Single-nucleotide resolution of N 6-adenine methylation sites in DNA and RNA by nitrite sequencing

Abstract

A single-nucleotide resolution sequencing method of N ⁶-adenine methylation sites in DNA and RNA is described. Using sodium nitrite under acidic conditions, chemoselective deamination of unmethylated adenines readily occurs, without competing deamination of N ⁶-adenine sites. The deamination of adenines results in the formation of hypoxanthine bases, which are read by polymerases and reverse transcriptases as guanine; the methylated adenine sites resist deamination and are read as adenine. The approach, when coupled with high-throughput DNA sequencing and mutational analysis, enables the identification of N ⁶-adenine sites in RNA and DNA within various sequence contexts.

Fig. 1. Similar Watson–Crick–Franklin base-pairing observed between adenine and thymine (left) and N⁶-methyladenine and thymine (right) limits direct high-throughput sequencing.

Fig. 2. Nitrite reaction with aromatic amines. (a) Reaction of 2-aminopyridine with sodium nitrite under acidic aqueous conditions. Reaction of nitrite with (b) adenine and (c) N⁶-methyladenine bases in RNA or DNA. Inset: inosine nucleobases are read as guanine by polymerases.

Fig. 3. Nitrite-mediated deamination of free nucleosides. HPLC analysis of the conversion of (a) adenosine into inosine; (b) guanosine into xanthosine; and (c) cytidine into uridine using NaNO₂ and 1.7% aqueous AcOH, at 22 °C over 12 h.

Fig. 4. Nitrite-mediated deamination of methylated nucleosides. (a) HPLC analysis of the conversion of m⁶A into nitrosylated m⁶A (m⁶A-NO) using NaNO₂ and 1.7% aqueous AcOH, at 22 °C over 3.5 h. m⁶A-NO was confirmed by ESI-MS. Note that inosine was not detected by HPLC analysis of nitrosylation of m⁶A. No reaction was observed over a 12 h period for (b) m¹A or (c) m³C.

Fig. 5. Optimisation of nitrite-mediated deamination on RNA and DNA. “>” denotes corresponding transition or transversion. (a) Recovery of DNA and RNA with respect to acid concentration during the nitrite reaction. Error based on assessment in duplicates. Dotted line represents 80% threshold of recovery. (b) High-throughput sequencing of RNA after nitrite reaction at varying acid concentrations. Mutations are represented in legend, and correspond to the specific type of mutation per expected nucleobase. (c) High-throughput sequencing of DNA after nitrite reaction at varying acid concentrations. Note that high-throughput DNA analysis above 2.3% AcOH was not processed due to undesirably low isolation (per Fig. 4a). (d) Quantification of methylation fraction of an adenosine site within an RNA sequence. See ESI for sequences.

Fig. 6. Normalised sequencing representation of the ratio of (d)A → (d)G mutation at each nucleobase following treatment with 1 M sodium nitrite in the presence of acetic acid for 5 h at 22 °C. The DNA sequences contain a single 6mA site at position 63 (a) and three 6mA sites at positions 35, 36, and 55 (b). The RNA sequences contain a single m⁶A site at position 26 (c) and two m⁶A sites at positions 31 and 32 (d). The 23S rRNA from E. coli contains a single m⁶A site at position 2030 (e). Primer sequence regions are not shown for clarity. See ESI for complete sequences and predicted folded structures determined with 1 M Na⁺ at 22 °C using MFold.