Researchers studying the relationship between the kinase PINK1 and the ubiquitin ligase parkin have discovered that the two interact in a surprising way. Both PINK1 and parkin are risk factors for parkinsonism. According to a report in the March 25 Biochemical Journal and separate meeting presentations earlier this year, PINK1 phosphorylates ubiquitin, which then activates parkin. This is the first evidence that ubiquitin, a small protein that tags other proteins for destruction, is subject to phosphorylation. The work hints at a new way to regulate mitophagy, a way of disposing of mitochondria that is mediated by PINK1 and parkin and is thought to malfunction in PD. If the discovery holds up in vivo, it could hasten the development of new tools, such as antibodies that recognize phosphorylated ubiquitin, to measure mitochondrial dysfunction in PD patients, said Miratul Muqit of the University of Dundee in Scotland, who led the study.
Muqit was surprised to find that a major protein modifier like ubiquitin could itself be modified. He called the discovery paradigm-challenging. Two other researchers created a buzz at “Parkinson’s Disease: Genetics, Mechanisms, and Therapeutics,” a symposium held March 2-7 in Keystone, Colorado, when they presented data supporting the same conclusion. Using different methods, Wade Harper of Harvard Medical School in Boston and Richard Youle of the National Institutes of Health in Bethesda, Maryland, independently discovered that PINK1 activates parkin through the phosphorylation of ubiquitin.
Recessive mutations in PINK1 and parkin genes trigger early onset PD; both proteins facilitate mitophagy. When chemical or other toxins stress mitochondria, PINK1 accumulates on the outer mitochondrial membrane, where it recruits parkin (see Matsuda et al., 2010). The E3 ubiquitin ligase then transfers ubiquitin molecules onto myriad proteins lining the outer membrane. These decorations flag the mitochondria for disposal.
In 2010, researchers worked out that PINK1 kinase activity was necessary for activation of parkin (see Geisler et al., 2010). In 2012, Muqit’s lab reported that the crucial phosphorylation event occurs at parkin’s serine 65 (see Kondapalli et al., 2012), which lies within the protein’s N-terminal ubiquitin-like (Ubl) domain, a region known for keeping parkin in an inactive state. However, phosphorylation of serine 65 only partially activated parkin. Muqit hypothesized that perhaps PINK1 acted on another protein that completed the job.
To find that missing link, first author Agne Kazlauskaite and colleagues examined the phosphoproteome. The researchers induced mitochondrial stress in cells expressing either wild-type PINK1 or a kinase-inactive version, then subjected mitochondrial extracts to mass spectrometry. In cells that expressed wild-type PINK1, one phosphorylated peptide was enriched 14-fold, and this phosphopeptide came from none other than ubiquitin. Curiously, the researchers found that ubiquitin was phosphorylated on its serine 65, the same spot PINK1phosphorylated in parkin’s Ubl domain.
The researchers next ran a multitude of biochemical experiments to nail down whether PINK1 directly phosphorylates ubiquitin, and if so, if this alters parkin activity. The researchers used recombinant PINK1 protein extracted from a flour beetle called Tribolium castaneum because of its proven kinase activity in vitro. Muqit’s lab had developed this tool in 2012 to overcome the poor performance of mammalian versions of the recombinant protein. In vitro kinase assays revealed that TcPINK1 specifically phosphorylated serine 65 of both parkin and ubiquitin.
By this point the researchers knew that PINK1 could phosphorylate both parkin and ubiquitin, but not which phosphorylation was more important to parkin. To address this, the researchers swapped serine 65 on either or both proteins for an alanine, which cannot be phosphorylated. The researchers then mixed the recombinant proteins with TcPINK1 and monitored parkin activity by measuring the accumulation of polyubiquitin chains on the ligase. It turned out that phosphorylation of both proteins was necessary to fully activate parkin.
The researchers also found that without PINK1, phosphorylated ubiquitin (phospho-Ub) only partially activates parkin, suggesting that parkin phosphorylation is required for full activity. In fact, researchers previously had shown that when they removed the serine 65-containing parkin Ubl domain, the ligase functioned, albeit poorly, in the absence of PINK1. Now, Kazlauskaite and colleagues report that far less phospho-Ub is required to stimulate the Ubl-negative ligase than to activate the native ligase. This could mean that the ubiquitin-like domain acted as a built-in inhibitor of parkin and that phosphorylation of the domain relieves this blockade. Muqit proposed a two-step model in which phosphorylation of the Ubl domain by PINK1 changes parkin’s conformation such that phosphorylated ubiquitin fully activates the ligase. “The phosphorylation of parkin likely triggers a profound alteration to open it for business,” Muqit said.
The model makes good evolutionary sense, said Mark Cookson of the National Institute on Aging in Bethesda. “Phosphorylation of both parkin and ubiquitin likely coordinates a reaction that is much more robust than phosphorylation of either protein alone,” Cookson said. “When you see ideas this crisp and neat, you tend to gravitate toward them and hopefully they hold up.”
The basic findings of the study will likely prove true, wrote Harper. That said, “The mechanism by which phosphorylated ubiquitin activates parkin, and whether it is important in vivo, was not answered by this paper,” Harper added. Muqit’s lab is deciphering the mechanism of parkin activation through phosphorylated ubiquitin, and conducting cell-based studies to assess the role of the pathway in mitochondrial function.
Youle’s findings, in press at the Journal of Cell Biology, suggest that phosphorylated ubiquitin binds parkin and facilitates its recruitment to the mitochondria within cells.
Youle said that if PINK1 phosphorylates ubiquitin bound to proteins on the surface of the mitochondria, then it could kick-start a feed-forward loop. “Products of the parkin reaction (ubiquitylated proteins) would become the substrate for PINK1,” he said. These phosphorylated ubiquitin proteins would then serve to further activate parkin. This hypothetical loop could rapidly trigger the mitophagy pathway beyond the point of return, Youle added.
Muqit, Youle, and others have their sights set on developing a phospho-Ub antibody. In conjunction with antibodies specific for phospho-parkin, that could reveal reductions in PINK1/parkin activity in people with Parkinson’s. “My prediction is that even in sporadic PD, a subset of patients have disruptions in the PINK1/parkin axis,” Muqit said.—Jessica Shugart
- Matsuda N, Sato S, Shiba K, Okatsu K, Saisho K, Gautier CA, Sou YS, Saiki S, Kawajiri S, Sato F, Kimura M, Komatsu M, Hattori N, Tanaka K. PINK1 stabilized by mitochondrial depolarization recruits Parkin to damaged mitochondria and activates latent Parkin for mitophagy. J Cell Biol. 2010 Apr 19;189(2):211-21. PubMed.
- Geisler S, Holmström KM, Skujat D, Fiesel FC, Rothfuss OC, Kahle PJ, Springer W. PINK1/Parkin-mediated mitophagy is dependent on VDAC1 and p62/SQSTM1. Nat Cell Biol. 2010 Feb;12(2):119-31. Epub 2010 Jan 24 PubMed.
- Kondapalli C, Kazlauskaite A, Zhang N, Woodroof HI, Campbell DG, Gourlay R, Burchell L, Walden H, Macartney TJ, Deak M, Knebel A, Alessi DR, Muqit MM. PINK1 is activated by mitochondrial membrane potential depolarization and stimulates Parkin E3 ligase activity by phosphorylating Serine 65. Open Biol. 2012 May;2(5):120080. PubMed.
No Available Further Reading
- Kazlauskaite A, Kondapalli C, Gourlay R, Campbell DG, Ritorto MS, Hofmann K, Alessi DR, Knebel A, Trost M, Muqit MM. Parkin is activated by PINK1-dependent phosphorylation of ubiquitin at Ser65. Biochem J. 2014 May 15;460(1):127-39. PubMed.