Using End-Point RT-PCR Analyses for Gene Expression Profiling

Reverse transcription-polymerase chain reaction (RT-PCR) based methodology has been widely used to quantify differential gene expression. End-point analysis by agarose gel electrophoresis has proven to be the simplest and most-cost effective method for profiling the relative expression of genes by RT-PCR. However, variability in the quantity or quality of the original RNA samples limits the accuracy of these RT-PCR analyses. The co-amplification of the target gene with an unrelated endogenous gene as a reference minimizes some of this systematic error (1). For example, glyceraldehyde-3-phosphate-dehydrogenase (GAPD) is one of the most commonly used "housekeeping" genes for expression analysis.

Unfortunately, most of the well-known reference or housekeeping genes are expressed much more abundantly than the typical gene of interest. When detecting a housekeeping gene in the same tube as another gene, the signal from the housekeeping gene tends to saturate in an RT-PCR analysis before genes with lower levels of expression can be detected. Also, amplification of the housekeeping gene in the same reaction expends PCR reagents interfering with amplification of the gene of interest. Figure 1A illustrates this point. Despite the use of two different amounts of template, standard GAPD primers yield equally intense and saturated signals. Furthermore, the signals from the genes of interest appear much weaker than the GAPD signal suggesting inhibition of amplification.

Unfortunately, most of the well-known reference or housekeeping genes are expressed much more abundantly than the typical gene of interest. When detecting a housekeeping gene in the same tube as another gene, the signal from the housekeeping gene tends to saturate in an RT-PCR analysis before genes with lower levels of expression can be detected. Also, amplification of the housekeeping gene in the same reaction expends PCR reagents interfering with amplification of the gene of interest. Figure 1A illustrates this point. Despite the use of two different amounts of template, standard GAPD primers yield equally intense and saturated signals. Furthermore, the signals from the genes of interest appear much weaker than the GAPD signal suggesting inhibition of amplification.

Under such conditions, the reliability of determining the relative amount gene of interest is called into question because the housekeeping signal cannot be used as a normalization factor. The results cannot be corrected for systematic errors yielding inaccurate relative gene expression profiles. Other studies have also demonstrated that using these genes to normalize data may actually increase error and make quantification more unreliable (2). Several tips have been offered to assist some specialized but not all applications (3, 4).

Figure 1: The Use of a Specially Designed Primer Mix Attenuates RT-PCR Amplification of GAPD. Two different amounts of a ReactionReady™ First Strand cDNA Synthesis Kit (C-01) reaction (1 µl in lane 1 and 3, 2 µl in lanes 2 and 4) were used as template for PCR to co-amplify either the human MX1 (lanes 1 and 2) or the human IRF7 gene (lanes 3 and 4) and GAPD in the same tube. Standard GAPD primers were used in Panel A, while the specially formulated Human GAPD Internal Normalizer (HN-01) primer mix was used in Panel B. The target genes yield the larger amplicons, while the GAPD primers and Internal Normalizer yield the smaller amplicons.

How the Internal Normalizer Can Help

The Internal Normalizer technology from SuperArray Bioscience makes it feasible to use the housekeeping gene GAPD as a universal reference for both the quality and quantity of your RNA sample. The specially modified primer mix attenuates the GAPD signal placing it in the same detection range as many other genes and allowing its accurate detection in the same tube as your gene of interest. Figure 1B shows that the GAPD Internal Normalizer provides a signal that now responds to different amounts of template because the signal no longer reaches saturation. Furthermore, use of the GAPD Internal Normalizer yields stronger signals from the genes of interest indicating that co-amplification of GAPD no longer interferes with their detection. In this way, this special primer mix serves as an effective internal control for systematic errors such as a misjudgment in the amount of template used. As a result, more accurate relative gene expression results can be obtained.

Relative Gene Expression Profiling with the Internal Normalizer

When using the Internal Normalizer to analyze RT-PCR results, determine the band intensity of the gene of interest and GAPD in the same agarose gel lane (that is, from the same PCR tube) by densitometry. Calculate the target-to-GAPD signal intensity ratio to obtain the relative expression of the gene of interest. Then, compare these ratios between different experimental conditions to obtain relative expression values or profiles. To illustrate the application of the Internal Normalizer, we monitored the time-dependent changes in the expression of three different genes upon treatment with TNFalpha. The results in Figure 2 clearly demonstrate that expression of all three of these genes increases quickly and dramatically up to a 4- to 8-fold induction, peaking at around 45 minutes. Expression then more slowly tapers off to a level still 3- to 5-fold elevated relative to untreated cells.

Figure 2: SingleGene™ PCR Kits Containing the GAPD Internal Normalizer Detects Induction of TNFa-Inducible Genes. HeLa cells were treated with 20 ng/ml TNFalpha for various times, and harvested for total RNA isolation. Templates were generated from 150 ng of total RNA with the ReactionReady™ First Strand cDNA Synthesis Kit. The relative expression of the human TNFAIP3, NFKBIA, and IL6 genes were determined using the respective SingleGene™ PCR Kits (Catalog Numbers SPH-0004A, SPH-0111A, and SPH-0503A, respectively) according to the manufacturer's specifications. After 30 PCR cycles, 10 µl of each reaction was analyzed by agarose gel electrophoresis (Panel A). The time-dependent response of the three target genes to TNFa treatment is plotted in B. The target-to-GAPD ratios were all normalized to time zero.