α-METALLOCENYL CARBONIUM IONS1,2 - American Chemical Society

calculated to be 1430 c.p.m. and 1560 c.p.m. on the basis of ninhydrin and enzymatic assays (see below), respectively (original specific activity of K...
0 downloads 0 Views 238KB Size
COBNUSIC.~TIO;";S TO THE EDITOR

3484

chloride (82 p M . ) was isolated and crystallized as described before.@ MeZtzng point: 1S0-184° (authentic sample of L-glutamic acid HC1 184-188'; mixed m.p. 182-186'). Anal. Calcd. for CjH904N HC1: C, 32.71; H, 5.49; N. 7.63. Found: c, 32.91; H, 5.39; N,'7.6;. [ a ] " D = 31..7'. Specific activity was calculated to be 1430 c.p.m. and 1560 c.p.m. on the basis of ninhydrin and enzymatic assays (see below), respectively (original specific activity of K.4, 2460 c.p.m.). Further evidence for the identity was provided by paper chromatography10 after and before enzymatic decarboxylation by L-glutamic decarboxylase. l i Preliminary experiments with KA-2-CL4, carboxyl-labeled KA and K=1-9-C14 indicated that glutamic acid is probably derived from the carboxyl carbon and carbon 2, 3, 4 and 10. Further studies are now in progress in order to identifv other products and intermediate steps.

+

(9) H. Tabor and 0. Hayaishi, J . B i d . Chem., 194, 171 (1952). (10) Paper chromatographic analysis was carried o u t with three different solvent systems and examined with an automatic scanner and ninhydrin spray. T h e isolated material before and after enzymatic decarboxylation gave exactly t h e same Ry values as those of authentic samples of L-glutamic acid and y-aminobutyric acid, respectively Ry values for glutamic acid a n d y-aminobutyric acid were as follorvs: 0.28 and 0.63 (butanol-acetic arid-water, 4:1: l ) , 0.15 and 0.56 (water saturated phenol), 0 35 and 0 52 (ethanol-ammonia-water, 18: 1 ' l ) , respectively. DL-y-Hydroxyglutamic acid, kindly furnished from nr. T. Kaneko, gave R f values of 0.14, 0.07 and 0.29, respectively. (11) 0. Schales, V. l f i m s and S . S Schnles, . l i r h . . R i w h e ; i ! . , 10, 12) KA-2 C'4 a n d KA-9 C'4 werc synthesized f r o m r,L-tryptc,phan2-C" and ~ ~ - t r y p t o p h a n ~ 7 a - Crespectively, '4, as described above. T h e latter compound mas a generous gift of Dr. XI. Rothstein of University of California. Carboxyl-labeled KA was kindly furnished b y Dr. J. XI. Price of t h e University of Wisconsin.

DBPARTMEST OF MEDICALCHEMISTRY KYOTOUNIVERSITY FACULTY OF MEDICINE RYCJTO, JAPAS RECEIVED MAY6, 1

Vol. 81

produced the corresponding ethyl ether and one mole of acetic acid. The addition of 7.16 X 1 0 F acetate ion reduced the solvolysis rate of 111 by a factor of nearly five. This was not caused simply by a change in thc ionic strength of the M lithium inedium as the addition of 7.24 X perchlorate had a negligible effect on the solvolysis rate. The comnion ion rate depression is, therefore, further evidence for alkyl oxygen f i ~ s i o n . ~ More importantly the large magnitude of the effect in an 80% acetone-water mixture indicates the intermediacy of an extremely stable ion which is quite selective in its recombination with nucleophiles.6 Rel. rate 30' in 80% acetone/water

I

Trityl acetate 1 I1 Ferrocenvlcarbinyl acetate' 0.63 I11 hlethylferrocenylcarbinql acetated 6 7 I V ?riethylruthenocenrlcarbinylacetate" 9 0 V Methylosmocenylcarbinyl acetatee 34 IJI a-Acetoxg-1,l'-trimethyleneferrocene' 0 23 The synthesis of all compounds was accomplished by standard methods from known intermediates.c,d,eJ Satisfactory analyses were obtained for all acetates. * Rates were determined by aliquot titration of the acetic acid produced. T h e rates followed the first order law faithfully to at least %To completion. F.S. Ari~notoarid -4.C. Haven, j r . , THIS JOL-RXAL, 77, 6295 (1955). P. J . Graham, R. V. Lindsey, G. \V. Parshall, M. L. Peterson and G. M. TVhitman, i b i d . , 79, 3416 (1957). e Ref. 7 . f K. L. Rinehart, Jr., and R. J. Curby, Jr., ibid.,79, 3290 ( 19571,

OSAMUHAVAISHI The requirement for coplanarity of the cationic HIROSHITASIUCHI MINORUTASHIRO center and the cyclopentadienyl ring is demonHIROMIYANADA strated by the markedly slower solvolysis rate of SICERUKrwo the bridged acetate VI (slower by a factor of

a-METALLOCENYL CARBONIUM IONS','

Sir: The results of the solvolyses reported in Table I demonstrate that a-metallocenyl carbonium ions are remarkably stable. Indeed, the rates indicate t h a t these a-metallocenyl cations are of the same order of stability as the triphenylmethyl cation. Qualitative observations such as the solubility of ferrocenecarboxaldehyde in dilute hydrochloric acid3 and the facile conversion of phenylferrocenylcarbinol to the corresponding methyl ether by refluxing aqueous methanol4 previously have iniplied a high order of stability for such ions. T h a t these solvolyses indeed proceeded by alkyloxygen fission was conclusively shown by the ethanolysis of 111 and VI, Each of these acetates (1) Presented in p a r t a t t h e 135th Xfeetinp of t h e American Chemical Society, Boston, Mass., April 5-10, 1959, cf. Abstracts p. 8 6 - 0 . ( 2 ) T h i s work mas supported in part by t h e National Scienre Foundation (3) G. D. Broadhead, J. M. Osperby and P. L. Pausou. J . Clieirt. .%IC., Ci50 (19.58). (4) N. Weliky and E. S. Gould, THISJ o u R P ; , ~ ~ .79, ~ 2712 (1057).

132 than the corresponding secondary acetate 111). I t is to be emphasized, however, t h a t the reaction proceeded by a carbonium ion mechanism as was indicated by the occurrence of alkyl-oxygen fission. This leads us to propose the possibility t h a t this bridged, non-planar ion is stabilized by a direct participation of the iron electrons.

L

d

Another interesting feature of the relative rates recorded in Table I is the order of effectiveness of rnetallocene derivatives of Group VI11 metals in stabilizing adjacent carbonium ion centers; ferrocene < ruthenocene < osmocene. This is just the inverse of the order of reactivity of these sub( 3 ) F o r a general discussion, c f , C K Ingold, "Structure and Xechanism in Organic Chemistry," Cornell University Press, Ithaca, X. Y . , 1 9 3 , p. 360. (6) C. G. Swain, C. F3 Scott and K . H. Lohman, THIS101-RNAI., 76, 136 (1953)

July 5, 1959

COMMUNICATIONS TO

THE

EDITOR

3485

log E 4.35). Similarly, progesterone, 17a-hydroxyprogesterone caproate and '3"17,21-diacetate were converted to their respective 6a-chloro derivatives 111, IV and V. Conversion of I1 t o its 1-dehydro analog VI COXTRIBUTIOS KO, 2461 GATESAND CRELLINLABORATORIES JOHN H. RICHARDS was accomplished by selenium dioxide oxidation. Treatment of I with N-bromosuccinimide in CALIFORNIA INSTITUTE O F TECHNOLOGY PASADENA, CALIFORXIA E. ALEXANDERHILL^ aqueous buffered acetone yielded 6&bromo-l7aacetoxyprogesterone (m.p. 178-180°, [ a ] ~ (7) M. D. Rausch, E. 0. Fischer and H. Gruhert, Chem. and I n d . , O o , Xma, 246 mp, log E 4.14) inverted by acid to the 756 (1958). a-derivative VI1 ( A = 1). Bromination of 1(8) J. H. Richards and D. C. Garwood, unpublished results. dehydro-17a-acetoxypr~gesterone~ with N-bromo(9) National Science Foundation Predoctoral Fellow. succinimide in carbon tetrachloride gave the 6PRECEIVED MAY6, 1969 bromo derivative (m.p. 170-172" [ff]D +23", Xmax 250 mp, log E 4.24) inverted by acid to 1dehydro-6a-bromo-17cu-acetoxyprogesterone (VSTEROIDS. CXXVII.' 6-HALO PROGESTATIONAL 111) ( A = 6). AGENTS 6 - Dehydro - 6 - chloro - 17a - acetoxyprogesteSir: rone (IX) and 6-dehydro-6-chloro-"S" diacetate 17a-Acetoxyprogesterone 3-ethyl enol ether (I) (X) were obtained by chloranil oxidationsf of I1 (rn.p. 160-163", [ a ] D -146" (all rot. in CHC13), and V, while selenium dioxide oxidation of IX and Xgt,"," 241 mp, log E 4.32)2 on reaction with N- X yielded the 6-~hloro-A~~~~~-trienones XI and XII. chlorosuccinimide gave 6~-chloro-17a-acetoxypro- In multi-dose Clauberg assays3 it was found that gesterone (A = 0.5)3 (m.p. 221-222", [ a ]- l~o o , the progestational activity of 6a-chloro-17a-aceXmax 239 mp, log E 4.19) inverted by hydrogen toxyprogesterone was increased by AI- or by A6-

stances toward electrophilic substitution as judged by ease of acylation7 and by competitive acetylations.8 The reactivity order in this case is ferrocene > ruthenocene > osmocene.

TABLE I

Compound

17a-Ethynyl-19-nortestosterone 17a-Acetoxyprogesterone 6a-Methyl-17a-acetoxyprogesterone~~8 6-Dehydro-6-methy1-17a-acetoxyprogesteror1e~ 1-Dehydro-6a-fluoro-17a-acetoxyprogesterone 6-Dehydro-6-fluoro-17a-acetoxyprogesterone 6a-Chloro-17a-acetoxyprogesterone (11)

M.p. OC.

[u]DCHCla

EtOH max. (md

log e

Oral progert. 4 activity (Clauberg assay)

1 0.07

6a-Chloroprogesteronc ( I 11) 6a-Chloro-17a-hydroxyprogesterone caproate (IV) l-Dehydro-6a-chloro-17a-acetoxyprogesterone (VI) 6-Dehydro-6.chloro-17a-acetoxyprogesterone ( IX) 1,6-Bisdehydro-6-chloro-17a-acetoxyprogesterone (XI)

polyrnor. 183-184 2 15-2 16 130-132 Oil 203-205 212-214 168-170

6cu-Chloro-"S" diacetate (V) 6-Dehydro-6-chloro-"S" diacetate (X) 1,6-Bisdehydro-6-chloro-"S" diacetate (XII)

197-198 248-250 201-203

f34" +37O -36'

6a-Bromo-17a-acetoxyprogesterone (VII) 1-Dehydro-6a-hromo-17a-acetoxyprogesterone( V I I I )

163-166 172-175

f42" -7O

+400

+133' +zoo

-77"

t8" -83'

236

4.20

237 238 24% 285 229 258 297 237 285 228 260 295 236 244

4.19 4.15 4.21 4.36 4.00 4.00 4.03 4.19 4.31 4.00 4.03 4.07 4.12 4.15

2-3 12 6 15 2-3

8 50 35

0.5 1.5 1

1 6

This substance was first described by J. C. Bahcock, E. S. Gutsell, M. E. Herr, J. ,4.Hogg, J. C. Stucki, L. E. Barnes and W. E. Dulin, THISJOURNAL, 80, 2904 (1958).

chloride-acetic acid to 6a-chloro-17a-acetoxyprogesterone (11) ( A = 2-3); I1 ethyl enol ether (d = 1) (m.p. 180-181°, [ f f ] D -146" Xmax 251 p , (1) Paper C X X V I , A Bowers, E. Denot, R . Urquiza and L. M. Sanchez Hidalgo, Tetrahedron, in press (1959). (2) Correct elemental analyses were obtained for all new compounds. (3) Clauberg assays, by t h e Endocrine Laboratories, A = oral progestational activity, 17a-ethynyl-19-nortestosterone(Norlutin) = 1.

double bond introduction with the latter modification inducing the more pronounced effect. T h e oral progestational activity of IX was 50X that of Norlutin or 7 0 0 X that of 17a-acetoxyprogesterone and would thus appear to be the most potent (4) M.p. 229-230°, [u]D +19O, Xmax 243 m r , log c 4.20. (5) A. J. Agnello a n d G. Laubach, THIS JOURNAL, 79, 1257 (1957).

(6) H.J. Ringold, J. Perez Ruelas, E. Batres and C. Djerassi, i b i d . , 81, io press (1959).