1 June 2009. Mutations in the PTEN-induced novel kinase 1 (Pink1) gene cause a recessive form of early-onset parkinsonism, and all the evidence points to mitochondria as the prime locus of the damage. From its first identification, Pink1 was found to be important at keeping a healthy electrical balance across the mitochondrial membrane (see ARF related news story and Valente et al., 2004). Further studies in fruit flies showed that Pink1 knockout results in severe swelling and disorganization of mitochondria. These morphological effects were attributed to a lack of normal mitochondrial fission, as expression of the fission-promoting GTPase dynamin-related protein 1 (Drp1) compensated for the loss of Pink1 (see Park et al., 2006; Clark et al., 2006; and ARF related news story on Yang et al., 2008 and ARF related news story on Poole et al., 2008).
But data are accumulating in support of the controversial idea that the target for Pink1’s attack may lie elsewhere, that is, not in mitochondrial fission/fusion dynamics but at the heart of mitochondrial function itself, in energy production. The latest installment comes from Mark Cookson and colleagues at the National Institutes of Health in Bethesda, Maryland. In a paper published May 27 in PLoS ONE, the researchers show that knocking out Pink1 in a human neuroblastoma cell line adversely affects mitochondrial membrane potential and renders cells more sensitive to mitochondrial toxins. That finding was expected. But the morphological changes in the mitochondria, and the role of Drp1, were exactly opposite to that seen in flies. The lack of Pink1 produced smaller, more fragmented organelles, and promoting fission by expressing of Drp1 exacerbated cell toxicity. The morphological effects resolved when the cells were treated with the phosphatase inhibitor FK506, which blocked dephosphorylation and activation of Drp1; however, FK506 was unable to restore the membrane potential. The new results echo recent work from Bart De Strooper’s lab that suggested Pink1 inhibits the respiratory chain at Complex I, resulting in energy-starved dysfunctional synapses. This effect occurred in the absence of morphological changes in the organelles (see ARF related news story).
Cookson explained his group’s results in terms of primary and secondary effects of Pink1 loss. “The important thing we see is that although we can use FK506 to rescue some of the morphological effects, the functional effects were still there. This says that there is an order to things; there are early first events and there are later secondary events. Our interpretation of our data is that what Pink1 is really doing is maintaining mitochondrial function. And when you see subtle morphological effects, or grossly distended mitochondria, or nothing at all, that’s probably because those changes are secondary.”
To look at the effects of Pink1 on mitochondria function and morphology in living cells, first authors Anna Sandebring and Kelly Thomas made human dopaminergic M17 cell lines that overexpressed Pink1 or Pink1 shRNAs, and imaged mitochondria by a variety of techniques. Consistent with previous work, overexpression of Pink1 protected cells from the mitochondrial toxin rotenone, while the knockdown showed more cell death and mitochondrial fragmentation after treatment. Without Pink1, mitochondria had lower membrane potential and exhibited subtle morphological changes. The mitochondria appeared “a bit shorter and more fragmented,” Cookson told ARF. “It was not a huge effect, but it let us go into the mechanism.”
The scientists next examined the effect of Drp1. Partial knockdown of Drp1 prevented mitochondrial fragmentation in Pink1-minus cells, whereas overexpression of Drp1 made it worse. The results indicate that lack of Pink1 renders mitochondria sensitive to Drp1-induced fission. This was associated with a loss of Drp1 phosphorylation and increased GTPase activity. At the same time, cellular activity of the calcium-dependent Drp1 phosphatase calcineurin increased. Treating cells with the calcineurin inhibitor FK506 restored Drp1 phosphorylation levels, decreased the number of truncated or fragmented mitochondria, and restored the pattern of normal mitochondrial connection. However, FK506 did not cure the lower membrane potential.
From these results, the authors propose a model where loss of Pink1 has as its primary effect a loss of mitochondrial membrane potential. As a result, cells not only accumulate more reactive oxygen species, but they also fail to buffer calcium effectively. Increased calcium could activate calcineurin, leading to dephosphorylation of Drp1 and increased mitochondrial fission. In another species or cell type or developmental stage, the result on mitochondrial morphology might differ, Cookson said, depending on the regulation of Drp1 or other downstream effectors. In some cases, as in Pink1 knockout mice, there may be no morphological changes at all (Gautier et al., 2008). Nonetheless, Cookson says the primary deficit—a loss of mitochondrial health—is conserved.
Support for this idea comes from another recent paper from the lab of Charleen Chu at the University of Pittsburgh (Dagda et al., 2009). In that work, Pink1 knockdown in a different neuronal cell line, SH-SY5Y, also caused mitochondrial fragmentation. The paper suggests a role for mitochondria turnover via autophagy, which was also stimulated by Pink1 knockdown. The destruction of (presumably unhealthy) mitochondria was protective for the cells, and was enhanced by overexpression of the parkin protein, which has been implicated in autophagy (Narendra et al., 2008). The finding that parkin can compensate for Pink1 deficiency matches genetic studies in Drosophila, and Cookson reports that his group has similar, unpublished results in their cell lines showing a conservation of parkin function.
“If you look across all these studies, the observation that’s most clear is that the mitochondria are not at their full potential,” Cookson said. “It’s the details that are not clear.” The biggest detail Cookson is eager to settle is what Pink1 actually phosphorylates. “All of the evidence so far is that the kinase does not act directly on any of these fusion or fission proteins, but we haven’t looked at all of them. There is some evidence that it might phosphorylate parkin (see Kim et al., 2008), but I don’t know if that’s been reproduced.”—Pat McCaffrey.
Sandebring A, Thomas KJ, Beilina A, van der Brug1 M, Cleland MM, Ahmad R, Miller DW, Zambrano I, Cowburn RF, Behbahani H, Cedazo-Mínguez A, Cookson MR. Mitochondrial Alterations in PINK1 Deficient Cells Are Influenced by Calcineurin-Dependent Dephosphorylation of Dynamin-Related Protein 1. 2009 May 27. PLoS ONE4(5):e5701. Abstract