Myotonic dystrophy is an autosomal dominant disease. It involves the electrical instability of muscle tissue, resulting in conductive heart defects, cataracts in the eye, weak muscles, and rapid discharges that make coordinated movement extremely difficult. Skeletal muscle depends on membrane action potential. When there aren’t enough ClC-1 channels in the membrane of skeletal muscle cells, Cl- conductance is reduced. Cl- is unable to stabilize the resting membrane potential, resulting in erratic contractions and delayed muscle relaxation.
There are two types of, DM type 1 is caused by 50 to >3000 CTG repeats in the 3’ untranslated region of the DMPK gene on chromosome 19, and DM type 2 is caused by 75 to 11,000 CCTG repeats in the first intron of the ZNF9 gene on chromosome 3. The repeats are amplified after every generation, and the severity of the symptoms of DM1 and DM2 are proportional to the size of the repeats. Most troublesome is the autosomal dominant nature of the disease; even if just one parent has the disease, 50% of the children will get the disease. How do microsatellite repeats in a region that does not code for protein can cause a dominantly inherited disease?
MBNL proteins, encoded for by the MBNL1, MBNL2, and MBNL3 genes, have been shown to bind to expanded CUG repeats on the DMPK and ZNF9 transcripts and accumulate in the nucleus in vivo. Does the sequestering of MBNL proteins in the nucleus contribute to the development of DM and change ClC-1 splicing?
In order to test this, they made heterozygous Mbnl1 +/ΔE3 mice and homozygous Mbnl1 ΔE3/ΔE3 mice. To make the heterozygous, they generated a pMbnl1ΔE3neo targeting plasmid with HSV-TK and a loxP-flanked phosphoglycerate kinase-neomycin (PGK-neo) to select for recombinants, and an EcoRV cut site. ES cells that had successfully incorporated the vector were transfected with Cre recombinase, and the floxxed neo cassette was excised. To make mice homozygous for the mutation, two heterozygous males were crossed to produce chimeric progeny.
They used RT-PCR to test for the expression of E3 in wild-type, Mbnl1 ΔE3/ΔE3 mice and Mbnl1 +/ΔE3 mice. In one PCR reaction, primers that anneal to E3 and E6 were used. Mbnl1 ΔE3/ΔE3 did not produce any detectable PCR product, confirming that E3 is not expressed in the Mbnl1 gene. Transcripts without E3 have low amplicon yield in this reaction because with only the E6 primer able to anneal to the transcript, there is only a linear increase after every PCR cycle. Since exon 3 is required for CUG-binding MBNL isoforms, so without it, Mbnl1 ΔE3/ΔE3 mice cannot produce these MBNL proteins. This was confirmed by an immunoblot analysis with antibodies specific for the 41- to 42-kD CUG-binding isoform.
In Mbnl1 ΔE3/ΔE3 mice and HSALR mice, aberrant splicing results in the incorporation of a novel 7a and 8a exons in the ClC-1 gene encoding for chloride channels. Expression of exon 7a and 8a in CIC-1 causes a frameshift and results in a truncated protein with only 5 transmembrane helices, instead of 10. In cells with expression of E7a and E8a, there is no ClC-1 in the membrane, and subsequent electromyography testing shows delayed muscle contraction. The Mbnl1+/+ and Mbnl1 +/ΔE3 mice did not express E7a or E8. In these cells, ClC-1 is found on the membrane and there are myotonic discharges in Mbnl1 ΔE3/ΔE3 mice. In all the cells, membrane-associated proteins dystrophin and α-sarcoglycan were expressed equivalently. This is to show that Mbnl1 affects strictly ClC-1, and the loss of ClC-1 is not due to some defect in the export of transmembrane proteins to the cell membrane. These results suggest the presence of E7a and E8a in ClC-1 results in a non-functional protein that does not localize to the membrane, and that the loss of function is what causes the myotonic dystrophy.
ClC-1 isn’t the only protein affected by aberrant splicing in DM; cardiac troponin T (TNNT2) and insulin receptor (INSR) are affected, as well. They want to find out how similar the mouse model is to DM, but the insulin receptor’s splicing isn’t conserved in mice, so only changes to the splicing of the mouse homolog Tnnt2 was tested. RT-PCR amplifying the Tnnt2 gene showed a fetal 5 exon consistently for Mbnl1 ΔE3/ΔE3 mice, just like in humans.The authors want to see if Mbn1l loss affects the splicing of other genes involved in the development of muscle, so they tested fast skeletal muscle troponin T (Tnnt3). In people with DM1, TNNT3 retains the fetal (F) exon.
In the Mbnl1 ΔE3/ΔE3, but not the normal or Mbnl1 +/ΔE3 mice, Tnnt3 also retains the fetal exon (located between exon 8 and 9), detected when RT-PCR was amplified between exon 2 and 11. However, splicing of exon 16 and 17 is unaffected among all three cell types. These results show Mbnl1 loss affect other genes in the muscle. And because the loss of Mbnl1 results in splicing only specific sites, Mbnl1 is probably involved in splice site selection.
The major conclusion is that the CUG-binding isoform of Mbnl1 is directly involved with proper RNA splicing. In Mbnl1 ΔE3/ΔE3 mice where Mbnl1 does not bind to the expanded repeats in DMPK or ZNF9, consistent, aberrant splicing of RNA transcripts occurs. In the mouse model, ClC-1 deficiency causes the myotonic dystrophy. However, the lack of Mbnl1 does not result in the congenital muscle and heart problems present in DM, so there are other factors involved in DM, perhaps ones do not involve mRNA pathogenicity.
Kanadia, R. N., K. A. Johnsone, A. Mankodi, C. Lungu, C. A. Thornton, D. Esson, A. M. Timmers, W. W. Hauswirth, and M. S. Swanson. 2003. A muscleblind knockout model for myotonic dystrophy. Science, 302, 1978-1980. Link: http://www.ncbi.nlm.nih.gov/pubmed/14671308
For further information and to help those affected by this tragic disease, go to http://www.myotonic.org/