April 20, 1959
1913
S T R U C T U R E .4ND C O N F I G U R A T I O N O F G E R h l I X E
[CONTRIBU?I O N
F R O M THE
DEPARTMENT OF
XXVIII.
Veratrum Alkaloids.
PHARMACEUTICAL
CHEMISTRY
O F THE r N I T ’ E R S I T Y O F
WISCOSSIS]
The Structure and Configxation of Ge~rnine*-~
BY S. MORRISKUPCHAN A N D C. R. N;\R.~YAN.W RECEIVEDOCTOBER 2, 1958 The alkaloid germine has been shown to have structure and configuration I . Xlkaline isomerization of I leads to isogermine ( X I ) and thence to pseudogermine ( S I I ) . Acetic anhydride-pyridine acetylation of I affords a tetraacetate ( X I I I ) ; acetylation under more vigorous conditions gives a pentaacetate (XIV). I forms an acetonide ( S V ) which is oxidized by periodic acid to an aldehydo-y-lactone (XVI) also obtained from pseudogermine acetonide (XVII ) ; XVI yields a diacetate ( X X ) . The oxime X I X derived from XVI also yields a diacetate (XXII). Compound X I yields an acetonide ( X V I I I ) which is acetylated to a diacetate ( X X I ) . Germine acetonide ( X V ) affords a diacetate ( X X I I I ) which is hydrolyzed by mineral acid to germine diacetate (XXIV). Periodic acid oxidation of X X I V yields a cyclopentenvne aldehyde (XXV). Methanolysis of X X I I I yields germine acetonide monoacetate ( X X V I ) which is hydrolyzed t o germine monoacetate ( X X V I I ) by acid. Chromic acid oxidation of X X I I I affords a dehydrogermine acetonide diacetate (XXVIII) which affords a dehydrogermine diacetate ( X X I X ) on acid hydrolysis. Chromic anhydride-pyridine oxidation of XI1 I yields a formamido ketone derivative ( X X X ) ; similar oxidation of X I I ’ affords a n analogous product ( X X X I ) .
Germine, CziH43OJJ, is the alkamine present in Germine was first obtained by Poethke, in 1 9 3 i , many polyester alkaloids which occur in V e r u t r ~ m ~ -by ~ alkaline hydrolysis of its esters protoveratridine and Zygadenus’o species. The structure of germine and germerine.13 Poethke described several xystalis of particular interest in view of the powerful line salts and proposed the formula C?6H4105Nfor hypotensive action of its ester derivatives“ and of the alkamine. Craig and Jacobs further evaniined the use of this antihypertensive action in clinical the question of germine’s empirical formula and they conditions associated with high blood pressure. advanced a C27H4308Nformulation on the basis of In this paper evidence is presented for assignment their analytical work on the alkamine and on of structure and configuration I to germine. germine acetonide h y d r ~ c h l o r i d e . ~The ~ ” latter authors also demonstrated a close structural analogy to cevine by isolating @-picoline,2-ethyl-5-methylpyridine, cevanthrol and cevanthridine from the products of selenium dehydrogenation of germine. These results led Jacobs and Pelletier to propose OH skeletal structure I1 for germine as well as cevine.’j Germine undergoes a series of isomerizations (gerniine+sogermine+pseudogerniine~ which parallels H Of J y H those of zygadeninelBand veracevine.I7-l9 OH
”
27
I (1) P a r t X X V I I . S. h l . Kupchan and C. I. Ayres, THISJOURNAL, 81, 1 0 0 9 (1959). ( 2 ) T h e investigations which f o r m t h e subject of this paper were first outlined in p a r t in two preliminary communications: Chemistry b I n d u s l r y . 2 5 1 (1955); 1092 (1956). P a r t s of t h e investigations were performed by t h e a u t h o r s a t t h e D e p a r t m e n t s of Chemistry of Harvard University a n d t h e University of Wisconsin. (3) P a r t s of t h e work were presented a t t h e Symposium on t h e Chemistry of N a t u r a l Products, Technion, H a i f a , Israel, June 28-29, 1955; a t t h e X I V t h International Congress of Pure a n d Applied Chemistry, Zurich, Switzerland, July 21-27, 195,5; a t t h e 8 t h S u m m e r Seminar in t h e Chemistry of A-atural Products, Grand AIanan I s l a n d , New Brunswick, C a n a d a , August 6-10, 1056; and a t t h e 130th Meeting of t h e American Chemical Society, .4tlantic C i t y , S . J . . September 16-21, 1950. (4) T h i s investigation was supported by research g r a n t s (H-15ti3 a n d H-2275) from t h e National H e a r t I n s t i t u t e , of t h e Rational I n s t i t u t e s of Health, U S. Public Health Service. ( 5 ) W . Poethke, A r c h . P h a r m . , 275, 3 5 7 (1937). ( 6 ) ( a ) J . Fried, H . I,. White a n d 0. Wintersteiner, THISJ O U R N A L , 7 2 , 4G?l (1950); J. Fried, P.Numerof a n d N H. C o y , i b i d . . 74, 3 0 4 1 (1952). ( 7 ) R I . W Klohs, e l nl , i b i d . . 7 5 , 4905 (1953); 7 6 , 1152 (1954). (8) G . S . Myers, el ol., < b i d . , 7 7 , 3 3 4 8 (1955); 7 8 , 1 6 2 1 (1956). (9) H . A . Piash and R . M. Brooker, i b i d . , 7 5 , 1942 (1953). (10) S. 51. K u p c h a n a n d C . 1’.Deliwala, i b i d . , 74, 3 2 0 2 (1952); 7 6 , 5.i.15 ( I g i 4 ) ; S. M . K u p c h a n , C. V. DeIiwala a n d R . D. Zonis, i b i d . , 7 7 , 7 5 5 (1955). (11) E . D . Fries, J. R . S t a n t o n and F. C. Moister. J. Phnrmncol. E x f i l l . T h e r o p . , 98, 166 (l950), G. L. hIaison, E . Gotz a n d J . W. S t u t z m a n , i b i d . , 103, 74 (1951). ( 1 2 ) L. S Goodman a n d A G i l m a n , “ T h e Pharmacological Basis of Therapeutics,” T h e Macmillan Co.,New Y o r k , N. Y., second edition, 1955, pp. 747-754; 0. K r a y e r in V. A . Drill, “Pharmacology in Medicine.” McGraw-Hill Book co.,N e w Y o r k , N. Y . , second edition, 1958, pp. 515-524.
I1 Our earlier work was concerned with the nature of the germine-isogermine-pseudogermine isomerizations. The oxidation of germine acetonide and pseudogerrnine acetonide to the same aldehydo-7lactone showed that these compounds contain the same a-ketol-5-membered hemiketal system. Fur. therinore, the close analogy of the isomerizations to the v~race;-iiie-cel.agenine-cevine?’ isonierizations (13) W. Pocthke, A r c h . P h a n i i . , 276, 5 7 1 (1937). (14) L . C . Craig and W. A. Jacobs; ( a ) J . BiL.1. C h e i i t . , 148, :)7 (1943); (b) 149, 4 5 1 (1943). ( 1 5 ) a‘. A . Jacobs a n d S. N. Pelletier. J. Org. Chem.s 18, 70,; (1953); THISJ O U R N A L , 78, 1914 (1956). (1G) S. M. K u p c h a n a n d C . V. Deliwala, i b i d . , 7 6 , 1025 (1953). (17) S. W. Pelletier a n d W. A. Jacobs, i b i d . , 7 6 , 3 2 4 8 (1953). (18) S. M. K u p c h a n , D. Lavie, C . V . Deliwala a n d B. Y .A . Andoh. i b i d . , 75, 5519 (1953). (19) S.hf.K u p c h a n , If,Fieser. C. R . Karayanan. L. F. Fieser and J. Fried, i b i d . , 7 7 , 5896 (1955). (20) D. H. R. Barton, 0. Jeger, V. Prelog and R. B. Woodward, E r p r r i r n l i n , 10, 8 1 (1954).
1914
S. MORRISKUPCHAN AND C. R. NARAYANAN
suggested that germiiie possesses a 3-hydroxy-4hemiketal system similar to that found in veracevine. At least two early experiments support this view concerning the ring A system of germine. First, the presence of the masked ketol system in ring A, with its potentiality for ring contraction, accounts for the appearance of cyclopentanofluorene derivatives such as cevanthridine (111) among the dehydrogenation products of germine.14a.16.20 Second, the formation of the hexanetetracarboxylic acid IV by chromic acid oxidation of g e r m i ~ ~ supports el~~ the location of the masked ketol system a t the 3,4position. Incidentally, the obtention of acid IV
111 IY from germine excludes the attachment of oxygen functions a t CI, CP,Cg and C19 in the germine molecule. Germine readily formed a tetraacetate which was stable to chromic acidlg: hence, four non-tertiary hydroxyl groups are present in the molecule. Under more vigorous acetylating conditions, germine formed a pentaacetate, and, by analogy with the behavior of veracevine21 and cevineIz2i t may be assunled that the difiicultly acylable hydroxyl is the Cd-hemiketal hydroxyl. Suppoi-t for this view follows from the formation of a pentaacetate from dihydrogermine (presumably formed by reduction of the masked ketone a t C,) under non-forcing acetylating conditions.l9 Germine consumed three moles of periodic acid; germine acetonide consumed one mole of periodic acid (in the cleavage of the 3-hydroxyl-4-hemiketal system). Hence the acetonide blocks the uptake of periodic acid by a 1,2,3-triol system or a 1,2,4,5tetraol system. The former situation was shown to prevail by the fact that periodic acid oxidation of germine was accompanied by the production of one mole of formic acid. In addition, this result revealed that the central hydroxyl of the triol system is secondary. Germine acetonide was readily acetylated to a diacetate which consumed one mole of chromic acid and yielded a crystalline ketone (dehydrogermine acetonide diacetate) which was stable to chromic acid. Mineral acid hydrolysis of gerrriine acetonide diacetate afforded germine diacetate, which consumed one mole of periodic acid. (21) Two products with rather widely different physical constants have been assigned a veracevine triacetate formulation in the literature, The first oi these, m.p. 273-274O, wag obtained b y acetylation with acetic anhydride-pyridine a t O n (S. W. Pelletier and W. A. Jacobs. T E I SJOURNAL, 76, 3248 (1953)). This product has been re-examined and found t o be veracevine diacetate (Dr. S. W. Pelletier, private communication). The second product, m.p. 239-241°, obtained by acetylation with acetic anhydride-pyridine a t steam-bath temperature, is the true veracevine triacetate (reference 18). (22) D. H. R . Barton, C. J. W. Brooks and J. S. Fawcett, J . Chcm. Sac., 2137 (1064).
Vol. 81
These facts show that: (1) the acetonide is tertiarysecondary, for two of germine’s original four nontertiary groups are still available for acetylation, and a third subsequently available for oxidation to a ketone: and (2) the triol system is tertiary-secondary-secondary, for one of the acetate groups in germine diacetate clearly occupies a terminal secondary hydroxyl of the triol system. The arguments presented thus far lead to the following tentative disposition of the functional groups of germine c
c
The problem of locating these functional groups in the skeletal framework (11) is considerably simplified by the facile exclusion of all locations outside of ring D for the triol system. Rings A and B may be excluded on the basis of formation of the acid IV by chromic acid oxidation of germine. Rings C and E will not accommodate a tertiary-secondarysecondary triol. Ring F may be excluded on the bases: (1) that germine shows no carbinolamine properties, hence ruling out sites alpha to nitrogen for hydroxyl location and eliminating a C22, c23, c24 triol; and ( 2 ) that if the triol system were located a t CZ3,C24, (226, coiisumption of periodic acid by germine would not stop a t three moles (cf. ref. 20). Spedic location a t c14, ‘215, Cl6 is made possible by the following results. When the crude product of periodic acid oxidation of germine diacetate was exposed to dilute ammonia, an unsaturated ketone was obtained, A, 238 mp (e 10,000); ,A, 2.90(s), 5.78-5.85(s), 5.92(s), 6.05(rn)p. The ultraviolet spectrum of the product is incompatible with thepresence of a CIb-, CIe-, C,,-triol system in germine.2a The ultraviolet spectrum is, however, compatible with two alternative enones derived from a C14-, CiaClo-triol (V and VI).23 T h e infrared spectral
v characteristics of the product indicate that the A8j9-14-one structure V is the correct one. T h e relative intensities of the carbonyl band ( 5 . 9 2 ~ ) and the double bond band (6.05,) are those characteristic of a transoid system. -4 cisoid system such as VI would be expected to show a C=C band of exalted intensity equal to or exceeding the intensity of the C-0 band.24 The absence of ab(23) R . B . Woodward. T ~ I JOURNAL. S 63, 1123 (1941); 64, 76 (1942); A . E. Gillam and T.F. West, J . Chcm. SOC.,486 (1942). (24) R. Hirschrnann. C. 5. Snoddg, Jr.. C. F. Hiskey and N. L. Wendler, THIS JOURNAL, 76, 4013 (1954); 0. Wintersteiner and M. Moore, ibid., 78, 8193 (1968); D. H. R . Barton End C. R. Narayanan, 1. Chrm. SOC.,Q6a (19158).
April 20, 1959
STRUCTURE AND CONFIGURATION OF GERMINE
1915
Preliminary assignment of the remaining tertiary sorption characteristic of a trisubstituted double is made on the following basis. bond (the A73 in the case of VI) in the 11.9-12.5 p hydroxyl group to CZO range2j also mitigates against the A7J-14-ketone Neogermitrine, a naturally-occurring triester of gerstructure. The A8v9location of the double bond in mine, is convertible in two steps to a diosphenol the unsaturated ketone indicates that a tertiary for which structure X is derived.ls The simplest hydroxyl (or equivalent) is a x e d a t CS in germine. explanation for the formation of the 17-20 double Incidentally, the fact that isogermine and dihydrogermine consumed three moles of periodic acid (like germine) indicates the absence of a hydroxyl group a t CI1. Treatment of dehydrogermine acetonide diacetate with dilute mineral acid afforded dehydrogermine diacetate. The latter compound consumed five moles of sodium periodate in the course of twentyHO,' QdQH On four hours. The result is readily compatible with the view that oxidation proceeds eVia initial cleavage of the C14,C15-glyco1 system to generate a cyclic X 1,3-diketone. Further cxidation of the &diketone v i a hydroxylation a t Cg would account for the addi- bond of X envisions elimination of a hydroxyl beta tional periodate consumption (see Chart 1).26 to the 16-keto group, viz., the Czo-hydroxyl. Support for the location of a hydroxyl group at CZO CHART I follows from a study of the facile methanolysis of the acetate grouping a t ClO in certain germine esters -an effect which has been shown to be attributable { & W O A 0e OH to a 1,a-diaxial interaction of the acetate a t Cle with the hydroxyl a t C ~(see O below). YII The foregoing preliminary considerations served I imol as a basis for the development of the structure (apart from configurational relationships) of germine represented by formulation I. Additional corroborative evidence for this structure as well as for the configuration indicated will now be discussed in the light of the proposed formulation. The germine-isogermine-pseudogermine isomeriWe believe that the five-mole periodate uptake of zations are now to be formulated as I-+X14XII.1s~m dehydrogermine diacetate is explicable uniquely The tetraacetates formed by acetic anhydrideon the basis of a structure with a keto group beta to c14. Since Cl1has been excluded as an oxygenated pyridine acetylation of the respective isomers1B position (see above), dehydrogermine diacetate has are 3,7,15,16-tetraacetates (e.g., XIII). The pentaa keto group a t C7 (see structure VII) and germine acetate formed by more vigorous acetylation catahas a hydroxyl group a t C,. The failure to detect the lyzed by sodium acetate or perchloric acid is to be lactone tricarboxylic acid VIIIa among the products represented as the 3,4,7,15,1G-pentaacetateXIV. of chromic acid oxidation of germinella is readily Some stereochemical aspects of the isomerizations explicable on the basis of a germine fonnula con- and the acylations are discussed below. taining a hydroxyl group a t Cr. Expression IX, CHART 2 which incorporates seven of the eight oxygen atoms of germine, summarizes the conclusionsdrawn to this point .27
$cy
A
--_
i
-\
,
i l + I
A. n
nv
n
on ( 2 5 ) L. J . Rellamy. "The Infra-red Spectra of Complex iMolecules." Mcthuen and Co., Ltd., London, 1954. ( 2 6 ) C/ M . L. Wolfrom and J. M . Bobbitt, TEXS J O U R N A L , 7 8 , 2489 (IF(56).
(27) The sequel will show that the bulk of the evidence strongly favors a l,%hemiketal structure for those derivatives having a group other than hydroxyl at C7. I n those compounds which possess a free hydroxyl group a t C, (e.g., the alkamines), an unequivocal choice between the 4,9-hemiketal and the 4.7-hemiketal structure considered earlier's is not possible. We shall discuss our results on the hasis of the preferred 4,g-hemiketal (Itructure.
.-
xrn
OAc
XIV
The acetonide derivative of germine,148which has proved to be a key compound in the development of the structural argument, is formulated as the 14,15acetonide XV. The periodic acid oxidation of germine 14,15-acetonide (XV) and pseudogerminc 14,lBacetonide (XVII) to the same aldehydo-ylactone XVI is represented as in Chart 3. Acetyla(28) S. M. Kupchan. rbid.. 81, 1921 (1959).
S. MORRISKUPCHAN AND C. R. NARAYANAN
1916
CHART
3
I
0
XXI
VOl. 81
XI1
II
0
tion of XVI with acetic anhydride-pyridine afforded a diacetate, stable to Fehling solution. This diacetate showed infrared absorption a t 5.62 (ylactone), 5.80 (acetate) and 5.97 p (double bond) and is represented by formula XX.29 Acetylation of the oxime of the aldehydo-y-lactone XIX yielded the acetonide-aldehyde-oxime-y-lactone diacetate XXIT. Gennine acetonide (XV) isomerized readily on alkahne treatment t o give isogennine 14,15acetonide (XVIII). Attempted preparation of XVI11 from isogerniine (XI) by reaction with acetone and concentrated hydrochloric acid in the usual was unsuccessful. However, reaction in acetone in the presence of hydriodic acid did yield XVIII. Acetylation of XVIII with acetic anhydride pyridine gave isogcrniine 14,15-acetonide 3,lGdiacetate (XXI). The diacetate formed by acetylation of geriiiine 14,15-acetonide% is now formdated as XXIII. Hydrolysis of XXIII with 2,4-dinitrophenylhyIrazine in strong alcoholic sulfuric acid30 afforded in small yield a germine monoacetate. The monoacetate is formulated as germine 16-acetate on the basis of its periodic acid consumption (Table I) and rotation in pyridine.28'31Hydrolysis of XXIII with dilute hydrochloric acid gave gennine 3,16diacetate (XXIV). As noted earlier, periodic acid oxidation of XXIV followed by exposure of the initial product to dilute base gave a product with (29) Cf.A . h l . Sladkov a n d G . S. Petrov, J . Gcn.Chem. U . S . S . R . , 24, 459 (1955)
(30) F. L. Weisenborn, J. W. Rolger, D. B. Rosen, L. T . M a n n , Jr., I,. Johnson a n d H. L. Holmes, THISJ O U R N A L , 76, 1792 (1954). (31) Cf.t h e hydrolysis of cevine 3.16-diacetate to cevine 16-acetate in t h e presenre of perchloric acid: n. H R . B a r t o n , C 1. W . Brooks * i t i t 1 1'. De R I . I ~ o J . C h c m . . 5 < ~3Y,vdl , (l!).j4).
XXIl
partial structure V. The formation of the cyclopentenone is clearly explicable in terms of initial glycol cleavage a t C14, CIS to a p-hydroxycyclopenTABLE I
PERIODIC L4cInOXIDATIONS Substrate
Mole equivalents of HI04 consumed' (hr.)
2 . 7 ( I ) , 2 . 9 (4) Gerniine ( I ) Isogermiiie ( S I ) 2 . 8 (l),2 . 9 (4) 2 . 7 ( l ) , 2 . 9 (4) Pseudogermine (S11) Dihydrogerminel4b 2 . 7 (1) Germine 3,7,lj,lG-tetraacetatc ( S I I I ) 0 (24) 0 . 8 ( l ) , 1 . 0 (1) Germine 14,15-acetonide (XV) Pseudogerminc 14,lfi-ncetouide ( S V I I ) 1 . 0 ( l ) , 1 . 2 (4) 0 . 9 ( l ) , 1 . 1 (4) Isogermine 14,15-acctotlide ( X V I I I ) Isogermine 14,15-acctonide 3,16diacetate ( S S I ) 0 (18) Gerniine 14,15-acetonide 3,16tliacetate ( X X I I I ) 0 (18) Germine 3,16-diacetate (XXII-) l . u (1) 1 . 7 ( l ) , 2.2 ( 6 ) Germine 16-acetate 1 . 8 ( l ) , 2 . 0 (6) Gerniine 3-acetate (XXVII) 1.4(2). 1 . 9 (4); 7-Dehydrogermine 3,16-diacctatc 3 . 4 (241, 4 . 6 (48) (XXIX )b a The last uptake recorded in each case is the one beyond which n o significant change occurred on further standing. * To test the effect of pH on the rate of oxidationZeofXXIX, the compound was oxidized also with sodium p e r i o d a t e dilute acetic acid; results: 1.9 (2 hr.), 2.7 ( 5 hr.), 4.7 (24 hr.), 5.3 (48 hr.).
tanone, followed by p-eliinination t o XXV. hlethanolysis of XXIII gave germine 14,15-acetonide 3acetate (XXVI). Dilute acid hydrolysis of XXVI afforded gcrrni~ie3-acc.tate (XXVII 1.
STRUCTURE AND CONFIGURATION OF GERMINE
April 20, 1959
CHART
XY
AcO
1917
4
ACU
AcO
OH
OH
XXYIIT
XXlX
derivatives support this view. Thus, the infrared spectrum of germine 14,15-acetonide 3,16-diacetate (XV) shows (in addition to the usual hydroxyl bands a t 2.90-2.95~)a band a t 3 . 0 1 which ~ is attributable to a strong intramolecular hydrogen bond.25.34 The band is absent from the spectra34of isogermine 14,15-acetonide 3,16-diacetate (XXI), 7-dehydrogermine 14,15-acetonide 3,16-diacetate (XXVIII) and zygadenine 14,15-acetonide 3 , l G - d i a ~ e t a t e . ~ ~ The latter compounds lack either the basic ethereal been reiated wiih the configuration a t in steroids oxygen of the 4,9-hemiketal bridge or the 7-av i a the hexanetetracarboxylic acid IV.32 The con- hydroxyl group necessary for the postulated strong figurations a t Cy, C4, C, and Cg are related to CIOas intramolecular hydrogen bond. Additional support for assignment of a-orientation shown in formula I by the steric requirements of the to the hydroxyl group a t C7 follows from another isomerization sequence I+XI+XII.20 Alkaline equilibration studies of the germine experiment. Acetylation of gemiine with acetic isomers show that the order of stability is isogermine anhydride-pyridine (conditions which lead to acetyl(3-P-hydroxy-4-keto-9-a-hydroxy-A/B-trans, XI) ation of the C4-hemiketal hydroxyl in veracevine'*,a) i ) 130- 131.
I'czl
Vol. 81
red spectrum of this material (Nujol mull) was identical with t h a t of germine. The chloroform mother liquors above yielded a second crop of crystalline product (40 mg.) upon concentration. This material showed a rotation ( [ a I z 8-33", ~ c 1.00, alc.) and infrared spectrum (Nujol) indicative of impure isogermine. Evaporation of the chloroform mother liquor to dryness gave an amorphous residue (301 mg.), [ C Y ] ~ * D0" (c 1.00, alc.). The infrared spectrum of this residue suggested that i t consisted of pseudogermine (predominantly) and isogermine. Alkaline treatment of germine under the same conditions yielded a mixture practically indistinguishable in composition from that above. Germine Acetonide Monoacetate (XXVI).-Germine acetonide diacetate (500 mg.) was dissolved in methanol (15 ml.) and water (7 ml.) and the solution was allowed to stand a t room temperature for 18 hours. Conrentration of the solution on the steam-bath led t o crystallization of needles (370 mg., m.p. 259-262" dec.). Recrystallization from methanol-water afforded colorless needle clusters (250 mg.), n1.p. 263-265" dec., [ a I z 3 D 1-40' ( C 1.36, pyr.). Anal. Calcd. for C~OH&K(COCH~):C , 64.95; H , 8.35; acetyl, 7.26. Found: C, 64.33; H, 8.11; acetyl, 7.15. Germine 3-Acetate (XXVII).-Gerniine acetonide monoacetate (480 mg.) was dissolved in 1 : 4 dilute hydrochloric acid (10 ml.) and the solution was allowed to stand a t room temperature for 15 minutes. Dilute ammonia was added t o p H 8.5 and the solution was extracted with chloroform (ten 15-ml. portions). The chloroform was dried over sodium sulfate and brought to dryness i n vacuo, leawng a residue which crystallized from ether. Filtration afforded colorless clusters of needles (288 nig. j , m . p . 219-221", [ a I Z 3 D +IO" (c 1.05, pyr.). AvaZ. Ctlcd. for C?,H4208N(COCH3): C , 63.13; H, 8.22; acetyl, 7.80. Found: C, 63.19; €1, 8.25; acetyl, 7.19. 7-Dehydrogermine Acetonide Diacetate (XXVIII).Germine ncetonide dhcetate ( 2 9 , ) in pyridine (20 1111.) was added t o chromic aiihydridc ( 4 g.i in pyridine (40 mi.) and tlie solution was allowed to st,uitl a t roum ternpcrature for 3 hours. LVater (40 m1.j and ammonium hydroxide ( 4 ml.) were added and the solution was shaken with chloroform (100 mi.) and filtered through Supercel t o clarify the emulsion. -1fter eight additional extractions with chloroform (50 ml.) the combined chloroform extracts wcre brought to dryness in zacuo. The residue crystallized from alcohol; yield 690 mg.. ii1.p. 261-263" dec. Recrystallization from acetone-petroleuin ctlier gave colorless elongated prisms (530 mg.), ni.p. 267-269' dec., [ c Y ] * ~ D-42' (c 0.86, ovr.1. . .41zol, C,ilcd. for Ca,I14308S(COCIIB\.I: C , 64.64; 11, 7.52; acetyl, 13.63. I?ntiiicl: C , N S l ; I f , 8.03; acetyl, 13.19. Siniilar yields c!f product I\ crc obt,iiiic.cl by oxidatioii of gerniiiie acctouitle tliacetate ( 2 8 . ) in ncctic acid (5 ml.) antl carbon tetraeliloridc [CJU 1111.) with O.66 -V chromic anliydride in 98.5:10 acetic 'tcitl ( 7 0 ml.).sG 7-Dehydrogermine Diacetate (XXIX).-;-L)eliydrogermine acetoiiide di:icet,tte (1.25 g . ) \vas dissolved in acetic ticid ( 2 I i 1 l . j and 1:1 dilute liydrocliloric acid (10 nil.) and I V X S :illo\vctl t o s t m d a t room temperature for the s~iiiiti~iii 4 hours. \\'vrk-up with dilute amirioiiia ant1 chloroform in the usual nianner gave a residue which crystatlized from acetone-ether. Recrystal1iz:rtion from the same solvents gave clusters of plates (55U mg.), ni.1). 235-237' dec., [ a ] D -68" ( 6 3.23, pyr.). Anal. Calcd. for C Z ~ E ~ ~ ~ O ~ N ( C O C ,C 62.92; H ~ ) ~ : H, 7.67; :icetyl, 14.55. Found: C, 62.56; H , 7.65; acetyl, 13.55. The corresponding oxime, prepared by the pyridine procedure, crystallized froiii alcohol, m.p. chars a t 250°, does not melt to 300'. Anal. Calcd. for C I ~ I 1 , ~ O I ~ S ?C: , 61.31; H, 7.64; S , 4.62; acetj-I, 14.19. I ~ o u n d : C , 61.53; H, 7.52; N, 4.92; acetyl, 13.05. Sodium Borohydride Reduction of 7-Dehydrogermine Acetonide Diacetate.-7-Ikli>.tlrogcrniine acetonide dittc (?ill)iiig.') iii invtli,ui ( j ) , ( 1 I) S. h l . K u p c h a n and C . V . Deliwala, ibid., 7 6 , 53i5 ( I ! l S ) (151 S. 11, K i i i x h a n , C . V . neiiwala and 11, D %onis, i b i d . , 77, 7;): (IBi5).
the alkamine germine (I).I In all, eleven wellcharacterized germine esters have been isolated froin the alkaloidal extracts of plants.I6 Some of t h e reported physical and chemical properties of these ester alkaloids are summarized in Table I. It is t t e purpose of this paper to present the structure elucidation of several of the naturally-occurring gerniine ester alkaloids. The series of esters relutetl to iieogcrniitrilic has received thc most clieniical attcwtioii (Chart 1). hfethanolysis of neogeriiiitriiic was shown to result in loss of one acetate grouping with coiiversioii to the tlicster gerniidi~ie.~Furthermore, 011 acetylation both alkaloids were converted to the smie product, nionoacetyliieogerniitriiie. Acid hydrolvsis of neogermitrine led to loss of an acetate group with conversion to the diester neogermidine. The germ;il.ine isomer neogermidine was also shown to undergo acetylation to monoacetylneogermitrine. Finally, a synthetic mono-(/)-2-methylbutyrate of germine has been shown to be convertible to the same acetylation product." These relationships reveal that the site of attachment of the ( / ) - 2 (IC,) I n view of the faciic di,.icylation of 5ome germine ester alkaluicls ( w i t i r . :>>,fro), the prisiibility exists t h a t scitue of the di- a n d rnnnucsters iii.iy lie artifacts f o r m e d d u r i n g the isul:ition pruccdures. 17) F. 1.. 1Veisenbc)rn :Ind J. \!' Ii
