*3.4. Facioscapulohumeral Muscular Dystrophy (FSHD)-Involvement of Tandem Repeats and Long ncRNA*

Repeat elements in primate genomes are primarily of two types: repeats of transposable element integration into the genome and tandem repeats (microsatellites, minisatellites and macrosatellites). There is evidence that ~25 percent of minisatellite tandem repeats are derived from TEs [18]. The evolutionary origin of most macrosatellites is uncertain, however the macrosatellite tandem repeat at human chromosome region 4q35 discussed below has an interesting origin and possibly related to a retrotransposition. It represents a remnant of an unprocessed mRNA from a primate ancestral retrogene. The retrogene was lost but the retrotranscribed unprocessed mRNA (which included introns) was retained. It proceeded through an expansion in the last 25 million years and was greatly multi-copied in Homo sapiens [40–43].

The tandem repeat macrosatellite sequences at chromosome region 4q35 and a chromatin associated long ncRNA are intimately involved in a human genetic disorder termed facioscapulohumeral muscular dystrophy (FSHD) [17,44]. This is a process perhaps best described as a "cause and effect" mechanism involving loss of repeat sequences, the activation of a lncRNA and resultant epigenetic changes in FAHD patients.

Normal individuals carry a certain number of repeats (11–100) in the facioscpulohumeral (FSHD) locus. FSHD patients have a decreased number of repeats [17]. The FSHD locus maps to the chromosomal region 4q35. The region containing the repeat sequences is termed D4Z4 (Figure 4). A lncRNA termed DBE-T is partly encoded by the D4Z4 repeat locus at its 3' end. The genomic repeat sequences serve as binding sites for Polycomb proteins (PcG). PcG proteins have multiple functions but are epigenetic suppressors and are needed to suppress the *FSHD* genes that are silent in normal individuals. However, when there is a loss of genomic repeats resulting in less than 11 repeats, as is found in FSHD patients, DBE-T lncRNA transcription is activated. The RNA is normally silent and is only detected in FSHD patients. DBE-T lncRNA recruits Ash1L, a histone methytransferase to the FHDS locus, which subsequently results in transcription of locus genes due to epigenetic changes in chromatin. Thus these are changes involving the loss of repeat sequences in the D4Z4 region, transcription of a lncRNA with resultant chromatin remodeling and transcription of normally silent genes from the FSHD locus (Figure 4).

**Figure 4.** Involvement of repeat units (in orange) and lncRNA (in red) in FSHD muscular dystrophy. DBE-T, lncRNA; TRG, Trithorax proteins; other terms are defined in text**.** From Casa and Gabellini [17]. Reprinted with permission of publisher.

This fascinating study raises questions regarding the cell's reliance of repeat sequences for crucial epigenetic regulation. The D4Z4 region is highly variable in normal individuals, again, 11–100 copies of repeats. Genetic rearrangements appear to have take place in the 4q35 region in FSHD patients [45]. It seems that the cell's reliance on variable number of repeat sequences, as in the D4Z4 region for regulation appears to be a flawed mechanism used for epigenetic silencing. There are also other genetic abnormalities involving deletions in unstable chromosomal regions that have repeat sequences, as in chromosome locus 22q11.2 in DiGeorge Syndrome [46]. Although these repeat regions widely differ in structure and properties, e.g., DiGeorge syndrome can involve rearrangements between palindromic AT-rich repeats, it appears that genomic repeat sequences may constitute weak points in terms of maintenance of genomic fidelity, even though they have their important functions in the cell.
