FSHD is linked to the misexpression of the DUX4 gene contained within the D4Z4 repeat array on chromosome 4. The gene encodes the DUX4 protein, a cytotoxic transcription factor that presumably causes the symptoms of the disease. However, individuals have been identified who express DUX4 in their muscle biopsies, but who remain asymptomatic, suggesting that there are other factors that modify FSHD penetrance or severity. We hypothesized that an FSHD-modifying factor would physically interact with DUX4, and we took a proteomic approach to identify DUX4-interacting proteins. We identified the multifunctional C1QBP protein as one such factor. C1QBP is known to regulate several processes that DUX4 affects, including gene expression, oxidative stress, apoptosis, and pre-mRNA splicing. We used siC1QBP knockdown assays to determine if C1QBP affects DUX4 activity. While C1QBP had little effect on DUX4 activity in myotubes, we found that it inhibits the kinetics of DUX4-target gene activation during myogenic differentiation. This identifies C1QBP as a regulator of DUX4 activity and a potential target for FSHD therapeutics. Importantly, C1QBP is regulated by binding to the signaling molecule hyaluronic acid (HA). Decreasing HA by treating cells with 4-methylumbelliferone (4MU), an inhibitor of HA synthesis, resulted in a sharp decline in DUX4 activity and also greatly reduced its cytotoxicity. We have found that DUX4-induced cytotoxicity is associated with severe mislocalizaton of C1QBP, which is prevented by 4MU. This defect is not a downstream result of DUX4-induced oxidative stress, as it could not be prevented by treating cells with an antioxidant, nor could it be recapitulated by exposing cells to oxidants. This identifies C1QBP as a target for the treatment of FSHD, and in particular indicates that 4MU, already an approved drug in Europe and currently under investigation for other indications, may be an effective C1QBP-targeting FSHD therapeutic compound.

INTRODUCTION: We provide a comprehensive overview of bone health in facioscapulohumeral muscular dystrophy (FSHD). METHODS: Ninety-four adult individuals with FSHD1 from two sites were included in this cross-sectional study. Clinical characteristics and determinants of bone health were examined. Relationships between bone mineral density (BMD), strength and function were explored. RESULTS: Nearly a third of subjects were deficient in vitamin D3. Mean whole body BMD z-score was -0.7; 11% had greater than age-related reductions in whole body BMD (z-score < -2.0). Whole body and regional BMD were associated with strength and function. Thirty-six percent had a history of fractures. Likelihood for fractures was reduced for those with normal whole body BMD (OR=0.25, 95% CI: 0.04-0.78). DISCUSSION: A diagnosis of FSHD is not necessarily predictive of reduced BMD or increased fracture rate. Given the considerable variability of bone health in the FSHD population, strength and function can serve as predictors of BMD. This article is protected by copyright.

Facioscapulohumeral Muscular Dystrophy (FSHD) is an autosomal dominant neuromuscular disease affecting 1 in 20,000 to 1 in 15,000 individuals and is characterized by progressive weakness in the facial, scapular, humeral, truncal, and lower extremity muscles (Tawil and Van Der Maarel Muscle Nerve 2006). FSHD is associated with the contraction of the D4Z4 microsatellite repeat below a threshold number of repeats (Wijmenga et al., Nat. Genet, 1992), allowing the transcription of the DUX4 gene contained within the last repeat (Snider et al., PLoS Gen, 2010). The disease only develops when DUX4 is expressed from a chromosome with the permissive 4qA allele, which contains a polyadenylation signal (PAS) that stabilizes the DUX4 transcript (Lemmers et al., Science, 2010). We are using CRISPR technology to investigate the possibility that disruption of the PAS in cells derived from FSHD patients could prevent expression of the DUX4 protein and restore the cell to a less affected phenotype. We will then take advantage of the high reprogramming efficiency of FSHD cells and generate iPSC from FSHD muscle cells with the repressed DUX4 allele, and determine if they have a similar phenotype to iPS cells derived from non-affected individuals. Finally, we will use the highly-engraftable iPS cells in xenograft experiments to determine if the DUX4-silenced iPSCs repopulate injured muscle more efficiently than unaltered FSHD-derived iPSC, and evaluate their potential for use as therapeutics.