Dynamics of co-transcriptional pre-mRNA folding influences the induction of dystrophin exon skipping by antisense oligonucleotides.
Bottom Line: In our analysis, to approximate transcription elongation, a "window of analysis" that included the entire targeted exon was shifted one nucleotide at a time along the pre-mRNA.Possible co-transcriptional secondary structures were predicted using the sequence in each step of transcriptional analysis.A nucleotide was considered "engaged" if it formed a complementary base pairing in all predicted secondary structures of a particular step.
Affiliation: Bioinformatics Institute, Singapore, Singapore.
Antisense oligonucleotides (AONs) mediated exon skipping offers potential therapy for Duchenne muscular dystrophy. However, the identification of effective AON target sites remains unsatisfactory for lack of a precise method to predict their binding accessibility. This study demonstrates the importance of co-transcriptional pre-mRNA folding in determining the accessibility of AON target sites for AON induction of selective exon skipping in DMD. Because transcription and splicing occur in tandem, AONs must bind to their target sites before splicing factors. Furthermore, co-transcriptional pre-mRNA folding forms transient secondary structures, which redistributes accessible binding sites. In our analysis, to approximate transcription elongation, a "window of analysis" that included the entire targeted exon was shifted one nucleotide at a time along the pre-mRNA. Possible co-transcriptional secondary structures were predicted using the sequence in each step of transcriptional analysis. A nucleotide was considered "engaged" if it formed a complementary base pairing in all predicted secondary structures of a particular step. Correlation of frequency and localisation of engaged nucleotides in AON target sites accounted for the performance (efficacy and efficiency) of 94% of 176 previously reported AONs. Four novel insights are inferred: (1) the lowest frequencies of engaged nucleotides are associated with the most efficient AONs; (2) engaged nucleotides at 3' or 5' ends of the target site attenuate AON performance more than at other sites; (3) the performance of longer AONs is less attenuated by engaged nucleotides at 3' or 5' ends of the target site compared to shorter AONs; (4) engaged nucleotides at 3' end of a short target site attenuates AON efficiency more than at 5' end.
Related in: MedlinePlus
Mentions: As an illustration, an AON (novelAON57) target sequence was selected to skip exon 57. All three reported AONs (h57AON1, h57AON2, h57AON3) designed to induce exon 57 skipping by targeting the intra-exonic sequences failed to skip exon 57 . Interestingly, exon 57 manifests an overwhelming occurrence of engaged nucleotides (Figure S3 of the Online Supporting Information); hence, it is relatively difficult to locate a suitably long sequence that has ESE activity as well as co-transcriptional binding accessibility that fulfils the four insights (as described above). For instance, the 3′ ends of the target sites of both h57AON1 and h57AON2 AONs manifest substantial engaged nucleotides whereas the first half of the target site of h57AON3 AON manifests extensive engaged nucleotides (Table S3 of the Online Supporting Information). We designed novelAON57 to have a target site with the following characteristics: negligible occurrence of engaged nucleotides, presence of ESE motifs predicted by ESE-Finder  and RESCUE-ESE , and location at the first half of the exon. Notably, novelAON57 targets a completely different site from the published AONs, as shown in Figure 7. At all AON concentrations tested, i.e. 100nM, 200nM, and 400nM, novelAON57 demonstrates selective skipping of exon 57 with an efficiency of (++) (Figure 7).