Final Activity Report Summary - PME (Molecular genetic and functional studies of Lafora Progressive Myoclonus Epilepsy) To gain insight into the possible candidate substrate of laforin and then to understand its role in the pathology of the disease we used different approaches such as yeast two-hybrid, immune-precipitation and immune-localisation. Our first round of experiments using EPM2AIP1 as bait; in the yeast two-hybrid screen yielded 58 clones and some of them are represented more than one time. We analysed their function in the literature and none of them seemed obviously correlated with the disease. We transformed them again in the yeast with our bait and with the vector and some of them were interacting with the vector so we excluded them. EPM2B protein product, malin, has been predicted to have ubiquitination activity since it contains a zing finger of the RING-HC type domain in the N-terminus. In our proposal we wanted to confirm this activity but during the reviewing process of this application, in collaboration with Dr. Minassian, we were able to show that malin self-ubiquitinates and, introducing a Cys26Ser mutation in the malin RING domain that would abolish the function, we obtain no ubiquitination activity (Lohi et al., Hum.Mol.Gen. 2005). These new data in addition to the discovery that laforin interacts even with glycogen synthase kinase 3 (GSK3) (Lohi et al., Hum.Mol.Gen. 2005) opened a scenario on what we had already thought that means the involvement of the glycogen metabolism in the Lafora pathogenesis. We were checking whether the known laforin interacting proteins (EPM2AIP1, HIRIP5 and R5) interact among them and also with malin since now we know they interact with laforin. Our results are showing that only yeast transformed colonies EPM2AIP1-R5 grow on selective medium and turn blue in the beta-gal assay. This is an exciting result since we know that R5 is a subunit of the protein phosphatase 1 (PP1) and together activate GS. What this interaction means we still don't know co-localisation experiments of R5 and GRP94 (a specific marker of the endoplasmic reticulum, ER) were positive confirming what we showed in yeast since EPM2AIP1 co-localises with the ER too. While we were doing these experiments, our Canadian collaborators created a knock out mouse EPM2AIP1. They are analysing the mouse tissues to see how the glycogen metabolism is involved and what happen to the R5 protein when EPM2AIP1 is absent. This will be an additional proof of the two proteins interaction and will allow us to elucidate the meaning of the interaction and finally write the paper. While we were getting these exciting news, a yeast two hybrid screening was initiated using EPM2B full-length gene as a bait. We know from our experiments and from the literature that malin interacts with laforin and probably with glycogen synthase and it seems that it is able to ubiquitinate both of them although this is not very clear. To clarify the pathway in which malin is involved we looked at other protein interactions screening a liver cDNA library. More than 60 clones were isolated but only 17 had a correct frame and since some of them were represented more times in total we obtained 10 transcript units. Among them there is the beta-contractin (actin-related protein 1B) involved in microtubule based vesicle motility and since other proteins such as GSK3 beta that has been found to interact with laforin has been seen to be involved with vescicles motility could be a start point for analysing by immunoprecipitation malin candidate proteins. To identify the EPM2C gene we were using a strategy similar to that employed successfully for cloning EPM2A and EPM2B. The 10 cM genome-wide linkage scan in five LD families is completed. Analysis, still in progress, has so far generated positive lod scores for four markers flanking up to 8 cM on each side of marker D13S788. Once found, the gene will be located in the same manner that was used to clone EPM2A and EPM2B.
