A feedback loop between nonsense-mediated decay and the retrogene DUX4 in facioscapulohumeral muscular dystrophy

Genes are sequences of DNA that contain instructions for the cell that must be carefully controlled because it is not always appropriate or safe for these instructions to be followed. When genes are active, copies of the DNA are made using molecules of ribonucleic acid (RNA) and these can then be used as templates to make proteins. One way genes can be controlled is by adding small chemical tags that mark them out to be activated or deactivated, known as epigenetic control. The muscle disease facioscapulohumeral muscular dystrophy (FSHD) is caused by the loss of the chemical tags that normally keep certain genes switched off in many cell types. One of these genes is DUX4, which in healthy males is normally only active in the testes, but in FSHD patients it is also active in other parts of the body. Another way to control genes is by nonsense-mediated decay, where incorrect or incomplete RNA molecules are destroyed before they can be used to make defective proteins. In this study, Feng et al. show that when DUX4 is activated following the failure of epigenetic control in FSHD patients, the effectiveness of nonsense-mediated decay is also reduced. This results in the build-up of incorrect RNA molecules inside muscle cells, which can harm the cell. In fact, 13% of the incorrect RNA molecules that are normally destroyed in cells were found at higher levels when DUX4 was active. To investigate how DUX4 could work, Feng et al. focused on another gene called UPF1 because cells without the protein encoded by this gene have similar defects in nonsense-mediated decay as cells with active DUX4. No difference was found in how often the UPF1 gene is activated in FSHD cells and normal cells. However, the amount of the protein encoded by UPF1 was lower in cells with FSHD than in normal muscle cells. The experiments show that the protein encoded by UPF1 is broken down as a result of the activation of the DUX4 gene, leading to problems with nonsense-mediated decay, which may result in the worsening of FSHD symptoms. The twist in the tale is that DUX4 itself is also controlled by nonsense-mediated decay under normal circumstances. Therefore, in diseased cells, a failure in epigenetic control allows DUX4 to prevent its own destruction by tampering with nonsense-mediated decay. These findings offer new insights into the role of the DUX4 gene in FSHD. The next step is to test whether these defects in nonsense-mediated decay can explain any of the symptoms of FSHD, such as muscle inflammation.