Obstacles in quantifying A-to-I RNA editing by Sanger sequencing

Adenosine-to-Inosine (A-to-I) RNA editing is the most prevalent type of RNA editing, in which adenosine within a completely or largely double-stranded RNA (dsRNA) is converted to inosine by deamination. RNA editing was shown to be involved in many neurological diseases and cancer; therefore, detection of A-to-I RNA editing and quantitation of editing levels are necessary for both basic and clinical biomedical research. While high-throughput sequencing (HTS) is widely used for global detection of editing events, Sanger sequencing is the method of choice for precise characterization of editing site clusters (hyper-editing) and for comparing levels of editing at a particular site under different environmental conditions, developmental stages, genetic backgrounds, or disease states. To detect A-to-I editing events and quantify them using Sanger sequencing, RNA samples are reverse transcribed, cDNA is amplified using gene-specific primers, and then sequenced. The chromatogram outputs are then compared to the genomic DNA sequence. As editing occurs in the context of dsRNA, the reverse transcription step is performed at a temperature as high as 65 °C, using thermostable reverse transcriptase to open double-stranded structures. However, this measure alone is insufficient for transcripts possessing long stems comprised of hundreds of nucleotide pairs. Consequently, the editing levels detected by Sanger sequencing are significantly lower than those obtained by HTS, and the amplification yield is low. We suggest that the reverse transcription is biased towards unedited transcripts, and the severity of the bias is dependent on the transcript's secondary structure. Here, we show how this bias can be significantly reduced to allow reliable detection of editing levels and sufficient product yield.