We have studied non sequential splicing and the occurrence of multi-step splicing for the largest reported DMD gene, using Capture-pre-mRNA-sequencing. Employing this approach, we have been able to shows for the first time experimental evidence of recursive splicing in a human gene and provides a detailed insight in the co-transcriptional splicing of the dystrophin transcript which occurs in a non-sequential manner for many introns.. Non-sequential and multi-step splicing of the dystrophin transcript

The long (2.2Mb) human dystrophin transcript (DMD) takes 16 hours to be transcribed and is co-transcriptional spliced. The presence of long introns (24 over 10kb long, 5 over 100kb long) poses challenges to the splicing machinery. Additionally, the heterogeneity in intron size suggests that intron removal not always takes place consecutively. We explored the order of intron removal and potential multi-step splicing for the DMD transcripts in human skeletal muscle cell lines. A customized library of probes (120bp) covering all exons and introns, has been generated capturing pre-mRNA and targeting DMD, followed by Hiseq sequencing. SplicePie pipeline has been used to analyze capture-pre-mRNA-sequencing data. Analysis showed that DMD introns can be removed non-sequentially generating exon blocks – joined blocks of exons flanked by unspliced introns. Exon blocks were detected by analysis of the coverage of paired-end reads and validated experimentally using PCR and Sanger sequencing. No correlation between intron length and speed of intron removal was observed. Computational analysis and experimental validation revealed that intron removal takes place in several steps for the majority of dystrophin introns. We found two mechanisms of multi-step intron removal in DMD– recursive and nested splicing. Non-sequential and multi-step splicing events were found throughout the DMD gene across three cell lines. We believe that our findings of non-sequential and multi-step splicing provide insight in the splicing mechanism and will be useful to optimize therapeutic strategies that interfere with the splicing process.