1-Benzyl-1,2,3,4-tetrahydroisoquinoline binds with tubulin β, a substrate of parkin, and reduces its polyubiquitination

Article abstract of DOI:10.1111/j.1471-4159.2010.06576.x

Substances that mimic the actions of causative gene products of familial Parkinson's disease (PD) are candidate as causative agents of idiopathic PD. 1-Benzyl-1,2,3,4-tetrahydroisoquinoline (1BnTIQ), an endogenous neurotoxin, is present at three times hig

Full text of DOI:10.1111/j.1471-4159.2010.06576.x

JOURNAL OF NEUROCHEMISTRY
|
2010
|
114
|
1291–1301
doi: 10.1111/j.1471-4159.2010.06576.x
, ,
,
,
,
*Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan
Graduate School of Medicine, Gifu University, Gifu, Japan
àGraduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, Yokohama, Japan
§Graduate School of Biomedical Science, Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Tokyo,
Japan
¶Pro Phoenix, Towa Environment Science, Osaka, Japan
**RIKEN Center for Molecular Imaging Science, Kobe, Japan
United Graduate School of Drug Discovery and Medical Information Sciences, Gifu University, Gifu, Japan
Abstract
mutation of parkin is reported to result in loss of tubulin b
ubiquitination. Therefore, we examined the effect of 1BnTIQ
on ubiquitination of tubulin b. The polyubiquitinated tubulin b
level in human neuroblastoma SH-SY5Y cells was reduced in
the presence of 1BnTIQ, even at concentrations as low as
those detected in parkinsonian CSF. In vitro ubiquitination
assay gave similar results. It is suggested that 1BnTIQ has the
same effect on tubulin ubiquitination as does mutant parkin in
familial PD. Taken together, substances which reduce poly-
ubiquitination of tubulin such as 1BnTIQ are supposed to be
candidates of etiological factors of PD.
Substances that mimic the actions of causative gene products
of familial Parkinson’s disease (PD) are candidate as causa-
tive agents of idiopathic PD. 1-Benzyl-1,2,3,4-tetrahydroiso-
quinoline (1BnTIQ), an endogenous neurotoxin, is present at
three times higher levels in CSF of PD patients than in CSF of
control subjects. However, the mechanism of 1BnTIQ’s neu-
rotoxicity is unclear. In this study, we tried to identify 1BnTIQ-
binding proteins by using a diazido-functionalized 1BnTIQ
analog, 1-(3-azido-5-azidomethylbenzyl)-1,2,3,4-tetrahydro-
isoquinoline, designed and synthesized as a probe for radio-
isotope-free photoaffinity labeling. One major photolabeled
protein identified using this probe was tubulin b, which has
been reported to be a substrate of parkin, a ubiquitin E3 ligase
and a causative gene product of familial PD. Loss of function
Keywords: parkin, Parkinson’s disease, polyubiquitination,
radioisotope-free photoaffinity labeling, tetrahydroisoquino-
line, tubulin.
J. Neurochem. (2010) 114, 1291–1301.
Parkinson’s disease (PD) is an age-related neurodegenerative
disorder characterized by progressive loss of dopaminergic
neurons and the appearance of Lewy bodies in the substantia
nigra, leading to clinical symptoms of rigidity, resting tremor,
and bradykinesia. A few percent of PD is of genetic origin, and
over ten genes have been identified as causes of familial PD so
Abbreviations used: 1BnTIQ, 1-Benzyl-1,2,3,4-tetrahydroisoquino-
line; 1DAzBnTIQ, 1-(3-azido-5-azidomethylbenzyl)-1,2,3,4-tetrahydro-
isoquinoline; 1MeTIQ, 1-methyl-1,2,3,4-tetrahydroisoquinoline; 3¢,4¢
DHBnTIQ, 1-(3,4-dihydroxybenzyl)-1,2,3,4-tetrahydroisoquinoline; AR-
JP, autosomal recessive juvenile parkinsonism; BSA, bovine serum albu-
min; CBB, Coomassie Brilliant Blue; GST, glutathione-S-transferase; IP,
immunoprecipitation; N-Boc, N-tert-butoxycarbonyl; Pael-R, parkin-
associated endothelin receptor-like receptor; PBS, phosphate-buffered
saline; PD, Parkinson’s disease; PVDF, polyvinylidene difluoride; SDS–
PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; TIQ,
1,2,3,4-tetrahydroisoquinoline; TNE, tris-NaCl-EDTA; TOF, time of
flight; Tris, tris(hydroxymethyl)aminomethane; TTBS, Tween 20-con-
taining tris-buffered saline; WB, western blotting.
Received October 9, 2009; revised manuscript received November 27,
2009; accepted December 1, 2009.
Address correspondence and reprint requests to Dr Y. Kotake,
Graduate School of Biomedical Sciences, Hiroshima University, 1-2-3
Kasumi, Minami-ku, Hiroshima 734-8553, Japan.
E-mail: yaichiro@hiroshima-u.ac.jp
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301
1291

1292
| R. Kohta et al.
Association for Laboratory Animal Science. In the presence or
absence of 1BnTIQ or other TIQ analogs (50 mM), 1DAzBnTIQ
(50 lM) was incubated with each subcellular fraction, i.e., nuclear
fraction (800 g precipitate), mitochondrial fraction (9000 g precip-
itate), microsomal fraction (105 000 g precipitate), or cytosol
fraction (105 000 g supernatant), prepared from rat whole brain
homogenate, in 0.1 M phosphate buffer (pH 7.4) at 30ꢁC for
10 min. Each fraction placed in a 1.5 mL polypropylene tube was
exposed to 254 nm UV light [SVL-6, Vilber Lourmat Lamp (Vilber
Lourmat, Marne-la-Valle´e, France), 4 W] from a distance of 5 cm
for 2 min, followed by incubation with a fluorescent triarylphos-
phine derivative, GIF-0373 (50 lM), at 30ꢁC for 4 h (Hosoya et al.
2004, 2005). This step specifically converts the photolabeled
proteins to fluorescent products. The fluorescent proteins were
detected with a luminescence image analyzer (LAS-1000; Fuji
Photo Film Co. Ltd., Tokyo, Japan) after sodium dodecyl sulfate–
polyacrylamide gel electrophoresis (SDS–PAGE) or 2-dimensional
electrophoresis, and their primary structures were identified by
matrix-assisted laser desorption-time of flight (TOF)/TOF mass
spectroscopy as described in the Appendix S1.
far, including a-synuclein (PARK1), DJ-1 (PARK7) and parkin
(PARK2) (Dawson and Dawson 2003; Thomas and Beal 2007).
Parkin is a ubiquitin E3 ligase which ubiquitinates tubulin a/
b, synphilin-1, synaptotagmin XI, and other proteins via two
really interesting new gene finger domains (Chung et al. 2001;
Huynh et al. 2003; Ren et al. 2003). It is recognized as the
causative gene product of autosomal recessive juvenile
parkinsonism (AR-JP), a type of familial PD, and biochemical
analysis has shown that parkin mutation is associated with
impaired ubiquitination of the above substrates. Parkin null
micehaveanambiguousphenotype, showingabnormalrelease
andmetabolism ofdopamineand severedopaminergiccellloss
with aging (Goldberg et al. 2003; Itier et al. 2003; von Coelln
et al. 2004;PerezandPalmiter2005). Studiesongeneproducts
linked to AR-JP are expected to be helpful for under-
standing the mechanism of neurodegeneration in PD patients.
However, in the case of idiopathic PD (accounting for over
90% percent of PD cases), it is believed that the etiology
involves a combination of genetic predisposition and
environmental factors. It has been suggested that endogenous
or exogenous environmental dopaminergic neurotoxins, such
as 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)
and rotenone, play a role in the pathogenesis of idiopathic
PD (Langston et al. 1983; Nagatsu and Yoshida 1988;
Betarbet et al. 2000, 2006). Candidate toxicological mech-
anisms include inhibition of mitochondrial complex I,
impairment of proteasome activity, and other pathways.
1-Benzyl-1,2,3,4-tetrahydroisoquinoline (1BnTIQ) is an
endogenous neurotoxin that has been proposed as one of the
etiological factors of idiopathic PD (Kotake et al. 1995). The
level of 1BnTIQ in CSF of patients with idiopathic PD was
found to be three times higher than that in CSF of
neurological control subjects (Kotake et al. 1995). Further,
chronic administration of 1BnTIQ induces parkinsonism in
mice and primates (Kotake et al. 1995, 1996). 1BnTIQ is
toxic to human SH-SY5Y neuroblastoma cells and cultured
primary neurons (Kotake et al. 2003; Shavali and Ebadi
2003; Shavali et al. 2004; Wasik et al. 2009).
Two-dimensional electrophoresis
The first electrophoresis was performed using a 7 cm pH 4–7
Immobiline DrystripTM (GE Healthcare Ltd., Buckinghamshire,
UK). Destreak rehydration solution (GE Healthcare Ltd.) diluted
with 0.5% immobilized pH gradient buffer pH 4–7 (GE Healthcare
Ltd.) was co-incubated with the fluorescent lysate for 30 min at 15–
25ꢁC. This solution was used to swell the drystrip overnight.
Isoelectric focusing was performed using a Multiphor II isoelectric
focusing Unit (GE Healthcare Ltd.), initially with a gradient from
200 V to 3500 V for 1.5 h, and then at 3500 V for 1.5 h. The
drystrip was equilibrated with 10 mg/mL dithiothreitol (Nacalai
Tesque Inc., Kyoto, Japan) in SDS-equilibration buffer – 6 M urea
(Nacalai Tesque Inc.), 30% (v/v) glycerol (Kanto Chemical Co. Inc.,
Tokyo, Japan), 2% (w/v) SDS (Nacalai Tesque Inc.), bromophenol
blue (0.002%; Nacalai Tesque Inc.), 50 mM tris(hydroxymeth-
yl)aminomethane (Tris; Nacalai Tesque Inc.) – HCl pH 8.8 for
20 min. The immobilized pH gradient strips (GE Healthcare Ltd.)
were placed on 12% SDS–polyacrylamide gel (90 · 80 · 1 mm)
and sealed with 0.5% (w/v) agarose (Nacalai Tesque Inc.). The gels
were run at 15 mA/gel for 15 min for initial migration. The current
was then increased to 30 mA/gel for separation until the dye reached
the bottom of the gel. The gels were stained with Coomassie
Brilliant Blue (CBB) using Rapid CBB KANTO reagent (Kanto
Chemical Co. Inc.).
In this study, to elucidate the molecular action of 1BnTIQ,
we searched for binding proteins, using a diazido-function-
alized 1BnTIQ analog, 1-(3-azido-5-azidomethylbenzyl)-
1,2,3,4-tetrahydroisoquinoline (1DAzBnTIQ), as a probe
for radioisotope-free photoaffinity labeling (Hosoya et al.
2004, 2005). We identified tubulin b as a candidate 1BnTIQ-
binding protein. Based on this result, we investigated the
effect of 1BnTIQ on the level of polyubiquitinated tubulin b
in SH-SY5Y human neuroblastoma cells and in in vitro.
1BnTIQ treatment and sample preparation
After SH-SY5Y cells were treated with 1BnTIQ (1 and 10 lM) for
48 h, cells were lysed and collected for immunoprecipitation (IP)–
western blotting (WB) experiment (Figs 3a and 4). In another
experiment, cells were treated with a low concentration of 1BnTIQ
(25 and 50 nM) for 1 month. Cells were dispersed after weekly
passage. Medium change was performed every 2 or 3 days. Cells
were lysed and collected with 0.1% tris-NaCl-EDTA (TNE) buffer,
pH 7.4 [20 mM Tris-HCl, 150 mM NaCl (Wako Pure Chemical
Industries Ltd., Osaka, Japan), 2 mM EDTA (Nacalai Tesque Inc.)],
0.1% Igepal CA630 (Sigma-Aldrich Inc., St. Louis, MO, USA)
containing protease inhibitor cocktail (Nacalai Tesque Inc.), 50 mM
Materials and methods
Photoaffinity labeling of target proteins with 1DAzBnTIQ
The animals were handled in accordance with the guidelines for the
care and use of experimental animals published by the Japanese
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

Reduction of Ub-tubulin by 1BnTIQ
| 1293
NaF (Nacalai Chemicals Ltd.), and 1 mM Na3VO4 (Wako Pure
Chemical Industries Ltd.) for IP–WB experiment (Fig. 4).
derivative, GIF-0373 (50 lM), at 30ꢁC for 4 h as described above.
These fluorescent proteins were subjected to SDS–PAGE, trans-
ferred to PVDF membrane (Immobilon-FLꢂ; Millipore Ltd.,
Bedford, MA, USA) and quantitated with a luminescence image
analyzer (LAS-1000; Fuji Photo Film Co. Ltd.). After washing each
PVDF membrane with TTBS buffer for 20 min thrice, colloidal gold
protein staining of each PVDF membrane was performed by
incubation with colloidal gold staining solution (Bio-Rad Labora-
tories Inc.) for a sufficient time to obtain bands. The bands were
detected by above described analyzer as well.
Immunoprecipitation
Immunoprecipitation was performed using monoclonal tubulin b
antibody (TUB 2.1; Sigma-Aldrich Inc.). Protein G Sepharose Fast
Flow beads (GE Healthcare Ltd.) were washed thrice with
phosphate-buffered saline (PBS)(-) and suspended in 20 lL of
50% slurry in PBS-containing protease inhibitor cocktail (Nacalai
Tesque Inc.). The antibody was bound to protein G Sepharose beads
with the aid of rotation at 4ꢁC for 1 h. After the beads were washed
with 0.1% TNE buffer, the beads (20 lL volume) were rotated with
brain homogenate or SH-SY5Y lysate (300 lg) in 0.1% TNE buffer
at 4ꢁC. Subsequent to wash the beads at 4ꢁC three times, elution was
performed using Laemmli sample buffer at 100ꢁC for 5 min. The
supernatant was subjected to SDS–PAGE.
In vitro ubiquitination assay of tubulin
In vitro ubiquitination assay was performed using a commercially
available ubiquitinylation assay kit (UW9920; Biomol International
L.P.). E1 (0.1 lM), UbcH5c (E2, 50 lg/mL), maltose binding
protein-parkin (E3, 0.1 lM), tubulin (1 lM; Cytoskeleton Inc.,
Denver, CO. USA), biotinylated ubiquitin (2.5 lM), Mg-ATP
(50 mM), dithiothreitol (1 mM), inorganic pyrophosphatase (50 U/
mL; Sigma-Aldrich Inc.), and 1BnTIQ (10 lM) were mixed on ice.
The mixture was incubated at 37ꢁC for 1 h and added to non-reducing
gel loading buffer. The sample mixture was subjected to western
blotting, and ubiquitinated proteins were detected with streptavidin-
conjugated horseradish peroxidase and Vectastain Elite ABC standard
kitꢂ (Vector Laboratories Inc, Burlingame, CA, USA).
Western blotting
After having been exposed to test compounds, the SH-SY5Y cells
were washed with PBS(-), scraped off, and transferred to 1.5 mL
tubes. Lysis was done with 0.1% TNE buffer-containing protease
inhibitor cocktail at 4ꢁC for 30 min. The loading samples were
obtained from supernatant by centrifugation at 800 g as a loading
sample. SDS–PAGE (8% acrylamide gel (Kanto Chemical Co. Inc.)
was run at 15 mA/gel for 20 min in running buffer [14.4 g/L glycine
(Nacalai Tesque Inc.), 3.0 g/L Tris, 1.0 g/L SDS]. The current was
then increased to 30 mA/gel for separation until the dye reached the
bottom of the gel. The polyacrylamide gel was transferred to Immun-
Blotꢂ polyvinylidene difluoride (PVDF) membrane (Bio-Rad Lab-
oratories Inc, Hercules, CA, USA) at 100 mA/cm2 for 2 h in transfer
buffer (14.4 g/L glycine, 3.0 g/L Tris, 7.5% (v/v) methanol; Wako
Pure Chemical Industries Ltd.). Blocking was performed with
blocking buffer containing 2% or 5% skim milk (Wako Pure
Chemical Industries Ltd.) in Tween 20-containing tris-buffered saline
(TTBS) buffer (50 mM Tris-HCl pH 7.6, 150 mM NaCl, 0.1% (v/v)
Tween 20; Nacalai Tesque Inc.) at 15–25ꢁC for 30 min. Next, the
membrane was incubated with 2% or 5% skim milk containing
primary antibody against b-actin (AC-15; Sigma-Aldrich Inc.),
polyubiquitinated proteins (FK1; Biomol International L.P., Ply-
mouth Meeting, PA, USA), and ubiquitin (sc-8017; Santa Cruz
Biotechnology Inc, Santa Cruz, CA, USA) in blocking buffer at 4ꢁC
overnight. Next day, the PVDF membrane was washed with TTBS
buffer at 15–25ꢁC for 8 min three times at 15–25ꢁC. Horseradish
peroxidase-conjugated secondary antibodies (Sigma-Aldrich Inc., St.
Louis, MO, USA) in blocking buffer were used to detect ECL plus
solutions (GE Healthcare Ltd.). For densitometric analysis, images
were scanned and quantitated using the LAS-1000 software (Scion
Co. Ltd., Frederick, MD, USA).
Fluorescence imaging of 1DAzBnTIQ and tubulin b in SH-SY5Y
cells
SH-SY5Y cells were dispersed at 1 · 105/well on a glass plate in
which above described 24-well plate was placed. After 24 h, the cells
were washed with PBS(-) thrice and treated with 4% paraformalde-
hyde in PBS(-) for 10 min. After fixing, cells were treated with 0.1%
Triton-X 100 for 5 min, and then washed with PBS(-) thirce. The cells
were incubated with 1 lM 1DAzBnTIQ in the presence or absence of
10 mM 1BnTIQ in PBS(-) containing 1% bovine serum albumin
(BSA; Intergen Co., Purchase, NY, USA) for 1 h. They were then
irradiated with UV light (Vilcer Lourmat, SVL, 4W) of 254 nm
wavelength for 10 s at a distance of 5 cm, washed with PBS(-) three
times, incubated with GIF-0373 (1 lM) in PBS(-) containing 1%
BSA at 15–25ꢁC for 4 h, washed with PBS(-) three times, and
incubated with mouse monoclonal anti-tubulin b antibody (T4026;
Sigma-Aldrich Inc.) in PBS(-) containing 1% BSA overnight. The
cells were washed with PBS(-) three times and incubated with Alexa
Fluorꢂ 555 goat anti-mouse IgG antibody (A31261, Invitrogen Ltd.)
and 500 nM 4¢,6-diamino-2-phenylindole, Invitrogen Ltd.) in PBS(-)
for 1 min. The cells were again washed with PBS(-) five times, and the
glass plate was covered with ProLongꢂGold antifade reagent
(Invitrogen Ltd.). Finally, the fluorescence was detected with a
confocal laser scanning microscope (LSM5 PASCAL Apotome; Carl
Zeiss Inc., Go¨ttingen, Germany).
Photolabeling study of glutathione-S-transferase (GST)-tagged
human recombinant tubulin b, isoforms with 1DAzBnTIQ
Each GST tagged human recombinant tubulin b subtype (1 lg;
Abcam plc., Cambridge, UK) was incubated with 1DAzBnTIQ
(50 lM) in the presence or absence of 1BnTIQ (50 mM) in 0.1 M
phosphate buffer (pH 7.4) at 30ꢁC for 10 min. Each sample, placed
in a 1.5 mL polypropylene tube, was exposed to 254 nm UV light
(SVL-6, Vilber Lourmat Lamp, 4 W) from a distance of 5 cm for
2 min, followed by incubation with a fluorescent triarylphosphine
Results
Design and synthesis of the photoaffinity probe,
1DAzBnTIQ
To elucidate the molecular mechanism of neurotoxicity
induced by endogenous 1BnTIQ, we planned to identify the
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

1294
| R. Kohta et al.
target protein of 1BnTIQ by applying a radioisotope-free
photoaffinity labeling technique, which allows direct analysis
of photolabeled proteins (Hosoya et al. 2004, 2005). Based
on this strategy, we designed a diazido-functionalized
1BnTIQ analog, 1DAzBnTIQ, as a candidate photoaffinity
probe (Fig. 1). This compound possesses two different kinds
of azido groups, a photoreactive aromatic one and a
relatively photo-stable aliphatic one, on the benzyl moiety
of 1BnTIQ. It was anticipated that the bioactivity of the
designed photoaffinity probe would be retained, because
introduction of azido and azidomethyl groups is a relatively
slight modification. This probe is expected to capture
specifically its target protein by covalent bonding via
photoreaction of the aromatic azido group (Fig. 1d). The
photolabeled protein can be specifically modified to a
detectable form by azido-selective reaction, such as Stau-
dinger–Bertozzi ligation (Saxon and Bertozzi 2000; Kiick
et al. 2002) or click chemistry (Rostovtsev et al. 2002;
(a)
(b)
(c)
(d)
Fig. 1 Our strategy to identify 1-benzyl-1,2,3,4-tetrahydroisoquinoline
(1BnTIQ) binding proteins using the diazido-functionalized 1BnTIQ
analog, 1-(3-azido-5-azidomethylbenzyl)-1,2,3,4-tetrahydroisoquino-
line (1DAzBnTIQ), as a radioisotope-free photoaffinity probe. (a)
Structure of 1BnTIQ. (b) Synthetic scheme of 1DAzBnTIQ. (c) Neuro-
toxicity of 1DAzBnTIQ. After 24 h exposure of cells to 1DAzBnTIQ, 2-
(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium,
monosodium salt solution was added for 3 h. The absorption of gen-
erated formazan was measured at 415 nm. The experiments were
repeatedly performed and representative data are shown. Data are
expressed as mean + SD. (d) Prospective scheme of radioisotope-free
photoaffinity labeling using 1DAzBnTIQ. The first step is the formation
of a covalent bond with the target proteins via the photoreactive aro-
matic azido group (red) by exposure to UV light at 254 nm. The second
step is specific fluoresceination of photolabeled proteins through the
remaining aliphatic azido group (blue) by Staudinger–Bertozzi ligation
using a fluorescein–triarylphospine derivative, GIF-0373. The target
proteins were separated by sodium dodecyl sulfate–polyacrylamide gel
electrophoresis or 2-dimensional electrophoresis, and their primary
structures were identified by the MS/MS ion search method using
matrix-assisted laser desorption-time of flight (TOF)/TOF mass spec-
troscopy and MASCOT analysis.
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

Reduction of Ub-tubulin by 1BnTIQ
| 1295
Tornøe et al. 2002), using the remaining aliphatic azido
group of the probe. Indeed, Staudinger–Bertozzi ligation
with a fluorescein–triarylphosphine derivative, GIF-0373,
was effectively utilized to detect photolabeled proteins in
previous studies (Hosoya et al. 2004, 2005).
Identification of 1BnTIQ binding proteins
With a promising photoaffinity probe 1DAzBnTIQ in hand,
we performed photoaffinity labeling experiments (Fig. 1d).
Rat whole brain homogenate was prepared, and subcellular
fractions were obtained by centrifugation. Each homogenate
fractions were incubated with 1DAzBnTIQ and photo-
irradiated with UV light (254 nm) to activate the aromatic
azido group and covalently bind the probe to its target
proteins. Next, the fluorescein–triarylphosphine derivative,
GIF-0373, was added to specifically modify the photolabeled
proteins with fluorescein through the remaining aliphatic
azido group of the probe. As a result, a fluorescent protein of
approximately 50 kDa, was detected in the cytosolic fraction
(Fig. 2a, lane 4). In the presence of an excess of 1BnTIQ, this
fluorescent band disappeared (Fig. 2a, lane 5). Therefore, this
1DAzBnTIQ-binding protein of 50 kDa is a candidate for the
target protein of 1BnTIQ. To analyze what structural features
of 1BnTIQ are essential to recognize the 50 kDa protein, we
performed photolabeling experiments in the presence of other
TIQ analogs: 1,2,3,4-tetrahydroisoquinoline (TIQ), 1-methyl-
1,2,3,4-tetrahydroisoquinoline (1MeTIQ), and 1-(3,4-dihydr-
oxybenzyl)-1,2,3,4-tetrahydroisoquinoline (3¢,4¢DHBnTIQ).
Co-treatment with 1BnTIQ or 3¢,4¢DHBnTIQ blocked 1DAz-
BnTIQ binding to the 50 kDa protein (Fig. 2b, lanes 2 and 4).
On the other hand, TIQ and 1MeTIQ only slightly weakened
the fluorescent band derived from 1DAzBnTIQ (Fig. 2b,
lanes 3 and 4). These results suggest that the benzyl moiety is
essential for binding to the 50 kDa protein.
Syntheses of TIQ analogs have so far been performed by
utilizing the Bischler–Napieralski reaction, which is an
intramolecular condensation of N-phenethylacylamide to
give cyclized 3,4-dihydroisoquinoline, followed by reduc-
tion of the carbon–nitrogen double bond by hydrogenation
or with hydride reagents, such as sodium borohydride
(Kotake et al. 1995; Kawai et al. 2000). If this method was
applicable for the synthesis of 1DAzBnTIQ, we would
simply have to prepare 2-(3-azido-5-(azidomethyl)phenyl)-
N-phenethylacetamide and carry out the two-step procedure
shown above. This linear route, however, was obviously
unsuitable because the reaction conditions of Bischler–
Napieralski reaction (usually heating with phosphorus
oxychloride) and also of imine reduction, are too severe
to leave the azido groups intact. Introduction of azido
groups at late stages of the synthesis after these reactions
may avoid this problem, but apparently time-consuming
steps are required (Nikulin et al. 2006). We therefore
sought for an alternative route and succeeded in developing
a quite straightforward route as shown in Fig. 1b. The key
step is Suzuki–Miyaura cross-coupling reaction of known
1-azido-3-azidomethyl-5-iodobenzene (3) (Hosoya et al.
2009) with B-alkyl-9-borabicyclo[3.3.1]nonane (Miyaura
et al. 1989; Chemler et al. 2001; Vice et al. 2001),
prepared in situ by hydroboration of N-tert-butoxycarbonyl
(N-Boc)-protected 1-methylene-1,2,3,4-tetrahydroisoquino-
line 2 with 9-borabicyclo[3.3.1]nonane, which afforded N-
Boc-protected 1DAzBnTIQ 4. Finally, deprotection of the
N-Boc group of 4 under acidic conditions yielded the
desired 1DAzBnTIQ. Compound 2 can be easily prepared
from commercially available starting material, 1-methyl-3,4-
We next performed 2-dimensional electrophoresis to
separate this protein more precisely, and obtained broad
fluorescent spots at 50 kDa and pI 4–5 (Fig. 2c). The CBB-
stained spots corresponding to the fluorescent spots were
analyzed by mass/mass (MS/MS) ion search using matrix-
assisted laser desorption-TOF/TOF mass spectroscopy
(Fig. 2d), and spots 1 and 2 were identified as tubulin b
and tubulin a, respectively, from the MASCOT analysis
(Table S1).
dihydroisoquinoline hydrochloride (1), according to
a
similar method to that previously reported (Lenz et al.
1982; Lenz and Lessor 1992). Although we have not
optimized the reaction conditions for the cross-coupling
step, this three-step convergent route provides a novel
general method for synthesizing TIQ analogs and related
compounds.
Confirmation of 1DAzBnTIQ-binding to tubulin b
To confirm whether 1DAzBnTIQ actually binds to tubulin a
or b, we applied IP method to the photolabeled fluorescent
proteins. As the results, the fluorescence of the tubulin b
antibody eluate was stronger than that in the case of tubulin a
(Fig. 3a). These results indicate that 1DAzBnTIQ interacts
with a/b heterodimers and efficiently binds covalently with
tubulin b upon photo-irradiation.
Since the 1BnTIQ analog, 1DAzBnTIQ, binds with
tubulin b, we analyzed the effect of 1BnTIQ on microtubule
polymerization. It has been reported that 1-methyl-4-
phenylpyridinium MPP+, a PD-related neurotoxin, binds
with microtubules and inhibits microtubule polymerization
(Cappelletti et al. 1999). Indeed, MPP+ inhibited microtu-
bule polymerization at 2 mM. 1BnTIQ also inhibited micro-
tubule polymerization at 1 and 2 mM (data not shown).
Neurotoxicity of 1DAzBnTIQ in SH-SY5Y cells
To examine whether 1DAzBnTIQ shows similar neurotox-
icity to 1BnTIQ in SH-SY5Y human neuroblastoma cells, we
performed 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulf-
ophenyl)-2H-tetrazolium, monosodium salt assay. 1DAzBn-
TIQ reduced the viability of SH-SY5Y cells almost to the
same degree with 1BnTIQ as reported previously (Fig. 1c).
This result indicates that 1DAzBnTIQ is a promising
candidate for a probe to photolabel specifically the target
protein of 1BnTIQ.
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

1296
|
R. Kohta et al.
(a)
(b)
(c)
(d)
Fig. 2 1-(3-Azido-5-azidomethylbenzyl)-1,2,3,4-tetrahydroisoquino-
line (1DAzBnTIQ)-binding protein of approximately 50 kDa is present in
cytosolic fraction of rat brain. (a) Photolabeling experiments with
1DAzBnTIQ (50 lM) were performed using fractionated samples of rat
whole brain homogenate, followed by specific fluoresceination of
photolabeled proteins with GIF-0373 (50 lM). The fluorescence signals
were detected with a luminescence image analyzer (LAS-1000; Fuji
Photo Film Co. Ltd.) after sodium dodecyl sulfate–polyacrylamide gel
electrophoresis. Lane 1: nuclear fraction, lane 2: mitochondrial fraction,
lane 3: microsomal fraction, lane 4: cytosolic fraction, lane 5: cytosolic
fraction + high conc. of 1-benzyl-1,2,3,4-tetrahydroisoquinoline
(1BnTIQ). The strongest fluorescent band at 50 kDa was detected in the
cytosolic fraction (lane 4), and it disappeared in the presence of 50 mM
1BnTIQ (lane 5). (b) The effects of some TIQ derivatives (50 mM) on
1DAzBnTIQ binding to the 50 kDa protein were examined. The signal of
50 kDa 1DAzBnTIQ-binding protein (lane 1) was considerably de-
creased by co-incubation with 1BnTIQ (lane 2) and 3¢,4¢DHBnTIQ (lane
4), but not so much with TIQ (lane 3) or 1MeTIQ (lane 5). ** and free
indicate 1DAzBnTIQ-binding protein and remaining fluorescent re-
agent, GIF-0373, respectively. (c) Separation of 1DAzBnTIQ-binding
proteins using 2-dimensional electrophoresis. Fluorescence imaging
with LAS-1000 (Fuji Photo Film Co. Ltd.) of 2-dimensional electropho-
retically separated rat whole brain homogenate, photolabeled with and
fluorescenated by 1DAzBnTIQ and GIF-0373, respectively. (d) The gel
stained with Coomassie Brilliant Blue. The arrowhead indicates two
major 1DAzBnTIQ-binding proteins of approximately 50 kDa in molec-
ular weight and with pI 4–5. All the experiments were repeatedly per-
formed and representative data are shown.
Separately, we also examined which subtype of tubulin b is
photolabeled with 1DAzBnTIQ using the isotypes of tubulin
b fused with GST. It has been reported that there are seven
isoforms for tubulin b (Sullivan and Cleveland 1986; Lopata
and Cleveland 1987). In neurons of the human brain, tubulin
b3 is highly expressed (Woulfe and Munoz 2000). We
prepared four isoforms of tubulin b1, b2, b3, and b4a, fused
with GST and performed photolabeling study with 1DAz-
BnTIQ. Among these recombinant tubulin b isoforms,
1DAzBnTIQ bound more strongly with tubulin b2 or b4a
(Fig. 3b), whereas binding with tubulin b3 was observed in
SH-SY5Y lysate (Fig. S1).
To check the behavior of 1DAzBnTIQ in SH-SY5Y
cells, we performed immunocytochemistry. 1DAzBnTIQ/
GIF-0373-derived fluorescence, as well as tubulin b, was
detected in whole cytoplasm (Fig. 3c), supporting the
conclusion that 1DAzBnTIQ binds to tubulin b. On the
other hand, co-incubation with high concentration of
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

Reduction of Ub-tubulin by 1BnTIQ
| 1297
(a)
(b)
Lanes:
1
+
2
+
3
+
4
–
5
–
Whole brain cytosol (WBC):
kDa
GST-tubulin β3 GST-tubulin β4a
250-
150-
100-
75-
GST-tubulin β1 GST-tubulin β2
GST
kDa
116
1BnTIQ
1BnTIQ
1BnTIQ
–
1BnTIQ
1BnTIQ
kDa
116
–
+
–
+
+
–
+
–
+
Fluorescent
detection
66
66
kDa
97
kDa
97
50-
Colloidal gold
protein staining
66
66
IP:
α
β
-
α
β
(c)
Fig. 3 Confirmation of 1-(3-azido-5-azidomethylbenzyl)-1,2,3,4-tetra-
hydroisoquinoline (1DAzBnTIQ) binding to tubulin b by (a) IP, (b)
photolabeling study of recombinant tubulin b isoforms, (c) fluores-
cence imaging in SH-SY5Y cells. (a) 1DAzBnTIQ binds preferentially
with tubulin b rather than tubulin a. In the first step, IP was performed
using rat whole brain cytosol reacted with 1DAzBnTIQ and GIF-0373,
using monoclonal anti-tubulin b antibody. Immunoprecipitated eluates
were subjected to sodium dodecyl sulfate–polyacrylamide gel elec-
trophoresis. In the second step, bands were transferred to poly-
vinylidene difluoride membrane. Finally, the 1DAzBnTIQ-bound
fluorescent protein was detected with LAS-1000 (Fuji Photo Film Co.
Ltd.). (b) 1DAzBnTIQ binds preferentially with tubulin b2 and b4a
among the four subtypes of tubulin b examined. 1DAzBnTIQ was co-
incubated with recombinant tubulin b instead of whole brain cytosol in
(a). All recombinant proteins are glutathione-S-transferase-tagged
tubulin b. (c) Immunocytochemistry of 1DAzBnTIQ-bound protein and
tubulin b was performed using mouse monoclonal tubulin b antibody.
1DAzBnTIQ-bound proteins co-localize with tubulin b. Tubulin b (red),
1DAzBnTIQ-bound proteins (green), nuclear staining with 4¢,6-diami-
no-2-phenylindole (blue). A high concentration of 1-benzyl-1,2,3,4-
tetrahydroisoquinoline (1BnTIQ) reduced the 1DAzBnTIQ/GIF-0373-
derived green signal. All the experiments were repeatedly performed
and representative data are shown. Bars indicate 20 lm.
1BnTIQ (10 mM) reduced the 1DAzBnTIQ/GIF-0373-
dependent fluorescence. These results suggest that 1BnTIQ
associates with tubulin b, a component of microtubules, in
this neuronal cell line.
IP and WB (Fig. 4a). These results suggest that 1BnTIQ
may decrease the level of polyubiquitinated tubulin b in
human dopaminergic cell lines.
Reduction of polyubiquitinated tubulin b in the presence of
proteasome inhibitor
Reduction of polyubiquitinated tubulin b by 1BnTIQ
in SH-SY5Y cells
To examine the mechanism of the reduction of polyubiq-
uitinated tubulin b induced by 1BnTIQ, we analyzed the
effect of 1BnTIQ on the amount of polyubiquitinated
tubulin b in the presence of a proteasome inhibitor,
MG-132. 1BnTIQ efficiently reduced the level of polyubiq-
uitinated tubulin b in the presence of 1 lM MG-132
(Fig 4b). An immunocytochemical examination indicated
that 1BnTIQ weakly reduced total polyubiquitinated pro-
teins (not polyubiquitinated tubulin b) accumulated in the
presence of 1 lM MG-132 (Fig. S2). Further, 1BnTIQ at
concentrations of less than 100 lM did not alter the
proteasome activity in SH-SY5Y cells (data not shown).
These results indicate that 1BnTIQ reduces polyubiquitinat-
ed tubulin b without increasing proteasome function.
Parkin, a member of the E3 ubiquitin ligase family, has been
recognized as the etiological agent of AR-JP (Kitada et al.
1998). Therefore, it is possible that loss of function arose by
knocking it out would result in impaired ubiquitination of
tubulin. On the other hand, over-expression of parkin is
known to induce polyubiquitination of tubulin b, while
over-expression of mutant parkin does not (Ren et al. 2003).
We expected that 1BnTIQ would also impair polyubiquiti-
nation of tubulin b, so we examined the effect of 1BnTIQ
on ubiquitination of tubulin b in SH-SY5Y cells for 48 h. In
the presence of 1BnTIQ (1 and 10 lM), the signal of
polyubiquitinated tubulin b detected at 100 kDa was
reduced, but that at 75 kDa was not reduced by means of
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

1298
| R. Kohta et al.
1BnTIQ, not only strong monoubiquitination, but also weak
polyubiquitination of tubulin was induced (Fig. 4c). How-
ever, 1BnTIQ reduced parkin-dependent polyubiquitination
of tubulins (Fig. 4c). On the other hand, T240R parkin
mutant also showed reduced polyubiquitination of tubulins,
as reported previously (Matsuda et al. 2006) (Fig. 4c). These
results suggest that 1BnTIQ reduces parkin-dependent poly-
ubiquitination of tubulin in in vitro ubiquitination assay, and
this is reminiscent of the reduction of polyubiquinated
tubulin in the presence of T240R parkin mutant, which is
recognized in AR-JP.
(a)
(b)
(c)
(d)
Reduction of polyubiquitinated tubulin b by long-term
exposure to pathological concentrations of 1BnTIQ
Under pathological conditions, 1BnTIQ exists at nM level in
CSF of PD patients. Therefore, we examined the effect of
pathological concentrations of 1BnTIQ on polyubiquitination
of tubulin b. SH-SY5Y cells were incubated with 1BnTIQ at
concentrations of 5–50 nM for 1 month. Consequently,
1BnTIQ reduced high-molecular-weight (approximately
100 kDa) polyubiquitinated tubulin b at concentrations of
25–50 nM (Fig. 4d) in agreement with the result obtained
from the 48 h experiment using lM order of 1BnTIQ
(Fig. 4a). Thus, at pathological concentrations, 1BnTIQ
might cause the reduction of polyubiquitination of tubulin b.
Fig. 4 Reduction of polyubiquitinated tubulin b in the presence of 1-
benzyl-1,2,3,4-tetrahydroisoquinoline (1BnTIQ) in vitro and in vivo. (a)
Treatment of SH-SY5Y cells with 1BnTIQ (1 lM and 10 lM) for 48 h
reduced polyubiquitinated tubulin b (at about 100 kDa). Cells were
treated with 1BnTIQ and lysed with tris-NaCl-EDTA (TNE) buffer.
Immunoprecipitation (IP) was performed with mouse monoclonal anti-
tubulin b antibody and western blotting (WB) was performed with
monoclonal anti-ubiquitin antibody. (b) Effect of 1BnTIQ exposure for
48 h on accumulated polyubiquitinated tubulin b in the presence of a
proteasome inhibitor, MG-132, in SH-SY5Y cells. Polyubiquitinated
tubulin b was detected by IP-WB analysis. (c) Effect of 1BnTIQ on
in vitro ubiquitination assay of tubulin. T240R parkin mutant was used
as a control for impaired polyubiquitination of tubulin. E1 (0.1 lM),
UbcH5c (0.05 mg/mL), maltose binding protein-parkin (0.1 lM),
tubulin protein (1 lM), biotinylated ubiquitin (Ub, 2.5 lM), Mg-ATP
(50 mM), dithiothreitol (1 mM), inorganic pyrophosphatase (50 U/mL),
and 1BnTIQ (10 lM) were mixed on ice. Proteins were denatured
under non-reducing conditions, and WB was performed using avidin-
biotin-peroxidase complex (ABC method). (d) Treatment of SH-SY5Y
cells with a low concentration of 1BnTIQ (25 and 50 nM) for 1 month
reduced polyubiquitinated tubulin b (approximately 100 kDa). Cells
were dispersed after weekly passage. Medium change was performed
every 2 or 3 days. Cells were lysed and collected using TNE buffer. IP
was performed with mouse monoclonal anti-tubulin b antibody and WB
of IP eluates was performed using monoclonal anti-ubiquitin antibody.
All the experiments were repeatedly performed and representative
data are shown.
Discussion
It has been considered that idiopathic PD is caused by an
interaction of environmental and genetic factors. There are
many candidates for environmental factors, including pesti-
cides, herbicides, dopamine, dopamine metabolites, and other
endogenous compounds. 1BnTIQ was detected at three times
higher concentration in the CSF of idiopathic PD patients as
compared with control subjects (Kotake et al. 1995), and has
been proposed as a candidate causative substance of idiopathic
PD (Kheradpezhouh et al. 2003; Kotake et al. 2003; Shavali
and Ebadi 2003; Shavali et al. 2004; Wasik et al. 2009). In this
study, we explored the molecular mechanism of 1BnTIQ’s
action by searching for binding proteins.
Diazido-functionalized drugs can serve as efficient phot-
oaffinity probes, which covalently bind with their target
proteins (Hosoya et al. 2004, 2005). In this study, to apply
this method, we developed a photoaffinity probe, 1DAzBn-
TIQ (Fig. 1), which is a diazido-functionalized 1BnTIQ
analog, and succeeded in identifying tubulin b and a as
candidate binding proteins. Tubulin a was previously
identified as one of the binding proteins of 2-methylnorhar-
man, a PD-related b-carboline derivative, using a phage
display system (Gearhart et al. 2002). The results in the
present study suggest that 1BnTIQ also binds with tubulin, a
component of microtubules, as well as 2-methylnorharman.
It was shown that 1DAzBnTIQ binding to tubulin b in
cytosolic fraction requires the benzyl moiety by means of
Reduction of polyubiquitinated tubulin b by 1BnTIQ in
in vitro ubiquitination assay
It has been reported that self-ubiquitination of parkin is
observed in in vitro ubiquitination assay (Matsuda et al.
2006). Here, using the reported methods previously reported,
we examined whether tubulin a/b can be ubiquitinated in the
presence of 1BnTIQ in vitro. As the results, in the absence of
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

Reduction of Ub-tubulin by 1BnTIQ
| 1299
competitive experiments with several TIQ analogs (Fig. 2a
and b). The benzyl moiety of 1BnTIQ might be important for
neurotoxicity, because TIQs, which have the benzyl moiety,
are toxic to neuronal cells (Kotake et al. 2007). These results
suggest that other TIQs bearing a benzyl moiety may also
bind with tubulin b and modulate the neurotoxicity.
ubiquitination, similarly to mutant parkin in AR-JP. This
would also be consistent with a report that tubulin is a
component of Lewy bodies in PD patients (Galloway et al.
1992).
One possible explanation for the reduction of polyubiq-
uitinated tubulin b, shown in Fig. 4a, would be an increase
of proteasome activity. Therefore, we examined the effect
of 1BnTIQ on tubulin b polyubiquitination in the presence
of MG-132, a proteasome inhibitor. We found that 1BnTIQ
reduced the accumulation of polyubiquitinated tubulin b
induced by MG-132 (Fig. 4b, Fig. S2). We clarified that
1BnTIQ does not alter the proteasomal activity in SH-
SY5Y cells (data not shown). Moreover, we examined the
protein interaction between tubulin b and parkin by IP–WB
using rat brain cytosol. However, 1BnTIQ treatment did not
lead to the reduction of parkin–tubulin b interaction
(Fig. S3). Therefore, the reduction of polyubiquitinated
tubulin b by1BnTIQ may not be owing to the change of
parkin–tubulin b interaction. Accordingly, these results
suggest that 1BnTIQ reduces the polyubiquitination of
tubulin b upstream of proteasome, e.g., by indirect mech-
anisms such as the inhibition of carboxyl terminus of the
Hsc70-interacting protein recruitment.
To check whether 1BnTIQ reduces parkin-dependent
polyubiquitination of tubulin in vitro, we constructed an
in vitro ubiquitination assay system consisting of E1, E2,
parkin as E3, and its substrate tubulin. In vitro ubiquitination
assay showed that 1BnTIQ reduces parkin-dependent poly-
ubiquitination of tubulin b as efficiently as T240R parkin
mutation, which is recognized in AR-JP patients (Fig. 4c).
Monoubiquitination was detected most strongly, in agree-
ment with previous findings (Matsuda et al. 2006). However,
1BnTIQ did not reduce this monoubiquitination. Hence, this
result not only suggests that 1BnTIQ selectively reduces
polyubiquitination, not monoubiquitination, but also sup-
ports the idea that 1BnTIQ reduces polyubiquitination of
tubulin b in SH-SY5Y cells (Fig. 4a). Considering the results
of in vitro ubiquitination assay, which consists of the
minimum components required for ubiquitination of tubulin,
it seems likely that 1BnTIQ acts to reduce polyubiquitinated
tubulin b via direct binding of 1BnTIQ to tubulin b, not via
an indirect pathway.
Tubulin b has seven isoforms (Sullivan and Cleveland
1986; Lopata and Cleveland 1987). In human neuroblas-
toma SH-SY5Y cells, tubulin b3 is highly expressed
(He´raud et al. 2004), and we also detected tubulin b2 and
b3 mRNAs in SH-SY5Y (data not shown). We chose and
prepared four GST-fused tubulin b isoforms, b1, b2, b3, and
b4a, to confirm which isoforms are photolabeled with
1DAzBnTIQ. We found that 1DAzBnTIQ binds quite
strongly to tubulin b2 and b4a, and also binds weakly to
tubulin b1 and b3 (Fig 3b). Therefore, 1DAzBnTIQ may
bind predominantly to tubulin b2 and b4a in SH-SY5Y
cells. Tubulin b3 was detected in midbrain dopaminergic
neurons in C57BL mice (Lamba et al. 2005), and we also
confirmed 1DAzBnTIQ binding to tubulin b3 in SH-SY5Y
cells by IP (Fig. S1). Furthermore, in an immunocytochem-
ical study, 1DAzBnTIQ/GIF-0373-derived fluorescence
(green) and tubulin b (red) were completely co-localized
in SH-SY5Y neuroblastoma cells (Fig. 3c), in agreement
with the 1DAzBnTIQ/GIF-0373-derived fluorescence de-
tected by IP using anti-tubulin b antibody (Fig. 3a). These
results suggest that 1BnTIQ binds with tubulin b in
midbrain neurons.
AR-JP is caused by loss of function mutation of parkin, a
ubiquitin E3 ligase (Kitada et al. 1998). More than ten
proteins have been reported to be substrates of parkin so far.
For example, it has been reported that parkin-associated
endothelin receptor-like receptor (Pael-R) is a substrate of
parkin (Imai et al. 2001). In AR-JP, it is possible that Pael-R
is accumulated in neurons via impairment of its ubiquitina-
tion. A strain of Pael-R over-expressing mice crossed with
parkin-deficient mice exhibits catecholaminergic neuronal
loss in the brain (Wang et al. 2008). Consequently, accumu-
lation of a parkin substrate may be involved in the
pathogenesis of idiopathic PD. For example, there is a report
that dopamine covalently binds to parkin and inactivates its
E3 ligase activity in substantia nigra of idiopathic PD
patients (LaVoie et al. 2005). These results suggest that
environmental factors related to idiopathic PD inactivate
parkin activity or reduce ubiquitination of parkin substrates,
i.e., an effect similar to that of mutant parkin in AR-JP.
Tubulin a/b is reported to be a substrate of parkin (Ren
et al. 2003) and parkin tightly binds to tubulin a/b (Yang
et al. 2005). Polyubiquitination of tubulin a/b was reduced
by over-expression of mutant parkin compared with that of
wild-type parkin (Ren et al. 2003; Yang et al. 2005). Our
data suggest that 1BnTIQ reduced the formation of high-
molecular-weight polyubiquitinated tubulin b (Fig. 4a). In
Impairment of mitochondrial complex I is considered as
one of the etiological pathways of idiopathic PD. 1BnTIQ
is reported to inhibit complex I and to induce neuronal cell
death at concentrations between a few dozen lM and a
few hundred lM (Shavali and Ebadi 2003; Kotake et al.
2007). However, only a few dozen nM 1BnTIQ is detected
in parkinsonian CSF (Kotake et al. 1995). Therefore, we
examined the effect of 1BnTIQ at 50 nM on polyubiqui-
tination of tubulin b for 1 month and observed that tubulin
b polyubiquitination was reduced (Fig. 4d), in agreement
with the result using 10 lM 1BnTIQ for 48 h. These
results suggest that even low concentrations of 1BnTIQ
other words, 1BnTIQ appeared to impair tubulin
b
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

1300
| R. Kohta et al.
Betarbet R., Canet-Aviles R. M., Sherer T. B. et al. (2006) Intersecting
pathways to neurodegeneration in Parkinson’s disease: effects of
the pesticide rotenone on DJ-1, a-synuclein, and the ubiquitin-
proteasome system. Neurobiol. Dis. 22, 404–420.
Burke D., Gasdaska P. and Hartwell L. (1989) Dominant effects of
tubulin overexpression in Saccharomyces cerevisiae. Mol. Cell.
Biol. 9, 1049–1059.
Cappelletti G., Maggioni M. G. and Maci R. (1999) Influence of MPP+
on the state of tubulin polymerisation in NGF-differentiated PC12
cells. J. Neurosci. Res. 56, 28–35.
Chemler S. R., Trauner D. and Danishefsky S. J. (2001) The B-alkyl
Suzuki-Miyaura cross-coupling reaction: development, mechanis-
tic study, and applications in natural product synthesis. Angew.
Chem. Int. Ed. 40, 4544–4568.
Chung K. K., Zhang Y., Lim K. L., Tanaka Y., Huang H., Gao J., Ross
C. A., Dawson V. L. and Dawson T. M. (2001) Parkin ubiquitinates
the a-synuclein-interacting protein, synphilin-1: implications for
Lewy-body formation in Parkinson disease. Nat. Med. 7, 1144–
1150.
von Coelln R., Dawson V. L. and Dawson T. M. (2004) Parkin-associ-
ated Parkinson’s disease. Cell Tissue Res. 318, 175–184.
Dawson T. M. and Dawson V. L. (2003) Molecular pathways of neu-
rodegeneration in Parkinson’s disease. Science 302, 819–822.
Galloway P. G., Mulvihill P. and Perry G. (1992) Filaments of Lewy
bodies contain insoluble cytoskeletal elements. Am. J. Pathol. 140,
809–822.
Gearhart D. A., Toole P. F. and Warren Beach J. (2002) Identification of
brain proteins that interact with 2-methylnorharman. An analog of
the parkinsonian-inducing toxin, MPP+. Neurosci. Res. 44, 255–
265.
can reduce polyubiquitination of tubulin b if present for a
long period. It was reported that over-expression of tubulin
b is toxic (Burke et al. 1989). Although we have not
observed accumulation of tubulin b caused by 1BnTIQ in
the present study, long exposure to endogenous amounts of
1BnTIQ in parkinsonian CSF might lead to the accumu-
lation of insoluble tubulin b.
In summary, we found that 1BnTIQ binds directly with
tubulin b and reduces the level of polyubiquitinated tubulin b
both in SH-SY5Y cells and in vitro, similarly to the effect of
mutant parkin in AR-JP. Substances such as 1BnTIQ, which
cause impairment of tubulin b polyubiquitination, are
candidates for environmental factors contributing to idio-
pathic PD.
Acknowledgements
We thank Dr N. Hattori (Juntendo University) for providing parkin
constructs. We also thank Dr T. Inuzuka (Gifu University) for help
in HRMS analyses. We thank the Analysis Center of Life Science
and the Research Center for Molecular Medicine, Hiroshima
University and the Life Science Research Center, Gifu University
for the use of their facilities. This study was supported in part by
Grant-in-Aids for Scientific Research from the Japan Society for the
Promotion of Science, Creative Scientific Research, and Strategic
Programs for R&D in RIKEN.
Goldberg M. S., Fleming S. M., Palacino J. J. et al. (2003) Parkin-
deficient mice exhibit nigrostriatal deficits but not loss of dopa-
minergic neurons. J. Biol. Chem. 278, 43628–43635.
He´raud C., Hilairet S., Muller J. M., Leterrier J. F. and Chade´neau C.
(2004) Neuritogenesis induced by vasoactive intestinal peptide,
pituitary adenylate cyclase-activating polypeptide, and peptide
histidine methionine in SH-SY5Y cells is associated with regulated
expression of cytoskeleton mRNAs and proteins. J. Neurosci. Res.
75, 320–329.
Supporting Information
Additional Supporting Information may be found in the online
version of this article:
Appendix S1. Supplementary Materials and methods.
Figure S1. We confirmed the binding of 1DAzBnTIQ to tubulin
b and b3 in SH-SY5Y cell lysate.
Hosoya T., Hiramatsu T., Ikemoto T., Nakanishi M., Aoyama H., Hosoya
A., Iwata T., Maruyama K., Endo M. and Suzuki M. (2004) Novel
bifunctional probe for radioisotope-free photoaffinity labeling:
compact structure comprised of photospecific ligand ligation and
detectable tag anchoring units. Org. Biomol. Chem. 2, 637–641.
Hosoya T., Hiramatsu T., Ikemoto T., Aoyama H., Ohmae T., Maruyama
K., Endo M. and Suzuki M. (2005) Design of dantrolene-derived
probes for radioisotope-free photoaffinity labeling of proteins in-
volved in the physiological Ca+ release from sarcoplasmic reticu-
lum of skeletal muscle. Bioorg. Med. Chem. Lett. 15, 1289–1294.
Hosoya T., Inoue A., Hiramatsu T., Aoyama H., Ikemoto T. and Suzuki
M. (2009) Facile synthesis of diazido-functionalized biaryl
compounds as radioisotope-free photoaffinity probes by Suzuki-
Miyaura coupling. Bioorg. Med. Chem. 17, 2490–2496.
Huynh D. P., Scoles D. R., Nguyen D. and Pulst S. M. (2003) The
autosomal recessive juvenile Parkinson disease gene product,
parkin, interacts with and ubiquitinates synaptotagmin XI. Hum.
Mol. Genet. 12, 2587–2597.
Figure S2. Immunocytochemistry of polyubiquitinated proteins
(not polyubiquitinated tubulin b) in the presence of 1BnTIQ and/
or MG-132, using mouse monoclonal anti-tubulin b antibody
(green) and monoclonal anti-polyubiquitinated proteins antibody
(red).
Figure S3. The effect of 1BnTIQ on parkin-tubulin b interaction
was examined by IP-WB experiments.
Table S1. Identification of separated fluorescent spots by MS/MS
ion search using MALDI-TOF/TOF mass spectroscopy and MAS-
COT analysis.
As a service to our authors and readers, this journal provides
supporting information supplied by the authors. Such materials are
peer-reviewed and may be re-organized for online delivery, but are
not copy-edited or typeset. Technical support issues arising from
supporting information (other than missing files) should be
addressed to the authors.
Imai Y., Soda M., Inoue H., Hattori N., Mizuno Y. and Takahashi R.
(2001) An unfolded putative transmembrane polypeptide, which
can lead to endoplasmic reticulum stress, is a substrate of Parkin.
Cell 105, 891–902.
Itier J. M., Ibanez P., Mena M. A. et al. (2003) Parkin gene inactivation
alters behaviour and dopamine neurotransmission in the mouse.
Hum. Mol. Genet. 12, 2277–2291.
References
Betarbet R., Sherer T. B., MacKenzie G., Garcia-Osuna M., Panov A. V.
and Greenamyre J. T. (2000) Chronic systemic pesticide exposure
reproduces features of Parkinson’s disease. Nat. Neurosci. 3, 1301–
1306.
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301

Reduction of Ub-tubulin by 1BnTIQ
| 1301
Kawai H., Kotake Y. and Ohta S. (2000) Inhibition of dopamine
receptors endogenous amines; binding to striatal receptors and
pharmacological effects on locomotor activity. Bioorg. Med. Chem.
Lett. 10, 1669–1671.
Kheradpezhouh M., Shavali S. and Ebadi M. (2003) Salsolinol causing
parkinsonism activates endoplasmic reticulum-stress signaling
pathways in human dopaminergic SK-N-SH cells. Neurosignals
12, 315–324.
Kiick K. L., Saxon E., Tirrell D. A. and Bertozzi C. R. (2002) Incor-
poration of azides into recombinant proteins for chemoselective
modification by the Staudinger ligation. Proc. Natl Acad. Sci. USA
99, 19–24.
Kitada T., Asakawa S., Hattori N., Matsumine H., Yamamura Y., Mi-
noshima S., Yokochi M., Mizuno Y. and Shimizu N. (1998)
Mutations in the parkin gene cause autosomal recessive juvenile
parkinsonism. Nature 392, 605–608.
Kotake Y., Tasaki Y., Makino Y., Ohta S. and Hirobe M. (1995) 1-
Benzyl-1,2,3,4-tetrahydroisoquinoline as a parkinsonism-inducing
agent: a novel endogenous amine in mouse brain and parkinsonian
CSF. J. Neurochem. 65, 2633–2633.
Kotake Y., Yoshida M., Ogawa M., Tasaki Y., Hirobe M. and Ohta S.
(1996) Chronic administration of 1-benzyl-1,2,3,4-tetrahydroiso-
quinoline, an endogenous amine in the brain, induces parkinsonism
in a primate. Neurosci. Lett. 217, 69–71.
Kotake Y., Ohta S., Kanazawa I. and Sakurai M. (2003) Neurotoxicity of
an endogenous brain amine, 1-benzyl-1,2,3,4-tetrahydroisoquino-
line, in organotypic slice co-culture of mesencephalon and stria-
tum. Neuroscience 117, 63–70.
Kotake Y., Sekiya Y., Okuda K. and Ohta S. (2007) Cytotoxicity of 17
tetrahydroisoquinoline derivatives in SH-SY5Y human neuro-
blastoma cells is related to mitochondrial NADH-ubiquinone oxi-
doreductase inhibition. Neurotoxicology 28, 27–32.
Lamba W., Prichett W., Munoz D., Park D. S. and Woulfe J. M. (2005)
MPTP induces intranuclear rodlet formation in midbrain dopami-
nergic neurons. Brain Res. 1066, 86–91.
Langston J. W., Ballard P., Tetrud J. W. and Irwin I. (1983) Chronic
Parkinsonism in humans due to a product of meperidine-analog
synthesis. Science 219, 979–980.
LaVoie M. J., Ostaszewski B. L., Weihofen A., Schlossmacher M. G. and
Selkoe D. J. (2005) Dopamine covalently modifies and function-
ally inactivates parkin. Nat. Med. 11, 1214–1221.
Lenz G. R. and Lessor R. A. (1992) Tetrahydro-3-benzazepin-2-ones:
lead tetraacetate oxidation of isoquinoline enamides. Org. Synth.
70, 139–150.
Lenz G. R., Woo C.-M. and Hawkins B. L. (1982) Reaction of isoquinoline
enamides with electrophiles. J. Org. Chem. 47, 3049–3053.
Lopata M. A. and Cleveland D. W. (1987) In vivo microtubules are
copolymers of available b-tubulin isotypes: localization of each of
six vertebrate b-tubulin isotypes using polyclonal antibodies elic-
ited by synthetic peptide antigens. J. Cell Biol. 105, 1707–1720.
Matsuda N., Kitami T., Suzuki T., Mizuno Y., Hattori N. and Tanaka K.
(2006) Diverse effects of pathogenic mutations of Parkin that
catalyze multiple monoubiquitylation in vitro. J. Biol. Chem. 281,
3204–3209.
arenes, and cycloalkenes via a hydroboration-coupling sequence.
J. Am. Chem. Soc. 111, 314–321.
Nagatsu T. and Yoshida M. (1988) An endogenous substance of the
brain, tetrahydroisoquinoline, produces parkinsonism in primates
with decreased dopamine, tyrosine hydroxylase and biopterin in
the nigrostriatal regions. Neurosci. Lett. 87, 178–182.
Nikulin V. I., Rakov I. M., De Los Angeles J. E., Mehta R. C., Boyd
L. Y., Feller D. R. and Miller D. D. (2006) 1-Benzyl-1,2,3,4-tet-
rahydroisoquinoline-6,7-diols as novel affinity and photoaffinity
probes for b-adrenoceptor subtypes. Bioorg. Med. Chem. 14,
1684–1697.
Perez F. A. and Palmiter R. D. (2005) Parkin-deficient mice are not a
robust model of parkinsonism. Proc. Natl Acad. Sci. USA 102,
2174–2179.
Ren Y., Zhao J. and Feng J. (2003) Parkin binds to a/b tubulin and
increases their ubiquitination and degradation. J. Neurosci. 23,
3316–3324.
Rostovtsev V. V., Green L. G., Fokin V. V. and Sharpless K. B. (2002) A
stepwise Huisgen cycloaddition process: copper(I)-catalyzed reg-
ioselective ‘‘ligation’’ of azides and terminal alkynes. Angew.
Chem. Int. Ed. 41, 2596–2599.
Saxon E. and Bertozzi C. R. (2000) Cell surface engineering by a
modified Staudinger reaction. Science 287, 2007–2010.
Shavali S. and Ebadi M. (2003) 1-Benzyl-1,2,3,4-tetrahydroisoquinoline
(1BnTIQ), an endogenous neurotoxin, induces dopaminergic cell
death through apoptosis. Neurotoxicology 24, 417–424.
Shavali S., Carlson E. C., Swinscoe J. C. and Ebadi M. (2004) 1-Benzyl-
1,2,3,4-tetrahydroisoquinoline, a Parkinsonism-inducing endoge-
nous toxin, increases a-synuclein expression and causes nuclear
damage in human dopaminergic cells. J. Neurosci. Res. 76, 563–571.
Sullivan K. F. and Cleveland D. W. (1986) Identification of conserved
isotype-defining variable region sequences for four vertebrate b
tubulin polypeptide classes. Proc. Natl Acad. Sci. USA 83, 4327–
4331.
Thomas B. and Beal M. F. (2007) Parkinson’s disease. Hum. Mol. Genet.
16, 183–194.
Tornøe C. W., Christensen C. and Meldal M. (2002) Peptidotriazoles on
solid phase: [1,2,3]-triazoles by regiospecific copper(I)-catalyzed
1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org.
Chem. 67, 3057–3064.
Vice S., Bara T., Bauer A. et al. (2001) Concise formation of 4-benzyl
piperidines and related derivatives using a Suzuki protocol. J. Org.
Chem. 66, 2487–2492.
Wang H. Q., Imai Y., Inoue H., Kataoka A., Iita S., Nukina N. and
Takahashi R. (2008) Pael-R transgenic mice crossed with parkin
deficient mice displayed progressive and selective catecholamin-
ergic neuronal loss. J. Neurochem. 107, 171–185.
´
Wasik A., Romanska I. and Antkiewicz-Michaluk L. (2009) 1-Benzyl-
1,2,3,4-tetrahydroisoquinoline, an endogenous parkinsonism-
inducing toxin, strongly potentiates MAO-dependent dopamine
oxidation and impairs dopamine release: ex vivo and in vivo neu-
rochemical studies. Neurotox. Res. 15, 15–23.
Woulfe J. and Munoz D. (2000) Tubulin immunoreactive neuronal in-
tranuclear inclusions in the human brain. Neuropathol. Appl.
Neurobiol. 26, 161–171.
Yang F., Jiang Q., Zhao J., Ren Y., Sutton M. D. and Feng J. (2005)
Parkin stabilizes microtubules through strong binding mediated by
three independent domains. J. Biol. Chem. 280, 17154–17162.
Miyaura N., Ishiyama T., Sasaki H., Ishikawa M., Satoh M. and Suzuki A.
(1989) Palladium-catalyzed inter- and intramolecular cross-coupling
reactions of B-alkyl-9-borabicyclo[3.3.1]nonane derivatives with
1-halo-1-alkenes or haloarenes. Syntheses of functionalized alkenes,
ꢀ 2010 The Authors
Journal Compilation ꢀ 2010 International Society for Neurochemistry, J. Neurochem. (2010) 114, 1291–1301