Abstract
Pathogenic variants in PINK1 cause early-onset Parkinson's disease. Although many PINK1 variants have been reported, the clinical significance is uncertain for the majority of them. To gain insights into the consequences of PINK1 missense variants in a systematic manner, we selected 50 PINK1 missense variants from patient- and population-wide databases and systematically classified them using Sherloc, a comprehensive framework for variant interpretation based on ACMG-AMP guidelines. We then performed functional experiments, including mitophagy and Parkin recruitment assays, to assess the downstream consequences of PINK1 variants. Analysis of PINK1 missense variants based on Sherloc showed that the patient databases over-annotate variants as likely pathogenic. Furthermore, our study shows that pathogenic PINK1 variants are most often linked to a loss-of-function for mitophagy and Parkin recruitment, while this is not observed for variants of unknown significance. In addition to the Sherloc framework, the added layer of evidence of our functional tests suggests a reclassification of 9/50 missense variants. In conclusion, we suggest the assessment of multiple layers of evidence, including functional data on top of available clinical and population-based data, to support the clinical classification of a variant and show that the presence of a missense variant in PINK1 in a Parkinson's disease case does not automatically imply pathogenicity.