This report continues the in-depth evaluation of methyl 4-[(p-chlorophenyl)amino]-6-methyl-2-oxocyclohex-3-en-1-oate , 1 (ADD 196022), and methyl 4-(benzylamino)-6-methyl-2-oxocyclohex-3-en-1-oate, 2, two potent anticonvulsant enaminones. These compounds were evaluated employing the amygdala kindling model. Neither 1 nor 2 was active against amygdala kindled seizures, further supporting the corneal kindled model as a definitive tool for antielectroshock seizure evaluation as previously reported. Additional intraperitoneal (ip) data on 1 revealed toxicity at 24 h at 100 mg/kg. Several active analogs have been prepared with the view to minimizing toxicity. In a special ip rat screen developed by the Antiepileptic Drug Development (ADD) Program, these newer analogs were evaluated for protection against maximal electroshock seizures (MES) at 10 mg/kg and neurotoxicity at 100 mg/kg. From this screen, several compounds were shown to be safer alternatives, the most notable was methyl 4-[(p-bromophenyl)amino]-6-methyl-2-oxocyclohex-3-en-1-oate, 13. Compound 13 had an ip ED50 of 4 mg/kg in the rat and a TD50 of 269 mg/kg, providing a protective index (TD50/ED50) of > 67. By variation in the ring size, additional aromatic substitutions and the synthesis of acyclic analogs, these newer compounds provide a more definitive insight into the structure-activity correlation. CLOGP evaluation and molecular modeling studies are also provided to further elaborate the molecular characteristics of potential anticonvulsant enaminones.

In a continuing evaluation of the aniline-substituted enaminones, the synthesis of additional para-substituted analogs has been made in an attempt to further quantify the electronic (sigma) and lipophilic (pi) requirements for anticonvulsant activity in this series. In addition, meta- and ortho-substituted and polysubstituted compounds have been synthesized and evaluated for anticonvulsant activity. In the para-substituted series, 4-cyano analogs (32 and 33) (+ sigma, - pi), which were highly active via intraperitoneal (ip) injection in mice, were inactive on oral (po) administration in rats. The para-substituted trifluoromethoxy (+ sigma, + pi) analog (8) had significant potency by both routes. Meta substitution limited the activity due to steric factors. Bromo and iodo substituents produced active para-substituted analogs (5 and 17) but were inactive when substituted in the meta position (37 and 41, respectively). Ortho substitution provided no clear relationship due to nonparametric deviations. Neither 1, the prototype enaminone, nor 2, the putative metabolite, produced significant nephrotoxicity or hepatotoxicity. Sodium channel binding of 1 and 8 indicated that 8 displayed relatively potent sodium channel binding but 1 showed weaker effects with IC50 values of 489 and 170 microM respectively against [3H]batrachotoxinin A 20 alpha-benzoate ([3H]BTX-B).

Previous results of anticonvulsant activity in several imidooxy carboxylates related to (aminooxy)acetic acid in young chicks, prompted an in-depth reinvestigation of these analogues in mice. A series of 22 succinimidooxy, phthalimidooxy, and naphthalimidooxy carboxylates were synthesized and evaluated for anticonvulsant activity by the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS). Methyl (succinimidooxy)acetate (2d), ethyl (succinimidooxy)acetate (2e), methyl (phthalimidooxy)acetate (3d), ethyl (phthalimidooxy)acetate (3e), and ethyl 2-(phthalimidooxy)propionate (3g), which were initially found to be active as anticonvulsants in young chicks were uniformly inactive in the Phase I seizure tests involving maximal electroshock (MES), pentylenetetrazol (scMet), or neurologic toxicity toxicity (Tox). Several newer analogues, ethyl (succinimidooxy)formate (2c) and methyl 3-(phthalimidooxy)-2-methylacrylate (4h) were found to be active in the scMet (3a) or both (4h) evaluations. Most interesting was the anticonvulsant results of N-(benzyloxy)-2-azaspiro[4,4] nonane-1,3-dione (5b), which displayed anti-MES activity and a protective index (TD50/ED50) of greater than 4.5.

Continuing structure-activity studies on the anticonvulsant activity of analogs of N-(benzyloxy)-2-azaspiro[4.4]nonane-1,3-dione (2a), which displayed anti-electroshock seizure (MES) activity and a protective index (TD50/ED50) of > 4.5 are reported. An in-depth analysis of this moiety was studied employing the Topliss structure activity and the Craig plot analytical approaches as well as a semiempirical method. CLOG P analysis was also applied to this series after experimentally determining the NOR fragment. All compounds were minimized and these physicochemical parameters correlated to anticonvulsant activity. Several interesting substituted benzyloxy compounds emerged from this study: the 2',4'-dichloro (2b), 4'-(trifluoromethyl) (2c), 2'-bromo (2d), 3'-chloro (2o), 2'-chloro (2r), 2'-fluoro (2p), and 3'-fluoro (2w) analogs, all of which had comparable, or better activity than the parent unsubstituted analog (2a). X-ray crystal analysis of the active 2a versus inactive N-benzyl-2-azaspiro[4.4]nonane-1,3-dione (10) is discussed.

No abstract