| Summary: | 博士 === 國立臺灣大學 === 藥學研究所 === 91 === 1. Synthesis and pharmacology of N-arylated pyrrolidin-2-ones and morpholin-3-ones as potassium channel openers
(-)-N-(4-Benzoylphenyl)-3,3,3-trifluoro-2-hydroxy-2-methylpropanamide (ZD6169, (-)-1) was recently reported to be a novel potassium channel opener (KCO), which significantly reduced micturition frequency in rats (ED50 = 0.16 mg/kg) with minimal effect on cardiovascular parameters (ED20 = 30 mg/kg). In our search for tissue-selective KCOs, N-(arylated)-1,3-oxazolidin-2,4-dione 2, morpholin-3-ones 3 - 5, pyrrolidin-2-ones 7 - 13, have been synthesized as conformational restricted analogs of ZD6169.
Target compounds 2-13 were synthesized using three different routes. N-(4-Benzoylphenyl)-5-methyl-5-trifluoromethyl-2,4-oxazolidinedione (2) was prepared from racemic ZD6169 via treatment with 1,1-carbonyldiimidazole under basic conditions; while N-arylated-2-methyl-2-trifluoromethyl- morpholin-3-ones (3-5) were obtained from the appropriate aniline tertiary carbinols by treatment with 1-bromo-2-chloroethane. N-(4-Benzoylphenyl)piperidin-2-one (6) and N-(4-benzoylphenyl)-3-hydroxy- pyrrolidin-2-one (11) were synthesized by condensation of 4-aminobenzophenone with the appropriate lactones in the presence of anhydrous aluminum chloride. However, when 4-aminobenzophenone was reacted with a-trifluromethyl- or a-methyl-g-butyrolactone in the presence of AlCl3, only small amounts of the corresponding amides could be isolated from a complex mixture of products. Thus, 4-aminobenzophenone was coupled with a-methyl-g-butyrolactone in the presence of Al(CH3)3, a milder Lewis acid, to give the g-hydroxyamide intermediate, which then underwent lactam formation under Mitsunobu reaction condition to provide N-(4-benzoylphenyl)-3-methylpyrrolidin-2-one (7). Compound 8, the trifluoromethyl analog of 7, was prepared in a similar fashion, albeit in lower yield. N-(4-Benzoylphenyl)-3,3-dimethylpyrrolidin-2-one (9) and N-(4-benzoylphenyl)-3-hydroxy -3-methylpyrrolidin-2-one (12) were prepared from 7 via base-catalyzed a-methylation and a-hydroxylation respectively. When the chiral Davis’ reagent was used in the above hydroxylation reaction, (-)-12 was obtained in 73% yield. However, when compound 8 was subjected to similar reaction conditions, instead of the desired hydroxylation and methylation products, only the defluorinated product 10 was obtained. Alternatively, ethyl trifluoropyruvate was subjected to allylation, ester hydrolysis, and coupling with 4-aminobenzophenone to give an amide intermediate, which was treated with ozone, followed by NaBH4 reduction, and lactam formation under Mitsunobu reaction conditions to provide N-(4-benzoylphenyl)-3-hydroxy-3-trifluoromethylpyrrolidin-2-one (13).
The KCO activity and selectivity of target compounds 2-13 were evaluated by in vitro tissue assays with preparations of male Wistar rat portal vein and rat bladder detrusor strips based on literature procedures. All compounds tested, except compound 8, demonstrated significant relaxant activity on both rat portal vein and detrusor strips; however, as compared to lemakalim and (±)-1, these rigid analogs are less potent. Among these newly synthetic compounds, 6 and 9 demonstrated potent and selective relaxant activity at the bladder detrusor muscle (IC50, bladder = 7.4 and 6.7mM respectively; IC50 ratio (portal vein/ bladder) = 41 and 51 respectively).
Based on the SAR study of the pharmacological data, it may also be concluded that (1) the KCO activity on the detrusor muscle is not sensitive to the nature of 3-substituents of pyrrolin-2-ones; (2) Compound (-)-12 is about 2-fold more potent than its recemate 12, indicating that, as with the case of ZD6169, the KCO activity of this type of compounds also resides in the (-)-isomers; (3) When the benzoylphenyl moieties in the X-ray diffraction structures of compound 9 and ZD6169 were superimposed, the corresponding carbonyl and the methyl groups are found to occupy different space areas. The conformational behavior of compound 9 may contribute to its high bladder-selectivity.
2.Synthesis and Pharmacology of galanthamine analogs as potential agents for the treatment of Alzheimer’s disease
Galanthamine, a tertiary alkaloid has been characterized as an acetyl cholinesterase (AChE) inhibitor and marketed as a therapeutic agent for the treatment of Alzheimer’s disease (AD). The clinical use of galanthamine may be restricted by its limited supply either from nature or by a relatively expensive total synthesis. In the search of agents improved with anti-AChE activity, there has been great interest in galanthamine derivatives. Therefore, we proposed to synthesize 2-methyl-7-methoxy-1,2,3,4-tetrahydrospiro[5H-2- benazepin-5,1'-cyclohex-1’-en-4’-ol] (31) as simplified analog of galanthamine. Based on computer modeling, 31 retained the rigidity of galanthamine.
The synthesis of compound 31 started from 1,4-cyclohexanedione monoethylene ketal, which underwent Reformatsky reaction with ethyl bromoacetate to yield ester 55. Compound 55 was reduced with LAH to give 8-(2-hydroxyethyl)-1,4-dioxa -spiro[4,5]decan-8-ol (56), which was reacted with phthalimide under Mitsunobu condition (Ph3P and DEAD). 2-[2-(8-Hdroxy-1,4-dioxa-spiro[4,5]dec -8-yl)-ethyl]isoindole-1,3-dione (57) and dehydrated 2-[2-(1, 4-dioxa-spiro [4,5]dec-7-en-8-yl)ethyl] isoindole-1, 3-dione (58) were obtained at the same time in 75 % total yield. The mixture was treated with p-TsOH to provid pure 58. Deprotection of the phthalimide analog 58 with hydrazine furnished 2-(1, 4-dioxa-spiro [4,5] dec-7-en-8-yl)ethylamine (51) in 65% yield. 2-Iodo-4-methoxy-benzaldehyde (101) prepared from p-anisaldehyde (53) was oxidized to its acid 102. Acid 102 was treated with SOCl2, and then reacted with amine 51 to furnish 2-iodo-4-methoxy-N-{2-[4’,4’-(ethyldioxy)cyclohex-1’- ene]ethyl}-benzamide (103). Methylation of 103 with NaH and CH3I gave 2-iodo-4-methoxy-N-methyl-{2-[4’,4’-(ethyldioxy)cyclohex-1’-ene]ethyl}-benzamide (104). 104 was subjected to Heck reaction condition to get the desired 7-exo product 2-Methyl-7-methoxy-3,4-dihydrospiro{5H-benzoazepine -1-one-5,1’-[4’,4’-(ethyldioxy)-cyclohex-2’-ene]} (105) in 67% yield. Thereafter, 105 was easily hydrolyzed to the corresponding a,b-unsaturated ketone 107. Reduction of 107 with L-selectride gave diastero-selectively the allylic acohol 109. 109 was then reduced with LAH to give the desired product 31.
Based on the synthetic methodology of 31, the total synthesis of lycoramine (138) was started from 2-iodoisovanillin (129) and 51. 2-Iodoisovanillin (129), prepared from isovanillin, was protected with benzyl bromide and then oxidized with KMnO4 to give the corresponding acid 132. Acid 132 was treated with thionyl chloride, and then reacted with amine 51 to furnish 3-benzyloxy-2-iodo-4-methoxy-{2-[4’,4’-(ethyldioxy)-cyclohex-1’-ene]- ethyl}-benzamide (133). Subjection of 133 to methylation with NaH and CH3I gave 134. The formation of the N-ring was achieved when 134 was subjected to Heck reaction conditions, and the tricyclic 135 was obtained in 28% yield. The low yield of the above intramolecular cyclization may be due to the stereo-hindrance of the structure. Acidic work up of 135 then afforded 6-benzyloxy-2-methyl-7-methoxy-3,4-dihydrospiro[5H-benzoazepine-1-one-5,1’-cyclohex- 2’-en-4’-one] (136). Deprotection of 136 with SnCl4 gave oxolycoraminone (137). Oxolycoraminone (137) was treated with LAH to give the final product (±)-lycoramine (138)
The anti-AChE activity (IC50) of compound 31 was 151 mM while that of galanthamie was 2.4 mM. Additionally, its activity against butyrylcholinesterase (BuChE) was 75 mM. This resulted in a 2-fold selectivity to BuChE, while galanthamine was claimed to have selectivity at AChE. the above results reflect the stringent structure requirements for binding to AChE since 31 showed much reduced activity at AChE although it retains most of rigidity of galanthamine.
|
|---|
