Identification of Proteins Bound to RNA

Identification of Proteins Bound to RNA linearized DNA sequence T7 promoter + in vitro transcription of RNA and treatment with sodium m-periodate binding to adipic acid dehydrate beads binding to adipic acid dihydrazide beads incubation with protein extract repeated cycles of centrifugation and washing SDS-PAGE analysis (and Coomassie staining) Nanospray mass spectrometric analysis Western blot Outcome: System to identify proteins that bind RNA. Question answered: Which RNA-binding proteins bind a particular RNA sequence? j 27 Identification of Proteins Bound to RNA Emanuele Buratti Abstract Defects at the level of pre-mRNA processing pathways are a major cause of human disease. Until now, most of these mutations have been detected in the relatively conserved basic splicing elements such as the donor, acceptor, and branch site regions, where they affect the binding of well-known splicing determinants such as snRNP factors. Increasingly, however, splicing mutations are being described within intronic and exonic regions of the pre-mRNA molecule, far from the canonical splicing signals, where they disrupt the binding of accessory splicing regulatory proteins. As these proteins play crucial roles in determining alternative and constitutive splicing levels, the establishment of their identity becomes essential to differentiate between potentially harmful mutations and harmless polymorphisms. Moreover, it allows a better understanding of the pathological mechanisms and, eventually, to plan for specific therapeutic strategies. In this chapter, brief practical guide is provided to identify these proteins, using an easy-to-set-up affinity purification procedure. In this technique, any RNA sequence of interest can be used to derivatize agarose beads that are then incubated with protein mixes/cellular extracts to identify potentially interacting factors. 27.1 Theoretical Background RNA-binding proteins (RBPs) regulate all aspects of post-transcriptional gene expression by affecting the biogenesis, stability, function, transport, and localization of all cellular RNAs produced in the eukaryotic nucleus (for a recent review, see Ref. [1] &&and Chapter 4 Allain&&). These aggregates between RBPs and cellular RNAs – which are referred to as ribonucleoprotein (RNP) complexes – are generally formed by stacking, electrostatic, and hydrogen-bonding interactions between regions of the various RBP proteins and selected nucleotides of an RNA molecule. In the case of proteins, the regions responsible for the direct interaction are often arranged in evolutionarily conserved motifs that provide a specific three-dimensional conformation. To date, several major types of RNA-binding structure have been described to mediate RNA–protein recognition: double-stranded RNA-binding motif (dsRBM); the Pumilio (Puf) homology domain (HD) and RGG repeats; zinc-binding domains; KH domains; and RNA recognition motif (RRM) domains. The molecular mechanisms that make some of these domains particularly suitable to bind specific RNA sequences/structures, and the way in which they differ from each other, have been the subjects of numerous structural studies. These have been recently reviewed in a number of reports, to which the reader is referred for additional details [2,3] (see also &&Chapter 4 Allain&&). The flexibility of RNA target sequences and the presence of many RBP proteins in vertebrates means that, very probably, every RNA present in the eukaryotic nucleus will be complexed with a variety of proteins in a more or less specific fashion. Beside Alternative pre-mRNA Splicing: Theory and Protocols, First Edition. Edited by Stefan Stamm, Chris Smith, and Reinhard Lührmann. Ó 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2012 by Wiley-VCH Verlag GmbH & Co. KGaA. 295 296 j 27 Identification of Proteins Bound to RNA the primary nucleotide sequence, a major modifier of RNA protein-binding properties is represented by the presence of RNA secondary structures [4]. For example, it has been shown recently that proteins such as MBNL1 and U2AF65 can selectively compete for binding to the same RNA region, depending on the presence of mutually exclusive RNA structures [5]. With regards to RBPs involved in splicing regulation, a place of honor should be reserved for the well-known class of hnRNP factors that are among the most abundant RNA-binding proteins in the human nucleus, and are responsible for forming the core of most RNP complexes described to date [6–8]. Another class of RBPs that is important for splicing regulation (though not only this) is represented by the serine–arginine (SR) class of factors that, in many cases, operate antagonistically to hnRNP proteins in regulating the splicing processes [9–11]. It is the combinatorial presence of all these factors, often binding very close to each other on a very narrow stretch of RNA sequence (such as a typical exon or selected regions therein), that determines the final functional outcome [12,13] (&&see Chapter 3, Hertel&&). In this respect, it should be noted that the unraveling of complex RNA–protein compositions is important not only to understand how pre-mRNA or mRNA molecules are processed. It is now clear that, in order to function correctly, all of the small RNA families that have been discovered during recent years (small nuclear RNAs, small nucleolar RNAs, microRNAs, siRNAs, and shRNAs) – including the ever-increasing number of regulatory noncoding RNAs (ncRNAs) – are known to assemble as RNP complexes, the composition of which is almost certain to regulate several aspects of their expression pathway and functional properties [14] &&(see Chapter 2, Meister).&& Last, but not least, mention should also be made of the increasing role played by RNA-binding proteins in the pathological mechanisms mediated by pathogenic RNAs that result from the expansion of repeats in noncoding and coding regions [15]. Taken together, it is clear that the methodologies used to unravel the composition of RNP complexes represent an essential tool in present-day research. Paradoxically, some of the more useful methodologies are based on classical biochemical techniques that had almost gone out of fashion during the late 1980s. Yet, this unexpected revival has been made possible by recent advances in mass spectrometric analysis [16]. Today, these novel techniques allow the protein compositions of RNP complexes to be obtain with a speed, resolution, accuracy, and economic cost that makes their use “affordable” for many laboratories. The affinity purification technique described here has been used successfully for the identification of splicing regulatory factors in various NF-1 donor sites [17], repeat sequences [18,19], splicing regulatory regions [20–22], and pseudoexon sequences [23]. Further details on the potential application and results obtained with the technique are available in these publications. 27.2 Protocol 27.2.1 RNA Templates RNA templates can obtained in the following ways: 1) 2) They can be synthesized by commercial suppliers; this is generally the best course of action for sequences less than 50 nucleotides (nt) in length (Biosyn; Sigma-Aldrich). The sequence of interest can be cloned in a pBluescript KS þ plasmid (Stratagene) or any other plasmid that contains a T7 promoter. In this case, it is advisable to ensure that the 50 end of the sequence to be transcribed is placed as close as 27.2 Protocol possible to the end of the T7 promoter, and that a suitable restriction enzyme to linearize the plasmid is present at its 30 end. This is to minimize the length of plasmid-related RNA that will eventually be transcribed together with the sequence of interest. In this case, it is important to choose an enzyme that cleaves efficiently and leaves a 50 overhang (i.e., BamH1, HindIII, XbaI). 3) By amplifying the sequence of interest using a forward primer carrying a T7 polymerase target sequence at the 50 end and 12–15 complementary nucleotides at the 30 end (50 -taatacgactcactatagg(n)12–15-30 ), and a reverse primer carrying 12–15 nt of the target sequence. The products from steps 2 and 3 should be purified by phenol/chloroform extraction, precipitated using standard protocols (1/10 3 M NaAc, 2.5 volumes ethanol, 20  C for 1 h), and resuspended in RNase-free water to a concentration of approximately 1 mg ml1. The procedure is continued as follows: 4) Approximately 2 mg of linearized plasmids/amplified products are transcribed using T7 RNA Polymerase (Stratagene) in the presence of transcription buffer (350 mM HEPES, pH 7.5, 30 mM MgCl2, 2 mM spermidine, 40 mM dithiothreitol; DTT), 40 units RNasin, 7 mM each of the four NTPs, and 60 units of T7 polymerase (1.5 U mg1). 5) In general, three 40 ml reactions should be perform for each RNA of interest, placing each reaction in a 1.5 ml Eppendorf tube together with 2 mg of linearized plasmids/amplified products. 6) Following incubation for 2 h at 37  C, the reactions are pooled, purified by a cycle of acid phenol/chloroform extractions, precipitated according to standard protocols, and then resuspended in 40 ml RNase-free water. Usually, this approach yields the desired 15 mg of transcribed RNA for the following steps (see Section 27.2.2). At this stage it is strongly recommended that the production/ integrity of the RNA is checked on a standard agarose gel. 27.2.2 Loading the Beads with RNA 1) 2) 3) 4) 5) Q1 6) 7) 8) The 500 pmol of T7-transcribed RNA (ca. 15 mg of a 100-mer RNA) previously dissolved in 40 ml of water are placed in a 1.5 ml Eppendorf test tube. To each sample is added 360 ml of a 5 mM sodium m-periodate (Sigma, #S11448) solution in 0.1 mM NaOAc, pH 5.0 (to prepare 50 ml of this reagent, dissolve 53 mg sodium m-periodate in 0.1 M NaOAc, pH 5). Note: this reagent must be prepared freshly before use. The 400 ml reaction mix is incubated for 1 h in the dark (each test tube should be wrapped in aluminum foil) at room temperature, in a rotator wheel. Each RNA is then ethanol-precipitated according to standard protocols, washed once with 70% ethanol, and resuspended in 100 ml of 0.1 M NaOAC, pH 5.0. Note: The pellet will be very small; great care is required for its handling. In the meantime, 100 ml of adipic acid dehydrazide/agarose bead 50% slurry (Sigma, #A0802) is taken for each RNA sample to be conjugated, and placed in a 15 ml Falcon tube. The beads are washed four times with 10 ml of 0.1 M NaOAc, pH 5.0, and pelleted by centrifugation at 3000 rpm (1200  g) for 5 min in a clinical centrifuge (e.g., Eppendorf 581R). After the final wash, the bead pellet is resuspended in a 10 ml tube, using 300 ml of 0.1 M NaOAc, pH 5.0, for each RNA sample prepared in step 4. After mixing well, separate 300 ml aliquots are taken and added to the 100 ml of periodate-treated RNA from step 4. The resulting 400 ml mixture is incubated overnight in the dark at 4  C on a rotator (the tubes should be wrapped in aluminum foil). j 297 298 j 27 Identification of Proteins Bound to RNA 27.2.3 Incubation with Protein Mix (Buffer A) 1) After the overnight incubation, the beads are pelleted by centrifuging at 4000 rpm for 5 min (bench-top Eppendorf minifuge; from this stage on the 5415 D centrifuge is used). The RNA-loaded beads will tend to cling to the side of the test tube, and must be shaken free so that they collect at the tube bottom (this is done by tapping the tubes gently on the side of the rack). 2) The supernatant is discarded and the RNA-loaded beads washed twice with 2 M NaCl. The beads are then deposited at 4000 rpm for 5 min in the minifuge. 3) The beads are washed three times with 1.0 ml of Sol.D 1 (20 mM HEPES, pH 7.9, 100 mM KCl, 0.2 mM EDTA, pH 8.0, 100 mM DTT, 6% (v/v) glycerol), and then deposited at 4000 rpm for 5 min (minifuge). The supernatant is discarded after each wash. 4) During the final spin of step 3, the following 500 ml mix is prepared for each RNA sample to be tested: T 50 ml Sol.D 10 (200 mM HEPES, pH 7.9, 2 mM EDTA, pH 8.0, 1 M DTT, 60% (v/v) glycerol). T 50 ml 1 M KCl (added separately). T 100 ml NE (ca. 10–15 mg ml1), or any other protein mix of interest. T 300 ml water. T Heparin (200 mg ml1 stock solution) is added to a desired final concentration (0.5–2.5–5.0 mg ml1 of the final volume). 5) The Nuclear Extract/Protein mix (500 ml) is added to the individual Eppendorf tubes, and the contents mixed gently by manually shaking. 6) The tubes are incubated on a rotor for 30 min at room temperature. 7) The beads are deposited at 4000 rpm (Eppendorf minifuge), and as much of the protein mix as possible is removed. 8) The beads are washed four times with 1.5 ml Sol.D 1  , by incubating them each time for 5 min on a rotating wheel at room temperature. For each wash the beads are deposited at 4000 rpm (Eppendorf minifuge) and the supernatant discarded. 9) SDS loading buffer (50 ml) is added, and the sample is denatured and loaded onto a SDS-PAGE gel. When loading, it is recommended that a glass Hamilton syringe is used, to avoid loading the beads into the well. 27.2.4 Incubation with Protein Mix (Buffer B) 1) After the overnight incubation, the beads are pelleted by centrifuging at 4000 rpm for 5 min (Eppendorf minifuge). Ensure that the RNA-loaded beads will collect at the tube bottom (tap the tubes gently on the side of the rack). 2) The supernatant is discarded and the RNA-loaded beads washed twice with 2 M NaCl. The beads are then deposited at 4000 rpm for 5 min (Eppendorf minifuge). 3) The beads are washed three times with 1.0 ml of Buffer B (5 mM HEPES, pH 7.9, 1 mM MgCl2, 0.8 mM magnesium acetate), and deposited at 4000 rpm for 5 min. The supernatant is discarded after each wash. 4) During the last spin of step 3, the following 500 ml mix is prepared for each RNA sample to be tested: T 50 ml binding buffer 10 (50 mM HEPES, pH 7.9, 10 mM MgCl2, 8 mM Mg acetate, 5.2 mM DTT, 7.5 mM GTP, 10 mM ATP, and 38% (v/v) glycerol). T 100 ml NE (ca. 10–15 mg ml1) or any other protein mix of interest. T 350 ml water. 27.4 Troubleshooting (a) RNA oligos: 5) WT ATM 5' UGGCCAGGUAAGUGAUAUAU 3' WT ∆ 5' UGGCCAG----GUGAUAUAU 3' (b) +N The Nuclear Extract/Protein mix (500 ml) is added to the individual Eppendorf tubes, and mixed gently by manual shaking. 6) The tubes are incubated on a rotor for 30 min at room temperature. 7) The beads are deposited at 4000 rpm (Eppendorf minifuge) and as much of the protein mix as possible is removed. 8) The beads are washed four times with 1.5 ml of buffer B (5 mM HEPES, pH 7.9, 1 mM MgCl2, 0.8 mM magnesium acetate) by incubating them each time for 5 min on a rotating wheel at room temperature. After each wash, the beads are deposited at 4000 rpm (Eppendorf minifuge), and the supernatant discarded. 9) SDS loading buffer (50 ml) is added, and the sample is denatured and loaded onto a SDS-PAGE gel. When loading, it is recommended that a glass Hamilton syringe is used, to avoid loading the beads into the well. 299 E +N E+ +N Ab E + ant Ab i U co 1A nt . T Heparin (200 mg ml1 stock solution) to a desired final concentration (0.5–2.5–5.0 mg ml1 of the final volume). j RNA+U1snRNP+Ab RNA+U1snRNP 27.3 Example Experiment In previous studies from the present author’s laboratory, it has been reported that the binding of a U1snRNP molecule to an intronic splicing processing element (ISPE) in intron 20 of the ATM gene was capable of inhibiting pathological pseudoexon inclusion. Inactivation of this element through a 4 nt deletion (GTAA) caused inactivation of this binding, pseudoexon inclusion, and the occurrence of ataxia telangiectasia in a patient [24]. Using synthetic RNAs carrying either the wild-type or mutated RNA sequence (Figure 27.1a), this loss in U1snRNP binding activity that was originally demonstrated through band-shift analysis (Figure 27.1b), can also be easily observed using the pulldown affinity technique (Figure 27.1c). In order to identify their identity, the bands of interest were cut from the Coomassie-stained gel. An internal sequence analysis from the Coomassie blue-stained bands excisedfromthe SDS-PAGE gel was performedusing electrospray ionization mass spectrometry (LCQ DECA XP; ThermoFinnigam). The bands weredigestedwithtrypsinandtheresultingpeptidesextractedwithwaterand60% acetonitrile/1% trifluoroacetic acid. The fragments were then analyzed with mass spectrometry, and the proteins identified by analysis of the peptide MS/MS data with Turbo SEQUEST (ThermoFinnigam) and MASCOT (Matrix Science). This example shows how it is usually better to use as a control a related RNA sequence. In fact, the low/medium amount of background present in the two lanes can even be considered a “useful” feature, as it allows specific differences to stand up more sharply and can also act as a loading/pulldown control. Of course if, rather than binding differences, the interest was focused on characterizing all RNA–protein interactions of a specific RNA sequence, then a better approach would have been to use naked beads or beads loaded with a completely unrelated RNA (usually the antisense strand of the intended target). 27.4 Troubleshooting Problem Reason þ Solution Too-strong protein binding signals or background in beads too high Increase heparin concentration added to the protein mix Shorten size of RNA targets bound to the beads (ideal length is normally less than 200 nt) RNA Labeled ATM WT (c) RNAs bound to agarose beads WT WT ∆ U1-70K U1-A SmRNP B/B' Coomassie staining Fig. 27.1 Example of RNA pulldown experiment using synthetic RNA oligonucleotides. (a) The synthetic RNAs that carry either the wild-type (ATM) or the deleted sequence (ATM D); (b) A band-shift experiment using labeled ATM WT RNA (lane 1, left) incubated in the presence of nuclear extract (lane 2), nuclear extract plus an antibody specific against the U1snRNP U1A protein (lane 3), and a nuclear extract plus a control antibody (lane 4, right). The samples were run on a 6% nondenaturing PAGE gel; (c) A pulldown analysis using the ATM WT (lane 1, left) and ATM D (lane 2, right) RNAs bound to the adipic acid dehydrazide beads following incubation with a commercial HeLa nuclear extract. Following the addition of SDS-PAGE running buffer, the beadderived proteins were separated in a 12% denaturing SDS-PAGE gel and stained with Coomassie blue, according to standard protocols. The bands indicated by arrows refer to the several U1snRNP components that are differentially bound to these two RNAs (U170K, U1A, and SmRNP proteins B and B0 ), as determined by mass spectrometric analysis. 300 j 27 Identification of Proteins Bound to RNA Ensure that the protein mix added to the mix is NOT cloudy. If, after heparin addition, the solution does not clear to near-transparency, it is advisable to centrifuge briefly (ca. 5 min at 4000 rpm in Eppendorf minifuge) and to discard any eventual pellet Control (i.e., empty) beads have the tendency to absorb high-molecular-weight proteins (>100 kDa) Too-weak protein binding signals to beads Failure of synthesized/synthetic RNAs binding to beads. Use fresh reagents. If the problem persists, the binding reactions to beads can be followed using a radioactively labeled RNA on a small experimental scale Decrease heparin concentration added to the protein mix Increase protein extract concentration added to the protein mix Use Binding Buffer B (see Section 27.2.3). This buffer tends to yield more protein signals than Sol. D (Warning: it will also raise background binding levels, especially with empty beads, if used as a control) Small or no differences detected in band intensities between different samples Increase the size of RNA sequence analyzed (maximum length is >1000 nt) Compare RNA sequences that display clear functional differences (i.e., gross deletion mutants, etc.) Preincubate the protein mix with semi-specific RNA competitors (in addition to heparin) Use a mass spectrometry-compatible silver stain procedure to stain SDS-PAGE gels References 1 Glisovic, T., Bachorik, J.L., Yong, J., and 2 3 4 5 Dreyfuss, G. (2008) RNA-binding proteins and post-transcriptio nal gene regulation. FEBS Lett., 582, 1977–1986. Auweter, S.D., Oberstrass, F.C., and Allain, F.H. (2006) Sequence-specific binding of single-stranded RNA: is there a code for recognition? Nucleic Acids Res., 34, 4943–4959. Chang, K.Y. and Ramos, A. (2005) The double-stranded RNA-binding motif, a versatile macromolecular docking platform. FEBS J., 272, 2109–2117. Buratti, E. and Baralle, F.E. (2004) Influence of RNA secondary structure on the premRNA splicing process. Mol. Cell. Biol., 24, 10505–10514. Warf, M.B., Diegel, J.V., von Hippel, P.H., and Berglund, J.A. (2009) The protein factors MBNL1 and U2AF65 bind alternative RNA structures to regulate 6 7 8 9 10 splicing. Proc. Natl Acad. Sci. USA, 106, 9203–9208. Krecic, A.M. and Swanson, M.S. (1999) hnRNP complexes: composition, structure, and function. Curr. Opin. Cell Biol., 11, 363–371. Martinez-Contreras, R., Cloutier, P., Shkreta, L., Fisette, J.F., Revil, T., and Chabot, B. (2007) hnRNP proteins and splicing control. Adv. Exp. Med. Biol., 623, 123–147. He, Y. and Smith, R. (2009) Nuclear functions of heterogeneous nuclear ribonucleoproteins A/B. Cell. Mol. Life Sci., 66, 1239–1256. Ram, O. and Ast, G. (2007) SR proteins: a foot on the exon before the transition from intron to exon definition. Trends Genet., 23, 5–7. Lin, S. and Fu, X.D. (2007) SR proteins and related factors in alternative splicing. Adv. Exp. Med. Biol., 623, 107–122. 11 Sanford, J.R., Gray, N.K., Beckmann, K., 12 13 14 15 16 and Caceres, J.F. (2004) A novel role for shuttling SR proteins in mRNA translation. Genes Dev., 18, 755–768. Buratti, E., Baralle, M., and Baralle, F.E. (2006) Defective splicing, disease and therapy: searching for master checkpoints in exon definition. Nucleic Acids Res., 34, 3494–3510. Cooper, T.A., Wan, L., and Dreyfuss, G. (2009) RNA and disease. Cell, 136, 777–793. Mendes Soares, L.M. and Valcarcel, J. (2006) The expanding transcriptome: the genome as the ‘Book of Sand’. EMBO J., 25, 923–931. O’Rourke, J.R. and Swanson, M.S. (2009) Mechanisms of RNA-mediated disease. J. Biol. Chem., 284, 7419–7423. Tate, E.W. (2008) Recent advances in chemical proteomics: exploring the References post-translational proteome. J. Chem. Biol., 1, 17–26. 17 Buratti, E., Baralle, M., De Conti, L., Baralle, D., Romano, M., Ayala, Y.M., and Baralle, F.E. (2004) hnRNP H binding at the 50 splice site correlates with the pathological effect of two intronic mutations in the NF-1 and TSHbeta genes. Nucleic Acids Res., 32, 4224–4236. 18 Buratti, E., Dork, T., Zuccato, E., Pagani, F., Romano, M., and Baralle, F.E. (2001) Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo CFTR exon 9 skipping. EMBO J., 20, 1774–1784. 19 Mercado, P.A., Ayala, Y.M., Romano, M., Buratti, E., and Baralle, F.E. (2005) Depletion of TDP 43 overrides the need for exonic and intronic splicing enhancers in the human apoA-II gene. Nucleic Acids Res., 33, 6000–6010. 20 Skoko, N., Baralle, M., Buratti, E., and Baralle, F.E. (2008) The pathological splicing mutation c.6792C > G in NF1 exon 37 causes a change of tenancy between antagonistic splicing factors. FEBS Lett., 582, 2231–2236. 21 Pagani, F., Buratti, E., Stuani, C., and Baralle, F.E. (2003) Missense, nonsense, and neutral mutations define juxtaposed regulatory elements of splicing in cystic fibrosis transmembrane regulator j 301 exon 9. J. Biol. Chem., 278, 26580–26588. 22 Marcucci, R., Baralle, F.E., and Romano, M. (2007) Complex splicing control of the human Thrombopoietin gene by intronic G runs. Nucleic Acids Res., 35, 132–142. 23 Raponi, M., Buratti, E., Llorian, M., Stuani, C., Smith, C.W., and Baralle, D. (2008) Polypyrimidine tract binding protein regulates alternative splicing of an aberrant pseudoexon in NF1. FEBS J., 275, 6101–6108. 24 Pagani, F., Buratti, E., Stuani, C., Bendix, R., Dork, T., and Baralle, F.E. (2002) A new type of mutation causes a splicing defect in ATM. Nat. Genet., 30, 426–429. Keywords Dear Author, Keywords will not be included in the print version of your chapter but only in the online version. Please check and/or supply keywords. Keywords: RNA; RNA-binding proteins; affinity purification; hnRNP; SR proteins; U1snRNP.; Author Query 1. Please check this value. You say that 4000 rpm = 1700g?