Title: Real‐Time Reverse Transcription Polymerase Chain Reaction to Measure mRNA: Use, Limitations, and Presentation of Results
Abstract: Quantitative polymerase chain reaction (PCR) was originally developed for DNA quantitation. There are various applications for quantitative PCR, and it is now increasingly used for measurement of messenger RNA (mRNA) levels (VanGuilder et al., 2008). Quantitative PCR became well known with the introduction of real-time quantitative PCR methods. The focus of this article is the use of quantitative real-time reverse transcription PCR for the measurement of mRNA. Real-time PCR is a technique based on PCR, with the key feature that the amplified product is detected as the reaction progresses in “real-time.” Real-time PCR is combined with the method of reverse transcription to quantify mRNA, which is often used to infer gene expression. There is a lack of consensus on the best method for performing and interpreting real-time reverse transcription PCR experiments for measurement of mRNA. This review focuses on clarifying the principles behind the real-time reverse transcription PCR, indicates what data should be used for analysis and briefly suggests what details of the method should be presented in publications. Real-time PCR, also called “quantitative real-time PCR,” has a number of different abbreviations used in the literature. This can be confusing to the reader. Different abbreviations include RTQ-PCR, Q-PCR, qRT-PCR, or qPCR. The abbreviation “RT-PCR” should denote “reverse transcription PCR” and NOT “real-time PCR” (see below for the difference between the two methods); but this is not always the case, which can add further confusion to the nomenclature. When real-time PCR is combined with the method of reverse transcription to quantify mRNA, the suggested preferred abbreviation is “real-time RT-PCR”, which will be used throughout this article. Reverse transcription PCR is an older method used for detection of low-abundance mRNA. An RNA strand is first reverse transcribed into its DNA complement (cDNA) and the resulting cDNA is amplified using traditional PCR. However, conventional RT-PCR is a time-consuming technique that has limitations compared with the real-time RT-PCR method (Mackay et al., 2002). In RT-PCR, the product is often detected using ethidium bromide, which has a low sensitivity, and results are not always reliable. In addition, results are variable because they are measured at the plateau phase (as opposed to the exponential phase) of the reaction. There is also an increased risk of sample cross-contamination because of post-amplification processing. One of the biggest disadvantages of RT-PCR is that it is a semi- or a low-quantitative technique, where the amplified product (amplicon) can be visualized only after the amplification ends. Real-time RT-PCR is an extremely sensitive cost-effective method. In contrast to the conventional RT-PCR, the amplicons can be visualized as the amplification progresses by using a fluorescent reporter molecule (see below). Other advantages include the following: there is an increased dynamic range of detection, no post PCR processing is required, detection is capable of a twofold change, and data are collected during exponential amplification of the PCR. Three other methods are commonly used to measure quantification of mRNA transcription. These include the gold standard of northern blotting (which is not as sensitive as real-time RT-PCR), in situ hybridization and RNAse protection assays (Bustin, 2000). All of these procedures are lengthy and involved and the amount of RNA required can be large. Real-time reactions are performed in a thermacycler that measures a fluorescent detector molecule. This is achieved by the use of one of the fluorescent-based technologies as follows: (1) probe sequences that fluoresce upon hydrolysis (TaqMan; Applied Biosystems) or hybridization (LightCycler; Roche Diagnostics Corporation), (2) fluorescent hairpins, and (3) intercalating dyes (SYBR green; Applied Biosystems). Errors in the quantification of mRNA transcripts are compounded by variation in the amount of starting material between samples. This is particularly the case when samples have been obtained from different individuals. Therefore, what is an appropriate standard? The current method for minimizing errors and correcting for sample variation is to simultaneously amplify, along with the target gene, a gene that serves as an internal reference to normalize against target RNA values (Bustin, 2000). This reference RNA should have the following qualities: (1) it is expressed at a constant level among different tissues; (2) it is not affected by the experimental treatment; (3) it is expressed at a similar level as the RNA being studied; and (4) it undergoes all steps of the reaction with the same kinetics as the target gene. In selecting a reference gene, it is also important to note that that there can be wide variation in the gene's level between individuals. For example, glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) is frequently used as a reference gene because its expression is often constant at different times and after experimental manipulation. However, GAPDH can significantly vary under certain conditions including between different individuals (Bustin, 2000), during pregnancy (Cale et al., 1997) and during the cell cycle (Mansur et al., 1993). Other commonly used RNAs to normalize gene transcript levels include β-actin and 18S, as well as histone H3 and cyclophilin. It is also possible to use multiple reference genes to minimize the impact of treatment effects, especially when multiple target genes are assessed (Udvardi et al., 2008). Reference genes are highly specific for a particular experimental model and must be validated (i.e., it meets the above requirements) prior to experimentation. This can be performed through various methods including real-time RT-PCR (Dheda et al., 2004) and low density array technology (Cai et al., 2007). The chosen reference gene is used as an endogenous control, which is used in the 2−ΔΔCT method for relative quantitation as discussed in “Data analysis” below. Real-time RT-PCR focuses on the exponential phase of the PCR reaction because it provides the most precise and accurate data for quantitation. Within the exponential phase, the real-time PCR instrument calculates two values. The “threshold line” is the level of detection at which a reaction reaches a fluorescent intensity above background. The PCR cycle at which the sample reaches this level is called the “cycle threshold” (Ct). The concept of Ct is the basis of accurate and reproducible quantification of mRNA. The more template there is at the beginning of the reaction, the fewer number of cycles it takes to reach the point in which the fluorescent signal is first recorded as being statistically significant above background. When small amounts of RNA are converted to cDNA for amplification, this poses a challenge for accurately quantifying the initial amount of cDNA. For example, a PCR of only 20 cycles amplifies the initial cDNA over a million-fold, which has the potential for introducing large error. When using the 2−ΔΔCT method for data analysis (see below), it is important to use a low cycle number when the reaction is still in the exponential phase, that is, Ct should be set at the earliest cycle possible. There is a lack of consensus on the best method for performing and interpreting real-time PCR experiments. Analysis of real-time RT-PCR data can be either as (1) absolute levels, where RNA transcript levels are determined by comparing results to a standard curve or computation method; or (2) relative levels, which compare mRNA from one sample with another sample. Most analyses use relative methods because they are easier, and in many cases, it is more relevant than the absolute amount. The most common relative method of quantification is the 2−ΔΔCT method (Livak and Schmittgen, 2001; Schmittgen and Livak, 2008), which will be discussed in more detail in this review. This method (VanGuilder et al., 2008) uses an RNA standard curve of the gene of interest to calculate the number of copies. Serial dilutions of a known amount of RNA are performed and subjected to amplification. This method is based on two assumptions as follows. (1) The reaction occurs with 100% efficiency and is one of the reasons for using a low cycle number when the reaction is still in the exponential phase. (2) An endogenous control (reference gene) must be included to normalize the target gene. mRNA transcript levels are calculated as follows (based on a study design of control and treated samples). The Ct values generated from the PCR reaction are put in a spreadsheet. The Ct value for the reference gene is subtracted from the Ct value of the target gene, which gives the ΔCt. It is important to note that the two genes must be amplified with comparable efficiencies for this calculation to be accurate. Unfortunately, many investigators do not address this issue. The ΔCt of the control sample is then subtracted from a treated sample, giving the ΔΔCt. Note that the ΔCt of the control sample is usually a mean value to be subtracted from the ΔCt of each treated sample. The negative value of this subtraction (−ΔΔCT) is used as the exponent of 2 in the equation and represents the fold change of the treated target gene relative to the control. What should you use for statistical analysis? Sometimes investigators use Ct values directly for statistical analysis, but this is mathematically incorrect (VanGuilder et al., 2008). The ΔCt value can be used for statistical analysis, but nonparametric testing should be used. The final 2−ΔΔCT value can be used to graphically represent the data as relative fold changes just for visual purposes. Currently, it is controversial as to how much detail of the real-time PCR method should be presented in publications. The minimum information for publication of quantitative real-time PCR experiments guidelines has been established to help ensure the integrity of the results, promote consistency between laboratories and increase experimental transparency (Bustin et al., 2009). It is advocated that these guidelines should be published in abbreviated form or as online supplementary material. So this leads to the question of just how much detail should you report? Generally, enough detail to allow others to repeat the experiment should be included. However, if the method used has been previously reported, a reference will suffice. However, any new experiment would require the following information: primer/probe sequences, or if commercially available, catalogue numbers, type of fluorescent-based technology (e.g., Taqman, etc), PCR conditions and method of analysis (relative or absolute).