Studies on 3-substituted 1, 2-benzisoxazole derivatives. 6. Syntheses

Feb 1, 1979 - A Useful Pd-Catalyzed Negishi Coupling Approach to Benzylic Sulfonamide Derivatives. Gang Zhou , Pauline Ting , Robert Aslanian and John...
1 downloads 0 Views 515KB Size
(‘no et al.

180 Journal of Medicinal Chemistry, 1979, Vol. 22, IVO.2

I. Hagedorn and W. Hohler, Arzrieim.-Forsch., 26, 1515 (1976). H. Oldiges and K. Schoene, Arch. Toxicol., 26, 193 (1970). K. Schoene, J. Steinhanses. and H. Oldiges, Hiochem. I’harrnacol., 25, 1955 (1976). D. Kuhnen-Clausen, E u r . J . Pharrnacol., 9, 85 (1970). D.Kuhnen-Clausen, Toxicol. A p p l . Pharrnacol., 23, 443 (1972). D. Kuhnen-Clausen, F E B S Lett., 39, 61 (1974). I.). Kuhnen-Clausen, in “Proceedings of the International Meeting on Cholinesterases and Cholinergic Receptors”, Split, Yugoslavia, April, 1975, Vol. 47, E. Reiner, Ed., Croatica Chemica Acta, Zagreb, Yugoslavia, 1975, p 465. R.Belleau. H. Tani, and F. Lie, J. Am. Chern. Snc., 87, 2283 (1965). R. M. Krupka, ref I O , p 299. E. J . Ariens and A. M. Simonis, i n “hlolecular Pharmacology”. E. J Ariens, Ed., Vol. I, Section II.A., ecademic Press, New York and London, 1964. d. F. Moran and D. J. Triggle, in “Cholinergic Ligand Interactions”, D. J. Triggle, J. F. Moran, and E. A. Barnard, Eds., Academic Press, New York and London, 1971. R . Rill, Diplomarbeit, University of Freiburg, i.Br., 1975. M. J. Astle and F. J. Donat. .I. Org. C h ~ m .25. . 507 (1960). H. Kuhnen, ref 10, p 371. We are gratefull to Dr. H. Kuhnen, Institut fur Aerobiologie, Schmallenberg-Grafschaft, who kindly supplied us with the enzymological data. The toxicological data were determined in the laboratory of Dr. H. Oldiges, Institut fur Aerobiologie, Schmallenberg-Grafschaft, to whom we express our sincere thanks. IJachesine was a kind gift of the Gerhardt-Penick l,td., Thornton Laboratories, Croydon, U.K. E. .J. Ariens and A. M. Simonis. A n n . N . Y . Acad. .%i. 144, A42 11967). E. cJ. Ariens and A. .J. Beld?Biochern. Pharniacol., 26, 913

excluded, with an excess of the respective pyridine component. The precipitate which formed when the mixture cooled was quickly sucked off and dried under vacuum (51 mmHg) at 60 “C. After several hours of drying, recrystallization from a small volume of ethanol yielded a colorless hygroscopic product. The reaction of 1 (Table I) with NaI in absolute ethanol yielded compound la. General Procedure for the Preparation of the N - ( @ Acetoxyethy1)pyridinium Salts 4-6. la, 2, or 3 (0.1 mol) was dissolved or suspended in acetonitrile. Acetyl chloride (0.15 mol), also dissolved in a small amount of acetonitrile, was dropped into this mixture, which was being refluxed. This mixture was stirred. and all moisture was excluded. After a maximum of 4 h of heating, the solvent and the excess of acetyl chloride were evaporated under reduced pressure. The residue was solid or oleaginous. It was ground with dry ether or acetone. After washing, it was crystallized from ethanol, 2-propanol. or acetone. if necessary. with the addition of ethyl ester. Pharmacology. Longitudinal muscle strips of the guinea pig ileum were isolated according to Paton and Rang.?“ They were derived from animals of 350--600 g and suspended in an organ bath at 37 “ C , containing 10 mL of Tyrode‘s solution of the following composition (mM): NaCl. 137: KCI, 3.7: Carl,, 1.8: MgC12, 1~05:NaH2P04,0.2; NaHC03, 11.9; glucose, 5.5: hexamethonium chloride, 0.01. The bath was gassed with 5 % C 0 2 in Olr the pH was 7.4 f 0.5. Isotonic contractions were recorded on a kymograph with B magnification of 1% The load of the lever was 150 mg.

Acknowledgment. T h e authors a r e indebted to t h e skilled assistance of Miss M. Krugel, Miss I. Muller, Mr. F. Bruggemann, Mr. H . Rorig, M r . E. Schmidt, a n d Mr. A. Schrichten. References and Notes (1) E. Dirks, A. Scherer, M. Schmidt, and G. Zimmer. Arzn e k - F o r s c h . , 20, 35 (1970). ( 2 ) E. Dirks, A. Scherer, h,l. Schmidt, and G. Zimnier. Arzneirn.-Forsch., 20, 197 (1970). (3) I. Hagedorn, W.-H. Giindel, J. Hoose. and Chr. ,renter. Arrneirn.-Forsch,, 26. 1273 (1976).

(1977). S. K. Gupta, J. F. Moran, and D. ,J. Triggle, Moi. Pharrnacoi.. 12, 1019 (1976). W. D.M.Paton and H. P. Rang, Proc. H . Sile. London. Ser. R 163, l(1964).

Studies on 3-Substituted 1,2-Benzisoxazole Derivatives. 6. Syntheses of 3-( Sulfamoylmethyl)-1,2-benzisoxazole Derivatives and Their Anticonvulsant Activities Hitoshi Uno,* Mikio Kurokawa, Yoshinobu Masuda, a n d H a r u k i Nishimura

Research Laboratories, Dainippon Pharmaceutical Cornpan\ L t d , 33-94 Enoki-cho, Suita, Osaka, J a p a n Received J u n e 30 1978 Several 3-(sulfamoylmethyl)-1,2-benzisoxazole derivatives were synthesized from 3-(bromomethyl~-l,2-benzisoxazole by the reaction with sodium bisulfite followed by chlorination and amination. Some of them displayed marked anticonvulsant activity in mice. The introduction of a halogen atom to the 5 position of the benzisoxazole ring caused increased activity and neurotoxicity; the substitution of a sulfamoyl group caused decreased activity. The activity of monoalkylated compounds might be the result of biotransformation. Among these compounds, 3-(sulfamoylmethvl)-1.2-benzisoxazole(la) was thought to be the most Dromising as an anticonvulsant from the ratio of NTDio and EDjo. I

During the course of’ routine testing, i t was noted t h a t 8-(sulfamoylmethyl)-1,2-benzisoxazole1 ( l a , S c h e m e I) exerted a p o t e n t anticonvulsant effect as measured by

only a weak effect on carbonic anhydrase i n vitro.’ Therefore, derivatives of la might b e of interest as novel anticonvulsants. Chemistry. I n the early work,’ la was prepared b y t h e chlorosulfonation and t h e successive amination of 1,2benzisoxazol-3-acetic acid (11). I n these reactions, 5sulfamoyl-3-(sulfamoylmethyl)-l,2-benzisoxazole was also obtained as a side product, a n d t h e yield of l a was very poor ( 7 % ) . Therefore, another m e t h o d was chosen t o prepare derivatives of la.

protection against maximal electroshock (MES) seizure. T h e present paper deals with t h e syntheses of several 3-(sulfamoylmethyl)-l,2-benzisoxazole derivatives and t h e results of their biological evaluation. It is well known that some arylsulfonamides show anticonvulsant activity which is supposed to b e due to t h e inhibition of carbonic anhydrase. However, l a showed 0022-2623, 79)1822-018~$01.00/0

G

1979 American Chemical Society

3-Suhstituted 1,2-Renrisoxazole Derioatiues

Journal of Medicinal Chemistry, 1979, Vol. 22, No. 2

181

Table I

compd

X

n

mp, "C

recrystn solvent

yield, %

formulaa

160-16 3 AcOEt 65 H 1 2" benzene 113-115 65 1 H NHCH, benzene-hexane 76-78 66 C 1 H NHC,H, benzene 114-117 46 1 H d NH-i-C,H, MeOH 72 105-107 H 1 e N(CH3)2 31 MeOH f 86-88 H NH-n-C,H, 1 97 ether 140-143 H NHOH 1 g 47 149-150 dec AcOEt H NHNH, h 1 benzene 29 113-114 dec i 1 H "N(CH3)2 24 AcOEt 138-14 1 H 1. NHCH,CH,OH j benzene 48 135-1 37 1 H k NHCH2C6H, benzene 46 139-141 H 1 1 NHC6H, benzene-ether 10 1 122-123 H m NH-C,H4-2-COOCH, benzene 113-115 22 H 1 n "(CH,)"H,), benzene 85 134-137 0 H 1 c-NC ,H,, 70 MeOH 114-1 1 9 1 H C-N(CH,CH,),O P benzene-hexane 27 119-121 H C-N(CH,CH, ),N-CH, 1 93 MeOH 140-142 r H 1 c-N(CH,CH,),N-C,H, benzene-hexane 71 S H 1 c-N(CH,CH,),N-CH,C,,H, 125-127 71 AcOEt 182-185 1 2a 5-F 2" benzene 67 141-144 b 1 NHCH, 5-F benzene 114-1 17 48 1 C NHC,H, 5-F benzene 42 127-130 d 1 NH-i-C,H, 5-F benzene 33 145-14 8 e 1 5-F N(CH3)2 benzene 34 151-153 f 1 c-N(CH,CH,),NCH, 5-F AcOEt 26 192-195 1 3a 5-cl 2" benzene 34 148-151 1 b NHCH, 5-cl benzene 37 1 150-15 2 NHC2H, C 5-cl benzene 20 114-116 1 d NH-i-C,H, 5-a 37 benzene 176-1 7 9 e 1 5-cl N(CH3)2 AcOEt 56 221-225 4a 5-Br 1 " : benzene 48 1 152-15 4 b 5-Br NHCH, benzene 73 144-147 NHC,H, C 1 5-Br benzene 52 95-97 d 1 NH-i-C,H, 5-Br benzene 32 183-185 e 1 5-Br N(CHA benzene-hexane 47 118-121 1 f 5-Br c - N ( ~ H ; ),NCH ~~2 EtOH 70 167-170 1 5 NH, 5-CH, 75 EtOH 1 226-229 6 NH, 5-NO, 25 AcOEt NH; 17 8-181 7 1 5-OCH3 52 AcOEt 1 187-190 8 6-F 2" 48 AcOEt 1 147-150 9 7-CH3 2" 16 AcOEt 159-162 H 16 2 ": 49 AcOEt 136-1 38 3 H 17 NH, .~ a All compounds were analyzed for C, H, N,S, and halogen; analytical results were within tO.4% of the theoretical values. la

b

3-(Bromomethyl)-1,2-benzisoxazole3(12), which was prepared from 11 by bromination a n d successive decarboxylation, was reacted with sodium bisulfite a n d gave 1,2-benzisoxazole-3-methanesulfonic acid (13) (see Scheme I). T h e chlorination of 13 with phosphorus oxychloride afforded sulfonyl chloride 14, which was converted t o several derivatives of la (1-9) by reactions with approp r i a t e amines. 3-(Sulfamoylethyl)- (16) a n d 3-(sulfamoylpropyl)-1,2-benzisoxazole(17) (see S c h e m e I) were also p r e p a r e d f r o m 3-(chloroethyl)-3 a n d 3-(chloropropyl)-1,2-benziso~azole,~ respectively. T h e reduction of 6 with stannous chloride gave 5amino-3-(sulfarnoylmethyl)-1,2-benzisoxazole (18), which was easily acetylated with acetic anhydride t o afford t h e 5-(acetylamino) derivative 19. T h e reaction of la with ethyl chloroformate a n d potassium carbonate gave N(ethoxycarbonyl)-3- (sulfamoylmethy1)-1,2-benzisoxazole (20). Sulfonylurea derivative 21 was obtained from t h e ammonium salt of 20 by heating. T h e N-acetyl derivative of la (22) was obtained by t h e acetylation of la with acetyl chloride.

Biology a n d Discussion. Protection against, MES was determined according t o Swinyard? and neurotoxicity was determined by t h e rotarod test5 using mice. Compounds l f , h - - p , r , s , 5-7, a n d 9 (see T a b l e I) were ineffective a t 100 mg/kg PO. Median effective doses (EDa) a n d median neurotoxic doses JNTDs0) were determined for t h e compounds which showed some activity at 100 mg/kg. EDso values a n d NTD5, values summarized in Table I1 indicated marked activity against MES following administration of la-e,g,q, 2 a - f , 3a-e, 4a-f, a n d 8 (see Table I). T h e introduction of a halogen atom t o t h e 5 or t h e 6 position of t h e benzisoxazole ring increased t h e potency. T h e order of increasing potency of those compounds was as follows: H < F < C1 < Br. T h e order of increasing neurotoxicity followed t h e order of increasing potency, and t h e increasing r a t e of neurotoxicity was larger t h a n t h a t of potency. As far as 5-halogeno derivatives were concerned, t h e order of increasing potency followed t h e order of increasing urnvalues6 (H, 0.00; F, 0.34; C1,0.37; Br, 0.39). However, t h e introduction of a sulfamoyl (6, = 0.46) or

Uno et al.

182 eJournal of Medicinal Chemistry, 1979, Vol. 22, N o . 2

-SO2NHCHB> -SO,NHEt = -S02(CH& > -S02-i-Pr.

Scheme I

This order is in general agreement with the relative extent of in vivo dealkylation of monoalkylated sulfonamide derivatives reported by S m i t h e t aL7 Therefore, t h e potency of these compounds might be t h e result of biotransformation. Smith e t ala7also reported t h a t t h e parent sulfonamide was formed hy in vivo reduction of N-hydroxybenzenesulfonamide. T h e potency of t h e N-hydroxy derivative of l a (lgj might be t h e result of in vivo reduction. Sulfamoylethyl derivative 16 a n d N-methyl derivat,ive IC exhibited approximately equivalent ED,, values, a n d t,he ED,, value of sulfamoylpropyl derivative 17 was comparahle to those of the N-ethyl derivative I C a n d .V-dimet,hyl derivative le. Among these compounds, la was thought to be the most promising compound as a n anticonvulsant from the ratici of NTD,, and EDSo. T h e further investigation of la is now in progress in our laboratories.

la

12, n =: 1-3

X - Br, C1

14

Experimental Section

Y

1-9, 16, 17

'Table 11. Ant,iconvulsant Activity and Neurotoxicity -

.

la

___-_-___-__I-

compd _____

anti-MES; ED,,,, mglkg, PO

neurotox; NTD,,, m d k g , PO

20 23 39 56 37 32 57 15 35 32 38 32

292 (252-330)

(16-24)' (14-37) (29-52) (44-69) (28-49) (19-54) (46-66) (10-20) (25-45) (23-41) (32-46) (27-38)

1 OOh 14 (lG-21) 20 (16-25) 2 1 (17-27)

223 (140-356) 340b

154 (134-175) 168 (141-202)

200b 109 (69-174) 350" 142 (64-317) 52 (38-70) 60 (44-83)

lOOb

57 (41-69) 250b 1 4 (10-18) 60 (42-77) 15 (10-22) 66 (55-79) 1 8 (14-23) 9 3 (59-146) 22 (16-32) 100-200 45 (33-58) 88 (72-108) 87 (60-127) 1 9 (8-46) 16 26 (17-40) 17 39 (23-67) diphenylhydantoin 10.0 (7.6-12.3) 111.0 (87.2-150) metharbital 60 8 (79.2-97.6) 87.9 (79.2-97.6) __ _________ 95% confidence limits. Graphically calculated.

a nitro group (a, = 0.71) to t h e 5 position contributed to the loss of potency a n d no acceptable QSAR analysis could be obtained among these compounds. T h e s t u d y of N-substituted compounds revealed t h a t t h e i~itroductionof simple monoalkyl substituents did not dbolish activity, hut when Ihe substituent was a n amino ( l h ) , a dimethylamino ( l i ) ,a benzyl ( l k ) or larger t h e activity was generally lost. With t h e exception of diillethylamino ( l e , 2e, 3e, a n d 4e) a n d N-(methylpiperah y l ) derivatives (lq and 4f), the disubstituted compounds were inactive in the anticonvulsant test. T h e order of t h e decreasing potency for alkylated derivatives in each series of c o m p o u n d s 1-4 was as follows: -S02NH2 >

Melting points were determined in a Yanagimoto apparatus and are uncorrected. The IR, NMR, and mass spectral data of all cornpounds were consistent with structure. Where elemental analyses are indicated only by symbols of the elements, aiialytical results obtained for these elements were withi!i -k0.4%* of the theoretical values. 3-(Sulfamoylalkyl)-l,2-benzisoxazoles la, 2a, 3a, 4a, a n d .i-9 (Table I). General Procedure. T o a solution of 12 (0.038 mol) in MeOH (130 mL) was added Na2S03(8.1 g, 0.064 mol) in H,O (1;lO i d . ) . The mixture was stirred a t 50 "C for 4 h aiid evaporated in vacuo. 'I'o the residue was added MeOH (250 mL). and the insoluble material in MeOH was removed. The soluticm was concentrated, and the residue was washed with ether arid dried. l ' o the dried residue was added I'OC13 (100 mL), t h r mixture was refluxed for 3 h. and then the excess of PO(& was removed in vacuo. Thus, crude 14 was obtained. Crude 1 4 was Ived in AcOEt 1200 mL). Under cooling, NH,,was saturated in the solution. The solid separated and was filtered off. and t h e solvent was removed. The residue was washed with AcOEt arid recrystallized from the appropriate solvent. N-Substituted 3-(Sulfamoylmethy1)-1,2-benzisoxazoles l h s, 2h f, Rb-e, and 4b ~f(Table I). General Procedure. To the AcOEt mlutioii of 14. above mentioned. was added the appropriate amine (0.111 mol, 3 molar equiv), and the mixture was stirred at room temperature for 30 rnin. The solid separated and was filtered off, arid the filtrate was evaporated in vaciio. 7'he resulting resitlne was recrystallized froni the appropriate solvent. 5-Amino-3-(sulfamoylmethyl)-1,2-benzisoxazole (18). To the suspension of G i l l g) in concentrated HCI was added a solution c f Sn( '1, (60 g in G O mI, of concentrated HCl), and the mixture was stirred at 25 30 "C for :4 h. The solution was made alkaline with NaHC(& and extracted with AcOEt. After being dried over Sa2S04:the solvent was removed in vnrilo. 'I'h? recrystallization of the residue frotn MeOH gave 6 g of' 18 (62%'), tnp 224 227 "C. Anal. ((:HHSNROiS) C, €1, N, S . d-(Acetylamino)-3-(sulfamoylmethyl)-l ,Z-henzisoxazole (19). 'l'(i the solution of 18 ( 2 . 5 g) in pyridine (20 mL) was added . k , O (3.4 gi. :\fter being kept overnight a t room temperature. ( h e mixture was poured intc 1-1,20 arid extracted with AcOEt. 'rhp extract was washed with H 2 0 and evaporated. ' h e residue was recrystallized from Me011 to give 0.7 g of 19 (24%). mp 261 266 '(', Anal. ((710Hl,N:,04SI C:, H, N , S. N-(Ethoxycarbonyl)-3-(sulfamoylmethyl)-l,2-henzisoxazole (20). T o a solution of l a (4.0 g in 80 i n l , o f acetone) were added ethyl chlorvformate (6.1 g) and K2CO:j(7.8 8 ) . 'T'he mixtiire was refluxed for 2.1 h and then filtered. 'The filtrate was evaporated, and the residual oil was dissolved in H 2 0 . The solution was acidified with HCI. The resulting precipitate was collected. washed ivith H,O. a n d recrystallized frotn AcOKt t o give 20 (4.0 g. 75%). rnp 1:49 141 '(', i\na\. l(',,Fl,7>J2()5s) (',

H.N . S. N-Carbamino-3-(sulfamoylmethyl)-1,2-benzisoxazole (21 ). A solution of 20 (2.0 g in 150 mT, of XfeOH) w a s saturated with

Azetidine Deriuatiues of Antidepressant Agents NHB.After being kept at room temperature overnight, the solution was concentrated in vacuo to give the ammonium salt of 20 (2.0 g) which was heated on an oil bath (150-160 "C) under reduced pressure for 10 min. To the cooled residue was added AcOEt. The insoluble solid was collected and recrystallized from MeOH to give 21 (0.9 g, 53%), m p 189-194 "C. Anal. (C9HBN3O4S) C, H, N, S. N-Acetyl-3-(sulfarnoylmethylb 1,2-benzisoxazole (22). To 50 mL of AcCl was added la (2.3 9). The mixture was refluxed for 48 h and then evaporated. The residue was washed with benzene, dried, and recrystallized from acetone to give 22 (2.5 g, 91%), mp 183-185 "C. Anal. (C10H10N204S) C, H, N, S. Biological Methods. All experiments were carried out in male mice of STD-dd strain weighing 20-22 g. Diet and water were given ad libitum to animals until the time of experiment. All compounds were administered by gavage as a suspension of 5% tragacanth solution. In preliminary screenings, all compounds were tested for anticonvulsant activity a t 100 mg/kg. For the determination of EDMand NTDw, groups of ten mice were used per dosage level, using a t least four dosage levels. Anticonvulsant Activity. Drugs were evaluated for their ability to prevent the hind-limb extensor component of maximal electroshock seizure induced using a 60-Hz, 25-mA current for 0.2 s, delivered through corneal electrodes 2 h after dosing. EDSo values were calculated by the method of Litchfield and Wilcoxon? Neurotoxicity. NTD,, WBS determined employing the rotarod. The end point for minimal neurotoxicity was muscle incoordination and was based upon the inability of the mouse to retain

Journal of Medicinal Chemistry, 1979, Vol. 22, No. 2 183 on a horizontal rod (2.5-cm in diameter) rotating at 11 rpm 2 h after dosing. NTDSO values were calculated by the method of Litchfield and Wilcoxon.8 Acknowledgment. T h e authors are grateful to Dr. M.

Shimizu, the director of these laboratories, and Dr. S. M i n a m i for their encouragement t h r o u g h o u t the course of this work. Thanks are also d u e to Dr. T. Karasawa for his helpful discussion and members of the analytical section of these laboratories for elemental analyses and spectral measurements.

References and Notes (1) H. Uno and M. Kurokawa, unpublished results. (2) Y. Masuda, T. Karasawa, Y. Shiraishi, M. Hori, K. Yoshida, and M. Shimizu, unpublished results. (3) H. Uno, M. Kurokawa, K. Natsuka, Y. Yamato (the late), and H. Nishimura, Chem. Pharm. Bull., 24, 632 (1972). (4) E. A. Swinyard, J. Am. Pharm. Assoc., Sci. Ed., 38, 201 (1963). (5) N. W. Dunham and T. S. Miya, J . Am. Assoc., Sci. Ed., 46, 208 (1957). (6) C. Hansch, A. Leo, S.H. Unger, K. H. Kim, D. Nikaitani, and E. J. Lien, J . Med. Chem., 16, 1207 (1973). ( 7 ) D. L. Smit,h,H. H. Keasling, and A. A. Forist, J. Med. Chem., 8, 520 (1965). (8) J. T. Litchfield and F. Wilcoxon, J . Pharmacol. Exp. Ther., 96, 99 (1949).

Azetidine Derivatives of Tricyclic Antidepressant Agents P i e r o Melloni,* Arturo Della T o r r e , Maurizio Meroni, Anna Ambrosini, and Alessandro C. Rossi

Carlo Erba Research Institute, 20159 Milan, Italy. Received July 17, 1978 Tricyclic derivatives of azetidine were synthesized and screened for their potential antidepressant activity. The active series had the tricyclic rings attached to position 1 and a basic group in position 3 of the azetidine. The most interesting compounds were comparable to the reference standards for reserpine antagonism in mice, the most active being the dextrorotatory methylamino derivative 84.' The pharmacological profile classifies it as a CNS stimulant devoid of peripheral anticholinergic activity.

One of the best represented classes i n the field of psychoactive drugs with antidepressant effects is without doubt that of the tricyclic antidepressant agents. There has been a g r e a t deal of research on these structures i n attempts to separate the documented therapeutic activity from the side effects, which are for the most part associated with their anticholinergic a n d cardiotoxic activities.' The novelty of the products synthesized is based on the association of the azetidine ring with the tricyclic structure. Some derivatives of t h i s series were f o u n d in pharmacological screening to have potential antidepressant activity, the most encouraging results coming from products with t w o aliphatic a m i n e groups, which is uncommon i n t h e field of d r u g s with antidepressant activity. Chemistry. The terf-butyl derivatives in T a b l e I were o b t a i n e d b y alkylation of suitable substrates with 3chloro-1-tert-butylazetidine,prepared b y t h e m e t h o d of Gaertnere6 T h e other compounds were prepared according to Scheme I. The alkylation was f o u n d to be difficult for n = 1 and A = -CH2CH2-. C o m p o u n d 11 was so obtained b y reductive dealkylation of the corresponding unsaturated derivative (A = -CH=CH-), which alkylates more easily. The yields of alkylation were 40--60%. Reductive dealkylation was, as expected, difficult for A = so that only t h e t e r t - b u t y l derivatives 2-4 and the unsubstituted azetidine 1 were prepared. To carry o u t the reductive

s,

Scheme Ia

Table I a

For n = 0, the alkylating halide was a b r ~ r n i d e for ;~ n =

1, a chloride.

amination? i t was necessary t o carefully control times and temperatures for the compounds with A = -CH20-, in order to avoid cleavage of the benzyl ether. T h e azetidines of T a b l e V were synthesized from the corresponding mesylates p r e p a r e d b y t h e m e t h o d of Anderson'. Tables 11-IV list the intermediates, with their physical properties and yields. The general scheme (Scheme 11) of synthesis is shown below. T h e methods

0022~2623/79/1822-0183$01,00/0 0 1979 American Chemical Society