Mouse Model Of Myotonic Dystrophy Research Articles

Altered drug metabolism and increased susceptibility to fatty liver disease in a mouse model of myotonic dystrophy

Myotonic Dystrophy type 1 (DM1), a highly prevalent form of muscular dystrophy, is caused by (CTG)n repeat expansion in the DMPK gene. Much of DM1 research has focused on the effects within the muscle and neurological tissues; however, DM1 patients also suffer from various metabolic and liver dysfunctions such as increased susceptibility to metabolic dysfunction-associated fatty liver disease (MAFLD) and heightened sensitivity to certain drugs. Here, we generated a liver-specific DM1 mouse model that reproduces molecular and pathological features of the disease, including susceptibility to MAFLD and reduced capacity to metabolize specific analgesics and muscle relaxants. Expression of CUG-expanded (CUG)exp repeat RNA within hepatocytes sequestered muscleblind-like proteins and triggered widespread gene expression and RNA processing defects. Mechanistically, we demonstrate that increased expression and alternative splicing of acetyl-CoA carboxylase 1 drives excessive lipid accumulation in DM1 livers, which is exacerbated by high-fat, high-sugar diets. Together, these findings reveal that (CUG)exp RNA toxicity disrupts normal hepatic functions, predisposing DM1 livers to injury, MAFLD, and drug clearance pathologies that may jeopardize the health of affected individuals and complicate their treatment.

Altered Behavioral Responses Show GABA Sensitivity in Muscleblind-Like 2-Deficient Mice: Implications for CNS Symptoms in Myotonic Dystrophy.

Considerable evidence from mouse models and human postmortem brain suggests loss of Muscleblind-like protein 2 (MBNL2) function in brain is a major driver of CNS symptoms in Myotonic dystrophy type 1 (DM1). Increased hypersomnia, fatigue, and surgical complications associated with general anesthesia suggest possible sensitivity to GABAergic inhibition in DM1. To test the hypothesis that MBNL2 depletion leads to behavioral sensitivity to GABAA receptor (GABAA-R) modulation, Mbnl2 knock-out (KO) and wild-type (WT) littermates were treated with the anesthetic sevoflurane, the benzodiazepine diazepam, the imidazopyridine zolpidem, and the benzodiazepine rescue agent, flumazenil (Ro 15-1788), and assessed for various behavioral metrics. Mbnl2 KO mice exhibited delayed recovery following sevoflurane, delayed emergence and recovery from zolpidem, and enhanced sleep time at baseline that was modulated by flumazenil. A significantly higher proportion of Mbnl2 KO mice also loss their righting reflex [loss of righting reflex (LORR)] from a standard diazepam dose. We further examined whether MBNL2 depletion affects total GABAA-R mRNA subunit levels and validated RNA-sequencing data of mis-spliced Gabrg2, whose isoform ratios are known to regulate GABA sensitivity and associated behaviors. While no other GABAA-R subunit mRNA levels tested were altered in Mbnl2 KO mouse prefrontal cortex, Gabrg2S/L mRNA ratio levels were significantly altered. Taken together, our findings indicate that loss of MBNL2 function affects GABAergic function in a mouse model of myotonic dystrophy (DM1).

Reversible cardiac disease features in an inducible CUG repeat RNA-expressing mouse model of myotonic dystrophy.

Myotonic dystrophy type 1 (DM1) is caused by a CTG repeat expansion in the DMPK gene. Expression of pathogenic expanded CUG repeat (CUGexp) RNA causes multisystemic disease by perturbing the functions of RNA-binding proteins, resulting in expression of fetal protein isoforms in adult tissues. Cardiac involvement affects 50% of individuals with DM1 and causes 25% of disease-related deaths. We developed a transgenic mouse model for tetracycline-inducible and heart-specific expression of human DMPK mRNA containing 960 CUG repeats. CUGexp RNA is expressed in atria and ventricles and induced mice exhibit electrophysiological and molecular features of DM1 disease, including cardiac conduction delays, supraventricular arrhythmias, nuclear RNA foci with Muscleblind protein colocalization, and alternative splicing defects. Importantly, these phenotypes were rescued upon loss of CUGexp RNA expression. Transcriptome analysis revealed gene expression and alternative splicing changes in ion transport genes that are associated with inherited cardiac conduction diseases, including a subset of genes involved in calcium handling. Consistent with RNA-Seq results, calcium-handling defects were identified in atrial cardiomyocytes isolated from mice expressing CUGexp RNA. These results identify potential tissue-specific mechanisms contributing to cardiac pathogenesis in DM1 and demonstrate the utility of reversible phenotypes in our model to facilitate development of targeted therapeutic approaches.

HNRNPA1-induced spliceopathy in a transgenic mouse model of myotonic dystrophy

Studies on myotonic dystrophy type 1 (DM1) have led to the RNA-mediated disease model for hereditary disorders caused by noncoding microsatellite expansions. This model proposes that DM1 disease manifestations are caused by a reversion to fetal RNA processing patterns in adult tissues due to the expression of toxic CUG RNA expansions (CUGexp) leading to decreased muscleblind-like, but increased CUGBP1/ETR3-like factor 1 (CELF1), alternative splicing activities. Here, we test this model in vivo, using the mouse HSALR poly(CUG) model for DM1 and recombinant adeno-associated virus (rAAV)-mediated transduction of specific splicing factors. Surprisingly, systemic overexpression of HNRNPA1, not previously linked to DM1, also shifted DM1-relevant splicing targets to fetal isoforms, resulting in more severe muscle weakness/myopathy as early as 4 to 6 wk posttransduction, whereas rAAV controls were unaffected. Overexpression of HNRNPA1 promotes fetal exon inclusion of representative DM1-relevant splicing targets in differentiated myoblasts, and HITS-CLIP of rAAV-mycHnrnpa1-injected muscle revealed direct interactions of HNRNPA1 with these targets in vivo. Similar to CELF1, HNRNPA1 protein levels decrease during postnatal development, but are elevated in both regenerating mouse muscle and DM1 skeletal muscle. Our studies suggest that CUGexp RNA triggers abnormal expression of multiple nuclear RNA binding proteins, including CELF1 and HNRNPA1, that antagonize MBNL activity to promote fetal splicing patterns.

Better living through peptide-conjugated chemistry: next-generation antisense oligonucleotides.

Two different antisense oligonucleotide-based (ASO-based) therapies are currently in clinical use to treat neuromuscular diseases. This success, for Duchenne muscular dystrophy and spinal muscular atrophy, offers hope not only for additional neuromuscular diseases, but also for other disorders that could benefit from RNA-targeted therapies. A major limitation for more widespread application of ASOs relates to relatively poor tissue penetration. In this issue of the JCI, Klein et al. showed that conjugating an ASO with an arginine-rich cell-penetrating peptide, Pip6a, enhanced delivery, resulting in corrective outcome for a mouse model of myotonic dystrophy. Linking ASOs to cell-penetrating peptides, or even other moieties, is an approach currently under development with treatment potential to expand to other disorders.

Altered levels of the splicing factor muscleblind modifies cerebral cortical function in mouse models of myotonic dystrophy

Myotonic dystrophy (DM) is a progressive, multisystem disorder affecting skeletal muscle, heart, and central nervous system. In both DM1 and DM2, microsatellite expansions of CUG and CCUG RNA repeats, respectively, accumulate and disrupt functions of alternative splicing factors, including muscleblind (MBNL) proteins. Grey matter loss and white matter changes, including the corpus callosum, likely underlie cognitive and executive function deficits in DM patients. However, little is known how cerebral cortical circuitry changes in DM. Here, flavoprotein optical imaging was used to assess local and contralateral responses to intracortical motor cortex stimulation in DM-related mouse models. In control mice, brief train stimulation generated ipsilateral and contralateral homotopic fluorescence increases, the latter mediated by the corpus callosum. Single pulse stimulation produced an excitatory response with an inhibitory-like surround response mediated by GABAA receptors. In a mouse model of DM2 (Mbnl2 KO), we observed prolonged and increased responsiveness to train stimulation and loss of the inhibition from single pulse stimulation. Conversely, mice overexpressing human MBNL1 (MBNL1-OE) exhibited decreased contralateral response to train stimulation and reduction of inhibitory-like surround to single pulse stimulation. Therefore, altering levels of two key DM-associated splicing factors modifies functions of local cortical circuits and contralateral responses mediated through the corpus callosum.

Downregulation of the Glial GLT1 Glutamate Transporter and Purkinje Cell Dysfunction in a Mouse Model of Myotonic Dystrophy

SUMMARYBrain function is compromised in myotonic dystrophy type 1 (DM1), but the underlying mechanisms are not fully understood. To gain insight into the cellular and molecular pathways primarily affected, we studied a mouse model of DM1 and brains of adult patients. We found pronounced RNA toxicity in the Bergmann glia of the cerebellum, in association with abnormal Purkinje cell firing and fine motor incoordination in DM1 mice. A global proteomics approach revealed downregulation of the GLT1 glutamate transporter in DM1 mice and human patients, which we found to be the result of MBNL1 inactivation. GLT1 downregulation in DM1 astrocytes increases glutamate neurotoxicity and is detrimental to neurons. Finally, we demonstrated that the upregulation of GLT1 corrected Purkinje cell firing and motor incoordination in DM1 mice. Our findings show that glial defects are critical in DM1 brain pathophysiology and open promising therapeutic perspectives through the modulation of glutamate levels.

Reduced cytoplasmic MBNL1 is an early event in a brain-specific mouse model of myotonic dystrophy.

Myotonic dystrophy type 1 (DM1) is caused by an expansion of CTG repeats in the 3' untranslated region (UTR) of the dystrophia myotonia protein kinase (DMPK) gene. Cognitive impairment associated with structural change in the brain is prevalent in DM1. How this histopathological abnormality during disease progression develops remains elusive. Nuclear accumulation of mutant DMPK mRNA containing expanded CUG RNA disrupting the cytoplasmic and nuclear activities of muscleblind-like (MBNL) protein has been implicated in DM1 neural pathogenesis. The association between MBNL dysfunction and morphological changes has not been investigated. We generated a mouse model for postnatal expression of expanded CUG RNA in the brain that recapitulates the features of the DM1 brain, including the formation of nuclear RNA and MBNL foci, learning disability, brain atrophy and misregulated alternative splicing. Characterization of the pathological abnormalities by a time-course study revealed that hippocampus-related learning and synaptic potentiation were impaired before structural changes in the brain, followed by brain atrophy associated with progressive reduction of axon and dendrite integrity. Moreover, cytoplasmic MBNL1 distribution on dendrites decreased before dendrite degeneration, whereas reduced MBNL2 expression and altered MBNL-regulated alternative splicing was evident after degeneration. These results suggest that the expression of expanded CUG RNA in the DM1 brain results in neurodegenerative processes, with reduced cytoplasmic MBNL1 as an early event response to expanded CUG RNA.

Phenylbutazone induces expression of MBNL1 and suppresses formation of MBNL1-CUG RNA foci in a mouse model of myotonic dystrophy.

Myotonic dystrophy type 1 (DM1) is caused by abnormal expansion of CTG repeats in the 3′ untranslated region of the DMPK gene. Expanded CTG repeats are transcribed into RNA and make an aggregate with a splicing regulator, MBNL1, in the nucleus, which is called the nuclear foci. The nuclear foci sequestrates and downregulates availability of MBNL1. Symptomatic treatments are available for DM1, but no rational therapy is available. In this study, we found that a nonsteroidal anti-inflammatory drug (NSAID), phenylbutazone (PBZ), upregulated the expression of MBNL1 in C2C12 myoblasts as well as in the HSALR mouse model for DM1. In the DM1 mice model, PBZ ameliorated aberrant splicing of Clcn1, Nfix, and Rpn2. PBZ increased expression of skeletal muscle chloride channel, decreased abnormal central nuclei of muscle fibers, and improved wheel-running activity in HSALR mice. We found that the effect of PBZ was conferred by two distinct mechanisms. First, PBZ suppressed methylation of an enhancer region in Mbnl1 intron 1, and enhanced transcription of Mbnl1 mRNA. Second, PBZ attenuated binding of MBNL1 to abnormally expanded CUG repeats in cellulo and in vitro. Our studies suggest that PBZ is a potent therapeutic agent for DM1 that upregulates availability of MBNL1.

G.P.90 - Neuroglial miscommunication in the cerebellum of a mouse model of myotonic dystrophy

G.P.90 - Neuroglial miscommunication in the cerebellum of a mouse model of myotonic dystrophy

Therapeutic impact of systemic AAV-mediated RNA interference in a mouse model of myotonic dystrophy.

RNA interference (RNAi) offers a promising therapeutic approach for dominant genetic disorders that involve gain-of-function mechanisms. One candidate disease for RNAi therapy application is myotonic dystrophy type 1 (DM1), which results from toxicity of a mutant mRNA. DM1 is caused by expansion of a CTG repeat in the 3' UTR of the DMPK gene. The expression of DMPK mRNA containing an expanded CUG repeat (CUG(exp)) leads to defects in RNA biogenesis and turnover. We designed miRNA-based RNAi hairpins to target the CUG(exp) mRNA in the human α-skeletal muscle actin long-repeat (HSA(LR)) mouse model of DM1. RNAi expression cassettes were delivered to HSA(LR) mice using recombinant adeno-associated viral (rAAV) vectors injected intravenously as a route to systemic gene therapy. Vector delivery significantly reduced disease pathology in muscles of the HSA(LR) mice, including a reduction in the CUG(exp) mRNA, a reduction in myotonic discharges, a shift toward adult pre-mRNA splicing patterns, reduced myofiber hypertrophy and a decrease in myonuclear foci containing the CUG(exp) mRNA. Significant reversal of hallmarks of DM1 in the rAAV RNAi-treated HSA(LR) mice indicate that defects characteristic of DM1 can be mitigated with a systemic RNAi approach targeting the nuclei of terminally differentiated myofibers. Efficient rAAV-mediated delivery of RNAi has the potential to provide a long-term therapy for DM1 and other dominant muscular dystrophies.

Sleep phenotype characterization of peripheral and central mouse models of myotonic dystrophy

Excessive daytime sleepiness and associated alterations in REM sleep patterns are among the most characteristic non-muscular features of myotonic dystrophy (DM). The symptoms of DM are caused by an expanded C (C) UG repeat that sequesters muscleblind-like protein 1 (MBNL1) and MBNL2, proteins that regulates alternative splicing required for fetal to adult developments. A recent study strongly suggests that major pathological changes in the DM brain are attributable to MBNL2 sequestration by toxic RNAs and dysregulation of specific alternative splicing events required for normal adult CNS function. We thus performed sleep EEG/EMG evaluations on Mbnl1 and Mbnl2 knockout (KO) mice. Adult wild and KO mice of Mbnl1 or Mbnl2 at 6 month of age (Mbnl1; n = 7, Mbnl2; n = 8) were implanted with EEG and EMG electrodes along with E-mitters. Sleep deprivation was performed for 6 h after one full day of baseline by gentle handling. Mbnl1 and Mbnl2 KOs had normal amounts, and natural diurnal distributions, of wakefulness and NREM sleep in the 12:12 LD condition. Shorter sleep latency and modest wake fragmentation during dark periods (frequent/shorter wake episodes) were observed in the Mbnl2 KO mice. However, the most profound sleep phenotypes observed in the Mbnl2 KO mice were an increase in REM sleep amounts, associated with increased numbers of REM sleep episodes, and increased EEG theta power. This change was most notable during the dark/active period. No direct transition from Wake to REM sleep, an EEG/EMG phenotype equivalent to behavioral cataplexy, was seen in either WT or Mbnl2 KO mice. A change in REM sleep in Mbnl2 KOs was also observed during rebound sleep after a 6-h sleep deprivation period initiated at ZT 0, where a more profound REM sleep rebound was observed in Mbnl2 KO mice, compared to WT mice. These REM specific changes were not observed in the Mbnl1 KO mice. Our results indicate that Mbnl2, but not Mbnl1, KO mice exhibit increased REM sleep propensity, suggesting REM-associated sleep abnormality in Mbnl2 KO are caused by dysregulation of specific alternative splicing events in the brain, and this may be one of the most important sleep abnormalities in DM. REM sleep characteristics in DM might be a residual of infant-type REM sleep, as infants of altricial species, including human and mice, spend large majority of time in REM sleep.

Age of Onset of RNA Toxicity Influences Phenotypic Severity: Evidence from an Inducible Mouse Model of Myotonic Dystrophy (DM1)

Myotonic dystrophy type 1 (DM1) is the most common muscular dystrophy in adults. It is caused by an expanded (CTG)n tract in the 3′ UTR of the Dystrophia Myotonica Protein Kinase (DMPK) gene. This causes nuclear retention of the mutant mRNA into ribonuclear foci and sequestration of interacting RNA-binding proteins (such as muscleblind-like 1 (MBNL1)). More severe congenital and childhood-onset forms of the disease exist but are less understood than the adult disease, due in part to the lack of adequate animal models. To address this, we utilized transgenic mice over-expressing the DMPK 3′ UTR as part of an inducible RNA transcript to model early-onset myotonic dystrophy. In mice in which transgene expression was induced during embryogenesis, we found that by two weeks after birth, mice reproduced cardinal features of myotonic dystrophy, including myotonia, cardiac conduction abnormalities, muscle weakness, histopathology and mRNA splicing defects. Notably, these defects were more severe than in adult mice induced for an equivalent period of exposure to RNA toxicity. Additionally, the utility of the model was tested by over-expressing MBNL1, a key therapeutic strategy being actively pursued for treating the disease phenotypes associated with DM1. Significantly, increased MBNL1 in skeletal muscle partially corrected myotonia and splicing defects present in these mice, demonstrating the responsiveness of the model to relevant therapeutic interventions. Furthermore, these results also represent the first murine model for early-onset DM1 and provide a tool to investigate the effects of RNA toxicity at various stages of development.

Longitudinal in vivo muscle function analysis of the DMSXL mouse model of myotonic dystrophy type 1

Myotonic dystrophy is the most common adult muscle dystrophy. In view of emerging therapies, which use animal models as a proof of principle, the development of reliable outcome measures for in vivo longitudinal study of mouse skeletal muscle function is becoming crucial. To satisfy this need, we have developed a device to measure ankle dorsi- and plantarflexion torque in rodents. We present an in vivo 8-month longitudinal study of the contractile properties of the skeletal muscles of the DMSXL mouse model of myotonic dystrophy type 1. Between 4 and 12months of age, we observed a reduction in muscle strength in the ankle dorsi- and plantarflexors of DMSXL compared to control mice although the strength per muscle cross-section was normal. Mild steady myotonia but no abnormal muscle fatigue was also observed in the DMSXL mice. Magnetic resonance imaging and histological analysis performed at the end of the study showed respectively reduced muscle cross-section area and smaller muscle fibre diameter in DMSXL mice. In conclusion, our study demonstrates the feasibility of carrying out longitudinal in vivo studies of muscle function over several months in a mouse model of myotonic dystrophy confirming the feasibility of this method to test preclinical therapeutics.

Respiratory failure in a mouse model of myotonic dystrophy does not correlate with the CTG repeat length

Myotonic dystrophy (DM1) is a multisystemic disease caused by an expansion of CTG repeats in the region of DMPK, the gene encoding DM protein kinase. The severity of muscle disability in DM1 correlates with the size of CTG expansion. As respiratory failure is one of the main causes of death in DM1, we investigated the correlation between respiratory impairment and size of the (CTG)n repeat in DM1 animal models. Using pressure plethysmography the respiratory function was assessed in control and transgenic mice carrying either 600 (DM600) or >1300 CTG repeats (DMSXL). The statistical analysis of respiratory parameters revealed that both DM1 transgenic mice sub-lines show respiratory impairment compared to control mice. In addition, there is no significant difference in breathing functions between the DM600 and DMSXL mice.In conclusion, these results indicate that respiratory impairment is present in both transgenic mice sub-lines, but the severity of respiratory failure is not related to the size of the (CTG)n expansion.

453. Systemic Delivery of a Peptide-Linked Morpholino Oligonucleotide Neutralizes Mutant RNA Toxicity in a Mouse Model of Myotonic Dystrophy

453. Systemic Delivery of a Peptide-Linked Morpholino Oligonucleotide Neutralizes Mutant RNA Toxicity in a Mouse Model of Myotonic Dystrophy

RNA Interference Targeting CUG Repeats in a Mouse Model of Myotonic Dystrophy

Myotonic dystrophy type 1 (DM1) is an RNA dominant disease caused by expression of DM protein kinase (DMPK) transcripts that contain an expanded CUG repeat (CUG(exp)). The toxic mRNA localizes to nuclear foci and sequesters proteins involved in the regulation of alternative splicing, such as, muscleblind-like 1 (MBNL1). Here, we used synthetic short interfering RNAs (siRNAs) to target CUG repeats and test the concept that inhibiting the expression of CUG(exp) RNA can mitigate features of DM1 in transgenic mice. Intramuscular injection and electroporation of siRNA resulted in ~70-80% downregulation of CUG(exp) transcripts. A limited survey of endogenous mouse transcripts that contain nonexpanded CUG or CAG repeats showed that most were not affected, though Txlnb containing (CUG)(9) was significantly reduced. By this strategy, the number and intensity of CUG(exp) nuclear foci were reduced and splicing of MBNL1-dependent exons was improved. These data suggest that the expanded CUG repeats are a potential target for allele-selective RNA interference.

Systemic Delivery of a Peptide-Linked Morpholino Oligonucleotide Neutralizes Mutant RNA Toxicity in a Mouse Model of Myotonic Dystrophy

Expansions of CUG trinucleotide sequences in RNA transcripts provide the basis for toxic RNA gain-of-function that leads to detrimental changes in RNA metabolism. A CTG repeat element normally resides in the 3' untranslated region of the dystrophia myotonica-protein kinase (DMPK) gene, but when expanded it is the genetic lesion of myotonic dystrophy type 1 (DM1), a hereditary neuromuscular disease. The pathogenic DMPK transcript containing the CUG expansion is retained in ribonuclear foci as part of a complex with RNA-binding proteins such as muscleblind-like 1 (MBNL1), resulting in aberrant splicing of numerous RNA transcripts and consequent physiological abnormalities including myotonia. Herein, we demonstrate molecular and physiological amelioration of the toxic effects of mutant RNA in the HSA(LR) mouse model of DM1 by systemic administration of peptide-linked morpholino (PPMO) antisense oligonucleotides bearing a CAG repeat sequence. Intravenous administration of PPMO conjugates to HSA(LR) mice led to redistribution of Mbnl1 protein in myonuclei and corrections in abnormal RNA splicing. Additionally, myotonia was completely eliminated in PPMO-treated HSA(LR) mice. These studies provide proof of concept that neutralization of RNA toxicity by systemic delivery of antisense oligonucleotides that target the CUG repeat is an effective therapeutic approach for treating the skeletal muscle aspects of DM1 pathology.

Functional and histopathological identification of the respiratory failure in a DMSXL transgenic mouse model of Myotonic Dystrophy

SUMMARYAcute and chronic respiratory failure is one of the major and potentially life-threatening features in individuals with myotonic dystrophy type 1 (DM1). Despite several clinical demonstrations showing respiratory problems in DM1 patients, the mechanisms are still not completely understood. This study was designed to investigate whether the DMSXL transgenic mouse model for DM1 exhibits respiratory disorders and, if so, to identify the pathological changes underlying these respiratory problems. Using pressure plethysmography, we assessed the breathing function in control mice and DMSXL mice generated after large expansions of the CTG repeat in successive generations of DM1 transgenic mice. Statistical analysis of breathing function measurements revealed a significant decrease in the most relevant respiratory parameters in DMSXL mice, indicating impaired respiratory function. Histological and morphometric analysis showed pathological changes in diaphragmatic muscle of DMSXL mice, characterized by an increase in the percentage of type I muscle fibers, the presence of central nuclei, partial denervation of end-plates (EPs) and a significant reduction in their size, shape complexity and density of acetylcholine receptors, all of which reflect a possible breakdown in communication between the diaphragmatic muscles fibers and the nerve terminals. Diaphragm muscle abnormalities were accompanied by an accumulation of mutant DMPK RNA foci in muscle fiber nuclei. Moreover, in DMSXL mice, the unmyelinated phrenic afferents are significantly lower. Also in these mice, significant neuronopathy was not detected in either cervical phrenic motor neurons or brainstem respiratory neurons. Because EPs are involved in the transmission of action potentials and the unmyelinated phrenic afferents exert a modulating influence on the respiratory drive, the pathological alterations affecting these structures might underlie the respiratory impairment detected in DMSXL mice. Understanding mechanisms of respiratory deficiency should guide pharmaceutical and clinical research towards better therapy for the respiratory deficits associated with DM1.

O.7 Systemic delivery of RNase H-active antisense oligonucleotides reverses RNA dominance in a mouse model of myotonic dystrophy

O.7 Systemic delivery of RNase H-active antisense oligonucleotides reverses RNA dominance in a mouse model of myotonic dystrophy