Neuron, Vol. 36, 1007–1019, December 19, 2002, Copyright 2002 by Cell Press
Parkin Protects against the Toxicity Associated
with Mutant ?-Synuclein: Proteasome Dysfunction
Selectively Affects Catecholaminergic Neurons
C-terminal hydrolase (UCHL1, OMIM 191342) has been
described (Farrer et al., 2000; Leroy et al., 1998), al-
though pathogenicity of this mutation has not been fully
established. However, polymorphic variability in UCHL1
has been associated with altered risk for development
of PD in case-control studies (Maraganore et al., 1999;
Wintermeyer et al., 2000).
Mutations in ?-synuclein have been reported in au-
tosomal dominant pedigrees (OMIM 601508, Kruger et
al., 1998; Polymeropoulos et al., 1997). Several pieces
teasome function may be related. Whether ?-synuclein
turnover is regulated by proteasome function is contro-
versial, with both positive (Bennett et al., 1999; Tofaris
et al., 2001) and negative (Ancolio et al., 2000; Paxinou
et al., 2001) results reported. Forced overexpression of
?-synuclein, especially mutant forms, sensitize PC12
(Stefanis et al., 2001; Tanaka et al., 2001), NT2, and SK-
N-MC (Lee et al., 2001b) neuroblastoma cell lines to
The mechanism by which this occurs is not clear, but
overexpression of mutant ?-synuclein produces an inhi-
bition of proteasome-associated proteolytic activities.
The A30P mutant ?-synuclein inhibits the postacidic
proteasome activity by 25% and the trypsin-like and
chymotrypsin-like activities by slightly smaller amounts,
with wild-type ?-synuclein having a much smaller effect
(Tanaka et al., 2001). The A53T mutant form of ?-synu-
clein also inhibits the chymotrypsin-like activity of the
proteasome (Stefanis et al., 2001). Finally, it has been
in human brain is ubiquitinated by parkin (Shimura et al.,
2001), raising the possibility that loss of parkin function
might result in ?-synuclein accumulation. ?-synuclein-
positive Lewy bodies have been noted in a case of Par-
kin-related PD (Farrer et al., 2001).
Overall, the above studies suggest that proteasome
inhibition might be a common link between the different
genetic triggers of PD. Furthermore, there is evidence
that proteasome function is impaired in sporadic PD
(McNaught and Jenner, 2001). However, the hypothesis
that proteasome dysfunction is an explanation for PD
remains conjecture. For example, as cell loss in PD is
not uniform, any attempt to link proteasome function
to disease should account for selective vulnerability of
specific subgroups of neurons. The selective vulnerabil-
ity of different neuronal types to cell death or formation
of the pathological hallmarks of the disease is complex
(reviewed in Braak and Braak, 2000), but it is clear that
functional loss of dopaminergic neurons in the substan-
tia nigra (SN) pars compacta is important. The move-
ment-related symptoms of PD patients are related to
dopaminergic cell loss, and loss of these cells not only
precedes symptom development, it is also progressive
throughout the course of the disease (Pakkenberg et
between overexpression of ?-synuclein and parkin with
toxicity associated with proteasome inhibition. We have
also used primary cell cultures to distinguish effects on
Leonard Petrucelli,1Casey O’Farrell,1
Paul J. Lockhart,1Melisa Baptista,3
Kathryn Kehoe,1Liselot Vink,1
Peter Choi,2Benjamin Wolozin,2
Matthew Farrer,1John Hardy,3
and Mark R. Cookson3,4
Mayo Clinic Jacksonville
Jacksonville, Florida 32224
2Department of Pharmacology
Loyola University Medical Center
Maywood, Illinois 60153
3Laboratory of Neurogenetics
National Institute on Aging
Bethesda, Maryland 20892
(PD) is that subsets of neurons are vulnerable to a
we show that overexpression of mutant ?-synuclein
increases sensitivity to proteasome inhibitors by de-
kin decreases sensitivity to proteasome inhibitors in
a manner dependent on parkin’s ubiquitin-protein E3
ligase activity, and antisense knockdown of parkin in-
creases sensitivity to proteasome inhibitors. Mutant
?-synuclein also causes selective toxicity to catechol-
aminergic neurons in primary midbrain cultures, an
effect that can be mimicked by the application of pro-
teasome inhibitors. Parkin is capable of rescuing the
tionin these cells.Therefore,parkinand ?-synucleinare
selective cell death in catecholaminergic neurons.
The identification of genes linked to familial forms of
Parkinson’s disease (PD) provides an important tool for
modeling the pathways leading to neurodegeneration
in this disorder. To date, eight linkages have been re-
ported, with three genes identified as causal, or proba-
bly causal, in different families. Two of these encode
proteins whose function is related to ubiquitin-depen-
dent protein degradation through the proteasome (for
review, see Hershko and Ciechanover, 1998). Parkin
(OMIM 600116) is an E2-dependent E3 protein-ubiquitin
ligase (Shimura et al., 2000; Zhang et al., 2000), and
mutations in this gene are generally associated with
recessive early onset parkinsonism (Kitada et al., 1998).
Parkin mutations reported to date appear to be loss-
of-function mutations reducing the ability of parkin to
regulate degradation of substrate removal (Shimura et
al., 2000; Zhang et al., 2000). One mutation in ubiquitin-
Figure 1. Effects of Proteasome Inhibitors on Cell Viability in M17 Cells Overexpressing ?-Synuclein
antibody 42. Untransfected cells (lane 2) or cells transfected with vector alone (lane 3) show moderate expression of ?-synuclein compared
to cells overexpressing wild-type (lane 4), A30P (lane 5), or A53T synuclein (lane 6). Human cerebral cortex extract was used as a positive
control (lane 1). A reprobe of the same blot using ?-actin is shown below to demonstrate similarity of loading across the lanes. Quantitation
of ?-synuclein expression is shown in (B) and is expressed as a ratio of the major synuclein band to ?-actin (n ? 4 serial passages of the
cells, error bars represent the SEM).
(C and D) Overexpression of ?-synuclein produces increased sensitivity to the proteasome inhibitor MG132 (C) or lactacystin (D). Cells were
exposed to either inhibitor for 24 hr, after which cell viability was estimated using the MTT assay (see Experimental Procedures). Cell lines
included untransfected cells (open squares) or cells transfected with empty vector (open circles), WT (open triangles), A30P (closed squares),
or A53T (closed circles) ?-synuclein. Results are expressed as a percentage of untreated cells for each clonal line (n ? 8, each curve is
representative of three or more experiments). Similar increased sensitivity to proteasome inhibition was seen in a second set of clonal lines.
Statistical significance was estimated using two-way ANOVA using cell lines and concentration of inhibitors as independent variables. **p ?
0.001 for differences between cell lines, both inhibitors having a significant effect on viability at p ? 0.001. Representative data from one of
TH-negative neurons. We show that parkin and mutant
associated with impaired proteasome function and that
parkin is capable of reducing toxicity associated with
?-synuclein overexpression. We also show that knock-
down of parkin increases the sensitivity of cells to pro-
tations in parkin would cause cell death by the same
mechanism as gain-of-function ?-synuclein mutations.
Furthermore, the effects of either mutant ?-synuclein or
proteasome inhibition are both selective for TH-positive
Overexpression of ?-Synuclein and Sensitivity
to Proteasome Inhibitors
We used human M17 neuroblastoma cells to explore
the relationship between mutant forms of ?-synuclein
and proteasome function. Cell lines stably overexpress-
ing wild-type or either of the two mutants expressed
about a 5-fold increased level of ?-synuclein compared
with untransfected cells (Figure 1A). Although the ex-
pression levels are high compared with untransfected
9-fold higher than untransfected M17 cells and approxi-
Proteasome Failure and Parkinson’s Disease
Figure 2. Measurements of Proteasome Ac-
tivity in Living Cells
?-synuclein decreases proteasome activity.
Stable cell lines (as in Figure 1) were tran-
siently transfected with the GFPuconstruct
(see text for description of the construct).
These included vector (lanes 1–3), cells ex-
pressing wild-type ?-synuclein (4–6), A30P
(7–9), or A53T (10–12) mutant ?-synuclein.
Protein extracts were blotted for GFP-CL1
peptide using anti-GFP (top panel) then re-
probed sequentially for ?-synuclein (middle
panel) and ?-actin (bottom panel). A semi-
quantitative analysis was performed by den-
sitometry, correcting GFP-CL1 levels for
?-actin. Each bar is the mean of the samples
shown in (A) plus a duplicate set (hence n ?
6), and error bars indicate the SEM. Both mu-
tant forms of synuclein increased the amount
of reporter construct, which was significant
between cell lines at p ? 0.001 by ANOVA.
(C–E) Additive effects of mutant ?-synuclein
expression and proteasome inhibition. (C)
Vector (lanes 1–6) or A30P mutant ?-synu-
clein cell lines (7–12) were transfected with
GFPuthen, after 48 hr, treated with 5 ?M lac-
for GFP, ?-synuclein, and ?-actin (top, mid-
dle, and bottom panels, respectively) and
semiquantitative analysis for GFP immunore-
activity (D) or ?-synuclein (E) normalized to
?-actin for the same gel, bars are the mean
of the samples shown in (C) plus a duplicate
set (hence n ? 6), and error bars indicate the
SEM. Similar results were obtained in dupli-
mately 30% higher than the highest expressing clone
of A30P cells. We also generated a second set of lines
with similar levels of expression (data not shown).
ity induced by proteasome inhibition (Figure 1C). Expo-
sure to 10 ?M MG132 caused cell viability to be de-
creased to 74% ? 1% of untreated for untransfected
cells and 76% ? 0.8% for vector cells (n ? 8). For wild-
tivity was noted at the highest tested concentration,
where cell viability was decreased to 68% ? 5%. Cells
transfected with either of the mutant forms of ?-synu-
clein were much more sensitive to MG132. The largest
effect was seen in cells overexpressing A30P where cell
viability was30% ? 0.4%at 10 ?MMG132. Overexpres-
sion of A53T ?-synuclein decreased cell viability to
50% ? 1% at the same concentration. Similar results
were seen with a second set of clones. For example,
using two independent A30P-expressing clonal lines,
cell viability after exposure to 10 ?M MG132 was de-
Figure 3. Effects of Proteasome Inhibitors on Cell Viability in M17 Cells Overexpressing Parkin
(A) Parkin protein was measured by Western blotting using a C-terminal antibody to human parkin in clonal lines transfected with vector (lane
1) or expression vector for human parkin (lane 2). Full-length parkin runs at ?50 kDa (arrow) although a C-terminal fragment (?42 kDa) is also
noted in the overexpressing cell line (asterisk). Equal loading was demonstrated by reprobing the same blot with ?-actin.
(B and C) Overexpression of parkin protects cells against the toxic effects of the proteasome inhibitor MG132 (B) or lactacystin (C). Cells
stably transfected with parkin (open symbols) or vector (closed symbols) were exposed to inhibitors for 24 hr and cell viability assessed as
in Figure 4. Statistical significance was estimated using two-way ANOVA using cell lines and concentration of inhibitors as independent
variables. *p ? 0.05; **p ? 0.001 for differences between cell lines, both inhibitors having a significant effect on viability at p ? 0.001. Similar
protective effects of parkin overexpression were seen in two independent experiments (each n ? 8).
(D) In an independent set of experiments, cells overexpressing mutant Parkin (R42P) were not protected from exposure to 10 ?M (filled bars)
or 25 ?M (striped bars) lactacystin. (n ? 8, representative of two experiments.)
(E) Cells were transfected with vector, the E3 ligase E6-AP, Parkin alone, or Parkin in the presence of a dominant-negative inhibitor of the E2
enzyme UbCH7 and exposed to 10 ?M lactacystin. Cell death was quantified as above (n ? 8) and expressed as a percentage of MTT
conversion for transfected cells that had not been exposed to lactacystin. The differences between vector and parkin transfected cells were
(F) Antisense knockdown of Parkin increases steady-state levels of heterologous substrates. M17 cells stably transfected with wild-type Parkin
(lanes 3 and 4) or an antisense parkin construct (lanes 5 and 6) were transiently transfected with GFPureporter as above (upper panel) or
transfected and then treated with MG132 (middle panel). Vector-only cells were included as controls (lanes 1 and 2, duplicate clonal lines).
Parallel samples were blotted for parkin (lower panel) to demonstrate the level of overexpression and the effect of antisense knockdown.
GFP-CL1 levels were unaffected by expression of parkin (similar effects were seen in two independent experiments), although antisense
parkin cell lines did show an accumulation of GFP-CL1.
(G) MTT assays in the same cell lines show that cells transfected with antisense parkin (open circles) are more sensitive to MG132 toxicity
compared to vector-only lines (closed circles) or cells transfected with WT parkin (closed triangles). Results are expressed as a percentage
of untreated cells for each clonal line (n ? 8, each curve is representative of two or more experiments).
Proteasome Failure and Parkinson’s Disease
Figure 4. Characterization
and ?-Synuclein Expression in Primary Post-
natal Midbrain Neurons
Primary neuronal cultures from postnatal
mouse midbrain were stained with polyclonal
antibodies totyrosine hydroxylase(TH, green
in [A]–[D]) and monoclonal antibodies to
?-synuclein ([A] and [B] shows higher mag-
nification) or synaptophysin ([C], higher mag-
nification shown in [D]). The monoclonal
antibody clone 42 (A and B) recognizes en-
dogenous (mouse) ?-synuclein. Staining for
?-synuclein was performed using a poly-
[F]). All three proteins were expressed in both
TH-positive (arrows) and TH-negative cells
with similar patterns. Higher magnification of
TH-positive cells (B, D, and F) demonstrated
localization at the cell membrane, reminis-
cent of synaptic structures (arrowheads). All
images were captured using a confocal mi-
croscope, and merged images of both chan-
nels are shown, yellow indicating overlap be-
represent 50 ?m.
creased to 27% ? 1% in one clone and 30% ? 0.4%
in the second. Consistent results were also obtained for
not shown). The differences between different clonal
cell lines were significant (p ? 0.0001) using two-way
ANOVA, as was the effect of MG132 across all cell lines
(p ? 0.0001). We also exposed cells to the structurally
unrelated inhibitor, lactacystin (Figure 1D). Lactacystin
required concentrations of up to 25 ?M to produce loss
of cell viability to 70% (Figure 1D). Increased toxicity
was noted in cell lines expressing either A30P or A53T
?-synuclein. A small effect of wild-type ?-synuclein was
also seen. The differences between different clonal cell
lines were significant (p ? 0.004), as was the effect of
iments using a second set of clonal cell lines. Results
presented here using MTT conversion are similar to pre-
vious reports usingTrypan blue dye exclusionas a mea-
sure of cell death (Tanaka et al., 2001). To address the
possibility that such effects might be due to inhibition
of other proteases, we exposed cells to the cell-soluble
calpain inhibitor E64d. This compound was without ef-
fect on cell viability up to 100 ?M, which approached
the limit of solubility (data not shown).
We next examined the mechanism by which mutant
?-synuclein increases cellular sensitivity to proteasome
inhibitors. We measured net proteasomal activity in liv-
ing cells using the GFPureporter construct (Bence et
al., 2001). When transfected into cells, the CL1 peptide
(Gilon et al., 1998) sequence fused to GFP leads to rapid
destruction of the protein, and in control cell lines we
found only small amounts of GFP-CL1 peptide (Figure
2). The amount of reporter construct accumulation in
cells transfected with wild-type ?-synuclein was similar
to that in vector-transfected cells but much higher in
cells expressing A30P or A53T mutants (Figures 2A and
2B). The differences in amounts of GFP-CL1 between
the cell lines were statistically significant (p ? 0.01 by
with 5 ?M lactacystin for 5 hr, which increased steady-
state levels of the GFP-CL1 fusion protein. This effect
was enhanced in the presence of A30P mutant ?-synu-
clein (Figure 2D). Therefore, mutant ?-synuclein inhibits
proteasome function in a manner that is additive to the
effect of pharmacological inhibition of the proteasome.
clein levels in experiments where an increased GFP-
CL1 reporter protein demonstrated unequivocally that
Figure 5. Overexpression of ?-Synuclein in Primary Neurons
(A and B) Midbrain or hippocampal neurons were transduced with human ?-synuclein (WT or A53T) and protein extracts (10 ?g/lane) blotted
with (A) either monoclonal 42 (recognizes rodent and human ?-synuclein) or (B) LB509 (human ?-synuclein only). Similar levels of expression
were obtained with each of the viral constructs. Untransfected cells (UT) and cells transduced with a LacZ construct were used as controls.
Proteasome Failure and Parkinson’s Disease
there was proteasome inhibition (Figure 2E). Given the
possible inhibitory effect of ?-synuclein on ?-synuclein
toxicity (Hashimoto et al., 2001), we also examined
?-synuclein expression in the same cell extracts. We
were unableto detect ?-synucleinprotein ineither basal
conditions or after proteasome inhibition, although the
antibodies used did detect expression of this protein in
human brain (data not shown) and in primary mouse
cultures (see below).
cell was reduced, as occurs in ARJP. To model this
effect, we generated cell lines with an antisense con-
struct that had lower steady-state levels of parkin pro-
tein. This antisense knockdown of parkin does increase
GFPulevels within the cell, and MG132 has an additive
ity increases the level of heterologous substrates, pre-
sumably due to increased levels of parkin substrates
competing for ubiquitination and/or proteasome-medi-
ated degradation. Proteasome inhibitors are more toxic
in cell lines transfected with an antisense parkin con-
struct (Figure 3G).
Manipulation of Parkin Expression Levels
and Sensitivity to Proteasome Inhibitors
The above data demonstrate that the overexpression of
mutant ?-synuclein results in an increased sensitivity of
cells to loss of viability induced by proteasome inhibi-
tion. We examined what effect increased parkin activity
might have on the same parameters. Stable cell lines
transfected with the parkin cDNA had substantially in-
creased parkin protein expression (Figure 3A). Over-
expression of parkin partially rescued cells from the
toxic effects of MG132 (Figure 3B) or lactacystin (Figure
3C), with cell viability being 10%–20% higher in the par-
kin cell lines compared to controls at all doses of either
for MG132, and p ? 0.004 for lactacystin).
Further experiments confirmed that the observed res-
cue of parkin is dependent on its E3 ligase activity. A
mutant form of parkin associated with loss of E3 ligase
activity (R42P) was not able to rescue cells in an inde-
ences between wild-type parkin and R42P parkin were
statistically significant (p ? 0.001 for differences be-
tween wild-type parkin and either vector alone or R42P).
also ameliorated the protective effect of parkin (Figure
3E). We also examined the effects of overexpression of
a second E3 ligase, E6AP, which was not protective
in these cells (Figure 3E). Experiments using the GFPu
construct demonstrated that there was no alteration in
the amount of GFP-CL1 in the cells when transfected
treatment (Figure 3F), although MG132 did increase
GFP-CL1 levels in this experiment as in the experiments
shown in Figure 2 . Parkin does not, therefore, act as
an E3 ligase for this artificial proteasome reporter (see
Discussion). However, the above experiments do not
address what would occur if parkin activity within the
Mutant ?-Synuclein Triggers Selective Cell Death
in Primary Neuronal Culture
of ?-synuclein overexpression and proteasome inhibi-
tion on different neuronal groups. These cultures are
predominantly from the nigral area and have a higher
proportion of TH-positive catecholaminergic neurons
than commonly used embryonic mesencephalic neuron
preparations, with TH-positive cells representing about
20%–30% of the total neuronal population. The TH anti-
bodies we used in these experiments recognized an
appropriately sized band in extracts from whole mouse
brain or from midbrain cultures but not in hippocampal
cultures (data not shown). We characterized the expres-
sion of (mouse) ?- and ?-synuclein in these cells (Figure
4). Both TH-positive and TH-negative neurons expressed
?-synuclein (Figures 4A and 4B) at the cell surface in a
punctate pattern reminiscent of synaptic proteins such
as synaptophysin (Figures 4C and 4D), similar to hippo-
campal neurons (Murphy et al., 2000). In addition, we
also noted ?-synuclein immunoreactivity in the cyto-
plasm. This was not limited to either TH-positive or TH-
negative cells but was seen in many midbrain neurons.
Expression of ?-synuclein (Figures 4E and 4F) was also
synaptic and seen in both cell populations at similar
We next transduced primary midbrain cultures with
wild-type or mutant ?-synuclein and monitored expres-
sion by Western blotting or immunocytochemistry. Two
monoclonal antibodies, clone 42 and LB509, were used
to distinguish overexpressed human ?-synuclein from
endogenous mouse ?-synuclein, respectively. Trans-
duction with wild-type or mutant ?-synuclein produced
Left-hand panels show controls using brain lysates from mouse (mo) or human (hu) cerebral cortex. Blotting using monoclonal antibody 42
yielded a major band at 19 kDa (arrow) plus a smaller degradation product with an apparent molecular weight of 16–17 kDa (arrowhead): this
smaller product was not seen with LB509.
(C–K) Expression of human ?-synuclein in catecholaminergic neurons. Primary midbrain catecholaminergic neurons (TH-positive, green) were
transduced at multiplicities of infection (MOIs) of 5–10 with HSV1 expressing lacZ (C, D, and E) as a negative control or ?-synuclein (WT,
[F–H]; A53T, [I–K]). Transduction was demonstrated using a human specific monoclonal antibody LB509 (red), and more than 95% of cells
were LB509 positive. Merged images are shown on the right of each set of photomicrographs. Scale bar in (K) represents 50 ?m and applies
to all panels. Representative data is shown from one of four experiments.
(L–M) Overexpression of mutant ?-synuclein in primary midbrain neurons is associated with selective cell death of TH-positive cells. Cells
were transduced with either LacZ (negative control), WT, or mutant (A53T) ?-synuclein and cell numbers estimated by counting using TH and
cells (N) were not affected. Each open circle is the average cell counts from one experiment with six fields counted in each of three cultures
(hence n ? 18): closed circles represent the mean from each of three experiments, with error bars representing the SEM between experiments.
Statistical significance was assessed using one-way ANOVA with Student-Neuman-Kuells post-hoc tests between each group. ns, not
similar levels of overexpression (Figures 5A and 5B). To
confirm that TH-positive neurons in midbrain cultures
were transduced, cells were costained for TH and hu-
man ?-synuclein using LB509 (Figures 5C–5K). We esti-
(MOI) of 10, more than 95% of TH-positive neurons ex-
pressed human ?-synuclein. TH-negative neurons were
sion in this acute model does not result in the formation
of microscopically visible ?-synuclein aggregates (Fig-
Cell counts were performed to assess whether the
overexpressed ?-synuclein induced any toxicity in TH-
positive neurons. In each experiment, we counted six
fields in each of three cultures. We also repeated the
whole series three times with independent purifications
of viral particles, and the data presented (Figure 5L)
shows the interexperiment variation. We were able to
demonstrate a clear toxic effect of A53T ?-synuclein in
TH-positive cells. We were not able to demonstrate an
effect with wild-type ?-synuclein under these condi-
tions. TH-negative midbrain neurons (Figure 5M) or hip-
pocampal neurons (Figure 5N) were unaffected by the
presence of A53T ?-synuclein.
Parkin Rescues the Toxicity of Mutant
?-Synuclein in Primary Neurons
We reasoned that as parkin protected cells against pro-
teasome inhibition and mutant ?-synuclein overexpres-
sion inhibited the proteasome then parkin might be pro-
tective against toxicity associated with overexpression
of mutant ?-synuclein. We repeated the experiments
using A53T ?-synuclein and coexpressed either lacZ as
a control or parkin (Figures 7A and 7B). Cotransduction
of parkin restored the number of counted neurons back
one-way ANOVA with post-hoc tests as above, the loss
of TH-positive neurons was significantly different (p ?
0.05)from controlsandfromcultures treatedwithparkin
and ?-synuclein (p ? 0.05), but the difference between
cotransduced cultures and control cells was not signifi-
cant. Similarly to experiments in cell lines, Parkin was
able to rescue to selective toxicity of MG132 to primary
cells (Figures 7C and 7D).
In the current study, we have examined manipulation of
two genes that show association with familial PD on
cellular sensitivity to proteasome inhibition and have
examined aspects of neuronal selectivity. There was an
et al., 2001b; Tanaka et al., 2001). Mutant ?-synuclein
also sensitizes cells to other insults (Junn and Moura-
dian, 2002; Ko et al., 2000; Lee et al., 2001b; Ostrerova-
Golts et al., 2000; Zhou et al., 2000), but we and others
(Stefanis et al., 2001; Tanaka et al., 2001) have demon-
the net proteasomal activity in living cells. Therefore,
proteasome inhibition is likely to make a significant con-
tribution to cell death induced by mutant ?-synuclein.
The inhibition of proteasome function by mutant ?-synu-
clein may be a direct inhibition of proteasome activity,
of the proteasome (Ghee et al., 2000) or the presence
as mutant ?-synuclein, might inhibit the ubiquitin-pro-
teasome pathway indirectly (Bence et al., 2001).
Overexpression of parkin protected against toxicity
associated with reduced proteasome function. The lack
of effect of the recessive R42P parkin mutation, which
lacks ubiquitination activity (Shimura et al., 2001), dem-
onstrates that mutant forms of parkin are unable to pro-
tect dopaminergic neurons against proteasome failure.
Therefore, both genes implicated in familial PD alter
the ability of neurons to tolerate reduced proteasome
activity. We have shown that the E3 ligase activity of
E2 mutant could ameliorate this effect. Experiments us-
net proteasome activity, consistent with the role of this
E3 ligase in controlling entry of target proteins into the
proteasome via ubiquitination, and does not alter the
steady-state levels of ?-synuclein. Parkin is also protec-
tive in some other models of cell death, such as ER
stress, but is not protective against all insults, including
staurosporine (Imai et al., 2000). We suggest that parkin
TH-Positive Neurons Are Selectively Vulnerable
to Proteasome Inhibition
We next examined whether proteasome inhibition was
sufficient to produce selective neuronal cell loss in the
same manner as ?-synuclein overexpression. The num-
bers of TH-positive neurons in MG132 were decreased
compared to controls, with remaining cells often show-
ing shrinkage of cell bodies and retraction of neuritic
processes (Figures 6A and 6B). Previous reports of cell
death induced by proteasome inhibitors in the presence
of mutant ?-synuclein have given contradictory results
(Tanaka et al., 2001) or autophagy (Stefanis et al., 2001).
In our cultures exposed to proteasome inhibitors, coun-
in damaged cells the nuclei remained intact, unlike apo-
ptosis. The numbers of TH-positive cells were signifi-
cantly reduced at higher concentrations of lactacystin
or MG132 (Figures 6C and 6E), while the number of TH-
negative neurons was unaffected (Figures 6D and 6F).
Using one-way ANOVA with Student-Newman-Kuells
bition on numbers of TH-positive cells remaining, MG132
had a significant (p ? 0.05) effect at both 1 and 5 ?M,
whereas the effect of lactacystin was significant only at
10 ?M. We also examined the effects of proteasome
(Figures 6Gand 6H). Thetwo treatments hadan additive
effect (p ? 0.01 by ANOVA for all groups), although the
loss of TH-positive cells was not complete after 24 hr,
the time point used for this study. We examined cells
after staining with Hoechst 33342 (as in Figures 6A and
6B) and again did not find evidence for apoptosis (data
Proteasome Failure and Parkinson’s Disease
Figure 6. Catecholaminergic
Preferentially Susceptible to Proteasome In-
hibition in Primary Culture
(A and B) Primary midbrain cultures were left
untreated (A) or exposed to 10 ?M MG132
for 24 hr (B) and stained for MAP2 (red), TH
(green), and Hoechst 33342 (blue) to demon-
strate nuclear morphology. Exposure to the
proteasome inhibitor caused TH-positive
cells to become shrunken and retract pro-
cesses,but nuclei remained intact (inset in[B]).
andD)or lactacystin(EandF). Bothinhibitors
substantially affected the TH-positive neu-
ronal population (C,E), whereas TH-negative/
MAP2-positive neurons (D and F) were unaf-
fected. Data are shown as cell counts from
six randomly selected microscope fields in
each of three replicate cultures (hence n ?
18) and is representative of two to four inde-
pendent experiments with different batches
of primary cells. Statistical significance was
assessed using one-way ANOVA with Stu-
dent-Neuman-Kuells post-hoc tests between
each group (*p ? 0.05).
(G and H) Additive effect of ?-synuclein over-
expression and proteasome inhibition. Pri-
mary cells as above were transduced with
mutant ?-synuclein as in Figure 5 for 24 hr
then exposed to MG132 for a further 24 hr.
Cell counts revealed a loss of TH-positive
tive effect of both treatments together (*p ?
0.05 by one-way ANOVA with Student-New-
man-Kuells post-hoc test).
protects against the accumulation of its specific protein
teins may also be promoted through ER stress due to
The nature of the targets for parkin’s E3 ligase activity
is still under investigation, although several candidates
have been reported (Chung et al., 2001; Imai et al., 2001;
Shimura et al., 2001; Zhang et al., 2000). We also show
that knockdown of parkin using a stable antisense con-
tion. This is, in many ways, a better model for loss-of-
function mutations than overexpression of the wild-type
protein. In ARJP, for example, homozygous large-scale
deletions are predicted to reduce enzyme activity to
is closer to this disease model than overexpression. In
this experimental setting, there is a clear accumulation
of heterologous substrates, as evidenced by accumula-
tion of the GFP-CL1 reporter peptide, suggesting that
loss-of-function alleles would decrease the ability of
nigral neurons to regulate levels of proteasome sub-
strates. It has been shown recently that proteasome
inhibition in vivo damages nigral neurons (McNaught et
al., 2002). Our results predict that loss of parkin function
would have the same effect.
Both overexpression of mutant ?-synuclein and pro-
Figure 7. Parkin Rescues Toxicity Associ-
ated with Mutant ?-Synuclein or Proteasome
Inhibition in Primary Midbrain Cells
Primary midbrain neurons were transduced
with both mutant ?-synuclein and wild-type
human parkin. Where either virus was absent
(?), LacZ was substituted to keep the total
number of viral particles similar. There was
a significant reduction in the number of TH-
positive neurons (A), which was ameliorated
ative cells was noted (B). Data are shown as
(hence n ? 18), and statistical significance
was assessed using one-way ANOVA with
Student-Neuman-Kuells post-hoc tests be-
tween each group (*p ? 0.05; ns, not signifi-
cant). Similar protective effects were noted
using exposure to MG132 to induce toxicity
(C and D) where selective toxicity to TH-posi-
tive cells was noted (C). The numbers of cell
counts and statistical tests are as in (A)
neurons. Therefore, proteasome inhibition is sufficient
to mimic the effects of mutant ?-synuclein. It is not the
case that all catecholaminergic neurons are severely
affected in PD, while all non-dopaminergic neurons are
spared (Braak and Braak, 2000). Nonetheless, we have
shown that the SN contains a population of TH-positive
cells that are particularly sensitive to both proteasome
inhibition and ?-synuclein overexpression. The higher
sensitivity of catecholaminergic neurons to damage in-
duced by overexpression of mutant ?-synuclein may
be due to the ability of catecholamines to promote the
formation of protofibrillar forms of ?-synuclein (Conway
et al., 2001). It has been suggested that ?-synuclein
protofibrils are a major toxic species of this protein,
and mutations in ?-synuclein also promote protofibril
formation (Conway et al., 2000). Recently, it has been
shown that, in cultured human dopaminergic neurons,
dopamine is required for the toxic effects of mutant
?-synuclein, supporting this hypothesis (Xu et al., 2002).
We have not seen a protective effect of the same tyro-
sinehydroxylase inhibitor(?-methyl-para-tyrosine; AMPT)
in the stable cell lines we used in this study, but this is
confounded by a small toxic effect of AMPT alone in
our hands (M.R.C. and M.B., unpublished data). In our
experiments, mutant ?-synuclein is more toxic to TH-
of ?-synuclein in embryonic mesencephalic cultures
(Zhou et al., 2000) or human dopaminergic cells (Xu
et al., 2002) produces a similar effect. In some mouse
transgenic models, there is pathology associated with
a substantial overexpression of wild-type ?-synuclein,
and there are reports of dopaminergic cell loss in some
(Masliah et al., 2000) but not all (Matsuoka et al., 2001;
Rathke-Hartlieb et al., 2001; VanDerPutten et al., 2000)
models. In a Drosophila model, loss of dopaminergic
neurons is also seen (Feany and Bender, 2000) with
mutant ?-synuclein having a more substantial effect
than wild-type. Our data are similar to viral-mediated
gene transfer experiments where mutant ?-synuclein
produces nigral cell loss (Kirik et al., 2002).
We did not find microscopically visible protein aggre-
gates, nor did we see the formation of higher molecular
weight species of ?-synuclein as seen in some models
(Lee et al., 2001a). We believe that this is compatible
with the idea that protein aggregation to the extent of
formation of insoluble fibrillar species is not a required
step for ?-synuclein toxicity. Soluble protein complexes
appear to mediate the toxic effects of mutant ?-synu-
clein in human dopaminergic cells (Xu et al., 2002). The
formation of protein aggregates is clearly relevant to the
human disease, as the formation of insoluble protein
deposits in the form of Lewy bodies occurs in surviving
neurons. Previous results using iron/dopamine-medi-
ated toxicityhave shownthat althoughformation ofpro-
tein aggregates and toxicity can be seen under similar
conditions, these are dissociatable phenomena (Ostrer-
ceivably affectthe fibrillization propertiesof ?-synuclein
throughthe phenomenonof“molecular crowding”(Ellis,
2001). Two recent studies have demonstrated that in-
creasing the concentration of macromolecules in the
immediate surroundings of ?-synuclein increases its
propensity to form protofibrillar and fibrillar species
(Shtilerman et al., 2002; Uversky et al., 2002). By inhib-
crowding effect. Therefore, if formation of protofibrillar
forms of ?-synuclein is important for the toxicity of the
mutant forms, proteasome inhibitors are likely to accel-
erate this process, without having an effect on net
?-synuclein protein concentrations. Protection by par-
kin of TH-positive neurons exposed to either mutant
?-synuclein or proteasome inhibition suggests that
In the results reported here, antisense-mediated knock-
down of parkin also increased levels of heterologous
substrates, again potentially inducing accumulation of
several toxic proteins and inducing a crowding effect.
The protective effect of parkin on loss of TH-positive
neurons mediated by mutant ?-synuclein demonstrates
that these two proteins have interrelated effects. The
Proteasome Failure and Parkinson’s Disease
blocking (5% v/v goat serum plus 5% FBS in DPBS, 30 min), primary
antibodies were appliedovernight at 4?C. Thesewere a combination
MAP2 (clone AP-20, Sigma) both at a dilution of 1:200. Secondary
antibodies were goat anti-rabbit conjugated to Alexafluor488 (Mo-
lecular Probes) and goat anti-mouse conjugated to AlexaFluor 568.
a 20? objective) were counted in each of three sister cultures, for
a total of 18 fields per condition, counting all neuronal cells within
each field. Each experiment was repeated two to four times with
clonal anti-human ?-synuclein LB509 (Zymed) or monoclonal anti-
rodent ?-synuclein (clone 42, listed above) were used in the above
protocol at a dilution of 1:200. A polyclonal antibody to ?-synuclein
(Chemicon) was used at 1:500 in conjunction with monoclonal anti-
body to TH (also Chemicon, 1:200).
above data suggest that an increased sensitivity of cells
to the toxic effects of proteasome inhibition link ?-synu-
clein and parkin as well as providing an explanation
for the selective loss of a subgroup of dopaminergic
neurons in PD. Whether proteasome inhibition will pro-
vide a full explanation for neuronal damage in PD is not
clear. There are several genes linked to familial PD that
remain to be identified, and it will be critical to evaluate
whether manipulations of these products also produces
increased sensitivity to proteasome inhibition.
Human Neuroblastoma Cell Lines
The production of stable cell lines overexpressing wild-type or mu-
tant ?-synuclein from parental BE (2)-M17 human dopaminergic
neuroblastoma cells has been detailed elsewhere (Ostrerova-Golts
et al., 2000). Full-length parkin cDNA was cloned into the same
vector and transfections performed as described previously (Ostre-
the first 100 bp of coding sequence for parkin into pCDNA3.1 in the
reverse orientation relative to the vector promoter. Stable clones
were prepared as above and screened for reduced parkin expres-
sion. For the present study, clonal lines for both ?-synuclein and
parkin were made by limiting dilution and were maintained on 500
?g mL?1G418. Stable clonal lines were screened for ?-synuclein
expression by Western blotting using monoclonal antibody to
?-synuclein (Clone 42, Transduction Labs). Cell lysates (10 ?g total
protein per lane) were separated on 16% SDS-PAGE gels (Novex)
extract of adult human cerebral cortex was used as a positive con-
trol. After probing with primary antibody (1:1000), blots were devel-
oped with peroxidase-labeled secondary antibodies (Jackson Im-
munochemicals) using enhanced chemiluminescence substrates
(Amersham). Blots were reprobed with ?-actin (Sigma, clone AC15,
1:5000) to verify equal loading. Quantitation of ?-synuclein expres-
sion was performed by capturing enhanced chemiluminescence us-
ing a CCD camera-based system (AlphaImager, Alpha Innotech
Corp). Parkin expression was also monitored by Western blotting
using a rabbit polyclonal antibody to the C terminus of parkin (Cell
Signaling Technology, 1:2000 dilution). Cell viability was assessed
usingthe MTTassay,as describedpreviously(Cooksonet al.,1998).
For each experiment, 8 wells were used per concentration of either
compound, and each experiment was repeated three times with
?-synuclein or parkin cDNAs were cloned into pHSVPrpUC and
packaged into recombinant viral particles using 5dl1.2 helper virus
and the 2-2 packaging cell line as described (Neve et al., 1997) and
purified using sucrose gradients. A control virus expressing LacZ
(from pHSVlacZ, Coopersmith and Neve, 1999) was prepared at the
same time. Recombinant viruses were titred on human neuro-
at a multiplicity of infection (MOI) of 10. In a small series of experi-
ments, we extracted cultures grown in 6-well plates and blotted for
?-synuclein as above.
Differences in the responses of cell lines to proteasome inhibitors
were evaluated using two-way analysis of variance (ANOVA) with
For primary cell counts, one-way ANOVA with Student-Neuman-
Kuells post-hoc test was used to assess differences between treat-
ments with proteasome inhibitors or with transduction with different
viral constructs. For each of these experiments, six fields were
counted in each of three independent cultures, hence n ? 18. In
experiments comparing the toxicity associated with overexpression
of ?-synuclein, we repeated this whole set of experiments three
The authors would like to thank Dr. David Sulzer, Columbia Univer-
sity, for sharing the methods for culturing postnatal mouse midbrain
neurons. We would also like to acknowledge the generous gifts
of reagents for HSV-1 packaging from Dr. Rachael Neve, Harvard
Medical School, Boston, and Neil Bence and Dr. Ron Kopito of
Stanford University forthe gift of the GFPuplasmid.Dr. Ted Dawson,
Johns HopkinsMedical Institute,kindly providedthe dominant-neg-
ative UbCH7 construct. This work was supported by NINDS grants
P01-NS40256 and RO1-NS41816-01.
Measurement of Proteasome Function in Living Cells
for 6 hr or left untreated as controls. Protein extracts and Western
blotting wereperformed asabove to measuresteady-state amounts
of GFP-CL1 fusion protein in the cells using a monoclonal antibody
to GFP (Clontech). Blots were reprobed with monoclonal anti-
?-synuclein and subsequently with monoclonal anti ?-actin as de-
Received: January 16, 2002
Revised: November 21, 2002
Primary Cell Cultures
Primary cell cultures were prepared from postnatal mouse midbrain
using methods described by Burke et al. (Burke et al., 1998). Briefly,
midbrains containing SN and ventral tegmental area were dissected
from 2-day-old postnatal mouse pups, using anatomical landmarks
as described (Burke et al., 1998; Mena et al., 1997). Neurons from
these areas were dissociated with papain and plated on top of
preestablished cortical glia cell monolayers at a density of 80,000
cells per well in growth medium which had been preconditioned by
adding to glial feeder layers 24 hr prior to plating neurons. Neurons
from 3- to 4-week cultures were exposed for 24 hr to MG132 or
lactacystin (Calbiochem) and toxicity assessed by staining for tyro-
sinehydroxylase (TH)andmicrotubule-associatedprotein 2(MAP2).
Cells were fixed and permeabilized with 4% paraformaldehyde in
saponin. After quenching (0.1 M glycine in DPBS, 20 min) and
Ancolio, K., Alves da Costa, C., Ueda, K., and Checler, F. (2000).
alpha-synuclein do not undergo proteasomal degradation in
HEK293 and neuronal cells. Neurosci. Lett. 285, 79–82.
Bence, N.F., Sampat, R.M., and Kopito, R.R. (2001). Impairment of
the ubiquitin-proteasome system by protein aggregation. Science
Bennett, M.C., Bishop, J.F., Leng, Y., Chock, P.B., Chase, T.N., and
Mouradian, M.M. (1999). Degradation of alpha-synuclein by protea-
some. J. Biol. Chem. 274, 33855–33858.
Braak, H., and Braak, E. (2000). Pathoanatomy of Parkinson’s dis-
ease. J. Neurol. Suppl. 247, 3–10.
neurotrophic growth factor inhibits apoptotic death of postnatal
substantia nigra dopamine neurons in primary culture. J. Neuro-
chem. 71, 517–525.
Chung, K.K., Zhang, Y., Lim, K.L., Tanaka, Y., Huang, H., Gao, J.,
Ross, C.A., Dawson, V.L., and Dawson, T.M. (2001). Parkin ubiquiti-
nates the ?-synuclein-interacting protein, synphilin-1: implications
for Lewy-body formation in Parkinson disease. Nat. Med. 7, 1144–
Conway, K.A., Lee, S.J., Rochet, J.C., Ding, T.T., Williamson, R.E.,
and Lansbury, P.T. (2000). Acceleration of oligomerization, not fibril-
lization, is a shared property of both alpha-synuclein mutations
linked to early-onset Parkinson’s disease: implications for patho-
genesis and therapy. Proc. Natl. Acad. Sci. USA 97, 571–576.
Conway, K.A., Rochet, J.-C., Bieganski, R.M., and Lansbury, P.T.
(2001). Kinetic stabilization of the ?-synuclein protofibril by a dopa-
mine-?-synuclein adduct. Science 294, 1346–1349.
Cookson, M.R., Ince, P.G., and Shaw, P.J. (1998). Peroxynitrite and
hydrogen peroxide induced cell death in the NSC34 neuroblastoma
x spinal cord cell line: Role of poly(ADP-ribose) polymerase. J. Neu-
rochem. 70, 501–508.
Coopersmith, R., and Neve, R.L. (1999). Expression of multiple pro-
teins within single primary cortical neurons using a replication defi-
cient HSV vector. Biotechniques 27, 1156–1160.
Ellis, R.J. (2001). Macromolecular crowding: an important but ne-
glected aspect of the intracellular environment. Curr. Opin. Struct.
Biol. 11, 114–119.
Farrer, M., Destee, T., Becquet, E., Wavrant-De Vrieze, F., Mouroux,
V., Richard, F., Defebvre, L., Lincoln, S., Hardy, J., Amouyel, P., and
Chartier-Harlin, M. (2000). Linkage exclusion in French families with
probable Parkinson’s disease. Mov. Disord. 15, 1075–1083.
Farrer, M., Chan, P., Chen, R., Tan, L., Lincoln, S., Hernandez, D.,
Forno, L., Gwinn-Hardy, K., Petrucelli, L., Hussey, J., et al. (2001).
Lewy bodies and parkinsonism in families with parkin mutations.
Ann. Neurol. 50, 293–300.
son’s disease. Nature 404, 394–398.
Ghee, M., Fournier, A., and Mallet, J. (2000). Rat alpha-synuclein
interacts with tat binding protein 1, a component of the 26S protea-
somal complex. J. Neurochem. 75, 2221–2224.
Gilon, T., Chomsky, O., and Kulka, R.G. (1998). Degradation signals
for ubiquitin system proteolysis in Saccharomyces cerevisiae.
EMBO J. 17, 2759–2766.
E. (2001). beta-Synuclein inhibits alpha-synuclein aggregation: a
possible role as an anti-parkinsonian factor. Neuron 32, 213–223.
Rev. Biochem. 67, 425–479.
Imai, Y., Soda, M., and Takahashi, R. (2000). Parkin supresses un-
folded protein stress-induced cell death through its E3 ubiquitin-
protein ligase activity. J. Biol. Chem. 275, 35661–35664.
Imai, Y., Soda, M., Inoue, H., Hattori, N., Mizuno, Y., and Takahashi,
R. (2001). An unfolded putative transmembrane polypeptide, which
can lead to endoplasmic reticulum stress, is a substrate of Parkin.
Cell 105, 891–902.
Junn, E., and Mouradian, M.M. (2002). Human a-synuclein over-
expression increases intracellular reactive oxygen species levels
and susceptibility to dopamine. Neurosci. Lett. 320, 146–150.
Kirik, D., Rosenblad, C., Burger, C., Lundberg, C., Johansen, T.E.,
Muzyczka, N., Mandel, R.J., and Bjorklund, A. (2002). Parkinson-
like neurodegeneration induced by the targeted overexpression of
?-synuclein in the nigrostriatal system. J. Neurosci. 22, 2780–2791.
Kitada, T., Asakawa, S., Hattori, N., Matsumine, H., Yamamura, Y.,
Minoshima, S., Yokochi, M., Mizuno, Y., and Shimizu, N. (1998).
Mutations in the parkin gene cause autosomal recessive juvenile
parkinsonism. Nature 392, 605–608.
Ko, L., Mehta, N.D., Farrer, M., Easson, C., Hussey, J., Yen, S.,
Hardy, J., and Yen, S.H. (2000). Sensitization of neuronal cells to
oxidative stress with mutated human alpha-synuclein. J. Neuro-
chem. 75, 2546–2554.
Kruger, R., Kuhn, W., Muller, T., Woitalla, D., Graeber, M., Kosel, S.,
Przuntek, H., Epplen, J.T., Schols, L., and Riess, O. (1998). Ala30Pro
mutation in the gene encoding alpha-synuclein in Parkinson’s dis-
ease. Nat. Genet. 18, 106–108.
Lee, H.J., Shin, S.Y., Choi, C., Lee, Y.H., and Lee, S.J. (2001a).
Formation and removal of alpha-synuclein aggregates in cells ex-
posed to mitochondrial inhibitors. J. Biol. Chem. 277, 5411–5417.
Lee, M., Hyun, D., Halliwell, B., and Jenner, P. (2001b). Effect of
the overexpression of wild-type or mutant alpha-synuclein on cell
susceptibility to insult. J. Neurochem. 76, 998–1009.
Leroy, E., Boyer, R., Auburger, G., Leube, B., Ulm, G., Mezey, E.,
Harta, G., Brownstein, M.J., Jonnalagada, S., Chernova, T., et al.
(1998). The ubiquitin pathway in Parkinson’s disease. Nature 395,
Maraganore, D.M., Farrer, M.J., Hardy, J.A., Lincoln, S.J., McDon-
nell, S.K., and Rocca, W.A. (1999). Case-control study of the ubiqui-
tincarboxy-terminal hydrolaseL1gene inParkinson’s disease.Neu-
rology 53, 1858–1860.
Masliah, E., Rockenstein, E., Veinbergs, I., Mallory, M., Hashimoto,
M., Takeda, A., Sagara, Y., Sisk, A., and Mucke, L. (2000). Dopamin-
tions for neurodegenerative disorders. Science 287, 1265–1269.
Matsuoka, Y., Vila, M., Lincoln, S., McCormack, A., Picciano, M.,
LaFrancois, J., Yu, X., Dickson, D., Langston, W.J., McGowan, E.,
et al. (2001). Lack of nigral pathology in transgenic mice expressing
human alpha-synuclein driven by the tyrosine hydroxylase pro-
moter. Neurobiol. Dis. 8, 535–539.
McNaught, K.S., and Jenner, P. (2001). Proteasomal function is im-
paired in substantia nigra in Parkinson’s disease. Neurosci. Lett.
McNaught, K.S., Bjorklund, L.M., Belizaire, R., Isacson, O., Jenner,
P., and Olanow, C.W. (2002). Proteasome inhibition causes nigral
degeneration with inclusion bodies in rats. Neuroreport 13, 1437–
Mena, M.A., Khan, U., Togasaki, D.M., Sulzer, D., Epstein, C.J., and
superoxide dismutase on neuronal survival and L-DOPA-induced
toxicity in postnatal midbrain culture. J. Neurochem. 69, 21–33.
Murphy, D.D., Rueter, S.M., Trojanowski, J.Q., and Lee, V.M. (2000).
Synucleins are developmentally expressed, and alpha-synuclein
campal neurons. J. Neurosci. 20, 3214–3220.
Neve, R.L., Howe, J.R., Hong, S., and Kalb, R.G. (1997). Introduction
of the glutamate receptor subunit 1 into motor neurons in vitro and
in vivo using a recombinant herpes simplex virus. Neuroscience 79,
Ostrerova-Golts, N., Petrucelli, L., Hardy, J., Lee, J.M., Farrer, M.,
and Wolozin, B. (2000). The A53T ?-synuclein mutation increases
iron-dependent aggregation and toxicity. J. Neurosci. 20, 6048–
Pakkenberg, B., Moller, A., Gundersen, H.J., Mouritzen Dam, A.,
and Pakkenberg, H. (1991). The absolute number of nerve cells in
substantia nigra in normal subjects and in patients with Parkinson’s
Neurosurg. Psychiatry 54, 30–33.
Paxinou, E., Chen, Q., Weisse,M., Giasson, B.I., Norris, E.H., Rueter,
S.M., Trojanowski, J.Q., Lee, V.M., and Ischiropoulos, H. (2001).
Induction of alpha-synuclein aggregation by intracellular nitrative
insult. J. Neurosci. 21, 8053–8061.
Polymeropoulos, M.H., Lavedan, C., Leroy, E., Ide, S.E., Dehejia, A.,
Dutra, A., Pike, B., Root, H., Rubenstein, J., Boyer, R., et al. (1997).
Mutation in the ?-synuclein gene identified in families with Parkin-
son’s disease. Science 276, 2045–2047.
Rathke-Hartlieb, S., Kahle, P.J., Neumann, M., Ozmen, L., Haid, S.,
Okochi, M., Haass, C., and Schulz, J.B. (2001). Sensitivity to MPTP
is not increased in Parkinson’s disease-associated mutant alpha-
synuclein transgenic mice. J. Neurochem. 77, 1181–1184.
Shimura, H., Hattori, N., Kubo, S., Mizuno, Y., Asakawa, S., Mino-
shima, S., Shimuzu, N., Iwai, K., Chiba, T., Tanaka, K., and Suzuki,
Proteasome Failure and Parkinson’s Disease
T. (2000). Familial parkinson disease gene product, parkin, is a ubi-
quitin-protein ligase. Nat. Genet. 25, 302–305.
Shimura, H., Schlossmacher, M.G., Hattori, N., Frosch, M.P., Trock-
enbacher, A., Schneider, R., Mizuno, Y., Kosik, K.S., and Selkoe,
D.J. (2001). Ubiquitination of a new form of ?-synuclein by parkin
from human brain: implications for Parkinson’s disease. Science
Shtilerman, M.D., Ding, T.T., and Lansbury, P.T. (2002). Molecular
crowding accelerates fibrillization of alpha-synuclein: could an in-
crease in the cytoplasmic protein concentration induce Parkinson’s
disease? Biochemistry 41, 3855–3860.
Stefanis, L., Larsen, K.E., Rideout, H.J., Sulzer, D., and Greene, L.A.
(2001). Expression of A53T mutant but not wild-type ?-synuclein in
PC12 cells induces alterations of the ubiquitin-dependent degrada-
tion system, loss of dopamine release, and autophagic cell death.
J. Neurosci. 21, 9549–9560.
Tanaka,Y., Engelender,S., Igarashi,S.,Rao, R.K.,Wanner, T.,Tanzi,
R.E., Sawa, A.L., Dawson, V., Dawson, T.M., and Ross, C.A. (2001).
Inducible expression of mutant alpha-synuclein decreases protea-
some activity and increases sensitivity to mitochondria-dependent
apoptosis. Hum. Mol. Genet. 10, 919–926.
Tofaris, G.K., Layfield, R., and Spillantini, M.G. (2001). ?-Synuclein
radation by the proteasome. FEBS Lett. 509, 22–26.
Accelerated ?-synuclein fibrillation in crowded milieu. FEBS Lett.
VanDerPutten, H., Widerhold, K.-H., Probst, A., Barbieri, S., Misti,
C., Danner, S., Kauffmann, S., Hofele, K., Spooren, W.P., Ruegg,
M.A., et al. (2000). Neuropathology in mice expressing human
?-synuclein. J. Neurosci. 20, 6021–6029.
Wintermeyer, P., Kruger, R., Kuhn, W., Muller, T., Woitalla, D., Berg,
D., Becker, G., Leroy, E., Polymeropoulos, M., Berger, K., et al.
Xu, J., Kao, S.-Y., Lee, F.J.S., Song, W., Jin, L.-W., and Yankner,
B.A. (2002). Dopamine-dependent neurotoxicity of ?-synuclein: a
mechanism for selective neurodegeneration in Parkinson disease.
Nat. Med. 8, 600–606.
Zhang, Y., Gao, J., Chung, K.K., Huang, H., Dawson, V.L., and Daw-
son, T.M. (2000). Parkin functions as an E2-dependent ubiquitin-
proteinligaseand promotesthedegradationof thesynapticvesicle-
associated protein, CDCrel-1. Proc. Natl. Acad. Sci. USA 97, 13354–
Zhou, W., Hurlbert, M.S., Schaack, J., Prasad, K.N., and Freed, C.R.
(2000). Overexpression of human alpha-synuclein causes dopamine
neuron death in rat primary culture and immortalized mesencepha-
lon-derived cells. Brain Res. 866, 33–43.