mRNA Processing and Metabolism

Volume 257 of the series Methods in Molecular Biology™ pp 47-64

Tobramycin Affinity Tag Purification of Spliceosomes

  • Klaus HartmuthAffiliated withDepartment of Cellular Biochemistry, Max Planck Institute for Biophysical Chemistry
  • , Hans-Peter VornlocherAffiliated withRibopharma AG
  • , Reinhard LührmannAffiliated withDepartment of Cellular Biochemistry, Max Planck Institute for Biophysical Chemistry



The ability to isolate native ribonucleoprotein (RNP) particles is of fundamental importance in the study of processes such as pre-messenger RNA (mRNA) processing and translation. We have developed an RNA affinity tag that allows the large-scale preparation of native spliceosomes in a solid-phase assembly scheme. A tobramycin-binding aptamer cotranscriptionally added to the 3′ end of the pre-mRNA is used to bind the pre-mRNA to tobramycin immobilized on a matrix. Incubation of the pre-mRNA thus immobilized allows the assembly of spliceosomes, which can be released from the matrix under native conditions by competition with tobramycin. Further density-gradient centrifugation affords highly purified spliceosomes suitable for the characterization of associated proteins by mass spectrometry as well as for studies using biochemical and biophysical methods. Although the method was developed for the preparation of spliceosomes, it is likewise applicable to the preparation of other RNP particles.

Key Words

RNA affinity tag tobramycin splicing pre-mRNA RNP spliceosome pre-spliceosome aminoglycoside antibiotic aptamer solid-phase spliceosome assembly glycerol-gradient centrifugation

1 Introduction

The splicing of pre-messenger RNA (mRNA), which yields mRNA, is catalyzed by the spliceosome, which is composed of U snRNPs (U1, U2, U4/U6.U5) and a large number of proteins (for review, see refs. 14). The spliceosome is a highly dynamic molecular machine, which forms anew on each intron in a process termed the spliceosomal cycle. During this cycle, distinct subcomplexes, namely complexes E, A, B, and C, can be discerned in vitro (for review, see ref. 3). The A complex—also called the prespliceosome—is composed of pre-mRNA, U1 and U2 snRNPs, and a large number of additional non-snRNP proteins (5). Integration of the U4/U6.U5 tri-snRNP into the A complex leads to formation of the precatalytic B complex. A hallmark feature of this complex is the presence of all U snRNAs. Subsequent dynamic RNA-RNA rearrangements (2) result in the formation of the C complex, which lacks U1 and U4 snRNA. The first step of pre-mRNA splicing occurs concomitantly with formation of the C complex, and the C complex itself then catalyzes the second step. The spliceosome finally disintegrates, releasing the mRNA and intron-lariat products as RNP particles.The isolation of defined functional stages of the spliceosomal cycle is of fundamental importance for understanding the molecular biology and biochemistry of pre-mRNA splicing. Recently, a number of affinity-based selection procedures have been developed to this end. One such procedure uses a pre-mRNA into which MS2 coat protein binding sites have been inserted. An MS2:maltose binding fusion protein is then used as an affinity tag to isolate spliceosomal complexes from a solution-phase splicing reaction through reversible interaction with an amylose resin (68). Another approach is the immunoaffinity selection of spliceosomes with a peptide antibody directed against a protein that is known to associate with the spliceosome at a defined assembly stage. Spliceosomes can then be released form the antibody by competition with the peptide. This approach was recently shown to be a powerful tool for the isolation of functional activated mammalian spliceosomes (9). We have developed a method for affinity purification of spliceosomes by using an RNA affinity tag (10). The tag is a tobramycin-binding aptamer that is added cotranscriptionally to the pre-mRNA. The tagged pre-mRNA binds to a tobramycin matrix with high specificity and can be released by competition with tobramycin. In the method outlined below, this pre-mRNA is used in a solid-phase assembly procedure for the preparation of spliceosomes. Major advantages of the method are the fact that it uses readily available, inexpensive reagents, and the ease with which the procedure is conducted. Furthermore, the procedure can be scaled up to any desired level.

1.1 The Tobramycin RNA Affinity Tag

Generally, an RNA affinity tag for use in the isolation of native RNP particles should conform to the following criteria: (1) the sequence must be short; (2) it must have a high affinity for a given ligand immobilized on a solid support; (3) the binding of the RNA affinity tag to the immobilized ligand must be stable under the conditions of the reaction studied; and (4) the interaction of the tagged RNA must be reversible under conditions that preserve the integrity of the RNP under study. Numerous short RNA aptamers are known that bind aminoglycosidic antibiotics reversibly, with moderate to high affinities (1015). For our studies, we focused on the 40-nt-long tobramycin-binding J6f1 RNA aptamer (10) because of its high affinity for tobramycin (KD approx 5 nM) at 145 mM KCl and 1.5 mM Mg2+.

The J6f1 aptamer was attached to the 3′ end of the pre-mRNA by cloning it into the transcription template ( Fig. 1 A and B; see Note 1 ). The pre-mRNA thus modified was bound to a tobramycin matrix, with high specificity, through the attached tobramycin aptamer ( Fig. 1 C). Furthermore, it was efficiently processed in a standard in vitro splicing reaction ( Fig. 1 D). However, we were unable to select the pre-mRNA from the solution-phase splicing reaction by passing it over a tobramycin matrix (see Note 2 ). We therefore devised a solid-phase spliceosome assembly procedure ( Fig. 2 A). Here, the pre-mRNA is first immobilized on the matrix, and then spliceosome assembly is initiated. After completion of the reaction, nonspecific components are washed from the matrix and spliceosomes are released by competition with tobramycin. Figure 2 B shows the kinetics of a typical splicing reaction taking place on the matrix. A dramatic reduction of the reaction velocity is observed in comparison with the splicing reaction in solution ( Fig. 1 D). This is inferred from the appearance of intermediates (e.g., the “exon 2-intron” lariat) and products (mRNA and intron lariat). By choosing the appropriate reaction time, large amounts of a specific spliceosomal complex can thus be prepared by scaling up the solid-phase splicing reaction. Because we were interested in the spliceosomal A complex, we chose to terminate the reaction after 45 min, well before the first appearance of intermediates at around 90 min ( Fig. 2 B, lane 2).
Fig. 1.

(A), Diagrammatic representation of the pre-mRNA. The lengths of exons (boxes) and the intron (line) are indicated. The dashed box highlights schematically the J6f1 tobramycin-binding aptamer attached to the 3′ end. (B), Secondary structure of the 3′ end of the pre-mRNA. The tobramycin aptamer attached to the 3′ end of the pre-mRNA is enclosed in the box and flanking sequences are shown. The BamHI site, used to cleave the J6f1 aptamer from the transcription template (see Subheading 3.1.2 ) is indicated. (C), Specificity of the interaction of the tobramycin-aptamer-tagged pre-mRNA with the tobramycin matrix. The radioactivity of the fractions described in the text (see Subheading 3.3.2 ) is shown. Fractions referred to as “bound” and “matrix” correspond to the amounts of the different RNAs that remained bound to matrix after washing and elution, respectively. (D), Kinetics of the solution-phase splicing reaction of the pre-mRNA (see Note 5 ). The RNA was analyzed on a 10% polyacrylamide/8.3 M urea gel.
Fig. 2.

(A), Outline of spliceosome assembly on the matrix. T, tobramycin. (B), Kinetics of the solid-phase spliceosome assembly. The reaction was performed for the times indicated under the conditions outlined in Subheading 3.4. except that the reaction was scaled down by a factor of two. RNA was extracted (see Subheading 3.5. ) and analyzed on a denaturing 15% polyacrylamide/8.3 M urea gel. To visualize the minor RNA species present in the supernatant, three times more supernatant was analyzed than eluate. The position of the pre-mRNA and splicing intermediates/products are indicated on the right. I, input pre-mRNA. (Reprinted with permission from ref. 5, copyright 2002, National Academy of Sciences.)

1.2 Fractionation of Spliceosomes by Glycerol Gradient Centrifugation

The crude eluate from the tobramycin matrix consists of a mixture of spliceosomal complexes with the A complex as the predominant species. In addition, minor nonspliceosomal contaminants are present. A further fractionation is therefore required if pure spliceosomal A and/or B complexes are to be prepared. For this purpose, the eluate from the tobramycin matrix is fractionated by ultracentrifugation on a glycerol gradient (see Note 3 ). This affords a clean separation of the spliceosomal A and B complexes and the lesser contaminating proteins. The assignment of the spliceosomal A and B complexes to particular fractions is done by first inspecting the RNA pattern of the individual fractions ( Fig. 5 B). The position of the spliceosomal A complex can be narrowed down further by performing psoralen crosslinking (16) on the gradient fractions (ref. 5; see Note 4 ).
Fig. 5.

Gradient fractionation of spliceosomes assembled on the matrix. (A), Distribution of the pre-mRNA, expressed as pmol, across the gradient. The fraction numbers are indicated above the lanes; P, pellet fraction. (B), RNA composition of the gradient fractions. Above the lanes, the amount of RNA analyzed from the total RNA recovered from a 100-μL gradient fraction is expressed as a the fraction of the total. Fractions corresponding to the spliceosomal A and B complex are indicated at the bottom. The RNA species are labeled on the right. Contaminating RNAs are indicated with asterisks.

2 Materials

2.1 Materials and Basic Buffers

  1. 1.

    Oligonucleotide primers.

  2. 2.

    pGEM-3Zf(+) vector (Promega).

  3. 3.

    pMINX clone with the MINX pre-mRNA (17).

  4. 4.

    Restriction enzymes, Taq polymerase, T7 RNA polymerase, RNase-free DNase I.

  5. 5.

    Agarose gel and DNA-sequencing equipment.

  6. 6.

    m7GpppG (Kedar sc, Warsaw, Poland).

  7. 7.

    S-300 HR spin columns (Amersham Biosciences).

  8. 8.

    Coupling buffer: 0.2 M NaHCO3, 0.5 M NaCl, pH 8.3 (NaOH).

  9. 9.

    Tobramycin (Fluka); it is dissolved in coupling buffer (buffer 8) at a final concentration of 100 mM. The solution is stable at −20°C.

  10. 10.

    NHS-activated Sepharose, fast flow (Amersham Biosciences).

  11. 11.

    10% NaN3.

  12. 12.

    10 mg/mL tRNA (Escherichia coli; Roche).

  13. 13.

    20 mg/mL Bovine serum albumin (BSA) (Roche).

  14. 14.

    Gradient mixer (Bio Comp).

  15. 15.

    HeLa cell nuclear extract, prepared according to Dignam (18) in Dignam’s buffer D (20 mM HEPES-KOH, pH 7.9; 100 mM KCl; 1.5 mM MgCl2; 0.2 mM ethylenediamine tetraacetic acid (EDTA), pH 8.0; 0.5 mM dithiothreitol (DTT); 0.5 mM PMSF; 10% glycerol); see Note 5 for a standard activity assay.

  16. 16.

    0.1 M ATP.

  17. 17.

    0.85 M Creatine phosphate.

  18. 18.

    Hybridization oven.

  19. 19.

    Rotating device for slow head-over-tail inversion of tubes.

  20. 20.

    Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) equipment.

  21. 21.

    Denaturing PAGE equipment.

  22. 22.

    CE buffer: 10 mM cacodylic acid-KOH, pH 7.0; 0.2 mM EDTA, pH 8.0.

  23. 23.

    1 M Tris-HCl, pH 8.1, at 21.5°C (add 22.5 mL of 37% HCl for each 0.5 L of 1 M Tris base).

  24. 24.

    1 mM HCl.

  25. 25.

    1 M CaCl2.

  26. 26.

    1 M MgCl2.

  27. 27.

    0.5 M DTT.

  28. 28.

    Phosphate-buffered saline (PBS): 130 mM NaCl, 20 mM potassium phosphate, pH 8.0.

  29. 29.

    PBS containing 0.02% NaN3.

  30. 30.

    1% NP-40 (Nonidet P-40).

  31. 31.

    10X G75: 200 mM HEPES-KOH, pH 7.9; 750 mM KCl, 15 mM MgCl2.

  32. 32.

    10% Glycerol (v/v) in 1X G75.

  33. 33.

    30% Glycerol (v/v) in 1X G75.


2.2 Buffers That Must Be Freshly Prepared Before Use

  1. 1.

    Blocking buffer A: 0.2 M NaHCO3, 0.1 M NaCl, 1 M ethanolamine, pH 8.0 (HCl).

  2. 2.

    4X Binding buffer (4X BP): 80 mM Tris-HCl (pH 8.1 at 21.5°C), 4 mM CaCl2, 4 mM MgCl2, 0.8 mM DTT (see Note 6 ).

  3. 3.

    Blocking buffer B: 1X BP, 300 mM KCl, 0.1 mg/mL tRNA, 0.5 mg/mL BSA, 0.01% NP-40.

  4. 4.

    Binding buffer: 1X BP, 145 mM KCl, 0.1 mg/mL tRNA.

  5. 5.

    145 mM KCl washing buffer (W145): 1X BP, 145 mM KCl, 0.1% NP-40.

  6. 6.

    Elution buffer (E145T): 1X BP, 5 mM tobramycin, 145 mM KCl, 2 mM MgCl2.

  7. 7.

    75 mM KCl washing buffer (W75): 1X BP, 75 mM KCl, 0.1% NP-40.


3 Methods

3.1 Template Construction

The generation of the transcription template for the synthesis of the tobramycin aptamer-tagged pre-mRNA is outlined in Subheadings 3.1.1. and 3.1.2. This comprises cloning of the J6f1 tobramycin aptamer into the 3′ end of the pre-mRNA template and large-scale polymerase chain reaction (PCR) amplification of the transcription template suitable for synthesis of the tagged pre-mRNA ( Fig. 2 ). Unless otherwise stated, all molecular–biological procedures should be performed according to ref. 19 or, when a commercial kit is used, according to manufacturers’ instructions.

3.1.1 Cloning Procedures

The J6f1 aptamer (10) was attached to the 3′ end of the pre-mRNA transcription template (see Note 7 ) in two steps. A pre-mRNA coding sequence was brought under the control of the T7 RNA polymerase promoter by first excising the EcoRI–BamHI fragment from the pMINX plasmid originally described by Zillmann et al. (17). The fragment was then transferred to the pGEM-3Zf(+) vector, resulting in the pGEM-MINX plasmid. The following oligonucleotides were used to insert the J6f1 aptamer: HPV23 GATCCGGCTT AGTATAGCGAGGTTTAGCTACACTCGTGCTGAGCCGGATCCGCATG and HPV24 CGGATCCGGCTCAGCACGAGTGTAGCTAAACCTCGCTAT ACTAAGCCG containing the aptamer coding and noncoding sequences, respectively. The two oligonucleotides were annealed and cloned through their sticky BamHI/SphI ends into BamHI/SphI-linearized pGEM-MINX, yielding pGEM-MINX-T5. The correctness of the complete construct was verified by DNA sequencing.

3.1.2 PCR Amplification of the Transcription Template

The transcription template for the (+)-aptamer pre-mRNA is generated in a PCR by using the pGEM-MINX-T5 template and the oligonucleotides HPV24 (see Subheading 3.1.1. ) and M13f (GTAAAACGACGGCCAGT). The latter oligonucleotide primes upstream of the T7 promoter sequence. Ten 100-μL PCRs are performed (conditions: 94°C, 30 s; 30 cycles of 94°C, 30 s, 62°C, 30 s, 68°C, 30 s; final 68°C, 2 min), using 2 μg/mL template and 1 μM of each oligonucleotide. The PCR product is fractionated by 1% agarose gel electrophoresis and further purified (QIAquick Gel extraction kit, Qiagen). To generate the template for the (−)-aptamer, pre-mRNA control used in Subheading 3.3.2 , the (+)-aptamer pre-mRNA template is cleaved with BamHI ( Fig. 2 A).

3.2 RNA Synthesis

The (+)-aptamer and (−)-aptamer pre-mRNAs are synthesized with T7 RNA polymerase (MegaScript, Ambion) from the templates prepared in Subheading 3.1.2 Capping of the transcripts is achieved by inclusion of 5 mM m7GpppG and lowering the GTP concentration to 1.5 mM. ATP, CTP, and UTP are each added, to a concentration of 7.5 mM of each, and the RNA was tracer-labeled by adding [α-32P] (3000 Ci/mmol) to a concentration of 0.23 μM. After transcription (approx 6 h) and DNase I treatment, transcripts are purified by overnight precipitation at −20°C in the presence of 2.5 M LiCl and subsequent spin-column chromatography (S-300 HR column, Pharmacia). The RNA is dissolved at approx 10 pmol/μL in CE buffer and stored at −20°C.

3.3 Preparation and Testing of the Tobramycin Matrix

Subheading 3.3.1. details the preparation of the tobramycin matrix in an adaptation of the protocol by Wang and Rando (11), and Subheading 3.3.2 outlines the procedure that is used to test the binding activity of the matrix with the two RNAs prepared according to Subheading 3.2.

3.3.1 Preparation of the Tobramycin Matrix

Freshly prepare blocking buffer A (buffer 1, Subheading 2.2. ). The procedure is conveniently performed in a 12-mL capped tube, and centrifugation is performed in a minifuge. All manipulations are performed at 4°C unless otherwise stated.

  1. 1.

    Wash 2 mL (packed volume) NHS-activated Sepharose (fast flow) four times with 9 mL of 1 mM HCl by gentle resuspension, brief centrifugation at 250g (1 min), and by carefully decanting the supernatant.

  2. 2.

    Prepare 1 mL of a 5-mM tobramycin solution by mixing 50 μL of tobramycin in coupling buffer with 950 μL of coupling buffer. Then add this mixture to the matrix (see Note 8 ). Incubate by head-over-tail rotation overnight.

  3. 3.

    Centrifuge at 250g (5 min), remove the supernatant, and add 8 mL of blocking buffer A. Incubate by head-over-tail rotation for 2 h at room temperature.

  4. 4.

    Centrifuge at 250g (5 min), decant the supernatant, and wash three times with PBS and twice with PBS containing NaN3. Store the tobramycin matrix in 2 mL of PBS containing NaN3 at 4°C. The tobramycin matrix is stable for approx 3 mo.


3.3.2 Testing of the Tobramycin Matrix

This section describes the procedure for testing the tobramycin matrix. Furthermore, the specificity of the interaction between the tobramycin aptamer on the RNA and the matrix-bound tobramycin is assayed. It is crucial for the solid-phase splicing procedure that the RNA only binds through the tobramycin aptamer to the tobramycin on the matrix, and that levels of non-specific interactions are below background (see Note 6 concerning the importance of pH in the binding). To test this, the (+)- and (−)-aptamer substrates prepared in Subheading 3.2. are required. Specificity is only given when the (+)-aptamer RNA binds and the (−)-aptamer RNA fails to bind.

Buffers 2–6 (see Subheading 2.2. ) should be freshly prepared. The procedure comprises the following: preblocking the tobramycin matrix (hereafter referred to as matrix) with tRNA and BSA (steps 1 and 2); binding of the RNA and subsequent washing of the matrix to remove nonspecifically bound RNA (step 3); and finally elution of the bound RNA (step 4). RNA is traced by counting the radioactivity in the various fractions and the experiment is performed in duplicate for each of the two RNAs. All steps are performed at 4°C.

  1. 1.

    Take an aliquot of the matrix corresponding to 15 μL of packed volume of beads for each assay and collect the beads by centrifugation.

  2. 2.

    Block the matrix by resuspension in 250 μL of blocking buffer B with head-overtail rotation overnight.

  3. 3.

    For one matrix aliquot, prepare 400 μL of binding buffer containing 60–80 pmol of RNA (see Note 9 ). Save 5 μL (sample A0) and then add the mixture to the collected matrix. Incubate by head-over-tail rotation for 1–1.5 h, collect the matrix, save 5 μL of the supernatant (sample A1), and then wash the matrix three times with 1 mL of the W145 buffer. Save the supernatants of the washes in one tube (sample W145). Determine the amount of RNA bound to the matrix.

  4. 4.

    To elute the bound RNA, add 50 μL of E145T elution buffer to each matrix aliquot and incubate by head-over-tail rotation for 10 min, then collect the matrix by centrifugation. The supernatant (sample E) contains the eluted RNA. Determine the amount of RNA left on the matrix.


Count all fractions saved (samples A0, A1, W145, and E) and determine the proportion of RNA in each. Specificity with the above RNAs (i.e., (+)-aptamer vs (−)-aptamer pre-mRNAs) has been achieved when 60–70% of the tagged RNA, but less than 2% of the untagged RNA, is bound to the matrix ( Fig. 1 C). Furthermore, more than 80% of the immobilized RNA must be released upon elution in order for the solid-phase assembly scheme to work.

3.4 Solid-Phase Splicing

Buffers 2–7 (see Subheading 2.2. ) must be freshly prepared. The procedure for isolation of spliceosomes by splicing on immobilized pre-mRNA is subdivided into three steps: production of the matrix-bound pre-mRNA (step 1); initiating and performing the splicing reaction (step 2); and washing the matrix and elution of the matrix assembled spliceosomes (step 3). During the whole procedure, small aliquots are saved to monitor the RNA content of each fraction.

Because the (−)-aptamer pre-mRNA does not bind to the matrix under the conditions outlined in Subheading 3.3.2 , a mock incubation with this pre-mRNA has no value. However, to control for nonspecific binding of nuclear extract components to the matrix during the procedure, it is essential to perform a mock assembly by omitting the pre-mRNA in step 1 ( Fig. 3 ). The steps of the procedure are as follows:
Fig. 3.

RNA and protein content of splicing complexes eluted with tobramycin. Solid-phase splicing was performed for 45 min at 30°C in the presence (lanes 1 and 3) or absence (lanes 2 and 4) of pre-mRNA, and the complexes were subsequently eluted with tobramycin. RNA and protein were recovered, analyzed by PAGE, and visualized by silver staining. (Reprinted with permission from ref. 5, copyright 2002, National Academy of Sciences.)

  1. 1.

    Prepare the matrix-bound pre-mRNA exactly as detailed in Subheading 3.3.2 , steps 1–3. For a standard experiment, 4X 15 μL matrix-bound pre-mRNA is prepared (see Note 9 ). The same amount of matrix material is required for a mock assembly. Each aliquot is sufficient for one 1.5-mL splicing reaction (see Note 10 ).

  2. 2.

    On ice, prepare a 6-mL splicing mix, consisting of 35% HeLa cell nuclear extract (see Note 9 ) supplemented with 25 mM KCl, 3 mM MgCl2, 2 mM ATP, and 20 mM creatine phosphate. Gently mix and distribute 1.5 mL onto each matrix aliquot. Fix the tubes securely to the rotating wheel of a hybridization oven set at 30°C. Incubate for the required time with rotation (approx 150 rotations/min); a 45-min incubation is sufficient to allow the assembly of large amounts of spliceosomal A complex (see Note 11 ).

  3. 3.

    At the end of the incubation the reaction mixtures are placed on ice. All further operations are performed at 4°C. The matrix is collected and the reaction supernatant is saved (fraction SN). The matrix is now washed three times with 750–800 μL aliquots of W75 buffer. For ease of elution the matrix aliquots are combined in pairs during the first and second washes, by adding the W75 buffer to only one tube and transferring the matrix material together with the W75 buffer to its twin tube. The washes are saved separately (W75, fractions 1–3). Assembled spliceosomes are released by adding 250 μL of E145T elution buffer to the single matrix aliquot and further incubation with head-over-tail rotation for 10 min at 4°C. The matrix is collected and the supernatant containing the assembled spliceosomes (fraction E) is used for further analysis (see Subheading 3.5. ) and fractionation (see Subheading 3.6. ).


Aliquots are withdrawn from all fractions and their radioactivity is measured. The amount of RNA in each fraction is determined. It is convenient to prepare a balance sheet listing the volumes next to amounts of RNA in each fraction.

A comparison of the eluates from two parallel assembly reactions, in the presence of the same matrix but with and without pre-mRNA, is shown in Fig. 3 . This demonstrates clearly the high level of purity obtained in the eluted spliceosomes. Figure 4 shows the RNA (A and B) and protein (C) analysis of the fractions collected during the actual purification procedure: the reaction supernatant (SN), the 75-mM KCl washes (W75), and the eluate (E).
Fig. 4.

Purification of spliceosomes assembled on the matrix. The gels show the RNA (A, silver stain), pre-mRNA (B, autoradiography of A), and protein (C) analysis of the different fractions in the course of purification of spliceosomes. I, input pre-mRNA (A0; see Subheading 3.3.2 ). SN, reaction supernatant; W75, 75 mM washes 1, 2, and 3; E, eluate. In (C), the W75 first wash was not analyzed, and the lane marked X is spillover from W75 fraction 3. B, Bio-Rad broad-range molecular-weight marker (250, 150, 100, 75, 50, 37, 25, and 15 kDa); N, Novex Mark12 wide-range protein standard (200, 116, 97, 66, 55, 36, 31, 22, and 14 kDa).

3.5 Protein and RNA Analysis

  1. 1.

    Aliquots of equal volume are taken from the reaction supernatant (SN), the three washes (W75 1–3), and the eluate (E). The volume to be taken is chosen in such a way that approx 1 pmol RNA is withdrawn from the eluate. After one phenol/chloroform extraction, the RNA is precipitated from the aqueous phase by adding ethanol, and the protein from the organic phase by adding acetone.

  2. 2.

    The RNA is analyzed by denaturing PAGE on a 15% polyacrylamide/8.3 M urea gel and the protein by SDS-PAGE on a 10%/13% step gel. Both RNA and protein gels are stained with silver (20) and dried. The RNA gel is also autoradiographed.


3.6 Glycerol-Gradient Fractionation of Assembled Spliceosomes

The assembled spliceosomes present in the eluate are further fractionated by glycerol-gradient centrifugation to separate the different spliceosomal complexes (see Note 3 ). One gradient is used to analyze the eluate from one assembly reaction (6 mL). Eluate (220 μL) is loaded onto each gradient, and the remainder is used for protein/RNA analysis (see Subheading 3.5. ). We use the Sorvall TH660 rotor (equivalent to the Beckman SW60 rotor), with a nominal capacity of 4.2 mL per tube. All operations are performed at 4°C.

  1. 1.

    First add DTT, to a final concentration of 0.5 mM, to aliquots of the 10% and 30% glycerol solutions, and then use these to prepare a gradient in a BioComp gradient mixer (SW60, 10–30% v/v, glycerol program). Let the gradient equilibrate at 4°C for at least 1 h.

  2. 2.

    Remove the rubber cap from the gradient tube, remove 220 μL from the top of the gradient and then carefully layer 220 μL of the eluate on top of the gradient. Centrifuge for 1 h 45 min at 448,579g.

  3. 3.

    Manually fractionate the gradient from top to bottom in 175-μL fractions.

  4. 4.

    Use 100 μL of each fraction for first determining the RNA content by measuring the cpm and then, with the same aliquot, for the protein/RNA analysis described in Subheading 3.5. The remaining 75 μL of each fraction can be used for psoralen crosslinking analysis (5).

  5. 5.

    For the gradient profile, plot the quantity of RNA against the fraction number.


A typical gradient profile is show in Fig. 5 A, and the RNA distribution across the gradient is shown in Fig. 5 B. Fractions 9–13 contain exclusively pre-mRNA, U1, and U2 snRNA. Through our previous psoralen crosslinking analysis, fractions 11–13 could be assigned unequivocally to the A complex (5). These fractions contain only pre-mRNA, U1, and U2 snRNA in approximately equal stoichiometric amounts. The spliceosomal B complex migrates in fractions 16–18. It is readily apparent from Fig. 5 B that these fractions contain equal stoichiometric amounts of pre-mRNA, U1, U2, U4, U5, and U6 snRNAs, a hallmark of the intact spliceosomal B complex.

4 Notes

  1. 1.

    Depending on the experimental context, the aptamer may have to be inserted at different positions in the RNA under study. Although we had variable success when it was inserted into positions within the intron of the pre-mRNA (H.-P. Vornlocher, unpublished), a 5′ position was found to be optimal for the isolation of the mRNP from a glycerol-gradient-fractionated solution-splicing reaction similar to the one shown in Fig. 1 D (C. Merz, personal communication).

  2. 2.

    The inability to affinity-select spliceosomes from a solution reaction may be owing to masking of the aptamer tag and/or the tobramycin matrix by factors present in the nuclear extract. We observed that the affinity selection is somewhat improved (15% recovery) after fractionation of the splicing reaction mixture either by glycerol-gradient centrifugation or by Superdex 200 (Amersham Biosciences) column chromatography (C. Merz, personal communication).

  3. 3.

    In principle, any procedure based on separation by size could be used. However, gel filtration on Sephacryl S500 (21) was impracticable owing to the dilution effect of the sample (not shown) and because the A, B, and C complexes migrate as a single peak. Furthermore, the spliceosomal A complex was found to be unstable when subjected to size-exclusion chromatography on Sephacryl S400 (7).

  4. 4.

    Psoralen crosslinking is used to detect RNA-RNA interactions diagnostic for a particular spliceosomal complex. A description of the method is outside the scope of the present chapter on the isolation of spliceosomes, and the reader is referred to ref. 5 for details.

  5. 5.

    For a standard splicing reaction in vitro, we use protocol 7 of ref. 22. As this protocol is exhaustive, it is not reproduced here in detail. Briefly, pre-mRNA synthesis is essentially as described in Subheading 3.2. , except that (1) the concentration of cold UTP is lowered to 100 μM; (2) [α-32P]UTP is added to 0.825 μM in a 20-μL transcription reaction, and (3) the transcript is gel-purified on a 9.6% polyacrylamide/8.3 M urea gel. A 100-μL in vitro splicing reaction is set up essentially as described in Subheading 3.4. , step 2, except that pre-mRNA is added to a concentration of 50–60 fM and incubation takes place at 30°C without agitation. For each time point in Fig. 1 D a 15-μL aliquot is withdrawn after the specified time and placed on ice. After treatment with proteinase K (19), RNA is recovered by phenol/chloroform extraction and analyzed as in Subheading 3.5. , step 2.

  6. 6.

    The most critical parameter for obtaining specificity in binding is the correct pH. The aminoglycosidic antibiotic tobramycin itself is positively charged at neutral pH (2326). The pK a values of the primary amino groups are (positions in brackets; refs. 23 and 24): 7.6 (6′), 8.6 (2′), 6.2 (3), 7.4 (1), 7.4 (3″). The molecule is only close to neutrality at a pH of around 9. Hence a basic pH is required to suppress non-specific binding of the RNA to matrix. The Tris buffer used (pH 8.1 at 21.5°C) will attain the correct pH upon equilibration at 4°C in the blocking, binding, washing and elution buffers (buffers 2–7, Subheading 2.2. ). To facilitate handling, a 4X BP without DTT may be prepared and kept at 4°C. DTT (2 mM) is then added just before use.

  7. 7.

    In the context of the RNA under study, it is essential that the tobramycin aptamer folds correctly. Partial complementarities with other regions of the RNA are best excluded by using the RNA folding program “mfold” (at http://​www.​bioinfo.​rpi.​edu; refs. 27 and 28). All calculated substructures should have the aptamer correctly folded.

  8. 8.

    A number of different tobramycin concentrations, ranging from 0.05 to 10 mM, were tested in the matrix coupling reaction. Concentrations below 5 mM dramatically reduced the binding capacity of the aptamer-RNA to the matrix, probably because of a reduced coupling efficiency of tobramycin to the matrix. Although higher concentrations somewhat improved the binding efficiency of the aptamer-RNA, they were not used, because we reasoned that too many binding sites on the matrix could lead to a molecular crowding effect and thus compromise the assembly of macromolecular structures such as spliceosomes.

  9. 9.

    The three critical parameters for the solid-phase assembly of spliceosomes are (1) the RNA concentration on the matrix, (2) the HeLa cell nuclear extract concentration, and (3) the time of incubation at 30°C (see Note 11 ). The values given for each parameter are derived mainly from optimization experiments. In the solid-phase assembly procedure, a HeLa cell nuclear extract should be used that processes at least 80% of the pre-RNA when assayed in a standard solution splicing reaction (see Fig. 1 D and Note 5 ) at an optimized concentration. Using this condition (35% nuclear extract in most cases), it was found that the optimal pre-mRNA concentration is 22–25 nM, that is, 33–38 pmol of pre-mRNA bound to a 15-μL matrix aliquot. Not all pre-mRNA added is eventually bound to the matrix and that minor variations in binding efficiency is observed in different tobramycin matrix preparation. Hence, the amount of pre-mRNA to use in the initial binding (60–80 pmol) has to be calibrated experimentally to the desired value of 22–25 nM in the final reaction. A lower amount of RNA (10 or 20 pmol/15-μL bead aliquot) affects the overall yield of spliceosomes, whereas a higher amount (50–75 pmol/15-μL bead aliquot) affects the yield of spliceosomal B complex.

  10. 10.

    The amount of splicing mix added to one 15-μL bead aliquot can be decreased. However, when using less than 1 mL of splicing mix/15-μL bead aliquot the overall yield of spliceosomes was found to be reduced.

  11. 11.

    It is essential that the splicing reaction per se has not yet started when the aim is the preparation of the spliceosomal A and/or B complexes. In the glycerol gradient the C complex would migrate close to the B complex and the mRNP close to the A complex (K. Hartmuth, unpublished data).


Copyright information

© Humana Press Inc., Totowa, NJ 2004