RNA polymerase II synthesizes a precursor RNA that extends beyond the 3′ end of the mature mRNA. The end of the mature RNA is generated cotranscriptionally, at a site dictated by RNA sequences, via the endonuclease activity of the cleavage complex. Here, we detail the method to study cleavage reactions in vitro.
The 3’ end of mammalian mRNAs is not formed by abrupt termination of transcription by RNA polymerase II (RNPII). Instead, RNPII synthesizes precursor mRNA beyond the end of mature RNAs, and an active process of endonuclease activity is required at a specific site. Cleavage of the precursor RNA normally occurs 10-30 nt downstream from the consensus polyA site (AAUAAA) after the CA dinucleotides. Proteins from the cleavage complex, a multifactorial protein complex of approximately 800 kDa, accomplish this specific nuclease activity. Specific RNA sequences upstream and downstream of the polyA site control the recruitment of the cleavage complex. Immediately after cleavage, pre-mRNAs are polyadenylated by the polyA polymerase (PAP) to produce mature stable RNA messages.
Processing of the 3’ end of an RNA transcript may be studied using cellular nuclear extracts with specific radiolabeled RNA substrates. In sum, a long 32P-labeled uncleaved precursor RNA is incubated with nuclear extracts in vitro, and cleavage is assessed by gel electrophoresis and autoradiography. When proper cleavage occurs, a shorter 5’ cleaved product is detected and quantified. Here, we describe the cleavage assay in detail using, as an example, the 3’ end processing of HIV-1 mRNAs.
The biosynthesis of most mature eukaryotic message RNAs (mRNAs) requires several post-transcriptional modifications such as capping, splicing and polyadenylation. These modifications are generally coupled to ensure correct processing1 and strongly increase the stability of the mRNA.
The 3’ end formation of mammalian pre-mRNAs is generated by endonucleolytic cleavage of the nascent RNA followed by addition of adenylate residues to the 5’ cleaved product by poly(A) polymerase (PAP)2-4. In mammals, cleavage is accomplished by a multicomponent protein complex, of approximately 800 kDa, which assembles on specific pre-RNA sequences. The poly(A) signal sequence, a highly conserved canonical hexanucleotide sequence AAUAAA, directs the cleavage site at approximately 10-30 nt downstream. This site is specifically recognized by the cleavage and polyadenylation specificity factor (CPSF) and the 73 kD subunit of CSPF contains the endonuclease activity. The cleavage stimulation factor (CstF) binds a more degenerate GU- or U- rich element sequence downstream of the poly(A) site. Also required for cleavage is the mammalian cleavage factor I (CFIm), and the mammalian cleavage factor II (CFII). CFIm binds the specific UGUA(N) sites in upstream sequence elements (USEs) that have been defined for a number of genes and seem to be involved in important physiological processes5-8.
In vitro, RNA processing reactions are commonly analyzed by the use of radiolabeled RNA substrates9-12. These may be synthesized by run-off transcription from the bacteriophage promoter T7 or SP6. When studying a polyadenylation site that has not been characterized before, it is necessary to use genomic DNA rather than cDNA to generate the RNA substrate, as important downstream sequences might not be present in cDNA. Design substrates to include at least 150 nt upstream and 50 nt downstream from the cleavage site/end on the mature mRNA. The cleavage product migrates faster than the substrate; however, because other fragments may be generated by non-specific nuclease action, the specificity of the reaction has to be verified by its dependence on the correct processing signal sequences. Therefore, RNA substrates with a point mutation in the AAUAAA sequence (e.g. AAGAAA) serve as a negative control for the cleavage reaction.
Given that a small amount of radiolabeled RNA is used for the cleavage reactions, RNases present in high abundance in most nuclear extracts can be problematic, and limit the choice of the starting material for extract preparation. HeLa cells tend to contain low levels of endogenous RNases, and thus perform well in these assays.
The endonucleolytic cleavage of the RNA substrates in vivo and in vitro is immediately followed by the poly(A) addition, thus the cleaved intermediate is not present in detectable quantities. Therefore, to study either a specific RNA sequence or proteins involved in a cleavage reaction, experiments are done in conditions that prevent polyadenylation from occurring. There is no dependence of cleavage on polyadenylation, or vice versa, so one can stop polyadenylation without harming the cleavage reaction. Thus, ATP is replaced with a chain terminating analogue that lacks the 3’ hydroxyl group so that only a single nucleotide can be incorporated at the poly(A) site and just cleaved RNA can be detected.
Given the complexity and high degree of particularity of this type of assay, we describe a detailed video protocol to study endonucleolytic cleavage by the cleavage/Poly(A) machinery of mRNA precursors in vitro. We describe how to prepare competent nuclear extracts, generate radiolabeled RNA substrates, perform the cleavage reaction, and analyze and interpret the resulting products. Figure 1 shows an example of substrate RNAs encoding for the 3’ end of HIV-1 pre-mRNAs to be used for a cleavage assay. The 3‘ end of the HIV RNA genome is composed of many important regulatory sequences such as the poly(A) site, a G+U rich region, and the USE element, which are all necessary for efficient maturation of the viral mRNA transcripts13. In this example we would expect the input RNA substrate to be 338 nt and upon cleavage 237 nt. If polyadenylation was allowed to occur, a smear of products would be observed between 237 and 437 nt.
1. Adaptation of Adherent Cells into Suspension
(This is an optional step. Suspension cells generally make better nuclear extracts, however cells grown in plates may also be used.)
2. Nuclear Extracts
3. RNA Probe Synthesis
4. 3’ mRNA Cleavage Reaction
Representative results of a cleavage assay of the RNA poly(A) sites of HIV-1 (Figure 2). We can observe the uncleaved RNA substrate, which is the slowest migrating band at the top of the gel. The specific cleaved product is the most intense shorter fragment band that runs faster in the gel at the expected size, and is specifically absent from the cleavage assay of the mutpoly(A) control that contains a point mutation in the polyA sequence (mutPolyA) RNA substrate. Degradation products of the input substrate may sometimes be observed. Quantification of the cleavage activity may be determined by densitometric plot of the ratio of uncleaved to cleaved RNA normalized to 100% of uncleaved RNA.
Figure 1. Schematic depiction of the HIV-pre-mRNA 3’ end processing substrates, and expected products of in vitro cleavage and polyadenylation reactions. The cleavage site, the dinucleotide –CA-, lies 25 nt downstream from the polyA site AAUAAA. LTR regions: U3 (unique 3’ sequence), R (repeated sequence), and U5 (unique 5’ sequence).
Figure 2. (A) Poly(A) site cleavage. 32P-labeled RNAs containing the poly(A) of HIV-1 and a point mutation of the poly(A) signal that abolishes cleavage (mutPolyA) were incubated in nuclear extracts of HeLa-S3 cells or BSA as a negative control. Bolded arrow indicates 5’ cleaved product. The 3’ fragment often runs off the gel or is degraded and not always visible. (B) Densitometric plot of the cleavage activity expressed as a ratio of uncleaved to cleaved RNA normalized to 100% of uncleaved RNA.
Buffer A (100 ml) | 10 mM Tris (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol (DTT). | Add DTT just prior to using buffer A during the extraction. Add 1 μl of 1 M DTT per 1 ml of buffer A. |
Buffer C (50 ml) | 20 mM Tris (pH 7.9), 25% (v/v) glycerol, 0.42 M NaCl, 1.5 mM MgCl2, 0.2 mM ethylenediaminetetraacetic acid (EDTA), 0.5 mM DTT, 0.5 mM phenylmethylsulfonyl fluoride (PMSF). | Add a protease inhibitor cocktail tablet the morning of the extractions. |
Buffer D50 (2-4 L) | 20 mM Tris (pH7.9), 20% (v/v) glycerol, 0.2 mM EDTA, 0.5 mM DTT, 50 mM ammonium sulfate. | HEPES-KOH may be substituted for Tris in buffer A, C and D. |
RNA Elution Buffer | 0.5 M amonium acetate, 10 mM MgCl2, 1 mM EDTA, 0.2% sodium dodecyl sulfate (SDS) | Store at room temperature. |
Stop Buffer (ETS) | 10 mM Tris (pH8.0), 10 mM EDTA, 0.5% SDS | Store at room temperature. |
Cleavage Loading Buffer | 90% formamide, 5 mM EDTA pH8, 0.025% Bromophenol Blue | Store at -20 °C. |
Table 1. Composition of buffers.
The in vitro pre-mRNA 3’ cleavage reaction, carried out in HeLa cell nuclear extracts or with cleavage factors fractionated from these extracts, has enabled identification of the core cleavage factors and their main complexes16-21. Many more proteins associated with these factors have recently been identified22, and the in vitro reaction may continue to shed light on how these proteins contribute to the reaction. Perhaps because the reaction in vivo appears to be cotranscriptional23, or because some contributing factors may be lost during extraction and dialysis, the efficiency is often poor, and rarely is more than 30% cleavage is achieved. In addition, the reaction efficiency may suddenly drop from day to day for no apparent reason. Therefore, it is important to appreciate which steps are the most critical, especially when attempting to adapt the reaction to modified or virally-infected HeLa cells, to other cell types or to new substrates.
Without doubt, the quality of the extract and the freshness of the in vitro transcribed RNA are paramount, as is the need to work under scrupulously RNase-free conditions. The health of the cells at the time they are harvested for extraction is also important. The cells should appear healthy, they should be doubling in less than 24 hr, they should be in or approaching mid log phase, and there should be few dead cells or other unexplained debris. Obviously, the cells should not be contaminated by mycoplasma or other bacteria. The substrate should not be made with too high a specific activity, as this can lead to apparent degradation.
When a large enough sequence on either side of the cleavage site is considered, and alternative polyadenylation is accounted for, the number of different poly(A) sequence signals in the human genome undoubtedly exceeds the number of genes24. As there is variability in the “strength” of different signals that have been tested, when adapting the procedure to a new cell type, it is best to begin with one of the stronger polyadenylation signals, such as the often-used Simian virus 40 (SV40) late poly(A) signal25. The combination of a weak poly(A) signal sequence with cells other than standard HeLa cells can prove frustrating, and it is better first to use a strong substrate with new cells, or unmodified HeLa cells with a potentially weak polyA substrate.
Even after so many years of successful use, there are still minor mysteries associated with the in vitro 3’ cleavage reaction. For example, why is creatine phosphate needed? It has been demonstrated that the original reason for including it – to regenerate the ATP pool – is not valid26. Interestingly, it can be omitted with only partial loss of activity when nuclear extracts are used, but becomes progressively more necessary as the extract is fractionated into partially purified cleavage factors that are used to reconstitute the reaction. In fact, phosphocholine, another small molecule with a phosphate group and a positively charged amine, but likely having no in vivo role in the reaction, has been found to be more effective than creatine phosphate, at least in the reconstituted reaction27. PVA is another ingredient whose requirement is not fully explained. It is assumed to be a crowding agent, leading to an effective increase in cleavage factor concentration, but other crowding agents, like polyethylene glycol, do not work nearly as well. Understanding these factors might lead to improved efficiencies, enabling the extension of the method to less efficient cell types and RNA substrates, and might yield clues to how the various factors work.
The authors have nothing to disclose.
S.V. is grateful for the funding support from the NIH K22AI077353 and the Landenberger Foundation. K.R. gratefully acknowledges funding from the NIH (5SC1GM083754).
1L Celstir Flask | Wheaton | 356884 | Different sizes available |
4 Position Slow Speed Stirrer | VWR | 12621-076 | |
Swinging Bucket Centrifuge | Beckman Coulter | Allegra x-15R with SX4500 Rotor | |
Ultra Centrifuge | Beckman Coulter | Optima L-100 XP Ultra with SW41 Rotor | |
Table Top Centrifuge 5417R | Eppendorf | Refridgerated | |
Thermomixer incubator | Eppendorf | ||
250ml Conicle Tubes | Corning | 430776 | |
50ml Conicle Tubes | BD Falcon | 352098 | |
Ultra-clear Centrifuge Tubes | Beckman Coulter | 344059 | |
JOKLIK Modified MEM | Lonza | 04-719Q | |
Fetal Bovine Serum | Atlas | F-0500-A | Heat inactivated |
L-Glut:Pen:Strep | Gemini Bio-Products | 400-110 | |
1M Tris-HCl pH 8.0 | Mediatech | 46-031-CM | |
Magnesium Chloride | Fisher | BP214-500 | |
Potassium Chloride | MP Biomedicals | 194844 | |
HEPES | Fisher | BP310-500 | |
DTT | Alexis Biomedicals | 280-001-G-010 | |
Glycerol | Fisher | BP229-4 | |
5M Sodium Chloride Solution | Mediatech | 46-032-CV | |
EDTA 0.5M Solution | Sigma-Aldrich | E7889-100ml | |
EDTA | Fisher | BP120-500 | |
PMSF | Thermo Scientific | 36978 | |
Ammonium Sulfate | Fisher | A702-500 | |
cOmplete EDTA-Free Protease inhibitor cocktail tablets | Roche | 04 693 132 001 | |
Slide-A-Lyzer Dialysis Cassettes Kit | Thermo Scientific | 66372 | MWCO 7000 0.5ml-3ml |
15ml Dounce Tissue grinder set | Sigma-Aldrich | D9938-1SET | Different sizes available |
Expand High Fidelity PCR Kit | Roche | 11 732 650 001 | |
10mM dNTP Mix | Invitrogen | Y02256 | 10mM each nucleotide |
MaxiScript SP6/T7 Kit | Ambion | AM1322 | |
m7G(5')ppp(5') G RNA Cap | New England Biolabs | S1404S | |
Century Marker Template Plus | Ambion | AM7782 | |
Easytides UTP [alpha-32p]-250uCi | Perkin Elmer | BLU507H250UC | |
Gel loading buffer II | Ambion | 8546G | |
DEPC treated water | Ambion | AM9906 | |
10x TAE | Fisher | BP13354 | |
10x TBE | Ameresco Life Sciences | 0658-4L | |
10x PBS | Fisher | BP399-20 | |
Urea | Fisher | BP169-212 | |
Ammonium Persulfate | Bio-Rad | 161-0700 | |
TEMED | Fisher | BP150-20 | |
Ammonium Acetate | Fisher | A637-500 | |
40% 19:1 Acrylamide:Bis-acrylamide | Bio-Rad | 161-0144 | |
Glycogen | Roche | 10 901 393 001 | |
100% Absolute Ethanol 200 Proof | Acros | 61509-0040 | |
Acid Phenol-Chloroform | Ambion | 9720 | For RNA |
Scintilation Fluid | Fisher | SX18-4 | |
Rnase Inhibitor | Promega | N261B | |
Poly (Vinyl alcohol) PVA | Sigma-Aldrich | P8136-250G | |
Creatine Phosphate | Calbiochem | 2380 | |
100mM dATP | Fisher | BP2560-4 | |
SDS | Acros | 23042-5000 | |
Proteinase K | Fisher | BP1700-100 | |
Adjustable Sequencing Unit | Sigma-Aldrich | Z351881-1EA | |
Binder Clips | Office Depot | 838-056 | |
20cmx42cm glass plates | Sigma-Aldrich | Z352543 | 1SET |
20cmx22cm glass plates | Sigma-Aldrich | Z35252-7 | 1SET |
20cmx42cm Aluminum Cooling Plates | Sigma-Aldrich | Z352667 | 1EA |
0.4mmx22cm Spacers | Sigma-Aldrich | Z35230-6 | 1SET |
0.4mmx42cm Spacers | Sigma-Aldrich | Z352314-1 | 1SET |
8-well Comb | Sigma-Aldrich | Z35195-4 | 1EA |
16-well Comb | Sigma-Aldrich | Z351962 | 1EA |
32-well Comb | Sigma-Aldrich | Z351970 | 1EA |
Gel Repel Coating | C.B.S. Scientific | SGR-0401 | or SGR-0101 for individual bottle |
Gel Loading Tips | Rainin | GT-10-4 | 0.1-10uL |
Sequencing PowerPac HV | Bio-Rad | PowerPac HV | 5000V/500mA/400W |
Gel Dryer Model 583 | Bio-Rad | Model 583 | |
Hydrotech Vacuum Pump for gel dryer | Bio-Rad | ||
Glogos II Glow-in-the-dark Markers | Agilent | 420201 | |
Film 8×10 | Midsci | BX810 | |
Film 14×17 | Phenix | F-BX1417 | |
Autoradiography Cassette | Fisher | FBCA 1417 | 8×10 size available |