gossyverdurin: a newly isolated pigment from cottonseed pigment glands

5
OCTOBER, 1963 COLEM ~ AN : ~ANCREATIC HYDROLYSIS OF NATURAL ]~ATS 571 positions of the triglyeerides. For some fats at least, the free fatty acids do not, therefore, provide a satis- factory measure of the composition of the acids occu- pying the 1 and 3 positions of the triglycerides. It must of course be remembered when using pan- creatic lipase for the investigation of the fatty acid distribution of natural fats, that rates of hydrolysis for individual fatty acids are only comparable for the higher members of the series. Savary and Desnuelle (20) have reported that similar rates are obtained for lauric, and acids of greater molecular weights, whether saturated or unsaturated. Entressangles etal. (21) and Clement et al. (22) have clearly shown that the presence of short chain acids, particularly butyric, obscure the results of fatty acid distribution deduced from hydrolysis data. Entressangles has drawn atten- tion (23) to the unsuitability of pancreatic hydrolysis for the investigation of such fats as milk fat, which contain appreciable anmunts of butyric acid. ACKNOWLEDGMENTS Assistance in preparing this paper by F. R. J~eobsberg and B. R. W. Pinsent. G~s-chromatographic analyses by Mrs. J. Clark a~d J. K. Messenger. REFERENCES 1. Savary, P., J. Flanzy, and P. Desnuelle, Biochim. Biophys. Acta., 24, 414 (1957). 2. Mattson, F. H., and E. S. Lutton, J. Biol. Chem., 233, 868 (1958). 3. Desnuelle, P., and P. Savary, Fotte-Seifen Anstrichmittel, 61, 871 (1959). 4. Mattson, F. H., and R. A. Volpenheim, J. Biol. Chem., 236, 189t (1961). 5. Youngs, C. G., JAOCS, 86, 665 (1959). 6. P~eiser, R., and H. G. R. R,eddy, Ibld., 86, 97 (1959). 7. CIoleman, M. H., and W. C. Fulton, in "The Enzymes of Lipid Metabolism," ed. P. Desnuelle, Pergamon Press, Oxford, p. 127, 1961. 8. Coleman, h~. It., JAOCS, 88, 685 (1961). 9. Ast, H. J., and R. J. VanderWal, Ibid., 88, 67 (1961). 10. Savary, P., and P. Desnuelle, Compt. Bend., 240, 2571 (1955). 15. As Ref. 7, p. 135. 12. Hilditeh, T. P., "The Chemical Constitution of Natural Fats," 3rd Ed., Chapman & Hall, London, pp. 357 & 238, 1956. 13. Quinlin, Patricia, and It. J. Weiser, JAOCS, 35, 325 (1958). 14. James, A. T., J. Chromatography, 2, 552 (1959). 15. Cbleman, M. It., Biochim. Biophys, Acta., 67, 146 (1963). 16. Laidler, K. J., "The Chemical Kinetics of Enzyme Action," Ox- ford University Press, Oxford, p. 40, 1958. 17. Constantin, ~r J., L. Pasero, and P. Desnuelle, Biochim. Bio- phys. Acta., 48, 103 (1960). 18. VanderWal, 1~. J., JAOCS, 37, 18 (1960). 19. ~d[attson, F. H., and R. A. Volpenheim, J. Lipid Res., 2, 58 (1961). 20. Savary, P., and P. Desnuelle, Arch. Sci. Biol., 89, 689 (1955). 21. Entressangles, B., L. Pasero, P. Savary, P. Desnuelle, and L. Sarda, Bull. Soc. Chim. Biol., 48, 583 (1961). 22. Clement, G., J. Clement, and J. Bezard, Biochem. Biophys. Res. Commun., 8, No. 3, 238 (1962). 23. As Ref. 7, p. 31. [Received January 7, 1963--Accepted April 3, 1963] Gossyverdurin A Newly Isolated Cottonseed Pigment Glands Pigment from C. M. LYMAN, A. S. EL-NOCKRASHY, and J. W. DOLLAHITE, The Departments of Biochemistry and Nutrition and Veterinary Medicine, A. & M. College of Texas, College Station, Texas Abstract The constituents of cottonseed pigment glands were fraetionated by the use of column chroma- tography with DEAE cellulose ion exchanger and silicie acid, and a new green pigment was isolated. The acute oral toxicity of the new pigment was determined using rats as experimental animals. The LD-50 value obtained was 0.66 g/kg of body weight indicating that the new pigment which was named gossyverdurin is the most toxic of any cottonseed pigment so far reported. Gossyver- durin showed absorption maxima at 250, 370, and 560 m~. Reaction with para-anisidine under the conditions used for ti~e determination of gossypoi gave an absorption peak similar to that obtained with gossypol indicating that the new compound is structurally related to gossypol. In addition a second peak at 342 mt~ appears on reaction with para-anisidine indicating important structural differences between gossypol and gossyverdurin. Introduction I T HAS BEEN reported that cottonseed pigment glands, which are separated in an essentially unaltered con- dition from cottonseed kernels by a flotation process (1,2) retard the growth of chicks when the glands or products containing them are included in the diet (3,4). 0ral administration of these glands in rela- tively large doses resulted in the death of rats, mice, guinea pigs, and rabbits (5). Gossypol, a polyphenolic yellow pigment, is the prin- cipal component and gossypurpurin, a gossypol deriva- tive, has been reported to be the most abundant sec- ondary component of cottonseed pigment glands (6). Eagle et al. (5) studied the relative toxicity of pure gossypol and a number of preparations of pigment glands using rats as experimental animals and re- ported that the glands were more toxic than an equiv- alent amount of gossypol, that the toxicity of different gland preparations was not proportional to their con- tent of gossypol, and that toxicity decreased with an increasing content of gossypurpurin. They concluded: "The toxicity of cottonseed pigment glands is at- tributable to some component or components of the glands other than, or in addition to, gossypol and gossypurpurin." In these studies an acetone-soluble, water-soluble fraction was obtained from the glands which had an LD-50 value of approximately 700 rag/ kg body weight, using rats as experimental animals. Recently the reports of Eagle et al. to the effect that cottonseed pigment glands are more toxic than an equivalent amount of gossypol have been confirmed by E1-Nockrashy, Lyman, and Dollahite (7). The purpose of the present communication is'to report the isolation of a bright green colored pigment from cottonseed glands which is more toxic than gossypol. This new pigment, which is a gossypol related compound, has been named gossyverdurin. Experimental Fractionation of Cottonseed Pigment Glands. Cot- tonseed pigment glands were separated from rolled deeorticated kernels by a flotation technique (8). Twenty-five grams of the pigment glands were ex- tracted with 200 ml of acetone in a Waring Blendor for 5 min, and the mixture was filtered through a Buehner funnel. The residue was re-extracted with 200 ml, and finally with 50 ml of aeetone. The corn-

Upload: c-m-lyman

Post on 19-Aug-2016

249 views

Category:

Documents


20 download

TRANSCRIPT

Page 1: Gossyverdurin: A newly isolated pigment from cottonseed pigment glands

OCTOBER, 1 9 6 3 C O L E M ~ AN : ~ A N C R E A T I C H Y D R O L Y S I S OF N A T U R A L ]~ATS 571

positions of the triglyeerides. For some fats at least, the free fat ty acids do not, therefore, provide a satis- factory measure of the composition of the acids occu- pying the 1 and 3 positions of the triglycerides.

It must of course be remembered when using pan- creatic lipase for the investigation of the fat ty acid distribution of natural fats, that rates of hydrolysis for individual fat ty acids are only comparable for the higher members of the series. Savary and Desnuelle (20) have reported that similar rates are obtained for lauric, and acids of greater molecular weights, whether saturated or unsaturated. Entressangles etal . (21) and Clement et al. (22) have clearly shown that the presence of short chain acids, particularly butyric, obscure the results of fat ty acid distribution deduced from hydrolysis data. Entressangles has drawn atten- tion (23) to the unsuitability of pancreatic hydrolysis for the investigation of such fats as milk fat, which contain appreciable anmunts of butyric acid.

ACKNOWLEDGMENTS

Assistance in preparing this paper by F. R. J~eobsberg and B. R. W. Pinsent. G~s-chromatographic analyses by Mrs. J. Clark a~d J. K. Messenger.

REFERENCES

1. Savary, P., J. Flanzy, and P. Desnuelle, Biochim. Biophys. Acta., 24, 414 (1957).

2. Mattson, F. H., and E. S. Lutton, J. Biol. Chem., 233, 868 (1958). 3. Desnuelle, P., and P. Savary, Fotte-Seifen Anstrichmittel, 61, 871

(1959). 4. Mattson, F. H., and R. A. Volpenheim, J. Biol. Chem., 236, 189t

(1961). 5. Youngs, C. G., JAOCS, 86, 665 (1959). 6. P~eiser, R., and H. G. R. R, eddy, Ibld., 86, 97 (1959). 7. CIoleman, M. H., and W. C. Fulton, in "The Enzymes of Lipid

Metabolism," ed. P. Desnuelle, Pergamon Press, Oxford, p. 127, 1961. 8. Coleman, h~. It., JAOCS, 88, 685 (1961). 9. Ast, H. J., and R. J. VanderWal, Ibid., 88, 67 (1961). 10. Savary, P., and P. Desnuelle, Compt. Bend., 240, 2571 (1955). 15. As Ref. 7, p. 135. 12. Hilditeh, T. P., "The Chemical Constitution of Natural Fats," 3rd

Ed., Chapman & Hall, London, pp. 357 & 238, 1956. 13. Quinlin, Patricia, and I t . J. Weiser, JAOCS, 35, 325 (1958). 14. James, A. T., J. Chromatography, 2, 552 (1959). 15. Cbleman, M. It., Biochim. Biophys, Acta., 67, 146 (1963). 16. Laidler, K. J., "The Chemical Kinetics of Enzyme Action," Ox-

ford University Press, Oxford, p. 40, 1958. 17. Constantin, ~r J., L. Pasero, and P. Desnuelle, Biochim. Bio-

phys. Acta., 48, 103 (1960). 18. VanderWal, 1~. J., JAOCS, 37, 18 (1960). 19. ~d[attson, F. H., and R. A. Volpenheim, J. Lipid Res., 2, 58

(1961). 20. Savary, P., and P. Desnuelle, Arch. Sci. Biol., 89, 689 (1955). 21. Entressangles, B., L. Pasero, P. Savary, P. Desnuelle, and

L. Sarda, Bull. Soc. Chim. Biol., 48, 583 (1961). 22. Clement, G., J. Clement, and J. Bezard, Biochem. Biophys. Res.

Commun., 8, No. 3, 238 (1962). 23. As Ref. 7, p. 31.

[ R e c e i v e d J a n u a r y 7, 1 9 6 3 - - A c c e p t e d A p r i l 3, 1963]

Gossyverdurin A Newly Isolated Cottonseed Pigment Glands

Pigment from

C. M. LYMAN, A. S. EL-NOCKRASHY, and J. W. DOLLAHITE, The Departments of Biochemistry and Nutrition and Veterinary Medicine, A. & M. College of Texas, College Station, Texas

Abstract

The constituents of cottonseed pigment glands were fraetionated by the use of column chroma- tography with DEAE cellulose ion exchanger and silicie acid, and a new green pigment was isolated. The acute oral toxicity of the new pigment was determined using rats as experimental animals. The LD-50 value obtained was 0.66 g/kg of body weight indicating that the new pigment which was named gossyverdurin is the most toxic of any cottonseed pigment so far reported. Gossyver- durin showed absorption maxima at 250, 370, and 560 m~. Reaction with para-anisidine under the conditions used for ti~e determination of gossypoi gave an absorption peak similar to that obtained with gossypol indicating that the new compound is structurally related to gossypol. In addition a second peak at 342 mt~ appears on reaction with para-anisidine indicating important structural differences between gossypol and gossyverdurin.

Introduction

I T HAS BEEN reported that cottonseed pigment glands, which are separated in an essentially unaltered con-

dition from cottonseed kernels by a flotation process (1,2) retard the growth of chicks when the glands or products containing them are included in the diet (3,4). 0ral administration of these glands in rela- tively large doses resulted in the death of rats, mice, guinea pigs, and rabbits (5).

Gossypol, a polyphenolic yellow pigment, is the prin- cipal component and gossypurpurin, a gossypol deriva- tive, has been reported to be the most abundant sec- ondary component of cottonseed pigment glands (6).

Eagle et al. (5) studied the relative toxicity of pure gossypol and a number of preparations of pigment glands using rats as experimental animals and re- ported that the glands were more toxic than an equiv- alent amount of gossypol, that the toxicity of different gland preparations was not proportional to their con- tent of gossypol, and that toxicity decreased with an increasing content of gossypurpurin. They concluded: "The toxicity of cottonseed pigment glands is at- tributable to some component or components of the glands other than, or in addition to, gossypol and gossypurpurin." In these studies an acetone-soluble, water-soluble fraction was obtained from the glands which had an LD-50 value of approximately 700 rag/ kg body weight, using rats as experimental animals.

Recently the reports of Eagle et al. to the effect that cottonseed pigment glands are more toxic than an equivalent amount of gossypol have been confirmed by E1-Nockrashy, Lyman, and Dollahite (7). The purpose of the present communication is'to report the isolation of a bright green colored pigment from cottonseed glands which is more toxic than gossypol. This new pigment, which is a gossypol related compound, has been named gossyverdurin.

Experimental Fractionation of Cottonseed Pigment Glands. Cot-

tonseed pigment glands were separated from rolled deeorticated kernels by a flotation technique (8). Twenty-five grams of the pigment glands were ex- tracted with 200 ml of acetone in a Waring Blendor for 5 min, and the mixture was filtered through a Buehner funnel. The residue was re-extracted with 200 ml, and finally with 50 ml of aeetone. The corn-

Page 2: Gossyverdurin: A newly isolated pigment from cottonseed pigment glands

572 T H E ~ O U R N A L O F T H E A ~ I E R I C A N O I L C H E M I S T S ' ~ O C I E T Y VoL. 40

A c e t o n e S o l u b l e $

W a t e r E x t r a c t i o n I

P I G M E N T G L A N D S

l A c e t o n e E x t r a c t i o n

l

W a t e r E x t r a c t ( C o l l o i d a l )

DEAE Co lumn

t A c e t o n e Insoluble

Resi~due

t W a t e r E l u e t e

t 6 0 % A c e t o n e E l u o t e

t 7 5 % A c e t o n e E l u a t e

t 1 0 0 % A c e t o n e E l u a t e

t Ch loro form E l u a t e

+ 15% Acet ic Ac id + Chloroform t

t Acet ic Acid Chloroform Phase

S I L I C I C A C I D [ Chloro form E l u a t e (Almost ent i re ly t ree (Jossypol}

Absotute Methano l E luote

Recrysta l l~zot ion

i G O S S Y V E R D U R I N

(2 ~ of weight of glands 1

FIG. 1. Diagram of fractionatlon of cottonsecd pigment glands and isolation of gossyverdurin.

bined extracts were evaporated by passing a s t ream of ni trogen over the surface. The deep red viscous acetone-soluble residue was then re-extracted with dis- tilled water in a War i ng Blendor unt i l the extract became clear. The combined water extract (500 ml) which was yellow, and appeared to be a colloidal sus- pension, was then passed over a column of' D E A E cellulose ion exchanger 1 in. diam and 12 in. long, and the column was washed with water unti l a clear efflu- ent was obtained. The combined water effluent was lyophilized and a pale yellow sticky mater ial was ob- tained (ca. 0.40 g) . The D E A E column was then eluted using the following eluants in succession: 60% acetone, 75% acetone, 100% acetone, and chloroform. The eluted f rac t ions were collected, and evaporated under vacuum on a warm water bath (40C). The 60% and 75% acetone cluates (0.60 and 0.85 g, respec- t ively) were yellowish brown oily solids. The 100'% acetone and the chloroform eluates (1.58 and 0.45 g, respectively) were brown oily liquids. The D E A E was then t r ans fe r red to a separa tory funnel with two volumes of 15% acetic acid solution, and two volumes of chloroform and the mixture was shaken vigorously. This t rea tment resulted in complete elution of all col- ored mater ia l adsorbed on the D E A E . The eluted ma- terial was found in the chloroform layer. The D E A E was washed twice with chloroform, and the chloroform eluate and wash was then concentrated to 100 ml by evaporat ion under vacuum using a wa rm water bath.

The chloroform eluate, yellowish green in e01or, was then passed over a silieie acid column, 1 in. diam and 12 in. long. Two bands separated on the column, a green band at the top and a yellow band below. The yellow band, which proved to be gossypol, rap id ly passed through the column and was collected in the chloroform eluatc. The column was then exhaustively washed with chloroform. The chloroform eluate and

T A B L E I C h a r a c t e r i z a t i o n of F r a c t i o n s f r o m Cottonseed P i g m e n t Glands

% of origi- na l p i g m e n t T o t a l F r e e

D e s c r i p t i o n o~ f r a c t i o n g l ands con- gossypol gossypol t a i ned in con ten t a c o . t e n t a

the f r a c t i on

O r i g i n a l p i g m e n t g lands .. . . . . . . . . . . . . . . Acetone-soluble f r ac t ion .. . . . . . . . . . . . . . . W a t e r ex t r ac t of acetone

soluble f r a c t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . E l u a t e s f rom D E A E column

W a t e r effluent . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 0 % Ace tone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 5 % Acetone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 0 % Ace tone . . . . . . . . . . . . . . . . . . . . . . . . . . . Ch lo ro fo rm .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

E l u a t e s f rom si l icic acid column Chloroform .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ :e thano i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 0 0 . 0 0 60 .40

4 0 . 3 0

1 .20 2 .40 3 .40 6 .70 1 .40

% 4 3 . 0 0 7 0 . 1 0

60 .00

0 .10 0 .35 0 .23 O.09 0 .11

9 2 . 0 1 3 2 . 5 0

% 38 .10 64 .70

43 .20

O,OO 0.25 0 .15 0 .04 0 .00

88 .39 2 5 . 0 0

a As d e t e r m i n e d co lo r lmet r i ea l ly a f t e r r eac t ion w i t h p a r a a n i s i d i n e ( 9 ) .

washing was evaporated under vacuum at low tem- pera ture (40C). The green colored band was then eluted readi ly and completely with absolute methanol. Evapora t ion of the methanol eluate at 40C, under vac- uum, lef t a dark green crystalline m a t e r i a l The green colored p igment was recrystall ized f rom absolute meth- anol solution, and then dried in a vacuum desiccator containing phosphorous pentoxide. A d iagram of the fraet ionat ion procedure is given in F igure 1.

Characterization of the Fractionation Products. The various fract ions were analyzed for free and total gos- sypol using the method of Pons et al. (9,10). Phos- phorous was determined by the method of Fiske and SubbaRow (11). The absorption spectra of the green pigment was obtained by the use of the Beckman D.U. Spectrophotometer.

Physiological Studies. Rats were used as experi- mental animals for the s tudy of the physiological sig- nificance of the isolated fractions. The care of the ani- mals, the method of adminis ter ing the doses, and the procedures for evaluat ing the results have been de- scribed previously (8). LD-50 values were calculated according to the method of Bliss ( ]2) .

Results and Discuss ion Fractionation of Pigment Glands. Tile percentages

by weight of the original p igment glands obtained in the various fractions and the content of free and total gossypol as determined by the colorimetrie procedure af ter reaction with para-anisidine are given in Table I.

The acetone-soluble fract ion showed high total gossy- pol content mostly as free gossypol. The five fractions collected f rom the D E A E column constituted 15.1% of the p igment glands. These five fract ions were oily or waxy in texture. I t was thought that these might be phospholipid-bound gossypol, but analysis did not show any phosphorous, and the peculiar thing is that the total gossypol content of the five fractions was very small, ranging f rom 0.10-0.35%. The chloroform eluate, a yellowish green solid obtained by t reat ing the D E A E with acetic acid and chloroform, consti- tuted 23.5% of the original p igment glands weight. The chloroform effluent collected f rom the silieie acid column, a br ight yellow, showed as high as 92.01% total gossypol, almost all as free gossypol. The green colored pigment, gossyverdurin, on analysis by the s tandard methods for the determinat ion of free and total gossypol, showed an apparen t content of 25.00'% free gossypol and 32.50% total gossypol. The yield of gossyverdurin was ca. 2.0% of the weight of the origi- nal p igment glands.

Physical and Chemical Characteristics of Gossyver- durin. Gossyverdurin is ve ry soluble in chloroform,

Page 3: Gossyverdurin: A newly isolated pigment from cottonseed pigment glands

OCTOBER, 1963 LYMAN ET AL.: GOSSYVERDURIN: NEWLY ISOLATED PIGMENT 573

W L~ Z < rn n-

O 09 nn <

1.0

0.8

0.6

0.4

0.2

I

IN C H L O R O F O R M

�9 IN 70~o A C E T O N E

A F T E R A D D I N G O X A L I C A C I D A N D H E A T T R E A T - M E N T

3 0 0 4 0 0 500 600

WAVE LENGTH mp :Fie. 2. Absorption spectra of gossyverdurin, concn 2 rag/

100 ml.

methanol, acetone, diethylether, and ethanol, and in- soluble in petroleum ether. I t became brown in color at 210C indicating decomposition, but did not show any melting point when heated to a temperature as high as 310C.

Microana]ytical results of the pigment showed: C, 62.93; H, 6.19; N, ].90; O, 21.09; ash, 8.20%.

Several qualitative tests (13) were applied to gossy- verdurin, in an effort to determine to what extent it was related to gossypol and gossypurpurin. Positive reactions were obtained with the same reagents which were found to give positive results with gossypol, viz., those indicating the presence of one or more carbonyl groups such as Tollen's reagent and Fehl ing 's solu- tion, and the presence of two ortho-phenolic hydrox- yls, and a hydroxy para or ortho to a carbonyl (using ferric chloride). Concentrated sulphuric acid reacts with the green pigment to produce a permanent green color, in contrast to the dark red color developed im- mediately with gossypol, and the yellow-green color which becomes orange af ter ten minutes with gos- sypurpurin. A deep blue color is obtained when the green pigment is t reated with glacial acetic acid, in contrast to the yellow precipitate obtained with gossypo], and the green precipitate obtained with gossypurpurin.

The absorption spectrum of gessyverdurin in chlo- roform solution has maxima at 250, 370, and 560 m~ (Fig. 2). Gossypol in chloroform has maxima at 288 and 363 m/~ (Fig. 3).

Considerable difference in the spectra of gossypol and gossyverdurin in chloroform at the lower wave lengths will be noted. While the sharp decrease in the case of gossyverdurin, below 250 m/~ might sug- gest the absence of a maxima at 236 m/~ and hence the absence of a diene conjugated system, chloroform is not the ideal solvent for examination in this range of the spectrum; fur ther work would be required to establish such a conclusion.

The shift in the absorption band of gossypol at 363 m/~ in chloroform to somewhat longer wave lengths in acetone (Fig. 3) can be accounted for, we believe, by the formation of the gossypol acetone complex. This bigger complex molecule will vibrate at a some- what lower f requency or longer wave length.

Figure 4 shows the absorption spectra of the reac-

W

Z

nn n-

O

~n

1.4

1.2

1.0

0 . 8

.---O-- IN CHLOROFORM

IN. 70% ACETONE II II II �9

AFTER ADDING OXALIC ACID AND HEAT TREATMENT

0.6

0.4

0.2

I i

300 4 0 0 500

W A V E L E N G T H mp FIG. 3. Absorpt ion spectra of gossypo], concn 2 mg/ lO0 mL

tion products of gossypol and gossyverdurin with para anisidine at the same concentrations under the conditions used in the determination of gossypol. The similarity of the shape of the peaks at 447 m~ sug- gests a s t ructural relationship between gossypol and gossyverdurin. The apparent free gossypol content of gossyverdurin as determined colorimetriea]ly a f t e r reaction with para anisidine was 25.00%. The ap- parent total gossypol content obtained after hy- drolysis with oxalic acid was 32.50%.

F igure 5 shows the absorption spectra with concen- trat ions adjusted so as to give approximately the same height of the peak at 447 m/~. TheSe curves were pre- pared by subtracting the absorption of the unreacted material as is done in the determination of free and total gossypol. Of par t icular interest is the appear- ance of a second peak in the curves for gossyverdurin at 342 m~ which does not occur in the curve for gossy- poL Since this peak also does not occur in the curve for unreacted gossyverdurin, it may offer a possibility for the quanti tat ive determination of gossyverdurin in the presence of gossypol.

When gossyverdurin is hydrolyzed with oxalic acid in preparat ion for the s tandard determination of to- tal gossypol, the green color quickly disappears and the solution becomes brown. Under the same condi- tions gossypol is not changed.

This difference in behavior of gossypol and gossy- verdur in on heating in oxalic acid solution is illus- t ra ted in the absorption spectra shown in Figures 2 and 3.

Page 4: Gossyverdurin: A newly isolated pigment from cottonseed pigment glands

574

0 . 4

0.3

W L) Z < rn 0 . 2 n- O 03 rn <

0 . 1

T H E J O U R N A L OF T I l E A M E R I C A N O I L C I : [ E M I S T S ' S O C I E T Y

GOSSYPOL-- Io-AN ISIDINE REACTION PRODUCT

�9 GOSSYVERDURIN-- Io-ANISIDINE REACTION PRODUCT

----a---- HYDROLYZED GOSSYVERDURIN-- p -ANIS I - DINE REACTION

PRODUCT ]3Q~

3 0 0 4 0 0 5 0 0

WAVE LENGTH rnp FIG. 4. A b s o r p t i o n s p e c t r a o f g o s s y p o l a n d g o s s y v e r d u r i n - -

p a r a - a n i s i d i n e r e a c t i o n p r o d u c t s , conen 0.1 rag . i n 25 ml .

0,4

0.3

GOSSYPOL-- p-ANISIDINE REACTION PRODUCT

�9 G O S S Y V E R D U R I N - - " " "

----,e--- HYDROLYZED GOSSYVERDURI N " " "

t['

3 0 ( 4 0 0 5 0 0

WAVE LENGTH mp

W 0 Z < rn I~ 0 . 2 0 03 rn <

0.r

FiG. 5. A b s o r p t i o n s p e c t r a o f g o s s y p o l a n d g o s s y v e r d u r i n - - p a r a - a n i s i d i n e r e a c t i o n p r o d u c t s , conch g o s s y p o l 0.1 r a g / 2 5 m ] ; g o s s y v e r d u r i n 0.4 r a g / 2 5 m l ; h y d r o l y z e d g o s s y v e r d u r i n 0.3 m g / 25 ml . T h e s e c u r v e s w e r e p r e p a r e d b y s u b t r a c t i n g the a b s o r p - t i o n o f t he u n r e a c t e d m a t e r i a l a s is done i n t h e d e t e r m i n a t i o n o f f r e e a n d t o t a l g o s s y p o l .

VOL. 40

TABLE I I

Nutritional Study of Fractions Obtained from Cottonseed P igment Glands a

Material fed

Acetone soluble fraction of pigment glands

Fractions Muted from DEAE column

Water effluent ..........

60% Acetone. ......... 75% Acetone ..........

100% Acetone ..........

Chloroform ............... Gossyverdurin ......... : ......

No. ose as of rats /kg of given the )dy wt I dose

2 . 0 0 10 1.25 10 0.75 10 0.50 10

1.00 2.00 2.00 2.00 1.00 2.00 4.00 2.00 2.00 1.00 0.75 0.50

No. of rats which died

10 8 6 0

% gain or loss in wt

among nrvlvln animMs

(avg)

_fb 19.4 12.2

9.4

4,12.4 4- 3.8 4- 1.4 4" 0.2 4.10.4 4- 5.8 -- 3.2 - - ] . 4

_fb 21.2 18.9 20.0

Calcu- lated

LD-50 g / k g

0.79

0.66

" The doses were given in corn oil to male rats weighing 160 ( • 10) g. and the rats were kept under observation for 7 days or until death.

t) Fatal to all rats tested.

Phys;~ological Studies. Table I I shows physiological effects of the different fract ions obtained f rom the pig- ment glands. The LD-50 value for the original glands was 1.12 g / k g body weight. Details of this assay have been repor ted previously (8). The value obtained for the acetone extract was 0.79 g /kg body weight and the free gossypol content of this extract was 64.70'%. I f gossypol was assumed to be the only physiologically active compound in the extract then the calculated LD-50 value for gossypol would be 0.55 g /kg body weight. I t has been shown previously that this amount of gossypol does not kill rats (8). There was also a direct relationship between the dose given and the per cent loss in weight among the surviving animals. Autopsy of dead rats revealed symptoms sinfilar to those observed in rats fed p igment glands (8) where consistent hyperemia of the gastrointestinal t rac t was observed. The livers showed prominent lobules. The kidneys and the lungs were congested.

None of the five fractions eluted f rom the D E A E proved to be fa ta l to rats at the doses given. The ra ts given the water effluent, at the two levels of 1.00 and 2.00 g/kg, gained weight. The most interesting point is that these five fract ions which were isolated f rom the acetone-soluble, water-soluble fract ion of the pig- ment glands were not lethal at the levels tested al- though they constituted ca. 37.5% of the acetone- soluble, water-soluble fraction.

The gossyverdurin, when examined for its physio- logical effects, proved to be the most toxic material obtained f rom cottonseed which has so fa r been re- ported. While the LD-50 value for the pigment glands, gossypol, and gossypurpur in were 1.12, 2.57, and 6.68 g /kg body weight, respectively (8), the gossyverdurin showed an LD-50 value of 0.66 g / k g body weight. Almost all ra ts on the four gossyverdurin doses were immobilized, and showed severe diarrhea with blood in the stools. The rats which did not die by the end of the exper iment showed considerable losses in weight. Autopsy of the dead animals revealed consistent hy- peremia of the gastrointest inal tract. Severe inflam- mations and hemorrhages were observed in the pyloric region of the stomach. Congestions throughout the in- testines, highly inflamed caecum and, in severe cases blood and mucus, were observed. The liver was light in color and the lobules were prominent . The kidneys showed badly congested medulla. The lungs were congested.

Page 5: Gossyverdurin: A newly isolated pigment from cottonseed pigment glands

OCTOBER, 1963 L Y M A N ET AL. : GOSSYVERDURIN:

With respect to the quant i ta t ion of the total toxicity of the pigment glands and the various constituents, it should perhaps be pointed out that gossyverdurin is a very unstable compound and it may be that a significant amount of the gossyverdurin decomposed and was lost dur ing the isolation process.

ACKNOWLEDGMENT

This investigation supported in par t by a grant-in-aid from the National Cottonseed Products Association.

1%EFEl~E NCE S

1. Spadaro, J. J., i%. M. Persell, C. H. Murphey, Jr., H. L. E. Vix, E. G. McCourtney. J. L. Hecker, E. F. Pollard, and E. A. Gastrock, JAOCS, 25, 345-353 (1948).

2. Spadaro, J. J., R. M. Persel|, C. G. 1%euther, Jr., I t . L. E. Vix,

N E W L Y ISOLATED P I G M E N T 5 7 5

E. J. Laborde, J. W. Latham, 1%. L. Jaeger, E. F. Pollard, and E. A. Gastrock, Ibid., 27, 336-343 (1950).

3. Conch, J. 1%., W. Y. Chang, and C. M. Lyman, Poultry Sci., 34, 178-188 (1955).

4. Lillie, i%. J., and H. 1%. Bird, 1bid., 29, 390-393 (1950). 5. Eagle, E., L. E. Castillon, C. M. Hall, and C. H. Boatner, Arch.

Biochem., 18, 271-277 (1948). 6. Moore, A. T., and M. L. 1%ollins, JAOCS, 88, 156-160 (1961). 7. Eagle, E., C. M. Hall, L. E. Castillon, and C. M. Boatner, Ibid.,

27, 300-303 (1950). 8. E1-Nockrt~shy, A. S., C. M. Lyman, and J. ~V. Dollahite, Ibid.,

40, 14-17 (1963). 9. Po.ns, W. A., Jr., and J. D. Guthrie, Ibid., 26, 671-676 (1949). 10. Pons, W. A., Jr., C. L. Hoffpauir, and 1%. T. O'Connor, Ibid., ~7,

390-393 (1950) . 11. Fiske, C. I-I., and V. Subba1%ow, J. Biol. Chem., 66, 375 (1925). 12. Bliss, C. I., J. Pharm. and Pharmocol., 11, 192-256 (1938). 13. Shriner, 1%. L., I~. C. Fuson, and D. Y. Ourtin, "The Systematic

Identification of Organic Compounds," 4th Ed., New York, John Wiley & Sons, Inc., 1956.

[ :Received O c t o b e r 26, 1 9 6 2 - - A c c e p t e d A p r i l ]6 , 1963]

The Separation of Glycerides by Liquid-Liquid Column Partition Chromatography I B. C. BLACK and E. G. H A M M O N D , Department of Dairy and Food Industry, Iowa State University of Science and Technology, Ames, Iowa

Abstract A l iqu id - l i qu id p a r t i t i o n c h r o m a t o g r a p h y

method was developed to separate triglycerides. The solvent was a two phase mixture of acetone, heptane, and water suppor ted on silane t reated eelite. A s tudy was made of the best means for equil ibrating the solvents and support , packing the column, and introducing the sample. The effect of various operat ing variables such as flow rate, sample size, column length, and solvent com- positions was studied using t r i laur in and t r imy- ristin as model glyeerides. Under the best condi- tions achieved, it was calculated that glycerides differing by two carbon atoms or one double bond would not separate completely, but glycer- ides differing by two double bonds or four carbon atoms would be separated. Cocoa butter , a rela- t ively simple triglyceride, was fractionated, and the fa t ty acid composition of each f ract ion was determined by gas chromatography. The g]yeeride composition was calculated and compared with theoretical compositions. The results indicate that useful glyeeride separation can be obtained with this system. Probably even more useful separa- tions could be obtained if a more sensitive device were used to detect the tr iglyeerides in the effluent. This would allow the use of solvent com- positions which give larger retention volumes and more plate efficiency.

Introduction

L IQUID-LIQUID P A R T I T I O N C'I~IROMATOG1%API~Y has dem- onstrated its abil i ty to give analyt ical ly useful

separat ions of f a t t y acids and tr iglycerides according to chain length and unsaturat ion. The reversed-phase system based on siliconized celite, developed by How- ard and Mart in (1), has been par t icu lar ly useful f o r separat ing long chain f a t ty acids. Various paper and thin-layer par t i t ion chromatography methods f o r separat ing glycerides have been reported, but they

1 Journal Paper No. J-4517 of the Iowa Agricultural and Home Eco- nomics Experiment Station, Ames, Iowa, Project No. 1517. Presented at the AOCS meeting in Toronto, Canada, 1962. Based on a thesis presented by B. O. Black in partial fulfillment of the requirements for a master 's degree.

are seriously limited in the amount of glyceride they will separate (2,3,4). This has lef t quant i ta t ive esti- mat ion of the glycerides subject to considerable error. Recently Hirsch (5) has repor ted a liquid-liquid col- umn chromatography method for tr iglycerides and other lipids using factice as a support .

This paper reports a s tudy of factors affecting the separat ion of the synthetic simple triglycerides, tri- laur in and t r imyris t in , by liquid-liquid par t i t ion chro- ma tography with silane t reated Celite as a support . This informat ion was then applied to the fract ionat ion of cocoa butter .

Experimental The appara tus used to f raet ionate the t r ig lyce r ides

is shown in F igure 1. Columns 1.8 cm inside diameter of various lengths were used. The s ta t ionary phase was silane treated celite p repared according to How- ard and Mart in (1). The solvent system was a two- phase mixture of acetone, heptane, and water. The percentage of each component depended on the tri- glycerides to be fractionated. The packing was pre- pared by equil ibrating the celite with the solvent in a large separa tory funnel in a 30r water bath. The eelite and the lower (or mobile) phase were allowed to run into the column, and the celite was packed f requent ly with a plunger. The remaining lower (mobile) phase was put into a jacketed reservoir and used to elate the glycerides. The sample was intro- duced into the column (10 mg each of t r i laur in and t r imyr is t in or 17 mg of cocoa but ter dissolved in pure heptane, 100 m g / m l ) , and the reservoir of eluting solvent was placed at the top of the column. The frac- tions eluted f rom the column were collected by an 'automatic f ract ion collector. The flow rate of column was controlled by a teflon plug stopcock. The Solvent in each fract ion was evaporated, and the elution of the tr iglycerides f rom the column was monitored by the colorimetric ester method of Hack (6).

When cocoa bu t te r was analyzed, duplicate runs were made. One run was analyzed colorimetrically to construct a weight curve. In the duplicate run, the solvent in each f ract ion was evaporated under a s t ream of purified nitrogen. The fract ions were corn-