Proteasomal inhibition reduces parkin mRNA in PC12 and SH-SY5Y cells

Parkinsonism and Related Disorders 15 (2009) 220e225 www.elsevier.com/locate/parkreldis

Proteasomal inhibition reduces parkin mRNA in PC12 and SH-SY5Y cells Andreas Koch a,1, Klaus Lehmann-Horn a,1,2, Justus C. Da¨chsel a,1,3, Thomas Gasser b, Philipp J. Kahle b, Christoph B. Lu¨cking a,* a

Department of Neurology, Klinikum Grosshadern, Ludwig Maximilians University, Marchioninistr. 15, 81377 Munich, Germany b Hertie Institute for Clinical Brain Research, Department for Neurodegenerative Diseases, University Clinics Tu¨bingen, Otfried-Mu¨ller-Strasse 27, 72076 Tu¨bingen, Germany Received 15 January 2008; received in revised form 10 May 2008; accepted 10 May 2008

Abstract Mutations in the gene encoding the E3 ubiquitin-protein ligase parkin have been shown to be a common genetic cause of familial early-onset Parkinson’s disease (PD). In addition to its function in the ubiquitineproteasome system (UPS), parkin has been ascribed general neuroprotective properties. Stress and mutation induced decreases in parkin solubility leading to compromised cytoprotection have recently been reported. We systematically investigated whether PD-related stresses including MG132 and epoxomicin (proteasomal impairment), tunicamycin (unfolded protein stress), and rotenone (mitochondrial dysfunction) resulted in expressional changes of parkin and other E3 ubiquitin ligases (dorfin, SIAH-1). Rotenone and tunicamycin did not change parkin mRNA levels, whereas proteasomal inhibition resulted in a reduction of parkin mRNA in PC12 cells as well as in SH-SY5Y cells. Therefore, surprisingly, cells did not react with a compensatory parkin upregulation under proteasomal inhibition, although, in parallel, parkin protein shifted to the insoluble fraction, reducing soluble parkin levels in the cytosol. Since the mRNA of the parkin-coregulated gene PACRG paralleled the parkin mRNA at least partly, we suspect a promoter-driven mechanism. Our study, therefore, shows a link between proteasomal impairment and parkin expression levels in cell culture, which is intriguing in the context of the described and debated proteasomal dysfunction in the substantia nigra of PD patients. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Parkinson’s disease; Ubiquitineproteasome system (UPS); Proteasomal dysfunction; Parkin; Siah-1; Dorfin

1. Introduction Parkinson’s disease (PD) is one of the most common neurodegenerative disorders. Although the majority of cases are considered sporadic, mutations in several genes have been shown to be the disease cause in rare familial forms of PD [1]. Based on the cellular function of the proteins encoded

* Corresponding author. Tel.: þ49 89 70950; fax: þ49 89 7095 3677. E-mail address: [email protected] (C.B. Lu¨cking). 1 These authors equally contributed to this work. 2 Current address: Department of Neurology, Klinikum rechts der Isar, Technical University Munich, Ismaninger Str. 22, 81675 Munich, Germany. 3 Current address: Mayo Clinic, Department of Neuroscience, 4500 San Pablo Road, Jacksonville, FL 32224, USA. 1353-8020/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.parkreldis.2008.05.005

by these genes various mechanisms leading to the dopaminergic degeneration observed in PD are currently discussed (for review see e.g. [2]). Mutations in the gene coding for parkin are associated with autosomal-recessive, early onset PD. As an E3 ubiquitin-protein ligase, parkin is responsible for the ligation of ubiquitin to specific substrate proteins. This leads, among other effects, to targeting of the substrates to the proteasome [3]. An important role for the ubiquitineproteasome system (UPS) in PD has been supported by the finding of impaired proteasomal activity in the substantia nigra of PD brains [4]. In addition, parkin was shown to possess neuroprotective properties against various neurotoxic insults [5e8]. This might be due to parkin-mediated ubiquitinylation of key components of cellular viability regulating signalling cascades, such as the Jun kinase and NF-kB pathways [9e11].

A. Koch et al. / Parkinsonism and Related Disorders 15 (2009) 220e225

To further elucidate these protective properties we were interested in the regulation of parkin under PD-related stress conditions, namely proteasome impairment, unfolded protein stress, and mitochondrial failure. Two other E3 ubiquitinligases, dorfin and Siah-1, were included as controls. These enzymes, like parkin, have been shown to ubiquitylate Synphilin-1 [12e14], an interactor of a-Synuclein [15], and were found in Lewy bodies, the pathological hallmark of PD [13, 14]. 2. Materials and methods Non-differentiated PC12 cells (tet-off, Clontech, Karlsruhe, Germany) were cultured as previously described [16]. SH-SY5Y cells (ATCC, Manassas, USA) were cultured according to the manufacturer’s instructions. Cells were seeded out at an approximate density of 1.5  104 cells/cm2 (PC12) or 0.5  104 cells/cm2 (SH-SY5Y). Twenty-four hours later, fresh medium containing the stressors was added. Cells were harvested, centrifuged, and pellets immediately frozen at 80  C at the time points specified below. All experiments were performed at least in triplicates. In a first set of experiments, mRNA expression was analyzed after 3, 6, 9, and 18 h in PC12 cells stressed with 125 mM MG132, 4.7 mM tunicamycin or 50 mM rotenone. These concentrations were established to obtain cell death rates of 20e30% after 24 h, which allowed (a) comparable toxic conditions and (b) satisfactory mRNA amounts and quality. In a second set of experiments, MG132 was used in lower concentrations (60 and 30 mM) as well as in the commonly used concentrations of 10 mM. In addition, another e more specific e proteasome inhibitor, epoxomicin, was introduced, concentrated at 0.5 mM, which is within the range used in the literature (0.1e1.0 mM). Controls received the amount of DMSO which was used to dissolve the respective toxin. For cell death quantification, either FACS analysis (FACSCaliburÔ-Cytometer, Becton Dickinson Biosciences, Franklin Lakes, USA) after propidium iodide staining (0.5 mg/ml, for MG132 treatment) or analysis of chromatin condensation/ nuclear fragmentation after Hoechst staining (0.5 mg/ml, for tunicamycin and rotenone treatment) was performed. mRNA was isolated with RNeasy MiniKit including an on-column DNA digest (both Qiagen, Hilden, Germany) out of approximately 3 million PC12 cells and 1 million SH-SY5Y cells. For reverse transcription, cDNA synthesis kits AMV (Roche Diagnostic, Mannheim, Germany) and QuantiTect (Quiagen, Hilden, Germany) were used with random primers following the manufacturer’s instructions. Quantitative RT-PCR was performed with the Light Cycler Instrument 2.0 using Light Cycler FastStart DNA Master SYBR Green I for hot start application (both Roche Diagnostic, Mannheim, Germany), as described by the manufacturer. The genes analyzed and the primers used are given in Table 1. All primers extended over exon boundaries to exclude amplification of genomic DNA. PCR products showed a single peak in the Light Cycler melting curve. To quantify mRNA expression, we first calculated for each sample the ratio between the gene of interest and a housekeeping gene, HPRT (hypoxanthine phosphoribosyltransferase) in PC12 or GAP-DH (glyceraldehyde-3-phosphate dehydrogenase) in SH-SY5Y. We could therefore

221

control for slight differences in RT-PCR efficiency or template amount. Then, the mean and the 95% confidence interval of these relative expression values were calculated for the three independent reactions. Only expression changes of more than 2-fold between stressed vs. control cells were considered as relevant. Unless otherwise indicated, HPRT and GAP-DH values did not differ for more than 0.5 cycles in treated vs. untreated samples. For Western Blot analysis, three pellets of 3  106 PC12 cells and 1  106 SH-SY5Y cells each were pooled, lysed (RIPA buffer: Triton-X-100 1%, sodium deoxycholate 1%, SDS 1%, NaCl 0.15 M, TriseHCl 50 mM pH 7.2) homogenized and centrifuged. Equal amounts of protein (50 mg) were subjected to denaturing 10% polyacrylamide gel electrophoresis. For analysis of the insoluble fraction, RIPA-buffer containing 8 M urea was added to the pellets. Anti-parkin antibodies (Cell Signalling, Beverly, USA) diluted 1:1000, anti-actin (Sigma, Deisenhofen, Germany) diluted 1:2000 (PC12) or 1:4000 (SH-SY5Y) and secondary antibodies (anti-rabbit and anti-mouse, Dako, Hamburg, Germany) diluted 1:2000 and 1:8000, respectively, were used. For quantification of the Western blot bands, the blots were scanned with a flat bed scanner. Using the software TINA 2.08e by Raytest, the optical density was calculated after standardized manual definition of each band. Then the background was subtracted and the result divided by the surface. The resulting values were used to calculate the ratio between each parkin band and its corresponding actin band, yielding a normalized relative parkin value. We also tried to quantify the PACRG protein by using the only existing commercially available PACRG-antibody (Rockland Immunocytochemicals, Gilbertsville, PA, USA, Code 600-401-474) in dilutions down to 1:500. To determine the 258T/G promoter polymorphism in SH-SY5Y cells, the region of interest was amplified by PCR (primers: forward 5´-GCATTTGTTT AAGCTCAGGGTCTC-3´ and reverse 5´-CCTGCTGGGAGTCGTAGTTCTAA C-3´). The resulting 441 bp PCR product was analyzed by restriction enzyme analysis with AlwNI (New England Biolabs) and by direct sequencing on an Abi Prism 310 Gene Analyzer (Applied Biosystems).

3. Results In the first set of experiments, under treatment with tunicamycin or rotenone, parkin and dorfin mRNA levels did not change more than 2-fold in PC12 cells at the investigated time points (3, 6, 9, 18 h, data not shown). However, under proteasomal inhibition with MG132 (125 mM), we observed a distinct, time-dependent and strong reduction of parkin mRNA in PC12 cells, whereas dorfin mRNA levels were not regulated. A similar reduction was seen under 60, 30 and 10 mM MG132 (data not shown). Prompted by these observations, we used a more specific proteasome inhibitor (epoxomicin, 0.5 mM) and observed again a parkin mRNA reduction (Fig. 1a). We than analyzed later time points (14, 24, 38 and 48 h) which still showed a strong reduction of parkin mRNA (Fig. 1b). Finally, we analyzed a second neuronal cell model

Table 1 Genes analyzed and primers used for RT-PCR Gene

Forward primer

Reverse primer

Rat HPRT Human GAP-DH Human parkin Rat parkin Human PACRG Rat PACRG Human dorfin Rat dorfin Human Siah-1 Rat Siah-1

GAC TTT GCT TTC CTT GGT CA CAG GGC TGC TTT TAA CTC TG CGT GGA GAA AAG GTC AAG AA ACC ACA GAG GAA AAG TCA CG GAT GGG CTT TGT GAA ATG AC TGC CTG GAA GGT AGA GAT TG GCA GCA GCT GAT GAT ATA AAG TGG GAC TAA CAC AGC CAT AGA TAT GTG TTA CCG CCC ATT CT TTG GCG AGT CTT TTC GAG TGT

TTC AAA TCC CTG AAG TGC TC GAG TCC TTC CAC GAT ACC AA GGG CCT TTG CAA TAC ACA TA GGC CTT TGC AGT AGA CAA AA TCT CAG CTG ACA CAA CCA GA CTT TGA GCG TGA CAC AGA TG ACA CAG GAA TGC CAA TAA TC ATT CAG AAT GGA CCC TGC CAT TCA CAT CCA GAA GAC GCA TA AAG GCC TGA ACT CAC AGA GCT

A. Koch et al. / Parkinsonism and Related Disorders 15 (2009) 220e225

90 80 70 60 50 40 30 20 10 0

PC12 - Parkin Epoxomicin control

0

3

6

9

copies per 100 HPRT

a

copies per 10.000 HPRT

222

PC12 - PACRG

1,8 1,6 1,4 1,2 1,0 0,8 0,6 0,4 0,2 0,0

Epoxomicin control

0

18

PC12 - Dorfin Epoxomicin control

0

3

6

9

copies per 100 HPRT

copies per 100 HPRT

160 140 120 100 80 60 40 20 0

18

40 0,63

20 0

0

14

24

38

2,1

48

PC12 - Siah-1 Epoxomicin

100

control

80 60 40 20 0

0

6

9

PC12 - PACRG Epoxomicin

1,5

control

1,0

0,63

2,1

60 40 Epoxomicin

14

24

control

38

48

incubation time (h)

2,1

0,5 0,0

0

14

24

38

PC12 - Siah-1

80

18

2,0

PC12 - Dorfin 0,63

0

3

incubation time (h)

100

0

18

incubation time (h)

120

20

copies per 100 HPRT

60

copies per 100 HPRT

copies per 10.000 HPRT copies per 100 HPRT

control

Epoxomicin

9

incubation time (h)

PC12 - Parkin

80

6

120

incubation time (h)

b

3

incubation time (h)

incubation time (h)

60

48

Epoxomicin

50

control

40

2,1 0,63

30 20 10 0 0

14

24

38

48

incubation time (h)

Fig. 1. mRNA expression levels of parkin, PACRG, dorfin and Siah-1 under proteasomal impairment. (a, b) PC12 cells were treated for the indicated times with 0.5 mM of the proteasome inhibitor epoxomicin. (c) SH-SY5Y cells were treated for the indicated times with 0.5 mM epoxomicin. RNA was prepared and qRT-PCR performed for parkin, PACRG, dorfin and Siah-1, as indicated. Depicted are the mean expression values of the indicated genes relative to the respective housekeeping gene with their 95% confidence intervals. When the Light Cycler cycle number of the housekeeping gene differed for more than 0.5 cycles between the stressor and the control condition, the differences are indicated in cycle number just above the mean values. Due to the strong downregulation of parkin, these differences were considered negligible.

(SH-SY5Y cells), which also showed a parkin mRNA decrease (Fig. 1c). Dorfin and Siah-1, the latter being an additional specificity controls, were unaltered under these conditions (Fig. 1aec). To assess whether the observed decrease in parkin mRNA was possibly due to a transcriptional downregulation or was, e.g., a consequence of decreased mRNA stability, we also analyzed the mRNA levels of PACRG, the parkin co-regulated gene, driven by the same head-to-head promoter as parkin

[17]. PACRG was, at least for the time points up to 18 h coregulated with parkin in both cell lines (Fig. 1aec). In PC12 cells, this co-regulation appeared to become partly obscured at later time points (Fig. 1a,b). To investigate whether the reduction of parkin mRNA also resulted in a reduction of the parkin protein, we performed Western Blot analyses. Increasing with time, we found a distinct reduction of soluble parkin protein and a shift of parkin to the insoluble fraction, consistent with previous reports

A. Koch et al. / Parkinsonism and Related Disorders 15 (2009) 220e225

30 25 20 15 10 5 0

SY5Y - Parkin Epoxomicin

copies per 100 GAP-DH

copies per 10.000 GAP-DH

c

control

0

14

24

38

48

0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00

SY5Y - PACRG Epoxomicin control

0

14

24

38

48

incubation time (h)

SY5Y - Dorfin

SY5Y - Siah-1 5

copies per 100 GAP-DH

copies per 100 GAP-DH

incubation time (h) 3,0 2,5 2,0 1,5 1,0 0,5 0,0

223

Epoxomicin control

0

4 3 2

Epoxomicin

1

control

0

14

24

38

48

0

14

incubation time (h)

24

38

48

incubation time (h) Fig. 1 (continued).

[18] (Figs. 2 and 3). We could not detect a considerable decrease of total parkin protein (Fig. 3), which was well explained with a blocked proteasome. A statistical analysis of these values was not possible, because only one value existed for each condition due to the necessary pooling of the replicate samples.

different PD-associated cell stress conditions at different time points up to 48 h in distinct cell lines. Under the chosen conditions, rotenone did not result in any differential regulation (>2-fold) of parkin, suggesting that the pathomechanism exerted by rotenone might be independent of parkin in PC 12 cells. Also under tunicamycin treatment, parkin mRNA was not upregulated, in contrast to data of Imai et al. [19]. They found an almost 12-fold upregulation of parkin mRNA after 24 h in SH-SY5Y neuroblastoma cells treated with 10 mg/ml tunicamycin. In contrast, West et al. [20] saw no upregulation of parkin in SH-SY5Y cells after

4. Discussion To our knowledge, this is the first study that systematically investigated the expression of endogenous parkin under

PC12 500nM Epoxo

hours

14

24

38

control

10µM Mg-132 48

14

24

38

48

14

24

38

48

Parkin 52 kDa

soluble

Actin 42 kDa Parkin 52 kDa

insoluble

Actin 42 kDa

SH-SY5Y 500nM Epoxo

hours

14

24

38

10µM Mg-132 48

14

24

38

control 48

14

24

38

48

Parkin 52 kDa Actin 42 kDa

soluble

Parkin 52 kDa

Actin 42 kDa

insoluble

Fig. 2. Parkin protein levels under proteasomal impairment. Western Blot analysis showed a time-dependent decrease of soluble parkin protein in treated samples vs. controls (pooled triplicates) in PC12 and SH-SY5Y cells stressed with 10 mM MG132 or 0.5 mM epoxomicin after 14, 24, 38 and 48 h. Actin served as loading control.

A. Koch et al. / Parkinsonism and Related Disorders 15 (2009) 220e225

224

12 and 18 h of 10 mg/ml tunicamycin exposure neither in RTPCR nor in Western Blot. In our experiments, we used a different cell line (PC12 cells) and a smaller drug concentration (4 mg/ml ¼ 4.7 mM). Still, the concentration used was sufficient to cause 20e30% cell shrinkage, nuclear chromatin condensation and nuclear fragmentation, indicating apoptosis, after 24 h. Therefore, our results are in accordance with West et al. [20], underlining that the unfolded protein response does not uniformly regulate parkin at the transcriptional level in different investigated cell lines [21]. Nevertheless, parkin may be regulated post-transcriptionally as part of the unfolded protein response, as evidenced by reduced parkin phosphorylation and increased ubiquitin ligase activity after tunicamycin treatment [22]. The role of the proteasomal impairment in the pathogenesis of PD is discussed very controversially. McNaught described a specific proteasomal impairment in the substantia nigra in PD patients [4]. Furthermore, experimentally induced proteasomal impairment in rodents could reproduce several clinical and pathological features found in PD patients [23,24]. However, these findings could not be reproduced by other groups [25,26]. In our assay, proteasomal inhibition resulted in a decrease of soluble parkin protein over time, primarily attributable to a shift into the insoluble fraction. This is consistent with the recent literature where reduced parkin solubility and formation of aggregates under stress conditions were reported [27,28]. The shift to the insoluble fraction results in a reduction of soluble, functionally active parkin in the cytosol [28]. To our surprise, our cells did not show a compensatory up-regulation of parkin mRNA. Thus, a decrease of soluble parkin protein together with reduced parkin mRNA expression may result in a vicious circle that deprives proteasome-impaired cells of functional, cytoprotective parkin. PC12 + 500nM Epoxomicin

2,5

insoluble soluble

2

To our knowledge, only one study exists that investigated endogenous parkin levels under proteasomal inhibition. Biasini at al. [30] focused their interest on the analysis of parkin accumulation (endogenous vs. overexpressed) within 24 h rather than on parkin downregulation. They did not mention parkin downregulation, but their presented data were compatible with reduced soluble parkin protein and decreased parkin-mRNA under proteasomal inhibition after 24 h, in line with our observation. The mechanism of the observed parkin mRNA reduction after proteasome inhibition remains to be clarified. A negative feedback mechanism due to the accumulating insoluble parkin is possible, but less likely because the effect on parkin mRNA preceded the protein accumulation seen in the Western blots. Since PACRG, a gene regulated by the same head-to-head promoter as parkin [17], was, at least up to 18 h, co-regulated with parkin in both cell lines, we suspect a promotor-driven regulatory effect, e.g. due to decreased degradation of a transcriptional repressor under acute proteasomal inhibition. At later time points, in the PC12 cells, additional, cell specific effects have to suspected, because the PACRG mRNA levels also decreased under control conditions. This reduction might be due to the solvent DMSO, but we could not observe an increased cell death rate in the DMSO treated control cells (rarely exceeding 2%, with a maximum of 3.5% for PC12 cells at 48 h). Thus, also other factors, such as cell stress linked to the medium changes, might interfere with PACRG mRNA levels, e.g. destabilizing the RNA. An alternative explanation could be that PACRG might not be tightly co-regulated with parkin in the PC12 cells, because PC12 cells are of rat origin, where the co-regulation has not yet been formally shown. Still, the rat is closely related to the mouse, for which a coregulation has been described. Unfortunately, we were unable to detect a specific Western blot band for PACRG, which

2,5 2 1,5

1,5 1

1

0,5

0,5

0

PC12 + 10µM MG-132 insoluble soluble

14

24

38

48

0

14

incubation time (h) SY5Y + 500nM Epoxomicin 0,6 0,5 0,4 0,3 0,2 0,1 0

insoluble soluble

14

24

38

incubation time (h)

24

38

48

incubation time (h)

48

0,6 0,5 0,4 0,3 0,2 0,1 0

SY5Y + 10µM MG-132 insoluble soluble

14

24

38

48

incubation time (h)

Fig. 3. Densitometric evaluation of soluble and insoluble parkin protein levels under proteasomal impairment. Using the TINA 2.08e program, the bands of Fig. 2 were quantified in respect to the actin levels. On the y-axis, the resulting parkin/actin ratios are given. The results show that total parkin protein levels did not relevantly decrease over time under proteasomal impairment. The decrease of soluble parkin protein level was compensated by an increase of insoluble parkin protein.

A. Koch et al. / Parkinsonism and Related Disorders 15 (2009) 220e225

might have helped to better understand the co-regulation of parkin and PACRG in our cells. This lack of a specific PACRG Western blot signal was most likely due to insufficient specificity of the antibody in combination with low PACRG amounts in the our cells. Finally, genetic variation of the promoter region could further influence parkin expression [29], but we did not find the described 258T/G promoter variant in SH-SY5Y cells. In conclusion, our study shows in cell culture a negative correlation between proteasomal impairment and parkin expression levels. This is intriguing in the context of the described and debated proteasomal dysfunction in the substantia nigra of PD patients. This proteasomal dysfunction in PD brains might therefore not only result in an accumulation of toxic proteins but also in the reduction of soluble parkin levels. The verification of this speculation in PD brains will be the challenge of the future. Not only parkin protein amounts are low in neurons, but these studies should also be performed using isolated dopaminergic neurons, to prevent the observation of changes in glial cells or other neuronal populations. Acknowledgements This work was supported by the German National Genome Network (NGFN; Grant number 01GS0116, German Ministry for Education and Research). We want to thank Dr. Heike Pohla (Laboratory for tumor immunology of the Department of Urology, Klinikum Großhadern LMU Munich) for providing the Light Cycler and FACS instruments and Dr. Zsolt Ruzsics (Gene Center of the LMU Munich) for providing the Nanodrop photometer. References [1] Gasser T. Genetics of Parkinson’s disease. Curr Opin Neurol 2005;18: 363e9. [2] Tan EK, Skipper LM. Pathogenic mutations in Parkinson disease. Hum Mutat 2007;28:641e53. [3] Shimura H, Hattori N, Kubo S, Mizuno Y, Asakawa S, Minoshima S, et al. Familial Parkinson disease gene product, parkin, is a ubiquitinprotein ligase. Nat Genet 2000;25:302e5. [4] McNaught KS, Belizaire R, Isacson O, Jenner P, Olanow CW. Altered proteasomal function in sporadic Parkinson’s disease. Exp Neurol 2003;179:38e46. [5] Haywood AF, Staveley BE. Mutant alpha-synuclein-induced degeneration is reduced by parkin in a fly model of Parkinson’s disease. Genome 2006;49:505e10. [6] Lo Bianco C, Schneider BL, Bauer M, Sajadi A, Brice A, Iwatsubo T, et al. Lentiviral vector delivery of parkin prevents dopaminergic degeneration in an alpha-synuclein rat model of Parkinson’s disease. Proc Natl Acad Sci USA 2004;101:17,510e5. [7] Yang H, Zhou HY, Li B, Chen SD. Neuroprotection of Parkin against apoptosis is independent of inclusion body formation. Neuroreport 2005;16:1117e21. [8] Yang YX, Muqit MM, Latchman DS. Induction of parkin expression in the presence of oxidative stress. Eur J Neurosci 2006;24:1366e72. [9] Jiang H, Ren Y, Zhao J, Feng J. Parkin protects human dopaminergic neuroblastoma cells against dopamine-induced apoptosis. Hum Mol Genet 2004;13:1745e54. [10] Cha GH, Kim S, Park J, Lee E, Kim M, Lee SB, et al. Parkin negatively regulates JNK pathway in the dopaminergic neurons of Drosophila. Proc Natl Acad Sci USA 2005;102:10,345e0.

225

[11] Henn IH, Bouman L, Schlehe JS, Schlierf A, Schramm JE, Wegener E, et al. Parkin mediates neuroprotection through activation of IkappaB kinase/nuclear factor-kappaB signaling. J Neurosci 2007;27:1868e78. [12] Chung KK, Zhang Y, Lim KL, Tanaka Y, Huang H, Gao J, et al. Parkin ubiquitinates the alpha-synuclein-interacting protein, synphilin-1: implications for Lewy-body formation in Parkinson disease. Nat Med 2001;7:1144e50. [13] Ito T, Niwa J, Hishikawa N, Ishigaki S, Doyu M, Sobue G. Dorfin localizes to Lewy bodies and ubiquitylates synphilin-1. J Biol Chem 2003;278:29,106e14. [14] Liani E, Eyal A, Avraham E, Shemer R, Szargel R, Berg D, et al. Ubiquitylation of synphilin-1 and alpha-synuclein by SIAH and its presence in cellular inclusions and Lewy bodies imply a role in Parkinson’s disease. Proc Natl Acad Sci U S A 2004;101:5500e5. [15] Engelender S, Kaminsky Z, Guo X, Sharp AH, Amaravi RK, Kleiderlein JJ, et al. Synphilin-1 associates with alpha-synuclein and promotes the formation of cytosolic inclusions. Nat Genet 1999;22: 110e4. [16] Da¨chsel JC, Lu¨cking CB, Deeg S, Schultz E, Lalowski M, Casademunt E, et al. Parkin interacts with the proteasome subunit alpha4. FEBS Lett 2005;579:3913e9. [17] West AB, Lockhart PJ, O’Farell C, Farrer MJ. Identification of a novel gene linked to parkin via a bi-directional promoter. J Mol Biol 2003; 326:11e9. [18] Junn E, Lee SS, Suhr UT, Mouradian MM. Parkin accumulation in aggresomes due to proteasome impairment. J Biol Chem 2002; 277:47,870e7. [19] Imai Y, Soda M, Takahashi R. Parkin suppresses unfolded protein stressinduced cell death through its E3 ubiquitin-protein ligase activity. J Biol Chem 2000;275:35,661e4. [20] West AB, Gonzalez-de-Chavez F, Wilkes K, O’Farrell C, Farrer MJ. Parkin is not regulated by the unfolded protein response in human neuroblastoma cells. Neurosci Lett 2003;341:139e42. [21] Wang HQ, Imai Y, Kataoka A, Takahashi R. Cell type-specific upregulation of Parkin in response to ER stress. Antioxid Redox Signal 2007;9:533e42. [22] Yamamoto A, Friedlein A, Imai Y, Takahashi R, Kahle PJ, Haass C. Parkin phosphorylation and modulation of its E3 ubiquitin ligase activity. J Biol Chem 2005;280:3390e9. [23] McNaught KS, Perl DP, Brownell AL, Olanow CW. Systemic exposure to proteasome inhibitors causes a progressive model of Parkinson’s disease. Ann Neurol 2004;56:149e62. [24] Schapira AH, Cleeter MW, Muddle JR, Workman JM, Cooper JM, King RH. Proteasomal inhibition causes loss of nigral tyrosine hydroxylase neurons. Ann Neurol 2006;60:253e5. [25] Kordower JH, Kanaan NM, Chu Y, Suresh Babu R, Stansell 3rd J, Terpstra BT, et al. Failure of proteasome inhibitor administration to provide a model of Parkinson’s disease in rats and monkeys. Ann Neurol 2006;60:264e8. [26] Manning-Bog AB, Reaney SH, Chou VP, Johnston LC, McCormack AL, Johnston J, et al. Lack of nigrostriatal pathology in a rat model of proteasome inhibition. Ann Neurol 2006;60:256e60. [27] Hampe C, Ardila-Osorio H, Fournier M, Brice A, Corti O. Biochemical analysis of Parkinson’s disease-causing variants of Parkin, an E3 ubiquitin-protein ligase with monoubiquitylation capacity. Hum Mol Genet 2006;15:2059e75. [28] Wang C, Ko HS, Thomas B, Tsang F, Chew KC, Tay SP, et al. Stressinduced alterations in parkin solubility promote parkin aggregation and compromise parkin’s protective function. Hum Mol Genet 2005;14: 3885e97. [29] Tan EK, Puong KY, Chan DK, Yew K, Fook-Chong S, Shen H, et al. Impaired transcriptional upregulation of Parkin promoter variant under oxidative stress and proteasomal inhibition: clinical association. Hum Genet 2005;118:484e8. [30] Biasini E, Fioriti L, Ceglia I, Invernizzi R, Bertoli A, Chiesa R, et al. Proteasome inhibition and aggregation in Parkinson’s disease: a comparative study in untransfected and transfected cells. J Neurochem 2004;88:545e53.