Comparison of neuroprotective and neurorestorative capabilities of rasagiline and selegiline against lactacystin-induced nigrostriatal dopaminergic degeneration

,

*Department of Neurology, The 1st Affiliated Hospital of Sun Yat-sen University, Guangzhou, China

�Department of Neurology, Baylor College of Medicine, Houston, Texas, USA

�Technion-Rappaport Family Faculty of Medicine, Haifa, Israel

Parkinson’s disease (PD) is a progressive disorder in whichneurons producing dopamine (DA), norepinephrine, andserotonin degenerate and intracellular inclusions, Lewybodies, accumulate. Various abnormalities, such as mito-chondrial defects, oxidative stress, and ubiquitin-proteasomesystem (UPS) dysfunction have been suggested to contributeto the degenerative process underlying PD (Siderowf andStern 2003). Pharmacological strategies designed to interferewith these pathological mechanisms may effectively coun-teract the degeneration.

Proteasomes, are the primarily enzymes involved in thedegradation of unwanted proteins within cells, whichconsist of enzyme complexes mediating the rapid ATP-dependent degradation of abnormal intracellular proteinsafter they have been covalently bound to ubiquitin(Ciechanover et al. 2000). Several studies have described

the abnormal structure and dysfunction of proteasomes inthe brains of PD patients (McNaught and Olanow 2006).Furthermore, systemic or striatal inhibition of the UPSby carbobenzoxy-L-y-t-butyl-L-glutamyl-L-alanyl-leucinal

Received August 6, 2007; revised manuscript received January 31, 2008;accepted February 1, 2008.Address correspondence and reprint requests to Weidong Le, MD

PhD, Department of Neurology, NB 205, Baylor College of Medicine,Houston, TX 77030, USA. E-mail: [email protected] or MoussaB. H. Youdim, PhD, Technion-Rappaport Family Faculty of Medicine,Efron St., PO Box 9697, Haifa 31096, Israel.E-mail: [email protected] used: DA, dopamine; DOPAC, 4-dihydroxy-phenyla-

cetic acid; HVA, homovanillic acid; MAO, monoamine oxidase; MFB,medial forebrain bundle; PBS, phosphate-buffered saline; PD, Parkin-son’s disease; SN, substantia nigra; SOD, superoxide dismutase; TH,tyrosine hydroxylase; UPS, ubiquitin-proteasome system.

Abstract

Nigrostriatal neurodegeneration in Parkinson’s disease (PD)

has been postulated to be caused by various pathological

conditions, such as mitochondrial defects, oxidative stress,

and ubiquitin-proteasome system (UPS) dysfunction. Phar-

macological strategies designed to interfere with these

pathological pathways may effectively counteract the

degeneration. Rasagiline and selegiline are selective and

irreversible monoamine oxidase-B inhibitors that possess

significant protective properties on dopamine neurons in var-

ious pre-clinical models of PD. In the present study, the

neuroprotective and neurorestorative effects of rasagiline and

selegiline were compared in an animal model of PD produced

by inhibition of the UPS. C57BL/6 male mice were microin-

jected bilaterally with UPS inhibitor lactacystin (1.25 lg/side),

into the medial forebrain bundle. Administration of rasagiline

(0.2 mg/kg, i.p. once per day) or selegiline (1 mg/kg, i.p. once

per day), started 7 days before or after (up to 28 days) after

lactacystin microinjection. We found that both rasagiline and

selegiline exerted a significant neuroprotective effect against

lactacystin-induced neurodegeneration; but only rasagiline

managed to restore the nigrostriatal degeneration. Further-

more, rasagiline showed a modest protection against lacta-

cystin-induced inhibition of proteasomal activity. Our study

indicates that compared with selegiline, rasagiline is more

potent in protecting neurodegeneration induced by UPS

impairment and may, therefore, exert disease-modifying ef-

fects in PD.

Keywords: lactacystin, neuroprotection, neurorestoration,

rasagiline, selegiline, ubiquitin-proteasome system.

J. Neurochem. (2008) 105, 1970–1978.

JOURNAL OF NEUROCHEMISTRY | 2008 | 105 | 1970–1978 doi: 10.1111/j.1471-4159.2008.05330.x

1970 Journal Compilation � 2008 International Society for Neurochemistry, J. Neurochem. (2008) 105, 1970–1978� 2008 The Authors

(PSI) has been shown to cause parkinsonism, which ischaracterized by the formation of neuronal inclusions(Fornai et al. 2003 and McNaught et al. 2003; McNaughtand Olanow 2006). Our previous studies have documentedthat unilateral stereotaxic injection of proteasome inhibitorlactacystin into the medial forebrain bundle (MFB) of micecauses degeneration of DA neurons, protein aggregation,and inclusion body formation (Zhang et al. 2005). Inaddition, bilateral stereotaxic injection with lactacystin intothe MFB of mice induces behavioral deficits, nigrostriataldegeneration, protein aggregation, and proteasomal inhibi-tion, similar to PD (Zhu et al. 2007). Thus, impairment ofproteasome function may play an important role in proteinaggregation and neurodegeneration in PD.

Monoamine oxidase type B (MAO-B) inhibitors ofpropargylamine family have been extensively investigatedfor their potential neuroprotective properties (Blandini 2005).Selegiline, the prototype of this class of compounds, hasdemonstrated antioxidant and neuroprotective effects inexperimental studies (Le et al. 1997; Youdim and Bakhle2006), although the neuroprotective activity in PD patientshas remained controversial (Shoulson 1998, Palhagen et al.2006). Its restricted derivate, rasagiline, which is differentfrom selegiline in that it is not metabolized to amphetamineand/or methamphetamine and possesses the neuroprotectivemetabolite aminoindan (Bar-Am et al. 2007), has shown tobe the most potent propargylamine against a variety of insultsin both in vitro and in vivo PD models (Youdim et al. 2003,2005; Sagi et al. 2007). Recent clinical studies have reportedefficacy of rasagiline for PD treatment, either as adjuncttherapy to L-DOPA or as monotherapy in early PD (Bayeset al. 2006). Patients treated with rasagiline from onset showsignificantly less functional decline than those in whomtreatment is delayed (Bayes et al. 2006). These findingssuggest that rasagiline may possess disease-modifyingactivity, which is supported by numerous in vitro andin vivo studies (Weinreb et al. 2006; Sagi et al. 2007). Asinhibition of proteasomal function contributes to the neu-rodegeneration in PD, our goal is to explore and compare theneuroprotective and neurorestorative potentials of rasagilineand selegiline in mice against nigrostriatal DA neuronsdegeneration induced by the proteasome inhibitor lactacystin.

Materials and methods

Animals and treatmentsThe proposed animal study was approved by the Baylor College of

Medicine Animal Use and Care Committee and conducted in

accordance with the Guide for the Care and Use of Laboratory

Animals as adopted and promulgated by the National Institute of

Health. Male C57BL/6 mice, aged 12 weeks, were randomly

assigned into six groups: control, lactacystin, pre-Ras, post-Ras,

pre-Sel, and post-Sel, respectively. They were housed five animals

per cage in a colony room maintained at constant temperature and

humidity, with a 12-h light/dark cycle, and allowed at least 7 days to

acclimate before any treatment. Administration of rasagiline

(0.2 mg/kg, i.p. q.i.d.) or selegiline (1 mg/kg, i.p. q.i.d.) started

7 days before (pre-treatment) or after (post-treatment) microinjection

with lactacystin, up to the end of the study (28 days after

microinjection of lactacystin), while the administration of a same

volume saline was served as a control. The time point chosen for

post-treatment are based on our previous finding that there was

noticeable injury in nigrastriatal system 7 days after lactacystin

microinjection. The dosage regimens of rasagiline and selegiline

were elected according to several studies (Youdim and Tipton 2002;

Muralikrishnan et al. 2003; West et al. 2006) and they were

postulated to have same abilities on MAO-B inhibitory. For the

stereotaxic injection of lactacystin, mice were deeply anesthetized

and placed in a Kopf stereotaxic frame (Kopf Instruments, Tujunga,

CA, USA). An injection cannula was inserted through a hole drilled

in the skull, into MFB, using the following coordinates (in mm):

(1.34 posterior, ±1.17 lateral, and 5.1 ventral from bregma) of each

mouse. Two microliters of either phosphate-buffered saline (PBS;

0.1 M) as control or lactacystin (1.25 lg; A.G. Scientific, San Diego,CA, USA) in PBS was injected into the MFB of each mouse. All the

mice were healthy during the experiment period and were killed

28 days after microinjection of lactacystin by terminal anesthesia

followed by transcardial perfusion with ice-cold PBS. The mice were

decapitated and the brains were immediately removed, placed on ice,

and transected coronally at the infundibular stem. The midbrain

blocks were fixed with 4% p-formaldehyde in PBS 2 days and

cryoprotected in 30% sucrose for 2 days at 4�C followed by

histological analysis. Striatal tissues and ventral midbrains were

rapidly dissected out and stored at )80�C until analysis. The samples

were divided to perform different sets of experiments.

Locomotive activities and rotarod performanceLocomotive activities and rotarod performance were tested 1 day

prior and every 7 days after the microinjection of lactacystin.

Locomotive activities were monitored by the AccuScan Digiscan

system (AccuScan Instruments Inc., Columbus, OH, USA). Data

collected by computer included total distance traveled and moving

time. The measurements were carried out during the period between

9 AM and 11 AM in a dark room. Each mouse was placed in the

testing chamber for 30 min for adaptation, followed by a 60 min

recording by the computer-generated automatic analysis system.

Motor coordination was determined with an accelerating rotarod

treadmill (Columbus Instruments, Columbus, OH, USA). Initially,

the mice were required to perch on the stationary rod for 30 s to

accustom themselves to the environment. Then the animals were

trained at a constant speed of 5 rpm for 90 s. After this pre-training,

mice were tested three times at 1-h intervals on three consecutive

days for a total of nine tests, a mean of which undergoes statistical

analysis. During each test, the rotarod was set at a starting speed of

5 rpm for 30 s, and the speed was increased by 0.1 rpm per second.

All animals were tested three times for each experiment, and the

mean of the test results were subjected statistical analysis.

Immunohistochemistry and stereology-based cell countingMidbrain blocks were cut into 30 lm sections and systematically

picked at 150 lm intervals. This procedure yielded seven sections

� 2008 The AuthorsJournal Compilation � 2008 International Society for Neurochemistry, J. Neurochem. (2008) 105, 1970–1978

Neuroprotection and neurorescue of rasagiline and selegiline | 1971

for each animal covering the whole substantia nigra (SN). Free-

floating sections were incubated successively for 15 min with

0.05% H2O2 in 0.1 M PBS to remove endogenous peroxidase

activity, for 1 h with 2% goat serum/0.1% Triton X-100 in 0.1 M

PBS to block non-specific binding sites, and for 24 h 4�C with

the primary rabbit anti-tyrosine hydroxylase (TH, 1 : 1500; Protos

Biotech, New York, NY, USA) to detect DA neurons. Sections

were then incubated for 2 h at 25�C with the appropriate

biotinylated secondary antibody (anti-rabbit or anti-rat IgG,

1 : 200; Vector Laboratories Inc., Burlingame, CA, USA). The

avidin–biotin method was used to amplify the signal (ABC Kit;

Vector Laboratories Inc.) and 3, 3¢-diaminobenzidine tetrachloride

was used to visualize bound antibodies. Stereological methods

were used to evaluate the number of DA neurons (TH-positive

cells) in the SN. The stereological setup was composed of a

computer, a microscope (Axioskop 2; Carl Zeiss Inc., Thornwood,

NY, USA), a color camera (Olympus America Inc., Center Valley,

PA, USA), and an electronic microcator (Heidenhain Corp.,

Schaumburg, IL, USA). Counting of cell numbers was performed

with the computer-assisted stereological toolbox software program

Stereo Investigator 7.0 (MicroBrightField, Inc., Willston, VT,

USA). The total number of TH-positive neurons was estimated

using the optical fractionator method (Glaser and Glaser 2000).

For each tissue section analyzed, the guard zones of 3 lmthickness were used at the top and bottom of each section. The

SN was outlined under 2.5· magnification, and 50% of the

outlined region was analyzed using a sampling design generated

with the following stereologic parameters: grid size,

200 · 200 lm; counting frame size, 150 · 150 lm; and dissector

height, 14 lm. The neurons were counted under 40· magnifica-

tion and the total number was calculated by using the formula

previously described for this method (West et al. 1991).

Determination of striatal DA and its metabolitesThe concentration of DA, 4-dihydroxy-phenylacetic acid (DOPAC),

homovanillic acid (HVA), serotonin, and 5-hydroxyindolacetic acid

were quantified in striatal tissues by HPLC. Briefly, striatal tissues

were homogenized (10% wt/vol) by sonication in ice-cold 0.1 M

perchloric acid. Homogenates were centrifuged at 10 000 g for

10 min at 4�C and the supernatants were collected and filtered

through acro-disc filters (0.25 lm; Fisher Scientific, Pittsburgh, PA,

USA) and subjected to HPLC (HTEC-500; Eicom, Kyoto, Japan)

with the column (EICOMPAK SC-3ODS; Eicom) and detected by

an electrochemical detector (ADInstruments Pty Ltd., Castle Hill,

NSW, Australia). The mobile phase consisted of 0.1 mM citric acid,

0.1 M sodium acetate, 220 mg/L octane sulfate sodium, 5 mg/L

EDTA, and 20% methanol, pH 3.5.

Proteasome activity assayVentral midbrains were placed on ice and homogenized in lysis

buffer (50 mM HEPES, pH 7.5, 5 mM EDTA, 150 mM NaCl,

0.5 mM ATP, and 1% Triton X-100). The lysates were centrifuged at

14 000 g at 4�C for 20 min. The resulting supernatants were placed

on ice and assayed for protein concentrations by the Bradford’s

method (Bio-Rad, Hercules, CA, USA). The 20S Proteasome

Activity kit (Chemicon International Inc., Temecula, CA, USA) was

used to measure the chymotrypsin-like activity. Assays were carried

out with 50 lg of midbrain lysates and the appropriate substrate at

37�C for 90 min incubation. The activity was measured by detection

of the fluorophore 7-amido-4-methylcoumarin after cleavage from

the synthetic fluorogenic peptide: Leu-Leu-Val-Tyr-7-amido-4-

methylcoumarin, using a spectrofluorimeter (Cytofluor II; PerSep-

tive Biosystems, Framingham, MA, USA) at excitation/emission

wavelengths of 380/460 nm. The results are expressed as fluores-

cence units/mg protein.

Bcl-2 immunoblotTissues of mice were extracted with mammalian tissue lysis/

extraction reagent (Sigma-Aldrich, St Louis, MO, USA) supple-

mented with complete protease inhibitor cocktail (Sigma-Aldrich).

Equal amounts of lysate protein were denatured in sodium dodecyl

sulfate sample buffer, subjected onto a 12% sodium dodecyl sulfate–

polyacrylamide gel electrophoresis gel, and transferred onto a

polyvinyl difluoride membrane. After being blocked in 6% non-fat

dry milk, membranes were incubated in the presence of respective

primary antibodies, Bcl-2 (1 : 1000; Chemicon International Inc.),

or b-actin (1 : 5000; Sigma-Aldrich), and followed by incubation

with horseradish peroxidase-labeled secondary antibodies (1 : 2000;

Chemicon International Inc.). Signals were detected using enhanced

chemiluminescence (Amersham, Arlington Heights, IL, USA).

Statistical analysisAll the results of the study were from groups of 5–10 mice. Data

were analyzed using SPSS 11.0 software (SPSS Inc., Chicago, IL,

USA). Statistical significance of difference between parameters was

determined at the 0.05 level, using one- or two-way ANOVA as

appropriate. ANOVA were followed when allowed by post hoc t-testcorrected for multiple comparisons by the method of Tukey or

Bonferroni. Mean values were quoted with the respective standard

error.

Results

Rasagiline and selegiline improved behavioral performancein lactacystin-lesioned miceIn the mice injected with lactacystin on day 7, compared withthe control, locomotive activities (total distance traveled andmoving time) were significantly decreased by 67.9% and71.8%, respectively (Fig. 1), and the rotarod performance (thetime remaining on the rod) was significantly reduced by 86.2%(Fig. 2). The changes in the locomotive tests and rotarodperformance remained unchanged at the end of the study (day28). However, pre-treatment with rasagiline and selegilinesignificantly attenuated the behavioral impairments on day 7,by 76.4% and 94.4% in total distance traveled, by 95.1% and139.0% in moving time, and by 87.6% and 54.6% in rotarodtime, respectively [Figs 1(a and c) and 2a]. Marked recoverywas also seen in the mice post-treated with rasagiline (by54.4% in total distance traveled, by 70.7% inmoving time, andby 69.1% in rotarod time) as well as in the mice post-treatedwith selegiline in total distance traveled (by 45.5%) and inmoving time (by 59.8%), but not in rotarod performance(remaining at 39.3% of control) [Figs 1(b and d) and 2b].

Journal Compilation � 2008 International Society for Neurochemistry, J. Neurochem. (2008) 105, 1970–1978� 2008 The Authors

1972 | W. Zhu et al.

Effects of rasagiline and selegiline againstlactacystin-induced DA neuron loss in SNCompared with the vehicle control, the number of DAneurons was reduced in lactacystin-injected mice by 71.1%on day 28 (Fig. 3). Pre-treatment with either rasagiline orselegiline significantly protected the DA neurons againstlactacystin-induced injury at the end of the study with 72.8%and 59.3% reduction in the DA neuron loss, respectively(Fig. 3). Furthermore, compared with selegiline (1 mg/kg,i.p. once per day), rasagiline (0.2 mg/kg, i.p. once per day)was more potent in protecting DA neurons against lactacy-stin. Only rasagiline was associated with an apparent rescueof DA neurons after a 21-day post-treatment followingmicroinjection of lactacystin, preventing the loss by 55.2%(Fig. 3). In contrast, selegiline failed to restore DA neurons

when it was applied after the microinjection of lactacystin(Fig. 3).

Rasagiline and selegiline restored lactacystin-induceddepletion of DACompared with the control, the reduction of DA, DOPAC, andHVA induced by lactacystin was significant on day 28, whichwas 51.0%, 52.2%, and 66.4%, respectively (Fig. 4a and b). Atthe end of the study, we found that pre-treatment of rasagilineand selegiline significantly attenuated the lactacystin-inducedreduction of striatal DA by 86.7% and 46.5%, respectively.Post-treatment with rasagiline and selegiline also rescued up to78.7% and 58.3% of control striatal DA level, respectively(Fig. 4a), but only rasagiline was able to restore the depletionsof DOPAC and HVA induced by lactacystin (Fig. 4b).

Control2500

(a) (b)

(c) (d)

2000

Tota

l dis

tan

ce t

rave

led

(cm

/60

min

)To

tal m

ovin

g t

ime

(s/6

0 m

in)

Tota

l mov

ing

tim

e (s

/60

min

)To

tal d

ista

nce

tra

vele

d (

cm/6

0 m

in)

1500

1000

300

250

200

150

100

50

300

250

200

150

100

50

2500

2000

1500

1000

500day-1 day 7 day 14 day 21 day 28 day-1 day 7 day 14 day 21 day 28

day-1 day 7 day 14 day 21 day 28 day-1 day 7 day 14 day 21 day 28

*

***

* * * ***

*

#

#

##

##

## #

#

#

#

LactacystinPre-RasPre-Sel

ControlLactacystinPost-RasPost-Sel


ControlLactacystinPre-RasPre-Sel

#

##

#

# #

###

##

#

*

*

*

* * ** * *

*

Fig. 1 Rasagiline and selegiline improved locomotive activities in

lactacystin-lesioned mice. Two microliters of either phosphate-buf-

fered saline (PBS; 0.1 M) as control or lactacystin (1.25 lg/side) in

PBS was injected into the MFB of each mouse. Administration of ra-

sagiline (0.2 mg/kg, i.p. per day) or selegiline (1 mg/kg, i.p. per day)

started 7 days before (pre-treatment) or after (post-treatment) micro-

injection with lactacystin, while the administration of a same volume

saline was served as control. The day of microinjection with lactacystin

was set to be day 0. The changes of locomotive activities were pre-

sented by total distance traveled (a and b) and moving time (c and d).

Data were expressed as mean ± SEM (n = 7). Two-way ANOVA was

applied to analysis the difference (F = 51.551 in total distance trav-

eled, F = 49.321 in total moving time, df for group = 5), followed by

post hoc tests corrected for multiple comparisons by the method of

Tukey. In general, there was no significant difference between ra-

sagiline and selegiline in improving the locomotive activities. *p < 0.05

versus control and #p < 0.05 versus lactacystin.



Rasagiline and selegiline failed to alleviatelactacystin-induced proteasome inhibitionA 40.6% inhibition caused by microinjection of lactacystinof the chymotrypsin-like proteasome activity in the ventralmidbrain was evident in the ventral midbrain 28 days afterlactacystin injection (Fig. 5). Neither rasagiline nor selegilinetreatment was able to alleviate the proteasome inhibitioninduced by lactacystin; although a modest, but insignificantrestoration of proteasome activity was seen after both pre-treatment and post-treatment with rasagiline (Fig. 5).

Rasagiline and selegiline reversed the reduction of Bcl-2protein level caused by proteasome inhibitionCompared with vehicle control, a significant reduction ofBcl-2 (38.5% of control) was observed in the ventralmidbrain on day 28 (Fig. 6). Protein level of Bcl-2 in micetreated with rasagiline (regardless of pre-treatment or post-treatment) was elevated to around 106% of control. While

selegiline seemed less effective in counteracting the Bcl-2reduction compared with rasagiline. Protein level of Bcl-2 inmice pre-treated and post-treated with selegiline was 98.8%and 95.4%, respectively (Fig. 6).

Discussion

Recently, we have shown that bilateral microinjection oflactacystin into the MFB of mice induces degeneration of DA

Control

200

150

Ro

taro

d p

erfo

rman

ceT

ime

stay

ing

on

th

e ro

d (

s)(a)

(b)

Ro

taro

d p

erfo

rman

ceT

ime

stay

ing

on

th

e ro

d (

s)

100

50

0day-1 day 7 day 14 day 21 day 28

#

#

##

#

#

#

* *

*

*

* *

*

*

*

*

**

#

#

#

*

#

day-1 day 7 day 14 day 21 day 28

200

150

100

50

0

LactacystinPre-RasPre-Sel


Fig. 2 Effects of rasagiline and selegiline on rotarod performance in

mice injected with lactacystin. Rotarod performance was represented

time staying on the rod. Data were expressed as mean ± SEM (n = 7).

Two-way ANOVA was applied to analysis the difference (F = 27.15, df

for group = 5), followed by post hoc tests corrected for multiple com-

parisons by the method of Tukey. *p < 0.05 versus control and#p < 0.05 versus lactacystin.

(a)

(b)

Fig. 3 Effects of rasagiline and selegiline against lactacystin-induced

loss of DA neurons in SN. Administration of rasagiline (0.2 mg/kg, i.p.

per day) or selegiline (1 mg/kg, i.p. per day) started 7 days before

(pre-treatment) or after (post-treatment) microinjection with lactacy-

stin, while the administration of a same volume saline was served as

control. (a) Representative photomicrographs of SN with TH-immu-

nohistochemistry. (b) Quantitative analysis of TH-immunopositive

neurons in the SN. Each value was presented by the mean ± SEM

based on the number of TH-immunopositive neurons in right SN

(n = 8). There was no difference in the number of TH-positive cells

between left and right SN (data not showed). Group differences were

assessed using one-way ANOVA (F = 105.662, df = 47), allowed by

post hoc tests corrected for multiple comparisons by the method of

Tukey. *p < 0.001 versus control, #p < 0.001 versus Lactacystin, and$p < 0.05 versus selegiline counterparts.



neurons, which can be prevented by pre-treatment with noveliron chelator VK-28 (Varinel Inc.,Westtown, PA, USA) and itsderivate, a multifunctional drug, M30. Furthermore, post-lactacystin treatments with these drugs induce restoration ofDA neurons and alleviate the inhibition of proteasome activity(Zhu et al. 2007). The present study was undertaken todetermine and compare the neuroprotective and neurorestor-ative activities of two anti-Parkinson drugs, rasagiline (azilect)and selegiline, in the lactacystin parkinsonism model. Thiswas performed as rasagiline has been suggested to have adisease modifying activity in PD patients (Parkinson StudyGroup 2004). Our study provides evidence supporting bothneuroprotective and neurorestorative activities for rasagiline,

Fig. 4 Effects of rasagiline and selegiline against lactacystin-induced

depletion of DA and its metabolites. Administration of rasagiline

(0.2 mg/kg, i.p. per day) or selegiline (1 mg/kg, i.p. per day) started

7 days before (pre-treatment) or after (post-treatment) microinjection

with lactacystin, while the administration of a same volume saline was

served as control. The figures represented the abilities of rasagiline

and selegiline to restore the lactacystin-induced depletions of striatal

DA (a) and its metabolites DOPAC and HVA (b). The results were

expressed as mean ± SEM (n = 7). Group differences were assessed

using one-way ANOVA (F = 5.766, df = 41), allowed by post hoc tests

corrected for multiple comparisons by the method of Tukey. *p < 0.01

versus control; #p < 0.05 and ##p < 0.01 versus Lac lactacystin;$p < 0.05 versus selegiline.

Fig. 5 Rasagiline and selegiline failed to alleviate lactacystin-induced

proteasome inhibition. Proteasome activities were examined on day

28 and were indicated by the changes on chymotrypsin-like activity

induced in ventral midbrain. Both rasagiline and selegiline failed to

restore the proteasome activity. The results were expressed as

mean ± SEM (n = 5). Group differences were assessed using one-

way ANOVA (F = 18.532, df = 29), allowed by post hoc tests corrected

for multiple comparisons by the method of Tukey. *p < 0.001 versus

control.

Bcl-2

β-actin

Fig. 6 Rasagiline and selegiline reversed the reduction of Bcl-2

protein level caused by proteasome inhibition. Protein level of Bcl-2 in

the ventral midbrain was analyzed by western blot on day 28. (a)

Representative figures of changes of Bcl-2 in ventral midbrain in mice

of different group. The loading of the lanes was normalized to levels of

b-actin and the experiment is representative of five independent

experiments. (b) The calculated densitometry intensities of the

respective bands were presented as percentage of the control. The

results were expressed as mean ± SEM (n = 5). Group differences

were assessed using one-way ANOVA (F = 41.779, df = 29), allowed by

post hoc tests corrected for multiple comparisons by the

method of Tukey. *p < 0.001 versus control and #p < 0.05 versus

lactacystin.



but only neuroprotective effect for selegiline, against lactacy-stin-induced neurodegeneration. For neuroprotection, a con-tinuous 7-day administration together with a continuous 28-day administration of either rasagiline or selegiline followingthe lactacystin lesion improved behavioral deficits, attenuatedthe severe degeneration of DA neurons and increased striatalDA content, suggesting possible neuroprotective effects of thedrugs. Only rasagiline, however, was able to restore thedegeneration of DA neurons and the depletion of DA contentwhen administered for 21 days following the lactacystinlesion. Thus, although both rasagiline and selegiline may exertbeneficial effects against lactacystin-induced neurodegenera-tion, it is apparent that MAO-B inhibitory activity cannotexplain either the neuroprotective or neurorescue properties ofthese MAO-B inhibitors in the lactacystin-induced neurode-generation. In fact, MAO-B inhibitors have been reported topossibly enhance neurotoxicity by proteasome inhibition(Fornai et al. 2003). In addition, MAO-A is suggested to bethe only isoform involved in striatal DA oxidative metabolism(Gesi et al. 2001). Selegiline, at 1 mg/kg, failed to modify theextracellular DA concentration in rat (Tipton et al. 2004).Furthermore, the optical isomers of rasagiline and selegilineare devoid of MAO-B inhibitory activity, but they have thesame potency in neuroprotection (Youdim et al. 2001). Thus,both rasagiline and selegiline may protect DA neurons andrestore DA concentration against lactacystin-induced neu-rodegeneration through other mechanisms.

Proteasome dysfunction has been suggested to play a role inPD (Grunblatt et al. 2004; McNaught and Olanow 2006), andproteasome inhibitors have been shown to induce degenera-tion of DA neurons in some in vitro and in vivomodels (Fornaiet al. 2003; McNaught et al. 2003; Zhang et al. 2005). Themechanisms whereby proteasome dysfunction leads to deathof DA neurons are, however, are not well understood. PC12cells and DA neurons have been reported to undergo apoptosisfollowing proteasome inhibition (Rideout et al. 2005; Nairet al. 2006), and may be rescued by antiapoptotic agents(Rideout et al. 2005). In linewith these findings, we found thatstereotaxic injection of lactacystin into MFB of micedecreased protein levels of Bcl-2, the over-expression ofwhich has been shown to suppress apoptosis in a wide range ofcell types (Lee et al. 2001). These findings support thatapoptosis occurs in neurodegeneration induced by UPSimpairment and the possibility that antiapoptotic drugs mighthave neuroprotective effects.

Among the agents that have been demonstrated to haveantiapoptotic properties in PD model systems, propargylam-ines are considered to be the most potent and promising(Olanow 2006). Both rasagiline and selegiline are two of thepropargylamines that have been shown to have neuroprotec-tive and antiapoptotic effects and are currently used in theearly treatment of PD (Chen and Swope 2006). Bothrasagiline and selegiline promote free radical scavengingby enhancing superoxide dismutase (SOD1 and 2) and

catalase activities or by increasing the SOD protein levels(Tabakman et al. 2004), increase production of neurotro-phins such as nerve growth factor, brain-derived neurotroph-ic factor, glial cell line-derived neurotrophic factor, andprotect neurons from inflammatory process (Nagatsu andSawada 2006). Furthermore, rasagiline and selegiline areable to bind to glyceraldehyde-3-phosphate dehydrogenase todecrease synthesis of pro-apoptotic proteins like Bcl-2associated X protein (BAX), c-JUN, and glyceraldehyde-3-phosphate dehydrogenase and increase synthesis of anti-apoptotic proteins like Bcl-2, Cu-Zn-SOD, and heat-shockprotein 70 (Tatton et al. 2002). Moreover, rasagiline canprevent the collapse of the mitochondrial membrane potential(Akao et al. 2002) and activate Ras-PI3K-Akt survivalpathway (Sagi et al. 2007). As neurodegeneration inducedby proteasomal inhibition has been shown to be mediated byapoptotic mechanisms (Rideout et al. 2005; Nair et al. 2006)and down-regulation of Bcl-2 has been documented inventral midbrain following lactacystin lesion, aforemen-tioned neuroprotective, and antiapoptotic actions of rasagi-line and selegiline may contribute to the neuroprotection or/and rasagiline-mediated neurorescue in the UPS impairedmouse model.

Our results suggest that rasagiline is superior to selegiline inboth neuroprotection and neurorescue against lactacystin-induced nigrostriatal degeneration. Although both rasagilineand selegiline are propargylamine derivatives and have similarstructures, rasagiline is metabolized to its major metaboliteaminoindan, while selegiline’s major metabolite is L-meth-amphetamine (Finberg et al. 1999; Am et al. 2004). Thedifference of the metabolites of these two drugs may accountfor their different neuroprotective and neurorescue abilities.Aminoindan, the major metabolite of rasagiline, has beenshown to be neuroprotective under the condition of serum andnerve growth factor withdrawal in PC12 cells (Am et al.2004). Thus, the in vivo generation of aminoindan fromrasagiline may contribute to its synergetic neuroprotective andneurorestorative effects. In contract to rasagiline, selegiline isa sympathomimetic amine and its major metabolite L-meth-amphetamine has been shown to be neurotoxic to PC12 andneuroblastoma cells in culture andDAneurons in vivo (Gassenet al. 2003). In addiction, L-methamphetamine has beenshown to abolish the neuroprotective activity of selegilinein vitro (Am et al. 2004). Furthermore, several studies havereported that both in vitro and in vivo administration ofmethamphetamine can induce cytosolic inclusions sharingsimilar components with Lewy bodies occurring in PD (Fornaiet al. 2004) and cause a dose-dependant inhibition of protea-some activity (Fornai et al. 2006; Lazzeri et al. 2006). Theabove findings may explain the observed inferior ability ofselegiline compared with rasagiline in neuroprotection and itsinability to rescue against lactacystin-induced nigrostriataldegeneration, as well as the failure of selegiline in restoringproteasome inhibition induced by lactacystin. In contrast, the



behavioral improvement documented after the treatment ofselegiline may be explained by its sympathomimetic proper-ties mediated via L-methamphetamine. Moreover, rasagilinehas been found to activate the UPS in DA-derived neuroblas-toma SH-SY5Y cells (Youdim et al. 2002), although only amodest restoration of proteasome activity was found in thecurrent study. This may be a dose-dependant effect, the studiesof which are on the way.

The lactacystin-induced inhibition of proteasome activity,demonstrated in the ventral midbrain, was moderately but notsignificantly reversed by the pre- or post-treatment withrasagiline, but was not counteracted by the treatment withselegiline. Thus, the degree of neuroprotection or rescueagainst DA neuron degeneration does not appear to correlatewith the recovery rate of proteasome activity after treatmentwith rasagiline or selegiline. These findings suggest thatrasagiline and selegiline may act relatively independent ofproteasome system. Instead, they may achieve their neuropro-tective and rescue effects though interfering with the down-stream pathways of proteasome system, such as by reversingthe lactacystin-induced Bcl-2 down-regulation as seen in ourstudy. On the other hand, inhibition of proteasome system innigrostriatal pathway inmice results in a significant increase ofiron concentration in ventralmidbrain (Zhu et al. 2007), whichhas been suggested to have a pivotal role in neurodegeneration(Zecca et al. 2004). Iron may further contribute to the processof neurodegeneration, via its ability to induce oxidative stressdependent inhibition of proteasome (Ding et al. 2006). Theinability of rasagiline and selegiline to fully protect againstlactacystin neurotoxicity indicates that chelation of iron maybe important for neuroprotection. This is born by the fact thatthe multifunctional neuroprotective iron chelator-brain selec-tive MAO inhibitor drug, M30, which possess the samepropargylamine moiety as rasagiline and selegiline (Zhanget al. 2005), is potent neuroprotective and neurorestorativeagent in the lactacystin-induced PD model (Zhu et al. 2007).Thus, additional studies into the neuroprotective and possiblydisease-modifying mechanisms are needed.

The present study has investigated the neuroprotective andneurorescue properties of rasagiline and selegiline in vivoagainst nigrostriatal degeneration induced by lactacystin. Ourfindings suggest that rasagiline is superior to selegiline inboth neuroprotection and neurorestoration against neurode-generation in nigrostriatal pathway induced by UPS impair-ment. As rasagiline has been approved by the Food and DrugAdministration for the treatment of PD, well designed long-term observational trials are needed to study its effects on theprogression of the disease.

Acknowledgements

This work was supported by research grants from NIH (NS043567)

and from Chinese 863 project (2007 AA022460). We are grateful for

the support of Teva Pharmaceurtical Co. (Israel).

References

Akao Y., Maruyama W., Shimizu S., Yi H., Nakagawa Y., Shamoto-Nagai M., Youdim M. B., Tsujimoto Y. and Naoi M. (2002)Mitochondrial permeability transition mediates apoptosis inducedby N-methyl(R)salsolinol, an endogenous neurotoxin, and isinhibited by Bcl-2 and rasagiline, N-propargyl-1(R)-aminoindan.J. Neurochem. 82(4), 913–923.

Am O. B., Amit T. and Youdim M. B. (2004) Contrasting neuropro-tective and neurotoxic actions of respective metabolites of anti-Parkinson drugs rasagiline and selegiline. Neurosci. Lett. 355(3),169–172.

Bar-Am O., Amit T. and Youdim M. B. (2007). Aminoindan and hy-droxyaminoindan, metabolites of rasagiline and ladostigil, respec-tively, exert neuroprotective properties in vitro. J. Neurochem. 103,500–508.

Bayes M., Rabasseda X. and Prous J. R. (2006) Gateways to clinicaltrials. Methods Find. Exp. Clin. Pharmacol. 28(9), 657–678.

Blandini F. (2005) Neuroprotection by rasagiline: a new therapeuticapproach to Parkinson’s disease? CNS Drug Rev. 11(2), 183–194.

Chen J. J. and Swope D. M. (2006) Clinical pharmacology of rasagiline:a novel, second-generation propargylamine for the treatment ofParkinson disease. J. Clin. Pharmacol. 45(8), 878–894.

Ciechanover A., Orian A. and Schwartz A. L. (2000) Ubiquitin-mediatedproteolysis: biological regulation via destruction. Bioessays 22(5),442–451.

Ding Q., Dimayuga E. and Keller J. N. (2006) Proteasome regulation ofoxidative stress in aging and age-related diseases of the CNS.Antioxid. Redox Signal. 8(1–2), 163–172.

Finberg J. P., Lamensdorf I., Weinstock M., Schwartz M. and Youdim M.B. (1999) Pharmacology of rasagiline (N-propargyl-1R-aminoin-dan). Adv. Neurol. 80, 495–499.

Fornai F., Lenzi P., Gesi M. et al. (2003) Fine structure and biochemicalmechanisms underlying nigrostriatal inclusions and cell death afterproteasome inhibition. J. Neurosci. 23(26), 8955–8966.

Fornai F., Lenzi P., Gesi M. et al. (2004) Similarities between meth-amphetamine toxicity and proteasome inhibition. Ann. NY Acad.Sci. 1025, 162–170.

Fornai F., Lazzeri G., Bandettini Di Poggio A., Soldani P., de Blasi A.,Nicoletti F., Ruggieri S. and Paparelli A. (2006) Convergent rolesof alpha-synuclein, DA metabolism, and the ubiquitin-proteasomesystem in nigrostriatal toxicity. Ann. NY Acad. Sci. 1074, 84–89.

Gassen M., Lamensdorf I., Armony T., Finberg J. P. and Youdim M. B.(2003) Attenuation of methamphetamine induced dopaminergicneurotoxicity by flupirtine: microdialysis study on dopamine re-lease and free radical generation. J. Neural. Transm. 110(2), 171–182.

Gesi M., Santinami A., Ruffoli R., Conti G. and Fornai F. (2001) Novelaspects of dopamine oxidative metabolism (confounding outcomestake place of certainties). Pharmacol. Toxicol. 89(5), 217–224.

Glaser J. R. and Glaser E. M. (2000) Stereology, morphometry, andmapping: the whole is greater than the sum of its parts. J. Chem.Neuroanat. 20, 115–126.

Grunblatt E., Mandel S., Jacob-Hirsch J. et al. (2004) Gene expressionprofiling of parkinsonian substantia nigra pars compacta; altera-tions in ubiquitin-proteasome, heat shock protein, iron and oxida-tive stress regulated proteins, cell adhesion/cellular matrix andvesicle trafficking genes. J. Neural. Transm. 111(12), 1543–1573.

Lazzeri G., Lenzi P., Gesi M., Ferrucci M., Fulceri F., Ruggieri S., BrunoV. and Fornai F. (2006) In PC12 cells neurotoxicity induced bymethamphetamine is related to proteasome inhibition. Ann. NYAcad. Sci. 1074, 174–177.

Le W., Jankovic J., Xie W., Kong R. and Appel S. H. (1997) (-)-Dep-renyl protection of 1-methyl-4 phenylpyridium ion (MPP+)-



induced apoptosis independent of MAO-B inhibition. Neurosci.Lett. 224(3), 197–200.

Lee M., Hyun D. H., Marshall K. A., Ellerby L. M., Bredesen D. E.,Jenner P. and Halliwell B. (2001) Effect of overexpression ofBCL-2 on cellular oxidative damage, nitric oxide production,antioxidant defenses, and the proteasome. Free Radic. Biol. Med.31(12), 1550–1559.

McNaught K. S. and Olanow C. W. (2006) Protein aggregation in thepathogenesis of familial and sporadic Parkinson’s disease. Neuro-biol. Aging 27(4), 530–545.

McNaught K. S., Perl D. P., Brownell A. L. and Olanow C. W. (2003)Systemic exposure to proteasome inhibitors causes a progressivemodel of Parkinson’s disease. Ann. Neurol. 56(1), 149–162.

Muralikrishnan D., Samantaray S. and Mohanakumar K. P. (2003) D-deprenyl protects nigrostriatal neurons against 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced dopaminergic neurotoxicity.Synapse 50(1), 7–13.

Nagatsu T. and Sawada M. (2006) Molecular mechanism of the relationof monoamine oxidase B and its inhibitors to Parkinson’s disease:possible implications of glial cells. J. Neural Transm. Suppl. 71,53–65.

Nair V. D., McNaught K. S., Gonzalez-Maeso J., Sealfon S. C. andOlanow C. W. (2006) p53 mediates nontranscriptional cell death indopaminergic cells in response to proteasome inhibition. J. Biol.Chem. 281(51), 39550–39560.

Olanow C. W. (2006) Rationale for considering that propargylaminesmight be neuroprotective in Parkinson’s disease. Neurology 66(10Suppl. 4), S69–S79.

Palhagen S., Heinonen E., Hagglund J., Kaugesaar T., Maki-Ikola O.,Palm R. and Swedish Parkinson Study Group (2006) Selegilineslows the progression of the symptoms of Parkinson disease.Neurology 66(8), 1200–1206.

Parkinson Study Group (2004) A controlled, randomized, delayed-startstudy of rasagiline in early Parkinson disease. Arch. Neurol. 61(4),561–566.

Rideout H. J., Lang-Rollin I. C., Savalle M. and Stefanis L. (2005)Dopaminergic neurons in rat ventral midbrain cultures undergoselective apoptosis and form inclusions, but do not up-regulateiHSP70, following proteasomal inhibition. J. Neurochem. 93(5),1304–1313.

Sagi Y., Mandel S., Amit T. and Youdim M. B. (2007) Activation oftyrosine kinase receptor signaling pathway by rasagiline facilitatesneurorescue and restoration of nigrostriatal dopamine neurons inpost-MPTP-induced parkinsonism. Neurobiol. Dis. 25(1), 35–44.

Shoulson I. (1998) DATATOP: a decade of neuroprotective inquiry.Parkinson Study Group. Deprenyl and tocopherol antioxidativetherapy of Parkinsonism. Ann. Neurol. 44(3 Suppl. 1), S160–S166.

Siderowf A. and Stern M. (2003) Update on Parkinson disease. Ann.Intern. Med. 138(8), 651–658.

Tabakman R., Lecht S. and Lazarovici P. (2004) Neuroprotection bymonoamine oxidase B inhibitors: a therapeutic strategy for Par-kinson’s disease? Bioessays 26(1), 80–90.

Tatton W. G., Chalmers-Redman R. M., Ju W. J., Mammen M., CarlileG. W., Pong A. W. and Tatton N. A. (2002) Propargylaminesinduce antiapoptotic new protein synthesis in serum- and nerve

growth factor (NGF)-withdrawn, NGF-differentiated PC-12 cells.J. Pharmacol. Exp. Ther. 301(2), 753–764.

Tipton K. F., Boyce S., O’Sullivan J., Davey G. P. and Healy J. (2004)Monoamine oxidases: certainties and uncertainties. Curr. Med.Chem. 11(15), 1965–1982.

Weinreb O., Amit T., Bar-Am O., Sagi Y., Mandel S. and Youdim M. B.(2006) Involvement of multiple survival signal transduction path-ways in the neuroprotective, neurorescue and APP processingactivity of rasagiline and its propargyl moiety. J. Neural Transm.Suppl. 70, 457–465.

West M. J., Slomianka L. and Gundersen H. J. (1991) Unbiased stere-ological estimation of the total number of neurons in the subdivi-sions of the rat hippocampus using the optical fractionator. Anat.Rec. 231(4), 482–497.

West B. D., Shughrue P. J., Vanko A. E., Ransom R. W. and KinneyG. G. (2006) Amphetamine-induced locomotor activity isreduced in mice following MPTP treatment but not followingselegiline/MPTP treatment. Pharmacol. Biochem. Behav. 84(1),158–161.

Youdim M. B. and Bakhle Y. S. (2006) Monoamine oxidase: isoformsand inhibitors in Parkinson’s disease and depressive illness. Br. J.Pharmacol. 147(Suppl. 1), S287–S296.

Youdim M. B. and Tipton K. F. (2002) Rat striatal monoamine oxidase-Binhibition by L-deprenyl and rasagiline: its relationship to 2-phenylethylamine-induced stereotypy and Parkinson’s disease.Parkinsonism Relat Disord. 8(4), 247–253.

Youdim M. B., Wadia A., Tatton W. and Weinstock M. (2001) The anti-Parkinson drug rasagiline and its cholinesterase inhibitor deriva-tives exert neuroprotection unrelated to MAO inhibition in cellculture and in vivo. Ann. NY Acad. Sci. 939, 450–458.

YoudimM. B., MaruyamaW. and Naoi M. (2002) Neuroprotective actionof rasagiline is dependent of rasagiline is dependent on activation ofBcl-2, PKCandproteasomecomplex and inhibition ofmitochondrialpermeability transition (PT). Neural Plast. 9(2), 125–126.

Youdim M. B., Amit T., Falach-Yogev M., Am O. B., Maruyama W. andNaoi M. (2003) The essentiality of Bcl-2, PKC and proteasome-ubiquitin complex activations in the neuroprotective-antiapoptoticaction of the anti-Parkinson drug, rasagiline. Biochem. Pharmacol.66(8), 1635–1641.

Youdim M. B., Maruyama W. and Naoi M. (2005) Neuropharmaco-logical, neuroprotective and amyloid precursor processing prop-erties of selective MAO-B inhibitor antiparkinsonian drug,rasagiline. Drugs Today (Barc.) 41, 369–391.

Zecca L., Youdim M. B., Riederer P., Connor J. R. and Crichton R. R.(2004) Iron, brain ageing and neurodegenerative disorders. Nat.Rev. Neurosci. 5(11), 863–873.

Zhang X., Xie W., Qu S., Pan T., Wang X. and Le W. (2005) Neuro-protection by iron chelator against proteasome inhibitor-inducednigral degeneration. Biochem. Biophys. Res. Commun. 333(2),544–549.

Zhu W., Wenjie Xie, Pan T., Xu P., Fridkin M., Zheng H., Jankovic J.,Youdim M. B. and Le W. (2007) Novel brain permeable ironchelators prevent lactacystin induced ubiquitin-proteasome inhibi-tion and degeneration of nigrostriatal dopamine neurons. FESAB J.21(14), 3835–3844.



Documents

Comparison of neuroprotective and neurorestorative capabilities of rasagiline and selegiline against lactacystin-induced nigrostriatal dopaminergic degeneration