© 2006 Nature Publishing Group
Drosophila pink1 is required for mitochondrial
function and interacts genetically with parkin
Ira E. Clark1*, Mark W. Dodson1*, Changan Jiang1*, Joseph H. Cao1, Jun R. Huh2, Jae Hong Seol3, Soon Ji Yoo4,
Bruce A. Hay2& Ming Guo1
Parkinson’s disease is the second most common neurodegenera-
tive disorder and is characterized by the degeneration of dopa-
minergic neurons in the substantia nigra. Mitochondrial
dysfunction has been implicated as an important trigger for
Parkinson’s disease-like pathogenesis because exposure to
environmental mitochondrial toxins leads to Parkinson’s dis-
ease-like pathology1. Recently, multiple genes mediating familial
forms of Parkinson’s disease have been identified, including
PTEN-induced kinase 1 (PINK1; PARK6) and parkin (PARK2),
which are also associated with sporadic forms of Parkinson’s
disease2–6. PINK1 encodes a putative serine/threonine kinase
have been reported for pink1 in any model system. Here we show
that removal of Drosophila PINK1 homologue (CG4523; hereafter
called pink1) function results in male sterility, apoptotic muscle
sensitivity to multiple stresses including oxidative stress. Pink1
localizes to mitochondria, and mitochondrial cristae are frag-
mented in pink1 mutants. Expression of human PINK1 in the
morphology in a portion of pink1 mutants, demonstrating func-
Drosophila parkin shows phenotypes similar to loss of pink1
function7,8. Notably, overexpression of parkin rescues the male
sterility and mitochondrial morphologydefects of pink1 mutants,
whereas double mutants removing both pink1 and parkin func-
tion show muscle phenotypes identical to those observed in either
mutant alone. These observations suggest that pink1 and parkin
function, at least in part, in the same pathway, with pink1
functioning upstream of parkin. The role of the pink1–parkin
pathway in regulating mitochondrial function underscores the
importance of mitochondrial dysfunction as a central mechanism
of Parkinson’s disease pathogenesis.
Drosophila melanogaster contains a single PINK1 homologue
(CG4523). As with human PINK1, Drosophila Pink1 has a predicted
amino-terminal mitochondrial targeting sequence and a serine/
threonine kinase domain that shares 43% amino acid identity and
60% similarity with human PINK1 (Supplementary Fig. S1a). More
Parkinson’s disease patients1,4,6. Many of the residues altered by mis-
sense mutations4are conserved in Drosophila Pink1 (Supplementary
pink1 messenger RNA was detectable at all developmental stages
with the highest expression levels in the adult head and testes
(Supplementary Fig. S1b). We generated two chromosomal de-
letions; both showed very similar if not identical phenotypes and
are probably null alleles (Fig. 1a and Supplementary Fig. S1c). Both
pink1 deletion strains were viable; however, they were completely
male sterile and almost completely female sterile. Because proper
mitochondrial function is required for spermatogenesis, we exam-
ined the pink1 mutant testes for mitochondrial defects. During
spermatogenesis, stem-cell differentiation is followed by mitosis
and meiosis with incomplete cytokinesis, creating syncytial cysts of
64 spermatids9. Subsequently, mitochondria exhibit marked morpho-
logical changes. Specifically, early spermatids undergo mitochondrial
aggregation and fusion, creating two giant mitochondria that form a
spherical structure known as the nebenkern9. Under phase contrast
microscopy, such ‘onion stage’spermatids can be identified as having
two adjacent spherical structures: the nucleus and the nebenkern (Fig.
1b). During spermatid elongation, the nebenkern unfurls to yield two
nuclei appeared normal; however, nebenkerns were vacuolated
elongation (Fig. 1k).
Toanalysefurther thesemitochondrial defects,weexaminedpink1
mutant testes using transmission electron microscopy (TEM). After
elongation, spermatids undergo a process known as individualiza-
tion, in which the cytoplasmic bridges that link the 64 spermatids
within a cyst are broken and excess cytoplasm is extruded9. After
individualization, each spermatid consists largely of the central
axoneme, a microtubule-based structure required for motility, and
Spermatids in pink1 mutant testes underwent elongation (Fig. 1k);
however, numerous defects in individualization were observed
(Fig. 1o). Although the number of spermatids in each cyst, the
shape and the size of the axoneme, and the spatial relationship
between the axoneme and mitochondria were unchanged (Fig. 1o),
the overall architecture of pink1 mutant cysts was disorganized, and
the mitochondria were of variable size and were frequently smaller
than those in wild-type cysts (Fig. 1, compare panels o and n). pink1
material, which may represent aberrant mitochondria (Fig. 1o).
To ensure that the mitochondrial and individualization pheno-
types were indeed due to lack of pink1 function, we generated
multiple lines bearing a pink1 genomic rescue transgene (Fig. 1a).
pink15or pink19males carrying a single copy of this transgene were
fertile (98% fertile, n ¼ 40). Moreover, defects in mitochondrial
morphology and individualization were almost completely sup-
pressed (Fig. 1e, h, l, p). Thus, the male sterility phenotype is due
to lack of pink1 function.
To determine the subcellular localization of Pink1, we generated a
carboxy-terminal myc-tagged version of the pink1 genomic rescue
1Department of Neurology, Brain Research Institute, The David Geffen School of Medicine, University of California, Los Angeles, California 90095, USA.2Division of Biology,
California Institute of Technology, Pasadena, California 91125, USA.3Department of Biological Sciences, Seoul National University, Seoul 151-742, Korea.4Department of Biology,
Kyung Hee Institute of Age-related and Brains Disease, Kyung Hee University, Seoul 130-701, Korea.
*These authors contributed equally to this work.
Vol 441|29 June 2006|doi:10.1038/nature04779
© 2006 Nature Publishing Group
transgene. pink1 mutant males with the pink1-9myc transgene were
fertile (100%, n ¼ 30), suggesting that the Myc tag does not interfere
with Pink1 function. Immunofluorescence on pink1-9myc testes
demonstrated that Pink1-9Myc protein co-localized to nebenkerns
with manganese superoxide dismutase (Mn SOD), a mitochondrial
marker10(Fig. 1q–s). Together, these observations suggest that pink1
functions in spermatogenesis to regulate mitochondrial morphology
To determine whether pink1 has a more general role in mitochon-
drial function, we examined other tissues in pink1 mutants. We were
unable to detect a significant change in the number of dopaminergic
neurons between pink1 mutant and wild-type flies at 50days old
(Supplementary Fig. S2). However, striking phenotypes were
observed in muscle, which also has high energy demands requiring
robust mitochondrial function. pink1 mutants have ‘held-up’ wings
(Fig. 2e) and poor flight performance (Supplementary Fig. S3),
suggesting a defect in indirect flight muscles. In wild-type adults,
muscle fibres were well organized in parallel stripes with regular
age-matched 14-day-old pink1 mutants were disorganized with
prominent vacuoles (Fig. 2f). Ultrastructural TEM analysis showed
that the mitochondrial cristae were fragmented in pink1 mutants,
with some mitochondria appearing nearly hollow (Fig. 2g, h), as
compared to wild-type mitochondria, which were filled with densely
by the pink1 genomic rescue transgene (Fig. 2i–l and Supplementary
Fig. S3). Previous immuno-electron microscopy studies have shown
that mammalian PINK1 is localized to mitochondrial cristae11.
in spermatids. a, Genomic map of pink1 (cytological location 6C6).
P element insertion (triangle), pink1 coding and untranslated regions (dark
and shaded rectangles), and nearby genes (open rectangles) are depicted.
70% of the kinase domain, including motifs required for ATP binding,
catalysis and metal binding30. The pink19deletion removes the entire 5
untranslated region (UTR) and part of the kinase domain. The pink1
genomic rescue construct does not include the full coding region of nearby
genes. b–l, Schematics and phase contrast micrographs of mitochondrial
morphogenesis in spermatids during the ‘onion stage’ (b–h) and spermatid
elongation (i–l). In both stages, pink1 mutants show vacuolation of
single post-individualization cysts containing 64 spermatids. Each
spermatid contains an axoneme (orange arrow) and mitochondrial
derivative (red arrowhead) within an individual plasma membrane. The
pink1 mutant cyst (o) shows individualization defects, mitochondria of
variable sizes, and a mass of electron-dense material (yellow arrow).
q–s, Double labelling of pink1-9myc testes with anti-Myc (q) and anti-Mn
SOD (r) demonstrates that Pink1 is localized to the nebenkerns. Scale bars:
10mm (c–e, j–l), 4mm (f–h) and 1mm (n–p).
Figure 2 | pink1 mutants undergo apoptotic muscle degeneration and
fragmentationof mitochondrialcristae. pink1mutantshave held-upwings
(compare panels a and e). b–d, f–h, j–l, Toluidine blue staining (b, f, j) and
TEMs(c,d,g,h, k, l)of indirectflight musclefrom14-day-old flies. Muscles
of pink1 mutants show numerous vacuoles (compare f and b) and
mitochondriawith fragmented cristae(compareg,hwithc, d). m–r, At96h
after puparium formation, pink1 mutant muscle fibres and mitochondria
appear normal (compare p–r and m–o). pink15mutants show many
TUNEL-positive nuclei in indirect flight muscle (compare t and s). Each of
these mutant phenotypes is rescued by the pink1 genomic rescue transgene
(i–l, u). Scale bars: 1.0mm (c, g, k, n, q) and 0.5mm (d, h, l, o, r).
NATURE|Vol 441|29 June 2006
© 2006 Nature Publishing Group
Together, these observations suggest that mitochondrial cristae are a
site of Pink1 action, either directly or indirectly.
To investigate whether these muscle phenotypes are developmen-
tal or degenerative, we examined indirect flight muscles in pink1
mutants during development and shortly after eclosion. At 96h after
puparium formation, muscle and mitochondrial morphology of
pink1 mutants (Fig. 2p–r) was indistinguishable from that of age-
matched wild-type flies (Fig. 2m–o). At 1–2days after eclosion,
however, pink1 mutants already showed muscle degeneration and
fragmentation of mitochondrial cristae (see below), which was
completely suppressed by the pink1 genomic rescue transgene (data
not shown). Furthermore, indirect flight muscles from pink1
mutants 1–2days after eclosion demonstrated a marked increase in
TdT-mediated dUTP nick end labelling (TUNEL)-positive nuclei
(Fig. 2s, t). Taken together, these findings suggest that muscle
phenotypes of pink1 mutants are rapidly degenerative and apoptotic
in nature. Finally, ATP levels in pink1 mutants were significantly
reduced (Fig. 3f), demonstrating that mitochondrial function is also
pink1 mutants were fully suppressed by the pink1 genomic rescue
transgene (Figs 2u and 3f).
Mitochondrial dysfunction can lead to decreased resistance to
reactive oxygen species, which has been implicated in Parkinson’s
disease pathogenesis. Sensitivity to reactive oxygen species has also
been observed in flies lacking either parkin8or DJ-112–15, another
familial Parkinson’s disease gene16. To explore the role of pink1 in
resistance to oxidative stress, we analysed the survival of pink1
mutants after exposure to paraquat, a free radical inducer, or
rotenone, which impairs complex I activity in the mitochondrial
respiratory chain. Both agents can induce Parkinson’s disease-like
pathology in mammals17. To avoid variability due to genetic back-
ground, we included two control strains: a wild-type line with
excision), and a line of pink1 mutants with the genomic rescue
transgene. pink1 mutants showed 59% reduced resistance to para-
quat, and 71% reduced resistance to rotenone (Fig. 3a, b). pink1
mutants were also sensitive to a hyperoxic environment, a stress that
does not rely on feeding (data not shown). The stress sensitivity of
pink1 mutants was not limited to oxidative stress, however, as they
were also sensitive to dithiothreitol (DTT), a protein folding inhibi-
tor, and high concentrations of salt, an osmotic stress (Fig. 3c, d).
Finally, pink1 mutants were shorter lived. At 56days, only 12% of
pink1 mutants were alive, compared with 70–80% of wild-type flies
The sensitivity to multiple stresses in pink1 mutants may reflect a
generalized sickness or may be a consequence of mitochondrial
Drosophila Pink1 shares significant homology with human PINK1
(Supplementary Fig. S1a). To test functional conservation of the
human and Drosophila homologues, we investigated whether human
PINK1 could rescue fly pink1 mutant phenotypes. We expressed
human PINK1 using the testes-specific b2-tubulin promotor18
(TMR-human PINK1), which is expressed in developing spermatids
beginning just before the onion stage. Fertility was restored in 17%
(n ¼ 93) of pink1 mutant males carrying the TMR-human PINK1
and human Pink1 share at least some functional conservation.
parkin mutants show phenotypes similar to pink1 mutants: male
drial cristae and sensitivity to multiple stresses including oxidative
stress7,8. We explored whether pink1 and parkin function in the same
Figure 3 | pink1 mutants are sensitive to multiple stresses, and have
reduced lifespan and ATP levels. a–d, Survival of pink15mutants (pink),
wild type (precise excision, blue) and pink15mutants with the genomic
rescue transgene (red) after exposure to 20mM paraquat (a), 5mM
rotenone (b), 100mM DTT (c) and 500mM NaCl (d). Mean survival times
(in hours) for pink15, wild-type and pink15;P[pink1]/þ flies, respectively,
are: paraquat 29.0 ^ 6.8, 65.7 ^ 1.0 and 70.8 ^ 5.5; rotenone 33.6 ^ 11.6,
115 ^ 14.3 and 116.2 ^ 3.6; DTT 25.8 ^ 4, 92.4 ^ 12.6 and 84.4 ^ 9.8;
NaCl40.5 ^ 1.5,68.8 ^ 1.8and66.6 ^ 1.9.e,pink19mutantshavereduced
lifespan. f, Mean ATP levels (mmolmg21protein) of pink15mutants
(0.13 ^ 0.02)are significantly reduced, as compared withwild-type (precise
excision, 0.32 ^ 0.07; P , 0.02) and pink15flies with the rescue transgene
(0.25 ^ 0.05; P , 0.05). Error bars indicate standard deviation. Student’s t-
test was used.
Figure 4 | Fly pink1 is functionally conserved with human PINK1, and acts
(arrowheads) in pink15testes (a, d) is almost completely suppressed
(arrows) by the TMR-human PINK1 (b, e) and TMR-parkin (c, f)
transgenes. g, Overexpression of parkin in muscle suppresses the pink1
mutant phenotype; some mitochondria contain densely packed cristae
(arrows), whereas others (arrowhead) still contain fragmented cristae.
h–k, Toludine blue staining (h, i) and TEM (j, k) of 2-day-old adult indirect
flight muscles. Double mutants of pink1 and parkin show muscle (i) and
mitochondria (k) phenotypes identical to those of pink1 (h, j) or parkin
single mutants7,8(data not shown). Scale bars: 10mm (a–c), 4mm (d–f) and
1.0mm (g, j, k).
NATURE|Vol 441|29 June 2006
© 2006 Nature Publishing Group
in pink1 mutant testes could compensate for loss of pink1 function
and restore male fertility. If parkin positively regulates pink1, we
would expect that overexpression of parkin would still result in the
pink1 null phenotype and thus be sterile. Conversely, if parkin acts
downstream of pink1, overexpression of parkin might suppress the
pink1 null phenotype and restore fertility. parkin overexpression in
pink1 mutants resulted in significant suppression of sterility (62%
fertile, n ¼ 65) and mitochondrial phenotype (Fig. 4; compare
panels c and f with a and d). In contrast, overexpression of pink1
in the parkinD21null background8failed to suppress sterility due to
lack of parkin (0% fertile, n ¼ 60). As controls, TMR-pink1 and
TMR-parkin restored the fertility in pink1 and parkin mutants,
respectively (100% fertile, n ¼ 40; 97% fertile, n ¼ 40, respectively).
Second, when parkin was specifically expressed in pink1 mutant
muscle using the UAS/GAL4 system19, a subset of 1–2-day-old flies
contained mitochondriawith densely packed cristae (Fig. 4g), which
was never seen in age-matched pink1 mutants (Fig. 4j). When pink1
was specifically expressed in pink1 mutant muscle, normal mito-
rescue in mitochondrial morphology seen with parkin overexpression
in indirect flight muscles may reflect low expression levels of the
ment for only a subset of pink1 functions. In either case, these results
suggest that parkin acts downstream of pink1, although it is formally
possible that they act in parallel on shared targets.
To test further this hypothesis, we generated double mutants that
removed both pink1 and parkin function. If pink1 and parkin act in a
linear pathway, double mutants should show phenotypes similar to
those observed in either mutant alone. Conversely, if these genes
function in parallel pathways to regulate a common process, double
mutants would show phenotypes stronger than those of either
mutant alone. Double mutants removing both pink1 and parkin
function showed mitochondrial and muscle degeneration pheno-
types identical to those of either single mutant (Fig. 4h–k and data
not shown). Together, these experiments suggest that parkin acts, at
least in part, downstream of pink1. This conclusion is particularly
interesting in light of recent work demonstrating that clinical pres-
entations of Parkinson’s disease patients harbouring pink1 or parkin
mutations are indistinguishable6.
Although mild dopaminergic neuronal loss has been observed in
parkin mutants20, we failed to observe any dopaminergic neuronal
loss in pink1 mutants. Similarly, knockout mice for parkin and
DJ-1generated by multiple groups fail to show any dopaminergic
neuronal loss, although functional impairments of the nigrostriatal
system and mitochondrial dysfunction have been observed21,22. It is
possible that a more sensitive assay, such as mitochondrial mor-
phology or function, or dopaminergic neuronal physiology, may
reveal a defect in dopaminergic neurons in pink1 mutants. However,
it is important to note that Parkinson’s disease is a multi-system
disease affecting more than dopaminergic neurons. Degeneration of
neurons predates that of dopaminergic neurons23. Moreover, patho-
logical changes and defects in mitochondrial respiratory chain
function have been observed in muscle biopsies of Parkinson’s
disease. Our finding that pink1 acts upstream of parkin to regulate
mitochondria strengthens the accumulating evidence that parkin has
an important role in mitochondrial function, and underscores
mitochondrial dysfunction as a central mechanism for Parkinson’s
disease pathogenesis. The pink1–parkin pathway provides an entry
point for isolating other genes related to regulation of mitochondria,
which may include components that function in Parkinson’s disease
Molecular biology. To express genes specifically in testes during spermatid
differentiation, the pGMR vector27was modified to replace the glass-binding
sites with part of the regulatory region of b2-tubulin18, yielding pTMR
(b2-tubulin mediated expression).
Genetics and Drosophila strains. pink1 deletions were generated by imprecise
excision of the G900 P element obtained from GenExel, which is viable and
fertile. Breakpoints were mapped by genomic polymerase chain reaction (PCR)
was driven by 24B-GAL4 in the pink15mutant background.
Immunofluorescence and confocal microscopy. We used the following anti-
bodies: mouse anti-Myc (Upstate, 1:400), rabbit anti-Mn SOD (Stressgen,
1:300), mouse anti-tyrosine hydroxylase (Immunostar, 1:300).
Light and electron microscopy. For light and electron microscopic analyses,
testes were prepared as described previously (refs 9 and 28, respectively). For
muscle TEM, thoraces were fixed in paraformaldehyde/glutaraldehyde, post-
sections were stained with Toluidine blue, and 50–80-nm sections with uranyl
acetate and lead citrate. At least three testes or thoraces of each genotype were
examined by TEM.
aged for 48h, starved for 6h and subjected to 5% sucrose plus each agent. Three
vials of 30–40 flies were assayed simultaneously for each genotype. Flight assays
For longevity measurements, 100 males of each genotype were divided into five
ATP and TUNEL assays. For each genotype, ATP levels were determined from
lysates of three groups of five 2–4-day-old flies using the ATP bioluminescence
sections using the in situ cell death detection kit from Roche.
Received 15 February; accepted 7 April 2006.
Published online 30 April 2006.
1.Moore, D. J., West, A. B., Dawson, V. L. & Dawson, T. M. Molecular
pathophysiology of Parkinson’s disease. Annu. Rev. Neurosci. 28, 57– -87 (2005).
Valente, E. M. et al. Hereditary early-onset Parkinson’s disease caused by
mutations in PINK1. Science 304, 1158– -1160 (2004).
Kitada, T. et al. Mutations in the parkin gene cause autosomal recessive
juvenile parkinsonism. Nature 392, 605– -608 (1998).
Bonifati, V. et al. Early-onset parkinsonism associated with PINK1 mutations:
frequency, genotypes, and phenotypes. Neurology 65, 87– -95 (2005).
Klein, C. et al. PINK1, Parkin, and DJ-1 mutations in Italian patients with
early-onset parkinsonism. Eur. J. Hum. Genet. 13, 1086– -1093 (2005).
Ibanez, P. et al. Mutational analysis of the PINK1 gene in early-onset
parkinsonism in Europe and North Africa. Brain 129, 686– -694 (2006).
Greene, J. C. et al. Mitochondrial pathology and apoptotic muscle degeneration
in Drosophila parkin mutants. Proc. Natl Acad. Sci. USA 100, 4078– -4083 (2003).
Pesah, Y. et al. Drosophila parkin mutants have decreased mass and cell size
and increased sensitivity to oxygen radical stress. Development 131, 2183– -2194
Fuller, M. T. in The Development of Drosophila melanogaster 71– -147 (Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1993).
10. Van Steeg, H., Oudshoorn, P., Van Hell, B., Polman, J. E. & Grivell, L. A.
Targeting efficiency of a mitochondrial pre-sequence is dependent on the
passenger protein. EMBO J. 5, 3643– -3650 (1986).
11.Silvestri, L. et al. Mitochondrial import and enzymatic activity of PINK1 mutants
associated to recessive parkinsonism. Hum. Mol. Genet. 14, 3477– -3492 (2005).
12. Meulener, M. et al. Drosophila DJ-1 mutants are selectively sensitive to
environmental toxins associated with Parkinson’s disease. Curr. Biol. 15,
1572– -1577 (2005).
13. Menzies, F. M., Yenisetti, S. C. & Min, K. T. Roles of Drosophila DJ-1 in survival
of dopaminergic neurons and oxidative stress. Curr. Biol. 15, 1578– -1582 (2005).
14. Yang, Y. et al. Inactivation of Drosophila DJ-1 leads to impairments of oxidative
stress response and phosphatidylinositol 3-kinase/Akt signaling. Proc. Natl
Acad. Sci. USA 102, 13670– -13675 (2005).
15. Park, J. et al. Drosophila DJ-1 mutants show oxidative stress-sensitive
locomotive dysfunction. Gene 361, 133– -139 (2005).
16. Bonifati, V. et al. Mutations in the DJ-1 gene associated with autosomal
recessive early-onset Parkinsonism. Science 299, 256– -259 (2003).
17. Przedborski, S. & Ischiropoulos, H. Reactive oxygen and nitrogen species:
weapons of neuronal destruction in models of Parkinson’s disease. Antioxid.
Redox Signal. 7, 685– -693 (2005).
18. Huh, J. R. et al. Multiple apoptotic caspase cascades are required in
nonapoptotic roles for Drosophila spermatid individualization. PLoS Biol. 2, E15
19. Brand, A. H. & Perrimon, N. Targeted gene expression as a means of altering cell
fates and generating dominant phenotypes. Development 118, 401– -415 (1993).
20. Whitworth, A. J. et al. Increased glutathione S-transferase activity rescues
dopaminergic neuron loss in a Drosophila model of Parkinson’s disease. Proc.
Natl Acad. Sci. USA 102, 8024– -8029 (2005).
NATURE|Vol 441|29 June 2006
© 2006 Nature Publishing Group
21. Fleming, S. M., Fernagut, P. O. & Chesselet, M. F. Genetic mouse models of
parkinsonism: strengths and limitations. NeuroRx 2, 495– -503 (2005).
22. Palacino, J. J. et al. Mitochondrial dysfunction and oxidative damage in
parkin-deficient mice. J. Biol. Chem. 279, 18614– -18622 (2004).
23. Braak, H. et al. Staging of brain pathology related to sporadic Parkinson’s
disease. Neurobiol. Aging 24, 197– -211 (2003).
24. Ahlqvist, G., Landin, S. & Wroblewski, R. Ultrastructure of skeletal muscle in
patients with Parkinson’s disease and upper motor lesions. Lab. Invest. 32,
673– -679 (1975).
25. Bindoff, L. A., Birch-Machin, M. A., Cartlidge, N. E., Parker, W. D. Jr & Turnbull,
D. M. Respiratory chain abnormalities in skeletal muscle from patients with
Parkinson’s disease. J. Neurol. Sci. 104, 203– -208 (1991).
26. Wiedemann, F. R., Winkler, K., Lins, H., Wallesch, C. W. & Kunz, W. S. Detection
of respiratory chain defects in cultivated skin fibroblasts and skeletal muscle of
patients with Parkinson’s disease. Ann. NY Acad. Sci. 893, 426– -429 (1999).
27. Hay, B. A., Wolff, T. & Rubin, G. M. Expression of baculovirus P35 prevents cell
death in Drosophila. Development 120, 2121– -2129 (1994).
28. Tokuyasu, K. T., Peacock, W. J. & Hardy, R. W. Dynamics of spermiogenesis in
Drosophila melanogaster. I. Individualization process. Z. Zellforsch. Mikrosk. Anat.
124, 479– -506 (1972).
29. Benzer, S. Genetic dissection of behaviour. Sci. Am. 229, 24– -37 (1973).
30. Hanks, S. K. & Hunter, T. Protein kinases 6. The eukaryotic protein kinase
superfamily: kinase (catalytic) domain structure and classification. FASEB J. 9,
576– -596 (1995).
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank G. Mardon and L. Pallanck for parkin cDNA and
mutant flies; A. Simon, D. Walker, X. Zhan and A. Kiger for technical advice;
L. Zipursky, L. Toro and D. Krantz for access to equipment and space; Guo
laboratory members for discussions; and the EM core facilities at UCLA Brain
Research Institute and at Caltech. We are indebted to R. Young in Seymour
Benzer’s laboratory for assistance with EM, and F. Laski for his phase contrast
microscope. This work was supported by a National Institute of Health (NIH)
grant to B.A.H. and an Alfred P. Sloan Foundation Fellowship in Neuroscience
and a NIH grant to M.G.
Author Contributions I.E.C., M.W.D., C.J. and J.H.C. in the Guo laboratory
conceived and performed the experiments. J.R.H. and B.A.H. in the Hay
laboratory assisted with experiments involving TEM in testes and with TUNEL
staining; J.H.S. and S.J.Y. provided crucial reagents; and M.G. conceived and
performed experiments, supervised the work, and wrote the manuscript with
helpful comments from B.A.H. and authors from the Guo laboratory.
Author Information Reprints and permissions information is available at
npg.nature.com/reprintsandpermissions. The authors declare no competing
financial interests. Correspondence and requests for materials should be
addressed to M.G. (email@example.com).
NATURE|Vol 441|29 June 2006