Diagnosis : Richard Lemmers (NL), Nicolas Levy (F) and Kevin Flanigan (USA) discussed their research advances concerning the genetic characterization of FSHD. Diagnostic tools have clearly improved over the past few years and the relationship between the D4Z4 region is now well-known, even if some doubts still exist. Unaffected individuals have between 11 and 150 D4Z4 repetitions. FSHD patients have less than ten. It seems that the severity of symptoms is related to the number of repeats remaining. Thus, patients with between 8 and 10 repetitions can only have very mild symptoms. The effect of modifying genes cannot be excluded since less than 10 D4Z4 repetitions is frequently synonymous with disease in the Caucasian population, whereas Japanese individuals having 6 to 10 repetitions are not affected. Finally, it is now well established that the disease is always linked to allele 4qA and not allele 4qB.
All FSHD cases are related to D4Z4 contractions. In the 5% of cases which do not involve D4Z4, the majority were reclassified into other pathologies. Richard Lemmers presented the characterization of new biomarkers that will allow a more precise diagnosis. These new markers are much closer to the D4Z4 locus and allow to distinguish several 4qA allelic variables.
Transcriptome and proteomics : Kevin Flanigan (USA) and Cecilia Gelfi (I) presented their already published data. They highlighted the difficulty in comparing results coming from different laboratories and using different techniques. Several genes involved in or at least candidates for the physiopathological mechanism of FSHD (i.e. FRG2 and Dux4) are not reproduced on transcriptome chips and are not detectable in proteomics. Consequently, it is not surprising that modifications in their expression levels were not reported. Similarly, different FRG1 isoforms that are not related to D4Z4 exist, making its detection difficult. Finally, under the experimental conditions used, it is not possible to identify the ANT1 gene product in proteomics.
Cecilia Gelfi compared proteins expressed in the vastus lateralis muscles (affected) and deltoïdes (less affected). The expression patterns vary according to age but it seems that the contractile proteins are decreased, the glycolytic proteins are increased, and the proteins involved in oxidative pathways are decreased. These results are being verified by Western blot. Dalila Laoudj (F) is carrying out similar work.
Physiopathological mechanism : Rossella Tupler (USA) reminded us of her work on the effect of over expressing human FRG1 in transgenic mice. Remarkably, these mice present a pathological index very close to FSHD. These mice represent an appropriate model to search for drugs or approaches aimed at blocking the effect of “too much” FRG1. In vitro screening studies are in progress (chemical database, MIT, Harvard) in order to identify molecules that attenuate the FRG1 promoter or directly interact with FRG1.
Alexandra Belayew (B) showed that Dux4 is over expressed in myoblasts of FSHD patients. Her team has developed an antibody directed against Dux4. She also demonstrated that Dux4 is toxic after transfection into TE671cells. This is probably because Dux4 induces an increased synthesis of certain caspases, whose transcription is strongly activated by the C terminal of Dux4. Last of all, Dux4 might also activate the PitX1 gene, a homeobox gene involved in laterisation in the embryo and muscle development. The deregulation of PitX1 might be harmful to muscle cells.
To end this section on physiopathological mechanisms, Julie Dumonceaux (F) partly confirmed the results of Rossella Tupler by showing that the over expression of human or murine FRG1 by AAV gene transfer in the mouse causes a severe dystrophic phenotype. This shows that the disease can begin after birth and thus indicates a minor role for FRG1 over expression during embryonic development. She is currently developing shRNAs capable of destroying the excess FRG1 mRNA produced in pathological situations.
Chromatin modification : Yegor Vassetzky’s (F) research seemed to merge all of the proposed physiopathological hypothesis. The normal D4Z4 locus could act like a silencer by fixing to the nuclear envelope matrix. The neighbouring genes FRG1, FRG2, ANT1 and Dux4 will therefore be turned off. He also showed that the D4Z4 region contains an enhancer activated by the Kruppel like factor 15 (KLF15). D4Z4 contractions would render the silencer inoperative (i.e. more S/MAR (scaffold/matrix attached region) interactions with the nuclear matrix) and would expose the aforementioned neighbouring genes to the enhancer/KLF15.
Conclusion
Several proteins are now well characterized as being over expressed in FSHD patients. It is very likely that these proteins belong to the same cascade of events. In vitro, D4Z4 controls the FRG1 promoter (not the FRG2 promoter) and D4Z4 contractions induce over expression of Dux4. Finally, over expression of FRG1 in vivo, results in a muscular dystrophy similar to FSHD. A deregulation of FRG1 is thus sufficient to cause this pathology in the mouse. Furthermore, in these mice, over expression of ANT1 has been observed. It should be noted that deregulation of FRG2 is not possible because this gene does not exist in the mouse.
To finish, Claude Desnuelle (F) spoke about a phase I autologous myoblast transplantation clinical trial that he has initiated. Three patients have already been selected and no major side effects were observed six months after transplantation. Magnetic resonance imaging (MRI) and force tests are in progress.
