27 January 2008. Parkinson disease (PD) is a bit of an enigmatic mix. Inherited forms of the disease are caused by mutations in any one of at least nine different genes, while environmental and genetic risk factors are both ingredients for sporadic Parkinson’s. It is not fully clear how these different etiologies blend into the pathology of what we know as Parkinson disease, but one recurrent theme does emerge—mitochondria are somehow involved. That fact is brought home once again by two papers in this week’s PNAS. Researchers led by David Park at the University of Ottawa, Canada, report that Pink1, a protein linked to familial PD, protects neurons from the mitochondrial toxin MPTP, which causes PD-like pathology in humans and animals alike. This protection is afforded even when Pink1 is restricted to the cytoplasm, the researchers found, though it is unclear how. In the second paper, a group led by Leo Pallanck at the University of Washington, Seattle, offers some clues. The researchers report that Pink1—and parkin, another PD-linked protein—prevent mitochondria from growing too large. In fact, they report that some of the toxic effects of mutant Pink1 and mutant parkin are rescued by a gene that promotes mitochondrial fission. The findings suggest that gross morphological changes to mitochondria may underlie at least some forms of Parkinson disease.
PD mutations in the Pink1 gene were first identified in 2004 (see ARF related news story) and since then scientists have struggled to explain how Pink1, short for PTEN-induced putative kinase 1, fits into the pathology of the disease. Since the kinase is localized, at least in part, in mitochondria, speculation has been rife that that is where Pink1 and PD intertwine. This seems an attractive idea, since the environmental toxin MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) and other mitochondrial toxins, such as rotenone, can cause neurodegeneration in the dopaminergic neurons that are lost in PD.
To address this idea, Park and colleagues tested if loss of Pink1 makes life any more difficult for mitochondria challenged with MPTP. First author Emdadul Haque and colleagues used RNA interference to knock down Pink1 expression in NIH 3T3 cell lines and in primary cortical neurons. In both cases, cell survival after treatment with MPTP was significantly reduced compared to wild-type cells. The reverse experiment, overexpression, showed that Pink1 protects cells from MPTP and that a Pink1 mutation found in familial PD cases (G309D) fails in that regard. The researchers also found that abolishing the kinase activity of Pink1 scuppered its ability to protect against the mitochondrial toxin.
These experiments indicate that Pink1 contributes to the health of mitochondria in cell cultures, but what about live animals? The researchers used adenoviruses to introduce Pink1 genes into the substantia nigra of adult mice and then examined whether increased expression of the kinase is protective. Again, they found that wild-type protein, but not the G309D or kinase mutants, spared tyrosine hydroxylase-positive dopaminergic neurons from MPTP.
One of the intriguing things about this work is that Pink1 constructs lacking a mitochondrial localization signal were every bit as good at protecting cells as those with the targeting sequence. This suggests that Pink1 does not have to be translocated into mitochondria to be effective. In fact, Haque and colleagues found that the majority of Pink1 ends up in the cytoplasm, even when it has a mitochondrial signal. This speaks to an ongoing debate about how and where Pink1 operates. While data suggest that the protein is equally at home in the cytosol and the mitochondria, there are also indications that the protein is proteolytically cleaved in the latter, making interpretation of Western blots a little tricky (see ARF related news story and commentary). In this paper Haque and colleagues also report smaller species that cross-react with Pink1 antibodies, though it is not clear if these are processed fragments of Pink1 or other proteins that non-specifically interact with the antibodies used. “However, it is clear that a significant amount of full-length Pink1 is present in the cytosol,” write the authors.
While the “where” of Pink1 will probably continue to be debated, the “how” is even murkier. A consensus is yet to emerge on which of Pink1’s potential substrates is linked to PD, though there is evidence that Pink1 acts upstream of parkin, since the latter can rescue Pink1 mutants in Drosophila (see Park et al., 2006) and human cells (see Exner et al., 2007). The paper by Pallanck and colleagues suggests that both are involved in mitochondrial fission.
When mutated Pink1 or parkin is expressed in fruit fly flight muscles, mitochondria become enlarged. This could be because the organelles are not dividing normally. Alternatively, they may have a greater propensity to fuse together. To test these possibilities, first author Angela Poole and colleagues looked at the interplay between Pink1/parkin and recently discovered genes that regulate mitochondrial fission and fusion. The researchers found that loss of functional dynamin-related protein 1 (drp1), a cytoplasmic protein that promotes mitochondrial fission, exacerbated both parkin- and Pink1-null phenotypes. While almost all flies made it to adulthood in parkin- and Pink1-null backgrounds, the combination of heterozygous drp1 mutation and parkin loss is fully lethal, while in the case of drp1/Pink1-nulls, only about 10 percent of larvae made it to adulthood and they died prematurely. Heterozygous drp1 mutants in a wild-type background had no effect on viability.
These results suggested that mitochondrial fission is compromised by loss of parkin or Pink1. The researchers tested this in two ways, by increasing expression of drp1 to promote fission and by suppressing two other genes, optic atrophy 1 (OPA1) and Mitofusin2 (Mfn2), involved in mitochondrial fusion. Both approaches partially rescued parkin and Pink1 phenotypes. For example, only about 10 percent of adults can fly in the Pink1 background, but about 50 and 60 percent get airborne when either drp1 is overexpressed or OPA1 is mutated. Similar improvements were seen in climbing abilities. Mutation of Mfn2 rescued flying and climbing abilities, too, though not as well. Where Mfn2 mutation did excel, however, was in correcting an anatomical malformation. Mutant parkin and Pink1 cause thoracic indentations, and these were completely rescued by the Mfn2 mutation.
All told, these results indicated that poor mitochondrial morphology is at least partly to blame for parkin and Pink1 phenotypes in flies. In fact, the authors confirmed this by electron microscopy. In Pink1 or parkin mutants, adding drp1 or mutating OPA1 returned the organelles to their normal size. The results also hint that drp1 might be a substrate of Pink1/parkin, but this will have to be tested.
What about humans? Do these findings have any relevance to the loss of dopaminergic neurons that causes Parkinson disease? The authors point out that mitochondrial changes seen in Pink1-deficient human cells are different from those in Drosophila (see Exner et al., 2007). “Also, there is currently no direct evidence for the involvement of dysfunctional mitochondrial dynamics in PD,” they authors write. So it remains to be seen if mitochondrial morphology is germane to human PD. However, the pathological consequences of mitochondrial fusion/fission failures is only just becoming appreciated (see review by David Chan). In fact, a single case was recently reported of a lethal mutation in the human dynamin-like protein 1 that caused abnormal brain development (see Waterham et al., 2007). —Tom Fagan.
Poole AC, Thomas RE, Andrews LA, McBride HM, Whitworth AJ, Pallanck LJ. PINK1/Parkin pathway regulates mitochondria morphology. PNAS online. 2008 January 21. Abstract
Haque ME, Thomas KJ, D’Souza C, Callaghan S, Kitada T, Slack RS, Fraser P, Cookson MR, Tandon A, Park DS. Cytoplasmic Pink1 activity protects neurons from dopaminergic neurotoxin MPTP. PNAS online. 2008 January 21. Abstract