Current Biology, Vol. 15, 1578–1582, September 6, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.cub.2005.07.036
Roles of Drosophila DJ-1 in Survival
of Dopaminergic Neurons and Oxidative Stress
Fiona M. Menzies,1Sarat C. Yenisetti,1
and Kyung-Tai Min*
Building 35, Room 2A1002
National Institutes of Health
Bethesda, Maryland 20892
The loss of dopaminergic neurons in the substantia
nigra is the pathological hallmark of Parkinson’s dis-
ease (PD). While the etiology of sporadic PD remains
elusive, an inherited form of early-onset familial PD
is linked to mutations of DJ-1 . To understand the
biological function of DJ-1 and its relevance to the
pathogenesis of PD, we investigated the function of
DJ-1 using Drosophila. Drosophila possesses two ho-
mologs of human DJ-1: DJ-1? and DJ-1?. We found
that DJ-1a is expressed predominantly in the testis,
while DJ-1b is ubiquitously present in most tissues,
resembling the expression pattern of human DJ-1.
Loss-of-function DJ-1b mutants demonstrated an ex-
tended survival of dopaminergic neurons and resis-
tance to paraquat stress, but showed acute sensitiv-
ity to hydrogen peroxide treatment. We showed a
compensatory upregulation of DJ-1a expression in
the brain of the DJ-1b mutant and demonstrated that
overexpression of DJ-1a in dopaminergic neurons is
sufficient to confer protection against paraquat insult.
These results suggest that Drosophila homologs of
DJ-1 play critical roles in the survival of dopaminergic
neurons and response to oxidative stress.
Results and Discussion
DJ-1a and DJ-1b Show Distinct Temporal
and Spatial Expression Patterns
Two Drosophila genes, DJ-1b (CG1349) and CG6646,
share significant homology with human DJ-1 (both
show 69% similarity to human; see Figure S1 in the
Supplemental Data available with this article online).
We will subsequently refer to CG6646 as DJ-1a. The
existence of multiple Drosophila homologs for a single
human gene is rare, and therefore we sought to estab-
lish the relationship between DJ-1a and DJ-1b in terms
of their expression. Developmental expression patterns
for both genes were determined by quantitative real-
time PCR. RNA samples were collected from defined
developmental stages: 2 hr embryos (at which time only
maternally contributed RNA is present), 2–16 hr em-
bryos, 1–3 instar larvae, late pupae, and adult flies (10
days old). The two genes showed distinct develop-
1These authors contributed equally to this work.
mental expression (Figures 1A and 1B). DJ-1a was ex-
pressed at high levels only in the later stages of devel-
opment, i.e., pupae and adults. DJ-1b did not show
such large changes in expression levels; it is maternally
contributed, after which its levels fall slightly in the
embryo but increase again through development. We
also established the spatial expression of DJ-1a and b
in adults. Comparing DJ-1a expression levels in male
and female heads and bodies showed that DJ-1a
mRNA expression was greatly increased in male bod-
ies, strongly suggesting localization of the DJ-1a tran-
script in the testis. Fly bodies were therefore dissected
to remove the reproductive organs (either testis or
ovary). DJ-1a transcripts were presented mainly in the
testis (Figure 1E). However, DJ-1b showed relatively
constant expression levels throughout heads and bod-
ies (Figure 1D).
Loss of DJ-1b Expression Results in Increased
Survival of Dopaminergic Neurons
Since the expression pattern of Drosophila DJ-1b re-
sembles human DJ-1, which is widely expressed in
most tissues , we generated mutant flies lacking ex-
pression of DJ-1β. Using P element local hopping from
the EP line, EP3700, which contains a single P element
insertion located 700 bp downstream of DJ-1b, we iso-
lated two mutant lines in which DJ-1β expression was
disrupted due to the insertion of the P element into the
gene. Loss of protein expression was confirmed by
Western blotting using an antibody generated to a pep-
tide region of DJ-1β that shows no homology to DJ-1α
(Figures 1F and 1G).
As the major pathological hallmark of PD is the loss
of dopaminergic neurons from the substantia nigra, we
investigated the effect of loss of DJ-1b expression on
dopaminergic neurons in Drosophila. Total dopamin-
ergic neuron number was counted in whole brains fol-
lowing staining with an antibody to tyrosine hydrox-
ylase (TH), an enzyme in the dopamine synthesis
pathway, which is a marker for dopaminergic neurons
(Figure 2). In DJ-1b mutant flies (aged 20 or 40 days at
25°C), no difference was seen in the total number of
dopaminergic neurons compared with control. This is
in contrast to transgenic flies overexpressing human
α-synuclein where a specific loss of dorsomedial dopa-
minergic neurons was seen as early as 10 days of age
[2, 3]. An age-related decrease in the number of TH-
positive neurons was seen in normal flies as previously
reported . The intensity of immunoreactivity against
TH of the remaining dopaminergic neurons was also
reduced (Figure 2). Surprisingly, we found that dopa-
minergic neurons were mostly sustained in the older
DJ-1b mutant flies, suggesting that the absence of DJ-
1β extended the survival of the dopaminergic neurons.
Immunoreactivity of these cells was also decreased to
a lesser extent.
showed selective regional loss of dopaminergic neu-
rons [2, 3], we examined whether the DJ-1b mutants
also show protection of dopaminergic cells in specific
Roles of Drosophila DJ-1
Figure 1. Expression Pattern of Drosophila
DJ-1a and DJ-1b and Generation of the DJ-
(A) DJ-1a mRNA transcript is most abundant
in later developmental stages.
(B) DJ-1b mRNA is maternally contributed
and shows less variation throughout devel-
(C) DJ-1a is present at highest levels in
(D) DJ-1b is present in head and body tissue
at similar levels.
(E) DJ-1a is greatly concentrated in the tes-
tis. Expression levels are shown as fold
change relative to expression in embryos (A,
B), male body (C, D), and testis (E). Quantita-
tive RT-PCR results were obtained from
cDNA isolated from defined developmental
stages, and values were normalized to the
level of Actin5C, which remain constant
throughout development. Values represent
mean ± SEM. Three independent experi-
ments were done in triplicate.
(F) Two alleles of the DJ-1b mutants were
isolated using P element local hopping. A P
element is inserted in DJ-1baat base pair
178 and in DJ-1bbat base pair 739 of the DJ-
(G) Western blot analysis shows no DJ-1b
protein in the DJ-1b mutants. Tubulin was
used as loading control.
areas of the brain. We found that dopaminergic cell
clusters in PPL1, PPM1/2, and PPL2 regions survived
better compared to other regions (Figure 3).
To determine if this protective effect is universal or
selective for dopaminergic neurons, we investigated
survival of serotoninergic neurons (Figure 3D). The
number of serotoninergic neurons in normal and DJ-1b
mutant flies at 60 days of age was not changed com-
pared to that of 10-day-old flies, although immunoreac-
tivity against serotonin was reduced in both flies. This
suggests that the effect of DJ-1β loss is not general,
but specific to dopaminergic neurons.
Next, we investigated whether the protection of these
neurons improved motor function in the mutant flies
using negative geotaxis assays. Although 60-day-old
mutant flies have more TH-positive neurons, they dis-
played sedentary behavior similar to normal flies of the
same age (data not shown). This implies that extended
survival of specific dopaminergic neurons does not in-
fluence locomotor vigor in old flies.
Loss of DJ-1? Expression Protects Flies
against Paraquat Insult
A role for DJ-1 in oxidative stress response has been
proposed using cell-culture systems [5–8]. We investi-
gated the effect of paraquat stress on DJ-1b mutant
flies. Flies raised for 5 days from eclosion at 25°C were
administered 20 mM paraquat in 5% sucrose (Figure
4A). Notably, the mutant flies showed significant resis-
tance to paraquat treatment, while control flies exhib-
ited acute sensitivity to paraquat, suggesting that the
DJ-1b mutants are less vulnerable to oxidative insult.
To ensure that changes in sensitivity to paraquat did
not result from altered feeding behavior, we measured
the level of radiolabeled sucrose consumption in flies
used in the oxidative stress test, and no changes in
consumption were found (Figure S2).
Decreased Sensitivity to Paraquat in the DJ-1b
Mutants Is Due to a Compensatory Upregulation
in DJ-1a in the Brain
Given that DJ-1β shares significant homology with DJ-
1α, it is plausible that DJ-1α may play a role in the phe-
notypes found in the DJ-1b mutant. We therefore inves-
tigated the possibility of a compensatory upregulation
of DJ-1a in the DJ-1b mutant. The levels of DJ-1a
mRNA transcripts, as measured by quantitative real-
time PCR, were doubled in the mutant brains (Figure
4B), but not in the whole flies (data not shown). To con-
firm whether this upregulation of DJ-1a is indeed rele-
Figure 2. Tyrosine Hydroxylase Immuno-
staining of the Fly Brain
Tyrosine hydroxylase (TH) immunostaining
was used to identify dopaminergic neurons
(DA) in the brain. Individual dompaminergic
neurons can be clearly seen in whole-mount
Drosophila brains. Immunoreactivity to TH is
decreased between 20 days and 60 days of
age in control flies, but not in the DJ-1b mu-
tant. Images shown were collected by con-
focal microscopy using a Zeiss LSM 520
Figure 3. The Number of DA in the DJ-1b Mu-
tant Is Not Reduced as the Fly Ages
(A) DA was counted in whole-mount Dro-
sophila brains. Total number of TH-positive
neurons is reduced at 60 days in control flies
(2+p = 0.0028, 40-day-old versus 60-day-
old). The DJ-1b mutant shows survival of DA
neurons and little change in immunoreactiv-
ity (*p = 0.01, **p = 0.0009 versus control 60-
(B) DA can be classified by their localization
in readily identifiable groups.
(C) Cells in PPL1, PPM1/2, and PPL2 clus-
ters show extended survival in the DJ-1b
mutant (*p < 0.05; **p < 0.01; ***p < 0.001).
(D) Staining of whole-mount brains with anti-
serotonin antibodies reveals no age-depen-
dant decrease in serotoninergic neurons
(SN) in the brains of control and the DJ-1b
mutant. Six brains were used for DA or SN
counting at each stage. Values represent
mean ± SEM.
Roles of Drosophila DJ-1
Figure 4. Sensitivity of Drosophila DJ-1 Mu-
tants to Oxidative Stress
(A) The DJ-1b mutants are resistant to para-
quat insult (p < 1 × 10−4for either DJ-1b mu-
tant versus either control).
(B) The DJ-1a mRNA transcript levels are
doubled in the brain of DJ-1b mutants (*p =
0.027, **p = 0.001).
(C) DJ-1a is overexpressed in dopaminergic
neurons using the TH-GAL4 driver (*p =
(D) Overexpression of DJ-1a in dopaminergic
neurons is sufficient to protect flies against
paraquat treatment (p < 1 × 10−4for flies
overexpressing DJ-1a versus each control).
(E) The DJ-1b mutants exhibit acute sensitiv-
ity to 1% H2O2treatment (p < 1 × 10−4for
DJ-1baor DJ-1bbversus each control).
(F) Overexpression of DJ-1a in dopaminergic
neurons is not sufficient to protect flies
against H2O2 treatment. Quantitative RT-
PCR was repeated three times in triplicate,
and statistical analysis was carried out using
Student’s t test. Values represent mean ±
SEM. Oxidative stress tests were performed
at least three times with an average of 80
male flies each time, and statistical analysis
was performed using Kaplan-Meier analysis
with log rank post-test.
vant to the altered response of the DJ-1b mutant to
oxidative stress, particularly in dopaminergic neurons,
we created transgenic flies overexpressing DJ-1a in
dopaminergic neurons using the binary UAS/GAL4 sys-
tem. Flies containing the DJ-1a transgene were crossed
with the TH-GAL4 (TH) driver lines  to yield DJ-1a
overexpression specifically in dopaminergic neurons.
Figure 4C shows that the DJ-1a transcript level was
elevated about 3.5-fold in the brain of transgenic flies
containing the driver. We then measured the survival of
transgenic flies following paraquat treatment (Figure
4D). TH-GAL4/UAS-DJ-1a flies showed significant re-
sistance to paraquat toxicity similar to that of the DJ-
1b mutants, suggesting that overexpression of DJ-1a in
dopaminergic neurons alone can preserve the animal
against oxidative stress.
DJ-1b Mutant Flies Show an Increase in Sensitivity
to Hydrogen Peroxide Stress
To investigate whether DJ-1b mutant flies are also in-
sensitive to other oxidative stress condition, we exam-
ined the effect of hydrogen peroxide on these flies.
Flies were treated with H2O2and their survival deter-
mined as for paraquat exposure. The DJ-1b mutant flies
were extremely sensitive to H2O2(Figure 4E). These
data are consistent with cell-culture experiments dem-
onstrating DJ-1-deficient cells to be highly sensitive to
oxidative stress . We also measured the survival of
TH-GAL4/UAS-DJ-1a flies against H2O2treatment and
found no effect of DJ-1a overexpression (Figure 4F).
These results suggest that DJ-1β plays a critical role in
H2O2-induced oxidative stress response.
Notably, the DJ-1b mutant flies exhibit difference in
response to different forms of oxidative stress. The flies
show less sensitivity to paraquat but increased sensi-
tivity to H2O2. The upregulation of DJ-1a is therefore
insufficient to protect against H2O2, unlike paraquat.
This suggests that DJ-1α is crucial for protecting cells
against paraquat-mediated stress, while DJ-1β plays a
role in responding to H2O2-induced stress. The exact
mechanism by which paraquat results in the production
of reactive oxygen species is unknown, although the
major species produced is superoxide. The two homo-
logs of DJ-1 in Drosophila may therefore be required
to protect against different species of reactive oxygen.
Evidence is also emerging from mouse models to sup-
port the possibility of selective response of DJ-1 to oxi-
dative stress as DJ-1-deficient mice demonstrate an in-
creased sensitivity to MPTP , but no such increase
is seen in response to paraquat . These results may
Current Biology Download full-text
provide new insights into the mechanism associated
with DJ-1 in PD.
While mammalian DJ-1 apparently plays multiple
roles, such as sensing oxidative levels, chaperone ac-
tivity, and fertility, in Drosophila these roles may be di-
vided between the two homologs. It remains to be
determined whether DJ-1β is required for the mainte-
nance of dopaminergic neurons in the absence of
DJ-1α upregulation and whether DJ-1α has a role in
dopaminergic neurons under normal circumstances. In
normal fly brains, the level of DJ-1a transcripts ap-
peared to be low, perhaps suggesting that the major
role of DJ-1α is in the testis. However, it is also possible
that DJ-1α is expressed at significant levels in specific
regions of the brain where it may also have an impor-
An accompanying paper by Meulener et al. in this issue
 shows that mutant flies lacking both DJ-1a and b
are sensitive to paraquat stress, while the DJ-1b mu-
tants in this report are resistant to the stress. This may
offer an interesting insight into the nature of the interac-
tion between the DJ-1 homologs. The absence of a de-
crease in sensitivity to paraquat in the DJ-1b mutants
created by Meulener et al. suggests that no compensa-
tory upregulation in DJ-1a occurs in these flies, in con-
trast to that seen in our mutant lines. A major difference
between the mutant lines described is that those of
Meulener et al. are created by a deletion, which also
removes adjacent genomic DNA regions that are still
present in the mutants generated in our lab. Within
these DNA regions, it is likely that regulatory elements
exist. A recent report by Spilianakis et al.  showed
that regulatory regions of related genes on different
chromosomes are important in the coordinate regula-
tion of multiple genes. It is therefore possible that the
two Drosophila DJ-1 homologs are also coordinately
regulated. A second possibility for the variation be-
tween the two mutant lines may be that the P element
insertion lines created in our lab could express a trun-
cated version of DJ-1β. No evidence of such a product
was observed by Western blotting; however, the anti-
body generated for this study recognizes a region close
to the C terminus of DJ-1β, so its existence cannot be
ruled out. It is possible that a truncated protein product
could result in the compensatory upregulation of the
DJ-1a gene we observe. Further work will help to eluci-
date the exact role of DJ-1 homologs in paraquat resis-
Supplemental Data include three figures and Supplemental Experi-
mental Procedures and can be found with this article online at
This work was supported by an intramural funding from NINDS,
NIH to K.-T.M.
Received: May 2, 2005
Revised: July 12, 2005
Accepted: July 12, 2005
Published: September 6, 2005
1. 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,
2. Feany, M.B., and Bender, W.W. (2000). A Drosophila model of
Parkinson’s disease. Nature 404, 394–398.
3. Auluck, P.K., Chan, H.Y., Trojanowski, J.Q., Lee, V.M., and Bon-
ini, N.M. (2002). Chaperone suppression of alpha-synuclein
toxicity in a Drosophila model for Parkinson’s disease. Science
4. Neckameyer, W.S., Woodrome, S., Holt, B., and Mayer, A.
(2000). Dopamine and senescence in Drosophila melanogaster.
Neurobiol. Aging 21, 145–152.
5. Mitsumoto, A., Nakagawa, Y., Takeuchi, A., Okawa, K., Iwa-
matsu, A., and Takanezawa, Y. (2001). Oxidized forms of per-
oxiredoxins and DJ-1 on two-dimensional gels increased in re-
sponse to sublethal levels of paraquat. Free Radic. Res. 35,
6. Canet-Aviles, R.M., Wilson, M.A., Miller, D.W., Ahmad, R.,
McLendon, C., Bandyopadhyay, S., Baptista, M.J., Ringe, D.,
Petsko, G.A., and Cookson, M.R. (2004). The Parkinson’s dis-
ease protein DJ-1 is neuroprotective due to cysteine-sulfinic
acid-driven mitochondrial localization. Proc. Natl. Acad. Sci.
USA 101, 9103–9108.
7. Taira, T., Saito, Y., Niki, T., Iguchi-Ariga, S.M., Takahashi, K., and
Ariga, H. (2004). DJ-1 has a role in antioxidative stress to pre-
vent cell death. EMBO Rep. 5, 213–218.
8. Martinat, C., Shendelman, S., Jonason, A., Leete, T., Beal, M.F.,
Yang, L., Floss, T., and Abeliovich, A. (2004). Sensitivity to oxi-
dative stress in DJ-1-deficient dopamine neurons: an ES-
derived cell model of primary parkinsonism. PLoS Biol. 2, e327.
9. Friggi-Grelin, F., Iche, M., and Birman, S. (2003). Tissue-specific
developmental requirements of Drosophila tyrosine hydrox-
ylase isoforms. Genesis 35, 260–269.
10. Kim, R.H., Smith, P.D., Aleyasin, H., Hayley, S., Mount, M.P.,
Pownall, S., Wakeham, A., You-Ten, A.J., Kalia, S.K., Horne, P.,
et al. (2005). Hypersensitivity of DJ-1-deficient mice to 1-methyl-
stress. Proc. Natl. Acad. Sci. USA 102, 5215–5220.
11. Goldberg, M.S., Pisani, A., Haburcak, M., Vortherms, T.A., Ki-
tada, T., Costa, C., Tong, Y., Martella, G., Tscherter, A., Martins,
A., et al. (2005). Nigrostriatal dopaminergic deficits and hypo-
kinesia caused by inactivation of the familial Parkinsonism-
linked gene DJ-1. Neuron 45, 489–496.
12. Meulener, M., Whitworth, A.J., Armstrong-Gold, C.E., Rizzu, P.,
Heutink, P., Wes, P.D., Pallanck, L.J., and Bonini, N.M. (2005).
Drosophila DJ-1 mutants are selectively sensitive to environ-
mental toxins associated with Parkinson’s disease. Curr. Bio.
15, this issue, 1572–1577.
13. Spilianakis, C.G., Lalioti, M.D., Town, T., Lee, G.R., and Flavell,
R.A. (2005). Interchromosomal associations between alterna-
tively expressed loci. Nature 435, 637–645.