A STUDY OF T1IE STEROLS, STEROLINS, AMD CERTAIN jiLCOHOLS OF SOME LEO-TT.Ei SEED OILS by L. Carroll King A THESIS Submitted to the Graduate School of Michigan State College of Agriculture and Applied Science in partial fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1942 I wish to thank Professor C. D. Ball CONTENTS Introduction------------------ 1 Presentation of Methods and Data I. II. The Sterols from the Unsaponifiable Fraction of Hardigan Alfalfa Seed Oil.---------------- 3 Isolation of a Hydrocarbon and of a Fourth Alcohol from the Unsaponifiable Fraction of Hardigan Alfalfa Seed Oil.------ 19 III. Isolation and Comoosition of a Phytosterolin from Hardigan Alfalfa Seed Oil.-----------------IV* V* VI* The Sterols from the Unsaponifiable Fraction of Grim Alfalfa Seed Oil.----------------Isolation and Composition of a Phytosterolin from Grim Alfalfa Seed Oil.------------------ 24 32 37 The Sterols from the Unsaponifiable Fraction of Dutch White Clover Seed Oil.------------------ 42 VII. Isolation and Composition of a Phytosterolin from Dutch White Clover Seed Oil.------------ 46 VIII. The Sterols from the Unsaponifiable Fraction of Medium Red Clover Seed Oil.------------------ 50 IX. X* Isolation and Composition of a Phytosterolin from Medium Red Clover Seed Oil.-------------- 53 Observations on the Reactions of Sterols of the Spinasterol Type with Halogens.--------------- 56 Discussion 1. A Comparison of the Sterol Com­ position of Hardigan Alfalfa Seed Oil, Grim Alfalfa Seed Oil, Dutch White Clover Seed Oil and Medium Red Clover Seed Oil, — - 64 A Consideration of the Structure and Homogeneity of Sterols of the Spinasterol Type.------------------ VI T3it>liography--------------------------------- 74 2, Introduction The intent of this study was to compare the amount and identity of sterols occurring in plant seeds with variety, species and genus differences of plants. The sterol composition of seeds of two varieties of alfalfa (Hardigan alfalfa seed and Grim alfalfa seed) and of two species of clover (Medium Red clover seed and Dutch White clover seed) was examined. The results of this investig­ ation, although limited in scope and as yet in­ complete, indicate that sterol composition is the same for different varieties of single species and for different species of a single genus. The sterol composition of the two genera studied was distinctly different. There have been previous attempts^ to classify sterols as characteristic of certain orders or families of plants* In view of the occurrence of 0 £»-spinasterol2 , 3 , / i >5 , 6 , 7 in very different plant types it is probable that such classifications are unjustified. There is need for a detailed char­ acterization of the sterols of many more plants before any general relationship can be defined. It is well known that sterols form a major portion of the unsaponifiable fraction of most —2— seed oils. Accordingly the crude sterols were isolated from the unsaponifiable fraction of the oil from each of the seeds mentioned and some progress was made in separating the crude material into its component chemical individuals. In each of the oils studied there was observed a characteristic phytosterolin which could be hydrolyzed into a mixture of sterols and a reduc­ ing substance. These compounds are resistant to alkaline hydrolysis and do not accompany the un­ saponif iable fraction. They were, therefore, iso­ lated and worked up separately. Phytosterolins probably occur very generally in plant soed oils and any study of sterols should provide for ob­ servation of them separate and apart from the sterols in the unsaponifiable fraction. In order to present the experimental part clearly it has been written as a series of short Sections, I-X . In a special addendum, Section XI, the sterol composition of the four seed oils has been summarized. -3- I. The Sterols from the Unsaponifiable Fraction of Hardigan Alfalfa Seed Oil. Three isomeric sterols, CL -spinasterol, (3 -spinasterol and a new sterol of the same gener­ al type which for purposes of this thesis was designated S -spinasterol, have been isolated from the unsaponifiable portion of Hardigan alfalfa seed oil. In order to separate these isomers the crude sterols were dissolved in a large excess of acetic anhydride. On cooling this mixture, the acetates of 01 -spinasterol and /3 -spinasterol separated as flakey crystals and were filtered off while the & -spinasteryl acetate remained for the most part in the acetic anhydride mother liquors. oc-spinasterol and (3 -spinasterol were separated by saponifying the crude acetates and repeatedly recrystallizing the resultant sterols from 85% alco­ hol. (3 -Spinasterol is considerably more soluble than CL -spinasterol in this solvent and was con­ centrated in the mother liquors. CL -Spinasterol w a s purified by extensive recrystallization from methanol and from chloroform-mehtanol, in an attempt to bring its rotation and melting point to the values reported by Kuwada -4- and Yosiki 8 *9 for bessisterol. This effort was not successful. QL -spinasterol has previously 5 been isolated from alfalfa seed oil . Several investigators0 ,-L0 have been able to reduce ql -spin­ asterol to QL -spinastenol which was subsequently identified as CL -stigmastenol by Fernholz and Ruigh^1 . The CL -spinasterol isolated from al­ falfa seed oil gave CL -stigmastenol on catalytic reduction. The purification of -spinasterol presented some difficulty since traces of CL -spinasterol persisted. This a. -spinasterol was removed by re­ peated recrystallization from 85^ ethyl alcohol. (3 -Spinasterol has been previously observed by 1P as a constituent of spinach fat. Heyl and Larsen-^ These authors noted that it absorbed one mole of hydrogen to give the same "spinastanol" as QL -spin­ asterol. Their analysis indicated the formula was C2 7 H4 5 O • Since recent work has established the reduction product of CL -spinasterol as CL-stigmastenol'*'^, it is apparent that the (3 -spinasterol of spinach fat-^2 has a basic structure containing twenty-nine carbon atoms. The (3 -spinasterol isolated from Hardigan alfalfa seed oil has the formula CggH^^OH • On catalytic reduction it gave (X. -stigmastenol (C2 9 H4 9 °H ). The (3 -spinasterol of Hardigan al- -5- falfa seed is therefore a doubly unsaturated sterol, isomeric with OL -spinasterol, and identi­ cal with the f t -spinasterol of spinach fat. £ -Spinasterol was best purified by saponi­ fying the crude acetates and recrystallizing from eth3rl alcohol and from methanol. Anal^/sis indic­ ated the formula C2 9 H 4 7 OII . On catalytic reduction it gave (X -stigmastenol. -Spinasterol is there­ fore a doubly unsaturated sterol, isomeric with QL -spinasterol. Sobotka-1' tacitly identified the |S -spin­ asterol of spinach fat*^ with the iso-spinasterol^-3 produced when OL -spinasterol was heated with chloroacetyl chloride. This assumption is not necessarily true since ^ -spinasterol has es­ sentially the sane melting point as that reported for iso-spinasterol. Summaries of the separation and of the re­ lationships of the three isomeric spinasterols are presented in Fig. 1 and Fig. 2 . Experimental 1. Separation of Crude Sterols from the Unsaponifiable Fraction. One hundred and fifty grams of crude unsaponifiable material from Hardigan alfalfa seed oil, Crude Hardigan Alfalfa Seed Sterols, m.p. 120-157° dissolve in acetic anhy­ dride (30 ec per gm). Let stand at room temperature soluble insoluble OL-spinasteryl aeetate ^-Spinasteryl acetate -spinasteryl aoetate Saponify, recrystalllze 10 times from 85# ethyl alcohol. Hydrolyze aoetic anhydride, saponify acetates, reorystallize from 95# ethyl alcohol and from methanol. More soluble material 4* -spinasterol m.p.143-145° sol*n orude crude <9-spina- ot-spinasterol sterol RecrystalEvapor­ ate sol­ lize from vents, Methanol recrysand MeOHtallize from 95# eth­ yl alc­ ohol. CHCIq P-Spinasterol m.p.147-149° & -spinasterol m.p.166-168° Fig. 1 Stannary of the separation of O L -spinasterol, and S -spinasterol p -spinasterol lb ciin C -C I I - C lI r C K - C J I - C H - C H z « s Np\ c J_ I c kr di3 CL -srlBast»r 1 '■-erzout-*, n.p. 197° KOH h3 ™ 3 ch3 C - C l - ' dccii-fH-CH-a;3 CIi3 HO CL -spinasterol*, u..n. 135° j§-3rinast®r:'l t*nzoate m.p* 180° >h->H ■ cotic anhydride <5 -spinasteryl benzoate m.p. 164° % P l3 9«3 ;!i-Ch=CM-Cl:-CH-CH3 benzoyl e h ' oride ch3 KOli KlH benzoy1 chloride c h 3-c -u Y CL -spinasteryl acctat"*, m.p. ISC'0 /0 -spinasterol ru.p* 148° g -sp'nasterol m.p. 143° HC ?t acetio anhydrl ?e KOH c p-spinasterj/l acetate m.p. 157° fHs -CK-CK- - "3 — «2 rt KOH 9% ;- :- c h 2- ci ci ucct! c an­ hydride ch3 ch3 «2 CH3-C-0- Ft g-spinasteryl acetate m.p. 133° ot -spinastonyl acetate CL -stlgnastenyl ac-tate!, rr..p. 115° KCH ic-stl c an;-ydrl la H 3 p!3 C ck 3 -C U -C Ilg -C K E -IJ ’i-CIl-CILe, « 3l f 1I CJ A f Kg ch3 HO- d. -splnaston'1 ( CZ-stlgmastenol), m.p. 112° Fie. 2 Chemical reactions, ard relationships of Ct -spinasterol, and g -spinasterol * Formula a coord! re t' F“raho'z end Rurg/i /3 -spinasterol -8- prepared as directed by King and Ball^, was dis­ solved in enough ethyl ether to make about 500 cc, A stream of water vapor* was passed through the solution until a slight turbidity occurred and the mixture let stand over night at 5°. The crystal­ line mass which separated was filtered off and recrystallized from ethyl ether. The original mother liquors, and those from the subsequent recrystallizations were concentrated somewhat and the whole process repeated. In this way £6 frac­ tions were obtained. The £6 solid fra.ctions weighing about 47 g. were nearly free of colored oily material. The first fraction isolated melted about 158°, sub­ sequent fractions had lower and lower melting points, while the last fractions had very indefi­ nite melting ranges in the vacinity of 60°. These crude fractions were combined and classified into two groups, those melting above 1£0° (Fraction A) and those melting below 1£0° (Fraction B). Frac­ tion B will be considered in Section II of this thesis. * A small amount of water greatly facilitates the separation of the sterols from the oily mixture. These sterols tend to crystallize with one-half mole of water, if water is available. -9- 2. Fractionation of the Crude Steryl Acetates from Cold Acetic Anhydride. The crude sterol fractions melting above 120° (Fraction A) were combined and dissolved in acetic anhydride (30 cc. per gram). This mixture was heat­ ed one hour, let stand over night, and then filter­ ed. The solid steryl acetates (Fraction A i ) so obtained consisted mostly of (X -spinasteryl and f3 — spinasteryl acetates. Yield 24.5 g., m.p. 152- 157°. The acetic anhydride mother liquors from the above were hydrolyzed by heating with water. The solid steryl acetates precipitated by this treat­ ment (Fraction A g ) were filtered off and recrystal­ lized from 95$ ethyl alcohol. Yield 2.9 g., m.p. 122-127°. 3. Separation and Identification of OL -spinasterol and p — spinasterol. & -Spinasterol— The crude acetates, m.p. 152-157°, (Fraction A ^ ) were hydrolyzed by boiling 1 hour with 5$ alcoholic potassium hydrox­ ide. The reaction mixture was poured into water and extracted with ethyl ether. The ether solution was washed with water and evaporated to dryness on the steam bath. The crude sterols obtained were 10- dissolved in just sufficient boiling 85$ ethyl alcohol to effect complete solution. This mixture was let stand over night and the crystalline maerial separated. The solid material was then re­ dissolved in a minimum amount of boiling 85% ethyl alcohol and let stand again. This proceedure was repeated 10 times. The resultant crystalline material was then recrystallized from methyl alcoh­ ol, and from chloroform-methvl alcohol; m.p. 168.5169°; M form, age q 7- “ 2.68° (556 mg., 10 cc. chloro- - 2 dcm., 27 0Lo m - 0.298°, aver­ reading). All the attempts to bring the ro­ tation of OL -spinasterol to the value - 8.5° or - 15.5°, as reported by Kuwada and Yosiki® * 9 failed. Ok -Spinasterol isolated in this way forms a pre­ cipitate with digitonin. QL -Spinasteryl acetate— A quantity of OL -spinasterol was dissolved in acetic anhydride and the mixture heated one hour. On standing the crystalline acetate separated. The product was filtered off and recrystallized from 95$ ethyl 57 alcohol; m.p. 180-182°; W o (52.9 rag., 2 cc. chloroform, SL7 oL q = = - 6.35° — 2 dcm., - 0.336°, average reading)* Attempts to bring the rotation to the value - 13.46° reported by Kuwada and Yosiki9 were not successful. -11 CL -Spinasteryl benzoate— milligrams of Five hundred OL -spinasterol was dissolved in 1.5 cc. pyridine and 0.5 cc. benzoyl chloride added. The mixture was heated in a boiling water bath 2 hours, let stand over night, poured into ice cold 5% sulfuric acid and the product extract­ ed with ethyl ether. The ether solution was washed first with 1% sodium carbonate, then with water, and finally evaporated to dryness on the steam bath. The product was recrystallized twice from 95% ethyl alcohol. Yield 350 mg.; m.p. 196-199°; 2.20° (51.6 mg., 2 cc. chloroform si*« 0I n /9-^Spinasterol— = 0.114 , average r The combined mother liquors from the isolation of CL -spinasterol were evapor­ ated to a small volume and water added. The pre­ cipitate was filtered off and taken up in the smallest volume of boiling 85% ethyl alcohol which would effect solution. After standing over night at room temperature the solid material was filter­ ed off. From the mother* liquors a fraction corre­ sponding to the |3 — spinasterol of Ileyl et al.^* was isolated. On recrystallization from 95% ethyl alcohol it gave flakey transparent crystals; m.p. 148-150°. In melting they lost water of -12- crystallization at 110-125 . r \0L\ (52.7 mg., 2 cc. chloroform, 20 0^/3 ^ — 5.91° = 2 dcm., 0.5116°, average reading). The sub­ stance formed sin insoluble precipitate with digitonin. Anal. 2.391 mg. gave 2.582 mg. II0H; 7.240 mg. C02 C, 82.56 ; H, 11.98 Calcd. for C2 9 H 4 7 0H*VHgO C, 82.58 : H, 11.72 Anhydrous |3 — spinasterol— p — Sp inast erol as isolated alovo was heated at 50° in vacuo for 7 days. The product was free of water of crystal­ lization; m.p. 148-150°. Anal. 2.232 mg. gave 2.447 mg. HOII; 6.904 mg. C02 C, 84.35 ; H, 12.12 Calcd. for C2 9 H4 7 0H C, 84.38 ; H, 11.73 This anhydrous product on recrystallization from 95$ ethyl alcohol gave the original product C2 9 H 4 7 0H*-?rH20 . p — Spinasteryl acetate— grams of One hundred milli­ (3 — spinasterol was dissolved in acetic anhydride and the mixture heated 1 hour. On stand­ ing several hours at room temperature the crystal­ line acetate separated. The product was filtered off and recrystallized from 95$ ethyl alcohol; 13- m.p. 153-155°; chloroform, 5.10° (44.7 mg., 2 cc. , = dcm., 2 /9 0^ D ~ n 0.228 , average reading). Anal. 2.332 mg. gave 2.292 mg. HOH; 6.998 mg. COg 2.235 mg. gave 2.290 mg. HOH; 6.683 mg. COg C, 81.83, 81.54 ; H, 10.92, 11.38 Calcd. for C3 1 H 5 Q02 C, 81.88 ; H, 11.097 |3-Spinasteryl benzoate— milligrams of Five hundred p — spinasterol in 1.5 cc. pyridine was treated with 0.5 cc. benzoyl chloride. The mixture was heated 2 hours on a boiling water bath, let stand over night, and then worked up as usual. The product was recrystallized from a mix­ ture of ethyl alcohol and ethyl ether. Yield 450mg.; m.p. 181-183°; chloroform, lf Q 0 =• 7.51° (56.0 mg., 2 cc. = 2 dcm., p :=• 0.421°, ave­ rage reading). Anal. 2.127 mg. gave 1.963 mg. HOH; 6.540 mg. COg C, 83.81 ; H, 10.25 Calcd. for C3 gH^gOg C, 83.65 ; H, 10.15 H 3''drogenation of p — spinasteryl acetate— Nine hundred and fifty milligrams of f3 — spin­ asteryl acetate was dissolved in 25 cc. glacial acetic acid and shaken in an atmosphere of hydrogen for 2 hours in the presence of 100 mg Adam’s cata­ lyst-*-4 . An additional portion of catalyst was added and the reaction continued about 2 hours -14- more* The reaction mixture was diluted with water and the product extracted with ethyl ether. The ether solution was washed with water, then with 1less soluble top frac­ tions gave a product; m.p. 142-143°. This substance, after repeated crystallization from methanol, gave a substance which appeared to be a chemical in­ dividual; m.p. 143-145°; 2 cc. chloroform, average reading). /i ~ zz 6.15° (49.6 mg., f9 2 dcm., CLq zz 0.305°, fo C J -16- Anal. S.466 mg. gave E.670 mg. HOH; 7.490 mg. COg C, 82.80 ; H, 12.03 Calcd. for C2 9 H 4 7 0 H-&H2 0 C, 82.58 ; H, 11.72 This substance forms an insoluble precipitate with digitonin and gives the comnon sterol color tests. S -Spinasteryl acetate— The sterol, m.p. 143-145°, was dissolved in a small amount of acetic anhydride and heated one hour. On standing at room temperature the product crystallized out. It was recrystallized from 95% ethyl alcohol; m.p. 132 Anal. 2.526 mg. gave 2.570 mg. HOH; 7.542 tvg> COg . C, 81.42 ; H, 11.30 Calcd. for C 3 1 H 5 0 0 g C, 81.88 ; H, 11.097 This acetate on hydrolysis gave the sterol; m.p. 143-145°. This acetate was also obtained from the mother liquors when crude iso-spinasteryl acetates were fractionated from 95% ethyl alcohol. S -Spinasteryl benzoate— Two hundred and ninety milligrams of the sterol was dissolved in 1 cc. pyridine and reacted with 0.5 cc. benzoyl chloride. The mixture was heated 2 hours on a boiling water bath and let stand over night. The product was recovered as usual, and recrystallized 4 times from ethyl alcohol; m.p. 165-168° (soften­ ing gradually to a viscous liquid) ^ - 11.17°; -17- (50.4 mg., 2 cc. chloroform, CL 4 9 rr JC - 2 dcm. y 0.563°, average reading). filial. 2.391 mg. gave 2.232 mg. HOH; 7.294 mg. C0? C, 83.20 ; H, 10.37 Calcd. for G3 6 H 5 2 ° 2 c » 8 5 «6 5 5 H f 10.15 Hydrolysis with 5% alcoholic potassium hydroxide gave the original sterol, m.p. 143-145°, and the sterol recovered from the benzoate was converted to the acetate; m.p. 131-134°. Hydrogenation of <£ -spinasteryl acetate— One hundred milligrams of was dissolved in 10 § -spinasteryl acetate cc. of glacial acetic acid and reacted with an atmosphere of hydrogen in the presence of 50 mg. of Adam*s catalyst. The product was recovered in the usual manner and recrystal­ lized twice from 95yo ethyl alcohol. Yield 40 mg.; m.p. 111-112°; JctJg4 rr 8.62° (44.5 mg., 2 cc. chloroform, rr 2 dcm., (Xp 4 ~ 0.384°, aver­ age reading). This product when mixed with an authentic specimen of Q b -stigmastenyl acetate showed no depression in melting point. Summary 1. From the unsaponifiable portion of Hardigan alfalfa seed oil three isomeric sterols of formula C g g ^ g O ^ H g O ; namely, (X.-spinasterol, -18- (3 -spinasterol and & -spinasterol have been isolated# 2. Several derivatives of each of the sterols have been prepared and the physical constants and analysis observed. 3# Each of the three isomers can be reduced to o c -stigmastenol. -19- II* Isolation of an Hydrocarbon and of a Fourth Alcohol from the Unsaponifiable of Hardigan Alfalfa Seed Oil* From the crude solid fractions of low melting point, obtained from the unsaponifiable fraction of Hardigan alfalfa seed oil (Fraction B), an hydrocarbon and a fourth alcohol were isolated* The hydrocarbon was separated by means of its low solubility in acetone after removal of most of the sterols* It was not characterized* The alcohol was separated as the acetate from acetone after removal of most of the hydro­ carbon. Analysis indicates as a probable formula C2 9 H 4 7 0H , isomeric with the three sterols men­ tioned in Section I. It is highly dextro rotatory. It reacts smoothly with halogens in such a way as to indicate one double bond* It did not precipi­ tate with digitonin and could not be epimerized with sodium amylate to give a compound which did precipitate with digitonin* It gave a slight pink color to the sulfuric acid layer in the LiebermanBurchard test* A summary of the separation procedure used to obtain this hydrocarbon and this alcohol is given in Fig. 3 . Crude Sterol Fractions from Hardigan Alfalfa Seed Oil, m.p. 120° or below Fractionate from 959& ethanol. soluble constituents sterols etc. sterols iinsoluble) Reflux v.-ith aoetio anhydride, cool. inso .u'uie Dissolve in a large vol\une of boiling acetone, cool. 1 solution flocculent precipi­ tate (hydrocarbon) m.p. 66 Concentrate, let stand. mix'.ure o± crystals ISeparate manually acetate c.. .n alcohol m.p. 235° sterol (probably /3-spinasterol) Fig. ? Scheme for isolation of an hydrocarbon and a high melting Sf — form, A — 5 *97° (51.1 mg., 2 cc. chloro- 2 dcm., 0Lo — 0.305°, average reading). Anal. 2.712 mg. gave 3.187 mg. CO2 ; 2.917 mg. HOH C, 82.33 ; H, 11.94 Calcd. for C29H470H*-^HgO C, 82.58; H, 11.72 -36- S -Spinasteryl acetate— A quantity of the above sterol was dissolved in acetic anhydride and the mixture heated* The acetate was recovered in the usual way and recrystallized from 95$ ethyl alcohol; m.p* 151-131.5°; 2 cc. chloroform, # Y (£ ] ' * — p — 2 dcm., 0.85° (50.7 mg., /if (X q 0.044°, average reading)• Anal. 2.169 mg. gave 2.254 mg. HOH; 6*502 mg. CO2 C 81.74 * H 11.54 Calcd. for CgiH^oOg* C* 81.88 ; H, 11.097 Reduction of S -spinasteryl acetate— Two hundred milligrams of the acetate was reduced with an atmosphere of hydrogen in the presence of Adam’s catalyst. The product was recovered as previously described and recrystallized repeatedly from ethyl alcohol. Yield 50 mg.; m.p. 111-112°; (43.8 mg., -2 cc. chloroform, 0L.q ~ 8*740 ;= 2 dcm., 0.383°, average reading). This acetate gave no depression in melting point when mixed with authentic QL -stigmastenyl acetate. Summary Three isomeric sterols, OL -spinasterol, spinasterol and S -spinasterol, p— have been isolated from the unsaponifiable of Grim alfalfa seed oil. -37- V* Isolation and Composition of a Phytosterolin from Grim Alfalfa Seed Oil, From Grim alfalfa seed oil a phytosterolin identical in all properties with the correspond­ ing compound found in Hardigan alfalfa seed oil has been obtained. The acetate of this sterolin prepared with acetic anhydride and nyridine has a saponification equivalent of about 200 . From this value, after making allowance for the sterol residue, and for the minimum requirements of the possible polyhydroxy alcohols, It appears that the molectilar weight of the acetate is approximately 800 . The sterolin is, therefore, a tetrahydroxy alcohol. This observation, when considered in con­ nection with the other properties noted, indicates that the sterolin is a glycoside of a hexose sugar Experimental The concentrated ether extracts of 28 kg. of Grim alfalfa seed gave 4.8 g. of crude phytosterol in. On recrystallization from n-amyl alcohol this substance gave a product which turned brown at 240° and decomposed at 260-270°. The product was slightly soluble in p-dioxane, ethyl ether, and n-amyl alcohol and easily soluble in hot acetic -38- anhydride or acetyl chloride. It was nearly in­ soluble in most other common solvents. Acetylation of the sterolin— To two grams of crude sterolin was added 30 cc. acetic anhy­ dride and 15 cc. pyridine. The mixture was let stand 24 hours at room temperature and then filt­ ered. The filtrate was decomposed with water and the preoipitated acetate filtered off. This residue was washed thoroughly with water and recrystallized from methyl alcohol. Yield 2 g.; m.p. 162-163°; 7.92° (45 mg., 2 cc. chloroform, 2 dcm., ~ 0.357°, average read­ ing). Saponification of the sterolin acetate— To 106 mg. of the above sterolin acetate weighed quantitatively into a 50 cc. Erlenmeyer flask, 20 cc. of ethyl alcohol containing about 100 mg. of potassium hydroxide was added. The mixture was refluxed 20 minutes and the excess potassium hydroxide titrated with standard acid to a phenolphthalein end point. An amount of potassium hy­ droxide equivalent to 5.5 cc. of 0.0952 N hydro­ chloric acid was used up in the reaction, 'Aie saponification equivalent calculated from this -39- data was 202, the percentage acetyl (H3 C-CO-) was 21.2 . The insoluble sterolin precipitated by the above saponification was collected quantitatively, washed with water, and with ethyl alcohol, and dried; yield 83.5 mg. The saponification equi­ valent calculated from this data was 197 . While the percentage acetyl (H^C-CO-) was 21.7 • This latter value was calculated by means of the formula 106 — 83.5 43 % acetyl = x— 106 x 100 42 The recovered sterolin turned brown at 240° and decomposed at 260-270°. The required values for the saponification number and percentage acetyl (H3 C-C0-) in the tetraacetate are 186 and 23.1$ respectively, as­ suming the formula of the sterol hexoside tetra­ acetate to be C4 3 H 0 6 Oqo • Hydrolysis ofthe sterolin— 3.9 g. A mixture of thissterolin dissolved boiling n-amyl alcohol and 200 of in 1.5liters of cc. of 15/S HC1 solution*1-5 , was refluxed 1 hour. The amyl alcohol was then removed by steam distillation, and the suspension extracted with ether. The ether extract was washed with sodium carbonate, then with water -40- and the solvent evaporated. The brownish residue was taken up in 95$ ethyl alcohol and the crude sterols allowed to crystallize. (X -Spinasterol— The crude sterols from hydrolysis of the sterolin were repeatedly re­ crystallized from 85$ ethyl alcohol and twice from methanol. Yield 243 mg, ; m,p. 164-165.5°; — - 2.56° (48.0 mg., 2 cc, chloroform, 2 dcm., OLq m - 0.125°, average read- ing). CL--Spinasteryl acetate— Seventy milli­ grams of the above sterol was reacted with acetic anhydride. The product was worked up as usual, and recrystallized from 95$ ethyl alcohol. Yield 2 cc, chloroform, ~ 2 dcm,, ca. q ~ - u.iyu f average reading)• Observation of reducing substance— The water layer obtained from the above hydrolysis was neu­ tralized with silver carbonate to a pH of 5 , The insoluble silver chloride was filtered off and the solution concentrated. The remaining aqueous sol­ ution reduced Benedict’s solution, A portion of the solution was treated with phenylhydrazine, acetic acid and the mixture heated. A greenish- -41- yellow crystalline osazone separated; m.p. about 197°. Summary 1. A phytosterolin identical with the correspond­ ing substance in Hardigan alfalfa seed oil, has been isolated from Grim alfalfa seed oil. 2. This substance had the properties of a tetrahydroxy alcohol and on acid hydrolysis gave a re­ ducing substance and a sterol mixture containing CL-spinasterol. 42- VT. The Sterols from the Unsaponifiable Fraction of Dutch White Clover Seed Oil. Several authors20»2^->SS»S3»24 have made ob­ servations on the oil obtained from clover flowers and plants. Rogerson mentioned the isolation of a phytosterol from Trifolium incarnatum while Finnemore et al. observed a similar substance in Trifolium repens. Fink and Richter2^ extracted the oil from Trifolium incarnatum seeds and determined its constants. As far as can be determined there is no other account of an investigation of clover seed oil. About 8$ of the weight of Dutch White clover seed may be extracted with ethyl ether as a clear yellow green oil and about 3.8$ of the oil is un­ saponif iable. The unsaponifiable portion consists of about 53$ of crude sterols and of about 67$ of reddish oily liquid. A portion of the sterol fraction obtained from the unsaponifiable of Dutch White clover seed oil after recrystallization from 85$ ethyl alcohol was acetylated and brominated. About 10$ of the weight of crude sterols precipitated on standing as an insoluble tetrabromide. This substance on 43- debromination and saponification gave stigmasterol. Experimental Fourteen kilograms of Dutch White clover seed was finely ground and extracted seven times with ethyl ether. The solvent was evaporated, giving 1075 g., or 7.6$ of a yellow green oil. The fat constants of this oil determined by the usual methods 32 were as follows: Iodine number 153.2 , saponification number 186.6 , unsaponifiable 3.83$. preparation of the unsaponifiable fraction— Fifteen hundred grams of Dutch White clover seed oil was saponified in 200 gram lots by dissolving in 1500 cc. ethyl ether and slowly adding to the vigorously stirred mixture, a solution of freshly prepared sodium ethylate (made by dissolving 40 g. of sodium in about 500 cc. 95$ ethyl alcohol). When the reaction was complete the ether was filter­ ed from the soaps in a closed vessel. The solid soaps were extracted several times with ether and finally dissolved in 2-3 liters of water and ex­ tracted repeatedly with ether. The combined ex­ tracts were washed free of alkali and the solvent distilled off. Yield 58 g. of a reddish oil. -44- Separation of sterols— Fift:^-eight grams of the above unsaponifiable fraction was dissolved in 600 cc, of ethyl ether and 5 crude sterol fractions separated out in a manner similar to that described for the separation of Hardigan alfalfa seed sterols. Yield 19 g.; m.p. 120-130°. The crude sterols were fractionated from 85$ ethyl alcohol. A top fraction weighing 15 g., m.p. 136-137°, was ob­ tained. Bromination of the sterols— Ten grams of the sterol fraction, m.p. 136-137°, was heated with acetic anhydride. The acetates obtained were dissolved in 100 cc. ethyl ether and treated with 7.0 g. of bromine in glacial acetic acid (5 g. bromine per 100 cc. glacial acetic acid) as re­ commended by Y/indaus28. After standing over night in the refrigerator, the insoluble tetrabromides were filtered off and washed with cold ether. Yield 1.07 g.; m.p. 192-196°, with browning. Stigraasteryl acetate— To 0.986 g. of the above tetrabromide, dissolved in 10 cc. glacial acetic acid, was added 1 g. of zinc dust, the mixture was refluxed 1 hour. The reaction mixture was filtered while hot, diluted with water and ex­ tracted with ethyl ether. The ether layer was washed with sodium carbonate and then with water 45- and the solvent evaporated. The residue was re­ crystallized from 95% ethyl alcohol. Yield 342 mg.; m.p. 137-139°; — chloroform, ~ - 58.4° (26.2 mg., 2 cc. 2 dcm., oL q ~ - 1.53°, ave­ rage reading). A mixed melting point of this sub­ stance with authentic stigmasteryl acetate gave no depression. Stigmasterol— A portion of the above acetate was saponified with alcoholic potassium hydroxide. The product was recrystallized from 95% ethyl alco­ hol and from methyl alcohol; m.p. 162-164°; 50.76° (47.5 mg., 2 cc. chloroform, r= 2 dcm., CLq — - 2.411 , average read­ ing). Summary 1. The oil has been extracted from Dutch White clover seed and some of its common constants determined. 2. From the unsaponifiable fraction of this oil stigmasterol has been isolated. 46- VII* Isolation and Composition of a Phytosterolin from Dutch White Clover Seed Oil* In 1910 Power and Salway^ and later RogerOR son00 isolated a high melting alcohol from clover. They named the substance Trifolianol and character­ ised it as a dihydroxy alcohol. In 1913 Power and Salway-*-^ recognized Trifolianol as a sterol gly­ coside and suggested the term phytosterolin for alcohols of this type. From the concentrated ether extracts of Dutch White clover seed a phytosterolin was isolated. This substance has properties which agree well with those given by Power and Salway29 for Trifolianol. On acid hydrolysis this substance gave a sterol mixture from which stigmasterol was obtained. Experimental Twelve kilograms of Dutch White clover were extensively extracted with ethyl ether and the ex­ tracts concentrated to about 75% oil. On stand­ ing, 6 g. of a slightly colored material separated. It was filtered off, dissolved in n-amyl alcohol and the hot solution filtered. On standing, a white gelatinous precipitate settled out. It was filtered off and the excess n-amyl alcohol 47- was removed by heating in vacuo; yield 4.82 g. It turned brown at 260° and decomposed at 280290°. This compound gave tests for carbon and hydrogen only. It gave positive Lieberman-Burchard and positive Molisch tests. Acetylation of the sterolin— About 1 g. of the above sterolin was heated with 50 cc. of acetic anhydride until the solid had all gone into sol­ ution. The product was recovered as usual, and recrystallized from methyl alcohol. Yield 1 g.; y m.p. 161-165°; chloroform, im — - 26.5° (51.0 mg., 2 cc. = 2 dcra., 3i / xr - 1.343°, average reading). Hydrolysis of the sterolin— Three grams of sterolin from Dutch White clover was suspended in 150 cc. ethanol containing 1 cc. of concentrated sulfuric acid and the mixture refluxed until com­ plete solution was effected. The reaction mixture was worked up in the usual way; yield 2 g. crude sterols. Bromination of sterols— The crude sterols weighing 2 g. were converted to the acetates. The acetates were dissolved in 40 cc. ethyl ether and treated with 2 g. of bromine in 40 cc. glacial acetic acid. The reaction mixture was placed in the -48- refrigerator over night and the crude tetrabromides filtered off. Yield 390 mg.; m.p. 180-185°. On recrystallization frora chloroform-ethyl alcohol the melting point improved to 190-195°. Stigmastervl acetate— above A portion of the tetrabromide fraction was debrominated in the usual v.ray and the product recrystallized from ethyl alcohol; m.p. 136-138°; (44 mg., 2 cc. chloroform, ^ = jjQtJ ~ - 54.6° ~ 2 dcm., - 2.406°, average reading). Stigmasterol— A portion of the above ace­ tate was saponified with alcoholic potassium hy­ droxide and the product recrystallized frora ethyl alcohol; m.p. 159-161°; 2 cc. chloroform, - — - 49.5° (34.5 mg., 2 dcm., OLq rr. - 1.711°, average reading). Detection of a reducing substance— The aqueous layer from the above hydrolysis was heated as recommended by Thornton et al. 17 . A portion of this solution when neutralized and heated with Fehling*s solution caused reduction of the copper solution. -49- Suimnary 1. A phytosterolin probably identical with Tri folianol 27 has been isolated from Dutch White clover seed oil. 2. This phytosterolin on acid hydrolysis gave reducing substance and a sterol mixture from which stigmasterol was isolated. -50- VIII. The Sterols from the Unsaponifiable Fraction of Medium Red Clover Seed Oil. Medium Red clover seed oil is very similar in properties to that obtained from Dutch White clover seed. From the unsaponifiable fraction of Medium Red clover seed oil, a quantity of crude sterols was separated. A fraction, m.p. 135-156°, obtained from these sterols by recrystallizing from 85$ ethyl alcohol, was acetylated and brominated. About 10$ of the weight of crude sterols separated on standing as an insoluble tetra bromide. This substance on debromination and saponification gave stigmasterol. Experimental Extraction of oil— Thirteen and eight tenths kilograms of Medium Red clover seed was finely ground and extracted eight times with ethyl ether. The solvent was evaporated; yield 1250 g. of a greenish-yellow oil. The common fat constants of this oil were as follows; unsaponifiable 5.41$, iodine number 129.4, saponification number 187.6 . Preparation of the Unsaponifiable fraction— Two kilograms of Medium Red clover seed oil was -51- saponified in 800 g. lots in a manner similar to that described in Section VI; yield 55 g. of a reddish oil. Separation of sterols— Sixty-five grams of the above unsaponifiable fraction was dissolved in 500 cc. of ethyl ether and 4 crude sterol fractions separated out in a manner similar to that described for the separation of Hardigan alfalfa seed oil sterols, Section I . Yield 88 g.; m.p. 130-156°. The crude sterols on fractionation from 35% ethyl alcohol gave a fraction weighing 10 g.; m.p. 155136°. Bromination of the sterols— One gram of the sterol fraction was acetylated using acetic an­ hydride; yield 0.9565 mg. The acetates obtained were dissolved in 80 cc. of ethyl ether and treat­ ed with 0.7 g. bromine in 15 cc. glacial acetic acid. The mixture was let stand over night in the refrigerator, and the insoluble tetrabromides filtered off and washed with ether; yield 96 mg. On recrystallization from chloroform-alcohol the product melted at 195° with browning. Stigmasteryl acetate— The tetrabromide above was debrominated and the product recovered -52- as in the case of the corresponding compound, Section VI. The product was recrystallized from 95$ ethyl alcohol; m.p. 138-140°; (37.0 mg., 2 cc. chloroform, — d p - 58.6° 2 dcm., - 2.170°, average reading). This pro- duct gave no depression in melting point when mixed with an authentic specimen of stigmasteryl acetate. Stigmasterol— The above acetate was hydro­ lyzed with alcoholic potassium hydroxide. The product was recovered in the usual way and re­ crystallized from ethyl alcohol; m.p. 160-162°; (38.5 mg., 2 cc. chloroform, — 2 dcm., o i0 = - 1.904°, average read­ ing). This substance gave no depression in melting point when mixed with authentic stigmasterol. Summary 1. The common fat constants of Medium Red clover seed oil have been determined. 2. From the unsaponifiable fraction of Medium Red clover seed oil, stigmasterol has been isolated. -53- IX. Isolation and Composition of a Phytosterolin from Medium Red Clover Seed Oil. From the concentrated ether extracts of Medium Red clover seed oil a phytosterolin has been isolated. This substance appears to be identical with Trifolianol 27 and with the corresponding compound isolated from Dutch White clover seed oil. Experimental Twenty-seven kilograms of Medium Red clover seed was repeatedly extracted with ethyl ether and the extracts concentrated to about 75$ oil. On standing 5.5 g. of an isoluble compound separated. This was filtered off and recrystallizeu from n-amyl alcohol. Yield 2.8 g.; m.p. 285-295°, with decomposition. This substance gives a positive Lieberman-Burchard and a positive Molisch test. Acetylation of the stero3.in- About 1 g. of the above sterolin was acetylated at room temper­ ature using acetic anhydride and pyridine. The product was recovered as usual and recrystallized from methanol; m.p. 162-163°; (50.5 mg., 2 cc. chloroform, a$ CL0 ~ ZZ - 28.5° z z 2 dcm., 1-444°, average reading). -54- Hydrolysis of the sterolin— About 1 g. of the Medium Red clover seed sterolin was suspended in 100 cc. of ethanol containing 1 cc. of con­ centrated sulfuric acid. The mixture was refluxed until complete solution was effected. The mixture was then diluted and the sterol fraction extracted with ether; yield 0.6 g. Broraination of sterols— The crude sterols weighing 0.6 g. were converted to the acetates. The crude acetates were dissolved in ether and treated with 0.6 g. bromine in 10 cc. of glacial acetic acid. The crude tetrabromides settled out on stand­ ing over night. Yield 50 mg.; m.p. 185-190° . The crude product after recrystallization from chloroform-eth:*l alcohol melted at 194-198°. This prodt^pt gave no depression in melting point when mixed with evade stigmasteryl acetate tetrabromide, prepared from soybean sterols. Detection of a reducing substance— The aqueous layer from the above hydrolysis was heated as recommended by Thornton et al.-1-7. A portion of the resultant solution reduced Rehling*s solution. - 55- Summary 1* A phytosterolin probably identical with Trifolianol*^ has been isolated from Medium Red clover seed oil. 2. This phytosterolin on acid hydrolysis gave a reducing substance and a sterol mixture. 3. Bromination of the sterol mixture gave a pro­ duct that appears to be stigmasteryl acetate tetrabromide. -56- X. Observations on the Reactions of* Sterols of the Spinasterol Type with Halogens, Several attempts have been made to prepare K IA halogen derivatives of Ct-spinasterol , all of them resulted in oily decomposition products or in unreacted starting substances, A preliminary study of the action of halogens on Ct-spinasterol, asterol and Ot -stigmastenol, spin­ & -spinasterol indicated that steroid compounds which have double bonds in ring "c " near or at positions 8 and 14- are extensively decom­ posed by reaction with halogens. This decomposition was so consistent that a green color, which devel­ oped, was used to detect sterols of the spinasterol type. It is probable that the irregular results obtained by Eck^° in his attempts to determine the number of double bonds in hydrocarbons related to cholestane, are in part due to this anomalous re­ action of halogens with unsaturation in positions 8 and 14. In a preliminary study it has also been ob­ served that compounds of the spinasterol type re­ act with perbenzoic acid in such a way that the amount of perbenzoic acid used depends on the excess reagent. It is possible that the anomalous -57- results of Fernholz and Moore® for OC -spinasterol and OC -stigmastenol can be explained on this basis. Experimental Reaction of 06* -spinasteryl acetate with bromine— To 0.1 g. of CC -spinasteryl acetate in 2 cc. ether was added 0.1 g. bromine in 2 cc. of glacial acetic acid and the mixture let stand at room temperature. After about 5 minutes the mix­ ture began to darken rapidly. The sane experiment was repeated except that the reaction was carried out with cold reagents and let stand in the ice box. After about 20 minutes a rapid darkening occurred. The same Phenomena occurred if CL -stigmastenyl acetate, (3 — spinasterol or 8 -spinasterol were used. Reaction of OC -spinasteryl acetate with Hanus iodine solution— About 5 mg. of CL -spinasteryl acetate was weighed quantitatively into a 10 cc. iodine flask, 0.2 cc. of chloroform, and a measured excess of Hanus iodine solution were added. After standing about 1 minute the mixture was observed to darken rapidly. It was let stand 30 minutes, treated with 2 cc. of saturated potassium iodide, diluted, and titrated with 0.0988 M sodium thio- -58- sulfate to a starch end point. This reaction was repeated several times with variation in the excess of Hanus iodine solution present during the reac­ tion. The chloroform layer of the above titration mixture had a characteristic dark green color. f3 — Spinasteryl acetate, $ CC -stigmastenyl acetate, -spinasteryl acetate and the free sterols as well react with Hanus iodine solution in the same manner. Data Data obtained for and CL -spinasteryl acetate OC -stigmastenyl acetate by means of the above prodecure are listed in Tables X and II . For purposes of illustration the data given in Table I is recalculated in Table III so that the milliequivalents for halogen added for each gram of sample nay be compared with the milliequivalents of halogen used by each gram of sample, and with the percentage excess halogen left at the end of the reaction. These values are plotted in Fig. 6 . Summary 1. From preliminary investigation it appears that sterols which are unsaturated in position 8 and 14 -59- react with halogens in such a way that the extent of the reaction is dependent on the excess of halogen present. S. A characteristic green color remaining in the chloroform layer after addition of Hanus iodine solution and discharging the excess iodine color, was used as a test for sterols of the spinasterol type. -60- Table I Data for the Reaction of CL -Spinasteryl Acetate with vari- ous Excess Amounts of Hanus Iodine Solution Weight of Sample mg, cc. Hanus solution added Titration cc. of 0.0988 M Na^S^Og 4.14 .750 .997 0.0 .750 4.176 .900 1.198 0.0 .900 1.862 5.0782 .500 .665 0.0 .500 1.050 5.148 .500 ,.660 0.0 .500 1.046 4.413 .500 0.0 .500 I No. Calculated 163.0 1.53 202.0 95.8 95.1 * 5.5105 0.0 .687 .652 103.5 91. 0 -61- Table II Data for the Reaction of 0L -Stigmastenyl Acetate with Vari­ ous Excess Amounts of Hanus Iodine Solution. V/eight of Sample mg, cc. Hanus solution added Titration cc. of 0.0988 M Na gS g0 g I No. Calculated 5.058 .3 .370 0.0 .3 .630 4.805 .5 *620 cm 1.037 5.070 .5 .650 0.0 .5 1.040 5.026 .5 .637 ini.8 5.0693 .3 .373 65.1 0.0 .3 .636 5.0305 .3 .372 0.0 .3 .636 0.0 65.2 105.0 97.6 65.8 -62- Table III Values for Oi -Spinasteryl Acetate, Calculated fron Table I These data are Plotted in Fig. 6 Weight of Sample mg. Milli­ equivalents of Halogen added per g. sample Milli­ equivalents of Halogen used per g. sample $ Halogen used $ Excesi Halogen left 4.14 .036 .012 33# 2 00$ 4.176 .044 .015 34$ 193$ 5.078 .0206 .0074 kjU/O rz arf 179$ 5.148 .0183 .0074 40# 147$ 4.413 .0236 .0081 34$ 192$ 5.51 .0189 .0071 37$ 166$ OF HsLoqeN U sed OF SAMpLe MiLLiequiVALeNTs Per G tam ru M/LLiequivALtNTs OF HALoqeN Added P e r Gtam OF SArrpte -64- XI* Discussion 1* A comparison of the Sterol Composition of Hardigan Alfalfa Seed Oil, Grim Alfalfa Seed Oil, Dutch White Clover Seed Oil and Medium Red Clover Seed Oil. The seed used in this work was purchased from the Michigan State Farm Bureau and under their guarantee had the characterstics indicated in Table IV. Certain fat constants of the oil obtained from each of these seeds by ethyl ether extraction are listed in Table V. A comparison of Sections I, III, IV and V shows clearly that Grim alfalfa seed and Hardigan alfalfa seed are Identical in steroid composition. To facilitate the comparison, the yields of the various fractions are tabulated in Tables VI and VIj . This complete correspondence in steroid composition is the basis of the introductory state­ ment, that the steroid composition is the same for different varieties within a single species. A comparison of Sections VI, VII, VIII and IX in­ dicates that Medium Red clover seed and Dutch White clover seed are identical in steroid com­ position. The yields of the various fractions are tabulated in Tables VII and VIII. The correspond­ -65- ence in steroid composition is complete as far as the work has progressed and forms the basis of the introductory statement, that the sterol composition is the same for different species within a single genus* The sterol composition of the seed of the genus TTedicago (alfalfa) is obviously quite differ­ ent from that of the genus Trifolium (clover). The former probably consists entirely of sterols of the spinasterol type, while the latter contains stigmasterol. -66- Table IV Hardigan Alfalfa C-rim Alfalfa Medium Clover Dut ch Wh it e clover Lot No. C9622 C9732 C9600 C9422 Purity 99.5$ 90.02$ 99.40$ 99.15^ .78$ .34$ .66$ .10$ .10$ .12$ .15$ .14$ .04$ Crop Seed Inert Weed Seed Where grown. When grown .2$ .2$ .1$ Michigan Michigan Michigan 1939 1939 1939 Wisconsin 1939 Table V Hardigan Alfalfa Grim Alfalfa Medium Clover Dutch ’White Clover $ Ether Extract 10.6 10.2 S.9 7.7 Iodine Humber 172.15 172.0 129.4 153.2 Saponification Number 187.9 1S7.2 187.6 186.6 Unsaponifiable 4.32$ 3.41$ 3.83$ -67- Table VI A Summary of Yields of the Various Fractions and Compounds Isolated from 3.75 kg. of Hardigan Alfalfa Seed Oil and from 2.67 kg. of Grim Alfalfa Seed Oil. Fraction or Compound Yields Hardigan Alfalfa Seed Oil Total unsaponifiable Grim Alfalfa Seed Oil 150 g. 92 g. Total crude sterols 47 g. 35 g. OL -Sninasterol isolated 11.1 g. 5.8 g. /3— Spinasterol isolated 18.7 g. 9.4 g. b -Spinasterol isolated 2.9 g. 1.5 g. Yield of unsaponifiable material 4.0$ 5.44$ Yield of crude sterols on the basis of oil saponified 1.2$ 1.2$ 31.0$ 35.0$ 0.29$ 0.2$ on the basis of total unsaponifiable 7.4$ 6.3$ on the basis of total crude sterols 23.0$ 17.0$ on the basis of total unsaponifiable Yield of & -spinasterol on the basis of oil saponified -68- Table VI (cont’d) Fraction or Compound Yield of f?— spinasterol on the basis of oil saponified Yields Hardigan Alfalfa Seed Oil Grim Alfalfa Seed Oil 0.5$ 0.55$ on the basis of total unsaponifiable 12.0$ 10.0$ on the basis of total crude sterols 59.0$ 28.0$ Yield cf & -spinasterol on the basis of oil saponified 0.07$ 0.05$ on the basis of total unsaponifiable 1.8$ 1.6$ on the basis of total crude sterols 6.1$ 4.5$ -69- Table VII Summary of the Yield of Phytosterolin from the Various Seeds Hardigan Alfalfa Amount of seed extracted Grim Alfalfa Medium Red Clover Dutch White clover £6 kg. £8 kg. £7 kg. 14 kg. 4.9g. 5.8 g « 2.79 g. 4.8£ g. I\ r\n a rtf u •yjo*±G /o .Amount of phyto­ sterolin ob­ tained Yield of phyto­ sterolin on the basis of seed 0.018A Yield of phyto­ sterolin on the basis of oil. 0.17% r\ r\r>rtf 0.19% rf \i % '- U./0 0.113% 0.44% -70- Table VIII A Summary of the Various Fractions and Compounds Isolated from 1500 g, of Dutch White Clover Seed Oil, and from 2000 g. of Medium Red Clover Seed Oil. Fraction or Compound Yields Dutch White Clover Seed Oil Medium Red Clover Seed Oil Total unsaponifiable 5S g. 65 g. Total crude sterols 19 g. 28 g. Sterols brominated 10 g. Total crude tetrabromides 1.07 g. 1.0 g. 0.096 g. Yield of unsaponi­ fiable 3.86% 3.25% Yield of crude sterols based on oil saponi­ fied 1.2% 1.4% based on total unsap­ onifiable 33.0% 43.0% Yield of crude stigmasteryl acetate tetrabromide based on sterols brominated 10.7% 9.09% -71- 2. A Consideration of the Structure and Homogeneity of Sterolc of the Spinasterol Type* In this work the isolation of a new sterol which was called S -spinasterol was described. It should be noted that the ordinary criteria of purity, such as constancy of melting point, re­ covery of original products from derivatives and so forth, are not completely reliable in the steroid field. The tendency of these compounds to form mixed crystals of constant properties makes the recognition of a chemical individual very uncertain. Recently Bernstein, Kauzmann and Wallis®**' have formulated a working method for calculating the specific rotation of steroids from the struc­ tures assigned. The extension and revision of the idea will probably provide an excellent criterion for the establishment of the structure and^einheitlichkeit*of members of the steroid group. In connection with this, it is of interest to note that the formula for a -spinasterol I, proposed by Fernholz and Ruigh-*-*** cannot be correlated with the specific rotation calculated according to Bernstein et al.®**-. The calculated rotation using the constants of these authors is: -72- ED = _ rnoL lOeyht “ C -tMP&iy+stid Dm ;a m o L UeifhC or W O + C -M ty + C -H r tO ) & _ 9.J U-tX whereas the observed value is - 2.7°* CHj tf.c ^ cH - c HJ c h - c% c. H2 C«j I " U On the other hand the specific rotations of p — spinasterol, 5.91°, and of $ -spinasterol, 6.15°, are in fair agreorient with the value calculated for I . The bessisterol isolated by Kuwada and Yosiki agrees closely in properties7 with CL-spinasterol except that it has a specific rotation of - 13°. Nov; the ■'alculated rotation of the structure II is: O -73- Mo C+M: Dnit+Sti’9 Dm.23 Ytko L UJtif AC Vrvol- k/e)yAt or < 1 /1 ,0 + 6 m o h c - k & o ) 1+ 13. This value is in fair agreement with the observed value for bessisterol. Tt seems possible that bessisterol in II and either ^ spinasterol or £ -spinasterol is I , and that CL -spinasterol as ordinarily isolated is a mixture of I and II . However all attempts in this laboratory to separate CL -spinasterol into two such components failed. -74- B Bibliography 1* Sobotka, Chemistry of the Sterids, page 86, Willians and Wilkins, Baltimore (1958). 2. Heyl, Wise and Speer, J. Biol. Chem. 82, 111 (1929). 3. Simpson, J. Chem. Soc. 730 (1937). 4. Dam, Geiger, Glavind, Karrer, Karrer, Roths­ child and Solomon, Helv. Chim. Acta. 2 2 , 510 (1939). 5. King and Ball, J. Am. Chem. Soc. 63L, 2910 (1939). 6. Fernholz and Moore, J. Am. Chem. Soc. 61, 2467 (1939). 7. Ifuwada and Yosiki, J. Pharm. Soc. (Japan) 161 (1940). 60, 8. Kuwada and Yosiki, J. Pharm. Soc. (Japan) 57, 155 (1937). 9. Kuv/ada and Yosiki, J. Pharm. Soc. (Japan) 59, 282 (1939). 10. Larsen and Heyl, J. Am. Chem. Soc. 5(3, 2663 (1934). 11. Fernholz and Ruigh, J. .Am. Chem. Soc. 62, 2341 (1940). 12. Heyl and Larsen, J. Pharm. Assoc. 22, 510 (1933). 13. Hart and Heyl, J. Biol. Chem. 95, 311 (1932). 14. Gilman and Blatt, Organic Synthesis, Collect­ ive Volume I . 2nd e a , f page 463, John Wiley and Sons, New York. 15. Power and Salway, J. Chem. Soc. 103, 399 (1913). 16. Jantzen and Gohdes, Biochem. Z. 272, 166 (1934). -75- Bibliography (contfd) 17. Thornton, Kraybill and Mitchell, J, Am* Chem. Soc. 62, 2006 (1940). 18. Thornton, Kraybill and Broome, J. Am. Chem. SOC. 63, 2079 (1941). 19. Moore and Link, J. Biol. Chem. 153. 293 (1940). 20. Mitumaru Tuzimoto, J. Soc. Chem. Ind., (Japan) 43, suppl. binding 208 (1940). C. A.. 54. 7974 (1940). 21. Guido Colombo, Boll, ufficiale staz. sner. seta 4, 57 (1934). C. A. 29, 1849 (1935). 22. Kirk, J. Am. Soc. Agron. JL8, 385 (1926). 23. Dunbar and Wells, J. Oil and Bat Ind. 382 (1926). 24. Stoeckert, Oel und Fettind (1920) 61. 15. 2991 (1921). 25. Rogerson, J. Chem. Soc. _97, 1004 (1910). 26. Finnemore, Cooper, Stanley, Cobcroft and Harris, J. Soc. Chem. Ind. _57, 162 (1938). 27. Fink and Richter, Casopis <5eskoslov Lekamictva 17, 262 (1937). C. A. 32, 1501 (1938). 28. Windaus and Hauth, Ber. 3£, 4378 (1906). 29. Power and Salway, J. Chem. Soc. _97, 231 (1910). 30. Eck and Hollingsworth, J. Am. Chem. Soc. 64. 140 (1942). C. A. 31. Bernstein, Kauzmann and Wallis, J. Organic Chem. 6 , 319 (1941). 32. Official and Tentative Methods of Analysis of the Association of Official Agricultural Chemists. 4-th ed.,1955. Association of Official Agricultural Chemists. Washington D.C.