Epigenetic Regulation and Epigenomics

Part I

Analytical Methods

1

RNA Methodologies

Robert E. Farrell, Jr.

Penn State University, Department of Biology, 1031, Edgecomb Avenue, NY, PA 17403, USA

1 Introduction

2 Subpopulations of RNA

2.1 Messenger RNA (mRNA)

2.2 Transfer RNA (tRNA)

2.3 Ribosomal RNA (rRNA)

2.4 Nuclear RNA

2.5 Organellar RNA

2.6 Noncoding RNA

3 Goals in the Purification of RNA

3.1 Goal 1: Select an Appropriate Method for Membrane Solubilization

3.2 Goal 2: Ensure Total Inhibition of Nuclease Activity

3.3 Goal 3: Remove Contaminating Proteins from the Sample

3.4 Goal 4: Concentrate the Sample

3.5 Goal 5: Select the Correct Storage Conditions for the Purified RNA

4 Methods of Cellular Disruption and RNA Recovery

4.1 Gentle Lysis Buffers

4.2 Harsh Lysis Buffers

4.3 Silica Separation Technology

4.4 Affinity Matrices

5 Inhibition of Ribonuclease Activity

5.1 Preparation of Equipment and Reagents

5.2 Inhibitors of RNase

6 Methods for the Analysis of RNA

6.1 RT-PCR

6.2 Northern Analysis

6.3 Nuclease Protection Assay

6.4 Transcription Rate Assays

6.5 Dot-Blot Analysis

6.6 High-Throughput Transcription Analysis

6.7 Suppression Subtractive Hybridization (SSH)

6.8 RNAi

6.9 In Vitro Translation

7 Summary

References

Keywords

Chaotropic

Biologically disruptive. Chaotropic lysis buffers disrupt the cell and organelle membranes and destroy enzymatic activity on contact.

Complementary DNA (cDNA)

DNA synthesized in vitro from an RNA template by an enzyme known as a reverse transcriptase. cDNA can be either single- or double-stranded, and is used for RT-PCR, nucleic acid probe synthesis, or library construction. Because cDNA can only be made from transcripts present at the moment of cellular disruption, it is a permanent biochemical record of the cell.

Dot-blot

A membrane-based technique for the quantification of specific RNA or DNA sequences in a sample. The sample is usually dot-configured onto a filter by vacuum filtration through a manifold. Dot blots lack the qualitative component associated with electrophoretic assays.

Functional genomics

Response of the genome, such as changes in gene expression, as a consequence of experimental challenge. This most often involves the up- and downregulation of specific genes.

Heterogeneous nuclear RNA (hnRNA)

The primary product of RNA polymerase II transcription in eukaryotic cells. hnRNA alone is processed and matured into mRNA which, in turn, is able to support the synthesis of proteins, though some hnRNA molecules fail to mature and are degraded in the nucleus.

Housekeeping gene

A gene that is expressed, at least theoretically, at a constant level in the cell. The products of these genes are generally required to maintain cellular viability or normal function. Housekeeping genes are often assayed as purportedly invariant controls, compared to the modulation of other genes in response to experimental challenge. Almost all known housekeeping genes show varying levels of gene expression under specific circumstances, so there is no single all-purpose housekeeping gene.

Hybridization

The formation of hydrogen bonds between two complementary nucleic acid molecules. The specificity of hybridization is a direct function of the stringency of the system in which the hybridization is being conducted.

Messenger RNA (mRNA)

The mature product of RNA polymerase II transcription. mRNA is derived from heterogeneous nuclear RNA (hnRNA) and, in conjunction with the protein translation apparatus, is capable of directing the synthesis of the encoded polypeptide.

Noncoding RNA (ncRNA)

A diverse population of transcripts in the cell that do not encode proteins or polypeptides. Certain classes of noncoding RNAs have been shown to profoundly regulate the expression of other genes.

Northern blot analysis

A technique for transferring RNA from an agarose gel matrix, after electrophoresis, onto a filter paper for subsequent immobilization and hybridization. The information gained from Northern blot analysis is used to assess, both qualitatively and quantitatively, the expression of specific genes, though much more sensitive methods are available.

Nuclear runoff assay

A method for labeling nascent RNA molecules in isolated nuclei. The rate at which specific RNAs are being transcribed can then be assayed based upon the degree of label incorporation. See Steady-state RNA for comparison.

Nuclease protection assay

A method for mapping and/or quantifying RNA transcripts. In general, hybridization between probe and target RNA takes place in solution, followed by nuclease digestion (with S1 nuclease or RNase) of all molecules or parts thereof which do not actually participate in duplex formation. Nucleic acid molecules which are locked up in a double-stranded configuration are relatively safe or protected from nuclease degradation. The undigested RNA : RNA or RNA : DNA hybrids are then precipitated and/or electrophoresed for quantification.

Poly(A)+ tail

A tract of up to 250 adenosine residues enzymatically added to the 3′ terminus of mRNA by the nuclear enzyme poly(A) polymerase. The addition of a poly(A) tail involves cleavage of the primary transcript, followed by polyadenylation. Most (but not all) eukaryotic mRNAs exhibit this structure which stabilizes their 3′ terminus. The poly(A) tract is commonly targeted by oligo(dT) for selection of these transcripts, as well as for priming the synthesis of first-strand cDNA.

Polymerase chain reaction (PCR)

Primer-mediated, enzymatic amplification of specific cDNA or genomic DNA sequences. This technology revolutionized molecular biology in the early and mid-1990s; it is the best known and perhaps most widely used molecular biology technique.

Primer

An artificially synthesized, short single-stranded nucleic acid molecule that can base-pair with a complementary sequence and which provides a free 3′-OH for any of a variety of primer extension-related reactions, especially PCR.

Probe

A DNA or RNA molecule which carries a label allowing it to be localized and quantified throughout an experiment. Probes are used most often to hybridize to complementary sequences present among a plethora of different molecules in a nucleic acid sample, as in Northern analysis, Southern analysis, nuclease protection analyses, or DNA library screening.

Proteome

The full complement of proteins produced by a cell at a particular time. Proteome maps are typically generated and assessed by two-dimensional electrophoresis and other techniques designed to identify, quantify, and characterize the products of translation.

Real-time PCR

A state-of-the art method for measuring PCR product accumulation as it is produced in each cycle, rather than measuring the final product mass at the end of the reaction (end-point PCR). Real-time PCR is widely regarded as the premier quantitative molecular biology technique and, as such, is often referred to as quantitative PCR (qPCR).

Relative abundance

The quantity of a particular RNA transcript relative to some other transcript in the same sample, or relative to the amount of the same transcript in other experimentally related samples. This determination is most often made using PCR-based analysis, though less quantitative, non-PCR assays may also be used.

Ribonuclease (RNase)

A family of resilient enzymes which rapidly degrade RNA molecules. The control of ribonuclease activity is a key consideration in all manipulations involving RNA.

Ribonucleic acid (RNA)

A polymer of ribonucleoside monophosphates, synthesized by an RNA polymerase. RNA is the product of transcription.

RNA interference (RNAi)

A novel method by which specific mRNA transcripts can be transiently prevented from participating in translation, or which are destroyed altogether through the formation of a dsRNA molecule. RNAi is loss-of-function approach used to determine the role of a specific gene; it is also known as post-transcriptional gene silencing.

Reverse transcription PCR (RT-PCR)

The technology by which RNA molecules are converted into their complementary DNA (cDNA) sequences by any one of several reverse transcriptases, followed by the amplification of the newly synthesized cDNA using PCR. Not to be confused with real-time PCR, which may or may not involve the use of RNA.

Steady-state RNA

The final accumulation of RNA in the cell. For example, measurement of the prevalence of a particular species of mRNA in a sample does not necessarily correlate with the rate of transcription or RNA degradation in the cell (see Nuclear runoff assay).

Transcription

The process by which RNA molecules are synthesized from a DNA template.

Transcriptome

The complete set of RNA molecules produced by a particular cell under a particular set of circumstances.

Cellular biochemistry is reflected in the abundance of cellular RNA species which, inevitably, drives the phenotype of the cell. In order to understand more readily the cellular response to experimental or environmental challenges, various subpopulations of RNA are harvested and characterized to gain insight to differential expression of genes, and possibly also the subcellular level at which these genes are modulated. RNA is isolated to answer transcription questions by measuring the prevalence of one or more RNA species. The observed changes in transcript abundance may then be related to morphological or physiological differences in the cells or tissues under investigation. The expedient isolation of high-quality RNA is essential to support all downstream applications, and the methods to be used are dictated by the nature of the biological source material. The RNA methodologies are diverse, with each providing a glimpse of some aspect of gene regulation with a characteristic level of sensitivity. Each technique has both advantages and limitations, often requiring a combination of RNA-based assays to provide a more complete picture of the upregulation and downregulation of specific genes and gene families. Data from transcription-based assays are often complemented by quantifying the cognate protein(s), the levels of which often—but not always—correlate. Most investigators use RNA, rather than protein, as a parameter of gene expression because RNA is often easier to isolate than proteins, and because very rare transcripts can be detected via cDNA synthesis and PCR amplification. Presently, there is no such powerful amplification method for proteins.

1 Introduction

The isolation and characterization of ribonucleic acid (RNA) from cells and tissue samples is a central and recurrent theme in molecular biology. In particular, the purification of chemically stable and biologically functional RNA is the starting point for the systematic evaluation of cellular biochemistry by standard molecular methods, including all forms of reverse transcription polymerase chain reaction (RT-PCR), as well as time-honored methods such as Northern analysis, nuclease protection (S1 and ribonuclease (RNase) protection assays), nuclear runoff assay, complementary DNA (cDNA) library construction, and even dot-blot analysis. Messenger RNA (mRNA) abundance is a useful parameter of gene expression; therefore, the expedient recovery of RNA from a biological source is a critical first step for the derivation of meaningful data. Difficulties in the purification, handling, and storage of RNA are intrinsic to the labile chemical nature of these molecules. These difficulties are further compounded by the aggressive character of resilient RNases, the apparent ubiquity of which is undisputed. Indeed, the novice quickly learns of the absolute requirement for management of RNase activity at each level of RNA isolation and characterization. Failure to do so will almost certainly compromise the integrity of the resulting RNA and its probable utility in various downstream applications.

2 Subpopulations of RNA

Prior to the onset of cellular disruption, the investigator must determine which RNA subpopulation is of experimental interest. For example, the precise questions being asked of a particular set of experiments may require characterization of the total cellular RNA, the cytoplasmic RNA alone, nuclear RNA alone, poly(A)+ RNA, or even noncoding RNA species. Transcriptional activity is generally assayed using one of the methods described below, such as Northern analysis, and the data are then validated using another method, such as nuclease protection or RT-PCR. The variegated RNA classifications are delineated in Tab. 1.

Tab. 1 RNA types and functions.

2.1 Messenger RNA (mRNA)

mRNA molecules are destined to serve as templates for protein synthesis via the action of the translation apparatus in the cell. In eukaryotes, the overwhelming majority of mRNA transcripts are characterized by the presence of a tract of adenosine nucleotides known as the poly (A) tail, and all mRNAs so-endowed are known collectively as poly(A)+ RNA. As needed, these molecules can be purified from previously isolated cellular RNA, cytoplasmic RNA, or directly from a whole-cell lysate by using affinity chromatography. For this, oligo(dT)12–18 linked to one of several popular matrices, including paramagnetic beads, biotin, cellulose beads or microcrystalline cellulose, is used to sequester those transcripts that are polyadenylated. The perceived enrichment is often used to increase the ability to detect very low-abundance transcripts. It is important to note, however, that transcript enrichment performed to increase sensitivity may actually be counterproductive in some cases, because the loss of some mRNA during the enrichment procedure may serve only to further under-represent very low-abundance mRNA. Due in no small measure to the power of the polymerase chain reaction (PCR), and the clever design of the required primers, poly(A)+ selection is viewed by many investigators as unnecessary for most contemporary applications.

Poly(A)− RNA is that subpopulation of RNA lacking the tract of adenosine residues at the 3′ terminus; it includes a small number of mRNA molecules, a noteworthy example of which are the histone mRNAs. The predominant members of this class, however, include ribosomal RNA (rRNA), transfer RNA (tRNA), and other noncoding transcripts. In instances where poly(A)− mRNA might not be detected due to exclusion from a sample, either the poly(A)− fraction or a sample of total RNA from the same biological source will contain these naturally nonadenylated transcripts for assay, assuming their respective genes are transcriptionally active. Moreover, the depletion of poly(A)+ mRNA from a sample renders the resulting poly(A)− fraction an excellent negative control in the assay of poly(A)+ mRNA species. For all of these reasons, it should be noted that the terms poly(A)+ mRNA and mRNA are not always synonymous. Finally, mRNAs in eukaryotic cells exhibit an unusual 5′ → 5′ linkage between the first two nucleotides, known as the 5′ cap. This structure not only stabilizes the 5′ end of the transcript but it also efficiently identifies mRNAs as candidates for translation, as these caps are found on mRNAs only, and not on other types of transcripts.

2.2 Transfer RNA (tRNA)

tRNA transcripts are small (74–95 nt) molecules with the responsibility of shuttling amino acids from the cytosol to the aminoacyl site of the ribosome, in order to support the process of translation. These tRNAs are not consumed during this process but are simply returned to the cytosol in order to acquire and transport additional amino acid molecules. The cognate amino acid that specific tRNA species will transport is encoded in its anticodon. Although tRNAs are single-stranded molecules, they fold into a characteristic three-dimensional (3-D) clover-leaf shape, and are immediately recognizable.

2.3 Ribosomal RNA (rRNA)

rRNA transcripts form the backbones of the large and small ribosomal subunits. Depending on the organism, as many as 80 or more proteins decorate the rRNAs in order to form functional protein-synthesis factories. In prokaryotes, the small and large ribosomal subunits are known as the 30S and 50S, respectively, and their eukaryotic counterparts are known as the 40S subunits and 60S subunits, where S represents the Svedberg unit, which is a sedimentation coefficient.

In the cell, the ribosome subunits are dissociated until just prior to the initiation of translation but, upon the completion of translation the ribosome again separates into its constituent subunits. rRNA is the most abundant type of transcript in the cell, often contributing up to 80% of the total RNA. As such, the major rRNAs species are useful as molecular weight standards for RNA electrophoresis, as indicated in Tab. 2.

Tab. 2 Comparison of the traditional Northern analysis, nuclease protection assay, nuclear runoff assay, and RT-PCR.

2.4 Nuclear RNA

Nuclear RNA is often studied in conjunction with the independent characterization of cytoplasmic RNA as a means of assessing the level (transcriptional versus post-transcriptional) and the degree of regulation of various genes. It is well documented that a large mass of transcribed RNA is degraded in the nucleus; this precursor RNA never matures into mRNA capable of supporting translation in the eukaryotic cytoplasm. By comparing the nuclear abundance and cytoplasmic abundance of a particular RNA, a cause–effect relationship may be discerned between an experimental manipulation and the regulation of gene expression in that system with respect to RNA biogenesis, because heterogeneous nuclear RNA (hnRNA), produced by the action of the enzyme RNA polymerase II, matures into mRNA. The analysis of nuclear RNA may also be performed in order to determine the rate at which genes are transcribed (e.g., in the nuclear runoff assay; see below), as opposed to the assay of steady-state RNA levels; these data can then be used to assess the level of regulation of gene expression.

Small nuclear RNAs (snRNAs) represent another class of nuclear RNA. These molecules typically exist as the RNA–protein complexes, known as U1, U2, U4, U5, and U6, and are confined to the nucleus where they are generically referred to as small nuclear ribonucleoproteins (snRNPs, or snurps). snRNPs are now known to form enormous complexes referred to as spliceosomes; these have responsibility for the removal of noncoding intron sequences found in hnRNA and concomitant exon ligation during mRNA biogenesis. Yet another class of small nucleolar RNAs (snoRNAs) is associated with rRNA biogenesis in the nucleolar region, where transcription of the rRNA genes occurs.

2.5 Organellar RNA

Both mitochondria and chloroplasts have their own circular chromosomes (mitochondrial DNA, mtDNA and chloroplast DNA, ctDNA, respectively) which are inherited independently of nuclear chromatin, and in a non-Mendelian manner. These unique genomes encode proteins that remain in the organelle, although mitochondria and chloroplasts each import proteins encoded by nuclear genes to support normal organellar function. In contrast to cytoplasmic mRNAs, neither mitochondrial nor chloroplast mRNAs exhibit a 5′ cap structure. Most mitochondrial transcripts exhibit a 3′ relatively short poly(A) tail, while most chloroplast mRNAs are not polyadenylated. Mitochondrial mRNAs often possess unusual AUA and AUU translation start codons, rather than AUG. These start codons are usually observed very close to the 5′ terminus, although there is considerable variation from one cell type to the next.

2.6 Noncoding RNA

Noncoding RNA refers to a population of small transcripts that do not encode proteins but, interestingly, are often intimately involved in the regulation of protein synthesis. This RNA category includes an abundant group of small cytoplasmic RNAs (scRNAs) found in the eukaryotic cytoplasm and, technically, also the well-known rRNA and tRNA species described above. The small cytoplasmic transcripts are known to exist as RNA–protein complexes (scRNP, or scyrps), and to have a role in regulating the synthesis, sorting, and secretion of proteins, as well as possible mRNA degradation. Of greatest contemporary interest to the molecular biologist are the microRNAs (miRNAs), which function as noncoding antisense regulators of protein synthesis. The formation of double-stranded RNA (dsRNA) structures via miRNA : mRNA base-pairing (either perfectly or with a mismatch) most commonly occurs near the 3′ end of the cognate transcript, and is able transiently to block the translation of that mRNA, or to direct its destruction altogether.

3 Goals in the Purification of RNA

Concise and thoughtful planning prior to beginning laboratory investigations is an absolute requirement for the recovery of high-quality RNA that is capable of supporting biochemical analyses. During the preliminary stages, an experimental design for the purification of nucleic acids must in general address five specific goals (adapted, in part, from Ref. [1]), the successful achievement of which will have a profound influence on the yield, quality, and utility of the sample.

3.1 Goal 1: Select an Appropriate Method for Membrane Solubilization

The first decision to be factored into an RNA isolation strategy is based on which population of RNA or subcellular compartment is to be studied. For example, the aim might be to determine whether an observed modulation of gene expression in a model system is regulated transcriptionally, or by certain post-transcriptional event(s). In such an instance, the methods selected for cellular disruption and subsequent RNA isolation must permit the analysis of salient nuclear transcripts independently of those localized in the cytoplasm.

The method of cell lysis will determine the extent of subcellular disruption in a sample, and is a direct function of the lysis buffer. For example, a lysis buffer that is used successfully with tissue culture cells may be entirely inappropriate for whole-tissue samples due to the presence of a cell wall (in the case of plants and yeast) or tenacious proteins found in the extracellular matrix (in animal tissues). The method by which membrane solubilization is accomplished will also dictate which additional steps will be required to remove DNA and protein from the RNA preparation, and whether compartmentalized nuclear RNA and cytoplasmic RNA species can be purified independently of one another. While DNA can be purged from an RNA preparation with minimal fanfare, it is not possible to determine the relative contribution of transcripts from the nucleus and from the cytoplasm, once the RNAs from these two subcellular compartments have mingled and copurified. A particular lysis procedure must likewise demonstrate compatibility with ensuing protocols. The main lesson is always to think two steps ahead: the correct method of solubilization is dependent on the plans for the RNA after purification, and the questions being asked of a particular study.

3.2 Goal 2: Ensure Total Inhibition of Nuclease Activity

The imperative for controlling nuclease activity is non-negotiable. This includes purging RNase from reagents and equipment (extrinsic sources of nuclease activity) and controlling the RNase activity in a cell lysate (intrinsic source of nuclease activity). Whilst harsh lysis buffers inhibit nuclease activity in their own right, gentle lysis buffers often require the addition of nuclease inhibitors to safeguard the RNA during the isolation procedure. Steps for the inhibition or elimination of RNase activity must, first and foremost, demonstrate compatibility with the lysis buffer.

3.3 Goal 3: Remove Contaminating Proteins from the Sample

The complete removal of protein from a cellular lysate is of paramount importance in the isolation of both RNA and DNA. Meticulous attention to this detail is required, both for accurate quantification and precision in hybridization, ligation, or reverse transcription into cDNA. The removal of proteins from nucleic acid samples may be accomplished by:

1. Protein hydrolysis with proteinase K

2. Salting-out of proteins

3. Solubilizing proteins in guanidinium-based buffers

4. Repeated extraction with mixtures of phenol and chloroform

5. Any combination of the above.

RNA molecules are much less fragile than high-molecular-weight DNA, and consequently more aggressive methods can be employed for the removal of proteins, including the use of phenol : chloroform extraction. While deproteinization is in itself a means of controlling RNase activity, purified RNA samples will be once again susceptible to nuclease degradation following removal of the protein denaturant, especially as a consequence of latent RNase contamination.

3.4 Goal 4: Concentrate the Sample

This is the final step in nearly all RNA purification schemes. The most versatile method for concentrating nucleic acids is precipitation, using various combinations of salt and alcohol (the most common method is to add sodium acetate and ethanol). Nucleic acids and the salt that drives their precipitation form complexes which have a greatly reduced solubility in high concentrations of alcohol. Unlike the precipitation of genomic DNA, that of RNA typically requires longer incubation periods, often at −20°C. In addition, when centrifuging samples a greater g-force must be applied in order to completely recover an RNA precipitate for subsequent analysis. Other concentration procedures include the use of commercially available concentrating devices, dialysis, centrifugation under vacuum, and binding to silica column matrices in high-salt. Today, silica-based purification formats are widely used and have all but replaced the salt and alcohol precipitation method. In the column format, the purified RNA can be eluted in as small a volume as a few microliters, thereby ensuring a favorably high concentration of nucleic acid that can be used directly. Care must be taken, however, when handling the RNA at this stage of purification, as it will once again be susceptible to nuclease attack when the residual, strongly denaturing lysis buffer components and deproteination reagents have been removed.

3.5 Goal 5: Select the Correct Storage Conditions for the Purified RNA

Because of the naturally labile character of RNA, the incorrect storage of excellent RNA samples will often result in degradation within a relatively short time. Many proposals have been made as to the correct temperature, buffer, and storage form for RNA but, as a general rule, RNA is most stable as an ethanol precipitate at −80°C. Large samples or RNA stocks should be stored in convenient aliquots in sterile Tris-EDTA buffer (10 mM Tris, pH 7.4; 0.1 mM EDTA) in order to avoid repeated freeze–thaw cycles. Long-term storage in water is not recommended because, over time, the slightly acidic pH environment will favor RNA degradation by acid depurination. Moreover, it is incumbent upon the investigator to ensure that added RNase inhibitors for either long-term or short-term storage will not interfere with any subsequent manipulations and/or reactions involving the RNA.

4 Methods of Cellular Disruption and RNA Recovery

As suggested above, in order to select a suitable method for cellular disruption or solubilization, consideration must be given as to which subpopulation of RNA is desired for study, as well as the nature of the biological material to be used (cells grown in tissue culture versus whole tissues). Beyond cell and tissue disruption, the absolute necessity for the highest purity, and highest quality, RNA cannot be understated. RNA molecules bind a variety of cytoplasmic and nuclear proteins, any one of which is capable of interfering with most downstream applications, including PCR. Consequently, lysis buffers that effectively strip away RNA-binding proteins are strongly preferred.

The removal of protein during RNA recovery from its biological source often begins with an application of the lysis buffer. In other cases, the addition of protein denaturants post-lysis is performed, particularly when organellar integrity must be maintained. In either case, thorough attention to this facet of nucleic acids isolation will minimize any subsequent purity-associated problems. While the details of many lysis buffer formulations have been reported, they may all be classified as being either gentle or harsh.

4.1 Gentle Lysis Buffers

Gentle lysis buffers are used when a specific subpopulation of RNA is desired (e.g., cytoplasmic RNA alone) and nuclear integrity must be maintained, as with the isolation of cytoplasmic RNA. Gentle lysis buffers, which often are slightly hypotonic, frequently contain the nonionic detergent NP-40 (Nonidet P-40; today known as Igepal CA-630). Because osmotic lysis is the least aggressive method of cellular disruption, NP-40 lysis buffers are ideal for solubilization of the plasma membrane alone, while the inclusion of low concentrations of magnesium helps to maintain nuclear integrity [2]. Thus, the nucleus and its contents (DNA and nuclear RNA) can be separated from the cytosol by using differential centrifugation. The resultant supernatant will be rich in cytoplasmic RNA and proteins, with the latter being easily removed by repeated extraction with phenol : chloroform, or using one of the above-described alternatives. If desired, the nuclear pellet may be processed separately for the recovery of nuclear transcripts. This method of cellular disruption is ideally suited to cells harvested from tissue culture; unfortunately, owing to the complex geometry and formidable nature of whole-tissue samples, nonionic lysis buffers are not effective with tissue samples unless they are coupled with limited, nonshearing homogenization (e.g., using a Dounce homogenizer).

The clear advantage of this isolation strategy is that, ultimately, the material recovered is cytoplasmic RNA alone (mRNA, tRNA, and rRNA). A disadvantage, however, is that the lysis buffer is not sufficiently inhibitory toward RNase. When cell lysis occurs, those RNases which normally are sequestered will be liberated, and their activity will greatly compromise the integrity of the RNA, despite the investigator seeking diligently to maintain its purity. At this point it may be helpful to keep the samples on ice at all times (unless the protocol specifically dictates otherwise); it might also help to use reagents and tubes that have been pre-chilled on ice before use. If desired, an exogenous RNase inhibitor such as RNasin® (Promega) can be added to the lysis buffer. Alternatively, hnRNA (nuclear RNA) alone can be isolated using this same gentle lysis buffer which, when used correctly, does not cause nuclear breakage. This facilitates the recovery of intact nuclei that can be washed free from any residual cytoplasmic transcripts.

4.2 Harsh Lysis Buffers

There is probably no better way to deal with seemingly recalcitrant RNases than to disrupt cells in a guanidinium lysis buffer [3]. On contact, guanidinium-containing buffers distort the tertiary folding of RNases, which results in their inactivation. Other chaotropic lysis buffers which contain high concentrations of ionic detergents, such as sodium dodecylsulfate (SDS), have also been described. The inclusion of additional RNase inhibitors to these lysis buffers is not necessary, and such procedures for RNA isolation are usually carried out at room temperature.

In the presence of chaotropic agents, organelle lysis accompanies disruption of the plasma membrane. Thus, nuclear RNA, genomic DNA and mitochondrial DNA will all be copurified with cytoplasmic RNA, such that additional steps will be required to remove the DNA from the sample. In the past, the most prevalent of these methods was isopycnic centrifugation [4], which involved gradient centrifugation using either cesium chloride (CsCl) [5] or cesium trifluoroacetate (CsTFA) [6]. Isopycnic separation of the biomolecules is possible because of their differing buoyant densities (DNA, 1.5–1.7 g ml−1; RNA, 1.7–2.0 g ml−1; protein, 1.1–1.2 g ml−1).

The differential partitioning of DNA, RNA and protein by acid–phenol extraction, which was first described by Chomczynski and Sacchi [7], led to a dramatic change in the way that RNA (in particular) could be purified from cells and tissues. Succinctly, the organic extraction of nucleic acids at acidic pH causes DNA to partition to the interphase and organic phase, while RNA remains in the aqueous phase. This approach precludes the requirement for ultracentrifugation, and thus greatly reduces the required amount of hands-on time, to the obvious benefit of the investigator. The popularity of acid–phenol extraction has resulted in the development of a number of nucleic acid isolation reagents that support the unceremonious purification of RNA from both tissues and tissue cultured cells alike.

In order to take full advantage of the disruptive nature of the guanidinium isolation procedures, whilst maintaining the subcellular compartmentalization of RNA, one worthwhile strategy is to start the isolation procedure with gentle nonionic lysis, followed by the recovery of intact nuclei, which are then lysed with guanidinium buffer. The purification of nuclear (or cytoplasmic) RNA then proceeds as if working with intact cells. This approach is particularly suited to the isolation of nuclear RNA for Northern analysis.

The principal drawback when applying these chaotropic methods to intact cells is the loss of any ability to discriminate between cytoplasmic and nuclear RNA. There is no method by which nuclear RNA can be separated from mRNA once mixing has occurred, although size fractionation may result in a partial separation. Moreover, it is unfortunate that many seasoned investigators begin to show signs of sloppiness with respect to the control of RNase activity when working routinely with guanidinium buffers. Whilst it is true that RNA is safe from nuclease degradation in the presence of these agents, the purified RNA is once again susceptible to nuclease degradation.

4.3 Silica Separation Technology

One of the more important improvements in the area of nucleic acid isolation has been the development of silica filters that are small enough to be used with a standard microcentrifuge. The filters consist of glass microfibers positioned in the bottom of small plastic insert that fits inside a standard 1.5 ml microfuge tube. The filters are widely available, and may be used for the efficient purification of RNA directly from biological sources. They can also be used to clean up nucleic acids after restriction enzyme digestion, ligation reactions, cDNA synthesis, and PCR amplifications. In general, the RNA (or DNA) is bound to silica in a high-salt, chaotropic environment that is produced by diluting a nucleic acid sample in guanidinium thiocyanate. Following a series of washes, the purified material is eluted from the matrix under very low-salt conditions. The main benefit of this procedure is that the nucleic acid purification and clean-up can be performed within a remarkably short time, and using small volumes.

4.4 Affinity Matrices

In addition to the methods described above for the isolation of total cellular RNA or total cytoplasmic RNA, certain products are available which capture polyadenylated transcripts directly. For example, many mRNA isolation kits feature tracts of oligo(dT) that have been linked covalently to a solid support such as cellulose, polystyrene, latex, or paramagnetic beads. The polyadenylated transcripts are then captured through canonical base-pairing between the poly(A) tail and the oligo(dT) tract in a high-salt environment. The main benefit associated with affinity selection is an enrichment of a nucleic acid sample in favor of mRNA by minimizing the carryover of rRNA and tRNA; enrichment in this manner may also increase the sensitivity of an assay. An older variant of affinity selection involved poly(A)+ mRNA being affinity-captured by using a column packed with poly(U) linked to Sepharose beads [8]. Although still available, this process is no longer generally used because of a perception that is a less-efficient matrix, and that the quantitative recovery of RNA from a poly(U) matrix normally requires the use of formamide-based elution buffers.

Yet another variant of the affinity matrix approach is designed to study nucleic acid–protein interactions by passing a heterogeneous protein mixture over a column packed with either RNA or DNA oligonucleotides, in order to capture proteins with some level of binding affinity to the sequences on the column. The nucleic acid is often referred to as the bait, while the proteins that can bind to it are known as the prey. The procedure, which may be referred to as a pull-down method, is still popular for the characterization of RNA- or DNA-binding proteins, despite the advent of glass or plastic arrays (also known as chips) that can be used for proteome profiling.

5 Inhibition of Ribonuclease Activity

The difficulties associated with the isolation of full-length, intrinsically labile RNA are further compounded by ubiquitous RNase activity. The RNases are a family of enzymes which degrade RNA molecules through both endonucleolytic and exonucleolytic activity cleavage. These small, remarkably stable enzymes resist denaturation under harsh conditions such as extremes of pH and autoclaving that would easily destroy the activity of many other enzymes [9]. It is incumbent upon the investigator to ensure that both the equipment and the reagents to be used are purged of nucleases from the onset of an experiment. For most RNA-minded molecular biologists, to say that a reagent or apparatus is sterile is more than likely a statement that it is RNase-free.

The method selected for controlling the RNase activity must, first and foremost, demonstrate compatibility with the cell lysis procedure. Occasionally, nuclease inhibitors are added to gentle lysis buffers when subcellular organelles (nuclei especially) are to be purified intact, as in the partitioning of nuclear RNA from cytoplasmic RNA. However, keeping the reagents and microfuge tubes ice-cold throughout the procedure is also an effective means of controlling nuclease activity. Second, the method of nuclease inhibition must support the integrity of the RNA throughout the subsequent fractionation or purification steps. Third, the reagents used to inhibit the RNase activity must be easily removed from the purified RNA, so as not to interfere with any subsequent manipulations. In all cases—and especially when characterizing a system for the first time—the control of nuclease activity should be aggressive. Failure to do so is likely to yield a useless sample of degraded RNA.

5.1 Preparation of Equipment and Reagents

Rule number one when working with RNA is to wear gloves during the preparation of reagents and equipment, and especially during the actual RNA extraction procedure. Finger greases are notoriously rich in RNase, and are generally accepted as the single greatest source of RNase contamination. There should be no hesitation in changing gloves several times during the course of an RNA-related experiment. Door knobs, micropipettors, computer keyboards, iPods, refrigerator door handles, containers in which chemicals are packaged, and other unassuming surfaces are all potential sources of nuclease contamination.

With respect to laboratory consumables, any plasticware that is certified as being tissue culture-sterile is always preferred when working with RNA. This includes individually wrapped serological pipettes and conical 15 and 50 ml tubes. In any event, these items should be handled only when wearing gloves. Bulk-packed polypropylene products (e.g., microfuge tubes and micropipette tips) are potential sources of nuclease contamination, due mainly to their being handled and distributed with ungloved hands from a single bag. These consumables are best purchased as being certified both DNase- and RNase-free. Any plastic product or other implement that will come into contact with an RNA sample at any time, either directly or indirectly, and which can withstand autoclaving, should be so treated and set aside exclusively for RNA studies.

When the use of glassware is unavoidable (as when using organic reagents such as phenol and chloroform), the use of individually wrapped borosilicate glass pipettes is strongly preferred. Any glassware that must be re-used should be set aside for RNA work, and not allowed to enter general circulation in the laboratory. Contrary to popular belief, the temperature and pressure generated during the autoclaving cycle are usually insufficient to eliminate all RNase activity. Fortunately, however, RNases can be destroyed quite effectively by baking in a dry heat oven; glassware to be used should be rinsed with RNase-free water and then baked for 3–4 h at 200 °C. Baking pertains to glassware alone; any problems regarding the heating of plastics or other materials can usually be resolved by the manufacturers' technical department. Finally, it is vital to pay attention to the expiry dates of all compounds and solutions in the laboratory. Older bottles of stock solutions in particular serve as excellent breeding grounds for microorganisms, which shed their RNase into the solution. The use of such a contaminated stock solution could lead to the obliteration of an entire RNA sample.

5.2 Inhibitors of RNase

Endogenous RNase activity varies tremendously from one biological source to the next, and the degree to which action must be taken to inhibit nuclease activity is a direct function of the cell type. Knowledge of the extent of intrinsic nuclease activity is derived from two principal sources: the salient literature, and personal experience. The method of RNase inhibition is to a great extent a function of the type of lysis buffer. Whereas, nondenaturing, osmotic lysis buffers often include a nuclease inhibitor, strongly denaturing (chaotropic) lysis buffers generally do not. Such chaotropic compounds include guanidinium thiocyanate, guanidinium HCl, sarcosyl, SDS, 8-hydroxyquinoline, CsCl, CsTFA, and/or various formulations of organic solvents.

RNasin® may be used to inhibit nuclease activity and circumvent some of the problems commonly associated with the use of a vanadyl ribonucleoside (VDR) complex, and is compatible with a variety of in vitro reactions. RNasin® inactivates RNase A, RNase B, and RNase C, but not RNase T1, S1 nuclease, nor RNase from Aspergillus. Care must be taken to avoid any strongly denaturing conditions that will cause the uncoupling of RNase − RNasin® complexes and the reactivation of RNase activity. RNasin® is widely used in reverse transcription reactions in order to protect the integrity of the template RNA prior to the synthesis of first-strand cDNA.

At one time, a VDR was a popular addition to nonionic lysis buffers which alone are ineffective for the control of RNase. In the absence of a VDR, the RNase-mediated cleavage of the phosphodiester backbone of RNA results in the transient formation of a dicyclic transition state intermediate that is subsequently opened by reaction with a water molecule. In its capacity as an RNA transition state analog, the VDR complex forms a highly stable dicyclic species to which the enzyme remains irreversibly bound. Thus, nuclease activity is eliminated by locking RNase and pseudo-substrate in the transition state. The VDR binds tightly to a broad spectrum of cellular RNases, including RNase A and RNase T1, but not to RNase H, and is compatible with a variety of cell fractionation methods. It is important that a VDR is used selectively, however, as even trace carry-over quantities are sufficient to inhibit the in vitro translation of purified mRNA. It can also interfere with reverse transcriptase activity, thereby excluding its use with any RT-PCR applications. For this reason, the VDR is no longer used by most molecular biologists as an RNase inhibitor.

Diethyl pyrocarbonate (DEPC), which at one time was used widely to purge RNase from solutions prepared in-house, has also fallen out of favor with molecular biologists. This is due to the widespread availability of certified nuclease-free reagents, including sterile H2O, from virtually all biotech vendors. DEPC is a well-known nonspecific inhibitor of RNase that is used to purge reagents of nuclease activity, due to the unreliability of autoclaving alone. Strict precautions (as indicated by the manufacturer) must be taken when using DEPC, however, as it is carcinogenic and potentially explosive. Clearly, it should be avoided unless there is an absolutely compelling reason for its use.

Hydrogen peroxide (H2O2) is a powerful oxidizing agent that can render common laboratory surfaces nuclease-free by soaking for 20–30 min, followed by rinsing with copious amounts of water that, at the very least, has been autoclaved. The soaking of glass pipettes, gel box casting trays, electrophoresis combs, graduated cylinders, and similar implements in a 3% H2O2 solution is a very effective and inexpensive measure. H2O2 is readily available in pharmacies and similar stores. It is important NOT to use the more concentrated forms of H2O2 (e.g., 30% H2O2) that are commonly available from chemical supply companies since, at this higher concentration H2O2 is extremely dangerous, perhaps causing irreparable damage to acrylic gel box components and other equipment, as well as tissue damage to the investigator. Old solutions of H2O2 must also be avoided, as they may no longer be solutions of H2O2!

Since many RNases manage to renature following removal of the denaturing reagent(s), it is prudent to maintain separate containers of chemicals and stock solutions for exclusive use as RNA reagents. Chemical solids should be weighed out with an RNase-free spatula, while stock solutions should be aliquoted into suitable volumes; any aliquots that have been used must be discarded. While, initially, such actions may seem excessive, they may well preclude the accidental introduction of RNase and facilitate an expedient recovery of high-quality RNA. All laboratories should have established standard operating procedures (SOPs) in place regarding RNA-related studies, and these protocols should be followed meticulously.

6 Methods for the Analysis of RNA

The evaluation of gene expression by the hybridization of RNA is possible in a variety of formats, as is the analysis of DNA. Methods range from the traditional to the contemporary, with each procedure having an applicability under a defined set of experimental conditions, as well as a characteristic level of sensitivity. The relative merits of four such standard methods are listed in Tab. 3.

Tab. 3 Comparison of the traditional northern analysis, nuclease protection assay, nuclear runoff assay, and RT-PCR.

6.1 RT-PCR

The PCR is a primer-mediated, enzymatic method for the quasi-exponential amplification of nucleic acid sequences. This method requires any one of several thermostable DNA polymerases, two short oligonucleotides acting as nucleic primer sequences, a dNTP cocktail, and the appropriate chemistry to support the activity of the enzyme. The primers are designed to base-pair to opposite strands of the DNA template with their respective 3′-OH ends facing each other. This leads to the amplification of that sequence which is framed by the 5′ ends of the respective primers through a series of heating, cooling, and primer extension stages, the mechanics of which are discussed in great detail elsewhere in the Encyclopedia of Molecular Cell Biology and Molecular Medicine (EMCBMM).

RT-PCR is a two-step process. First, high-quality RNA acts as the template for the synthesis of first-strand cDNA with the enzyme reverse transcriptase. The components and mechanics of this reaction are almost identical to any other first-strand cDNA synthesis reaction, an example being the construction of a traditional cDNA library. Second, the products of the first-strand synthesis reaction are then amplified using the PCR. Traditionally, the first-strand synthesis products are added to a second tube which provides all of the cofactors necessary to support the amplification of these products by PCR. A more recently developed method for performing RT-PCR, which is now widely used in clinical and diagnostic laboratories, requires only one enzyme in a single reaction tube format (one-tube RT-PCR). In either case, the newly synthesized cDNA is amplified as would be the DNA from any other source, predicated upon the availability of a set of gene-specific primers. The PCR-amplified cDNA products can then be quantified, or in some other way analyzed, in order to more fully understand some aspect of normal, or abnormal, cell function. RT-PCR is advantageous because the very labile character of RNA does not favor its long-term storage. The synthesis of cDNA provides a template for a DNA polymerase-mediated amplification on an immense scale; only those transcribed RNAs which are purified from the cell can be converted into cDNA. Different tissues—even from the same biological source—will yield different cDNA products, such that cDNA may be best thought of as a permanent biochemical record of the cell. cDNA represents a means by which the molecular physiology of the cell can be studied in great detail over a period of months or years—much longer, and with much greater sensitivity, than the assay of purified RNA directly.

In addition to its obvious utility for the quantification of gene expression, the judicious design of primers permits RT-PCR to be used to map the 5′ and 3′ ends of transcripts—a method known as the rapid amplification of 5′ complementary DNA ends (5′ RACE) [10] and the rapid amplification of 3′ complementary DNA ends (3′ RACE) [11], respectively. RACE is used to detect alternative transcript initiation, splicing, and poly(A)+ polymerization sites, and to identify induced and repressed genes under a defined set of environmental conditions.

Finally, RT-PCR can be performed using two different platforms, namely end-point PCR and real-time PCR; the latter method may also be referred to as the quantitative polymerase chain reaction (qPCR). End-point PCR involves amplifying the template over 25–30 cycles, with a theoretical amplification of 2n-fold, where n is the number of cycles. When all of the cycles have been completed, the reaction tube is opened and the resulting products are analyzed using agarose gel electrophoresis. In this case, the band intensity is associated with product abundance, which in turn mirrors the abundance of the starting material. Both, the mechanics of end-point PCR and the method of detection can limit the sensitivity of end-point PCR. For example, the intensity of two bands representing two vastly different samples may appear identical on electrophoresis when one reaction depletes the primers (the so-called plateau effect) after 15 cycles, and the other reaction depletes the primers after 30 cycles.

Real-time PCR is widely regarded as the gold standard with respect to nucleic acid detection sensitivity. In the real-time format, the accumulation of product in the reaction vessel is measured at the end of every cycle—that is, in real-time. As the PCR product accumulates, however, there will be a directly proportional increase in fluorescence, due to the inclusion of fluorescent precursors in the reaction chemistry. The fluorescence detection system permits an extremely early detection in the amplification process, while the accumulation of product is reliably exponential. With each passing cycle, however, inefficiencies in the reaction itself compromise the amplification efficiency of subsequent cycles. As a consequence, the true abundance relationships among genes and among samples may be distorted, or even lost altogether, by waiting until the end of all cycles before the products are analyzed. Moreover, the fact that real-time quantification occurs in a sealed tube that is not opened at all greatly minimizes the risk of carry-over contamination—an unfortunate occurrence where the product from one PCR experiment inadvertently becomes the template in a subsequent experiment.

It is also important to note that, following recovery from the cell, intramolecular base-pairing that results in secondary and tertiary RNA structures is problematic. Molecules in which higher-level structures form are often resistant to reverse transcription, which thereby diminishes their ability to be quantified or otherwise assayed. This issue is often addressed by heat denaturation in the presence of one or more compounds that interfere with hydrogen bonding, and is performed prior to reverse transcription. Further, performing the first-strand cDNA synthesis reaction at elevated temperatures also helps to reduce any intramolecular base-pairing; this is possible because of the availability of thermostable reverse transcriptases.

6.2 Northern Analysis

The quintessential method for the assay of gene expression is a method referred to as Northern analysis [12] (it is also known colloquially as Northern blotting, the Northern blot analysis, and/or RNA blot analysis. Northern analysis involves the electrophoretic separation of RNA molecules under denaturing conditions, with subsequent transfer or blotting of the sample onto a solid filter support (the so-called filter membrane). The RNA on the blot is then hybridized to an appropriately labeled nucleic acid probe which will support subsequent detection by autoradiography, or by chemiluminescence. Because the samples of RNA undergo electrophoresis prior to their hybridization, the Northern analysis provides both quantitative and qualitative biochemical profiles of the sample. Denaturation of the RNA prior to electrophoresis is necessary to ensure that the migration of the sample through the gel occurs only with respect to molecular weight, and is not distorted by the formation of any secondary structure that is commonly associated with single-stranded molecules. Thus, the length of the transcript(s) can be determined—a datum that cannot be discerned using other methods.

The objective of the Northern analysis is to quantify gene expression by detecting the relative abundance of those mRNAs in the sample which are of immediate interest to the investigator. Whereas, in the Southern analysis [13] the resulting data pertains to the structure and organization of genes, data derived by Northern analysis reflects the transcriptional activity of genes.

The principal shortcoming associated with Northern blot data is the limited sensitivity of the assay. The physical application and immobilization of an RNA sample onto a filter membrane renders some of those molecules incapable of base-pairing to a complementary nucleic acid probe. Neither is the Northern analysis intended to discern the absolute mass of RNA in the cell. Rather, such data may be measured far more accurately by using solution hybridization-based methods, especially real-time PCR. Hence, data derived from the Northern analysis must be interpreted in the context of the relative abundance of a particular RNA among all of the samples involved; hence, the method is semi-quantitative at best.

The electrophoresis of RNA is itself an important diagnostic tool, with a host of information being made available regarding the integrity and probable utility of an RNA sample by examining a representative aliquot. RNA has a highly characteristic profile on a denaturing gel (Fig. 1), whereby the appearance of the predominant species—the 28S and 18S rRNAs—being an indicator of the integrity of the sample. Ideally, a very light smearing above, between, and just barely below the rRNAs indicates that sample is intact and is probably capable of supporting nucleic acid hybridization. Heavier smearing, especially below the level of the 18S rRNA is quite ominous, being indicative of partially or fully degraded RNA (Fig. 2). The complete absence of the rRNAs indicates a completely degraded sample. As it is clearly desirable to ascertain the integrity of a sample before moving on to sophisticated and often time-consuming techniques, a brief period of electrophoresis to assess the quality of the sample should become a standard procedure in any molecular biology setting.

Fig. 1 Assessment of RNA quality. The sharp definition of the 28S and 18S rRNA species in lanes a and b demonstrates the integrity of the sample. RNA in lanes c and d is also high quality, although an excessive amount of RNA was applied to these lanes. Lanes a and b: 20 µg of total cytoplasmic RNA prepared by NP-40 lysis. Lanes c and d: 25 µg of total cellular RNA (nuclear and cytoplasmic), prepared by guanidinium–acid–phenol extraction. Note the higher molecular weight nuclear RNA species in the sample.

Reproduced with permission from Farrell Jr, R.E. (1993) RNA Methodologies: A Laboratory Guide for Isolation and Characterization, Academic Press, San Diego, CA, p. 60).

Fig. 2 Going, going, gone…degraded RNA. A representative aliquot from four different samples of human fibroblast RNA was electrophoresed in a 1.2% agarose-formaldehyde gel and then stained with ethidium bromide. The RNA molecular weight standard is visible in lane 1. RNA in lanes 2–5 shows increasing degrees of degradation, most likely due to RNase contamination during the isolation procedure. Especially noteworthy is the complete absence of the 28S and 18S rRNA species expected in high-quality RNA. This is an excellent example of what not to do.

Reproduced from with permission from Farrell Jr, R.E. (2010) RNA Methodologies: A Laboratory Guide for Isolation and Characterization, 4th edn, Elsevier, Academic Press, San Diego, CA, p. 149) [1].

6.3 Nuclease Protection Assay

The intrinsic shortcomings of the Northern analysis mandate a different format for the assay of gene expression when very exacting quantitative data are required. In contrast to the assay format of the Northern analysis, at the heart of an assay by nuclease protection is a high stringency hybridization between the target and probe molecules, both of which are free-floating in solution (solution hybridization) as opposed to having the target mRNA fixed on the filter paper (mixed-phase hybridization). The driving forces behind solution hybridization are the random molecular collisions, the kinetics of which are related directly to the total mass of nucleic acid in the reaction tube (probe + target + carrier = total mass). Because of the solution hybridization format, all complementary nucleic acid molecules are presumed to be capable of hybridization. The S1 nuclease protection assay (Fig. 3) and the RNase protection assay (Fig. 4) are methods of greatly enhanced sensitivity and resolution, and are universally considered to be more quantitative than Northern analysis.

Fig. 3 S1 nuclease assay for the quantification of specific RNA species. Purified RNA is hybridized in solution with a labeled probe sequence to form thermodynamically stable hybrid molecules. Any RNA or probe molecules that do not participate in the formation of hybrid molecules are digested away by the single-strand-specific nuclease S1, followed by electrophoresis of the intact hybrid molecules. The size and abundance of protected RNAs are then deduced by autoradiography, performed directly from the gel. Lane 1: undigested probe; lanes 2 and 3: experimental samples; lane 4: molecular weight standards.

Reproduced with permission from Farrell Jr, R.E. (2010) RNA Methodologies: A Laboratory Guide for Isolation and Characterization, 4th edn, Elsevier, Academic Press, San Diego, CA, p. 323) [1].

Fig. 4 RNase protection assay for the quantification of specific RNA species. Purified RNA is hybridized in solution with a labeled antisense probe sequence to form thermodynamically stable double-stranded RNA molecules. Any RNA or probe molecules that remain single stranded are digested by an RNase cocktail. Following electrophoresis, the size and abundance of protected RNAs are then deduced by autoradiography, performed directly from the gel. Lane 1: undigested probe; lanes 2 and 3: experimental samples; lane 4: molecular weight standards. The general approach is identical to that for the S1 nuclease assay.

Reproduced with permission from Farrell Jr, R.E. (2010) RNA Methodologies: A Laboratory Guide for Isolation and Characterization, 4th edn, Elsevier, Academic Press, San Diego, CA, p. 324) [1].

The best nucleic acid probes for these assays are substantially shorter than the target mRNA. Upon molecular hybridization, a short double-stranded region is generated, while the 5′ and 3′ regions of the target molecule flanking the double-stranded area remain single-stranded. The enzyme S1 nuclease, or a combination of RNases, is then used to digest all of the nucleic acid molecules that did not participate in nucleic acid hybridization. Only double-stranded nucleic acid molecules are resistant to nuclease attack. The resulting product of this assay—the so-called protected fragment—is then resolved by electrophoresis. By virtue of the mechanics of this assay, the size of the protected fragment is expected to be similar to the size of the probe sequence itself, which is often substantially shorter than the native RNA target, and can be visualized by using autoradiography. As a direct result of solution hybridization and the digestion of all nonhybridized nucleic acid molecules, the investigator can expect an at least 10-fold enhancement in sensitivity, compared to Northern analysis, particularly when performed using antisense RNA probes.

6.4 Transcription Rate Assays

The modulation of key regulatory molecules is an integral cellular response to both intracellular and extracellular challenge. One fundamental goal in the assessment of any biological model system is an elucidation of the level of gene modulation. While potential levels of regulation are infinite, they are broadly categorized as transcriptional or due to some post-transcriptional event. The initial characterization of these systems commonly involves the isolation, hybridization and subsequent detection of specific RNA species by RT-PCR, nuclease protection analysis, or even Northern analysis. While these approaches may provide reliable qualitative and quantitative data with respect to steady-state levels of message, RNA prepared by total cellular lysis does not provide information about the rate of transcription, the subcellular compartmentalization (nuclear or cytoplasmic) of the RNA under investigation, or the translatability of the RNA in the cytoplasm. Knowledge of these aspects of gene expression is necessary to elucidate the level of gene regulation, because the half-lives among RNA species are variable and because the half-life of many mRNA species can be modified in response to a particular xenobiotic regimen or environmental stimulus.

In order to address these questions, two basic approaches have been employed to study the mechanism of transcription and the processing of the resulting transcripts in eukaryotic cells. In one approach, the rate of transcription is measured in intact nuclei by the incorporation of labeled precursor nucleotides into RNA transcripts initiated on endogenous chromatin at the time of nuclear isolation. Elongated, labeled nuclear RNA is then purified for hybridization to complementary, membrane-bound DNA sequences. This technique, which is known as the nuclear runoff assay (Fig. 5), is a superbly sensitive method for measuring transcription rate as a function of cell state [14, 15], and consequently is widely used. Because it is the RNA transcripts, rather than the probes used to quantify their abundance, that are radiolabeled the basic format of this assay can be likened to a reverse dot-blot, as the probe is membrane-bound and nonradiolabeled.

Fig. 5 Nuclear runoff assay. The relative rate of transcription of all genes can be assessed by incubation of intact nuclei with an NTP cocktail containing labeled UTP. Elongated, radiolabeled transcripts are then hybridized to nonradioactive cDNA probes immobilized on a nylon filter. On autoradiography, the intensity of the signal from each dot is indicative of the degree of label incorporation, and thus the relative rate of transcription of specific genes under a define set of experimental conditions.

Reproduced with permission from Farrell Jr, R.E. (2010) RNA Methodologies: A Laboratory Guide for Isolation and Characterization, 4th edn, Elsevier, Academic Press, San Diego, CA, p. 343) [1].

The principal advantage of the nuclear runoff assay is that labeling occurs whilst maintaining the natural geometry of the transcription apparatus. The mechanics and reaction conditions of the assay promote the elongation of initiated transcripts, but are not believed to support new initiation events. The degree of labeling of any particular RNA species, which is indicative of the relative transcription rate of a specific gene, may then be assessed by liquid scintillation counting (a specific type of radioactive detection), coupled with autoradiography. These data correlate directly with the number of RNA polymerase molecules engaged in transcribing a specific gene, and indirectly with the transcriptional efficiency of regulatory sequences associated with the gene under a defined set of experimental conditions. When used in conjunction with a steady-state analysis of cytoplasmic RNA species, data from the nuclear runoff