Jpn. J. Pharmacol. 88, 174 – 182 (2002) Novel Monoamine Oxidase Inhibitors, 3-(2-Aminoethoxy)-1,2- benzisoxazole Derivatives, and Their Differential Reversibility Kenji Yoshimi1,*, Masao Kozuka1, Jyunichi Sakai1, Tomoko Iizawa1, Yuki Shimizu1, Isao Kaneko1, Kouichi Kojima2 and Nobuyoshi Iwata3 1Neuroscience and Immunology Research Laboratories, Sankyo Co., Ltd., 2-58, Hiromachi 1-chome, Shinagawa-ku, Tokyo 140-8710, Japan 2Institute of Science and Technology, Inc., 10-2, Kitashinagawa 3-chome, Shinagawa-ku, Tokyo 140-0001, Japan 3Science Information Co., Ltd., 2-58, Hiromachi 1-chome, Shinagawa-ku, Tokyo 140-8710, Japan Received August 3, 2001 Accepted November 19, 2001 ABSTRACT—Although possible usefulness of non-selective monoamine oxidase (MAO) inhibitors for Parkinson’s disease therapy has been suggested in the literature, MAO inhibitors whose inhibition is reversible and have dual action to both MAO-A and -B subtypes is not available yet. Subtype selectivity and reversibility of a series of novel MAO inhibitors, 3-(2-aminoethoxy)-1,2-benzisoxazole derivatives, were studied. Several dual MAO inhibitors, which inhibit both MAO-A and -B, were obtained. When admin- istered to mice, their effects were generally reversible. Among the derivatives, RS-1636 and RS-1653 had much longer duration of brain MAO-B inhibition than that of MAO-A. In vitro, the inhibited MAO-A activity by these compounds was partially recovered by buffer change at 4°C, while little MAO-B activity was recovered. Although it is not fully elucidated yet, the reversibility of these inhibitors is probably determined primarily by this dissociation profile. This unique differential reversibility indicates that optimization of the balance of actions can be achieved by differentiating reversibility to each target molecule. Keywords: Monoamine oxidase, Tight-binding inhibitor, Reversibility, Parkinson’s disease Monoamine oxidase (MAO, EC 1.4.3.4) inhibitors were lazabemide (9) are some such selective inhibitors to MAO- discovered by their effect on mood disorders in the 1950s B that have a mild anti-Parkinsonian effect. However, (1, 2), and have been used as antidepressants since then. potential usefulness of reversible MAO inhibitors that Two isozymes, MAO-A and MAO-B have been identified have dual action to both MAO-A and MAO-B (10) have (1, 2). MAO-A is selectively inhibited by clorgyline and not been verified yet. preferentially deaminates serotonin (5-HT) and noradrena- In this paper, a new family of MAO inhibitors, 3-(2- line (NA). MAO-B is selectively inhibited by l-deprenyl aminoethoxy)-1,2-benzisoxazole derivatives, is presented. and preferentially deaminates 2-phenylethylamine (PEA) Several reversible MAO inhibitors to both MAO-A and -B and benzylamine. Recently, marked advances have been were obtained. Among the reversible dual inhibitors, made in the understanding of selective binding of substrates RS-1636 and RS-1653 had much longer duration of with MAO (3, 4). The old generation of MAO inhibitors, MAO-B inhibition than that of MAO-A. The profile of such as isocarboxazid, phenelzine and tranylcypromine, MAO inhibition in vitro was studied to verify the mecha- inhibit both A and B isozymes irreversibly. Several in- nism of this differential reversibility. hibitors selective to only MAO-A or MAO-B, with both reversible and irreversible characteristics, were developed MATERIALS AND METHODS later. Clorgyline, moclobemide (2), brofaromine (5), toloxatone and RS-8359 (6, 7) are some such MAO-A Animals selective inhibitors that are strong antidepressants. l- Male C57BL/6 mice (5 – 7 week-old, 18 – 22 g body Deprenyl (selegiline) (8), rasagiline, MDL-72974 and weight, from Charles River, Tokyo), which are widely used in rodent models of Parkinson’s disease, were used *Corresponding author. FAX: +81-3-5436-8566 throughout the pharmacological studies on the MAO in- E-mail: [email protected] hibitors. Male ddY mice weighing 30 – 35 g from SLC 174 Differential Reversibility of MAO Inhibitor 175 (Shizuoka) were used to obtain brain mitochondria, follow- and sensitive MAO assay was determined in our previous ing our previously reported method (6, 7). Animals were experiments (6). Then radiolabeled substrate, [14C]-5-HT housed in our temperature- and humidity-controlled animal (final concentration of 100 mM and radioactivity of quarters with free access to food and water for more than 15 mCi/mmol) or [14C]-PEA (final concentration of 20 mM 1 week before use. All experiments were carried out in and radioactivity of 25 mCi/mmol), was added to the tubes, accordance with the guidelines provided by the Institutional and incubation was continued for another 20 min. After the Animal Care and Use Committee of Sankyo Co., Ltd. reaction was stopped with HCl, the labeled metabolites (Tokyo). were extracted into ethyl acetate-toluene (1:1 V/V) by vigorous mixing using a vortex mixer and the organic layer Chemicals was counted for radioactivity with Topcount® (Packard [14C]5-Hydroxytryptamine creatinine sulphate (5-HT) Instrument Company, Inc., Meriden, CT, USA) using (55 mCi/mmol) and [14C]2-phenylethylamine hydro- Microscint-0® (Packard). In the experiments of in vitro chloride (PEA) (56 mCi/mmol) were obtained from reversibility (Figs. 3 and 4), pre-incubation and incubation Amersham Pharmacia Biotech (Buckinghamshire, UK). times were changed as described in the text. 5-HT, tranylcypromine hydrochloride and PEA were pur- In ex vivo MAO determination of the brain, C57BL/6 chased from Sigma Chemical Co. (St. Louis, MO, USA), mice were orally administered the test compounds. The Aldrich (Milwaukee, WI, USA) and Tokyo Kasei (Tokyo), brains were removed 1 h later and frozen on dry ice and respectively. l-Deprenyl hydrochloride and clorgyline kept at -80°C until further assay. The brain tissue was hydrochloride were purchased from Research Biochemi- homogenized with Polytron® in tenfold volume of cold cals, Inc. (Natick, MA, USA). RS-8359 ((±)-4-(cyanoani- 0.32 M sucrose and 10 mM sodium phosphate buffer lino)-7-hydroxycyclopenta[3,2-e]pyrimidine) and moclo- (pH 7.4). Determination of MAO activity was carried out bemide were synthesized by Ube Kosan Co., Ltd. (Yama- essentially by the methods described above. The brain guchi) and the Institute of Science and Technology, Inc. homogenate was further diluted ninefold with 50 mM (Tokyo), respectively. Reserpine solution (Apoplon®) was phosphate buffer, and the temperature of the diluted homo- purchased from Daiichi Pharmaceutical Co., Ltd. (Tokyo). genate and tubes stabilized for 10 min at 38°C in a water- Novel 3-(2-aminoethoxy)-1,2-benzisoxazole derivatives bath incubator. Then [14C]-5-HT or [14C]-PEA was added and lazabemide hydrochloride were synthesized in our and the homogenate was further incubated exactly 3 min or laboratories. Compounds were dissolved in saline for 2 min, respectively. The final concentration of the brain administration to animals, except RS-8359 and moclo- homogenate was 0.9% (w/v). bemide which were suspended in 0.5% carboxymethyl- To evaluate simple dissociation of inhibitors from the cellulose solution. enzyme, recovery of MAO activity of mitochondria was determined after repeated washing (Fig. 3). Mitochondria Radiochemical assay of MAO activity were pre-incubated with test compounds for 90 min at MAO activity was assayed as described previously (6, 7, 38°C, and then unbound inhibitors were removed by 11). Pooled whole brains of ddY mice were homogenized repeated buffer change five times by centrifugation at in tenfold volume of cold 0.32 M sucrose and 10 mM 15,000´g and re-suspension in 50 mM phosphate buffer at sodium phosphate buffer (pH 7.4) with a teflon-glass 4°C. Concentrations of test compounds were adjusted to homogenizer. The mitochondrial fraction was obtained by give about 75% inhibition, IC75, calculated according to the a differential centrifugation technique. Supernatant centri- dose-response curve for IC50 determination. Compounds fuged at 600´g and 4°C for 10 min was re-centrifuged at and concentrations used for assay were 1 nM clorgyline, 15,000´g and 4°C for 10 min, the pellet was suspended 2000 nM RS-8359, 1000 nM RS-1622, 30 nM RS-1636, and centrifuged again, and then diluted to yield a protein 20 nM RS-1650 and 70 nM RS-1653 for MAO-A and 8 nM concentration of 1 mg/ml and stored at -80°C. The protein l-deprenyl, 200 nM lazabemide, 200 nM RS-1622, 20 nM content was determined according to the method of Lowry RS-1636, 20 nM RS-1650 and 20 nM RS-1653 for (12), using bovine serum albumin as a standard. MAO-B. Control mitochondria were treated using the same MAO activity was determined by the traditional radio- procedure without inhibitors to define 100% MAO activity. chemical technique essentially as described by Wurtmann For the control of inhibition without washing, mitochondria and Axelrod (11). To determine the IC50 values of MAO were stood on ice during repeated centrifuge and re-suspen- inhibition, incubation tubes containing 0.1 mg-protein/ml sion. Percent inhibition was separately calculated for mitochondrial preparation in 50 mM sodium phosphate samples with and without washing. To evaluate incubation- buffer (pH 7.4) with various concentrations of test com- time dependency (Fig. 4), pre-incubation time was varied pounds were pre-incubated in a water bath at 38°C for from 0 to 180 min and the MAO activity was determined 20 min. Optimum substrate concentration for selective as described above. 176 K. Yoshimi et al. Inhibition of reserpine-induced ptosis obtained by changing the residues on the benzisoxazole To evaluate MAO-A inhibition in vivo, the antagonistic ring (Table 1). Several MAO inhibitors, which have dual effect on reserpine-induced ptosis was tested (13, 14). action to both MAO-A and MAO-B, were obtained as well Reserpine at 2 mg/kg was given to C57BL/6 mice sub- as selective inhibitors.