The synthesis of a structurally novel series of 6,6a,7,8,9,13b-hexahydro-5H-benzo[d]naphtho[2,1-b]azepines (2), conformationally restricted analogues of the dopamine D1 antagonist (5R)-8-chloro-2,3,4,5-tetrahydro-3-methyl-5-phenyl-1H-3-benzazepin -7-ol (SCH 23390, 1c), is described. Affinity for D1 receptors was determined by competition for rat striatal binding sites labeled by [3H]SCH 23390; affinity for D2 receptors was similarly determined by competition experiments using [3H]spiperone. Compounds in this series having the B/C-trans ring junction (2b and related analogues), where the D ring is unequivocally fixed in an equatorial orientation, possess considerably more D1 receptor affinity and selectivity vs the D2 receptor than the conformationally mobile cis stereoisomers (2a), thus leading to the conclusion that axial substituents at the 4- or 5-positions of the benzazepine nucleus are detrimental to D1 receptor affinity. Resolution and X-ray analysis demonstrated that D1 receptor affinity was preferentially associated with the (-)-6aS,13bR enantiomer of 2b.

Three new pathways to the antitumor drug temozolomide (4) have been explored via intermediates 3, 6, and 7. The key intermediate 5-amino-1-(N-methylcarbamoyl)imidazole-4-carboxamide (6) has been successfully converted to 4 in 45% yield by employing sodium nitrite in aqueous tartaric acid at 0-5 degrees C. Compound 6 is prepared from nitrophenyl carbamate 14a and methylamine or directly from 5-aminoimidazole-4-carboxamide (13) and either methyl isocyanate or N-methylcarbamoyl chloride. Temozolomide (4) is also prepared from 8-cyano-3-methylimidazo[5,1-d]-1,2,3,5-tetrazin-4(3H)-one (7) by hydrolysis to the hydrochloride salt of 4 in 10 M hydrochloric acid. Compound 7is prepared from either 5-diazoimidazole-4-carbonitrile (28) and methyl isocyanate or by diazotization of 5-amino-1-(N-methylcarbamoyl)imidazole-4-carbonitrile (25). Attempts to cyclize 5-(3-methyltriazen-1-yl)imidazole-4-carboxamide (3) with phosgene or phosgene equivalents were unsuccessful: only 2-azahypoxanthine (11) was isolated.

A novel enantioselective alkylation of double benzylic substrates with secondary electrophiles is reported. A simple norephedrine-based chiral ligand was synthesized that gives alkylation product in 95% yield and 95% ee. A unique water effect on the enantioselectivity was unveiled. Good to excellent ee values were obtained with a number of double benzylic substrates and secondary electrophiles. This novel reaction has been applied to the synthesis of a promising anticancer agent.

Several novel enantioselective syntheses of the dopamine D 1 antagonist (6aS,13bR)-11-chloro-6,6a,7,8,9,13b-hexahydro-7methyl-5H-benzo[d]naphth[2,1-b]azepin-12-ol (2) are described in which the key intermediate was 1-(2,2-dimethoxyethyl)-1methyl-1a,2,3,7b-tetrahydro-1H-naphth[1,2-b]aziridinium salt (20). The latter species was prepared either from 1-(2,2dimethoxyethyl)-1a,2,3,7b-tetrahydro-1H-naphth[1,2-b]azirine (18) by methylation or from the tertiary amino alcohols 1-[(2,2-dimethoxyethyl)methylamino]-1,2,3,4-tetrahydro-2naphthalenol (23) or 2-[(2,2-dimethoxyethyl)methylamino]-1,2,3,4-tetrahydro-1-naphthalenol (24) by tosylation and in situ ring closure. Regioselective trapping of 20 with Grignard reagent (4-chloro-3-methoxyphenyl)magnesium bromide (10) then afforded the trans amine 1-(4-chloro-3-methoxyphenyl)-N-(2,2-dimethoxyethyl)-1,2,3,4-tetrahydro-N-methyl-2-naphthalenamine (22), which was cyclized to give 11-chloro-6,6a,7,8,9,13b-hexahydro-7-methyl-12-methoxy-5H-benzo[d]naphth[2,1-b]azepine (9), a known precursor of 2. Several enantioselective syntheses, including a Jacobsen epoxidation route, a de noWo synthesis from L-homophenylalanine, and a classical salt resolution sequence, were developed for the preparation of the key intermediates in chiral form.

Two enantioselective syntheses of the fused benzazepine dopamine D 1 antagonist (6aS,13bR)-11-chloro-6,6a,7,8,9,13b-hexahydro-7-methyl-5H-benzo[d]naphth[2,1-b]azepin-12-ol (1) are described in which the starting material is (+)-L-homophenylalanine (6). In the first approach, methyl (2S)-(1,2,3,4-tetrahydro-1-oxo-2-naphthalenyl)carbamate (5) is prepared by intramolecular Friedel-Crafts cyclization of N-carbomethoxy (+)-L-homophenylalanine (9). Subsequent alkylation of 5 with (4-chloro-3-methoxyphenyl)magnesium bromide, deoxygenation with Et 3 SiH, reduction, alkylation, and epimerization yields (+)- trans-(1R,2S)-1-(4-chloro-3-methoxyphenyl)-N-(2,2-dimethoxyethyl)-1,2,3,4-tetrahydro-N-methyl-2-naphthalenamine (2), a key intermediate in the previouslydescribed route to 1 (Draper, R. W.; Hou, D.; Iyer, R.; Lee, G. M.; Liang, J. T.; Mas, J. L.; Tormos, W.; Vater, E. J.; Gu 1nter, F.; Mergelsberg, I.; Scherer, D. Org. Process Res. DeW. 1998, 2, XXXXX). A complementary route to 2 is also described in which arylation of an N-protected, carboxyl-activated (+)-L-homophenylalanine affords (2S)-1-(4-chloro-3-methoxyphenyl)-2-(methoxycarbamoyl)-4-phenyl-1-butanone (26). Reduction of the latter compound followed by an acid-catalyzed, diastereoselective cyclization affords (+)-(1R,2S)-[1-(4-chloro-3-methoxyphenyl)-1,2,3,4-tetrahydro-2-naphthalenyl]carbamate (16), which is reduced and alkylated as before to produce 2.