Evaluating the effects of CELF1 deficiency in a mouse model of RNA toxicity

Human Molecular Genetics Evaluating the effects of CELF1 deficiency in a mouse model of RNA toxicity Yun Kyoung Kim 1 2 Mahua Mandal 1 2 Ramesh S. Yadava 1 2 Luc Paillard 0 1 Mani S. Mahadevan 1 2 0 Institut de Ge ne tique et 1 De veloppement de Rennes, Universite de Rennes 1 , Rennes F-35000 , France 2 Department of Pathology, University of Virginia , Charlottesville, VA 22908 , USA Myotonic dystrophy type 1 (DM1), the most common form of adult-onset muscular dystrophy, is caused by an expanded (CTG)n repeat in the 3 untranslated region of the DM protein kinase (DMPK) gene. The toxic RNA transcripts produced from the mutant allele alter the function of RNA-binding proteins leading to the functional depletion of muscleblind-like (MBNL) proteins and an increase in steady state levels of CUG-BP1 (CUGBP-ETR-3 like factor 1, CELF1). The role of increased CELF1 in DM1 pathogenesis is well studied using genetically engineered mouse models. Also, as a potential therapeutic strategy, the benefits of increasing MBNL1 expression have recently been reported. However, the effect of reduction of CELF1 is not yet clear. In this study, we generated CELF1 knockout mice, which also carry an inducible toxic RNA transgene to test the effects of CELF1 reduction in RNA toxicity. We found that the absence of CELF1 did not correct splicing defects. It did however mitigate the increase in translational targets of CELF1 (MEF2A and C/EBPb). Notably, we found that loss of CELF1 prevented deterioration of muscle function by the toxic RNA, and resulted in better muscle histopathology. These data suggest that while reduction of CELF1 may be of limited benefit with respect to DM1-associated spliceopathy, it may be beneficial to the muscular dystrophy associated with RNA toxicity. - Myotonic dystrophy type 1 (DM1) is the most common cause of adult-onset muscular dystrophy with an incidence of 1 in 8000 and is a multisystemic disorder (1). Major features of this autosomal dominant disorder are myotonia, muscle weakness, atrophy, smooth muscle dysfunction, cardiac defects and insulin resistance (2). DM1 is triggered by the expanded (CTG)n triple repeat in the 3-untranslated region (UTR) of the DM protein kinase (DMPK) gene (1). This is transcribed to produce mutant RNAs containing CUG repeats that are retained in the nucleus to form RNA foci (3,4). This affects the nuclear and cytoplasmic activities of RNA-binding proteins such as muscleblind-like 1 (MBNL1) and CELF1 (5 8). MBNL1 binds to the expanded CUG repeat and co-localizes with the RNA foci, causing a local reduction of MBNL1 (6,7). Consequently, the activity of MBNL1 as a splicing regulator is impaired, resulting in aberrant alternative splicing of target genes (9). Consistent with the loss of MBNL1 function, an analysis of MBNL1 knockout mice (MbnlDE3/DE3) showed that these mice developed some of the characteristic features of DM1, including misregulated mRNA splicing, muscle histopathological changes, cataracts and myotonia (10). Moreover, an adenoviral delivery of MBNL1 reversed the splicing changes and myotonia, again underscoring the importance of MBNL1 in the DM1 disease phenotype (11). In addition to MBNL1, CELF1, a member of CUG-BP and ETR-3-like factor (CELF) family, is implicated in the disease process. CELF1 has been reported to have multiple functions in RNA metabolism including regulation of alternative splicing, RNA stability and translational regulation of its RNA targets (12 16). The mutant DMPK transcript is thought to activate PKC activity, leading to a hyperphosphorylation of CELF1 (17), resulting in the stabilization of and increased steady state levels of CELF1 in DM1 skeletal muscle and heart tissues. Transgenic mice overexpressing CELF1 showed neonatal lethality, skeletal muscle histological changes and misregulated splicing pattern observed in DM1 patients (18,19). Heart specific overexpression of CELF1 caused premature lethality, histopathological and echocardiographic abnormalities, and splicing defects (20). Two different transgenic mouse models which overexpressed CELF1 in skeletal muscle developed DM1 features. One mouse model showed that CELF1 responsive alternative splicing events were misregulated in skeletal muscle and that CELF1 affects muscle integrity and function (21). In the other model, muscular dystrophy, fiber type switching and delayed muscle development were seen in conjunction with increased levels of p21 and MEF2A (both of which are translational targets of CELF1) (18). Both of these models exhibited muscle loss, impaired muscle function and dystrophic muscle histology. These data demonstrate that increased CELF1 contributes to DM1 pathogenesis and suggests that reduction of CELF1 in a DM1 mouse model may have beneficial effects. To test this, we generated CELF1 knockout mice that also carried an inducible toxic RNA transgene (5-313) (22). We found muscle function in these double-transgenic mice was protected from the effects of RNA toxicity and the histopathological features were milder as compared with mice carrying only the toxic RNA transgene, despite the fact that many alternative splicing events known to be misregulated in DM1 were not corrected by the absence of CELF1. This was associated with decreased Nkx2-5 levels in skeletal muscle from Celf12/2 mice expressing the toxic RNA, and corresponded to the milder histopathology of the muscle in these mice. Additionally, the protein levels of MEF2A and C/EBPb were reduced in the absence of CELF1. These data suggest reduction of CELF1 may be beneficial to the muscular dystrophy associated with DM1. Levels of CELF1 correlate with skeletal muscle histopathology in DM1 We first evaluated CELF1 levels in skeletal muscles from patients with DM1. Western blot analyses of muscle extracts showed increased CELF1 (up to 11-fold) that correlated well with muscle histopathology (Fig. 1A and C, Supplementary Material, Fig. S1A). No significant changes were seen in CELF mRNA levels between DM1 patient groups (Supplementary Material, Fig. S1B). Next, we investigated CELF1 in a doxycyclineinducible RNA toxicity mouse model of DM1 (5-313), which expresses an enhanced green fluorescent protein (eGFP) gene fused to the DMPK 3UTR (CTG)5 (22). CELF1 was increased in mice expressing the toxic RNA (between 2- and 6-fold) and the expression was highest in the muscles with the most severe histopathology (Fig. 1B and D, Supplementary Material, Fig. S1A). Thus, the mouse model accurately represented these aspects of the disease phenotype providing a basis for further investigations. Absence of CELF1 does not affect MBNL1 or CELF2 in mice with RNA toxicity To determine whether the reduction of CELF1 will improve the muscle pathology, 5-313 mice were crossed to Celf12/2 mice to generate Celf12/2/5-313+/2 mice and Celf12/2/5-313+/+ mice. The ratio of +/+, +/2 and 2/2 Celf1 mice from the intercrossing Celf1+/2/5-313 mice was different from the expected 1:2:1 Mend (...truncated)