C-terminal truncation and Parkinson’s
disease-associated mutations down-regulate
the protein serine/threonine kinase activity of
Chou Hung Sim1,2, Daisy Sio Seng Lio1,2, Su San Mok3,4,5, Colin L. Masters2,3,5,
Andrew F. Hill1,2,3, Janetta G. Culvenor3,4,5and Heung-Chin Cheng1,2,*
1Department of Biochemistry and Molecular Biology,2Bio21 Molecular Science and Biotechnology Institute,
3Department of Pathology and4Centre for Neuroscience, University of Melbourne, Parkville, Victoria 3010, Australia
and5Mental Health Research Institute, Parkville, Victoria 3052, Australia
Received July 9, 2006; Revised and Accepted September 19, 2006
The Parkinson’s disease (PD) causative PINK1 gene encodes a mitochondrial protein kinase called
PTEN-induced kinase 1 (PINK1). The autosomal recessive pattern of inheritance of PINK1 mutations suggests
that PINK1 is neuroprotective and therefore loss of PINK1 function causes PD. Indeed, overexpression of
PINK1 protects neuroblastoma cells from undergoing neurotoxin-induced apoptosis. As a protein kinase,
PINK1 presumably exerts its neuroprotective effect by phosphorylating specific mitochondrial proteins
and in turn modulating their functions. Towards elucidation of the neuroprotective mechanism of PINK1,
we employed the baculovirus-infected insect cell system to express the recombinant protein consisting of
the PINK1 kinase domain either alone [PINK1(KD)] or with the PINK1 C-terminal tail [PINK1(KD 1 T)]. Both
recombinant enzymes preferentially phosphorylate the artificial substrate histone H1 exclusively at serine
and threonine residues, demonstrating that PINK1 is indeed a protein serine/threonine kinase. Introduction
of the PD-associated mutations, G386A and G409V significantly reduces PINK1(KD) kinase activity. Since
Gly-386 and Gly-409 reside in the conserved activation segment of the kinase domain, the results suggest
that the activation segment is a regulatory switch governing PINK1 kinase activity. We also demonstrate
that PINK1(KD 1 T) is ?6-fold more active than PINK1(KD). Thus, in addition to the activation segment, the
C-terminal tail also contains regulatory motifs capable of governing PINK1 kinase activity. Finally, the avail-
ability of active recombinant PINK1 proteins permits future studies to search for mitochondrial proteins that
are preferentially phosphorylated by PINK1. As these proteins are likely physiological substrates of PINK1,
their identification will shed light on the mechanism of pathogenesis of PD.
Parkinson’s disease (PD) is the most common neurodegenera-
tive movement disorder affecting humans. Clinical symptoms
of PD develop when dopamine levels in corpus striatum of PD
patients falls below 70% of the healthy level (1). Autopsy
studies on idiopathic PD patients revealed that the movement
disorder usually correlated with progressive preferential loss
of dopaminergic neurons in substantia nigra pars compacta
of the mid-brain (1,2). However, the cause of progressive
preferential cell death of dopaminergic neurons in PD is still
Neurotoxins such as the mitochondrial complex I inhibitor,
1-methyl-4-phenylpyridium ion (MPPþ) are capable of indu-
cing PD, indicating involvement of mitochondrial dysfunction
in the pathogenesis of PD. This notion is further supported by
identification of the PINK1 gene encoding a mitochondrial
protein kinase as a causative PD gene in several pedigrees
of autosomal recessive familial PD (3,4). Patients carrying
the PD-associated PINK1 mutations exhibited early onset of
# The Author 2006. Published by Oxford University Press. All rights reserved.
For Permissions, please email: firstname.lastname@example.org
*To whom correspondence should be addressed. Tel: þ613 83442254; Fax: þ613 93481421; Email: email@example.com
Human Molecular Genetics, 2006, Vol. 15, No. 21
Advance Access published on September 25, 2006
by guest on May 30, 2013
the disease (before 50 years of age) (5,6). Their clinical pheno-
types are also somewhat different from those of patients suffer-
ing sporadic late-onset Parkinsonism—patients suffering from
the early-onset Parkinsonism exhibit slow progression of the
disease, good response to L-dopa treatment, dystonia at onset
and lesser cognitive decline. Exactly how alteration of PINK1
functions by PD-associated PINK1 mutations contributes to
the aforementioned clinical phenotypes remains unknown.
PTEN-induced kinase 1 (PINK1, the numbers in PINK1
constructs denote PINK1 residue numbers of the N- and
C-terminal boundaries) is expressed throughout the brain
(3,7). Bioinformatics analysis reveals that PINK1 contains
an N-terminal mitochondrial targeting sequence and a
protein kinase domain (Fig. 1A) (3,8,9). Silvestri et al. (8)
also proposed the segment containing residues 101–107 as a
putative transmembrane domain. Both endogenous and recom-
binant PINK1 expressed in mammalian cells are localized to
the mitochondria. In addition to the presence of the full-length
PINK1 (63 kDa), several proteolysed forms of PINK1 with
molecular masses ranging from 30 to 50 kDa also exist
(3,8–10), the functional role of these fragments is currently
Figure 1. Putative functional domains and motifs of PINK1 and conserved subdomains of the PINK1 kinase domain. (A) PINK1 sequence consists of (i) a
mitochondrial targeting sequence, (ii) a putative transmembrane domain and (iii) a putative protein serine/threonine kinase domain. The positions of amino
acid missense mutations (single-letter code) associated with familial forms of PD are indicated by each arrow. Non-sense mutations are indicated by X,
which denotes the introduction of a premature stop codon. Frameshift mutation is indicated by fsX. All except two of the PD-associated mutations are
mapped to the protein kinase domain of PINK1. (B) Multiple alignments of the catalytically critical motifs of PINK1 orthologues and other protein kinases.
PD-associated A168P mutation resides in the conserved ATP-anchoring b-hairpin P-loop (GXGXXGXV). Two PD-associated mutations were chosen in this
study—G386A and G409V mutations. Gly-386 corresponds to the glycine residue in the DFG motif at the start of the activation segment, which is responsible
for chelating the Mg2þion in the active site. Gly-409 is mapped to the protein substrate-binding loop upstream of the conserved APE motif at the end of the
activation segment. ‘Start’ and ‘End’ denote the residues at the N-terminal and C-terminal boundaries of the kinase domain, respectively. (C) Schematic
representation of PINK1 constructs generated for expression in Sf9 cells. Introduction of 6-histidine tag and FLAG-tag facilitates purification of the PINK1
proteins by Ni-NTA affinity column chromatography and immunoprecipitation by anti-Flag antibody, respectively.
3252 Human Molecular Genetics, 2006, Vol. 15, No. 21
by guest on May 30, 2013
Down-regulation of PINK1 expression by siRNA in
SH-SY5Y neuroblastoma cells increased the mortality of the
cells by 3-fold when the mitochondrial complex I was inhib-
ited with a neurotoxin (11). Furthermore, PINK1-deficient
Drosophila mutants display mitochondrial defects leading to
degeneration of flight muscles (12–14) and more importantly
show loss of dopaminergic neurons (13,14). Overexpression
of PINK1 in SH-SY5Y neuroblastoma cells resulted in a
neurotoxin-induced cell death (3,15). Collectively, these data
suggest that PINK1 is neuroprotective—it protects neuronal
cells against neurotoxin- and cell stress-induced cell death.
How might PINK1 exert its neuroprotective function?
Treatment of SH-SY5Y neuroblastoma cells with neurotoxins
causes loss of mitochondrial membrane potential and release
of cytochrome c (3,15). Both events are typically triggered
by apoptotic signals to initiate cell death (reviewed in 16).
Overexpression of PINK1 in these cells suppresses this loss
of mitochondrial membrane potential and subsequent release
of cytochrome c induced by neurotoxins (3,15). These data
indicate that PINK1 prevents cell death by inhibiting the
release of cytochrome c and maintaining the mitochondrial
membrane potential. Relevant to this notion, Petit et al. (15)
reported that fibroblasts from PINK1 familial PD patients
have caspase-3 activity at levels 33% higher than their age-
matched controls. Since caspase-3 can be activated by cyto-
chrome c released from mitochondria, this result implies that
these PD patients may have a higher level of released cyto-
chrome c. The mechanism by which PINK1 suppresses mito-
chondrial cytochrome c release and maintains mitochondrial
membrane potential remains an important outstanding ques-
tion. Presumably, PINK1 governs these two events by phos-
phorylating specific cellular proteins in the mitochondria.
Thus, identifying the physiological substrates of PINK1 will
be important in answering this question.
PD-associated PINK1 mutations map to the protein kinase
domain (Fig. 1A) (3–5,17–24). These mutations may signifi-
cantly reduce or abrogate kinase activity and in turn abolish
the neuroprotective function of PINK1. Previous biochemical
studies performed by two groups of investigators on PINK1
kinase activity showed that the recombinant protein kinase
domain of PINK1 (residues 112–496) expressed in E. coli
was capable of undergoing autophosphorylation and of phos-
phorylating casein (8,9). Unexpectedly, introduction of
enzyme gave conflicting effects on kinase activity: (i)
several missense mutations resulted in a slight reduction in
autophosphorylation and kinase activity towards casein, but
(ii) the C-terminal truncating W437X non-sense mutation
leads to activation (8). Thus, it is uncertain from these obser-
vations whether PD-associated mutations indeed inactivate
PINK1 kinase activity in PINK1 familial PD. These recombi-
nant PINK1 proteins were generated with the E. coli
expression system. Recombinant mammalian proteins are
prone to misfolding and/or forming insoluble aggregates
when expressed in E. coli (25). Furthermore, prokaryotic
E. coli lacks the enzymatic machinery for proper post-
translational modifications of mammalian proteins such as
fatty acid acylation and glycosylation. For these reasons, the
eukaryotic baculovirus-infected insect cell system has been
widely used to express recombinant mammalian protein
kinases for functional and structural studies (26–29). It is
possible that recombinant PINK1 proteins expressed in
E. coli were not properly folded and the data obtained may
not truly reflect the effects of PD-associated mutations on
PINK1 kinase activity. Since PINK1 is expressed in
metazoans ranging from nematodes to mammals, it is more
appropriate to use a eukaryotic expression system such as
the baculovirus-infected insect cell system to generate recom-
binant PINK1 proteins for functional studies (25,30).
Protein kinase domains are folded into a bilobal structure.
Catalysis of the phosphorylation reaction occurs within the
cleft between the two lobes (reviewed in 31–33). Conserved
regions within the kinase domain supply catalytically critical
residues for binding of substrates and for the phosphorylation
reaction (Fig. 1B). Mutations in the second PD-associated
kinase gene, leucine-rich repeat kinase 2 (LRRK2) cause auto-
somal late-onset familial and sporadic disease with disease
phenotype similar to typical PD (34,35). The most common
disease causing mutation G2019S for this large protein
kinase maps to the critical kinase activation loop DYG
motif, the effect of this change on kinase activity for
LRRK2 function is controversial requiring further investi-
gation. Two PD-associated mutations also map to the acti-
vation loop of PINK1—Gly-386 and Gly-409 in the
activation segment of PINK1 were recently found mutated
in two pedigrees of autosomal recessive familial PD (4).
Given that the activation segment is a likely regulatory
switch governing PINK1 kinase activity and substrate selectiv-
ity, we investigated how these two PD-associated mutations
affect PINK1 kinase activity.
Herein we show that active recombinant PINK1 proteins
can be expressed in baculovirus-infected insect Spodoptera
frugiperda (Sf9) cells. Using these active recombinant
PINK1 proteins, we have developed an in vitro assay to
measure PINK1 kinase activity, allowing us to assess the
impact of PD-associated mutations on PINK1 kinase activity.
With the active PINK1 constructs, we were able to examine
the substrate amino acid preference for phosphorylation by
PINK1 and define the regions in PINK1 that govern its
Expression of intact recombinant PINK1 proteins
in baculovirus-infected Sf9 cells
To study the enzymatic properties of the PINK1 kinase
domain, it is important to establish the N- and C-terminal
boundaries of this domain. The start of a kinase domain is
usually marked by a hydrophobic residue located about
seven residues upstream of the P-loop (GXGXXGXV)
(Fig. 1B) (36). The end of a kinase domain is usually rep-
resented by a loosely conserved His-X-aromatic-hydrophobic
motif (Fig. 1B) (36). This C-terminal boundary motif is
located 9–13 residues downstream of an invariant arginine
(Arg-497 of PINK1) that salt-bridges with the glutamate
(Glu-417) in the conserved APE motif of the activation
segment to maintain the structural integrity of the kinase
Human Molecular Genetics, 2006, Vol. 15, No. 21 3253
by guest on May 30, 2013
domain (36). On the basis of sequence alignment of PINK1
with other protein kinases, we postulate that the Tyr-156 and
Leu-511 of PINK1 correspond to N- and C-terminal bound-
aries of the kinase domain, respectively.
To study PINK1 kinase activity in the absence of possible
influence posed by its adjacent domains, we generated the
recombinant protein kinase domain of PINK1 (residues
148–515). We termed this construct as PINK1 kinase
domain [PINK1(KD)]. PD-associated G386A and G409V
mutations were introduced into PINK1(KD) to study their
effects on kinase activity. PINK1(KD) bearing G386A or
G409V mutations are referred to as KD(G386A) and
KD(G409V), respectively. Figure 1C depicts the schematics
of recombinant full length PINK1, the truncated PINK1 con-
structs and mutants produced for this study. These recombi-
nant proteins were purified as intact proteins of the predicted
size of ?42 kDa (Fig. 2A and B).
To investigate the influence of the C-terminal tail on PINK1
kinase activity, we also created a recombinant PINK1 protein,
named as PINK1[KD þ T], containing the protein kinase
148-581). The purified PINK1[KD þ T] has the predicted
size of ?48 kDa (Fig. 2A).
We also attempted to investigate the influence of regulatory
motifs located N-terminal to the kinase domain on PINK1
kinase activity. Sequence analysis of PINK1 reveals that the
segment (residues 121–147) located at the N-terminal end
of the kinase domain has the propensity to adopt an amphi-
pathic a-helix similar to the A-helix segment preceding the
kinase domain of Protein Kinase A (PKA) (37). To examine
the influence of this segment on PINK1 kinase activity, we
tried generation of a recombinant PINK1 protein (residues
121–581), which lacks the mitochondrial targeting sequence
and the putative transmembrane domain but retains the puta-
tive a-helix segment preceding the kinase domain. Upon
PINK1(121–581) was degraded. In addition, expression of
full length PINK1 was also tried. Again, upon expression in
Sf9 cells, almost all the recombinant PINK1 protein was
degraded to the degradation product of ?30 kDa (data not
tail of PINK1(residues
all the recombinant
PINK1(KD) is a protein serine/threonine kinase capable
of phosphorylating histone H1 and casein in vitro
Histone H1 and casein have been employed as in vitro artifi-
cial substrates of many protein serine/threonine kinases. For
protein tyrosine kinases, poly(Glu,Tyr) copolymer was
chosen as the in vitro substrate (38). To examine if recombi-
nant PINK(KD) is active and to assess the in vitro substrate
selectivity of PINK1, histone H1, casein and poly(Glu/Tyr)
were used as the substrates and the extent of their phosphoryl-
ation by PINK(KD) was compared. As shown in Figure 3,
PINK1 is catalytically active, and it prefers to phosphorylate
the lysine-rich histone H1 to the acidic residue-rich casein,
suggesting that PINK1 prefers basic proteins as in vitro sub-
strates. Furthermore, PINK1(KD) is unable to phosphorylate
poly(Glu,Tyr) (data not shown), indicating that PINK1 is not
a protein tyrosine kinase. Phosphoamino acid analysis of
PINK1-phosphorylated histone H1 reveals that histone H1
was phosphorylated exclusively at serine and threonine resi-
dues, confirming for the first time that PINK1 is a protein
serine/threonine kinase (Fig. 3B).
The PD-associated G386A and G409V mutations
significantly reduce PINK1 kinase activity
Most PD-associated mutations map to the conserved catalyti-
cally critical regions of the PINK1 kinase domain (5). It is
therefore plausible that the loss of kinase activity is the
common cause of PINK1 familial PD. To test this hypothesis,
we introduced PD-associated G386A and G409V mutations
into PINK1(KD) to examine the effects of these mutations
on its activity.
Gly-386 and Gly-409 are located in the activation segment
of the PINK1 kinase domain (Fig. 1B). The functional signifi-
cance of these two residues of PINK1 can be predicted by
examining the functions of homologous residues in PKA. As
shown in Figure 4A, Gly-386 is within the conserved DFG
motif responsible for chelating Mg2þ—its mutation is likely
to disrupt the Mg2þ-chelating of ATP and consequently
perturb the orientation of g-phosphate of ATP for the phos-
phorylation reaction. The G409V mutation maps to the
Figure 2. Western blot and SDS–PAGE analyses of purified recombinant
PINK1 proteins used in the enzymatic characterization. (A) PINK1(KD),
KD(G386A) and KD(G409V) were purified sequentially by nickel–NTA
agarose affinity chromatography and immunoprecipitated prior to enzymatic
analysis. PINK1(KD þ T) was purified sequentially by DEAE anion exchange
chromatography and immunoprecipitated prior to enzymatic analysis. The
purified preparations were analyzed by immunoblotting with anti-PINK1
antibody. The recombinant PINK1 proteins were immunoprecipitated as
intact proteins as follows: PINK1(KDþ T), 48.3 kDa; PINK1(KD) and its
PD-associated mutants, 42 kDa. (B) The purity of the final preparation of
PINK1(KD) was analyzed by silver staining of the SDS–PAGE gel.
A single prominent protein band of 42 kDa was detected.
3254 Human Molecular Genetics, 2006, Vol. 15, No. 21
by guest on May 30, 2013
protein substrate binding loop near the C-terminal end of the
activation segment. This mutation may upset the binding of
the protein substrates.Therefore,
KD(G409V) are likely to be much less active than the wild-
type PINK1(KD). In agreement with the prediction, the
specific kinase activity of PINK1(KD) is about 7- and
20-fold higher than those of KD(G409V) and KD(G386A),
respectively (Fig. 4), supporting the notion that PD-associated
mutations cause PD by significantly reducing PINK1 kinase
Figure 3. Analysis of the substrate preference and amino acid selectivity of
PINK1(KD) and its mutants. (A) Comparison of the extent of phosphorylation
of histone H1 and casein by PINK1(KD) and its PD-associated mutants. Auto-
radiogram of casein phosphorylated by PINK1(KD) and its PD-associated
mutants (left). Autoradiogram of histone H1 phosphorylated by PINK1(KD)
and its PD-associated mutants (right). On the basis of densitometry compari-
son, the level of histone H1 phosphorylation by PINK(KD) is 8-fold higher
than that of casein. (B) Phosphoamino acid analysis of histone H1 phosphory-
lated by PINK1(KD). The radioactively phosphorylated histone H1 was
subject to acid hydrolysis. The hydrolysate was mixed with the phosphoamino
acid standards, spotted onto a TLC plate before electrophoresis. Ninhydrin-
stained phosphoamino acid standards on the TLC plate (left). Autoradiogram
of the TLC plate (right).
Figure 4. PD-associated G386A and G409V mutations reduce PINK1 kinase
activity. (A) Mutation of Gly-386 and Gly-409 is expected to cause
reduction in the PINK1 kinase activity: Gly-386 and Gly-409 of PINK1
reside in the activation loop, which governs both the activity and substrate
recognition of many protein kinases. On the basis of the sequence alignment
between PKA and PINK1, Gly-386 and Gly-409 of PINK1 falls within the
conserved subdomains VII (blue) and VIII (pale green) of PKA crystal struc-
ture (PDB ID: 1ATP). Gly-186 of PKA (equivalent to Gly-386 of PINK1) is
hydrogen bonded to Asp-184 of PKA (equivalent to Asp-384 of PINK1) to
stabilize its electrostatic interaction with Mg2þ-ATP. Gly-200 of PKA
(equivalent to Gly-409 of PINK1) is hydrogen bonded to the backbone of
the protein substrate analog, which dictates substrate recognition and
binding. These interactions are important for bringing g-phosphate of ATP
and the target hydroxyl group in the substrate [represented by the methyl
group (magenta) of the substrate analog] together in close proximity for
the phosphorylation reaction. Therefore, mutations in Gly-386 and Gly-409
of PINK1 are likely to disrupt the ATP and substrate binding, respectively.
Green-dotted lines indicate hydrogen bonds. The image was generated with
the Swiss pdb viewer. (B) Purified PINK1(KD) (9.2 pmol), KD(G386A)
(20 pmol) and KD(G409V) (35 pmol) were incubated with 15 mg histone
H1 in the presence of kinase assay buffer and 5 mM [g-32P]ATP. The
extent of histone H1 phosphorylation over time is shown by autoradiograms.
(C) The protein bands corresponding to histone H1 were excised and the
extent of phosphorylation was quantified and expressed in picomoles of
32P incorporated in histone H1 per micromole of recombinant PINK1
protein and the data were plotted over time. The specific kinase activity
of PINK1(KD), KD(G386A) and KD(G409V) was determined from the
data presented to be 25, 1.2 and 3.4 pmol of 32P incorporated per minute
per micromole of recombinant PINK1 protein, respectively.
Human Molecular Genetics, 2006, Vol. 15, No. 213255
by guest on May 30, 2013
The C-terminal tail governs the kinase activity of PINK1
The PD-associated W437X mutation is a non-sense mutation,
which causes truncation of the C-terminal tail and a portion of
the kinase domain (residues 437–511) of PINK1. Another
PD-associated mutation D525fsX562 is a frame-shift mutation
starting at codon 525. Theoretically, this mutation would give
rise to a PINK1 mutant lacking most parts of the C-terminal tail.
These mutations are expected to cause PD by reducing PINK1
activity. To investigate the validity of this notion and decipher
the functional role of the C-terminal tail, we generated PINK1
(KDþ T) and compared its kinase activity with PINK1(KD).
Figure 5 shows that the specific kinase activity of PINK1
(KDþ T) is ?6-fold higher than that of PINK1(KD), revealing
for the first time that the C-terminal tail contains one or more
regulatory motifs capable of enhancing PINK1 kinase activity.
Our result also supports the notion that PD-associated truncating
mutations can cause PD by reducing PINK1 kinase activity.
The C-terminal tail influences PINK1’s selectivity of
phosphorylation sites in histone H1 but not in casein
The C-terminal tail of PKA is responsible for recruiting
protein substrates and positioning them in the catalytic cleft
(32). To examine the possible role of the C-terminal tail of
PINK1 in regulating PINK1 substrate recognition, we perfor-
med tryptic phosphopeptide mapping of casein and histone
H1 phosphorylated by PINK1(KD) and PINK1(KD þ T).
For most protein kinases, phosphorylation of substrates is
governed by the structural features in residues surrounding
the phosphorylation site of the protein substrates. Given that
casein and histone H1 have 23 and 17 potential phosphoryl-
ation sites, respectively, it is expected that only some of
these sites are preferentially phosphorylated by PINK1. This
notion is supported by the data shown in Figure 6. The
tryptic phosphopeptide maps of casein phosphorylated by
the two PINK1 proteins are almost identical (Fig. 6A and
B), indicating that PINK1(KD) and PINK(KD þ T) displayed
similar patterns of preference for phosphorylation sites in
casein. In contrast, histone H1 phosphorylated by the two
PINK1 proteins gave distinct tryptic phosphopeptide maps
(Fig. 6C and D). The tryptic phosphopeptide map (Fig. 6E)
of a mixture of histone phosphorylated by PINK1(KD) and
that phosphorylated by PINK1(KD þ T) reveals that the two
PINK1 proteins phosphorylated common (Sites a–i) and
distinct (Sites j–n) sites in histone H1 (depicted schematically
in Fig. 6F and G). Close inspection reveals that PINK1(KD)
phosphorylated nine sites in histone H1 (Sites a–i). Among
them, Site a and Site b were the preferred sites (Fig. 6C and
F). PINK1(KD þ T) also phosphorylated the same nine sites
in histone H1 with Sites b, d, f and g being the preferred
Figure 5. The effect of C-terminal tail on PINK1 kinase activity. Purified
PINK1[KD] (9.2 pmol) and PINK1(KD þ T) (6.9 pmol) were incubated with
15 mg histone H1 in the presence of kinase assay buffer and 5 mM
[g-32P]ATP. (A) The extent of histone H1 phosphorylation over time is
shown by autoradiograms. (B) The protein bands corresponding to histone
H1 were excised and the extent of phosphorylation was quantified and
expressed in picomoles of32P incorporated in histone H1 per micromole of
recombinant PINK1 protein and the data were plotted over time. The specific
kinase activity of PINK1(KD) and PINK1(KDþ T) were calculated to be 25
and 140 pmoles 32PO4
nant PINK1 protein, respectively.
¼incorporated per minute per micromole of recombi-
Figure 6. The C-terminal tail influences PINK1’s selectivity of phosphoryl-
ation sites in histone H1 but not casein. (A) Tryptic phosphopeptide map of
casein phosphorylated by PINK1(KD). (B) Tryptic phosphopeptide map of
casein phosphorylated by PINK1(KDþ T). (C) Tryptic phosphopeptide map
of histone H1 phosphorylated by PINK1(KD). (D) Tryptic phosphopeptide
map of histone H1 phosphorylated by PINK1(KD þ T). (E) A mixture of
the samples analyzed in (C) and (D). (F) Schematic of the phosphopeptide
maps shown in (C). (G) Schematic of the phosphopeptide maps shown in
(D). O: Origin. The arrows in (G) indicate the phosphopeptide fragments
that were preferentially phosphorylated by PINK1(KDþ T).
3256 Human Molecular Genetics, 2006, Vol. 15, No. 21
by guest on May 30, 2013
sites of phosphorylation (Fig. 6D and G). In addition to these
nine sites, PINK1(KD þ T) phosphorylated five other sites
(Sites j–n) in histone H1. Thus, it is clear from the data that
the C-terminal tail can influence the phosphorylation site
selectivity of the PINK1 kinase domain. Although histone
H1 is also phosphorylated by PKA, the tryptic phosphopeptide
map of histone H1 phosphorylated by PKA is very different
from those of the two PINK1 constructs (Supplementary
Material, Fig. S3). Taken together, the data suggest that
(i) PINK1 recognizes specific
surrounding the target phosphorylation sites in its substrates
as substrate selectivity determinants and (ii) the C-terminal
tail of PINK1 participates in determining the substrate
selectivity of PINK1.
With the baculovirus-infected insect cell expression system,
we were able to generate intact recombinant PINK1 constructs
for investigation into the enzymatic properties of PINK1.
These PINK1 proteins are catalytically active; they preferen-
tially phosphorylate histone H1 with specific enzymatic activi-
ties ranging from 25–140 pmol PO4
of enzyme. Although histone H1 is not the physiological
substrate of PINK1, it allowed us to develop an assay
method to investigate the catalytic properties, the impact of
PD-associated mutations on PINK1 kinase activity as well as
the regulatory properties of PINK1. Indeed, with this assay
method we found that (i) PINK1 is a protein serine/threonine
kinase—it phosphorylates histone H1 exclusively at serine and
threonine residues, (ii) PINK1 kinase activity is significantly
reduced by two PD-associated mutations mapped to the
kinase domain, supporting the hypothesis that loss-of-function
mutations of PINK1 induce PD by abrogating PINK1 kinase
activity (3), (iii) the C-terminal tail contains regulatory
motifs that enhance PINK1 kinase activity, partly by assisting
the kinase domain in substrate recognition. It is reported that
PINK1 can protect neuronal cells from undergoing apoptosis
through an unknown mechanism when the cells are treated
with neurotoxins (11,15). One of the avenues to decipher
this mechanism is to identify the physiological substrates of
PINK1. The availability of active recombinant PINK1 proteins
permits the search for PINK1 substrates.
Catalytic properties of PINK1
It is clear from the results shown in Figure 6 that the subset of
serine and threonine residues targeted by PINK1 in histone H1
is distinct from those phosphorylated by PKA. Most protein
kinases recognize the specific features in the primary struc-
tures around the phosphorylation sites in protein substrates
as substrate selectivity determinants. For PKA, the optimal
phosphorylation site sequence is R/K-R/K-x-S/T-Hb, where
x represents any amino acid and Hb represents a hydrophobic
residue (reviewed in 39). The results shown in Figure 6 and
Supplementary Material, Figure S3 suggest that the optimal
phosphorylation site sequence of PINK1 is different from
that of PKA. The availability of active recombinant PINK1
proteins allows future studies employing the peptide library
approach to define the optimal phosphorylation target
sequence of PINK1 (reviewed in 40). Knowledge of the
optimal phosphorylation sequence of PINK1 will facilitate
the search for PINK1 protein substrates because mitochondrial
proteins with sequences similar to that of the PINK1 optimal
phosphorylation sequence are potential physiological sub-
strates of PINK1.
The results shown in Figure 6 indicate that the C-terminal
tail of PINK1 can influence phosphorylation site selectivity
in substrates. It is possible that the C-terminal tail contains
functional motif(s) that bind to specific sequences in a
protein substrate and in turn direct the kinase domain to phos-
phorylate a specific site in the substrate. In light of this, it
would be worthwhile in future investigation to use recombi-
nant C-terminal tail of PINK1 as the ‘bait’ to search mitochon-
drial extract of neuronal cells for potential PINK1 protein
substrates that bind to the C-terminal tail.
Regulatory properties of PINK1
Similar to many protein kinases, the kinase activity of PINK1
is likely governed by configuration of the activation segment
in the kinase domain. Indeed, mutations of Gly-386 and
Gly-409 in this segment induce significant reduction of
PINK1 kinase activity (Fig. 4). As shown in Figure 4A,
Gly-386 corresponds to the conserved DFG motif that forms
the Mgþþ-chelating loop of protein kinases. It is therefore
conceivable that mutation of Gly-386 to alanine can disrupt
the structural integrity of the Mgþþ-chelating loop and conse-
quently induces significant reduction in PINK1 kinase activity
(Fig. 4). Relevant to this, the homologous Gly (Gly-2019) in
the Mgþþ-chelating loop of another PD-related protein
kinase LRRK2 was found mutated to serine in many PD
patients (reviewed in 35). Biochemical analysis revealed
that, in contrast to the inactivating effect of G386A mutation,
the G2019S mutation induces a significant increase in kinase
activity of LRRK2, suggesting that this missense mutation
of PINK1 causes PD by activating LRRK2. What is the
structural basis for the opposing effects arising from subs-
titution of the conserved Gly in the Mgþþ-chelating loop in
PINK1 and LRRK2? Determination of the three-dimensional
structures of PINK1 and LRRK2 kinase domain will reveal
the answers to this question.
In many protein kinases, configuration of the activation
segment is regulated by phosphorylation of a conserved
segment. Phosphorylation of this residue in most cases acti-
vates the enzyme. For example, the active conformations of
PKA and c-Src kinase are stabilized by phosphorylation of
Thr-197 and Tyr-416, respectively, in the activation segment
(reviewed in 31,41,42). Using the recombinant PINK1
construct spanning the segment corresponding to residues
112–496, Beilina et al. (9) and Silvestri et al. (8) detected
autophosphorylation of the recombinant kinase in vitro. In
contrast to their findings, we failed to detect autophosphoryla-
tion of PINK1(KD) and PINK1(KDþ T). Thus, the autopho-
sphorylation site of PINK1 is unlikely to be located in the
activation segment and other parts of the kinase domain.
The PINK1 construct used by these groups of researchers
contains the segment corresponding to residues 112–147,
Human Molecular Genetics, 2006, Vol. 15, No. 213257
by guest on May 30, 2013
which is not included in both PINK1 constructs used in our
study. It is possible that the autophosphorylation site resides
in this segment. The other explanation for the discrepancy of
our results and those reported by Beilina et al. (9) and Silvestri
et al. (8) relates to the intracellular compartment in which
expressed. It is noteworthy that owing to the lack of a mito-
chondrial targeting motif, these two purified recombinant
PINK1 proteins were expressed and folded in the cytosol of
Sf9 cells (Supplementary Material, Fig. S1). Similar to other
mitochondrial proteins encoded by nuclear genes, native
PINK1 is likely synthesized in the cytosol and then imported
into the mitochondria in an unfolded configuration. Presum-
ably, once the unfolded PINK1 protein enters the mito-
chondria, it is processed and folded to the functional
configuration with the assistance of chaperones (43). It is
possible that PINK1 acquires the ability to undergo autopho-
sphorylation only when it is folded in the functional
configuration in the mitochondria. Alternatively, PINK1 can
undergo autophosphorylation only when it binds to a regulat-
ory protein in mitochondria. In view of this, the enzymatic
properties such as the ability to undergo autophosphorylation
and kinetics of phosphorylation of histone H1 should be
compared between recombinant PINK1(KD þ T) used in the
current studies and endogenous PINK1 purified from the
mitochondria of mammalian cells.
PINK1(KD þ T)were
PINK1 interplays with Parkin and superoxide dismutase 1
to maintain survival of neurons in Drosophila
In addition to PINK1, the Parkin gene encoding an E3 ubiquitin
ligase is also a recessive PD-causative gene. Thus, in addition
proteasome system can also cause PD. Using Drosophila as
a model system, three groups of researchers recently reported
that PINK1 and Parkin act in a common signalling pathway to
maintain mitochondrial integrity (12–14,44). Their studies
revealed that Parkin operates downstream of PINK1 in this
pathway (12–14). In addition, Wang et al. (44) demonstrated
that expression of human superoxide dismutase 1 (SOD1) sup-
pressed degeneration of neurons in Drosophila induced by
Drosophila PINK1 inactivation. Their results suggest that
PINK1 maintains survival of neuronal cells by protecting the
cells from undergoing oxidative stress. Hence, there are two
important outstanding questions: (i) How might PINK1,
which is located in mitochondria, regulate the activity of
Parkin in the cytosol? (ii) How might PINK1 protect
neurons from undergoing oxidative stress? Identifying the
physiological substrates of PINK1 will provide the answers
to these questions. The availability of active recombinant
PINK1 constructs will permit a search for the protein sub-
strates, which upon phosphorylation, mediate regulation of
Parkin by PINK1.
The four groups of researchers mentioned above also
(dPINK1) gene induced pathogenic phenotypes including
mitochondrial dysfunction, shortened lifespan and dopamin-
ergic neuron degeneration (12–14,44). These phenotypes
can be rescued by expressing wild-type human PINK1 in the
dPINK1-deficient Drosophila (14), indicating that the role of
an impaired ubiquitin–
PINK1 in maintaining mitochondrial integrity and preventing
degeneration of dopaminergic neurons is conserved from
Drosophila to man. However, upon deletion of the C-terminal
72-residue segment, the ability of human PINK1 to rescue
these phenotypes in dPINK1-deficient Drosophila is lost. On
the basis of these results, Yang et al. (14) suggested that the
C-terminal tail may affect the following enzymatic properties
of PINK1: (i) kinase activity, (ii) the ability to bind to its
substrates, (iii) the ability to bind to other cofactors (14).
The data shown in Figures 5 and 6 give further credence to
their prediction of the role of the C-terminal tail in governing
PINK1 kinase activity and binding of PINK1 to substrates.
Our future investigation will include examining whether the
C-terminal tail mediates PINK1 binding to some unknown
regulatory cofactors in vivo.
How do heterozygous C-terminal truncation mutations
of PINK1 contribute to PD?
Intriguingly, heterozygous PINK1 mutations can also give rise
to PD (5). For example, the Q456X non-sense mutation has
been reported as a heterozygous PD-associated mutation (5).
Truncated PINK1 resulting from this mutation lacks the
C-terminal tail and a small portion of the kinase domain. On
the basis of our data presented in Figure 5, it is expected
that this truncated form of PINK1 exhibits very low or no
kinase activity. Since Q456X is observed in PD patients as a
heterozygous mutation, it is possible that both the wild-type
PINK1 encoded by the normal allele, and the truncated
PINK1 mutant encoded by the mutant allele are co-expressed
in patients carrying this mutation. Bonifati et al. (5) suggested
that the truncated PINK1 mutant may act as a dominant
negative mutant, which interferes with the activation and/or
function of wild-type PINK1 encoded by the normal allele.
Future investigation into the role of the C-terminal tail in
regulating PINK1 function may provide the answer to this
In summary, our study has laid the ground work for eluci-
dating the molecular basis of regulation and neuroprotective
function of PINK1, both are areas critical to understanding
the mechanism of PD pathogenesis.
MATERIALS AND METHODS
Bovine histone H1 and dephosphorylated casein were from
Sigma-Aldrich. Poly(Glu,Tyr), a random copolymer of
glutamate and tyrosine (molar ratio of Glu: Tyr ¼ 4:1) was
purchased from Sigma-Aldrich. It is an efficient substrate for
most known protein tyrosine kinases (38). Bacfectinw,
pBacPAK9 vector and Bsu361-digested BacPAK6 viral
DNA were from BD Biosciences. QuickChangewII site-
directed mutagenesis kit was from Stratagene. The human
PINK1 cDNA clone was from OriGene Technology. The anti-
PINK1 antibody raised against the segment corresponding to
residues 258–274 of PINK1 was from Imgenex Corporation.
agarose and the catalytic subunit of PKA were from
3258Human Molecular Genetics, 2006, Vol. 15, No. 21
by guest on May 30, 2013
Generation of pBacPAK9-recombinant PINK1
Full length PINK1 gene was retrieved from the human PINK1
cDNA clone by restriction digestion with EcoR1 and Xba1.
Truncated PINK1 constructs encoding PINK1(121–581),
PINK1(148–581) [also referred to PINK1(KD þ T)] and
amplified by PCR from the human PINK1 cDNA clone.
PCR reactions were used to introduce the restriction sites as
well as the sequences encoding (i) the poly-His tag at the
N-terminal end of PINK(121–581) and the C-terminal end
of PINK1(148–515), and (ii) a FLAG sequence (Asp-Tyr-
Lys-Asp-Asp-Asp-Asp-Lys) at the N-terminal end of both
PINK1(148–581) and PINK(148–515). The PCR products
were digested with EcoRI and XbaI endonucleases and then
ligated to pre-digested pBacPAK9 vector to generate the fol-
lowing plasmids: pBacPAK9-PINK1,
(121–581), pBacPAK9-PINK1(148–581) and pBacPAK9-
PINK1(148–515). Authenticity of the plasmids containing
the cDNA encoding full length and the truncated PINK1
constructs was confirmed by DNA sequencing.
Generation of pBacPAK9-PINK1(148–515)[G386A]
and pBacPAK9-PINK1(148–515)[G409V] plasmids
by site-directed mutagenesis
The glycine residues at positions 386 (G386) and 409 (G409) of
PINK1 were mutated to alanine (G386A) and valine (G409V)
residues, respectively, in the PINK1(148–515) construct, by
using the QuickChangewII site-directed mutagenesis kit
according to manufacturer’s instructions. Authenticity of the
resultant plasmids, pBacPAK9-PINK1(148–515)[G386A] and
pBacPAK9-PINK1(148–515) [G409V] was confirmed by
Generation of recombinant PINK1 baculovirus
To generate the recombinant PINK1 baculovirus expression
vector, 1 ? 106Sf9 cells were co-transfected with respective
viral DNA and Bacfectinw, according to the manufacturer’s
instructions. The cell lysates were analyzed by immunoblot-
ting with anti-PINK1 antibody and anti-Flag mouse mono-
clonal antibodies to confirm the production of recombinant
PINK1 in the infected Sf9 cells.
Purification of recombinant PINK1 proteins
Sf9 cells (1 l and at cell densities ranging 0.6–0.9 ? 106cells/
ml) were cultured in Grace’s medium (Invitrogenw) in the
presence of 7% ‘Cosmic’ calf serum (Hyclone) before they
were infected with the recombinant PINK1 baculovirus at a
multiplicity of infection ?1. At 50 h after infection, the cells
were harvested for protein purification by centrifugation at
1000g for 5 min. The cell pellet was washed with serum-free
Grace’s medium (Invitrogenw) and centrifuged at 1000g for
5 min again. All subsequent protein purification procedures
were carried out at 48C. The cell pellet was homogenized in
Lysis Buffer [50 mM Tris–HCl, pH 7, 15% glycerol, 0.1 mg/
ml soybean trypsin inhibitor, 0.2 mg/ml benzamidine–HCl,
0.1 mg/ml phenylmethylsulfonylfluoride (PMSF)] and centri-
fuged at 100 000g for 30 min. The supernatant was collected
for subsequent protein purification.
Supernatants derived from cell lysates containing recombi-
nant full-length PINK1, PINK1(121–581) and PINK1(148–
581) [also called PINK(KD þ T)] constructs, were subjected
to DEAE anion exchange column chromatography. These
supernatants were diluted 1:1 with DEAE Buffer (25 mM
Hepes, pH 7.0, 15% glycerol, 0.2 mg/ml benzamidine–HCl,
0.1 mg/ml PMSF, 1 mM dithiothreitol) and loaded onto a
DEAE Sepharose column (80 ml bed volume) pre-equilibrated
with DEAE buffer. The proteins were eluted with a 0–1 M
NaCl gradient in DEAE buffer. The elution profiles of recom-
binant PINK1 proteins were monitored by immunoblotting
with anti-PINK1 antibody.
PD-associated mutants contain a 6-histidine tag to allow puri-
fication by affinity chromatography onto a column with
(QIAGENw). Supernatants derived from cell lysates contain-
ing these recombinant PINK1 proteins were incubated in
5 ml of nickel–NTA agarose gel for 2 h. The nickel–NTA
agarose gel was washed with 10 ml of wash buffer (25 mM
Hepes, pH 7.0, 15% glycerol, 0.2 mg/ml benzamidine–HCl,
0.1 mg/ml PMSF, 1 M NaCl). The proteins were eluted with
10 ml of Elution Buffer (wash buffer as above containing
300 mM imidazole, at pH 7.5). The elution profiles of recom-
binant PINK1 proteins were monitored with anti-Flag western
Flag-tagged recombinant PINK1 proteins were immunopre-
cipitated with anti-Flag antibody agarose (Sigma-Aldrich).
2 ? 1 ml of DEAE Buffer with 1 M NaCl and 2 ? 1 ml
DEAE Buffer. All subsequent experiments were performed
with immunoprecipitated recombinant PINK1 proteins.
Quantitation of Flag-tagged PINK1 proteins purified
by anti-Flag antibody agarose
All Flag-tagged PINK1 proteins including PINK(KD þ T),
PINK(KD), KD(G386A) and KD(G409V) were affinity puri-
fied by immunoprecipitation with anti-Flag antibody agarose
prior to enzymatic analysis. The Flag-Hck(222–503), contain-
ing the kinase domain and the C-terminal regulatory domain
of the Hck tyrosine kinase were generated and purified as
described previously (45). It was used as the Flag-protein stan-
dard for quantitation of the PINK1 proteins in the immunopre-
cipitates. Briefly, aliquots of the immunoprecipitates and
known amounts (4–40 pmol) of the Flag-protein standard
were analyzed by anti-Flag western blotting.
The anti-Flag immunoreactivities of the PINK1 proteins and
the protein standard were quantitated and expressed as densi-
tometry units. The values from the Flag-protein standard were
used to generate the standard curve. From the curve and the
immunoreactivities of the PINK1 proteins, their amounts
were determined. Supplementary Material, Figure S1 illus-
trates how we determined the amounts of the immunoprecipi-
tated PINK1 proteins with this method.
Human Molecular Genetics, 2006, Vol. 15, No. 213259
by guest on May 30, 2013
Phosphorylation of histone H1, casein and poly(Glu,Tyr)
copolymer by recombinant PINK1 proteins
Phosphorylation of exogenous substrates (1 mg each of histone
H1, dephosphorylated casein or poly(Glu,Tyr) copolymer)
was performed at 308C for 1 h in a final volume of 25 ml con-
taining the Assay Buffer (20 mM Tris–HCl, pH 7.5, 10 mM
MgCl2, 1 mM
[g-32P]ATP and 10 ml (bed volume) of recombinant PINK1
proteins bound to the anti-Flag antibody agarose. The
amounts of PINK1 proteins used in the experiment were quan-
titated by anti-Flag western blotting to be 4.9 pmol, 11 pmol
and 19 pmol for PINK(KD), KD(G386A) and KD(G409V),
respectively. The reaction was stopped with the addition of
5? SDS PAGE sample buffer and resolved by 10% Tris–
Glycine SDS–PAGE. The SDS–PAGE gel was dried and
exposed to a phosphoimager plate for subsequent quantitative
50 mM Na3VO4), 10 mM
Time course of phosphorylation of histone H1 by
recombinant PINK1 proteins
Histone H1 (15 mg) was phosphorylated at 248C for 56 min by
the resuspended immunoprecipitated recombinant enzymes
(6.9–35 pmol PINK1 proteins in 30 ml of anti-Flag antibody
agarose) in a final volume of 75 ml of the assay buffer with
5 mM [g-32P]ATP. Aliquots (5 ml) of the reaction were taken
out and mixed with 5? SDS–PAGE sample buffer at
8 minute intervals. The proteins were resolved by SDS–
PAGE. The SDS-gel was dried and exposed to a phosphoima-
associated radioactivity was determined by scintillation
The amounts of recombinant PINK1 proteins present in the
above reactions were quantified by comparing their immuno-
reactivity with anti-Flag antibody and that of a Flag-tagged
protein standard of known concentration.
32P-labelled histone H1 bands were excised and the
Phosphoamino acid analysis
Histone H1 and casein were phosphorylated by immunopreci-
pitated Flag-tagged recombinant PINK1 as mentioned earlier.
After SDS–PAGE, proteins in the gel were transferred onto
32P-labelled histone H1 was detected by autoradiography
and the corresponding portion of the membrane was excised.
The32P-labelled histone H1 was hydrolyzed by 6 M HCl at
1108C for 90 min. After repeated freeze-drying to remove
HCl, the hydrolysate was mixed with phosphoamino acid
standards [1 mg each of phosphotyrosine, phosphoserine and
phosphothreonine (Sigma-Aldrich)] and spotted onto a
cellulose thin-layer chromatography (TLC) plate (Macherey-
Nagel). The amino acid mixture was separated by thin-layer
electrophoresis (TLE) in TLE buffer [5% (v/v) acetic acid,
0.5% (v/v) pyridine in H2O] at 500 V for 80 min, as described
previously (46). The phosphoamino acids were detected
by ninhydrin spray [0.1% (w/v) in acetone], followed by
heating of the TLC plate. The
acids were detected by phosphoimager.
Phosphopeptide mapping of radioactively
phosphorylated casein and histone H1
PINK1(KD) (?100 pmol) and PINK1(KD þ T) (?40 pmol)
volume ¼ 10 ml) were used to phosphorylated histone H1 or
casein (15 mg each) at 308C for 30 min in a final volume of
40 ml of the assay buffer with [g-32P]ATP. Since casein is a
much poorer substrate of the PINK1 (Fig. 3A), ATP of a
lower concentration (1 mM) and high specific radioactivity
(?145 000 cpm/pmol) was used in the reaction. For histone
H1 phosphorylation, the ATP concentration and specific radio-
activity used for the reaction were 10 mM and ?36 000 cpm/
pmol, respectively. The reactions were stopped by the addition
of SDS sample buffer. Following SDS–PAGE, proteins in
the gel were transferred onto a nitrocellulose membrane.
The32P-labelled casein and histone H1 were detected by auto-
radiography and the corresponding portions of the membrane
were excised and digested with 1 mg/ml trypsin in the
presence of 5% (v/v) CH3CN in 100 mM NH4HCO3. In the
presence of 5% CH3CN, the tryptic fragments including
the32P-labelled peptides were detached from the membrane.
The resulting solution was freeze-dried and the tryptic
peptide fragments were separated by TLE in the first dimension
and TLC in the second dimension as described previously (33).
Aliquots containing ?2000 cpm of the phosphopeptide
fragments each were spotted on the TLC plates for analysis.
PINK1(KD) and that phosphorylated by PINK19(KD þ T)
(Fig. 6E), an aliquot containing ?1000 cpm of phosphopep-
tide fragments was taken from each sample. The two aliquots
were mixed together prior to spotting onto the TLC plate. To
maximize the space on the TLC plate (24 ? 24 cm) available
for the separation of the positively charged tryptic phospho-
peptide fragments, the samples were spotted 16 cm from the
cathode end [labelled as (2) in Fig. 6] and 8 cm from the
anode end [labelled as (þ) in Fig. 6].
Supplementary Material is available at HMG Online.
We are grateful to Bruce Kemp for constructive suggestions
for this project. We thank Ryan Mills for critical reading of
this manuscript. This project was supported by grants from
the Cancer Council of Victoria and National Health and
Medical Research Council of Australia.
Conflict of Interest statement. None declared.
1. Bernheimer, H., Birkmayer, W., Hornykiewicz, O., Jellinger, K. and
Seitelberger, F. (1973) Brain dopamine and the syndromes of Parkinson
and Huntington. Clinical, morphological and neurochemical correlations.
J. Neurol. Sci., 20, 415–455.
3260Human Molecular Genetics, 2006, Vol. 15, No. 21
by guest on May 30, 2013
2. Hughes, A.J., Daniel, S.E., Kilford, L. and Lees, A.J. (1992) Accuracy of
clinical diagnosis of idiopathic Parkinson’s disease: a clinico-
pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry, 55,
3. Valente, E.M., Abou-Sleiman, P.M., Caputo, V., Muqit, M.M., Harvey, K.,
Gispert, S., Ali, Z., Del Turco, D., Bentivoglio, A.R., Healy, D.G. et al.
(2004) Hereditary early-onset Parkinson’s disease caused by mutations in
PINK1. Science, 304, 1158–1160.
Agid, Y., Durr, A. and Brice, A. (2006) Mutational analysis of the PINK1
gene in early-onset parkinsonism in Europe and North Africa. Brain, 129,
5. Bonifati, V., Rohe, C.F., Breedveld, G.J., Fabrizio, E., De Mari, M.,
Tassorelli, C., Tavella, A., Marconi, R., Nicholl, D.J., Chien, H.F. et al.
(2005) Early-onset parkinsonism associated with PINK1 mutations:
frequency, genotypes, and phenotypes. Neurology, 65, 87–95.
6. Schrag, A. and Schott, J.M. (2006) Epidemiological, clinical, and
genetic characteristics of early-onset parkinsonism. Lancet Neurol., 5,
7. Taymans, J.M., Van den Haute, C. and Baekelandt, V. (2006) Distribution
of PINK1 and LRRK2 in rat and mouse brain. J. Neurochem., 98,
8. Silvestri, L., Caputo, V., Bellacchio, E., Atorino, L., Dallapiccola, B.,
Valente, E.M. and Casari, G. (2005) Mitochondrial import and enzymatic
activity of PINK1 mutants associated to recessive parkinsonism. Hum.
Mol. Genet., 14, 3477–3492.
9. Beilina, A., Van Der Brug, M., Ahmad, R., Kesavapany, S., Miller, D.W.,
Petsko, G.A. and Cookson, M.R. (2005) Mutations in PTEN-induced
putative kinase 1 associated with recessive parkinsonism have
differential effects on protein stability. Proc. Natl Acad. Sci. USA, 102,
10. Gandhi, S., Muqit, M.M., Stanyer, L., Healy, D.G., Abou-Sleiman, P.M.,
Hargreaves, I., Heales, S., Ganguly, M., Parsons, L., Lees, A.J. et al.
(2006) PINK1 protein in normal human brain and Parkinson’s disease.
Brain, 129, 1720–1731.
11. Deng, H., Jankovic, J., Guo, Y., Xie, W. and Le, W. (2005) Small
interfering RNA targeting the PINK1 induces apoptosis in dopaminergic
cells SH-SY5Y. Biochem. Biophys. Res. Commun., 337, 1133–1138.
Hay, B.A. and Guo, M. (2006) Drosophila pink1 is required for
mitochondrial function and interacts genetically with parkin. Nature, 441,
13. Park, J., Lee, S.B., Lee, S., Kim, Y., Song, S., Kim, S., Bae, E., Kim, J.,
Shong, M., Kim, J.M. et al. (2006) Mitochondrial dysfunction in
Drosophila PINK1 mutants is complemented by parkin. Nature, 441,
Beal, M.F., Vogel, H. and Lu, B. (2006) Mitochondrial pathology and
muscle and dopaminergic neuron degeneration caused by inactivation of
Drosophila Pink1 is rescued by Parkin. Proc. Natl Acad. Sci. USA, 103,
15. Petit, A., Kawarai, T., Paitel, E., Sanjo, N., Maj, M., Scheid, M., Chen, F.,
Gu, Y., Hasegawa, H., Salehi-Rad, S. et al. (2005) Wild-type PINK1
prevents basal and induced neuronal apoptosis, a protective effect
abrogated by Parkinson disease-related mutations. J. Biol. Chem., 280,
16. Adrain, C. and Martin, S.J. (2001) The mitochondrial apoptosome: a
killer unleashed by the cytochrome seas. Trends Biochem. Sci., 26,
17. Hatano, Y., Li, Y., Sato, K., Asakawa, S., Yamamura, Y., Tomiyama, H.,
Yoshino, H., Asahina, M., Kobayashi, S., Hassin-Baer, S. et al. (2004)
Novel PINK1 mutations in early-onset parkinsonism. Ann. Neurol., 56,
18. Healy, D.G., Abou-Sleiman, P.M., Gibson, J.M., Ross, O.A., Jain, S.,
Gandhi, S., Gosal, D., Muqit, M.M., Wood, N.W. and Lynch, T. (2004)
PINK1 (PARK6) associated Parkinson disease in Ireland. Neurology, 63,
19. Rohe, C.F., Montagna, P., Breedveld, G., Cortelli, P., Oostra, B.A. and
Bonifati, V. (2004) Homozygous PINK1 C-terminus mutation causing
early-onset parkinsonism. Ann. Neurol., 56, 427–431.
20. Li, Y., Tomiyama, H., Sato, K., Hatano, Y., Yoshino, H., Atsumi, M.,
Kitaguchi, M., Sasaki, S., Kawaguchi, S., Miyajima, H. et al. (2005)
Clinicogenetic study of PINK1 mutations in autosomal recessive
early-onset parkinsonism. Neurology, 64, 1955–1957.
21. Criscuolo, C., Volpe, G., De Rosa, A., Varrone, A., Marongiu, R.,
Mancini, P., Salvatore, E., Dallapiccola, B., Filla, A., Valente, E.M. et al.
(2006) PINK1 homozygous W437X mutation in a patient with
apparent dominant transmission of parkinsonism. Mov. Disord., 21,
22. Atsumi, M., Li, Y., Tomiyama, H., Sato, K. and Hattori, N. (2006) A
62-year-old woman with early-onset Parkinson’s disease associated with
the PINKi gene deletion. Rinsho Shinkeigaku, 46, 199–202.
23. Klein, C., Grunewald, A. and Hedrich, K. (2006) Early-onset
parkinsonism associated with PINK1 mutations: frequency, genotypes,
and phenotypes. Neurology, 66, 1129–1130.
24. Deng, H., Le, W.D., Zhang, X., Pan, T.H. and Jankovic, J. (2005) G309D
and W437OPA PINK1 mutations in Caucasian Parkinson’s disease
patients. Acta. Neurol. Scand., 111, 351–352.
25. Marston, F.A. (1986) The purification of eukaryotic polypeptides
synthesized in Escherichia coli Biochem. J., 240, 1–12.
26. Messerschmidt, A., Macieira, S., Velarde, M., Badeker, M., Benda, C.,
Jestel, A., Brandstetter, H., Neuefeind, T. and Blaesse, M. (2005) Crystal
structure of the catalytic domain of human atypical protein kinase C-iota
reveals interaction mode of phosphorylation site in turn motif. J. Mol.
Biol., 352, 918–931.
27. Yang, J., Cron, P., Good, V.M., Thompson, V., Hemmings, B.A. and
Barford, D. (2002) Crystal structure of an activated Akt/protein kinase B
ternary complex with GSK3-peptide and AMP-PNP. Nat. Struct. Biol., 9,
28. Nowakowski, J., Cronin, C.N., McRee, D.E., Knuth, M.W., Nelson, C.G.,
Pavletich, N.P., Rogers, J., Sang, B.C., Scheibe, D.N., Swanson, R.V.
et al. (2002) Structures of the cancer-related Aurora-A, FAK, and EphA2
protein kinases from nanovolume crystallography. Structure, 10,
29. Sicheri, F., Moarefi, I. and Kuriyan, J. (1997) Crystal structure of the Src
family tyrosine kinase Hck. Nature, 385, 602–609.
30. Kidd, I.M. and Emery, V.C. (1993) The use of baculoviruses as expression
vectors. Appl. Biochem. Biotechnol., 42, 137–159.
31. Taylor,S.S., Knighton, D.R.,Zheng, J., Ten Eyck,L.F. and Sowadski,J.M.
(1992) Structural framework for the protein kinase family. Annu. Rev.
Cell. Biol., 8, 429–462.
32. Johnson, D.A., Akamine, P., Radzio-Andzelm, E., Madhusudan, M. and
Taylor, S.S. (2001) Dynamics of cAMP-dependent protein kinase. Chem.
Rev., 101, 2243–2270.
33. Chong, Y.P., Mulhern, T.D., Zhu, H.J., Fujita, D.J., Bjorge, J.D.,
Tantiongco, J.P., Sotirellis, N., Lio, D.S., Scholz, G. and Cheng, H.C.
(2004) A novel non-catalytic mechanism employed by the C-terminal
Src-homologous kinase to inhibit Src-family kinase activity. J. Biol.
Chem., 279, 20752–20766.
34. Kachergus, J., Mata, I.F., Hulihan, M., Taylor, J.P., Lincoln, S., Aasly, J.,
Gibson, J.M., Ross, O.A., Lynch, T., Wiley, J. et al. (2005) Identification
of a novel LRRK2 mutation linked to autosomal dominant parkinsonism:
evidence of a common founder across European populations. Am. J. Hum.
Genet., 76, 672–680.
35. Mata, I.F., Wedemeyer, W.J., Farrer, M.J., Taylor, J.P. and Gallo, K.A.
(2006) LRRK2 in Parkinson’s disease: protein domains and functional
insights. Trends Neurosci., 29, 286–293.
36. Hanks, S.K. and Hunter, T. (1995) Protein kinases 6. The eukaryotic
protein kinase superfamily: kinase (catalytic) domain structure and
classification. Faseb J, 9, 576–596.
37. Knighton, D.R., Zheng, J.H., Ten Eyck, L.F., Xuong, N.H., Taylor, S.S.
and Sowadski, J.M. (1991) Structure of a peptide inhibitor bound to the
catalytic subunit of cyclic adenosine monophosphate-dependent protein
kinase. Science, 253, 414–420.
38. Braun, S., Raymond, W.E. and Racker, E. (1984) Synthetic tyrosine
polymers as substrates and inhibitors of tyrosine-specific protein kinases.
J. Biol. Chem., 259, 2051–2054.
39. Kemp, B.E., Cheng, H.C. and Walsh, D.A. (1988) Peptide inhibitors of
cAMP-dependent protein kinase. Methods Enzymol., 159, 173–183.
40. Manning, B.D. and Cantley, L.C. (2002) Hitting the target: emerging
technologies in the search for kinase substrates. Sci. STKE, 2002, PE49.
41. Chong, Y.P., Ia, K.K., Mulhern, T.D. and Cheng, H.C. (2005) Endogenous
and synthetic inhibitors of the Src-family protein tyrosine kinases.
Biochim. Biophys. Acta, 1754, 210–220.
Human Molecular Genetics, 2006, Vol. 15, No. 213261
by guest on May 30, 2013
42. Sotirellis,N.,Johnson,T.M.,Hibbs,M.L.,Stanley,I.J.,Stanley,E.,Dunn,A.R. Download full-text
and Cheng, H.C. (1995) Autophosphorylation induces autoactivation and
a decrease in the Src homology 2 domain accessibility of the Lyn protein
kinase. J. Biol. Chem., 270, 29773–29780.
43. Herrmann, J.M. and Neupert, W. (2000) Protein transport into
mitochondria. Curr. Opin. Microbiol., 3, 210–214.
44. Wang, D., Qian, L., Xiong, H., Liu, J., Neckameyer, W.S., Oldham, S.,
Xia, K., Wang, J., Bodmer, R. and Zhang, Z. (2006) Antioxidants
protect PINK1-dependent dopaminergic neurons in Drosophila.
Proc. Natl Acad. Sci. USA, 103, 13520–13525.
45. Chong, Y.P., Chan, A.S., Chan, K.C., Williamson, N.A., Lerner,
E.C., Smithgall, T.E., Bjorge, J.D., Fujita, D.J., Purcell, A.W.,
Scholz, G. et al. (2006) C-terminal SRC kinase-homologous kinase
(CHK)—a unique inhibitor inactivating multiple active
conformations of SRC-family tyrosine kinases. J. Biol. Chem. Epub
ahead of print.
46. Boyle, W.J., van der Geer, P. and Hunter, T. (1991) Phosphopeptide
mapping and phosphoamino acid analysis by two-dimensional
separation on thin-layer cellulose plates. Methods Enzymol., 201,
3262 Human Molecular Genetics, 2006, Vol. 15, No. 21
by guest on May 30, 2013