A nuclear matrix attachment site in the 4q35 locus has an enhancer-blocking activity in vivo: implications for the facio-scapulo-humeral dystrophy.

Facio-scapulo-humeral muscular dystrophy (FSHD) is an autosomal dominant neuromuscular disease with a prevalence of 1 in 20,000 (Lunt and Harper 1991). FSHD is characterized by progressive weakness and atrophy of muscles of the face, upper arms, and shoulder girdle. The disorder is associated with a shortened repeat array that remains present at a subtelomeric position on chromosome 4q after deletion of an integral number of 3.3-kb tandem repeats. The size of the D4Z4 polymorphic locus varies in normal individuals from 35 to 300 kb but is consistently less than 35 kb in length in FSHD patients (van Deutekom et al. 1993). D4Z4 elements have been shown to contain a cryptic DUX4 gene potentially coding for a double homeodomain protein (van Geel et al. 1999). An overall perturbation of mRNA expression profiles and protein content has been observed in FSHD patients (Tupler et al. 1999; Winokur et al. 2003; Wohlgemuth et al. 2003; Laoudj-Chenivesse et al. 2005). However, since no gene has been found altered by the intrachromosomal deletion, the mechanism leading to FSHD remains unexplained and alternative hypotheses are clearly needed. In eukaryotic nuclei and metaphase chromosomes, the DNA is organized into loop domains (for review, see Vassetzky et al. 2000), which are anchored to the nuclear skeleton or matrix via specific sequences called S/MARs, for scaffold/matrix-associated regions (Mirkovitch et al. 1984; Cockerill and Garrard 1986). Generally A/T-rich, S/MARs are DNA fragments between 200 and 1000 bp in length. Some S/MARs are present in nontranscribed regions or within introns. Others found in the vicinity of enhancers, insulators, replication origins, or transcribed genes are defined as function-related (Vassetzky et al. 2000). Previously, we demonstrated the existence of a nuclear matrix attachment site (S/MAR) in the immediate vicinity of the D4Z4 repeat (Fig. 1). We then demonstrated that the S/MAR adjacent to the D4Z4 array was prominent in normal human myoblasts and nonmuscular human cells but much weaker in muscle cells derived from FSHD patients. We also reported that the D4Z4 repeat array and upstream genes reside in a single loop in FSHD myoblasts but are located in two distinct loops in nonmuscular cells and normal human myoblasts (Petrov et al. 2006). In FSHD muscle, the decreased number of D4Z4 repeats results in an inappropriate up-regulation of adjacent 4q35 genes, including FRG1, FRG2, and SLC25A4 (previously known as ANT1) (Gabellini et al. 2002; Rijkers et al. 2004; Laoudj-Chenivesse et al. 2005). Interestingly, overexpressing FRG1 in transgenic mice provokes an FSHD-like phenotype (Gabellini et al. 2005). It has also been proposed that overexpression would result from the absence of a D4Z4-binding transcriptional repressor complex (Gabellini et al. 2002). Other hypotheses include the existence within D4Z4 of a transcriptional enhancer, which could up-regulate the expression of FRG2 (Petrov et al. 2003; Rijkers et al. 2004). The presence of a S/MAR (referred to as FR-MAR) between the D4Z4 array and proximal genes raises the question as to whether it functions as a border element insulating adjacent genes from the effect of D4Z4 in normal human cells. Indeed, within a given chromatin domain, border elements have been shown to protect genes from stimulatory and repressive effects exerted by flanking genomic regions (for review, see Gaszner and Felsenfeld 2006), and S/MARs have previously been characterized as implicated in bordering functions (Bode et al. 2000). Figure 1. The D4Z4 repeat contains a strong transcriptional enhancer. (A) Schematic representation of the D4Z4 repeat in its chromosome 4q35 environment. The SLC25A4 (previously known as ANT1), FRG1, and FRG2 genes are shown as gray rectangles. Telomeric and D4Z4 ... Here, we have studied in detail the ability of the D4Z4 repeat to regulate transcriptional activity. We have found evidence of an exceptionally strong transcriptional enhancer at the 5′-end of the repeat. This enhancer could up-regulate transcription from the FRG1 gene promoter. We have also found that the S/MAR located in the vicinity of the D4Z4 array exerts an enhancer blocking activity in vivo, thus insulating the FRG1 and FRG2 genes from the effect of D4Z4. We propose a model whereby this S/MAR regulates chromatin accessibility and expression of genes implicated in the genesis of FSHD.