mg of heparin/¨l. (C) Left, in vitro splicing assays of the N and NΔISS1 substrates depicted in the upper diagram. Lane 1, size markers (M) (plasmid pBR322 cut with HpaII and 59 end labeled); lanes 2 and 3, construct N; lanes 4 and 5, construct NΔISS1; lanes 2 and 4, negative controls without ATP. Right, in vitro splicing assays of N(IDX→E3) and N(IDX→E4a) depicted in the upper diagram. Lane 1, size markers (M); lanes 2 and 3, substrates N; lane 4, N(IDX→E3); lane 5, N(IDX→E4a). E3 and E4a are 92 and 101 nt long. Lane 2, negative control without ATP. Band identity is indicated in the right of each panel.

FIG. 2.

The inhibitory sequence within rasISS1 maps to nt 2731 to 2750. (A) rasISS1 covers the region 2721 to 2772 nt downstream of exon IDX. The upper diagram shows the five regions (1 to 5) into which rasISS1 was divided and the corresponding deleted substrates assayed in in vitro splicing assays. The five sequences deleted and the five substrates are depicted. In addition, pre-mRNAs 1N-ISS1Δ23 (lacking nt 2731 to 2750) and 1N-ISS1Δ234 (lacking nt 2731 to 2760) are also shown. (B) Left, in vitro splicing assays of the five consecutive substrates deleted and the undeleted substrate 1N. Odd lanes, negative controls without ATP. Quantification of each splicing reaction (% spliced) is shown below each lane. Right, splicing of pre-mRNAs 1N-ISS1Δ23 (lanes 5 and 6) and 1N-ISS1Δ234 (lanes 7 and 8), in comparison to that of the undeleted pre-mRNA 1N (lanes 1 and 2) and the substrate lacking the whole rasISS1 sequence (pre-mRNA 1NΔD2) (lanes 3 and 4). Odd lanes, controls without ATP. (C) RT-PCR analysis of total RNA extracted from HeLa cells transiently transfected with pcDNA3 carrying the different minigenes shown (lanes 2 to 5) or with the empty vector (lane 6). Lane 1, molecular size markers (M) with sizes indicated at left. u, unspliced; s, spliced.

FIG. 3.

rasISS1 forms a specific complex in HeLa nuclear extracts and binds hnRNP A1. (A) Sequences corresponding to ISS1, ISS1Δ23, and IDX RNA. The sequence deleted in ISS1Δ23 RNA with respect to ISS1 is depicted as a thin line. Nucleotides in bold type in ISS1 represent one putative A1-binding motif. (B) Electrophoretic mobility shift assay performed with homogeneously labeled (*) ISS1 or ISS1Δ23 RNA in nuclear extract (NE), either alone or in competition conditions with increasing amounts of cold competitor RNAs. Lanes 1 and 11, free ISS1 and ISS1Δ23 RNA without extract; lanes 2 and 12, each RNA incubated with NE; lanes 3 to 6, ISS1 reactions with increasing amounts (2, 4, 8, and 16 pmol, respectively) of cold-transcribed ISS1 RNA added; lanes 7 to 10, ISS1 reactions with equivalent amounts of cold ISS1Δ23 added; lanes 13 to 16, ISS1 reactions with equivalent amounts of cold IDX RNA added. Competitor RNAs were preincubated for 10 min in the extract before addition of the labeled ISS1. (C) Left, Coomassie staining of RNA affinity-purified factors from HeLa nuclear extracts separated by 12.5% SDS-polyacrylamide gel electrophoresis. Agarose beads, either coupled to ISS1 RNA (lane 2) or ISS1Δ23 (lane 3) or alone (lane 4), were incubated with nuclear extract. The arrow indicates a specific ISS1-binding protein. The numbers on the left indicate protein molecular mass standards in kilodaltons. Right, Western blot analysis of the protein gel shown on the left with the anti-hnRNPA1 monoclonal antibody (Ab) 4B10.

FIG. 4.

Reconstitution of 1N splicing inhibition by addition of hnRNP A1 recombinant protein. (A) Gel mobility shift assay with labeled ISS1 or ISS1Δ23 RNAs and nuclear extract (NE) or recombinant GST-A1 or GST-UP1. RNA probes, indicated above the autoradiogram, were incubated either with nuclear extract (lanes 1 and 8), with 2, 4, or 8 pmol of recombinant protein GST-A1 (lanes 2 to 4 and 9 to 11), or with the same increasing amounts of recombinant GST-UP1 (lanes 5 to 7 and 12 to 14). Migration of the free RNA probes and the complexes is indicated on the left. (B) Depletion of hnRNP A1 from the HeLa nuclear extract. Left, Coomassie staining of HIV-ESS affinity-purified factors. Lanes 1 and2, eluted factors of the first and second rounds of depletion from nuclear extract incubated with HIV-ESS-coupled beads; lanes 3 and 4, 5 μl of mock-depleted extract (incubated with uncoupled agarose beads) or of the A1-depleted nuclear extract. The band corresponding to hnRNP A1 is indicated. Right, Western blot, performed with monoclonal antibody (Ab) 4B10, of the SDS-polyacrylamide gel on the left. The lane numbers correspond to the same samples on both panels. (C) Reconstitution of splicing inhibition by addition of recombinant (GST-fused) hnRNP A1. In vitro splicing reactions were performed with the HeLa NE depleted twice with HIV-ESS-bound beads (A1-depleted NE) or mock-depleted NE as shown in panel B. Lanes 7 to 10 and 17 to 20, splicing reactions performed in A1-depleted NE supplemented with 200 ng of total HeLa cell SR proteins in the absence or presence of 4, 8, or 16 pmol of purified recombinant hnRNP A1; lanes 2 to 5 and 12 to 15, identical reactions with mock-depleted NE without additional SR proteins; lanes 1 and 11, negative controls without ATP; lanes 6 and 16, control reaction for the A1-depleted NE without additional SR proteins. hnRNP A1 was added to the reaction mixtures prior to the addition of the splicing substrates.

FIG. 5.

SC35 and SRp40 activate the splicing of 1N pre-mRNA in vitro and in vivo. (A) In vitro splicing of 1N and 1N-ISS1Δ23 pre-mRNAs with increasing amounts of total HeLa native SR proteins. Lanes 4 to 6 and 9 to 11, SRs (100, 200, and 400 ng) were assayed; lanes 2 and 7, control reactions without ATP; lanes 3 and 8, control reactions without additional SRs; lane 1, size markers (M). Quantitation of each reaction is shown below each lane. (B) In vitro splicing assays of 1N and 1N-ISS1Δ23 pre-mRNAs with total or individual baculovirus-expressed SR proteins. Lanes 3 and 16, 100 ng of total native SRs added to the splicing reaction; lanes 4 and 17, 200 ng of total native SRs added to the splicing reaction; all other lanes, 8 and 16 pmol of each recombinant baculovirus-expressed SR protein assayed in each couple of reactions, with the identity of theproteins indicated above each lane; lanes 2 and 15, no extra SR proteins added; lane 1, size markers (M). The quantitation of each reaction is shown below each lane. (C) Gel shift assay performed with homogeneously labeled IDX RNA and increasing amounts (3, 6, and 12 pmol) of recombinant baculovirus expressed SC35 (lanes 2 to 4), SRp40 (lanes 5 to 7), ASF/SF2 (lanes 8 to 10), and hnRNP A1 (lanes 11 to 13). Lane 1, labeled RNA only. (D) RT-PCR analysis of total RNA extracted from HeLa cells transiently cotransfected with pCG carrying the different SR proteins shown (lanes 3 to 6) and pcDNA3 carrying 1N minigene; lane 1, molecular weight markers (M); lane 2, control cotransfection with the empty vector pCG; lane 7, control transfection with vector pCG alone. We have checked that all constructs express significant amounts of each SR protein by Western blot analysis, using an anti-T7 tag antibody, of the HeLa cell lysates transiently transfected with the T7-tagged SR proteins (not shown).

FIG. 6.

FUS/TLS, hnRNP H, and the ATP-dependent RNA helicase p68 are associated to rasISS1 and IDX. (A) Coomassie blue staining of RNA affinity-purified factors from HeLa nuclear extracts separated by 8% SDS-polyacrylamide gel electrophoresis. Agarose beads coupled to rasISS1 RNA (lane 2), ISS1Δ23 (lane 3), or IDX (lane 4) or alone (lane 5) were incubated with nuclear extract. Lane 1, size markers (M). hnRNP A1 band is indicated (lane 2). Four bands were excised from the gel and microsequenced (P1 to P4). (B) Identification of the four protein bands. MM, molecular mass. Protein identity was assessed either by matrix-assisted laser desorption time-of-flight mass spectrometry (MALDI-TOF) or by nanospray ion trap tandem mass spectrometry (nESI-ITMS/MS). (C) Western blot of the gel in panel A assayed with each of the specific antibodies depicted on the right. The values on the left indicate the molecular mass in kilodaltons.

FIG. 7.

RNA interference of p68 RNA helicase in HeLa cells promotes an increase in p19H-ras mRNA. (A) Western blot analysis of HeLa cell lysates transiently expressing the plasmid pSUPER-p68 (lane 2) or the empty vector (lane 1). Cells were harvested 48 h posttransfection, and the blot was probed with the monoclonal antibodies PAb204 (anti-p68) and Sigma (mouse anti-α-actin) as a loading control. (B) Northern blot analysis of total RNA (5 μg) from HeLa cells transiently transfected with construct pSUPER-p68 (lanes 2 and 3) or vector alone (lane 1). RNA was extracted 24 h (lane 2) or 48 h (lanes 1 and 3) posttransfection. The identity of each probe used is described on the right. mRNA levels shown below were determined by normalizing against the levels of α-actin mRNA, which was assessed in the same blot. The level of each mRNA in the control sample was arbitrarily given a value of 1.

FIG.8.

IDX and ISS1 act in concert to repress 1N splicing. (A) Conservation of rasISS1 sequences across species. The alignment and comparison of rasISS1 sequences from human (ACN V00574), hamster (ACN M84166.1), rat (ACN M13011), and mouse (AF081118) sources was performed with MacMolly Tetra version 3.10. (B) Prediction of the secondary structure of IDX-rasISS1 obtained from the same species and folded by the Mfold program, version 3.1. It is hypothesized that an IDX sequence is seized by a long interaction with the intronic rasISS1. Shown are the IDX sequence (green) and contiguous and downstream 5′ SS and rasISS1 (black). Panels show human (i) (ΔG = −63.7 kcal/mol), hamster (ii) (ΔG = −61.4 kcal/mol), rat (iii) (ΔG = −62.4 kcal/mol), and mouse (iv) (ΔG = −56.5 kcal/mol) structures. (C) Competing IDXISS1 RNA can restore splicing activity of 1N substrate. In vitro splicing assay of 1N pre-mRNA was performed in the presence of 5, 10, 20, or 40 pmol of cold IDX RNA (lanes 2 to 5), ISS1 RNA (lanes 6 to 9), or IDXISS1 RNA (lanes 10 to 13).