June 20, 1952
COMMUNICATIONS TO THE EDITOR
3201
generally weaker than those of the triazines and tion gave tigogenin (isoallospirostan-3 a-01) m.p. perhaps qualitatively different. Thus V is only a 203-205', yield 3.9 g. ; (2) manogenin (isoallospiroweak inhibitor of L. cas& and its activity is not stan-12-one-2a,3/3-diol) m.p. 240-244", yield clearly blocked by folic acid. Furthermore, its 1.3 g., on Wolf-Kishner reduction gave gitogenin antimalarial activity is minimal. (isoallospirostan-2c-30-diol) m.p. 268-270'. InThe above observations lend support to the pre- frared spectra of all sapogenins and their Wolfviously expressed views concerning the relationship Kishner reduction products were identical (with between antifolic acid and antimalarial activity2 exception of manogenin which showed slight diverand further delineate the chemical structures with gencies) with those of authentic specimens. Hecowhich this activity is associated. genin and manogenin had a carbonyl peak in the G.H . HITCHINGSinfrared at 1708-1710 cm.-', absent in reduced THEWELLCOME RESEARCH LABORATORIESA. MAGGIOLO products. Manogenin also showed conjugated carTUCKAHOE, NEWYORK P. B. RUSSELL H. VANDERWERFPbonyl a t 1678 cm.-l from which it is deduced that l0-20% of the 9(11)dehydro component was present. THEWELLCOME LABORATORIES OF TROPICAL MEDICINE I. M. ROLLO From I V after subsequent acid hydrolysis and LONDON, ENGLAXD chromatography were isolated 2.4 g. of sapogenin RECEIVED APRIL29, 1952 similar in composition to those found in 111,indicating that the enzymatic hydrolysis was 7840% complete under our experimental conditions. STEROIDAL SAPOGENINS. V.' ENZYMATIC In a duplicate experiment, the sapogenins were HYDROLYSIS OF STEROIDAL SAPONINS isolated entirely by means of acid hydrolysis. The Sir : yield of total sapogenin was similar and the sapoIt is well known that plants containing cardio- genins isolated identical with the enzymatic hyactive glycosides have enzyme systems which can drolysis. partially hydrolyze these substances.2 We have Using methods similar to those described above, recently found that leaves of some Agave and Yucca acid or enzyme hydrolysis of Agave serulata5 gave species which contain steroidal saponins have simi- hecogenin, manogenin and small quantities of the lar enzyme systems. In contrast to the enzymes 9(11)-dehydroanalogs of both sapogenins; an uniacting on cardio-active glycosides, those found in dentified Yucca speciesfiyielded with both methods the sapogenaceous plants tested cleave the saponin sarsasapogenin (spirostan-3 p-01) and an unidentisubstrate to the sapogenin aglycone and component fied Dioscorea species gave diosgenin (A5-isospirosugars or polysaccharides. sten-3p-01) in both cases. An enzyme preparation was obtained by grinding In agreement with Marker, et al.,I we find that 2 kg. of frozen Agave toumeyana3leaves, followed by sapogenins occur in plants only as glycoextraction with water at 10'. The filtered extract steroidal sides as evidenced by the fact that free sapogenin was shaken with benzene I and the aqueous layer could not be found prior to hydrolysis. Our data containing the enzyme fraction held a t 4'. The are not in accordance with the views of Marker and leaf residue remaining from the aqueous extraction Lopez8 that the spiroketal side chain and the 12was refluxed with 95% ethanol to extract residual keto group in sapogenins are artifacts produced by saponins. The ethanol solution was concentrated, acid hydrolysis. The fact that the much milder diluted with water, extracted with benzene 11, and enzyme hydrolysis gives the same products as acid the aqueous saponin fraction heated on the steam cleavage tends to cast doubt on Marker's hypothebath to drive off residual alcohol. sis. Our previous findings0 based on infrared studThe aqueous enzyme and saponin solutions were ies of saponins have also shown that the structure of combined, adjusted to pH 5.25 and held a t 37' for the steroidal moiety of saponins and their, sappgen90 hours. The resultant turbid suspension was ex- ins derived by acid hydrolysis are in all probability tracted first with a mixture of 90% benzene--lO% identical. ethanol I11 to recover enzyme hydrolyzed sapogenBecause of the mild conditions used, tars, resins ins and then with butanol IV to obtain unhydro- and pigments produced by acid hydrolysis are not lyzed or partially hydrolyzed saponins. in the enzymatic method, thus simplifying Only a trace of sapogenin was found in the pre- found the isolation of purified sapogenins. hydrolysis benzene extracts I and 11. The extract distribution, occurrence, and physico-chemiI11 was concentrated and chromatographed on ac- calThe properties of the enzyme systems found in sapotivated alumina. Using methods presented in de- genaceous plants are being studied and will be retail elsewhere4 the following sapogenins were iso- ported in detail a t a later date. lated : (1) hecogenin (isoallospirostan-12-one-3 PREGIONAL RESEARCH LABORATORY ol), m.p. Kofler 255-260°, on Wolf-Kishner reduc- EASTERN BUREAUOF AGRICULTURAL AND INDUSTRIAL CHEMISTRY
(1) Paper I V of thisseries submitted for publication in THIS JOURNAL. This work was done as part of a co6perative arrangement between the Bureau of Plant Industry, Soils, and Agricultural Engineering and the Bureau of Agricultural and Industrial Chemistry, United States De partment of Agriculture, and the National Institutes of Health, Federal Security Administration. (2) For pertinent references see A. Stoll and associates, Hclv. Chim. A c f o , 1933-1951,and W. A, Jacobs, Physiol. Rev., 13, 222 (1933). (3) Collected by H. S. Gentry, #S 2046, Superior, Arizona, December, 1951. (4) M. E. Wall, M . M. Krider, E. 9. Rotbman and C.R. Eddy, J . Bfol. Chem.,in press.
-
AGRICULTURAL RESEARCH ADMINISTRATION U. S. DEPARTMENT OF AGRICULTURE MERLEM. KRIDER E. WALL MONROE 18, PA. PHILADELPHIA RECEIVED MAY9, 1952
(5) Collected by H. S. Gentry, #1865, San Ignacio, Baja California, November, 1951. (6) Collected by H. S. Gentry, #Zl6l,Jecabasco, Mexico, December, 1961. (7) R . E. Marker, el aZ.,TEISJOURNAL,69, 2167 (1947). ( 8 ) R. E. Marker and J. Lopez, ibid., 69 2390 (1947). (9) E. S. Rothman, M. E . Wall and H . Walens, i b i d . , in preas.
