Gas Chromatographic Separation of Cholesteryl Esters of Fatty Acids of Different Degrees of Unsaturation TORU TAKAGI, AKIRA SAKAI, YUTAKA ITABASHI, and KENJI HAYASHI, Department of Chemistry, Faculty of Fisheries, Hokkaido University, Hakodate, Japan

ABSTRACT

Cholesteryl esters prepared from the fatty acid methyl esters of linseed oil, pig liver lipids, and Japanese anchovy oil have been separated on the basis of their chain lengths and number of double bonds by gas liquid chromatography on a cyano- siloxane SILAR 10C column. The equiva- lent chain lengths of cholesteryl esters having acyl groups with 14-22 carbons and 0-6 double bonds are presented. A s i g n i f i c a n t influence of the column temperature on the equivalent chain lengths of the polyenoic fatty acid cho- lesteryl esters has been found. Separation of the cholestanyl and epicholestanyl esters of linseed oil fatty acids, respec- tively, under the same conditions is also described.

INTRODUCTION

Developments in gas liquid chromatography (GLC) now permit the direct analysis of the fatty acid cholesteryl esters as described in reviews (1,2). The fatty acid cholesteryl esters r e q u i r e columns and operating conditions similar to those employed in GLC of triglyc- erides and are usually carried out on a ther- mally stable nonpolar phase such as SE-30 (3), JXR (4), and OV-17 (4). The results have usually been presented in compositions based on the chain length only.

Kuksis presented two methods for the separation of saturated and unsaturated fatty acid cholesteryl esters in his review (1): GLC on a nonpolar column following a preliminary separation of the mixture on the basis of unsaturation by argentation chromatography, and GLC on thin-film packing containing polyester liquid phase. However, the former techniques are not simple and the latter seems to be impractical since thermal decomposition of the polyester takes place at temperatures near 300 C.

Recently, Murata et al. reported the direct analysis of fatty acid cholesteryl esters on the basis of the chain length and number of double bonds with a mass spectrometer and a chemical ionization source (5). The method seems to be rapid and simple. However, the availability of such instruments is limited and some of the pure specimens required for the analysis are not available at the present time.

This paper presents a new GLC technique for the resolution of cholesteryl esters of satu- rated and unsaturated fatty acids on SILAR 10C, which was formerly APOLAR 10C as described in the previous paper (6). Though the conditions required are more severe than those used for the wax esters (6), reproducible results were obtained without the base line drift, and the liquid phase could be used over a fairly long time.

MATERIALS AND METHODS

The preparation of fatty acid esters from

P x

TIPI[ ( m n ) , ~ ~ _

r l ~ ( H I M ) r l ~ ( m N )

FIG. 1. Gas chromatograms of methyl esters (A) and cholesteryl esters (B) of linseed oil fatty acids. Column; length 0.5 m, A 150 C, B 240-270C (2 C/rain).

FIG. 2. Gas chromatograms of methyl esters (A) and cholesteryl esters (B) of pig liver lipid fatty acids. Column; length A 1.5 m, B 0.5 m, A 180C, B 240-270 C (2 C/min).

228

GC SEPARATION OF CHOLESTERYL ESTERS 2 2 9

c h o l e s t e r y l a c e t a t e a n d l i n s e e d oil m e t h y l e s t e r s was c a r r i e d o u t b y a m o d i f i e d t r a n s e s t e r i f i c a - t i o n m e t h o d (7 ,8 ) . A m i x t u r e o f l i n s e e d oil m e t h y l e s t e r s ( 1 . 0 g ) , c h o l e s t e r y l a c e t a t e ( 1 . 5 g), d i o x a n e (2 m l ) , a n d p o t a s s i u m t- b u t o x i d e ( 5 0 m g ) wa s h e a t e d a t 70 C fo r 1 h r u n d e r n i t r o g e n . T h e c h o l e s t e r y l e s t e r s o f t h e f a t t y a c id s f r o m pig l iver l i p id s a n d J a p a n e s e a n c h o v y oil , a n d t h e c h o l e s t a n y l a n d e p i c h o - l e s t a n y l e s t e r s o f t h e f a t t y a c i d s f r o m l i n s e e d oil were p r e p a r e d b y t h e s a m e p r o c e d u r e . T h e w a x e s t e r s were s y n t h e s i z e d as d e s c r i b e d in t h e l i t e r a t u r e (7 ) b y h e a t i n g o f t h e m i x t u r e s o f m e t h y l e s t e r s ( 1 . 0 g) a n d f a t t y a l c o h o l (0 .5 g) w i t h s o d i u m m e t h y l a t e ( 4 0 m g ) a t 70 C fo r 4 0 m i n . T h e m e t h y l e s t e r s we re p r e p a r e d f r o m l i n s e e d oil, J a p a n e s e a n c h o v y oil, a n d e r u c i c acid. C e t y l , s t e a r y l , o ley l , a n d e r u c y l a l c o h o l s

T I ! (Mill

FIG. 3. Gas chromatograms of methyl esters (A) and cholesteryl esters (B) of Japanese anchovy oil fa t ty acids. Column; length 0.5 m, A 150C, B 240-270 (2 C/rain).

TABLE I

Relative Retent ion Times and ECL Values o f Cholesteryl Esters in Comparison with Those of Methyl and Stearyl Esters a

Cholesteryl ester b Stearyl ester c Methyl ester d

Acyl group RRT ECL RRT ECL RRT ECL

14:0 t .00 --- 1.00 --- 1.00 --- 16:0 1.30 - - 1.36 --- 1.62 --- 16:1 1.56 1"/.5 1.59 16.9 1.99 16.8 18:0 1.68 --- 1.86 --- 2.69 --- 18:1 1.98 19.3 2.13 18.9 3.24 18.7 18:2 2.44 20.9 2.55 20.0 4.17 19.8 18:3 3.03 22.6 3.11 21.2 5.59 20.9 20:0 2.17 . . . . . . . . . . . . . . 20:4 3.62 24.0 . . . . . . . . . . . 20:5 4.47 25.6 5.12 24.3 12.9 24.4 22:6 6.44 28.5 7.68 26.9 25.0 27.1

aECL = equivalent chain length based on the carbon number of acyl groups, RRT = rela- tive retention t ime (standard, myristate) . Retent ion t imes of myris tates (rain); cholesteryl 5.14, stearyl 3.99, methyl 2.50. Glass co lumns of inside diameter 3 m m packed with 5% SILAR 10C on Gas Chrom Q (100-120 mesh) were used to obtain the data in every case.

bColumn temperature 270 C. Column length 0.5 m. CColumn tempera ture 250 C. Column length 1.5 m. dColumn temperature 180 C. Column length 1.5 m.

TABLE II

Separation Factors of Fat ty Acid Esters on SILAR 10C a

Molecular species Cholesteryl ester Stearyl ester Methyl ester

Monoene/Satura te b 1.19 1.16 1.22 Diene/Monoene 1.23 1.20 1.29 Triene/Diene 1.24 1.22 1.34 Cn+2/Cn c 1.29 1.36 1.64

aSeparation factors are average values calculated from relative retention t imes (RRT) listed in Table I.

bCalculated from the RRT of each ester having the aeyl groups of same carbon number . eSeparation factors of esters having the same number of double bonds; n = carbon num-

ber of acyl groups.

LIPIDS, VOL. 12, NO. 2

230 TORU TAKAGI, AKIRA SAKAI, YUTAKA ITABASHI AND KENJI HAYASHI

TABLE III

Comparison of a t ECL of Cholesteryl and Methyl Esters a

Cholesteryl ester b Methyl ester c

Acyl group 240-255 250-260 255-265 DEGS SILAR 10C

18:1 0.0040 0.0083 0.0157 0.0040 0.0068 18:2 0.0058 0.0167 0.0272 0.0050 0.0108 18: 3 0.0112 0.0263 0.0410 0.0080 0.0135

aA t ECL; increase of equivalent chain length (ECL) per one degree rise in column temperature.

b240-255 indicates that the A t ECL was calculated from ECL at 240 and 255 C. eThe A t ECL was calculated from ECL at 150 and 190 C.

TABLE IV

Compositions of Cholesteryl Esters in Comparison with Those of Original Methyl Esters a

Acyl group Cholesteryl ester Methyl ester

16:0 5.8 6.0 18:0 2.8 3.5 18:1 16.0 15.3 18:2 15.5 15.7 18:3 59.9 59.5

apercents of peak area.

were iden t i f i ed by compar i son of the syn the - sized cho les te ry l esters wi th appropr ia t e ref- e rence s tandards . Af t e r the c o l u m n was cooled to the ini t ia l t empe ra tu r e , suf f ic ient t ime was t a k e n for the S ILAR 10C to rever t back to the original polar i ty . An analogous a l t e ra t ion of the l iqu id phase po la r i ty , d e p e n d e n t u p o n t empera - ture , was r e p o r t e d w i th the cyanos i loxane XE-60 (9).

RESULTS AND DISCUSSION

were used for the es ter i f ica t ion . The GLC of cho les te ry l esters was per-

f o r m e d wi th a Sh imadzu GC 6AM i n s t r u m e n t e q u i p p e d wi th dual glass c o l u m n s ( 0 . 5 m length , 3 m m inside d iamete r ) and f lame ioniza- t ion de tec tors . The co lumns were packed wi th Gas C h r o m Q (100-120 mesh ) coa ted wi th 5% SILAR 10C (Appl i ed Science Labora to r ies , Sta te College, PA). The car r ier gas was n i t r ogen at 40 ml /min . Prior to use, the co l um ns were c o n d i t i o n e d at 270 C for 10 hr. D e t e c t o r and in jec t ion hea te r s were m a i n t a i n e d at 320 C. The co lumn t e m p e r a t u r e was k e p t at 270 C or p r o g r a m m e d f rom 240-270 C at the ra te of 2 C /min f rom the s ta r t of the analysis. Dif- f icult ies in d e t e r m i n i n g peak areas due to par t ia l over lapp ing were ove rcome by a digital i n t e g r a t o r ( S h i m a d z u ITG-4AX) . The peaks

Figures 1-3 compa re the typ ica l c h r o m a t o - grams o b t a i n e d for the cho les te ry l esters pre- pa red f rom m e t h y l esters of l inseed oil, pig liver lipids, and Japanese a n c h o v y oil, w i th those for the or iginal m e t h y l esters, respect ively. Af te r e x a m i n i n g co lumns 0.3-1.5 m in length , a 0.5 m c o l u m n was se lec ted to f inish the to ta l analysis f rom choles te ry l myr i s t a t e (14 :0 -Ch) to cho- l e s t e ry l d o c o s a h e x a e n o a t e ( 2 2 : 6 - C h ) wi th in 1 hr. Fo r good r e so lu t ion o f the cho les te ry l esters of some vegetable oil f a t t y acids, a 1.5 m c o l u m n is preferable . The e lu t ion pa t t e rn s are charac te r i zed b y a fair ly good r e so lu t i on of the var ious c o m p o n e n t s as s h o w n in Figures 1-3, excep t for the p o o r separa t ion of s teara te and h e x a d e c e n o a t e peaks in Figure 3. The c o l u m n t e m p e r a t u r e used in this s t udy is s o m e w h a t low c o m p a r e d wi th those wh ich have been r e p o r t e d

TABLE V

Comparison of Retention Times and ECL Values of Fatty Acid Epicholestanyi, Cholestanyl, and Cholesteryl Esters

Epicholestanyl ester Cholestanyl and cholesteryl esters

Acyl group Retention time (min) ECL a Retention time (rain) ECL a

16:0 4.25 -- 6.75 -- 18:0 5.45 -- 8.78 --- 18:1 6.27 19.1 10.3 19.3 18:2 7.56 20.6 12.6 20.9 18:3 9.28 22.3 15.5 22.6

aEquivalent chain length (ECL) based on the carbon number of acyl groups.

LIPIDS, VOL. 12, NO. 2

GC SEPARATION OF CHOLESTERYL ESTERS 231

in previous papers (1-4). This is permitted by the low affinity of the SILAR I 0C polar liquid phase to the cholesteryl esters which have a low polarity. Table I compares the relative retention times (RRT) with respect to myristate and equivalent chain lengths (ECL) based on the chain lengths of the acyl groups for cholesteryl, stearyl, and methyl esters. Cholesteryl esters of unsaturated fatty acids show more or less higher ECL than the corresponding stearyl and methyl esters while ECL of the latter shows quite similar values. This is to be expected since the slopes of the lines in plotting log 10 reten- tion time against chain lengths for wax esters and cholesteryl esters are different as described later. The separation factors calculated for the cholesteryl, stearyl, and methyl esters are sum- marized in Table II. The separation factors of the cholesteryl esters, based on similar degrees of unsaturation and fatty acid chain lengths, are close to those of wax esters but lower than those of methyl esters. This is evidently due to the decrease of resolution from the higher molecular weights of the cholesteryl esters.

In comparison with SILAR 5CP, SILAR 10C liquid phase showed high separation factors for the double bond homologues and the low sepa- ration factors for the chain length homologues (10,11). This characteristic is most significant in cholesteryl esters. In practice, this is recog- nizable from the fact that linoleate elutes after arachidate in GLC of cholesteryl esters, whereas the former usually emerges before the latter in the methyl esters, and the former overlaps with the latter at 250 C in the wax esters. The influence of the column temperature on the separation patterns of the methyl esters on the polyester columns was reported in detail by Ackman (12).

The wax es te r s , s t e a r y l b e h e n o a t e (18:0-22:0), the saturated components of 52 carbons in beeswax and erucyl docosahexa- enoate (22:1-22:6) showed the retention times corresponding to ca. l:0-Ch, 14:0-Ch, and 1 l:0-Ch in the cholesteryl ester series, respec- tively (13). Thus, mixtures of wax esters and cholesteryl esters can be separated on SILAR 10C, if the sample does not contain cholesteryl esters shorter than laurate and wax esters longer than 22:1-22:6 on the retention time base. The influence of column temperature on the ECL can be shown by A t ECL, the increase of ECL for one degree rise of column temperature. The A t ECL values of cholesteryl and methyl esters are shown in Table III. It indicates that A t ECL va lues of cholesteryl esters increase with increasing degree of unsaturation and increase of column temperature. The data demonstrate that the overlapping components having dif-

ferent degrees of unsaturation could be sepa- rated by selection of a more appropriate column temperature, and that the ECL of polyenoic acid cholesteryl esters should be reported together with a record of the column temperature.

The facts concerning the influence of column temperature on the separation factors were reported in the GLC of the esters on polar columns in a few papers. In the GLC of methyl esters on DEGS and an organosilicone polyester EGSS, it was found that the ECL values of unsaturated esters change with the temperature since the slopes of the lines due to unsaturated esters relative to the saturated esters alter with temperature when plotting log10 retention time against carbon number (12,14,15). In the G LC of trimethylsilyl ethers of diacylglycerols on SILAR 5CP, the relative temperature depen- dence of separation by carbon number and degree of unsaturation can be illustrated by plotting the ECL of diacylglycerols versus the sum of the ECL values of methyl esters of the fatty acids (10). Although nearly straight lines are obtained for the diacylglycerols from seed oils, the polyunsaturated species present in phospholipids have been found to deviate from this relationship. It was suggested that the esters containing polyunsaturated species do not possess the same temperature dependence (10).

The results in Table III show that A t ECL of an unsaturated methyl ester on SILAR 10C phase is higher than the A t ECL on DEGS. This suggests that SILAR 10C is more sensitive to temperature effects. This effect can be ascribed to conformational changes of polymers, which have bulky polar side chains, with increases of thermal energy as described in GLC on XE-60 (9).

The peak area compositions of cholesteryl esters of linseed oil fatty acids were determined by GLC on SILAR 10C columns. As shown in Table IV, the proportions of the cholesteryl esters were in fair agreement with those of original methyl esters. Further work, however, is now in progress to demonstrate more reliable results for the quantification of cholesteryl esters on SILAR 10C.

Table V shows the ECL of the cholesteryl esters of linseed oil fatty acids. The ECL values based on the carbon number of the acyl groups are in both cases in agreement with those of the corresponding cholesteryl esters. The steric con- figuration of the 3-OH has a significant influ- ence on the ECL values, but the absence of 5:6-double bond has little influence in the cholestanyl esters. Thus, the GLC on SILAR 10C is expected to be useful for the characteri-

LIPIDS, VOL. 12, NO. 2

232 TORU TAKAGI, AKIRA SAKAI, YUTAKA ITABASHI AND KENJ1 HAYASHI

zation of the fatty acid esters of various sterols.

REFERENCES

1. Kuksis, A., J.J. Myher, L. Marai, and K. Gehr, J. Chromatog. Sci. 13:423 (1975).

2. Kuksis , A., Fette Seifen Anstr iehm. 75:420 (1973).

3. Gold, M., and G. Mathew, Anal. Chem. 47:752 (1975).

4. Kuksis, A., L. Marai, and D.A. Gornall, J. Lipid Res. 8:352 (1967).

5. Murata, T., S. Takahashi, and T. Takeda, Anal. Chem. 47"577 (1975).

6. Takagi, T., Y. Itabashi, T. Ota, and IC Hayashi, Lipids 11:354 (1976).

7. Phillips, F., and C.V. Viswanathan, Lipids 2:437

(1967). 8. Mahadevan, V., and W.O. Lundberg, J. Lipid Res.

3:106 (1962). 9. Chen, C., and D. Gacke, Anal. Chem. 36:72

(1964). 10. Myher, J.J., and A. Kuksis, J. Chromatog. Sci.

13:138 (1975). 11. Myher, J.J., L. Marai, and A. Kuksis, L Lipid Res.

15:586 (1974). 12. Ackman, R.G., Lipids 2:502 (1967). 13. Takagi, T., and Y. Itabashi, submi t ted for publi-

cation in L Chromatog. Sci. 14. A c k m a n , R.G., J. Gas Chromatog. 1(6):11

(1963). 15. Ackman, R.G., JAOCS 40:558 (1963).

[ Received September 14, 1976]

LIPIDS, VOL. 12, NO. 2