Contribution of cell culture, RNA extraction, and reverse transcription to the measurement error in quantitative reverse transcription polymerase chain reaction-based gene expression quantification.

Quantitative polymerase chain reaction (qPCR) instruments are known to be reliable. However, many authors have underlined the poor reliability of the procedures that precede the measurement of gene expression--cell culture, RNA extraction, and reverse transcription. Here we quantified the measurement errors due to each step and estimated the correction that would accrue from replicating any of those steps. We measured the relative expression of the APC-11 gene (the catalytic anaphase-promoting complex/cyclosome subunit suspected to be involved in breast cancer) with step replication in 18 breast cancer cell lines. The final qPCR step was found to be reproducible (standard deviation [SD]=0.26). In comparison with the between-cell-line variability (SD=1.7), the variability due to the previous steps (cell culture, RNA extraction, and reverse transcription) was on the same order of magnitude (SD=1.2-2.0). Misclassification rates were used to assess the impact of replicating each manual procedure. The misclassification rates improved with replication of cell culture, RNA extraction, and reverse transcription (90.0, 60.9, and 61.1% decreases, respectively). The results point out a high error level in the quantification of gene expression, and these errors may stem from all steps of the procedure. The best correction would accrue from replicating cell culture.