A fascinating new paper has turned PWS genetics on its head, and has important implications for understanding the underlying disruptions that cause PWS. As anyone who is familiar with PWS knows, the genetics of this disorder is very complex. The PWS region encompasses a pretty big chunk of DNA on chromosome 15 and includes five genes that code for proteins (MKRN3, MAGEL2, NECDIN, NPAP1/C15orf12, and SNURF-SNRPN) followed by a series of unusual “snoRNAs”.These snoRNAs sequences code for short RNA molecules that are thought to regulate other RNAs in the cell, in a way that’s not completely understood. A major focus of PWS genetic research has been to examine the PWS region and sort through which of the stretches of DNA (protein coding genes, snoRNAs) are most important in conferring the PWS features (phenotype). One way for geneticists to narrow down the PWS ‘critical region’ is to carefully characterize rare PWS patients with very small deletions of the region. Over the past several years, this approach, combined with good mouse genetics, has led to a general (but not complete) consensus in the field that the SNORD116 snoRNAs are ‘the’ critical portion of the DNA – lose SNORD116s and you get all of the major characteristics of PWS (eg., Sahoo et al, deSmith et al, Duker et al).
The new paper by Schaaf and colleagues at Baylor College of Medicine, Truncating mutations in MAGEL2 cause Prader-Willi phenotypes and autism, throws a major wrench in that understanding. The Baylor team specializes in using next generation DNA sequencing technology to sequence the genomes of people who apparently have a genetic disorder, but for whom a diagnosis has proven difficult to make. Using this large scale DNA sequencing approach, the research team identified four individuals with many of the major features of PWS whose only genetic mutation is a defect in the MAGEL2 gene in the PWS region.
This is surprising given the previous evidence SNORD116 was key. Up until this report, MAGEL2 was thought perhaps to contribute to some of the minor features of PWS, but this new evidence suggests disruption of MAGEL2 may be a major contributor to the PWS phenotype. The four cases presented in the Schaaf paper had some variability in the presentation of PWS symptoms, which may be complicated in part because of the young age (5 years old) of one patient. Each of the patients showed several, but not all, of the PWS major clinical symptoms, including poor muscle tone at birth, early feeding problems, excessive weight gain after infancy and before 6 years of age, hypogonadism and intellectual disability.
As well, many of the PWS “minor” clinical criteria (weak cry, speech articulation problems, skin picking, and challenging behavior) were present in the individuals studied. Interestingly, all of these cases described were also diagnosed with autism spectrum disorder (ASD). It’s not yet clear yet whether all individuals with a mutated MAGEL2 gene will have autism - there may be some bias in the population that was included in this study – but it does raise new questions about whether a deficit in MAGEL2 production is the reason for the increased incidence of autism in PWS. As well, it implicates MAGEL2 as a new autism susceptibility gene.
So, what does this paper mean for our community? The biggest short-term impacts are likely to be in the diagnostic and research settings. Kids who “look” like they have PWS, but who tested negative by methylation testing, might now be retested to see if they actually have a mutation in their MAGEL2 gene (all of the individuals described in the paper had normal methylation status and would not have been diagnosed as having PWS using a standard methylation test). As far as understanding PWS, there is likely to be a more focused and intense work aimed at understanding the normal role of MAGEL2 in the cell.
There is currently limited knowledge about the normal function of the MAGEL2 protein, but it may play an important general role in protein trafficking in the cell (Hao et al). We are fortunate that Magel2 deficient mouse models have been produced and characterized by the Wevrick and Muscatelli labs, and show many, but not all, features consistent with PWS - including failure to thrive at birth (rescued by oxytocin administration), obesity and increased fat mass (although due to inactivity as opposed to overeating), circadian rhythm defects, poor glucose regulation, altered brain chemistry and impaired fertility. Intellectual deficits and autistic behaviors in this mouse model have not yet been fully explored and warrant additional focus given the current findings.
Looking ahead towards potential therapeutics, one important question to address will be - if you replace the MAGEL2 gene in the setting, for example, of a large PWS deletion, will it be able to correct enough of the genetic defect to have a beneficial effect? Would there be a positive effect if this were done after the birth of an individual? In a child or adult? Or would the correction have to come early in development? If a gene replacement strategy proved technically too difficult, could replacement of the MAGEL2 function be achieved by drugs? These are challenging questions that will take some time to address, but if the concepts can be demonstrated in an animal model of Magel2 deficiency, it would open up new therapeutic avenues.
One final ‘big picture’ question is why is there a discrepancy between prior reports that pointed to the SNORD116 snoRNAs as being the main culprit in PWS, and this paper, which puts the focus back on MAGEL2? This is another area that will need to be addressed experimentally, but it suggests that the entire chromosome 15 PWS region may be regulated in a more coordinated and tightly linked way than previously appreciated. Like all good science, we expect this paper will help spark an entirely new set of investigations, which will bring about a better understanding of the mechanisms causing PWS and eventually lead to new therapeutic approaches that might benefit those struggling with the disorder.