July 1965
515
3-AMINO-2,~-BENZISOTHIAZOLES
crystallize upon addition of a few drops of petroleum ether (b.p. 30-60"). The crystals were caollected, washed with a few drops of ethyl acetate, and recrystallized from the same solvent after clarification with charcoal; m.p. 124-130'; yield 50 mg. (30%); ultraviolet absorption spectra: in 0.1 S NaOH, Amax 219 mp (log E 4.42),275 mw (log E 4.11);in 0.1 S hydrochloric acid, A, 277 mw (log E 3.90). Anal. Calcd. for C13H,&LN,Os: C, 44.72; H, 4.04; C1, 20.31. Found: C, 44.38;H, 3.86; C1, 19.89. When chromatogrammed on Whatman 3 RIM paper and developed with 0.1 31 acetate buffer of p H 4.4, ascending flow, 3,5-dichloro-4-methylaminobenzoylglutamic arid showed Ri 0.83.
Discussion Although derived from methotrexate, DCAI differs from it in several aspects. I n comparison with methotrexate, D C N is superior as a therapeutic agent in mouse leukemia L1210 and less toxic toward the host animals.I8 Another distinct feature revealed by our work is that DCM is rapidly oxidized to the 7-hydroxy derivative both in vivo and in vitro while methotrexate apparently is 1 1 o t . ~ ~ ~ ~ The metabolic fate of pteroylglutamic acid and derivatives in man and other animals has not been extensively investigated previously. 2o Because of its occurrerice in human urine, xanthopterin has been suggestedZ1as a final catabolic product of pteroylglutamic acid in man. It was postulated that initially pteroyl(18) ( a ) A . Goldin, S. 11. Humpiireys, J. 11.Venditti, and K,Mantel, J . Xatl. Cancer Inst.. 22, 811 (1959); (b) J. M. Venditti, S. R. Humphreys, N . Mantel, I . Kline. and 1.Goldin, Cancer Res., 20, 698 (1960). (19) (a) E. S. Henderson and C . Denham, Proc. A m . Assoc. Cancer Res., 4 , 27 (1963); see, however: ( b ) D. G. Johns, .J. IV. Hollingsworth, -1.R. Cashmore, I. IT. Plenderleith, and J . R . Bertino, J . Clin. Invest.. 43, 621 (1964); ( e ) D . G . Johns, .'i.T. Iannotti, A. C. Sartorelli, B. A. Booth, and J. R. Bertino, L i f e Sci., 3, 1383 (1964). (20) B. Anderson, E. H. Belcher. I. Chanarin, and D . L. hlollin, Brit. J . Haematol.. 6 , 439 (1960). (21) J. A. Blair, Biockem. J . , 68, 385 (1958).
glutamic acid was cleaved between position 0 and Slot followed by oxidation of the pteridine part, finally i'csulting in 7-hydroxylation. Xanthine oxidase was implicated in the last step even though the enzyme is inert toward intact pteroylglutamic acid. 2 2 There was no evidence that pteroylglutamic acid was oxidized a t position 7 as in our studies with D C l I although pteroylglutamic acid probably indirectly gave rise to simple pteridines such as xanthopterin.': Incubation of pteroylglutamic acid with either chicken liver or rat liver slices (but not homogenate)23b afforded a diazotizable amine, most likely p-aminobenzoylglutamic acid. The fate of the pteridine moiety was not defined. In any event, 7hydroxylation of pteroylglutaniic acid was not encountered during incubation. Simple pteridines, on the other hand, are subjected to 7-hydroxylation when incubated with either liver preparations or xanthine oxidase. For example, 2amino-4hydroxypteridiiie is oxidized to isoxarit hopterin by chicken liver as well as by xanthine oxidase from milk24b;likewise, xanthopterin is oxidized to l e ~ c o p t e r i n . ? ~ b -In ~ all c a m , an oxygen atom is introduced a t position 7, parallel to DCM. \'et interestingly, DClI is not oxidized by milk or calf liver xanthine ~ x i d a s e . ~Although methotrexate and aminopterin were not oxidized by rat liver hon~ogenate,~ such an oxidation was recently reported by the incubation of these drugs with an erizymc system isolated from rabbit liver. I9c (22) J. -1.Blair, i b i d . , 66, 209 (1957). ( 2 3 ) (a) 8. Flitterman and 11.Silverman, J . Biol. Ckem., 224, 31 (1957); (b) .J. S. Dinning, J. T. Sime, and P. L. Day, Federatzon Proc.. 16, 549 (1956) (24) (a) T. Taira, .Vatwe, 189, 231 (1961); (b) F. Bergmann and H. Kwietny, Biochim. Biophys. Acta, 28, 613 (1958); ( e ) H. n'ieland and R . Leibig, Ann., 555, 146 (1944); (d) H. G. Petering and J. -4. Schmitt, J . A m . Chem. Soc., 72, 2995 (1950); (e) F. Korte and H. Bannuscher, Ann.. 62,2 126 (1959).
-
Synthesis and Pharmacological Action of 3-Amino-2,1-benzisothiazoles ROBERT F.l I E T E R , BETTYL. CUI\IJlISGS,p . 4 1 ~BASS,^'
H. 0. J. COLLIER'^
AND
Research Division, Parke, Davis and Company, A n n Arbor, Michigan, and Hounslow, England Received January 7 , 1965
A series of 21 novel 3-amino-2,l-benzisothiazoleswas prepared and evaluated for certain pharmacological effects. Some members show potent gastric antisecretory activity in the rat, antinociceptive action in the mouse, and antibradykinin activity in the guinea pig. The structure-activity relationships indicate that these three biological actions are not interrelated to a specific chemical moiety.
Continuous effort is exerted to develop novel heterocyclic systems that possess interesting pharmacological activities. This had led us to the synthesis and testing of a series of 3-amino-2,1-benzisothiazoles. Pharmacological observations indicated that certain independent members of the series possess gastric antisecretory, antibradykinin, antinociceptive, and mild antierythema activity. Chemistry.-Although 2,l-benzisothiazole (thioanthranil) itself has long been known2 and was prepared both by reductive cyclization2 from o-nitro-a-toluene(1) (a) Department of Pharmacology.
(b) Department of Pharmacology,
Hounslow, England. (2) S. Gabriel and T. Posner, Ber., 28, 1027 (1895); Stelzner, zbad., 99, 161 (18Q6).
S. Gabriel and
R.
thiol and recently by oxidative cyclization3 from oamino-a-toluenethiol, no 3-amino- or 3-substituted amino-2,1-benzisothiazoles are reported in the literature. Goerdeler and co-workers oxidized some paminothiocrotonamides to give 5-aminoi~othiazoles.~ We prepared 3-amino- and 3-monosubstituted amino-
I
NHR
NHR
(3) J. Goerdeler and J. Kandler. ibid., 92, 1679 (1959). W. Pohland, Ber., Sl, 2950 (1961);
(4) J. Goerdeler and H. (1963).
S6, 526
Starting materia! fur ?,Ibenaiaothiazole
1 2
,
-1 1
1i
-,
Jl.p.,
or.
121-122 107.-IOS' ti7 (ih' !J1 !I;'
\i!l -1011 !).4--!16 SS--S(I''
s 122-22:; 160-1 61' 129-130r 114- 115 159-1 60 1.50-1 51 15 99-100 12B--l27 1(i 17 182-184 102-10:i' 1s l!) 92--93 20 114-1 1.5 13-132 21 a B = benzene, l i = ethanol, T = tetrahydrofuran, \V = water, I' = "-proparioI, .I = ac~eto~iitrile, 1;t = ether, ('T = (,arbon tetraA. Iteissert arid F. Grube [Be?., 42, 3710 (1909)l report m.p. 121.5'. < This riielting point was takeii o n :L Fisher-Joliiis chloride. block and wa:, not corrected. In order to obtain the corrected ineltiiig point, ti" kins tu be sulitracted from the given uncorrected InrltLow-melting, crude thioamide was used without purifiratioii f o r tht. nest step. ' Ch. Grundmann :md H. Ulric4h [,I. Org. ing point. Chem., 24, 272 (10.59)1 report m.p. 150--151°. 0 10 11 13 1,'; 14
concd. HCI
.1 The N-mono- arid S,K-disubstituted thioamides obtained readily from the corresponding amides hy reaction with phoiphorus pentasulfide in pyridine follo~vedby hydrolysis of the phosphorus compound hy heating it i n water preferably ill presence of an organic solvent. The 2,l-hctizisothiazoles were yellow crystalline compounds which formed stable salts with mineral acids arid could be alkylated with reactive haloalkyls. In three cases the alkylation was carried out with niethyl iodide. llethylation in two of the three cases was proved to occur at the ring nitrogen by hydrolysis to a known derivative of arithrariilic acid. l-AIethyl-3methylamino-2,l-berizisothiazolium iodide on hydrolysis with refluxing concentrated HC1 gave K-methylarithranilic acid ; on the other hand, 1-methyl-3-amino2,l-brrizisothiazolium iodide on treatment with ac-jiieoiis ammonia gavc o-met hylaminobenzonitrile. were'
aNHCH3 COOH
XI1 2,l-benzisot Iiiazoles prepared show-cd vhararteristic absorptioii in the ultraviolet a t ca. 380 and ca. 230 mp. I:urthermore, the blue-green color reaction of all 2,l-beiizisothiazoles prepared in the presence of alcoholic ferric chloride proved very useful for identification purposes. This blue-green color faded rapidly on addition of a few drops of water which suggested that a metal complex (but no chelate) was formed which breaks up i n water. Absorption of the metal complex it1 the ultraviolet occurred at the same wave length as the 2,l-henzisothiazoles. Metal coniplexirig was observed with other metal salts too, such as coha1toils chloride. 13w r y t~alliza t ion from a hsolu t P e t ha-
~ - A ~ I I N 1-BENZISOTHIAZOLES o-~,
July 1965
517
TABLE I1 3-AhlINO-2,1-BENZISOTHIAZOLES
NO. 1 1ad 1bd
2 2 ad %bd 3 4 5 6 7 8 9 10 11 llhd 12 13 14 15 16
16ad 17
18 18 20 20a 21
R3, R4
R I , Rz H, H, H, H, H. H. 11, 11, 11,
H,
H
n
H €1 H H €1 11 H €1 H H H H
H, H, H. H, H. H H, H H , 5-Br H , 5-CI H, 6-C1 H , 6-C1 H , 6-C1 H , 6-CI H . 6-C1 H . 6-CH3 H. 6-OCH3 5-OCH.v 6-OCH3 5-OCH3, 6-OCH3 5-OCHa. 7-OCH3
H, H, H, r-I, H, H, H,
H H H ('113
CH3 CHI CIFIS IT, (CH2)zCHz 11, CH(CH3)z H, CH?CH(CHa)? H, CH(CHdC2H5 n, C H G H ~ H , C6H5 CH3, CH3 (CH44 (CH2)4 H, CH3 CHI. CH3 H, H H, CH3 CH3, CHI CHI, CHI (CHZCH?)~~ H, C2Hs H , C?Hb H, CvH6 H, CzHr H. CZH5
M.p., "C. 178-17Qb >ZOO dec. 112-113 206-20Th 210 dec.b >200 drc. 195- 19eb 107-168h 188-180 179-181 206-207h 18 1-1 82 >240 dec. 120-121h 144-14jh 2 10-215h 244-249 140-141b 191-192 2 50-2 5 1 134-13sh 254-260 125-1 26 225-22fih 181-182 187-188 2 40-2 4 2 181-1 85
Recrystn. solventa R
E T
+W
P A P E
P 13
P E
c E E t or W
P P E
P B
P P .A
P E R E
P E
Yield,
% 62 88 76 46
--66 13
52 54 45 69 .5 8 59 50 24 22 72 30 29 87 33 57 92 33 60 43 62 78 58
---% C 55.97 32.89 45.04 58.51 35.30 47.87 00.64 02.46 62.46 64.04 64.04 69.97 69.00 60.64 64.67 54.88 39.52 50.92 45.53 48.37 50.92 33.87 51.86 62.46 57.65 55.44 48.08 55.44
ca1cd.-H X 4.04 18.65 3.11 3 . 7 8 19 01 4.91 17.06 3.62 4.62
5.60 6.2!l 6.29 6.84 6.84 5.03 4.46 5.66 5.92 5.45 2.90 4.2i 2.73 3.55 4.27 3.41 4 35 6.28 5.82 5.92 5.50 5.92
I5 i 2
15.72 13.T1 11.64 13.17 15 17 14.10 13.17 11.00
11.75
~7~
found--H K 4 . 1 8 18.48 32.84 3 . 0 0 4 5 . 1 7 3 86 1 9 . 0 9 5 8 . 4 6 4 64 I T 22 35.41 3 . i 2 48.13 4.51 60.70 5 . 5 9 1 5 . 7 7 02.71 6.27 62 66 6 . 2 4 63.98 6.00 64.06 6 . 8 0 i0.24 5.06 68.99 4.20 6 0 . 6 8 5 74 1 5 . 5 8 64.65 5.83 13.66 55 03 5 . 2 2 1 1 . 7 3 39.74 2.87 50.83 4 . 0 4 1 3 . 0 5 45.79 2.49 15.21 48.60 3 . 6 1 1 3 . 9 2 50.82 4 . 4 7 13.38 33 99 3 . 3 4 51.60 4 . 1 2 11.12 62.50 6 . 2 1 57.69 5.72 55.48 5.97 48.08 5 . 5 0 5 5 . 6 5 5 . 8 9 11.80
C 56.28
Footnote a, Table I. This melting point was taken on a Fisher-Johns block and was not corrected. In order to obtain the corS: calcd., 21.35; found, 21.30. d a = rected melting point, 6' has to be subtracted from the given uncorrected melting point. C1-: calcd., 14.73; found, 14.30. methiodide, b = hydrochloride. e C1-: calcd., 17.67; found, 17.25. a
no1 gave a 2 :1 complex of the 3-dimethylaminobenzisothiazole with cobaltous chloride. Titration in 50% NeOH indicated that 3-amino-2,lbenzisothiazoles are basic (comparable in strength with nz- or p-chloroaniline) , 3-amino-2,1-benzisothiazole having pK.' = 3.8, 3-methylamino-2,1-benzisothiazole pK,' = 3.9, and 3-dimethylamino-2,1-benzisothiazole pK.' = 3.5.
Experimentalj 3-Alkylamino-2,l-benzisothiazoles. o-Amino-N-ethylthiobenzamide.-A mixture of 82 g. (0.5 mole) of o-amino-N-ethylbenzamidea and 111 g. (0.5 mole) of phosphorus pentasulfide in 200 ml. of pyridine was refluxed for 1.5 hr., cooled, and poured into 2 1. of ice-water. The yellow solid was removed by filtration and washed with water, then heated a t reflux with a mixture of 1 1. of benzene and 500 ml. of water until two clear phases were obtained (5-6 hr.). The organic layer was separated, washed with nTater, concentrated, and cooled. The thioamide was obtained as a yellow crystalline solid (see Table I). 3-Ethylamino-2,l-benzisothiazole (3).-A solution of 36 ml. (0.315 mole) of 30% HzOz was added dropwise with stirring to a (5) Melting points were determined on a Thomas-Hoover capillary melting point apparatus and are corrected unless stated otherwise. Some intermediates were prepared b y reference methods: 4-methyl-2-nitrobenaoic acid [A. Claus and J. Joachim, Ann., 266, 210 (189211 and 4-methoxy-2-nitrohenzoic acid were prepared by the method of L. Katz, L. S. Karger, W. Sohroeder, and M. S. Cohen, J . 0x7.Chem., 18, 1399 (19531, and 4,5-dimethoxy-2-nitrobenmoic acid [R. Pschorr and C. Sumuleanu, Ber., 32, 3412 (189911and 2-amino-3,5-dimethoxy-iV-ethylbenzamide according to P. Friedlander, "Fortschritte der Teerfarbenfabrikation,"Vol. 17, J. Springer, Berlin, 1932, p. 473. (6) R . H. Clark and E. D. Wagner. J. Org. Chem. 9, 55 (1914).
solution of 54 g. (0.3 mole) of o-amino-N-ethylthiobenzamide in 100 ml. of pyridine a t 3.5'. The reaction mixture was allowed to stand for 16 hr.; the yellow crystalline solid was removed by filtration, washed with water, and recrystallized from ethanol (see Table 11): ultraviolet in absolute methanol, Amax 378 230 mp (E,,, 478,1948). 3-Dialkylamino-2,1-benzisothiazoles. o-Nitro-N,N-dimethylthiobenzamide.-A mixture of 194 g. of o-nitro-N,N-dimethylbenzamide, 222 g. of phosphorus pentasulfide, and 800 ml. of pyridine was refluxed for 1.5 hr. The reaction mixture was poured into 3 1. of ice-water, and the yellow solid was filtered and washed with water. The solid was suspended in a mixture of 2 1. of benzene and 500 ml. of water and refluxed until a clear two-phase system was obtained (about 4 hr.). The organic layer was washed with water and concentrated to small volume. 3-Dimethylamino-2,l-benzisothiazole (lo).-A solution of 248 g. (1.1 moles) of stannous chloride dihydrate in 250 nil. of concentrated HC1 was added dropwise to a well-stirred suspension of 105 g. (0.5mole) of finely powdered o-nitro-N,N-dimethylthiobenzamide in 1 1. of concentrated HC1 at 25". The mixture was stirred an additional 3 hr. a t 25O, then cooled in ice. The solid tin complex was removed by filtration and washed with benzene. The solid was suspended in a mixture of 1 1. of benzene and 500 ml. of water and, below 20°, an excess of 507, aqueous NaOH was added. The benzene layer was separated, washed with water, and concentrated to small volume to give, on cooling, a yellow crystalline solid which was recrystallized readily from ether or water; ultraviolet in absolute methanol, A, 378, 232 mp (E,,, 499, 2200). The hydrochlorides were prepared by adding slightly more than 1 equiv. of concentrated HCl t o the alcoholic solution of the benzisothiazole. The methiodides were obtained by refluxing the benzisothiazole with an excess of CHJ in acetonitrile for 3 hr . Cobaltous Chloride Complex of 3-Dimethylamino-2,l-benz isothiazo1e.-A solution of 2 g. of 3-diniethplamino-2,1-benziso
3 - ~ I N O - ~-BENZISOTHIAZOLES 2,
.Jdy 19h5
Gastric antisecre-
No.6
la 2 2 :I 21) >
4 >
6
7 8
1.v.
in
tory.
guinea
in r a t , ED&?
Pi&!. LlED
58 >25 1.5
2c 16
2.4
32 8 1 8 8 >32 16 >16 >16
S.C.
1
intibradykinin.
1.8
3 .0 >23 3.6 23 19 >50 >25 4.8 >50 >25 2.8 11 54 3.1 50 6 15 4.2 7.4 11 >25
8
AntinociceptiT e,
__ S.C. in mouse-ED60
87 39 152 34 36 37 128 66 >200 >400 >200 >200 79 203 >200 >400 >200
9 10 8 16 11 12 >8 4 13 8 14 >lo0 16c 1.5 >200 8 16 >25 >4 16a 2 >200 17 4 250 18 63 19 >32 4 56 20 c1 i i 20b >200 21 >32 Aminopyrine 82 8 99 a All values are in mg./kg. * a = methiodide, b chloride. c Nonspecific response.
F Dan
70 78 >400 59 ,5 4 25.5 239
360 >200 >400 >200 >200 115 >400 >200 >400 >200
>loo
>200 >25 >200 >800 136 136 114 >200
=
160 hydro-
tions12 were determined between pairs of properties where the potencies had been tested. The coefficients (r,) were not significant a t a 0.5 level between the compounds' antisecretory-antibradykin,antisecretoryantinociceptive, or antibradykinin-antinociceptive potencies. This would indicate no relationship of pharmacological properties among compounds in this series based on their potencies in the individual tests. (12) R . G . D. Steel and J. H. Torrie, "Principlw and Procedures of Statistics " McGraa-Hill Book Co., Inc.. New York. N. y., 1960, p. 409.
519
Some of thcse compounds mere tested for their ability to delay the onset of erythema upon exposure to ultraviolet i r r a d i a t i ~ n . ' ~ Previous experience with established nonsteroidal antiinflammatory compounds had shown some correlation between antiultraviolet erythema and antibradykinin activity.1° A few menibers of the benzisothiazole series effectively delayed skin erythema, 16 being the most active and showing potency of the order of phenylbutazone. However, 16 was not the most active antibradykinin compound. Generally, within this series, the relationships between activity against ultraviolet erythemr. and against bradykinin were poor. Xevertheless, there is evidence of the existence of both properties in this series, but to test this relationship more closely would require more precise experimental values. Nonsteroid antiinflammatory drugs are active in the antibradykinin test," but, except for aminopyrine and antipyrine (phenazole), they are inactive in the antinociceptive test used.lO On the contrary, narcotic analgesics are inactive in the antibradykinin and active in the antinociceptive test. Some of the benzisothiazoles resemble aminopyrine and antipyrine in being active in both tests. Aminopyrine also resembles these benzisothiazoles in showing antisecretory activity. Highest potency of the gastric antisecretory property was obtained with lower 3-monoalkylamino-2, l-benzisothiazoles. The site of action of these compounds has not been elucidated. It apparently is not related to any anticholinergic or ganglionic-blocking properties since the series is devoid of such autonomic activity. Acknowledgments.-We
would like to thank Nisses
11.A. Patterson and R. A. Purdon for their assistance with the gastric antisecretory tests, Dr. D. A. AlcCarthy for the autonomic nervous system experiments, and 311. C. Schneider. Mrs. P. J. Piper, A h . R. J. Skerry, llisses S. Horwood-Barrett, G. C. Clarke, and 1'. Shorley for the antibradykinin and antinociceptive tests. Appreciation is also extended to Dr. E . F. Elslager for encouragement, to N r . C. Childs and associates for the microanalytical data, to Dr. J. 31. Vandenbelt and associates for spectral data, Miss C. Eady for tabulating pharmacological results, and A h . W. 11. Pearlman for catalytic hydrogenations. (13) C. 1'. Kinder, J Wax, V. Burr, 11. Been, and C. E. Rosiere, Aroh Intern. Pharmacodgn , 116, 261 (1958).
