Parkin mediates nonclassical, proteasomal-independent ubiquitination of synphilin-1: implications for Lewy body formation

Abstract

It is widely accepted that the familial Parkinson's disease (PD)-linked gene product, parkin, functions as a ubiquitin ligase involved in protein turnover via the ubiquitin-proteasome system. Substrates ubiquitinated by parkin are hence thought to be destined for proteasomal degradation. Because we demonstrated previously that parkin interacts with and ubiquitinates synphilin-1, we initially expected synphilin-1 degradation to be enhanced in the presence of parkin. Contrary to our expectation, we found that synphilin-1 is normally ubiquitinated by parkin in a nonclassical, proteasomal-independent manner that involves lysine 63 (K63)-linked polyubiquitin chain formation. Parkin-mediated degradation of synphilin-1 occurs appreciably only at an unusually high parkin to synphilin-1 expression ratio or when primed for lysine 48 (K48)-linked ubiquitination. In addition we found that parkin-mediated ubiquitination of proteins within Lewy-body-like inclusions formed by the coexpression of synphilin-1, alpha-synuclein, and parkin occurs predominantly via K63 linkages and that the formation of these inclusions is enhanced by K63-linked ubiquitination. Our results suggest that parkin is a dual-function ubiquitin ligase and that K63-linked ubiquitination of synphilin-1 by parkin may be involved in the formation of Lewy body inclusions associated with PD.

Figure 1.

Effect of parkin on synphilin-1 turnover. A, Pulsechase analysis of the turnover rate of myc-tagged synphilin-1 in HEK 293 cells in the absence or presence of FLAG-tagged parkin. Data from three independent experiments are expressed as means ± SEM. B, Top panel, Anti-myc immunoblot of equal amount of total cell lysates prepared from HEK 293 cells transfected with myc-tagged synphilin-1 and treated with puromycin (100 μm) for different durations as indicated. Middle panel, Anti-myc and anti-FLAG immunoblots of equal amount of total cell lysates prepared from HEK 293 cells transfected with myc-tagged synphilin-1 without or with an equivalent amount of FLAG-tagged parkin and treated with puromycin (100 μm) for 24 h. The anti-myc immunoblots from three independent experiments were used to derive the relative densitometric units of synphilin-1, which is presented as a histogram in the bottom panel. Closed and opened bars represent units of synphilin-1 in the absence and presence of parkin respectively (^*p > 0.1; ^**p < 0.05; Student's t test). C, Anti-HA and anti-myc immunoblots of synphilin immunoprecipitates (IPmyc) prepared from HEK 293 cells transfected with myc-tagged synphilin-1 in the presence of HA-tagged wild-type ubiquitin without or with FLAG-tagged parkin. Lysates prepared from these variously transfected cells were also subjected to anti-myc or anti-FLAG immunoblotting to show their expression (INPUT). These experiments were replicated at least three times.

Figure 2.

Dose-dependent effect of parkin on synphilin-1 steady-state levels. A, Anti-myc and anti-FLAG immunoblots of total cell lysates prepared from HEK 293 cells transfected with myc-tagged synphilin-1 (left panels) or myc-tagged α-synuclein (right panels) in the absence or presence of increasing relative amounts of FLAG-tagged parkin, as indicated. B, Anti-myc, anti-FLAG, and anti-HA immunoblots of total cell lysates prepared from HEK 293 cells transfected with myc-tagged synphilin-1 alone or with various FLAG-tagged or HA-tagged E3 ligases (as indicated) at an E3 ligase/synphilin-1 cDNA cotransfection ratio of 4:1. Equal loading of the different cell lysates was verified by anti-β-actin immunoblotting. These experiments were replicated at least three times.

Figure 3.

Degradation of synphilin-1 by parkin is dependent on their relative expression levels and on parkin catalytic competency. A, Anti-myc and anti-FLAG immunoblots of total cell lysates prepared from HEK 293 cells transfected with myc-tagged synphilin-1 without or with FLAG-tagged parkin at different cotransfection ratios (as indicated) and treated 24 h later with puromycin (100 μm) for 24 h. B, Anti-myc and anti-FLAG immunoblots of total cell lysates prepared from HEK 293 cells transfected with myc-tagged synphilin-1 without or with FLAG-tagged, wild-type parkin, or familial-PD parkin mutants, at a parkin/synphilin-1 cDNA cotransfection ratio of 4:1 and treated with puromycin (100 μm) for 24 h. In both cases, equal loading of the different cell lysates was verified by anti-β-actin immunoblotting. C, HEK 293 cells transfected with the indicated combinations of myc-tagged synphilin-1 and FLAG-tagged parkin were left untreated or treated with 5 μm clasto-lactacytstin β-lactone (lactacystin) for 16 h. Lysates prepared from these cells were immunoblotted with anti-myc or anti-FLAG as indicated. These experiments were duplicated with similar results.

Figure 4.

Parkin is not an efficient degradation-associated E3 ligase. A, Anti-HA and anti-myc immunoblots of total cell lysates prepared from HEK 293 cells transfected with HA-tagged synphilin-1 alone or with myc-tagged parkin, siah-1, or siah-2 at 1:1 cotransfection ratio. Asterisk indicates nonspecific band. B, Anti-myc and anti-FLAG immunoblots of total cell lysates prepared from HEK 293 cells cotransfected with the indicated combinations of myc-tagged synphilin-1, FLAG-tagged parkin, and various HA-tagged ubiquitin species. C, Anti-myc immunoblots of total cell lysates prepared from HEK 293 cells cotransfected with the indicated combinations of myc-tagged synphilin-1, myc-tagged siah-1, and various HA-tagged ubiquitin species. Asterisk indicates nonspecific band. For both B and C, equal loading of the different cell lysates was verified by anti-β-actin immunoblotting. These experiments were duplicated with similar results.

Figure 5.

Parkin mediates K63-linked polyubiquitination of synphilin-1. A, A portion of lysates prepared from HEK 293 cells transfected with FLAG-tagged parkin alone or with various HA-tagged ubiquitin species were subjected to anti-FLAG immunoprecipitation followed by anti-HA immunoblotting (left panel). The blot was stripped and reprobed with anti-FLAG to show the amounts of immunoprecipitated parkin (top right panel). Lysates prepared from these variously transfected cells (INPUT) were also subjected to anti-FLAG immunoblotting to show FLAG-parkin expression levels (bottom right panel). B, A portion of lysates prepared from HEK 293 cells transfected with the indicated combinations of myc-tagged synphilin-1, FLAG-tagged parkin, and various HA-tagged ubiquitin species were subjected to anti-myc immunoprecipitation followed by anti-HA immunoblotting (top left panel). The blot was stripped and reprobed with anti-myc (bottom left panel) to illustrate that relatively equal amounts of synphilin were immunoprecipitated. The remaining portion of the transfected cell lysates was immunoblotted with anti-HA (top right panel) and anti-myc (bottom right panel) to show their respective expression levels. These experiments were replicated at least three times.

Figure 6.

Parkin mediates predominantly K-63-linked ubiquitination of cytosolic inclusions formed by coexpression of synphilin-1 and α-synuclein. A, Coexpression of α-synuclein, myc-tagged synphilin-1, FLAG-tagged parkin, and either HA-tagged K48 ubiquitin or K63 ubiquitin results in the formation of myc-synphilin-1-positive inclusions (red). Positive anti-HA staining of the inclusion bodies is observed only in transfections containing HA-tagged K63 ubiquitin (green). The merged pictures show colocalization of the anti-myc and anti-HA staining (yellow). B, Coexpression of mutant huntingtin and either HA-tagged K-48 ubiquitin or HA-tagged K-63 ubiquitin results in the formation of huntingtin-positive inclusions (red). Positive anti-HA staining of the inclusion bodies is observed only in transfections containing HA-tagged K-48 ubiquitin (green). The merged pictures show colocalization of the anti-huntingtin and anti-HA staining (yellow). C, Histograms showing quantitative data of inclusion bodies in SH-SY5Y cells cotransfected with the various construct as described in A and B. These experiments were replicated at least three times.