Title: TDP-43 Binds Heterogeneous Nuclear Ribonucleoprotein A/B through Its C-terminal Tail
Abstract: TDP-43 is a highly conserved nuclear factor of yet unknown function that binds to ug-repeated sequences and is responsible for cystic fibrosis transmembrane conductance regulator exon 9 splicing inhibition. We have analyzed TDP-43 interactions with other splicing factors and identified the critical regions for the protein/protein recognition events that determine this biological function. We show here that the C-terminal region of TDP-43 is capable of binding directly to several proteins of the heterogeneous nuclear ribonucleoprotein (hnRNP) family with well known splicing inhibitory activity, in particular, hnRNP A2/B1 and hnRNP A1. Mutational analysis showed that TDP-43 proteins lacking the C-terminal region could not inhibit splicing probably because they were unable to form the hnRNP-rich complex involved in splicing inhibition. Finally, through splicing complex analysis, we show that splicing inhibition mediated by TDP-43 occurs at the earliest stages of spliceosomal assembly. TDP-43 is a highly conserved nuclear factor of yet unknown function that binds to ug-repeated sequences and is responsible for cystic fibrosis transmembrane conductance regulator exon 9 splicing inhibition. We have analyzed TDP-43 interactions with other splicing factors and identified the critical regions for the protein/protein recognition events that determine this biological function. We show here that the C-terminal region of TDP-43 is capable of binding directly to several proteins of the heterogeneous nuclear ribonucleoprotein (hnRNP) family with well known splicing inhibitory activity, in particular, hnRNP A2/B1 and hnRNP A1. Mutational analysis showed that TDP-43 proteins lacking the C-terminal region could not inhibit splicing probably because they were unable to form the hnRNP-rich complex involved in splicing inhibition. Finally, through splicing complex analysis, we show that splicing inhibition mediated by TDP-43 occurs at the earliest stages of spliceosomal assembly. In its most basic form, the splicing process has the task of removing from the primary RNA transcript all of those sequences (introns) that will not be present in the mature mRNA (1Krainer A.R. Hames B.D. Glover D.M. Frontiers in Molecular Biology: Eukaryotic mRNA processing Oxford University Press, New York. 1997Google Scholar, 2Adams M.D. Rudner D.Z. Rio D.C. Curr. Opin. Cell Biol. 1996; 8: 331-339Crossref PubMed Scopus (117) Google Scholar, 3Burge C.B. Tuschl T. Sharp P.A. Gesteland R.R. Cech T.R. Atkins J.F. The RNA World. 1999: 525-560Google Scholar). Alternative splicing, i.e. the inclusion/exclusion of selected exonic sequences in particular tissues or developmental stages, has been heavily exploited by evolution to generate multiple mRNA transcripts from the same pre-mRNA sequence (4Modrek B. Lee C.J. Nat. Genet. 2003; 34: 177-180Crossref PubMed Scopus (423) Google Scholar, 5Lareau L.F. Green R.E. Bhatnagar R.S. Brenner S.E. Curr. Opin. Struct. Biol. 2004; 14: 273-282Crossref PubMed Scopus (258) Google Scholar, 6Boue S. Letunic I. Bork P. BioEssays. 2003; 25: 1031-1034Crossref PubMed Scopus (108) Google Scholar). However, all of this flexibility also means that the splicing process is prone to mistakes following even minor changes (7Pagani F. Baralle F.E. Nat. Rev. Genet. 2004; 5: 389-396Crossref PubMed Scopus (452) Google Scholar), and alterations of splicing are being increasingly reported as the underlying cause of many genetic diseases (8Faustino N.A. Cooper T.A. Genes Dev. 2003; 17: 419-437Crossref PubMed Scopus (963) Google Scholar, 9Caceres J.F. Kornblihtt A.R. Trends Genet. 2002; 18: 186-193Abstract Full Text Full Text PDF PubMed Scopus (530) Google Scholar, 10Cartegni L. Chew S.L. Krainer A.R. Nat. Rev. Genet. 2002; 3: 285-298Crossref PubMed Scopus (1708) Google Scholar, 11Nissim-Rafinia M. Kerem B. Trends Genet. 2002; 18: 123-127Abstract Full Text Full Text PDF PubMed Scopus (178) Google Scholar, 12Garcia-Blanco M.A. Baraniak A.P. Lasda E.L. Nat. Biotechnol. 2004; 22: 535-546Crossref PubMed Scopus (427) Google Scholar). At the molecular level, the removal of introns and the joining of exons are catalyzed by the spliceosome, which contains several hundred different proteins in addition to the five spliceosomal small nuclear RNAs (13Nilsen T.W. BioEssays. 2003; 25: 1147-1149Crossref PubMed Scopus (279) Google Scholar, 14Zhou Z. Licklider L.J. Gygi S.P. Reed R. Nature. 2002; 419: 182-185Crossref PubMed Scopus (713) Google Scholar). This complex arrangement of factors has two functions: first, to define the exact boundaries of an exon; and second, to catalyze the cut-and-paste generation of the mature mRNA. However, many external factors can also contribute to its workings, such as RNA secondary structure (15Buratti E. Baralle F.E. Mol. Cell. Biol. 2004; 24: 10505-10514Crossref PubMed Scopus (311) Google Scholar), transcription rates (16Kornblihtt A.R. De La Mata M. Fededa J.P. Munoz M.J. Nogues G. RNA (N. Y.). 2004; 10: 1489-1498Crossref PubMed Scopus (369) Google Scholar), the presence of splicing enhancer and silencer elements (17Smith C.W. Valcarcel J. Trends Biochem. Sci. 2000; 25: 381-388Abstract Full Text Full Text PDF PubMed Scopus (747) Google Scholar, 18Blencowe B.J. Trends Biochem. Sci. 2000; 25: 106-110Abstract Full Text Full Text PDF PubMed Scopus (530) Google Scholar), and even external stimuli (19Stamm S. Hum. Mol. Genet. 2002; 11: 2409-2416Crossref PubMed Scopus (174) Google Scholar, 20Shin C. Manley J.L. Nat. Rev. Mol. Cell. Biol. 2004; 5: 727-738Crossref PubMed Scopus (232) Google Scholar). It is the combinatorial effect of all of these factors that will decide when, where, and to what degree a specific sequence will be included or not in the mature mRNA (17Smith C.W. Valcarcel J. Trends Biochem. Sci. 2000; 25: 381-388Abstract Full Text Full Text PDF PubMed Scopus (747) Google Scholar). Recently, the finding that at least 5% of all human alternative exons are derived from the highly repeated dimeric retrotransposons Alu elements has focused a lot of attention on the potential splicing modulatory ability of repeated nucleotide sequences (21Kreahling J. Graveley B.R. Trends Genet. 2004; 20: 1-4Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). In previous works, we have focused our attention on clarifying the pathological role played by (ug)m-repeated sequences near the 3′-splice site of cystic fibrosis transmembrane conductance regulator (CFTR) 3The abbreviations used are: CFTRcystic fibrosis transmembrane conductance regulatorhnRNPheterogeneous nuclear ribonucleoproteinGSTglutathione S-transferaseHPLChigh pressure liquid chromatographyPBSphosphate-buffered salineNi-NTAnickel-nitrilotriacetic acidTemedN,N,N′,N′-tetramethylethylenediaminewtwild-typePTBpolypyrimidine tract-binding proteinRRMRNA recognition motifsnRNPsmall nuclear ribonucleoprotein 3The abbreviations used are: CFTRcystic fibrosis transmembrane conductance regulatorhnRNPheterogeneous nuclear ribonucleoproteinGSTglutathione S-transferaseHPLChigh pressure liquid chromatographyPBSphosphate-buffered salineNi-NTAnickel-nitrilotriacetic acidTemedN,N,N′,N′-tetramethylethylenediaminewtwild-typePTBpolypyrimidine tract-binding proteinRRMRNA recognition motifsnRNPsmall nuclear ribonucleoprotein exon 9 (22Niksic M. Romano M. Buratti E. Pagani F. Baralle F.E. Hum. Mol. Genet. 1999; 8: 2339-2349Crossref PubMed Scopus (128) Google Scholar, 23Buratti E. Dork T. Zuccato E. Pagani F. Romano M. Baralle F.E. EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar, 24Buratti E. Brindisi A. Pagani F. Baralle F.E. Am. J. Hum. Genet. 2004; 74: 1322-1325Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar), as they have been known to promote skipping and to correlate well with disease penetrance (25Groman J.D. Hefferon T.W. Casals T. Bassas L. Estivill X. Des Georges M. Guittard C. Koudova M. Fallin M.D. Nemeth K. Fekete G. Kadasi L. Friedman K. Schwarz M. Bombieri C. Pignatti P.F. Kanavakis E. Tzetis M. Schwartz M. Novelli G. D'Apice M.R. Sobczynska-Tomaszewska A. Bal J. Stuhrmann M. Macek Jr., M. Claustres M. Cutting G.R. Am. J. Hum. Genet. 2004; 74: 176-179Abstract Full Text Full Text PDF PubMed Scopus (212) Google Scholar). In particular, we have shown that (ug)m repeats promote skipping of this exon through their interaction with TDP-43, a protein that, although first described as a DNA-binding protein, has all of the characteristics of a heterogeneous nuclear ribonucleoprotein (hnRNP) (23Buratti E. Dork T. Zuccato E. Pagani F. Romano M. Baralle F.E. EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar, 26Buratti E. Baralle F.E. J. Biol. Chem. 2001; 276: 36337-36343Abstract Full Text Full Text PDF PubMed Scopus (475) Google Scholar, 27Ou S.H. Wu F. Harrich D. Garcia-Martinez L.F. Gaynor R.B. J. Virol. 1995; 69: 3584-3596Crossref PubMed Google Scholar). Significantly, depletion of this protein in transient transfections using antisense oligonucleotides (23Buratti E. Dork T. Zuccato E. Pagani F. Romano M. Baralle F.E. EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar) or in in vitro assays (24Buratti E. Brindisi A. Pagani F. Baralle F.E. Am. J. Hum. Genet. 2004; 74: 1322-1325Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar) results in increasing CFTR exon 9 inclusion, whereas add-back experiments in a depleted nuclear extract restore exon inhibition (24Buratti E. Brindisi A. Pagani F. Baralle F.E. Am. J. Hum. Genet. 2004; 74: 1322-1325Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar, 28Ayala Y.M. Pantano S. D'Ambrogio A. Buratti E. Brindisi A. Marchetti C. Romano M. Baralle F.E. J. Mol. Biol. 2005; 348: 575-588Crossref PubMed Scopus (263) Google Scholar). In keeping with this, it has been recently shown through minigene analysis that CFTR exon 9 inclusion in a wide panel of different cell lines correlates well with the endogenous levels of TDP-43 in each respective cell line (29Disset A. Michot C. Harris A. Buratti E. Claustres M. Tuffery-Giraud S. Hum. Mutat. 2005; 25: 72-81Crossref PubMed Scopus (46) Google Scholar). It should be noted, however, that an alternative explanation based on RNA secondary structure has been offered by Hefferon et al. (30Hefferon T.W. Groman J.D. Yurk C.E. Cutting G.R. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 3504-3509Crossref PubMed Scopus (113) Google Scholar) in which the role of TDP-43 binding to the ug repeats has been questioned. cystic fibrosis transmembrane conductance regulator heterogeneous nuclear ribonucleoprotein glutathione S-transferase high pressure liquid chromatography phosphate-buffered saline nickel-nitrilotriacetic acid N,N,N′,N′-tetramethylethylenediamine wild-type polypyrimidine tract-binding protein RNA recognition motif small nuclear ribonucleoprotein cystic fibrosis transmembrane conductance regulator heterogeneous nuclear ribonucleoprotein glutathione S-transferase high pressure liquid chromatography phosphate-buffered saline nickel-nitrilotriacetic acid N,N,N′,N′-tetramethylethylenediamine wild-type polypyrimidine tract-binding protein RNA recognition motif small nuclear ribonucleoprotein In addition to CFTR exon 9 splicing, the isolation of TDP-43 as a highly specific (ug)m-binding protein is relevant to other genes in which (ug)m-repeated sequences have been described at the 3′-splice site, such as in the case of apolipoprotein AII exon 3 (31Arrisi-Mercado P. Romano M. Muro A.F. Baralle F.E. J. Biol. Chem. 2004; 279: 39331-39339Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar) and intron 2 of the human cardiac Na+-Ca2+ exchanger (32Gabellini N. Eur. J. Biochem. 2001; 268: 1076-1083Crossref PubMed Scopus (50) Google Scholar). In addition, ug-repeated sequences have been predicted to function as intronic splicing enhancer elements in the fish Fugu rubripes (33Yeo G. Hoon S. Venkatesh B. Burge C.B. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 15700-15705Crossref PubMed Scopus (181) Google Scholar). All of these examples indicate that this kind of dinucleotide repeat may well play an important biological role in other systems besides the CFTR gene, something that may well be reflected also in the observation that (ug)m-binding proteins have been found in many different organisms. In fact, (ug)m-binding proteins have been described in Chlamydomonas reinhardtii, in which they participate in the regulation of circadian-controlled processes (34Zhao B. Schneid C. Iliev D. Schmidt E.M. Wagner V. Wollnik F. Mittag M. Eukaryot. Cell. 2004; 3: 815-825Crossref PubMed Scopus (48) Google Scholar); in Caenorhabditis elegans, in which a novel ug-specific binding protein (SUP-12) has been shown to control muscle-specific splicing of the actin-depolymerizing factor/cofilin gene (35Anyanful A. Ono K. Johnsen R.C. Ly H. Jensen V. Baillie D.L. Ono S. J. Cell Biol. 2004; 167: 639-647Crossref PubMed Scopus (41) Google Scholar); and also in humans (36Takahashi N. Sasagawa N. Suzuki K. Ishiura S. Biochem. Biophys. Res. Commun. 2000; 277: 518-523Crossref PubMed Scopus (66) Google Scholar, 37Faustino N.A. Cooper T.A. Mol. Cell. Biol. 2005; 25: 879-887Crossref PubMed Scopus (77) Google Scholar, 38Perez Canadillas J.M. Varani G. EMBO J. 2003; 22: 2821-2830Crossref PubMed Scopus (121) Google Scholar). Interestingly, this kind of repeated dinucleotide sequence can play a functional role irrespective of the direction that its gets transcribed. In fact, (ca)m repeats have also been described to have an effect on splicing by promoting splicing inclusion of exon 13 in the endothelial nitricoxide synthase gene (39Hui J. Stangl K. Lane W.S. Bindereif A. Nat. Struct. Biol. 2003; 10: 33-37Crossref PubMed Scopus (131) Google Scholar, 40Bilbao D. Valcarcel J. Nat. Struct. Biol. 2003; 10: 6-7Crossref PubMed Scopus (7) Google Scholar). Also in this case, the trans-acting factor binding to these repeats has been identified and shown to correspond to the hnRNP L nuclear protein. Not unexpectedly and in a way analogous but opposite to TDP-43, the length of the (ca)m repeats correlates with hnRNP L binding and exon inclusion (39Hui J. Stangl K. Lane W.S. Bindereif A. Nat. Struct. Biol. 2003; 10: 33-37Crossref PubMed Scopus (131) Google Scholar). However, besides its role in CFTR exon 9 splicing, very little is known concerning the normal biological function of TDP-43. At present, experimental data show only that it is a highly conserved protein, as closely related proteins have been found in Drosophila and Caenorhabditis (28Ayala Y.M. Pantano S. D'Ambrogio A. Buratti E. Brindisi A. Marchetti C. Romano M. Baralle F.E. J. Mol. Biol. 2005; 348: 575-588Crossref PubMed Scopus (263) Google Scholar, 41Wang H.Y. Wang I.F. Bose J. Shen C.K. Genomics. 2004; 83: 130-139Crossref PubMed Scopus (236) Google Scholar), that its Xenopus homolog may be involved in the timing of events in mitosis (42Georgi A.B. Stukenberg P.T. Kirschner M.W. Curr. Biol. 2002; 12: 105-114Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar), and that it may act as a bridge between the various nuclear bodies possibly through an interaction with the SMN protein (43Wang I.F. Reddy N.M. Shen C.K. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 13583-13588Crossref PubMed Scopus (163) Google Scholar). Recently, its expression has also been shown to be up-regulated in cells infected by respiratory syncytial virus (44Brasier A.R. Spratt H. Wu Z. Boldogh I. Zhang Y. Garofalo R.P. Casola A. Pashmi J. Haag A. Luxon B. Kurosky A. J. Virol. 2004; 78: 11461-11476Crossref PubMed Scopus (76) Google Scholar). Analogously, very little is also known regarding the mechanism through which TDP-43 can inhibit splicing; and at the moment, the available data show only that its C-terminal tail plays an essential role in inhibiting this process (28Ayala Y.M. Pantano S. D'Ambrogio A. Buratti E. Brindisi A. Marchetti C. Romano M. Baralle F.E. J. Mol. Biol. 2005; 348: 575-588Crossref PubMed Scopus (263) Google Scholar, 41Wang H.Y. Wang I.F. Bose J. Shen C.K. Genomics. 2004; 83: 130-139Crossref PubMed Scopus (236) Google Scholar). Here, we have investigated the reason for this observation; and in this work, we show that the C-terminal region of TDP-43 is capable of interacting with several proteins of the hnRNP family, in particular, hnRNP A2/B1, hnRNP A1, hnRNP C1/C2, and hnRNP A3. This interaction is probably essential for the splicing inhibitory activity of TDP-43. RNA Secondary Structure Determination—RNA secondary structure determination using limited V1 RNase, T1 RNase, and S1 nuclease digestion has been described in detail (45Buratti E. Muro A.F. Giombi M. Gherbassi D. Iaconcig A. Baralle F.E. Mol. Cell. Biol. 2004; 24: 1387-1400Crossref PubMed Scopus (98) Google Scholar). Glutathione S-Transferase (GST) Overlay, HPLC Analysis, and Nanoelectrospray Mass Spectrometry—Western blots containing both HeLa nuclear and cytoplasmic extracts (150 μg) were incubated for 2 h with GST-TDP-43 (10 μg of protein in 20 ml of phosphate-buffered saline (PBS) and 10% (w/v) nonfat dried milk) for 1 h. The blots were washed four times with PBS plus 0.2% Tween 20 and incubated for 1 h with a commercial anti-GST antibody (Sigma) at a dilution of 1:2000. The blots were then washed again four times with PBS plus 0.2% Tween 20 and incubated for 1 h with horseradish peroxidase-conjugated anti-goat antibody (DakoCytomation A/S) at a dilution of 1:2000. After four final washes, the Western blots were developed by ECL (Amersham Biosciences). Fractionation of nuclear extracts was performed on an HPLC Agilent 1100 series using a Phenomenex C4 300A reverse phase column (250 × 4.6 mm). The proteins were eluted using a solution of 95% acetonitrile and 0.1% trifluoroacetic acid in a 20–70% linear gradient over 38 min. The different fractions were first speed-dried for 2 h to reduce volume and then concentrated using Microcon YM-10 filters. Each fraction was divided in two and loaded onto 10% SDS-polyacrylamide gels; one was stained with Coomassie Blue, whereas the other was blotted on Optitran BA-S 83 nitrocellulose membrane (Schleicher & Schüll) and subjected to GST overlay as described above. Internal sequence analysis with the Coomassie Blue-stained bands excised from the SDS-polyacrylamide gel was performed using an electrospray ionization mass spectrometer (LCQ Deca XP, Thermo Electron Corp.). The bands were digested with trypsin, and the resulting peptides were extracted with water, 60% acetonitrile, and 1% trifluoroacetic acid. The fragments were then analyzed by mass spectrometry, and the proteins were identified by analysis of the peptide tandem mass spectrometry data with Turbo SEQUEST (Thermo Electron Corp.) and Mascot (Matrix Science). Reverse Transcription-PCR Amplifications of TDP-43, hnRNP Factors, and Mutants and Their Expression—The GST-TDP-43 and GST-TDP-43-(101–261) expression plasmids have been described (23Buratti E. Dork T. Zuccato E. Pagani F. Romano M. Baralle F.E. EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar). The GST-TDP-43(NR-1) and GST-TDP-43-(101–414) mutants are protein fragments containing residues 1–190 and 101–414 of TDP-43, respectively. The fragments were amplified by PCR using oligonucleotides 571SmaRV (5′-tcccccggggcttctcaaaggctcatcttgg-3′) and pGEX5′ (5′-gggctggcaagccacgtttggtg-3′) for GST-TDP-43(NR-1) and oligonucleotides TDPDNf (5′-cgcggatccagaaaacatccgatttaatagtg-3′) and pGEX3′ (5′-ccgggagctgcatgtgtcagagg-3′) for GST-TDP-43-(101–414). The products were cloned into the pGEX-3X expression vector (Amersham Biosciences) using BamHI and SmaI in the case of GST-TDP-43(NR-1) and BamHI and EcoRI in the case of GST-TDP-43-(101–414). Expression and purification of the mutants were carried out as described previously (23Buratti E. Dork T. Zuccato E. Pagani F. Romano M. Baralle F.E. EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar). The GST-TDP-43(ΔCter) mutant has been used in a recent study (28Ayala Y.M. Pantano S. D'Ambrogio A. Buratti E. Brindisi A. Marchetti C. Romano M. Baralle F.E. J. Mol. Biol. 2005; 348: 575-588Crossref PubMed Scopus (263) Google Scholar). The plasmid expressing His-hnRNP A1 and rabbit antisera directed against this protein were a kind gift of Alexander Ochem and Raffaella Klima (International Centre for Genetic Engineering and Biotechnology). The hnRNP C2 sequence (to obtain GST-hnRNP C2) was amplified by reverse transcription-PCR from total HeLa RNA using primers hnRNPCS (5′-aataaagaattctcatcctccattggcgctgtctctgtcatcctc-3′, sense) and hnRNPCAS (5′-tcgactcactgggctggcacaggctgaggggtctctcgcccactatcattaggcccctcgg-3′, antisense). The amplified product was then sequenced and inserted in the pGEX-3X plasmid for expression. The same procedure was followed to obtain the GST-hnRNP A2 plasmid using primers hnRNPA2S (5′-aatttaaggatccccatggagagagaaaagga-3′, sense) and hnRNPA2AS (5′-aattaagaattctcagtatcggctcctcccaccataa-3′, antisense). Mutant hnRNP A2 sequences were produced as follows. GST-hnRNP A2(ΔCter) was obtained by amplifying the hnRNP A2 cDNA with primers hnRNPA2S and hnRNPA2/631AS (5′-cctcctctaaagttacttcctggtcc-3′, antisense), whereas hnRNP A2(ΔRRM2) was obtained by a two-round amplification procedure using primers hnRNPA2/ΔRRM2S (5′-ctcatgtaactgtggaaatgcaggaagtt-3′, sense) and hnRNPA2/ΔRRM2AS (5′-aacttcctgcatttccacagttacatgag-3′, antisense). All proteins were expressed and purified as described previously (23Buratti E. Dork T. Zuccato E. Pagani F. Romano M. Baralle F.E. EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar). Polyclonal antibodies were then obtained by immunizing a 3-month old New Zealand white rabbit according to standard immunization protocols. GST Pull-down Analysis—Approximately 40 μg of each recombinant protein were bound to 500 μl of glutathione-Sepharose 4B resin and incubated with 0.5 mg of HeLa nuclear protein extract (C4 Biotech) for 2 h at room temperature in PBS and 0.2% Tween 20. As a control, an equal amount of GST protein was bound to a second batch of glutathione-Sepharose 4B resin. After four wash cycles, the proteins bound to the resin were incubated with SDS-PAGE loading buffer and loaded onto a 10% SDS-polyacrylamide gel. Western blotting was performed following standard protocols using rabbit anti-TDP-43 sera at a dilution of 1:2000, and blots were developed by ECL. Protein/Protein Interaction by Pull-down Analysis—200 μl of nickel-nitrilotriacetic acid (Ni-NTA; for His-TDP-43) or Sepharose 4B (for GST-TDP-43) resin were washed with 1 ml of PBS and incubated for 1 h at 4 °C with 4 μg of dialyzed recombinant protein. The coated beads were then collected by centrifugation at 3000 rpm for 5 min in a Eppendorf minicentrifuge; and after two additional washes with PBS and 0.2% Tween, each batch was incubated for 1 h at room temperature with 2 μg of its respective recombinant protein (GST-hnRNP A2 and GST-hnRNP C2 for the His-TDP-43-coated resin and His-TDP-43 for the GST-TDP-43-coated resin). After four washes with PBS and 0.2% Tween, the beads were collected, and 60 μl of SDS loading buffer were added before loading onto a 10% polyacrylamide gel. Western blotting was performed to check for the presence of the added recombinant protein. As a control, uncoated beads were used in each experiment. Band Shift Analysis—Commercial (ug)12 and (ug)6 RNA oligonucleotides (200 ng, ∼25 pmol; MWG Biotech) were labeled by phosphorylation with [γ-32P]ATP and T4 polynucleotide kinase (Stratagene) according to a standard protocol and resuspended in 400 μl of water. Each binding reaction was performed at room temperature for 15 min by mixing the purified protein with the labeled RNA in a 20-μl final volume. To obtain the (uaggg(u/a))3 RNAs, we cloned, into the SacI-XbaI restriction fragment of Bluescript KS+ sites, oligonucleotides (after annealing) 5′-ctagggatagggttagggat-3′ (sense) and 5′-ctagatccctaaccctatccctagagct-3′ (antisense); and to obtain the u21 RNAs, oligonucleotides 5′-cttttttttttttttttttttt-3′ (sense) and 5′-ctagaaaaaaaaaaaaaaaaaaaaagagct-3′ (antisense). The plasmids were linearized with XbaI, and in vitro transcription was performed as described previously (26Buratti E. Baralle F.E. J. Biol. Chem. 2001; 276: 36337-36343Abstract Full Text Full Text PDF PubMed Scopus (475) Google Scholar). Binding reactions were performed in 1× bind shift binding buffer (20 mm Hepes (pH 7.9), 72 mm KCl, 1.5 mm MgCl2, 0.78 mm magnesium acetate, 0.52 mm dithiothreitol, 3.8% glycerol, 0.75 mm ATP, and 1 mm GTP) and 5 μg/μl heparin and electrophoresed on a 5% polyacrylamide gel at 100 V for 1 h in 0.5 × Tris borate/EDTA buffer at 4 °C. The amounts of each protein used in these experiments are reported in the figure legends. The gel were then dried and exposed to autoradiographic XAR film (Eastman Kodak Co.). In Vitro Splicing Analysis and Separation of Spliceosomal Complexes— The pY7 and pY7(UG12U5) plasmids have been described previously (24Buratti E. Brindisi A. Pagani F. Baralle F.E. Am. J. Hum. Genet. 2004; 74: 1322-1325Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). In vitro splicing reactions were performed using capped SP6-transcribed RNAs according to standard protocols. Recombinant proteins (400 ng each) were added to the depleted nuclear extract prior to assembling the remainder of the splicing reaction mixture. Splicing reactions were incubated at 30 °C for 2 h. The RNA was extracted using RNAwiz (Ambion, Inc.); ethanol-precipitated; and analyzed on a denaturing 6% polyacrylamide gel, followed by autoradiography. Spliceosomal complexes at selected time points during the in vitro splicing reactions were separated by electrophoresis on a native 3.5% acrylamide gel (2 ml of 1% bisacrylamide, 3.7 ml of 40% acrylamide, 2 ml of 1 m Tris (pH 10), 2 ml of 1 m glycine (pH 6), 28.5 ml of H2O, 0.28 ml of ammonium persulfate, and 28 μl of Temed) following addition of 1 μl of heparin (50 μg/μl). RNA Secondary Structure of CFTR Exon 9 and the (ug)m-repeated Motif—Before analyzing in detail the interactions of TDP-43 with cellular proteins, we decided to address the alternative explanation for the negative role played by (ug)m repeats in CFTR exon 9 splicing proposed recently by Hefferon et al. (30Hefferon T.W. Groman J.D. Yurk C.E. Cutting G.R. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 3504-3509Crossref PubMed Scopus (113) Google Scholar). In this model, the (ug)m repeats have been suggested to fold upon themselves to inhibit CFTR exon 9 recognition by blocking 3′-splice site usage. This mechanism could thus occur without the need to invoke the presence of any trans-acting factor such as TDP-43. Therefore, to address this issue experimentally, we extended a preliminary structural probing that was performed on CFTR exon 9 in a previous study (22Niksic M. Romano M. Buratti E. Pagani F. Baralle F.E. Hum. Mol. Genet. 1999; 8: 2339-2349Crossref PubMed Scopus (128) Google Scholar). The probed RNA thus comprised the entire CFTR exon 9 sequence carrying a (ug)11u7 repeat and a significant portion of the intervening sequence 8 intron. Structural probing was performed according to a methodology recently described for the EDA exons of the mouse fibronectin gene (45Buratti E. Muro A.F. Giombi M. Gherbassi D. Iaconcig A. Baralle F.E. Mol. Cell. Biol. 2004; 24: 1387-1400Crossref PubMed Scopus (98) Google Scholar), which integrates limited RNase digestion with mFold structural predictions (46Zuker M. Nucleic Acids Res. 2003; 31: 3406-3415Crossref PubMed Scopus (9968) Google Scholar). As shown in Fig. 1, in this extended RNA context, the enzymatic cleavages confirmed the existence of several stem-loop elements (I–IV) that had already been suggested to exist in a previous study (22Niksic M. Romano M. Buratti E. Pagani F. Baralle F.E. Hum. Mol. Genet. 1999; 8: 2339-2349Crossref PubMed Scopus (128) Google Scholar) and excluded any tendency for the (ug)11 sequence to fold upon itself, even in the absence of interacting proteins. Moreover, mFold analyses performed using this extended context and after varying the number of ug repeats in no instance predicted this sequence as being capable of folding upon itself (data not shown). The discrepancy between these experimental probing data and the conclusions of Hefferon et al. (30Hefferon T.W. Groman J.D. Yurk C.E. Cutting G.R. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 3504-3509Crossref PubMed Scopus (113) Google Scholar) may be explained by the fact that their in silico predictions were performed by taking into account only the ug-repeated sequences and a very limited region surrounding this region (10 nucleotides on either side). Given these limitations, it is not surprising that mFold predictions suggested the possibility that (ug)m repeats might fold upon themselves. Indeed, the length of RNA examined represents a severe limitation in assigning significance to in silico results, which has been reported for the neurofibromin-1 gene, where the RNA window analyzed by even a few nucleotides could yield striking differences in the prediction results (47Vandenbroucke I. Callens T. De Paepe A. Messiaen L. BMC Genomics. 2002; PubMed Google Scholar). Therefore, we decided to investigate further the splicing inhibitory properties of TDP-43. In Vitro Effects of TDP-43 Mutants on CFTR Exon 9 Splicing and Analysis of Spliceosomal Complex Formation—In a recent study, we highlighted the importance of the TDP-43 C-terminal region in splicing inhibition (28Ayala Y.M. Pantano S. D'Ambrogio A. Buratti E. Brindisi A. Marchetti C. Romano M. Baralle F.E. J. Mol. Biol. 2005; 348: 575-588Crossref PubMed Scopus (263) Google Scholar). To confirm and extend these data, we used a pY7 plasmid containing two tropomyosin exons separated by an ∼120-nucleotide-long intron sequence that was modified to contain the 3′-splice site of CFTR exon 9 (Fig. 2A, underlined). Detailed in vitro anal