COMMUNICATIONS TO THE EDITOR
1896
stituent. This approach has been used successfully for structure proposals of many indole alkaloids.*
Vol. 86
Nucleosides. VI. The Introduction of Unsaturation into the Carbohydrate of a Pyrimidine Nucleoside via a 2,d’-Anhydro Bond
Sir: 205-206
212-216
272475 201
I
244446
111
J z & + -f:l-
205-206
272-216 244-246
CHa I CHO
We wish to report the introduction of a 2’,3’-double bond into the carbohydrate moiety of a pyrimidine nucleoside via a novel, base-catalyzed elimination reaction in which a uracilyloxy group behaves as a leaving group. The unsaturated derivatives are viewed as potentially useful intermediates for the synthesis of a number of unusual pyrimidine nucleosides. 3‘-0-Mesyl-5’-O-trityI-2’-deoxyuridine2(I) was readily converted to 2,3’-anhydro-l-(2’-deoxy-5’-0trityl-P-D-1yxosyl)uracil (11) in high yield on treatment with 1 equiv. of sodium hydroxide in ethan01.~ The reaction of I1 with potassium t-butoxide in dimethyl sulfoxide at room temperature for 0.5 hr. afforded a solid (70% yield) with properties consistent with 1- ( 5 ’-0-trityl-2’,3’-dideoxy-2’-ene-P-~-g~ycero-pentofuranosyl)uracil5 (TV), m.p. 194-196”, CY]''^' 1560 0-
1v H-N &OCH,
0
I1
Second, the noncrystalline product (11) from the methylation of tubulosine with diazomethane gave a mass spectrum in which all the peaks originating from the quinolizidine moiety were also present, but the peaks a t m / e 201 and 187 were shifted t o m/e 215 and 201 (base peak). The observed peak movements (see 11) are in full accord with the proposed structure of tubulosine. Mechanistic proposals for the formation of the fragments indicated schematically in structures 1-111 may be found in the original literature4q5dealing with emetine. In the biogenesis of Ipecacuanha alkaloids i t is assumed that the isoquinoline nucleus is formed from a phenylethylamine precursor, which (formally speaking) condenses with another intermediate which can be represented by protoemetine (IV).9 By substituting a tryptamine precursor for the phenylethylamine component, the biogenetically plausible, but hitherto unencountered, structure of tubulosine (I) is created. Acknowledgment.-We thank Dr. W. Simon (E. T. H . , Zurich) for the p K ~ c sdeterminations, Prof. A. R. Battersby (University of Liverpool) for the generous gift of a sample of the hydrochloride of 111, Dr. M. J. Vernengo (University of Buenos Aires) for assistance in part of this work, and Dr. D. P. Hollis (Varian Associates for the n.m.r. spectrum. The work a t Stanford University was supported by grant No. AM-04257 from the National Institutes of Health of the U. S. Public Health Service. ( 8 ) See H. Budzikiewicz, C Djerassi, and D. H . Williams,“ S t r u c t u r e Elucidation of Natural Products b y Mass Spectrometry,” Vol. I , HoldenDay, Inc , San Francisco, C a l i f , 1964. (9, A. R . Battersby and B. J . T . Harper, J . Chem. S o c , 1748 (1958); C. SzLntay and I, Take, Tetvahedron L e t t e r s , 1323 (1963).
DEPARTAMENTO DE QUIMICA ORGANICA
P. BRAUCHLI
FACULTAD DE CIENCIAS EXACTAS Y XATURALES V. DEULOFEU
ARGENTISA BUENOSARES, DEPARTMENT OF CHEMISTRY H. BUDZIKIEWICZ STANFORD UNIVERSITY CARLDJERASSI STANFORD, CALIFORNIA RECEIVED MARCH6, 1964
IV, R=Tr V,R=H
I
\:B 111 (mp) 261, 242 ( t 9770, 6000). (c 0.4, ethanol), The same base-solvent system ( 2 equiv. of base) applied to either 1-(3’-0-mesyl-5’-0-trityl-2’-deoxy~-D-lyxosyl)uracil(III)Eor I gave a product identical in every respect with IV. The corresponding detritylated product V , m.p. 155-156’ dec., [ C Y ] ~ ~-88” D (c 0.3, water), (mp) 261, 231 ( t 10,380, 2300), was obtained in 81% yield on treatment with 1 equiv. of hydrogen chloride in chloroform (1) T h e concept of a n anhydro bond comprising a leaving group was first advanced by K . C. Murdock a n d R. B. Angiers [ J . A m . Chem. SOL.,84, 3748 (1962)l in connection with t h e formation of a halogenated l-cyclopentane derivative of thymine ( a thymidine isostere) from an anhydro intermediate. I t was proposed t h a t a positive charge (acid solution) on t h e anhydro oxygen a t o m enables a thyminyloxy group t o behave a s a leaving group. (2) J. P. Horwitz, J. Chua, M . Noel, and M . A DaRooge, J . Med. Chem., I , 385 (1964). (3) Analytical values f o r all compounds described in this work were Consistent with t h e indicated structures. (4) T h e procedure is essentially t h e same a s t h a t described previously [J. P. Horwitz, J . C h u a , J A. Urbanski. and M Noel, J . Org. Chem., 98, 942 (1963) ] for t h e synthesis of 5’-0-trityl-2,3’-anhydrothymidine. (5) T h i s nomenclature is in accord with t h a t employed by R.K Ness and H . G’ Fletcher: Jr:! ibid., 98, 435 (1963). ( 6 ) T h e mesylation of 1-(5’- 0-trityl-2 ’-deoxy-&D-lyxof uranosyl) uracil (see ref. 2) gives I11 in good yield.
May 5 , 1964
COMMUNICATIONS TO THE EDITOR
1897
The identity of V was established by independent Its photolytic behavior was found to differ considerably synthesis which utilized 1-(5'-0-benzoyl-2',3'-epoxy- from that of benzophenone2; irradiation with near0-u-1yxosyl)uracil (VI)' as starting material and a ultraviolet light in the absence of oxygen gave the method described recently for the introduction of a reduction products benzhydryl mercaptan, dibenz2,3-double bond into a hexose.8 Scission of the epoxide hydryl disulfide, and a tetrasulfide from thiobenzoring in VI with sodium iodide in acetone containing phenone, whereas dibenzopinacol is produced from 1 equiv. of acetic acid gave a single isolable iodohydrin benzophenone. (55% yield), m . p . 198-200' dec., [ C Y ] * ~ D+41° (c The addition of carbonyl compounds to olefins under the influence of light to give oxetanes is a well known 1.0, acetone) ",",A:: (mp) 261, 229 ( e 8710, 19,640), r e a ~ t i o n . ~ -In ~ view of the contrast presented by the ":A: (mp) 248, 214 ( e 7330, 16,730), to which the photochemical results for thiobenzophenone and benzostructure 1-(5'-0-benzoy1-3'-deoxy-3'-iodo-~-~-arab~nosy1)uracil (VII) has been tentatively a ~ s i g n e d . ~ phenone which have been noted above, i t was of interest to us to determine what course the irradiation of the The latter was converted to a 2'-O-mesylate (VIII, SOYo yield), m.p. 114-116', [ c Y ] ? O D f41' (c 0.8, ace- thioketone in the presence of olefins would take. In the present communication we report the novel phototone), A",: (mp) 260, 232 ( E 10,100, 14,300), "A:!: lytic reaction which thiobenzophenone undergoes with (mp) 249 ( e 8'760), and treated with sodium iodide in olefins, as described below for cis-2-butene and l-hexacetone. Reaction occurred a t 0 ' with elimination ene . of iodine and the methanesulfonyloxy group to give 1-(5'-0-benzoyl-2',3'-dideoxy2'-ene-P-~-glycero-pentohu (CgHj)zC=S RCH=CHR' --f furanosy1)uracil (IX) in 70yo yield, m.p. 139-140', RCH=C( CeHs)z R'CH=C( C B H S ) ~ [ c Y ] ~ ~-DI O i o (c 0.5, acetone), XO,:", (mp) 261, 230 (E 9330, 14,320), (mp) 246, 217 ( e 7300, cis-2-Butene (Matheson) and nitrogen were bubbled 12,005). Saponification of the benzoate ester in I X continuously through a solution of thioben~ophenone~ gave a solid which proved to be identical in every re(2.2 g.) in cyclohexane (170 ml.) which was irradiated spect with V. with a Hanovia No. 608A-36 lamp.8 After 4 hr. Studies are currently in progress to expand the scope the thioketone had completely reacted, as evidenced by of elimination reactions as applied to anhydro nucleothe disappearance of its blue color. The solvent and sides. The results will be described in detail in the excess cis-2-butene were removed under reduced presnear future. sure, and the residue was distilled a t 0.5 mm. giving a light blue-tinged liquid. Chromatography of the liquid on Woelm Activity Grade 1 acid alumina with petroleum ether (b.p. 30-60') as the eluent gave a white crystalline solid (0.71 g.), m.p. 47-49', Recrystallization of the solid from 95yo ethanol yielded VI, B = uracil VII, R = H IX, R = BZ pure 1,l-diphenyl-1-propene (0.5 g., 23% yield based on Bz= CeHsCO VIII, R = Ma V,R=H thiobenzophenone), m.p. 49-50'; a mixture melting point with an authentic sample prepared by the deAcknowledgment.-This work was supported in part hydration of l ,l-diphenyl- l -propanol was undepressed by research grants CA-02903 and CY-5943 from the (lit.9,10m.p. 51') 51-52'), When the photolysis was National Cancer Institute, Public Health Service, and carried out using trans-2-butene, the same reaction in part by an institutional grant from the United occurred, yielding an equal amount of 1,l-diphenyl-lFoundation of Greater Detroit allocated through the propene. Michigan Cancer Foundation. In order to ascertain what, if any, specificity thio17) J F Codington, R Fecher, and J J Fox, J Org Chem., 37, 163 benzophenone exhibits in its reactions with olefins, it (1962). was treated photochemically with the terminal olefin, ( 8 ) S . F. Taylor and G M . Riggs, J . Chem Soc.. ,5600 (1963). 1-hexene. A mixture of thiobenzophenone (2.0 g.) and ( H I This assignment is in accord with the normal course of opening of sugar- or nucleoside-2,3-epoxides which occurs predominantly a t C-3 [see 1-hexene (65 ml., Phillips 99 mole %) was irradiated for C I ) Anderson, L. Goodman, and B R . Baker, J A m Chem S o c , 81, 5 hr. while a slow stream of nitrogen was bubbled 898 11959) 1. However, i t is t o be noted t h a t either of the two possible iodohydrins should lead t o the introduction of a 2',3'-double bond in this through. The solvent and 1-hexene were removed alternate synthesis of V under reduced pressure, and the residue was distilled ROLLISH. STEVENSMEMORIAL JEROMEP. HORWITZ a t 0.1 mm. giving a clear liquid. Chromatography on
+
+
~YQ#.T
JONATHAN CHUA LABORATORY DETROITINSTITUTE O F CANCER RESEARCH IRWIN L. KLUNDT DETROIT, MICHIGAN48201 MARGARET .A. DAROOGE MICHAEL NOEL RECEIVED FEBRCARY 26, 1964
Photolysis of Thiobenzophenone in the Presence of Olefins: A Novel Reaction
Sir: A study of the light-catalyzed reactions of thiobenzophenone in alcoholic solvents was reported recently. I (1) G Oster, L Citarel. and M Goodman, J (19fi21
A m Chem Sor , 84, 703
(2) J K. Pitts, R 1. Letsinger, R P Taylor, J . M . Patterson, G Recktenwald, and R . B Martin, i b i d . , 81, 1068 (19591 ( 3 ) E Paterno and G Chieffii, CQZZchim i l d , 39, 341 (19091. (4; C Buchi, C G. I n m a n , and E S Lipinsky, J A m . Chem SOL.,76, 4327 119.541 ( 5 ) R Srinivasan, ibid.. 83, 775 (1960). (61 J F Harris, J r , and D. D Coffman, i b i d . , 84, 1553 (1962). (7) R H Abeles, R F H u t t o n , and F H . Westheimer, ibid., 79, 712 11857) (81 T h e reaction did not proceed when a Pyrex filter was placed between the lamp and the photolysis mixture However, the reaction was not a f fected hy the placement of a Vycor filter between the lamp and the photolysis mixture This suggests t h a t the reaction observed is d u e t o irradiation with light in the range 210-280 mp I t has been previously shown in ref. 1 t h a t thiohenzophenone i n a n inei-t solvent, cyclohexane, and in t h e absence of oxygen undergoes a net chemical change when ii-radiated with far-ultraviolet light (2.54 mpJ. Howevei-, the products of this reaction were not isolated. ( 9 ) K I.agrave, A n n chim (Paris), [IO] 8 , 386 (1927). (101 K l' Auwet-6, B i r , (IO, 693 (1929)