We performed BAP1 mutation screening in a cohort of 140 consecutive UM patients treated in the Department of Ophthalmology at the University Hospital Essen. Variant analysis of BAP1 was performed first on the tumor DNA followed by analying of the detected variant in blood DNA from the same patient. This approach allows to determine or exclude germline heterozygosity for the tumorigenic variants. Using this approach, we identified 2 UM patients with BAP1 tumor predisposition syndrome (BAP1-TPDS). In all other M3 patients with only somatic BAP1 mutations but absence of this mutation in blood DNA, the tumor predisposition syndrome can be excluded. This is clinically of great importance for the patients and their relatives.  In relatives of BAP1 germline carriers, prognosis can be improved by intensive early checkup. In contrast, such extensive examinations are not necessary in the relatives of UM patients with somatic mutations only. Based on this study, we roughly estimated the proportion of UM patients with M3 tumors that carry a BAP1 germline mutation at about 3% (2/61).  For these patients, individual medical and family history is not always apparently distinct from that of the other patients in the cohort, supporting the need for genetic testing.

To test if BAP1 inactivation might be associated with the formation of distinct methylation patterns in UM, we established a set of methylation markers (DMRs) suited for methylation based class assignment of UM by deep amplicon sequencing. Using these markers, we performed classification of genetically well characterized UM, including tumors with partial M3, and metastatic tumor samples. The results show that the methylation pattern correlated with the BAP1 mutation status thus providing further evidence that BAP1 might contribute to the formation of the tumor class specific DNA methylation patterns.

We established an isogenic cell culture model using CRIPR/Cas genome editing of BAP1. Genetic characterization of the genome edited cells revealed absence of wild type BAP1 alleles and by western blot analysis no BAP1 protein could be detected. Comparing the methylation patterns of BAP1 wild type and BAP1 knock out cells using the set of 11 DMRs revealed no significant difference between both cell lines after two weeks of culture was detected. From this result it, can be concluded that other or additional factors beyond BAP1 are essential to form the characteristic methylation pattern of the UM cells with M3. Alternatively, it is possible that the formation of the class-specific methylation pattern takes longer to be fully developed. This has to be tested by methylation analysis after long term cultivation of the BAP1 knock out cells lines.