The RNA Binding Domains of the Nuclear poly(A)-binding Protein
2003; Elsevier BV; Volume: 278; Issue: 19 Linguagem: Inglês
10.1074/jbc.m209886200
ISSN1083-351X
AutoresUwe Kühn, Anne Németh, Sylke Meyer, Elmar Wahle,
Tópico(s)RNA modifications and cancer
ResumoThe nuclear poly(A)-binding protein (PABPN1) is involved in the synthesis of the mRNA poly(A) tails in most eukaryotes. We report that the protein contains two RNA binding domains, a ribonucleoprotein-type RNA binding domain (RNP domain) located approximately in the middle of the protein sequence and an arginine-rich C-terminal domain. The C-terminal domain also promotes self-association of PABPN1 and moderately cooperative binding to RNA. Whereas the isolated RNP domain binds specifically to poly(A), the isolated C-terminal domain binds non-specifically to RNA and other polyanions. Despite this nonspecific RNA binding by the C-terminal domain, selection experiments show that adenosine residues throughout the entire minimal binding site of ∼11 nucleotides are recognized specifically. UV-induced cross-links with oligo(A) carrying photoactivatable nucleotides at different positions all map to the RNP domain, suggesting that most or all of the base-specific contacts are made by the RNP domain, whereas the C-terminal domain may contribute nonspecific contacts, conceivably to the same nucleotides. Asymmetric dimethylation of 13 arginine residues in the C-terminal domain has no detectable influence on the interaction of the protein with RNA. The N-terminal domain of PABPN1 is not required for RNA binding but is essential for the stimulation of poly(A) polymerase. The nuclear poly(A)-binding protein (PABPN1) is involved in the synthesis of the mRNA poly(A) tails in most eukaryotes. We report that the protein contains two RNA binding domains, a ribonucleoprotein-type RNA binding domain (RNP domain) located approximately in the middle of the protein sequence and an arginine-rich C-terminal domain. The C-terminal domain also promotes self-association of PABPN1 and moderately cooperative binding to RNA. Whereas the isolated RNP domain binds specifically to poly(A), the isolated C-terminal domain binds non-specifically to RNA and other polyanions. Despite this nonspecific RNA binding by the C-terminal domain, selection experiments show that adenosine residues throughout the entire minimal binding site of ∼11 nucleotides are recognized specifically. UV-induced cross-links with oligo(A) carrying photoactivatable nucleotides at different positions all map to the RNP domain, suggesting that most or all of the base-specific contacts are made by the RNP domain, whereas the C-terminal domain may contribute nonspecific contacts, conceivably to the same nucleotides. Asymmetric dimethylation of 13 arginine residues in the C-terminal domain has no detectable influence on the interaction of the protein with RNA. The N-terminal domain of PABPN1 is not required for RNA binding but is essential for the stimulation of poly(A) polymerase. ribonucleoprotein glutathione S-transferase poly(A)-binding protein, nuclear 1 poly(A)-binding protein, cytoplasmic 5-iodouridine nickel-nitrilotriacetic acid N-[2-hydroxy-1,1-bis(hydro xymethyl)ethyl]glycine In the cell, mRNA molecules and their precursors are always bound by proteins. These proteins not only protect the RNA from nucleases and undesirable interactions of its highly charged surface but influence enzymes and other proteins that act upon the RNA at all stages of its maturation, function, and decay (1Varani G. Nagai L. Annu. Rev. Biophys. Biomol. Struct. 1998; 27: 407-445Crossref PubMed Scopus (252) Google Scholar). Characteristically, a single RNA-binding protein very often contains more than one RNA binding domain. Different kinds of RNA binding domains have been described (2Burd C.G. Dreyfuss G. Science. 1994; 265: 615-621Crossref PubMed Scopus (1726) Google Scholar). Among them, the RNA recognition motif or RNP1-type RNA binding domain is probably the best understood (1Varani G. Nagai L. Annu. Rev. Biophys. Biomol. Struct. 1998; 27: 407-445Crossref PubMed Scopus (252) Google Scholar, 3Nagai K. Oubridge C. Ito N. Avis J. Evans P. Trends Biochem. Sci. 1995; 20: 235-240Abstract Full Text PDF PubMed Scopus (201) Google Scholar, 4Nagai K. Curr. Opin. Struct. Biol. 1996; 6: 53-61Crossref PubMed Scopus (159) Google Scholar). The RNP domain consists of ∼90 amino acids forming a औαऔऔαऔ fold, in which a four-stranded औ-sheet is backed by two α-helices. The two central antiparallel औ-strands carry the highly conserved amino acids of the RNP1 and RNP2 motifs. Different members of the RNP protein family can bind structured or extended RNA molecules in a sequence-specific manner. As seen in several co-crystals, the RNA is bound on the surface of the औ-sheet by hydrogen bonds and stacking interactions between bases and amino acid side chains (5Oubridge C. Ito N. Evans P.R. Teo C.-H. Nagai K. Nature. 1994; 372: 432-438Crossref PubMed Scopus (781) Google Scholar, 6Price S.R. Evans P.R. Nagai K. Nature. 1998; 394: 645-650Crossref PubMed Scopus (306) Google Scholar, 7Handa N. Nureki O. Kurimoto K. Im I. Sakamoto H. Shimura Y. Muto Y. Yokoyama S. Nature. 1999; 398: 579-585Crossref PubMed Scopus (316) Google Scholar, 8Deo R.C. Bonanno J.B. Sonenberg N. Burley S.K. Cell. 1999; 98: 835-845Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar). Another common RNA binding domain is the so-called RGG domain, characterized by multiple copies of the amino acid sequence arginine-glycine-glycine, interspersed with phenylalanine and tyrosine residues (9Kiledjian M. Dreyfuss G. EMBO J. 1992; 11: 2655-2664Crossref PubMed Scopus (506) Google Scholar). The structure of the domain is not known, although a spiral of औ-turns has been proposed based on spectroscopic data (10Ghisolfi L. Joseph G. Amalric F. Erard M. J. Biol. Chem. 1992; 267: 2955-2959Abstract Full Text PDF PubMed Google Scholar). A possibly related arginine-rich domain found at the C termini of several of the spliceosomal Sm core proteins was not ordered in a crystal structure in the absence of RNA (11Kambach C. Walke S. Young R. Avis J.M. de la Fortelle E. Raker V.A. Lührmann R. Li J. Nagai K. Cell. 1999; 96: 375-387Abstract Full Text Full Text PDF PubMed Scopus (362) Google Scholar). The RGG domain usually occurs in proteins in conjunction with one or more other RNA binding domains, e.g. of the RNP or K homology type (2Burd C.G. Dreyfuss G. Science. 1994; 265: 615-621Crossref PubMed Scopus (1726) Google Scholar) and is often considered unable to discriminate between different RNA sequences. Instead, it is thought to increase the RNA binding affinity of a protein in a nonspecific manner, the specificity being determined by the other RNA binding domain(s) (12Ghisolfi L. Kharrat A. Joseph G. Amalric F. Erard M. Eur. J. Biochem. 1992; 209: 541-548Crossref PubMed Scopus (101) Google Scholar, 13Bagni C. Lapeyre B. J. Biol. Chem. 1998; 273: 10868-10873Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 14Adinolfi S. Bagni C. Musco G. Gibson T. Mazzarella L. Pastore A. RNA. 1999; 5: 1248-1258Crossref PubMed Scopus (74) Google Scholar). However, sequence- or structure-specific binding by means of an RGG domain has been proposed for several proteins (9Kiledjian M. Dreyfuss G. EMBO J. 1992; 11: 2655-2664Crossref PubMed Scopus (506) Google Scholar, 15Vanhamme L. Perez-Morga D. Marchal C. Speijer D. Lambert L. Geusken M. Alexandre S. Ismaili N. Göringer H.U. Benne R. Pays E. J. Biol. Chem. 1998; 273: 21825-21833Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 16Abdul-Manan N. O'Malley S.M. Williams K.R. Biochemistry. 1996; 35: 3545-3554Crossref PubMed Scopus (46) Google Scholar, 17Darnell J.C. Jensen K.B. Jin P. Brown V. Warren S.T. Darnell R.B. Cell. 2001; 107: 489-499Abstract Full Text Full Text PDF PubMed Scopus (779) Google Scholar). A characteristic feature of the RGG domain is the asymmetric dimethylation of the arginine side chains within RGG sequences (18Gary J.D. Clarke S. Prog. Nucleic Acid Res. Mol. Biol. 1998; 61: 65-131Crossref PubMed Google Scholar). A possible modulation of RNA binding by arginine methylation has frequently been discussed, but binding of the yeast protein Hrp1p to a specific RNA sequence was not affected by arginine methylation (19Valentini S.R. Weiss V.H. Silver P.A. RNA. 1999; 5: 272-280Crossref PubMed Scopus (72) Google Scholar). The affinity of a synthetic RGG domain peptide for nonspecific RNA was also independent of arginine methylation, although CD spectroscopy suggested that the methylated peptide had a different structural effect on the RNA compared with the unmethylated peptide (20Raman B. Guarnaccia C. Nadassy K. Zakhariev S. Pintar A. Zanuttin F. Frigyes D. Acatrinei C. Vindigni A. Pongor G. Pongor S. Nucleic Acids Res. 2001; 29: 3377-3384Crossref PubMed Google Scholar). Several RGG domains are involved in protein-protein interactions; RGG domain-dependent self-association of the hnRNP A1 protein leads to moderate cooperativity of RNA binding (21Cobianchi F. Karpel R.L. Williams K.R. Notario V. Wilson S.H. J. Biol. Chem. 1988; 263: 1063-1071Abstract Full Text PDF PubMed Google Scholar, 22Casas-Finet J.R. Smith J.D. Kumar A. Kim J.G. Wilson S.H. Karpel R.L. J. Mol. Biol. 1993; 229: 873-889Crossref PubMed Scopus (53) Google Scholar), but the same domain can also interact with other proteins (23Cartegni L. Maconi M. Morandi E. Cobianchi F. Riva S. Biamonti G. J. Mol. Biol. 1996; 259: 337-348Crossref PubMed Scopus (159) Google Scholar). Similarly, RGG domains of other proteins serve in protein-protein interactions (24Bouvet P. Diaz J.-J. Kindbeiter K. Madjar J.-J. Amalric F. J. Biol. Chem. 1998; 273: 19025-19029Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 25Yun C.Y. Fu X.-D. J. Cell Biol. 2000; 150: 707-717Crossref PubMed Scopus (136) Google Scholar). The poly(A) tails at the 3′-ends of eukaryotic mRNAs are bound by two different proteins. Cytoplasmic poly(A)-binding protein 2In the designation of poly(A)-binding proteins, we follow the recommendations of the HUGO Gene Nomenclature Committee (www.gene.ucl.ac.uk/nomenclature/). The cytoplasmic poly(A)-binding protein (PABPC) exists in several variants (PABPC1 through PABPC4 in humans) and is usually called PAB in the literature. The nuclear poly(A)-binding protein (now called PABPN1) was initially described as PAB II (33,34) and later renamed PABP2 (46Brais B. Bouchard J.P. Xie Y.G. Rochefort D.L. Chretien N. Tome F.M. Lafreniere R.G. Rommens J.M. Uyama E. Nohira O. Blumen S. Korcyn A.D. Heutink P. Mathieu J. Duranceau A. Codere F. Fardeau M. Rouleau G.A. Nat. Genet. 1998; 18: 164-167Crossref PubMed Scopus (634) Google Scholar). 2In the designation of poly(A)-binding proteins, we follow the recommendations of the HUGO Gene Nomenclature Committee (www.gene.ucl.ac.uk/nomenclature/). The cytoplasmic poly(A)-binding protein (PABPC) exists in several variants (PABPC1 through PABPC4 in humans) and is usually called PAB in the literature. The nuclear poly(A)-binding protein (now called PABPN1) was initially described as PAB II (33,34) and later renamed PABP2 (46Brais B. Bouchard J.P. Xie Y.G. Rochefort D.L. Chretien N. Tome F.M. Lafreniere R.G. Rommens J.M. Uyama E. Nohira O. Blumen S. Korcyn A.D. Heutink P. Mathieu J. Duranceau A. Codere F. Fardeau M. Rouleau G.A. Nat. Genet. 1998; 18: 164-167Crossref PubMed Scopus (634) Google Scholar). (PABPC; Pab1p in Saccharomyces cerevisiae) (26Sachs A.B. Bond M.W. Kornberg R.D. Cell. 1986; 45: 827-835Abstract Full Text PDF PubMed Scopus (272) Google Scholar, 27Adam S.A. Nakagawa T. Swanson M.S. Woodruff T.K. Dreyfuss G. Mol. Cell. Biol. 1986; 6: 2932-2943Crossref PubMed Scopus (344) Google Scholar) is found in all eukaryotes. Its main functions are in the initiation of translation (28Sachs A. Sonenberg N. Hershey J.W.B. Mathews M.B. Translational Control of Gene Expression. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY2000: 447-465Google Scholar) and in mRNA decay (29Schwartz D.C. Parker R. Sonenberg N. Hershey J.W.B. Mathews M.B. Translational Control of Gene Expression. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY2000: 807-825Google Scholar). PABPC contains four copies of the RNP domain, the first and second being mainly responsible for specific binding to poly(A) (30Burd C.G. Matunis E.L. Dreyfuss G. Mol. Cell. Biol. 1991; 11: 3419-3424Crossref PubMed Scopus (204) Google Scholar, 31Kühn U. Pieler T. J. Mol. Biol. 1996; 256: 20-30Crossref PubMed Scopus (185) Google Scholar, 32Deardorff J.A. Sachs A.B. J. Mol. Biol. 1997; 269: 67-81Crossref PubMed Scopus (84) Google Scholar). In a co-crystal of these two domains with oligoadenylate, A11, the औ-sheet surfaces of the two RNP domains form an almost continuous platform that binds an extended conformation of the oligonucleotide. The 3′-half of the oligonucleotide is associated with the N-terminal RNP domain (8Deo R.C. Bonanno J.B. Sonenberg N. Burley S.K. Cell. 1999; 98: 835-845Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar). Nuclear poly(A)-binding protein2 (PABPN1) (33Wahle E. Cell. 1991; 66: 759-768Abstract Full Text PDF PubMed Scopus (233) Google Scholar, 34Nemeth A. Krause S. Blank D. Jenny A. Jenö P. Lustig A. Wahle E. Nucleic Acids Res. 1995; 23: 4034-4041Crossref PubMed Scopus (80) Google Scholar) stimulates synthesis of the poly(A) tails of pre-mRNAs by increasing the processivity of poly(A) polymerase (33Wahle E. Cell. 1991; 66: 759-768Abstract Full Text PDF PubMed Scopus (233) Google Scholar, 35Bienroth S. Keller W. Wahle E. EMBO J. 1993; 12: 585-594Crossref PubMed Scopus (186) Google Scholar) and also plays a role in poly(A) tail length control, i.e. in limiting processive poly(A) tail synthesis to ∼250 nucleotides (33Wahle E. Cell. 1991; 66: 759-768Abstract Full Text PDF PubMed Scopus (233) Google Scholar,35Bienroth S. Keller W. Wahle E. EMBO J. 1993; 12: 585-594Crossref PubMed Scopus (186) Google Scholar, 36Wahle E. J. Biol. Chem. 1995; 270: 2800-2808Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar). In addition, the protein may be involved in mRNA export into the cytoplasm (37Chen Z. Li Y. Krug R.M. EMBO J. 1999; 18: 2273-2283Crossref PubMed Scopus (326) Google Scholar, 38Calado A. Kutay U. Kühn U. Wahle E. Carmo-Fonseca M. RNA. 2000; 6: 245-256Crossref PubMed Scopus (87) Google Scholar). Although PABPN1 is conserved in most organisms, it does not appear to exist in S. cerevisiae, as its closest homolog in yeast is a cytoplasmic protein (Rbp29p) possibly involved in translation (39Winstall E. Sadowski M. Kühn U. Wahle E. Sachs A.B. J. Biol. Chem. 2000; 275: 21817-21826Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar). In vitro, PABPN1 binds with high affinity and specificity to poly(A) and, almost equally well, to poly(G) (33Wahle E. Cell. 1991; 66: 759-768Abstract Full Text PDF PubMed Scopus (233) Google Scholar, 40Wahle E. Lustig A. Jenö P. Maurer P. J. Biol. Chem. 1993; 268: 2937-2945Abstract Full Text PDF PubMed Google Scholar). In vivo data are consistent with its binding to poly(A) tails in nuclear RNA and support a role in mRNA polyadenylation (37Chen Z. Li Y. Krug R.M. EMBO J. 1999; 18: 2273-2283Crossref PubMed Scopus (326) Google Scholar, 38Calado A. Kutay U. Kühn U. Wahle E. Carmo-Fonseca M. RNA. 2000; 6: 245-256Crossref PubMed Scopus (87) Google Scholar, 41Krause S. Fakan S. Weis K. Wahle E. Exp. Cell Res. 1994; 214: 75-82Crossref PubMed Scopus (104) Google Scholar, 42Calado A. Carmo-Fonseca M. J. Cell Sci. 2000; 113: 2309-2318PubMed Google Scholar, 43Calado A. Tomé F.M.S. Brais B. Rouleau G.A. Kühn U. Wahle E. Carmo-Fonseca M. Hum. Mol. Genet. 2000; 9: 2321-2328Crossref PubMed Scopus (196) Google Scholar). Binding to poly(A) is moderately cooperative (44Meyer S. Urbanke C. Wahle E. Biochemistry. 2002; 41: 6082-6089Crossref PubMed Scopus (23) Google Scholar). In binding to long poly(A), PABPN1 can form spherical particles of a defined size that accommodate ∼250 nucleotides (45Keller R.W. Kühn U. Aragon M. Bornikova L. Wahle E. Bear D.G. J. Mol. Biol. 2000; 297: 569-583Crossref PubMed Scopus (56) Google Scholar). These particles appear to be in equilibrium with filamentous complexes. The structure of PABPN1 is also of interest as short expansions of an oligoalanine tract at the N terminus of the protein lead to the human genetic disease oculopharyngeal muscular dystrophy, which is characterized by the formation of insoluble PABPN1 aggregates in the nuclei of muscle cells (43Calado A. Tomé F.M.S. Brais B. Rouleau G.A. Kühn U. Wahle E. Carmo-Fonseca M. Hum. Mol. Genet. 2000; 9: 2321-2328Crossref PubMed Scopus (196) Google Scholar, 46Brais B. Bouchard J.P. Xie Y.G. Rochefort D.L. Chretien N. Tome F.M. Lafreniere R.G. Rommens J.M. Uyama E. Nohira O. Blumen S. Korcyn A.D. Heutink P. Mathieu J. Duranceau A. Codere F. Fardeau M. Rouleau G.A. Nat. Genet. 1998; 18: 164-167Crossref PubMed Scopus (634) Google Scholar). Upon sequence inspection, an RNP-type RNA binding domain is evident approximately in the middle of the PABPN1 amino acid sequence. Although the C-terminal domain of the protein contains no RGG sequences, it is arginine-rich, and all of its 13 arginines are asymmetrically dimethylated (47Smith J.J. Rücknagel K.P. Schierhorn A. Tang J. Nemeth A. Linder M. Herschman H.R. Wahle E. J. Biol. Chem. 1999; 274: 13229-13234Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar). The contributions of different domains of the protein to RNA binding have not been investigated so far. In this paper we present evidence that both the RNP domain and the C-terminal arginine-rich domain of PABPN1 contribute to RNA binding. The N-terminal domain is essential for the stimulation of poly(A) polymerase. Sequences of DNA oligonucleotides used in the following procedures are available upon request. All constructs were verified by DNA cycle sequencing with a Prism 310 genetic analyzer (Applied Biosystems). The bovine PABPN1 coding sequence (GenBankTMaccession number X89969) (34Nemeth A. Krause S. Blank D. Jenny A. Jenö P. Lustig A. Wahle E. Nucleic Acids Res. 1995; 23: 4034-4041Crossref PubMed Scopus (80) Google Scholar) inserted into the NdeI andBamHI sites of pGM10 (48Martin G. Keller W. EMBO J. 1996; 15: 2593-2603Crossref PubMed Scopus (168) Google Scholar) was initially used for the production of His6-tagged PABPN1 and variants inEscherichia coli. Later, a modified pET19b expression vector (Novagen) was used, in which the NcoI/NdeI fragment encoding the tag was replaced by the corresponding fragment from pGM10 encoding the peptide MAH6. The resulting vector, which resulted in higher levels of recombinant proteins compared with pGM10, will be referred to as pUK. Gel-purified PABPN1 cDNA fragments were cloned into pUK according to standard procedures. For silent mutagenesis of the PABPN1 coding sequence with the aim of reducing the GC content, 18 overlapping and phosphorylated oligonucleotides spanning the first 360 bp were synthesized by TIB Molbiol, Berlin, Germany. The oligonucleotides were designed such that the desired ligation product contained overhangingNdeI/XhoI ends. A mixture of all oligonucleotides (20 nm each in 100 ॖl of ligation buffer without ATP) were melted at 95 °C for 5 min. Annealing took place by slow cooling of the reaction mix to room temperature in a water bath. After addition of 1 mm ATP and 800 units of T4 DNA ligase, aliquots of this mixture were incubated at six different temperatures between 12 and 40 °C for 15 min to 16 h, the incubation time depending on the temperature. After DNA recovery by ethanol precipitation, one-half of each ligation reaction was used for ligation into theNdeI/XhoI-opened and dephosphorylated pGM-PABPN1 plasmid in 20-ॖl standard reactions. The presence of an additional RsaI restriction site in the synthetic gene fragment allowed for the initial identification of clones in which the synthetic sequence had replaced the beginning of the authentic open reading frame. One such clone was then confirmed by DNA sequencing. After subcloning of the synthetic open reading frame into pUK, the resulting plasmid pUK-synPABPN1 was used for the generation of C-terminal deletion constructs, as well as for fusion protein constructs. Deletions mutants of PABPN1 were generated by PCR using PwoDNA polymerase (Hybaid AGS, Heidelberg, Germany) and primers introducing a new start codon as part of an NdeI site or a stop codon followed by a new BamHI site, respectively. Phosphorylated and purified PCR fragments were subcloned intoSmaI-cut pGEM3z (Promega). After double digestion withNdeI or XhoI combined with BamHI, the shortened fragments of the PABPN1 coding region were cloned into the pUK vector or the pUK-synPABPN1 construct opened at the same restriction sites. For GST pull down experiments, theXhoI/BamHI fragments coding for C-terminal truncations were subcloned into the pUK-PABPN1-ΔN113 construct opened at the same sites. The clone expressing the RNP domain of PABPN1 encodes the amino acids 161–257, and the C terminus consists of amino acids 258–306. Single amino acid substitutions were made with the use of a PCR-based method (49Picard V. Ersdal-Bardju E. Lu A. Bock S.C. Nucleic Acids Res. 1994; 22: 2587-2591Crossref PubMed Scopus (216) Google Scholar). Positive clones were identified with the help of newly created restriction sites and verified by DNA sequencing. For the generation of GST and protein A fusion proteins, the sequences encoding the respective tags were PCR-amplified with Pwo DNA polymerase and primer pairs containing additional 5′-NdeI sites. The plasmid pGEX5 × 1, bp 258–945 (Amersham Biosciences) was used as a template for the GST gene, and the plasmid pBS1761, bp 785–1225 (50Puig O. Caspary F. Rigault G. Rutz B. Bouveret E. Bragado-Nilsson W. Wilm M. Seraphin B. Methods. 2001; 24: 218-229Crossref PubMed Scopus (1419) Google Scholar) was used for the protein A tag. PCR products were inserted into the NdeI site of pUK-synPABPN1. Protein concentrations were determined with Bradford reagent (Bio-Rad) and/or by SDS-polyacrylamide gel electrophoresis followed by Coomassie staining and gel imaging with bovine serum albumin as a standard. Calf thymus PABPN1 was isolated as described (40Wahle E. Lustig A. Jenö P. Maurer P. J. Biol. Chem. 1993; 268: 2937-2945Abstract Full Text PDF PubMed Google Scholar). The untagged wild-type PABPN1 used in the experiment of Fig. 9 and related assays was produced in E. coli from the authentic cDNA sequence cloned into pT7–7 (34Nemeth A. Krause S. Blank D. Jenny A. Jenö P. Lustig A. Wahle E. Nucleic Acids Res. 1995; 23: 4034-4041Crossref PubMed Scopus (80) Google Scholar) and purified essentially as described (34Nemeth A. Krause S. Blank D. Jenny A. Jenö P. Lustig A. Wahle E. Nucleic Acids Res. 1995; 23: 4034-4041Crossref PubMed Scopus (80) Google Scholar, 40Wahle E. Lustig A. Jenö P. Maurer P. J. Biol. Chem. 1993; 268: 2937-2945Abstract Full Text PDF PubMed Google Scholar). For expression of His-tagged PABPN1 variants, pUK or pGM10 constructs were transformed into electrocompetent BL21 (DE3) pUBS520. The plasmid pUBS520 facilitates the translation of genes containing rare arginine codons by co-expression of the corresponding tRNAArg (51Brinkmann U. Mattes R.E. Buckel P. Gene. 1989; 85: 109-114Crossref PubMed Scopus (336) Google Scholar). Growth conditions, further treatment of cells, and protein purification were according to Benoit et al. (52Benoit B. Nemeth A. Aulner N. Kühn U. Simonelig M. Wahle E. Bourbon H.-M. Nucleic Acids Res. 1999; 27: 3771-3778Crossref PubMed Scopus (40) Google Scholar) with the following variations: the Ni-NTA column was washed with 50 mm sodium phosphate, 10 mm Tris-HCl, pH 8.0, and an additional step with 6 column volumes of buffer containing 75 mm imidazole. Proteins were eluted with 5 ml of buffer containing 500 mmimidazole and further purified on a 1-ml MonoQ fast protein liquid chromatography column (Amersham Biosciences) to remove nucleic acids. The variant PABPN1-ΔN160 was additionally loaded onto a Superdex-200 fast protein liquid chromatography gel filtration column to separate monomeric protein from aggregates. The His-tagged C terminus of PABPN1 was insoluble under the conditions described above. Therefore, Ni-NTA purification was performed in the presence of 8 m urea following the protocol supplied by Qiagen. Elution was done with 2 ml of buffer containing 250 mm imidazole, 8 murea, 100 mm sodium phosphate, 10 mm Tris-HCl, adjusted to pH 8.0. The protein concentration of the His-tagged C terminus was measured photometrically (1 A280 = 530 ॖg/ml). Bovine poly(A) polymerase was a kind gift of Georges Martin (Biozentrum, Basel, Switzerland). DNA modifying enzymes were from New England Biolabs. Homopolymers were from Sigma, and E. coliribosomal RNA was from Roche Diagnostics. Chemically synthesized RNA oligonucleotides were from IBA (Göttingen, Germany). Size-fractionated homopolymers were prepared by gel purification (45Keller R.W. Kühn U. Aragon M. Bornikova L. Wahle E. Bear D.G. J. Mol. Biol. 2000; 297: 569-583Crossref PubMed Scopus (56) Google Scholar) or by ion exchange chromatography (40Wahle E. Lustig A. Jenö P. Maurer P. J. Biol. Chem. 1993; 268: 2937-2945Abstract Full Text PDF PubMed Google Scholar). RNA concentrations were determined as described (36Wahle E. J. Biol. Chem. 1995; 270: 2800-2808Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 40Wahle E. Lustig A. Jenö P. Maurer P. J. Biol. Chem. 1993; 268: 2937-2945Abstract Full Text PDF PubMed Google Scholar). 5′-Labeling of RNA was performed using [γ-32P]ATP (Amersham Biosciences) and T4 polynucleotide kinase according to standard procedures (53Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1989Google Scholar). Incorporated radioactivity and RNA yields were measured using DE81 filter binding (54Stayton M. Kornberg A. J. Biol. Chem. 1983; 258: 13205-13212Abstract Full Text PDF PubMed Google Scholar). Filter binding assays were carried out essentially as described (40Wahle E. Lustig A. Jenö P. Maurer P. J. Biol. Chem. 1993; 268: 2937-2945Abstract Full Text PDF PubMed Google Scholar). For determination of the binding constants, 112 fmol (as mononucleotides) of radioactively labeled RNA was incubated with increasing amounts of PABPN1 variants in 40 ॖl of RNA binding buffer (50 mm Tris-HCl, pH 8.0, 107 glycerol, 0.2 mg/ml methylated bovine serum albumin, 0.017 Nonidet P-50, 1 mm EDTA, 1 mm dithiothreitol, 100 mm KCl). After 30 min of incubation at room temperature, 35 ॖl of each reaction were applied to nitrocellulose filters (Schleicher & Schuell) pre-treated with 1 ml of wash buffer (10 mm Tris-HCl, pH 8.0, 100 mm NaCl) containing 5 ॖg/ml rRNA. After rinsing with 5 ml of ice-cold wash buffer, the filter-bound radioactivity was measured by scintillation counting. Apparent KD or K50 values were determined both from direct and double-reciprocal plots. For the electrophoretic mobility shift assay, 5′-labeled gel-purified RNA was incubated with increasing amounts of PABPN1 variants in 20 ॖl of RNA binding buffer (see above). After incubation for 30 min at room temperature, 15-ॖl aliquots of the reactions were loaded onto a native agarose/polyacrylamide composite gel (55Rüegsegger U. Beyer K. Keller W. J. Biol. Chem. 1996; 271: 6107-6113Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). Gels were dried and analyzed with a PhosphorImager (Amersham Biosciences). C-spiked A12 was made by IBA (Göttingen) using 957 A- and 57 C-precursors for each step of synthesis. As the adenosine at the 3′-end was covalently coupled to the support during synthesis, there was no C substitution at this position. Gel-purified C-spiked A12and homogeneous A12 used as a control were 5′-labeled, and about 2.5 nm labeled oligonucleotides (as 5′-ends) were incubated at ambient temperature for 15 min in 100 ॖl of RNA binding buffer containing different concentrations of calf thymus PABPN1. The reaction mixtures were applied to a pre-treated nitrocellulose filter and washed once with 2 ml of ice-cold wash buffer (see above). Each filter was then treated for 30 min at 37 °C with 20 ॖg of proteinase K (Merck) in 300 ॖl of elution buffer containing 100 mm Tris-HCl, pH 7.5, 12.5 mm EDTA, 150 mm NaCl, 17 SDS, and 2 ॖg of rRNA. The eluted RNA was precipitated with 3 volumes of ethanol and digested for 30 min at 37 °C with 1 ng of RNase A in 10 ॖl of 5 mm Tris-HCl, pH 8.0, 1 mm EDTA. The recovered radioactivity was determined by scintillation counting of 1-ॖl aliquots. Equal amounts of radioactivity were analyzed on a 207 polyacrylamide gel (40-cm-long) containing 8.3 m urea. Autoradiography and quantification of RNA fragments was done with the help of a PhosphorImager (Amersham Biosciences). Digestion of the C-spiked A12 was complete, and unsubstituted A12 was found to be resistant to RNase A under the conditions used. Desalted 5-iodo-UMP-modified oligonucleotides (A2-5iU-A10 and A10-5iU-A2) were purchased from IBA (Göttingen) and used without additional purification for 5′-labeling with minimal exposure to light. Binding reactions were done in 100 ॖl of RNA binding buffer (see above) that contained 2 nm (as 5′-ends) of either of the two modified oligonucleotides in the presence of either 50 nm His-tagged wild-type PABPN1 or 500 nm of the deletion variants, respectively. After 10 min of incubation at room temperature, each binding reaction was evenly distributed to five wells of a 96-well microtiter plate. The plate was placed on top of an ice-cooled aluminum block at a distance of 4–4.5 cm to an inverted UV table (Fluolink; Renner GmbH). After 30 min of irradiation at 312 nm, the aliquots from each binding reaction were recombined, and 20 ॖl of each irradiated RNA/protein mix were digested with 200 ng of protease Lys-C (sequencing grade; Roche Diagnostic) at ambient temperature. Aliquots were taken as indicated and analyzed on a 107 Tricine-SDS-
Referência(s)