2906
COMMUNICATIONS TO THE EDITOR
Vol. 78
of I was also attempted by the coupling of S-benzylThe N-tosyl and S-benzyl groups were removed N - carbobenzoxy - L - cysteinyl -L- tyrosyl -L- phenyl- from I11 (300 mg.) by reduction with sodium in alanyl-~-glutaminyl-~-asparagine~ with S-benzyl-L- liquid ammonia and the resulting product was cysteinyl-L-prolyl-etosyl-L-lysylglycinamide (11) oxidized to the disulfide form by aeration in aqueous according t o the pyrophosphite p r o c e d ~ r e ,fol~ solution a t pH 6.4. The yield of pressor activity" lowed by removal of the protecting groups from ranged in several experiments from 50,000 to the resulting nonapeptide derivative and subse- 70,000 units (per 300 mg. of 111). After concenquent oxidation to the disulfide form. This ap- tration and lyophilization, the product was puriproach yielded a synthetic product assaying ap- fied by countercurrent distribution between secproximately 100 pressor units/mg. after purification butyl alcohol and 0.08 M p-toluenesulfonic acid by electrophoresis a t pH 5.6 in pyridine acetate and followed by electrophoresis in pyridine acetate countercurrent distribution between sec-butyl al- buffer (pH 5.6) on a cellulose-supporting mecohol and 0.08 M p-toluenesulfonic acid. d i ~ m . ' ~ ,The ' ~ purified material had a pressor In addition to these approaches, another method activity of 250-290 units/mg.'* when assayed was utilized which has led to material which is as as usual against the U.S.P. Standard P ~ w d e r . ' ~ biologically active as the most potent preparation The ratios between pressor, antidiuretic, milkof lysine-vasopressin obtained from natural sources. ejecting and avian depressor activities for the synThe key intermediate in this synthesis is S-benzyl- thetic material are the same as those for natural N - tosyl-L-cysteinyl-L-tyrosyl-L-phenylalanyl-L-glu-lysine-vasopressin.12t14r16 Starch column chromataminyl-L-asparaginyl-S-benzyl -L - cysteinyl - L - protography" of a hydrolysate of this material showed lyl-e-tosyl-L-lysylglycinamide (111), which has solu- the eight amino acids to be present in molar ratios bility properties favorable for its purification. to each other of approximately 1 : 1 and ammonia I11 was synthesized by the coupling of S-benzyl- in a ratio to any one amino acid of approximately N-tosyl-L-cysteinyl-L-tyrosyl-L-phenylalanyl-L-glu-3 :l. The natural hormone and the synthetic prodtaminyl-L-asparagine (IV) with the tetrapeptide uct showed the same infrared spectrum.'* They amide 11. For the synthesis of 11, ethyl L-prolyl- showed the same behavior on countercurrent disc-tosyl-L-lysylglycinate,6 m.p. 81-84', was coupled tribution between sec-butyl alcohol and 0.08 :tlp with S-benzyl-N-carbobenzoxy-L-cysteine6 by the toluenesulfonic acid and also had the same electroN,N'-dicyclohexylcarbodiimide method? and the phoretic mobility on Whatman No. 1 paper in resulting protected tetrapeptide ester was con- pyridine acetate buffer a t pH 5.6 and 4.0. verted t o the corresponding crystalline amide with (11) J. Dekanski, B r i f . J. Pharmacal., 7 , 567 (1952). ammonia in ethanol; m.p. 101-104°, [ c Y ] ~ ' D-29.3' (12) D. N. Ward and V. du Vigneaud, J . B i d . Chcm., in press. (13) H. G. Kunkel in "Methods of Biochemical Analysis," 1'01. 1, (c I, CHC13) (calcd. for C 3 8 H 4 8 0 s N & : C, 58.5; D. Glick, Ed., Interscience Publishers, Inc., New York, p . 141. H, 6.19; N, 10.8; S, 8.21. Found: C, 58.3; (14) A sample of natural lysine-vasopressin with approximately the H, 6.34; N, 10.6; S, 8.10); the carbobenzoxy same pressor activity as the synthetic material was kindly supplied group was then removed with H B ~ - H O A C . ~by Dr. Albert Light of this laboratory. This material was obtained IV was prepared by the coupling of L-phenyl- by purification by countercurrent distribution followed by chromatogalanyl-L-glutaminyl-L-asparagineswith S-benzyl- raphy on Amberlite IRC-50. (15) "The Pharmacopeia of the United States of America," fifteenth N-tosyl-~-cysteinyl-~-tyrosine~ by the isobutyl revision, black Printing Co., Easton, Pa., 1955, p. 776. These assay chlorocarbonate mixed anhydride procedurelo in a values are based on an activity of 0.4 U.S.P. Posterior Pituitary Unit/ mg for the Standard Powder. yield of 55% after recrystallization; m.p. 203-204', (16) For the ratios between these four activities in natural lysine[ a I z 1 D +4.4' (c 2.08, dimethylformamide) (calcd. see H. B. van Dyke, s. L. Engel and K. Adamsons, Jr., C, 57.0; H, 5.65; N, vasopressin, for C44HS1N7011S2-1/2H20: Proc. Soc. Expri. B i d . M c d . , 91, 484 (1956). The ratios of antidiuretic 10.6. Found: C, 57.1; H, 5.66; N, 10.6). and milk-ejecting activities to pressor activity for the synthetic maCompounds IV and I1 were coupled by the N,N'- terial were determined by Professor H. B. van Dyke and Mr. s. L. of the College of Physicians and Surgeons. dicyclohexylcarbodiimide method? to give a 39% Engel (17) S. Moore and W. H. Stein, J . B i d . Chem., 178, 53 (1949). yield of analytically pure 111, m.p. 226-230°, [ci]"D (18) The authors wish to thank Dr. Julian R. Rachele of this laboratory for determination of the infrared spectra. -23.0' (c 2.11, dimethylformamide) (calcd. for (274(19) This work was supported in part by grants from the Kational H 9 i 0 i 6 N i 3 S 4 . H 2 0 : C, 56.8; H, 5.99; N, 11.6; HzO, Heart Institute, Public Health Service, Grant H-1675, and Lederle 1.15. Found: C,56.7; H,6.02; N, 11.6; HaO, 1.02). Laboratories Division, American Cyanamid Co. (3) V. d u Vigneaud, D. T. Gish and P. G. Katsoyannis, ibid., 76, 4751 (1954). (4) G. W. Anderson, J. Blodinger and A. D. Welcher, ibid., 74, 5309 (1952). (5) This tripeptide ester was prepared from a-carbobenzoxy-r-tosylL-lysine, m.p. 85-88', and ethyl glycinate, removal of the carbobenzoxy group, and coupling of the product with carbobenzoxyL-proline by the o-phenylene chlorophosphite procedure [G. W. Anderson and R. W. Young, THISJOURNAL, 74, 5307 (1952) I followed by removal of the carbobenzoxy group with HBr-HOAc. (6) C. R. Harington and T. H. Mead, Biochem. J . , 8 0 , 1598 (1936). 77, 1067 (1955). (7) J. C. Sheehan and G. P. Hess, THIS JOURNAL, (8) E. A. Popenoe and V. du Vigneaud, i b i d . , 76, 6202 (1954). (9) This compound was prepared independently b y a method similar to t h a t recently published by J. Honzl and J. Rudinger [Collection Czachoslou. Chem. Communs., 20, 1190 (1955) I They also reported the synthesis of a product of high oxytocic activity from a nonapeptide derivative containing a tosylcysteine residue. (10) J. R . Vaughan, Jr., and J. A. Eichler. THISJOURNAL, 75, 5556 (1963).
M. FREDERICK BARTLETT ALBERTJOHL ROGER ROESKE ~ E P A R T M R N TOF BIOCHEMISTRY CORNELL UNIVERSITYMEDICALCOLLEGE R. J. STEDMAN F. H. C. STEWART h-EW Y O R K , Tu'. 'ET. DARRELL N. WARD \'INCENT
DU \'IGNEAUD''
RECEIVED MAY18, 1956
T H E PREPARATION O F ACID-STABILIZED SUBHALIDES FROM MOLTEN METAL-METAL HALIDE SOLUTIONS' Sir :
The formation of slightly stable subhalides by the solution of certain metals in their fused halides (1) Work was performed in the -4mes Laboratory of the Atomic Energy Commission.
June 20, 1956
COMMUNICATIONS TO THE EDITOR
2907
has been suggested several time^.^!^,^ Among the All the solid products described are diamagnetic, post-transition metal-metal halide systems, the and give powder pattern lines different from those variations in metal solubility both within the obtained with any binary mixture of the comgroup and with change in halide ion are consistent ponents. with such an interpretation, and, moreover, there INSTITUTE FOR ATOMIC RESEARCH is a direct correspondence between appreciable DEPARTMENT J O H X D. CORBETT OF CHEMISTRY STATE COLLEGE RICHARD K MCMULLAN solubility in the molten halide and the existence of IOWA AMES, Ion-A a known gaseous subhalide.6 However, with the RECEIVED M A Y3, 1956 possible exception of CaCl,6 the isolation of lower oxidation states from these melts apparently has not been achieved; the systems characteristically THE STRUCTURE OF T H E ANTIBIOTIC METHYMYCIN revert to the original components on solidification. The effect of certain foreign salts on the metal Sir : solubility has been reported for the cadmium7 and Methymycin' belongs chemically to a class of bismuth8 systems. An interpretation of the results antibiotics2 which contains such therapeutically that appears more satisfactory than those pre- important representatives as erythromycin3 and viously presented3,' can be obtained from a consid- carbomycin (magnamycin) and partial structures eration of the possible acid-base or ''complexing" have been advanced for two of them (erythromyinteractions between the added salt and the two cin3 and pikromycin5). We should now like t o oxidation states present. Added base, ;.e., halide report certain degradation experiments which, ion, would be expected to reduce the amount of coupled with earlier results,6 lead us to propose subhalide formed through stabilization of the more structure I as a complete expression for methyacidic, higher oxidation state. Conversely, ad- mycin. dition of an acid capable of complexing halide ion Mild hydrolysis of methymycin with aqueous would increase the amount of subhalide formed. sulfuric acid led t o the desosamine free fragment I1 Use of the strongly acidic A1C13 in such systems (m.D. 163-165'. la1 -1- 79' (all rotations in chlorohas resulted in the preparation of stable Bi(I), form), 225 mp, log E 4.03, " : : :A 2.80, Cd(1) and Ga(1) compounds. 2.95, 5.76, 5.91 and 6.08 p ; Found: C, 65.17; The solubility of Bi in BiC13 a t 260' corresponds H, 9.11; C-CHS, 19.34) which was further to 46y0 conversion to BiC1; a black, asphalt-like characterized as the acetate I11 (m.p. 198-200', mixture is obtained on solidification. At the same [(Y]D 93'; Found: C, 64.39; H, 8.29; acetyl, temperature, addition of AlCl, results in the reac- 12.11) and as the ketone IV (m.p. 173-179', 2Bi 3A1C13 = 3BiAlC14 taking tion BiC13 224 mp, log e 3.96; Found: ~ place quantitatively [56.3, 55.7% Bi, 55.3% theor.]. [ a ] +177', C, 65.87; H, 8.47; C-CH3, 19.95). The ketone The product, m.p. ca. 253', is maroon in bulk and reddish-brown as a powder. I t disproportionates IV gave a negative Schiff test and yielded 58% of t o metal and trihalide essentially quantitatively in carbon dioxide upon treatment with alkali followed water, dioxaneor alcohol, and darkens rapidly in air. by acidification (a parallel experiment with I1 furnished no carbon dioxide). Similarly, the solubility of Cd in CdC1, a t 740' When methymycin or I1 was treated with corresponds to 17.6y0 conversion of Cd2C12; a methanolic sulfuric acid, there was produced the black mixture is obtained on quenching. With spiroketal V (m.p. 79-81', [ Q ] D - 68', no high 2AlC&/CdC12, the solubility of cadmium a t 330' " : : :A 5.77 F ; indicates 71.5y0 conversion to the Cd(1) oxidation selective ultraviolet absorption, state. The solid, very light gray in bulk, white as Found: C, 66.51; H, 9.25; OCH3, 9.81; active a powder, also readily disproportionates in con- hydrogen, O.OO), which upon lithium aluminum tact with the above solvents. Similar results have hydride reduction led to the corresponding diol (m.p. 159-161', [ Q ] D 118'; Found: c, 65.77; been obtained in the iodide system. The compound GazCh, which has been found t o H, 10.17; OCH3, 9.40). Permanganate oxidation of I1 in acetone solution be Ga(I)[Ga(III)C14],5 is an example of such an acid-stabilized lower oxidation state. Further re- furnished three products: VI (m.p. 164-172', duction of the Ga(II1) therein by metal gives a [(Y]D +52' (acetone), A"',: 2.90, 5.66 and 5.78 p ; solution containing 7.4y0 GaCl a t I n the Found: C, 58.35; H, 7.79; C-CHI, 21.01; neut. presence of sufficient A1C13 the amount of metal dis- equiv., 318), VI1 (m.p. 55-58', [ a ] ~ 69', solved a t 180' is within 0.1% of that for the reac-" : : :A 5.68-5.76 p (broad); Found: C, 63.40; tion Ga(GaCl4) 2Ga 4%1cl3 = 4GaA1C14. H, 8.39; C-CH,, 22.51) and VIIIa (m.p. 126-128', The white product, m.p. 175', is, as expected, (1) M. N. Donin, J. Pagano, J. D. Dutcher and C. M . McKee. physically very similar to Ga(GaC14). "Antibiotics Annual 1953-1954," Medical Encyclopedia, Inc., New \
+
I
I
L
2
.
+
+
+
+
+
+
(2) G. von Hevesy and E. Lawenstein, 2 anorg. allgem Chem , 187, 266 (1930). (3) K. Grjotheim, F. Gronvold and J. Krogh-Moe, THTSJOURNAL, 77, 5824 (1955). (4) J. D. Corbett and S. von Winbush. ibid., 77, 3964 (1955). (5) S. von Winbush, R. K. McMullan and J. D. Corbett, to be
published. (6) (a) P.Ehrlich and L. Gentsch, Nafurwiss., 40, 460 (1953); (b) The compound has been found t o be CaHC1, P. Ehrlich, B. Ait, and L. Gentsch, Z . anovg. allgem. Chcm.. 883, 58 (1956). (7) D.Cubicciotti, THISJOURNAL, 74, 1198 (1952). ( 8 ) G. Cleary and D. Cubicciotti, ibid., 74, 557 (1952).
York, p. 179. ( 2 ) For leading references see R. Corbaz, L. Ettlinger, E . Ggumann, W. Keller-Schierlein, F. Kradolfer, E. Kyburz, L. Neipp,V. Prelog, A. Wettstein and H. ZBhner, Helw. Chim. A d a , 89, 304 (1956). (3) Cf.P.F. Wiley, K. Gerzon, E. H. Flynn, M. V. Sigal and U. C. 77, 3677 (1955). Quarck, THISJOURNAL, (4) R.L.Wagner, F. A. Hochstein, K. Murai, N. Messina and P. P. Regna, ibid., 76, 4684 (1953). (5) H. Brockmann and R. Oster, Nafurwiss., 4 8 , 155 (1955). (6) (a) C. Djerassi, A. Bowers and H. N. Khastgir, THISJOURNAL. 78, 1729 (1956); (b) C. Djerassi, A. Bowers, R. Hodges and B. Riniker, ibid., 78, 1733 (1956).
