2590
COMMUNICATIONS TO THE EDITOR
1701.
80
OXOHAEMANTHIDINE : A BICYCLIC LACTAM POSSESSING A BRIDGEHEAD NITROGEN
Sir: Haemanthidine, C17H1&Oj, has been considered to be N-demethyltazettine (I, R = H ) since methylation of the base with either methyl iodide' or formaldehyde and formic acid3 affords tazettine (I, R = CH3).4 Data obtained recently in this Laboratory show this assignment to be incorrect. 0
Y k
R
r
I1
111
The revised structure (11, R = OH) for haernanthidine is supported by these data. HaemanthiY dine has been found to form an 0,O-diacetate ( A ~ 5.73 ~ ~ p" ) , not an 0,Xdiacetate as previously reported.' 6 iV Hydrochloric acid a t 90' converted haemanthidine to apohaenianthidine (111, R = OH), m.p. 193-10Go dec., [ Q ] ? ~ D 12305, (found for C16HI5iYO4:C, 67.34; H, 5.2'7; N, 4.95; OCHa, 0.00), 240 my (3700) and 294 mp (5050). Catalytic hydrogenation of I11 (R = OH) afforded a dihydro derivative, imp. 25S-260°, [ Q ] ' ~ D 20.7", (found for C16HliN O r : C, 66.63; H , 5.90; iY, &SA), hE2H 241 ETHYLENE MOLECULES FROM I 2-E No E - 8 (3700) and 293 m p (4580),which was treated with R I N G S C A N 0 0 OF T H E : 4 ?-C 6 8-C thionyl chloride and then reduced directly with FOUR LANOSTEROL S P E C I E S 3 C-C 2 c-c lithium aluminum hydride in tetrahydrofuran.6 ETHYLENE FROM REMAINING 16 c-c The product (111, R = H, no unsaturation 1, 2), 16 C-C PARTS OF FOUR LANOSTEROLS m.p. 159-160°, [ c ~ ] ~ ~ D10s" (EtOH), was identiFig. 1.-Since squalene is a symmetrical molecule, the cal in all respects (infrared, ultraviolet spectra ; species BA and AB are identical in the free state. How- rotation; m.p. and m.1n.p. determinations) with ever, the two are distinct in the form of a complex with the dihydroapohaemanthamine: obtained by the action of 6 hydrochloric acid on haemanthamine enzyme which may be assumed t o have asymmetry. (11, R = H)' and then catalytic reduction. Although the conversion of haemanthidine to Experimentally O.lSl% excess C1?Hz=Ci3H2 and 1.6527, C13Hz=CH2were found.12 The ace- haemanthamine has not been achieved by this tic acid from lanosterol thus contained doubly route, the positions of the functional groups of labeled molecules, proving methyl migration from haemanthidine must be as in I1 (R = OH) to acCy to (214, ;.e., two l,%methyl shifts in the course of count for its unusual conversion to tazettine (I, R = CH3). I n agreement with this structure, a cyclization. chloroform solution of haemanthidine was oxidized COXVERSE MEMORIAL LABORATORY HARVARD USIVERSITY R. K , MALIDGAL,by manganese dioxide to oxohaemanthidine, (11, DEPARTMENT O F CHEMISTRY T. T. TCHEN'? R = 0=), m.p. 104-19Go,[ Q I z 3 D -41.Ao, '$A':: 12 OXFORDSTREET KOXRAD I3Locr-i 3.16 (OH), 5.90 (C=O) arid (i.20 (Ar) p , h ~ ~ CAMBRIDGE, MASSACHCSETTS 232 (20,000),275 ((5300)and 326 x n p (510U), (found RECEIVED MARCH 27, 1958 for CliHliNOS: C, 64.94; H, 5.64; h-, 4.33). Oxohaemanthidine formed a non-crystalline 0(11) T h e second term in the numerator represents t h e amount of acetyl derivative 5.75, 5.88 p ) . Under singly labeled ethylene as indicated in Fig. 1. T h e first term arises from t h e f a c t t h a t t h e starting material contained only S370 C13, similar conditions dihydrohaemanthidine and apotherefore, t h e "doubly labeled" acetic acid or ethylene actually con. haemanthidine afforded oxodihydrohaeinnnthiclitie tains singly labeled molecules t o the extent of 2 X 0.65 X 0.35. I t is
+
+
~
true t h a t t h e "singly" labeled acetic acid also makes a contribution t o doubly labeled species because of t h e normal abundance of ' 2 ' 3 b u t this will be less t h a n of t h e amount of singly labeled acetic acid and can therefore be neglected. ( 1 2 ) T h e experimental values f o r ethylene of both mass 29 and 30 are higher t h a n those calculated. This m a y be attributed t o the unequal yield of acetic acid from different parts of t h e lanosterol molecule. This result does not affect our arguments which are based o n t h e qualitative presence of doubly labeled ethylene rather than o n t h e absolute amounts of this species. (13) Schular i i C.incer Research o f the American Cancer Society.
lx,
+
(1) H. G. Boit, Chzm. B e y . , 87, 1339 (1934). ( 2 ) H. G. Boit and W.Stender, i b i d . , 89, 1 6 1 (1956). 13) \\'. C. Wildman, C h e m i s f v y and J n d u s l v y , 123 (19SG). ( 4 ) T. Ikeda, W. I . Taylor, Y . T s u d a , S. Lgeo and 13. Ynjima, J . Chem. Soc., 47-19 (1936). is) Rotations obsd in chloroform solution unless otherwise noted. (6) I n one a t t e m p t t o isolate t h e intermediate chloroamine, only dihydroapohaemanthidine was recovered, presumably through the facile hydrolysis of t h e benzylic chloride. (7) H. M. Pales and LV. C XVildman, C'heinisii.) arid Indristvy, 561 (1958).
~
"
259 1
COMMUNCATIONS TO THE EDITOR
May 20, 1938
(11,R = 0=, no 1,2-unsaturation), m.p. 254-255', [c~]*~D +36", 2.75, 5.90 and 6.18 P, kz2H 234 (22,900), 274 (6500) and 325 mp (53001, (found for C1;HlSNO5: C, 64.42; H, 5.96), and oxoapohaemanthidine, (111, R = O=), m.p. 142-145", [ c Y ] ~ ~+230°, D ' ' : : :A 5.88 and 6.20 1.1, 236 (24,200), 276 (6000) and 322 mp (5000), (found for CI6H13NO4: C, 67.89; H, 4.53), respectively. These oxidation products represent three compounds of proven structure which, for theoretical reasons, should possess the properties of amino ketones rather than lactams.* Preliminary evidence supporting this view includes the positions of the infrared carbonyl maxima, the ultraviolet spectra and the facile reduction in methanol of each oxo compound to the respective starting alcohol by sodium borohydride.
pounds I, 11, and 111. The results obtained in aqueous solution a t 25' are plotted in Fig. 1 and show that I and I1 hydrolyze in the p H range 3-9
(8) C j . t h e reported synthesis of 2-quinuclidone, L. N . Yakhontov and hl. V Rubtsov, Zhur. Obshchei K h i m . , '27, 72 (1957); C.A., 61, 12085b (1957). (9) Visiting Scientist, National H e a r t Institute, on leave from the University of Osaka, Osaka, Japan.
LABORATORY OF CHEMISTRY s. U Y E O ~ OF NATURAL PRODUCTS H. M. FALES NATIOXAL HEARTISSTITUTE R. J. HICHET BETHESDA 14, MARYLAND W. C. WILDMAN RECEIVED APRIL2 , 1958 INTRAMOLECULAR BIFUNCTIONAL CATALYSIS OF ESTER HYDROLYSIS
Sir: It has been demonstrated previously that the solvolysis of phenyl esters is powerfully catalyzed by suitably spaced neighboring ionized carboxyl group^.^,^,^ Such reactions should be further accelerated, if the ester carries a second substituent which will stabilize the transition state, e.g., by hydrogen bonding or by ion pair formation. The principle is similar to the "bifunctional catalysis" of Swain and Brown4 except that all three functional groups form part of a single molecule. We have tested this assumption by comparing the PH dependence of the solvolysis rate of com-
1
w
I 4
2
0
I
I
I
I
6
q
10
I2
PH,
data for com-
Fig. 1.-Hydrolysis rate constants a t 25'; pound I from reference 3.
a t rates proportional to their carboxyl ionization, while I11 hydrolyzes a t a rate corresponding to the concentration of the singly ionized species, assuming pK = 3.62 and 4.5 for the salicylic and succinic carboxyls, respectively. Two pathways are pos0
0
0
/I
0
0
0
I,'1
/I
A/CoH I
N
\
H.0
O
11 0
0
\/
0
C-CHz
B
It
0
It
0
I11 (1) J. D. Chanley, E. M. Gindler and H. Sobotka, THISJOURNAL, 74, 4347 (1952). (2) (a) H. Morawetz and P. E. Zimmering, J. P h y s . Chem., 68, 753 (1954). (b) P. E. Zimmering, E. W. Westhead, Jr., and H. Marawetz, Biochim. Biophys. Acla, 25, 376 (1957). (3) E. R. Garrett, THISJ O U R N A L , 79, 3401 (1957). (4) C. G. Swain and J. F. Brown, Jr., i b i d . , 74, 2538 (1952).
t0-CEO 9
2
sible and no distinction may be made between them from kinetic data. If the reaction proceeds exclusively through A, this intermediate would decompose 24,000 times as fast as ionized I which has the carboxylate in the same position but lacks the second, un-ionized, carboxyl. Conversely, if the reaction proceeds through B, this intermediate is 66 times as reactive as ionized 11. Intramolecular catalysis by neighboring functional groups, such as observed with compound
