A high resolution transcriptome map for both wild-type and NMD defective C. elegans

While the genome sequence of many animals is now complete, their transcriptomes are less well characterised. Both genome-scale tiling arrays and massively parallel sequencing now allow transcriptomes to be mapped at unprecedented depth. We used both technologies to map the C. elegans transcriptome across development. This unbiased overview can serve as a framework for assessing transcriptome changes in a mutant animal and we compared the wild-type data with that of animals that have lost the nonsense-mediated decay (NMD) pathway. Results We find that while the great majority of detectable transcripts map to known gene structures, over 5% of transcribed regions are novel, falling outside current gene annotations. We show that at least 40% of these are novel exons. We also used both technologies to assess isoform complexity and estimate that at least 17% of genes change their major isoform across development. Having mapped the wild-type transcriptome, we examined how this is perturbed in animals lacking nonsense -mediated decay (NMD). NMD prevents expression of prematurely truncated proteins by degrading transcripts containing premature termination codons (PTCs). We find that ~20% of all genes produce transcripts that appear to be targets for NMD. While most of these arise from splicing errors, NMD targets are also enriched for transcripts that contain short open reading frames upstream of the predicted translational start (uORFs). We find an intriguing relationship between the strength of Kozak consensus surrounding the true start codon and the degree to which these uORF containing transcripts are targeted by NMD, suggesting that translational efficiency may be coupled to transcript turnover via the NMD pathway for many transcripts. Conclusions We have generated a high-resolution map of the C. elegans transcriptome and have used it to identify transcripts that are endogenous targets of the NMD machinery. We find that these targets arise principally through splicing errors and suggest that splicing and NMD are highly interlinked processes.

11

Ramani AK, Nelson AC, Kapranov P, Bell I, Gingeras TR, Fraser AG