DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
(titres de publication, de périodique et noms de colloque inclus)
2003-09-30 - Colloque/Présentation - poster - Anglais - 1 page(s)

Ansseau Eugénie , Sauvage Sébastien, Mattéotti Christel, Marcowyzc Aline, Belayew Alexandra , Coppée Frédérique , "Functional study of the DUX4 gene" in FSHD International Consortium Research Meeting, Los Angeles, USA, 2003

  • Codes CREF : Biologie moléculaire (DI3111), Pathologies particulières (DI3370)
  • Unités de recherche UMONS : Biologie moléculaire (M122)
  • Instituts UMONS : Institut des Sciences et Technologies de la Santé (Santé)

Abstract(s) :

(Anglais) The FSHD candidate gene we are studying (DUX4; Gabriëls et al 1999, Gene 236, 25-32) is located within the D4Z4 unit itself. Shortening of the D4Z4 array would open an inhibitory chromatin and allow DUX4 expression in patient myoblasts. The DUX4 protein has a double homeodomain and locates to the cell nucleus, partly overlapping with emerin and lamin. Hundreds of 3.3-kb elements with homologous DUX genes not linked to FSHD are scattered in the human genome: their expression constitutes a background against which DUX4 expression is difficult to single out. DUX1 is a non pathological protein expressed in vivo from such gene (Ding et al 1998, Hum.Mol.Genet. 7, 1681-94). A rabbit antiserum was raised against a peptide of the DUX4 carboxy-terminal (tail) domain lacking in DUX1 and probably in other DUX proteins. This serum was used to show DUX4 expression from its natural gene promoter: mouse myoblast C2C12 cells were transfected with a pGEM plasmid containing either a 13.5-kb EcoRI genomic fragment with two D4Z4/DUX4 units corresponding to a patient locus (Gabriëls et al 1999), or a 3.3-kb KpnI fragment containing a single DUX4 gene. As a positive control, we used the pCIneo-DUX4 expression vector (DUX4 ORF under control of a strong CMV promoter). Nuclear extracts of these cells were analysed by electrophoresis on SDS-PAGE 10% followed by Western blotting: in each sample a 52 kDa protein was detected with the apparent molecular weight of bona fide DUX4 produced by transcription/translation in vitro. We could not identify DUX4 protein partners with the yeast two hybrid system (MatchMaker, Clontech), because it showed a strong transcriptional activity that could be mapped to its tail domain. In contrast, DUX1 gave no background activation and was used to screen a human muscle cDNA library. Among the 42 positives, 35 encoded desmin. Since the DUX1 double homeodomain is very similar to the DUX4 one, desmin might also interact with DUX4. This is being checked by GST pull down experiments. Desmin is a muscle intermediate filament protein that interacts with nuclear envelope proteins (Stromer and Bendayan 1990, Cell Motil Cytoskeleton. 17, 11-8). It is mutated in some muscular dystrophies and is a plausible DUX partner to be further evaluated. Gabellini et al (2002, Cell 110, 339-348) identified a repressing cis-element in D4Z4 that interacted with YY1 and inhibited expression of a linked reporter gene in HeLa cells. This element precisely mapped within the DUX4 promoter, just 5' from the Sp1 site that mediated its basal expression (Gabriëls et al 1999). We inactivated the YY1 site according to Gabellini et al (2002) in our DUX4-LUC construct, only weakly affecting its transient expression in C2C12 cells. Because this mutation had created a new binding site, we used a different mutation to prevent YY1 binding, and observed a 1.5-fold activation in C2C12 cells. These experiments demonstrated a weak inhibitory role of YY1 on the DUX4 promoter activity in myoblasts. Additionally, a putative MyoD cis-element was found onto the DUX4 transcription start site. This element was shown to be repressing: its mutation increased DUX4-LUC expression 2-3 fold in C2C12 cells and co-transfection with a MyoD expression vector inhibited 4-fold the wild type DUX4-LUC construct. These observations suggested that DUX4 expression was to be expected in cells with low MyoD levels, i.e. myoblasts but not myotubes. In a way to evaluate DUX4 function, we transfected TE671cells with pCIneo-DUX4, extracted nuclear proteins 48 hours later, and detected a 3-fold increase in MEF2 and p53 concentrations. This observation is in agreement with recent data of Winokur et al (2003, Neuromuscul Disord. 4, 322-33) demonstrating premature differentiation of FSHD myoblasts. In addition, a strong decrease in expression of genes involved in resistance to oxidative stress was detected in these transfected cells by RNA analysis on microarrays, as shown recently for FSHD myoblasts (Winokur et al 2003). In conclusion, our data showed that DUX4 expression in non-affected muscle cells could impinge on the activity of genes known to be disturbed in FSHD myoblasts. We thank the MDA (USA), AFM (France), and ABMM (Belgium) for funding. Graduate student fellowships to E.A. and C.M. came from the FRIA (Belgium), and to S.S. and A.M. from the AFM.