Previews PINK1-PARKIN Interplay: Down to Ubiquitin Phosphorylation Alexandra Stolz1 and Ivan Dikic1,2,* 1Institute
of Biochemistry II, Goethe University, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany Institute for Molecular Life Sciences, Max-von-Laue-Strasse 15, 60438 Frankfurt am Main, Germany *Correspondence: [email protected]
By using quantitative proteomics, Ordureau et al. (2014) provide a comprehensive view on the regulatory steps by which PINK1-mediated phosphorylation of PARKIN and ubiquitin triggers the recruitment of the ubiquitin ligase PARKIN to damaged mitochondria. Parkinson’s disease (PD) is a chronic neuropathy characterized by a specific loss of dopaminergic neurons. Genetic evidence has shown that PINK1 (PTENinduced putative kinase 1; PARK6) and PARKIN (PARK2) are closely linked to the pathogenesis of autosomal-recessive juvenile Parkinsonism (AR-JP). Several mutations found in PD patients result in loss of function of either of these two enzymes due to truncation, instability, impaired recruitment, or inactivation of their catalytic center. PINK1 is a mitochondrial serine/threonine kinase, while PARKIN is a cytosolic ubiquitin E3 ligase. Biochemical and molecular studies have indicated an epistatic relationship between these two genes involved in mitochondrial homeostasis (Youle and Narendra, 2011). In this pathway PINK1 acts as an upstream kinase that recruits PARKIN, in a phosphorylation-dependent manner, onto dysfunctional mitochondria. However, it is unclear how, and to what extent, PINK1 and PARKIN contribute to mitochondrial homeostasis in vivo. Is it the clearance of oxidized protein aggregates from mitochondria? Do they actively promote mitochondrial removal or are they primarily influencing fusion, fission, and transport of mitochondria (Scarffe et al., 2014)? In this issue, Ordureau et al. (2014) suggest a feedforward mechanism that explains how phosphorylation events mediated by PINK1 put defective mitochondria on an irreversible path to degradation. On healthy mitochondria, newly synthesized PINK1 is imported into the intermembrane space where it undergoes rapid degradation. In dysfunctional mitochondria, PINK1 degradation is impaired,
leading to the accumulation of the kinase on the mitochondrial outer membrane and to sustained phosphorylation of downstream components, including PARKIN (Kondapalli et al., 2012). PARKIN phosphorylation within its ubiquitin-like (UBL) domain (later referred to as p-PARKIN) influences PARKIN ligase activity and has been implicated in PARKIN’s recruitment to mitochondria (Shiba-Fukushima et al., 2012). Moreover, Ordureau et al. (2014), as well as several recent studies, demonstrated that PINK1 is actually able to directly phosphorylate ubiquitin (later referred to as p-UB) on serine 65 (similar to the UBL domain of PARKIN) conjugated to mitochondrial proteins and that this phosphorylation event is critical for efficient PARKIN-dependent ubiquitylation of mitochondrial substrates (Kane et al., 2014; Kazlauskaite et al., 2014; Koyano et al., 2014). However, the precise role of p-UB in the regulation of mitochondrial PARKIN translocation and ubiquitin chain synthesis in vivo remained poorly understood. Using quantitative proteomics, Harper and colleagues further dissect this process and provide evidence to support a feedforward mechanism in PINK1-PARKIN activation, which relies on PINK1-dependent phosphorylation of PARKIN and ubiquitin chains. The proposed feedforward mechanism initiates with the accumulation of PINK1 on dysfunctional mitochondria and subsequent phosphorylation of cytosolic PARKIN (Figure 1) leading to a tremendous (2,308-fold) increase of its catalytic activity and to initial ubiquitylation events on multiple outer mitochondrial membrane proteins (Sarraf et al., 2013; Ordureau et al., 2014). PINK1 then targets these
newly attached ubiquitin moieties for phosphorylation at serine 65 (later referred to as poly-p-UB). Recent models suggest p-UB as a PARKIN activator (Kane et al., 2014; Kazlauskaite et al., 2014; Koyano et al., 2014). However, Ordureau et al. show that catalytic activity of PARKIN is a prerequisite of UB phosphorylation, that the activity of p-PARKIN remains essentially unchanged in the presence of p-UB chains, and that increasing amounts of p-UB have an inhibitory effect on chain synthesis. Therefore, the authors propose a model where p-UB primarily functions as a localization anchor point for PARKIN on damaged mitochondria, preventing its diffusion back to the cytoplasm (Figure 1). Along these lines, the authors show that p-PARKIN has no detectable binding affinity toward unmodified UB, but strongly binds to p-UB chains. The suggested feedforward mechanism does not refute the previous working model of p-UB unlocking the autoinhibition of PARKIN in an allosteric manner but might actually complement it. Also, Ordureau et al. (2014) confirmed the activation of the catalytic activity of unmodified PARKIN by p-UB. Hence, to a certain extent, poly-pUB might also serve as an anchor point on mitochondria for unmodified PARKIN, simultaneously inducing conformational changes and catalytic activity, respectively. The suggested PINK1-PARKIN interplay propels the removal of dysfunctional mitochondria from the cell interior: the feedforward mechanism enables locally restricted PARKIN to spread a wave of further ubiquitylation events on the dysfunctional organelle, while it at the same time prevents transgression
Molecular Cell 56, November 6, 2014 ª2014 Elsevier Inc. 341
it exclusively linked to PARKIN activity? What are the associated kinases? Are deubiquitinases regulated by a similar mechanism? Are there deubiquitinating enzymes that selectively target p-UB and counteract specific ubiquitylation/ phosphorylation events? And also, are there other posttranslational modifications of ubiquitin, such as acetylation or methylation, that could expand the complexity of UB regulatory networks and have an important contribution to regulation of biological processes?
Figure 1. Phosphorylation of Ubiquitin Contributes to a Feedforward Mechanism for Ubiquitylation Events on Dysfunctional Mitochondria PINK1-dependent phosphorylation of PARKIN on its ubiquitin-like domain promotes a conformational change within the ligase, converting it from an autoinhibitory to an active state. Activated PARKIN ubiquitylates specific substrates on the outer mitochondrial membrane. PINK1 kinase phosphorylates UB chains on Ser65. As a consequence, PARKIN is tightly bound to poly-p-UB chains and is locally restricted to mediate efficient ubiquitylation of neighboring substrates.
of PARKIN-mediated ubiquitylation to healthy mitochondria, e.g., in the case of incomplete fission or spatial contiguity. If phosphorylation of PARKIN or the presence of ubiquitin chains was sufficient to activate and efficiently recruit more PARKIN molecules to mitochondria, then this could result in the complete overgrowth of the mitochondrial network with ubiquitin chains and subsequent breakdown of the whole organelle network. The mechanism described by Ordureau et al. (2014) may prevent this unfavorable event from happening, as PARKIN’s preferential binding to p-UB chains over unmodified UB chains ensures local restriction of PARKIN to PINK1-positive, dysfunctional mitochondria. In this way efficient removal of a damaged organelle combined with high selectivity is ensured. The dual role of PINK1 in PARKIN activation through both phosphorylation of PARKIN and ubiquitin chains is intriguing and well documented in cultured cells, but its functional relevance in vivo remains to be analyzed and validated in primary neuronal cells.
In addition to putting forward this conceptually novel model, the authors have deciphered the PARKIN ubiquitylation code in a quantitative manner showing that PARKIN synthesizes UB chains with K6, K11, K48, and K63 linkages. The ability of PARKIN to intrinsically produce four different kinds of ubiquitin chains raises the question of which pathways are attracted to such a combination of UB chains to determine the fate of damaged mitochondria, including their transport, fusion/fission, and mitophagy (Scarffe et al., 2014; Stolz et al., 2014; Rogov et al., 2014). Most analyses presented in the study are not only performed with modified and unmodified PARKIN but, in addition, with several diseaserelated PARKIN variants, thereby providing a valuable foundation for future clinical studies. The findings of Ordureau et al. (2014) and others (Kane et al., 2014; Kazlauskaite et al., 2014; Koyano et al., 2014) are also raising more general questions: Is phosphorylation of ubiquitin involved in the regulation of other Ub ligases, or is
342 Molecular Cell 56, November 6, 2014 ª2014 Elsevier Inc.
Kane, L.A., Lazarou, M., Fogel, A.I., Li, Y., Yamano, K., Sarraf, S.A., Banerjee, S., and Youle, R.J. (2014). J. Cell Biol. 205, 143–153. Kazlauskaite, A., Kondapalli, C., Gourlay, R., Campbell, D.G., Ritorto, M.S., Hofmann, K., Alessi, D.R., Knebel, A., Trost, M., and Muqit, M.M. (2014). Biochem. J. 460, 127–139. Kondapalli, C., Kazlauskaite, A., Zhang, N., Woodroof, H.I., Campbell, D.G., Gourlay, R., Burchell, L., Walden, H., Macartney, T.J., Deak, M., et al. (2012). Open Biol. 2, 120080. Koyano, F., Okatsu, K., Kosako, H., Tamura, Y., Go, E., Kimura, M., Kimura, Y., Tsuchiya, H., Yoshihara, H., Hirokawa, T., et al. (2014). Nature 510, 162–166. Ordureau, A., Sarraf, S.A., Duda, D.M., Heo, J.M., Jedrychowski, M.P., Sviderskiy, V.O., Olszewski, J.L., Koerber, J.T., Xie, T., Beausoleil, S.A., et al. (2014). Mol. Cell 56, this issue, 360–375. Rogov, V., Do¨tsch, V., Johansen, T., and Kirkin, V. (2014). Mol. Cell 53, 167–178. Sarraf, S.A., Raman, M., Guarani-Pereira, V., Sowa, M.E., Huttlin, E.L., Gygi, S.P., and Harper, J.W. (2013). Nature 496, 372–376. Scarffe, L.A., Stevens, D.A., Dawson, V.L., and Dawson, T.M. (2014). Trends Neurosci. 37, 315–324. Shiba-Fukushima, K., Imai, Y., Yoshida, S., Ishihama, Y., Kanao, T., Sato, S., and Hattori, N. (2012). Sci. Rep. 2, 1002. Stolz, A., Ernst, A., and Dikic, I. (2014). Nat. Cell Biol. 16, 495–501. Youle, R.J., and Narendra, D.P. (2011). Nat. Rev. Mol. Cell Biol. 12, 9–14.