**Facioscapulohumeral Muscular Dystrophy: From Clinical Data to Molecular Genetics and Return**

Monica Salani2, Elisabetta Morini1, Isabella Scionti1 and Rossella Tupler1,2 *1Universita' degli Studi di Modena e Reggio Emilia, 2University of Massachussets Medical School, 1Italy 2USA* 

#### **1. Introduction**

20 Neuromuscular Disorders

Vasyutina, E., Martarelli, B., Brakebusch, C., Wende, H. and Birchmeier, C. (2009) 'The small

Vorgerd, M., van der Ven, P. F., Bruchertseifer, V., Lowe, T., Kley, R. A., Schroder, R.,

myofibrillar myopathy', *American journal of human genetics* 77(2): 297-304. Wang, A., Patrone, L., McDonald, J. A. and Sheppard, D. (1995) 'Expression of the integrin

Wang, H. V., Chang, L. W., Brixius, K., Wickstrom, S. A., Montanez, E., Thievessen, I.,

Wickstrom, S. A., Lange, A., Montanez, E. and Fassler, R. (2010) 'The ILK/PINCH/parvin

Xu, W., Baribault, H. and Adamson, E. D. (1998) 'Vinculin knockout results in heart and brain defects during embryonic development', *Development* 125(2): 327-37. Zaidel-Bar, R., Itzkovitz, S., Ma'ayan, A., Iyengar, R. and Geiger, B. (2007) 'Functional atlas

Zemljic-Harpf, A. E., Miller, J. C., Henderson, S. A., Wright, A. T., Manso, A. M., Elsherif, L.,

Zervas, C. G., Gregory, S. L. and Brown, N. H. (2001) 'Drosophila integrin-linked kinase is

Zhou, A. X., Hartwig, J. H. and Akyurek, L. M. (2010) 'Filamins in cell signaling, transcription and organ development', *Trends in cell biology* 20(2): 113-23. Ziegler, W. H., Liddington, R. C. and Critchley, D. R. (2006) 'The structure and regulation of

Ziober, B. L., Vu, M. P., Waleh, N., Crawford, J., Lin, C. S. and Kramer, R. H. (1993)

*publication of the American Association of Anatomists* 204(4): 421-31.

damage', *The Journal of cell biology* 180(5): 1037-49.

of the integrin adhesome', *Nature cell biology* 9(8): 858-67.

membrane', *The Journal of cell biology* 152(5): 1007-18.

vinculin', *Trends in cell biology* 16(9): 453-60.

87(2): 169-74.

8935-40.

281-91.

7522-37.

268(35): 26773-83.

both hypertrophic and dilated cardiomyopathy', *Molecular genetics and metabolism*

G-proteins Rac1 and Cdc42 are essential for myoblast fusion in the mouse', *Proceedings of the National Academy of Sciences of the United States of America* 106(22):

Lochmuller, H., Himmel, M., Koehler, K., Furst, D. O. et al. (2005) 'A mutation in the dimerization domain of filamin c causes a novel type of autosomal dominant

subunit alpha 9 in the murine embryo', *Developmental dynamics : an official* 

Schwander, M., Muller, U., Bloch, W., Mayer, U. et al. (2008) 'Integrin-linked kinase stabilizes myotendinous junctions and protects muscle from stress-induced

complex: the kinase is dead, long live the pseudokinase!', *The EMBO journal* 29(2):

Dalton, N. D., Thor, A. K., Perkins, G. A., McCulloch, A. D. et al. (2007) 'Cardiacmyocyte-specific excision of the vinculin gene disrupts cellular junctions, causing sudden death or dilated cardiomyopathy', *Molecular and cellular biology* 27(21):

required at sites of integrin adhesion to link the cytoskeleton to the plasma

'Alternative extracellular and cytoplasmic domains of the integrin alpha 7 subunit are differentially expressed during development', *The Journal of biological chemistry*

Facioscapulohumeral muscular dystrophy (FSHD or Dejerine–Landouzy muscular dystrophy, OMIM #158900) is the third most common hereditary myopathy, with prevalence of 1 in 20,000 (Padberg, 1982; Mostacciuolo et al., 2009). This disease is characterized by the progressive wasting of a highly selective set of muscle groups (Padberg, 1982) and it has been traditionally classified as an autosomal dominant trait (Lunt, 1998; Padberg, 1992). FSHD genetic locus has been mapped on chromosome 4q35 by genetic linkage analysis (Wijmenga et al., 1990). Interestingly, this muscular dystrophy has not yet been related to a classical mutation within a protein-coding gene, but rather the disease has been associated with DNA rearrangements in a polymorphic genomic region consisting of an array of tandemly repeated 3.3 kb segments, named D4Z4 (Wijmenga et al., 1992b). D4Z4 contains an ORF encoding a putative homeobox protein called "DUX4." The existence of native transcripts of DUX4 from D4Z4 single repeats is still controversial (Gabriels et al., 1999; Hewitt et al., 1994; Lyle et al., 1995), although recent data show evidences of the presence of the DUX4 transcript from the last D4Z4 unit in FSHD myoblasts (Lemmers et al., 2010a). The number of D4Z4 repeats varies from 11 to 100 in the general population, whereas less than 11 repeats are usually present in sporadic and familial FSHD patients. A very low copy number of 4q35 D4Z4 repeats (1–3) often correlates with an earlier onset and more severe disease. However no FSHD-linked array has been found to have zero copies of the repeat unit (Tupler et al., 1996; van der Maarel et al., 2007) suggesting that the repeat itself plays a critical role in the disease. Alleles with 4-7 repeats are the most frequent in the FSHD population and are associated with the more common form of FSHD that usually presents in adulthood, whereas alleles with 8-10 repeats typically display milder disease phenotypes with reduced penetrance. Nearly identical and equally polymorphic D4Z4 sequences reside on the subtelomere of chromosome 10q (Bakker et al., 1995; Deidda et al., 1995). The proportion of individuals in the general population carrying 4q or 10q chromosome ends with repeat arrays entirely or partially transferred between both chromosomes, is considerable (van Deutekom et al., 1996b; Lemmers et al., 1998; van Overveld et al., 2000). Rearrangements between 4q and 10q subtelomeres occur in 20% of

Facioscapulohumeral Muscular Dystrophy: From Clinical Data to Molecular Genetics and Return 23

groups (Padberg, 1982). The onset of the disease is in the second/third decade of life and usually involves weakness of facial mimic muscles. The clinical presentation is characterized by an initially restricted distribution of weakness starting with asymptomatic facial weakness followed by scapular fixator, humeral, truncal, and lower extremity weakness. The onset of lower-extremity weakness is typically in the anterior leg compartment, presenting with footdrop. Extraocular and bulbar muscles are typically spared. Although facial weakness is perhaps the most recognizable aspect of FSHD, affected individuals from otherwise clinically typical families may have no or minimal facial weakness (Flanigan et al., 2004). Weak abdominal muscles result in a protuberant abdomen and contribute to the lumbar lordosis. Lower abdominal muscles are weaker than the upper abdominal muscles, causing a strikingly positive Beevor's sign, a physical finding fairly specific for FSHD. In advanced cases, hip girdle may be as affected or more than shoulder girdle muscles, making difficult the clinical distinction between FSHD and limb-girdle muscular dystrophy. A notable distinctive feature of FSHD is that muscle weakness displays an asymmetric distribution, which does not correlate with the handedness of the individual (Brouwer et al., 1992). The chronology of disease progression is unpredictable; for example, long periods of stability can be followed by sudden and dramatic worsening. In addition, there is a wide variability in the spectrum of disease among patients, ranging from subjects with very mild muscle weakness, who are almost unaware of being affected, to those who are wheelchair-dependent. This variability in disease penetrance was exemplified by a set of monozygotic male twins who carried the same genetic mutation but were affected by FSHD to a dramatically different extent (Tupler et al., 1998). Electromyography and histological analysis reveal non-specific myopathic changes associated, in some cases, with neurologic aspects. The creatine kinase (CK) level can be moderately increased or normal. Thus, diagnosis of FSHD is mainly based on the clinical phenotype in combination with a molecular test (Lunt et al 1995; Lunt, 1998). Ancillary features such as sensorineural deafness or retinal vasculopathy have been also reported in FSHD patients, but they are not to be considered decisive criteria for FSHD diagnosis

(Fitzsimmons et al., 1987; Padberg et al., 1995; Trevisan et al., 2008).

The FSHD genetic defect does not reside in mutation of any protein-coding gene. Instead FSHD has been causally associated with reduction of the number of 3.3 kb DNA elements at the subtelomeric region of chromosome 4q. When digested with the restriction enzyme EcoRI the FSHD allele originates a fragment of 35 kb in size or shorter that can be detected by Southern analysis using the probe p13E-11 (Wijmenga et al., 1992b). The polymorphic genomic region, identified by probe p13E-11, consists of an array of tandem repeat 3.3-kb segments (hereafter referred to as D4Z4 repeats), such that the variation in the size of EcoRI fragments is due to variability in the number of D4Z4 repeats (van Deutekom et al., 1993) (Figure 1). Normal subjects carry p13E−11 EcoRI alleles longer than 50 kb (>10 D4Z4 units) originating from chromosome 4 whereas alleles of 35 kb or shorter (≤ 8 D4Z4 units) are present in the majority of either de novo or familial FSHD patients (Lunt, 1998). *De novo* reduced allele transmitted by an affected parent to the offspring co-segregate with the disorder (Griggs et al., 1993). New mutations account for a surprisingly high percentage of FSHD patients (10%−33%) (Padberg et al., 1995; Zatz et al., 1995). This high incidence can be partly explained by the presence of parental mosaicism for 4q short alleles that has been reported in 19% of *de novo* cases (Upadhyaya et al.; 1995, Köhler et al.; 1996, Lemmers et al.,

**2.2 Molecular diagnosis** 

subjects and represent an additional complication to FSHD molecular diagnosis. However, only subjects with reduced number of D4Z4 repeats on chromosome 4, but not on chromosome 10, develop FSHD. Thus, molecular diagnosis of FSHD is, at this time, based on the analysis of the D4Z4 polymorphic alleles (Lunt, 1998; Tawil et al., 2010). Despite the identification of the genetic defect associated with FSHD, the pathologic effects of the reduction of D4Z4 elements remain largely unknown. The observation of the unique linkage of the disease with chromosome 4, led to the hypothesis that in *cis* DNA elements must be present on D4Z4 reduced alleles to explain disease onset. An element within D4Z4 has been shown to behave as a silencer that provides a binding site for a transcriptional repressing complex (Gabellini et al., 2002). These results suggest a model in which reduction of D4Z4 leads to the inappropriate transcriptional derepression of proximal chromosome 4-specific genes. Indeed, closely located genes such as FSHD Region Gene 2 (*FRG2*), FSHD Region Gene 1 (*FRG1*), and Adenine Nucleotide Translocator (*ANT1*)*,* with high myopathic potential, were observed to be transcriptionally upregulated in FSHD muscle (Gabellini et al., 2002). However, studies testing this model obtained conflicting results with some showing support (Gabellini et al., 2002; Rijkers et al., 2004; Klooster et al., 2009; Bodega et al., 2009), and others not (Winokur et al., 2003b; Jang et al., 2003; Osborne et al., 2007; Masny et al., 2010). These discrepancies prevented a general consensus on the role of 4q35 genes upregulation in FSHD pathogenesis. Recently it has been proposed that FSHD arises from chromosomes bearing a short D4Z4 repeat array associated with a novel polyadenylation signal capable of stabilizing *DUX4* mRNA from the normally transcriptionally inactive DUX4 gene. However, there is reason to believe that this model does not account for all FSHD cases. For example, several studies described FSHD patients carrying full-length D4Z4 alleles that are clinically indistinguishable from FSHD subjects carrying D4Z4 reduced alleles. Conversely, there is a high percent of normal individuals, unrelated to any FSHD patients, with reduced D4Z4 alleles (Scionti et al., 2012b). Most importantly it has been recently showed that the polymorphism adding the polyadenylation signal to *DUX4* RNA, is common in the general population. Thus, just a portion of subjects carrying this molecular signature develops FSHD (Scionti et al., 2012b). Consequently, the number of D4Z4 repeats at 4q35 by itself or in association with a specific haplotype, does not to fully explain FSHD development. The non linear correlation between the number of D4Z4 elements and FSHD expression raises the question whether or when FSHD can be considered a simple autosomal dominant mendelian disorder, or if a more complex picture should be considered in clinical practice. Indeed, recent findings suggest that epigenetic modification at the disease locus, such as DNA methylation (de Greef et al., 2008), histone modifications (Zeng et al., 2009) or chromosomal architecture (Petrov et al., 2006) can be altered in FSHD patients, adding further levels of complexity. Undeniably, the incomplete knowledge of molecular mechanism(s) leading to FSHD onset has hampered the possibility of developing effective therapeutic strategies. This chapter aims to discuss the recent advances in our understanding of FSHD pathology viewed against its clinical heterogeneity, considering all the factors that can contribute to the complexity of this elusive disease.

#### **2. FSHD Clinical features: Role of D4Z4 repeats reduction**

#### **2.1 Clinical features**

Facioscapulohumeral muscular dystrophy is considered an autosomal dominant myopathy characterized by progressive atrophy and weakness of a highly selective set of muscle groups (Padberg, 1982). The onset of the disease is in the second/third decade of life and usually involves weakness of facial mimic muscles. The clinical presentation is characterized by an initially restricted distribution of weakness starting with asymptomatic facial weakness followed by scapular fixator, humeral, truncal, and lower extremity weakness. The onset of lower-extremity weakness is typically in the anterior leg compartment, presenting with footdrop. Extraocular and bulbar muscles are typically spared. Although facial weakness is perhaps the most recognizable aspect of FSHD, affected individuals from otherwise clinically typical families may have no or minimal facial weakness (Flanigan et al., 2004). Weak abdominal muscles result in a protuberant abdomen and contribute to the lumbar lordosis. Lower abdominal muscles are weaker than the upper abdominal muscles, causing a strikingly positive Beevor's sign, a physical finding fairly specific for FSHD. In advanced cases, hip girdle may be as affected or more than shoulder girdle muscles, making difficult the clinical distinction between FSHD and limb-girdle muscular dystrophy. A notable distinctive feature of FSHD is that muscle weakness displays an asymmetric distribution, which does not correlate with the handedness of the individual (Brouwer et al., 1992). The chronology of disease progression is unpredictable; for example, long periods of stability can be followed by sudden and dramatic worsening. In addition, there is a wide variability in the spectrum of disease among patients, ranging from subjects with very mild muscle weakness, who are almost unaware of being affected, to those who are wheelchair-dependent. This variability in disease penetrance was exemplified by a set of monozygotic male twins who carried the same genetic mutation but were affected by FSHD to a dramatically different extent (Tupler et al., 1998). Electromyography and histological analysis reveal non-specific myopathic changes associated, in some cases, with neurologic aspects. The creatine kinase (CK) level can be moderately increased or normal. Thus, diagnosis of FSHD is mainly based on the clinical phenotype in combination with a molecular test (Lunt et al 1995; Lunt, 1998). Ancillary features such as sensorineural deafness or retinal vasculopathy have been also reported in FSHD patients, but they are not to be considered decisive criteria for FSHD diagnosis (Fitzsimmons et al., 1987; Padberg et al., 1995; Trevisan et al., 2008).

#### **2.2 Molecular diagnosis**

22 Neuromuscular Disorders

subjects and represent an additional complication to FSHD molecular diagnosis. However, only subjects with reduced number of D4Z4 repeats on chromosome 4, but not on chromosome 10, develop FSHD. Thus, molecular diagnosis of FSHD is, at this time, based on the analysis of the D4Z4 polymorphic alleles (Lunt, 1998; Tawil et al., 2010). Despite the identification of the genetic defect associated with FSHD, the pathologic effects of the reduction of D4Z4 elements remain largely unknown. The observation of the unique linkage of the disease with chromosome 4, led to the hypothesis that in *cis* DNA elements must be present on D4Z4 reduced alleles to explain disease onset. An element within D4Z4 has been shown to behave as a silencer that provides a binding site for a transcriptional repressing complex (Gabellini et al., 2002). These results suggest a model in which reduction of D4Z4 leads to the inappropriate transcriptional derepression of proximal chromosome 4-specific genes. Indeed, closely located genes such as FSHD Region Gene 2 (*FRG2*), FSHD Region Gene 1 (*FRG1*), and Adenine Nucleotide Translocator (*ANT1*)*,* with high myopathic potential, were observed to be transcriptionally upregulated in FSHD muscle (Gabellini et al., 2002). However, studies testing this model obtained conflicting results with some showing support (Gabellini et al., 2002; Rijkers et al., 2004; Klooster et al., 2009; Bodega et al., 2009), and others not (Winokur et al., 2003b; Jang et al., 2003; Osborne et al., 2007; Masny et al., 2010). These discrepancies prevented a general consensus on the role of 4q35 genes upregulation in FSHD pathogenesis. Recently it has been proposed that FSHD arises from chromosomes bearing a short D4Z4 repeat array associated with a novel polyadenylation signal capable of stabilizing *DUX4* mRNA from the normally transcriptionally inactive DUX4 gene. However, there is reason to believe that this model does not account for all FSHD cases. For example, several studies described FSHD patients carrying full-length D4Z4 alleles that are clinically indistinguishable from FSHD subjects carrying D4Z4 reduced alleles. Conversely, there is a high percent of normal individuals, unrelated to any FSHD patients, with reduced D4Z4 alleles (Scionti et al., 2012b). Most importantly it has been recently showed that the polymorphism adding the polyadenylation signal to *DUX4* RNA, is common in the general population. Thus, just a portion of subjects carrying this molecular signature develops FSHD (Scionti et al., 2012b). Consequently, the number of D4Z4 repeats at 4q35 by itself or in association with a specific haplotype, does not to fully explain FSHD development. The non linear correlation between the number of D4Z4 elements and FSHD expression raises the question whether or when FSHD can be considered a simple autosomal dominant mendelian disorder, or if a more complex picture should be considered in clinical practice. Indeed, recent findings suggest that epigenetic modification at the disease locus, such as DNA methylation (de Greef et al., 2008), histone modifications (Zeng et al., 2009) or chromosomal architecture (Petrov et al., 2006) can be altered in FSHD patients, adding further levels of complexity. Undeniably, the incomplete knowledge of molecular mechanism(s) leading to FSHD onset has hampered the possibility of developing effective therapeutic strategies. This chapter aims to discuss the recent advances in our understanding of FSHD pathology viewed against its clinical heterogeneity, considering all

the factors that can contribute to the complexity of this elusive disease.

**2. FSHD Clinical features: Role of D4Z4 repeats reduction** 

Facioscapulohumeral muscular dystrophy is considered an autosomal dominant myopathy characterized by progressive atrophy and weakness of a highly selective set of muscle

**2.1 Clinical features** 

The FSHD genetic defect does not reside in mutation of any protein-coding gene. Instead FSHD has been causally associated with reduction of the number of 3.3 kb DNA elements at the subtelomeric region of chromosome 4q. When digested with the restriction enzyme EcoRI the FSHD allele originates a fragment of 35 kb in size or shorter that can be detected by Southern analysis using the probe p13E-11 (Wijmenga et al., 1992b). The polymorphic genomic region, identified by probe p13E-11, consists of an array of tandem repeat 3.3-kb segments (hereafter referred to as D4Z4 repeats), such that the variation in the size of EcoRI fragments is due to variability in the number of D4Z4 repeats (van Deutekom et al., 1993) (Figure 1). Normal subjects carry p13E−11 EcoRI alleles longer than 50 kb (>10 D4Z4 units) originating from chromosome 4 whereas alleles of 35 kb or shorter (≤ 8 D4Z4 units) are present in the majority of either de novo or familial FSHD patients (Lunt, 1998). *De novo* reduced allele transmitted by an affected parent to the offspring co-segregate with the disorder (Griggs et al., 1993). New mutations account for a surprisingly high percentage of FSHD patients (10%−33%) (Padberg et al., 1995; Zatz et al., 1995). This high incidence can be partly explained by the presence of parental mosaicism for 4q short alleles that has been reported in 19% of *de novo* cases (Upadhyaya et al.; 1995, Köhler et al.; 1996, Lemmers et al.,

Facioscapulohumeral Muscular Dystrophy: From Clinical Data to Molecular Genetics and Return 25

Through years, additional findings have emerged that need to be considered in the molecular diagnosis of FSHD. First, translocated 4-type repeats residing on chromosome 10q as well as translocated 10-type repeats residing on chromosome 4q are found in 10% of the population (van Deutekom et al., 1996b; van Overveld et al., 2000; Matsumura et al., 2002;). Therefore, FSHD-sized D4Z4 alleles may be attributed incorrectly to chromosome 10 and viceversa. Second, deletions at 4q encompassing the genomic sequence recognized by probe p13E-11 have been detected in FSHD cases. Thus D4Z4 short arrays might not be detected by using the standard diagnostic procedure. The frequency of such extended deletions has been estimated around 3% (Lemmers et al., 2003) and represents a possible caveat of FSHD molecular diagnosis. Third, 5-10% of subjects showing FSHD clinical features do not carry D4Z4 reduced alleles. Possible explanations for such anomalous cases include a different mechanism at 4q35, such as D4Z4 hypomethylation (De Greef et al, 2009) or the presence of other mutations not linked to the FSHD locus at 4q35. At present, no FSHD families linked to other chromosomal loci have been described. Figure 2 summarizes the diagnostic flow

chart that should be used to study the 4q35 region in FSHD patients.

Fig. 2. **Schematic representation of FSHD diagnostic approach at 4q35 locus**.

**N**: restriction enzyme NotI; **H**: restriction enzyme HindIII.

in solution; **PFGE**: Pulse Field Gel Electrophoresis; **LGE**: Linear Gel Electrophoresis; **E**: restriction enzyme EcoRI; **B**: restriction enzyme BlnI; **X**: restriction enzyme XapI;

Abbreviations: **Ch.DNA**: chromosomal DNA embedded in plugs; **Gen.DNA**: genomic DNA

2004). The presence of somatic mosaicism for a rearrangement of D4Z4 was found in as much as 3% of the general population (van der Maarel et al., 2000), suggesting that the D4Z4 repeat is highly recombinogenic.

A complication of molecular testing by Southern analysis is represented by the presence of a polymorphic region recognized by probe p13E-11 at the subtelomeric region of chromosome 10q, which shares numerous homologies with the 4q subtelomere (Bakker et al., 1995; Deidda et al., 1996). The repeat element at 10q is 98% identical to D4Z4 at 4q, and the size of 10q EcoRI alleles varies between 11 and 300 kb (1-100 D4Z4 units). Moreover, 10% of these alleles are shorter than 35 kb (8 D4Z4 units) (Bakker et al., 1996; Bakker et al., 1995), overlapping the 4q alleles. Clearly these overlapping features can interfere with the molecular diagnosis of FSHD. Nevertheless the presence of a BlnI restriction site within the 3.3 kb element associated with chromosome 10q allows the discrimination between 4q and 10q alleles (Deidda et al., 1996). As a result, Southern blot hybridization of EcoRI and EcoRI/BlnI digested genomic DNA is used for the molecular diagnosis of FSHD (Lunt, 1998) (Figure 1).

