Cases of inherited Parkinson disease, while rare, have been eagerly studied for clues to causes of the far more prevalent sporadic form of the disease. So far, three genes—parkin, DJ1, and PINK1—have been identified that cause a recessive form of early onset PD. PINK1, the newest member of the trio (see ARF related news story) bears all the sequence hallmarks of a mitochondrial targeted protein kinase. Like parkin (a ubiquitin ligase), and DJ1 (a mitochondrial protein of unknown function; see ARF related news story), PINK1 plays a role in protecting cells from stress, but the details of its role in normal cells and in PD are not yet understood. Now, Mark Cookson and his colleagues at the National Institute on Aging have characterized the activity and cellular localization of wild-type PINK1 and two PD-causing mutants. The authors' work, which appeared in PNAS online on April 11, confirms that the PD mutations compromise PINK1 kinase activity and raises some interesting questions about the protein stability and cellular localization of the protein.
Natural substrates for PINK1 have not yet been identified, but first author Alexandra Beilina and colleagues were able to purify a fragment of the protein that retains autophosphorylation activity. While perhaps not physiologically relevant, the authors found that this activity was halved by the G309D mutation found in a Spanish family, while the L347P mutation found in the Philippines practically abolished kinase activity. But the authors also found that the L347P mutation dramatically reduces protein stability, in either bacteria or mammalian COS cells, suggesting that loss of PINK1 can be causal for PD. The link between the G309D mutant and PD is less obvious, however, given that the protein is relatively stable and retains substantial activity, though this mutation has been shown to leave cells susceptible to mitochondria-induced apoptosis (see ARF related news story).
Of course, evidence linking mitochondria and Parkinson disease has steadily grown, and PINK1 does have a mitochondrial localization signal, which prompted Beilina and colleagues to test the protein's cellular distribution. They found that when overexpressed in mammalian cells, PINK1 is localized mainly to the mitochondria, but that the protein is proteolytically cleaved and a small amount of the C-terminal ends up in the cytosol. The significance of the cytosolic location of the protein is unknown. Though the authors used two different C-terminal tags, both giving similar localization patterns, the cytosolic protein could be an artifact of overexpression. They also found that localization was not affected by PD mutations, suggesting changes in cell localization do not underlie the pathogenic effect of the mutations.
“The outstanding problem now is that we don’t know what the kinase targets are,” said Cookson, who is now fishing for PINK1-associated proteins that could be cellular substrates.—Pat McCaffrey
- Beilina A, Van Der Brug M, Ahmad R, Kesavapany S, Miller DW, Petsko GA, Cookson MR. Mutations in PTEN-induced putative kinase 1 associated with recessive parkinsonism have differential effects on protein stability. Proc Natl Acad Sci U S A. 2005 Apr 19;102(16):5703-8. PubMed.