1

Title: The assembly of spliceosomes in vitro

Alternative: assembly and isolation of spliceosomal complexes in vitro (most of it is really about the isolation)

Authors: Klaus Hartmuth, Maria A. van Santen, Peter Odenwälder, Reinhard Lührmann

Affiliation: Max Planck Institute for Biophysical Chemistry, Am Fassberg 11, D-37077 Göttingen, Germany

1. Abstract

Spliceosome dynamics comprise a complex, but regulated interplay of many different complexes and factors that interact sequentially with the pre-mRNA. The purification of spliceosomal complexes at distinct stages has been extremely helpful for the compositional, structural and functional analysis of the spliceosome. Currently, the best method for purification is affinity selection using a molecular tag on the pre-mRNA substrate. We make use of a pre-mRNA tagged with three MS2 RNA aptamers. This RNA is incubated with the MS2-MBP fusion protein, which interacts (i) with the pre-mRNA by binding strongly to the MS2 hairpins; and (ii) with an amylose affinity matrix through the MBP (maltose-binding protein) portion of the protein. The latter interaction is fully reversible, under mild conditions, by competition with maltose. Our affinity purification procedure consists of the assembly of a particular spliceosomal complex on the tagged pre-mRNA, followed by size fractionation and affinity selection. The procedure is exemplified here for the purification of a spliceosomal C complex.

Keywords

Spliceosome, C complex, affinity selection, MS2-MBP

2. Theoretical background

Spliceosomal assembly in HeLa cell nuclear extracts is a sequential process entailing the stepwise addition of pre-formed complexes to the pre-mRNA. This process was initially observed by analysing splicing-reaction kinetics by gradient centrifugation (Frendewey and Keller, 1985; Grabowski et al., 1985) or native gel electrophoresis (Konarska and Sharp, 1986). The sequence of events has been worked out in great detail (Staley and Guthrie, 1998; Wahl et al., 2009) and is outlined in Figure 1 Figure 1 is already shown by the introduction given by Reinhard, I would refer then to this figure. (see Figure xxx, chapter 5, Luhrmann) Briefly, the U1 snRNP interacts with the 5' splice site (ss) in an ATP-independent manner and recruits the U2 snRNP to form the earliest discernible complex, the E complex (Das and Reed, 1999). Upon ATP hydrolysis, the branch-point interaction site of U2 snRNA interacts with the branch point of the pre-mRNA, forming the A complex. This complex, also termed the pre-spliceosome, serves as the landing platform for the tri-snRNP (consisting of the U4, and U5 and U6 snRNPs), thereby generating the B complex, which contains all the five snRNAs. In an activation step, the catalytically active spliceosomal RNA network is formed, in which specific interactions take place between the U2 and U6 snRNAs, the pre-mRNA and also the U5 snRNA. In this complex, termed B*, in which U1 and U4 snRNPs are no longer present, the first step of splicing takes place, leading to the C complex. This complex then performs the second step of the splicing reaction, resulting in the formation of the mRNP and the post-spliceosomal complex.

The spliceosome is a protein-rich machine, containing many snRNP-specific proteins as well as numerous non-snRNP proteins. While the functions of many of these non-snRNP proteins are known (Guthrie, 2009; Wahl et al., 2009), details of their mechanism of action and the manner in which this is synchronised with the other events on the spliceosome remain largely unknown. To investigate this, it is essential to have reliable procedures to isolate key spliceosomal intermediate complexes. These can then be investigated in terms of composition, structure, and enzymatic activities.

By far the most powerful procedure for the isolation of defined spliceosomal complexes is affinity selection, and a number of methodological approachesmethods? have been developed to this end (see (Jurica and Moore, 2002) for review). Initially, the antibody affinity of a spliceosomal protein, present in only one particular state of the spliceosome was exploited to immunopurify activated spliceosomes (Makarov et al., 2002). The antibody was directed against the SKIP protein, can you explain what the SKIP protein is? and immunoselected spliceosomes could be released specifically by competition with the original immunogenic peptide. In a different approach, an RNA affinity tag, the tobramycin aptamer, was used to assemble an A complex on a pre-mRNA that itself was immobilised on a tobramycin matrix (Hartmuth et al., 2002). The A complex, thus assembled, could be eluted specifically by competition with tobramycin and used for further studies. This method was also coupled with immunoaffinity purification to isolate a highly purified A complex for compositional and structural studies (Behzadnia et al., 2007). Although powerful, immunopurification methods depend on the availability of large amounts of peptide antibodies against a particular spliceosomal protein. This is not always guaranteed. The alternative, therefore, would be a robust RNA affinity selection that uses only readily available materials and that works under mild conditions in solution.

At present, the method of choice for the isolation of homogenous spliceosomes employs the MS2-MBP affinity purification (Zhou and Reed, 2003). This approach has been used extensively in the characterisation of spliceosomal complexes (Das et al., 2000; Jurica and Moore, 2002; Zhou et al., 2002a) and is based on the engineered high-affinity binding of a bacteriophage MS2 coat-protein fragment to a minimal MS2 RNA hairpin (LeCuyer et al., 1995). Three copies of the MS2 RNA hairpin are inserted into the pre-mRNA (Zhou et al., 2002b). The MS2 coat-protein fragment, expressed as a fusion protein with the maltose-binding protein (MBP; (Jurica and Moore, 2002)) is then bound to the hairpins. Affinity selection is through binding of the MBP portion of the fusion protein to an amylose resin. Pre-mRNA and any complexes assembled on it can thus be captured on the resin, washed and subsequently released under native conditions by simply adding maltose.

In our laboratory, this method is used to purify the mRNP (Merz et al., 2007), and the spliceosomal B (Deckert et al., 2006) and C (Bessonov et al., 2008) complexes assembled in HeLa and yeast nuclear extracts (Fabrizio et al., 2009; Warkocki et al., 2009). In principle, the preparation of particular complexes has to be optimised for specific pre-mRNAs, but the procedure is general enough to be easily adaptable to a variety of different situations. Preparation of nuclear extracts is outlined in chapter 25, Harthmuth and the analysis their fractionation by gradient centrifugation in Chapter 12, Hartmuth.

3. Protocol

Klaus, in the protocols, can you insert pipetting scheme-type protocols (1ul, x, 5 ul Y, for 30 min at 30, etc.), it would also be helpful if you indicate how may cpm of a minx or other substrate of a certain length represents the molar concentration you need.

Can you also show diagrams of the substrates used?

3.1 The core splicing reaction

A standard analytical splicing reaction mixture contains: 2–5 nM 32P-labelled pre-mRNA (specific activity 50,000–500,000 cpm/pmole), 2 mM ATP, 20 mM creatine phosphate (CP), 3.6 mM MgCl2, and nuclear extract what protein concentration does the NE have 10 mg/ml? in a total volume of 15 µL. The percentage of nuclear extract depends on the quality of the nuclear extract obtained what are the criteria for the quality of the nuclear extract? and is usually between 30–50% (v/v). The assessment of nuclear extract quality is discussed in Chapter 25, Hartmuth, Figure xxx. Can you clarify what a good extract is? The total KCl concentration is adjusted to 60 mM by adding the required amount of KCl from a 0.1 M KCl solution. Further variations in KCl or MgCl2 concentrations can be assayed (Krainer et al., 1984), and the improvements can be important in scale-up experiments. For splicing, reactions are mixed on ice and then placed at 30°C for the required time and reactions are stopped by chilling them to 4°C. In highly active splicing extracts ( what is their definition x% splicing of a minx substrate after 2hrs?), but only in these, the spliceosomal A complex may form after prolonged periods (>2h) on ice. Klaus, can you give this info in a pipetting scheme type: 1 ul x, 1ul y etc, and add the time

For analytical purposes, RNA is extracted from the splicing reactions and analysed by denaturing PAGE as described in Chapter 26, mayeda (Chapter II.16). The efficiency of spliceosome formation and of spliceosome transformations is investigated by native electrophoresis as detailed in the next section.

The basic reaction scheme outlined above can be scaled up to any extent desired. An important variable that should be optimised is the amount of pre-mRNA added to the splicing reaction, taking into account the objective of the experiment. For a preparative assembly of a splicing complex, where the end product accumulates (as is the case with the preparation of the C complex described below), pre-mRNA concentrations as high as 30 nM can be optimal to maximise the yield (Bessonov et al., 2008). These conditions have to be determined separately for each particular extract.

3.2 Splicing complex analysis by agarose gel electrophoresis

The analysis of splicing complex assembly by native gel electrophoresis on agarose gels was first introduced by Das and Reed (Das and Reed, 1999). Samples for agarose gel electrophoresis are prepared by adding 5x Stop buffer (50% glycerol, 1.25 mg/mL heparin, 0.001% (w/v) of bromophenol blue and xylene cyanol) to the splicing reactions.

Can you also give this here in a pipetting type protocol to give a timecourse that shows E, A, B and C complexes, as well as a picture of the gel? Do you stop the reaction in any way prior to loading on the gel or just put it on ice?

An additional incubation at 30°C for a fixed time (1 min) is normally performed to standardise the exposure to the heparin. Heparin is essential for the analysis of spliceosomes (Konarska and Sharp, 1986), but it has to be omitted in the analysis of the E complex, because of the sensitivity of this complex towards heparin (Das and Reed, 1999). Samples are loaded onto a standard 1.5% low-melting agarose gel, and are run in 0.5 x TBE at room temperature, at 9 V/cm until the bromophenol blue has reached the bottom of the gel. The agarose gel is fixed for 30 min in 10% methanol, 10% acetic acid, then gently dried on Whatman 3MM paper (60°C for approximately 6 h), and visualised by phosphorimager screens or medical X-ray film.

3.3 Preparation of MS2-MBP-tagged spliceosomal complexes

3.3.1 Preparation of MS2-MBP- and MS2-tagged pre-mRNA

The preparation of the MS2-MBP fusion protein is performed essentially as described elsewhere (Jurica and Moore, 2002), except that the final heparin column is eluted with 100 mM KCl. Klaus can you give for this a brief pipetting type protocol? The three MS2 hairpins are inserted by standard PCR cloning into the desired position of the transcription template for the pre-mRNA. Can you give here the sequences of the MS2 hairpins or the primer sequence? This is done either by de novo cloning (Zhou et al., 2002b) or by transferring the three loops from a pre-existing plasmid. For B complex purification (Deckert et al., 2006), we inserted the tag 3' of the downstream exon of pMINX (Robberson et al., 1990). For the pM5 substrate for preparation of the C complex (Bessonov et al., 2008), the MS2 tag was inserted at the 5' end of exon 1 of the PYP pre-mRNA (Wollerton et al., 2004). What was the rationle of using different pre-mRNAs? The pre-mRNAs we use are m7GpppG-capped and 32P-labelled co-transcriptionally (see chapter, 26 mayeda) Following purification of the RNA (gel purification or spin column), the amount of MS2-MBP required to saturate completely the MS2-binding sites on the RNA of interest should be determined empirically (Zhou and Reed, 2003) at least once for a particular MS2-MBP preparation. To this end, increasing amounts of MS2-MBP protein are incubated with the tagged pre-mRNA – untagged pre-mRNA is used as a control – and the resulting complexes are analysed on a native gel (see above) do you have an example picture for this?. Usually, a 20-fold molar excess of MS2-MBP is sufficient to shift all the pre-mRNA on the gel, indicating that all the tagged pre-mRNA has bound the MS2-MBP protein.

3.3.2 Assembly of the C complex

The pM5 substrate allows for the accumulation of the C complex because of the missing 3' exon and an unusually long polypyrimidine tract (Bessonov et al., 2008; Wollerton et al., 2004). To prepare the C complex, the standard splicing reaction is scaled up to 12 mL, using 30 nM pre-mRNA. To facilitate tracking, the pre-mRNA is 32P-labelled co-transcriptionally (specific activity: 1,000–20,000 cpm/pmole). For this purpose, the pre-mRNA is first bound to the MS2-MBP protein by incubating the two components on ice for a minimum of 30 min. Then, a splicing reaction is assembled by adding a splicing mixture, consisting of nuclear extract and KCl as required, and MgCl2, CP, and ATP (see above). To ensure homogenous conditions, the reaction is best performed in an array of standard 1.5 mL tubes by incubation with mild agitation at 30°C for 3 h. Pre-mRNA not assembled into C complex is specifically destroyed by RNase H (vendor, final concentration?). To this end, a 30-fold excess of one or two DNA oligonucleotides complementary to nucleotides between –6 and –18 relative to the 5' splice site is added and the incubation is continued for another 20 min. We use two oligonucleotides simultaneously (M6: (GGCGGTCTCGTC and M12 (CTCGTCGGCAGC) complementary to pre-mRNA positions starting at –6 and –12 nt upstream of the 5' splice site). The incubation is stopped by placing the tubes on ice.

Can you give this again as a pipetting protocol, can you also give the sequence and structure of the pM5 construct, so the binding sites for the antisense oligos become clear

3.3.3 Purification of the C complex

Splicing reactions are first fractionated by centrifugation on a 10–30% glycerol gradient (see chapter 12, Hartmuthfor details on rotors and run parameters), and fractions containing the C complex are identified how are the identified? and pooled. The C complex is further purified from the pool by amylose affinity chromatography as follows. C complex fractions are loaded onto a chromatography column (1.2 mL bed volume; Bio-Spin Chromatography Columns, Bio-Rad catalogue #732-6008) containing 0.25 mL amylose beads (New England Biolabs) pre-equilibrated with standard G buffer. The sample is allowed to pass into the matrix by gravity flow. The matrix is then washed with 25 column volumes of standard G buffer (composition of G buffer, temperature?). C complexes are then eluted dropwise with G buffer containing 12 mM maltose. The concentration of spliceosomes in each eluted fraction is measured by determining the 32P activity usingCherenkov counting. Peak fractions are identified and pooled and constitute the Ccomplex material that can be used for further downstream analysis.

3.3.4 Assembly and purification of the B complex

B complexes can be assembled on the pM5 substrate (see 3.3.2), and also on a regular??what is regular? biexonic substrate, such as the MINX pre-mRNA(Robberson et al., 1990). In both cases, the pre-mRNA concentration in the splicing reaction should be decreased to 10 nM. The time of incubation is dependent on the kinetic activity of the nuclear extract. As a rule, reaction mixtures are incubated for 8 minutes, because during this time mainly complexes A and B, but no activated complexes, have formed. Immediately after this incubation, reaction mixtures are placed on ice and fractionated on a 10–30% (v/v) glycerol gradient (chapter 12, Hartmuth). This will result in A and B complexes in separate gradient fractions, which can be used for affinity purification or further downstream analysis.

3.3.5 Characterisation of the purified complexes

To check the purity of the spliceosomal complexes, RNA is extracted (how, proteinase K, phenol, choroform?) from the affinity-purified spliceosomes and analysed by denaturing PAGE (?%gel, 8mUrea, 1xTBE). Similarly, the protein composition can be analysed by SDS-PAGE. For structural investigation, spliceosomal complexes are investigated by electron microscopy, as has been described for the B complex (Deckert et al., 2006). For functional analysis, spliceosomes can be tested for their activity either in a chase experiment using appropriately depleted nuclear extract (as described for the B complex), or in a bimolecular ligation assay (as described for the C complex, (Bessonov et al., 2008)). If the purity of the complexes needs to be improved further, for example for structural and functional studies, then the affinity purified spliceosomes can be subjected to a second round of gradient centrifugation and affinity selection.

4. Example of an experiment

A 12 mL C complex preparation was initiated as described above and, after the RNase H digestion, the sample was fractionated on a TST 41.14 rotor (vedor?)(2 mL reaction/gradient) and centrifuged for 12h at 25,000 rpm at 4°C. The gradient was fractionated manually into 23 fractions of 0.5 mL each and the gradient profile was determined by Cherenkov counting (Figure 2A). Fractions 17–21, corresponding to the C complex, were pooled and amylose affinity selection was performed as described above. An aliquot of the purified material was rerun on a second gradient (Figure 2B); the single peak observed attests to the high purity of the C complex obtained. The RNA (Figure 3) composition of the gradient- and affinity-purified C complex shows only the U2, U5, and U6 snRNAs; the first step of the splicing has already occurred, as evidenced by the presence of a free exon 1 together with the lariat structure, the hallmarks of the splicing reaction after the first step of splicing. The purity of the C complex is ascertained by the absence of U1 or U4 snRNAs, and the almost complete absence of pre-mRNA. The proteins detected in the gradient- and affinity-purified material (Figure 3) are therefore genuine Ccomplex proteins.

5. Trouble-shooting

  1. If the amount of MS2-MBP protein required to shift all the pre-mRNA on the gel is substantially more than a 20-fold molar excess, contamination of the MS2-MBP protein by RNA should be considered. The solution is either to repeat the heparin-column chromatography or to make a new protein preparation.
  2. On the average, 8–10% of the input pre-mRNA is recovered in the Ccomplex region of the gradient. Should this value be much smaller, re-determination of the required pre-mRNA concentration may prove helpful. Also, the use of a different extract should be considered. We find that such optimisations can best be performed by investigation of Ccomplex formation on native gels.
  3. On average, 66% of the material loaded onto the amylose column is recovered in the maltose eluate. Should this value drop, the most likely cause is an old amylose resin. Testing a new batch of amylose resin is therefore recommended. Next, the binding of the MS2-MBP protein to the pre-mRNA should be tested with a bandshift assay. If this does not help, a fresh sample of MS2-MBP protein should be tested.
  4. The B and C complexes cannot be separated on a glycerol gradient; therefore, they have to be prepared in separate experiments. Note that B complexes are deliberately destroyed during the preparation of the C complex, during the digestion by RNase H.
  5. When preparing the B complex, the A complex peak should be treated with caution because it can potentially be contaminated with an H complex. Additional methods must be used to estimate the degree of contamination (e.g. proteomics), or other methods should be used or adapted for preparation of the A complex (Behzadnia et al., 2007). Note that the A complex must be purified at 75 mM salt, as it loses components at higher salt concentrations.

Figure legends