Figure 1

Antisense-mediated multiexon 45–55 skipping. A. The introns flanking the intended multiexon skip are extremely large (248 kb and 120 kb for intron 44 and intron 55, respectively), whereas the internal introns are shorter (2.3 – 54 kb). B. It is possible that internal exons are joined before intron 44 is spliced out, resulting in an "exon block" of exons 45–55. Here, the splice acceptors of exons 45 and 56 are competing for the splice donor of exon 44, and the splice donors of exons 44 and 55 are competing for the splice acceptor of exon 56. While the donor sites of exons 44 and 55 are of similar strength (1 and 0.99, respectively, calculated with the Berkely Drosophila Genome Project for human splice site prediction software; http://www.fruitfly.org/seq_tools/splice.html), the acceptor splice site of exon 45 (0.76) is weaker than that of exon 56 (0.97), thus explaining why occasionally exon 44 is joined to exon 56 rather than exon 45. C. Exon 45–55 skipping levels and frequency might be enhanced by AONs targeting exon 45 and exon 55, which should increase the chance that exon 44 and exon 56 are joined. D. Regardless of whether the exon block hypothesis is correct, exon 45–55 skipping can be induced using a cocktail of AONs targeting each of the individual AONs. However, it is more likely that some but not all exons are skipped and that this cocktail gives rise to many intermediate products where one or more (but not all) intended exons are skipped.