Abstract
The poly(A) tail, co-transcriptionally added to most eukaryotic RNAs, plays an important role in post-transcriptional regulation through modulating mRNA stability and translational efficiency. The length of the poly(A) tail is dynamic, decreasing or increasing in response to various stimuli through the action of enzymatic complexes, and changes in tail length are exploited in regulatory pathways implicated in various biological processes.
To date, assessment of poly(A) tail length has mostly relied on protocols targeting only a few transcripts. We present PASP (‘poly(A) tail sequencing protocol’), a whole-transcriptome approach to measure tail lengths — including a computational pipeline implementing all necessary analyses. PASP uses direct Illumina sequencing of cDNA fragments obtained through G-tailing of poly(A)-selected mRNA followed by fragmentation and reverse transcription.
Analysis of reads corresponding to spike-in poly(A) tracts of known length indicated that mean tail lengths can be confidently measured, given sufficient coverage. We further explored the utility of our approach by comparing tail lengths estimated from wild type and Δccr4-1/pan2 mutant yeasts. The yeast whole-transcriptome tail length distributions showed high consistency between biological replicates, and the expected upward shift in tail lengths in the mutant samples was detected. This suggests that PASP is suitable for the assessment of global polyadenylation status in yeast.
The correlation of per-transcript mean tail lengths between biological and technical replicates was low (higher between mutant samples). Both, however, reached high values after filtering for transcripts with greater coverage. We also compare our results with those of other methods. We identify a number of improvements that could be used in future PASP experiments and, based on our results, believe that direct sequencing of poly(A) tails can become the method of choice for studying polyadenylation using the Illumina platform