Selegiline increases on time without exacerbation of dyskinesia in 6-hydroxydopamine-lesioned rats displaying l-Dopa-induced wearing-off and abnormal involuntary movements

3,4-Dihydroxy-l-phenylalanine (l-Dopa) remains the most effective drug for treating the motor symptoms of Parkinson's disease (PD). However, its long-term use is limited due to motor complications such as wearing-off and dyskinesia. A clinical study in PD patients with motor complications has demonstrated that selegiline, a monoamine oxidase type B inhibitor, is effective in reducing off time without worsening dyskinesia, although another study has shown worsening dyskinesia. Here, using unilateral 6-hydroxydopamine-lesioned rats showing degeneration of nigrostriatal dopaminergic neurons and l-Dopa-induced motor complications, we determined the efficacy of selegiline in controlling l-Dopa-induced motor fluctuations and exacerbated dyskinesia. Repeated administration of l-Dopa/benserazide (25/6.25 mg/kg, intraperitoneally, twice daily for 22 days) progressively shortened rotational response duration (on time) and augmented peak rotation in lesioned rats. Single subcutaneous injection of selegiline (10 mg/kg) extended l-Dopa-induced shortened on time without augmenting peak rotation. Furthermore, l-Dopa/benserazide (25/6.25 mg/kg, intraperitoneally, once daily for 7 days) progressively increased abnormal involuntary movements (l-Dopa-induced dyskinesia, LID) and peak rotation. Single subcutaneous injection of selegiline (10 mg/kg) did not exacerbate LID or alter mRNA expression of prodynorphin (PDy) and activity-regulated cytoskeleton-associated protein (Arc), both mRNAs associated with LID in the lesioned striatum. Despite undetectable plasma concentrations of selegiline and its metabolites at 24 h post-administration, these on time and LID effects did not decrease, suggesting involvement of irreversible mechanisms. Altogether, these results indicate that selegiline is effective in increasing on time without worsening dyskinesia.


Introduction
Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by motor symptoms such as bradykinesia, rigidity, tremor, and postural instability, which mainly arise from dysfunction of the nigrostriatal dopaminergic pathway [1,2]. 3,4-Dihydroxy-L-phenylalanine (L-Dopa) is widely recognized as the most efficacious medication for the motor symptoms of PD. However, motor complications appear in approximately 40% of patients with PD who have taken L-Dopa for 4-6 years [3]. Motor complications such as dyskinesia and motor fluctuations, including wearing-off (unpredictable switching between on and off phases), have a strong negative influence on patients' quality of life [4]. The mechanisms underlying motor complications have not yet been fully elucidated, but may be related to loss of dopaminergic nerve terminals with disease progression and to pulsatile stimulation of dopamine (DA) receptors by short-acting dopaminergic agents, such as L-Dopa [5][6][7][8][9]. Recently, the concept of continuous dopaminergic stimulation (CDS) has emerged as a therapeutic strategy for PD management [10,11]. Several studies have shown that continuous infusion of short-acting or long-acting dopaminergic agents leads to a reduction in the risk of motor complications [12][13][14][15][16]. In addition, combined treatment of L-Dopa and its metabolic inhibitors, such as catechol-O-methyltransferase (COMT) inhibitors or monoamine oxidase type B (MAO-B) inhibitors, theoretically provides longer CDS of DA receptors than treatment with standard L-Dopa treatment, and thereby might reduce the risk of motor complications. However, in a prospective double-blind study on the risk of developing L-Dopa-induced dyskinesia (LID) in PD patients, combined treatment of L-Dopa and a COMT inhibitor, entacapone, did not delay the time of onset or reduce the frequency of dyskinesia [17]. The authors speculated that the dosing frequency of L-Dopa and entacapone may not have sufficiently provided continuous L-Dopa delivery [18]. The potential effects of CDS provided by combined treatment of L-Dopa and its metabolic inhibitors have not been fully elucidated.
As an adjunct to L-Dopa, a MAO-B inhibitor, selegiline, has been reported to improve motor fluctuations [19][20][21][22]. Indeed, an orally disintegrating formulation of selegiline (using Zydis ® technology) bypasses the first-pass effect [23], and reduces off time without worsening dyskinesia [24]. Furthermore, a conventional formulation of selegiline has been reported to ameliorate off time symptoms, although there is insufficient evidence regarding off time [25,26]. Regarding dyskinesia, a long-term trial has demonstrated that selegiline administration improves off time symptoms, but with somewhat worsening dyskinesia [26]. In contrast, other studies have shown that adjunctive therapies with MAO-B inhibitors do not increase the incidence of dyskinesia compared with L-Dopa therapy [27,28]. Thus, the aim of the present study was to determine whether selegiline extends rotational response duration (on time) shortened by repeated administration of L-Dopa without worsening dyskinesia, and whether such effects correlate to plasma concentrations of selegiline and its metabolites. Herein, we used unilateral 6-hydroxydopamine (6-OHDA)-lesioned rats showing degeneration of nigrostriatal dopaminergic neurons and L-Dopa-induced motor complications to demonstrate that selegiline exerts a significant increasing effect on shortened on time without worsening dyskinesia. This effect occurred without augmentation of mRNA expression in the lesioned striatum of prodynorphin (PDy) and activity-regulated cytoskeleton-associated protein (Arc), both postsynaptic striatal factors associated with LID. Furthermore, in lesioned rats at 24 h after selegiline administration, plasma concentrations of selegiline and its metabolites were below the lower limit of quantification, but MAO inhibitory effects and increasing on time effects were observed. Altogether, the mechanisms underlying the beneficial effects of selegiline on L-Dopa-induced motor complications may be attributable to irreversible mechanisms such as MAO inhibition.

Animals
Male Sprague-Dawley rats (7 weeks old, Nihon SLC, Shizuoka, Japan) were maintained in a facility with controlled humidity (50 ± 20%) and temperature (23 ± 2°C), under a 12-h light/dark cycle (lights on at 7:00 a.m.) with free access to food (Oriental Yeast, Tokyo, Japan) and water. Rats were acclimated for 5-7 days before use for experiments. This study was performed in accordance with the National Institute of Health (NIH) guide for the care and use of laboratory animals, and the institutional guideline for the care and use of laboratory animals.

Compounds
L-Dopa (Sigma-Aldrich, St. Louis, MO, USA) was suspended in saline containing 0.25% sodium carboxymethylcellulose. Benserazide hydrochloride (HCl) (Sigma-Aldrich) was dissolved in saline. L-Dopa and benserazide were intraperitoneally (i.p.) administered at a dose of 25 mg/kg and 6.25 mg/kg, respectively. Selegiline HCl (Fujimoto Pharmaceutical Corporation, Osaka, Japan) was dissolved in saline and administered by a single subcutaneous (s.c.) injection at 3 or 10 mg/kg, doses reported to have anti-parkinsonian effects in several animal models [29,30]. Desipramine HCl (Sigma-Aldrich) was dissolved in distilled water and injected i.p. at 25 mg/kg. 6-OHDA HCl (Sigma-Aldrich) was dissolved in saline containing 0.02% ascorbic acid (vehicle). Apomorphine HCl hemihydrate (Sigma-Aldrich) was dissolved in saline containing 0.2% ascorbic acid, and administered by a single s.c. injection at 0.05 mg free base/kg. The unilateral 6-OHDA-lesioned rat has been widely used as a model of hemi-parkinsonism. This model shows degeneration of nigrostriatal dopaminergic neurons and rotational behavior contralateral to the lesion in response to DA agonists and L-Dopa. Repeated L-Dopa injections into unilateral 6-OHDA-lesioned rats result in reduced rotational response duration, and increased peak rotation and abnormal involuntary movements (AIMs) [31][32][33][34][35][36], which resemble the wearingoff phenomenon and LID that are observable in PD patients chronically treated with L-Dopa. Under anesthesia with sodium pentobarbital (25.9 mg/kg, i.p.) and isoflurane, each rat received a single vehicle (4 μL) or 6-OHDA injection (8 μg free base/4 μL) into the right medial forebrain bundle (anteroposterior = -4.0, mediolateral = +1.3, dorsoventral = −8.4 to the skull surface, according to the atlas of Paxinos and Watson [37]) at a rate of 0.8 μL/min. Rats were administered with desipramine HCl (25 mg/kg, i.p.), which preserves noradrenergic neurons, 30 min prior to 6-OHDA lesion. Following a 3-week recovery period, rats exhibiting > 100 contralateral turns over 60 min in response to apomorphine (0.05 mg/kg, s.c.) were selected as successfully 6-OHDA-lesioned rats for subsequent tests. Rats meeting this apomorphine screening criterion are reported to show > 95% loss of dopaminergic neurons [31].

Rotational behavior
To induce a wearing-off-like phenomenon, 6-OHDA-lesioned rats were administered chronic treatment of L-Dopa/benserazide (25/ 6.25 mg/kg, respectively, i.p.) twice daily for 24 or 25 days from the 7th day after apomorphine screening. Rotational behavior was measured for 4 h after L-Dopa injection on days 1,8,15,22,24, and 25. The number of complete (360°) turns during each 5-min period was recorded using an automated rotameter (MED Associates, St. Albans, VT, USA), or manually counted using a multiple tally counter by an observer blinded to the treatment regimen. Lesioned rats received a single s.c. injection of selegiline HCl (3 or 10 mg/kg) 30 min prior to first administration of L-Dopa on day 24.
Changes in L-Dopa response in 6-OHDA-lesioned rats were evaluated using the following previously described method [32,38]: (1) the maximum number of rotations in any 5-min interval was designated the peak rotation; and (2) the rotational response duration following L-Dopa injection (on time) was calculated between the timepoint elapsing from the 5-min interval when rotations exceeded 20% of peak rotation and the timepoint when rotations fell below 20% of peak rotation (Fig. 1B). Alteration rate of on time in selegiline-treated rats (day 24) was expressed as a percentage of on time following L-Dopa treatment alone (day 22) (Fig. 2A).
On days 1, 5, 6, and 7 of L-Dopa treatment, AIMs were individually observed in 6-OHDA-lesioned rats after L-Dopa injection. Rats were placed in a plastic cage and AIMs recorded in a single blind manner for 1 min every 20-min interval for a duration of 180 min. AIMs were classified into three subtypes, as described previously [33,34]: (a) axial AIM, contralateral twisted position of the neck and upper body; (b) forelimb AIM, repetitive jerking or grabbing motion of a contralateral forelimb; and (c) orolingual AIM, repetitive purposeless jaw movements and tongue protrusions. For each of these three subtypes, LID severity was scored on a four-point scale: 0 = not present; 1 = occasional; 2 = frequent; 3 = continuous but interrupted by sensory distraction; and 4 = continuous and not interrupted by sensory distraction. Axial, forelimb, and orolingual (ALO) AIM scores were calculated by adding the respective three individual dyskinesia scores. To compare anti-parkinsonian effects between saline and selegiline treatment following L-Dopa injection, rotational behavior was measured on days 1, 5, 6, and 7 of L-Dopa treatment using an automated rotameter (MED Associates) as described in 2.3.2.

MAO activities in rat brain
Brain MAO activity in rats 30 min or 24 h after single injection of saline or selegiline was measured according to the previously described method [39,40]. Whole brains were removed immediately after Aliquots of extracts were subjected to liquid scintillation counting to determine radioactivity.

Plasma concentrations of selegiline and its metabolites following selegiline administration
Under anesthesia with isoflurane, blood samples (150 μL) were collected from the jugular vein 30 min and 24 h after selegiline administration, and stored at −20°C after centrifugation (2,000 × g, 10 min). Each plasma sample was extracted with 1-chlolobutane/acetonitrile (4/1, v/v), and then 0.5% HCl (back extraction). Plasma concentrations of selegiline and its metabolites (N-desmethylselegiline, methamphetamine, and amphetamine) were determined by liquid chromatography-tandemmass spectrometry [41] with an Inertsil C8 column (GL Sciences, Tokyo, Japan) without chiral derivatization [42]. A lower limit (1 ng/mL) of quantification was used for each measurement.

Statistical analysis
Statistical analyses were performed using SPSS 23.0 (IBM, Armonk, NY, USA). Data are expressed as mean ± SEM. Paired t-tests were Fig. 2. Effect of single subcutaneous injection of selegiline on L-Dopa-induced shortened on time in unilateral 6-OHDA-lesioned rats. (A) Drug treatment and behavioral test schedule. On day 22, 6-hydroxydopamine (6-OHDA)-lesioned rats were grouped into three groups based on percentage of on time following L-Dopa treatment (day 1). (B) Single subcutaneous (s.c.) injection of selegiline extended on time following L-Dopa administration (day 24). Saline-treated rats (opened circles, n = 17) and selegiline (10 mg/kg)-treated rats (closed squares, n = 13). Values represent mean + SEM. (C, D) Single s.c. injection of selegiline at 10 mg/kg, but not 3 mg/kg, resulted in a significant increase of L-Dopa-induced shortened on time (C) without increasing peak rotation (D). Alteration rate of on time in selegilinetreated rats expressed as percentage of on time following L-Dopa treatment alone (day 22). Dashed line indicates alteration rate on day 1 expressed as a percentage of day 22. Saline-treated group (n = 17), selegiline (3 mg/kg)-treated group (n = 12), and selegiline (10 mg/kg)-treated group (n = 13). Values represent mean + SEM. ** P < 0.01 vs. saline-treated group (Tukey's test). (E, F) On time (E) and peak rotation (F) following L-Dopa injection in lesioned rats that received a single s.c.
injection of selegiline at 30 min or 24 h prior to L-Dopa administration. Saline-injected group (n = 5) and selegiline (10 mg/kg)-injected group (n = 8). Values represent mean + SEM. Table 1 Brain monoamine oxidase inhibition and plasma concentration of selegiline and its metabolites following selegiline administration. Values represent mean ± SEM (n = 5-6). Percentage of brain monoamine oxidase (MAO) inhibition in rats following administration of 10 mg/kg selegiline was calculated from MAO activity in saline-treated rats. N.D.; not detected (below the lower limit of quantification: < 1 ng/mL). performed to evaluate differences in rotational response between designated L-Dopa treatment days in 6-OHDA-lesioned rats. Comparisons between saline-and selegiline (3 and 10 mg/kg)-treated groups were performed by analysis of variance (ANOVA) followed by post hoc Tukey's test. Saline-and selegiline (10 mg/kg)-treated groups were compared using Student's t-test. Data on mRNA expression were analyzed using ANOVA followed by Student's t-test. Differences were considered statistically significant at P value < 0.05.

Results
In unilateral 6-OHDA-lesioned rats that showed marked loss of THpositive neurons in the substantia nigra (Fig. 1A), repeated i.p. injections of L-Dopa (25 mg/kg with 6.25 mg/kg benserazide, twice daily for 22 days) resulted in reduced on time (P < 0.01; Fig. 1D) and increased magnitude of rotational behavior and peak rotation (P < 0.01; Fig. 1E). This appearance is consistent with previous reports for 6-OHDA-lesioned rats [31,43], and possibly resembles the wearing-off phenomenon that is observable in PD patients chronically treated with L-Dopa. On day 24 of L-Dopa treatment, single s.c. injection of selegiline (10 mg/kg) resulted in significant recovery of shortened on time (P < 0.01), without affecting peak rotation (Fig. 2B-D). Alteration rates in L-Dopa-induced on time in lesioned rats that received a single s.c. injection of selegiline (3 and 10 mg/kg) were 127.3 ± 7.6% and 153.8 ± 12.9%, respectively, of on time post-L-Dopa alone treatment (day 22). To investigate the mechanism underlying this beneficial effect of selegiline on L-Dopa-induced shortened on time, we evaluated L-Dopa-induced on time in lesioned rats 30 min and 24 h following selegiline administration. On day 24, on time and peak rotation following L-Dopa injection in lesioned rats at 24 h post-selegiline administration was comparable to the same rats at 30 min post-selegiline administration ( Fig. 2E and F). Although plasma concentrations of selegiline and its metabolites were below the lower limit of quantification at 24 h following selegiline administration, MAO inhibitory activity was comparable with those 30 min post-selegiline administration (Table 1). These results suggest that selegiline increases L-Dopa-induced shortened on time without augmenting maximum L-Dopa efficacy, and that its effects are due to plasma concentration-independent and irreversible mechanisms.
One therapeutic strategy for management of the wearing-off phenomenon in PD patients is to select adjunctive (add-on) medications that do not augment the risk of development or exacerbation of dyskinesia, and do not reduce L-Dopa efficacy. In rats exhibiting a wearingoff-like phenomenon, a single s.c. injection of selegiline (10 mg/kg) did not lead to a significant increase in peak rotation of L-Dopa (Fig. 2), suggesting that selegiline does not exacerbate LID. Thus, we further investigated whether selegiline administration exacerbates LID using unilateral 6-OHDA-lesioned rats. Repeated i.p. injections of L-Dopa (25 mg/kg with 6.25 mg/kg benserazide, once daily for 6 days) resulted in progressive increases in peak rotation and total ALO AIM score, as described in previous reports [35,44]. On day 6, peak rotation induced by combined selegiline and L-Dopa treatment was comparable with L-Dopa treatment alone (Fig. 3B), suggesting that selegiline does not affect peak efficacy of L-Dopa. There were no significant differences in total ALO AIM scores post-L-Dopa administration between salinetreated and selegiline-treated rats (Fig. 3C). Interestingly, in selegilinetreated rats, a reduction in ALO AIM score (P < 0.05; Fig. 3D-G) was observed at 0-100 min, and especially in orolingual and limb AIM scores (day 6: orolingual, P < 0.01; limb, P < 0.01, Fig. 3E; day 7: limb, P < 0.01, Fig. 3G) compared with saline-treated rats, indicating that selegiline does not exacerbate LID. Selegiline-treated rats showed L-Dopa-induced ALO AIM for 180 min, while scores in saline-treated rats rapidly decreased at 160-180 min post-L-Dopa injection ( Fig. 3D and F). There were significant differences in L-Dopa-induced axial AIM scores at 120-180 min between saline-treated and selegiline-treated rats (P < 0.05; Fig. 3E). This suggests that decreased AIM score at ≥120 min post-L-Dopa injection in saline-treated rats is due to substantial reduction of L-Dopa efficacy.
We further evaluated the effect of L-Dopa and/or selegiline administration on upregulation of PDy and Arc mRNA in the striatum of 6-OHDA-lesioned rats expressing LID. In unilateral 6-OHDA-treated rats, decreased PDy and Arc mRNA expression was found in the lesioned striatum compared with sham-operated rats (P < 0.01). Repeated i.p. injections of L-Dopa induced a significant and marked increase in PDy and Arc mRNA expression in the lesioned striatum (P < 0.01; Fig. 4A and B), coincident with previous findings [36,[45][46][47][48]. Single L-Dopa injection enhanced Arc mRNA expression, but not PDy mRNA expression. Single selegiline injection did not modify increased PDy and Arc mRNA expression in the lesioned striatum of rats that received repeated L-Dopa injections ( Fig. 4A and B). These data support our behavioral test results indicating that selegiline does not exacerbate LID (Fig. 3).

Discussion
The wearing-off phenomenon is attributed to reductions in presynaptic DA handling and storage, and buffering capacity of striatal DA receptors to fluctuating plasma concentrations of L-Dopa, due to progressive degeneration of nigrostriatal dopaminergic neurons [9]. In the present study, repeated L-Dopa injections in 6-OHDA-lesioned rats showing marked loss of nigral dopaminergic neurons progressively resulted in shortened L-Dopa-induced on times, and increased peak rotations and AIM scores. These changes in L-Dopa response possibly mimic the wearing-off phenomenon and dyskinesia observed in most advanced PD patients who have been chronically treated with short-acting dopaminergic medications such as L-Dopa. Addition of DA agonists, COMT inhibitors, or MAO-B inhibitors are implemented in management of L-Dopa-induced wearing-off [4,49]. A randomized, parallel group, double-masked study demonstrated that an orally disintegrating formulation of selegiline (using Zydis ® technology) significantly shortened off time by 2.2 h/day (32%) and extended dyskinesia-free on time by 1.8 h/day [24]. Rascol et al. [50] demonstrated that another MAO-B inhibitor, rasagiline, shortened off time by 1.2 h/day and extended on time without troublesome dyskinesia by 0.85 h/day. Furthermore, the efficacy magnitude of rasagiline was similar to a COMT inhibitor, entacapone (off time: 1.2 h/day, on time: 0.85 h/day). In the present wearing-off animal model, single selegiline injection (10 mg/kg) extended on time without augmenting peak rotation (Fig. 2). Because entacapone also had a similar effect on L-Dopa-induced shortened on Fig. 3. Effect of single subcutaneous injection of selegiline on L-Dopa-induced AIMs in unilateral 6-OHDA-lesioned rats. (A) Drug treatment and behavioral test schedule. On day 5, 6-hydroxydopamine (6-OHDA)-lesioned rats were grouped based on percentage of peak rotation following L-Dopa treatment (day 1). (B) Single subcutaneous (s.c.) injection of selegiline (arrow) did not influence peak rotation induced by repeated intraperitoneal (i.p.) injections of L-Dopa (once daily, for 7 days) in unilateral 6-OHDA-lesioned rats. Saline-treated group (opened circles, n = 12) and selegiline (10 mg/kg)-treated group (L-Dopa alone, opened squares; L-Dopa and selegiline co-treatment, closed squares; n = 10). Values represent mean + SEM. (C) Single s.c. injection of selegiline (arrow) did not influence total axial, forelimb, and orolingual (ALO) abnormal involuntary movement (AIM) score induced by repeated i.p. L-Dopa injections. Saline-treated group (opened circles, n = 10) and selegiline (10 mg/kg)-treated group (L-Dopa alone, opened squares; L-Dopa and selegiline co-treatment, closed squares; n = 9). (D, F) Change in L-Dopainduced ALO AIM score in rats at 30 min (day 6, D) and 24 h (day 7, F) following administered saline or selegiline. Saline-treated group (opened circles, n = 10) and selegiline (10 mg/kg)-treated group (closed squares, n = 9). Values represent mean + SEM. * P < 0.05, ** P < 0.01 vs. saline-treated group (Student's t-test). (E, G) The effect of selegiline on AIM subtypes (axial, limb, and orolingual AIM) induced by L-Dopa injection in rats at 30 min (day 6, E) and 24 h (day 7, G) following administered saline or selegiline. Sal, saline; Sele, selegiline. Values represent mean. * P < 0.05, ** P < 0.01 vs. saline-treated group (Student's t-test).
time in this animal model (Supplementary figure), our results are consistent with the above-mentioned clinical studies [24,50]. To control L-Dopa-induced motor complications in clinical practice, dopaminergic medications are used to improve the wearing-off phenomenon, but unfortunately may provoke or exacerbate dyskinesia. Microdialysis studies have shown that selegiline significantly increases extracellular DA concentration in the striatum of rats and primates following L-Dopa injection [51][52][53], suggesting that elevated peak DA concentration may exacerbate dyskinesia. A long-term trial demonstrated that co-administration of selegiline and L-Dopa improves off time symptoms but worsens dyskinesia [26]. In contrast, no increase in the incidence of dyskinesia was recently reported with adjunctive therapy using MAO-B inhibitors [27,28], and a retrospective study suggested that long-term use of MAO-B inhibitors was associated with a lower risk of dyskinesia [54]. In our dyskinesia model, selegiline administration augmented neither L-Dopa-induced AIM score nor PDy and Arc mRNA expression in the lesioned striatum of rats that showed LID (even on day 1) and progressive exacerbation. Thus, these results suggest that selegiline improves the wearing-off phenomenon without worsening dyskinesia, consistent with findings from previous clinical studies [24,27,28]. Interestingly, selegiline treatment led to a decrease in orolingual and limb AIM scores during 100 min post L-Dopa administration. However, single administration of selegiline did not suppress L-Dopa-induced upregulation of PDy or Arc mRNA, yet significant correlations are reported between upregulating degrees of PDy or Arc mRNA expression in the lesioned lateral striatum and ALO AIM score [36,45,48,55,56]. Further investigation is required to clarify the mechanisms underlying the beneficial effects of selegiline on orolingual and limb AIMs.
Selegiline exerts enhancing effects on dopaminergic transmission, mediated through irreversible MAO-B inhibition and weak reversible inhibition of DA reuptake [57,58]. Consequently, we determined whether irreversible mechanisms (such as MAO inhibition) or reversible mechanisms (such as DA reuptake inhibition) by selegiline and its metabolites contribute to the recovery effects on motor complications. On time and dyskinesia score in lesioned rats at 30 min following single administration of selegiline were comparable to those at 24 h (Figs. 2E and 3C). In addition, lesioned rats at 24 h post-selegiline administration showed orolingual and limb AIM scores at 0-100 min post-L-Dopa administration that were comparable to scores at 30 min ( Fig. 3E and G). Plasma concentrations of selegiline and its metabolites at 24 h following single s.c. injection of selegiline (10 mg/kg) were below the lower limit of quantification (1 ng/mL). Meanwhile, brain MAO activity at 24 h post-injection of selegiline (10 mg/kg) was comparable with rats at 30 min following selegiline administration (Table 1). These results suggest that ameliorating effects of selegiline on L-Dopa-induced shortened on time and LID are mediated by irreversible action of selegiline, probably via MAO inhibition. Thus, if addition of selegiline in PD patients taking L-Dopa induces dyskinesia, reducing L-Dopa dose may efficaciously counteract the dyskinesia, as previously described [59]. Heinonen and Rinne [60] suggested that the use of selegiline in patients taking maximal L-Dopa doses would allow a 20% reduction in L-Dopa dose, which may reduce peak-dose side effects. Conventional and Zydis ® -type formulations of selegiline might be useful for treating motor complications at sufficient doses through irreversible actions of selegiline such as MAO inhibition. The effective dose (10 mg/ kg) of selegiline in our PD model was higher than the therapeutic doses in PD patients (5 -10 mg/day). This difference in effective dose may reflect different physiological roles of striatal MAO-A and MAO-B between rodents and humans, and pathological differences between unilateral 6-OHDA-lesioned rats and PD patients. Moreover, because of our single (and not chronic) treatment design, we cannot exclude the possibility that a higher dose of selegiline was required to exert its beneficial effect on wearing-off. Extrapolation to chronic clinical doses of selegiline requires the use of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated primates, in which MAO-B distribution in the brain is similar to humans.

Declarations of Interest
All authors are employees of Fujimoto Pharmaceutical Corporation.