Preventative Effects of 1-Methyl-1,2,3,4- Tetrahydroisoquinoline Derivatives (N-Functional Group Loading) On MPTP-Induced Parkinsonism In Mice

Hiroko Munakata Ohu University: Ohu Daigaku Risa Ishikawa Ohu University: Ohu Daigaku Toshiaki Saitoh Nihon Pharmaceutical University: Nihon Yakka Daigaku Toshie Kambe Showa Pharmaceutical University: Showa Yakka Daigaku Terumasa Chiba Nihon Pharmaceutical University: Nihon Yakka Daigaku Kyoji Taguchi Showa Pharmaceutical University: Showa Yakka Daigaku Kenji Abe (  k-abe@nichiyaku.ac.jp ) Nihon Pharmaceutical University: Nihon Yakka Daigaku

Selegiline ((R)-N,α-dimethyl-N-2-propynyl-phenethylamine), an inhibitor of monoamine oxidase B (MAO-B), is similar in structure to TIQ derivatives, and has been used as a PD treatment (Gerlach et al., 1996). Selegiline decreases endogenous 1-BnTIQ content in the mouse brain (Kotake et al., 1998) and shows stereoselective and neuroprotective features in glutamate receptor-mediated toxicity of mesencephalic dopamine neurons (Mytilineou et al., 1997). Therefore, we expect that TIQ derivatives with structures similar to selegiline could be more effective as PD therapeutics. We have previously studied the pharmacological characteristics of several arti cially synthesized TIQ derivatives. These results showed that the cytotoxicity of 1-alkyl-TIQs depended on the lipophilicity of the alkyl group (Kitabatake et al., 2009), and the neuroprotective effects of certain synthetic TIQ derivatives were enhanced by the loading of N-propargyl groups, which is a component of the selegiline chemical structure (Katagiri et al., 2010;Saito et al., 2013).
In the present study, we examined the neuroprotective effects of 1-MeTIQs loaded with a different Nfunctional groups (N-propyl, N-propenyl, N-propargyl, or N-butynyl) to estimate the role of bond number and carbon number in the functional groups. These functional groups possess three carbons with a single bond (N-propyl), one double bond (N-propenyl), one triple bond (N-propargyl), and four carbons with one triple bond (N-butynyl). The chemical structures of selegiline, 1-MeTIQ, and the four 1-MeTIQs loaded with N-functional groups are shown in Fig. 1.

Animals and drug administration
Seven-weeks-old male C57BL/6N strain mice were purchased from Charles River Japan Inc. All mice were handled in accordance with the guidelines for animal care and use provided by the National Institutes of Health and the code of ethics for laboratory animals of Ohu University (Fukushima, Japan). All possible means were used to eliminate pain for the experimental animals. The mice were kept in a temperatureand humidity-controlled environment where they could freely consume food and water. After an acclimation period, various doses of 1-MeTIQ (80 mg/kg), 1-Me-N-propyl-TIQ, 1-Me-N-propenyl-TIQ, 1-Me-N-propargyl-TIQ, and 1-Me-N-butynyl-TIQ (20, 40, and 80 mg/kg) were administered intraperitoneally (i.p.) twice a day, every 12 hours, for 4 consecutive days (days 1-4). Three days later, 30 mg/kg MPTP or saline was given intraperitoneally twice a day for 4 consecutive days (days 7-10). The hydrochloride salt of each compound was dissolved in physiological saline and used for administration. The volume used for administration was 10 mL/kg. Experiments were performed using 7-9 animals/group beginning on day 19. A diagram summarizing the drug administration is shown in Fig. 2.

Evaluation of bradykinesia
Bradykinesia is an indicator that is indicative of movement disorders. To assess bradykinesia, the pole test was performed on day 9 after completion of drug administration as described by Ogawa et al. (1985). Mice were placed on the top of a rough iron bar (10 mm in diameter and 55 cm in height) with their heads facing upward. We then measured the time it took for the animal's head to completely turn downward (de ned as the turn time: T turn ) and the time it took for its limbs to completely touch the pole base (the locomotor activity time: T LA ). This measurement was performed ve times consecutively, and the average value was used as the measurement value.
Measurement of monoamine content in the striatum Nine days after the last dose of drug, animals were euthanized by decapitation under sodium pentobarbital (50 mg/kg, i.p.) anesthesia. The brain was quickly removed and the striatum was dissected from the both hemispheres. Monoamine content in the striatum was measured using the method of Kanthasamy et al. (1997).
Ag/AgCl. Immunohistochemical detection of tyrosine hydroxylase (TH) positive cells Perfusion xation was performed by injecting 0.9% physiological saline into the cardiovascular vessels of mice (under sodium pentobarbital anesthesia; 50 mg/kg, i.p.), followed by perfusion with cold 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS, pH 7.4). The brain (including the brainstem) was quickly removed and embedded in OCT compound and rapidly frozen in liquid nitrogen. Twenty µm thin sections of the substantia nigra pars compacta (bregma −3.0 to −3.1 mm) were prepared according to the atlas of Franklin and Paxinos (1997). TH immunostaining was performed using a VECTASTAIN Elite ABC kit and a DAB Substrate kit for peroxidase (Vector Laboratories, CA, USA) according to the manufacturer's instruction manual. Sections were xed with acetone and incubated with normal goat serum for 20 min at room temperature, then treated with anti-TH antibody (1:500, rabbit polyclonal antibody raised against amino acids 1-196 of TH of human origin; sc-14007, Santa Cruz Biotechnology, CA, USA) and incubated overnight at 4°C.
Sections were washed with PBS and incubated with biotinylated secondary antibody (1:200) for 30 min at room temperature. After washing with PBS, an avidin-biotin complex was applied to the sections for 30 min. The sections were washed with PBS again and diaminobenzidine was applied until the appropriate staining intensity was obtained. The number of cells stained by the TH antibody was counted throughout the entire substantia nigra pars compacta area.
Western blot analysis of striatal dopamine transporter (DAT) levels Dissection of the striatum was performed as described above. The striatum was weighed and homogenized in ice-cold 0.1 M PBS (containing 0.32 M sucrose, 0.2 mM PMSF, 2 mM EDTA, and protease inhibitor cocktail) at a ratio of 1000 µL/100 mg tissue weight. Sample preparation for western blot analysis was performed according to Kobayashi et al. (2012), with slight modi cations. Tissue homogenates were centrifuged at 1,000 × g for 5 min at 4°C, and the supernatant was transferred to a fresh tube. Ice-cold homogenization buffer was added to the resulting pellet, which was re-homogenized and centrifuged. Both supernatants were mixed and recentrifuged at 15,000 × g for 20 min at 4°C. The resulting pellet was dissolved in ice-cold radio-immunoprecipitation assay lysis buffer (EzRIPA Lysis kit, ATTO, Tokyo, Japan) for sample determination. Samples were applied to 10% SDS polyacrylamide gel electrophoresis, and then transferred to polyvinylidene di uoride membranes and the blots were incubated with anti-DAT antibody (1:5000, MAB369, Merck Millipore, MA, USA). The membranes were subsequently incubated with peroxidase-conjugated secondary antibody. The densities of DAT protein (immunoreactive bands) were analyzed using image analysis software (ImageJ).

Thiobarbituric acid reactive substances (TBARS) levels in the substantia nigra
The animals were sacri ced by decapitation under sodium pentobarbital anesthesia, as above. The brains were removed and the substantia nigra was isolated. The substantia nigra was weighed and homogenized using ice-cold radio-immunoprecipitation assay lysis buffer (containing protease and phosphatase inhibitors, EzRIPA Lysis kit, ATTO). TBARS levels in the substantia nigra were assayed using a TBARS Assay Kit (Cayman Chemical, MI, USA) with malondialdehyde as the standard compound.

Data analysis
Data were expressed as mean ± standard error (SE) for each group. Initially, determination of signi cant differences between the control (saline + saline) and MPTP-treated groups (saline + MPTP) was performed using the Student's or Aspin-Welch's t-test.
Where signi cant differences were identi ed between the control and MPTP-treated group, the MPTPtreated group was used as a control in a one-way analysis of variance (one-way ANOVA) with a subsequent Dunnett's multiple comparison test to identify signi cant differences. A value of p < 0.05 was considered statistically signi cant.

Effects of N-loaded 1-MeTIQs on MPTP-induced bradykinesia
The effect of 1-MeTIQ and 1-MeTIQ derivatives loaded with an N-functional group on inhibition of MPTPinduced bradykinesia was examined. The MPTP treated group (saline + MPTP) T turn value (2.47 ± 0.11 s) was signi cantly prolonged relative to that of the control group (saline + saline, 1.33 ± 0.03 s, p < 0.01). 1-MeTIQ and its derivatives showed a signi cant effect on MPTP-induced bradykinesia (T turn , p < 0.01).
Effects of N-loaded 1-MeTIQs on striatal monoamine content Striatal dopamine content in the saline + MPTP treatment group was signi cantly decreased compared to the saline + saline group at 9 days after completion of drug administration. Striatal dopamine content in the MPTP-treated group was decreased to 29.5% compared to the control group (MPTP: 3.7 ± 0.2 ng/mg vs. control: 12.5 ± 0.4 ng/mg, p < 0.01). Although pretreatment with 1-MeTIQ showed a tendency to prevent the MPTP-induced decrease in dopamine content, a statistically signi cant difference was not observed (Table 1). Only 1-Me-N-propargyl-TIQ prevented the MPTP-induced decrease in dopamine content in a dose-dependent manner that was statistically signi cant. The other N-loaded 1-MeTIQs did not inhibit the effect of MPTP. In particular, the dopamine content in the high-dose 1-Me-N-propargyl-TIQ treatment group was nearly equal to that of the control group (11.5 ± 1.0 ng/mg tissue, Table 1).

Effects of N-loaded 1-MeTIQs on nigral TH-positive cells
At 9 days after completion of MPTP, the number of TH-positive cells in the substantia nigra was counted following immunohistochemical staining. The TH-positive cell count in the control group was 108.8 ± 8.3 cells/slice (10-13 slices/brain, n = 5-8, Table 2, Fig. 4A). MPTP signi cantly diminished TH-positive cells to 62.6 ± 7.0 cells/slice (57.5% of control, 11-14 slices/brain, p < 0.01, Table 2 propyl-TIQ potently blocked the MPTP induction of TBARS levels (Fig. 6). TBARS values were comparable to controls when 1-MeTIQ or N-loaded TIQs were administered in the absence of MPTP (data not shown).

Discussion
Because the structure of TIQ derivatives is similar to that of MPTP, which causes parkinsonism-like symptoms, they have been studied extensively as candidates for PD-inducing agents. Most TIQ derivatives, including TIQ, 1-BnTIQ, and N-methyl-(R)-salsolinol, have been reported to be parkinsonisminducing compounds, while the TIQ derivative 1-MeTIQ prevents the development of bradykinesia induced by TIQ or MPTP (Nagatsu et al., 1988) and inhibits 1-methyl-4-phenylpyridinium ion (MPP + )-induced cell death (Parrado et al., 2000). Thus, 1-MeTIQ may be an endogenous PD-preventing substance (Okuda et al., 2006). In addition, selegiline, which has a structure similar to TIQ, is used for the treatment of PD. We . Thus, these results suggest that the propargyl moiety is essential for the neuroprotection mediated by rasagiline. However, since this neuroprotective effect is also observed in propargylamine-containing molecules that do not inhibit MAO-B, we speculate that it may not be due to the inhibition of MAO-B. For example, TVP1022 (S-optical isomer of rasagiline) was shown to be neuroprotective, with its activity attributable to the propargyl moiety (Bar-Am et al., 2005; Youdim, 2013). Our previous study also suggested that 1-Me-N-propargyl-TIQ does not disturb MAO-B in vitro (Kitabatake et al., 2009). Thus, MAO-B inhibition may not be necessary for neuroprotection. In the present study, the order of neuroprotection e cacy according to the functional groups is N-propargyl (signi cant effects against all parameters) > N-butynyl (signi cant against bradykinesia, DAT, and TBARS) > Npropenyl (signi cant against bradykinesia and DAT) > N-propyl (signi cant against bradykinesia and TBARS, but lethal at high doses). These results suggest that the number of bonds in the N-functional group is crucial and plays a pivotal role in neuroprotection, and that the presence of a triple bond is the most effective in our study system. Selegiline enters the active site cavity (core) in the center of MAO-B and inhibits monoamine oxidation. Xray crystallography showed that when present in the active site, the position of the propargyl group of selegiline was very close to avin adenine dinucleotide (FAD), which is a coenzyme of MAO-B (De Colibus  et al., 2005). Thus, the propargyl group may affect FAD function. FAD is a co-factor of oxidation-reduction reactions, is an energy carrier, and is required for oxidative phosphorylation during metabolism. We speculate that the interaction between a propargyl functional group and FAD but not MAO-B may participate in mediating the neuroprotective effects of 1-Me-N-propargyl-TIQ. Further investigations will be required to determine whether this speculation re ects the mechanism of N-functional groups. In addition, the length of the N-functional group is also an important factor because N-butynyl, which has a triple bond and a long length compared to propargyl, showed decreased neuroprotective properties.
A possible mechanism by which selegiline slows the progression of PD symptoms is that it protects neurons from the effects of oxidative stress and various neurotoxins (Lahtinen et al., 1997; Przuntek et al., 1999). However, its neuroprotective effects may be diminished by the formation of neurotoxic metabolites such as amphetamine and methamphetamine (Bar-Am et al., 2004,2007). Although the structures of the metabolites of 1-MeTIQ derivatives are unclear, the absence of the production of neurotoxic metabolites and/or the presence or production of neuroprotective metabolites may contribute to the protective function of 1-MeTIQ derivatives. In addition, TIQ derivatives are metabolized to Nmethylated-TIQ by N-methyltransferase in the brain, and further oxidized to N-methylisoquinolium ions, which may exert neurotoxic effects (Naoi et al., 1989a, 1989b, 1993, Niwa et al., 1990). On the other hand, a possible neuroprotective mechanism of TIQ derivatives was reported by Antkiewicz-Michaluk et al. In conclusion, our study suggests that the neuroprotective properties of 1-MeTIQ derivatives loaded with an N-functional group mainly depends on the number of bonds. The length of the functional group appears to also be partially involved. These results suggest that compounds intended to prevent MPTPinduced PD-like symptoms should include an N-propargyl group. Therefore, we should investigate how structural differences affect pharmacological properties in subsequent studies, which may reveal the reason that certain TIQ derivatives are neurotoxic and others are neuroprotective. Biochemical data regarding N-propargylated compounds is accumulating, including studies of its role in apoptosis and its neurotrophic mechanism. Some N-propargylated compounds are expected to be useful for PD treatment as well as in treating Alzheimer's disease (Youdim, 2013). In addition, although 1-MeTIQ did not prevent MPTP-induced decreases in dopamine content in our experimental system, it did prevent NMDA receptor antagonist (MK-801)-induced reduction in dopamine in the prefrontal cortex (Białoń et al., 2021), and 6hydroxydopamine-induced striatal dopamine reduction (Wąsik et al., 2018). Because the pharmacological potential of TIQ derivatives is high and their actions are broad, combining their structurally similar skeletons with N-propargyl groups and various substituents may lead to the development of new PD therapeutics.  Table 2 Brains were sectioned 9 days after completion of administration of saline or MPTP. The number of TH-positive cells in the substantia nigra are expressed as the mean ± standard error (SE, n = 5-8).