To better understand the mechanism of splicing in CLCN we
To better understand the mechanism of splicing in CLCN1, we began by constructing a minigene spanning exons 5–7 of CLCN1, resulting in the synthetic CLCN1 (5–7) minigene. We also established a new assay using real-time reverse transcription (RT)-PCR, which can distinguish between splicing variants based upon the presence of the TAG-inserted pattern. We found significant differences between the human and mouse orthologue, which may have important implications for the study of DM.
Materials and methods
Discussion Here, we investigated human specific splicing of CLCN1, the gene responsible for myotonia in DM, using an CLCN1 (5–7) minigene expression construct. Expression of this minigene revealed various splicing patterns, including the TAG-inserted pattern, which encodes for a stop Iodoacetyl-LC-Biotin and is readily detected in human skeletal muscle biopsies. To distinguish between these various transcripts, we developed an RT-PCR assay specific for the 5–6–7 splicing pattern, regardless of the presence of other variants. This assay was then used to quantify cellular splicing. MBNL1 was shown to activate the expression of the 5–6–7 splicing pattern, with no effects seen for either MBNL2 or MBNL3. This result marks an important difference from that of our previous reports using mouse Clcn1. In contrast to MBNL1, overexpression of each CELF isoform decreased the expression of the 5–6–7 splicing pattern. The TAG-inserted pattern is likely the cause of myotonia, as it produces a stop codon just before exon 7 and can suppress the expression of the 5–6–7 splicing pattern. Although we detected the acceptor site responsible for this insertion, the overall function of the TAG-inserted pattern remains unclear. In terms of its role in DM, no significant differences were observed in the percentage of 5–6–7 and TAG-inserted patterns between DM and controls. However, as sequencing was performed only 15 times per individual, comparisons between these two groups may not be fully reliable. We therefore needed to establish an assay that could be used to specifically quantify the TAG-inserted pattern, as well as investigate the TAG-inserted pattern in more detail. In this study, we identified two tandemly repeated TAG sequences immediately upstream of exon 7, one of which served as the splice acceptor site for this exon. These two TAG sequences are conserved across a range of species, including chimpanzees, dogs and mice, and have been used in a variety of experiments investigating Clcn1. Hence, a more accurate assay method like the one described here may be necessary for future examinations of Clcn1. The assay system described here is able to directly quantify the production of 5–6–7 splicing pattern transcripts, an important distinction as the expression of this protein is directly related to the onset of myotonia. However, this assay was unable to quantify other splicing variants. We will therefore need to establish additional assays such as direct RNA sequencing as a means of quantifying the percentage of each splicing variant to better understand the effects of splicing in CLCN1. Furthermore, this assay relied on real-time PCR performed at relatively high annealing and extension temperatures to achieve transcript-specific amplification of the normal splicing pattern. As a result, variance in Ct-values between triplicates was detected. Further improvements will be necessary to fully quantify the various splicing products produced by this gene. Using this new assay, we were able to detect differences in terms of CLCN1 regulation between MBNL1 and MBNL2-3. All three MBNL proteins bind CHG/CHHG sequences of RNA and have the same amount of zinc finger domains, which are critical for recognising a common consensus sequence in pre-mRNA and mRNA targets . Splicing efficacy therefore depends on differences in the target genes themselves, as opposed to any specific MBNL protein. In fact, our previous reports showed that the effects of each isoform differed based upon its target genes , . On the other hand, the expression levels of MBNL2 and MBNL3 are weaker than that of MBNL1 (Fig. 4A), which may have mediated the lower overall effects seen with these homologues. However, in our previous study using Clcn1, we observed a similar tendency in terms of protein expression; however, in this model, both MBNL2 and MBNL3 did enhance the expression of the 5–6–7 splicing pattern . Moreover, the effects of MBNL are regulated by a variety of other factors, such as the secondary structure of mRNAs .