NEW Ambion Silencer® Select siRNAs
TaqMan® Gene Expression Assays

Tips from the Bench: Calculating Gene Expression Changes and Variability in siRNA Experiments

RNA interference (RNAi), using synthetic small interfering RNA (siRNA) to target mRNA transcripts for degradation, has become an effective method of inhibiting gene expression. Scientists rely heavily on quantitative RT-PCR (qRT-PCR) using Applied Biosystems TaqMan Gene Expression Assays to evaluate specific mRNA levels in their samples after treatment with siRNA. These RT-PCR data are most commonly interpreted using the delta delta CT (2^–ΔΔCT) method to quantitate the relative change in gene expression attributable to delivery of the target-specific siRNA. The 2^–ΔΔCT method uses data from both transfection of a negative control siRNA and RT-PCR of an endogenous control mRNA. In this article, Applied Biosystems scientists describe basic experimental design to address technical and biological variability in siRNA experiments. They also discuss ways to interpret siRNA-induced knockdown data obtained using qRT-PCR analysis.

Number of Replicates and Appropriate Controls
In the context of siRNA experiments in cultured cells, biological replicates (BRs) are different cell populations that are transfected with equal amounts of the same siRNA. Data from BRs show the variation in the measured amounts of target mRNA levels in the different populations. Technical replicates (TRs ) are replicate PCRs from reverse transcription of RNA from the same cell population; they provide a representation of the variability of the analysis process.

It is important that siRNA experimental design includes appropriate controls and replicates to provide enough statistical power for discrimination of “real” results from both biological and technical variability. A good rule of thumb is to include three BRs of transfection with each siRNA targeting the gene of interest, three BRs of samples transfected with a negative control siRNA (NC), and three TRs of the real-time PCR analysis of each BR (Figure 1). In order to interpret the real-time PCR data, each sample should be amplified for both the experimental target and an endogenous control gene that is not expected to vary among samples (for example 18S rRNA, cyclophilin, or β-actin).

Figure 1. Experimental Design With Appropriate Biological Replicates (BR) and Technical Replicates (TR). "Experimental" represents wells transfected with target-specific siRNA. "Neg. Control" represents wells transfected with nontargeting negative control siRNA. "Endogenous control" represents either Experimental or Neg. Control samples in which the endogenous qRT-PCR control will be measured (e.g., 18S rRNA). Each BR is used in 3 TRs of each qRT-PCR.

Steps for Data Analysis
Step 1. Calculate the mean C_T of the technical replicates.
Step 2. Calculate the standard deviation and coefficient of variation for the C_T of the technical replicates.
Step 3. Calculate the ΔC_T for both Experimental and C_T Control siRNA Transfections:
ΔC_T = Experimental C_T – Endogenous control C_T
Step 4. Calculate the mean ΔC_T for the negative controls (NC ΔC_T).
Step 5. Calculate the ΔΔC_T:
ΔΔC_T = (Experimental ΔC_T – NC ΔC_T)
Step 6. Calculate percent knockdown:
(%KD) = ([1–2^–ΔΔC_T] x 100)
Step 7. Calculate the mean %KD and the standard deviation of the biological replicates.

Figure 2. Sample Experimental, Real-Time RT-PCR Data from an RNAi experiment using siRNA. Steps correspond to those in Figure 1 on page 10. CLCT=Gene of interest in this example; BR=Biological replicate; TR=Technical replicate; StDev=Standard deviation; CV=Coefficient of variation; NCDCT=MeanDCT for Negative Controls; KD=Knockdown. [Larger Version]

Data Analysis and Interpretation
Figure 2 illustrates how RT-PCR data from biological and technical replicates, and both negative and endogenous controls, is analyzed to determine percent knockdown induced by transfection with the experimental siRNA. In this analysis the coefficient of variation (CV) of the technical replicates plays a key role. CV values under 3% indicate that the variability of the technical replicates is not significant to the overall result of the experiment. If the variation is above this benchmark, confidence in the experiment is drastically decreased, and it would be difficult to make any conclusions from the results. Since the technical replicates in the example pass this criterion, the variability of the biological replicates can be calculated with confidence from the linear conversion of the ΔΔC_Ts (2^–ΔΔC_T).

What Level of Target Knockdown is Necessary?
The goal of RNAi experiments is to examine the biological effect (phenotype) that results from knocking down a target of interest. To elicit that phenotype, however, a target-dependent minimum threshold level of knockdown is required. These sample data and calculations demonstrate how to evaluate target knockdown using a single experimental siRNA. We recommend evaluating the effects of three siRNAs to the target of interest, and then selecting the two siRNAs that provide the most complete knockdown for further experimentation. To be confident that the phenotypic result is due to knocking down the target of interest and not due to an off-target effect, it is important to use two distinct siRNAs to a target in parallel to confirm results.

Compared to siRNAs from other vendors, Silencer® Select siRNAs more reliably elicit maximum knockdown levels that reach the threshold level of knockdown required to see a loss-of-function phenotype. These new siRNAs also have chemical modifications that minimize off-target effects. By using Silencer Select siRNAs to induce knockdown, and analyzing their effects with industry-leading TaqMan® Gene Expression Assays and trusted Applied Biosystems Real-Time PCR Systems, scientists can easily generate reproducible data and arrive to accurate conclusions from their siRNA experiments.

Scientific Contributors
Rajeev Varma, Angie Cheng • Applied Biosystems, Austin, TX