Reagent volume and plate bias in real-time polymerase chain reaction.
چکیده
When optimal conditions are employed, real-time polymerase chain reaction (PCR) is a sensitive and accurate technique enabling the quantiWcation of low-copynumber transcripts [1]. However, as with conventional PCR, small variations in initial reaction conditions are ampliWed exponentially and can signiWcantly aVect results [2]. Uniform reaction conditions are therefore essential to achieve accuracy and reproducibility during transcript quantiWcation. Reagent costs are considerable, and in a high-throughput setting a reduction in the reagent volume used in each reaction would signiWcantly reduce the cost of real-time PCR. However, the eVect of reduced volume on the accuracy of results in a plate-based system has not, to our knowledge, been examined. Here we show that lower reagent volumes can reduce reproducibility by enhancing a bias in results across a plate. Total RNA was isolated from conXuent A549 lung epithelial type II cells (ATCC, Manassas, VA) using the TRIzol method (Invitrogen Canada, Burlington, ON) and quantiWed using the RiboGreen Reagent and Kit (Molecular Probes, Eugene, OR). RNA was reverse transcribed using MuLV reverse transcriptase and random hexamers (Applied Biosystems, Mississauga, ON) according to the manufacturer’s instructions. Primers for human endothelin-1 (NM_001955), the gene encoding a potent vasoconstrictor peptide relevant in a number of cardiovascular pathologies [3], were designed using Vector NTI (Informax), and double-desalted primers were ordered from Invitrogen. Annealing conditions were optimized, and high reaction eYciency (90%) was conWrmed using a dilution series of A549 cDNA over a 80to 0.625-ng range. We compared three reagent mixes,