Siah-1 Facilitates Ubiquitination and Degradation of Synphilin-1*

Parkinson's disease is a common neurodegenerative disorder characterized by loss of dopaminergic neurons and appearance of Lewy bodies, cytoplasmic inclusions that are highly enriched with ubiquitin. Synphilin-1, α-synuclein, and Parkin represent the major components of Lewy bodies and are involved in the pathogenesis of Parkinson's disease. Synphilin-1 is an α-synuclein-binding protein that is ubiquitinated by Parkin. Recently, a mutation in the synphilin-1 gene has been reported in patients with sporadic Parkinson's disease. Although synphilin-1 localizes close to synaptic vesicles, its function remains unknown. To investigate the proteins that interact with synphilin-1, the present study performed a yeast two-hybrid screening and identified a novel interacting protein, Siah-1 ubiquitin ligase. Synphilin-1 and Siah-1 proteins were endogenously expressed in the central nervous system and were found to coimmunoprecipitate each other in rat brain homogenate. Confocal microscopic analysis revealed colocalization of both proteins in cells. Siah-1 was found to interact with the N terminus of synphilin-1 through its substrate-binding domain and to specifically ubiquitinate synphilin-1 via its RING finger domain. Siah-1 facilitated synphilin-1 degradation via the ubiquitin-proteasome pathway more efficiently than Parkin. Siah-1 was found to not facilitate ubiquitination and degradation of wild type or mutant α-synuclein. Synphilin-1 inhibited high K+-induced dopamine release from PC12 cells. Siah-1 was found to abrogate the inhibitory effects of synphilin-1 on dopamine release. Such findings suggest that Siah-1 might play a role in regulation of synphilin-1 function.

formation of ubiquitin-positive cytoplasmic inclusions that resemble LBs, suggesting that synphilin-1 might link ␣-synuclein and Parkin to a common pathogenic mechanism (20). A recent report demonstrated that Dorfin, an E3 for mutant superoxide dismutase-1, also interacts with and ubiquitinates synphilin-1 (21). Like Parkin, Dorfin contains two RING finger domains and an IBR domain. Furthermore, Dorfin is colocalized with ubiquitin in LBs of PD, suggesting that Dorfin is also implicated in the pathogenesis of PD.

EXPERIMENTAL PROCEDURES
Yeast Two-hybrid Screening-The full-length human synphilin-1 cDNA was cloned by a library screening as described previously (39). The coding region of synphilin-1 cDNA was subcloned into the yeast two-hybrid vector pGBKT-7 (Clontech), which was in-frame fused to the GAL4-binding domain sequence. The recombinant plasmid was introduced into the yeast strain AH109. A rat brain cDNA library constructed in pGAD10 (Clontech) was introduced into the yeast strain expressing the synphilin-1 fusion protein, and ϳ1.0 ϫ 10 6 transformants were screened for growth on SD plate media lacking tryptophan, leucine, histidine, and adenosine. Positive clones were detected by a ␤-galactosidase assay. An ␣-Synuclein construct in pGAD10 was used as a positive control for the screening. To eliminate false positives, plasmid DNA from positive clones was purified, amplified, and retransformed into the yeast strain expressing synphilin-1 protein fused to the GAL4-binding domain. The positive clones in this second screening were subjected to DNA sequencing. A BLAST search revealed that one of the isolated positive clones contained a fragment nearly identical to the Siah-1a gene.
Vectors and Antibodies-Full-length human Siah-1 cDNA was amplified from a human brain cDNA library (Stratagene) by PCR using a Pfu DNA polymerase (Stratagene) and the following forward and reverse primers: 5Ј-GAA TTC TCG AGA TGA GCC GTC AGA CTG CTA C-3Ј (F1) and 5Ј-GCG ATC TAG ATC AAC ACA TGG AAA TAG TTA CAT TGA TGC C-3Ј (R1). The DNA fragment obtained from PCR was subcloned in a pcDNA3 vector (Invitrogen) in-frame with the Myc tag sequence (pcDNA3-Myc-Siah-1). Human Siah-1 contains an N-terminal RING finger domain (amino acids (aa) 40 -75), followed by a conserved cysteine/histidine-rich region (aa 98 -152), which might represent two zinc fingers (ZF). Siah-1 contains a substrate-binding domain (SBD) (aa 90 -282) at the C terminus that interacts with a number of substrate proteins. The cDNA encoding Siah-1 mutants, such as RING finger domain-deleted Siah-1 (Siah⌬N) and SBD-deleted Siah-1 (Siah⌬C), were amplified by PCR using the following forward and reverse primers: 5Ј-GAA TTC TCG AGA CAT GTT GTC CAA CTT GCC GG-3Ј (F2) and R1 for Siah⌬N and F1 and 5Ј-GCG ATC TAG ATC AGG TTG TAA TGG ACT TAT GCT G-3Ј (R2) for Siah⌬C. The DNA fragment obtained from PCR was subcloned into a pcDNA3 vector in-frame with the Myc tag sequence (pcDNA3-Myc-Siah⌬N and pcDNA3-Myc-Siah⌬C). To construct the tetracycline (Tet)-repressible expression vectors for Siah-1 and Siah⌬C, each cDNA encoding Siah-1 or Siah⌬C was inserted into a pTet splice vector (Invitrogen) in-frame with the Myc tag sequence (pTet-splice-Myc-Siah-1 and pTet-splice-Myc-Siah⌬C). Ubiquitin (Ub) cDNA was amplified by PCR and subcloned into a pcDNA3 vector in-frame with the FLAG tag sequence (pcDNA3-FLAG-Ub). The sequences of all constructs were confirmed by DNA sequencing for both complementary strands. The plasmid pcDNA3.1-Myc-Parkin vector was provided by Drs. N. Hattori and Y. Mizuno. The plasmid pcDNA-Myc-␤-TrCP/FWD1 was provided by Drs. S. Kishida and A. Kikuchi. Expression vectors for HA-tagged synphilin-1 (pcDNA3-HAsynphilin-1) and HA-tagged ␣-synuclein (pcDNA3-HA-␣-synuclein; wild type, A53T, A30P) were generated as described previously (39,40). An anti-synphilin-1 polyclonal antibody was provided by Dr. E. Iseki. An anti-␣-synuclein monoclonal antibody was obtained from Pharmingen. An anti-FLAG monoclonal antibody (M2) was obtained from Sigma. An anti-HA monoclonal antibody (F-7), an anti-HA polyclonal antibody (Y-11), an anti-Myc monoclonal antibody (9E10), and normal rabbit IgG (NRI) were obtained from Santa Cruz Biotechnology, Inc. Siah-1 antiserum was raised against Siah-1 protein fused with glutathione Stransferase (GST) at the N terminus (GST-Siah). The antigen was thoroughly mixed with Freund's complete adjuvant to produce a suspension, which was intradermally injected into rabbits. Identical immunogen was injected five times every week. Phlebotomy was ultimately performed to collect serum. Antiserum was filtered through a GST coupling Hitrap NHS activated HP column (Amersham Biosciences) to eliminate anti-GST antibodies. To purify anti-Siah-1 antibodies, pre-cleared serum was filtered through another column coupling with GST-Siah-1. Siah-1 antibodies were subsequently eluted from the column and dialyzed with phosphate-buffered saline.
Pulse-Chase Assay-A pulse-chase assay was performed as described previously (42,43). HEK 293 cells were transfected with pcDNA3-HAsynphilin-1 and pcDNA3-Myc-Siah-1, pcDNA3-Myc-Siah⌬N, pcDNA3-Myc-Parkin, pcDNA-Myc-␤-TrCP/FWD1, or empty pcDNA3-Myc vectors. After 24 h, cells were cultured with 20 M MG132 for 8 h; cycloheximide (Sigma) was subsequently added to the medium to yield a final concentration of 40 M, which would inhibit new synthesis of synphilin-1. The cells were cultured for chase intervals of 0, 2, 4, 6, 8, 12, 18, 24, and 48 h and harvested in the lysis buffer after the appropriate chase time. An equal amount of protein from each lysate was separated on SDS-PAGE and immunoblotted with an anti-HA antibody. The degree of synphilin-1 expression was quantitated by densitometric analysis with NIH Image software.
For the precipitation assay, HEK 293 cells transfected with pcDNA3-HA-synphilin-1, pcDNA3-Myc-Siah-1, or pcDNA3-HA-␣synuclein were lysed and precipitated with various GST fusion proteins or GST alone as described previously (39). Each binding assay was conducted with 10 g of GST fusion protein bound to glutathione-Sepharose 4B. After the lysate protein content was normalized and precipitated with GST fusion proteins, bound proteins were separated on SDS-PAGE and immunoblotted with anti-HA or anti-Myc antibodies. GST fusion proteins used for the binding assays were stained by Coomassie Brilliant Blue R250 (Sigma).
Dopamine Release Assay-PC12-Tet-Siah cells, PC12-Tet-⌬C cells, or PC12-Tet cells were plated onto 35-mm dishes at a density of 10 6 cells per dish. In a standard experiment, cells were subsequently transfected with the indicated plasmid DNA using LipofectAMINE 2000 and grown with or without Tet for 24 h. For a dopamine release assay, cells were washed twice with a low K ϩ solution (20 mM HEPES-NaOH, pH 7.4, 140 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , and 11 mM glucose) and incubated in the low K ϩ solution for 2 min. The medium was subsequently replaced with 1 ml of the low K ϩ solution or 1 ml of a high K ϩ solution (20 mM HEPES-NaOH, pH 7.4, 85 mM NaCl, 60 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , and 11 mM glucose) for 0, 2, 4, 6, 10, and 16 min; the solution was then collected. Concentrations of dopamine, dihydroxyphenylacetic acid (DOPAC), and homovanilic acid (HVA) were determined by high pressure liquid chromatography (HPLC) using a reverse-phase column and FIG. 1. Siah-1 interaction with synphilin-1. A, synphilin-1 and Siah-1 coimmunoprecipitation. Lysates prepared from HEK 293 cells transfected with either empty pcDNA3-HA vector (HA vector) or pcDNA3-HA-synphilin-1 (HAsynph) and either pcDNA3-Myc-Siah-1 (Myc-Siah) or pcDNA3-Myc-Siah⌬N (Myc-Siah⌬N) were immunoprecipitated (IP) with an anti-HA antibody or control normal rabbit IgG (NRI). Immunoprecipitates were divided into two parts and analyzed by Western blotting (WB) with anti-Myc and anti-HA antibodies. Molecular mass markers are indicated on the right. B, coimmunoprecipitation of Siah-1 and synphilin-1 in rat brain homogenate. Rat brain homogenate was subjected to IP with NRI, anti-GST, anti-synphilin-1, or anti-Siah-1 antibodies followed by anti-synphilin-1 and anti-Siah-1 WB. Synphilin-1 migrates near the 83-kDa marker. Molecular mass markers are indicated on the right. an electrochemical detector system (Eicom) as described previously (45). Dopamine, DOPAC, and HVA concentrations for each sample were normalized for the total cell number for each condition. Dopamine release was calculated as a percentage of the total intracellular dopamine contents. Data were calculated from four independent experiments. Statistic analysis was performed with one-way analysis of variance.

Identification of Siah-1 as a Protein That Interacts with
Synphilin-1-To identify proteins that interact with synphilin-1, yeast two-hybrid screening of a rat brain cDNA library was performed using full-length synphilin-1 as bait. From 1.0 ϫ 10 6 library transformants, a positive clone that contained a fragment nearly identical to the Siah-1a gene was obtained. Rat Siah-1a is a RING finger-type E3 that shares 99.6% amino acid identity with human Siah-1. The full-length human Siah-1 cDNA was isolated from a human brain cDNA library by PCR.
Identification of Domains Involved in Siah-1-Synphilin-1 Association-To define the specific domain of Siah-1 responsible for interaction with synphilin-1, a series of GST fusion proteins containing various truncations of the Siah-1-conserved region were generated. HEK 293 cells transfected with expression vectors for HA-synphilin-1 were lysed and precipitated with various GST fusion Siah-1 proteins, such as GST-Siah, GST-RING-ZF, GST-ZF, GST-SBD, GST-SBD-S, and GST-SBD-SS, as well as GST alone as a control ( Fig. 2A). Anti-HA immunoblotting revealed that GST fusion Siah-1 proteins containing the SBD exclusively precipitated synphilin-1 (Fig. 2B). Such results suggest that Siah-1 binds to synphilin-1 through its SBD. Furthermore, the centrally located polypeptide in the SBD (aa 180 -240; SBD-SS) was found to be necessary and sufficient for binding to synphilin-1 (Fig. 2B).
To identify the binding site of synphilin-1 to Siah-1, HEK 293 cells transfected with expression vectors for Myc-Siah-1 were lysed and precipitated with various GST fusion synphilin-1 proteins ( Fig. 2A). Anti-Myc immunoblotting revealed that only the N-terminal residues 1-202 of synphilin-1 (synphilin-N1) could precipitate Siah-1 (Fig. 2C). In contrast, other regions of synphilin-1, including the ANK repeats or C-terminal domain, were unable to bind to Siah-1. It was concluded that the SBD of Siah-1 and the N terminus of synphilin-1 interact with each other.

FIG. 2. Identification of the binding sites involved in Siah-1-synphilin-1 interaction.
A, schematic representation of the primary structures and constructs of Siah-1 and synphilin-1 utilized for the binding assay. Siah-1 contains a RING finger domain, two zinc fingers, and a substrate-binding domain (SBD). Synphilin-1 contains six ankyrin (ANK) repeats, a coiled-coil domain, and an ATP/GTPbinding site. B, mapping of the synphilin-1-binding site of Siah-1. HEK 293 cells transfected with pcDNA3-HA-synphilin-1 were lysed and precipitated (PT) with either the indicated GST fusion Siah proteins or GST alone as a control. Top, bound proteins were analyzed by WB with an anti-HA antibody. Bottom, the amount of GST fusion proteins used for the assay was detected by Coomassie staining. C, identification of the Siahbinding site of synphilin-1. HEK 293 cells transfected with pcDNA3-HA-Siah-1 were lysed and precipitated with either the indicated GST fusion synphilin-1 proteins or GST alone as a control. Top, bound proteins were analyzed by WB with an anti-HA antibody. Bottom, the amount of GST fusion proteins used for the assay was detected by Coomassie staining.
Siah-1 was expressed in all regions in the central nervous system, including the cerebral cortex, hippocampus, striatum, cerebellum, medulla, and spinal cord. In the rat brain, the anti-Siah-1 antibody recognized a single band of ϳ35 kDa, similar to the predicted molecular mass of Siah-1 protein (32 kDa), that disappeared when the antibody was preincubated with antigen (GST-Siah protein) (Fig. 3). Synphilin-1 was enriched in all regions in the brain, whereas ␣-synuclein was enriched in the cerebral cortex, hippocampus, and striatum. Such results demonstrate that both Siah-1 and synphilin-1 are coexpressed in the brain.
Colocalization of Siah-1 with Synphilin-1 in the Cytoplasm-Because Siah-1 interacted with synphilin-1 in the brain, endogenous Siah-1 and synphilin-1 localization was investigated with an immunofluorescent study. Confocal microscopic analysis revealed that Siah-1 localized in the cytoplasm of human dopaminergic SH-SY5Y cells. Synphilin-1 was observed as a cytoplasmic ring-like appearance as described previously (11) (Fig. 4A). Such immunofluorescence disappeared when the antibodies were preincubated with each antigen. Furthermore, a double-staining immunofluorescent study was performed to determine whether Siah-1 colocalized with synphilin-1. Myc-Siah-1 was concomitantly expressed with HA-synphilin-1 by transient transfection of HEK 293 cells. Confocal microscopic analysis revealed that synphilin-1 formed cytoplasmic inclusions (Fig. 4B) and that Siah-1 was distributed peripheral to the synphilin-1 inclusions. An overlay of Siah-1 and synphilin-1 staining demonstrated colocalization of both proteins.
pression vectors for FLAG-Ub and Myc-Siah-1, and the cell lysates were immunoprecipitated with an anti-Myc antibody. The procedure was followed by immunoblotting with an anti-FLAG antibody to detect ubiquitin-conjugated Siah-1. In cells coexpressing Myc-Siah-1 and FLAG-Ub, ubiquitinated high molecular mass bands were detected in the anti-Myc immunoprecipitates, suggesting that Siah-1 was also ubiquitinated (Fig. 5B). Such results suggest that Siah-1 itself was also degraded through the ubiquitin-proteasome pathway as described previously (46).
In addition, to investigate whether Siah-1 would facilitate synphilin-1 degradation, a pulse-chase assay was performed. In cells transfected with pcDNA3-HA-synphilin-1 and pcDNA3-Myc-Siah-1, the synphilin-1 degradation rate was much higher than that of cells transfected with pcDNA3-HAsynphilin-1 and empty pcDNA3-Myc vectors. Moreover, the half-life of synphilin-1 was 8 h for cells cotransfected with expression vectors for Siah-1, whereas that for cells cotransfected with empty vectors was 44 h (Fig. 5C). Nonetheless, the synphilin-1 degradation rate in cells coexpressing Myc-Siah⌬N was slower than that in cells cotransfected with empty vectors, suggesting that synphilin-1 degradation was inhibited by a dominant-negative effect of Siah⌬N on endogenous Siah-1 (Fig.  5C). Such results are consistent with the notion that the RING finger domain of E3 is required for substrate ubiquitination that leads to degradation. As a negative control, ␤-TrCP/ FWD1, an E3 for ␤-catenin, was utilized for the pulse-chase assay. Both Siah-1 and ␤-TrCP/FWD1 independently facilitate ␤-catenin degradation (35,47,48). In cells transfected with HA-synphilin-1 and Myc-␤-TrCP/FWD1, the synphilin-1 degradation rate was almost the same as that of cells transfected with empty vectors (Fig. 5C), suggesting that Siah-1 specifically facilitated synphilin-1 degradation. Because Parkin has also been reported to regulate synphilin-1 degradation (20), the present study compared the effects of Siah-1 and Parkin on the synphilin-1 degradation rate. It was noted that the synphilin-1 degradation rate for cells coexpressing Myc-Siah-1 was higher than that for cells coexpressing Myc-Parkin. Moreover, the half-life of synphilin-1 was 8 h for cells expressing Siah-1 and 18 h for cells expressing Parkin (Fig. 5C). Siah-1 and Parkin cellular expression levels were similar (data not shown). Such results suggest that Siah-1 facilitates synphilin-1 degradation more efficiently than Parkin.
Effects of Siah-1 on Dopamine Release-Siah-1 is known to be distributed in the central nervous system. In addition, recent papers have demonstrated that Siah-1 is involved in synaptic transmission (12,38,49). The present study examined whether Siah-1 affects dopamine release. Monoclonal cell lines, i.e. PC12-Tet-Siah (clones PS1-4) and PC12-Tet-⌬C (clones PDC1-4), that have Tet-repressible expression systems for Siah-1 and Siah⌬C, respectively, were established. Immunoblotting with an anti-Myc antibody demonstrated that Tet in the culture medium could negatively regulate Siah-1 and Siah⌬C protein cellular expression (PS1 and PDC1). Expressions were found to be very low at 500 ng/ml of Tet (SiahϪ or ⌬CϪ), moderate at 50 ng/ml of Tet (Siahϩ), and high in the absence of Tet (Siahϩϩ or ⌬Cϩϩ) (Fig. 8A). Siah-1 reduced the total intracellular dopamine content without increasing dopamine metabolites such as DOPAC or HVA in PC12-Tet-Siah cells (PS1). In contrast, the total intracellular dopamine content in PC12-Tet-⌬C cells (PDC1) in the absence of Tet was similar to that in PC12-Tet cells, suggesting that Siah-1 might inhibit dopamine biosynthesis (Fig. 8A). Similar to PC12-Tet cells, high K ϩ (60 mM) stimulation induced dopamine release from PC12-Tet-Siah (PS1) and PC12-Tet-⌬C (PDC1) cells, whereas low K ϩ (4.7 mM) stimulation induced little dopamine release (Fig. 8B). High K ϩ -induced dopamine release was rapid and reached a plateau at ϳ6 min. Nonetheless, maximum extracellular dopamine levels were decreased in relation to the decreased total intracellular dopamine contents resulting from Siah-1 expression (Fig. 8B). Siah-1 did not affect dopamine release kinetics, which was calculated as a percentage of the total intracellular dopamine content. Independent from Siah-1 expression, ϳ80 -90% of the total intracellular dopamine content was released within 6 min (Fig. 8B). Similar results were obtained from other monoclonal cell lines (PS1-4, PDC1-4) (Fig. 8C).
Synphilin-1 Inhibits High K ϩ -induced Dopamine Release-The role of synphilin-1 in dopamine release is poorly understood. The present study also examined whether synphilin-1 might be involved in dopamine release. PC12-Tet-Myc-Siah-1 cells were transfected with empty expression vectors or expression vectors for either synphilin-1 or ␣-synuclein, which were grown in the presence of Tet to inhibit Siah-1 expression (SiahϪ) and then stimulated with high K ϩ solution for 1 min. Synphilin-1 expression reduced the total intracellular dopamine content (Fig. 9A) without increasing dopamine metabo-FIG. 6. Siah-1-mediated ubiquitination of synphilin-1. A, lysates prepared from HEK 293 cells cotransfected with pcDNA3-HA-synphilin-1, pcDNA3-FLAGubiquitin, pcDNA3-Myc-Siah-1, pcDNA3-Myc-Siah⌬N, pcDNA3-Myc-Siah⌬C, or pcDNA3.1-Myc-Parkin were immunoprecipitated with an anti-HA antibody (HA-IP). Immunoprecipitates were divided into two parts and analyzed by WB with anti-FLAG and anti-HA antibodies. Expression of Myc-Siah-1, Myc-Siah⌬N, Myc-Siah⌬C, or Myc-Parkin in each total cell lysate was detected by WB with an anti-Myc antibody. FLAG-Ub-conjugated proteins in each total cell lysate were detected by WB with an anti-FLAG antibody. Molecular mass markers are indicated on the right. B, lysates prepared from HEK 293 cells cotransfected with pcDNA3-HA-synphilin-1, pcDNA3-FLAG-ubiquitin, and pcDNA-Myc-␤-TrCP/FWD1 were subjected to HA-IP; the ubiquitination assay was performed in the same manner as A.
lites, such as DOPAC and HVA (data not shown). ␣-Synuclein expression also reduced the total intracellular dopamine content as described previously (10). Because the present work suggests that Siah-1 facilitates synphilin-1 degradation, it can also be said that Siah-1 might negatively regulate synphilin-1 function. Upon Siah-1 coexpression with synphilin-1 and ␣-synuclein, a state attained by removing Tet from the medium (Siahϩ), the total intracellular dopamine content was further reduced (Fig. 9A). Expression of synphilin-1 and ␣-synuclein inhibited the high K ϩ -induced increase in extracellular dopamine levels, in proportion to the decreased total intracellular dopamine contents (Fig. 9B). With Siah-1 coexpression, the high K ϩ -induced increase in extracellular dopamine levels was further reduced. Nonetheless, when dopamine release kinetics were evaluated by percentage of dopamine release, high K ϩinduced dopamine release from cells expressing synphilin-1 was depressed compared with controls, whereas release from cells expressing ␣-synuclein was almost equal to controls (Fig.  9C), suggesting that synphilin-1 might inhibit dopamine release. Furthermore, when Siah-1 expression in the same cells was induced by removing Tet from the medium (Siahϩ), dopamine release increased to control levels (Fig. 9C). Such results suggest that Siah-1 might abrogate the inhibitory effect of synphilin-1 on dopamine release by facilitating synphilin-1 degradation via the ubiquitin-proteasome pathway. Similar results were obtained from other monoclonal cell lines (Fig. 9D). DISCUSSION The present study identified Siah-1 as a binding partner for synphilin-1 that is implicated in the pathogenesis of Parkinson's disease. Siah/Sina family proteins are evolutionarily conserved E3 ubiquitin ligases that participate in regulating ubiquitination and proteasome-dependent degradation of multiple proteins. Siah-1 contains a RING finger domain at the N terminus that is required for interacting with E2s and a SBD at the C terminus that is required for substrate binding. Binding assays using GST fusion proteins indicated that Siah-1 also bound to synphilin-1 via the SBD. In particular, amino acid residues 180 -240 in the SBD are necessary and sufficient for synphilin-1 binding. Recently, a binding motif for Siah, RPVAXVXPXXR, was identified (50). Furthermore, the core sequence PXAXVXP was found in the Siah interacting proteins SIP, OBF-1, DCC, and TIEG1, with more degenerate consensus sequences found in NUMB, Vav, Kid, and N-CoR (50). The most conserved residues in the motif appear to be VXP; mutagenesis of both of these residues reduced or abrogated Siah binding (50). The GST fusion protein binding assays of the present study indicate that the SBD of Siah-1 binds to the N-terminal region (aa 1-202) of synphilin-1, within which a consensus sequence, PXXXVXP, is located at residues 74 -80, suggesting the presence of a binding motif.
Pulse-chase and ubiquitination assays demonstrated that Siah-1 specifically facilitates synphilin-1 ubiquitination and degradation. Although Parkin also exerts a similar function, synphilin-1 ubiquitination levels achieved with Siah-1 were higher than those attained with Parkin. In addition, synphilin-1 degradation rates achieved with Siah-1 were also faster than those attained with Parkin. Such results suggest that Siah-1 functions as a more efficient E3 for synphilin-1 than Parkin. Because Siah proteins interact with E2s, such as UbcH5, UbcH8, or UbcH9, and Parkin interacts with UbcH6, UbcH7, or Ubc12, interaction with different E2s might account for the distinct degradation rates. In addition to Siah-1 and Parkin, a recent study demonstrated that Dorfin also ubiquitinates synphilin-1 (21). Accordingly, three RING finger-type E3s, i.e. Siah-1, Parkin, and Dorfin, have been found to target synphilin-1 for ubiquitination and degradation.
The functional similarity among Siah-1, Parkin, and Dorfin raises important questions regarding redundancy and physiological significance of multiple pathways facilitating synphilin-1 degradation. Parkin and Dorfin bind the central portion of synphilin-1, which contains ANK repeats, a coiled-coil domain, and an ATP/GTP-binding site. Nonetheless, Siah-1 was not found to bind to that portion. As mentioned above, a specific peptide motif that mediates interaction of each E3 with a range of substrate proteins has been elucidated. Accordingly, it remains possible that if a single protein contains multiple peptide motifs for different E3s, such a protein could be ubiquitinated by multiple E3s, leading to degradation by multiple, synergistic mechanisms. It appears uncertain whether three proteins can simultaneously bind synphilin-1, but it remains possible that Siah-1 might synergistically ubiquitinate synphilin-1 with Parkin or Dorfin.
On the other hand, synphilin-1 is initially distributed in the cell bodies of immature neurons, subsequently becoming redistributed toward presynaptic nerve terminals during development after birth (51). It remains possible that Siah-1, Parkin, and Dorfin independently act as E3s for synphilin-1, depending on the distribution or developmental stage, despite concurrent FIG. 8. Effects of Siah expression on high K ؉ -induced dopamine release. A, a PC12-Tet-Siah cell line, which exhibits a Tet-repressible Siah-1 expression system, was grown in the presence of 0, 50, or 500 ng/ml of Tet for 24 h. PC12-Tet-⌬C cell line and PC12-Tet cell line (Cont and Cont-T) were grown in the presence of 0 or 500 ng/ml of Tet. Anti-Myc WB demonstrated that Siah-1 or Siah⌬C protein expression is very low at 500 ng/ml of Tet (SiahϪ or ⌬CϪ), moderate at 50 ng/ml of Tet (Siahϩ), and high in the absence of Tet (Siahϩϩ or ⌬Cϩϩ) (left). The total intracellular dopamine, DOPAC, and HVA contents were measured by HPLC and normalized for total cell number (right). Data are presented as means Ϯ S.D. from four independent experiments. Statistical analysis was performed by one-way analysis of variance. *, p Ͻ 0.05. B, cells were treated with either high K ϩ solution (H) or low K ϩ solution (L). Extracellular dopamine levels were measured at the indicated time points by HPLC and either normalized for total cell number (left) or calculated as a percentage of the total intracellular dopamine content (% dopamine release) (right). C, the total intracellular dopamine and % dopamine release at 1 min in the presence and absence of Tet (500 ng/ml) in other monoclonal cell lines (PS1-4 and PDC1-4). Data are presented as means from four independent experiments in each cell line. expression of all three E3s in the adult brain. Future studies will need to elucidate the reason underlying synphilin-1 ubiquitination by multiple E3s.
A previous study demonstrated that synphilin-1 strongly associated with synaptic vesicles; synphilin-1 was found to be located close to synaptic vesicles using electron microscopy (51). The present study demonstrated that Siah-1 is distributed in the central nervous system, and recent papers have shown that Siah-1 binds to group 1 metabotropic glutamate receptors (mGluR1 and mGluR5), which are involved in the regulation of synaptic transmission and regulates mGluR-mediated signaling (38). In addition, Siah-1 attenuates mGluR-mediated calcium current modulation at the synaptic terminal (49). Furthermore, it was demonstrated both that Siah-1 binds to synaptophysin and that endogenous Siah-1 is localized on synaptic-like microvesicles in PC12 cells (12). Such findings suggest that Siah-1 might play a role in neurotransmitter release by facilitating ubiquitination of synaptic vesicle proteins, to include synphilin-1. In the present work, Siah-1 and synphilin-1 reduced intracellular dopamine content without increasing dopamine metabolites, suggesting that both proteins might inhibit dopamine biosynthesis. Siah-1 did not affect high K ϩinduced dopamine release from PC12 cells. Synphilin-1 was found to moderately inhibit high K ϩ -induced dopamine release from cells, whereas coexpression of Siah-1 abrogated the inhibitory effect of synphilin-1 on dopamine release. Such a result suggests that association of Siah-1 with synphilin-1 might be involved in the regulation of dopamine release. It must be said FIG. 9. Synphilin-1-mediated inhibition of dopamine release was abrogated by Siah-1. A, PC12-Tet, PC12-Tet-Siah, or PC12-Tet-⌬C cells were transfected with empty expression vectors (Mock) or expression vectors for either HA-synphilin-1 (synph) or HA-␣synuclein (␣S). Transfected cells were grown in either the presence or absence of 500 ng/ml of Tet to inhibit (SiahϪ and Siah⌬CϪ) or induce (Siahϩϩ and Siah⌬Cϩϩ) Siah-1 expression, respectively. The total intracellular dopamine contents were measured by HPLC and normalized for total cell number. B and C, cells were treated with high K ϩ solution for 1 min. Extracellular dopamine levels were measured by HPLC and either normalized for total cell number (B) or calculated as a percentage of the total intracellular dopamine contents that were shown in A (C). Siah-1, Siah⌬C, synphilin-1, and ␣-synuclein expression was confirmed by WB with anti-Myc and anti-HA antibodies (C). Data are presented as means Ϯ S.D. from four independent experiments. Statistical analysis was performed by one-way analysis of variance. *, p Ͻ 0.05. D, dopamine release assays were performed using other monoclonal cell lines. The table presents the results of % dopamine release at 1 min from multiple monoclonal cell lines in the presence of 0 or 500 ng/ml of Tet. Data are presented as means from four independent experiments in each cell lines. Statistical analysis was performed by one-way analysis of variance. *, p Ͻ 0.05. that the effects of Siah-1 and synphilin-1 on dopamine release is relatively small. Because multiple proteins close to synaptic vesicles are involved in neurotransmitter release, our observation might only represent one aspect of regulation of dopamine release. Further investigation should be performed to clarify the role of Siah-1 and synphilin-1 in neurotransmission.
Recently, Siah-1 has been shown to form an SCF-type complex with Skp1, Ebi, SIP, and APC, to facilitate ␤-catenin degradation in a p53-dependent manner (35,48,52). A recent report demonstrated that Parkin also functions in a multiple E3 complex that includes the F-box/WD repeat protein hSel-10 and Cullin-1 (53). In the present study, Siah-1 was found to directly bind to synphilin-1, but it remains possible that SCF box-type E3s containing Siah-1 or Parkin also facilitate synphilin-1 ubiquitination and affect synaptic transmission. It is becoming increasingly clear that the ubiquitin-proteasome pathway is involved in synaptic function and neurodegeneration. Nonetheless, it is not understood how alteration of synphilin-1 expression levels by multiple E3s relates to synaptic function and PD pathogenesis. We are currently attempting to produce Siah knockout mice, which should prove very useful for enhancement of the understanding of both Siah protein function and the mechanisms by which Siah regulates synaptic function through the ubiquitin-proteasome pathway.