Introduction
The function of an RNA molecule in vivo and in vitro will often depend on its tertiary structure, as well as the information encoded in its Watson-Crick base-pairing potential. There are now a number of powerful methods that can be used simultaneously to determine structural properties of small and large RNAs. This and subsequent Units provide methods for RNA footprinting - detecting positions sensitive to chemical and enzymatic attack in the presence or absence of bound proteins. RNA footprinting differs profoundly from DNA footprinting, in that the starting structure of the RNA cannot be taken for granted even in general terms, and the footprinting reactions are often carried out with structure-sensitive reagents precisely for the purpose of elucidating the RNA solution structure. When interpreting these structures, though, it is necessary to always consider whether the functional form of the RNA molecule is being examined.
Physiological and functional relevance
Sensitivity to chemical and enzymatic reagents can often be used to test hypotheses regarding solution exposure and single-stranded vs. double-stranded character along the folded RNA chain, but the data generated in these experiments is most effective when used in reference to a secondary structure model for the RNA generated by other approaches. Computer-based energy minimization algorithms have proven extremely useful in predicting simple RNA structures or in providing multiple local folding possibilities in larger RNAs (Gautheret et al 1990; Gorodkin et al 1997; Major et al 1991; Malhotra et al 1993; Walter et al 1994; Zuker 1989). The folding algorithms suffer from several severe limitations in predicting longer RNAs, however, tending to miss long-distance interactions and lacking sufficient predictions of non-standard interactions among nucleotides. In some instances, the physiologically relevant RNAs will also contain nucleoside modifications that are not necessarily known or taken into account. Perhaps most importantly, the functional form of the molecules might not be the most stable folded form under the conditions used to fold the RNA in vitro. It is possible that manipulation of the folding conditions might identify methods that drive the majority of a purified RNA into an appropriate structure, assuming that there is some method to determine what constitutes an appropriate structure. Manipulation of folding conditions for RNAs synthesized in vitro will be discussed below. However, folding of many RNAs in vivo is not readily reproduced in vitro, in many instances because the RNAs require association of proteins to assume physiologically appropriate structures. In these cases it might be necessary to either isolate the RNAs from cells, or to associate the appropriate proteins in vitro in a manner that can be demonstrated to cause the RNA to assume a functional structure. Lastly, it should be noted that the RNA might need to adopt more than one structure to carry out its function(s) in vivo. Driving all of the RNA sample (or protein-RNA complex) into a single form might reflect only one aspect of the functional RNA structure.
Strategic Planning
Footprinting with an Appropriate RNA
Before undertaking these often labor-intensive footprinting experiments it is worthwhile to evaluate the extent to which all or most of the RNA sample to be examined is in an appropriate folded state. For RNA which is to be probed in vitro, there are a number of ways of determining this, depending on what experimental system is available to assess the structure of the RNA. In the simplest case, the RNA might have an assayable function that can be demonstrated in vitro. An example of this is an artificial RNA ligand selected for binding to a protein target in vitro. RNAs isolated from cells are generally more difficult to interpret. If an RNA is stripped of its native protein structure in the process of being isolated from cells, or if an RNA is made synthetically, it should be assumed that the folding of that RNA is suspect until proven otherwise. Small RNAs that are tightly structured, such as tRNAs, can often refold in functional form, but this cannot be assumed. The chances of misfolding increase dramatically for larger RNAs where there are more potential isoforms. An RNA that is isolated as part of a ribonucleoprotein complex, such as a ribosomal subunit, is generally assumed to be correctly folded, but even that assumption should be viewed with caution.
Choice of structure-sensitive reagents
The object in RNA footprinting is to determine what positions in the RNA are accessible to specific types of attack by chemical reagents or nucleases. Cleavage or modification by these reagents is normally performed under conditions where the target RNA is cleaved at most once per molecule, thus lowering the chance that the first cleavage/ modification causes an RNA rearrangement that alters access for a second attack. The cleavage or modification specificities of individual nucleases and commonly used chemicals are listed in Table 1, and it is worth considering the general advantages and disadvantages of different classes of reagents. Nucleases have the lowest resolution because their steric radii do not allow access to many sites that are solvated, but nucleases are easily used in a wide range of physiological buffers. Most of the nucleases described below are specific for nucleotides not involved in Watson-Crick pairing, with variable sensitivity to other types of structure and base specificity. The exception is cobra venom ribonuclease (V1), which is the only reagent described here that is specific for double-stranded regions. Even V1 is not exceptionally helpful in identifying double-stranded structures, however, since the helix needs to be of a minimum length. Cleavage by V1 is infrequent, and V1 can cleave adjacent to, rather than within helices.
Table 1.
RNA Structure Probing Reagents and Specificities
Cleavage Reagent | Target |
DMS (dimethyl sulfate)* | G’s (N-7) |
DEPC (diethy pyrocarbonate)* | A’s (N-7) |
ENU (ethylnitroso urea)* | phosphate oxygens |
hydroxyl radicals | ribose sugar backbone |
Pb2+ | phosphates of SS nucleotides |
mung bean nuclease | SS nucleotides |
RNase A | SS C’s and U’s |
RNase CL3 | primarily C |
RNase I(ONE) | SS nucleotides |
RNase Phy M | A’s and U’s |
RNase T1 | G’s |
RNase T2 | SS nucleotides |
RNase U2 | SS A’s |
RNase V1 | DS nucleotides |
S1 nuclease | SS |
Modification Reagent | Target |
CMCT** | primarily U’s (N-3) and possibly G’s (N-1) |
DMS (dimethyl sulfate) | A’s (N-1) and C’s (N-3) |
kethoxal | G’s (N-1, 2-NH2) |
SS = single-stranded
DS = double-stranded
this reagent requires further chemical processing to cleave the RNA.
1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide metho-p-toluenesulfonate
The available chemical reagents discussed in this and subsequent Units allow better access to all regions of the RNA tertiary structure and cumulatively can attack a wide range of nucleotide positions. In general, the larger the number of different reagents used to probe, the more detailed the view of solvent-exposed positions. Exposure of phosphates, sugars, and aromatic ring positions can be probed, and reagents specific for Watson-Crick positions are often used as diagnostic for the existence of standard base pairing at contiguous nucleotides. Some of the protocols described below are based on methods described previously (Knapp 1989; Krol & Carbon 1989, Peattie & Gilbert 1980).
Location of chemical and nuclease cleavages can be determined using either end-labeled RNA or by primer extension. Hydroxyl radicals and Pb2+ will directly hydrolyze the phosphodiester backbone, as will the various nucleases according to their sequence and structural preferences. Other reagents, including DMS, DEPC, and ENU require additional chemical steps, ultimately resulting in strand scission.
Detection of Modifications or Cleavages
The method used to detect cleavages or modified positions varies with both the nature of the RNA being probed and the reagents being used. Short RNAs (<200 nt) can be labeled at either the 5′ or 3′ end, then folded and subjected to reagents that result in RNA strand cleavage. Separation of the cleaved RNA on denaturing polyacrylamide gels and detection of labeled fragment sizes identifies the position of cleavages. Alternatively, a labeled DNA oligomer primer can be annealed at any point along the length of an unlabeled RNA and extended to a point of cleavage or Watson-Crick base modification. Either case terminates extension and produces a labeled DNA fragment corresponding to the length from the label to the termination site. There are a variety of labels that can be used with these two general detection schemes. The most common label is 32P, which can be detected with either X-ray film or a phosphorimager.
The question of whether to use end-labeling or primer extension methods for detection depends on the RNA to be analyzed and how many reagents are needed. Primer extension is more often useful for longer RNAs, although end-labeling is still widely used to probe small RNAs. The advantage of end-labeling is that it requires fewer manipulations (i.e. - no primer extension reactions) after the cleavage reactions. Its disadvantages include that it can only be used for relatively short RNAs that can be labeled before cleavage reagent treatment and it requires cleavage of the RNA strand for detection. Transfer RNAs, 5S rRNAs, and other small, structured RNAs have typically been investigated by end-labeling. Primer extension has four main disadvantages. The first is that the extreme 3′ end of the RNA cannot be probed because the primer must anneal 3′ of the region to be extended across. The second disadvantage is that sequence-dependent pausing and termination by the reverse transcriptase, even on intact RNA, tends to give a high background at some positions that obscures true signal from cleavages or modifications. This problem can be reduced by using different reverse transcriptases and/or extension conditions, but is never completely eliminated with structured RNAs. The third disadvantage is that additional manipulations are required to perform the primer extension reactions following the cleavage or modification reactions. Finally, cellular RNAs may contain post-transcriptional modifications that will terminate the extension reaction. However, primer extension RNA footprinting is potentially useful in many situations. For example, RNA of any size can be probed because a labeled primer can be placed anywhere along the length of RNA, and multiple primers can be used in separate reactions to detect cleavages or modifications along the entire length. The RNA also does not need to be purified or labeled in advance, allowing probing of pre-formed RNA-protein complexes even in crude cell lysates. Following the cleavage/modification reactions, protein is removed by organic extraction and the purified RNA is subject to primer extension. Lastly, the extension reactions can be used to detect several chemical modifications at Watson-Crick positions, in addition to any form of RNA strand cleavage.
Basic Protocol 1: Dimethyl Sulfate (DMS) Modification and Cleavage of RNA
Dimethyl sulfate alkylates the N-7 cyclic amine of guanosine. The methylated nucleoside is reduced, opening the imidizole ring and weakening the glycosidic bond. Aniline treatment catalyzes the b-elimination of the ribose sugar leaving the nucleotide 5′ of the displaced guanosine with a 3′ phosphate and the nucleotide 3′ of the guanosine with a 5′ phosphate. The N-7 position is pointed into the major groove of an A-form RNA helix and is accessible to modification. Therefore, the N-7 of double-stranded and single-stranded guanosines should be detected unless participating in tertiary structure or non-Watson-Crick base pair interactions.
Materials
DMS (dimethyl sulfate)
HEPES, 1 M pH 7.8
KCl, 1 M
MgCl2, 0.1 M
carrier RNA, 1 mg/mL
Tris-HCl, 1 M pH 8.0
NaBH4, 0.2 M (fresh)
aniline-acetate buffer, 1 M
FEXS
Methylation of Guanosine N-7
1. Reaction mix:
reagent | mL | final conc. |
---|---|---|
HEPES, 1 M pH 7.8 | 8 | 200 mM |
KCl, 1 M | 4 | 100 mM |
MgCl2, 0.1 M | 4 | 10 mM |
carrier RNA, 1 mg/mL | 8 | 0.2 mg/mL |
2. Combine 50,000 cpm end-labeled RNA of interest or 0.5 mg unlabeled RNA (for primer extension) with H2O to make 15 mL total and add to reaction.
3. Add 1 mL DMS stock, mix, and incubate at room temperature. Stop the reaction after 2 min by precipitating with 4 mL of 3 M Na-acetate, pH 5.2 and 120 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
Reduction of methylated RNA and aniline cleavage
4. Resuspend pellets with 20 mL of 1 M Tris, pH 8.0 and 20 mL of 0.2 M NaBH4.
5. Incubate on ice in the dark for 30 min.
6. Add 4 mL of 3 M Na-acetate, pH 5.2 and 120 mL of 100% ethanol and precipitate on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, dry the pellet.
7. Resuspend pellet in 40 mL of 1 M aniline-acetate buffer and incubate in the dark at 60°C for 15 min.
8. Add 4 mL of 3 M Na-acetate, pH 5.2 and 120 mL of 100% ethanol and precipitate on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
9. For end-labeled RNA, resuspend in 12 mL FEXS for direct gel electrophoresis (Ellington 1989). See Support Protocol 3 for end-labeled RNA standards. See Support Protocol 1 for primer extension of unlabeled RNA samples.
Basic Protocol 2: Diethyl Pyrocarbonate (DEPC) Modification and Cleavage of RNA
Diethyl pyrocarbonate alkylates the N-7 cyclic amine of adenosine. Mechanism is essentially identical to that described in the introduction for Basic Protocol 1: Dimethyl Sulfate (DMS) Modification and Cleavage of RNA.
Materials
DEPC (diethyl pyrocarbonate)
HEPES, 1 M pH 7.8
KCl, 1 M
MgCl2, 0.1 M
carrier RNA, 1 mg/mL
Tris-HCl, 1 M pH 8.0
NaBH4, 0.2 M (fresh)
aniline-acetate buffer, 1 M
FEXS
Alkylation of Adenosine N-7
1. Reaction mix:
reagent | mL | final conc. |
---|---|---|
HEPES, 1 M pH 7.8 | 8 | 200 mM |
KCl, 1 M | 4 | 100 mM |
MgCl2, 0.1 M | 4 | 10 mM |
carrier RNA, 1 mg/mL | 8 | 0.2 mg/mL |
2. Combine 50,000 cpm end-labeled RNA of interest or 0.5 mg unlabeled RNA (for primer extension) with H2O to make 15 mL total and add to reaction.
3. Add 1 mL DEPC stock, mix, and incubate at room temperature. Stop the reaction after 45 min by precipitating with 4 mL of 3 M Na-acetate, pH 5.2 and 120 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
Reduction of DEPC modified RNA and aniline cleavage
See steps 4-9 of Basic Protocol 1: Dimethyl Sulfate Modification and Cleavage of RNA
Basic Protocol 3: Ethylnitrosourea (ENU) Modification and Cleavage of RNA
Ethylnitrosourea alkylates phosphate oxygens that are not involved in tertiary structure interactions. These include phosphates of both single-stranded and double-stranded nucleotides not participating in higher-ordered structure. Following alkaline treatment, the phosphotriester hydrolyzes resulting in RNA strand scission.
Materials
ENU-ethanol
HEPES, 1 M pH 7.8
KCl, 1 M
MgCl2, 0.1 M
carrier RNA, 1 mg/mL
Tris-HCl, 0.1 M pH 9.0
FEXS
Phosphate alkylation reaction
1. Reaction mix:
reagent | mL | final conc. |
---|---|---|
HEPES, 1 M pH 7.8 | 2 | 200 mM |
KCl, 1 M | 1 | 100 mM |
MgCl2, 0.1 M | 1 | 10 mM |
carrier RNA, 1 mg/mL | 2 | 0.2 mg/mL |
2. Combine 50,000 cpm end-labeled RNA of interest or 0.5 mg unlabeled RNA for primer extension with H2O to make 3 mL total and add to reaction.
3. Add 1 mL of ENU-ethanol, mix, incubate 30 min at room temperature. Stop the reaction by precipitating with 1 mL of 3 M Na-acetate, pH 5.2 and 30 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
Phosphotriester hydrolysis
4. Resuspend pellet in 10 mL of 0.1 M Tris-HCl, pH 9.0 and incubate at 50°C for 5 min.
5. Precipitate with 1 mL of 3 M Na-acetate, pH 5.2 and 30 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
6. Resuspend in 12 mL of FEXS for gel electrophoresis of end-labeled RNA (Ellington 1989). See Support Protocol 3 for end-labeled RNA standards. See Support Protocol 1 for primer extension of unlabeled RNA samples.
Basic Protocol 4: Nuclease Cleavage of RNA
Nucleases hydrolyze the RNA phosphodiester backbone resulting in strand scission. Cleavages can be identifed using either primer extension or by using end-labeled RNA. See Table 1 for nuclease specificities and cleavage locations. Some nucleases use divalent metal cofactors to catalyze the reaction. Divalent independent nucleases can be used to probe RNA in the presence and absence of divalent dependent tertiary structure. Most nucleases are specific for single stranded bases and require access to the base determinants for cleavage site recognition. Uncleaved nucleotide targets may be obscured by tertiary structure or local steric hindrince.
Being relatively large, nucleases may not penetrate some RNA structures completely, and it is common to observe some potential targets uncleaved. The resulting data obtained from one nuclease will identify general structural trends, single- or double-stranded character for example, but the data often needs to be corroborated using additional nucleases or chemical probes.
Materials
mung bean nuclease
RNase A
RNase CL3
RNase I(ONE)
RNase Phy M
RNase T1
RNase T2
RNase U2
RNase V1
S1 nuclease
Tris-HCl, 0.1 M pH 7.5
KCl, 1 M
MgCl2, 0.1 M
carrier RNA, 1 mg/mL
FEXS
Standard nuclease reaction
1. Reaction mix:
reagent | mL | final conc. |
---|---|---|
Tris-HCl, 0.1 M pH 7.5 | 1 | 10 mM |
KCl, 1 M | 1 | 100 mM |
MgCl2, 0.1 M | 1 | 10 mM |
carrier RNA, 1 mg/mL | 2 | 0.2 mg/mL |
2. Combine 50,000 cpm end-labeled RNA of interest or 0.5 mg unlabeled RNA for primer extension with H2O to make 4 mL total and add to reaction.
3. It is necessary to titrate the amount of each nuclease to cleave approximately 10% of the RNA of interest. Nuclease activity varies between lot and manufacturer. Set up six 5-fold serial dilutions, covering four orders of magnitude, as an initial screen. A second titration consisting of narrower increments may be necessary once the desired activity is found. Fresh nuclease dilutions should be used each time, although dilutions of RNases T1, A, and I(ONE) can be frozen with minimal loss of activity. Dilute nucleases with 10 mM Tris-HCl pH 7.5, except dilute RNase V1 with 10 mM Tris-HCl pH 7.5, 10 mM MgCl2.
The following activity Units have produced 10% cleavage using the reaction conditions described above and should be used as guidelines only:
RNase V1 | 0.047 U |
RNase T1 | 0.138 U |
RNase A | 0.2 ng |
RNase U2 | 1.0 U |
RNase Phy M | 1.0 U |
RNase CL3 | 0.025-0.100 U |
RNase I(ONE) | 0.012 U |
4. Add 1 mL of diluted nuclease, mix, and incubate at room temperature for 10 min.
5. Add 1 mL of 3 M Na-acetate, pH 5.2 and 30 mL of 100% ethanol and precipitate on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
6. Resuspend in 12 mL of FEXS for gel electrophoresis of end-labeled RNA (Ellington 1989). See Support Protocol 3 for end-labeled RNA standards. See Support Protocol 1 for primer extension of unlabeled RNA samples.
Basic Protocol 5: Modification of RNA at Watson-Crick Positions with Dimethyl Sulfate
Watson-Crick positions of adenosines (N-1) and cytidines (N-3) not involved in base-pairing or tertiary structure hydrogen bonding are alkylated by dimethyl sulfate (DMS). Primer extension (see Support Protocol 1) of modified RNA and comparison with dideoxynucleotide sequencing reactions identifies the modified bases. Used together, dimethyl sulfate (A & C), kethoxal (G), and CMCT (U) provide a complete analysis of the base-pairing status (i.e. - secondary structure) of the RNA of interest.
Materials
DMS (dimethyl sulfate)
HEPES, 0.5 M pH 7.8
KCl, 1 M
MgCl2, 0.1 M
Modification reaction
1. Combine:
reagent | mL | final conc. |
---|---|---|
HEPES 0.5 M pH 7.8 | 1 | 50 mM |
KCl, 1 M | 1 | 100 mM |
MgCl2, 0.1 M | 1 | 10 mM |
RNA of interest, 1 mg/mL | 2 | 0.2 mg/mL |
H2O | 4 |
2. DMS will need to be titrated to attain desired RNA modification. Dilute with 100% ethanol just prior to use. The following dilutions are provided as examples only. Make two-fold serial dilutions for initial titrations:
1 mL of stock (10.56 M) and 16.6 mL of 100% ethanol (= 600 mM).
Further dilute 2, 4, and 8-fold with 100% ethanol to make 300, 150, and 75 mM, respectively.
Final reaction concentrations are 60, 30, 15, and 7.5 mM.
3. Add 1 mL of DMS dilution to reaction, mix, and incubate for 20 min at room temperature.
4. Precipitate with 1 mL of 3 M Na-acetate, pH 5.2 and 30 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
5. Resuspend pellet in 40 mL of H2O; use 4 mL (0.2 mg) per primer extension reaction.
Basic Protocol 6: Modification of RNA at Watson-Crick Positions with Kethoxal
Kethoxal alkylates the N-1 and 2-NH2 Watson-Crick positions of guanosines not involved in base pairing or tertiary structure hydrogen bonding. Primer extension (see Support Protocol 1) of modified RNA and comparison with dideoxynucleotide sequencing reactions identifies the modified bases. Used together, dimethyl sulfate (A & C), kethoxal (G), and CMCT (U) provide a complete analysis of the base-pairing status (i.e. - secondary structure) of the RNA of interest.
Materials
kethoxal
HEPES, 0.5 M pH 7.8
KCl, 1 M
MgCl2, 0.1 M
Modification reaction
1. Combine:
reagent | mL | final conc. |
---|---|---|
HEPES 0.5 M pH 7.8 | 1 | 50 mM |
KCl, 1 M | 1 | 100 mM |
MgCl2, 0.1 M | 1 | 10 mM |
RNA of interest, 1 mg/mL | 2 | 0.2 mg/mL |
H2O | 4 |
2. Kethoxal will need to be titrated to attain desired RNA modification. Dilute kethoxal with H2O just prior to use. The following dilutions are provided as examples only. Make two-fold serial dilutions for initial titrations: 1 mL of stock (4.27 M) and 20.35 mL of H2O (= 200 mM)
Further dilute 2, 4, and 8-fold with H2O to make 100, 50, and 25 mM, respectively. Final reaction concentrations are 20, 10, 5, and 2.5 mM.
3. Add 1 mL kethoxal dilution to reaction, mix, and incubate for 20 min at room temperature.
4. Precipitate with 1 mL of 3 M Na-acetate, pH 5.2 and 30 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
5. Resuspend pellet in 40 mL of H2O; use 4 mL (0.2 mg) per primer extension reaction.
Basic Protocol 7: Modification of RNA at Watson-Crick Positions with CMCT
CMCT alkylates the N-3 Watson-Crick position of uridines not involved in base pairing or tertiary structure hydrogen bonding. Primer extension (see Support Protocol 1) of modified RNA and comparison with dideoxynucleotide sequencing reactions identifies the modified bases. Used together, dimethyl sulfate (A & C), kethoxal (G), and CMCT (U) provide a complete analysis of the base-pairing status (i.e. - secondary structure) of the RNA of interest.
Materials
CMCT (1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide metho-p-toluenesulfonate), 0.5 M K-borate, 0.5 M pH 8.0
KCl, 1 M
MgCl2, 0.1 M
Modification of uridine (N-3)
1. Combine:
reagent | mL | final conc. |
---|---|---|
K-borate, 0.5 M pH 8.0 | 1 | 50 mM |
KCl, 1 M | 1 | 100 mM |
MgCl2, 0.1 M | 1 | 10 mM |
RNA of interest, 1 mg/mL | 2 | 0.2 mg/mL |
2. CMCT will need to be titrated to attain desired RNA modification. All dilutions should be made just prior to use. Heat 0.5 M CMCT at 37°C and vortex to ensure the reagent is completely solubilized.
The following dilutions can be used for the initial titration:
4 mL of 0.5 M CMCT + 6 mL of H2O = 200 mM CMCT
1 mL of 0.5 M CMCT + 9 mL of H2O = 50 mM CMCT
Final reaction concentrations are 250, 100, and 25 mM.
3. Add 5 mL of CMCT dilution to reaction, mix, and incubate for 20 min at room temperature.
4. Precipitate with 1 mL of 3M Na-acetate, pH 5.2 and 30 mL of 100% ethanol on dry ice. Spin 10,000 rpm for 30 min, remove supernatant, rinse with 70% ethanol, dry the pellet.
5. Resuspend pellet in 40 mL of H2O; use 4 mL (0.2 mg) per primer extension reaction.
Support Protocol 1: Detection of RNA Cleavage or Modification by Primer Extension of DNA Oligomers with Reverse Transcriptase
Primer extension of a radiolabeled DNA oligomer will synthesize a complementary DNA strand to the RNA of interest. Reverse transcriptase will stop upon encountering alkylated Watson-Crick determinants, resulting in a product one base short of the modification. The enzyme will also run-off RNA cleavage products producing a DNA corresponding to the site of cleavage. Strong secondary structure, such as hairpin loops within the RNA, can cause the enzyme to pause, creating a prematurely terminated product. It is therefore essential to perform the reaction on unmodified/uncleaved RNA for use as a control. This enables pause sites from such secondary structure to be subtracted from the modification/cleavage analysis. Sometimes carrier RNA can affect the reverse transcription reaction, therefore control reactions should include the carrier RNA as well. The reaction requires a DNA oligomer complementary to the RNA of interest, 3′ of the region to be examined. It might be necessary to use several DNA oligomers to examine an RNA several hundred nucleotides long. The extension products are resolved on standard DNA sequencing gels, allowing approximately 200 nucleotides to be examined per DNA primer. Location of primer extension terminations is determined by comparison to standard dideoxynucleotide sequencing reactions using the same primer and a template of the same sequence as the RNA being examined.
The primer extension reaction requires a 32P-labeled DNA oligomer which is generated using bacteriophage T4 polynucleotide kinase. Reaction conditions for labeling the primer are identical to those described in Support Protocol 2: Labeling the 5′ RNA terminus with T4 polynucleotide kinase and g-32P-ATP, substituting the DNA oligomer for the RNA.
Materials
dNTPs (25 mM each dATP, dCTP, dGTP, and dTTP)
DTT (dithiothreitol), 0.1 M
5′-32P-DNA oligomer (see Support Protocol 1)
5X first strand buffer
Superscript II ® (RNase H(-) MMLV reverse transcriptase, GIBCO-BRL)
FENXB
Anneal primer
1. Mix the following to make 18.5 mL total:
100,000 cpm of 5′-32P-DNA oligomer
0.2 mg of RNA of interest (cleaved or modified at Watson-Crick positions)
H2O
A control reaction, using unmodified/uncleaved RNA, should be performed to identify structure induced terminations and to ensure the majority of the DNA oligomer is being fully extended through the region of interest. It may be necessary to titrate the 32P-DNA oligomer and RNA concentrations to optimize the extension reaction conditions.
2. Incubate: 70°C for 10 min
42°C for 30 min
3. Briefly spin down any condensation that forms.
Reaction mix
4. Add: 3 mL of dNTPs
2 mL of DTT, 0.1 M
6 mL of 5X first strand buffer
Total volume is now 29.5 mL.
5. Add 0.5 mL of Superscript II ® and mix.
6. Incubate at 42° for 45 min.
7. Add 3 mL of Na-acetate, 3 M pH 5.2 and 90 mL of 100% ethanol.
8. Precipitate on dry ice, centrifuge at 10,000 rpm for 30 min. Remove supernatant, rinse pellet with 70% ethanol, and dry the pellet. Resuspend in 6 mL of FENXB for resolution on sequencing gels (6% acrylamide (20:1 mono-:bis-), 8 M urea, 1X TBE). Use dideoxynucleotide sequencing reactions as markers.
Typically a cDNA, rather than an RNA template, is used to allow sequencing reactions to be performed with a DNA-dependent DNA polymerase (e.g. Sequenase®, United States Biochemical, Cleveland, OH).
Support Protocol 2: Labeling the 5′ RNA Terminus using T4 Polynucleotide Kinase and g-32P-ATP
Bacteriophage T4 polynucleotide kinase will transfer the gamma phosphate of g-32P-ATP to an RNA substrate containing a 5′ hydroxyl group; or, will exchange the radiolabeled g-phosphate with a single 5′ phosphate present on the RNA substrate. The 5′ end of cellular RNAs may contain methyl guanosine cap structures that prohibit labeling by this method. Chemical solid-phase synthesis produces an RNA with a 5′ hydroxyl available for radiolabeling using g-32P-ATP and polynucleotide kinase, whereas RNA transcribed in vitro has a 5′ triphosphate that must first be removed with a phosphatase. Dephosphorylation is accomplished with bacterial alkaline phosphatase and accompanying buffer (GIBCO BRL, Cat. No. 18011-015, see also technical bulletin 8011-1 from same for reaction conditions). An alternate method to generate 5′ labeled RNA by enzymatic synthesis is to transcribe in vitro with T7 RNA polymerase and g-32P-GTP, incorporating 32P at only the 5′ terminal triphosphate (Milligan & Uhlenbeck 1989).
Materials
BAP (bacterial alkaline phosphatase), 150 U/mL (GIBCO-BRL Cat. No. 18011-015) 10X dephosphorylation buffer (100 mM Tris pH 8.0)
EDTA, 10 mM
DTT, 0.1 M
10X kinase buffer
T4 polynucleotide kinase, 10,000 U/mL
Stop mix
g-32P-ATP, 6000 Ci/mmol
Phosphatase treatment of RNA (optional)
1. Calculate pmoles of RNA in 0.2 mg. Use 70 U BAP for each pmole of RNA, dilute enzyme with 1X dephosphorylation buffer.
2. Set up 100 mL reaction: 10 mL of 10X dephosphorylation buffer 90 mL of RNA, BAP, and H2O
3. Incubate at 65°C for 1 hour. Stop reaction by extracting with 100 mL of phenol:chloroform:isoamyl alcohol (25:24:1).
4. Remove the aqueous phase to a new tube. Add 10 mL of Na-acetate, 3 M pH 5.2 and 300 mL of 100% ethanol. Precipitate on dry ice, centrifuge at 10,000 rpm for 30 min. Remove supernatant, rinse with 70% ethanol, and dry the pellet
Denature the RNA sample (optional)
Denaturation will increase labeling efficiency of base-paired or recessed 5′ termini; however, some RNA samples may be labeled more efficiently when folded.
5. Bring 0.2 mg RNA to 18 mL with H2O, then add 2 mL of 10 mM EDTA.
6. Heat at 90-100°C for 2 min, then quickly transfer to ice-water bath.
7. After 5 min, briefly spin down any condensation.
Kinase reaction
8. Take 13 mL of denatured RNA and add:
2 mL of 10X kinase buffer
2 mL of 0.1 M DTT
2 mL of g-32P-ATP, 6000 Ci/mmol
1 mL of T4 polynucleotide kinase
Incubate at 37°C for 1 hr.
9. Add 4 mL of Stop mix and 100 mL of 100% ethanol (5 volumes).
10. Precipitate on dry ice, then centrifuge 10,000 rpm for 30 min.
11. Carefully remove supernatant (very radioactive) and dry the pellet.
If the RNA is not going to be used immediately, resuspend the pellet in the appropriate buffer or H2O to reduce radiation-induced damage and freeze at −80°C.
12. It is necessary to purify the radiolabeled RNA by denaturing gel electrophoresis if a heterogeneous population exists (Ellington, 1989). The RNA must have a discrete length to accurately identify nucleotide cleavages/modifications in the subsequent structure probing experiments.
Support Protocol 3: Labeling the 3′ RNA Terminus using T4 RNA Ligase and [32P]pCp
Bacteriophage T4 RNA ligase will catalyze phosphodiester formation between a nucleotide 3′ hydroxyl and a nucleotide 5′ phosphate. The RNA of interest requires a 3′ hydroxyl which will be coupled to [5′-32P]cytidine 3′,5′-bisphosphate ([32P]pCp). [32P]pCp can be purchased (Amersham, Arlington Heights, IL) or generated using 3′-cytosine monophosphate, g-32P-ATP, and T4 polynucleotide kinase (England et al 1980). The latter method is recommended because it generates the concentrated form needed at a reasonable cost. Enzymatically synthesized RNAs have a 3′ hydroxyl that can be directly ligated with radiolabeled [32P]pCp. However, most chemically synthesized RNAs have a 3′ phosphate present which must first be removed with a phosphatase. Dephosphorylation is accomplished with bacterial alkaline phosphatase and accompanying buffer (GIBCO BRL, Cat. No. 18011-015, see also technical bulletin 8011-1 from same for reaction conditions). Breakdown of RNA by contaminating nucleases, metal or alkaline hydrolysis, and cleavage by ribozymes (excl. RNase P) produces 2′-3′ cyclic phosphates, which subsequently hydrolyze to 3′ phosphates and cannot directly be 3′ end-labeled.
Materials
BAP (bacterial alkaline phosphatase), 150 U/mL (GIBCO-BRL Cat. No. 18011-015) 10X dephosphorylation buffer (100 mM Tris pH 8.0)
HEPES, 0.5 M, pH 7.9 @ 50 mM
MgCl2, 0.1 M
BSA (bovine serum albumin), 0.1 mg/mL
DTT, 30 mM
ATP, 1 mM
DMSO (dimethyl sulfoxide), 100%
500 mM [32P]pCp
T4 RNA ligase, 20-25 U/mL
Phosphatase treatment of RNA (optional)
1. Calculate pmoles of RNA in 2 mg. Use 70 U BAP for each pmole of RNA, dilute enzyme with 1X dephosphorylation buffer.
2. Set up 100 mL reaction: 10 mL of 10X dephosphorylation buffer 90 mL of RNA, BAP, and H2O
3. Incubate at 65°C for 1 hour. Stop reaction by extracting with 100 mL of phenol:chloroform:isoamyl alcohol (25:24:1).
4. Remove the aqueous phase to a new tube. Add 10 mL of Na-acetate, 3 M pH 5.2 and 300 mL of 100% ethanol. Precipitate on dry ice, centrifuge at 10,000 rpm for 30 min. Remove supernatant, rinse with 70% ethanol, and dry the pellet
5. Resuspend in 10 mL of H2O to make 0.2 mg/mL (≈ 6 mM for a 100 nucleotide RNA).
Ligation reaction
6. Mix the following (10 mL total vol.):
reagent | mL | final conc. |
---|---|---|
RNA of interest @ 10 mM | 1 | 1 mM |
HEPES, 0.5 M, pH 7.9 @ 50 mM | 1 | 50 mM |
MgCl2, 0.1 M | 2 | 20 mM |
BSA, 0.1 mg/mL | 1 | 10 mg/mL |
DTT, 30 mM | 1 | 3 mM |
ATP, 1 mM | 1 | 0.1 mM |
DMSO, 100% | 1 | 10% (v/v) |
500 mM [32P]pCp | 1 | 50 mM |
T4 RNA ligase, 20-25 U/ mL | 1 | 2-2.5 U/ mL |
7. Incubate at 4°C for 12+ hr; typically overnight.
8. Add 90 mL of H2O, mix, then extract with 100 mL of phenol:chloroform:isoamyl alcohol (25:24:1).
9. Add 10 mL of 3 M Na-acetate, pH 5.2 and 300 mL of 100% ethanol.
10. Precipitate on dry ice and centrifuge 10,000×g for 30 min. 11. Remove supernatant and dry the pellet.
If the RNA is not going to be used immediately, resuspend the pellet in the appropriate buffer or H2O to reduce radiation-induced damage and freeze at −80°C.
12. It is necessary to purify the radiolabeled RNA by denaturing gel electrophoresis if a heterogeneous population exists (Ellington 1989). The RNA must have a discrete length to accurately identify nucleotide cleavages/modifications in the subsequent structure probing experiments.
Support Protocol 4: Standards for Electrophoresis of End-Labeled RNA Cleavage Products
Limited alkaline hydrolysis and digestion with RNase T1 under denaturing conditions are used to determine sites of end-labeled RNA cleavage. Alkaline hydrolysis results in a relatively even ladder of truncated RNAs. A denaturing RNase T1 digest cleaves the RNA 3′ of each guanosine. Comparison of the two reactions products after electrophoresis identifies guanosines along the alkaline hydrolysis ladder and therefore the intervening products present in the alkaline hydrolysis reaction can be determined from the RNA sequence. End-labeled RNA cleavage products from structure probing reactions can thus be identified.
Materials
Na2CO3-EDTA
CU buffer
RNase T1
tRNA carrier, 2 mg/mL
CEU buffer
FEXS
[32P]-end-labeled RNA of interest
Alkaline hydrolysis
1. Ethanol precipitate 50,000 cpm of end-labeled RNA, spin 10,000 rpm for 30 min, and dry the pellet.
2. Resuspend with 1.2 mL of Na2CO3-EDTA and mix with pipette.
3. Incubate: 95°C for 90-120 seconds ice for 1 min
Vary incubation time at 95°C to achieve a relatively even ladder of products following electrophoresis.
4. Neutralize by adding 2.8 mL of CU buffer and mix.
5. Add 8 mL of FEXS. Store frozen at −80°C, if necessary.
Denaturing RNase T1 digest
1. Add 5 mL of tRNA carrier to 50K cpm of end-labeled RNA. Ethanol precipitate with 1/10 volume 3 M Na-acetate, pH 5.2 and 3 volumes 100% ethanol, spin 10,000 rpm for 30 min, and dry the pellet.
2. Resuspend with 3 mL of CEU buffer and mix with pipette.
3. Incubate at 50°C for 5 min.
4. Add 1 mL of diluted RNase T1 and incubate at 50°C for 15 min.
RNase T1 dilution will need to be titrated, 1-3 U typically produces good results. Alternatively, other urea tolerant nucleases, such as RNase A (C & U specific) can be used.
5. Add 8 mL of FEXS. Store frozen at −80°C, if necessary.
REAGENTS AND SOLUTIONS
g-32P-ATP, 6000 Ci/mmol.
DuPont NEN, Cat. No. NEG-035C, 10 mCi.
Crude preparation is sufficient for kinase reaction and less expensive.
aniline-acetate buffer, 1 M
10 mL of aniline, 99.5% (HIGHLY TOXIC)
93 mL of H2O
6 mL of acetic acid, glacial
carrier RNA, 1 mg/mL
Carrier RNA can be unlabeled RNA of interest or total yeast tRNA. Extract yeast tRNA with phenol:chloroform:isoamyl alcohol (25:24:1) and ethanol precipitate to remove any protein contaminants.
CEU buffer
25 mM citrate | for 100 mL: 10.25 mL of 0.1 M citric acid |
1 mM EDTA | 14.75 mL of 0.1 M NaCitrate |
7 M Urea | 0.2 mL of 0.5 M EDTA |
pH 4.7-5.0 | ~ 30 mL of H2O |
42.04 g of urea | |
pH w/ HCl |
CMCT (1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide metho-p-toluenesulfonate) for 0.5 M: 212 mg in 1 mL of H2O store at −20°C
[32P]pCp, 500 mM ([5′-32P]cytidine 3′,5′-bisphosphate)
Available from Amersham or make using protocol from England et al 1980.
CU buffer
25 mM citrate | for 100 mL: 10.25 mL of 0.1 M citric acid |
10 M urea | 14.75 mL of 0.1 M NaCitrate |
pH 4.7-5.0 | ~ 30 mL of H2O |
60.06 g of urea | |
pH w/ HCl |
DEPC (diethyl pyrocarbonate)
TOXIC, store at 2-8°C
DMS (dimethyl sulfate)
HIGHLY TOXIC, CANCER SUSPECT AGENT
ENU-ethanol
Add ENU (N-nitroso-N-ethylurea) to 100 mL of 100% ethanol until solution is saturated; spin to pellet insoluble reagent. ENU is HIGHLY TOXIC and a CANCER SUSPECT AGENT.
FENXB
formamide, 95%
EDTA, 20 mM
NaOH, 2 mM
xylene cyanol, 0.05%
bromophenol blue, 0.05%
FEXS
formamide, 95%
EDTA, 10 mM
xylene cyanol, 0.05%
0.1% SDS
first strand buffer, 5X
Tris-HCl, 250 mM, pH 8.3
KCl, 375 mM
MgCl2, 15 mM
kethoxal (Pharmacia-Upjohn, Kalamazoo, MI)
kinase buffer, 10X
Tris-HCl, 700 mM, pH 7.6
MgCl2, 100 mM
DTT, 50 mM
Na2CO3-EDTA
50 mM Na2CO3, pH 11.7 | for 1 mL: 50 mL of 1 M Na2CO3 |
1 mM EDTA | 2 mL of 0.5 M EDTA |
948 mL of H2O |
NaBH4, 0.2 M (fresh)
7.6 mg in 1 mL H2O
RNase I(ONE) (Promega, Madison, WI 53711-5399)
RNase Phy M (Pharmacia, Piscataway, NJ 08854)
RNase T1 (GIBCO BRL, Gaithersburg, MD)
RNase U2 (Pharmacia, Piscataway, NJ 08854)
RNase V1 (Pharmacia, Piscataway, NJ 08854)
Note: this enzyme requires divalent cations for activity, 10 mM MgCl2 final concentration is recommended.
S1 nuclease (Boehringer Mannheim, Indianapolis, IN 46250-0414)
Stop mix
EDTA, 20 mM
NaCl, 0.3 M
glycogen, 1 mg/mL
Superscript II®, RNase H(-) MMLV reverse transcriptase
(GIBCO BRL, Gaithersburg, MD)
tRNA carrier, 2 mg/mL
see also carrier RNA
water, RNase free
Test water source for contaminating nuclease activity by incubating labeled RNA in water and buffers for intended reaction time. Ascertain integrity of sample by electrophoresis. Often deionized and glass-distilled water will be free of nucleases. However, if water and/or buffer reagents contain nuclease activity, treat with DEPC to inactivate. Add 0.2 mL of DEPC per 100 mL of solution. Shake vigorously to get the DEPC into solution. Autoclaving will then neutralize most of the remaining DEPC. Autoclaved solutions are sometimes allowed to stand at room temperature, with loose caps, for several days to remove all DEPC. Trace DEPC can inhibit reverse transcriptase and other enzymes.
COMMENTARY
Background Information
Structural analysis of RNAs and ribonulceoproteins (RNPs) can be used to determine secondary structure, facilitate modeling of tertiary structure, and identify RNA-protein contacts. The source of the material to be probed will often dictate which labeling and probing techniques are necessary. Primer extension is the preferred method for analysis of large or cellularly derived RNAs. The native RNA or RNP complex can be assayed without manipulations that may disrupt integrity of the sample. Post-modification or cleavage products are assayed by extension of labeled DNA oligomers with reverse transcriptase. End labeling techniques can sometimes be successful. However, in vivo derived RNAs often possess 5’ methyl cap structures preventing labeling at that terminus, and one or both termini in the native RNA can be recessed or obscured by protein in RNP structures. Chemical or enzymatic synthesis of RNA permits incorporation of labeling groups during synthesis, or if necessary, labeling sites can be engineered in the synthetic RNAs.
Studies involving novel sequences, subregions of larger RNAs, or RNP reconstitution experiments sometimes require the use of synthetic RNA. Chemical and enzymatic synthesis of RNA also provides an opportunity to incorporate nucleosides modified with fluorescent, antigenic, or crosslinking groups during synthesis.
Critical Parameters
The first principle in RNA structure analysis is to begin with an RNA properly folded into its physiologically relevant structure. Gentle purification techniques should retain the in vivo conformation of cellularly derived RNAs and RNPs. Synthetic RNA may or may not be in its properly folded state. The best insurance that the RNA or RNP of interest is correctly folded is to test its functional or enzymatic activity. Other folding assays will be necessary if no functional assay exists.
The number of folding isoforms of an RNA sample can sometimes be determined using native polyacrylamide gel electrophoresis. RNAs with extensive self-complementarity often form multimers which can be detected on the native gel. Heat denaturation can be used to melt present structures. Cooling rate, RNA concentration, and the presence of mono- and divalent cations all affect the subsequent folding isoform populations. In general, a faster cooling rate, the absence of cations, and low RNA concentrations favor intramolecular folding. Likewise, slow cooling in high salt with high RNA concentrations favors intermolecular folding and the formation of multimers.
Once the RNA is properly folded, the structure probing experiments can begin. There are four factors that are key to experiment reproducibility. The first is a consistent reaction mix. Make a batch of all ingredients minus the reagent, mix, then aliquot to the necessary number of tubes. Always make enough for the number of desired reactions plus one. This will ensure the amount of each component, especially radiolabeled RNA, is identical between reactions. The second factor is reaction time. Many of the probing reactions are short (i.e. 10 min) to minimize the effect of any contaminating nuclease activity. The number of reactions attempted at once should not compromise strict adherence to the reaction time. Longer reaction times may be used effectively, but short times have been found to reduce secondary or non-specific activities. The third factor is the use of unlabeled carrier RNA. Carrier RNA provides a constant substrate concentration for the probing reagents. The concentration of radiolabeled RNA is much less than the carrier and is negligible in most reactions. Therefore, any changes in labeling efficiency between RNA preparations will have little effect. Lastly, it is necessary to titrate the amount of reagent for each particular RNA of interest so that approximately 10% of the RNA is cleaved. The likelihood of multiple cleavages per molecule is thereby reduced. Typically, the concentration of RNA is held constant and the amount of reagent is titrated until the desired extent of modification is achieved. If the RNA sample of interest is extremely limited, reagents are titrated using a different RNA with the identical unlabeled carrier RNA and buffer conditions. The presence of excess carrier RNA facilitates reproducibility of cleavage conditions.
Troubleshooting
Primer extension analysis of probing reactions can encounter several pitfalls. For example, breakdown in the sample results in termination products not produced by the probing reagents. Extension of unmodified/uncleaved RNA should identify these sites and allow for their subtraction from the probing data. However, extensive degradation or strong secondary structure induced terminations may obscure cleavages or modifications at several nucleotides of interest. Synthetic RNA sample integrity can be improved by changing the original purification strategy, or by further purification to ensure a discrete population exists. RNA samples in crude cellular fractions should be prepared as rapidly as possible at 0-4°C and stored frozen (−80°C) when not in use to minimize degradation by endogenous nucleases. To alleviate secondary structure induced terminations, the primer extension reaction conditions and/or the reverse transcriptase can be changed. Superscript II® has demonstrated an ability to proceed quite well through secondary structure, but other reverse transcriptases, such as AMV and Retrotherm®, may function better on a particular RNA. The heat stable Retrotherm® can also be used at elevated temperatures to minimize RNA secondary structure.
Achieving an interpretable amount of modification or cleavage is another problem often encountered. The desired modification or cleavage goal is 10% of the total RNA, in the realm of single-hit kinetics. This rule minimizes the presence of products produced by multiple modifications or cleavages. Typically, the concentration of reagent is adjusted until the 10% goal is reached for a batch of RNA, carrier RNA, and reagent. Single cleavages or modifications per RNA molecule usually allow for detection of all available targets within the region of interest. However, a hyper-reactive site may exist that prevents further primer extension or observation of end-labeled RNA products larger than the hyper-reactive nucleotide. In this scenario, a new primer complementary to the region 5′ of the RNA modification will be needed, or in the case of end-labeled RNA, labeling the opposite end will allow analysis of the remaining RNA region of interest.
All of the probing reactions are purposefully designed to have nearly identical buffer and salt conditions. RNA structure can change considerably based on the amount and nature of counterions and pH. Consistency in reaction conditions between different probing reagents helps to ensure the same structure is being examined. The buffers and pH are changed only when they may interfere with the chemical nature of the reagent. It should be noted that the maximal activities of several reagents and most of the nucleases occur under different conditions than those listed, but all have been used successfully as described. The standard salt conditions can be altered as desired but consistency is important. Also, formation of RNA tertiary structures is often very dependent on the amount and type of counter-ions present.
Uncharacteristic modifications or cleavages from probing reagents are sometimes observed. Interpretation of atypical activities should be regarded with appropriate caution. For example, reagents used outside their optimal reaction conditions or the presence of contaminating activities could be responsible. It is also possible to observe inconsistencies between reagents with the same targets. Thus, using two reagents that should cleave or modify an unpaired nucleotide might give the result that the nucleotide is sensitive to one, but not the other. In all probability, the reagents are functioning properly, but subtle differences in substrate binding and recognition can favor one over the other.
Anticipated Results
There are several cautionary notes that should be considered when interpreting RNA footprinting data. First, it should always be remembered that both purified RNAs and RNA-protein complexes can exist in multiple isoforms, whether folded and bound in vitro or isolated intact from cells. As far as possible, the degree to which the RNA sample of interest is heterogeneous should be investigated and taken into account when interpreting sensitivity. Second, sensitivity to cleavage or modification reagents is far more useful in showing solution exposure than “protection” from reagents. Inaccessibility of even the same region of a base to an individual enzyme or chemical can vary from reagent to reagent, and can have a number of causes including involvement in hydrogen bonding, folding into the interior of the RNA where it is less exposed to the solvent, inappropriate chemical states, and direct blockage by interacting proteins. Thus, in an example where protection is observed in a particular region when a protein is added, the protection might result either from direct coverage by the protein or indirect effects of the bound protein acting at a distance. It should be re-emphasized that even detailed footprinting results are often consistent with multiple hypothetical structures, and for any complex RNA it is likely that additional, independent types of evidence will be needed to sort out the possibilities.
Results of structure probing experiments are therefore useful to qualitatively model RNA structure. The more reagents employed, the more descriptive the analysis. Suspected anomalous modifications or cleavages by a single reagent can therefore be subtracted. Modification/cleavage products can be quantitated by phosphorimager analysis and analyzed using a computer (ImageQuant™, from Molecular Dynamics, Sunnyvale, CA). The accessibility of an RNA target is directly proportional to the modification/cleavage product band intensity. End-labeled RNA or labeled primer extension products have one label per product molecule. However, the reason one target is preferred over another is not a simple interpretation. Targets on the surface of a molecule are likely to be more accessible than those buried within the interior of a tertiary structure, although local steric effects may have as much impact on reagent reactivity as global structure. Data obtained using multiple reagents will better uncover the true solution structure of a target, and the trend of several reagents should be regarded as the rule, rather than a single hit by one reagent. One should also keep in mind that solution structure is dynamic, and RNA molecular motion is dependent on temperature and ionic environment.
Time Considerations
The design of structure probing experiments is dictated by the stability of the RNA sample of interest and the half-life of radiolabeled material. Cellularly derived RNAs or RNPs often contain contaminating nuclease activities, and these can be considerable depending on the purification scheme employed. It is therefore imperative to preserve samples frozen at −80°C and use immediately upon thawing. Synthetic RNAs are often free of contaminating nucleases, depending on the handling and manipulations the sample has received. The structure probing reactions are usually short (i.e. ≤10 min.) to reduce the effects of any contaminating activities. Control reactions should identify extraneous cleavages, providing the majority of the sample is not degraded in the reaction interval. A time course incubation of the RNA of interest in the desired reaction buffer may prove beneficial in identifying sample integrity vs. incubation time. Longer reaction times, allowing for more samples, might then be used.
Modified/cleaved RNA intended for primer extension analysis can be stored frozen at −80°C while the radiolabeled DNA primer is made. End-labeled RNA should be resolved by electrophoresis promptly after modification/cleavage reactions are complete. Note that only 10% of the RNA sample should be modified/cleaved, therefore detectable radiolabel signal from the modified/cleaved products will only be 10% of the sample’s specific activity. Experiments should proceed readily to maximize signal from a batch of radiolabeled RNA.
Once all reaction components are in hand, setting up the probing reactions, incubating, and terminating the reactions for a single reagent should take approximately 1-3 hours. The primer extension procedure typically takes 2-3 hours and gel electrophoresis of standard sequencing (6%) polyacrylamide gels requires 2-3 hours.
Footnotes
This chapter describes the analysis of RNA solution structure using structure-sensitive chemicals and nucleases. Detection methods are also included.
Literature Cited
- Ellington A. Purification of Oligonucleotides Using Denaturing Polyacrylamide Gel Electrophoresis. Current Protocols in Molecular Biology, John Wiley & Sons, NY. 1989 doi: 10.1002/0471142727.mb0212s42. Unit 2.12. [DOI] [PubMed] [Google Scholar]
- England T, Bruce A, Uhlenbeck O. Specific labeling of 3′ termini of RNA with T4 RNA ligase. Methods in Enzymology. 1980;65:65–74. doi: 10.1016/s0076-6879(80)65011-3. [DOI] [PubMed] [Google Scholar]
- Gautheret D, Major F, Cedergren R. Computer modeling and display of RNA secondary and tertiary structures. Methods in Enzymology. 1990;183:318–30. doi: 10.1016/0076-6879(90)83021-z. [DOI] [PubMed] [Google Scholar]
- Gorodkin J, Heyer L, Stormo G. Finding the most significant common sequence and structure motifs in a set of RNA sequences. Nucleic Acids Research. 1997;25:3724–3732. doi: 10.1093/nar/25.18.3724. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Knapp G. Enzymatic approaches to probing of RNA secondary and tertiary structure. Methods in Enzymology. 1989;180:192–212. doi: 10.1016/0076-6879(89)80102-8. [DOI] [PubMed] [Google Scholar]
- Krol A, Carbon P. A guide for probing native small nuclear RNA and ribonucleoprotein structures. Methods in Enzymology. 1989;180:212–27. doi: 10.1016/0076-6879(89)80103-x. [DOI] [PubMed] [Google Scholar]
- Major F, Turcotte M, Gautheret D, Lapalme G, Fillion E, Cedergren R. The combination of symbolic and numerical computation for three-dimensional modeling of RNA. Science. 1991;253:1255–60. doi: 10.1126/science.1716375. [DOI] [PubMed] [Google Scholar]
- Malhotra A, Gabb H, Harvey S. Modeling large nucleic acids. Current Opinion in Structural Biology. 1993;3:241–246. [Google Scholar]
- Milligan JF, Uhlenbeck OC. Synthesis of Small RNAs Using T7 RNA Polymerase. Methods in Enzymology. 1989;180:51–62. doi: 10.1016/0076-6879(89)80091-6. [DOI] [PubMed] [Google Scholar]
- Peattie DA, Gilbert W. Chemical probes for higher-order structure in RNA. Proceedings of the National Academy of Sciences of the United States of America. 1980;77(8):4679–82. doi: 10.1073/pnas.77.8.4679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Walter AE, Turner DH, Kim J, Lyttle MH, Muller P, et al. Coaxial stacking of helixes enhances binding of oligoribonucleotides and improves predictions of RNA folding. Proceedings of the National Academy of Sciences of the United States of America. 1994;91:9218–22. doi: 10.1073/pnas.91.20.9218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zuker M. On finding all suboptimal foldings of an RNA molecule. [Review] [32 refs] Science. 1989;244:48–52. doi: 10.1126/science.2468181. [DOI] [PubMed] [Google Scholar]