tRNA base methylation identification and quantification via high-throughput sequencing

Abstract

Eukaryotic transfer RNAs contain on average 14 modifications. Investigations of their biological functions require the determination of the modification sites and the dynamic variations of the modification fraction. Base methylation represents a major class of tRNA modification. Although many approaches have been used to identify tRNA base methylations, including sequencing, they are generally qualitative and do not report the information on the modification fraction. Dynamic mRNA modifications have been shown to play important biological roles; yet, the extent of tRNA modification fractions has not been reported systemically. Here we take advantage of a recently developed high-throughput sequencing method (DM-tRNA-seq) to identify and quantify tRNA base methylations located at the Watson-Crick face in HEK293T cells at single base resolution. We apply information derived from both base mutations and positional stops from sequencing using a combination of demethylase treatment and cDNA synthesis by a thermophilic reverse transcriptase to compile a quantitative "Modification Index" (MI) for six base methylations in human tRNA and rRNA. MI combines the metrics for mutational and stop components from alignment of sequencing data without demethylase treatment, and the modifications are validated in the sequencing data upon demethylase treatment. We identify many new methylation sites in both human nuclear and mitochondrial-encoded tRNAs not present in the RNA modification databases. The potentially quantitative nature of the MI values obtained from sequencing is validated by primer extension of several tRNAs. Our approach should be widely applicable to identify tRNA methylation sites, analyze comparative fractional modifications, and evaluate the modification dynamics between different samples.

Base-Resolution Mapping Reveals Distinct m1A Methylome in Nuclear- and Mitochondrial-Encoded Transcripts

“…Our results confirmed that m 1 A58 is globally present in the cytosolic tRNAs (Figure 1C and S2F). A recent study reported m 1 A at position 9 in cytosolic tRNA Asp(GUC) (Clark et al, 2016); this site is also detected by m 1 A-MAP (Figure 1D). Collectively, these observations suggest that m 1 A-MAP is highly sensitive in detecting m 1 A at single-base resolution.…”

Structure of a Thermostable Group II Intron Reverse Transcriptase with Template-Primer and Its Functional and Evolutionary Implications

“…As a result, group II intron RTs have higher fidelity, processivity and strand displacement activity than retroviral RTs, along with a highly proficient end-to-end template switching activity that is minimally dependent upon base pairing (Mohr et al, 2013). Recently, new methods for producing group II intron RTs in soluble form with high yield and activity have enabled their use for biotechnological applications, including new approaches for next-generation RNA sequencing (RNA-seq), identification of RNA post-transcriptional modifications, and RNA structure mapping (Clark et al, 2016; Nottingham et al, 2016; Zheng et al, 2015; Zubradt et al, 2016). …”

Genome recoding by tRNA modifications

“…This method holds substantial promise for the comprehensive characterization of transcriptome-wide RNA methylation patterns. Similarly, ARM-seq (AlkB-facilitated RNA methylation sequencing) or DM-tRNA-seq (demethylase tRNA sequencing) revealed a complex modification landscape of full-length tRNAs and tRNA fragments [30–32]. …”