#apaperaday: Clinical and genetic interpretation of uncertain DMD missense variants: evidence from mRNA and protein studies.

In today’s #apaperaday, Prof. Aartsma-Rus reads and comments on the paper titled: Clinical and genetic interpretation of uncertain DMD missense variants: evidence from mRNA and protein studies.

Today’s post Easter pick is from Xie et al in @ojrarediseases on ‘missense’ variants causing dystrophinopathies. The ‘ are used because some of the variants actually cause missplicing! DOI: 10.1186/s13023-024-03128-7.

Dystrophinopathies can cause Duchenne (absence of functional dystrophin in skeletal muscle) or Becker (production of partially functional dystrophins in skeletal muscle). Most pathogenic variants are deletions or duplications of 1 or more exons.

Small mutations within exons are also reported, nonsense and small deletions or insertions, but also rarely missense variants. Because dystrophin is a huge protein, the pathogenicity of missense variants is often difficult to predict.

There are a few known missense variants, e.g. in the actin binding domain of dystrophin causing Becker and in the dystroglycan binding domain causing Duchenne. However, when missense are present in the rod domain the pathogenicity is difficult to predict.

Authors here dived into this, by studying the effects in more detail for 9 patients. 3 presented with Duchenne-like symptoms, severe pathology before age 4 and quick progression and very high CK at presentation. 6 others presented with Becker-like symptoms: the onset was still early (around age 5) but less severe and also progression was slower. CK was elevated but in the 1000s rather than 10000s. Authors specify that these individuals could have Becker but could also have a sarcoglycanopathy.

Missense variants were found for these individuals. Biopsies were done to study dystrophin protein & mRNA. On immunohistochemistry, no dystrophin was detected for 3 severe patients, while some staining, especially for the C-terminal part was detected for the 6 milder patients.

Sarcoglycans were reduced as well for the 6 milder patients, still not ruling out a sarcoglycanopathy. MRI showed fat infiltration and histology showed severe dystrophic changes for the Duchenne patients and dystrophy/myopathic changes for the other 6.

DNA analysis had revealed missense variants. 3 of these, found for the 3 Duchenne patients, had been reported before as Duchenne causing. However, in those studies only predictions were made on missplicing and no mRNA and protein analyses were done.

Authors state they thought this was insufficient evidence, & they performed mRNA analysis showing missplicing for all 3, causing out-of-frame transcripts instead of a missense transcript. Note that 2/3 events were predicted with software, but 1/3 was not predicted correctly.

For the other 6 patients authors ruled out missplicing on the sarcoglycanopathy genes and they observed the missense variants in thedystrophin transcripts. They argue that these variants likely cause the Becker symptoms. Based on their data they classify 2 as likely pathogenic

The other 4 are still variants of unknown significance. For the 3 Duchenne patients authors could ‘upgrade’ the variant from likely pathogenic to pathogenic. Authors discuss that having a confirmed diagnosis is important, e.g. to initiate steroids for Duchenne diagnoses.

They also stress that missense variants found in non-critical domains of dystrophin in patients presenting with Duchenne symptoms should trigger mRNA analysis as it is more likely these cause missplicing. This means that for those patients, biopsies are still needed.

In many cases missense variants are not further studied and prediction softwares are often not correct so mRNA analysis is really a crucial confirmatory step (unless someone else has done this and reported it, like authors do here).

Authors list the limitations: only 9 patients were included (I think this is still quite a lot for rare variants in a rare disease) and that they did not perform Western blot so there is limited information on protein amounts.

The most important message however is the consider that not all missense variants may be that. They may affect splicing. Common sense is needed here. If missense variants are in important protein domains, this is likely detrimental.

However, when they are not, also missplicing should be considered. Splicing software can help but it not optimal and rather inaccurate in many cases. For some diseases, e.g. those affecting brain, mRNA analysis is not always easy though.

Back to the paper: kudos for the authors for studying these variants in detail and reporting their results. It will help classify others finding the same variants and hopefully will make people more alert to the missplicing possibility.