Biochemical Characterisation of DJ-1 (PARK7) in Idiopathic
Parkinson’s Disease and Multiple System Atrophy
13.08.2007



Laura L. Kilarski
16C Parkway
NW1 7AA London
l.kilarski@ucl.ac.uk


7 Figures; 1 Table

30 Pages


Keywords: DJ-1, Parkinson’s Disease, Multiple System Atrophy, oxidative stress,
isoelectric point (pI), expression levels


Acknowledgements: I would like to thank Dr. Rina Bandopadhyay and Ravindran
Kumaran for their assistance in and supervision of this study.

2

Biochemical Characterisation of DJ-1 (PARK7) in Idiopathic Parkinson’s Disease and
Multiple System Atrophy
Laura L. Kilarski


Abstract
Loss-of-function mutations of DJ-1 (PARK7), a highly conserved, ubiquitously expressed,
dimeric protein have recently been found to cause early-onset autosomal recessive
Parkinson’s disease (PD). As the role of DJ-1 in idiopathic PD and related
synucleinopathies is unknown, we investigated DJ-1 expression levels by western blotting
in PD medulla, midbrain and cortical tissue, as well as in soluble and insoluble fractions of
multiple system atrophy (MSA) brain homogenate. We showed that expression levels in
diseased tissue did not differ significantly from controls. By 2D gel electrophoresis we
also showed that DJ-1 is expressed in several different pI isoforms, undergoing an acidic
shift in MSA. In addition we showed DJ-1 immunoreactivity in a subset of glial inclusions
in cortical and cerebellar MSA sections, which are considered a pathological alpha-
synuclein containing hallmark of MSA. Attempts to co-immunoprecipitate alpha-synuclein
and DJ-1 failed. We investigated DJ-1 expression levels under oxidative stress conditions
putatively mimicking PD pathology, by western blotting of DJ-1 in astroglioma cells which
had been subjected to MPTP, paraquat, hydrogen peroxide, rotenone and bisphenol A.
We showed no significant changes in DJ-1 expression levels, but were able to
demonstrate an acidic shift in pI following paraquat and hydrogen peroxide
administration. Based on our and previous findings, we speculate that DJ-1 might be
associated with oxidative stress responses in PD and MSA. Further research is necessary
to elucidate the role and function of DJ-1 in the brain.

Introduction
Parkinson’s Disease (PD) is a common neurodegenerative disease, characterised by
selective and progressive loss of dopaminergic (DA) neurons in the substantia nigra (SN),
consequent loss of nigrostriatal projections, and the presence of intracytoplasmic

3
proteinaceous inclusions, called Lewy bodies (Moore et al 2005a; Spillantini et al 1997).
Multiple System Atrophy (MSA) displays similar pathology, including glial inclusions. Both
PD and MSA manifest as motor function disturbances, followed by cognitive decline (Lang
& Lozano 1998a), but MSA also affects the autonomic nervous system (Wenning et al
2004).
While MSA is considered purely idiopathic, 5-10% of PD is genetically inherited. 10
different loci have been identified to date, including alpha-synuclein and parkin (Dawson
& Dawson 2003). The PARK7 locus hosts the DJ-1 gene, mutations of which have been
shown to cause autosomal recessive early onset PD (Bonifati et al 2003). A number of PD
causing mutations resulting in loss of function have been identified to date (eg Miller et al
2003; Olzmann et al 2004).
DJ-1 exists as a homodimer (Huai et al 2003) and has been implicated with several roles
in systemic tissue (Bandyopadhyay & Cookson 2004), but its function in the brain is
unknown. DJ-1 is abundant especially in subcortical regions, in both neurones and
astrocytes (Bandopadhyay et al 2004). DJ-1 localises to the cytoplasm, but can
translocate to mitochondria under oxidative stress conditions (Zhang et al, 2005).
Several lines of evidence suggest that DJ-1 counteracts oxidative stress. DJ-1 deficiency
in vitro (Martinat et al, 2004), in Drosophila DJ-1 mutants (Meulener et al, 2005a) and
KO mice (Kim et al, 2005) confers hypersensitivity to reactive oxygen species (ROS). DJ-
1 is thought to be neuroprotective and neutralise ROS by readily oxidising its C106 to
cysteine sulfinic acid (Taira et al, 2004; Canet-Aviles et al, 2005), by preventing
apoptosis (Junn et al, 2005), increasing cellular glutathione (GSH) concentration and
upregulating heat shock protein 70 (Hsp70), which is part of the ubiquitin proteasome
system (Zhou & Freed 2005, Zhou et al 2006, Li et al, 2005).
Malfunctioning of the UPS, oxidative stress and mitochondrial defects have been
implicated as some of the putative causes underlying PD and related neurodegenerative
disorders (Betarbet et al 2006; Dauer & Przedborski 2003; Giasson & Lee 2003; Li et al
2005), thus determining the function of DJ-1 and its relevance to idiopathic PD could

4
advance our understanding of the mechanisms responsible for nigral cell death and LB
formation.
Using western blot we investigated DJ-1 expression levels in IPD and MSA, as well as in
astroglioma cell culture treated with paraquat, rotenone, H2O2, MPTP, and Bisphenol A.
We also investigated shifts in isoelectric point of DJ-1 isoforms in MSA and astroglioma
by 2D gel electrophoresis. Localisation of DJ-1 and alpha-synuclein in MSA cortex and
cerebellum was visualised with immunohistochemistry. Immunoprecipitation was used to
detect a putative interaction between DJ-1 and alpha-synuclein in MSA/PD.

Materials and Methods
Tissues
Brain tissue of PD, MSA and control cases (see table 1 for details) were obtained from the
brain tissue archive at Queen Square Brain Bank in accordance with the London
Multicentre Research Ethics Committee and the National Hospital for Neurology and
Neurosurgery/Institute of Neurology Joint Research Ethics Committee.

Tissue Preparation
Approximately 1g frozen brain tissue per sample was mechanically homogenised in
homogenisation buffer A (50mM Tris-HCl pH 7.5, 1mM EGTA, 1mM DTT, protease
inhibitor cocktail (Roche Diagnostics, Mannheim, Germany), dH2O) and centrifuged at
1000g for 3 min. The supernatant was ultracentrifuged at 350000g (45000 rpm) for 20
min and frozen until further use. The pellet was resuspended in buffer A additionally
containing 1% sarkosyl, 10% sucrose and 0.5M NaCl, incubated at 37°C for 1h, and
ultracentrifuged at 350000g (45000 rpm) for 20 min. Protein quantity in both soluble and
insoluble fractions was determined by the bicinchoninic acid protein assay method (Bio-
Rad, UK) using BSA (Sigma-Aldrich, St. Louis, MO, USA) as standard.

5
Cell Culture Preparation
U373 astroglioma cells (European Collection of Cell Cultures) were cultivated, incubated
with toxins and kindly provided by Ravindran Kumaran. Untreated cells from the
respective batch constituted the controls for each set of samples. For sample set A, cells
were treated with 25, 50, 100, 150 or 200 µM rotenone, hydrogen peroxide (H2O2)
paraquat, bisphenol A (BPA), or MPTP (all Sigma-Aldrich), overnight, totalling 25 distinct
samples plus controls (n = 1). For sample set B, cells were treated with 100 or 200 µM
H2O2 or paraquat for either 2, 4, or 6h in triplicate, totalling 12 distinct samples plus
controls (n = 3).
Cells were suspended in a small volume of 50mM Tris-HCl buffer including protease
inhibitor complex, and sonicated (Sonics Vibracell™, Sonics & Materials Inc., Newtown,
CT, USA) for approximately 10 sec. Samples were centrifuged at 5000 rpm for 10 min,
and the pellet was discarded. Protein quantity was determined as above.

Western Blot
Regional Changes in DJ-1 Expression Levels in IPD: 10 µg of protein from homogenates
of medulla (P3, P6, P7, C1, C3, C5), midbrain and cortex (both P1, P2, P5, C1, C3, C4) of
PD cases and controls were resuspended in 50% Novex Tris-Glycine SDS sample buffer
(Invitrogen, Paisley, UK) including 10% reducing agent (Invitrogen), loaded onto Tris-Gly
gels (Invitrogen) and run at 120V with Tris-Glycine running buffer (Invitrogen) in
duplicate. Proteins were blotted onto Hypond P PVDF transfer membrane (Amersham
Biosciences, UK) and membranes were blocked with 5% milk (Marvel, UK) in phosphate-
buffered saline (Roche) plus 0.1% Tween 20 (Sigma) (PBS-T). Subsequently, blots were
incubated with DJ-1 antibody (clone 3E8; Stressgen, San Diego, CA, USA) at 1:5000
dilution at 4°C overnight and duplicate blots were incubated with monoclonal anti-ß-actin
(clone AC-15, Sigma) at 1:1000 dilution at 4°C overnight. This was followed by
incubation with 1:2500 goat anti-mouse horseradish peroxidase-conjugated
immunoglobulin (IgG-HRP) (sc-2031, Santa Cruz Biotechnology, Santa Cruz, CA, USA)
for 1h at room temperature. Primary and secondary antibody application in this and

6
subsequent experiments was always followed by 3x5min washes with PBS-T.
Immunoreactivity was detected by enhanced chemiluminescent reaction (ECL) (Super
Signal West Pico Luminol Enhancer & Pico Stable Peroxide Solution, Pierce, Rockford, IL,
USA) and blots were developed onto Kodak Biomax autoradiography film (Eastman
Kodak Company, NY, USA). Results were quantified with Quantity One™ (Bio-Rad).

Changes in DJ-1 Expression Levels and Isoelectric Point in MSA: 10µg of protein from the
soluble fractions and 8 µg of protein from the insoluble fractions of MSA tissue samples
(P10, P11, P12, P14) and controls (C2, C6, C7, C8) were resuspended in 25% NuPage
sample buffer (Invitrogen) incl. 10% reducing agent (Invitrogen) and loaded onto 10%
Bis-Tris gels (Invitrogen). Using MES (morpholinoethane sulphonic acid) buffer
(Invitrogen), gels were run at 200V for approx. 45 min. Blots were transferred and
blocked with milk as described above, and subsequently incubated with either DJ-1
antibody (Stressgen) at 1:5000 dilution or alpha-synuclein antibody (MAb, BD
Biosciences, San Jose, CA, USA) at 1:300 dilution at 4°C overnight. Anti-mouse IgG-HRP
(Santa Cruz Biotechnology) was applied to the blots at 1:2000 for one hour at room
temperature. Blots were developed and quantified as above. The soluble fractions of said
samples were run in duplicate as above and incubated with β-actin antibody (Sigma-
Aldrich) at 1:1000 dilution and then HRP-IgG at 1:1000 before developing and
quantifying as above.

Effects of Toxin Treatment on DJ-1 expression levels: 3µg protein from each sample in
both sample set A and B were run on Bis-Tris gels, transferred to membrane and blocked
as above. Blots were incubated with DJ-1 antibody (Stressgen) at 1:2000 dilution
overnight at 4°C, followed by anti-mouse IgG-HRP (Santa Cruz Biotechnology) at 1:5000
for 1h at room temperature. Blots were developed and quantified as above. In order to
reprobe with β-actin antibody, blots were stripped by bathing in stripping buffer (2%
Sodium dodecyl sulfate (SDS), 50mM Tris-HCl pH 6.8, 100mM β-mercaptoethanol
(Sigma)) at 50°C for 30 minutes and then washed thoroughly with PBS-T. Blots were

7
then incubated with 1:1000 β-actin antibody (Sigma) overnight at 4*C, followed by
1:5000 IgG-HRP for one hour at room temperature. Blots were developed and quantified
as above. Density values obtained from DJ-1 blots were normalised against actin.

Immunoprecipitation (IP) of DJ-1 and alpha-synuclein in MSA and PD
Protein G-agarose (Pharmacia, UK) was washed by addition of lysis buffer (50mM Tris-
HCl, 150mM NaCl, 5mM EDTA, 0.25% NP-40, protease inhibitor complex (Roche)) and
centrifugation at 1000g for 1min, removal of the buffer, new addition and repeated
centrifugation (repeat 5 times). 2x 200µg PD/MSA (P4, P13) and 2x control (C6, C7)
protein samples were suspended in 300µl lysis buffer and incubated with 50µl washed
protein G-agarose (Invitrogen) bead slurry for 2h at 4°C. Samples were centrifuged at
10000g for 10min, and the supernatant was kept. 20µl supernatant were stored at –20°C
overnight for the cleared lysate (CL) portion of the IP. Of the PD/MSA and control
samples, one each was incubated with 10µl DJ-1 AB (Stressgen), the other with 10µl
alpha-synuclein antibody (BD Biosciences) at 4°C overnight. The next day, 100µl bead
slurry was added to each sample, incubated for 2h. A mock IP was prepared containing
500µl lysis buffer and 100µl bead slurry. Samples were washed with wash buffer (50mM
Tris-HCl, 150mM NaCl, 0.5mM EDTA, 0.05% NP-40, protease inhibitor complex (Roche))
via centrifugation as above, removing all supernatant after the last centrifugation. For
the IP samples, 20µl 8M urea and 16µl Nu-Page sample buffer (Invitrogen) were added
to the bead slurry and heated at 65°C for 15min. For the CL, 25µl Nu-Page sample buffer
and 5µl reducing agent were added per sample, then heated for 10min at 70°C. All
samples were briefly spun down and loaded on Tris-Gly gels, and run/transferred as
outlined in the WB section under experiment 1. One blot was probed with 1:500
polyclonal goat alpha-synuclein antibody (C-20, Santa Cruz Biotechnology), the other
with 1:2500 DJ-1 (Stressgen) for 2h at room temperature. For the alpha-synculein
probed blot, polyclonal anti-goat IgG-HRP (Santa Cruz Biotechnology) was applied at
1:1000, the DJ-1 probed blot was incubated with anti-mouse IgG-HRP (Santa Cruz

8
Biotechnology) at 1:5000, both for 1h at room temperature. Blots were developed as
above.

2D Gel Electrophoresis
Changes in Isoelectric Point (pI) of DJ-1 in MSA: 25µg protein from soluble MSA and
control homogenate samples (P10, P12, P14, C2, C6, C7) was separated on 7cm
immobilised pH gradient (IPG, Amersham Biosciences) strips with 4.0-7.0 IPG buffer
(Amersham Biosciences) on the IPGPhor system (Amersham Biosciences). After
incubation with SDS-Page buffer, first containing 0.1g/10ml dithiothreitol (DTT,
Amersham) and second containing 0.25g/10ml iodoacetamide (IDT, Amersham) at room
temperature for 20min each, strips were resolved in the 2nd dimension on Tris-Gly gels
(Invitrogen), transferred, blocked and incubated with DJ-1 antibody as above for western
blots of the same experiment.
Effects of Toxin Treatment on pI of DJ-1: 10µg protein from soluble rotenone, paraquat
and H2O2 treated (see sample set B for details) cell culture samples was separated as
above, except that the 2nd dimension was resolved on Bis-Tris gels according to the WB
procedure for experiment 3.

Immunohistochemistry
8µm formalin-fixed wax-embedded sections of basal ganglia and cerebellum (C5, P8, P9,
P15, P16, P17) were dewaxed in xylene. Endogenous peroxidase reactions were blocked
with 0.5% H2O2 in methanol for 10min.
For DJ-1 staining, sections were pressure cooked in citrate buffer (pH 6; 5.8g Tri-
Sodium-Citrate, 0.48g Citric Acid, 2l dH2O), while for alpha-synuclein staining, sections
were treated with 95% formic acid for 5min. Sections were blocked with 5% milk
(Marvel) in phosphate buffer (Roche), before incubating overnight with either 1:1000 DJ-
1 antibody (1130, a generous gift of P. Rizzu and P. Heutinck, University of Rotterdam),
1:500 DJ-1 antibody (Stressgen) or 1:500 alpha-synuclein antibody (LB509 mouse,
Zymed, San Francisco, CA, USA). Sections were then incubated with either 1:500 anti-

9
mouse monoclonal IgG-HRP (Santa Cruz Biotechnology), followed by avidin-biotin-
peroxidase complex (ABC; Vector Laboratories, Peterborough, UK) and diamino benzidine
(DAB) (Sigma) for visualisation. Sections were lightly counterstained with haematoxylin
(Sigma), dehydrated and fixed with mounting medium (Merck, UK).


Results
Regional Changes in DJ-1 Expression Levels in IPD
Western blotting of medulla, midbrain, and cortex homogenates from IPD (IPD) and
control tissue with DJ-1 antibody revealed a single band at 20 kD, corresponding to the
DJ-1 monomer (Fig.1 a, upper panel). Specificity for DJ-1 had previously been
established (Bandopadhyay et al 2004). 10 µg protein were loaded per lane, and in order
to account for putative loading differences, DJ-1 levels were normalised against β-actin
(Fig.1 a, lower panel). There was no significant change in DJ-1 expression levels between
controls and IPD, in any of the three brain regions examined (Fig.1 b). In medulla and
midbrain, expression levels in IPD decreased by 33.78% and 3.84% respectively, as
compared to controls. In cortex, DJ-1 levels increased by 1.08% (n=3, p>0.05).


Changes in DJ-1 Expression Levels and Isoelectric Point in MSA.
Immunoblot analysis of soluble and insoluble fractions of MSA and control brain
homogenate revealed the characteristic 20 kD DJ-1 band (Fig.2 a,b), as well as two
higher molecular weight bands corresponding to the DJ-1 homodimer in soluble fractions
(Fig.2a). Dimers were not expected in the insoluble fraction, as DJ-1 is thought to
aggregate only in monomeric form. DJ-1 protein levels were normalised against β-actin,
and by comparing MSA cases to control it was show that mean DJ-1 expression levels in
the soluble and insoluble fractions in MSA increased by 42.91% and 36.00%, respectively
(Fig.2c). However, a two-tailed t-test did not determine the change in expression levels
to be statistically significant (n=3, p>0.05).

10
Next, we separated 25 µg protein from three MSA samples and three controls by 2D gel
electrophoresis on 4.00-7.00 IPG strips. Representative images are shown in Fig. 2d. It
has been previously found that DJ-1 exists in at least six different pI isoforms, ranging
from 5.5 to 6.6 (Bandopadhyay et al 2004). We were able to replicate this result in
controls. In addition, we show accumulation of acidic DJ-1 isoforms, at around 5.8 in
comparison to control. Also, one of the MSA samples missed the pI 6.6 isoform,
comparable to findings by Bandopadhyay et al (2004) in IPD samples. The MSA samples
investigated also appear to show higher concentrations of DJ-1 as compared to control,
further strengthening the indication that DJ-1 expression levels are increased in MSA as
hinted at by the western blot experiments described above.
Immunohistochemistry
Cerebellar and cortical slices of MSA and control brains were immunostained for either
DJ-1 or alpha-synuclein (Fig.3). Alpha-synuclein positive glial inclusions are a
pathological hallmark of MSA (Wenning et al 2004), similar to LBs in PD. Both DJ-1 and
alpha-synuclein positive oligodendroglial inclusions were found in cortex and cerebellum,
although alpha-synuclein immunoreactivity (IR) was observed more frequently. DJ-1 IR
was observed in a number of glial inclusions and, less frequently, in neurons (Fig.3 b, g).
The cytoplasm was most commonly affected, although rarely, nuclei also stained positive
for DJ-1. While it was not possible to detect a co-localisation of DJ-1 and alpha-synuclein
with the type of staining used, the fact that glial inclusions containing DJ-1 were found
suggests that both proteins aggregate together at least to some extent.

Immunoprecipitation of DJ-1 and alpha-synuclein in MSA and PD
In order to determine whether DJ-1 and alpha-synuclein co-localise and possibly even
interact in MSA, we attempted to immunoprecipitate DJ-1 with alpha-synuclein and vice
versa, in samples from one PD and three MSA cases. While immunoprecipitating DJ-1
alone was successful (Fig.4), DJ-1 and alpha-synuclein could not be co-
immunoprecipitated in any of the samples (data not shown).

11
DJ-1 Expression Levels in Astroglioma Cells after Toxin Treatment
For the first set of experiments, astroglioma cell cultures were incubated overnight with
varying concentrations of H2O2, paraquat, rotenone, Bisphenol A (BPA) and MPTP (see
methods: sample set A). Western blotting of 3µg soluble protein from each sample
revealed the typical 20kD DJ-1 band (Fig.5 f), the density of which was normalised
against actin, and expressed as percentage of control. (Fig.5 a-e). Paraquat is a
herbicide, implicated as one of the putative causes for IPD (Liou et al 1997) and paraquat
treatment (Fig.5a), appeared as a promising method of changing DJ-1 expression levels,
with treated cells expressing approximately 50% less DJ-1 than control. However, no
obvious dose-dependency was observed. Results obtained from treating cells with MPTP
(Fig.5b) were highly variable, ranging from 40-160% of control, with a mean of
approximately 100%, with no dose-dependency. MPTP is a mitochondrial electron
transport chain complex I inhibitor, and has been widely used to model PD pathogenesis
in mice and non-human primates (reviewed in Bové et al 2005). BPA, an environmental
oestrogen, has been shown to induce hydroxyl radical formation in mouse brain (Kabuto
et al 2003), and BPA treatment (Fig.5c) resulted in a general increase in DJ-1 expression
levels, from approximately 138-178%, however, the distribution of values along
increasing concentrations of BPA was very variable. H2O2 treatment did not result in any
alterations of DJ-1 levels in comparison to control. On the other hand, rotenone displayed
an increase in DJ-1 expression levels, from about 120% to 170% of control, in a dose-
dependant manner.
Following this experiment, we tried to determine whether the duration of incubation with
a given toxin could influence DJ-1 expression levels. Astroglioma cells were incubated
with either 100µM or 200µM paraquat or H2O2 for either 2, 4, or 6h (n=3 per
combination). 3 µg protein were loaded per lane, resulting in the characteristic 20kD
bands (Fig.6c). After stripping, the blots were probed with β-actin antibody (also Fig. 6c),
to normalise DJ-1 levels against actin levels. While paraquat treated cells (Fig.6a)
displayed a slight decrease in comparison to controls, the finding was not statistically
significant (two-tailed t-test, p>0.05) and seemingly non-related to concentration or

12
duration of incubation. Incubation with H2O2 resulted in a very small, statistically
insignificant increase in DJ-1 expression levels, however it would be interesting to see
the effects of either substances in a larger sample size.

Influence of Rotenone and H2O2 on the Isoelectric Point of DJ-1
We separated 10 µg protein from three rotenone treated (control, 50 µM/overnight
incubation, 100 µM/overnight incubation) and three H2O2 treated cell culture samples
(control, 100 µM/4h incubation, 200 µM/4h incubation) by 2D gel electrophoresis on
4.00-7.00 IPG strips. Representative images are shown in Fig. 7. In rotenone treated
samples, there is a clear shift in pI, with acidic isoforms of DJ-1 accumulating at 5.5-5.8.
In addition, rotenone treated samples show a dose-dependent loss of more basic
isoforms as compared to control, and a slight decrease of DJ-1 protein levels (Fig.7 a).
H2O2 treated samples, on the other hand, displayed less clear results. At a concentration
of 100 µM H2O2, one acidic isoform of DJ-1 (pI ~5.8) was enriched, while the other (pI
~5.5) was missing in comparison to control. As expected, the pI 6.2 isoform was less
prominent as compared to control. However, at 200 µM H2O2, the pI 6.2 isoform was only
slightly less detectable in comparison to control, while more acidic isoforms were
abundant in comparison.


Discussion

Regional Changes in DJ-1 Expression Levels in IPD
DJ-1 mutations conferring loss of function have been shown to cause autosomal
recessive PD (Bonifati et al 2003, Lev et al 2006) and DJ-1 KO mice display motor
deficits reminiscent of PD symptoms (Chen et al 2005). Since increased oxidative
damage has been implicated as one of the putative causes of neurodegeneration in PD
(reviewed in Jenner 2003) and DJ-1 has been shown to act protectively against ROS
induced cell death (Zhou et al 2006), DJ-1 could theoretically be involved in the cellular
pathology in idiopathic PD.

13
Medulla, midbrain (eg SN and striatum), and cortex are brain regions known to be
affected in PD. Our investigation of DJ-1 expression levels in medulla, midbrain and
cortex revealed no significant changes in PD, although there appears to be a downwards
trend of DJ-1 expression levels in medulla. As the sample size in this study was small
(n=3) and resulted in large standard mean error values, a larger sample size could
possibly demonstrate a significant change in DJ-1 expression levels. However, since DJ-1
is mostly expressed in astrocytes (Bandopadhyay et al 2004), neuronal cell death may
only have a modest effect on overall tissue expression levels.

Changes in DJ-1 Expression Levels and Isoelectric Point in MSA.
MSA is a synucleinopathy displaying a number of similarities to IPD, including nigral
neurodegeneration and proteinaceous inclusions containing alpha-synuclein. Co-
localisation of DJ-1 and alpha-synuclein, and increased levels of DJ-1 in sarsosyl-
insoluble fractions in a small number of MSA cases have recently been observed
(Neumann et al 2004). Whether DJ-1 plays a role in MSA is currently unclear. We aimed
to further investigate DJ-1 biochemistry and localisation in MSA. Immunoblot analysis
showed that mean DJ-1 expression levels in the soluble and insoluble fractions in MSA
increased (Fig.2c), but these changes were not statistically significant. As before, an
increase in sample size might have rendered the observed trend significant.

Investigation of isoelectric point of DJ-1 in MSA revealed the presence of several different
isoforms, and shows that DJ-1 undergoes an acidic shift in pI (Fig.2 d) similar to that
previously observed in PD (Bandopadhyay et al 2004). An acidic shift in pI can result
from oxidation by ROS and is thought to render DJ-1 non-functional (Kinumi et al 2004).
Furthermore, oxidative stress has been implicated in alpha-synuclein aggregation in glial
inclusions (Riedel et al 2007), the pathological hallmark of MSA (reviewed in Wenning &
Jellinger 2004).
Using immunohistochemistry techniques, we showed glial inclusions in cerebellar and
cortical MSA sections to immunostain positive for alpha-synuclein and observed that

14
some inclusions, and rarely also neurons, stained positive for DJ-1 as well (Fig. 3).
Although these findings do not demonstrate co-localisation of DJ-1 and alpha-synuclein
directly, it is likely that those inclusions positive for DJ-1 were also positive for alpha-
synuclein. Thus our results support the findings of Neumann et al (2004), who showed
co-localisation using double fluorescent immunohistochemistry.
However, co-localisation does not necessarily translate into direct interaction of alpha-
synuclein and DJ-1. While Meulener et al (2005b) showed that 3% of alpha-synuclein co-
immunoprecipitated with DJ-1, we failed to repeat this finding despite numerous trials.
Two other groups also failed to co-immunoprecipitate DJ-1 and alpha-synuclein (Macedo
et al 2003, Miller et al 2003). Yet, it is possible that the putative interaction is transient
and thus not easily detectable, that our samples did not contain enough protein to show
the relatively small degree of co-immunoprecipitation, or that the epitopes targeted by
the antibodies used were hidden due to conformational changes following an interaction.

DJ-1 Expression Levels in Astroglioma Cells after Toxin Treatment
Environmental factors, including certain toxins, are thought to play an important role in
PD etiology (eg Gorell et al 1998, Semchuk et al 1992). In addition, astrocytes are
thought to assist neurons in battling oxidative stress (Bandopadhyay et al 2004). In
terms of a preliminary investigation, we thus treated an astroglioma cell culture with a
number of different toxins, to see whether these would have an effect on DJ-1 expression
levels.
DJ-1 expression levels after MPTP treatment were highly variable, with no mean
difference compared to control (Fig.5 b). It should be noted that MPTP induced cell death
could have had a negative impact on protein levels as a whole, indeed it was difficult to
obtain sufficiently large samples for western blotting. The active metabolite of MPTP,
MPP+, induces generation of superoxide radicals and peroxynitrite, arrest of ATP
synthesis and activation of apoptotic pathways in DAergic neurons of the SN, thus
mimicking some of the putative mechanisms of PD pathology (reviewed in Przedborski &
Vila 2001). Tyrosine hydroxylase and alpha-synuclein are selectively nitrated by

15
peroxynitrite (reviewed in Przedborski & Vila 2001) and DJ-1 KO mice have been shown
to be hypersensitive to MPTP (Kim et al 2005), thus it would be of interest to repeat the
experiment.
Paraquat treatment (Fig. 5 a, 6 a) resulted in a moderate mean DJ-1 expression level
decrease as compared to controls, however, no dose- or time-dependency was observed.
Paraquat exerts its neurotoxic effect by inducing oxidative stress and inhibiting
proteasome activity, and eg causing aggregation of alpha-synuclein (reviewed in Sun et
al 2007). Furthermore, a shift in pI of DJ-1 isoforms has been observed following
paraquat treatment of human umbilical vein endothelial cells (Mitsumoto et al 2001).
Repetition of the experiment with a larger sample size is highly recommended.
BPA is thought to induce hydroxyl radical formation by acting on mitochondria (reviewed
in Obata 2002). Furthermore, BPA is thought to inhibit anti-apoptotic nuclear factor
kappa B, thus facilitating apoptosis (Lee et al 2007), and to influence responsiveness of
neurons and astrocytes to dopaminergic signals (Miyatake et al 2006). BPA treatment
(Fig.5c) resulted in a mean increase in DJ-1 expression levels by ~50% compared to
control, however, the change did not appear to be dose-dependent. Previously, Ooe et al
(2005) also showed increased DJ-1 expression levels both in mice as well as in mouse
Neuro2a and GC1 cells, however they were able to demonstrate dose-dependency.
Furthermore, Ooe et al observed a significant increase in acidic pI isoforms of DJ-1 in cell
culture after treatment with BPA.
Wild-type DJ-1 aids to eliminate H2O2 by oxidising itself (Canet-Aviles et al 2005, Taira et
al 2004, Zhou & Freed 2005), the most sensitive residue being Cysteine 106 (Kinumi et
al 2001). It has been suggested that DJ-1 has chaperone activity, preventing alpha-
synuclein aggregation. However, this is only the case when C106 of DJ-1 is oxidised, not
when DJ-1 is in its native isoform or further oxidised (Zhou et al 2006). Furthermore, DJ-
1 has been shown to interact with parkin after incubation with H2O2 (Moore et al 2005b).
In our first set of cell culture experiments, H2O2 treatment did not result in any mean
changes in DJ-1 expression levels (Fig.5 d), and the second set showed a very small,
statistically insignificant increase. While upregulation of DJ-1 after H2O2 treatment has

16
been previously observed in Drosophila (Menzies et al 2005), these results cannot
necessarily be extrapolated to human cell culture. 2D gel electrophoresis of H2O2 treated
samples (Fig. 7 b) showed a general increase of DJ-1 pI 5.8 isoform levels. pI 5.8
corresponds to the DJ-1 isoform in which C106 has been oxidised to cysteine-sulfinic acid
(Canet-Aviles et al 2005).
Rotenone inhibits mitochondrial electron chain complex I, similar to MPTP, but cells are
capable of upregulating proteasomal function as a defence mechanism given that some
level of ATP supply is maintained (reviewed in Sun et al 2007). Rotenone induced
apoptosis can be halted by the use of antioxidants or by replacing complex I with a
bacterial variant that is unsusceptible to inhibition by rotenone (Sherer et al 2003).
However, prolonged exposure to rotenone eventually results in accumulation of ROS and
suppression of proteasomal activity, leading to alpha-synuclein accumulation and DJ-1
oxidation in vivo (Betarbet et al 2006). In our study we found DJ-1 expression levels to
increase after treatment with rotenone in a roughly dose-dependant manner (Figure 5 e).
This could possibly be a cellular attempt to counteract accumulation of ROS. 2D gel
electrophoresis revealed a clear acidic shift in pI (Fig. 7 a), in concordance with Betarbet
et al (2006).

While the causes of neurodegeneration in PD and MSA remain incompletely understood,
an increasing amount of evidence points towards the involvement of oxidative stress,
mitochondrial dysfunction and proteasome defects, all of which intersect with one
another, resulting in accumulation of defective proteins, eg alpha-synuclein aggregation
and ultimately apoptosis. While our study of DJ-1 in PD, MSA and cell culture yielded
limited results, it does show an effect of oxidative stress conditions (as in MSA and in
culture) on DJ-1 isoforms. Further research on a bigger scale is needed to assess DJ-1
expression levels in PD and MSA. Detailed characterisation of DJ-1 including its function
in healthy and diseased brain tissue, and its interactions with other proteins, could aid us
in understanding the molecular causes of PD and other synucleinopathies, ultimately
resulting in finding novel therapeutic strategies.

17

References

Bandopadhyay R, Kingbury A, Cookson MR, Reid AR, Evans IM, Hope AD, Pittman AM,
Lashley T, Canet-Aviles RM, Miller DW, McLendon C, Strand C, Leonard AJ, Abou-Sleiman
PM, Healy DG, Ariga H, Wood NW, de Silva R, Revesz T, Hardy AJ, Lees A (2004). The
expression of DJ-1 (PARK7) in normal human CNS and idiopathic Parkinson’s disease.
Brain 127: 420-430.

Bandyopadhyay S, Cookson MR (2004) Evolutionary and functional relationships within
the DJ-1 superfamily. BMC Evolutionary Biology 4: 6.

Betarbet R, Canet-Aviles RM, Sherer TB, Mastroberardino PG, McLendon C, Kim J-H, Lund
S, Na H-M, Taylor G, Bence NF, Kopito R, Seo BB, Yagi T, Yagi A, Klinefelter G, Cookson
MR, Greenamyre JT (2006). Intersecting pathways to neurodegeneration in Parkinson’s
disease: Effects of the pesticide rotenone on DJ-1, A-synuclein, and the ubiquitin–
proteasome system. Neurobiology of Disease 22: 404-20.

Bonifati, V., Rizzu, P., van Baren, M.J., Schaap, O., Breedveld, G.J., Krieger, E., Dekker,
M.C., Squitieri, F., Ibanez, P., Joosse, M., et al. (2003). Mutations in the DJ-1 gene
associated with autosomal recessive early-onset parkinsonism. Science 299, 256–259.

Bové J, Prou D, Perier C, Przedborski S (2005) Toxin-Induced Models of Parkinson’s
Disease. NeuroRx 2: 484–494.

Canet-Aviles RM, Wilson MA, Miller DW, Ahmad R, McLendon C, Bandyopadhyay S,
Baptista MJ, Ringe D, Petsko GA, Cookson MR (2005) The Parkinson's disease protein DJ-
1 is neuroprotective due to cysteine-sulfinic acid-driven mitochondrial localization. PNAS
101: 9103–9108.

Chen L, Cagniard B, Mathews T, Jones S, Koh HC, Ding Y, Carvey PM, Ling Z, Kang UJ,
Zhuang X (2005) Age-dependent motor deficits and dopaminergic dysfunction in DJ-1
null mice. J Biol Chem 280: 21418-21426.

Cookson, M.R. (2004). The biochemistry of Parkinson’s disease. Annu. Rev. Biochem. 74,
29–52.

Dauer W, Przedborski S (2003) Parkinson's Disease: Mechanisms and Models. Neuron
39: 889-909.

Dawson, TM, Dawson, VL (2003). Molecular pathways of neurodegeneration in
Parkinson’s disease. Science 302: 819–822.

De Silva R, Lashley T, Gibb G, Hanger D, Hope A, Reid A, Bandopadhyay R, Utton M,
Strand C, Jowett T, Khan N, Anderton B, Wood N, Holton J, Revesz T, Lees A (2003)
Pathological inclusion bodies in tauopathies contain distinct complements of tau with
three or four microtubule-binding repeat domains as demonstrated by new specific
monoclonal antibodies. Neuropathol Appl Neurobiol 29: 288-302.

Giasson BI, Lee VMY (2003). Are ubiquitination pathways central to Parkinson’s disease?
Cell 114: 1–8.

Gorell JM, Johnson CC, Rybicki BA, Peterson EL, Richardson RJ (1998) The risk of
Parkinson’s disease with exposure to pesticides, farming, well water, and rural living.
Neurology 50: 1346–1350.

18
Huai Q, Sun Y, Wang H, Chin L-S, Li L, Robinson H, Ke H (2003) Crystal structure of DJ-
1/RS and implication on familial Parkinson’s disease. FEBS Letters 549: 171-175.

Jenner P (2003) Oxidative stress in Parkinson's disease. Annals of Neurology 53: S26 -
S38.

Junn E, Taniguchi H, Jeong BS, Zhao X, Ichijo H, Mouradian MM (2005) Interaction of DJ-
1 with Daxx inhibits apoptosis signal-regulating kinase 1 activity and cell death. PNAS
102: 9691-9696.

Kabuto H, Hasuike S, Minawaga N, Shishibori T (2003) Effects of bisphenol A on the
metabolisms of active oxygen species in mouse tissues. Environmental Res 93: 31-35.

Kim RH, Smith PD, Aleyasin H, Hayley S, Mount MP, Pownall S, Wakeham A, You-Ten AJ,
Kalia SK, Horne P, Westaway D, Lozano AM, Anisman H, Park DS, Mak TW (2005).
Hypersensitivity of DJ-1-deficient mice to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyrindine
(MPTP) and oxidative stress. PNAS 102: 5215–5220.

Kinumi, T., Kimata, J., Taira, T., Ariga, H., and Niki, E. (2004). Cysteine-106 of DJ-1 is
the most sensitive cysteine residue to hydrogen peroxide-mediated oxidation in vivo in
human umbilical vein endothelial cells. Biochem. Biophys. Res. Commun. 317, 722–728.

Lang AE, Lozano AM (1998) Parkinson’s Disease – First of two parts. N Engl J Med. 339:
1044-1053.

Lee YM, Seong MJ, Lee JW, Lee YK, Kim TM, Nam SY, Kim DJ, Yun YW, Kim TS, Han SY,
Hong JT (2007) Estrogen receptor independent neurotoxic mechanism of bisphenol A, an
environmental estrogen. J Vet Sci 1: 27-38.

Lev N, Roncevich D, Ickowicz D, Melamed E, Offen D (2006). Role of DJ-1 in Parkinson’s
disease. J Mol Neurosci 29: 215-225.

Li HM, Niki T, Taira T, Iguchi-Ariga SMM, Ariga H (2005) Association of DJ-1 with
chaperones and enhanced association and colocalization with mitochondrial Hsp70 by
oxidative stress. Free Radical Research 39: 1091–1099.

Liou HH, Tsai MC, Chen CJ, Jeng JS, Chang YC, Chen SY, Chen RC (1997) Environmental
risk factors and Parkinson’s disease: a case-control study in Taiwan. Neurology 48:
1583–1588.

Macedo MG, Anar B, Bronner IF, Cannella M, Squitieri F, Bonifati V, Hoogeveen A,
Heutink P, Rizzu P (2003) The DJ-1L166P mutant protein associated with early onset
Parkinson's disease is unstable and forms higher-order protein complexes. Hum Mol
Genet 12: 2807–2816.

Martinat C, Shendelman S, Jonason A, Leete T, MF Beal, Yang L, Floss T, Abeliovich A
(2004). Sensitivity to oxidative stress in DJ-1-deficient dopamine neurons: An ES-derived
cell model of primary Parkinsonism. PLoS Biol 11: e327.

Menzies FM, Yenisetti SC, Min KT (2005) Roles of Drosophila DJ-1 in Survival
of Dopaminergic Neurons and Oxidative Stress. Current Biology 15: 1578–1582.

Meulener MC, Whitworth AJ, Armstrong-Gold CE, Rizzu P, Heutink P, Wes PD, Pallanck LJ,
Bonini NM (2005a) Drosophila DJ-1 Mutants Are Selectively Sensitive to Environmental
Toxins Associated with Parkinson’s Disease. Curr Bio 15: 1572–1577.

19
Meulener MC, Graves CL, Sampathu DM, Armstrong-Gold CE, Bonini NM, Giasson BI
(2005b) DJ-1 is present in a large molecular complex in human brain tissue and interacts
with α-synuclein. J Neurochem 93: 1524-1532.

Miller DW, Ahmad R, Hague S, Baptista MJ, Canet-Aviles R, McLendon C, Carter DM, Zhu
PP, Stadler J, Klinefelter GR, Blackstone C, Cookson MR (2003) L166P mutant DJ-1,
causative for recessive Parkinson’s disease, is degraded through the ubiquitin-
proteasome system. J Biol Chem 38: 36588-36595.

Mitsumoto A, Nakagawa Y, Takeuchi A, Okawa K, Iwamatsu A, Takanezawa Y (2001)
Oxidized forms of peroxiredoxins and DJ-1 on two-dimensional gels increased in response
to sublethal levels of paraquat. Free Radic. Res. 35, 301–310.

Miyatake M, Miyagawa K, Mizuo K, Narita M, Suzuki T (2006) Dynamic changes in
dopaminergic neurotransmission induced by a low concentration of bisphenol-A in
neurones and astrocytes. J Neuroendocrinol 18: 434-44.

Moore DJ, West AB, Dawson VL, Dawson TM (2005a) Molecular Pathophysiology of
Parkinson’s Disease. Annu Rev Neurosci 28: 57-87.

Moore DJ, Zhang Li, Troncoso J, Lee MK, Hattori N, Mizuno Y, Dawson TM, Dawson VL
(2005b) Association of DJ-1 and parkin mediated by pathogenic DJ-1 mutations and
oxidative stress. Human Mol Genetics 14: 71–84.

Neumann M, Müller V, Görner K, Kretzschmar HA, Haass C, Kahle PJ (2004) Pathological
properties of the Parkinson’s disease-associated protein DJ-1 in a-synucleinopathies and
tauopathies: relevance for multiple system atrophy and Pick’s disease. Acta Neuropathol
107: 489–496.

Obata T (2002) Environmental estrogen-like chemicals and hydroxyl radicals induced by
MPTP in the striatum: a review. Neurochem Res 27: 423-431.

Olzmann JA, Brown K, Wilkinson KD, Rees HD, Huai Q, Ke H, Levey AI, Li L, Chin L-S
(2004) Familial Parkinson’s Disease-associated L166P Mutation Disrupts DJ-1 Protein
Folding and Function. JBC 279: 8506-8515.

Ooe H, Taira T, Iguchi-Ariga SMM, Arig H (2005) Induction of Reactive Oxygen Species
by Bisphenol A and Abrogation of Bisphenol A-Induced Cell Injury by DJ-1. Toxicological
Sciences 88: 114-126.

Przedborski S, Vila M (2001) MPTP: a review of its mechanisms of neurotoxicity. Clinical
Neuroscience Res 1: 407-418.

Riedel M, Goldbaum O, Uryu K, Bruce J, Trojanowski JQ, Lee VMY, Richter-Landsberg C
(2007) Alpha-synuclein aggregate formation in oligodendroglia OLN-t40 cells stably
transfected with alpha-synuclein. BMC Neuroscience 8 (Suppl 1): 13.

Semchuk KM, Love EJ, Lee RG (1992) Parkinson’s disease and exposure to agricultural
work and pesticide chemicals. Neurology 42: 1328–1335.

Sherer TB, Betarbet R, Testa CM, Seo BB, Richardson JR, Kim JH, Miller GW, Yagi T,
Matsuno-Yagi A, Greenamyre JT (2003) Mechanism of toxicity in rotenone models of
Parkinson's disease. J Neurosci 23: 10756-10764.

Spillantini MG, Schmidt ML, Lee VM, Trojanowski JQ, Jakes R, Goedert M (1997) Alpha-
synuclein in Lewy bodies. Nature 388: 839-840.

20
Sun F, Kanthasamy A, Anantharam V, Kanthasamy AG (2007) Environmental neurotoxic
chemicals-induced ubiquitin proteasome system dysfunction in the pathogenesis and
progression of Parkinson's disease. Pharmacology & Therapeutics 114:327–344.

Taira T, Saito Y, Niki T, Iguchi-Ariga SMM, Takahashi K, Ariga H (2004) DJ-1 has a role in
antioxidative stress to prevent cell death. EMBO Reports 5: 213–218.

Wenning G, Colosimo C, Geser F, Poewe W (2004) Multiple System Atrophy. The Lancet
Neurology 3: 93-103.

Wenning GK, Jellinger KA (2005) The role of α-synuclein in the pathogenesis of multiple
system atrophy. Acta Neuropathologica 109: 129-140.

Zhang L, Shimoji M, Thomas B, Moore DJ, Yu S-W, Marupudi NI, Torp R, Torgner IA,
Ottersen OP, Dawson TM, Dawson VL (2005) Mitochondrial localization of the Parkinson’s
disease related protein DJ-1: implications for pathogenesis. Human Mol Genetics 14:
2063–2073.

Zhou W. and Freed C. R. (2005) DJ-1 up-regulates glutathione synthesis during oxidative
stress and inhibits A53T alpha-synuclein toxicity. J Biol Chem 280: 43150-43158.

Zhou W, Zhu M, Wilson MA, Petsko GA, Fink AL (2006) The oxidation state of DJ-1
regulates its chaperone activity toward alpha-synuclein. J Mol Biol 356: 1036-1048.

21
Figures and Tables

Table 1: Case Details
Case Sex Age PMD* pH Cause of Death Diagnosis
P1 F 53 48.30 n/a Steady decline IPD**
P2 M 68 15:35 6,5 Steady decline IPD
P3 F 66 55:30 6,3 IPD IPD
P4 M 77 44:30 7,2 Gradual decline IPD
P5 M 75 44:15 6,7 Bronchopneumonia IPD
P6 F 71 50:10 6,4 Aspiration Pneumonia IPD
P7 F 78 37:15 6,5 Gradual decline IPD
P8 F 73 18:00 6,1 PSP** IPD, PSP, AD pathology**
P9 F 83 43:00 n/a Septic shock Incidental LPD, mild CVD**
P10 F 79 27:35 6,4 MSA, sudden death MSA**
P11 M 69 14:30 6,5 Bronchopneumonia MSA
P12 F 57 20:00 6,3 Overdose & asphyxiation MSA
P13 M 70 46:15 6,5 Bronchopneumonia MSA
P14 M 64 67:20 6,4 Bronchopneumonia MSA
P15 M 72 49:45 6,5 MSA, Chest Infection MSA
P16 F 60 66:25 5,7 Cerebellar Ataxia MSA
P17 M 64 39:10 n/a Pneumonia MSA
C1 M 86 53:00 6,7 Bronchopneumonia, heart failure
C2 M 83 117:05 6,8 Heart attack
C3 F 84 81:45 6,3 Cancer, Heart failure
C4 M 91 48:00 6,5 Bronchopneumonia
C5 M 81 40:00 6,5 n/a
C6 F 83 20:00 6,6 Bowel resection w/ complications
C7 F 88 49:25 6,2 Chronic obstructive airway disease
C8 M 81 42:00 6.5 Chronic obstructive airway disease
*PMD - post-mortem delay **IPD - idiopathic Parkinson’s disease, MSA - Multiple System atrophy, PSP -
Progressive Supranuclear Palsy, AD - Alzheimer’s disease, LPD - lymphoproliferative disease, n/a - not
available

22

23

24

25

26













Fig. 7(a) Rotenone treated samples show a shift in isoelectric point (pI), lower protein levels and less
abundancy of more basic isoforms (right side of panel) as compared to controls. (b) H2O2 treated samples
display an acidic shift at 200 µM but not at 100 µM concentration. All bands observed were at 20kD, as
expected from DJ-1.

27
Appendix: Merits and Limitations of the Techniques Employed in this Study

Choice of Sample
Part of our experiments were done using human brain tissue as sample. It is obviously an
advantage that the data thus obtained is more applicable to human disease, as compared
to in vitro or animal models of disease. However, the supply of human brain tissue is
restricted, and this resulted in a very small sample size in our studies, thus limiting the
significance of our findings. In addition, it is unclear how individual post-mortem delay
and pH differences affect protein expression levels and protein pI isoforms expressed.


Western Blot
Western blotting (WB) was the technique most frequently utilised in the present study.
Given an antibody specific to the protein in question, one can detect the molecular
weight and expression levels of said protein in tissue homogenates as well as cell
samples with relative ease and precision. However, there are several issues one needs to
consider.
Starting with sample preparation, it is important to note that the cells/tissue to be
investigated might not be completely soluble during homogenisation, and insoluble
protein may be lost in the cell debris. Depending on the amount of homogenisation buffer
needed, the resulting sample solution may be very dilute, making it more difficult to load
WB gel wells with an appropriate amount of protein without exceeding their volume
capacity.
WB protocols need to be optimised to achieve a certain standard of the data obtained.
Optimal incubation times, antibody dilutions and exposure times may differ from
experiment to experiment even with the same protein, as the protein concentration of
the sample is only an estimate and while BCA protein assays provide accurate intra-
sample values, there may be inter-sample differences from assay to assay. Thus an

28
antibody dilution that worked well with blot X, might not be enough or too much for blot
Y. This issue also complicates quantitative comparisons between blots.
Furthermore, the relatively high sensitivity of WBs can become a disadvantage, since
even small pipetting errors when preparing and loading the sample can result in falsified
data. A way to circumvent this problem is to normalise band densities of the protein of
interest against a ubiquitous protein eg actin. Probing for actin on a duplicate blot
however, does not account for the loading error. It is more convenient to strip and
reprobe the original blot with another antibody than to run duplicate blots. However,
stripping a blot does not only remove the unwanted antibodies but usually also results in
some of the proteins bound to the membrane being lost.


2D Gel Electrophoresis
2D gel electrophoresis is a relatively simple technique for separating proteins in a sample
according to both molecular weight and pI. In our case, we used immobilized pH gradient
(IPG) strips, which have the advantage of ease of handling and a pH gradient that is
fixed, thus allowing for loading of large protein samples. However, membrane or
hydrophobic proteins cannot be represented in the 2nd dimension well, probably owing to
certain protein/gel interacting during isoelectric focusing (Görg et al 2000). Since DJ-1 is
a soluble protein we did not experience any problems of this kind. Our main limitation
was the maximum sample size, since some of our original protein concentrations were
very dilute and we could only load half of the recommended amount of protein in some
cases. Furthermore, quantification of IPG 2D gel electrophoresis blots is difficult, since
bands often tend not to be well segregated, thus we had to rely on visual inspection of
the blots obtained.
Apart from these issues, the problems applying to WB are also applicable to 2D gel
electrophoresis, since resolution of the 2nd dimension is in essence a WB.

29
Co-Immunoprecipitation
Co-Immunoprecipitation allows the detection of protein-protein interactions, even in the
absence of knowledge regarding their function. Essentially, all that is needed are
antibodies specific to the proteins in question. Polyclonal antibodies are to be preferred to
monoclonal antibodies where ever these are available, since polyclonal antibodies are
raised against a larger number of epitopes on a given protein, and are thus more likely to
detect protein whose epitopes have been partially masked by the interaction eg due to
conformational changes during binding to the other protein. In our case the nature of the
proteins in question was known, i.e. DJ-1 and alpha-synuclein. The main advantage of
co-immunoprecipitation is that the protein interaction studied will be endogenous, and
not artifically altered as can be the case in other quantitative proteomics methods
(reviewed in Berggård et al 2007).


Immunohistochemistry
Immunohistochemistry allows for the visualisation of protein distribution and localisation
in a given tissue section, based on antigen recognition by an antibody specific to the
protein under investigation. The antibody-antigen interaction can be visualised directly by
a labeled antibody, that is for example, conjugated to a fluorophore. This has the
advantage of being a rapid staining method. Furthermore, direct staining eg with
fluorophores allows to visualise more than one protein at the same time, thus allowing
for co-localisation studies. However, it requires a custom labelled antibody to be
synthesised for every protein of interest (Ramos-Vara 2005). In our study, we used an
indirect method of staining. An unlabeled primary antibody was used first to react with
the protein in question, followed by a biotinylated secondary antibody that was coupled
with streptavidin-horseradish peroxidase and visualised by applying diaminobenzidine
(DAB). This type of visualisation provides greater sensitivity than direct staining, due to
signal amplification by the secondary antibody. Staining with DAB also has the advantage
that after fixation and mounting, sections can be easily stored for several years after the

30
experiment, since the signal does not fade. Fluorescent staining on the other hand
renders the section sensitive to light and fades relatively quickly. However, one
disadvantage of DAB staining is that, especially in the case of DJ-1, a lot of background
staining is observed, which can make localisation of relevant stained structures difficult
for the untrained eye. Furthermore, only one protein can be visualised at any one time.


References

Berggård T, Linse S, James P (2007) Methods for the detection and analysis of
protein–protein interactions. Proteomics 7. early view. (DOI 10.1002/pmic.200700131)

Görg A, Obermaier C, Boguth G, Harder A, Scheibe B, Wildgruber R, Weiss W (2000) The
current state of two-dimensional electrophoresis with immobilized pH gradients.
Electrophoresis 21: 1037-1053.

Ramos-Vara JA (2005) Technical Aspects of Immunohistochemistry. Vet Pathol 42: 405-
426.