The cloning of the gene for the PARK6 locus by Enza-Maria Valente and her colleagues now gives us a triumvirate of recessive genes that cause parkinsonism in humans; parkin, DJ-1 and now Pink1. Logically, the identification of three recessive mutations with similar phenotypes suggests that either 1) these three genes now delineate a single pathogenic pathway or 2) they point to different pathogenic processes that happen to all cause loss of a small group of neurons in the substantia nigra.

PINK1 looks very much like a serine/threonine-directed protein kinase, and thus has no immediate connection to the E3-ligase activity of parkin or the varied putative activities of DJ-1. This suggests the involvement of diverse cellular pathways. However, there are one or two intersecting observations that may indicate some similarities among these different gene products. The first is that Pink1 localizes to mitochondria. DJ-1 can localize to mitochondria under some circumstances, and a fraction of parkin is also found in this organelle. The proportion of parkin that localizes to...  Read more

The cloning of the gene for the PARK6 locus by Enza-Maria Valente and her colleagues now gives us a triumvirate of recessive genes that cause parkinsonism in humans; parkin, DJ-1 and now Pink1. Logically, the identification of three recessive mutations with similar phenotypes suggests that either 1) these three genes now delineate a single pathogenic pathway or 2) they point to different pathogenic processes that happen to all cause loss of a small group of neurons in the substantia nigra.

PINK1 looks very much like a serine/threonine-directed protein kinase, and thus has no immediate connection to the E3-ligase activity of parkin or the varied putative activities of DJ-1. This suggests the involvement of diverse cellular pathways. However, there are one or two intersecting observations that may indicate some similarities among these different gene products. The first is that Pink1 localizes to mitochondria. DJ-1 can localize to mitochondria under some circumstances, and a fraction of parkin is also found in this organelle. The proportion of parkin that localizes to mitochondria is small, but experiments in mice and flies suggest that there are mitochondrial effects of parkin knockout. The other thing that links these three proteins is that all three protect cells against “stress” in a very broad sense. Thus, parkin protects against proteasome inhibition and mitochondrial stress, DJ-1 against oxidative events (which may be mitochondrial in nature) and proteasome inhibition, and Pink1 protects against proteasomal dysfunction and mitochondrial damage. Which leaves us with a number of important questions to answer. The mechanism by which Pink1 protects cells against mitochondrial damage secondary to proteasome inhibition is unclear and Valente et al. evoke a mitochondrial substrate. So what are the kinase substrates of Pink1? And is Pink1 somehow intertwined with parkin and DJ-1; or is the concept of one pathogenic cascade a red (or pink) herring?

View all comments by Mark Cookson

The two exciting reports in ScienceExpress of two discoveries, one, the mutant Pink1 gene at the root of PARK6-linked autosomal recessive Parkinson disease; and two, the functional inactivation of parkin's ubiquitin ligase activity by S-nitrosylation, provide strong support for an integrated picture of Parkinson's disease. The characterizations of Pink1 localization (and thus, likely, function) in mitochondria and parkin's inactivation as a result of excess oxidative stress cement two cornerstones of PD pathogenesis, mitochondrial impairment and sustained oxidative stress. They also highlight the relevance of wild-type parkin in the development of sporadic, late-onset PD, given its role in regulating steady-state levels of both mitochondrial enzymes and antioxidant proteins in parkin-deficient mouse brain (see Palacino et al., 2004 in ARF related news story).

View all comments by Michael Schlossmacher