Introduction

Reverse transcription of RNA into cDNA is a core method for analysis of gene expression using DNA microarray or real-time PCR (1,2,3,4,5). The reverse transcription should result in a cDNA population that reflects the original messenger RNA (mRNA) population in terms of transcript abundance and complexity. Furthermore, the reverse transcription reaction should be as efficient as possible to give maximum sensitivity in the final assay. Sensitivity is often problematic in microarray analysis of unamplified mRNA. Sensitivity is also an issue when following fusion gene markers in cancer patients where, for instance, it is desired to detect a single cancer cell expressing the BCR/ABL fusion transcripts in a background of 100,000 normal cells in chronic myeloid leukemia (CML) patients (6,7). This suggests that improvements in reverse transcription could have large impact in assay results.

RNA quality (8,9,10,11,12), priming strategy (13), and enzyme efficiency (14,15) are important parameters for obtaining high yield cDNA of good quality. Less studied is the impact of different primers in the reaction. Reverse transcription reactions can be primed using specific primers (16,17) if relatively few mRNA species are targeted. This approach is not practical for whole transcriptome analysis using microarray, because it would require synthesis and mixing of thousands of specific primers. In those cases, the reverse transcription reaction is primed with oligo(dT) (18,19) random hexamers (18)(20,21,22,23) or random nonamers (23,24). Oligo(dT) priming has the virtue of producing cDNA from the 3′ end of poly(A) mRNA, allowing total RNA to be used as a template. The drawback is that oligo(dT) priming often results in a 3′ bias compared with random priming (14). Poly(A)-selected RNA like an isolated mRNA fraction or amplified RNA (aRNA) are preferably reverse transcribed with random primers, because random priming is less likely to give a 3′ end bias in the resulting cDNA.

Here we report a novel priming method for reverse transcription reactions and a study of the implication of random primer length in reverse transcription reactions on the yield of the resulting cDNA and the influence on the amount of detectable genes on oligonucleotide microarrays.

Materials and Methods

Cells

Human carcinoma cells (HeLa) were cultured in 75 cm² culture flasks (Easy Flask™; Nalge Nunc International, Rochester, NY, USA) in 25 mL RPMI 1640 media supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 µg/mL; all from Sigma, St. Louis, MO, USA). Cells were cultured at 37°C in an atmosphere containing 5% CO₂ (AGA, Copenhagen, Denmark) in a CO₂ incubator (Assab; Don Whitley Scientific Ltd., Shipley, West Yorkshire, UK). The cells were seeded at a density of 6667 cells/cm², and the cells were subcultured every 3 to 4 days when approximately 90% confluence was reached.

RNA Isolation and Amplification

Total RNA was isolated by using RNeasy® Mini Kit spin columns (Qiagen, Valencia, CA, USA). Quantification of total RNA was performed in an Ultraspec 3000 spectrophotometer (Pharmacia Biotech, Cambridge, UK), and validation of RNA quality was performed by using the RNA 6000 Nano Assay on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA). Amplification was performed with the RiboAmp® T7-based RNA amplification kit (Arcturus Engineering, Mountain View, CA, USA) according to the manufacturer's instructions using 2 µg total RNA as a starting material, yielding 35–40 µg aRNA after one round of amplification. For each labeling reaction, 3 µg aRNA were used unless otherwise mentioned. The mRNA fraction was isolated from 100 µg total RNA using oligo(dT)₂₅ Dynabeads® (Dynal, Oslo, Norway) according to the manufacturer's instructions. The mRNA (2 3 µg) was eluted in 20 µL buffer (10 mM Tris-HC1, pH 7.5).

cDNA Synthesis

Synthesis of cDNA was performed as described elsewhere (23), with the following modifications. In brief, 2–3 µg aRNA were mixed with 3.35 nmol random primers (Sigma-Genosys, Haverhill, UK or Invitrogen, Paisley, UK), unless otherwise mentioned, in a total volume of 18.5 µL. The mixture was heated at 70°C for 10 min and snap-cooled on ice for at least 1 min. Reverse transcription reaction mixture lacking enzyme was added to a final concentration of 500 µM each dATR dCTR and dGTR 300 µM dTTP (Larova Biochemie GMBH, Teltow, Germany), 200 µM 5-aminoallyl-dUTP (Sigma-Aldrich, Steinheim, Germany), 40 U RNasin® (Promega, Mannheim, Germany), 1× first-strand buffer (Invitrogen), 10 mM dithiothreitol (DTT; Invitrogen) in a final volume of 29 µL. The reaction mixture was incubated for 3–5 min at room temperature to let the primers anneal. SuperScript™ II (400 U, 2 µL; Invitrogen) were added, and the mixture was incubated at 42°C for 5 h. Avian myeloblastosis virus (AMV)-catalyzed reactions were performed in 100-µL reactions containing 3 µg aRNA, 1× reaction buffer, 6 µg either Inv or R15 primer, 60 U RNasin, 50 µg/mL BSA, 500 µM each dATR dCTR and dGTR 300 µM dTTR 200 µM 5-aminoallyl-dUTR and 40 U AMV enzyme (USB, Cleveland, Ohio, USA). Moloney murine leukemia virus (MmLV)-catalyzed reactions were performed in 50-µL reactions containing 3 µg aRNA, 6 µg either Inv or R15 primer, 1× reaction buffer, 40 U RNasin, 500 µM each dATR dCTR and dGTR 300 µM dTTR 200 µM 5-aminoallyl-dUTR and 500 U MmLV enzyme (USB). Annealing of primers was performed identically to SuperScript II-catalyzed reactions, and the enzymes were added last to all reactions.