Experimental Section All rucliing poitits were detet,riiirietl iisiiig a IIel-Temp meltitig poitit appalatiis arid are iiricoriwted. The ir arid timr sperrra tlrterriiitted o n all rompoiuids ~ v e r e RS espected. Arialytical 1,esiiltu obtaiiied for C, Tf, arid S were withiii 10.4'7 of the theoretical values. Pharmacological Methods for Intrathecal Studies.-Polyet hy1i t i t : glycol '200 (PEG 200) \vv:w iisetf :LS :I vehirle I J ~ ~ of Z t hL I I S ~ aqiieoiis insoliibility of the carbanmte~, It was known that 0.2 nil/kg of PEG 200 irtjected into ihe cer~ebrospirialfliiid caiised only transient motor incoordiliat ion. Soliiticnis were prepared by dimilving tlie (wbamttes (29, 31, 48, 49, 50) in war111 (%') PEG 200 in conrentrations v:trying fi,orii 12.5 t o 50.0 tiig/ml. 1118ome experiments t h e dogs were Iitiaiiesthetixed; iii o t h t w , iidrniiiistratioii of the catbarnates was perfomied iitider intrav e t i i ~ i t s methohexitttl sodiiim (12.5 mg,,kg) miesthesia. The ( ~ : i t ~ l ~ t t i :~oliitiiiiis it~~ rvi'w riiiititi(d1). :itlrnitiistered i r i O.I-rti1 kp
During the course of our invc,itigntion of carbamo) I dcrivnt ives of tricyclic amines, we prepared 10.11tlihydrodibenz [b,f][ 1,4]oxaze~~i1i-l0-carboxylic acid hydrazide (1) arid found it to have potent airticoir\-uli:~i1t and analgetic properties. The scope of this nctivit: waq examined by preparing n series of similar hydrazide.: in which substitution on the nitrogeii fui1ctiotih :ind aryl groups was iiirwtigated as rwll as ~ ( ~ r n p o u n d s i n which the oxygen bridge wzi? replaced 1)y 8, SCEI,, CH2CH2,CH-CH, t n d a single bond. emicarbazides were iynt hesized from ires via the carbarno) 1 chloride- a i h e ~ i i in Hchemc I. l l o 5 t of t h c tricyclic arnineq cniploycd :itid the cartximoyl chloridei of thew :miries have I)ce1I described hjr us previously.' Trentinent of the cxrliamoyl chlorideh with h~~draziiie or substituted hydrazint.s gave thc expected products in good yield. AcylnIion of the -ernicarbazides obt:iinc~l above gave the terminal acyl derivatives :ii confirmed by spectral conip:irison with st :tnd:ird compomidi. The unsubstitutcd scmicnrbnzides condensed rcadily with nltlehydes but were inert to ketones w(~h:is acetune, ncetopheiroiic, and cyclohexanoiie. Two new tricyclic amines were prepared for structurc-activity studies. S-Trifluoromethyl-l0,ll-dihydrotlibenz [!I.$] [1,4]oxnzepinc~ vas olitnincd in n tn-o-
I
€I
cor1
KNP I
CONHNH,
1.
x=o
H
ztrp synthesis 4 o ~ v nin Scheme 11. Condens:itiori of 2-nitro-4-trifluoromethylchlorobenzene with potasium salicylaldehyde gave O-(2-1iitro-~-trifluorometl~~lphc1iyl)salicylaldeh\.de which was subsequeiitlj. l i j clrogenated with l'd-C to give an excellent yield of tlw clcqiretl nmiiicb.
Sovember 19ti8
AX~~ICOSVULSANT SEAIICAHBAZIDES
1159
The syiithesis of ~-chloro-3,B-dihydro~)heiiaiithridiiie chloric acid induced writhiug in the muube9 \vas measured for these compounds. Coinpourid l had an was accomplished as seen in Scheme 111. The isoEDa, of 26 mg/kg, equal to its ariticorivulsarit potency. SCHEME I11 However, the other derivatives were only weakly active in this test. To compare the activity of this series with the krlomr~ ariticorivulsants (18, 19),8 two ureas were synthesized Nco (20, 21). There lvas no observable anticonvulsant activity for theie derivatives indicating the hydrazides are the important compounds in this series. cyanate of 2-amino-4-chlorobipheIiyl~was treated with hlC133-4t o give 3-chlorophenanthridone. Reduction of this lactam with LiAlH4 gave the desired amine. A similar sequence starting with 2-aminodiphenyl sulfide gave an excellent yield of 10,ll-dihydrodibenz[b,f] [ 1,4I t h i a ~ e p i n e . ~ Biological Activity.-Anticonvulsant potencies (EDj0) were determined for two types of convulsant challenge : one electrical and one chemical. The methods employed were essentially the same as those reported by Swinyard, et d . , 5 for the maximal electroshock beizure test (AlES test, 50 m h ac, 0.2 sec, corneal electrodes) and the subcutaneous pentylenetetrazole test (sc l l e t test, ST, mg,lig). The end point employed in the N E S test mas abolition specifically of the hind leg tonicextensor component of the maximal seizure, while in the sc Met test, inhibition of all seizure components was required. Acute neurotoxicity (iYDjo) was measured by the method of Dunham and ;\Iiya,6 and was based on a subject's inability to remain for 1 miri on a slowly rotating rod. EDjovalues for anticonvulsant and I'Djo values for neurotoxicity measures were calculated according to the method of Litchfield arid W i l c ~ x o n . ~ The results of the anticonvulsant testing for the more interesting compounds is heen in Table I. None of the compounds was active against pentyleiietetrazoleinduced convulsions. Compounds 18 and 19 which are known anticonvulsants* were included for comparison. In general, the oxygen-bridged compounds were the most interesting with some activity residing in compounds where the oxygen bridge was replaced by a methylene or a single bond. Addition of a chlorine atom meta to the amino group in 1 increased the anticonvulsant potency but also the neurotoxicity (2). Only minor substitution of the hydrazine moiety retained activity, the most interesting substituents being methyl (7 arid 10) and acetyl (5 and 8). These derivatives retained their anticonvulsant activity but were less neurotoxic. Compounds with peak time less than 1 hr have too short a duration of activity to be promising. The best compound is probably 8 which ha5 peak time a t 2.5 hr with a 10: 1 neurotoxic :anticonvulsant ratio. The analgetic potency, as determined by the hydro( 2 ) K. Hoegerle and D. L'Ecuyer, Can. J . Chem.. 87, 2068 (1959). ( 3 ) .I. M. Butler, J . Am. Chem. Soc., 71, 2578 (1949). (4) J. Schmutz, F. Kunzle, F. Hunsiker, a n d A. Burki, Helu. Chim. Acta, 4 8 , 336 (1965). ( 5 ) E. A . Sminyard, I\'. C. Brown, a n d L. S. Goodman, J . Pharmacol. Ezptl. T h e r a p . , 106, 319 (1952). (6) N. W. Dunham a n d T. S. Miya, J . Am. Pharm. Assoc., 4 6 , 208 (195i). (7) J. T. Litchfield, Jr., and F. Wilcoxon, J . Pharmacol. Ezptl. Therap., 96, 99 (1949). ( 8 ) 11. .i. Davis. 5 . 0. TVintlirop, R . -1. Thomas, F. Herr, AI. Cliareut. and R . Gaudry. J . .\fed. Cliem., 7 , 88 (1964).
Experimental Section 8-Trifluoromethyl-l0,ll-dihydrodibenz[ b , f ][ 1,4]oxazepin-10carbonyl Chloride.-To a stirred solution of 13 g of phosgene in 50 ml of toluene a t j0 was added 150 ml of anhydrous ether followed by a dropwise addition of a solution of 18.9 g (0.0714 mole) of 8-trifluoromethyl-10,ll- dihydrodibenx [b,f] [ 1,4]oxazepine and 7.2 g (0.0714 mole) of triethylamine in 200 ml of ether. After addition, the suspension was stirred for 2 hr a t 5" and filtered, and the filt,rate was evaporated. The residue obtained was dissolved in 200 ml of hot petroleum ether ( b p 60-go"), filtered, and cooled to yield 19.25 g of light yellow crystals, mp 102-105'. Anal. (CijHgClF3N02) C , H, N. I n a similar manner, 10,ll-dihydrodibenz [b,f][ 1,4]thiazepine was converted to lO,ll-dihydrodibenz[b,f] [ 1,4]thiazepin-10carbonyl chloride, mp 111-114'. Anal. (C14H,,C1&O) C, H, N. Similarly, 3-chloro-5,6-dihydrophenanthridinewas converted t,o 3-chloro-5,6-dihydrophenanthridine-5-carbonylchloride, mp 137-139". A n d . (CiIHsC12SO) C , H, N. 3-Chloro-5,6-dihydrophenanthridone.-To a stirred solution of 39.0 g of 2-amino-4-chlorobiphenyl in 300 ml of toluene a t 0" was added dropwise a solution of 36 g of COClz in 75 ml of toluene. After the addition the solution was a cloudy red color. The solut,ion was warmed slowly to reflux while a slow stream of COC12 was passed through, followed by a reflux period of 15 min. The solvent was dihlled leaving an amber oil identified by ir as 4-chlorobiphenyl 2-isocyanate. The isocyanate in 100 ml of o-dichlorobenzene was added to a stirred suspension of 25.0 g of A1C13in 200 ml of o-ClzCsH4 a t 90-100". The resulting mixture Fas heated a t 120" for 1 hr, cooled, and decomposed by the addition of ice. Excess o-ClrC~H4was removed by steam distillatioii. The aqueous suspension was filtered and the residue was triturated with hot acetone and filtered to give 38.0 g of white crystals, mp 292". Anal. Calcd for CIsHsClNO: C, 67.98; H, 3.51; N, 6.10. Found: C, 67.50; H, 3.86; N, 5.56. 3-ChIoro-5,6-dihydrophenanthridine.-To a stirred suspension under N2 of 20 g of LiAlH4 in 500 ml of dioxane a t 95' was added a suspension of 38 g of 3-chlorophenanthridine in 750 ml of dioxane. ilfter refluxing for 18 hr the mixture was decomposed by careful successive addition of 20 ml of H20, 20 ml of 155; aqueous NaOH solution, and 60 ml of HzO. The suspension was filtered and the solvent was removed leaving a yellow solid. After two recrystallizations from EtOH, 13.1 g of yellowish white crystals were obtained, mp 120-122". 9nal. ( C I ~ H I O C ~C,N H, ) N, C1. 0-(2-Nitro-4-trifluoromethylphenyl)salicylaldehyde.-To 200 g of 4-chloro-3-nitrobenzotrifluoride a t 160" was added 160 g of potassium phthalaldehyde, with stirring, over a period of 0.3 hr. After addition an exothermic reaction occurred and the temperat,ure rose t,o 19.5" to give an amber liquid. T h e mixture was heated a t 150" for 1 hr, cooled, ice and H?O were added, and the suspension was extracted (EtrO). The ether extract (1..5 1.) was dried, charcoaled, and filtered. On evaporation an oil was obtained which crystallized from benzene-petroleum ether (bp 30-60'). Recrystallization from the same solvent mixture gave 122.1 g of yellowish white crystals, mp 79-81". 8na2. (C14HSFSNO4) C, H, N.
8-Trifluoromethyl-10,ll-dihydrodibenz[b,f][ 1,4]oxazepine.-A solution of 55.0 g of 0-(2-nitro-4-trifluoromethylpheny1)sal icylaldehyde was hydrogenated a t 0.703 kg/crnZ using 15 g of Kaney nickel catalyst. Filtration and evaporation of the EtOH gave an oil which crystallized on standing. Recrystallization from petroleum ether (bp 30-60') gave 40.65 g of yellowish white crystals, mp 86-88'. Anal. ( C U H I ~ F ~ SC, O )H, N, F. (9) E. T. Eckliardt. F. Clieplovita, RI. Lipo. and \V. 11. Govier. t'ruc. Suc. Ezptl. B i d . M e d . , 9 8 , 186 (1958).
1160
-x
I
cox N
I
K
0
11
II
fl 11
/K 'K?
Csllt,
