Double homeobox 4 activates germline genes, endogenous retroelements and immune modulators: Implications for facioscapulohumeral muscular dystrophy

Abstract

Double homeobox 4 (DUX4) is a candidate disease gene for causing facioscapulohumeral dystrophy (FSHD), a condition characterized by progressive degeneration of specific skeletal muscle groups. While the genetic lesion associated with the disease has been known for decades, the molecular mechanism(s) leading to muscular dystrophy remained unclear. DUX4 was detected in FSHD muscle, but the abundance was extremely low and the downstream consequence of DUX4 in human muscle cells was not known. This work demonstrates that the low level of DUX4 expression in a population of FSHD muscle cells represents a relatively high expression of DUX4 mRNA and protein in a subset of cells at any given point in time. While the full length DUX4 induces toxicity in muscle cells, the shorter isoform without the C-terminal end, DUX4-s, exhibited no obvious detrimental effects on muscle cells. Genome-wide binding studies revealed that DUX4 protein binds to a DNA target sequence that contains two closely-spaced canonical homeodomain binding sequences TAAT in tandem. These binding sites are present in both unique regions of the genome as well as within LTRs of a family of endogenous retrotransposons called MaLR. DUX4-s can bind to the same sequences, but, interestingly, DUX4-s was unable to activate transcription in luciferase reporter assays and, in fact, could act as a dominant negative to inhibit full length DUX4's activity. Microarray expression analysis showed that DUX4 can upregulate over a thousand genes involved in multiple processes, such as germ cell development, RNA splicing and immune modulation. These targets are indeed upregulated in cultured FSHD muscle cells as well as biopsies, providing further support for the causal role of DUX4 in FSHD. One of the DUX4 targets, DEFB103, not only suppresses the innate immune response to viral infection but also inhibits myogenic differentiation. Therefore, FSHD represents a disease where incomplete developmental silencing of a retrogene DUX4 results in inappropriate expression of its targets in skeletal muscle cells. These findings suggest specific mechanisms of FSHD pathology and identify candidate biomarkers for disease diagnosis and progression.